Dams, Ecosystem Functions and Environmental Restoration · • Lotek Engineering • Manitoba Hydro • National Wildlife Federation, ... Reviews I.1 Social Impacts of Large Dams

WCD Thematic Review Environmental Issues II.1

Dams, Ecosystem Functions and Environmental Restoration

Final Version: November 2000

Prepared for the World Commission on Dams (WCD) by:

Ger Bergkamp, Matthew McCartney, Pat Dugan, Jeff McNeely and

Mike Acreman

Based on contributions from:

M.C. Acreman (Institute of Hydrology, UK) E. Barbier (University of York, UK)

G. Bernacsek (FAO) M. Birley (University of Liverpool, UK) J.R. Bizer (MESAS Consultants, USA)

C. Brown (University of Cape Town, South Africa) K Campbell (Natural Resources Institute, UK)

J. Craig (Consultant, UK) N. Davidson (Wetlands International, The Netherlands);

S. Delany (Wetlands International, The Netherlands) C. Di Leva (IUCN Environmental Law Center, Germany)

F. Farquharson (Institute of Hydrology, UK) N Hodgson (Natural Resources Institute, UK)

D.C. Jackson (Mississippi State University, USA); J. King (University of Cape Town, South Africa)

M. Larinier (Institut de Mecanique des Fluides, France) J. Lazenby (Gibb Ltd, UK )

D.E. McAllister, (Ocean Voice International, Canada); G. Marmulla, (Fisheries Department, FAO)

M.P. McCartney (Institute of Hydrology, UK) J. Morton (Natural Resources Institute, UK)

D. Murray (OPIRG, Carleton University, UK) M.B. Seddon (National Museum of Wales, UK)

L. Sklar (University of California, Berkeley, USA); D. Smith (Natural Resources Institute, UK)

C. Sullivan (Institute of Hydrology, UK) R. Tharme (University of Cape Town, South Africa)

Secretariat of the World Commission on Dams P.O. Box 16002, Vlaeberg, Cape Town 8018, South Africa

Phone: 27 21 426 4000 Fax: 27 21 426 0036. Website: http://www.dams.org E-mail: [email protected]

Dams, Ecosystem Functions, and Environmental Restoration i

This is a working paper prepared for the World Commission on Dams as part of its information gathering activities. The views, conclusions, and recommendations contained in the working paper are not to be taken to represent the views of the Commission

Disclaimer This is a working paper of the World Commission on Dams - the report published herein was prepared for the Commission as part of its information gathering activity. The views, conclusions, and recommendations are not intended to represent the views of the Commission. The Commission's views, conclusions, and recommendations will be set forth in the Commission's own report. This manuscript has been compiled by current and former staff members of IUCN in their personal capacity, based on contributions from a wide range of sources, and comments received from the review panel and WCD Forum. It does not therefore represent any official IUCN policy. Please cite this report as follows: Berkamp, G., McCartney, M., Dugan, P., McNeely, J., Acreman, M. 2000. Dams, Ecosystem Functions and Environmental Restoration Thematic Review II.1 prepared as an input to the World Commission on Dams, Cape Town, www.dams.org The WCD Knowledge Base

This report is one component of the World Commission on Dams knowledge base from which the WCD drew to finalize its report �Dams and Development-A New Framework for Decision Making�. The knowledge base consists of seven case studies, two country studies, one briefing paper, seventeen thematic reviews of five sectors, a cross check survey of 125 dams, four regional consultations and nearly 1000 topic-related submissions. All the reports listed below, are available on CD-ROM or can be downloaded from www.dams.org

Case Studies (Focal Dams) • Grand Coulee Dam, Columbia River Basin, USA • Tarbela Dam, Indus River Basin, Pakistan • Aslantas Dam, Ceyhan River Basin, Turkey • Kariba Dam, Zambezi River, Zambia/Zimbabwe • Tucurui Dam, Tocantins River, Brazil • Pak Mun Dam, Mun-Mekong River Basin,

Thailand • Glomma and Laagen Basin, Norway • Pilot Study of the Gariep and Van der Kloof

dams- Orange River South Africa

Country Studies • India • China

Briefing Paper • Russia and NIS

countries

Thematic Reviews • TR I.1: Social Impact of Large Dams: Equity

and Distributional Issues • TR I.2: Dams, Indigenous People and Vulnerable

Ethnic Minorities • TR I.3: Displacement, Resettlement,

Rehabilitation, Reparation and Development • TR II.1: Dams, Ecosystem Functions and

Environmental Restoration • TRII.1: Dams, Ecosystem Functions and

Environmental Restoration • TR II.2: Dams and Global Change • TR III.1: Economic, Financial and

Distributional Analysis • TR III.2: International Trends in Project

Financing

• TR IV.1: Electricity Supply and Demand Management Options

• TR IV.2: Irrigation Options • TR IV.3: Water Supply Options • TR IV.4: Flood Control and Management

Options • TR IV.5: Operation, Monitoring and

Decommissioning of Dams • TR V.1: Planning Approaches • TR V.2: Environmental and Social Assessment

for Large Dams • TR V.3: River Basins � Institutional Frameworks

and Management Options • TR V.4: Regulation, Compliance and

Implementation • TR V.5: Participation, Negotiation and Conflict

Management: Large Dam Projects

• Regional Consultations � Hanoi, Colombo, Sao Paulo and Cairo • Cross-check Survey of 125 dams

Dams, Ecosystem Functions, and Environmental Restoration ii


Acknowledgment The WCD acknowledges contributions to this thematic review by the United Nations Environment Programme (UNEP) and the IUCN - The World Conservation Union - through their respective work programmes. The submissions to the WCD thematic review were supported by the partnership agreement between United Nations Foundation, UNEP and the WCD.

Dams, Ecosystem Functions, and Environmental Restoration iii


Financial and in-kind Contributors Financial and in-kind support for the WCD process was received from 54 contributors including governments, international agencies, the private sector, NGOs and various foundations. According to the mandate of the Commission, all funds received were �untied�-i.e. these funds were provided with no conditions attached to them. • ABB

• ADB - Asian Development Bank

• AID - Assistance for India's Development

• Atlas Copco

• Australia - AusAID

• Berne Declaration

• British Dam Society

• Canada - CIDA

• Carnegie Foundation

• Coyne et Bellier

• C.S. Mott Foundation

• Denmark - Ministry of Foreign Affairs

• EDF - Electricité de France

• Engevix

• ENRON International

• Finland - Ministry of Foreign Affairs

• Germany - BMZ: Federal Ministry for Economic

Co-operation

• Goldman Environmental Foundation

• GTZ - Deutsche Geschellschaft für Technische

Zusammenarbeit

• Halcrow Water

• Harza Engineering

• Hydro Quebec

• Novib

• David and Lucille Packard Foundation

• Paul Rizzo and Associates

• People's Republic of China

• Rockefeller Brothers Foundation

• Skanska

• SNC Lavalin

• South Africa - Ministry of Water Affairs and

Forestry

• Statkraft

• Sweden - Sida

• IADB - Inter-American Development Bank

• Ireland - Ministry of Foreign Affairs

• IUCN - The World Conservation Union

• Japan - Ministry of Foreign Affairs

• KfW - Kredietanstalt für Wiederaufbau

• Lahmeyer International

• Lotek Engineering

• Manitoba Hydro

• National Wildlife Federation, USA

• Norplan

• Norway - Ministry of Foreign Affairs

• Switzerland - SDC

• The Netherlands - Ministry of Foreign Affairs

• The World Bank

• Tractebel Engineering

• United Kingdom - DFID

• UNEP - United Nations Environment

Programme

• United Nations Foundation

• USA Bureau of Reclamation

• Voith Siemens

• Worley International

• WWF International

Dams, Ecosystem Functions, and Environmental Restoration iv


Executive Summary Introduction. The impact of dams upon natural ecosystems and biodiversity has been one of the principal concerns raised by large dams. Over the course of the past 10 years in particular, considerable investments have been made in the development of measures to alleviate these impacts. Yet today widespread concern remains that despite improvements in dam planning, design, construction and operation, they continue to result in significant negative impacts to a wide range of natural ecosystems and to the people that depend upon them for their livelihood. WCD Thematic Reviews I.1 Social Impacts of Large Dams Equity and Distributional Issues, I.2 Dams, Indigenous People and vulnerable ethnic minorities and I.3 Displacement, Resettlement, rehabilitation, reparation and development examine this complex set of issues. It does so by first reviewing the importance of natural river basin ecosystems and examining the impact of dams on these ecosystems. It then examines the current status of approaches being taken to addressing these impacts through the continuum of �avoidance-mitigation-compensation-restoration�. Based upon this analysis the report concludes with an assessment of the areas of convergence and divergence on these issues within the dams debate and provides a set of recommendations to the Commission. River Basin Ecosystems and Biodiversity. Each river basin contains many natural ecosystems including not only the aquatic habitats associated with water in the river channel, but all of the elements of the river catchment that contribute water, nutrients and other inputs to the river. These ecosystems include: the headwaters and the catchment landscapes; the channel from the headwaters to the sea; riparian areas; associated groundwater in the channel/banks and floodplains; wetlands; the estuary and any near shore environment that is dependent on freshwater inputs. These ecosystems perform functions such as flood control and storm protection, yield products such as wildlife, fisheries and forest resources, and are of aesthetic and cultural importance to many millions of people. The total global value of ecosystem goods and services is estimated at US$ 33 trillion per year of which roughly 25% relates directly to freshwater ecosystems. With widespread and still growing recognition of these ecosystem values, river basin development needs to determine how much water is required for the maintenance of ecosystems to provide environmental goods and services, and how much water should be used to support agriculture, industry and domestic services. Ecosystem Impacts of Large Dams. The current state of knowledge indicates that the impacts of dams on ecosystems are profound, complex, varied, multiple and mostly negative. By storing or diverting water dams alter the natural distribution and timing of stream flows. This in turn changes sediment and nutrient regimes and alters water temperature and chemistry, with consequent ecological and economic impacts. Reduction in downstream annual flooding in particular affects the natural productivity of floodplains and deltas. These ecosystem impacts result in a significant impact of dams on freshwater biodiversity, which is already under special threat. Global estimates of endangered freshwater fish reach 30% of the known species. And in North America detailed studies indicate that dam construction is one of the major causes of freshwater species extinction. Dramatic reductions in bird species are also known, especially in downstream floodplain and delta areas. Some reservoirs also provide habitats for birds and other fauna but this often does not outweigh the loss of habitat downstream. Multiple dams on a river significantly aggravate the impact on ecosystems. Sediment entrapment can reach 99% if a cascade of dams is developed. Fish migration is affected even by a single dam, and multiple dams worsen this situation dramatically. In the Northern hemisphere 77% of the largest rivers are affected by dams and on many rivers fully natural reaches are restricted to headwaters. The global impacts of dams on the global water cycle are increasingly recognised. The review highlights the complexity of the processes that occur when a dam impacts an ecosystem. It is therefore extremely difficult and rarely possible to predict in precise detail the magnitude and

Dams, Ecosystem Functions, and Environmental Restoration v


nature of impacts arising from the construction of a dam or a series of dams. The precise impact of any single dam is unique and dependent not only on the dam structure and its operation, but also upon local hydrology, fluvial processes, sediment supplies, geomorphic constraints, climate, and the key attributes of the local biota. There is therefore no normative or standard approach to address ecosystem impacts and these have to be looked at on a case-by-case basis. In addition the acceptability of ecosystem changes will vary with the nature of human societies, cultures, and expectations. The Economic and Social Implications of Ecosystem Impacts. Because natural ecosystems fulfil functions and yield a range of services that are of substantial economic and cultural value to society, the ecosystem changes that result from the creation of dams lead in turn to substantial economic and social impacts. Entire communities depend on the functions provided by freshwater wetlands, yet it is still difficult to translate the value into monetary terms. As a result the value of ecosystem functions is not properly accounted for in conventional market economics, and the value of these functions and the cost of their loss, is excluded from the economic decision-making process. This externalisation of costs is a major factor leading to the loss of natural ecosystems. By reducing or eliminating access to resources flooded by the reservoir, through degradation and loss of agricultural and grazing resources on downstream floodplains, and through loss of riverine and coastal fisheries dependent upon the river flood, many dams have very high external costs. Policy-makers need to identify the value of this loss of welfare and implement financial and institutional mechanisms to assimilate these costs into the accounting structure. The review stresses however that, even when these steps are taken, the valuation of ecosystems and the consideration of development options is not a straightforward accounting exercise. It needs to be recognised that not all ecosystem values can be expressed in economic terms. Ethical and societal considerations also need to be included. The monetary value serves as an input to multi-criteria decision-making and raises awareness of costs that are currently hidden and negated in the accounting exercise. Responding to the Ecosystem Impacts of Dams. There are four principal categories of measures that may be incorporated into dam design or operating regime in order to respond to the environmental impacts identified through an EIA. These are: i) measures that avoid anticipated adverse effects of a dam; ii) mitigation measures that are incorporated into a new or existing dam design or operating regime in order to eliminate, offset or reduce ecosystem impacts to acceptable levels; iii) measures that compensate for existing or anticipated adverse effects that cannot be avoided or mitigated; iv) de-commissioning of the dam and restoration of the riverine ecosystem. Within this framework of avoidance, mitigation, compensation and restoration, there are a wide range of specific measures that can be taken appropriate to specific circumstances of each dam. The Thematic Review evaluates experience in each approach and reveals that the most widely used approach, mitigation, is problematic. It concludes that there are always residual impacts that cannot be mitigated, simply by the nature of the dam�s impact on ecosystems themselves. Whether these impacts are significant varies from case to case. While there is experience of good mitigation, this success is nevertheless contingent upon stringent conditions of: • a good information base and competent professional staff available to formulate complex choices

for decision-makers; • an adequate legal framework and compliance mechanisms; • a co-operative process with the design team and stakeholders; • monitoring of feedback and evaluation of mitigation effectiveness, and • adequate financial and institutional resources;

Dams, Ecosystem Functions, and Environmental Restoration vi


If any one of these conditions is absent, then the ecosystem values are likely to be lost. In practice the extent to which these conditions are met varies enormously from country to country and dam to dam. The review therefore concludes that mitigation, though often possible in principle, has many uncertainties attached to it in field situations and is therefore at present not a credible option in all cases and all circumstances. In addition the weaknesses of the EIA process for many projects (cf Thematic Review V.2) reduce the possibilities for positive outcomes. This would tend to encourage a strategy of avoidance and minimisation rather than one of mitigation if the aim is to maintain biodiversity and ecosystem functions and services for the foreseeable future. Alternative tools for maintaining ecosystem health therefore need to be pursued. The review argues that improved scientific predictive capacity and improved institutional and human capacity will take several decades. In the short term therefore focused attention needs to be given to the development and application of effective tools that can allow environmentally sound development of river water resources and the management of dams within this context. Three such tools are described: i) Indicators for Hydro-project selection; ii) Indicators of Ecological Integrity; iii) Environmental Flow Requirements. Trends in the International Debate/Approach to Dams. The Thematic Review examines current trends in the international debate over dams and their environmental impacts. It concludes that considerable steps have been taken to address the environmental concerns and that there are today many areas of broad agreement between those who are generally supportive of building dams and those who are generally philosophically opposed to large dams. However differences remain. At the most general level these differences concentrate on the value systems adhered to by the different groups involved and especially the value to be attached to the intrinsic value of nature. This highlights the importance of ensuring that project approval be based on multi-criteria decision-making, not just economic cost-benefits analyses or on a purely eco-centric view of the world. Techniques also need to be improved to offer better methods of economic valuation that are acceptable to both proponents and opponents of dams. Clearer guidelines on how costs and benefits can be distributed among those people affected by a dam may necessitate the establishment of appropriate institutions to promote equitable water use, especially between upstream and downstream ecosystems and livelihoods. The Review argues that most success in bridging the differences outlined is likely to be made by strengthening options assessment and the evaluation of the true cost and benefits of projects for the short and medium term. Discrepancies are likely to remain on value systems and development paradigms for decades to come. Therefore efforts to deal with environmental impacts of dams should concentrate on developing legitimate and accepted processes for dam planning, design and management within the river basin context. Secondly, much effort could be invested in improving the economic tools for analysis and improving incentives for better dam design and operation. Policy Recommendations. The review concludes by providing ten policy recommendations to the WCD.

Dams, Ecosystem Functions, and Environmental Restoration vii


Table of Contents List of Figures............................................................................................................................x

List of Tables ............................................................................................................................xi

List of Boxes............................................................................................................................ xii

1. Introduction .....................................................................................................................1

2. River Basin Ecosystems and Biodiversity .....................................................................3 2.1 Introduction ...................................................................................................................................3 2.2 Why Ecosystems are Valuable ......................................................................................................4 2.2.1 Ecosystems as Regulators..............................................................................................................5 2.2.2 Ecosystems as Habitats..................................................................................................................6 2.2.3 Ecosystems as Providers of Resources ..........................................................................................7 2.2.4 Ecosystems as Providers of Information .......................................................................................8 2.3 The value of ecosystem goods and services ..................................................................................8 2.3.1 Economic valuation techniques .....................................................................................................8 2.3.2 The Monetary Value of Freshwater Ecosystems ...........................................................................9 2.4 Ecosystems and River Basin Development .................................................................................10 2.5 International and national recognition of ecosystem values ........................................................11 2.6 Conclusions .................................................................................................................................13

3. Ecosystem Impacts of Large Dams ..............................................................................14 3.1 Introduction .................................................................................................................................14 3.2 Scale and Variability of Impacts..................................................................................................14 3.3 Framework for Analysis ..............................................................................................................18 3.4 Information Constraints ...............................................................................................................20 3.5 Upstream Impacts ........................................................................................................................21 3.5.1 First-Order Impacts on Key Parameters ......................................................................................21 3.5.2 Second Order Impacts � Changes in Primary Production ...........................................................23 3.5.3 Third-Order Impacts on Fauna ....................................................................................................25 3.6 Downstream Impacts on Rivers, Floodplains and Deltas ............................................................26 3.6.1 First-Order Impacts on Ecosystem Driving Variables.................................................................27 3.6.2 Second Order Impacts on Primary Production ............................................................................33 3.7 Third-Order Impacts on Fauna ....................................................................................................36 3.7.1 Freshwater Species Diversity Changes........................................................................................36 3.7.2 Bivalve and Gastropod Molluscs.................................................................................................40 3.7.3 Impact of Dams on Fish Diversity...............................................................................................42 3.7.4 Dams and Waterbirds ..................................................................................................................44 3.8 Cumulative Impacts of Dams ......................................................................................................46 3.8.1 Conceptual Framework for cumulative impact assessment.........................................................47 3.8.2 Case studies on cumulative impacts ............................................................................................48 3.9 Estimating the Costs of the Impacts of Dams on Ecosystems .....................................................51 3.9.1 Externalities and Livelihoods ......................................................................................................52 3.9.2 Trade-offs between Economic and Ethical Considerations .........................................................52 3.10 Conclusions .................................................................................................................................53

4. Responding to the Ecosystem Impacts of Dams .........................................................55

Dams, Ecosystem Functions, and Environmental Restoration viii


4.1 Introduction .................................................................................................................................55 4.2 Types of Response.......................................................................................................................55 4.2.1 Avoidance....................................................................................................................................56 4.2.2 Mitigation ....................................................................................................................................58 4.2.3 Compensation ..............................................................................................................................61 4.2.4 Dam Decommissioning and River Restoration ...........................................................................62 4.3 How Effective is Mitigation?.......................................................................................................64 4.3.1 The Example of Fish Ladders......................................................................................................66 4.3.2 Why Avoidance, Mitigation, and Compensation are Difficult ....................................................68 4.4 How to Make Mitigation More Effective? ..................................................................................69 4.4.1 Indicators for Hydro-Project Site Selection.................................................................................70 4.4.2 Indicators of Ecological Integrity ................................................................................................71 4.4.3 Environmental Flow Requirements (EFRs).................................................................................72 4.5 Conclusions .................................................................................................................................73

5. Trends in the International Debate/Approach to Dams ............................................75 5.1 Introduction .................................................................................................................................75 5.2 Summary of the debate ................................................................................................................75 5.3 Summary of Trends .....................................................................................................................76 5.3.1 IEA 76 5.3.2 International Commission on Large Dams (ICOLD) ..................................................................78 5.3.3 The World Bank ..........................................................................................................................79 5.3.4 New approaches of the Organisation for Economic Co-operation and Development (OECD) ..80 5.3.5 The International Movement Against Large Dams .....................................................................80 5.3.6 Requirements of International Conventions ................................................................................81 5.4 Areas of Convergence/Divergence ..............................................................................................82

6. Conclusions and Policy Recommendations for WCD ................................................84 6.1 Conclusions .................................................................................................................................84 6.2 Recommendations........................................................................................................................85 6.3 Options for Operationalising the Recommendations...................................................................86

7. References ......................................................................................................................89

Annex 1: Potential Environmental Impacts of Dams, Reservoirs and Hydroelectric Projects .........................................................................................................................100

Annex 2: Reservoir Fisheries ..............................................................................................104

Annex 3: Comparison of Pre vs. Post Impoundment Conditions.....................................107

Annex 4: Sediment Discharges ...........................................................................................108

Annex 5: Large Dam Projects: Adverse Environmental Impacts and Mitigation Options..........................................................................................................................111

Annex 6: Environmental Flow Requirements (EFR) ......................................................114

Annex 7: Example of Mitigation Measures ......................................................................116

Dams, Ecosystem Functions, and Environmental Restoration ix


Appendix I � List of Contributing Papers to Thematic Review II.1 ...............................117

Appendix II - Submissions for Thematic Review II.1 .......................................................119

Appendix III � Comments Received for Thematic Review II.1 Dams, Ecosystem Functions & Environmental Restoration ..................................................................129

Dams, Ecosystem Functions, and Environmental Restoration x


List of Figures Figure 2.1: General description of the inter-relationships between a river basin and the water cycle

(after Shiklomanov 1999). ..............................................................................................................4 Figure 2.2: Cumulative number of countries ratifying the main environmental Conventions(World

Heritage (1), Ramsar Convention on wetlands (2), trade in endangered species (3), biological diversity (4), and climate change (5)). ..........................................................................................12

Figure 3.1: Distribution of reservoir area for dams over 15m high in Turkey(DSI, 1999)..................15 Figure 3.2: Total area of large-dam reservoirs (1000's km2) by region (ICOLD 1999).....................16 Figure 3.3: Average area of large dam reservoir (km2) per region(ICOLD 1999)..............................16 Figure 3.4: Comparison of pre and post impoundment flows in the Murray River, Australia:variation

in the average monthly flow at a) Albury (2225 km from the mouth) and b) at Barrages (1 km from the mouth). Source: Murray-Darling Basin Ministerial Council, 1995...............................28

Figure 3.5: Daily Streamflow Variations in the Colorado River at Lee's Ferry in September. Peak flows are associated with the power generation between 14.00 and 19.00 daily, with minima at 04.00 am, and the fluctuation in demand also varies from day to day..........................................29

Figure 3.6: Fish species richness decreases at higher latitudes indicating that dam construction in tropical regions could potentially have more impacts than at higher latitudes (WCMC 1998)...36

Figure 3.7: Fragmentation of rivers in 225 basins in the world (Source: Nilsson et al. 2000). ..........46 Figure 3.8: Dams in the river systems of Sweden. Only four major rivers remain undammed...........47 Figure 3.9: The impact of dams on the hypothesised downstream pattern exhibited by a given river

characteristic in situationa) with no cumulative impacts and b) with cumulative impacts (modified from Ward and Stanford, 1995). ..................................................................................48

Figure 3.10: a) Cumulative useable storage in reservoirs in the Platte River basin and b) associated cumulative change in island and channel area ..............................................................................50

Figure 3.11: a) Longitudinal and altitudinal profile of the Gunnison River and b) changes in speciosity and biomass with distance downstream. .......................................................................................51

Figure 4.1: Number of dams removed in the USA, as a function of a) dam height and b) year of removal (after, Doyle et al., 2000)...............................................................................................63

Figure 4.2: Figure 4.1: Distribution of EMP preparation & environmental problem evaluations for dam projects in Latin America co-financed by the IDB from 1960-1999(IDB 1999)..................65

Dams, Ecosystem Functions, and Environmental Restoration xi


List of Tables Table 2.1: Natural ecosystems provide many goods and services (functions) to humankind that are

often neglected in (economic) planning and decision making. .......................................................6 Table 2.2: Global monetary values of freshwater and wetland functions (in US$ billion, 1994)..........9 Table 3.1: The potential scale of the impacts of dams.........................................................................16 Table 3.2: Some databases of dams .....................................................................................................17 Table 3.3: Upstream and downstream impacts according to first, second, and third order as described.

......................................................................................................................................................18 Table 3.4: A framework for assessing the impact of dams on river ecosystems(modified from Petts,

1984). ............................................................................................................................................19 Table 3.5: Fragmentation of rivers in 225 basins in the world (Source: Nilsson et al. 2000).............46 Table 5.1: Distillation of arguments used by proponents and opponents of large dams......................77 Table 5.2: Trends in the Planning of Hydropower Projects (IEA, 2000)............................................78 Table 6.1: Options for establishing sub-principles under each recommendation ................................87

Dams, Ecosystem Functions, and Environmental Restoration xii


List of Boxes Box 1.1: Why large dams are built ........................................................................................................1 Box 2.1: Important ecological concepts of rivers and floodplains.........................................................3 Box 2.2: Why biological diversity is important .....................................................................................7 Box 2.3: Watershed Management........................................................................................................10 Box 2.4: International recognition of ecosystem values ......................................................................12 Box 3.1: Types of dams, in descending order of impacts on ecosystems ............................................15 Box 3.2: Construction impacts.............................................................................................................20 Box 3.3: Invasive species and large dams............................................................................................24 Box 3.4: Species richness of the planet�s major environments (Source: McAllister et al. 1997).......37 Environment % area of % of known Relative species ......................................................................37 Box 3.5: Global hotspots for freshwater molluscs ...............................................................................40 Box 3.6: Mollusc species present within reservoir region, USA (Source: Neves, 1999) .....................41 Box 3.7: Fish species richness in selected river basins (after: World Bank 1998, WCMC 1998).......43 Box 3.8: Dams as Wildlife Habitats.....................................................................................................45 Box 3.9: Example of cumulative affect on first order impacts (i.e. the hydrology) of the Murray

River, Australia (after Maseshwari et al., 1995)...........................................................................49 Box 3.10: Example of cumulative affect on second order impacts (i.e. the geomorphology) of the

Platte River, USA (after Hadley et al., 1987) ..............................................................................50 Box 3.11: Example of cumulative affect on third order impacts (i.e. zoobenthos) in the Gunnisson

River, USA (after Hauer et al., 1989). .........................................................................................51 Box 3.12: Ethical principles for decision-makers involved in water and energy planning (di Leva

1999) .............................................................................................................................................53 Box 4.1: Demand management, water recycling and rainwater harvesting: examples of ....................57 Box 4.2: Avoidance of impacts on sensitive species during blasting ..................................................58 Box 4.3: Decommissioning of the Edwards Dam, USA.......................................................................63 Box 4.4: Possible consequences for salmon of removal of the Elwha Dam, Washington (after Doyle

et al., 2000) ...................................................................................................................................64 Box 4.5: Improving fish passage design to make them work better ....................................................67 Box 4.6: Why fish passes may fail.......................................................................................................67 Box 4.7: Environmental Indicators To Guide Site Selection................................................................70 Box 4.8: Indicators of Ecological Integrity..........................................................................................71 Box 4.9: Case study: The Colorado River ...........................................................................................72 Box 4.10: Case study: Kromme River .................................................................................................73 Box 5.1: The Curitiba Declaration.......................................................................................................81 Box 5.2: Ramsar Convention: Guidelines for Contracting Parties relating to reducing the impact of

water development projects on wetlands ......................................................................................82 Box A2.1: Fisheries yields of selected reservoirs ..............................................................................104

Dams, Ecosystem Functions, and Environmental Restoration 1


1. Introduction Dams, large and small, are planned, constructed and operated to meet human needs in the generation of energy, irrigated agricultural production, flood control, supply of drinking water, and various other purposes. While seen originally as a relatively straightforward solution to many of these needs, the history of dams over the past 100 years has shown that their many benefits to society come together with an array of environmental and social costs. The decision to construct a dam and the design and operation of its management regime therefore need to be based upon a rigorous analysis of these costs and benefits. There is today widespread recognition of this challenge amongst governments, industry, the development assistance community, NGOs, community groups, and many others concerned with the issues of large dams. Indeed over the course of the past 10 years there has been substantial change in the approach to dams with much greater attention to environmental and social issues. Various types of environmental and social guidelines now exist and are increasingly applied. Differences in judgements towards dam development are based on different value systems, development paradigms, options analysis and practical actions. At the level of value systems, the dilemma focuses on whether �environmental conservation� and �development� are antagonistic. Within the environmental conservation movement one can distinguish �conservationists� and �preservationists� (Norton 1991). Conservationists see natural ecosystems and species as resources and are concerned mainly with the wise use of them. �Preservationists� on the other hand are committed to protecting large areas of landscape from any human alterations. In a simplified way, the development community can be divided in those seeking �development per se� and those that are looking for �sustainable development�. Obviously the preservationist and those seeking �development per se� adhere to a different development paradigm. For conservationists and those seeking sustainable development, paradigms often lie close to each other, with the debate focussing on what constitutes �acceptable� change. The majority of dams (75%) are developed and operated to irrigate land and generate power or are used for both (Box 1.1). In many cases, these dams have provided profits for a range of beneficiaries. At the same time, dams have negatively impacted affected people and the environment. As such the development of water resources using dams has created many conflicts of interest and it is becoming increasingly clear that environmental and social dimensions need to be addressed more substantially. Box 1.1: Why large dams are built Irrigation only 37% Multi-purpose 22% Electricity generation only 16% Water supply only 12% Flood control only 6% Recreation only 3% Other 4% TOTAL: 100% (Source: ICOLD World Register of Large Dams, 1998) Despite this progress there remain significant and widespread concerns about the environmental impacts of dams at the more practical level linked with option analysis and practical actions. The conservationist view in short argues that dams, even when designed to minimise environmental impacts, result in significant negative impacts to a wide range of natural ecosystems and to the people



that depend upon them for their livelihood. At a time when pressures upon the diversity and productivity of the world�s natural resources continue to rise, it is argued that firm action is required to prevent loss of these resources through further dam construction (McCully, 1996). In response, those engaged in the planning, construction and operation of dams argue that with continuing improvements in knowledge and technology it is increasingly possible to avoid, mitigate or compensate for the environmental impacts of dams, so yielding win-win solutions in most cases (ICOLD, 1997). As a contribution to improved decision-making about large dams, this report examines the nature of the effects of dams upon upstream and downstream ecosystems and the reported experience with methods to avoid, mitigate and compensate those effects. It does not attempt an exhaustive assessment of the impact of dams on ecosystems world-wide as data for this task are currently unavailable. The focus of the report is deliberately on the medium to long-term impacts of dams on ecosystems rather than the short term impacts of construction. The paper approaches these issues by first examining the nature of river basin ecosystems, asking why they are important and why the international community has signed up for promoting their protection within the framework of sustainable development. It then reviews current understanding of the nature of the impact of dams upon these ecosystems and their associated values. This includes both their intrinsic spiritual, ethical and biodiversity values and their economic values to local people and their livelihoods, as well as wider ecosystem values for society as a whole. The report then examines the current status of approaches to addressing the consequences of dam impacts on ecosystems through the continuum of �avoidance-mitigation-compensation-restoration� of ecosystem losses. Based upon this analysis the report concludes with an assessment of the areas of convergence and divergence on these issues within the dams debate and set of recommendations to the Commission.



2. River Basin Ecosystems and Biodiversity 2.1 Introduction Freshwater covers only 2.7% of the Earth�s surface, of which 66% by volume is in snow and ice, 29% in groundwater, 2.5% in lakes and rivers and less than 0.001% in reservoirs (Shiklomanov 1999). Rivers therefore cover a very small part of the Earth, yet they are intricately linked to the vast area of the planet that lies within river basins, as well as the coastal and near-shore marine ecosystems that are dependent on freshwater inputs. Rivers are central elements in many landscapes. Their string-like shape and dendritic drainage pattern mean they are effectively interspersed into the landscape despite their small total area (Nilsson and Jansson, 1995). They are important natural corridors for the flows of energy, matter and species (Malanson, 1993). Viewed holistically, river- related ecosystems encompass all the components (both biotic and abiotic) of the environment linked to that river, including people. This includes not only the aquatic habitats associated with water in the river channel, but all the elements of the river catchment that contribute water, nutrients and other inputs to the river. Thus the complex of ecosystems that constitute a river basin includes: the headwaters and the catchment landscapes; the channel from the headwaters to the sea; riparian areas; associated groundwater in the channel/banks and floodplains; wetlands; the estuary and any near shore environment that is dependent on freshwater inputs. Each of these environments is dependent to a greater or lesser degree on connectivity with the active channel of the river and the ecological character of the main channel depends on the interactions with those environments (Petts and Amoros 1996) (Box 2.1). The hydrological cycle provides an important linkage between the component parts (Figure 2.1). Another important linkage is formulated in the �flood-pulse� concept which describes the periodic, two-way exchange of nutrients between the main river channel and riparian ecosystems (Junk et al. 1989, Bayley 1995, Sparks 1995) (Box 2.1). In order to understand the relationship between large dams and the rivers on which they are built, it is essential to understand the nature and values of the different ecosystems along a river�s course from its catchment to the sea. Box 2.1: Important ecological concepts of rivers and floodplains 1. The �river continuum concept� encompasses the linkages upstream and downstream from a river�s source to the coastal zone, including any deltas or lagoon systems. This concept includes the gradual natural changes in river flows, water quality and species, that occur along the rivers length. Nutrients and sediment generated in the headwaters are recycled downstream, driving plant growth and biotic productivity. One of the most obvious characteristics of the river continuum concept is the migration of fish from the sea to spawning grounds in the headwaters. River engineering projects, such as dams, can break this continuum causing radical changes in flows, water quality and stopping the movement of species. 2. The �flood pulse� concept is based on the importance of lateral connectivity between rivers and their floodplains and sees the inundation of floodplains as the main driving force behind river life, not as a problem that needs eradicating. Rivers provide the floodplain with nutrients and sediment, whilst the floodplain provides a breeding ground for river species and improves water quality through settlement of sediment and absorption and re-cycling of nutrients and pollutants. Reviews of river ecosystems can be found in a number of volumes, for example Petts (1984) and Davies and Day (1998).



Figure 2.1: General description of the inter-relationships between a river basin and the water cycle (after Shiklomanov 1999). 2.2 Why Ecosystems are Valuable Each ecosystem is composed of a number of physical, biological or chemical components such as soils, water, plant and animal species, and nutrients. Processes among and within these components allow the ecosystem to perform certain functions such as flood control and storm protection, and generate products such as wildlife, fisheries and forest resources. There are also ecosystem scale attributes such a biological diversity and cultural uniqueness/heritage, that have value either because they induce certain uses or because they are valued themselves. It is the combination of these functions, products, and attributes that make ecosystems important to society. Whether a natural or man-made ecosystem performs a certain function, yields specific products, or possesses certain attributes, is determined by the interaction between chemical and physical characteristics of the site. Characteristics vary greatly between and within each major ecosystem group. Thus forests perform different functions from wetlands and amongst wetlands there is variation both in terms of the types of functions, and the degree to which they are performed.



Ecosystem functions can be grouped into four categories (after De Groot, 1992): regulation functions, habitat functions, production functions, and information functions (Table 2.1). The following sections summarise examples of functions of river basin ecosystems under each of these categories. 2.2.1 Ecosystems as Regulators Ecosystems along the course of a river serve both as regulators of water quantity and water quality. Several types of ecosystems, notably forests and wetlands, are known to act as hydrological buffers, absorbing water when it rains and releasing it gradually over several weeks and months. This not only helps to protect downstream communities from flooding, but helps to ensure that water continues to flow during the drier periods of the year. For example, the forest of La Tigra National Park (Honduras) sustain a well-regulated, high quality water flow throughout the year, yielding over 40% of the water supply of the capital city (Acreman and Lahmann, 1995). Wetland ecosystems are able to reduce rates of water flow and store water above the surrounding water table (for example in a raised bog). The vegetation and hydrology enables the wetland ecosystem to function as a �sponge� and provide the services of flood prevention and water storage. The value of these services may be considerable. Often technical alternatives to regulate the quantity of flow are much more expensive. New York City ensures the quality of its water supply through the protection of the biological and hydrological processes of the upper parts of the catchment on which the water supply depends. Building water treatment plants would cost ten times as much, US$ 7 billion (Abramovitz, 1997). Ecosystems also regulate water quality. On sloping ground, for example, vegetation anchors soil and prevents it from being washed into the watercourse where it would cause siltation and nutrification and reduce light penetration. This would reduce water quality, the health of aquatic ecosystems and the suitability of the water for aquaculture and other uses. The physical structure of watercourses and the organisms that inhabit it also regulate water quality. For example, waterfalls, rapids and aquatic vegetation oxygenate the water, and riverbanks, river beds and vegetation trap sediment. These hydrological and biological processes enable the watercourse to function as a water purification unit providing fresh water. Riverine wetlands play an important role in regulating water quality. They remove toxins and excessive nutrients from the water both by processes of decomposition and uptake by vegetation (Baker and Maltby, 1995). As wetlands hold water for long periods of time, decomposition processes and vegetation are given enough time to remove nutrients and toxins from the water. For example, vegetation found in the Melaleuca wetlands in SE Asia reduces the acidity of polluted water and removes toxic metal ions making the water suitable again for the irrigation of rice (Ni et al., 1997). In this way, the combination of hydrological and biological processes allows these wetlands to function as filtration and purification systems and to provide the service of water purification.



Table 2.1: Natural ecosystems provide many goods and services (functions) to humankind that are often neglected in (economic) planning and decision making. (adapted from de Groot 1992).

REGULATION FUNCTIONS The capacity of natural and semi-natural ecosystems to regulate essential ecological processes and life support systems

PRODUCTION FUNCTIONS Resources provided by natural and semi-natural ecosystems

• Food (e.g. edible plants and animals) • Raw materials (e.g. thatch, fabrics) • Fuel and energy (renewable energy

resources) • Fodder and fertiliser (e.g. krill, litter) • Medicinal resources (e.g. drugs, models,

test organisms) • Genetic resources (e.g. for crop resistance) • Ornamental resources (e.g. aquarium fish,

souvenirs) INFORMATION FUNCTIONS Providing opportunities for reflection, spiritual enrichment and cognitive development

• Maintenance of biogeochemical cycling

(e.g. air-quality regulation and CO2-buffering)

• Climate regulation (e.g. buffering extremes)

• Water regulation (e.g. flood protection) • Water supply (filtering & storage) • Soil retention (e.g. erosion control) • Soil formation & maintenance of fertility • Bioenergy fixation • Nutrient cycling (i.e. maintenance of the availability of essential nutrients) • Waste treatment (e.g. water purification) • Biological control (e.g. pest control and

pollination)

HABITAT FUNCTIONS Providing refugia to wild plants and animals (and native people) in order to maintain biological and genetic diversity • Refugium function (for resident &

migratory species) • Nursery function (reproduction habitat for

harvestable species)

• Aesthetic information (e.g. valued scenery) • Recreation and (eco-) tourism • Religious and cultural values • Cultural & artistic inspiration (i.e. nature

as a motive and source of inspiration for human culture and art)

• Spiritual and historic information (based on ethical considerations and heritage values)

• Scientific educational information (i.e. nature as a natural field laboratory and reference area)

2.2.2 Ecosystems as Habitats Riverine floodplains and river courses, together with their catchments are important habitats for many species of plants, fish, birds and others animals. Wetlands areas are known as prime areas for biodiversity conservation and as important nursery and feeding areas for many aquatic and terrestrial migratory species. In contrast to their fringing wetlands, the main watercourses of rivers function as habitats for animals that require fast-flowing oxygen-rich water. Together, freshwater ecosystems support over 10,000 species of fish and over 4,000 species of amphibians described so far. Freshwaters support a relatively high proportion of species, and more per unit area than other



environments; 10% more than land and 150% more than oceans (McAllister et al., 1997; WCMC, 1998). Coastal deltas and the estuaries of the major rivers, are also important providers of habitats. They provide food and shelter for marine animals that require freshwater conditions for part of their life cycle. Consequently, these coastal wetlands function as habitats for crabs, oysters and shrimp, and provide the service of supporting fisheries based on these goods. For example, it has been calculated that over 90% of the shrimp harvest of the Gulf of Panama is dependent upon the estuaries and mangroves of the region (D�Croz and Kwiecinski, 1980) that are, in turn, dependent on fresh water inflows. Box 2.2: Why biological diversity is important Biological diversity, or biodiversity, is a measure of the variability of genes, species, and ecosystems in a region. This region can vary from a small forest patch to a sub-contintent. Biodiversity is important because plants and animals have made our planet fit for the forms of life we know today. They help maintain the chemical balance of the Earth, stabilise climate, protect watersheds and renew soil. All societies continue to draw on a wide array of ecosystems, species and genetic variants to meet their ever-changing needs. The diversity of nature is a source of beauty, enjoyment, understanding, and knowledge. It is the source of all biological wealth, supplying all our food, much of our raw materials, and a wide range of goods and services and genetic materials for agriculture, medicine and industry worth many billions of dollars per year. Biological diversity should be conserved as a matter of principle, because all species deserve respect regardless of their use to humanity, and because they are all components of our life support system. Prudence dictates that we keep as much biodiversity as possible, but the trend is steadily downward, as more habitats are converted to exclusively human uses. While we are still uncertain about how many species now exist, leading experts calculate that if present trends continue, up to 25% of the world's species could become extinct, or be reduced to tiny remnants, by the middle of the next century. Many more species are losing a considerable part of their genetic variation, making them increasingly vulnerable to pests, disease, and climatic change. 2.2.3 Ecosystems as Providers of Resources Many riverine ecosystems provide large quantities of water, food and energy for direct human consumption, agriculture, fisheries, watering livestock, industry and energy production. Harvesting these goods while respecting the production rate and the regenerative capacity of each species can generate great benefits to human society. One of the most important products of riverine ecosystems is fish. In many areas, river-dependent fisheries form a fundamental pillar of the local and national economy. Direct harvest of forest resources of many floodplains also yields important products, ranging from fuelwood, timber and bark to resins and medicines, which are common non-wood �minor� forest products (Dugan, 1990). Wildlife provides important commercial products such as meat, skins, eggs and honey. Extensive riverine floodplain ecosystems also support substantial seasonal grasslands that are grazed by livestock. For example the Brazilian Pantanal supports over 5 million cattle (Adamoli, 1988). Wetlands also contain a large genetic reservoir for certain plant species, fish and other animals. For example, wild rice continues to be an important resource of new genetic material used in developing disease resistance and other desirable traits.



2.2.4 Ecosystems as Providers of Information Water-based ecosystems provide many opportunities for recreation, aesthetic experience and reflection. Recreational uses include fishing, sport hunting, birdwatching, photography, and water sports. The economic value of these can be considerable. For example in Canada the value of wetland recreation was estimated in 1981 to exceed US$ 3.9 billion (Dugan, 1990). Maintaining the wetlands and capitalising on these uses can be a valuable alternative to more disruptive uses and degradation of these ecosystems. They are important repositories and stores of palaeontological information. Under anaerobic conditions biological material such as pollen, and diatoms and even human bodies can be preserved in peats and lake sediments. In addition, many people gain spiritual or aesthetic benefits from visiting, appreciating and experiencing free-flowing rivers. The symbolism of nature untamed, the bubbling of mountain streams and the majesty of lowland rivers can be an uplifting personal experience, and also provides inspiration for literature and music. 2.3 The value of ecosystem goods and services We all depend on functioning ecosystems for our survival. For many of the World�s poorest people the biological resources of river ecosystems often provide the single most important contribution to their livelihoods and welfare in the form of food supplies, medicines, income, employment and cultural integrity. Such communities often have limited alternative livelihood options and this makes them particularly vulnerable to changes in the condition of the natural resources on which they depend. 2.3.1 Economic valuation techniques Attempts to quantify economic values for ecosystems have been made, both at the micro (Echeverria et al., 1995; Sullivan, 1999) and at the macro levels (Costanza et al., 1997; Alexander et al., 1998). These have demonstrated that replacement costs for ecosystems and their functions are likely to be far higher than the opportunity cost of maintaining the natural system intact. However they have highlighted the difficulties faced by those trying to assess environmental values in monetary terms. The complexity of such systems in particular makes accurate assessments very difficult, since feed-back effects and interactions are not yet fully understood. While immature, the science of valuation of ecosystems has, however, allowed a clear distinction to be made between: i) ecosystem products that can be sold on the market (and for which prices exist, revenue is generated and jobs maintained); ii) non-marketable services (such as water quality maintenance or groundwater recharge) that are more difficult to price in evolving circumstances; and iii) intrinsic values such as the beauty of natural landscapes. Less clear guidance is available on how to price "free services", particularly in developing economies that need to generate real, not virtual, incomes and where society may value these services more highly 20 years hence as the economy develops. The assessment of intrinsic, cultural and aesthetic values cannot usually be addressed in monetary terms in the same way, as there is no replacement when these values are lost. These values are therefore usually addressed through a political or ethical process rather than a process of economic valuation. Some current valuation techniques are based on preferences, and money provides a measure of value. Preferences are both subjective and dynamic, and measurement of such �fuzzy� variables is a difficult task. The fact that preferences are subjective means that different groups within society are likely to have different values, and this demonstrates the importance of consulting a wide variety of stakeholders when considering the question of environmental values.



It should be stressed that this review does not see the valuation of ecosystem services as a straightforward accounting exercise where the calculated value is simply added to the cost:benefit balance sheet. As the ecosystem dimensions of dam projects also include ethical and societal values, the monetary value serves as an input to multi-criteria decision-making and raises awareness of costs that are currently hidden and negated in the accounting exercise. 2.3.2 The Monetary Value of Freshwater Ecosystems Many functions of freshwater ecosystems and wetlands have direct and indirect economic importance. Entire communities depend on the functions provided by freshwater ecosystems and, as such, ecosystems have enormous value. It is still difficult to translate this value into monetary terms, leading to the continuing loss and degradation of water systems due to undervaluation and neglect in economic accounting procedures. In Nigeria, for example, it has been shown that the net economic benefits of the Hadejia-Nguru floodplain ecosystem are much larger than those from irrigated land: US$ 32 versus US$ 0.15 per 1,000m3 of water, not including benefits of floodplain inundation for groundwater recharge and water supply to the productive ecosystem of Lake Chad (Adams 1992). A first attempt to synthesise existing knowledge on the monetary benefits of the services of ecosystems on a global scale was published in 1997 (Costanza et al. 1997). Table 2.2 gives a summary of the main functions, and monetary values, of freshwater and wetland ecosystems. Table 2.2: Global monetary values of freshwater and wetland functions (in US$ billion, 1994). (functions based on de Groot 1997; values based on Costanza et al. 1997). Ecosystem functions (goods & services) Active or direct

use values (mainly

market prices)

Passive or indirect use

values (mainly shadow

price)

Per cent of Global Total

(for a particular function)

1. REGULATION FUNCTIONS 1.1 Climate regulation & biogeochemical cycling (e.g. CO2)

? 44 3 %

1.2 Water buffering (e.g. flood prevention) ? 350(a) 40 % 1.3 Waste treatment ? 5,300 31 % 1.4 Biological control ? 14 3 % 2. HABITAT FUNCTIONS 2.1 Refugium function ? (c) (c) 2.2 Nursery function 62 62(a) 100 % 3. PRODUCTION FUNCTIONS 3.1 Water 840 840(a) 99 % 3.2 Food (mainly fish) 186 (b) 13 % 3.3 Raw materials & energy 40 (b) 6 % 3.4 Genetic material & medicines (d) (d) (d) 4. INFORMATION FUNCTIONS 4.1 Aesthetic information (e.g. views) ? 5 2 % 4.2 Recreation and tourism 304 (b) 37 % 4.3 Cultural values (e.g. art, science) (d) (d) (d) Total (in US$ billion/year) 1,782 + 6,905 Average 26% Notes: (a) The total value of the flood prevention, nursery function and water supply given in Costanza et al. (1997) was based on a

combination of market and shadow prices. For simplicity, it has been estimated that 50% of the calculated value is included in market prices.



(b) The values given for food, raw materials and tourism are based only on market prices. However, these resources also have an unknown (direct) consumptive use value (many people depend on freshwater systems for these resources directly, without market intervention).

(c) In addition to active and passive use values, many ecosystem functions have so-called non-use or intrinsic value. In this study it is not attempted to place a monetary value on the intrinsic importance of nature but it could, in part, be derived from the money people are willing to spend to maintain the refugium function of natural ecosystems.

(d) Freshwater and wetland systems are important sources of genetic material, medicines and cultural values but little or no information is available on the monetary value of these ecosystem functions.

This analysis shows that, world-wide, freshwater and wetland systems account for approximately 26% of the total economic value of all ecosystem services (which vary substantially by function, as the last column shows). It can be concluded that still only about 20% (US$ 1,782 billion) of the economic value of coastal and freshwater systems is accounted for in market pricing mechanisms. All other values, which mainly relate to regulation and habitat functions, are not yet (properly) accounted for. 2.4 Ecosystems and River Basin Development The central issue of river basin development is to decide how to allocate water to maximise the benefits it provides to society as a whole. In the past little consideration was given to the importance of ecosystems and their multiple values to society. Very regularly these were overlooked when single-use developments were predominantly considered. Today, in some societies high value is placed on sustaining healthy �pristine� river ecosystems because they are believed by many to have an intrinsic value in themselves. Human � nature interactions within river basins are so strong that the system as a whole is the logical level for environmental and water management measures. The sustainable development of river basins requires the development and implementation of management plans at the level of the entire basin (Newson 1997, Mostert 1999). The conservation of valuable ecosystem goods and services forms an essential element in these (Box 2.3). To manage a river basin implies to optimally allocate scarce resources among competing users now and in the future. This requires political will, accurate information and knowledge of the basin, sustainable technologies, appropriate institutional and legal arrangements, stakeholder participation and economic viability (Burton 1999). There is no single best approach for river basin development as each basin is unique in its configuration and state of development. Box 2.3: Watershed Management Watershed management programs generally include a variety of subprograms designed to reduce erosion through establishment or expansion of protected areas, improved management of protected areas, restoration and rehabilitation of forest or other biotypes, and the introduction of improved agricultural technologies or alternative types of production. One of the more promising approaches is to introduce agroforestry practices that have the dual benefit of increasing forest cover in the basin and replacing existing agricultural practices with cultivation and promotion of non-destructive, but economically beneficial use of those resources. As the importance of the inter-relationship between dam development and the surrounding watershed has become more fully realised, measures to protect and manage the watershed are being promoted in association with the construction of new dams. These include management of agricultural, urban and natural areas throughout the basin. While initially viewed as compensation to ameliorate the negative impacts of dams (section 4.2.3), these measures for protecting and enhancing environmental resources in the basin are increasingly seen as also sustaining the operational life of the project. In 1996 the Compania Nacional de Fuerza y Luz, a private utility in Costa Rica (but owned by the state utility), started a management program for the upper watershed (142 km2) of its Brasil hydropower plant on the Virilla River. The project consists of reforestation, forest conservation, and



environmental education components, and will cost US$9 million over a ten-year period. It is expected that such a change in land use management upstream of the dam will improve the water regime by 2%, therefore resulting in an increased power production of 9%. The reforestation component will change the land use from pastures to tree plantations in 1,250 ha, while an additional 1,250 ha will be managed as agroforestry systems. 3,400 ha of forest will be conserved, which together with the plantations will yield the additional benefit of the fixing of 584,000 tonnes of carbon (Mora 2000/ENV223). Dams have been an important technology in river basin development. Planning for new dams and upgrading of existing dams should be carried out within the context of river basin development and management plans. National sector reviews, plans and efforts for the implementing of an integrated water resources management approach need to be taken into account in dam development and management. The WCD�s Thematic Review V.3, �River Basins- Institutional Frameworks and Management Options� deals with this subject in more detail. Also, Thematic Review V.1 �Planning Approaches� also looks at the planning level of the dam building decision process. 2.5 International and national recognition of ecosystem values During the past two decades, legal experts have attempted to understand and clarify the basic concept underlying the governing principles regarding respect for all forms of life. The three resulting �over-arching� principles are to be read in conjunction with principles regarding human needs � development and poverty eradication. Di Leva (1999) defines the three principles as: 1. Recognise that the enjoyment of basic human rights is dependent on the continued existence of a

ecologically sustainable natural environment; 2. Recognise that decisions impact on future generations which have inherent rights (inter �

generational equity); 3. Respect all life forms independent of their value to humanity (UN 1982). To date there is a growing recognition of these values (Box 2.4). At the international level, this has lead to the development of a UN Charter for Nature and a range of environmental conventions to protect species and specific ecosystems. The UN Charter for Nature (UN 1982) was adopted by consensus by the UN General Assembly to provide a high-level guiding principle to govern humankind�s responsibility for nature conservation and management. The Charter has several principles. One of these includes �Ecosystems and organisms as well �. resources utilised by men shall be managed to achieve and maintain optimum sustainable productivity, but not in such a way as to endanger the integrity of those ecosystems or species with which they coexist.� As Figure 2.2 illustrates the attention given to ecosystem conservation by national governments from all parts of the developing and developed world, as reflected in the signatures of international treaties, has increased markedly in recent years. For example, The Convention on Wetlands (Ramsar 1971) has been ratified by 110 contracting parties and 177 countries have ratified the Convention on Biological Diversity to date, that is 96% of all UN-recognised countries in the world (185) (UN-CBD 2000). This argues strongly that every effort should be made to avoid irreversible loss of resources that are likely to become more valuable to all societies in future.



Box 2.4: International recognition of ecosystem values Contracting Parties give the following motivations for respecting ecosystem values: ��the intrinsic value of biological diversity and of the ecological, genetic, social, economic, scientific, educational, cultural, recreational and aesthetic values of biological diversity and its components�� the importance of biological diversity for evolution and for maintaining life sustaining systems of the biosphere�� that conservation and sustainable use of biological diversity is of critical importance for meeting the food, health and other needs of the growing world population, for which purpose access to and sharing of both genetic resources and technologies are essential�� Convention on Biological Diversity of June 5, 1992 ____________________ ��that wetlands constitute a resource of great economic, cultural, scientific, and recreational value, the loss of which would be irreparable�� Convention on Wetlands of International Importance Especially as Waterfowl Habitat-The Ramsar Convention of 1971

Figure 2.2: Cumulative number of countries ratifying the main environmental Conventions(World Heritage (1), Ramsar Convention on wetlands (2), trade in endangered species (3), biological diversity (4), and climate change (5)). Currently, national governments and civil society are faced with the challenge of deciding how the continuous process of the implementation of these conventions can be strengthened. Often this relates to strengthening national legislation and its implementation / enforcement. An important practical issue to resolve is defining how much water should be used for the maintenance of ecosystems to provide environmental goods and maintain elemental services. Recent changes in the South African



law defined specific allocations for maintaining river flows for ecological reasons. The basic human needs and environmental requirements are now identified as �The Reserve� and have priority of use by right (Asmal 1998). Increasingly, a range of techniques are available to determine the requirements of downstream ecosystems (King et al. 1999). This has to be determined along with how much water should be used to support agriculture, industry and domestic services to provide basic goods. Obviously, the value that society places on these alternative goods and services will determine the pattern of allocation. It is important therefore that the costs and benefits to society of allocating water to maintain ecosystems or to support agriculture, industry and domestic uses are well understood. 2.6 Conclusions Ecosystems provide goods and services to human society. These have high values and provide the basis for sustainable livelihoods. The goods and services these systems provide, such as food, timber, fisheries and drinking water, form an important natural resource base for many societies throughout the world. The total global value of ecosystem goods and services is estimated at US$ 33 trillion per year, of which roughly 25% relates directly to freshwater ecosystems. Freshwater ecosystems are also known to regulate water quality and quantity and to provide habitats for tens of thousands of species. To maintain natural ecosystem goods and services it is essential to conserve and sustainably manage species and ecosystem processes. Together they form the integrity of healthy ecosystems. For the maintenance of healthy river ecosystems, the integrated management of land and water resources is required within an entire river basin. Dams therefore cannot be an objective in themselves, but should be seen as a tool that should be used with great care and prudence. A large majority of sovereign states have committed themselves to conservation of nature through a range of international conventions and national legislation and policies, as exemplified by the extensive ratification of the five major international conventions on nature conservation. The Convention on Biological Diversity alone is ratified by over 96% of all countries. The effective implementation of these, however, often remains weak in relation to dam development.



3. Ecosystem Impacts of Large Dams 3.1 Introduction Dams are structures designed to store or divert water. They are intended to alter the natural distribution and timing of streamflows in order to meet human needs. As such, they also alter essential processes for natural ecosystems. Dams constitute obstacles for longitudinal exchanges along rivers. By altering the pattern of downstream flow (i.e. intensity, timing and frequency), they change sediment and nutrient regimes and alter water temperature and chemistry. Storage reservoirs flood terrestrial ecosystems, killing terrestrial plants and displacing animals. As many species prefer valley bottoms, large scale impoundment may eliminate unique wildlife habitats and extinguish entire populations of endangered species (Nilsson and Dynesius, 1994). Terrestrial ecosystems in reservoir areas are replaced by lacustrine, littoral and sublittoral habitats and pelagic mass-water circulations replace riverine flow patterns. As such, dams and their reservoirs also provide new opportunities as they create new habitats and over time could be considered to become part of the new environment. The degree to which these �new� habitats can compensate the loss of original habitats, species and ecosystem goods and services is however often contested. Within these broad patterns of change, there is a wide diversity of specific impacts that vary from dam to dam, catchment to catchment, ecosystem to ecosystem, and species to species. For example loss of some ecosystems may benefit some species (e.g. waterfowl and fish that favour deep water), but others may suffer significant loss of population, or even extinction. The purpose of this chapter is to review current understanding of these impacts. An initial assessment of the scale and variability of impacts is followed by an analysis of specific ecosystem impacts divided according to Petts (1984) � Fig 3.4. He suggests the following breakdown that is used for the basic structure of this chapter. 1. Assessment of first-order impacts that influence the key abiotic driving variables of the riverine

ecosystem (e.g. temperature and hydrological flows). 2. Definition of second-order impacts that include primary productivity as the basis for the food

chain. 3. Third-order impacts on the food web � implications for fauna. This approach is adopted for both the upstream and downstream areas, followed by an assessment of specific impacts on biodiversity and cumulative impacts within the catchment. 3.2 Scale and Variability of Impacts There are different types of dams each with their own operating characteristics. Similarly, dams have been built in a wide array of conditions, from highlands to lowlands, temperate to tropical regions, fast-flowing to slow-flowing rivers, urban and rural areas, etc. The combination of dam types, operating systems, and the contexts where they are built, yields a multitude of conditions that are site-specific and very variable. This complexity makes it difficult to generalise about the impacts of dams on ecosystems as each specific context is likely to have different types of impacts and to different degrees of intensity. However, at a certain level of generality, some indications can be given of the most likely impacts and their relative order (Box 3.1).



Box 3.1: Types of dams, in descending order of impacts on ecosystems Storage dams Large reservoirs with or without river diversions. Diversion (run-of-river) Uses flow with limited or no storage; diverts all or part of in-stream

or across catchments. Run-of-river Uses flow with limited or no storage and no river diversion. In addition to dam type, the height of dams and their reservoir areas are extremely variable. ICOLD recognises a large dam as one that is higher than 15 m and/or, between 5-15 m high and impounding more than 3 million cubic meters of water. Within any individual country there is a wide range of different dam types, dam heights and reservoir sizes. For example, Figure 3.1 shows the distribution of reservoir size for dams in Turkey where reservoir size varies from a few hectares to over 80,000 ha.

Figure 3.1: Distribution of reservoir area for dams over 15m high in Turkey(DSI, 1999) Although dam and reservoir size are highly variable, it is possible to examine the broad scope of the impacts. Figure 3.2. shows the total area of reservoirs by continent, while Figure 3.3. shows the average area of reservoirs in each continent. These highlight both the large total areas involved and the great variation between regions in terms of the average size of the reservoirs.



Figure 3.2: Total area of large-dam reservoirs (1000's km2) by region (ICOLD 1999)

Figure 3.3: Average area of large dam reservoir (km2) per region(ICOLD 1999) Table 3.1 gives some examples of the scale of the impacts of water diversion on water flows from countries where data is available, which indicate the significant effects of man�s activities on water flows in major catchments. These impacts are likely to differ between northern countries, where temperate climates and little irrigation mean that there is little water diversion and semi-arid countries, which may have extensive out-of-river uses and high evaporation rates to contend with. Table 3.1: The potential scale of the impacts of dams River Example of Scale of Impact Source Indus, India Only 28% of the Indus� total

annual streamflow reaches its delta. For dry season flows it is only 10%. The dams along the river retain ~75% of the silt carried by the river.

Anonymous 1997, WCD Tarbela report 1999

Various, South Africa There are 520 major regulating structures in South Africa that capture nearly 50% of the mean annual runoff

Davies & Day 1998



Murray-Darling, Australia Mean annual outflow from the Murray to the sea has been reduced from some 13,700 GL/yr under natural conditions to 4,900 GL/yr, or as low as 35% of natural flows.

Murray Darling Basin Commission 1999, quoted in WWF Australia 1999/ENV220, www.mdbc.gov.au

Japan Of 35,000 rivers; only two have not been either dammed or modified in any way.

McAllister et al. 1997, Dams Yearbook 2000

North America north of Mexico, Europe, and former Soviet Union

77% of the total water discharge of the 139 largest river systems is strongly or modestly affected by fragmentation of the river channels by dams and by water regulation resulting from reservoir operation, inter-basin diversion, and irrigation.

Dynesius and Nilsson 1994

United States Only 42 free-flowing rivers longer than 125 miles remain � less than 2% of the country�s 3.1 million miles of rivers and streams

Abramovitz 1996

Europe There are more than 10,000 major reservoirs in Europe, covering a total surface area of ca 140,000km2, which is equal to app. 4 times the national territory of the Netherlands.

Kristensen & Hansen 1994, ICOLD 1999

Columbia, USA 5% to 14% of adult salmon are killed at each of the eight dams through which they pass on their way up the river

Collier et al 1996

Studying the importance of dam impacts on ecosystems at a global scale, and incorporating the complexity of temporal and spatial variability is made difficult by the lack of available information. Although there are several databases containing some information on dams (Table 3.2.) there is, at present, no comprehensive database of dams with geo-referenced locations in the public domain. This means that there is no global data set that can easily be used to locate and map large dams. Hence, it is not possible, except in very general terms, to relate dam distribution to major biotopes and to deduce the different effects dams have in different regions of the world. Table 3.2: Some databases of dams Name Source World Register of dams ICOLD 1999 World atlas of large dams International Journal on Hydropower and Dams,

Hydropower & Dams 1999 European Lakes, Dams and Reservoir database (ELDRED)

European Environment Agency 1999

National registers of dams e.g. USA register of dams

ICOLD national committees, USCOLD 1999



Including the temporal and spatial variability of ecosystems and dams increases the complexity of the analysis. Temporal scales are important because some ecosystem changes caused by dams may not become evident for many years, or may evolve through time (see for example section 3.7). Similarly, the time factor is essential when considering some impacts which might become irreversible with time, such as the accumulation of toxic sediments in a reservoir that can make dam decommissioning more difficult or even economically impossible. Spatial scales are also important for ecosystems as some impacts of dams are felt far away from the dam site. For example, if the dam reduces the amount of detritus in the streamflow, this might affect the fisheries production in the river�s estuary many kilometres downstream. The cumulative effects of many dams in a catchment might also be different from the sum of the impacts of each individual structure, reinforcing the need for a spatial assessment (see section 3.8).

3.3 Framework for Analysis

In considering the impact of dams on riverine ecosystems it is important to recognise the interconnected nature of the ecosystems concerned and the often far-reaching consequences of change in individual ecosystem components. Analysis of this complexity can be approached from various different perspectives. Many groups and individuals, such as McCully (1996), Davies & Day (1998), Veltrop (1999), and USGS (1996), World Bank, have attempted this in different ways. Two of these approaches are presented in Annex 1 and Annex 5 � A summary of impacts derived from ICOLD Bulletins and USCOLD, and another from the World Bank�s Environmental Assessment Source Book. Overall such an analysis is only possible by breaking down the impacts into categories and there is no agreed or definitive way of doing this. This report adopts the approach of Petts, 1984. (Figure 3.4) as he dissagregates the components of ecosystem complexity by structuring impacts according to their level.

First-order impacts are the immediate abiotic effects that occur simultaneously with dam closure and influence the transfer of energy and material into and within the downstream river and connected ecosystems (e.g. changes in flow, water quality and sediment load).

Second-order impacts are the abiotic and biotic changes in upstream and downstream ecosystem structure and primary production, which result from first-order impacts. These depend upon the characteristics of the river prior to dam closure (e.g. changes in plankton, macrophytes and periphyton), and these changes may take place over many years.

Third-order impacts are the long-term biotic changes resulting from the integrated effect of all the first- and second-order changes, including the impact on species close to the top of the food chain (e.g. changes in invertebrate communities and fish, birds and mammals). Complex interactions may take place over many years before any new �ecological equilibrium� is achieved. Table 3.3: Upstream and downstream impacts according to first, second, and third order as described. Location in Relation to the Dam

Category of Impact (as in Petts 1984)

Impact

Modification of the Thermal Regime Accumulation of Sediment in the Reservoir Changes in Water Quality

First-Order Impact

Groundwater along reservoir

Plankton and Periphyton Growth of Aquatic Macrophytes

Second-Order Impact

Riparian Vegetation

Upstream

Third-Order Impact Invertebrates, Fish, Birds and Mammals



Daily, Seasonal and Annual Flows Water Quality Reduced Sediment Flows Changes to Channel, Floodplain and Coastal Delta Morphology Groundwater in riparian zone Water temperature � thermal pollution

First-Order Impact

Ice formation Plankton and Periphyton Growth of Aquatic Macrophytes Riparian Vegetation

Second-Order Impact

Carbon flows and cycle distortions

Invertebrates, Fish, Birds and Mammals Estuarine Impacts Marine Impacts

Downstream

Third-Order Impact

In general terms the complexity of interacting processes increases from first- to third- order impacts. Since ecosystem functioning is guided by abiotic steering variables related to hydrology (i.e. water quantity and flow regime), geomorphology and water quality, observations related to these ecosystem components can be used as primary indicators of river ecosystem conditions. Changes in abiotic steering variables are key to understanding the long-term ecological consequences of dams as they are the underlying mechanisms by which many habitats are maintained. As Ligon et al (1995) stated , �If [a] stream�s physical foundation is pulled out from under the biota, even the most insightful biological�program will fail to preserve ecosystem integrity.�

BirdsMammals

Fish

Invertebrates

Aquatic MacrophytesChannel formSubstrate composition

Plankton

Water quality

Water quantity

Flow regime

Sediment load

Barrier effects

Algae

THIRD-ORDERIMPACTS

SECOND-ORDERIMPACTS

FIRST-ORDERIMPACTS

Terre

strial

env

ironm

ent

Hydrology

Primary production

Morphology

Table 3.4: A framework for assessing the impact of dams on river ecosystems(modified from Petts, 1984).



Box 3.2: Construction impacts The framework presented in Figure 3.4 and used here does not include the often significant impacts on non-aquatic ecosystems caused by the construction of physical infrastructure, transmission lines and roads. Power transmission lines are typically cleared of shrubs and trees by cutting or use of herbicides. Habitat is thus modified on the cleared swathes and runoff or winds may result in off-site effects of the herbicides. Hydro-Québec reports that the length of its transmission lines totals 32,000 km (HQ and GDG 1999). To deal with this problem, it has established, as outlined in its environmental code, specific measures to minimise construction impacts on the environment (Hydro Quebec 1991). HQ and GDG (1999) point out that powerlines have less environmental impact than roads and railway lines, which are essentially vegetation free. Some powerline constructions support species, like the smoky shrew and southern bog lemming, that are in danger of extinction in Québec. The lines and roads, at times, fragment habitat for organisms both large and small. Access roads to dam sites can also cause a significant direct impact on natural ecosystems, while also providing access to previously remote areas for settlers and hunters. Blasting at construction sites can also be a major source of disturbance, in particular during certain times of the life cycles of animals such as calving caribou in Canada (Kiell 2000/ENV202). A comprehensive analysis of indirect impacts due to the development of dams, for example impacts of new irrigation schemes, development of navigation or tourism, and human health impacts, do not form part of this review. Other papers prepared for the WCD secretariat will deal with these issues, such as Thematic Review IV.2 Assessment of Irrigation Options and I.1 Social Impacts of Large Dams: Equity & Distributional Issues. 3.4 Information Constraints Over the last 30 years, the finding of numerous scientific studies relating to the environmental impacts of dams have been reported in the scientific literature. Some of these findings have been summarised within wide-ranging compilations (e.g. ICOLD, 1981; Petts, 1984; McCully 1996, USGS 1996, ICOLD, 1988). Research continues and research findings are constantly being up-dated. Other sources of information include a significant body of �grey� literature (e.g. consultant reports), usually written during the planning of a river impoundment. Most of these case studies consist of pre-regulation investigations. Finally, there is now an increasing amount of information and related �position papers� published by various organisations. To an extent the perspective of the people and organisations involved cloud the latter and the information presented may be selective in nature. In order to effect a thorough investigation of the impacts of dams on ecosystems, data are required on both the abiotic and the biotic components of ecosystems (eg Annex 3). Pre- and post- impoundment information is required on: the hydrology of the river (both at the site of the dam and downstream); hydraulic characteristics of the river; water quality; geomorphological characteristics (i.e. sediment transport); aquatic biota and their habitat requirements; riparian vegetation and associated fauna; vegetation and associated fauna in the upper watershed; and the direct use of the river and its associated resources by local people. To date however, most studies have investigated the impact of one dam or a few dams on specific components of ecosystems rather than on the ecosystem as a whole. Most studies are focussed primarily on the abiotic, primarily first-order impacts. Relatively few studies have assessed second- and third-order impacts, possibly because of the longer time frame required before new equilibrium states are attained and total change becomes apparent. At higher trophic levels (e.g. impact on



terrestrial vertebrates), very limited amounts of data relate to long-term change caused by dam construction, though possible impacts are subject to much speculation (Nilsson and Dynesius, 1994). Further, most studies of the environmental impacts of dams have been conducted in temperate climates. Relatively little is known of the possible third-order impacts in tropical climates, where biological processes often proceed faster and so ecological changes become apparent more quickly (Bardach and Dussart, 1973). Scientists and conservationists suspected that the flooding of large areas in the tropics is especially likely to contribute to global species extinction. Although largely unproven because the data are unavailable, this suspicion exists because of the high species richness and possible endemicity of many of the affected areas. One well-documented example is the case of the Kihansi Spray Toad in Tanzania (Finlow-Bates, Gentle and Lovett, 2000). This report therefore draws on the available literature while being aware of the dangers of generalising from �worst case� examples that have been well studied. It seeks to lay out and illustrate the generic impacts that are known to occur while recognising that the nature and scale of these impacts will vary from site to site. It emphasises the need to look holistically at the impacts on a case-by-case basis rather than addressing them piecemeal. 3.5 Upstream Impacts The construction of a dam results in post-impoundment phenomena that are specific to reservoirs and do not occur in natural lakes. One difference is that with first reservoir filling terrestrial habitats are submerged and destroyed. Another difference is that level fluctuations may be much larger than those normally found in a natural lake. Non-earth storage dams often have a bottom outlet. This may allow both sediment flushing and water releases from deep below the surface. Both management measures cannot be carried out with most natural lakes. Nevertheless some older reservoirs can be considered as lakes and the challenges presented in managing them are often the same (Dinar et al., 1995), such as the management of the riparian wetland habitats and fisheries. In this section a summary is given of the first-, second- and third-order impacts on upstream ecosystems comprising the reservoir and upper reaches of the river. 3.5.1 First-Order Impacts on Key Parameters 3.5.1.1 Modification of the Thermal Regime Temperature is an important regulator of many important physical, chemical and biological processes. In particular temperature, in conjunction with nutrient dynamics and seasonal availability of minerals and light conditions, controls primary productivity. Reservoirs act as thermal regulators that may fundamentally alter the seasonal and short-term fluctuations in temperature that are characteristic of many natural rivers. The relatively large mass of still water in reservoirs allows heat storage and produces a characteristic seasonal pattern of thermal behaviour. Depending on geographical location, water retained in deep reservoirs has a tendency to become thermally stratified (Hutchinson, 1957). Typically, three thermal layers are formed: i) a warm, well-mixed, upper layer (the epilimnion); ii) a cold, dense, bottom layer (the hypolimnion) and iii) an intermediate layer of maximum temperature gradient (the thermocline). Water in the hypolimnion may be up to 10oC lower than in the epilimnion and in the thermocline the temperature gradient may be up to 2oC for each metre. A range of factors, including climatic characteristics, controls the exact nature of thermal stratification. Reservoirs closest to the equator are least likely to become stratified. At higher latitudes the overall controlling factor is the variable input of solar energy. Considerable variability may occur within a region as a consequence of different topographies and different reservoir-catchment morphometrics. Shallow reservoirs respond most rapidly to fluctuations in atmospheric conditions and are less likely to become stratified. Strong winds can affect rapid thermocline



oscillations. The patterns of inflows into, as well as the nature of outflows from, the reservoir also influence the development of thermal stratification. Currents generated by large water level fluctuations in reservoirs caused by the operating regime can sometimes prevent thermal stratification. Many deep reservoirs, particularly at mid and high latitudes do become thermally stratified as do natural lakes under similar climatic and morphological conditions. However the release of cold water into the receiving river from the hypolimnion of a reservoir is the greatest �non-natural� consequence of stratification. This is addressed in Section 3.6.1. 3.5.1.2 Accumulation of Sediment in the Reservoir River currents transport particles from the fine ones in turbid water to the coarser ones such as sand, rocks, and boulders. The speed and turbulence of currents enable transport of the geologically-derived materials. As waters slow down and turbulence declines, the particles tend to drop out. Lowered currents and turbulence occur when the river bed gradient diminishes, as in the lower reaches of many rivers, upon entry into lakes or the sea. This also happens when river flow reaches man-made reservoirs. Many reservoirs retain a large proportion of the sediment load supplied by the drainage basin. About 1,100 km3 of sediment has accumulated in the world�s reservoirs, taking up almost 20% of the global storage capacity (Mahmood, 1987). The Glen Canyon Dam on the Colorado River, USA, traps 66 million tons of sediment per year, equivalent to 95% of the sediment load. (Collier et al., 1996). As with a natural lake, the �trap efficiency� of a reservoir depends on: i) the size of the reservoir�s catchment; ii) the characteristics of the catchment that affect the sediment yield (i.e. geology, soils, topography, vegetation and human disturbance) (see Kettab and Remini, 1999/ENV048); and iii) the ratio of the storage capacity to the river flows into the reservoir. However, unlike a natural lake, the type of outlet on the dam will also affect the trap efficiency of a reservoir. Sediment transport shows considerable temporal variation, both seasonally and annually. The amount of sediment transported into reservoirs is greatest during floods but also depends largely on the management of the upper catchment. Sediment transport and deposition have both positive and negative impacts. Sedimentation can create new habitats in the reservoir, especially at the mouth of the river, while sedimentation reduces storage capacity. For example, Nepal�s Kulekhani hydro dam, estimated to have a useful life of 85 years when commissioned in 1981, had lost nearly half of its 12 million cubic metres of dead storage capacity by 1993, while El Salvador�s Cerron Grande reservoir was found to have a useful life of 30 years, instead of the originally expected 350 years (Dorsey et al 1997). In North Africa, severe autumn rains and a mountainous terrain mean that reservoirs receive enormous sediment loads. For example, the Mellegue reservoir in Tunisia has lost 92% of its storage capacity since filling in 1954, and the Mohamed V reservoir in Morocco has lost 58% of its storage capacity since filling in 1967 (Kettab and Reminin, 1999/ENV048). 3.5.1.3 Changes in Water Quality Water storage in reservoirs induces physical, chemical and biological changes in the stored water and in the underlying soils and rocks, all of which affect water quality. The chemical composition of water within a reservoir can be significantly different from that of the inflows. The size of the reservoir, its location in the river system, its geographical location with respect to altitude and latitude, the storage retention time of the water and the source(s) of the water all influence the way that storage detention modifies water quality.



Major biologically-driven changes occur within thermally stratified reservoirs. In the surface layer, phytoplankton often proliferate and release oxygen thereby maintaining concentrations at near saturation levels for most of the year. In contrast, the lack of mixing and sunlight for photosynthesis in conjunction with the oxygen used in decomposition of submerged biomass can result in anoxic conditions in the bottom layer. Nutrients, (i.e. phosphorous and nitrogen) are released biologically and leached from flooded vegetation and soil. Although oxygen demand and nutrient levels generally decrease over time as the organic matter decreases, some reservoirs require a period of more than 20 years to develop stable water-quality regimes (Petts, 1984). After maturation, reservoirs, like natural lakes, can act as nutrient sinks particularly for nutrients associated with sediments. Eutrophication of reservoirs may occur as a consequence of large influxes of organic loading and/or nutrients. In many cases these are a consequence of anthropogenic influences in the catchment (e.g. application of fertilisers) rather than a direct consequence of the presence of the reservoir. For example, eutrophication of the heavily-regulated Waikato River system in New Zealand was enhanced by sewage and stormwater discharges (Chapman, 1996). Nutrient pulses, in conjunction with the specific environmental conditions, can result in water blooms of blue-green algae which (in addition to being aesthetically unpleasant) can cause oxygen depletion and increased concentrations of iron and manganese in the bottom layer and increased pH and oxygen in the upper layers of stratified reservoirs (Zakova et al., 1993). Mercury and other heavy metal contamination has recently been highlighted as a major reservoir problem in some countries (Friedl, 1999/ENV079). Mercury is naturally present as a harmless inorganic form in many soils. However, bacteria breaking down decomposing matter under a new reservoir transforms this inorganic mercury into methylmercury, a toxin of the central nervous system. Plankton and other creatures at the bottom of the aquatic food chain absorb the methylmercury. As the methylmercury passes up the food chain it becomes increasingly concentrated in the bodies of the animals eating contaminated prey (Paterson et al. 1998). Through this process of bio-accumulation, levels of methylmercury in the tissues of large fish-eating fish or birds at the top of the food-chain can be several times higher than in the small organisms at the bottom of the chain. The degree to which fauna have been intoxicated with mercury has been shown to be variable (Friedl, 1999). In other reservoirs no effects are reported (Lucotte et al. 1999). Water quality changes due to the reservoir will be reflected throughout the downstream watercourse, affecting primary productivity and the invertebrate fauna that provide the basis for the foodweb. Annex 3 provides a comparison of a series of water quality variable parameters under pre and post impoundment conditions of two reservoirs in the Mekong river basin, together with corresponding data on second and third order variables, notably phytoplankton, zooplankton and fish. 3.5.2 Second Order Impacts � Changes in Primary Production 3.5.2.1 Plankton and Periphyton Within natural fast-running river (lotic) systems, phytoplankton production is often negligible, only derived from lakes, low velocity backwaters and benthic algal communities. Natural rivers, particularly clean, slow-moving lowland rivers, do contain free-floating micro-organisms, but the plankton populations are inherently unstable and dependent upon the frequency of high discharges. The introduction of a reservoir into a river system, particularly in headwater areas, can markedly alter its primary productivity. The hydrological characteristics and thermal and chemical regimes of reservoirs are unique, so the character of primary production within reservoirs is highly site- and catchment-specific. Upon dam closure, the river (lentic) system resets itself as the reservoir fills. Often a microbial population explosion releases nutrients as the newly submerged organic matter begins to decompose.



This stimulates the rapid development of the phytoplankton.. The enrichment of reservoir water by large amounts of nitrogen and phosphorous, as a result of the decay and mineralisation of organic matter flooded by the reservoir, may lead to a multiplication of blue-green algae. This in turn drives invertebrate productivity and fisheries production. In many reservoirs, fisheries thrive for 4-5 years after closure and then decline as primary productivity drops. The occurrence of lacustrine plankton assemblages varies seasonally and with individual reservoirs, depending upon their geographical location and catchment inputs. In temperate and high latitude climates plankton populations are lowest in the cold winters and greatest during the warm summers. Tropical reservoirs have no seasonal check to plankton growth comparable to winter in temperate regions. Given the favourable thermal regime, the productivity of tropical reservoirs is mainly limited by the introduction of highly turbid waters and wind-induced turbulence during the wet season. Periphyton are layers of algae attached to any submerged object, including larger plants. Diatoms normally dominate the attached algae of lotic systems. Conversion from a lotic to a lentic environment will provide opportunity for some species of periphyton, while destroying the habitat for others. Periphyton are most likely to proliferate where light penetrates, in the shallow water close to the reservoir edge. The exact species composition will be determined by the nature of the substrate, the presence or absence of aquatic macrophytes, the temperature and chemistry of the reservoir water and the operation of the dam. 3.5.2.2 Growth of Aquatic Macrophytes There may be increased opportunity for aquatic macrophytes in the littoral and sub-littoral zone of reservoirs. The rapid build up of delta deposits near river inlets to the reservoir reduces water depths and can encourage macrophyte growth. However, their ability to colonise these areas may be limited if there are large fluctuations in reservoir level. Further out in the reservoir opportunity for aquatic macrophytes may be limited by lack of light penetration to depth, yet in windless conditions with high nutrient levels, colonisation by floating invasive species is possible (Box 3.3). The growth of macrophytes can be an advantage as they create wetland-like conditions with biodiversity values, support fisheries and assist in structuring habitats. However, they may also provide habitat for disease vectors such as bilharzia-carrying snails, mosquitoes and intermediate hosts for flukes. Box 3.3: Invasive species and large dams The modified habitats resulting from large dams often create environments that are more conducive to non-native and exotic plant, fish, snail, insect and animal species. These resulting non-native species often out-compete the native species and end up developing ecosystems that are unstable, nurture disease vectors, and are no longer able to support the historical environmental and social components. The short-term gain in having a reservoir or hydroelectric plant may not compensate for the loss of critical ecosystem functions. Species of floating and submerged weeds that are particularly virulent when introduced into new habitats (so-called "alien invasive species") such as water hyacinth Eichornia crassipes, water lettuce Pistia stratiotes, and water fern Salvinia molesta, pose a major threat to the efficiency of dams and irrigation systems. These floating plants can form thick mats that cover the surface of the reservoir completely. By shading out phytoplankton and through increased input of organic matter (when they die and sink), they add to oxygen depletion, which in turn has impacts on fish and may have other ecologically detrimental impacts and serious economic implications (Joffe and Cooke, 1999/ENV057). Managing invasive species that threaten dam and water systems in a proactive manner is far more cost efficient than the usual reactive, crisis-driven manner that is expensive and typically has had only limited success.



3.5.2.3 Riparian Vegetation The riparian ecosystem will inevitably change when its adjoining aquatic environment changes. The largest upstream impact of dam construction on riparian vegetation is biomass submergence. In arid locations the shallow groundwater in the vicinity of a reservoir provides opportunity for vegetation that require access to water throughout the year. A study of radial stem growth of coniferous trees near Swedish reservoirs found a significant increase in the variation in growth following dam construction in trees located close to a reservoir where regulation produced short-term (daily and weekly) variation in water levels. Variation in the water levels of reservoirs can have a negative impact on plants in the immediate vicinity of the reservoir. For example, in Sweden, regulated water level fluctuations may exceed 30 m in height. This has resulted in riparian corridors that are several hundred meters wide. However, because the pattern of water level fluctuations is not synchronised with the natural regime, the riparian vegetation cover is extremely sparse and the riparian ecosystem gives the impression of a barren strip across the landscape (Nilsson and Jansson, 1995). The impact of reservoir level fluctuations are directly related to the gradient of the drawndown zone. Where fluctuations are significant, steep gradient drawndown zones are often characterised by baren strips along the reservoirs. With flat gradients much wider areas can be affected, causing both a disappearance of species and the creation of new habitats for amphibians, birds and drawdown-area plants. 3.5.3 Third-Order Impacts on Fauna 3.5.3.1 Invertebrates, Fish, Birds and Mammals Filling of the dam reservoir results in permanent flooding of riverine and terrestrial habitat, and depending upon the topography and habitats of the river valley upstream from the site of the dam, these impacts can vary greatly in extent and severity. For example, the 500-megawatt Chile Pehuenche Hydroelectric Project floods only 400 hectares of land (with minimal damage to forest or wildlife resources) and has no water quality problems. By contrast, the Suriname Brokopondo Dam Project inundated about 160,000 hectares of biologically valuable tropical rainforest and suffers from severe water quality and aquatic weed problems (Ledec et al., 1997). The effects of inundation are especially severe when the reservoirs are situated close to mountains, in dry areas, or at higher latitudes where the river valleys are usually the most productive landscape elements. Due to impoundment, all terrestrial animals disappear from the submerged areas and populations decrease within a few years in proportion to the habitat area that is lost (Nilsson and Dynesius, 1994). Flooding can result in both local and global extinctions of animal and plant species. Particularly hard hit are the species dependent upon riverine forests, and other riparian ecosystems, and those adapted to the fast-flowing conditions of the main river course (McAllister et al. 1999). Dams also serve as a physical barrier to movement of migratory species, notably fish. This prevents broodstock from reaching their spawning grounds during the breeding season, resulting in massive failure of recruitment and eventual extinction of the stock above the dam. Dams in coastal locations prevent fingerlings and juveniles migrating from brackish water in breeding and nursery areas from reaching freshwater habitats upstream, leading to similar impacts (Bernacsek, 1999). This issue is dealt with further in Section 3.7.3. Flooding of the dam impoundment creates a new ecosystem, which can vary enormously in ecological value and productivity according to the physical and biological characteristics of the site and the management regime of the dam. Reservoirs have been described as an �ecological hodgepodge� (Helfman in preparation). When a dam is built, some riverine species trapped behind the structure



survive although most of the lotic species cannot tolerate the lentic conditions. Most of the riverine fish stay close to the shores of the reservoir, the mouth of tributaries and in shallows. The pelagic and deep water is poorly used unless fishes adapted to these conditions were present before the reservoir was formed. Exotic species are often introduced to fill these vacant niches and these increase the number of species. To determine their effectiveness in providing environmental services and enlarging biodiversity requires a rigorous examination of initial biophysical conditions and comparison with altered conditions especially in terms of species diversity and presence of indigenous species. During the first years of submergence, reservoirs may experience an initial increase in aquatic productivity as a result of nutrients released from decomposing plant biomass. The species of colonising fish capable of rapidly using this new and abundant food will tend to dominate until biomass decomposition has stabilised. Submerged vegetation provides habitats favourable to some invertebrates, which attract in their wake fish capable of deriving benefit from them, followed by predatory species that feed off these fish (De Silva, 1988; CIGB, 1985). Fish catch undergoes an �evolution� in terms of quantity during the first 10 years after the dam closure. Typically (and assuming there is sufficient fishing effort), catch rises very quickly to a peak level 3-5 years after dam closure and then declines to a more-stable level thereafter. This is a normal feature of almost all reservoirs throughout the world and should not be misconstrued as fishing effort-induced stock depletion or that the reservoir is losing its productivity (Bernacsek, 1997). However, it should not be assumed that reservoir fish biomass will in all cases exceed pre-dam river system biomass. In addition to their importance for fisheries many reservoirs are also important for waterbirds (see Section 3.7.4.) and other wildlife, in particular in drier regions. 3.6 Downstream Impacts on Rivers, Floodplains and Deltas Rivers are part of the hydrological cycle and it is the variable nature of runoff processes that give rivers their dynamic characteristics. The ecological integrity of river ecosystems is dependent on the variation in flow regime to which they are adapted. Floods cause hydraulic disturbance that determines the composition of biotic communities within the channel, the riparian zone and the floodplain (Junk et al., 1989; Webb et al., 1999). The spatio-temporal heterogeneity of river systems is responsible for a diverse array of dynamic aquatic habitats and hence ecological diversity, all of which is maintained by the natural flow regime. It is flooding and the consequent transfer of material that makes rivers and floodplains among the most fertile, productive and diverse ecosystems in the world. Floodplain communities are characterised by resilience and the ability to respond quickly to changing hydrological conditions. The rich productivity of floodplains allows them to sustain large populations of organisms that are interdependent on one another. Regular floods keep the vegetational successions in young, productive stages, creating excellent conditions for an abundant wildlife. The diverse vegetation favours animal diversity. Consequently, floodplains are also rich in species endemic to small geographical areas. Coastal marine wetlands are often highly dependent on inputs of freshwater and associated nutrients and sediments from rivers. Coastal wetlands are ecologically and environmentally diverse because of the gradual and often fluctuating dynamic boundaries between salt, brackish and freshwaters. Salt water may penetrate considerable distances upstream, but boundary patterns vary with flow regimes and landscape forms. These patterns influence not only vegetation, but also animal behaviour, such as the extent to which marine species can range into the food-rich wetlands. Dams constitute obstacles for longitudinal exchanges along fluvial systems. Dams not only alter the pattern of downstream flow (i.e. intensity, timing and frequency) they also change sediment and nutrient regimes and alter water temperature and chemistry. These changes and others directly and indirectly influence a myriad of dynamic factors that affect habitat heterogeneity and successional



trajectories and, ultimately the ecological integrity of river ecosystems. The changes induced by large dams may affect ecosystems and the people who depend on them for tens to thousands of kilometres downstream. In this section the first-, second-, and third-order impacts on downstream ecosystems are summarised. 3.6.1 First-Order Impacts on Ecosystem Driving Variables 3.6.1.1 Daily, Seasonal and Annual Flows In general, discharge control resulting from the development and operation of storage dams changes flow variability downstream from the dam. For major floodplain rivers, dams may increase flood peaks by altering the timing of the floodpeak to coincide with floodpeaks from tributaries downstream. Peak discharges can also increase when reservoirs are used for generating peak power. In most cases, however, the magnitude and timing of flood peaks is reduced by storage dam development and operation. The effect of a reservoir on individual flood flows depends on both the storage capacity of the dam relative to the volume of flow and the way the dam is operated. Reservoirs having a large flood-storage capacity in relation to total annual runoff can exert almost complete control upon the annual hydrograph of the river downstream. However, even small-capacity detention basins can achieve a high degree of flow regulation through a combination of flood forecasting and management regime. An example of the changes in average annual flow regime following dam construction on the Murray river (Australia) is shown in Figure 3.5. Discharge close to the Yarrawonga weir has no resemblance to natural flow pattern. At the mouth of the river, the timing of the annual peak discharge under natural conditions is similar to altered conditions (Figure 3.5b). However, river discharge is reduced to 21 % and especially medium size flow peaks are affected. At Albury, a seasonal inversion of river flows is observed due to releases for rice, dairy and orchard irrigation (Figure 3.5b). Reduction in flow velocity in weir pools is considered a key cause of the decline in silver perch in Murray river basin. A decline in river mouth wetlands due to reduced flows is also observed (MDBMC 1995). a)

0

200

400

600

800

1000

Jan

Feb

Mar Ap

r

May Jun

Jul

Aug

Sep

Oct

Nov

Dec

Flow

(GL)

Natural Conditions Current Conditions



b)

Figure 3.4: Comparison of pre and post impoundment flows in the Murray River, Australia:variation in the average monthly flow at a) Albury (2225 km from the mouth) and b) at Barrages (1 km from the mouth). Source: Murray-Darling Basin Ministerial Council, 1995. A consequence of reduced flood peaks is reduction in the frequency of overbank flooding and reduced extent of flooding when it does occur. For example, in the Hadejia-Nguru wetlands (Nigeria) annual flooding prior to construction of dams for irrigation was typically about 3,000 km2 and this was reduced to less than 1000 km2 after construction (Hollis et al., 1993). Reduced floodplain inundation and altered hydrology downstream of dams may reduce groundwater recharge in the riparian zone, resulting in lowering of the groundwater table, with consequent impacts on riparian vegetation. Equally there is a direct and significant relationship between flood extent and the number of wintering ducks in these wetlands (WWF 2000/ENV224). A range of operational procedures can result in fluctuations in discharge that occur at non-natural rates. Hydroelectric power and irrigation demands are the most usual causes, but peak-discharge waves have been utilised for navigational purposes and to meet recreational needs (e.g. white water kayaking and rafting). For many purposes, so called �pulse releases� are made regularly (e.g. daily releases through power turbines which reflect diurnal variation in power demand). Downstream from the West Point Dam (USA), discharge ranges from 14 m3s-1 during low flow generation to 445 m3s-1 during peak generation, resulting in changes in stage height of more than 2 m. The pattern of daily fluctuation in the Colorado River is shown in Figure 3.5. As hydropower represents one of the most easily activated form of peaking power available to most national electricity grids, these kind of fluctuations are frequently associated with hydropower dams.

0

300

600

900

1200

1500

1800

Jan

Feb

Mar Ap

r

May Jun

Jul

Aug

Sep

Oct

Nov

Dec

Flow

(GL)

Natural Conditions Current Conditions



-

5 ,0 0 0

1 0 ,0 0 0

1 5 ,0 0 0

2 0 ,0 0 0

2 5 ,0 0 0

1-Se

p

3-Se

p

5-Se

p

7-Se

p

9-Se

p

11-S

ep

13-S

ep

15-S

ep

17-S

ep

19-S

ep

21-S

ep

23-S

ep

25-S

ep

27-S

ep

29-S

ep

1-O

ct

D a y

Stre

am F

low

(cub

ic fe

et p

er s

econ

d)

Figure 3.5: Daily Streamflow Variations in the Colorado River at Lee's Ferry in September. Peak flows are associated with the power generation between 14.00 and 19.00 daily, with minima at 04.00 am, and the fluctuation in demand also varies from day to day. (U.S. Bureau of Reclamation, Upper Colorado Region, 2000) In addition to altering the flow regime of rivers, dams also affect the total volume of runoff. These changes may be either temporary and permanent. Temporary changes arise primarily from filling the reservoir, which may take several years where reservoir storage greatly exceeds the mean annual runoff. Permanent changes occur because: i) water is removed for direct human consumption and not returned to the river (e.g. for

irrigation or interbasin transfers); ii) water is lost from the reservoir through evaporation � worldwide it is estimated that

evaporation from reservoirs is of the order of 188 km3 y-1, which equates to more than 8% of the total human consumption of freshwater (Shiklomanov 1999).

The hydrological effects of a dam become less significant the greater the distance down stream (i.e. as the proportion of discharge from the uncontrolled catchment increases). The frequency of tributary confluences below the dam and the relative magnitude of the tributary streams, largely determine the length of river affected by an impoundment. Catchments in semi arid and countries with significant storage may never recover their natural hydrological characteristics even at the river mouth, especially when dams divert water for agriculture or municipal water supply. Flow regimes, including volume, duration, timing, frequency and lapse time since last flooding, are the key driving variables for downstream aquatic ecosystems and are critical for the survival of communities of plants and animals living downstream. Small flood events may act as biological triggers for fish and invertebrate migration, major events create and maintain habitats, and the natural variability of most river systems sustains complex biological communities that may be very different from those adapted to the stable flows and conditions of a regulated river. It should also be noted however, that natural flood events can also be detrimental to ecosystems. After the Saguanay flood (Canada) in 1996, for example, salmon habitats had to be restored over a large area (Gaétan pers. com.). Changes in river discharge can have significant effects on downstream groundwater resources. A reduction in flooding can considerably reduce the amount of recharge to downstream aquifers.



3.6.1.2 Water Quality Water storage in reservoirs induces physical, chemical and biological changes in the stored water. As a result the water discharged from reservoirs can be of a different composition to that flowing into the reservoir. Reservoirs act as thermal and chemical regulators so that seasonal and short-term fluctuations in water quality are altered. The salinization of water below dams in arid climates (arising from increased evaporation) is particularly problematic and is exacerbated in areas of marine sediments and where saline drainage water from irrigation schemes is returned to rivers downstream of dams. Salinization has also proved to be a problem on floodplain wetlands in the absence of periodic flushing and dilution by flood water. If sufficiently high and prolonged, elevated salinity will affect aquatic organisms (Hart et al., 1991). Water temperature is an important quality parameter for the assessment of reservoir impacts on downstream aquatic habitats because it influences many important physical, chemical and biological processes. In particular, temperature drives primary productivity. It has been proposed that thermal changes caused by water storage have the most significant effect on in-stream biota (Petts, 1984). Temperatures downstream of the dam may be affected by the reservoir level from which the discharge is drawn, e.g. cool deep temperatures or warm surface temperatures. In New South Wales (NSW), cold water pollution impacts on average a river stretch of 300 km below each dam with water temperatures 5 degrees or more below normal. The total amount of river stretch affected in NSW amounts to 2650 km (Lugg 2000). Changed temperatures may affect spawning, growth rates and length of the growing season for many species. For example juvenile silver perch grown in the cold water released from Burrendong Dam, Australia, increased only 16% in weight over one month compared to a 112% increase in water warmed to natural levels (Blanch 1999/ENV204). In the case of the Gariep Dam in South Africa, for example, the temperature changes due to impoundment extend for 130 to 180 km downstream (Davies 1999). Even without stratification of the storage, water released from dams may be thermally out of phase with the natural temperature regime of the river. The Hume dam on the Murray River, Australia alters the thermal regime of the river and its effect is still discernible 200 km downstream (Walker, 1979). Water temperature changes have often been identified as a cause of the reduction in native species of fish, particularly impacting spawning success negatively (Petts, 1984). Cold-water release from high dams of the Colorado River is still measurable 400km downstream and this has resulted in a decline in native fish abundance (Holden and Stalnaker, 1975). The fact that various introduced trout species replaced some twenty native species of fish has been attributed to the change from warm water to cold water. The quality of water released from a stratified reservoir is determined by the elevation of the outflow structure relative to the different layers within the reservoir. Water released from near the surface of a stratified reservoir is often well-oxygenated, warm, nutrient depleted water. In contrast water released from near the bottom of a stratified reservoir is often cold, oxygen-depleted, nutrient-rich water which may be high in hydrogen sulphide, iron and/or manganese. Water depleted of dissolved oxygen is not only a pollution problem in itself, affecting many aquatic organisms (e.g. salmonid, fish that require high levels of oxygen for their survival), but one that may be exacerbated because such water has a reduced assimilation capacity and so a reduced flushing capacity for domestic and industrial effluents (ICOLD, 1994). The problem of low dissolved oxygen levels is sometimes mitigated by the turbulence generated when water passes through turbines. Water passing over steep spillways may become supersaturated in nitrogen and oxygen and this may also be fatal to fish immediately below a dam (ICOLD, 1994; Fidler and Miller 1997: Bouck, G.R. 1980). This is known as the gas bubble disease. It is for example a problem on the Columbia river (USA), where very high dams in the upper catchment generate high total dissolved gases that are not dissipated downstream (Bell and DeLacy 1967 in Bizer 2000).



Davies (1999) argues that some of these key water quality parameters can recover to their natural levels. If the river flows freely for long enough, they can be �reset�, although each has its own �recovery distance�. Some, such as depleted oxygen levels, may recover within several hundred metres. Others, such as temperature, may take hundreds of kilometres. 3.6.1.3 Changes in Sediment Loads Under natural conditions sediment feeds floodplains, creates dynamic successions, and maintains ecosystem variability and instability (Petts and Amoros 1996). Changes in sediment transport have been identified as one of the most important environmental impacts of dams. The reduction in sediment transport in rivers downstream of dams not only has impacts on channel, floodplain and coastal delta morphology (section 3.6.1.4), and so alters habitat for fish and other groups of plants and animals, but through changes in river water turbidity may effect biota directly. For example, plankton production is influenced by many variables, including turbidity. Turbidity interferes with photosynthesis and algal development may be attenuated by the presence of suspended inorganic particles. If turbidity is reduced, as a consequence of impoundment, plankton development may be enhanced and may even be stimulated to appear in new sections of rivers. The selective release of highly turbid waters from a reservoir is a technique often used to reduce sedimentation. Sediment sluicing involves drawing down a reservoir at the start of the flood season and then allowing as much sediment-laden water as possible to pass through the dam before it has a chance to settle. The sudden release of tonnes of sediment can be disastrous for some biota. For example, the introduction of large quantities of fine silts and clays into permeable gravel substrates can have a catastrophic effect on fish eggs and fry. Thus, even though reservoirs generally trap sediments, reservoir operations can result in extreme and unnaturally high concentrations of sediment, which may produce a major stress effect on downstream aquatic ecosystems. Contaminated sediments in particular form a potential threat to downstream ecosystems if sediment flushing is carried out. Reservoirs tend to serve as sediment traps because river velocities and therefore carrying capacities for particles decrease in reservoirs. However, sometimes, fluctuating water levels in reservoirs erode the shores and add to the turbidity of the reservoir discharge. Furthermore, the selective release of highly turbid waters from a reservoir is a technique often used to reduce sedimentation. 3.6.1.4 Changes to Channel, Floodplain and Coastal Delta Morphology Complex relationships exist between channel form and processes. In general the frequency of flood discharges and the magnitude and particle-size distribution of the sediment load are the dominant controls of channel and floodplain morphology. Reservoirs alter the processes operating in the downstream river system by isolating upstream sediment sources, reducing the frequency of floods and regulating the flow regime (section 3.6.1.1). A unique combination of climate, geology, vegetation, size of impoundment and operational procedures produce the effect of any individual dam upon the fluvial processes downstream. Hence, a wide range of geomorphological responses can be generated by river regulation. Some physical changes caused by dams are immediate and obvious while others are so gradual that they may go unrecognised by humans using the river for many years. Three examples of these slow and not always intuitive impacts are: • Reduced sediment transport can result in lowering of the riverbed downstream and deepening of

the channel as a result of sediment starvation. This channel incision impacts the frequency of floodplain inundation, as the deeper channel requires a higher discharge to overtop its banks and spill out over the floodplain.;



• A reduction in lateral migration of the river channel can reduce the recruitment of spawning gravel from the floodplain. Lack of channel avulsion and bank cutting eliminates coarse sediment recruitment (Dietrich 1999/ENV082);

• Where non-transportable materials are present in the bed sediments, the selective transport of the smaller sediment sizes results in the formation of a coarse sediment layer at the surface that protects the underlying material from erosion, a phenomenon known as channel armouring. A single grain thickness of coarse material may effectively prevent degradation although rare high magnitude floods may disturb this surface layer. Degradation occurs most rapidly in the upstream reaches closest to the dam so the armouring and degradation shifts progressively downstream.

Channel Erosion and Sedimentation The geomorphological effects of changes in flow and sediment regime have been analysed by many, for example Galay (1983), Williams and Wolman (1984) and Carling (1996). If the post-regulation flows remain competent to move bed material, the initial effect is degradation downstream from the dam, because the entrained sediment is no longer replaced by material arriving from upstream. According to the relative erodibility of the streambed and banks, the degradation may be accompanied by either narrowing or widening of the channel. A result of degradation is a coarsening in the texture of material left in the streambed; in many instances, a change from sand to gravel is observed and, in some, scour proceeds to bedrock. Channel degradation below a dam persists until the reduction of channel slope reduces the flow velocity below the threshold for sediment transport. However, degradation is rarely able to progress freely. It is complicated by interrelated hydraulic, sedimentological and biotic factors. For example, degradation may be limited by the local hydraulic conditions within the channel: the interaction of a low channel slope, large cross-section and rough boundary can reduce flow velocity below the threshold for sediment transport. Consequently on many rivers these effects are constrained to the first few kilometres or tens of kilometres below the dam. Degradation of up to 7.5 m has been observed on large rivers immediately below the dam and decreasing downstream (e.g. the Colorado below the Hoover Dam). Typically, 1-3 m of degradation occurs within a decade or two of regulation (Church, 1995). Further downstream, increased sedimentation (aggradation) may occur because material mobilised below a dam and material entrained from tributaries cannot be moved so quickly through the channel system by the regulated flows. Channel widening is a frequent concomitant of aggradation. Most degradation is observed during the first 10-15% of the period of adjustment as a certain armouring and stabilisation starts to occur (Brookes 1996). Thus both channel erosion and sedimentation take place in response to dam construction and operations. Floodplains Damming a river can alter the character of floodplains as the reduction in high-magnitude flows reduces the number of occasions and extension of floodplain inundation. In this sense the river becomes divorced from it floodplain. Effects on floodplain ecosystems are specifically critical as they often are matured systems with a large biological diversity and complicated foodweb structures that are difficult to restore once lost (if at all). In some circumstances the depletion of fine suspended solids reduces the rate of overbank accretion so that new floodplains take longer to form and soils remain infertile. In other circumstances channel bank erosion results in loss of floodplains. For example, between 1966 and 1973, some 230 ha of land were lost from 10% of the total bank length of the Zambezi below the Kariba dam. Erosion was particularly pronounced at alluvial sites with non-cohesive sandy bank materials and was attributed to: the release of silt free water; the maintenance of unnatural flow-levels, sudden flow fluctuations, and out-of-season flooding (Guy, 1981). However, in some places the reduction in the frequency of flood flows and the provision of stable low flows may encourage vegetation encroachment which will tend to stabilise new deposits, trap further sediments and reduce floodplain erosion. Hence, depending on specific conditions, dams can either increase or decrease floodplain deposition and erosion.



Coastal deltas In contrast to the impact on river and floodplain morphology, where aggradation may occur, impounding rivers invariably results in increased degradation of at least part of coastal deltas, as a consequence of the reduction in sediment input. For example, the slow accretion of the Nile Delta was reversed with the construction of the Delta Barrage in 1868. Today, other dams on the Nile including the Aswan High Dam have further reduced the amount of sediment reaching the delta. As a result much of the delta coastline is eroding at rates of up to 5-8 metres per year, but in places this exceeds 240 metres per year (Khafagy and Fanos 1993; AbdelMegeed and Aly Makky 1993; Stanley and Warne 1993). Similarly, erosion of parts of the Rufiji Delta, by up to 40 metres per year, is attributed to the construction of dams (Horrill, 1993). The consequence of reduced sediment may also extend to long stretches of coastline eroded by waves which are no longer sustained by sediment inputs from rivers. It is estimated that the entire coastlines of Togo and Benin are being eroded at a rate of 10-15 metres a year because the Akosombo Dam on the Volta River in Ghana has halted the sediment supply to the sea (Bourke 1988). Another example is the Rhone River, where a series of dams retain much of the sediment that was historically transported into the Mediterranean and fed the dynamic processes of coastal accretion there. It is estimated that these dams and associated management of the Rhone and its tributaries have reduced the quantity of sediment transported by the river to 12 million tons in the 1960s and only 4-5 million tons today. This has contributed to erosion rates of up to 5 meters per year for the beaches in the regions of the Camargue and the Languedoc (Balland 1991), requiring a coastal defence budget running into millions of dollars. Further consideration of these issues is given in Annex 4. 3.6.2 Second Order Impacts on Primary Production 3.6.2.1 Plankton and Periphyton The introduction of a reservoir into a river system as a result of impoundment can markedly alter the plankton component of the river system below the dam. Dams affect the plankton component of the river system in two ways: 1. by changing the conditions affecting the development of riverine plankton (e.g. through

modification of the flow regime and alteration of chemical, thermal and turbidity regimes), and 2. by usually, but not always, augmenting the supply of plankton into the downstream system. These changes will affect not only the total plankton present, but also plankton assemblages. Three factors govern the contribution of lentic plankton to the river downstream: the rate of water replacement within the reservoir (i.e. retention time); the seasonal pattern of lentic plankton development, and the character of outflows from the reservoir. Pulses of plankton output from reservoirs are often linked to season, hydrological conditions, nutrient supply and reservoir operation. The flood mitigating characteristics of dams tend to promote the maintenance of higher than natural plankton populations within regulated rivers, by both sustaining populations released from the reservoir and promoting conditions for plankton development. For example, flow regulation imposed by the Eildon Reservoir, Murray River, Australia has allowed increased development of phytoplankton within backwaters, billabongs and fringing reed beds (Shiel, 1978). Furthermore, dams tend to enhance plankton development through temperature moderation, reduction of turbidity and reduction of effluent dilution (from incoming downstream tributaries etc.).



Within impounded rivers, in temperate climates, the maintenance of higher summer discharges, the reduction of flood magnitude and frequency, reduced turbidities and the regulation of the thermal regime (i.e. higher winter temperatures) often promotes algal growth (Petts, 1984). Moderately swift currents and stable flow favour the growth of periphyton, but the effect of flow regulation on substrate stability may be the most important control. The periodic disruption of periphytic communities under, natural, variable flow conditions may be eliminated, or decreased in frequency, as a result of flow regulation. This allows the full development of a periphyton assemblage, at least in channels of relatively steep slope where moderate current speeds can be maintained. Downstream from deep release reservoirs the composition of the attached algae and the proportion of the substrate covered changes as temperature, turbidity and substrate stability vary in response to tributary and anthropogenic inputs. Typically, algal growth occurs in the channel immediately downstream from dams, because of the nutrient loading of the reservoir releases, and diminishes downstream due to processes of self-purification. Increased algal density has been observed immediately below the Veyriers dam, on the Fontaulière River (France). However, although algal biomass was up to 30 times greater than at an upstream reference site, species composition was considerably altered. The differences have been attributed to nutrient pollution, lowered water temperature, flow constancy and substrate stability (Valentin et al., 1995). 3.6.2.2 Growth of Aquatic Macrophytes Water depth and light penetration are important controls upon the composition and spatial patterns of higher plants. Together with current velocity and the susceptibility of the substrate to scouring, they are the dominant controls upon plant distribution. Thus it is the influence of dams on these factors that tends to dominate their impact on aquatic plants. Of particular significance is the often general increase in bed stability downstream from dams. Compared with the situation in the natural river, the root systems of plants experience reduced effects of scour, the plants themselves suffer less stress from high discharges and the rate of channel migration is reduced, so that an area of the channelbed available for the development of aquatic plants can be stabilised. For example, in the years since the creation of Lake Kariba, flow regulation has allowed the rapid development of rooted plants (Panicum repens and Phragmites mauritanus) within the Zambezi (Jackson and Davies, 1976) where previously there were unstable sandbanks. Flow regulation not only decreases the frequency of high flows and inhibits bed-material movement, but also induces the deposition of finer sediments where supplies are available from tributary or effluent sources. Channel sedimentation, particularly involving nutrient-rich silt, can markedly alter plant distributions. For example, sedimentation is often associated with the invasion and spread of Zannichellia palustris, which traps further sediments as it develops. The elimination of high discharges to flush systems has allowed the extensive development of the aquatic weeds Water Hyacinth Eichornia. crassipes and Water Fern Salivinia molesta in both Africa and Australia. E. crassipes infested the lower reaches of the Fitzroy River, Australia after upstream dam construction stabilised flows thereby reducing floods and preventing salt water incursions to the upper tidal reaches (Mitchell, 1978). These growths may be supplemented by the discharge of floating weeds from infested reservoirs. Thus it is estimated that 150 000 S. molesta mats per hour, supplied by Lake Kariba, passed the Luanga confluence on the middle Zambezi in January 1974 (Davies, 1979). Colonisation by reeds of 41,000 ha of riverbed has occurred as a result of stablised flows on the Orange River, South Africa (Davies 1999).



3.6.2.3 Riparian Vegetation The characteristics of riparian communities are controlled by the dynamic interaction of flooding and sedimentation. Many riparian species are dependent on shallow floodplain aquifers that are recharged during regular flood events. Dams can have significant and complex impacts on downstream riparian plant communities. An important downstream manifestation of river impoundment is the loss of pulse-stimulated responses at the water-land interface of the riverine system. High discharges can retard the encroachment of true terrestrial species, but many riparian plants have evolved with, and become adapted to the natural flood regime. Species adapted to pulse-stimulated habitats are often adversely affected by flow-regulation and invasion of terrestrial weeds in these habitats is frequently observed (Malanson 1993). Typically riparian forest tree species are dependent on river flows and a shallow aquifers. Therefore the community and population structure of riparian forests is related to the spatial and temporal patterns of flooding at a site. For example, the Eucalyptus forests of the Murray floodplain, Australia, depend on periodic flooding for seed germination and regeneration has been curtailed by headwater impoundment (Walker, 1979). Conversely, artificial pulses generated by dam releases at the wrong time � in ecological terms � have been recognised as a cause of forest destruction. For example, Acacia xanthophloea is disappearing from the Pongolo system below Pongolapoort Dam, South Africa as a result of mis-timed floods (Furness, 1978). The direct loss of annual silt and nutrient replenishment as a consequence of upstream impoundment is thought to have contributed to the gradual loss of fertility of formerly productive floodplain soils. It has been shown that given sufficient time after dam construction, riparian forest vegetation may be replaced by forest types more characteristic of unflooded upland areas (Thomas, 1996). Similar effects caused by the decoupling of basin hydrology from riparian vegetation (i.e. caused by changes in both high and low flow regimes) have been documented in the USA (e.g. Crawford, et al., 1994; Rood et al., 1995; Miller et al., 1995; Johnson, 1992). A study in Sweden indicated that both storage reservoirs and run-of-river impoundments permanently altered and reduced the diversity of riparian vegetation. In comparison to natural river reaches there were one-third fewer species around storage reservoirs and 15% fewer species near run of river sites (Nilsson et al., 1997). The Kariba dam has reduced downstream flood magnitudes within the Zambezi valley by about 24% (Masundire, in press). Within Zimbabwe�s Mana Pools National Park, flood extent has declined since the construction of Kariba Dam, reducing regeneration of the floodplain woodlands (Anonymous 1997). 3.6.2.4 Delta and Coastal Vegetation Reduction in streamflow can also have considerable impacts on vegetation in downstream delta and coastal areas. Dam construction and operation in the Indus basin, for example, has reduced flow by more than 80% (McCully 1986). With the increased abstraction of water upstream, the quantity of silt reaching the delta has been reduced. Especially impacted are estuarine mangroves that once covered over 1 millon ha. The sediment brought down to the Delta is now estimated at about 60 Mt per year, about one fifth of original quantities. The active delta is only 10% of its original area and the reduction in the sediment discharge has meant that the balance between erosion due to high energy waves and sediment deposition has changed towards erosion. Mangrove forest needs sediment as part of its habitat renewal mechanism that provides direct benefits to people such as fuel, fodder and fibre and forms rich nursery grounds for fish. The reduced freshwater and sediment flow plus human encroachment contributes to further mangrove degradation. These changes have even affected the total site biodiversity as mangrove species in the Delta have decreased from the eight recorded species to a virtually mono-specific mangrove stand (WCD Tarbela Case Study 2000).



3.7 Third-Order Impacts on Fauna 3.7.1 Freshwater Species Diversity Changes Only a modest fraction � perhaps 10% � of the planet�s species have been discovered by science, named and classified: the known species. Of the 1.87 million recorded species of plants, animals, and micro-organisms, 44 000 or 2.4% occur in freshwater, 14.7% in the sea, and 77.5% on land (Box 3.4). However, the diversity of freshwater species is 10% higher than that on land when the fact that freshwaters comprise only 0.8% of the surface area of the planet is taken into account. The disproportion is even greater for the fishes; about 42% of known fish species occur in the tiny fresh water area, compared to 58% in the far greater marine area. Freshwater fish species diversity generally increases at lower latitudes. This has specific consequences for dam construction impacts at these latitudes as their impact on species loss can be potentially much higher than at higher latitudes (Figure 3.6).

Figure 3.6: Fish species richness decreases at higher latitudes indicating that dam construction in tropical regions could potentially have more impacts than at higher latitudes (WCMC 1998).



Species populations may be located along a spectrum from common to rare. Declining species status can be measured in simple terms by lowered populations, extirpations (loss of populations from a part of the species range), or extinctions (loss of all individuals of a species). IUCN (1994) has developed a scientifically objective means for assessing such populations, and many governments have enshrined species status considerations in their national legislation.

Box 3.4: Species richness of the planet�s major environments (Source: McAllister et al. 1997) Environment % area of % of known Relative species Planet living richness surface species (%species / %area) Fresh water 0.8% 2.4% 3.0 Terrestrial 28.4% 77.5% 2.7 Marine 70.8% 14.7% 0.2 Symbiotic N.A. 5.3% N.A.

According to IUCN's 1996 Red List, 1 107 bird species (11% of the total) are threatened and 104 (1%) are extinct. Among the more threatened of bird groups are the aquatic rails and cranes with 54 species threatened, and the partially aquatic kingfishers and bee-eaters with 11.5% threatened, while 18% of the grebes are threatened. Extinct aquatic birds include the Colombian Grebe (Podiceps andinus) and the Atitlan Grebe (Podilymbus gigas). The IUCN Red List of Threatened Plants concluded that 33 375 species or 13.8% of the world�s 242 000 vascular plant species are threatened, and 376 are extinct. At regional scale few detailed data are available. The exception is North America where freshwater animals have been shown to be the most endangered species group on the continent, dying out five times faster than those that live on land, with a rate similar to the loss of rainforest species. Since 1900, at least 123 species have been lost from North America�s waters. A further 190 fish, 27 amphibian, 35 reptile, 84 bird and 94 mammal species are currently threatened with extinction, as 51% of species decline in numbers (Riccardi and Rasmussen, 1999). In the United States alone data on the conservation status of freshwater species groups give an alarming picture: • 67% of freshwater mussels are vulnerable to extinction or are already extinct • 303 fish species � 37% of the US freshwater fish fauna � are at risk of extinction • 51% of US crayfishes are imperilled or vulnerable • 40% of amphibians are imperilled or vulnerable • at least 106 major populations of salmon and steelhead trout on the west coast have been

extirpated, and an additional 214 salmon, steelhead trout, and sea-run cut-throat trout stocks are at risk of extinction (Nehlsen et al. 1991).

While these figures give an indication of the scale of the threats to freshwater biodiversity, the information constraints highlighted in section 3.4. mean that there are limited data available on the specific impacts of dams on species diversity. However, useful studies have been carried out on some groups and these can serve as indicators. The following sections on molluscs, fish and waterbirds therefore serve to illustrate how dams impact upon the biology of individual freshwater species and thus lead to changes in species diversity.



3.7.1.1 Invertebrates, Fish, Birds and Mammals The preceding sections have shown that when dams are constructed the variability in water discharge over the year is reduced; high flows are decreased and low flows may be increased. Reduction of flood peaks reduces the frequency, extent and duration of floodplain inundation. Reduction of channel-forming flows reduces channel migration. Truncated sediment transport (i.e. sedimentation within the reservoir) results in complex changes in degradation and aggregation below the dam. These changes and others directly and indirectly influence a myriad of dynamic factors that affect the diversity and abundance of invertebrates, fish, birds and mammals downstream of dams. In light of the information constraints outlined in section 3.4., comprehensive data are not available. It is only therefore possible to provide here a first indication of the third-order impacts on ecosystem functioning and productivity. Section 3.7 then examines the specific issues of the impact of dams on species diversity using three more thoroughly-studied groups as indicators. Most aquatic species cannot live for long without water, e.g. those breathing with gills. When a dam closes off river flow, some species may avoid dehydration for short periods, e.g. snails by closing their operculum. Some downstream populations will be reduced but may manage to hang on in pools or tributaries. Survival in such pools may be reduced by predation as individuals are more accessible because they are concentrated in the shallows. These effects can lead to declines in downstream fisheries. Larger aquatic species such as sturgeons, crocodiles and dolphins require minimal flows in which to navigate, feed, etc. Such species may be seriously affected by reduced flows which mean reduction of area of habitat. Habitat reduction may mean simply smaller populations or reduced growth rates, or where populations are already at risk, it may lead to extirpation (loss of a population) or extinction (loss of an entire species). Large woody debris plays an important and until recently unrecognised role in providing fish and food base habitat. Wood contributes to complexity of channel form and habitat in many rivers. In some cases, woody debris is removed from the downstream environment by the storage dam operations. If the original amount of debris input into the river is large compared to the downstream input the impacts of dam construction and operation can be considerable (Kondolf 1999/ENV083/ENV085/ENV088). River-dwelling species have several migratory patterns. These include the well-known anadromous fishes like salmon and catadromous fishes like eels. Adults of the first migrate up rivers to spawn and the young descend, while the reverse occurs with the latter. But many other freshwater fishes move up rivers or their tributaries to spawn, while the glochidia larvae of freshwater mussels hitch rides on host fishes. Migration between marine and freshwater ecosystems and within freshwater ecosystems are known. Dams block these migrations to varying degrees. Biological linkages also extend laterally away from the river, extending the effect of river changes to a band of varying width, parallel to the river. As long as the river flow is sufficient, other wildlife such as deer, antelope and elephants will come to the water, especially in the dry/hot season, for drinking water. These lateral movements can extend to several kilometres from the river. Many wildlife species in a fairly wide strip of land on either side of the river depend upon it, and they may all be affected when the flow of the river is disrupted by the construction of a large dam. The blockage of fish movements upstream is probably the most significant and negative impact of dams on fish survival and biodiversity. Many stocks of Salmonidae and Clupeidae have been lost as a consequence. In the Columbia River alone, more than 200 stocks of anadromous salmonids have been



extirpated. Sturgeon populations in the Caspian Sea now rely on hatcheries, mainly in Iran, since Russian dams block natural spawning migrations. The control of floodwaters by large dams, which usually reduces flow during natural flood periods and increases flow during dry periods, leads to a discontinuity in the river system. This together with the associated loss of floodplain habitats normally has a marked negative impact on fish diversity and productivity. The connection between the river and floodplain or backwater habitats is essential in the life history of many riverine fishes that have evolved to take advantage of the seasonal floods and use the inundated areas for spawning and feeding. Loss of this connection can lead to a rapid decline in productivity of the local fishery and to extinction of some species. An assessment of 66 case studies of the impacts of dams on fish biodiversity concluded that 27% of cases had positive impacts (i.e. increase in species richness) compared to 73% having negative impacts (i.e. decrease in species richness) . Of the latter, 53% were downstream of the dam, affecting upward fish migrations and connections to floodplains. Within regions, negative impacts of this kind are more common in temperate than in tropical zones. In tropical regions, the extent of positive impacts is much greater than in temperate ones, particularly in reservoirs upstream of the dams (McAllister et al., 2000). Where fish biodiversity increases it occurs because the reservoir provides �new� habitat for fish species preferring lentic habitats. In many cases people introduce exotic (i.e. non-native) species to improve fisheries. Fish migrations in the tropics are probably best known in the Neotropical region. Hydroelectric dams in the Amazon basin as a whole have halted the long distance migrations of several species of catfish although the available data are not quantified (Ribeiro et al.,1995). The dams have also interrupted the downstream dispersal of catfish larvae. On the Araguaia-Tocantins River Basin, several species of fish which undergo long distance migrations have been drastically reduced in abundance as a result of dams blocking their routes. Downstream fisheries have been reduced by 70%, probably as a result of recruitment failure. In Africa the recent droughts have made it difficult to differentiate between the effects of reduced flow resulting from dams and from lack of rainfall, for example in the central delta of the Niger River (Läe 1995). However, substantial losses to overall fishery production in river basins have been reported in Africa as a result of dam construction. For example, 11 250 tonnes of fish per year from the Senegal River system were lost following dam construction (Reizer, 1971). A major concern throughout Asia is that movements of migratory fishes along river courses are being blocked by dams. Dams can enhance some riverine fisheries, particularly tailwater fisheries immediately below dams that result from discharge of nutrients (seston) (primarily plankton) from the upstream reservoir. However, discharge of seston is typically attenuated quickly downstream from the dam, with corresponding attenuation of the associated fisheries. If discharge is from the hypolimnion of the reservoir, lowered temperatures in the receiving tailwater can curtail or eliminate warmwater river fisheries and require stocking of exotic coldwater species such as salmonids (assuming that the water is sufficiently oxygenated). Productive tailwater fisheries targeting these coldwater fishes can result but generally require supplemental hatchery programs and introduction of coldwater invertebrates to serve as food for these fish. In North America, yields from cold tailwater fisheries have been recorded for up to 753 kg/ha/year with fishing effort 7-16 times higher than the respective upstream reservoir. Estuarine Impacts. Reduction in freshwater flow can result in an increase in salinity in estuarine areas and upset the complex nature of water currents which in turn can alter fish biodiversity. Increased salinity has occurred in the Nile Delta although the effects on fishes have not been well documented (Aleem, 1972; Stanley and Warne, 1993). Marine fishes were found in higher reaches of the Eastman Estuary after 90% of the river water was diverted to the La Grande River (Canada) (Ochman and Dodson, 1982).



Marine Impacts. Freshwater flows support marine fish production. The effect of reduced freshwater flow is probably greatest in the first year of life of a fish population. Fish abundance is normally determined during the egg and larval stages (Drinkwater and Frank 1994). Thus, although the annual discharge through a hydropower dam may not differ much from unregulated flow, unless water is diverted, the seasonal timing of discharge may be significantly different and have negative impacts on marine fishes. Many marine fishes spawn in estuaries or floodplains generally at times of peak run-off. A decrease in freshwater flow and in nutrients may affect the nursery areas in a number of ways including increasing salinity, allowing predatory marine fishes to invade and reducing the available food supply. These impacts are well illustrated by the effect of the Aswan High Dam on the coastal waters of the Mediterranean (Aleem 1972; Drinkwater and Frank 1994). Here reduction in nutrients transported to the sea has reduced production at all trophic levels, resulting in a decline in catches of sardines and other fish. In the Zambezi delta the impact of modified seasonal flows on shrimp fisheries has been estimated at 10 million dollars per year (Gammelsrod 1992a, 1992b). 3.7.2 Bivalve and Gastropod Molluscs Bivalve molluscs are especially important elements of riverine ecosystems because of their ecosystem functions and economic value.

Box 3.5: Global hotspots for freshwater molluscs River System Species % Endemic Mobile Bay, USA 192 78 Balkans region 190 95 Lake Baikal, Russia ±180 67 Lower Mekong 160 72 Lower Zaire 96 25 Lower Uruguay/Rio de la Plata 93 37 Lake Tanganyika 83 64 Western Ghats, India 71 18 They are also highly endemic, and therefore subject to extinction (Box 3.5). The ecology and life history traits of one group of molluscs, the freshwater mussels (Unionoidea), makes them an important indicator of ecosystem health and of the impact of physical and biological changes in the ecosystem on species diversity. Freshwater mussels are filter feeders requiring a rich and plentiful supply of diatoms, desmids, filamentous algae and other algal species. They are therefore especially vulnerable to the second and third order ecosystem impacts described above. In addition their reproductive cycle may be seriously disrupted by dams. This involves a larval stage (called the glochidia), which is retained in the female brood pouch or gills and released for their intermediate stage as a parasite of a host fish before being transformed to bottom-dwelling juveniles. Dam building activity which blocks migratory fish or changes fish communities can also reduce the reproductive success of the freshwater mussel communities which depend on the fish as glochidial hosts. For example, dam construction at Lake Pepin on the Mississippi River led to the demise of mussels upstream of the dam, as the runs of skipjack herring, their host species, were blocked (Eddy and Underhill, 1974). Unfortunately few of the host fish for mussels have been identified. Stresses associated with dam construction resulting in physical disturbance of the river bed may also cause mussels to prematurely empty their brood pouches of glochidia, resulting in reproductive decline (Howells et al., 1996). The changes in water chemistry that occur in reservoirs may in turn



affect spawning of mussels (Isom, 1971). Bivalve species are very sensitive to water chemistry. Translocation experiments on endangered mussel species in Europe have shown that changes in water chemistry can lead to stress and trigger release of glochoidea during unsuitable flow/water conditions. Changes in temperature may affect spawning of mussels, growth and the duration of the growing season (Isom, 1971). Low dissolved oxygen levels may cause stress to freshwater mussels although some species can withstand brief periods of low oxygen levels. Increased siltation can be a major problem in some areas. The greatest diversity in the prosobranch gastropod fauna in the USA is found in the Mobile Bay river basin, and the Tennessee river basin, although 7% of total taxa are now extinct (Bogan, 1998). Most of the extinctions in this group (38 out of 42 taxa) are in the Mobile Bay fauna, and occurred when the river shoal fauna was impounded and covered by deep standing water and subsequent siltation. Many mussel species also have extended life cycles, some of which span over 100 years, where maturity is delayed until the individual reaches 6�15 years of age (Bauer, 1993; Chesney and Oliver 1998). This can lead to the impression that populations are secure, when in fact no active recruitment is taking place and the populations may well be functionally extinct. Box 3.6 demonstrates the impacts of dams and reservoirs, using data from the USA. Impacts are evident from construction, after construction, and downstream. Species richness of molluscs has declined between 40% and 80% from the original diversity levels in certain USA rivers, over a period of 50 years. The figures in post-dam richness also indicate the stretches downstream of the dam where the bed may be devoid of mussels. Box 3.6: Mollusc species present within reservoir region, USA (Source: Neves, 1999) Reservoir Date

Preimpoundment Richness

Postimpoundment Richness

Norris, Clinch R. (1937) 40 species (1935-37) 12 species (1990�s) Center Hill, Caney Fork (1948) 39 species (pre-1940) 2 species (1993) Cumberland, (1952) 59 species (1947-49) 16 species (1961) Wheeler, Tennessee R. (1936) >60 species (pre-1935) 18 species (1991) Demopolis, Tombigbee R. (1936) 50 species (1933-35) 8 species (1954) Demopolis/Warrior (1954/57) 48 species (pre-1950) 13 species (1972-75)

Stein and Flack (1996) conclude that the current decline of freshwater mussels in the Mississippi Basin will have a detrimental impact upon the entire ecosystem. They point out that the freshwater mussels play an important role in sediment mixing and nutrient recycling, and given their dominance in terms of biomass, their removal could have long-term repercussions that are as yet unknown. They are also a major food source for aquatic vertebrates. Water level fluctuation also affects gastropod species. Brown (1994) described the gastropod diversity of several African reservoirs which are comparable in size to large natural lakes. The outflow from these reservoirs differs from the natural lakes, with most suffering large seasonal draw-down as outflows from the lakes are regulated to ensure that the rainy season floods can be contained. This gives a very unstable littoral zone, which stresses aquatic life at the margins, restricting the number of mollusc species which can survive in the lake. Conversely, stabilisation of flows in the Senegal river following construction of the Manantali and Diama dams allowed colonisation by bilharzia-carrying snails, that were previously absent from the dynamic flood river ecosystem. Riparian habitats also hold unique species of molluscs. Disturbance during the construction phase, especially the destruction of habitats for temporary roads, can lead to loss of these species. These losses may be permanent or temporary, depending on the degree of degradation and the amount of habitat fragmentation. In South America, possible species extinctions have been related to loss of



gallery forest adjacent to rivers which are now submerged following construction of the Salto Grande Dam. The land-snail Anthinus albolabiatus (Jaeckel, 1927) was formerly endemic to gallery forests of the Uruguay River and has been submitted for inclusion in the IUCN Red List of Threatened Animals. 3.7.3 Impact of Dams on Fish Diversity Fish are the most species-rich of all vertebrates. Valid scientific descriptions exist for about 24 600 living species of fishes in 482 families (Nelson 1994). One third of the fish families have at least one member spending at least part of their life in freshwater. Freshwater fish diversity is therefore large compared to other systems since freshwater lakes and rivers account for only 0.8% of the earth�s surface and less than 0.01% of its water. Approximately 10 100 species are entirely freshwater and 2 500 move between the sea and freshwater during their life cycles (Helfman et al.,1997). An indication of fish species diversity in some river systems is presented in Box 3.7. The largest number of species occurs in the tropics and the diversity of fishes, in general, increases from the poles to the tropics. Southeast Asia, South America and Africa have the most freshwater fishes. However, many have not yet been described, so taxonomists are needed to describe unknown species especially in these species-rich areas. It is also important to protect genetically distinct stocks within a species. For example, Ryman et al.,(1995) suggested that it is just as important to protect the intraspecific diversity of the Atlantic salmon as to protect the cichlid flock in Lake Malawi. The 1996 IUCN Red List of Threatened Animals lists 617 freshwater fishes (including euryhaline � salinity-level tolerant � species), about 7% of the known number of freshwater fish species. Studies that take into account the fact that the Red List has evaluated only a fraction of freshwater fishes estimate conservatively that 20% of freshwater fishes are either extinct, endangered or vulnerable; a more realistic estimate might reach 30-35% (Stiassny, 1996). Fish populations are highly dependent upon the characteristics of their aquatic habitat that support their biological functions. Migratory fish require different environments for the main phases of their life cycle: reproduction; production of juveniles; growth; and sexual maturation. The life cycle of diadromous species takes place partly in fresh water and partly in seawater; the reproduction of anadromous species takes place in freshwater; and catadromous species migrate to the sea for breeding purposes and back to freshwater for trophic purposes. There are also migrations of potadromous species, whose entire life cycle is completed within the inland waters of a river system.

The disruption of movement of species upstream has probably been the most significant and negative impact on fish biodiversity and many examples illustrate the point from all regions. Large dams halt long distance migrations and the fish fail to reach their spawning grounds. Many anadromous fish populations such as Salmonidae and Clupeidae (e.g. shads) have died out as a result. The sturgeon populations in the Caspian Sea now rely on stocking from hatcheries (mainly in Iran) as natural spawning migrations were halted by dams built by the former USSR on rivers entering the sea. The best-documented examples of disrupted migrations are from the west coast rivers of the USA, in particular the Colorado and Columbia Rivers. In the Columbia River more than 200 stocks of anadromous salmonids have become extinct. Catadromous species such as Anguillidae have been less affected although adults are often killed in hydroelectric turbines. Eels are not restricted to specific rivers, like salmonids, and can move into new rivers if their path is blocked by a dam (Drinkwater and Frank 1994). Even when fish passes have been installed successfully, migrations can be delayed by the absence of navigational cues such as strong currents. This causes stress on the energy reserves of the fish as anadromous fish such as salmonids do not feed during migration. Mortality resulting from fish passage through hydraulic turbines or over spillways during their downstream migration can be significant. The risk of injury varies according to fish size and species, but typically ranges from <1% for young fry to between 5-20% for 15cm Atlantic Salmon smolt for



example. Mortality in adult fish may reach 100% without special protection measures. Problems associated with downstream migration can also be a major factor affecting anadromous or catadromous fish stocks. Habitat loss or alteration, discharge modifications, changes in water quality and temperature, increased predation pressure, and delays in migration caused by dams are significant issues. Box 3.7: Fish species richness in selected river basins (after: World Bank 1998, WCMC 1998) Watershed/Continent Number of fish species Number of species 100/000

km2 of watershed

Kapuas, Indonesia 320 360 Mekong � Asia 1,200 147 Chao Phrya, Thailand 222 124 Xi Jiang (Pearl), China 290 71 Amazon - South America 3,000 49 Orinoco - South America 318 33 Yangtze � China 322 19 Paraná - South America 355 14 Congo - Africa 900 13 Mississippi � USA 375 12 In Australia, dams have generally resulted in negative impacts upon native riverine fishes while encouraging exotic species. This has been attributed, in part, to disruption of seasonal flood cycles, and to dams acting as barriers to fish movements. The Murray-Darling, which has 84 main reservoirs with capacities of 10 000 ML capacity and over, now has the lowest commercial fish yield per sq km of floodplain of any of the world�s major rivers, although historical catches were comparable (Jackson, 1999). Fish diversity in reservoirs is usually not as extensive as in natural lakes, because natural lakes have more stable conditions under which the fishes evolve. Riverine species have to live under harsher and more variable conditions. During reservoir formation the river and possibly associated wetland areas become inundated. As the reservoir fills, riffles, runs and pools of the river are lost beneath the rising waters leading to the extinction of habitat-sensitive riverine species with tightly defined niche requirements (e.g. species of darter (Percidae) found in streams above dams in the Tennessee River system (Neves and Angermeier, 1990). During construction, downstream flow may be severely restricted, as at Cahora Bassa, Mozambique (Jackson 1999), eliminating the fishes present below the dam. However many fishes can quickly recolonise once a flow is re-established. The filling of reservoirs may take a few months (e.g. Kainji) or years (e.g. Volta, Kariba and Nasser/Nubia), and fishes adapt better to prolonged filling. Reduced number of species in reservoirs may also be an artefact created by inappropriate timing of dam closure and poor control of environmental impacts during dam construction. The initial natural stocking with native species is of high importance in determining the species composition of the stabilised reservoir. If dam closure occurs during the dry season, the number of naturally stocked species will likely be minimised and not be representative of the full complement of fish species which occur in the river all year round. This is because many larger fish species migrate downstream to refuge habitats during the dry season and only migrate upstream into low order tributaries during the rainy season for spawning purposes. The disruption in normal hydrological flows which can occur during the dam construction phase, compounded by excessive erosion and siltation of the river in the vicinity of the dam site, may result in disturbance of fish stocks and migrations, and reduce the magnitude of fish biodiversity and quantity available for initial natural stocking (Bernacsek, 1997).



3.7.4 Dams and Waterbirds Waterbirds, both migratory and non-migratory, are important components of the biodiversity of wetlands throughout the world. This is recognised in international conventions and agreements, which place requirements on Parties to safeguard waterbirds throughout their range and distribution. This is achieved in several ways, notably through the designation of wetlands of international importance for waterbirds through the Ramsar Convention on Wetlands of International Importance, and the development and implementation of flyway-scale migratory waterbird conservation strategies, notably the Bonn Convention African-Eurasian Migratory Waterbird Agreement (AEWA) and the Asia-Pacific Migratory Waterbird Conservation Strategy. Such flyway-scale initiatives recognise the vital need to safeguard the international networks of key sites upon which these birds depend throughout the year for their survival, and to put in place a range of management measures to maintain these populations. Waterbirds may use natural and dammed open water wetlands for breeding, and during their non-breeding seasons for feeding and for roosting. Wildfowl (divers, grebes, cormorants, swans, geese, ducks, coots and rails) are particularly characteristic of open water systems. Large numbers of waterfowl migrate south and south-west from arctic, sub-arctic and boreal breeding areas in Europe, North America and Russia to overwinter in the relatively mild climate of western Europe, Africa, and tropical America. Other major waterbird guilds such as waders (shorebirds) chiefly use the shallow emergent shorelines of such wetlands for feeding during migration staging or wintering. Of 957 Ramsar sites designated by December 1998, 10% included artificial wetland types, compared to 25% including natural lake types (Frazier, 1999). Many of the designated artificial wetlands are dammed sites: of the almost 100 artificial wetlands designated as internationally important, 78 are listed as having water storage areas either as a primary or occurring wetland type. Of these 78 sites 57 were designated either wholly or partly for their internationally important waterbird populations. Nineteen regularly support over 20 000 waterbirds (Ramsar Criterion 5), 13 sites regularly support more than 1% of the biogeographic population of one or more waterbird species (Ramsar Criterion 6), and a further 22 sites meet both of these waterbird criteria (Frazier, 1999)). In inland South Africa for example, almost all permanent waterbodies are dammed sites, constructed for water storage purposes. The total capacity of these impoundments amounts to some 52% of annual run-off. These range from large scale impoundments several kilometres long to many small farm dams. At least 517 major reservoirs were constructed by 1986, along with many tens of thousands of farm dams of a few hectares each in area (Taylor et al., 1999). The overall impact of these many artificial open water bodies has been to greatly increase the year-round availability of permanent lakes in inland South Africa (Cowan and van Riet, 1998) and this has undoubtedly had very major effects on the distribution and numbers of waterfowl in the region. Artificial wetlands are included in many Important Bird Areas (IBAs) identified in South Africa (BirdLife International, in prep.), and at least 12 impoundments support major and important concentrations of waterbirds. Overall large dams have provided increased areas of suitable habitat for several species that favour deep open-water conditions. The suitability of such dammed lakes for other species depends, as elsewhere, on the extent to which they provide areas of fringing emergent vegetation and shallow shorelines, features which generally are found in the upper parts of impoundments. Large dams in South Africa have provided generally beneficial conditions for Pelecaniformes (pelicans, darters and cormorants). They provide suitable habitats for moulting sites for waterfowl: for example at least 70% of the global population of the South African Shelduck Tadorna cana moults at only 23 localities in South Africa, 21 of which are large dams. Dams also provide dry season or drought refuges for many waterfowl species, and breeding sites for many South African waterfowl, including some species of national conservation concern, notably the Pink-backed Pelican Pelecanus rufescens and Caspian Tern Hydroprogne caspia.



While dams in South Africa have increased the amount of suitable year-round habitat for species of waterbird that prefer open-water habitats, and in some cases species that feed along the shallow margins of the dams, the overall waterbird assemblage that naturally occurs in southern Africa has suffered from major negative impacts. These include: ! The loss, on most river systems, of many of the former natural marshes and riverine habitats,

which has impacts on a larger assemblage of species that depend on such habitats than the number of waterbird species that have benefited from the creation of open water dams.

! Major changes to downstream riverine habitats, first by reduction of river flow and removal of

much of its previous seasonal variability, causing changes to sediment movement and stabilising channel morphology. Poor dam capacity management during major floods can also lead to sudden major releases of water, so creating major floods in downstream areas in river systems that have had little or no flood activity for years. This has affected the suitability of the river systems between dams for those species that breed chiefly on unvegetated river banks and sandbanks between river channels.

Few studies have assessed the waterbird assemblage before and after the construction of a large dam. Allan (1999) reported on waterbirds present before and after construction of the large Katse Dam in Lesotho. Of 13 waterbird species present before inundation, two disappeared (including Black Stork, a Red Data Book listed species), two decreased in abundance, seven showed little change in status and two common species increased in abundance. Two widespread open water species colonised the area: Little Grebe and Red-knobbed Coot. Unlike organisms that are unable to move in response to the construction of dams, waterbirds are highly mobile and capable of exploiting such new open water systems and their margins if they provide suitable breeding, feeding and roosting conditions. Hence reservoirs are often used by large numbers of waterbirds. On the evidence of an analysis in temperate regions where natural lakes also occur, the waterbird assemblage using dams is broadly similar to those of natural lakes. Although in some situations (e.g. Switzerland) the species diversity is generally lower and artificial sites generally support the more common and ubiquitous species, some dams (e.g. in the United Kingdom) support an assemblage as large or larger than many natural lakes and provide important habitats for internationally important waterbird populations. However, even in the UK, dams overall are of much less international significance than the natural lake systems, and in this instance are also strongholds of some alien invasive species that are the cause of conservation problems. The before and after responses by waterbird assemblages to dam construction needs further evaluation, as do the characteristics of reservoirs (in comparison with natural lakes) that determine their suitability for different waterbird species. Such characteristics are most likely to include water depth, steepness of shoreline and the presence of fringing vegetation. In many dams the water regime and slopes do not favour colonisation by plants and this creates barren and sterile shorelines, equally unfavourable to a range of bird species. Box 3.8: Dams as Wildlife Habitats While birds may be the most visible and well-surveyed fauna, there are cases of reservoirs also supporting significant wildlife under particular circumstances. For example, the only place the common caiman, Caiman crocodilus is found in Tobago is at Hillsborough Dam, while the Indian mugger, Crocodylus palustris, an endangered species in Pakistan, thrives at Hub Dam (DFID-Mott Macdonald 2000/ENV203). The National Park along the shores of Lake Kariba is also a famous wildlife refuge.



3.8 Cumulative Impacts of Dams Many of the major catchments in the World now contain multiple dams. Within a basin, the greater the number of dams the greater the fragmentation of river ecosystems. It is estimated that 61% of the worlds river basins are highly or moderately fragmented (Table 3.5, Figure 3.7, Dynesius and Nilsson, 1994; Nilsson et al., 2000). Table 3.5: Fragmentation of rivers in 225 basins in the world (Source: Nilsson et al. 2000). # of Basins assessed % of Basins assessed Highly fragmented 83 37% Moderately fragmented 54 24% Unfragmented 88 39% Total Basin assessed 225 100%

Figure 3.7: Fragmentation of rivers in 225 basins in the world (Source: Nilsson et al. 2000). The magnitude of river fragmentation can be very high. In Sweden, for example, only four major (longer than 150 km) and 6 minor (70-150 km) first order rivers have not been affected by dams (Figure 3.8) (Lovggren, 1999).



Figure 3.8: Dams in the river systems of Sweden. Only four major rivers remain undammed (Lovgren 1999/ENV136). River impoundment affects the downstream environment so dams built in the same catchment, either in series (i.e. along the same river) or in parallel (i.e. on different tributaries) will inevitably result in cumulative impacts. A cumulative impact can be defined as the incremental effect of an impact added to other impacts. An individually insignificant impact may, when combined with others, produce a major change within a river ecosystem. The total effect on a river ecosystem of cumulative impacts may be greater than the sum of each individual impact. This is particularly the case for those second and third order impacts that are contingent on a number of lower order impacts. 3.8.1 Conceptual Framework for cumulative impact assessment A fundamental tenet of the river continuum concept (Box 2.1) is that within river ecosystems biotic communities are structured along resource gradients and downstream communities are dependent, at least in part, on upstream processes. The serial discontinuity concept (SDC) is a theoretical construct that perceives dams as major disruptions of longitudinal resource gradients along river courses (Ward and Stanford, 1983; Ward and Stanford, 1995). According to the SDC, dams result in upstream-downstream shifts in physico-chemical parameters that in turn affect biotic patterns. The SDC proposes that dams reset or shift riverine characteristics in predictable ways and that the magnitude of such shifts depend on the variable being considered, the stream order, the position of a dam within a catchment and the dam operation. Riverine characteristics include both the biotic and abiotic characteristics e.g. flow, sediment load, dissolved oxygen, thermal heterogeneity, channel stability, biodiversity etc. In all cases, the dynamic and self-correcting nature of rivers means that with increasing distance downstream, characteristics altered by the dam tend to return to their unregulated level (i.e. baseline). Within this conceptual framework, cumulative impacts can be visualised as occurring when any riverine characteristic, altered as the consequence of one



dam, fails to fully recover, before it is �re-set� as a consequence of either the presence of a downstream dam or the operation of a dam on another tributary. Two parameters can be used to measure the magnitude of biotic and abiotic shifts. These are: i) the discontinuity distance (DD) which refers to the longitudinal shift and ii) intensity (PI) which refers to the impact of regulation relative to its equivalent unregulated position. For any given attribute, cumulative impacts affect both DD and PI. In most cases, cumulative impacts result in an increase in both DD and PI. Figure 3.9, illustrates the SDC and the effect of cumulative impacts. The thick curve represents the conceptualisation of spatial gradients along a natural river system for a hypothetical river characteristic. The arrows on the curves indicate the position of dams situated in headwaters, middle and lower reaches. The thin lines associated with the arrows indicate how the variable changes as a consequence of river impoundment. The upper graph shows the situation where the dams are far enough apart for there to be no cumulative affects. Downstream of each dam the characteristic being considered returns to the unregulated baseline before being reset by the next dam. The lower graph shows the situation where cumulative impacts occur. In this case the dams are closer together and the recovery distance is not attained before the characteristic is reset by the next impoundment. The result is an extended discontinuity distance and increased intensity.

Riv

erin

e ch

arac

teris

tic

Headwaters Lower Reaches

Riv

erin

e ch

arac

teris

tic

Headwaters Lower Reaches

DD

PI

DD

PI

Figure 3.9: The impact of dams on the hypothesised downstream pattern exhibited by a given river characteristic in situationa) with no cumulative impacts and b) with cumulative impacts (modified from Ward and Stanford, 1995). 3.8.2 Case studies on cumulative impacts There has been relatively little research into the cumulative affects of dams and almost none in tropical climates. The most frequently mentioned type of cumulative impact is the combined effects of multiple dams on river discharge and water quality (Kvernevik and Ghazali, 1998). Cada and Hunsaker (1990) investigated the cumulative effects of hydropower development and grouped the



impacts into four potential pathways ranging from simple, additive effects of a single project to synergistic effects arising from multiple projects. In several countries, the importance of cumulative impacts is increasingly recognised. Several of them, most notably the United States and Canada have made efforts to study and define cumulative impacts for incorporation of their impacts assessment into legal guidelines for environmental impact assessment. Consideration of cumulative impacts became a formal requirement in the National Environmental Policy Act in the United States in the late 1970s (Irving and Bain, 1989) and in Canada in 1992 (Bunch and Reeves, 1993; Brink 2000/ENV130). A major constraint on assessing the cumulative effects on higher order impacts is the paucity and low quality of available data (Irving and Bain, 1989). However, research has been conducted that demonstrates cumulative impacts at all three levels of impact caused by impoundment (Boxes 3.10 to 3.12). A third order cumulative impact often cited is that of mortality of migratory fish. On the Columbia River, USA between 5% and 14% of adult salmon are killed at each of the eight dams through which they pass. Consequently, the cumulative mortality is 70% to 90% in every salmon run. Fish mortality is attributed to navigational problems for migrating salmon in the still water behind dams, killing of salmon within dam turbines and increased predation in the warmed still waters along fish ladders. Box 3.9: Example of cumulative affect on first order impacts (i.e. the hydrology) of the Murray River, Australia (after Maseshwari et al., 1995) The flow regime of the Murray River, Australia has changed markedly over the last century, and especially the last 50 years, as a consequence of increased diversions, construction of dams, weirs and levees and changes in operational procedures. A model developed by the Murray-Darling Basin Commission has been used to compare simulated natural (unregulated) flows at seven consecutive stages in the development of regulation. The model simulated flow from 1891 to 1986 by adjusting storage and diversions as appropriate, and thereby estimated the monthly flows that would have occurred had the given stage of development prevailed throughout the simulation period. The study results illustrate the cumulative effects of the dams on flow characteristics. Changes in the peak flow, low flow and flow-duration characteristics of the system increased both with distance downstream and the stage of development of regulation. Flow is often tightly coupled with other environmental characteristics such as temperature and oxygen, channel morphology and substrate particle size. Consequently, it is believed the cumulative alteration to the natural regime will have profound implications for communities of native plants and animals in both the riverine and the floodplain environment of the river.



Box 3.10: Example of cumulative affect on second order impacts (i.e. the geomorphology) of the Platte River, USA (after Hadley et al., 1987) Major changes have occurred in the hydrological regime and morphology of the channels of the Platte River, USA as consequence of impoundment and water resource development. Reservoir storage increased from zero to more than 8000 hm3 between 1885 and 1983 (Figure 3.10a). Trends in the hydrological regime of the river are indicated by increases in the magnitude of low flows in the flow duration curves and attenuation of annual mean flows and peak flows. These changes have affected the discharge of sediment in the stream channel network. New sediment transport regimes have resulted in sand bars that are not scoured or removed annually. When vegetation stabilises the sand bars, many become permanent islands. The cumulative effect of the water resource development on island and channel area are illustrated in Figure 3.10b. a) b)

0South Platte River

North Platte River

SEMINOE

PATHFINDER

MCCONAUGHY

GLENDO

1885 1895 1905 1915 1925 1935 1945 1955 1965 1975 1985

Cum

ilativ

e us

able

sto

rage

(cub

ic h

ecto

met

res)

8000

7000

6000

5000

4000

3000

2000

1000

9000 1000

900

800

700

600

500

400

300

200

100

0

Area

(squ

are

hect

omet

res) Island area

Channel area

1920 1930 1940 1950 1960 1970 1980 Figure 3.10: a) Cumulative useable storage in reservoirs in the Platte River basin and b) associated cumulative change in island and channel area



Box 3.11: Example of cumulative affect on third order impacts (i.e. zoobenthos) in the Gunnisson River, USA (after Hauer et al., 1989). Caddisflies often contribute significantly to zoobenthic biomass and energy processing in running water ecosystems. The Gunisson river system in the Rocky Mountains Colorado has been impounded by a series of high, deep release dams (Figure 3.11a). Distinct patterns in caddisfly distribution and abundance result from longitudinal changes in temperature, flow, substrata, trophic dynamics and other ecosystem attributes. Figure 3.11b illustrates the cumulative downstream effect of the dams on the total number of species and biomass. It has been demonstrated that the process of resetting the river continuum downstream of dams requires tens of kilometres. a) b)

3300

3100

2900

2700

2500

2300

2100

1900

1700

1500

13000 60 90 150 210 270 330120 180 240 30030

Taylor Park Reservoir

UnregulatedRegulatedReregulated

Kilometres from source

Elev

atio

n (m

etre

s ab

ove

sea

leve

l)

Blue Mesa ReservoirMorrow Point Reservoir

Crystal Reservoir

Blank Canyon

3700

3500

Num

ber o

f Spe

cies

Species per site

Biomass

DensityD

ry w

eigh

t (M

g m

)-2

Lava

e (m

)-2

20

15

10

5

020 80 140 200 260 320

4000

3000

2000

1000

0

4000

3000

2000

1000

0

20 80 140 200 260 320

20 80 140 200 260 320Kilometres from source

Figure 3.11: a) Longitudinal and altitudinal profile of the Gunnison River and b) changes in speciosity and biomass with distance downstream. 3.9 Estimating the Costs of the Impacts of Dams on Ecosystems This section provides a minimal treatment of this subject as most of this work is presented in WCD Thematic Review I.1 (Social Impacts of Large Dams) and WCD Thematic Review III.1 (Economics, Financial, and Distributional Analysis ). From an economic perspective, ecosystem changes due to dam development and operations may be considered as changes in the natural capital of a region. It is increasingly recognised that to achieve a sustainable development, measures of natural capital, and the income flows from it, must be assimilated into national and corporate accounting systems. This requires their identification and valuation usually in monetary terms. In the case of water resources, these may constitute part of what may be described as a nation�s �Critical Natural Capital� (Faucheux and O�Connor, 1998). Those responsible for water management need to be able to measure and value the ways that dams affect the environment. Clearly if the values of environmental impacts are incorporated into power/agricultural sector planning, they may well tip the balance and turn what appears to be an economically viable project into one that is likely to have a net economic cost. If the financial benefits of the dam are marginal, perhaps other development strategies would produce a better result. If, on the other hand, these environmental costs are not included at the project appraisal stage of dam planning,



the costs are hidden, but still need to be paid. The end result may have spatial and temporal impacts which create a situation of both intragenerational and intergenerational inequity. The economic impacts of these ecosystem changes tend to be greatest where they affect livelihood opportunities and industrial production. However in the case of dams the costs and benefits of resettlement of human populations, human and veterinary health, and the loss of cultural or natural capital all have to be addressed, and incorporated into the accounting process. In order for this to be done, these impacts must first be defined and measured, and a measure of value assigned to them. Where economic values can not be easily ascribed, multi-criteria decision-making is essential. Resettlement and health issues are addressed in detail in other WCD reviews (I.3) 3.9.1 Externalities and Livelihoods Many ecosystem functions are not properly accounted for in conventional market economics so the value of these functions and the cost of their loss, is external to the economic decision-making process. This externalisation of costs is a major factor driving the loss of natural ecosystems. Large-scale projects such as dams and canalisation of river-beds, have very high �hidden� environmental and consequent social costs that may only become visible after construction. These external costs are manifest in the loss of livelihood incurred by certain sectors of society rather than by those entities building and operating the dam. Examples of these losses include loss of access to resources from flooding for reservoir construction, degradation and loss of agricultural and grazing land on downstream floodplains, and loss of riverine and coastal fisheries dependent upon the river flood. The social consequences of these changes are dealt with in more detail in WCD Thematic Reviews I.1 Social Impacts of Large Dams Equity and Distributional Issues, I.2 Dams, Indigenous People and vulnerable ethnic minorities and I.3 Displacement, Resettlement, rehabilitation, reparation and development. Policy-makers need to identify the value of this loss of welfare, and implement financial and institutional mechanisms to assimilate these costs into the accounting structure faced by the agent responsible for bringing about that loss into the accounting process. This is particularly important when the distributional impacts of these external costs and benefits may be spatially and temporally inequitable. Compensation for such losses may also be included within the project as a redistribution mechanism. 3.9.2 Trade-offs between Economic and Ethical Considerations The vast majority of the world�s dams are owned by governments, central, regional or local. The value system that prevails within most governments or parastatal organisations is to increase the welfare of the constituency, which can either be a country, the population of a state, or that of a smaller region. An important premise is that growth in employment and material wealth is essential for the progress of humanity and the improvement of health and living conditions. Another value system maintains that nature should be disturbed as little as possible. People are part of nature, and nature has an intrinsic value that goes beyond the use to which it can be put by people. We must try to understand and respect natural cycles and live in harmony with them. Ecological systems are delicately balanced, and technological interventions can cause irreversible damage to the Earth�s life-sustaining processes. Clearly there is no consensus when it comes to codes of moral conduct. In our modern, cosmopolitan, multiethnic societies, we must learn to deal with this (IEA, 2000). However the United Nations has recognised the inter-dependence of the world's ecosystems, and noted that even purely local environmental crises are of common concern to all humankind. Environmental treaties repeatedly urge countries to take actions to protect environmental values within their borders because global values transcend those borders. Thus, ethical concerns run throughout the international instruments and pertain to projects that affect more than the immediately intended beneficiaries. Decisions must be proven to have taken these common concerns into account.



A set of ethical principles for decision-makers is presented in Box 3.12. Box 3.12: Ethical principles for decision-makers involved in water and energy planning (di Leva 1999) A. Decision-makers should acknowledge that their decisions affect a common concern of

humanity.

B. Decision-makers must respect all life forms.

C. Decision-makers must recognise that their decisions impact on future generations, and that those generations have inherent rights � the principle of �inter-generational equity�.

D. While environmental concerns must be respected and protected, decision-makers must also recognise the basic human right to development and to achieve a life free from poverty.

E. Decision-makers must also recognise that all people have a right to a livelihood and as far as possible to a stable, protective and unthreatening environment

F. Decision-makers must respect that actions taken within their jurisdiction should not lead to the harm of others outside their jurisdiction.

G. Decision-makers should ensure that in the event project development proceeds, that mechanisms are in place to enact the polluter pays principle, including the principle that those who cause damage are liable to ensure that the environment is restored to its former state.

H. Decision-makers should ensure that before decisions are taken, those who could be adversely affected have an opportunity to engage in prior informed consent and that decisions are enacted consistent with appropriate due process.

This short review of the economic and social implications of ecosystem impacts has highlighted the great value of natural ecosystems to society, yet underlined both the current challenges inherent in measuring this value accurately, and the need to ensure that these values are included in economic decision making. In the face of these constraints it is important that ecosystem values and the impacts of dams upon these be quantified wherever possible, and this should be pursued in parallel with ethical judgements as to the importance of the natural environment. Decisions need to be made not just through traditional cost-benefit analysis but also taking the environmental and social dimensions into account. This work should seek to ensure that the hidden costs of dams are made explicit and integrated into the decision-making process. Special importance needs to be given to the consequences of ecosystem impacts for those communities whose livelihoods are dependent upon healthy ecosystems. 3.10 Conclusions The processes that occur when a dam impacts an ecosystem are complex and, as a result, predicting in precise detail the nature and magnitude of any impact arising from the construction and operation of a dam or a series of dams is difficult in many situations. The detail of the impact of any single dam is unique and dependent not only on the dam structure and its operation but also upon local hydrology,



sediment supplies, geomorphic constraints, climate, and the key attributes of the local biota. Furthermore, fluvial processes will operate differentially even within an individual catchment. There is therefore no normative or standard approach to addressing ecosystem impacts and these have to be looked at on a case-by case-basis. The importance attached to some ecosystem changes will vary with the nature of human societies, cultures, and expectations, others are more straightforward to assess and quantify. Dams have a significant and measurable impact on man-made and natural ecosystems. These impacts are not documented for all 40 000 large dams in the world in all details . Despite the fact that each dam is unique, several generic environmental impacts are documented for a large number of cases. Negative impacts of dam construction and operations on ecosystems include in a majority of cases adjustments of river flow regime, negative effects on the quality of the water and reductions in sediment transport. The latter is leading in several deltas to coastal erosion of between 5 and 250 metres per year, requiring millions of dollars investment in coastal defence. Reduction in downstream annual flooding affects the natural productivity of floodplains and deltas. The impact of dams on freshwater species is significant. Global estimates of endangered freshwater fish reach 30% of the known species. Detailed studies in North America indicate that dam construction is one of the major causes for freshwater species extinction. Dramatic reductions in bird species are also known, especially in downstream floodplain and delta areas. Positive impacts on ecosystems and species also occur. The creation of reservoirs has in several cases lead to the development of productive and species-rich riparian ecosystems along reservoirs . In other cases, an increase in species numbers has been observed due to reservoir construction, especially in areas with low number of species per surface area as can be found at higher latitudes. In many cases the habitat gains in and around the created reservoir do not compensate nor outweigh the loss of habitats either upstream (i.e. inundated terrestrial ecosystem) or downstream. Multiple dams on a river significantly aggravate the impact on ecosystems. To date 61% of a large sample of world river basins have been found to be highly or moderately fragmented. In many cases, high percentages of sediment are entrapped if a cascade of dams is developed. Fish migration is affected even by a single dam, and multiple dams worsen this situation dramatically.



4. Responding to the Ecosystem Impacts of Dams 4.1 Introduction All dams, through disruption of physiochemical processes modify, in both obvious and subtle ways and across a wide range of spatio-temporal scales, the conditions to which ecosystems have adapted. Chapter 3 has shown that large dams can have major impacts on riverine ecosystems and processes and the changes in ecological processes can have profound social and economic repercussions for people whose livelihoods depend on the natural resources and ecosystem functions of river ecosystems. Consequently, the impacts have been at the forefront of the debate regarding the desirability of continued dam development (Chapter 5), and have led to a major investment by scientists and engineers to identify measures to ameliorate the most damaging impacts of dams. The analysis of potential ecosystem impacts of dams and the development of appropriate responses to these is one focus of the Environmental Impact Assessment (EIA) process. The strengths and weaknesses of EIA are dealt with in another WCD Thematic Review (V.2 Environmental and Social Assessment for large dams ) and will not be revisited in detail here. It is important to recognise however that while in principle an effective EIA allows the majority of potential effects of a proposed dam to be anticipated prior to construction, it may not be possible to determine the precise magnitude of impacts. Furthermore, because of the complex interaction of different effects, it will always be the case that some changes will not be anticipated and even when impacts are correctly predicted it is not technically possible to completely annul them all. It should be recognised that an EIA is the start of the amelioration process; implementation of measures to address negative impacts requires resources, capability and willingness from both the dam owner/builder/operator and the full range of political and stakeholder institutions. The present chapter reviews the methodologies that seek to respond to ecosystem impacts of dams. It presents information on their effectiveness, examines constraints, and discusses tools submitted to the Commission that can help improve the sustainability of water resource use and of large dams. 4.2 Types of Response When considering the construction of a new dam a prioritised, three-tier response to the environmental impacts identified through an EIA approach is widely advocated: 1) avoidance: measures that completely exclude as many negative impacts as possible; 2) mitigation: measures that attempt to minimise those negative impacts that cannot be

completely avoided, and 3) compensation: measures that attempt to enable reparation for those negative impacts that

cannot be either avoided or mitigated. For existing dams the same three types of response can be considered together with decommissioning and restoration of pre-dam ecosystems. Within this framework of avoidance, mitigation, compensation, and restoration, there are a wide range of specific measures that can be taken appropriate to the specific circumstances of each dam. The present section provides a summary of these. A more extensive listing of options that can be used to meet specific environmental impacts are dealt with in greater detail in Bizer (2000). Annex 5 presents a summary of World Bank experience and lists examples of mitigation measures.



4.2.1 Avoidance The best way to manage negative environmental impacts is to avoid them in the first place. The adoption of avoidance measures results in no change to existing functioning of a particular ecological area or resource. Normally, these measures are adopted to prevent loss of significant or sensitive ecological areas and consequently can only be incorporated during the conceptualisation or planning stages of the project (i.e. pre-feasibility, feasibility, or design). In order that avoidance be given due consideration, it is important that an adequate assessment of options be carried out at the conceptualisation stage of projects. The most obvious option for avoiding impacts to aquatic and terrestrial ecosystems is to either not construct a new project or construct the project at another location. Alternatives to constructing a dam should always be considered. For example, demand management, water recycling, rainwater harvesting (Box 4.1) as well as increasing efficiency of energy use or electricity production from alternative sources (e.g. solar, wind, thermal or nuclear) are all approaches that reduce the need for dam construction. However, all alternatives have environmental impacts that need to be weighed against those arising from dam construction. Such evaluation must consider the inevitable impacts both �upstream� and �downstream� of the site of immediate intervention (e.g. the impact of an oil based thermal power station must include the impacts associated with oil extraction, oil processing, transportation and storage as well as the impacts of the power station itself). Thus, when comparing environmental impacts the full �life-cycle� of alternatives must be considered. These options are further explored in Thematic Reviews IV.1 Planning Approaches; IV.2 Environmental and Social Impact Assessment for Large Dams; IV.3 River Basin Management: Its Role in Major Water Infrastructure Projects and IV.4 Regulation, Compliance and Implementation.



Box 4.1: Demand management, water recycling and rainwater harvesting: examples of methods that may avoid dam construction Demand Management � refers to actions for affecting the ways in which water is used and seeks to improve the efficiency of the direct use of water by people. It is increasingly coming on the agenda in regions where concerns have arisen over water sufficiency, allocation, high costs of expanding supplies and environmental preservation. Water supply and sanitation systems in many urban areas are characterised by high leakage and low efficiencies. A great deal of water is not returned directly (treated or untreated) to the source from which it was withdrawn. In some communities water losses can reach 40-60%. In contrast in modern cities with centralised water pipeline systems and relatively new sewage systems, losses do not usually exceed 5-10% (Shiklomanov, 1997). These figures demonstrate that actions to improve operational efficiency and reduce leaks could make a significant impact on water demand in urban areas. Water recycling � refers to the capture and reuse of water prior to ejecting wastewater to ground or surface water receptors. Hence, recycling reduces water withdrawal requirements for a given set of final uses. For example, end-use water requirements for an industrial process may be met totally by withdrawals in a once-through process, or by lower levels of withdrawals in combination with internal recycling which captures and reuses waste water. Similar considerations apply in the capture and reuse of water in municipal and irrigation systems. Water harvesting � is the direct capture of rainfall or collection of surface runoff to use as drinking water and to water crops and livestock. In some circumstances traditional methods of harvesting water may present opportunities for relieving water shortages in arid areas. Techniques include: • the capture of rainfall draining from the roofs of buildings; • tanks to capture surface runoff; • low earth embankments built across drainage channels to divert runoff onto fields, and • bunds constructed on fields to promote the infiltration of surface runoff. One approach that is advocated to avoid widespread adverse effects to environmental resources at the national level is to initiate a development strategy that commits a single river basin to development while limiting development in other river basins. This has most potential in countries where there are several undeveloped river basins, some of which have more environmental significance than others. Currently, the national hydropower strategies of Switzerland (Truffer, 1999) Sweden (Lövgren 2000) and Norway (Larsen, 1999; Flatby and Konow, 1999) involve the �set aside� of particular river basins for the purpose of environmental protection. However, the addition of new installations on regulated rivers can lead to cumulative impacts (section 3.8) and harm remaining habitats. Consequently, this policy should not allow a �free for all� in those basins designated for development; amelioration of negative environmental impacts should still be viewed as a key aspect of all projects. Beyond the actual selection of a site, adjustments to the alignment of the dam or the configuration of the dam and ancillary facilities may be made on the basis of environmental criteria. Factors that may be considered in developing the configuration of the project include dam height (maximum elevation of the water surface of the impoundment) and dam alignment. In many cases, constructing the dam to a somewhat lower elevation may avoid inundation of areas that are relatively densely populated, provide habitat for unique ecological or geological areas, or include sensitive or critical habitats for rare, endangered or endemic species. For example, Hydro Quebec modified the height of the St Marguerite-3 dam to reduce the flooded area by 20% (315 to 253 km2) while only suffering a 4% reduction in power generation (2.9 TWh to 2.8 TWh) (IEA 2000). Such adjustments must be based on



an adequate understanding of the environmental conditions within the river basin as well as outside the project area. Avoiding the potential for adverse effects associated with construction of the project and ancillary facilities is somewhat more flexible. Re-alignment of access roads or transmission lines to prevent fragmenting associations of native vegetation and protection of wildlife is normally possible. In the Upper Salmon Hydroelectric project, for example, Newfoundland and Labrador Hydro avoided disturbance to calving caribou by curtailing construction activities when sufficient numbers of caribou were in close proximity (Kiell 2000/ENV202). Box 4.2: Avoidance of impacts on sensitive species during blasting During rehabilitation of the spillway of the Shongweni dam (S Africa), the EIA identified 5 000-10 000 bats roosting in the Mlazi by-wash tunnel. A significant population of Temmincks hairy Bat, Myotis tricolor, use the tunnel as a maternity site in January, which was the originally-intended construction period. Blasting was therefore delayed to avoid the risk of activity-related starvation during the breeding season. After project completion, monitoring showed no adverse effects on the population. (English 1999/ENV205) Selection of locations for acquisition of aggregate or fill for construction of the dam or for disposal of excavated material may also be based on ecological considerations. Construction materials may be obtained from, or spoil materials disposed of, within the impoundment zone, thereby avoiding changes to the topography in the area around the reservoir and reducing the potential for erosion from the disturbed sites, and reducing the aesthetic impacts after construction is complete (EHDC, 1994). Further, consideration of placing construction staging areas and other necessary areas within the impoundment zone may be a means to avoid disturbance to lands that otherwise would be affected in addition to the areas that are inundated. However careful assessment is required to ensure that such measures do not increase water pollution or have other negative impacts 4.2.2 Mitigation The second category concerns mitigation measures that are incorporated into a new or existing dam design or operating regime in order to reduce negative ecosystem impacts. For new dams, mitigation measures attempt to reduce the occurrence of anticipated adverse effects while at existing dams, mitigation measures seek to rectify as far as possible adverse effects through modification of structural or operational components of the development. Ideally, mitigation measures are identified through the EIA process so that adverse effects are minimised from the outset of a project. Mitigating construction impacts Disturbance to soils at the construction site is one of the first and more easily-handled aspects of dam construction simply through the effective use of berms that direct runoff to settling basins. Areas adjacent to the primary areas needed to construct the project and which are disturbed as a consequence of site preparation may be revegetated with grasses or other rapidly-growing plants to prevent erosion from the sites. Other types of containment facilities (primarily berms) are frequently used to detain runoff from batch plants, equipment maintenance areas and equipment storage facilities (Bizer 2000). Most of these types of protection measures are described in numerous compilations of �Best Management Practices� and generally do not add significantly to the cost of the site preparation (e.g. Hydro Quebec 1991). The behaviour of the labour force is a more difficult aspect of managing the potential adverse effects of a construction site, especially in remote, unpopulated or sensitive areas. This is particularly true where workers are prone to hunt either to supplement their food or to supplement their income through sale of plants and animals from the area immediately surrounding the labour camp. The



impacts of non-labour force migrants that gather at construction sites are even more difficult to manage. While the labour force itself may be provided with sufficient food, water and alternative activities, the migrant, non-labor force population will likely forage in the adjacent areas for food and fuel and may create significant problems with sanitary and solid wastes. Few if any construction projects are equipped to deal with the potential environmental damage caused by non-workforce migrants. In most cases, the selection of routes for access roads and transmission lines can avoid passing through ecologically sensitive areas. Where it is not possible to avoid such areas, a route should be selected to avoid disturbance and fragmentation of the biologically significant areas to the extent possible. For transmission lines in particular, it may be possible to construct the transmission towers and string the lines without completely clearing the right-of-way between the towers. In this way the fragmentation of the area may be minimised. As soon as construction of the access road or transmission towers is completed, the disturbed areas should immediately be re-vegetated with local vegetation. Access roads may also be closed to the public to reduce frequentation of sensitive areas (Kiell 2000/ENV202). A mitigation measure sometimes implemented is the rescue of terrestrial animals from the area to be inundated. For example, 10 000 animals were rescued from drowning prior to the filling of the Afokaba reservoir on the Surinam River in South America (Nilsson and Dynesius, 1994). However, while the approach often generates wide publicity, it is rarely a success. Often the rescued animals are relocated to areas where there is insufficient carrying capacity to support the influx of new animals. Mitigating Operational impacts Scientists and engineers have developed numerous methods for mitigating many of the environmental impacts that occur both upstream and downstream of dams after construction. One measure that is commonly adopted is the release of �compensation� flows for the maintenance of in-channel ecosystems downstream of a dam. In the past these were often simply a continuous minimum flow release that had to be maintained throughout the year. In recent years it has been more common for seasonally varying flows regimes to be designed to meet �instream� or �environmental flow� requirements. Thus release regimes are planned to mimic the natural seasonal variation in flow. In some situations (e.g. in parts of Australia), where rivers would naturally dry up at certain times of year, this may mean very small or even zero release. Even more recently consideration has been given to the release of high flows to maintain floodplain (out-of-bank) and deltaic ecosystems (Acreman et al, 2000). The design of Environmental Flow Requirements is dealt with in further detail in Section 4.4.3. Measures to minimise erosion in the draw-down zone of the dam impoundment include minimisation of the magnitude of the draw-down, and/or planting of annual vegetation along the margins of the impoundment as the reservoir is drawn down. When operation of the project is such that water level fluctuation in the reservoir is unavoidable, measures that have been implemented have included construction of submersible barrages at the mouths of inlets along the margins of the reservoir (Casinader, 1999; Seattle City Light, 1985). These maintain a water level within the inlet at some level greater than the maximum drawdown of the impoundment. When the impoundment is at maximum operating level, the water in the inlets is contiguous with the impoundment allowing movement of organisms between the main body of the reservoir and the inlet area. The changes in physical and chemical properties of water associated with dam construction, both in the reservoir and downstream from the reservoir, may be minimised by integrating measures to mitigate for changes in temperature and dissolved gas concentrations. Options include the installation of variable level off takes or submersed surface impellers and limiting plunge port depths. The



retrofitting of variable level off takes is very expensive but has been attempted in various reservoirs, including in Japan (Miyanaga et al., 1994). The design technology is now well supported by field experience. Other major considerations relative to water quality include effects to nutrient loading and salinization associated with return flows from irrigated agricultural areas and use of the river for domestic waste disposal upstream. In all cases an integrated upstream catchment management strategy is required to minimise anthropogenic inputs of pollutants into the reservoir. Measures to mitigate for the potential effects of nutrient accumulation in an impoundment have focused on reducing the amount of inflow of nutrients to the reservoir, and increasing the removal of nutrients from the water. In some countries (e.g. Switzerland) waste water treatment is specifically designed to try and limit nutrient influx into lakes and reservoirs. There are various interventions that can reduce the nutrient load of streams that flow into reservoirs (Bernhardt, 1994). However, in many countries eutrophication of reservoirs remains a major problem. In many cases, there is much scope for improving up-stream basin wide management to reduce nutrient input into the reservoir. Reduction of the inflow of nutrients has been accomplished through the construction of wastewater treatment facilities at communities along the margins of the impoundments as well as in the watershed upstream and the promotion of the use of biogas units in more rural areas of the watershed (EHDC, 1998). Other methods of reducing continuing inflow of nutrients to an impoundment include training of local farmers in the use of fertilisers, or seasonal flushing of the reservoir (i.e. drawdown of the impoundment and refilling to dilute the concentration of nutrients). The effectiveness of this process however is dependent upon the volume of the reservoir relative to inflow (Jobin, 1999). Mitigation for the accumulation of sediments in the impoundment has been achieved in a small number of dams in several ways. A direct approach to reducing the accumulation is to mechanically remove the sediments by periodic dredging. In other cases, the sediments have been removed through periodic flushing of the reservoir by releasing large volumes of water through low-level outlets in the dam. However, this approach is limited to removal of sediments that have accumulated close to the dam, in many cases it is impossible to flush sediments deposited further upstream in the reservoir. Furthermore, although the flushing of sediment may have some geomorphological benefits, the extreme and unnaturally high concentrations of sediment caused by flushing may produce a major stress on downstream aquatic ecosystems and can be disastrous for some biota (section 3.7). Furthermore, sediments flushed from the hypolimnion of stratified reservoirs may be contaminated (e.g. with mercury and other trace metals). Partial removal of accumulated sediments at the mouths of tributaries is another method often used. It maintains access to the tributaries for movement of fish from the reservoir into the tributaries for spawning and rearing of juvenile fish. The accumulation of sediments in the river channel downstream from a dam due to the altered hydrologic regime may be mitigated through periodic flushing of the river channel with artificial, high flow events, if the sluice gate design and water levels allow. For many dams, sediment accumulation remains a major concern and one that is not often financially viable to mitigate by dredging. The configuration and bathymetry of most reservoirs means that sediment frequently accumulates at the head of the reservoir, a long way from the dam wall and the bottom outlet. Many feasibility studies acknowledge the concept of a �life span� for a dam based on inevitable sediment accumulation. Sediment arrangement and safety issues currently head industry concerns about dam management. Most measures to mitigate for the altered hydrologic regime have been designed primarily to prevent impacts to the biological components of downstream aquatic communities dependent on the natural hydrologic, physical and chemical regimes. This is addressed further in section 4.4.3.



Effective measures for mitigating the blockage to migration of fish include the installation of fish passage facilities to facilitate movement of fish from below the dam to the reservoir and further upstream. Designs of fish passage facilities have evolved through the years as scientists and engineers learn more of the requirements for encouraging fish to use the passage facilities, the specific hydraulic conditions that various fish species use to orient their migration, and the climbing capabilities of the target species or groups of species (Bizer, 2000). However, much of this research has been conducted in Europe and North America and data is lacking on appropriate designs for tropical and subtropical species. Most designs for fish passage facilities are passive: it is up to the fish to move themselves into and through the fishway. Other fish passage designs have included fish elevators that collect the fish in boxes and then lift the fish to the level of the impoundment. In other instances, trap-and-haul techniques have been used to move fish from below the dams to the impoundment. The effectiveness of fish ladders is discussed further in Section 4.3.1. 4.2.3 Compensation The third category concerns compensation measures. These consist primarily of some form of �repayment� for anticipated (or realised) adverse effects that cannot be either avoided or mitigated to some �acceptable� level . Such measures include direct monetary payment to a governmental or non-governmental organisation for the adoption of resource management programmes that provide for protection, replacement, rehabilitation, or restoration of natural resources in repayment for the loss of natural resources due to the development. Compensation for lost resources may take the form of: i) guaranteed preservation or restoration of existing ecologically important areas to �replace�

those lost as a consequence of dam construction (e.g. establishment of nature reserves), and ii) alternative approaches to ensure the continuance of some ecosystem services (e.g.

construction of a fish hatchery to replace lost fish spawning areas) Some authorities will only accept the former type of compensation. Compensation may also be paid �in-basin� (e.g. restoration of forest area within the river basin for forest lost to inundation) or �out-of-basin� (e.g. assistance in expanding management capability at similar locations in another river basin). Compensation measures are normally adopted during the planning process. However, compensation programmes may also be implemented after construction of the project. They may be included as part of the capital costs of the scheme and/or funded from revenue in the case of hydro-power projects (Bizer 2000). Generally, the entire loss of riverine and terrestrial areas to inundation is neither avoidable nor directly mitigable. Consequently, the only options readily available are compensation, either through increased management of other existing areas in the basin, restoration and rehabilitation of other lands, or through direct monetary compensation for improved management by national resource agencies in other parts of the country. For example, Zimbabwe created a National Park along the shores of Kariba reservoir, and the Nam Theun II project proposes support for a 3 710 km2 forest National Park in the catchment. Many ecologically important terrestrial areas are in danger of continued exploitation regardless of whether or not a dam is constructed. Compensation therefore should be directed primarily at the restoration or rehabilitation of areas that have been damaged by other human activities (e.g. timber harvest or agricultural practices). Restoration, rehabilitation and conservation efforts have been successfully applied to a number of projects (e.g. Braund, 2000; Hamerlynck et. al., 1999, Rivero, 1999). However, it is of course impossible to fully compensate for loss of unique environments (e.g. virgin rainforest) and/or species.



In situations where installation of fish passage facilities is not feasible, other measures that may be used to compensate for the loss of spawning areas include artificial enhancement of breeding populations in hatcheries designed specifically for native riverine fish and artificial breeding and stocking of appropriate fish species for introduction into the reservoir (e.g. Idaho Power, 1999; Chelan County PUD, Committee on Dams and Environment; 1999). Critics of hatcheries point however to the need to maintain the genetic diversity and fitness of natural stocks if compensation is to be truly effective. While they do not address issues of biodiversity loss, the introduction of native species from other adjacent areas or the introduction of non-native species to a reservoir can bring considerable economic and social benefits through the development of subsistence and commercial fisheries (e.g. Indonesia, Laos, China) and the evolution of sport fishing industries as in the US (Jackson, 1999), the United Kingdom (Binnie, 1999), and France (Masson, 1999). The decision to compensate for loss of fish species in an inundated reach of river through introduction of non-native commercial species is generally made in situations where principle concerns exist for providing a source of protein for local residents or where pressure for recreational opportunities is present (Jackson, 1999). Such introductions, however, bring with them the risk that the new species will predate or invade the ecological niches of native species, so reducing biological diversity further. For example, the introduction of brown trout to Lake Pedder Dam in Tasmania is believed to have contributed to the decline of two endemic Galaxia species, one of which is now endangered (WWF Australia 2000/ENV220). It must be recognised that no scheme will ever fully compensate for the natural resources and functions that are lost as a consequence of dam construction. Consequently, compensation should be viewed as the least acceptable of the amelioration approaches and one that is used solely to diminish the negative consequences of impacts which can neither be avoided nor mitigated to an acceptable level. 4.2.4 Dam Decommissioning and River Restoration Increasingly rivers once developed and altered for human benefits are being restored, in a variety of ways, to an approximation of pre-disturbance conditions. Within this context dam removal is increasingly being considered as a realistic option for the potential restoration of river ecosystems. In some cases people have modified their livelihood strategies or the river ecosystem has been developed to such an extent that decommissioning is at present not a viable option. However, in many cases decommissioning is possible and should be considered. In the USA 465 dams have been removed in recent years. The majority are small (i.e. < 5 m) and medium (i.e. 5 to 15 m) sized dams. To date only a handful of dams greater than 20 m have been decommissioned (Figure 4.1). The removal of dams in part indicates changing societal standards and in part reflects simple economic considerations. Increasingly society is placing greater value on �natural� systems and in many cases environmental enhancement is the primary consideration in dam removal (Box 4.3). However, it also often the case that when dams reach the end of their design life and/or represent a safety hazard, decommissioning is often less expensive than alternative options. The cost of repairing a dam is often significantly greater, than the cost of removal, particularly when fish passage must be provided. In the USA it has been computed that on average the cost of repair is 3.4 times greater than the cost of removal (Born et al. 1998).



Figure 4.1: Number of dams removed in the USA, as a function of a) dam height and b) year of removal (after, Doyle et al., 2000) To date there has been very little research into the environmental impacts associated with dam removal. Nearly all that has been conducted has been done in the USA. The primary concern in most, if not all, dam decommissioning cases is the fate of sediment stored in the reservoir and the subsequent physical changes in the river channel that occur following removal. Although there has been relatively little scientific study, the most consistent observation is that drastic geomorphological changes have occurred immediately following dam removal. At the 1969 removal of the Newaygo Dam on the Muskegon River, Michigan, approximately 40% of the stored sediment moved downstream immediately in the form of an elongating sediment wave, travelling at a rate of approximately 1.6 km per year (Simons and Simons, 1991). At the Fort Edward dam on the Hudson River (removed in 1973) approximately 33% of the stored sediment moved downstream within a year of removal and was highly publicised due to the presence of PCBs in the transported sediment (Shuman, 1995). Box 4.3: Decommissioning of the Edwards Dam, USA.

When in 1997 the Federal Energy Regulatory Commission (FERC) in the USA ordered the removal of the Edwards dam (built in 1870) it cited �compelling environmental� considerations. The Commission said its actions were based on the following key considerations: • power produced at the dam can easily be replaced by existing resources in the region; • removal will provide 9 species of fish with continuous access to 15 miles of spawning habitat; • removal will provide 4 species of fish that do not use fishways with access to their entire historic

range within the Kennebec river; • wetland habitats, recreational boating and fishing will benefit, and • there will be no major environmental or social drawbacks.



Although it is often widely assumed that changes are substantial, relatively rapid and positive, there is little solid evidence regarding rates of biotic recovery following dam removal. Some preliminary results indicate that changes can be extremely rapid. For example, striped bass returned to inaccessible sections of the Kennebeck River within 3 months of breaching of the Edwards dam in Maine (American Rivers et al., 1999). Similarly, there was increased diversity of invertebrate communities in the Baraboo River, Winconsin within one year of dam removal (Doyle et al., 2000). However, in contrast, modelling results from studies of dam removals suggest a period of decades to centuries will be required for complete recovery of fish and riparian plant populations (National Park Services, 1996). It has been suggested that massive changes in the physical structure of the river (e.g. caused by sediment redistribution) could result in dramatic ecological changes with long-term consequences (Doyle et al., 2000) (Box 4.4). Box 4.4: Possible consequences for salmon of removal of the Elwha Dam, Washington (after Doyle et al., 2000) An evaluation of the potential impacts of removal of the Elwha Dam, Washington, concluded that dam removal would have �major adverse short-term impacts on salmon attempting to return or spawn in the river�. Suspended sediment loads associated with breaching are expected to reach lethal levels during some phases of the removal, consequently the removal schedule will be designed such that sediment releases occur when the salmon are not in the river. If scheduling is successful it is reasonable to expect populations to recover to pre-impoundment densities. However, mismatches between expected time of fish runs and the removal schedule, or an unanticipated flood that mobilises large amounts of sediment could potentially devastate remaining populations. In addition the ecosystems upstream of the dam may take many years to recover from inundation. Predicting or controlling the alignment and location of the upstream channel in the reservoir sediment after dam removal is difficult. Surveys and soil coring data indicate that in four out of five cases, post-removal channels do not follow the pre-dam channel alignment, but rather develop a new course (Lenhart, 2000). Furthermore, plant communities in former impoundment sites may not resemble naturally occurring plant assemblages. In Winconsin, floral communities are dominated by monocultures of weedy pioneer species five to six years after dam removal. The extreme habitat alteration (i.e. the sudden availability of extensive amounts of nutrient-rich sediments) means that plants that initially colonise the exposed sediments are able to persist for several years and prevent other species from becoming established (Lenhart, 2000). There are however methods of ameliorating the negative environmental impacts of dam decommissioning that will improve as experience grows. For example, reservoir sediment can be removed (e.g. by dredging) prior to decommissioning. Alternatively the dam removal can be staged so that there is a more gradual influx of sediment into the downstream system. Upstream of the former impoundment active management to both stabilise sediments and encourage the return of naturally occurring plant species is possible. Given that all dams have a finite life, the issue of decommissioning is central to dam construction. There is a requirement for much more research to improve methods of decommissioning and post-dam restoration of ecosystems. Furthermore the costs of decommissioning must be considered within the life-cycle analysis of all dam projects. 4.3 How Effective is Mitigation? Dams are the source of significant and unavoidable environmental impacts. As section 4.2.2 has shown, the severity of many, but not all, impacts can be reduced through implementation of a variety of mitigation measures. The critical question that emerges from this review and the many submissions (58) on mitigation received by the Commission is � how effective is mitigation? To attempt to answer this question we have drawn upon all reviews available to us. There is however



only limited published information in the literature on how effective mitigation plans have been in meeting their objectives in developing and developed countries. While it is obvious that mitigation measures can work in at least some specific cases it is certainly not the case that mitigation measures are designed, implemented, or are effective in all cases (see example in Annex 7). To a large extent the effectiveness of mitigation measures depends on the abilities both of those people who determine which impacts need to be mitigated and those who design and implement a particular mitigation strategy. The effectiveness of mitigation is, like the impacts themselves, very site specific. Reviews of environmental impact assessment, mitigation and monitoring by the World Bank, the Asian Development Bank, African Development Bank and Interamerican Development Bank highlight the gap that exists between the potential to identify and mitigate against environmental impacts and current practice, one that has also been documented (Bizer 2000). For example. the Asian Development Bank report �Special valuation study on the social and environmental impacts of (four) selected hydropower projects� (ADB, 1999) highlights the importance of distinguishing between both adequate formulation of environmental clauses and compliance with these clauses. The ADB report notes that �compliance with environmental clauses in construction contracts has not been satisfactory, because many have very modest clauses�, and highlights the limited implementation of measures, even where they are well designed. The report then goes on to stress the important role played by regular monitoring missions and recommends strengthening ADB�s practice in this sector as weak oversight can lead to unsatisfactory outcomes. These issues also emerge in the IDB assessment of dam projects (Dam projects 1960-1999, IDB 1999) which strives to distinguish whether proposed environmental mitigation plans were implemented as planned. It also determines whether additional actions were needed to correct what were seen as either badly-formulated Environmental Mitigation Programmes, or as corrective actions for unforeseen impacts. In fact most of the IDB examples do not have any data on environmental impacts as these projects were constructed prior to EIA regulations (51 out of 88). Figure 4.2 shows the distribution of different categories of environmental problem evaluations that indicates that only 4 out of the 16 more recent EMPs (Environmental Management Programmes) have effectively addressed the environmental issues; a further two require additional inputs and these are underway; the remainder have residual impacts that are not being appropriately addressed.

3

2

5

1

0

3

0

2

0

1

2

3

4

5

6

7

P 1 P 2 P 3 N 1 N 2 N 3 P x N x

Num

ber o

f Cas

es

P 1 : a n E M P w as p re p a re d . C u rren tly th e re a re n o s ig n ific a n t e n v iron m e n ta l p ro b lem s in th e p ro je c t a re a o f in f lue n c e . S o m e m ino r e n v iro n m e n ta l a lte ra tio n s h av e b e e n a pp ro p r ia te ly a d d re s s e d a nd a p p ro p r ia te m e a s u res to p ro te c t the e n v iron m e n t a re b e in g im p le m en te d in a tim e ly a n d s a tis fa c to ry m a tte r.P 2 : a n E M P w as p re p a re d . T he p ro jec t p ro v o k e d o r is th re a te n in g to p rov ok e a s ig n if ic an t d e te r io ra tio n in the e n v iron m e n t, a n d /o r s o m e u n fo rs e e n c u m u la tiv e n eg a tiv e e nv iro n m en ta l im p a c ts h av e a r is e n . H ow e v e r, a p p ro p r ia te m e a s u res a re b e in g ta k e n b y th e b o rro w e r/im p le m e n t in g a g en c y to a d d res s th e p ro b lem s .P 3 : a s in P 2 , b u t p ro b le m s a re n o t be in g ad d re s s ed a p p rop r ia te ly .N 1 : n o s p ec ific E M P w a s p rep a red o r it is u n k n o w n if an E M P w as p re p a re d . C u rren tly the re a re n o s ig n if ic an t e n v iron m e n ta l p ro b lem s in the p ro je c t a re a o f in f lu e n c e . S o m e m in o r e n v iro n m e n ta l a lte ra t ion s h a v e be e n a pp ro p ria te ly a d d re s s e d a nd a p p ro p r ia te m e a s u res to p ro te c t th e e n v iro n m e n t a re b e ing im p lem e n te d in a t im e ly a n d s a t is fac to ry m a tte r .N 2 : n o s p ec ific E M P w a s p rep a red o r it is u n k n o w n if an E M P w as p re p a re d . T h e p ro jec t p ro v o k e d o r is th re a te n in g to p ro v ok e a s ig n ific an t d e te r io ra tion in th e e nv iro n m en t. H ow e v e r, a p p ro p r ia te m e a s u re s a re b e in g ta k e n by th e lo c a l a u tho r it ie s to a d d re s s th e p rob le m s .N 3 : a s in N 2 , b u t p ro b le m s a re n o t b e in g a d d re s s e d a pp ro p r ia te ly .P x : an E M P w a s p re p a red , b u t th e re is ins u ff ic ie n t a n d /o r n o n -c le a r b a s is to e v a lu a te th e m ag n itu de o f p ro b le m s th a t m a y a ffe c t th e p ro je c t a re a o f in flu en c e .N x : n o s p e c ific E M P w a s p re pa re d o r it is un k n o w n if a n E M P w a s p rep a red . T h e re is in s u ffic ie n t a nd /o r n o n -c le a r ba s is to e v a lu a te the m a gn itu d e o f p ro b le m s th a t m ay a ffec t th e p ro je c t a rea o f in flu e nc e .

Figure 4.2: Figure 4.1: Distribution of EMP preparation & environmental problem evaluations for dam projects in Latin America co-financed by the IDB from 1960-1999(IDB 1999).



These bank reviews give insights on the effectiveness of mitigation in developing countries and the difficulties experienced in ensuring that clients respect operational guidelines on environmental issues. It is significant that these problems arise despite the internal scrutiny of the banks, their regular supervisory missions, independent panels, and their commitment to ensuring good environmental outcomes according to their own internal policies. This raises the concern that dam projects without donor input that are less well documented and monitored may have even lower performance ratings than those described above. In industrialised countries where the information base and the scientific and financial capacity to mitigate is generally greater, increased investments are now being made in post construction mitigation of impacts, frequently in response to strong public demand. Recent crises such as the salinity crisis and algal blooms in the Murray Darling basin (A) in the early 1990s, or the declaration of once common migratory fish stocks as �endangered� in the US, have focused minds and allowed mitigation measures to be improved. Salmon and trout issues dominate mitigation measures in the US where several tools exist for retrospective fitting to reduce impacts including technical measures for the reduction of mortality due to Total Dissolved Gases, turbine design improvements to make them fish friendly, fish screening and guidance technology to guide them over the dam, and improved fish ladders and barging schemes. These measures do not prevent all negative dam impacts on fish. For those people campaigning for the removal of dams, technological �fixes� will never be the solution. The real issue for mitigation of dam impacts is what residual effects are acceptable. This is a question of societal preference, the answer to which has and will continue to change over time. 4.3.1 The Example of Fish Ladders Fish ladders are perhaps the most obvious form of mitigation on rivers where the dam creates an insurmountable obstacle to migratory fish, and serve as an interesting example where mitigation practice from different countries serves to inform the discussion on how effective such measures are. Upstream passage for anadromous species is provided for through several types of fishway: pool-type fish passes, Denil fish passes, nature-like bypass channels, fish lifts or locks and collection and transportation facilities. Only a few special designs have been developed in Europe, Japan, New Zealand and Australia for catadromous species, namely for eels. Most of the effective fish passage facilities have been designed and installed at existing temperate and north-temperate-zone projects. Installation of fish passage facilities in tropical or subtropical zones has been less successful. A major cause for ineffective installations at dams in the tropical and subtropical areas is that the designs are typically adaptations of facilities designed for temperate species. As was discovered in North America, the designs are generally species or family specific in that designs such as pool and weir ladders that work for salmonids do not necessarily work for other migratory species such as the alosids, cyprinids or ictalurids (Committee on Dams and Environment, 1999). In Thailand 50 species have been recorded to use the Pak Mun fishpass out of 265 present in the Mun river. This may be because of non-optimal design due to a steep gradient resulting in high water velocities. It is also important to note that for only two species do a high proportion of the migrating fish manage to swim through the pass (Ek, 2000 pers comm., Bizer 2000, WCD Case Studies � The Pak Mun Dam & Mekong river basin) Between 1950 and 1994, only 16 known fishways (mostly of the fish ladder type) were constructed in South Africa. Fifteen of these were formally planned, with only one properly modelled and model tested, before construction (B Bernade pers com). Fish passages have been considered singularly ineffective by some experts in Brazil, which has led to widespread mitigation of hydroelectric effects with fish hatcheries (Carolsfeld 2000, pers comm). The first ladder was constructed on a small dam in Pirassununga, SP, over 60 years ago, along the



lines of salmonid ladders, followed by another 30 in this state. Godoy (1945) estimated 10% passage of up to six species for the first of these ladders, and they appear to have assisted significantly in maintaining the populations of some species that jump well (Prochilodus spp. in particular). However, a more recent evaluation of one of these ladders by Godinho et al. (1991) suggested a much lower efficiency (2% reached the top for only a single species out of 34). Few of the 430 dams over 15 metres in Australia have fishways of any description as there were no technical solutions at the time, although with current technology fishways / fishlifts could be retro-fitted on these structures (Flanders 1999, pers comm.). Recent research has shown that design improvements can increase effectiveness of mitigation (see Box 4.5). Only 9.5 % of the 1 825 hydropower dams in the US have an upstream fish pass facility. A DOE report from 1994 looked in detail at 16 case studies of dams with fish passes and concluded that half of those with upstream facilities reached their stated mitigation goal. The others either do not monitor or were affected by factors such as low stream flows that impaired success. The report concludes that monitoring should be a requirement in order to improve assessment of mitigation investments that may cost hundreds of thousands of dollars per year (DOE 1994). Box 4.5: Improving fish passage design to make them work better In 1976 a pool-and-weir type fishway was incorporated into a tidal barrage on the Burnett River in SE Queensland, Australia. Assessment of the fishway in 1984 and 1994 showed it to be ineffective with only 2 000 fish of 18 species ascending over a 32-month period. The fishway was modified to a vertical-slot design with low water velocity and turbulence. The new design saw 52 000 fish of 34 species using the fishway over 17 months. The modified design provided access for non-leaping fish precluded from the pool-and-weir design by high water velocities and high head loss between pools. The local fish species were clearly less adept at leaping than the northern salmonid species for which the pass was originally designed (Flanders 1999/ENV219). Box 4.6: Why fish passes may fail When the causes of poor performance (in terms of effectiveness and/or efficiency) of fish facilities are analysed, certain factors are frequently revealed (Larinier, 1998 ; Nakamura, 1993 ; OTA, 1995) : • Lack of attraction of the facility, resulting from a poor position of the fish pass or insufficient

flow at the entrance of the facility in relation to the flow discharge into the river; • Poor design of the facility with regard to the variations in water levels upstream and downstream

during the migration period, resulting in under or oversupply of flow to the fish pass, or excessive drop at the entrance. This may be due to poor appreciation of the range of the upstream and/or downstream water levels during the project planning phase, or a subsequent change in these levels;

• Poor dimensions: pools with insufficient volume causing excessive turbulence and aeration, excessive drop between pools, insufficient depth for the fish, or the flow pattern in the pools not suitable for the target species;

• Frequent clogging up or obstruction of the fish passage facility, resulting from inadequate protection against debris, or too exposed a position, or quite simply inadequate maintenance on the part of the operator;

• Malfunctioning of parts which regulate the flow discharge and the drops between pools (automatic sluice gates, etc.), or which ensure the functioning of the facility in the case of fish lifts and fish locks (automatic sluice gates, hoist for the tank, moving screens, etc.).



4.3.2 Why Avoidance, Mitigation, and Compensation are Difficult 4.3.2.1 Dealing with scientific uncertainty The constraints encountered in ameliorating the ecosystem impacts of dams are hardly surprising. In order to design fully-effective amelioration systems a good information base and a strong predictive capacity are essential. These are simply not yet available for most riverine ecosystems. For example, while it has rarely proved practical to survey all biota at potential dam sites, this is clearly essential if major impacts on biodiversity are to be avoided. Not only do we need to know what species are present, but also whether they are unique compared to neighbouring areas and other riverine ecosystems in the region. The capacity to rank the importance of species richness for alternative dam sites is a key contribution to the decision-making process and is essential if the policy of �avoidance� or �compensation of damage� at one site through creation of protected areas in similar sites is to be effective. Future restoration projects will also be more successful if the state of the original ecosystem is known, and can provide a benchmark for fixing future restoration management objectives. In contrast to the engineering profession, the details of ecosystem solutions tend not to be generic and transferable. Fish ladders designed for salmon migration in the Pacific NW of the USA will not be appropriate for Mekong Catfish in Thailand, and flow releases to maintain threatened Lesotho Minnow would be inappropriate for the Indus Dolphin. This requires a case-by-case response that can be inspired by what works elsewhere but can never simply be copied. This faulty assumption has been the basis of many ineffective and expensive fish ladders (Section 4.3.1). This uncertainty has a knock-on effect on the effectiveness of mitigation measures such as Environmental Flow Releases (EFRs). Their success depends, in large part, on predicting and modelling the theoretical ecosystem response prior to construction. However the real experiment will only begin when the dam is actually closed and data can be gathered on the observed responses, some of which may only stabilise over many years. It is therefore essential to include a �monitoring and feedback� element in the EFR measures that provides for constant assessment of ecosystem response and feedback to modify dam operation rules as appropriate. There is a need for fundamental research linking abiotic processes to changes in ecology, particularly in tropical climates. In the absence of much basic understanding, a cautious approach should be adopted in decision-making. 4.3.2.2 Capacity constraints Lack of scientific understanding is at present the primary constraint on amelioration in both the developing and the developed world. In part this has often been due to a lack of commitment from planners and donors to funding the necessary research. However, in many countries effective conservation, mitigation and compensation measures face a combination of additional constraints relating to human, financial, and institutional capacity. While in general these are often exacerbated in developing countries, they are far from uncommon in the developed world. Primary issues include: Funding • Experience of successful mitigation in industrialised countries indicates that developing the

scientific knowledge needed to mitigate and manage impacts involves significant financial resources. In many projects funds to conduct the necessary research are very limited.

• Funds for conducting environmental impact assessments and for post-project monitoring are often only a very small fraction of the total project costs and in many cases are insufficient.



• Environmental impact assessments are very often viewed simply as an �add-on extra�, a hurdle that has reluctantly to be overcome to enable the project to go ahead.

Institutional capacity • Often the responsibilities for planning, monitoring and regulating dams are spread across a large

number of institutions. Disparate organisation can result in problems relating to management co-ordination and the identification of responsibility.

• Institutions without the technical expertise to conduct acceptable environmental assessments, and

to develop and monitor adequate amelioration strategies. Very often the ecological and socio-economic monitoring required both for design prior to construction and afterwards to assess the effectiveness of amelioration measures is inadequate.

• Many developing countries have neither the necessary framework to ensure legal compliance nor

organised civil society to enforce recommended amelioration measures. In such situations the contractual arrangement with the donor may be the major means for ensuring compliance. However, in the absence of a transparent accountable system of compliance, this arrangement is rarely successful. More supervisory missions and closer monitoring have been recommended. Unfortunately this may cease when the contract expires and the donor considers the project �completed�.

Human capacity ! Adequate mitigation and compensation measures rely heavily on professional judgement and

adequate data. In many parts of the world, adequate pre-dam surveys are rarely undertaken, and the background information base for comparing options is not systematically available. Where such professionals do not exist, the data for making judgements are unavailable, or interdisciplinary working habits among professionals are weak, amelioration is less likely to be successful. Lack of knowledge and understanding hinders good planning and decision-making.

! In some countries biophysical and social science specialists are rare, while those recruited from

other countries may not understand the local rivers, human communities, or the likely constraints of working within the country. Foreign specialists also tend to eventually leave, often leaving inadequately trained local specialists to face any developing problems. In developing countries, failure rates of mitigation measures may be higher than in developed countries simply because of the lack of experienced professionals who can detect emerging problems.

4.4 How to Make Mitigation More Effective? From consideration of the monitoring reports referred to above and analysis of the constraints to mitigation, it is clear that even when technically feasible mitigation is not always successful. The wide range of constraints mean that successful mitigation is rarely achieved in practice. The survey has indicated that successful mitigation requires: • A good information base and competent professional staff able to formulate complex choices for

decision-makers; • An adequate legal framework and adequate compliance mechanisms; • Sophisticated and transparent co-operation between design teams and all stakeholders; • Adequate long-term monitoring and evaluation of mitigation effectiveness; • Adequate financial and institutional resources.



Only when all these aspects are in place do mitigation measures, even when technically feasible, have a chance of success. In theory, as the science base develops and the capacity gaps are filled, the amelioration of negative dam impacts should also improve. However, this process could take several decades. In the immediate future there is an urgent requirement for effective tools that allow environmentally sound development of river resources and the management of dams within this context. These tools should not only inform decision-making, but also form part of what is currently termed Environmental Management Programmes (EMP) that have become common practice as an outcome of the Environmental Impact Assessment requirement for the approval of new dam projects (WCD Thematic Review V.2 Environmental and Social Assessment for Large Dams). EMPs gather together all the mitigation and compensation measures for social and environmental effects. These programmes inform the decision making process for the planning, options assessment, design, and operation of all dam projects. Three tools have been highlighted in the information gathered by this review, or submitted directly to the Commission. 4.4.1 Indicators for Hydro-Project Site Selection. In a submission to WCD (INS082), the World Bank reviewed more than 20 completed hydroelectric dam projects in Latin America, along with several well-known projects from other regions (Ledec et al., 1997). They found that some large dams may be relatively benign in terms of environmental impact, while others have caused tremendous environmental damage. On the basis of this analysis, Ledec et al (1997) distilled six quantitative, easily-calculated indicators useful for hydro-project site selection. These six indicators (Box 4.7) are particularly useful because they have especially high predictive value for likely adverse environmental and biodiversity impacts and they require data which are relatively easy to obtain. Box 4.7 is augmented with additional indicators (not identified by Ledec et al., 1997) which are also useful in ranking hydroelectric project sites according to the expected severity of their adverse environmental impacts. Such indicators have recently been used in Colombia and Brazil to incorporate environmental concerns within power expansion plans. The exact ranking of potential new hydroelectric dam sites will vary somewhat according to the indicators used and the relative weight accorded to each. However, this methodology is remarkably robust, in that most dam sites tend to get broadly similar ratings, regardless of which combination of the environmental indicators in Box 4.7 are used. Box 4.7: Environmental Indicators To Guide Site Selection

Indicators identified by Ledec et al,(1997)

Good Site Other Useful Indicators Good Site

Reservoir Surface Area Small Critical Natural Habitats Affected*

Small

Water Retention Time in Reservoir

Short Fish and other species diversity

Low

Biomass Flooded Low Length of River Left Dry None/Short Length of River Impounded Small Likelihood of Reservoir

Stratification Low

Number of inflows to mainstream from undammed downriver tributaries

Large Useful Reservoir Life Long

Access Roads through sensitive natural areas (eg forests)

Short Lost Land-Based Production Small



* Not all habitats are of equal quality. Habitats that support large numbers of individuals and high biodiversity should be affected as little as possible.

An environmentally �good� large dam site will receive favourable ratings from most of these indicators (including small reservoir surface area with low hectares per megawatt ratio, short water retention time, short stretch of river impounded, low fish diversity, etc.) while a particularly �bad� site will receive unfavourable ratings from the same indicators (large flooded area with high hectares per megawatt ratio, long water retention time, long stretch of river impounded, high fish diversity, etc.). 4.4.2 Indicators of Ecological Integrity The universal challenge facing any ecosystem management or large-scale restoration programme is to translate the general and often amorphous goal defined by the public e.g. a healthy ecosystem � into a set of measurable attributes that, taken as a whole, add up to an assessment of overall ecological integrity. To achieve this, a reproducible and methodical procedure that draws heavily on scientific understanding, expert opinion and local knowledge needs to be developed, and applied in the specific political and social context aimed at forging a plan for ecosystem management (and restoration as appropriate). This approach of an ecologically-based conceptual framework has been used in the San-Francisco Bay-Delta Watershed (Box 4.8) to translate the abstract concepts of �ecosystem health� and �ecological integrity� into usable tools to guide the long-term process of large-scale ecosystem conservation and restoration. The purpose of such a suite of ecosystem indicators is to provide a scientifically valid definition of ecological integrity that can be used to help develop ecosystem management programmes and ultimately to determine whether its goal has been met. Such an approach can provide an important tool for river basin and water resource development. By establishing a set of key indicators from the outset it will be possible to assess development investments against these. The information and capacity constraints described for mitigation also apply to this approach however and success can only be incremental. Few areas have the technical information that is available for the San Fransisco Bay-Delta watershed. However by assembling even limited data in this focused manner a much more effective framework can be developed for examining water resource management options and resources focused upon filling critical information gaps. Box 4.8: Indicators of Ecological Integrity In a collective effort to work towards the restoration of the San Francisco Bay-Delta-River watershed system, the Environmental Defense Fund, CALFED staff, universities, research institutes, scientists, and other stakeholders have been working towards developing a suite of ecological indicators for the system. The goal is to try to translate the abstract concepts of ecosystem health and ecological integrity into usable tools to guide the long-term process of large-scale restoration. This approach is different to others because it focuses on positive ecological attributes rather than solely on indicators of problems. While an indicator of stress or response can show progress, it may leave out system components that are not currently seen as problems. Overall, the purpose of the suite of indicators is to provide a scientifically valid definition of ecological integrity that can be used to help develop a restoration program and ultimately to determine whether its goal has been met. Two essential characteristics of the framework are that both structural and functional attributes of ecosystems need to be assessed and that this should be done at a variety of scales. Structural attributes refer to the physical components of the system and their spatial relationships to one another. Functional attributes are the processes at work in the system. Determining ecological indicators at many scales helps to ensure consideration of the whole as well as the parts and ensures that large-scale processes work in harmony with processes and structures at smaller scales. Three different levels of scale can be the entire landscape, ecological zones, and habitat types for example. (Fujita 1999 /ENV096/ENV095/ENV090)



4.4.3 Environmental Flow Requirements (EFRs) One vital area of emerging scientific knowledge is the assessment of environmental flow requirements, which deals with the amount, timing, and conditions under which water should be released by dams, to enable downstream river ecosystems to retain their natural integrity and productivity. It is important to recognise that these are releases specifically for environmental purposes. They do not include flows necessary for downstream commercial or water supply purposes. Where flows are released for commercial as well as environmental purposes, the term instream flows should be used (Petts, 1996). Assessments of Environmental Flow Requirements (EFRs) are now used in 25, mostly developed, countries. Their use has increased in importance during the last three decades as it has become apparent that flow manipulations are causing serious degradation of river ecosystems. The level of costs involved is now sufficiently high for EFRs to be increasingly accepted worldwide as an essential tool for water-resource management, especially where downstream livelihoods may be threatened. EFRs in future need to be developed as part of dam design. However, for existing dams they can be introduced as a means of monitoring or restoring downstream ecosystems (see Boxes 4.9 & 4.10). Box 4.9: Case study: The Colorado River Between 22 March 1996 and 7 April 1996, water was released from Hoover Dam on the Colorado River, Colorado USA, into the downstream river in an effort to restore some of the features of the downstream river that had been lost as a result of a reduction in flooding. For the first four days, a steady release of 8 000 cubic feet per second (cfs) was made; on March 26th the release was increased at a rate of 4 000 cfs per hour (cfs/h) until 45 000 cfs was achieved at about mid-day, March 27th. The release was maintained at that level for seven days. Between April 2nd and 7th, they were decreased in a three-step fashion to maximise sediment deposition in the river. A team, consisting of aquatic scientists, engineers and managers, closely monitored the releases and their effects on the riverine ecosystem. The results were hailed as a success, with the releases achieving many of their objectives. These included the following: • at least 55 large, new beaches in the Grand Canyon were created; • more than half of the existing canyon beaches increased in size due to the flood, 37% remained

approximately the same size, and 10% lost small amounts of sediment; • the flood caused scouring of clay and vegetation bases in backwaters and marshes, thus providing

habitat for the humpback chub and other endangered fish species, and • in numerous backwater areas, the increased organic debris (primarily non-native plant species

growing very close to the banks of the river), which would not occur on the natural river, was cleared by the floodwaters.

Source : Press release: May 1996: Office Of The Interior Secretary. River scientists have developed a range of processes for assessing environmental flows that reflect their multidisciplinary understanding of rivers. Few would question that rivers with these flows in place would be in better condition than those that were simply exploited for water or as waste disposal facilities. Thus the process of establishing environmental flow assessments can be pronounced successful. Further success of EFRs can be judged at two other levels: the effect of EFRs on national attitudes, and the effect of a specific EFR on an individual river. At the national level, there is no doubt that the concept of EFRs has been successful. A growing number of countries now recognise the need for EFRs, and are either searching for or developing suitable methodologies to do them or adopting tried and tested approaches from elsewhere. The Instream Flow Incremental Methodology (IFIM) has long enjoyed legal status in America and the US Federal Energy Regulation Commission requires that operators of many hydro-dams release



environmental flows as a condition of renewing their dam licenses. Box 5.7 shows one innovative example on the Colorado river. In Spain, 10% of the mean annual runoff (MAR) of a river should be released from dams as environmental flows, which although probably insufficient to sustain the downstream environment, at least acknowledges the need for environmental flows (McCully 1996). The South African Building Block Methodology (BBM, King and Louw 1998) convinced lawyers re-writing the country�s Water Law that environmental flows could be calculated in a scientific and defensible way. This led to flows for maintenance of aquatic systems being recognised within the country�s new Water Law as one of only two sectors with a right to water, the other being basic human needs. It is important to remember however that even the most successful EFR will only partially mitigate against the effects of a dam on a river. The physical presence of a dam will, in itself, inevitably result in impacts on the downstream environment related to, inter alia, trapping of sediment, reduction in flow variability, and changes in the temperature and chemical composition of the water, with knock-on social and economic impacts. Nothing is gained at no cost; if flow regimes are manipulated the targeted rivers will change. Society decides, pro-actively or through neglect, the extent of that change. More detail of EFRs are found in Annex 6 and in King et al (1999) report to WCD. Box 4.10: Case study: Kromme River The Kromme River estuary is situated in the Eastern Cape, South Africa. Under natural conditions, aperiodic floods scoured out the estuarine channels, and maintained the biotic diversity in the estuary. However, reservoirs in the Kromme River catchment currently dampen all the floods smaller than the 1:30 year events. The existing environmental flow allocation allows for a single release of 2m3/s-1 per annum for maintenance of the Kromme estuary. As part of the SA Department of Water Affairs and Forestry�s investigations for the new Water Law in that country, a multi-disciplinary study was commissioned to evaluate the response of abiotic and biotic components in the Kromme estuary to an experimental release of 2m3s-1 from Mpofu Reservoir. The objective of the release was to create freshwater conditions throughout the upper half of the estuary. However, the release resulted in the water column becoming highly stratified for about two weeks after which the estuary returned to its marine-dominated pre-release condition. The release also provided no direct or indirect advantages to zooplankton or other biota in the estuary. No scouring took place. It was concluded that the release was too small to effect the desired changes, and that a regular baseflow combined with freshwater pulses into the estuary would be more beneficial to the estuary (Wooldridge and Callahan, in prep).

4.5 Conclusions In response to the identified impacts of dams on natural ecosystems and species, four principal approaches: avoidance, mitigation, compensation and restoration have been developed, and are now promoted as solutions to these impacts. Obviously since avoidance results in no change to the existing functioning of a particular ecological area or resource it is the most acceptable approach. This chapter has reviewed the use of these different measures, and indicates that the most widely used approach, mitigation is particularly problematic. It is clear that there are always residual impacts that cannot be mitigated. Many impacts are technically impossible to mitigate if dams are to provide their planned outputs. While there is experience of good mitigation, this success is nevertheless contingent upon stringent conditions of: • a good information base and competent professional staff able to formulate complex choices for

decision-makers;



• an adequate legal framework and adequate compliance mechanisms; • a co-operative process with the design team and stakeholders; • monitoring of feedback and evaluation of mitigation effectiveness, and • adequate financial and institutional resources. If any one of these conditions is absent, then the ecosystem functions will likely be lost. In practice the extent to which these conditions are met varies enormously from country to country and dam to dam. The review therefore concludes that mitigation, though often possible in principle (as demonstrated by IEA�s thorough review), has many uncertainties attached to it in field situations and is therefore at present not a credible option in all cases and all circumstances. In addition the weaknesses of the EIA process for many projects (cf Thematic review V.2) reduce the possibilities for positive outcomes. There should therefore be a strategy of avoidance and minimisation rather than one of mitigation if the aim is to maintain biodiversity, and ecosystem functions and services for the foreseeable future.



5. Trends in the International Debate/Approach to Dams 5.1 Introduction For much of modern human history water management has focused on the direct provision of enough water for people to drink, grow their food and support industries. This has changed however in recent years as the inter-relationship between renewable natural resources and human development has received greater international attention. Drawing upon this philosophy, and based upon principles developed at water conferences in Mar del Plata (UN, 1977) and in Dublin (WMO, 1992), Chapter 18 of Agenda 21 develops the concept of Integrated Water Resources Management (IWRM). This treats �water as an integral part of the ecosystem, a natural resource and a social and economic good, whose quantity and quality determine the nature of its utilisation� (UN, 1992). Central to an IWRM approach is the protection of surface water and groundwater resources, water quality and aquatic ecosystems and the management of land and water in an integrated way. The issues raised by dams are central to IWRM and to the wider debate on sustainable use of the world�s water resources. This is particularly so as in recent years the costs and benefits of dams to human society have been questioned. Over the past two decades the multiple values of natural ecosystems to human society and the environmental consequences of dams have become more widely understood. Opponents of dams argue that in many instances the environmental and social costs outweigh the economic benefits gained (by some) from dam construction and that better options than dams are available. They further argue that the benefits and costs are distributed unequally and that as knowledge of the impact of dams increases the opposition to dams continues to grow. Dam proponents maintain that large dams are essential for the well-being of many millions of people and have played a key role in human development. They argue that many of the alternatives to dams are at present either uneconomic (eg desalination), constrained by technical limitations (eg. options for solar and wind power) or are potentially more environmentally damaging (eg thermal and nuclear power stations). They also argue that many of the negative impacts of dams can be mitigated and that increasing human population will require increased dam construction in the future. Although often presented as a simple dichotomy, the argument is in reality very complex and less polarised than often presented; positions fall out along a "dams-no dams" continuum, and may vary with the dam. The key issue is whether in the long run any particular dam will or will not provide a net benefit to humankind, and how such a judgement is reached. Science can help illuminate the discussion, but the decision ultimately is a political one. 5.2 Summary of the debate Looking into the future, it seems highly likely that the scarcity value of freshwater will greatly increase, as will the value of energy produced in ways that do not contribute significantly to atmospheric carbon dioxide and other greenhouse gases. Those who promote the hydropower perspective argue that hydropower provides 20% of electricity globally. Further, hydropower is critically important for many countries; hydropower produces more than 50% of electricity for 65 countries; for 24, more than 90%; and for 10, practically all energy for everything except transportation. The irrigation sector points to increased population growth and the need to improve supplies that lead it to predict a 17 % increase in water storage for irrigation purposes in the next two decades (Report of the World Commission on Water, February 2000). Those concerned with rural poverty and access to common property resources point to the essential role played by floodplains and deltas in sustaining livelihoods in many African, Asian and South American countries. These communities often see water diverted to other uses, frequently depriving them of a secure and diverse resource base that comes with combining terrestrial and aquatic production systems(see Thematic review I.1).



Since the mid-1980s a well-organised international movement against current dam building practices has evolved. Dam opponents do not view themselves as simply anti-dams, but rather as advocates of what they feel are more sustainable, equitable and efficient technologies and management practices. For this reason they have in many countries become aligned with environmental and human rights organisations. Their principal argument is that existing dams have largely failed to meet their economic and social objectives and at the same time resulted in significant environmental and social damage (McCully, 1994). Anti-dam groups have called for a moratorium on the construction of all new large dams that fail to comply with a set of 17 conditions listed in the San Francisco Declaration (1988). Conversely, dam proponents maintain that large dams have played a key role in human development throughout the world and significantly improved the well-being of humankind. They contend that as human population grows and expectations of higher standards of living increase, global water demand inevitably will increase. They argue that to meet this increased demand more dams will be needed in future (ICOLD, 1997; Le Cornu, 1998b). They maintain that the anti-dam movement has tended to focus on very narrow issues, and through sustained campaigning has swayed public opinion against dams. They feel that educating the public about wider water resource issues will provide them with a more balanced view of dams. Table 5.1 highlights the main points of each side. 5.3 Summary of Trends In the face of this evolving debate within civil society a number of significant steps have been taken at international level and today set the context within which future steps to address the ecological impacts of dams need to be considered. There are four principal groups: • those institutions related to dam construction, (e.g. IEA and ICOLD); • those institutions that facilitate funding for construction (e.g. World Bank and OECD); • those organisations largely opposed to continued large-scale dam construction (e.g. International

Rivers Network), and • governments or �international community� whose stance is expressed through international

Conventions that set international environmental standards. The current approach of each is summarised in Table 5.1. 5.3.1 IEA In a review of hydropower and the environment (IEA, 2000) IEA has summarised trends in industry�s perspective on the planning of hydropower projects (Table 5.2). They highlight that in the past large dam projects generally recognised a future demand (water, power) to be met in a least cost manner, and that planning procedures consequently minimised mitigation of environmental and social impacts. Today the emphasis has switched to avoiding unnecessary delays or expenditure and effort on projects which in the end will not be carried out, and consequently planning procedures in developed countries are often geared toward minimising business risk and maximising acceptance. This new planning approach presents and discusses as early in the planning stage as possible all the costs and benefits of competing scenarios with interested parties, including the persons directly affected by the project and NGOs, taking into account technical, economic, financial, environmental, social institutional, political and risk factors. This is followed by formulation and review of alternative plans and major investment in trying to reach consensus about the best plan to be adopted for implementation.



Table 5.1: Distillation of arguments used by proponents and opponents of large dams Proponents Opponents Through provision of reliable water supplies, production of energy and creation of recreational opportunities, dams have improved the economic and social well-being of many millions of people.

Through inundation of huge tracts of inhabited land and destruction of the natural services provided by downstream ecosystems, dams have destroyed the livelihoods and reduced the well-being of many millions of people. The social and economic benefits promised for large dams have in many cases not been realised.

Dams are the most important means of making surface water available at the place and time of demand. Although there are non-structural alternatives (eg demand management), more dams will be needed in the future to manage the world�s limited water resources.

Dams are only one way of providing water when required. Other options such as demand management, rainfall harvesting and the tapping and recharging of groundwater or desalination of seawater could reduce our dependence on the construction of dams.

Dams create new habitat through the creation of reservoirs, which, although detrimental to some species, provides opportunities for others.

Through creation of reservoirs, dams have flooded and so destroyed many pristine biotopes, with consequent negative impacts on biodiversity. Dams transform �healthy� river ecosystems into impoverished reservoirs.

Downstream from dams, the destruction of ecosystems resulting from the disruption of the natural flow regime can be mitigated by release of compensation flows that simulate both the high and the low discharges of the natural flow regime.

Downstream from dams, disruption of natural flow, sediment and energy dynamics destroys the integrity of many ecosystems. Although it is possible to mitigate against some of the negative effects, it is impossible to undo all the damage.

Large dams provide flood protection, and so increase the security of many millions of people who live downstream from them.

Large dams protect from regular annual floods but often fail to hold back floods of longer return periods. Dams lead people to believe that floods are controlled and so lead to increased development of floodplains. Then when a large flood does come, damage caused is often greater than it would have been without the dam. If a dam fails the consequences can be devastating. Thus dams reduce, in a very tangible way, the security of people living downstream.

The health risks associated with dams and associated projects were not appreciated in the past. We now understand the risks and so can mitigate against the causes. Furthermore dams, by increasing economic status, can provide the impetus for improved health care.

Many large dams and the projects associated with them (eg irrigation schemes) create health risks for many people who live in their vicinity. The health risks associated with the workforce bringing in disease during dam construction are an additional hazard.

Hydropower represents a �clean� sustainable energy source. Many of the alternatives to hydropower (eg nuclear and coal fired power stations) create greater environmental and social-economic problems.

Hydropower is not a �clean� energy. By altering chemical and thermal regimes, reservoirs effectively pollute rivers and destroy downstream ecosystems. Furthermore, reservoirs may contribute to greenhouse gases (i.e. decomposition of submerged vegetation releases carbon dioxide and methane). Modern technologies (eg solar power) provide new opportunities that enable us to reduce our dependence on large dams.

Over the last 20 years, environmental issues have come to the fore, but people (especially in the developed world) are not prepared to make the changes to their life styles that doing away with large dams would entail.

Over the last 20 years societal values have changed (especially in the developed world). Environmental damage is no longer accepted as an inevitable consequence of human development. Indeed, it is now recognised that continued environmental degradation is non-sustainable.

Overall: In the past mistakes have been made. We now have a greater understanding of the negative ecological, socio-economic and health consequences of large dams, and to a large extent these can be mitigated against. The scope for reducing any detrimental impacts on the environment through alternative solutions, project modifications in response to particular needs, or mitigating measures should be thoroughly investigated, evaluated and implemented (ICOLD, 1997). The benefits of dams outweigh the costs. In many cases the alternatives are associated with far greater costs. The negative impacts of large dams on the environment are sometimes overstated. Dams can sometimes enhance environmental conditions. Dams are now an essential part of the basis for human survival. More dams will be needed in future. The future for large dams should be bright.

Overall: Large dams should be a last resort after �less damaging and costly alternatives for flood management, transportation, water supply, irrigation and power supply are exhausted" (IRN, 1994). In the past, dams have not lived up to the promises made for them. The ecological, socio-economic and health costs associated with their construction are now recognised. In many cases these costs outweigh the benefits. The price paid is too high. Alternatives to large dams exist. The era of large dams should be brought to an end.



While this is a time consuming approach, it brings the major advantage that by attempting to reach a consensus amongst all parties concerned at an early a stage, it is possible to avoid last minute surprises after years of development expenditures, as has happened with several large dam projects in the recent past. Table 5.2: Trends in the Planning of Hydropower Projects (IEA, 2000)

Old Planning Concept New Planning Concept A hydro project is a technical scheme to: ! Provide basic technical infrastructure to

improve supply of power/water

A hydro project is part of an integrated set of technical, environmental and social measures to: ! cover basic needs of people in a sustainable manner

(water, light, power) ! accelerate rural development to improve the welfare of

people in the region � particularly those directly affected by the project

! improve environmental and flood protection ! combat global warming

Planning is government responsibility, often assisted by international development agencies

Planning involves many partners/stakeholders: ! government ! people affected ! non-governmental organisations ! private sector developers ! financing institutions

Least-cost planning procedure: ! identify least-cost project to cover

power/water needs ! carry out unavoidable social and

environmental impact mitigation at minimum cost

! carry out detailed studies

Multi-criteria planning procedure: ! project(s) must be part of sectoral development plan ! rigorous study of project alternatives, including the

No-Project option ! prepare comprehensive comparison matrix showing

pros and the cons of each alternative from technical, environmental, social, economic, financial, risk and political perspectives

! quantify secondary and external costs and benefits as well as risk

! reach consensus among stakeholders about overall best alternative to be developed

! carry out detailed studies Public Sector Project: ! developed and owned by government ! funding partly from international

development agencies

Private/Public Sector Project: ! developed and owned by private sector, with or

without government participation ! finance largely from commercial sources ! international development agencies act as catalyst for

project funding by providing guarantees ! access to semi-concessional funding if stringent

international guidelines for social and environmental impact mitigation are followed

5.3.2 International Commission on Large Dams (ICOLD) In a recent position paper on dams and the environment, ICOLD gave great importance to addressing the environmental and social aspects of dams and reservoirs (ICOLD, 1997). It recognised that the impact of dams and reservoirs on both the natural as well as the socio-economic environment is �inevitable and undeniable�. It argued that in the quest to provide growing numbers of people with a better life, the need to develop natural resources, including water, means that the natural environment cannot be preserved completely unchanged. However, the paper accepts that �great care must be



taken to protect the environment from all avoidable harm or interference� and the aim of humankind must be to balance the need for the development of water resources with the conservation of the environment in a way that will not compromise future generations. The paper argues that all dam projects must be judged by the state-of-the-art of the technologies involved and by current standards of environmental care. It agrees that a comprehensive Environmental Impact Assessment should be a standard procedure as part of the project conceptualisation. However, it does not provide guidance on exactly what should be included in this assessment, nor how findings should be used to assist the decision making process. Nevertheless, it does state that �special attention should be paid to any effects on biodiversity or the habitat of rare or endangered species�. The document states that in future, environmental and social aspects of dams �should be addressed with the same concern that has made the question of dam safety a predominant concept� for the industry. 5.3.3 The World Bank The World Bank advocates the application of environmental assessments to provide information about the ways new economic activities (including projects that involve dam construction) may directly or indirectly affect ecosystems. The Bank requires that the environmental assessments carried out for the projects it supports reflect the views of persons affected by the project, including the poor, indigenous people, and disadvantaged groups. The Bank will not support projects which involve the significant conversion or degradation of critical natural habitats unless there are no other feasible alternatives to the project and its siting and the overall benefits from the project substantially outweigh its environmental costs. The current Bank policies and guidelines that are of most relevance specifically to ecosystem impacts of large dams are published in the Bank's Operational Handbook (World Bank, 1999) and include: World Bank Operational Policy 4.04 (Natural Resources): The Bank� supports the protection, maintenance and rehabilitation of natural habitats and their functions�and expects borrowers to apply a precautionary principle to natural resource management�.The Bank promotes and supports natural habitat conservation and improved land use by financing projects designed to integrate into national and regional development the conservation of natural habitats and the maintenance of ecological functions. The Bank does not support projects that in the Bank's opinion involve the significant conversion or degradation of critical natural habitats. World Bank Operational Policy 4.01 (Environmental Assessment): The Bank requires environmental assessment (EA) of projects proposed for Bank financing�EA evaluates a project�s potential environmental risks and impacts in its area of influence, examines project alternatives, identifies ways of improving project selection, siting, planning, design and implementation by preventing, minimising, or compensating for adverse environmental impacts and enhancing positive impacts�EA considers natural and social aspects in an integrated way. World Bank Procedure 4.01 (Environmental Procedures); This contains details of how environmental assessments should be conducted and notes that EA reports will be made public through its Infoshop. An appendix (Annex B) provides specific details on the application of EA to Dam and Reservoir projects.



World Bank Operational Policy 4.07 (Water Resources Management): World Bank policies are to promote economically viable, environmentally sustainable and socially equitable water management. 5.3.4 New approaches of the Organisation for Economic Co-operation and

Development (OECD) OECD members agree to ensure that �development assistance projects and programmes which, because of their nature, size and/or location, could significantly affect the environment, should be assessed at as early a stage as possible and to an appropriate degree from an environmental standpoint�. The exploitation of hydrological resources is highlighted as the type of project where an environmental impact assessment (EIA) should be conducted. In conducting EIAs, the following main elements are particularly relevant to large dam construction and should be included: ! consideration of alternative project designs (including the no action alternative) as well as

required mitigation and monitoring measures; ! an assessment of off-site effects including transboundary, delayed and cumulative effects, and ! a clear statement of significant beneficial and adverse environmental and related social effects and

risks of the project. The OECD guidelines recommend that in conducting EIA�s donors should �use the standards that will achieve the minimum level of �acceptable�, non-mitigable negative effects and maximise the positive effects. However, no definition of �acceptable� is given. 5.3.5 The International Movement Against Large Dams Throughout the world there are a large number of non-governmental organisations that oppose dam construction for a wide variety of social, economic and environmental reasons. However, the most prominent of these groups, one that campaigns at a global scale and perhaps best articulates the overall position of the anti-dam lobby, is the International Rivers Network (IRN). The IRN states its mission as: ��to halt and reverse the degradation of river systems; to support local communities in protecting and restoring the well-being of the people, cultures and ecosystems that depend on rivers; to promote sustainable, environmentally sound alternatives to damming and channelling rivers, to foster greater understanding, awareness and respect for rivers�� IRN works with environmental and human rights groups on specific key projects around the world whilst at the same time working to alter global policies. It argues that critical review of large dam projects indicates that organisations such as the World Bank rarely, if ever, fulfil their own guidelines relating to dam development. IRNs position on dams is summarised through a series of declarations including the San Francisco Declaration (1988), the Manibeli Declaration (1994) and most recently the Curitiba Declaration (1997). All these declarations call for a moratorium on large dam building until certain criteria are fulfilled (Box 5.1) .



Box 5.1: The Curitiba Declaration This declaration signed by representatives of dam-affected people and of opponents of �destructive� dams from 20 countries calls for a moratorium on the building of large dams until: i) There is a halt to all forms of violence and intimidation against people affected by

dams and organisations opposing dams; ii) Reparations, including the provision of adequate land, housing and social

infrastructure, be negotiated with the millions of people whose livelihoods have already suffered because of dams;

iii) Actions are taken to restore environments damaged by dams � even when this requires the removal of the dams;

iv) Territorial rights of indigenous, tribal, semi-tribal and traditional populations affected by dams fully respected through providing them with territories which allow them to regain their previous cultural and economic conditions;

v) An international independent commission is established to conduct a comprehensive review of all large dams financed or otherwise supported by international aid and credit agencies, and its policy conclusions implemented. The establishment and procedures of the review must be subject to the approval of the international movement of people affected by dams.

vi) Each national and regional agency which has financed or otherwise supported the building of large dams have commissioned independent reviews of each large dam project they have funded and implemented the policy conclusions of the reviews. The reviews must be carried out with the participation of representatives of affected people�s organisations.

vii) Policies on energy and freshwater are implemented which encourage the use of sustainable and appropriate technologies and management practices, using the contributions of both modern science and traditional knowledge. These policies need also to discourage waste and over consumption and guarantee equitable access to these basic needs.

5.3.6 Requirements of International Conventions The Convention on Biological Diversity (CBD), to date ratified by 178 countries, explicitly recognises the links between biodiversity conservation and sustainable development. It acknowledges that biological diversity is more than just the sum of species numbers; it encompasses the variety, variability and uniqueness of genes and species and of the ecosystems in which they occur. The Convention�s overall objectives include the conservation of biological diversity, the sustainable use of its components, and the fair and equitable sharing of the benefits arising from the utilisation of genetic resources. As of February 2000, 122 countries have also ratified the Convention on Wetlands (Ramsar 1971) which provides broad guidelines on reducing the impact of development projects (eg dams) on wetlands (Box 5.2).



Box 5.2: Ramsar Convention: Guidelines for Contracting Parties relating to reducing the impact of water development projects on wetlands 1. Ensure that proposals for water development projects are carefully reviewed at their initial

stages to determine whether non-structural alternatives may be feasible. 2. Take all necessary actions in order to minimise the impact of water development projects

on biodiversity and socio-economic benefits during the construction phase and longer term operation.

3. Ensure that project design/planning processes includes a step-by-step process to integrate

environmental issues, especially initial biodiversity/resource surveys, and post project evaluation and monitoring.

4. Incorporate long-term social benefit and cost considerations into the process from the very

initial stages of project preparation. 5.4 Areas of Convergence/Divergence It is clear from this review of current trends that considerable steps have been taken to address the environmental concerns surrounding dams. Indeed there are today many areas of broad agreement between those who are generally supportive of building dams and those who are generally philosophically opposed to building large dams. These include: ! The public policy issues affecting dam construction are to be addressed with the active

participation of governments, the private sector, and civil society. ! The determination of the size, location, and type of dam depends on the economic, technical,

social, and environmental flexibility of the proposal. ! Dams require a physical area for the dam body and the reservoir. They change the natural water

conditions, inducing changes in both physical and biological aspects of the river basin affected, including the humans living in the area. They are built to change the way water is managed, with the objective of satisfying important needs identified by people.

! Projects should be judged everywhere by �current standards of environmental care� (ICOLD,

1997). In general opponents of dams do not believe that no dams should ever be built (IRN, 2000) and they accept that at least some of the adverse environmental effects of dams can be mitigated.

! In general, proponents accept that decisions about new dam projects must be based on

�unequivocally realistic economic analysis� that neither �overstates the benefits� nor �understates the costs� and that such analysis requires �taking the impacts on the natural and social environment into account� (ICOLD, 1997). Opponents of dams state that �the claims of project promoters of the economic, environmental and social benefits and costs� of projects should be �verified by independent experts� (IRN, 1997)

At the same time, the analysis carried out indicates that besides agreements, a wide range of important and fundamental differences remain. At the most general level these differences relate to the value systems adhered to by the different groups involved and especially the value to be attached to the intrinsic value of nature.



In very broad terms the dam debate is emblematic of the wider on-going debate about the future direction of human development. In the past the argument that more people require more water, more food and more energy and thus more dams was widely accepted. Today, this apparently logical argument is broadly accepted by proponents of dams but is questioned by dam opponents. There remain four main areas of difference: • the integration of social, economic and environmental concerns; • the question of equity of distribution of costs and benefits; • the methods of the decision-making process, and • the extent to which impacts can be mitigated or compensated. Differences arise when it comes to evaluating the �true� costs and benefits (both direct and indirect) of projects, particularly those environmental and social impacts which are difficult to quantify in monetary terms. While proponents accept that these need to be taken into account, opponents of dams argue that existing methods of project appraisal result in unrealistically optimistic assessments of the benefits of dam projects and that not all values can be expressed in economic terms. This debate is driven by the different value systems of the different ethical perspectives described in Section 3.9.2 and referred to above. Project life-cycle analysis does not eliminate the need for value judgements and arbitration because many impacts are at present impossible to compare directly. Clearer guidelines are therefore required on how costs and benefits can be distributed among those people affected by all development projects (not just dams). Techniques need to be improved to offer better methods of economic valuation that are acceptable to both proponents and opponents of dams. It is important that development projects are appraised in a transparent manner using multi-criteria analysis, not just economic cost-benefit analyses or using a purely eco-centric view of the world.



6. Conclusions and Policy Recommendations for WCD 6.1 Conclusions This review has highlighted the value of natural ecosystems to human society, giving particular attention to the specific goods and services provided by those ecosystems that are impacted most by dams. While some of these ecosystem values are non-monetary in nature, such as the aesthetic, cultural and heritage value of specific habitats and landscapes, the direct and indirect economic value of these services is highly significant to local, national and regional economies. In most cases one or more sector of society depends upon these values (e.g. fisheries, grazing) while in some the total value of the benefit of natural ecosystems can exceed the value of the benefits derived from dams and associated investments in agriculture (Barbier 1996). In the past, the failure to take into account the cost of the consequences of dams has resulted in the benefits of many dams being overstated. The importance of these natural ecosystems is today widely recognised by national governments and the importance of efforts to preserve these ecosystems and harness their values sustainably is enshrined in a series of international agreements, notably the Convention on Biological Diversity and the Ramsar Convention on Wetlands of International Importance. See section 5.3.6. The review has underlined that dams have a wide range of major impacts upon natural ecosystems, that most of these are negative, that many are irreversible, and that they are manifest in economic and social costs. Perhaps surprisingly, the review has noted that there is today widespread, but not complete, agreement as to the reality and importance of these impacts and their costs. The review has also recognised the growing understanding of the threats to the world�s biodiversity, and the particularly acute threats to those species that are dependent upon freshwater. By altering the quantity and quality of water available to natural riverine ecosystems, dams add to these already significant threats. In response to the identified impacts of dams on natural ecosystems and species, four principal approaches: avoidance, mitigation, compensation and restoration have been developed, and are now promoted as solutions to these impacts. A review of these approaches highlights, however, that while there is good evidence through practical experience that each of the individual measures can be successful in specific cases, there are problems with them all. The most widely used approach, mitigation, is particularly problematic. As with any kind of human development, whatever amelioration measures are utilised, dam building will always result in some environmental and ecosystem impacts While there is experience of good mitigation, this success is nevertheless contingent upon stringent conditions of: • a good information base and competent professional staff able to formulate complex choices for

decision-makers; • an adequate legal framework and compliance mechanisms; • a co-operative process with the design team and stakeholders; • monitoring of feedback and evaluation of mitigation effectiveness, and • adequate financial and institutional resources. If any one of these conditions is absent, then the ecosystem values will be lost. In practice the extent to which these conditions are met varies enormously from country to country and dam to dam. The review therefore concludes that mitigation, though often possible in principle, will, under present political, economic and institutional conditions, rarely be successfully implemented. Alternative approaches to maintaining ecosystem health therefore need to be pursued.



6.2 Recommendations In light of the above, ten recommendations are submitted to the WCD. 1. Recognise the important role of natural ecosystems in contributing to sustainable development. If river basin development is to be truly sustainable it needs to recognise the wide range of goods and services that are provided to human society by natural ecosystems. All major development investment, including dam construction, should seek to conserve and enhance these ecosystems and their value to society. Actions that diminish these values should be minimised. 2. Recognise the importance of biodiversity and promote its conservation. Biodiversity is recognised internationally as a uniquely important, but endangered, feature of our planet. In the face of unprecedented rates of species extinction in recent decades every effort needs to be made to minimise threats to biodiversity. In the past dams have contributed significantly to endangerment and extinction of species. In future no dam should proceed if it is shown to have a high probability of having a significant detrimental effect on species diversity. 3. Dams must be considered within a framework of river basin management plans and international/national/regional policies. They must be evaluated alongside other options for water supply, irrigation and electricity production. In any situation, the environmental costs and benefits of the full �life-cycle� of the various options must be compared. This must include the costs of decommissioning dams that have come to the end of their useful life. 4. Recognise and manage for uncertainty. There is enormous variation between river basins, rivers, ecosystems, dams and associated projects. This diversity, together with the seriously limited quantity and quality of information on the functioning of specific natural ecosystems, and on species diversity and resilience in different habitats affected by dams, contributes to a very limited capacity to predict the precise impact of dams on natural ecosystems and biodiversity. Such a high degree of uncertainty and limited predictive capacity argue forcefully for adoption of a precautionary approach to dam development. Wherever possible dams and their impacts should be avoided. Where avoidance is not possible, capacity to manage the dam in a flexible manner and so adapt to improved understanding of ecosystem requirements, should be incorporated into dam design. This precautionary approach should be recognised as a central feature of planning, design and management of dams, especially as many are probably irreversible. 5. Ensure effective multi-sector participation in planning, design and management of dams. In order to help recognise and reduce uncertainty it is essential that all dam projects and their impacts are subject to intensive analysis during planning and design. This needs to be pursued through open processes that ensure that there is full sharing of available information, and recognition of areas where that information is not sufficient to predict the impact of dams or design successful mitigation measures with any confidence. This participatory process also needs to identify who should assume responsibility for the ecosystem impacts of dams and therefore take on their true costs, ensure their mitigation or compensation (as appropriate) and restore, where possible, the river at the end of a project�s life. 6. Maximise adaptive capacity. When the participatory design processes recommended above leads to a decision to construct a dam as the best option for sustainable development in the river basin, design features should include the capacity to adjust operation to adapt to the lessons of experience, improved knowledge, or changing ecological requirements. Such design features include in particular sluices or gated spillways that will allow Environmental Flow Releases of appropriate water quality. This approach needs to be accompanied by a programme of independent environmental monitoring that will allow continuous tracking and regular assessment of the impact of the dam and its operation upon downstream ecosystems. This information needs to be fed back into to an adaptive decision



making process. Mechanisms must be established to ensure compliance with recommendation on dam operation from monitoring bodies. 7. Promote incorporation of environmental management features into dam design. In addition to features that provide for adaptive management as a permanent element of the dam cycle, dams should also be designed to include all appropriate environmental features for improving water quality. These include variable level off-takes, shallow plunge pools, fish passes, regulating weirs etc.. 8. Promote the development of national legislative frameworks. Ultimately the measures recommended here, together with recognition of the need to fulfil international commitments with regard to ecosystems and biodiversity, need to be enshrined in national legislation governing dams and river basin development. This should be promoted together with measures to strengthen enforcement, such as the use of environmental bonds, direct compensation revenue sharing (hydropower), or environmental trust funds as a guarantee of compliance. 9. Promote application of tools to foster ecosystem health.

(I) Environmental Flow Releases. EFRs are being used in 25 countries and today serve as the single most important tool for managing the ecosystem and associated impacts of dams. EFRs should be a requirement for all future dams. Blanket minimum flow requirements, such as �10% minimum flow� do not address the needs of riverine ecosystems. Taking account of the dynamic nature of rivers requires optimum flows, often including periodic managed floods. An intensive investment should be made in developing further the knowledge-base required to improve this tool adapting it to local needs, extending it to include explicit support for social downstream needs.

(II) Ecosystem Health Indicators. In order to engage in a proactive discussion on the

requirements for maintaining (or restoring) healthy ecosystems, greater investment should be made in the development of indicators of ecosystem health. These can be used for setting targets for mitigation, compensation and restoration of ecosystems impacted by dams.

(III) Site Selection Indicators. The World Bank has identified six key indicators of site

selection that help minimise ecosystem impacts: reservoir surface area; water retention time in the reservoir; biomass flooded; length of river impounded; number of inflows to mainstream from undammed down-river tributaries; and access roads through sensitive areas. Use of these Indicators should be promoted and refined on the basis of experience.

10. The role of every dam should be periodically reviewed and its value to society re-evaluated. Consideration should be given to decommissioning, retrofitting modern technologies and/or altering dam operation so that where feasible, dams are improved to comply with up-to-date standards of environmental care. 6.3 Options for Operationalising the Recommendations It is recognised that the higher level recommendations in this chapter provide a framework for addressing ecosystem issues, yet for their effective implementation they require a level of operational detail that these formulations do not allow. Some also link directly to similar recommendations emanating from other WCD thematic reviews. The Commission may wish to attach measurable norms or standards to these recommendations in order to move from the policy framework towards implementation. This document does not intend to



be prescriptive at this level, however Table 6.1 explores some of the options open to the Commission in establishing more operational sub-principles for dams and ecosystems. Table 6.1: Options for establishing sub-principles under each recommendation

Recommendation Operationalisation Option 1. Recognise the important role of

natural ecosystems in contributing to livelihoods and sustainable development.

• Assess the value of natural ecosystem functions and services during feasibility studies for both upstream and downstream communities. Water does not have a zero opportunity cost.

• Adopt a river basin approach to water and ecosystem management.

• Use multi-criteria decision-making that recognises values other than costs and not simply a cost-benefit analysis.

• Rivers are a public trust and extractive use rights should not be permanently allocated.

• Identify users of the downstream and upstream natural resources and ensure that their needs are incorporated by the project (link with thematicI.1).

2. Recognise the importance of

biodiversity and promote its conservation.

• Ensure that dam projects comply with the provisions and guidance of the Convention on Biological Diversity, the Convention on Wetlands and other related nature conservation Conventions.

• Dams should not negatively impact any Red Data Book species.

• Dams should not be built in, nor should reservoirs impinge on, declared National Parks or Nature Reserves.

• Undertake comprehensive biodiversity surveys of rivers in order to allow the least ecosystem-damaging choices and trade-offs to be made. Ecological investigations should be placed on the same footing as engineering and economic assessments during project planning and not be �add-on� extras.

• River flows should not be reduced during commissioning to zero or levels likely to have a negative impact on biodiversity.

• Dams built on tributaries will have fewer impacts on migrating fish than those on the main stream.

3. Recognise and manage for

uncertainty • Recognising that the precautionary principle should

apply as it is at present impossible to predict all consequences of dam construction.

• Undertake baseline assessments of the riverine ecosystem and its biodiversity down to the river mouth, during feasibility studies to provide the information base needed to improve predictive capacity.

• Assess possible cumulative impacts. 4. Ensure effective participation in

planning, design and management of dams.

• All EIA studies should be public documents. • All stakeholders must participate in EIA and decision-

making process. Utilise decision-making tools that



encourage interdisciplinary discussion and stakeholder participation.

• Base decision-making on multi-criteria analyses, not simple cost-benefit analyses.

• Link with �participation� thematic review : V.5. 5. Maximise adaptive capacity. • Allow for regular monitoring of ecosystems to ensure

that management objectives are met. • Every existing dam should have an Environmental flow

requirement. • Review operating rules every five years to incorporate

findings of monitoring programmes and mitigate unexpected ecosystem changes.

• Undertake research programmes to solve outstanding problems.

• Ensure every dam has a proposal for how it will eventually be decommissioned, especially with regard to design features for reservoir drainage, the treatment of accumulated sediment, and appropriate financial measures for ecosystem restoration.

• Include direct revenue sharing (hydropower), or environmental trust funds as tools to ensure funds for monitoring and repairing ecosystem damage are available throughout the project�s lifetime.

6. Promote incorporation of

environmental management features into dam design

• Include biologists and ecologists in the design team. • Every dam on a river with migratory fish should have an

effective fishpass and monitoring programme. • Dams that have pulsing flood releases due to hydropower

should systematically have a downstream re-regulating weir that levels out day to day flow oscillations.

• Where water quality is, or is likely to be, an issue, variable level offtakes should be mandatory.

7. Promote the development of

national legislative frameworks • Use environmental bonds as a guarantee of compliance. • Ensure the developer/owner is responsible for managing

dam-related ecosystem impacts and restoring the site at the end of the project�s lifetime.

• Ensure that the environmental components of dam tenders are at a fixed cost and are not subject to the competitive tendering process that is used for the infrastructure.

• Undertake regular independent audits of environmental performance of major projects where complex mitigation measures are planned.

• National legislation should include provision for an intact rivers policy.



7. References Abdel Megeed, A. and Aly Makky, E. 1993. �Shore Protection of the Nile Delta After the Construction of High Aswan Dam�, in Egyptian Committee on Large Dams (ed), High Aswan Dam Vital Achievement Fully Controlled, Cairo: ENCOLD. In, McCully 1996. Abramovitz, N.N. 1996. �Imperiled Waters, Impoverished Future: The Decline of Freshwater Ecosystems�, Worldwatch Paper 128, Worldwatch Institute, Washington D.C., 80 pp. Acreman, M.C. and Lahmann, E. 1995. �Overview: hydrological management and protected areas�, in Parks 5 Vol 2: 1-5. Acreman, M.C., Farquharson, F.A.K., McCartney, M.P. Sullivan, C., Campbell, K., Hodgson, N., Morton, J., Smith, D., Birley, M., Knott, D., Lazenby, J., Wingfield, D., Barbier, E.B. 2000. Managed Flood Relases from Reservoirs: Issues and Guidance. Report to DFID and the World Commission on Dams, Centre for Ecology and Hydrology, Wallingford, UK. Adamoli, J. 1988. �Sistema de Alarma para Prevencion de Peligros de Inundaciones en el Pantanal Matogrossense (Brazil)�, in Proc. Of Workshop �Mitigacion de Peligros Debidos a Eventos Naturales Extremos en America�. Natural Science Foundation, University de Puerto Rico: pp 177 � 194. ADB, 1999. Special Valuation Study on the Social and Environmental Impacts of (Four) Selected Hydropower Projects, Manilla: Asian Development Bank. Agostinho, A.A. & Zalewski, M. 1995. �The Dependence of Fish Community Structure and Dynamics on Floodplain and Riparian Ecotone Zone in Parana River, Brazil�, in Hydrobiologia. Vol 303: 141-148. Agostinho, A.A., Julio, Jr., H.F. and Petrere, M. 1994. �Itaipu Reservoir (Brazil): Impacts of the Impoundment on the Fish Fauna and Fisheries�, in I.G. Cowx. (ed) Rehabilitation of Freshwater Fisheries, Oxford: Fishing News Books.. Aleem, A.A. 1972. �Effect of River Outflow Management on Marine Life�, in Marine Biology. Vol 15: 200-208. Alexander A.M., List, J.A., Margolis, M. and d�Arge, R.C. 1998. �A Method for Valuing Global Ecosystem Services�, in Ecological Economics. Vol 27: 161-170. Allan, D.G. 1999. �Mega-Developments and Birds: the Waterbirds Impacted by the Lesotho Highlands Water Scheme as an Example�, in N.J. Adams & R.H. Slotow (eds) Proc. 22nd International Ornithological Congress, Durban., Pp. 1556-1578, Johannesburg: BirdLife South Africa. American Rivers, Friends of the Earth and Trout Unlimited. 1999. Dam Removal Success Stories: Restoring Rivers through Selective Removal of Dams that Don�t Make Sense. Final Report. Amsler, M.L. & Drago, E.C. 1999. �A Review of the Suspended Sediment Budget at the Confluence of the Parana and Paraguay Rivers,� paper presented at Hydrological and Geochemical Processes in Large Scale River Basins Conference, Manaus, Brazil, 15-19 November. Anonymous. 1997. �Wetlands and Integrated River Basin Management: Experiences in Asia and the Pacific�, in UNEP/Wetlands International - Asia Pacific, Kuala Lumpur. Baker, C.J. and Maltby, E. 1995. �Nitrate Removal by River Marginal Wetlands: Factors Affecting the Provision of a Suitable Denitrification Environment�, in: Hughes, J.M.R. and Heathwaite, A.L.



(eds) Hydrology and Hydrochemistry of British Wetlands, Chichester (UK): John Wiley and Sons. pp. 291-313. Balland, P. 1991. Le Littoral Mediterraneen Francais. Evolution Physique. - Qualite Generale. Agence de l�Eau Rhone-Mediterranee-Corse, p. 23. Barbier, E.B. 1996. �Valuing Floodplain Benefits: Economic and Hydrological Studies Conducted at the Hadejia-Nguru Wetlands Conservation Project, Nigeria�, in IUCN Wetlands Programme Newsletter. Vol 14: 10-12. Bardach, J.E. and Dussart, B. 1973. �Effects of Man-Made Lakes on Ecosystems�, in Geophys. Monogr. Vol 17: 811-817. Bell. M. C., DeLacy A. C. and Paulik G. J. 1967. A Compendium on the Success of Passage of Small Ffish through Turbines, Report on Contract DA-35-026-CIVENG-66-16 to North Pacific Division, U.S. Army Corps of Engineers. Bernacsek, G.M. 1997. Large Dam Fisheries of the Lower Mekong Countries: Review and Assessment, MKG/R.97023 Vol.I Prepared for The Mekong River Commission Project on Management of Fisheries Reservoirs in the Mekong Basin. Bernhardt, H. 1994. �Control of Water Quality in Lakes and Reservoirs� in Hino, M (ed) Water Quality and its Control, Rotterdam: A.A. Balkema. pp. 77-109. Binnie, C. 1999. Personal Communication to Mr. J. Skinner, World Commission on Dams. Blanch ̧S. 1999. Environmental Flows: Present and Future, Sidney, Australia: Inland Rivers Network. pp. 8. Bogan, A.E. 1998. �Freshwater Molluscan Conservation in North America: Problems and Practices�, in Killeen, I.J., Seddon, M.B. and Holmes, A. (eds) �Molluscan Conservation: a Strategy for the 21st Century�. Journal of Conchology Special Publication no. 2: pp. 223 � 230. Born, S.M., Genskow, K.D., Filbert, T.L., Hernandez-Mora, N., Keefer, M. and White, K. 1998. �Socioeconomic and Institutional Dimensions of Dam Removals: the Wisconsin Experience�, in Environmental Management.Vol 22: 359-370. Bourke, G. 1988. �Subduing the Sea�s Onslaught�. South, July. In McCully 1996. Brown, D.S. 1994. Freshwater Snails of Africa and their Medical Importance. 2nd edition, London: Taylor & Francis. Bunch and Reeves (eds). 1993. �Proceedings of a Workshop on the Potential Cumulative Impacts of Development in the Region of Hudson and James Bays, 17-19 June 1992�, in Canadian Technical Report of Fisheries and Aquatic Sciences Vol. 1874. Cada, G.F. and Hunsaker, C.T. 1990. �Cumulative Impacts of Hydropower Development: Reaching a Watershed in Impact Assessment�, in The Environmental Professional, Vol. 12: 2-8. Casinader, Ranji. 1999. �Sugarloaf Dam, Australia: Note on Environmental and Social Mitigation Mechanisms in Planning, Design and Construction�, Personal Communication to Mr. J. Skinner, World Commission on Dams.



Chapman, M.A. 1996. �Human Impacts on the Waikato River System, New Zealand� in Geojournal. Vol. 40 (1-2): 85-99. Chelan County Public Utility District. 1999. Performance-Based Resource Standards in the Hydroelectric Relicensing Process. Chelan County, Washington, USA. 19 pp. (Draft). Chesney, H.G.R. & Oliver, P.G. 1998. �Conservation Issues for the Margaritiferidae in the British Isles and Western Europe�, in Killeen I.J., Seddon M.B. & Holmes A. (eds) �Molluscan Conservation: a Strategy for the 21st Century�, in J. Conchology Special Publication no. 2: pp. 231-242. Church, M. 1995. �Geomorphic Response to River Flow Regulation: Case Studies and Time-Scales�, in Regul. Rivers: Res. Manage. Vol. 11: 3-22. Collier, M. Webb, R.H. and Schmidt, J.C. 1996. �Dams and Rivers: a Primer on the Downstream Effects of Dams�, in US Geological Survey Circular Vol. 1126: p 94. Costanza, R., d�Arge, R., de Groot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K., Naeem, S., O�Neill, R.V., Paruelo, J., Raskinet R.G., Sutton, P. and van den Belt, M. 1997. �The Value of the World�s Ecosystem Services and Natural Capital�, in Nature Vol. 387:253-260. Cowan, G.A. & van Riet, W. 1998. A Directory of South African Wetlands, Pretoria: Dept. of Environmental Affairs & Tourism. Crawford, C.S., Molles, M.C., Valett, H.M. 1994. �Flooding and Conservation in the Albuquerque Bosque. Tech. Rep. New Mex. Water Resour. Res. Inst. 290: The Future of Albuquerque and Middle Rio Grande Basin�, in Proceedings of the 39th Annual New Mexico Water Conference. pp 101-105. Davies, B.R. 1979. �Stream Regulation in Africa: a Review�, in: Ward, E.J. and Stanford, J.A. (eds) The Ecology of Regulated Streams. New York, USA: Plenum Press. pp 113-142. D�Croz, L. and Kwiecinski, B. 1980. �Contribucion de los Manglares a las Pesquerias de la Bahia de Panama�, in Revista Biologia Tropical, Vol. 28: 13-29. De Groot, R.S., 1992. Functions of Nature. Evaluation of Nature in Environmental Planning, Management and Decision Making. Wolters Noordhoff, Deventer (The Netherlands). 315 pp. De Silva, S. S. 1988a. �Reservoirs of Sri Lanka and their Fisheries�. FAO Fisheries Technical Paper, No. 298, Rome. De Silva, S. S. 1988b. �Reservoir Fishery Management and Development in Asia�, in Proceedings of a workshop held in Kathmandu, Nepal, 23-28 November 1987, Ottawa: International Development Research Centre. Dinar, A., Seidl, P., Olem, H., Jorden, V., Duda, A. and Johnson, R. 1995. �Restoring and Protecting the World�s Lakes and Reservoirs�. World Bank Technical Paper No. 289. Washington, D.C.: The World Bank. 85 pp. + maps. Dixon, J.A., Talbot, L.M. and Le Moigne, G.J.M. 1989. �Dams and the Environment: Considerations in World Bank Projects�. World Bank Technical Paper No. 110, Washington, D.C. Doyle, M.W, Stanley, E.H. Luebke, M.A. and Harbor, J.M. 2000. �Dam Removal: Physical, Biological and Societal Considerations�. American Society of Civil Engineers Joint Conference on



Water Resources Engineering and Water Resources Planning and Management, Minneapolis, MN, July 30-August 2, 2000. Drinkwater, K.F. & Frank, K.T. 1994. �Effects of River Regulation and Diversion on Marine Fish and Invertebrates�. Aquatic Conservation : Freshwater and Marine Ecosystems. Vol. 4: 135-151. Dugan, P.J. 1990. Wetland Conservation: A Review of Current Issues and Required Action. Gland (Switzerland): IUCN, 95 pp. Dunne, T., Mertes, L.A.K., Meade, R.H., Richey, J.E. & Forsberg, B.R. 1999. �Sediment Transport and Floodplain Storage along the Amazon River Valley in Brazil�, paper presented at Hydrological and Geochemical Processes in Large Scale River Basins Conference, Manaus, Brazil, 15-19 November. Dynesius, M. and Nilsson, C. 1994. Fragmentation and Flow Regulation of River Systems in the Northern Third of the World, in Science. Vol. 266 (4 November 1994):753-762. Echeverria, J., Hanrahan, M. and Solorzano, R. 1995. �Valuation of Non-Priced Ameneties Provided by the Biological Reserves within the Monte Verde Cloud Forest Preserve�, in Ecological Economics. Vol. 13: 43-52. Eddy, S. & Underhill, J.C. 1974. Northern Fishes with Special Reference to the Upper Missippi Valley. Minneapolis: University of Minnesota Press. Ertan Hydroelectric Development Corporation (EHDC). 1994. �Ertan Hydroelectric Project: Environmental Assessment�. Submitted to World Bank. Chengdu, Sichuan Province, P. R. China. European Environment Agency 1999. �Lakes and Reservoirs in the EEA Area�. Topic Report No. 1/1999. EEA, Copenhagen, 108 pp. Faucheux, S. and O�Connor, M. 1998. Valuation for Sustainable Development. Cheltenham: Edward Elgar. Finlow Bates, K.S., Gentle J. and Lovett J.C. 2000. �Environmental Mitigation of the Lower Leihansi Project, Tanzania�, Process Conference International Hydropower Association, Berne 2000 pp 427 - 436 Aqua - Media International, UK. Flanders, D. Principal Engineer, International Business. Department of Natural Resources, State Water Projects, Australia. 1999. Personal Communication to Mr. J. Skinner, World Commission on Dams. Flatby, R. and Konow. T. 1999. �Environmental Impact Assessments, Licensing Procedures and Terms of Licence (sic) � A Review of the Norwegian Experience�, in Proceedings of Workshop on Benefits of and Concerns about Dams � Case Studies. Pp. 69 � 76. Antalya, Turkey: International Commission on Large Dams. Frazier, S. 1999. Ramsar Sites Overview. Wageningen: Wetlands International. Furness, H.D. 1978. �Ecological Studies on the Pongola River Floodplain�, in Working Document IV, Workshop on Man and the Pongolo Floodplain. Pietermaritzburg, South Africa: C.S.I.R. No. 14/106/7C. Galay, V.J. 1983. �Causes of River Bed Degradation�, in Wat. Resour. Res. Vol. 19: 1057-1090.



Gammelsrod, T. 1992. �Variation in Shrimp Abundance on the Sofala Bank, Mozambique, and its Relation to the Zambezi River Runoff�, in Estuarine, Coastal and Shelf Science. Vol. 35: 91-103. Godinho, H. P., Godinho A. L., Formagio P. S. and Torquato V. C.. 1991. �Fish Ladder Efficiency in a Southeastern Brazilian river, in Ciencia e Cultura Vol. 43(1): 63-67. Goldsmith, E. and Hildyard N. 1984. The Social and Environmental Effects of Large Dams. San Francisco, CA, USA:A Sierra Club Book. 404 pp. Gourene, G., Teugels G.G., Hugueny B. & Thys van den Audenaerde D.F.E. 1999. �Evaluation de la Diversite Ichthyologique d'un Bassin Ouest-Africain Apres la Construction d'un Barrage�, in Cybium Vol. 23: 147-160. Guy, P.R. 1981. �River Bank Erosion in the mid-Zambezi Valley, Downstream of Lake Kariba�, in Biological Conservation. Vol. 20: 199-212. Hadley, R.F., Karlinger, M.R., Burns, A.W. and Eschner, T.R. 1987. �Water Development and Associated Hydrologic Changes in the Platte River, Nebraska, USA�, in Regul. Riv: Res. and Manage. Vol. 1: 331-341. Hauer, F.R., Stanford, J.A. and Ward, J.V. 1989. �Serial Discontinuity in a Rocky Mountain River II. Distribution and Abundance of Trichoptera�, in Regul. Riv: Res. and Manage. Vol. 3: 177-182. Hamerlynck, O., Ould Baba M. L., and Duvail S.. (In press). The Diawling National Park, Mauritania: Conflict and development around a newly established protected area. Submitted to Vida Sylvestre Neotropical. Hart, B.T. Bailey, P., Edwards, R., Hortle, K., James, K., McMahon, A., Meredith, C. and Swaling, K. 1991. �A Review of the Salt Sensitivity of the Australian Freshwater Biota�, in Hydrobiologia. Vol. 210: 105-144. Helfman, G.S., Collette B.B. & Facey D.E. 1997. The Diversity of Fishes. Malden, Massachusetts, USA: Blackwell Science, Inc. Holden, P.B. & Stalnaker C.B. 1975. �Distribution and Abundance of Mainstream Fishes of the Middle and Upper Colorado River Basins�, in Transactions of the American Fisheries Society. Vol. 104: 217-231. Hollis, G.E., Penson, J.R., Thompson, J.R. and Sule, A.R. 1993. �Hydrology of the River Basin� in Hollis, G.E., Adams, W.M. and Kano, M.A. (eds) The Hadejia-Nguru Wetlands: Environment, Economy and Sustainable Development of a Sahelian Floodplain Wetland. Gland, Switzerland and Cambridge, UK: IUCN. xviii + 244pp. Howells, R.G., Neck, R.W. & Murray, H.D. 1996. Freshwater Mussels of Texas. Texas Parks and Wildlife Press. Hydro Quebec 1991. Environmental Code. Montreal (Canada): Hydro Quebec. HQ and GDG, 1999. Biodiversity and Hydroelectricity: World Outlook and Québec Context. A joint report by Hydro-Québec and GDG Conseil, Inc., Montreal (Canada). Hynes, H.B.N. 1970. �The Ecology of Flowing Waters in Relation to Management�, in Journal of The Water Pollution Control Federation. Vol. 42 (3): 418-424.



ICOLD. 1981. Dam Projects and Environmental Success. ICOLD Bulletin 37. Paris, 30 pp. ICOLD. 1988. Dams and Environment. Case Histories. ICOLD Bulletin 65. Paris 112 pp. ICOLD. 1994. Dams and Environment: Water Quality and Climate. Paris. 85 pp. ICOLD. 1997. Position Paper on Dams and Environment. http://genepi.louis-jean.com/cigb/chartean.html ICOLD. 1999. Database. http://genepi.louis-jean.com/cigb/anglais/html International Energy Agency. 2000. Hydropower and Environment: Present Context and Guidelines for Future Action. Annex 3. Implementing Agreement for Hydropower Technologies and Programmes. IRN. 1994. Manibeli Declaration. http://www.irn.org/programs/finance/manibeli.html IRN, 2000. Questions and Answers on the International Movement Against Large Dams. http://www.irn.org/basics/qanda.html Irving, J.S. and Bain, M.B. 1989. �Assessing the Cumulative Impact on Fish and Wwildlife in the Salmon River Basin, Idaho�, in The Scientific Challenges of NEPA: Future Direction Based on 20 Years of Experience. Knoxvill, TN, October 24-27: 357-372. IUCN. 1994. Wasa Logone Flood Restoration Study. Delft Hydraulics/IUCN, Gland. Jackson, D. C. 1999. �The Influence of Dams on River Fisheries, Consultant�s Report to the Unitied Nations Food and Agriculture Organization. 36 pp. Jackson, P.B.N. and Davies, B.R. 1976. �Cabora River in its First Yyear: Some Ecological Aspects and Comparisons�. Rhodesian Science News.Vol. 10 (5): 128-133. Jobin, W., Blue Nile Associates. 1999. Personal Communication to Mr. J. Skinner, World Commission on Dams. Johnson, W.C. 1992. �Dams and Riparian Forests: Case Study from the Upper Missouri River�, in Rivers. Vol. 3 (4): 229-242. Junk, W.J., Bayley, P.B. and Sparks, R.E. 1989. �The Flood Pulse Concept in River-Floodplain Systems� in Doge, D.P. (ed) Proceedings of the International Large River Symposium (LARS). Can. J. Fish. Aquat. Sci. Vol. 106: 110-127. Khafagy, A.A. and Fanos A.M.. 1993. �Impacts of Irrigation Control Works on the Nile Delta Coast�, in Egyptian Committee on Large Dams (ed) High Aswan Dam Vital Achievement Fully Controlled. Cairo: ENCOLD. In, McCully 1996. King, J., Tharme R., and Brown C. 1999. �World Commission on Dams Thematic Report: Definition and Implementation of Instream Flows�. Final Draft. Prepared by Southern Waters Ecological Research and Consulting under contact to World Commission on Dams, Cape Town, South Africa. 64 pp. King, JM. and Louw, M.D. 1998. �Instream Flow Assessments for Regulated Rivers in South Africa Using the Building Block Methodology�, in Aquatic Ecosystem Health and Management. Vol. 1: 109-124.



Kirchner, J.W., Finkel, R.C. Riebe, C.S. Granger, D.E. Clayton J.L. & Megahan, W.F. 1998. �Episodic Erosion of the Idaho Batholith Inferred from Measurements over 10-Year and 10,000 Year Timescales�, in EOS Transactions, American Geophysical Union. Vol 79: F338. Kristensen, P. and Hansen, H.O. (eds), 1994. �European Rivers and Lakes. Assessment of Their Environmental State�. EEA Environmental Monograph 1. Kvernevik, T.I. and Ghazali, N. 1998. Cumulative Effects of Dams � a Resource File. 10 pp. Läe, R. 1995. �Climatic and Anthropogenic Effects on Fish Diversity and Fish Yields in the Central Delta of the Niger River�, in Aquatic Living Resources 8: 43-58. Larinier, M. 1998. �Upstream and Downstream Fish Passage Experience in France� in Jungwirth, M., Schmutz, S. and Weiss, S. (eds) Fish Migration and Fish Bypasses. Oxford (UK): Fishing New Books, Blackwell Science Ltd.. Larsen, T. B. 1999. �Norwegian Hydropower Resources: Use, Management and Planning�, proceedings of Workshop on Benefits of and Concerns about Dams � Case Studies. International Commission on Large Dams, Antalya, Turkey. 6 pp. Le Cornu, J. 1998b. �ICOLD�s Challenges�, report of ICOLD Secretary General to Hydrovision �98: the World Hydro Forum. Reno, Nevada, July 28. http://genepi.louis-jean.com/cigb/reno.html Ledec, G., Quintero, J.D. and Mejia, M.C. 1997. �Good Dams and Bad Dams: Environmental and Social Criteria for Choosing Hydroelectric Project Sites�, World Bank Sustainable Dissemination Note 1. Lenhart, C.F. 2000. �The Vegetation and Hydrology of Impoundments after Dam Removal in Southern Winconsin�, unpublished MSc thesis, University of Winconsin � cited in Doyle et al. 2000. Ligon, F.K., Dietrich, W.E. & Trush, W.J. 1995. �Downstream Ecological Effects of Dams,� in Bioscience. Vol 45: 183-192. Maheshwari, B.L., Walker, K.F. and McMahon, T.A. 1995. �Effects of Regulation on the Flow Regime of the River Murray, Australia�, in Regul. Riv: Res. and Manage. Vol 10: 15-38. Mahmood, K. 1987. �Reservoir Sedimentation: Impact, Extent and Mitigation�, World Bank technical Paper 71. Malanson, G.P. 1993. Riparian Landscapes, Cambridge: Cambridge University Press. Masson, J. 1999. �Advantages and Impacts of Dams: The example of the Durance and Verdon Scheme�, Proceedings of Workshop on Benefits of and Concerns about Dams � Case Studies. International Commission on Large Dams, Antalya, Turkey, pp. 295 � 305. McAllister, D.E, Hamilton A.L., and Harvey B.. 1997. �Global Freshwater Biodiversity: Striving for the Integrity of Freshwater Ecosystems�, in Special Edition of Sea Wind-Bulletin of Ocean Voice International. Vol 11(3): 1-140. McCully, P. 1994. �Damming the Rivers: the World Bank�s Lending for Large Dams�, International Rivers Network. http://www.irn.org/pubs/wp/damming.html



McCully, P. 1996, Silenced Rivers: the Ecology and Politics of Large Dams, London: Zed Books. 350 pp. Meade, R.H. & Parker, R.S. 1985. �Sediment in Rivers of the United States,� in Water Supply Paper. Vol 2275: 49-60, United States Geological Survey, Washington D.C. Miller, J.R. Schulz, T.T. Hobbs, N.T., Wilson, K.R. Schrupp, D.L. and Baker, W.L. 1995. �Changes in the Landscape Structure of a Southeastern Wyoming Riparian Zone Following Shifts in Stream Dynamics�, in Biol. Conserv. Vol 72 (3): 371-379. Mitchell, D.S. 1978. Aquatic Weeds in Australian Inland Waters, Canberra: Australian Government Publishing Service. 189 pp + xxii. Miyanaga, Y. Tanaka, M. and Hino, M. 1994. �Thermal Stratification: their Prediction and Control� in Hino, M (ed) Water Quality and its Control, Rotterdam: A.A. Balkema. 3-43. Moore, D. 2000. �Dams and Wetlands Worldwide: The World Commission on Dams to Examine Methods for Wetlands Protection, Mitigation, and Restoration�, in National Wetlands Newsletter. January-February 2000. Nakamura, S. 1993. "A Review of Fish Passage Facilities in East Asia. Fish Passage Policy and Technology", in Proceedings of the symposium, Portland, Oregon, USA. National Park Services. 1996. �Elwha River Ecosystem Restoration Implementation� summary draft EIS. Olympic National Park, Washington. Nehlsen, W., Williams J.E., and Lichatowich J.A. 1991. �Pacific Salmon at the Crossroads: Stocks at Risk from California, Oregon, Idaho, and Washington�, in Fisheries. Vol 16: 4-21. Nelson, J.S. 1994. Fishes of the World. 3rd Ed., New York: John Wiley & Sons, Inc.. 620pp. Neves, R.J. and Angermeier, P.L. 1990. �Habitat Alteration and its Effects on Native Fishes in the Upper Tennessee River System, East-central U.S.A�, in Journal of Fish Biology. Vol 37: 45-52. Nilsson, C. and Jansson, R. 1995. �Floristic Differences Between Riparian Corridors of Regulated and Free-Flowing Boreal Rivers�, in Regul. Rivers: Res. Manage.Vol 11: 55-66 Nilsson, C., Jansson, R. and Zinko, U. 1997. �Long-term Responses of River-Margin Vegetation to Water-Level Regulation�, in Science. Vol 276: 798-800. Ochman, S. & Dodson J. 1982. �Composition and Structure of the Larval and Juvenile Fish Community of the Eastmain River and Estuary, James Bay�, in Le Naturaliste Canadien. Vol 109: 803-813. Office of Technology Assessment (OTA). 1995. �Fish Passage Technologies : Protection at Hydroelectric Facilities�, report OTA-ENV-641, Washington DC. Petts, G.E. 1984. Impounded Rivers, Chichester, UK: John Wiley &Sons Ltd Publishers. Petts, G.E. 1996. �Water Allocation to Protect River Ecosystems�, in Regulated Rivers: Research & Management. Vol 12: 353 � 365. Prescott, J. P. Vice President Generation, Idaho Power Company. 1999. Personal Communication to Mr. J. Skinner, World Commission on Dams.



Reizer, C. 1988. �Les Peche Continentales du Fleuve Sénégal. Environment et Impacts des Aménagement�. Musée Royal de l�Afrique Centrale, Tervuren, Belgique, in Annales Série 1N-8 Sciences Zoologiques Vol. 254: 1-10. Ribeiro, M.C.L.B., Petrere, M., Jr & Juras, A.A.. 1995. �Ecological Integrity and Fisheries Ecology of the Araguaia-Tocantins River Basin, Brazil, in Regulated Rivers: Research and Management. Vol 11: 325-350. Rivero, C. ENDESA, Spain. 1999. Personal Communication to Mr. J. Skinner, World Commission on Dams. Rood, S.B., Mahoney, J.M. Reid, D.E. and Zilm, L. 1995. �Instream Flows and the Decline of Riparian Cottonwoods along the St. Mary River, Alberta�, in Can. J. Bot. Vol 73 (8): 1250-1260. Rosales, J., Bradley, C., Petts, G., & Gilvear, D. 1999. �A Conceptual Model of Hydrogeomorphological-Vegetation Interactions within Confluence Zones of Large Tropical Rivers, Orinoco, Venezuela,� paper presented at Hydrological and Geochemical Processes in Large Scale River Basins Conference, Manaus, Brazil, 15-19 November. Ryman, N., Utter, F. & Laikre, L.. 1995. �Protection of Intraspecific Biodiversity of Exploited Fishes�, in Reviews in Fish Biology and Fisheries. Vol 5: 417-446. Sanz Montero, M.E., Avendaño Salas, C. & Cobo Rayan, R. 1999. �Influencia de los Embalses en el Transporte de Sedimentos Hasta de Delta del Rio Ebro, España,� paper presented at Hydrological and Geochemical Processes in Large Scale River Basins Conference, Manaus, Brazil, 15-19 November. Seattle City Light. 1985. �Application for Amendment to License for Boundary Dam Hydropower Project�, Submitted to U. S. Federal Energy Regulatory Commission. Shiel, R.J. 1978. �Zooplankton Communities of the Murray-Darling System�, in Proceedings of the Royal Society of Victoria. Vol 90: 193-202. Shiklomanov, I.A. (ed) 1997. Assessment of Water Resources and Water Availability in the World. Comprehensive Assessment of the Freshwater Resources of the World, Stockholm: Stockholm Environment Institute. Shiklomanov, I.A. 1999. �World Water Resources: Modern Assessment and Outlook for the 21st Century.� Federal Service of Russia for Hydrometeorology and Environment Monitoring: State Hydrological Institute. (Summary of the monograph World Water Resources at the Beginning of the 21st Century, prepared in the framework of IHP-UNESCO, 51 pp.) Shuman , J.R. 1995. �Environmental Considerations for Assessing Dam Removal Alternatives for River Restoration�, in Regul. Rivers: Res. and Manage. Vol 11: 249-261. Simons R.K. and Simons, D.B. 1991. �Sediment Problems Associated with Dam Removal � Muskegon River, Michigan�, in Shane, R.M. (ed) Hydraulic Engineering, Proceedings of the National Conference on Hydraulic Engineering American Society of Civil Engineers, New York. 680-685. Smith, L.C. & Sidorchuk, A. 1999. �Hydrologic Variability and Floodplain Sedimentation of Major West Siberian Rivers,� paper presented at Hydrological and Geochemical Processes in Large Scale River Basins Conference, Manaus, Brazil, 15-19 November.



Stanley, D.G. & Warne, A.G.. 1993. �Nile Delta: Recent Geological Evolution and Human Impact�, in Science (Washington, D.C.). Vol 260: 628-634. Stein, B.A. & Flack, S.R. (eds). 1996. America�s Least Wanted: Alien Species Invasions of U.S. Ecosystems, Arlington, Virginia: The Nature Conservancy. Sullivan, C.A. 1999. �Review of the Economic Dimensions of Downstream Environmental Impacts of Controlled Flood Releases�, Report to DfID and the WCD. Institute of Hydrology, Wallingford. Sullivan, C.A. 1999. �Valuation of Non-timber Forest Products: A Case Study of Three Amerindian Villages in Guyana�, Ph.D. Thesis, Dept. of Economics, Keele University, UK. Taylor, R., Mcginnis, S. and Elliott, C.. 1999. �Certification without Green Buyers?: A Discussion Paper�, presented at the World Bank Alliance for Forest Conservation and Sustainable Use: Forest Certification/ Verification Systems Workshop. Washington, DC, USA. 19 pp. Truffer, B. 1999a. �Green Electricity Products and the Sustainable use of Water Resources�, Forum of the UNESCO International School of Science for Peace on �Water security in the Third Millenium. Mediterranean Countries as a case�. Landau Network � Centro Volta, Como, Italy. 10 pp. Truffer, B. 1999b. Personal Communication to Mr. J. Skinner, World Commission on Dams. USCOLD, 1999. Dam, Hydropower and Reservoir Statistics. http://www2.privatei.com/ ~uscold/uscold_s.html USGS, 1996. �Dams and Rivers. Primer on the Downstream Effects of Dams�, U.S. Geological Survey Circular 1126. USGS, Tucson (USA). Valentin, S., Wasson, J.G. and Philippe, M. 1995. �Effects of Hydropower Peaking on Epilithon and Invertebrate Community Trophic Structure�, in Regulated Rivers: Research & Management. Vol 10: 105-119. Walker, K.F. 1979. �Regulated Streams in Australia: the Murray-Darling River System� in Ward, J.V. and Stanford, J.A. (eds) The Ecology of Regulated Streams, New York: Pelnum Press. 398 pp. Walling, D. & Fang, D. 1999. �Longer-term Variability of Sediment Transport to the Oceans,� paper presented at Hydrological and Geochemical Processes in Large Scale River Basins Conference, Manaus, Brazil, 15-19 November. Ward, J.V. and Stanford, J.A. 1995. �The Serial Discontinuity Concept: Extending the Model to Floodplain Rivers. Regul. Riv: Res. and Manage. Vol 10: 159-168. Ward, J.V and Stanford, J.A. 1983. �The Serial Discontinuity Concept of Lotic Ecosystems in Fontaine�, in T.D. and Bartell, S.M (eds) Dynamics of Lotic Ecosystems, Collingwood: Ann Arbor Science, MI 29-42. Watson, R. T., Zinyowera, M. C., and Moss, R. H. 1996. (eds). �Climate Change 1995. Impacts, Adaptations and Mitigation of Climate Change: Scientific-Technical Analyses�, contribution of Working Group II to the Second Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge, UK: Cambridge University Press. 878 pp. WCMC 1998. �Freshwater Biodiversity: a Preliminary Global Assessment�, WCMC Biodiversity Series no. 8, WCMC/UNEP, Cambridge (UK).



Webb, R.H., Schmidt, J.C. Marzolf, G.R and Valdez, R.A. 1999. �The Controlled Flood in Grand Canyon�, Geophysical Monograph Series, 110, Geological Survey 368 pp.

Williams, G.P. and Wolman, M.G. 1984. �Downstream Effects of Dams on Alluvial Rivers�, US Geol. Surv. Prof. Pap., 1286, 83 pp. World Bank, 1991a. �Environmental Assessment Sourcebook, Volume I, Policies, Procedures, and Cross-Sectoral Issues�, World Bank Technical Paper No. 139, Washington. World Bank, 1999. The World Bank Operational Manual. http://wbln0018.worldbank.org Zahar, Y. & Albergel, J. 1999. �Hydrodynamique Fluviale de l�Oued Medjerdah a l�Aval du Barrage Sidi Salem. Evolution Recente,� paper presented at Hydrological and Geochemical Processes in Large Scale River Basins Conference, Manaus, Brazil, 15-19 November. Zakova, Z., Berankova, D., Kockova, E., Kriz, P., Mlejnkova, H. and Lind, O.T. 1993. �Investigation of the Development of Biological and Chemical Conditions in the Vir Reservoir 30 Years After Impoundment�, in Water Sci. Technol. Vol 28 (6): 65-74.



Annex 1: Potential Environmental Impacts of Dams, Reservoirs and Hydroelectric Projects Based on ICOLD�s Bulletins 35, 37, 50, 65, 66, 86, 90, 96 and 100, observations from USCOLD and the World Bank�s �Environmental Assessment Sourcebook�, Vol. III, pp 69 � 73, [8] ten categories of environmental impacts have been established: A. Impacts on the Natural Environment (Flora, Fauna and Aquatic Fauna) B. Social / Economic / Cultural Aspects (resettlement) C. Land D. Dam Construction Activities E. Sedimentation of Reservoirs F. Downstream Hydrology G. Water Quality H. Tidal Barrages I. Climate J. Human Health Note: ICOLD Bulletins are identified by a �B� followed by the number of the bulletin, USCOLD by �U�; and the World Bank�s Sourcebook by �S� and the numerical listing. A. Impacts on the Natural Environment 1. Negative environmental effects due to construction activities (S1); 2. Loss of wildlands, wetlands and wildlife habitat, extinction of plant and animal life (S15, B50);

threats to endangered species (U); 3. Effects of stopping the flow of nutrients downstream (U, B50); 4. Reduced biological activity downstream (B50) (In arid areas often an increase in quantity of flora

and fauna (B50); 5. Reduction in downstream flooding may result in less natural submergence for flood-recession

agriculture, reduction in groundwater recharge and reduction in removal of parasites by natural flooding (B50);

6. Impacts on quantity of water needed for maintaining downstream ecology (B35); 7. Anaerobic decomposition of vegetation and production of greenhouse gasses (high cost of

cleaning up); 8. Environmental degradation from increased pressure on land such as irrigated agriculture,

industries and municipalities (S23); 9. Dams form obstacles to passage of trees, floating debris, ice and ships (B35); 10. Waterloss due to evaporation; 11. Induced seismicity; 12. Changed morphological character of rivers (flow volume, surface area and water levels) (B50); 13. Rivers may dry up (B35). A* Flora 1. In severely cold climates, impact is limited to direct inundation and nearby changes in

groundwater levels (B50); 2. Aquatic weeds (floating and submerged) may proliferate, especially in tropical areas: Water

hyacinth and water lettuce (B35); 3. Prolific vegetation impedes navigation and fishing, and affects hydraulic structures (B50); 4. Tourism may adversely affect flora and fauna and also create social problems (B50).



A** Fauna 1. Accommodation of amphibians, riparian fauna and birds to a new environment (B35); 2. Migration of animals to new areas, where new equilibrium may favour some species over others

(B50). A** Aquatic Fauna 1. Blocking fish migration (S14, B35); 2. Disruption of riverine fisheries due to changes in patterns, duration, velocity and volume of flow

(S14, U, B35); 3. Introduction of new species of fish in the reservoirs (B35); 4. Inappropriate reservoir operation with large variations in water levels could threaten fish by

drying up shallow-breeding and flood-producing areas (B50); 5. Destruction of spawning beds in shallow areas at the margins of reservoirs due to enhanced

turbidity as a result of land erosion caused by wave action. B. Social / Economic / Cultural Aspects (resettlement) 1. Dislocation of people living in the inundation zone (S2); 2. Disruption of livelihood of private lives and tribal / indigenous groups (S20); 3. Destruction of lifestyles and customs (S20); 4. Social disruption and decrease in standard of living of resettled people (S18); 5. Impacts on host communities (receiving resettlers); 6. Uncontrolled migration of people into the project area made possible by access roads and

transmission lines corridors (S22); 7. Migration of displaced people from rural to urban areas (B50); 8. Loss of local land marks, historic, cultural or aesthetic features by inundation (S4); 9. Effects on recreation, instream fishing and whitewater rafting (U); 10. Loss of aesthetic values and scenic beauty, such as white water, waterfalls and active streams (U); 11. Influx of construction workers: interaction with local population may cause social and health

problems (B50); 12. Fluctuating water levels create bare slopes and, in arid conditions, also dust (B50); 13. Frazil ice in cold winter climates may clog downstream installations (B50); 14. Blockage of land transportation routes. C. Land 1. Loss of land through inundation: agricultural, forest, range, wildlands and wetlands (S3, B35); 2. Decrease in floodplain (recession) agriculture (S11), but expanded irrigated agriculture; 3. Salination of floodplains (S13); 4. Saltwater intrusion in estuaries (S13); 5. Loss of riparian soils (U) and erosion of borrow areas (B50); 6. Land slides may occur as a result of wetting and rapid drawdown (B35, 50); 7. Increased seismic activity and reservoir-induced earthquakes (B35); 8. Permafrost warming could cause considerable ground deformations (B50). D. Dam Construction Activities 1. Air and water pollution; 2. Soil erosion; 3. Creation of borrow and spoil areas; 4. Access roads open up new areas;



5. Workforce: boomtown effects, squatter settlements, strain on local resources and services, open up new markets for local goods.

E. Sedimentation of Reservoirs 1. Sedimentation of reservoirs causes loss of storage capacity (S8) (often not in severe winter

climates, except in the Himalayas) (B50); 2. Formation of sediment deposits at reservoir entrance creates backwater effect, flooding and water-

logging upstream (S9); 3. Capture of nutrients causes deficiencies downstream; 4. Scouring of river bed below dam due to lower sediment content of released water (S10) Also less

soil replacement; 5. Poor land use practices in catchment areas (such as deforestation and incautious agricultural

development (B35), and inflow of untreated industrial effluents and municipal wastes (B50); 6. Release of captured sediments (e.g. heavy metals). F. Downstream Hydrology 1. Change of riverflow patterns; 2. Oxygen deficiency, changes in water temperature and pH; 3. Salination and saltwater encroachment; 4. Changes in tidal prism in estuaries resulting from increased siltation; 5. Changes in water quality. G. Water Quality 1. Changes to water quality and limnology (S14) due to inflow of saline water (B35, 37) (retention

time is important); 2. Effects of changes in groundwater levels, higher around the reservoir and lower downstream (U,

B35). These may also affect ground water quality; 3. Proliferation of aquatic weeds in reservoir and downstream, causing clogging and impairing

navigation, recreation, fisheries and irrigation (S6); 4. Deterioration of water quality in the reservoir due to rotting of submerged vegetation and

hydrogen sulphide (B50) (Normally not in areas with a severe winter climate (B50) (S7); 5. Water quality deterioration is due to lack of dissolved oxygen near the bottom of reservoirs (B50).

This is toxic to fish and can lead to death of aquatic life (B35). It is also corrosive to turbines; 6. Snagging of fishing nets due to submerged vegetation in reservoir; 7. Deoxygenation (especially at lower levels in the reservoir) due to submergence of forests and

other vegetation with a high content of degradable matter (B50); 8. Fish may die downstream from nitrogen and oxygen supersaturation (B35); 9. Thermal stratification in deep reservoirs (due to heating and cooling of the surface layer (B50)

may result in low temperature water released through low level outlet works. This is detrimental to fish, and affects home and industrial water supplies and cooling ponds (B35);

10. Eutrophication results from sediment inflow enriched with nutrients (B35); 11. Pollution of reservoirs by humans and animals (B35) (industrial affluents, mercury release from

the soil and raising or contaminating watertables); 12. Macrophytes cause high water losses due to evaporation and depletion of oxygen and creates

difficulties for fishing (B50); 13. Agriculture on marginal lands near the water level (e.g. islands), may introduce pesticide,

rendering fish inedible (B50); 14. Evaporation in arid areas increased salinity, chlorides, carbonates and sulphates (B50);



15. Ice may impede proper aeration. Oxygen deficiency seldom occurs in severely cold climates (B50).

H. Tidal Barrages (B35) 1. Reduce inflow of tidal waters and scouring due to tides; 2. Formation of bars affects navigation; 3. Salt and fresh water regimes may change from stratified to mixed, or vice versa, which may affect

siltation and aquatic life; 4. Changes in natural wave patterns may affect flora and fauna in the tidal range; 5. Upland drainage and fresh water runoff may require pumping schemes; sewage treatment may

also be required; 6. Locks are needed for shipping, fish and ice passage. I. Climate 1. Large reservoirs produce a microclimate due to storage of heat and cold and may cause changes in

local rainfall patterns (B35); 2. Formation of early morning and winter fog in temperate climates (B35, 50); 3. Shallow reservoirs may show thicker mist on cool days (B35, 50); 4. Local increase in humidity and fog may create favourable habitats for insect disease vectors

(mosquitoes, tsetse) (S21); 5. Reservoirs in cold climates have very little effect because of the presence of many natural lakes

(B50). J. Human Health 1. Health is affected by dissemination of water-related diseases and from environmental changes, or

population movement or settlement (S16, B50); 2. In hot and damp regions endemic parasites find an increase in �host� habitat, such as those that

cause malaria, schistosomiasis, filariasis, bilharzia (in still water), and onchoceriasis (in flowing water) (B35, 50) also, typhoid fever, viral hepatitis and other parasitic and infectious endemic diseases (B50);

3. The health of workers and people in adjacent neighbourhoods may endanger the social and economic environment (B50).

Source : Jan Veltrop, ICOLD



Annex 2: Reservoir Fisheries The high production of fish (kg/ha) observed in reservoirs often exceeds that of natural bodies of water (Dixon et al., 1989; De Silva, 1988; Goldsmith and Hildyard, 1984) (Box A2.1). The vast areas and volumes of water made available during the creation of reservoirs provide additional habitats for aquatic wildlife. The net increase in the number of hectares of aquatic habitats generates high fish production, which generally encourages recreational and commercial fishing (World Bank, 1991a; Dixon et al., 1989; De Silva, 1988; Goldsmith and Hildyard, 1984). In Canada, this phenomenon has given rise to the development of numerous recreational fishing outfitters on the periphery of reservoirs. Aquatic biomass produced in reservoirs in north temperate zones is higher than the terrestrial wildlife biomass that is available for harvesting. In Lakes Kariba and Ayame, although several riverine species have disappeared, the overall number of fish species has increased (Gourène 1999; Kolding and Karenge in press). Tilapias (Cichlidae) are usually the most successful in these lakes. The siting of the reservoir can have an important influence on whether the impact is positive or negative. The Itaipu Reservoir, Brazil, is sited below a floodplain which enhances migratory species that inhabit the floodplain and when mature migrate down into the reservoir (Agostinho et al.,1994; Agostinho and Zalewski 1995). However the floodplain will disappear when a new dam being built permanently inundates the floodplain. Well-managed reservoir fisheries can be very productive. For example, an innovative programme introduced at the Saguling (1986) and Cirata (1991) dams in Indonesia makes use of the recent creation of the reservoir for fishery development. Fish-farming technologies have been adapted for small producers with an established tradition of aquaculture in ponds or rice fields. In 1992, follow-up activities demonstrated that the floating cage farming system of both reservoirs made possible the hiring of 7 500 families and produced 10 000 tons of fish. This harvest far exceeds the ten tons produced annually by the river of origin. Fisheries revenues from both reservoirs have exceeded $10 million per year, higher than the value of rice harvests from the farm lands submerged. Furthermore, 21 000 additional jobs have been created in enterprises for fish food production, cage maintenance, marketing, etc. Fishery management in other reservoirs has generated a significant number of jobs and major revenues, among them the Akosombo in Ghana, the Kedung Ombo in Indonesia, the Kariba in Zambia and Zimbabwe, the Magla and Tarbela in Pakistan, and the Nam Ngum in Laos. Box A2.1: Fisheries yields of selected reservoirs

Reservoir/Country/Region Yield Kg/ha/yr

Large African reservoirs (Kariba, Nassar, Nubia) 27-65 Lake Kainji 3.5-4.7 Medium-sized African reservoirs 80 China (with intensive stocking) 127-152 India 11.4-49.5 Malaysia 15 Kazakhstan 15 Sri Lanka (with intensive stocking) 40-650 Cuba 125 Dominican Republic 29-75 Brazil 2.1-11.5 Panama 4.8-63.2 North America 24 Europe 21-76



Reservoir fishery yields tend to be higher for the Caribbean than is generally recorded for Central and South American reservoirs. Records available for Brazil and Panama suggest that reservoirs can have quite variable yields, depending on flushing rates, elevation, and basin morphology. Higher yields throughout the region typically result from stocking of exotic species. In short, reservoirs resulting from construction of dams can in some situations result in productive fisheries. This is particularly true for locations where river fisheries contribute little to overall national fishery yields. The exact position of the dam with respect to fertile floodplain and delta fisheries will affect the degree to which the reservoir replaces or exceeds the natural river fishery. The available data seems to indicate that lowland dams have more impact on fisheries than highland dams. Beneficial reservoir fisheries also exist in drier regions where they are constructed for agricultural irrigation, and fisheries are secondary considerations. Benefits seem more pronounced for smaller, shallower reservoirs that have reasonably high concentrations of dissolved solids and that are located in the upper reaches of their respective river ecosystems. Stocking of exotic species (both in reservoirs and in tailwaters) can enhance yields, as long as the exotic fishes are environmentally sound and culturally acceptable to the surrounding human population. In this regard, caution is warranted in cultures where fishing and fish consumption are non-traditional activities. Building reservoirs in the context of such cultures may not achieve projected fishery benefits even though exploitable fish stocks may exist. See also contributing papers: Contributing Paper/ Synthesis Paper Writers Contract co-ordinator(s)/

funder Molluscan Biodiversity and the Impact of Large Dams

M.B. Seddon IUCN/UNEP

D. C. Jackson The Influence of Dams on River Fisheries G. Marmulla

FAO/WCD

M.P. McCartney C. Sullivan

Ecosystem Impacts of Large Dams

M.C. Acreman

IUCN/UNEP

D.E. McAllister J.Craig N. Davidson M. Seddon

Biodiversity Impacts of Large Dams

D.Murray

IUCN/UNEP

Large Dams and Freshwater Fish Biodiversity J.F.Craig FAO/WCD J. King R. Tharme

Definition and Implementation of Instream Flows

C. Brown

WCD

J. King Information Needs for Appraisal and Monitoring of Ecosystem Impacts C. Brown

WCD

Dams and Fish Migration M. Larinier FAO/WCD N. Davidson Biodiversity Impacts of Large Dams:

Waterbirds S. Delany IUCN/UNEP

Fundamental Legal and Ethical Principles In Adjudging the Merits of Development Projects

C. Di Leva IUCN/UNEP



Contributing Paper/ Synthesis Paper Writers Contract co-ordinator(s)/

funder M.C. Acreman E. Barbier M. Birley K. Campbell F. Farquharson N. Hodgson J. Lazenby M. McCartney J. Morton D. Smith

Managed Flood Releases from Reservoirs � A Review of Current Problems and Future Prospects

C. Sullivan

DFID

Report on the Conference on Hydrological and Geochemical Processes in Large Scale River Basins, 15-19 November, 1999, Manaus, Brazil

L. Sklar WCD

International Mechanisms for Avoiding, Mitigating and Compensating The Impacts of Large Dams on Aquatic and Related Ecosystems and Species

J. R. Bizer IUCN/UNEP

Capacity and Information Base Requirements for Effective Management of Fish Biodiversity, Fish Stocks and Fisheries Threatened or Affected by Dams During the Project Cycle

G. Bernacsek FAO/WCD



Annex 3: Comparison of Pre vs. Post Impoundment Conditions Comparison of a series of variables in pre and post impoundment conditions of two reservoirs in the Mekong River Basin. Adapted from Bernacsek 1997.

Reservoir and Year of Study Trin An Thac Ba

Variable Pre-Impoundment 1983 & 1985

Post-Impoundment

1990

Pre-Impoundment 1964 & 1971

Post-Impoundment 1976 & 1985

ToC 30-33.2 26.5-28 - 16-32 Transparency 100-243 15-30 - 38-169 pH 7.5-7.6 6.5-7.5 7.8-7.9 7-8.2 O2 (mg/l) 3.2-10.24 4.0-6.08 9.12-9.29 0.72-8.64 CO2 (mg/l) 1.75-2.64 3.52-6.16 - 0.88-23.76 Hardness 0.67-2.35 0.84-1.12 4.81-4.92 1.9-7.6 SiO2 (mg/l) 5-12 7-9 11 7-73 Ca (mg/l) 2.4-7.2 4 23.2-25 23.6-24.4 Mg (mg/l) 0.96-5.78 1.2-2.2 7.29 4.45-5.87 Fe (mg/l) 0.05-0.35 0.20-0.30 0.10-0.40 0.40-0.59 HCO3 (mg/l) 24.4-42.7 61 - 105.2-111.4 PO4 (mg/l) 0.15-0.30 0.05-0.10 Trace-0.13 Trace NH4-N (mg/l) 0 Trace-0.10 0 Trace Phytoplankton Species 97 38 306 101 Phytoplankton Quantity nos./l

46,600 100,000-500,000 295,000 40,000-200,000

Zooplankton Species 24 20 100 61 Zooplankton Quantity nos./m3

1,710 4,750-12,000 302 4,300-81,000

Zoobenthos Species 13 7 170 60 Zoobenthos Quantity nos/m2

1,706 100 - -

Zoobenthos Biomass g/m2

53.82 0.125-0.200 - 0.337-0.766

Fish Species 81 50 53 80

Fishermen 190 2,700 78 430 Fisheries Production (tonnes)

42 1,000 2.8-7.5 430

Aquaculture cages 0 600 0 20 Aquaculture production (tonnes)

0 300 0 9

Change in # of Species -38% +51% Change in fish catch Increased 24 times Increased 57 times



Annex 4: Sediment Discharges (Based on Sklar 2000) An illustrative example of temporal variability in sediment delivery is the observation that in a single three-day storm in the winter of 1982, the Eel River in Northern California delivered more sediment to the Pacific Ocean than had passed through the river in the previous seven years combined (Meade & Parker, 1985). At longer time scales the magnitude of temporal variability can also be quite large. A recent study of sediment flux from the mountains of central Idaho (Kirchner et al, 1998) showed that long term average erosion rates, as inferred from reliable cosmogenic nuclide dating techniques, are as much as an order of magnitude larger than the rate calculated from several decades of reservoir sedimentation monitoring. These examples show that sediment supply is often dominated by infrequent, catastrophic events that are difficult to anticipate but which should be incorporated into sediment supply studies for dam projects. Walling & Fang (1999) presented an analysis of a data set of suspended sediment flux measurements from 142 large rivers, with the goals of quantifying the range of variability in each river and revealing any global trends. Each river basin had more than 25 years of record and a contributing drainage area greater than 10 000 km2; all were in the northern hemisphere. They found that for nearly every basin, the range of annual sediment discharges was far wider than the range of annual water discharges, when normalised by the average value for each river. In statistical terms, the coefficient of variation (standard deviation divided by the mean) of the distribution of annual fluxes, was typically more than twice as large for sediment than water discharge. There are several practical implications of this result. First, the uncertainty in any estimate of mean annual sediment flux is likely to be very large, particularly for gauge records of limited duration. This potential error in sediment flux estimation is in addition to the uncertainty in any single measurement of sediment flux, which is often large for suspended load, as will be discussed in more detail below, and even more so for bedload, which is notoriously difficult to measure. Second, downstream river channels, floodplains, and resident ecosystems may be adapted as much to the natural variations in sediment supply as to the annual sediment flux as averaged over decadal time scales. Walling & Fang�s analysis also suggests that there is a global trend toward reduced sediment supply, which they attributed to the widespread trapping of sediment behind dams. Of the 142 rivers studied, 68 showed a decrease in sediment flux, 70 showed no change, and only 4 showed an increase in sediment flux over the period of record. Some case study examples Sidi Salem Dam, Medjerdah River, Tunisia Zahar & Albergel (1999) reported on the downstream effects of the Sidi Salem Dam, on the Medjerdah River. The dam, closed in 1981, intercepts water draining from 77% of the basin, and effectively traps nearly all sediment entering the reservoir. Mean annual discharge downstream has been reduced by 65% due to diversion for irrigation and evaporative losses. Peak sediment concentrations have been reduced from 60 to 2 g/l, leading to severe erosion of the coastal delta just north of the city of Tunis. Prior to dam construction, the delta had been prograding, having advanced 3000 m since Roman times. In addition to loss of delta land to coastal erosion, river bed erosion has undermined several bridges and other structures immediately downstream of the dam. Mequinenza-Ribarroja-Flix Dam Complex, Ebro River, Spain Sanz-Montero et al (1999) studied the loss of reservoir storage volume due to sedimentation in the major dams of the Ebro River, and the effects on the delta and estuary downstream. Over 180 dams were constructed in the basin during the 20th century. Pre-dam sediment flux to the delta was about 30 megatons per year but is now reduced to 0.2 megatons per year, less than 1% of the historic sediment supply. Mean annual flow in the Ebro River has declined by 30% during the same period, due to diversion and evaporative losses from reservoirs. Nearly half the total basin reservoir storage volume has been lost to sedimentation, with some reservoirs more than 80% filled. Although some



erosion of the delta has occurred, the primary downstream concern is salt water intrusion in the estuary and coastal groundwater. Itaipu and Yacyreta Dams, Parana River, Brazil, Paraguay and Argentina Amsler & Drago (1999) presented a sediment budget for the middle Parana River, downstream of Itaipu and Yacyreta Dams, and compared their results with a similar study done in the 1970s, prior to construction of the two dams. Nearly all suspended sediments are trapped by the two dams, resulting in a decrease in sediment flux of 80% since closure of the dams. Most of the remaining sediment is derived from bank erosion below the dams, where the channels are eroding due to the reduction in sediment supply. Downstream of the confluence of the Parana and Paraguay rivers the sediment load returns to its pre-dam level because of an increase in sediment supply of 85% on the Bermejo river, a tributary of the Paraguay that drains the Andean foothills. Amsler & Drago attribute the increase in sediment flux from the Bermejo to a wetter climate that has caused an increase in mean annual runoff throughout the upper Paraguay and Parana basins. Sediment Exchange between Floodplains and River Channels Dunne et al (1999) presented results of the most comprehensive sediment budget ever constructed for a large floodplain river system (Dunne et al, 1998). Dividing the main stem Amazon river into ten reaches, each about 200 km in length, Dunne et al measured the sediment fluxes between the river and floodplain and downstream along the river. Sediment enters each reach by in-channel suspended load and bedload transport, from local tributaries and by bank erosion. Sediment leaves the reach by deposition on bars, diffuse overbank flow, by flow into floodplain channels leading to lakes and other off-channel waterbodies, and by in-channel transport. They found that the magnitude of annual sediment exchange between the river and floodplain typically exceeds the magnitude of downstream annual sediment flux, often by a factor of nearly two. This result shows that sediment supply in the channel can be dominated by interaction with the floodplain. It also suggests that the floodplain is closely coupled to the channel system and thus is vulnerable to even subtle changes in channel sediment transport capacity and supply caused by construction of dams upstream. For example, reduction in sediment supply from upstream could lead to channel bed erosion and deepening of the channel cross-section, which in turn would reduce the frequency and duration of overbank flooding and limit sediment flux to the floodplain. Floodplain ecosystems would then experience a reduction in the supply of vital nutrients carried by fine-grained suspended sediments. This process has been documented by Ligon et al (1995) on the Oconee River in the southern United States. Smith & Sidorchuk (1999) looked at the Ob, Yenisey and other large Siberian rivers that drain into the Arctic Ocean. In these rivers, the timing of the annual ice break-up strongly influences the duration and extent of floodplain inundation, and thus the rate of sediment delivery to the floodplain. Compared with river water entering the floodplain, the water that drains from the floodplain wetlands system has a much lower suspended sediment concentration and an elevated organic carbon content, factors that are important for the coastal ecosystems of the arctic ocean. Rosales et al (1999) studied the influence of river confluences on floodplain ecosystem diversity in two large tributaries of the Orinoco River in Venezuela. They found a maximum in species diversity near tributary junctions that they attribute in part to more active, and more temporally variable, sediment exchange between channel and floodplain. Junctions are more dynamic because of differences in flood timing, flood magnitude, sediment load and sediment grain size between the main stem and tributary channels. Cumulative Impact Analysis Cumulative impact analysis is an essential tool for understanding and predicting the impacts of dams on large river basins. Unfortunately, little scholarly research been devoted to developing a theoretical framework or practical methodologies for cumulative impact analysis. Cumulative impacts have been defined in two different and useful ways. First, the indirect impacts which result from interaction of direct impacts, originating with a single intervention in the river system, can be considered cumulative



impacts. For example, the reduction in sediment flux downstream of a dam could lead to channel bed coarsening, while flow regulation by the same dam could result in the elimination of infrequent large discharge events and lead to channel narrowing due to vegetation encroachment. The combined effect of these two impacts may be sufficient to eliminate the spawning habitat for an endangered fish species, although either impact taken alone may not have had any major effect. A second type of cumulative impact results from the additive effects of multiple interventions in different places within the river system. For example, cold water releases from a high dam combined with a large reduction in suspended sediment flux downstream of a large volume storage dam on the same river, may result in water too nutrient poor and cold to allow the spring bloom of algae, which form the base of the aquatic food chain. The effects of either dam, taken individually, may not produce this result. The cumulative impacts illustrated above result from the existence of thresholds and feedbacks within river systems. Anticipating cumulative impacts involves more than the summing of individual impacts. The relevant thresholds need to be identified and the state of the system relative to those thresholds needs to be assessed. Cumulative impact analysis is difficult because it requires cross-disciplinary interaction among experts who are usually trained in reductionist approaches to science. Cumulative impacts are particularly important to assess in large rivers when a large number of dams may be built within a single basin.



Annex 5: Large Dam Projects: Adverse Environmental Impacts and Mitigation Options (Source: World Bank Submission to WCD/Ledec et al. 1997/INS082)

ENVIRONMENTAL IMPACTS MITIGATION OPTIONS

Flooding of Natural Habitats. Large reservoirs can permanently flood natural areas, with local and even global extinctions of animal and plant species. Particularly hard-hit are riverine forests and other riparian ecosystems, which naturally occur only along rivers and streams.

Compensatory Protected Areas. To compensate for the loss of natural habitats to reservoir flooding or other project components (such as borrow pits), one or more protected areas can be established and managed under the project. If the compensatory area is protected �on paper� only, a useful project option can be to strengthen its on-the-ground management. The area protected under the project should be ecologically similar to, and no smaller than, the natural area lost to the project. Hydroelectric and other projects should not be sited where they would cause the significant conversion or degradation of critical natural habitats that do not occur elsewhere (and, hence, cannot be adequately compensated).

Loss of Terrestrial Wildlife. An inherent consequence of the flooding of terrestrial natural habitats, the drowning of wildlife during reservoir filling, is often treated as a separate impact.

Wildlife Rescue. Widely practised for public relations purposes, the capture and relocation of wild animals during reservoir filling is usually of little or no conservation value. Instead of drowning, the �rescued� animals typically starve, are killed by competitors or predators, or fail to reproduce successfully, due to the limited carrying capacity of their new habitats. The money spent on rescue would usually do much more for wildlife conservation if it were invested in compensatory protected areas.

Downriver Hydrological Change. Major changes in downriver flows can destroy riparian ecosystems dependent on periodic natural flooding, exacerbate water pollution during low-flow periods, and increase saltwater intrusion near river mouths. Reduced sediment and nutrient loads downriver of dams can increase river-edge and coastal erosion and damage the biological and economic productivity of rivers and estuaries. Induced desiccation of rivers below dams (when the water is diverted to another portion of the river, or to a different river) kills fish and other fauna and flora dependent on the river; it can also damage agriculture and human water supplies.

Management of Water Releases. Objectives to be considered in optimising water releases from the turbines and spillways include adequate downriver water supply for riparian ecosystems, reservoir and downriver fish survival, reservoir and downriver water quality, aquatic weed and disease vector (eg mosquito) control, irrigation and other human uses of water, downriver flood protection, recreation, and power generation. From an ecological standpoint, the ideal water release pattern would usually closely mimic the natural flooding regime (although this may not be feasible for densely settled floodplains where flood protection is a high priority). Environmental management plans for hydroelectric projects should specify environmental water releases, even for dams owned or operated by the private sector.



Water Quality Deterioration. Damming of rivers can cause serious water quality problems due to reduced oxygenation and dilution of pollutants by relatively stagnant reservoirs (compared to fast-flowing rivers), flooding of biomass (especially forests) and the resulting underwater decay, and/or reservoir stratification (where deeper lake waters lack oxygen).

Water pollution control measures (such as sewage treatment plants or enforcement of industrial regulations) may be needed to improve reservoir water quality. Where poor water quality would result from the decay of flooded biomass, selective forest clearing within the impoundment area should be completed before reservoir filling.

Fish and Other Aquatic Life. Reservoirs can positively affect some fish species (and fisheries) by increasing the area of available aquatic habitat. However, the net impacts are often negative because (a) the dam blocks upriver fish migrations, while downriver passage through turbines or over spillways is often unsuccessful, (b) many river-adapted fish and other aquatic species cannot survive in reservoirs, (c) changes in downriver flow patterns adversely affect many species, and (d) water quality deterioration in or below reservoirs (usually low oxygen levels; sometimes gas supersaturation) kills fish and damages aquatic habitats. Freshwater molluscs, crustaceans, and other benthic organisms are even more sensitive to these changes than most fish species, due to their limited mobility.

Fish Management Measures. Management of water releases (see above) may be needed for the survival of certain fish species, in and below the reservoir. Fish passage facilities (fish ladders, elevators, or trap-and-truck stations) are intended to help migratory fish move upriver past a dam; they are usually of limited effectiveness for various reasons (including the difficulty of ensuring safe downriver passage for many adults and fry). Fish hatcheries can be useful for maintaining populations of native species which can survive but not successfully reproduce within the reservoir. They are also often used for stocking the reservoir with economically desired species, although introducing non-native fish is often devastating to native species and not ecologically desirable. Fishing regulation is often essential to maintain viable populations of commercially valuable species, especially in the waters immediately below a dam where migratory fish species concentrate in high numbers and are unnaturally easy to catch.

Floating Aquatic Vegetation. Floating aquatic weeds can rapidly proliferate in eutrophic reservoirs, causing problems such as (a) degraded habitat for most species of fish and other aquatic life, (b) improved breeding grounds for mosquitoes and other nuisance species and disease vectors, (c) impeded navigation and swimming, (d) clogging of electro-mechanical equipment at dams, and (e) increased water loss from some reservoirs.

Pollution control and preimpoundment selective forest clearing will make reservoir conditions less conducive to aquatic weed proliferation. Physical removal of floating aquatic weeds is effective, though an expensive recurrent cost for large reservoirs. Where compatible with other objectives (power generation, fish survival, etc.), occasional drawdown of reservoir water levels may be used to kill aquatic weeds. Chemical poisoning of weeds or related insect pests requires much environmental caution and is usually best avoided.

Greenhouse Gases. While relatively small reservoirs (which flood little or no forest) are essentially �carbon neutral,� larger reservoirs can release significant quantities of carbon dioxide and methane into the atmosphere, either slowly (as flooded organic matter decomposes) or rapidly (if forest is cut and burned before reservoir filling). These �greenhouse gases� are strongly suspected of causing human-induced climate change.

None, other than choosing reservoir sites which minimise the flooding of large tracts of land in general, and forests in particular. Although carbon releases could be slightly reduced by thorough salvage of commercial timber, in practice this rarely happens because of (a) high extraction and transportation costs, (b) marketing constraints, and (c) pressures to fill the reservoir quickly.



Reservoir Sedimentation. Over time, reservoir storage and power generation are reduced by sedimentation, such that much of a project�s hydroelectric energy may not be renewable in the long term.

Watershed Management. To minimise reservoir sedimentation rates, it may be necessary (though often difficult) to control road construction, mining, agriculture, and other human activities in the watershed. Protected areas are sometimes established in upper watersheds to reduce sediment flows into reservoirs.

Access Road Impacts. New access roads to hydroelectric dams can indirectly result in major land use changes � particularly deforestation � with resulting loss of biodiversity, accelerated erosion, and other environmental problems. In some projects, the environmental impacts of access roads can greatly exceed those of the reservoir.

Careful Road Siting, Design, and Construction. Any new access roads should be sited in the environmentally and socially least damaging corridors. Forests and other environmentally sensitive areas along the chosen road corridor should receive legal and on-the-ground protection. Environmental rules for contractors (including penalties for non-compliance) should cover proper drainage, construction camp siting, gravel extraction, waste disposal, avoiding water pollution, worker behaviour (such as no hunting), etc.

Transmission Line Impacts. Power line corridors can directly reduce and fragment forests; indirectly, they occasionally facilitate further deforestation by improving physical access. Large birds are sometimes killed in collisions with power lines. Power lines can also be aesthetically objectionable.

Siting of power lines should be in the environmentally, socially, and aesthetically least damaging corridors, with good environmental practices used during construction (as with roads, above). In areas with concentrations of vulnerable bird species, the top (grounding) wire should be made more visible by using special devices.

Quarries and Borrow Pits. Used to provide material for construction of the dam and other civil works, these pits can increase the area of natural habitats or agricultural lands that are lost to a hydroelectric project.

To the maximum extent feasible, quarries and borrow pits should be sited within the future inundation zone. Where this is not feasible, the pits should be rehabilitated after use, ideally for ecological purposes (such as wetland habitats).

Associated Development Projects. Hydroelectric projects often make possible other projects with high environmental impacts, including irrigation and new residential or industrial development (due to new water supplies).

Environmental Planning and Mitigation. Major new irrigation, water supply, or industrial development projects should be planned to minimise adverse environmental and social impacts. Environmental impact assessment studies should be carried out in the early stages of project planning; the resulting environmental mitigation plans should be fully implemented.

Additional Dams. The construction of the first dam on a river can make the subsequent construction of additional dams more economical, because flow regulation by the upriver dam can enhance power generation at the downriver dam(s).

Cumulative Environmental Assessment. The environmental impact assessment study for the first dam on any river should include a cumulative assessment of the likely impacts of proposed additional dams on the same river system. Implementation of mitigation measures for cumulative (rather than dam-specific) environmental impacts should be well underway prior to construction of the second dam on the river.



Annex 6: Environmental Flow Requirements (EFR) (Based on King et al. 1999) The review commissioned by the WCD (Southern Waters 1999) examines whether scientific knowledge is sufficiently advanced to permit development of guidelines on how much of the original flow regime of a river should continue to flow in order to maintain specified valued features of the riverine ecosystem. Such an assessment is usually linked to a proposed water-resource development or river rehabilitation scheme, probably because interest and funds are focussed on that specific river at that time; but an assessment of flow requirement may be done for any river at any time. It is used to assess how much water could be abstracted from a river without an unacceptable level of degradation of the riverine ecosystem or, for a highly modified river with much abstraction, how much of its original flow should be reinstated in order to rehabilitate the ecosystem to some desired condition. In the context of the wider river environment, neither of the historical terms, "instream flow assessment" or "biophysical flow assessment", seem appropriate. The terminology assessment of �environmental flow requirement� (EFR) is more all encompassing, and is adopted in this report. EFRs have two main areas of focus: 1) the different flow regimes that would maintain a river ecosystem at various levels of health (condition); and 2) the ways those different levels of river health will affect people. The second area of focus can be subdivided in various ways: the people using the river for sustenance versus all other stakeholders (regional, national and international); or similarly, but not necessarily quite the same, the issues that can be costed (loss of resources, cost of development) versus those that cannot (moral and ethical issues, legalities, intangible river values). Each area of focus is addressed by different kinds of specialists. A comprehensive assessment of EFR will employ them all, combining their inputs in a structured and transparent way. A flow assessment produces one or more descriptions of possible future flow regimes for a river, each linked to an objective which this achieves in terms of the condition or health of the riverine ecosystem. Each possible future flow regime provides the environmental flow requirement for achieving that objective. For instance, the requirement may be stated as simply as �a water depth of at least 50 cm throughout the year, to provide adequate wetted habitat areas for fish species A�. Alternatively, it may be described with much greater complexity, detailing a comprehensive flow regime, with specified magnitudes, timing and duration of low flows and floods at both intra-annual and inter-annual scales of variability, all designed to maintain fundamental functioning of all ecosystem components (eg, fish, riparian trees, water chemistry) at a specified level of condition. The linking of �condition� with �flow regime� indicates that rivers may be maintained in a range of conditions. Rivers maintained close to natural require more of their natural flow regime than those for which extensive modification is acceptable. Recognising this, the assessment and resulting EFR can be viewed and used from two perspectives. First, the assessment can be made, and the flow requirement stated, by any stakeholder group, in order to present in a negotiating forum their aspirations for the river. Second, as other stakeholders may have different aspirations and thus different EFRs for the river, compromises may be sought and agreed upon. In this situation, an agreed compromise solution reflects the eventual EFR and condition for the river. If no compromise can be agreed upon, a decision-maker would have to make a decision on the future river condition and associated flow regime (EFR), and be accountable for that decision. To assess EFRs, river scientists link valued features of the river to the amount of water required for their maintenance. Such valued features could include a Red Data Book fish species threatened with extinction, riparian forests, a harvestable resource, or sufficient water of a certain quality for a specific use such as washing clothes or watering livestock. The links can be made at various levels of complexity. The simplest level may be a desk-based study of the past and present hydrological character of the river, linked to a review of any published literature on the riverine ecosystem, providing a coarse calculation of the kinds of flow needed in a generic way to support riverine biotas. The highest level of complexity may require an intensive, interdisciplinary, long-term study, with



extensive fieldwork, to enhance understanding of the nature and functioning of the river. These data would be used to provide clear quantitative descriptions of the consequences for all ecosystem components of different potential manipulations of the flow regime. Usually, the greater the investment in the fieldwork and other specialist inputs, the higher the confidence in the output. Measuring the success of an EFR is difficult because change takes place over such long distances and time-spans, and also because changes that are not flow-related will also be occurring in the river. As an example, if a channel is becoming narrower, it could be because an upstream dam has harnessed scouring floods, or because alien trees have infested riverbanks and fallen into the water, or both. The former situation is flow-related. The latter may not initially be flow-related, but due to disturbance of the surrounding land, with this in turn allowing invasion into the riparian zone of shallow-rooted alien vegetation with poor bank-stabilising abilities. This condition could then evolve into flow-related degradation of the river as the invading trees would be more likely to collapse into the river during floods than would native riparian ones, thus further destabilising banks. Similarly, the objective to maintain a Red Data Book species within a river may fail despite a very favourable flow regime simply because catchment activities caused deterioration of water quality in the river beyond that which the species could tolerate. In these kinds of complex pictures � which are probably the norm � it is not easy to assess the success of an EFR. Post-development monitoring of river health is an integral part of the water-resource development, especially if the assessment of the EFR is linked to a proposed re-structuring of releases from an extant dam or intended to guide the control of run-of-river abstractions. Four basic groups of methodology for assessment of EFRs are widely recognised: hydrological index methodologies; hydraulic rating methodologies; habitat simulation methodologies; and holistic methodologies. They are fully explained in Annex 2 of Southern Waters 1999. The evidence presented in Annex 2, drawing from many countries, leads to the conclusion that environmental flows are not a waste of water. They are working. Rivers are becoming "healthier"; more birds are breeding, more fish are swimming, more wetlands are filling. The decline of many rivers and wetlands has been slowed down in some areas, and time will tell if they have been turned around. But generally speaking, the volumes being released fall far short of what is needed and many physical and institutional and psychological hurdles of the type described in Chapter 4 remain. All types of methodology, for all applications, however, have in common a considerable reliance on professional judgement.



Annex 7: Example of Mitigation Measures Mitigation, Enhancement and Compensation Measures Applicable on the Physical Environment: Water Quality (Presented by IEA 2000)

Impacts Measures Type Climatic Zone

Biome Effective-ness

Comments Measures applied

Measures proposed

Eutrophication of reservoir Mechanical removal of macrophytes and

phytoplankton on the reservoir surface to limit eutrophication of the reservoir

M TR-T F-S-W LE Measure that is simple to apply but requires on-going follow-up. Alone, it is not sufficient to achieve the objectives of limiting the eutrophication of the reservoir. Must be applied in combination with other measures, such as forced aeration, discharge measurement, etc.

● 42, 104 ◆ 128

● 26

Mercury methylation Removal and controlled burning of vegetation before impounding.

M TR-T-C M-F E-LE This measure seems to effectively reduce the decomposition of organic matter that results in organic enrichment of reservoir waters. On the other hand, it is often very expensive and its secondary impacts on fishing are not negligible. Moreover, in tropical regions, cleared zones are recolonized at a very high rate, often recreating bush that is similar to that cut before impounding.

● 42, 53 ◆ 133 ✱ 2, 21, 101, 130, 131 # 129

# 129

Gas saturation Modification of the exit of the tailrace canal to avoid oversaturation of water with gases

M T C E Meets intended objectives. ◆ 12 ✱ 2, 16

Thermal shock Installation of water intakes at various levels in the reservoir to mitigate the impact of thermal shock downstream of the reservoir.

M T C E Simple mitigation measure that is easy to apply. It completely mitigates the impact of thermal shock downstream of the dam. Should be planned as part of project planning.

◆ 128 ✱ 2 # 129

# 129

Management of releases to reduce salinity of waters feeding the diversion canal.

M TR W ND ● 26

Saline intrusion Construction of protective dikes in the diversion canal to control inflows of salt water in an estuary.

M TR W NE This measure seems inadequate on its own. ● 38

Sealing the diversion canal walls to reduce rise of water table and risks of waterlogging and salinization of surrounding land.

M TR S-W ND ● 26

Transmission lines and energy transformation substations and refurbishing power stations Measure Type M: mitigation, E: enhancement, C: compensation

Biome M: mountain, F: forest, S: savannah, W: wetland

Effectiveness E: effective, LE: limited effectiveness, NE: non effective, ND: undetermined

References as per climatic zones # general, $ (TR) tropical region, % (T) temperate, ✱ (C) cold region



Appendix I � List of Contributing Papers to Thematic Review II.1 Contributing Paper/ Synthesis Paper Writers Affiliation Contract co-

ordinator(s)/ funder Molluscan Biodiversity and the Impact of Large Dams

M.B. Seddon National Museum & Galleries of Wales, UK

IUCN/UNEP

D. C. Jackson Mississippi State University, USA The Influence of Dams on River Fisheries G. Marmulla Fisheries Department, FAO, U.N.

FAO/WCD

M.P. McCartney Institute of Hydrology, UK C. Sullivan Institute of Hydrology, UK

Ecosystem Impacts of Large Dams

M.C. Acreman Institute of Hydrology, UK

IUCN/UNEP

D.E. McAllister Consultant on biodiversity J.Craig Consultant on Fish and Fisheries N. Davidson Wetlands International M. Seddon National Museum of Wales, UK

Biodiversity Impacts of Large Dams

D.Murray OPIRG, Carleton University, UK

IUCN/UNEP

Large Dams and Freshwater Fish Biodiversity J.F.Craig Whiteside, Dunscore, UK FAO/WCD J. King University of Cape Town, South Africa R. Tharme University of Cape Town, South Africa

Definition and Implementation of Instream Flows

C. Brown University of Cape Town, South Africa

WCD

J. King University of Cape Town, South Africa Information Needs for Appraisal and Monitoring of Ecosystem Impacts C. Brown University of Cape Town, South Africa

WCD

Dams and Fish Migration M. Larinier Institut de Mecanique des Fluides, Toulouse

FAO/WCD

N. Davidson Wetlands International Biodiversity Impacts of Large Dams: Waterbirds S. Delany Wetlands International

IUCN/UNEP

Fundamental Legal and Ethical Principles In Adjudging the Merits of Development Projects

C. Di Leva IUCN Environmental Law Center, Bonn IUCN/UNEP



M.C. Acreman Institute of Hydrology, UK E. Barbier University of York, UK M. Birley University of Liverpool, UK K. Campbell Natural Resources Institute F. Farquharson Institute of Hydrology, UK N. Hodgson Natural Resources Institute J. Lazenby Gibb Ltd M. McCartney Institute of Hydrology, UK J. Morton Natural Resources Institute D. Smith Natural Resources Institute

Managed Flood Releases from Reservoirs � A Review of Current Problems and Future Prospects

C. Sullivan Institute of Hydrology, UK

DFID

Report on the Conference on Hydrological and Geochemical Processes in Large Scale River Basins, 15-19 November, 1999, Manaus, Brazil

L. Sklar University of California, Berkeley, USA WCD

International Mechanisms for Avoiding, Mitigating and Compensating The Impacts of Large Dams on Aquatic and Related Ecosystems and Species

J. R. Bizer Independent Consultant IUCN/UNEP

Capacity and Information Base Requirements for Effective Management of Fish Biodiversity, Fish Stocks and Fisheries Threatened or Affected by Dams During the Project Cycle

G. Bernacsek Fisheries and Environment Specialist FAO/WCD



Appendix II - Submissions for Thematic Review II.1 The WCD is committed to an open and consultative process. To broaden the scope for participation and input from all interested groups and stakeholders the Commission invites submissions on all aspects related to its work programme. As they are received, submissions are classified according to the area(s) of the work programme to which they are relevant. Therefore the submissions used here are those that have been identified as applicable to the Thematic Review (II.1) on Dams, Ecosystem Functions, and Environmental Restoration. Submissions arrive in parallel to the drafting process of the WCD�s reports. Those listed here are the 203 submissions specifically for TR II.1 which were received by March 31, 2000. Note that submissions are listed alphabetically by author and are therefore not numbered sequentially. Every submission has been read carefully. Some are informed individual perspectives on which the WCD can not mediate. For example, there are some submissions that seek the endorsement of the WCD, and the WCD�s mandate is neither to adjudicate nor to mediate on specific dams or disputes. Therefore, the submissions received for Thematic Review II.1 have been used as background information. Many have been cited as references throughout the text. Others have been used as part of a text box to illustrate a point. All submissions have informed the WCD as to the different positions on the dams debate. A few submissions only included an abstract or an outline for a presentation at one of the consultations with insufficient detail to be included.

Author Serial Title Abbueid, Abdalla M ENV099 Water Aquifers in Palestine Adil, M Ali ENV104 El Girba Dam and its Environmental Effects Agarwal, A ENV012 Private Power: Maheshwar Project in Narmada Valley Anand, A ENV040 Appeal Against Installation of the Barge Mounted Hydroelectric Project Proposed to be set up in Tadadi, Kumta

Taluk, Uttara Kannada District Anane, Mike ENV100 Ghana's Proposed Bui Dam to Cause Havoc: 150 Endangered Hippos in and National Park in Danger Anastacio Afonso Juras ENV065 Fishing studies on Tucurui Dam Angelucci, Carlo ENV197 Comments from ITCOLD Armando Llop, Maria Valeria Mendoza ENV058 Assessing the Environmental Impact of Large Dams on the Agricultural Sector in Southern Mendoza, Argentina Ashish Kothari and Rahul N. Ram ENV132 Environmental aspects of the Sardar Sarovar Project Baird, Ian B ENV154 Indigenous fish, Artisanal Fisheries and the Cycles of the Mekong River in Southern Lao PDR Bakhsh, H; Ghaffar, A; Sario, P ENV046 Stop Damming the Indus Bank, Robert ENV179 Comments on Pine River Dam




Taluk, Uttara Kannada District Bank, Robert ENV180 Kangaroo Rat & Seven Oaks Dam Bansuri Taneja ENV128 Restoration of Rivers : Lessons From Environmental Recovery and Restoration in Northwest India Barnard, Bob ENV213 Information on Mitigation on Dams Belanger, Robert ENV212 Environmental Mitigation Measures on Dams Bell, JE, Gilkes, PW and Millmore, JP ENV036 The Roadford Scheme: Planning, Reservoir Construction and the Environment Billore, R. ENV004 Narmada Sagar: A Case for Review Binnie, Chris ENV052 Ghazi-Barotha HydroPower Project: Environmental Assessment Binnie, Chris ENV172 Various Comments Birley,M ENV044 Comment of Impacts Table Bozek, Jacek ENV228 Cascade Punch for the Vistula Braga, Maria Isabel ENV171 Integrating Freshwater Ecosystem Function & Services with Water Development Projects Bridle, Rodney ENV166 The Benefits of Dams to British Society Bridle, Rodney ENV190 Environmental mitigation at Pollan Dam-Design and construction implications Bridle, Rodney ENV191 Roadford Lake-Leisure Development Bridle, Rodney ENV192 Roadford Reservoir-enhanced flows, fisheries and hydroelectric power generation Bridle, Rodney ENV193 Maximizing the ecological benefit of a new small reservoir Bridle, Rodney ENV194 Environmental assessment-legislation and planning requirements within the European framework Bridle, Rodney ENV195 Reservoir, environment and private sector Brink, Elizabeth ENV054 Dam Removal and Sedimentation Brink, Elizabeth ENV130 Cumulative effects of Dams - a resource file Brink, Elizabeth ENV131 Changes in the habitat and fish community of the Milwaukee River, Wisconsin, following removal of the Woollen

Mills Dam Broome, Kenneth ENV185 Alternatives of small dams Canese, Ricardo ENV074 Preliminary profile of Water Resources Management of Parana River Project Cappato, J ENV030 Declaration of San Jose, Costa Rica on Dams and Wetlands Casinader, Ranji ENV200 Note on Environmental & Social mitigation mechanisms in planning design & construction




Taluk, Uttara Kannada District Chang Cheng-Yang ENV150 The Meinung Dam: Eleven Major Concerns Chang, Cheng-Yang ENV084 Experiences from Meinung Dam, KaoPing River, Taiwan Choudhury, GA ENV019 Experiences with Large Dams in Bangladesh Comision Nacional Del Agua ENV094 Los Consejos De Cuenca En Mexico Definiciones Y Alcances Cooke, Sanjiva & Joffe, Steen ENV057 Management of the Water Hyacinth and Other Invasive Aquatic Weeds: Issue For the World Bank Cruz, Marcos Orellana ENV225 Failed Mitigation and Compensation Schemes in dam Construction: The Pehuen Foundation in the Pangue/Ralco

Project in the Alto BioBio Datye, K R ENV221 Biomass Strategy and a Scenario for 2025 Dave, JM. ENV005 Environmental Issues related to Large Dams and Alternatives David Kiell ENV202 Mitigation and Compensation Measures/Programs at Hydroelectric Developments-Newfoundland and Labrador

Hydro De Alwis, D ENV020 Balancing Integrated Rural Development with Environmental Conservation around Reservoirs Delaunay, Alexis ENV189 Effacement du Barrage de Maisons Rouges Delio,Julio & Bacchiega, Jorge & Fattor, Claudio

ENV116 Massive death of fishes by supersaturation of total dissolved gases: the case of Yacyreta Dam

Department of Landscape Planning- Swedish University of Agricultural Sciences

ENV214 Review of Environmental Mitigation and Compensation Measures in Sida Financed Hydropower Projects

Desarda; HM ENV016 Toward Sustainable Water and Energy Resource Development DFID-Mott Mcdonald ENV203 Mitigating Large Dams Dietrich, William ENV082 Impacts of dams on River geomorphology Dr J.D. Cadman ENV061 The Environmental Aspects of Six Hydro Reservoirs in the Amazon Basin Dr Kefialew Abate ENV103 Large Dams and Sustainable Development in Africa: An Overview DSI -Turkey ENV142 Archaeological Investigations of Illisu Dam, Turkey E.A.K. Kalitsi ENV101 Dams and Ecosystems: Assessing and Managing Environmental Impacts (Ghana's Experience) Earthlife Africa ENV091 Water Demand Management in tha Mgeni River Catchment - Kwazulu Natal English, Graham ENV205 Environmental Mitigation Measures-Shongweni Dam




Taluk, Uttara Kannada District European Rivers Network - for living rivers- (ERN)

ENV144 Information on Dam Construction in a Russian National park on the Belaya River

Fearnside, Philip M. ENV059 Publications on Human Carrying Capacity, Agroecosystems, Deforestation, and Development Planning in Brazilian Amazonia

Fearnside, PM ENV034 China's Three Gorges Dam: "Fatal" Project or Step Toward Modernisation? Fernando Esquivel Rafael Galo ENV069 The Pacuare Dams Fernando Francis ENV064 The Corumba Hydroelectric Plant and the Geothermal Resources of Caldas Novas Fick, Loraine ENV199 Skuifraam Dam: Outlet Works-DWAF Discussion Report Fields, Daryl ENV231 BC Hydro Environmental Measures Flaim, Sam ENV167 Flaim, Sam ENV168 Optimal Provision of Hydroelectric Power Under Env & Regulatory Constraints Flanders, Doug ENV219 Effectiveness of Fishways and Fish lifts in Queensland Fone, Doug ENV226 Environmental Mitigation Measures for Dams Franzin, R ENV053 The Piave River: A Case of Contested Waters Friederich, Hans ENV148 The Biodiversity of the Wetlands in the Lower Mekong Basin Fujita, Rodney M. ENV090 Thematic Review Submission on Ecological Indicators for Monitoring Ecosystem health Fujita, Rodney M. ENV095 Essential Ecological Indicators for the San Francisco Bay-Delta-River System Fujita, Rodney M. ENV096 Conceptual Framework for Indicator Development: Development of key Ecological Attributes for the San

Francisco bay-Delta Watershed Gabriela Fried1 ENV079 Swiss Federal Institution for Environmental Science and technology (EAWAG) Getzen, Beverly ENV176 Information Paper: Lower Snake River Fish & Wildlife Compensation Plan Ghimire, DJ ENV018 Environmental Monitoring of the Construction of the Upper Bhotekoshi Project Ghosh, A ENV008 Water & Power: A People-Centric Development Alternative Giancarlo Fanelli ENV140 Dams and the Environment Gornitz, Vivien ENV050 Effects of anthropogenic intervention in the land hydrologic cycle on global sea level rise Gracia, CL ENV022 Biscarrues Dam Project. Gallego River. Ebro Basin. SPAIN




Taluk, Uttara Kannada District Gracia, CL ENV023 Santaliestra Dam Paroject, Esera River, Ebro Basin, Spain Gracia, CL ENV024 Itoiz Dam Project. Irati River. Ebro Basin. Gracia, CL ENV025 Valle del Genal Dam Project. General River. Sur Basin Gracia, CL ENV026 Hozgarganta Dam Project. Hozgarganta River. Sur Basin Gracia, CL ENV027 Lechago Dam Project. Pancrudo River. Ebro Basin Gracia, CL ENV047 Melonares Dam Project. Viar River. Guadalquivir Basin (Spain) Grethel Aguilar Rojas ENV070 Impact assessment and hydroelectric dams in Central America. A legal perspective Groupe Caisse Francaise De Developpement ENV093 Urban and Periurban Water Supply and Sanitation Groupe Caisse Francaise De Developpement ENV117 Water Supply in Villages and Small Towns Groupe Caisse Francaise De Developpement ENV118 Hydro-electricity Groupe Caisse Francaise De Developpement ENV119 Irrigation and Livestock Water Supply Guha, SK. ENV002 Earthquakes Induced Following Impoundment Gujja, Biksham ENV224 Dams: Impact on River's Life Gujja, Biksham & Hunziker, Diwata Olalia ENV230 The Impact of Dams on Life in Rivers Hagenbucher, Thomas ENV198 Mitigation in three projects Hamerlynck, Olivier ENV164 The Diawling National Park, Mauritania: conflict and development around a newly established� Harris, Garth ENV163 Hloele, Thabo ENV215 LHWP- Mitigation and Compensation Measures Hori, Hiroshi ENV147 Recommendation for the Study of Development and the Environment in the Lower Mekong Basin Horowitz, Michael M ENV159 Environment and Society in the Lower Mekong Basin hydro-Quebec & GDG CONSEIL inc ENV051 Biodiversity and Hydroelectricity: World outlook and Quebec Context ICOLD ENV141 Benefits of and Concerns about Dams: The Italian Case: Case Studies, Antalya, Turkey, 1999. Infrastructure Development Insitute- Japan ENV143 Rivers in Japan '98 International Network on Small Hydro Power ENV208 Environmental Mitigation Measures for Dams Iriyagolle, G ENV039 Curse of Large Dams- The Sri-Lankan Experience- A Compilation on the Mahaweli Master Plan




Taluk, Uttara Kannada District ITAIPU Binacional ENV076 Itaipu and the Environment Jacques de Boissezon ENV106 Pour un gestiondes plans d'eau naturels et artificiels respectue Jobin, William ENV181 Various examples John L. Mckern ENV137 Improving Salmon Passage: Draft: The Lower Snake River Juvenile Salmon Migration Feasibility Report/

Environmental Impact Statement Jose Antonio Ribeiro ENV111 Flavio Nenflidio Carvalho Jose Roberto Fontes Castro ENV075 Statement on Pilar dam project, Minas Gerais State, Brazil Kallar Protection Group ENV014 The Vamanapuram Irrigation Project Karim S Numayr ENV102 Sedimentation and Water Quality Problems at the King Talal Reservior Karmacharya, Janak ENV178 General Comments Karpowicz, Charles ENV211 Highlights of NPS Dams Program Report to Congress for 98-99 Kataoka, N ENV031 The Nagara River Estuary Dam Kato, Akira ENV222 The Environment at Dams Kettab, A & Remini, B ENV048 Reservoir Sedimentation in Meghreb's Dams Khwaja, Aslam ENV045 Some Aspects of Opposing Kalabagh Dam Kim, Young-Sook ENV158 Balance Destruction of Tides by Blocking the Flow- Hagu Dam Naktong River Kondolf, G Mathias ENV085 Dams and Ecosytems, Regarding the Impacts of Dams on River Geomorphology, and the Feasibility of

Restoration Strategies Kondolf, G. Mathias ENV088 Thematic Review on Dams and Ecosystems, regarding the impacts of dams on river geomorphology, and the

feasibility of restoration strategies Kulkarni, P. ENV003 Sharavathy Tail Race Project: Yet Another Monument of Destruction Larsen, Thrond Berge ENV165 Norwegian Hydro Power Resources: Use, Management & Planning Lawson, JD; Sambrook, HT; Solomon, JD and Weilding, G

ENV037 The Roadford Scheme: Minimising the Environmental Impact on Affected Catchments.

Le, Dien Duc ENV156 Dam - An Unsustainable Way of Development Lemaire, Bernard ENV183 Protected Areas & Dams: The case of the Senegal River Delta Lemaire, Bernard ENV184 La voie fonciere et administrative an appui a la voie ecologique et agricole pour une gestion decent




Taluk, Uttara Kannada District Lovgren, Lars ENV136 Moratorium in Sweden: An account of the dams debate Mahoney, James ENV182 Specific examples from China & Venezuela Manning, Ian ENV173 Comments from own experience Marcus Aurelius Minervino ENV066 Integrated management of landed-based activities in the Sao Francisco River basin: Phase I Marcus Aurelius Minervino ENV068 Integrated management of landed-based activities in the Sao Francisco River basin: Phase I Marwa Doudy ENV108 The Development of the Euphrates & Tigris Basins: An Assessment of Upstream Development (Turkey) on

Downstream Riparians (Syria) Matola, S ENV001 Pristine Belize Habitat under Serious Threat McCully, P ENV033 Why Ecological Knowledge is Ignored in Dam Planning Mediwake, LW ENV017 Negative Effects of Victoria Dam Project Mehta, MB ENV006 Environmental Impacts & Issues of Large Dams & the Alternatives Micceslau Kudlavicz ENV063 Porto Primavera Dam in Rio Parana Mieceslau Kudlavicz ENV129 Relato: Barragem de Porto Primavera no Rio Parana Ministry of Construction-Japan ENV145 MIYAGASE: Affluent in Nature Forever Mishra, G and Singh, TP ENV041 Dams in North Bihar Mitchell, Thomas ENV187 Customized response from US Bureau of Reclamation Mitchell, Thomas ENV188 Native American Indian rural water systems required because of impacts to Missouri River lands Moores, Nial ENV157 Estuarine Dams: Declining Waterfowl Populations and Diversity in the Nakdong Estuary, South Korea Mora, Dennis ENV223 Plan de Mejoramiento Ambiental de la Parte Alta de la Cuenca del Rio Virilla Mosley, JG ENV028 Why lake Pedder should be Restored Mosley, JG ENV029 How Lake Pedder can be Restored Mothepu, Mahlape ENV175 Env Mitigation Measures for Dams: The Lesotho Highlands Water Project Experience Nature Conservation Council of NSW ENV049 Dams, Ecosystem Functions and Environmental Restoration Neuhauser, C ENV186 Special Conditions from Construction Contract Nicolaas van Zalinge, Nao Thuok & Sam Nuov ENV151 Cambodia's Inland Fisheries and the Dams of the Mekong Basin Nuttall, Peter ENV196 Leading Edge Approach to Instream Flow Requirements for the Lesotho Highlands Water Project




Taluk, Uttara Kannada District Nyoman, I & Suryadiputra, N ENV153 Biodiversity and Water Quality of the Ciujun Ciliman River Basin, West Java- Indonesia Onta, I R ENV209 Environmental Mitigation Measures for Dams Owen, Philip ENV109 Tree Plantations and Water in South Africa Potnis; VD ENV009 Impact on Narmada River Ecosystem of Namada Valley Project (NVDA): Limnological Issues & Aspects Prescott, John P ENV210 World Commission on Dams Survey Prof. Emmanuel Obot ENV098 Large Dams, Biodiversity and Local Livelihood: A summary of Experience from 3 dams, Kainji, Jebba and

Shihoro in Northern Nigeria Purdom, Roger ENV161 Mitigation in Columbia River Rafael Bastos ENV067 The question of water in the evaluation of environmental impacts of dam construction Rivero, Cristina ENV177 General comments Robert Belanger, P.Eng. ENV139 Dam Monitoring and Instrumentation Rodomiro Ortiz ENV092 Potential for Improving Agricultural Production through Biotechnology in the Semi-Arid Tropics Rofe, BH ENV038 Reservoirs: An Environmental Gain Ron Gee ENV201 Yukon Energy Corporation's response to the WCD's request on information on environmental mitigation Rozengurt, Michael ENV152 Controversial Influence of Impounded Large Rivers on their Delta-Estuary-Coastal Ecosystems Rudkowski, Clarice Blake ENV056 My River, My Home Sale, M J et al ENV206 Environmental Mitigation at Hydroelectric Projects. Volume 1: Current Practices for Instream Flow Needs,

Dissolved Oxygen and Fish Passage Sasidharan; M ENV011 Large Dams and Alternatives: A Perspective View Saunders, Rob ENV162 Mitigation & Small Dams Shaojun Xiong ENV149 Fragmentation and Flow Regulation of the Large Rivers in China and South East Asia Siazo, Mugiel ENV207 Environmental Mitigation Measures for Dams Singh, Arun ENV133 A Seismo-Tectonic Framework of the Narmada Valley and its Implications for the Sardar Sarova Project Singh, Arun ENV135 Manasiwakal Project Siwakoti, Gopal ENV081 Smith, Brian D ENV227 Downstream Effects on Biodiversity of a Planned High Dam in the Karnali River, Nepal




Taluk, Uttara Kannada District Smith, Richard ENV174 General comments Sonia Santos Baumgratz, Environmental Control Plan Coordinator

ENV071 Environmental management of the Guilman-Amorim HHP, located in Minas Gerais, Brazil (Doce river basin)

Stuart Blanche ENV204 Environmental Flows: Present and Future Sullivan, Dr. Caroline ENV089 The Economic Impact of Changes in River Profiles. Upstream impacts Tabeth Chiuta ENV105 Vision for Water and Nature: Link with Environmental Issues of Dams Takehiro Nakamura et al ENV107 Implication of Dams on the Freshwater and Coastal Environment and its Resources Taylor, Richard ENV087 Research needs for water quality issues relating to hydro reservoirs The Common Wealth ENV216 Salinity and Drainage Strategy- Ten Years On The Common Wealth ENV218 The Salinity Audit of the Murray Darling Basin: A 100 Year Perspective Toyishiki Adachi ENV146 Conserving a Valuable Resource in Harmony with the Environment (Video Tape) Truffer, Bernhard ENV169 Truffer, Bernhard ENV170 Visvanathan; N ENV010 Impacts of Construction of Large Dams: A Case Study Vladut, T ENV007 Specific Environmental Issues of Dams Vladut, T ENV042 Seismicity Concerns Vladut, T ENV043 Response to Martin Birley's Comment Vu, Van Tuan ENV155 Some Preliminary Assessments of the Influences of Hoabinh Dam/Reservoir on Environment Waage, Jeff ENV097 Alien Invasive Waterweeds- A Threat to Dams and Water Resources in Developing Countries Walker, Glen, Gilfedder, Mat & Williams, John ENV217 Effectiveness of Current Farming Systems in the Control of Dryland Salinity

Wegner, David L ENV086 Submission to the World Commission on Dams' Thematic Review On Dam and Ecosystems, regarding the Impacts of Dams on ecosystem sustainability and integrity

Williams, Philip[ B. ENV160 Reviving Living Rivers WWF-Australia ENV220 The Ecological Effects of Large Dams in Australia: A review of literature Yong-Woon, M ENV032 Tong River Preservation Campaign




Taluk, Uttara Kannada District Young, Terry & Fujita, Rod ENV055 How Healthy is Our Estuary? 13 Essential Indicators Tell All Countries from which Submissions were Received for TR II.1

Country Total Country Total Country Total Country Total Country Total Argentina 2 Costa Rica 4 Lesotho 3 Peru 1 Syria 1 Australia 10 France 4 Malaysia 1 Poland 1 Taiwan 2 Bangladesh 2 Ghana 2 Mauritania 1 Russia 1 Thailand 3 Belize 1 India 20 Mexico 1 Senegal 2 Turkey 1 Blank 15 Indonesia 1 Nepal 4 South Africa 4 UK 17 Brazil 13 Italy 4 Nigeria 1 South Korea 1 USA 23 Cambodia 1 Japan 5 North Africa 1 Spain 7 Vietnam 2 Canada 10 Jordan 1 Norway 1 Sri Lanka 3 W. Africa 1 Chile 1 Kenya 3 Pakistan 3 Sudan 1 Zimbabwe 1 China 4 Korea 2 Palestine 1 Sweden 2 Colombia 1 Lao PDR 2 Paraguay 1 Switzerland 5



Appendix III � Comments Received for Thematic Review II.1 Dams, Ecosystem Functions & Environmental Restoration The WCD is committed to an open and consultative process. To broaden the scope for participation and input from interested groups and stakeholders, the Commission invited specialists, centers of excellence and WCD Forum members to prepare comments on the thematic drafts. Comments were received throughout the progression of the thematic review. The comments were incorporated to the extent possible into subsequent drafts of the thematic. Every comment has been read carefully. Some are informed individual perspectives on which the WCD can not mediate. For example, there are some comments that seek the endorsement of the WCD, and the WCD�s mandate is neither to adjudicate nor to mediate on specific dams or disputes. Others may go beyond the scope of the individual thematic review. Please note that section numbers referred to in individual commentaries will have changed in the final version of the report. I: Comments on Circulation Draft of March 2000 a) Guy Lanza Director, Environmental Sciences Program, University of

Massachusetts, Amherst, USA b) Wulf Klohn Food and Agriculture Organization c) Himanshu Thakkar South Asia Network on Dams, Rivers and People d) Ikuko Morishita Institute of Freshwater Biology, Osaka, Japan e) Montri Suwanmontri Environmental Division, Electricity Generating Authority of

Thailand f) Patrick McCully &

collaborators International Rivers Network, USA

g) Yogi Carolsfeld Research Director, World Fisheries Trust, Canada

h) Gaetan Guertin ICOLD, Chairman of the environment committee i) Goran Ek &

Collaborators Swedish Society for Nature Conservation

j) Henk Saeijs Ministry of Transport, Public Works and Water Management, The Netherlands

k) Martin Perusse Senior Advisor - Environmental Issues Hydro Quebec, Canada l) Robert Dobias Asian Development Bank, Philippines m) Stuart Blanch Coordinator of the Inland Rivers Network, Australia n) Musonda Mumba Programme Officer, Freshwater Programme, WWF lnternational o) Tor Ziegler & Hans

Olav Ibrekk World Bank

p) Takehiro Nakamura United Nations Environment Programme a) Comments by Guy Lanza I recently received a draft of the WCD Thematic Review "Dams, Ecosystem Functions, and Environmental Restorations." The draft manuscript indicates that it "presents a synthesis of knowledge on the environmental impacts of large dams and incorporates inputs (reports and submissions) received by the WCD." I assume that statement refers to the selected and limited knowledge provided by " a series of reports commissioned directly by WCD, and through a co-operative agreement with UNEP." I would like to offer the following brief comments on the report: (1) In spite of the fact that the conclusions of the draft report clearly indicate that most impacts from



dam construction have negative effects on ecosystems (Page 80), the recommendations clearly favour a policy of continued global dam construction. (2) The increasing appropriation of water resources by large scale engineering schemes, including dams and their impoundments, represents a major threat to global water supplies and biodiversity. At a time when we desperately need a new paradigm of global water resource use premised on the conservation of water resources, water quality, and biodiversity in natural ecosystem settings, the WCD draft recommendations emphasise a policy to 'build better dams' and future fragment the already abused and fragile global hydrological cycle. (3) Although tropical terrestrial and freshwater ecosystems represent only 26 percent of the land surface of the globe, they generate almost 60 percent of the world's essential primary productivity, and support two-thirds of the identified species of vascular plants and more than two-thirds of the planet's total biodiversity. The severe damage to ecosystems from dam construction and operation , much of it irreversible, is well-documented and well-understood. And, the history of damage is clearly stated in many sections of the report body and in the report conclusions. I find the lack of correspondence between the report conclusions and the report recommendations to be a very serious flaw. If one of the main purposes of the theme report is - - "To identify good practices (including tools, methodologies and procedures) and options that could improve the decision-making process within the broader context of sustainable management of water and energy resources." -- then the report falls far short of being a useful document. b) Comments by Wulf Klohn Comments: The WCD Dams and Ecosystem Functions document provides a very good and timely introduction to the many aspects of IWRM regarding biological functions of the aquatic environment. The difficulty of the IWRM is highlighted by the fact, stated in the draft report, that to this date perhaps only 10% of all living organisms have been scientifically described. Progress in describing and understanding the workings of the ecosystems into which human activity is embedded is slow, also because science does not offer an attractive career for young people. At a time when major and irreversible human-induced changes are taking place, engineers and other practitioners have a duty to apply the precautionary principle and not to dismiss the dangers involved in taking decisions that have largely unknown consequences. I would recommend this report for reading to all colleagues. Comments to WCD can be brief and commendatory - good progress is being achieved. c) Comments by Himanshu Thakkar 1. This is, unlike many other WCD papers, remarkably open paper and inclusive of all perspectives

on large dams and their impact on ecosystems. The authors of the paper and WCD deserve to be congratulated for this excellent paper.

2. Some of the areas that the paper has neglected include: impacts of irrigation (brought about by dams) in command areas of irrigation projects (Only page 44 it is mentioned that "The impact of the irrigation infrastructure can in some cases be as great as the dam".), the impact on environment, people and communities due to submergence of lands, forests and rivers, brought about by dams (this is supposed to be looked at by other papers, but this paper needs to mention them), the impact of downstream impacts on people and communities, issue if dams are sustainable solutions for water and energy needs, the health impacts of dams in all its dimensions and seismologic & the geo-hydrological issues. It is true that some of these impacts are to be



looked into by other papers, but still a paper that looks at the Ecosystem impacts in its entirety would be expected to make a comprehensive statement on impacts of large dams. Many of the issues described here should be in table 3.3 (page 15), but are absent. Similarly, while summarising the impacts of dams on ecosystems on page 46, impacts like loss of forests, lands and rivers due to submergence, irrigation impacts of dams, etc. may also be mentioned. That irrigation canals can have very serious negative environmental impacts is well known.

3. One other area that such a paper would be expected to look into would be the issue of basin wide management issues. Many of the justifications of dams emanate from improper management of river basins. And hence solution to many of the problems would not be dams, as posed by supporters of dams, but proper management at basin wide level. (This again, it would be said, is to be looked into by other papers, but a paper looking at ecosystems impacts of dams in its entirety would be expected to state this.) One example is the issue of floods. Dams are proposed as flood control options in many cases. But many times the flood result due to improper ecosystem management. Hence, solution would not be dams, but proper management of ecosystems. This issue should be a part of the ecosystem paper.

4. One issue that needs to be clearly brought about is that the poorest people are worst affected (and many times the only people so affected) by ecosystems degradation as their livelihood, most of the times, directly depends on ecosystems. This is particularly so in countries like India, where inequities and poverty is much more wide spread. Thus, preservation of ecosystems should be of that much more importance in such countries, not only for the reasons stated in the paper but also because the degradation will have greatest impact on the poor people.

5. Thus, on page 80, first para of chapter 7, it should be mentioned that poorest people are most affected by dams.

6. The statement on page 24 that "As hydropower represents the cheapest and most easily activated form of peaking power�." needs qualifications as hydropower is seen to be cheapest only when crucial costs are not included in the cost calculations.

7. The statement on page 23 in section 3.6.1.1 that "In general discharge control resulting from the damming of rivers reduces flow variability downstream from the dam" is not always correct. The dams that divert water away from the river or its basin could in fact increase flow variability. Moreover, this is true only when dams are operated for the purpose of maintaining uniform or minimum flows downstream of rivers. But most dams are not operated for this purpose.

8. The next statement on page 23-24 that "Although for major floodplain rivers, dams may increase flood peaks it is normally the case that the magnitude and timing of flood peaks is reduced" is also not always correct. The magnitude of flood peaks may be reduced till a certain magnitude of flows (this will depend on available storage capacity and reservoir operation rules) beyond which floods cannot be moderated. In other cases, the floods downstream may actually increase due to wrong operation of dams. As far as duration of floods are concerned, there are cases (e.g. Hirakud dam in Orissa and Damodar dams in Bihar-Bengal in India, some of the few dams in India that had flood control as one of the objectives) where both duration and frequency of floods have increased.

9. Estuarine impacts discussed briefly on page 33-34 needs elaboration as they can be quite serious. For example, in case of upcoming Sardar Sarovar Project on Narmada river in western India, the dam is to totally cut off the non monsoon (8 months in a year) flows downstream of the dam (there is zero provision for downstream releases), seriously affecting the rich estuarine fisheries, most likely, destroying it completely. The estuarine fish production here is in access of 12,000 tons per annum, most of it comprising of species Hilsa and Tiger prawns, both commercial varieties with rich returns. But the over 10,000 fisherfolk families who depend on fisheries here have no idea (they have not been told anything about the dam) about how a dam is going to snatch away the livelihoods from them.

10. One of the conclusions of chapter 3, mentioned on page 45 that, "There is therefore no normative or standard approach to address ecosystem impacts and these have to be looked at on a case by case basis" could be a bit problematic and could be conveniently interpreted by dam supporting perspective. The statement may be qualified.

11. On page 47, while discussing valuation of ecosystems, the issue of valuation of livelihood losses may also be discussed. What price to put on livelihoods lost in situations where alternative



livelihoods are just not available? Similarly, the issue of cultural values, particularly for tribal (indigenous) communities.

12. It would be useful to know the source of statement on page 52 where it is said, "Nepal is also currently devising a national policy which involves set aside of particular river basins, or portions thereof, from hydropower development".

13. The statement on page 54 that, "This latter method has the added advantage of renewing the sediment load to the downstream channel�.". It may be clarified how this is an advantage.

14. On page 58, at the end of section 5.2 (Decommissioning and Restoration), it may be mentioned that what will be the fate of dam after its (finite) useful life gets over. Decommissioning costs and options should be part of dam proposals. This is even more relevant when paper notes that in many cases, decommissioning may not be viable option.

15. The Socio-economic impacts of dams mentioned in table 5.1 on page 63 should also include the impacts on upstream areas and in command areas.

16. Among the capacity constraints mentioned in section 5.3.2 on page 64, it may be mentioned that absence of existence of an institute for planning, monitoring and regulation and absence adequate legal system and civil society could be serious constraints.

17. While discussing "Why Mitigation can Fail" in section 5.3.3 on page 64, it may be mentioned that non dam options may not have been looked into even when they are known to exist. This may be due to lack of transparent, accountable institutional back up.

18. On page 65 it is stated (under "Adequate legal framework and compliance mechanisms") that "the contractual arrangement with the donor is the major means for ensuring compliance". But it needs to be noted that in absence of transparent, accountable system of compliance, this arrangement has generally not succeeded.

19. Regarding box on "Indicators of Ecological Integrity" on page 68, there is need for elaboration, particularly about the results of various steps mentioned, before this can be a convincing tool for Environmentally sound management of water resources.

20. The trends in international debate/ approach to dams described in chapter six is quite inadequate. The chapter gives elaborate space to the trends in IEA, ICOLD, World Bank, OECD and international conventions. But it does not adequately describe the moving force behind these trends, that is the movements world over that has questioned dam centric, engineering centric water resource development and have shown directions to alternatives. Due to this serious inadequacy, the chapter reaches a conclusion that "Therefore efforts to deal with environmental impacts of dams should concentrate on developing legitimate and accepted processes for dam planning, design and management within a river basin context. Secondly, much effort could be invested in improving the economic tools for analysis and improving incentives for better dam design and operation". This is very inadequate and inappropriate conclusion at the end of and due to an incomplete narration and analysis of trends.

21. On page 80, para 3, it is stated that, "there is today, widespread, but not complete, agreement as to the reality and importance of these impacts and their costs". It may as well be noted that this "agreement" has actually translated little into agreed/ compatible actions. Two paragraphs later, it is stated, "While there is experience of good mitigation�.". Here the word experience may be qualified by words such as: "some limited".

22. On page 81, at the end of para 7.2(2) it may be noted that credible, independent (of developers) and transparent assessment of impacts on biodiversity is first important step for this to succeed.

23. It should be mentioned in para 7.2 (3) on page 81 that the precautionary approach mentioned there should also be applied to ongoing and past projects (for their operation and decommissioning).

24. In para 7.2 (5) on page 82 it should be mentioned that participation of local communities is essential for this recommendation to succeed.

25. Among the measures to strengthen enforcement mentioned in para 7.2 (7) on page 82, transparent and participatory processes should also be mentioned.

26. In table 7.1, page 84, point 5, second bullet in right hand side column, it may be added that "if necessary, change operation of dams for this". In fifth bullet here, it may be added after "Ensure every dam" that "(including ongoing and existing dams)". Same words also need to be added before "lifetime" in last bullet of this point.



27. In Annex 3.2, page 92, it is mentioned that "Well-managed reservoir fisheries can be very productive". After this it may be mentioned that "but if not properly maintained/ operated, the fish production can also be very low and much below potential as is happening in most reservoirs, as in India.

28. The other important point here is the comparison between reservoir fisheries and fisheries loss due to dam. In many cases, the base line information about actual fisheries production from resources (river (upstream and downstream) and estuary) in absence of reservoir is generally not known. Where it is known, it is not taken note of. For example, in Narmada estuary average fish production is over 12,000 ton per annum. The SSP reservoir, 39,000 ha in area, even at peak production figure mentioned in Box A3.1 for India, that is 49.5 kg/ ha/ year (it would be useful to have specific reference for these figures), can produce no more than 2,000 ton per annum fish. Here the fish production (upstream and downstream) in absence of dam is not even accounted. Such losses should be taken into account, which is not done, as in the case of SSP.

29. The paper does not discuss the impact of dams on downstream mangrove forests and the role of mangrove forests in ecosystems. The beneficial role played by mangrove forests in ecosystems is known and it is also known that dams lead to destruction of mangroves. In a recent (1999) Orissa (eastern India) cyclone, it has been noted that the coastal areas that had mangrove forests intact had less damage due to cyclone compared to areas that did not have mangrove forests.

d) Comments by Ikuko Morishita Comments I consider the report as a whole presents an image of maintenance and conservation of natural environment as it ought to be against the situation of the world in a very organised manner and with a degree of elaboration that almost lets us wonder if it has really been prepared in a short period of one and half years. However, whereas the report describes the remedy for environment distinctively with expressions of avoidance, mitigation, compensation and restoration in its basic framework, I should say that a dam creates new environment at its site, so it is very seldom to consider restoring the environment afterward. In addition, it should be kept in mind that compensation (meaning creation of a similar environment) is frequently impossible when virgin forest or natural forest is involved. In this respect, avoidance and mitigation are considered the only major means for a dam. The natural and social characteristics and conditions of each country should be respected. In this respect, I propose the recommendations and options for operationalisation not be too restrictive. We have examples of good mitigation measures in Japan, while the paper gives very limited value to mitigation. In some cases, mitigation is only possible remedial option. I propose the role of mitigation be more appreciated in the paper. The followings are some further comments on individual texts. P-13, Table 3.1: It is necessary to consult original information sources for the descriptions concerning Japan. There are approximately 21,000 rivers in Japan which are managed by the State or prefectures, and approximately 35,000 if those managed by other municipalities are put together. It is, therefore, not clear where the number of 30,000 cited in the report comes from. I also wonder how the writer could have come to describe only two rivers had not been either dammed or modified in anyway. For your information, the total area of reservoirs in Japan is approximately 1,800 km2, corresponding to 0.5% of the total area of Japan. (Source: Dam Yearbook 2000)



P15 Table3.3: I do not agree with the way of impact variables classification, which is simply based on the interconnectedness nature of ecosystem functions. However, if I follow your classification, I would add in the "Second Order Impact" categories filter feeding invertebrates such as hydropsyched caddisflies and simulid black flies because they normally increase their numbers followed by impoundment of rivers. These filter feeders feed on plankton and other organic particle discharge from the upstream reservoirs. The colonies may dominate the benthic communities. Because caddisfly and black fly population emerge as adult during a short period in spring or summer, often the abundance of the flies cause nuisance to local communities. P16 Box3.3: In Japan, we conduct detail research not only on distribution of raptorial birds, but also on their habitat use, such as home range and foraging area which are important for population existence, to understand the structure of habitat. Then the structure of habitat is compared with the area affected by dam construction. When loss or fragmentation of the habitat is predicted, suitable conservation programs are conducted according to the predicted effects of the dam construction on the raptorial birds. The site of the Kurosagawa Sabo Dam Project in Japan was located near the nest of a golden eagle that is categorised as an endangered species in the red list. Because the construction work was performed avoiding its breeding season, it took five years to complete the project. There are, therefore, cases where construction work has been carried out in Japan considering the raptorial birds with their nest located in gorges etc. P19 L7: It is unbelievable that the sediment has reached almost 1/5 of its global storage capacity. The source book must be consulted again. In Japan, the amount of sediment in reservoirs with a total storage capacity of 1 million m3 comes to 1.25 billion m3 that is only about 7% of the total storage capacity of 17.71 billion m3, and there are few dams whose sediment exceeds the design sediment storage capacity (as of 1997). And the following corroborative experiments are conducted to resolve dam sedimentation problems in Japan. The reduction of sand and gravel flux by dams may cause degradation and armouring of river bed materials. As a result, water blooms have occurred and fish spawning sites have been effected. Countermeasures have been carried out at the Naka River from 1991 to 1995, where they dug 2,000 m3/year of gravel and sand at the Nagayasuguchi Dam Reservoir and placed them on two river shores 2 km downstream from the dam. At the upper site, trucks placed the gravel and sand directly, and at the lower site, the sand and gravel were pushed into the watercourse by bulldozers after they were placed. Some investigations have been carried out and the analysis is now underway. Some newspapers say, �sand�s sedimentation favourably received,� �we can see sand between the rocks and the fish are increasing,� or �there are some places where a lot of sand has accumulated.� P20 L5: We would like to add: �In Japan there are cases such as that mentioned below where red tides caused by concentrations of flagellatae also occurs in dam reservoirs where eutrophication is not very advanced, harming the scenic beauty of the sites.� Red tides are observed in the northern portion of Lake Biwa, Japan�s largest lake, which is less populated and not as eutrophicated as the southern portion of the lake. Red tides are seldom observed in southern Lake Biwa. P21 L9: In Japan, during winter through spring, diatoms such as Melosira sp. are common. However, in summer, commonly, blue-greens such as Homoeothrix sp. become dominant.



P23 L12�16 "Promoting (Kohler et al. 1986)": In Japan, among ecologists and biologists, exotic species are treated as organisms threatening existing native fauna. For example, black bass and blue gills were introducing into natural lakes and artificial lakes throughout Japan for recreational purposes. The damage on native fauna, ecological integrity of the system and fisheries production is enormous. P26 L2: In Japan�s rivers� whose natural flow rates fluctuate widely, downstream aquatic ecosystems are generally preserved by supplying maintenance water from dams during dry seasons. For example, at the Tonegawa River, at least 30 m3/s is supplied to a downstream river mouth weir, and at the Yodogawa River, water is supplied from a dam to guarantee a flow of 60 m3/s in the same way. P27 L8: Because the temperature problem with the dam discharge is caused by reservoir stratification, it is possible to almost resolve this problem by establishing a selective withdrawal facility. In Japan, selective withdrawal facilities have been adopted at many dams recently. And there are cases of old dams with water intakes located only at the bottom layer where selective withdrawal facilities have been added after the start of operation. P28 L30 Illinois River floodplain example: I suggest reviewing literature regarding flood pulse and river-floodplain concepts described in Sparks (1995), etc. River-floodplain is a species rich ecosystem. Many tropical forests are on floodplains as in the Amazon River basin. Large river-floodplains tend to have high biodiversity because in many cases, the system is mature, large, possesses complex habitat structures and is variable (Sparks 1995). For example, large river floodplains serve as important nursery and spawning grounds and refuges during droughts for many species of fish. From a human perspective, flood pulses in natural rivers have been important because they carry organic matter and nutrients into the floodplains that become fertile and enhance biological productivity. Therefore, large river-floodplain integrity is maintained by flood pulses and river-floodplain connectivity (Junk et al, 1989). Bayley, P. B. 1995. Understanding large river-floodplain ecosystems. Bioscience. Vol.45.no.3: 153-158. Sparks, R. E. 1995. Need for ecosystem management of large rivers and their floodplains. Bioscience. Vol.45.no.3: 168-182. Junk, W. J., P. B. Bailey, and R. E. Sparks. 1989. The flood pulse concept in river-floodplain systems. Can. Spec. Publ. Fish. Aquatic. Sci. 106: 110-127. P29 L3 Mississippi Delta: Please refer to reference as follows: National Research Council. 1992. Restoration of Aquatic Ecosystems, Science, Technology, and Public Policy. National Academy Press. Washington, D.C. p.p.33 and 177. P33 L5�7: The source should be indicated. It is not clear that the case studies are carried out globally or not P44 Conclusion: Conclusions should be based on the discussions in the main text. It is rather strange that the conclusions discuss the rate of coastal erosion the cost incurred specifically, while there is no indication in the main text. Other examples are "sediment entrapment can reach 99% ...." and "the global impact of dams for the global water cycle are ...". �51: The basic framework is categorised as avoidance, mitigation, compensation, and restoration. We think that at dams, basically a new environment for the place is created, with restoration rarely considered.



And regarding compensation, it is difficult to provide compensation (creation of a new similar environment) in the case of a primeval forest or natural forest . For this reason, at dams the principal methods adopted should be avoidance or mitigation. P54 L21 ��.with changes in water quality.�: For example, current control system combining selective outlet and aeration circulation mitigates total water quality problems including water temperature, eutrophication, dissolved oxygen and turbidity in some reservoirs in Japan.. P58 L9: In a number of countries and districts, suitable sites for dam construction are limited to some extent by geographic, geologic, and environmental constraints. Therefore, redeveloping an existing dam while conforming to the public water demand is another effective method of protecting the ecosystem by avoiding constructing dams at new sites. In Japan, many efforts have been made to redevelop existing dams; the Shinmaruyama Dam Project and the Tsugaru Dam Project for example. The Shinmaruya Dam Project will increase the effective storage capacity to 105, 220 X 103 m3, which is about 2.7 times as large as that of the existing dam. The Tsugaru Dam Project will increase the effective storage capacity to 128,600 X 103 m3, which is about 3.9 times as large as that of the existing dam. These points should also be added to the description. P81L28: The statement, "wherever possible dams and their impacts should be avoided" sounds as if the utility of dams itself is denied. The expression should be changed. P80-85 Chapter 7 general: Such words as "any" and "every" are used abundantly in this chapter. If a standard which can be acceptable to all countries is intended, assertive expressions meaning "dams should" needs to be changed without exception in all cases because countries differ from each other in climate, natural features, topographical and social conditions, the present state of social capital, and legal systems. P83 Table 7.1 "Recognise the important role of natural ecosystem in ....": In environmental economics, it is considered that on the value assessment of function and role of natural ecosystem, an evaluation technique which is scientific and can meet an objective consensus has not yet been established. Consider the use of multi-standard decision-making that has no established technique for weighing different standards. P83 Table 7.1 "Recognise the importance of biodiversity and promote its conservation": In the statement, "dams should not negatively impact any Red Data Book species," the definition of negative impact is vague. Since most important subject for conservation is to continue existence of populations, the expression should be put emphasis on this point and be changed to: "dams should not negatively impact population viability of any Red Data Book species. Despite the statement, �dams should not be built in declared National Parks or Nature Reserves,� the standards of designating national parks and the like differ from country to country. Therefore, across the board prohibition cannot be supported. With reference to the statement, �dams on the main stem should be avoided,� there will be varied conditions depending on sizes and shapes of rivers. If the gradient of a river is high and its length is rather short as in Japan, it is believed that there are many cases where building dams in main stems is effective from the standpoint of flood control.



P84 Table 7.1 �4. Ensure effective participation ....�: Although it is stated that �all EIA studies should be public documents,� it should also be taken into consideration that as in the case of Japan, some documents are unable to be published for the protection of valuable living things. P84 Table 7.1 �5. Maximise adaptive capacity�: Addition of the description, �redevelopment of existing outworn dams should be discussed in some cases� is requested. P84 Table 7.1 "6. Promote incorporation of environmental management features into dam design": Although it is stated that �every dam on a river with migratory fish should have an effective fishpass and monitoring programme,� considering that some dams are unable to be provided with fishpasses for structural reasons (topography, height, etc.) and that there still is a question about the downstream migration of fishes through the fishpass for a high dam, it cannot be supported that this applies to all dams uniformly. As a measure to preserve water quality, other means than variable level off takes is conceivable. Therefore, to make it a duty is not supported. P84 Table 7.1 "7. Promote the development of national legislative framework": The modificative phrase, �except cases of redeveloping� is requested to be added to �ensure the developer/owner...lifetime.� e) Comments by Montri Suwanmontri Here are some comments: 1) The construction of storage dam with man-made lake not only provide negative impacts to biodiversity but also creates the new man-made ecosystem that sometimes benefit to biodiversity such as more birds and amphibians and some new species of drawdown-area plants. This issue should be incorporated. 2) Some policy recommendation (Table 7.1 Item 2)is likely a close-door policy that sometime may not fair to developing countries to develop the nations which primarily rely on water resources development. The report should address or discuss about the appropriate options and ways for development and conservation as well. 3) How far should we go for project/sector EIA on biodiversity assessment aspect? Should it address the relationships of the sun, the moon and flora & fauna with a project? How many experts on this issue available on planet? This should also be discussed in the report. f) Comments by Patrick McCully and Collaborators Patrick McCully, International Rivers Network with extensive help from David L Wegner, Ecosystem Management International, Inc., and Philip Williams, Philip Williams and Associates.



The authors have marshalled together a large amount of material on the impacts of dams and set it within a sound analytical framework. They set out a very useful set of recommendations. They have also made good use of submissions made to the WCD. There are however several important areas of weakness which are outlined below. A key part of the document which is missing is the Executive Summary. This will set the tone for the entire document (and may be the only part many people will read) so it is vital than the Executive Summary properly represents the rest of the report. General comments are: • the paper needs to assess the impacts of infrastructure and socio-economic changes associated

with dams, in particular irrigation schemes, as well as resettlement sites and informal settlement and deforestation induced by dam development ;

• the authors have a tendency to over-emphasise uncertainties on the nature of environmental

impacts while ignoring the preponderance of evidence; • there is a need to stress the fact that dams are simplifying riverine ecosystems by replacing a large

diversity of dynamic riverine ecosystems world-wide with more homogenous and stable reservoir and regulated river ecosystems. One impact of this is that "weed species" such as rainbow trout and carp have greatly expanded at the expense of native fish species with smaller ecological niches;

• the paper needs to stress that in many parts of the world healthy ecosystems are vital for the

livelihoods of people from the most economically and politically marginalized sections of society; • the section on decommissioning is extremely short and weak in content. Material on

decommissioning which has been sent to the WCD by IRN should be used to expand this section. The link between decommissioning and river restoration in particular needs expanding. The subject of decommissioning is dealt with much better in the thematic on Operations, Monitoring and Decommissioning and some of the material from that paper should be used here;

• a clearer distinction should be made between findings and recommendations on the planning,

monitoring and effectiveness of mitigation measures for new projects, and those on restoration measures for already completed projects.

• estuarine and offshore impacts are poorly treated. Reference should be made, for example, to how

Iron Gates Dam has changed the chemistry of the Black Sea with consequent increases in toxic algal blooms (see Humborg et al (1999) Silicon retention in river basins: far-reaching effects on biogeochemistry and aquatic food webs in coastal marine environments. Ambio 28(7); Humborg et al (1997) Effect of Danube River dam on Black Sea biogeochemistry and ecosystem structure. Nature 386, 385-388); Milliman (1997) Blessed dams or damned dams. Nature 386, 325-327) Also the extensive writings of Rozengurt on estuarine impacts.

• the paper is weak on global impacts eg extent of land flooded world-wide, and impact on global

sediment flows. The important study of the extent of fragmentation and regulation of N. hemisphere rivers by Dynesius and Nilsson is mentioned in Table 3.1 but should also be noted in the text.

• more attention should be given to the need to assess cumulative impacts in assessments of new

dams and operations of existing projects;



• Another area of weakness is the lack of discussion of inter-basin transfers (see eg. �An Analysis of the Effects of Inter-Basin Water Transfers in Relation to the New Water Law�, Kate Snaddon and Bryan Davies, Freshwater Research Unit, Department of Zoology, University of Cape Town);

• Mention should be made • a glossary of terms would be helpful. One terminological problem is that reservoirs are variously

referred to as "reservoirs", "artificial lakes", "man-made lakes", and "lakes" (eg first para p.18 "lake level fluctuations may be much larger than are normal in a natural lake" should be "reservoir fluctuations may be much larger than are normal in a lake"). As the text states there are distinct differences between lakes and reservoirs and it would be best to refer to "reservoirs" as reservoirs throughout the text.

Specific Comments: Section 2: River Basin Ecosystems and Biodiversity The section should have an overview of global status and trends in river basin ecosystem and biodiversity conservation - eg % of discharge affected by dams, trends in watershed condition (deforestation, urbanisation etc.), trends in wetland loss, trends in sediment flows, trends in biodiversity, trends in pollution, trends in river and estuarine fisheries, % of land flooded by reservoirs etc. Some of this information is later in section 3.7 on consequence of dams for species diversity, but an overview on this section would be helpful in showing interrelation between dam and non-dam impacts and also stressing need to protect riverine ecosystems in general. 2.3 The first para should address the importance of healthy riverine ecosystems to the large number of people who depend directly upon them for their livelihoods. This is particularly important as these people tend to be among the most marginalized in society in economic and political terms. The statement that "The central issue of river basin development is to decide how to allocate water to maximise the benefits it provides to society as a whole." should note that for reasons of equity the livelihood benefits of the poorest sections should be prioritised over benefits for richer sections of society. This statement is also unduly utilitarian - RBD should also seek to sustain healthy riverine ecosystems because these have an intrinsic value in themselves. Section 3: Ecosystem Impacts of Large Dams Box 3.1 The example shows how dams can benefit alien species to the detriment of natives. This should be explained in the heading for the box. 3.2 The statement "This complexity makes it difficult to generalise about the impacts of dams on ecosystems" should be reworded. There are a number of characteristics of dam impacts which are consistent and can be generalised. Box 3.2. The typology of dams is confused - eg it is not what is the difference between a barrage and a run-of-river dam (I would argue that they're essentially the same), and a run-of-river diversion does not necessarily divert water through turbines. A better classification would be 1) storage 2) RoR and 3) diversion. "ICOLD recognises a large dam as one that is greater than 15m high"



should read "ICOLD recognises a large dam as one that is higher than 15m. Dams between 10-15m may also be defined as "large" depending on other parameters such as reservoir size or crest length." Table 3.1 The statement that there are more than 10,000 major reservoirs in Europe should explain how "major reservoir" is defined. 3.3 Should discuss the river continuum concept. Box 3.3 Construction Impacts: This box is more about transmission line impacts than construction impacts. The material on transmission lines should be included in a section on impacts of infrastructure and land use changes associated with dams. This would also include impacts of resettlement, increased migration and irrigation schemes. 3.4: The note on consultants reports should state how these reports are often written as pro-project advocacy documents by consultants linked to or hired by project developers and therefore of limited use as scientific studies. 3.5: The paper overstates the similarity between older reservoirs and lakes. There are many differences in behaviour as described elsewhere in this section. 3.5.1.1 "In particular temperature drives primary productivity" should be "In particular temperature impacts primary productivity along with nutrient dynamics, and seasonal availability of minerals and light conditions." 3.5.1.2 "Many reservoirs store almost . . . " should read "Many reservoirs capture almost . . . " The sentence on Glen Canyon should note that the sediment trapped equals approximately 95% of pre-dam sediment flows. The type of outlets on the dam also impacts reservoir trap efficiency. Are degraded watersheds also a reason for high sediment loads in N. Africa? 3.5.1.3: "The size of the dam," should read "The size of the reservoir". "Major biologically-induced changes" should read "Major biologically-driven changes". "(but not always)" is tautological. "particularly phosphorus" should read " particularly phosphorus and nitrogen" "Eutrophication can result in" should read " Eutrophication and nitrogen pulses can result in" "can cause oxygen depletion" should read "can cause additional oxygen depletion" "Mercury contamination" should read "Mercury and other heavy metal contamination"



3.5.2.1: 1st para, 3rd & 4th sentences should read: "The introduction of a reservoir into a river system, can markedly alter its primary productivity. The hydrological . . . regimes of reservoirs are unique, so the . . . highly site and watershed specific." The final sentence should be deleted. 2nd para 1st, 2nd and 3rd sentences should read: "Upon dam closure, the river (lentic) system resets itself as the reservoir fills. Usually, a microbial . . . nutrients as the . . . matter begins decomposition.. This stimulates a rapid development of phytoplankton." 3rd para, 1st sentence add "geographical location AND WATERSHED INPUTS". Last sentence "productivity of tropical lakes is limited by the introduction of highly turbid waters, wind-induced turbulence during the wet season, and chemical stratification". 3.5.2.2: Should include role of macrophytes in increasing evaporation and greenhouse gas emissions, impeding navigation, and having negative biodiversity and fisheries impacts. The second para in Box 3.4 on invasive species concerns macrophytes and should be pasted into this section. "support for disease vectors" should read "habitat for disease vectors" 3.5.3 The characterisation of the Pehuenche project is based solely on World Bank literature on the dam. A less rosy picture is given by the IDB, the project�s other main funder in �Synthesis Report: Environmental Regulation and Supervision of Infrastructure Investments, Office of Evaluation and Oversight, June 1999� (Chris Clarke at WCD has a copy of the relevant section). This notes, for example, that for 78% of the time the project dries up 12km of one river and 5km of another and that this has led to ongoing conflicts with downstream water users. 3.5.3.1 Needs further explanation of dangers of introducing exotics. 5th para, 1st sentence should read "During . . . an initial increase . . plant biomass which results in a pulse of nutrients". The following sentence should be added at end of this para "Nor should it be assumed that reservoir fish biomass will exceed pre-dam river system biomass." 3.6 Needs to stress the ecological importance of floods and the dynamic variability of the river. Also needs to mention the importance of the primary downstream food base, the macroinvertebrates which may be severely impacted by dams. 3.6.1.1 Dams may increase flood peaks on any river, not just major floodplain rivers. 2nd para delete "may in some cases" p25. An explanation is needed of "increased transmission losses downstream" 3.6.1.2 1st para "Reservoirs act as thermal AND CHEMICAL regulators" The statement that "seasonal and short-term fluctuations in water quality are regulated" is misleading as reservoir releases can cause pulses of poor water quality. "Seasonal and short-term fluctuations in water quality are altered" might be more accurate. "irrigation streams" should just read "irrigation". 4th para "reservoir will be cold" should read "reservoir often is cold" p27, 1st sentence "resolved" should read "dissipated"



3.6.1.3 "Sedimentation can degrade habitat downstream". The meaning here is unclear. Eg does it refer to lack of sediment downstream due to reservoir sediment trapping? 3.6.1.4 1st para "controlling floods" should read "reducing floods" p28 1st bullet should read "Reduced sediment transport can result in lowering of the riverbed downstream . . . " Armouring of the river bed should be added to these examples. Channel Erosion: Degradation impacts may extend far beyond "tens of kms". Coastal Deltas: Loss of coastal wetlands should be stressed (eg Mississippi). There seems to be a typo in the Rhone example (from 12 million tons to 12 million tons). 3.6.2.1 Should mention that algal assemblages will change in regulated rivers. Second point should read "by augmenting or decreasing the supply of plankton" Plankton pulses are also linked to nutrient supply. 3.6.2.2 3rd para "composition" should be "frequency" (?) 3.6.2.3: note should be made of the important study of dam impacts on downstream riparian vegetation in Nilsson et al (1997) Long-Term Responses of River-Margin Vegetation to Water-Level Regulation. Science 276, 2 May. 3.6.3.1 4th para: "Large woody debris . . . role in providing fish AND FOOD BASE habitat". 6th para: "fly in to drink AND FEED ON EMERGING INSECTS" 7th para: "fish biodiversity AND SURVIVAL" 8th para: "spawning, feeding, and juvenile rearing." 9th para: what is citation for the 66 case studies? What types of positive and negative impacts occurred? 11th para: citation for Senegal example? Migratory fishes have already been blocked on many Asian rivers - this is not just a problem for the future. Sentence here on stream dewatering should be expanded and moved into a separate para - this is an important issue. 3.7.2: 6th para: Small dams can block migrations as well as small ones. 3.7.3: 3rd para: citation required. 4th para: how are "major reservoirs" defined? 3.8: the comment on "several dams" understates the issue - it may be dozens or even hundreds of dams. 6th para notes the major impacts of irrigation infrastructure. Coverage of this issue should be greatly expanded. 3.9: 1st para: 1st sentence should state "ADVERSE impact". Dams do not form a "natural" resource base.



5th para: it is incorrect to state that there is "no normative or standard approach to address ecosystem impacts . . . ". There are for example standard approaches based on physical processes. Section 4. Economic and Social Implications "disruption of ecosystem processes" should be added to the list of ecosystem changes. 4.4: This section should note that those directly dependent on riverine ecosystems for their livelihoods tend to be among the most economically and politically marginalized sectors of society. It should also note the need to negotiate the consent of affected communities before their resources are expropriated. 3rd sentence: should note that these costs may only become visible many years after the project is built. 4.5: This section should recognise the role of corruption and political and economic vested interests in promoting dam construction (i.e. the forces promoting dam construction are not just related to values). Box 4.1 should be titled "Ethical principles for decision-makers involved in water and energy planning" These are a good set of principals. Some other internationally recognised rights which should be added to this box are the right to livelihood, the right to a healthy environment, and the distinct rights of indigenous and tribal peoples. The Box should also mention respect for civil rights. 5 Responding to the Ecosystem Impacts A frequently promoted mitigation measure which is not covered here but which gets a lot of publicity are animal rescue operations during reservoir filling. This issue should be discussed here with case studies. (A brief mention of the issue is made in 3.5.3.1 and it is covered in a little more detail in annex 5.1). 5.1: 2nd para: The final sentence needs edited to make sense. 3rd para: should read "reviews the METHODOLOGIES" 5.2.1: This section should attempt to identify where the mitigation measures described have actually been carried out and with what level of success. If the evidence for this is not available it should be stated. 2nd para, 1st sentence: Another "obvious option" is to find non-dam options to meet the identified needs. 5.2.2: 1st para, 2nd sentence "mitigation measures ATTEMPT TO prevent the . . . " and "mitigation measures ATTEMPT TO rectify continuing . . . " 1st para, last sentence is unclear in meaning and should be deleted. 7th para: encouraging agriculture in the draw-down may increase rather than minimise erosion. 9th para: the changes in physical and chemical properties may be MINIMIZED but are very rarely likely to be "resolved".



10th para: have measures ever been taken to remove nutrients from reservoirs? 12th para: needs more explanation of viability of measures such as dredging and to give examples where these sediment management techniques have been applied and under what circumstances they may work. 13th para: grossly exaggerates the viability of dredging. "not always financially viable to mitigate by dredging" should be replaced with "is very rarely financially viable to mitigate by dredging". The phrase "a fatalistic assessment" should be deleted. A para should be added to discuss measures to ensure downstream fish passage. 5.2.3: 3rd para: statement on Nam Theun 2 should recognise that the area supposed to be protected if the dam is built was already to have been protected under a GEF project which was dropped because of the proposed dam. 4th para: more space should be given to the failure of hatcheries as mitigation measures, particularly in the US West where this is a major issue in fisheries management. 5.2.4: A separate section (5.4) should be added on river restoration. There should be a greatly expanded section on decommissioning within this section which should include case studies of successful decommissioning cases and dams proposed for decommissioning. Mention should be made of the need to establish decommissioning funds for new and existing dams. Safety reasons for decommissioning dams should also be explained. IRN has submitted to the WCD a large amount of material on this issue which should be used here. 2nd para: decisions on decommissioning not just due to changing values but also ageing of dams, safety concerns, and increased awareness of decommissioning successes. 4th para should mention strategies other than removal for dealing with sediment, eg stabilisation in place or gradual erosion. It should also mention strategies for dealing with sediment contamination eg removal or stabilisation. 5th para: the statement that "restoration may not be an environmentally acceptable option" is nonsensical. No matter how bad a condition the catchment is in, there will surely always be scope for restoration, and possibly more scope the worse the catchment is. The final sentence is wrongheaded and should be deleted. Decommissioning has been shown to be viable in practice not just "in theory". The fact that the issues are complex does not mean that decommissioning is not viable. 5.3 The IEA reference is not credible. How many of these "successful" mitigation measures were reported by IEA members (dam operators, funders etc.) and how many by independent scientists? This example also does not explain the significance of the impacts mitigated (eg mitigation of oil spills during construction vs. mitigation of long-term fisheries impacts). Another study which should be cited here is the "World Survey on Environmental Management Practice" (International Water Power and Dam Construction, May 1991). Out of 31 national dam agencies more than 60% stated they had no formal system for monitoring dam impacts. Without monitoring, how can it be known whether or not mitigation is successful? 2nd para: surely that fact that the environmental clauses were "very modest" should have made them easier to comply with?



6th para: the wording of the final sentence is biased and implies that environmental groups opinions are unreasonable. The para should discuss how effectively the "common tools" discussed have worked. 5.3.1: The section should also refer to the problems of ensuring downstream fish migration (it is a common misconception that it is upstream rather than downstream passage which is the major problem reducing salmonid migrations in the Western US). The para on Pak Mun should refer to the WCD case study. 6th para: hatcheries have failed to mitigate fish impacts in the US. What is evidence for their success in Brazil? 5.3.2: 1st para: mitigation is also difficult because of the inherent nature of the impacts of dam technology on riverine ecosystems. Many impacts are technically impossible to mitigate if dams are to provide their planned outputs. 5.3.3.2: The lack of political will to implement mitigation measures should also be discussed. 5.3.3: This section is wrongly titled. It should be something like How to Make Mitigation More Effective. A good list of conditions is given. 1st para: "is not a foregone conclusion" should be reworded to "is unlikely". 2nd para: "is less likely to be successful" should read "is very unlikely to be successful". 3rd para: more supervision and monitoring make little difference in the absence of the recommendation of appropriate actions and the will and ability to ensure compliance. For an example of this see the numerous World Bank monitoring missions to the Sardar Sarovar Project in India. 7th para: "some countries" should read "many countries" 5.4.1: The last 2 sentences of the 1st para repeat those in the 1st para of 3.5.3.1. My comments on the Pehuenche project given above are thus also valid here. 2nd para: More info should be given on how these indicators were used in Colombia and Brazil and what were the results. 5.4.2: It would be useful to have some examples of Ecological Integrity Indicators. The present text gives only an overview of the concept. Box 5.8: shouldn't this be 2m3/s-1 rather than 2mm3/s-1 ? 5.5 Final para, 2nd sentence should read: "The review therefore concludes that it is very unlikely that all impacts can be mitigated for any project. While some mitigation measures may succeed in practice and others may be possible in principle, under actual political, economic and institutional conditions mitigation measures will rarely be successfully implemented." The final sentence should read: "There should therefore be a strategy of avoidance and minimisation . . . . future." 6 Trends in the International Debate



6.1 2nd para: It should be added that dam opponents also argue that better options than dams are available. 6.2: 2nd para: change "improving livelihood security" to "sustaining livelihoods" Table 6.1: This is a good distillation of arguments. In the seventh box down is should be noted that dams are also not clean because they contaminate rivers and destroy riverine ecosystems. 6.3: This gives only the positions of dam promoters. The positions of dam critics should also be given (eg the Curitiba Declaration). 6.3.3: 1st para: It should be noted that the Bank has never been shown to have complied with its requirement that its EAs reflect the views of affected people. 6.4: it should be noted that the progress noted in policy statement has not been reflected in practice (cf ongoing and proposed projects such as Three Gorges, Sardar Sarovar, Maheshwar, San Roque, Ilisu etc. etc). 3rd bullet: "physical area" rather than "space allocation" penultimate para: Another important area where improved techniques are needed is in cumulative impact assessments. final 2 paras: Both of these are overly focused on dams - the focus should be on water and energy planning in general. 7. Conclusions and Recommendations 7.1: 5th para: "mitigation is particularly problematic AND USUALLY INEFFECTIVE". It should also be noted that compensation in problematic. 7.2 These are very well thought out recommendations. Recommendation 5: Should include language on participatory monitoring and mechanisms to ensure compliance with recommendations on dam operation from monitoring bodies. Recommendation 7: Should include language on the need for developers to set aside decommissioning funds. Recommendation 8: Reword 2nd sentence to: "Blanket minimum flow requirements such as "10% minimum flow" do not address the needs of riverine ecosystems." Taking account of the dynamic nature of rivers requires optimum flows, including periodic controlled flood flows. Language should be added on the use of retrofitting dams, changing dam operation and decommissioning dams as tools for fostering ecosystem health. Table 7.1: The options here are also well thought out. Recommendation 1: Should include an option on the need to improve ability to predict cumulative impacts and to make cumulative impact assessments obligatory for new projects and review of operations of existing projects. Recommendation 2: The option to avoid main stem dams is important and should be given more visibility by moving it into section 7.2. Such an option would make a major difference in reducing the negative impacts of dams.



The final option should read "should not be reduced during commissioning to zero or levels likely to harm biodiversity" Recommendation 4: Add option for developers to be forced to provide funds to communities to do their own EIAs and review developer EIAs. Joint community-developer EIAs are another option. Recommendation 5: 1st bullet should be reworded to: "Provide mechanisms and financial support for regular monitoring . . . met. Monitoring should be transparent and mechanisms should be established to ensure dam operators comply with recommended changes in dam operation. 3rd bullet: "Revise operating rules" should be "review operating rules". Additional bullet: "Establish a fund from dam revenues for mitigation and dam decommissioning". Annex 3.2 Reservoir Fisheries This annex is highly one-sided and ignores numerous important issues. These include: - the drop in fishery yields after initial filling. - the frequent overestimation of reservoir fisheries by project proponents (eg Pak Mun and Aslantas case studies). - the difficulties for river fishers to adapt to reservoir fishing techniques and afford reservoir fishing technology (and consequence that reservoir fishery benefits are often captured by migrants, eg at Kariba and Akosombo). - the need to compare reservoir yields not just with pre-impoundment yields in the flooded stretch of river but also with the post-impoundment change in fisheries in the watershed as a whole, including estuarine and near-shore areas. - the systematic underestimation of yields from rivers and wetlands compared to reservoir yields due to the more commercial nature of reservoir fisheries compared with the more subsistence oriented river and wetland fisheries. - exotic introductions are generally of a small number of species world-wide. As these invariably displace diverse wild fisheries they lead to a reduction of global biodiversity. Annex 3.3. Comparison of Pre vs. Post Impoundment Conditions It is not clear what the point of this annex is. The comparisons do not seem particularly relevant as they appear to compare the flooded stretch of river with the reservoir and do not account for the impacts on the river up- and downstream. Annex 3.4: Sediment Discharges Much of this material is important and should be moved into the main text, in particular the section on Cumulative Impact Analysis Annex 5.1 Env Impacts and Mitigation Options The material on 'Wildlife Rescue' and Cumulative Impact Assessment should be moved into the main text.



g) Comments by Yogi Carolsfeld The items that I feel important to address are these: Biodiversity Maintaining natural biodiversity is currently recognised as central to ensuring sustainable aquatic resources, as mentioned in several places in the review. However, two components I think are being missed: a) biodiversity has an important historical element to it: i.e. the statement that �ecosystem functioning is guided by abiotic steering variables� (p. 15) appears to be true up to the trophic level only, but is not really true in terms of the specific biotic composition of ecosystems. It is very much this specific composition that determines how an ecosystem responds to disturbances b) the original biodiversity or ecosystems are not necessarily retrieved simply by repairing habitat (corollary of (a)). Both items are alluded to in the text, but I think not clearly enough nor with sufficient emphasis. I think the ramifications of these issues are at the core of the controversy between biological uncertainty and engineering precision discussed in section 5.3.2.1, and not adding enough emphasis to them are what keeps biodiversity at the �second table� despite all kinds of rhetoric to the contrary. Biodiversity is a distinct element that has resulted from many historical and present interactions; it is not an unresolved engineering issue related to today�s physical world. Specifically: Biodiversity should be included as one of the KEY indicators in assessing impact (Box 5.5 and section 7.2 (8) III). Leaving it out at this level or relegating it to a secondary role as in Box 5.5 has already created considerable problems in getting the biodiversity issue on the development table. This has occurred specifically in Brazil, for example, where despite many advances in water management, legislation and biodiversity conservation, aquatic biodiversity in practice still remains an afterthought in discussions of development projects. It also still appears to be a problem in the World Bank, as members of their environmental group will no doubt attest, and it is pretty clear that biology will continue to be relegated to the second level until it is included specifically in such things as these key indicators. Exotic Species: The introduction of non-native ("exotic") species is mentioned several times as a means to improve the aquatic productivity of modified reservoir ecosystems and cold raceways (e.g. p. 57 in section 5.2.3 and p. 93 of Annex 3.2), including the statement on p. 33 that this introduction is �required�. There are, of course, very large risks associated with the introduction of species, which are also discussed in the review, but one of the biggest risks not mentioned is that the effects are largely IRREVERSIBLE, and may not be acceptable to future generations. There are many examples globally of introductions that have gone awry, but it is also a hotly argued topic in many parts of the world, and portions of this review could well be cited out of context in numerous situations. I don't think that the introduction of "exotics" should be described as �required� anywhere in the review, though the review of their use is appropriate. However, in ALL cases the risk component, including irreversibility, should also be referred to, particularly as the precautionary approach to development is strongly promoted elsewhere in the review in relation to dam planning (p. 81).



Annex 3.2 indicates that the highest reservoir fish production comes from exotic fish in Brazil and Panama. In fact, I believe that the most productive hydroelectric reservoirs in Brazil rely on native fish species that also bring a higher local market value. In at least the case of the Sobradinho reservoir the introduction of exotic species was recommended, but rejected (Petrere, 1996), though one of the species fished is an invader from elsewhere in Brazil. Some irrigation reservoirs in the Northeast of the country are stocked with tilapia, but this practice is also now being questioned by many. The treatment of exotic introductions in this annex should be much more precautionary and specifically emphasise the likely irreversibility of such introductions. Effects on Fish Biodiversity A significant potential effect by dams on fish biodiversity not mentioned in the review is that natural obstacles to fish dispersion are flooded. In Angelo Agostinho�s publications, he describes the example of the Itaipu dam in Brazil that flooded Seite Quedas, a natural barrier to fish, and allowed at least sixteen species to invade the upper Parana river, which previously held a considerably unique fish fauna (Agostinho et al., 1995). Diversity of Dams Impacts The conclusions at the end of section 3 could be interpreted to say that all dams are different so their impacts cannot be predicted. While this may be true at a certain level, and is a very important point, the whole preceding section appears to be primarily devoted to describing impacts that are predictable. A list of these would be a good reinforcement of the final paragraph, for example: 1) assess importance of habitats to be impacted, 2) assess migratory value, and 3) retain stretches of natural habitat between dams. �Cultural� vs �information� services Goods provided by ecosystems include a category termed "information services" (e.g. pp. 5, 7, Table 2.1). I realise that this is a citation from other authors, but it strikes me as an unfortunate misnomer: these are actually "cultural services", with information being one of them. Mitigation Mitigation is discussed at length, but the conclusion to the section reads that in most situations it is ineffective or impractical and other options should be pursued for maintaining ecosystem health. Unfortunately, these options are not specified. The implication is that they should include avoidance, compensation or restoration, but these are unlikely practical solutions on their own. Participatory planning: Heading 4 of section 7.2 should read: "Ensure effective multisector participation in planning ...." Conclusions and recommendations: Conclusions and recommendations are naturally going to be key elements of the review. Wording is, of course very important, but the items must also be supported by the text of the review. Some suggestions that I have for table 7.1 of Section 7.3 (p. 83): Item 2 of the table: The Red Data Book appears to be presented as the final authority on endangered species in this table. As discussed on page 38, the book does not represent aquatic species very well. This should be clear in this table as well as in the text.



Item 3 of the table: baseline assessments of ecosystems: it should be emphasised in this recommendation that this assessment occurs over a long enough period to accurately reflect what happens in the ecosystem. Migratory species, for example, could easily be missed in a swat-team style of assessment, and surveys during a particularly dry year could be very misleading. Item 6 of the table: downstream passes or structures for fish should also be specified, as should the design for pulse flooding where appropriate. Item 7: the fourth point on fixed cost vs tendered process comes out of the blue and may be confusing to the reader. Presumably the authors mean that the money available for environmental work should be secure and applied in a manner that ensures quality work. If this is what is meant, it should be couched in these terms; the tendering part could be a parenthetical explanation for people that are already in the system, or this point could be explained more fully in the text of the review. Item 7: the "intact rivers policy" should be spelled out more in this table. Without question this table will often be read without the accompanying text and as thus should be able to stand on its own. Cumulative effects of dams is missed in these tables and reviews, other than through this reference to the "intact river policy". This is particularly important for migratory fish species and biodiversity and needs to be said again clearly here. Natural vs modified ecosystems Section 2 of the review appears to be largely dedicated to a description of natural ecosystems and their value, as distinct from disturbed ecosystems. This distinction comes out later in the review, but it is quite ambiguous in this section. The authors should make it clearer in this section when they are referring to natural ecosystems and when they are referring to all ecosystems (natural and disturbed). Cumulative effects of dams The cumulative effects of dams are particularly important to migratory freshwater fish that, as in South America, require migration in order to mature, and floodplains for juvenile growth. The example of the Itaipu dam, mentioned in Annex 3.2, is particularly telling: reservoir fisheries here are still quite productive, but only because there are still floodplains upstream (Agostinho et. al., 1994). Once the floodplains are permanently submerged by another dam, this fishery is likely to collapse as it has in other Brazilian rivers with multiple dams. This sort of scenario is being repeated over and over in Brazil, and probably also in the rest of the world, and should be more emphasised in the review: perhaps in the summary tables of Section 7 and by citing the Itaipu example in the main text on page 43 (section 3.8). Privatisation: The review states that most dams are associated with governments and that, as such, their function is generally meant for the public good. In actual fact, a significant proportion of the international hydroelectric industry is private; in Brazil, the majority of dams are presently being privatised. Globalisation of the economy is likely to carry this trend to many other parts of the world. A discussion of the actual, perceived and likely impacts of privatisation on how dams operate with respect to the environment is missing in the review, and could represent a big hole in how the WCD looks to the future. This could be particularly important given the discussion on low levels of compliance with regulations of international donors on p. 59. Gas Bubble disease The discussion on gas bubble disease and swim bladders on page 27 is wrong: swim bladder problems arise as fish pass through the turbine and are exposed to rapid pressure changes. This results in over-expansion of the swim bladder and frequently in death, but shows up only immediately at the dam. Gas bubble disease is independent of the swim bladder, resulting from supersaturation of the water



with dissolved air when spillways drop into deep pools. In this situation, the water is forced to hold more air than it should. The dissolved air passes into fish tissues, but then slowly starts coming out of solution, forming bubbles in a variety of the tissues. This is comparable to the bends of human divers. The problem can extend many miles downstream of the dam. Good references are:

Fidler, L.E. and Miller, S.B. 1997. British Columbia Water quality guidelines for the protection of aquatic biota from dissolved gas supersaturation: technical report. BC Ministry of Environment, Lands and Parks, Victoria, B.C.

Bouck, GR 1980. Etiology of gas bubble disease h) Comments by Gaetan Guertin This report is not a bad document, but it suffer from the authors bias towards a conservationist perspective that gives to untouched environment an higher value than anything else. The authors tends to forget that man made ecosystem are capable of responding to the same function and value as natural ecosystem. They also forget that human are part of the ecosystem and by that are interacting with it in a positive and a negative way. Considering the level of the actual population in the world, I am not sure that by relying only on natural environment, we could be able to satisfy mankind fundamental needs without the technology interventions.

Even though the report is well documented, a large part of the literature is not used. As for an example there is very few citations on the LaGrande Complex in Québec Canada were we have follow-up the environment for more than 20 years and published more than 350 articles.

At the beginning of the section on impacts of large dams, I was happy to read the authors comments about the difficulty to generalise because of the complexity of the issues (p.11 this complexity makes it difficult to generalise and p.12 need to treat each new dam separately), but in many case they have put aside that comment and did to many generalisation themselves. The following examples illustrate my point:

• Page 17 �The flooding of large areas in the tropics is especially likely to contribute to global species extinction�

• Page 19 �Many reservoirs store almost the entire sediment load supplied by the drainage basin.�

• Page 22 � Large-scale impoundments are likely to extinguish entire populations of species.�

Biodiversity Box 2.1. �Is a measure� in a region� The authors should be more specific about the term region, by reading their conclusion we understand that a region for them is only the flooded area by a reservoir. We agree with the principle of keeping as much biodiversity as possible but a geographical or ecological unit must be chosen before concluding. As an example many scientist use the watershed or the bioregion as the base unit for evaluation. The species concept must also be consider because many opponents to development use the concept of population and micro stock in biodiversity.

Ecosystem Values. I agree with the authors that ecosystems are valuable but they should consider that man made ecosystem (reservoir as a case) can perform the same function and have the same value or sometime more value than untouched environment.



Box 3.2 Types of dams. I do not agree with your class of order. For me, there are two categories: storage dams and run-of-river. A third category could include diversion: in-stream and cross-watershed. The ICOLD criteria to recognise a large dam also include the volume of the dam and the size of the reservoir.

For the databases you should consider that each ICOLD National Committee as its own register of dams, not only USA.

Box 3.3 Construction impacts. This box is an over simplification and reflect the bias tendency of the writers to overcharge the impacts. The width of the right-of-way varies with the tension of the line. Not all powerline are 735 kV and not all right-of-way are 100m wide. In much case we have multiple line in the same right-of-way. As for roads during construction and maintenance, in the southern part of the province we use existing roads and in the northern part the road are build in the right-of-way. We also have for the construction activities a list of mitigation measures and a code of conduct to reduce the impacts.

I am not an expert in sedimentology but I can say that this section is badly covered. The authors are always considering erosion and sedimentation as a negative situation. The sedimentation and erosion processes are natural and normal but they some time are accelerated or reduced by man�s interventions. Sedimentation in reservoir is not worse than in a lake or in an enlargement of a river. In many case the sedimentation issue is created by a bad watershed management and by human activities (agriculture, forestry, etc.) upstream of the reservoir. A recommendation on watershed management should be proposed.

In the discussion on water quality, you should use the word retention time instead of detention time. The monitoring we have done in LaGrande complex does not reflect your conclusions. It is the leaching of the vegetation and of the soil that creates the increase productivity and most of the plant biomass does not decompose. I was also surprised that you used (McCully, 1996) as a reference for mercury, not because the description is bad but because there is better experts than him on that issue. The book �Mercury in the Biogeochemical Cycle Natural Environments and Hydroelectric Reservoirs of Northern Québec� by Lucotte and all, Springer 1999, is a better reference.

Box 3.4 invasive species. This situation is not specific to reservoir it is also the case in natural environment, why emphasis more in reservoir?

Filling of reservoir does not always cause drowning of animal, in Québec we never observed that for large mammals. For the salvage operation the study done by EDF on Petit-Sault in French Guyana should be review by the authors for a better perspective.

I agree that small flood events may act as biological triggers for fish migration but extreme conditions are very stressful for the fish habitat. Following the Saguenay flood in 1998, salmon habitat had to be restored by man intervention.

On page 32, I do not understand the conclusion � Dams block these migrations to varying degrees� in the case of mayflies and stoneflies the adults are aerial.



In the section on species diversity, it does not mean that losses of habitat or of animals by construction activities or flooding will directly or automatically lead to loss of species.

The conclusions on the impacts of dams are very dramatics and very affirmative but the preceding demonstration are not as evident considering that not all existing documentation was used,

The same thing can be said about the conclusion on mitigation measures and the affirmation on page 81 concerning them is also very rigid and dogmatic considering the demonstration that was done. Since mitigation measures are site specific we cannot generalise about their effectiveness in other situation so categorically.

Section 6 is a good description of the present situation of the debate.

As for the recommendations and the options for operation even though I agree and fully support most of them, some of them surprises me, considering the lack of demonstration from the document. They are:

• Dams on the main stem should be avoided. Why it should be a case by case evaluation?

• Dams should be seen as a last resort rather�For many uses it is the contrary. In the case of electricity I would personally prefer hydroelectricity instead of thermal.

• Fish pass and Environmental flow are not always needed it should be site specific.

• Variable level intakes are not always the best solution for water quality issues why make it mandatory?

• Not only biologist should be include in the design team but also hydrologist, geomorphologist, etc.

I also attach a file of my presentation in La Hague march 2000 where there is recommendations and guidelines for better integration of dams in the environment.

i) Comments from the Swedish Society for Nature Conservation SSNC Review team Göran Eklöf, head International department of SSNC Göran Ek, international secretary at SSNC Per Isakson, freshwater biologist at SSNC Gopal Siwakoti, INHURED Chainarong Sretthachau, Southeast Asia Rivers Network(SEARIN) Jan Wallinder, SSNC network for sustainable large dams 1. General Comments on the report: 1.1. We find the review�s conclusions generally good and that the paper gives a frank and

honest description on how harmful large dams are to ecosystems 1.2 The report draws from a large number of references to present the environmental effects

of large dams in an extensive way



1.3 However we find that the report is not strong enough in making recommendations to

the WCD on some of these most crucial issues dealing with dams and their effects on ecosystems. We can ask for more guidelines and suggestions from the WCD, but, in our view, we do not lack policies and procedures and the knowledge about what has to be done from UN to the national level. The problem is their implementation, monitoring, compensation and punishment in case of losses and violations. The WCD must speak on these issues. It is an opportunity for all of us. We have to make sure that dams are not going to be built in the future unless all these issues and problems as mentioned in the report are fully addressed.

1.4. The report is not sufficiently concerned with the infrastructure issues on the

construction of large dams which also affect the non-riverine parts of the ecosystem. Particularly the impact of the irrigation systems, and the change of the land utilisation. According to the Thai NGO SEARIN in Thailand�s case, as well as other countries, the development of irrigation dams includes the construction of long irrigation canals, as well as roads along the canal, all on wetland ecosystems. So this sort of development creates environmental impacts that reach much further than the actual dam site or reservoir.

1.5 In a separate comment on this TR Himanshu Thakkar of SSNC�s partner South Asia

Network on Dams, Rivers and People has expressed that �One issue that needs to be clearly brought about is that the poorest people are worst affected (and many times the only people so affected) by ecosystems degradation as their livelihood, most of the times, directly depends on ecosystems. This is particularly so in countries like India, where inequities and poverty is much more wide spread. Thus, preservation of ecosystems should be of that much more importance in such countries, not only for the reasons stated in the paper but also because the degradation will have greatest impact on the poor people�.

SSNC endorses this comment.

. 2. SSNC´s comments on Conclusions and Recommendations 2.1 p.80 �In most cases one or more sector of society depends upon these (ecosystem)

values (e.g. fisheries, grazing) while in some the total value of the benefit of natural ecosystems can exceed the value of the benefits derived from dams and associated investments in agriculture�

This is especially true regarding the case on the dependence of very large groups of people in the South on functioning freshwater ecosystems for fishery, livelihood and subsequent good health (as is widely known, 80-85% of the protein intake of the people in the Mekong region is estimated to come from freshwater fish).

2.2 p.80 �The review has underlined that dams have a wide range of major impacts upon

natural ecosystems, that most of these are negative, that many are irreversible, and that they are manifest in economic and social costs. Perhaps surprisingly, the review has noted that there is today widespread, but not complete, agreement as to the reality and importance of these impacts and their costs. However, in the abovementioned separate comment on this TR Himanshu Thakkar has expressed that " It may as well be noted that this "agreement" has actually translated little into agreed/ compatible actions. Two paragraphs later, it is stated, "While there is experience of good mitigation�.". Here the word experience may be qualified by words such as: "some limited".



SSNC endorses this comment. 2.3 p.80 . �The review has also recognised the growing understanding of the threats to the

world�s biodiversity, and the particularly acute threats to those species that are dependent upon freshwater. By altering the quantity and quality of water available to natural riverine ecosystems dams add to these already significant threats.�

2.4 p.81-82 SSNC support all recommendations submitted. Regarding recommendation #

1, We wish to highlight, again, the connection between natural ecosystems and sustainable development This is especially true regarding the case on the dependence of very large groups of people in the South on functioning freshwater ecosystems for fishery, livelihood and subsequent good health. In the Mekong and Amazonas regions a number of large dams have been constructed in recent years which pose a serious threat to the sustainable development and livelihood for the rural populations here. We would rather have recommendation # 1 read as:

�Recognise the basic role of natural ecosystems in contributing to sustainable development.�

2.5 p.81-82 SSNC support all recommendations submitted. 2.6 p.83-84 SSNC also support the options for Operationalising (with one

exception, see 3.2) the submitted recommendations In particular the following - Assess the price of natural ecosystem functions and services during feasibility studies for both upstream and downstream values. Water does not have a zero opportunity cost. - Dams should be seen as a last resort rather than the first choice (precautionary principle) - Revise operating rules every 5 years to incorporate findings of monitoring programmes and mitigate unexpected ecosystem changes. - Ensure every dam has a proposal for how it will eventually be decommissioned especially with regard to design features for reservoir drainage, the treatment of accumulated sediment, and appropriate financial measures for ecosystem restoration - Include biologists in the design team But, we want to emphasise that in p.84 - Section 6 "Promote Incorporation�" The report states that every dam should be equipped with an effective fishpass; we feel that the technology of fishpasses and knowledge of the migration patterns of fish is limited. And in those terms, no fishpass is truly effective. So the recommendation does not quite relate to what is really practised. 3. Comments on views shared by SSNC 3.1 p.16 We appreciate the report�s highlighting construction impacts of dams on ecosystems as

they are seldom mentioned in this context 3.2 p.33. We find all the examples mentioned on loss of fish species diversity

and numbers very relevant and to the point. 3.3 p.38.We support your conclusion that the number of endangered freshwater fishes

is underestimated by IUCN



3.4 p. 44-45 Overall we find that you have made the correct conclusions from the data presented on the ecosystem impacts of large dams

3.5 p.46 Last paragraph. You are correct in highlighting the need for �those responsible for water

management need to be able to measure and value the ways that dams affect the environment. Clearly if the values of environmental impacts are incorporated into power/agricultural sector planning, they may well tip the balance and turn what appears to be an economically viable project into one that is likely to have a net economic cost� This is certainly true in the case of dams like Pak Mun and Theun Hinboun where over-optimistic prognosis of the economic output of the dam and chronic underestimation of the value of the ecosystem affected by the dams has caused both economical and ecological failures.

3.6 p.49. Paragraph 4.4. This is a very good summary of what has been sadly lacking in most of

the decision-making processes around large dams 3.7 p.50. Paragraph 4.6. See above! 3.8 p.51 Section 5.1. Thanks for stating so clearly that all impacts cannot be mitigated! 3.9 p.51-52 Section 5.2 This is the only way to rank responses to the negative impacts of

large dams 3.10 p.62 All chapter 5.3.2.1 provides an extensive overview of this sticky issue with a

remarkable insight. We especially support the views expressed in the 2nd-3rd paragraph, the salmon ladder in Pak Mun is a very well chosen example.

3.11 p.65 top and 5th paragraph. The sad story of the World Bank handling �completed�

dam projects comes to mind here 3.12 p.70-71. We support in full the conclusions expressed here especially the closing

statement �In addition the weaknesses of the EIA process for many projects (cf Thematic review V.2) reduce the possibilities for positive outcomes. This would tend to encourage a strategy of avoidance and minimisation rather than one of mitigation if the aim is to maintain biodiversity, and ecosystem functions and services for the foreseeable future.�

4. Critical comments on some views and findings 4.1 p.10 Box 3.1 :The patterns of change described in this section are in an unbalanced

(and unnatural) ecosystem, due to the dam and the beneficiary of the impact is an introduced species. Thus, we find this as an example of positive impact of a dam highly irrelevant.

4.2 p. 18 ; First Paragraph : The report makes the comparison that a reservoir, once completed, has a level of stability similar to that of a natural lake. We do not think that there is enough information to back up this statement, and we feel that an important point for this case is to make a comparison between pre- and post impoundment, within the same area. 4.3 p. 21 ; Box 3.4 Regarding the floating and submerged weeds described in the 2nd paragraph: The report mentions that the weeds pose a major threat to the efficiency of dams and irrigation systems, also as in Thailand's case (Kew Loom Dam), some reservoirs cannot be used for recreation and transportation because of the large number of floating weeds. Also, the treatment of the floating



weeds makes the water downstream contaminated with suspended solids, and it can damage the fishing gear and cause health problems for fishermen downstream (as in the case of Pak Mun Dam). 4.4 p.24 In the case of floodplains, the loss of wetlands not only results from the control of the water flow of the dam, but also from the dams which are built to prevent floods, which allow the people to occupy the wetlands for developing for agriculture and resettlement. This leads to floods downstream that are more severe than before the dam, and this is another reason for the loss of wetlands. Before the dams are built, water would flow into the flood plains and be stored there during the rainy season, before returning to the river. By storing precipitation and releasing runoff evenly, wetlands can diminish the destructive onslaught of flood crests downstream. But after dams are built upstream, villagers would move into the now-dry flood plains and modify the land (by agriculture and other development). Then, when the rainy season comes again, there is no protective flood plain to prevent severe flooding. 4.5 ; Pg. 32 Paragraph 5: In this section, there is not much mention of fish in the tropical rivers. The fish in tropical rivers are also migratory fish, but they differ from anadromous fish because they migrate within freshwater systems. In the case of the Mekong River, during the rainy season fish will migrate from the Great Lake (located in the lower Mekong) and Mekong mainstream up to its tributaries for spawning and feeding. The fish will return to the Mekong mainstream near the end of the rainy season, and in the summer season will feed and grow in the mainstream. So when dams are built, they block the migration pattern of fish both from the mainstream to the tributaries, and back. Many fish are killed by the dam turbines on the migration back to the mainstream. 4.6 p.33 2nd paragraph. It would be interesting to know where the case studies reporting positive impacts from dams on fisheries are made. Are they solely from Northern waters? In the case of Thailand, according to Mr Chainarong Srettachau of SEARIN, there has never been a study that shows any dam to have a positive impact on the fish. For example: In the case of Bhumipol Dam, fish species in the Ping River decreased 50% after the dam. Reference EGAT 1987. In the case of Khao Lam dam,59 fish species remained out of 78. And in Srinakarin dam, 20 fish species were lost. 4.7 Overall in section 3.7 there is a regrettable lack of referred studies on benthic fauna 4.8 p.55 3rd paragraph. Unfortunately we know of no current research neither any good examples- especially from the South supporting this statement: �Designs of fish passage facilities have evolved considerably through the years as scientists and engineers learn more of the requirements for encouraging fish to use the passage facilities, the specific hydraulic conditions that various fish species use to orient their migration, and the climbing capabilities of the target species or groups of species� 4.9 p.58 1st paragraph. Decommissioning costs and options should be part of dam proposals. This is even more relevant when paper notes that in many cases, decommissioning may not be viable option. 4.10 p.58 Section 5.3 1st paragraph �To attempt to answer this question (effectiveness of mitigation) we have drawn upon all reviews available to us. There is however only limited published information in the literature on how effective mitigation plans have been in meeting their objectives in developing and developed countries.� See comment 5.8 4.11 p. 60 Paragraph 4 : According to SEARIN, The information on Thailand is not correct. Referring to WCD Pak Mun Case Study, there are 265 fish species in the Mun River, not 109. . In the Case Study mentioned, it was stated that a vertical slot fishpass or a Denil fishpass would be more effective than a pool and weir fishpass. Also, the location of the existing fishpass is not optimal.



Same Paragraph : Regarding the older fishpass that the report states to be more effective, this was not at Pak Mun because there was no older fishpass at Pak Mun. The older fishpass mentioned in the report is the fishladder at Kwan Payao Dam, which was built 70 years ago. The numbers mentioned in the report (29 out of 33 species) are not accurate. The latest survey of the Ing River has identified 90 fish species, not 33 species. When we talk of the fish species which can pass the fish ladder, we also need to have the amount of fish that pass, not only the species. In the case of Pak Mun dam, only 2 species can pass a large number of fish through the ladder, while other species can only pass a few fish through. 4.12 P. 63 Table 5.1 under the socio-economic "To assess the number of people��" The table should include the people upstream, as well as in the reservoir area, not only those who live downstream. 4.14 Pg. 64 Section 5.3.3 According to SEARIN the report raises technical issues on the capacity constraints for conservation, mitigation etc. While these constraints do exist, it should be noted that in 3rd world countries the most major constraint is related to the lack of responsibility on the dambuilders part, as well as the failure to enforce mitigation measures, or the mitigation plan is never implemented, merely written out on paper. Also, the report takes a very 1st world viewpoint on such topics as the "lack of 4 wheel-drive vehicles and proper road networks" needed to access the rivers. In developing nations such as Thailand, the people don't think like that, and we hope that the report can take this into account. 4.15 p.66. 1st paragraph The Pangue dam may have drowned little wildlife but has evicted thousands of Pehuenche Indians from their homes. We wouldn�t call this dam �benign� 5. Recommendations from the review to be included in WCD´s final report Taking into account the comments and views expressed by us in chapter 1-4 above SSNC supports the suggested recommendations to the WCD of the review. j) Comments by Henk Saejis The time allowed me to assess the report was very short. It is quite possible that I have missed sections, or failed to notice some item. Please regard my remarks as a modest contribution to a good piece of work. I make my remarks from the perspective of: My experiences with the major hydraulic engineering projects that have been realised in The Netherlands in the 20th century; My experiences with the issues impinging on the Rhine riverbed area; Numerous visits to and recommendations for dam projects in Korea, Brazil, Egypt, Turkey, South Africa, Chile, Russia, Italy etc. General My overall impression of the report is: �a valuable piece of works with many facts, opinions and standpoints, most of which I share The report does however have the appearance of being a hastily assembled study document, in which I miss the connecting structure and any synthesis. The issues are presented in reasonable and sufficient detail, if you know where to look, while the conclusions and



recommendations are left hanging. The key message is largely present but it takes time and concentrated reading to find it. There really needs to be an extra level of explanation. I miss: main conclusions. Main conclusions in my opinion are: There is an urgent need for management plans at the level of entire river basins. Dams could play a role in it, but they are not the first essential matter of concern in relation to sustainable development. [In my opinion, the report of the World Commission for Dams concentrates its attention too strictly on reviewing the development effectiveness of dams and guidelines concerning �all aspects of dam building and utilisation�. Although these are very important subjects, there is an urgent need�] When damming parts of rivers the determining abiotic circumstances of the river basin shouldn't be altered beyond the natural fluctuations of the natural river basin. These natural flions caused by human activities, river basins having, as dynamic ecosystems, extensive. When the effects due to artificial reservoirs remain within the limits of this natural resilience no harm is done to the natural system. In that way irreversible effects on ecosystems, and following these the negative impacts on socio-economic aspects, are avoided. The challenge for each new project is to find out what the resilience is and how not to go beyond the limits of that resilience. (It is important to realise that the resilience for abiotic and biotic changes fluctuates in with the seasons.) The challenge for each existing project is to investigate what should be corrected in order to restore a situation in which the effects of the reservoirs stay within the limits of the resilience of the natural ecosystem of the river basin. This conclusion also provides a logical explanation of the concept sustainable use. Disturbing a river basin by human activity should never have a greater impact than the natural ranges and frequencies of disturbances that the natural river basin has to deal with. The word �natural� is important here: the disturbances caused by human abuse of the river basin in the past shouldn't be taken into account of course. Following this overall conclusion the complete flow of a river should never be dammed, because making a stagnant lake of a stretch of river clearly goes beyond the borders of the resilience of a natural river. Only parts of the flow should be used for reservoirs made in bypasses of the main stream. Dams may offer many advantages but as soon as they are on such a scale that the resilience of the natural system is eroded, they cease to be sustainable. So, consider recommending small-scale dams instead of large-scale dam-building In addition the following marginal notes Increased awareness of the natural environment and its endangered situation is one of the most important developments of the late twentieth century. The United Nations �Declaration on the Environment� and the Club of Rome�s message on the �Limits to Growth� left their mark on our thinking in 1972, followed in 1987 by immediate and world-wide agreement on the convincing concept of <<sustainable development>> as propagated in the Brundtland Report of the United Nations on <<Our Common Future>>. In 1992 the United Nations Conference on Environment and Development (UNCED) put the issue into a global perspective and drew up a comprehensive action program in Agenda 21. In my opinion initiatives such as the World Commission for Dams� may be considered in part as the (indirect) outcome of UNCED. The most important strategic principles formulated at this conference were: • Policy and management need an integrated approach at the level of an entire river basin.



• Management of water resources needs to be developed within a total package of policy measures on human health, production, protection and distribution of food, prevention and solution of accidental events, environmental protection and conservation of natural resources.

• Integrated water management is based on the awareness that water is an inextricable part of the ecosystem and that water is also a social and economic asset.

• Priority must be given to (1) fulfilling basic human needs and (2) at the same time protecting the earth�s ecosystems.

These four main recommendations should be constrains for your report. And for dam building In Europe the water policy of the European Commission is evolving from individual guidelines concerning different aspects (like water quality standards, pollution control, swimming water etc.) to a more integrated framework directive on water management. An important basic principle within the (draft) guideline is the organisation of water management at the level of an entire river basin. River basin authorities should propose and implement (not only with respect to water quality issues!) action programmes in order to solve the problems. Five important steps should be undertaken: 1. As a first step, a clear description should be made of the state of the art in a river basin. What are

the general features, what are the ecological and economical characteristics, what are the human interests? What are the borders of the natural resilience of the river basin? What sort of dams and other infrastructure are present in the river basin.

2. Secondly, a problem definition should be made at the level of an entire river basin, including five major area of concern: protection from flooding (1), transport (2), energy demand (3), water availability and distribution (4) and maintenance of biodiversity and ecological services. The abiotic factors are the basis for the biodiversity that is present. This biodiversity will continue as long as there is abiotic diversity. Or as long as there is no substantive change to it. Within the resilience of the ecosystem they can survive without permanent injury. (5). Socio-economic and ecological problems should be identified. Key-factors in inhibiting sustainable development should be identified.

3. Thirdly, a long-term cost-benefit analysis, based on sustainable development of natural resources, should be set up for the entire river basin in its present state. Also long term cost-benefit analyses should be made for proposed solutions (step 4).

4. Fourthly an inventory should be carried out of possible solutions to solve the problems. Active participation of stake-holders is highly desirable. In this respect it is very important to consider a (large) dam or a series of dams as one of the possible instruments/alternatives for solving (social) problems, rather than as an objective in itself. When considering a dam project, the long-term costs to society and the environment should be studied and compared with these of alternative solutions.

5. Finally, recommendations should be formulated for a sustainable management approach for the entire river basin. A river basin action programme should be formulated and approved by the governments and stake-holders involved.

In order to meet sustainable development objectives one should try to strike a balance between water and other natural resources. I would like to make some statements on the role of dams in the sustainable development of river basins In my opinion, the report of the World Commission for Dams concentrates its attention too strictly to reviewing the development effectiveness of dams and guidelines concerning �all aspects of dam building and utilisation�. Although these are very important subjects, there is a more urgent need for management plans at the level of entire river basins. Dams could play a role in it, but they are not the issue of primary concern in relation to sustainable development. They are not an objective as such, but



an instrument, which can and should be used in some cases. Unfortunately, there are not many river systems left in which we can implement alternative solutions as well. In less than a century, virtually all the major rivers in the world have been �reconstructed� as a result of the building of dams and reservoirs. There are now some 45,000 large dams in the world, and many more are under construction or in preparation. On the other hand, dams and reservoirs can in some cases offer new opportunities if one takes advantage of changing environmental boundary conditions. The Netherlands for example, is protected and divided by dikes and dams. The impact of this infrastructure on the original environment has been tremendous. There have been some fundamental errors, but in the mean time by compartmentalisation of the land has created numerous new opportunities with social and ecological prospects. At this moment there are ongoing studies to see whether the mistakes made are amenable to correction. A differentiation of management alternatives in neighbouring compartments (polders, lakes, estuaries) is a distinct possibility. A diversity of ecosystems has been created, a diversity of environments with a large variety of land and water uses. As a result, an impressive nature and biodiversity have developed and for the inhabitants a prosperous and fertile land, with high potential. Point by point The report quite properly devotes a great deal of attention to the maintenance of the existing (natural) environment. This is almost universally undervalued in decision taking and practically nowhere is there an �awareness of ecological costs�. That while the costs of a dam are generally underestimated and the benefits too low. Long-term cost-benefit analyses are never made, while the damage in the long-term can be huge (For example soil pollution Rhine outlet). Large sections of the report have been written with the underlying emphasis on dams with reservoirs in rivers serving for storage and timed distribution of the water. There are indeed quite a variety of dams. In my opinion there is too much emphasis on dams and collection reservoirs for providing water storage and its timed distribution. Too little attention is paid to other types of dams designed for safety in estuaries for example, for hydroelectricity, for collection of downstream mining residues of for example copper mines with all their consequences. For hydroelectricity there�s quite a difference between the dam being on the mountain or in the middle or lower reaches of the river. [The Netherlands for example] Of the original 8,660 km2 of estuaries in the Rhine-Maas delta in 1900, there remain only 3,930 km2 in 2000. In one century the construction of dams has led to the loss of 4,730 km2 of estuary area. When Costanza�s key figures (1998) are applied to these estuaries and the new systems, the gross National Nature Product of the estuaries in 1900 is estimated to have been ca. USD 19 billion / annum. These water systems now represent a capital value of ca. USD 386 billion. The 4730 km2 of estuaries are responsible for a loss of national nature product of USD 10.5 billion / annum and a cash loss of ca. USD 210 billion. Compensating for this is a profit of 2,784 km2 of new lakes and new land with a combined national nature product of USD 2.5 billion / annum and a capital value of USD 51 billion. The net production loss can then be estimated to be USD 8 billion and the net capital value loss to be USD 159 billion (all amounts in 1998 exchange rate values). These costs are never included in cost-benefit analyses. Current policy is aimed towards where possible restoring estuaries environment. This too will cost a fortune. What are we trying to do?



There is quite rightly a great deal of attention being devoted to threats and negative effects. But too little attention is being given to opportunities and possibilities of the new systems and their potential. See �Environmental considerations��..� and �Creative in a changing delta� Based on experiences in the Delta Project in the Netherlands new directions are suggested for large-scale hydraulic engineering projects in estuaries. Emphasis is put on greater awareness of the many factors involved, especially on the ecological implications of the project, and the necessity to incorporate flexibility that will make it possible to cope with changes and unexpected developments, and on recognition of the fact that processes are being dealt with. The alternative comprising preservation of the existing situation (the To alternative) must play a much more important role in future preparations and decision making. If a civil engineering structure is inevitable the process of modification and transformation undergone by an area as a result of hydraulic engineering works must be considered at least as worthy of attention as the process of designing and constructing the works themselves. Both processes must play an important part in decision-making, from the preparatory stage to the after-care. A sectored approach must be avoided. Every effort must be made to achieve an integrated approach, taking the basin as a whole as model. This demands smooth administrative co-operation. Multifunctional considerations should be given greater weight in the design of hydraulic projects. In this connection, more attention should be given to their function as an eco-technical management tool. Wide freedom of management must be incorporated into the design, to offer greater flexibility in response to changes, unanticipated events, new views on management; and such hydrodynamic works must be seen as subsidiary (as regulatory instrument) to the ecological and social functioning of the systems they can exert an influence on. A probabilistic approach to the design must therefore be related not only to the primary functions and the existence of the construction but also to the (future) requirements for modification, transformation, and management of the region that is affected. Decisions to execute hydraulic projects in regions for which (future) management plans do not exist must therefore be considered premature. The objective is not to resist change but to guide it properly. Ways to achieve these goals are indicated in the thesis �Changing estuaries�, chapter attached) The learning process too is receiving too little attention both within the project and between dam projects. [Experiences with the Zuiderzee project and the Delta project in the nineteen-eighties have resulted in The Netherlands in a revolution in water management that is now known as an �Integrated water system approach�. This new type of water management has had its impact on the conferences in Rio de Janeiro (1992), Dublin etc. (See also �Creative in a changing delta� attached)] Dams receive too much attention. A dam is just an instrument for controlling the physical environment. The discussion should in fact be focussed on the problems of the river basin and alternative solutions. A dam may certainly be appropriate but other solutions should be investigated too. The issue of dams can moreover not be considered in isolation. There are many other reasons for modifying river systems (shipping, land reclaim, safety). The effects are usually a result of these modifications combined. There are for example in the Rhine basin and its tributaries ca. 450 structures (dams, weirs, locks etc) all separately managed without any co-ordination. Result greater discharges with greater chance of flooding and of summer droughts (See article A Quest etc.) The almost universally adopted sectored approach to decision-taking for dams generally results in erroneous conclusions. The requirement for and desirability of a dam should in my opinion be judged against the background of the management of the entire basin area and all human intervention taken as a whole.



[Currently for example in the Rhine and its tributaries expensive projects are being implemented in an attempt to restore lost environments. I take for example �giving the river space as an alternative to dike reinforcement�; or �restoration of the flood plains and estuaries�. These recovery projects run into the millions and should be a lesson for new projects. It�s open to question whether a profit will result from a long-term costs-benefit analysis.] The soil at the outlet of the Rhine and the Meuse, the Hollands Diep Haringvliet area is heavily polluted from contaminated silt from upstream. The costs for remediation are estimated to be in excess of USD 75 billion. An important lesson is that one should never dam a river before dealing with upstream pollution. Dams are too often seen as an objective in themselves, resulting in other permanent solutions being ignored. Appendix A Statements on the role of dams and reservoirs in the sustainable development of river basins. • Sustainable development, management and use of natural (water) resources require integrated river

basin planning. The long-term conservation of natural resources and the services they deliver to humankind (e.g. productivity, water retention, energy, clean water for all kinds of purposes, biodiversity) is the main objective of river basin plans, in order to safeguard a multiple and wise use for us and future generations.

• Dams and reservoirs are instruments to meet an objective and are not an objective in themselves.

During the initial stages of planning a dam project, the question should be asked whether alternative solutions exist for problems in the river basin with lower long-term costs to society and the environment. A dam project must fit in an integrated river basin management plan; if not and/or if better solutions are available, the project should not be carried out or the dam should even be demolished.

• International organisations like ICOLD and the World Commission for Dams should promote

examples of dams that contribute to sustainable development. Examples of (existing and planned) projects, which are not contributing to sustainable development, should be mentioned to the world explicitly.

• Dams and their environments interrelate with a degree of complexity that makes the task of the

dam engineer particularly difficult but of great interest. All dams and reservoirs become a part of their environment and influence, modify or even transform it to a degree and within a range that varies from project to project. Dams and reservoirs play an (unplanned) role in landscape development, with both negative and positive effects on the quality and the use of the landscape. Dams and reservoirs can offer new opportunities if one takes advantage of changing environmental boundary conditions. This approach can be very important for the existence, design and execution of a dam, a real challenge for the ingenuity of engineers in co-operation with ecologists, environmentalists, social experts and people from many other disciplines.

• The larger a dam project, the greater will be its likely effects on the natural and social environment,

and the wider the scope of the multidisciplinary, holistic studies which it requires. Large-scale development demands integrated planning for an entire river basin before the implementation of the first individual project(s). When river basins are located in more than one country, such planning presupposes international co-operation. Integrated planning includes the development of management plans for the operation of dams and reservoirs within a river basin.



• A dam and/or reservoir project is not finished after construction. After construction, the management of that construction starts and effects become clear. There are more alternatives to operating a dam, taking more or less advantage of changing environmental boundary conditions.

• The market will grow for new sustainable and inventive solutions. There is a growing demand for

ingenious infrastructure and alternative solutions as a tool for achieving sustainable development, management and use of the river basins in the world. Some existing dams will have to be adjusted to these new insights or even have to be reconsidered.

• Decisions for the construction of water infrastructure or the management of water systems appear

to have a timeless value. In any case, they usually have consequences for centuries. Therefore long term costs and benefits should be studied for individual project proposals together with other possible alternatives before any decision is made at all.

• Dams and reservoirs can contribute to a better distribution of available water resources, but they are

not the only solution. A lot of attention needs to be paid to measures for reducing the growth of the world population and raising awareness of sensible use of available water resources. Often, better control of the quantities of water in a river only generates a feeling of abundance, so that incentives for more efficient use are not considered. The average efficiency of irrigation in the tropics has been estimated by the FAO to be only 30%. Water should be priced to cover its true cost.

• One of the central problems is that too much priority is given to building even more and even

bigger constructions. The emphasis should be placed much more on management than on construction. The key question should not be �where can we build even more�, but �how do I use efficiently and effectively the infrastructure that I already have. And how can this be improved in relation to those questions of management�.

• Many projects are carried out to serve local, regional or national needs without taking into account

the real causes of problems and without taking into account the effects on the entire river basin. • Decision-making in water management is quite frequently dictated by disasters. The challenge is to

meet decisions based on long-term cost-benefit analyses. • The time has come for economists and ecologists to work together. This will certainly result in new

instruments for ecosystem management and perhaps in new applications of economics. By analogy with the concept of eco-technology, we could call the new applications of economics for ecosystem management �eco-economics� or abbreviated to �ecoconomics�.

• Wide-ranging consideration must be given to the question of how the river basins should be

managed internationally in the next century. • Management of water systems should be organised at the river basin level. The basis should be laid

by an international (holistic) evaluation study, followed by a policy analysis at the river basin level. In order to activate this international management of the river system, a step-by-step approach is perhaps the correct method: First set-up a Co-ordinating Committee; Then set-up a River Basin Commission; Finally, when the time comes for this, set-up a Management Authority, with appropriate powers. The task package of this management authority might include: quantitative and qualitative water management; environmental protection; integrated management of existing infrastructure; ecological recovery of the river system; and co-ordination and harmonisation of new infrastructure construction.

• Political decisions about water management issues should be motivated short-term motive, but

should be based primarily on an explicit long-term strategy.



k) Comments by Martin Perusse General Comments 1. Considering the scope of this report and the needs of the WCD to document and to review

different thematics in order to make sound recommendations, I think this report misses the point and is not very useful in its actual format. In fact, this report is more a pasting of anecdotes and big numbers from which general conclusions are drawn (even if it should not be done). This approach has no scientific ground, is tendentious and misleading. For each example that the report takes to argue I could give an example showing the opposite.

2. The report argues that dams are a threat in every continents, especially for biodiversity. But

curiously, surveys made by international agencies such as the United Nations and the « Global environment outlook 1997 and 2000 » do not come to the same conclusions. Global environment outlook 1997 shows that the greatest threats to biodiversity are :

Africa : agriculture, grazing land, energy production from wood ; Europe : agriculture, industry, transportation, tourism ; Asia-Pacific : strong demographic growth, strong economic growth, clearing of forests, introduction of exotic species ; Western Asia : over-fishing, marine pollution Latin America and the Caribbean : agriculture, animal husbandry, forestry operations, cutting of firewood, road construction ; North America : thermal generating stations, burning of fossil fuels. As you can see, there are many sources of threats but dams are not the biggest problem on earth, according to the UN.

3. The report concludes that dams have major impacts and is a threat to natural ecosystems. If we

follow on this chain of logic, countries that use massively dams for energy production for example should see their natural capital destroyed. Canada and Quebec especially is a good example to look at for a demonstration. Quebec produces more than 95% of its electricity from dams. Quebec, as all of North America, is a big consumer of energy. So Quebec should be an ecological disaster ! Well, let me tell you that it is not the case. Natural capital in Quebec is very well preserved. Moreover, just think if Quebec would have to replace all its dams to produce electricity... by fossil fuels ! Now you might get an ecological disaster don�t you think.

4. The report makes it clear that the creation of a dam and a reservoir on a river will change (or

destroy if you wish) forever this ecosystem. This is a fact, a reservoir-lake is not a river. Physical, chemical and biological characteristics are not the same. But this is not the real problem because a reservoir has its own set of characteristics making it a different but productive ecosystem and achieving natural functions of regulation, habitat, production and information. There would be a problem if all rivers would ultimately be dammed and transformed into a reservoir. That would be a loss of natural capital. But this is not the case. Again, as an example, Quebec has only about 0,1% of its river length turned into hydroelectric reservoirs, has less than 0,4% of its rivers dammed, has only 15 of its 430 large drainage basins with Hydro-Quebec dams and has about 1,2% of its surface area covered with reservoirs. Worldwide dams cover about 0,3% of the earth (HQ and GDG, 1999).

5. One must be careful concluding that one option, dams, is not a good option. One cannot

concludes that way without assessing the alternatives. It is not dams or nothing but rather



dams or something else. And in a deregulated market with more interconnections this trend will just increase. Life cycle analysis of electricity generation options must be used before such conclusions are to be made.

6. I think the conclusion that dams should be the last resort must be deleted. Again, one would

need a life cycle analysis of options before arriving at such conclusion. Moreover, I don�t understand this conclusion because in the text examples of good dams are presented (Fortuna and Pehuenche) and criteria to select good dams are shown.

7. You should have a look at a news from Reuters at the end of my comments. Worlwide

electricity consumption is expected to increase by more than 70% in the next two decades. Fossil fuels has the lead and is expected to continue its domination. Is it what the report means when it says that dams should be the last resort ? I am not sure than our natural capital will be better off with such a strategy. This is rather a wake-up call that more renewable is desperately needed including dams because new renewables like wind and solar will not be enough considering the needs and the limitations of these options in terms of potential, economics and level of services.

Specific Comments

3.2 page 12, The descending order of dams and impacts may be misleading since it does not account for the level of service of each type of dams. Storage dams do not provide for the same service as a run-of-river dam. By doing so, one compares apples and oranges and does not acknowledge for the well recognised fact that many small dams are more detrimental for the environment than one big dam. Box 3.3 page 18, This box does not acknowledge for the fact that impacts from transmission lines are not typical of dams but of any type of production. And again, such as in table 3.1 numbers are used out of context. What is the importance of 32 000 km of lines, or 3 200 sq km (there is an error in the text the number should be 1350 sq km), for the province of Quebec with its 1,6 millions sq km. The use of big numbers, mainly because they are big, has no scientific basis and is misleading. This box should also be revised since facts from HQ and GDG study and conclusions made by IUCN authors are put together but not always with clear reference to each other. 3.5 Upstream impacts page 20, I do not like to read in a scientific analysis a phrase such as « Dams also often have a bottom outlet. » Why not : Dams also often have a upper layer outlet ? ? (which is also true, I could give you many examples). Such phrases are imprecise and tendentious. Stick to facts. Most earth dams have no bottom outlets. 3.5.1.3 Changes in water quality, page 22, The paragraph on mercury should be revised. It is too general to be of any use. It does not talk about real reservoirs and populations dealing with this issue. It sticks to the chemical toxic potential but without explaining that in many countries populations and fish have not been intoxicated with mercury. And the use of McCully as a reference is not very serious since I do not recall him as being an expert on the issue. Major studies done in Quebec, Sweden could be used more efficiently (Mercury in the biogeochemical cycle by Lucotte et al.1999 Springer-Verlag). 3.5.3.1 Invertebrates, fish, birds and mammals page 25, The phrase « As many species prefer valley bottom habitats, large-scale impoundments are likely to extinguish entire populations of species » is not very convincing. This does not make this report a serious scientific analysis of dams, especially without reference and clear examples. Moreover, even



with examples it is always scientifically dangerous to jump from a few examples to a well accepted and supported fact. Same for the other phrases of this paragraph. 3.6.1.2 Water quality page 29, The problems of low dissolved oxygen and gas bubble disease are not universal problems at all. This is specific to some type of dams and in Quebec for example we do not have these problems. Same for sediment transport and turbidity in 3.6.1.3. 3.6.3 Third order impacts on fauna page 35, Again, I do not see a rigorous scientific analysis here about large woody debris. The real contribution to streams and rivers is yet to be known, the impacts of dams not known. Moreover, wood is not systematically removed by dam operator, this is not necessary for dam operation and it would be too costly anyway. The text also suggest that only one point on a river contributes to downstream woody debris, which is not true. Debris fall all along the river, including downstream of dams. Another theoretical and not supported analysis concerns the hippos. The text talks about potential impacts to hippos without references and without acknowledging the possible positive impacts. In fact regulating rivers may be good for hippos since water will be provided all year long, which may be crucial in dry season when hippos die because of the scarcity of water. 3.7 Consequences of dams for species diversity page 38, This section is tendentious because it relates indirectly species extinction to dams but without talking about the many other threats. Agriculture, grazing, land conversion for development and outdoor recreation vehicles (ORV) have been identified as more serious threats that dams. Saying that North America freshwater species are the most endangered species group is false. First, the text should refer to USA only and not to Canada. Secondly, it is said that it is the most endangered group based on the rate of disappearance but without saying that other groups such as plant have more species endangered. The report cannot use statistics only when it suits the argument. 3.7.1 Bivalve and gastropod molluscs page 39-41, The report takes a few examples to relate dams and the disappearance of molluscs but without acknowledging for other threats that are at least as important as dams, agriculture and pollution. And the Mississippi is a very good example of multiple sources of impacts, where dams have probably one small contribution to its deterioration. 3.7.3 Dams and waterbirds page 44, A scientific analysis would have said that agriculture and land conversion are the biggest threats to endangered birds not dams. 3.8 Cumulative impacts page 46-49, Again, the use of numbers is misleading. Figure 3.7 says that only four major rivers remain undammed. What is a major river ? Are major rivers more important than not major rivers ? What is the ecological significance and contribution of major rivers to the Sweden environment ? And the not major rivers of Sweden ? The phrase « In view of the severity of these cumulative impacts.... » should be deleted. There has been no serious and scientifically demonstrated analysis showing the cumulative impacts of dams on the environment. The fact that cumulative impacts are included in guidelines and policies underlines the importance to consider multiple and additive sources of impacts (and not just dams) on the environment.



Specialists in that field acknowledge the fact that cumulative impacts is a very seductive concept but still with major methodological constraints, suggesting that interpretation is subject to caution and that conclusions cannot be made. 4 The economic and social implications of ecosystem impacts page 51, Introduction : The phrase « Chapter 3 has in turn shown that dams impact upon natural ecosystems to change these functions, usually diminishing the value of the benefits that they provide to society » should be deleted. Chapter 3 (and I read it carefully) does not provide for such a demonstration. To achieve such a conclusion, the report would first need the right method (which it does not have), the report would have to consider the positive impacts of dams (which it did not do) and the report would have to consider the alternatives to produce irrigation and electricity (if not dams then what candles or coal...). The report seems to have an hypothesis right in mind but there is no conclusive scientific demonstration of that hypothesis in your text. This is not good science. This section lacks the following : -multicriteria requires the comparison of alternatives (if you don�t build dams then what ?) -alternatives must provide for the same level of service -impacts of dams can also be positive (avoided emissions, watershed protection, potable water...) -determination of all costs and benefits of all alternatives. 5 Responding to the ecosystem impacts of dams page 63, This section talks a lot about decommissioning as a measure but without mentioning one major consequence that would happen with its wide use. At some point the removal of dams that produce electricity could mean producing electricity with other sources such as fossil fuels. 5.3 How effective is mitigation ? page 64-65, The first paragraph refers to the IEA study (2000) by saying that even if many measures are proven to be efficient one cannot imply that these measures would work in all cases. You have to admit that this study refers to many measures in many different countries and situations. But I agree that we have to be cautious. What I do not understand is why the report is so cautious here while in many places in the text it is not, concluding with a few examples, often without references ? This does not seem to be a well-balanced analysis. One cannot easily generalise on the efficiency of mitigation measures, and neither on the impacts of dams. Page 67 : « Fish passages have been considered singularly ineffective by some experts in Brazil... ». Who are these experts, references ? This is not very serious. 5.3.2 Why mitigation is difficult ? page 69, The phrase « The science of wetland and river ecosystem management and the downstream impact of dams is still in its infancy and falls far short of 100% predictive capacity » should be deleted. First, uncertainty is a fact of science and especially of ecology. No real scientist and ecologist is looking for 100% prediction. Any result that would achieve 100% would be considered suspicious by the scientific community ! Secondly, I have some difficulties to follow the logic here because if it is still in its infancy and poor in predictive power, how can it be so conclusive with all those impacts of dams in chapter 3 or when chapter 4 says that « Chapter 3 has in turn shown that dams impact upon natural ecosystems to change these functions, usually diminishing the value of the benefits that they provide to society ». The analysis shows no scientific basis.



Table 5.1 page 70, this table is not very useful. We need water quality data to assess water quality impacts. We need aquatic biota data to assess impacts on aquatic species and so on. Sure but not very informative ! 5.4 Tools for environmentally sound management of water resources page 73, I disagree totally with the first phrase. I do not see how the report can say that mitigation is only rarely achievable in most situations. On the contrary the IEA study shows that many measures work in different situations and countries. 6 Trends in the international debate page 79, I have some problems with this section because all come down to pros or cons. This is too manichean. It is too simplistic and it does not reflect the complexity of such questions. What about the various perspectives of stakeholders, agencies, governments, etc. ? 7 Conclusions page 88, The scope of the conclusions is misleading because there is no clear demonstration that all dams in many countries of the world have all those impacts and that it is always negative. There is also no clear demonstration about mitigation. Moreover the report cannot conclude in such a way without assessing the alternatives. It is not dams or nothing but rather dams or something else. Because of that the recommendation that dams should be seen as a last resort is made out of the global context and must be deleted. Globally your recommendations are too general or too vague to be very useful, even with the operationalisation options. l) Comments by Robert Dobias 1. The draft document provided interesting reading and is an important contribution to the WCD

process. A few comments are provided below for the authors� consideration, and for the most part focus on Section 7: Conclusions and Policy Recommendations for WCD. I don�t believe that any of the comments below mention concepts not already somewhere in the paper, but the comments may present minor changes in emphasis or suggestions for moving the concepts more explicitly into the paper�s policy recommendations.

A. Incorporating ecosystem considerations early in the planning process 2. The need for early incorporation of ecosystem/biodiversity concerns in the planning of dam

projects is mentioned in several places (though, unfortunately, not explicitly in Section 7.2). This certainly is a key consideration in improving the environmental acceptability of dams. For countries that have an active hydropower and/or irrigation development sector, there would seem to be two primary policy measures that might be taken, one short term (1-5 years) and one longer term (5-15 years). (Also pertains to Section 5.3)

3. Short term Where there are national energy and irrigation development plans (virtually

all countries), these plans should be assessed for their treatment of ecosystems/biological



resources (and of course social issues). If ecosystem concerns continue to be ignored in these plans, then much of the battle is already lost in terms of improving environmental conditions in water resource development � we will have to continue to rely on mitigation measures that come so late in the process. Where these considerations are found to be weak or lacking, the plans should be revised based on rapid river basin assessments to be commissioned as part of the revision process. Because the sector development plans will already give a reasonable indication of where and when major projects are to be implemented, the rapid assessments can be targeted at what are likely to be just a handful of basins.

4. In most cases, the issue of financial and human resource capacity raised by the authors may in

reality not be an issue at all. This is because the work need not be done in all basins over the short term (for reasons stated above), and because those countries with financial and human resource constraints are also likely to be countries receiving support from bilateral and international aid agencies. Thus the policy should be directed particularly at aid agencies as well as governments. The private sector would, of course, be required to follow the revised plans.

5. Longer term The longer-term policy recommendation would be for the preparation and

implementation of river basin management plans that, by definition, would consider ecosystem conservation. This has already been mentioned by the authors, and they provide an excellent recommendation earlier in the report regarding the preparation of plans to �set aside� certain basins. However, this might be expanded to state that budget approval for sectoral development should be contingent on the sectoral plans being in line with the river basin management plans. This way, central budget agencies would be required to assess hydropower and irrigation proposals in light of the river basin plans, and withhold funding from any proposal that contravened these plans. At minimum, this would eliminate �surprise projects� which so often are most destructive of the environment.

B. Use of strategic environmental assessment (SEA) 6. I believe this is mentioned in the report, but it does not seem to be explicitly mentioned in

Section 7. National policy requiring SEA for hydropower/irrigation policies and development plans would be an important step in the early identification of unacceptable projects or project components and related avoidance recommendations. (See also Section 5.3.3)

C. Translating EIA/EMP recommendations into project contracts 7. This would seem to be another key to limiting the potential environmental damage produced

by dams (also see Section 5.3). So much of the unsatisfactory environmental effects of dams, at least in this part of the world, are due not to poor EIAs/EMPs (though this certainly is a factor), but to the inability to translate EIA/EMP recommendations into project contractual documents, especially those related to civil works. Most agencies have procurement guidelines that include templates for tender documents, contracts, bills of quantity, and so forth. These agencies (including the international donor agencies, which tend to use common procurement guidelines) should be requested to review these guidelines for their effectiveness in getting environmental recommendations into contracts. In many cases, I suspect that this will be found sorely lacking, and perhaps limited to sweeping statements on the need to �add a section� on environmental requirements or �the EIA must be implemented� (which has virtually no legal teeth with contractors).

D. Adequate legal framework and compliance mechanisms 8. Section 5.3.3 correctly states that more supervisory missions and closer monitoring have been

recommended to ensure compliance in countries where there may be a weak legal framework



or will to implement existing legal requirements, and that these monitoring activities may unfortunately cease when the contract expires and the donor considers the project complete. However, this issue shouldn�t be left at that. In addition to legal capacity building, donors could play a key role in ensuring that long-term monitoring mechanisms are included in the project design that would carry through to the project operation stage. This is not particularly difficult to do, but is not often done perhaps because the donors tend to lose �leverage� once a project is completed. However, if a portion of the project proceeds are contractually committed to monitoring and reporting during project operation, for example, then this could be continued without further donor presence.

E. A few other observations 9. Why does the discussion of cumulative impacts of interbasin water transfers merit just a single

short paragraph on page 44? These projects are often the most destructive of biological resources and beneficial uses.

10. In Box 4.1 on page 50, is it realistic to state that those who cause damage are liable to ensure

that the environment is restored to its former state? Clearly, this will not be possible in many �good� projects no matter the desire of the project proponent to do so.

11. On page 54, the authors may also mention that biomass clearance is another option to deal

with nutrient accumulation. 12. I�m a bit surprised that the environmental indicators in Box 5.5 are touted so highly by the

authors. m) Comments by Stuart Blanch Comments have been made in reference to specific sections and issues, with page numbers given. Page

Comment

10 I thought the chinook salmon example was inappropriate given that the species is exotic to NZ.

13 Table 3.1. I am pretty sure that mean annual flows at the mouth of the Murray R in the Murray-Darling Basin have fallen to 21% of natural - not 35%. For clarification suggest you read Murray-Darling Basin Commission website.

15 Table 3.3. Thermal pollution and carbon flows/cycling should be included here in respectively first order and second order impacts.

24 Para 2. Small to medium sized flows, which inundate floodplains and wetlands in the Murray-Darling Basin, have been reduced more than any other aspect of the hydrograph. This warrants special mention. For example, these sized flows have fallen to 30% of natural in the lower Murray R.

Also, reductions in flow velocity in weir pools in the Murray R have been given as a key cause of the decline in silver perch in NSW, leading to its recommendation



as a vulnerable species (see accompanying document).

Also, seasonal inversions of flows due to releases for rice, dairy and orchard irrigation in the Murray are environmentally damaging. Suggest inclusion of a graph from the Murray-Darling Basin Commission's website.

24-26 There is insufficient mention of the impacts of many weirs, acting in sequence, in reducing instream flow variability. For example, papers by Thoms and Walker and others in the journal Regulated Rivers Research and Management.

As in much of the document there is too much focus on data from the Colorado R and other US rivers - there is a lot of good information from Australia that wasn't mentioned and the report is the weaker for it.

Also, turkey nest dams or ring tanks should be mentioned. These are 4-5 metre high dams built out of dirt on floodplains to store water, mainly for cotton. They commonly store about 5, 000 to 10, 000 megalitres. Extractions of water for these can greatly impact flows.

25 Second last para (The hydrological effects�) should include mention of terminal wetlands and their shrinking from intercapetion of flows upstream. This is particularly important in wetlands like the Ramsar listed Macquarie Marshes and Gwydir wetlands in northern NSW.

Last para (Flow regimes are the key�.). The period since last flooding is another key feature of the hydrologic regime that should be included in the second sentence (Flood timing, duration�). Also mention should be made of recruitment into the seedbank and sporebank in addition to mention for fish etc,.

26 Water Quality - 600 and 1500 electrical conductivity units are emerging salinity thresholds used in NSW for guiding flow management. Also a Q10 of 10 degrees Celsius is probably relevant for biogeochemical processes including decomposition rates.

Cold water pollution impacts on average 300 km below each large dam - hence water temperatures are 5 degrees Celsius or more below normal - with a total of about 2650 km across every major inland river in western NSW (see accompanying paper).

28 Under Floodplains section: need to mention hydrological connectivity and divorcing of floodplains from rivers as key impacts of flow regulation.

Also losses in floodplain area - the Ramsar listed Macquarie Marshes formed by the Macquarie River in northern NSW are now about 50-60% of pre-dam size.

Also terrestrialisation of higher floodplains and invasion by terrestrial weeds due to loss in flood frequency.

29 Section 3.6.2.1. Need to discuss impacts on carbon dynamics, bacterial and fungal decomposition and higher impacts via food webs. Also filamentous blue-green algae are important.

30 Disagree with sentence beginning Although water depth and light�. In floodplain rivers with low velocity, water level changes are far more important than current velocity and scouring.

Mention needs to be made of the rate at which water levels change in relation to the photic depth - this is the key impact on photosynthesis in submersed plants.

There is no mention of submersed species in section 3.6.2.2 e.g. Vallisneria!

Water temperature impact on germination, photosynthesis and growth in



submersed species needs to be mentioned.

Also, carp favour weir pools with stable water levels - carp disturb aquatic plants and increase turbidity.

31 Under section 3.6.3.1 A 40% reduction in flows sufficient to allow the passage of native fish in the Lachlan River in NSW has occurred from flow regulation and irrigation diversions.

In the Murray R floodplain mussels have invaded the main channel and the species preferring deeper, colder water have reduced in distribution and abundance.

32 Para starting River-dwelling species� should mention impacts of slower water on the survival of pelagic eggs of fish e.g. see the attached report on silver perch.

41 100, 000 fewer pairs of colonial nesting waterbirds are estimated to have bred in the Macquarie Marshes in northern NSW over 12 years due to flow regulation from Burrendong Dam and irrigation extractions.

50 Chapter 4 is too short and simplified. This is a key issue and needs more information and analysis.

51 Section 5.2 Types of response - the first question is 'Is the dam really absolutely necessary and are there no other options?".

53 Under Mitigating for Operational Impacts - a 'continuous supply �' is definitely not what Australian rivers need from dams. Environmental flows here are modelled and released according to natural flows and inflow triggers. A zero release in summer is probably much more appropriate than a continuous release over summer when many of the rivers here would be low or even dry.

54 Variable level offtakes are the most expensive option for mitigating thermal pollution impacts e.g. submersed surface impellers are being investigated here and are a fraction of the cost.

56 In 'Because much of the�' - Stocking of fish generally does not produce a self-sustaining population. See the conclusion of the NSW Fisheries Scientific Committee's on the effects of stocking of silver perch.

63 Under Table 5.1 - Carbon should be mentioned under 3. Water Quality.

Also, floodplains, groundwater and wetlands should be mentioned under Biotic.

Under 5.3.2.2. Feasibility is the main constraint of mitigation. Once a river is impounded, no amount of mitigation will restore (i.e. to natural) it.

n) Comments by Musonda Mumba A. General assessment of report Date, authorship and structure The draft provided for review is dated 10 March 2000 and compiled by G. Bergkamp, P. Dugan and J. McNeely of IUCN HQ. Contributing papers were prepared by experts contracted by IUCN/UNEP, FAO/WCD, WCD alone, and the UK Department for International Development (DFID). The report includes 120 pages, of which approximately 85 pages are devoted to the seven main sections: 1. Introduction



2. River Basin Ecosystems and Biodiversity 3. Ecosystem Impacts of Large Dams 4. The Economic and Social Implications of Ecosystem Impacts 5. Responding to the Ecosystem Impacts of Dams 6. Trends in the International Debate/Approach to Dams 7. Conclusions and Policy Recommendations to WCD An Executive Summary and References will be added to the final version. There are seven technical annexes to the report, which are cross-referenced in chapters 3 & 5. An eighth annex lists all relevant submissions received by the WCD. Observations on presentation and style A key weakness is that in trying too hard to be �objective� and �independent� from the positions advocated by more �radical� environmental bodies, the IUCN review has lost much of its potential weight and clarity. The compilers constantly seem to be �bending over backward� to present arguments �for and against�, so that it is frequently difficult for the reader to determine what conclusion or direction the report is really advocating. Whilst this may help keep the proponents of dams open to constructive debate, many environmental NGOs, including WWF will feel that IUCN should reach stronger, clearer positions � particularly in the concluding chapter. This impression is reinforced by the lack (in this version) of an executive summary and the limited number of conclusions from individual chapters that make it into the final conclusions in chapter 7. Highly significant statements are often buried in a mass of dense text, which is sometimes quite difficult to follow � not because of its technical content, but because of a convoluted style. A thorough re-edit is needed to fix this. The structure is basically sound, though the report strays quite far from its mandate at times, with chapter 6, in particular, indulging in a recapitulation of the overall �dams debate�. Whilst it is important to recognise the progress that has been made by some international institutions in giving attention to environmental issues, the �broad-brush� approach of chapter 6 detracts from the focus of the Thematic Review. The immediate emphasis on biodiversity at the start of chapter 2 is misplaced, since biodiversity of freshwater ecosystems is introduced before the ecosystems themselves have been defined and described. The early emphasis on biodiversity may also �play into the hands� of those interest groups that would like to portray environmentalists as having a narrow nature conservation perspective. There are many references quoted in the text. On one hand these lend authenticity and seriousness, but they sometimes detract from the flow of the text. More significantly, a lot of key issues are ONLY covered through a brief bracketed reference and it is assumed that interested readers have ready access to some pretty obscure sources to get further information on these issues. Overall conclusions and recommendations (Chapter 7) This chapter presents some general conclusions, together with eight Recommendations to the WCD and �Options for Operationalising the Recommendations� (the latter boil down to suggestions for implementation). Conclusions The conclusions are weak and poorly presented as continuous text, rather than concise, numbered conclusions. We do not need a 120 page review to tell provide such generalities as, �there remain significant and widespread concerns about the environmental impacts of dams�, or �the importance of



natural ecosystems is widely recognised by national governments...and international agreements�. Amongst the few really useful statements in this short section is that: �The review has underlined that dams have a wide range of major impacts upon natural ecosystems, that most of these are negative, that many are irreversible and that they are manifest in economic and social costs�. However, it is regrettable that the there is no clear statement to the effect that large dams ALWAYS have at least some adverse impacts on ecosystems and biodiversity. Avoidance, mitigation, compensation and restoration are listed as the four main categories of �solutions� proposed for dam impacts on ecosystems, with mitigation being described rather weakly as �particularly problematic�, and having �usually residual impacts�. This paragraph (bottom of p.80 and top of p.81) continues by giving an impression that mitigation is a generally attainable and acceptable solution, providing that various conditions are met, but then concludes (more helpfully) �mitigation, though often possible in principle, is at present not a credible option in many cases�. Important findings in the body of the report, which should be imported into a significantly strengthened conclusion section, are dealt with on a chapter-by-chapter basis in section B below. Recommendations The eight recommendations to the WCD do not really say anything very new and are presented in very general terms. They can be summarised in only five points as: • recognise the value of freshwater ecosystems and biodiversity conservation for sustainable

development; • recognise and manage for uncertainty (i.e. there is no �one size fits all� approach); • ensure effective participation in planning, design and management of dams; • when it is decided to build a dam, implement design and management measures to ensure that its

environmental impacts are minimised; • promote development of better legislation and application of tools such as environmental flow

releases, ecosystem health indicators and site selection indicators. The report fails to recommend clearly that the WCD should set international standards or guidelines for any of the above issues. Instead there is a much more cautious formulation on p.83 that �the Commission may wish to attach measurable norms or standards to these recommendations in order to move from the policy framework to implementation�. Important statements contained in the Recommendations heading, but which should be given more �up front� prominence as stand-alone conclusions are: �In future, no dam should proceed if it is shown to have a high probability of having a significant detrimental effect of species biodiversity� (Comment: the same should apply for ecosystem functioning and environmental goods and services. More work is needed to define the meaning of �high probability� and �significant detrimental effect�. The debate will continue to be subjective, and ecosystems and biodiversity will continue to be lost and degraded unless international definitions and guidelines are adopted by the WCD). �...A high degree of uncertainty and limited predictive capacity argue forcefully for adoption of a precautionary approach to dam development. Wherever possible dams and their impacts should be avoided�. (Comment: this is the first of only two references to the precautionary approach in the entire review and merits much more attention).



�It is essential that all dam projects and their impacts are subject to intensive analysis during planning and design. This needs to be pursued through open process....�. (Comment: the report fails to follow up this statement with meaningful �Options for Operationalisation� � see below). Options for Operationalising the Recommendations Table 7.1 �Options for Operationalising the Recommendations� (pp.83-85) needs substantially more refinement if it is not to do more harm than good. For example, the Options suggested for Recommendation 2 (�Recognise the importance of biodiversity....�) include statements including: �Dams should not be built in declared National Parks or Nature Reserves�. Does this mean that they are perfectly OK just outside the border of a National Park/Nature Reserve? Or that they can be built within other categories of protected area? Or that they can be built within proposed National Parks and Nature Reserves? �Dams should not negatively impact any Red Data Book species�. At what scale (regional, national, continental, and global)? What about areas that are poorly covered by the Red Data Book system? �River flows should not be reduced to zero during commissioning�. Does this mean that something just above zero is OK, e.g. 1%, 5%, 10%, 25%... Without any further elaboration, this is at best meaningless, and at worst highly dangerous. �Undertake full biodiversity surveys of rivers in order to allow the least ecosystem damaging choices and trade offs to be made� Biodiversity surveys, though important, do not in themselves provide the necessary information base for making such decisions. A full understanding of ecosystem functioning is required, making e.g. hydrological and sediment studies of at least equal importance. (This is in fact recognised in the Options suggested under Recommendation 3). Buried in the Options for Recommendation 3 (�Recognise and manage for uncertainty�) is the welcome statement that: �Dams should be seen as a last resort rather than the first choice (precautionary principle)�. This crucial finding, together with greater prominence for the precautionary approach should be clearly expressed in the main conclusions of the review. There is only one Option for Recommendation 4 (�Ensure effective participation in planning, design and management....�), namely to make EIA reports public documents. It appears to be assumed that everything else will be taken care of by the �participation� Thematic Review. The Options for Recommendation 7 (�Promote the development of national legislative frameworks�) mix economic and legal instruments in a confusing way. There needs to be explicit recognition that many countries simply do not have the political leadership, institutional capacity and/or human/financial resources to implement better environmental legislation, even if it is developed on paper.

B. Detailed comments on chapters 1 to 6 Chapter 1 � Introduction Strengths: • reasonably balanced concise statement of the �dams and environment� debate. Weaknesses:



• doesn�t really address the issue of �environmental restoration', which is given prominence in the title of the Thematic Review.

Chapter 2 � River Basin Ecosystems and Biodiversity Strengths: • Table 2.1 is a helpful summary, which could be usefully expanded to a full page to make it more

prominent and less cramped. Weaknesses: • Fails to provide a convincing and forceful enough exposé of the functions, values and attributes of

freshwater ecosystems. • Postpones adequate analysis of information on social and economic values of freshwater

ecosystems to chapter 4. • Gives too much early emphasis to biodiversity (an ecosystem attribute) early on, before the

ecosystems themselves have been adequately introduced. • Does not give a schematic representation of river basin ecosystem types (would be useful for non-

specialists, as per the river basin/water cycle diagram � Figure 2.1). • Insufficient attention is given to the regional diversity of river basin ecosystems (would be useful

to have a brief summary of �typical� riverine ecosystems of e.g. circumpolar, temperate, arid/semi-arid, and tropical/sub-tropical regions.

• Religious values are not included explicitly in Table 2.1 or the accompanying text. • The statement on p.7 that �wetlands in semi-arid and arid areas are known as prime areas for

biodiversity conservation� is misleading, since the non-specialist may conclude that wetlands in other areas are of lesser importance for biodiversity conservation.

• Insufficient attention is given to the special importance of river basin ecosystems for migratory species (passing mention only on p.7).

• Section 2.3 �Ecosystems and River Basin Development� is poorly written and argued, especially the contention that increased membership of environmental conventions is, in itself, a reason why efforts should be made to avoid irreversible loss of resources.

Chapter 3 � Ecosystem Impacts of Large Dams Strengths: • Table 3.1 is clear and forceful in its message, as are Figures 3.5 and 3.6. • Section 3.4 rightly draws attention to information constraints and the lamentable lack of

comprehensive studies. • Sections 3.5, 3.6 and 3.7 present a generally convincing picture of the negative impacts of large

dams on ecosystems. • Stresses (p. 27) that �a unique combination of climate, geology, vegetation, size of impoundment

and operational procedures produce the effect of any individual dam upon the fluvial processes downstream. Hence a wide range of geomorphological responses can be generated by river regulation�. (Strengthens the argument for thorough, tailor-made, case-by-case review of every dam proposal).

• The importance of adequately assessing cumulative impacts is recognised. • The chapter concludes that �dams have a significant and measurable impact on ecosystems. The

current state of knowledge indicates that the impacts of dams on ecosystems are profound, complex, varied, multiple and mostly negative�.

Weaknesses:



• gets off to a bad start by stating in the second paragraph of p.10 that �loss of some ecosystems

may benefit some species but others may suffer significant loss of population or even extinction�. This may easily be misinterpreted or deliberately distorted as implying that ecosystem loss can have its advantages. This is not a conclusion that IUCN should be encouraging!

• Table 3.3 needs more prominence and clarity as it is effectively the key to understanding sections 3.5, 3.6. and 3.7 which are in many ways the heart of the review;

• Tends to �bend over backwards� too much to show that there are sometimes positive impacts from dams (e.g. the almost comical note in Box 3.3 that �some species of wildlife use access roads for travel�). Also the statement on p.17 that �this report draws on the available literature while being aware of the dangers of generalising from �worst case� examples�. Some of the possible �worst case� and therefore, presumably, �too controversial� examples have been conveniently overlooked, so that e.g. impacts of the Three Gorges Dam are not mentioned anywhere.

• Incorrect contention on p.18 that �once a reservoir has formed and has reached a state of stability, its subsequent dynamic behaviour is very similar as that of a natural lake� (this is subsequently shown to be incorrect, but still needs putting right).

• Not all asserted impacts are supported by examples. (Value and credibility of review would be enhanced by fixing these gaps, e.g. sedimentation impacts asserted at end of section 3.6.1.3).

• Insufficient care is given to identifying (for the benefit of non-ecologists) which impacts are most likely to occur in which regions of the world. e.g. clarification needed in section 3.5.2.2 that the aquatic macrophytes may support disease vectors in tropical regions.

• Introduction to section 3.6 should recognise that downstream impacts may be felt thousands of kilometres away (not only tens or hundreds) and that the impacts are frequently transboundary.

• The sections on floodplains and coastal deltas on pp.28-29 are rather sparse in comparison with the very serious known impacts. The importance of FUNCTIONING floodplains and deltas for human beings is not adequately addressed either here or in chapter 4.

• The assessment of 66 case studies mentioned in paragraph two of p.33 is not referenced. • The section on dams and waterbirds should refer specifically to the transboundary issue of

migratory species. The potential loss of tundra of international importance for migratory geese due to flooding by a proposed dam in Iceland is a good example.

Chapter 4 � The Economic and Social Implications of Ecosystem Impacts Strengths: None discernible. Weaknesses: • This chapter is highly deficient in terms of presenting the available evidence for the social and

economic importance of naturally functioning riverine ecosystems. There is no real sense here of the immense number of people that depend on freshwater ecosystems for their livelihoods, with only one, partial, example: Hadejia-Nguru floodplain in Nigeria.

• Section 4.5 suggests � highly simplistically and naively � that central, regional or local governments are likely to base decisions about dams around the goal of �the greatest possible happiness for the greatest number of people�.

• Box 4.1 �Ethical principles for decision-makers� is parachuted-in with no reference. Are these IUCN�s ethical principles? If not, whose?

Chapter 5 � Responding to the Ecosystem Impacts of Dams Strengths:



• Useful classification of avoidance, mitigation, compensation and restoration. • The limitations of mitigation are well presented; it is clearly not the panacea that some interest

groups portray (however, there is an incorrect statement at the foot of p.52 where it is claimed that mitigation measures for new dams prevent the occurrence of anticipated adverse effects. By definition, mitigation only reduces adverse effects; it does not avoid them).

• Emphasises that the main measure to reduce the ecosystem impacts associated with dam construction is �to provide a continuous supply of water to the reaches downstream from the project and to the extent possible to provide that water in a way that mimics the natural hydrologic regime of the river�.

• Fair assessment of indicators for hydro-project site selection (section 5.4.1) and indicators of ecological integrity (section 5.4.2).

• Recognition in relation to Environmental Flow Requirements (EFRs � section 5.4.3) that �even the most successful EFR will only partially mitigate against the effects of a dam on a river� (foot of p.69).

Weaknesses: • The wrap-up sentence for the whole chapter (p.71) is another classic example of the review

�bending over backwards� not to upset pro-dam interest groups. After setting out five stringent conditions that ALL have to be fulfilled for mitigation to be likely to succeed, and stressing that if ANY ONE of these conditions is absent, then the ecosystem values will likely be lost, it is concluded �This would tend to encourage a strategy of avoidance and minimisation rather than one of mitigation if the aim is to maintain biodiversity, and ecosystem functions and services for the foreseeable future�. �...would tend to encourage� should be replaced with �dictates�!

• Avoidance, compensation, and restoration are not given the same in-depth treatment as mitigation. • There is a tendency to suggest that mitigation works OK in developed countries, but not in

developing countries (see section 5.3.2.2). This is dangerous, since it implies that mitigation is mainly limited by resource availability. This is clearly wrong in the light of imperfect scientific knowledge of ecosystem behaviour and mitigation �science�, as emphasised at the beginning of section 5.4.

• p.53 mentions �numerous compilations� of best practice guidelines for dam construction, but does not reference any of them.

• Insufficient attention given to the shortcomings of �compensation� as a solution; the review is not sufficiently critical of �out of basin� compensation, or �in kind� versus �out of kind compensation�. There is a pervasive implication that compensation is possible. Study after study shows that so-called compensation schemes rarely, if ever compensate for the functions and values which have been lost. The sterile �no net loss� debate in the US is a good example.

• In the framework of �avoidance, mention of �set aside� (p. 52) of some river basins for environmental protection may imply that other basins can then be freely developed without regard for environmental protection.

• Restoration (i.e. dam decommissioning) is given a somewhat dismissive treatment. The review appears to take the approach that restoration is only likely to be an option in a tiny number of cases, probably smaller dams, and in any case brings lots of problems and risks. IUCN and the rest of the conservation community should be taking a more cutting edge approach, calling for improved techniques for dam decommissioning to be developed.

Chapter 6 � Trends in the International Debate/Approach to Dams Strengths: • Summarises some of the key arguments of the �anti� and �pro� dam lobbies.



• Acknowledges the policy positions and guidelines etc. issued by IEA, ICOLD, The World Bank, and OECD.

• Underlines that better methods of economic valuation are needed, together with clearer guidelines on how costs and benefits can be distributed (e.g. appropriate institutions for promoting equitable water use.

Weaknesses: • Ranges into territory beyond the scope of this Thematic Review and therefore loses some of the

focus of previous chapters. • Through characterising the dam debate as �anti� and �pro�, tends to promote further polarisation of

the argument. • Implies that IUCN is a completely neutral bystander, which is, of course (or shouldn�t be) the

case. • Does not critically evaluate the extent to which good words in position papers and guidelines of

international organisations have been put into practice. • The �international conventions� section does not give sufficient coverage to the relevant

requirements of Ramsar or CBD, and does not mention that other global conventions (e.g. Climate Change, CCCD, CITES and CMS are all relevant too.

C. Comparison of report findings with WWF International Policy Paper Element of WWF Strategy on Dams: Conserving and protecting important ecosystems & sites The review strengthens the internationally authoritative literature on the potential adverse impacts of large dams on ecosystems. It also clearly sets out the limitations of mitigation and gives more credibility to avoidance as a strategy. The review is weak in its references to dams and protected areas (or proposed protected areas) which will be some of the key sites of interest to WWF. WWF should press for a much fuller treatment of the issue of dams and protected areas, as well as more in-depth coverage of the socio-economic values of naturally functioning freshwater ecosystems. Element of WWF Strategy on Dams: Supporting removal of redundant dams The review is rather negative about restoration of dams through decommissioning, and gives little help on other restoration and/or rehabilitation options for upstream downstream ecosystems damaged by dam construction. WWF should call for better leadership in restoration science and practice from IUCN and WCD. Element of WWF Strategy on Dams: Mitigate adverse effects of other existing dams The review sets clear criteria which have to be met if a mitigation effort is likely to stand a high chance of success. Element of WWF Strategy on Dams: Supporting the World Commission on Dams The review strengthens the position of WWF with regard to its calls for WCD to:

- adopt a catchment approach; - avoid a universal solution to dam construction; - focus on general policy guidelines.



BUT, the report does not go far enough towards stating clearly the areas in which WCD should show leadership by developing and adopting guidelines relating specifically to environmental impacts. It also fails to address WWF�s calls to support the development and transfer of the most appropriate and efficient technologies, or the estimation and recovery of full costs. o) Comments by By Tor Ziegler & Hans Olav Ibrekk 1) General Next to the resettlement issue, - ecosystem functions, and particularly the question of Operationalising environmental reserve and maintenance flows, is one of the most contentious issues related to dams. For the operational life-span of the dam, this is probably the issue where the need for a sound decision-process, knowledge-base for trade-offs and management of water allocations is most needed, and where science is yet in its infancy. The draft report provides considerable perspective on the ecosystem functions related to dams, and how these concerns should be integrated into decision-making, implementation and operation of dams. In our view, the paper still needs considerable work if it is to provide balanced state-of-the-art knowledge of this issue in a compressed form (as one may regard about 80 pages or less to be for such a comprehensive subject). Parts of the report still has character of a compilation of excerpts from submissions and not a good enough attempt to carefully review these and to prioritise them. Most important are however the RECOMMENDATIONS made in the report, see below. In its approach, the paper emphasises maintaining biodiversity (function-oriented). The document is less explicit on the bio-productivity of regulated rivers (service-oriented). To local stakeholders, the latter is the main concern, although the former certainly is one key prerequisite for long-term sustainability of maintaining services. The introductory chapter mentions the "pressures on diversity and productivity of the world's natural resources", and hence provides readers with expectations of more comprehensive treatment of the productivity-issue than can be found in subsequent chapters. It would have been good to expand on the productivity issue to make the report more "down-to-earth". We do however realise the significance of also providing the broader and long-term oriented biodiversity picture in relation to environmental sustainability of water resources management. For the introductory chapter we believe it would have been helpful to readers if a paragraph had pointed to how awareness of environmental qualities and values of societies evolve/change as transition from traditional rural to more market-based, science-oriented and industrial forms of economy takes place. Of course, the Commissioners know this. The definition of the impacted ecosystems (only water or terrestrial or both) is not clear. In some chapters reference is made to impacts due to construction activities on terrestrial ecosystems. The paper would have benefited from a more stringent definition of what components of the various impacted ecosystems it is addressing. A key measure for maintaining ecosystem services is often the flow required in the river for such maintenance (freshwater, wetlands, riparian areas, aquifers and estuaries). The draft report refers to minimum flows, instream flows and environmental flows as more or less synonymous, which they to our understanding are not. A clearer definition of "environmental flows" is missing. Instream flows for instance (as opposed to offstream flows) are what is released into the existing river bed where several "commercial" uses may be nested together with "environmental flows", e.g. like irrigation water on its way to its use area or released for navigational purposes, or as underwater from a hydro-plant. A typology of regulated flows should be provided. It is necessary to distinguish between river-stretches where an environmental flow is maintained solely for environmental reserve, e. g. bypassing



a hydropower diversion, or released for environmental maintenance in a river that is transferred elsewhere by a dam, - and those where multipurpose releases are nested. There is for instance usually a difference between what flows are ideal for angling/fishing, and what flows are required for maintaining biodiversity and fish productivity. Are both the fish rearing and angling requirements implicit in the notion environmental flows? To the dam-owner/operator a release that does not meet the utilitarian purpose for which the investment is made is seen as more expensive than what is a nested release, where trade-offs/compromises are considered less costly. In several cases, the owner and regulatory agencies are also often willing to combine measures like fish-stocking, construction of weirs to create pools, etc for maintenance of ecosystem services, if that can "save" water for primary purpose use. Needless to say this will increasingly be an issue, as private sector more and more comes into ownership and operation of dams. Diversions of water and of inter-basin transfers that certainly have biodiversity, -productivity and other impact implications in the delivering- as well as in the receiving end are dealt with only superficially in the report. There is a reference to that the present paper interphases with other WCD thematics. However, there seems to be little harmonisation between the various thematic papers in the sense that there will be some key issues that several review papers will deal with and overlaps are unavoidable. To what extent have the various authors harmonised the content of their papers? (e.g. the effectiveness of EIA and mitigation measures - an issue which the conclusion of this paper mainly rests on, see below). ----- In the following, comments are first given about the final recommendations, and subsequently on other substance and conclusions of the draft report: 2) The Recommendations to the Commission Chapter 7 concludes that ecosystem goods and services have direct and indirect economic value to local, national and regional economies, and also contingency value. It is of course essential that these concerns be factored into decision-making and management of dams and more broadly into natural resources management. It is also pointed to that total value of such benefits can exceed the benefits from dams and associated investments in agriculture. The latter must more often be the exception than the rule, in which case the dam should not be built. It is also pointed to how governments widely have committed to international agreements on biodiversity. It may be argued to what extent specific sites, stretches or rivers are critical, and to what extent specific dam options lead to conversion or degradation of critical natural habitats. Some countries have adopted the principle of maintaining some rivers (and their biodiversity and habitats) intact, while they develop others "harder", still observing that stakeholders dependent on ecosystem services are considered through compensation or mitigation measures. In the trade-off decision processes, our experience is that productivity counts more than diversity, in spite of their interdependence. This is not likely to change before a real compelling reason can be given that will convince investors �why biodiversity in our project?�. Biodiversity "sells" best at the national regulatory level, less well at the local level where it is productivity that is the concern of stakeholders. Today, there is general consensus on that dams have a number of adverse impacts that to varying degrees have become manifest in economic and social costs. It is not so surprising that there in much of the dam building sector globally is fairly widespread agreement as to the reality of these impacts



and costs. Surprising to some may be that there is convergence among champions in the different camps in the dams debate of how to reach better decision-making and management of dams. It is in practical implementation that capacity to do what is "right" first and foremost is lacking. The review succeeds in raising a general understanding and awareness of that dam building presents a threat to the world's freshwater biodiversity. It would have been good if one ore two regions of the world had been analysed (even cursory) to even better document such cumulative effects. In the case of Sweden with all its dams shown in figure 3.7, it is hotly debated between biologists there whether the dams have led to a reduction in biodiversity as such, although there is agreement on that there is a threat and that numbers of e.g. salmon have decreased. However, it is not clear whether this is due to the dams alone, or whether pollution of the Baltic, overfishing in the seas, etc, are contributing and perhaps even more important causes. In relation to finding solutions to dealing with impacts the approaches should indeed be measures for avoidance, mitigation, compensation and restoration as outlined, and that there exist numerous examples of how such approaches can be implemented. In general it can also be agreed that mitigation is particularly problematic or rather, challenging. However, the statement as it reads in chapter 7 is sweeping, and does not reflect best practice examples of which there are a number. There will usually remain residual impacts, but by applying more adaptive approaches to planning and management, setting river use objectives and monitoring adequately as management tools, singular or combined mitigation measures can succeed if objectives are realistic. The 5 bullet-points that outline the prerequisite for good mitigation represent the ideal case, and these conditions unfortunately exist only or mainly in OECD countries as find expression in the moves towards river restoration in these countries. A consorted effort is needed to get developing countries up to this level. However, to write off mitigation altogether in other (developing countries) cases is again a bit sweeping. This does not contradict that alternative approaches as those recommended should also be pursued. The 8 points outlined next represent best practice approach. For clarity, it would be good to add bio-productivity to biodiversity in recommendation no 2. This will "sell" better in developing countries where there in many cases are large poor riparian populations having freshwater products as a basis for their subsistence economies. However good this safeguard approach may be, experience shows that it is in implementation that these ideals often are not complied with. Even more needed than these 8 recommendations is a recommendation for HOW TO START MOVING IN THE DIRECTION INDICATED IN THE PRESENT DRAFT RECOMMENDATIONS. Some of the bullet-points in section 7.3, meant to illustrate operationalisation options, are very good and may indicate how to move in the "right" direction. But there is a danger in becoming to prescriptive, and not reach results on the ground. Would it be an idea to prioritise what are experienced to be the most important points, or should that prioritisation be left to the Commission itself (if they wish to pursue such and approach)? The paper concludes that the effectiveness of mitigation is little and subsequently encourages a strategy of avoidance and minimisation rather than one of mitigation. This is perhaps the most important conclusion in the paper, one which needs to be clearly supported in terms of findings. Mitigation or the Environment Management Plan (EMP) are not flawed as instruments, but they are not being implemented the way they should. Implementation failures do not necessarily lead to the conclusion cited above. By increasing the effectiveness and efficacy of mitigation (institutional capacity, enforcement and all these other catch words) a lot can be accomplished. Some impacts can not be mitigated properly (to make a fire you need to burn something or to make an omelette you have to crush some eggs) but the main decisive question should be if the residual impacts are acceptable or not. Of the 8 recommendations, some also belong in other thematic reviews (e.g. participation, legal framework). The recommendations follow from the paper's general conclusion of avoidance and



minimisation, e.g. the precautionary principle. From applying the recommendations strictly, one might easily conclude that no dams should be built at all. Statements like - wherever possible dams and their impacts should be avoided! - are not helpful as a recommendation! However, the attempt at Operationalising the recommendations are much better although the options could as said have been prioritised. 3) Examining the nature of river basin ecosystems The introduction tells us that 99% of the volume of freshwater is in lakes and 1% in rivers. How much is in reservoirs? (about 0,13 % or 5 500km3 according to numbers found elsewhere in the draft). There could have been a reference to the productivity of biomass in wetlands which represent some of the highest values per m2 in the world. Box 2.2 refers to the economic value of biomes. It should be more correct to term it annual economic rent. This value (if correct) is a moving target, and the �price� is likely to rise with increasing scarcity as these areas are encroached upon further. Question is, what purpose does this illustration serve in practical terms. Theoretical magnetising is sometimes a fad. The categorisation of ecosystem functions into regulation, production, habitat and information functions is helpful. Section 2.3 gets to the crux of the matter: �Understand well the costs and benefits to society in decisions for allocating water to maintain ecosystems or to support agriculture, industry and domestic uses, and recognise that societal values usually undergo changes with time.� 4) Current understanding of the nature of the impact of dams on ecosystems Chapter 3 is quite comprehensive and in the conclusion argues values at the global level. This is an illustration that creates some perspective on the significance of ecosystem goods, but as with the reminder about international conventions it remains rather academic, and raises the question of how to operationalise measures for maintenance of these services in developing and industrialised economies. Hopefully this is what the Commission will do. As earlier stated, in our opinion it would have been even more helpful to have an analysis of cumulative impacts in one or two regions of the world to get more meat on the bone. The closest the report gets is with the reference to molluscs in North America. However, the many cases of adverse impacts sited leaves no doubt that that dams generally impact biodiversity adversely. Although dams undeniably create adverse impacts, there are usually other factors at work with land-use, pollution, etc too. There is no mention of how to separate between such impacts and those created solely by dams. Box 3.2 could have listed Transfers. Table 3.3 with the Petts Categories could also have referred to ice-formation (important in managed flows in northern rivers), groundwater along reservoirs and in riparian areas, and water temperature issues. On Valuation of ecosystems in Chapter 4, the valuation of the global environment is as said interesting from an academic point of view but not very operational. This is a critical issue since we are not able to convince decision-makers, or our colleagues for that matter, to integrate environment based on global importance or ethical values. We often will need hard facts at project level, preferably of a quantitative nature, to be heard. The paper provides little guidance on how to arrive at such figures. Furthermore, the discussion of ethical principles for decision-makers is interesting but probably



deserves a more thorough discussion in another place, probably outside this document, but within the WCD-process. The problem that EIAs are not integrated in the decision-making process or that EMPs are not implemented does not necessarily lead to opting for avoidance. We assume that the EIA paper will adequately address this issue. 5) Status of the approaches to addressing consequences Chapter 5 on responding to the ecosystem impacts of dams aims to bring out where practice and the ability to address consequences of dams on ecosystems stands today. First of all, the sequence: avoid the worst; mitigate and reduce where acceptable; compensate what can not be mitigated; and restore what went wrong; - is advised in all good textbooks today. Again, the crux of the matter is to implement adequate measures and make sure that good intentions work. The draft text largely presents impacts and corresponding mitigation measures without any attempt at prioritising them. The report comes out with a negative assessment of experiences with mitigation, as commented on above. In the conclusion a strategy of avoidance and minimisation rather than one of mitigation is encouraged - �if the aim is to maintain biodiversity, ecosystem functions and services�. We agree on that this should be the strategy if you want to maintain these values 100%. However, in many real-world cases trade-offs are made, and in the assessment of tangible and intangible costs and benefits, it frequently happens that strictly economic/utilitarian benefits are seen as sufficiently large to accept 10- 20-30-50 75-100% reduction, as long as mitigation that reduces impacts or compensation in- or out-of-kind is conceived as adequate by affected stakeholders. One approach that can be used (where capacity is present) is to set an attainable environmental objective/target for mitigation, and commit the dam owner to meet the objective in dam implementation and operation. There exist good practice on this approach. Avoid (no change) Examples of good practice are cited in the text, we have little to add other than that the distinction between what normally is conceived as avoidance and what is mitigation is not well demonstrated in the examples sited for avoidance (e.g. the caribou case). Mitigate or reduce Here the text advises design of operating regime to eliminate, offset or reduce ecosystem impacts to acceptable levels. This is where setting objectives rather than prescribing set �window-dressing� measures that do not work satisfactorily may provide an option that will make dam owners find the best measures to meet objective ( or be penalised). This may in some cases imply application of a combination of mitigation measures if that is most cost-effective. Use of weirs and stream bed engineering to create/improve habitats and use less water was not mentioned among the list of measures mentioned in the text. Compensate The border between mitigation and compensation is blurred and should be defined, (the in-kind compensation measures mentioned contribute to the blur). An out-of-kind type that was not mentioned is creation of a fund for discretionary use by environmental interests. However, it is mentioned elsewhere that reservoir aqua-culture provides ecosystem services/products that even may create considerable development opportunities like in the cases cited from Indonesia. Restore or decommission The Glen Canyon case is mentioned where an estimated 3 million people/ recreationers visit the reservoir every year. It would be good if the concluding statement: �It is perhaps because of these



considerations that all dams removed today are relatively small� could be elaborated a bit on. It sounds like the recreational attraction is the only use that justifies many dams today!!! Effectiveness of mitigation Like farming, effectiveness of cultivating soil or mitigating impacts on ecosystems depends very much on the skill of those who design and implement mitigation measures. One important factor is indeed to adequately formulate environmental clauses (or objectives) and compliance with these, and the cost of mitigation should be justified by the outcome. Bank reviews and mitigation effectiveness With due respect for the kind of multilaterals the World Bank represents, we are a typical develop-new-projects institution with little experience in long-time operation of dams. This may account for some of our limitations in ensuring compliance, especially as there are no carrots or sticks available for enforcement. Fish passes need skilled expertise and management (design, monitoring and feed-back) to work. Scientific uncertainty/ It may be of interest to draw attention to experimental and adaptable rules of operations for dams that are used for some dams on the northern hemisphere, and like the artificial floods in the Manantali dam (Senegal) and several other cases in Africa. The northern hemisphere cases rely heavily on stakeholder consultation, monitoring and feed-back in order to make amendments for scientific uncertainty in assessing flow requirements. DAMS ARE NOT LIKE CLOCKS YOU CAN SET AND WIND, AND THAT WILL RUN ACCORDING TO �SCHEDULE� REGARDLESS. MOST LARGE DAMS REQUIRE CAREFUL MANAGEMENT THROUGHOUT THEIR LIFETIMES! Capacity-building for planning, design, implementation and operation is absolutely essential, but takes time and resources for building adequate skills. 6)Assessment of the areas of convergence and divergence on issues To our understanding, the principles and guidelines of IEA, ICOLD and the World Bank show considerable convergence. Bilateral and national EIA guidelines are equally important and could have been elaborated more on. The role of the private sector, commercial banks' EA-systems etc. are not addressed. 7) Concluding remarks This draft review is very interesting reading, has brought a considerable number of interested people on board the WCD process, and we trust, will inform the Commissioners (who themselves have a lot of practical knowledge) on the very important issue of ecosystem functions and environmental restoration, - enough for them to make their own wise recommendations. In order for this material to be published as a knowledge base and guidance for future decision-making (if that is contemplated), it will in our view still need significant firming up/quality control/balancing, as well as further peer review and editing, before any publication is advisable. p) Comments by Takehiro Nakamura I have earlier commented on the Instream flows paper and on the three draft reports prepared under the UNEP/IUCN sub-project (for improvements of the drafts prepared by IUCN consultants). Therefore, my comments below are related more to integrity of the whole report, or on the contents concerned about major technical issues.



1. The issue of �first filling� is not given sufficient attention. During the period of first filling, strict

control of how much water is used for filling and how much is used for release to downstream to maintain downstream ecosystem and flood regimes. The condition during this period is more severe than the operation period of dams. Further, during this period, nutrients and other materials from the areas being inundated enter the reservoir water body. Especially, under the tropical conditions, poor operation of first filling may lead to eutrophication and/or outbreak of malaria, bilharzia and other diseases.

2. Public health issues such as malaria and bilharzia are considered as one of environmental health

issues in relation to dams. 3. In association with the water level changes in both upstream and downstream of dams,

groundwater tables fluctuate. This has impacts on the �river basin ecosystem� (part of the title of Chapter 2). This aspect has not been sufficiently covered.

4. The term �wetlands� are used in several places in Chapter 2. In the Ramsar definition, reservoirs,

lakes and rivers themselves can be wetlands, and special care should be taken of the terminology. 5. In terms of relevant data base, UNEP and International Lake Environment Committee (ILEC) has

developed a survey of World Lakes, which also includes limited numbers of dams. Further, UNEP and WCMC recently published a book, �Freshwater Biodiversity: a preliminary global assessment�, which can also serve as data/information sources.

6. Another environmental issue relevant to dams is impacts of floods on riverine and floodplain

ecosystems, caused by uncoordinated release of reservoir waters. In particularly, when there are a series of dams along a river, uncoordinated release of water may cause floods or inundation in some part of the river, which might have impacts on floodplain ecosystem. Such a case was observed in the Aral Sea Basin.

7. Accumulation of pollutants adsorbed to sediments settling in the reservoirs may take place,

providing the reservoirs with the nature of �chemical reservoirs�. Depending on equilibrium between the pollutant concentration in the sediments and that in water, such pollutant may be released to water.

8. We have observed the cases (such as San Francisco River or Senegal River) where construction of

dams impacts not only delta areas but also a wider range of coastal areas, due to change in balance between freshwater input and coastal current.

9. As indicated, there are mitigation measures and compensations schemes. What is the general

trend of international financial mechanism towards development of dams and towards mitigation measures in monetary term? What financing mechanism goes towards what areas?

10. River basin management is a mechanism not only for sediment control but also for cost-efficient

and effective water management mechanism, if it is carried out in an integrated manner. Under the river basin management schemes, dam options can be compared with other options.

Documents

Dams, Ecosystem Functions and Environmental Restoration · • Lotek Engineering • Manitoba Hydro • National Wildlife Federation, ... Reviews I.1 Social Impacts of Large Dams