Living in the Water Living in the Water EnvironmentEnvironment

How can we protect ecosystems and better How can we protect ecosystems and better manage and predict water availability and manage and predict water availability and quality for future generations, given changes quality for future generations, given changes to the water cycle caused by human activities to the water cycle caused by human activities and climate trends?and climate trends?

Jeff Dozier, UC Santa BarbaraJeff Dozier, UC Santa Barbara

WATERSWATERS = WATWATer and EEnvironmental RResearch SSystems

Dynamic nature of atmospheric waterDynamic nature of atmospheric water

• Atmospheric precipitable water ~25 mm, annual precipitation/evapotranspiration ~1000 mm– So atmospheric residence time for water ~9

days– Climate models show much greater

agreement on temperature in a 2x CO2 environment than on precipitation• Coquard et al. [2004, Climate Dynamics]: 15 GCMs,

downscaled to western U.S., don’t agree on the sign of a future precipitation change

Water-energy nexusWater-energy nexus

• Including hydroelectric, US water withdrawals for energy are 80% of US agriculture’s withdrawal– Much non-consumptive (although

perhaps at higher temperature)– But consumptive withdrawals for energy

are ~20% of non-agricultural use

• Life cycle assessment of biofuels?

Coupled human-environment problemCoupled human-environment problem

• Incentives and institutional/governmental arrangements to manage water?– Not much behavioral information about response

to drought in early 1990s– Why is worldwide foreign aid directed toward

water quality ~4% of the bottled water industry?

• Water demand not as well understood as supply

• Value (saliency) of information – what decision can you make based on information, vs based on statistics?

Grand ChallengesHydrologic Sciences:

Closing the water balance

Social Sciences: People, institutions, and their

water decisions

Engineering: Integration of built environment

water system

Measurement of stores, fluxes, flow paths and

residence times

Water quality data throughout natural and

built environment

Synoptic scale surveys of human behaviors and

decisions

How is fresh water availability changing, and

how can we understand and predict these changes?

How can we engineer water infrastructure to be

reliable, resilient and sustainable?

How will human behavior, policy design and institutional decisions affect and be affected by changes

in water?

Resources needed to answer these questions and transform water science to address the Grand Challenges

Status as NSF MREFC “horizon” Status as NSF MREFC “horizon” project project ((Major Research Equipment and

Facilities Construction))• This year: produce science plan

– 15th May, in review by NRC, briefing on 15th June

• Conceptual design (2 years)

– Requirements definition, prioritization, review

– Identify critical enabling technologies and high risk items

– Top-down parametric cost and contingency estimates and risk assessment

– Draft Project Execution Plan

• Preliminary design/ readiness stage (2-3 years)

– Site selections in this stage

• National Science Board approves – final design

• Construction and Construction and

CommissioningCommissioning

– From MREFC accountFrom MREFC account

• Operation and maintenance

– From Directorates

• Renewal/termination7

Context: the NSF budgetContext: the NSF budget

Account FY09 ($M) Stimulus ($M)

FY10 request

($M)

Research & related activities

$5,183 $2,500 $5,733

Education & human resources

845 100 858

MREFC 152 400 117

Operations & management 294 318

National Science Board 4 4

Inspector General 12 2 14

TotalTotal $6,490$6,490 $3,002$3,002 $7,045$7,045

8

NSF FY2010 Budget Summary http://tinyurl.com/qebc7u

ExamplesExamples

9

Accumulation and ablation inferred from snow pillow Accumulation and ablation inferred from snow pillow data, Tuolumne Meadows (TUM) and Dana Meadows data, Tuolumne Meadows (TUM) and Dana Meadows (DAN)(DAN)

10

Snow Redistribution and DriftingSnow Redistribution and Drifting

12

13

Global hypoxiaGlobal hypoxia

14

Hypoxic volume per unit N is Hypoxic volume per unit N is increasingincreasing

16

Total phosphorus and total suspended Total phosphorus and total suspended solids loadingsolids loading• TSS and TP

from turbidity using surrogate relationships

• ~50-60% of the annual load occurs during one month of the year

• Provides information about flow pathways Horsburgh 2009

Effects of sampling frequencyEffects of sampling frequency

Spring 2006

Urban stormwater and wastewaterUrban stormwater and wastewater

Field scale green infrastructure (e.g.,

EPA’s Urban Watershed Research Facility in

Edison, NJ)

Field scale sewershed management

Alcosan Wastewater Treatment Facility

Pittsburgh Drinking Water Treatment Facility

Ohio River BasinDrinking Water Intake Sensors

Pittsburgh RegionUrban Water Treatment Plants

Pittsburgh RegionWastewater Collection System and Drinking

Water Distribution System

Pittsburgh Drinking Water Treatment FacilityNested sensor design Nested sensor design for drinking water for drinking water

systemssystems

19

Network considerationsNetwork considerations

20

Science progress vs funding Science progress vs funding (conceptual)(conceptual)

21

The water information value ladder

Monitoring

Collation

Quality assurance

Aggregation

Analysis

Reporting

Forecasting

Distribution

Done poorly

Done poorly to well

Generally done well, by many groups, butcould be vastly improved with new IT

>>> Incr

easing v

alue >

>>

Integration

Data >

>> Info

rmatio

n >>> In

sight

Slide Courtesy CSIRO, BOM, WMO

The data cycle perspective, from The data cycle perspective, from creation to curationcreation to curation

23

• The science information user:— I want reliable, timely, usable

science information products• Accessibility• Accountability

• The funding agencies and the science community:— We want data from a network of

authors• Scalability

• The science information author:— I want to help users (and build my

citation index)• Transparency• Ability to easily customize and

publish data products using research algorithms

The Data CycleThe Data Cycle

Collect

Store

Search

Retrieve

Analyze

Present

Organizing the data cycleOrganizing the data cycle

• Progressive “levels” of data– EOS, NEON, WATERS

Network0 Raw: responses directly

from instruments, surveys1 Processed to minimal

level of geophysical, engineering, social information for users

2 Organized geospatially, corrected for artifacts and noise

3 Interpolated across time and space

4 Synthesized from several sources into new data products

• System for validation and peer review– To have confidence in

information, users want a chain of validation

– Keep track of provenance of information

– Document theoretical or empirical basis of the algorithm that produces the information

• Availability– Each dataset, each

version has a persistent, citable DOI (digital object identifier)

24

Observatories and facilitiesObservatories and facilities

25

Structure: similar environmental Structure: similar environmental themes for sampling designthemes for sampling design• Objectively identify “similar” thematic places

that are comparable and can be intensively studied at a few (1-4) “observatories” in each– Capture the diverse hydrologic, engineering and

social conditions that exist across the U.S.– Set of variables that quantify hydrologic setting,

both physical and human-influenced

• Example: ISODATAclustering based on theHuman-Influenced WaterEnvironmentClassification(HIWEC)

Hutchinson et al. 200926

Human-impacted water environment classes USGS hydrologic regions

US EPA ecoregions NEON domains

Information products, hydrologic Information products, hydrologic exampleexample

Example of virtual sewershed rainfall sensorExample of virtual sewershed rainfall sensor

NEXRAD reflectivity data on local server

Temporal averaging of sewershed rainfall rate produces 15 minute sewershed average rainfall data

Use Z-R relationship to convert average reflectivity (Z) into sewershed rainfall rate (R)

Spatially interpolate NEXRAD reflectivity from polar grid to polygon sample points. Then average over sample points.

Uniformly sample area within polygon on 500 m by 500 m grid

Geospatial sewershed polygon encoded as KML

<Polygon><tessellate>1</tessellate><outerBoundaryIs><LinearRing><coordinates>-87.8891714325616,41.8093419197642,0-87.8858482809939,41.7695567017372,0 -87.8284450634905,41.7684213670785,0 -87.8462880396867,41.8051753689394,0 -87.8891714325616,41.8093419197642,0 </coordinates></LinearRing></outerBoundaryIs></Polygon>

Community workflow publication & provenance enable customized virtual sensors using experimental algorithms and alternative data/resolutions

Information products, social science Information products, social science exampleexample

30

Data available operationally

now, but dispersed

WATERS Network

original data

Resulting high-level

data products

Census-derived

Demographics

Agency Budgets, Policies,

and Authorities

Water Uses and

Withdrawal Rates

Catalog of watershed-

level interest groups and politicians

Spatially distributed

politics

Spatially distributed water user

data

Interpolate network data

with geospatial

data

Interpolate political data

with geospatial

data

Survey data on agency and key

actor interaction;

experimental data on

management decisionsSpatially

distributed information

and management

networks

Survey data on water

information sources,

attitudes, values, and

behaviors; experimental

data on decisions

Spatial Distribution of Socio-

Political Water

Systems

Geography of

Water Informati

on Networks

Geography of Water

Information Sources

Attitudes, Behaviors,

and Decisions

Interpolate user data

with geospatial

data

Network of Network of sites, sensors, facilities, sites, sensors, facilities, surveys, tools, surveys, tools, and and people,people, and the and the links among themlinks among them• Within a specific observational or experimental facility

– scale through a sampling design, which may contain elements of nested and gradient-based design

• In multiple places within a water environment class– extend from intensively observed places to similar environments– validate our capability to extend the knowledge gained from the

detailed sites or facilities to other places in the same class

• Throughout the set of water resource classes– test the generality of proposed hypotheses across

heterogeneous environments– provide multiple lines of evidence to constrain theory– accumulate community data, models, and information to

support multiple investigators working in interdisciplinary water science.

• Mesh with existing national-scale data– national water survey of human behavior– remotely sensed observations of continental or global extent– existing data of operational agencies

31

Test bed HISServers

Central HIS servers

ArcGIS

Matlab

IDL, R

MapWindow

Excel

Programming (Java, C, VB,

…)

Desktop clients

Customizable web interface (DASH)

HTML - XMLW

SD

L - SO

AP

Modeling (OpenMI)

Global search (Hydroseek)

Water Data Web Services, WaterML

HIS LiteServers

External data

providers

Deployment to test beds

Other popular online clients

ODM DataLoader

Streaming Data Loading

Ontology tagging

(Hydrotagger)

WSDL and ODM registration

Data publishing

QA/QC

Server config tools

HIS CentralRegistry & Harvester

Hydrologic Information System Service Oriented Architecture

Why now?Why now?

• Addresses Grand Challenges in environmental research and integrates natural, engineering, and social science

• Water couples humans and natural systems as a balancing mechanism between human activity and sustainability

• Given the current state of water issues, the need to understand and predict is urgent

• Other federal agencies are making investments, so leveraging opportunities exist over the next decade

• Because of community readiness and technological advances, the ability to address this need (finally) exists 33

What Next?What Next?

• Thematic division (the many-colored map)– Multi-disciplinary workshop to

get to community consensus

• Integration with mission agencies

• Review enabling technologies and high-risk items– Sensors, satellites, surveys

• Review & select models to use in network design– Review datasets, models, and

experience from testbeds & CZO investigations

• Education and outreach plan

• Cyberinfrastructure– Framework and facility to

support higher-level data products

– Community network– Identify research &

development needs for WATERS Network

• Execution plan and strategy for selection in Preliminary Design phase– Observatories, facilities, surveys

• Cost and contingency estimates

• Examine options and identify management structure for MREFC

34