United State* Science Advisory Board EPA-SAB-EPEC-94-012
Environmental 1400F May 1994
Protection Agency Washington, DC
&EPA AN SAB REPORT: REVIEW
OF THE MIDWEST
AGRICHEMICAL
SURFACE/SUBSURFACE
TRANSPORT AND
EFFECTS RESEARCH
(MASTER) PROGRAM
PREPARED BY THE
ECOLOGICAL PROCESSES AND
EFFECTS COMMITTEE
f
ENVIRONMENTAL
V PROTECTION
f AGENCY
DALLAS, TEXAS
LIBRARY
-------
Y.lA.v
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
EPA-SAB-EPEC-94-012
Mav 4 1994 OFFICE OF THE ADMINISTRATOR
J ' SCIENCE ADVISORY BOARD
Honorable Carol M. Browner
Administrator
U.S. Environmental Protection Agency
401 M Street, S.W.
Washington, DC 20460
Subject: Review of Midwest Agrichemical Surface/Subsurface Transport
and Effects Research (MASTER) Program
Dear Ms. Browner:
On October 28-29, 1993, the Ecological Processes and Effects Committee
met to review the Midwest Agrichemical Surface/Subsurface Transport and Effects
Research (MASTER) Program. The MASTER Program is an inter-agency effort,
involving the EPA, U.S. Department of Agriculture, and the U.S. Geological
Survey, designed to assess the impacts of agricultural practices on the watershed
scale. EPA's participation, funded to-date under the Office of Research and
Development's (ORD) Nonpoint Source Issue Plan at approximately $2 million in
FY92 and approximately $1.5 million in FY93 and FY94, has focused on the
ecological effects of agricultural management practices. The research program is
being conducted in the Walnut Creek Watershed in Iowa. The Charge to the
Committee from ORD was to evaluate the MASTER Program with respect to the
following issues: a) overall program goals and objectives; b) assessment approach;
c) interdisciplinary integration; d) suitability for evaluating and reducing ecological
effects of agrichemicals and farm practices; e) balance of research between
assessment in the pilot watershed vs. extension of the assessment techniques to
the regional scale; and f) other watershed studies which should be incorporated
into the program.
The Committee strongly supports EPA's involvement in research to assess
the impacts of nonpoint source (NFS) pollution in agroecosystems and in seeking
ways to attain sustainability and ecological quality in agriculture. The MASTER
Program offers a unique opportunity to develop and test the effectiveness of
alternative ecosystem design and management scenarios, and thus may yield an
excellent model for management at the watershed level. We agree that EPA's
unique contribution to MASTER should be in the area of ecological effects; of the
federal collaborators, only EPA is likely to provide this focus. The MASTER
Program provides the opportunity to consider not only the ecological effects of
toxic chemicals, but to include a broader consideration of stressors. For example,
habitat alterations associated with various agricultural management practices
Rtcyctodfftocyclabte
PrfnMd OB p«p«r m* contains
-------
(e.g., loss of stream-side shading vegetation and increased turbidity from
sedimentation) may have ecological consequences equal to or greater than those
from toxic chemicals. The scale-up from watershed to regional scales will be a
major challenge and will require validation of the Walnut Creek Watershed models
at selected other sites.
Among the potentially significant advantages of the MASTER approach is
the use of an in-depth case study to test the effect of changes in management
practices and then to select practices which provide optimal agricultural and
ecological benefits. We stress, however, that the MASTER Program's objective of
reducing the ecological effects of agrichemicals and farm practices will not be
realized without a long-term commitment to the program by the Agency. After
two years of a planned five-year endeavor, the MASTER Program has invested
considerable effort in developing a study baseline for the Walnut Creek Watershed
and in developing models to predict how proposed management changes would
affect ecological indicators. The potential of the case study approach will not be
realized until the predicted effects of management changes can be compared to
results from field tests.
As described in the Agency's Framework for Ecological Risk Assessment, the
first step in ecological risk (ecorisk) assessment is problem formulation, which
includes the identification and selection of appropriate endpoints by which the
health of the ecosystem can be evaluated. The analysis phase of ecorisk
assessment then requires characterization of the changes in those selected
ecological endpoints in response to the stress regime. Although the MASTER
Research Plan describes the program as following the ecorisk assessment
paradigm, both the problem formulation and ecological effects analyses for selected
endpoints are not sufficiently developed. In our report, we have made a number
of recommendations for strengthening the MASTER Program, including the
addition of an ecological modeler to the MASTER team, development of an
ecological effects conceptual model, and development of an integrated assessment
framework to guide interdisciplinary interactions, provide a basis for selecting
priority research activities, and relate the projects being leveraged with other
agencies to the goals of MASTER.
Overall, the MASTER Program, if carried out with the guidance of an
integrated assessment framework and adequate agency funding and time
commitment, is an appropriate way to assess the ecosystem effects of various
agricultural management practices. We strongly urge EPA to maintain its
commitment and financial support for the program; we believe that the program
offers an excellent opportunity to improve our capability to assess the effectiveness
of agricultural "best management practices" and to manage ecological resources at
the watershed scale.
-------
We appreciate the opportunity to review this important program, and we hope that our
comments will be helpful to you.
Sincerely,
Dr. Genevieve M. Matanoski, Chair
Executive Committee
Dr. Kenneth L. Dickson, Chair Dr. Alan W. Maki, Chair
Ecological Processes and MASTER Review Subcommittee
Effects Committee
-------
U.S. Environmental Protection Agency
NOTICE
This report has been written as part of the activities of the Science
Advisory Board, a public advisory group providing extramural scientific
information and advice to the Administrator and other officials of the
Environmental Protection Agency. The Board is structured to provide balanced,
expert assessment of scientific matters related to problems facing the Agency.
This report has not been reviewed for approval by the Agency and, hence, the
contents of this report do not necessarily represent the views and policies of the
Environmental Protection Agency, nor of other agencies in the Executive Branch
of the Federal government, nor does mention of trade names or commercial
products constitute a recommendation for use.
-------
ABSTRACT
The Ecological Processes and Effects Committee met on October 28-29,
1993, to review the Midwest Agrichemical Surface/Subsurface Transport and
Effects Research (MASTER) Program. The MASTER Program is an inter-agency
effort between EPA, the U.S. Department of Agriculture, and the U.S. Geological
Survey, designed to assess the impacts of agricultural practices on the watershed
scale. EPA's participation in MASTER has focused on the ecological effects of
agricultural best management practices (BMPs). The Committee strongly supports
EPA's involvement in research such as MASTER to assess the impacts of nonpoint
source (NFS) pollution in agroecosystems and in seeking ways to attain
sustainability and ecological quality in agriculture. The MASTER Program
provides the opportunity to consider not only the ecological effects of toxic
chemicals, but to include a broader consideration of stressors such as habitat
alterations associated with various agricultural BMPs which may have ecological
consequences equal to or greater than those from agrichemicals. The Committee
urges the Agency to continue support for the MASTEP. Program beyond the
development of models and baseline data in the pilot watershed so that the
predicted effects of management changes in the watershed can be compared to
results from field tests.
KEYWORDS: agroecosystems, best management practices, nonpoint source
pollution, watershed assessment
11
-------
US ENVIRONMENTAL PROTECTION AGENCY
SCIENCE ADVISORY BOARD
ECOLOGICAL PROCESSES AND EFFECTS COMMITTEE
October 28-29, 1993
ROSTER
CHAIR
Dr. Kenneth L. Dickson, Institute of Applied Sciences, University of
North Texas, Denton, Texas
MEMBERS/CONSULTANTS
Dr. William E. Cooper, Institute Environmental Toxicology, Michigan State
University, East Lansing, Michigan
Dr. Virginia Dale, Environmental Sciences Division, Oak Ridge National
Laboratory, Oak Ridge, Tennessee
Dr. Rolf Hartung, School of Public Health, University of Michigan, Ann
Arbor, Michigan
Dr. Mark A. Harwell, Rosenstiel School of Marine and Atmospheric Science,
University of Miami, Miami, Florida
Dr. Robert J. Huggett, Virginia Institute of Marine Science, School of
Marine Sciences, College of William and Mary, Gloucester Point, Virginia
Dr. Alan W. Maki, Exxon Company, USA, Houston, Texas
Dr. Anne McElroy, SUNY at Stony Brook, Stony Brook, New York
Dr. Frederic K. Pfaender, Carolina Federation for Environmental Studies,
University of North Carolina, Chapel Hill, North Carolina
Dr. William H. Smith, School of Forestry and Environmental Studies, Yale
University, New Haven, Connecticut
Dr. Terry F. Young, Environmental Defense Fund, Oakland, California
111
-------
SCIENCE ADVISORY BOARD STAFF
Ms. Stephanie Sanzone, Designated Federal Officer, US EPA, Science
Advisory Board (1400F), 401 M Street, SW, Washington, DC 20460
Mrs. Marcia K. Jolly (Marcy), Staff Secretary, US EPA, Science Advisory
Board (1400F), 401 M Street, SW, Washington, DC 20460
IV
-------
TABLE OF CONTENTS
1. EXECUTIVE SUMMARY 1
2. INTRODUCTION 5
2.1 Background 5
2.2 Charge to the Committee 5
3. GENERAL COMMENTS 7
3.1 The Importance of MASTER 7
3.2 Ecological Risk Assessment in MASTER 9
3.3 Integrated Assessment Framework 9
3.4 Hypothesis Testing and Experimental Design 11
3.5 Regionalization and Extrapolation 11
3.6 Coordination With Other Watershed Studies 12
3.7 Variability and Uncertainty 13
4. MASTER COMPONENT PROCESS RESEARCH 14
4.1 Ground Water and Surface Water Quantity and Quality 14
4.2 Soil Quality and Subsurface Ecology 14
4.3 Terrestrial Biota and Habitat Structure 15
4.4 Aquatic Habitats and Biotic Communities 16
4.5 Decision-Support Tools 16
5. SUMMARY OF FINDINGS 17
6. REFERENCES CITED 19
-------
1. EXECUTIVE SUMMARY
The Ecological Processes and Effects Committee has reviewed the goals and
progress-to-date of the Midwest Agrichemical Surface/Subsurface Transport and
Effects Research (MASTER) Program. The MASTER Program is a cooperative
effort between EPA, the U.S. Department of Agriculture (USDA) and the U.S.
Geological Survey (USGS) to assess how agricultural practices impact overall
environmental quality, including ground and surface water quality and ecological
processes and populations in Midwestern farmlands. The research program is
being conducted in the Walnut Creek Watershed in Iowa. EPA's participation in
MASTER has been funded to-date under the Office of Research and Development's
(ORD) Nonpoint Source Issue Plan at approximately $2 million in FY92 and
approximately $1.5 million in FY93 and FY94. The overall goal of MASTER is to
provide the scientific and ecological bases for the development and implementation
of agricultural best management practices (BMPs) that both promote economically
viable agroecosystems and prevent degradation of the nation's water quality and
ecological resources.
The Committee strongly supports EPA's involvement in research to assess
the impacts of nonpoint source (NPS) pollution in agroecosystems and in seeking
ways to attain sustainability and ecological quality in agriculture. The MASTER
Program represents a unique inter-agency effort to scientifically develop and test
the effectiveness of alternative ecosystem design and management scenarios, and
thus offers an excellent model for management at the watershed level. In
addition, it was evident to the Committee that the interdisciplinary team is well-
qualified, enthusiastic, and committed to the program. The team has clearly had
the benefit of strong leadership and has been productive over the two years that
EPA has been involved in MASTER.
We agree that EPA's unique contribution to MASTER should be in the area
of ecological effects. Developing the ecological effects capability at the
watershed/landscape level is an important, innovative purpose for MASTER and
among the federal collaborators, only EPA is likely to provide this focus. The
MASTER Program offers the opportunity to advance beyond the toxic chemical
focus to a broader consideration of stressors. For example, habitat alterations
associated with various agricultural management practices (e.g., loss of stream-side
shading vegetation and increased turbidity from sedimentation) may have
ecological consequences equal to or greater than those from toxic chemical stresses.
Among the potentially significant advantages of the MASTER approach is
the use of an in-depth case study to test the effect of changes in management
-------
practices and then to select practices which provide optimal agricultural and
ecological benefits. We stress, however, that the MASTER Program's objective of
reducing the ecological effects of agrichemicals and farm practices will not be
realized without a long-term commitment to the program by the Agency. After
two years of a planned five-year effort, the MASTER Program has invested
considerable effort in developing the study baseline for the Walnut Creek
Watershed and in developing models to predict how proposed management changes
would affect the ecological indicators. The primary benefit of the case study
approach (the ability to predict ecological consequences of changes in management
practices in the watershed) will not be realized until the predicted effects of
management changes can be compared to results from field tests.
Our recommendations for strengthening the MASTER Program are
summarized as follows:
a) In accordance with the Agency's Framework for Ecological Risk
Assessment, as part of the problem formulation stage, we recommend
that the MASTER research team give immediate attention to a
strategic planning exercise. The output of this exercise should be a
plan for the integrated assessment which lays out the interactions
between the various research components of the MASTER Program.
An integrated assessment framework can serve as the "road map" for
interdisciplinary interactions, provide a basis for selecting priority
research activities, and relate the numerous projects being leveraged
with other agencies to the goals of MASTER. In addition, the
assessment strategic plan should emphasize the role that MASTER
can play as an intensive monitoring location for agroecosystems in the
Environmental Monitoring and Assessment Program (EMAP); clearly,
the MASTER Program can make a very effective contribution to
EMAP and vice versa. We urge the Agency to fully explore the links
between these two programs.
b) The MASTER Program's integrated assessment framework must
clearly demonstrate the ability to couple ground water and surface
water parameters to watershed/ecosystem characteristics. An
important consideration is the degree to which the models can
separate effects due to changing management practices from natural
effects due to weather variations, and the degree to which observed
changes can be related to practices at the farm level, particularly for
management practices which do not involve changes in land use or
cropping patterns. We are concerned over the lack of discussion of
variability, both in terms of measurement error and heterogeneity
-------
over space and over time. Given the complex nature of the MASTER
assessment structure, which includes a mosaic of models, GIS
descriptive parameters, exogenous influences, and inventories of biota,
it is critical that the MASTER Program develop a well articulated
statistical design to ensure the appropriate level of effort, location of
sampling sites, and duration of experiments.
c) A stated objective of MASTER is to provide regional assessment by
"scaling-up" from field or farm scales to watershed, ecoregion, and
regional scales. From the information presented by the MASTER
team, however, it is clear that more effort should be put into the
development of "regionalization assessment techniques." We urge the
MASTER Program to utilize data from other watershed research sites
for independent validation of the models developed for the Walnut
Creek Watershed. The scale-up from watershed to regional scales will
be a major challenge and model validation at selected other sites is a
necessary step in this process.
d) We commend the MASTER Program for including the soil
environment as a legitimate ecosystem worthy of independent study.
We encourage the MASTER Program to incorporate soil productivity
and soil quality degradation endpoints in the assessment. As
measures of soil quality are developed, it must be recognized that
naturally occurring soils have a broad range of chemical and physical
characteristics and vary greatly in their capacity to support micro-
and macro-flora and fauna. Also, the assessment of soil
quality/degradation should include not only the assessment of
productive capacity as used in agriculture, but broader measures of
ecological health as well (e.g., diversity of soil fauna and flora,
processing of carbon, and nitrogen fixation).
e) While the study recognizes the significant role of habitat
characteristics in determining the diversity and abundance of
terrestrial wildlife, the researchers need to clarify the tune delay
between instituting certain management practices and observing an
effect on bird (and/or mammal) populations. Furthermore, it appears
that increased abundance and diversity of terrestrial wildlife was
regarded as a desirable condition or trend. However, agricultural
ecosystems are highly managed systems and the selection of endpoints
for such systems must be based on a recognition that land
management processes are now inherent components of the
ecosystem.
-------
f) And finally, in assessing aquatic habitats and biotic communities, we
urge the MASTER Program to develop a balanced assessment design
which evaluates the role(s) of both chemical and non-chemical
stressors, e.g., via an ecological effects model that integrates the
impacts of the various exogenous forcing functions.
-------
2. INTRODUCTION
2.1 Background
The Midwest Agrichemical Surface/Subsurface Transport and Effects
Research (MASTER) Program is a cooperative effort between EPA, the U.S.
Department of Agriculture (USDA) and the U.S. Geological Survey (USGS) to
assess how agricultural practices impact overall environmental quality, including
ground and surface water quality, ecological processes, and biological communities
in Midwestern farmlands. The research program is being coaducted in the Walnu
Creek Watershed in Iowa. The overall goal of MASTER is to provide the scientif
and ecological bases for the development and implementation of agricultural best
management practices (BMPs) that both promote economically viable
agroecosystems and prevent degradation of the nation's water quality and
ecological resources. EPA's participation in MASTER has been funded to-date
under the Office of Research and Development's (ORD) Nonpoint Source Issue
Plan at approximately $2 million in FY92 and approximately $1.5 million in FY
and FY94. EPA's focus in the program is on developing the methods to design
and evaluate BMPs at the watershed level, emphasizing ecological benefits, and
assess the ecological effects of agricultural practices at the regional scale (e.g.,
Western Corn Belt Plains Ecoregion).
2.2 Charge to the Committee
The Ecological Processes and Effects Committee met in Washington, E
October 28-29, 1993, to review the overall goals and progress of the MASTE1
Program. In particular, the MASTER Research Team requested that the
Committee consider the following questions:
a) Are the goals and objectives of the program clear and appropriafc
the environmental problem being addressed?
b) Is the assessment approach appropriate for defining long-term
research in the Walnut Creek Watershed?
c) Is progress toward interdisciplinary integration apparent?
d) Are the ongoing and proposed projects appropriate for evaluating i
reducing the ecological effects of agrichemicals and farm productio
practices?
-------
e) Is the research balanced between field testing, monitoring, and
development of assessment methodologies within the Walnut Creek
Watershed vs. the development and application of "regionalization
assessment techniques" using Walnut Creek data and results?
f) Are there other watershed sites and cooperators that should be
incorporated into the MASTER Program?
-------
3. GENERAL COMMENTS
3.1 The Importance of MASTER
The Committee strongly supports EPA's involvement in research to assess
the impacts of nonpoint source (NFS) pollution in agroecosystems and in seeking
ways to attain sustainability and ecological quality in agriculture. Nonpoint
pollution and habitat alterations associated with current agricultural activities are
among the greatest stressors to the integrity of ecological systems. In order to
address the difficult environmental problems and issues related to nonpoint source
pollution, the Agency will need increased capabilities to assess the cumulative
effects of a variety of diffuse sources within a watershed. The MASTER Program
represents a unique inter-agency effort to scientifically develop and test the
effectiveness of alternative ecosystem design and management scenarios, and thus
offers an excellent model for management at the watershed level. It appears that
an excellent working relationship has been established between EPA, USDA and
USGS scientists, as well as among the EPA researchers on the team (from the
Environmental Research Laboratories in Athens, Duluth, Corvallis and Ada, the
Environmental Monitoring Systems Laboratory in Las Vegas, and a number of
university investigators and contractors). By advancing environmental science at a
watershed/landscape scale, MASTER will provide a scientific basis for evaluating
policy and management options for addressing NPS pollution.
The relevance of watershed-based research like MASTER is underscored by
legislative proposals currently before the U.S. Congress. The reauthorization of
the Clean Water Act is correctly emphasizing the need to implement proactive
watershed design and management practices. Pollution prevention, using BMPs or
economic incentives, is considered the most effective method for managing
agricultural pollution. This requires knowledge about the effectiveness of various
BMPs on both watershed and landscape scales.
In addition to the importance of the MASTER Program's goals and
objectives, it was evident to the Committee that the EPA interdisciplinary team is
well-qualified, enthusiastic, and committed to the program. The team has clearly
had the benefit of strong leadership and has been productive over the two years
that EPA has been involved in MASTER.
The USDA has a great deal of data, expertise, and methodologies to
characterize the effects of alternative agricultural management practices on
chemical levels in field runoff, surface and ground water. The central reason for
its collaboration with EPA on this project is to attain an ability to link these
-------
chemical responses with ecological effects, a capability currently lacking in the
USDA program. We agree that the central focus for MASTER should be on the
ecological effects linkages, only EPA is likely to provide this capability, the highly
leveraged collaboration with USDA allows that focus for EPA, and developing the
ecological effects capability at the watershed/landscape level is an important,
innovative purpose for MASTER. Applied assessments such as MASTER should
also serve as an important "reality check" for the assessment of agrichemicals
required under the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA),
which currently relies heavily on interpretation of extensive laboratory studies
with only limited field studies.
Additionally, the MASTER Program offers the opportunity to advance
beyond the toxic chemical focus to a broader consideration of natural and
anthropogenic stresses. In particular, habitat alterations associated with various
agricultural management practices (e.g., loss of stream-side shading vegetation and
increased turbidity from sedimentation) may have ecological consequences equal to
or greater than those from toxic chemical stresses. While there is some attention
at present in MASTER to non-chemical stresses, these appear to be treated as
confounding factors which interfere with the assessment of chemical
exposure/response relationships in the real world. An alternate perspective,
consistent with the goals of MASTER, would be that each stressor should be
examined for its relationship to ecological effects manifested in changes hi
endpoints. That perspective, and the experimental design considerations
commensurate with it, should be incorporated into the MASTER Program.
Among the potentially significant advantages of the MASTER approach is
the use of an in-depth case study to test the effect of changes in management
practices and then to select practices which provide optimal agricultural and
ecological benefits. We stress, however, that the MASTER Program's objective of
reducing the ecological effects of agrichemicals and farm practices will not be
realized without a long-term commitment to the program by the Agency.
Ecosystem-level assessments such as MASTER, covering an entire watershed,
require data inputs from several years to test cause-effect hypotheses accurately in
the face of natural physical, meteorological and year-to-year variability. Agency
budgetary and staffing commitments need to reflect the longer time frame required
to complete this program successfully.
After two years of a planned five-year effort, the MASTER Program has
invested considerable effort in developing a study baseline for the Walnut Creek
Watershed and in developing models to predict how proposed management changes
would affect the ecological indicators. The primary benefit of the case study
approach will not be realized, however, until the predicted effects of management
8
-------
changes can be compared to results from field tests. The confirmation step is
important not only for scientific, but also for policy reasons, and we urge the
Agency to maintain its commitment to MASTER through this important stage of
the program.
3.2 Ecological Risk Assessment in MASTER
The goals of the MASTER Program require the explicit linkage of changes
in the physical and chemical environment, whether natural or anthropogenic, to
changes in the health of the ecosystem. This should be done in the context of the
Agency's Framework for Ecological Risk Assessment. The first step in the
ecological risk (ecorisk) assessment process is problem formulation, which includes
the identification and selection of appropriate ecological endpoints by which the
health of the ecosystem can be evaluated. The analysis phase of ecorisk
assessment then requires characterization of the changes in those selected
ecological endpoints in response to the stress regime.
Although the MASTER Research Plan describes the program as following
the ecorisk assessment paradigm, both the problem formulation and ecological
effects analyses for selected endpoints are not sufficiently developed. This can best
be remedied by the development of an integrated assessment framework, as
discussed in the following section of our report. More thought should be given to
how ecological effects will be characterized, what models and other analytical
methodologies will be used, and how these effects assessments will fit into the
overall decision-support system. The ecological effects approach should be detailed
now, not delayed until data are collected or analyzed. To this end, we recommend
that an ecological modeler be added to the MASTER team and that an ecological
effects conceptual model be developed with explicit linkages to the physical and
chemical modeling activities. This will help ensure the intimate co-development of
the ecological modeling tasks with the field studies.
3.3 Integrated Assessment Framework
The MASTER Program has a clear goal to assess the impact of agricultural
management practices on the sustainability of agroecosystems and ecological
resources. However, the means of attaining this goal are not clear. Therefore, we
recommend that the MASTER research team give immediate attention to a
strategic planning exercise. The output of this exercise should be a plan for the
integrated assessment which lays out the interactions between the various research
components of the MASTER Program.
-------
The assessment framework should start with potential management
scenarios and end with potential effects on the system (i.e., ecological endpoints).
Of course, the choice of ecological endpoints depends on the variability associated
with those endpoints (i.e., highly variable endpoints will be difficult to evaluate).
Therefore, going through a "strawman" assessment and considering a rough
estimate of uncertainties will contribute to the selection of appropriate endpoints.
In addition, the development of an assessment framework should help to clarify
potential management practices and their cascading effects through the system,
potential feedbacks, uncertainties in the system, spatial and temporal
discongruities between research components (e.g., how data and model output from
one scale will relate to information and questions at another scale), and how the
MASTER Program relates to other watershed studies. The decision-support
system should provide the framework for relating the research components to the
ecological endpoints of interest.
Another benefit of an integrated assessment framework is that it can serve
as the "road map" for interdisciplinary interactions, provide a basis for selecting
priority research activities, and relate the numerous projects being leveraged with
other agencies to the goals of MASTER. While we commend the MASTER
program for creatively leveraging their limited funds with other research projects,
it is crucial to clarify how the various projects relate to the goal of assessing the
ecological impacts of farm management practices. In addition, the MASTER
Program's assessment strategic plan should emphasize the role that MASTER can
play as a Tier 3 (or 4?) intensive monitoring location for agroecosystems in the
Environmental Monitoring and Assessment Program (EMAP) and should take into
account the sampling design utilized by EMAP and other larger assessment
programs to allow effective program integration.
The strategic plan should also clarify how the effects of incorporating
different management practices can be traced through the models to each of the
assessment endpoints. At this stage of the MASTER Program, it appears that
selected elements of the project are relatively well-developed, such as the surface
water and ground water models, and the baseline characterization of land uses and
management practices. It is not clear, however, to what degree the effects of
changing the various management practices can be characterized, particularly at
the farm level and for those management practices which do not involve changes
in land use or cropping patterns. In short, there are a number of apparent
conceptual "disconnects" between the front and back ends of the system which
could be remedied by the development of an overall strategic plan for the
integrated assessment.
10
-------
3.4 Hypothesis Testing and Experimental Design
The Walnut Creek Watershed represents a discrete, well-characterized
ecosystem that provides a unique opportunity to test hypotheses about ecosystem
design and management. The hypotheses associated with model validation involve
the toxic effects of agrichemicals and the ecological effects of habitat alterations.
The hypotheses associated with alternative management scenarios involve BMPs
and alterations in landscape design.
The overall assessment structure is characterized by a mosaic of models
(surface water and ground water), GIS descriptive parameters (surface and
subsurface horizons, cropping patterns, ecological landscapes), exogenous influences
(e.g., weather, pesticide applications) and inventories of biota (IBI and ICI). The
system in the aggregate is too complex and interactive to test for validation and/or
robustness in the total response. Therefore, modular components must be
evaluated either by using opportunistic events (e.g., floods, drought) or direct
perturbation experiments. Once the causal relationships have been demonstrated,
then the responses to alternative BMPs and/or alternative landscape designs can
be articulated. These predictions constitute hypotheses that are verifiable by
direct field observations.
Due to the complexity of the system, however, the potential for confounding
variables to contaminate the diagnostic ability of the experimental design is high.
The effects of pesticides, habitat alterations, extreme weather conditions and
natural variability are all co-variants in space and time. Thus, it is critical that
the MASTER Program develop a well articulated statistical design to ensure the
appropriate level of effort, location of sampling sites, and duration of experiments.
The sequence of investigations should be articulated within the framework of the
integrated assessment plan.
We encourage the MASTER researchers to make greater use of existing data
as a means to develop hypotheses, extend their results to other regions or
management practices, extend their results to other species, and extrapolate their
results in both space and tune.
3.5 Regionalization and Extrapolation
A stated objective of MASTER is to provide regional assessment by "scaling-
up" from field or farm scales to watershed, ecoregion, and regional scales. From
the information presented by the MASTER team, however, it is clear that more
effort should be put into the development of assessment methodologies with
particular focus on the "regionalization assessment techniques." Although the
11
-------
existing research program is using the most advanced GIS techniques to "scale-up"
patterns, it is not clear how processes will be translated between scales. The
scientific community is grappling with the problem of how to scale processes (e.g.,
King, 1991), and the MASTER Program can make a significant contribution in
this area. We encourage the MASTER researchers to work with EMAP scientists
in addressing the scaling issues.
Distinct from the issue of regionalizing results is that of extrapolating the
results and methodologies developed in the Walnut Creek Watershed to other
similar-scale watersheds. As mentioned previously, one suggestion is to couple the
MASTER Program with EMAP to examine heterogeneity at the watershed level
across the region (e.g., using the Thematic Mapper (TM) data base, being acquired
by EMAP, to characterize the Walnut Creek and other watersheds in the region).
Although the large-scale EMAP hexagons dwarf the Walnut Creek Watershed, this
presents the opportunity to link to EMAP's finer-scale, more process-oriented goals
for intensive examination of specific systems. Further, the EMAP experimental
design could well be applied to the watershed level in Walnut Creek by scaling
down the EMAP hexagon structure to the scale appropriate for MASTER studies.
Clearly, the MASTER Program can make a very effective contribution to EMAP
and vice versa. We urge the Agency to fully explore the links between these two
programs.
3.6 Coordination With Other Watershed Studies
While the MASTER Program represents a successful multi-agency, multi-
disciplinary research program, there are several other sites/research centers across
the country that could prove beneficial to the MASTER program if effective
liaisons were established. For example, research at Clemson University's Edith
Angel Research Center at Chariton, Iowa, is closely analogous to MASTER.
Largely funded by the agrichemical industry, the experimental sites are located
just south of the Walnut Creek Watershed. Another possible source of data and
coordination is Michigan State University's Kellogg Agricultural Experiment
Program, a Long-Term Ecological Research (LTER) site funded in part by the
National Science Foundation.
We urge the MASTER Program to utilize data from sites such as these for
independent validation of the models developed for the Walnut Creek Watershed.
Validation at other MSEA sites may not be feasible since the level of information
available is considerably less than the data collected in the Walnut Creek
Watershed. The scale-up from watershed to regional scales will be a major
challenge and model validation at selected other sites is a necessary step in this
process. A more detailed discussion of the importance of confirming models with
12
-------
field data can be found in the SAB report, Resolution on the Use of Mathematical
Models by EPA for Regulatory Assessment and Decision-Making (EPA-SAB-EEC-
89-012).
3.7 Variability and Uncertainty
The Committee was concerned with the inattention to uncertainties. One
example was the lack of error terms on hazard assessment data. We presume
such statistical aspects are being addressed, but the absence of error terms in
presentations of results and conclusions diminished confidence. More important,
however, is a lack of discussion of variability in the broader sense, not just in
terms of measurement error, but especially in terms of heterogeneity over space
and over time. As mentioned previously, the problem formulation step in the
ecorisk assessment framework includes identification of the full range of potential
uncertainties. This should provide the basis for the experimental design and for
setting priorities among specific research hypotheses and activities. In addition,
the limitations for extrapolation to other watersheds and to regional scales should
be identified.
13
-------
4. MASTER COMPONENT PROCESS RESEARCH
4.1 Ground Water and Surface Water Quantity and Quality
The research underway to model the subsurface geology and hydrology in
the watershed is innovative and interesting. It is not clear, however, that the
ground water and surface water models are linked to ecological endpoints of
interest. This linkage is critical since the ultimate goal of the MASTER Program
is the ability to quantify the impacts of BMPs on ecosystem values. The MASTER
Program's integrated assessment framework must clearly demonstrate the ability
to couple ground water parameters to watershed/ecosystem characteristics. Once
ecological endpoints of interest have been selected, chemical and habitat stressor
effects can be evaluated against these endpoints. For example, the ground water
model will allow prediction of pesticide exposure in surface water systems of the
watershed. In the case of herbicides, exposures may alter aquatic plant
communities, causing insect communities associated with these plants to be
adversely impacted, and this in turn (due to foodweb linkages) may have an
adverse impact on bird communities. In this example, bird, small mammal and
arthropod species richness and abundance represent possible ecological endpoints
of concern.
An important consideration is the degree to which the models can separate
effects due to changing management practices from natural effects due to weather
variations. Similarly, we recognize that, while the model may distinguish the
effects of changes in management practices at a field scale, testing of the model
predictions is limited by the larger scale of the monitoring networks. It would be
useful to determine up front how the limits imposed by the scale of the monitoring
system, as well as variables such as the timing and amount of precipitation, will
effect MASTER'S ability to test' model predictions. This relates to our
concerns/comments about hypothesis testing and experimental design in section 3.4
of this report.
4.2 Soil Quality and Subsurface Ecology
We commend the MASTER Program for adding a soil microbiological
component to the conceptual framework for this agroecosystem; we agree that the
soil environment is a legitimate ecosystem worthy of independent study.
Treatment of soil as an ecosystem recognizes the living and dynamic nature of
soils and the need to assess the effects of BMPs on the below-ground environment.
This is a relatively new idea for EPA and one that could add significantly to this
and future projects.
14
-------
We encourage the MASTER Program to incorporate soil productivity and
soil quality degradation endpoints in the assessment. Soil ecosystem endpoints of
direct relevance are total microbial biomass (e.g., bacterial, fungal and nematode
biomass) and rates of specific soil processes (e.g., soil respiration, nitrification, and
denitrification). Sites of important action (biologically relevant) in the soil
ecosystem are focused in the rhizosphere and in the uppermost soil horizons.
The stated goal of this effort is to assess the "assimilative capacity" of the
soil ecosystem for agricultural chemicals. We strongly recommend that "fate and
effects" terminology be used rather than the term "assimilative capacity."
"Assimilative capacity" could also be replaced by the elements of which it is
composed, including adsorption and biotic and abiotic degradation.
As measures of soil quality are developed, it must be recognized that
naturally occurring soils have a broad range of chemical and physical
characteristics and vary greatly in their capacity to support micro- and macro-flora
and fauna. Clearly, one cannot apply the same standards of productivity to silica
sand as to Iowa loess soils. In addition, soil characteristics are not static-they
respond to climatic changes, are modified by agricultural practices designed to
influence the productivity of crops and the diversity of soil organisms.
Given this broad range of status for soils, the MASTER Program's
conceptual framework for the assessment of soil quality/degradation should include
not only the assessment of productive capacity as used in agriculture, but broader
measures of ecological health as well.
4.3 Terrestrial Biota and Habitat Structure
The MASTER Program's assessment of the impacts of agricultural practices
on terrestrial wildlife is at an early stage. Potential effects on species diversity
and numbers of individuals are being explored for birds, but not for mammals.
The assessment is focusing on measures of abundance for species considered the
most likely to be affected by changes in agricultural practices based upon existing
information about the life histories and habitat preferences of bird species in
central Iowa. Many of the land management techniques to increase avian and
mammalian diversity and abundance are well understood for game animals and
have been studied extensively by wildlife managers. While the study recognizes
the significant role of habitat characteristics as a determinant of bird diversity and
abundance, it is also important to clarify the time delay between instituting certain
management practices and observing an effect on bird (and/or mammal)
populations. For instance, increasing habitat structure (e.g., creation of farm
15
-------
woodlots, hedgerows or wetlands) will increase species diversity, but the impact
will be difficult to assess in the 5-year timeframe of the MASTER Program.
Furthermore, it appears that increased abundance and diversity of wildlife
was regarded as a desirable condition or trend. In the absence of other stressors,
bird and mammal populations respond well to habitat management. However,
when dealing with agroecosystems, several questions may be posed: how much
habitat management is desirable? how abundant and diverse should terrestrial
wildlife be? Obviously, agricultural ecosystems are highly managed systems and
the selection of endpoints for such systems must be based on a recognition that
land management processes are now inherent components of the ecosystem.
4.4 Aquatic Habitats and Biotic Communities
The aquatic ecology component of the MASTER Program takes a
multimetric approach to establishing the condition of Walnut Creek. By including
assessments of habitat, ambient toxicity, and instream community structure (fish
and macro-invertebrates), it may be possible to identify stressors affecting aquatic
life. However, it was not explicitly clear that the aquatic assessment experiments
included diagnostic indicators of non-chemical stressors. Rather, the assessment
design appears to be weighted toward assessing xenobiotic chemical impacts (i.e.,
toxicological impacts of residual agrichemicals). Non-chemical stressors like
riparian vegetation, siltation, and continuity of stream flow are used as "blocking"
(random block design covariants) criteria to enhance the observability of
toxicological effects. Habitat characteristics and alterations are not explicitly being
analyzed as important forcing functions or "drivers". There are a number of
stream studies, however, that show that habitat alterations can produce far greater
effects than residual agrichemicals. We therefore urge the MASTER Program to
develop a balanced assessment design which evaluates the role(s) of both chemical
and non-chemical stressors (e.g., via an ecological effects model that integrates the
impacts of the various exogenous forcing functions).
4.5 Decision-Support Tools
The Committee is concerned that the decision-support component of
MASTER was presented as just a "black box." In part this may reflect the mixture
of MASTER personnel present at the review meeting in October, but supporting
written materials also indicate an insufficient attention to the objectives, use and
structure of the decision-support system. As a part of the overall assessment
framework development exercise discussed in section 3.3, there should be explicit
attention to specific aspects of the decision-support system.
16
-------
5. SUMMARY OF FINDINGS
Overall, the MASTER Program, if carried out with the guidance of an
integrated assessment framework and adequate agency funding and time
commitment, is an appropriate way to assess the ecosystem effects of various
agricultural management practices. We strongly urge EPA to maintain its
commitment and financial support for the MASTER Program; we believe that the
program offers an excellent opportunity to improve our capability to assess the
effectiveness of agricultural BMPs and to manage ecological resources at the
watershed scale.
With regard to the specific questions in the Charge to the Committee, our
responses are summarized as follows:
a) Are the goals and objectives of the program clear and appropriate for
the environmental problem being addressed?
Yes. Furthermore, the focus on agroecosystems and the assessment
approach (watershed to regional analysis) are important and as yet
under-represented areas of investigation for the Agency.
b) Is the assessment approach appropriate for defining the longer-term
research in the Walnut Creek Watershed?
In the absence of an integrated assessment framework, it is not
possible to determine whether or not the details of the proposed
research program are appropriate and defensible. The development of
such a strategic framework should be the top priority of the MASTER
team. The framework should clarify the linkage between BMPs,
hydrologic parameters, and ecosystem values (endpoints).
c) Is progress toward interdisciplinary integration apparent?
Yes. The multi-laboratory, interagency, and interdisciplinary nature
of the MASTER Program is a great strength of the program.
However, the effectiveness of these interactions in supporting overall
program goals will be compromised until an integrated assessment
framework is developed. Very effective cooperation with USDA has
allowed the program to be productive with a relatively small budget.
The obvious mutual respect between the primary investigators greatly
enhances the potential for MASTER results to have wide-ranging
influence on the management of agroecosystems in this country.
17
-------
d) Are the ongoing and proposed projects appropriate for evaluating and
reducing the ecological effects of agrichemicals and farm production
practices?
Yes, in large part the MASTER components are appropriate for
accomplishing the stated goals and are based on sound scientific
approaches. The most significant benefits of the program will not be
realized, however, unless the research components are successfully
integrated via an assessment framework and the program continues
long enough to allow predicted effects of management changes in the
pilot watershed to be compared to results from field tests.
e) Is the research balanced between field testing, monitoring and
development of assessment methodologies within the Walnut Creek
Watershed vs. the development and application of "regionalization
assessment techniques" using Walnut Creek data and results?
It is difficult to determine if MASTER is effectively balancing the
research between field testing, monitoring, and developing assessment
methodologies without seeing the strategic plan. However, it is clear
that more effort should be put into development of assessment
methodologies with particular focus on the "regionalization assessment
techniques".
The ability to connect measurements made of different scales is
critical to MASTER'S ability to meet its goals and objectives and the
eventual application of this approach to other watersheds and regions.
This is not a trivial undertaking. How this will actually be
accomplished should be laid out in the integrated assessment
framework. Effort should be made to incorporate complimentary
design elements with other regional programs such as EMAP.
f) Are there other watershed sites and cooperators that should be
incorporated into the MASTER Program?
Yes. Data from sites like the Kellogg Agricultural Experiment
Program in Michigan and the Edith Angel Research Center in Iowa
should be used to test and validate models developed for the Walnut
Creek Watershed.
18
-------
6. REFERENCES CITED
King, A.W. 1991. Translating Models Across Scales in the
Landscape. In Turner, M.G. and R.H. Gardiner (ed.), Quantitative Methods
in Landscape Ecology. New York. Springer-Verlag. pp. 479-518.
19
------- |