EDA u*s- Environmental Washington, DC
fc* " Protection Agency iPA-SAB-EPEC-91-001
Report of The
Ecological Monitoring
Subcommittee of The
Ecological Processes and
Effects Committee
Evaluation of The
Ecological Indicators
Report for EMAP
A SCIENCE ADVISORY BOARD REPORT November 1990
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, 0,C, 20460
QFFIGi OF
THE ADMINISTRATOR
November 6, 1990
EPA-SAB-EPEC-91-001
The Honorable William Reilly
Administrator
U.S. Environmental Protection Agency
401 M. Street, S.W.
Washington, D.c. 20460
Dear Mr. Reilly:
The Ecological Monitoring Subcommittee of the Science
Advisory Board (SAB) has completed its review of the interim
conceptual plan entitled "Ecological Indicator Report for the
Environmental Monitoring and Assessment Program (EMAF)". As
you are aware, the Science Advisory Board recommended that a
program be implemented within EPA to monitor the status, extent,
and geographic distribution of our ecological resources and
assess changes in the condition of those resources on a regional
and national scale. The selection of indicators is a critical
step in the program because indicators will define the condition
of ecosystems.
We are pleased that the Office of Research and Development
(ORD) is developing EMAP to address this need, and we look
forward to future reviews of its principal components and
strategies. We believe, in concert with ORD, that the SAB
reviews will complement the focused Peer Reviews for particular
ecosystems and the long-term assessment that is proposed for the
National Research Council.
The Subcommittee met on May 8-9, 1990, to review and
evaluate the Indicator report. ORD asked the Subcommittee to
address the following questions as part of their review:
a. Are the concepts of the indicators clearly articulated?
b. Are the data in the appendices on individual indicators
relevant and accurate?
c. Is the suite of indicators sufficient and appropriate
for each ecosystem? Will the results of such monitoring
Prixttd an Recycled Paper
-------
be relevant given the temporal and spatial scale of the
program?
d. Is the current classification of ecosystems appropriate?
The Subcommittee supports the EMAF concept and commends the
program for the careful and iterative approach that has been
taken to selecting indicators. While this promotes extensive
reviews it has also led to a proliferation of terminology and a
divergence in the concept of indicators that are applied among
different ecosystems. We recommend that the revised report
contain a glossary and an explicit rationale for the selection
of the indicators, and clarifications of the term habitat
indicators.
The use of data in the text and appendices is quite uneven
for different parts of the report. In addition to our ecosystem
specific comments, we recommend that the report include a matrix
of selection criteria for each ecosystem strategy. The
sufficiency of'the indicators and the stage of development and
testing varies significantly among ecosystems. We have provided
specific comments and encourage the program to accelerate the
schedule for research on risk characterization. ORD must clarify
the use of the terms "nominal" and "subnominal". This
incorrectly implies that the indicators will identify a cause.
The relevance -of the results should be evaluated after each
demonstration pilot? pilot monitoring studies and other
techniques for midcourse corrections are essential due to the
temporal scale of EMAP and its reliance on many Agencies.
The current ecosystem classification system is in some cases
artificial ' (e.g., agro-ecosystems that include fish farms and
wheat fields) and in some cases illogical (e.g., wetlands have
little in common that range from tundra to mangrove swamps). But
most importantly it may inhibit a focus on the integration of
regional environmental factors and the pooling of monitoring
resources that an assessment of ecological functions needs.
We recommend that EMAP consider a pilot to test the use of
biogeographic regions as assessment units as a possible
alternative to ecosystem classes.
Although we expect the Program Office will address all of
the issues presented in this report, we respectfully direct your
attention to these:
-the need to integrate and use historical data
-the implication that EMAP will identify causality
-the artificial nature of the current ecosystem
classification and the plans to evaluate the use of
ecoregions as assessment units
-------
-the rationale for selection of indicators
-the plans to evaluate results of pilot studies
-the inconsistent use of terms such as nominal and
subnominal.
The SAB appreciates the opportunity to conduct this
scientific review and looks forward to receiving your response to
the scientific advice transmitted herein.
Sincerely,
Dr. Raymond Loehr, Chairman Dr» Kenneth Dickson, Chairman
Executive Committee Ecological Processes and
Science Advisory Board Effects Committee
-------
ABSTRACT
This report presents the conclusions and recommendations of
the U.S. Environmental Protection Agency's science Advisory Board
following a review of the draft document "Ecological Indicator
Report for the Environmental Monitoring and Assessment Program"
(EMAP). EMAP, the Agency's program to monitor the status and
trends of regional and national ecological resources, will
evaluate and use indicators to measure and describe the overall
condition of ecosystems. Specific sets of indicators which
quantify response to anthropogenic stress (response indicators),
exposure to pollutants or causes of poor condition (exposure
indicators), and measure human activities which are suggestive of
environmental stress (stressor indicators) are being evaluated
for each class of ecosystem. This approach is evolving and the
Subcommittee considered that EMAP had made good progress through
its peer reviews of the overall approach and the interim plan for
indicators. The development of indicators was in different
stages for each of the ecosystems; however the use of a field
demonstration pilot is a good technique which could promote
interactions between the ecosystem scientists. The terminology
for defining "nominal", "subnominal", and habitat indicator
should be clarified and selection of indicators by all ecosystem
classes should be better coordinated and integrated with
ecological risk assessment principles. The EMAP team should rely
on historical data, where possible, to estimate sampling
requirements and assess trends and they should reassess their
ability to identify causal relationships through this regional
scale of monitoring. Other comments and suggestions for
clarification are provided in the report.
Key words.;. Indicators? Environroental Quality,* Monitoring;
Stress.
-------
U.S. ENVIRONMENTAL PROTECTION AGENCY
NOTICE
This report has been written as a 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 a 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 or
other agencies in Federal Government. Mention of trade names or
commercial products does not constitute a recommendation for use.
11
-------
U.S. ENVIRONMENTAL PROTECTION AGENCY
SCIENCE ADVISORY BOARD
ECOLOGICAL PROCESSES AND EFFECTS COMMITTEE
ECOLOGICAL MONITORING SUBCOMMITTEE
ROSTER
Dr. Kenneth Dickson
Director
University of North Texas
Institute of Applied Sciences
P.O. BOX 13078
Denton, Texas 76203
VICE-CHAIRMAN
Dr. Stanley Auerbach
Director
Environmental Sciences Division
Oak Ridge National Laboratory
Oak Ridge, Tennessee 37831
MEMBERS
Dr. Yoram Cohen
Associate Professor
Engineering Department
UCLA Rm. 5531
Boelter Hall
Los Angeles, California 90024
Daniel Goodman
Montana State University
Department of Biology
Louis Hall
Bozeman, Montana 59717
111
-------
Dr. Rolf Hartung
Professor
School of Public Health
3125 Fernwood Avenue
University of Michigan
Ann Arbor, Michigan 48108-1955
Dr. Allan Hirsch
Director, Washington Operations
Midwest Research Institute
Falls Church, Virginia 22041
Dr. Kenneth JenJcins
Director .
Molecular Ecology. Institute
California State University
Long Beach, California 90840
Dr. Richard Kimerle
Monsanto Corporation
800 H* Lindbergh Boulevard
St. Louis, Missouri 63167-5842
Dr. Richard F. Lee
Skidaway institute of Oceanography
P.O. Box 13687
Savannah, Georgia 314IS
Dr. John Neuhold
Department of wildlife Sciences
College of Natural Resources
' Utah State University
Logan, Utah 84322
Dr. William H. Smith
Professor of Forest Biology
School of Forestry
and Environmental studies
Yale University
370 Prospect Street
New Haven, Connecticut 06511
IV
-------
Dr. Leonard H. Weinstein
Boyce Thompson Institute Plant Res
at Cornell
Tower Road
Ithaca, New York 14853
S-CI.EH.CE-JIOVISORY STAFF
Dr. Edward S* Bender
Biologist & Executive Secretary
U.S. Environmental Protection Agency
Science Advisory Board
401 M Street, SW.
' Washington, D.C. 20460
Ms. Frances A. Dolby
Secretary to the Executive Secretary
-------
E PF CONTENTS
1.0 Executive Summary. ... ........... i
2,0 Introduction . . , , 2
2.1 Charge to the Subcommittee . . 3
3.0 Evaluation of the Indicator Report ........... 4
3.1 General Comments 4
3,1,1 The EMAP Concept ,,....,. 4
3,1.2 EMAP Strategy , . , ......... 5
3.1,3 Inter-agency Coordination 5
3.1.4 Peer Review. ...... 6
3.1.5 Discussion of the Charge ............ 5
3.1.6 Assessment Systems ......... 7
3-1.7 Habitat Indicators . 8
3.1.8 Sampling strategy ........... s
3.1.9 Historical Data. .....,.....,..,, 9
3.1.10 Model Ecosystem Approach. ........... 10
3.1.11 Coordination and Integration 10
3.1.12 Biogeographic Regions ...... n
3.1.13 Cost Estimates. ... ..... 12
3.2 Recommendations for Revising the Indicator Report. . 13
3.2.1 Cross-cutting Indicators ............ 13
3.2.2 Wetlands 14
3.2.3 Near-Coastal Waters 15
3.,2.4 Forests, .................... 17
3.2.5 Arid Lands 18
3.2.6 Agro-Ecosystems. ................ 13
3.3 Testing Indicators 19
3.3.1 Document Rationale .»»»..,,....... 19
3.3.2 Resource Types 19
3.3.3 Consistency of Terms .............. 20
3,3*4 Indicator Selection Criteria ... 20
3.3.5 Causation Versus Correlation 20
3,3.6 Indicator Development Process. ......... 20
3.3,7 Indicator Research 21
3.3.8 Summary Chapter 21
4.0 Summary of Recommendations 21
vi
-------
i.o
The Environmental Monitoring Subcommittee of the Ecological
Processes and Effects Committee of the science Advisory Board met
on May S-9, 1990, to review the draft document "Ecological
Indicator Report for the Environmental Monitoring and Assessment
Program" (EMAP).
The Subcommittee recognizes the inherent difficulties
involved in identifying EMAP indicators and appreciates the
substantial effort and progress reflected in the report. We
recognize that this report is being developed as an interim
document to summarize work to date in developing EMAP indicators
and to elicit further comment and we reviewed the document from
that standpoint. We have identified a number of conceptual and
substantive issues that require further clarification and
resolution. Our Subcommittee's comments and recommendations
address both fundamental questions with respect to selection of
indicators, and questions relating to improvement and
clarification of the way-in which such information was presented
in the report.
The Subcommittee also commented on broader aspects of the
EMAP conceptual design and sampling strategy which we consider
inextricably linked to selection of appropriate indicators, we
reiterate support for the EMAjP concept, and commend the EMAP team
for its efforts toward inter-agency coordination and peer review*
We further stress the importance of testing IMAP concepts through
conduct of pilot studies and for carefully evaluating results of
those studies before proceeding with full scale implementation.
Among our more salient findings and recommendations are;
1. Clarification is required on the concept of assessing
ecosystems as "nominal11 or "subnominal"? this concept has not
been clearly developed.
2. Clarification is required on the definition of "habitat"
indicators. There is ambiguity in the way the concept of habitat
condition is being applied as an indicator of environmental
deterioration.
3, The report presents unrealistic expectations concerning
information that can be provided through the current monitoring
strategy with respect to determination of causality for observed
trends and evaluation of the effectiveness of EPA*s regulatory
programs.
4. We strongly recommend that the EMAP team make maximum use of
historical data and data from other sources to help guide design
-------
of the monitoring program and complement EMAP monitoring data,
and that it synthesize such information to provide timely
assessments.
5. There is a need for better coordination and integration of
indicator selection among the ecosystem classes, and between EMAP
and the Ecorisk program.
6. We recommend that EMAP consider a pilot to test use of
biogeographic regions as an assessment units as a possible
alternative to assessment organized by ecosystem classes.
7. We recommend that EPA undertake an accelerated research
program to develop improved ecological indicators that will
identify changes in the ecosystem with sufficient sensitivity to
be useful in assessing the state of environmental quality,
8. Subsequent versions of the Ecological Indicator Report
should be revised to;
« Provide more explicit documentation of rationale for
indicator selection and the process used.
Discuss the classification of ecosystem types within
each ecosystem class,
* Provide a glossary, and ensure consistent use of
terminology.
Include matrices of selection criteria for indicators.
• Clarify that EMAP is not designed to demonstrate
causality,
Because the indicator concept is central to success of the
EMAP program, and because the concept is still evolving and
undergoing testing in pilot projects, we recommend that SAB
revisit the indicators issue later in 19iO or early in 1991.
2.0 INTRODUCTION
The Environmental Monitoring Subcommittee of the Ecological
Processes and Effects Committee of the Science Advisory Board met
on May 8-9, 1990, to review the draft document "Ecological
Indicator Report for the Environmental Monitoring and Assessment
Program" that was prepared by the office of Research and
Development (QRD) of EPA. In addition to the draft report, the
Subcommittee received a series of briefings concerning
development of indicators from key EMAP participants,
The goal of the Environmental Monitoring and Assessment
2
-------
Program (EMAP) ia_to provide unbiased estimates concerning the
current and changing condition of ecological resources at
regional and national scales. The EPA, U.S. Congress, and
private environmental organizations have long recognized the need
to improve our ability to document the condition of our
environment. In 1988, the EPA Science Advisory Board recommended
that EPA initiate a program that would monitor ecological
resource status and trends, as well as develop innovative methods
for anticipating emerging problems before they reach crisis
proportions. EKAP is being initiated in 1990 by EPA in response
to these needs.
A critical step within the program is the selection of
indicators that effectively define the status and condition of
ecosystems. The Indicator Report serves as an interim conceptual
plan for the indicator component of EMAP until more detailed
plans can be prepared for each ecosystem. The purpose of this
document is as follows:
-- To inform potential EMAP data users of the approach
proposed to describe ecological condition.
— To outline a framework for selecting and evaluating
indicators for further testing;
-- To seek expert advice and ecological data sets that
might aid in these evaluations,
2.1 Charge to the subcommittee
At the December 12, 1989, meeting of the Ecological
Processes and Effects Committee, the Director of the office of
Modeling, Monitoring Systems, and Quality Assurance of ORD
invited the Science Advisory Board to review critical aspects of
EMAP. The Committee accepted the invitation and requested that
ORO submit the draft document for Ecological Indicators to the
Science Advisory Board for review. On March 28, 1990, ORO
provided the draft report to the SAB for review. The Office of
Research and Development asked that the Committee address the
following questions as part of their review:
a. Are the concepts of the indicators clearly articulated?
b. Are the data in the appendices on individual indicators
relevant and accurate?
c. is the suite of indicators sufficient and appropriate
for each ecosystem? Will the results of such monitoring be
relevant given the temporal and spatial scale of the program?
d. Is the current classification of ecosystems appropriate?
-------
3.0 EVALUATION OF THE INDICATOR REPORT
In evaluating the proposed approach to indicators, the
Subcommittee found it necessary to consider the broader context
of the EMAP framework. The selection of appropriate indicators
is inextricably linked to EMAP's fundamental conceptual design
and sampling strategy, as outlined in the initial sections of the
Ecological Indicator Report.
Therefore, our comments address the general EMAP approach as
well as the specific issue of indicators. To a large extent,
these general comments reiterate or expand upon the Ecological
Processes and Effects Committee's report on Evaluation of the
Core Research Program for Ecology, transmitted to EPA on July 11,
1990,
Section 3.1 presents the Subcommittee's most significant
findings,
Section 3.2 presents more specific comments with respect to
indicators for the various ecosystem classes. Because of the
breadth of subject matter and relatively limited time for review,
the Subcommittee focused most of its attention to the issues
reflected in 3.1, which we considered to be of a strategic
nature. Our comments in Section 3.2 should be considered more
supplementary in nature; they do not represent a definitive
review of each of the ecosystem classes.
Section 3.3 presents recommendations concerning more
effective presentation of information in subsequent versions of
the Ecological Indicator Report.
3.1 General comments
3.1.1 Tlie EMAP Concept. The Subcommittee reiterates
support for the EMAP concept. The 1MAP strategy is to impose a
grid on the U.S. and examine remote sensed {satellite and aerial
photo) information in every grid cell to obtain a probability
sample of landscape characteristics. This information will then
be used to define strata and select sites for later sampling of
chemical and biological variables on ground visits. In concept
this strategy can offer a good combination of statistical
sophistication, efficiency, flexibility, and practicability.
Through its landscape characterization component, EMAP can be an
important step toward establishing an integrated natural
resources inventory for the nation and we believe this aspect of
the program should be emphasized.
However, the Subcommittee has not reviewed statistical
design of the program, At this stage, we have only been
presented with a strategy. We do not yet know:
-------
what variables will be measured?
what resolution is required?
what measurement technology will be employed?
how many observations will be made per stratum?
what is the natural variation in these variables?
In concept, when these are known, the evaluation of the
statistical design will rest essentially on estimation of whether
the proposed design (as specified in terms of measurement
technology, number of observations) is in fact expected to
deliver the required resolution (in light of the measurement
error, sampling error, background variation).
Current understandings of natural spatial and temporal
variability are probably inadequate to fully support such an
evaluation? although this will vary among ecosystem classes and
biogeographic provinces. This highlights the need for the EMAP
planners to evaluate existing data sets and information on
ecosystem variability in assessing the adequacy of the sampling
design. We are concerned whether 800 sites per ecosystem class
will be adequate to characterize variability over a large spatial
scale or will only yield largely site-specific results, we are
also concerned whether, without further stratification, the grid
will give sufficient coverage to characterize conditions in
smaller, but ecologically significant, ecosystems. We believe
that the alternate approach of developing a stratified sampling
plan based on existing knowledge ecosystems should be fully
evaluated. Area sampling techniques may be required.
3,1.2 The EMAP Strategy. The Subcommittee supports the EMAP
strategy of conducting pilot studies and stresses the importance
of conducting and carefully evaluating the results of pilot
studies to test EMAP concepts. Because of many unresolved
questions, which will be outlined below, we wish to emphasize the
importance of conducting and fully evaluating the results of
initial pilot studies before beginning full-scale implementation
of the monitoring program.
3,1.3 Interagency Coordination. The Subcommittee reiterates
the Ecological Processes and Effacts Committea's recognition of
the importance of in'ter-agency coordination. We commend efforts
to achieve interageney cooperation and participation in EMAP, and
wish to emphasize the importance of continuing these efforts if
EMAP is to succeed. Integration of monitoring efforts of other
Federal agencies into EMAP, along with associated cost sharing,
is both efficient and essential. In the case of forests, for
example, compatibility with the U.S. Forest Service»s Forest
Inventory and Analysis program is very valuable and Forest
-------
Service representatives should be active participants in the
planning and implementation of the forest pilot studies.
At the same time, we recognize the potential risks in
dependency of EMAP on other agencies such as the U*S» Forest
Service, Soil Conservation Service, Geological survey, Fish and
Wildlife Service, and NOAA to produce much of the status and
trends monitoring data. Therefore, formal arrangements between
EPA and other agencies must be developed and commitments made to
their scope and time commitments in regard to reporting results
if complete and timely state-of-the-environment reports are to be
published*
3.1.4 Peer Review. The subcommittee feels that peer review
is essential for all aspects of EMAP and commends the EMAP team
for the leval of peer review that baa been incorporated into the
development of th* pr«??rftIflT It encourages the continued use of
appropriate peer reviews as components of the program are
developed, refined, and implemented. In future reviews of EMAP,
the SAB will focus on critical components (e.g., strategic plans,
landscape characterization, demonstration projects, and risk
characterization) and issues related to the relationship of EMAP
to other EPA programs and objectives.
3.1.5 Discussion of the Charge, The Subcommittee addressed
the four questions posed by ORD In its charge to us and ve have
specific concerns related to each, both with underlying content
and with the manner of report presentation. The "Ecological
Indicator Report for the Environmental Monitoring and Assessment
Program" may be the first formal output of EMAP. EMAP is a major
component of the Agency's core ecological research program and it
will be the largest funded portion of that research effort.
Since EMAP is based on monitoring, which of itself has always
been controversial in the research community, its methods,
approaches, and techniques will be scrutinized closely by the
scientific community. This selection of ecological indicators
and the rationale for sampling, is not only fundamental to the
entire program but also may be an effort by which the entire
program is initially judged a. priori.
We recognize the inherent difficulties involved in
identifying indicators for SMAP, and appreciate the substantial
effort and progress -reflected in the report. We further
recognize that the report is being issued as an interim document
and that many of the conceptual and substantive issues that we
will pose below may not be quickly resolved, but we urge that
they receive continuing attention as the program evolves. We
also recommend an accelerated research effort to develop improved
ecological indicators that will identify changes in the ecosystem
with sufficient sensitivity to be useful in assessing the state
of environmental quality.
-------
3.1.6 Ecosystem Assessment* Clarification is required em
the concept of assessing acosyatams m.m "nominal" ox "subnominal*"
The report describes the EMAP assessment goals for determining
the percentage of each ecosystem class that is in nominal versus
subnominal condition nationally. However, the concept of
defining nominal conditions has not been clearly developed.
A core difficulty seems to be a desire that the EMAP survey
deliver an overall assessment of "ecosystem status" as a measure
of ecological "health" or "integrity." This goal then motivated
generation of a long list of indicators in the hope that these
would somehow add up to an adequate measure of ecosystem status*
However, there is no monolithic ranking of ecosystem status
which captures whether the state of the ecosystem is good
("nominal") or not. A piece of landscape can be in good
condition in one respect, and bad in another? it can be good for
some uses, and bad for others.
Even where the reference site concept is used to describe
"nominal" conditions, we must recognize that ecosystems are
dynamic and are influenced by short and long-term natural forcing
functions. Thus, the characteristics of an ecosystem may be
fluctuating in ways that confound defining "nominal," unless
response to natural variation can be incorporated in that
definition. Since terrestrial ecosystems will be identified with
vegetative type, the state of succession will be a factor. If by
"nominal" we mean the potential ecosystem or the seer leading to
it, then it will be necessary to identify the state of succession
at the time of the assessment. In addition, it will be necessary
to predict how an unperturbed site would change with time.
EMA'P does not yet appear to have adequately addressed the
question of change induced by natural stress, A major issue
generic to all the ecosystem classes is the need to understand
natural cycles which would affect the proposed indicators, as
distinct from anthropogenic stress. For example, in the case of
coastal systems, these would include diseases that result in
mortality or death of bivalves and submerged vegetation, and
cycles of fish population changes. The need, to understand
natural variability emphasizes the importance of utilizing
historical data sets, where these are available, to help
interpret data collected by EMAP,
One possible approach inight be for the EMAP planners to
abandon the concept of ranking ecosystems as nowinal or
subnominal. Instead, the focus could be on generating a short
list of assessment endpoints which individually are directly
justifiable in terms of societal/management interest, to adopt
measurement endpoints that can be measured and related in some
meaningful way to the assessment endpoints, and to rank the
condition of ecosystems in those terms.
-------
3.1*7 Habitat indicators, clarification is required on the
definition of fcabitat indicators. Quantitative habitat loss is
one measure of environmental deterioration that can be described
clearly, but loss of habitat quality is less clearly defined.
There is some ambiguity in the way the concept of habitat
condition is being applied as an indicator of environmental
deterioration. Ecosystem characteristics which describe the
"natural" structure and function of ecosystems and which can
serve to characterize ecosystems dp not appear to be
distinguished from those which are measures of physical
deterioration (e.g., indices of range condition).
3.1.S Sampling Strategy. Th* draft Indicators Report
presents unrealistic expectations concerning the information that
can be provided through the current monitoring strategy. The
fundamental EMAP concept proposes annual monitoring at a
statistically designed pattern of fixed stations. Adherence to
the pre-determined sampling design, will limit the selection of
indicators and some otherwise relevant information will not be
included.
An example is use of information on fisheries in the coastal
demonstration; the Subcommittee agrees that it is not feasible to
collect meaningful data on commercial fish populations in the
EMAP sampling pattern. However/ some mechanism for addressing
the status of fisheries is essential to assess the condition of
the marine environment in terms that will be most meaningful to
policy makers and the general Public. ,
The Subcommittee has not reviewed the Near-Coastal Program
Plan for 1990, but as we understand it, the pilot will not
include assessment endpoints of general societal concern, such as
status of fish populations, shellfish bed closures, recreational
water quality, sea bird and coastal wading bird populations.
Further, it will be inherently difficult, if not impossible, to
link the proposed measurement indicators directly to such
endpoints in any quantitative way. For example, the confounding
effects of exploitation, such as sport and commercial fishing,
will limit our ability to relate trends in raarine environmental
deterioration to changes in fish populations*
We understand the difficulty of including such assessment
endpoints within the basic IMAP sampling design. At the same
time, however, these are the questions that will be asked by the
public and policy makers. This dilemma suggests two possible
alternatives relating to the fundamental EMAP strategy;
a. Make a strong effort to supplement the EMAP field
sampling of the limited suite of measurement indicators with
relevant information collected from other sources and a broader
base to provide interpretative reports on the condition of near-
coastal resources. Examples are: NMFS and State fisheries catch
8
-------
statistics; Fish and Wildlife Service information on breeding
shorebird and waterfowl populations and on contaminants in
colonial nesting wading birds. Information from such sources can
help EMAP to interpret natural variation and population cycles in
relation to various measurement indicators and assessment,
endpoints (e.g., declines in oysters and sea grass in the
Virginian Province of the pilot study, which take place
periodically due to disease outbreaks), or
b. Be more explicit that the EMAP coastal monitoring
program is simply measuring some sentinels of marine
environmental condition, and will not provide information that
directly correlates with the status of endpoints of management
concern. It could be extremely useful to meet this more limited
monitoring objective, which can alert deeisionmakers to trends in
environmental deterioration and indicate the need for intensive
field studies and research to determine the causes and societal
consequences.
However, currently there is inconsistency between what the
report indicates that EMAP will provide, and what the coastal
pilot as proposed will actually provide. For example, the draft
indicator report seems to imply that the monitoring program will
explain causal relationships in many cases, and that it can be
used to evaluate the effectiveness of EPA's regulatory programs.
This could lead to false expectations and a loss of credibility.
3,1.9 Historical Datft, The Subcommittee reiterates the
Ecological Processes and Effects Committee's finding concerning
use of existing data and recommends that th« EMAP team make the
maximum use of Historical data and data from other sources to
complement data from EMAP monitoring, and that it synthesize such
information to provide assessments in a timely fashion. EMAP
provides for incorporation of data from other monitoring networks
and sources. However, it appears that EMAP may not be providing
adequate resources or priority to analysis and use of historical
data in interpreting the status and trends of ecological
resources. Some provision is being made for use of "found" data,
but this effort seems to be receiving only secondary attention in
EMAP planning. The Subcommittee wishes to stress the importance
of making adequate provision both in time and resources for
incorporating, interpreting, and synthesizing all relevant
information, including historical data (e.g., water quality
monitoring, forest resource inventories, and agricultural
resources),
No matter how well EMAP is designed, as we discussed above
with coastal ecosystems, IPA's ability to interpret the findings
of a fixed point sampling program with a relatively limited
number of sampling points will be greatly enhanced if IMAP data
are viewed in the light of other existing information. Perhaps
even more important, analysis of existing knowledge may lead to
-------
important modification in the EMAP program design. Such
information may point to the need for adopting stratified
sampling approaches or selecting other indicators without biasing
statistical integrity.
Further, an emphasis on integration and synthesis will
enable the program to begin to issue useful interpretative
reports and findings early on* This will help assure continued
interest and support for the major budgetary expenditures
necessary to implement and maintain EMAP? it may not be possible
to sustain the necessary commitment if many years elapse before
some findings are issued.
3.1.10 Model Ecosystem Approach. Th« Subcommittee supports
the approach described by th« arid, lands group. In relation to
the above recommendation, the Subcommittee was impressed by the
presentation made for the arid lands group, even though it was
informal and preliminary in nature. Information available froa
other agencies, e.g., the Bureau of Land Management and the Soil
Conservation Service, was used to prepare interpretative maps of
arid lands for various parameters such as greenness and
evapotranspiration. The approach of using remote sensing data
seemed best developed in the arid land study. The indicators
proposed seem veil suited for a broad-scale study like EMAP,
We recommend that the other ecosystem groups examine and
apply the approach used by the arid lands group. For example,
the utility of the satellite mapping and aerial photography needs
to be more clearly stated in each ecosystem. The present general
plan is to use these tools for establishing the extent and
boundary patterns of various resources* In the specific case of
the arid ecosystems, it also planned to use information derived
from remote sensing to estimate environmental impacts associated
with such factors as grazing and decertification. The potential
for broader use of remote sensing information should also be
explored for other ecosystems classes.
3.1,11 coordination and Integration, There is a need for
better coordination and integration of indicator selection among
the ecosystem classes, and between EMAP and the leorisk program.
We observed inconsistency in the use of indicator terminology and
concepts among the ecosystem groups, particularly in the oral
presentations made to the Subcommittee. Criteria for choice of
indicators also seemed to be unclear or specified unevenly
applied among the various ecosystem groups. Further, although
the report identified a number of cross-cutting indicators
relevant to a number of ecosystem classes (Chapter 9), there was
little indication that these would be incorporated as appropriate
into the individual ecosystem classes.
In addition, we believe that there is a need for the EMAP
program and the Ecorisk program, both are components of EPA's
10
-------
Core Research program, to coordinate with respect to selection of
endpoints and techniques for risk characterization. As the
Ecological Processes and Effects Committee pointed out in its
evaluation of the core research program, it is important to have
a strong linkage and feedback between the field monitoring and
risk characterization research* p Monitoring can frame and define
some research topics and research can focus monitoring
procedures. We saw little evidence of such coordination.
3.1.12 Biogeographic Regions, fits Subcommittee recommends
that EMAP consider us* of biogeographic regions as assessment
units as a possible alternative to ecosystem classes.
Development of EMAP around the concept of ecosystem classes
(e.g., wetlands, forests, inland waters, etc.) poses some
inherent problems:
a. Many "real-life" ecosystems include elements of pore
than one ecosystem category. For example, a watershed may
include inland waters {rivers and lakes), various wetland types,
agro-ecosystems, and forested areas. To characterize that system
and understand ecosystem functions, interactions among the
various elements must be understood.
b. In many cases, there is greater commonality among the
elements within an ecoregion or biogeographic province than there
is among different types within the same ecosystem category. For
example, Sp.artina marshes relate much more closely to their
adjacent estuarine waters than they do to mangrove swamps or bogs
even though these are all categorized as "wetlands ecosystems."
The wetlands category of classification covers a
particularly broad range of systems, including such disparate
wetlands as salt marshes, northern cedar swamps, peat bogs, and
southern cypress swamps. Their commonality is limited, to the
presence of water and the presence of plants that are adapted to
water saturated soils. As a consequence of this very broad
definition, ecosystems with only remote affinities are
aggregated. Consequently, the actual ecological affiliations
between a northern cedar swamp and a New Jersey salt marsh are
probably less than the affiliations between a Michigan pine
forest and a Kansas prairie. The individual members in the
present category of wetlands clearly have their closest
affinities to the adjacent upland ecosystems on the one hand, and
the adjacent aquatic ecosystems on the other hand. Consequently,
it is particularly important that the wetlands category should
take cognizance of the relationships of the individual wetland
ecosystems to their affiliated ecosystems.
c. Some ecosystems may have dual classifications because
of affinities to more than one category. For instance, tundra
may be classified as arid on the basis of low rainfall, but as
wetlands on the basis of the occurrence of standing water, the
11
-------
presence^of plants adapted, to water saturated soils, and the
utilization as breeding territories by water birds. The
extensive areas devoted to aguaeulture in Arkansas and
Mississippi can be classified in the category of agro-ecosystems
or in the category of inland surface waters. The northern cedar
swamps or the southern cypress • swamps can be categorized as
wetlands or as forests.
d. EMAP is attempting to monitor and assess ecological
functions on a grand scale. The ecosystem classes ignore areas
that are geographically and ecologically related and in some
cases interdependent* Their nature and potential condition may
be linked to a common, widespread geological, Meteorological, or
hydrological phenomena which are easily revealed through mapping.
In addition, with the ecosystem classes, fundamentally different
approaches are being taken to define indicators for different
ecosystem categories. As a result, it will be difficult to link
and compare conditions among the various ecosystem categories
even where they occur in the same biogeographic region.
Therefore, we recommend that a pilot b* developed for at
least one inland area, selected from Omeruik's ecoregions 07
other similar ecological classification system, in which aa
integrated EMAP approach be developed and tested. Based on the
outcome of this pilot, a determination should be made concerning
whether to move to an ecoregion approach instead of the current
ecosystem categories approach as the basis for further program
development.
3*1.13 Cost Estimates, Th* cost •stimates in the report
ar« questionable The sampling and analytical costs presented in
each of- the appendices are open to question and criticism. They
are difficult to evaluate in the context in which they are
presented. Moreover, they are misleading because no
summarization of total costs is provided for each ecosystem type.
The numbers shown are without meaning because the component
institutional costs, direct and indirect, are not reflected.
without these, one cannot arrive at an overall realistic estimate
of the total program effort.
3.2 Specific Comments on Proposed indicator*
3.2.1 Cross-Cutting Indicators. The subcommittee has
additional specific comments as follows;
a. Biomarkers, The biomarker section in Section 9 of the
report is generally well written. However, biomarkers also
should be included more specifically within the individual
ecosystem class discussions. In particular, the needs or uses of
biomarkers in each ecosystem class should be specified.
The list on page 9-16 should distinguish changes in gene
expression (i.e., synthesis of proteins in response to stressors)
12
-------
from changes in enzyme activity/protein function.
The use of wildlife biomarkers to signal significant changes
in the health of the ecosystem (or region) under consideration
could be very valuable. There is, however, a spatial (and
temporal) resolution issue that has not been clearly addressed.
Different species (e.g., honey bees, frogs, birds) sample very
different spatial scales and temporal scales. Therefore,
biomarkers associated with animal species will have to be
selected and monitored with knowledge of the temporal variations
in species behavior (e*g», change in location as a function of
the season). For example, for migratory species (e.g., birds),
effects as revealed by biomarkers evaluated for a given species
in a given site and a given time could be the consequence of
events of prior time. Thus, the time-effects correlations may,
in some cases, be extremely difficult due to the coupling of
spatial and temporal features associated with specific indicators
and regions.
An alternate approach to use of species-specific biomarkers
might be to identify biological guilds and monitor the rate of
resource utilization, as has been done successfully for aquatic
systems by Cummins efc aI (1989, Shredders and Riparian
Vegetation, Bioscience ^9;24-30),
b. Regulatory criteria or standards, such as closing of
shellfish beds and beaches due to fecal contamination, or health
advisories on fish consumption due to chemical contamination, are
important indices by which environmental decisionmakers and the
public assess the status of the environment. These regulatory
requirements are not uniformly stated (e.g., individual states
have different criteria for advisories, fishing bans, and
closures-) . Further compliance or non-compliance may be unrelated
to environmental contamination or effects that can be monitored
directly. Despite these limitations, compliance with
environmental criteria represents an important way of describing
environmental status and trends. Therefore, EMAP should consider
how to incorporate information on such compliance in its
assessments.
c. Sediment/soils represent sinks of contaminants for all
of the ecosystems evaluated by EMAP. These data can thus provide
an important link between ecosystems. The collection of
sediment/soil contamination data should be incorporated as an
exposure indicator in all of the ecosystem types and the
methodologies standardized to facilitate comparisons between
ecosystems.
d, Climate and atmospheric deposition are driving
variables that are the root of much ecosystem variation and
should be measured along with the ecosystem indicators. The
coupling of atmospheric deposition monitoring with ecosystem
13
-------
monitoring is essential. Over time, it will permit correlation
studies of the relationships between air quality and ecosystem
responses. These correlations may prove very valuable in efforts
to partition stress effects between natural and anthropogenic
stress factors,
Atmospheric deposition is the source of a significant
portion of toxicants coming onto a site whether terrestrial or
aquatic. In addition to efforts to calculate deposition loadings
at reference sites for all regions, there should be an effort to
tie deposition to sources by considering meteorological patterns.
3.2.2 Wetlands. Tha Subcommitt** had some specific
comments on the wetlands chapter aa follows:
a. This chapter is generally well done. The choice of
response indicators appears appropriate but the exclusion of
primary production as a response indicator should be discussed
further by the team.
b. Unlike the other ecosystem discussions, the
developmental indicators are only presented in the table in 5-7
but not discussed in the text. The relevance of these
developmental indicators in evaluating wetlands should be
reviewed at least briefly.
c. There is a major problem with the lack of connection
between the various types of wetlands and the ecosystems to which
they are physically linked. An example is the salt marsh-wetland
ecosystem and its linkages to the adjacent estuary. As we have
previously indicated, these issues should be discussed and the
various types of indicators should be compatible so that the data
can be compared and the linkages evaluated.
d. Chemical concentrations in the sediments are called for
in Figure 5-1 and on page 5-10 and, as already indicated, these
are important contaminant sinks. However, there is some question
about the validity of such measurements within the limited
sampling periods proposed by EMAP. Chemical activity in wetlands
sediments is extremely complex. What one finds one day can be
quite different the next. Since wetlands are shallow water
systems, the sediment quality is subject to changes related to
atmospheric conditions. High evaporative and transportational
losses will modify both ionic and cationic migration as will
sudden dilution of the surface water with precipitation.
Therefore, these short-term variations could mask significant
trends unless provision is made for more intensive sampling.
In any case, sediment monitoring should be linked to ongoing
work in the agency on sediment quality criteria (i.e., AET and
EqP methods) so that the data can be integrated into these
programs and to facilitate its interpretation.
14
-------
e» Some wetlands are cm a relatively short successions!
timeline (i.e., acid bogs). This further illustrates the issues
discussed in 3.1,6, above, with respect to the need to
differentiate between anthropogenic change and man-mediated
changes, and problems associated with definition of "nominal."
3.2.3 Near-Coastal Waters. Tb* subcommittee comments on
the Near-Coastal chapter do not reflect subsequent EMAP work done
in preparing the Near-Coastal Program Plan for 1990, which we
have not reviewed. We recognize that son* of our comments nay
have been superseded by that document. Our comments on the
Ecological indicator draft are as follows:
a. Choice of assessment endpoints. The Near-Coastal
program identified assessment endpoints related to the public's
use of near-coastal ecosystems for commercial and/or recreational
fishing. Primary endpoints are the health of fish and shellfish
populations. A second major endpoint of concern is the
maintenance of near-coastal habitat structure. Public perception
of the aesthetic value of the resource is the third endpoint,
with clarity of water as influenced by suspended solids and/or
algae as the measurement endpoints for this area. The
Subcommittee agrees that these three categories of endpoints
relate to aspects that the public values and therefore are
reasonable assessment endpoints toward which to direct a status
and trends monitoring program.
b. Response indicators. The core response indicators
proposed are dissolved oxygen and benthic abundance, biomass, and
species composition. These indicators are related to the
assessment endpoints, even though as previously indicated it will
not be possible to establish quantitative relationships, and they
seem to be reasonable choices.
c. Interpretation of benthic community data will be
extremely difficult without information on the benthic habitat.
The distribution of benthic organisms is highly dependent on the
nature of the habitat. Some effort should be made to physically
characterize the habitat present at the sampling sites with
respect to both grain size and organic carbon content. Sediment
grain size alone may not be adequate.
d. While the Subcommittee recognizes the difficulties in
collecting fish, it recommends that fish abundance and species
composition {NOTE: Contrast with acknowledged problem on fish
population data under 3.1.8,}, and fish gross pathology be
considered within the program. These two measurements most
directly relate to the public's assessment endpoints. We
recognize that fish abundance is affected by natural cycles,
fishing effort, and anthropogenic chemicals, and that it will be
difficult to separate the influence of these different factors on
15
-------
species abundance. We have discussed fundamental issues
associated with inclusion of fish population sampling in our
comment 3.1*8., above *
With respect to the gross pathology, the fish health
protocol developed by R.w. Goude (Experimental Fisheries Research
Laboratory, Division of wildlife Resources, State of Utah, Logan,
Utah) should be investigated for use in place of (or in addition
to) the gross pathology indicator. This protocol has been tested
on cold and warmwater, freshwater, and marine species with
considerable effect. It requires little more effort than a gross
pathology examination and will give an earlier warning signal
than gross pathology. Fish abundance is affected by natural
cycles, fishing effort and anthropogenic chemicals. It may be
difficult to separate the influence of these different factors on
a species abundance.
f. The report identifies several promising developmental
endpoints. The Subcommittee agrees with the report that
additional research is needed before biological mixing depth and
presence of large infaunal bivalves can be routinely used. The
report also indicates that extent and density of submerged
aquatic vegetation cannot be routinely implemented, and proposes
it as a developmental indicator. While it is possible to
routinely sample for aquatic vegetation, it may not be suitable
as an indicator within the EMAP sampling scheme since submerged
vegetation is not present in many coastal areas of the
southeastern United states,
This issue nay apply to other indicators as well: dissolved
oxygen sampling would not be useful in highly flushed areas where
there are high daily tide changes, and water clarity studies
would not be useful in naturally turbid. Areas of sewage
outfalls often have large, healthy bivalves due to organic
enrichment. This suggests that the suite of proposed indicators
may be useful for some sites but not for others.
g. Shellfish growth was eliminated as an indicator (page
3-13) because changes in growth could not be related to specific
environmental problems. This is not an appropriate reason for
eliminating this endpoint. It is important to acknowledge that
other response indicators, including benthic abundance suffer
from the same lack of specificity.
h. An important biological process which was not covered
by an indicator, except for some toxicity tests, was
reproduction. A variety of events are important in reproductive
success of vertebrates and invertebrates which have received
intensive study. These include ovary formation, synthesis of
vitellin (major egg protein), and hatching of young and the
levels of fecundity and fitness. If normal reproduction is
occurring in a species, it is probably one of the best indicators
16
-------
that the species will maintain its population. Indicators based
on reproduction should be considered for inclusion in the
program.
it Exposure indicators. Core exposure indicators include
acute sediment toxicity, chemical contaminants in sediments, and
water clarity. The Subcommittee is concerned that a suite of
acute sediment toxicity tests are not proposed; use of a single
amphipod species for evaluating sediment toxicity is
inappropriate. Experience gained from ambient toxicity testing
in freshwater and marine ecosystems has demonstrated that it is
prudent to use a battery of assay organisms to determine water
and sediment toxicity because of the range of sensitivities of
assay organisms to toxicants. Lacking knowledge on the
toxicant(s) present in coastal waters and sediments, it seems
wise to include a battery of assays. Further, EPA's guidelines
for establishment of water quality criteria call for bioassays en
an array of organisms,
The report proposes that sediments be assayed for NOAA's
list of proposed measurements in surficial sediments, while many
of the chemicals on MOAA'S list are important chemicals of
concern, it might be advisable to use a broad screening approach
such as that used in the Chesapeake Bay. This approach can
incorporate specific chemicals of concern as well as provide
information on a much wider range of contaminants.
It is unclear to the Subcommittee why the chemical residues
in fish and shellfish were not included as a Core Indicator.
Levels of contaminants in fish and shellfish directly relate to
an assessment endpoint valued by the Public.
j. Use of EMAP indicators for "early warning." It is
difficult to predict what future problems lie ahead for the
coastal zone, but if the EMAP approach can provide an "early
warning," it could prove very useful. One way of assessing the
usefulness of the proposed indicators is to assess whether they
would have been able to predict subsequent environmental problems
in coastal waters, e.g., problems with PCBs, Kepone, and tributyl
tin. other than through some of the proposed chemical analysis,
it is not clear that the indicators would have "caught" these
problems. Many of the proposed indicators have been useful only
where pollution is quite dramatic. These include gross fish
pathology and most of the biomarkers. We have no specific
recommendations concerning the "early warning" issue, but believe
that it merits continuing1 attention by the EMAP planners.
3.2*4 Forests. fftt lav* the following comments with respect
to the Forests strategy: The primary measurements most
appropriate for forest monitoring include tree height and
diameter, leaf area, and analyses of soil contaminants. From
these, core indicators including growth efficiency, leaf area
17
-------
index, and integrity of foliar can be calculated. These primary
measurements are superior because they are simple, standardized,
time-tested, and they integrate forest tree response with
multiple stress factors. The technology for leaf area
measurements is developing rapidly, and soon, it is highly
probable that strategies for land-based, low altitude, and
satellite Measurements will be improved. Leaf area is the most
useful leaf characteristic available. Foliar symptoms are not
always specific to different stresses, but forest damage that has
been induced by ozone and certain other gaseous pollutants can be
identified with some degree of certainty. Other than these few
instances where foliar symptoms can be identified, elaborate
efforts to inventory various types of symptoms may not be cost
effective. Leaf bioraass is the critical factor that determines
forest productivity,
The analysis of persistent contaminants (e.g., heavy metals
and chlorinated organics) in forest soils (particularly those of
the forest floor) is especially appropriate for a long-term
infrequent sample cycle as proposed for EMAP. However, the
variances associated with foliar and soil nutrient analyses may
render these measurements of limited usefulness. Needles of
conifers also offer a valuable method of measuring amounts of
deposition over shorter time scales because there are usually
several year classes of needles present, and the needle
configuration is favorable for collecting gases and particles.
Proposed pilot studies in the northeast and southeast represent
an excellent opportunity to test or evaluate procedures.
3,2.5 Arid Lands, As already indicated, the Committee was
impressed with tfce concept developed for arid lands and
particularly vita tha utilisation of remot* sensing techniques
for eliciting many of the indicator variables.
a* EMAP issues are, to a large extent, resource issues
that have application on a global scale. It is to the credit of
the arid lands group that they are taking advantage of
international experience in the formulation of their plan.
Particularly useful will be the reference to the experience and
data of the IBP and LTER efforts.
b» Much of the arid west is under little plant cover and
is subjected to substantial fetches of wind reaches. These
forces may be instrumental in redeposition of materials
accumulated in friable playa soils. Consideration should be
given to establishing meteorological sites to measure this
deposition and its distribution patterns.
3.2.6 Agro-ecosystem. Our comments on agro-ecosystems are
as follows:
a. The category of agro-ecosystems represents a somewhat
18
-------
artificial aggregation of systems from an ecological point of
view. The commonalities between rangelands, row crops, orchards,
and aqua-culture appear to be remote, and it may be advisable to
assign some of these agro-ecosystems to ecosystem categories with
which they may have closer affinities, such as rangelands with
prairies and aqua-culture with aquatic systems. Alternative, a
clearly established mechanism to foster integrated assessments
among categories of ecosystems is required. (See comment 3,1.12,
above),
b. Agriculture in the arid regions of the country poses
fundamentally different problems for the environment than
agriculture in those regions of the country where water is not a
major limiting resource. Consequences of irrigation practices,
such as increased rates of salinization and the transport and
accumulation of elements that result in toxicity, such as
selenium and boron, are examples of the overlap and special
coordination that is needed between the groups dealing with agro-
ecosystems and arid land ecosystems.
c* It is important that the assessment of agro-ecosystems
take account of social-political actions to be able to assess
changes in farming activities due to actions such as incentives
to let land lie fallow (e.g., the Soil Bank), trends toward low-
tillage agriculture, "clean" versus "dirty" farming, etc.
3,3 Recommendations for Revising tn* Indicator Report.
The subcommittee has several recommendations toward a more
effective presentation of information in tn* report in addition
to more substantive comments on sanction and testing of
indicators. These comments were conveyed orally to EPA
representatives after our May 8-9, 1990 review, and we understand
that they were considered in revising the draft report. If not,
we recommend that they be reflected in any future reports on this
topic.
3*3.1 Document Rationale. The sections dealing with
indicator strategies for particular ecosystem types suffer from
insufficient description and rationale. The audience that may be
reading this report is lifcely to be a critical one, and a
scientific case has to be made for each suggested indicator or
developmental indicator for each ecosystem type regardless of
length. This may involve including more information from the
appendices in the body of the report,
3.3.2 Resource Types. Include information on resource
types for each ecosystem. In the draft report, very little
information is included in each ecosystem category chapter on the
types of ecosystem within that category. However, it became
evident during the briefing that the classification of ecosystems
within each category has a direct bearing on the choice of
19
-------
response and exposure indicators, Each ecosystem category
chapter should include a discussion of the classification
approach used (e.g., biogeographie regions, biomes, domains,
etc.) and the role this classification played in the selection of
indicators. As it currently exists, the report is opaque in this
regard and the reader is left to speculation.
3.3.3 Consistency of t*rms. The Subcommittee found the
report to be generally consistent in its use of terminology.
However, during the briefings, it was evident that different
ecosystem teams had different definitions for concepts such as
assessment endpoints, response indicators, etc. These
definitions produced some confusion within and among team members
and for the individuals being briefed. A well-defined glossary
of terms should be included in the report which is used by all
members of the EMAP team.
3.3,4 indicator Selection critaria. The Subcommittee
recommends that each ecosystem chapter have a matrix added to it,
which includes all candidate indicators considered for that
ecosystem and all evaluation criteria applied by the team to
screen candidate indicators. This would effectively communicate
to a reader the universe of candidate indicators considered for
each ecosystem and the rationale for identifying final core
indicator candidates and the selection of developmental
indicators. It would also assist in evaluating the consistency
of selection criteria and indicators between ecosystems. The
Forest Ecosystem team presented a very informative diagram in
which they ordinated their candidate indicators by feasibility
versus interpretability. The Subcommittee recommends that
consideration be given to a similar presentation in each
ecosystem chapter.
3,3.5 Causation versus correlation. The subcommittee feels
that the report should be carefully edited to clearly point out
that EMAP is not designed to demonstrate causality. In several
chapters, this point is not as strongly made as the Subcommittee
feels it should be. For example, many of the figures, if
examined out of context, imply that SMAP monitoring may
demonstrate causality for subnorainal conditions? whereas, in
general, such monitoring can probably only demonstrate
environmental trends.
3.3.6 Indicator Development Procass. The Subcommittee
recommends that the report include in each ecosystem chapter, a
section which discusses how the team arrived at the final
candidate indicators. Information should be presented on who was
on the team, the role of outside reviews, name and affiliation of
reviewers, etc. Inclusion of such information, helps place the
product of each chapter in perspective and makes the process
transparent„
20
-------
3.3.7 indicator Research, it could be useful to include,
perhaps as a separate appendix, a more penetrating explanation of
important research topics related to ecological indicators. This
could be valuable input for ORD to use in developing its core
research program? as already indicated, we recommend an
accelerated research effort in this area.
3,3.9 Summary chapter. The present Executive Summary needs
work. In part, it seems to be a shortened and cut version of a
number of the sections. The summary should cover the rationale,
the bioindicators concepts, and an explanation of the differences
between categories of indicators without lifting text material
directly from the sections which is what the present summary
reflects. Further, it is recommended that the summary chapter
compare and contrast the assessment endpoints, response,
exposure/ habitat, and stressor indicators by ecosystems. A
summary matrix such as was presented by Dr. Hunsaker during the
briefing could be used as the centerpiece of such a chapter. The
Subcommittee recommends that nor* effort be made to analyze,
compare, and contrast tha various paradigm* directing the
selection of indicators between the ecosystems, and that the
results of this analysis be included in the summary chapter.
4.0 SUMMARY OF RECOMMENDATIONS
The Subcommittee, reiterates its support for the EMAP
concept. We recognize that the Ecological Indicator Report is
being issued as an interim document to summarize work to date and
to elicit further comment, and we reviewed it as such. We have a
number of fundamental questions with respect both to the
selection of indicators for EMAP, and to the way in which
information is presented in the report, our findings and
recommendations are:
l. Clarification is required on the concept of ranking systems
as "nominal11 or "subnominal"j this concept has not been clearly
developed.
2. Clarification is reguired on the definition of "habitat"
indicators. There is ambiguity in the way the concept of habitat
condition is being applied as an indicator of environmental
deterioration.
3. The report presents unrealistic expectations concerning
information that can be provided through the current monitoring
strategy with respect to determination of causality for observed
trends and evaluation of the effectiveness of EPA's regulatory
programs,
4, We strongly recommend that the 1MAP team make maximum use of
historical data and data from other sources to help guide design
21
-------
of the monitoring program and complement EMAF monitoring data,
and that it synthesize such information to provide timely
assessments.
5. There is a need for better coordination and integration of
indicator selection among the ecosystem classes, and between EMAP
and the Ecorisk program. We saw little evidence of such
coordination.
6. We recommend that EMAP consider a pilot to test use of
biogeographic regions as assessment units as a possible
alternative to ecosystem classes*
7. We recommend that EPA undertake an accelerated research
program to develop improved ecological indicators that will
identify changes in the ecosystem with sufficient sensitivity to
be useful in assessing the state of environmental quality.
8. We have a number of specific comments and suggestions
concerning indicators for the individual ecosystem classes.
9, Subsequent versions of the Ecological Indicators Report
should be revised to:
* Provide more explicit documentation of rationale for
indicator selection and the process used.
Discuss the classification of ecosystem types within
each ecosystem class.
Provide a glossary, and ensure consistent use of
terminology.
Include matrices of selection criteria for indicators.
Clarify that EMAP is not designed to demonstrate
causality.
Because the indicator concept is central to success of the
EMAP program, and because the concept is still evolving and
undergoing testing in pilot projects, we recommend that SAB
revisit the indicators issue later in 1990 or early in 1991.
22
------- |