8 Advances in Environmental Research Vol 32 - Reyes & Kneeshaw 2014

Complimentary Contributor Copy


ADVANCES IN ENVIRONMENTAL RESEARCH

ADVANCES IN ENVIRONMENTAL

RESEARCH

VOLUME 32

No part of this digital document may be reproduced, stored in a retrieval system or transmitted in any form orby any means. The publisher has taken reasonable care in the preparation of this digital document, but makes noexpressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. Noliability is assumed for incidental or consequential damages in connection with or arising out of informationcontained herein. This digital document is sold with the clear understanding that the publisher is not engaged inrendering legal, medical or any other professional services.



Additional books in this series can be found on Nova‟s website

under the Series tab.

Additional e-books in this series can be found on Nova‟s website

under the e-book tab.



ADVANCES IN ENVIRONMENTAL

RESEARCH

VOLUME 32

JUSTIN A. DANIELS

EDITOR

New York


Copyright © 2014 by Nova Science Publishers, Inc.

All rights reserved. No part of this book may be reproduced, stored in a retrieval system or

transmitted in any form or by any means: electronic, electrostatic, magnetic, tape, mechanical

photocopying, recording or otherwise without the written permission of the Publisher.

For permission to use material from this book please contact us:

Telephone 631-231-7269; Fax 631-231-8175

Web Site: http://www.novapublishers.com

NOTICE TO THE READER

The Publisher has taken reasonable care in the preparation of this book, but makes no expressed or

implied warranty of any kind and assumes no responsibility for any errors or omissions. No

liability is assumed for incidental or consequential damages in connection with or arising out of

information contained in this book. The Publisher shall not be liable for any special,

consequential, or exemplary damages resulting, in whole or in part, from the readers‟ use of, or

reliance upon, this material. Any parts of this book based on government reports are so indicated

and copyright is claimed for those parts to the extent applicable to compilations of such works.

Independent verification should be sought for any data, advice or recommendations contained in

this book. In addition, no responsibility is assumed by the publisher for any injury and/or damage

to persons or property arising from any methods, products, instructions, ideas or otherwise

contained in this publication.

This publication is designed to provide accurate and authoritative information with regard to the

subject matter covered herein. It is sold with the clear understanding that the Publisher is not

engaged in rendering legal or any other professional services. If legal or any other expert

assistance is required, the services of a competent person should be sought. FROM A

DECLARATION OF PARTICIPANTS JOINTLY ADOPTED BY A COMMITTEE OF THE

AMERICAN BAR ASSOCIATION AND A COMMITTEE OF PUBLISHERS.

Additional color graphics may be available in the e-book version of this book.

Library of Congress Cataloging-in-Publication Data

ISSN: 2158-5717

Published by Nova Science Publishers, Inc. † New York

ISBN: 978-1-63117-330-1 (eBook)


CONTENTS

Preface vii

Chapter 1 Improved Gas and Vapor Detection Using Denuder Technology 1 Amanda L. Jenkins

Chapter 2 Environmental Protection from Thermodynamic Properties,

the Performance of Refrigeration Cycles and Air Conditioning 23 Abdeen Mustafa Omer

Chapter 3 Some Aspects of Solar and Wind Energy Resources 103 Abdeen Mustafa Omer

Chapter 4 Rationalization of Salt-Related Processes in the

Leather Industry as a Tool for Minimization of Their

Environmental Impact 137 Karel Kolomazník, Michaela Bařinová, Jiří Pecha and Dagmar Janáčová

Chapter 5 Defuzzification of Fuzzy Concepts to Support Vulnerability

Assessments of Climate Change Impacts in the Philippines 165 Lilibeth A. Acosta and Jemimah Mae A. Eugenio

Chapter 6 Ecological Resilience: Is It Ready for Operationalisation

in Forest Management? 195 Gerardo Reyes and Daniel Kneeshaw

Chapter 7 Climate Change and an Australian Rainforest Conifer 213 Rohan Mellick

Chapter 8 A Novel Thermal Seawater Desalination Process

Based on Self-Heat Recuperation 267 Hiroyuki Mizuno, Yasuki Kansha and Atsushi Tsutsumi

Chapter 9 Biochemical Gas Sensor “Bio-Sniffer” and Imaging System

for Medical and Health Care Science 279 Takahiro Arakawa and Kohji Mitsubayashi


Contents vi

Chapter 10 Desalination of Brackish Water by Electrodialysis:

Effects of Operational Parameters and Water Composition

on Process Efficiency 295 Mourad Ben Sik Ali, Amor Hafiane, Mahmoud Dhahbi

and Béchir Hamrouni

Chapter 11 Appropriate Time Scale for Aggregating Climatic Data

to Predict Flowering and Boll Setting Behavior of Cotton 331 Zakaria M. Sawan

Index 347


PREFACE

This series covers leading-edge research focusing on the environment, including current

research data on improved gas and vapor detection using denuder technology; environmental

protection from thermodynamic properties; solar and wind energy resources; salt-related

processes in the leather industry; climate change impacts in the Philippines; ecological

resilience; climate change and Australian rainforest; seawater desalination; biochemical gas

sensors; desalination of brackish water; and time scales for aggregating climatic data.

Chapter 1 - An improved method for accomplishing air/vapor sample collection and

transport has been achieved through the use of denuder technology. Denuders use gas

injection technology to continuously introduce a gas or aerosol into a solvent stream where

the analyte is taken into a “trapping” solution preselected for a specific analyte. This

extraction has been demonstrated to have greater than 96% efficiency. Unlike extraction or

bubbler methods denuder extraction takes place in real time and with less loss. Once the

materials are in solution they can be transported to many different types of analytical

detection platforms including LC-MS and ion selective electrodes. This chapter describes

several different applications of the denuder to enhance gas/aerosol detection as well as a

novel vapor generation system to provide a means to challenge the detection methods.

Chapter 2 - Over the years, all parts of a commercial refrigerator, such as the compressor,

heat exchangers, refrigerant, and packaging, have been improved considerably due to the

extensive research and development efforts carried out by academia and industry. However,

the achieved and anticipated improvement in conventional refrigeration technology are

incremental since this technology is already nearing its fundamentals limit of energy

efficiency is described is „magnetic refrigeration‟ which is an evolving cooling technology.

The word „green‟ designates more than a colour. It is a way of life, one that is becoming more

and more common throughout the world. An interesting topic on „sustainable technologies for

a greener world‟ details about what each technology is and how it achieves green goals.

Recently, conventional chillers using absorption technology consume energy for hot water

generator but absorption chillers carry no energy saving. With the aim of providing a single

point solution for this dual purpose application, a product is launched but can provide

simultaneous chilling and heating using its vapour absorption technology with 40% saving in

heating energy. Using energy efficiency and managing customer energy use has become an

integral and valuable exercise. The reason for this is green technology helps to sustain life on

earth. This not only applies to humans but to plants, animals and the rest of the ecosystem.

Energy prices and consumption will always be on an upward trajectory. In fact, energy costs


Justin A. Daniels viii

have steadily risen over last decade and are expected to carry on doing so as consumption

grows. Refrigerants such as hydrochloroflurocarbons (HCFCs) are present in the ground

source heat pump (GSHP) systems and can pose a threat to the environment through being

toxic, flammable or having a high global warming potential.

However, new types and blends of refrigerant with minimal negative impacts are being

developed. A correctly fitted system will also greatly reduce the potential for leakage, which

is why using a professional installer is highly recommended. Significant CO2 savings can be

gained by displacing fossil fuels. Even compared to the most efficient gas or oil condensing

boilers, a well-designed heat pump with COP of 3-4 will reduce emissions by 30-35%.

Further carbon savings can be made if the electricity used to power the pump comes from a

renewable energy source such as photovoltaic or a renewable electricity tariff. Also, measures

can be taken to reduce the impact of pollution from using grid electricity generated through

fossil fuel. For example, one can purchase dual tariff green electricity from a number of

suppliers. However, even if ordinary grid electricity is used to run the compressor, the system

will still produce less CO2 emissions than even the most efficient condensing gas or oil boiler

with the same output. The term “vapour compression refrigeration” is somewhat of a

misnomer, it would be more accurately described as 'vapour suction refrigeration'. Vapour

compression is used to reclaim the refrigerant and is more aptly applied to heat pumps.

Vapour compression refrigeration exploits the fact that the boiling temperature of a liquid is

intimately tied to its pressure. Generally, when the pressure on a liquid is raised its boiling

(and condensing) temperature rises, and vice-versa. This is known as the saturation pressure-

temperature relationship.

Chapter 3 - Renewable energy resources (solar, wind and biomass) is one of the most

fundamental of natural resources that Sudan must harness in its efforts for rapid economic

development. The role of renewable energy technologies in the development process cannot

be over – emphasised. The demand for energy in Sudan has increased tremendously over the

years and will continue to increase in view of the accelerating pace of population growth,

urbanisation and industrialisation. Comprehensive renewable energy resources management

is a necessity. Human resource development should be based on education and training

programmes funded both by the private and public sector. Promotion research and

development, demonstration and adaptation of energy resources amongst national, regional,

and international organisations which seek clean, pure, safe, and abundant energy sources.

Results, suggest that, wind pumps, solar stills, and biogas energy must be encouraged,

invested, and implemented, but especially for remote rural areas of Sudan.

Chapter 4 - The main purpose of the authors‟ contribution is reducing the existing

environmental burden related to the worldwide use of sodium chloride for raw hide curing.

The essential step is rationalization of the curing process itself, since it determines the amount

of salt to deal with in the subsequent operations such as pre-soaking, soaking, after-soaking

and desalting of animal fleshings during their complex processing into valuable products.

Leather industry has been known as major producer of total dissolved solids (TDS),

particularly sodium chloride. The salt often gets to water environment, or right to the soils

resulting in arid conditions in the spilling areas, from where it can be again washed away to

water sources. Despite intensive search for alternative methods, sodium chloride remains the

most common way of raw hide curing worldwide. This method not only represents

considerable environmental burden, but insufficient application of the theory of the related

transport phenomena also has economic impacts, such as high consumption of the


Preface ix

preservation agent, of water and electric power. Rationalization of the curing process brings

considerable reduction in the time necessary for proper curing and thus reduction in the

amount of salt which in subsequent soaking operation would get to tannery effluents.

Understanding the related transport processes, particularly diffusion of sodium chloride into

or from the hide inner volume, gives ground for their optimization and thus to reduction of

their adverse environmental impact as well as minimization of the consumption of sodium

chloride, water and energies. The authors‟ approach lies in application of theoretical tools of

chemical engineering, namely indirect modeling based on quantitative relations from the

theory of transport phenomena. This method is in particular cases supported by

experimentally measured data, mainly estimation of the value of the effective diffusion

coefficient of sodium chloride in animal fleshings and its application in calculation the

minimal time necessary for proper curing of raw hides. The acquired knowledge is applied in

newly emerging industrial field, complex processing of animal fleshings into valuable

products such as quality biodiesel and gelatin. The fleshings contain considerable amount of

preservation salt, therefore desalting represents the key operation in their processing and the

quality of the resulting products is highly dependent on the precise performance of the

desalting process. The essential step leading to the optimization was the determination of the

time dependence of sodium chloride concentration (from the hide surface towards its internal

volume), in other words, a non-stationary concentration field of sodium chloride in the hide

during the curing process. Subsequently, the solution of the model allowed the calculation of

the optimum time necessary for sufficient preservation.

Chapter 5 - Climate change is one of the most pressing global issues that require

innovative methods to address the complexity of human-environment interactions. Many

aspects of vulnerability to climate change and adaptation measures to address its adverse

impacts remain vague and unquantifiable. Vulnerability assessments require methods that can

reduce the vagueness and imprecision of interpretations of data and information in human and

environment systems. This chapter illustrates an empirical application of fuzzy logic analysis

and the utility of this analytical tool in integrated modeling assessments in the context of

climate change. Using the intervulnerability framework, fuzzy logic models can be used to

assess trade-offs of adaptive capacity and hotspots of vulnerable regions. In particular, the

authors used data on socio-economic and ecological indicators that are relevant for the

assessments of adaptive capacity and vulnerability in the different provinces and regions in

the Philippines. The empirical application in this country shows the advantages of fuzzy set

theory in terms of its (1) transparency, which allows explicit presentation of model

assumptions through inference rules; and (2) flexibility, which allows direct inclusion of

informal and expert knowledge in combining various indicator sets. The results of the fuzzy

models show that the types of indicators and determinants traded-off depend on the social and

economic conditions in the regions. Vulnerable provinces are mostly located in northern and

southern parts of the Philippines. Vulnerability in the former can be reduced through

improving water availability for agriculture, whilst in the latter through improving peace and

order condition that affects socio-economic development.

Chapter 6 - Given the physiographic variability, variation in socio-political landscapes,

and differences in connectedness of people and communities associated with boreal forest

ecosystems, approaches to forest management that are flexible enough to accommodate this

variation are needed. Moreover, to ensure sustainable forest resource use, we need to embrace

the inherent complexity of boreal forest ecosystems rather than eliminate it, and be prepared


Justin A. Daniels x

to adapt and adjust as environmental conditions change. While ecological resilience may be a

useful forest management objective to this end, developing general guidelines to integrate it

into practice remains elusive. The authors address a number of questions often posed by

managers when attempting to include ecological resilience into forest management planning.

The authors‟ goal is to determine if the theoretical foundation of ecological resilience is

sufficiently developed to provide a general framework that can be applied for boreal forest

management.

Chapter 7 - Revealing the evolutionary trends of the recent past (late Quaternary) will

allow climate change science to anticipate the demographic response of species to future

disturbance. The genetic disjunctions and distribution of a long-lived rainforest conifer

provide a valuable signature of past demographic response to climate change – a biological

autograph of time, climatic cycles and the environment. This chapter reviews literature

pertaining to three case studies. Case study one determines the level of genetic diversity and

structure within naturally occurring populations of Podocarpus elatus and resolves the

influence of historic and contemporary drivers on divergence. Case study two explores the

impact of glacial cycle climatic changes on the palaeodistribution of the species: by

combining population genetic analysis, coalescent-based analysis, the observed fossil record

and environmental niche modeling. Case study three hypothesizes that future adaptive

potential is indicative of genetic/demographic change and range expansion/contraction trends

in Podocarpus elatus associated with the Clarence River Corridor. The final section of the

chapter concludes the case studies and suggests conservation strategies applicable to long-

lived species allied with threatened communities.

Chapter 8 - Thermal desalination processes, which separate pure water from seawater by

evaporation, have been used throughout the world. Thermal desalination typically comprises

either multi-stage flash or multi-effect desalination processes. These processes have the

following characteristics: the product water salinity is low enough for industrial use and

seawater quality does not seriously affect the operating conditions. However, heat is not

recovered perfectly by either process and is partly discarded to cooling water. Therefore,

energy consumption in conventional thermal processes is large.

A novel thermal desalination process based on self-heat recuperation, which is a

technology designed to save energy, has been proposed recently. In this process, all of the

heat is recirculated and energy is required only for producing a temperature difference in the

heat exchangers. As a result, the energy consumption can be dramatically reduced. At the

same time, it has been reported that the specific energy consumption, defined as the total

energy added divided by the product water flow rate, may be further decreased by increasing

the recovery ratio (the product water flow rate / the feed water flow rate).

In this chapter, the authors describe a novel desalination process based on self-heat

recuperation technology and explain its potential for saving energy compared with its

conventional counterparts. In addition, they summarize recent development of the process.

Chapter 9 - The authors introduce biochemical gas sensors “bio-sniffer” which leverage

biochemical reactions and characteristics of various enzymes. Volatile substances associated

with diseases and breath odors are released from human body. Hence, highly sensitive and

highly selective bio-sniffers have been developed and demonstrated in biomedical and health

care applications utilizing enzymes as biomarkers of related diseases. The authors present

three topics of bio-sniffer techniques for monitoring of gaseous components of methyl

mercaptan, formaldehyde and ethanol. A bio-sniffer for gaseous methyl mercaptan (MM) is


Preface xi

applied to measure halitosis (oral odor) in breath. MM is known as one of the important

chemicals of halitosis in oral care. And then, a fiber-optic biochemical gas sensor for

assessment of indoor environment is fabricated and demonstrated in monitoring of

formaldehyde. Finally, two-dimensional visualization system of gaseous ethanol is

demonstrated in spatial and temporal imaging of gaseous components of exhaled breath. In

this chapter, the authors describe the details and future prospects of biochemical gas sensors

for environmental analysis, medical and health care.

Chapter 10 - Electrodialysis is an electro-membrane process for separation of ions across

charged membranes from one solution to another under the influence of an electrical potential

difference used as a driving force. The transfer of the charged species is carried out according

to a mechanism of exchanges of ions between the ions of the solution and the counter ions of

the membrane.

This chapter is dealing with the effectiveness of the desalination of brackish water by this

process. A laboratory scale electrodialysis cell was used for this purpose.

The desalination of brackish solutions containing only one salt according to the

continuous mode (single pass process) was carried out initially. This study revealed that the

effectiveness of the process is dependent on the operational parameters of the electrodialysis

cell. But the rate of desalination is relatively weak. This rate does not exceed 50% and

sometimes does not allow reaching the awaited results.

Study of another configuration was carried out. This configuration is the discontinuous

mode or total recirculation mode (batch process). In this mode, the solutions are recycled in

the cell until the desired concentration is reached. While working with this mode, the

desalination rates of various solutions reached quickly high values. Indeed, they exceeded the

75% rate in few minutes.

This study concluded also that the composition of the solution itself influences the

effectiveness of desalination process. Indeed, the transfer of the ions from one compartment

to another depends on the various ionic species present in solution.

The presence of the calcium ions largely decreases the transfer of the magnesium ions

and vice versa. This is explained by the competition effect between two ions of same valence

during their transfers through exchanging membranes.

Moreover, the rates of transfer of the cations and more precisely calcium and magnesium

ions undergo reductions in the presence of anions and in particular sulfate ions. The desired

rates of desalination are also obtained for longer handling times which have an immediate

effect on energy consumption. In fact, the consumption of energy is more important for more

complex solutions.

Chapter 11 - This study covers the predicted effects of climatic factors during convenient

intervals (in days) on cotton flower and boll production compared with daily observations.

Evaporation, sunshine duration, relative humidity, surface soil temperature at 1800 h, and

maximum air temperature, are the important climatic factors that significantly affect flower

and boll production. The least important variables were found to be surface soil temperature

at 0600 h and minimum temperature. The five-day interval was found to be more adequately

and sensibly related to yield parameters. There was a negative correlation between flower and

boll production and either evaporation or sunshine duration, while that correlation with

minimum relative humidity was positive. Higher minimum relative humidity, short period of

sunshine duration, and low temperatures enhanced flower and boll formation.



In: Advances in Environmental Research. Volume 32 ISBN: 978-1-63117-329-5

Editor: Justin A. Daniels © 2014 Nova Science Publishers, Inc.

Chapter 1

IMPROVED GAS AND VAPOR DETECTION USING

DENUDER TECHNOLOGY

Amanda L. Jenkins, Ph.D.

ASK Inc., Hebron, MD, US

ABSTRACT

An improved method for accomplishing air/vapor sample collection and

transport has been achieved through the use of denuder technology. Denuders use gas

injection technology to continuously introduce a gas or aerosol into a solvent stream

where the analyte is taken into a “trapping” solution preselected for a specific

analyte. This extraction has been demonstrated to have greater than 96% efficiency.

Unlike extraction or bubbler methods denuder extraction takes place in real time and

with less loss. Once the materials are in solution they can be transported to many

different types of analytical detection platforms including LC-MS and ion selective

electrodes. This chapter describes several different applications of the denuder to

enhance gas/aerosol detection as well as a novel vapor generation system to provide

a means to challenge the detection methods.

INTRODUCTION

A large number of analytical methods require gas and aerosol concentrations to be

measured in real-time using validated procedures. Unfortunately, many of these methods have

only been developed for aqueous samples. Getting the air/vapor sample into a liquid vehicle

for detection is time consuming, highly problematic and often has significant loss. One way

of solving this problem is to use impingers, which are glass bubble tubes used to collect

chemicals such as acids and anhydrides. A known volume of air is actively pumped through

an impinger, which contains a liquid medium, causing the liquid to bubble. The liquid

chemically reacts with, or physically dissolves, the chemical being sampled. The liquid is

ASK Inc. 7447 Rockawalkin Rd, Hebron, MD 21830. Tel: 410-543-7642. E-mail: [email protected].


Amanda L. Jenkins 2

then analyzed to determine the airborne concentration. However, there are many issues

associated with using impingers including the difficulty of use, the time required, and the lack

of reproducibility making them less than optimal [1-3]. Instrumentation like GC-MS does

have the capability of detecting gas samples directly without the need for introduction into a

liquid vehicle, but getting the sample to the instrument provides its own set of problems. For

example, sampling of many gases through tubing can lead to severe sample losses from the

material sticking to the walls of the tubing.

Typically, air samples are ported to the instrumentation via heated transfer lines and

special tubing to address this issue, but line losses can still be significant. If the line is too hot,

the analyte can break down, and if it is too cold, the material being sampled will tend to

adsorb to these regions. A novel method for accomplishing air/vapor sample collection and

transport has been achieved through the use of denuder technology. This technique was

adapted from a method developed by Steve Hoke and coworkers at the Aberdeen test Center

to detect acid gases in fire fighting applications [4- 6] Denuders are similar to continuous

impingers using gas injection analysis technology (similar to flow injection analysis) to

continuously introduce a gas or aerosol sample into a solvent stream where the analyte of

interest is taken up into a “trapping” solution. The trapping solution is delivered to the

denuder using a chemical pump, i.e., a solenoid, syringe, peristaltic, rotary or high-

performance liquid chromatography (HPLC) pump. Airflow on the order of 1-10 L/min is

obtained using a mass flow controller (rotometers or critical orifice) and a vacuum source.

This vacuum source, located external to the analyzer, draws sample air into the denuder

where the chemical agent gases and aerosols are immediately extracted into the liquid phase

of the trapping solution to be detected. The solution can then be discarded or re-circulated to

continue collection and provide a cumulative detection scheme.

Another challenge in doing gas analysis and detection is how to generate standards to

evaluate the analytical methods. Traditional means of generating chemical vapors such as

standard saturator cells and aerosol generators (which could be combined with heat and air to

create vapors), are not always suitable for compounds that may be temperature-sensitive, have

a low volatility, or have a high cost of custom synthesis. This type of chemical is often cost-

prohibitive and/or difficult to obtain in quantities large enough to operate standard saturator

cells or other aerosol–vapor types of generators. Additionally, many of these materials can be

highly toxic, which makes working with them in large quantities undesirable.[7] A novel

approach was developed to address these issues and create a stable high- flow vapor generator

for low-volatility materials that require only a minimum amount of chemical consumption

(milligrams). The technique was based on the science of static-coating capillary columns. [8]

Like methods for column preparation, a packed bed of inert particles was filled with a

solution of the test chemical. As the volatile solvent was stripped away under vacuum, the test

chemical remained behind as a film, coating the contacted surfaces, and a generator was

created when a carrier gas flowed through the packed bed. In this manner, an operator can

predetermine the amount of chemical needed to conduct the required testing. With well-

established engineering principles used in the packed-bed design, [9] the resulting generator

could be scaled to meet the flow-rate requirements of the test. When an inert carrier gas is

used, and the temperature of the bed is constant, the vapor–chemical film-coating equilibrium

is not exceeded at the outlet end of the generator, and the vapor generator will produce stable

output-chemical concentrations based on the vapor pressure of the test chemical. The output


Improved Gas and Vapor Detection Using Denuder Technology 3

concentration will remain constant until the chemical is either consumed or the dynamic

factors change [10].

This chapter describes the preparation and evaluation of packed bed vapor generators for

validation of air detection methods. It also describes several diverse applications of the

denuder to demonstrate how it can help solve some important gas/aerosol detection problems.

In the first application, the denuder is coupled to an LC-MS for rapid high throughput

detection samples. In the second method, the denuders are coupled with ion selective

electrodes (ISEs) for acid gas detection. Lastly the use of denuders with molecularly

imprinted polymer sensors (MIPs) will be discussed. Detection limits in all the examples are

dependent on the sensitivity of the analytical platform.

VERIFICATION OF DENUDER COLLECTION EFFICIENCY

Extraction efficiency of the denuders was evaluated using a double-needle atomizer that

sprayed a solution of 25 mg/mL of sodium fluorescein in PEG200. Two aerosol filters were

installed on the upstream side of the denuder and one filter was installed on the downstream

side. Fluorescein aerosol samples were generated and collected for 15 min. Sample flow rates

were varied between 2-3 L/min. The filter pads were removed after each test and saved for

analysis. The fluorescein was extracted from the pads using 3 mL of phosphate buffered

saline solution (PBS). The samples were evaluated against a series of freshly prepared

standards using a Sequoia-Turner Model 450 fluorometer. (Sparks, NV) The extraction

efficiency of the denuder was determined by comparing the fluorescein captured on the

upstream and downstream aerosol pads. Particle size distribution was also determined for

each run. An example of this is given in Figure 1. The replicate data showed that a flow rate

between 2 and 2.5 was optimal for denuder extraction efficiency. Flows below 2 L/min were

too low to provide the necessary amount of air needed for extraction and flows at 3 and above

were too high. The extraction efficiencies using the denuder were determined to be between

92.5 and 96.8 % with the highest extraction efficiency at a flow rate of 2ml/min. [11].

Figure 1. The particle size distribution of the aerosols generated for the test.


Amanda L. Jenkins 4

VAPOR GENERATOR PREPARATION

The vapor-generation techniques were validated using VX. VX was chosen since its

vapor pressure is lower than many compounds, so if the method worked for VX it should

work even better for higher volatility materials. The high flow (liters per minute) generator

was constructed using a 40 cm length of 5.0 cm diameter 316 stainless steel pipe, packed with

an inert material, to serve as a structural support for the test chemical. The pipe was filled

with 4 mm borosilicate glass beads supported between two laminated, 100 mesh stainless

steel screens that were machined into the surface of the corresponding flanged O-ring end

caps. Each end cap was ported and equipped with 9.5 mm ball valves (Swagelok Co., Solon,

OH) and sealed with Viton O-rings. The exterior of the pipe was wrapped with a single layer

of 6.4 mm o.d. thin-walled copper refrigeration tubing (Cerro Flow Products, St. Louis, MO)

and insulated with 1.25 cm closed-cell foam pipe insulation. A Neslab RTE-111 temperature

bath (Thermo Neslab, Newington, NH) was connected to the copper tubing to maintain

temperature control of the generator.

Once the vapor generator was assembled, the inner walls of the pipe were rinsed several

times with dichloromethane to remove impurities. The filling solution was prepared by

dissolving milligram quantities of the analyte in approximately 50 mL of dichloromethane.

With all the ball valves open, the resulting solution was transferred into the generator, filling

it from the inlet endcap using a funnel fitted with a two-way, 6.4 mm ball valve and a vertical

length of 6.4 mm o.d. × 3.2 mm i.d. polytetrafluoroethylene (PTFE) tubing that served as a

“sight-glass” during this process, as shown in Figure 2. Repeated solvent rinses were added

through the funnel to wash any residual filling solution into the generator. Solvent was added

until the solution level was within 2 cm of the top retaining screen.

To remove the solvent, the two-way ball valve was closed at the base of the funnel, and a

vacuum was slowly applied to the top outlet end cap using a suitable pressure controller rated

at approximately 1 Lpm flow rate at 100 torr. The gradual reduction from ambient pressure to

200 torr over 20 min allowed the chemical solution to be mixed during the solvent

outgassing. The evaporation was monitored using a surface-mounted thermocouple placed on

the bottom inlet end cap of the generator canister. As the leading quantity of the liquid solvent

evaporated past the thermocouple, that area of the canister cooled. Once the solvent was

completely removed, the end cap warmed to room temperature as shown in Figure 3. The

entire solvent-stripping process took approximately 5 h.

To prepare the vapor generator for use, the filling-tube and funnel assembly were

removed from the bottom inlet ball valve and replaced with the inert-carrier stream hardware

and corresponding MFC used for the ensuing filter test. The vapor generator bed temperature

was set to subambient temperature conditions, and the top outlet end cap was heated

(approximately 20 °C higher than ambient) with a heating tape to avoid condensation of the

test chemical downstream of the generator. The generator was then conditioned for

approximately 24 h under airflow into a waste-scrubber filter to remove high-volatility

impurities and to purge residual test chemical from above the retaining screen. The generator

was integrated into the test system at this point. The airstream entered the base of the

generator and passed through the thin-film coating of the chemically wetted surfaces of the

glass beads, which produced chemical-vapor outlet concentrations equal to the vapor pressure

of the test chemical at the temperature of the bed. The chemical stream then exited the top of



the generator and entered a chemical-intake manifold. The stream was combined with a

premixed humidified airstream and a chemical-free diluent airstream through a static mixer to

produce the target feed concentration that was required for testing.

Figure 2. Schematic of the packed-bed vapor generator.

The output of the VX generator was monitored using a GC. Initial generator output

showed several peaks that corresponded to higher volatility impurities coming from the

sample. However, after 24 h, the only peak observed in the GC results was that of the VX

(Figure 4). This result provided verification that the sample being generated was pure VX.

Figure 5 shows the stable output of VX over several days. The success of the VX testing

provided validation of the concept. Since then, the procedure has been used for various other

materials including other chemical agents and toxic industrial chemicals (TICs). The time

between regeneration of the generator is dependent on the amount of loading on the column,

the flow rate used to deliver the chemical and the vapor pressure of the material.


Amanda L. Jenkins 6

Figure 3. Temperature response resulting from evaporative cooling of the solvent front as it moves past

the sensor. The minimum indicates completion of the solvent evaporation process.

Smaller scale versions of the vapor generator have also been developed for lower flow

applications, typically on the order of 10-100mL/min of flow. These vapor generators have

been prepared for VX, dimethyl methyl phosphonate (DMMP), tributyl phosphate (TBP) as

well as many other materials, using a custom designed plastic coated borosilicate glass

saturator cells filled with 4mm borosilicate beads as shown in Figure 6. (Glassblowers Inc.

Turnersville, NJ). Just like the larger version, once the vapor generators were filled with

beads, the inner walls of the cell were rinsed several times with dichloromethane to remove

any impurities.

Filling solutions were prepared by dissolving 50-100 mg of analyte to be generated (VX,

DMMP, etc) in approximately 60 mL of dichloromethane. The resulting solution was then

transferred into one arm of the saturator cell. Repeated solvent rinses were added to wash any

residual filling solution into the generator. Solvent was added until the solution level was

within 2 cm of the top of the beads [11].

For the smaller version, the solvent was removed by slowly applying a vacuum to the exit

arm of the saturator using a pressure controller rated at approximately 1 Lpm flow rate at 100

torr. The gradual reduction from ambient pressure to 100 torr over 20 min allowed the

chemical solution to be mixed during the solvent out gassing. The evaporation was monitored

visually to ensure the bubble rate in the beads was not violent enough to cause splashing on

the walls.



Figure 4. GC-MS from VX during generation demonstrating purity of the sample.

The entire solvent-stripping process took 4-5 h per cell. Solvents other than

dichloromethane such as methanol have been used when necessary for solubility. The time

required for the stripping process varied based on the solvent.

Figure 5. Output of the VX generator at 16 °C.


Amanda L. Jenkins 8

Figure 6. Schematic of the packed-bed vapor generator.

Once filled, the generators were prepared for use, by placing Swagelok parts over the

inlet and outlet ports. Inert-carrier stream gas (nitrogen) was flowed into the bed using a mass

flow controller (MFC) (Brooks Inst., Hatfield, PA). The vapor generator bed temperature was

set to sub-ambient temperature conditions, and the top outlet end cap was heated

(approximately 20 °C higher than ambient) with a heating tape to avoid condensation of the

test chemical downstream of the generator. The generator was then conditioned for 24-48 h

under airflow into a waste-scrubber filter to remove high-volatility impurities and to purge

residual test chemical from the arm of the generator.

Once conditioned, these smaller generators were placed in a water bath (Thermo Neslab,

Newington, NH) so the temperature could be carefully controlled for testing. Then the

generator was integrated into the test system. The airstream entered the side arm of the

generator and passed through the thin-film coating of the chemically wetted surfaces of the

glass beads, which produced chemical-vapor outlet concentrations equal to the vapor pressure

of the test chemical at the temperature of the bed. The chemical stream then exited the center

arm of the generator and was delivered to the denuder for analysis. Once the generator was

placed in the water bath and connected to the denuder system, the arm carrying the vapor

stream was wrapped in heat tape. Then the entire system was wrapped again in aluminum

foil. These precautions were taken to prevent any of the sample from condensing out in a

“cold spot” before it reached the denuder. The temperature of the water bath was controlled

through the same Labview software that controlled the timings of the collection. Temperature

of the water bath at every set point was verified by an external mercury thermometer. The

nitrogen flow rate through the generator was 100 sccm/min (standard cubic centimeters/min)

for temperatures 30°C and above, and 10 sccm/min for temperatures below 30°C.



DENUDER-LCMS SYSTEM

The denuder-LCMS system was used to analyze concentrations of an analyte from air

samples. [7, 11] The limits of detection and range of compounds depended entirely upon the

LC-MS. The data presented here describes the use of the system to determine vapor pressure

of low volatility materials, but it has also been successfully used to monitor breakthrough of

filter bed systems and as a method for chamber analysis. Vapor sampling (shown in Figures

7-8) was conducted using A-7 denuders (St. John‟s Associates, Beltsville, MD) with 1.25 cm

diameter bodies and 6.4 mm tapered pipe threads. The denuders were fabricated of

chemically resistant Kel-F material. Except where noted, all tubing used to connect this

sampling system was 3.2 mm o.d. × 1.6 mm i.d. PTFE. The trapping solution used was either

HPLC grade acetonitrile (Aldrich) or deionized water depending on analyte solubility. Any

aqueous or organic solvent can be used depending on the analyte. The trapping solution was

transported to the denuder via a Lab Alliance 1500 Series HPLC pump (Thomas Scientific,

Swedesboro, NJ) set to a flow rate of 2.0 mL/min.

Figure 7. The denuder system flow schematic.


Amanda L. Jenkins 10

Figure 8. The denuder sampling apparatus.

Figure 9. Sample transfer schematic.

The pump delivered the solution to the denuder through an inline 500 psi back-pressure

check valve to maintain pumping efficiency via 1.6 mm PTFE tubing that was connected

directly to the base of the denuder. The vacuum return line of the denuder was connected to

the top of a specially designed U-tube bottle (Glassblowers Inc., Turnersville, NJ) fitted with

a three-hole GL45 vacuum cap (Diba Industries, Danbury, CT) to coalesce the returning

trapping solution. Vacuum tubing connected to the opposite top end of the U-tube was used to

siphon off the excess trapping solution, once the U-tube was filled, and to direct it to a glass

waste container that was fitted with a similar two-hole vacuum cap (Diba Industries). The

sample vacuum through each of the denuders (2.0 sLpm) was controlled by a suitable MFC



protected by an in-line filter. The in-line filter was useful in creating a slight pressure drop

between the waste container and MFC to prevent vapor condensation inside of the controller.

A 3.2 mm, two-position, six-port 316 SS switching valve (Valco Instruments Company,

Inc.; Houston, TX), shown in Figure 9, allowed for convenient sampling during testing.

Solenoid pumps with 50 µL displacement (model 130SP2450-1TP, Bio-Chem Fluidics,

Boonton, NJ), operating on a 3 s cycle, were used to pump a portion of the trapping solution

from their respective U-tubes to the stream-selector valve SV1. The selected denuder stream

(output form from SV1) was connected to a similar six-port valve, SV2, and configured as

shown in Figures 9-10.

Figure 10. Second stage of the denuder sample-collection system.

Table 1. LC-MS Instrument Settings

Parameter Test Conditions

Mobile phase 0.1% formic acid in acetonitrile, run at 400 µL/min

Run time 4 min

Tune type zoom scan; range ±2 u, centered on the parent ion of the test

chemical

Due to the high pressure of the LC-MS injection system, a separate HPLC pump was

used to transfer the sample loop (0.5 mL) contents from the selected denuder sample to the

(25 µL) LC-MS injection loop. The same trapping solution that was used in the denuder

sample collection was used as the mobile phase (0.40 mL/min) for the HPLC pump to fill the

injection loop of the LCQ Advantage Max LC-MS (Thermo Finnegan, San Jose, CA) for

analysis. The LC-MS mobile phase was 0.1% formic acid of 98+% purity (Sigma Aldrich, St.

Louis, MO) in the same solvent as the trapping solution (water, acetonitrile, etc). All valve-

switching, temperature controls on the water bath, and start signals were provided through an

automated LabView (National Instruments, Austin, TX) process-control program that was



written in-house. The filling of the LC-MS analysis loop was timed to receive the center cut

of the denuder sampling loop of SV2. Data was calculated and collected using the Thermo

LC-MS software that was available with the instrument.

The denuder-LC-MS method was validated by performing vapor pressure measurements

and comparing those with vapor pressure measurements collected using the dynatherm

method. [12-13] VX was selected as the material for comparison as that material had been

well characterized in our lab. A small VX generator was prepared and placed in the water

bath. (Thermo Neslab, Newington, NH). The nitrogen flow rate through the generator was

100 sccm/min (standard cubic centimeters/min). Temperature was varied between -10°C and

30°C using the water bath. Temperature in the bath was verified with a mercury thermometer.

Concentration values at each temperature were collected every 5 minutes for 4 hours

(approximately 48 data points per temp). The last 15 points were averaged to give the value

for each temperature. Each experiment consisted of 6 temperatures, with a set of standards

run in between each experiment. This ensured that instrument reproducibility would not affect

the results. The entire process of temperature measurements and standards was repeated 4

times to get a statistical average. Five freshly prepared standards in trapping solution

(acetonitrile) were used for each analysis. The standards were made from stock solution in

trapping solution also prepared fresh daily. The fresh standards were used to prevent any bias

from degradation of the sample over time. Each standard was measured at least 24 times and

the last 12-15 measurements were averaged for each calibration point. Figure 11 shows an

example of the VX calibration curve.

Figure 11. Example of the VX calibration curve.



Figure 12. Comparison of the VX vapor pressure from several methods.

Using the concentration data at each temperature, the vapor pressure of VX at each

temperature was calculated. This data was then plotted and compared with existing data as

shown in Figure 12. [12-13] The denuder-LC-MS method gave values comparable to the

existing dynatherm method and was accepted as an alternate but equally acceptable method.

Since this validation, the vapor pressure of many more materials has been evaluated. Many of

the materials were unable to be measured using any other method due to thermal instability,

low vapor pressure or other factors. The addition of the denuder-LC-MS method has provided

a valuable tool in the measurement of vapor pressure for difficult materials.

THE DENUDER ION SELECTIVE ELECTRODE (ISE) SYSTEM

Vapor sampling for acid gases using ions selective electrodes (ISEs) was also conducted

using specially designed A-7 denuders (St. John‟s Associates, Beltsville, MD) with tapered

pipe threads.[14-15] Except where noted, all tubing used to connect this sampling system was

3.2 mm o.d. × 1.6 mm i.d. PTFE. The trapping solution used for in this study was a pH 7.6

ionic strength adjuster buffer (TISAB). The trapping solution was prepared fresh weekly by

dissolving 54.4 g of potassium hydrogen phosphate (KH2PO4)[0.2 M], 24 g of sodium

chloride (NaCl) [0.2 M] (ACS certified from Fisher Scientific Company, Fair Lawn, NJ

07410) in 1000 mL of distilled water. Once a clear solution was obtained, 22-23 g of

potassium hydroxide (KOH) (ACS certified from Fisher Scientific Company, Fair Lawn, NJ

07410) was added, and the pH adjusted to 7.6 by the slow addition of 1.0 M KOH. Twenty

four milliliters of a 20:1 Tween-20R solution (Fisher Scientific, Fair Lawn, NJ 07410) was

added as a nonionic surfactant wetting agent to suppress the electrical noise produced by the

electrode surfaces. To retard biological growth, 250 ppm chlorine (8 mL of 5% Bleach) was

also added. The volume was then diluted to the 2L mark of a volumetric flask. A schematic of

the denuder-ISE setup is shown in Figure 13.



The trapping solution was transported to the denuder via a Lab Alliance 1500 Series

HPLC pump (Thomas Scientific, Swedesboro, NJ) set to a flow rate of 2.0 mL/min. The

pump delivered the solution to the denuder through an inline 500 psi back-pressure check

valve to maintain pumping efficiency via 1.6 mm PTFE tubing that was connected directly to

the base of the denuder. A vacuum source connected to a mass flow controller (Brooks Inst.,

Hatfield, PA) was used to draw sample air into the denuder at a rate of 2 sLpm, where it

combined with the trapping solution, flowed through the flow cell (in the bottom and out the

top), and out into a vacuum bottle fitted with a two hole screw cap (Diba Industries, Danbury,

CT). The liquid portion was removed through tubing placed into one of the holes. The

sampled air then passed through a particulate filter to reduce the pressure of this saturated gas

stream to avoid condensation and protect the mass flow controller (MFC) before the vapor

stream was vented. Fluoride ion selective electrodes (ISEs) (Thermo-Orion, Beverly, MA)

were inserted into this flow stream by wetting the outer surface of the electrode barrel with

trapping solution to help the electrode slide through dual sealing o-rings of the flow cell until

it touched the rim at the bottom of the flow cell. Output from the ion selective electrodes was

collected into a LabView program and converted into concentration using the calibration

curve, an example of the data output is shown in Figure 14.

A complete calibration study to verify the performance of the denuder–ISE system was

conducted using six standards in the concentration range of 0.1 to 1000ppm. After this, a

three-point calibration was conducted at the start of each day. Standards were prepared fresh

daily by dilution of sodium fluoride in the trapping solution. Standards were analyzed by

pumping the calibration solution through the denuder in an effort to eliminate any bias.

Calibration was performed by turning on the gas analyzer and pumping trapping solution

through the flow cell for about 5 min. Then flow control valve was switched from the

trapping solution to the low level standard. Once 10-15 stable readings had been collected, the

process was repeated with the other standards. Once the data was collected, the electrode

response in millivolts (mV) at each concentration (an average over the stable readings) was

plotted against the log of the concentration in Molarity (M). At the end of the day the HPLC

pumps and the vacuum supply were turned off and the electrodes were removed from the

system and stored in a solution of 1x10-2

to 1x10-3

M NaF. HF calibration curves for 3

different electrodes are shown in Figure 15. The HF denuder method demonstrated excellent

reproducibility and reliability for near real-time measurement of HF gas. Trapping solution

flow rates of 2 mL/min and an air flow rate of 2 L/min were determined to be optimal for this

setup and they also provided convenient calculation values. Initially, the trapping solution

was prepared without the bleach but significant bacterial growth was seen in the trapping

solution after a few days. The bleach proved to be successful in retarding the growth. Fresh

standards and trapping solutions were prepared on a weekly basis to eliminate concerns over

sample degradation. Subsequent studies have been performed using the denuder with

chlorine, chloride, and bromide ion selective electrodes. [14] The trapping solutions vary

somewhat depending on the electrode but all had similar performance characteristics to the

one presented here.



Figure 13. Scematic of the denuder-ISE setup.

Figure 14. Example of how the electrode responds to changes in concentration.



Figure 15. Typical Calibration curves for the HF electrodes.

DENUDER BASED IMPRINTED POLYMER SENSOR

Molecularly imprinted polymers (MIPs) are synthetic materials made using the molecular

imprinting technique. MIP cavity sites are produced so that they only recognize the

corresponding target molecule based on its size, shape, and chemical functionality. Molecular

imprinting is often called “lock and key” technology. In essence, the polymer matrix becomes

an artificial lock for a specific molecule that serves as the key. This process mimics many

basic biological functions from signaling between nerve and muscle cells to antibody-antigen

reactions. One of the greatest advantages of artificial receptors over naturally occurring ones

is freedom of molecular design. These artificial receptors are also much more chemically and

thermally stable than their biological counterparts. Functionality and responses can be

provided by using appropriate functional groups and/or counter ions like metals. These

polymers have an enhanced affinity for the original molecule and have been used in

applications such as chemical separations, catalysis, or molecular sensors.

The MIP-denuder work was initially undertaken to see if existing MIP sensors designed

for the detection of chemical agents in aqueous environments could be coupled with the

denuder to allow detection of gases and aerosols. Previous attempts to use the MIP sensors as

a standalone air detection method had been unsuccessful. Pinacolyl methylphosphonate,

(PMP), the hydrolysis product of the nerve agents sarin and soman was chosen as the target

since that MIP sensor has been well characterized in prior publications. [16-21] Initial tests

were conducted to verify that the PMP sensors prepared for this work were performing as

well as previously reported.

The PMP-MIP was synthesized by preparing complexes of the target using a

stoichiometric ratio of 1:1:3 europium, target (PMP) and monomer, vinyl benzoate. (The

number of binding species was chosen to accommodate the nine coordinate Eu (III).)

Europium nitrate (Eu (NO3)3) was prepared by dissolving europium oxide in water with just

enough nitric acid to produce a clear solution. The PMP was diluted and dissolved in a 50/50



water-methanol mixture to which the vinyl monomer was subsequently added. The resulting

solution was added to the Eu (NO3)3 and the pH adjusted with sodium hydroxide to a pH of 9

and 10 for complexation. The resulting solutions were stirred approximately 2 hours, then

covered with a watch glass and left to crystallize overnight. The crystals were then removed

and the spectra interpreted.

Once the complex was made, polymeric coatings were prepared by dissolving 3 mole

percent complex compound in styrene with approximately 0.1 mole percent

azobis(isobutyronitrile) (AIBN) added as an initiator, and 2 mole percent divinyl benzene

(DVB) added as a cross linking agent. The resulting solution was placed in a glass vial,

purged with nitrogen, and sealed using parafilm and screw-on tops. The solution was

sonicated at approx. 60C until it became viscous. The fiber optic sensors consisted of a

400micron multimode optical fiber (Thor Labs, Newton, NJ) with the polymeric sensing

element chemically bound on the distal end. The fibers were prepared by terminating one end

with an SMA connector. The end to be coated was tapered by heating it in an air/acetylene

flame and manually pulling the stripped end. (Tapered fibers are much more efficient at

coupling the evanescent field to the polymer.) The tapered fiber tips were dipped into the

chemically initiated viscous copolymer leaving a uniform layer on the fiber. The polymer

finished curing, under a small UV lamp, overnight.

Once cured, the polymers were swelled in water with gradually increasing amounts of

methanol to remove unreacted monomer and expand the polymer pores which produces

accessible sites and facilitates the removal of the imprinting molecules. The imprinted

molecules were removed by washing with 1.0M nitric acid which leaves a weakly

coordinated nitrate ion in the cavity. Template removal was verified spectroscopically.

Measurements for the calibration data and response time for all tests were performed

using the same fiber to demonstrate the reversibility of the sensors and maintain continuity.

For the first test, the large monochromator laser system was used. The MIP fiber was inserted

directly into each solution and spectra collected at one minute intervals. The response leveled

off after about 11 minutes and that was taken to be our response time. A calibration curve was

generated based on the 11minute data. Analytical figures of merit were calculated and the

sensor was determined to have a detection limit of about 1ppt (0.7ppt was the reported LOD

in the previous work). Detection limit is calculated as 3sigma or 3 times the standard

deviation of the blank. Standards were analyzed in the order of both increasing and

decreasing concentration in order to demonstrate the reversibility of the sensor. This response

was taken as the baseline measurement for the technology and used for comparison with the

denuder.

The detection platform was built on commercially available off the shelf (COTS)

technologies. Polymer luminescence was excited using a model 60X-argon ion laser (Melles

Griot, Carlsbad, CA). A 488 nm holographic filter (Kaiser Optical Systems, Ann Arbor,

Michigan) turned to pass the 465.8 nm line, was used to exclude all other laser lines. Spectra

were collected using an f/4, 0.5m monochromator (CVI, Albuquerque, NM) equipped with a

Model ST-6 CCD (Santa Barbara Instruments Group, Santa Barbara, CA) using Kestrel Spec

Software (K&M Co., Torrance, CA, USA). Spectra were also collected with a StellarNet Blue

Wave miniature fiber optic UV-vis Spectrometer (StellarNet Inc, Tampa, FL).

It has an integrated thermo electric cooler (TEC) to stabilize references and allow long

detector integration times. The spectral range is 500-850nm with a single element holographic

concave grating that produces a flat field image for uniform resolution. The grating is



aberration corrected for spectral imaging. The detector has a 25um slit for 1nm resolution,

and runs on Spectra Wiz software. A second StellarNet system identical to the one described

above but also containing an integrated 15 Watt Deuterium light source with 4000 hour

Hamamatsu L2D2 bulb, and 0.5inch diameter filter to allow only the 470-490nm light

through was also evaluated.

The MIP-Denuder system (Figure 16) combined the MIP setup (either the large bench top

version or the miniature StellarNet setup) with the denuder. [22] The trapping solution (water

adjusted to pH =9.5 with 1M NaOH) was delivered to the denuder at a flow rate ranging from

2 mL/min using an HPLC pump as previously described. Airflow on the order of 2 L/min was

obtained using a vacuum source that drew sample air with trapping solution into the denuder

where the target gases and aerosols were immediately extracted into the trapping solution.

The resulting mixture was then passed over the MIPs coated fiber that senses the captured

CWA/NTA. A series of 6 standards were incorporated into the system to provide an internal

calibration check.

Once the sensor and detector were determined to be working properly the denuder study

was undertaken. The MIP was placed into “T” connector so the effluent from the denuder

would pass over it. The standards were then pumped through the denuder and over the MIP

inline. The response from the MIP was collected at one minute. In this phase of the work,

maximum response was seen in 9 minutes (so the denuder decreases the response time).

Detection limits for the sensor were calculated based on the 9 minute data. (Figure 17) The

detection limit for the standards through the denuder was calculated to be 54ppt. This rise in

detection limit was believed to be a result of the fast flow rate over the MIP which possibly

decreased the contact time. Several different air and liquid flows were evaluated to see if the

detection limit was influenced by the flow. Lower flows did lower the detection limit of the

MIP sensor but were at a level that they decreased the efficiency of the denuder.

Figure 16. The MIP-Denuder system.



Figure 17. Results from the denuder calibration testing with Pinacolyl methylphosphonate.

Figure 18. Headspace analysis of empty PMP bottle. Spectra were collected every 2 minutes until it

reached steady state at 10 minutes.

In an attempt to the improve the contact time of the solution with the fiber without

compromising the denuder, a specially designed glass U shaped tube was developed to collect

and hold the solution near the MIP for a slightly longer time. (See Figures 8-9) This design

was then incorporated into the LC-MS system setup as well. Testing with this tube in the

system improved the detection limits of the system to make the MIP-denuder system as



sensitive as the MIP system alone is. In an effort to verify the ability to detect vapors, a

headspace study was conducted. The intake portion of the denuder system was placed over an

empty (but not cleaned) bottle of the PMP. Signals were collected every minute using the

large scale system. Positive detection of the PMP vapor was seen within the first minute of

analysis and the maximum signal was detected within 7 minutes. This represents the first time

that this type of MIP sensor previously used for liquid samples has ever detected a vapor.

(Figure 18) The amount of PMP in the headspace was calculated to be 0.9 ppm. Headspace

analysis measurements were also conducted on the same sample using the Stellarnet

miniature spectrometer. Using the miniature fiber optic system the headspace concentration

was determined to be 0.8 ppm.

The miniature spectrometer system works almost as well the large monochromator

system with approximately a 10-15% loss in sensitivity. In most cases, the convenience and

portability of the miniature spectrometer outweigh the slight loss in sensitivity.

CONCLUSION

This chapter presented three completely different applications of the denuder to improve

the process of getting a gas/aerosol sample into a liquid vehicle for sample transport and

detection as well as a new method for generating high and low flow calibration standards to

evaluate the methods. Coupling the denuder with the LC-MS allowed real time detection of

analytes for monitoring and breakthrough processes. It also provided a means of determining

vapor pressure for materials that had been previously difficult to work with due to

temperature instability and or low vapor pressure. In the hydrogen fluoride work, the denuder

ion selective electrode method was sensitive enough to provide detection of the breakthrough

in filter test systems as well as handle the high concentrations required for feed monitoring.

The methods have also been used with hydrogen chloride (HCl) and hydrogen bromide (HBr)

electrodes but should work well with any ion selective electrode.

The denuder in the molecularly imprinted polymer (MIP) system allowed the sensors to

detect air/vapor samples previously undetectable by the method. This represented the first

time a chemical agent has been able to be detected in liquid and gas/vapor using a single

molecularly imprinted polymer platform. These polymers have been demonstrated to have

unprecedented sensitivity for over 30 chemical agents and related compounds to date and

more are being added all the time. The applicability of this technique to air and vapor opened

up new, more sensitive methods for monitoring acid gases for a wide range of applications.

The addition of a small relatively inexpensive device, the denuder, has improved the

detection capabilities of methods. It reduces the time and difficulties in using traditional

sample extraction techniques and serves as a method of pre-concentration. Denuders are real-

time, and can be used in a variety of applications and with a wide range of analytes since they

are chemically resistant and thermally stable. The technology could be modified and used

with almost any type of instrumentation requiring a liquid sample not just the ones described

in this chapter. Development of the denuder technique has led to an overall improvement in

the way we sample and detect gases and aerosols.



REFERENCES

[1] White, J,; Ann. Occup. Hyg. 2006, 50(1), p.15-27.

[2] Richards, J.; Holder, T.; Goshaw, D.; “Optimized Method 202 Sampling Train to

Minimize the Biases Associated with Method 202 Measurement of Condensable

Particulate Matter Emissions” Proceedings from Hazardous Waste Combustion

Specialty Conference St. Louis, MO Nov. 2-3, 2005.

[3] Linak, William P.; Ryan, Jeffrey V.; Ghorishi, Behrooz S.; Wendt, Jost O. L.; “Issues

Related to Solution Chemistry in Mercury Sampling Impingers” Journal of the Air &

Waste Management Association; 2001, Vol. 51(5), p.688.

[4] Hoke, S.H.; Analytica Chimica Acta, 2002, 460, p.219-225.

[5] Hoke, S.H.; Herud, C.; Proceedings of the Halon Alternatives Technical Working

Conference, 11-13 May 1993; Albuquerque, NM, 1993, p.185-190.

[6] Hoke, S.H.; Arnold, M.F.; Holmes, T.D.; Proceedings of the Halon Options Technical

Working Conf., 14th, 4-6 May 2004; NIST SP-984-2; Albuquerque, NM, 2004; p. 1-7.

[7] Jenkins, Amanda L., Ellzy, M.W.; Buettner, Leonard C.; Bruni, Eric J., “New Approach

for Evaluating Military Filter Performance with Low Volatility Chemical Vapors Using

a Denuder-LC/MS Technique” ECBC-TR-1094; U.S. Army Edgewood Chemical

Biological Center: Aberdeen Proving Ground, MD 2013.

[8] Grob, K. Making and Manipulating Capillary Columns for Gas Chromatography,

Huethig: Heidelberg, 1986.

[9] Geankoplis, C.J. Transport Processes and Unit Operation, 2nd ed.; Allyn and Bacon,

Inc.: Boston, 1983, p. 371–414.

[10] Buettner, L.C. Solid State Vapor Generator, U.S. Patent 6,722,182 B1, 2004.

[11] Jenkins, A.L.; Buettner, L.C. “Vapor and Aerosol Capture and Transport Using

Denuder Technology”; ECBC-TR-866; U.S. Army Edgewood Chemical Biological

Center: Aberdeen Proving Ground, MD, 2011; UNCLASSIFIED Report.

[12] Tevault, D.E.; Brozena, A.; Buchanan, J.H. Abercrombie-Thomas, P.E.; Buettner, L.C.

“Thermophysical Properties of VX and RVX”; J. Chem. Eng. Data 2012, 57, p.

1970−1977.

[13] Buchanan, J. H..; Buettner, L. C.; Tevault, D. E. “Vapor Pressure of VX”; ECBC-TR-

068; U.S. Army Edgewood Chemical Biological Center: Aberdeen Proving Ground,

MD, 1999; UNCLASSIFIED Report.

[14] Jenkins, Amanda L.; Buettner, Leonard C.; Maxwell, Amy H.; and Bruni, Eric J.,

“Methodology Assessment for Chlorine and Hydrogen Chloride Gas Analysis” ECBC-

TR-948; U.S. Army Edgewood Chemical Biological Center: Aberdeen Proving

Ground, MD 2012, UNCLASSIFIED Report.

[15] Peterson, G.W.; Buettner, L.C.; Jenkins, A.L. “Hydrogen Fluoride Removal

byImpregnated, Activated Carbons”; ECBC-TR-837; U.S. Army Edgewood Chemical

Biological Center: Aberdeen Proving Ground, MD, 2011; UNCLASSIFIED Report.

[16] Jenkins, A.L.; Murray, G.M. Anal. Chem.; 1996, 68(17), p. 2974-2980.

[17] Jenkins, A.; Uy, M.; Murray, G. Anal. Chem. 1999, 71, p. 373-378.

[18] Jenkins, A.; Yin, R.; Jensen, J. Analyst 2001, 126, p. 798–802.

[19] Murray, G.M.; Uy, O.M.; Jenkins, A.L. Polymer Based Lanthanide Luminescent

Sensors for the Detection of Organophosphorus Compounds. U.S. Patent 2008.



[20] Jenkins, A.; Bae, S. Analytica Chimica Acta. 2005, 542(1), p. 32-37.

[21] Jenkins, A.L., Buettner, L.C., Ellzy, M.W. “Molecularly Imprinted Polymer Sensors for

Detection in the Gas, Liquid and Vapor Phase” J. Molec. Recog. Vol. 25 No. 6 p. 330-

335; 2012.

[22] Jenkins, A.L., Buettner, L.C. “Molecularly Imprinted Polymer Sensors for the Near

Real-time Detection of Gases, Aerosols and Vapors”, Patent accepted 2013.




Chapter 2

ENVIRONMENTAL PROTECTION

FROM THERMODYNAMIC PROPERTIES,

THE PERFORMANCE OF REFRIGERATION

CYCLES AND AIR CONDITIONING

Abdeen Mustafa Omer Energy Research Institute (ERI), Nottingham, UK

ABSTRACT

Over the years, all parts of a commercial refrigerator, such as the compressor, heat

exchangers, refrigerant, and packaging, have been improved considerably due to the

extensive research and development efforts carried out by academia and industry.

However, the achieved and anticipated improvement in conventional refrigeration

technology are incremental since this technology is already nearing its fundamentals limit

of energy efficiency is described is „magnetic refrigeration‟ which is an evolving cooling

technology. The word „green‟ designates more than a colour. It is a way of life, one that

is becoming more and more common throughout the world. An interesting topic on

„sustainable technologies for a greener world‟ details about what each technology is and

how it achieves green goals. Recently, conventional chillers using absorption technology

consume energy for hot water generator but absorption chillers carry no energy saving.

With the aim of providing a single point solution for this dual purpose application, a

product is launched but can provide simultaneous chilling and heating using its vapour

absorption technology with 40% saving in heating energy. Using energy efficiency and

managing customer energy use has become an integral and valuable exercise. The reason

for this is green technology helps to sustain life on earth. This not only applies to humans

but to plants, animals and the rest of the ecosystem. Energy prices and consumption will

always be on an upward trajectory. In fact, energy costs have steadily risen over last

decade and are expected to carry on doing so as consumption grows. Refrigerants such as

hydrochloroflurocarbons (HCFCs) are present in the ground source heat pump (GSHP)

systems and can pose a threat to the environment through being toxic, flammable or

having a high global warming potential.

However, new types and blends of refrigerant with minimal negative impacts are

being developed. A correctly fitted system will also greatly reduce the potential for


Abdeen Mustafa Omer 24

leakage, which is why using a professional installer is highly recommended. Significant

CO2 savings can be gained by displacing fossil fuels. Even compared to the most efficient

gas or oil condensing boilers, a well-designed heat pump with COP of 3-4 will reduce

emissions by 30-35%. Further carbon savings can be made if the electricity used to power

the pump comes from a renewable energy source such as photovoltaic or a renewable

electricity tariff. Also, measures can be taken to reduce the impact of pollution from

using grid electricity generated through fossil fuel. For example, one can purchase dual

tariff green electricity from a number of suppliers. However, even if ordinary grid

electricity is used to run the compressor, the system will still produce less CO2 emissions

than even the most efficient condensing gas or oil boiler with the same output. The term

“vapour compression refrigeration” is somewhat of a misnomer, it would be more

accurately described as 'vapour suction refrigeration'. Vapour compression is used to

reclaim the refrigerant and is more aptly applied to heat pumps. Vapour compression

refrigeration exploits the fact that the boiling temperature of a liquid is intimately tied to

its pressure. Generally, when the pressure on a liquid is raised its boiling (and

condensing) temperature rises, and vice-versa. This is known as the saturation pressure-

temperature relationship.

Keywords: Absorption cycles, environment, heat pumps, refrigeration cycles,

thermodynamic

NOMENCLATURE

A surface area (m2)

COP coefficient of performance

Cp specific heat (J kg-1

k-1

)

Cp specific heat at constant pressure (kJ kg-1

k-1

)

E exergy rate (W)

g local acceleration of gravity ((ms-2

)

h enthalpy (j kg-1

)

I irreversibility (W)

m mass flow rate (kg s-1

)

Ns entropy generation number

NTU number of heat transfer units

P pressure (Pa)

R gas constant (j kg-1

K-1

)

Sgen entropy generation rate (W K-1

)

T temperature (K)

To reference temperature (k)

U overall heat transfer coefficient (W m-2

K-1

)

Greek Symbols

τ effectiveness

ψ rational efficiency

ε heat exchanger effectiveness


Environmental Protection from Thermodynamic Properties … 25

ε efficiency

ρ density (kg/m3)

Subscripts

c cold stream

h hot stream

in inlet

max maximum

min minimum

out outlet

Superscripts

cond condenser or cooling mode

abs absorber

cg condenser to generator

evap evaporator

fg fluid to vapour

gen generator

gh high-temperature generator

p pressure energy component

p pump

R refrigerating or evaporator conditions

rhx refrigerant heat exchanger

shx solution heat exchanger

sol solution

t heat transfer energy component

1. INTRODUCTION

One of the most energy efficient methods of domestic heating is to use heat pumps. Heat

pumps use electrical energy to reverse the natural flow of environmental heat from cold to

hot. A typical heat pump requires only 100 kWh of electrical power to turn 200 kWh of freely

available environmental heat into 300 kWh of useful heat [1]. In every case, the useful heat

output will be greater than the energy required to operate the pump itself. Heat pumps also

have a relatively low carbon dioxide output, less than half that of oil, electric and gas heat

production.

Heat pumps for domestic heating are a relatively new concept in Britain; however the

technology is widely used in an industrial capacity. Across Europe, hundreds of thousands of

domestic heat pump units are in use, and the technology is tried, tested and reliable.

Ideally, a refrigerant will have the following characteristics.



Non-toxic - for health and safety reasons.

Non-flammable - to avoid risks of fire or explosion.

Operate at modest positive pressures - to minimise pipe and component weights (for

strength) and avoid air leakage into the system.

Have a high vapour density – to keep the compressor capacity to a minimum and

pipe diameters relatively small.

Easily transportable - because refrigerants are normally gases at SSL conditions they

are stored in pressurised containers.

Environmentally friendly - non-polluting and non-detrimental to the atmosphere,

water or ground.

Easily recycleable.

Relatively inexpensive to produce.

Compatible with the materials of the refrigeration system - non-corrosive, miscible

with oil, chemically benign.

In practice, the choice of a refrigerant is a compromise, e.g., Ammonia is good but toxic

and flammable. R12 is very good but detrimental to the Ozone layer [2]. An air-source heat

pump is convenient to use and so it is a better method for electric heating. The ambient

temperature in winter is comparatively high in most regions, so heat pumps with high

efficiency can satisfy their heating requirement. On the other hand, a conventional heat pump

is unable to meet the heating requirement in severely cold regions anyway, because its

heating capacity decreases rapidly when ambient temperature is below -10oC. According to

the weather data in cold regions, the air-source heat pump for heating applications must

operate for long times with high efficiency and reliability when ambient temperature is as low

as -15oC. Hence, much researches and developments have been conducted to enable heat

pumps to operate steadily with high efficiency and reliability in low temperature

environments [2]. For example, the burner of a room air conditioner, which uses kerosene,

was developed to improve the performance in low outside temperature [3]. Similarly, the

packaged heat pump with variable frequency scroll compressor was developed to realise high

temperature air supply and high capacity even under the low ambient temperature of –10 to –

20oC [4]. Such a heat pump systems can be conveniently used for heating in cold regions.

However, the importance of targeting the low capacity range is clear if one has in mind that

the air conditioning units below 10 kW cooling account for more than 90% of the total

number of units installed in the EU [5].

2. ENERGY EFFICIENCY CONSIDERATIONS

IN HEAT EXCHANGERS DESIGNS

Heat exchangers are devices, designed to efficiently transfer heat, from one medium to

another, i.e., water-to-air, refrigerant-to-air, refrigerant-to-water, stream-to-water. Heat

exchangers are widely used in power engineering, chemical industries, petroleum refineries,

food industries and in HVAC technology. Therefore, heat transfer and the design of heat

transfer equipment continue to be a centrally important issue in energy conservation. With

increasing worldwide awareness of the serious environmental problems due to fossil fuel



consumption, efforts are being made to develop energy efficient and environmentally friendly

systems by utilisation of non-polluting renewable energy sources, such as solar energy,

industrial waste heat or geothermal water. The GSHPs are suitable for heating and cooling of

buildings and so could play a significant role in reducing CO2 emissions. Ground source or

geothermal heat pumps are a highly efficient, renewable energy technology for space heating

and cooling. This technology relies on the fact that, at depth, the Earth has a relatively

constant temperature, warmer than the air in winter and cooler than the air in summer.

2.1. Heat Transfer Mechanisms

Single-phase convection on both sides

Single-phase convection on one side

Two-phase convection on other side

Two-phase convection on both sides

Examples: condensers, boilers, evaporators and radiators (Figure 1).

Naturally, it would be preferred, for comfort reasons that this index would be small,

preferably nil. It may be seen that the variable is directly related to temperature discomfort:

the larger the value of the index, the farthest will inside conditions be from expected

wellbeing. Also, the use of electricity operated air conditioning systems will be more

expensive the higher this variable is. Hence, energy expenditure to offset discomfort will be

higher when comparing two index values; the ratio of them is proportional to the expected

energy savings.

When the external shade blocks the windowpane completely, the excessive heat gains

belong to the lowest values in the set, and the dimensionless index will be constant with

orientation. For the climate conditions of the locality, it can be seen that a naked window can

produce undesirable heat gains if the orientation is especially unfavourable, when the index

can have an increase of up to 0.3 with respect to the totally shaded window.

Figure 1. Diagram of a phase change heat pump: 1) condensator coil 2) expansion valve 3) evaporator

coil and 4) compressor.



2.2. Brief Methodology of Heat Exchanger Design Based on the Log Mean

Temperature Difference (LMTD) and Effectiveness

A heat exchanger is usually referred to as a micro heat exchanger (μHX) if the smallest

dimension of the channels is at the micrometer scale, for example from 10 μm to 1 mm.

Beside the channel size, another important geometric characteristic is the surface area density

ρ (m2/m

3), which is defined as the ratio of heat exchange surface area to volume for one fluid.

It reflects the compactness of a heat exchanger and provides a criterion of classification. Note

that the two parameters, the channel size and surface area density, are interrelated, and the

surface area density increases when the channel size decreases. The exchangers that have

channels with characteristic dimensions of the order of 100 μm are likely to get an area

density over 10 000 m 2/m

3 and usually referred to as μHXs [6-10].

By introducing efficiency (α) in the specific heat exchanger performance equation, the

volumetric heat transfer power P/V (W/m 3) can be expressed as follows:

P= FUA ΔTm= FUA ρ α V ΔTm (1)

P/V = ρ FU ΔTm (2)

)()(

)]()[(

,,,,

,,,,

incinhoutcouth

incinhoutcouth

mTTTTLn

TTTTT

(3)

where, U, ΔTm and F refer to the overall heat transfer coefficient (W/m2 K), the mean

temperature difference (K) and the dimensionless mean temperature difference correction

factor for flow configuration respectively. A is the heat transfer surface area. Note that for a

specific heat exchanger performance, high values of α lead to a corresponding high

volumetric heat transfer power, larger than that of the conventional equipment by several

orders of magnitude. As a result, heat exchanger design by miniaturisation technology has

become a common research focus for process intensification [11].

The main advantages of the μHX design are “compactness, effectiveness and dynamic”.

These properties enable exact process control and intensification of heat and mass transfer

[12]:

Compactness: The high surface area density reduces the volume of the heat exchanger

needed for the same thermal power substantially. As a result, the space and costly material

associated with constructing and installing the heat exchanger could be reduced significantly.

Moreover, the fluid hold-up is small in a μHX; this is important for security and economic

reasons when expensive, toxic, or explosive fluids are involved.

Effectiveness: The relatively enormous overall heat transfer coefficient of the μHXs

makes the heat exchange procedure much more effective. In addition, the development of

microfabrication techniques [13] such as LIGA, stereolithography, laser beam machining, and

electroformation allows designing a μHX with more effective configurations and high

pressure resistance.

Dynamic: The quick response time of a μHX provides a better temperature control for

relatively small temperature differences between fluid flows. The quick response (small time

constant) is connected to the small inertia of the heat transfer interface (the small metal



thickness that separates the two fluids). On the other hand, the exchanger as a whole,

including the “peripheric” material, usually has a greater inertia than conventional

exchangers, entailing a large time-constant. Thus the response of one fluid to a temperature

change of the other fluid comprises two “temperature-change waves”, with very distinct time-

constants. In conventional exchangers, it is possible that the two responses are blurred into

one.

However, the μHXs are not without shortcomings. On the one hand, the high

performance is counterbalanced by a high pressure drop, a rather weak temperature jump and

an extremely short residence time. On the other hand, those fine channels (~100 μm) are

sensitive to corrosion, roughness and fouling of the surfaces. Moreover, the distinguishing

feature of the μHXs is their enormous volumetric heat exchange capability accompanied with

some difficulties in realisation. The μHXs design optimisation lies, on the one hand, in

maximising the heat transfer in a given volume taking place principally in microchannels,

while, on the other, minimising the total pressure drops, the dissipations, or the entropy

generation when they function as a whole system. Moreover, difficulties such as the

connection, assembly, and uniform fluid distribution always exist, all of which should be

taken into account at the design stage of the μHXs. All these make the optimisation of the

μHXs design a multi-objective problem, which calls for the introduction of multi-scale

optimisation method in order to bridge the microscopic world and the macroscopic world

[14]. In recent years, the fractal theory [15] and constructal theory [16] were introduced to

bridge the characteristics of heat and mass transfer that mainly takes place in micro-scale and

the global performance of the heat exchanger system in macro-scale [17].

The concept of multi-scale heat exchanger is expected to have the following

characteristics [18]:

A relatively significant specific heat exchange surface compared to that of traditional

exchangers;

A high heat transfer coefficient, as heat transfer is taking place at micro-scales and

meso-scales;

An optimised pressure drop equally distributed between the various scales;

A modular character, allowing assembly of a macro-scale exchanger from

microstructured modules.

Some difficulties still exist. On the one hand, the properties of flow distribution in such

an exchanger are still unknown [19]. A lot of research work still needs to be done for the

equidistribution optimisation. On the other hand, 3-D modelling of heat transfer for such an

exchanger requires a thorough knowledge of the hydrodynamics and profound studies on

elementary volume (smallest scale micro channels). Finally, maintenance problems for this

type of integrated structures may become unmanageable when fouling; corrosion, deposits or

other internal perturbations are to be expected. Figures 2-4 show the connections of the heat

exchanger, water pump, heat rejection fan expansion valve, and the power supply to heat

injection fan (Figure 5).

Renewable energy is the term to describe a wide range of naturally occurring, and

replenishing energy sources. The use of renewable energy sources and the rational use of

energy are the fundamental inputs for a responsible energy policy. The energy sector is



encountering difficulties because increased production and consumption levels entail higher

levels of pollution and eventually climate changes, with possibly disastrous consequences.

Moreover, it is important to secure energy at acceptable cost to avoid negative impacts on

economic growth. On the technological side, renewables have an obvious role to play. In

general, there is no problem in terms of the technical potential of renewables to deliver energy

and there are very good opportunities for renewable energy technologies to play an important

role in reducing emissions of greenhouse gases into the atmosphere-certainly far more than

have been exploited so far. But there are still technical issues to be addressed to cope with the

intermittency of some renewables, particularly wind and solar. However, the biggest problem

with replying on renewables to deliver the necessary cuts in greenhouse gas emissions is more

to do with politics and policy issues than with technical ones.

Figure 2. Shows the heat exchanger.

Figure 3. Shows the connections of the heat exchanger, water pump, heat rejection fan and expansion

valve.



The single most important step governments could take to promote and increase the use

of renewables would be to improve access for renewables to the energy market. That access

to the market would need to be under favourable conditions and possibly under favourable

economic rates. One move that could help-or at least justify-better market access would be to

acknowledge that there are environmental costs associated with other energy supply options,

and that these costs are not currently internalised within the market price of electricity or

fuels. It could make significant difference, particularly if, appropriate subsidies were applied

to renewable energy in recognition of environmental benefits it offers. Cutting energy

consumption through end-use efficiency is absolutely essential. And this suggests that issues

of end-use consumption of energy will have to come onto the table in the foreseeable future.

Thermal comfort is an important aspect of human life. Buildings where people work

require more light than buildings where people live. In buildings where people live the energy

is used for maintaining both the temperature and lighting. Hence, natural ventilation is rapidly

becoming a significant part in the design strategy for non-domestic buildings because of its

potential to reduce the environmental impact of building operation, due to lower energy

demand for cooling. A traditional, naturally ventilated building can readily provide a high

ventilation rate. On the other hand, the mechanical ventilation systems are very expensive.

However, a comprehensive ecological concept can be developed to achieve a reduction of

electrical and heating energy consumption, optimise natural air condition and ventilation,

improve the use of daylight and choose environmentally adequate building materials. Energy

efficiency brings health, productivity, safety, comfort and savings to homeowner, as well as

local and global environmental benefits. The use of renewable energy resources could play an

important role in this context, especially with regard to responsible and sustainable

development. It represents an excellent opportunity to offer a higher standard of living to

local people and will save local and regional resources. Implementation of the ground source

heat pump systems offers a chance for maintenance and repair services. It is expected that the

pace of implementation will increase and the quality of work to improve in addition to

building the capacity of the private and district staff in contracting procedures. The financial

accountability is important and more transparent.

Figure 4. Shows the connections of the heat exchanger and expansion valve.



Figure 5. Shows the power supply to heat injection fan.

Various passive techniques have been put in perspective, and energy saving passive

strategies can be seen to reduce interior temperature and increase thermal comfort, and

reducing air conditioning loads. The scheme can also be employed to analyse the marginal

contribution of each specific passive measure working under realistic conditions in

combination with the other housing elements. In regions where heating is important during

winter months, the use of top-light solar passive strategies for spaces without an equator-

facing façade can efficiently reduce energy consumption for heating, lighting and ventilation.

The use of renewable energy resources could play an important role in this context, especially

with regard to responsible and sustainable development. It represents an excellent opportunity

to offer a higher standard of living to local people and will save local and regional resources.

Implementation of the GSHPs offers a chance for maintenance and repair services.

An annual sinusoidal ambient temperature profile and an exponentially decaying

sinusoidal temperature profile as a function of depth are assumed. Temperature at any given

depth in a moment can be estimated on the basis of the following equation.

365

2365

2cos

365),(

xttetxT o

xsA (4)

where:

T(x, t) is the soil temperature at the depth (x) and time (t) (oC), Tm is an average soil

temperature (oC), As is the thermal wave amplitude (

oC), x is the depth (m), t is the day of

year (in days, where t=0 at midnight on 31 December), to is the phase constant (days), and α

is the apparent thermal diffusivity (m3/day).

The heat exchanger effectiveness is defined as the ratio of actual heat transfer versus

maximum possible heat transfer. The actual heat transfer may be computed by calculating

either the energy lost by the hot fluid or the energy gained by cold fluid. For counter flow and

parallel flow heat exchangers it is given as:

Q = (mcp) h (Th, in-Th, out)

= (mcp) c (Tc, out-Tc, in) (5)



The maximum possible heat transfer expressed as:

Q = (mcp) min (Th, in-Tc, in) (6)

The minimum fluid may be either the hot or cold fluid, depending on the mass flow rates

and specific heats. For counter flow heat exchanger, the effectiveness is given as:

)(

)(

)(

)(

,,

,,

min incinh

incoutc

p

cp

TT

TT

mc

mc

)(

)(

)(

)(

,,

,,

min incinh

outcinc

p

cp

TT

TT

mc

mc

(7)

For given effectiveness and maximum heat transfer rate, actual heat transfer rate be

obtained from:

Q min)( pmc )( ,, incinc TT (8)

The number of heat transfer units designates the non dimensional heat transfer size of the

heat exchangers is defined as:

NTU = UA/Cmin (9)

Defining capacity rate as the product of mass flow rate and specific heat as:

(mCp) c = Cc and (mCp) h = Ch (10)

According, Cmin and Cmax will be minimum and maximum capacity rate respectively. The

relationship between effectiveness and number of heat transfer units, NTU is given for

counter flow heat exchanger configuration as follows:

max)min/1(maxmin

max)min/1(

)/(1

1)(

CCNTU

CCNTU

eCC

e

(11)

Compared to the LMTD method of analysis of heat exchanger, effectiveness- NTU

method provides a direct solution. LMTD method requires the outlet temperature of both the

streams which is not so in effectiveness- NTU method.

2.3. Thermodynamic Efficiency Analysis of Heat Exchanger

The maximum useful work that could be obtained from the system at a given state in a

specified environment is known as exergy also called available energy. The property exergy

serves as a valuable tool in determining the quality of energy and comparing work potential



of different energy sources or systems. From the second law point of view, a measurement

procedure is required to compare the performance of different processes or equipment.

The type of exergy efficiency called the rational efficiency is defined by Kotas, 1985 as

the ratio of desired exergy output to exergy used:

used

putdesiredout

E

E)( (12)

Ψ is the sum of all exergy transfers from the system, which must regarded as constituting

the desired output, plus any by-product, which is produced by the system. The desired output

is determined by examining the function of the system. Eused is the required exergy input for

the process to be performed which can be expressed in terms of irreversibilities as:

usedE = putdesiredoutE + I (13)

Alternative form of the rational efficiency can be obtained from:

IE

E

used

putdesiredout

)( (14)

From the consideration of the heat exchanger, Kotas considered that the desired exergy

output as the increase of the thermal component of exergy of the cold stream:

ET

cputdesiredoutE (15)

where:

EEET

inc

T

outc

T

c ,, (16)

with reference to equation (12), in which rational efficiency is formulated, the following

identify the required exergy input, Eused as:

EEEEP

h

P

c

T

htused (17)

By using equation (12) and equation (14), the rational efficiency of the heat exchanger is

obtained as:

EEE

EP

h

P

c

T

h

T

h

)( =

IE

ET

c

T

c

(18)

The exergy change of the hot and cold streams can be written with the help of ideal gas

relations as follows:



Eout – Ein = m [hout – hin + To (sout – Sin)]

= mCp (Tout – Tin) – To mCp Ln (Tout/Tin) + mToR Ln (Pout/Pin) (19)

The desired exergy output which is the increase of the thermal component of exergy of

the cold stream is obtained by equation (19) as:

ET

c= )]()[()(

,

,,,

inc

outcpincoutccp

T

TLnToCTTmC (20)

Expressing the outlet temperature in terms of inlet temperature and effectiveness from

equation (7), the above equation becomes:

ET

c= Cc [ ]]1(/1[)(/

,

,min,,min

inc

outcccoutcincc

T

TCCLnTTTCC (21)

The irreversibility, also called exergy destruction or exergy loss, is calculated by exergy

balance and taking the difference between all incoming and outgoing exergy flows given by

Kotas, 1985:

I = ∑ Ein-∑Eout (22)

Another way of calculating the irreversibility can be done by the Gouy-Stodola formula,

in which the entropy generation rate is multiplied by the environmental temperature as:

I = To Sgen (23)

which, in turn can be written in terms of summation of irreversibilities due to temperature

difference between the fluid streams and pressure drop respectively as:

Sgen = )( pmC )(,

,

inc

outc

T

TLn + cpmC )( )(

,

,

inc

outc

T

TLn - (mR) c Ln (Pc, out/Pc, in) - (mR) h

Ln (Ph, out/Ph, in)

(24)

The entropy generation rate in the heat exchanger is formulated using first and second

law statements:

I = I∆T + I∆P (25)

Defining entropy generation number by dividing entropy generation by minimum heat

capacity, i.e., Cmin as given by:

Ns = Sgen/Cmin (26)



This number is second-law „„relative‟‟ of the order concept of the NTU, which is used in

traditional first-law analyses of heat exchangers. Ns represent a high or a low entropy

generation rate depends on the following factors:

The size of heat exchanger Ns that can be economically tolerated;

On the magnitude of the remnant irreversibility, Ns, imbalance;

On the entropy generation levels shown by the other components that make up the

greater system.

Heat exchangers are generally inefficient from an energy conservation point of view

because they have been designed in the past on the basis of low cost that dictates a minimum-

size unit. To achieve the small-size heat exchanger, the temperature difference between the

fluid streams is maximised.

However, the larger is the temperature difference in a heat exchanger, the greater will be

the loss during heat transfer. Also, capacity mismatch is used between the streams to increase

the performance.

3. SOURCES OF HEAVY METALS IN THE ENVIRONMENT

Various industries generate heavy metal containing wastewater includes tanning, battery,

glassware, ceramics, electroplating, fertiliser, mining, paints, photographic industries. The

waste contain heavy metals such as chromium, lead, cadmium, arsenic, copper, iron,

manganese, nickel, mercury, and cobalt among others. The amount and the number of metals

present in any wastewater are related directly to the operations carried out in an industry. For

example, tanneries discharge chromium in wastewater; copper, chromium, zinc, and cadmium

are widely generated form metal plating; the production of electrical equipment and mining,

smelting, and fossil fuel combustion contribute to mercury pollution; and lead is generated

from a number of industrial and mining sources. In most wastewater, the concentration of

heavy metals present is much larger than the safe permissible limits and therefore, they need

to be removed. Table 1 summarises the anthropogenic sources of heavy metal in the

environment.

Magnetic nanoparticles offer tremendous opportunity to treat wastewater containing toxic

metal ions because of their high surface area, high physicochemical stability, multi

functionality, ease in coating, and low-cost of synthesis. Equilibrium isotherm models are

usually classified into the empirical equation and the mechanistic model. The mechanistic

models are based on mechanism of metal ion adsorption, which are able not only to represent

but also to explain and predict the experimental behaviour. Some empirical models for single

solute systems are listed in Table 2.

Desorption is very much necessary when the material synthesis is costly. Regenerability

of loaded magnetic nanoparticles is a key factor for improving the economy of adsorption

process. The economic feasibility of using adsorbent based on magnetic nanoparticles to

remove heavy metals from aqueous solution relies on its regeneration ability during multiple

adsorption/desorption cycle. Elutant must be none damaging to the magnetic nanoparticles,

less costly, eco-friendly and effective. Figure 6 Effect of pH on the adsorption of metal ions.



Table 1. Significant anthropogenic sources of heavy metal in the environment

Industry Metals Pollution arising

Electroplating Cr, Ni, Zn, Cu liquid effluents from plating processes

Batteries Ps, Sb, Zn, Cd, Ni, Hg Waste battery fluid, contamination of soil

and groundwater

Paints and pigments Pb, Cr, As, Ti, Ba, Zn Aqueous waste from manufacture, old

paint deterioration and soil pollution

Landfill leachate Zn, Cu, Cd, Pb, Ni, Cr, Hg Landfill leachate, contamination of ground

and surface water

Electronics Pb, Cd, Hg, Pt, Au, Cr, As, Ni, Mn Aqueous and solid metallic waste from

manufacturing and recycling process

Metalliferous mining Cd, Cu, Ni, Cr, Co, Zn, As Acid mine drainage, tailings, slag heaps

Fertilisers Cd, Cr, Mo, Pb, U, V, Zn Run-off, surface and groundwater

contamination, plant bio-accumulation

Manures sewage sludge Zn, Cu, Ni, Pb, Cd, Cr, As, Hg Land spreading threat to ground and

surface water

Specialist alloys and

steels Pb, Mo, Ni, Cu, Cd, As, Te, U, Zn

Manufacture, disposal and recycling of

metals, tailings and slag heaps

Paper and pulp Zn, Cu, Cd, Pb, Ni, Fe, Mn Wastewater effluents

Table 2. Summary of widely used isotherms for systems with their advantages

and disadvantages

Isotherms Langmuir Freudlich Temkin Redlich-Peterson

Functional

form

qe = Qm KL Ce (1 +

KL Ce) qe = KF (Ce)

1/n qe = (RT/b) In (KT

Ce)

qe = KRP Ce (1+aRP

(Ce)B

Linear form Ce/qe = 1/(Kl qm) +

1/qm Ce

In qe = In KF + 1/n In

Ce

qe = RT/b In KT +

RT/b In Ce

In [(KRPCe/qe)-1] = In

aRP + b In Ce

Plot Ce/qe VS.Ce In qe VS. In Ce qe VS. In Ce In [(KRP Ce)-1] VS. In Ce

Advantages

has Henry law and

finite saturation limit

so valid over a wide

range of

concentration

Simple expression

and has parameter

for surface

heterogeneity

Simple expression Approaches Freudlich at

high concentration

Disadvantages Based on monolayer

assumption

Does not have Henry

law and no

saturation limit, not

structured, not

applicable over wide

range of

concentration

Same as Freudlich.

It does not have

correct Henry law

limit and finite

saturation limit,

not applicable over

wide range of

concentration

No special advantages

The following are the more remarkable ones:

National decisions must be based on scientific knowledge such that the measures are

designed to consider the potential sources of accidental admixture and the likelihood

of contamination from each source.



The measures must be effective, financially feasible and proportional to the labelling

parameters established.

The balance among producers‟ interests must be guaranteed.

The civil liability regulations have to ensure that legal frameworks offer equal and

adequate compensation for economic damages to farmers.

Regarding the latter point, farmers who introduce genetically modified (GM) crops are

responsible for any damages and the implementation of the necessary agricultural

management practices to limit genetic contamination. The recommendation identifies the

primary potential accidental admixture sources including seed impurities, natural regeneration

or seed remaining in the soil after a harvest, cross-pollination, harvest and post-harvest

processes, transport and storage, and the final processing of the products. Similarly, it

complies a catalogue of coexistence measures related with the above-mentioned accidental

admixture sources (Table 3).

0

20

40

60

80

100

120

1 2 3 4 5 6 7 8 9

Inital pH

Rem

ov

al e

ffic

ien

cy (

%)

Zn

Cu

Pb

Cd

Figure 6. Effect of pH on the adsorption of metal ions; adsorbent 0.05 g, concentration ions 100 mg L-1

,

volume of metal ions solution 50 ml, time 2 hr, at 298 K.

Table 3. Classification of the main coexistence measures according to the sources

accidental admixture

Source of admixture Coexistence measures at the farm-level Coexistence measures

at the region-level

Impurities in the

propagation material Certification and registration of GM * seeds

Cross pollination

Isolation distances

Buffer zones

Coordination between neighbours to stagger flowering

periods

Voluntary agreements

among producers

Sowing, harvest and

post-harvest activities

Conditions that enable

Production segregation

Training in good agricultural practices: cleaning machinery, installations, means

of transport, etc.

Notifications to public administrations and neighbours

* GM Genetically modified.



Table 4. Traffic related lead concentration levels in selected African countries

Country Air (μg/m3) Soil (μg/g) Food (μg/g) Water (μg/l) Blood (μg/dl)

Egypt 0.6-4.8 (1.9) N.A. N.A. N.A. 11-36 (19)

Kenya 0.4-1.3 26.73-4000 (105) 0.45-85.5 (10.15) 0.11-19.1 (5.65) N.A.

Nigeria 0.5-45 N.A. 0.01-1.6 (0.1) 0.9-9.8 (4.05) 8.7-60 (30.1)

Senegal N.A. N.A. N.A. N.A. 6.1-10.67 (8.4)

South Africa 0.36-2.1 (0.76) 76.7-80 N.A. N.A. 3.8-12 (9.7)

Uganda N.A. 2.5-703 (25.5) N.A. N.A. N.A.

Zambia 0.15 16-1000050 (1830) 0.4-66 N.A. N.A.

Source: (UNEP, 2006) [20].

0

50

100

150

200

250

300

350

400

450

W Europe E Europe Pacific Africa World

Average

Regions

gC

O2

/km

2000 2020 2050

Figure 7. The GHG emissions intensity of automobile stock by regions.

The growth in vehicle emissions has, to a large extent, diminished Africa‟s carrying

capacity, posing a significant threat to human health, ecosystems on which livelihoods

depend, materials and infrastructure, climate change and biodiversity (Table 4) [20].

Figures in the parentheses are median concentrations.

The reported values are from available literature published in the period 1982-2005.

The total emissions emanating from the regions are projected to surpass those of their

industrialised counterparts, i.e., Western Europe, North America, and Japan (Figure 7)

(WBCSD, 2004) [21].

3.1. Co-Precipitation

Co-precipitation is very facile and convenient way to synthesise iron oxide nanoparticles

(either Fe3O4 or γ-Fe2O3) from aqueous Fe2+

/Fe3+

salt solutions by addition of a base under

inert atmosphere at room temperature or at elevated temperature. The chemical reaction of

Fe3O4 formation may be written as:



Fe2+ + 2Fe3+

+ 8OH- → Fe3O4 + H2O (27)

According to the thermodynamics of this reaction, complete precipitation of Fe3O4 should

be expected at a pH between 8 and 14, with a stoichiometric ratio of 2:1 (Fe3+

/ Fe2+

) in non-

oxidising oxygen environment. The size, shape and composition of magnetic nanoparticles

very much depends on the type of salt used (e.g., chlorides, sulphates, nitrates), the Fe2+

/Fe3+

ratio, the reaction temperature, the pH value and ionic strength of the media. With this

synthesis, one the synthetic conditions are fixed, the quality of the magnetite nanoparticles is

fully reproducible.

However, magnetite nanoparticles are not very stable under ambient conditions and are

easily oxidised to maghemite or dissolved in an acidic medium.

Fe3O4+ 2H+ → γ-Fe2O4 + Fe2+

+ H2O (28)

3.2. Hydrothermal Syntheses

Several scientists have reported the hydrothermal synthesis of magnetic nanoparticles for

wide range of applications. The reaction between Fe(NO3)3 solution and NH3.H2O is shown

in equation:

Fe(NO3)3 + 3NH3.H2O → Fe(OH)3 + 3NH4.NO3 (29)

and α-Fe2O3 particles will form through a two-step phase transformation, according to the

following equation:

Fe(OH)3 + β – FeOOH → α – Fe2O3 (30)

Ethylene glycol will have an effect on preventing Fe(OH)3 the nanoparticles from

agglomeration. Therefore, uniform α – Fe2O3 nanocrystals were obtained after being heated at

200-289oC to complete the phase transformation.

4. THERMODYNAMICS

Thermodynamics is the study of energy, its transformations and its relation to states of

matter. Thermodynamic system is a region in space or a quantity of matter bounded by a

closed surface.

The surroundings include everything external to the system and the system is separated

from the surroundings by the system boundaries. These boundaries can be movable or fixed,

real or imaginary.



4.1. The Pseudo-First Order Kinetic

The Lagergreen‟s first-order rate expression based on solid capacity is generally

expressed as follows:

qqdt

dqek

1 (31)

where q and qe are amounts of adsorbate adsorbed (mg/g) at time t and at equilibrium

respectively, k1 is the rate constant of adsorption (l/min). Integration of the above equation

with the boundary conditions: t=0, q=0 and t=t, q=q gives:

)exp(1 1tkq qe

(32)

Kinetic adsorptions by numerous magnetic nanomaterials have been studied using the

pseudo-first-order kinetic model.

qqekdt

dq

2

2 (33)

Integration of above equation with the boundary conditions t=0, q=0 and t=t, q=q result

in:

)(1

qqe

tkqe

21

)( (34)

This equation can be stated in the linear form as:

)(q

t

qekq

t

e2

2

1)( (35)

where k2 is the equilibrium rate constant of pseudo-first-order adsorption (g/mg/min).

Pseudo-first-order model is derived on the basis of the sorption capacity of the solid

phase expressed as in equation (33).

4.2. The Weber and Morris Sorption Kinetic Model

The Weber and Morris sorption kinetic model was initially employed by Pavasant to

describe their biosorption experimental data. This model has the following form:

tKWMq (36)



The intraparticle diffusion can be estimated with:

D =

qe

dpKWM

2

)8640

(

(37)

The external mass transfer process was determined by:

Cckis

idt

dq

2

(38)

4.3. Freundlich Isotherm

Freundlich isotherm is an empirical equation. This equation is one among the most

widely used isotherms for the explanation of adsorption equilibrium. This equation has the

following form:

CKn

eFeq

/1 (39)

It can also be expressed in the linearised logarithmic form:

eFe Cn

q K log1

log log (40)

By applying these assumptions and a kinetic principle (rate of adsorption and desorption

from the surface is equal), the Langmuir equation can be written in the following form:

)1

(max

eL

eLe

CK

CKq q

(41)

The linear form of this equation is often written as:

)1

(max.

max

1qK

Cq

C

L

e

e

e

q

(42)

Langmuir isotherm constant can be used in biosorption to estimate the thermodynamic

parameters Gibbs free energy (ΔG), change in enthalpy (ΔH) and change in entropy (ΔS). The

negative (ΔG) indicates the spontaneity of the adsorption process. The free energy of

adsorption (ΔG) can be related with the Langmuir equilibrium constant by the following

expression:

LRTLnKG (43)



Enthalpy and entropy changes are also related to the Langmuir equilibrium constant by

the following expression:

RT

H

R

sLK

ln (44)

Thus, a plot of Ln KL versus 1/T should be a straight line. ΔH and ΔS values could be

obtained from the slope and intercept of this plot. Magnetic nanoparticles offer tremendous

opportunity to treat wastewater containing toxic metal ions because of their high surface area,

high physiochemical stability, multi functionality, ease in coating and low-cost of synthesis.

5. STORED ENERGY

Thermal (internal) energy is caused by the motion of molecules and/or intermolecular

forces. Potential energy (PE) is caused by attractive forces existing between molecules, or the

elevation of the system.

PE = mgz (45)

where: m is the mass; g is the local acceleration of gravity; and z is the elevation above

horizontal reference plane.

Kinetic energy (KE) is the energy caused by the velocity of molecules and is expressed

as:

KE = ½mV2 (46)

where: V is the velocity of a fluid stream crossing the system boundary.

Flow work is energy carried into or transmitted across the system boundary because a

pumping process occurs somewhere outside the system, causing fluid to enter the system.

Flow work also occurs as fluid leaves the system (Figure 8).

Flow work (per unit mass) = pv (47)

Enthalpy h is an important property that includes internal energy and flow work and is

defined as:

pvuh (48)

5.1. First Law of Thermodynamics

The first law of thermodynamics is often called the law of conservation of energy. The

following form of the first-law equation is valid only in the absence of a nuclear or chemical

reaction.



Figure 8. Energy flows in general thermodynamic system.

Based on the first law or the law of conservation of energy for any system, open or

closed, there is an energy balance as:

[Net amount of energy added to system] = [Net increase of stored energy in system] (49)

or

[Energy in] – [Energy out] = [Increase of stored energy in system] (50)

For the general case of multiple mass flows with uniform properties in and out of the

system, the energy balance can be written:

systemiiff

outoutin

gzpvumgzpvum

WQgzpvumgzpvu

VV

VV

])2

()2

([

)2

()2

(m

22

22

in

(51)

where subscripts i and f refer to the initial and final states respectively.

Nearly all important engineering processes are commonly modelled as steady-flow

processes. Steady flow signifies that all quantities associated with the system do not vary with

time. Consequently:

0)2

(m)2

(m

2

in

2

in WQigzhgzpvh nmouin

enteringallstreams

VVleavingallstreams

(52)

where: h = u +pv as described in equation (48).

A second common application is the closed stationary system for which the first law

equation reduces to:

Q – W = [m (uf – ui] system (53)



5.2. Second Law of Thermodynamics

The second law of thermodynamics differentiates and quantifies processes that only

processed in a certain direction (irreversible) from those that are reversible. Reducing total

irreversibility in a cycle improves cycle performance. In the limit of no irreversibilities a

cycle attains its maximum ideal efficiency. In open system, the second law of

thermodynamics can be described in terms of entropy as:

dIsmsmT

Qeeii

systemdS

(54)

where: dSsystem is the total change within system in time dt during process; δmisi is the entropy

increase caused by mass entering (increasing); δmese is the entropy decrease caused by mass

leaving (exiting); δQ/T is the entropy change caused by reversible heat transfer between

system and surroundings at temperature T; dI is the entropy change caused by irreversibilities

(always positive).

Equation (54) accounts for all entropy changes in the system, Rearranged, this equation

becomes:

])[( dIdSsmsmTQ systemiiee (55)

5.3. Thermodynamics and Refrigeration Cycles

In integration form, if inlet and outlet properties, mass flow and interactions with the

surroundings do not vary with time, the general equation for the second law is:

ImsmsT

QSS outin

rev

systemif )()()(

(56)

In many applications, the process can be considered to operate steadily with no change in

time. The change in entropy of the system is therefore zero. The irreversibility rate, which is

the rate of entropy production caused by irreversibilities in the process, can be determined by

rearranging equation (56):

surr

inout

T

QmsmsI )()( (57)

Equation (57) is commonly applied to a system with one mass flow in, the same mass

flow out, no work and negligible kinetic or potential energy flows, combining equation (52)

and (57) yields:



surr

inoutout

T

hhsmI

)s( in (58)

In a cycle, the reduction of work producing by a power cycle (or the increase in work

required by a refrigeration cycle) equals the absolute ambient temperature multiplied by the

sum of irreversibilities in all processes in the cycle. Thus, the difference in reversible and

actual work for any refrigeration cycle, theoretical or real, operating under the same

conditions, becomes:

ITWW oreversiblecatual (59)

5.4. Thermodynamic Analysis of Refrigeration Cycles

Refrigeration cycles transfer thermal energy from a region of low temperature TR to one

of higher temperature.

Usually the higher temperature heat sink is ambient air or cooling water, at temperature

To, the temperature of the surroundings. The first and second laws of thermodynamics can be

applied to individual components to determine mass and energy balances and the

irreversibility of components.

Performance of a refrigeration cycle is usually described by a coefficient of performance

(COP), defined as the benefit of the cycle (amount of heat removed) divided by the required

energy input to operate the cycle:

Ne

UeCOP (60)

where: Ue is the useful refrigeration effect, and Ne is the net energy supplied from external

sources.

W

Q

net

evapCOP (61)

In an absorption refrigeration cycle, the net energy supplied is usually in the form of heat

into the generator and work into the pump and fans, or:

QW

Q

gennet

evapCOP

(62)



In many cases, work supplied to an absorption system is very small compared to the

amount of heat supplied to the generator, so the work term is often neglected. Applying the

second law to an entire refrigeration cycle shows that a completely reversible cycle operating

under the same conditions has the maximum possible COP. Departure of the actual cycle

from an ideal reversible cycle is given by the refrigerating efficiency:

COPrevR

COP (63)

The Carnot cycle usually serves as the ideal reversible refrigeration cycle. For multistage

cycles, each stage is described by a reversible cycle.

5.5. Equation of State

The equation of state of a pure substance is a mathematical relation between pressure,

specific volume and temperature. When the system is in thermodynamic equilibrium:

0),,( Tvpf (64)

The principles of statistical mechanics are used to (1) explore the fundamental properties

of matter, (2) predict an equation of state based on the statistical nature of a particular system,

or (3) propose a functional form for an equation of state with unknown parameters of a

substance.

A fundamental equation with this basis is the virial equation, which is expressed as an

expression in pressure p or in reciprocal values of volume per unit mass v as:

...'''132

pp DCpBRT

pv (65)

...///132 vv DCvB

RT

pv (66)

where coefficients B‟, C‟, D‟, etc., and B, C, D, etc., are the virial coefficients. B‟ and B are

the second virial coefficients; C‟ and C are the third virial coefficients, etc. The virial

coefficients are functions of temperature only and values of the respective coefficients in

equation (65) and (66) are related. For example, B‟=B/RT and C‟=(C-B2)/ (RT)

2. The

universal gas constant R‟ is defined as:

T

pvR

T

p

)'(' lim 0 (67)



where (pv‟)T is the product of the pressure and the molar specific volume along an isotherm

with absolute temperature T. The current best value of R‟ is 8314.41 J/(kg mol-1

K). The gas

constant R is equal to the universal gas constant R‟ divided by the molecular mass M of the

gas or gas mixture.

The quantity pv/RT is also called the compressibility factor Z, or:

...///132 vv DCvBZ (68)

The Benedict-Webb-Rubin (B-W-R) equation has been extensively for hydrocarbons:

Tvevv

vvTvca

abRTCARTBvRTP ooo

23)/(26

322

/])/1([/(

/)(/)/()/(2

(69)

eBA

bv

eCBA

bv

eCBA

bv

eCBA

bv

eCBA

av

TckTTckT

TckTTckT

T

TT

TT

bv

RTP

)(

66

5

)/(

555

4

)/(

444

3

)/(

333

2

)/(

222

)(

)()(

)()(

(70)

where the constant coefficients are A1, B1, C1, k, b and a.

The Strobridge equation is accurate within the uncertainty of the measured p-v-T data:

6

15

2

16

1413

2

12

2

16

1110

2

9

543

)(]43

5

)(]43

3

4

8

3

76

2

4221

exp[

exp[

)(

][

nnnn

T

n

nnn

T

n

TnnTRn

nnnnTRnRTP

TT

TT

TTTp

(71)

The 15 coefficients of this equation‟s linear terms are determined by a least-square fit to

experimental data.

5.6. Calculating Thermodynamic Properties

Entropy may be considered a function of T and p and from calculus an infinitesimal

change in entropy can be written as:



dpp

sdT

T

sds Tp )()(

(72)

Likewise, a change in enthalpy can be written as:

dpp

hdT

T

hdh Tp )()(

(73)

At least two intensive properties (properties independent of the quantity of substance,

such as temperature, pressure, specific volume and specific enthalpy) must be known to

determine the remaining properties. If two known properties are (p, v, or T) (these are

relatively easy to measure and are commonly used in simulations), the third can be

determined throughout the range of interest using an equation of state.

Using the Gibbs relation Tds = dh = vdp and the definition of specific heat at constant

pressure, Cp = (δh/δT)p, Equation (73) can be rearranged to yield:

T

dpv

p

hdT

T

Cpds p ])[()(

(74)

Equation (72) and (74) combine to yield (δs/δT)p = cp/T. Then using the Maxwell

relation (δs/δp)T = -(δv/δT)p, Equation (72) may be written as:

dpT

vdT

T

Cpds p)[()(

(75)

This is an expression for an exact derivative, so it follows that:

dpvTp

Cpp

T

T )()(2

2

(76)

Table 5. Thermodynamic property values

State Temperature

(oC)

Pressure

(kPa)

Specific volume

(m3/kg)

Specific enthalpy

(kJ/kg)

Specific entropy

(kJ/(kg K)

1 -20.0 132.73 0.14739 386.55 1.7413

2 2.8 292.80 0.07097 401.51 1.7413

3 0.0 292.80 0.06931 398.60 1.7282

4 33.6 770.20 0.02726 418.68 1.7282

5 30.0 770.20 0.00084 241.72 1.1435

6 0.0 292.80 0.01517 241.72 1.15297

7 0.0 292.80 0.000772 200.00 1.0000

8 -20.0 132.73 0.01889 200.00 1.00434



Table 6. Measured and computed thermodynamic properties of R-22

Measured Computed

State Temperature

(oC)

Pressure

(kPa)

Specific

volume (m3/kg)

Specific enthalpy (kJ/kg) Specific entropy

(kJ/(kg K)

1 -10.0 310.0 0.07558 402.08 1.7810

2 -4.0 304.0 0.07946 406.25 1.7984

3 82.0 1450.0 0.02057 454.20 1.8165

4 70.0 1435.0 0.01970 444.31 1.7891

5 34.0 1410.0 0.00086 241.40 1.1400

6 33.0 1405.0 0.00086 240.13 1.1359

7 -12.8 320.0 0.019010 240.13 1.1561

Table 7. Thermodynamic property data

State t, (oC) p, (kPa) v, (m

3/kg) h, (kJ/kg) s, (kJ/(kg K)

1 -20.0 132.73 0.14739 386.55 1.7413

2 37.8 770.20 0.02798 423.07 1.7413

3 30.0 770.20 0.000842 241.72 1.1435

4 -20.0 132.73 0.047636 241.72 1.1692

Table 8. Energy transfers and irreversibility rates for refrigeration system

Component q, kW W, kW I, W/K I/Itotal (%)

Evaporator 7.000 0 0.4074 9

Suction line 0.1802 0 0.1575 3

Compressor -0.4276 2.5 2.1928 46

Discharge line -0.4274 0 0.2258 5

Condenser -8.7698 0 0.8747 18

Liquid line -0.0549 0 0.0039 ≈0

Expansion device 0 0 0.8730 18

Total -2.4995 2.5 4.7351

Table 9. EU criteria pollutant standards in the ambient air environment

Pollutant EU limit

CO

NO2

O3

SO2

PM10

SO2 + PM10

Pb

Total suspended particulate (TSP)

HC

30 mg/m2; 1h

200 μg/m2; 1h

235 μg/m2; 1h

250-350 μg/m2; 24 h

80-120 μg/m2; annual

250 μg/m2; 24 h

80 μg/m2; annual

100-150 μg/m2; 24 h

40-60 μg/m2; annual

2 μg/m2; annual

260 μg/m2; 24 h

160 μg/m2; 3 h



Table 10. Significant EU environmental directives in water, air

and land environments

Environment Directive name

Water Surface water for drinking

Sampling surface water for drinking

Drinking water quality

Quality of freshwater supporting fish

Shellfish waters

Bathing waters

Dangerous substances in water

Groundwater

Urban wastewater

Nitrates from agricultural sources

Air Smokes in air

Sulphur dioxide in air

Lead in air

Large combustion plants

Existing municipal incineration plants

New municipal incineration plants

Asbestos in air

Sulphur content of gas oil

Lead in petrol

Emissions from petrol engines

Air quality standards for NO2

Emissions from diesel engines

Land Protection of soil when sludge is applied

Thermodynamic property data are summarised in Table 5. Table 6 lists the measured and

computed thermodynamic properties of the refrigerant, neglecting the dissolved oil. The

property data are tabulated in Table 7.

This result is within computational error of the measured power input to the compressor

of 2.5 kW. The analysis demonstrated in Table 8 can be applied to any actual Vapour

compression refrigeration system. The only required information for second law analysis is

the refrigerant thermodynamic state points and mass flow rates and the temperatures in which

the system is exchanging heat.

Technological progress has dramatically changed the world in a variety of ways. It has,

however, also led to developments of environmental problems, which threaten man and

nature. During the past two decades the risk and reality of environmental degradation have

become more apparent. Growing evidence of environmental problems is due to a combination

of several factors since the environmental impact of human activities has grown dramatically

because of the sheer increase of world population, consumption, industrial activity, etc.

throughout the 1970s most environmental analysis and legal control instruments concentrated

on conventional effluent gas pollutants such as SO2, NOx, CO2, particulates, and CO (Table

9). Recently, environmental concerns has extended to the control of micro or hazardous air

pollutants, which are usually toxic chemical substances and harmful in small doses, as well to

that of globally significant pollutants such as CO2. Aside from advances in environmental

science, developments in industrial processes and structures have led to new environmental

problems. For example, in the energy sector, major shifts to the road transport of industrial

goods and to individual travel by cars has led to an increase in road traffic and hence a shift in

attention paid to the effects and sources of NOx and volatile organic compound (VOC)



emissions. Environmental problems span a continuously growing range of pollutants, hazards

and ecosystem degradation over wider areas.

The main areas of environmental problems are: major environmental accidents, water

pollution, maritime pollution, land use and sitting impact, radiation and radioactivity, solid

waste disposal, hazardous air pollutants, ambient air quality, acid rain, stratospheric ozone

depletion and global warming (greenhouse effect, global climatic change) (Table 10). The

four more important types of harm from man‟s activities are global warming gases, ozone

destroying gases, gaseous pollutants and microbiological hazards (Table 11).

Table 11. The external environment

Damage Manifestation Design

NOx, SOx

CO2

O3 destruction

Legionnellosis

Irritant

Acid rain land damage

Acid rain fish damage

Global warming

Rising sea level

Drought, storms

Increased ultra violet

Skin cancer

Crop damage

Pontiac fever

Legionnaires

Low NOx burners

Low sulphur fuel

Sulphur removal

Thermal insulation

Heat recovery

Heat pumps

No CFC‟s or HCFC‟s

Minimum air conditioning

Refrigerant collection

Careful maintenance

Dry cooling towers

Table 12. Refrigerant/absorbent pairs

Refrigerant Absorbents

H2O Salts

Alkali halides

(LiBr, LiClO3, CaCl2, ZnCl2, ZnBr)

Alkali nitrates

Alkali thiocyanates

Bases

Alkali hydroxides

Acids (H2SO4, H3PO4)

NH3 H2O

Alkali thiocyanates

TFE (organic) NMP

E181

DMF

SO2 Organic solvents

5.7. Thermodynamics and Refrigeration Cycles

However, one advantage of absorption cycles is they can maximise benefit from low-

temperature, high-glide heat sources. That ability derives from the fact that the desorption

process inherently embodies temperature glide, and hence can be tailored to match the heat

source glide. Similarly, absorption also embodies glide, which can be made to match the glide

of the heat rejection medium.



Table 12 lists refrigerant/absorbent pairs. Several appear suitable for certain cycles and

may solve some problems associated with tradition pairs. However, information on

properties, stability and corrosion is limited. Also, some of the fluids are somewhat

hazardous.

Integrating this expression at a fixed temperature yields:

T

T

p

o

p dpvTT

CpoC )(

2

2

(77)

where Cpo is the known zero-pressure specific heat and DpT is used to indicate that

integration is performed at a fixed temperature. The second partial derivative of specific

volume with respect to temperature can be used determined from the equation of state. Thus,

equation (77) can be used to determine the specific heat at any pressure.

Using Tds = dh-vdp, and equation (76) can be written as:

dpT

vTvdTCdh pp )([

(78)

Equations (74) and (78) may be integrated at constant pressure to obtain:

dTpT

CpPoTosPoTs

T

To

)(),(),1(

1

(79)

and

dTCpPoTohPoThT

To

1

),(),1( (80)

Integrating the Maxwell relation (δs/δp)T = - (δv/δT)p gives an equation for entropy

changes at a constant temperature as:

dpTT

vPoTosPTs

p

po

1

)(),()1,0(

(81)

Likewise, integrating equation (78) along an isotherm yields the following equation for

enthalpy changes at a constant temperature:

1

])([),()1,0(

p

po

p dpT

vTvPoTohPTh

(82)

Therefore, the Clapeyron equation is of particular value; for equation evaporation or

condensation, it gives:



)()()(fg

fgsat

fg

fg

Tv

h

v

s

dT

dp (83)

where: Sfg is the entropy of vaporisation; hfg is the enthalpy of vaporisation; and vfg is the

specific volume difference between vapour and liquid phases.

5.7.1. Phase Equilibria for Multi-Component Systems

To understand phase equilibria, consider a container full of a liquid made of two

components; the more volatile component is designated i and the less volatile component j

(Figure 9A). Heat added at a constant pressure raises the mixture‟s temperature and a

sufficient increase causes vapour to form, as shown in Figure 2B. If heat at constant pressure

continues to be added, eventually the temperature becomes so high that only vapour remains

in the container (Figure 9C).

Figure 10 is a typical T-x diagram valid at a fixed pressure. The case shown in Figure 9A,

a container full of liquid mixture with mole fraction xi,o at temperature To, is point 0 on the T-

x diagram. Most mixtures have T-x diagrams that behave in this fashion, but some have a

markedly different feature. If the dew-point and bubble-point curves intersect at any point

other than at their ends, the mixture exhibits azeotropic behaviour at that composition. This

case is shown as position a in the T-x diagram of Figure 11.

Figure 9. Mixture of i and j component in constant pressure container.

5.7.2. Carnot Cycle

The Carnot cycle, which is completely reversible, is a perfect model for a refrigeration

cycle operating between two fixed temperatures, or between two fluids at different

temperatures and each with infinite heat capacity. Reversible cycles have two important

properties: (1) no refrigerating cycle may have a coefficient of performance higher than that

for a reversible cycle operated between the same temperature limits and (2) all reversible

cycles, when operated between the same temperature limits, have the same coefficient of

performance. Figure 12 shows the Carnot cycle on temperature-entropy coordinates. Heat is

withdrawn at constant temperature TR from the region to be refrigerated.



Figure 10. Temperature –concentration (T-x) diagram zerotropic mixture.

Figure 11. Azerotropic behaviour shown on T-x diagram.



Figure 12. Carnot refrigeration cycle.

Qo = To (S2-S3)

Qi = TR (s1-S4) = TR (S2-S3)

Wnet = Qo-Qi

Thus, by equation (62):

TTT

Ro

RCOP

(84)

Flow of energy and its area representation in Figure 13 are:

Energy kJ Area

Qi 125 a

Qa 150 a+b

W 25 a

The Carnot cycle in Figure 14 shows a process in which heat is added and rejected at

constant pressure in the two-phase region of a refrigerant. Saturated liquid at state 3 expands

isentropically to the low temperature and pressure of the cycle at state d. heat is added

isothermally and isobarically by evaporating the liquid-phase refrigerant from state d to

state 1.

The cold saturated vapour at state 1 is compressed isentropically to the high temperature

in the cycle at state b.

However, the pressure at state b is below the saturation pressure corresponding to the

high temperature in the cycle. The compression process is completed by an isothermal

compression process from state b to state c. the cycle is completed by an isothermal and

isobaric heat rejection or condensing process from state c to state 3. Applying the energy

equation for a mass of refrigeration m yields (all work and heat transfer are positive.



Figure 13. Temperature-entropy diagram for Carnot refrigeration cycle.

3Wd = m (h3-hd) (85a)

1Wb = m (hb-h1) (85b)

bWc = To (sb-sc) – m (hb-hc) (85c)

dQ1 = m (h1-hd) = Area def1d (85d)

The net work for the cycle is:

Wnet = 1Wb + bWc – 3Wd = Area d1bc3d (85e)

and

TTT

Ro

R

net

d

W

QCOP

1 (85f)

5.7.3. Theoretical Single-Stage Cycle Using a Pure Refrigerant or Azeotropic Mixture

A system designed to approach the ideal model shown in Figure 14 is desirable.

A pure refrigerant or azeotropic mixture can be used to maintain constant temperature

during phase changes by maintaining constant pressure. Because of concerns such as high

initial cost and increased maintenance requirements, a practical machine has one compressor

instead of two and the expander (engine or turbine) is replaced by a simple expansion valve,

which throttles refrigerant from high to low pressure.



Figure 14. Carnot vapours compression cycle.

Figure 15 shows the theoretical single-stage cycle used as a model for actual systems.

Applying the energy equation for a mass m of refrigerant yields:

4Q1 = m (h1-h4) (86a)

1W2 = m (h2-h1) (86b)

2Q3 = m (h2-h3) (86c)

h3 = h4 (86d)

Constant-enthalpy throttling assumes no heat transfer or change in potential or kinetic

energy through the expansion valve.

The coefficient of performance is:

hhh h

W

QCOP

12

41

21

14

(87)

The theoretical compressor displacement (CD) (at 100% volumetric efficiency) is:

CD = mv1 (88)

Table 13. Thermodynamic property data for figure 17

state t, oC p, kPa v, m

3/kg h, kJ/kg s, kJ(kg.K)

1 -20.0 132.73 0.14739 386.55 1.7413

2 37.8 770.20 0.02798 423.07 1.7413

3 30.0 770.20 0.000842 241.72 1.1435

4 -20.0 132.73 0.047636 241.72 1.16918



Figure 15. Theoretical single-stage vapours compression refrigeration cycle.

Figure 16 shows a schematic p-h diagram for the problem with numerical property data.

The property data are tabulated in Table 13. The saturation temperatures of the single-stage

cycle strongly influence the magnitude of the coefficient of performance (Figure 17).

5.7.4. Lorenz Refrigeration Cycle

The Carnot refrigeration cycle includes two assumptions that make it impractical. Figure

18 is a schematic of a Lorenz cycle.

Qo = (To + ΔT/2) (S2-S3) (89a)

Q1 = (TR –ΔT/2) (S1-S4) = (TR-ΔT/2) (S2-S3) (89b)

Wnet = Qo – QR (89c)

T

TCOP

TTT

Ro

R

)2/( (89d)



Figure 16. Schematic p-h diagrams for the problem with numerical property data.

Figure 17. Areas on T-s diagrams representing refrigerating effect and work supplied for theoretical

single-stage cycle.

A practical method to approximate the Lorenz refrigeration cycle is to use a fluid mixture

as the refrigerant and the four system components. When the mixture is not azeotropic and

phase change occurs at constant pressure, the temperatures change during evaporation and

condensation and the theoretical single-stage cycle can be shown on T-s coordinates as

Figure 19.



Figure 18. Processes of Lorenz refrigeration cycle.

Figure 19. Areas on T-s diagrams representing refrigerating effect and work supplied for theoretical

single-stage cycle using Zerotropic mixture as refrigerant.

5.7.5. Multistage Vapour Compression Refrigeration Cycles

Multistage or multipressure vapour compression refrigeration is used when several

evaporators are needed at various temperatures, such as in a supermarket, or when evaporator

temperature becomes very low.

Thermodynamic analysis of multistage cycles is similar to analysis of single-stage cycles,

except that mass flow differs through various components of the system. For multistage

cycles, the expression for the coefficient of performance should be written as:



net

f

W

QCOP (90)

When compressors are connected in series, the vapour between stages should be cooled

to bring the vapour to saturated conditions before proceeding to the next stage of

compression. Intercooling usually minimises the displacement of the compressors, reduces

the work requirement, and increases the COP of the cycle. If the temperature is below

ambient, which is the usual case, the refrigerant itself must be used to cool the vapour. This is

accomplished with a flash intercooler. Figure 20 shows a cycle with a flash intercooler

installed.

Figure 20. Schematic and pressure-enthalpy diagram for dual-compression, dual-expansion cycles.



5.7.6. Actual Refrigeration Systems

Actual systems operating steadily differ from the ideal cycles considered. Refrigerant

pressures and temperatures are measured at seven locations shown in Figure 21. An air-

cooled, direct-expansion, single-stage mechanical vapour-compression refrigerator uses R-22

and operates under steady condition.

A pressure-enthalpy diagram of cycle is shown in Figure 22. The following performance

data are obtained:

Ambient air temperature to = 30oC

Refrigerated space temperature tg = -10oC

Refrigeration load Qevap = 7.0 kW

Compressor power input Wcomp = 2.5 kW

Condenser fan input WCF + 0.15 kW

Evaporator fan input WEV = 0.11 kW

Figure 21. Schematic of real, direct-expansion, single-stage mechanical vapour-compression

refrigeration system.

6. THERMODYNAMICS AND REFRIGERATION CYCLES

The following cycles are summarised:

6.1. Absorption Refrigeration Cycles

An absorption cycle is a heat –activated thermal cycle. It exchanges only thermal energy

with its surroundings; no appreciate mechanical energy is changed. Absorption cycles are

used in applications where one or more of the exchanges of heat with the surroundings is the



useful product (e.g., refrigeration, air conditioning and heat pump). The two great advantages

of this type of cycle in comparison to other cycles with similar product are:

Figure 22. A pressure-enthalpy diagram of cycle.

No large, rotating mechanical equipment is required.

Any source of heat can be used, including low-temperature sources (e.g., waste heat).

6.2. Ideal Thermal Cycle

All absorption cycle include at least three thermal energy exchanges with their

surroundings (i.e., energy exchanger at three different temperatures). The highest and lowest-

temperature heat flows are in one direction and the mid-temperature one (or two) is in the

opposite direction. In the forward cycle, the extreme (hottest and coldest) heat flows are into

the cycle. Figure 23 illustrated both types of thermal cycles.

This fundamental constraint of heat flow into or out of the cycle at three or more different

temperatures established the first limitation on cycle performance. By the first law of

thermodynamics (at steady state):

QQQmidcoldheat

(91)

(Positive heat quantities are into the cycle)

The second law requires that:

0T

Q

T

Q

T

Q

mist

mis

cold

cold

hot

hot (92)

with equality holding in the ideal case.



Figure 23. Thermal cycles.

From these two laws alone (i.e., without invoking any further assumptions) it follows

that, for the ideal forward cycle:

)

)

(

(

TTTQ

Q

coldmid

cold

hot

midhot

hot

coldideal

TTx

TCOP

(93)

The heat ratio

Q

Q

hot

cold is commonly called the coefficient of performance (COP), which is

the cooling realised divided by the driving heat supplied.

Heat rejected to ambient may be at two different temperatures, creating a four-

temperature cycle. The ideal COP of the four-temperature cycle is also expressed by equation

(93), with Tmid signifying the entropic mean heat rejection temperature. In that case, Tmid is

calculated as follows:

T

Q

T

Q

Q Q

midcold

midcold

midhot

midhot

midcoldmidhot

midT

(94)

This expression results from assigning all the entropy flow to the single Tmid.



The ideal COP for the four-temperature cycle requires additional assumptions, such as

the relationship between the various heat quantities. Under the assumptions that

Q Qmidcoldcold and Q Qmihot

hot the following expression results:

TTT midcold

cold

midcold

cold

hot

midhothot

iidealTTT

xxT

COP)(

(95)

Working fluid phase change constraints

Absorption cycles require at least two working substances: a sorbent and a fluid

refrigerant, these substances undergo phase changes. As illustrated in Figure 24, the

refrigerant phase changes occur in an evaporator and a condenser and the sorbent phase

changes in an absorber and desorber (generator). For the forward absorption cycle, the

highest-temperature heat is always supplied to the generator.

QQgenheat

(96)

and the coldest heat is supplied to the evaporator:

QQevapcold

(97)

The second result of the phase change constraint is that, for all known refrigerants and

sorbents over pressure ranges of interest.

QQevapcond

(98)

and

QQabsgen

(99)

The ideal single-effect forward-cycle COP expression is:

TTTT abs

cond

evapmidcold

evap

gen

absgen

iidealTTT

xxT

COP)

)

(

(

(100)

Practical liquid absorbents for absorption cycles have a significant negative deviation

from behaviour predicted by Raoult‟s law. This has the beneficial effect of reducing the

required amount of absorbent recirculation, at the expense of reduced lift (Tcond-Tevap) and

increased sorption duty. In practical terms, most absorbents:



Figure 24. Single-effect absorption cycle.

3.12.1 toQ

Q

condt

abs (101)

and

)(2.1 TTTT evapcondabsgen (102)

The net result of applying these approximations and constraints to the ideal-cycle COP

for the single-effect forward cycle is:

8.0)

)2.1

(

(

Q

Q

T abs

cond

absgen

condevap

ideal

T

TCOP

T (103)

In practical terms, the temperature constraint reduces the ideal COP to 0.9 and the heat

quantity constraint further reduces it to 0.8. Another useful result is:

TTTT evapabscondgen

min (104)

Alternative approaches are available that lead to nearly the same upper limit on ideal-

cycle COP. This leads to the expression:



)

)

(

(

TT

Tgen

cond

abs

evapideal

TCOP (105)

6.3. Working Fluids

Working fluids for absorption cycles fall into four categories, each requiring a different

approach to cycle modelling and thermodynamic analysis. Liquid absorbents can be non-

volatile (i.e., vapour phase is always pure refrigerant, neglecting condensables) or volatile

(i.e., vapour concentration varies, so cycle and component modelling must track both vapour

and liquid concentration). Solid sorbents can be grouped by whether they are physisorbents

(also known as absorbents), for which, as for liquid absorbents, sorbent temperature depends

on both pressure and refrigerant loading (bivariance) or chemisorbents, for which sorbent

temperature does not vary with loading at least over small ranges. Beyond these distinctions,

various other characteristics are either necessary or desirable for suitable liquid

absorbent/refrigerant pairs as follows:

6.3.1. Absence of Solid Phase (Solubility Field)

The refrigerant/absorbent pair should not solidify over the expected range of composition

and temperature. If a solid forms, it will stop flow and shut down equipment. Controls must

prevent operation beyond the acceptable solubility range.

6.3.2. Relative Volatility

The refrigerant should be much more volatile than the absorbent so the two can be

separated easily. Otherwise, cost and heat requirements may be excessive. Many absorbents

are effectively non-volatile.

6.3.3. Affinity

The absorbent should have a strong affinity for the refrigerant under conditions in which

absorption takes place.

Strong affinity allows less absorbent to be circulated for the same refrigeration effect,

reducing sensible heat losses, and allows a smaller liquid heat exchanger to transfer heat from

the absorbent to the pressurised refrigerant/absorption solution. On the other hand, as affinity

increases, extra heat is required in the generators to separate refrigerant from the absorbent

and the COP suffers.

6.3.4. Pressure

Operating pressures established by the refrigerant‟s thermodynamic properties should be

moderate.

High pressure requires heavy-walled equipment and significant electrical power may be

needed to pump fluids from the low-pressure side to the high-pressure side. Vacuum requires

large-volume equipment and special means of reducing pressure drop in the refrigerant

vapour paths.



6.3.5. Corrosion

Most absorption fluids corrode materials used in construction. Therefore, corrosion

inhibitors are used.

6.3.6. Stability

High chemical stability is required because fluids are subjects to severe conditions over

many years of services. Instability can cause undesirable formation of gases, solids or

corrosive substances. Purity of all components charged into the system is critical for high

performance and corrosion prevention.

6.3.7. Safety

Precautions as dictated by code are followed when fluids are toxic, inflammable or at

high pressure. Codes vary according to country and region.

6.3.8. Transport Properties

Viscosity, surface tension, thermal diffusivity and mass diffusivity are important

characteristics of the refrigerant/absorbent pair. For example, low viscosity promotes heat and

mass transfer and reduces pumping power.

6.3.9. Latent Heat

The refrigerant latent heat should be high so the circulation rate of the refrigerant and

absorbent can be minimised.

6.4. Eco-Friendly Natural Refrigerants

Over the years, all parts of a commercial refrigerator, such as the compressor, heat

exchangers, refrigerant, and packaging, have been improved considerably due to the extensive

research and development efforts carried out by academia and industry. However, the

achieved and anticipated improvement in conventional refrigeration technology are



The word „green‟ designates more than a colour. It is a way of life, one that is becoming more

and more common throughout the world. An interesting topic on „sustainable technologies for

a greener world‟ details about what each technology is and how it achieves green goals.










grows.



The word „green‟ designates more than a colour. It is a way of life, one that is becoming

more and more common throughout the world. An interesting topic on „sustainable

technologies for a greener world‟ details about what each technology is and how it achieves

green goals. Over the years, all parts of a commercial refrigerator, such as the compressor,

heat exchangers, refrigerant, and packaging, have been improved considerably due to the

extensive research and development efforts carried out by academia and industry. However,

the achieved and anticipated improvement in conventional refrigeration technology are












grows. On hot summer days in particular, many of us appreciate a refreshing drinks. The

word „green‟ designates more than a colour. It is a way of life that is becoming more and

more common throughout the world.

6.4.1. Ozone Depletion Potential (ODP)

The ozone layer is damaged by the catalytic action of chlorine, fluorine and bromine in

compounds, which reduce ozone to oxygen and thus destroy the ozone layer (Table 14). The

ozone depletion potential (ODP) of a compound is shown as chlorine equivalent (ODP of a

chlorine molecule = 1).

6.4.2. Global Warming Potential (GWP)

The greenhouse effect arises from the capacity of materials in the atmosphere to reflect

the heat emitted by the earth back onto the earth. The direct global warming potential (GWP)

of a compound is shown as a CO2 equivalent (GWP of a CO2 molecule = 1)

The reason for this is green technology helps to sustain life on earth. This not only

applies to humans but to plants, animals and the rest of the ecosystem. Energy prices and

consumption will always be on an upward trajectory. In fact, energy costs have steadily risen

over last decade and are expected to carry on doing so as consumption grows (Tables 15-17).

Today, buildings are largest consumers of energy. Air conditioning and heating consume

about 40% of the power in the buildings. Demand to conserve energy has become necessity as

there has been rising costs of energy consistently and this make us to think to go green and

innovate the greener concept for buildings. A green building uses less water, optimises energy

efficiency, conserves natural resources, generates less waste and provides healthier spaces for

occupants. And, a green home can have benefits, such as reduction in water and operating

energy costs of the building.



Table 14. Global warming potential (GWP)

Subject Ozone depletion

potential (ODP)

Global warming

potential (GWP)

Ammonia (NH3) 0 0

Carbon dioxide (CO2) 0 1

Hydrocarbons (propane C3H8, propene C3H6, isobutene C4H10) 0 <3

Water (H2O) 0 0

Chlorofluoro-hydrocarbons (CFCs) 1 4680-10720

Partially halogenated Chlorofluoro-hydrocarbons (HCFCs) 0.02-0.06 76-12100

Per-fluorocarbons (PFCs) 0 5820-12010

Partially halogenated fluororinated-hydrocarbons (HFCs) 0 122-14310

Table 15. Energy saving in terms of MW/h and economic saving % for annual operation

between an intelligent free-cooling system and traditional free-cooling

City Energy saving (kWh) Energy saving (%)

Frankfurt 34560 5%

Rome 31587 3%

Milan 29132 4%

Manchester 46008 6%

Paris 36954 4%

Amsterdam 42558 7%

Stockholm 28167 5%

Madrid 36743 4%

Berlin 31525 4%

London 46018 6%

Copenhagen 38077 6%

Table 16. Comparison between the energy consumption of standard chillers, free-

cooling chillers and intelligent free-cooling (Absorbed power (MWh))

City Standard chillers Free-cooling chillers Intelligent free-cooling

Frankfurt 1107 747 712

Rome 1245 1056 1025

Milan 1144 815 786

Manchester 1116 789 743

Paris 1178 910 873

Amsterdam 1062 651 609

Stockholm 1001 541 513

Madrid 1204 938 902

Berlin 1088 707 676

London 1103 154 708

Copenhagen 1061 638 599

This may also mean refrigerant-based chillers and compressors to be shut off or to be

operated at reduced capacity. With the environmental protection posing as the number one



global problem, man has no choice but reducing his energy consumption, one way to

accomplish this is to resort to passive and low-energy systems to maintain thermal comfort in

buildings.

Traditionally hot water generators have been used to generate hot water, which are either

fossil fuel fired like gas, oil, etc., or steam fired (Tables 18-21).

Table 17. Comparison between the energy consumption of standard chillers and

intelligent free-cooling

City Standard chillers (MWh) Intelligent free-cooling (%)

Frankfurt 394 36%

Rome 220 18%

Milan 357 31%

Manchester 373 33%

Paris 304 26%

Amsterdam 452 43%

Stockholm 487 49%

Madrid 301 25%

Berlin 411 38%

London 394 36%

Copenhagen 462 44%

Table 18. Energy consumption and corresponding expenses occurred

Water cooler system

Per day Per day Per month Per month Per year Per year

Energy

(W)

Expenditure

($)

Energy

(W)

Expenditure

($) Energy (W)

Expenditure

($)

Conventional water

cooler with air

cooled condenser

4.6 19.32 138 743.4 1656 8920.8

Modified water

cooler with water

cooled condenser

3.94 16.55 118.2 658.26 1418.4 7899.12

The conventional and modern designs of wind towers can successfully be used in the hot

arid regions to maintain thermal comfort (with or without the use of ceiling fans) during all

hours of the cooling season, or a fraction of it. Climatic design is one of the best approaches

to reduce the energy cost in buildings. Proper design is the first step of defence against stress

of climate. Buildings should be designed according to climate of the site for reducing the need

of mechanical heating or cooling hence maximum natural energy can be used for creating

pleasant environment inside the built envelope. Technology and industry progress of the last

decade diffused electronic and informatics‟ devices in many human activities and now appear

also in building construction. The utilisation and operating opportunities components,

increase the reduction of heat losses by varying the thermal insulation, optimise the lighting

distribution with louver screens and operate mechanical ventilation for coolness in indoor

spaces. In addiction to these parameters the intelligent envelope can act for security control



and became an important part of the building demotic revolution. Application of simple

passive cooling measure is effective in reducing the cooling load of buildings in hot and

humid climates. 43% reductions can be achieved using a combination of well-established

technologies such as glazing, shading, insulation, and natural ventilation. More advanced

passive cooling techniques such as roof pond, dynamic insulation, and evaporative water

jacket need to be considered more closely.

The building sector is a major consumer of both energy and materials worldwide, and the

consumption is increasing. Most industrialised countries are in addition becoming more and

more dependent on external supplies of conventional energy carriers, i.e., fossil fuels. Energy

for heating and cooling can be replaced by new renewable energy sources. New renewable

energy sources, however, are usually not economically feasible compared with the traditional

carriers. In order to achieve the major changes needed to alleviate the environmental impacts

of the building sector, it is necessary to change and develop both the processes in the industry

itself, and to build a favourable framework to overcome the present economic, regulatory and

institutional barriers.

The greenhouse effect is one result of the differing properties of heat radiation when it is

generated at different temperatures. Objects inside the greenhouse, or any other building, such

as plants, re-radiate the heat or absorb it. Because the objects inside the greenhouse are at a

lower temperature than the sun, the re-radiated heat is of longer wavelengths, and cannot

penetrate the glass. This re-radiated heat is trapped and causes the temperature inside the

greenhouse to rise. Note that the atmosphere surrounding the earth, also, behaves as a large

greenhouse around the world.

Table 19. High efficiency chiller heater

Description Units Hot water generator or

conventional chiller heater

High

efficiency

chiller heater

Cooling capacity, i.e., Q1 TR

kW

100

352

100

352

Chilled water temperature profile oC 12/7 12/7

Chilled water flow rate m3/h 60.5 60.5

Heating capacity, i.e., Q3 kW 260 260

Hot water temperature profile oC 80-90 80-90

Hot water flow rate m3/h 22.9 22.9

Fuel type Steam Steam

Steam pressure kg/cm2 g 8 8

Fuel consumption

For chilling

For heating

kg/h

kg/h

380

391

380

235

External heat input i.e., Q2 kW 251 406

Cooling water inlet temperature oC 32 32

Cooling water flow m3/h 100 100

Cooling tower rejection i.e., Q4 kW Q1+Q2

603

Q1+Q2-Q3

498

Cooling COP considering no saving in

heating

1.40 2.4



Table 20. Gas based operation

Description Units Compression chiller Chiller heater

Chilling capacity TR 100 100

Heating capacity kW 260 260

Energy source for chilling Electricity Natural gas

Energy source for heating Natural gas Natural gas

Natural gas consumption for chilling Nm3/h - 26.3

Natural gas consumption for heating Nm3/h 27.2 16.3

Total natural gas consumption Nm3/h 27.2 42.6

Cooling tower duty

Electrical power requirement

Cooling tower fan

Chiller power

Total power

kW

kW

kW

3.0

80.0

83.0

3.7

5.6

9.3

Gas cost $/Nm3 30.0 30.0

Electrical power rate

Hourly operational cost

Gas cost

$/kWh

$/h

8.0

816

8.0

1278

Electrical power cost

Total hourly operational cost

Hourly saving

$/h

$/h

$/h

664

1480

127

1278

75

127

Assumptions: natural gas calorific value 9650 kCal/Nm3, burner efficiency 85%, and higher power cost

because of commercial applications.

Table 21. Steam based operation

Description Units Compression chiller Chiller heater

Chilling capacity TR 100 100

Heating capacity kW 260 260

Energy source for chilling Electricity Steam

Energy source for heating Steam Steam

Steam consumption for chilling kg/h - 380

Steam consumption for heating kg /h 391 235

Total Steam consumption Kg/h 432 498

Cooling tower duty

Electrical power requirement

Cooling tower fan

Chiller power

Total power

kW

kW

kW

3.0

70.0

73.0

3.7

3.0

6.7

Steam cost $/Nm3 0.75 0.75

Electrical power rate

Hourly operational cost

Gas cost

$/kWh

$/h

6.0

293.25

6.0

461.25

Electrical power cost

Total hourly operational cost

Hourly saving

$/h

$/h

$/h

438

731.25

229.75

40.2

501.5

229.75



Changes to the gases in the atmosphere, such as increased carbon dioxide content from

the burning of fossil fuels, can act like a layer of glass and reduce the quantity of heat that the

planet earth would otherwise radiate back into space. This particular greenhouse effect,

therefore, contributes to global warming. The application of greenhouses for plants growth

can be considered one of the measures in the success of solving this problem. Maximising the

efficiency gained from a greenhouse can be achieved using various approaches, employing

different techniques that could be applied at the design, construction and operational stages.

The development of greenhouses could be a solution to farming industry and food security.

6.4.3. Vapour Absorption Chiller Heater

Three important indices of energy efficiency are:

Specific electric power (SP) (kW of electric power input per tone of refrigeration =

kW/TR).

Energy efficiency ratio (EER) = cooling load in W/electric input in W

Expressed in W/W it is generally in the range of 2.3 to 3.1.

The coefficient of performance (COP) is given by:

COP = 3.516/ (kW/TR)

Table 22 gives the energy efficiency performance of typical split and window ACs.

Table 22. Inherent energy efficiency

Particular Window ACs Split ACs

Nominal capacity (TR) 1.5 2.0

EER (W/W) 2.80 2.50

SP (kW/TR) 1.26 1.30

EER (W/W) 2.80 2.48

SP (kW/TR) 1.20 1.23

Advantages of water/lithium bromide include high (1) safety, (2) volatility ratio, (3)

affinity, (4) stability and (5) latent heat. However, this pair tends to form solids and operates

at deep vacuum. Because the refrigerant turns to ice at 0oC, it cannot be used for low-

temperature refrigeration.

6.5. Absorption Cycle Representations

The quantities of interest to absorption cycle designers are temperature, concentration,

pressure and enthalpy. The most useful plots use linear scales and plot the key properties as

straight lines. Some of the following plots are used:



Absorption plots embody the vapour-liquid equilibrium of both the refrigerant and

the sorbent. Plots on linear pressure-temperature coordinates have a logarithmic

shape and hence are little used.

The Duhring (solution temperature versus reference temperature) retains the

linearity. The primary drawback is the need for a reference substance (Figure 25).

Temperature-entropy coordinates are occasionally used to relate absorption cycles to

their mechanical vapour compression counterparts.

Figure 25. Double-effect absorption cycle.

Table 23. Assumptions for single-effect water/lithium bromide model

Assumptions

Generator and condenser as well as evaporator and absorber are under same

pressure

Refrigerant vapour leaving the evaporator is saturated pure water

Liquid refrigerant leaving the condenser is saturated

Strong solution leaving the generator is boiling

Refrigerant vapour leaving the generator has the equilibrium temperature of the

weak solution at generator pressure

Weak solution leaving the absorber is saturated

No liquid carryover from evaporator

Flow restrictors are adiabatic

Pump is isentropic

No jacket heat losses

The LMTD (log mean temperature difference) expression adequately estimates the

latent changes



Figure 26. Generic triple-effect cycles.

Figure 26 shows a double-effect absorption cycle formed by coupling the absorbers and

evaporators of two single-effect cycles into an integrated, single hermetic cycle. Figure 27

shows twelve generic triple-effect cycles. Table 23 shows a simulation of the chiller starts by

specifying the assumptions and the design parameters and operating conditions at the design

point (Table 24).

Figure 27. Single-effect water/lithium bromide absorption cycle.



Figure 28 shows double-effect water/lithium bromide absorption cycle with state points.

Figure 29 shows single-effect ammonia/water absorption cycle.

With the assumptions and the design parameters and operating conditions as specified in

Table 24, the cycle simulation can be conducted by solving the following set of equations

Figure 28. Double-effect water/lithium bromide absorption cycle with state points.

Figure 29. Single-effect ammonia/water absorption cycle.



Table 24. Design parameters and operating conditions for single-effect water/lithium

bromide absorption chiller

Item Design parameters Operating conditions

Evaporator

Condenser

Absorber

Generator

Solution

General

USevap = 319.2 kW/K, countercurrent film

UScond = 180.6 kW/K, countercurrent film

USabs = 186.9 kW/K, countercurrent film-

absorber

USgen = 143.4 kW/K, countercurrent

USsol = 33.8 kW/K, pool-generator

mweak = 12 kg/s

tchill in = 12oC

tchill out = 6oC

tcool out = 35oC

tcool in = 27oC

mhot = 74.4 kg/s

Qevap = 2148 kW

6.5.1. Mass Balances

mmm weakstronrefr (106)

weakweakstrongstrong mm (107)

6.5.2. Energy Balances

)(,, hhmQ condliqevapvapourrefrevap

)( hhmQ chilloutchillinchillevap (108)

)(,, hhmQ condliqgenvapourrefrevap

)( hhmQ coolmeancooloutcoolevap

(109)

QhmhmhmQsolabsweakweakgenstrongstrongevapvapourrefrabs

,,,

)( hhmQ coolmeancooloutcoolevap

(110)

QhmhmhmQsolabsweakweakgenstrongstronggenvapourrefrgen

,,,

)( hhmQ hotouthotincoolhot (111)

)(,, hhmQ solstronggenstrongstrongsol



)(,, hhmQ absweaksolweakweaksol

(112)

6.5.3. Heat Transfer Equations

))(

ln(

)(

)( ,

,

tt

tttt

UAQ

evapvapourchillout

evapvapourchillin

chilloutchillin

evapevap

(113)

))(

ln(

)(

)( ,

,

tt

tttt

UAQ

cooloutcondliq

coolmeancondliq

coolmeancoolout

condcond

(114)

))(

ln(

)()(

)( ,

,

,,

tt

tttttt

UAQ

coolinabsweak

coolmeanabsstrong

coolinabsweakcoolmeanabsstrong

absabs

(115)

))(

ln(

)()(

)( ,

,

,,

tt

tttttt

UAQ

genweakhotout

genstronghotins

genweakhotoutgenstronghotin

gengen

(116)

))(

ln(

)()(

)( ,,

,,

,,,,

tt

tttttt

UAQ

absweaksolstrong

solweakgenstrong

absweaksolstrongsolweakgenstrong

solsol

(117)

Table 25. Simulation results for single-effect water/lithium bromide absorption chiller

Item Internal parameters Performance parameters

Evaporator

t vapour evap = 1.8 oC

P out evap = 0.697 kPa

Qevap = 2148 kW

m chill = 85.3 kg/s

Condenser

T liq cond = 46.2 oC

P sol cond = 10.2 kPa

Q cond = 2332 kW

m cool = 158.7 kg/s

Absorber

δ weak = 59.6%

t weak = 40.7 oC

t strong abs = 49.9 oC

Q abs = 2984 kW

t cool mean = 31.5 oC

Generator

δ strong = 64.6%

t strong gen = 103.5 oC

t weak gen = 92.4 oC

t weak sol = 76.1 oC

Q gen = 3158 kW

t hot in = 125 oC

t hot cond = 115 oC

Solution

t strong sol = 62.4 oC

t weak sol = 76.1 oC

Q sol = 825 kW

δ = 65.4%

General m vapour = 0.93 kg/s

m strong = 11.06 kg/s

COP = 0.68



6.5.4. Fluid Property Equations at Each State Point

Thermal equations of state: h water (t, p), h sol (t, p, δ)

Two-phase equilibrium: t water out (p), t sol out (p, δ)

The results are listed in Table 25.

Table 26. Inputs and assumptions for double-effect water-lithium bromide model

Inputs

Capacity Q evap = 1760 kW

Evaporator temperature t10 = 5.1 oC Desorber solution exit temperature t14 = 170.7 oC

Condenser/absorber low temperature t1, t8 = 42.4 oC

Solution heat exchanger effectiveness e = 0.6

Assumptions

steady state

refrigerant is pure water

no pressure changes except through flow restrictors and pump state points at 1, 4, 8, 11, 14 and 18 are saturated

state point 10 is saturated vapour

temperature difference between high-temperature condenser and low-temperature generator is 5 K parallel flow

both solution heat exchangers have same effectiveness

upper loop solution flow rate is selected such that upper condenser heat exactly matches lower generator heat requirement

flow restrictors are adiabatic pumps are isentropic

no liquid carryover from evaporator to absorber

vapour leaving both generators is at equilibrium temperature of entering solution stream

Table 27. State point data for double-effect lithium bromide/water cycle

Point h (kJ/kg) m (kg/s) p (kPa) Q Fraction t (oC) x (% LiBr)

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

117.7

117.7

182.3

247.3

177.2

177.2

2661.1

177.4

177.4

2510.8

201.8

201.8

301.2

378.8

270.9

270.9

2787.3

430.6

430.6

9.551

9.551

9.551

8.797

8.797

8.797

0.320

0.754

0.754

0.754

5.498

5.498

5.498

5.064

5.064

5.064

0.434

0.434

0.434

0.88

8.36

8.36

8.36

8.36

0.88

8.36

8.36

0.88

0.88

8.36

111.8

111.8

111.8

111.8

8.36

111.8

111.8

8.36

0.0

0.0

0.004

0.0

0.063

1.0

0.0

0.0

0.008

0.0

0.105

42.4

42.4

75.6

97.8

58.8

53.2

85.6

42.4

5.0

5.0

85.6

85.6

136.7

170.7

110.9

99.1

155.7

102.8

42.4

59.5

59.5

59.5

64.6

64.6

64.6

0.0

0.0

0.0

0.0

59.5

59.5

59.5

64.6

64.6

64.6

0.0

0.0

0.0

COP = 1.195 Q evap = 1760 kW

∆t = 5 K Q gen = 1472 kW

δ = 0.600 Q abs1 = 617 kW

Q abs = 2328 kW Q abs2 = 546 kW

Q gen = 1023 kW Wp1 = 0.043 kW

Q cond = 905 kW Wp2 = 0.346 kW



Table 28. Inputs and assumptions for single-effect ammonia/water cycle

Inputs

Capacity Q evap = 1760 kW

High-side pressure p high = 1461 kPa

Low-side pressure p low = 515 kPa

Absorber exit temperature t1 = 40.6 oC

Generator exit temperature t4 = 59 oC

Rectifier vapour exit temperature t1 = 55 oC

Solution heat exchanger effectiveness δ shx = 0.692

Refrigerant heat exchanger effectiveness δ rhx = 0.629

Assumptions

steady state

no pressure changes except through flow restrictors and pump

state points at 1, 4, 8, 11 and 14 are saturated liquid

state point 12 and 13 are saturated vapour

flow restrictors are adiabatic

pumps are isentropic

no jacket heat losses

no liquid carryover from evaporator to absorber

vapour leaving both generators is at equilibrium temperature of entering solution stream

Table 29. State point data for single-effect ammonia/water cycle

Point h (kJ/kg) m (kg/s) p (kPa) Q Fraction t (oC) x, fraction

NH3

1

2

3

4

5

6

7

8

9

10

11

12

13

14

-57.2

-56.0

89.6

195.1

24.6

24.6

1349

178.3

82.1

82.1

1216

1313

1429

120.4

10.65

10.65

10.65

9.09

9.09

9.09

1.55

1.55

1.55

1.55

1.55

1.55

1.59

0.04

515.0

1461

1461

1461

1461

515.0

1461

1461

1461

515.0

515.0

515.0

1461

1461

0.0

0.006

1.000

0.0

0.049

0.953

1.000

1.000

0.0

40.56

40.84

78.21

95.00

57.52

55.55

55.00

37.82

17.80

5.06

6.00

30.57

79.15

79.15

0.50094

0.50094

0.50094

0.41612

0.41612

0.41612

0.99809

0.99809

0.99809

0.99809

0.99809

0.99809

0.99809

0.50094

COP = 0.571 Q evap = 1760 kW

∆t rhx = 7.24 K Q gen = 3083 kW

∆t rhx = 16.68 K Q rhx = 149 kW

δ shx = 0.629 Q abs = 170 kW

δ rhx=0.692 Q shx = 1550 kW

Q abs = 2869 kW W = 12.4 kW

Wp1 = 0.043 kW



Double-effect cycle calculations can be performed in a manner similar to that for the

single-effect cycle. Mass and energy balances of the model shown in Figure 28 were

calculated using the inputs and assumptions listed in Table 26. The results are shown in

Table 27.

Figure 29 shows the diagram of a typical single-effect ammonia/water absorption cycle.

The inputs and assumptions in Table 28 are used to calculate a single-cycle solution, which is

summarised in Table 29.

CONCLUSION

The building sector is a major consumer of both energy and materials worldwide, and the

consumption is increasing. Most industrialised countries are in addition becoming more and

more dependent on external supplies of conventional energy carriers, i.e., fossil fuels. Energy

for heating and cooling can be replaced by new renewable energy sources. New renewable

energy sources, however, are usually not economically feasible compared with the traditional

carriers. In order to achieve the major changes needed to alleviate the environmental impacts

of the building sector, it is necessary to change and develop both the processes in the industry

itself, and to build a favourable framework to overcome the present economic, regulatory and

institutional barriers. Today, buildings are largest consumers of energy. Air conditioning and

heating consume about 40% of the power in the buildings. Demand to conserve energy has

become necessity as there has been rising costs of energy consistently and this make us to

think to go green and innovate the greener concept for buildings. A green building uses less

water, optimises energy efficiency, conserves natural resources, generates less waste and

provides healthier spaces for occupants. And, a green home can have benefits, such as

reduction in water and operating energy costs of the building. This may also mean refrigerant-

based chillers and compressors to be shut off or to be operated at reduced capacity. With the

environmental protection posing as the number one global problem, man has no choice but

reducing his energy consumption, one way to accomplish this is to resort to passive and low-

energy systems to maintain thermal comfort in buildings.

Naturally, it would be preferred, for comfort reasons that this index would be small,

preferably nil. It may be seen that the variable is directly related to temperature discomfort:

the larger the value of the index, the farthest will inside conditions be from expected

wellbeing. Also, the use of electricity operated air conditioning systems will be more

expensive the higher this variable is. Hence, energy expenditure to offset discomfort will be

higher when comparing two index values; the ratio of them is proportional to the expected

energy savings. When the external shade blocks the windowpane completely, the excessive

heat gains belong to the lowest values in the set, and the dimensionless index will be constant

with orientation. For the climate conditions of the locality, it can be seen that a naked window

can produce undesirable heat gains if the orientation is especially unfavourable, when the

index can have an increase of up to 0.3 with respect to the totally shaded window.



TECHNICAL DEFINITION

Refrigeration

The achievement of a temperature below that of the immediate surroundings.

Latent Heat of Fusion

The quantity of heat (Btu/Ib) required changing 1 Ib of material from the solid phase into

the liquid phase.

Sensible Heat

Heat that is absorbed/rejected by a material, resulting in a change of temperature.

Latent Heat

Heat that is absorbed/rejected by a material resulting in a change of physical state

(occurring at constant temperature).

Saturation Temperature

That temperature at which a liquid starts to boil (or vapour starts to condense). The

saturation temperature (boiling temperature) is constant at a given pressure (except for

zoetrope refrigerant) and increase as the pressure increases. A liquid cannot be raised above

its saturation temperature. Whenever the refrigerant is present in two states (liquid and

vapour) the refrigerant mixture will be at the saturation temperature.

Subcooling

At a given pressure, the difference between a liquid‟s temperature and its saturation

temperature.

Ton of Refrigeration

The amount of cooling required to change (freeze) 1 ton of water at 32oF into ice at 32

oF,

in a 24 hour period.



Btu

British thermal unit: The amount of heat required to raise 1 Ib of water 1oF.

1 Ton

12,000 Btu/hr

Refrigeration Effect (RE)

This is the total heat transfer in Btu/Ib from the refrigeration space to the refrigerant.

Heat of Compression (HOC)

This is the amount of heat added to the refrigerant from the compression process.

Refrigerant Circulation Rate (RCR)

The amount of refrigerant in Ib/min, which must circulate in the system to meet the

demands of the load.

Compressor Horsepower Required

The horsepower/ton required to meet the load demand.

Heat of Rejection (HOR)

This is the amount of heat that has to be rejected at the condenser. The heat transferred to

the refrigerant from the refrigerated space (RE) and the heat transferred to the refrigerant

during compression (HOC).

Compressor Volume Required

The compressor cylinder volume required need to pump the RCR in cu ft/ton. The vapour

specific volume is read on the lines of constant volume.



Compression Ratio

The ratio of absolute discharge pressure/absolute suction pressure. The compressor motor

amperage will increase as the compression ratio increases.

Discharge Temperature

A good measure of the relative health or sickness of the system.

Supplemental Heat Lockout

Outdoor thermostats sense the outdoor temperature and lock out the use of secondary

heating devices (supplemental heat) unless the temperature drops below a preset point (except

when the heat pump is not working and needs emergency heat). The advantage of this

strategy is that it is a positive lockout that will not be defeated by "thermostat fiddlers".

Smart Thermostats

Smart thermostats (microprocessor controlled) sense only the indoor temperature. They

will not turn on supplemental heat unless the heat pump is unable to keep the house at the

desired temperature. The advantage of this strategy is that it is tied to indoor comfort. There is

considerable debate about which is the better strategy. But either one is clearly better than no

lockout at all.

Staged Supplemental Heat

Most heat pumps have controls that cause all of the supplemental heat to be on at the

same time. However, there are important benefits to staging supplemental heat. Staging refers

to turning the supplemental heat on in two or more stages. The first stage comes on in mild

temperatures just below the heat pump balance point. If the temperature falls below a point

that the first stage is not enough, the second stage provides more heat. There are two

important benefits of staging supplemental heat. Firstly, is more comfortable and secondly, it

benefits utility ratepayers. One of the major costs of operating an electric utility is the cost of

"peak" power as defined by a high point of electrical customer demands. This peak occurs

during the winter when lots of heating systems need power. By staging the backup heat, the

utility's peak is likely to be lower, reducing utility operating costs. These savings are

particularly important when they can help to avoid the cost of expensive new electrical

generating facilities. Depending on the control system, staging supplemental heat may reduce

the on/off cycles of the heat pump. Reduced cycling can increase the longevity of the heat

pump.



Staging supplemental heat increases the comfort of a heat pump. As it gets colder outside

the refrigeration components produce less heat, so then, when the supplemental heat is not on,

the temperature of the air coming out of the registers gets lower as it gets colder outside. The

supplemental heat adds more heat to the air, making it more comfortable. If all supplemental

heat comes on at once, it may run only for a short time. The delivered air temperature drops

as soon as it goes off. When only one stage of supplemental heat comes on, it stays on longer.

Staged supplemental heat provides higher temperature air for longer periods of time, so the

house will be more comfortable during winter cold spells.

Defrost Control

There are many types of defrost controls. They can be grouped into two categories. First,

is the time and temperature and, second is the demand. Time and temperature controls turn on

the defrost cycle at specified intervals whenever the outdoor temperature reaches a

predetermined point. It is assumed that below a certain outdoor temperature there will be

frequent frost formation and defrost will be necessary. Unfortunately this can result in

unnecessary defrost cycles, which waste energy. There may not actually be frost during the

times specified. Demand controls actually detect the presence of frost on the outdoor coils.

When the controls sense frost, they initiate the defrost cycle. When the frost is melted, the

defrost cycle is terminated. Since the defrost cycle is only used when needed, this is much

more efficient and reduces heating costs.

Emergency Heat Indicator

"Emergency heat" comes on when the heat pump breaks down. Auxiliary heat proves all

the heat to the house. A break down could be caused by a mechanical failure or by the

operation of the safety switches. Since the house will still be warm when emergency heat is

on, occupants may not realise that the heat pump is not working. The emergency electric heat

is more expensive to operate than the heat pump. Many thermostats come with a light that

turns on when the emergency heat is on; indicating that the heat pumps is not working when it

should. This is a highly desirable feature as it serves as a reminder that the heat pump needs

repairs. The terms "emergency heat," "auxiliary heat" and "supplemental heat" are often used

interchangeably, since they all refer to the heating unit(s) that add to, or take over from, the

heat pump when needed. Manufacturers are understandably reluctant to put a light called

"emergency heat" on their equipment, so the emergency heat indicator may have a different

name. A thermostat with an emergency heat light should be specified, even if it goes by a

different name. A supplemental heat indicator, called "auxiliary heat" indicator on some

thermostats, tells when the supplemental heat is on. Unlike emergency heat, supplemental

heat is a normal occurrence. It comes on when the heat pump alone cannot keep the house

warm enough. The heat pump keeps working while the supplemental heat is on.



Safety Switches

To prevent compressor damage, a heat pump should have pressure sensors that indicate

either excessively high or dangerously low refrigerant pressures. If either condition occurs,

the heat pump should automatically shut down, and switch on the emergency heat if needed.

These adverse conditions are often accompanied by high temperature. Accordingly, those in

the heat pump business often call these pressure/temperature switches.

Accumulator

Heat pump compressors are designed to compress gases, not liquids. Liquids are much

more difficult to compress than gases. If liquid refrigerant enters the compressor, it may

damage the compressor. Since compressors are very expensive to replace, heat pumps should

be protected with an accumulator. The accumulator traps liquid refrigerant to prevent it from

entering the compressor. However, scroll compressors are the exception to this rule as they

can handle some liquid refrigerant without damage, so they do not require an accumulator.

Filter/Drier

The filter/drier does two things: It filters the refrigerant to remove dirt and other

impurities that can cause damage to the compressor and other heat pump parts. Also, it

removes moisture from the refrigerant. Moisture can cause a variety of problems, so it is

important to remove it from the system quickly.

Crankcase Heater

When the heat pump is off during cold weather, liquid refrigerant can migrate to the

compressor crankcase, reducing lubrication effectiveness. When the heat pump comes on, the

refrigerant evaporates rapidly. Foam forms in the oil preventing adequate lubrication of the

compressor and shorten its life. The compressor should be equipped with a crankcase heater,

which prevents the refrigerant from liquefying in the oil. Most crankcase heaters are on at all

times. Some are designed to operate only when needed. Again, the scroll compressor is the

exception, as it does not require a crankcase heater to operate safely.

Salt Air Models

The outdoor unit may have a shorter life in coastal locations where salt air corrosion is a

problem. Consequently, some companies manufacture outdoor units designed and built to

resist salt air corrosion.



Types of Heat Pump Technologies

Characteristics of two types of heat pumps are summarised below:

Variable Speed Heat Pumps

A long awaited heat pump has just come on the market. It is called a "variable speed"

heat pump because it adjusts its output to match the heating or cooling requirement of the

home. This allows the heat pump to run continuously rather than starting and stopping

frequently during mild weather. This is more efficient and reduces wear. It can also improve

comfort since the fan speed is matched to the heat output, reducing drafts when the output is

low.

Scroll Compressor

Another recent development is a new type of compressor called a "scroll compressor". It

has a rotary motion that reduces noise more than any other rotary compressors, due to its

unique design. It has a higher efficiency at lower temperatures than reciprocating compressors

for better seasonal heating performance. It also tolerates small amounts of liquid refrigerant,

so it does not require an accumulator or crankcase heater.

Proper Sizing and Installation

It is difficult for the consumer to verify that the contractor properly sizes and installs the

system, and they should at least let the contractor know the expected criteria to be met. If it is

in a utility-sponsored heat pump programme, the utility may help ensure the system meets

these criteria.

Heat Pump Sizing

There are three reasons that a heat pump should be properly sized for the house, cost,

durability and efficiency.

Cost: Large equipment is more expensive. If the system is too large, too much money

will be spent on it. On the other hand, if the heat pump is undersized for heating,

supplemental heat will operate too often, increasing the electric bill.

Durability: Most wear and tear on a compressor occurs when it starts up. Oversized

equipment will cycle on and off more often than accurately sized systems.

Efficiency: Oversized systems have shorter "on" times, which means a greater portion of

"on" time is spent getting started, an inefficient part of the heat pump cycle.

Load Calculations

The only way to properly size a heat pump is to do heating and cooling load calculations

and then match the equipment to the calculated loads. These load calculations should take the

following into account:



The dimensions of the floors, basement walls, above ground walls, windows, doors

and ceilings.

The energy efficiency of these components (insulation, window types, air tightness,

etc.).

Local weather: Loads should be calculated for a cold winter day (but not the coldest

on record) and a hot summer day (but not the hottest on record). The local electric

utility may be able to recommend appropriate design temperatures.

If the heat pump is for heating use only, then only heat load calculations are needed. If

the plan is to do both heating and cooling, both heating and cooling load calculations are

needed. It is important to properly calculate the heating and cooling loads and not to guess for

the reasons explained in the previous section (Heat pump sizing).

Load Calculation Methods

There are several widely accepted methods of calculating heating and cooling loads. The

most popular are based on methods and data developed by the American Society of Heating,

Cooling and Air- conditioning Engineers (ASHRAE). One of the best is called Manual

Journal: Load Calculation developed by the Air Conditioning Contractors of America

(ACCA). Manufacturers often simplify ASHRAE and ACCA methods so that the process can

be speeded up. Most major manufacturers provide forms that help contractors‟ size their

heating equipment. If the forms are based on ASHRAE or Manual Journal, the contractor will

probably do a good job of estimating the sizing requirements.

Room-by-Room Load Calculations

Load calculations should be done for each "room" in the house, because each room has

its own heating and cooling requirement. It is the only way contractors know how much

heating or cooling to deliver to each room. Otherwise, they can only guess at duct sizes, and

deliver incorrect amounts of heating or cooling. Areas that are open to each other are treated

as one "room". For example, one "room" could include kitchen, family and dining areas.

Sizing for Heating and Cooling

In most homes, heat pumps provide both heating and cooling. Proper sizing requires

calculating both heating and cooling loads as explained above. A heat pump that matches

both loads as closely as possible should be chosen.

First, choose equipment that meets the cooling needs of the house. Then check the heat

output of the equipment against the heating requirement "Choosing the best size for heating".

If the heat output is too small, larger equipment can be chosen as long as it does not oversize

the cooling load by more than 25 percent (some utility financing programmes allow 50

percent over-sizing). When sizing for cooling, both sensible and latent cooling loads should



be considered. Sensible cooling is most important, and its the types of cooling people are

familiar with - simply reducing the air temperature. Latent cooling is reducing the humidity

so that occupants can be comfortable at the temperature they choose for indoor living. Since

European summers are typically not humid, latent load is not as important. However, in

general, equipment should be chosen to meet sensible plus latent loads. If the equipment

matching the heating requirement is not available without over-sizing for cooling, then one of

the following strategies may need to be considered:

1. Upgrade the insulation and windows and reduce the air leakage from the house to

reduce the winter heating requirement; or

2. Choose a unit that is undersized for heating, realising that the supplemental heat will

operate more than optimal, hence, reducing the potential savings.

Choosing the Best Size

Appendix 1 shows the relationship between the heating requirement of the house and the

heat output of four hypothetical heat pumps.

Appendix 1. The relationship between the heating requirement of the house and the heat output of four

hypothetical heat pumps.

The four heat pumps are different sizes (they have different heat outputs), so the balance

point occurs at different outdoor temperatures for each. This graph can help to choose the best

heat pump for heating. Assume that the ideal heating unit in the climate has a typical balance

point near 30oF. There is a range of 3 to 5 degrees on either side. Two heat pumps in the

graph are close enough. The smaller heat pump has a balance point of 32oF, and the larger

one, 27oF. None of the units give the exact balance point and the cooling load should be

checked to see which one is acceptable. If the larger unit is outside the boundary of 125-150

percent of the cooling load, then the smaller unit would be chosen. If both were in the

acceptable range for cooling, the larger unit would be chosen for increased heating economy.

The Duct Systems

The issues relating to the duct systems are summarised below:



Duct Sizing

One of the most important aspects of proper installation is to ensure that the ductwork

can deliver adequate airflows to maintain comfort in the house. But even more importantly,

the airflow needs to be adequate to prevent damage to the heat pump.

Airflow must meet the manufacturer's specifications otherwise the refrigerant will not be

able to get rid of excess heat. If the refrigerant gets too hot, refrigerant pressure may exceed

the limits of the compressor and cause it to fail. Replacing a compressor is expensive; it costs

more than the added cost of properly designing and installing the duct system.

A large volume of air moving through the ducts may be needed. It is important that it

does not move too fast. Air velocity in supply (warm air) ducts should not exceed 235 metres

per minute (mpm) for smooth metal ducts or 200 mpm for a flexible duct. Return duct air

(room temperature air returning to heat pump) velocities should not exceed 200 mpm,

regardless of duct type. As a consumer it is necessary to verify these criteria are met. If the

system exceeds these velocities, it will be noisy.

Return Air System

Most customers are aware of the warm air supply outlets because they deliver warm air to

the house. And it is easy to figure out that there are usually ducts that connect the heat pump

to those outlets. However, it is easy to overlook the return side of the system - the ducts and

grills that carry air back to the heat pump. In fact, the return airside is often installed as

though it is an afterthought. Ideally, there is a return grill and duct in each room where there

is a warm air supply. That is usually expensive, but it is worthwhile to get as many returns as

practical. At a minimum, there should be one for the main living areas and one for the

bedrooms. It is also worth considering separate returns for large living rooms and master

bedrooms. A good return air system can help reduce differences in pressure throughout the

house, reduce drafts and improve comfort.

Terminals

Terminals are the registers, vents and grills at the ends of the ducts that deliver air to

rooms or return air to the heat pump. Terminals should distribute conditioned air properly in

each room of the house. The room-by-room heat load calculations mentioned earlier tell the

contractor how much airflow each room needs.

An important consideration for selecting terminals is the air velocity going through them.

In general, larger terminals produce slower velocities. If the velocity is too high, terminals

will be noisy and cause drafts. Supply registers and vents should have a velocity of less than

250 mpm and the correct "throw" (distance the air is projected from the terminal-usually

about three metres). Return grills should have a velocity of 130 mpm or less. Again, recognise

that consumers probably will not be able to verify that these criteria are met.

Terminals should direct airflow away from sitting areas, work areas and other occupied

places. The best design is to deliver air from the floor parallel to an outside wall.

Dampers

Adjustable duct dampers must be installed so airflow can be set for each room, according

to the room-by-room heat load calculations. After the system is installed, the heating

contractor should "balance" the system by adjusting each damper for correct air flows. These



dampers are usually found in branch ducts near where they take off from the main duct. Once

the dampers have been adjusted to balance the system, they are usually not moved unless the

system is modified.

Duct Air Sealing

Air leakage from ducts is typically one of the main sources of heat loss in the house.

Ducts should be sealed at joints between sections, along seams in individual duct sections,

and where ducts penetrate from unheated to heated areas. Ducts are typically sealed with duct

tape. Aluminum tape is more durable than cloth varieties.

The energy savings are worth the expense. Sealing is also important for comfort; one may

not be comfortable if warm air leaks out of the duct before it gets to the room. Also,

eliminating duct leaks is critical for heat pump system efficiency.

Duct Insulation

Ducts passing through unheated areas such as garages, crawl spaces and attics should be

insulated. Northwest regional conservation standards require sheet metal ducts to be insulated

with R-11 insulation and insulated flex duct with two layers of R-4 or one layer of R-11

insulation.

Flexible Duct

Some contractors use flexible ducts with insulation already built in. Because flexible

ducts can be installed by less experienced trades‟ people than required by sheet metal ones,

there is some concern that flexible ducts encourage poor installation practices. Flexible ducts

should be installed according to the following guidelines.

1. Use R-11 or "double wrap" ducts (two layers of R-4).

2. Make all bends gradual so airflow is not restricted.

3. Make connections between sheet metal and flexible ducts with metal or nylon

clamps.

4. Seal connections (aluminium duct tape) between flexible and sheet metal ducts.

5. Support flexible ducts so they do not sag (restricting air flow) using supports at least

2.5 cm wide.

6. Stretch duct to its full length so air passages are as smooth as possible.

The Outdoor Unit

The outdoor units may require the followings.

Outdoor Unit Installation

The outdoor portion of the heat pump should be installed on a concrete pad (unless the

manufacturer specifies other support), separate from the house foundation. Some units are

elevated on legs so air can flow under the unit. In cold climates, the legs should be tall enough

so that snow will not block the airflow. The unit should be far enough out from the eaves so



snow falling from the roof will not land on it. If possible, someone needs to place the unit so

that it is sheltered from prevailing winter winds, and the shrubs do not block airflow.

Outdoor Unit Location

The outdoor unit should be located where its noise will not bother the neighbours.

Likewise, make sure that it is not located under bedroom windows. If it is difficult to find a

good location, then one may need a heat pump with special sound reduction features. The

outdoor unit should be located to minimise the length of the copper refrigerant lines that

connect the indoor and outdoor units.

Refrigerant Lines

The lines should be reasonably straight. Any excess lines that need to be coiled should be

coiled horizontally so they do not form an oil trap. Vertical coils may prevent lubricant from

returning to the compressor. Both refrigerant lines should be insulated; to reduce unwanted

heat loss and heat gain in order to save energy.

Condensate Drain

When the heat pump is in its summer cooling mode, water from air inside the house often

condenses on the indoor coil. This water must be drained from the house (but not into the

crawl space). Locating the drain where it will work by gravity alone may be difficult or

impossible. Some systems require condensate pumps.

Heat Distribution System

The heat pump works by promoting the evaporation and condensation of a refrigerant to

move heat from one place to another. A heat exchanger transfers heat from the

water/antifreeze mixture in the ground loop to heat and evaporate refrigerants, changing them

to a gaseous state. A compressor is then used to increase the pressure and raise the

temperature at which the refrigerant condenses. This temperature is increased to

approximately 40oC. A condenser gives up heat to a hot water tank, which then feeds the

distribution system.

Lengths of plastic pipe are buried in the ground, either in a borehole or a horizontal

trench. The pipe is a closed loop, which is filled with a water/antifreeze mixture. This mixture

circulates in the pipe, absorbing heat from the ground. Horizontal trenches are dug to a depth

of 1-2 metres and can cost less than boreholes, but require a greater area of land. Placing

coiled piping in horizontal trenches will enhance the performance compared with straight

piping. A borehole is drilled to a depth of between 15-100 metres and will benefit from higher

ground temperatures than the horizontal trench, although installation costs will be greater.

Because the GSHPs raise the temperature to approximately 40oC, they are most suitable

for under-floor heating systems, which require temperatures of 30-35oC, as opposed to

conventional boiler systems, which require higher temperatures of 60-80oC. The GSHPs can

also be combined with radiator space heating systems and with domestic hot water systems.

However, top-up heating would be required in both cases in order to achieve temperatures

high enough for these systems. Some systems can also be used for cooling in the summer.



Installation

The Installation of a GSHP should be carried out by a trained engineer. At present, the

UK market is small and there is currently no network of accredited installers as with other

technologies. Manufacturers and suppliers should also be able to provide trained engineers

but geographical limitations may increase installation costs.

The cost of a professionally installed GSHP is ranging from about £1,200-£1,700 per kW

of peak heat output. This includes the cost of the distribution system. Vertical borehole

systems would be at the higher end of this scale, due to greater installation costs. A typical 8

kW system would therefore vary between £9,600-£13,600. The costs will vary from property

to property.

Depending on the size of the system installed, the heat distribution system chosen and the

resulting coefficient of performance (COP), the GSHPs can be a cheaper form of space

heating than oil, LPG or electric storage heaters. It is, however, comparably more expensive

than natural gas. However, the GSHP technology is low in maintenance as systems have very

few moving parts. Systems can have an operating life of over 20 years.

Running Costs

COP is an indicator of the efficiency of a GSHP system. This is the ratio of the number of

units of heat output for each unit of electricity input used to drive the compressor and pump

for the ground loop.

Typical COPs range between 2.5-4. The higher end of this range is for underfloor

heating, because it works at a lower temperature (30-35oC) than radiators. If grid electricity is

used for the compressor and pump, then an economy 7 tariff usually gives the lowest running

costs.

Environmental Impacts

The main environmental impacts of a GSHP system can be summarised as follows:

(1) Emission of Green House Gases

Significant CO2 savings can be gained by displacing fossil fuels. Even compared to the

most efficient gas or oil condensing boilers, a well-designed heat pump with COP of 3-4 will

reduce emissions by 30-35%. Further carbon savings can be made if the electricity used to

power the pump comes from a renewable energy source such as photovoltaic or a renewable

electricity tariff.

Also, measures can be taken to reduce the impact of pollution from using grid electricity

generated through fossil fuel. For example, one can purchase dual tariff green electricity from

a number of suppliers. However, even if ordinary grid electricity is used to run the

compressor, the system will still produce less CO2 emissions than even the most efficient

condensing gas or oil boiler with the same output.



(2) Use of Refrigerants in the System

Refrigerants such as hydrochloroflurocarbons (HCFCs) are present in the GSHP systems

and can pose a threat to the environment through being toxic, flammable or having a high

global warming potential. However, new types and blends of refrigerant with minimal

negative impacts are being developed. A correctly fitted system will also greatly reduce the

potential for leakage, which is why using a professional installer is highly recommended.

Heat Pump Operation

Using direct combustion (gas or oil) to generate heat is never the most efficient use of

fuel. Heat pumps are more efficient because they use renewable energy in the form of low-

temperature heat.

Geothermal heat pumps (GHPs) are a relatively new technology application that can save

homeowners money. These GHPs use the natural heat storage capacity of the earth or ground

water to provide energy efficient heating and cooling. The GHPs should not be confused with

air-source heat pumps that rely on heated air.

The GHPs use the relatively constant temperature of the ground or water several feet

below the earth's surface as source of heating and cooling. They are appropriate for retrofit or

new homes, where both heating and cooling are desired. In addition to heating and cooling,

geothermal heat pumps can provide domestic hot water. Furthermore, they can be used for

virtually any size home or lot in any region of the world.

A GHP system consists of indoor heat pump equipment, a ground loop, and a flow center

to connect the indoor and outdoor equipment. The heat pump equipment works like a

reversible refrigerator by removing heat from one location and depositing it in another

location. The ground loop, which is invisible after installation, allows the exchange of heat

between the earth and the heat pump.

The GHPs can be open or closed-loop. Open-loop systems draw well water for use as the

heat source or heat sink, and after use, return the well water to a drainage field or another

well. Closed-loop or earth-coupled systems use a water and antifreeze solution, circulated in a

ground loop of pipe to extract heat from the earth. Ground loops can be installed in a vertical

well or a horizontal loop. Vertical wells are usually more expensive and used where space is

limited. The length of loop pipe required will vary with soil type, loop configuration, and

system capacity. Loop length can range from 80-330 metres per ton of capacity.

Special heat pump features can include variable speed blowers and multiple-speed

compressors. These features can improve comfort and efficiency in areas where heating and

cooling loads are quite different. Additional features include the capability to produce hot

water. Desuperheaters can be added to supplement the production of domestic hot water when

there is a demand for space heating or cooling. These devices make use of excess heat during

the cooling cycle and use some of the heat during the heating cycle to supplement hot water

production. Dedicated water heaters can be added which operate whenever there is a demand

for hot water.



Factors Affecting Heat Pump Performance

The performance of heat pumps is affected by a large number of factors. For heat pumps

in buildings these include:

1. The climate - annual heating and cooling demand and maximum peak loads.

2. The temperatures of the heat source and heat distribution system.

3. The auxiliary energy consumption, e.g., pumps, and controls.

4. The technical standard of the heat pump.

5. The sizing of the heat pump in relation to the heat demands and the operating

characteristics of the heat pump.

6. The heat pump control system.

Heat Pump Characteristics

The following needs to be considered to enhance heat pump characteristics:

A Constant Heat

A heat pump delivers a lower supply air temperature than a furnace over a longer period

of time to provide a more constant heat. It may give the impression that the system "never

stops running", or "it feels like cold air". At times, the temperature of the air coming out of

the vents is less than the body temperature so it feels like cold air. But it is still providing heat

for the house. And when it can no longer keep-up with the heat loss of the structure, the

second stage or auxiliary heat will automatically energise, bringing on a much warmer heat.

Water Run-off

During the heating cycle, one may notice water running off the outdoor coil. Moisture

from the air is condensed on the outside surface of the coil where it gathers and runs off. This

is normal.

Outdoors Coil Defrosting

At certain conditions (low temperature, high humidity), frost, even ice, may build up on

the coil of the outdoor unit. In order to maintain heating efficiency, the system will

automatically defrost itself. Steam rising from the outdoor unit is normal and is an indication

of proper operation. The vapour cloud will only last for a few minutes. When the defrost

cycle is completed, the system will automatically switch back to heating. Supplemental heat

is automatically energised to maintain comfort during defrost.

Heat flows naturally from a higher to a lower temperature. Heat pumps, however, are

able to force the heat flow in the other direction, using a relatively small amount of high

quality drive energy (electricity, fuel, or high temperature waste heat). Thus heat pumps can

transfer heat from natural heat sources in the surroundings, such as the air, ground or water,

or man-made heat sources such as industrial or domestic waste, to a building or an industrial

application. Heat pumps can also be used for cooling. Heat is then transferred in the opposite



direction, from the application that is cooled, to surroundings at a higher temperature.

Sometimes the excess heat from cooling is used to meet a simultaneous heat demand.

In order to transport heat from a heat source to a heat sink, external energy to drive the

heat pump is needed. Theoretically, the total heat delivered by the heat pump is equal to the

heat extracted from the heat source, plus the amount of drive energy supplied. Electrically

driven heat pumps for heating of buildings typically supply 100 kWh of heat with just 20-40

kWh of electricity. Many industrial heat pumps can achieve even higher performance, and

supply the same amount of heat with only 3-10 kWh of electricity. But simply, during the

cooling cycle, a heat pump will remove heat and humidity from the home and will transfer

this heat to the outdoor air. Likewise, during the heating cycle, a heat pump will remove heat

and humidity from the outdoor air and will transfer this heat to homes. This is possible

because, even at 32 degrees Centigrade, outdoors air contains a great deal of heat. The heat

pump does not generate much heat; it merely transfers it from one place to another.

Loops can be installed in three ways: horizontally, vertically or in a pond or lake

(Appendix 2). The type chosen depends on the available land area, soil and rock type at the

installation site. These factors help to determine the most economical choice for installation

of the ground loop. The GSHP delivers 3-4 times as much energy as it consumes when

heating, and cools and dehumidifies for a lower cost than conventional air conditioning. It can

cut homes or business heating and cooling costs by 50% and provide hot water free or with

substantial savings. The GSHPs can reduce the energy required for space heating, cooling and

service water heating in commercial/institutional buildings by as much as 50%.

Appendix 2. The GSHPs extract solar heat stored in the upper layers of the earth.

The GSHP is an electrically powered system that takes advantage of the earth's relatively

constant ground temperature to provide heating, cooling and hot water for homes or business.

The water-to-water system is especially designed for supplying the hot water for radiant floor

heating or for heating swimming pools. The water to air system is an excellent furnace

replacement or enhancer.

The GSHP is called such, because it pumps heat. Through the use of a simple, yet tried

and tested refrigeration system, it pumps heat from the warm earth in the winter and places it

in homes or businesses. In the summertime the process is reversed. Since it costs far less to

move heat than to make it, much less energy is consumed. This results in huge reductions in

energy costs and greatly reduces the environmental impact of space conditioning. For closed

loop systems, water or an antifreeze solution is circulated through plastic pipes buried beneath

the earth's surface. During the winter the fluid collects heat from the earth and carries it



through the system and into the building. During the summer the system reverses itself to

cool the building by pulling heat from the building, carrying it through the system and placing

it in the ground. Open systems operate on the same principle as closed loop systems and can

be installed where an adequate supply of suitable water is available and open discharge is

feasible. Benefits similar to the closed loop system are realised.

Benefits:

Requires less mechanical room space.

Requires less outdoor equipment.

Does not require roof penetrations, maintenance decks or architectural blends.

Quiet operation.

Reduces operation and maintenance costs.

Limitations:

Requires surface area for heat exchanger field.

Higher initial cost.

Requires additional site co-ordination/supervision.

Higher design cost.

The GSHPs replace the need for a boiler in winter by utilising heat stored in the ground;

this heat is upgraded by a vapour-compressor refrigeration cycle. In summer, heat from a

building is rejected to the ground. This eliminates the need for a cooling tower or heat

rejecter, and also lowers operating costs because the ground is cooler than the outdoor air.

Water-to-air heat pumps are typically installed throughout a building with ductwork serving

only the immediate zone; a two-pipe water distribution system conveys water to and from the

ground-source heat exchanger. The heat exchanger field consists of a grid of vertical

boreholes with plastic u-tube heat exchangers connected in parallel. Simultaneous heating and

cooling can occur throughout the building, as individual heat pumps, controlled by zone

thermostats, can operate in heating or cooling as required.

Unlike conventional boiler/cooling tower type water loop heat pumps, the heat pumps

used in the GSHP applications are generally designed to operate at lower inlet water

temperature. They are also more efficient than conventional heat pumps, with higher COPs.

Because there is lower water temperatures in the two-pipes loop, piping needs to be insulated

to prevent sweating. In addition, a larger circulation pump is needed because the units are

slightly larger in the perimeter zones requiring larger flows.

The GSHPs reduce energy use and hence atmospheric emissions. Conventional boilers

and their associated emissions are eliminated, since no supplementary form of energy is

usually required. Typically, single packaged heat pump units have no field refrigerant

connections and thus have significantly lower refrigerant leakage compared to central chiller

systems. The GSHP units have life spans of 20 years or more and the two-pipe water-loop

system typically used, allows for unit placement changes to accommodate new tenants or

changes in building use. The plastic piping used in the heat exchanger should last as long as

the building itself. When the system is disassembled, attention must be given to the removal

and recycling of the hydrochlorofluorocarbon (HCFC) or Higher heating value



hydrofluorocarbon (HFC) refrigerants used in the heat pumps themselves and the anti-freeze

solution typically used in the ground heat exchanger.

REFERENCES

[1] Luo, L.; Tondeur, D.; Le Gall, H.; and Corbel, S. (2007). Constructal approach and

multi- scale components. Applied Thermal Engineering, 27, 1708-1714.

[2] Luo, L.; Fan, Y.; and Tondeur, D. (2007). Heat exchanger: from micro to multi- scale

design optimisation. International Journal of Energy Research, 31, 1266-1274.

[3] Philappacopoulus, A.J.; and Berndt, M.L. (2001). Influence of de-bonding in ground

heat exchangers used with geothermal heat pumps. Geothermics, 30(5): 527-545.

[4] Jo, H.Y.; Katsumi, T.; Benson, C.H.; and Edil, T.B. (2001). Hydraulic conductivity and

swelling of non-prehydrated GCLs permeated with single- species salt solutions.

Journal of Geotechnical and Geo-environmental Engineering, 127(7): 557-567.

[5] Anandarajah, A. (2003). Mechanism controlling permeability changes in clays due to

changes in pore fluids. Journal of Geotechnical and Geo-environmental Engineering,

129(2): 163-172.

[6] Fridleifsson, I. B. (2003). Status of geothermal energy amongst the world‟s energy

sources. Geothermics, 30: 1-27.

[7] ASHRAE. (1995). Commercial/Institutional Ground Source Heat Pump Engineering

Manual. American Society of heating, Refrigeration and Air conditioning Engineers,

Inc. Atlanta, GA: USA.

[8] Kalbus, E.; Reinstrof, F.; and Schirmer, M. (2006). Measuring methods for

groundwater surface water interactions: a review. Hydrology and Earth System

Sciences, 10, pp. 873-887.

[9] Shah, R. K. (1991). Compact Heat Exchanger Technology and Applications, in Heat

Exchange Engineering, Volume 2: Compact Heat Exchangers: Techniques of Size

Reduction, eds. Foumeny E. A.; and P. J. Heggs, pp. 1–23, Ellis Horwood Limited,

London.

[10] Ramshaw, C. (1995). Process Intensification in the Chemical Industry, Mechanical

Engineering Publications Ltd, London.

[11] Bergles A. E. (1988). Some perspectives on enhanced heat transfer - second generation

heat transfer technology. Journal of Heat Transfer, 110, 1082-1096.

[12] Bowman, W. J.; and Maynes, D. (2001). A Review of Micro-Heat Exchangers Flow

Physics, Fabrication Methods and Application. Proc. ASME IMECE, New York, USA,

HTD-24280.

[13] Li, J.; Zhang, J.; Ge, W.; and Liu, X. (2004). Multi-scale methodology for complex

systems. Chemical Engineering Science, 59, 1687-1700.

[14] Mandelbrot, B. (1982). The Fractal Geometry of Nature, 2nd ed., W. H. Freeman, San

Francisco, California.

[15] Bejan A. (2000). Shape and Structure, from Engineering to Nature. Cambridge

University Press: London. The many faces of protease-protein inhibitor interaction.

EMBO J. 7: 1303-1130.



[16] Luo, L.; Tondeur, D. (2005). Multiscale optimisation of flow distribution by constructal

approach. Particuology, 3, 329-336.

[17] Omer, A. M. (2011). Principle of cooling and heating with ground source energy,

Cooling India, Vol. 7, No. 7, p. 64-77, India, October 2011.

[18] Omer, A. M. (2012). Cooling and heating with ground source energy, International

Journal of Energy Optimisation and Engineering, Vol. 1, No. 2, p. 41-58, Malaysia,

April-June 2012.

[19] Tozer, R.M; and James, R.W. (1997). Fundamental thermodynamics of ideal absorption

cycles. International Journal of Refrigeration 20 (2): 123-135.

[20] United Nations Environment Programme (UNEP). (2006). African regional

implementation review for the 14th session of the Commission on Sustainable

Development (CSD-14). Report on atmospheric and air pollution. New York: UNEP.

[21] World Business Council for Sustainable Development (WBCSD). (2004). Supporting

IEA/SMP transport model documentation and reference case projections. The

sustainable mobility project. World Business Council for Sustainable Development.





Chapter 3

SOME ASPECTS OF SOLAR AND

WIND ENERGY RESOURCES

Abdeen Mustafa Omer Energy Research Institute (ERI), Nottingham, UK

ABSTRACT

Renewable energy resources (solar, wind and biomass) is one of the most

fundamental of natural resources that Sudan must harness in its efforts for rapid

economic development. The role of renewable energy technologies in the development

process cannot be over – emphasised. The demand for energy in Sudan has increased

tremendously over the years and will continue to increase in view of the accelerating pace

of population growth, urbanisation and industrialisation. Comprehensive renewable

energy resources management is a necessity. Human resource development should be

based on education and training programmes funded both by the private and public

sector. Promotion research and development, demonstration and adaptation of energy

resources amongst national, regional, and international organisations which seek clean,

pure, safe, and abundant energy sources. Results, suggest that, wind pumps, solar stills,

and biogas energy must be encouraged, invested, and implemented, but especially for

remote rural areas of Sudan.

Keywords: Renewable technologies, solar energy, wind energy, biomass energy, Sudan,

environment

1. INTRODUCTION

Sudan has a population of about 26 million with an annual growth rate of 2.8% with a

population density of less than 10 per square kilometres. In the last two decades, Sudan has

been suffering from an imbalance of trade. This has led to serious energy problems and

environmental destruction.



Energy supply in Sudan is about 11.7 million tons of oil equivalents (TOE). Out of this

only 57% reached the end users due to conversion losses. About 87% of this energy supply is

biomass [1]. Petroleum energy supply represented only 12% of the total energy supply.

Electricity supply represented almost 1% of the total energy (hydropower supplies 58%, and

the rest is supplied by thermal and gas turbines). As Sudan is a tropical country with high

solar radiation and moderate wind; solar and wind energies seem to be attractive sources of

energy. In Sudan, great attention is given to renewable energy utilisation since the country has

potential for this.

This communication presents a review of three projects carried by national company for

manufacturing water equipment limited (NCMWE). Wind pumps (two are locally

manufactured, installed and tested), solar stills, and biogas technology. The future efforts

must be toward the use of renewable environmentally – friendly, and appropriate technologies

in Sudan.

With increasing urbanisation in the world, cities are growing in number, population and

complexity. At present, 2% of the world‟s land surface is covered by cities, yet the people

living in them consume 75% of the resources consumed by mankind [2]. Indeed, the

ecological footprint of cities is many times larger than the areas they physically occupy.

Economic and social imperatives often dictate that cities must become more concentrated,

making it necessary to increase the density to accommodate the people, to reduce the cost of

public services, and to achieve required social cohesiveness. The reality of modern

urbanisation inevitably leads to higher densities than in traditional settlements and this trend

is particularly notable in developing countries.

Today, the challenge before many cities is to support large numbers of people while

limiting their impact on the natural environment. Buildings are significant users of energy and

materials in a modern society and, hence, energy conservation in buildings plays an important

role in urban environmental sustainability. A challenging task of architects and other building

professionals, therefore, is to design and promote low energy buildings in a cost effective and

environmentally responsive way. Passive and low energy architecture has been proposed and

investigated in different locations of the world [3-4]; design guides and handbooks were

produced for promoting energy efficient buildings [5-8]. However, at present, little

information is available for studying low energy building design in densely populated areas.

Designing low energy buildings in high-density areas requires special treatment to the

planning of urban structure, co-ordination of energy systems, integration of architectural

elements, and utilisation of space. At the same time, the study of low energy buildings will

lead to a better understanding of the environmental conditions and improved design practices.

This may help people study and improve the quality of built environment and living

conditions.

However, the term low energy is often not uniquely defined in many demonstration

projects and studies [9]. It may mean achieving zero energy requirements for a house or

reduced energy consumption in an office building. A major goal of low energy building

projects and studies usually is to minimise the amount of external purchased energy such as

electricity and fuel gas. Yet, sometimes the target may focus on the energy costs or a

particular form of energy input to the building. As building design needs to consider

requirements and constraints, such as architectural functions, indoor environmental

conditions, and economic effectiveness, a pragmatic goal of low energy building is also to

achieve the highest energy efficiency, which requires the lowest possible need for energy


Some Aspects of Solar and Wind Energy Resources 105

within the economic limits of reason. Since many complicated factors and phenomena

influence energy consumption in buildings, it is not easy to define low energy building

precisely and to measure and compare the levels of building energy performance. The loose

fit between form and performance in architectural design also makes quantitative analysis of

building energy use more difficult. Nevertheless, it is believed that super-efficient buildings,

which have significantly lower energy consumption, can be achieved through good design

practices and effective use of energy efficient technology [10].

In an ideal case, buildings can even act as producers rather than consumers of energy.

Besides the operational energy requirements of buildings, it is important to consider two

related energy issues. The first one is the transport energy requirements as a result of the

building and urban design patterns and the second one is the embodied energy or energy

content of the building materials, equipment or systems being used. Transport energy is

affected by the spatial planning of the built environment, transport policies and systems, and

other social and economic factors. It is not always possible to study the effect of urban and

building design on transport energy without considering the context of other influencing

factors. The general efficiency rules are to promote spatial planning and development, which

reduce the need to travel, and to devise and enforce land-use patterns that are conducive to

public transport [11]. Embodied energy, on the other hand, is the energy input required to

quarry, transport and manufacture building materials, plus the energy used in the construction

process. It represents the total life-cycle energy use of the building materials or systems and

can be used to help determine design decisions on system or materials selection [12]. At

present, the field of embodied energy analysis is generally still only of academic interest and

it is difficult to obtain reliable data for embodied energy. Research findings in some countries

indicate that the operating energy often represents the largest component of life-cycle energy

use. Therefore, most people, when studying low energy buildings, would prefer to focus on

operating energy, and perhaps carry out a general assessment of embodied energy only.

To handle population growth on a limited land basis, the word density is unavoidable.

Instead of expanding the boundary, cities often respond to development pressure by setting

targets for increased urban densities. This, however, results in the establishment of a high-rise

cityscape and compact urban settings. The effects of urban concentrated load centres and

compactness of land use patterns will bring benefits to energy distribution and transport

system design, but crowded conditions may create congestion and undesirable local

microclimate. Burchell and Listokin [13] have discussed the urban energy advantage and

believed that cities are more energy efficient for the following reasons:

1. The urban building stock, due its density and compactness, consumes less energy.

2. Cities benefit from advantageous transportation and commutation characteristics.

3. Cities can easily capitalise from emerging more efficient energy systems, and

4. High densities and mixing of land uses may contribute to better efficiency.

2. RENEWABLE ENERGY TECHNOLOGIES

The increased exploitation of renewable energy sources is central to any move towards

sustainable development. However, casting renewable energy thus carries with it an inherent



commitment to other basic tenets of sustainability, openness, democraticisations, etc. Due to

increasing fossil fuel prices, the research in renewable energy technologies (RETs) utilisation

has picked up a considerable momentum in the world. The present day energy arises has

therefore resulted in the search for alternative energy resources in order to cope with the

drastically changing energy picture of the world. The environmental sustainability of the

current global energy systems is under serious question. A major transition away from fossil

fuels to one based on energy efficiency and renewable energy is required. Alternatively

energy sources can potentially help fulfill the acute energy demand and sustain economic

growth in many regions of the world. The mitigation strategy of the country should be based

primarily ongoing governmental programmes, which have originally been launched for other

purposes, but may contribute to a relevant reduction of greenhouse gas emissions (energy-

saving and afforestation programmes). Throughout the study several issues relating to

renewable energies, environment and sustainable development are examined from both

current and future perspectives. The exploitation of the energetic potential (solar and wind)

for the production of electricity proves to be an adequate solution in isolated regions where

the extension of the grid network would be a financial constraint.

The provision of good indoor environmental quality while achieving energy and cost

efficient operation of the heating, ventilating and air-conditioning (HVAC) plants in buildings

represents a multi variant problem. The comfort of building occupants is dependent on many

environmental parameters including air speed, temperature, relative humidity and quality in

addition to lighting and noise. The overall objective is to provide a high level of building

performance (BP), which can be defined as indoor environmental quality (IEQ), energy

efficiency (EE) and cost efficiency (CE).

Indoor environmental quality is the perceived condition of comfort that building

occupants experience due to the physical and psychological conditions to which they

are exposed by their surroundings. The main physical parameters affecting IEQ are

air speed, temperature, relative humidity and quality.

Energy efficiency is related to the provision of the desired environmental conditions

while consuming the minimal quantity of energy.

Cost efficiency is the financial expenditure on energy relative to the level of

environmental comfort and productivity that the building occupants attained. The

overall cost efficiency can be improved by improving the indoor environmental

quality and the energy efficiency of a building.

2.1. Wind Energy

Forty years ago, wind pumps were very common in central Sudan. Unfortunately, they

disappeared gradually due to scarce spare parts, lack of maintenance skills as well as stiff

competition from relatively cheap diesel pumps. The government reintroduced wind pumps to

counter the high prices of imported diesel fuel and the difficulties in transporting it to remote

areas.

In 1985, the energy research institute (ERI) in cooperation with a consultancy services

firm started a wind pumps project. The project was financed by the Netherlands ministry of

foreign affairs [14]. During the 14 months of the project life, ten imported wind pumps were



installed in Khartoum area, while one was locally manufactured for demonstration. Results

suggested that wind energy would be more profitably used for local applications in Sudan.

After the termination of the project, the ERI continued monitoring and testing the

performance of the installed pumps.

The consultancy firm set out to produce wind pumps for low head pumping applications

which could be built in developing countries. The wind pump produced consists of an eight –

bladed rotor which is directly coupled to a specially designed piston pump. This pump is

fabricated with a small leak – hole (to facilitate starting) and a continuous replenished air

chamber (served by a small air pump driven by a wind machine). The design is relatively

simple and is within the capabilities of the local manufacturers. So far, two wind pumps have

been manufactured locally at a cost of us $ 2500 each. The test results show that the design

has some deficiencies. The wind pump‟s performance is about 50% of that predicted by the

firm. This could be due to low pump efficiency and high start – up wind speed (3 ms-1

). The

frequency of maintenance required is high (at least once every two months).

For wind pumps to be effective, it is recommended that more research should be carried

out on adapting the design to the materials available in the local market, quality control

guidelines ought to be set up and users should be trained on how to utilise the pumps more

efficiently.

2.2. Solar Energy

Sudan enjoys bright sunshine and dry weather most of the year. It receives 10 to 12 hours

of sunshine per day with a high level of radiation. The solar radiation is abundant and can

contribute to the country‟s energy use if suitable and appropriate technologies are introduced.

Solar energy can be utilised directly for water desalination or water heating. Due to the

rise in the prices of conventional fuels, solar desalination is more promising and cheaper to

use. It offers an attractive alternative as it is hygienic and helps to reduce the consumption of

fuel wood. Fortunately the areas which lack potable water supplies have abundant solar and

wind energies.

Solar desalination was initiated in Sudan as a possible answer for converting the

underground brackish water into potable water to contribute to the anti – thirst campaign

especially at those isolated arid areas lacking both fresh water and power.

Solar stills are used for distilling the water. A solar still unit consists of a basin, insulating

bottom layer, black lining, transparent cover arranged in such a way that the surface slope

downwards and rests on a collecting trough at the sides. The design is simple in construction,

operation, and maintenance; rigid and firm enough to resist the worst prevailing

environmental conditions; and attempts to use locally available materials (essential foreign

materials are minimised to the least). The energy research institute (ERI) set up a bilateral

project in conjunction with the national company for manufacturing water equipment limited

(NCMWE) in February, 1995.

This project aims at solving the salinity problem in isolated areas. It has been under –

taking research with a view to developing a solar still design that is economical and

technically suitable for use in Sudan. It has also carried out tests on the performance of

existing solar stills under varying prevailing operational conditions.



The parameters that affect the average productivity of the solar stills include: solar

radiation, ambient temperature, wind velocity, depth of the brine in the basin, cover material

and its shape, and construction; and insulating materials used (the life span of a still is 5 – 10

years).

Tests were done in kilo eight area of Khartoum show that the average productivity of a

solar still is about one gallon per square meter per day. The daily productivity of solar stills is

only slightly affected by dust, clouds, and cold weather. Solar stills are suitable for use in

laboratories, medical purposes, charging and topping batteries, supplying drinking water to

small communities in isolated sunny areas as well as local markets.

2.3. Biogas Energy

Many countries with agriculturally based economics face the growing problem of human,

animal and plant waste. This waste is on one hand a very dangerous and continuous source of

pollution, but is on the other hand a very useful source of energy.

A very common technique that utilises such waste is anaerobic fermentation, also called

anaerobic digestion. The gas produced by the digestion of organic waste is known as biogas.

It is colourless, flammable, and generally contains 50 – 70% methane (CH4) and 30 – 40%

carbon dioxide (CO2), with small amounts of other gases such as Hydrogen (H2), Nitrogen

(N2), and Hydrogen Sulphide (H2S). Its energy is more than 20,000 kJm–3

.

Among the many uses of biogas are water heating, space heating, lighting and cooking.

Conversion of internal combustion engines to run on biogas can be relatively simple; thus the

gas can also be used for pumping water and small – scale electric power generation. From the

point of view of better thermal energy utilisation, it is more economical and convenient to use

biogas to generate electricity for lighting, than to burn the biogas directly in biogas lamps.

In rural areas, it is very practical to change a small internal combustion engine with an

asynchronous generator. This has many advantages such as simple structure, ease of operation

and maintenance, less trouble, more safety and low installation cost. It is suitable for farms

whose inhabitants are not too widely scattered. But such an engine – generator is usually

regarded as unsuitable to supply a wider area on account of its low efficiency. Moreover, the

digester can produce not only an excellent gas fuel but also a large quantity of digested sludge

which is an excellent pollution – free organic fertiliser. Thus the promotion and development

of agricultural and animal husbandry in rural areas can be optimised by coordination of

biogas, fertiliser and pollution control.

3. EFFECTS OF URBAN DENSITY

As the quality of living and built environments has become a critical issue in many urban

areas, it is useful to investigate low energy design and evaluate it against the social and

environmental objectives. From psychological and sociological points of view, high

population density and the effect of crowding are interesting topics, which have attracted

much attention. A crowded and stressful urban environment may have unhealthy effects on

the occupants due to air pollution and noise problems. On the other hand, the level of mobility



and traffic speed will benefit the working and living of the people. Therefore, it should be

noted that density and crowding are not necessarily found together. People who live under

crowded conditions may not suffer from being crowded if the built environment has been

designed to provide enough personal space and functional open space.

Compact development patterns can reduce infrastructure demands and the need to travel

by car. As population density increases, transportation options multiply and dependence

areas, per capita fuel consumption is much lower in densely populated areas because people

drive so much less. Few roads and commercially viable public transport are the major merits.

On the other hand, urban density is a major factor that determines the urban ventilation

conditions, as well as the urban temperature [15]. Under given circumstances, an urban area

with a high density of buildings can experience poor ventilation and strong heat island effect.

In warm-humid regions these features would lead to a high level of thermal stress of the

inhabitants and increased use of energy in air-conditioned buildings.

However, it is also possible that a high-density urban area, obtained by a mixture of high

and low buildings, could have better ventilation conditions than an area with lower density

but with buildings of the same height. Closely spaced or high-rise buildings are also affected

by the use of natural lighting, natural ventilation and solar energy. If not properly planned,

energy for electric lighting and mechanical cooling/ventilation may be increased and

application of solar energy systems will be greatly limited.

Table 1. Effects of urban density on city‟s energy demand

Positive effects Negative effects

Transport:

Promote public transport and reduce the need

for, and length of, trips by private cars.

Infrastructure:

Reduce street length needed to accommodate a

given number of inhabitants.

Shorten the length of infrastructure facilities

such as water supply and sewage lines, reducing

the energy needed for pumping.

Thermal performance:

Multi-story, multiunit buildings could reduce the

overall area of the building‟s envelope and heat

loss from the buildings.

Shading among buildings could reduce solar

exposure of buildings during the summer period.

Natural lighting:

Energy systems:

District cooling and heating system, which is

usually more energy efficiency, is more feasible

as density is higher.

Ventilation:

A desirable in flow pattern around buildings may

be obtained by proper arrangement of high-rise

building blocks.

Transport:

Congestion in urban areas reduces fuel

efficiency of vehicles.

Vertical transportation:

High-rise buildings involve lifts, thus increasing

the need for electricity for the vertical

transportation.

Ventilation:

A concentration of high-rise and large buildings

may impede the urban ventilation conditions.

Urban heat island:

Heat released and trapped in the urban areas may

increase the need for air conditioning.

The potential for natural lighting is generally

reduced in high-density areas, increasing the

need for electric lighting and the load on air

conditioning to remove the heat resulting from

the electric lighting.

Use of solar energy:

Roof and exposed areas for collection of solar

energy are limited.

Table 1 gives a summary of the positive and negative effects of urban density. All in all,

denser city models require more careful design in order to maximise energy efficiency and

satisfy other social and development requirements. Low energy design should not be



considered in isolation, and in fact, it is a measure, which should work in harmony with other

environmental objectives. Hence, building energy study provides opportunities not only for

identifying energy and cost savings, but also for examining indoor and outdoor environment.

3.1. Energy Saving in Buildings

The admission of daylight into buildings alone does not guarantee that the design will be

energy efficient in terms of lighting. In fact, the design for increased daylight can often raise

concerns relating to visual comfort (glare) and thermal comfort (increased solar gain in the

summer and heat losses in the winter from larger apertures). Such issues will clearly need to

be addressed in the design of the window openings, blinds, shading devices, heating system,

etc. In order for a building to benefit from daylight energy terms, it is a prerequisite that lights

are switched off when sufficient daylight is available. The nature of the switching regime;

manual or automated, centralised or local, switched, stepped or dimmed, will determine the

energy performance. Simple techniques can be implemented to increase the probability that

lights are switched off [16]. These include:

Making switches conspicuous.

Loading switches appropriately in relation to the lights.

Switching banks of lights independently.

Switching banks of lights parallel to the main window wall.

There are also a number of methods, which help reduce the lighting energy use, which, in

turn, relate to the type of occupancy pattern of the building [16]. The light switching options

include:

Centralised timed off (or stepped)/manual on.

Photoelectric off (or stepped)/manual on.

Photoelectric and on (or stepped), hotoelectric dimming.

Occupant sensor (stepped) on/off (movement or noise sensor).

Likewise, energy savings from the avoidance of air conditioning can be very substantial.

Whilst day-lighting strategies need to be integrated with artificial lighting systems in order to

become beneficial in terms of energy use, reductions in overall energy consumption levels by

employment of a sustained programme of energy consumption strategies and measures would

have considerable benefits within the buildings sector. The perception often given however is

that rigorous energy conservation as an end in itself imposes a style on building design

resulting in a restricted aesthetic solution. Better perhaps would be to support a climate

sensitive design approach which encompassed some elements of the pure conservation

strategy together with strategies which work with the local ambient conditions making use of

energy technology systems, such as solar energy, where feasible. In practice, low energy

environments are achieved through a combination of measures that include:



The application of environmental regulations and policy.

The application of environmental science and best practice.

Mathematical modelling and simulation.

Environmental design and engineering.

Construction and commissioning.

Management and modifications of environments in use.

While the overriding intention of passive solar energy design is to achieve a reduction in

purchased energy consumption, the attainment of significant savings is in doubt. The non-

realisation of potential energy benefits is mainly due to the neglect of the consideration of

post-occupancy user and management behaviour by energy scientists and designers alike.

Buildings consume energy mainly for cooling, heating and lighting as shown in Table 2. The

energy consumption shown in the table was based on the assumption that the building

operates within ASHRAE-thermal comfort zone during the cooling and heating periods [17].

Most of the buildings incorporate energy efficient passive cooling, solar control, photovoltaic,

lighting and day lighting, and integrated energy systems. It is well known that thermal mass

with night ventilation can reduce the maximum indoor temperature in buildings in summer

[18]. Hence, comfort temperatures may be achieved by proper application of passive cooling

systems. However, energy can also be saved if an air conditioning unit is used [19]. The

reason for this is that in summer, heavy external walls delay the heat transfer from the outside

into the inside spaces.

Table 2. Energy-saving in buildings

Passive Comfort

Measures

Active Comfort

Measures

Climatic zones

Mediterranean Subtropical Tropical Desert

Natural ventilation 6 7 7 7

Mechanical

ventilation

4 5 6 6

Night ventilation 6 7 7 7

Artificial cooling 3 5 5 6

Evaporative cooling 3 2 2 7

Free cooling 5 6 6 7

Heavy-weight

construction

6 2 2 6

Light-weight

construction

3 5 5 4

Artificial heating 4 0 0 1

Solar heating 6 0 0 0

Free heating 5 0 0 0

Incidental heat 4 0 0 0

Insulation/permeability 5 0 0 4

Solar control/shading 6 6 6 7

Daytime

artificial lighting

3 3 3 2

Day lighting features 6 5 5 4

* 0 = not important, 4 = important, and 7 = very important (importance is rated from 0 to 7).



Moreover, if the building has a lot of internal mass the increase in the air temperature is

slow. This is because the penetrating heat raises the air temperature as well as the temperature

of the heavy thermal mass. The result is a slow heating of the building in summer as the

maximal inside temperature is reached only during the late hours when the outside air

temperature is already low. The heat flowing from the inside heavy walls can be removed

with good ventilation in the evening and night. The capacity to store energy also helps in

winter, since energy can be stored in walls from one sunny winter day to the next cloudy one.

One can define four levels of thermal mass as follows:

Light building: no thermal mass, e.g., a mobile home.

Medium-light building: light walls, but heavy floor, e.g., cement tiles on concrete

floor, and concrete ceiling.

Semi-heavy building: heavy floor, ceiling and external walls (20 cm concrete blocks)

but light internal partitions (Gypsums boards).

Heavy building: heavy floor, ceiling, external and internal walls (10 cm concrete

blocks, with plaster on both sides).

The exact reduction in the maximum indoor temperature depends on the amount of

thermal mass, the rate of night ventilation, and the temperature swing between day

and night.

3.2. Energy Efficiency and Architectural Expression

The focus of the world‟s attention on environmental issues in recent years has stimulated

response in many countries, which have led to a closer examination of energy conservation

strategies for conventional fossil fuels. Buildings are important consumers of energy and thus

important contributors to emissions of greenhouse gases into the global atmosphere. The

development and adoption of suitable renewable energy technology in buildings has an

important role to play. A review of options indicates benefits and some problems [20-29].

There are two key elements to the fulfilling of renewable energy technology potential within

the field of building design; first the installation of appropriate skills and attitudes in building

design professionals and second the provision of the opportunity for such people to

demonstrate their skills. This second element may only be created when the population at

large and clients commissioning building design in particular, become more aware of what

can be achieved and what resources are required.

Terms like passive cooling or passive solar use mean that the cooling of a building or the

exploitation of the energy of the sun is achieved not by machines but by the building‟s

particular morphological organisation. Hence, the passive approach to themes of energy

savings is essentially based on the morphological articulations of the constructions. Passive

solar design, in particular, can realise significant energy and cost savings. For a design to be

successful, it is crucial for the designer to have a good understanding of the use of the

building. Few of the buildings had performed as expected by their designers. To be more

precise, their performance had been compromised by a variety of influences related to their

design, construction and operation. However, there is no doubt that the passive energy

approach is certainly the one that, being supported by the material shape of the buildings has



a direct influence on architectural language and most greatly influences architectural

expressiveness [30]. Furthermore, form is a main tool in architectural expression. To give

form to the material things that one produces is an ineluctable necessity. In architecture, form,

in fact, summarises and gives concreteness to its every value in terms of economy, aesthetics,

functionality and, consequently, energy efficiency [31]. The target is to enrich the expressive

message with forms producing an advantage energy-wise. Hence, form, in its geometric and

material sense, conditions the energy efficiency of a building in its interaction with the

environment. It is, then, very hard to extract and separate the parameters and the elements

relative to this efficiency from the expressive unit to which they belong. By analysing energy

issues and strategies by means of the designs, of which they are an integral part, one will,

more easily, focus the attention on the relationship between these themes, their specific

context and their architectural expressiveness. Many concrete examples and a whole literature

have recently grown up around these subjects and the wisdom of forms and expedients that

belong to millennia-old traditions has been rediscovered. Such a revisiting, however, is only,

or most especially, conceptual, since it must be filtered through today‟s technology and

needs; both being almost irreconcilable with those of the past. Two among the historical

concepts are of special importance. One is rooted in the effort to establish rational and

friendly strategic relations with the physical environment, while the other recognises the

interactions between the psyche and physical perceptions in the creation of the feeling of

comfort. The former, which may be defined as an alliance with the environment deals with

the physical parameters involving a mixture of natural and artificial ingredients such as soil

and vegetation, urban fabrics and pollution [32]. The most dominant outside parameter is, of

course, the sun‟s irradiation, our planet‟s primary energy source. All these elements can be

measured in physical terms and are therefore the subject of science. Within the second

concept, however, one considers the emotional and intellectual energies, which are the prime

inexhaustible source of renewable power [33]. In this case, cultural parameters, which are not

exactly measurable, are involved. However, they represent the very essence of the

architectural quality. Objective scientific measurement parameters tell us very little about the

emotional way of perceiving, which influences the messages of human are physical sensorial

organs. The perceptual reality arises from a multitude of sensorial components; visual,

thermal, acoustic, olfactory and kinaesthetics. It can, also, arise from the organisational

quality of the space in which different parameters come together, like the sense of order or of

serenity. Likewise, practical evaluations, such as usefulness, can be involved too. The

evaluation is a wholly subjective matter, but can be shared by a set of experiencing persons

[33]. Therefore, these cultural parameters could be different in different contexts in spite of

the inexorable levelling on a planet- wide scale. However, the parameters change in the

anthropological sense, not only with the cultural environment, but also in relation to function.

The scientifically measurable parameters can, thus, have their meanings very profoundly

altered by the non-measurable, but describable, cultural parameters.

However, the low energy target also means to eliminate any excess in the quantities of

material and in the manufacturing process necessary for the construction of our built

environment. This claims for a more sober, elegant and essential expression, which is not

jeopardising at all, but instead enhancing, the richness and preciousness of architecture, while

contributing to a better environment from an aesthetic viewpoint [34]. Arguably, the most

successful designs were in fact the simplest. Paying attention to orientation, plan and form

can have far greater impact on energy performance than opting for elaborate solutions [35].



However, a design strategy can fail when those responsible for specifying materials for

example, do not implement the passive solar strategy correctly. Similarly, cost-cutting

exercises can seriously upset the effectiveness of a design strategy. Therefore, it is imperative

that a designer fully informs key personnel, such as the quantity surveyor and client, about

their design and be prepared to defend it. Therefore, the designer should have an adequate

understanding of how the occupants or processes, such as ventilation, would function within

the building. Thinking through such processes in isolation without reference to others can

lead to conflicting strategies, which can have a detrimental impact upon performance.

Likewise, if the design intent of the building is not communicated to its occupants, there is a

risk that they will use it inappropriately, thus, compromising its performance. Hence, the

designer should communicate in simple terms the actions expected of the occupant to control

the building. For example, occupants should be well informed about how to guard against

summer overheating. If the designer opted for a simple, seasonally adjusted control; say,

insulated sliding doors were to be used between the mass wall and the internal space. The

lesson here is that designers must be prepared to defend their design such that others

appreciate the importance and interrelationship of each component. A strategy will only work

if each individual component is considered as part of the bigger picture. Failure to implement

a component or incorrect installation, for example, can lead to failure of the strategy and

consequently, in some instances, the building may not liked by its occupants due to its poor

performance.

3.3. Sustainable Practices

Within the last decade sustainable development and building practices have acquired

great importance due to the negative impact of various development projects on the

environment. In line with a sustainable development approach, it is critical for practitioners to

create a healthy, sustainable built environment [36-37]. In Europe, 50% of material resources

taken from nature are building-related, over 50% of national waste production comes from the

building sector and 40% of energy consumption is building-related [38]. Therefore, more

attention should be directed towards establishing sustainable guidelines for practitioners.

Furthermore, the rapid growth in population has led to active construction that, in some

instances, neglected the impact on the environment and human activities. At the same time,

the impact on the traditional heritage, an often-neglected issue of sustainability, has not been

taken into consideration, despite representing a rich resource for sustainable building

practices.

Sustainability has been defined as the extent to which progress and development should

meet the need of the present without compromising the ability of the future generations to

meet their own needs [38]. This encompasses a variety of levels and scales ranging from

economic development and agriculture, to the management of human settlements and

building practices. This general definition was further developed to include sustainable

building practices and management of human settlements. The following issues were

addressed during the Rio Earth Summit in 1992 [39]:



The use of local materials and indigenous building sources.

Incentive to promote the continuation of traditional techniques, with regional

resources and self-help strategies.

Regulation of energy-efficient design principles.

International information exchange on all aspects of construction related to the

environment, among architects and contractors, particularly non-conventional

resources.

Exploration of methods to encourage and facilitate the recycling and reuse of

building materials, especially those requiring intensive energy use during

manufacturing, and the use of clean technologies.

The objectives of the sustainable building practices aim to:

Develop a comprehensive definition of sustainability that includes socio-cultural,

bio-climate, and technological aspects.

Establish guidelines for future sustainable architecture.

Predict the CO2 emissions in buildings.

The proper architectural measure for sustainability is efficient, energy use, waste

control, population growth, carrying capacity, and resource efficiency.

Establish methods of design that conserve energy and natural resources.

A building inevitably consumes materials and energy resources. The technology is

available to use methods and materials that reduce the environmental impacts, increase

operating efficiency, and increase durability of buildings. Literature on green buildings

reveals a number of principles that can be synthesised in the creation of the built environment

that is sustainable. According to Lobo [40], these are: land development, building design and

construction, occupant considerations, life cycle assessment, volunteer incentives and

marketing programmes, facilitate reuse and remodelling, and final disposition of the structure.

These parameters and many more are essential for analysis, making them an important

element of the design decision-making process.

Today, architects should prepare for this as well as dealing with existing buildings with

many unfavourable urban environmental factors, such as many spaces have no choice of

orientation, and, often, set in noisy streets with their windows opening into dusty and polluted

air and surrounding buildings overshadowing them.

3.4. Buildings and CO2 Emission

To achieve carbon dioxide, CO2, emission targets, more fundamental changes to building

designs have been suggested [41]. The actual performance of buildings must also be

improved to meet the emission targets. To this end, it has been suggested that the

performance assessment should be introduced to ensure that the quality of construction,

installation and commissioning achieve the design intent. Air-tightness and the

commissioning of plant and controls are the main two elements of assessing CO2 emission.

Air-tightness is important as uncontrolled air leakage wastes energy. Uncertainties over



infiltration rates are often the reason for excessive design margins that result in oversized and

inefficient plants. On the other hand, commissioning to accept procedures would significantly

improve energy efficiency. The slow turnover in the building stock means that improved

performance of new buildings will only cut CO2 emissions significantly in the long term.

Consequently, the performance of existing buildings must be improved. For example,

improving 3% of existing buildings would be more effective in cutting emissions than, say,

improving the fabric standards for new non-domestic buildings and improving the efficiency

of new air conditioning and ventilation systems [42]. A reduction in emissions arising from

urban activities can, however, only be achieved by a combination of energy efficiency

measures and a move away from fossil fuels.

3.5. Low Energy Buildings

There is no single, simple formula for achieving low energy buildings. The basic

principle is to minimise energy demand and to optimise energy supply through a greater

reliance on local and renewable resources. Cities need to take a close look at how to make

more efficient use of resources while fulfilling the needs of the people. An energy dimension

should be included in the development process to measure the sustainability of urban and

building design and growth planning models. Previous experience in public transport systems

indicates that density is conductive to profitability and efficiency [43]. A compact urban form

with vertical zoning through multi-level and multi-functional urban clusters may be an

efficient option for high-density living. There are, however, opportunities for high-density

cities to explore and develop effective energy technologies, which can take full advantage of

the concentrated loads and high-rise context, such as using district energy systems and

vertical landscapes. Designing and constructing low energy buildings require the design team

to follow an energy design process that considers how the building envelope and systems

work together [44].

As low energy design is becoming more and more complicated, there is a need to develop

analytical methods and skills, such as simulation and modelling techniques, for the evaluation

of energy performance of buildings and the analysis of design options and approaches [45].

Kausch [46] pointed out that low energy building design is compatible with a wide range of

architectural styles. Studio Nicoletti [47] also illustrated the methods of architectural

expression for low energy buildings in their projects. For high-density conditions, some of

their methods are still valid but adaptation or modification may be needed to satisfy the local

requirements. Climate consideration is a key element and starting point for formulating

building and urban design principles that aim at minimising the use of energy for

environmental control. In densely populated areas, analysis of the climatic and solar

conditions is critical for the design optimisation. It should be noted that in urban areas, the

group of buildings would in fact modify the climatic conditions surrounding it.

Measures to maximise the use of high-efficiency generation plants and on-site renewable

energy resources are important for raising the overall level of energy efficiency. For

renewable energy systems, energy storage is still the major technical constraint to their

applications [48]. Loads concentration in high-density cities might provide opportunities for

better utilisation of renewable energy systems. At present, lack of incentives and shortage of

land and space are the key factors limiting the deployment of renewable energy systems.



High-rise buildings and high population density make it difficult to find suitable locations for

solar collectors and equipment. As the demand for heating energy is relatively low in many

buildings because of the warm climate throughout the year, the economic advantage of

directly using solar heat is weakened. To promote renewables, it is necessary to create new

development patterns and shift from a centralised view of energy sector to a regional

perspective [49]. One important aspect often being overlooked is the raising of awareness and

the education about low energy design. More efforts are needed to educate the people and

establish the culture so that more people would accept and consider low energy buildings an

important element of their living and working environment. It is important to recognise that

solutions to the energy problems are not simply a matter of applying technology and

enforcement through legislation [50]. It requires public awareness and participation as well.

Therefore, measures to promote public awareness and education are crucial for the

implementation of energy efficiency and renewable energy policies.

In summary, achieving low energy building requires comprehensive strategy that covers;

not only building designs, but also considers the environment around them in an integral

manner. Major elements for implementing such a strategy are as follows.

3.5.1. Efficiency use of energy

Climate responsiveness of buildings.

Good urban planning and architectural design.

Good house keeping and design practices.

Passive design and natural ventilation.

Use landscape as a means of thermal control.

Energy efficiency lighting.

Energy efficiency air conditioning.

Energy efficiency household and office appliances.

Heat pumps and energy recovery equipment.

Combined cooling systems.

Fuel cells development.

3.5.2. Utilise renewable energy

Photovoltaics.

Wind energy.

Small hydros (Table 3).

Waste-to-energy.

Landfill gas.

Biomass energy.

Biofuels.

3.5.3. Reduce transport energy

Reduce the need to travel.

Reduce the level of car reliance.

Promote walking and cycling.

Use efficient public mass transport.

Alternative sources of energy and fuels.



3.5.4. Increase awareness

Promote awareness and education.

Encourage good practices and environmentally sound technologies.

Overcome institutional and economic barriers.

Stimulate energy efficiency and renewable energy markets.

4. DISCUSSIONS

A number of years of data on the solar radiation on horizontal surface, sunshine duration,

and wind speed in Sudan have been compiled, evaluated and presented in this study.

Measurements of global solar radiation on horizontal surface at 16 stations for couples of

years are compared with predictions made by several independent methods. In the first

method, Angstrom formula was used to correlate relative global solar irradiance to the

corresponding relative duration of bright sunshine. Regression coefficients are obtained and

used for prediction global solar irradiance. The predicted values were consistent with

measured values ( 8.01% variation). In the second method, by Barbaro et al., sunshine

duration, and minimum air mass were used to derive an empirical correlation for the global

radiation. The predicted values compared well with measured values ( 12% variation). The

diffuse solar irradiance is estimated using Page‟s, Lui and Jordan‟s correlations. The results

of the two formulas have a close agreement. Radiation map of Sudan was prepared from the

estimated radiation values. The annual daily mean global radiation ranges from 3.05 to 7.62

kWhm-2

per day. Routine wind data from 70 stations were analysed. Monthly average wind

speeds, and average powers were determined for each station. The derived annual average

speeds range from 1.53 to 5.07 ms-1

. Maximum extractable average wind powers were found

to vary between 1.35 and 49.5 wm-2

. Wind map of Sudan was also prepared (Figure 1).

Figure 2 Yearly wind probability density (k = 1.99, c = 7.81ms-1

. Figure 3 Schematic of

the construction of the wind pump.

4.1. Factors Determine Consumption

The following factors determine energy consumption in Sudan:

4.1.1. Urban and Rural

Substitution options for household energy in Sudan urban dwellings are electricity, LPG,

kerosene/gasoline and fuelwood. Rural towns and villages are the viable consumers of wood

and charcoal. Due to unavailability or un-affordability of fuelwood in these areas household

consumption was shifted towards agriculture residues.


Table 3. Recent assessments of comparative costs of turbine and diesel pump

Year 0 1 2 3 4 5 6 7 8 9 10 11 12

Water Current Turbine

Cost 6750

Installation & training 450

3 yrs spares 972 972 972

Annual Maintenance 90 90 90 90 90 90 90 90 90 90 90 90

3 yrs maintenance 90 90 90

Annual cost 7200 90 90 90 1152 90 90 1152 90 90 1152 90 90

Accumulation Cost 7200 7290 7380 7470 8622 8712 8802 9954 10044 10134 11286 11376 11466

3” Diesel Pump

Cost 2000

Fuel 1080 1080 1080 1080 1080 1080 1080 1080 1080 1080 1080 1080

Oil 135 135 135 135 135 135 135 135 135 135 135 135

Grease 6 6 6 6 6 6 6 6 6 6 6 6

Spares & Maintenance 200 400 400 400 400 200 400 400 400 400

Annua1 cost 2000 1221 1421 1621 1621 1621 1621 1221 1421 1621 1621 1621 1621

Accumulation Cost 2000 3221 4642 6263 7884 9505 11126 12347 13768 15389 17010 18631 20252

WCT – Diesel Pump 5200 4069 2738 1207 738 -793 -2324 -2393 -3724 -5255 -5724 -7255 -8786

Note: Figures shown in D.S * 100.



Figure 1. Annual average wind speeds of Sudan (ms-1

).

Figure 2. Yearly wind probability density (k = 1.99, c = 7.81ms-1

.



Figure 3. Schematic of the construction of the wind pump.

4.1.2. Occupation

Occupation pattern are different in urban/rural areas. Housewives consider themselves

unemployed, through they are occupied by household management and children raising

(especially in rural area) are active in farms assisting their husband.

4.1.3. Income

Highest consumption of LPG and electricity is found in higher income households. For

wood and charcoal the situation is the reverse, highest consumption by low-income group and

lowest by high-income groups.

4.1.4. Education

It is quite evident that the share of illiteracy developed from 2.4% in high income

households up to 58.9% in rural low income in systematic matter which shows a direct

correlation between level of income, mode of living and education (result of household

survey 1994). In general illiteracy rates are higher among rural population compared to urban,

with levels around 40–45% except for Khartoum rural with 21% illiterate.



4.1.5. Family Size

The increase in energy supplies especially biomass was mainly observed due to

population growth which directly related to family size.

Figure 4 Performance of the CWD 5000. Figure 5 Locally-manufactured wind pump

installed at kilo 8 site.

4.2. Synthesis of the Renewable Energy

Although the overall impact of renewables has been necessarily low, the experience has

clearly demonstrated their potential as sustainable energy alternatives. There has been

substantial learning in disseminating and managing various technologies on account of:

Scale: with increasing numbers, teething problems have been overcome and better

knowledge has been gained in different aspects related to planning, implementation, operation

and maintenance.

Indigenisation: through joint ventures with international industry, the technology transfer

process has been facilitated, helping in developing local production capacities.

Infrastructure: a strong infrastructure has been created over the years to provide the

technical, operational and managerial support to intervention programmes. This includes

research institutions, training agencies, NGOs, financial intermediaries, etc.

Figure 4. Performance of the CWD 5000.

Diverse strategies: though the whole renewable energy programme started with the same

technology push approach, diversification occurred over a period of time in terms of

strategies and to promote different technologies according to market conditions.



Figure 5. Locally-manufactured wind pump installed at kilo 8 site.

0

2

4

6

8

10

12

14

16

18

20

1 2 3 4 5 6 7 8 9 10 11

Wind speed (m/s)

Ou

tpu

t (l

/min

)

0

10

20

30

40

50

Output (l/min) Efficiency (%)

Figure 6. Wind pumps output and efficiency.



30%

18%16%

8%

8%

6%14%

Solar

Efficiency

Biofuels

Biomass

Wnd

Other renewables

Low carbon

technologies

Figure 7. Renewable systems over the world.

4.2.1. Wind Power Calculation

The theoretical maximum power that can be extracted from the wind is:

P = Π * Cp * ρa * A * V3

(1)

Where:

P is the power available from wind Wm-2

; Cp = 16/27 = 0.593; ρa is the average density of

air in Sudan, at the height of 10 m, taken as 1.15 kgm-3

; A is the area swept by rotor,

projected in a plane perpendicular to the direction of wind m2 and V is the average wind

speed ms-1

.

Annual mean wind speeds were derived from the original monthly mean wind speeds.

Annual mean wind powers were derived from monthly mean wind speeds, which were

calculated according to the following procedure: given a monthly mean wind speed V, the

maximum extractable monthly mean wind power per unit cross-sectional area, P is given by

[51]:

P = 0.3409 * V3 (2)

Where:

V is in ms-1

and P is in Wm-2

.

Figure 6 Wind pump output and efficiency. Figure 7 Renewable systems over the world.

A note should be added on a distinction between wind pumps for different purposes:

(1) Low lift (<6 m), high volume applications (2 pumps are available).

(2) Medium lift application (<50 m) (10 pumps).

(3) Deep-well applications (>50 m) (more than 13 pumps).



46%

18%

10%

1%

1%

3%

1%20% Germany

Spain

Europe

China

India

Asia

Australia

World

Figure 8. Wind distributions over the world.

In the beginning, priority has to be given to further industrial improvement of the

technology around Khartoum. Once the technology is sufficiently reliable, the focus has to

shift to applications in more remote areas for irrigation and for water supply purposes. The

overall specifications of modified CWD 5000 wind pump are as follows (modified by ERI):

Wind machine:

Rotor diameter = 5 m

Number of blades = 8

Tower height = 9 m

Transmission = Crankshaft & connecting rod

Safety mechanism = Furling system

Pump:

Diameter = 76 mm

Stroke = 20 cm

Static head = 20 m

Cut-in wind seed = 4 ms-1

Rated wind speed = 9 ms-1

Cut-out wind speed =12 ms-1

Figure 8 Wind distributions over the world.

The actual value that can be achieved practically is less than the above because of

mechanical losses and aerodynamic problems, which are not considered in collection the

0.593 value.

The most obvious region to start with seems to be the northern region because of a

combination of:

(1) Good wind regime.

(2) Shallow ground water level 5-10 m depth.

(3) Need for additional rural water supply.

(4) Existing institutional infrastructure: National Company for Manufacturing Water

Equipment Limited (NCMWE), Sahara Engineering Company, Sudanese

Agricultural Bank (SAB).



4.2.2 Cost Comparison of Diesel and Wind Pumps

Two types of water lifting from ground water (40 m deep) are distinguished:

(1) A borehole of 35-40 m deep with 18 HP = 13.3 kW diesel engine-turbine pump.

(2) A borehole of 25-30 m deep with CWD 5000 wind pump.

A tentative cost comparison was made of a diesel engine pump and the CWD 5000 wind

pump as shown in Table 4, using the most common formula:

CT = (A + F x P + M)/v (3)

Where CT is the total cost-based on initial capital cost, life, and discount rate; F is the

total fuel consumption; P is the fuel cost per litre; M is the maintenance cost; and v is the

volume of water pumped.

The annual cost is the function of the capital cost which is calculated by the interest rate

and the life of the system [52].

A = [C x I x (I+1)]T/[I+1]

T-1 (4)

Where A is the annual cost; C is the capital cost; I is the interest rate or discount rate; and

T is the lifetime.

Figure 9 Conventional energies. Figure 10 Wind plants in Europe. Figure 11 Energy

conservation measures.

The comparison indicates that the necessary fuel and maintenance needed to run the

diesel pump unit long-term is the main factors, which govern the overall cost, and not the

capital cost of the diesel pump itself. Therefore, in the case of Sudan where the fuel is

expensive, the supply is uncertain, the infrastructure is poor, and areas are remote, the use of

wind machines is the ideal.

The following is concluded:

(1) The initial investment cost of wind pumps is high; this may be a scale problem.

(2) Maintenance costs in some areas are too high for the user.

(3) The pumping costs are more or less the same.

(4) Parallel and integrated projects could reduce costs.

(5) Local production versus import: one of the perceptions is the installation of local

production.

(6) Utilities and water authorities should set up in and take over responsibilities

regarding technology and investment.

(7) There are substantial power production fluctuations due to variation in wind speed,

and using storage devices can smooth these out.



Table 4. Cost comparison of diesel and wind pumps in Sudanese Dinar (S.D)

Specification Diesel pump Wind pump

Cost of borehole deep well 182400 114000

Cost of the system (purchased or fabricated in Sudan) 93600 440000

Cost of storage tank - 420000

Cost of annual fuel consumption 343700 -

Cost of maintenance and repair 120000 110000

Total annual cost 1582100 1084000

Specific water pumping cost 79.1 per m3 54.2 per m

3

1 US $ = S.D 250 (Sudanese Dinar), in July 2000.

Annual output 15000-20000 m3 of water.

Annual fuel consumption: Average 491 gallons and price S.D 350-700 per gallons.

Interest rate 15%.

0 0.2 0.4 0.6 0.8 1

United States

European Union

Japan

Russia

Other developed countries

China and Hong Kong

India

Other developing countries

Million tons of oil equivalent

Coal

Oil

Natural Gas

Figure 9. Conventional energies.

0

1000

2000

3000

4000

5000

6000

Total plant capacity

(1000 t)

Ger Ita Spn UK Fra Aus Fin

Under construction

Operational

Figure 10. Wind plants in Europe.



0 0.2 0.4 0.6 0.8 1

Nuclear

Hydro

Renewables

Coal

Oil

Gas

Generation output (GWh)

BU

Conservation measures

continued

Maximum conservation

Figure 11. Energy conservation measures.

0

10

20

30

40

50

60

70

80

1 2 3 4 5 6 7 8

Battery lifetime

Rel

ativ

e an

nu

al c

ost

(%

)

Maintenance

Charge controler

PV generator

Battery

Figure 12. Battery lifetime versus relative annual cost (%).

Figure 12 Battery lifetime versus relative annual cost (%). Figure 13 PV installed

capacity (MW). Figure 14 Landfill, recycled, composted and incineration in European

countries. Figure 15 PV generations in some leading countries.

The increased exploitation of renewable energy sources is central to any move towards

sustainable development. However, casting renewable energy thus carries with it an inherent

commitment to other basic tenets of sustainability, openness, democraticisations, etc. Due to

increasing fossil fuel prices, the research in renewable energy technologies (RETs) utilisation

has picked up a considerable momentum in the world. The present day energy arises has

therefore resulted in the search for alternative energy resources in order to cope with the

drastically changing energy picture of the world. The environmental sustainability of the



current global energy systems is under serious question. A major transition away from fossil

fuels to one based on energy efficiency and renewable energy is required. Alternatively

energy sources can potentially help fulfill the acute energy demand and sustain economic

growth in many regions of the world. The mitigation strategy of the country should be based

primarily ongoing governmental programmes, which have originally been launched for other

purposes, but may contribute to a relevant reduction of greenhouse gas emissions (energy-

saving and afforestation programmes). Throughout the study several issues relating to

renewable energies, environment and sustainable development are examined from both

current and future perspectives. The exploitation of the energetic potential (solar and wind)

for the production of electricity proves to be an adequate solution in isolated regions where

the extension of the grid network would be a financial constraint.

The provision of good indoor environmental quality while achieving energy and cost

efficient operation of the heating, ventilating and air-conditioning (HVAC) plants in buildings

represents a multi variant problem. The comfort of building occupants is dependent on many

environmental parameters including air speed, temperature, relative humidity and quality in

addition to lighting and noise. The overall objective is to provide a high level of building

performance (BP), which can be defined as indoor environmental quality (IEQ), energy

efficiency (EE) and cost efficiency (CE).

Indoor environmental quality is the perceived condition of comfort that building

occupants experience due to the physical and psychological conditions to which they

are exposed by their surroundings. The main physical parameters affecting IEQ are

air speed, temperature, relative humidity and quality.

Energy efficiency is related to the provision of the desired environmental conditions

while consuming the minimal quantity of energy.

Cost efficiency is the financial expenditure on energy relative to the level of

environmental comfort and productivity that the building occupants attained. The

overall cost efficiency can be improved by improving the indoor environmental

quality and the energy efficiency of a building.

1

10

100

1000

10000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Installed capacity (MW)

(%)

Base

Intermediate

Reserve

Peak

Figure 13. PV installed capacity (MW).



0% 20% 40% 60% 80% 100%

Greece

UK

Finland

Spain

Austria

Germany

Belgium

Netherlands

% of total waste

Landfill

Recycled/composeted

Incineration

Figure 14. Landfill, recycled, composted and incineration in European countries.

0 0.5 1 1.5 2 2.5 3 3.5 4

USA

EU-15

Netherlands

Australia

Germany

Switzeralnd

Japan

Power (Wp)

Figure 15. PV generations in some leading countries.



An expression for the volume of airflow induced by wind is:

Qwind = K A V (1)

Qwind is the volume of airflow (m3/h)

A is the area of smaller opening (m2)

V is the outdoor wind speed (m/h)

K is the coefficient of effectiveness



CONCLUSION

Conclusion

Many cities around the world are facing the problem of increasing urban density and

energy demand. As cities represent a significant source of growth in global energy demand,

their energy use, associated environmental impacts, and demand for transport services create

great pressure to global energy resources. Low energy design of urban environment and

buildings in densely populated areas requires consideration of a wide range of factors,

including urban setting, transport planning, energy system design, and architectural and

engineering details. It is found that densification of towns could have both positive and

negative effects on the total energy demand. With suitable urban and building design details,

population should and could be accommodated with minimum worsening of the

environmental quality.

Wind Energy

Application of wind energy available in Sudan is a major issue in the future energy

strategic planning for the alternative to fossil conventional energy to provide part of the local

energy demand. This communication presents potential of wind regimes available in many

sites in Sudan; northern states (Dongola); eastern states (Port Sudan); and central states (Wad

Medani). The meteorological parameters must be reported, and can be considered as nucleus

information for executing research and development of wind energy projects; as the same

time, they could determine sites that are likely to have a better prospect, and will serve as a

good source of information for statistical analyses and correlation among various stations.

Also, it highlights future plans concerning optimum technical, economical, and environmental

utilisation of all wind energy available in Sudan. This theme presents a review of wind energy

activities, including research work, studies, and pilot projects in wind energy applications,

which are feasible technically and economically in Sudan. It highlights the promotion,

development and demonstration of wind energy resources amongst national, regional and

international organisations which involved; seek clean, safe, and abundant energy sources.

Sudan is rich in wind, about 50% of Sudan area is suitable for generating electricity

(annual average wind speed more than 5 ms-1

), and 75% of Sudan area is suitable for

pumping water (annual average wind speed range between 3 – 5 ms-1

).

In areas where there is wind energy potential, and not connected to the electric grid,

the challenge is the simplicity in the design, and higher efficiency.

The research and development in the field of wind pumps should be directed towards

utilising the local skills, and the local available materials.

Local production of wind pump should be encouraged on both public and private

organisations.



Solar Energy

Provide incentives to encourage the household sector to use solar energy.

Invest in research and development.

Assist in launching public awareness programmes with respect to solar energy.

Availing training opportunities to personnel at different levels.

Provide loans and / or grants to achieve the above objectives.

Biogas Energy

Biogas technology cannot only provide fuel, but also important for comprehensive

utilisation of biomass forestry, animal husbandry, fishery, evoluting the agricultural

economy, protecting the environment, realising agricultural recycling, as well as

improving the sanitary conditions, in rural areas.

The biomass energy, one of the important options, which might gradually replace the

oil in facing the increased demand for oil and may be an advanced period in the

coming century. Sudan can depend on the biomass energy to satisfy part of local

consumption.

Development of biogas technology is vital component of alternative rural energy

programme in Sudan, whose potential is yet to be exploited. A concerted effort is

required by all if this has to be realised. The technology will find ready use in

domestic, farming, and small – scale industrial applications.

The diminishing agricultural land may hamper biogas energy development but

appropriate technological and resource management techniques will offset the

effects.

Support biomass research and exchange experiences with countries that are advanced

in this field. In the meantime, the biomass energy can help to save exhausting the oil

wealth.

REFERENCES

[1] Eisa, E. I. (1981). A design study for wind pumps system for use in the Sudan, PhD

Thesis, University of Reading, Reading.

[2] Hamid, Y. H. and Jansen, W. A. M. (1981). Wind energy in Sudan, CWD – report 81, –

2, The Netherlands.

[3] Posoriski, et al. (1984). Case studies of the wind potential and its utilisation for water

pumping in Northern Sudan, Special Energy Programme (SEP), Khartoum, Sudan,

August.

[4] Joop Van, Meel, Paul, H. and Abdeen Mustafa, Omer. (1987). Evaluation of Sudan

wind energy project, CWD – ERI, October, 5 – 16.

[5] Abu Bakr, E. H. (1988). The boundary layer wind regime at a representative tropical

Africa region, Central Sudan, PhD Thesis, Eindhoven, University of Technology, The

Netherlands, December.



[6] Abdeen Mustafa, Omer. (1993). Wind speeds and wind power potential in Sudan, 4th

Arab International Solar Energy Conference, Amman, Jordan, November.

[7] Hamid, E. B. (1995). Design and testing of a wind pump with pneumatically transmitted

power, M.Sc. Thesis, University of Khartoum, Sudan, December.

[8] Abdeen Mustafa, Omer. (1997). Compilation and evaluation of solar and wind energy

resources in Sudan, Renewable Energy, Vol. 12, No. 1, 39 – 69.

[9] Naga, A.M. and Amin, M. (1996). Towards a healthy urban environment in hot-humid

zones-information systems as an effective evaluation tool for urban conservation

techniques. In: Proceedings of XXIV IAHS World Congress: Ankara. Turkey.

[10] Steele, J. (1997). Sustainable architecture: principles, paradigms, and case studies.

New York: McGraw-Hill Inc.

[11] Sitarz, D. Editor. (1992). Agenda 21: The Earth Summit Strategy to save our planet.

Boulder (CO): Earth Press.

[12] Omer, A.M. and Yemen, D. (2001). Biogas an appropriate technology. Proceedings of

the7th

Arab International Solar Energy Conference, pp.417, Sharjah, UAE, 19-22

February.

[13] Sarafiddis, Y., Diakoulaki, D., Papayannabis, L. and Zervos, A. (1999). A regional

planning approach for the promotion of renewable energies. Renewable Energy, 18(3),

317-30.

[14] Tong, C.O. and Wong, S.C. (1997). Advantages of a high density, mixed land use,

linear urban development. Transportation, 24(3), 295-307.

[15] Lam, J.C. (2000). Shading effects due to nearby buildings and energy implications.

Energy Conversion and Management, 47(7), 647-59.

[16] Martin, W. (1996). Air quality control in commercial buildings. WREC.

[17] Limb, M.J. (1995). Air intake positioning to avoid contamination of ventilation. AIVC.

[18] Burge, P., Jones, P. and Robertson, A. (1993). Sick building syndrome. Proceedings

Indoor Air Quality. Finland.

[19] Persily, A.K. (1989). Ventilation rates in office buildings. ASRAE, 128-136. Atlanta:

USA.

[20] Givoni, B. (1976). Man climate and architecture. Applied Science Publishers Ltd, 289-

306. London, UK.

[21] Koenigsberger, O., Ingersoll, T., Mayhew, A. and Szokolay, S. (1973). Manual of

tropical housing and building. Part 1: Climate design. Longmas, 119-130. London, UK.

[22] Boulet, T. (1987). Controlling air movement: a manual for architects and builders.

McGraw-Hill, 85-138, New York: USA.

[23] Szokolay, S. (1990). Design and research issues: passive control in the tropic.

Proceedings First World Renewable Energy Congress, 2337-2344, Reading: UK.

[24] Borda-Daiz, N., Mosconi, P. and Vazquez, J. (1989). Passive cooling strategies for a

building prototype design in a warm-humid tropical climate. Solar and Wind

Technology, 6, 389-400.

[25] Givoni, B. (1965). Laboratory study of the effect of window sizes and location on

indoor air motion. Architectural Science Review, 8, 42-46.

[26] Fordham, M. (2000). Natural ventilation. Renewable Energy, 19, 17-37.

[27] Hirunlabh, J. and Khedari, J. (1999). Study of natural ventilation of houses by a

metallic solar wall under tropical climate. Renewable Energy, 18, 109-119.



[28] Mats, S. and Barhram, M. (1998). Ventilated solar roof airflow and heat transfer

investigation. Renewable Energy, 15, 287-292.

[29] Campbell, J. (1988). Use of passive solar energy in office. In: O‟Sullivan P, editor,

Passive Solar Energy in Buildings. The Watt Committee on Energy. New York:

Elsevier Applied Science.

[30] ASHRAE. (1993). Principle of the heating and ventilation and air conditioning.

Handbook-Fundamentals. American Society of Heat and Refrigerating and Air

Conditioning Engineers. USA.

[31] Molla, M. (1997). Air pollutants and its probable transmutation in the ionosphere.

Renewable Energy, 10 (2/3), 327-329.

[32] Jones, P.G. and Cheshire, D. (1996). Bulk data for marking non-domestic building

energy consumption. CIBSE/ASHRAE Joint National Conference. Harrogate, 29

September – 1 October, 03-213.

[33] BRECSU. (1998). Energy use in offices. Energy Consumption Guide 19. Watford:

Building Research Energy Conservation Support Unit.

[34] Farm Energy Centre. (2000). Helping agriculture and horticulture through technology,

energy efficiency and environmental protection. Warwickshire.

[35] Randall, M. (1992). Environmental Science in Building. Third Edition.

[36] Tiwari, G.N. and Goyal, R.K. (1998). Greenhouse technology. New Delhi: Narosa

Publishing House.

[37] Santamouris, M., Balaras, C.A., Dascalaki, E. and Vallindras, M. (1994). Passive solar

agricultural greenhouse; a worldwide classification evaluation of technologies and

systems used for heating purpose. Solar Energy, 53(5), 411-26.

[38] Santamouris, M., Arigirious, A. and Vallindras, M. (1993). Design and operation of a

low energy consumption passive greenhouse. Solar Energy, 52(5), 371-8.

[39] Mercier, I. (1982). Design and operation of a solar passive greenhouse in the South

West France. In: Proceedings of the International Congress on Energy Conservation of

Agriculture and Fishculture. London.

[40] Grafiadellis, M. (1987). Greenhouse heating with solar energy. In: Von Zabettitz C

(Editor), FAO, Rome.

[41] Lobo, C. (1998). Defining a sustainable building. In: Proceedings of 23rd National

Passive Conference, ASES‟98. Albuquerque (USA): American Solar Energy Society.

[42] Pacheco, M., Marreivos, S. and Rosa, M. (1987). Energy conservation and renewable

energies for greenhouse heating. In: Von Zabettitz C (Editor), FAO, Rome.

[43] Santamouris, M., Mihalakakow, G., Belaras, C.A., Lewis, J.O., Vallindras, M. and

Argiriou, A. (1996). Energy conservation in greenhouses with buried pipes. Energy,

21(5), 353-60.

[44] Garzoli, K.V. and Blackwell, J. (1981). An analysis of the nocturnal heat loss from a

single skim plastic greenhouse. J. Agric. Engg. Res., 26, 203-14.

[45] Chandra, P. & Albright, L.D. (1980). Analytical determination of the effect on

greenhouse heating requirements of using night curtains. Trans ASAE, 23(4), 9994-

1000.

[46] Blackwell, J. and Garzoli, K.V. (1981). Australian development of a low energy

greenhouse based on thermal storage rock piles. Acta Horticulture, 115, 113-23.



[47] Kausch A. (1998). Is low-energy building compatible with any architectural style? In:

Environmental friendly cities. Proceedings of PLEA 98, Lisbon: Portugal, June, 217-

220.

[48] Studio Nicoletti. (1998). Architectural expression and low energy design.

RenewableEnergy, 15, 32-41.

[49] Sutar, R.F. and Tiwari, G.N. (1995). Analytical and numerical study of a controlled

environment agricultural system for hot and dry climatic conditions. Energy and

Building, 23, 9-18.

[50] Carpenter G. (1986). Dust in livestock buildings- a review of some aspects. Journal of

Agricultural Engineering Research, 33, 227-241.

[51] Dawson J. (1990). Minimising dusts in livestock buildings: possible alternatives to

mechanical separation. Journal of Agricultural Engineering Research, 47, 235-248.




Chapter 4

RATIONALIZATION OF SALT-RELATED PROCESSES

IN THE LEATHER INDUSTRY AS A TOOL FOR

MINIMIZATION OF THEIR ENVIRONMENTAL IMPACT

Karel Kolomazník, Michaela Bařinová, Jiří Pecha

and Dagmar Janáčová Faculty of Applied Informatics, Tomas Bata University in Zlin, nám.

T. G. Masaryka, Czech Republic

ABSTRACT

The main purpose of our contribution is reducing the existing environmental burden

related to the worldwide use of sodium chloride for raw hide curing. The essential step is

rationalization of the curing process itself, since it determines the amount of salt to deal

with in the subsequent operations such as pre-soaking, soaking, after-soaking and

desalting of animal fleshings during their complex processing into valuable products.


particularly sodium chloride. The salt often gets to water environment, or right to the

soils resulting in arid conditions in the spilling areas, from where it can be again washed

away to water sources. Despite intensive search for alternative methods, sodium chloride

remains the most common way of raw hide curing worldwide. This method not only

represents considerable environmental burden, but insufficient application of the theory

of the related transport phenomena also has economic impacts, such as high consumption

of the preservation agent, of water and electric power. Rationalization of the curing

process brings considerable reduction in the time necessary for proper curing and thus

reduction in the amount of salt which in subsequent soaking operation would get to

tannery effluents. Understanding the related transport processes, particularly diffusion of

sodium chloride into or from the hide inner volume, gives ground for their optimization

and thus to reduction of their adverse environmental impact as well as minimization of

the consumption of sodium chloride, water and energies. Our approach lies in application

of theoretical tools of chemical engineering, namely indirect modeling based on

quantitative relations from the theory of transport phenomena. This method is in

particular cases supported by experimentally measured data, mainly estimation of the

value of the effective diffusion coefficient of sodium chloride in animal fleshings and its


Karel Kolomazník, Michaela Bařinová, Jiří Pecha et al. 138

application in calculation the minimal time necessary for proper curing of raw hides. The

acquired knowledge is applied in newly emerging industrial field, complex processing of

animal fleshings into valuable products such as quality biodiesel and gelatin. The

fleshings contain considerable amount of preservation salt, therefore desalting represents

the key operation in their processing and the quality of the resulting products is highly

dependent on the precise performance of the desalting process. The essential step leading

to the optimization was the determination of the time dependence of sodium chloride

concentration (from the hide surface towards its internal volume), in other words, a non-

stationary concentration field of sodium chloride in the hide during the curing process.

Subsequently, the solution of the model allowed the calculation of the optimum time

necessary for sufficient preservation.

INTRODUCTION

As well as in medicine it is said that prevention is better than cure, the same applies for

industrial utilization of sodium chloride (common salt, NaCl). Examples of that utilization are

preservation of meat products, cheese manufacturing or processes in which raw hide is

transformed into leather.

Raw hides must be technologically processed by 12 hours after flaying [1] to prevent

microbiological and autolytical degradation which negatively affects the properties of the

final product – leather. Since such quick processing is not possible in many areas, the hides

must be preserved before transportation to tanneries.

Since the authors have gained extensive experience in optimization and rationalization of

leather production from raw hides, this contribution will deal in particular with the processes

which involve sodium chloride – namely raw hide curing and a specific operation called

“pickling” of white hide, which precedes the most important process – tanning.


particularly sodium chloride. The salt often gets to water environment, or right to the soils

resulting in arid conditions in the spilling areas, from where it can be again washed away to

water sources. According to FAO [2], the worldwide leather production in 2010 was 6,214

ths. tons of bovine hides and 726 ths. tons of sheep and goatskins. For example in Kenya,

with yearly production of 2,8 mil. hides and 6 mil. skins, 85% of them are preserved with

sodium chloride, with an average of 11 ths. tons of salt spent every year [3]. Considering that

the amount of salt is about 300 kg of salt for one ton of hides, and by rough estimation more

than 60 % of world hides and skins are salt-cured, we get some million tons of sodium

chloride to deal with every year, most of it coming to waste waters.

Ozgunay et al. in their work [4] describe the main environmental impacts of salt coming

from leather industry, focusing mainly on the problem of limited water availability of saline

water for plants that are not salt-tolerant and soil degradation in the areas of storage of salt-

cured solid wastes. Due to concentration of leather industry within the last decades into

clusters limited to a relatively narrow geographical areas, the problem with increased salinity

applies especially to these areas and affects both soils and fresh waters (lakes, rivers, ground

water). It should be also considered that high salt content in rivers may temporarily or in long

time periods change the osmotic conditions especially when the river empties into water

reservoirs with higher evaporation rate than the influx.


Rationalization of Salt-Related Processes in the Leather Industry ... 139

There are various methods for optimization of salt consumption in raw hide curing. In our

contribution, we applied indirect modeling, which is based on mathematical analysis of

physical or physical-mechanical mechanisms of processes. On the contrary, direct modeling

is based only on experimental procedures without knowing the exact mechanisms and their

mathematical –physical description, which is used so far in most of published tannery

research. The decision whether to apply direct or indirect modelling is a key step in a

proposal of a new technology or optimization of an existing one [5].

In direct modelling, we choose such an experimental laboratory (or pilot scale) procedure

that, without calculation and mathematical analysis of physical or physical-mechanical

processes, gives us direct answer to how the operating plant will behave. The experiment

therefore should be conducted in such a manner that the relationship between laboratory and

real operating process is as close as possible and the proposed design and operating mode of

the industrial apparatus do not require high abstraction. In conclusion, direct modelling builds

on long-time experience and comprises extensive and demanding experimental work.

Compared to this, the indirect modelling is based on mathematical models describing the

industrial processes or technological devices according to the nature of elementary actions,

that is physical-chemical processes, transport phenomena, sorption processes, etc. This

approach is based on highly abstracted data, i.e., data independent on the employed

experimental apparatus and its mode of operation. The experimental design in this way of

modelling can substantially differ from the real industrial setup as well as of laboratory

measuring potentially being quite different.

At present, the direct method is becoming hardly applicable in many industries including

leather manufacturing. The main reasons are high financial and time demands, and with

increasing production it is also getting difficult to approximate a laboratory experiment to real

industrial conditions, resulting in often unacceptable extent of intuitive extrapolation. On the

other hand, relying only on indirect modelling based exclusively on published data can also

be treacherous. This particularly applies to leather industry where well-conducted

experiments still play indispensable role because in most cases it represents the only way how

to obtain the elementary mechanism and properties of the studied process. For effective

combination of mathematical modelling and experimental work, it is necessary to perform

such experiments that allow precise evaluation and subsequent design of a technological

device on the basis of the experiment results.

An example of successful application of indirect modelling is given in [6]. The authors

use mathematical modeling of the soaking operation, i.e., the operation in which the

preservation salt has to be removed from cured hides for their further processing. In the series

of processes in which raw hide is transformed into leather, soaking is of great importance in

view of the quality of the final product, especially in the cases when the hides are cured with

solid rock salt. Incorrectly performed desalination may cause extensive damage to thinner

parts of cured hides and consequently lead to substantial economical losses. The damage can

be observed under scanning electron microscope as a collapse of the hide texture in the areas

close to the surface and is caused during rapid desalination due to the presence of great

osmotic (concentration) changes, so-called concentration shock. As a result, there are

considerable losses in the hide matter and also subsequent increase in the COD in the

effluents. In the said publication, mathematical models were worked out to define the

conditions under which the concentration shock occurs, and based on the models and

experimental verification, a technology of through-flow washing was proposed to preserve



the quality of fine fiber structure of the hides and eliminate the increase in COD in tannery

effluents.

OPTIMIZATION OF RAW HIDE PRESERVATION

Curing and soaking of raw hides are heterogeneous processes in which transport

phenomena often play crucial role. Description of the transport phenomena accompanying the

curing or desalting is closely related to the hide texture which predetermines not only the

properties of final leather, but also the demands on the hide curing and soaking [Bailey].

The hide texture is in close relation with the inner arrangement of collagen fibers. This

arrangement gives the hides their unique rheological properties [7] particularly their

substantial rigidity and tensile strength. On the other hand, the elasticity of hides results from

the presence of elastin, which unlike collagen shows relatively low rigidity. These properties

influence the mechanical action of the technological devices (tanning drums), which can

positively affect the operation times. The overall mechanical properties of hides are then

given mostly from the ratio of the said components and their spatial arrangement [8]. Other

important factor is the interfiber and intercellular space, the major part of which is liquid. The

presence of this free space filled with liquid ranks hides among porous materials

characterized by physical quantity called porosity, which is defined as a ratio of free volume

(pores) to the total volume of the solid material.

In addition to the “free” liquid (water) present in the pores, the hide also contains

considerable amount of water bound in the native structure of collagen. This water plays

important role in stabilization of the collagen macromolecule conformation by influencing the

chemical and physical properties of the hides as well as the value of the effective diffusion

coefficients. There are generally two ways of how water is bound in collagen structure –

intramolecularly within the collagen triple helix, and intermolecularly between peptide chains

of different macromolecules [9]. Free water can be removed from hides by increased

pressure, resulting in the reduction in the hide volume. The reverse process is called

collagen/hide swelling. On the contrary, bound water remains in collagen structure even after

pressure increase over 70 MPa. In this stage, the hide contains about 45 % of water from the

original 60-70 %. This water shows some characteristic features different from “common”

water, such as losing the ability of electrolyte dissolving, freezing at -20°C and change in

collagen structure (roentgenogram) after its removal [9]. About 2.5 % of this water is in

bound very firmly and can only be removed by drying at tempreratures above 100°C, which

leads to structural changes in collagen macromolecules [9].

From the viewpoint of water distribution, cured hides can be classified as of “good,

medium and poor consistence”, which substantially affects the soaking (desalting) processes

[10]. The authors of this work observed slower removal of NaCl from hides of low moisture

content, the reason being supposedly the closure of the majority of pores due to conformation

changes in dehydrated collagen as a result of overdrying. Decelerated release of NaCl from

cured hides also occurs in the hides with higher content of fat on the flesh side.

One of indispensable functions of skin (hide) is preventing undesirable microorganisms

from getting into the body. As soon as the animal is slaughtered, microorganisms start to

penetrate through this protective barrier, reproduce and cause damage to the hide structure



[1]. For this reason it is crucial to protect fresh hides from microbial decay by preservation

(curing).

There are generally two ways of preservation – short- and long-term. Short-term

preservation is effective within days up to few weeks and involves various methods of

cooling and/or antiseptics and biocides. Long-term preservation comprises drying, freezing

and salting. Wherever possible, fresh hides should be fleshed in order to dispose of fat and

excess subcutaneous tissue (so-called green fleshings). The fleshing operation results in

considerable reduction in the hide weight, facilitates preservation and the generated organic

waste does not contain additional chemicals, which makes it more valuable for its possible

processing.

If a beamhouse is located within reasonable distance from the abattoir, fresh hides are

cooled to preserve their natural, chemically untreated state. The cooling can be performed in

several ways such as chilling (cooling to 10-15°C), refrigeration (deep cooling to the freezing

point) or freezing (at temperatures dropping from -10 to -30°C) [11]. Chilled hides must be

processed by 3 days, refrigerated hides can be stored up to 3 weeks and frozen hides will last

for months. Disadvantages of the said preservation methods have rather practical aspect – the

hides must be cooled individually immediately after flaying and cannot be exposed to higher

temperatures during transportation. Moreover, short-term chilling (when the chilled hides

must be delivered for further processing to tanneries within up to 24 hours) leads to

insufficient sorting of the hides by their weight and type. For this reason, advanced method of

cooling using dry ice was developed to preserve the hides for weeks (depending on

temperature) and allowing adequate sorting before their further processing [12]. Other long-

term preservation is drying. This environmentally friendly method is still being used in the

countries with suitable climatic conditions. Drying can be applied on pelts and skins of

smaller animals (e.g., goatskins), but is not suitable for cattle hides which substantially limits

the general use of this method. There are also problems resulting from too fat or too slow

drying and last but not least, the drying skins can be attacked by insects [1].

Curing with common salt (sodium chloride, NaCl) is the most common way of long-term

preservation of raw hides [13]. At moderate temperatures, salt-cured hides can be stored up to

1 year, depending on the way of salting [1]. The use of crystalline NaCl is more common in

Europe, while in the United States the most prevalent method is brine curing with saturated

solution of NaCl [9]. In his comprehensive study [1] Bailey describes the dual effect of salt

curing – for one thing, addition of salt removes water from the hide and thus reduces bacteria-

friendly conditions inside the hide, and for another the presence of salt kills the organisms

adapted to normal osmotic conditions. Preservation with sodium chloride is advantageous

from several points of view – it is relatively easy, cheap and suitable for transportation.

However, it also entails numerous disadvantages.

One of them is that hides cured with saturated brine solution are susceptible to attacks of

halophilic bacteria, it means bacteria adapted to habitats with increased salt concentrations.

There is a wide range of microorganisms capable of growing in saline environments. These

organisms can be generally classified as halotolerant, which are able to grow at moderate salt

concentrations although their optimal growth is in the absence of NaCl, and halophilic, the

growth of which is directly fixed on saline conditions. Halophilic organisms can be further

classified as mild, moderate and extreme halophils on the basis of the range of optimal salt

concentrations. Mild halophiles require 1-6% salt, moderate halophiles 6-15% salt and

extreme halophiles require 15-30% NaCl [14]. Some of these bacteria called haloarchaea can



be found even in saturated brine solutions used for raw hide curing. They contain high

amount of carotenoid compounds and cause red stains on brine cured hides, called the red

heat. While it was believed in the past that that the presence of halophils only indicates the

presence of other microorganisms which deteriorates the quality of final leather, more recent

publications imply otherwise. Bailey and Birbir [15] carried out detailed experimental study

on the action of halophilic bacteria and its correlation with the damage of hide structure. The

results of their research suggest that halophils themselves have an adverse effect on the

quality of leather made of attacked hides. This damage is mild or even negligible when cured

hides are stored at moderate temperatures and humidity, but with increased time of storage (as

early as 7 weeks and longer) and at higher temperatures the damage can be substantial.

The occurrence of red heat can be also influenced by the origin of salt used for curing.

Sea salt and salt from salt lakes generally contain higher numbers of halophilic bacteria

compared to rock salt. Birbir and Eryilmaz [16] found out that crude salt from salt lakes

contains up to 105-10

6 cfu/g of extremely halophilic bacteria (where cfu stands for colony

forming units and the authors of [17] describe about 82 strains of halophilic bacteria living in

natural sources of salt including salt lakes. High occurrence of these organisms substantially

limits direct utilization of this salt for raw hide preservation. On the contrary, sea salt dried in

drying oven contains practically no halophilic bacteria [18]. Other limitation of utilization of

salt from salt lake is higher concentration of particularly calcium and magnesium ions, which

in the case of calcium salts can lead to formation of salt stains and in the case of magnesium

salts to the formation of salt stippen (star-like elevations visible after de-hairing) [19].

Preservation method also plays a role in the occurrence of halophilic microorganisms.

Hides cured with solid salt are less susceptible to halophilic bacteria attack than those cured

in brine solution. It is said that rock salt cured hides generally contain 103-10

4 cfu/g, while

brine cured hides contain up to 106-10

7 cfu/g [20]. The difference comes from the fact that in

curing with dry salt only fresh and clean salt is used; on the contrary, in brine curing many

hides go through one brine solution, which makes it a bacteria reservoir that inoculates every

other hide put into the solution. The degree of bacterial contamination rapidly increases with

higher temperatures and the amount of hides going through one brine solution.

Red heat can be prevented by addition of specific substances aimed directly at halophilic

strains of bacteria. Orlita [21] proposes a combination of 2 % sodium carbonate and 2 %

naphthalene, or possibly 2 % zinc diethyldithiocarbamate (related to the weight of curing

salt). Vreeland et al. [22] inhibited halophilic microorganisms bile salts applied directly onto

cured hides, or as an additive in brine solution. Birbir and Eriylmaz [16] describe the

possibility of using natural antimicrobial substances called halocins which are produced by

some halophilic bacteria strains and serve as inhibitors of competitive halobacteria inhabiting

the same ecosystem.

Another disadvantage of salt curing is corrosion on transportation devices and generally

on all industrial equipment used in curing [1]. However, the most serious problem related to

salt curing is its adverse effect on the environment, especially on the amount of total

dissolved solids (TDS) in tannery effluents [23]. For this reason, alternatives to salt curing

have been sought intensively, based both on substantial reduction in the amount of salt used

or its complete substitution with other methods or chemicals.

According to the Leather International [24], the global trends in reduction in the amount

of salt in tannery effluents are the following:



low-salt pickling with polysulphonic acid

salt-free liquid chemicals

recycling of salt from waste waters (e.g., with membrane technologies)

Our approach lies in optimization of the brine curing process to reduce the total amount

of salt entering the hide-processing technological chain, keeping all the other advantages of

this method of preservation.

One of the alternatives is preservation with chemicals with antimicrobial and/or

fungicidal properties, which at the same time meet the required hygienic, economic and

environmental parameters. Antibacterial action is of major importance in raw hide curing,

because raw hides are susceptible to bacterial deterioration, while processed hides (for

example those tanned to the wet blue) are more likely attacked by mould [25]. An important

environmental parameter of mainly bactericidal substances is whether they show an adverse

effect on the activated sludge used in waste water treatment plants. Hides treated with

antibacterial chemicals last for weeks. Some of these chemicals were or have been used

commercially, although none of them have found global application. Examples can be

bacteriostatic substance Silostan developed in former Czechoslovakia [26] or Liricure

developed in South Africa, which was originally used for preservation of goatskins and later

on its application was extended to cattle hides [27]. Its wider application was limited by high

operation costs, though its environmental impact was considerably lower than that of sodium

chloride.

Mitchell [28] introduced a biocidal preparation Proxel X in combination with boric acid,

which according to Bailey [1] represents the best option for short-term raw hide preservation.

Other commercially used chemicals are for example didecyldimethylammonium chloride [25]

and products on the basis of dimethyldithiocarbamate [29] or 2-thiocyanomethy-

lthiobenzothiazole [30], especially because of their minimal environmental impacts and low

health risks. These substances are usually used in the amount of 0.02-0.1 % related to the

weight of raw hides. At the said dosage and usual dilution in waste waters, the concentration

of these substances is generally below the levels which may negatively affect activated sludge

in waste water treatment plants and their residues in waste waters are usually eliminated

before biological treatment [25]. Hughes [31] describes raw hide preservation in a bath

containing concentrated solution of boric acid. Bailey and Hopkins [32] applied sodium

sulfite in combination with acetic acid and successfully tested this method on industrial scale

in a small abattoir in Illinois. Another promising alternative is according to Vankar and

Dwivedi [33] treatment of raw hides with sodium sulfate. Their study is supported by

successful experiments on both goatskins and buffalo hides, which represent one of the major

raw materials for Indian leather industry. The method with sodium sulfate is considered

prospectively sustainable also in more recent bibliographical sources [34].

Another chemical alternative to sodium chloride is its replacement with potassium

chloride (KCl) [35]. Hides cured with KCl have analogical properties as those cured with

NaCl and the excess KCl can be used as potassium fertilizer. Moreover, hides treated in this

way do not show signs of halophilic bacteria attacks. Its disadvantage is relatively high price

compared to NaCl and higher dependence of KCl solubility on temperature.

Radiation is another environmentally friendly alternative. It can be performed either in

the form of gamma [36] or electron beam irradiation [37]. While gamma irradiation has not



found wider application due to limited sources of gamma rays, electron beam irradiation has

been used commercially to a limited extent (its limitations come from high investment costs

and the need for additional biocidal treatment of irradiated hides). However, the latter method

is listed in the reference document of the European Commission [29] among BATs (best

available techniques).

Raw hides can also be preserved by pickling [1,9]. This operation leads to partial removal

of non-fibrous proteins which are most susceptible to bacterial degradation. Hides are

commonly pickled in 10 % solution of NaCl and 1% solution of sulfuric acid. Pickling can be

applied to light skins such as lambskins and goatskins. Relatively low pH (around 3.5)

protects the skins from bacterial degradation and the skins can be stored for several months.

The authors of the American patent [38] describe other “non-salt” method, namely

fatliquoring prior to tanning, which is subsequently followed by de-liming, bating and

pickling. This procedure leads to dry, flexible and preserved hides ready for the transportation

and tanning.

Quite common and efficient preservation method is chrome tanning to a product called

wet blue [1]. This method is suitable also for cattle hides and other heavy hides and reduces

the content of salt in waste water. The wet blue stock can be further treated with fungicides or

antibacterial preparations based on dimethyldithiocarbamate. In his comprehensive study [1]

Bailey also mentions the possibility of bio-preservation of raw hides with bacteria-produced

peptides (so-called bacitracins). Bacitracins were first isolated in 1945 from the bacteria

species Bacillus subtilis var Tracy [39]. Although their medicinal application is limited due to

high toxicity of these substances, they show biocidal effect even on bacteria species resistant

to most of the existing antibiotics including strains of exteremely halophilic bacteria (for

example in combination with novobiocin [17]). As far as the authors are concerned, this

research has not been brought to a commercial use so far, despite excellent preliminary results

mentioned by Bailey [1].

Other group of methods does not avoid utilization of NaCl completely, but works on the

basis of partial substitution of the salt load with other substances. Salt curing can be

combined with other preservation agents such as superabsorbent (co)polymers (SAP) [40], p-

chloro-meta-cresol after prior dehydration of the hides or sodium metabisulfite Na2S2O5 [41,

42]. The authors of the U. S. Patent [43] describe application of preservation additives such as

essential oils with antibacterial properties in combination with non-ionic surfactants that

enhance the diffusion of NaCl into the hides and thus make the preservation procedure more

efficient. Kanagaraj et al. [44] tested raw hide preservation with 5 % solution of boric acid, as

well as a preservation system composed of 2 % boric acid and 5% solution of NaCl. The

reference document of the European Commission [29] describes a method called “flo-ice”

which is commonly used in the fishing industry for fish preservation. “Flo-ice” is made in a

refrigerator device containing 3-5 % solution of NaCl to stay in the liquid state, at the

temperatures between 0 to -10°C. Hides cured in this way can be stored in waterproof

containers for up to 3 days for which the temperature remains under 10°C. This period can be

modified with the change of NaCl concentration. This method requires initial investment in

the “flo-ice” generating machinery.

Research in less-salt and salt-less alternatives is only one of the possible approaches how

to reduce the salt content in tannery effluents. Close attention is also paid to salt curing itself

since it still is the most commonly used and available way of preservation. Generally there are

two main research areas. We opened our contribution saying that the proverb “prevention is



better than cure” applies to many industrial areas and raw hide preservation is no exception.

Let us briefly start with the “cure”, in other words methods of salt regeneration once it is

dissolved in tannery effluents. The authors of [23, 45] describe methods for salt regeneration

from technological waters after soaking and its re-use in the preservation process. However,

the authors of [46] point out that generally this approach is not profitable from the economic

point of view. Re-use of such salt is limited due to organic contamination and fresh salt is

widely available at reasonable prices. Up to 10% of the spent salt can be mechanically shaken

out in suitable drums from hides cured with solid salt. After elimination of solid impurities,

this salt can be re-used in pickling, but cannot serve for curing due its contamination with

halophilic bacteria. The authors of [23] proposed a laboratory-tested procedure of NaCl

regeneration from brine solutions and its re-use in goatskin preservation; however, the

publication does not provide detailed analysis on the presence of halophilic bacteria. Reverse

osmosis and other membrane technologies have been discussed recently in relation to

reducing the salt content in tannery effluents. Basically, reverse osmosis (RO) has proved a

powerful tool in water desalination; for example, RO-based technologies are used in the

majority of sea water desalination plants [47]. Despite the fact that in principle RO also

significantly reduces the salt content in tannery effluents [48], the nature of tannery effluents

brings challenging tasks particularly from the viewpoint of optimization and economical

feasibility of RO applications. High organic content of tannery effluents leads to rapid fouling

of RO membranes with a consequent reduction of performance, which makes it necessary to

involve pre-treatment steps such as ultrafiltration or membrane bioreactors [48]. General

problem of pressure-driven membrane technologies including RO is the occurrence of

concentration polarization, which is an increase in concentration of salts dissolved in the feed

solution in the stagnant boundary layer at the membrane surface. The boundary concentration

thus exceeds the concentration of the bulk feed, leading to reduction of the feed flux rate

and/or formation of crystals that plug the membrane pores [49]. This undesirable

phenomenon cannot be eliminated completely, but can be reduced to some extent by

implementation of turbulators. This phenomenon also complicates mathematical modeling of

membrane systems since experimental determination of the boundary concentration is

exacting. Despite rapid progress in membrane technologies and gradual reduction of

operating costs, this way of desalination can still be financially unavailable particularly in

developing countries [50].

Another approach falls into the category "prevention" and lies in the application of

theoretical tools of chemical engineering, i.e., mathematical modeling of the chemical-

physical processes involved in particular technological procedures. In the case of salt curing it

means optimization of the main operating costs which is related to the determination of the

time necessary for proper curing and the consumption of salt, technological water and

energies. Mathematical description of the transport phenomena related to the transport of

NaCl in the hides has been so far mostly limited to soaking, which is in simple terms the

opposite of curing.

The first mathematical description of cattle hide soaking was published by O'Brien [51].

Detailed mathematical models of the soaking operation in a real industrial device (drum or

mixer) at repeated washing and a relatively low ratio of the volumes of the liquid and solid

phase, respectively, is given in [10]. The authors provided individual description of the initial

and later phase of soaking due to the different mechanisms of the said processes. They also

compared the values of transport parameter λ obtained from experimental data in fleshed



hides and hides without prior fleshing. The results of their work contain some important

conclusions that can be generalized also the curing; the authors state that the sorption of

sodium chloride to the hide matter (collagen) can be neglected, the diffusion takes place along

the real pore length which is not significantly influenced by hide swelling, the presence of

subcutaneous tissue substantially decelerates NaCl transport and soaking is slower in hides

with high percentage of dry matter. The said results were further applied in optimization and

control of the soaking operation [52]. Kolomazník et al. [53] modeled soaking in tanning

drum as a non-isothermal and non-adiabatic reactor with implementation of the continuous

reaction model [54].

Most of published mathematical models of salt curing have been related to food-

processing industry; sodium chloride diffusion has been described in meat and meat products

[55-57] and cheese [58]. There is also information on the experimental determination of the

effective diffusion coefficients, for example in cheese the values range between 1.0-5.3 ×

10−10

m2·s

−1 at the temperatures between 10-15°C, regardless on the cheese type and its

composition [59]. Experimental determinations of transport parameters are mostly based on

the relationships stated in Crank‟s excellent monograph [60]. Dehkordi et al. [61] presented

numerical solution of mathematical model of salt diffusion in potatoes and the description of

NaCl concentration fields in various times of curing.

There are practically no bibliographic records of mathematical modeling of raw hide

curing, with the exceptions of [13, 62]. The latter publication gives mathematical description

of concentration field and time dependence of NaCl concentration in a homogeneous sodium

chloride solution for various ratios of the liquid phase volume to the volume of raw hide

(soaking number). Minimal time necessary for proper curing is achieved by high values of

concentration gradient of NaCl on the hide surface, which is in the case of the said

publication possible at high soaking numbers. High concentration gradient leads to faster

curing and thus to saving on power consumption; on the other hand, it requires large quantity

of water and salt. Nevertheless, high values of concentration gradient can be also achieved by

the addition of solid salt into saturated brine solution. After termination of the curing process

there is a saturated brine solution left, which means that all the solid salt has been spent on the

curing [62].

DESALTING OF ANIMAL FLESHINGS

The "salt issue" in the leather industry is not restricted only to curing and soaking, but is

of great importance also in the processing of wastes generated by the said industry. Saline

liquid wastes and methods of salt regeneration have been already discussed above, but there is

also huge amount of solid waste to be taken into account. The possibilities of effective and

commercially interesting utilization of this waste, particularly primary waste generated in the

initial phase of raw hides processing, is a long-term global topic. Animal fleshings are a

typical example of such waste, representing more than 50 % of the total waste generated by

the leather industry [63] and at the same time the main protein solid waste of the said industry

[64]. The fleshings are produced during the fleshing operation, in which the remaining flesh

and fat are removed. The fleshings are referred to as “green fleshings” or “pre-fleshings” if

they are obtained before liming, or “limed fleshings” if hides are fleshed after the liming



procedure. Fleshing after liming brings practical advantages since the hides are de-haired, free

of major impurities and swollen after being placed for several hours in concentrated solution

of calcium hydroxide Ca(OH)2. However, this way also leads to creation of limed waste with

specific requirements for treatment and disposal, and limited utilization of the fat fraction

from limed fleshings. The fat partially hydrolyzes in strong alkali conditions of Ca(OH)2,

resulting in the losses of valuable raw stock. Moreover, with reduced fat increases the content

of free fatty acids which deteriorates the quality of the fat fraction.

Fleshings generally contain high portion of water and a balanced ratio of fat, protein and

carbohydrates [63]. In addition to that, green fleshings contain considerable amount of the

curing salt – sodium chloride [65]. The protein content substantially complicates industrial

utilization of fleshings since it decomposes quickly and the decomposition products produce

unpleasant odors. For this reason, the majority of waste fleshings is incinerated or rendered.

Both ways are demanding from the viewpoint of both time and power consumption and lead

to economically uninteresting products. Incineration is frequently burdened with disposal fees

and also disputable from the environmental and public health aspect due to formation of toxic

dioxins in incineration products.

For the reasons described above, great efforts are being made recently to reduce the

health and environmental risks of fleshings processing as well as to find technologies to

convert fleshings into products with high economical potential. Examples of such attempt can

be found in [66], the authors of which treated fleshings with pancreatic enzymes; a

combination of enzymatic treatment and ultrasound is presented in [67] and fleshing

processing with acid hydrolysis using formic and propionic acids is described in [68].

Microbial treatment has been described as an interesting alternative to the chemical treatment

and is reported to be effective on both green and limed fleshings. The authors of [69] used

fleshings as a substrate for bacteria of the species Pseudomonas aeruginosa for the

production of alkaline protease, and Kumar et al. [70] tested microbial processing of tannery

protein waste for the production of mesophilic protease. To make a conclusion from the

above cited papers, the most frequently used ways of fleshings processing are chemical and

biological (or combinations of both), the products being used as fertilizers, feed additives, or

production of amino acids or enzymes. Great attention has been paid recently to utilization of

the fat fraction as an alternative source of energy (e.g., [71, 72]), the most promising [73] and

studied of which has been the production of biodiesel as an alternative to the existing

conventional biodiesel production from vegetable oils.

It is obvious from the available bibliographic sources on fleshing utilization that usually

only one of the two fractions (fat or protein) is processed. For overall and economically

profitable fleshing treatment it is advantageous to process both fractions simultaneously.

Research in this area has been very rare so far, with the exception of [63,64], or possibly [65].

The reasons for this are technological challenges to be overcome. One of the main

challenging tasks is the content of sodium chloride in the fleshings, which particularly

complicates effective processing of the protein fraction to commercial products such as

gelatin. Gelatin quality (and consequently its price) is derived from the gel strength (Bloom

value). This value decreases with higher ash content, in the case of fleshings the content of

sodium chloride. Desalting of fleshings is then one of the key operations in the technology for

its complex processing and gaining the maximal profit from the products.

In the following paragraph we would like to briefly summarize the background of our

research and set the objectives of our contribution. With reference to raw hide preservation,



most of the published papers deal with the properties of a particular preservation agent. The

most common way of preservation so far is curing with sodium chloride or its saturated

solution. Very few studies provide an engineering description of the curing process,

particularly by application of theoretical tools of chemical engineering for the description of

non-stationary concentration fields and time dependence of the salt concentration in the hides

or in the aqueous bath in the case of desalting. Quantitative relations of the said dependencies

can be then implemented in practical optimization for particular operation conditions. The

main purpose of our contribution is reducing the existing environmental burden related to the

worldwide use of sodium chloride for raw hide curing. The essential step is rationalization of

the curing process itself, since it determines the amount of salt to deal with in the subsequent

operations such as pre-soaking, soaking, after-soaking and desalting of animal fleshings

during their complex processing into valuable products. We focus on general application of

quantitative description of sodium chloride diffusion in various operations the mathematical

description is basically the same and differs only in the initial and boundary conditions.

MATHEMATICAL MODELING

Mathematical Model of Raw Hide Curing in Permanently Saturated Brine

Solution

Mathematical model of raw hide curing with pseudo-homogeneous sodium chloride

solution was first presented in [62] and is based on the model of continuous reaction [54].

Taking into account the observations from [1, 13], let us presume that the diffusion proceeds

asymmetrically into the hide inner volume from the flesh towards the grain side, forming a

non-stationary concentration field (Figure 1).

Figure 1. Mathematical model of sodium chloride diffusion into the hide iner volume during curing with

permanently saturated brine solution.



The model is represented by a partial differential equation of parabolic type (Flick‟s

Second Law) (1) under specific initial and boundary conditions. The complexity of the real

diffusion process is projected in the value of effective diffusion coefficient (Def) of sodium

chloride ions into the hide inner volume. The value of Def can be dependent on concentration

as well as on time. For a “first approximation”, let us assume that the value of Def does not

differ significantly with concentration. Experimental verification of this assumption is

described in the Experimental part. Considering the above stated limitation, the quantitative

model of raw hide curing can be described in the following form:

,,2

2

xx

cDx

c

0 (1)

0,0

x

c (1a)

scbc , (1b)

00, xc (1c)

Boundary condition (1a) says that the problem is not solved symmetrically, in other

words that the diffusion proceeds only from the flesh side towards the grain. Condition (1b)

expresses ideal mass transfer between the brine solution and the raw hide surface (the bath is

intensively stirred and the NaCl concentration in the bath is equal to the salt concentration in

the pores on the raw hide surface. Moreover, it is an open system where NaCl concentration

in the bath does not change with time and represents a permanently saturated brine solution

(which in practice is created by addition of solid salt into saturated brine solution. Initial

condition (1c) (i.e., when τ = 0) says that at the beginning of the curing process the

concentration of salt in the hide is zero. For the solution of the mathematical model it is

advantageous to introduce the following dimensionless parameter and variables:

sc

cC

(2a)

b

xX (2b)

2b

DFo

(2c)

The dimensionless mathematical model together with the initial and boundary conditions

is represented by equations (3) – (3c):



FoXX

CFoX

Fo

C,,

2

2

0Fo (3)

1,1 FoC (3a)

0,0

Fo

X

C (3b)

00, XC (3c)

Taking into account that the presented mathematical model is linear, it is possible to

apply Laplace transformation for the solution of the partial differential equation. The

analytical solution (4) describes a dimensionless non-stationary concentration field of sodium

chloride in raw hide:

1 112

212cos

11

4

2212

n n

enX

Cn

nFo

(4)

Graphical expression of equation (4) is given in Figure 2:

Figure 2. Dimensionless concentration field of sodium chloride in hide during the curing process.

For the determination of time necessary to achieve certain degree of NaCl concentration

in the hide, we have introduced the integral mean concentration, which can be calculated

according to equation (5):



dXFoXCC

0

, (5)

After integration, we get (6):

122

4

2212

32

12

121

n n

eC

nFo

n

(6)

Dimensionless time dependence of the integral mean concentration is depicted in

Figure 3:

Figure 3. Dimensionless time dependence of the integral mean concentration of sodium chloride in raw

hide.

Mathematical Model of the Desalting of Animal Fleshings

Desalting is the first from the fleshing pre-treatment procedures in the complex

technology of industrial processing of fleshings into valuable products, both fat- and protein-

based. The fleshings are stirred in cold water, resulting in the formation of small spheres –

pellets. Mathematical model of the desalting of fleshings was first published in [74] and

describes diffusion of sodium chloride from the pellet inner volume into the surrounding

aqueous bath (7):



r

rc

rr

rcD

rc

,2,,

2

2

10 Rr (7)

The model was solved under initial and boundary conditions given by equation (7a-e).

The following condition says that the problem is solved as symmetrical, in other words that

NaCl concentration is highest in the center of the pellet:

0

,0

r

c (7a)

We further assume that NaCl concentration in the beginning of the desalting process, i.e.,

when τ = 0, is constant for any point in the solid phase. This assumption is expressed by

boundary condition (7b).

pcrc 0, (7b)

The following initial condition (7c) says that in the beginning of the desalting process

there is no salt present in the aqueous bath.

000 c (7c)

Equation (7d) stands for the assumption of an ideal mass transfer, which can be ensured

by ideal stirring in the reactor.

01, cRc (7d)

Equation (7e) is the balance boundary condition which says that the diffusion flux of

NaCl on the pellet boundary is equal to the accumulation of the diffusion substance in the

volume of the surrounding aqueous bath (V0).

0

01, c

Vr

RcSD (7e)

Again, we have introduced dimensionless parameters and variables defined by equations

(8a – d).

p

p

c

ccC

(8a)

pc

cC 0

0

(8b)



2

1R

DFo

(8c)

1R

rR (8d)

By substitution of the said dimensionless variables into (7) we obtained mathematical

model of the desalting operation in dimensionless form (9), together with the related initial

and boundary conditions (9a-e):

0 10

,2,,2

2

FoR

R

FoRC

RR

FoRC

Fo

FoRC (9)

0

,0

R

FoC (9a)

00, RC (9b)

000 C (9c)

01,1 CFoC (9d)

2

1

3

1

00 4 3

4 , kde

3,1 , RSRV

V

VNa

Fo

FoCNaFo

R

CS

S

(9e)

By integration of (9) and implementation of the said conditions we obtained analytical

solutions of the model (10-13). Equation (10) represents dimensionless concentration field of

sodium chloride in the fleshing pellets:

13

1cos

11

3sin

expsin

3

2

23

2

n Na

qq

qqq

Naq

qFoqR

qR

Na

Na

NaC

n

n

nnn

n

n

n

n

(10),

where qn are roots of transcendent equation (11):

n

nn

q

qNaq

1

3

cotg

(11)

Graphical expression of equation (10) is given in the following Figure 4.



Equation (12) defines dimensionless concentration of NaCl in the surrounding aqueous

bath.

Figure 4. Dimensionless concentration field of sodium chloride in the fleshing pellets.

Figure 5. Time dependence of dimensionless salt concentration in the aqueous bath on dimensionless

water consumption (Na).



19

1

exp

3

2

2

22

2

0

n qNaNa

qFoNa

NaC

n

n

(12)

The following Figure 5 shows the time dependence of dimensionless concentration of salt

in the bath for various dimensionless water consumptions (soaking number, Na).

One of the most important parameters in optimization of the desalting process is the

efficiency – the degree of desalting (y). The efficiency can be calculated according to the

following equation (13):

Figure 6. Time dependence of the efficiency of the desalting process (y) for various soaking numbers

(Na).

19

1

exp

3

2

2

22

2

2

2

000

n qNaNa

qFoNa

Na

NaNaC

VC

CVy

n

n

Sp

(13)

Symbol qn stands for the roots of transcendent equation (11) and Na denotes the ratio of

the volume of aqueous bath to the volume of solid phase (fleshings), also referred to as so-

called soaking number.

Graphical expression of equation (13) is shown in Figure 6.



EXPERIMENTAL

Determination of the Effective Diffusion Coefficient of NaCl

As a reference value, we calculated the value of diffusion coefficient of sodium chloride

at infinite dilution according Nernst equation (14) [75]. This equation is sometimes referred to

as Nernst-Haskell equation [76], where the value of 8.9 × 10-10

is expressed as the universal

gas constant divided by Faraday constant, and the expression ʌ0 stands for

00

ll .

zz

zzllTD

0

0010

0 10931,8 (14),

where D0 [cm2·s

−1] is diffusion coefficient of NaCl at infinite dilution, T [K] is absolute

temperature, [S·cm2·mol

−1] stands for limit ionic conductances for cations and anions, ʌ0

[S·cm2·mol

−1] represents the sum of limit ionic conductances of cations and anions and z+, z−

[1] stand for the valences of cations and anions.

The value of diffusion coefficient of NaCl at infinite dilution and 20°C calculated

according to equation (14) is 1.58 × 10-9

m2·s

-1.

Critical dimensionless time (Fourier diffusion criterion) FoK corresponding to the

required salt concentration in the hide can be read from Figure 3. For the calculation of the

real time (τ) of the curing operation we need to know the value of the effective diffusion

coefficient (Def) and the thickness of raw hide (b). By implementation of the definition of

dimensionless Fourier criterion given by equation (2c), we obtain the following relation (15)

for the real time of curing:

K

Fo (15),

where 2b

D . (15a)

We have introduced the important transport parameter λ that involves the hide thickness,

tortuosity and the effective diffusion coefficient. The transport parameter can be calculated

from experimental data obtained during the desalting of fleshings. In the case of the initial

phase of desalting and the assumption of very low values of effective diffusion coefficient

and relatively large particles of the solid phase, the time dependence of dimensionless salt

concentration in the bath can be calculated according to the following modified equation from

Crank [60]:

Na

NaC

12 (16)



With rearrangement of (16) we get linear dependence of dimensionless concentration on

the square root of time (17) and the transport parameter λ can be estimated from the slope of

the straight line.

2

14

Na

Nak (17)

The transport parameter λ is defined as diffusion coefficient at low concentrations divided

by the second power of the pore half-length. In our case, we can replace diffusion coefficient

with the value of the effective diffusion coefficient and the pore half-length with the

equivalent half-thickness of the fleshing pellets (Rm). Effective diffusion coefficient can then

be calculated as:

2

mef RD (18)

For the determination of the effective diffusion coefficient of salt in the inner volume of

the solid phase it was necessary to measure the conductivity of the bath on the square root of

time. After recalculation of conductivity to dimensionless concentration, we obtained the

graphical expression of equation (16). After linear regression of the data, we estimated the

value of the transport parameter from the slope of the straight line according to equation (17).

The following Figure 7 gives an example of the graph obtained from the experimental data:

Figure 7. Graph of the dependence of dimensionless salt concentration in the bath on the square root of

time during the desalting of fleshings.

The experiment work was carried out as follows: the electric conductivity was measured

in a mixture of animal fleshings obtained from a local tannery and distilled water. Initial

conditions of the experiment are given below. The heterogeneous mixture was intensively



stirred to eliminate the effect of inner diffusion. The conductivity was measured by a probe

every minute for one hour.

Example of the experiment:

Weight of the solid phase (fleshings) 202 g

Density of the solid phase 0.900 g∙cm-3

Equivalent half-thickness of the solid phase 0.125 cm

Volume of the aqueous bath 600 ml

Soaking number 2.67

Conductivity corresponding to the initial

concentration of the bath 2 µS∙cm-1

Conductivity corresponding to the equilibrium

concentration of the bath 160 µS∙cm-1

Dimensionless concentration was calculated from the measured conductivity divided by

the conductivity corresponding to equilibrium concentration, where from each member was

subtracted the value of the conductivity corresponding to the initial salt concentration in the

bath.

The resulting effective diffusion coefficient was 5.582 × 10-10

m2∙s

-1.

OPTIMIZATION OF RAW HIDE CURING

The obtained effective diffusion coefficient can be approximated also for the case of raw

hide curing. Proceeding from equations (15) and (15a), the minimal time necessary for proper

curing can be calculated for raw hides of various thicknesses. The 85 % hide saturation with

salt, which is according to [33] the minimal value for proper curing, corresponds to the value

of FoK = 0.125. Implementing the value of Def = 5.582 × 10-10

m2∙s

-1 and for the hide

thicknesses of b = 5; 10 and 15 mm, the corresponding minimal times for proper curing are

1.6; 6.2 and 14 h, respectively. In common practice according to [13] the hides are cured in

practically saturated brine solution for the minimum of 18 h at soaking numbers around 5.

The results of our work thus not only helps to save the time for which the hides must

remain in the curing raceway, but termination of the curing process right at the time of

reaching the 85 % saturation leads to considerable savings in the amount of the preservation

agent and electric power consumption. The fact that the cured hides do not contain

unnecessarily higher amount of salt also substantially reduces the amount of salt to deal with

in the effluents generated during the soaking operation and thus reduces the environmental

burden. Moreover, modeling of raw hide curing under the assumption that the maximal

concentration gradient is maintained by permanently saturated NaCl solution and not by high

soaking numbers is beneficial also from the viewpoint of water consumption.

The experimentally determined value of the effective diffusion coefficient is lower by

one order of magnitude than the theoretical value calculation according to Nernst equation at

infinite dilution (14), but is in good agreement with experimental values calculated for

soaking of cured hides [10, 77]. The speed of diffusion can be influenced by many factors,

one of the most important of which is the nature of the porous material itself, in other words



the arrangement of the collagen fibers in the hide, and its homogeneity. For comparison,

experimentally determined effective diffusion coefficients in meat and meat products, it

means materials composed mostly of highly organized molecules of myosin, were generally

in the same order of magnitude (× 10-10

m2∙s

-1) as those estimated for raw hides, while in

potatoes (consisting mostly of polysaccharide – starch) the diffusion coefficients were

ranging between 3.45 – 4.39 × 10-9

m2·s

-1 [61] in dependence on the initial salt concentration

in the curing bath. Other factors which can affect the speed of diffusion are for example poor

fleshing, agglutination of protein fibers, or high portion of dry matter in the hides (fleshings).

In contrast to soaking, the latter factor probably does not play important role in the curing

since it is unlikely that fresh hides overdry before curing.

ECONOMIC PARAMETERS OF THE DESALTING OF FLESHINGS

The basic economic parameters of the fleshing desalting operation have been described in

[74]. The main operating costs of the desalting involve the sum of the unit electric power

consumption necessary for stirring, and the overall washing water consumption:

VVE VKPKN (19)

Consumption of the washing liquid (water) (VV) [m3] is an implicit function of time (τ)

[h], which is calculated according to equation (13) for the required desalting efficiency (y).

Symbol KE [USD·kWh-1

] denotes unit price of electric power, P [kW] is the input power the

electromotor for the mixer and KV [USD·m-3

] represents unit price of washing water. A

computer program has been elaborated which allows calculating the intersections for the

required desalting efficiency and various soaking numbers. These intersections are

subsequently implemented in equation (5.17), where the resulting cost function shows a

minimum. The purpose of rationalization is to control the desalting operation in this

minimum. The said publication gives also detailed description of experimental determination

of the fleshing desalting efficiency (y) during decantation washing and the related cost

functions for various levels of the required efficiency and various soaking numbers. The

results show that the proposed mathematical model can be implemented in the industrial

design of the desalting procedure for the calculated optimal conditions. The importance of

incorporation of the desalting step in the technology of complex processing of fleshings into

valuable products was proved by observation of the properties of gelatin prepared according

to patented technology of fleshing pre-treatment [78]. Gelatin obtained by the innovative

technology showed very low ash content and at the same time very high Bloom value

compared to gelatin produced without implementation of the desalting step.

CONCLUSION

Despite of global efforts to find economically viable and environmentally friendly

methods of raw hide curing with reduced or no involvement of sodium chloride, curing with

common salt still remains the most frequently used worldwide. Rationalization of the curing



process leads to shorter times necessary for proper curing and consequent reduction in the

consumption of water, energies and the preservation agent. Our contribution shows the

advantages of application of indirect modeling, particularly the theory of transport

phenomena for mathematical description of raw hide preservation for optimization of the said

technological operation. Based on the analytical solution of the mathematical model and

experimental determination of the effective diffusion coefficient of sodium chloride in the

hide (fleshings), we were able to estimate minimal time periods necessary for proper curing

of hides of various thicknesses. The acquired knowledge can be used in rationalization of

other technologies related to the leather industry and processing of its waste; in our case, we

give an example of application of the model on the desalting of fleshings during its

processing into valuable fat- and protein-based products such as quality biodiesel and glycerin

and commercial grade gelatin.

ACKNOWLEDGMENT

This work is a part of PhD thesis Barinova, M. “Optimization of Raw Hide Preservation”

[in Czech], Tomas Bata University in Zlin, Czech Republic, 2013. The work was financed by

the Ministry of Education, Youth and Sports of the Czech Republic, grant No. 7088352102

and ERDF Project CEBIA Tech No. CZ.1.05/2.1.00/03.0089. The authors would also like to

express great appreciation to the people from the Eastern Regional Research Center, USDA,

Wyndmoor, Pennsylvania, USA for a long-term beneficial cooperation.

REFERENCES

[1] Bailey, D. G. The 44th John Arthur Wilson memorial lecture: Preservation of hides and

skins. J. Am. Leather Chem. As. 98, 308-319, 2003. ISSN 0002-9726.

[2] Medium-term prospects for agricultural Commodities Projections To The Year 2010.

URL <http://www.fao.org/docrep/006/y5143e/y 5143e00.htm#Contents>.

[3] Mwinyihija, M., Magero, J., Chamining'wa, G. N., Gnanamani, A. Characterization of

hides and skins curing process and rusultant effluent impact on soil profile at

Mariakani, Kenya. In II. International Leather Engineering Congress Innovative

Aspects for Leather Industry, May 12-13 2011, Izmir, Turkey.

[4] Ozgunay, H., Colak, S., Mutlu, M. M., Akyuz, F. Characterization of Leather Industry

Wastes. Pol. J. Environ. Stud. 2007, vol. 16, p. 867-873.

[5] Kolomazník, K., Janáčová, D., Vašek, V., Blaha, A. Chemical Engineering and

Automatic Control in Leather Technology. In Advanced technologies research-

development-application. Mammendorf: Verlag Robert Mayer-Scholz, 2006, s. 475-

516. ISBN 3-86611-197-5.

[6] Kolomazník, K., Furst, T., Uhlířová, M. Relationship between mass transport and the

quality of cured hide. The Canadian Journal of Chemical Engineering, 2009, roč. 87, č.

1, s. 60-68. ISSN 0008-4034.

[7] Jeyapalina, Sujeevini. Studies on the hydro-thermal and viscoelastic properties of

leather. PhD thesis, University of Leicester, Great Britain, 2004.



[8] Lopot, František. Reologie a biomechanika. Aplikace modelů. Praha: Katedra anatomie

a biomechaniky FTVS UK. [in Czech]

[9] Blažej, A., Galatík, A., Galatík, J., Mládek, M: Technologie kůže a kožešin. SNTL

Praha, ALFA Bratislava, 1984, 451 pp. [in Czech]

[10] Blaha, A., Kolomazník, K. Mathematical model of soaking. Part I. J. Soc. Leath. Tech.

Ch. 1989, vol. 73, p. 136-140. ISSN 0144-0322.

[11] Adzet, J. M. Short term preservation of raw hides and skins. World Leather 1997, vol.

10, No. 3, p. 59-60.

[12] Endlweber, Hans. Preserving animal hides. United States Patent 4985039, 1991-01-15.

[13] Hernandez Balada, E., Marmer, W. N., Kolomazník, K., Cooke, P. H., Dudley, R. L.

Mathematical model of raw hide curing with brine. J. Am. Leather Chem. As. 2008,

vol. 103, p. 167-173. ISSN 0002-9726.

[14] Todar, K. Todar`s Online Textbook of Bacteriology. University of Wisconsin, 2009.

[15] Bailey, D. G., Birbir, M. The impact of halophilic organisms on the grain quality of

brine-cured cattle hides. J. Amer. Leather Chem. Ass. 1998, vol. 93, p. 121-131. ISSN

0002-9726.

[16] Birbir, M., Eryilmaz, S. Prevention of halobacterial damage on hide caused by lipolytic

halophylic archaea with halocins, 2006. www.aaqtic.org.ar/congresos/istanbul2006/

[17] Birbir, M., Sesal, C. Extremely halophilic bacterial communities in Sereflikochisar salt

lake in Turkey. Turk. J. Biol. 2003, vol. 27, p. 7-22.

[18] International Union of Leather Technologists and Chemists Societies (IULTCS).

www.iultcs.org/leather_terms/r.asp.

[19] CSIRO Division of Food Research, Meat Research Laboratory. Salt for hide

preservation: The preventiv of salt stains and salt stippen. Meat Research News Letter

1974, vol. 74, no. 2. http://www.meatupdate. csiro.au/data/MEAT_ RESEARCH_

NEWS_LETTER_74-2.pdf

[20] The American Leather Chemists Association, Leather Discussion (online diskusní

forum). www.leatherchemists.org/forum/, September 30, 2007. Search: “halophilic

bacteria”.

[21] Orlita A. Microbial biodeterioration of leather and its control: a review. Int.

Biodeterior. Biodegradation 2004, vol. 53, p. 157-163 ISSN: 0964-8305.

[22] Vreeland R. H., Bailey D. G., Claunch R. W. Method of using bile salts to inhibit red

heat in stored brine-cured hides and skins. United States Patent No. 5,945,027, 1999-

08-31.

[23] Preethi V., Ramesh R., Chandra Babu N. K., Sehgal P. K. Efficient salt recovery and

total dissolved solids reduction in leather processing. J. Am. Leather Chem. As. 2006,

vol. 101., no. 4, p. 117-122. ISSN 0002-9726.

[24] Salt in wastewater from leather manufacturing. URL <http://www. leathermag.com/

features/featuresalt-in-wastewater-from-leather-manufacturing/>

[25] Hauber, C. Microbicide applications in the leather industry. Directory of Microbicides

for the Protection of Materials: A Handbook. Paulus, W. (Editor), 2005, p. 317-324,

online ISBN: 978-1-4020-2818-2.

[26] Orlita, A., Ludvík, J., Navrátil, V: Krátkodobá konzervace vepřovic. Kožařství 1986,

vol. 36, p. 78.

[27] Russell, A, E., Tandt, H., Kohl, R: Liricure – Powder biocide composition for hide and

skin preservation. J. Soc. Leath. Tech. Ch. 1987, vol. 81, p. 137-142. ISSN 0144-0322.



[28] Mitchell, J. W. Prevention of bacterial damage on brine cured and fresh cattlehides. J.

Am. Leather Chem. As. 1987, vol. 82, no. 12, 372-382. ISSN 0002-9726.

[29] Integrated Pollution, Prevention and Control (IPPC). Reference Document on Best

Available Techniques (BAT) for the Tanning of Hides and Skins. European

Commission, February 2003.

[30] Gattner, H., Lindner, W., Neuber, H. U: Mikrobieller Befall bei der Lederherstellung

und seine Kontrolle mit modernen Konservierungsmitteln. Das Leder 1988, vol. 39, no.

4, p. 66-73. ISSN 0024-0176.

[31] Hughes, I. R. Temporary preservation of hides using boric acid. J. Soc. Leath. Tech.

Ch. 1974, vol. 78, p. 18-53. ISSN 0144-0322.

[32] Bailey, D. G., Hopkins, W. J. Cattle hide preservation with sodium sulphite and acetic

acid. J. Am. Leather Chem. As. 1977, vol. 72, p. 334. ISSN 0002-9726.

[33] VANKAR, P. S., DWIVEDI, A. K. Sulphates for skin preservation - A novel approach

to reduce tannery effluent salinity hazards. J. Haz. Mater. 2009, vol. 163, no. 1, p. 207-

212. ISSN: 0304-3894.

[34] Gorems, A. Reduction of Total Dissolved Solids from Tannery Effluent by Optimum

Substitution of Common Salts with Sodium Sulphate for Hide/skin preservation. LAP

LAMBERT Academic Publishing 2012. 120 p. ISBN 978-3-8473-3387-6.

[35] Bailey D. G., Gosselin, J. The preservation of animal hides with potassium chloride. J.

Am. Leather Chem. As. 1996, vol. 91, p. 317-333. ISSN 0002-9726.

[36] Bailey, D. G. Gamma radiation preservation of cattle hides. A new twist on an old

story. J. Am. Leather Chem. As. 1999, vol. 94, no.7, p. 259-267. ISSN 0002-9726.

[37] Bailey, D. G., Dimaio, G. L., Gehring, A. G., Ross G. D. Electron beam irradiation

preservation of cattle hides in a commercial scale demonstration. J. Am. Leather Chem.

As. 2001, vol. 96, no. 10, p. 382-392. ISSN 0002-9726.

[38] Holicza, P. J. Method of preserving hides and skins. United States Patent US

2011/0135951 A1, 2011-06-09.

[39] Johnson, B., Anker, H., Meleney, F. Bacitracin: a new antibiotic produced by a

member of the B. subtilis group. Science 1945, vol. 102, p. 376-377. ISSN 0036-8075.

[40] Brosse, J., Sabatier, B. Use of superabsorbent polymers for treating raw skins,

corresponding compositions and methods and resulting treated skins. United States

Patent 6964745, 2005-11-15.

[41] Kanagaraj, J., Chandra Babu, N. K., Sadulla, S., Suseela RAJKUMAR, G., Visalakshi,

V., Chandrakumar, N. A new approach to less-salt preservation of raw skin/hide. J.

Am. Leather Chem. As. 2000, vol. 95, no. 11, p. 368-374. ISSN 0002-9726.

[42] Kanagaraj, J., Sastry, T. P., Rose, C. Effective preservation of raw goat skins for the

reduction of total dissolved solids. J. Clean. Prod. 2005, vol. 13, no. 9, p. 959-964.

ISSN: 0959-6526.

[43] Johnson, D. J., Mcaninch, T. L., Holzhauer, F. W. Hide-curing additive. United States

Patent 6451062, 2002-09-17.

[44] Kanagaraj, J., Sundar, V. J., Muralidharan, C., Sadulla, S. Alternatives to sodium

chloride in prevention of skin protein degradation - a case study. J. Clean. Prod. 2005,

vol. 13, no. 8, p. 825-831. ISSN: 0959-6526.

[45] Gautam, S. P. Recovery of better quality reusable salt from soak liquor of tanneries in

solar evaporation pans. Control of Urban Pollution Series: Cups/ 2009-10. India,

Central Pollution Control Board Ministry of Environment & Forests, January, 2009.



[46] International Union of Leather Technologists and Chemists Societies (IULTCS). IUE-

1: Recommendations on cleaner technologies for leather production, 2008.

www.iultcs.org/IUE1_2008.pdf.

[47] IDA Desalination Yearbook 2012-2013, ISBN 978-1-907467-19-6.

[48] Scholz, W. G., Rougé, P., Bódalo, A., Leitz, U. Desalination of Mixed Tannery

Effluent with Membrane Bioreactor and Reverse Osmosis Treatment. Environ. Sci.

Technol. 2005, vol. 39, p. 8505–8511.

[49] HO, W. Membrane Handbook. 2nd ed. New York: Chapman & Hall, 1992, 954 s.

ISBN 0412988712.

[50] UNIDO. Introduction to treatment of tannery effluents. URL <http://www.unido.org/

resources/publications/poverty-reduction-through-productive-activities/agribusiness-

and-rural-entrepreneurship/introduction-to-treatment-of-tannery-effluents.html?

ucg_no64=1%2Fdata%2Fproject%2FDATA1%2Fdata%2Ffileadmin%2Ftemplates%2

Ftypo3conf%2Fext%2Fdoc%2Findex.php>

[51] O‟Brien, D. J. A mathematical model for unsteady state salt diffusion from brine-cured

cattlehides. J. Am. Leather Chem. As. 1983, vol. 78, p. 286-299. ISSN 0002-9726.

[52] Kolomazník, K., Prokopová, Z., Vašek, V., Bailey, D. Development of a Control

Algorithm for the Optimized Soaking of Cured Hides. J. Am. Leather Chem. As. 2006,

vol. 101, no. 9, p. 309-316. ISSN 0002-9726.

[53] Kolomazník, K., Fürst, T., Uhlířová, M. Relationship between mass transport and the

quality of cured hide. Can. J. Chem. Eng. 2009, vol. 87, no. 1, p. 60-68. ISSN 0008-

4034.

[54] Levenspiel, O. Chemical Reaction Engineering, 3rd Ed., John Wiley & Sons, Inc., New

York, 1999. 578 pp. ISBN 978-0-4712-5424-9.

[55] Gravier, N., Pinotti, A., Califano, A., Zaritzky, N. Mathematical modeling of the

uptake of curing salts in pork meat. J. Food Eng. 2009, vol. 95, no. 4, p. 533-540. ISSN

0260-8774.

[56] Barat J. M., Baigts, D., Aliño, M., Fernández, F. J., Pérez-García, V. M. Kinetics

studies during NaCl and KCl pork meat brining. J. Food Eng. 2011, vol. 106, p. 102-

110. ISSN 0260-8774.

[57] Bertram, H. C., Holdsworth, S. J., Whittaker, A. K., Andersen, H. J. Salt diffusion and

distribution in meat studied by 23Na nuclear magnetic resonance imaging and

relaxometry. J. Agric. Food Chem. 2005, vol. 53, p. 7814-7818. ISSN 1520-5118.

[58] Turhan, M., Kaletunç, G. Modeling of salt diffusion in white cheese during long term

brining. J. Food Sci. 1992, vol. 57, no. 5, p. 1082-1085. ISSN 1750-3841.

[59] Floury, J., Jeanson, S., Aly, S., Lortal, S. Determination of the diffusion coefficients of

small solutes in cheese: A review. Dairy Sci. Technol. 2010, vol. 90, no. 5, p. 477-508.

ISSN 1958-5586.

[60] Crank, J. Mathematics of Diffusion, 2nd ed., USA, 1975, Oxford University Press, Inc.,

New York. 414 pp. ISBN 0-19-853411-6.

[61] Dehkordi, B. M., Hashemi, F., Mostafazadeh, R. Numerical solution of the equations of

salt diffusion into the potato tissues. International Journal of Chemical and Biological

Engineering 2010, vol. 3, no. 2, p. 78-81. ISSN 2010-3867.

[62] Bařinová, M., Kolomazník, K., Vašek, V., Matyašovský, J., Jurkovič, P. Optimization

of raw hide curing using two-component counter-current diffusion model. J. Am.

Leather Chem. As. 2009, vol. 104, no. 12, p. 397-404. ISSN 0002-9726.



[63] Kanagaraj, J., Velappan, K., Chandra Babu, N. K., Sadulla, S. Solid wastes generation

in the leather industry and its utilization for cleaner environment-A review. J. Sci. Ind.

Res. 2006, vol. 65, p. 541-548. ISSN 0975-1084.

[64] Sundar, V. J., Gnanamani, A., Muralidharan, C., Chandra Babu, N. K., Mandal, A. B.

Recovery and utilization of proteinous wastes of leather making: A review. Rev.

Environ. Sci. Biotechnol. 2011, vol. 10, p. 151–163. ISSN 1572-9826.

[65] Dalev, P. G., Simeonova, L. S. An enzyme biotechnology for the total utilization of

leather wastes. Biotechnol. Lett. 1992, vol. 14, p. 531-534.

[66] Kumaraguru, S., Sastry, T. P., Rose, C. Hydrolysis of tannery fleshings using

pancreatic enzymes: a biotechnological tool for solid waste management. J. Am.

Leather Chem. As. 1998, vol. 93, p. 32–39. ISSN 0002-9726.

[67] Jian, S., Wenyi, T., Wuyong, C. Ultrasound-accelerated enzymatic hydrolysis of solid

leather waste. J. Clean. Prod. 2008, vol. 16, p. 591-597. ISSN 0959-6526.

[68] Rai, A. K., Nived, C., Sakhare, P. Z., Suresh, P. V., Bhaskar, N., Mahendrakar, N. S.

Optimization of acid hydrolysis conditions of delimed tannery fleshings by response

surface method. J. Sci. Ind. Res. 2009, vol. 68, p. 967-974.

[69] Kumar, G. A., Swarnalatha, S., Sairam, B., Sekaran, G. Production of alkaline protease

by Pseudomonas aeruginosa using proteinaceous solid waste generated from leather

manufacturing industries. Bioresour. Technol. 2008, vol. 99, p. 1939–1944. ISSN

0960-8524.

[70] Kumar, G. A., Venkatesan, R., Rao, B. P., Swarnalatha, S., Sekaran, G. Utilization of

tannery solid waste for protease production by Synergistes sp. In solid-state

fermentation and partial protease characterization. Eng. Life Sci. 2009, vol. 9, p. 66–73.

ISSN 1618-2863.

[71] Aloy, M. Methane gas from tannery effluent. Leather 1987, vol. 189, p. 53–54. ISSN

0023-9739.

[72] Shanmugham, P., Horan, N. J. Optimising the biogas production from leather fleshing

waste by co-digestion with MSW. Bioresour. Technol. 2009, vol. 100, p. 4117–4120.

ISSN 0960-8524.

[73] Kolomazník, K., Pecha, J., Bařinová, M., Šánek, L. Economic aspects of biodiesel

production from tannery waste fats. J. Am. Leather Chem. As. 2010, vol. 105, p. 327-

333. ISSN 0002-9726.

[74] Pecha, J., Kolomazník, K., Bařinová, M., Šánek, L. High Quality Biodiesel and

Glycerin from Fleshings. J. Am. Leather Chem. As. 2012, vol. 107, no. 10, p. 312-322.

ISSN 0002-9726. [75] Chilton C. H., Perry R. H. (Eds.). Chemical Engineers Handbook, 5th ed., USA, 1973,

McGraw-Hill, Inc., New York. 1920 pp. ISBN 978-0-0704-9478-9.

[76] Reid, R. C., Prausnitz, J. M., Sherwood, T. K. The Properties of Gases and Liquids, 3rd

ed., USA, 1977, McGraw-Hill, New-York. 688 pp. ISBN 978-0-0705-1790-5.

[77] Janáčová, D. Nepřímé modelování námoku surových kůží. PhD. Thesis. Vysoké učení

technické v Brně, Zlín, 1993. 115 p. [78] Kolomazník, K., Pecha, J., Vašek, V., Friebrová, V., Podzimek, P. Způsob

deproteinizace odpadních tuků a olejů. Czech Patent, CZ303310B6. 2012-07-25. [in

Czech]




Chapter 5

DEFUZZIFICATION OF FUZZY CONCEPTS

TO SUPPORT VULNERABILITY ASSESSMENTS

OF CLIMATE CHANGE IMPACTS

IN THE PHILIPPINES

Lilibeth A. Acosta1,2*

and Jemimah Mae A. Eugenio3

1Potsdam Institute for Climate Impact Research (PIK),

Telegraphenberg, Potsdam, Germany 2School of Environmental Science and Management,

University of the Philippines in Los Banos (UPLB), Philippines 3Institute of Mathematical Sciences and Physics,

College of Arts and Sciences,

UPLB, Philippines

ABSTRACT

Climate change is one of the most pressing global issues that require innovative

methods to address the complexity of human-environment interactions. Many aspects of

vulnerability to climate change and adaptation measures to address its adverse impacts

remain vague and unquantifiable. Vulnerability assessments require methods that can

reduce the vagueness and imprecision of interpretations of data and information in human

and environment systems.

This chapter illustrates an empirical application of fuzzy logic analysis and the utility

of this analytical tool in integrated modeling assessments in the context of climate

change. Using the intervulnerability framework, fuzzy logic models can be used to assess

trade-offs of adaptive capacity and hotspots of vulnerable regions. In particular, we used

data on socio-economic and ecological indicators that are relevant for the assessments of

adaptive capacity and vulnerability in the different provinces and regions in the

Philippines.

* Corresponding author: Dr. Lilibeth A. Acosta, Potsdam Institute for Climate Impact Research (PIK),

Telegraphenberg A62, 14473 Potsdam, Germany. Tel.: +49 331 2882643; Fax: +49 331 2882695; E-mail:

[email protected].


Lilibeth A. Acosta and Jemimah Mae A. Eugenio 166

The empirical application in this country shows the advantages of fuzzy set theory in

terms of its (1) transparency, which allows explicit presentation of model assumptions

through inference rules; and (2) flexibility, which allows direct inclusion of informal and

expert knowledge in combining various indicator sets. The results of the fuzzy models

show that the types of indicators and determinants traded-off depend on the social and

economic conditions in the regions.

Vulnerable provinces are mostly located in northern and southern parts of the

Philippines. Vulnerability in the former can be reduced through improving water

availability for agriculture, whilst in the latter through improving peace and order

condition that affects socio-economic development.

1. INTRODUCTION

Climate change is one of the most pressing global issues that require innovative methods

to address the complexity of human-environment interactions. The Intergovernmental Panel

on Climate Change (IPCC), which was established in 1988 among others to provide scientific

support to climate change negotiations, is the leading international scientific body for the

assessment of climate change under the auspices of the United Nations (UN). The IPCC

(2007) defines climate change as: “A change in the state of the climate that can be identified

(e.g., by using statistical tests) by changes in the mean and/or the variability of its properties

and that persists for an extended period, typically decades or longer. It may be due to natural

internal processes or external forcings, or to persistent anthropogenic changes in the

composition of the atmosphere or in land use”. The IPCC differentiates climate change

problems in terms of its immediate (short-term) impacts like storms, drought, etc., and long-

term impacts like sea level rise. Following its Assessment Reports (First AR, 1990; Second

AR, 1995; Third AR, 2001; Fourth AR, 2007), approaches to solving climate change are

generally categorised into two – adaptation and mitigation. On the one hand, adaptation refers

to solutions for short-term impacts and, on the other hand, mitigation refers to measures for

reducing long-term impacts of climate change. Human and environmental systems that are

easily or highly exposed to the adverse effects of climate change impacts, whether in the

short-term or long-term, are considered to be vulnerable to climate change impacts.

Efforts to advance human dimensions in climate change research resulted in a shift from

impact assessments based on climate model scenarios to adaptation analyses based on an

understanding of vulnerability. The consequences of environmental change are now well

understood as a function of the ability of people and social systems to adapt to global

environmental change.

Thus, adaptation, like mitigation, became an important issue in addressing vulnerability

to global environmental change. For this reason, concepts and methods contributing to

increased knowledge on and understanding of vulnerability and adaptation have become

major foci of human dimensions research in the context of climate change. (Acosta and Galli,

2013) explained that new theoretical constructs and empirical methods that consider human

interaction with its environment are important for improving vulnerability assessments. In the

past decade, a number of interdisciplinary studies have captured the complexities of human-

environment links such as the Social Vulnerability (Adger, 1999), Double Exposure (O'Brien

and Leichenko, 2000), Vulnerability-Resilience (Moss et al., 2001), Risk-Chain (Heitzmann

et al., 2002), Vulnerability for Sustainability (Turner et al., 2003), Eight-Step Approach


Defuzzification of Fuzzy Concepts to Support Vulnerability … 167

(Polsky et al., 2003), Intervulnerability (Acosta-Michlik and Rounsevell, 2005; Acosta-

Michlik and Espaldon, 2008; Acosta-Michlik and Espaldon, 2011), and Crisis Probability

Curves (CPC) (Acosta-Michlik et al., 2006; Acosta and Galli, 2013).

Among these studies, Acosta-Michlik and her colleagues applied fuzzy logic as one of

the important methods for the application of Intervulnerability and CPC frameworks. This

chapter will however focus only on the latter framework because it allows the application of

fuzzy logic models on the assessments of two important aspects of vulnerability – trade-offs

pattern and hotspot regions.

The trade-offs pattern shows the sources of adaptive capacity for the society and hotspot

regions refer to the location of the most vulnerable people. The chapter aims to illustrate the

advantages of using fuzzy theory and methods for empirical research on adaptation and

vulnerability in view of its (1) transparency, which allows explicit presentation of model

assumptions through inference rules; and (2) flexibility, which allows direct inclusion of

informal and expert knowledge in combining various indicator sets. The chapter is structured

as follows: section 2 presents the concept of vulnerability and framework for

intervulnerability, section 3 discusses the fuzzy theory and methods that relates to their

applications in intervulnerability framework, section 4 checks the robustness of the fuzzy

logic models in terms of sensitivity of model outputs to changes in model inputs, section 5

presents the trade-offs pattern and hotspot regions generated from fuzzy logic models, and

section 6 provides conclusions.

2. CONCEPT AND FRAMEWORK

Vulnerability is not a new science, but it remains a much debated concept. As a result,

there are divergent definitions and frameworks for vulnerability available in literature.

Understanding vulnerability is very much influenced not only by the level, focus and methods

of analysis but also by the researchers‟ field of expertise. The most widely used definition is

that of IPCC, which defines vulnerability as a function of adaptive capacity, sensitivity and

exposure (IPCC, 2001).

Exposure is the nature and degree to which a system is exposed to significant

climatic variations.

Sensitivity is the degree to which a system is affected, either adversely or

beneficially, by climate-related stimuli.

Adaptive capacity is the ability of a system to adjust to climate change (including

climate variability and extremes) to moderate potential damages, to take advantage

of opportunities, or to cope with the consequences.

The element of exposure is an integral part of vulnerability concept because it answers

the question “vulnerable to what”, the “what” being external stressors in the environment.

The degree of vulnerability to some environmental stressors depends on the sensitivity of the

system, from which the society depends for its existence (i.e., social, economic and natural

resources).



Vulnerability is also dependent on the adaptive capacity of the society. Sensitivity is the

most poorly defined element in vulnerability concept because it can hardly be disassociated

from exposure or adaptive capacity1. In view of this, many studies have combined sensitivity

with exposure to characterise natural environment that is fragile to the impacts of climate

change (e.g., Metzger et al., 2006; Acosta-Michlik et al., 2008).

The intervulnerability framework builds on the IPCC‟s vulnerability definition and

considers exposure and sensitivity as combined components of vulnerability. The framework

emphasises the importance of considering the interaction of the impacts of global processes

and the interconnection of global to local changes in assessing vulnerability (Acosta-Michlik

and Rounsevell, 2005).

Specifically, it aims to assist in merging the relevant socio-economic and biophysical

attributes of an agent‟s environment, and thus in assessing the differential vulnerability of

local communities to interacting impacts of globalisation and climate change. The

intervulnerability framework follows stepwise approach to assess vulnerability – indicator-

based, profile-based and agent-based (Figure 1).

This chapter focuses on the indicator-based approach, which applies fuzzy logic analysis

to understand the structure of adaptive capacity trade-offs and generate maps of hotspot

regions. In intervulnerability assessment, maps are used to guide the selection of case study

areas for the profile-based approach and the trade-offs structure can assist the selection of

adaptation options for the agent-based approach. The profile-based approach applies cluster

analysis to generate vulnerability profiles. The profiles are required to build the behavioural

model for the agent-based approach. Because agent-based models are processed-based, they

are useful for dynamic analysis of future vulnerabilities.

Estimating dynamic vulnerability is hardly possible using indicator-based and profile-

based approaches as they only aggregate and stratify past and present information. Studies

that have projected vulnerability using indicator-based approach are based on comparative-

static, and not dynamic analysis.

Although it can only provide static view, the indicator-based approach remains very

useful in vulnerability assessments for the following reasons: (1) provides a synthesis of

complex state-of-affairs such as the vulnerability of regions, households or countries into a

single number that can then be easily used by policy (Hinkel, 2011); (2) provides scoping or

“first look” assessment to support identification of priority vulnerable coastal areas (Ramieri

et al., 2011), and (c) provides generalised knowledge across multiple spatial scales even

where data sources are limited (Acosta-Michlik and Rounsevell, 2009).

1 This statement is based on the discussions during the workshop of the START Advanced Institute on Vulnerability

to Global Environmental Change held in IIASA, Laxenburg, Austria on May 3-21, 2004.


Figure 1. Stepwise methods of the intervulnerability framework.

B a ta n g a sB a ta n g a s

Secondary data

• socio-economic indicators

• biophysical indicators

• GIS maps of land features

Primary data

• household socio-economic

information (social survey)

• farm boundaries (GPS)

Data Collection

Fuzzy logic analysis

• social capacity

• economic capacity

• institutional capacity

• soil quality

• water availability

Cluster analysis

• housedhold income & assets

• farming & non-farming jobs

• farm size and crops planted

• age and educational level

• source of information, etc.

Vulnerability maps

(2002)

Vulnerability profiles

(2005)

Vulnerability futures

(2005 - 2030)

Data Inputs

Methods

Model OutputsProportion of village residence per cluster

0%

20%

40%

60%

80%

100%

traditional diversified subsistence commercial

GonzalesNatatasCalle

Agent-based model

Case study areas

• market structure

• farming system

• social network

Agents„ profiles

• cognitive strategies

• adaptive decisions

• adaptation scenarios

Select vulnerable region

Build behavioural model

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

2020

2021

2022

2023

2024

2025

no policy irrigation market both policy

very low

moderate

low

high

very highB a ta n g a sB a ta n g a s

Secondary data




Secondary data




Primary data




Primary data




Data Collection


• social capacity



• soil quality



• social capacity



• soil quality


Cluster analysis






Cluster analysis






Vulnerability maps

(2002)

Vulnerability maps

(2002)


(2005)


(2005)


(2005 - 2030)


(2005 - 2030)

Data Inputs

Methods

Model OutputsProportion of village residence per cluster

0%

20%

40%

60%

80%

100%

traditional diversified subsistence commercial

GonzalesNatatasCalle

Agent-based model

Case study areas

• market structure

• farming system

• social network

Agents„ profiles

• cognitive strategies

• adaptive decisions

• adaptation scenarios

Select vulnerable region

Build behavioural model

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

2020

2021

2022

2023

2024

2025

no policy irrigation market both policy

very low

moderate

low

high

very high



3. METHODS

3.1. Indicator-Based Approach

Indicator-based approaches to vulnerability assessments in climate change studies can be

traced back to earlier work on poverty mapping and food insecurity (Downing, 2001;

Ramachandran and Eastman, 1997; Henninger, 1998). Under this approach, the exposure

units are usually geographical areas, the vulnerabilities of which are captured by the

differences in their social, economic, institutional and environmental structure at a given point

in time. Composite vulnerability index, which is derived from aggregating socio-economic

and ecological indicators, is useful for identifying vulnerable countries, regions or

communities (Figure 2). Vulnerability indices are often presented in maps, which are a

convenient way of presenting spatial variation in vulnerability (e.g., O‟Brien et al., 2004;

Metzger et al., 2006; Liu et al., 2008; Fraser et al., 2013). The selection of indicators is

influenced by the availability of data and understanding of vulnerability concept. In this

chapter, relevant indicators of adaptive capacity were selected using theory-based (deductive)

approach. Quantitative assessment of a society‟s capacity to adapt to global environmental

changes has been largely based on the use of proxy variables for economic wealth, social

capital, and institutional infrastructure (e.g., Acosta-Michlik et al., 2008; Adger et al., 2004;

O‟Brien et al., 2004; Acosta et al., 2013; etc.). The indicators used for the vulnerability

assessment are presented in Figure 2 and the data used to represent these indicators are

presented in Appendix 1. The relevance of these indicators for adaptive capacity and

exposure/sensitivity are discussed below.

Groundwater supply

indicators

Rainwater supply

indicators

Soil erosion

indicators

Landuse change

indicators

Water

availability

Soil

quality

Sensitivity/

Exposure

Income

indicators

Infrastructure

indicators

Good governance

indicators

Human security

indicators

Economic

capacity

Institutional

capacity

Education

indicators

Health

indicators

Social

capacity

Adaptive

Capacity

Vulnerability

index

Groundwater supply

indicators

Rainwater supply

indicators

Soil erosion

indicators

Landuse change

indicators

Water

availability

Soil

quality

Sensitivity/

Exposure

Groundwater supply

indicators

Rainwater supply

indicators

Soil erosion

indicators

Landuse change

indicators

Water

availability

Soil

quality

Groundwater supply

indicators

Groundwater supply

indicators

Rainwater supply

indicators

Rainwater supply

indicators

Soil erosion

indicators

Soil erosion

indicators

Landuse change

indicators

Landuse change

indicators

Water

availability

Soil

quality

Water

availability

Water

availability

Soil

quality

Soil

quality

Sensitivity/

Exposure

Sensitivity/

Exposure

Income

indicators

Infrastructure

indicators

Good governance

indicators

Human security

indicators

Economic

capacity

Institutional

capacity

Education

indicators

Health

indicators

Social

capacity

Adaptive

Capacity

Income

indicators

Infrastructure

indicators

Good governance

indicators

Human security

indicators

Economic

capacity

Institutional

capacity

Income

indicators

Income

indicators

Infrastructure

indicators

Infrastructure

indicators

Good governance

indicators

Good governance

indicators

Human security

indicators

Human security

indicators

Economic

capacity

Institutional

capacity

Economic

capacity

Economic

capacity

Institutional

capacity

Institutional

capacity

Education

indicators

Education

indicators

Health

indicators

Health

indicators

Social

capacity

Social

capacity

Adaptive

Capacity

Adaptive

Capacity

Vulnerability

index

Vulnerability

index

Figure 2. Indicator-based framework for vulnerability assessment.



Income per capita, measured by the ratio of GDP to total population, is the most widely

used indicator for economic well-being. However, GDP per capita is a crude measurement of

the population‟s personal wealth. This is because it does not disclose how much of the total

GDP accrue to what percentage of the population. The empirical relationship between income

per capita and the sense of basic need satisfaction can be extremely weak particularly in low-

income countries (Fuentes and Rojas, 2001; Shen and Williamson, 2001). In this chapter,

alternative indicators of economic capacity to adapt were thus used (Figure 2). Economic

development theorists argue that skewed distribution of income limits access to sufficient

food, although there are adequate food supplies at the national level, and restricts access to

basic services such as health and education. The shape of the distribution rather than the total

size of the national income help to account for observed variations in common indicators of

population health (Judge et al., 1997). Distribution of income and extent of poverty are thus

relevant indicators of economic capacity. In addition, indicators that directly reveal the

income level and purchasing power of a household are useful measure of economic capacity

to adapt to the impacts of climate change.

Infrastructure that supports effective transport system and clean water/energy supply is

important to increase adaptive capacity. Quality of roads and availability of ports are crucial

for creating cost-effective links between supplier and consumer agents in different

geographical scales. In the Philippines where there are rich water resources, hydropower,

which accounts for about 6 percent of the total energy produced domestically in 2009 (DOE,

2010), is one of the important sources of not only cheap but also clean energy. The river dams

are used not only for energy generation, but also to provide water for irrigation systems that

are vital for sustaining agricultural productivity. More recently, climate mitigation efforts to

reduce greenhouse gas emissions have promoted the use of renewable sources of energy like

hydropower. Access to clean water has long been considered as a key indicator of human

development because it improves health and standard of living.

Indicators for governance capacity were selected based on the work of Taenzler et al.

(2008). They used indicators such as tax revenue, social expenditure, and financial resources

to measure the government‟s ability to respond to crisis. According to Tänzler et al.

(2008:165), “the general ability of states to collect taxes is a key function in order to pursue

specific policy goals. … [T]he availability of financial resources is crucial to introduce

certain policy programs, administrative structures and more specifically early warning

systems in preparation for the occurrence of extreme weather events”. However, financial

availability should be matched with state willingness to provide the services needed by the

population. The government‟s commitment to protect the people can be measured by the

proportion of fiscal resources spent for social services. Furthermore, indicators relating to

human security are also important measure of institutional capacity because they are

manifestations of weak governance.

Health and education are core to improving the social well-being of the people. They

provide the foundation for generating social capital. Education is essential for improving the

capacity of the people, for achieving awareness and skills, and for effective participation in

decision-making. “Regular education, provided in state-governed institutional frameworks,

is, although not the sole provider of knowledge and competence, a key element in strategies to

ensure a broad participation of people in processes of socio-cultural, economic and

technological innovation” (Kaiser et al., 2000:7). Through social services, the government

helps the people to acquire knowledge, create awareness and be innovative. These are all



indispensable to increase the ability to adapt to environmental stress. Similarly, more money

spent on health services will help to improve the physical condition of the population and to

increase their capacity to cope with health risks associated with environmental stress.

However, a number of studies point out that public subsidies on health have favoured the

better-off and failed to reach the deprived (Filmer et al., 1997; Filmer and Pritchett, 1999;

Castro-Leal et al., 2000; Gwatkin, 2003; Wang, 2003). For this reason, it is useful to consider

indicators that could assess the outreach of health and education services to the local people.

Impacts of climate change are affecting water availability and quality, whilst the impacts

of globalisation are influencing the land availability and land use. A significant part of the

global land area is already used for agriculture and most of the land use conversions between

1970 and 2000 were in favour of agriculture and related activities. According to Gomiero et

al. (2009), agriculture activities including improved pasture and co-adapted grassland account

for about 40% of land surface and nearly 85% of annual global water withdrawals. Thus, in

addition to extreme weather events (e.g., droughts and floods) due to changes in climate,

changes in land use are increasingly becoming a source of environmental exposure or causing

ecological sensitivity.

3.2. Fuzzy Logic Model

The analytical tool applied to derive composite indices of vulnerability based on the

above-mentioned indicator-based framework is fuzzy logic. Whilst fuzzy logic is now

increasingly used in other fields of agriculture and environment (e.g., Roberts, 1996; Wu et

al., 1996; Mays et al., 1997; Silvert, 2000; Mackay and Robinson, 2000; Kangas and Kangas,

2002), its application in vulnerability to climate change impacts is very limited. The first

fuzzy logic models for vulnerability have been used to assess susceptibility or lack of

adaptive capacity in selected regions in Portugal, Russia and India (Acosta-Michlik et al.,

2008; Krömker et al., 2008). These studies were extended to use the outputs from fuzzy

models to develop thresholds for vulnerability (Acosta-Michlik et al., 2006; Acosta & Galli,

2013). Another important application of fuzzy logic models is the development of scenarios

for adaptive capacity (Acosta et al., 2013), which are useful for integrated assessment

modelling of climate change impacts (Metzger et al., 2006). In these studies, fuzzy logic

models were only applied to develop adaptive capacity indices. But by using

intervulnerability framework in this chapter, the indices for both components of vulnerability

– adaptive capacity and exposure/sensitivity, were generated from fuzzy logic models.

The fuzzy logic models are based on multi-valued logic producing indices between 0 and

1. Classical set theory that produces only two-logic values (i.e., either 0 or 1) will be able to

inform whether vulnerability exists or not. But the multi-logic approach of fuzzy set theory

can provide information on the degree of vulnerability, which is useful for identifying hotspot

regions through comparison of indices of adaptive capacity and exposure/sensitivity across

regions. Moreover, fuzzy logic models allow the use of linguistic values such as low,

moderate and high to certain index value ranges. Linguistic values are useful for fuzzy

concepts to quantify the vagueness and imprecision of interpretations (Mays et al., 1997;

Darbra et al., 2008). Such linguistic values have practical use for evaluating vulnerability,

which does not have an objective yardstick to assess its relative magnitude (Acosta-Michlik et

al., 2008). In this case, vulnerability assessments require transparency of model assumptions



and flexibility of model structure. On the one hand, assumptions are made transparent

because indicators can be combined to represent specific concepts (i.e., as in Figure 2) and

these concepts can be presented in inference rules (see section 3.1). On the other hand,

structure is made flexible because fuzzy logic allows inclusion of several indicators that

represent specific concept and these indicators can be included in different sets of fuzzy

models (i.e., different aggregation levels) to simplify the model. Vulnerability assessment

tools that are transparent and flexible translate vague and diverse information into a common

and comprehensible language, which is important in highly interdisciplinary fields like

climate change.

A fuzzy logic model has three components - input variables, inference rules and model

output. Figure 3 summarizes how these components are used in fuzzy logic analysis. The

analysis in fuzzy logic models consists of three steps - fuzzification, fuzzy inference and

defuzzification. Fuzzification translates data into linguistic values and defines the degree of a

membership μA. A membership function is a curve that defines how each point in the input

space is mapped into a membership value (or degree of membership) between 0 and 1. The

fuzzy inference involves an implication and aggregation process (Cornelissen et al., 2001).

The implication process evaluates inference rules, which are “if-then” statements describing

the possible relationships between the input variables. For example, if water supply is low and

soil quality is low, then sensitivity/exposure is high. The aggregation process combines the

fuzzy conclusions (i.e., the area under the truncated membership functions) in each inference

rule. Finally, defuzzification converts the aggregated fuzzy conclusions into numerical

assessment, giving values between 0 and 1. The numerical assessment represents the index

value of the output variable. Details of these steps are available elsewhere (Eierdanz et al.,

2008). The fuzzy logic module of the MatLab software (The MathWorks, Inc., 2001) was

used in implementing the fuzzy logic models.

Figure 3. Components of fuzzy logic model and steps in fuzzy logic analysis.



In generating adaptive capacity, exposure/sensitivity and vulnerability indices, five levels

of aggregation (i.e., data, indicators, determinants, components, and vulnerability index) were

used to increase the transparency and decrease the complexity of vulnerability assessment

(Table 1). Each aggregation level represents different fuzzy logic models. In the first level of

aggregation, the input variables were the socio-economic and biophysical data, or proxy

variables, and the model outputs were the indices of the indicators. For example, data on

poverty incidence of families, food subsistence incidence of families and Gini coefficient

were used as inputs to the fuzzy logic model, which was applied to generate index for poverty

and inequality. At this level of aggregation, there were 14 fuzzy logic models corresponding

to 10 socio-economic indicators and 4 biophysical indicators. In the second level of

aggregation, the input variables were the model outputs from the first level of aggregation.

The model outputs from the second level of aggregation were the indices of the determinants

and/or components. There were 4 fuzzy logic models at this level of aggregation

corresponding to income level, infrastructure facilities, education level and services, and

health condition and services. For example, poverty and financial indices were used as inputs

to fuzzy logic model, which was applied to generate index for income level. There were no

corresponding determinants for indicators of governance and security, water supply, soil

erosion and land degradation because they were directly used as inputs to the third level of

aggregation.

In the third level of aggregation, the input variables were the model outputs from the

second level of aggregation, i.e., indices of the determinants. The model outputs were the

indices of vulnerability components. At this level of aggregation, there were 5 fuzzy logic

models corresponding to economic, institutional and social capacity as well as water

availability and soil quality. The fourth level of aggregation aimed to generate the indices of

adaptive capacity and exposure/sensitivity. There were thus 2 fuzzy logic models at this level

of aggregation. The input variables for the adaptive capacity model were economic,

institutional and social capacity and those for the exposure/sensitivity were water availability

and soil quality. Finally, the fifth level of aggregation consisted of one fuzzy logic model. It

combined the indices of adaptive capacity and exposure/sensitivity, which were the model

outputs from the previous aggregation level, to generate index of vulnerability, the output

from the last level of aggregation. At this point, there were 26 fuzzy models applied to

generate vulnerability index (Table 1). The models were applied in 81 provinces of the

Philippines for the year 2000, for which there was a complete dataset for all provinces.

Altogether, there were thus 2,106 fuzzy logic models to produce vulnerability indices for the

81 provinces.

3.3. The Case Study Area

The Philippines is an archipelago consisting of more than 7000 islands with a total land

area of 298,170 square kilometers (BAS, 2013). About 32 percent of the area is agricultural

land. The agricultural land is mainly planted to arable (51 percent) and permanent (44

percent) crops. Agriculture contributed about 11 percent of the 10,565 billion GDP in 2012.

Crops contribute to about 50 percent of the agricultural GDP, in particular rice, corn, banana

and coconut. The main agricultural exports are coconut oil (20 percent), banana fresh (13

percent) and pineapple and products (8 percent). Agricultural activities and services account



for only 7 percent of the agricultural GDP. The share of agriculture to total employment is

continuously declining, reaching only 32 percent in 2012. The Philippines has three major

islands – Luzon, Visayas and Mindanao (Figure 4). It has a total 15 administrative regions, 6

in Luzon, 3 in Visayas and 6 in Mindanao.

RegionsIslandsLuzon

Visayas

Mindanao

RegionsIslandsLuzon

Visayas

Mindanao

Figure 4. Different regions in Luzon, Visayas and Mindanao islands in the Philippines.

In terms of land area, region 4 in Southern Luzon, which consists of Calabarzon and

Mimaropa, is the biggest with a total of 4.7 million hectares (Table 2). It has also the largest

number of population with more than 15 million people. This is followed by the National

Capital region (NCR) that covers the area of Metro Manila and has a population of 12 million

in 2010 (BAS 2013). The top agricultural export product (i.e., coconut) is widely cultivated in

region 4 (Calabarzon), region 5 (Bicol), region 8 (Eastern Visayas), region 11 (Davao) and

region 13 (Caraga). Prices of raw coconut products are however one of the lowest among crop

commodities in the Philippines so that daily wage is also low. The highest daily farm wage is

found in Central Luzon which is one of the regions with largest rice production. Rice is the

main staple food in the Philippines. The lowest daily farm wage is found in Central Visayas at

less than PhP100. Many regions in Mindanao have the highest poverty incidence of families –

39.8 percent in region 13, 38.1 percent in region 15 and 36.6 in region 9. Poverty in these

regions is further exacerbated by local insurgence and civil conflicts. In region 15 or

Autonomous Region in Muslim Mindanao (ARMM), almost 70 percent of the population still

relies on agriculture for their living, more than double of the average employment figures in

the Philippines.

The Philippine climate is tropical and maritime, characterised by relatively high

temperature, high humidity and abundant rainfall (PAGASA, 2013). The average temperature

is 26.6 degrees centigrade, with the lowest mean temperature of 25.5 in January and highest

mean temperature of 28.3 in May. There is essentially no difference in the mean annual

temperature in the different regions in Luzon, Visayas and Mindanao. The high relative

humidity in the Philippines is due to high temperature and the surrounding bodies of water.

The average monthly relative humidity varies between 71 percent in March and 85 percent in



September. Rainfall is the most variable climatic element in the Philippines with different

rainfall distribution in different regions depending on wind direction and mountain locations.

The mean annual rainfall varies from 965 to 4,064 millimeters. Philippine climate is

categorised into four types based on the distribution of rainfall (Table 2): Type I has two

pronounced season - dry from November to April and wet during the rest of the year; Type II

has no dry season with a pronounced rainfall from November to January; Type III has no

pronounced seasons - relatively dry from November to April, and wet during the rest of the

year; and Type IV has more or less evenly rainfall distribution throughout the year. Type I is

typical climate in regions 1 and 6, type II in regions 7, 8, 10, 11 and 13, type III in region 2

and Type IV in region 12. All other regions have combined climate types.

4. MODEL VERIFICATION

Prior to the analysis of the results of the fuzzy logic models, we first verified the

sensitivity of the outputs to the input variables. A sensitivity analysis can help to gauge the

robustness of the indicators (Phillis and Andriantiatsaholiniaina, 2001; Baraldi et al., 2009;

Singh et al., 2012) and fuzzy logic models. It thus informs about the correctness of not only

the model structure, particularly in terms of constructing the membership functions, but also

the model inputs. Errors may result from the computation of standardized values and/or

preparation of database. It was necessary to do standardization1 because of large variability of

data range not only among indicators but also across regions. Because standardisation makes

data dimensionless (Gan et al., 2007), it is useful in defining standard indices for the different

indicators. Huge number and range of model inputs can cause uncertainties concerning the

data quality because information may be incomplete, imprecise, contradictory, not fully

reliable or vague (Janssen et al., 2010; Singh et al., 2012). Errors could have thus occurred in

up-/downloading and encoding data both by the data providers and users. The relevance and

accuracy of the data can be checked through sensitivity analysis. Information collected from

the analysis allow experts to see if model outputs are decreasing, increasing or stationary in

correspondence of a progressive increment or decrement of the input variables, and to verify

if such behaviour is in agreement with the experts‟ expectations (Baraldi et al., 2009). We

checked for the sensitivity of the input variables to the fuzzy logic models by changing the

raw data for the indicators to generate new datasets with values corresponding to ±5%, ±10%,

±25%, and ±50%. Thus, in the addition to the reference model (i.e., actual data for 2000), the

26 fuzzy logic models were run eight times for 81 provinces using the new datasets.

The results of the sensitivity analysis for the indices of adaptive capacity,

exposure/sensitivity and vulnerability are presented in Figure 5. The adaptive capacity indices

show increasing trend as values of input variables are increased and decreasing trend as they

are reduced from 5 to 50 percent.

1 The following formula was used to standardise the data:

minmax

min1 0,

XX

XX ii

where Xi refers to the original value of data point i, Xmin is the minimum or lowest value in the dataset X, Xmax

is the maximum or highest value in the same dataset and Xi represents the standardised data for each data

point i with values between zero and one.


Table 1. Fuzzy logic models for five levels of aggregation

Data

1st level

Indicators

2nd level

Determinants

3rd level

Components

4th level

Vulnerability26

5th level

Family Poverty Incidence Poverty &

inequality1

Income index15

Economic

Capacity19

ADAPTIVE

CAPACITY24

Food Subsistence Incidence

Gini Coefficient

Average Annual Family Income Income &

purchasing power2

Average Annual Family Savings

Inflation Rate

Length Of National Roads

Road infrastructure3

Infrastructure

index16

Length Of Poor Quality Roads

Total Number Of Ports

Large Scale Dams Water/Energy

infrastructure4

Hydropower Schemes

% Population with Water Service

Literacy Rate

Education situation5

Education

index17

Social Capacity20

Higher Education Enrolment

Secondary Cohort Survival Rate

Number of Public Schools Education services

6

Public Teacher-Student Ratio

Number of Water-Born Diseases

Health condition7 Health index

18 Access To Sanitary Toilet

Access To Safe Water Supply

Available Village Health Stations Health services

8

Active Barangay Health Workers

Total Financial Resources perCapita

Good governance index9 Institutional

Capacity21

Non-Tax Revenue

% Social Services To Expenditures

Unemployment Rate Human security index10


Table 1. (Continued)

Data

1st level

Indicators

2nd level

Determinants

3rd level

Components

4th level

Vulnerability26

5th level

Average Monthly Crime Rate

Crime Solution Efficiency

Amount Water Abstracted By Source Groundwater supply index

11

Water

availability22

EXPOSURE/

SENSITIVITY25

Physical Accounts Of Groundwater

Average Rainfall Rainwater supply index

12

Rainy Days Per Year

Lands Suffering From Erosion Soil erosion index

13

Soil quality23

Nutrient Loss From On-Site Erosion

Forest Destruction Land degradation index

14

Land Use Conversion By Region

Note: Numbers in superscript refer to the number of fuzzy models.

Table 2. Different regions in the Philippines

Region Number

and Name

Land

Area1

Population Daily Farm-

Wage2

% Poverty

Incidence

Climate

Types

% Agric.

employ

Main Crops

1 Ilocos Region 1.3 4,748,372 161.20 17.8 I 38.5% Tobacco,

Mango

2 Cagayan

Valley 2.8 3,229,163

140.76 14.50 III 57.9% Corn

3 Central Luzon 2.2 10,137737 172.87

12.0 I, II, III 21.6% Rice

4a Calabarzon 1.7 12,609,803 160.23 10.30 I, II, III, IV 15.9% Sugarcane,

Coconut

4b Mimaropa 3.0 2,744,671 138.23 27.60 I, II 48.6% Calamansi

5 Bicol Region 1.8 5,420,411

111.69 36 II, IV 40.7% Rice,

Coconut


Region Number

and Name

Land

Area1

Population Daily Farm-

Wage2

% Poverty

Incidence

Climate

Types

% Agric.

employ

Main Crops

6 Western

Visayas 2.1 7,102,438 118.23 23.80 I

39.7% Sugarcane,

Rice

7 Central

Visayas 1.6 6,800,180

99.52 30.2 II 31.2% Sugarcane

8 Eastern

Visayas 2.3 4,101,322

105.79

33.2 II 44.7% Coconut

9 Zamboanga

Peninsula 1.7 3,407,353 105.09 36.6

III (West)

IV (East) 49% Rubber

10 Northern

Mindanao 2.1 4,297,323

109.22 32.80 II

43.1% Pineapple,

Banana

11 Davao Region 2.0 4,468,563 102.00

25.60 II

40.9% Banana,

Coconut

12 Central

Mindanao 2.2 4,109,571

115.50 28.1

IV 49.5% Pineapple,

Banana, Corn

13 Caraga 2.1 2,429,224 126.12 39.80

II 38.6% Coconut,

Banana

14 CAR 2.0 1,616,867 130.17

17.1

II, III

50.1% Cabbage

15 ARMM 3.4 3,256,140 102.28 38.10

III, IV

69.8% Cassava

Notes: 1In Million hectares, 2In Philippine Pesos (PhP).

Source: BAS 2013.



Economic, institutional and social capacities have positive relationship with adaptive

capacity so both model inputs and outputs follow the same direction. At ±50% level of

change the adaptive capacity indices increase and decrease by ±12 percent. The

exposure/sensitivity indices show a decreasing trend as the input variables are increased, and

increasing trend as they are reduced from 5 to 50 percent. This is conceptually correct

because improving soil quality and increasing water availability should decrease

exposure/sensitivity. The vulnerability indices increase by 11 percent at +50% and by 7

percent at -50% level of change. Figure 4 show that vulnerability has positive relationship to

exposure/sensitivity and negative relationship to adaptive capacity, which are both

conceptually correct. The structure of the fuzzy logic models is thus robust and the model

inputs have no significant errors.

5. RESULTS AND DISCUSSION

5.1. Adaptive Capacity Trade-Offs

The results of fuzzy logic models for the determinants of adaptive capacity are presented

in web diagrams (Figure 6) and they correspond to the second level of aggregation (Table 1).

The larger the area of web diagram, the higher is the adaptive capacity with respect to these

socio-economic indicators. The selected provinces in Luzon and Visayas islands have

relatively higher level of adaptive capacity than those in Mindanao Island. Income indices are

lowest and education indices are highest in Luzon, particularly in the Batangas province.

Moreover, this province has the highest infrastructure index. Education and infrastructure are

thus traded-off against other determinants of adaptive capacity. Batangas belongs to region 4

in Calabarzon. Batangas has a very high level of economic capacity with an index of 0.65.

This can be attributed to the proximity and accessibility of the province to Manila, the

country‟s center for foreign capital and investment. Many municipalities in Batangas province

are experiencing rapid rate of urbanization growth. Moreover, Batangas has an international

port that is suitable for industrial and commercial activities. The port is currently used for

shipment of foreign traded goods. Because the province is closely linked to international

market, it is very open to future opportunities and/or risks related to globalisation. Due to

these economic opportunities, which extend to the other provinces, poverty incidence is

lowest in the region (Table 2). Cagayan, a province in region 2 (Cagayan Valley) has the

lowest adaptive capacity among the selected provinces in Luzon (Figure 6). Unlike

Calabarzon, Cagayan and the rest of the region remain very much dependent on agriculture

with the sector providing about 58 percent of employment (Table 2). Health is however given

priority in this province as shown by its high index (Figure 2).

Some provinces in Visayas have higher adaptive capacity than those in Luzon (Figure 6).

Unlike in Luzon, there is no obvious trade-offs pattern among the different determinants. For

example, two provinces have high income indices and the other two have low income indices.

The sources of adaptive capacity are thus quite variable in the different provinces in Visayas.

Iloilo province in region 6 has the highest adaptive capacity in the Visayas Island. Except for

infrastructure, all adaptive capacity indices in Iloilo are higher than those in Cebu (region 7),

the capital of Visayas.


Defuzzification of Fuzy Concepts to Support Vulnerability … 181

Adaptive Capacity

0,25

0,30

0,35

0,40

0,45

0,50

0,55

0,60

-50% -25% -10% -5% Reference 5% 10% 25% 50%

Percentage Level of Sensitivity Analysis

Me

an

Va

lue

s o

f In

dic

es

Adaptive capacity

economic capacity

institutional capacity

social capacity

Exposure/Sensitivity

0,25

0,30

0,35

0,40

0,45

0,50

0,55

0,60

0,65

-50% -25% -10% -5% Reference 5% 10% 25% 50%


Me

an

Va

lue

s o

f In

dic

es

Exposure

soil quality

water availability

Vulnerability

0,25

0,30

0,35

0,40

0,45

0,50

0,55

0,60

0,65

-50% -25% -10% -5% Reference 5% 10% 25% 50%


Mea

n V

alue

s of

Indi

ces

Vulnerability

Exposure/Sensitivity

Adaptive capacity

Figure 5. Results of the sensitivity analysis using ±5%, ±10%, ±25%, and ±50% changes from the

reference values.


Luzon Island

0,00

0,20

0,40

0,60

0,80

1,00Income

Infrastructure

EducationHealth

Governance

Ilocos Norte CagayanQuezon Batangas

Visayas Island

0,00

0,20

0,40

0,60

0,80

1,00Income

Infrastructure

EducationHealth

Governance

Siquijor LeyteIloilo Cebu

Mindanao Island

0,00

0,20

0,40

0,60

0,80

1,00Income

Infrastructure

EducationHealth

Governance

Bukidnon SuluCompostela Valley Davao del Norte

Luzon Island

0,00

0,20

0,40

0,60

0,80

1,00Income

Infrastructure

EducationHealth

Governance

Ilocos Norte CagayanQuezon Batangas

Visayas Island

0,00

0,20

0,40

0,60

0,80

1,00Income

Infrastructure

EducationHealth

Governance

Siquijor LeyteIloilo Cebu

Mindanao Island

0,00

0,20

0,40

0,60

0,80

1,00Income

Infrastructure

EducationHealth

Governance

Bukidnon SuluCompostela Valley Davao del Norte

Figure 6. Trade-offs in determinants of adaptive capacity in selected provinces, by island, 2000.



Region 6 continued to be one of the top five fastest growing economies in the Philippines

attributed to a remarkable growth in agriculture sector at 14 percent and services sector at 4.3

percent between 2010 and 2011 (NEDA Region VI, 2013). Iloilo has an international port that

supports economic growth. Moreover, economic growth is supported by economic activities

and job-creating investments, which were generated by the business process outsourcing

companies, hotels and restaurant businesses, the region‟s top employment drivers in the last

three years. In Mindanao Island, the trade-offs among the different determinants of adaptive

capacity are more apparent. The province of Bukidnon in region 10 has highest index for

health but one of the lowest for governance and education. Davao del Norte in region 11 has

the highest index for education which can be ascribed to its proximity to Davao City, the

capital of Mindanao, where most of the big private and public universities are located. Health

is however traded-off against other determinants of adaptive capacity in Davao del Norte. The

province of Sulu has the lowest level of adaptive capacity, with very low indices particularly

for health. Sulu is a province in region 15 or Autonomous Region in Muslim Mindanao

(ARMM) which has the most severe social unrest due to ethnic/religious issues.

Figure 7 summarises the results of fuzzy models for the socio-economic and

environmental components of vulnerability (i.e., 4th

level of aggregation). Social capacity is

highest in Luzon, revealing a general pattern of trading-off education and health against other

determinants of adaptive capacity in many regions in this island. Although social capacity is

also highest in Visayas, the level of trade-offs against the determinants of economic and

institutional capacity is not very high. In Mindanao, determinants of institutional capacity or

governance are traded-off against those of economic and social capacity. This is not

surprising because Mindanao remains to be affected by various political problems that require

effective governance. Improving institutional capacity is thus an important precondition to

improving the overall adaptive capacity in many regions in Mindanao. Another challenge in

Mindanao is poor soil quality, which affects economic productivity in many regions in this

island. Unlike in Mindanao, spatial differences in environmental problems related to water

availability and soil quality are not very significant in Luzon and Visayas.

0,00

0,10

0,20

0,30

0,40

0,50

0,60

economiccapacity

institutionalcapacity

social capacity wateravailability

soil quality

Luzon

Visayas

Mindanao

Figure 7. Indices of vulnerability components for three major islands in the Philippines, 2000.



5.2. Vulnerability Hotspots

Figure 8 presents maps of adaptive capacity, exposure/sensitivity and vulnerability in the

Philippines. The first two maps were generated from 5th level and the latter from 6

th (or last)

level of fuzzy aggregation. The geographical scale of aggregation is provincial level. In

Luzon, the provinces with high level of adaptive capacity are located in region 1 (i.e., Ilocos

Sur), region 4 (i.e., Cavite, Rizal, Romblon, Oriental Mindoro and Marinduque), region 5

(i.e., Catanduanes and Sorsogon) and region 14 (i.e., Abra and Kalinga). In Visayas, the

provinces with high level of adaptive capacity are located in region 6 (i.e., Aklan, Capiz and

Negros Occidental) and region 8 (i.e., Western Samar). Only the provinces of Zamboanga del

Norte in region 9 and Camiguin in region 10 have high level of adaptive capacity in

Mindanao. No province has very low adaptive capacity in Visayas Island. In Luzon Island

four provinces have very low adaptive capacity including Nueva Viscaya, Pampanga,

Palawan and Masbate. The most number of provinces with very low adaptive capacity is

located in Mindanao Island including Lanao del Sur, Sultan Kudarat, Davao del Sur, Basilan,

Tawi-Tawi and Sulu. Except for the last three provinces which are all located in the same

region, the other provinces belong to different regions. There is thus no pertinent spatial

pattern in adaptive capacity in the Philippines because it differs from one region to the other.

Unlike the levels of adaptive capacity, those of exposure/sensitivity show a similar

pattern for groups of provinces in the same regions (Figure 8). This is not surprising because

indicators of water availability and soil quality are influenced by geographical locations. The

highest number of provinces with very high level of exposure/sensitivity is found in

Mindanao. All provinces in regions 11 and 12 show very high level of exposure/sensitivity. In

addition, two provinces each in region 9 (i.e., Zamboanga del Norte and Zamboanga del Sur)

and region 10 (i.e., Lanao del Norte and Misamis Oriental) show very high level of

exposure/sensitivity. In Luzon, several provinces also show very high level of

exposure/sensitivity. For example in region 1 and region 5 (except for the province of

Catanduanes), all provinces have very high level of exposure/sensitivity. The NCR also

shows very high level of exposure/sensitivity of exposure, which is not surprising because of

the large number of population demanding land and water resources in very small area.

Whilst there is no province with very high level of exposure/sensitivity in Visayas Island,

there is no province with very low level of exposure/sensitivity in Mindanao Island. The areas

with very low level of exposure/sensitivity are mostly found in the middle of the Philippines,

covering several provinces in Visayas and Luzon Islands. All provinces in region 6 in Visayas

and almost all provinces in region 4 in Luzon have very low level of exposure/sensitivity.

The most vulnerable provinces in the Philippines are located in many regions in

Mindanao, followed by Luzon. The capacity of the people to adapt to the impacts of climate

change will be lower in Mindanao than in Luzon. As compared to many provinces in Luzon,

those in Mindanao have lower purchasing income and farmers depend mainly on their own

crops for food supply (Vinck and Bell, 2011). Moreover, nowhere in the Philippines is

poverty more widespread than in Mindanao where people are confronted not only with

economic but also social instability (Ringuet, 2002). In 2009 the families in National Capital

Region in Luzon had the highest average annual family income at 356,000 Pesos, whilst the

families in Autonomous Region in Muslim Mindanao (ARMM) had the lowest average

annual family income at 113,000 Pesos (NSO, 2011). In terms of source of employment,

more than 70% of the business establishments are located in Luzon and only 13% in



Mindanao (NSO, 2010). Vulnerability in many provinces in Mindanao can thus be reduced by

improving the institutional problems, which hinders socio-economic development. In Luzon,

vulnerability in many provinces can be reduced by reducing the level of exposure/sensitivity.

Improvement in irrigation systems in northern Luzon, particularly in region 2 where 57.9

percent of the labour force depends primarily on agriculture will help people to adapt to the

impacts of climate change. After region 3 (Central Luzon – 17.32 percent), region 2 (Cagayan

Valley – 16.48 percent) account for the largest irrigated areas in the Philippines in 2012

(BAS, 2013). But like in region 3, the remaining potential for irrigation development is

highest in region 2 at 14 percent. Provinces with very low level of vulnerability are located in

the southern regions in Luzon and western regions in Visayas, where levels of

exposure/sensitivity are very low.

Adaptive Capacity Exposure/Sensitivity VulnerabilityAdaptive Capacity Exposure/Sensitivity Vulnerability

Figure 8. Spatial patterns in adaptive capacity, exposure/sensitivity and vulnerability in the Philippines,

2000.

CONCLUSION

Indicator-based approach to vulnerability assessment provides simple synthesis of

complex issues, rapid system for scoping and mapping, and generalisation of multiple spatial

scales. These advantages of the approach are often considered its weaknesses. Under this

approach, the indices are often generated through aggregation of socio-economic and

biophysical indicators. Such indicators, however, only inform either about the state of human

condition (e.g., level of income, health and education) and that of the environment (e.g., soil

quality, water stress) or about global environmental change (e.g., future climate and price

variability). The vulnerability indices are thus no more than aggregated information about

human and natural environment. Moreover, whilst the vulnerability maps provide information

about vulnerable regions, they are limited when identifying the most vulnerable people in

vulnerable places. The maps can provide only little knowledge on appropriate adaptation

policies to help the vulnerable people. Finally, maps can only describe the spatial variation in

vulnerability but not the dynamics of human (individuals or communities) vulnerability. Even



temporal progression of maps cannot measure the dynamics of vulnerability, unless human

adaptive behaviour, which contributes a lot to these dynamics, is taken into account in the

development of vulnerability indices. The vulnerability maps for the Philippines, for example,

can inform policy of the potential vulnerable provinces but not of vulnerable people. It will be

difficult to identify appropriate adaptation policy unless the groups of vulnerable people are

also known.

The forgoing does not imply however that indicator-based approach is completely

inappropriate for vulnerability assessment. Due to the simplicity, quickness and generality of

the approach, it is a useful component of an integrated vulnerability assessment framework

such as intervulnerability. As adaptation issue is core to human dimensions research on

vulnerability to global environmental change, vulnerability maps should be considered the

starting, and not the finishing point for vulnerability assessment. Knowledge generated from

indicator-based approach will make implementation of other approaches like agent-based

more efficient and focused. Whilst still in its infancy in the field of climate change, an agent-

based approach is a very promising tool for vulnerability assessment in a changing global

environment because it provides great potential for extending our understanding of complex

social systems and how aggregate structures emerge from local rules of interaction. However,

models based on multi-agent systems are time-consuming and data hungry, which limit their

application to well documented case study areas. It is thus not a real alternative to indicator-

based approach, but a complementary approach to understand the knowledge generated from

indicators.

Vulnerability assessments using indicator-based approach rely on methods that can

combine a diverse and wide range of information on adaptive capacity and exposure/

sensitivity. The information is represented by various factors like social, economic,

institutional and ecological and covers various types like qualitative and quantitative.

Vulnerability assessments based on diverse and multitude indicators thus require a flexible

tool for aggregation of information. This chapter shows that fuzzy logic is a very useful and

flexible tool for assessing vulnerability using indicator-based approach. The empirical

application in the Philippines showed the advantages of fuzzy set theory in terms of its (1)

transparency, which allows explicit presentation of model assumptions through inference

rules; and (2) flexibility, which allows direct inclusion of informal and expert knowledge in

combining various indicator sets. Because of its flexibility, vulnerability assessment using

fuzzy logic facilitates not only identification of hotspots of vulnerable regions but also

comparison of trade-offs among sets of adaptive capacity indicators representing economic,

social, and institutional capacity.

REFERENCES

Acosta, L.A. & Galli, F. (2013) Crisis Probability Curves (CPCs): A Model for Assessing

Vulnerability Thresholds Across Space and Over Time. Journal of Environmental

Science and Management, 16, 36-49.

Acosta, L.A., Klein, R.J., Reidsma, P., Metzger, M.J., Rounsevell, M.D., Leemans, R. &

Schröter, D. (2013) A spatially explicit scenario-driven model of adaptive capacity to

global change in Europe. Global Environmental Change, 23, 1211–1224.



Acosta-Michlik, L. & Espaldon, V. (2008) Assessing vulnerability of selected farming

communities in the Philippines based on a behavioural model of agent‟s adaptation to

global environmental change. Global Environmental Change, 18, 554–563.

Acosta-Michlik, L. & Espaldon, V. (2011) Reducing vulnerability of rural communities in the

Philippines: Modelling social links between science and policy. Integrating Science and

Policy. Kasperson, R.E. & Berberian, M.), pp. 335-362. Earthscan, London, UK.

Acosta-Michlik, L. & Rounsevell, M. (2005) From Generic Indices to Adaptive

Agents:Shifting Foci in Assessing Vulnerability to the Combined Impacts of Climate

Change and Globalization. IHDP Update, 01, 14-16.

Acosta-Michlik, L. & Rounsevell, M. (2009) An Agent-based Framework for Assessing

Vulnerability Futures. Assessing Vulnerability to Global Environmental Change: Making

information useful for adaptation policy and decision-making. Patt, A., Schröter, D.,

Klein, R. & Vega-Leinert, A.D.), pp. 147-172. Earthscan, London.

Acosta-Michlik, L., Eierdanz, F., Alcamo, J., Krömker, D., Galli, F., Klein, R.J., Kumar,

K.K., Campe, S., Carius, A. & Tänzler, D. (2006) An Empirical Application of the

Security Diagram to Assess the Vulnerability of India to Climatic Stress. Die Erde, 137,

223-252.

Acosta-Michlik, L.A., Kumar, K.S., Klein, R.J. & Campe, S. (2008) Application of fuzzy

models to assess susceptibility to droughts from a socio-economic perspective. Regional

Environmental Change, 8, 151-160.

Adger, W.N. (1999) Social Vulnerability to Climate Change and Extremes in Coastal

Vietnam. World Development, 27, 249-269.

Adger, W.N., Brooks, N., Bentham, G., Agnew, M. & Eriksen, S. (2004) New indicators of

vulnerability and adaptive capacity. Tyndall Centre for Climate Change Research

Technical Report 7, Norwich, England.

Baraldi, P., Librizzi, M., Zio, E., Podofillini, L. & Dang, V.N. (2009) Expert Systems with

Applications. Expert Systems with Applications, 36, 12461-12471.

BAS (2013) CountrySTAT, Bureau of Agricultural Statistics (BAS), Quezon City,

Philippines. http://countrystat.bas.gov.ph/

Castro-Leal, F., J. Dayton, et al. (2000): "Public Spending on health care in Africa: do the

poor benefit?" Bulletin of the World Health Organization 78(1): 66-74.

Cornelissen, A., Van Den Berg, J., Koops, W., Grossman, M. & Udo, H. (2001) Assessment

of the contribution of sustainability indicators to sustainable development: a novel

approach using fuzzy set theory. Agriculture, Ecosystems and Environment, 86, 173-185.

Darbra, R.M., Eljarrat, E. & Barcelo, D. (2008) How to measure uncertainties in

environmental risk assessment. Trends in Analytical Chemistry, 27, 377-385.

DOE (2010) "Renewable Energy." City. http://www.doe.gov.ph/IPO web/linkedfiles/

2010/EIK10/biomass_biofuels.pdf.

Downing, T.E. (2001) Vulnerability to hunger in Africa: A climate change perspective.

Global Environmental Change, 1, 365–380.

Eierdanz, F., Alcamo, J., Acosta-Michlik, L., Krömker, D. & Tänzler, D. (2008) Using fuzzy

set theory to address the uncertainty of susceptibility to drought. Regional Environmental

Change, 8, 197-205.

Filmer, D. and L. Pritchett (1999): "The impact of public spending on health: does

monmatter?" Social Science & Medicine 49: 13 09-1323.



Filmer, D., J. Hammer & L. Pritchett (1997): Health Policy in Poor Countries: Weak Links in

the Chain, Policy Research Working Papers, Washington, DC, World Bank: pp. 63.

http://www.worldbank.org/ poverty/health/library/papers.htm, 01/27/04.

Fraser, E.D., Simelton, E., Termansen, M., Gosling, S.N. & South, A. (2013) “Vulnerability

hotspots”: Integrating socio-economic and hydrological models to identify where cereal

production may decline in the future due to climate change induced drought. Agricultural

and Forest Meteorology, 170, 195-205.

Fuentes, N. and M. Rojas (2001): "Economic theory and subjective well-being: Mexico,

Social Indicators Research 53: 289-314.

Gan, G., C. Ma and J. Wu (2007) Data clustering: theory, algorithms and applications, ASA-

SIAM Series on statistics & applied probability, SIAM, Philadelphia, ASA, Alexandria,

VA.

Gomiero, Tiziano, Maurizio G Paoletti, and David Pimentel (2010) Biofuels: Efficiency,

Ethics, and Limits to Human Appropriation of Ecosystem Services. Journal of

Agricultural and Environmental Ethics 23:403-434. http://www.springerlink.com/

content/v5267n80055l26q1/fulltext.pdf.

Gwatkin, D. R. (2003) Free government health services: are they the best way to reach the

poor? Washington, DC, World Bank: pp. 13. http://www.worldbank.org/ poverty/

health/library/papers.htm, 01/27/04.

Heitzmann, K., Canagarajah, R.S. & Siegel, P.B. (2002) Guidelines for Assessing the Sources

of Risk and Vulnerability. Washington, D.C.

Henninger, N. (1998) Mapping and Geographic Analysis of Human Welfare and Poverty --

Review and Assessment. World Resources Institute, Washington, D.C.

Hinkel, J. (2011) „„Indicators of vulnerability and adaptive capacity‟‟: Towards a clarification

of the science–policy interface. Global Environmental Change, 21, 198-208.

IPCC. (2001) Climate Change 2001: Impacts,Adaptation, And Vulnerability (Contribution of

Working Group II to the Third Assessment Report of the Intergovernmental Panel on

Climate Change) (JJ McCarthy, OF Canziani, NA Leary, DJ Dokken, and KS White).

Cambridge University Press, Cambridge, United Kingdom.

IPCC. (2007) Climate Change 2007: The Physical Science Basis (Contribution of Working

Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate

Change) (S Solomon, D Qin, M Manning, Z Chen, M Marquis, K Averyt, M Tignor, and

H Miller). Cambridge University Press, Cambridge, United Kingdom.

Janssen, J., Krol, M., Schielen, R., Hoekstra, A. & Kok, J.D. (2010) Assessment of

uncertainties in expert knowledge, illustrated in fuzzy rule-based models. Ecological

Modelling, 221, 1245-1251.

Judge, K., J.-A. Mulligan, et al. (1997) Income inequality and population health. SociScience

& Medicine 46(4-5): 567579.

Kaiser, F., L. v. d. Maat, et al. (2000) Signs of the times monitoring the position of

dutch education. Enschede, Center for Higher Education Policy Studies (CHEPS):

pp.123. http://www.utwente.nl/cheps/ publications/downloadable_publications/Down

loadablesenglish.doc/, 01/27/04.

Kangas, A. S. and J. Kangas (2004) Probability, possibility and evidence: approaches to

consider risk and uncertainty in forestry decision analysis. Forest Policy and Economics

6, 169–188.



Krömker, D., Eierdanz, F. & Stolberg, A. (2008) Who is susceptible and why? An agent-

based approach to assessing vulnerability to drought. Regional Environmental Change, 8,

173-185.

Liu, J., Fritz, S., Fuchs, M., You, L., Obersteiner, M. & Yang, H. (2008) Global and Planetary

Change. Global and Planetary Change, 64, 222-235.

Mackay, D. S. and V. B. Robinson (2000): A Multiple Criteria Decision Support System for

Testing Integrated Environmental Models. Fuzzy Sets and Systems 113: 53-67.

Mays, M. D., I. Bogardi and A. Bardossy (1997): Fuzzy Logic and Risk-Based Soil

Interpretations. Geoderma 77: 299-315.

Mays, M., Bogardi, I. & Bardossy, A. (1997) Fuzzy logic and risk-based soil interpretations.

Geoderma, 77, 299–315.

Metzger, M., Rounsevell, M., Acosta-Michlik, L., Leemans, R. & Schröter, D. (2006) The

vulnerability of ecosystem services to land use change. EAgriculture, Ecosystems and

Environment, 114, 69-85.

Moss, R.H., Brenkert, A.L. & Malone, E.L. (2001) Vulnerability To Climate Change: A

Quantitative Approach. Online, 12.

NEDA Region VI 2013, Regional Development Council, National Economic Development

Authority (NEDA), Iloilo City, Province of Iloilo. www.neda-rdc6.ph/website/

index.php/99-latest-news-articles/24

NSO (2010) Annual survey of philippine business and industry (2008 Preliminary results),

National statistics office (NSO), Manila, Philippines; 2010. http:// www.census.gov.ph/

data/sectordata/2008/ aspbi08_04b.htm

NSO (2011) Family income and expenditure survey (Final results updated 2011). Income and

employment statistics division, national statistics office, Republic of the Philippines;

2011. http://www.census.gov.ph/ data/ pressrelease/2011/ie09frtx.html

O‟Brien, K.L., Leichenko, R., Kelkar, U., Venema, H., Aandahl, G., Tompkins, H., A. Javed,

S. & Bhadwal, S. Barg, L.N. (2004) Mapping vulnerability to multiple stressors: climate

change and globalization in India. Global Environmental Change, 14, 303-313.

O'Brien, K.L. & Leichenko, R.M. (2000) Double exposure: assessing the impacts of climate

change within the context of economic globalization. Global Environmental Change, 10,

221-232.

PAGASA (2013) Climate of the Philippines, Philippine Atmospheric, Geophysical and

Astronomical Services Administration (PAGASA). http://kidlat.pagasa. dost.gov.ph/cab/

climate.htm

Phillis, Y.A. & Andriantiatsaholiniaina, L.A. (2001) Sustainability: an ill-defined concept and

its assessment using fuzzy logic. Ecological Economics, 37, 435- 456.

Polsky, C., Patt, A., Gaffin, S., Martello, M.L., Neff, R., Pulsipher, A. and Selin, H. (2003)

Assessing Vulnerabilities to the Effects of Global Change: An Eight-Step Approach,

Belfer Center for Science & International Affairs. Cambridge, MA.

Ramachandran, M. and J.R. Eastman (1997): Applications of GIS to Vulnerability Mapping:

A West African Food Security Case Study. Applications of Geographic Information

Systems (GIS) Technology in Environmental Risk Assessment and Management, UNEP

and the Clark Labs for Cartographic Technology and Geographic Analysis.

Ramieri, E., Hartley, A., Barbanti, A., Santos, F.D., Gomes, A., Laihonen, P., Marinova, N. &

Santini, M. (2011) Methods for assessing coastal vulnerability to climate change.



Copenhagen. ETC CCA Technical Paper 1/2011, European Environment Agency,

Copenhagen.

Ringuet D.J. (2002) The continuation of civil unrest and poverty in Mindanao. Contemp

Southeast Asia 24(1):33–49.

Roberts, D. W. (1996): Modelling Forest Dynamics with Vital Attributes and Fuzzy

SysteTheory. Ecological Modelling 90: 161-173.

Shen, C. and J. B. Williamson (2001): Accounting for crossnational differences in infant

mortality decline (1965-1991) among less developed countries: effects of women's status,

economic dependency, and state strength. Social Indicators of Research 53: 257-288.

Singh, R.K., Murty, H.R., Gupta, S.K. & Dikshit, A.K. (2012) Ecological Indicators.

Ecological Indicators, 15, 281-299.

Taenzler, D., Carius, A. & Maas, A. (2008) Assessing the susceptibility of societies to

droughts: a political science perspective. Regional Environmental Change, 8, 161-172.

The MathWorks, Inc. (2011) MATLAB and Fuzzy Logic Toolbox Release 2011b, The

MathWorks, Inc., Natick, Massachusetts, United States.

Turner, B.L., Kasperson, R.E., Matson, P.A., Mccarthy, J.J., Corell, R.W., Christensen, L.,

Eckley, N., Kasperson, J.X., Luers, A., Martello, M.L., Polsky, C., Pulsipher, A. &

Schiller, A. (2003) A framework for vulnerability analysis in sustainability science,

PNAS 100 (14):8074–8079.

Vinck P and Bell E. (2011)Violent conflicts and displacement in central Mindanao:

challenges to recovery and development. The World Food Programme and the World

Bank.

Wang, L. (2003): Determinants of child mortality in LDCs empirical findings from

demographic and health surveys. Health Policy, 2003, 65:277-299.

Wu, H.I.; B.L. Li, R. Stoker and Y. Li (1996): A Semi-Arid Grazing EcModel with

Probabilistic and Fuzzy Parameters. Ecological Modelling 90 : 147-160.

APPENDIX

Appendix 1. Data sources of socio-economic

and biophysical indicators

Determinants Indicators Data source

Poverty and

inequality

Poverty Incidence Of Families

(Unit: Percent)

National Statistical Coordination

Board (2005)

Food Subsistence Incidence Of

Families

(Unit: Percent)


Board (2005)

Gini Coefficient

(Unit: Range 0-1)


Board (2005)




Income and

purchasing

power

Average Annual Family Income

(Unit: Pesos)


Board (2002b)

Savings: Average Annual

Family Less Expenditures

(Unit: Pesos)


Board (2002b)

Inflation Rate

(Unit: Percent)


Board (2002b)

Good

governance

Total Financial Resources Per

Capita

(Unit: Pesos)


Board (2002b)

Non-Tax Revenue

(Unite: Pesos)


Board (2002b)

Share Of Social Services

Expenditure To Total

Expenditures

(Unit: Percent)


Board (2002b)

Human

security

Unemployment Rate

(Unit: Percent)


Board (2002b)

Average Monthly Crime Rate

(All Types)

(Unit: Number Of Incidents Per

100,000 Population

Philippine Institute for

Development Studies (PIDS)

Crime Solution Efficiency –

Percentage Of Solved Cases Out

Of Total Crime Cases

(Unit: Percent)



Education

situation

Literacy Rate

(Unit: Percent)



Higher Education Enrolment

(Unit: Number Of Students)


Board (2002b)

Cohort Survival Rate For

Secondary Level

(Unit: Percent)



Health

condition

Water-Born Diseases By Type

(Unit: Number Of Cases)


Board (2002a)

Proportion Of Households With

Access To Sanitary Toilet

(Unit: Percent)


Board (2002b)

Proportion Of Households With

Access To Safe Water Supply

(Unit: Percent)


Board (2002b)



Appendix 1. (Continued)


Education &

health services

Public Elementary + Secondary

Schools

(Unit: Number)


Board (2002b)

Teacher-Student Ratio In Public

Secondary Schools

(Unit: Ratio)


Board (2002b)

Percent Of Barangay Health

Stations

To Total Barangays

(Unit: Percent)


Board (2002b)

Active Barangay Health

Workers

(Unit: Number)


Board (2002b)

Road

Infrastructure

Length Of National Roads

(Unit: Kilometer)


Board (2002a)

Length Of Poor Quality Roads

(Unit: Kilometer)


Board (2002a)

Total Number Of Ports

(Unit: Number)


Board (2002a)

Water

infrastructure

Large Scale Dams

(Unit: Number And Capacity)


Board (2002a)

Hydropower Schemes

(Unit: Storage Capacity)


Board (2002a)

Water District Percent

Population Served

(Unit: Percent)


Board (2002a)

Groundwater

supply

Amount Of Water Abstracted By

Source Of Water Supply

(Unit: Liters Per Second)


Board (2002a); National Statistical

Coordination Board (2000)

Physical Accounts Of

Groundwater

(Unit: Million Cubic Meters)


Board (1998)

Rainwater

supply

Average Rainfall

(Unit: Millimeter Per Year)


Board (2002a)

Rainy Days Per Year

(Unit: Number Of Days)


Board (2002a)

Soil erosion

Lands Suffering From Moderate

(E2) To Severe (E3) Erosion

(Unit: Degree)


Board (2002a); National Statistical

Coordination Board (1998)

Estimated Nutrient Loss From

On-Site Erosion (ONSE) Due To

Upland Palay Farming


Board (2000)




(Unit: Metric Tons)

Land use

change

Forest Destruction

(Unit: Hectares)


Board (1998)

Land Use Conversion By Region

(Unit: In Hectares)


Board (1998)





Chapter 6

ECOLOGICAL RESILIENCE: IS IT READY FOR

OPERATIONALISATION IN FOREST MANAGEMENT?

Gerardo Reyes1,2* and Daniel Kneeshaw

2

1Department of Interdisciplinary Studies, Lakehead University,

Orillia, Ontario, Canada 2Centre for Forest Research, Department of Biological Sciences,

University of Quebec in Montreal, Montreal, Quebec, Canada

ABSTRACT

Given the physiographic variability, variation in socio-political landscapes, and

differences in connectedness of people and communities associated with boreal forest

ecosystems, approaches to forest management that are flexible enough to accommodate

this variation are needed. Moreover, to ensure sustainable forest resource use, we need to

embrace the inherent complexity of boreal forest ecosystems rather than eliminate it, and

be prepared to adapt and adjust as environmental conditions change. While ecological

resilience may be a useful forest management objective to this end, developing general

guidelines to integrate it into practice remains elusive. We address a number of questions

often posed by managers when attempting to include ecological resilience into forest

management planning. Our goal is to determine if the theoretical foundation of ecological

resilience is sufficiently developed to provide a general framework that can be applied for

boreal forest management.

Keywords: Boreal forests, ecological resilience, stability and change, adaptation, forest

ecosystem management

* E-mail : [email protected]; [email protected].


Gerardo Reyes and Daniel Kneeshaw 196

1. INTRODUCTION

Given the overwhelming environmental, social, and economic importance of forests to

humankind, preoccupation is growing for promoting the sustainable use of forest ecosystem

resources. Because of expected rapid changes in global conditions over the next century

(Sokolov et al., 2009) and the potential consequences of these changes to our well-being, it is

imperative that we develop and implement forest management strategies that ensure forests

will continue to provide us with needed resources and services. Ecological resilience, an

ecosystem‟s ability to re-organize and adapt to disturbance or environmental change without

shifting to an undesirable alternative state (Holling, 1973; Gunderson, 2000), is a concept that

has been proposed to help us to achieve this objective.

Ecological resilience was conceptualized to help explain unexpected and nonlinear

dynamics observed in complex adaptive systems (Holling, 1973; Gunderson and Holling,

2002), thus providing a theoretical foundation towards spatio-temporal understanding of how

boreal forest ecosystems may respond to any changes in climate, natural and anthropogenic

disturbances, invasive species, resource utilisation, and so forth. Managing for ecological

resilience is said to promote sustainability by enhancing a forest ecosystem‟s adaptive

capacity (Gunderson, 2000; Allen and Holling, 2010), defined as the magnitude of an

ecosystem‟s component species‟ ability to respond and adapt to disturbance or change before

collapsing and shifting to a new stability domain, even as the shape or breadth of the domain

changes (Figure 1). In other words, maintaining or improving the ability of species within

ecosystems to respond to episodic disturbances or gradual change will improve an

ecosystem‟s chance of avoiding progression towards an unwanted ecological state (Holling,

1973; Walker et al., 2004). Thus, forest ecosystems adapted to natural as well as imposed

anthropogenic disturbance regimes will have greater capacity to re-organize and retain

desired characteristics and functions, and by consequence be more resilient. Central to this

tenet is that while post-disturbance conditions in resilient ecosystems are not expected to be

exactly like those that existed prior to disturbance, as structural and compositional changes

occur, the same critical processes driving the system are upheld (e.g., photosynthetic capacity,

nutrient cycling, disturbance regime, etc.). Changes in critical processes can drive an

ecosystem into a new stability domain and thus it is imperative that we focus our attention on

understanding their roles in ecosystem maintenance.

Along with the direct changes to forest structure and natural ecological processes caused

by forest management, climate change presents new and unique challenges that will make

sustainable management of boreal forest ecosystems far more difficult to achieve given the

potential for it to interact with processes such as nutrient and hydrological cycling,

disturbance regimes, pollination, etc. (Bonan, 2008; Berggren et al., 2009, Huntington et al.,

2009). Even now, fundamental changes to environmental conditions are occurring at an

unprecedented rate (Bentz et al., 2010; Kilpeläinen et al., 2010; Fettig et al., 2013) and we are

uncertain about the nature, magnitude, and timing of the effects. Given this uncertainty, an

adaptive approach for boreal forest ecosystem management is essential. To this end, the idea

of making forest ecosystems resilient to these challenges is certainly appealing. However,

operationalizing the concept; i.e., actively managing for resilience has a number of stumbling

blocks that require attention.


Ecological Resilience 197

Figure 1. Ball and cup conceptual model of ecological resilience and ecosystem state change. Balls

represent different stable ecological states, each within a domain of attraction controlled by a unique set

of processes. A threshold point is exceeded when a disturbance or change is so intense, severe, or

frequent that an ecosystem is driven into a qualitatively different stable ecosystem state and controlled by

a different set of ecological processes. In this example, grassland, boreal forest, and heathland

ecosystems are shown. In (A) ecological states shift into alternate stable states when disturbances of

sufficient magnitude or gradual changes drive them beyond ecological threshold points and into different

domains of attraction; (B) the domain itself can change in depth or width over time due to slow changes

in controlling processes (e.g., climate, acid rain), also potentially causing shifts in ecological state as well

as changing an ecosystem‟s overall resilience. Red arrows with solid lines indicate changes within an

ecosystem‟s natural range of variation. These shifts may be caused by natural disturbances such as fire or

insect outbreaks for which ecosystem components have developed adaptations to; i.e., the disturbances

have historical precedents. Red arrows with dotted lines indicate that restoration effort may be required if

attempting to shift a system from an unwanted ecological state back into another.

Modified from Gunderson (2000).

Our purpose here is to examine the applicability of ecological resilience as a management

option in boreal forest ecosystems. We address a number of questions directly related to

„putting the concept on the ground‟ for a hypothetical forest management unit. Ultimately, we

wish to determine if the theoretical foundation of ecological resilience is developed enough to

provide a general framework that can be applied for any boreal forest management unit.

A

B



2. INTEGRATING ECOLOGICAL RESILIENCE

INTO FOREST MANAGEMENT

Those involved with boreal forest management decisions usually raise questions such as:

i. For what components of an ecosystem should we build ecological resilience?

ii. What do we need to know to manage for ecological resilience?

iii. At what spatio-temporal scales should we focus our management efforts? and

iv. How do we determine if a system is resilient or not?

Proponents of resilience thinking have responded by stating that managing for ecological

resilience requires:

i. clearly defined stakeholder objectives;

ii. knowledge of critical processes and drivers that promote ecosystem stability or

ecosystem change;

iii. knowledge of the ecological impacts of cultivating, harvesting, or using various

ecosystem resources or services at multiple scales; and

iv. indicators of the adequacy of resilience via proxies such as biological diversity,

structural heterogeneity, response diversity, and ecological redundancy.

(Fischer et al. 2006, Campbell et al. 2009, Thompson et al. 2009)

For the remainder of this section, we address each of the above questions in relation to

the responses in more detail.

2.1. For What Components of an Ecosystem Should We Build Ecological

Resilience?

A starting point for operationalising ecological resilience is for stakeholders to determine

what objectives to manage for. The question of „resilience of what to what?‟ (Carpenter et al.,

2001) forces managers to clearly define objectives for the entire forest management unit and

explicitly specify their relative importance and spatio-temporal impacts across the landscape.

This can include managing for timber supply, maintaining biodiversity or old-growth forest,

provisioning of water, or providing opportunities for recreational activities. However, it

should be recognized that managing for one desired aspect of an ecosystem may reduce

resilience of another. This apparent paradox stems from what Holling and Meffe (1996)

called a „command and control‟ approach to managing resources. They note that managers

have simplified ecosystems to maximize the production of a desired resource; and that it this

simplification that reduces the adaptability of a system and thus the resilience of its non-

targeted components.

When entire forest management units are managed for only one purpose, tradeoffs are

inevitable. We cannot maintain resilience for everything everywhere because of fundamental

differences in species‟ life history requirements, feedbacks and interactions among species,

and conflicting stakeholder interests. For example, management plans may include provisions



to enhance white-tailed deer (Odocoileus virginianus) habitat. Large herd sizes provide

greater opportunities for hunters and naturalists but also result in heavy browsing damage to

regenerating commercial tree species (Rooney and Waller, 2003). Moreover, improving

hunting opportunities entails having a mix of favorable habitat types across the landscape that

includes conifer forest cover for shelter during winter, an abundance of clearings that provide

herbaceous plants, forbs, and browse for deer to forage, as well as maintaining logging roads

for human access (Voigt et al., 1997). Conversely, protecting pine marten (Martes americana)

populations in the same forest management unit may require maintaining large tracts of intact

mature mixed-coniferous forest containing spruce (Picea spp.), fir (Abies spp.), or cedar

(Thuja occidentalis), and limiting the fragmentation across the landscape that favours deer

(Watt et al., 1996). Many other associated plant and animal species also draw benefits or are

negatively impacted by conditions that promote elevated deer population densities (de Calesta

1994, Gill and Beardall, 2001). Extremely high population densities have, for example,

shifted the forest state on Anticosti Island from a balsam fir (Abies balsamea) to white spruce

(Picea glauca) dominated forest with concomitant losses or decreases of many herbaceous

species palatable to deer (Potvin et al., 2003; Morissette et al., 2009). Managing for a single

resource invariably reduces habitable conditions for other elements in the ecosystem and may

be a critical driver for shifting ecological states.

2.2. What Do We Need to Know to Manage for Ecological Resilience?

Whether our desire is to simply maintain a functioning forest ecosystem or to maintain a

specific type of forest ecosystem, building ecological resilience entails identifying the critical

processes that drive the ecosystem (Table 1). Species and ecosystems are adapted to

ecological processes that have historical precedents (Peterson, 2000; Read et al., 2004;

Johnstone et al., 2010). Retaining these processes is thus an approach that can be proactively

used to maintain ecological resilience. This is in fact, the original premise behind the

Emulating Natural Disturbance (END) concept (Gauthier et al., 2008). Moreover, if the focus

is centered on emulating processes rather than patterns END would escape some (but perhaps

not all) of the critiques of managing for past patterns in a changing environment (Klenk et al.,

2009). Understanding the natural variability in processes and species adaptations to them can

identify the type and range of processes that will maintain the stability of a desirable state, as

well as those that will lead to unwanted ecosystem state changes.

Natural processes that can lead to an ecosystem state change includes paludification,

which results in the conversion of conifer forests into peat bogs over time (Lavoie et al.,

2005). The process can be magnified by human activity when dominant or correcting

processes are not understood. For example, severe fires that burn into the moss layer can

reduce or reverse paludification whereas partial or less severe disturbances such as windthrow

or senescence (e.g., pathogen caused tree mortality) that do not disturb the soil (moss) layer

accelerate the process. Consequently, the blanket approach of using harvesting that protects

soils and advance regeneration (Leblanc and Pouliot, 2011) creates conditions favourable for

stand conversion whereas more aggressive silvicultural techniques that include scarification

would better emulate the soil disturbing processes that naturally control paludification.



Table 1. Factors that impact ecological resilience at various spatial scales in boreal

forest ecosystems

scale Process Structure Other

environmental

factors

anthropogenic

impacts

Stand seed dispersal,

natural

regeneration,

competition,

pollination,

herbivory, disease,

photosynthesis,

respiration, evapo-

transpiration,

nutrient cycling,

allelopathy,

mycorrhizal

association

vertical,

horizontal, stand

density, relative

species mixes,

patch size &

shape

soil moisture,

pH, light

availability,

temperature,

nutrient

availability,

slope-aspect,

altitude, latitude,

Timber harvest,

soil erosion,

compaction,

land conversion,

invasive species,

climate change,

conversion,

structural and

compositional

simplification,

pollution

Landscape Natural

disturbance,

succession,

nutrient cycling,

hydrological

cycling,

paludification

Variation in

forest types &

age class, stand

pattern &

connectivity

Soil moisture,

nutrient

availability,

physiography

Fragmentation,

homogenization,

sedimentation &

waterflow

alteration,

climate change,

pollution

Region Primary

production climate

regulation

Variation in

forest types &

age class, patch

pattern &

connectivity

temperature,

precipitation,

CO2, ozone, N

deposition &

uptake,

physiography

Fragmentation,

homogenization,

sedimentation &

waterflow

alteration,

climate change,

pollution

Understanding the dominant processes and their interactions is clearly an important step

to effective management. Such an understanding will be critical when dealing with novel

combinations of disturbances such as the interaction of allelopathy, clearcutting, and fire that

have resulted in some conifer forest ecosystems to be converted to heathlands (Mallik, 1995;

Payette and Delwaide, 2003), invasive insect pests that can substantially alter forest structure

and composition (Dukes et al., 2009), and use of other inappropriate harvesting analogues

(Nitschke, 2005; Salonius, 2007; Taylor et al., 2013).

Unprecedented changes to the historical frequency or severity of natural disturbances is

also problematic. Fire regimes that are more frequent than the age of sexual maturity of tree

species, for example, can lead to ecosystem change. Increased frequency of stand-replacing

fires has resulted in conversion of aspen woodland to conifer forest (Strand et al., 2009) and

conifer forests to grasslands (Heinselman, 1981; Hogg and Hurdle, 1995; Beckage and

Ellingwood, 2008). Noble and Slatyer (1980) used knowledge of these processes and tree



functional attributes to identify when and why species shifts would occur as disturbance

processes changed. In fact, ecological research throughout history has been about identifying

shifts in ecosystem states due to changes in natural disturbances as well as those caused by

humans (Frelich and Reich, 1998). Clements (1928) was concerned about how the

agricultural practices of his time influenced the integrity of mid-plains ecosystems. Holling

(1978) identified the importance of disturbance in maintaining resilience; exemplified by the

spruce budworm (Choristoneura fumiferana) maintaining balsam fir forests in the Maritimes

by killing the canopy and releasing understory trees whereas fire, which is a rare disturbance

in this ecosystem, can lead to a different forest type. Thus, processes that can cause ecosystem

collapse usually do not have historical precedents and are often the result of anthropogenic

changes to the timing or severity of natural processes. Accordingly, human disturbances

should be evaluated in light of the processes they affect and the subsequent impacts on

species present across the landscape.

2.3. At What Spatio-temporal Scales Should We Focus Our Management

Efforts?

Ecological resilience changes over time and space. Thus, understanding the critical

processes driving a system must include knowledge of the spatio-temporal scales over which

they operate and interact (e.g., Heinselman, 1981; Gunderson and Holling, 2002; Mladenov et

al., 2008). Different ecological processes influence community structure and composition at

different spatial and temporal scales (Ricklefs, 1987; Herzog and Kessler, 2006; Seppä et al.,

2009) (Figure 2). Certain processes can also have impacts across scales of measure. These

processes often do not function in a simple linear fashion, nor do they function independently

of one another (Peterson et al., 1998; Frelich and Reich, 1999; Groffman et al., 2006).

Extrapolating ecosystem responses to these processes by scaling up or down may result in

erroneous assumptions and predictions due to non-linear relationships, differences in

environmental characteristics at different scales, and emergent properties (Peterson, 2000;

Turner et al., 2001). Therefore, we need to understand if and how critical processes impact

our forests at stand, management unit, and regional levels.

Effective management requires careful planning of how each desired objective is

distributed across the forest management unit. Thus, the scope should be large enough to

generate region-wide ecological benefits that compensate for impacts of an objective at a

single site as well as the cumulative impacts of multiple interventions of this and various

other objectives over time. For example, while the effects of logging are site specific, we need

to consider the spatio-temporal impacts on the forest management unit as a whole; not just

accommodate short-term and local needs or demands. If an associated objective is to maintain

structural complexity across the landscape, including large tracts of mature forest to provide

core habitat for wildlife and various aesthetic values, then a mixture of large and small cuts

arranged in an aggregated pattern across the management unit could allow for more intact,

interior forest conditions to be retained across the landscape relative to a strategy creating

smaller, uniform patches distributed systematically. Over time, a more fragmented landscape

with a greater edge-to-interior ratio may develop utilising a systematic approach (Turner et

al., 2001).



Figure 2. Some important processes affecting boreal forest ecosystems across spatio-temporal scales.

2.4. How Do We Determine If a System Is Resilient or Not?

At this time, ecological resilience can only be coarsely quantified using a proxy; i.e.,

measured in terms of the amount of biodiversity, structural heterogeneity, response diversity,

and ecological redundancy. Biodiversity and structural heterogeneity are defined as the

amount of variation in biological (genes, species, and ecosystems) and structural elements

(vertical strata of extant vegetation, spatial arrangement of patches, snags, coarse woody

debris, pit & mound topography, etc.), respectively (Hunter, 1999). Response diversity is the

variation in responses of functionally similar species to disturbance (Elmqvist et al., 2003);

for example, black spruce (Picea mariana) regenerates almost exclusively from the abundant

seed rain after severe fire while white birch (Betula papyrifera) and poplars (Populus spp.)

can reproduce via seed, but can also regenerate vegetatively. Ecological redundancy is the

extent to which a forest ecosystem structure, process, or function is substitutable if a

degradation or loss in the main species that provides that particular attribute occurs (Folke et

al., 2004). A system having greater quantities of a proxy is thought to be more resilient

(Loreau et al., 2003; Fischer et al., 2006). Response diversity and ecological redundancy are

deemed particularly important as multiple species performing the same critical function can

replace or compensate for substantial losses in a dominant species, as well as display

variation in responses to disturbance or gradual change (Thompson et al., 2009).



An abundance of research shows that the chances of shifting into another stability domain

increase when removal, reduction, or drastic changes to any of these proxies occurs (e.g.,

Naeem et al., 1995; Loreau et al., 2003; Contamin and Ellison, 2009). Larger impacts on

critical ecosystem processes are typically observed when there are fewer species present,

when the dominant or keystone species are strongly affected, or when functional redundancy

is low (Pastor et al., 1996; Lavorel et al., 2007; Rinawati et al., 2013). Thus, greater species

diversity may confer greater ecological resilience (Hooper et al., 2005; Fischer et al., 2006).

Yet this may not always be the case (e.g., Petchey and Gaston, 2009). Some boreal systems

with relatively low species diversity levels are also resilient. For example, black spruce

(Picea mariana) and balsam fir (Abies balsamea) forest ecosystems both have low functional

diversity and redundancy, yet are both highly resilient to catastrophic fire and insect

disturbance, respectively (Pollock and Payette, 2010; Boiffin and Munson, 2013).

Black spruce and balsam fir trees are well adapted to these severe disturbances and have

a broad genetic diversity that can tolerate a wide range of habitat conditions (Thompson et al.,

2009). Thus, while high levels of diversity may not be expressed at the species or community

levels of organization, at the genetic level, these species have the necessary components for

renewal and reorganization. However, questions remain as to how these ecosystems will

respond to climate change. Balsam fir, for example, regenerates poorly after fire (Asselin et

al., 2001) while jack pine (Pinus banksiana) regenerates poorly in its absence (Parisien et al.,

2004). Boiffin and Munson (2013) observed shifts in species dominance from black spruce to

jack pine after a period of unusually high fire activity that caused changes in microhabitat

suitability for germination. Large scale changes to species distribution patterns will likely

occur across the landscape if these periods of large fire years become more frequent. Other

concomitant effects of climate change are also of concern. Changes to habitat suitability for a

number of spruce beetle species (Dendroctonus spp.) along the west coast of North America

have expanded the potential for their impacts in both altitude and latitude (Bentz et al., 2010)

for example.

So how much diversity is enough to maintain resilience? Clearly there is still much to be

resolved with this aspect of ecological resilience. It is difficult to ascertain the quantity of a

proxy required for stability or which proxy is most important for any particular forest

management unit given that differences in local physiographic attributes, disturbance regimes,

and the spatial or temporal scale of measurement can change expected contributions (Loreau

et al., 2002; Lavorel et al., 2007). Further, knowledge of the functional roles of many species

remains incomplete (Grime, 1998; Scherer-Lorenzen et al., 2005), and thus it can be difficult

to judge the adequacy of response diversity or ecological redundancy.

Management is facilitated by clear objectives and by concrete numbers that support and

validate them; and ecological resilience theory, at this stage of its conceptual development

cannot provide them. In the ball and cup model of Figure 1, this equates to determining

exactly how close to a threshold edge an ecosystem‟s current state is, how quickly it can

tumble towards it, and how much a proxy can keep it from drawing nearer or can drive it

away from collapse. Modeling that projects changes in critical processes into the future is

only beginning (e.g., Hirota et al., 2011; Gustafson, 2013; Lafond et al., 2013) so detecting or

predicting critical changes such as shifts between stable ecosystem states is still problematic.

Thus, it remains an enormous task to shift knowledge of the adequacy of ecological resilience

from hindsight to a useful predictive tool as we still don‟t know where thresholds are until

after they are crossed.



But do we really need to know exactly where thresholds lie or just the impacts of the

processes that lead to them? Shouldn‟t we be able to identify signals of tumbling towards

shifts in forest ecosystem states, and use these signals as qualitative indicators of the risks of

surpassing undesirable threshold points? As it is, we are not even able to effectively identify

key species or their response functions (e.g., be it whole plant, stem and below ground, or

regeneration functional traits) (Grime 1998, Scherer-Lorenzen et al., 2005; Lavorel et al.,

2007). Thus, the precautionary principle; n.b., Leopold‟s (1949) argument that the intelligent

rule to tinkering is not to get rid of any of the pieces, suggests that all species should be

maintained. Moreover, a recent synthesis (Cardinale et al., 2012) suggests that increased

diversity also begets increased ecosystem productivity, which implies that a call for

maintaining biodiversity does not need to be based on altruism or ethical considerations but

may be for our own best interest.

Perhaps another issue is that we‟re expecting that managing for ecological resilience (or

any other management option) should account for everything a priori. Questions arise such

as: is management that promotes maintaining processes such as disturbance regimes within

natural historical ranges of variation even useful if the resultant patterns and relationships are

expected to change or decouple altogether with global change? How do we know if oncoming

novel disturbance types and/or disturbance interactions will be beyond what our forest

management unit can absorb? These are questions that perhaps no amount of management or

management approach can truly account for a priori. This may also require us to accept that

domain shifts will occur as conditions exceed the adaptations of local species. For example, if

conditions become too xeric for moisture sensitive species such as balsam fir. The ability to

adapt human institutions that depend on natural ecosystems will thus be tantamount to socio-

ecological resilience as the ecosystems themselves re-organize.

3. PUTTING IT ALL TOGETHER: THE WAY FORWARD

Ecological resilience may eventually be an important management option. But at its

current conceptual iteration, there are too many details that require development or resolution

prior to it being used as a general operational tool. In particular, the lack of knowledge of a

number of critical processes and how they function and interact across spatio-temporal scales,

the uncertainty associated with relationships between resilience and the quantity of

biodiversity needed to maintain stability, as well as the lack of quantitative approaches to

determine an ecosystem‟s position in state space relative to threshold points need addressing.

Despite this, ecological resilience can and does have an indirect role in managing our forests;

as there are several elements of ecological resilience already being used in contemporary

forest management paradigms. Objectives such as maintaining native biodiversity and

improving knowledge of processes that impact ecosystem functioning, for example, are

consistent with the guiding principles of Ecosystem Management, Emulating Natural

Disturbance, and Managing for Complexity (Holling, 1978; Grumbine, 1994; Perera et al.,

2007; Gauthier et al., 2008).



Figure 3. Conceptual diagram of the balance of social, economic, and environmental objectives under

Ecosystem Management linked across scales. We cannot manage for everything everywhere. In our

example here, the ball represents an objective. In (a) an economic objective takes precedence but overall

effects are within the adaptive capacity of the ecosystem at the microsite scale; (b) priority on social

measures is concomitantly beneficial to conservation but negatively impacts economic interests at the

stand scale. But since there are many stands across the landscape, stand-level impacts of this objective

will be balanced out by applying more intensive forestry on some sites and other social or conservation

measures on others; (c) shows multiple objectives across the region, each having a specific focus, but

impacts on other objectives are always considered. Ecosystem function is maintained by processes that

can interact and affect the ecosystem at one or a number of scales. Elimination of an objective occurs

when disturbance or slow change drives species responsible for providing objective beyond the „tipping

point‟ (i.e., threshold limit of resilience) at the regional scale (indicated by the dashed arrow).

Restoration is now required to re-introduce source or basis of objective. Ecosystem collapse can occur

when detrimental impacts of an objective are beyond the adaptive capacity of the groups of species

responsible for regulation of key ecosystem processes driving the system. Link between scales is

dependent on the connectivity and pattern of forest patches across scales and the processes controlling

them.

An approach based on Ecosystem Management (Grumbine, 1994), one that integrates

various aspects of other contemporary paradigms at multiple scales of focus will help

minimize risk of changing stability domains as well as maintain processes and attributes

identified when asking „from what to what‟. Managing forest resources so that processes

remain within historic natural ranges of variability are stressed, but stakeholders should be

flexible enough to adapt strategies as more information becomes available. As in the TRIAD

approach to forestry management (Seymour and Hunter, 1992), the forest management unit

could be partitioned into zones where either social, economic, or conservation objectives are

emphasized, the proportion of which are pre-determined by stakeholder agreement, and this



pattern repeated across the landscape at various spatio-temporal scales (Figure 3). At each

scale of focus, an adaptive strategy is used. This is an iterative approach wherein the effects

of management policy and stakeholder actions are periodically evaluated and modified as

necessary; essentially, as outcomes from natural events and management actions become

better understood (Holling, 1978). Thus, it is a multidisciplinary, dynamic, and multi-scalar

approach to Ecosystem Management based on processes responsible for natural historic

ranges of variation. It emphasizes frequent communication, research, and information

exchange among stakeholders. The use and modification of procedures derived from

continuously updated knowledge of ecosystem dynamics is the underlying premise for

stakeholder exchanges.

Consistent with the requirements for building ecological resilience, this strategy

recognizes the importance of a range of variability in natural processes in contributing to

forest ecosystem functioning. The strength of the approach would be in the ability to identify

changes in conditions created by anthropogenic disturbances from multiple viewpoints and at

multiple scales. Moreover, complexity and variation of forest ecosystems are emphasized

rather than avoided while modeling and forecasting could incorporate spatial structure and

processes, in addition to traditional modeling parameters (Baskent and Yolasiǧmaz, 2000). So

rather than focusing on attaining a single optimal ecosystem condition, a range of acceptable

outcomes is managed for, and thus, potentially reducing vulnerability to unforeseen

disturbance and gradual change across the entire forest management unit; n.b., similar to the

ball and cup metaphor, this is analogous to having a number of balls moving around in the

desired ecosystem state space at the same time (Figure 3).

CONCLUSION

Operationalising ecological resilience is an admirable goal. But at this stage of its

conceptual development, its use in management planning is limited. Instead it is perhaps best

used as a monitoring tool to evaluate the success of other strategies (e.g., TRIAD, Ecosystem

Management, Emulating Natural Disturbances). Until our understanding of critical processes

and ability to predict shifts in ecological states improves, current management approaches

that draw attention to the processes driving ecosystem dynamics across spatio-temporal

scales, as well as linking these processes with societal uses and values should be emphasized.

ACKNOWLEDGMENTS

Lively discussions and feedback from a number of individuals, including H. Archibald,

H. Chen, B. Freedman, T. Gooding, B. Harvey, K. Hylander, T. Jain, M. Kennedy, H.

Kimmins, N. Klenk, D. Kreutzweiser, L. Leal, C. Messier, A. Miller, A. Mosseler, A. Park,

K. Peterson, K. Puettmann, M. Willison, L. Van Damme, S. Woodley, and R. Tittler were

important in helping to develop ideas and clarify concepts presented here.



REFERENCES

Allen, C.R., and C. Holling 2010. Novelty, adaptive capacity, and resilience. Ecology and

Society 15(3): 24. [online] URL: http://www.ecologyandsociety.org/vol15/iss3/art24/

Asselin, H., Fortin, M.-J., and Y. Bergeron. 2001. Spatial distribution of late-successional

coniferous species regeneration following disturbance in southwestern Quebec boreal

forest. Forest Ecology and Management 140: 29-37.

Başkent, E.Z., and H.A. Yolasiǧmaz. 2000. Exploring the concept of a forest landscape

management paradigm. Turkish Journal of Agriculture and Forestry 24: 443-451.

Beckage, B., and C. Ellingwood. 2008. Fire feedbacks with vegetation and alternate stable

states. Complex Systems 18: 159-172.

Bentz, B.J., Régnière, J., Fettig, C.J., Hansen, E.M., Hayes, J.L., Hicke, J.A., Kelsey, R.G.,

Negron, J.F., and S.J. Seybold. 2010. Climate change and bark beetles of the western

United States and Canada: direct and indirect effects. BioScience 60: 602–613.

Berggren, A., Björkman, C., Bylund, H., and M.P. Ayres. 2009. The distribution and

abundance of animal populations in a climate of uncertainty. Oikos 118: 1121-1126.

Boiffin, J., and A.D. Munson. 2013. Three large fire years threaten resilience of closed crown

black spruce forests in eastern Canada. Ecosphere 4(5):56, 20 p. http://dx.doi.org/

10.1890/ES13-00038.1

Bonan, G.B. 2008. Forests and climate change: forcings, feedbacks, and the climate benefits

of forests. Science 320: 1444-1449.

Campbell, E.M., Saunders, S.C., Coates, K.D., Meidinger, D.V., MacKinnon, A.J., O‟Neill,

G.A., MacKillop, D.J., and S.C. DeLong. 2009. Ecological resilience and complexity: a

theoretical framework for understanding and managing British Columbia‟s forest

ecosystems in a changing climate. Technical Report 055. BC Ministry of Forests and

Range Forest Science Program, Victoria, BC, 36 p.

Campbell, J.L., Rustad, L.E., Boyer, E.W., Christopher, S.F., Driscoll, C.T., Fernandez, I.J.,

Groffman, P.M., Houle, D., Kiekbusch, J., Magill, A.H., Mitchell, M.J., and S.V.

Ollinger. 2009b. Consequences of climate change for biogeochemical cycling in forests

of northeastern North America. Canadian Journal of Forest Research 39(2): 264–284.

Cardinale, B.J., Duffy, J.E., Gonzalez, A., Hooper, D.U., Perrings, C., Venail, P., Narwani,

A., Mace, G.M., Tilman, D., Wardle, D.A., Kinzig, A.P., Daily, G.C., Loreau, M., Grace,

J.B., Larigauderie, A., Srivastava, D.S., and S. Naeem. 2012. Biodiversity loss and its

impact on humanity. Nature 486: 59-67.

Carpenter, S.R., Walker, B., Anderies, J.M., and N. Abel. 2001. From metaphor to

measurement: resilience of what to what? Ecosystems 4: 765-781.

Clements, F.E. 1928. Plant Succession and Indicators: a definitive edition of plant succession

and plant indicators. The H.W. Wilson Co., New York, USA.

Contamin, R., and A.M. Ellison. 2009. Indicators of regime shifts in ecological systems: what

do we need to know and when do we need to know it? Ecological Applications 19(3):

799-816.

de Calesta, D.S. 1994. Effects of white-tailed deer on songbirds within managed forests in

Pennsylvania. Journal of Wildlife Management 58: 711–718.

Dukes, J.S., Pontius, J., Orwig, D., Garnas, J.R., Rodgers, V.L., Brazee, N., Cooke, B.,

Theoharides, K.A., Stange, E.E., Harrington, R., Ehrenfeld, J., Gurevitch, J., Lerdau, M.,



Stinson, K., Wick, R., and M. Ayres. 2009. Responses of insect pests, pathogens, and

invasive plant species to climate change in the forests of northeastern North America:

what can we predict? Canadian Journal of Forest Research 39: 231-248.

Elmqvist, T., Folke, C., Nyström, M., Peterson, G., Bengtsson, J., Walker, B., and J. Norberg.

2003. Response diversity, ecosystem change, and resilience. Frontiers in Ecology and the

Environment 1(9):488-494.

Fettig, C.J., Reid, M.L., Bentz, B., Sevanto, S., Spittlehouse, D.L., and T. Wang. 2013.

Changing climates, changing forests: a western North American perspective. Journal of

Forestry 111(3):214–228. http://dx.doi.org/10.5849/jof.12-085

Fischer, J., Lindenmayer, D.B., and A.D. Manning. 2006. Biodiversity, ecosystem function,

and resilience: ten guiding principles for commodity production landscapes Frontiers in

Ecology and the Environment 4(2): 80-86.

Folke, C., Carpenter, S., Walker, B., Scheffer, M., Elmqvist, T., Gunderson, L., and C.S.

Holling. 2004. Regime shifts, resilience and biodiversity in ecosystem management.

Annual Review in Ecology, Evolution, and Systematics 35: 557–58.

Frelich, L., E., and P.B. Reich. 1998. Disturbance severity and threshold responses in the

boreal forest. Conservation Ecology [online] 2(2):7. http://www.consecol.org./

vol2/iss2/art7

Frelich, L.E., and P.B. Reich 1999. Neighbourhood effects, disturbance severity, and

community stability in forests. Ecosystems 2: 151-166.

Gauthier, S., Vaillancourt, M.-A., Leduc, A., De Grandpré, L., Kneeshaw, D., Morin, H.,

Drapeau, P., and Y. Bergeron (eds.). 2008. Aménagement Écosystemique en Forêt

Boréale. University of Quebec Press, Quebec, Canada, 600 p.

Gill, R.M.A., and V. Beardall. 2001. The impact of deer on woodlands: the effects of

browsing and seed dispersal on vegetation structure and composition. Forestry 74: 209–

218.

Grime, J.P. 1998. Benefits of plant diversity to ecosystems: immediate, filter, and founder

effects. Journal of Ecology 86(6): 902-910.

Groffman, P.M., Baron, J.S., Blett, T., Gold, A.J., Goodman, I., Gunderson, L.H., Levinson,

B.M., Palmer, M.A., Paerl, H.W., Peterson, G.D., Poff, N.L., Rejeski, D.W., Reynolds,

J.F., Turner, M.G., Weathers, K.C., and J. Weins. 2006. Ecological thresholds: the key to

successful environmental management or an important concept with no practical

application? Ecosystems 9: 1-13.

Grumbine, R.E.1994. What is ecosystem management? Conservation Biology 8(1): 27-38.

Gustafson, E.J. 2013. When relationships estimated in the past cannot be used to predict the

future: using mechanistic models to predict landscape ecological dynamics in a changing

world. Landscape Ecology 28:1429–1437.

Gunderson, L.H. 2000. Ecological resilience – in theory and application. Annual Review of

Ecology and Systematics 31: 425-439.

Gunderson, L.H., and C.S. Holling. 2002. Panarchy: Understanding Transformations in

Human and Natural Systems. Island Press, Washington, DC, USA, 507 p.

Heinselman, M.L. 1981. Fire and succession in the conifer forests of northern North America

In Forest succession: concepts and application Edited by D.C. West, H.H. Shugart, and

D.B. Botkin, Springer-Verlag, New York, USA, pp. 374-406.



Herzog, S.K., and M. Kessler. 2006. Local vs. regional control on species richness: a new

approach to test for competitive exclusion at the community level. Global Ecology and

Biogeography 15: 163–172.

Hirota, M., Holmgren, M., van Nes, E.H., and M. Scheffer. 2011. Global resilience of tropical

forest and savanna to critical transitions. Science 334: 232–235.

Hogg, E.H., and P.A. Hurdle. 1995. The aspen parkland in western Canada: a dry-climate

analogue for the future boreal forest? Water, Air and Soil Pollution 82: 391-400.

Holling, C.S. 1973. Resilience and stability of ecological systems. Annual Review of Ecology

and Evolution S4: 2-23.

Holling, C. S. 1978. Adaptive environmental assessment and management. John Wiley,

London, 377 p.

Holling, C.S., and G.K. Meffe. 1996. Command and control and the pathology of natural

resource management. Conservation Biology 10:328-337.

Hooper, D.U., Chapin III, F.S., Ewel, J.J., Hector, A., Inchausti, P., Lavorel, S., Lawton, J.H.,

Lodge, D.M., Loreau, M., Naeem, S., Schmid, B., Setälä, H., Symstad, A.J., Vandermeer,

J., and D.A. Wardle. 2005. Effects of biodiversity on ecosystem functioning: a consensus

of current knowledge. Ecological Monographs 75(1): 3-35.

Hunter, M.L. 1999. Maintaining biodiversity in forest ecosystems. Cambridge University

Press, Cambridge, UK, 698 p.

Huntington, T.G., Richardson, A.D., McGuire, K.J., and K. Hayhoe. 2009. Climate and

hydrological changes in the northeastern United States: recent trends and implications for

forested and aquatic ecosystems. Canadian Journal of Forest Research 39(2): 199–212.

Johnstone, J., Chapin III, F.S., Hollingsworth, T.N., Mack, M.C., Romanovsky, V., and M.

Turetsky. 2010. Fire, climate change, and forest resilience in interior Alaska. Canadian

Journal of Forest Research 40: 1302-1312.

Kilpeläinen, A., Kellomäki, S., Strandman, H., and A. Venäläinen. 2010. Climate change

impacts on forest fire potential in boreal conditions in Finland. Climatic Change 103(3):

383-398 DOI 10.1007/s10584-009-9788-7

Klenk, N.L., G.Q. Bull and J.I. MacLellan. 2009. The “emulation ofnatural disturbance”

(END) management approach in Canadian forestry: A critical evaluation. The Forestry

Chronicle 85(3): 440–445.

Lafond, V., Lagarrigues, G., Cordonnier, T., and B. Courbaud. 2013. Uneven-aged

management options to promote forest resilience for climate change adaptation: effects of

group selection and harvesting intensity. Annals of Forest Science [online first] DOI

10.1007/s13595-013-0291-y, 14 p.

Lavoie, M., Paré, D., Fenton, N., Groot, A., and K. Taylor. 2005. Paludification and

management of forested peatlands in Canada: a literature review. Environmental Reviews

13(2): 21-50.

Lavorel, S., Díaz, S., Cornelissen, J.H.C., Garnier, E., Harrison, S.P., McIntyre, S., Pausas,

J.G., Pérez-Harguindeguy, N., Roumet, C., and C. Urcelay. 2007. Chapter 13. Plant

Functional Types: are we getting any closer to the holy grail? In Terrestrial Ecosystems

in a Changing World, IGBP Series Edited by J.G. Canadell, D. Pataki, and L. Pitelka,

Springer-Verlag, Berlin, Germany, pp 149-160.

Leblanc, M., and B. Pouliot. 2011. La coupe avec protection de la régénération et des sols

avec rétention de bouquets - Fondements et exécution opérationnelle, Ministère des

Ressources Naturelles et de la Faune, Québec, Canada, 9 p.



Leopold, A. 1949. A Sand County Almanac: and sketches here and there. Oxford University

Press, UK, 240 p.

Loreau, M., Naeem, S., and P. Inchausti. 2002. Biodiversity and Ecosystem Functioning:

synthesis and perspectives, Oxford University Press, New York, USA, 294 p.

Loreau, M., Mouquet, N., and A. Gonzalez. 2003. Biodiversity as spatial insurance in

heterogenous landscapes. Proceedings of the National Academy of Sciences of the United

States of America 100(22): 12765-12770.

Mallik, A.U. 1995. Conversion f temperate forests into heaths: role of ecosystem disturbance

and ericaceous plants. Environmental Management 19(5): 675-684.

Mladenov, N., Pulido-Villena, E., Morales-Baquero, R., Ortega-Retuerta, E., Sommaruga, R.

and I. Reche. 2008. Spatiotemporal drivers of dissolved organic matter in high alpine

lakes: Role of Saharan dust inputs and bacterial activity. Journal of Geophysical

Research 113(G2): 9 p.G00D01, DOI: 10.1029/2008JG000699

Morissette, È.-M., Lavoie, C., and J. Huot. 2009. Fairy slipper (Calypso bulbosa) on Anticosti

Island : the occurrence of a rare plant in an environment strongly modified by white-

tailed deer. Botany 87: 1223-1231.

Naeem, S., Thomson, L.J., Lawlor, S.P., Lawton, J.H., and R.M. Woodfin. 1995. Empirical

evidence that declining species diversity may alter the performance of terrestrial

ecosystems. Philosophical Transactions: Biological Sciences 347(1321): 249-262.

Nitschke, C.R. 2005. Does harvesting emulate fire disturbance? A comparison of effects on

selected attributes in coniferous-dominated headwater systems. Forest Ecology and

Management 214: 305-319.

Noble, I.R., and R.O. Slatyer. 1980. The use of vital attributes to predict successional changes

in plant communities subject to recurrent disturbances. Vegetatio 43: 5-21.

Parisien, M.-A., Sirois, L., and M. Babeau. 2004. Distribution and dynamics of jack pine at its

longitudinal range limits in Quebec In Proceedings of the 22nd Tall Timbers Fire

Ecology Conference: fire in temperate, boreal and montane ecosystems, October 15-18,

2001, Kananaskis, Alberta Edited by R.T. Engstrom, K.E.M. Galley, and W.J. de Groot,

Tall Timbers Research Station, Tallahassee, Florida, USA, pp 247-257.

Pastor, J. Mladenoff, D.J., Hyla, Y., Bryant, J., and S. Payette. 1996. Biodiversity and

ecosystem processes in boreal regions In Functional Roles of Biodiversity: A Global

Perspective Edited by H.A. Mooney, J.H. Cushman, E. Medina, O.E. Sala, and E.-D.

Schulze, John Wiley and Sons, New York, NY, USA, pp 33-69.

Payette, S., and A. Delwaide. 2003. Shift to conifer boreal forest to lichen-heath parkland

caused by successive stand disturbances. Ecosystems 6: 540-550.

Perera, A.H., Buse, L.J., Weber, M.G., and T.R. Crow. 2007. Emulating natural forest

landscape disturbances – a synthesis In Emulating Natural Forest Landscape

Disturbances: Concepts and Applications Edited by A.H. Perera, L.J. Buse, and M.G.

Weber, Columbia University Press, NY, USA, pp 265-274.

Petchey, O.L., and K.G. Gaston. 2009. Effects on ecosystem resilience of biodiversity,

extinctions, and the structure of regional species pools. Theoretical Ecology 2: 177-187.

Peterson, G.D. 2000. Scaling ecological dynamics: self-organisation, hierarchical structure,

and ecological resilience. Climatic Change 44: 291-309.

Peterson, G., Allen, C.R., and C.S. Holling. 1998. Ecological resilience, biodiversity, and

scale. Ecosystems 1: 6-18.



Pollock, S. L., and S. Payette. 2010. Stability in the patterns of long-term development and

growth of the Canadian spruce-moss forest. Journal of Biogeography 37:1684–1697.

Potvin, F., Beaupré, P., and G. Laprise. 2003. The eradication of balsam fir stands by white-

tailed deer on Anticosti Island, Québec: a 150-year process. Écoscience 10(4): 487-495.

Read, D.J., Leake, J.R., and J. Perez-Moreno. 2004. Mycorrhizal fungi as drivers of

ecosystem processes in heathland and boreal forest biomes. Canadian Journal of Botany

82: 1243-1263.

Ricklefs, R.E. 1987. Community diversity: relative roles of local and regional processes.

Science 235: 167-171.

Rinawati, F., Stein, K., and A. Lindner. 2013. Climate change impacts on biodiversity – the

setting of a lingering global crisis. Diversity 5: 114-123.

Rooney, T.P., and D.M. Waller. 2003. Direct and indirect effects of white-tailed deer in forest

ecosystems. Forest Ecology and Management 181: 165-176.

Salonius, P. 2007. Silvicultural discipline to maintain Acadian forest resilience. Northern

Journal of Applied Forestry 24(2): 91-97.

Scherer-Lorenzen, M., Körner, C., and E.,-D. Schulze. 2005. Forest Diversity and Function:

temperate and boreal systems. Ecological Studies 176, Springer-Verlag Inc., New York,

NY, 400 p.

Seppä, H., Alenius, T., Bradshaw, R.H.W., Giesecke, T., Heikkilä, M., and P. Muukkkonen.

2009. Invasion of Norway spruce (Picea abies) and the rise of the boreal ecosystem in

Fennoscandia. Journal of Ecology 97: 629-640.

Seymour, R.S., and M.L. Hunter Jr. 1992. New forestry in eastern spruce-fir forests:

principles and applications in Maine. Maine Agricultural and Forest Experiment Station

Miscellaneous Publication 716, 36 p.

Sokolov, A., Stone, P., Forest, C., Prinn, R., Sarofim, M., Webster, M., Paltsev, S., Schlosser,

C.A., Kicklighter, D., Dutkiewicz, S., Reilly, J., Wang, C., Felzer, B., Melillo, J., and H.

Jacoby. 2009. Probabilistic forecast for 21st century climate based on uncertainties in

emissions (without policy) and climate parameters. Journal of Climate 22(19): 5175-

5204.

Strand, E.K., Vierling, L.A., Bunting, S.C., and P.E. Gessler. 2009. Quantifying successional

rates in western aspen woodlands: current conditions, future predictions. Forest Ecology

and Management 257: 1705-1715.

Taylor, A.R., Hart, T., and H.Y.H. Chen. 2013. Tree community structural development in

young boreal forests: a comparison of fire and harvesting disturbance. Forest Ecology

and Management 310: 19-26.

Thompson, I., Mackey, B., McNulty, S., and A. Mosseler. 2009. Forest Resilience,

Biodiversity, and Climate Change: a synthesis of the biodiversity/resilience/stability

relationship in forest ecosystems. Technical Series No. 43, Secretariat of the Convention

on Biological Diversity, Montreal, QC, 67 p

Turner, M.G., Gardner, R.H., and R.V. O‟Neil. 2001. Landscape Ecology in Theory and

Practice. Springer-Verlag Inc., New York, NY, 401 p.

Voigt, D.R., Broadfoot, J.D., and J.A. Baker. 1997. Forest management guidelines for the

provision of white-tailed deer habitat. Queen‟s Printer for Ontario, Ontario, Canada, 33 p.

Watt, R.W., Baker, J.A., and D.M. Hogg. 1996. Forest management guidelines for the

provision of marten habitat. Queen‟s Printer for Ontario, Ontario, Canada, 24 p.



Walker, B., Holling, C.S., Carpenter, S.R., and A. Kinzig. 2004. Resilience, adaptability and

transformability in social–ecological systems. Ecology and Society 9 (2): 5.

[online] URL: http://www.ecologyandsociety.org/vol9/iss2/art5/




Chapter 7

CLIMATE CHANGE AND AN AUSTRALIAN

RAINFOREST CONIFER

Rohan Mellick National Herbarium of NSW, The Royal Botanic Gardens and Domain Trust,

Sydney, NSW, Australia

ABSTRACT

Revealing the evolutionary trends of the recent past (late Quaternary) will allow

climate change science to anticipate the demographic response of species to future

disturbance. The genetic disjunctions and distribution of a long-lived rainforest conifer

provide a valuable signature of past demographic response to climate change – a

biological autograph of time, climatic cycles and the environment. This chapter reviews

literature pertaining to three case studies. Case study one determines the level of genetic

diversity and structure within naturally occurring populations of Podocarpus elatus and

resolves the influence of historic and contemporary drivers on divergence. Case study

two explores the impact of glacial cycle climatic changes on the palaeodistribution of the

species: by combining population genetic analysis, coalescent-based analysis, the

observed fossil record and environmental niche modeling. Case study three hypothesizes

that future adaptive potential is indicative of genetic/demographic change and range

expansion/contraction trends in Podocarpus elatus associated with the Clarence River

Corridor. The final section of the chapter concludes the case studies and suggests

conservation strategies applicable to long-lived species allied with threatened

communities.

1. INTRODUCTION

The Podocarpaceae are an ancient coniferous family harboring traits that evolved with

extreme seasonality in high latitudes epochs ago (Brodribb and Hill, 1999, 2003; Quinn and

Price, 2003; Hill, 1994). These characters have proven resilient and have persisted despite

mass extinction events and many catastrophes. Although conifers are currently only a small

proportion of the total arboreal diversity on the planet, they once dominated the landscape


Rohan Mellick 214

with tall relatively closed forest. A high extinction rate has resulted in a rich fossil legacy and

record of biological response to environmental change.

Over time, the lack of landmass in the Southern Hemisphere and the inability to migrate

in response to pressure may have resulted in varied adaptive response and past diversification.

Many atypical coniferous taxa endure despite the vast majority having become extinct. In

particular, broad-leaved podocarps are incredibly diverse and are well represented in the

Gondwanan fossil record for over 144 million years (My) (Hill, 1994; Quilty, 1994).

These conifers are a remnant of a world dramatically different from the world we live in

today. In high latitudes, rainforests were then extremely seasonal and trees evolved to sub-

annual periods of light and darkness (McLoughlin, 2001). Broad-evergreen foliage needed to

be tough and versatile to endure months of relentless cold and darkness.

The browsers and dispersers that evolved with Podocarpus in Australia (~65 million

years ago; Ma) are long extinct (Bartholomai and Molnar, 1981; Molnar, 1996), but elements

of this ancient environment remain. Dacrydium (Podocarpaceae) and associated swamp forest

in Australian Wet Tropics is believed to have gone extinct only six thousand years ago (ka)

(Kershaw et al., 2007a). Currently, Lagarostrobos franklinii (Podocarpaceae) forests extend

through South East Tasmania (the first Australasian landmass north of Antarctica), and of

coarse another southern hemisphere conifer Wollemia nobilis (Araucariaceae) was recently

rediscovered and thought to be extinct for 65 My - one population currently defies extinction

and persists 129km northwest of Sydney.

The Southern Hemisphere is dry due to a number of causes, such as northern tectonic

drift and the establishment of the Antarctic circumpolar current about 18 Ma (Barker and

Thomas, 2004). This has dramatically limited the fossil record in Australia due to the lack of

deposition, especially for the Quaternary (1.8 Ma to present). Interdisciplinarity may support

the current understanding of historical community structure.

Here we study a long-lived rainforest conifer, Podocarpus elatus (R.Br. ex Endl.). The

species ranges over 20 degrees of latitude, east of the Great Dividing Range (Eastern

Australia). The principle aim being to understand the effect of late Quaternary and future

climate change on the distribution, diversity and genetic disjunctions of a long-lived conifer

restricted to rainforest. Slow morphological adaptation (evolutionary senescence), restriction

to naturally fragmented mesic communities, and broad distribution makes P. elatus a valuable

tool to understand changes in community structure according to climate.

Understanding the effect of Quaternary climate change on the distribution, diversity and

divergence of Podocarpus elatus will contribute to the conservation of the east Australian

rainforests in light of increasing ecological damage, rapid human population growth and

anthropogenic-induced global warming. To know how intraspecific diversification in a wide-

ranging long-lived species is impacted by climate change will improve our understanding of

how climatic-drivers are involved in the evolutionary process.

This chapter discusses the history of climate change in Australia, the family

Podocarpaceae, vegetation change during the Cenozoic, event histories, the East Australian

rainforest, palynology, marker development, domestication, cytoplasmic inheritance,

coalescent-based analyses environmental niche modeling, interdisciplinarity, general points of

interest and the following published case studies.

Case study one determines the level of genetic diversity and structure within naturally

occurring populations of Podocarpus elatus and resolves the influence of historic and

contemporary drivers on divergence.


Climate Change and an Australian Rainforest Conifer 215

The objectives were to:

1 Determine genetic structure and how it corresponds to geographic and/or

environmental patterns.

2 Evaluate landscape-level habitat fragmentation and if there is loss of gene flow

among populations.

3 Identify genetic diversity within populations to see if there are measurable

differences across demographic age cohorts.

Case study two explores the impact of glacial cycle climatic changes on the

palaeodistribution of the species: by combining population genetic analysis, Bayesian

coalescence-based analysis, the observed fossil record and Environmental Niche Modeling

(ENM).

We endeavored to answer these questions:

1 Are the distributional changes predicted by ENM in agreement with the available

fossil record?

2 Are ENM and population genetic-based estimates of range contractions/expansions

and disjunctions in agreement?

3 In the context of the broad latitudinal and climatic distribution of the species, are the

northern and southern ranges likely to have experienced similar

expansion/contraction dynamics?

Case study three hypothesizes that genetic/demographic change and range

expansion/contraction trends in Podocarpus elatus associated with the Clarence River

Corridor are indicative of the future distribution and degree of adaptive potential in this long-

lived species. Future conservation strategies may be more successful if distributional and

demographic inference based upon reliable observational data is made.

Also we aimed to answer:

1 What are the coalescent-based estimates of ancestral demographic patterns and

divergence times in Podocarpus elatus and how do these relate to climate change?

2 What is the predicted distribution of P. elatus for the IPCC 4th Assessment Report

climatic estimates of 2050?

3 Based on these findings, what are the appropriate conservation and management

strategies for P. elatus?

The final section of the chapter will conclude the case studies and suggest conservation

strategies applicable to threatened species and habitats.

2. THE AUSTRALIAN VEGETATION

The Australian vegetation today is the consequence of a dynamic history of climate

change, latitudinal change, continental isolation, interaction with an evolving fauna, fire and


Rohan Mellick 216

more recently anthropogenic affects (Hill, 1994). Community turnover (including

distributional change) between drier rainforest types, containing Araucaria and Podocarpus

species, and wetter types containing Nothofagus species have been successional in response

to recent glacial cycles along the east coast of Australia (Mellick et al. 2013a, Kershaw et al.

1994; Hill, 1994).

The Quaternary (1.8 Ma to present) is a period of particular importance due to extant

plants and animals coming to establish broader communities and to reside within their

environmental limits of tolerance (environmental niche). The fossil record indicates some

species, including Podocarpus, have a similar ecology, co-occur with palaeo-counterparts and

rely on mesic habitats such as they did since and during the Tertiary (65 - 0 Ma), largely

unaffected by the changing environment (Brodribb and Hill, 2003) – that is not to say changes

in distribution have not facilitated their survival and trait preservation.

Throughout the Quaternary in the Australian Wet Tropics (AWT), rainforest was the

dominant vegetation type with large tracts of dry rainforest fringing areas of complex

rainforest in wetter areas (Kershaw et al., 2005) similar in type to present-day dry rainforest

communities typified by Araucarian emergents. Information from the fossil record indicates

these dry Araucarian forests were the major habitat type of Podocarpus species. These

transitional communities fringed floristically complex rainforest core areas, and where

reduced angiosperm competition allowed for slower growing rainforest gymnosperms to

establish, and in some circumstances thrive.

Australian vegetation is very diverse and reflective of the heterogeneous landscape and

diversity of ecosystems that occur across the continent. This diversity is primarily due to the

huge variation in rainfall across the landscape which over time has changed considerably and

promoted rapid adaptation and extinction of a variety of plant taxa.

Increasing aridity during the Neogene (33.7–1.8 Ma) into the Quaternary is responsible

for the transformation of the Australian landscape from one dominated by broad-leaf

rainforest communities to one composed predominantly of open vegetation communities

(sclerophyllous-type) with wetter rainforest restricted to present day locally moist areas

(Kershaw et al., 1994; Hill, 2004; Martin, 2006). Detailed information of the Australian

climate throughout the Cretaceous and Cenozoic (145.5 Ma to present) has emerged over the

last few decades but there is limited data to document short term climatic cycles (Kershaw et

al., 2007a), particularly for the Quaternary from eastern Australia where fossil evidence is

sporadic and poorly dated (Quilty, 1994).

The change in distribution of major forest types provides a good assessment of abiotic

change (e.g., climate, soil and fire). Biotic conditions (e.g., competition, pathogens,

symbionts and people) also influence the expansion/contraction of plant communities. A plant

taxon requires a suitable abiotic environment for growth yet biotic interactions are varied.

Adaptability governs response to change so migration out of a residing area can be viewed as

an inability to adapt to that area over time.

Past climatic change is inferred from evidence incorporated into deposits and especially

from contained fossils (Hill, 1994; Quilty, 1994). If palaeoclimate is reconstructed solely on

that indicated by fossil plants, it is susceptible to false interpretation. It is ideal to reach the

same conclusion from a number of sources of evidence that are independent. Australia is

deprived of palaeoclimatic data sources, especially ice cores that provide unique archives of

past climate and environmental changes. Environmental niche modeling, used in this study, is



reliant on known global fluctuations of climate, which for the large part has been sourced

from Antarctic ice cores.

Australia is a continent where continuous onshore sedimentary records, particularly of

those deposits that accumulated in a non-marine environment, are uncommon, and research

often depends on the offshore marine sequences, themselves incomplete, around the continent

margins, supplemented by other diverse convoluted records (Quilty, 1994). While carbonate

deposition has characterized the offshore history during the Cretaceous to Cenozoic, they are

gallingly incomplete in so far as records of onshore conditions (Quilty, 1994). The Australian

paucity of reliable chronostratigraphic records emphasizes the need for exploration of further

palaeo-data sources and techniques for inferring Quaternary community structure.

3. THE HISTORY OF PODOCARPACEAE

Extant lineages are only a minutiae of past coniferous diversity - when angiosperms were

rare and conifers covered Gondwana (~65 million years ago; Ma). Gymnosperms were once

dominant in the Australian landscape, now, Podocarpaceae are one of only a few

gymnosperm families to still occur in angiosperm dominated tropical rainforests (Hill, 1994).

Historically they were a major component of the Gondwanan flora, but a changing physical

environment, a mass extinction event (Creataceous-Paleogene boundary ~ 65 Ma), and rapid

diversification of the angiosperms, resulted in the displacement and consequential extinction

of many coniferous lineages. In comparison to the flowering plants, which when all taxonomy

is finally done may add up to 300 000 species, conifers are currently a diminutive group of

woody seed plants, with only 630 species (Hill, 1994).

Extant Podocarpaceae in particular are only a very small representation of a once highly

diverse group (Hill, 1994). The family-wide morphology has stayed conserved through

evolutionary time (Biffin et al., 2012), yet recent radiations of the family suggest Podocarpus

is more dynamic and adaptive than other podocarp genera (Biffin et al., 2011). Over time the

lack of landmass in the Southern Hemisphere and the inability to migrate in response to

pressure may have resulted in varied adaptive response and past diversification in the group.

The community associations of the podocarps have also stayed conserved (e.g.,

Araucariaceae, Nothofagaceae and Cunoniaceae); even in peripheral ranges such as Central

America they continue co-occurring with their palaeo-counterparts (Brodribb and Hill, 2003).

The ecological and environmental requirements of Podocarpaceae are specific, with

almost all species restricted to wet montane and rainforest environments (Brodribb and Hill,

2003). Regeneration is usually continual recruitment of shade-tolerant seedlings, or in less

shade tolerant species it is reliant on small-scale disturbances or topographic features such as

ecotones, rivers and ridgelines to open the canopy (Brodribb and Hill, 2003).

3.1. Taxonomy of Podocarpaceae

Podocarpaceae is comprised mainly of Southern Hemisphere conifers and is a large

family of 18 genera and 173 species (Hill, 1994; Quinn et al., 2003). Podocarpus is by far the

largest genus with 110 species, with the next largest genus being Dacrydium with 16. Of the


Rohan Mellick 218

18 genera of Podocarpaceae only seven are represented in Australia, with most species

restricted to northeast Queensland and Tasmania (Hill, 1994). The family was endemic to the

ancient super continent of Gondwana and is a classic member of Antarctic flora (Quinn et al.,

2003). The main centre of diversity is Australasia, mainly New Caledonia, Tasmania and

New Zealand, and to a smaller extent, Malesia and South America.

Podocarpus has a lancolate leaf with a prominent mid-vein with no secondary venation

(Fig. 1). Retrophyllum has opposing leaflets arranged so the abaxial surface on one side is up

and on the other is down (Hill and Pole, 1992). This possibly is a light-harvesting trait left

over from its high-latitude origin, where the sun remains low in the sky (Ed Biffin, University

of Adelaide, pers. comm.).

Figure 1. Variation in leaf morphology of shade-tolerant genera from the family Podocarpaceae

(Modified from source: Tim Brodribb, University of Tasmania).

Podocarpaceae members are evergreen shrubs or trees usually with a straight trunk and

mostly horizontal branches. The leaves are usually spirally arranged and are sometimes

opposite. The leaves are scale-like, needle-like or flat and leaf-like, and are linear to

lanceolate. Members are dioecious or rarely monoecious trees or shrubs with spirally inserted

(opposite in Microcachrys) oblong to scale-like leaves (functionally replaced by flattened

branches in Phyllocladus). Pollen cones are catkin-like and have many stamens. Male cones

are terminal on axillary shoots including numerous spirally arranged sporophylls every one

with two abaxial microsporangia (Mcarthy, 1998; http://www.ibiblio.org/pic/GymnospKey

/gymnosperm_key_glossary.html).

Pollen grains are winged (saccate or bisaccate). Female cones are pendant and mature in

one year. They are terminal on branches or terminal on short axillary shoots. Scales are



persistent or deciduous and including axes are fleshy or dry and not woody at maturity. The

seeds are completely covered by a fleshy structure referred to as an epimatium and are

wingless (Mcarthy, 1998). Epimatium and integument sometimes connate and form a leathery

testa. Germination is phanerocotular with two cotyledons and two parallel vascular bundles

(Van Royen, 1979; Pocknall, 1981; Silba, 1984).

3.2. The Origin of Podocarpaceae

Podocarpaceae originated in the Southern Hemisphere about 242 million years ago (Ma)

and later migrated north of the equator probably during the last 15 million years (My) when

the Australian plate contacted the Southeast Asian plate (Quilty 1994; Hill and Brodribb

1999). They also migrated further north through Central America to Mexico during the same

period. The podocarps appeared in the fossil record at the beginning of the Triassic when the

great super continent Pangaea broke up. They grew alongside araucarians, ginkophytes,

cycads, tree ferns, giant club mosses and horsetails (Quilty, 1994; Hill and Brodribb, 1999),

and were likely to be browsed and dispersed by dinosaurs (Bartholomai and Molnar, 1981;

Molnar, 1996).

The Cenozoic macrofossil record of the Podocarpaceae is extensive, especially in

southeastern Australia, where the majority of the extant genera have been recorded (Hill and

Brodribb, 1999). A few extinct genera (i.e., Podosporites, Willungia and perhaps Coronelia)

have also been reported from across high southern latitudes, confirming an extremely diverse

and widespread suite of Podocarpaceae during the Cenozoic in the region (Hill and Brodribb,

1999). The origins and relationships of the Podocarpaceae are unclear (Hill 1994) but

attempts to explain the phylogeny have improved over time (Kelch, 1997; Conran et al.,

2000; Biffin et al., 2011). Essentially the Podocarpaceae are a Southern Hemisphere family

though macrofossils are found in the Northern Hemisphere (Hill and Brodribb, 1999; Hill,

1994) suggesting a northern expansion. The earlier Mesozoic and Tertiary floras of the

Northern Hemisphere contain no Nothofagus or Podocarpaceae (Couper, 1960).

By the onset of the Mesozoic, the evolution of the Podocarpaceae is shown by the genus

Rissikia, evident from the Triassic (248-206 Ma) of Madagascar, South Africa, Australia and

Antarctica (Hill, 1994). Other early southern conifers, such as the Jurassic Nothodacrium and

Mataia (Townrow, 1967a; 1967b) are most probably podocarps, although, their relationship

to extant genera warrants classification (Stockey, 1990). Podocarps were still prominent in the

Cretaceous, albeit in Australia they were members of extinct genera (Hill, 1994), e.g.,

Bellarinea barklyi from the Early Cretaceous of Victoria (Drinnan and Chambers, 1986). Mill

(2003) discusses the biogeography of Podocarpaceae pertaining to extant/fossil taxa and the

palaeogeography of areas presently occupied by podocarps.

Molecular systematic studies using the locus rbcL for the Podocarpaceae show they are

monophyletic but with low overall main branch support, although most genera in the

phylogeny hold together as clades (Conran et al., 2000). Phyllocladaceae are nested inside

Podocarpaceae and Podocarpus is one genus with both subgenera positioned as clades

(Conran et al., 2000).

Podocarpus were abundant in Antarctic/Australian rift valley during the Cretaceous

(Dettmann and Jarzen, 1990; Hill, 1994). A number of extant podocarpaceous genera are

recorded as macrofossils in Australia during the early Tertiary (Hill, 1994). The majority of


Rohan Mellick 220

those genera, e.g., Podocarpus, Falcatifolium, Acmopyle, Dacrycarpus, Dacrydium,

Lepidothamnus and Phyllocladus, occur in southeastern Australia, but Retrophyllum has been

found in the southwest of Australia (Hill, 1994).

There are some shrubby species of Podocarpus that do presently occur within open-

canopied communities, particularly heath (i.e., P. spinulosus, P. drouynianus and P.

lawrencei), but like a number of taxa whose affinities are with rainforests, opinion is divided

as to whether these are regarded as particularly tolerant and adaptable remnants of rainforests

or are true components of open communities (Kershaw et al., 1994; Hill, 1994). Australasia

and Malaysia have the greatest diversity of living podocarps, where South America,

Antarctica and New Zealand have the greatest diversities of fossil podocarps. The fossil

podocarp flora of New Zealand is more recent and probably derived from that of Australia,

which has fewer living or endemic fossil genera (Hill, 1994; Mill, 2003; Wagstaff, 2004,

Jordan et al., 2011).

3.3. Podocarpaceae, Pinaceae and Pangaea

In the Southern Hemisphere today conifers achieve greatest abundance in wet forests,

where there ability to compete successfully with broad-leaved angiosperms is due in part to

their production of broad, flat photosynthetic shoots (Hill and Brodribb, 1999; Brodribb and

Hill, 2003). The Podocarpaceae produce large leaves (Fig. 1) and they have superior light

harvesting ability than the Pinaceae. The tall closed canopy forests of the equatorial region

have remained accessible due to this leaf morphology. The Pinaceae (with the exception of

Pinus krempfii) are shade-intolerant and have not been able to colonize south of the equator

(Hill and Brodribb, 1999; Brodribb and Hill, 2003).



Figure 2. The unsuccessful colonization of Pinaceae into Southeast Asia (red), and the successful

colonization of Podocarpaceae (black) (Modified from source: Tim Brodribb, University of Tasmania).

The colonization of Southeast Asia and subsequently Australia by Pinaceae has been

blocked by the equatorial evergreen rainforests (Fig. 2). Pines are physiologically incapable

of traversing this region due to shade-intolerance (Brodribb and Hill, 2003). In contrast, the

Podocarpaceae are shade-tolerant and can propagate under a canopy making them capable of

colonizing the equatorial zone and beyond (Hill and Brodribb, 1999; Brodribb and Hill,

2003).

3.4. Migration, Adaptation or Extinction

Shade tolerance, and the ability to propagate under a canopy, is a common character of

the Southern Hemisphere conifers. Possibly, high-latitude rainforest origin and the ability to

senesce over seasonal periods of darkness, is associated with the evolution of shade-tolerance

in the present members of Podocarpaceae. Relative to the southern supercontinent Gondwana,

the northern Laurasian supercontinent stayed together and allowed range shift across the

landmass in response to a changing environment (McLoughlin, 2001).

It has been shown from fossil and molecular systematic studies that conifers are a

monophyletic group and a common ancestor between the Northern and Southern Hemisphere

conifers occurred (Chaw et al., 1997: Stephanovic et al., 1998). The Northern Hemisphere

family Pinaceae is shade-intolerant, possibly a result of the ability to migrate in latitude,

where the relative lack of restriction did not necessitate significant adaptation to a changing

environment. While, the Southern Hemisphere conifers were restricted in distribution and

needed to adapt to the changing environment or, as many did, become extinct (Enright and

Hill, 1995).

Palynological records show that Northern Hemisphere tree populations are capable of

rapid migration in response to a warming climate. The Northern Hemisphere fossil pollen

record has revealed rapid migration rates of many temperate tree species of 100–1000 m/yr

during the early Holocene (McLachlan et al., 2005). The migration rates you would imagine

would be comparable to Southern Hemisphere tree populations, but detailed palynology to

record past migrations is absent, or of poor quality to record such events here in Australia.

Possibly, the relative lack of land mass in the Southern Hemisphere has necessitated

adaptation rather than migration in response to competitive exclusion. The high latitude


Rohan Mellick 222

circumpolar Taiga Forests of the Northern Hemisphere are devoid of arboreal angiosperms.

The distribution illustrates the ancestral retreat of gymnosperms away from increased

angiosperm competition at lower latitudes (Enright and Hill, 1995). While in the Southern

Hemisphere, the lack of migration routes south toward the pole may have caused the

remarkable adaptation of the southern conifers in response to increased angiosperm

competition (Fig. 1: broad leaves and shade tolerance; Enright and Hill, 1995; Brodribb and

Hill, 2003; Biffin, et al. 2011). Strong selection may account for the extinction of many

coniferous genera in the Southern Hemisphere, as recorded in the Mesozoic fossil record

(Hill, 1994; 2004).

3.5. Podocarpus

Podocarpus was first described in 1807 by L‟Heritier. The genus comprises 110 species

the majority of which are restricted to lowland and montane forests of warm temperate to

tropical areas in the Southern Hemisphere. Seven species are endemic to Australia: P.

dispermus, P. drouynianus, P. elatus, P. grayae, P. lawrencei, P. smithii, and P. spinulosus.

Podocarpus has two subgenera, subgenus Podocarpus and subgenus Foliolatus. The

subgenus Podocarpus have cones that are not subtended by lanceolate bracts and the seed

usually has an apical ridge, and a florin ring around stomata. While Foliolatus has a cone

subtended by two lanceolate bracts and the seed usually is without an apical ridge (Mcarthy,

1998).

Distribution of the subgenus Podocarpus is in the temperate forests of Tasmania, New

Zealand, southern Chile, with some species extending into the tropical highlands of Africa

and the Americas (Mcarthy, 1998). While Foliolatus, generally has a tropical and subtropical

distribution, concentrated in east and Southeast Asia and Malesia, overlapping with subgenus

Podocarpus in northeastern Australia and New Caledonia (Van Royen, 1979; Silba, 1984;

Harden, 1990).

3.5.1. Podocarpus: A Climatic Indicator

Podocarps are important to the study of the effect of long-term climatic change on, leaf

morphology (Sporne, 1965; Biffin, et al. 2011; Hill and Pole, 1992), physiology (Brodribb

and Hill, 2003), taxonomic make-up during the Cenozoic (Hill, 1994), distribution (Ledru et

al., 2007) and diversity (Quiroga and Premoli, 2007; Quiroga and Premoli, 2010). The

concentration of conifers in wet forest habitats left them vulnerable to Cenozoic climate

change and decreases in diversity have occurred since the Paleogene in all areas where fossil

records are available (Hill and Brodribb, 1999).

Podocarpus fossils have been used by palynologists and palaeo-climatologists to

reconstruct Quaternary climate in tropical South America even though the factors involved in

their modern distribution are not well understood (Enright and Hill, 1995). The moist

ecosystems where the vast majority of the genus grow are well defined climatically and their

bisaccate pollen grain are unmistakable. Pollen analysis using light microscopy has not yet

been able to identify to the species level reliably, so palynological interpretations are the

subject to several hypotheses (Enright and Hill, 1995).

For decades, palynologists working in tropical South America have been using the genus

Podocarpus as a climate indicator (Ledru et al., 2007). The combination of botany, pollen and



molecular analysis has proved to be reliable for determining population groups and their

regional evolution within tropical ecosystems. The refugia of rainforest communities

identified as crucial hotspots has allowed the Atlantic forests to survive under unfavorable

climatic conditions and are likely to offer the opportunity for this type of forest to expand in

the event of future climate change (Ledru et al., 2007).

Investigation of the long-term responses to climate changes in Podocarpus parlatorei, a

cold-tolerant tree species from the subtropics in South America, using distribution patterns of

isozyme variation has inferred northern expansion of the species during glacial periods

(Quiroga and Premoli, 2007). Podocarpus parlatorei is restricted to montane forests within

the Yungas, a cloud forest of the subtropics of northwestern Argentina and southern Bolivia.

It consists of disjunct population groups that are ecologically subdivided according to

latitude. These groups were expected to be genetically divergent from one another as a result

of historical isolation. The effective number of alleles and observed heterozygosity increased

with latitude, with the southern populations tending to be more variable and genetically

distinct. A positive association between genetic and geographic distances was detected and

reduction in diversity towards the north and high-elevation mountains are consistent with

evidence of patterns of forest migration resulting from climate change during the Late

Quaternary (Quiroga and Premoli, 2007).

3.5.2. Stand Dynamics

Stand dynamic studies on Podocarpus species have revealed that on poorly-drained,

nutrient-poor and high altitude sites where most of associated species were fairly shade

intolerant and light crowned, dense all-aged populations and the presence of numerous

saplings beneath the canopy suggested continuous regeneration (Lusk, 1996). Conversely, on

more favorable sites, were several of the associated angiosperms were highly shade-tolerant

and dense crowned, Podocarpus species were less abundant, and there regeneration from seed

appeared to be sporadic (Lusk, 1996).

Great longevity of shade tolerant conifers is probably crucial to their persistence strategy

in competition with shade-tolerant broad leaved species in undisturbed stands on favorable

sites (Lusk, 1996). A review of literature on southern temperate forests (Enright and Hill,

1995; Enright and Ogden, 1995; Lusk, 1996) disputes the hypothesis that heavily shaded,

infrequently disturbed habitats are an evolutionary refuge for conifers (Bond, 1989).

Sites likely to have high leaf area indices and infrequent disturbance are more

successfully exploited by shade-tolerant angiosperms (Lusk, 1996). Regeneration can take the

form of continual recruitment of shade-tolerant seedlings, or in less shade tolerant species it is

reliant on small-scale disturbances or topographic features such as rivers and ridgelines to

open the canopy (Brodribb and Hill, 2003).

The Myall Lakes and Jervis Bay populations of P. elatus are regularly burnt and

considerable recruitment is observed post-burning (Chris Quinn, The Royal Botanic Gardens

and Domain Trust, Sydney, pers. comm.). Podocarpus drouynianus and P. spinulosus are

probably fire adaptable remnants of rainforest. They resprout from lignotubers and are fire

tolerant. It is also shown for P. lawrencei that recruitment is favored after fire (Macdonald,

2004).


Rohan Mellick 224

3.6. Podocarpus elatus R.Br. Ex Endl

Eukaryota; Viridiplantae; Streptophyta; Embryophyta; Tracheophyta; Spermatophyta;

Coniferopsida; Pinales; Podocarpaceae; Podocarpus; Foliolatus; elatus.

Podocarpus elatus is an evergreen conifer that was abundant across the warmer wetter

parts of present day Australia before the arrival of Europeans. Indigenous people possibly

referred to P. elatus as „Djerren Djerren‟ in the Eora dialect of the Gadigal nation but there is

no known dreaming or song lines associated with the species (Clarence Slockee, The Royal

Botanic Gardens and Domain Trust, Sydney, pers. comm.). Podocarpus is derived from two

Greek words pous (foot) and karpos (fruit), referring to the fleshy fruit stems. The species

name elatus, is Latin for 'tall', refers to how tall the trees grow (http://www.

anbg.gov.au/anbg/conifers/Podocarpus-elatus.html).

Figure 3. The distribution of Podocarpus elatus based on a compilation of collections made for all major

herbaria in Australia since 1818. The Clarence River Corridor is a regional genetic boundary between

northern and southern populations (indicated by bar) (Source: Mellick et al. 2011). Photos: Podocarpus

elatus fruiting and mature tree.



The species ranges from New South Wales to Queensland and prefers habitat in and

around rainforest north of the Beecroft Peninsula, Jervis Bay, NSW (Harden, 1990), in the

south, to Cape York Peninsula, QLD, in the north (Fig. 3). It prefers subtropical, gallery and

littoral rainforest habitats, with rich, moist, non-alkaline soils but the species is usually more

successfully established on poorer sites where competition is less vigorous. Plants will

tolerate salt spray and frosts to -7oC so long as they have adequate moisture levels and

humidity. Maturing in 8-10 years, P. elatus is fast growing once established (Harden, 1990).

Populations observed along the East Coast of NSW exhibited little recruitment with most

populations being composed of a few mature trees with no seedlings or saplings present.

It is commonly referred to as the Plum Pine, Illawarra Plum or Pine, Brown Pine

(Harden, 1990) and Turpentine Pine in the timber industry. Growing to 40m with brown bark

(fibrous) that is often fissured and scaly on old trees. It is a dioecious species having separate

male and female trees. The fruits are a purple black color and composed of two segments. The

edible portion is a grape-like (10mm diameter), sweet with a juniper-like flavor but

mucilaginous that ripens from March to July (Harden, 1990). The inedible part of the fruit is

the hard dark seed that is situated externally on top of the swollen stalk. The fruit has a high

ascorbic acid content (vitamin C), contains minerals, fat, protein, and are high in energy. The

leaves are lanceolate, oblong to linear (6-18mm wide, 5-14cm long) and the midvein is

prominent (Harden, 1990). Leaves can grow to 25cm long on young trees. The male cones are

narrow-cylindrical, catkin-like, to 3cm long, and are in sessile clusters (Harden, 1990).

Female cones are stalked, and solitary; scales are few, fleshy, uniting with the stalk to form a

fleshy receptacle (Harden, 1990).

3.6.1. Reproductive Biology of Podocarpus elatus

Podocarpus elatus has a reproductive cycle of one year and is a dioecious species with

separate male and female plants. Information specific to P. elatus is limited. The closest

phylogenetic species with information available is P. neriifolius, which exhibits

underdeveloped embryos at time of dispersal, although germination is usually rapid with seed

quickly losing viability (i.e., 20-60 days) so that there is no persistent soil-stored seed bank

(Enright and Jaffr, 2011). Podocarpus totara is a large tree of New Zealand and is a member

of the sub-genus Podocarpus, while P. elatus is a member of Foliolatus. Podocarpus totara

in New Zealand has a reproductive cycle of two years, and strobilus initiation is in

September, followed by a nine-month period of winter dormancy until emergence during the

growth flush in July-August of the following year (Wilson, 1999). Pollination in P. totara

occurs in mid-October to mid-November at the megaspore tetrad stage, where female strobili

bear only one or two ovules. During December pollen germination and fertilization occur

quickly and the pollen tube carries the body cell which branches out after reaching the

archegonia. Embryo maturation is complete by February (Wilson, 1999).

3.6.2. Podocarpus elatus Timber and Cultivation

The early loggers targeted Podocarpus elatus extensively for it is a useful timber that has

wide applications from boat building to cabinet making. The wood has a very fine, even

texture and a straight grain with faint growth rings, making it desirable for tabletops,

furniture, musical instruments (piano keys and violin bellies) and wood turning. Commonly

referred to as Brown Pine, the yellow wood turns brown in color shortly after being exposed


Rohan Mellick 226

(Barker et al., 2004). It has not been widely planted as a timber tree due to not being as light

and strong as its relative Araucaria cunninghamii. It has however been widely clonally

cultivated for roadside plantings (David Bateman, Waverly Council, Sydney, pers. comm.), as

an ornamental shade tree and now more commonly for the native bush food market. There is

horticultural development research published on P. elatus (Ahmed and Johnson, 2000)

although the majority of cultivation of the species is now clonal with preference for male

ramets due to not dropping fruit. This practice may be homogenizing the P. elatus pollen

cloud.

3.7. Fire and Ecotonal Specialization

Bowman‟s book (Bowman, 2000), titled „Australian Rainforests: islands of green in a

land of fire‟ emphasizes the existence of Australian rainforests is largely owed to the patterns

and processes responsible for the sharp boundaries between drier rainforest-types and fire

promoting communities. The capacity of the fragmented East Australian rainforests to resist

destruction by fire is partially a consequence of the species composition and structure of the

surrounding, frequently disturbed drier rainforest communities.

Podocarpus elatus exhibits a comparative advantage in drier communities, fringing more

floristically complex core communities. Larger dense populations are usually found on the

southern side of a break in the rainforest canopy with a northern aspect. This drier more

frequently burnt rainforest boundary would seemingly select against longer-lived species,

though for P. elatus, thick bark and waxy leaf cuticle, gravity seed dispersal, shade-tolerance

(i.e., rapid and dense propagation under female trees), the ability to senesce as a sapling for

long periods and possibly leaf chemistry may aid it in occupying these fire prone areas where

other rainforest trees are less able.

Podocarpus elatus is known as Turpentine Pine, due to the high concentration of

turpentinol compounds in the foliage, as are found in Pinus species. The evolution of these

compounds is likely a response to increased herbivory. Turpentinols increase flammability

and possibly these compounds in the species may also be a result of frequent burning. It

would be interesting to know if P. spinulosus, a co-occurring fire adapted podocarp, has

similar leaf chemistry to P. elatus. Understanding the past expansion/contraction of dry

rainforest communities will aid in understanding the seemingly precarious existence of these

relatively fragile-wet communities in such a harsh-dry landscape.

Evidence suggests large herbivorous dinosaurs browsed on, and subsequently dispersed,

the Australian podocarps (Bartholomai and Molnar, 1981; Molnar, 1996), which may have

contributed to the podocarps ecotonal habitat preference (Hill and Brodribb, 1999), as heavily

forested areas would restrict the movement and dispersal from such browsers. It is worth

briefly mentioning that the rapid extinction of mega fauna possibly increased fire frequency

due to biomass gain from reduced browsing (Rule et al., 2012), indicating that extinct mega

fauna may have had a considerable influence on rainforest fragmentation.

Drier rainforest communities, typified by Araucarian emergents and Podocarpus species,

are sensitive to effects of fire. Kershaw et al. (2005) postulated that dry rainforest

communities fringed wetter core rainforest communities in the Australian Wet Tropics

(AWT), and expansion and contraction of these communities are seasonal and climate driven.



Precipitation gradients are largely responsible for successional community turnover from wet

rainforest to dry rainforest and eventually to sclerophyllous forest during dry periods.

During wet periods, wetter rainforest would encroach into previously dry rainforest and

the lack of fire would allow dry rainforest to expand into areas previously occupied by

sclerophyllous forest. Evidence for the latter can be found throughout the rainforests of

northern NSW (central range of P. elatus) where old Eucalyptus grandis trees tower over wet

rainforest communities far from fire prone boundaries.

These boundaries are an important parameter of climate change. With increasing

fragmentation (associated with aridification), the abundance of plant communities associated

with these boundaries may also increase, because of increasing surface area. Past changes in

species composition and extent of these communities have been showed to be correlated to

the glacial cycles of the Quaternary period (Kershaw et al., 2007a).

Podocarpus species are known to senesce for periods up to a decade with very little

growth (Brodribb and Hill, 2003). The species exhibits remarkable traits to survive in such a

highly diverse, dynamic and competitive environment. Possibly, high-latitude rainforest

origin of Podocarpaceae and the need to senesce over seasonal periods of darkness is

associated with the evolution of shade-tolerance in the present members of the family.

CASE STUDY ONE: CONSEQUENCES OF LONG- AND SHORT-TERM

FRAGMENTATION ON THE GENETIC DIVERSITY AND

DIFFERENTIATION OF A LATE SUCCESSIONAL RAINFOREST CONIFER

The east Australian rainforests provide an informative system with which to study

historic climate-driven habitat fragmentation. The long life cycle of rainforest conifers and

consequent lag effects on genetic variation, offer insights into demographic stochasticity in

small populations and persistence in increasingly fragmented systems. Microsatellite markers

were used to investigate the genetic diversity and structure of Podocarpus elatus

(Podocarpaceae), a long-lived rainforest conifer endemic to Australia (Mellick et al., 2011).

The markers used in the study were developed and characterized in Almany et al. (2009) and

assessed in Mellick et al. (2011). Twenty-seven populations throughout the east Australian

rainforests were screened and two divergent regions separated by the dry Clarence River

valley (New South Wales) were found (Fig. 4).

This new biogeographic barrier may be called the Clarence River Corridor (CRC). Niche

modeling techniques were employed to verify the incidence of habitat divergence between the

two population groups (Fig. 5).

Significantly high inbreeding was detected throughout the species range with no evidence

of contemporary bottlenecks. Most of the diversity in the species resides between individuals

within populations that suggest the species would be sensitive to the adverse effects of

inbreeding, though evidence suggests that these populations have been small for several

generations (Mellick et al., 2011). Higher diversity estimates were found in the southern

region, yet it is likely that the species survived historic population contraction in dispersed

refugia within each of these genetically differentiated regions.


Rohan Mellick 228

Figure 4. Results based on K= 2 using the Bayesian framework implemented by STRUCTURE across

individuals from the 27 populations of Podocarpus elatus.(ordered according to latitude). (a) All

individuals were used in the analysis that clustered them into two regions representing: Southern Region

(NSW) and Northern Region (Northern NSW/Qld). (b) Results based on K= 2 using the same Bayesian

framework on juvenile individuals and (c) on the mature individuals (Source: Mellick et al., 2011).

Figure 5. The regional climatic models for Podocarpus elatus based on collection data (National

Herbaria, YETI database and ATLAS database) and environmental data (WORLDCLIM). Darker areas

indicate a higher probability of occurrence (white indicates 0–20% and black indicates 80–100%). The

black bar shows the Clarence River Corridor (Source: Mellick et al., 2011).

Mellick et al. (2011) is the first study reporting regional genetic structure either side of

the Clarence River Corridor in a plant. Environmental niche modeling of the separate regions



has shown distinct differences between the climatic niches of each region. Primarily, the

southern model is associated with winter (uniform) rainfall patterns and northern with

summer rainfall patterns (Mellick et al., 2011). Some overlap between the two distinct niches

coincides geographically with the Macleay McPherson Overlap Zone that allowed for

validation of each regional model (i.e., each model predicted small areas of high probability

of occurrence in the neighboring region in the absence of occurrence data). These climatic

differences may underline long-term divergent processes between the two regional groups. In

the shorter term and at a regional scale, anthropogenic habitat clearing (i.e., Big Scrub

rainforest, Nth NSW) is also likely to contribute to the differences measured between regions

and among generational cohorts (Mellick et al., 2011).

The lower diversity and distributional pattern of the northern populations could suggest

that the current distribution is the result of an earlier northern expansion from a narrower

genetic pool (Mellick et al., 2011). This theory has been investigated in a phylogeographic

study combining genetic and palynological data that will helped identify migration routes and

date expansion/contraction events (case study two; Mellick et al. 2012). The projection of the

spatial distribution models generated for the northern and southern population groups onto

past climatic conditions provides support to adaptive divergence between the regions, and

elucidates the contrasting expansion/contraction climatic response of the species in relation to

early successional rainforest taxa (Mellick et al., 2011).

4. VEGETATION CHANGE IN AUSTRALIA DURING THE CENOZOIC

The Australian landscape has not always been arid, and the central desert was once well-

vegetated (McLoughlin, 2001; Martin, 2006). At the start of the Cenozoic (65 Ma), Australia

had a warm/humid climate and the vegetation was predominantly warm to cool temperate

rainforest (Martin, 2006). The history shaping the eastern Australian rainforest communities

involved, high southern latitude of the continent at the end of the Cretaceous period (65 Ma)

and its gradual movement north toward the equator; general global aridification over the last

65 My amidst short term fluctuations and the evolution from ancestral Gondwanan lineages

of most taxa, but also immigration onto the continent by both plant and animal lineages from

different geographic sources at several times (Martin, 1982; Greenwood and Christophel,

2005; Sniderman and Jordan, 2011).

Separation of Australia from Antarctica started at the end of the Paleocene (54.8 Ma)

with the formation of a narrow strait, but it was the mid-Oligocene (31.1 Ma) when the first

channel was formed between the continents (McLoughlin, 2001). This allowed for the

Antarctic Circumpolar Current (ACC) to be created due to the removal of the block caused by

the South Tasman Rise, which cooled the continent and increased the size of the west

Antarctic ice cap, which removed the available atmospheric moisture and reduced

precipitation globally (Martin, 1982). The magnitude of the ACC as we know it today was

established about 18 Ma. This started the rapid cooling of Antarctic and the eventual demise

of the rich Antarctic flora.


Rohan Mellick 230

4.1. Cenozoic Summary

In the Palaeocene (65–54.8 Ma), the Australian continent was warm/wet, and the

vegetation was mostly rainforest, except the northwest where there was limited aridity and

central Australia where aridity would have been seasonal (Martin, 2006). Gymnosperms were

the predominant vegetation type of south-eastern Australia but angiosperms were dominant in

central Australia (Martin, 2006). Podocarpaceous gymnosperms dominated the Palaeocene

vegetation due to the cool temperate climate similar to Tasmania and New Zealand today.

The early Eocene (54.8 Ma) was hot/humid ensuing angiosperm diversification and

dominance, and the increase of mega-thermal taxa in south-eastern Australia (Martin, 2006).

Rainforest gymnosperms were less abundant than the Palaeocene.

In the mid-late Eocene (49–41 Ma) there was a decrease in temperature – the vegetation

was mainly rainforest, but in central Australia there were some open vegetation and

sclerophyllous taxa (Martin, 2006). Sclerophylly developed as a response to infertile soils,

long before the climate became dry (Martin, 2006). Nothofagus became the dominant pollen

type, especially in the southeast, and Lauraceae the dominant leaf type (Martin, 2006).

There was a rapid cooling during the early Oligocene (33.7 Ma), the result of opening of

the seaway between Australia and Antarctica and strengthening of the Antarctic Circumpolar

Current (Martin, 2006; Quilty, 1994) Angiosperm diversity decreased and Nothofagus pollen

became more prominent (Martin, 2006).

During the Oligocene (33.7–23.7 Ma) temperature and angiosperm diversity increased,

and the occurrence of swamps in south-eastern Australia were more common, indicating that

rainfall was particularly high at this time (Martin, 2006).

The early Miocene (23.7 Ma) saw an increase in temperatures and the vegetation became

more variable in south-eastern Australia and in central Australia, rainforest was limited to

small pockets. The Miocene remains the warmest and most humid period during the Cenozoic

(Martin, 2006).

The mid-Miocene (16.4 Ma) is when regular flows in palaeo-drainage systems over much

of western and central Australia ceased, and alkaline lakes of inland basins in central

Australia deposited dolomite indicating high rates of evaporation and a well-marked dry

season (Martin, 2006).

The climate became colder and drier during the late Miocene (11.2 Ma), reducing the

abundance of rainforest in Australia and increasing the fire tolerant Myrtaceae flora especially

Eucalyptus (Martin, 2006).

A brief warming period and increased precipitation in the early Pliocene (5.3 Ma) caused

an expansion of rainforest in river valleys of the western slopes and south-eastern Australia

(Martin, 2006).

The climate increasingly became drier during the late Pliocene (3.6 Ma) (Fig. 6) and

rainforest taxa contracted further to the wetter coastal and highland regions. Grasslands

became more prevalent in inland areas and the modern climate was established, but it still was

considerably wetter than today (Martin, 2006).

The Pleistocene (1.8–0.01 Ma) fluctuated between drier glacial and wetter interglacial

periods. About 0.5 Ma, there was a marked shift to a dry climate (Martin 2006). The present

interglacial is drier than the previous interglacial (130 ka), with the last glacial maximum

(LGM: 21 ka) being particularly adverse. Although conditions have improved, precipitation

has not returned to the levels of the previous interglacial period (Martin, 2006).



Figure 6. The build-up of arid cycles during the last 5–6 million years (Source: Bowler 1982).

The earth's climate has become cooler through the Tertiary (65-1.8 Ma) with frequent

oscillations that increased in amplitude (Fig. 6) and lead to the series of major ice ages of the

Quaternary (Hewitt, 2000).

The surviving rainforest taxa have had to migrate into areas (refuges) that have conserved

favorable environmental conditions for survival, and the ranges of many Gondwanan lineages

has decreased dramatically. The present genetic structure of populations, species and

communities has been primarily formed by the Quaternary ice ages, and genetic, fossil and

environmental data combined can greatly help our understanding of how organisms were

affected (Bennett, 1997; Hewitt, 2000).

4.2. Quaternary Vegetation

The Quaternary is a period in which modern life as we know it have evolved, or have

persisted from the Tertiary largely unaffected by environmental change (Hope, 1994; Hewitt,

2000). The Quaternary chronology is not based on widespread evolutionary changes, as are

older periods, but instead uses climatic changes, such as the growth of ice sheets and the

spread of cold surface water (Hope, 1994; Wright Jr, 1984). Fluctuations of heavy isotope

oxygen (18

O) from marine sediments and ice cores, records a series of cyclic changes of

periods of 100 ky, being glacial maximums alternating with inter-glacial periods. This record

shows in the last 600 ky there have been six glacials periods (Fig. 6).

The two divisions of the Quaternary are the Pleistocene from 1.8 Ma to 10 ka, and the

Holocene, which includes the present inter-glacial. The Pleistocene represents the period of

establishment of our current landscapes, climatic patterns and diversity, and the adaptation of

the Tertiary biota to these new environments (Hope, 1994). Vegetation history of the

Quaternary is of particular importance when comparing extant flora, or recent extinctions, to

fossil flora (Hill, 1994; Hill, 2004; Hope, 1994; Wright Jr, 1984).


Rohan Mellick 232

4.3. East Australian Rainforests

World Heritage Listing of the east Australian rainforests (isolated sub-tropical and

tropical forests that occur above 300m) was awarded largely due to the rare assemblage of

Lepidozamia hopei, Agathis robusta, Prumnopitys ladei, Podocarpus grayae and Podocarpus

elatus which includes the closest living counterparts of Jurassic-age fossils. The members of

Podocarpaceae in the rainforests of tropical Australia are Podocarpus dispermus, Podocarpus

elatus, Podocarpus grayae, Podocarpus smithii, Prumnopitys ladei and Sundacarpus amara.

All occur in the Wet Tropics, although P. elatus is very rare in the Wet Tropics. It should

however be included, as it is more typical of somewhat drier (Araucarian) rainforest

(Greenwood and Christophel, 2005).

Tenuous persistence of the Australia Wet Tropics (AWT) rainforest communities under

increasing aridification can be attributed to orographic uplift and subsequent increase in

precipitation in those areas as a result of the Great Dividing Range (GDR). The vast tracts of

rainforest that once occurred throughout Australia contracted to these areas of high

precipitation (Kershaw et al., 2007b). Floristic interchange with Southeast Asian rainforests

may have occurred (Sniderman and Jordan 2011) but possibly was prevented by a dry

corridor in the northern Australian-Sahul Shelf region (Kershaw et al., 2005).

The aridification of the continent, decline of mega fauna and advent of indigenous

burning increased the exposure of the relatively fire susceptible drier rainforest communities.

“Araucarian forest appears to have remained intact until it was progressively replaced by

sclerophyllous vegetation as a result of increased burning over the last 200,000 years”

(Kershaw et al., 2005). Kershaw et al., (2005, 2007a) provides excellent evidence of

community turnover in the Australian Wet Tropics (AWT) in response to recent glacial cycles

of the Quaternary. His extensive work has provided invaluable records of fossil pollen over

the last 230 ky (Fig. 7).

During interglacial periods in the AWT the occurrence of Podocarpus pollen is

dramatically reduced when compared to glacial periods, indicating a preference for cooler

drier environments, and/or susceptibility to angiosperm competition. Possibly, the genus

persists during periods of angiosperm dominance (interglacials), due to association with drier

fringing communities around core rainforest communities, where angiosperms are present yet

are not as competitive (Lusk, 1996).

On another front it can be observed that during glacial periods the sclerophyllous taxa

become more abundant. This fire promoting vegetation could be encroaching on drier

rainforest, and accordingly drier rainforests communities may infringe wetter rainforests

during glacial periods.

Kershaw et al. (2007a) on completion of the Lynch‟s crater sequence (Fig. 7), maintains

the pattern of complex rainforest expansion during wetter interglacial periods is replaced by

drier rainforest and sclerophyll vegetation during drier glacials. Araucaria and Podocarpus

(drier rainforest) incur cyclic changes in abundance coinciding with the last two glacial

cycles. During the last glacial period, these genera are replaced by more fire-tolerant

elements, as is understood from the recorded increase in fire frequency during this period.

The decline in a more fire sensitive sclerophyll taxon Callitris, may suggest an earlier,

more regional increase in burning, a feature not inconsistent with the evidence in the offshore

fossil record, that regionally Araucariaceae had an initial decline about 130 ka, after some 10

My of near dominance in the landscape (Kershaw et al., 2007a).



Figure 7. A pollen record of the last 230 ka from Lynch's Crater (AWT), north-eastern Australia

(modified from Kershaw et al., 2007a), showing the relative proportions of the major groups of arboreal

forest taxa and percentages of major contributors to these groups. All percentages of individual taxa are

based on the arboreal forest taxon sum for individual spectra. Corresponding change in temperature (two

glacial cycles) according to Vostok (Antarctica) ice core data (modified from Petit et al., 1999) is

provided on the left (Source: Kershaw et al., 2007a; Petit et al., 1999).

The impact of these changes on the current landscape is that Araucaria (dry rainforest) is

now restricted to small isolated patches that are unlikely to regain their previous dominance

(Kershaw et al., 2005).

The extinction of Dacrydium from the continent, whose closest morphological type

occurs as in New Caledonia, is clearly related but its presence in the AWT record may

suggest it was associated with a different community – a peat land or swamp forest that is

now extinct in Australia (Kershaw et al., 2007a).

In AWT there are four occurring Podocarpus species (Fig. 7) associated with different

mesic habitats and this poses a problem when assessing species-level dynamics because the

pollen record is recorded at the generic-level, and does not account for differences in habitat

preference among the species.

Podocarpus elatus is very rare in AWT and only occurs in dry rainforest at high

elevations, where it co-occurs with Podocarpus grayae, which also expands in to lowland

areas. Podocarpus dispermus occurs in wet rainforest at lower elevations and Podocarpus

smithii occurs in wet rainforest at high elevations.

So there is a problem assessing species-level dynamics and community turn-over from a

generic profile, the signatures of four Podocarpus species are compounded within Fig. 7.

Shimeld (1995, 2004) provides evidence of long-term decline of Podocarpus elatus since

the previous interglacial (130 ka) from central areas of the species‟ current range (Fig. 8).

Two adjacent deposits were used and chronology was replicated. The only other co-occurring

podocarp is Podocarpus spinulosus, which is expected to contribute little to the pollen profile

because it is a dry adapted shrub with different community associations.


Rohan Mellick 234

Figure 8. Fossil observations at Moffat‟s Swamp/Creek, Port Stephens, NSW (130 ka to present)

showing an overall decline of Podocarpus elatus pollen since the previous interglacial (130 ka)

(Source: Shimeld 1995).

4.4. Palynology

Pollen analysis has been well-established as a means for reconstructing vegetation history

(Brewer et al., 2002; Hill, 1994, 2004; Kershaw et al. 2007b, 2007a, 2005, 1994; Shimeld

1995, 2004, Woods and Davis, 1989). The pollen grains and spores produced by plants are

preserved well in anoxic environments (e.g., lake deposits, forest hollows, peat bogs) and may

be extracted from these deposits by sampling exposed sections or by taking sediment cores



(Brewer et al., 2002). The variation in the pollen assemblage are normally presented in the

form of a pollen diagram (Fig. 7) with the changes in pollen percentages of each taxon plotted

against depth, enables the temporal changes in the representation of individual taxa and the

type of surrounding vegetation communities to be determined.

Kershaw et al. (1994) indicates that it is not always possible to separate different genera

on their pollen morphology. Ledru et al. (2007) has found it impossible to classify

Podocarpus pollen grains at the species-level using light microscopy. Techniques used in

forensic palynology (Lynn Milne, University of Western Australia, Perth, pers. comm.) made

need to be incorporated into current methodology, if species-level identification using light

microscopy is important in reconstructing past community structure.

Fossil pollen sampled from sediment cores taken from the bottom of Lake Euramoo

(AWT) has revealed that there has been a gradual change in the surrounding vegetation from

dry adaptive species to rainforest species (Haberle, 2005). This vegetation change peaked at

about 7300 years ago during the Holocene Climatic Optimum and has since been in decline.

This is consistent with the habitat becoming wetter until the shift towards competitive

dominance of taxa adapted to drought and fire, which may be a response to the intensification

of El Nino-related climatic cycles (Haberle, 2005).

In north-eastern Australia, recent pollen spectra from the Ocean Drilling Program (ODP)

have sufficient in common with the riverine samples to suggest that rivers are contributing a

major pollen component to the offshore sediments (Moss et al., 2005). This would suggest

riparian vegetation (gallery forest) and associated taxa would be well-represented in the

offshore fossil record. Recent pollen samples from core tops taken from the Grafton Passage

on the continental shelf, that was thought to be the major passage for pollen transport to ODP

Site 820 (16°38′S, 146°18′E), show significant differences to both riverine/ODP samples and

suggest that pollen is dispersed across the continental shelf and through the outer Great

Barrier Reef system in an unanticipated fashion (Moss et al., 2005).

4.4.1. Podocarpus Pollen Morphology

Species-level taxonomic resolution is limited for Podocarpus pollen and understanding of

past community structure and relative abundance of species from the fossil pollen record is

not reliable (M. K. Macphail, Australian National University, Canberra, pers. comm.).

Podocarpus pollen grains are monads, heteropolar, bilateral; vesiculate and bisaccate

(Pocknall, 1981). Podocarpaceae pollen mostly have sacci (Wodehouse, 1935), small bladder

like structures flanking the corpus (Fig. 9). Their likely function is to increase buoyancy to

facilitate migration of the pollen grain through the pollination fluid in the micropylar canal,

tending to orientate the grain so that the germinal furrow contacts the female gamete (Doyle,

1945). Sacci structures may also be involved with flight, increasing the dispersal distance of

pollen and aiding pollination (Proctor et al., 1996). Sacci are always associated anatomically

with the furrow and may function in protecting it in periods of water stress (Pocknall, 1981).

Podocarpus pollen is easily identifiable due to distinctive bi-saccate grains (Fig. 9).

Although bi-sacci is a synapomorphic trait shared with other coniferous flora, it is unique to

Podocarpus in Australian fossil deposits. The mesic ecosystems where the trees grow are

climatically well defined, so the broad distribution patterns can be extrapolated from point

locations of fossil remains, due to distinct climatological requirements.


Rohan Mellick 236

Figure 9. Bilateral cross-section of Podocarpus elatus bi-saccate pollen.

5. PHYLOGEOGRAPHY AND POPULATION GENETICS

Phylogeography is a term coined by Avise et al. (1987) to illustrate the spatial

distributions of phylogenies. Phylogeography is used to determine common responses of

organisms to historical biogeography so to understand divergence and speciation, and to

comprehend geographic distributions of genealogical lineages including those at the

intraspecific level (Avise, 2000). Phylogeography interprets current genetic structure,

phylogenetic relatedness of alleles and their spatial arrangement among and within

populations to reveal historical components of gene flow (Avise, 2000).

Individual species phylogeography provides identification of geographic areas containing

high genetic diversity or distinct evolutionary units for that species (Byrne, 2007). These

areas are usually inferred as refugia, although past migration of diversity may have an

influence. Phylogeographic data collected from wide-ranging species allows for phylogenetic

estimates of gene flow to be made (Lowe et al., 2004). Phylogeography is a critical juncture

between established fields and effectively weights the influences of dispersal versus

vicariance in shaping the geographic distributions of genetic traits (Avise, 2000).

The field takes a population genetics and phylogenetic perspective on biogeography. An

explicit focus on a species biogeographical past sets phylogeography apart from classical

population genetics and phylogenetics (Knowles and Maddison, 2002). Past events including

population expansion/contraction, bottlenecks, vicariance and migration can be identified.

Recently developed approaches integrating coalescent theory, environmental niche modeling

and genealogical history of alleles can accurately address the relative roles of these historical

events in shaping current patterns (Cruzan and Templeton, 2000).

The field can inform conservation strategies through the assessment of genetic diversity

that incorporates an evolutionary perspective, and allows evaluation within a geographical

context, so providing integration with other biogeographical information (Byrne et al., 2011).

A number of studies have been interested in the spatial and temporal response of organisms to

climate driven distributional fluctuation (e.g., Bell et al., 2007; Mellick, et al. 20011, 2012,

2013a, 2013b; Pease et al., 2009; Petit, 2002, Petit et al. 2002, 2003; Richards et al., 2007;

Rissler et al., 2006; Scoble and Lowe, 2010).



5.1. Population Genetics

Population genetics is the study of the allele frequency distribution and change under the

influence of the four evolutionary forces: natural selection, genetic drift, mutation and gene

flow (Hartl and Clark, 1997; Lowe et al., 2004). It takes account of population subdivision

and population structure in space, and attempts to explain such phenomena as adaptation and

speciation. Population genetics was a vital ingredient in the modern evolutionary synthesis;

its primary founders were Sewall Wright (Wright, 1921, 1931), J. B. S. Haldane and R. A.

Fisher (1930), who also laid the foundations for the related discipline of quantitative genetics

(Falconer and Mackay, 1996; Hartl and Clark, 1997).

5.1.1. Hardy-Weinberg Principal

The Hardy–Weinberg principle or equilibrium (HWE) states that the genotype

frequencies in a population remain constant or are in equilibrium from generation to

generation unless specific influences are introduced (Lowe et al., 2004). For example in a

random mating diploid organism, AA + 2Aa + aa = 1 genotype proportions are expected.

Those influences include non-random mating, mutations, natural selection, limited

population size, random genetic drift and gene flow. Genetic equilibrium is a basic principle

of population genetics (Emigh, 1980); it is achieved in one generation, and requires the

assumption of random mating with an infinite population size (Hartl and Clark, 1997).

5.1.2. Genetic Distance

Genetic distance is a measure of the dissimilarity of genetic material between different

species or individuals of the same species (Pollock and Goldstein, 1995). It is a degree of

similarity between a pair of individuals, populations or species were values range between

zero (identical) and one (completely different) (Lowe et al., 2004). By comparing the

percentage difference between the same genes or neutral loci of different species, a figure can

be obtained, which is a measure of genetic distance. Depending upon the difference and

correcting for known rates of evolution, genetic distance can be used as a cladistic tool.

5.1.3. Heterozygosity

Quantification of within population genetic variation is central in the interpretation of

genetic differentiation between populations. Heterozygosity (Ho) is a measure of genetic

diversity, being the frequency of heterozygotes for one locus within a population (Page and

Holmes, 1998). A heterozygote is an individual with two (diploid) or more (polyploidy)

alleles at a locus. In understanding how genetic diversity is partitioned within a population, it

is important to determine how many alleles are present at a single locus. Usually the observed

(Ho) and expected (He) heterozygosities are compared through the determination of allelic

frequencies. Deviation away from the expected HWE frequencies indicates that the

population is not randomly breeding and some degree of assortative mating is occurring

(Lowe et al., 2004).

5.1.4. The Fixation Index

In 1921 what is known as the fixation index was defined (Wright, 1921). The purpose of

which is to quantify the inbreeding effect of population sub-structure (Lowe et al., 2004). The


Rohan Mellick 238

index equals the reduction in heterozygosity expected with random mating at anyone level of

a population hierarchy relative to another (Hartl and Clark, 1997). It quantifies genetic

differentiation and proves useful because it allows for an objective comparison of the overall

effect of population substructure among different organisms (Lowe et al., 2004). F-statistics

describe genetic differentiation by partitioning variation into three levels: total population,

sub-population and individual (Lowe et al., 2004). FST and RST statistics refer to analysis of

variance in allele frequencies and in repeat numbers respectively and are used to illustrate

quantitatively the degree of genetic structure (Pérez et al., 2002).

5.1.5. Inbreeding Depression and Heterosis

Inbreeding depression is a term indicating reduced fitness in a given population as a

result of breeding of related individuals (Lowe et al., 2004). Although inbreeding to some

degree accounts for all the diversity that we see today, it is a double edged sword and

breeding between closely related individuals, called inbreeding, results in more recessive

deleterious traits manifesting themselves. The more related the breeding pair the more

deleterious genes the offspring will have (i.e., increased genetic load), resulting in unfit

individuals. Genetic load is a term widely used in domestication and one that will become

more familiar to conservation efforts as natural populations become smaller and more

fragmented.

The effects of inbreeding depression can be dramatic between selfed and out-crossed

families (Fig. 10) depending on the genetic load. In a population where inbreeding occurs

frequently, fatal recessives alleles would be rare, due to promptly being eliminated from a

population. With infrequent inbreeding, recessive deleterious alleles will be masked by

heterozygosity, and so heterozygotes will not be selected against (assuming dominance).

Obligate outcrossers, such as trees, are more susceptible to inbreeding depression.

Heterosis is the advantageous effects of hybridization (introgression), and to some

degree, admixture, between two inbred lines with different deleterious genes. By crossing

such lines in a diploid species, the genetic load is dramatically decreased and the fitness of the

offspring increases in comparison to the parents. Many domestication efforts have utilized

this phenomenon (Fig. 10: Shepherd et al. 2005). If an offspring of two highly inbreed lines is

selfed, the offspring produced (assuming sufficient fecundity) will largely exhibit genotypic

proportions expected under Hardy–Weinberg equilibrium.

5.2. Coalescent Theory

The application of coalescence theory to genealogical relationships within species

provides information into the process of diversification and the influence of biogeography on

distributional patterns (Cruzan and Templeton, 2000). The coalescent theory is a retrospective

model of population genetics that traces all alleles of a gene from a population to a single

ancestral copy shared by all members of the population, known as the Most Recent Common

Ancestor (MRCA) (Donnelly and Tavaré, 1995). Under conditions of genetic drift alone,

every finite set of genes or alleles has a coalescent point at which all descendants converge to

a single ancestor (Cannings, 1974; Donnelly and Tavaré, 1995).



Selfed Outcrossed

Figure 10. The effect of inbreeding depression (left) and heterosis (right) on the vegetative propagation

of the Pinus elliottii x P. caribaea hybrid (Study and photo: R. Mellick).

Some properties of the coalescent theory are: the larger the number of samples (genes)

the greater the rate of coalescence (the more lineages there are the greater the probability that

two will coalesce), the larger the population size the slower the rate of coalescence, time to

coalescence gets longer as the process moves toward the MRCA, smaller sample sizes have a

high probability of including the MRCA of the population, and time cannot be measured

directly with genetic data but genetic divergence can. Considering the assumption of selective

neutrality (Kimura, 1968) it is possible to model the history of a sample, that is, without

regard to the rest of the population. Selection can be accommodated easily if it is strong,

while coalescent models of weak selection are more complicated. Chloroplast and nuclear

genes will coalesce at different rates due to the difference in inheritance between chloroplast

and nuclear DNA as a result of recombination.

The coalescent, as it is typically presented in population genetics, makes all the usual

assumptions of the Wright-Fisher model of a population (Donnelly and Tavaré, 1995; Fisher,

1930; Wright, 1931). As coalescent theory was developed, so were the within-species

molecular data sets resulting from the development and availability of molecular techniques

(Crandall and Templeton, 1993; Kingman, 2000).


Rohan Mellick 240

We now have access to Second Generation Sequencing or Next Generation Sequencing,

where potential exists to develop 1000s of informative markers from a single sequencing run.

Recently the resolution and sheer number of molecular markers available has increased

confidence of inferred molecular history through coalescent-based analyses. Coupled with

environmental models that can be projected onto historical climatic estimates an

interdisciplinary evolutionary history can be acquired for a species (Mellick et al., 2012,

2013b).

CASE STUDY TWO: PALAEODISTRIBUTION MODELING AND GENETIC

EVIDENCE HIGHLIGHT DIFFERENTIAL POST-GLACIAL RANGE

SHIFTS OF A RAINFOREST CONIFER DISTRIBUTED ACROSS

A LATITUDINAL GRADIENT

We examined range dynamics during the last glacial cycle of the late-successional

tropical rainforest conifer Podocarpus elatus using a combination of modeling and molecular

marker analyses. In particular, we tested whether distributional changes predicted by

environmental niche modeling are in agreement with the glacial maximum contractions

inferred from the southern fossil record, and population genetic-based estimates of range

disjunctions and demographic dynamics. Also, we tested whether northern and southern

ranges are likely to have experienced similar expansion/contraction dynamics. The study

location is the Eastern Australian tropical and subtropical rainforests (Mellick et al., 2012).

Environmental niche modeling was completed for three time periods during the last

glacial cycle and was interpreted in light of the available fossil record (Fig. 11; Mellick et al.,

2012). We collected 109 samples from 32 populations across the entire range of P. elatus. Six

microsatellite loci and Bayesian coalescence analysis were used to infer population

expansion/contraction dynamics, and five sequenced loci (one plastid and four nuclear) were

used to quantify genetic structure/diversity (Fig. 12; Mellick et al., 2012). Environmental

niche modeling showed that the northern and southern ranges of P. elatus experienced

different expansion/contraction dynamics. In the northern range, after persisting in a small

refugial area during the Last Glacial Maximum (LGM; 21 ka), the habitat suitable for P.

elatus expanded during the post-glacial period. Conversely in the south, suitable habitat was

widespread during the LGM but since has contracted. These differential dynamics were

supported by coalescent-based analysis of the population genetic data (northern dispersal) and

are consistent with the higher genetic diversity in the south compared to the north. A contact

zone between the two genetically divergent groups (corresponding to the Macleay Overlap

Zone) was supported by environmental niche modeling and molecular analyses (Fig. 11 and

12; Mellick et al., 2012).

The climatic changes of the Quaternary have differentially impacted the northern and

southern ranges of this broadly distributed rainforest tree in Australia. Recurrent

contraction/expansion cycles contributed to the genetic distinction between northern and

southern distributions of P. elatus (Mellick et al., 2012). By combining molecular and

environmental niche modeling evidence, this study questions the general assumption that

broadly distributed species respond in a uniform way to climate change.



Figure 11. Environmental niche models for Podocarpus elatus north (a, b and c), and south (d, e and f),

of the Clarence River Corridor biogeographic barrier in eastern Australia based on the Model for

Interdisciplinary Research on Climate (MIROC) global climatic model for 21 ka Last Glacial Maximum

(a and d), 6 ka Holocene Climatic Optimum (b and e) and the 0 ka pre-industrial (c and f) time periods.

Dark blue indicates low probability of occurrence and warmer colours indicate high probability of

occurrence (Source: Mellick et al., 2012).

Niche-modeling indicated (Fig. 11) that at the Last Glacial Maximum (21 ka), the

habitats suiting the two genetically differentiated regions of P. elatus were geographically

disjunct (Mellick et al., 2012). The northern distributional region persisted through the LGM

in a small refugial area, which during post-glacial periods has expanded. Conversely, the

southern range followed the opposite trend and has contracted since the LGM, but overall had

higher genetic diversity (Fig. 12). Coalescent-based analysis supported these differential

dynamics across the distribution of the species (Mellick et al., 2012).

In this study we coupled molecular (cpDNA and nDNA) and environmental niche

modeling (ENM) data to compare the distribution histories of two genetically differentiated

groups of the widespread Australian rainforest tree Podocarpus elatus. Molecular and ENM

results were congruent in suggesting that north/south divergence can be explained by

differential range-shift responses between these two genetic clusters separated by the

Clarence River Corridor. The differential range shift response of the species in northern

versus southern distributions is likely to be illustrative of differing climatic drivers between


Rohan Mellick 242

the southern subtropical and northern tropical rainforests. Stronger seasonality in the south

might have contributed in maintaining this divergence, and while the species occupies such a

broad environmental envelope, the degree of gene flow between the population groups has

been sufficient to maintain the species‟ integrity (Mellick et al., 2012).



Figure 12. Podocarpus elatus sequence haplotype distribution and haplotype networks for each of the

five loci (a–e) used in the study. The Clarence River Corridor (CRC: curved line) is a microsatellite

genetic boundary (K = 2), where the north includes both Macleay Overlap Zone and northern

populations. Pie charts are coloured according to the haplotypes present in each population, and are

proportional to the sample size, were the smallest circles represent n = 1 and the largest n = 7. The

networks are shaded according to region and are sized relative to abundance of each haplotype. The

proportion of each haplotype present in the southern populations (1–12) is white, the Macleay Overlap

Zone (13–25) is grey and in the northern populations (26–32) is black. Only two haplotypes were found

at locus A45 (b) and therefore no network was generated. Locus B37 (c) is a chloroplast locus and the

remaining loci are nuclear (Source: Mellick et al., 2012).

6. MOLECULAR ANALYSIS

6.1. Nucleic Markers Including Microsatellites

The flanking sequence of nuclear microsatellites (SSRs) has proved informative to

evolutionary studies (Rossetto et al., 2008, 2009). Inferring genealogical relationships from

SSR allelic size data alone is weak due to the mutational mechanisms involved (Estoup et al.,

2002) but the regions flanking the repeat itself provide a tractable source of phylogenetic

information that is well suited to cross-population and phylogeographic studies (Rossetto et

al. 2002; Hey et al., 2004). The fact that these flanking regions are conserved to a degree

necessary to allow for primer annealing and Polymerase Chain Reaction (PCR), yet more

variable as the sequence approaches the SSR (within the amplicon), give these markers an

advantage over other co-dominant marker systems. This is due to being able to score


Rohan Mellick 244

haplotypes from the flanking sequence, and in some circumstances link to repeat length

variation, i.e., HapSTR (Haplotype Short Tandem Repeat; Hey et al., 2004, Mellick et al.,

2013b).

Microsatellites, or Simple Sequence Repeats (SSRs), are polymorphic loci present in

nuclear DNA and organelle DNA that consist of repeating units of one to four base pairs in

length (Turnpenny and Ellard, 2005). They are co-dominant molecular markers that have

wide-ranging applications in the field of genetics, including kinship and population studies.

These markers often show high levels of inter- and intra-specific polymorphism, particularly

when tandem repeats number ten or greater (Queller et al., 1993). Microsatellites owe their

variability to an increased rate of mutation compared to other neutral regions of DNA. These

high rates of mutation can be explained most frequently by slipped strand mispairing

(slippage) during DNA replication on a single DNA strand. Mutation may also occur during

recombination during meiosis (Blouin et al., 1996). Some errors in slippage are rectified by

biomolecular proofreading mechanisms within the nucleus, but some mutations can escape

repair.

Commonly it is found that a number of nucleic microsatellite loci distinguish between

individuals of a rare or threatened organism. Microsatellite library construction is problematic

in conifers due to their large genome size and the high proportion of duplicated DNA (Elsik

and Williams, 2001; Scott et al., 1999; Scott et al., 2003). Nucleic microsatellite libraries

were developed for Podocarpus elatus (Almany et al., 2009) and P. lawrencii.

6.2. Plastid DNA Markers

Plastid analysis provides a deeper historical observation than nucleic DNA, due to not

undergoing recombination and being of a more conserved nature (Powell et al., 1995). This

conservation allows for historical events to be recorded as signatures in the DNA sequence.

Chloroplast genomes are predominantly maternally inherited in angiosperms and paternally

inherited in gymnosperms. They are powerful in determining gene flow from seed versus

pollen movement and reconstructing phylogeographic colonization (Petit et al., 2003).

Intergenic spacers and introns of chloroplast DNA are commonly used as markers in

phylogenetic studies focusing on species-level relationships in plants. In order to attain

informative variation at this low taxonomic level, a large number of nucleotide bases and,

therefore, a great many regions must be investigated. The great majority of loci investigated

for this project were plastid, and only one polymorphic locus was found (PeB37BGT:

Almany et al., 2009).

6.3. Cytoplasmic Inheritance in Podocarpus

The pattern of cytoplasmic inheritance in Podocarpus totara confirms a predominantly

paternal inheritance of plastids, common in all conifer families, and inheritance of

mitochondria is dependant on the mechanism of fertilization and the structure of male and

female gametes found in each family (Owens and Morris, 1990; Wilson and Owens, 2003).

Wilson and Owen (2003) inform that Podocarpaceae display a cytoplasmic inheritance

mechanism similar to that of the Pinaceae and Taxaceae. Bi-parental inheritance in



Podocarpus elatus is therefore likely, with cpDNA being paternally inherited and mtDNA

being maternally inherited. An extensive search for polymorphic mtDNA loci in the P. elatus

was warranted due to populations being clumped and maternal structure being likely but none

were found. Discovery of polymorphic plastid loci has improved since the advent of Next

Generation Sequencing and will serve gene discovery in organisms with large sized genomes

where paraology is likely.

6.4. Interdisciplinary Studies

Interdisciplinary studies that use multiple sources of information to infer past genetic

structure, such as molecular and palynological data, are commonly used (Petit et al., 2002,

2003). The methods are complementary, so that the limitations from one method are often

compensated for by evidence from the other method. For instance, the taxonomic resolution

using palynological tools is limited, whereas some insights on the dynamics within species

can only be obtained using molecular tools. Furthermore, genetic surveys can have a greater

spatial resolution yet fossil data provide greater temporal resolution (Petit et al., 2002, 2003).

Macro-fossil data in most taxa provides more taxonomic certainty, while palynology provides

much more fossil data but is subject to scrutiny when it comes to reliable species

identification.

Petit et al. (2002) identified refugia and post-glacial colonization routes of European

White Oaks (Quercus spp.) based on both chloroplast DNA (cpDNA) and fossil pollen

evidence. Geographic distribution of 32 cpDNA variants belonging to eight Quercus species

sampled from 2613 populations is presented by Petit et al. (2002). Clear-cut geographic

patterns were revealed in the distribution of cpDNA haplotypes. Combined with

palynological data, colonization routes out of glacial period refugia were able to be identified.

Integration of genetic and climatic data (Pease et al., 2009; Scoble and Lowe, 2010, Mellick

et al., 2011, 2012, 2013a, 2013b) with fossil data will further improve our understanding of

the underlying environmental factors that maintain genetic structure.

7. ENVIRONMENTAL NICHE MODELING

Environmental Niche Modeling (ENM) techniques endeavor to define a conceptual

envelope that best describes the limits to a species actual distribution by correlating

occurrence records to environmental variables. It is a process using computer algorithms to

predict the geographic distribution of species on the basis of a mathematical representation of

their known distribution in environmental space (Beaumont and Hughes, 2002; Heikkinen et

al., 2006). Environmental suitability is largely governed by climate in particular temperature

and precipitation gradients. Variables can be relatively constant, such as soil type, or dynamic

like climate and land use, so a prior knowledge of the study area and study system need to be

considered for reliable predictions.

The development of Geographic Information Systems (GIS) has allowed researchers to

investigate these environmental factors through ENM in detail. This has made it possible to

model the spatial patterns of climatic suitability for organisms based on their current


Rohan Mellick 246

distributions and to project these conditions onto mapped estimates of historical and future

climates. Allowing for inference of refugial areas based solely on environmental data alone.

Figure 13. Climate-based prediction and constraints on species distributions (Modified from source:

Cunningham, 2004).

Prediction of species‟ distributions is integral to various ecological, evolutionary and

conservation applications in science (Elith et al., 2006; Elith and Leathwick, 2009).

Applications include field collection, predicting demographic history and forecasting future

threats. Environmental Niche Modeling is now a common tool for assessing the potential

impact of climate change on species‟ ranges (Beaumont et al., 2008).

Many of the questions in phylogeographic studies pertain to how environmental variation

over space and time shape patterns of genetic divergence (Pease et al., 2009; Scoble and

Lowe, 2010). One of these questions is the identification of environmental barriers to

dispersal and gene flow. Environmental niche modeling can answer the question: Is

environmental variation causing range disjunction and reproductive isolation of fragmented

populations? This can be tested by determining genetic connectivity between populations

located in different environmental envelopes.

Modelers are often faced with a problem when the „fundamental niche‟ (potential

distribution) of a species does not correspond with a „realized niche‟ (actual distribution) and

for this reason either an error in the construction of the fundamental niche (the abiotic

thresholds for a species) has been made or that there are genuine factors affecting the actual

distribution of the species, such as competition, predation and pathogens (Fig. 13). This

discrepancy between the two often provides the basis of the more interesting hypotheses.



7.1. Landscape Genetics

Environmental and historical data are now being combined to understand geographic

patterns of genetic diversity in the emerging discipline „landscape genetics‟ (Scoble and

Lowe, 2010). It integrates population genetics, spatial statistics and landscape ecology. Just

coming into existence and beginning to display signs of future potential, landscape genetics

research explicitly quantifies the effects of landscape composition, configuration and matrix

quality on gene flow and spatial genetic variation (Storfer et al., 2007). The „matrix‟ can be

perceived as the hostile are between areas of increased habitat suitability and associated

fragmented habitats. Niche modeling has contributed to such efforts by allowing species–

environmental associations to be projected into the past so that hypotheses about historical

vicariance can be generated and tested independently with genetic data (Pease et al., 2009).

Recent episodes of environmentally mediated divergence are unlikely to be shown as a

molecular signature, however, utilizing environmental models associated with divergent

population groups reveals the history of genetic disjunctions and how climate and the

landscape drives evolution.

Speculation has been made that rainforest communities (Hilbert, et al., 2007;

VanDerWal, et al., 2009) and alliances (Mellick et al., 2013a) fluctuate independently in

response to climate depending on their species‟ specific ecology in the context of their abiotic

environment. Byrne et al. (2011) suggests that plants respond to Quaternary climatic change

as species, not as associations or biomes. Population isolation/migration and

vicariance/admixture processes over time, brought about by enduring glacial periods to the

brief interglacials like the present, add to the evolutionary complexity in fragmented highly

biodiverse areas.

Figure 14. Environmental niche models (altitudinal range shift) for Nothofagus moorei (blue) and

Elaeocarpus grandis (yellow) in far-eastern Australia (Mt Warning Caldera) for 21 thousand years ago

(fossil record support) during the Last Glacial Maximum (LGM) time period based on the Model for

Interdisciplinary Research on Climate (MIROC 3.2.2) global climatic model, the current time period

based on WorldClim data (1966–present) and an average of 13 global climatic models for 2050 (Mellick

et al., 2013a; Mellick et al., 2013b ). Darker shading indicates areas of higher elevation and lighter

shading indicates areas of lower elevation (Source: Mellick et al., 2013a).


Rohan Mellick 248

Figure 15. a. Boxplots of area of overlap at each time for 10 paired replicate models. Area of overlap was

measured as number of overlapping grid cells marked as core climate conditions for each species. b.

Histograms of frequency of altitude values in the core climate condition grid cells for each species.

Elaeocarpus grandis data is to the left and Nothofagus moorei to the right in each panel. Histograms of

altitude in 100 m bands were based on the core climate suitability grid cells for each species pooled

across the 10 replicates at each time. The dashed line on each plot indicates the mean altitude for each

species. c. The entire (northern and southern) Nothofagus moorei occurrences (YETI and Atlas of NSW

Wildlife), including genetic structure and diversity across whole range that comprises four major

highland population groups. Darker shading indicates areas of higher elevation and lighter shading

indicates areas of lower elevation (Source: Mellick et al., 2013a).

7.2. Past-Current-Future Distribution

Distributional changes facilitate adaptation and forge the evolutionary trends responsible

for incipient speciation. Palaeodistribution modeling projects the envelope created from

current occurrence records onto historical climatic estimates (Mellick et al., 2012). Used in

conjunction with molecular data the method combines independent data to locate glacial and

interglacial refugia (Hilbert et al., 2007). Recently research on the turnover along altitudinal

gradients from cool-temperate (Nothofagus spp.) to warm subtropical rainforest (Elaeocarpus

spp.) in far-eastern Australia has revealed post-glacial trends (Mellick et al., 2013a).

The potential distributions of the two species closely associated with different rainforest

types were modeled to infer the potential contribution of post-glacial warming on spatial

distribution and altitudinal range shift (Mellick et al., 2013a: Fig. 14 and 15). Environmental

niche models were used to infer range shift differences between the two species in the past

(21 thousand years ago), current and future (2050) scenarios, and to provide a framework to

explain observed genetic diversity/structure (Fig. 15a, b and c). The models suggest

continuing contraction of the highland cool temperate climatic envelope and expansion of the

lowland warm subtropical envelope, with both showing a core average increase in elevation

in response to post-glacial warming (Figure 15b).



7.3. Predicting Future Distribution of Genetic Diversity

Modeling and monitoring studies over the last 20 years have provided considerable

evidence that global climate change is already affecting and will continue to affect many

species and ecosystems, resulting in declines and extinctions of many species (Dunlop and

Brown, 2008). It is usually the most threatened species and habitats that are going to respond

first to these climatic changes. Nonetheless, because of the interacting nature of ecological

systems the impacts on biodiversity to future environmental change are not clear.

We know that adaptive potential is associated with genetic diversity and that distributions

follow environmental gradients. Through modeling both molecular and environmental data,

we can anticipate future spatial changes in genetic diversity. Recently the resolution and sheer

number of molecular markers available has increased confidence of inferred molecular

history through coalescent-based analyses. Coupled with environmental models that can be

projected onto future climatic estimates an interdisciplinary forecast can be acquired for a

species. Next Generation Sequencing will provide additional tools to reliably predict future

changes in genetic diversity and therefore the adaptive potential of species to the rapid

environmental changes anticipated this century.

Table 1. Details of the 13 atmospheric oceanic global circulation models (AOGCMs)

used to create an ensemble (average) for projecting MaxEnt models onto 2050 future

climate conditions (Source: Mellick et al., 2013b)

Climate Model Inter-comparison

Project 3(CMIP3) identifier

Originating institution or collective

BCCR-BCM2.0 Bjerknes Centre for Climate Research, Norway

CGCM3.1(T63) Canadian Centre for Climate Modeling and Analysis, Canada

CNRM-CM3 Météo-France / Centre National de Recherches

Météorologiques, France

CSIRO-Mk3.5 CSIRO Atmospheric Research, Australia

ECHAM5/MPI-OM Max Planck Institute for Meteorology, Germany

ECHO-G Meteorological Institute of the University of Bonn,

Germany; Meteorological

Research Institute of the Korean Meteorological Agency,

Korea

GFDL-CM2.1 Geophysical Fluid Dynamics Laboratory, NOAA, Dept. of

Commerce USA

GISS-ER Goddard Institute for Space Studies NASA USA

INGV-SXG Instituto Nazionale di Geofisica e Vulcanologia, Italy

INM-CM3.0 Institute for Numerical Mathematics, Russia

IPSL-CM4 Institut Pierre Simon Laplace, France

MIROC3.2(medres) Center for Climate System Research (The University of

Tokyo), National Institute for Environmental Studies, and

Frontier Research Center for Global Change (JAMSTEC),

Japan

MRI-CGCM2.3.2 Meteorological Research Institute, Japan


Rohan Mellick 250

Studies have supported the prediction that species with narrower climatic niches will be

more sensitive to climate-induced range contraction (Sexton et al., 2009; Bell, 2007).

Potential impacts of projected climate change on biodiversity are often assessed using single-

species Environmental Niche Models (ENM) (Heikkinen et al., 2006). Predicting the

probability of successful establishment of plant species by correlating distribution to

climatic/environmental variables, has considerable potential as supporting methodology to

reinforce palynological and phylogeographic studies. In the future, deeper ecological and

physiological understanding may allow competition effects to be quantified and included in

species distribution modeling. Future biogeographical ranges of species are constructed by

applying the models based on climatic variables that best describe the current equilibrium

distributions to simulate future distributions under selected climate changes scenarios

(Bakkenes et al., 2002; Solomon et al., 2007). Bakkenes (2002) suggests that in reviewing

possible future trends, it was found that European plant species, in general, would find their

current climate envelopes further northeast by 2050, implying maximum temperature

threshold may restrict species‟ lower latitude range limits.

Araujo and New (2007) suggests added confidence in ENM output can be achieved

through projection of the trained environmental envelope onto an ensemble of different

atmospheric oceanic global circulation models (AOGCMs). The selection of climate

scenarios for impact assessments should not be undertaken arbitrarily - strengths and

weakness of different climate models should be considered (Beaumont et al., 2008). To

account for uncertainties in predictions of future climate data, we derived data from 13 global

climate models (Table 1: Mellick et al., 2013a, 2013b) used in the Fourth Assessment Report

of the Inter-governmental Panel on Climate Change (IPCC).

7.4. Correlative and Mechanistic Modeling

Environmental niche models are correlative (statistical) models. They relate observed

presences of a species to values of environmental variables across the geographic extent of

the species‟ range. Some models use absence data, which is very hard and expensive to

collect but the most commonly used models use presence-only data, perhaps together with

random background data. In contrast, mechanistic (or process-based) models assess the bio-

physiological aspects of a species to generate the conditions in which the species can ideally

survive, based on a species‟ traits and observations made in laboratory or controlled field

studies (Bresson et al., 2011). Mechanistic modeling defines the abiotic limits of a species

range through the physiological tolerances of the species, e.g., Podocarpus auxiliary xylem

tissue collapses at certain water-stress thresholds and therefore the species distribution is

limited by a trade-off between water use and photosynthetic efficiency (Brodribb and Hill,

1999).

It may be possible to combine mechanistic and correlative approaches to alleviate the

expense required for mechanistic data collection and gain the convenience of correlative

modeling. The development of phenotype-genotype correlation between the physiological

traits associated with climatic tolerance and a marker suite linked to such characters may

allow this. Possibly, selective genotyping (i.e., genotyping the best and worse individuals of

physiological quantitative traits), candidate gene analysis and/or quantitative trait loci

analysis could achieve the phenotype-genotype correlation required. Then the correlative



modeling approach could be applied to the geographic distribution of markers associated with

such traits.

Figure 16. The distribution of Podocarpus elatus showing genetically differentiated population groups;

southern group is red and northern is green (Mellick et al., 2011). The Clarence River Corridor dividing

the groups is shown. The three populations used for the coalescent-based analysis are shown and the

sizes of their representative circles are relative to allelic richness (Rs) x unbiased heterozygosity (uh)

(Source: Mellick et al., 2013b).


Rohan Mellick 252

CASE STUDY THREE: INTRASPECIFIC DIVERGENCE ASSOCIATED

WITH A BIOGEOGRAPHIC BARRIER AND CLIMATIC MODELS SHOW

FUTURE THREATS AND LONG-TERM DECLINE OF

A RAINFOREST CONIFER

A capacity to foresee the shift in species‟ range and the demographic response to future

climate change is integral to effective conservation planning (Mellick et al., 2013b). Here we

model the future climate-driven range shift, and compare it with past range shift, along a

latitudinal gradient in two population groups of a late-successional rainforest conifer

(Podocarpus elatus), genetically differentiated over the Clarence River Corridor

biogeographic barrier (Northern NSW, East Australian Rainforests). Environmental niche

modeling of the past-current-future distributions of the two groups and a coalescent-based

isolation-with-migration model investigated divergence times and effective population sizes

among the current genetic disjunctions in the species. This suggests differential range shift

(i.e., expansion in the north and contraction in the south) will continue in the future, with a

southern range shift also occurring in both climatic models. The origin of the Clarence River

Corridor (Fig. 16) dividing the two population groups was inferred by molecular analysis to

be prior to the last glacial maximum (Table 2). Another divergence in the south (19 ka) is

indicative of slow consistent habitat contractions since the last glacial maximum (LGM: 21

ka). We recommend the southern and Macleay Overlap Zone (far-eastern Australia)

populations as priority areas for protection based upon intraspecific diversity and past-

current-future habitat suitability. The integrated approach shows that this widely distributed

species is more at risk than expected from current climate change and other anthropogenic

effects (Mellick et al., 2013b).

Past-current-future modeling (Table 2 and Fig. 17) suggests that the two population

groups respond differently to climate change, and that both are subjected to

expansion/contractions cycles indicative of community turnover (Mellick et al., 2013a;

Kershaw et al., 2007a). This emphasizes the sensitivity of this species to climatic fluctuation

and, when considering its fragmented distribution along a broad latitudinal range, the

likelihood that this species may survive climatic cycles within dispersed refugia and

microhabitats (e.g., ecotones, ridgelines, rainforest remnants; Mellick et al., 2013b; Brodribb

and Hill, 2003).

Although the species has long generation periods, possibly of 600 years or more,

successful recruitment and establishment are reliant on abiotic factors (e.g., fire frequency

and climate) as well as biotic factors (e.g., competition and light availability). Observed

spatial displacement to boundary communities (Mellick et al., 2011, Harden et al., 2006)

away from the increased competition of core communities (Lusk, 1996) may have facilitated

the survival of populations in the past, yet may now expose P. elatus to increased

anthropogenic disturbance, such as more frequent burning and non-indigenous invasive

species (Mellick et al., 2013b).

The nature of these ecotonal communities (e.g., nutrient poor areas, fire prone, less

competition) affords Podocarpus an advantage (Brodribb and Hill, 2003; Lusk, 1996). The

micro-environmental character and the sheltered ecology of ecotonal micro-habitats may have

facilitated survival in areas of low climatic suitability (Mellick et al., 2013b). These

fragmented habitats considering the outcrossing nature of the tree are especially vulnerable to



the effects of isolation, genetic drift and inbreeding (Mellick et al., 2011) as shown in other

Australian rainforest conifers (Peakall et al., 2003, Shapcott, 1997). Although Podocarpus

elatus is common to the East Australian rainforests, its existence is closely allied to habitats

that are threatened by future climate change. Seeing localized extinction and population

decline has been documented in other rainforest podocarps (Shapcott, 1997) and drier

rainforest conifers (Peakall et al., 2003; Pye and Gadek, 2004), we should consider P. elatus

as potentially at future risk (Mellick et al., 2013b).

Table 2. The inferred genetic history of three Podocarpus elatus populations

representing the southern [1. southern New South Wales (Sth NSW) and 2. northern

New South Wales (Nth NSW)] and northern [3. southern Queensland (Sth QLD)] ranges

either side of the Clarence River Corridor (CRC) biogeographic barrier

DATA SET Sth NSW ζ1 Nth NSW ζ2 Sth QLD ζ3 NSW ancestral ζ4 All ancestral ζ5 t1 t2

MLE 0.1825 0.1673 0.5655 0.888 34.04 0.0475 0.0095

HPD95 Low 0.0075 0.01425 0.1905 0 21.8 0.0175 0.0015

HPD95 High 2.918 0.7748 1.556 12.11 54.44 0.1475 0.0435

rescaled population size (ζ: individuals) and time (t:1000 years) parameter units

MLE 938.6 837.3 2831 4443 170406 105.127 19.023

HPD95 Low 37.55 71.34 953.7 0 109132 35.042 3.004

HPD95 High 14605 3878 7787 60636 272530 295.357 87.105

The 95% highest posterior density intervals (HPD95 low – HPD95 high) are tabulated for effective

population sizes (ζ1- ζ5) and splitting times (t1; first split either side of the CRC between Sth QLD

and NSW ancestral populations: t2; second split between Sth and Nth NSW). The maximum

likelihood estimate (MLE) is the curve height (i.e., mode) of marginal posterior probability for

each parameter. The difference between the original parameter values (i.e., curve height of

marginal posterior probability) and the rescaled population size and time parameter values are that

the latter use marginal distribution values in demographic units. None of the migration parameters

converged and were left out of the table accordingly (Source: Mellick et al., 2013b)

CONCLUSION

The case studies underpinning this chapter aim to understand how the distribution of

Podocarpus elatus has responded to recent climatic change of the Quaternary and the likely

response to future climate change. Using the fossil record, molecular observation/inference

and Environmental Niche Modeling (ENM), an interdisciplinary signal has been obtained of

how past climate change has affected distribution and diversity of this rainforest conifer.

Seeing a common genetic and environmental boundary has been observed (Fig. 5), future

changes in the distribution of genetic diversity have been predicted (Fig. 16).


Rohan Mellick 254

Figure 17. Environmental niche models for Podocarpus elatus north (a, b, c and d) and south (e, f, g and

h) of the Clarence River Corridor biogeographic barrier in eastern Australia for 21 ka Last Glacial

Maximum (a and e), 6 ka Holocene Climatic Optimum (b and f), 0 ka Pre-industrial (c and g), including

the location of the three populations under study, and A2 2050 future time periods (d and h). Red borders

around projected distributions do not reflect probability of occurrence (Source: Mellick et al., 2013b).

Agreement between the climatic data (i.e., ENMs: Fig. 5, 11 and 16) and regional

distribution either side of the Clarence River Corridor (CRC) suggests genetic divergence is

climate-driven, as the decline in the observed fossil record suggests (Black et al., 2006; Black

and Mooney, 2007; Kershaw et al., 2007; Shimeld, 1995, 2004; Williams et al., 2006).

Rainforest herpetofauna (Burns et al., 2007; Schauble and Moritz 2001) show genetic

divergences over the CRC that suggest the differential climatic drivers identified in P. elatus

may support broader rainforest associations.

Due to the study being based on separate independent sources of evidence, the

conclusions made are reliable and will provide a basis to interpret the environmental drivers



and biological processes that have led to present-day genetic diversity and structure. The

method used here to reconstruct community turnover has been tailored to the inadequacies of

the east Australia Quaternary fossil record (i.e., abundance, chronology and taxonomy).

Interdisciplinarity will improve our theoretical understanding of the response of threatened

communities to climate change as well as informing future conservation strategies.

The majority of the genetic diversity (microsatellite, nDNA and cpDNA sequence) in P.

elatus is found in the southern and central ranges. The northern distribution is impoverished

and very fragmented. The species is rare in the Australian Wet Tropics (Fig. 16). The lower

diversity and distributional pattern of the northern populations could suggest that the current

distribution is the result of an earlier northern expansion from a narrower genetic pool (case

study one).

Podocarpus have traits that originated in extreme seasonality (Brodribb and Hill, 1999,

2003; Quinn and Price, 2003; Hill, 1994) and since have served expansion into equatorial

regions (Fig. 2). The surviving podocarps are remnants of a world dramatically different from

the one they occupy today. Although past Antarctic climates were seasonally hot and wet, the

change in day length for the genus over time has been dramatic.

The current wide latitudinal range of the species is linked to the availability of rainforest

boundary communities. This has given P. elatus a continental wide range yet a narrow

microhabitat requirement. Over time population expansion and contraction cycles have been

influence by climate. Differential range shift response is shown between the northern and

southern population groups (Fig. 11 and 17) illustrative of differing climatic drivers between

the southern subtropical and northern tropical rainforests. These differing climatic conditions

may underline long-term divergent processes between the two regional groups. In the shorter

term and at a regional scale, anthropogenic habitat clearing (i.e., Big scrub Rainforest) is also

likely to contribute to the differences measured between regions and among generational

cohorts (Fig. 4).

The location and aridity of the Clarence River Corridor would have restricted dispersal

during expansion phases of each latitudinally distributed region. These successional waves of

diversity in opposing directions resulted in the populations flanking the barrier to be

genetically distinct and of high diversity, with the majority of unique allele genotypes found

close to the barrier (Fig. 12). Due to the huge climatic variation across its current range,

periods of admixture (expansion) and vicariance (contraction) between populations have been

different between southern and northern ranges. Differential range shift response shown in P.

elatus may be caused by stronger seasonality in the south. Although the species occupies such

a broad environmental envelope, the degree of gene flow between the population groups has

been sufficient to maintain the species‟ integrity (case study two).

The future climate-induced range shift (Fig. 16 d and h) of the two genetically

differentiated regions were shown to follow a similar pattern to that observed during the

Holocene Climatic Optimum (6 ka: Fig. 11b and e, Fig. 17 b and f) when the climate was

hotter and wetter. Expansion/contraction dynamics suggest the southern diverse region is

under threat of future climate change and sea-level rises expected this century (case study

three).

Locally adapted genetic variants may be harbored by threatened populations and these

variants may offer genetic variation that may reduce the genetic load of other inbreed

populations (Mellick et al., 2011). Conservation strategies may involve the extension of

habitat corridors to accommodate future range-shift and assisted migration of genetically rich


Rohan Mellick 256

stock under threat of localized extinction into areas of high habitat suitability (Mellick et al.,

2013b).

Climatic change has made naturally disjunct populations of Podocarpus elatus

genetically inbreed and divergent from one another, suggesting local genetic seed should be

used in the restoration of degraded habitats. The approach of using locally sourced seed to

revegetate areas may be appropriate for areas of relatively high diversity but where

genetically impoverished, success will depend on seed viability and, once established, on

resilience to forecasted changes and further disturbance. We need to promote admixture

between divergent population groups to reduce genetic load and reduce the adverse effects of

inbreeding. If natural selective gradients are in place then this local increase in diversity will

not hinder selective filtering and the natural processes present in the species, rather it would

aid it through survival.

It was found that P. elatus is considerably more threatened than shown by its current

distribution, and I suggest the use, and extension, of habitat corridors to accommodate future

climate-induced range shift of fragmented rainforest habitats along the east coast of Australia.

The future design of habitat corridors could take on a broader evolutionary application by

linking fragmented habitats along predicted avenues of range shift, and understanding natural

gene flow patterns will allow for genotype selection and assisted migration along these

avenues. Range shift is part of the natural adaptive process (Moritz, 2002) but during previous

climate fluctuations, range shift has been unhindered by human land-use. In order to conserve

the natural genetic constitution and adaptive potential of species in general, protected avenues

along which migration is assisted will need to be integrated into the human land-use matrix.

Our data suggest that conservation and management should be focused around the

southern populations and Macleay Overlap Zone; where most of the species‟ diversity resides

and where sustained habitat suitability in response to post-glacial warming and future

scenarios occurs (Fig. 17 d and h). The southern populations (Fig. 17h) are under threat of

localized extinction, and therefore should be a target of conservation strategies. The general

decline of rainforest conifers from the fossil record is a consequence of a changing

environment and has occurred since the previous interglacial (130 ka: Kershaw, 2005; et al.,

2007a; Shimeld, 1995, 2004).

Predictive distributional modeling and the understanding of gene flow dynamics provide

a method to interpret current distribution patterns and potentially anticipate, and

accommodate, rapid migration rates as a result of projected anthropogenic-induced climate

change.

Species worldwide are a result of natural evolution that has afforded them a genetic

constitution and an ability to survive natural climatic cycles and environmental extremes. The

consequences of our interference (artificial selection through anthropogenic effects) may be a

reduction in the adaptive potential and natural genetic constitution to deal with natural

climatic cycles and environmental pressures.

Complex evolutionary processes occur in the World Heritage listed central east

Australian Rainforests, the majority of which are surrounded by a human land use matrix with

little means of altitudinal or latitudinal range shifts in response to anthropogenic-induced

climate change. Understanding the manner of intraspecific divergence and adaptation along

latitudinal temperature/precipitation gradients within these confined rainforests is integral to

their conservation.



ACKNOWLEDGMENTS

Thank you to Tim Brodribb, Robert S. Hill, Andrew Lowe, Ed Biffin, Chris Allen, Peter

D. Wilson and Maurizio Rossetto. I acknowledge the University of Adelaide, Faculty of

Science, for funding the research (Divisional Scholarship). This research was also funded by

the Australian Research Council Discovery Grant (DP0665859). Thank you to the National

Herbarium of NSW, The Royal Botanic Gardens and Domain Trust, Sydney. I particularly

thank Andrew Ford (CSIRO, Atherton), Rebecca Johnson, Bob Coveny, David Bateman,

Phillip Greenwood, NSW National Parks and Wildlife and Robert Kooyman for their

assistance in undertaking collections and observations; Simon Ho (The University of Sydney)

and Paul Rymer (The University of Western Sydney) for their support with regard to multi-

locus coalescence-based analysis; Peter Kershaw, Peter Shimeld, Sandy Harrison, Scott

Mooney and Nicole Williams for providing palynological data.

REFERENCES

Ahmed AK, Johnson KA (2000) TURNER REVIEW No. 3. Horticultural development of

Australian native edible plants. Australian Journal of Botany 48, 417-426.

Almany GR, De Arruda MP, Arthofer W (2009). Permanent Genetic Resources added to

Molecular Ecology Resources Database 1 May 2009-31 July 2009. Molecular Ecology

Resources 9, 1460-1466.

Araujo MB, New M (2007). Ensemble forecasting of species distributions. Trends in Ecology

and Evolution 22(1): 42-47.

Avise JC (2000). 'Phylogeography: the history and formation of species.' (Harvard University

Press)

Avise JC, Arnold J, Ball RM Jr., Bermingham E, Lamb T, Neigel JE, Reed CA, Saunders NC

(1987). Intraspecific phylogeography: the mitochondrial DNA bridge between population

genetics and systematics. Annual Review of Ecology and Systematics 18, 489–522.

Bakkenes M, Alkemade JRM, Ihle F, Leemans R, Latour JB (2002). Assessing effects of

forecasted climate change on the diversity and distribution of European higher plants for

2050. Global Change Biology 8, 390-407.

Barker NP, Muller EM, Mill RR (2004). A yellowwood by any other name: molecular

systematics and the taxonomy of Podocarpus and the Podocarpaceae in southern Africa.

South African Journal of Science 100, 629-632.

Barker PF, Thomas E (2004). Origin, signature and palaeoclimatic influence of the Antarctic

Circumpolar Current. Earth-Science Reviews 66, 143-162.

Bartholomai A, Molnar RE (1981). Muttaburrasaurus, a new iguanodontid (Ornithischia:

Ornithopoda) dinosaur from the Lower Cretaceous of Queensland. Memoirs of the

Queensland Museum 20, 319-349.

Beaumont LJ, Hughes L (2002). Potential changes in the distributions of latitudinally

restricted Australian butterfly species in response to climate change. Global Change

Biology 8, 954-971.

Beaumont LJ, Hughes L, Pitman AJ (2008). Why is the choice of future climate scenarios for

species distribution modeling important? Ecology Letters 11, 1135-1146.


Rohan Mellick 258

Bell KL, Moritz, C., Moussalli, A. and Yeates, D.K. (2007). Comparative phylogeography

and speciation of dung beetles from the Australian Wet Tropics rainforest. Molecular

Ecology 16, 4984 -4998.

Bennett KD (1997). 'Evolution and Ecology: The pace of life.' (Cambridge University Press:

Cambridge, U.K.)

Biffin E, Brodribb TJ, Hill RS, Thomas P, Lowe AJ (2012). Leaf evolution in Southern

Hemisphere conifers tracks the angiosperm ecological radiation. Proceedings of the

Royal Society B: Biological Sciences.

Biffin E, Conran J, Lowe A (2011). Podocarp Evolution: A Molecular Phylogenetic

Perspective. In „Ecology of the Podocarpaceae in Tropical Forests‟. (Eds BL Turner and

LA Cernusak) pp. 1–20. Smithsonian Contributions to Botany, No. 95. (Smithsonian

Institution Scholarly Press: Washington D.C.)

Black MP, Mooney SD (2007). The response of aboriginal burning practices to population

levels and El Nino-Southern oscillation events during the mid- to late-Holocene: a case

study from the Sydney basin using charcoal and pollen analysis. Austral. Geo. 38, 37-52

Black, M.P., Mooney, S.D., Martin, H.A. 2006. A >43,000-year vegetation and fire history

from Lake Baraba, New South Wales, Australia. Quaternary Science Review 25, 3003-

3016.

Blouin MS, Parsons M, Lacaille V, Lotz S (1996). Use of microsatellite loci to classify

individuals by relatedness. Molecular Ecology 5, 393-401.

Bond WJ (1989). The tortoise and the hare - ecology of angiosperm dominance and

gymnosperm persistence. Biological Journal of the Linnean Society 36, 227-249.

Bowler JM, (1982). Aridity in the late tertiary and quaternary of Australia. In: „Evolution of

the flora and fauna of arid Australia‟. Barker WR and Greenslade PJM (eds) (Peacock

Publications, South Australia).

Bowman D (2000) 'Australian Rainforests: Islands of Green in the Land of Fire.' Cambridge

University Press, Cambridge.

Bresson, CC, Vitasse Y, Kremer A, Delzon S (2011). To what extent is altitudinal variation

of functional traits driven by genetic adaptation in European oak and beech? Tree

Physiology 31, 1164-1174.

Brewer S, Cheddadi R, de Beaulieu JL, Reille M (2002). The spread of deciduous Quercus

throughout Europe since the last glacial period. Forest ecology and management 156, 27-

48.

Brodribb T, Hill RS (1999). The importance of xylem constraints in the distribution of conifer

species. New Phytologist 143, 365-372.

Brodribb TJ, Hill RS (2003). The rise and fall of the Podocarpaceae in Australia: a

physiological explanation. In 'Evolution of plant physiology'. (Eds A Hemsley and I

Poole) pp. 381-399. (Academic: London)

Burns EL, Eldridge MDB, Crayn DM, Houlden BA (2007). Low phylogeographic structure

in a wide spread endangered Australian frog Litoria aurea (Anura : Hylidae).

Conservation Genetics 8, 17-32.

Byrne M (2007). Phylogeography provides an evolutionary context for the conservation of a

diverse and ancient flora. Australian Journal of Botany 55, 316-325.

Byrne M, Steane DA, Joseph L, Yeates DK, Jordan GJ, Crayn D, Aplin K, Cantrill DJ, Cook

LG, Crisp MD, Scott Keogh J, Melville J, Moritz C, Porch N, Sniderman K, Sunnacks P,



Weston PH (2011) Decline of a biome: evolution, contraction, fragmentation, extinction

and invasion of the Australian mesic zone biota. Journal of Biogeography 38, 1635-1656.

Cannings C (1974). The Latent Roots of Certain Markov Chains Arising in Genetics: A New

Approach, I. Haploid Models. Advances in Applied Probability 6, 260-290.

Chambers TC, Drinnan AN, McLoughlin S (1998). Some morphological features of Wollemi

Pine (Wollemia nobilis: Araucariaceae) and their comparison to cretaceous plant fossils.

International Journal of Plant Sciences 159, 160-171.

Chaw SM, Zharkikh A, Sung HM, Lau TC, Li WH (1997). Molecular phylogeny of extant

Gymnosprems and seed plant evolution: analyisi of nuclear 18s rRNA sequences.

Molecular Biology and Evolution 14, 56-68.

Conran JG, Wood GM, Martin PG, Dowd JM, Quinn CJ, Gadek PA, Price RA (2000).

Generic relationships within and between the gymnosperm families Podocarpaceae and

Phyllocladaceae based on an analysis of the chloroplast gene rbcL. Australian Journal of

Botany 48, 715-724.

Couper RA (1960). Southern Hemisphere Mesozoic and Tertiary Podocarpaceae and

Fagaceae and their palaeogeographic significance. Proceedings of the Royal Society of

London. Series B, Biological Sciences 152, 491-500.

Crandall KA, Templeton AR (1993). Empirical tests of some predictions from the Coalescent

Theory with applications to intraspecific Phylogeny reconstruction. Genetics 134, 959-

969.

Cruzan MB, Templeton AR (2000). Paleoecology and coalescence: phylogeographic analysis

of hypotheses from the fossil record. Trends in Ecology and Evolution 15, 491-496.

Cunningham S (2004). Climatic impacts on biodiversity: how useful are current predictions.

CSIRO entomology presentation.

Demesure B, Sodzi N, Petit RJ (1995). A set of universal primers for amplification of

polymorphic noncoding regions of mitochondrial and chloroplast DNA in plants

Molecular Ecology 4, 129-131.

Dettmann ME, Jarzen DM (1990). The Antarctic/Australian rift valley: Late cretaceous cradle

of nortteastern Australasian relicts? Review of Palaeobotany and Palynology 65: 131-

144.

Donnelly P, Tavaré S (1995). Coalescents and genealogical structure under neutrality. Annual

Review of Genetics 29, 401-421.

Doyle J (1945). Development lines in pollination mechanisms in the Coniferales. Scientific

proceedings of the Royal Dublin Society 24, 43-62.

Drinnan AN, Chambers TC (1986). Flora of the Lower Cretaceous Koonwarra Fossil Bed

(korumburra Group) south Gippsland. T. C. Mem. Ass. Australas. Palaeontols 3, 1−77.

Dunlop M, Brown PR (2008). ' Implications of climate change for Australia's National

Reserve System: a preliminary assessment. Technical Report.' Department of Climate

Change, Canberra.

Elith J, Graham CH, et al. (2006) Novel methods improve prediction of species' distributions

from occurrence data. Ecography 29, 129-151.

Elith J, Leathwick JR (2009). Species Distribution Models: ecological explanation and

prediction across space and time. In 'Annual Review of Ecology Evolution and

Systematics' pp. 677-697. (Annual Reviews: Palo Alto)

Elsik CG, Williams CG (2001). Low-copy microsatellite recovery from a conifer genome.

TAG Theoretical and Applied Genetics 103, 1189-1195.


Rohan Mellick 260

Emigh TH (1980). A comparison of tests for Hardy-Weinberg Equilibrium. Biometrics 36,

627-642.

Enright NJ, Hill RS (1995). 'Ecology of the southern conifers.' (Melbourne University Press.:

Melbourne)

Enright NJ, Jaffr T (2011). Ecology and Distribution of the Malesian Podocarps. In „Ecology

of the Podocarpaceae in Tropical Forests‟. (Eds BL Turner and LA

Cernusak) pp. 57–77. Smithsonian Contributions to Botany, No. 95. (Smithsonian Institution

Scholarly Press: Washington D.C.)

Enright NJ, Ogden J (1995). The southern conifers- a synthesis. In 'Ecology of the southern

conifers'. (Eds NJ Enright and RS Hill) pp. 271-287. (Melbourne University Press:

Melbourne )

Estoup A, Jarne P, Cornuet JM (2002). Homoplasy and mutation model at microsatellite loci

and their consequences for population genetics analysis. Molecular Ecology 11, 1591-

1604.

Ewens W (1990). Population genetics theory - the past and the future. In 'Mathematical and

Statistical developments of Evolutionary Theory'. (Ed. S Lessard) pp. 177-227. (Kluwer

Academic Publishers: New York)

Falconer DS, Mackay TFC (1996). 'Introduction to quantitative genetics.' (Addison Wesley

Longman Limitied: Harlow)

Fisher RA (1930). 'The genetical theory of natural selection.' (Clarendon Press: Oxford)

Greenwood DR. Christophel DC (2005). The origins and tertiary history of Australian

Tropical Rainforests. Tropical Rainforests: past, present and future. D. a. M.

Bermingham.

Haberle SG (2005). A 23,000-yr pollen record from Lake Euramoo, Wet Tropics of NE

Queensland, Australia. Quaternary Research 64, 343-356.

Harden GJ (1990). 'Flora of New South Wales.' (University of NSW press: Sydney)

Hartl DL, Clark AG (1997). 'Principals of Population Genetics (3rd ed).' (Sinauer Associates,

Inc: MA)

Heikkinen RK, Luoto M, Araujo MB, Virkkala R, Thuiller W, Sykes MT (2006). Methods

and uncertainties in bioclimatic envelope modeling under climate change. Progress in

Physical Geography 30, 751-777.

Hewitt G (2000). The genetic legacy of the Quaternary ice ages. Nature 405, 907-913.

Hey J, Won YJ, Sivasundar A, Nielsen R, Markert JA (2004). Using nuclear haplotypes with

microsatellites to study gene flow between recently separated Cichlid species. Molecular

Ecology 13, 909-919.

Hilbert D, Graham A, Hopkins M (2007). Glacial and interglacial refugia within a long-term

rainforest refugium: The wet tropics bioregion of NE Queensland, Australia.

Palaeogeography Palaeoclimatology Palaeoecology 251(1): 104-118.

Hill RS (1994). The history of selected Australian taxa. In 'History of the Australian

Vegetation: Cretaceous to Recent '. (Ed. RS Hill) pp. 390-419. (Cambridge University

Press: Cambridge)

Hill RS (2004). Origins of the southeastern Australian vegetation. Philosophical Transactions

of the Royal Society of London Series B-Biological Sciences 359, 1537-1549.

Hill RS, Brodribb TJ (1999). Turner Review No. 2 - Southern conifers in time and space.

Australian Journal of Botany 47, 639-696.



Hill RS, Pole MS (1992). Leaf and shoot morphology of extant Afrocarpus, Nageia and

Retrophyllum (Podocarpaceae) species, and species with similar leaf arrangement, from

tertiary sediments in Australia. Australian Systematic Botany 5, 337-358.

Hope GS, Kershaw P, Kenyon C, McKenzie M, Williams N (2010). Late Quaternary fire

regimes of Australasia. Quaternary Science Reviews 30, 28-46.

Hope GS (1994). Quaternary vegetation. In 'History of the Australian Vegetation: Cretaceous

to Recent'. (Ed. RS Hill). (Cambridge University Press: Cambridge)

Jordan GJ, Carpenter RJ, Bannister JM, Lee DE, Mildenhall DC, Hill RS (2011). High

conifer diversity in Oligo-Miocene New Zealand. Australian Systematic Botany 24, 121-

136.

Kelch DG (1997). The phylogeny of the podocarpaceae based on morphological evidence.

Systematic Botany 22, 113-131.

Kershaw AP, van der Kaars S, Flenley JF (2007b). The Quaternary history of far eastern

rainforests. In 'Tropical Rainforest Responses to Climate Change'. (Eds MB Bush and JF

Flenley) pp. 77-115. (Springer-Praxis: Berlin)

Kershaw AP, Bretherton SC, van der Kaars S (2007a). A complete pollen record of the last

230 Ka from Lynch's Crater, north-eastern Australia. Palaeogeography

Palaeoclimatology Palaeoecology 251, 23-45.

Kershaw AP, Moss PT, Wild R (2005). Patterns and Causes of Vegetation Change in the

Australian Wet Tropics Region over the last 10 Million Years. In 'Tropical Rainforests:

past, present and future'. (Ed. DaM Bermingham))

Kershaw AP, Martin HA, McEwen Mason JRC (1994). The Neogene: a period of transition.

In 'History of the Australian Vegetation: Cretaceous to Recent'. (Ed. RS Hill) pp. 299-

328. Cambridge University Press, Cambridge.

Kimura M (1968). Evolutionary Rate at the Molecular Level. Nature 217, 624-626.

Kingman JFC (2000). Origins of the Coalescent: 1974Â–1982. Genetics 156, 1461-1463.

Knowles LL, Maddison WP (2002). Statistical phylogeography. Molecular Ecology 11(12):

2623-2635.

Ledru MP, Salatino MLF, Ceccantini G, Salatino A, Pinheiro F, Pintaud JC (2007). Regional

assessment of the impact of climatic change on the distribution of a tropical conifer in the

lowlands of South America. Diversity and Distributions 13, 761-771.

Lowe A, Harris S, Ashton P (2004). Ecological genetics: design, analysis and application.

(Blackwell Science Ltd.)

Lusk CH (1996). Stand dynamics of the shade-tolerant conifers Podocarpus nubigena and

Saxegothaea conspicua in Chilean temperate rainforest. Journal of Vegetation Science 7,

549-558.

Macdonald T (2004). Fire biodiversity in the Australian Alps Workshop proceedings, Albury

NSW.

Martin HA (2006). Cenozoic climatic change and the development of the arid vegetation in

Australia. Journal of Arid Environments 66, 533-563.

Martin HA (1982). Changing Cenozoic Barriers and the Australian Paleobotanical Record.

Annals of the Missouri Botanical Garden 69, 625-667.

McLachlan JS, Clark JS, Manos PS (2005). Molecular Indicators of Tree Migration Capacity

under Rapid Climate Change. Ecology 86(8): 2088-2098.

McLoughlin S (2001). The breakup history of Gondwana and its impact on pre-Cenozoic

floristic provincialism. Australian Journal of Botany 49, 271-300.


Rohan Mellick 262

Mellick R, Rossetto M, Allen C, Wilson PD, Hill RS, Lowe A (2013b) Intraspecific

divergence associated with a biogeographic barrier and climatic models show future

threats and long-term decline of a rainforest conifer. The Open Conservation Biology

Journal 7, 1-10.

Mellick R, Wilson PD, Rossetto M (2013a). Post-glacial spatial dynamics in a rainforest

biodiversity hot spot. Diversity 5: 124-138.

Mellick R, Lowe A, Allen CD, Hill RS, Rossetto M (2012). Palaeodistribution modeling and

genetic evidence highlight differential post-glacial range shifts of a rainforest conifer

distributed across a latitudinal gradient. Journal of Biogeography 39(12): 2292-2302.

Mellick R, Lowe A, Rossetto M (2011) Consequences of long- and short-term fragmentation

on the genetic diversity and differentiation of a late successional rainforest conifer.

Australian Journal of Botany 59, 351-362.

Mill RR (2003). Towards a biogeography of Podocarpaceae. In 'Acta Hort. (ISHS)' pp.

615:137-147.

Molnar RE (1996). Observations on the Australian ornithopod dinosaur Muttaburrasaurus.

Memoirs of the Queensland Museum 39, 639-652.

Mooney SD, Harrison SP, Bartlein PJ, Daniau AL, Stevenson J, Brownlie KC, Buckman S,

Cupper M, Luly J, Black M, Colhoun E, Costa D, Dodson J, Haberle S, Hope GS,

Kershaw P, Kenyon C, McKenzie M, Williams N (2010). Late Quaternary fire regimes of

Australasia. Quaternary Science Reviews 30, 28-46.

Moritz C (2002). Strategies to protect biological diversity and the evolutionary processes that

sustain it. Syst Biol 51: 238-54.

Moss PT, Kershaw AP, Grindrod J (2005). Pollen transport and deposition in riverine and

marine environments within the humid tropics of northeastern Australia. Review of

Palaeobotany and Palynology 134, 55-69.

Nix HA (1991). Biogeography: Patterns and process. In ' Rainforest animals. Atlas of

vertebrates endemic to Australia‟s Wet Tropics'. (Eds HA Nix and M Switzer) pp. 11-39.

(ANPWS: Canberra, Australia.)

Owens JN, Morris SJ (1990). Cytological Basis for Cytoplasmic Inheritance in Pseudotsuga

menziesii. I. Pollen Tube and Archegonial Development. American Journal of Botany 77,

433-445.

Page RDM, Holmes EC (1998). 'Molecular evolution: A phylogentic approach.' (Blackwell

Science: Cambridge)

Peakall, R, Ebert, D, Scott, L.J, Meagher, P.F, Offord, C.A. (2003). Comparative genetic

study confirms exceptionally low genetic variation in the ancient and endangered

relictual conifer, Wollemia nobilis (Araucariaceae). Molecular Ecology 12, 23-31.

Pease KM, Freedman AH, Pollinger JP, McCormack JE, Buermann W, Rodzen J, Banks J,

Meredith E, Bleich VC, Schaefer RJ, Jones K, Wayne RK (2009). Landscape genetics of

California mule deer (Odocoileus hemionus): the roles of ecological and historical factors

in generating differentiation. Molecular Ecology 18, 1848-1862.

Pérez T, Albornoz J, Dominguez A (2002). Phylogeography of chamois (Rupicapra spp.)

inferred from microsatellites. Molecular Phylogenetics and Evolution 25, 524-534.

Petit JR, Jouzel J, et al. (1999). Climate and atmospheric history of the past 420,000 years

from the Vostok ice core, Antarctica. Nature 399, 429-436.



Petit RJ (2002). Chloroplast DNA variation in European white oaks Phylogeography and

patterns of diversity based on data form over 2600 populations. Forest Ecology and

Management 156, 5-26.

Petit RJ, Brewer S, et al. (2002). Identification of refugia and post-glacial colonisation routes

of European white oaks based on chloroplast DNA and fossil pollen evidence. Forest

Ecology and Management 156, 49-74.

Petit RJ, Csaikl UM, et al. (2003). Chloroplast DNA variation in European white oaks

phylogeography and patterns of diversity based on data from over 2600 populations (vol

156, pg 5, 2002). Forest Ecology and Management 176, 595-599.

Pocknall DT (1981). Pollen morphology of the New Zealand species of Dacrydium Solander,

Podocarpus L'Heritier, and Dacrycarpus Endlicher (Podocarpaceae). New Zealand

Journal of Botany 19: 67.

Pollock DD, Goldstein DB (1995). A comparison of two methods for constructing

evolutionary distances from a weighted contribution of transition and transversion

differences. Molecular Biology and Evolution 12, 713-717.

Powell W, Morgante M, Andre C, McNicol JW, Machray GC, Doyle JJ, Tingey SV, Rafalski

JA (1995). Hypervariable microsateliites provide a general source of polymorphic DNA

markers for the chloroplast genome Current Biology 5, 1023-1029.

Proctor M, Yeo P, Lack A (1996). 'The natural history of pollination'. Timber Press, Portland.

Pye MG, Gadek PA (2004). Genetic diversity, differentiation and conservation in Araucaria

bidwillii (Araucariaceae), Australia's Bunya pine. Conservation Genetics 5, 619-629.

Queller DC, Strassmann JE, Hughes CR (1993). Microsatellites and kinship. Trends in

Ecology and Evolution 8, 285-288.

Quilty PG (1994). The background: 144 million years of Australian palaeoclimate and

palaeogeography. In ' History of Australian Vegetation: cretaceous to recent'. (Ed. RS

Hill))

Quinn CJ, Price RA (2003) Phylogeny of the Southern Hemisphere Conifers. In Proc. Fourth

International Conifer Conference 129-136.

Quiroga MP, Premoli AC (2007). Genetic patterns in Podocarpus parlatorei reveal the long-

term persistence of cold-tolerant elements in the southern Yungas. Journal of

Biogeography 34, 447-455.

Quiroga MP, Premoli AC (2010). Genetic structure of Podocarpus nubigena (Podocarpaceae)

provides evidence of Quaternary and ancient historical events. Palaeogeography,

Palaeoclimatology, Palaeoecology 285, 186-193.

Rahmstorf S (2007). A semi-empirical approach to projecting future sea-level rise. Science

315, 368-370.

Richards CL, Carstens, BC, Knowles, LL (2007). Distribution modeling and statistical

phylogeography: an integrative frame work for generating and testing alternative

biogeographical hypotheses. Journal of Biogeography 34, 1833-1845.

Rissler L, Hijmans, R.J., Graham, C.H., Moritz, C., Wake, D.B. (2006). Phylogeographic

lineages and species comparisons in conservation analyses: A case study of California

herpetofauna. The American Naturalist 167 (5), p. 655.

Rossetto M, Crayn D, Ford A, Mellick R, Sommerville K (2009). The influence of

environment and life-history traits on the distribution of genes and individuals: a

comparative study of 11 rainforest trees. Molecular Ecology 18, 1422–1438.


Rohan Mellick 264

Rossetto M, Kooyman R, Sherwin W, Jones R (2008). Dispersal limitations, rather than

bottlenecks or habitat specificity, can restrict the distribution of rare and endemic

rainforest trees. American Journal of Botany 95, 321-329.

Rossetto M, Jones RS, McNally J (2002). Isolation of microsatellite loci from a rainforest

tree, Elaeocarpus grandis (Elaeocarpaceae), and amplification across closely related taxa.

Molecular Ecology Notes 2(2): 179-181.

Rule S, Brook BW, Haberle SG, Turney CSM, Kershaw AP, Johnson CN (2012) The

Aftermath of Megafaunal Extinction: Ecosystem Transformation in Pleistocene Australia.

Science 335, 1483-1486.

Schauble CS, Moritz C (2001) Comparative phylogeography of two open forest frogs from

eastern Australia. Biological Journal of the Linnean Society 74, 157-170.

Scoble J, Lowe AJ (2010). A case for incorporating phylogeography and landscape genetics

into species distribution modeling approaches to improve climate adaptation and

conservation planning. Diversity and Distributions 16, 343-353.

Scott LJ, Shepherd M, Henry RJ (2003). Characterization of highly conserved microsatellite

loci in Araucaria cunninghamii and related species. Plant Systematics and Evolution 236,

115-123.

Scott LJ, Cross M, Shepherd M, Maguire T, Henry RJ (1999). Increasing the efficiency of

microsatellite discovery from poorly enriched libraries in coniferous forest species. Plant

Molecular Biology Reporter 17, 351-354.

Sexton JP, McIntyre PJ, Angert AL, Rice KJ (2009). Evolution and Ecology of Species

Range Limits. In 'Annual Review of Ecology Evolution and Systematics' pp. 415-436.

(Annual Reviews: Palo Alto)

Shapcott A (1997) Population genetics of the long-lived Huon pine Lagarostrobos franklinii:

An endemic Tasmanian temperate rainforest tree. Biological Conservation 80, 169-179.

Shepherd M, Mellick R, Toon P, Dale G, Dieters M (2005). Genetic control of adventitious

rooting on stem cuttings in two Pinus elliottii x Pinus caribaea hybrid families. Annals of

Forest Science 62, 403-412

Shimeld P (2004). The Last Interglacial at Port Stephens, New South Wales. In 'Australasian

Quaternary Association Biennial Conference'. Cradle Mountain Tasmania. (Eds S

Haberle and J Stevenson). (AQUA 2004 Program and Abstracts)

Shimeld P (1995) A late Quaternary vegetation and climate history of Port Stephens and the

lower Hunter Valley, N.S.W. Thesis, University of Tasmania, Hobart.

Silba J (1984). An international census of Coniferae. Phytologia memoir 7, 1-79.

Sniderman JMK, Jordan GJ (2011). Extent and timing of floristic exchange between

Australian and Asian rainforests. Journal of Biogeography 38, 1445-1455.

Solomon S, Qin D, Manning M, Marquis M, Averyt K, Tignor MMB, Miller Jr HL, Chen Z

(2007). Climate change 2007: The physical science basis. Cambridge University Press,

Cambridge, U.K.

Sporne KR (1965). 'The morphology of gymnosperms. ' (Hutchinson and Co. Ltd.: London)

Stockey RA (1990). Antarctic and Gondwana conifers. In 'Antarctic Paleobiology: Its Role in

the Recontruction of Gondwana' (Ed. T Taylor). 179-191. (Springer-Verlag: New York).

Storfer A, Murphy MA, Evans JS, Goldberg CS, Robinson S, Spear SF, Dezzani R, Delmelle

E, Vierling L, Waits LP (2007) Putting „landscape‟ in landscape genetics. Heredity 98,

128–142



Townrow JA (1967b). On Rissikia and Mataia podocarpaceous conifers from the Lower

Mesozoic of southern lands. Pap. Proc. Roy. Soc. Tasmania 101, 103-36.

Townrow JA (1967a). The Brachyphyllum crassum complex of fossil conifers. Pap. Proc.

roy. Soc. Tasmania 101, 149-72.

Turnpenny P, Ellard S (2005). 'Emery's Elements of Medical Genetics.' (Elsevier: London)

Van Royen P (1979). 'The alpine flora of New Guinea ' (J. Kramer: Berlin)

VanDerWal J, Shoo LP, Williams SE (2009). New approaches to understanding late

Quaternary climate fluctuations and refugial dynamics in Australian wet tropical

rainforests. Journal of Biogeography 36, 291-301.

Wagstaff SJ (2004). Evolution and biogeography of the austral genus Phyllocladus

(Podocarpaceae). Journal of Biogeography 31, 1569-1577.

Williams NJ, Harle KJ, Gale SJ, Heijnis H (2006) The vegetation history of the last glacial-

interglacial cycle in eastern New South Wales, Australia. Journal of Quaternary Science

21, 735-750.

Wilson VR, Owens JN (2003). Cytoplasmic inheritance in Podocarpus totara

(Podocarpaceae). Proceedings of the Fourth International Conifer Conference - Conifers

for the Future? R. R. Mill: 171-172.

Wilson VR, Owens JN (2003). Cytoplasmic inheritance in Podocarpus totara

(Podocarpaceae). Acta Hort. (ISHS).

Wilson VR, Owens JN (1999). The reproductive biology of totara (Podocarpus totara)

(Podocarpaceae). Annals of Botany 83(4): 401-411.

Wodehouse R (1935). 'Pollen Grains. Their Structure, Identification and Significance in

Science and Medicine.' (McGraw-Hill: New York)

Woods KD, Davis MB (1989). Paleoecology of range limits-Beech in the Upper Peninsula of

Michigan Ecology 70, 681-696.

Wright Jr HE (1984). Sensitivity and response time of natural systems to climate change in

the Late Quaternary. Quat. Sei. Rev. 3, 91-131.

Wright S (1931). Evolution in Mendelian populations. Genetics 16, 97-159.

Wright S (1921). Systems of mating. Genetics 6, 111-178





Chapter 8

A NOVEL THERMAL SEAWATER DESALINATION

PROCESS BASED ON SELF-HEAT RECUPERATION

Hiroyuki Mizuno, Yasuki Kansha and Atsushi Tsutsumi

Institute of Industrial Science, Collaborative Research Center for Energy Engineering,

The University of Tokyo, Komaba, Meguro-ku, Tokyo, Japan

ABSTRACT

Thermal desalination processes, which separate pure water from seawater by

evaporation, have been used throughout the world. Thermal desalination typically

comprises either multi-stage flash or multi-effect desalination processes. These processes

have the following characteristics: the product water salinity is low enough for industrial

use and seawater quality does not seriously affect the operating conditions. However,

heat is not recovered perfectly by either process and is partly discarded to cooling water.

Therefore, energy consumption in conventional thermal processes is large.

A novel thermal desalination process based on self-heat recuperation, which is a

technology designed to save energy, has been proposed recently. In this process, all of the

heat is recirculated and energy is required only for producing a temperature difference in

the heat exchangers. As a result, the energy consumption can be dramatically reduced. At

the same time, it has been reported that the specific energy consumption, defined as the

total energy added divided by the product water flow rate, may be further decreased by

increasing the recovery ratio (the product water flow rate / the feed water flow rate).

In this chapter, we describe a novel desalination process based on self-heat

recuperation technology and explain its potential for saving energy compared with its

conventional counterparts. In addition, we summarize recent development of the process.

Keywords: Desalination, self-heat recuperation, energy saving, fluidized bed

Corresponding Author address: Email: [email protected].


Hiroyuki Mizuno, Yasuki Kansha and Atsushi Tsutsumi 268

INTRODUCTION

Water shortages are becoming more serious globally. In fact, more than 1/10 of the global

population cannot access potable water (UN, 2012). Furthermore, because of the high growth

rates of both population and industrial development, it is predicted that increasing numbers of

people will face water shortage problems (Cotruvo et al., 2010). To solve this problem,

increased attention has been paid recently to desalination processes.

To date, many desalination processes have been investigated (Park et al., 2011;

Charcosset, 2009; Younos et al., 2005). Amongst these, membrane desalination and thermal

desalination processes have generally been used (Voutchkov et al., 2010). Reverse Osmosis

(RO) desalination is a membrane process that separates pure water from seawater using a

semi-permeable membrane. RO is the most widely used desalination process, accounting for

51% of global desalination capacity, simply because the energy consumption of this process

is relatively low (Lattemann, 2010). In fact, the reported energy required by RO is between

9.0 and 25.2 kJ/ kg-water (Lattemann, 2010). However, the product water salinity is high

(Fritzmann et al., 2007). Hence, to use the product as industrial water, multiple RO stages are

required (Greenlee et al., 2009). Moreover, when the feed water contains high solids content,

undesirable fouling can occur, which restricts the permeation rate through the membrane.

Thus, the RO process requires a higher degree of pretreatment than the thermal desalination

process, which separates potable water from seawater by evaporation. Thermal desalination

can be classified into two processes: multistage flash distillation (MSF) and multiple effect

distillation (MED) (El-Dessouky et al., 1998; Al-Shammiri et al., 1999). The major difference

between these is the evaporation method. In the MSF process, seawater is vaporized by flash

evaporation, whereas in the MED process, seawater evaporation is achieved by heat transfer

through a heat exchanger. Thus, in the MSF process, it is generally accepted that scale

deposition, which leads to a decrease in heat transfer rates in the heat exchanger, is less likely

to occur. Thus, MSF is used more widely than the MED process. In fact MSF comprises 32%

of the reported global desalination capacity, while MED comprises only 8% (Lattemann,

2010).

Advantages of thermal process are as follows: 1) seawater quality does not seriously

affect the operating conditions. Hence, they do not require physical pretreatment, such as

filters or screens, other than what is provided by the intake. 2) Product water salinity is low

because vapor contains little solids (Voutchkov et al., 2010). However, thermal processes

consume considerably more energy than RO process. The reported energy consumption of

MSF is between 250 and 330 kJ/kg-water and that of MED is between 145 and 390 kJ/kg-

water (Lattemann, 2010). Thus, energy saving technologies for thermal desalination processes

are required. Recently, Kansha et al. proposed self-heat recuperation technology to save

energy (Kansha et al., 2009). They reported that the energy requirements of industrial thermal

processes could be considerably reduced, by 1/3 – 1/22, using such technology (Aziz et al.,

2011; Kansha et al., 2010; Kishimoto et al., 2012; Matsuda et al., 2011; Matsuda et al., 2010).

Figure 1 shows a flow diagram and temperature-heat diagram of a vapor/liquid stream

thermal process, with heat recovery achieved using a feed-effluent heat exchanger as a

conventional thermal process (self-heat exchange process). The feed stream passes through

the heat exchanger (1→2), in which the heat of the effluent is recuperated. The stream is then

heated and flows into a reactor (2→3). The effluent stream from the reactor can be used to


A Novel Thermal Seawater Desalination Process … 269

preheat the feed stream (4→5). However, a finite temperature difference is needed to transfer

heat from one medium to another. Thus, most of the latent heat during phase transition cannot

be recovered. Hence, the heat exchange load, QHX, is small. The effluent stream is then

cooled using cooling water (5→6). Figure 2 shows a flow diagram and temperature-heat

diagram of the thermal process based on self-heat recuperation. The feed stream passes

through a heat exchanger (1→2) and flows into the reactor (2→3). The effluent stream from

the reactor is then compressed adiabatically (3→4) to provide the temperature difference in

the heat exchanger. Because the boiling temperature of the effluent stream is raised by

compression, all of the heat is circulated (4→5). Therefore, the energy requirement is

dramatically reduced. The effluent stream is depressurized by a valve (5→6) and then cooled

by cooling water (6→7).

Figure 1. Configuration of a conventional thermal process.

Figure 2. Configuration of a thermal process based on self-heat recuperation.



Recently, Mizuno et al. proposed a novel thermal desalination process based on self-heat

recuperation (Mizuno et al., in press). They reported that the energy consumption of their

proposed process could be reduced to approximately 1/4 that of a corresponding conventional

thermal process. Furthermore, they found that the energy required for the proposed thermal

desalination process is comparable with that of RO desalination.

In this chapter, we describe a thermal desalination process based on self-heat

recuperation technology and explain the reason for its low energy consumption compared

with the conventional thermal process (MSF process) using the law of conservation of energy.

Furthermore, parameters that influence the energy consumption of this process are examined

and differing features between the process and RO are summarized. Finally, the latest

research on thermal desalination using self-heat recuperation technology is introduced.

PROCESS DESCRIPTION OF SELF-HEAT

RECUPERATIVE THERMAL DESALINATION

Figure 3 shows the configuration of the conventional thermal process (MSF process). The

feed stream flows into a heat exchanger in each stage and it is preheated. Then, the stream is

heated by an additional heater. The heated feed stream flows into a flash column, with

depressurization to lower-than-saturated vapor pressure. Then, flash evaporation occurs. The

vapor is cooled by the heat exchanger and becomes product. The brine flows into the next

flash column whose operating pressure is slightly lower than the previous one, and flash

evaporation occurs again. By repeating flash evaporation, product water is separated from

seawater. The energy consumption of this process was calculated by the commercial process

simulator Pro/II™ Ver. 9.0 (Invensys). The reported energy was 325 kJ/kg-water (Mizuno et

al., in press).

Figure 4 shows the thermal desalination process based on self-heat recuperation. This

process comprises three heat exchangers (HX), one separator (S), one compressor (CM), one

valve (V) and two coolers (CL). The seawater is divided into two streams (1→2, 3) and flows

into the first two heat exchangers (2→2‟, 3→3‟), where the heat from the product and brine

streams is recovered. The stream then passes through heat exchanger 3 (2‟, 3‟→4) and

receives latent heat. The generated vapor and the brines are separated from the heated stream

by the separator (S) (4→5, 6). The vapor is adiabatically compressed by the compressor (CM)

(5→7) to produce a temperature difference in the heat exchangers. The latent heat of the

compressed vapor stream is exchanged in heat exchanger 3 (7→8) and the stream condenses.

These streams (6, 8) then exchange their sensible heat with the feed stream in heat exchangers

1 and 2 (6→10, 8→9). The pressure of stream 9 is adjusted to ambient pressure by the valve

(V) (9→11). These streams (10, 11) are cooled to ambient temperature by the coolers (CL1,

CL2)(10→13, 11→12). Stream 13 contains rejected brine and stream 12 is product water.

The energy consumption of this process was calculated under conditions where the adiabatic

efficiency of the compressor was 100%, the recovery ratio, defined as the product water flow

rate divided by the feed water flow rate, was 30% and the minimum internal temperature

difference in the heat exchanger was 5 K. The reported energy was 70 kJ/kg-water and the

energy was 1/4 of that of the conventional thermal process (MSF process).



Figure 3. A multi-stage flash desalination process.

Figure 4. The thermal desalination process based on self-heat recuperation.

The specific energy consumption of the thermal desalination process based on self-heat

recuperation is quite small compared with that of the conventional process because in the

former, the sensible heat of the product and rejected brine streams as well as latent heat is

recirculated. That is, all of the heat is recuperated and recycled. On the other hand, in the

conventional process, only latent heat is recovered. Thus, larger energy consumption is

required because of the no recovering of the sensible heat.

The specific energy consumption difference between these two processes can be

understood using temperature-heat diagrams. Figure 5 shows a temperature-heat diagram for

the conventional thermal process (MSF). The solid and dotted streams represent for the cold

(feed) and hot (product) side streams, respectively. From this figure, it can be seen that in

each stage, only latent heat is recovered, while sensible heat is not recovered because of the

structure of MSF. Thus, from conservation of energy, additional heat equal to the discarded

sensible heat is required.

Figure 6 shows the temperature-heat diagram of the thermal process based on self-heat

recuperation. The solid and dotted lines represent the cold (feed) and hot (product and

rejected brine) side streams, respectively. From this figure, it can be seen that the sensible

heat from the feed stream is exchanged with that of the product and rejected brine streams at

heat exchangers 1 and 2. The feed stream receives the latent heat from the product stream at

heat exchanger 3 and is evaporated.



Figure 5. Temperature-heat diagram for MSF.

Figure 6. Temperature-heat diagram for the thermal seawater process based on self-heat recuperation.

Thus, both sensible heat and latent heat can be recirculated, leading to the drastic

reduction of the energy consumption compared to the conventional thermal process. The

required energy is used only for producing temperature differences in the heat exchangers.

The energy consumption of MED is also large compared with thermal seawater

desalination process based on self-heat recuperation. This is because, in the MED process,

only latent heat is recovered at the heat exchanger and sensible heat is not recovered, the

same as for the MSF process because of the structure of the MED process. Thus, external

heating load is required and the specific energy consumption is also larger.

SIMULATION RESULTS AND DISCUSSION

Adiabatic Efficiency

In the simulation, the adiabatic efficiency was assumed to be 100%. However, in real

compressors, it is approximately 70%. In this section, the influence of adiabatic efficiency on

the energy consumption of the thermal process based on self-heat recuperation is examined.



Adiabatic efficiency is defined as the ratio of ideal to actual work. Thus, this value represents

the ratio of additional energy used to increase the pressure of the gas stream. The increased

pressure of the stream increases the condensation temperature and produces a temperature

difference in the latent heat exchanger. Hence, the temperature difference in the latent heat

exchanger decreases as adiabatic efficiency decreases for constant additional energy. Figure 7

shows the temperature-heat diagrams for the thermal seawater desalination process based on

self-heat recuperation with 100% and 70% adiabatic efficiencies. The solid and dotted lines

represent the cold (feed) and hot (product and rejected brine) streams, respectively. From this

figure, it can be seen that the temperature difference between the hot and cold streams in heat

exchangers 1 and 2 are the same in each process, while that in heat exchanger 3 varies with

adiabatic efficiency. In this simulation, the important thing to note is that the specific energy

consumption remains nearly unchanged, while the temperature difference in the heat

exchanger varies with adiabatic efficiency and the amount of heat exchanged in HX 3 is the

same at each process. This is because in these processes, perfect internal heat circulation is

achieved and conservation of energy demands that the energy required is used only to

produce a temperature difference in the heat exchangers.

The reported temperature difference during the change in phase of a real desalination

process is very small, at 2 K (Darwish et al., 2002). Thus, it can be considered that, while the

higher adiabatic efficiency a compressor has, the better it is, the current compressor

specification is not to low to develop a highly energy efficient thermal desalination process.

Figure 7. Temperature-heat diagram with different adiabatic efficiencies.

Recovery Ratio and Minimum Internal Temperature Difference

It has been reported that the recovery ratio has a large influence on specific energy

consumption (Mizuno et al., 2012). The recovery ratio is defined as the product water flow

rate divided by the feed water flow rate. In other words, a 30% recovery ratio for a process

means 70% of the feed stream is rejected as brine and the remainder is product water. We

mentioned above that specific energy consumption decreases as the recovery ratio increases.



In fact, the specific energy consumption with a 30% recovery ratio was 70 kJ/kg-water and

that with a 70% recovery ratio was 30 kJ/kg-water. This is because when the feed stream

(seawater) flow rate and the minimum internal temperature difference in heat exchangers are

fixed, the energy consumption of the compressor (kW) is constant for a given recovery ratio.

The specific energy consumption (kJ/kg-water) is defined as the required energy (kW), which

is equal to the amount of discarded heat from conservation of energy, divided by the product

water flow rate (kg/s) and the discarded energy can be expressed by the following equation:

TCmTCmQ pp brinerejectedproductdiscarded (1)

where brinerejectedproduct and mm represent the flow rates of product and rejected brine streams,

respectively. pC is the specific heat at constant pressure and T is the temperature

difference between the product and rejected brine streams and feed water stream. Although

the recovery ratio varies, the discarded heat (kW) from this process is almost constant

because pC is fairly constant. Moreover, T is reasonably constant for a process design

based on self-heat recuperation, which leads to perfect internal heat circulation and the

minimum internal temperature difference is assumed to be constant. Furthermore, when the

feed stream is constant, the values of productm plus

brinerejectedm are constant. Thus, because of

the conservation of energy, the energy required by the compressor (kW) is constant at any

recovery ratio. However, when the recovery ratio is changed, the product water flow rate also

changes. Therefore, the higher the recovery ratio, the less specific energy (kJ/kg-water) is

consumed.

Mizuno et al. also reported the influence of minimum internal temperature difference in

heat exchangers on specific energy consumption (Mizuno et al., in press). The specific energy

consumption grows proportionally with increasing minimum internal temperature difference

in the heat exchanger. In fact, when the recovery ratio was assumed to be 30%, the energy

consumption of the proposed process with a 2 K minimum internal temperature difference

was 28 kJ/kg-water, because as you can see from the above equation, the temperature

difference is proportional to the energy required in the compressor.

Thus, it can be said that the minimum internal temperature difference in the heat

exchanger has a large impact on specific energy consumption.

To develop a highly energy efficient thermal desalination process, one must increase the

recovery ratio and decrease the temperature difference in the heat exchangers.

Comparison with RO Process

The reported specific energy consumption of the proposed thermal desalination process

using self-heat recuperation technology is in the approximate range 30 – 70 kJ/kg-water.

Thus, it can be seen that an RO process is superior to thermal desalination process from the

energy consumption point of view because the reported energy consumption of an RO

process is in the range 9.0 – 25.2 kJ/kg-water. However, some conditions, such as product



water salinity and recovery ratio, differ between the two processes. Hence, when we evaluate

the specific energy consumption, these differences in condition must be taken into account.

Product water salinity from an RO process is approximately 10 times higher than that of

the proposed process because RO separation is based on a rate-governed mechanism, while

the thermal process is based on vapor-liquid equilibrium. In an RO process, both salts and

water molecules can permeate the membrane. Thus, the product water salinity is relatively

high. In contrast, in the thermal process, because salts are not volatile, the salinity of the

product water can be very low. The product water salinity is important, especially for

industrial and agricultural use, which require values below approximately 10 ppm (Tanaka et

al., 2009). Hence, to use the product of the RO process as industrial or agricultural water, the

permeate water must pass through multiple RO stages, which leads to an increase of the

specific energy consumption because of the need for additional pumps. In the thermal

desalination process, the product water salinity is sufficiently low for industrial and

agricultural usage. Thus, to gain low salinity water, there is no need for additional processes.

The recovery ratio also differs between the two processes. The high recovery ratio

process enables the cost of pretreatment to be reduced and the impact of rejected brine on the

environment to be minimized. Thus, from the viewpoints of cost reduction and environmental

protection, enhancement of the recovery ratio is required. The recovery ratio from an RO

process is restricted to below 50%. This value is determined by the total permeate flux

through the membrane, J , which can be expressed by the following equation (Meyer et al.,

2008):

pfpfp ppLJ (2)

where and,p pL represent water flux, stream pressure and osmotic pressure, respectively.

The subscripts f and p indicate feed and permeate streams. From this equation, under

conditions where the operating pressure and osmotic pressure of the permeate stream are held

constant, the total flux rate increases with increasing feed stream pressure, which is applied

by a pump. However, the osmotic pressure of the feed stream is not constant but increases as

the recovery ratio increases because of the concentration of seawater. Thus, as the pressure

applied by the pump increases, the rate of increase in total permeate flux rate slows.

Furthermore, the maximum applicable pressure to membrane is limited to approximately

7000 kPa because of limitations in the mechanical strength of the RO module. Thus, the

recovery ratio in RO processes is limited to 60% (Greenlee et al., 2009). On the other hand, in

the thermal desalination process based on self-heat recuperation, the recovery ratio can be

easily varied by changing the amount of heat exchanged in the latent heat exchanger. Thus,

for achieving a high recovery ratio desalination process, it can be concluded that the thermal

process is more suitable than an RO process. (The problem of scale deposition is discussed in

following section.)

As you have seen, it is difficult to definitely determine which desalination process is the

lowest energy consumption. Thus, in the future, a new model that includes these parameters

must be developed.

In any case, it can be concluded that the energy consumption of the proposed thermal

process is comparable with that of an RO process under some conditions and the novel



desalination process can potentially be installed in regions where it is not feasible to introduce

a desalination process because of problems with feed water quality and energy cost.

Self-Heat Recuperative Evaporation

To realize a high recovery ratio thermal desalination process, the problem of scale

deposition must be solved. Scale, which is caused by the precipitation of salts in seawater that

exceed their saturation limits, deposits on a heated surface and leads to a reduction in heat

transfer coefficient within heat exchangers (Andritsos et al., 2003; Wade, 1979; Paakkonen et

al., 2009). It has been reported that calcium carbonate, calcium sulfate and sodium chloride

are the main contributors to scale deposition, and these begin to precipitate when 50%, 70%

and 90% of seawater is evaporated, respectively (The salt industry center of Japan, 2006). In

an actual thermal desalination process, the recovery ratio is generally limited to 30% to avoid

scale deposition. However, as mentioned above, the specific energy consumption decreases as

the recovery ratio increases in thermal desalination processes based on self-heat recuperation.

Furthermore, to reduce operating costs and the impact of the desalination plant on the

environment, a novel heat exchange method that can work at a high recovery ratio is required.

In the latest research on thermal desalination based on self-heat recuperation, to solve the

scale deposition problem and develop a highly energy efficient thermal desalination process,

a novel desalination process that uses a fluidized bed evaporator has been investigated. In this

process, it can be considered that seawater evaporation occurs on heated particles. Thus, scale

does not deposit on the heated surface. Furthermore, even if the scale deposition takes place

on the heat transfer surface, fluidized particles should erode the scale from the surface.

Currently, experiments on the anti-scale performance of fluidized bed evaporators have been

conducted, aimed at enabling the practical use of this desalination process.

The self-heat recuperative desalination process using a fluidized bed evaporator can be

applied to waste water treatment, medicine manufacturing, food concentration and other

applications, in addition to desalination. Although self-heat recuperation technology has

significant energy saving potential, it has been applied to date only in processes in which the

scale deposition problem does not occur, such as distillation and gas separation processes.

When a novel thermal desalination process using fluidized bed evaporators is finally

implemented, it will be possible to widen the applications of self-heat recuperation

technology and to further enhance energy savings in production processes. Therefore, it can

be said that research on self-heat recuperative thermal desalination using a fluidized bed

evaporator has significant potential for realizing an eco-friendly society as well as solving the

water shortage problems around the world.

CONCLUSION

In this chapter, a new thermal seawater desalination process based on self-heat

recuperation technology was introduced and the differences between the proposed process

and conventional processes were examined. It was found that the energy consumption of the

proposed process is reduced to 1/4 of that of a conventional thermal process. Furthermore, it



was shown that the higher the recovery ratio, the less specific energy is consumed. To enable

a high recovery ratio desalination process, the problem of scale deposition must be overcome.

To this end, a novel self-heat recuperative desalination process using a fluidized bed

evaporator has been examined.

This research has barely begun and further research is needed to implement the process.

It can be said that this is a very promising process for solving water shortage problems around

the world.

REFERENCES

Al-Shammiri M., Safar M. (1999). Multi-effect distillation plants: state of the art,

Desalination, 126(1-3), 45–59. doi:10.1016/S0011-9164(99)00154-X.

Andritsos N., Karabelas A. J. (2003) Calcium carbonate scaling in a plate heat exchanger in

the presence of particles, International Journal of Heat and Mass Transfer 46(24), 4613–

4627. doi:10.1016/S0017-9310(03)00308-9.

Aziz M, Kansha Y, Tsutsumi A (2011) Self-heat recuperative fluidized bed drying of brown

coal, Chemical Engineering and Processing 50, 944-951. doi: 10.1016/j.cep.2011.07.005.

Cotruvo J., Abouzaid H. (2010) Desalination Technology. CRC press, Florida, chapter 1.

Charcosset C (2009) a review of membrane processes and renewable energies for

desalination, Desalination 245, 214-231. doi: 10.1016/ j.desal.2008.06.020.

Darwish M.A., Asfour F. Al., Al-Najem N. (2002) Energy consumption in equivalent work

by different desalting methods : case study for Kuwait, Deaslination 152, 83–92.

El-Dessouky H, Alatiqi I, Ettouney H (1998) Process synthesis: the multi-stage flash

desalination system, Desalination 115, 155-179. doi: 10.1016/ S0011-9164(98)00035-6.

Fritzmann C, Lowenberg J, Wintgens T, Melin T, (2007) State-of-the-art of reverse osmosis

desalination, Desalination 216, 1-76. doi: 10.1016/ j.desal.2006.12.009.

Greenlee L. F., Lawler D. F., Freeman B. D., Marrot B., Moulin P. (2009) Reverse osmosis

desalination: water sources, technology, and today‟s challenges, Water research, 43(9),

2317–48. doi:10.1016/ j.watres.2009.03.010.

Kansha Y, Tsuru N, Sato K, Fushimi C, Tsutsumi A (2009) Self-Heat recuperation

technology for energy saving in chemical processes, Industrial and Engineering

Chemistry Research 48,7682-7686. doi:10.1021/j.ces.2009.05.001.

Kansha Y, Tsuru N, Fushimi C, Shimogawara K, Tsutsumi A (2010) An innovative

modularity of heat circulation for fractional distillation, Chemical Engineering Science

65, 330-334. doi: 10.1016/ j.ces.2009.05.001.

Kishimoto A, Kansha Y, Fushimi C, Tsutsumi A (2012) Exergy recuperative CO2 gas

separation in pre-combustion capture, Clean Technol. and Environ. Policy 14, 465-474.

doi: 10.1007/s10098-011-0428-3.

Lattemann S (2010) Development of an environmental impact assessment and decision

support system for seawater desalination plants, Dissertation, Delft University.

Meyer D. E., Williams M., Bhattacharyya D. (2008) Separation Technology second edition,

Wiley interscience, New Jersey.

Matsuda K., Kawazuishi K., Hirochi Y., Sato R., Kansha Y., Shikatani Y., Kunikiyo H.,

Tsutsumi A. (2010) Advanced energy saving in the reaction section of the hydro-



desulfurization process with self-heat recuperation technology, Applied thermal

engineering 30, 2300-2305. doi:10.1016/ j.applthermaleng.2010.05.003.

Matsuda K., kawazuishi K., Kansha Y., Fushimi C., Nagao M., Kunikiyo H., Masuda F.,

Tsutsumi A. (2011) Advanced energy saving in distillation process with self-heat

recuperation technology, Energy 36, 4640- 4645. doi:10.1016/j.energy.2011.03.042.

Mizuno H., Kansha Y., Kishimoto A., Tsutsumi A. Thermal seawater desalination based on

self-heat recuperation, Clean Technology and Environmental Policy, in press.

doi:1007/s10098-012-0539-5.

Mizuno H., Kansha Y., Kishimoto A., Tsutsumi A. (2012) Thermal Desalination Process

Based on Self-heat Recuperation, Chemical Engineering Transactions 29, 379-384. doi:

10.3303/CET1229064.

Paakkonen T. M., Riihimaki M., Puhakka E., Muurinen E., Simonson C. J., Keiski R. L.

(2009) Crystallization fouling of CaCO3 –effecty of bulk precipitation on mass

deposition on the heat transfer suface, Proceedings of international conference on heat

exchanger fouling and cleaning, Published online www.heatexchanger-fouling.com.

Park K, Hong SY, Lee JW, Kang KC, Lee YC, Ha M, Lee JD (2011) A new apparatus for

seawater desalination by gas hydrate process and removal characteristics of dissolved

minerals (Na+, Mg2+, Ca2+, K+, B3+), Desalination 274, 91-96. doi: 10.1016/

j.desal.2011.01.084.

Tanaka K., Matsui K., Hori T., Iwahashi H., Takeuchi K., Ito Y. (2009) The First Large Full

Triple Pass RO Seawater Desalination Plant in the World, Mitsubishi Heavy Industries

Technical Review 46, 13-15 (in Japanese).

The Salt Industry Center of Japan (2006) Data Book for Seawater Science and Salt

Production, The Salt Industry Center of Japan, Tokyo, chapter 1 (in Japanese).

UN (2012) WATER, United Nations Web sites, http://www.un.org/en/ globalissues/water/

Accessed 29 July 2013.

Voutchkov N, Sommariva C, Pankratz T, Tonner J (2010) Desalination Technology. CRC

press, Florida, chapter 2.

Wade N M (1979) a review of scale control methods, desalination 31,309-320.

Younos T, Tulou KE (2005) Overview of desalination techniques, Journal of contemporary

water research and education 132, 3-10. doi: 10.1111/ j.1936-704X.2005.

mp132001002.x.




Chapter 9

BIOCHEMICAL GAS SENSOR “BIO-SNIFFER”

AND IMAGING SYSTEM FOR MEDICAL

AND HEALTH CARE SCIENCE

Takahiro Arakawa and Kohji Mitsubayashi Tokyo Medical and Dental University, Tokyo, Japan

ABSTRACT

We introduce biochemical gas sensors “bio-sniffer” which leverage biochemical

reactions and characteristics of various enzymes. Volatile substances associated with

diseases and breath odors are released from human body. Hence, highly sensitive and

highly selective bio-sniffers have been developed and demonstrated in biomedical and

health care applications utilizing enzymes as biomarkers of related diseases. We present

three topics of bio-sniffer techniques for monitoring of gaseous components of methyl

mercaptan, formaldehyde and ethanol. A bio-sniffer for gaseous methyl mercaptan (MM)

is applied to measure halitosis (oral odor) in breath. MM is known as one of the

important chemicals of halitosis in oral care. And then, a fiber-optic biochemical gas

sensor for assessment of indoor environment is fabricated and demonstrated in

monitoring of formaldehyde. Finally, two-dimensional visualization system of gaseous

ethanol is demonstrated in spatial and temporal imaging of gaseous components of

exhaled breath. In this chapter, we describe the details and future prospects of

biochemical gas sensors for environmental analysis, medical and health care.

1. INTRODUCTION

In last decade, a measurement of gaseous components from human body plays an

important role in a non-invasive, painless and attractive diagnostic method of monitoring

disease states without risk of blood collection to patients. Exhalation diagnosis and human

body odor were reported to be helpful in diagnosis of diseases by monitoring for volatile

organic compounds (VOCs) [1, 2]. Additionally, Health damages from formaldehyde, a well-

known representative of volatile organic compounds, are caused by human activities and


Takahiro Arakawa and Kohji Mitsubayashi 280

products. The main emission sources of indoor formaldehyde are urea-formaldehyde resins in

pressed wood products in cabinetry and furniture. A combination of respiratory disease and

allergic dermatitis, so-called sick building syndrome is associated with chronic exposure to

formaldehyde [3, 4]. Hence, instant and simple methods are strongly desired to discriminate

volatile organic compounds from human body, indoor environment and so on. However, gas

chromatography methods are required time consuming, expensive and large-scale equipment.

It is difficult for low selective semiconductor gas sensor to arrest density and a change in the

space precisely. Therefore, continuous and easy monitoring system for various gas

components behavior in spatial localization is strongly required. Some technology has been

developed for gas sensor employing enzymatic reaction, such as biochemical gas sensors for

ethanol, acetaldehyde, formaldehyde and various volatile organic compounds and a NADH-

dependent fiber-optic biosensor „bio-sniffer‟ for determination of gaseous components [5].

Also the enzyme-based biosensors are highly selective and sensitive for target chemicals.

We introduce biochemical gas sensors “bio-sniffer” by exploiting biochemical reactions

and characteristics of various enzymes. Firstly, optical bio-sniffer (gas sensor) for gaseous

methyl mercaptan (MM) was applied to measure halitosis (oral odor) in breath. MM is known

as one of the important chemicals of halitosis in oral care. The MM bio-sniffer was fabricated

a oxygen-sensitive optical fiber with a ruthenium organic complex, a monoamine oxidase

type-A (MAO-A) immobilized membrane and a reaction unit, which consists of liquid and

gaseous compartments, separated by a hydrophobic porous polytetrafluoroethylene (PTFE)

diaphragm membrane. This sensor was applied for halitosis measurement.

Secondly, gaseous formaldehyde is well known as one of the harmful volatile organic

compounds, which causes health damages of so-called sick building syndrome. A fiber-optic

biochemical gas sensor for assessment of indoor environment was fabricated and tested in

monitoring of formaldehyde. The bio-sniffer measures formaldehyde vapor as fluorescence of

reduced NADH, which is the product of enzymatic reaction of formaldehyde dehydrogenase

(FALDH). We employed a micro flow-cell with a FALDH immobilized membrane to prevent

the FALDH from deactivation at the optode. An ultraviolet light emitting diode (UV-LED)

with peak emission of 335 nm was employed as an excitation light source. Emission of the

UV-LED was introduced to the optode through an optical fiber and fluorescence of NADH

was picked up coaxially at the optode. Continuous formaldehyde monitoring with

biochemical method was successfully conducted with high selectivity at sub-ppb level.

Finally, two-dimensional imaging system of gaseous ethanol was demonstrated in spatial

and temporal imaging of gaseous components of exhaled breath for a non-invasive diagnostic

method of monitoring disease states without risk to patients. We applied alcohol oxidase

(AOD), catalyzes low molecular weight alcohols by molecular oxygen into aldehydes with

the production of hydrogen peroxide, which can be led to the HRP-luminol-H2O2 system for

the chemiluminescence analysis. The chemiluminescence generated by the catalytic reaction

of gaseous ethanol was analyzed. Rebuilding an image analysis of chemiluminesce allows it

to visualize spatial and temporal image of gaseous ethanol. The metabolism capability of

alcohol in volunteers was successfully evaluated by the illustrated images of ethanol

chemiluminescence noninvasively.


Biochemical Gas Sensor “Bio-Sniffer” and Imaging System … 281

2. BIO-SNIFFER FOR METHYL MERCAPTAN IN HALITOSIS

Halitosis diagnosis is important in the medical and dental fields. Methyl mercaptan (MM)

is one of typical causation for halitosis [6, 7]. The American Conference of Governmental

Industrial Hygienists and the Environment Agency Government of Japan had specified the

MM as a typical volatile organic compounds (VOCs) and malodorous substance. The

maximum permissible concentration of gaseous MM in the work place is 5.0 ppm (threshold

limit value-time weighted average concentration) and 0.5 ppm. We have constructed an

optical bio-sniffer for the measurement of MM as major odorous chemical in bad breath using

monoamine oxidase. After the evaluation of the sensor characteristics using a batch

measurement system, the optical sniffer was applied to the halitosis analysis as a

physiological application.

From Analytica Chimica Acta 573–574 (2006) 75–80.

Figure 1. Schematic image and photograph of methyl mercaptan bio-sniffer with a sensor cleaning

system in the gas phase and the device components.



The schematic image and photograph of an optical bio-sniffer for MM in the gas phase

was shown in Figure 1. The sniffer device constructed with an enzyme immobilized

membrane for detection of MM, a flow cell and a commercial available oxygen-sensitive

optical fiber with a ruthenium-organic complex. The optical fiber was coated in a sol–gel

process with the ruthenium-organic complex that is subjected to an optical quenching

(excitation wavelength: 470 nm, fluorescent wavelength: 600 nm) in the presence of oxygen.

Monoamine oxidase type A (MAO-A, E.C.1.4.3.4) was used as the MM recognition material

for the halitosis bio-sniffer. MAO-A was mixed with PVA-SbQ solution (UV cross-linkable

polyvinyl alcohol) to a dialysis membrane for enzyme immobilization, and then irradiated

with a fluorescent lamp for photo-crosslink the solution and immobilizing the enzyme

membrane. The MAO-A immobilized membrane was removed from the glass plate and

immersed in phosphate buffer. The flow cell was constructed by connecting two t-tubes to

both sides of a stainless steel pipe. The enzyme membrane, which was cut to the required

dimensions using a scalpel, was used to close one of the open edges of the flow cell and

secured with a supporting nylon mesh net and a rubber O-ring. Figure 2 illustrates a batch

flow measurement system with valve connecting to a gas generator for evaluating the

characteristics of the optical bio-sniffer and to a sampling bag for the breath analysis.

Standard gaseous MM and sample of human breath and phosphate buffer solution could be

injected individually through the flow cell in the detection system. A spectrophotometer was

optically connected to the bio-sniffer and monitoring the fluorescent quenching, resulting

from oxygen consumption caused by MAO-A catalytic reaction with MM.


Figure 2. Breath analysis procedure and system by the optical bio-sniffer with MAO-A for methyl

mercaptan in the gas phase.

The performance of the bio-sniffer for MM vapor was evaluated with the batch flow

measurement system. The output increased rapidly following application of MM vapor and

reached to a steady-state output within 3 minutes. The reduction of optical quenching of the

ruthenium-organic complex emission induced by the consumption of oxygen was caused by

the MAO-A catalytic reaction with methyl mercaptan. The calibration curve of the bio-sniffer

for methyl mercaptan in the gas-phase is shown in Figure 3. The calibration range of the bio-



sniffer for MM vapor was from 8.7 to 11500 ppb (correlation coefficient 0.977) including a

MM threshold (200 ppb) of pathologic halitosis and the human sense of smell level 3.5 (10.0

ppb). As the physiological application, breath samples from healthy male volunteer subjects

were provided to the gas measurement system with the optical bio-sniffer for monitoring

halitosis. Exhausted breath gases in five subjects were sampled every hour and the process of

concentration change of MM was monitored [8-11].


Figure 3. Calibration curve of the bio-sniffer for methyl mercaptan in the gas phase. (200 ppb MM as

threshold of pathologic halitosis).

3. BIO-SNIFFER FOR FORMALDEHYDE IN ENVIRONMENTAL ANALYSIS

Formaldehyde is a reactive and flammable chemical, which is well known as one of the

volatile organic compounds (VOCs). Formaldehyde was upgraded in classification from

group 2A (probably carcinogenic to humans) to group 1 (carcinogenic to humans) by

International Agency for Research on Cancer (IARC) in 2004 [12]. Health damages for long-

term exposure to formaldehyde are also reported. A combination of respiratory disease,

allergic dermatitis and other ailments, so-called sick building syndrome, is associated with

chronic exposure to low concentrations of formaldehyde [13]. For this reason, the World

Health Organization sets guideline level of formaldehyde exposure to 80 ppb. Nonetheless,

formaldehyde is widely used for many industrial purposes. It is typically used in pressed-

wood products, adhesives and coatings. Outgassing from resins in furniture, wallpaper or

paints can be the major indoor formaldehyde emission sources. Formaldehyde emissions in

residential atmosphere cause sick building syndrome even in lower concentrations than 80

ppb [14].

In the previous study, we reported a biochemical gas-sensor with formaldehyde

dehydrogenase (FALDH) for convenient analysis of gaseous formaldehyde with highly gas-

selectivity. The bio-sniffer for formaldehyde in the gas phase was constructed by

immobilizing FALDH to a Pt-electrode coated hydrophilic PTFE membrane. The oxidation



current of NADH produced by the enzymatic reactions was measured by amperometric

analysis [15]. Biochemical recognition system allows highly selective and highly sensitive

gas monitoring with a simplified structure.

We introduce a highly sensitive fiber-optic bio-sniffer using FALDH immobilized

membrane for gaseous formaldehyde monitoring. The bio-sniffer measures formaldehyde

concentration as fluorescence of NADH. In order to obtain sufficient sensitivity for indoor

formaldehyde monitoring, a photomultiplier tube (PMT) was employed as a fluorescence

detector. The UV-LED based excitation system enabled the bio-sniffer to be a simplified and

miniaturized gas sensor for its low power consumption and low heat generation. LEDs also

provided advantages in security and controllability compared with general UV lamps. The

construction, characteristics and possible applications of the fiber-optic bio-sniffer was

reported [16].

FALDH immobilized membrane was prepared by immobilizing FALDH (EC 1.2.1.1)

onto a hydrophilic PTFE membrane filter [17]. 2-methacryloyloxyethyl phosphorylcholine

(MPC) copolymerized with 2-ethylhexyl methacrylate (EHMA) was used for enzyme

immobilization. MPC-co-EHMA (PMEH) was synthesized using free radical-polymerization

method. NADH (-nicotinamide adenine dinucleotide, reduced from disodium salt) was

prepared for characterization of the NADH measurement system. Phosphate buffer solution

containing 20 mmol/L NAD+ (-NAD) was prepared in order to wet the enzyme-

immobilized membrane during gas measurement. a fiber-optic NADH measurement system

was constructed.

formaldehyde+NAD+H2OFALDH¾ ®¾¾ fomicacid +NADH +2H+

(1)

A UV-LED (= 335 nm) was employed as an excitation light source. Emission of the

UV-LED was coupled into a branched optical fiber using custom-fabricated UV-LED power

supply with an adjustable optical fiber connector. A PMT was connected to the other

branched terminal. An optical fiber probe was connected to the assembled terminal of the

branched optical fiber. In order to reduce background level, excitation light and fluorescence

were filtered using band-pass filters (330–350 nm and 490–510 nm), respectively. Thus,

fluorescence of NADH at the neighborhood of the optical fiber probe could be measured

coaxially by the PMT [18, 19].

From Biosensors and Bioelectronics 26 (2010) 854–858.

Figure 4 Construction of the fiber-optic bio-sniffer and FALDH immobilization for continuous

monitoring of formaldehyde.



A formaldehyde-sensitive optode was constructed by attaching a flow cell with FALDH

immobilized membrane onto the optical fiber probe. The flow cell was fabricated by

assembling a silicone tube (2 mm) and a cylindrical PMMA cell (4 mm) with inlet/outlet as

shown in Figure 4.

FALDH immobilized membrane was prepared by curing a mixture of PMEH solution (1

µL/cm2) and FALDH (50 units/cm

2), which was spread on the H-PTFE membrane filter, for

180 min under 4 degree C. The FALDH membrane was cut into 1 cm × 1 cm and tightly

fixed onto the end of the PMMA tube using a silicone O-ring.

Measurement of gaseous formaldehyde was carried out using the standard formaldehyde

measurement system. Gaseous formaldehyde was applied to the optode with a flow rate of

200 mL/min. Fluorescent signal immediately increased after application of gaseous

formaldehyde and attained steady state within 2 min. Fluorescent intensities of the steady

state are determined by the balance of NADH production rate and the rinse effect of PB.

Sufficient recovery of initial state when formaldehyde gas was eliminated from the optode

was also confirmed. The response curves indicated that the bio-sniffer was stable enough to

be used for continuous formaldehyde monitoring.


Figure 5. Calibration curve for formaldehyde vapor using formaldehyde bio-sniffer. The fluorescent

intensity related to formaldehyde concentrations from 2.5 to 10 ppb. The small box shows typical

responses to gaseous formaldehyde.

The calibration curve of the FALDH bio-sniffer for gaseous formaldehyde is shown in

Figure 5. The output fluorescence was related to the formaldehyde concentration from 2.5

ppb to 10 ppm. The bio-sniffer also showed high reproducibility. The calibration range

included both the guideline value of WHO (80 ppb, signed in the figure). Considering the

following capability of fluctuation of formaldehyde level, the bio-sniffer showed comparative

sensitivity with above system and higher detection limit than that of commercial

formaldehyde sensors. For instance, modern commercial solid-state sensors for real-time



monitoring are available from several tens of ppb. The lower detection limit of the FALDH

bio-sniffer can be improved by some realistic approaches. For instance, excitation with

multiple LEDs is considered to be one of the most effective approaches. Thicker optical fibers

are also useful to enhance fluorescence signal because the effective region of the enzyme

membrane increases. These improvements may enhance the sensitivity down to sub-ppb level

considering the present SN ratio. On the other hand, the optical system can be easily

miniaturized using a photo-diode as a detector instead of the PMT.

Gas selectivity for various chemical substances was also investigated. Acetaldehyde is a

useful reference to assess gas selectivity of a formaldehyde sensor because of similar

structure. Figure 6 shows the relative fluorescent intensities of the FALDH bio-sniffer for

formaldehyde, acetaldehyde, acetone, benzene, methanol and ethanol (5.0 ppm). As a result,

the fluorescent intensity for acetaldehyde was only 1.3% of output for formaldehyde, and no

fluorescence was detected for other gaseous substances. Considering the selectivity for

formaldehyde, the selectivity of the FALDH bio-sniffer was approximately 10 fold higher

than that of a conventional formaldehyde gas sensor. Such a high selectivity of the bio-sniffer

was induced by the specificity of FALDH. Since the FALDH bio-sniffer measures

formaldehyde as a concentration of NADH, it is also possible to measure other volatile

chemical substances with high selectivity by replacing enzyme with similar system (i.e.,

alcohol dehydrogenase for ethanol monitoring).

In addition, we are applying the NADH based biosensor technique with UV-LED to

many application of detection of gaseous ethanol [5] and amino acid solution of L-leucine

[20] and phenylalanine [21]. Especially, this NADH-detecting biosensor with enzyme

immobilized membrane that permits to contribute various diagnosis of amino acid in blood

and urine by applying to point of care testing. In the future, we will apply the biosensor with

UV-LED excitation system to the combined multiple dehydrogenase enzyme-immobilized

biosensor for determination of various amino acids.


Figure 6. Comparison of gas selectivity of FALDH bio-sniffer and conventional FA sensor.



4. TWO-DIMENSIONAL VISUALIZATION SYSTEM OF

GASEOUS ETHANOL

Breath gas analysis plays a role important method for non-invasive information of

clinical state of volatile organic compounds (VOCs) present in the exhaled breath, which is

faster and safer than blood or urine analyses. Compared to traditional diagnostic methods,

breath analysis is painless and requires no special medical skills [22]. Approximately 3000

VOCs have been found in human breath using gas chromatography; 200 of these have been

linked to various diseases such as lung cancer, breast cancer, diabetes, and liver diseases [23,

24]. For example, methanol, ethanol, acetone, and ethylene are all VOCs that are known to be

diagnostic for certain diseases. Ethane and methylated hydrocarbons are related to oxidative

stress and lung or breast cancers, respectively, and exhaled carbon monoxide is a marker for

cardiovascular diseases. Many investigations have been carried out to characterize and

identify the substances found in breath samples. Gas chromatography and mass spectrometry

was used to identify 42 VOCs that act as lung cancer biomarkers [25]. As a result of such

extensive studies, several breath markers have been successfully used in the diagnosis of

diseases. Breath gas screening is a highly desirable diagnostic technique because it is a non-

invasive, convenient, and rapid diagnostic method that links VOCs found in exhaled breath to

medical conditions [26, 27]. A straight-forward continuous monitoring system for various gas

components is therefore required. We fabricate and demonstrated a non-invasive spatial

visualization system of exhaled ethanol for real-time analysis of alcohol metabolism. We

have focused on alcohol oxidase (AOD). AOD catalyzes the reaction of low molecular weight

alcohols with molecular oxygen, producing aldehydes and hydrogen peroxide, which can in

turn be consumed in the HRP-luminol-H2O2 reaction for CL analysis (Figure 7). CL was

generated by the catalytic reaction of breath gaseous ethanol was detected and analyzed using

standard gas and breath gas sample. The system for imaging the spatial movement of gas was

developed. AOD and HRP enzyme and PVA-SbQ mixture was coated onto the mesh

substrate, spread, cured, and treated with ultraviolet radiation. Enzyme-immobilized supports

were stored at 4oC. AOD and HRP were used together to generate the CL signal from gaseous

ethanol. Tris-HCl buffer solution was the medium of the AOD catalytic reaction, and HRP

catalyzed the CL reaction. Ethanol is oxidized to acetaldehyde by AOD in the presence of

oxygen.

Figure 7. Schematic view of gaseous ethanol imaging system, consisting of ethanol gas generator, mass

flow controller, enzyme immobilized mesh and EM-CCD recoding system. Fabrication process of AOD

and HRP enzymes and UV-cross-linkable PVA-SVQ immobilized mesh substrate.



From Analyst, 136 (2011) 3680-3685.

Figure 8. Temporal changes of various concentrations of standard gaseous ethanol.

The peroxide then reacts with the luminol solution catalyzed by HRP, ultimately

generating chemiluminescence (λ= 460 nm). The reactions are as follows [28]:

ethanol+O2

AOD¾ ®¾¾ acetaldehyde+H2O2 (2)

luminol+H2O2+2OH- HRP¾ ®¾¾ 3-aminophthate+2H2O+N2+hn (3)

The temporal changes of various concentrations of standard gaseous ethanol are shown in

Figure 8 with the images of maximum intensity processed. As Figure 8 indicates, the CL

intensities increased rapidly following the loads of standard gaseous ethanol. The peaks

appeared within 90 s and faded away within 180 second. And the images showed that the area

became bigger and the CL intensities became higher as a result of the increments in the

amount of standard gaseous ethanol. The development of the system showed rapid and

accurate responses and a visible measurement in real time. Figure 9 shows the calibration

curve of the system for standard gaseous ethanol measurement. The peaks of average CL

intensities detected by the system were related to the concentrations of gaseous ethanol from

30 to 400 ppm. This system showed a wide calibration range covered the concentration of

exhaled breath ethanol from drunk driving to acute toxicosis of alcohol [29, 30].

CL of gaseous ethanol (300 ppm) and ethanol in exhaled breath sample (30 min after

ethanol oral administration) of volunteer with ALDH2 (-) are illustrated in Figure 6. The

spatiotemporal changes of ethanol in exhaled breath (right) were detected and recorded as the

gaseous ethanol (left). The upper images were processed to show the different information of

CL of exhaled ethanol in a 3D profile. The lower ones were processed to show CL intensity in

RGB. Ethanol in exhaled breath after oral administration was selectively detected and



measured. Compared with the gas detector tube for exhaled ethanol after oral administration,

the system showed a better sensitivity. Exhaled ethanol in breath can be visibility evaluated

with the system in real time, which cannot be evaluated with other ethanol sensors.

From Analyst, 136 (2011) 3680-3685.

Figure 9. Calibration curve of the imaging system for standard gaseous ethanol. The peak of average

intensity related to the concentration of gaseous ethanol from 30 to 400 ppm.

From Analyst, 136 (2011) 3680-3685.

Figure 10. Chemiluminescence of gaseous ethanol (left) and ethanol in exhaled breath (right). The

spatiotemporal changes of ethanol in exhaled breath were detected and recorded as gaseous ethanol.



The pharmacokinetic profiles of mean exhaled ethanol are shown in Figure 11. The

exhaled breath samples of 10 volunteers (ALDH2 (+), n=5; ALDH2 (-), n=5) were

demonstrated. The lines illustrate the regression line based on data points from 75 min and

onward according to zero-order kinetics for the time variation profile. Exhaled ethanol

concentrations in all volunteers rapidly increased after oral administration and the peaks

appeared at 30 min and then gradually decreased until the end of sample collection at 120

min. During the first 30 min, the time variation profile showed the absorption and distribution

function after oral administration, and the values thereafter showed the elimination function.

Due to the deficiency in ALDH2, acetaldehyde (oxidation product of ethanol) could not be

metabolized. The ethanol concentration in volunteers with ALDH2 (+) was lower than that in

volunteers with ALDH2 (-). Ethanol that had not even been metabolized was egested in

exhaled breath and was detected and illustrated by the visualization system. ALDH2 was

considered to be the main enzyme in acetaldehyde oxidization related to ethanol metabolism.

Hence, in volunteers with ALDH2 (-), it results in increased acetaldehyde levels and leads to

ethanol being metabolized with a slower rate than volunteers with ALDH2 (+) [29, 30]. We

applied this imaging system to a non-invasive method of directly imaging exhaled gaseous

ethanol using chemiluminescence and used in the analysis of alcohol metabolism [31]. This

imaging system will be useful and significant for the diagnosis of diseases using exhaled

human breath. In addition, the rate of ethanol metabolism by ALDH2(+) and ALDH2(-)

human subjects was also determined using direct gaseous ethanol imaging using flow control

unit, over 300 minutes after the oral administration of alcohol.

From Analyst, 136 (2011) 3680-3685.

Figure 11. Pharmacokinetic profile of exhaled ethanol concentration of 10 healthy volunteers.



CONCLUSION

We introduce biochemical gas sensors “bio-sniffer” using biochemical reactions and

characteristics of various enzymes for determination of methyl mercaptan, formaldehyde and

gaseous ethanol. We apply the bio-sniffer to highly sensitive and highly selective gas sensor

for biomedical and health care applications utilizing enzymes as biomarkers of related

diseases. In future work, the sensing and imaging system of gaseous components from human

body and breath will apply for imaging of volatile organic components information from

human, transdermal analysis, halitosis and non-invasive screening.

REFERENCES

[1] J. P. Spinhirne, J. A. Koziel, N. K. Chirase, A device for noninvasive on-site sampling

of cattle breath with solid phase microextraction, Biosystems Engineering, 84(2), 239–

246, 2003.

[2] A. W. Boots, J. J. B. N. Berkel, J. W. Dallinga, A. Smolinska, E. F. Wouters, F. J.

Schooten, The versatile use of exhaled volatile organic compounds in human health and

disease, J. Breath Res. 6 027108, 2012.

[3] J. Gamble, “Effects of formaldehyde on the respiratory system.” In: Gibson JE, ed.

Formaldehyde toxicity. Hemisphere & McGraw Hill, Washington DC, 1983, 175-197.

[4] M. Ritchie and R. G. Lehnen, “Formaldehyde-related Health Complaints of Residents

Living in Mobile and Conventional Homes,“ Am. J. Public Health, 77, pp. 323-328,

1987.

[5] H. Kudo, M. Sawai, Y. Suzuki, X. Wang, T. Gessei, D. Takahashi, T. Arakawa, K.

Mitsubayashi, Fiber-optic bio-sniffer (biochemical gas sensor) for high-selective

monitoring of ethanol vapor using 335 nm UV-LED, Sensor Actuator B Chem., 132(2),

676-680, 2010.

[6] J. Tonzetich, P. A. Carpenter, Arch. Oral Biol. 16 (1971) 599–607.

[7] N. F. Schmidt, S. R. Missan, W. J. Tarbet, Oral Surg. Oral Med. Oral Pathol. 45

(1978) 560–567.

[8] K. Mitsubayashi, T. Minamide, K. Otsuka, H. Kudo, H. Saito, “Optical bio-sniffer for

methyl mercaptan in halitosis”, Analytica Chimica Acta 573–574 (2006) 75–80.

[9] T. Minamide, K. Mitsubayashi, H. Saito, “Bioelectronic sniffer with monoamine

oxidase for methyl mercaptan vapor”, Sensors and Actuators B 108 (2005) 639–645.

[10] K. Mitsubayashi, Y. Hashimoto, “Bioelectronic nose for methyl mercaptan vapor using

xenobiotic metabolizing enzyme: avin-containing monooxygenase”, Sensors and

Actuators B 83 (2002) 35-40.

[11] T. Minamide, K. Mitsubayashi, N. Jaffrezic-Renault, K. Hibi, H. Endo, H. Saito,

“Bioelectronic detector with monoamine oxidase for halitosis monitoring”, Analyst 130

11 (2005) 1490-1494.

[12] V. J. Cogliano, Y. Grosse, R. A. Baan, K. Straif, M. B. Secretan, F. El Ghissassi,

Working Group for Volume 88, Environ. Health Perspect. 113, (2005) 1205–1208.

[13] L.E. Beane Freeman, A. Blair, J.H. Lubin, P.A. Stewart, R. B. Hayes, R.N. Hoover, M.

Hauptmann, “Mortality from lymphohematopoietic malignancies among workers in



formaldehyde industries: the National Cancer Institute cohort”, J. Natl. Cancer Inst.

101, (2009) 751–761.

[14] J.R. Beall, A.G. Ulsamer, “Formaldehyde, Formaldehyde and hepatotoxicity: a review”,

J. Toxicol. Environ. Health 14, (1984) 1–21.

[15] K. Mitsubayashi, G. Nishio, M. Sawai, E. Kazawa, H. Yoshida, T. Saito, H. Kudo, K.

Otsuka, M. Takao, H. Saito, “A biochemical sniffer-chip for convenient analysis of

gaseous formaldehyde from timber materials”, Microchim. Acta (2008) 160: 427–433.

[16] H. Kudo, T. Yagi, M.X. Chu, H. Saito, N. Morimoto, Y. Iwasaki, K. Akiyoshi, K.

Mitsubayashi, “Glucose sensor using phospholipid polymer based enzyme

immobilization method”, Anal. Bioanal. Chem. 391, (2008) 1269–1274.

[17] H. Kudo, Y. Suzuki, T. Gessei, D. Takahashi, T. Arakawa, K. Mitsubayashi,

“Biochemical gas sensor (bio-sniffer) for ultrahigh-sensitive gaseous formaldehyde

monitoring”, Biosensors and Bioelectronics 26 (2010) 854–858.

[18] H. Kudo, X. Wang, Y. Suzuki, M. Ye, T. Yamashita, T. Gessei, K. Miyajima, T.

Arakawa, K. Mitsubayashi, “Fiber-optic biochemical gas sensor (bio-sniffer) for sub-

ppb monitoring of formaldehyde vapor”, Sensors and Actuators B 161 (2012) 486–492.

[19] H. Kudo, T. Yamashita, K. Miyajima, T. Arakawa, K. Mitsubayashi,, “NADH-

fluorometric Biochemical Gas Sensor (Bio-Sniffer) for Evaluation of Indoor Air

Quality”, IEEE Sensors Journal, 13 8 (2013) 2828-2833.

[20] T. Koshida, T. Arakawa, T. Gessei, D. Takahashi, H. Kudo, H. Saito, K. Yano, K.

Mitsubayashi, “Fluorescence biosensing system with a UV-LED excitation for L-

leucine detection” Sensors and Actuators B 146 (2010) 177-182.

[21] T. Arakawa, T. Koshida, T. Gessei, K. Miyajima, D. Takahashi, H. Kudo, H. Saito, K.

Yano, K. Mitsubayashi, “Biosensor for L-phenylalanine based on the optical detection

of NADH using a UV light emitting diode”, Microchimica Acta, 173 (2011) 199-205.

[22] J. P. Spinhirne, J. A. Koziel, N. K. Chirase, A device for noninvasive on-site sampling

of cattle breath with solid phase microextraction, Biosystems Engineering, 84(2), 239–

246, 2003.

[23] A. W. Boots, J. J. B. N. Berkel, J. W. Dallinga, A. Smolinska, E. F. Wouters, F. J.

Schooten, The versatile use of exhaled volatile organic compounds in human health and

disease, J. Breath Res. 6 027108, 2012.

[24] G. Shuster, Z. Gallimidi, A. H. Reiss, E. Dovgolevsky, S. Billan, R. A-Bortnyak, A.

Kuten, A. Engel, A. Shiban, U. Tisch, H. Haick, Classification of breast cancer

precursors through exhaled breath, Breast Cancer Research and Treatment, 126, 3,

791-796, 2011.

[25] G. Peng, U. Tisch, O. Adams, M. Hakim, N. Shehada, Y. Y. Broza, S. Billan, R. A.

Bortnyak, A. Kuten, H. Haick, Diagnosing lung cancer in exhaled breath using gold

nanoparticles, Nature Nanotech., 14 669-673, 2010.

[26] P. J. Mazzone, Exhaled breath volatile organic compound biomarkers in lung cancer J.

Breath Res., 6, 027106, 2012.

[27] M. Philips, Method for the collection and assay of volatile organic compounds in breath

Anal. Biochem., 247, 2272-2278, 1997.

[28] P. Fletcher, K. N. Andrew, A. C. Calokerinos, S. Forbes, P. J. Worsfold, Applications

of Flow Injection Chemiluminescence Detection - A review Luminescence, 16, 1-23,

2001.



[29] X. Wang, E. Ando, D. Takahashi, T. Arakawa, H. Kudo, H. Saito, K. Mitsubayashi, 2D

spatiotemporal visualization system of expired gaseous ethanol after oral administration

for real-time illustrated analysis of alcohol metabolism, Talanta, 82, 892–898, 2010.

[30] X. Wang, E. Ando, D. Takahashi, T. Arakawa, H. Kudo, H. Saito, K. Mitsubayashi,

Non-invasive spatial visualization system of exhaled ethanol for real-time analysis of

ALDH2 related alcohol metabolism, Analyst, 136 (2011) 3680-3685.

[31] T. Arakawa, X. Wang, T Kajiro, K. Miyajima, S. Takeuchi, H. Kudo, K. Yano, K.

Mitsubayashi, A direct gaseous ethanol imaging system for analysis of alcohol

metabolism from exhaled breath, Sensors and Actuators B: Chemical, 186 (2013)

27–33.





Chapter 10

DESALINATION OF BRACKISH WATER BY

ELECTRODIALYSIS: EFFECTS OF OPERATIONAL

PARAMETERS AND WATER COMPOSITION ON

PROCESS EFFICIENCY

Mourad Ben Sik Ali*, Amor Hafiane,

Mahmoud Dhahbi and Béchir Hamrouni Desalination and Water Treatment Research Unit, Chemistry Department,

Faculty of Sciences of Tunis, University of El Manar, Manar II, Tunisia

Engineering Preparatory Institute of Nabeul, University of Carthage,

Merazka, Nabeul, Tunisia

Water Researches and Technologies Center, Soliman, Tunisia

ABSTRACT

Electrodialysis is an electro-membrane process for separation of ions across charged

membranes from one solution to another under the influence of an electrical potential

difference used as a driving force. The transfer of the charged species is carried out

according to a mechanism of exchanges of ions between the ions of the solution and the

counter ions of the membrane.

This chapter is dealing with the effectiveness of the desalination of brackish water by

this process. A laboratory scale electrodialysis cell was used for this purpose.

The desalination of brackish solutions containing only one salt according to the

continuous mode (single pass process) was carried out initially. This study revealed that

the effectiveness of the process is dependent on the operational parameters of the

electrodialysis cell. But the rate of desalination is relatively weak. This rate does not

exceed 50% and sometimes does not allow reaching the awaited results.

Study of another configuration was carried out. This configuration is the

discontinuous mode or total recirculation mode (batch process). In this mode, the

solutions are recycled in the cell until the desired concentration is reached. While

* Desalination and Water Treatment Research Unit, Chemistry Department, Faculty of Sciences of Tunis, 2092

Manar II, Tunisia. Email: [email protected].


Mourad Ben Sik Ali, Amor Hafiane, Mahmoud Dhahbi et al. 296

working with this mode, the desalination rates of various solutions reached quickly high

values. Indeed, they exceeded the 75% rate in few minutes.

This study concluded also that the composition of the solution itself influences the

effectiveness of desalination process. Indeed, the transfer of the ions from one

compartment to another depends on the various ionic species present in solution.

The presence of the calcium ions largely decreases the transfer of the magnesium

ions and vice versa. This is explained by the competition effect between two ions of same

valence during their transfers through exchanging membranes.

Moreover, the rates of transfer of the cations and more precisely calcium and

magnesium ions undergo reductions in the presence of anions and in particular sulfate

ions. The desired rates of desalination are also obtained for longer handling times which

have an immediate effect on energy consumption. In fact, the consumption of energy is

more important for more complex solutions.

Keywords: Brackish water, desalination, operational parameters, water composition, process

efficiency

INTRODUCTION

The drinking water is an invaluable and vital product. The vast majority of the water of

the sphere consists of sea water. Approximately 2.5 % are fresh water, and the two-thirds of

this one are in the form of icecaps (El-Dessouky & Ettouney, 2002).

Rivers and lakes contain a small percentage of planet water but this surface water is

crucial. Like certain aquifers, they are constantly renewed by the water cycle.

The groundwater reserves constitute an important source of water for many people. The

groundwater is often subject of water quality deterioration. The coastal aquifers can contain

sea water since prehistory, when their sediments settled. The sea water intrusion is also

possible because of the excessive pumping of the fresh water. The aquifers of the arid regions

can also contain brackish water. In this case, salts are usually concentrated by the evaporation

of closed basins or by depressions.

Modern technologies of desalination can extract salts from sea water and brackish water,

thus ensuring a new source of fresh water. The desalination processes available on the market

are divided into two main categories: thermal processes and membrane processes (Alghoul,

Poovanaesvaran, Sopian, & Sulaiman, 2009; Greenlee, Lawler, Freeman, Marrot, & Moulin,

2009; Sadrzadeh & Mohammadi, 2008).

Among the thermal processes, distillation is distinguished. It is one of the most frequently

employed methods. Approximately half of the de-salted water in the world is produced by

means of heat being used to distill fresh water starting from sea water. The distillation process

reproduces the natural cycle of water since it consists in heating salt water for the production

of steam which is in its turn condensed to form fresh water (El-Dessouky & Ettouney, 2002).

In nature, the membranes play a big role to separate salts. Indeed, dialysis and osmosis

processes occur in the living organisms (Baker, 2004; Noble & Stern, 1995). The membranes

are used for water treatment in two important processes: electrodialysis (ED) and reverse

osmosis (RO).

These membrane processes belong to new technologies which can play an important role

for the treatment of wastewater and the production of drinking water. However, the


Desalination of Brackish Water by Electrodialysis 297

application of the membrane processes and especially reverse osmosis encounter difficulties.

This is due to the formations of polarization layer and then membrane fouling. This

phenomenon presents the acutest problem of membrane filtration. It reduces the productivity

of the membranes and thus contributes to increase the energy expenses, to increase the

frequency of washings and to possibly reduce the lifespan of the membranes (Côté, Weirun,

Gierlich, Natarajan, & Taniguchi, 1996; Danis, 2003; Greenlee et al., 2009; Matsuura, 2001).

Electrodialysis is an electro-membrane process for separation of ions across charged

membranes from one solution to another under the influence of an electrical potential

difference used as a driving force. The transfer of the charged species is carried out according

to a mechanism of exchange of ions between the ions of the solution and the counter ions of

the membrane.

This process has been widely used for production of drinking and process water from

brackish water and seawater, treatment of industrial effluents, recovery of useful materials

from effluents and salt production (Danis, 2003; Matsuura, 2001; SenGupta, 2007;

Zagorodni, 2007). This chemical-free technology competes with reverse osmosis. It shows

better resistance to fouling and scaling. It also has an economical advantage in desalination of

low salinity solution (less than 5 g L-1

as total dissolved salts (TDS)). Also it should be kept

in mind that because of low chemicals consumption, ED is an environmental friendly process

for water desalination (Dardel, 1998).

In this chapter we are interested to study the effectiveness of the desalination process by

ED of synthetic solutions and real sample. The various parameters which can influence this

process were studied. These parameters include those related to the operation of the pilot of

electrodialysis and those related to the nature of the solution and its composition. An

energetically approach was carried out.

ELECTRO-MEMBRANE PROCESSES: OVERVIEW

The electro-membrane techniques are separative techniques implementing the ions

exchange membranes. The driving force in these processes is a gradient of electric potential

which causes an electric current and by consequence leads to the separation of ionic species.

Ion exchange is a process wherein ions of a certain charge contained in a solution (eg

cation) are removed from the solution by adsorption on a solid material (the ion exchanger),

to be replaced with an equivalent amount of other ions of similar charge emitted from the

solid. The opposite charged ions are not affected (Inglezakis & Poulopoulos, 2006; SenGupta,

2007; Zagorodni, 2007).

The ion exchange reactions are reversible and selective; they are governed by the law of

chemical equilibrium. They take place until the concentrations of various ions reach certain

specific proportions (Zagorodni, 2007).

An ion exchange membrane (IEM), also known as "ion-permeable membrane" or "ionic

membrane" generally consists of a macromolecular material, more or less crosslinked in a

three-dimensional network insoluble in water, to which are covalently attached ionized or

ionizable functional groups also called fixed ions. These groups are electrically neutralized by

mobile ions of opposite sign called compensating ions or counter ions. The co-ions are ions

having the same charge sign as the fixed sites. They are ideally excluded from transfers



through the membrane. Counter ions and co-ions are components of the electrolyte solution

(Figure 1). (Baker, 2004; C. Hannachi, Hamrouni, & Dhahbi, 2009; Ch Hannachi,

Bouguecha, Hamrouni, & Dhahbi, 2008; Kontturi, Murtomäki, & Manzanares, 2008; Noble

& Stern, 1995; Heiner Strathmann, 2004; Yoshinobu Tanaka, 2007). The fixed ions are the

basis of the specificity of the membrane. If they are negatively charged, membrane is then

selective to cations and is then called cation exchange membrane (CEM), otherwise it is an

anion exchange membrane (AEM).

Figure 1. Structure of an ion exchange membrane (case of CEM) (Krol, 1997).

The transfer of the charged species is carried out according to a mechanism of ions

exchange between the ions of the solution and counter ions of the membrane (Baker, 2004;

Hoffman, 2003; Kontturi et al., 2008; Matsuura, 2001; Nagarale, Gohil & Shahi, 2006; Noble

& Stern, 1995; Sadrzadeh & Mohammadi, 2008; Shaposhnik, Zubets, Mill & Strigina, 2001;

H. Strathmann, 2004; Yoshinobu Tanaka, 2007; Wilhelm, 2001).

In the case of a cation exchange membrane, under the effect of an electric field,

positively charged ions (for example Na+) will move in the network of the immobilized anion

functional groups and thus will cross the membrane.

The anions for their part (for example Cl-) will be retained by the cation exchange

membrane. Effectively they will be repulsed by the anionic functional groups forming the

membrane.

In the case of an anion exchange membrane, the anions (as Cl-) will move in the network

of the cation functional group of the membrane.

Figure 2. Principle of operation of an (a) anion and (b) cation exchange membrane.



Figure 2 illustrates the principle of operation of an (a) anion and (b) cation exchange

membrane.

Electro-membrane techniques have seen for a few years their potential fields of

application widening in an important way. This is explained by the appearance on the market

of new generations of membranes having an improved chemical resistance (Nagarale et al.,

2006).

Among these techniques we distinguish mainly:

- Electrodialysis (ED) known as conventional electrodialysis

- Electrodialysis with bipolar membrane (EDBP)

- Electro-electrodialysis (EED)

- Electrodeionisation (EDI)

THE CONVENTIONAL ELECTRODIALYSIS

1. Principle

Conventional electrodialysis (ED) is the most applied configuration among

electromembrane processes. Its basic principle was discussed earlier and is shown in figure 2.

In an electrodialysis unit, the cationic membranes (CEM, cation permeable) and anionic

membranes (AEM, anion permeable) are arranged in a parallel and alternate way.

The injection of current in the system is assured by two parallel electrodes to membranes

plan. They are placed at the ends of the electrodialysis cell.

Figure 3. Principle of conventional electrodialysis.



The ionized dissolved species, such as salts, acids or bases, are transported across the

ionic membrane under the action of an electric current. The CEM blocks the anions and let

passing only cations. Whereas AEM blocks the cations and let anions transfers. As

consequence, two différents compartiments are created: concentrating compartment

(concentrate) and dilution compartement (dilute). Solutions are renewed in the compartments

by a fluid circulation parallel to membrane plan.

The principle of conventional electrodialysis is presented by Figure 3.

In industrial plants, stacks can reach several hundreds of unitaire cells in filter press

assemblies (Heiner Strathmann, 2004).

2. Operation Modes

There are four different operation modes under which can works the electrodialysis

process (MOÇOTÉGUY, 1999):

a. Continuous Mode with Direct Passage

The solution passes only once in the cell (figure 4). To achieve the desired output

concentration, a cascade of cells is used. This type of operation is also called "single pass

process."

Figure 4. Schematic diagram of a continuous process with direct passage.

b. Discontinuous Mode or Total Recirculation

The solutions are recycled in the cell until the desired output concentration is reached

(figure 5). This type of operation is also called "batch process".

Figure 5. Schematic diagram of a "batch process".



c. Partial Recycling Mode

It is a combination of the previous two modes. A portion of the solution outlet of the cell

is recycled to the inlet (figure 6). It is also called "feed and bleed."

Figure 6. Schematic diagram of a "feed & bleed process".

d. Semi-Continuous Mode

One compartment operates continuously (direct or feed & bleed) while the other operates

in discontinuous. (Figure 7)

Figure 7. Schematic diagram of a "semi-continuous process".

3. Electrdodialysis Limitation: The Limiting Current

In electrodialysis, concentration polarization can take place at the membrane surface. The

magnitude of the concentration polarization is a function of various parameters including the

applied current density, the feed flow velocity parallel to the membrane surface, the cell

design, and the membrane properties (Ben Sik Ali, Mnif, Hamrouni, & Dhahbi, 2010; Lee,

Sarfert, Strathmann, & Moon, 2002; Lee, Strathmann, & Moon, 2006; Y. Tanaka, 2002,

2005). The transport of charged species to the anode or cathode through a set of ion exchange

membranes leads to a concentration decrease of counter-ions in the laminar boundary layer at



the membrane surface facing the dilute cell (Csd) and an increase at the surface facing the

concentrate cell (Csc). When the ion concentration at membrane surface will approach to zero,

the current density will approach to the maximum value. At this condition the applied current

is defined as the limiting current density (LCD) (Lee et al., 2006).

Figure 8. Schematic diagram of concentration polarization in electrodialysis process.

The limiting current density in the electrodialysis process is an important parameter

which determines the electrical resistance and the current utilization. It is generally believed

that an electrodialysis process shows higher electrical resistance or lower current utilization

when operated at above LCD. It also can give rise to problems such as water dissociation or

salt precipitation. It is necessary, therefore, to determine the LCD to prevent problems and

operate the electrodialyzer successfully (Y. Tanaka, 2002).

MATERIAL AND METHODS

1. Electrodialysis Equipment and Membranes

The ED setup consists of: a power DC, a brine tank, a feed tank, an electrode rinse tank,

three centrifugal pumps (P = 84W, total head = 4.2 m) equipped each with a flowmeter and

three valves to control feed flow rate in the compartments of ED cell. Figure 9 shows a

simplified diagram of the ED setup working in continuous mode and recirculation mode.

The electrodialysis operation was carried out on a laboratory stack “PCCell ED 64 002”

supplied by PCA-Polymerchemie Altmeier GmbH and PCCell GmbH, Heusweiler, Germany.

ED cell is packed with ion exchange membranes (cation and anion), spacers and a pair of

electrodes (anode and cathode). Both electrodes are made of Pt/Ir- coated Titanium. The

membranes and spacers are stacked between the two electrode-end blocks. Plastic separators

are placed between the membranes to form the flow paths of the dilute and concentrate

streams. These spacers are designed to minimize boundary layer effects and are arranged in

the stack so that all the dilute and concentrate streams are manifolded separately. In this way



a repeating section called a cell pair is formed; it consists of a cation exchange membrane, a

dilute flow spacer, an anion-exchange membrane, and a concentrate flow spacer. In this

chapter, experiments were carried out by this stack equipped with three cation exchange

membranes (CEM) and two anion exchange membranes (AEM).

Figure 9. Schematic of the ED system used in this study (a) continuous mode (b) batch recirculation

mode.

For each membrane, the active surface area is 0.0064 m2 and the flow channel width

between two membranes is 0.5 mm.

Figure 10. Schematic of the ED cell.

Table 1. Information on PCA standard cation and anion exchange membranes

Membrane Thickness

(μm)

Ion exchange

capacity

(meq g-1

)

Chemical

stability

(pH)

Permselectivity Functional

groups

Surface potential

(Ω cm2)

PC-SK 130 ~1 0-11 > 0.93 -SO3- 0.75-3

PC-SA 90-130 ~1.5 0-9 > 0.96 -NR4- 1-1.5



PC-SK standard cation exchange membranes and PC-SA standard anion exchange

membranes are used in the stack. Information about the membranes is given in Table 1, which

was supported by the manufacturer.

2. Reagents

Analytical grade sodium chloride, magnesium chloride, calcium chloride and sodium

sulfate salts are used in all experiments to produce solutions with known amount of salts and

electrode rinse solution. Distilled water was used throughout.

ANALYTICAL METHODS

In all the experiments, Magnesium (Mg) and calcium (Ca) amounts were determined by

atomic absorption spectroscopy using an analytical AAS Vario 6 spectrometer.

Sodium (Na) and potassium (K) were evaluated by atomic emission spectroscopy using a

Jenway PFP 7 spectrometer.

Sulfate and chloride anions were determined by ion chromatography using a Metrohm

761 Compact IC with conductivity detection (Metrohm AG, Switzerland) and chemical

suppression. The anion measurements were performed with a Metrosep anion dual 2 column

(4.6×75 mm) with a particle diameter of 6 μm.

A conductometer (712 Conductometer, Metrohm AG, made in Switzerland) was used to

measure the electrical conductivity and temperature of samples. Water conductivity directly

depends on its salt content.

654 pH-metre (Metrohm AG, Switzerland) equipped with a glass electrode was used for

measuring pH solutions

DATA ANALYSIS

The purpose of the experiments was to study the effects of operational parameters

(voltage, flow rate, circulation mode) and water composition on ED cell performance.

All figures and tables refer to concentration changes in the dilute. The mass balance of

the ions present in the feed solution was verified for dilute, concentrate and electrode rinse

solution. Thus, data for concentrate and electrode rinse solution will not be presented in this

chapter.

1. Determination of the Demineralization Rate

In this study, the quality characteristic was the demineralization rate which can be

calculated as follow (Casademont, Farias, Pourcelly, & Bazinet, 2008; Wang, Ying, Jiang,

Yang, & Jahangir, 2009):



0

(%) 100 1 tECDR

EC

Where DR is the demineralization rate expressed in percentage, EC0 and ECt are,

respectively, the initial and conductivity of the dilute, expressed in µS cm-1

.

2. Determination of the Ionic‟s Transport Flux

The flux values were evaluated for all experimental conditions in order to compare the

transport of ions from feed to receiver phase. The flux of ion (J) was determined by using the

following equation (Ergun, Tor, Cengeloglu, & Kocak, 2008):

2 1( )V C

J mol cm sA T

Where V is the volume of receiver phase (L), A is the effective membrane area (cm2),

C is the transported amount of ions at a time (mol L-1

) and T is the period time (s).

3. Determination of the Specific Power Consumption

Specific power consumption (SPC) can be described as the energy needed to treat unit

volume of solution. SPC was calculated for each experimental condition using the following

equation (Kabay, Arar, Samatya, Yüksel, & Yüksel, 2008):

0

( )

t

E I t dt

SPCV

Where E is the applied potential, I the current, V the dilute stream volume and t is the

time.

The SPC is a key parameter of desalination by electrodialysis in competing with the other

membrane processes.

4. Determination of the Limiting Current

The limiting current can be determined experimentally by plotting the electrical

resistance across the membrane stack (E I-1

) or the pH value in the dilute cell as a function of

the reciprocal electric current (I-1

). This is called a Cowan–Brown plot after its original

developers (Baker, 2004; Y. Tanaka, 2002).



Usually, the limiting current depends on membrane and solution properties as well as on

the electrodialysis stack construction and various operational parameters such as the flow

velocity (flow rate) of the dilute solution. The set of primary tests was done to determine the

limiting current density for the operating conditions to be adopted in the study when we work

at batch recirculation mode.

Figure 11. Experimental determination of Ilim by the Cowan–Brown method.

RESULTS AND DISCUSSION

Desalination of Synthetic Solution

Synthetic brackish water solutions with known amount of compounds were prepared by

dissolving Sodium chloride, Calcium chloride, Magnisium chloride and Sodium sulfate in

distilled water. Ionic strength was fixed by adjusting the concentration of sodium chloride.

Prior to the experiments, pH was adjusted by the addition of 1 M HCl and/or NaOH.

In order to prevent the generation of chlorine or hypochlorite, which could be hazardous

for the electrodes, 0.1 M Na2SO4 solution, prepared by dissolving Sodium sulfate salt in

distilled water, was used as electrode rinse solution circulating in electrode compartments.

Flow rate of electrode rinse solution was fixed to 80 L h-1

for all experiments.

During the experiments, the same synthetic brackish water solutions were used as initial

concentrate and dilute solutions. Their flow rates (dilute and concentrate) were fixed at the

beginning of experiment. Total voltage drop, including voltage drop in the membrane stack as

well as on the electrodes, was also fixed at the start of the experiment.

In this chapter, only single pass process and batch process modes were studied.

Samples, to be analyzed, were collected at the inlet (before treatment) and outlet (after

treatment) of each compartment of electrodialysis cell.

In order to remove any deposits, cleaning solutions of 0.1 M HCl, 0.1 M NaOH and

distilled water were circulated through the ED cell for 30 min each at the end of experiment.



a. Single Pass Process Mode

Effects of Applied Voltage

To study the effect of applied voltage, the ionic strength, pH, and the inlet velocity of

feed solution were fixed respectively to 0.025 M, 5, and 17 L h-1

.

Conductivity and consequently salts concentration change in the dilute solution as

function of applied voltage is presented in figure 12.

We note that applied potential difference has a direct influence on the evolution of the

conductivity in dilute compartment. Indeed, an increase of this parameter leads to a decrease

of the dilute conductivity and so the salt concentration in this compartment.

Figure 12. Effect of applied voltage on dilute solution conductivity (ionic strength 0.025 M; pH 5; feed

flow rate 17 L h-1).

Figure 13. Effect of applied voltage on the demineralization rate (ionic strength 0.025 M; pH 5; feed flow

rate 17 L h-1

).



This can be explained by the transfer of these entities between the two compartments:

dilute and concentrate. There will be a migration of ions from dilute to concentrate. This

migration is exacerbated by the increase of the applied potential across the stack.

The influence of applied voltages on the demineralization rate is presented in figure 13.

Figure 13 shows that the demineralization rate strongly depends on the applied voltage.

There are three distinguished parts in this curve. In the first part (E < 2 V) the

demineralization increases only marginally with the increasing voltage drop. Previous work

(Ben Sik Ali, Ennigrou, & Hamrouni, 2013; Ben Sik Ali et al., 2010) showed that in this

region the resistance of the stack is relatively high. The applied voltage is not so important to

overcome the resistance of membranes and to induce the transport of several ions. At these

conditions there are few transports of ions between the dilute and concentrate compartments.

This is also confirmed in figure14 which present the effect of applied voltage on ionic

transport or ionic flux. Effectively low ionic fluxes are obtained in this region.

Figure 14. Effect of applied voltage on the ionic transport (ionic strength 0.025 M; pH 5, feed flow rate

17 L h-1

).

In the second part (2 V < E < 12 V), Figure 13 and 14 shows a more or less linear

increase of the demineralization rate and ionic flux. The transport of ions between the

compartments of cell is proportional to current.

In the third part (E > 12 V), the increase in the applied voltage does not lead to a

significant increase in the demineralization rate and ionic flux. At this point the limiting

current density is reached and there is no increase in the current density neither ionic transport

from dilute to concentrate compartment when the applied voltage increases. In this range of

applied voltage and as it is shown in Figure 15, the current efficiency decreases by increasing

the applied voltage. The induced current is not only used to ions transport but also to split

water. Same results were obtained by Ben Sik Ali et al.(Ben Sik Ali et al., 2013; Ben Sik Ali

et al., 2010).

As shown in Figure 16, the specific power consumption increases by increasing the

applied voltage. It is low in the first region (0-2 V). This is expected because the specific

power consumption depends of the applied voltage and current which were weak.



Figure 15. Effect of applied voltage on the current efficiency (ionic strength 0.025 M; pH 5; feed flow

rate 17 L h-1

).

Figure 16. Effect of applied voltage on the specific power consumption (ionic strength 0.025 M; pH 5;

feed flow rate 17 L h-1

).

Effects of Flow Rate

The ionic strength and pH of feed solution were fixed respectively at 0.025 M and 7. The

flow rate of dilute compartment was varied from 5 to 15 L h-1

. Experiments were performed

by applying 2, 5 and 8 V each time across the entire assembly. The influences of the flow rate

on the ED performances are illustrated in figure 17, 18, 19 and 20.

In figure 17 we notice that the DR decreases with an increase of the flow rate. This can

be explained by the remaining time of ions inside the different compartments of the cell. In

fact the ions have more time to be transferred from one compartment to another thought the

membrane when the velocity or flow rate is lower. This can lead to an increase of the total

amount of salts transferred and consequently to an increase of separation percentage. These



interpretations are confirmed in figure 18. In fact the ionic flux transport of salts decreases by

the increase of the feed flow rates. For that, low flow rates are recommended to get maximum

separation

Figure 17. Effect of flow rate on the demineralization rate (ionic strength 0.025 M; pH 7).

Figure 18. Effect of flow rate on the ionic flux (ionic strength 0.025 M; pH 7).



Figure 19. Effect of flow rate on current efficiency (ionic strength 0.025 M; pH 7).

Figure 20. Effect of flow rate on the specific power consumption (ionic strength 0.025 M; ; pH 7).

In other hand and as depicted in figure 19 the current efficiency is not affected by the

variation of flow rate in these conditions. All current was used to transport salts.

As illustrated in figure 20 the specific power consumption is higher for low flow rate.

However, in this case, SPC amount is still relatively low.

Effects of Ionic Strength

The effect of the feed solution ionic strength on the removal of salts was investigated

using aqueous solutions at various concentrations of NaCl salt. The ionic strength of feed

solution was varied from 5 10-4

M to 5 10-2

M. The pH of prepared solutions was fixed to 7.

The flow rate of dilute and concentrate were fixed to 12 L h-1

. 2, 5 and 8 V was respectively



applied across the entire assembly. Figures 21, 22, and 23 represent the effects of the ionic

strength of feed solution on electrodialysis performances.

As shown in figure 21 the demineralization rate has a considerable dependency on the

feed solution ionic strength. Effectively at some hydrodynamique and electrical conditions an

increase of the initial ionic strength or salts concentration leads to the decrease of the DR. As

depicted on figure 22 the number of ions transported through the membranes is almost the

same but total amounts of salts are quite different from the different treated solution. As

known the calculation of DR depends strongly on the initial feed concentrations and the

amount of transported ions. So the DR evolves reciprocally to the initial feed concentration at

some hydrodynamique and electrical conditions.

Figure 21. Effect of ionic strength on the demineralization rate (flow rate 12 L h-1

; pH 7).

Figure 22. Effect of ionic strength on the ionic flux (flow rate 12 L h-1

; pH 7).



Figure 23. Effect of ionic strength on the specific power consumption (flow rate 12 L h-1

; pH 7).

Figure 24. Experimental determination of Ilim by the Cowan–Brown method. (ionic strength 0.01 M; feed

flow rate 15 L h-1

).



As depicted in figure 23 the specific power consumption are low for lowest initial feed

concentration and mainly for lowest applied voltage. In other hand the current efficiency is

not affected by the variation of flow rate in these conditions. All current was used to transport

salts.

It was not possible to increase the ionic strength of initial feed solution above 0.05 M

since an electrode overheating appeared in the present system as a result of the increase of

cell conductivity. The DR was very low and insignificant. For these reasons it is suitable to

use this process for the demineralization of solution with salinity under 0.05 M.

b. Batch Process Mode

Determination of the Limiting Current (Ilim)

In this study the limiting current (Ilim) was determined by measuring the potential and the

cell resistance as a function of the applied current. Figure 24 shows the curves illustrating the

determination of the limiting current from the experimental result obtained with a 0.01 M

NaCl solution and a feed flow velocity of 15 L h-1

. In the relationship between the dilute pH

and the current in figure 24 (a), the limiting current was determined where the slope was

changed due to water dissociation on the surface of ion exchange membranes (Baker, 2004;

Lee et al., 2006). In addition, the limiting current was determined from the graph showing the

cell resistance versus the reciprocal of the current (figure 24 (b)).

NaCl Solutions with different concentrations were used to determine these values at

different flow rates. The obtained results were summarized in Table 2.

Table 2. Limiting current values as function of flow rate and feed concentration

Flow rate (L h-1) Feed solution concentration (g L-1)

0.5 1 1.5 2.5

5 0.2 0.4 0.82 1.16

15 0.34 0.62 0.98 1.52

25 0.46 0.86 1.15 1.96

30 0.52 1 1.26 2.2

The applied current in all experiment was fixed from the beginning to a value below the limiting current

(I = 0.8 Ilim)(Balmann & Casademont, 2006).

Effect of Initial Feed Concentration

The effect of initial feed concentration on the desalination process was investigated using

aqueous solutions at various concentrations of NaCl salt (0.01, 0.02, 0.03 and 0.05 M) and

fixed flow rate of 25 L h-1

. As shown in Figures 25, 26 and 27, the initial salt concentration

has a significant effect on the process efficiency. The most important observation is that total

process time increased by increasing initial concentration in the feed. Consequently the

specific power consumption is closely depending on it as seen in Figure 27.

The ionic flux also is closely depending on the initial ionic strength of solution. Higher

ions transfert were obtained for highest initial concentration. These results can be explained

by the increase of number of ions in solutions when the concentration of salts increases. As a

result a better transfer throught membrane can appear.



Figure 25. Effect of ionic strength on the demineralization rate (flow rate 25 L h-1

).

Figure 26. Effect of ionic strength on the ionic flux (flow rate 25 L h-1

).

Effect of Feed Flow Rate

The operation was carried out using different feed flow rates (5, 15, 25 and 30 L h-1

) and

a solution with 0.05 M as ionic strength.

As seen in figure 28, flow rate has a significant effect on desalination process in the range

studied. Fewer flow rates were better for the performance of ED as was believed by

Sadrzadeh et al. (2007) (Sadrzadeh, Razmi, & Mohammadi, 2007). This was because the

treated solution could stay longer in the ED device and the ions could transfer freely in the

membrane. However, the ions had relative shorter retention time in the dilute tank at high

flow rate, which resulted in lower removal efficiency (Wang et al., 2009).

In other hand and as depicted in figure 29 the specific power consumption depends also

on this parameter. The SPC is lower for low flow rate. In fact the SPC depends mainly on two

parameters: the applied current and the time of experiment. Firstly the applied current (the

limiting current) is depending on the flow rate as shown in Table 3. It is lower for low flow

rate. Secondly the experiment time to obtain the same results (removal 85 % of salt)

augmented by increasing the flow rate.



Figure 27. Effect of ionic strength on the specific power consumption (flow rate 25 L h-1

).

Figure 28. Effect of flow rate on the demineralization rate (ionic strength 0.05 M).

Figure 29. Effect of flow rate on the specific power consumption (ionic strength 0.05 M).



Effect of Coexisting Ions

In this section, we will study the demineralization of saline solutions containing multiple

anionic and cationic species. These solutions were prepared by dissolving: NaCl, MgCl2,

CaCl2 and Na2SO4 in distilled water.

The total salt concentration of different solutions and their flow rates were set at the

starting of the experiment respectively to 3 g L-1

and 25 L h-1

.

The current between the electrodes was imposed and maintained constant. It was equal to

80% of the limiting current. The end of the experiment was announced by the obtention of a

dilute solution with a total salt concentration less than 0.5 g L-1

or conductivity lower than

400 µS cm-1

.

Figure 30 illustrates the variation of demineralization rates over time during the

desalination of solutions of different compositions. From these curves we find that the

presence of several salts affects the demineralization rate. Indeed, at same total dissolved salts

amount (TDS), demineralization rate decreases when more various salts are present. As

consequence, the duration of the desalination process increases.

The evolution of the energy consumption and the desalination process total duration of

studied solutions are presented in figure 31.

Figure 30. Demineralization rate variation as a function of time during the desalination of solutions of

different compositions (ionic strength 0.05 M; I: 1.96 A; feed flow rate 25 L h-1

)

From this figure we see that the total duration of the process increases with the addition

of salts. In fact, for a solution containing only sodium chloride, the dilute can achieve

desirable salinity after 60 min, when 90 min is required to achieve the same salinity for a

solution containing a mixture of several salts.

What is also remarkable is that the energy consumption increases for solutions containing

various salts. This is expected since the energy consumption is defined as a function of time

on the one hand and the increase in cell resistance on the other hand.

We proposed to study the behavior of each species during the desalination of a brackish

solution containing various salts. For this we have traced the evolution of the ion flux of each

species over time in Figure 32.



Figure 31. Evolution of the specific power consumption and the desalination process duration of

solutions with different compositions (ionic strength 0.05 M; I 1.96 A; feed flow rate 25 L h-1

).

Figure 32. Variation of the ionic flux of Ca2+

, Mg2+

and SO42-

versus time during the desalination

process. (Flow rate 25 L h-1

, TDS 3 g L-1

, I 1.96 A).

From this curve, we see that the ion fluxes of cations (Ca2+

and Mg2+

) are relatively

higher than those of anions (SO42-

). In addition, in the presence of calcium, magnisium ions

were preferentially transported. This is reflected by their higher fluxes.

It was found that all ion fluxes decreased over time. This is due to the depletion of the

solution from salts.

As shown in Figure 33, the effect of magnesium ions is clearly visible on the calcium

ions transfer. Indeed, the presence of magnesium ions greatly reduces the transfer of calcium

ions especially at the beginning of the desalination process.



Figure 33. Variation of the Ca2+

ionic flux versus time during the desalination process of solutions with

various compositions. (Flow rate 25 L h-1

; TDS 3 g L-1

; [Ca2+

] 0.01 M; I 1.96 A).

This indicates that there is competition between these two ions with same valence during

their transfer through ion exchange membranes. This competition is reduced at the end of the

process because the solution is depleted of ions at this level.

The competition between the two ionic species can also be deduced from the figure 34

showing the variation of the current efficiency for the transport of calcium ions as a function

of time for solutions containing different fractions of magnesium ions. Indeed, the increase in

this fraction causes the decrease of current efficiency. Thus, the current is no longer used only

for the transport of Ca2+

ions, but it is divided between the two species.

Figure 34. Variation of the current efficiency for the calcium ions transport as a function of time for

solutions containing different fractions of magnesium ions (Flow rate 25 L h-1

; TDS 3 g L-1

; [Ca2+

] 0.01

M; I 1.96 A).



Figure 35. Variation of specific power consumption as a function of time for solutions containing

different fractions of magnesium ions (Flow rate 25 L h-1

; TDS 3 g L-1

; [Ca2+

] 0.01 M; I 1.96 A).

Concerning energy consumption, the addition of Mg ions does not have a significant

effect on this parameter as shown in Figure 35. But, as the demineralization rate, it causes a

slight increase in the total duration of the desalination process.

The presence of divalent anions may also have an effect on the desalination process. To

perform the study we have prepared solutions with different amounts of sulfates. Calcium

concentration and total dissolved salts were respectively set at 0.01 M and 0.05 M. We added

sulfate ions at different fraction by dissolving Na2SO4 in these solutions.

Figure 36 illustrates the variation of demineralization rate as function of time during the

desalination of prepared solutions. From these curves we find that the presence of sulfate ions

affects the demineralization rate. Indeed, the addition of these anions decreases the process

efficiency. It causes also a slight increase in the total duration of the desalination process.

But the effect of sulfate ions is clearly visible on the calcium ionic transfer as is shown in

Figure 37. Indeed, the presence of these anions largely reduces the transfer of calcium ions

especially at the beginning of the desalination process. An increase in the fraction, especially

beyond seventh, sulfate ions reduces the transfer of calcium ions at very low fluxes. These

rates remain fairly constant over time. This result encouraged us to study the transfer of

calcium and sulfate ions analogous as shown in Figure 38.

From figure 38, we find that calcium ions transfer fluxes increases slightly over time

unlike that of sulfate ions.

We also note that these fluxes are relatively low (~10-8

mol cm-2

s-1

) when compared to

those in the absence of sulfate ions (~ 10-7

mol cm-2

s-1

).

This result can be interpreted by the fact that in the presence of bivalent anions, Ca ions

are retained in the dilute. Apparently the mobility of these cations is decreased. A mutual

interaction between the oppositely charged ions can take place.

As shown in figure 39, the effect of the addition of anions on the current efficiency for

the transport of Ca2+

is also clear.



This efficiency decreases more and more with increasing the molar fraction of the SO42-

ions in the solution. Thus the current is no longer used only for the transport of Ca2+

ions, but

it is divided between the two species.

Concerning energy consumption, the addition of sulfate ions does not have a significant

effect on this parameter as shown in Figure 40. But, as the demineralization rate, it causes a

slight increase in the total duration of the desalination process.

Figure 36. Variation of demineralization rate as a function of time for solutions containing different

fractions of sulfate ions (Flow rate 25 L h-1

; TDS 3 g L-1

; [Ca2+

] 0.01 M; I 1.96 A).

Figure 37. Variation of the Ca2+

ionic flux versus time during the desalination process of solutions with

various compositions. (Flow rate 25 L h-1

; TDS 3 g L-1

; [Ca2+

] 0.01 M; I 1.96 A).



Figure 38. Variation of the Ca2+ and SO42- ionic fluxes versus time during the desalination process.

(Flow rate 25 L h-1

, TDS 3 g L-1

, [Ca2+

] 0.01 M; [SO42-

] 0.002 M; I 1.96 A).

Figure 39. Variation of the current efficiency for the transport of calcium ions as a function of time for

solutions containing different fractions of sulfate ions (Flow rate 25 L h-1

, TDS 3 g L-1

, [Ca2+

] 0.01 M; I

1.96 A).

c. Desalination of the Real Water Sample

Brackish water sample with total dissolved salts TDS around 15 g L-1

was taken from

groundwater in the region of Borj-Cédria (Tunis City - Tunisia). Since an electrode

overheating can occur in the ED system for the treatment of solution with relatively high



salinity (TDS over 5 g L-1

), the sample was diluted by mixing it with raw water to reduce the

total salinity until 5 g L-1

. The physico-chemistry characteristics of obtained sample are given

in Table 3.

Figure 40. Variation of specific power consumption as a function of time for solutions containing

different fractions of sulfate ions (Flow rate 25 L h-1

, TDS 3 g L-1

, [Ca2+

] 0.01 M; I 1.96 A).

Table 3. Composition, pH and conductivity of brackish water Sample

Physico-chemistry parameter

Conductivity at 25 °C (µS cm-1

) 5 008

pH 7.2

F- (mg L

-1) 2.9

Cl- (mg L

-1) 2 674

HCO3- (mg L

-1) 202

NO3- (mg L

-1) 225

SO42-

(mg L-1

) 707

Na+ (mg L

-1) 1041

K+ (mg L

-1) 300

Ca2+

(mg L-1

) 235

Mg2+

(mg L-1

) 127.6

TDS (mg L-1

) 5 424

From this table, we can see that the water is rich in sulfate as well as chloride ions. It also

contains high levels of calcium and magnesium.

The present section is dealing with the desalination of the resulting solution by

electrodialysis in batch process mode. The flow rates were set at the outset to 25 L h-1

. An

electric current of 1.96 A is applied and maintained constant between the electrodes cell

terminals.



We followed the changes in the demineralization rate and ionic fluxes of predominant

species namely calcium, magnesium and sulfate in solution over time.

Figure 41 illustrates the variation in demineralization rate as function of time during the

desalination process.

Figure 41. Variation of demineralization rate as a function of time during the desalination of brackish

solution (I 1.96 A; Flow rate 25 L h-1).

From this graph we see that the demineralization rate increases over time. The desired

salinity is practically obtained after 85 min.

Figure 42. Variation of ionic flux as a function of time during the desalination of brackish solution (I

1.96 A; Flow rate 25 L h-1

).

We proposed to study the behavior of each species in these conditions. For this purpose,

we have traced the evolution of the ionic flux of each species over time in Figure 42.

From Figure 42, we note that the ionic fluxes of cations (Ca2+

and Mg2+

) are relatively in

the same order of scale (10-8

mol cm-2

s-1

) and even slightly lower than those of anions



(sulfates). This result is different from what was found in thedesalination of synthetic

solutions containing a mixture of these three ions. This can be explained by the fact that other

ions in solution, such as bicarbonates and nitrates, also influence the transport of the studied

species. Concerning the energy consumption, there is an increase in this parameter over time.

Energy consumption at the end of the experiment (about 35 W h L-1

) is equivalent to that

found in the case of desalination of a synthetic solution containing a mixture of calcium,

magnesium and sulfate ions.

Figure 43. Variation of specific power consumption as a function of time during the desalination of

brackish solution (I 1.96 A; Flow rate 25 L h-1

).

Table 4. Composition, pH and conductivity of brackish water sample,

treated solution and WHO recommended solution for drinking water

Physico-chemistry parameter Sample

Treated

solution Who recommendation

Conductivity at 25 °C (µS cm-1) 5 008 0.808 -

pH 7.2 5.1 6.5-8.5

F- (mg L-1) 2.9 0.22 1.5

Cl- (mg L-1) 2 674 189 250

HCO3- (mg L-1) 202 202 -

NO3- (mg L-1) 225 25.5 50

SO42- (mg L-1) 707 24.67 400

Na+ (mg L-1) 1041 49.92 250

K+ (mg L-1) 300 24.23 -

Ca2+ (mg L-1) 235 14.03 -

Mg2+ (mg L-1) 127.6 12.72 -

TDS (mg L-1) 5 424 350 500



The physico-chemical characteristics of the sample, the desalted solution and

recommendations of the World Health Organization (WHO) for drinking water are

summarized in the following table (WHO, 2008):

Comparing different solutions, we find that the resulting solution with a pH adjustment is

consistent with the recommendations of the World Health Organization for drinking water.

This proves the efficiency of electrodialysis process for the desalination of brackish water to

produce drinking water.

CONCLUSION

In this chapter it has been showed that the desalination of brackish water by conventional

electrodialysis is depending on several parameters.

The applied voltage, the feed flow rate and the ionic strength of the feed solution have a

significant effect on the process efficiency and mainly on the demineralization rate and power

consumption.

Working in single pass process mode, on which the solution passes only once in the

electrodialysis cell, the decrease of the flow rate and ionic strength induces better

performances. However an increase of applied voltage and taking into account to work under

limiting current density gives acceptable efficiency.

In this configuration, the maximum desalination rate does not exceed the 55% and a

second electrodialysis stage is required to enhance the process efficiency.

A study of another configuration was carried out. This configuration is the discontinuous

mode or total recirculation mode (batch process). In this mode, the solutions are recycled in

the cell until the desired concentration is reached.

In the first configuration, the applied voltage, the feed flow rate and the ionic strength of

the feed solution have also significant effect on the process efficiency. While working with

this mode, the desalination rates of various solutions reached quickly high values. Indeed,

they exceeded the 75% in a few minutes.

This study also revealed that the composition of the solution itself influences the

effectiveness of desalination process. In fact, the transfer of ions from a compartment to

another depends on the various ionic species present in solution.

The presence of the calcium ions largely decreased the transfer of the magnesium ions

and vice versa. This is interpreted by the fact that two ions of same valence enter in

competition during their transfers through exchanging membranes.

Moreover, the rates of transfer of the cations and more precisely calcium and magnesium

ions undergo reductions in the presence of anions and in particular sulfate ions. The desired

rates of desalination are also obtained for longer handling times. The consequence is

immediate on energy consumption. Indeed the consumption of energy is more important for

more complex solutions.

The process was applied for the treatment of real brackish water sample. With 30 W h L-1

as specific power consumption, the concentrations of different species in the obtained treated

water are below the recommended amounts by World Health Organization for drinking water.



REFERENCES

Alghoul, M. A., Poovanaesvaran, P., Sopian, K., & Sulaiman, M. Y. (2009). Review of

brackish water reverse osmosis (BWRO) system designs. Renewable and Sustainable

Energy Reviews, 13(9), 2661-2667.

Baker, R. W. (2004). Membrane technology and applications (2 ed.). England: John Wiley &

Sons, Ltd.

Balmann, Hélène Roux De, & Casademont, Ernest. (2006). Électrodialyse. Techniques de

l‟Ingénieur, J 2840(J 2840).

Ben Sik Ali, M., Ennigrou, D. J., & Hamrouni, B. (2013). Iron removal from brackish water

by electrodialysis. Environmental Technology (United Kingdom).

Ben Sik Ali, M., Mnif, A., Hamrouni, B., & Dhahbi, M. (2010). Electrodialytic desalination

of brackish water: effect of process parameters and water characteristics. Ionics, 1-9.

Casademont, C., Farias, M. A., Pourcelly, G., & Bazinet, L. (2008). Impact of electrodialytic

parameters on cation migration kinetics and fouling nature of ion-exchange membranes

during treatment of solutions with different magnesium/calcium ratios. Journal of

Membrane Science, 325(2), 570-579.

Côté, P., Weirun, Z., Gierlich, H. H., Natarajan, P., & Taniguchi, Y. (1996). State-of-the-art

techniques in reverse osmosis, nanofiltration and 1 electrodialysis in drinking-water

supply. Water Supply, 14(3-4), 289-319.

Danis, Patrick. (2003). Dessalement de l‟eau de mer Techniques de l‟Ingénieur, traité Génie

des procédés, J2700.

Dardel, François De. (1998). Échange d‟ions Principes de base. Techniques de l‟Ingénieur,

traité Génie des procédés, J 2 783.

El-Dessouky, Hisham T., & Ettouney, Hisham M. (2002). Fundamentals of Salt Water

Desalination. Amsterdam, The Netherlands: Elsevier Science B.V.

Ergun, E., Tor, A., Cengeloglu, Y., & Kocak, I. (2008). Electrodialytic removal of fluoride

from water: Effects of process parameters and accompanying anions. Separation and

Purification Technology, 64(2), 147-153.

Greenlee, Lauren F., Lawler, Desmond F., Freeman, Benny D., Marrot, Benoit, & Moulin,

Philippe. (2009). Reverse osmosis desalination: Water sources, technology, and today's

challenges. Water Research, 43(9), 2317-2348.

Hannachi, C., Hamrouni, B., & Dhahbi, M. (2009). Ion exchange equilibrium between cation

exchange membranes and aqueous solutions of K+/Na+, K+/Ca2+, and Na+/Ca2+.

Ionics, 15(4), 445-451.

Hannachi, Ch, Bouguecha, S., Hamrouni, B., & Dhahbi, M. (2008). Ion exchange equilibrium

between ion exchange membrane and electrolyte solutions. Desalination, 221(1-3), 448-

454.

Hoffman, E. J. (2003). Membrane Separations Technology: Single-Stage, Multistage, and

Differential Permeation: Elsevier Science & Technology Books

Inglezakis, V.J., & Poulopoulos, S.G. (2006). Adsorption, Ion Exchange and Catalysis:

Design of Operations and Environmental Applications. Amesterdam, The Netherlands:

Elsevier Science.



Kabay, Nalan, Arar, Özgür, Samatya, Saba, Yüksel, Ümran, & Yüksel, Mithat. (2008).

Separation of fluoride from aqueous solution by electrodialysis: Effect of process

parameters and other ionic species. Journal of Hazardous Materials, 153(1-2), 107-113.

Kontturi, Kyösti, Murtomäki, Lasse, & Manzanares, José A. (2008). Ionic Transport

Processes In Electrochemistry and Membrane Science. New York: Oxford University

Press Inc.

Krol, John Jacco. (1997). Monoplolar and bipolar ion exchange membranes: Mass transport

limitations. (Thesis), Université Twente.

Lee, Hong-Joo, Sarfert, F., Strathmann, H., & Moon, Seung-Hyeon. (2002). Designing of an

electrodialysis desalination plant. Desalination, 142(3), 267-286.

Lee, Hong-Joo, Strathmann, Heiner, & Moon, Seung-Hyeon. (2006). Determination of the

limiting current density in electrodialysis desalination as an empirical function of linear

velocity. Desalination, 190(1-3), 43-50.

Matsuura, Takeshi. (2001). Progress in membrane science and technology for seawater

desalination -- a review. Desalination, 134(1-3), 47-54.

Moçotéguy, Philippe. (1999). Modélisation du fonctionnement d'une électrodialyseur à

membrane à l'aide du logiciel Flux-Expert. (Thesis), École Centrale de Paris Paris.

Nagarale, R. K., Gohil, G. S., & Shahi, V. K. (2006). Recent developments on ion-exchange

membranes and electro-membrane processes. Advances in Colloid and Interface Science,

119(2-3), 97-130.

Noble, Richard D., & Stern, S. Alexander. (1995). Membrane separations technologies

Principles and Applications Amsterdam: Elsevier Science B.V.

Sadrzadeh, M., & Mohammadi, T. (2008). Sea water desalination using electrodialysis.

Desalination, 221(1-3), 440-447.

Sadrzadeh, M., Razmi, A., & Mohammadi, T. (2007). Separation of different ions from

wastewater at various operating conditions using electrodialysis. Separation and

Purification Technology, 54(2), 147-156.

SenGupta, Arup K. (2007). Ion exchange and solvant extraction A Series of Advances (Vol.

18). Boca Raton, Florida: CRC Press.

Shaposhnik, V. A., Zubets, N. N., Mill, B. E., & Strigina, I. P. (2001). Demineralization of

water by electrodialysis with ion-exchange membranes, grains and nets. Desalination,

133(3), 211-214.

Strathmann, H. (2004). Ion-Exchange Membrane Separation Processes (Vol. 9). Amsterdam,

The Netherlands Elsevier.

Strathmann, Heiner. (2004). Ion-exchange membranes separation processes (Vol. 9).

Amsterdam: Elsevier B.V.

Tanaka, Y. (2002). Current density distribution, limiting current density and saturation

current density in an ion-exchange membrane electrodialyzer. Journal of Membrane

Science, 210(1), 65-75.

Tanaka, Y. (2005). Limiting current density of an ion-exchange membrane and of an

electrodialyzer. Journal of Membrane Science, 266(1-2), 6-17.

Tanaka, Yoshinobu. (2007). Ion Exchange Membranes: Fundamentals and Applications (Vol.

12). Oxford, UK: Elsevier.

Wang, Q., Ying, T., Jiang, T., Yang, D., & Jahangir, M. M. (2009). Demineralization of

soybean oligosaccharides extract from sweet slurry by conventional electrodialysis.

Journal of Food Engineering, 95(3), 410-415.



WHO. (2008). Guidelines for drinking-water quality. In W. H. Organization (Ed.), (3 ed.,

Vol. 1, pp. 375-492).

Wilhelm, Friedrich Georg. (2001). Bipolar Membrane Electrodialysis: Membrane

development and transport caracteristics. (Ph.D. thesis), University of Twente,

Enschede, the Netherlands. Retrieved from www.tup.utwente.nl/uk/catalogue/

technical/electrodialysis

Zagorodni, Andrei A. (2007). Ion Exchange Materials Properties and Applications. Oxford,

UK: Elsevier BV.





Chapter 11

APPROPRIATE TIME SCALE FOR AGGREGATING

CLIMATIC DATA TO PREDICT FLOWERING

AND BOLL SETTING BEHAVIOR OF COTTON

Zakaria M. Sawan* Cotton Research Institute, Agricultural Research Center,

Ministry of Agriculture and Land Reclamation, Giza, Egypt

ABSTRACT

This study covers the predicted effects of climatic factors during convenient intervals

(in days) on cotton flower and boll production compared with daily observations.


maximum air temperature, are the important climatic factors that significantly affect

flower and boll production. The least important variables were found to be surface soil

temperature at 0600 h and minimum temperature. The five-day interval was found to be

more adequately and sensibly related to yield parameters. There was a negative

correlation between flower and boll production and either evaporation or sunshine

duration, while that correlation with minimum relative humidity was positive. Higher

minimum relative humidity, short period of sunshine duration, and low temperatures

enhanced flower and boll formation.

Keywords: Cotton flower and boll production, boll retention, evaporation, relative humidity,

sunshine duration, temperature

ABBREVIATIONS

ET Evapotranspiration

ETmax Maximum Evapotranspiration

* Email: [email protected]


Zakaria M. Sawan 332

PGR plant growth regulator

TKW thermal kinetic window

1. INTRODUCTION

Climate affects crop growth interactively, sometimes resulting in unexpected responses to

prevailing conditions. Many factors, such as length of the growing season, climate (including

solar radiation, temperature, light, wind, rainfall, and dew), cultivar, availability of nutrients

and soil moisture, pests and cultural practices affect cotton growth (El-Zik, 1980). The

balance between vegetative and reproductive development can be influenced by soil fertility,

soil moisture, cloudy weather, spacing and perhaps other factors such as temperature and

relative humidity (Guinn, 1982). Weather, soil, cultivars, and cultural practices affect crop

growth interactively, sometimes resulting in plants responding in unexpected ways to their

conditions (Hodges et al. 1993).

Water is a primary factor controlling plant growth. Xiao et al. (2000) stated that, when

water was applied at 0.85, 0.70, 0.55 or 0.40 ET (evapotranspiration) to cotton plants grown

in pots, there was a close relationship between plant development and water supply. The fruit-

bearing branches, square and boll numbers and boll size were increased with increased water

supply. Barbour and Farquhar (2000) reported on greenhouse pot trials where cotton cv. CS50

plants were grown at 43 or 76% relative humidity (RH) and sprayed daily with abscisic acid

(ABA) or distilled water. Plants grown at lower RH had higher transpiration rates, lower leaf

temperatures and lower stomatal conductance. Plant biomass was also reduced at the lower

RH. Within each RH environment, increasing ABA concentration generally reduced stomatal

conductance, evaporation rates, superficial leaf density and plant biomass, and increased leaf

temperature and specific leaf area.

Temperature is also a primary factor controlling rates of plant growth and development.

Burke et al. (1988) has defined the optimum temperature range for biochemical and metabolic

activities of plants as the thermal kinetic window (TKW). Plant temperatures above or below

the TKW result in stress that limits growth and yield. The TKW for cotton growth is 23.5 to

32°C, with an optimum temperature of 28°C. Biomass production is directly related to the

amount of time that foliage temperature is within the TKW.

Reddy et al. (1995) in growth chamber experiments found that Pima cotton cv. S-6

produced lower total biomass at 35.5°C than at 26.9°C and no bolls were produced at the

higher temperature of 40°C. Schrader et al. (2004) stated that high temperatures that plants

are likely to experience inhibit photosynthesis. Zhou et al. (2000) indicated that light duration

is the key meteorological factor influencing the wheat-cotton cropping pattern and position of

the bolls, while temperature had an important function on upper (node 7 to 9) and top (node

10) bolls, especially for double cropping patterns with early maturing varieties.

In Texas, Guo et al. (1994) found that plant growth and yield of the cotton cv. DPL-50

(Upland cotton) were less in a humid area than in an arid area with low humidity. Under arid

conditions, high vapor pressure deficit resulted in a high transpiration rates, low leaf water

potential and lower leaf temperatures. Gipson and Joham (1968) mentioned that cool

temperatures (< 20°C) at night slowed boll development. Fisher (1975) found that high

temperatures can cause male sterility in cotton flowers, and could have caused increased boll


Appropriate Time Scale for Aggregating Climatic Data to Predict Flowering … 333

shedding in the late fruiting season. Zhao (1981) indicated that temperature was the main

climatic factor affecting cotton production and 20-30°C was the optimum temperature for

cotton growth.

Hodges et al. (1993) found that the optimum temperature for cotton stem and leaf growth,

seedling development, and fruiting was almost 30°C, with fruit retention decreasing rapidly

as the time of exposure to 40°C increased.

Reddy et al. (1998) found that when Upland cotton (G. hirsutum) cv. DPL-51 was grown

in naturally lit plant growth chambers at 30/22°C day/night temperatures from sowing until

flower bud production, and at 20/12, 25/17, 30/22, 35/27 and 40/32°C for 42 days after flower

bud production, fruit retention was severely curtailed at the two higher temperatures

compared with 30/22°C. Species/cultivars that retain fruits at high temperatures would be

more productive both in the present-day cotton production environments and even more in

future warmer world.

This study aimed at predicting effects of climatic factors during different convenient

intervals (in days) on cotton flower and boll production compared with daily observations.

The study presents a rich effort focused on evaluating the efficacy of regression equations

between cotton crop data and climatic data grouped at different time intervals, to determine

the appropriate time scale for aggregating climate data to be used for predicting flower and

boll production in cotton (Sawan et al., 2006). This could result in formulating advanced

predictions as for the effect of certain climatic conditions on production of Egyptian cotton.

Minimizing the deleterious effects of the factors through utilizing proper cultural practices

will lead to improved cotton yield (Sawan et al., 2010).

2. DATA AND METHODS

Two uniform field trials were conducted at the experimental farm of the Agricultural

Research Center, Ministry of Agriculture, Giza, Egypt (30oN, 31

o: 28‟E at an altitude of 19

m), using the cotton cultivar Giza 75 (Gossypium barbadense L.) in 2 successive seasons (I

and II). The soil texture was a clay loam, with an alluvial substratum (pH = 8.07, 42.13%

clay, 27.35% silt, 22.54% fine sand, 3.22% coarse sand, 2.94% calcium carbonate and 1.70%

organic matter) (Sawan et al., 2010).

In Egypt, there are no rain-fed areas for cultivating cotton. Water for the field trials was

applied using surface irrigation. Total water consumed during each of two growing seasons

supplied by surface irrigation was about 6,000-m³ h-1

. The criteria used to determine amount

of water applied to the crop depended on soil water status. Irrigation was applied when soil

water content reached about 35% of field capacity (0-60 cm). In season I, the field was

irrigated on 15 March (at planting), 8 April (first irrigation), 29 April, 17 May, 31 May, 14

June, 1 July, 16 July, and 12 August. In season II, the field was irrigated on 23 March

(planting date), 20 April (first irrigation), 8 May, 22 May, 1 June, 18 June, 3 July, 20 July, 7

August and 28 August. Techniques normally used for growing cotton in Egypt were followed.

Each experimental plot contained 13 to 15 ridges to facilitate proper surface irrigation. Ridge

width was 60 cm and length was 4 m. Seeds were sown on 15 and 23 March in seasons I and

II, respectively, in hills 20 cm apart on one side of the ridge. Seedlings were thinned to 2

plants per hill 6 weeks after planting, resulting in a plant density of about 166,000 plants ha-1

.



Phosphorus fertilizer was applied at a rate of 54 kg P2O5 ha-1

as calcium super phosphate

during land preparation. Potassium fertilizer was applied at a rate of 57 kg K2O ha-1

as

potassium sulfate before the first irrigation (as a concentrated band close to the seed ridge).

Nitrogen fertilizer was applied at a rate of 144 kg N ha-1

as ammonium nitrate in two equal

doses: the first was applied after thinning just before the second irrigation and the second was

applied before the third irrigation. Rates of phosphorus, potassium, and nitrogen fertilizer

were the same in both seasons. These amounts were determined based on the use of soil tests

(Sawan et al., 2010).

After thinning, 261 and 358 plants were randomly selected (precaution of border effect

was taken into consideration by discarding the cotton plants in the first and last two hills of

each ridge) from 9 and 11 inner ridges of the plot in seasons I, and II respectively. Pest

control management was carried out on an-as-needed basis, according to the local practices

performed at the experimental station (Sawan et al., 2010).

Flowers on all selected plants were tagged in order to count and record the number of

open flowers, and set bolls on a daily basis. The flowering season commenced on the date of

the first flower appearance and continued until the end of flowering season (31 August). The

period of whole September (30 days) until the 20th of October (harvest date) allowed a

minimum of 50 days to develop mature bolls. In season I, the flowering period extended from

17 June to 31 August, whereas in season II, the flowering period was from 21 June to 31

August. Flowers produced after 31 August were not expected to form sound harvestable bolls,

and therefore were not taken into account (Sawan et al., 2010).

For statistical analysis, the following data of the dependent variables were collected:

number of tagged flowers separately counted each day on all selected plants (Y1), number of

retained bolls obtained from the total daily tagged flowers on all selected plants at harvest

(Y2), and (Y3) percentage of boll retention ([number of retained bolls obtained from the total

number of daily tagged flowers in all selected plants at harvest]/[daily number of tagged

flowers on each day in all selected plants] x 100).

As a rule, observations were recorded when the number of flowers on a given day was at

least 5 flowers found in a population of 100 plants and this continued for at least five

consecutive days. This rule omitted eight observations in the first season and ten observations

in the second season. The number of observations (n) was 68 (23 June through 29 August)

and 62 (29 June through 29 August) for the two seasons, respectively (Sawan et al., 2010).

The climatic factors (independent variables) considered were daily data of: maximum air

temperature (°C, X1); minimum air temperature (°C, X2); maximum-minimum air

temperature (diurnal temperature range) (°C, X3); evaporation (expressed as Piche

evaporation) (mm day-1

, X4); surface soil temperature, grass temperature or green cover

temperature at 0600 h (°C, X5) and 1800 h (°C, X6); sunshine duration (h day-1

, X7);

maximum relative humidity (maxRH) (%, X8), minimum relative humidity (minRH) (%, X9)

and wind speed (m s-1

, X10) in season II only. The source of the climatic data was the

Agricultural Meteorological Station of the Agricultural Research Station, Agricultural

Research Center, Giza, Egypt. No rainfall occurred during the two growing seasons. Range

and mean values of the climatic parameters recorded during the production stage for both

seasons and overall data are listed in Table 1. Daily number of flowers and number of bolls

per plant which survived till maturity (dependent variables) during the production stage in the

two seasons are graphically illustrated in Figures 1 and 2 (Sawan et al., 2006).



Table 1. Range and mean values of the independent variables for the two seasons

and overall data

Climatic factor's First season* Second season**

Over all data

(Two seasons)

Range Mean Range Mean Range Mean

Max Temp (°C), (X1) Min

Temp (°C), (X2) Max-Min

Temp (°C), (X3) ♦

Evap (mm d-1), (X4)

0600 h Temp (°C), (X5)

1800 h Temp (°C), (X6)

Sunshine (h d-1), (X7) Max

RH (%), (X8) Min RH (%),

(X9) Wind speed (m s-1),

(X10)

31.0-44.0

18.6-24.5

9.4-20.9

7.6-15.2

14.0-21.5

19.6-27.0

10.3-12.9

62-96

11-45

ND

34.3

21.9

12.4

10.0

17.8

24.0

11.7

85.4

30.8

ND

30.6-38.8

18.4-23.9

8.5-17.6

4.1-9.8

13.3-22.4

20.6-27.4

9.7-13.0

51-84

23-52

2.2-7.8

34.1

21.8

12.2

6.0

18.0

24.2

11.9

73.2

39.8

4.6

30.6-44.0

18.4-24.5

8.5-20.9

4.1-15.2

13.3-22.4

19.6-27.4

9.7-13.0

51-96

11-52

ND

34.2

21.8

12.3

8.0

17.9

24.1

11.8

79.6

35.1

ND

(Sawan et al. 2006). ♦Diurnal temperature range. ND not determined.

*Flower and boll stage (68 days, from 23 June through 29 August).

**Flower and boll stage (62 days,

from 29 June through 29 August).

Sawan et al., 2006.

Figure 1. Daily number of flowers and bolls during the production stage (68 days) in the first season (I)

for the Egyptian cotton cultivar Giza 75 (Gossypium barbadense L.) grown in uniform field trial at the

experimental farm of the Agricultural Research Centre, Giza (30°N, 31°:28'E), Egypt. The soil texture

was a clay loam, with an alluvial substratum, (pH = 8.07). Total water consumptive use during the

growing season supplied by surface irrigation was about 6000 m3ha

-1. No rainfall occurred during the

growing season. The sampling size was 261 plants.



Sawan et al., 2006.

Figure 2 Daily number of flowers and bolls during the production stage (62 days) in the second season

(II) for the Egyptian cotton cultivar Giza 75 (Gossypium barbadense L.) grown in uniform field trial at

the experimental farm of the Agricultural Research Centre, Giza (30°N, 31°:28'E), Egypt. The soil

texture was a clay loam, with an alluvial substratum, (pH = 8.07). Total water consumptive use during

the growing season supplied by surface irrigation was about 6000 m3ha

-1. No rainfall occurred during the

growing season. The sampling size was 358 plants.

2.1. Statistical Analysis

Statistical analysis was conducted using the procedures outlined in the general linear

model (GLM, SAS Institute, Inc., 1985). Data of dependent and independent variables,

collected for each day of the production stage (60 days in each season), were summed up into

intervals of 2, 3, 4, 5, 6 or 10 days. Data from these intervals were used to compute

relationships between the dependent variables (flower and boll setting and boll retention) and

the independent variables (climatic factors) in the form of simple correlation coefficients for

each season. Comparisons between the values of “r” were done to determine the best interval

of days for determining effective relationships. The α-level for significance was P < 0.15. The

climatic factors attaining a probability level of significance not exceeding 0.15 were deemed

important (affecting the dependent variables), selected and combined with dependent variable

in multiple regression analysis to obtain a convenient predictive equation (Cady and Allen,

1972). Multiple linear regression equations (using stepwise method) comprising selected

predictive variables were computed for the determined interval and coefficients of multiple

determinations (R²) were calculated to measure the efficiency of the regression models in

explaining the variation in data. Correlation and regression analyses were computed

according to Draper and Smith (1966) (Sawan et al., 2006).



3. RESULTS AND DISCUSSION

3.1. Response of Flower and Boll Development to Climatic Factors on the

Anthesis Day

Daily number of flowers and number of bolls per plant which survived to maturity

(dependent variables) during the production stage of the two seasons (68 days and 62 days in

the first and the second seasons, respectively) are graphically illustrated in Figures 1 and 2.

The flower- and boll-curves reached their peaks during the middle two weeks of August, and

then descended steadily till the end of the season. Specific differences in the shape of these

curves in the two seasons may be due to the growth-reactions of environment, where climatic

factors (Table 1) (Sawan et al., 2006) represent an important part of the environmental effects

(Miller et al., 1996).

3.2. Correlation Estimates

Significant simple correlation coefficients were estimated between the production

variables and studied climatic factors for different intervals of days (combined data of the 2

seasons) (Table 2) (Sawan et al., 2006).

Evaporation was the most important climatic factor affecting flower and boll production

in Egyptian cotton. The negative correlation means that high evaporation ratio significantly

reduced flower and boll production. High evaporation rates could result in water stress that

would slow growth and increase shedding rate of flowers and bolls (Sawan et al., 2006). Kaur

and Singh (1992) found in cotton that flower number was decreased by water stress,

particularly when existing at flowering stage. Seed cotton yield was decreased by about 50%

when water stress was present at flowering stage, slightly decreased by stress at boll

formation stage, and not significantly affected by stress in the vegetative stage (6-7 weeks

after sowing).

The second most important climatic factor was minimum humidity, which had a high

positive correlation with flower and boll production, and retention ratio. The positive

correlation means that increased humidity would bring about better boll production (Sawan et

al., 2006).

The third most important climatic factor in our study was sunshine duration, which

showed a significant negative relationship with flower and boll production only (Sawan et al.,

2006). The negative relationship between sunshine duration and cotton production may be

due to the fact that the species of the genus Gossypium are known to be short day plants

(Hearn and Constable, 1984), so, an increase of sunshine duration above that sufficient to

attain good plant growth will decrease flower and boll production. Bhatt (1977) found that

exposure to daylight over 14 hours and high day temperature, individually or in combination,

delayed flowering of the Upland cotton cv. J34. Although average sunshine duration in our

study was only 11.7 h, yet it could reach 13 h, which, in combination with high maximum

temperatures (up to 38.8°C), may have adversely affected reproductive growth.


Table 2. Significant simple correlation coefficient values between the production variables and the studied climatic factors for the daily

and different intervals of days combined over both seasons

Daily and intervals

of days

Production

variables

Climatic factorsz

Air temp (°C) Evap

(mm d-1

)

(X4)

Surface soil

temp (°C)

Sunshine

duration

(h d-1

)

(X7)

Relative humidity

(%)

Max

(X1)

Min

(X2)

Max-Min

(X3)

0600 h

(X5)

1800 h

(X6)

Max

(X8)

Min

(X9)

Daily (n = 120)

2 Days (n# = 60)

3 Days (n# = 40)

4 Days (n# = 30)

5 Days (n# = 24)

Flower

Boll

Boll ret. rat.

Flower

Boll

Boll ret. rat.

Flower

Boll

Boll ret. rat.

Flower

Boll

Boll ret. rat.

Flower

Boll

Boll ret. rat.

-0.15++

NS

NS

-0.31++

-0.29++

NS

-0.34*

-0.32*

NS

-0.31++

-0.31++

NS

-0.35++

-0.33+

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

-0.26**

-0.25**

NS

-0.32*

-0.30++

NS

-0.34*

-0.32*

NS

-0.35++

-0.33++

NS

-0.37++

-0.35++

NS

-0.33**

-0.43**

-0.56**

-0.36**

-0.46**

-0.61**

-0.33*

-0.48**

-0.63**

-0.33++

-0.48**

-0.64**

-0.39++

-0.49*

-0.66**

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

-0.20*

-0.19++

NS

-0.24+

-0.21+

NS

-0.28++

-0.24+

NS

-0.28+

-0.23+

NS

-0.39++

-0.35++

NS

-0.23*

-0.18++

NS

-0.36**

-0.31*

NS

-0.39*

-0.36*

NS

-0.39*

-0.38*

NS

-0.52**

-0.44*

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

0.30**

0.36**

0.34**

0.37**

0.44**

0.40**

0.34*

0.45**

0.40*

0.34++

0.45*

0.42*

0.41*

0.47**

0.43*


Daily and intervals

of days

Production

variables

Climatic factorsz

Air temp (°C) Evap

(mm d-1

)

(X4)

Surface soil

temp (°C)

Sunshine

duration

(h d-1

)

(X7)

Relative humidity

(%)

Max

(X1)

Min

(X2)

Max-Min

(X3)

0600 h

(X5)

1800 h

(X6)

Max

(X8)

Min

(X9)

6 Days (n# = 20)

10 Days (n# = 12)

Flower

Boll

Boll ret. rat.

Flower

Boll

Boll ret. rat.

-0.37++

-0.37++

NS

NS

NS

NS

NS

NS

NS

NS

NS

NS

-0.41++

-0.40++

NS

-0.45++

-0.43++

NS

-0.38++

-0.49*

-0.69**

-0.40+

-0.51++

-0.74**

NS

NS

NS

NS

NS

NS

NS

NS

NS

-0.55*

-0.53++

NS

-0.54**

-0.46*

NS

-0.65*

-0.57*

NS

NS

NS

NS

NS

NS

NS

0.42*

0.49*

0.45*

0.43++

0.51++

0.55*

(Sawan et al. 2006). z Wind speed did not show significant effect upon the studied production variables, so is not reported.

** Significant at 1 % probability level,

* Significant at 5 % probability level.

++ Significant at 10 % probability level,

+ Significant at 15 % probability level.

NS Means simple correlation coefficient is not significant at the 15% probability level. #n = Number of data pairs used in calculation.



Maximum air temperature, temperature magnitude and surface soil temperature at 1800 h

show significant negative relationships with flower and boll production only. Meanwhile, the

least important factors were surface soil temperature at 0600 h and minimum air temperature

(Sawan et al., 2006).

Our results indicate that evaporation was the most effective climatic factor affecting

cotton boll production. As the sign of the relationship was negative, this means that an

increase in evaporation caused a significant reduction in boll number (Sawan et al., 2006).

Thus, applying specific treatments, such as an additional irrigation or the use of plant growth

regulators (PGR) that would decrease the deleterious effect of evaporation after boll

formation, could contribute to an increase in cotton boll production and retention, and

consequently an increase in cotton yield. In this connection, Meek et al. (1999) in a field

experiment in Arkansas found that application of 3 or 6 kg glycine betaine (PGR) ha-1

to

cotton plants under mild water stress increased yield.

Comparing results for the different intervals of days with those from daily observation

(Table 2) (Sawan et al., 2006), the 5-day interval appeared to be the most suitable interval,

which actually revealed a more solid and more obvious relationships between climatic factors

and production characters. This was in fact indicated by the higher R2

values obtained when

using the 5-day intervals. The 5-day interval may be the most suitable interval for diminishing

the daily fluctuations between the factors under study to clear these relations comparing with

the other intervals. However, it seems that this conception is true provided that the

fluctuations in climatic conditions are limited or minimal. Therefore, it would be the most

efficient interval used to help circumvent the unfavorable effect of climatic factors. This

finding gives researchers and producers a chance to deal with condensed rather than daily

weather data (Sawan et al., 2006).

3. REGRESSION MODELS

Multiple linear regression equations were estimated using the stepwise multiple

regression technique to express the relation between cotton production variables [number of

flowers (Y1); bolls per plant (Y2); and boll retention ratio (Y3)] and the studied climatic

factors (Table 3) (Sawan et al., 2006).

Evaporation and surface soil temperature at 1800 h, sunshine duration and minimum

humidity accounted for a highly significant amount of variation (P < 0.05) in cotton

production variables, with the equation obtained for the 5-day interval showing a high degree

of certainty. The R² values for the 5-day interval were higher than those obtained from daily

data for each of the cotton production variables. Also, the 5-day interval gave more efficient

and stable estimates than the other studied intervals (data not shown) (Sawan et al., 2006).

The R² values for these equations clearly indicate the importance of such equations since

the climatic factors involved explained about 59 to 62% of the variation found in the

dependent variables (Sawan et al., 2006).



Table 3. The equations obtained for each of the studied cotton production variables for

the five-day intervals and daily intervals combined over both seasons

Equationz R² Significance

Five-day intervals

Y1 = 23.78 – 0.5362X4 – 0.1429X6 – 0.1654X7 + 0.0613X9 0.6237 **

Y2 = 15.89 – 0.4762X4 – 0.1583X6 – 0.1141X7 + 0.0634X9 0.5945 **

Y3 = 72.65 – 0.0833X4 – 0.1647X6 + 0.2278X9 0.6126 **

Daily intervals

Y1 = 19.78 – 0.181X3 – 0.069X4 – 0.164X6 – 0.182X7 + 0.010X9 0.4117 **

Y2 = 14.96 – 0.173X3 – 0.075X4 – 0.176X6 – 0.129X7 + 0.098X9 0.4461 **

Y3 = 52.36 – 3.601X4 – 0.2352X7 + 4.511X9 0.3587 **

Sawan et al., 2006. zWhere Y1 = number of flowers per plant, Y2 = number of bolls per plant, Y3 = boll retention ratio, X3 =

maximum – minimum temperature °C, X4 = evaporation mm day-1

, X6 = surface soil temperature

°C at 1800 h., X7 = sunshine duration h day-1

, and X9 = minimum relative humidity %.

During the production stage, an accurate weather forecast for the next 10 days would

provide an opportunity to avoid any adverse effect for weather factors on cotton production

through applying appropriate cultural practices such as adequate irrigation regime or

utilization of plant growth regulators. This proposal would be true if the fluctuations in

weather conditions were not extreme. Our recommendation would be the accumulation 5-day

climatic data, and use this information to select the adequate cultural practices (such as an

additional irrigation or utilization of plant growth regulators) that would help circumvent the

unfavorable effects of climatic factors. In case of sharp fluctuations in climatic factors, data

could be collected daily, and when stability of climatic conditions is restored, the 5-day

accumulation of weather data could be used again (Sawan et al., 2006).

Relative humidity showed the highest contribution to the variation in both flower and boll

production. This finding can be explained in the light of results found by Ward and Bunce

(1986) in sunflower (Helianthus annuus). They stated that decreases of relative humidity on

both leaf surfaces reduced photosynthetic rate of the whole leaf for plants grown under a

moderate temperature and medium light level.

Reddy et al. (1993) found that cotton (Gossypium hirsutum) fruit retention decreased

rapidly as the time of exposure to 40°C increased. Warner and Burke (1993) indicated that the

cool-night inhibition of cotton (Gossypium hirsutum) growth is correlated with biochemical

limitation on starch mobilization in source leaves, which result in a secondary inhibition of

photosynthesis, even under optimal temperature during the day. Gutiérrez and López (2003)

studied the effects of heat on the yield of cotton in Andalucia, Spain, during 1991-98, and

found that high temperatures were implicated in the reduction of unit production. There was a

significant negative relationship between average production and number of days with

temperatures greater than 40°C and the number of days with minimum temperatures greater

than 20°C. Wise et al. (2004) indicated that restrictions to photosynthesis could limit plant

growth at high temperature in a variety of ways. In addition to increasing photorespiration,

high temperatures (35-42°C) can cause direct injury to the photosynthetic apparatus. Both

carbon metabolism and thylakoid reactions have been suggested as the primary site of injury

at these temperatures.



Regression models obtained explained a sensible proportion of the variation in flower and

boll production, as indicated by their R2, which ranged between 0.53-0.72 (Sawan et al.,

2010). These results agree with Miller et al. (1996) in their regression study of the relation of

yield with rainfall and temperature. They suggested that the other R2 0.50 of variation was

related to management practices, which coincide with the findings of this study. Thus, an

accurate climatic forecast for the effect of the 5-7 day period during flowering may provide

an opportunity to avoid possible adverse effects of unusual climatic conditions before

flowering or after boll formation by utilizing additional treatments and/or adopting proper

precautions to avoid flower and boll reduction.

Temperature conditions during the reproduction growth stage of cotton in Egypt do not

appear to limit this growth even though they are above the optimum for cotton growth (Sawan

et al., 2010). This is contradictory to the finding of Holaday et al. (1997). A possible reason

for that contradiction is that the effects of soil moisture status and relative humidity were not

taken into consideration in the research studies conducted by other researchers in other

countries. Since temperature and evaporation are closely related to each other, the higher

evaporation rate could possibly mask the effect of temperature. Sunshine duration and

minimum relative humidity appeared to have secondary effects, yet they are in fact important

factors. The importance of sunshine duration has been eluded by Moseley et al. (1994) and

Oosterhuis (1997).

CONCLUSION


maximum temperature, were the most significant climatic factors affecting flower and boll

production of Egyptian cotton. The negative correlation between each of evaporation and

sunshine duration with flower and boll formation along with the positive correlation between

minimum relative humidity value and flower and boll production, indicate that low

evaporation rate, short period of sunshine duration and high value of minimum humidity

would enhance flower and boll formation. Temperature appeared to be less important in the

reproduction growth stage of cotton in Egypt than evaporation (water stress), sunshine

duration and minimum humidity. These findings concur with those of other researchers

except for the importance of temperature. A possible reason for that contradiction is that the

effects of evaporation rate and relative humidity were not taken into consideration in the

research studies conducted by other researchers in other countries. The matter of fact is that

temperature and evaporation are closely related to each other to such an extent that the higher

evaporation rate could possibly mask the effect of temperature. Water stress is in fact the

main player and other authors have suggested means for overcoming its adverse effect which

could be utilized in the Egyptian cotton. It must be kept in mind that although the reliable

prediction of the effects of the aforementioned climatic factors could lead to higher yields of

cotton, yet only 50% of the variation in yield could be statistically explained by these factors

and hence consideration should also be given to the management practices presently in use.

The 5-day interval was found to give adequate and sensible relationships between climatic

factors and cotton production growth under Egyptian conditions when compared with other

intervals and daily observations (Sawan et al., 2006). It may be concluded that the 5-day



accumulation of climatic data during the production stage, in the absence of sharp fluctuations

in these factors, could be satisfactorily used to forecast adverse effects on cotton production

and the application of appropriate production practices circumvent possible production

shortage.

Finally, the early prediction of possible adverse effects of climatic factors might modify

their effect on production of Egyptian cotton. Minimizing deleterious effects through the

application of proper management practices, such as, adequate irrigation regime, and

utilization of specific plant growth regulators could limit the negative effects of some climatic

factors (Sawan et al., 2010).

REFERENCES

Barbour MM, Farquhar GD (2000) Relative humidity- and ABA-induced variation in carbon

and oxygen isotope ratios of cotton leaves. Plant, Cell and Environment 23: 473-485.

Bhatt JG (1977) Growth and flowering of cotton (Gossypium hirsutum L.) as affected by

daylength and temperature. Journal of Agricultural Science 89: 583-588.

Burke JJ, Mahan JR, Hatfield JL (1988) Crop specific thermal kinetic windows in relation to

wheat and cotton biomass production. Agron J 80: 553-556.

Cady FB, Allen DM (1972) Combining experiments to predict future yield data. Agron J 64:

211-214.

Draper NR, Smith H (1966) Applied Regression Analysis. John Wiley & Sons Ltd., New

York, NY. 407 pp.

El-Zik KM (1980) The cotton plant - its growth and development. Western Cotton Prod.

Conf. Summary Proc., Fresno, CA, p. 18-21.

Fisher WD (1975) Heat induced sterility in Upland cotton. Proc 27th

Cotton Improvement

Conf. 85.

Gipson LR, Joham HE (1968) Influence of night temperature on growth and development of

cotton (Gossypium hirsutum L.): I. Fruiting and boll development. Agron J 60: 292-295.

Guinn G (1982) Causes of square and boll shedding in cotton. USDA Tech. Bull. 1672.

USDA, Washington, DC.

Guo Y, Landivar JA, Hanggeler JC, Moore J (1994) Response of cotton leaf water potential

and transpiration to vapor pressure deficit and salinity under arid and humid climate

conditions. In Proceedings Beltwide Cotton Conferences, January 5-8, San Diego,

California, USA. Memphis, USA, National Cotton Council, 1301-1308.

Gutiérrez Mas JC, López M (2003) Heat, limitation of yields of cotton in Andalucia.

Agricultura, Revista Agropecuaria 72: 690-692.

Hearn AB, Constable GA (1984) The Physiology of Tropical Food Crops. Chapter 14:

Cotton. P. 495-527 (Edited by Goldsworth, P.R.; Fisher, N.M.), John Wiley & Sons Ltd.,

NY. 664 pp.

Hodges HF, Reddy KR, McKinion JM, Reddy VR (1993) Temperature effects on cotton.

Bulletin Mississippi Agricultural and Forestry Experiment Station No. 990: 15.

Holaday AS, Haigler CH, Srinivas NG, Martin LK, Taylor JG (1997) Alterations of leaf

photosynthesis and fiber cellulose synthesis by cool night temperatures. In Proceedings



Beltwide Cotton Conferences, January 6-10, New Orleans, LA, National Cotton Council,

TN pp 1435-1436.

Kaur R, Singh OS (1992) Response of growth stages of cotton varieties to moisture stress.

Indian J Plant Physiol 35: 182-185.

Meek CR, Oosterhuis DM, Steger AT (1999) Drought tolerance and foliar sprays of glycine

betaine. In Proceedings Beltwide Cotton Conferences, January 3-7, Orlando, FL, USA.

Memphis, USA. National Cotton Council, 559-561.

Miller JK, Krieg DR, Paterson RE (1996) Relationship between dryland cotton yields and

weather parameters on the Southern Hig Plains. In Proceedings Beltwide Cotton

Conferences, January 9-12, Nashville, TN, USA, Memphis, USA, National Cotton

Council, 1165-1166.

Moseley D, Landivar JA, Locke D (1994) Evaluation of the effect of methanol on cotton

growth and yield under dry-land and irrigated conditions. In Proceedings Beltwide Cotton

Conferences, January 5-8, San Diego, CA, USA. Memphis, USA. National Cotton

Council, 1293-1294.

Oosterhuis DM (1997) Effect of temperature extremes on cotton yields in Arkansas. In

Proceedings of the Cotton Research Meeting, held at Monticello, Arkansas, USA,

February 13 [Edited by Oosterhuis, D.M.; Stewart, J.M.]. Special Report-Agricultural

Experiment Station, Division of Agriculture, University of Arkansas, No. 183: 94-98.

Reddy KR, Hodges HF, McKinion JM (1993) Temperature effects on Pima cotton leaf

growth. Agron J 85: 681-686.

Reddy KR, Hodges HF, McKinion JM (1995) Carbon dioxide and temperature effects on

pima cotton growth. Agriculture Ecosystems & Environment 54: 17-29.

Reddy KR, Robana RR, Hodges HF, Liu XJ, Mckinion JM (1998) Interactions of CO2

enrichment and temperature on cotton growth and leaf characteristics. Environ Exp Bot

39: 117-129

SAS Institute, Inc (1985) SAS User‟s Guide: Statistics. 5

th ed. SAS Institute, Inc., Cary, NC.

pp. 433-506.

Sawan ZM, Hanna LI, McCuistions WL (2006) Appropriate time scale for aggregating

climatic data to predict flowering and boll setting behaviour of cotton in Egypt.

Communication in Biometry and Crop Science 1: 11-19.

Sawan ZM, Hanna LI, McCuistions WL, Foote RJ (2010) Egyptian cotton (Gossypium

barbadense) flower and boll production as affected by climatic factors and soil moisture

status. Theoretical and Applied Climatology 99: 217-227.

Schrader SM, Wise RR, Wacholtz WF, Ort DR, Sharkey TD (2004) Thylakoid membrane

responses to moderately high leaf temperature in Pima Cotton. Plant, Cell and

Environment 27: 725-735.

Ward DA, Bunce JA (1986) Responses of net photosynthesis and conductance to independent

changes in the humidity environments of the upper and lower surfaces of leaves of

sunflower and soybean. J Exper Botany 37: 1842-1853.

Warner DA, Burke JJ (1993) Cool night temperatures alter leaf starch and photosystem II

Chlorophyll fluorescence in cotton. Agron J 85: 836-840.

Wise RR, Olson AJ, Schrader SM, Sharkey TD (2004) Electron transport is the functional

limitation of photosynthesis in field-grown Pima cotton plants at high temperature. Plant,

Cell and Environment 27: 717-724.



Xiao J-F, Liu Z-G, Yu X-G, Zhang J-Y, Duan A-W (2000) Effects of different water

application on lint yield and fiber quality of cotton under drip irrigation. Acta Gossypii

Sinica 12: 194-197.

Zhao Y-Z (1981) Climate in Liaoning and cotton production. Liaoning Agricultural Science

No. 5: 1-5.

Zhou Z-G, Meng Y-L, Shi Pei, Shen Y-Q, Jia Z-K (2000) Study of the relationship between

boll weight in wheat-cotton double cropping and meteorological factors at boll-forming

stage. Acta Gossypii Sinica 12: 122-126.



INDEX

#

21st century, 211

A

abstraction, 139

access, 31, 171, 199, 240

accessibility, 181

accountability, 31

acetic acid, 143, 162

acetonitrile, 9, 11, 12

acid, 2, 3, 11, 13, 16, 17, 20, 52, 143, 144, 147, 162,

164, 197, 225

ACs, 75

adaptability, 198, 212

adaptation(s), viii, ix, 103, 116, 165, 166, 167, 168,

186, 187, 188, 195, 197, 199, 204, 209, 214, 216,

221, 231, 237, 248, 256, 258

additives, 144, 147

adsorption, 36, 38, 41, 42

adverse conditions, 88

adverse effects, 166, 227, 256

aerosols, 2, 3, 16, 18, 20

aesthetic(s), 110, 113, 201

Africa, 39, 133, 143, 188, 219, 222, 257

age, 200, 215, 232

agencies, 122

agglutination, 159

aggregation, 173, 174, 178, 181, 184, 185, 186, 187

aggregation process, 173

agricultural exports, 174

agriculture, ix, 114, 118, 135, 166, 172, 175, 181,

184, 186

AIBN, 17

air pollutants, 51, 52

air quality, 52

air temperature, xi, 63, 87, 90, 97, 112

Alaska, 209

allele, 237, 238, 255

alternative energy, 106, 128

altruism, 204

aluminium, 93

ambient air, 46, 50, 52

amino acid(s), 147

ammonia, 78, 82, 83

amplitude, 32, 231

anaerobic digestion, 108

angiosperm, 216, 217, 222, 230, 232, 258

animal fleshings, viii, ix, 137, 148, 157

animal husbandry, 108, 133

annealing, 243

antibiotic, 162

antibody, 16

antigen, 16

appropriate technology, 134

architects, 104, 115, 134

Argentina, 223

argon, 17

arid conditions, viii, 137, 138, 332

arsenic, 36

ascorbic acid, 225

Asia, 190, 221, 222

assessment, xi, 105, 115, 166, 168, 170, 172, 173,

174, 186, 187, 188, 190, 209, 216, 236, 259, 261

assessment tools, 173

ATLAS, 228

atmosphere, 26, 30, 39, 70, 73, 75, 112, 166

attitudes, 112

Australasia, 218, 220, 261, 262

authorities, 126

avoidance, 110

awareness, 26, 117, 118, 133, 171


Index 348

B

Bacillus subtilis, 144

bacteria, 141, 142, 143, 144, 145, 147, 161

bacteriostatic, 143

banks, 110

barriers, 73, 83, 118, 246

base, 4, 10, 14, 39, 126, 190

base pair, 244

basic services, 171

batteries, 108

beetles, 207, 258

beneficial effect, 66

benefits, 31, 70, 83, 86, 105, 110, 111, 112, 199,

201, 207

benign, 26

benzene, 17

bias, 12, 14

bilateral, 236

bile, 142, 161

biodiesel, ix, 138, 147, 160, 164

biodiversity, 39, 198, 202, 204, 208, 209, 210, 211,

249, 250, 259, 261, 262

biogas, viii, 103, 104, 108, 133, 164

biogeography, 219, 236, 238, 262

biological processes, 255

biomarkers, x

biomass, viii, 103, 104, 122, 133, 188, 226

biotechnology, 164

biotic, 216

biotic factor, 252

blends, viii, 23, 96, 99

boilers, viii, 24, 27, 95, 99

Bolivia, 223

bonding, 100

boreal forest, ix, 195, 196, 197, 198, 200, 202, 207,

208, 209, 210, 211

boreholes, 94, 99

boric acid, 143, 144, 162

breeding, 237, 238

Britain, 25, 160

bromine, 70

browsing, 199, 208, 226

buffalo, 143

building blocks, 109

burn, 108, 199

businesses, 98, 184

C

cadmium, 36

calcium, xi, 142, 147

calculus, 48

calibration, 12, 14, 17, 18, 19, 20

cancer, 52

capacity indicators, 187

capillary, 2

carbohydrates, 147

carbon, viii, 24, 95

carbon dioxide, 25, 75, 108, 115

case study(ies), x, 134, 162, 168, 187, 213, 214, 215,

229, 253, 255, 258

casting, 105, 128

catalysis, 16

catastrophes, 213

cattle, 141, 143, 144, 145, 161, 162

CCA, 190

centigrade, 175

challenges, 147, 191, 196

changing environment, 199, 216, 221, 256

cheese, 138, 146, 163

chemical(s), ix, 1, 2, 4, 5, 6, 8, 11, 16, 20, 26, 39, 43,

51, 69, 137, 139, 140, 141, 142, 143, 145, 147,

148

chemical stability, 69

child mortality, 191

children, 121

Chile, 222

chlorine, 13, 14, 70

chloroplast, 239, 243, 244, 245, 259, 263

chromatography, 2

chromium, 36

circulation, 69, 85, 99, 249, 250

Clarence River Corridor, x, 213, 215, 224, 227, 228,

241, 243, 251, 252, 253, 254, 255

classification, 28, 135, 219

clean energy, 171

cleaning, 38

clients, 112

climate change, vii, ix, x, 30, 39, 165, 166, 167, 168,

170, 171, 172, 173, 186, 187, 188, 190, 196, 200,

203, 207, 208, 209, 213, 214, 215, 222, 223, 227,

240, 246, 249, 250, 252, 253, 255, 256, 257, 259,

260

climates, 73, 93, 208, 246, 255

climatic factors, xi

closure, 140

cluster analysis, 168

clustering, 189

clusters, 116, 138, 225, 241

CO2, viii, 24, 27, 51, 52, 70, 71, 95, 96, 108, 115,

200

coatings, 17

cobalt, 36

coconut oil, 174


Index 349

collagen, 140, 146, 159

colonization, 221, 244, 245, 263

color, 225

combustion, 36, 51, 96, 108

commercial, vii, 23, 69, 70, 74, 98, 134, 144, 147,

160, 162, 181, 199

commodity, 208

communication, 104, 132, 206

community(ies), ix, x, 108, 161, 168, 170, 186, 187,

195, 201, 203, 208, 209, 210, 211, 213, 214, 216,

217, 220, 223, 226, 227, 229, 231, 232, 233, 235,

247, 252, 255

compaction, 200

comparative advantage, 226

comparative costs, 119

compensation, 38

competition, xi, 106, 200, 216, 222, 223, 225, 232,

246, 250, 252

compilation, 224

complexity, ix, 104, 149, 165, 166, 174, 195, 201,

206, 207, 247

composition, xi, 40, 54, 68, 146, 161, 166, 200, 201,

208, 226, 227, 247

compounds, 2, 4, 9, 20, 70, 142, 226

compressibility, 48

compression, viii, 24, 51, 56, 58, 59, 61, 62, 63, 76,

85, 86

computation, 176

computer, 159, 245

conceptual model, 197

concreteness, 113

condensation, 4, 8, 11, 14, 54, 61, 94

conditioning, 26, 27, 32, 52, 63, 70, 83, 90, 98, 100,

106, 109, 110, 111, 116, 117, 129, 135

conductivity, 100, 157, 158

configuration, xi, 28, 33, 96, 247

congress, 134, 135, 160

conifer, x, 199, 200, 208, 210, 213, 214, 224, 227,

240, 244, 252, 253, 258, 259, 261, 262

connectivity, 200, 205, 246

consensus, 209

conservation, x, 26, 36, 43, 44, 93, 104, 110, 112,

126, 128, 134, 135, 205, 213, 214, 215, 236, 238,

244, 246, 252, 255, 256, 258, 263

construction, 69, 72, 75, 105, 107, 108, 111, 112,

113, 114, 115, 118, 121, 244, 246

consumers, 70, 83, 92, 105, 112, 118

consumption, vii, viii, x, xi, 2, 23, 27, 30, 31, 32, 51,

69, 70, 71, 72, 73, 74, 83, 97, 105, 107, 109, 110,

111, 114, 118, 121, 126, 127, 133, 135, 137, 139,

145, 146, 147, 154, 158, 159, 160

contact time, 18, 19

containers, 26, 144

contamination, 37, 38, 134, 142, 145

Convention on Biological Diversity, 211

cooking, 108

cooling, vii, x, 6, 23, 25, 26, 27, 31, 46, 52, 65, 69,

70, 71, 72, 73, 75, 83, 84, 89, 90, 91, 94, 95, 96,

97, 98, 99, 100, 109, 111, 112, 117, 134, 141,

229, 230

cooperation, 106, 160

coordination, 108

copolymer, 17

copper, 4, 36, 94

correlation(s), xi, 118, 121, 132, 142, 250

corrosion, 29, 53, 69, 88, 142

cost, 2, 30, 36, 43, 57, 68, 72, 74, 86, 89, 92, 94, 95,

98, 99, 104, 106, 107, 108, 114, 119, 126, 127,

128, 129, 159, 171

cost saving, 110, 112

cotton, xi

covering, 185

CPC, 167

crop(s), 38, 174, 175, 186

crown, 207

crystalline, 141

crystals, 17, 145

CSD, 101

CT, 10, 14, 126

cultivation, 226

culture, 117

cure, 138, 145

curing process, viii, ix, 137, 143, 146, 148, 149, 150,

158, 160

customers, 92

cuticle, 226

CWA, 18

cycles, x, 24, 46, 47, 52, 53, 54, 62, 63, 64, 65, 66,

68, 76, 77, 86, 87, 101, 213, 216, 227, 231, 232,

233, 235, 240, 252, 255, 256

cycling, 86, 117, 196, 200, 207

cytoplasmic inheritance, 214, 244

Czech Republic, 137, 160

D

damages, 38, 167

data collection, 250

data set, 239

database, 176, 228

decay, 141

decision-making process, 115

decomposition, 147

defence, 72

deficiencies, 107

degradation, 12, 14, 52, 138, 144, 162, 174, 179, 202


Index 350

dehydration, 144

demographic change, x, 213, 215

denuder extraction, vii, 1, 3

deposition, 200, 214, 217, 262

deposits, 29, 216, 217, 233, 234, 235

depression, 238, 239

depth, 27, 32, 94, 108, 125, 197, 235

desalting process, ix, 138, 152, 155

designers, 75, 111, 112, 114

desorption, 36, 42, 52

destruction, 35, 52, 103, 226

detection, vii, 1, 2, 3, 9, 16, 17, 18, 19, 20

developed countries, 191

developing countries, 104, 107, 145

deviation, 17, 66

dew, 54

diesel engines, 51

diesel fuel, 106

diffusion, ix, 42, 137, 140, 144, 146, 148, 151, 152,

156, 157, 158, 160, 163

diffusion process, 149

diffusivity, 32, 69

digestion, 108, 164

diluent, 5

dinosaurs, 219, 226

diploid, 237, 238

directives, 51

discomfort, 27, 83

diseases, x

displacement, 11, 58, 62, 191, 217, 252

disposition, 115

distilled water, 13, 157

distribution, x, 3, 29, 72, 94, 95, 97, 99, 100, 105,

140, 163, 171, 176, 203, 207, 213, 214, 215, 216,

221, 222, 223, 224, 229, 235, 237, 241, 243, 245,

246, 248, 250, 251, 252, 253, 254, 255, 256, 257,

258, 261

distribution of income, 171

divergence, x, 213, 214, 215, 227, 229, 236, 239,

241, 246, 247, 252, 254, 256, 262

diversification, 122, 214, 217, 230, 238

diversity, x, 198, 202, 203, 204, 208, 210, 211, 213,

214, 215, 216, 217, 218, 220, 222, 223, 227, 229,

230, 231, 236, 237, 238, 240, 241, 247, 248, 249,

252, 253, 255, 256, 257, 261, 262, 263

DMF, 52

DNA, 239, 244, 245, 259, 263

DOI, 209, 210

domestication, 214, 238

dominance, 203, 230, 232, 233, 235, 238, 258

dosage, 143

drainage, 37, 96, 230

drawing, 203

dreaming, 224

drinking water, 108

drought, 166, 188, 189, 235

dry ice, 141

dry matter, 146, 159

drying, 140, 141, 142

durability, 89, 115

dusts, 136

dynamic factors, 3

E

early warning, 171

Earth Summit, 114, 134

ecological indicators, ix, 165, 170

ecological processes, 196, 197, 199, 201

ecological systems, 207, 209, 212, 249

ecology, 216, 247, 252, 258

economic damage(s), 38

economic development, viii, ix, 103, 114, 166, 186

economic growth, 30, 106, 129, 184

economic indicator, 174, 181

economic well-being, 171

economics, 108

ecosystem, vii, 23, 52, 69, 70, 142, 190, 195, 196,

197, 198, 199, 200, 201, 202, 203, 204, 205, 206,

208, 209, 210, 211

education, viii, 103, 117, 118, 121, 171, 174, 181,

184, 186, 189

effluent, 18, 51, 160, 162, 164

effluents, ix, 37, 137, 139, 142, 144, 158, 163

Egypt, 39

elastin, 140

electric conductivity, 157

electric power, ix, 75, 108, 137, 158, 159

electrical potential, xi, 295, 297

electricity, viii, 24, 27, 31, 83, 95, 97, 98, 104, 106,

108, 109, 118, 121, 129, 132

electrode surface, 13

electrodes, vii, 1, 3, 13, 14, 16, 20

electrolyte, 140

electron, 139, 143

electroplating, 36

emergency, 86, 87

emission, 115

empirical methods, 166

employment, 110, 175, 181, 184, 186, 190

encoding, 176

endangered, 258, 262

energy conservation, 26, 36, 104, 110, 112

energy consumption, x, xi, 31, 32, 71, 72, 83, 97,

104, 110, 111, 114, 118, 135


Index 351

energy efficiency, vii, 23, 69, 70, 75, 83, 89, 104,

106, 109, 113, 116, 118, 129, 135

energy expenditure, 27, 83

energy input, 46, 104, 105

energy recovery, 117

energy supply, 31, 104, 116, 171

enforcement, 117

engineering, ix, 2, 26, 44, 111, 132, 137, 145, 148

England, 188

entrepreneurship, 163

entropy, 24, 29, 35, 36, 42, 43, 45, 48, 49, 50, 53, 54,

55, 57, 65, 76

environmental change, 166, 170, 186, 187, 196, 214,

216, 231, 249

environmental characteristics, 201

environmental conditions, x, 104, 106, 107, 129,

195, 196, 231

environmental control, 116

environmental degradation, 51

environmental factors, 115, 200, 245

environmental impact, ix, 31, 51, 73, 83, 95, 98, 115,

132, 137, 138, 143

environmental issues, 112

environmental management, 208

environmental protection, vii, 71, 83, 135

environmental quality, 106, 129, 132

environmental regulations, 111

environmental stress, 167, 172

environmental sustainability, 104, 106, 128

environmental variables, 245, 250

environments, 16, 26, 51, 108, 110, 111, 141, 217,

231, 232, 234

enzyme(s), x, 147, 164

equality, 64

equilibrium, 2, 41, 42, 43, 47, 76, 81, 82, 158, 237,

238, 250

equipment, 26, 28, 34, 36, 64, 68, 87, 89, 90, 96, 99,

104, 105, 107, 117, 142

erosion, 174, 179, 193, 200

ethanol, x

EU, 26, 50, 51

Europe, 25, 39, 114, 126, 127, 141, 187, 258

European Commission, 144, 162

europium, 16

evaporation, x, xi, 4, 6, 54, 60, 94, 138, 162, 230

evidence, 51, 189, 210, 216, 223, 227, 232, 233, 240,

245, 249, 254, 261, 262, 263

evolution, 219, 221, 223, 226, 227, 229, 237, 247,

258, 259, 262

exclusion, 209, 221

exercise, vii, 23, 69, 70

experimental design, 139

expertise, 167

exploitation, 105, 112, 128

exports, 174

exposure, 109, 167, 168, 170, 172, 173, 174, 176,

181, 184, 185, 186, 187, 190, 232

expressiveness, 113

external environment, 52

extinction, 213, 214, 216, 217, 222, 226, 233, 253,

256, 259

extraction, vii, 1, 3, 20

extreme weather events, 171, 172

F

fabrication, 100

families, 174, 175, 186, 217, 238, 244, 259

family income, 186

farmers, 38, 186

farms, 108, 121

fat, 140, 141, 146, 147, 151, 160, 225

fatty acids, 147

fauna, 215, 226, 232

feed additives, 147

fermentation, 108, 164

fertilization, 225, 244

fertilizers, 147

fever, 52

fiber(s), xi, 17, 18, 19, 140, 159

filters, 3, 88

financial, 31, 106, 129, 139, 174

financial intermediaries, 122

financial resources, 171

Finland, 134, 209

fire fighting, 2

fires, 199, 200

fish, 51, 52, 144

fishing, 144

fitness, 238

fixation, 237

flame, 17

flammability, 226

flavor, 225

flex, 93

flexibility, ix, 166, 167, 173, 187

flight, 235

floods, 172

flora, 217, 218, 220, 229, 230, 231, 235, 258

flora and fauna, 258

flowering period, 38

fluctuations, 126, 217, 229, 256

fluid, 25, 28, 29, 32, 33, 35, 36, 37, 43, 60, 66, 98,

235

fluorine, 70

food, 26, 75, 146, 170, 171, 174, 175, 186, 226


Index 352

food security, 75

forbs, 199

force, xi, 97, 186

forecasting, 206, 246, 257

foreign affairs, 106

forest ecosystem, ix, 195, 196, 197, 199, 200, 202,

203, 204, 206, 207, 209, 211

forest fire, 209

forest habitats, 222

forest management, ix, 195, 196, 197, 198, 201, 203,

204, 205, 206

forest resources, 205

formaldehyde, x, 279, 280, 283, 284, 285, 286, 291,

292

formation, xi, 39, 69, 87, 142, 145, 147, 151, 229,

257

formula, 35, 116, 118, 126

fossil record, x, 213, 214, 215, 216, 219, 222, 232,

235, 240, 247, 253, 254, 255, 256, 259

fossils, 216, 222, 232, 259

fouling, 29, 145

foundations, 237

founder effect, 208

fractal theory, 29

France, 135, 249

free energy, 42

free volume, 140

freedom, 16

freezing, 140, 141

frequency distribution, 237

freshwater, 51

frost, 87, 97

fruits, 225

fuel consumption, 27, 109, 126, 127

fuel efficiency, 109

fuel prices, 106, 128

funding, 257

fungi, 211

fuzzy set theory, ix, 166, 172, 187, 188

G

gamete, 235

gamma rays, 144

gas injection technology, vii, 1

gas sensors, vii, x

gaseous ethanol, xi, 279, 280, 286, 287, 288, 289,

290, 291, 293

GDP, 174

GDP per capita, 171

gel, 147

genes, 202, 237, 238, 239

genetic diversity, x, 203, 213, 214, 215, 227, 236,

237, 240, 241, 247, 248, 249, 253, 255, 262

genetic drift, 237, 238, 253

genetic load, 238, 255, 256

genetic traits, 236

genetics, 236, 237, 238, 239, 244, 247, 257, 260,

261, 262

genome, 244, 259, 263

genotype, 237, 250, 256

genotyping, 250

genus, 217, 219, 222, 225, 232, 255

Geographic Information System (GIS), 190, 245

geothermal heat pump, 27, 96, 100

Germany, 165, 209, 249

germination, 203, 225

GHG, 39

GHP, 96

global climate change, 249

global warming, viii, 23, 52, 70, 75, 96, 214

globalization, 190

glycerin, 160

glycol, 40

governance, 171, 174, 178, 184, 192

governments, 31

grants, 133

graph, 91, 157

grasslands, 200

gravity, 24, 43, 94, 226

Great Britain, 160

green buildings, 115

green electricity, viii, 24, 95

greener world, vii, 23, 69, 70

greenhouse gas(es), 30, 106, 112, 129, 171

greenhouse gas emissions, 30, 106, 129, 171

grid electricity, viii, 24, 95

groundwater, 37, 100

growth, viii, 13, 14, 30, 39, 75, 103, 105, 106, 114,

115, 116, 122, 129, 132, 141, 181, 184, 198, 211,

214, 216, 225, 227, 231

growth rate, 103

growth rings, 225

guidelines, x, 93, 107, 114, 115, 195, 211

guiding principles, 204, 208

gymnosperm, 217, 218, 258, 259

H

habitat(s), 141, 199, 201, 203, 211, 215, 216, 222,

223, 225, 226, 227, 229, 233, 235, 240, 241, 247,

249, 252, 255, 256

halitosis, xi

haplotypes, 243, 244, 245, 260

harmony, 110


Index 353

harvesting, 198, 199, 200, 209, 210, 211, 218, 220

hazardous air pollutants, 51, 52

hazards, 52, 162

health care, x, 188

health condition, 174

health risks, 143, 172

health services, 172, 189, 192

heat capacity, 35, 54

heat loss, 68, 72, 76, 82, 93, 94, 97, 109, 110, 135

heat pump cycle, 89

heat pumps, viii, 24, 25, 26, 27, 86, 87, 88, 90, 91,

96, 97, 98, 99, 100

heat transfer, 24, 25, 26, 28, 29, 32, 33, 36, 45, 56,

58, 85, 100, 111, 135

heavy metals, 36

height, 109, 124, 125, 253

hemisphere, 214

herpetofauna, 254

heterogeneity, 37, 198, 202

heterosis, 239

heterozygote, 237

highlands, 222

historical data, 247

history, 198, 201, 214, 215, 217, 229, 231, 234, 236,

239, 240, 246, 247, 249, 253, 257, 258, 260, 261,

262, 263

Holocene, 221, 231, 235, 241, 254, 255, 258

homeowners, 96

homes, 90, 96, 98

homogeneity, 159

hotels, 184

hotspots, ix, 165, 187, 188, 223

House, 95, 135

household sector, 133

housing, 32, 134

human, ix, x, 31, 39, 51, 72, 108, 113, 114, 165, 166,

171, 186, 187, 199, 201, 204, 214, 256

human activity, 199

human body, x

human condition, 186

human development, 171

human dimensions, 166, 187

human health, 39

human security, 171

humidity, xi, 90, 97, 98, 106, 129, 142, 175, 225

Hunter, 202, 205, 209, 211

hunting, 199

husband, 121

husbandry, 108, 133

hybrid, 239

hybridization, 238

hydrocarbons, 48, 71

hydrochloroflurocarbons, viii, 23, 96

hydrogen, 13, 20

hydrogen bromide, 20

hydrogen chloride, 20

hydrogen fluoride, 20

hydrolysis, 16, 147, 164

hydrothermal synthesis, 40

hydroxide, 13, 17, 147

hypothesis, 223

I

IAM, 189

ideal, 34, 45, 47, 57, 63, 64, 65, 66, 67, 91, 101, 105,

126, 149, 152, 216

identification, 168, 187, 235, 236, 245, 246

IEA, 101

illiteracy, 121

image, 17

immigration, 229

impact assessment, 166, 250

imprinting, 16, 17

impurities, 4, 5, 6, 8, 38, 88, 145, 147

inbreeding, 227, 237, 238, 239, 253, 256

incidence, 174, 175, 181, 227

income, 121, 171, 174, 181, 186, 190

increased competition, 252

India, 100, 162, 172, 188, 190

indirect effect, 207, 211

individuals, 186, 206, 227, 228, 237, 238, 244, 250,

253, 258

industrial chemicals, 5

industrial processing, 151

industrialisation, viii, 103

industries, 26, 36, 139, 164

industry, vii, viii, 23, 36, 69, 70, 72, 73, 75, 83, 122,

137, 138, 139, 143, 144, 146, 160, 161, 164, 190,

225

inequality, 174, 189, 191

inertia, 28

infancy, 187

infant mortality, 191

information exchange, 115, 206

infrastructure, 39, 109, 122, 125, 126, 170, 174, 181,

193

ingredients, 113

inheritance, 214, 239, 244

inhibitor, 100

initiation, 225

insects, 141

insecurity, 170

institutional infrastructure, 125, 170

institutions, 122, 204

insulation, 4, 52, 72, 89, 91, 93


Index 354

integration, 17, 45, 53, 104, 151, 153, 236

integrity, 201, 242, 255

integument, 219

interface, 28, 189

interference, 256

intermediaries, 122

internal processes, 166

internalised, 31

intervention, 122

introns, 244

investment(s), 126, 144, 181, 184

ion adsorption, 36

ions, xi, 13, 16, 36, 38, 43, 142, 149

IPO, 188

iron, 36, 39

irradiation, 113, 143, 162

irrigation, 125, 171, 186

islands, 174, 175, 181, 184, 185, 226

isolation, 110, 114, 215, 223, 246, 247, 252, 253

isotherms, 37, 42

isotope, 231

isozyme, 223

issues, ix, 2, 30, 31, 91, 105, 106, 110, 112, 113,

114, 129, 134, 165, 166, 184, 186

Italy, 249

iteration, 204

J

Japan, 39, 249

joint ventures, 122

joints, 93

Jordan, 118, 134, 220, 229, 232, 258, 261

K

Kenya, 39, 138, 160

kerosene, 26, 118

kinetic model, 41

kinship, 244, 263

KOH, 13

Korea, 249

L

labour force, 186

lakes, 138, 142, 210, 230

landscape(s), ix, 116, 117, 195, 198, 199, 201, 203,

205, 206, 207, 208, 210, 213, 215, 216, 217, 226,

229, 232, 231, 233, 247

laws, 46, 65

LC-MS, vii, 1, 3, 9, 11, 12, 13, 19, 20

LDCs, 191

lead, 2, 28, 36, 39, 67, 104, 109, 114, 139, 142, 147,

199, 200, 204, 231

leakage, viii, 24, 26, 91, 93, 96, 99, 115

leaks, 93

learning, 122

legislation, 117

legs, 93

lichen, 210

life cycle, 115, 227

lifetime, 126, 128

light, 18, 31, 32, 87, 110, 112, 144, 200, 201, 214,

218, 220, 222, 223, 226, 235, 240, 252

linear dependence, 157

liquid chromatography, 2

liquid phase, 2, 54, 84, 146

liquids, 88

lithium, 75, 76, 77, 78, 79, 80, 81

livestock, 136

living conditions, 104

loans, 133

loci, 237, 240, 243, 244, 245, 250, 258, 260

locus, 219, 237, 243, 244, 257

logging, 199, 201

longevity, 86, 223

low temperatures, xi

LPG, 95, 118, 121

luminescence, 17

Luo, 100

M

machinery, 38, 144

macromolecules, 140

magnesium, xi, 142

magnetic refrigeration, vii, 23, 69, 70

magnetic resonance, 163

magnetic resonance imaging, 163

magnitude, 28, 36, 59, 158, 172, 196, 197, 229

majority, 140, 145, 147, 214, 219, 222, 226, 244,

255, 256

Malaysia, 101, 220

man, 51, 52, 72, 83, 97

management, viii, x, 38, 103, 111, 114, 121, 133,

164, 195, 196, 197, 198, 199, 200, 201, 203, 204,

205, 206, 207, 208, 209, 211, 215, 256, 258

manganese, 36

manufacturing, 37, 104, 107, 113, 115, 138, 139,

161, 164

mapping, 170, 186

marginal distribution, 253

marine environment, 217, 262

market access, 31


Index 355

marketing, 115

mass, 2, 8, 14, 24, 28, 29, 33, 42, 43, 44, 45, 46, 47,

51, 56, 58, 62, 69, 111, 112, 114, 117, 118, 149,

152, 160, 163, 213, 217, 221

material resources, 114

materials, vii, 1, 2, 4, 5, 6, 9, 13, 16, 20, 26, 31, 39,

69, 70, 73, 83, 104, 105, 107, 108, 114, 115, 132,

140, 143, 159

matrix, 16, 247, 256

matter, 40, 47, 113, 117, 121, 139, 146, 159, 210

maximum likelihood estimate (MLE), 253

MB, 257, 259, 260, 261

measurement(s), 12, 13, 14, 17, 20, 34, 113, 171,

203, 207

meat, 138, 146, 159, 163

mechanical properties, 140

mechanical ventilation, 31, 72

media, 40

median, 39

medical, xi, 108

medicine, 138

Mediterranean, 111

megaspore, 225

meiosis, 244

membership, 173, 176

membranes, xi, 145

mercury, 8, 12, 36

Mercury, 21

messages, 113

metabisulfite, 144

metal ion(s), 36, 38, 43

metals, 16, 36, 37

metaphor, 206, 207

meter, 108

methanol, 7, 17

methodology, 100, 235, 250

Mexico, 189, 219

microclimate, 105

microfabrication, 28

microhabitats, 252

micrometer, 28

microorganisms, 140, 141, 142

microsatellites, 243, 260, 262

microscope, 139

microscopy, 222, 235

migration, 216, 221, 223, 229, 235, 236, 247, 252,

253, 255, 256

migration routes, 222, 229

milligrams, 2

miniature, 17, 18, 20

Ministry of Education, 160

Miocene, 230, 261

MIP, 16, 17, 18, 19, 20

mission(s), 27, 39, 96, 115

Missouri, 261

mitochondria, 244

mitochondrial DNA, 257

mixing, 105

modelling, 29, 68, 111, 116, 139, 172

models, ix, 36, 109, 116, 139, 145, 146, 165, 166,

167, 168, 172, 173, 174, 176, 178, 179, 181, 184,

187, 188, 189, 208, 228, 229, 239, 240, 241, 247,

248, 249, 250, 252, 254, 262

modern society, 104

modifications, 111

modules, 29

moisture, 88, 140, 200, 204, 225, 229

moisture content, 140

mole, 17, 54

molecular mass, 47

molecular sensors, 16

molecules, 17, 43, 159

momentum, 106, 128

monolayer, 37

morphology, 217, 218, 220, 222, 235, 261, 263

mortality, 191, 199

MPI, 249

MRI, 249

MSW, 164

mtDNA, 245

mutation(s), 237, 244, 260

myosin, 159

N

NaCl, 13, 138, 140, 141, 143, 144, 145, 146, 149,

150, 152, 154, 156, 158, 163

nanocrystals, 40

nanomaterials, 41

nanoparticles, 36, 39, 40, 43

naphthalene, 142

National Academy of Sciences, 210

national income, 171

natural disturbance, 197, 200

natural evolution, 256

natural gas, 74, 95

natural resource management, 209

natural resources, viii, 70, 83, 103, 115, 167

natural selection, 237, 260

negative effects, 109, 132

negative relation, 181

neglect, 111

nerve, 16

Netherlands, 106, 133

neutral, 237, 244

New South Wales, 225, 227, 253, 258, 260


Index 356

New Zealand, 218, 220, 222, 225, 230, 261, 263

NGOs, 122

nickel, 36

Nigeria, 39

nitrates, 40, 52

nitrogen, 8, 12, 17

NOAA, 249

nonlinear dynamics, 196

North America, 39, 203, 207, 208

Norway, 211, 249

Norway spruce, 211

nuclear magnetic resonance, 163

nucleus, 132, 244

nutrient, 196, 200, 223, 252

O

oil, viii, 24, 25, 26, 49, 51, 72, 88, 94, 95, 96, 104,

133, 174, 216

oil boiler, viii, 24, 96

openness, 106, 128

operating costs, 86, 99, 145, 159

operations, viii, 36, 137, 147, 148

opportunities, 30, 72, 110, 116, 133, 167, 181, 198,

199

optical fiber, 17

optimization, ix, 137, 138, 139, 143, 145, 146, 148,

155, 160

ores, 17, 234, 235

organelle, 244

organic matter, 210

organism, 237, 244

organize, 196, 204

organs, 113

oscillation, 258

osmosis, 145

outreach, 172

outsourcing, 184

overlap, 229, 248

oxide nanoparticles, 39

oxygen, 40, 70, 231

ozone, 52, 70, 200

P

paints, 36

palaeodistribution, x, 213, 215

parallel, 32, 81, 92, 99, 110, 219

parents, 238

pasture, 172

pathogens, 208, 216, 246

pathology, 209

PCR, 243

peace, ix, 166

peat, 199, 233, 234

peptide(s), 140, 144

peptide chain, 140

permeability, 100, 111

Perth, 235

pests, 200, 208

petroleum, 26, 104

pH, 13, 17, 18, 36, 38, 40, 144, 200

phase transformation, 40

phenotype, 250

Philadelphia, 189

Philippines, v, vii, ix, 165, 166, 171, 174, 175, 179,

184, 185, 186, 187, 188, 190

phosphate, 3, 6, 13

photosynthesis, 200

physical environment, 113, 217

physical properties, 140

physiology, 222, 258

piano, 225

Pie chart, 243

plants, vii, 23, 51, 69, 70, 73, 75, 106, 116, 126, 127,

129, 138, 143, 145, 199, 210, 216, 217, 225, 234,

244, 247, 257, 259

plastid, 240, 244, 245

platform, 3, 17, 20

Pliocene, 230

polarization, 145

policy, 29, 111, 168, 171, 186, 187, 188, 189, 206,

211

policy issues, 30

political problems, 184

politics, 30

pollen, 221, 222, 225, 226, 230, 232, 233, 234, 235,

236, 244, 245, 258, 260, 261, 263

pollen tube, 225

pollination, 38, 196, 200, 235, 259, 263

pollutants, 51, 52, 135

pollution, viii, 24, 30, 36, 37, 52, 95, 101, 108, 113,

200

polymer(s), 3, 16, 17, 20, 144, 162

polymer matrix, 16

polymorphism, 244

polyploidy, 237

polysaccharide, 159

pools, 98, 210

poor performance, 114

population density, 103, 108, 109, 117

population group, 223, 227, 229, 242, 247, 248, 251,

252, 255, 256

population growth, viii, 103, 105, 115, 122, 214

population size, 237, 239, 252, 253


Index 357

population structure, 237

porosity, 140

porous materials, 140

portability, 20

Portugal, 136, 172

positive relationship, 181

potassium, 13, 143, 162

potato, 163

poverty, 163, 170, 171, 174, 175, 181, 186, 188, 189,

190

power generation, 108

precedents, 197, 199, 201

precipitation, 39, 40, 200, 229, 230, 232, 245, 256

predation, 246

preparation, 2, 3, 143, 171, 176

preservation, ix, 137, 138, 139, 141, 142, 143, 144,

147, 158, 160, 161, 162, 216

prevention, 69, 138, 144, 145, 162

principles, 2, 47, 115, 116, 134, 204, 208, 211

prior knowledge, 245

probability, 110, 118, 120, 189, 228, 229, 239, 241,

250, 253, 254

probe, 158

process control, 28

producers, 38, 105

professionals, 104, 112

profit, 147

profitability, 116

project, 101, 106, 107, 133, 244, 246

propagation, 38, 226, 239

propane, 71

protection, vii, 71, 83, 135, 209, 252

proteins, 144

prototype, 134

Pseudomonas aeruginosa, 147, 164

PTFE, 4, 9, 10, 13, 14

public administration, 38

public awareness, 117, 133

public health, 147

public sector, viii, 103

public service, 104

pulp, 37

pumps, viii, 11, 14, 24, 25, 26, 27, 52, 81, 82, 86, 87,

88, 90, 91, 94, 96, 97, 98, 99, 100, 103, 104, 106,

107, 117, 123, 124, 126, 127, 132, 133

purchasing power, 171, 178, 191

pure water, x, 76, 81

purity, 7, 11

Q

quality control, 107, 134

quality standards, 51

Queensland, 218, 225, 253, 257, 260, 262

R

radiation, 52, 73, 104, 107, 108, 118, 143, 162, 258

rainfall, 175, 216, 229, 230

rainforest, vii, x, 213, 214, 216, 217, 221, 223, 225,

226, 227, 229, 230, 231, 232, 233, 235, 240, 241,

247, 248, 252, 253, 254, 255, 256, 258, 260, 261,

262

random mating, 237, 238

ratepayers, 86

raw materials, 143

reaction temperature, 40

reactions, x, 16

real time, vii, 1, 20, 156

reality, 51, 104, 113

receptacle, 225

receptors, 16

recognition, 31

recombination, 239, 244

reconstruction, 259

recovery, x, 52, 117, 161, 191, 259

recreational, 198

recycling, 37, 99, 115, 133, 143

redundancy, 198, 202, 203

refrigerant, vii, viii, 23, 25, 26, 49, 51, 53, 56, 57, 58,

60, 61, 62, 66, 68, 69, 70, 71, 75, 76, 81, 83, 84,

85, 87, 88, 89, 92, 94, 96, 99

regenerate, 202

regeneration, 5, 36, 38, 145, 146, 199, 200, 204, 207,

223

regions of the world, 106, 129

regression, 157

regulations, 38, 111

rejection, 29, 30, 53, 56, 65, 73

relative humidity, xi, 106, 129, 175, 331, 332, 334,

341, 342

relevance, 170, 176

reliability, 14, 26

remodelling, 115

renewable energy, viii, 24, 27, 29, 31, 32, 73, 83, 95,

96, 103, 104, 105, 112, 116, 117, 118, 122, 128

renewable energy technologies, viii, 30, 103, 106,

128

repair, 31, 32, 127, 244

replication, 244

requirements, 2, 57, 68, 90, 104, 105, 109, 116, 135,

147, 198, 206, 217, 235

research institutions, 122

researchers, 167, 245

residues, 118, 143


Index 358

resilience, vii, x, 195, 196, 197, 198, 199, 200, 201,

202, 203, 204, 205, 206, 207, 208, 209, 210, 211,

256

resistance, 28

resolution, 17, 204, 235, 240, 245, 249

resource management, 133, 209

resources, vii, viii, 31, 32, 70, 83, 103, 104, 106,

112, 114, 115, 116, 128, 132, 134, 163, 167, 171,

185, 196, 198, 205

respiration, 200

response, x, 6, 14, 17, 18, 28, 112, 164, 198, 202,

203, 204, 213, 214, 216, 217, 221, 226, 229, 230,

232, 235, 236, 241, 247, 248, 252, 253, 255, 256,

257, 258

response time, 17, 18, 28

responsiveness, 117

restoration, 197, 256

RETs, 106, 128

revenue, 171

reverse osmosis, 145

rings, 4, 14, 225

risk(s), 26, 51, 114, 143, 147, 172, 181, 188, 189,

190, 204, 205, 252, 253

risk assessment, 188

room temperature, 4, 39, 92

root(s), 153, 155, 157

roughness, 29

routes, 222, 229, 245, 263

rowing, 225

Royal Society, 258, 259, 260

rules, ix, 105, 166, 167, 173, 187

rural areas, viii, 103, 108, 121, 133

rural population, 121

Russia, 172, 249

S

SAB, 125

safety, 26, 31, 75, 87, 108

saline water, 138

salinity, x, 107, 138, 162

salt concentration, 141, 148, 149, 154, 156, 157, 158,

159

salts, 142, 145, 161, 163

SAP, 144

saturation, viii, 24, 37, 56, 59, 84, 158

savings, viii, 24, 27, 31, 83, 86, 91, 93, 95, 98, 110,

111, 112, 158

scale system, 20

scaling, 201

science, x, 2, 51, 111, 113, 167, 187, 189, 191, 213,

246

scientific knowledge, 37

scope, 201

sea level, 52, 166, 255

sea-level rise, 255

seasonality, 213, 242, 255

second generation, 100

second virial coefficient, 47

security, 28, 72, 75, 171, 174, 178, 192

sediment(s), 231, 234, 235, 261

sedimentary records, 217

sedimentation, 200

seed, 38, 125, 200, 202, 208, 217, 222, 223, 225,

226, 244, 256, 259

seedlings, 217, 223, 225

segregation, 38

senescence, 199, 214

senses, 18

sensing, 17

sensitivity, 3, 20, 167, 168, 170, 172, 173, 174, 176,

181, 182, 184, 185, 186, 187, 252

sensors, vii, xi, 3, 16, 17, 20, 87

sequencing, 240

services, 31, 32, 69, 104, 106, 132, 171, 172, 174,

184, 189, 190, 192, 196, 198

set theory, ix, 166, 172, 187, 188

settlements, 104, 114

sewage, 37, 109

shade, 27, 83, 217, 218, 220, 221, 222, 223, 226,

227, 261

shape, 16, 40, 76, 108, 112, 171, 196, 200, 246

sheep, 138

shelter, 199

shock, 139

shoot(s), 218, 220, 261

shortage, 116

showing, 233, 234, 248, 251

shrubs, 94, 218

signals, 11, 204

signs, 143, 247

simulation(s), 49, 77, 78, 111, 116

siphon, 10

skin, 140, 161, 162

slag, 37

sludge, 37, 51, 108, 143

small communities, 108

social capital, 170, 171

social expenditure, 171

social services, 171

society, 104, 167, 168, 170

sodium, viii, 3, 13, 14, 17, 137, 138, 141, 142, 143,

144, 146, 147, 148, 149, 150, 151, 153, 154, 156,

159, 162


Index 359

sodium chloride, viii, 13, 137, 138, 141, 143, 146,

147, 148, 149, 150, 151, 153, 154, 156, 159, 162,

276, 304, 306, 317

sodium hydroxide, 17

software, 8, 12, 18, 173

soil erosion, 174, 200

soil pollution, 37

soil type, 96, 245

solar collectors, 117

solid phase, 41, 84, 145, 152, 155, 156, 157, 158

solid waste, 52, 138, 146, 164

solubility, 7, 9, 68, 143

solution, vii, ix, xi, 1, 2, 3, 4, 6, 9, 10, 11, 12, 13, 14,

16, 17, 18, 19, 23, 25, 33, 36, 38, 40, 68, 69, 70,

75, 76, 81, 82, 83, 96, 98, 99, 106, 110, 129, 138,

141, 142, 143, 144, 145, 146, 147, 148, 149, 150,

158, 160, 163

solvent stream, vii, 1, 2

solvents, 52

sorption, 41, 66, 139, 146

sorption process, 139

South Africa, 39, 143, 219, 257

South America, 218, 220, 222, 223, 261

Southeast Asia, 190, 219, 221, 222, 232

speciation, 236, 237, 248, 258

species richness, 209

specific heat, 24, 28, 29, 33, 49, 53

specifications, 92, 125

spending, 188

spontaneity, 42

stability, 36, 43, 53, 69, 75, 195, 196, 198, 199, 203,

204, 205, 208, 209, 211

stabilization, 140

stable states, 197, 207

stakeholders, 198, 205

stamens, 218

standard deviation, 17

standard of living, 31, 32, 171

standardization, 176

starch, 159

state(s), 19, 33, 40, 44, 47, 49, 51, 53, 56, 58, 64, 78,

81, 82, 84, 94, 132, 141, 144, 146, 163, 164, 166,

168, 171, 186, 191, 196, 197, 199, 201, 203, 204,

206, 207, 237

statistics, 189, 190, 238, 247

steel, 4

stock, 12, 39, 116, 144, 256

stomata, 222

storage, 38, 95, 96, 116, 126, 127, 135, 138, 142

storms, 52, 166

strategic planning, 132

stress, 72, 109, 172, 186, 235, 250

stressors, 167, 190

structural changes, 140

structure, x, 97, 104, 108, 115, 140, 142, 168, 170,

173, 176, 181, 196, 200, 201, 202, 206, 208, 210,

213, 214, 215, 217, 219, 226, 227, 228, 231, 235,

236, 237, 240, 244, 245, 248, 255, 258, 259

style, 110, 136

styrene, 17

subcutaneous tissue, 141, 146

subjective well-being, 189

subsistence, 174

substitution, 142, 144, 153

substrate, 147

succession, 200, 207, 208

Sudan, viii, 103, 104, 106, 107, 118, 120, 124, 126,

127, 132, 133, 134

sulfate, xi, 143

sulfuric acid, 144

sulphur, 52

supervision, 99

supplier(s), viii, 24, 95, 171

surface area, 24, 28, 36, 43, 99, 227

surface tension, 69

surfactant, 13

surfactants, 144

survival, 216, 231, 252, 256

susceptibility, 172, 188, 191, 232

sustainability, 104, 106, 114, 115, 116, 128, 188,

191, 196

sustainable development, 31, 32, 105, 101, 114, 128,

188

sustainable energy, 122

swelling, 100, 140, 146

syndrome, 134

synthesis, 2, 36, 40, 43, 168, 186, 204, 210, 211,

237, 260

T

tandem repeats, 244

target, 5, 16, 18, 104, 113, 256

tariff, viii, 24, 95

taxa, 214, 216, 219, 220, 229, 230, 231, 232, 233,

235, 245, 260

taxes, 171

taxonomy, 217, 255, 257

TBP, 6

techniques, x, 4, 20, 28, 32, 73, 75, 110, 115, 116,

133, 134, 144, 199, 217, 227, 239, 245

technology transfer, 122

tenants, 99

tensile strength, 140

tension, 69

terminals, 92


Index 360

terrestrial ecosystems, 210

testing, 2, 5, 8, 11, 19, 107, 134

tetrad, 225

texture, 139, 140, 225

TFE, 52

Thermal desalination, x, 267, 268

thermal energy, 46, 63, 64, 108

thermodynamic equilibrium, 47

thermodynamic parameters, 42

thermodynamic properties, vii, 49, 50, 68

thermodynamics, 40, 43, 45, 46, 64, 101

threats, 246, 262

time periods, 138, 160, 240, 241, 254

tissue, 141, 146, 250

total energy, x, 104, 132, 171

toxicity, 144

tracks, 258

trade, ix, 103, 165, 167, 168, 181, 184, 187, 250

trade-off, ix, 165, 167, 168, 181, 184, 187, 250

traditions, 113

training, viii, 103, 119, 122, 133

traits, 204, 213, 227, 236, 238, 250, 255, 258

trajectory, vii, 23, 69, 70

transformation, 40, 150, 216

transformations, 40

transparency, ix, 166, 167, 172, 174, 187

transpiration, 200

transport, vii, viii, 1, 2, 20, 38, 51, 98, 101, 105, 109,

116, 117, 132, 137, 139, 140, 145, 146, 156, 157,

160, 163, 171, 235, 262

transport processes, ix, 137

transportation, 105, 109, 138, 141, 142, 144

treatment, 104, 143, 144, 145, 147, 151, 159, 163

tributyl phosphate, 6

tropical forests, 232

Turkey, 134, 160, 161

turnover, 116, 216, 227, 232, 248, 252, 255

twist, 162

U

ultrasound, 147

uniform, 17, 29, 40, 44, 201, 229, 240

universal gas constant, 47, 156

universities, 184

urban areas, 108, 109, 116

urban densities, 105

urbanisation, viii, 103, 104

urbanization, 181

USDA, 160

UV, 17

V

vacuum, 2, 4, 6, 10, 14, 18, 75

valence, xi

validation, 3, 5, 13, 229

valuation, 135

valve, 4, 10, 11, 14, 27, 29, 30, 31, 57, 58

vapor, vii, 1, 2, 3, 4, 5, 6, 8, 9, 11, 12, 13, 14, 20

vapour absorption, vii, 23, 69, 70

variables, xi, 149, 152, 153, 170, 173, 174, 176, 181,

245, 250

variations, 167, 171

varieties, 93

vascular bundle, 219

vegetable oil, 147

vegetation, 113, 202, 207, 208, 214, 215, 216, 229,

230, 232, 234, 235, 258, 260, 261

vehicles, 109

vein, 218

velocity, 43, 92, 108

ventilation, 31, 32, 73, 109, 111, 112, 114, 116, 117,

134, 135

vertebrates, 262

Vietnam, 188

viscoelastic properties, 160

viscosity, 69

visualization, xi

vitamin C, 225

volatility, 2, 4, 5, 8, 9, 75

vulnerability, ix, 165, 166, 167, 168, 170, 172, 174,

176, 181, 184, 185, 186, 187, 188, 189, 190, 191,

206

vulnerable people, 167, 186

W

Wales, 225, 227, 253

walking, 117

warning systems, 171

Washington, 188, 189, 208, 258, 260

waste, 4, 8, 10, 27, 36, 37, 52, 64, 70, 83, 87, 97,

108, 114, 115, 138, 141, 143, 144, 146, 147, 160,

164

waste disposal, 52

waste heat, 27, 64, 97

waste management, 164

waste water, 138, 143, 144

wastewater, 36, 43, 51, 161

water absorption, 78, 83

water heater, 96

water quality, 51

water resources, 171, 185


Index 361

water supplies, 107

wavelengths, 73

weakness, 250

wealth, 133, 170, 171

wear, 89

web, 181, 188

well-being, 171, 189, 196

wells, 96

West Africa, 190

Western Australia, 235

Western Europe, 39

wetting, 13, 14

white oak, 263

wildlife, 201

wind power, 118, 124, 134

wind speed(s), 107, 118, 120, 124, 125, 126, 131,

132

windows, 89, 91, 94, 115

wood, 107, 118, 121, 225

woodland, 200

World Bank, 188, 189, 191

World Health Organization, 188

worldwide, viii, 26, 73, 83, 135, 137, 138, 148, 159,

256

X

xylem, 250, 258

Y

Yemen, 134

yield, xi, 49

Z

zinc, 36, 142


Documents

8 Advances in Environmental Research Vol 32 - Reyes & Kneeshaw 2014