United States
Environmental Protection
Agency
Office of Research and
Development
Washington, DC 20460
EPA/600/R-01/014
April 2001
www.epa.gov/ncerqa
f/EPA Proceedings
2001 Water and Watersheds
Progress Review
•i&
April 17-19, 2001
San Francisco, California
AUC^NTEB FOR ENVIRONMENTAL RESEARCH
-------
United States Office of Research and EPA/600/R-01/014
Environmental Protection Development April 2001
Agency Washington, DC 20460 www.epa.gov/ncerqa
vvEPA Proceedings
2001 Water and Watersheds
Progress Review
April 17-19, 2001
San Francisco, California
iil
*
_o
'm
•
NATIONAL CENTER FOR ENVIRONMENTAL RESEARCH
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Table of Contents
Introduction vii
Section 1. Projects Initiated With Fiscal Year 1999/2000 Support
The Impact of Lawn Care Practices on Aquatic Ecosystems in Suburban Watersheds 3
Kevin L Armbrust, Larry Shuman, Judith Meyer, Marsha Black, Raymond Noblet,
Andrew Keeler, Ted Gragson, James B. Williams, Dee West
PULSES—The Importance of Pulsed Physical Events for Watershed Sustainability
in Coastal Louisiana 4
John Day, Jaye Cable, Dubravko Justic, Brian Fry, Paul Kemp, Enrique Reyes,
Paul Templet, Robert Twilley
Linking Environmental and Social Performance Measurement for Management at National
and Watershed Levels: Modeling and Statistical Approaches 6
Scott Farrow, Mitchell Small, Tim Bondelid, Andrew Solow, George Van Houtven,
James Sinnott, Martin Schultz
Alternative Urbanization Scenarios for an Agricultural Watershed: Design Criteria, Social
Constraints, and Effects on Groundwater and Surface Water Systems 8
Richard C. Lathrop, Kenneth W. Potter, Jean M. Bahr, Kenneth R. Bradbury, Steven R. Greb,
James A. LaGro, Jr., Edward B. Nelson, Peter Nowak, Joy B. Zedler
An Integrated GIS Framework for Water Reallocation and Decisionmaking
in the Upper Rio Grande Basin 10
Olen Paul Matthews, David Brookshire
The Spatial Pattern of Land Use Conversion: Linking Economics, Hydrology,
and Ecology To Evaluate the Effects of Alternative Future
Growth Scenarios on Stream Ecosystems 11
Margaret A. Palmer, Nancy E. BockstaelGlenn E. Moglen, N. LeRoy Poff,
James E. Pizzuto, Cameron Wiegand, Keith Van Ness
Integrating Coral Reef Ecosystem Integrity and Restoration Options With Watershed-Based Activities
in the Tropical Pacific Islands and the Societal Costs of Poor Land-Use Practices 13
Robert H. Richmond, Michael Hamnett, Eric Wolanski
Identification and Control of Nonpoint Sources of Microbial Pollution in a Coastal Watershed 15
Brett Sanders, Stanley Grant, Alex Home, Robin Keller, Mark Sobsey
Strategic Renewal of Large Floodplain Rivers: Integrated Analysis 16
.Richard E. Sparks, Deborah Beal, John B. •Braden, Misganaw Demissie, Andrew M. Isserman,
Douglas M. Johnston, Jungik Kim, Yanqing Lian, Da-Mi Maeng, Zorica Nedovic-Budic,
Daniel Schneider, Diane M. Timlin, David C. White
Watershed-Scale Assessments of E. coli Contamination Implications of Source
Identification for Public Policy Debate 17
Ronald Turco, S. Brouder, C. Nakatsu, A. Bhunia, J. Frankenberger, J. Harbor, G. Thomas
The Office of Research and Development's National Center for Environmental Research iii
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Table of Contents (continued)
An Acre an Hour: Documenting the Effects of Urban Sprawl in a Model
Watershed Near PhHadelphia, Pennsylvania 18
Claire Welty, Susan S. Kilham, Aaron I. Packman, Robert J. Brulle
Section 2. Projects Initiated With Fiscal Year 1998 Support
Development and Testing of a Decision Support System for River Rehabilitation 21
J. David Allan, Gloria Helfand, JoanNassauer
An Integrated Systems Approach to Watershed Restoration With Community Involvement
Applied to a Small Rural Watershed 22
J. Boll, J.D. Wuljhorst, S. Chen, C.O. Stockle, D.K. McCool, D.C. Brown,
D. Fetch finger, A.J. Vitale
Integrating Ecological, Economic, and Social Goals in Restoration Decisionmaking 24
John Bolte
Social Impact Assessment of Human Exposure to Mercury Related to Land Use
and Physicochemical Settings in the Mobile-Alabama River Basin 25
Jean Claude Bonzongo, Eric E. Roden, Milton G. Ward, C. Hobson Bryan,
W.B. Lyons, Indrajeet Chaubey
Applying the Patuxent and Gwynns Falls Landscape Models To Designing a Sustainable
Balance Between Humans and the Rest of Nature 27
Robert Costanza, Alexey Voinov, Roelof Boumans, Tom Maxwell, Ferdinando Villa,
Helena Voinov, Joshua Farley
Understanding the Social Context for Ecological Restoration in Multiple-Ownership
Watersheds: The Case of the Cache River in Illinois 28
Steven Kraft, Christopher Lant, Jeffrey Beaulieu, Leslie Duram, J.B. Ruhl, David Bennett,
Jane Adams, John Nicklow, Tim Loftus
Restoring and Maintaining Riparian Ecosystem Integrity in Arid Watersheds: Meeting
the Challenge Through Science and Policy Analysis 30
Thomas Maddock, III, Kathryn Baird, Victor Baker, Bonnie Colby,
Robert Glennon, Julie Stromberg
Development of an Urban Watershed Rehabilitation Method Using Stakeholder Feedback
To Direct Investigation and Restoration Planning 31
Marty D. Matlock, Charles D. Samuelson, William H. Neill, Tarla Rai Peterson,
Ann L. Kenimer, Guy D. Whitten
Combining Economic and Ecological Indicators To Prioritize Wetlands Restoration
Projects Within a Spatial GIS Framework 32
James J. Opaluch, MarisaJ. Mazzotta, Peter August, Robert Johnston, Frank Golet
Integrating Science and Technology To Support Stream Naturalization 33
Bruce L. Rhoads, David Wilson, Edwin E. Herricks, Marcelo Garcia, Rebecca Wade
iv The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Table of Contents (continued)
Selection of Wetland Restoration Sites in Rural Watersheds To Improve Water
Quality: Integrating Ecological and Economic Approaches 34
Curtis J. Richardson, Randall A. Kramer, NealE. Flanagan
When Do Stakeholder Negotiations Work? A Multiple Lens Analysis of Watershed
Restorations in California and Washington 35
Paul Sabatier, William Leach, Neil Pelkey
Integrating Models of Citizens' Perceptions, Metal Contaminants, and Wetlands
Restoration in an Urbanizing Watershed 36
Robert K. Tucker, George S. Hawkins, Peter R. Jaffe, Kerry Kirk Pflugh,
Branden B. Johnson
Changes in River-Land Uses and Management: Implications for Salmonid Habitat
Restoration in the Lower Cedar River, Washington 38
Robert C. Wissmar, Thomas M. Leschine, Ray Timm, David Fluharty, John Small
Section 3. Projects Initiated With Fiscal Year 1997 Support
Community Values and the Long-Term Ecological Integrity of Rapidly Urbanizing Watersheds 41
M. Bruce Beck, A.K. Parker, T. C. Rasmussen, B, C. Patten, K. G. Porter,
B.G. Norton, A. Steinemann
Connecting Ecological and Social Systems: Watershed Research Relating Ecosystem
Structure and Function to Human Values and Socioeconomic Behaviors 43
Gaboury Benoit, S. Kellert, M, Ashton, P. Barten, L. Bennett, D. Skelly, S. Anisfeld
Social and Ecological Transferability of Integrated Ecological Assessment Models 45
Linda A. Deegan, James Kremer, Thomas Webler
From Landscapes to Waterscapes: An Integrating Framework for Urbanizing Watersheds 47
P. Diplas, E.F. Benfield, DJ. Bosch, W.E. Cox, R. Dymond, D.F. Kibler, V.K. Lohani,
S. Mostaghimi, P.S. Nagarkatti, D.J. Orth, L.A. Shabman, K. Stephenson
Conversion of Science Into Management Decisions at Lake Tahoe (CA-NV) 48
Charles R. Goldman, John E. Reuter, S. Geoff Schladow, Alan Jassby, M. Levant Kavvas,
Alan C. Heyvaert, Theodore J. Swift, Jennifer E. Coker
An Integrated Ecological and Socioeconomic Approach To Evaluating and Reducing
Agricultural 'Impacts on Upper Mississippi River Watersheds 50
PrasannaH. Gowda, Roger J. Haro, TedL. Napier
Nutrient Sources, Transformations, and Budgets at the Watershed Scale
in Ipswich River, Massachusetts 51
Charles S. Hopkinson, E. Rastetter, J.V. Vallino, M. Williams, R.G. Pontius
Linking Watershed-Scale Indicators of Changes in Atmospheric Deposition
to Regional Response Patterns 52
J. Kohl, I. Fernandez, J. Rubin, J. Cosby, S. Norton, L. Rustad, D. Mageean, P. Ludwig
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Table of Contents (continued)
REKA, a New Comprehensive Watershed Management System 54
C. Gregory Knight, Jeffrey J. Carmichael, Heejun Chang, Dimitar Dimitrov,
Barry M. Evans, James M. Hamlett, Todor N. Hristov, Vania D. loncheva,
Ivan I. Nikolov, Marieta P. Staneva, Petko S. Varbanov
Coping With Nature: Accepting Risk, Adopting Technology, and Assuming Ignorance 56
James McManus, CourtlandL. Smith, Jesse Ford, PaitlD. Komar,
Debbie Colbert, Michael Styllas
Ecological Risks, Stakeholder Values, and River Basins: Testing Management
Alternatives for the Illinois River 57
MarkMeo, Baxter Vieitx, Blake Pettus, Edward Sankowski, Robert Lynch, WillFocht,
Keith Willett, Lowell Caneday
Balancing Risks of Flood Control and Ecological Preservation in Urban Watersheds 59
Vladimir Novotny, D. Clark, R. Griffin, A. Bartosovd, D. Booth
Impact of Social Systems on Ecology and Hydrology in Urban-Rural
Watersheds: Integration for Restoration 60
Steward T.A. Pickett, J.M. Grove, l.W. Band, K.T. Belt, G.S. Brush, W.R. Burch, Jr.,
M.L. Cadenasso, J.M. Carrera, G.T. Fisher, P.M. Groffinan, R.V. Pouyat, W.C. Zipperer
Index of Authors 61
vi The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSFAJSDA Water and Watersheds Progress Review
Introduction
The Environmental Protection Agency/National Science Foundation/United States Department of Agriculture
(EPA/NSFAJSDA) Water and Watersheds competition is one of four special extramural awards competitions
supported by the EPA and the NSF under a partnership for environmental research initiated in 1994. USDA began
sponsorship of the program during the 1998 competition.
The Water and Watersheds competition emphasizes interdisciplinary research that adopts a systems approach to
water and watershed issues. The goals are to: (1) develop an improved understanding of the natural and anthro-
pogenic processes that govern the quantity, quality, and availability of water resources in natural and human-
dominated systems; and (2) improve the understanding of the structure, function, and dynamics of terrestrial and
aquatic ecosystems within watersheds.
The 1995 Water and Watersheds competition reviewed 656 proposals and made 36 awards. In 1996, the focus was
narrowed to truly interdisciplinary research, and as a result, the 1996 competition reviewed 249 proposals and
made 12 awards. The announcement was narrowed further in 1997, primarily in response to concerns about the
competition's low rate of proposal success. Proposals were required to integrate physical, ecological, and social
science research. For the first time, investigators were encouraged to adopt a community-based approach. The 1997
competition, with an emphasis on urban/suburban research, reviewed 128proposals and made 13awards.In 1998,
the topical emphasis shifted to watershed restoration, a total of 125 proposals were reviewed, and 14 awards were
made. In 1999/2000, an additional 12 grants were funded.
The abstracts in this volume are organized alphabetically within three sections that correspond to the year of award
and reflect all active grants. The most recent awards (FY 1999/2000) appear in the first section. These projects
have only 1 year of research to report, and consequently, these abstracts indicate goals and plans rather than results.
The FY 98 cohort of projects appears next. These abstracts report early findings and describe plans for future years.
The projects that were initiated with FY 97 support are in the third section and will report results based on several
years of research.
Progress reviews, such as this one, will allow investigators to interact with one another and to discuss progress and
findings with program officers and other federal officials interested in the program.
Any opinions, findings, conclusions, or recommendations expressed in this report are those of the investigators
who participated in the research and in the Progress Review meeting, and are not necessarily those of the EPA,
the NSF, or the USDA. For further information on the EP A/NSF/USDA Water and Watersheds competition, please
contact the Program Coordinators: Ms. Barbara Levinson, EPA, levinson.barbara@epa.gov; Dr. Douglas James,
NSF, ldjames@nsf.gov; and Dr. Michael O'Neill, USDA, moneill@reeusda.gov.
The Office of Research and Development's National Center for Environmental Research vii
-------
Section 1.
Projects Initiated With Fiscal Year
1999/2000 Support
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
The Impact of Lawn Care Practices
on Aquatic Ecosystems in Suburban Watersheds
Kevin L. Armbrust', Larry Shuman', Judith Meyer2, Marsha Black3, Raymond Noblet4, Andrew Keeler 5,
Ted Gragson6, James B. Williams7, and Dee West*
' Department of Crop and Soil Science, University of Georgia, Griffin, GA;}Institute of Ecology, 3Department
of Environmental Health Sciences,4Department of Entomology, ^Department of Agricultural Economics,
Department of Anthropology, University of Georgia, Athens, GA; ?Peachtree City Developmental Services,
Peachtree City, GA; sAlpharetta Environmental Services, Alpharetta, GA
The working hypothesis of this project is that
homeowner beliefs, values, and socioeconomic sta-
tus will determine loads and ecological impacts of
turf-care chemicals (pesticides and nutrients) in aquatic
ecosystems in suburbanized watersheds. The objec-
tives of this research project are to: (1) measure the
loading to streams and temporal trends in concen-
trations of turf-care products and biological indica-
tors of stream ecosystem health in creeks receiving
stormwater drainage from residential neighborhoods
with different socioeconomic statuses; and (2) com-
pare the cultural models of lawn and lawn care held
by "experts" and "homeowners" to determine their
points of commonality and divergence, and establish
the nature of variation.
This investigation integrates the physical, eco-
logical, and social sciences to understand the impacts
of residential lawn care chemicals on aquatic eco-
systems at six locations in Metropolitan Atlanta as
well as at two locations on a golf course. A team of
university researchers and community-based stream
monitoring programs will monitor pesticide and nutri-
ent loads leaving residential neighborhoods and resi-
due levels in receiving water and sediment of streams
(physical/chemical); monitor aquatic organism pop-
ulations and multiple biological indices in these
streams to determine the impact of lawn care prac-
tices (ecological); and work with selected homeown-
ers in these neighborhoods to understand their gen-
eral beliefs and values of lawns and the lawn care
practices they display to assess the impact "expert"
groups have in forming these beliefs (social).
Confirmatory laboratory investigations of bio-
logical effects from individual and multiple stressors
will provide added confidence that observed in-
stream toxicity can be tracked to a particular chemi-
cal or chemicals. The results of this research will be
communicated to interested citizens via research ex-
hibits and educational materials produced by com-
munity-based environmental protection programs.
During the past summer, the State of Georgia
has experienced one of its worst droughts on record,
and extreme water restrictions were placed on the ir-
rigation of both golf courses and residential lawns.
However, water and sediment samples collected on a
monthly basis since July have contained detectable
residues of pesticides and pesticide degradation pro-
ducts associated with landscaping as well as nutri-
ents at all sites. Effects thus far observed hi leaf pack
decay rates and mussel biological indices have not
been found to be different between creeks draining
suburban areas and reference creeks.-
Laboratory testing has indicated that certain in-
secticides detected in the streams were more toxic to
black fly larvae and to mussels in mixtures than
when added individually. Preliminary data from sur-
veys of homeowners have shown that the demogra-
phics of the study location are typical of most other
communities of similar sizes in the Metropolitan At-
lanta area.
Due to drought conditions there were few rain-
storms, generating virtually no runoff events during
the periods of highest lawn care chemical use, which
is atypical of what would occur in a normal year.
The levels of the pesticides and nutrients detected
were below any level of concern, and no significant
effects were observed in any organism. Most impacts
to leaf-degrading fungi in different streams were at-
tributed to leaf pack burial by sediment.
Monthly sampling of water for pesticides and
nutrients and sediment for pesticides and metals will
continue during the coming year, as will additional
field monitoring experiments with mussels, aquatic
insects, and leaf packs. Additionally, storm event
sampling will occur, and more detailed surveys of
individual homeowner lawn care practices will be
conducted. Select homeowners will be keeping lawn
care diaries, and followup surveys will be conducted
to assess homeowner lawn care attitudes and prac-
tices.
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
PULSES—The Importance of Pulsed Physical
Events for Watershed Sustainabilitv in Coastal Louisiana
John DayJ, Jaye Cable', Dubravko Justic', Brian Fry', Paul Kemp 2, Enrique Reyes', Paul Templet3,
and Robert Twilley4
'Department of Oceanography and Coastal Sciences, 2Centerfor Coastal, Energy, and Environmental
Resources, 3Institute for Environmental Studies, Louisiana State University, Baton Rouge, LA; Department of
Biology, University of Louisiana, Lafayette, LA
Riverine inputs to coastal wetlands, floodplains,
and marshes are important to long-term ecological pro-
ductivity and development of watershed resources. In
many cases, levees and dams constructed during the
past 100 years have effectively isolated rivers from
their natural connections to adjacent floodplain and
deltaic wetlands. To help revitalize these productive
systems, the ecological restoration of historical river-
floodplain connections is being attempted.
The PULSES Project focuses on evaluating the
effects of pulsed river inputs in one such coastal wa-
tershed, the Breton Sound Watershed, just south of
New Orleans. In this area, Mississippi River water is
introduced through gated river diversion structures at
the head of the estuary at Caernarvon, LA. Diversions
have been ongoing for a decade (since 1991) at Caer-
narvon, but have received little scientific attention and
study.
The physical science objectives of this research
project are to evaluate marsh accretionary responses to
two different levels of river pulsing, Ix (14 nrVs) and
3 3-16x (184-227 m3/s). The diversions will be exper-
imentally conducted in 2-week episodes within the
winter/spring operating schedule of the Caernarvon
structure that is controlled by the Louisiana Depart-
ment of Natural Resources. Figure 1 shows replicated
high flow and low flow diversions scheduled for the
winter and spring of 2001; similar diversions are plan-
ned in 2002. In addition to marsh accretion studies,
historical down-core studies will evaluate the effects
of the great 1927 flood event (approximately 650x
base flow of 14 mVs) at this site.
The ecological science objectives are to evaluate
marsh and phytoplankton plant growth responses to
river pulses, and to evaluate marsh nitrogen nutrient
removal via denitrification. Stable isotope studies will
assay effects of river pulses on recreational and com-
mercial fisheries of this area.
The social science objectives are to make link-
ages between the human and natural systems more un-
derstandable in three separate modeling efforts: land-
scape simulation modeling, multicriteria analysis, and
cost/benefit economic analysis.
To document the effects of physical pulsing on
the overall ecosystem dynamics of this area, the ex-
perimental treatments of high and low river input di-
versions are being used. The research group is taking
advantage of natural storm and tide "pulse treatments."
Human reactions to natural pulsing (flooding) usually
are negative, and the various planned modeling inter-
faces will explore minimizing negative effects while
maximizing positive effects of natural flood events.
The field program began in the fall of 2000, and
some indication of moderate to high chlorophyll levels
has been found in the fresher bays and lakes in the
upper end of the estuary most impacted by the diver-
sion. State and federal agencies are monitoring results
of this project closely for possible management impli-
cations. The first main field season began in January
2001.
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
200 i
'o
8
S>
ra
£
U
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Linking Environmental and Social
Performance Measurement for Management at National
and Watershed Levels: Modeling and Statistical Approaches
Scott Farrow1, Mitchell Small1, Tim Bondelid2, Andrew Solow3, George Van Houtven2, James Sinnott2,
and Martin Schultz1
'Department of Engineering and Public Policy, Carnegie Mellon University, Pittsburgh, PA;2Research
Triangle Institute, Research Triangle Park, NC;3 Woods Hole Oceanographic Institution, Woods Hole, MA
The goal of this research project is to integrate
physical, ecological, and social science models and da-
ta to provide an evaluation tool for surface water
quality managers at various levels of spatial aggre-
gation. The objectives of the study are to: (1) estimate
year-to-year changes in water quality for conventional
water quality parameters at the national and watershed
level by using index numbers, and multivariate and
ordered mean rates of change; (2) estimate the net
benefits of alternative policies for total maximum daily
loads (TMDLs) trading; (3) estimate the economic
benefits of water quality improvement at the watershed
level; (4) improve modeling of wet weather events in
a policy model; and (5) estimate the link between wa-
ter quality pollution abatement and control expendi-
tures at the facility level and water quality performance
indicators for the nation and specific regions and wa-
tersheds.
Both modeling and statistical approaches are
being investigated. As a point of departure, the Na-
tional Water Pollution Control Assessment Model
(NWPCAM) is being used (see Figure 1). With that
model, sensitivity to various input parameters, the
linkage between its output, and other indices of sur-
face water quality are being investigated. Econometric
techniques are being applied to facility-specific data.
In the initial partial year of implementation, the
major sources of uncertainty in the model approach
have been characterized, investigation of the weak
statistical links between model-based and state reports
of water quality has begun, and the hypothesis about
the crossmedia pollution abatement control costs and
the actions of polluting firms at the local level have
been econometrically tested.
Although the results are preliminary, they point
the way to local, state, and regional integrated water
quality assessments that combine environmental and
social performance measures. Investigation of the be-
havioral modeling of state water quality reports and
integration of cost data into the water quality model
also have begun. With this integration, it is hoped to
move toward evaluation of alternative policies for wa-
ter quality management where TMDLs have been de-
fined.
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Water Quality Measures
8.5
I
ED305(b) '
• NWPCAM f
* State average values based on Lyon and Farrow Uniform estimate.
' Based on unweighted average state water quality estimates from NWPCAM 1.0.
Figure 1. Preliminary water comparison: 305(b) and NWPCAM.
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Alternative Urbanization Scenarios
for an Agricultural Watershed: Design Criteria, Social
Constraints, and Effects on Groiindwater and Surface Water Systems
Richard C. LathropJ'2, Kenneth W. Potter2, Jean M Bahr3, Kenneth R. Bradbury2, Steven R Greb',
James A. LaGro, Jr.2, Edward S. Nelson', Peter Nowak2, and Joy B. Zedler2
'Wisconsin Department of Natural Resources, Madison, WI; ^University of Wisconsin, Madison, Wl
The urbanization of agricultural landscapes is oc-
curring throughout the United States, resulting in the
degradation of aquatic systems. Fundamental changes
in watershed hydrology result from the construction of
impervious surfaces (roofs, streets, sidewalks). As im-
pervious surface area expands, runoff peaks and vol-
umes increase, and water quality and groundwater lev-
els decline. Increased runoff peaks cause channel ero-
sion and habitat degradation—increased erosion of soil
and associated nutrients causes sedimentation and eu-
trophication of lakes and wetlands. The diminished
groundwater levels desiccate wetlands and lead to a
reduction hi the discharge of high-quality groundwater
to lakes, streams, and springs, and wetland biodiversity
declines. These impacts of urbanization are exacer-
bated in regions where groundwater is pumped for
domestic use and irrigation.
This group will evaluate alternative management
practices and patterns of urbanization by considering
a range of urban development issues, including storm
runoff, groundwater depletion, wastewater treatment,
eutrophication, and wetland degradation. The inter-
action among these issues and the social and political
opportunities for, and constraints on, effective manage-
ment also will be addressed. The goal is to fill critical
knowledge gaps and extend (or develop) analytical and
modeling tools that will minimize the hydrologic and
ecological impacts of urbanization. This new know-
ledge and enhanced modeling tools will be applied to
a case study of the North Fork of Pheasant Branch near
Madison, WI (see Figure 1).
An abundance of data, several ongoing research
projects, and a high level of public interest make this
an excellent research site. Comparable land use/water
management scenarios for this watershed will be con-
structed, including "low-impact development" designs,
and their approximate economic costs, social/political
acceptability, and hydrologic and ecological impacts
will be evaluated. Extant groundwater and surface wa-
ter models calibrated for the region and structurally
modified by the research group wDl be used to correct-
ly simulate the infiltration practices, thermal pollution,
well locations and pumping schedules, and wastewater
treatment options.
Urban impacts on wetlands, especially their bio-
diversity, will be examined. Those native species that
can thrive in constructed urban bioretention wetlands
will be determined. Farmer behaviors needed to reduce
high soil P concentrations in agricultural lands that are
likely to be converted to urban development will be
evaluated, and water clarity and algal bloom responses
to scenarios of P loading changes downstream in Lake
Mendota will be modeled.
Finally, the social and institutional barriers to
low-impact development will be evaluated through in-
terviews and focus groups with engineers, planners,
homeowners, and other "key players. This multidisci-
plinary research will allow for recommendations to be
made that should help local governments and citizen
groups improve the management and protection of
critical aquatic resources in rapidly urbanizing land-
scapes.
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/'USDA Water and Watersheds Progress Review
Lakes
Urban Areas
Pheasant Branch Watershed
Figure 1. Map of Lake Mendota Watershed, including Pheasant Branch Subwatershed and the Madison metropolitan urban
area. The enlarged map of Pheasant Branch shows the major downstream wetland and the North Fork (.'reck
area that is still in agricultural land use (not shaded).
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
An Integrated GIS Framework for Water
Reallocation and Decisionmaking in the Upper Rio Grande Basin
Olen Paul Matthews1 and David Brookshire2
'Department of Geography, 2Department of Economics, University of New Mexico, Albuquerque, NM
Reallocating water is a politically sensitive issue
in the Western United States. Changes from agricul-
tural uses to urban or environmental uses are occur-
ring, but the process tends to polarize competing water
users, thus creating barriers to reallocation. Other bar-
riers are inherent in the appropriation doctrine, and
some barriers exist because of poor data or inadequate
science. These barriers could be more easily overcome
and the process made less political if the impacts of
change were better known.
The biophysical and behavioral models currently
used to predict the impacts of change do not account
for spatial complexity or information uncertainty in
ways that overcome barriers to reallocation. An inte-
grated approach that couples a spatial and temporal
framework to biophysical, institutional, and behavioral
science can reduce uncertainty. Process-based geogra-
phic information systems (GISs) can fill that role by al-
lowing impacts to be assessed more accurately.
A coupled physical, environmental, and human
system model is being developed in an integrated GIS
framework to simulate interactions and changes within
the Rio Grande Watershed, NM. The coupled model
will operate entirely within a GIS, unlike other models
that use a GIS mostly for display. This approach will
permit the evaluation of impacts if any component of
the model changes as a result of natural or anthropo-
genic causes. Because water law and economics will
be integrated with physical and biological components,
the coupled model can be used to evaluate the econom-
ic consequences of water reallocation and the impacts
of different environmental policies. Stakeholders will
use the model to evaluate policy questions.
The project has two components: (1) develop-
ment of the GIS model, and (2) stakeholder evaluation
of policy options. The modeling framework of this
study utilizes a raster-based distributed water balance
approach in which each raster element represents a
bucket through which inputs and outputs may be rout-
ed. The model utilizes a hierarchical resolution grid
scheme based on a quad-tree subdivision of the land-
scape. The raster data structure is designed to allow an
infinite number of process specific resolutions on an as
needed basis (i.e., finer cells where detail is required,
coarser cells where data limitations preclude the finer
scales or where processes operate on coarser scales).
Stakeholders will identify issues and create future
water use scenarios. The information gained during
these early stages will be hypothetical to a large extent.
A pseudo-real time decision analysis tool will be de-
veloped that incorporates real consequences, via mon-
etary payoffs, to minimize the potential bias in hypo-
thetical responses. Stakeholders will make water use
decisions in an experimental setting. The cumulative
effects of individual stakeholder decisions will be sim-
ulated using a GIS model developed during the first
2 years of this research.
The model's data structure is being developed
and stakeholders are identifying the issues. Preliminary
experiments with stakeholder participation are occur-
ring. These are the first steps needed for model devel-
opment. The routing mechanism for water movement
needs to be developed, different elements of the model
need to be linked, the decision analysis tool needs to
be refined, and water use scenarios need to be devel-
oped.
10
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
The Spatial Pattern of Land Use Conversion:
Linking Economics, Hydrology, and Ecology To Evaluate
the Effects of Alternative Future Growth Scenarios on Stream Ecosystems
Margaret A. Palmer1, Nancy E. Bockstael2, Glenn E. Moglen3, N, LeRoy Poff4, James E. Pizzuto5,
Cameron Wiegand6, and Keith Van Ness6
'Department of Biology, Department of Agricultural and Resource Economics, ^Department of Civil and
Environmental Engineering, University of Maryland, College Park, MD;4Department of Biology, Colorado
State University, Ft. Collins, CO; ^Department of Geology, University of Delaware, Newark, DE; ^Department
of Environmental Protection, Montgomery County, Rockville, MD
Conversion of land to residential use has degrad-
ed freshwater ecosystems throughout the United States.
This research project is interested in how the timing,
rate, and spatial configuration of land conversion in-
fluences stream habitat and ecosystem health.
The basic design is to contrast two watersheds
with an older development closer to Washington, DC,
with two watersheds in the rapidly expanding rural-
urban fringe of Montgomery and Howard Counties,
MD. This design provides a broad mix of intensity,
spatial configuration, and history of land use develop-
ment.
By working with these counties, access to exist-
ing high-resolution geographic information system
(GIS) and biomonitoring databases is maximized.
Through collaboration with the Montgomery County
Department of Environmental Protection, the effective-
ness of current land use policies and restoration pro-
grams in minimizing the ecological consequences of
land use conversion in urbanizing watersheds is being
evaluated.
The project objectives are to: (1) examine, using
past and current conditions, how the .type, timing, and
rate of development in conjunction with its spatial con-
figuration influence stream hydrology and geomorph-
ology, which influence the structure and function of
stream ecosystems; (2) evaluate the effectiveness of
local government policies in altering the pattern of de-
velopment and in mitigating the impact of develop-
ment on stream ecosystems; and (3) use empirical and
theoretical models from hydrology, geomorphology,
and economics to make and test projections about
future development and its ecological implications,
and to compare stream health measures under different
spatial and temporal patterns of development.
To accomplish these objectives, this research is
building on recent econometric work aimed at mod-
eling and forecasting not only the quantity of land use
change, but also future spatial development patterns.
This project will incorporate the temporal dynamics of
land conversion decisions and the details of Montgomery
and Howard Counties' regulatory instruments that
have been aimed at altering the spatial and temporal
characteristics of development. This extension to the
modeling approach allows for testing the; effectiveness
of these specific public policy instruments, assessing
the effect of increasingly stringent stormwater man-
agement plans on development costs and the amenity
value of stream-side properties, and forecasting future
development patterns.
Spatially explicit models are being developed that
predict, conditional on land use history and pattern, the
change hi hydrologic and geomorphic parameters that
influence conditions along stream flowpaths in mul-
tiple watersheds across an urban-rural gradient (see
Figure 1). These models either are embedded directly
within the framework of the GIS or are linked to the
GIS in a batch-style configuration so as to fully con-
sider the spatial and temporal dynamics of the land use
change within the study watersheds.
The flow of information from one discipline to
another likewise is mediated by the common frame-
work provided by the GIS, which is critical to provid-
ing a meaningful and precise linkage between econom-
ic, hydrologic, geomorphic, and ecological quantities.
Ecological and geomorphic data will be collected
over the 3 years of the project at multiple locations
within the watersheds. These locations are organized
along a series of nested subwatersheds allowing for the
aggregation of data at different spatial scales to follow
the natural structure of the drainage network. The in-
tent is to determine how land use patterns and history
of development influence local ecological conditions.
Specifically, this will allow for an assessment of whether
ecological responses occur at thresholds under various
combinations of extent, spatial configuration, and his-
tory of land use configurations.
This research offers a synthesized approach to
evaluate the environmental consequences of alternative
future development scenarios in urbanizing watersheds
where the spatial pattern and tempo of development is
a growing public policy issue.
The Office of Research and Development's National Center for Environmental Research
11
-------
2001EPA STAR/NSFAJSDA Water and Watersheds Progress Review
Hydrologic Modeling
Economic Modeling
Future Development Patterns
Land Use Change
Biological
Inventories
Geomorphic Modeling
\
Ecological
Modeling
Stream and Riparian
Ecosystem Structure
and Function
Figure 1. Predicting impacts of land use change in urban/suburban lotic ecosystems by integrating models operating on
different spatial-temporal scales.
12
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Integrating Coral Reef Ecosystem Integrity
and Restoration Options With Watershed-Based Activities in the
Tropical Pacific Islands and the Societal Costs of Poor Land-Use Practices
Robert H. Richmond1, Michael Hamnett*, and Eric Wolanski3
'Marine Laboratory, University of Guam, Mangilao, GU; 2Social Sciences Research Institute, University of
Hawaii atManoa, Honolulu, HI;3Australian Institute of Marine Sciences, Townsville, Queensland, Australia
The objectives of this research are to: (1) charac-
terize watershed discharges affecting coastal reefs
chemically, temporally, and spatially; (2) determine the
classes and concentrations of coastal pollutants that are
of greatest concern to coral reef sustainability, and pro-
vide quantitative data for revising local and regional
water quality standards; (3) develop techniques that
can identify sublethal stress in corals, before outright
mortality occurs; (4) determine if coral reef recovery
and restoration activities are practical following both
anthropogenic and natural disturbances; (5) quantify
the cultural and economic impacts of land-based devel-
opments that affect coastal resources, and incorpor-
ate this information into the decisionmaldng process;
(6) develop educational materials for a variety of users
and stakeholders; and (7) develop a set of recommen-
dations to prevent damage to coral reef ecosystems,
and when such occurs, mitigation measures that may
be undertaken.
These objectives are being met by addressing the
following questions: (1) What is being discharged on-
to reefs from within selected watersheds? (2) How ef-
fective are present mitigation measures in controlling
watershed discharges, and how can these be improved
to provide measurable results? (3) Of the five chem-
ically mediated steps essential for successful coral re-
production and recruitment, which are most sensitive
to diminished water quality and what are the threshold
levels? (4) When land-based development occurs, what
are the societal costs when coastal resources are af-
fected? What are the parameters to be considered when
attempting to balance economic development and cul-
tural preservation on small islands? (5) Are coral reef
restoration activities practical, and if so, under what
circumstances? (6) How can data from this and other
studies be presented to stakeholders, as well as tradi-
tional and elected leaders to allow for effective envi-
ronmental policy development and implementation?
The approach includes performing ecological
studies on coral reefs as well as laboratory-based bio-
assays; quantifying levels at which sedimentation and
selected classes of pollutants become problematic;
studying coastal water characteristics of flow, resi-
dence time, and spatial-extent of watershed discharge
to determine measures that can be implemented to re-
duce negative impacts; quantifying the societal costs to
island communities resulting from watershed and re-
lated reef degradation; and testing reef restoration tech-
niques coupled with land-based remediation.
Coral fertilization and recruitment bioassays were
performed during the summer spawning event. Copper
was found to inhibit both fertilization and larval re-
cruitment at levels of 100 ppb and below. Watershed
discharge effects, including impacts of reduced coastal
salinity (see Figure 1) and increased substratum cover
by algae, also were documented. Techniques for re-
cruitment bioassays using metamorphic inducers were
refined and are being tested for precision. Coral culti-
vation techniques were refined and applied to pro-
ducing additional colonies for bioassays and transplan-
tation/reseeding trials.
Current meters were purchased and deployed to
gather data on coastal circulation patterns and charac-
terize the extent of freshwater plumes being discharged
into reef areas from select watersheds. The first set of
circulation models was developed. A workshop was
held to set priorities and identify the deliverables that
would be of most value. The workshop was attended
by 14 resource managers, researchers, educators, and
community-based organizational representatives.
This information is being incorporated into the
research plan, and there will be a followup later this
year. The "deliverables" are expected to include tools
necessary for assessing coral reef health, for monitor-
ing changes related to human activities, as well as in-
formation to guide development and policy in a respon-
sible manner.
The next steps are to: (1) begin the assessment of
the societal costs associated with poor land-use prac-
tices; (2) begin work on the educational materials;
(3) continue characterization of coastal water quality
during the dry season, and prepare for sampling during
the onset of the rainy season; and (4) continue exper-
iments on the application of biomarkers in corals.
The Office of Research and Development's National Center for Environmental Research
13
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
120
C
o
-4—*
CO
ro
LL.
;j"
C.
(1)
100
80
60
40
20
control 32.3 31.4
Salinity
29.3
27.7
HKUIT 1 • l.lt'ccl of altered salinity on fertilization of Acropom surciiloxa. C'ontrol = filtered scawatcr with salinity of
34.1 o/oo. Statistically significant effects of reduced salinity from control; 15 percent dilution, p value 0.003;
20 percent dilution, p value < 0.0001.
1 I
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Identification and Control of Nonpoint
Sources of Microbial Pollution in a Coastal Watershed
Brett Sanders', Stanley Grant', Alex Home3, Robin Keller2, and Mark Sobsey4
'Department of Civil and Environmental Engineering, ^Graduate School of Management, University of
California, Irvine, CA; ^Department of Civil and Environmental Engineering, University of California,
Berkeley, CA; * School of Public Health, University of North Carolina, Chapel Hill, NC
The goals of this project are to: (1) characterize
the spatiotemporal variability of microbial pollution in
urban runoff and to identify the association between
pathogens and indicator organisms; (2) develop a nov-
el strategy to control the impact of urban runoff on the
microbial water quality of beaches and coastal wet-
lands during nonstorm periods; and (3) develop a mul-
tiple-objective decision model to aid stakeholders in
selecting strategies to mitigate microbial pollution prob-
lems in coastal waters.
A well-defined and controllable system of flood
control channels and a constructed marsh in southern
California will be utilized as the test site. The flood
control infrastructure includes a network of pump sta-
tions with forebays that are engineered to lift runoff
from below sea-level subbasins into tidally influenced
flood control channels that drain to the ocean.
A sampling survey of forebay water and channel
water will be undertaken to ascertain the spatiotempo-
ral variabilility of pathogens (enteric viruses) and indi-
cator microorganisms (Escherichia coli, enterococci,
spores ofClostridiumperfringens, fecal coliform> and
male-specific and somatic bacteriophage) present in
the watershed to address the goal of ascertaining the
association between pathogen levels and indicator
organisms both at the Met to open channel waterways,
and at the outlet where runoff drains to the near-shore
region. The control approach involves combining both
active and passive control strategies to mitigate the
impact of urban runoff that is transported by flood con-
trol channels, through a constructed wetland, and into
a recreational near-shore area.
Pump station operation schedules that minimize
the impact of urban runoff on coastal water quality will
be determined, and the principal mechanisms responsi-
ble for pathogen- removal by tidally influenced con-
structed wetlands will be identified through a series of
microcosm studies.
Stakeholders will be interviewed or surveyed to
evaluate preferences towards various objectives as-
sociated with active and passive control strategies, and
a decisionmaking model will be developed to assess
the efficacy of existing control alternatives and to iden-
tify previously unrecognized approaches for water qual-
ity control.
Because flood control channels are a ubiquitous
feature of urban watersheds and constructed wetlands
have become an important resource for pollution mit-
igation, the primary data, control strategies, and stake-
holder information obtained in this study could lead to
regional and national strategies for reducing the ad-
verse impact of urban runoff on coastal water quality.
The Office of Research and Development's National Center for Environmental Research
15
-------
2001 EPA STAR/NSE/USDA Water and Watersheds Progress Review
Strategic Renewal of Large
Floodplain Rivers: Integrated Analysis
Richard E. Sparks1, Deborah Seal5, John B. Braden2, Misganaw Demissie6, Andrew M. Isserman2,
DouglasM. Johnston', JungikKim3, YanqingLiane, Da-MiMaeng1, ZoricaNedovic-Budic3,
Daniel Schneider3'*, Diane M. Timlin4, and David C White2
'Department of Natural Resources and Environmental Sciences, ^Department of Agricultural and Consumer
Economics, Department of Urban and Regional Planning, ^Department of Geography, University of Illinois,
Urbana, 1L; Department of Environmental Studies, Illinois College, Jacksonville, IL; Illinois State Water
Survey, Champaign, IL; 'National Center for Supercomputing Applications, Champaign, IL; Illinois State
Natural History Survey, Champaign, IL
This research extends an earlier Water and Wa-
tersheds project (96-13562) by refining and linking
ecologic, hydrologic, and economic models to support
restoration planning for a large river floodplain system.
A suite of models is being developed to provide in-
sights to stakeholders concerning likely impacts of res-
toration strategies by simulating essential aspects of a
large floodplain-river ecosystem so that alternative res-
toration strategies can be evaluated.
Component models that are being linked include:
(1) one- and two-dimensional hydraulic models of the
river, (2) a floodplain forest simulator, (3) a herba-
ceous plant simulator, (4) numerous habitat suitability
indices, and (5) regional economic input/output mod-
els. An 80-mile section of the Illinois River serves as
the study site.
Water levels in the river govern the inundation
pattern on the floodplains, except where levees prevent
flooding of lands developed mostly for row crop agri-
culture. The inundation pattern determines the vegeta-
tion communities, which in turn provide habitat and
food for fish and wildlife.
Restoration involves "de-development," or con-
version of some land from existing agricultural and
other commercial uses, to uses such as flood convey-
ance, restoration of native plant and animal commu-
nities, and outdoor recreation. The impacts of such
conversions on local and regional economies are im-
portant political and policy issues.
A second issue is naturalization of the seasonal
flood regime. Models indicate that simply removing
some levees will not restore native plant and animal
communities. Operation of the navigation dams, and
other factors now cause excessive, unnatural water
fluctuations during the summer growing season, which
destroy valuable plant communities.
Policy options include opening the levees and
modifying the operation of the navigation dams to
achieve more natural flooding; or alternatively, keep-
ing the levees intact and using pumps to create water
regimes in the areas behind the levees that are "ideal"
(e.g., for waterfowl). Variations on each of these op-
tions (e.g., regulating flooding of land behind levees
using control structures installed in the levees) have
complex consequences both in terms of costs and de-
gree of restoration of plant and animal communities
and ecosystem functions.
The connections between the river and its flood-
plain, and therefore between the hydrology, ecology,
and economic processes, are fundamentally spatial in-
teractions. A geographic information system will serve
as a data repository, a link between the various models,
and a tool for analyzing the effects of alternative man-
agement strategies. State-of-the-art visualization tools
will translate the spatially integrated model results into
visual presentations to improve understanding of envi-
ronmental processes and facilitate communication with
stakeholders.
16
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Watershed-Scale Assessments of E. coll Contamination
Implications of Source Identification for Public Policy Debate
Ronald Turco, S. Brouder, C. Nakatsu, A. Bhunia, J. Frankenberger, J. Harbor, and G. Thomas
Environmental Sciences and Engineering Institute, Purdue University, West Lafayette, IN
In watersheds in Indiana (as in most of the Mid-
west), contamination from Escherichia coli exceeds
water quality standards in most locations where mon-
itoring has been conducted. This study area, in the
Tippecanoe River Watershed, drains into Lake Shafer.
Lake Shafer is a 522-ha water supply in north central
Indiana that has shown significant and repeated high
levels of bacterial contamination as well as some con-
tamination from plant nutrients, but fairly low levels of
pesticides. Since 1993, approximately 49 percent of
the 775 Lake Shafer water samples collected for gen-
eral coliform testing have tested over the acceptable
U.S. Environmental Protection Agency standard for
whole body contact. The significance of this concern
is in the exposure route. Drinking water supplies are
treated (i.e., with chlorine) before coming into contact
with humans. Recreational waters are not treated, and
exposure is through whole body contact, including
some ingestion.
This project has five specific objectives. Objec-
tive 1 is to facilitate the use of scientific evidence
(generated from other parts of this project) in practical
efforts to improve water quality in Lake Shafer by de-
fining the setting for information assessment and pub-
lic interaction. Objective 2 is to estimate locations and
types of fecal sources available to the water supply
through a pathway analysis, and to describe water flow
within the two subwatersheds. Objective 3 is to define
the sources of the bacterial pollution by developing a
comprehensive identification scheme and database for
E. coli strains using a sensitive DNA fingerprinting
technique (e.g., amplified fragment length polymor-
phism). Objective 4 is to fully characterize the role that
land application of manure plays (if any) in facilitating
the introduction of bacterial populations into surface
water. Objective 5 is to deliver, scientific evidence
(generated from other parts of this project) to stake-
holders in the watershed as part of the evolving dialog
established in Objective 1, and to assess how this is
used to reach consensus concerning the set of solutions
the community will adopt as the most cost-effective
and equitable approaches to addressing the E. coli
problem.
To meet Objectives I and 2, the contribution of
E. coli from different land uses along two rivers that
feed into the lake is being tested by comparing
samples collected at 21 different locations. A new
hypothesis on E. coli fate in the landscape is being
developed. To meet Objective 3, more than 1,900
environmental "E. coli," defined as E. coli from known
and unknown environmental sources, have been
collected. These presently are being genetically char-
acterized. To meet Objective 4, data from a controlled
field site receiving seasonal applications of manure are
being collected. From these data, clear evidence has
been found indicating that the longevity of E. coli in
the environment may be longer than anticipated.
Integrated within these programs has been a se-
ries of community/stakeholder events. A Web site and
newsletters have been developed, and a number of pre-
sentations describing the project have been given. The
final stages of data analysis for the baseline survey of
environmental awareness are underway. As these data
sets begin to be used in educational efforts, community
awareness and response to data collected in their wa-
tershed will be evaluated.
The Office of Research and Development's National Center for Environmental Research
17
-------
2001 EPA STAR/NSFAJSDA Water and Watersheds Progress Review
An Acre an Hour: Documenting the Effects of Urban
Sprawl in a Model Watershed Near Philadelphia. Pennsylvania
Claire Welty', Susan S. Kilham', Aaron L Packman2, and Robert J. Brulle'
'School of Environmental Science, Engineering, and Policy, Drexel University, Philadelphia, PA;
^Department of Civil Engineering, Northwestern University, Evanston, IL
The principal objective of this project is to doc-
ument the effects of urbanization on the Valley Creek
Watershed, which lies in a rapidly developing area of
suburban Philadelphia, PA. Valley Creek is a tributary
of the Schuylkill River and runs through Valley Forge
National Historic Park. The watershed lies in the Pied-
mont physiographic province and supports a reproduc-
ing brown trout population in its limestone-fed stream.
In addition to the common effects of development such
as increased surface runoff and sediment loading, the
watershed has experienced point-source pollution prob-
lems from Resource Conservation and Recovery Act
and Comprehensive Environmental Response, Com-
pensation, and Liability Act hazardous waste sites, and
dewatering of the aquifer due to quarrying operations
and pumping for a municipal water supply.
A historical review of land use is being conduct-
ed in an attempt to quantify the changes in the stream
caused by development during the last 200 years. Spe-
cifically, changes in population, building permits, road
mileage, and land use patterns from colonial times to
the present are being examined. After the effects of
prior development in the watershed have been estab-
lished, the effect of the continuing urbanization of the
area will be assessed. Because the watershed is ac-
tively undergoing urbanization, this research group
will attempt to directly examine the development-
induced geomorphologic changes in the stream over
time. The primary conditions being observed are:
stream flows (base and storm flows), channel morph-
ology, bed composition, and suspended sediment con-
centrations.
The degree and pattern of heterogeneity of hy-
draulic conductivity of this fractured rock aquifer will
be quantified at multiple scales from existing hydro-
geologic data. The effect of aquifer heterogeneity and
three-dimensional flow pathways on stream-subsurface
exchange rates and contaminant transport subsequently
will be evaluated using groundwater flow and transport
models. This information will be used to assess the
chemical loadings to the fish and other biota in the
stream. Stream tracer-injection experiments are being
used to directly assess stream-subsurface exchange in
Valley Creek and storage of tracer in the hyporheic
zone.
Several aspects of the interaction between envi-
ronmental quality and the biota in this watershed are
being investigated. First, sediment distribution data are
being collected to assess the impacts of changes in
channel characteristics on the community structure of
macroinvertebrates and fish. Second, polychlorinated
biphenyl (PCB) levels and supply rates are being com-
pared with PCB levels in the organisms from the same
areas. Third, the positioning of species in the food web
is being assessed to measure biomagnification via food
web processes. Fourth, general stress levels in the or-
ganisms are being measured—these measurements are
being related to PCB levels as an independent in-
dicator of environmental impacts on organisms. This
analysis will allow for the development of a compre-
hensive picture of how urbanization-induced changes
in the watershed affect invertebrate and fish com-
munities.
Political controversies in this watershed have left
a documented historical record of the political strug-
gles that develop in the process of urbanization. The
major development activities that have -had an impact
on the watershed over time will be identified, and the
political decisionmaking associated with these activi-
ties will be examined. By examining these political
decisions, an understanding of the political forces in-
volved hi urbanization can be developed.
The social science research is based on the use of
three social science perspectives: (1) Advocacy Coali-
tion Framework, (2) network analysis, and (3) dis-
course analysis. These sociological perspectives define
an image of watershed politics as the result of the for-
mation of different advocacy coalitions, each with a
specific network structure and unique belief system. A
comprehensive historical view of the process of urban-
ization, including the influence of social, economic,
and political factors will emerge from this research.
18
The Office of Research and Development's National Center for Environmental Research
-------
Section 2.
Projects Initiated With Fiscal Year 1998 Support
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Development and Testing of a Decision
Support System for River Rehabilitation
/. David Allan, Gloria Helfand, and Joan Nassauer
School of Natural Resources and Environment, University of Michigan, Ann Arbor, MI
The goal of this research project is to develop a
conceptual and quantitative watershed model that will
aid in identifying opportunities for the rehabilitation
of stream ecosystems that are socially acceptable,
ecologically beneficial, and cost effective. Research
is organized around a series of modules, each of
which can be linked to a geographic information sys-
tem (GIS).
The final product will be an integrated, spa-
tially explicit, multiscale model to display informa-
tion on land use, land cover, and ecological condition
of tributary streams in a 5,000 km2 agricultural and
urbanizing region. Using "build-out" projections of
local government master plans, as implemented by
conventional or alternative innovative landscape sce-
narios, the approach is intended to support decision-
makers' capacities to advocate for ecologically ben-
eficial landscape change and to anticipate ecological
effects of landscape changes that are likely to be
proposed for agricultural landscapes.
Module A: Landscape and Hydrological Pro-
cesses. The Huron (drainage area 2,320 km2) and
Raisin (2,780 km2) Watersheds together comprise
much of the landscape of southeastern Michigan ex-
ternal to Greater Detroit. The Huron has substantial
urban land, including small cities and outlying sprawl,
while the Raisin has less urban development. The
Huron Basin is less agricultural compared with the
Raisin Basin, and it contains more wetland and a
similar amount of forest. Considerable spatial hetero-
geneity is observable at the scale of the subcatch-
ments associated with headwater streams. Based on
approximately 50 small subcatchments from both wa-
tersheds, agricultural land use varied from 2-80 per-
cent, with the top quartile more than 60 percent agri-
cultural. Urban land use ranged from 0.2-70 percent,
with the top quartile more than 20 percent urban.
Comparison of 100-m riparian buffers to total
subcatchment area reveals overall strong correla-
tions. Wetland and grass are overrepresented in buf-
fers and forest, agriculture, and urban land uses are
underrepresented, on average. Patch number, not patch
size, appears to underlie differences among sub-
catchments in wetland and forest extent.
Module B: Social Acceptability and Economic
Feasibility of Alternative Landscape Scenarios.
The premise of this module is that forms of devel-
opment and forms of agriculture that are ecologically
beneficial, and viewed as desirable and affordable by
the public are more likely to be implemented and
sustained. Using alternative landscape designs as ex-
perimental treatments, suburban and rural southeast-
ern Michigan residents' perceptions of the attractive-
ness and prices for alternatives site and subdivision
designs were surveyed using traditional pencil-and-
paper questionnaires and a Web-based survey.
Survey results will be used to determine wheth-
er the more ecologically friendly designs are per-
ceived and valued similarly, in which case there is
no social cost (there is a social benefit) associated
with improving ecological quality in the rivers. If
people prefer the less ecologically friendly designs,
then this research will calculate how much people
might have to be compensated for more protective
designs. Survey results also have scale (landowner
versus subdivision) implications for planning devel-
opment hi rural areas.
Results to date indicate that at the subdivision
scale, design treatments characterized by the most
beneficial ecological function, with less lawn area
and more forest or prairie, were more attractive. In
contrast, at the front yard scale, the most ecologic-
ally beneficial designs were not found to be as at-
tractive as moderately ecologically beneficial designs
or the least ecologically beneficial designs. This scale
hierarchy may suggest an ecologically beneficial and
socially acceptable approach to the aggregation of
small-scale ecological improvements at the subdivi-
sion and watershed scale.
Module C: Ecological Integrity of Stream Eco-
systems. One or more field sites on 48 tributary
streams were assessed using biological (fish macro-
invertebrates), habitat metrics, and chemistry. Anal-
ysis of macroinvertebrate data is not complete. Land-
scape metrics (subcatchment land use, buffer land
use, subcatchment geology) alone accounted for ap-
proximately 30 percent of the variation hi the fish-
•based index of biotic integrity (IBI). Habitat metrics
accounted for 41-54 percent of variation in the IBI.
This analysis is preliminary, using simple combina-
tions of variables. Further analyses will use rnulti-
variate approaches to explore the interrelationships
among landscape, habitat, and biological measure-
ments required to determine the best set of predictive
relationships for model-building.
The project was initiated in June 1999. Work to
date has concentrated on module development, with
integration efforts slated for summer of 2001.
The Office of Research and Development's National Center for Environmental Research
21
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
An Integrated Systems Approach to Watershed Restoration
With Community Involvement Applied to a Small Rural Watershed
J. Boll', J.D. Wulfhorst2, S. Chen3, C.O. Stockle3, D.K. McCool3, D.C. Brown4,D. Feichtinger4,
andA.J. Vitale5
'Department of Biological and Agricultural Engineering, ^Department of Agricultural Economics and Rural
Sociology, University of Idaho, Moscow, ID;3Biological Systems Engineering Department, Washington State
University, Pullman, WA; ''Natural Resources Conservation Service, Coeur d'Alene, ID; sCoeur d'Alene
Tribe, Fish, and Wildlife Program, Plummer, ID
In watersheds today, it is imperative to develop a
scientifically based, integrated watershed restoration
process with community input. The geographic area
being studied in this research project is the Northwest
Wheat and Range Region in Idaho, Washington, and
Oregon. The objectives of this project are to: (1) de-
velop a geographic information system (GlS)-based
integrated systems approach for watershed restoration
(see Figure 1); (2) adapt the adoption-diffusion model
to identify institutional and attitudinal barriers to the
adoption of erosion and water quality control practices;
(3) test and improve an existing model for enhanced
ability to determine critical source areas and to eval-
uate management practices and climatic variation in
watersheds; and (4) develop an optimization technique
for integration with the systems approach, including
socioeconomic, physical, and ecological aspects. The
study watershed is Lake Creek Watershed in the Coeur
d'AJene Lake Basin in Idaho/Washington State.
Watershed restoration consists of the selection of
"best management practices," which consider a physi-
cal, socioeconomic, and cultural component. Restora-
tion efforts in Lake Creek Watershed since 1991 pro-
vide data on hydrology, water quality, and ecological
aspects. Using these data, a hydrology model coupled
with simple crop growth, erosion, and economic mod-
els has been developed and tested to determine the cost
of erosion without the use of control practices.
Current land use conditions are characterized by
bluegrass seed production and future conditions by the
conversion to winter wheat. This approach is being ex-
panded to include the cost of different tillage oper-
ations and implementation of control practices. Inter-
views were held in the watershed to strengthen the
sociological aspects of erosion control. Multiple meet-
ings with the Watershed Working Group were held
which, among others, consists of the local community
and tribal, state, and federal agencies.
Given the desire to transfer the integrated system
to county-level resource managers, two constraints were
placed on model development. First, input data are
based on publicly available data sources, and second,
the models do not require calibration. Application to
Lake Creek Watershed shows that using these con-
straints provided very reasonable results. The hydrol-
ogy model successfully simulated observed stream flow
and distribution of runoff-generating areas in the wa-
tershed.
Erosion distribution in the watershed determined
using the empirical Revised Universal Soil Loss Equa-
tion (RUSLE), which used a length-slope factor based
on upslope contributing area, agreed reasonably well
with the distribution of runoff-generating areas. In ad-
dition, total erosion amounts appeared to be in agree-
ment with observed sediment measured in the stream.
The crop growth modeling approach based on actual
evapotranspiration and water use efficiency yielded
reasonable, although somewhat low, spatially distri-
buted estimates of bluegrass and wheat yields. Yield
reductions after 75 years of soil erosion using the ad-
justed topsoil depths were somewhat low, resulting in
small yearly price reductions.
The example application shows the potential of
the integrated systems approach in yielding very useful
site-specific information. It is believed that mis tool
eventually will assist resource managers to identify
critical source areas within a watershed and properly
assign load reductions for individual landowners. An
optimization procedure is being developed based on
the RUSLE. The objective function is minimizing the
total cost of restoration of an entire farm in a water-
shed, or all the farms in the entire watershed. Integra-
tion of an expanded version of the physical-economic
systems approach with the sociological data will fol-
low another round of interviews with landowners in
the watershed.
22
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Resource Managers
1
No
r
Database/CIS
Indicators Assessment:
Socioeconomic: Ecological:
- offsite (urban) - water quality
- onsite - habitat
- farm income - biol. health
Implementation
• community involvement
Maintain status or
recommendations
climate
DEM, soils, land use,
current/alternative
practices
Physical Science Model
(hydrology and water quality)
- onsite soil and nutrient loss
- offsite soil and nutrient loss
Socioeconomic Model
-cost of erosion
- farm optimization
Optimization Scheme
- water quality plan
Figure 1. Schematic representation of the GIS-based integrated systems approach.
The Office of Research and Development's National Center for Environmental Research
23
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Integrating Ecological, Economic,
and Social Goals in Restoration Decisionmaking
John Bolte
Bioresource Engineering Department, Oregon State University, Corvallis, OR
The integration of ecological, economic, and
social goals is an important element of watershed
restoration planning and prioritization. It generally is
accepted that for restoration efforts to be successful,
each of these goals must be addressed in a manner
that reflects stakeholder priorities, objectives, and
constraints. Additionally, it is becoming increasingly
apparent that restoration strategies based solely on
opportunistic, site-scale activities frequently do not
accomplish watershed-scale goals.
Because watersheds are complex systems in-
volving integration of human, hydrologic, and eco-
logical processes, it can be difficult to understand the
consequences of particular restoration activities on
meeting restoration goals. Synthesis tools capturing
spatially explicit data are needed to couple human
and ecological processes with landscape features to
assist in developing effective restoration plans.
The overall goal of this project is to refine and
integrate spatially explicit models of watershed func-
tion and economic characterizations of restoration
options with stakeholder-determined constraints and
priorities. The resulting tool can assist stakeholders
in identifying feasible restoration strategies and eval-
uate the ecological and economic effectiveness of
these strategies at addressing watershed-level ecolo-
gical, economic, and social function.
A geographic information system-based multi-
objective decision support tool that contains a series
of rules that relate specific site-based restoration al-
ternatives, stakeholder goals, and site-specific land-
scape features is being developed to generate fea-
sible restoration plans that reflect stakeholder con-
cerns. This research group is cooperating with two
watershed councils representing diverse watershed
types and disturbance levels to evaluate the effec-
tiveness and transferability of the methodology be-
tween distinct ecological and economic systems. The
analysis framework uses a landscape generator to
apply design heuristics that embody ecological, eco-
nomic, and social constraints and preferences to al-
locate restoration activities to specific sites based on
site features. It then evaluates the resulting landscape
options using a series of ecological, social, and eco-
nomic watershed-scale models. The utility of the tool
for addressing stakeholder needs and its impact on
stakeholder decisionmaldng is being explicitly eval-
uated using sociological and applied anthropological
methods.
Sociological analysis of the stakeholder groups
represented in the two watershed councils has been
completed. Also, the landscape generation tool has
been completed, and a collection of approximately
400 rules relating the utility of particular restoration
strategies for meeting stakeholder preferences to site
features have been developed. Preliminary results
have been presented to the watershed councils with
positive results. Stakeholder suggestions currently are
being incorporated into the analysis framework.
Initial results indicate that using a rule-based
framework for capturing qualitative relationships be-
tween restoration strategies, stakeholder objectives,
and site features is an effective way of representing
these relationships in a manner that stakeholders can
readily understand. Making restoration recommenda-
tions at a site level, distributed across a watershed,
allows for the evaluation of the effectiveness of ba-
sinwide plans at meeting stakeholder goals. Further,
the use of muJtiobjective methodologies provide a
stakeholder-accessible method for weighing and bal-
ancing competing economic, social, and ecological
objectives. It is anticipated that these rules could be
readily adaptable to restoration strategies, other situ-
ations, and stakeholder goals. Although it is too soon
to evaluate the usage of the too] by the watershed
councils, the initial response has been very positive.
24
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Social Impact Assessment of Human Exposure to Mercury Related
to Land Use and Phvsicochemical Settings in the Mobile-Alabama River Basin
Jean Claude Bonzongo', Eric E. Roden1, Milton G. Ward2, C Hobson Bryan3, W.B. Lyons*,
andlndrajeet Chaubey5
'Department of Environmental and Engineering Sciences, University of Florida, Gainesville, FL; ^Department
of Biological Sciences, Department of Geography, University of Alabama, Tuscaloosa, AL; 4Byrd Polar
Research Center, Ohio State University, Columbus, OH; ^Department of Biological and Agricultural
Engineering, University of Arkansas, Fayatteville, AR
There four objectives of this research project
are to: (1) determine levels and speciation of mercu-
ry (Hg) in water, sediments, and fish from different
aquatic systems in the Mobile-Alabama River Basin
(MARB) (see Figure 1); (2) investigate the linkage
between land-use types or the presence of wetlands
and microbial Hg transformation and bioaccumu-
lation; (3) predict Hg levels in fish using recent and
historical land-use data; and (4) use a participatory
approach to environmental decisionmaking to ame-
liorate conflict, and achieve an effective understand-
ing and support for Hg policy.
Water, sediment, and fish (largemouth bass)
were collected from 52 sites with different land uses
across the MARB. Recreational tournament fisher-
men were utilized for part of the fish collection ef-
fort. Water chemistry, Hg levels, and speciation in
these samples are being determined. Following this
survey, several sites will be selected for more inten-
sive sampling and laboratory studies, directed to-
ward linking observed trends in fish Hg levels with
processes-controlling Hg transformation and bioac-
cumulation. Next, a database will be developed and
used to map Hg concentrations and to determine
their correlation with biogeochemical and physical
variables. Lastly, the public will be involved in and
informed of this research to aid in the assessment of
risk imposed by elevated Hg levels in fish and to
help formulate possible remedial policies.
Total mercury concentrations in the water sam-
ples (0.43-2.23 ng L~') fall among background levels
typically found in natural waters worldwide. After
analyzing the majority of the fish samples, at least
one fish from each major river basin within the
MARB was found to have a total Hg concentration
greater than 0.5 ppm, a level at which limited con-
sumption is recommended by some regulatory stan-
dards. Mercury concentrations of greater than 1 ppm
were detected in fish samples from three locations
thus far, two of which are categorized as being im-
pacted by wetland area. Project personnel have met
with recreational fishermen, municipal groups, and
several energy industries to exchange ideas and is-
sues. These groups have come to appreciate the com-
plexity involved in understanding Hg behavior in the
environment, and they are willing to remain involved
through the life of the project.
Preliminary results suggest that while mercury
concentrations in water samples from all sites are very
low, elevated levels of mercury are found in selected
fish samples across the watershed. The data support
the idea that wetland abundance may result in higher
concentrations of mercury in fish in those areas. It is
anticipated that the mapping and statistical analyses
will be helpful in predicting the potential for Hg ac-
cumulation in predatory fish hi other similar loca-
tions in the Southeast United States, and that the
approach of involving stakeholders throughout the
project will result in positive approaches to develop-
ing mercury policy.
After completion of all sample analyses, data-
base development, and statistical analyses and map-
ping, informed decisions will be made on which sites
to focus for the detailed biogeochemical studies. The
social assessment continues by gathering input from
stakeholders and identifying mutual concerns.
The Office of Research and Development's National Center for Environmental Research
25
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Figure 1. The Mobile-Alabama River Basin.
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSFAJSDA Water and Watersheds Progress Review
Applying the Patuxent and Gwynns Falls Landscape Models
To Designing a Sustainable Balance Between Humans and the Rest of Nature
Robert Costanza, Alexey Voinov, Roelof Boumans, Tom Maxwell, Ferdinando Villa, Helena Voinov,
and Joshua Farley
Institute for Ecological Economics, University of Maryland, Solomons, MD
As part of a previous Environmental Protection
Agency/National Science Foundation-funded project,
an integrated, spatially explicit model of the Patuxent
Watershed, MD, has been developed (http://iee.umces.
edu/PLM). The model is being further developed to
use as a tool for whole watershed analysis and resto-
ration. This includes development of methods to assess
the ecological health of ecosystems and watersheds,
development of preferred future states for the water-
sheds using broad stakeholder participation, and devel-
opment of dynamic links between the ecological and
socioeconomic sectors of the model. Based on the
above, the degree to which various management poli-
cies can restore the ecological health of the Patuxent
and Gwynns Falls Watersheds and achieve the pre-
ferred future states will be tested.
In the Patuxent Landscape Model (PLM), the wa-
tershed is represented as a grid of cells with a process-
based ecological model replicated in each of the cells.
The ecological model includes modules to simulate
local and spatially distributed hydrologic fluxes, nu-
trient dynamics, plant growth, dead organic matter
decomposition, and so on. These modules were tested
and calibrated separately, and then put together within
a Spatial Modeling Environment, created by this re-
search group, to represent the watershed as a whole.
The model runs are in good agreement with avail-
able data. Analyses for numerous scenarios of land-use
change and nutrient loading were performed. The
model output is compared in the different scenarios
examining nitrogen concentration in the Patuxent Riv-
er as an indicator of water quality, changes in the hy-
drologic flow, and changes in the net primary pro-
ductivity of the landscape, as indicators of ecosystem
services.
The PLM has been calibrated to mimic the hy-
drologic flows in the Villa Nova Subwatershed of
Gwynns Falls. This research project will be expanded
to other subwatersheds and the full watershed as the
data for calibration become available. The human
capital model has been further developed and tested.
Calibrations were successful against data from the U.S.
Census Bureau on population dynamics in Baltimore
between 1790 and 1994.
A stakeholder workshop (Patuxent and Gwynns
Falls Watersheds) was held on February 14 at the Uni-
versity of Maryland College Park Campus. Breakout
groups discussed the most appropriate uses of the mod-
el, assessed what model scenarios would prove most
useful to stakeholders, and sought common ground
concerning preferred future states of the Patuxent and
Gwynns Falls Watersheds as a step towards defining
endpoints for restoration efforts. The workshop proved
valuable in developing future directions for model
development to better meet stakeholder needs, and
initiated the task of defining preferred states for the
watersheds.
To ensure broad stakeholder participation, this
research group is working in close collaboration with
the Patuxent River Commission (PRC). A Scenario
Development Working Group has been formed with.
the PRC to oversee and coordinate the development of
the model with the needs of the stakeholders. As an
outcome of this effort, several focus subwatersheds
have been identified for case studies and specific ap-
plications for restoration projects, such as riparian buf-
fer design and stormwater management.
The model has been applied in an optimization
framework to find optimal patterns of land use and
fertilizer application in a watershed. The goal function
was chosen to take into account both economic and
ecological considerations. The integration of both of
these indicators makes the approach very promising
for purposes of valuation of landscapes and watershed.
Outreach efforts include maintaining a Web page with
all the significant project developments and applica-
tions for public participation and dissemination of
results. The project also became part of an educational
effort in collaboration with the Calvert Department of
Planning and Zoning and Calverton School to intro-
duce high school students to watershed dynamics and
modeling.
The Office of Research and Development's National Center for Environmental Research
27
-------
2001 EPA STAJR/NSF/USDA Water and Watersheds Progress Review
Understanding the Social Context for Ecological Restoration
in Multiple-Ownership Watersheds: The Case of the Cache River in Illinois
Steven Kraft1, Christopher Lant2, Jeffrey Beaulieu', Leslie Duram *, J.B. Ruhl3, David Bennett4,
Jane Adams5, John Nicklow6, and Tim Loftus2
'Agribusiness Economics Department, 2Geography Department, 3Law School, ^Department of Anthropology,
'Department of Civil Engineering, Southern Illinois University, Carbondale, IL;4Geography Department,
University of Kansas, Lawrence, KS
A conceptual framework that has been developed
to help focus this work on the watershed planning
process. As part of this process, the research team
identified 30 individuals who were significant players
in the recently completed watershed planning process
in the Cache River Watershed. Using an open-ended
questionnaire, indepth interviews of these key infor-
mants were conducted. The informants came from
three main groups: Technical Committee and other
agency personnel, local activists, and Resource Plan-
ning Committee members. Interviews were based on
questions relating to the individual, the planning group,
and outside influences. Interviews have been tran-
scribed and are being analyzed.
Based on the interviews and literature review, a
set of preliminary findings have been developed that
will help to guide the remainder of the research. As
analysis of the interviews continues and the focus
group and telephone surveys are developed, it is impor-
tant that these initial findings are used as a foundation
for broader investigation within the. watershed.
Findings deal with the following topics: (1) the
Watershed Plan provides agency legitimacy; (2) there
are divergent views between agency personnel and
farmers regarding the format and substance of the plan-
ning meetings; (3) differences in the power base be-
tween farmers and environmentalists in the region have
developed over time, so that environmentalists rely on
the agencies, while farmers rely on elected officials;
and (4) theories of power structure are relevant in the
Cache and must be illuminated to understand the plan-
ning process and its outcomes.
There has been a complete review of both state
and federal laws that impinge on watershed planning.
Identifying the legal framework within which water-
shed planning takes place in conjunction with the anal-
ysis of the indepth interviews and literature review is
expected to shed light on the question of the legitimacy
of the planning process and the resulting watershed
resource management plan. The question of the legit-
imacy of the planning process and the resulting resource
management plan has occupied many of the research
team meetings during the last year. A review of the
literature indicates that this is a critical question in the
planning process that has not been addressed.
The development and refinement of a spatial de-
cision support system (SDSS) has continued through-
out this phase of the research. For the SDSS, the re-
search team is developing a graphical user interface
that will facilitate the use of the SDSS with community
members who are part of a watershed planning process.
The goal is to have an SDSS that will show the eco-
nomic and environmental consequences of different
policy scenarios designed to enhance environmental
quality. Watershed planners then would be able to
develop a number of "what if scenarios and see their
economic consequences as well as the implications for
the watershed's landscape. To enhance the SDSS, use
of genetic algorithms (GAs) is being explored.
The development of the genetic algorithm code
is complete. The Soil and Water Assessment Tool
(SWAT) source code has been linked to the genetic
algorithm for the single objective function. This model
will directly evaluate the optimal land-use distribution
to minimize sediment yield. Figure 1 demonstrates a
solution convergence to a minimum sediment yield
from a hydrologic response unit. This model will di-
rectly evaluate the optimal land-use distribution to mini-
mize sediment yield.
In light of practical constraints, a database of
feasible land-use management alternatives currently is
being assembled for SWAT that are appropriate for the
study area. These will be incorporated into the GA-
SW AT optimal control model for the singular sediment
objective. Simultaneously, linkages between the SWAT
source code and MINOS (a mathematical program-
ming program) are being developed for an economic-
related objective. The next phase of work will consist
of integrating the sediment and economic model to
create one multiobjective optimal control model that
interfaces SWAT and the GA for minimizing sediment
yield and maximizing farming income.
28
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
130
I
1 120-
U
t
o
a
2
"3
no
| 100
B
*3
90
80
10
20 30
Generation Number
40
50
Figure 1. Solution convergence for a hydrologic response unit.
The Office of Research and Development's National Center for Environmental Research
29
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Restoring and Maintaining Riparian Ecosystem Integrity in Arid
Watersheds: Meeting the Challenge Through Science and Policy Analysis
Thomas Maddock, in1, Kathryn Baird', Victor Baker', Bonnie Colby1, Robert Glennon3,
and Julie Stromberg4
'Department of Hydrology and Water Resources, ^Department of Agricultural and Resource Economics,
^College of Law, University of Arizona, Tucson, AZ; ^Department of Plant Biology, Arizona State University,
Tempe, AZ
This research projectcombines expertise from four
disciplines whose knowledge is critical to restoring and
maintainingrivers in the Southwest: hydrology, ecology,
economics, and law, Hydrologic models, a riparian eco-
system integrity index, and economic analyses are being
integrated into a user-friendly decision support system
(DSS). Coupled with legal analysis, this integration is
designed to aid in understanding the impact of develop-
ment and in evaluating strategies for maintaining or
achieving environmental restoration.
To improve estimates of riparian evapotranspiration
(ET), ET measurements were taken in conjunction with
climate, surface water, and groundwater parameters on
the South Fork Kern River Basin. Detailed river eleva-
tion profiles and vegetation transects were surveyed. ET
(stem flux) measurements are being analyzed in con-
junction with the tree structure, hydrologic, and climatic
information. Joint ET and abiotic measurements eluci-
date how abiotic parameters combine with tree physio-
logy to determine the amount of water required by ri-
parian trees. Community water requirements are scaled
up from ET measurements, density and size class in-
formation, and coverage estimates from habitat maps
derived from recent aerial photographs. To simulate
surface water and groundwater behavior, the HEC-RAS
and MODFLOW models are used. Flood flows asso-
ciated with riparian recruitment events were determined
from tree ring and river gauge data.
To develop an index of biotic integrity (D3I), field
studies on two community types in the San Pedro Basin
are underway to identify metrics that are sensitive in-
dicators of site moisture availability. Research was con-
ducted on sites spanning a gradient from wet (perennial
stream flow) to dry (ephemeral flow). Vegetation var-
iables from three categories were measured: individual
productivity, population abundance and size structure,
and community composition. Abundance and composi-
tion of terrestrial arthropods also were measured.
Correlation and univariate regression analyses are
bebg used to determine the relationship of plant and
insect variables with depth to groundwater, surface flow
frequency, and soil moisture. Multiple regression anal-
ysis is used to determine whether the vegetation metrics
vary with geomorphology, site elevation, and site hydro-
logy. This allows for a determination of whether the ffll
will require stratification by geomorphic reach type and/
or site elevation. Threshold values, above or below
which vegetation metrics change sharply, or which spe-
cific plant associations do not occur, are identified.
An economic model of costs and benefits asso-
ciated with restoring riparian areas is being constructed.
Economic activities relating to instream and consumptive
uses of water, and estimates of local income related to
differing water uses have been identified. The physical
location and land ownership patterns along the South
Fork Kem have been documented, giving background to
the region's history, describing the population centers,
providing an economic setting, and outlining both the
surface water and groundwater users. A Kern River Pre-
serve visitor's survey was implemented and will allow
for an estimation of the local economic impacts of visita-
tion to the riparian preserve. A parallel study is underway
for the San Pedro. Both surveys assess visitor economic
benefits from riparian area protection.
California groundwater and surface water law and
the role of the Endangered Species Act (ESA) in man-
aging rivers are being studied. A lawsuit filed by The
Southwest Center for Biological Diversity that chal-
lenges, under the ESA, the administration of waterflow
from Lake Isabella is being examined.
30
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Development of an Urban Watershed Rehabilitation Method Using
Stakeholder Feedback To Direct Investigation and Restoration Planning
Marty D. Matlock1, Charles D. Samuelson2, WMam H. NeUl3, Tarla Rat Peterson4, Ann L. Kenimer5,
and Guy D. Whitten6
'Department of Agricultural Engineering, Department of Psychology, 3Department of Wildlife and Fisheries
Sciences, ^Department of Speech Communication, ^Department of Agricultural Engineering, Department of
Political Science, Institute for Science, Technology and Public Policy, George Bush School of Government
and Public Service, Texas A&M University, College Station, TX
This project has developed and is testing a meth-
od for restoring the ecological integrity of urban water-
sheds that integrates ecology, engineering, and social
science. Research is being conducted on two streams
in the San Antonio, TX, metropolitan area. Research
questions include: (1) Can a risk-based watershed
model linked with two ecoindicators in a regressive
ecological risk assessment for a complex watershed
quantify the uncertainty associated with ecosystem
rehabilitation? (2) Will stakeholders' understanding of
nonpoint source pollution issues, ability to use scientif-
ic information about watershed management strategies,
and communication competence improve as a result of
a~collaborative learning (CL) intervention? A water-
shed model of ecological risk assessment is being de-
veloped. Models of stakeholder knowledge levels and
environmental attitudes also are being developed.
A watershed model linked with an instream mo-
del has been developed to evaluate and optimize eco-
system management strategies. CL is being used to
structure and facilitate stakeholder group activities
among large, heterogeneous groups affiliated with the
Salado and Leon Creek Watersheds in San Antonio.
Two integrated ecoindicators are being used to eval-
uate and communicate risk to the stakeholder groups.
This risk-based approach is used to shape dis-
cussions between stakeholders and scientists in an iter-
ative process that results in an informed and stake-
holder-driven action plan for watershed rehabilitation.
Social dilemma/game-theoretic analysis will be used to
develop models of compliance under different assump-
tions about time and other actors' behavior. A sim-
ulation of the interactive effects of human and non-
human factors on watershed nutrient levels will be
developed.
A public opinion survey was designed to obtain
detailed measures of public opinion on general and
specific environmental issues and on local watershed
issues among others, and to identify potential partici-
pants in the stakeholder recruitment process. The
survey was administered to 1,017 randomly selected
residents of Bexar County, TX. These data currently
are being analyzed. The issues of representation and
implementation are being examined (e.g., issues of ef-
ficacy, trust, participation, and optimism).
In November 1999, through December 2000, the
project team held monthly meetings with the Salado
and Leon Creek stakeholder groups. Potential partici-
pants were identified in a number of ways, including
the public opinion survey and face-to-face contact with
community groups. A snowball technique was used for
additional recruitment. Three surveys—a pretest, fol-
lowup, and posttest—were given to each of the stake-
holder groups. These survey data allow for direct com-
parisons between the large sample telephone survey of
Bexar County residents and the stakeholder groups
participating in the CL meetings.
The statistical analysis of survey data will be con-
ducted during 2001 to assess the effectiveness of the
CL program hi meeting project objectives. Followup
interviews with stakeholder group members will be
conducted in 2001 as well. Data from individual inter-
views with stakeholders and CL workshops will be
analyzed to identify critical social processes and com-
munication practices.
Field investigations of ecological processes were
initiated in August 1999. Two sites were selected on
two streams (one upstream and one downstream) flow-
ing through San Antonio. Periphyton response to nu-
trient loading (nitrogen and phosphorus) was measured
in situ using the Matlock Periphytometer. The survival,
growth, and metabolic capacity of bluegill acclima-
tized in ambient stream water also were measured at
each of the sites. The data currently are being anal-
yzed, and field investigations will continue in 2001.
Geographic data have been compiled from a vari-
ety of sources (U.S. Geological Survey, Texas Natural
Resource Conservation Commission [TNRCC], Envi-
ronmental Protection Agency, National Aeronautics
and Space Administration) to develop input data files
for the Better Assessment Science Integration Point
and Nonpoint Sources-Hydrological Simulation Pro-
gram-Fortran (BASINS-HSPF) modeling program.
Calibrating BASINS-HSPF in Salado and Leon Creeks
has been completed in collaboration with the San An-
tonio River Authority and TNRCC.
The Office of Research and Development's National Center for Environmental Research
31
-------
2001 EPA STARyNSF/USDA Water and Watersheds Progress Review
Combining Economic and Ecological Indicators To Prioritize
Wetlands Restoration Projects Within a Spatial GIS Framework
James J. Opaluch2, Marisa J. Mazzotta', Peter August2, Robert Johnston *, and Frank Golet2
^Department of Environmental and Natural Resource Economics, ^Department of Natural Resources Science,
University of Rhode Island, Kingston, RI
Restoration and rehabilitation of damaged or de-
graded ecosystems is an important component of many
of today's environmental and natural resource manage-
ment strategies. It not only is important to protect and
preserve watershed ecosystems, but also to restore de-
graded components so that the functioning of the entire
system is maintained or improved. This project focuses
on developing methods for setting priorities for wet-
lands restoration projects and applying the tool to res-
toration of coastal wetlands. The research will address
three issues related to valuing wetlands: (1) spatial
aspects of value, (2) cost-effective methodologies for
valuation, and (3) transferability of methods and val-
ues. The research employs a three-phased approach to
provide sequential links from physical wetlands fea-
tures, to functions provided by the wetlands, to values
of those functions, and ultimately, to setting priorities
for alternative restoration projects.
Phase I develops the methodology of the study,
and provides initial steps in identifying and defining
indicators. This phase includes development of the
conceptual linkages between indicators, indices, and
valuation methods; linking indicators with benefits
transfer methods; coordinating the approach with data
collection efforts by the Rhode Island Department of
Environmental Management; and identifying an initial
set of indicators to serve as a starting point for the field
research, included in Phase II.
Phase II develops and implements two surveys—
a survey of technical experts, and a survey of the gen-
eral public. The survey of technical experts identifies
specific features that contribute to the potential for the
wetland to provide important functions. For example,
open water and tidal creeks can contribute to a wet-
land's potential for fish and bird habitat. The general
public survey elicits public values for important wetland
functions.
Phase IE uses the methods and data developed in
Phases I and n to develop and implement the geo-
graphic information system-based tool for identifying
wetland priorities based on indicators of social and
economic values. It also will test the transferability of
the method to other sites by applying results to ad-
ditional locations in Rhode Island and in another state.
To date, the project has focused on habitat func-
tions of wetlands, as they appear to be the principal
function of Rhode Island coastal wetlands. The re-
search group is in the process of completing focus
group pretests that are critical to development of the
public survey. Simultaneously, meetings have been
held with wetland experts to learn about important
wetland issues and to develop a method for eliciting
expert opinion on the linkage between physical fea-
tures and habitat quality for the various species groups.
When complete, the approach will provide input
into prioritizing wetlands restoration programs. The
research group is working closely with the state's
interagency restoration team, which is in the process of
developing tools to prioritize restoration actions. The
research project's efforts have been designed to be
complementary—in particular, to provide public input
that is needed by the state.
32
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Integrating Science and Technology To Support Stream Naturalization
Bruce L. Rhoads, David Wilson, Edwin E. Herricks, Marcelo Garcia, and Rebecca Wade
University of Illinois at Urbana-Champaign, Urbana, IL
Recent initiatives by federal agencies, including
the Environmental Protection Agency, have supported
a move toward integrated watershed management that
emphasizes community-level decisionmaking based on
sound science. The concept of stream naturalization,
which seeks to establish sustainable, morphologically
and hydraulically varied, yet dynamically stable fluvial
systems that are capable of supporting healthy, biolog-
ically diverse aquatic ecosystems, is consistent with
this new perspective. Naturalization integrates biolog-
ical, physical, and social science within a local deci-
sionmaking context over multiple temporal and spatial
scales.
This research project seeks to develop an inte-
grated scientific and technological framework for stream
naturalization. Empirical and modeling aspects of the
research are focusing on case studies of stream natural-
ization in two small watersheds in the Chicago metro-
politan area. These case studies highlight the scientific
and technological challenges associated with natural-
ized stream-channel designs as well as the vital role of
social interaction and community perceptions in the
naturalization process.
The research design combines social analysis,
both of community-based environmental visions and of
decisionmaking about stream naturalization, with a sci-
entific/technical analysis aimed at generating a pre-
dictive understanding of, and technical basis for, stream
naturalization. The social methodology includes anal-
ysis of the historical development of the environmental
vision within each community, and case-study investi-
gations of current components of this vision and the
role of scientific information in sustaining this vision.
Scientific/technical research is developing and integrating
engineering-based modeling of stream dynamics with
geomorphological analysis of stream processes and
ecological analysis of physical habitat and fish pop-
ulation dynamics.
The two study sites for this project are the West
Fork of the North Branch of the Chicago River, North-
brook, IL, and Poplar Creek, Elgin, IL. Results of the
social analysis in Northbrook highlight some tensions
associated with the implementation of the concept of
naturalization. These tensions stem from the dichoto-
my between stakeholder values and desires and "ex-
pert" sense of best usage and vision. Naturalization
centers on the implementation of pool-riffle sequences
within a section of the North Branch in downtown
Northbrook. The research team has assisted local deci-
sionmakers with project design. The pool-riffle struc-
tures are based on ecogeomorphological principles and
have been tested via hydraulic modeling.
Social analysis in Elgin is examining discourses
of environmental resource provision. Current plans call
for a channelized section of Poplar Creek to be re-
meandered and its floodplain restored to presettlement
conditions. The creek is viewed as a natural resource
for community use. Social analysis is examining the
values, ideologies and languages that underlie the pro-
vision of this resource.
Geomorphological research has focused on the
pattern and movement of large woody debris in the
creek and characterizing the influence of this debris on
patterns of three-dimensional flow and channel erosion
and deposition in meander bends. This analysis pro-
vides a framework for ecological studies of fish and
macroinvertebrate habitat and for modeling-based as-
sessments of stream remeandering.
The Office of Research and Development's National Center for Environmental Research
33
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Selection of Wetland Restoration Sites
in Rural Watersheds To Improve Water Quality:
Integrating Ecological and Economic Approaches
Curtis J. Richardson, Randall A. Kramer, andNealE, Flanagan
Nicholas School of the Environment, Duke University, Durham, NC
The primary objective of this study is to develop
a procedure for configuring mosaics of restored wet-
lands on the landscape to yield the greatest positive
cumulative effect on watershed-level water quality
given a set of ecological, economic, and political con-
straints. This study focuses on the development of a
decision support system (DSS) to assist land managers
in the site locations of restored wetlands with the
objective of maximizing watershed-level water quality
improvement.
The DSS will rely upon a water quality model
that examines the watershed-level water quality impact
of restoring wetland areas. An economic model also
will be included in the DSS, which assesses the will-
ingness of landowners to participate in wetland resto-
ration programs based on their socioeconomic char-
acteristics, various aspects of the program, and other
factors that affect land-use decisions. Economic data
for use in the DSS were collected through a survey of
roughly 500 landowners/operators in selected areas of
North Carolina.
The survey was developed through an extensive
period of literature reviews, collecting background in-
formation, and conducting focus groups. It used a con-
joint analysis methodology to assess the preferences of
landowners for wetland restoration programs with dif-
ferent options. Various socioeconomic and land-use in-
formation also was collected. Soil, hydrology, and land
cover data were used to identify potential wetland res-
toration sites.
A classification and regression tree (CART) mod-
el was used to relate water quality and watershed char-
acteristics at sites monitored by the U.S. Geological
Survey National Water Quality Assessment Project. A
dynamic nonpoint source water quality model was
used to explore alternative restoration scenarios, and
site-level data at a coastal wetland restoration site were
used to validate the regional simulation models.
Survey administration commenced hi November
2000, and was completed in January 2001. Survey data
are being analyzed through statistical and econometric
methods to develop a model predicting the partici-
pation decisions of landowners/operators. Preliminary
survey results indicate that landowners generally fa-
vored restoration programs that allowed for shorter-
term contracts (10-15 years), that allowed them to
lease land for undeveloped recreatipnal use, and that
were administered by state agencies.
The ecological portion of this study examined the
relationship between wetlands and water quality at
"local, watershed, and regional spatial scales. Prelim-
inary statistical models utilizing both CART and dis-
criminant analysis models classified 300 samples into
one of three water quality categories using watershed
characteristics and hydrologic flux. The model pre-
dictions were correct for 80 percent of the sample anal-
yses.
The key findings from these studies will be in-
strumental in determining which sites have the highest
potential for wetland restoration success in terms of
water quality improvement on the landscape. Most im-
portant are findings concerning the reasons for and
willingness of landowners to participate in programs
to convert current agriculture lands into restored wet-
lands. After survey data have been analyzed, they will
be integrated into the ecological modeling effort to
build the DSS. The DSS will combine both biophysical
data on the water quality impacts of wetland restora-
tion and socioeconomic data, such as the willingness
of landowners to participate in various restoration
scenarios, relevant demographic information, and their
land-use preferences.
34
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
When Do Stakeholder Negotiations Work? A Multiple Lens
Analysis of Watershed Restorations in California and Washington
Paul Sabatier, William Leach, and Neil Pelkey
Department of Environmental Science and Policy, University of California, Davis, CA
This paper reports some preliminary results from
the Watershed Partnership Project at the University of
California at Davis. The project uses a database con-
sisting of random samples of approximately 60 water-
shed partnerships in California and Washington. In
each case, 3-5 diverse participants are interviewed,
mail surveys are sent to all participants and some know-
ledgeable outsiders, and relevant documents are coded.
This results in about 350 variables per case.
The project seeks to: (1) measure partnership suc-
cess on five different dimensions, and then (2) explain
variation in success using three different conceptual
frameworks: Transaction Costs Economics, Ostrom's
Institutional Analysis and Deliberation Framework,
and a version of the Advocacy Coalition Framework
developed by Sabatier and Jenkins-Smith that has been
•expanded to include Alternative Dispute Resolution
(Bingham, Carpenter).
Preliminary results from an analysis of 30 cases
reveals that development of trust within the partnership
usually is the most important factor explaining most
success measures. It also was found that a certain age
(usually around 40 months) is necessary for partnership
success, but that success does not necessarily increase
with age above that threshold.
' Members' satisfaction with the quality and ac-
cessibility of technical information also tends to be cor-
related with several dimensions of success. Conversely,
the size of a partnerships' budget, whether ithas a paid
coordinator, and the extent of belief conflict within the
partnership do not appear to be consistently associated
with success measures. The findings of this research
project should, however, be viewed with great caution
until the number of cases analyzed can be increased to
approximately 50 (it is hoped to accomplish this by the
time of the Progress Review).
The Office of Research and Development's National Center for Environmental Research
35
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Integrating Models of Citizens' Perceptions, Metal
Contaminants, and Wetlands Restoration in an Urbanizing Watershed
RobertK. Tucker1, George S. Hawkins1, Peter R. Jaffe2, Kerry Kirk Pflugh3, andBranden B. Johnson3
1 Stony Brook-Millstone Watershed Association, Pennington, NJ; ^Princeton University, Princeton, NJ;
3Division of Science, Research, and Technology, New Jersey Department of Environmental Protection,
Trenton, NJ
The overall goal of this project is to use the scien-
tific information from this research to increase public
understanding and support for the vital role wetlands
play in the integrity of watersheds. The approach in-
volves scientific investigations of metals interactions
in wetlands, education, and social science assessment
of the outreach efforts. One focus of this research is
nonpoint source pollution, particularly toxic metal im-
pacts on wetland function and water quality.
The nonpoint contamination is closely related to
the degree of development and intensity of human ac-
tivity within the watershed. This has been documented
in a comprehensive assessment and characterization of
one of the subwatersheds, Beden Brook. Data on water
quality, threatened and endangered species, invasive
species, contaminated sites, land use and management,
and area geology and demographics are included.
Research conducted by investigators at Princeton
University will provide detailed information on metals
behavior in the oxygen-depleted soils of wetlands, par-
ticularly as affected by the roots of plants. The Univer-
sity has made significant progress with electrochemical
techniques to measure concentration profiles of elec-
tron acceptors and trace metals in pore water and on
developing the model to simulate trace metal dynamics
in wetland sediments. Figures la, Ib, and Ic illustrate
the simulated profiles in wetland sediments of the key
electron acceptors, their corresponding reduced spe-
cies, ammonia, and arsenic. Arsenic is used to illus-
trate the effect that the wetland rhizosphere has on the
speciation of a metalloid of concern in the environ-
ment.
New Jersey Department of Environmental Protec-
tion social scientists have interviewed selected experts
on wetlands from federal and state government, aca-
demic, consulting, and nonprofit sectors for their at-
titudes about wetlands. They also have started con-
ducting a survey of citizens on their attitudes about
wetlands. All experts thought preservation of existing
wetlands was by far the best management approach,
although they differed fiercely on how to prevent wet-
lands development.
Information from this research already has played
a key role for local citizens in opposing such environ-
mentally harmful projects as a sewer extension into a
largely undeveloped and environmentally sensitive
area, and in persuading the Governor of New Jersey to
mandate a comprehensive environmental impact study
of a proposed road adjacent to wetlands along the Mill-
stone River.
A model stream protection ordinance for area
municipalities has been drafted, and environmentally
protective "river friendly" strategies for residents, golf
courses, and other businesses have been developed.
Venues for dissemination of information include the
Natural Lands Network, which the Stoney Brook-Mill-
stone Watershed Association helped organize. The
Network meets with 40 local land trusts, environmen-
tal commissions, and planning board members.
A Watershed Institute also has been formed to
provide support and assistance to growing watershed
associations—this research group is working with the
New Jersey Council of Watershed Associations to
unify these associations and advocate for policies that
protect water quality and natural resources. On January
29-30,2001, this research group sponsored a seminar
at the Woodrow Wilson School, Princeton University,
for local municipal officials. The seminar was titled
"PreservingThis Place Called Homeland included in-
formation on wetlands protection.
Next steps include field measurements of metals
and intensive characterization of another of the sub-
watersheds, Rocky Brook, in the headwaters area of
the Millstone River. Additional educational efforts and
surveys of the effectiveness of the educational out-
reach will be conducted.
36
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
concentration, nM
50 100 150 200 250 300
Mn(ll) and Fe(ll) concentration, jiM
12345
16
16
Figure la.
0 2000 4000 6000 8000 10000 12000
NH4* and HS" concentration, (iM
Figure Ib.
concentration, (iM
0 0.02 0.04 0.06 0.08 0.1 0.12
16
Figure Ic.
Figures la, Ib, and Ic. Simulated concentration profiles in wetland sediments. Figure la: Electron acceptors; Figure Ib:
Reduced species; and Figure Ic: Arsenic. Figure Ic shows the effect of roots on the simulated
arsenic profiles (D = no roots, W = roots that penetrate 8 cm into the sediments, which are the same
conditions as for Figures la and Ib).
The Office of Research and Development's National Center for Environmental Research
37
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Changes in River-Land Uses and Management: Implications
for Salmonid Habitat Restoration in the Lower Cedar River. Washington
Robert C. Wissmar', Thomas M. Leschine2, Ray TimmJ, David Fluharty2, and John Small3
'School of Aquatic and Fishery Sciences, * School of Marine Affairs, University of Washington, Seattle, WA;
*King County Department of Natural Resources, Seattle, WA
This joint societal-ecological research project is
elucidating how the distribution of human development
and conflicts affect salmonid habitat restoration efforts
in the Lower Cedar River Basin, and what restoration
efforts make riverine ecosystems and habitats more
resilient to human influence. Any biodiversity in the
system is beneficial, even if it is short lived, function-
ally. Also, approaches to increasing longevity are es-
sential given restoration costs.
The goal is to develop societal-ecological ap-
proaches that can be applied to restoration initiatives
and conflicts ranging from human actions that alter
local habitats to growth management policies that in-
fluence ecosystem functions at larger geographic scales
(e.g., connectivity between habitats, fish migrations).
Humans today and historically have modified the river
(e.g., flow regulation-diversion and channel-confining
structures), resulting in unnatural channel forming pro-
cesses that homogenize habitats. These actions and
other uses have caused the Cedar River and floodplains
to lose habitat biodiversity and to be unavailable to fish
and wildlife.
Contemporary and historic changes must be ac-
counted for in restoration planning. Conflicts between
human development and salmonid habitat restoration
are very common due to direct competition for land
and water resources. Problems arise when restoration
of habitats in river channels depend on existing man-
dates (Endangered Species Act listings) that conflict
with provisions for reducing impacts of flood hazards
on human activities. Other problems arise when resto-
ration activities within the basin are based on an op-
portunistic model. For instance, where resources are
available for specific actions (e.g., habitat mitigation),
there is a higher likelihood of restoration taking place.
Together these issues indicate the need to bring com-
prehensive, risk-based approaches to restoration plan-
ning.
The research group currently is using a geogra-
phic information system-based approach to identify
areas where human influences most compromise the
functional ability of current and potential restoration
sites. Some research questions include: (1) How can
contemporary habitat restoration best compensate for
the heterogeneity that has been lost? (2) How suc-
cessful are restoration projects that take advantage of
opportunities offered by other management programs?
(3) What approaches provide effective ways of resolv-
ing conflicts between human development and salm-
onid habitat restoration?
38
The Office of Research and Development's National Center for Environmental Research
-------
Section 3.
Projects Initiated With Fiscal Year 1997 Support
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Community Values and the Long-Term
Ecological Integrity of Rapidly Urbanizing Watersheds
M. Bruce Beck', A.K Parker', T.C. Rasmussen1, B.C. Patten2, KG. Porter2, B.C. Norton3,
and A. Steinemann3
'Warnell School of Forest Resources, Institute of Ecology, University of Georgia, Athens, GA; * School of
Public Policy and College of Architecture, Georgia Institute of Technology, Atlanta, GA
This research project seeks to integrate ecologi-
cal, hydrological, and social/policy sciences in a study
of a rapidly urbanizing watershed (Lake Lanier, GA),
where preservation of long-term ecological integrity is
perceived as being at stake. More speifieally, the goals
are to: (1) develop a concept of environmental deci-
sionmaking in which science-based models are respon-
sive to identified community values, as they evolve in
both the short and long term; (2) develop and apply a
procedure for identifying those scientific unknowns
crucial to the "reachability" of the community's de-
sired/feared environmental futures; and (3) improve
understanding of basic aspects of lake ecosystem be-
havior, with special reference to the roles of the micro-
bial foodweb, sediment-nutrient interactions, and geo-
chemistry.
There are six steps involved in the overall pro-
cedure of environmental decisionmaking. These steps
ideally would be implemented in the following se-
quence: (1) elicitation and elaboration of stakeholder
concerns for the future; (2) development of the math-
ematical model cast in terms (state variables) compat-
ible with these concerns; (3) computational analysis of
target futures; (4) communication to stakeholders of
plausibility (implausibility) of target futures; (5) iden-
tification of key scientific unknowns upon which plau-
sibility hinges, and specification of experimental stud-
ies designed to reduce these uncertainties; and (6) entry
into a second iteration of the sequence, from (1) adapt-
ed in the light of (4). Progress can be reported accord-
ing to the logic of this procedure.
Thus, this research group has experimented with
two instruments (a survey and a participatory "Fore-
sight" Workshop) designed to elicit stakeholder hopes
and fears for the longer term target futures (step [1]),
with varying degrees of success. In general, the survey
revealed an extremely high concern for the well-being
of Lanier, and a possibly counterintuitively yet greater
concern for the longer term, as opposed to the shorter
term, future—a form of "reverse time preference."
There clearly was a tendency for respondents to be
most troubled by threats over which they perceived
they (and their cohorts) had least control. The Fore-
sight Workshop appears to have proved to be the more
successful instrument in anchoring comunity/stake-
holder concerns to the quantitative scales attaching to
the state variables of these models (step [2]). One of
these, a foodweb model incorporating the microbial
loop hypothesis of Pomeroy, has been designed ex-
pressly for the purpose of analyzing the reachability of
the so-derived futures (step [3]).
Another model, the Generalized Lake Lanier Eco-
system Model, has been constructed to serve the pur-
pose of communicating scientific concepts to a (sci-
entifically) lay audience (step [4]), wherein the issue of
quality assurance—given the notion of "models as
languages"—also can be addressed, as an important
question of research in its own right. To identify from
the foodweb model the key scientific unknowns (step
[5]), computational and methodological extensions of
the so-called Regionalized Sensitivity Analysis have
been explored, specifically the multivariate procedure
of Tree Structured Density Estimation.
On the basis of these computational studies and,
more importantly, from intensive field work on man-
ipulating a small pond system, this research group has
come to the view that the classical paradigm-of P
cycling (see Figure 1) does not apply to the case of
Piedmont impoundments with iron-rich sediments. In
response to clearly and strongly expressed stakeholder
concerns (from step [1]), a significant portion of the
research under step (5) has been devoted to imple-
menting field work on the pond system intended to
provide a qualitative conceptual model of the propa--
gation and fate of pathogens in a water-watershed sys-
tem.
Within the scope of the present project, it seems
unlikely that the features of this conceptual model can
be given more quantitative expression. However, de-
velopment and preliminary testing of a biogeochemical
impoundment model have been completed. This model
combines fairly detailed accounts of the carbonate-pH
and Fe-sediment subsystems with the more conven-
tional nutrient (C, N, P)-phytoplankton subsystem to
explore the behavior of the vertical dissolved oxygen
(DO) profile and sediment-water interactions of Lanier.
In the light of what may have to be a revised view of
P cycling in Piedmont impoundments, this research
group is especially interested in understanding the
scope for P being cleaved from Fe under substantial,
transient, phytoplankton-induced excursions in DO and
pH conditions.
The Office of Research and Development's National Center for Environmental Research
41
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
The Lake Phosphorus Cycle
Improved Paradigm
Recirculated DIP
at Mixis
No
Elevated
DIP at
Mixis
Figure 1. The phosphorus cycling program in lakes is based on data from systems in northern temperate regions. Phosphorus
sorbs to iron-rich sediments from tributaries and runoff to form particulate inorganic phosphorus (PIP). Anoxic
respiration in the hypolimnion of north-temperate lakes creates strongly reducing conditions that liberate dissolved
inorganic phosphorus (DIP) from settling particulates. Hypolimnetic DIP steadily increases during summer
stratification as phosphorus is liberated, and is mixed throughout the water column at fall overturn. This paradigm
fails to explain phosphorus cycling in Southeastern Piedmont lakes. No increase in DIP is found in the anoxic
hypolimnion during summer stratification, nor is an increase in DIP observed during fall overturn in Southeastern
Piedmont lakes. It is hypothesized that the conventional paradigm is not appropriate in iron-rich Southeastern
Piedmont systems because: (1) iron-oxide sorption reduces the bioavailability of DIP, and (2) the abundance of
oxidized iron prevents DIP accumulation in the anoxic hypolimnion.
42
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Connecting Ecological and Social Systems:
Watershed Research Relating Ecosystem Structure
and Function to Human Values and Socioeconomic Behaviors
Gaboury Benoit, S. Kellert, M. Ashton, P. Barten, L. Bennett, D. Shelly, and S. Anisfeld
Yale Environmental Studies, New Haven, CT
The goal of this research project is to elucidate
ways in which ecological and social systems shape
each other, and to understand the mechanisms by
which the structure and function of natural systems
influence, and in turn are influenced by, human values
and socioeconomic behaviors. In the current phase of
the research, a paired watershed approach was used,
involving the collection of original data from 18 sub-
watersheds within the basin formed by the three rivers
(Qunnipiac, West, Mill) entering New Haven Harbor,
CT.
The aggregated area of the subwatersheds studied
totaled more than 8,000 ha and included the homes of
18,000 people. Subwatersheds were selected from trib-
utaries of the three major rivers and range in size from
113cr-826 ha and are drained by low-order streams (first
or second). Watersheds were chosen to include a broad
range of human population density, from rural to ur-
ban.
Within each subwatershed, teams of researchers
measured hydrological, chemical, and biological at-
tributes (39 variables) of the streams and the surround-
ing uplands. Direct observation and responses to sur-
veys of individual watershed residents also were used
to quantify the values, beliefs, and behaviors of the hu-
man residents of each subwatershed. Human responses
included a total of 25 variables collected by direct ob-
servation (household quality and neighborhood quality
measured on an ordinal scale, number of households
per hectare), and in the form of two surveys conducted
in each subwatershed. Most answers were used to cal-
culate a set of 15 indices that previously were devel-
oped. Responses from a mail survey in each subwater-
shed were used to calculate valuations of water quality,
appearance, biological diversity, and the willingness to
pay for conservation of these values. Finally, previ-
ously collected information on land use and land cover
(34 mutually exclusive categories) was used within
each subwatershed to quantify landscape patterns.
Principal components analysis (PGA) was used to
reduce the dimensionality of each data set. The number
of components was determined using the broken-stick
method in each case. Canonical correlation analyses
were used to evaluate the null hypothesis that there is
no linear relationship between variables within the two
multivariate data sets. Three such tests were conducted
to evaluate the relationships between biophysical and
social data sets (p > 0.05), biophysical and landscape
data sets (p < 0.01), and between social and landscape
data sets (p< 0.01).
PCA results show strong correlations between
landscape and biophysical variation and between land-
scape and social variation. Thus, these data are con-
sistent with the hypothesis that human beliefs, atti-
tudes, and values are associated with the structure of
the landscapes we live in, and in turn, with the compo-
sition and functioning of the ecosystems those land-
scapes represent. The original conceptual model con-
sisted of three components (ecosystem health, human
environmental values, and socioeconomic benefits)
linked to each other by direct, but unknown, feedback
processes.
Results to date suggest a different set of linkages
(see Figure 1), with ecosystem health and human envi-
ronmental values connected through the intermediary
of landscape structure. Interestingly, these associations
seem not to be strong functions. of differences in
wealth or education among people living in the subwa-
tersheds. Further statistical tests (factor analysis) indi-
cated that the relationship between social and biophys-
ical variables was manifest in certain subwatersheds
more than others, and this result corroborated elements
of the hypothesized link between human values and
ecosystem structure and function.
Additional funding is being sought for the next
phase of this research. The plan is to conduct large-
scale field experiments to probe causal mechanisms for
correlations revealed to date.
The Office of Research and Development's National Center for Environmental Research
43
-------
2001 EPA STAR/NSFAJSDA Water and Watersheds Progress Review
Ecosystem
Health
Biodiversity
Biomass
Water quality
Hydrologic regulation
Human
Environmental Values
Aesthetic Dominionistic\
Humanistic Naturalistic
Negativistic Moralistic
Scientific Symbolic
Utilitarian
Landscape Factors
Vegetative cover
Road network
Highways
Forested land
Waterways
Commercial/Industrial
Residential
Cropland
Figure 1, Results of this project to date suggest that ecosystem health and human environmental values are connected through
the intermediate of landscape structure.
44
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Social and Ecological Transferability
of Integrated Ecological Assessment Models
Linda A, Deegan', James Kremer2, and Thomas feebler3
'Marine Biological Laboratory, Woods Hole, MA; ^Department of Marine Sciences, University of
Connecticut, Groton, CT; ^Social and Environmental Research Institute, Leverett, MA
The goals and objectives of this research project
are to: (1) create a model of broad generality that links
land-use patterns and nitrogen loading to ecologically
important and socially relevant endpoints of water
quality, eelgrass habitat, and fish diversity and abun-
dance; (2) measure estuarine fish habitat and com-
munity structure throughout a range of estuaries in
southeastern New England, and develop empirical re-
lationships of biotic integrity and habitat quality; and
(3) investigate perceptions of ecological models and
modeling science by town planners, with the objective
of determining the best and most efficient way to en-
courage scientifically aware decisions at the crucial,
local level of land-use debate.
The team is working to develop an estuarine eco-
system response model (Changing Land Use and Es-
tuaries [CLUE]) that will aid in understanding how
several biological and physical factors influence the
range of ecological responses to nutrient inputs. The
goal is to develop a model that is more readily trans-
ferred to other systems, and thus amenable to man-
agement applications when extensive and expensive
scientific studies are unavailable or not feasible. One
unique aspect of the output is its probabilistic char-
acter, an important and usually underemphasized as-
pect that planners should understand if they are to
make realistic, scientifically informed decisions. Data
to build and test the model derive from studies of water
quality, fish habitat, and community structure in 16 es-
tuaries of Buzzards Bay and Cape Cod, MA, sampled
from 1993-1999, and eight estuaries of south coastal
Rhode Island and Connecticut.
The social science has focused on three major
components of estuarine science in the community:
(1) attitudes and beliefs about ecosystem models among
local governmental officials in small towns in southern
New England; (2) performance of a participatory mod-
eling tool (MANAGE) as applied in Jamestown, RI, by
the cooperative extension service of the University of
Rhode Island; and (3) characterization of views on
ecosystems models and their application in local deci-
sionmaking by modelers and outreach professionals in
southern New England.
In several submitted manuscripts, this research
group has documented large, eutrophication-driven
changes in the fish community and fish habitat struc-
ture during the past decade in 24 estuaries of south-
eastern Massachusetts, Rhode Island, and Connecticut.
It was shown that the biotic integrity of many of these
estuaries has been severely degraded from historical
levels over a relatively short time interval (1-5 years)
(see Figure 1). The estuarine biotic integrity (EBI)
index that was developed has been shown to have
broad applicability within the southern New England
ecoregion, and could be a valuable monitoring tool to
assess the recovery of ecosystem function after eutro-
phication remediation.
Juvenile fish community characteristics and fish
abundance in the studied estuaries are strongly influ-
enced by the integrity and sustainability of eelgrass
(Zostera marina) habitat. Healthy eelgrass beds are
necessary to sustain estuarine carrying capacity and
biodiversity. In those communities bordering estuaries
with extant eelgrass beds, land-use decisions can jeop-
ardize the integrity of this essential fish habitat. The
CLUE model of nitrogen loading and ecosystem re-
sponse will be a useful land-use planning tool. Im-
proved understanding of the use and perceptions of
ecological models by planners will facilitate design of
presentation styles and formats by scientists to pro-
mote informed land-use decisionmaking at the local
level.
Recent results show that the inherent optical
properties and the contribution of colored dissolved or-
ganic matter and nonpigment particles to water clarity
also may be important in controlling estuarine plant
and algal production. Previous work has shown very
unsatisfactory relationships between chlorophyll and
the diffuse attenuation coefficient in these coastal wa-
ters, and resolution of this issue is a sine qua non for
the modeling effort.
This research group also has found a striking mis-
match between local policymakers' desires for models
to reveal consequences of development on the scale of
a single building lot and scientists' admonishments of
using models to justify decisions at this level. Mod-
elers perceive the endpoints local decisionmakers re-
quest as improper uses of models.
This research group will integrate the EBI index
into the CLUE ecosystem response model and test its
usefulness in new estuarine settings. Field data from
1998-2000 will be used to formulate the bathymetric
and watershed characteristics that force the model, and
to display observed estuarine data against which to
evaluate model goodness-of-fit in as many as 13 es-
tuaries. It is anticipated that the Jamestown, RI, study
will be completed this spring, and analysis of results
will be written.
The Office of Research and Development's National Center for Environmental Research
45
-------
2001 I-.I'A S I \R NSI- LSDA \\'ater and Watersheds Progress Review
0)
20-
CO 10
LLJ
0
30
CO O if) O ao O
oo oo O O O O
0s O^ 0s 0s 0s O^
oo O
CO OO O
O 0s O*
co O
O O
Buttermilk Bay Waquoit Bay
i: hiolic integrity (I-!1H) index nicasiiremcnts averaged over sampling sites in Buttermilk Bay (four sites) and
\V;K|UOII Bay (five sites). I'JSN-I W). Dark gray bars indicate years in which eelgrass was present in at least two
sites in Buttermilk Bay and three sites in Waquoit Bay. Light gray bars indicate the absence of eelgrass in any sites.
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAJR/NSF/USDA Water and Watersheds Progress Review
From Landscapes to Waterscapes:
An Integrating Framework for Urbanizing Watersheds
P. Diplas1, E.F. Benfield2, D.J. Bosch3, W.E. Cox', R. Dymond1, D.F. Kibler1, V.K Lohani4,
S. Mostaghimi5, P.S. Nagarkatti2, D.J. Orth6, L.A. Shabman3, andK Stephenson3
'Department of Civil and Environmental Engineering, ^Department of Biology, ^Department of Agricultural
and Applied Economics, ''Department of Engineering Fundamentals, ^Department of Biological Systems
Engineering, 6Department of Fisheries and Wildlife Sciences, Virginia Polytechnic Institute and State
University, Blacksburg, VA
Urbanization, farming, and other types of land-
use change activities can significantly alter storm hy-
drographs and sediment erosion rates within a water-
shed. These effects can cause more frequent flooding
and increased pollutant loading, which in turn might
result in the degradation of the physical and biological
integrity of streams and other aquatic ecosystems.
Regardless of these problems, watershed development
appears inevitable.
The central issue, therefore, is how to reconcile
development with acceptable environmental quality at
an affordable price. The overarching goal of this pro-
ject is to develop procedures for integrated assessment
of the hydrologic, ecological, and economic conse-
quences of alternative landscape scenarios occurring
during the urbanization/suburbanization process.
More specific objectives include: (1) develop-
ment of an integrated hydrologic framework for assess-
ing the impacts of alternative landscapes on surface
and subsurface water flows and movement of sedi-
ments and pollutants; (2) development of procedures
to predict the response offish and macroin vertebrate
communities to urbanization-induced changes in water
quantity, water quality, and other biological condi-
tions; and (3) identification and assessment of policy
and economics consistent with alternative landscape
scenarios as well as estimation of the effects of alter-
native landscapes on land values and fiscal consequen-
ces for local governments. A case study focusing on
the Upper Roanoke River Watershed is employed to
test the operation of the integrated framework that cur-
rently is under development.
The various components of this research project
are addressed by the following groups: hydrology/-
hydraulics, biology, economics/policy, and geographic
information systems/problemsolving environment. The
computer models HSPF and MODFLOW are utilized
to determine the changes in the surface and subsurface
runoff taking place within a subwatershed triggered by
a certain amount of development. The first tends to
respond more rapidly while the latter, depending on
the local geology, might take many years to reach qua-
sisteady-state conditions. It is, therefore, this longer
time horizon that needs to be considered when exam-
ining the impacts of land-use change.
The output from these models becomes input to
RMA-2V, the computer model that was used to sim-
ulate the river flow. Usually, a major difference be-
tween streams in heavily developed areas compared to
those found in areas of low level of development is in
terms of channel topography or complexity. Biologists
have identified channel complexity as an important
ingredient of stream ecology. New methods have been
devised for modeling the localized flow patterns gen-
erated by these topographic features. Spatial metrics
have been developed and found to better describe the
abundance of stream habitat.
Extensive fish and macroinvertebrate data col-
lected from streams representing different size water-
sheds indicate that changes in diversity and richness of
species are connected with land-use practices. Veg-
etation strips along the river significantly reduce the
adverse effects of land-use changes within the water-
shed. Immunotoxicological tests indicate that fish may
serve as excellent biomarkers for environmental pol-
lutants, and therefore may act as warnipg systems for
human health hazards.
Four development footprints, which vary from
low to high density, are used to estimate the effects of
alternative settlement patterns on public service costs
andland value estimates, and the concomitant environ-
mental consequences. A computer interface that inte-
grates hydrological and economic aspects of this study
was developed recently. A Web-based version of this
model currently is underway. During the remaining
part of the project, the integration of the various com-
ponents, as well as issues regarding the interfaces
between hydraulics and biology, hydrology, and eco-
nomics will be examined further. A specific subwater-
shed has been selected to apply the entire modeling
procedure for various degrees of urbanization and
alternative settlement patterns.
The Office of Research and Development's National Center for Environmental Research
47
-------
2001 EPA STAR/NgF/USDA Water and Watersheds Progress Review
Conversion of Science
Into Management Decisions at Lake Tahoe (CA-NV)
CharlesR. Goldman', JohnE. Renter1, S. Geoff Schladow2, Alan Jassby1, M. Levant Kavvas2,
Alan C. Heyvaert', Theodore J. Swift1, and Jennifer E. Coker2
'Department of Environmental Science and Policy,''Department of Civil and Environmental Engineering,
University of California, Davis, CA
Deteriorating environmental conditions at Lake
Tahoe have been documented since the 1960s and
include loss in transparency of 0.3 m annually, in-
creased algal growth at 5 percent per year, changes
in biodiversity, increased loading of nutrients and
fine sediment, wetland loss, invasion of normative
biota, air pollution, and decline in forest health.. Data
suggest that if degradation continues, the remarkable
clarity of Lake Tahoe will be lost in 30 years. This
precipitated the Presidential Forum in 1997, and ne-
cessitated a more rapid conversion of basic limnolo-
gical studies into management decisions.
To date, long-term data collection and basic re-
search have been key to better understanding and man-
aging the lake, its surrounding watershed, and air
quality. With this grant and related investigations,
the primary goal is for science (monitoring, research,
and modeling) to assist in the restoration of water
quality and ecosystem health at Lake Tahoe. Reduc-
tion hi phosphorus and fine-sediment loading is con-
sidered critical to reduce the accelerated decline in
water clarity.
The approach can be summarized in a series of
three questions that need to be understood to achieve
effective management of lake clarity: (1) What are
the sources and relative contributions of nutrients
and fine sediments? (2) How much of a reduction in
this material is needed to attain desired water clarity?
(3) How will this reduction be achieved?
A budget has been completed that shows di-
rect runoff and atmospheric deposition (AD) as im-
portant P sources and highlights the need for addi-
tional study. Phosphorus reduction strategies will
have to address multiple sources, including direct run-
off, AD, streamflow, and groundwater. The contribu-
tion of AD to the N budget clearly dominates. Using
dissolved P to approximate biologically available P
(BAP), the budget shows that BAP is 30-35 percent of
total P; this is not uncommon. Research to determine
P bioavailability and its relation to restoration efforts
has been proposed. Sedimentation losses to the
bottom of Lake Tahoe are 401.7 MT yr'1 (TN) and
52.8 MT yr"1 (TP), and agree with loading estimates.
Watershed mitigation at Lake Tahoe may take
15-20 years to complete. Because the lake also has a
retention time of decades for nutrients, the direct ef-
fect of restoration on lake clarity may not be detected
for many years. Watershed and lake modeling provides
tools to overcome this long time lag. To explore
management options for loading reduction, a one-
dimensional modeling approach has been adopted.
The model, DLM, is driven hydrodynamically by
daily inputs of meteorological and streamflow data.
Water quality inputs are from streams, surface runoff,
groundwater, and atmospheric loading (see Table 1).
The model seeks to predict the distribution of
nutrient concentration, algal concentration, and sus-
pended particle concentration. Water clarity, a func-
tion of light absorption and scattering, can in turn be
calculated from the algal concentration and the size
distribution and concentration of particles. Intensive
data collection has been initiated to provide suffi-
cient calibration and validation data for the optical
part of the model. Preliminary results from the op-
tical component are presented in Figure 1.
Achieving loading reductions requires adaptive
management, wherein data are collected to: (1) opti-
mize best management practice effectiveness for in-
dividual projects, (2) assign project priority based on
nutrient/fine sediment sources, (3) provide expecta-
tions of a project's load reduction to be directly
coupled with a clarity model, (4) facilitate timely
changes to project design, and (5) assess success/fail-
ure.
48
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Table 1. Summary of loading estimates to Lake Tahoe expressed as metric tons per year.
1 Total N Total P Dissolved P [
Atmospheric deposition
Stream loading
Direct runoff
Groundwater
Shoreline erosion
Total
234 (56%)
82 (20%)
42 (10%)
60 (14%)
1 (<1%)
419
12.4 (26%)
13.3 (28%)
15.5 (33%)
4 (9%)
1.6 (3%)
46.8
5.6 (37%)
2.4 (16%)
3 (29%)
4 (27%)
No Data
15.0
35
•0-30
o>
£25
•a
£20
Q.15
&
Q
210
o
0
I
I
I
I
I
0 5 10 15 20 25 30
Secchi Depth Observed, m
35
Figure 1. Preliminary comparison of predicted Secchi depth transparency to actual field measurements over a wide
range of observed values.
The Office of Research and Development's National Center for Environmental Research
49
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
An Integrated Ecological and Socioeconomic Approach To Evaluating
and Reducing Agricultural Impacts on Upper Mississippi River Watersheds
Prasanna H, Gowda', Roger J. Haro', and TedL. Napier2
'Department of Biology, University of Wisconsin, La Crosse, WI;2School of Natural Resources, Ohio State
University, Columbus, OH
The objectives of this project are to: (1) use a
spatial-process model to predict agricultural discharges
from two watersheds in the Upper Mississippi River
Basin; (2) use the model to evaluate potential water
quality benefits associated with the adoption of alter-
native management strategies on these watersheds;
(3) develop regional-scale predictive models of eco-
system "health," biodiversity, and sustainability by
relating information on biota and ecosystem functional
processes to current and potential landscape compo-
sition and structure; and (4) identify factors that affect
adoption of conservation production systems among
land owner-operators in the two watersheds. The Low-
er Minnesota River Watershed in eastern Minnesota
and the Maquoketa River Watershed in northeastern
Iowa are being studied. The study recently was ex-
tended to include the Big Darby Creek Watershed in
central Ohio for a socioeconomic survey of farm own-
ers-operators.
Calibration of the ADAPT model, was performed
for six tributaries of the Lower Minnesota River Wa-
tershed. Statistical evaluation of the modeling results
indicates that model performance is best for subwater-
sheds dominated by agricultural land use. Pollutant
loadings from the remaining ungauged watersheds
were estimated using the calibrated ADAPT model.
For this purpose, subwatersheds in the Lower Minneso-
ta River Watershed were grouped into four regions:
(1) southeastern, (2) western, (3) north-central, and
(4) northeastern.
Four alternative agricultural management prac-
tices were developed considering developments in the
Lower Minnesota River Watershed in consultation
with local experts, and evaluated. They were: (1) in-
creased adoption of conservation tillage, (2) conversion
of crop land to pasture, (3) varying N-fertilizer appli-
cation rates, and (4) increased subsurface tile drainage.
Of the simulated scenarios, greatest reductions in ni-
trate-N loads (up to 7%) were observed in southeastern
and western regions of the watershed when N-fertilizer
application rates were decreased by 20 percent. Crop
lands in these regions typically are poorly drained. A
major portion of the land is equipped with subsurface
tile drainage systems. Model results indicate that a
75 percent increase in adoption of conservation tillage
can reduce average annual sediment loads by up to an
additional 57 percent. Increased adoption of conserva-
tion tillage also is predicted to increase nitrate-N los-
ses. The impact of conservation tillage on nitrate-N
losses offsets reduced losses of nitrate-N due to re-
duction in N-fertilizer application rates.
Relationships between landscape and stream hab-
itat characteristics and benthic macroinvertebrate com-
munity compositions in study watersheds were eval-
uated using macroinvertebrate data collected in 1998.
Soil erosion potential was estimated for key landscape
features using the University Soil Loss Equation. In the
Maquoketa River Watershed, stream habitats were of
relatively low quality and were highly variable across
sites. Benthic macroinvertebrate-index of biological
integrity (BM-IBI) scores were strongly related to
stream habitat variables, but were not related to soil
erosion potential.
In the Lower Minnesota River Watershed, stream
habitats were of relatively higher quality and were
evenly distributed across sites. BM-IBI scores were not
related to individual habitat variables, but had a strong
relationship to soil erosion potential. In both study
watersheds, BM-IBI scores increased with increased
rates of soil permeability and conservation tillage;
however, percent of row crop agriculture or forested
land had no effect on BM-IBI scores.
Data collected by a socioeconomic survey of land
owner-operators were analyzed and compared. Study
findings revealed that farmers in all of the study wa-
tersheds had adopted some form of soil and water con-
servation production practices; however, they contin-
ued to use production practices that could negate the
positive environmental benefits of the conservation
practices employed at the time of the data collection.
Many farmers reported that they did not expect to
transfer their farm operations to their children.
At present, this research group is operating on a
1-year no-cost extension. Efforts will concentrate on:
(1) modeling the Maquoketa River Watershed for wa-
ter quality, and developing and evaluating various
alternative management practices; (2) processing, tax-
onomic identification, and analysis of the macroinver-
tebrate samples collected in 1999; and (3) developing
manuscripts for publication in peer-reviewed journals
and presenting research findings at water resources-
related professional conferences.
50
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Nutrient Sources, Transformations, and Budgets
at the Watershed Scale in Ipswich River, Massachusetts
Charles S. Hopkinson1, E. Rastetter1, J.V. Vallino1, M. Williams1, andR.G. Pontius2
'Marine Biological Laboratory, Woods Hole, MA; 2Clark University, Worcester, MA
Patterns of land-use change, nutrient and water
export from land, riverine processing and retention of
nutrients, and the overall mass balance for nutrients in
the Ipswich River Watershed axe being studied and
modeled. There have been major changes in land cover
and land use since European settlement. For example,
agricultural land has decreased by more than 70 per-
cent, and urban land now comprises about 32 percent
of the total cover.
• Land-use change is modeled based on spatial
physical factors, legal constraints, and extrapolations
of quantities of change. Maps of suitability for defor-
estation are calibrated with maps of real change be-
tween 1971 and 1985 by using multicriteria analysis.
The maps of 1971 and 1985 also serve as the basis to
extrapolate the quantity of predicted future deforest-
ation. The extrapolated quantities and calibrated suit-
ability maps predict the location of deforestation from
1985-1991. The predicted deforestation maps are val-
idated with the map of real forest area of 1991. To pre-
dict land-use change into the future, a validated sim-
ulation method is used to sketch various scenarios.
The relation between land use and nutrient export
is being evaluated in two ways. At the finest temporal
scale, flow-integrated sampling is used in first-order
streams draining 1-1.5 km2 catchments with homoge-
neous land cover: forest, urban, or agricultural. At the
broadest spatial scale, seasonal sampling of 60-80 first-
order streams draining 1-1.5 km2 catchments of vary-
ing land use is employed. All phosphorus fractions and
ammonium generally are not related to land use and
are present in low concentrations. NO3" concentrations
can exceed 100 |iM and are strongly related to the
fractional cover in urban and agriculture uses. Dis-
solved organic nitrogen and carbon (DON and DOC)
are related to the relative amount of wetlands within a
catchment, but only during late summer and fall. Total
dissolved nitrogen typically is most related to frac-
tional urban and agricultural land uses.
The land-use change model has been coupled
with the land-use/nutrient relationships to evaluate
how stream nutrient loading might increase in the fu-
ture with and without laws restricting development.
For every decade in the two land-use change scenarios,
an empirical nitrate loading relationship is applied to
several hundred subcatchments. The results are time
series of spatially distributed surfaces of nitrate load-
ing, with corresponding transects of potential nitrate
concentrations for the Ipswich River Stream Network.
Simulation results for realistic development scenarios
indicate a doubling in nitrate delivery to the Plum Is-
land Sound Estuary from 1991-2101.
The changing landscape and associated socio-
economic activities, together with a changing climate,
are having a major impact on the watershed hydrology.
Precipitation has increased at a rate of 3 mm/year since
1933, while streamflow has remained relatively con-
stant, even after correcting for increased net diver-
sions. The long-term water budget indicates an in-
crease in evapotranspiration. The analyses show that
during the period of 1949-1998, the effect of a chang-
ing climate on evapotranspiration is stronger than the
effect of a changing land use. Future water budgets
also have been predicted using global climate model
output and output from the land-use change model. At
the anticipated rate of urban expansion and climate
change, the Ipswich River Basin will experience seri-
ous problems in allocating its water resources within
the next 100 years. The research group is working with
the Massachusetts Executive Office of Environmental
Affairs, the Department of Environmental Protection,
the Ipswich River Watershed Association, and the U.S.
Geological Survey to develop sound, sustainable wa-
tershed management plans for the watershed.
The Hydrologic Simulation Program-Fortran
(HSPF) hydrologic and nutrient processing model is be-
ing used to examine spatial aspects of nutrient sources,
sinks, and export to the coastal zone. An extensive da-
tabase has been developed on in-stream nutrient con-
centrations that, when coupled to the HSPF model,
allows for the identification of critical habitats in the
Ipswich River Watershed where nutrient retention and
processing are strongest. First-order streams draining
urban land covers and extensive riverine wetlands are
the primary sites where nutrients (N) are retained.
Finally, a whole watershed budget of nitrogen in-
puts, losses, and export has been constructed. Nutrient
mass balance indicates that of about 500-1,000 Mt N
input to the watershed annually, about 64-82 percent
is retained on land and 8-16 percent is retained in the
river. In addition to substantial N losses during down-
stream transfer, there also are changes in the compo-
sition. Inorganic components comprise 58 percent of
total N in first-order streams, while DON and panicu-
late N comprise 61 percent of total N at the river mouth.
With a predicted doubling of nitrogen inputs to first-
order streams over the next 100 years, there is concern
about the uptake capacity of the river system and
whether nutrient loading to the productive Plum Island
Estuarine System will increase concomitantly.
The Office of Research and Development's National Center for Environmental Research
51
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Linking Watershed-Scale Indicators of Changes
in Atmospheric Deposition to Regional Response Patterns
J. Kahl1,1. Fernandez2, J. Rubin3, J. Cosby4, S. Norton 5, L. Rustad1- \ D. Mageean3, and P. Ludwig6
'Water Research Institute, * Department of Plant, Soil and Environmental Studies,3Margaret Chase Smith
Center for Public Policy, ^Department of Geological Sciences, University of Maine, Orono, ME;
4Department of Environmental Sciences, University of Virginia, Charlottesville, VA; 'international Paper,
Inc., Bucksport, ME; 7U.S, Department of Agriculture Forest Service, Durham, NH
This research group is determining the patterns
and indicators of response of a specific ecosystem re-
sponse to experimental watershed acidification. A
major goal is to scale this knowledge to the regional
level to determine the extent of acidification and N-
saturation in a sensitive subpopulation of high eleva-
tion lakes. Concurrently, the Tracking and Analysis
Framework (TAP) model used in the Adirondacks un-
der the National Acid Precipitation Program (NAPAP)
is being recalibrated to examine the effects of acid
deposition on Maine's high elevation lakes.
The project activities are hierarchical from site-
specific to regional. First, the indicators of and model
predictions for acidification and N saturation have
been examined in the whole-watershed N enrichment
experiment at the Bear Brook Watershed (BBWM),
ME. This site is in its 11th year of experimental treat-
ment with dry ammonium sulfate. From new and on-
going data collection on soils, stream chemistry, and
forest growth, the indicators of response at this site are
being assessed (see Figure 1).
Second, the site-intensive information is being
scaled to the region using data from High Elevation
Lakes in Maine (HELM) and their watersheds. Their
acidification status is similar to lakes in the Adiron-
dacks. The potential for parallel relationships between
soils and surface water in HELM and BBWM is being
tested.
Third, the HELM data are being used to recal-
ibrate the aquatic effects submodel within the TAP to
predict changes in acid neutralization capacity, pH,
aluminum, and calcium using the Model of Acidifi-
cation of Groundwater in Catchments (MODEL). The
possible effects of increases and decreases in N and S
on lake chemistry and the viability of HELM fish pop-
ulations are demonstrated.
The whole-watershed experiment at BBWM has
resulted in increases in S and N flux of 2x and 20x,
respectively. The process of acidification has altered
base cation flux and ratios, resulting in a depletion of
Ca relative to Mg in streamwater, which is inferred to
reflect changes in watershed soils. HELM lakewater
Ca is positively and significantly correlated with soil
Ca saturation. Aluminum concentrations are inversely
correlated with soil Ca saturation. Nitrate concentra-
tions remain high in these lakes in contrast to the re-
gional trend, confirming the acidified condition of
these lakes, but without significant correlations with
measured watershed N properties.
This information will be used by the Environ-
mental Protection Agency to meet the Congressional
mandate of determining the effectiveness of the Clean
Air Act Amendments on influencing these trends. Site-
specific data from BBWM scaled to the regional
HELM population also will provide a template for the
recognition and understanding of possible N saturation,
base cation depletion, and indicators of ecosystem re-
covery. The recalibration of the TAF for Maine's high
elevation lakes advances the science of integrated nat-
ural and social science research by providing the ability
to contrast the effects of acidification in Maine's lakes
with those of the Adirondacks.
The project is in the final stages of data analysis.
The relationships between soil chemistry in the two
watersheds at BBWM and in the HELM lakes are
being analyzed. The soils data will be related to surface
water chemistry to develop indicators of response.
Deliverables will include papers on the indicators of
recovery in natural systems, the controls of N chem-
istry in high elevation lakes, the experimental response
at BBWM, and the application of the TAF to Maine
10 years after the NAPAP.
52
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
g
o
o
^
C/)
1.0:
0.8 :
0.6 \
0.4 -
0.2 -i
o.o-l
High Elevation Lakes
1980s
an
A
' I ' 1 | • 1 1 r r l | .
-50 0 50 100 150 200 250 300 350 400
1
0
0
0
0
0
o-
8-;
6 •:
4 -
2 -
0J
D
D m£
+*
^^fen
A tt ti
A.-A B"
^% n n
n n n !5Ln ^A
ana A n
= ° *
n
a a
-50 0 50 100 150 200 250 300 350 400
ANC
Figure 1. The sulfate fraction is the ratio of sulfate to total anions. The decline in this ratio, especially the loss of the high
ratios, is an indicator of recovery in these lakes despite the lack of increase in alkalinity (ANC').
The Office of Research and Development's National Center for Environmental Research
53
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
REKA. a New Comprehensive Watershed Management System
C. Gregory KnightJ>2>3, Jeffrey J. Carmichael2'7, Heejun Changlt2, Dimitar Dimitrov s, Barry M. Evans3,
James M. Hamlett3-5-6, TodorN. Hristov5, Vania D. loncheva1, Ivan I. Nikolov5, Marieta P. Staneva3'4,
andPetkoS. Varbanov5
'Department of Geography, 2Centerfor Integrated Regional Assessment,3 Environmental Resources Research
Institute, 4Penn State Altoona, Pennsylvania State University, University Park, PA; Institute of Water
Problems, Bulgarian Academy of Sciences, Sofia, Bulgaria; 6University of Architecture, Civil Engineering and
Geodesy, Sofia, Bulgaria; ^University of British Columbia, Vanvcouver, BC, Canada
River Environmental Knowledge and Assess-
ment (REKA) is a spatial water quality assessment
system designed to answer two questions: (1) If pol-
lution control is invested in at one place, where and
to what degree will stream quality goals be achieved?
and (2) To attain certain quality standards for a given
reach (or reaches) of the basin, what alternative strat-
egies could be implemented under various probabil-
istic levels of stream flow resulting from climatic
variability and with input from the local community?
REKA was developed in the context of the Yantra
River Basin (see Figure 1) in Bulgaria (reka = "riv-
er" in Bulgarian) as a transferable geographic infor-
mation system (GlS)-based tool to link process and
decision models related to water quality in a compre-
hensive framework. The project is a collaboration be-
tween the Institute of Water Problems in the Bulgar-
ian Academy of Sciences and the Environmental
Resource Research Institute and the Center for Inte-
grated Regional Assessment at the Pennsylvania State
University.
REKA has three subcomponents (see Figure 2).
Basin Impacts of Simulated Transport from Rural
Areas (B1STRA; bistra = "clean" in Bulgarian) uses
a new articulation of the Generalized Watershed
Loading Function in the ArcView GIS program. This
component calculates river loads of nutrients (N, P)
and sediments based on weather, soil, topography, and
land use, including both nonpoint and point sources
of pollution.
Validation and Optimization for Decision Anal-
ysis (VODA; voda = "water" in Bulgarian) is a sub-
stantial revision of the STREAMPLAN model of the
International Institute of Applied Systems Analysis.
This component uses a reduced-form process model
to derive stream hydraulics and pollutant transport
and processing. REKA delivers loadings and water
volume data to VODA by stream reach. VODA then
computes flows, pollutant concentration, and loads by
reaches.
In VODA, the user has several model design
choices. VODA can be used to generate a scenario
based on current conditions (e.g., for calibration and
validation) or to simulate the water quality impacts
of proposed treatment facilities or discharge regula-
tions. Alternatively, VODA can calculate financially
optimal strategies to achieve water quality improve-
ment goals or to meet specified standards for all or
some of the stream reaches. Decision choices include
alternative allocations of reservoir water for dilution
(with lost value from alternative uses); temporary or
permanent closing of polluting entities; and capital
and operational costs of pollution pretreatment and
treatment.
VODA also provides for input of alternative
weather conditions, based on the validation period or
a typical wet, average, or dry year. It also is possible
to input temperature and precipitation change scenar-
ios derived from global climate change models. Typ-
ically, simulations will be framed in terms of low-
flow months during relatively dry years.
The numerical results of VODA's simulation
and optimization then are passed back to ArcView
for presentation of Protection Location and Action
Network (PLAN; the same meaning in Bulgarian as
in English). PLAN provides maps of water quality
status and tabular information on goal achievement,
cost, and decision variables that entered into the so-
lution. Dissemination of the model includes a CD in
preparation that will have the full model structure for
a simple, heuristic river basin, as well as selected
scenarios derived for the Yantra Basin itself.
54
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
YANTRA RIVER BASIN
BULGARIA
Figure 1. The Yantra River Basin, a Danube sub-basin in Bulgaria.
REKA
PEKA
River Environmental Knowledge and Assessment
A Spatial Water Quality Assessment System
BISTRA
BHCTPA
Basin Impacts of Simulated
Transport from Rural Areas
VODA
BOJJA
Validation and Optimization
for Decision Analysis
PLAN
HJIAH
Figure 2. REKA, a spatial water quality assessment system.
The Office of Research and Development's National Center for Environmental Research
55
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Coping With Nature: Accepting Risk,
Adopting Technology, and Assuming Ignorance
James McManus', Courtland L. Smith2, Jesse Fofd3, Paul D. Komar4, Debbie Colbert4,
and Michael Styllas4
'Large Lakes Observatory, University of Minnesota, Ditluth, MN; ^Department of Anthropology, ^Department
of Fisheries and Wildlife, 'College of Oceanic and Atmospheric Sciences, Oregon State University,
Corvallis, OR
The National Research Council (NRC) recom-
mends evaluating incentive-based approaches to envi-
ronmental regulation to replace command-and-control
measures. The NRC further recommends the use of
comparative examples that combine natural scientific
study with risk assessment and benefit-cost analysis.
Landowners in the Tillamook Basin, OR, have
150 years of adapting to natural processes that pose
risks of flooding, tidal currents, wind and wave action,
fire, disease, earthquake, and tsunami. In addition, land
uses have contributed to water quality problems that
affect the ecology of Tillamook Bay, particularly oys-
ter growing and fishing.
This group's research on sediments, water chem-
istry, ecosystem health, and social science has identi-
fied a number of situations in which landowners make
risk-benefit calculations. The results of these calcula-
tions include building the dairy industry and a major
commercial area in the flood plain. Another is plan-
ning a destination tourist resort on a sand spit breached
by ocean waves. In trying to explain these decisions, it
was found that landowners conducted their own risk-
benefit calculations, employed technology to minimize
risk, and accepted higher risk because of ignorance.
Why do many landowners make decisions that
appear to go against natural processes? One hypothesis
is that they are ignorant of the effects of natural pro-
cesses. A second hypothesis is that they think tech-
nology can overcome the forces of nature. A third hy-
pothesis is that they make a risk-benefit calculation.
Data support each of these hypotheses, but risk-benefit
calculations explain more than the others.
These hypotheses were tested in a number of sit-
uations in the Tillamook Basin. For example, historical
evidence shows a pattern of learning about the effects
of natural processes, the use of technology to minimize
risk, and most of all, calculated risktaking. Learning
began in the 1850s, when the first settlers brought
dairy cattle to the area. The dairy industry persists be-
cause of risk calculations made by dairy owners about
relative economic impacts of water quality problems
and the probability of floods wiping out their herds.
The dairy industry is the biggest contributor of
fecal coliform to the Bay, which is closed to oyster
harvest approximately 100 days per year. Agriculture
persists because the risk of major reductions in fecal
coliform levels is far greater than the oyster harvest
and fishing benefits. Further, much of the science as to
causes and processes still is uncertain. The other major
natural threat to agriculture is flooding. Here, the effort
is to make modifications in the basin to move flood-
waters out as quickly as possible and keep the number
of milkings missed to fewer than six. More than six
missed milkings results in the loss of productive cows.
Examination of other decisions about forest, ur-
ban, and recreation land use suggests that reduction of
ignorance, technological innovation, and risk-benefit
calculations all affect land-use decisions. The risk-ben-
efit calculation typically is the one hypothesis that ex-
plains most about landowner decisionmaking. These
examples of risk-benefit analysis by local landowners
responding to market incentives show that people will
take risks against natural hazards. They often are ig-
norant of risk probabilities, and these are not well com-
municated in the community.
The risk assessments are not quantitative. Land-
owners seek technology believing they can armor
themselves against risks. Because of ignorance and be-
lief in the benefits of technology, the risk-benefit cal-
culation is often wrong in the short term, but the col-
lective result over the long term is local adaptation to
the occurrence of natural processes.
56
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Ecological Risks, Stakeholder Values, and River
Basins: Testing Management Alternatives for the Illinois River
Mark Meo', Baxter Vieux*, Blake Pettus3, EdwardSankowski4, Robert Lynch s, Will Focht6,
Keith Wttlett', and Lowell Caneday s
'Science and Public Policy Program, ^College of Engineering, 3College of Architecture, ^College of Arts and
Sciences, University of Oklahoma, Norman, OK; ^College of Public Health, University of Oklahoma Health
Sciences Center, Oklahoma City, OK; 'College of Arts and Sciences, 'College of Business Administration,
sCollege of Education, Oklahoma State University, Stillwater, OK
The objective of this research project is to iden-
tify and compare different environmental and social
values held by stakeholders in Oklahoma's Illinois
River Watershed, and to test a management protocol
that is technically effective, economically efficient,
and socially and politically acceptable. In the first
phase of the project, baseline technical, economic, and
sociopolitical assessments were conducted that serve
as the basis for subsequent interactive visualization
workshops with policymakers and stakeholders to de-
fine alternative management policies that meet the
three criteria for acceptance.
In the second phase, alternative policy assess-
"ments are being modeled and prepared to help water-
shed stakeholders attain consensus about alternative
land and water uses. In the final phase of the project,
the project's acceptability will be determined through
a telephone survey of watershed stakeholders.
The technical analysis involves a geographic in-
formation system (GlS)-based modeling of watershed
hydrology and nutrient flows, and ecological risk char-
acterization of selected species. Distributed parameter
hydrologic modeling has been used to relate land-use
alternatives to impacts on water quality. Simulations
use digital representations of rainfall, soils, topogra-
phy, and land use in GIS format. To date, the project
team has collected water quality data (see Figure 1)
over the period for which the distributed runoff non-
point pollution (DRIP) model is run; developed model
data sets derived from GIS maps of soils, land use, top-
ography, and radar estimates of rainfall; assembled
GIS maps that depict population density, roads/trans-
portation, and political boundaries; and completed the
visualization of the digital elevation models and water
quality simulations using the AVS software. The ob-
jective of the ecological risk effort has been to identify
at-risk ecological communities that: (1) rely on avail-
able data, (2) enable quantification of visually com-
municable risk measures, (3) are relevant to groups of
stakeholders, and (4) are responsive to alternative man-
agement strategies.
Accomplishments include the collection of data
from sites on 24 tributaries within the river basin over
the past several years to monitor the influence of land
use on receiving waters, and a comprehensive survey
of all agency databases and the calculation of com-
munity integrity indices for fish, periphyton, and ben-
thic macroinvertebrates. The economic assessment has
been focused on the development of databases and
modeling structures that capture the most significant
economic activities in the watershed.
Regional impact of these activities is estimated
through the use of the IMPLAN input-output model,
which enables the calculation of the economic impact
of individual expenditures on total gross output, em-
ployment, employee compensation, property income,
value added, and indirect business taxes. In addition,
the role of poultry production in the regional environ-
ment and economy has been analyzed by incorporating
poultry feeding and production decisions for disposal
of waste litter. The sociopolitical assessment has been
undertaken and completed through a series of joint ac-
tivities that were based on stakeholder interviews in
each of eight study regions identified for the water-
shed. Concurrent activities included 150 in-person
interviews, 60 mental modeling interviews, and 120 Q
methodological interviews.
Computer visualization is being used as a deci-
sion support tool to facilitate stakeholder understand-
ing and as an aid for negotiating alternative land and
water use policies. The visualization team has focused
on: (1) developing a graphic infrastructure for organ-
izing research data; (2) collecting data on the back-
ground, history, and visual character of the watershed
and converting this into digital format; (3) developing
digital standards required to compile the work of the
different research teams; (4) developing an effective
approach to graphically compiling, rendering, and ani-
mating multiple large-format DEM data; and (5) crea-
ting color-relief imagery from high-resolution DEM
data. This work has been completed.
Future technical activities will focus on charac-
terizing policy options with the developed DRIP and
IMPLAN models and presenting them in a visual for-
mat for use in a planned sequence of stakeholder and
policymaker negotiation workshops.
The Office of Research and Development's National Center for Environmental Research
57
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Oklahoma
Arkansas
Crawford
Sequoyah
I I Illinoii River Subbatint
County Lines
.0. Water Testing Stations
Rivers
m Mii,.s
Figure 1. Illinois River water testing stations.
58
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Balancing Risks of Flood Control
and Ecological Preservation in Urban Watersheds
Vladimir Novotny, D. Clark, R. Griffin, A. Bartosovd, andD. Booth
Institute for Urban Environmental Risk Management, Marquette University, Milwaukee, WI
This study employs an interdisciplinary approach
to investigate community support for watershed man-
agement initiatives. Watershed professionals often ad-
dress two consequences of urbanization. First, urban
growth frequently exacerbates downstream flooding
problems (magnitude, frequency, spatial scope) due to
increased storm runoff. Second, urbanization also in-
creases ecological risks as less tolerant species disap-
pear due to loss of wetlands and diminished water
quality. When assessing policy alternatives, watershed
professionals must determine community support for
flood control objectives vis & vis ecological risk reduc-
tion.
This research has the following specific objec-
tives: (1) develop statistical flow, loading, and water
quality models for flood risk assessment; (2) develop
ecological risk assessment procedures to estimate eco-
Io"gical consequences of urbanization; (3) simulate
flooding/ecological risk assessment for urbanization
scenarios; (4) adapt a contingent valuation approach to
spatially assess individual willingness to pay (WTP)
from flood control and water quality improvements
within urban watersheds; (5) examine the impact of
drivers (spatial, demographic, attitudinal) of stated
WTP for flood/ecological risk reduction; (6) determine
temporal stability of WTP for flood/ecological risk re-
duction; and (7) derive communitywide benefit/cost
estimates for watershed management alternatives.
A hydrologic model of flooding risk within a
Milwaukee watershed was developed. The modified
methodology uses data from an existing hydraulic
model to delineate floodplain based on the flow of a
specified recurrence interval. Regression analysis ties
the flow to depth and floodplain width. Results then
are incorporated into a geographic information system
software package, permitting more precise evaluation
of existing flood risks within the Watershed. This meth-
odology also allows flood risk changes from urban-
ization to be computed.
A model of ecological risk also has been devel-
oped. The information on habitat suitability and eco-
logical risk due to chemical contamination is being
analyzed with respect to overall biotic integrity offish
and macroinvertebrates. A habitat index has been, used
to assess the effect of habitat on an index of biotic
integrity (EBI). All sites investigated show a decrease
in IBI greater than the decrease in habitat index, or the
level of habitat impairment, would indicate. This
shows significant chemical impairment of water bod-
ies. A simple software package was developed to facil-
itate calculation of ecological risks from hypothetical
chemical contamination.
A telephone survey was developed, and the first
of two waves was conducted in fall 1999/spring 2000
on 999 randomly selected adult residents of two Mil-
waukee watersheds. Respondents were provided with
a description of a hypothetical referendum for a water-
shed project based on the hydrologic models. They
stated their annual WTP for the next 20 years. Pr^lim-
inary findings suggest that: (1) WTP is independent of
the project scope (i.e., projects with more risk re-
duction are not more highly valued); (2) there is a re-
lationship between WTP and sociodemographic char-
acteristics of the respondent; (3) WTP is influenced by
location within the watershed (e.g., upstream/down-
stream); and (4) artitudinal factors related to subjective
norms and overall belief structures appear to influence
WTP.
The stability of responses will be reviewed after
the second survey wave. Benefit estimates will be com-
puted for each watershed and compared with expected
costs.
The Office of Research and Development's National Center for Environmental Research
59
-------
2001 EPA STAR/NSF/USDA Water and Watersheds Progress Review
Impact of Social Systems on Ecology and Hydrology
in Urban-Rural Watersheds: Integration for Restoration
Steward T.A. Pickett', J.M. Grove2, LW. Band3, K.T. Belt4, G.S. Brush5, W.R. Burch, Jr.6,
M.L. Cadenasso', J.M. Carrera7, G.T. Fisher*, P.M. Groffman1, R.V. Pouyat4, and W.C. Zipperer4
'Institute of Ecosystem Studies, Millbrook, NY; 2U.S. Department of Agriculture Forest Service, Burlington,
VT; ^University of North Carolina, Chapel Hill, NC; 4U.S. Department of Agriculture Forest Service,
Syracuse, NY; * Johns Hopkins University, Baltimore, MD; 6Yale University, New Haven, CT; ''Parks and
People Foundation, Baltimore, MD; SU.S, Geological Survey, Baltimore, MD
This collaborative project, representing hydrol-
ogists, social scientists, plant ecologists, engineers,
landscape ecologists, and outreach specialists, seeks to
develop novel hydro-ecological models that integrate
key social drivers of watershed function at various
scales in Metropolitan Baltimore, In addition, it is
sought to interact with stakeholders, managers, and
decisionmakers to improve and test the utility of the
model in the real world.
The project focuses on the 17,000-ha Gwynns
Falls Watershed that extends from the suburban fringe
to the densely built residential, and ultimately, indus-
trial areas at the mouth of the stream. U.S. Geological
Survey gauging stations, which are located at con-
trasting reaches of the stream, in representative but
contrasting subcatchments of the stream, and in a ref-
erence forested watershed, provide the substrate for
model development. The project is integrated by the
patch dynamics approach, the human ecosystem frame-
work, and hydrological models that take into account
human and social capital as well as the more tradition-
al inputs of built and natural capital.
The model is being developed to allow dynamic
feedbacks between the four major realms of capital.
The research project brings together specific data such
as microclimate, soils and slope form, riparian and up-
land vegetation, social-ecological spatial patch pattern,
and social structure and processes.
Testing the models will use, in part, paleoeco-
logical and historical data on infrastructure, land cover,
human population, and social indicators. Projections
will be based on contrasting land-use and economic
scenarios. The models have an explicit spatial com-
ponent that allows the role of biogeophysical and "so-
cial heterogeneity on watershed function to be as-
sessed.
The insights provided by recent additions to the
knowledge base include: (1) the failure of land-use
models to routinely include human decisionmaking;
(2) the role of social heterogeneity in increasing eco-
logical heterogeneity in the metropolis; (3) unexpected
locations of environmental hazards in the watershed;
(4) improved water quality downstream in the urban-
suburban watershed, with expected low impairment in
the forested reference watershed; (5) increased compo-
sitional and spatial heterogeneity in urban vegetated
plots; and (6) promising utility of software being de-
veloped to support the integrated hydro-ecological
models. Close interactions with the Patuxent Land-
scape Model is a feature of the project.
60
The Office of Research and Development's National Center for Environmental Research
-------
2001 EPA STAR/NSF/OSDA Water and Watersheds Progress Review
Index of Authors
Allan, ID., 21
Armbrust, K.L., 3
Beck,M.B.,41
Benoit, G., 43
Boll, I, 22
Bolte, I, 24
Bonzongo, J.C., 25
Costanza, R., 27
Day, I, 4
Deegan, L.A., 45
Diplas, P., 47
Farrow, S., 6
Goldman, C.R., 48
Gowda, P.H., 50
Hopkinson, C.S., 51
Kahl, J.5 52
Knight, C.G., 54
Kraft, S., 28
Lathrop, R.C., 8
Maddock, T., 30
Matlock,M.D.,31
Matthews, O.P., 10
McManus, J., 56
Meo, M., 57
Novotny, V., 59
Opalucb, J.J., 32
Palmer, M.A., 11
Pickert, S.T.A., 60
Rhoads, B.L., 33
Richardson, C.J., 34
Richmond, R.H., 13
Sabatier, P., 35
Sanders, B., 15
Sparks, R.E., 16
Tucker, R.K., 36
Turco,R., 17
Welty.C., 18
Wissmar, R.C., 38
The Office of Research and Development's National Center for Environmental Research
61
-------
&EPA
United States
Environmental Protection Agency
(8701R)
Washington, DC 20460
Official Business Only
Penalty for Private Use
$300
------- |