UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
                            WASHINGTON D.C. 20460
                                                    OFFICE OF THE DMINISTRATOR
                                                     SCIENCE ADVISORY BOARD
                                 April 18,2008

EPA-SAB-08-007

Honorable Stephen L. Johnson
Administrator
U.S. Environmental Protection Agency
1200 Pennsylvania Avenue, N.W.
Washington, D.C. 20460
          Subject: SAB Advisory on EPA's Draft Report on the Environment 2007:
                  Science Report

Dear Administrator Johnson:

   EPA's Office of Research and Development requested that the Science Advisory
Board (SAB) review the Agency's draft Report on the Environment 2007: Science Report
(ROE).  The ROE is an update of EPA's draft 2003 Report on the Environment which
was reviewed by the SAB in 2004.  In response to the Agency's advisory request, an
SAB panel conducted a peer review of the draft 2007 ROE. The enclosed advisory report
provides the advice and recommendations of the Panel.

   The ROE is intended to present status and trend information for indicators that reflect
the condition of human health and ecosystems in the United States.  The SAB commends
the Agency for its initiative in preparing this unique but ambitious report, and is pleased
that prior SAB  advice has been incorporated into the document. While the draft 2007
ROE represents an evolutionary advancement over the earlier 2003 version, the current
draft still does not fully meet its intended purpose. Although the ROE presents status
information to establish baselines for reporting future trends, the lack of long-term trend
information in the document precludes trend analysis for many indicators. The rigid
application of indicator selection criteria has resulted in the exclusion of valuable and
relevant information that could be used to further analyze trends.  The ROE is also
limited because it contains little data interpretation and no conclusions supported by
statistical analysis.  In addition, the ROE lacks a framework describing the scientific
understanding of relationships between indicators and the basis for including them in the
Report.  The SAB has provided recommendations to improve the  draft 2007 ROE and
strongly recommends that they be incorporated prior to releasing the final Report. In
addition, the SAB has provided recommendations to improve future Reports on the

-------
Environment with the expectation that those recommendations will require a longer time
frame to implement.

   The ROE has the potential to replace the sorely missed annual reports on the state of
the environment once published by the Council on Environmental Quality. The value
and importance of the ROE will continue to grow as pressures of population, energy use,
urbanization, and climate change lead to continued stress on environmental quality and
impacts on health and ecosystems. Therefore, as recommended in its review of the 2003
ROE, the SAB again strongly urges EPA to fully support and permanently embed the
ROE into its core mission-directed activities. This will require an investment of
resources beyond those currently devoted to the ROE.
                          Sincerely,
      /Signed/
Dr. M. Granger Morgan, Chair
Science Advisory Board
Attachments
      /Signed/

Dr. Deborah Swackhamer, Chair
Panel for the Review of EPA's
2007 Report on the Environment
                                       11

-------
                                  NOTICE

   This report has been written as part of the activities of the EPA 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 the 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. Reports of
the EPA Science Advisory Board are posted on the EPA website at
http://www.epa.gov/sab.
                                    in

-------
                  U.S. Environmental Protection Agency
                          Science Advisory Board
     Panel for the Review of EPA's 2007 Report on the Environment
CHAIR
Dr. Deborah Swackhamer, Professor and Interim Director, Institute on the
Environment, University of Minnesota, St. Paul, MN

MEMBERS
Dr. Henry Anderson, Chief Medical Officer, Division of Public Health, Wisconsin
Division of Public Health, Madison, WI

Dr. Fred Benfield, Professor of Ecology, Department of Biological Sciences, Virginia
Tech, Blacksburg, VA

Dr. Mark Borchardt, Director, Public Health Microbiology Laboratory, Marshfield
Clinic Research Foundation, Marshfield, WI

Dr. Timothy Buckley, Associate Professor and Chair, Division of Environmental Health
Sciences, School of Public Health, The Ohio State University, Columbus, OH

Dr. Aaron Cohen, Principal Scientist, Health Effects Institute, Boston, MA

Dr. David A. Dzombak, Walter J. Blenko Sr. Professor of Environmental Engineering
and Faculty Director, Steinbrenner Institute for Environmental Education and Research,
Carnegie Mellon University, Pittsburgh, PA

Dr. Dennis Grossman, Senior Policy Advisor, The Nature Conservancy, Arlington, VA

Dr. Philip Hopke, Bayard D. Clarkson Distinguished Professor, Department of Chemical
Engineering, Clarkson University, Potsdam, NY

Dr. George Lambert, Associate Professor of Pediatrics, Director, Center for Childhood
Neurotoxicology, Robert Wood Johnson Medical  School-UMDNJ, Belle Mead, NJ

Dr. Allan Legge, President, Biosphere Solutions,  Calgary, Alberta, CANADA

Dr. Maria Morandi, Assistant Professor, Division of Environmental and Occupational
Health, School of Public Health, University of Texas, Houston, TX

Dr. Deborah Neher, Associate Professor and Chair, Plant and Soil Science, College of
Agriculture and Life Sciences, University of Vermont, Burlington, VT

Dr. Duncan Patten, Research Professor, Land Resources and Environmental Sciences
Department, Montana State  University, Bozeman, MT, USA
                                      IV

-------
Dr. Ramesh Reddy, Graduate Research Professor and Chair, Soil and Water Science
Department, Institute of Food Agricultural Sciences, University of Florida, Gainesville,
FL

Dr. Gary Sayler, Beaman Distinguished Professor, Joint Institute for Biological
Sciences, Oak Ridge National Laboratory, University of Tennessee, Knoxville, TN

Dr. Alan Steinman, Director, Annis Water Resources Institute, Grand Valley State
University, Muskegon, MI

Dr. Chi-Yeung John Suen, Professor, Earth and Environmental Sciences, College of
Science and Mathematics, California State University, Fresno, Fresno, CA, USA

Dr. Robert Twiss, Professor Emeritus, University of California-Berkeley, Ross, C A

Dr. Judith S. Weis, Professor, Department of Biological Sciences, Rutgers University,
Newark, NJ

Dr. Barry Wilson, Professor, Animal Science and Environmental Toxicology, College
of Agriculture and Environmental Science, University of California, Davis, CA

SCIENCE ADVISORY BOARD STAFF
Dr. Thomas Armitage, Designated Federal Officer, U.S. Environmental Protection
Agency, Washington, DC

-------
                  U.S. Environmental Protection Agency
                          Science Advisory Board
CHAIR
Dr. M. Granger Morgan, Lord Chair Professor in Engineering, Department of
Engineering and Public Policy, Carnegie Mellon University, Pittsburgh, PA

SAB MEMBERS
Dr. Gregory Biddinger, Coordinator, Natural Land Management Programs, Toxicology
and Environmental Sciences, ExxonMobil Biomedical Sciences, Inc, Houston, TX

Dr. Thomas Burke, Professor, Department of Health Policy and Management, Johns
Hopkins Bloomberg School of Public Health, Johns Hopkins University, Baltimore, MD

Dr. James Bus, Director of External Technology, Toxicology and Environmental
Research and Consulting, The Dow Chemical Company, Midland, MI

Dr. Deborah Cory-Slechta, Professor, Department of Environmental Medicine, School
of Medicine and Dentistry, University of Rochester, Rochester, NY

Dr. Maureen L. Cropper, Professor, Department of Economics, University of
Maryland, College Park, MD

Dr. Virginia Dale, Corporate Fellow, Environmental Sciences Division, Oak Ridge
National Laboratory, Oak Ridge, TN

Dr. Kenneth Dickson, Regents Professor, Department of Biological Sciences, University
of North Texas, Aubrey, TX

Dr. David A. Dzombak, Walter J. Blenko Sr. Professor of Environmental Engineering,
and Faculty Director, Steinbrenner Institute for Environmental Education and Research,
Carnegie Mellon University, Pittsburgh, PA

Dr. Baruch Fischhoff, Howard Heinz University Professor, Department of Social and
Decision Sciences, Department of Engineering and Public Policy, Carnegie Mellon
University, Pittsburgh, PA

Dr. James Galloway, Professor, Department of Environmental Sciences, University of
Virginia, Charlottesville, VA

Dr. James K. Hammitt, Professor, Center for Risk Analysis, Harvard University,
Boston, MA
                                      VI

-------
Dr. Rogene Henderson, Senior Scientist Emeritus, Lovelace Respiratory Research
Institute, Albuquerque, NM

Dr. James H. Johnson, Professor and Dean, College of Engineering, Architecture &
Computer Sciences, Howard University, Washington, DC

Dr. Bernd Kahn, Director, Environmental Radiation Center, Georgia Tech Research
Institute, Georgia Institute of Technology, Atlanta, GA

Dr. Agnes Kane, Professor and Chair, Department of Pathology and Laboratory
Medicine, Brown University, Providence, RI

Dr. Meryl Karol, Professor Emerita, Graduate School of Public Health, University of
Pittsburgh, Pittsburgh, PA

Dr. Catherine Kling, Professor, Department of Economics, Iowa State University,
Ames, IA

Dr. George Lambert, Associate Professor of Pediatrics, Director, Center for Childhood
Neurotoxicology, Robert Wood Johnson Medical School-UMDNJ, Belle Mead, NJ

Dr. Jill Lipoti, Director, Division of Environmental Safety and Health, New Jersey
Department of Environmental Protection, Trenton, NJ

Dr. Michael J. McFarland, Associate Professor, Department of Civil and
Environmental Engineering, Utah State University, Logan, UT

Dr. Judith L. Meyer, Distinguished Research Professor Emeritus, Institute of Ecology,
University of Georgia, Athens, GA

Dr. Jana Milford, Associate Professor, Department of Mechanical Engineering,
University of Colorado, Boulder, CO

Dr. Rebecca Parkin, Professor and Associate Dean, Environmental and Occupational
Health, School of Public Health and Health Services, The George Washington University
Medical Center, Washington, DC

Mr. David Rejeski, Director, Foresight and Governance Project, Woodrow Wilson
International Center for Scholars, Washington, DC

Dr. Stephen M. Roberts, Professor, Department of Physiological Sciences, Director,
Center for Environmental and Human Toxicology, University of Florida, Gainesville,  FL

Dr. Joan B. Rose, Professor and Homer Nowlin Chair for Water Research, Department
of Fisheries and Wildlife, Michigan State University, East Lansing, MI
                                       vn

-------
Dr. James Sanders, Director and Professor, Skidaway Institute of Oceanography,
Savannah, GA

Dr. Jerald Schnoor, Allen S. Henry Chair Professor, Department of Civil and
Environmental Engineering, Co-Director, Center for Global and Regional Environmental
Research, University of Iowa, Iowa City, IA

Dr. Kathleen Segerson, Professor, Department of Economics, University of
Connecticut, Storrs, CT

Dr. Kristin Shrader-Frechette, O'Neil Professor of Philosophy, Department of
Biological Sciences and Philosophy Department, University of Notre Dame, Notre Dame,
IN

Dr. Philip Singer, Professor,  Department of Environmental Sciences and Engineering,
School of Public Health, University of North Carolina,  Chapel Hill, NC

Dr. V. Kerry Smith, W.P. Carey Professor of Economics , Department of Economics ,
W.P Carey School of Business , Arizona State University, Tempe, AZ

Dr. Deborah Swackhamer, Interim Director and Professor, Institute on the
Environment, University of Minnesota, St. Paul, MN

Dr. Thomas L. Theis, Director, Institute for Environmental Science and Policy,
University of Illinois at Chicago,  Chicago, IL

Dr. Valerie Thomas, Anderson Interface Associate Professor, School of Industrial and
Systems Engineering, Georgia Institute of Technology, Atlanta, GA

Dr. Barton H. (Buzz) Thompson, Jr., Robert E. Paradise Professor of Natural
Resources Law at the Stanford Law School and Director, Woods Institute for the
Environment Director, Stanford University, Stanford, CA

Dr. Robert Twiss, Professor Emeritus, University of California-Berkeley, Ross, C A

Dr. Lauren Zeise, Chief, Reproductive and Cancer Hazard Assessment Branch, Office
of Environmental Health Hazard Assessment, California Environmental Protection
Agency, Oakland, CA

SCIENCE ADVISORY BOARD STAFF
Mr. Thomas Miller, Designated Federal Officer, U.S.  Environmental Protection
Agency, Washington, DC
                                      Vlll

-------
                     TABLE OF CONTENTS

1.0   EXECUTIVE SUMMARY	x

2.0   INTRODUCTION	1

3.0   CHARGE TO THE REVIEW PANEL	2

4.0   REVIEW PROCESS	3

5.0   OVERARCHING RECOMMENDATIONS	3

6.0   AIR CHAPTER COMMENTS	13

7.0   WATER CHAPTER COMMENTS	19

8.0   LAND CHAPTER COMMENTS	29

9.0   HUMAN HEALTH CHAPTER COMMENTS	34

10.0  ECOLOGICAL CONDITION CHAPTER COMMENTS	38

11.0  REFERENCES	45

APPENDIX A:  SPECIFIC TECHNICAL COMMENTS AND CORRECTIONS	A-1

APPENDIXB: EDITORIAL COMMENTS	B-l

APPENDIX C: EXAMPLE CONCEPTUAL FRAMEWORK	C-l

APPENDIX D: EXAMPLE INDICATOR DESCRIPTION	D-l

APPENDIX E: THE USE OF ECOREGIONALLY DERIVED INDICATOR
INFORMATION FOR ACTION AND DECISION MAKING AT THE EPA
REGIONAL OFFICES	E-l

APPENDIX F:  TABLE OF RECOMMENDATIONS TO BE CONSIDERED
BEFORE FINALIZING THE 2007 REPORT	F-l

APPENDIX G:  TABLE OF RECOMMENDED IMPROVEMENTS FOR
FUTURE REPORTS ON THE ENVIRONMENT	G-l


                              ix

-------
1.0    EXECUTIVE SUMMARY

   The Science Advisory Board (SAB) Panel for the Review of EPA's 2007 Report on
the Environment has reviewed the Agency's draft Report on the Environment 2007:
Science Report (ROE 2007 Science Report or Report). The ROE 2007 Science Report
compiles and reports on scientific indicators of status and trends in human health and
ecological condition in the United States.  EPA initially presented this information in a
draft Report on the Environment Technical Document released in 2003 and reviewed by
the SAB. Since then, EPA has revised the Report in response to feedback from the SAB
and stakeholders.  The ROE 2007 Science Report will be used by EPA to inform strategic
planning, priority setting, and decision making across the Agency, and to communicate
with the general public.

   In developing the ROE 2007 Science Report, EPA compiled indicators to help answer
twenty-six policy-relevant questions deemed to be critically important to the Agency's
mission and national interest.  EPA sought the SAB's review of the adequacy of the
formulation and scope of the questions posed in the Report; the appropriateness of the
indicators in answering the questions; the accuracy of the characterization of indicator
data gaps and limitations; the degree to which the data gaps and limitations of the
indicators limit ability to answer the questions; the appropriateness of regionalization of
national indicators; the utility of regional indicators in the Report; and the overall quality
of the Report with respect to technical accuracy, clarity, and appropriateness of the level
of communication. In this advisory report, the SAB provides specific comments and
recommendations in response to the charge questions.

   The Panel emphasizes the tremendous value of the ROE 2007 Science Report and
concurs with the statement in the previous SAB review of the 2003 Report that there is an
urgent need for this kind of assessment. The Report is unique in providing a
comprehensive assessment of the time-varying quality of the environment including air,
land, and water in  relation to human and ecological health.  Such an assessment becomes
increasingly important as environmental pressures of population, urbanization, climate
change, and energy use grow. The Panel therefore strongly urges EPA to fully support
and permanently embed the Report on the Environment into its core mission-directed
activities.  This will require an investment in resources well beyond those currently
devoted to the ROE 2007 Science Report.

  The Panel finds that the ROE 2007 Science Report is a valuable collection of data and
impact indicators.  The Panel commends EPA for incorporating many of the SAB's
recommendations from the 2004 review to improve the organization and scope of the
Report.  EPA has structured the ROE 2007 around questions central to the Agency's
mission to protect  human health and ecological condition.  Generally, the SAB finds that
the questions in the Report are well developed. However, a conceptual framework
should be constructed and used as the basis for further developing appropriate questions.
In the Report, EPA has effectively identified many of the key indicator data gaps and
limitations; and regional analyses have made the Report more meaningful.  However, as
discussed below, the Panel has identified shortcomings in the document that limit its

-------
usefulness in fulfilling its stated purposes. While the Report may help inform strategic
planning and priority setting, its value is limited because the Report contains little
data interpretation and no conclusions supported by statistical analysis.
Recommendations for improvements in the Report are provided in response to EPA's
specific charge questions and to generally strengthen the document.  The Panel provides
recommendations at two levels, i.e., those to be considered before finalization of the
Report, and longer-term recommendations to be considered in subsequent reports.  These
recommendations are listed as bullets throughout this advisory report and presented in
summary tables in Appendices F and G respectively.  Additional comments and
suggestions are also provided in the text of this advisory report, and detailed comments
pertaining to specific indicators and technical issues are included in Appendix A.

Overarching "higher level" findings and recommendations that pertain to all chapters

•  The scientific underpinnings of the final Report should be strengthened to make
   it a "science report," as indicated by its title, rather than simply a data report.
   As discussed below, this can be accomplished by including greater synthesis,
   interpretation, statistical analysis, and discussion related to the literature.  An
   alternative would be to remove "science" from the title of the final Report so it is
   characterized as a status and trends report. When work is completed to strengthen the
   scientific underpinnings of future Reports, "science" could be included in the title.

•  The final Report should contain a greater degree of integrated discussion across
   the indicators and chapters.  The organization of the Report into individual media
   chapters (air, water, and land) and chapters on human health and ecosystem condition
   is consistent with EPA's programmatic configuration, and asking key scientific
   questions about the environment is a highly effective approach for organization and
   presentation. However, the Panel recommends that the final Report contain a greater
   degree of integrated discussion across the indicators and chapters than currently
   exists  A conceptual framework that illustrates the connectedness between the
   media chapters and the human health and ecological condition chapters should
   be  added to the introduction of the final Report.  In addition we recommend
   that a final synthesis chapter be added to future Reports. The synthesis chapter
   should fully integrate the entire Report and discuss health and ecosystem status,
   trends, and effects from a holistic perspective.  The synthesis chapter should
   include a discussion that interprets the observed trends, connects the trends  seen in
   the various indicators with cause/effect to the extent possible, and also connects the
   indicators with each other. EPA should add a brief section to the  final Report
   outlining how a synthesis chapter could be developed in future Reports.

•  An approach to statistical analysis should be incorporated into future Reports
   on  the Environment.  The ROE 2007 Science Report states that, due to time and
   resource limitations, statistical analyses of uncertainty and trends  in indicators were
   not included.  The Panel finds that this has limited the usefulness  of the Report, and
   that a statistical approach to analysis and presentation of the data is needed to add
   rigor to the Report.  The Panel notes that the Report provides a large amount of
                                        XI

-------
   valuable data and information that can be interpreted by readers, but it contains few
   clear conclusions and statements of significance of the findings. Future Reports on
   the environment should provide such conclusions and statements of significance.
   The Panel therefore recommends that EPA incorporate into future Reports on
   the Environment an approach to statistical analysis and reporting across all
   indicators. This should be part of the results presentation for each indicator. In
   some cases, this may involve formal statistical analyses, whereas in other cases it may
   involve the inclusion of additional information such as error bars around mean values.
   When there are insufficient data available for robust quantitative analyses, such
   statistical limitations should be reported.  Without such information, the ROE cannot
   fully meet its intended purpose of reporting scientifically established trends in human
   health and  environmental condition.

•  All questions in the final Report should address status and trends. The Panel was
   asked to comment on the adequacy of the formulation and scope of questions posed in
   the ROE 2007 Science Report. Although the scope of the questions posed in the
   Report is generally appropriate, questions are asked only about trends. Most of the
   information presented in the Report reflects indicator status rather than trends. The
   Panel recommends that all questions in the final Report address both status and
   trends. The discussions of "what the data show" should clearly reflect cases where
   trends cannot be presented because only status information is available. In addition,
   it is recommended that in the final Report, EPA should clarify whether trends are
   defined qualitatively or quantitatively. The word "trend" has a specific meaning in
   statistical science. The definition of trend can cover both statistical and qualitative
   assessment of change over time as long as the intended meaning in a particular
   situation is indicated. It is also recommended that EPA explicitly state how each
   question is related to the conceptual framework of the Report.

•  The Report on the Environment can be strengthened by selecting additional
   indicators to inform the stated questions.  The Panel was asked to comment on
   whether the indicators presented were used appropriately to answer questions
   contained in the Report.  Indicators were selected against a set of specified criteria.
   The Panel finds that, with some exceptions, appropriate indicators were selected.
   However, the rigid application of indicator selection criteria, particularly national
   representativeness, has resulted in the exclusion of valuable and relevant information.
   As discussed in various sections of this advisory report, the Report on the
   Environment can be strengthened by relaxing the restrictive indicator selection
   criteria to  enable the use of additional indicators to inform the stated questions.
   In this regard, it will be useful to select additional regional indicators and case
   studies by applying clear and transparent criteria. Indicators should also be
   included to show the response of more integrated components of the system or
   address missing issues.  For example, indicators should capture the status of and
   trends in ecosystem  services.  This will provide important information on how
   changes in environmental condition directly  affect human well being.  In addition,
   the Panel recommends  that for each indicator in the final Report, EPA provide a
   clear description of why the indicator is important, what it tells, and the
                                       xn

-------
   documented relationship between the indicator and human health and ecological
   condition.

•  Discussion of the indicator gaps, limitations, and challenges should be clarified
   in the final Report.  Each question contained in the Report is accompanied by a
   discussion of the most critical indicator gaps, limitations, and challenges that prevent
   the question from being fully answered. The Panel was asked to comment on the
   adequacy, objectiveness, and transparency with which the indicator gaps and
   limitations were characterized. In general, the Panel finds that most of the critical
   indicator data gaps and limitations have been identified.  However, to improve the
   final Report, additional clarification is required to differentiate data gaps (e.g., where
   indicator data are not available) from limitations (e.g., where limited data are
   available). In addition, various kinds of indicator data limitations could be grouped
   (e.g., geographic limitations, statistical limitations, and limitations in coverage of
   ecosystem types or components). In addition, the Panel recommends that in the final
   Report, the discussion of data gaps and limitations should be strengthened by
   including: 1) a discussion of the need for a transparent set of indicator metrics that
   can be well justified, 2) additional information, at the end of each chapter, on
   emerging issues such as chemicals of concern and invasive species, and 3) further
   justification of limitations associated with the intervals of time used to establish
   trends.  The Panel also recommends that the indicator gaps discussion in the final
   Report be expanded to identify some of the more prominent available data sets that
   were excluded,  and the reasons for their exclusion.  It is also recommended that in the
   final Report, EPA should identify the status of the monitoring programs (e.g., extant,
   on hold, or expired) that have provided indicator data used in the Report.  This will
   enable readers to determine whether additional trend information will be available in
   the future.

•  In future Reports on the Environment, regional analyses of indicator data
   should be presented by relevant geographic units such as ecoregions, airsheds,
   and watersheds.  The Report has broken out national-level data for some of the
   indicators by EPA region and provided ten regional indicators. The Panel was asked
   to comment on the utility of these approaches. The Panel finds that regional analysis
   of data makes the Report more meaningful.  The Panel also recognizes the pragmatic
   appeal of using EPA administrative regions for this  purpose.  However, the use of
   EPA administrative regions to scale national data has little ecological justification and
   does not provide particularly informative geographic descriptors of human health.
   Therefore, the Panel recommends that in future Reports on the Environment,
   indicator data  be presented by relevant geographic units such as ecoregions,
   airsheds, and watersheds. This would be a useful  approach for presenting both
   ecological and human health data.  The Panel supports the use of regional  indicators
   that can reflect important information for gauging the state of the U.S. environment.
   Key regional issues such as the ecological health of the Great Lakes or the Everglades
   should also be addressed in a national report on the environment, and the use of state
   and county data could increase the resolution for reporting the health indicators in
   future Reports.  In this regard, some states generate  and publish their own reports on
                                       Xlll

-------
    the environment. Such reports may provide useful data on indicators of human health
    as well as ecological condition.

•   More regional indicators and case studies with long-term, well-supported data
    sets could be used in future Reports on the Environment to illustrate trends
    when national data sets are not available. It is disappointing that the lack of
    available long-term data for many indicators precludes trend analysis and limits the
    usefulness of the Report   Regional data are not a substitute for national or  even
    representative national data. However, the Panel notes that with appropriate
    caveats, more regional indicators and case studies with long-term, well-supported
    data sets could be used in future Reports on the Environment to illustrate trends when
    national data sets are not available.  Some regional case studies are included in the
    Report, and it should be clearly stated that the specific case studies presented may not
    be representative of a general or national situation.  However,  this concern should not
    constrain the use of additional regional studies to demonstrate important examples of
    national importance or particular significance to local populations. In Section 5.0 of
    this advisory report, the Panel suggests criteria that might be applied to identify useful
    regional indicators and case studies.

   In addition to overarching findings and recommendations pertaining to all chapters of
the ROE 2007 Science Report, the Panel has provided specific recommendations
pertinent to individual chapters of the Report.

Air chapter findings and recommendations

   Although the questions in the air chapter of the Report are generally appropriate, a
science framework is needed to show interaction within, between  and among media,  as
well as between and among indicators.  The Panel  also notes that a short historical section
should be included in the air chapter of the final Report to provide background
information on the criteria pollutants. This information is needed to provide an
understanding of the importance of these pollutants as indicators, how they have been
tracked, and their relationship to other indicators in the Report. As discussed in Section
6.0 of this advisory report, the Panel has identified a number of missing air indicators that
should be added to the final Report because they represent important trends in air quality
or present a more holistic picture of atmospheric chemistry.  These include SO2
concentration and air toxics information.  The Report should also  discuss key trends in
the understanding of the atmosphere, such as the clear reduction of primary pollutants
(CO, SO2, and Pb) and much flatter trends in secondary pollutants (Os and
   Most of the gaps and limitations of air indicators have been appropriately identified in
the Report. However, the Panel finds that the pollutant-by-pollutant recounting approach
used in the air chapter does not show the interplay of the various criteria and toxic
pollutants with one another or the role of stratospheric ozone depletion and climate
change with respect to air quality. In addition, the pollutant information in the Report
does not demonstrate the relationship to human health.  An integrative description of
                                        xiv

-------
these air pollutants is needed in the final Report to provide public or other policy makers
with a full picture of the state of the atmospheric environment.

Water chapter findings and recommendations

   The Panel finds that some of the questions in the water chapter inappropriately call for
information on trends in both the extent and condition of certain indicators.  Therefore, it
is recommended that in the final Report, EPA refine the differentiation between extent
and condition for indicators where inclusion of both extent and condition measures does
not make sense. For example, it is not meaningful to  refer to the extent of coastal waters
because the issue of importance is  condition.  In addition, questions should be
incorporated into the water chapters of future Reports on the Environment to provide
information on critical habitats and missing thematic  elements such as trends in water
availability and usage of water for human activities.

   The Panel finds a lack of acceptable water indicators in the Report to answer some of
the questions posed.  The following additional types of indicators are recommended to
answer questions in future Reports on the Environment.  1) The freshwater indicators in
the Report have a strong lotic bias, and equal attention should be devoted to indicators
relevant to lentic systems. 2) EPA should identify and use indicators that have relevance
to human health as well as to ecology. Pathogen indicators should be included in the
future Reports.  The Panel notes that in the Report, concentrations of chemical indicators
have been inappropriately compared to drinking water maximum contaminant levels.
The comparison is inappropriate because people do not generally drink water from
agricultural streams where pesticide concentrations were measured for the comparison.
The data should be evaluated to assess toxicity to biota living in the streams. 3) EPA
should identify indicators of important ecosystem processes such as denitrification,
decomposition,  and primary production. In this regard, data on biogeochemical processes
in wetlands such as organic matter decomposition and accretion, denitrification, and
sulfate reduction can provide early indications of impending ecological changes. 4) EPA
should identify indicators that will aid in evaluating the impact of emerging issues such
as biofuel feedstock production on the quality and quantity of water.  5)  Some chemical
indicators, such as pesticides in agricultural streams, should be based on measured
concentrations in sediments and biota, if available, rather than the water column where
concentrations may be low but biota may be impacted by elevated levels in sediments.

Land chapter findings and recommendations

   The questions in the land chapter address land resource management  and land
contamination.  The questions are appropriate although somewhat peripheral to EPA's
mission. It is recommended that in future Reports on the Environment EPA consider
adding a question that addresses the important issue of soil quality and conservation. In
future Reports on  the Environment EPA should also 1) consider a range  of available land-
cover classification schemes with different levels of resolution (this is necessary because
the resolution of the data in the current draft of the Report is too coarse to completely
answer the  questions); 2) extend land cover characterization to all major ecosystem types,
                                       xv

-------
not just the forest land characterized in the current draft of the Report; and 3) adopt
standard approaches for land-use and land-cover analysis to evaluate information and
document trends across a range of available data sets.  Moreover, as further discussed in
Section 8.0 of this advisory report, the Panel finds that the questions in the land chapter
are not completely answered by the indicators presented, and the range of indicators in
the land chapter is not at the  same overall level of development as in the water and air
chapters.  For example, few land indicators provide direct measures of effects on human
health.  Some additional resources and an expanded set of disciplines are needed to bring
the land chapter to the level of evaluation provided in other chapters. To more
completely answer the questions posed in the land chapter, the Panel recommends that
EPA include the following additional indicators in the final Report: 1) a pesticide use
indicator (this would be particularly important from the standpoint of human exposure);
and 2) indicator data for persistent bioaccumlative toxics (PBTs) and mining wastes (e.g.,
Toxics Release Inventory derived information), radioactive wastes, and animal and other
wastes applied on agricultural land.

Human health chapter findings and recommendations

   The questions in the human health chapter are comprehensive, appropriate, and well-
developed. However, the Panel notes that they encompass both human health and
exposure. It is therefore recommended that in the final Report, the chapter be more
descriptively renamed "Human Exposures and Health." In addition, the questions in the
human health chapter of the final Report should be reordered to be consistent with the
event sequence in the environmental health paradigm as depicted in Figure 5.1 of the
Report. The indicators used  in the human health chapter are appropriate, but the Panel
recommends that in future Reports on the Environment, EPA consider using an expanded
suite of human health indicators that include National Health Interview Survey (NHIS)
and Behavioral Risk Factor Surveillance System (BRFSS) information (Centers for
Disease Control and Prevention, 2008a,b), hospital and emergency room admission data
available in National Hospital Discharge Survey reports (Centers for Disease Control and
Prevention, 2008c), and reports of infectious disease maintained by the Centers for
Disease Control and Prevention (Centers for Disease Control and Prevention, 2007).
These indicators would more effectively capture important health concerns such as
effects related to indoor air quality, use of pesticides, and exposure to pathogens.

   In addition, the Panel finds that there is a critical need to expand the indicator
discussion in the final Report to address indicator relevance to the stated questions.  Such
discussion is needed because the relevance of the indicators to the questions can be wide
ranging and it is important that the Report provide a characterization of the value or
importance of the indicator to the question.  Strong epidemiologic evidence is available in
the literature to support many of the indicators EPA has chosen (i.e., cancer incidence,
childhood cancer incidence, cardiovascular disease, chronic obstructive pulmonary
disease, asthma, infectious disease, birth defects, low birth weight, and preterm delivery),
and it is recommended that a qualitative or quantitative description of such information
be provided in the final Report. To further strengthen  the scientific credibility of the
Report, the Panel recommends that the discussion of indicator gaps and limitations also
                                       xvi

-------
be expanded in the final Report to include a more quantitative description of indicator
relevance by relying on the epidemiologic literature.  The discussion might be further
expanded to address how the limitations and gaps affect the interpretations of the
indicators.

Ecological condition chapter findings and recommendations

   The questions posed in the ecological condition chapter of the Report are generally
appropriate, but the Panel recommends that in the final Report the chapter be reorganized
to reflect an integrated focus on ecosystem health.  Some revision of the questions may
be needed as the chapter is reorganized. It is recommended that the chapter be organized
hierarchically according to 1)  major ecosystem type, 2) ecosystem processes and
services, and 3) ecosystem components (physical, chemical, biological). In addition, the
Panel finds that the scope of indicators in the ecological condition chapter needs
considerable broadening to cover all ecosystem types and fill specific gaps in the
indicator coverage (i.e., missing ecosystems, missing populations, and missing
processes). Specific indicators and indicator types have been suggested in Section 10.0
and Appendix A of this advisory report to broaden the coverage and fill gaps.  Easily
accessible data may be available for some of these indicators, and they could be  included
in the final Report, while others should be included in future Reports on the Environment.
It is recognized that EPA cannot develop an unlimited set of indicators but should select
those that address key ecological issues.
                                       xvn

-------
2.0    INTRODUCTION

   This report transmits the advice of the U.S. Environmental Protection Agency Science
Advisory Board (SAB) Panel for the Review of EPA's 2007 Report on the Environment.
The Panel conducted a peer review of EPA's draft Report on the Environment 2007:
Science Report (ROE 2007 Science Report or Report). The draft ROE 2007 Science
Report compiles and updates scientific indicators of status and trends in human health
and ecological condition in the United  States. EPA released its first draft Report on the
Environment in 2003. That report was reviewed by the SAB (U.S. EPA  Science
Advisory Board, 2004),  and the SAB's advice was used to develop the improved and
updated ROE 2007 Science Report. A second SAB Panel was formed to review the 2007
Report. EPA intends to use the ROE 2007 Science Report to inform strategic planning,
priority setting, and decision making across the Agency.  The ROE 2007 Science Report
is also intended to provide information that will enable the public to assess whether EPA
is succeeding in its overall mission to protect human health and the environment.

   In developing the ROE 2007 Science Report, EPA identified twenty-six policy-
relevant questions about environmental and human health deemed to be critically
important to the Agency's mission and national interest.  The Agency selected a suite of
indicators to answer these questions. The ROE 2007 Science Report consists of chapters
developed to answer status and trend questions concerning air, water, land, human health,
and ecological condition. In each of these five chapters, EPA described the scope of the
priority questions to be answered; provided a set of indicators to answer the questions;
and discussed indicator data gaps, limitations, and challenges that prevented questions
from being fully answered. In the ROE 2007 Science Report, EPA established an explicit
indicator  definition and  six indicator selection criteria. The metadata for indicators used
in the Report are available on EPA's website (U.S. Environmental Protection Agency,
2008). The metadata include EPA's evaluation of the extent to which procedures for
quality assurance and quality control of the data have been documented.  The Report
presents indicator status information to establish baselines for reporting future trends, but
it does not provide long-term trend information for many indicators. EPA stated that the
ROE 2007 Science Report was written for a target audience of environmental
professionals. The Agency developed a less detailed ROE 2007 "Highlights Document"
for the more general audience of concerned citizens, and a web-based "e-ROE" to
facilitate  electronic access to materials in the Report and provide timely updates in the
future. The SAB Panel was asked to review only the ROE 2007 Science Report.

   The Panel emphasizes the tremendous value of EPA's Report on the Environment.
This is a unique report with the objective of providing an assessment of changes in
environmental quality over time as related to human and ecological health. We concur
with the statement in the SAB's review of EPA's draft 2003 Report that there is an urgent
need for this kind of assessment. It can have an important impact on improving the state
of the environment by synthesizing relevant information from many sources for the
development of effective environmental monitoring, policy, and protection programs.
EPA's Report on the Environment can also provide the public with essential information
about environmental status and trends and their relevance to public health and ecological

-------
condition. The Panel therefore strongly urges EPA to fully support and
permanently embed the Report on the Environment in its core mission-directed
activities. This will require an investment in resources beyond those currently devoted to
the ROE 2007 Science Report. The EPA staff that produced the ROE 2007 Science
Report are commended for their remarkable productivity and output; however, a
sustained and increased investment in staff and expertise for the Report on the
Environment is essential and strongly recommended.  The Panel offers recommendations
for improvements in the ROE 2007 Science Report to make it more useful to EPA and
other intended audiences.
3.0    CHARGE TO THE REVIEW PANEL

   EPA gave the following six charge questions to the SAB Panel for its review of the
ROE 2007 Science Report.

Charge Question 1. Please comment on the adequacy of the formulation and scope of
the questions in the Chapters of the Report on the Environment 2007: Science Report.
Does the SAB have any specific recommendations on how to improve or clarify the
formulation of the questions? Does the SAB have recommendations on changing the
scope of the questions to better reflect EPA's mission?

Charge Question 2. Please comment on whether all of the relevant indicators in the
Report have been used appropriately to answer the questions. Please comment on
whether the integrity of the material in the indicator write-up is preserved in the chapter
narrative.

Charge Question 3. Please comment on the adequacy, objectivity, and transparency of
the identification and communication of gaps and limitations of the indicators in
answering the Report on the Environment questions.

Charge Question 4. Please comment on the utility,  comparability, and objectivity of the
regionalization of the national Report on the Environment indicators.  Does the use of
EPA Regions to scale national data accurately reflect, or does it inappropriately distort
the problem domain?

Charge Question 5. Please comment on the utility of the regional indicators in Report
on the Environment 2007: Science Report in answering the questions. Does the SAB
have recommendations for whether and how to build on this base in future versions of the
report?

Charge Question 6. Please comment on the overall quality of the Report on the
Environment 2007: Science Report with respect to technical accuracy, clarity, and
appropriateness of the level of communication.

-------
4.0    REVIEW PROCESS

   The Panel's review of EPA's ROE 2007 Science Report was structured to develop
responses to all of the charge questions for each chapter of the Report. Panel subgroups
were assigned lead responsibility for reviewing individual chapters of the Agency's draft
Report.  The Panel then discussed the subgroup responses and developed specific
findings and recommendations concerning the air, water, land, human health, and
ecological condition chapters.  The Panel has also provided "higher level" overarching
recommendations that pertain to all chapters of the ROE 2007 Science Report. The
overarching findings and recommendations in Section 5.0 of this advisory report address
EPA's specific charge questions as well as general improvements needed to make the
ROE 2007 Science Report a more effective assessment of status and trends in human
health and ecological condition. The Panel has recommended revisions that  should be
incorporated into the final Report as well as improvements that will require  a much
longer time frame to implement, and thus  should be incorporated in future Reports on the
Environment.  These recommendations are listed as bullets throughout this advisory
report and presented in summary tables in Appendices F and G respectively. Additional
comments and suggestions are also provided in the text of this advisory report, and
detailed comments pertaining to specific indicators and technical issues are included in
Appendix A. The Panel strongly recommends that EPA make the suggested near-term
changes prior to releasing the final Report.

5.0    OVERARCHING RECOMMENDATIONS

   The Panel finds that the ROE 2007 Science Report is a valuable collection of data and
impact indicators and strongly endorses continued development and dissemination of the
Report.  The Panel finds that the Report is an improvement over EPA's draft 2003 Report
on the Environment and commends the Agency for addressing many of the SAB's
comments and recommendations on the 2003 Report. As recommended by the 2004
SAB Review Panel, the ROE 2007  is free  from conclusions about the impacts of specific
policies or government initiatives, regional indicators have been incorporated into the
Report, and some key missing indicators have been added. The 2007 SAB Review Panel
notes, however, that some recommendations of the previous SAB review panel were not
addressed.  Additional funds and personnel have not been allocated to sustain
development of the Report on a continuing basis, and analyses of greater statistical rigor
have not been included in the Report. Generally, the formulation and scope of the
questions in the ROE 2007 Science Report are adequate, narratives in the text have
captured information about the indicators presented in the document, EPA has effectively
identified many of the key indicator data gaps and limitations, and regional analyses have
made the Report more meaningful.  However, as discussed below, the Panel has
identified numerous shortcomings in the document that limit its usefulness in fulfilling its
stated purposes. While the Report may help inform strategic planning and priority
setting, its value is limited because  the Report contains data with little interpretation and
no conclusions supported by statistical analysis.  Recommendations for improvements in
the Report are provided to make it more useful to EPA and other intended audiences.

-------
Organization of the ROE 2007 Science Report

    The organization of the Report into individual media chapters (air, water, and land)
and chapters on human health and ecological condition is consistent with EPA's
programmatic configuration, and asking key scientific questions is a highly effective
approach for presenting the information  in the Report.  However, the Panel finds that the
introduction of the Report should be revised to clearly articulate EPA's objectives in
developing the document and to more fully describe the structure of the document. As
further discussed below,  the introduction should also provide a conceptual framework
that illustrates the connectedness between the media, human health, and ecological
condition chapters. To articulate EPA's objectives and describe the structure of the
Report, the Panel recommends the following changes in the introduction:

•   In the final Report,  the introduction should be revised to clearly indicate that the
    first three chapters  are intended to address status and trends using specific
    indicators for the individual "media" of air, water, and land, and that the next
    two chapters are intended to provide integrated assessments of status and trends
    in human health and ecosystem condition.

•   In the final Report,  the introduction should also clearly state its purpose for
    intended  audiences  and EPA.  The introduction should discuss how the Agency
    plans to use the Report and its analyses and how the Agency wants the Report to
    be used by the broader public. In this regard, the final Report should state that
    the current version  of the Report provides status information to establish
    baselines for reporting future trends, but does not yet include long-term trend
    information for many indicators.

Strengthened scientific underpinnings

    The Panel finds that the scientific underpinnings of the final Report should to be
strengthened to  make it a "science report," as indicated by its title, rather than simply a
data report. As  discussed below, this can be accomplished by including greater synthesis,
interpretation, statistical  analysis, and discussion related to the literature. An alternative
would be to remove "science" from the title so that the report is characterized as a status
and trends report.  If work is completed to strengthen the scientific underpinnings of
subsequent versions of the Report, "science" could again be  included in the title.

Incorporation of a conceptual framework and synthesis chapter

   The Panel  finds that the final Report needs a greater degree of integrated discussion
across the indicators and chapters. Each chapter of the Report is currently designed to be
a stand-alone document for readers interested in the particular subject areas of land,
water, air, health, and ecology. Consequently,  the interconnections among these areas are
not well established or discussed. For example, the relationship between waste
management and chemical uses (addressed in Chapter 4) and water quality (addressed in
Chapter 3) is  mentioned  in the introduction of the water chapter, but this relationship is

-------
not obvious from the presentations of the individual indicator data. In addition, the
possible links between greenhouse gas emissions (in Chapter 2) and global sea level and
temperature changes (in Chapter 6) are not discussed. The Report currently contains a
discussion section after each question and related series of indicators, but there is not a
corresponding synthesis discussion across the questions to tie the document together and
make the whole greater than the sum of its parts.  The Panel also notes that, although the
Report provides a large amount of valuable data and information that can be interpreted
by readers, it contains few clear conclusions and statements of significance of the
findings. In future Reports on the Environment, EPA should provide such conclusions
and statements.  The Panel also urges that the following recommendations be
implemented.

•  In the final Report, EPA should incorporate a conceptual framework into the
   introduction to  illustrate the connectedness between the media, human health,
   and ecological condition chapters.  Development of the conceptual framework will
   enable selection  of better indicators by identifying the key stressors (drivers),
   responses, and outcomes that should be tracked to understand trends in the condition
   of air, water, land, human health, and ecosystems. The conceptual framework could
   be a short but comprehensive description and figure that demonstrates scientific
   understanding of relationships between the stressors, responses, and outcomes to
   human health and ecosystem condition.  An example  conceptual framework figure is
   provided in Appendix C of this advisory report. The conceptual framework should
   address relationships between source, transport, and fate of human and environmental
   health hazards, as well as exposure to receptors, dose, and impact. The description of
   the conceptual framework might discuss efforts underway to develop so-called linked
   indicators of environmental hazards and human health, such as the Environmental
   Public Health Tracking Project (National Association of County and City Health
   Officials, 2007). The  figure could be included in the introduction with appropriate
   similar figures at the beginning of each chapter to provide overall context for the
   chapter and illustrate how the chapters are connected.  For example, in each chapter
   the relevant parts of the figure that show the role and  importance of a given chapter
   could be highlighted in the graphic.  This would provide the clear basis for the use
   and prioritization of specific indicators, the choice of scale and boundaries in regional
   indicators, and selection of metrics (i.e., thresholds, benchmarks, etc.) The choice in
   scale and metrics would provide the appropriate context for future monitoring and
   assessment of status and trends.

•  In appropriate places in the final Report, interconnections between the
   indicators should be  established by cross-referencing the discussion of indicators
   in different chapters. EPA should elaborate wherever possible on the relationships
   between indicators and the outcomes with respect to human health and ecological
   condition

•  In future Reports on the Environment, a synthesis chapter should be included to
   fully integrate the Reports and to provide an overall assessment of health and
   ecosystem status, trends and effects. The synthesis chapter in future reports could

-------
   also analyze and discuss in more detail the connections among various related
   indicators as well as relationships among the media and health and ecology chapters.
   For example, the relationship between nitrogen and phosphorus in agricultural
   watersheds (in Chapter 3) and fertilizer use (in Chapter 4) could be discussed.  In this
   regard, a number of questions could be addressed, such as, Is there any evidence that
   indicators are correlated? Is it possible to use the indicator data for such an analysis?
   EPA should add a brief section to the final Report outlining how a synthesis chapter
   could be developed in future Reports.

•  In future Reports on the Environment, a summary section should be included
   after each media chapter to summarize information presented in the chapter
   and identify relevant emerging issues that could potentially affect human health
   and the environment.

Statistical analysis

   The ROE 2007 Science Report states that, due to time and resource limitations,
statistical analysis of uncertainty and trends in indicators was not included.  The Panel
finds that this has limited the usefulness of the Report, and that a statistical approach to
analysis and presentation of the data is needed.  Without such information, the Report on
the Environment cannot fully meet its intended purpose of reporting scientifically
established trends in human health and environmental condition. The Panel understands
that EPA has begun this work for some indicators and that the analysis for those
indicators  will be included in the final Report. It is our further understanding that this
work will  eventually be completed for all indicators. The  Panel understands that  some of
the most important indicators are not well developed and few high quality data sets may
be available for these indicators. The Panel suggests that these indicators could be used
with the explanation that a higher level of statistical analysis and reporting will be
developed in the future. We encourage the effort to develop statistically established
trends and recommend that:

•  In the final Report, EPA should clarify whether the document refers to
   qualitative or quantitative trends, or both (i.e., "trend" as used in the document
   needs  to be defined). The word "trend" (used in the questions) has a specific
   meaning in statistical science. The definition of trend used in the Report can cover
   both statistical and qualitative assessment of change over time, as long as the intended
   meaning in a particular situation is indicated. The Panel suggests that trend
   information be developed wherever possible, and that EPA use both qualitative as
   well as quantitative data to generate trend information for all indicators.

•  In future Reports on the Environment, EPA should incorporate statistical
   analysis and interpretation in the reporting of all indicators. This should  be part
   of the results presentation for each indicator. In some  cases, this may involve formal
   statistical analyses,  whereas in other cases it may involve the inclusion of additional
   information such as error bars around mean values.  The Panel notes that this
   approach should be developed taking into consideration the need for statistical

-------
    accuracy as well as the importance of using available information to report on
    indicators of human health and environmental condition.

Charge Question 1.  Adequacy of formulation and scope of questions in the ROE 2007
Science Report

   The Panel was asked to comment on the adequacy of the formulation and scope of
questions posed in the ROE 2007 Science Report.  The Panel notes that in the ROE
questions are asked only about trends. The scope of the questions should be broadened in
the final Report to focus on status as well as trends. This will reflect the importance of
capturing information to represent a baseline established as an initial step to evaluate
trends when more data become available. To help readers understand the importance of
the questions and associated indicators, it is also important to explain the relationship
between the questions and the conceptual framework in the final Report. The Panel
therefore recommends that:

•   In the final Report, all questions should be broadened to ask, "What are the
    status and trends..." rather than focusing only on trends. In some chapters of the
    Report, a few long-term data sets are presented. However, the information in the
    Report is focused more on status than trends.  The questions should therefore address
    both status and trends. In cases where a trend cannot be presented because only status
    information is available, this should be clearly reflected in the discussion of what the
    data show.

•   In the final Report, EPA should explicitly state how each question in the Report
    is related to a conceptual framework. The Panel recognizes that in the Report EPA
    has included "policy relevant" questions that are important to the Agency's program
    offices. However, the conceptual framework that demonstrates understanding of the
    relationships between stressors, responses, and outcomes to human health and
    ecosystem condition should be the basis for developing questions to be answered in
    the report.  Once the appropriate questions are developed, EPA can consider their
    relevance to Agency policy.

Charge Question 2.  Use of indicators to answer questions in the ROE 2007 Science
Report and presentation of indicator data in the chapter narratives

   The Panel was asked to comment on whether the indicators presented in the ROE
2007 Science Report were used appropriately to answer questions in the Report, and
whether narratives in the text accurately captured indicator information. The Panel has
responded to this question and has also identified a number of missing but appropriate
indicators  (discussed in Sections 6.0 - 10.0 and Appendix  A of this advisory report).  In
particular, as discussed in Section 10.0 below, indicators should be included in the final
Report to represent the status of and trends in ecosystem services. In addition, the Panel
recommends that the final Report contain further discussion of the relationships between
the indicators and human health and ecological condition.

-------
   EPA established a set of criteria that were used to drive the process of selecting the
indicators in the Report.  The criteria included rigorous data requirements for selection of
indicators.  The Panel finds that, with some exceptions, the narratives in the text of the
Report have accurately captured the indicator data. However, the high data standards
established by the indicator selection criteria are restrictive, and this has resulted in the
exclusion of many important indicators of status and trends in human and ecological
health. As further discussed in other sections of this advisory report, future Reports on
the Environment can be strengthened by including indicators and data sets that may not
meet the current selection criteria.  The Panel specifically recommends the following:

•  In the final Report, EPA should provide a clear description of why each
   indicator is important, the rationale for selecting the indicator, what it tells, and
   the documented relationship between the indicator and human health and
   ecological condition. An example indicator description is provided in Appendix
   C of this advisory report. For each indicator, the description could be provided in
   an introductory section that refers to the conceptual model or framework.  This is
   critical in order to enable the reader to interpret the meaning of the indicator relative
   to the question. The primary stressors (e.g., air emissions data) are important
   indicators but the Report should more fully explain how these stressors contribute to
   answering questions in the Report.

•  In the final Report, additional indicators (identified in various sections of this
   advisory report) should be included to show the response of more integrated
   components of the system or address missing issues.  For example, indicators
   should be included to capture the status of and trends in ecosystem services.
   This will provide important information on how changes in environmental condition
   directly affect human well-being. For information on this topic, EPA is referred to
   Meyerson et al., 2005.  Ecosystem services classification and indicators are further
   discussed in Section 10.0 of this advisory report.

•  In the final Report, additional trend data (classified as either qualitative or
   quantitative) should be included for as many indicators as possible. This is
   recommended as a revision for the final Report if data are available and certainly as a
   revision for future Reports on the Environment.

•  In the final Report, EPA should identify the status of the monitoring programs
   (e.g., extant, "on hold," or expired) that have provided indicator data used in the
   Report. This will enable readers to determine whether additional trend information
   will be available in the future.

•  In future Reports on the Environment, the indicators selected should be clearly
   related to the "big picture" fundamental questions, and not chosen just because
   of data availability or compliance with indicator criteria (i.e., they are the only
   indicators left after others have been eliminated).

-------
•  In future Reports on the Environment, EPA should consider relaxing the
   restrictive indicator selection criteria so that additional indicators can be
   included. The use of restrictive indicator criteria resulted in selection of indicators
   supported by nationally representative data. However, long-term data were not
   available for many of these indicators, and thus trend analysis was not possible.
   Relaxing the restrictive criteria will enable EPA to use additional indicators in order
   to better evaluate trends and answer questions in the Report.  In this regard, regional
   indicators supported by long-term data sets may be particularly useful. The Panel
   appreciates that EPA's indicator selectivity is in response to the 2004 SAB review,
   but the Panel feels the selection criteria have been made too restrictive and rigid such
   that useful data have been excluded.  One way to revise the selection criteria in order
   to identify useful regional indicators and data sets would be to classify indicators
   according to completeness or rigor. This could supplement the current approach of
   classifying the data as national or regional. For example, indicators could be
   classified as high, medium, or low  with respect to confidence in the ability to detect
   trends based on data continuity. Although this is recommended as a revision for
   future Reports on the Environment, some regional trend data may currently be
   available and easily obtained.  In these cases, revision of the final Report is
   recommended to use the available data.  On page 12 of this advisory report, the Panel
   has provided some additional guidance for selecting useful regional indicators, and a
   number of specific indicators that should be considered are identified in various
   sections of this advisory report. For example, a coral reef indicator  and National
   Oceanographic and Atmospheric Administration status and trends data could be
   included if restrictive selection criteria were relaxed.

   The Panel recognizes that it is not a simple task to change the indicator selection
   criteria to take into account the importance of additional long-term data sets and key
   indicators in various media and systems.  However, the conceptual framework of the
   Report should drive the design of criteria that will enable selection of the best
   indicators to answer questions posed in the Report.

Charge Question 3. Discussion of indicator data gaps and limitations in the ROE 2007
Science Report

   Each question in the ROE 2007 Science Report is accompanied by a discussion of the
most critical indicator gaps, limitations, and challenges that prevent the question from
being fully answered. The Panel was asked to comment on the accuracy of
characterization of the indicator gaps and limitations, and the degree to  which they limit
the ability to answer questions in the Report. In general, the Panel finds that most of the
critical indicator data gaps and limitations have been identified and clearly explained in
the Report. However, the Panel is troubled by the frequency of statements indicating that
long-term data were not available for many indicators and that this precluded trend
analysis.  The Panel appreciates this transparency but finds that there are too many
indicators in the ROE 2007 Science Report that use "snapshot" data.  The Panel
acknowledges that baseline data are essential but, as noted above, in future Reports on the

-------
Environment EPA should consider relaxing the indicator criteria, especially on a regional
basis, to allow the use of data sets that are amenable to trend analysis.

  It is somewhat problematic that many of the indicators in the Report aggregate data
over a prolonged period of time.  While this may be the result of the sampling
methodology, it should be mentioned and discussed as a weakness. For example, in the
presentation of the indicator "nitrogen and phosphorus in streams in agricultural
watersheds," the data are aggregated over nearly 10 years, but it is safe to assume that
agricultural practices and land cover in each of the watersheds have changed over that
time.  The Panel notes that these changes in agricultural practice may be a confounding
effect.

    The Panel also notes that it is not always clear which bullets in the Report refer to
"indicator limitations" or to "data gaps."  This should be clarified in the final  Report, and
in future Reports on the Environment it may be useful to subdivide the data gaps and
limitations section into different types of limitations instead of providing a laundry list
after each indicator. For example, the limitations could be grouped based on: 1)
geographic limitations; 2) statistical limitations; 3) data coverage limitations;  etc. The
following specific recommendations are provided to amplify and clarify the discussion of
indicator data gaps and limitations in the  Report:

•   In the final Report, EPA should clarify whether specific bullets in the indicator
    limitations sections refer to indicator limitations or data gaps.

•   In the final Report, the discussion of gaps and limitations should be expanded to
    identify  some of the more prominent available data sets that were excluded and
    the reasons for their exclusion (e.g., technical concerns, lack of statistical power,
    or other specific reasons). This discussion should refer to the indicator selection
    criteria and might identify indicators that could effectively narrow data gaps but may
    not meet specific stringent criteria (e.g., older data sets that can be used to show
    trends in important indicators, regional data sets that are of national interest or case
    studies demonstrating a framework for discussion or national applicability).  This
    would help address questions about some omissions, such as fish advisories issued by
    states and birth defect data.

•   In the final Report, the discussion of data gaps and limitations should be
    strengthened by adding or expanding existing information in several  areas.
    These areas include: 1) discussion of the need for a transparent set of indicator
    metrics that can be well justified (the current choices of metrics and benchmarks are
    not well justified);  2) the need to provide additional information at the end of each
    individual chapter on emerging issues such as chemicals of emerging concern, exotic
    wildlife diseases or invasive species (the Panel specifically notes that perfluorinated
    chemicals should be added to the list  of emerging contaminants of importance in
    Chapter 7 of the ROE 2007); and 3) further justification and discussion of limitations
    associated with the intervals of time used to establish trends.  To understand and
    account for such potential confounding effects, the description of each indicator
                                        10

-------
    should include a discussion of the relevant time periods that can be aggregated
    without losing integrity.

•   In the final Report, the implications of each indicator limitation should be
    discussed, and the uncertainties associated with each limitation should be
    quantified to the extent feasible. One possible approach to address uncertainty
    would be to assign a level of confidence to the inferences that can be drawn from the
    data sets. Even a subjective evaluation would provide helpful information.

•   In future Reports on the Environment, each of the sections that address data
    gaps and limitations should be separated into clear discussions of types of
    limitations (e.g., geographic, statistical, data coverage, etc.)

 Charge Question 4. Regionalization of national indicators in the ROE 2007 Science
Report

   The ROE 2007 Science Report has broken out national-level data for some of the
indicators by EPA region, and the Panel was asked to comment on the utility of this
approach. The panel notes that national-level indicators are by themselves insufficient
for gauging the state of the U.S. environment. Nationally aggregated data cannot reflect
local and regional environmental trends that are important to the quality of life and health
of the residents living in these areas. Exposures to environmental contaminants may be
relevant at three scales: national (e.g., mercury emissions), regional (e.g., contaminants in
lake fish), and local (e.g., contaminated land sites). Moreover, disasters such as
Hurricane Katrina and "9/11" taught us that while the immediate direct effects of such
events are regional or local in scale, the overall long-term effects reverberate through the
nation.  Similarly, a decline in the health of one region's environment could  affect the
entire nation. Therefore, national indicator data should be presented at the finest spatial
resolution that can be scientifically  supported. For example, it would be valuable to
examine national trends  in air quality as well as regional, state, and/or county trends.

   The disaggregation of the national indicator data in the Report by EPA administrative
regions is useful for some purposes. For example, indicator data for individual EPA
regions could be used for goal setting and performance evaluation. However, this should
be done independently from the primary environmental assessments because the use of
EPA administrative regions to scale national data has little ecological justification and
does not provide particularly informative geographic descriptors of human health.
Appendix D of this advisory report  provides further discussion of how ecoregionally
derived indicator information could be used for action and  decision making by EPA
regional offices.  The Panel finds that a preferable approach would be to analyze  the air,
water, land, human health, and ecological  condition indicators using appropriate airshed,
watershed, and ecoregional units. A useful approach to regionalization of indicators may
be to include two  subcomponents for each indicator: 1) a national metric of some kind,
with the obvious caveat that data aggregation can lead to masking of local trends; and 2)
a consistent (whenever possible)  approach to showing regional data,  preferably based on
                                        11

-------
ecologically justifiable regions, not EPA administrative regions.  The following specific
recommendations are provided regarding this approach.

•  In the final Report, if EPA administrative regions continue to be used as the
   basis for regionalizing data, the Panel recommends that this process be better
   explained. For example, it is unclear why the data are not presented consistently for
   each Region. Presenting these data consistently for each EPA Region would at least
   provide more comparability, although it will not address the bigger issue of
   ecological validity.  The strengths and limitations of using EPA administrative
   regions to regionalize data should also be discussed.

•  In future Reports on the Environment, EPA should analyze the air, water, land,
   human health, and ecological condition indicators using appropriate airshed,
   watershed, and ecoregional units. However, the appropriate scaling for indicator
   analysis and reporting should be considered on an indicator-by-indicator basis. This
   is true also for temporal  scaling issues and the appropriateness of data aggregation
   over time and space.

Charge Question 5.  Utility of regional indicators in the ROE 2007 Science Report

   EPA has included ten regional indicators in the ROE 2007 Science Report. The Panel
was asked to comment on the utility of regional indicators in answering the questions in
the Report. The Panel finds that regional indicators and case studies should be used in
future Reports on the Environment when they may be of particular value for use in trend
analysis,  or provide information that is  vital to the nation's interest (e.g., topsoil
preservation in the central Midwest). Examples will be most valuable if they can be
replicated across the U.S.  In addition, important regional issues, such as the ecological
health of the Great Lakes or the Everglades, should be addressed in a national report on
the environment. The Panel  notes, however, that the justification for the inclusion of
particular regional indicators is not clear in the current draft of the Report on the
Environment and therefore appears somewhat arbitrary. It is difficult to understand why
the current regional indicators have been chosen, as they do not appear to  provide value
for replication elsewhere.

The Panel finds that the use of regional examples is particularly useful in cases where:

       They present the successful application of an approach, model or tool that may
       have wider application. For example, the conceptual approach used for Biscayne
       Bay may have application to a wide range of problems in quite different
       environments, and the connectivity analyses done for EPA Region 4 may have
       broader applications.
   -   They serve to explain the functioning of the ecosystem and help build
       understanding of a conceptual framework of wider application. Diagrams of
       conceptual models or frameworks might be linked (especially in the web version
       of the "e-ROE") to regional examples that demonstrate processes or cause and
       effect relationships.
                                        12

-------
   -   They have wider applicability to areas within the same ecologically relevant
       region or type. Case examples can be very effective if the Report is built around
       natural systems (for example, tidal wetlands, dunes, tundra).
   -   They have long-term data sets that permit explanation of trends. This would be
       especially useful where nationwide data sets have limited time series.
       They represent an issue of national importance and deserve illumination even if
       they fail to meet the other criteria. Significance may stem from its natural
       resource value (e.g., Great Lakes), or from its importance as an emerging issue
       (nanotechnology, pharmaceuticals).
       They provide a higher resolution example of a nationwide indicator.

    The following specific recommendations are provided concerning the use of regional
data sets and indicators:

 •  In future Reports on the Environment, it is recommended that EPA identify
    and use, with appropriate caveats, more regional indicators and data bases to
    illustrate trends when national data sets are not available.  The Panel notes,
    however, that such regional data are  not a substitute for national or even
    representative national data and can be misleading if not carefully presented.
    Regional indicators should also be used in future Reports on the Environment when
    they have national  importance or are of particular significance to local populations.
    Long-term, well-supported data sets  are available for such regional indicators.
    Examples include data available from the National Science Foundation's Long-term
    Ecological Research Program sites, U.S. Geological Survey (USGS) groundwater
    basins, state agencies, and data collected on Lake Tahoe, Lake Mendota, and the
    Great Lakes.

•  In future Reports on the Environment, it is recommended that EPA develop
   clear and transparent criteria that are uniformly used for the selection of
   regional indicators and case studies, with the recognition that not all data will
   meet the criteria for these regional indicators. For example, regional indicators
   should have long-term well supported data sets, be of particular national or local
   significance, or represent an assessment approach that that could be replicated.
6.0    AIR CHAPTER COMMENTS

Charge Question 1.  Adequacy of formulation and scope of questions in the air chapter

   In general, the Panel finds that the scope of questions in the air chapter of the Report
is appropriate. However, it is problematic that the indicator data in the chapter are
presented in isolation.  A science framework consisting of a process model and
discussion is needed in the final Report to provide context for the components by
showing the interaction within, between, and among media and indicators as well  as the
effects on human health and ecosystem condition. The lack of such a framework is a
significant problem. It is critically important for EPA to understand that data presented
                                       13

-------
in isolation are not science. It is only when the data are explained as well as
appropriately interrelated across factors and chapters that one gains the scientific
understanding of what the data mean. The following recommendations are provided to
improve the formulation and scope of the questions in the air chapter.

•  In future Reports on the Environment, the discussion provided in the response
   to the indoor air quality question should be expanded. The Panel finds that the
   discussion of indoor air and related indicators is too limited considering the
   importance of the indoor environment and the amount of time spent by the population
   indoors. While indoor environments do not fall within the statutory mandate of EPA,
   exclusion of available and relevant data makes the Report incomplete. Because of the
   importance of consumer products (e.g. solvents, paints, glues, and building materials)
   as a determinant of indoor air quality and exposure, we encourage EPA to consider
   whether there are appropriate consumer product data available that satisfy the criteria
   for uses as an indicator. For example, data regarding  changes in the benzene content
   of gasoline, paints, and varnishes with time would provide a powerful indicator of
   human exposure. Similarly, data concerning changes in formaldehyde content of
   particle board and other building materials would be relevant and informative of
   indoor air quality and human exposure.

Charge Question 2. Use of indicators to answer questions in the air chapter of the ROE
2007 Science Report and presentation of indicator data in the chapter narrative

   Overall, the Panel finds that the integrity of the indicator information is maintained in
the air chapter narrative, but  as noted above, the indicators are not adequately linked to
information across the various other Report chapters.  A short historical section
containing background information on the criteria pollutants is needed in the final Report
to provide an understanding of the importance of these pollutants as indicators, how they
have been tracked, and their relationship to other indicators in the Report. Because the
Report contains no history of the air indicators, there is no indication of how long the air
monitoring networks have been in place. This knowledge would give the reader a sense
of the importance that EPA places on the air monitoring networks.  Further, it would
provide the opportunity for the reader to learn about the various types of air monitoring
networks. The air chapter then can have a discussion of questions that integrate  across
the pollutants. In addition, it is important to discuss conceptual links between trends in
climate (e.g., increased radiation from stratospheric ozone depletion) and secondary
pollutant1 problems.  There is a clear need to look at the air chapter from the whole
atmosphere perspective instead of simply isolated atmospheric components.

   The most significant shortcoming in the air chapter is the  fact that the pollutant-by-
pollutant recounting approach does not show the interplay of the various criteria and
toxic pollutants with one another or the role of stratospheric ozone depletion and climate
change on air quality.  Put another way, a holistic picture  of the chemistry of the
atmosphere is missing.  The Panel notes that substantial gains have been made in limiting
1  A "secondary pollutant" is not emitted directly. Rather, it forms in the air when directly emitted
pollutants react or interact.
                                        14

-------
the emissions of specific primary pollutants1. Additional research on the health effects of
air pollutant mixtures versus single substances is needed, however, because it has become
increasingly recognized that air pollutant mixtures are also playing an important role in
the impact of air pollutants on human health. Human activities have made the
atmosphere more oxidizing through increases in NOX emissions.  This leads to greater
ozone, more rapid conversion of SC>2 to SO/2, NOX to NCV, and biogenic and
anthropogenic volatile organic compounds (VOCs) to secondary organic aerosols. Thus,
one cannot really look at the problem of ozone and fine paniculate matter without
considering 862 and NOX emissions all together. NOX has been controlled to the point
where average ambient concentrations of NC>2 no longer violate the primary national
ambient air quality standard (NAAQS), but that approach fails to achieve control  of Os
and PM2.5 (Paniculate Matter less than 2.5 micrometers in diameter). The discussion of
VOCs in the air chapter is almost entirely focused on anthropogenic VOCs, but it is now
recognized that for many parts of the U.S., biogenic VOCs dominate and it is necessary
to think very differently about how to bring about continuing improvements in air quality.
Thus, the pollutant-by-pollutant evaluation or "stove piping" within the air chapter does
not really provide a clear picture of the current status of air quality and what should be
done in the future to continue the gains made over the past 35 years. Local sources have
been or are being controlled through either air quality state implementation plan (SIPs)
processes or maximum achievable control technology (MACT) and residual risk.  New
conceptualization of the problems is needed. Recitation of pollutant-by-pollutant gains
without a truly integrative description of their interplay fails to provide the public or other
policy makers of the full picture of the state of the atmospheric environment.

   The Panel notes that the Report contains some discussion of trends in air indicators,
but it is unfortunate that there is neither mention nor discussion of the possible direction
of trends in air indicators  10 to 20 years into the future.  Such a discussion would  provide
the reader with the rationale for the suggestion that improving air indicator trends will
continue  into the future. It should be made clear to the reader that EPA views air  quality
management as an ongoing process. The Panel also notes that EPA used data from sites
going back to 1990 to demonstrate declining trends, but it is not clear that the same data
for these  sites during the past 5 years would provide the same understanding of trends.
The issue of base year and site selection bias should be considered and a transparent
description of the  analysis should be provided.  In addition, the Panel notes that when
regional indicators are considered the picture of air quality may change.  It is important
for EPA to consider whether all of the available relevant information is being used in the
Report.

   A number of missing air indicators have been identified below and in Appendix A of
this advisory report.  These indicators should be added to the future Reports on the
Environment because they represent important trends in air quality, or present a more
holistic picture of atmospheric chemistry.  The Panel also notes that the acid deposition
discussion in the air chapter should be cross referenced and further discussed in the water
and land  chapters.  It is not unreasonable for given indicators to appear in different
1  A "primary pollutant" is directly emitted from a process, such as ash from a volcanic eruption or the
carbon monoxide gas from a motor vehicle exhaust.
                                        15

-------
chapters as long as there is a clear cross-reference across the chapters and the reason for
the cross-reference is clearly explained.

   The following specific recommendations are provided to improve the air indicators in
the final Report.

•  As stated in the overarching recommendations, a science framework should be
   incorporated into the air chapter of the final Report to show the interaction
   within, between and among media as well as between and among indicators. The
   data presented should be explained because data presented in isolation are not
   science.  In addition, the health/environmental relevance of the air indicators should
   be better documented with more extensive reference to the epidemiologic evidence as
   well as the environmental evidence.

•  In the final Report, a short historical section should be added to the air chapter
   to provide background information on the criteria pollutants.

•  In the final Report, SOi concentration should be added to the air chapter as an
   indicator. The Panel notes  that this is a "good news" story for both EPA and the
   environment. SC>2 emissions controls have resulted in significant reductions in
   ambient SO2 concentrations. This has also resulted in a reduction in the amount of
   acidic deposition attributable to 862 emissions.

•  In the final Report, the air toxics indicator should be expanded in the air
   chapter.  This is an important and rapidly  emerging human and environmental health
   issue  and it should be more  completely addressed in the Report. Currently the air
   chapter presents an air toxics emissions indicator as an aggregate of 188 compounds.
   A more informative description could be presented to provide additional information
   concerning specific toxics (see also the following recommendation concerning the
   National  Emissions Inventory). The Panel also notes a disconnect in data between
   1990  and 1999 and suggests that the Agency could look at the possibility of using
   estimates to determine trends. Trends in ambient concentrations of toxics could be
   developed by looking beyond the regional  scale to the local level where additional
   monitoring data are available. While it is true that in the current network the benzene
   data are the most robust, it should be anticipated by EPA that in the future the
   network will be more robust for additional chemicals of concern.

   Further, it is not clear in the text what the difference is between Persistent Organic
   Pollutants (POPS), Persistent Bioaccumulative and Toxic chemicals (PBTs), and
   Hazardous Atmospheric Pollutants (HAPS).  Sometimes the terms air toxics and
   HAPS are used as synonyms.  Since the ROE 2007 Science Report is to be read by
   the general public, it is essential that all of the terms used in the text be clearly and
   unambiguously defined and used  consistently.  This becomes an important integration
   issue  when chemicals and the responses to those chemicals appear in different media
   chapters.  Reference is made in the water chapter, for example, to compounds also
   found in  the air chapter but no cross-referencing is evident.
                                       16

-------
•  In the final Report, a broader explanation of what is in the National Emissions
   Inventory (NEI) should be added to the air chapter.  This is important because
   there is reference in the text to the Toxic Release Inventory (TRI) and Persistent
   Bioaccumulative and Toxic (PBT) chemicals.

•  In the final Report, further analysis of the trends in air indicators should be
   added to the air chapter. While it is important to know whether air indicator trends
   are increasing or not, it is important for the reader to understand the reason for the
   direction of indicator trends. The Report should state where have we been, where we
   are now, and where we are going. As it stands, there is no history provided on how
   the air indicators were developed or evolved, or what may have influenced a  certain
   trend (e.g., banning lead from gasoline resulted in a precipitous decline in
   atmospheric lead concentrations).

•  In the final Report, an indicator should be added to the air chapter to focus on
   the clear reduction of primary pollutants (CO, SOi, and Pb) but much flatter
   trends in secondary pollutants (Os and PM2.s), reflecting the growing importance
   of secondary air pollutants. These pollutants are becoming increasingly important
   as regulatory efforts have resulted in reductions of major primary pollutants.  Such an
   indicator would allow EPA to show the interaction of the atmospheric components
   and would help pull the pieces together conceptually. It also allows one to discuss
   more complex issues such as climate and ozone.

•  In the final Report, a small section should be added to the air chapter to  discuss
   how climate change is affecting aerosols. A paragraph would be appropriate. This
   paragraph would create the opportunities in the text to emphasize the interactions
   among pollutants, the importance of secondary pollutants, and the complexity of the
   atmospheric chemistry.

Charge Question 3.  Identification of gaps and limitations of the air chapter indicators

   Overall, the Panel finds that most of the critical gaps and limitations of air chapter
indicators have been identified. That being said, the Panel provides a number of
suggestions for informational improvements to the gaps and limitations to provide a
better understanding of the meaning and relevance of the indicators.  The Panel finds that
indicator limitations are presented in a generally pro forma and mechanical fashion.
There is virtually no discussion of whether, and how, these limitations should affect the
reader's interpretation of the estimates with regard to magnitude of point estimates or
shape of trends. With  the exception of the ambient concentration indicators for criteria
pollutants, benzene, and manganese in Region 5, quantitative estimates of uncertainty are
lacking, leaving unanswered questions concerning the robustness of the majority of the
indicators.

   The Panel also finds that in the discussion of gaps and limitations of the air indicators,
more emphasis should be placed on how limitations fit into the "big picture," or how
                                       17

-------
changes in outdoor concentrations may have increased or decreased the importance of
other contributors to exposure and health risk. For example, given what is known,
information should be provided to indicate how decreases or increases in ambient
contaminant concentrations are reflected in total exposure and human and ecosystem
health. It is important to know whether the trends in decreasing ambient concentrations
for certain contaminants are reflected to the same extent in bio-measurements (human
and other organisms) beyond Lead (Pb). These are questions that require thinking more
comprehensively than the media-by-media presentation. The Panel also notes that in the
air chapter, as well as other chapters, the final Report should offer approaches and/or
solutions to filling gaps and limitations. The following specific recommendations are
provided to improve the discussion of indicator limitations in the air chapter.

•  In the final Report, EPA should acknowledge and discuss the limitations of a
   single pollutant, local source approach to pollution control in the context of the
   marked reductions in individual pollutants documented by the indicators, and as
   exemplified by continuing challenges with regard to ozone and PM2.s. The
   significance of temporal trends viewed in the light of the importance of primary vs.
   secondary pollutants (specifically with respect to PM and ozone) should be discussed.

•  For the final Report, EPA should view the PM speciation network as the vehicle
   to provide the needed information on PM composition.

•  In the final Report, the bias that may result from the choice of base year for
   trends for a given air indicator should be discussed, as this has implications in
   the interpretation of the air indicator data.

•  In the final Report, the effects of trends in ambient concentrations of air
   pollutant indicators on exposure and dose should be discussed.

Charge Question 4. Regionalization of the national Report on the Environment
indicators in the air chapter

   The Panel finds that the concept of having "national" as well as "regional" air
indicators would be very informative if an appropriate approach were used.  The main
problem with the approach currently used in the air chapter is that the EPA regions are
artificial administrative units that do not reflect airsheds. In addition, the national air
quality data are dominated by  data from urban air quality monitoring stations.
Presentation of national data at the scale of EPA regions and subregions (e.g., states and
cities) could be extremely misleading unless the inherent limitations of the data are
clearly understood.

Charge Question 5. Utility of the regional indicators in answering the questions in the
air chapter

   The Panel finds that regional air indicators would be very useful as long as their
application has a sound scientific basis. Unfortunately, this is generally not the case in
                                       18

-------
the air chapter.  As noted above, the EPA regions do not correspond to airsheds but rather
artificial administrative units.  That being said, the Agency could get around this dilemma
by carefully defining the "region" according to an air issue.  While two examples of
regional indicators are provided in the chapter (Manganese within Region 5 and PM
along the U.S.-Mexico border), the basis for the selection of these indicators is not
evident. This illustrates the need for EPA to consider developing and providing air
indicators for 'hot-spot' locations/areas. For future Reports on the Environment, more
conceptual development is required by EPA with respect to applying regional and sub-
regional (i.e., hot spot) air indicators.

Charge  Question 6. Overall quality of the air chapter with respect to technical accuracy,
clarity, and level of communication

   The Panel finds that the air chapter fails to provide the critical links between the
observed changes in concentrations of pollutants and the understanding of the functioning
of the atmospheric environment. The air chapter benefits from a long record of
atmospheric monitoring that provides a wealth of data.  Data are an essential part of
science because they provide the basis  for developing an understanding of the sources,
processes  and fate of the measured constituents. However, the final Report should do
more than report data.  The pollutant-by-pollutant presentation does not adequately
reflect the understanding of the interrelationships among the measured species. As
mentioned above, there are key trends in the understanding of the atmosphere that should
be addressed in the final Report, such as the clear reduction of primary pollutants (CO,
SO2, lead) but much flatter trends in secondary pollutants (O3, PM2 5).  NOX has been
controlled to the point where average concentrations of NO2 do not exceed the primary
NAAQS, but concentrations permit formation of O3 and PM2 5 that lead to air quality
violations. As mentioned above, the discussion of VOCs in the air chapter is almost
entirely focused on anthropogenic VOCs.  However, it is now  recognized that for many
parts of the U.S., biogenic VOCs dominate. In addition, the relationships between
climate change and stratospheric ozone depletion, and tropospheric chemistry that
enhances key pollutants (O3 and PM2 5), provide an important link between these
currently isolated aspects of the chapter and other air pollutants which the EPA monitors.
Thus, to improve understanding of atmospheric processes and  achieve continuing
improvements in air quality, indicator data such as those currently presented in the air
chapter  should be treated as a valuable resource but not an end in themselves.  More
attention needs to be paid to the "one atmosphere" concept that EPA has been trying to
implement, and to using the data to demonstrate how they have improved our
understanding of the atmospheric system in the U.S. The "one atmosphere" concept
incorporates the dynamic complexities of emissions and their chemistry in the
atmosphere to predict air pollution and guide program implementation.
7.0    WATER CHAPTER COMMENTS

Charge Question 1.  Adequacy of formulation and scope of questions in the water chapter
                                        19

-------
   The Panel finds that the overall broadness and consistency of the questions in the
water chapter of the ROE 2007 Science Report are appropriate given EPA's mission and
the scope of the Report. However, the questions in the water chapter do not adequately
address the interconnectedness of different water systems and both land-water and air-
water interactions. The Panel also finds that additional questions are needed to
incorporate missing information on critical habitats and thematic elements.  The
following specific recommendations are provided to improve the formulation and scope
of the questions.

•  In the final Report, the questions in the water chapter should be expanded to
   focus on the interconnectedness of different systems (both within the different
   water types and across media).

•  In the final Report, additional questions should be included in the water chapter
   to incorporate missing information on availability and usage of water for human
   activities, especially with respect to both ground water and surface water
   withdrawals (see data in Roy et al., 2005 and Solley et al., 1995).

•  In the final Report, EPA should examine the relevance of measures of "Extent
   and Condition" across all aquatic ecosystem types. In this regard, the Panel finds
   that the question on the "extent" of coastal waters is not meaningful because for
   coastal waters, the issue of importance is their condition not their extent.

•  In future Reports on the Environment, additional questions should be included
   in the water chapter to incorporate missing information on critical habitats or
   thematic elements such as:

          Extent and condition of coral reefs;
          Wastewater management information (it is recommended that EPA review
          available National Pollution Discharge Elimination System data for possible
          useful indicators);
          Extent and condition of, and trends in, riparian zones and lake shoreline (i.e.,
          land-water interface, where much of the biological activity occurs), and their
          effects on human health and the environment; and
          More national indicators and analyses providing data and information on non-
          indigenous invasive species.

•  In future Reports on the Environment, some key model aquatic systems should
   be identified in several ecoregions of the U.S. and data collected from these
   systems should be mined and analyzed in the context of questions presented in
   the Report.

•  For future Reports on the Environment, EPA should examine the 2004 National
   Research Council Report on national  and global water resources and water
   infrastructure problems, and the importance of research in addressing them
   (National Research Council, 2004). In this regard, relevant questions to be
                                      20

-------
    considered include: (1) Will drinking water be safe? (2) Will there be sufficient water
    to support both the environment and future economic growth?  (3) Can effective
    water policy be made? (4) Can water quality be maintained and enhanced? (5) Will
    our water management systems adapt to climate change?  While the Panel recognizes
    that some, if not most, of these questions are outside the narrowly defined scope of
    the ROE 2007 Science Report, EPA should consider addressing these questions
    because they help place the water media chapter into the context of the rest of the
    report.

Charge Question 2. Use of indicators to answer questions in the water chapter of the
ROE 2007 Science Report and presentation of indicator data in the chapter narrative

   In general, the Panel finds that the narratives in the water chapter of the ROE 2007
Science Report have accurately captured the indicator data. However, there is a lack of
acceptable water indicators to provide answers to the questions in the chapter.  In this
regard, the following concerns are noted.

    The indicators selected to address freshwater issues are all based on streams and
    rivers. It is problematic that there is no mention of any indicators for lakes, ponds,
    and reservoirs.
-   The section in the water chapter on wetlands provides minimal analysis of available
    data. The Panel finds that addressing only loss or gain in wetland acreage as an
    indicator is not adequate. A measure of wetland quality should be incorporated into
    the final Report to improve this indicator.
-   Only total nitrogen and phosphorus are used as nutrient indicators in the water
    chapter. Other nutrient indicators mentioned below should be considered.
    The drinking water section of the water chapter needs additional critical analysis to
    consider the implications of drinking water quality to human health. For example, the
    water chapter indicator dealing with "drinking water" covers only the number of
    systems that have not reported exceedances of maximum  contaminant levels (MCLs).
    The Panel finds that it would be more informative to report this indicator in the final
    Report as the number of systems that have had exceedances, and include data on
    which contaminants were present and the  degree to which they exceeded the MCL.
    The lack of microbial indicators in the water chapter makes it difficult if not
    impossible to ascertain human health implications and  impairment of water resources
    due to fecal pathogen contamination, regulated contaminants, or EPA Contaminant
    Candidate List elements.  In the case of pathogens, this is an unfortunate void (as
    implied in the water chapter limitations and gap analysis) given that there is a non-
    ambiguous (etiological) link between pathogen exposure and disease, albeit an
    unclear dose-dependent relationship. In earlier U.S. EPA Water Quality Inventory
    Reports to Congress (U.S. EPA, 2000), pathogen  data were evaluated and used to
    classify contributions to pollution of water resources.  It was noted that pathogens
    were either the first or second primary pollutant contributing to non-attainment of
    water quality standards for estuaries, coastal shoreline, and rivers and streams. These
    data, once obtainable from the states, are apparently no longer accessible or have
    been judged statistically or probabilistically unreliable  for accurate trend analysis.
                                        21

-------
   EPA should look for ways to obtain these data again (perhaps collaboratively with
   states).
   It appears that many of the indicators used in the water chapter are composite or
   multi-metric in nature. These indicators are useful, but the Panel recommends that
   they be complemented with single metric  indicators that are easier to understand and
   require fewer caveats and assumptions.
   In the water chapter there is very limited inclusion of data on specific toxic industrial
   chemicals and contaminants, of either a regulated or an unregulated nature, for which
   EPA has statutory responsibility under the Clean Water Act. Analysis of specific
   toxic and bioaccumulating chemicals, other than pesticides, is largely confined to fish
   tissue contaminant concentration. The lack of such information for streams, rivers,
   and sediments makes it difficult to discriminate sources of contamination and
   impairment (e.g., urban/industrial vs. agricultural).
   The water chapter data on "pesticides in agricultural streams" are comprised of
   measurements of concentrations in the water only. However, the Panel notes that
   many of these chemicals are hydrophobic and are better analyzed in the sediments
   and biota rather than in the water column, where they may appear low even in
   situations where biota may be impacted by their elevated levels in the sediments.  It is
   also unclear why these concentrations were compared with EPA's MCLs for drinking
   water.  People are not generally drinking water out of agricultural streams, so the
   focus on pesticide concentrations should be their toxicity to biota living in the
   streams, not to human consumers of drinking water.
   The section of the water chapter on "coastal fish tissue contaminants" includes
   analyses of many species offish, and indicates that 22% of the sites showed high
   contamination. However, the contaminant data are pooled from many different
   species offish and shellfish from different habitats, trophic levels, and age classes.
   The Panel notes that these factors strongly influence the degree to which a particular
   species bioaccumulates various contaminants.

   The Panel suggests that in the water chapter of future Reports on the Environment it
should be possible to develop internally consistent local or regional indicators (covering
individual environmental units or ecological provinces) in those cases where data for
national indicators are not available or do not meet the criteria for inclusion in the ROE
2007 Science Report.  Indicator data from different watersheds or hydrological basins
may not be directly comparable with each other, but the local or regional sets of data can
provide meaningful temporal trends.

   The Panel also finds that the final Report should contain better justification for some
of the schemes used to grade indicators in the water chapter.  In some instances (e.g.,
trophic state of coastal waters) the grading of "high, medium and low" quality are quite
understandable. On the other hand, the low, medium, and high grading of "nitrogen and
phosphorus in wadeable streams" presented on pages 3-22 and 3-23 is confusing. It is
hard to understand why the grading is "low" when it is below the 75th percentile for the
reference.  It appears this system was used because of statistical analyses that are not
discussed in the Report. Providing only qualitative indication (such as low nitrogen,
medium nitrogen, and high nitrogen or low flow and high flow) is not adequate for those
                                        22

-------
who would like to use this report as a guide to determine the state of these systems. The
Panel suggests that it might be better to provide a range of values in the final Report for
each of these parameters presented.  The following specific recommendations are
provided to address the concerns noted above.

•  In the final Report, data for the indicator "pesticides in agricultural streams"
   should not be compared to human health benchmarks. In future Reports on the
   Environment, data should reflect pesticide toxicity to stream biota (e.g., sediment
   concentrations of pesticides could be considered).

•  In future Reports on the Environment, EPA should include appropriate
   indicators  of condition of lakes, ponds, and reservoirs.

•  In future Reports on the Environment, EPA should consider including the
   following important specific indicators:
       Snow pack (extent, condition, and volume)
       Pathogens (coliforms, enteric viruses, toxins, etc.)
       Storm water and wastewater (contaminant effects)
   -   Drinking water primary contaminants (e.g., microbial indicators  and
       pathogens: bacterial, viral or protozoan)
       Contaminants of emerging concern such as pharmaceutical and personal
       care products, perfluorinated chemicals, brominated flame retardants,
       nanoparticles, and others.

•  In future Reports on the Environment, additional wetland data should be used.
   In many areas, wetlands will indicate more efficiently the ecological integrity of the
   entire watershed than will any other portion of the landscape. New data on basic
   wetland soil, vegetation, and periphyton characteristics are now emerging in various
   ecoregions. These data can provide important information. In addition, some of the
   possible complementary or alternative wetland indicators may include
   biogeochemical processes, such as organic matter decomposition and accretion,
   denitrification, phosphorus saturation, sulfate reduction, and indices of biotic integrity
   (IBIs), which can provide early indications of impending ecological changes.

•  For future Reports on the Environment, EPA should evaluate whether nutrient
   indicators  based on bioavailable nitrogen and phosphorus or
   nitrogen:phosphorus ratios may be more useful.

•  For future Reports on the Environment, EPA should develop drinking water
   indicators  based on the available data from the Agency's own databases and the
   consumer confidence reports released to the public annually by community
   water systems.  Based on these data, EPA could formulate indicators that can
   delineate trends in drinking water quality.  The water chapter should include source
   water monitoring data in addition to treated water quality data.
                                      23

-------
•  For future Reports on the Environment, pathogen monitoring should be
   investigated as a primary indicator for water quality trends and human health
   effects across various water sources.  This recommendation would encourage more
   cooperation with states in providing data for analysis for longer term trends.

•  In future Reports on the Environment, composite or multi-metric indicators
   should be complemented with single metric indicators that are easier to
   understand and require fewer caveats and assumptions. For example, the coastal
   benthic communities indicator could be supplemented with data on the abundance of
   key reference organisms that are particularly important to ecosystem function in each
   region (i.e., keystone species), or species that have special value to the stakeholders
   of the region (e.g., manatees in Florida or Coho salmon in Pacific Northwest).

•  In future Reports on the Environment, EPA should incorporate more
   information on specific toxic industrial chemicals for which the Agency has
   statutory responsibility under the Clean Water Act.

•  In future Reports on the Environment, EPA should analyze fish tissue
   contaminant data by different species, or at least conduct separate analyses of
   fish from different trophic levels or different habitats (as was done for the "lake
   fish tissue" indicator) to see which species (e.g., piscivores) are more likely to
   have higher levels of contaminants than others.

   Additional technical comments and recommendations concerning the specific
indicators in the water chapter are provided in Appendices A and  B of this report.

Charge Question 3. Identification of gaps and limitations of the water chapter indicators

   In general, the Panel finds that EPA has effectively identified and communicated the
gaps and limitations of the indicators in answering questions posed in the water chapter
of the ROE 2007 Science Report.  However, it is disappointing that many of the
indicators used in the chapter are recent and do not include many  years of prior
monitoring to show trends, so this gap/limitation is cited frequently. This is in striking
contrast to the air chapter of the Report in which numerous graphs with downward trends
are presented showing the overall improvement in release and ambient concentrations of
various air pollutants (with the exception of greenhouse gases which are going up).  The
Panel finds it hard to understand why the data collected for the last three decades on
various water systems are not adequate to determine status of and trends in the ecological
condition of water systems.  The gaps identified in the water chapter (e.g., on page 3-40)
for freshwater systems highlight the need for more data. The Panel notes that more data
will not necessarily answer the questions presented in Report, but it may be helpful to use
additional data from well-planned and consistent monitoring of representative systems.

   In several instances the "indicator limitations" discussion in the water chapter
addresses or provides recommendations on how to interpret indicators.  In these instances
the discussion is most often focused on interpretation of indicators to show human health
                                       24

-------
effects.  The Panel finds that the discussion of how to interpret indicators or, show what
they mean, would fit better in the section of the water chapter titled, "what the data
show."  Alternatively, to address the need for cross-media linkages, it is suggested that
EPA could add a separate section titled, "what does this mean for human heath." An
example of such a limitation is on page 3-27 in the discussion of the nitrate in streams
indicator.  The text states that, "Drinking water treatment can significantly reduce
concentrations of nitrate, so the level of contaminants reported in this indicator is not
necessarily representative of exposures to people when these waters are used as public
water supplies."  The Panel  notes that this is a separate issue from the sample design and
temporal limitations of the data set, concerns that most commonly appear in the indicator
limitations list.  The interpretation statement included on page 3-27 raises important
human health questions that could well be addressed by providing additional information.
These include questions such as: How many communities rely on these streams for their
water supply? How many communities rely on the streams that had nitrates above the
MCL? How many communities treat their water for nitrate?  The Panel notes that while
treatment can reduce nitrate levels, it is often cost prohibitive and communities must find
an alternate water supply. In addition, a high percentage of residents in rural areas
depend on private water wells which have no treatment capability. Because surface water
contamination in streams often has a direct bearing on ground water quality, how are the
exposures of these people affected?  A similar issue is apparent in the limitations
discussion of the "pesticide  in streams" indicator on page 3-32. Important human health
questions that could be addressed include: How practical is it to treat a community water
supply for pesticides?  and How many communities do this?

   The Panel recognizes that the "Survey of the Nation's Lakes" will provide a valuable
database in the future for assessing conditions of ponds and reservoirs that are
representative of all lakes in the United States. However, in the interim, usable data that
already exist should not be overlooked. For example, there is a wealth of information
(and associated data) available on nutrients, especially for rivers, lakes, and coastal
waters.  The Panel recommends that staff visit (or revisit) EPA guidance manuals for
lakes, rivers, coastal waters, and wetlands for potential data sets, if they have not already
done so. In addition, long-term monitoring programs of EPA (e.g., Environmental
Monitoring and Assessment Program - EMAP) and other Federal Agencies (e.g., the U.S.
Geological Survey's National Water Quality Assessment Program, the National Oceanic
and Atmospheric Administration's Status and Trends and Mussel Watch Programs, and
the National Science Foundation's Long-term Ecological Research and Long-term
Research in Environmental Biology programs), and of states or universities should be
examined. Indicator criteria should be relaxed (within reason) to enable the use of
important trend data. It is important to be able to see the trends with appropriate caveats
about methodologies used.  This was done for the "SAV in the Chesapeake" indicator
discussed on pages 3-74 to 3-75.  In this case, data were adjusted to account for
methodological inconsistencies.  A similar approach should be adopted for other
parameters (e.g., sediment contamination, tissue contaminants, benthic communities,
etc.), if feasible. The following specific recommendations are recommended to address
indicator gaps and limitations in the water chapter.
                                        25

-------
 •  For future Reports on the Environment, EPA should visit (or revisit) the
    Agency's guidance manuals for lakes, rivers, coastal waters, and wetlands for
    potential data sets to fill identified data gaps.

 •  For future Reports on the Environment, long-term monitoring programs of
    EPA (e.g., Environmental Monitoring and Assessment Program - EMAP) and
    other Federal Agencies  (e.g., the U.S. Geological Survey's National Water
    Quality Assessment Program, and the National Oceanic and Atmospheric
    Administration's Status and Trends and Mussel Watch Programs), and of states
    or universities should be examined.  Indicator criteria should be relaxed (within
    reason) to enable use of important trend data.

Charge Question 4. Regionalization of the national Report on the Environment
indicators in the water chapter

   The Panel finds that regionalization of national indicators is an important component
of the water chapter of the ROE 2007 Science Report. However, as noted previously, the
Panel is concerned that the use of EPA administrative regions will distort true ecological
patterns or gradients.  If possible, in future Reports on the Environment the data should
be analyzed at more appropriate scales. For surface water, a more appropriate approach
may be to use watersheds or established hydrologic units that also account for altitudinal
gradients.  For groundwater, EPA should evaluate the validity of using U.S. Geological
Survey (USGS) groundwater  basins as regional units.  Contributing watersheds may  be
used as a scaling unit for estuaries.

   The Panel notes that a regional approach will also aid in evaluating indicators to be
used for various water systems during extreme events such as hurricanes, drought, and
possibly bioterrorism. As noted previously, it is important for EPA to mine existing  data
and find ways to use these data to develop indicators for different ecoregions. For
example, an enormous amount of data is collected by the five Water Management
Districts in Florida on various water systems.  Similar data sets exist for various
ecoregions. For future Reports on the Environment, these  data  can be used to identify
indicators.

Charge Question 5.  Utility of the regional indicators in answering the questions in the
water chapter

   The Panel finds that there  is considerable utility in using regional indicators to answer
questions in the water chapter of the ROE 2007 Science Report. The regional indicators
used in the water chapter answer parts of the questions to one degree or another but
certainly do not address all  aspects of the questions. The Panel suggests that additional
regional indicators could be used to answer questions in the water chapter. One indicator
used in the Report to respond to the question of the condition and extent of coastal waters
and their effects on human health and the environment is the occurrence of dinoflagellate
blooms on the west coast of Florida (e.g., Karina brevis). The Panel notes that
dinoflagellate blooms (Pfiesterid) have been strongly linked to nutrient input in the bays
                                       26

-------
of North Carolina and Virginia and could be possible regional indicators. In addition,
recurrent harmful algal blooms (HABs) of Alexandrium off the coast of New England,
brown tide (Aureococcus) in the middle Atlantic, and Pseudonitzschia off the coast of the
Pacific Northwest are being monitored, among others. The Panel questions why harmful
algal blooms in fresh waters and invasive species have not been included as indicators in
the discussion of extent and condition of fresh surface waters. The Panel notes that a
regional indicator would seem to make sense here, either based on Great Lakes or
Everglades long-term data (National Oceanic and Atmospheric Administration, 2007;
South Florida Water Management District, 2007). Occurrences of freshwater HABs such
as Microcystis could  also be used as indicators.  In future Reports on the Environment,
EPA should  consider incorporating these  and other monitored blooms into the HAB
indicator in the water chapter. In the water chapter, there are seven other indicators listed
in response to the question of the condition and extent of coastal waters and their effects
on human health and the environment. Even taken collectively, these indicators do not
answer all aspects of the question, although each indicator illuminates some facet of the
problem posed. If EPA continues to use regional indicators in answering this question in
future Reports on the Environment, it would be helpful to explicitly identify the benefits
and limitations associated with each regional indicator vis-a-vis national indicators.

   The Panel finds that for future Reports on the Environment, development of regional
indicators focusing on individual water systems would be a useful way to identify
common indicators across regions. For example, separate water systems could be divided
into groups:  lakes and reservoirs, streams and rivers, ground water aquifers, wetlands,
estuaries, and coastal waters. Indicators used in each of these groups could be evaluated
across ecoregions and climatic gradients.  Regional EPA offices, in collaboration with
USGS and state agencies in the region, could identify data sources and transform data
into useable  information for the Report on the Environment.

   The Panel notes that as indicators are developed, there  are a multitude of processes
that must be integrated, some of which can be described in deterministic/mechanistic
equations (e.g., water flux, sediment and contaminant transport) or stochastic models
(e.g., climate change). In contrast, other complex processes that affect water resources,
such as the behavior of population groups, are more difficult to incorporate into
quantitative  models.  The process of indicator development will require transdisciplinary
research and education to synergize expertise from various domains and develop holistic
approaches or models that are modular, scalable, and flexible in order to link land and
water resources to internal and external forcing functions. The following specific
recommendations are provided to strengthen the use of regional indicators in the water
chapter of the Report:

•  In future Reports on the Environment, EPA should utilize and build on existing
   databases that have been collected and existing local expertise that has been
   developed at benchmark sites in various ecoregions. Some specific examples are
   provided in the discussion above and in the following recommendations.  This effort
   should focus on addressing water quality and quantity issues that could potentially
                                       27

-------
   affect human, economic, and ecological health.  The specific proposed goals of such
   an effort should be to:

   -   Identify attributes of land and water resources that can serve as indices of
       sustainability, and develop field and laboratory methodologies to determine these
       attributes in space and time within different benchmark water systems;
       Investigate the sensitivity and dependence of basin factors to internal and external
       forcing functions such as climate change, extreme events, water law, land use
       policies, and social customs;
       Develop predictive tools that will aid in determining the interactions and linkages
       between hydrologic processes, biogeochemical processes and socio-economic
       factors; and
   -   Expand institutional collaborations through partners and maximize the utilization
       of available resources to promote interdisciplinary research and educational
       activities in benchmark water systems.

•  In future Reports on the Environment, EPA should give state data sets much
   closer scrutiny for possible inclusion. Some states have a wealth of area-specific
   data.  For example, private well testing data are available in states with a high
   proportion  of private wells,  cf the "Wellogic" system in Michigan (Michigan
   Department of Environmental Quality, 2008) and local sport fish testing in states with
   strong recreational fisheries may mesh well with the existing national indicators.
   Highlighting what some states have done might help advance interest in expanding
   the efforts to develop a national surveillance system.

•  For future Reports on the Environment, the Panel recommends that EPA
   consider the following as an example potential local/regional indicator for use in
   the water chapter. The State Water Resources Control Board of California is
   funding USGS to lead and conduct a Groundwater Ambient Monitoring and
   Assessment (G.A.M.A.) program (U.S. Geological Survey, 2008) under which
   groundwater samples from public and private water supply wells from California are
   analyzed for water quality.  The data collected will be integrated with existing water
   quality data (such as the public supply well water quality data of the California
   Department of Health Services). The monitoring program is scheduled to repeat the
   collection and analyses once every ten years and therefore it will provide the badly
   needed information for temporal trends. Although this type of data set may not be
   useful in developing a national  water quality indicator, it is nevertheless meaningful
   and very useful in answering many of the questions in the regional context.

•  In future Reports on the Environment, the Panel recommends that, in addition
   to the Gulf of Mexico and  Long Island Sound, other places where hypoxic
   conditions tend to occur and are well monitored (such as Chesapeake Bay,  the
   coastal waters off Oregon, and parts of Lake Erie) should be added to the
   hypoxia indicator.
                                       28

-------
•  For future Reports on the Environment, EPA should develop indicators for arid
   regions. In this regard the Agency should draw upon the numerous studies and data
   collection efforts conducted by various federal and state agencies in the western states
   where the climate is arid. Most areas in these states (EPA Region 9: California,
   Arizona, Nevada) can be classified as desert or semi-desert, and water resources
   issues (related to both water quality and quantity) are highly contentious.

Charge Question 6.  Overall quality of the water chapter with respect to technical
accuracy, clarity, and level of communication

   The Panel generally finds that the water chapter is technically accurate and that the
level of communication is appropriate.  As noted above, additional indicators are needed
to answer the questions in the water chapter.  The following specific recommendation is
provided to strengthen the overall quality of the water chapter and other parts of future
Reports.

•  In future Reports on the Environment, EPA should consider identifying the
   following emerging issues in the summary section of the water chapter:

   -   Effect of climate change on water quantity and quality
       Emerging pathogens associated with climate change
       Chemicals of emerging concern
   -   Nanoparticle waste products
   -   Water availability and sustainability
       Invasive species
       Algal toxins.
8.0    LAND CHAPTER COMMENTS

Charge Question 1. Adequacy of formulation and scope of questions in the land chapter

   In the land chapter of the ROE 2007 Science Report, indicators are presented to
address fundamental questions about the state of the nation's land and its effect on human
health and the environment.  The five questions in the chapter focus on trends in: the
extent of land cover, land use, wastes, chemicals used on land, and contaminated land.
The questions in the land chapter are appropriate although somewhat peripheral to the
mission of the EPA. The first two questions (addressing land cover and land use) relate
to land resource management, while the last three questions relate to land contamination.
The Panel finds that an additional question is needed to address the important issue of
soil quality and conservation. In addition, the Panel finds that, while the inclusion of the
phrase "and their effects on human health and the environment" in each question is
understandable given the mission of EPA, there are few land indicators in the Report that
directly measure effects on human health. The following specific recommendations are
provided to improve the overall formulation and scope of the questions in the land
chapter.
                                       29

-------
•  In the final Report, EPA should consider the following suggested revisions of the
   land chapter questions to improve their clarity.

       The waste deposition addressed in Question 3 (wastes) could be considered a
       "land use" issue and included as a subtopic of Question 2 (land use).  However,
       the separation of waste management is understandable, as it is recognized that the
       hazardous and solid waste management programs run by EPA are large and
       important land media activities for the agency.

   -   Waste deposition on land has impacts on groundwater that are likely of equal or
       greater significance than the direct impacts on land.  Thus, the topic encompassed
       by Question 3 has overlap with the fundamental question regarding groundwater
       in Chapter 3, and in the introduction there is a need for an explanation of
       integration among components of the Report.

       The indicators presented in relation  to Question 4 (addressing chemicals used on
       land) focus  on agriculture.  The agency may wish to explicitly identify agriculture
       as the focus of Question 4.  An alternative  would be to include agricultural land
       indicators under Question 2 (addressing land use), considering agriculture as a
       specific land use.

   -   Question 5 (addressing contaminated land) has some overlap with Questions 3
       and 4.  The  "contaminated land" issue that is addressed by Question 5 (e.g., from
       pesticide use, industrial waste disposal, etc.) can be viewed as subsidiary to
       Questions 3 and 4.  The factors distinguishing Question 5 (addressing
       contaminated land) from Questions  3 and 4 should be explained more fully.

•  In future Reports on the Environment, EPA should consider adding a
   fundamental question on soil quality and conservation to the land chapter  The
   structure of the question could be parallel to the others in the chapter. While it could
   be argued that soil quality is covered conceptually under one of the existing
   questions, it is not obvious which one, and the Panel believes that soil quality and
   conservation is  at the same level of importance as land cover, land use, etc.  A variety
   of indicators could be established in relation to this fundamental question, including
   soil properties such as ability to hold nutrients (as measured by cation exchange
   capacity [CEC] or organic matter content),  soil nutrient inventory (e.g., to assess
   loadings of nutrients and legacy phosphorus inventory), soil salinity (e.g., to assess
   long-term effects of irrigated agriculture), and others.

Charge Question 2. Use of indicators to answer questions in the land chapter of the ROE
2007 Science Report and presentation of indicator data in the chapter narrative

   The Panel finds  that the five fundamental land chapter questions are not completely
answered by the indicators presented, and in some cases are answered only in very small
part. Further, most of the indicators do not by themselves represent a direct causal
                                        30

-------
relationship to human and environmental health. However, the Panel recognizes that
presently it may not be possible to define land indicators that are directly linked to health
effects.

   The following recommendations are provided to improve the use of indicators to
answer questions in the land chapter. Additional specific technical comments concerning
the land chapter indicators are provided in Appendix A.

•  In the final Report, EPA should include more direct indicators of effects in the
   land chapter. For example, stream water quality associated with particular land uses
   could be used as an indicator. In addition, as in other chapters, a better explanation of
   the reasons for choosing the indicators used should be provided.

•  In the final Report, EPA should consider adding indicators for mining wastes as
   well as animal and other wastes applied on agricultural land. The Panel finds that
   the two waste indicators in the land chapter are appropriate, but adding these
   additional indicators would provide important information about waste on land.

•  In the final Report, EPA should add an indicator based on the generation and
   disposal of civilian radioactive waste. This will fill an important data gap. The
   Panel recognizes that some data on defense  radioactive waste may not be publicly
   available. However, it is recommended that EPA staff work with the U.S. Nuclear
   Regulatory Commission to obtain statistical information on status and trends
   concerning civilian radioactive waste generation, disposal, and management (U.S.
   Nuclear Regulatory Commission, 2007).

•  In the final Report, a pesticide use indicator should be added to the land
   chapter. This could be done by renaming the existing indicator, "fertilizer applied
   for agricultural  purposes," as "fertilizer and pesticide applied" and adjusting the type
   of data used to populate the indicator.  In this regard, one possible indicator that could
   be used is pesticide sales.  Pesticide sales  could likely be parsed into agricultural and
   residential/commercial landscape applications. The latter would provide a
   suburban/urban indicator, which is  important from the standpoint of human exposure.
   It is noted, however, that pesticide sales alone will not capture the complex
   interrelationships between agricultural chemical use and human health and
   environmental condition.  Sales of pesticides should be evaluated in the context of
   ecological and human health risks as well as benefits of pesticide use such as changes
   in crop yields (affecting land use), changes in food quantity and quality (e.g., price
   and availability of food), contamination of foods with natural toxins, changes in
   agricultural practices such as no-till farming which is dependent upon herbicide use,
   and other factors such as seasonal changes in weather which affect pesticide use.

•  In the final Report, the pesticide residues in food and reported pesticide incident
   indicators should be moved to the human health chapter. The Panel finds that the
   decline in reported pesticide incidents has a direct relationship with human health.
   However, the link between reported pesticide incidents and the human health impacts

-------
   of land management practices is tenuous. Reported pesticide incidents cover all sorts
   of uses of pesticides, and are based on calls to poison control centers. Many of these
   incidents are related to misuse of household products and activities far removed from
   land management.  Pesticide residues in food may have a direct relationship to human
   and environmental health.  However, the linkage of this indicator to land use is weak
   and this information should be included in the human health chapter.

•  In future Reports on the Environment, with respect to the land chapter
   indicators the Panel recommends that EPA should: 1) consider a range of land
   cover classification schemes with different levels of resolution. This is necessary
   because the resolution of the data in the current Report is too coarse to completely
   answer the questions; 2) characterize land cover of all major ecosystem types, not
   just the forest land cover characterized the current draft of the Report; and 3)
   adopt standard, established approaches for land use and land cover analysis to
   evaluate information and document trends across a range of available data sets.

Charge Question 3. Identification of gaps and limitations of the land chapter indicators

   The Panel finds that the  discussions of indicator information gaps and limitations in
the land chapter are objective, honest and insightful.  In many cases, these sections point
out why particular indicators do not provide the comprehensive picture that is needed or
are "not ready for prime time."  However, with respect to data gaps, much more could be
said for each question. The data gap topics chosen for discussion seem somewhat
arbitrary, though the data gaps discussions do uniformly address the lack of measures
needed to directly assess the relationship of the indicator values to human health.
Therefore the Panel recommends that:

•  In the final  Report, the discussions of the data  gaps in the land chapter should be
   modified to make it clear that the gaps mentioned are the highest priority  gaps
   determined by the agency, and that the list is not intended to be comprehensive

Charge Question 4. Regionalization of the national Report on the Environment
indicators in the land chapter

  Concerns about the use of EPA administrative regions to regionalize national data have
been noted previously. The Panel also notes that for future Reports on the Environment,
the Agency may wish to consider the utility of the land chapter for cross-media
evaluations if EPA regions  were keyed to important environmental factors.  The Panel
notes that no single regionalization approach fits all evaluation needs. In the age of
geographic information systems (GIS) there is no need to oversimplify.  Therefore, in
evaluating the condition of land, for example, EPA could select a particular level of
USGS Hydrologic Units and overlay an ecoregionalization scheme. Bailey's U.S. Forest
Service (USFS) Ecoregions of the U.S. (Bailey, 1995) or Omernik's Ecoregional schema
(Omernik, 1987) would be  appropriate because these combine soil, elevation, moisture,
vegetation, and other factors.
                                       32

-------
•  In future Reports, different types of regional groupings could be used to show
   the location and extent of features in various chapters of the ROE.  For example,
   as further discussed in Section 10 of this advisory report, ecologically relevant units
   such as watersheds, climatic provinces, and major coastal realms could be used to
   regionalize data. At the beginning of a future Report, it would be useful to discuss
   how the indicators have been regionalized (i.e., an ecologically relevant
   regionalization scheme has been selected based on a type of indicator).

Charge Question 5. Utility of the regional indicators in answering the questions in the
land chapter

   As further discussed in Appendix A of this advisory report, the Panel does not find the
one regional example in the land chapter (the Puget Sound/Georgia Basin example given
in the Land Cover subsection) to be very useful. It is sufficiently unique that it is not
seen as providing much value as a national model or case study. The Panel could not
determine why this example was included, nor was it clear how this example could be
standardized for use in other regional analyses. However the Puget Sound  case study
exemplifies how an indicator (impervious cover) in one medium (land) has clear
implications in another medium (water).  Unfortunately, there is no explicit linkage to the
water chapter of the ROE.  As discussed above, a conceptual model could  be used to
illustrate such linkages. It would also be useful to include examples from more than  one
region in the Report. Examples and case studies of significant national importance (e.g.,
from the Great Lakes region) should be given preference.

Charge Question 6. Overall quality of the land chapter with respect to technical
accuracy, clarity, and level of communication

   The Panel finds that the land chapter is generally clearly written and technically
accurate.  The data presented are interesting and will be useful for multiple purposes.
However, in most cases, the fundamental questions in the land chapter are far from
completely answered by the indicators and indicator data available, and the big picture
understanding that the public may expect is not achieved. The data gap discussions are
brief and the Panel recommends that they be reviewed and expanded where appropriate.
In addition, while the Report writers clearly made strong efforts to avoid statements
regarding influence of programs, some  such statements have made their way into the
Report and should be removed.  For example, the Report states that recycling efforts
related to municipal solid waste have increased "most likely due to the increased
awareness about the benefits of recycling and the implementation of policies by state and
local governments tying waste generation directly to the cost of waste services."

   The Panel also finds that the range of indicators in the land  chapter is not at the same
level of development as indicators in the water and air chapters. This is understandable
given that EPA does not have a land program comparable to its water and air programs.
The modest level of development of the land chapter must ultimately be addressed
through direction of additional resources and an expanded set of disciplines in the
Agency.
                                       33

-------
9.0    HUMAN HEALTH CHAPTER COMMENTS

Charge Question 1.  Adequacy of formulation and scope of questions in the human health
chapter

   The panel generally finds the questions in the human health chapter of the ROE 2007
Science Report to be comprehensive, appropriate, and well developed. There is strength
in the simplicity and clarity of the questions. However, the Panel recommends the
following specific revisions to improve the scope and clarity of the questions.

•   In the final Report, the questions in the human health chapter should be
    reordered to be consistent with event sequence in the environmental health
    paradigm  as depicted in Figure 5.1 of the Report (i.e., exposure precedes the
    health effect).

•   In the final Report, the human health chapter should  be more descriptively
    renamed as "Human Exposures and Health." This change is needed because the
    questions contained in the chapter encompass both human health and exposure. In
    addition to be being more descriptive, the inclusion of "exposure" within the chapter
    title offers  the following advantages:

       It appropriately elevates exposure assessment within the ROE as a central and
       critical domain within EPA;
    -   It is a key tenet of the ROE to link environmental change to human and ecological
       change; and
       It provides a more appropriate place to include National Health and Nutrition
       Examination Survey (NHANES) pesticide body burden measurements as well as
       the pesticide residues in food and pesticide incident information that is currently
       out of place in the land use chapter.

   There were differing opinions among panelists regarding the adequacy and scope of
the first question in the chapter, "What are the trends in health  status in the United
States?" Some panelists thought that because the environmental factors considered in the
Report play  relatively small roles in the epidemiology of major U.S.  health trends  (i.e.,
general mortality, life expectancy, and infant mortality), such broad health-related
conditions would have limited utility as environmental health indicators per se.  Others
felt that this question was appropriate in highlighting EPA's health mission.  There was
some consensus around a compromise suggestion to eliminate this question in the  final
Report but retain the content as introductory text to  the subsequent, more specific health
question, "What are the trends in human disease and conditions for which environmental
contaminants may be a risk factor, including across  population subgroups, and
geographic regions?" In contrast to the previous more general  question, there is strong
justification for the inclusion of this question in the  final Report.
                                       34

-------
Charge Question 2.  Use of indicators to answer questions in the human health chapter of
the ROE 2007 Science Report and presentation of indicator data in the chapter narrative

   The Panel finds that the indicators used in the human health chapter of the ROE 2007
Science Report are appropriate.  However, as discussed below, additional indicators are
recommended to more completely answer the questions. In addition, there is a critical
need to expand the discussion of the health indicators' relevance to the questions.  This
discussion can appropriately stem from the following indicator criterion on page 1-7 of
the Report.

       "The indicator is useful.  It answers (or makes an important contribution to
       answering) a question in the Report on the Environment."

Although there is strong epidemiologic evidence that supports the indicators chosen
(cancer incidence, childhood cancer incidence, cardiovascular disease, chronic
obstructive pulmonary disease, asthma, infectious disease, birth defects, low birth weight,
preterm delivery) the Panel finds that the Report fails to take advantage of this literature
to provide either a qualitative or quantitative description of the environmental
contribution.  For example, what is the estimated fraction of cardiovascular disease that
can be attributed to air pollution? Although the Report acknowledges that the health
questions are complex and have multiple causes, it fails to provide a quantitative or even
qualitative assessment of the relevance of the indicator to the question. This is an
important consideration in providing the reader with the necessary context for
understanding the meaningfulness of the indicator in the context of the health question.
For example, there are scientifically credible estimates for the contribution of the
environment to various cancers (Doll and Peto, 1981; Lichtenstien et al., 2000) and such
information should be provided in the Report.  There are similar estimates of air pollution
contributions to asthma and cardiovascular morbidity and mortality (U.S. Environmental
Protection Agency, 2005). The Panel therefore recommends that:

•  For the final Report, if credible quantitative impact estimates are available (e.g.,
   estimates of the mortality or morbidity impacts of particulate air pollution in selected
   locations in the U.S.), they should be included. Establishing the relevance of the
   indicator grounded in the literature will go a long way toward strengthening the
   science of the Report.

•  In future Reports on the Environment, EPA should consider using an expanded
   suite of human health indicators that would include the following:

       The National Health Interview Survey (NHIS) (Centers for Disease Control and
       Prevention, 2008a) and Behavioral Risk Factor Surveillance System (BRFSS)
       (Centers for Disease Control and Prevention, 2008b).  This is a population-base
       survey administered by states and includes the relevant domains of Secondhand
       Smoke Policy (Module 10),  Indoor Air Quality (Module 11), and the Home
       Environment (Module 12). These modules include salient indicators for indoor
       air quality: 1) the use of gas appliances; 2) use of a coal stove, fireplace, or
                                        35

-------
       kerosene heater; 3) use of pesticides; 4) whether smoking is allowed indoors at
       home and at work. Because these data are collected at a state level, there is
       sufficient resolution to the data for use as a regional as well as a national indicator
       (this recommendation also is relevant to the air chapter).
   -   Hospital and emergency room discharge data available in the National Hospital
       Discharge Survey reports (Centers for Disease Control and Prevention, 2008c).
       Reports of infectious disease maintained by CDC (Centers  for Disease Control
       and Prevention, 2007).

•  In future Reports on the Environment, EPA needs to adopt the suites of
   indicators  that other agencies have developed, but present them in relation to
   environmental factors.

Charge Question 3.  Identification of gaps and limitations of the human health chapter
indicators

   The Panel finds that the identification and communication of gaps and limitations of
the indicators in the health chapter are adequately addressed with some potential areas for
improvement.  The following recommendations are provided to improve the
identification of gaps and limitations:

•  In the final Report, the discussion of gaps and limitations should be expanded to
   include a more quantitative description of indicator relevance by relying on the
   epidemiologic literature (this is also addressed in the indicator discussion above).
   The discussion might be further expanded to address how the limitations and gaps
   affect the interpretations of the Report on the Environment indicators, or the larger
   framework of the disease state or indicator.

•  In the final Report, the concept statements in the indicator  limitations sections
   such as "the measurement of mercury or any other environmental chemical in a
   person's blood or urine does not by itself mean that the chemical has caused or
   will cause harmful  effects in that person" should be removed from each
   discussion of indicator gap and instead be placed in the conceptual framework section
   of the chapter.

Charge Questions 4 and 5. Regionalization of the national indicators and utility of the
regional indicators in answering the questions in the human health chapter

   As noted previously, the Panel finds that regional analysis will make the Report on the
Environment richer and more meaningful. Nationally aggregated data cannot reflect
local and regional environmental or health trends that are  important to the quality of life
and health of the residents living in these areas. Regional indicators as presented by EPA
administrative regions are not particularly informative geographic  descriptors of health.
The Panel notes that geographic units such as ecoregions and watersheds  are far more
useful for presenting regional information. Using such units to present health data would
be a novel approach that would set the Report on the Environment apart from the already
                                       36

-------
existing health data presentations. The finer the spatial scale of this analysis, the more
valuable it becomes.  The finest spatial resolution contained in the Report is at the EPA
regional level and trend analysis is shown simplistically as line graphs.  The Panel notes
that even for this relatively simple analysis, and certainly as the Report on the
Environment is developed to include indicators with greater spatial resolution, more
sophisticated and innovative means of analysis and presentation will be required.

   For some of the indicators, resolution can go down to the state and even the county
level (indicators derived from birth and death certificates) thus making it possible to
aggregate the data in many geographic patterns. The NHIS survey data and the Survey
Epidemiology and End Results (SEER) cancer data (National  Cancer Institute, 2008)
only have national resolution. However, state-based surveys such as the BRFSS can
provide much of the same disease prevalence data as the NHIS with resolution at the state
level.  State cancer reporting registries are available in nearly all states and, while not as
rigorous as the SEER program, provide credible cancer incidence data widely used by
states without SEER registries.  In addition, some states periodically publish cancer and
other disease reports that provide data on county and/or major metropolitan areas. It
would be helpful for EPA to provide "regional" reports that were integrative and
coherent.  The current approach does not provide much benefit.  Therefore the Panel
specifically recommends that:

•  In the final Report, EPA should build on the higher geographic resolution theme
   by presenting individual or multiple state data which could inform the gross
   national estimates presented and point toward the future. This should be done if
   possible, given the time constraints of revising this version of the ROE.

•  For future Reports  on the Environment, EPA should consider making use of
   county-level data available from the states. All of the vital statistic data presented
   and used for the EPA Regional indicators can and have been scaled to the county
   level and excellent maps have been generated and already published in books.
   Geographic differences in disease have been identified.  Virtually every  state provides
   tables  and maps of their vital statistics by county and they are used to identify local
   priorities to allocate targeted interventions and funding, yet on page 5-68 of the ROE
   2007 Science Report it is stated that "underlying data for most ROE indicators ... do
   not enable extensive analysis of disease trends within or across geographic regions."
   The Panel notes that this statement only pertains to the NHIS survey data.  Certainly
   cardiovascular disease, stroke, and chronic obstructive pulmonary disease mortality
   can be presented at the county level or certainly the state level. Data sets can be
   found  in the National Vital Statistics Reports (Centers for Disease Control and
   Prevention, 2008d).

Charge Question 6. Overall quality of the human health chapter with respect to
technical accuracy, clarity, and level of communication

   The Panel finds that the human health chapter is generally technically accurate
although limited in its assessment and synthesis.  As with the Report in general, there is a
                                        37

-------
need to further develop the chapter from its current form, which can be characterized as a
data report, to a more sophisticated scientific document that includes assessment based on
the primary literature and appropriate statistical analysis. The following specific
additional recommendations are provided to improve the overall quality of the human
health chapter.

•  In the final Report, Bullet #2 on page 5-5 should be rewritten to include
   biological agents. The following sentence should be added: "Infectious diseases
   associated with environmental exposures or conditions are also addressed."

•  In the final Report, expanded health indicator information should be provided
   for sensitive populations. Expanded information is needed because these
   populations are important in considerations of environmental health.  Although the
   current draft of the Report does discuss the importance of health indicator information
   for at-risk populations, more information should  be provided to understand trends in
   diseases such as asthma or autism in these populations.  Moreover, information is not
   provided to relate the contribution of environmental hazards to these diseases.
10.0   ECOLOGICAL CONDITION CHAPTER COMMENTS

   The ecological condition chapter of the ROE 2007 Science Report addresses an
extremely complex topic. The Panel recognizes that developing the chapter has been a
difficult task, as it covers millions of species as well as populations, biological
communities, and ecosystems, all of which interact with each other and are differentially
affected by environmental factors.  EPA is to be commended for tackling this important
task. Compiling this information and pointing out the gaps and limitations is a very
useful project for the Agency, the scientific community, and the general public.
However, the Panel finds that reorganization of the chapter is needed to reflect an
integrated focus on ecosystem health. The ecological condition chapter should be
reorganized hierarchically according to: 1) major ecosystem type, 2) ecosystem processes
and services, and 3) ecosystem components (physical, chemical, biological).  This is
discussed in more detail below.

   Structuring the chapter as recommended above will involve reorganization of material
presently covered in the chapter and the inclusion of additional indicators discussed
below. The Panel recognizes that many of the comments and recommendations provided
below in response to the specific charge questions probably cannot be addressed in the
final Report, but should be considered for future Reports on the Environment. However,
the Panel recommends that EPA complete as much of the reorganization as possible for
the final Report.  The Panel also suggests that in the final Report, the ecological condition
chapter include a synthesis of the independent indicators, and that it emphasize the
connections between ecosystems and stressors.

Charge Question 1.  Adequacy of formulation and scope of questions in the ecological
condition chapter
                                       38

-------
   In general, the Panel finds that the questions in the ecological condition chapter are
formulated appropriately, although some revision of the questions may be needed as the
chapter is reorganized as recommended below. An exception is the biomarker question
addressing the level of exposure of specific plant and animal species to different forms of
pollution and toxic chemicals. The Panel suggests that in the final Report, rather than
focusing on trends in biomarkers, the question should refer to trends in exposure and
effects of contaminants in organisms. Biomarker data are collected to analyze the trends.
In addition, the Panel notes that it is important to show the linkages between the effects
seen in the ecological condition chapter and the indicators discussed in the media
chapters. For example, EPA should strengthen the link between sea temperature and sea
level rise discussed in this chapter and greenhouse  gases in the air chapter. The Panel
therefore recommends that:

•  In the final Report, the climate indicator trends in the ecological condition
   chapter should be placed in a paleoclimatic context to distinguish between
   human induced changes and other long-term changes. References to the Report
   of the Intergovernmental Panel on Climate Change (IPCC, 2007a,b) should be
   included.

•  The Panel recommends that in the final Report,  a question should refer to trends
   in exposure and effects of contaminants in organisms rather than focusing on
   trends in biomarkers.

   Although most of the questions in the ecological condition chapter appear to be
germane, the associated indicators in the chapter seem to have been chosen because of
the availability of data, not always because of their appropriateness  to answer the
questions. In some cases there are significant gaps between the questions and the
corresponding indicators. As recommended previously for other chapters of the final
Report, EPA should provide the rationale for selection of these particular indicators. This
rationale may be that for many desired indicators of ecological condition the needed data
simply are not available. If a desired indicator has no data, the final Report should
contain a statement of the need for data.

   As further discussed below, the Panel also notes that the  scope of indicators used to
answer questions in the chapter needs considerable broadening to cover more  ecosystem
types, with the recognition that EPA cannot develop an unlimited set of indicators but
should select those that address key ecological issues. Easily accessible data may be
available for some of these indicators and could be included in the final Report,
while others will have to wait for future Reports on the Environment. A critical
issue to be considered is whether data must meet some test that many ecological studies
may not achieve. The final Report will be more useful if it includes more information,
and then discusses caveats about the methodology. Specific gaps in coverage (missing
ecosystems, missing populations, and missing processes) in the ecological condition
chapter of the ROE 2007 Science Report are  identified in Appendix A.
                                       39

-------
Charge Question 2.  Use of indicators to answer questions in the ecological condition
chapter of the ROE 2007 Science Report and presentation of indicator data in the
chapter narrative

   In reviewing indicators used in the ecological condition chapter, the Panel considered
the charge question in two parts:  "Are the current indicators appropriately used to answer
the questions?" and "Are these the correct indicators to answer the questions?"  The
Panel finds that the indicators in the ecological condition chapter provide relevant and
useful information as an initial attempt to answer the general questions posed, but many
of the indicators are not transparent.  The limited number of acceptable indicators in the
ecological condition chapter can offer only a narrow perspective or a snapshot, and many
do not show temporal trends.  They are hardly adequate.  This argues for an introductory
discussion of each indicator along with a conceptual process diagram so that the reader
can better understand the role of each indicator and its importance relative to the
questions asked.  The Panel's specific recommendations to address these concerns are as
follows:

•  In the final Report, EPA should reorganize the ecological condition chapter to
   focus on three major indicator categories: Ecosystems, Ecological Processes and
   Services, and Ecosystem Components.

•  In the final Report, appropriate indicators should be included in the ecological
   condition chapter to provide information on the ecosystem extent (e.g., land
   cover, land use, urbanization) and quality /condition (e.g., landscape integrity,
   connectedness, fragmentation, and contamination) of major ecosystem types.
   Examples of major ecosystem types  include: forests, grasslands, shrublands, arid
   lands, wetlands, farmlands, freshwater, and coastal, marine, and urban ecosystems.

•  In the final Report, indicators should be included in the ecological condition
   chapter to represent important ecosystem processes and services such as:
   provisioning (e.g., timber, fuel, minerals, and other services); regulating (e.g.,
   disease, climate, and flood processes); cultural (e.g., spiritual and aesthetic services);
   and supporting (e.g., soil formation, primary productivity, pollination,
   decomposition, disturbance, nutrient cycling, hydrological/chemical cycling, carbon
   sequestration processes, and services such as clean air, clean water, and net
   production).  These ecosystem services classifications were developed by the
   Millennium Ecosystem Assessment  (2005).  Potential indicators relevant to the
   ecosystem processes listed above include: fire frequency, floods, drought, algal
   blooms, invasive species, carbon storage,  soil salinity, nutrients, and erosion.

•  In the final Report, indicators should be included in the ecological condition
   chapter to represent physico-chemical components of ecosystems (e.g., soils,
   water, chemicals, snow pack,  and physical habitats).  Some physico-chemical
   indicators are already included in the Report (e.g.,  mean temperature and
   precipitation, sea surface  temperature, sea level, stream flows, and nitrogen and
   phosphorus discharge into rivers and streams). It is also noted that an indicator of
                                        40

-------
   light pollution is not currently included in the Report. Because artificial light may
   adversely affect some species that depend upon darkness as a physical attribute of
   habitat, it is specifically recommended that in future Reports, EPA consider including
   an indicator of light pollution. Noise pollution can also affect ecological condition.
   EPA should therefore consider including an indicator of noise pollution in future
   Reports.

•  In the final Report, indicators should be included in  the ecological condition
   chapter to represent biological components of ecosystems ranging from the
   genome to the community level of organization.  Such components include
   biodiversity, endangered species, invasive species, keystone species, and
   communities. Specific examples of biological component indicators include: the
   extent and range of communities (e.g., land cover, coastal benthic communities, and
   coral reefs) and particular taxa (e.g., birds, fish, macroinvertebrates, and submerged
   aquatic vegetation); the protection status of biological components (e.g., management
   policy and zoning information relevant to understanding status and future
   vulnerability); and threats.  The Panel finds that the current indicators in the
   ecological condition chapter have too much reliance on vertebrates and not enough
   emphasis on small organisms (e.g., microbes, invertebrates, and flora).

   In Appendix A the Panel has provided specific technical comments and suggested
improvements concerning individual indicators currently used in the ecological condition
chapter.

Charge Question 3. Identification of gaps and limitations of the ecological condition
chapter indicators

   The Panel finds that, in general, the limitations and gaps are assessed fairly and
objectively, and are presented in a clear and transparent way in the ecological condition
chapter. As in other chapters of the Report, it may be useful to subdivide this section into
different types of limitations, such as geographic limitations, statistical limitations, data
coverage limitations, etc. Limitations are often based on inadequate data or inability to
interpret data because they are "incomplete."  Often gaps or limitations are discussed
with an inadequate understanding of relationships between the indicator and the
environment. This concern can be addressed by including a conceptual model in the
chapter as recommended above.  The conceptual model should indicate how stressors
(drivers), responses and outcomes are perceived by the scientific community.  As
previously discussed, this will improve interpretation and discussion and help the reader
understand the importance of the indicators.

   As in other chapters of the Report, it is disappointing that so many of the indicator
data are recent and prior monitoring data are not available to see temporal trends. As
noted previously, there are many monitoring programs of EPA, other federal agencies,
and states that have long-term data sets.  These data sets may not be based on
probabilistic surveys and the statistical approaches that meet the indicator selection
criteria. However, they may provide good long-term data and, if appropriate, should be
                                        41

-------
incorporated into future Reports on the Environment. The sampling deficiencies
associated with the data should be discussed in the section on gaps and limitations.
Ignoring decades of prior monitoring information because methodologies were not "up
to" current standards results in the inability to see trends in many important parameters.
The Panel notes that it is important to show trends and include caveats about
methodology.  As methods, indices, and statistical design continue to improve, EPA
should not discard the present measurements in favor of the new and improved indices.
When methods are changed, there should be a time when both the old and new methods
are used in order to establish their comparability.

The Panel provides the following specific recommendations to improve the discussion of
indicator limitations in the Report.

•  In the final Report, the discussion of "trends in diversity and biological balance
   of the nation's ecological systems," (on page 6-29) should acknowledge that some
   systems inherently have  different numbers  and variety of species, making
   comparisons between these systems inappropriate.

•  In the final Report, the discussion of "fish faunal intactness," should explain why
   1970 is chosen as the reference year.

•  In the final Report, trend data should be adjusted to  account for methodological
   inconsistencies. For example, in the discussion of "SAV in the Chesapeake" which
   shows trends since 1978, the Report on the Environment states that "methods
   changed over the course of this study.  However, data have been adjusted to account
   for any methodological inconsistencies." The same should have been done with other
   parameters that are presented as a snapshot at  one time but could have been used to
   show trends with adjustment. The Panel recognizes that not all data sets will lend
   themselves to this type of adjustment, but when possible, EPA should calibrate or
   adjust data from different periods that use different methodologies to allow
   comparability over time.

•  In future Reports on the Environment, EPA should use available information
   from the Agency's water quality criteria guidance manuals. As noted in the water
   chapter discussion, EPA has previously conducted a detailed review of current
   information to develop water quality criteria guidance manuals for lakes, rivers, and
   coastal waters.  It is not clear whether this information was used in addressing some
   of the questions raised in the Report.

Charge Question 4. Regionalization of the national Report  on the Environment
indicators in the ecological condition chapter

   As discussed previously, regionalization is an important  element in the Report on the
Environment.  However, the EPA regions, while important for administrative purposes,
are not relevant for representation of regional indicators in the ecological condition
chapter. The  separation of data into the ten EPA regions may inadvertently convey
                                       42

-------
inaccurate ecological information to readers. For example, Exhibit 6-2 shows the
changes in acreage in the extent of forested land in the U.S. broken down by EPA
regions. However, the Report fails to recognize the differences in climate, biomes, and
the amount of total area among these ecologically distinct units.  The Panel finds the
basis of the division to be misleading.  Ecologically relevant units, such as watersheds,
climatic provinces, major coastal realms, forests, etc. provide a scientifically sound basis
for conceptual and statistical analyses. Results from ecoregional analysis could easily be
reported in the final Report for EPA administrative units by using current GIS
technology.  It could be mentioned early in the final Report that some indicators will be
regionalized based on the type of indicator (e.g., one that relates to large watersheds, such
as nutrient discharge to oceans; or to major climatic zones,  such as forest indicators).  In
this way objectivity of regionalization is addressed. Because there is little comparison
across indicators in the Report, comparability across regions is limited. This suggests a
future need for some kind of cross-reference table or section in the final Report that
addresses the issue of comparability of indicators, questions and regions.

Charge Question 5. Utility of the regional indicators  in answering the questions in the
ecological condition chapter

   The Panel finds that regional indicators in the ecological condition chapter have value
and should be retained but with qualifications.  Although regional examples have value
for the national report, caution should be used in applying interpretation of regional
examples on a national basis.  As discussed  above, the shortage of acceptable national
large-scale indicators can be remedied by developing regional or local indicators.
However, the justification of the  inclusion of these particular indicators in the  chapter is
not clear. The use of a region to  demonstrate some trend or change is useful if it
represents scaling of similar national data. Some of the  data sets are sufficiently
complete to support useful regional subdivision, while others are not.  Scaling decisions
should be made on an indicator-by-indicator basis. If a regional indicator has been
included in the Report only because a particular EPA region developed the methodology
and collected the data (e.g., ecological connectivity in EPA Region 4), the indicator
should be tested  in another region that is not geographically or physiognomically
equivalent. If the indicator represents  an "interesting" region (e.g., Puget Sound area)
where analysis of changes has been completed, it should be pointed out that the
uniqueness of the study may make it difficult to duplicate across the nation.

The following recommendations are provided to improve the use of regional indicators in
the ecological condition chapter.

•  In the final Report, it should be clearly stated that specific case studies in the
   Report may not be representative of a general or  national situation. These
   concerns should not constrain the use  of regional examples if developed in a fashion
   similar to other indicators with emphasis on the importance and applicability of the
   example.
                                        43

-------
•  In future Reports on the Environment, specific case studies using regional
   indicators should be selected for their ability to demonstrate the long-term
   trends that cannot be accomplished at the national level. It would be useful to
   pick well-studied sites (e.g., Lake Mendota, Lake Tahoe) where there are long-term
   data sets available for each region.

•  For future Reports on the Environment, some of the regional indicators should
   be expanded to become national indicators (e.g., SAV, invasive species, harmful
   algal blooms).

Charge Question 6.  Overall quality of the ecological condition chapter with respect to
technical accuracy, clarity, and level of communication

   As noted above, the ecological condition chapter provides relevant, accurate, and
useful information, but it is far too limited in scope.  The nation's ecosystems and key
ecosystem processes are far more extensive than represented in the Report. One problem
is the immense difference between the objectives and base questions for the Report and
the availability of applicable information to meet these objectives. The ecological
condition chapter of the final Report would benefit from improved organization, as
mentioned previously.  The general introduction of the final Report should include a
description of how all the themes are or can be integrated. To improve integration it
would be possible to take a regional approach (e.g., large watershed) and show how each
theme can be integrated within the region.  This is something that should be considered
for future Reports on the Environment. The Report also makes scaling difficult.
Regional data need to be scaleable to a larger region or nationally, and national  data need
to be scaled to regional levels for application and understanding of the data. A more
consistent and defensible approach is needed in future Reports on the Environment to
deal with regionalization of indicators.

   There is no easy way to develop ecological condition indicators, populate them with
data, and then interpret the results. One approach requires use of conceptual models that
show how indicator selection was achieved and how the indicator actually "indicates" the
consequences of changing stressors, processes and outcomes.  The authors should be
commended for their ecological condition paradigm diagram Exhibit 6-1.  A conceptual
model of flows between stressors and outcomes will look quite different from this general
interactive model but, as discussed previously, this type of diagram showing interactions
among many processes and attributes should be placed at the beginning of the document.
The ecological condition paradigm is an excellent conceptual framework, but not well
used in discussions of the indicators.  The interconnections of human  health and
ecological condition with each other and with the media chapters should be discussed and
expanded.  This approach would greatly improve the level of communication. For
example, the schematic that the SAB provided in its prior advisory report to demonstrate
interconnections should be consulted because it is still germane and would improve the
Report on the Environment. The inclusion of a statistical approach to analysis of the
data, and consistent use of metric measures would also add rigor and are needed in a
scientific document.
                                       44

-------
11.0   REFERENCES

Bailey, R.G. 1995. Description of the Ecoregions of the United States.  U.S. Department
of Agriculture Forest Service Miscellaneous Publication 1391, Washington, D.C., 108 p.

Centers for Disease Control and Prevention. 2007. Infectious Disease Surveillance.
http://www.cdc.gov/ncidod/osr/. [Accessed November 30, 2007]

Centers for Disease Control and Prevention. 2008a. Behavioral Risk Factor Surveillance
System, http://wonder.cdc.gov/wonder/sci_data/surveys/brfs/type_txt/brfs2.asp
[Accessed January 11, 2008]

Centers for Disease Control and Prevention. 2008b. National Health Interview Survey.
http://www.cdc.gov/nchs/nhis.htm [Accessed January 11, 2008]

Centers for Disease Control and Prevention. 2008c. National Hospital Discharge and
Ambulatory Surgery Data, http://www.cdc.gov/nchs/about/major/hdasd/nhds.htm
[Accessed March 24, 2008]

Centers for Disease Control and Prevention. 2008d. National Vital Statistics System.
http://www.cdc.gov/nchs/nvss.htm [Accessed March 24, 2008]

Doll, R. and R. Peto.  1981. The causes of cancer: quantitative estimates of avoidable
risks of cancer in the United States today. Journal of the National Cancer Institute,
66(6):1191-308.

IPCC.  2007a. Climate Change 2007'- The Physical Science Basis. Contribution of
Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on
Climate Change (IPCC) (WG1, AR4)

IPCC. 2007b. Climate Change 2007 - Impacts, Adaptation and Vulnerability.
Contribution of Working Group II to the Fourth Assessment Report of the
Intergovernmental Panel  on Climate Change (IPCC) (WG2, AR4)

Lichtenstein, P., U. de Faire, B. Floderus, M. Svartengren, P. Svedberg, andN. Pedersen.
2000. The Swedish Twin Registry: a unique resource for clinical, epidemiological and
genetic studies. Journal of Internal Medicine 252, 184 - 205.

Michigan Department of Environmental Quality.  2008. Wellogic System.
http://www.michigan.gov/deq/0,1607,7-135-6132_6828-16124--,00.html [Accessed
January 8, 2008]

Millennium Ecosystem Assessment. 2005. Ecosystems and Human Well Being:
Synthesis. Island Press, Washington, D.C. [Available at:
http://www.maweb.org/documents/document.356.aspx.pdf]
                                       45

-------
Meyerson, L. A., J Baron, J.M. Melillo, RJ. Naiman, R. I .O'Malley, G. Orians, M. A.
Palmer, A., S.P. Pfaff, S. W. Running, and O. E. Sala.  2005.  Aggregate measures of
ecosystem services: can we take the pulse of nature. Front. Ecol. Environ. 3(1): 56-59
[available at:
http://www.heinzctr.org/Programs/ReportingAVorking%20Groups/Ecosystem%20Servic
es/Frontiers%20In%20Ecology%20Eco%20Services%20Article.pdf]

National Association of County and City Health Officials. 2007.  Environmental Public
Health Tracking Project, http://www.naccho.org/topics/environmental/epht.cfm .
[Accessed November 30, 2007]

National Cancer Institute. 2008.  Surveillance Epidemiology and End Results.
http://seer.cancer.gov/about/ [Accessed January 11,2008]

National Oceanic and Atmospheric Administration. 2007. Harmful Algal Bloom Event
Response. http://www.glerl.noaa.gov/res/Centers/HABS/habs.html  [Accessed January 8,
2008]

National Research Council.  2004.  Confronting the Nation's Water Problems, the Role of
Research. Committee on Assessment of Water Resources Research. National Research
Council, National Academy of Sciences, National Academies Press, Washington, D.C.

Omernik, J.M.,  1987. Ecoregions of the Conterminous United States. Annals of the
Association of American Geographers., 77(1): 118-125.

Roy, S.B. et al. 2005. Evaluation of the sustainability of water withdrawals in the United
States, 1995 to 2025. J Am Water Res Assoc 41:1091-1108

Solley, W.B. et al. 1998. Estimated use of water in 1995. U.S. Geological Survey
Circular 1200, U.S. Geological Survey, Reston, VA

South Florida Water Management District. 2007.  Managing and Protecting our
Region's Water Resources.
https://my.sfwmd.gov/portal/page?_pageid=2754,19862620&_dad=portal&_schema=PO
RIAL [Accessed January 8, 2008]

U.S. Environmental Protection Agency. 2000.  National Water Quality Inventory: 1998
Report to Congress. EPA-841-R-00-001. U.S. Environmental Protection Agency,
Washington, D.C.

U.S. Environmental Protection Agency. 2005. Review of the National Ambient Air
Quality Standards for Paniculate Matter: Policy Assessment of Scientific and Technical
Information. EPA/422/R-05-005A. OAQPS  Staff Paper, Office of Air Quality Planning
and Standards, U.S. Environmental Protection Agency, Research Triangle Park, NC.
                                       46

-------
U.S. Environmental Protection Agency.  2008.  EPA Report on the Environment, A-Z
Indicators List. http://cfpub.epa.gov/eroe/index.cfm7fuseactionmst.listByAlpha
[Accessed March 10, 2008]

U. S. EPA Science Advisory Board. 2002. A Framework for Assessing and Reporting on
Ecological Condition: an SAB Report. EPA-SAB-EPEC-02-009.  U.S. EPA Science
Advisory Board, Washington, D.C.

U.S. Geological Survey. 2008. GAMA: Ground Water Ambient Monitoring and
Assessment.
http://ca.water.usgs.gov/gama/ [Accessed January 2. 2008]

U.S. Nuclear Regulatory Commission. 2007.  History and Framework of Commercial
Low-Level Radioactive Waste Management in the United States. NUREG-1853.  U.S.
Nuclear Regulatory Commission [available at:
http://hps.org/govtrelations/documents/nrc_nuregl 853 .pdf]

U. S. EPA Science Advisory Board. 2004. EPA 's Draft Report on the Environment
(ROE) 2003: An Advisory by the ROE Advisory Panel of the EPA Science Advisory
Board. EPA-SAB-05-004. U.S. Environmental Protection Agency Science Advisory
Board, Washington, D.C.
                                      47

-------
Appendix A:  Specific Technical Comments and Corrections

General
    -   In the final Report, EPA should identify, perhaps by using a letter (e.g., "H"),
       those indicators that explicitly relate to human health. Such identification will
       help differentiate those indicators from others that relate more to ecological
       health. For example, indicators presented on pages 3-32 (pesticides in
       agricultural watersheds), 3-44 (nitrate and pesticides in shallow ground water in
       agricultural watersheds), 3-90 (population served by community water systems
       with no reported violations of health-based standards), and 3-103 (coastal fish
       tissue contaminants) include a health component because of health based
       standards.
    -   In some cases, data are presented for each of the 9 ecoregions (e.g., Exhibit 3-3).
       The type  of panel in Exhibit 3-3, showing a map of the 9 ecoregions, with stacked
       bars emanating from each region, should be more frequently used in the Report
       because it is much more informative than aggregated data. This is recommended
       as a revision for future Reports on the Environment.

Air chapter

Ambient Concentrations of Lead (Figure B on page 2-22)
       The caveat regarding the lead trend sites above the NAAQS is not really useful
       since the  last year with a concentration above the NAAQS was 1982.  The
       description in the narrative is sufficient for describing this event and the Figure is
       not necessary. It is not clear if excluded sites could be useful for establishing
       trends in  more recent years.  In the final Report, perhaps the X  axis in Figure B
       could be modified in a manner similar to the NOX or other NAAQS emission
       trends as  presented in Figure A on page 2-24 or 862 on page 2-54.

NAAQS for PM2.5
    -   Exhibit 2-26 on page 2-48 of the ROE indicates that the 24-hour PM2.5 NAAQS is
       65 ug/m3. The NAAQS is now set at 35 ug/m3.

Particulate Matter (PM)
       With regard to PM, a significant fraction of the 2.5 size range results from
       secondary formation.  The Report on the Environment does not mention the
       potential  use of the PM  characterization sites (which also provide data on
       elemental composition and elemental carbon/organic carbon [EC/OC]).  These
       sites have been operational for several years and it would be worth considering in
       future Reports on the Environment.  At a minimum, there is probably enough data
       to provide a snapshot of regional differences in broad ranges of composition
       (North American Consortium for Atmospheric research in Support of Air Quality
       Management - NARSTO data, for example).

Acid Deposition Data
                                       A-l

-------
   -   The presentation of acid deposition data is visually attractive but the format is
       discordant with how data are presented for other pollutants (i.e., charts). The
       graphical format on pages 2-59 and 2-60 is difficult to follow in its detail as
       compared with the chart (bar graph, pie chart). Thus it would seem for added
       clarity in understanding national and regional trends, in the final Report it would
       be advisable to follow the format used for the other pollutants in the air chapter.

Water Chapter

Presentation of Data
       On rare occasions, the discussion text in the water chapter is not consistent with
       the data being presented. For example, in the discussion of "coastal benthic
       communities" on pages 3-71 to 3-73 it is shown that 17% of area of all the U.S.
       coastal waters have low index values, and that 27% of the area in U.S. EPA
       Region 3 has a low index value. The Panel notes that a substantial portion the
       area of U.S. coastal waters (20 to 25% of the area) has a low index, and in
       addition there are extensive areas with "moderate" rather than "high" condition.
       However, in the discussion on page 3-85 of the water chapter, the Report states
       that, "Benthic communities in the nation's estuaries are largely intact in terms of
       species diversity... which is critical because these organisms are a fundamental
       link in the coastal food web." While the second part of the sentence is true, the
       first part is the statement is not supported by the data. The Panel recommends
       that in the final Report this misrepresentation  of the data be corrected.

High and Low Stream Flows
       "High and low stream flows" is not an accurate characterization of this indicator.
       The data also address timing, but this is not intuitive from this heading. A more
       accurate title is recommended for the final Report.
       Page 3-15 lines 5-11: The text is confusing in this section. First,  the word
       "substantially" is vague, perhaps intentionally, and lacks rigor. Second, what
       does substantially "larger low flows" mean?  Is this an increase in volume for low
       flows?  A greater number of streams experiencing low flows? Or  does it mean
       something else? This should be clarified in the final Report.
       3-15/24-32: It might be instructive to know if the change in timing showed any
       type of pattern.  Was there more often a delay or an acceleration, or was there no
       distinct pattern? Distinct patterns may be useful to identify,  as they may be
       related to withdrawal patterns or climate change influences.
       Since this indicator comes directly from the Heinz Center Report, EPA should
       look into how Heinz has modified its data.

Nitrogen and Phosphorus in Wadeable Streams
       3-22/1: This indicator should be labeled Total N and P, not just N and P, to be
       accurate.
   -   Although there are geographic limits to the data, much of the land cover that
       sends waters to the oceans is covered.  Limitations on what is not included are
       explained. It might be mentioned in the final Report that many if not most of the
                                       A-2

-------
       rivers in the Southwest do not discharge into the ocean, or if they do, much of the
       river has already been diverted for other purposes.

Nitrogen and Phosphorus in Agricultural Watersheds
       3-25/10: The important point is not that nitrite and ammonium are notpresent—
       rather, they are present but in low concentrations.
       3-25/20: Clarify in the final Report that it is the decomposition of the excess algae
       that can deplete oxygen in water. Also, include internal P loading from sediments
       as a potential P source, especially in shallow lakes.
       3-26/3: The low range for phosphorus is still quite high, and indicates eutrophic
       conditions for most systems.  It is unclear why such a high threshold was chosen
       for the low end of this indicator. It is not surprising that such a high percentage
       fell into this low category, but its significance is debatable.  Clarify this in the
       final Report.
       3-26/10: Flow-weighting makes considerable sense, given the aggregation of data.
       However, it would be very instructive if the data were analyzed for base flow and
       storm event periods, assuming the  data set allows this type of analysis.

Nitrogen and Phosphorus Discharge from Large Rivers
   -   3-28/1: In the final Report replace  "Discharge" with "Load"
   -   Exhibit 3-9: As noted in the text, load is a function of both discharge and
       concentration—in the final Report it would be instructive to have discharge data
       also included in this figure, to see how much of the change in load is a function of
       discharge vs. concentration. While both drive load, changes driven by the former
       are more climate related, while changes driven by the latter are more a function of
       land use practices, and therefore more related to human activities.  This is an
       important distinction.
   -   In future Reports on the Environment statistical analysis (trend analysis) is
       recommended for these data to determine if these trends are significant or not.

Pesticides in Streams in Ag Watersheds
   -   Exhibit 3-11: There is considerable value in disaggregating the data into at least a
       few key pesticides. In the present format, there may be considerable
       improvement or declines in a key pesticide, but the trend would be masked.
       While there is presentation value in aggregated data, it also can lead to
       misinterpretations—if the aggregated data need to be retained, they should be
       enhanced in the final  Report by adding trends on a few key pesticides.
       3-33/25: Include a map of the watersheds in the final Report to show explicit
       geographic distribution of the data collection.

Benthic Macroinvertebrates in Wadeable Streams
       3-36: The explanation of the O/E model will be difficult for many readers to
       follow.  This may be  an acceptable limitation, especially if the intended audience
       of the Report on the Environment is scientists, but others will balk at the non-
       intuitive narrative.
                                       A-3

-------
       What is the justification for the inclusion of these 3 ecoregions? Why not more,
       less, others?
       Exhibit 3-13: Why are the O/E data not shown in a geographic context as well?
   -   Data from the wadeable stream sampling are suspect, not because of the sampling
       technique but rather the timing.  Samples are taken from April to November and
       then treated equally.  Certainly streams change considerably over this time, both
       in physical and chemical characteristics.

3.2.3 Discussion
       3-39/30: These indicators do not reveal the role of precipitation—the load
       indicators don't provide any precipitation information, and in fact, as currently
       presented, mask the role of precipitation.
   -   3-39/33: This should be rephrased in the final Report—the chemical and physical
       indicators are proxies, at best, for the biological condition of the fresh surface
       waters. The Report on the Environment provides a very limited picture, not a
       mixed picture, of biological conditions, simply because there are so few
       biological indicators to this point.  The only trophic level discussed for surface
       fresh waters in the entire U.S. is benthic invertebrates—nothing about bacteria,
       algae, macrophytes, fish, or waterfowl. Hence, it is misleading to state the
       biological condition index is mixed—there simply are insufficient data to draw
       any conclusions about the overall state of the nation's surface fresh water biology.
       The final Report should include something on waterborne pathogens in this
       section; even if it is not an indicator, perhaps there can be cross-references to
       other sections where this indicator is discussed.

Nitrate and Pesticides in Shallow Ground Water in Ag Watersheds
       The indicator "nitrate and pesticides in shallow groundwater in agricultural
       watersheds" does not match well with the question it addresses in the Report
       because as it fails to inform on "extent."  The Panel therefore  suggests that it may
       be appropriate to restrict the question in the final Report to just "condition of
       groundwater"
       See comments on the pesticides in streams  indicator—they apply here, as well.

3.3.3 Discussion
       There is a dire need for a national monitoring program to address groundwater
       extent; this is within the domain of the USGS and hopefully funding can be
       obtained to start this work. In the interim, why not use groundwater contribution
       to stream base flow as a measure?  Gauging stage data from appropriate streams
       across the nation might serve this purpose.
   -   3-48/15: Changes in water table elevations  are available in many groundwater
       basins. For example, California Department of Water Resources maintains the
       water table data.  Over-drafting groundwater resources is a major concern and
       pressing environmental issue in the central  and western states. Regional
       indicators should be developed to address the question of "extent."

Wetland Extent, Change, and Sources of Change
                                       A-4

-------
   -   3-54/19: An important limitation of these data, which is not stated, is that extent
       does not equate to quality—the increase in freshwater ponds results in a very
       different quality of habitat than an increase in wetlands.
   -   If possible, the data should also be shown in the final Report by region within the
       U.S.
       Exhibit 3-19: More attention should be paid in the final Report to the "other"
       category; relative to the identified land use categories, this change is very large
       and needs better classification.

Trophic State of Coastal Waters
       The Panel recommends that in the final Report the water chapter indicator called
       "trophic state of coastal waters" be renamed "nutrients" or "eutrophication" in
       coastal waters.  Trophic state is a larger concept that would encompass, for
       example, depleted stocks of large piscivores such as cod in New England that
       have altered food webs and the trophic state of the waters through top-down
       cascading effects. As written, the focus of this section of the water chapter is only
       on bottom up, nutrient-related issues. In future Reports on the Environment it
       would be useful to include considerations of the state of all trophic levels but the
       name of the indicator could be changed immediately.
   -   3-62/2: It appears from the exhibits that this indicator does not include the Great
       Lakes coastal regions; this should be denoted in the final Report.
       3-62/7: Note that algal blooms can also include attached, macroscopic algae, such
       as Cladophora or Enteromorpha blooms.
       3-62/12: If the definition of algal blooms stays strictly planktonic, this sentence is
       correct as is; however, if it is extended to include benthic algae, then this sentence
       will need amending to reflect that reduction of chlorophyll a by filtering activity
       is restricted to water column chlorophyll a.
   -   3-62/27: It may be instructive to include  a table in the final Report that lists the
       reference conditions for each region; this gives readers an idea of the thresholds,
       and how they vary with region.
       3-62/34: It does not appear that the composite U.S. score is weighted in any
       fashion. Is that correct?  Given the very different lengths of coastal areas in each
       Region, what is the rationale for giving them equal weights?
   -   Exhibits 3-20 to 3-25:
       A) In the final Report, a more effective graphic display would be to show the
       regional map of the U.S. in the center of the Exhibit (now in the bottom footnote),
       and have each region blown up as a pie chart, radiating out from the U.S. map.
       The pie chart would show the four water quality categories.
       B) As noted earlier, ecoregions are a much more scientifically defensible
       geographic approach for showing regional data than EPA regions.
       C) There may be value in applying statistical tests to determine if there are
       differences among regions; one would need to know more about the data
       computation and distribution, a Chi-square test may work.
   -   3-63/9: The indicator should be accurately defined in the final Report—this is
       dissolved inorganic nitrogen (DIN), not nitrogen, per se.
                                       A-5

-------
   -   3-64/1: As above, in the final Report this should be called dissolved inorganic
       phosphorus (DIP), not "phosphorus", which could mean a lot of different things to
       readers.

Dissolved Oxygen
       It is unclear when dissolved oxygen (DO) measurements were taken; because DO
       concentration is dependent on time of day, this point should be reflected in the
       discussion in the final Report.

Coastal Sediment Quality
       3-67/2: It is unclear why this indicator was not applied to fresh water systems, as
       well. This does not invalidate its merit for coastal systems, but readers may
       wonder why there is an apparent inconsistency among systems.
       Exhibits 3-26 and 3-27:  See comments above regarding 1) using a different
       graphic for these  exhibits,  using pie charts from each region emanating from a
       map of the U.S.; 2) revising composite U.S.  score to weight based on coastline
       within each region; and 3) statistical analysis for differences among regions.

Coastal Benthic Communities
       Exhibit 3-28: See comments above for Exhibits 3-26 and 3-27.

SAV in Chesapeake Bay
       3-74/2: In the final Report, EPA may want to generalize the importance of SAV
       beyond just Chesapeake Bay, similar to what is done for the introduction in the
       Hypoxia Indicator.
       3-74/28: In the final Report, it would be useful, either here or in indicator
       limitations, to identify what percent of total area was estimated based on prior
       years' surveys for those years with incomplete coverage. Is this a small amount
       (<10%) or something  more significant where the uncertainties have more
       significance?
       3-74/32:  The data show two distinct trends - increasing from  1984 to 1992 and
       leveling from 1992 to 2005.  These trends should be discussed.
   -   3-75/25: Species composition is also an important variable, as not all SAV species
       provide the same ecosystem functions.

Hypoxia in Gulf of Mexico and Long Island Sound
   -   This was  a very well structured indicator.

HAB Outbreaks Along the Western  Florida Coastline
   -   Other potential limitations to this indicator include: 1) cell density does not
       necessarily equate to toxicity; and 2) biovolume may be a better indicator than
       density, although this  may be too labor-intensive to compute.
       The selection of an indicator such as HAB should not be restricted to coastal
       waters; rather, the question is more appropriately aimed at surface waters—for
       future Reports on the Environment consideration should be given to reorganizing
                                      A-6

-------
       the questions in the water chapter around surface water (including both fresh and
       marine coastal), ground water, and drinking water.
       The HAB indicator is site specific. Perhaps the algal blooms are more common
       along the Florida coastline, but they are not unique to that region.

3.5.3 Discussion
       3-84/28: In the final Report the "location of a large city" should not be
       characterized as a human activity—the Report on the Environment is confusing
       state variables and flows. This should be reworded to identify the relevant
       activities (e.g., runoff from impervious surfaces, combined sewer overflows, etc.).

Population  Served  by Community Water Systems (CWS) with no Reported
Violation
       It is unclear why the indicator is not the inverse of what is presented—i.e., the
       number of CWS with reported violations.  This seems the more direct
       measurement.
       3-90/39: The value of reporting the number of regions above the national
       percentage of systems with no reported health based violations is unclear. Aren't
       the regional data used to develop the national percentage? What is the point of
       including this information?

3.7.3 Discussion
       Why not have a regional indicator based on number of beach closings (number of
       beaches or number of days)? This information is currently being collected in the
       Great Lakes, as part of the EPA Beach Act.

Coastal Fish Tissue Contaminants
    -   Exhibit 3-38: See comments given above regarding 1) using a different graphic
       for these exhibits, using pie charts from each region emanating from a map of the
       U.S.; 2) revising composite U.S. score to weight based on coastline within each
       region; and 3) statistical analysis for differences among regions.

Contaminants in Lake Fish Tissue
       Exhibits 3-40/41: In the final Report, it would be more instructive to represent
       these data by Region to show geographic differences.
    -   3-109/4: In the final Report, the absence of Great Lakes data should be  noted
       earlier in this section, not just as an indicator limitation bullet. This is important
       given the historical legacy of contaminants in this region, so the exclusion of
       these data may result in an underreporting of the degree of contamination.

Land Chapter

Land Cover
    -   The Panel's issues of concern for the land cover indicator are that: a) the data are
       old, b) the classification categories for land cover are too coarse, and c) to date
       there is no time series (trend) information, though EPA reports that time series
                                       A-7

-------
       information is under development.  There are a number of regional and national
       products; including the National Land Cover Data Set (NLDC), Landscape Fire
       and Resource management Planning Tools (LandFire), and others. These would
       provide greater resolution and time series for these analyses.
       The Panel recommends that in future Reports on the Environment, EPA consider
       using a range of land cover classification schemes with different levels of
       resolution based on what is most appropriate to answer specific questions.
   -   In the land chapter there is no discussion regarding the relationship between the
       status/trends in land cover and the effects on human health and the environment.
       In the final Report,  it would be helpful to elucidate what data collection and
       analysis will be required to answer these questions and steps that need to be taken
       to make this a practical and useful indicator.

Forest Extent and Type
       The Panel believes that EPA needs to characterize land cover of all major
       ecosystem types.  Therefore it is unclear why the Agency chose to only report on
       forest extent and type over other types of land cover. Other land cover types are
       distinguishable from existing imagery products data sets, the data presented for
       the land use indicator show trends in many different land cover (use) types. The
       rationale for only using forest extent and type needs to be clarified in the final
       Report.
       The forest extent and type indicator that is presented in Chapter 6 (ecosystem
       condition) only represents timberland. This presents only one category of forest
       land and others should be included in future Reports on the Environment.

Land Cover in the Puget Sound/Georgia Basin
       The land chapter regional example on Puget Sound using the National Oceanic
       and Atmospheric Administration Coastal Change Analysis Program (CCAP) data
       has very coarse classification information that translates to a low sensitivity
       indicator instrument.  The reference point of 10% impervious surface becomes an
       important metric to make a statement regarding what the indicator means, whether
       things are falling apart or improving, and when action needs to be taken. There
       are many changes brought up in this section that may be better suited to the land
       use indicator category.
       There is no good explanation why the Puget Sound example was chosen to be
       representative of "land cover." Such an explanation should be included in the
       final Report. The area encompasses many watersheds that have many different
       types of land cover, but the data only assess changes to forest and urban classes.
       The example does not provide much useful information or methodology that
       would describe an approach that should be used for assessment of land condition
       outside of the immediate area covered by the case study. The Panel suggests that
       regional indicators should provide this use through inclusion in this Report. The
       Panel supports the inclusion of regional examples, but believes that they should
       present data or methods that can be applied across the U.S. An example or case
       study should be chosen to demonstrate particular aspects of the conceptual model
       underlying the set of indicators and their linkage to the fundamental questions.
                                      A-8

-------
Land Use
       The Panel suggests that in future Reports on the Environment, EPA evaluate and
       adopt widely-used standardized approaches to classify land uses that have been
       developed through the National Resource Inventory and the Forest Inventory and
       Analysis programs as well as the National Agricultural Statistics Service and
       Economic Research Service.  The Panel finds that much of the introductory
       material in the land chapter discusses the differences between approaches rather
       than interpreting what they can tell us about the status of land resources. The
       Panel is pleased to see that this indicator provides a beginning of a time series that
       can be used to document trends.  The usefulness of the discussion will be
       improved by elaboration of what specific land uses changed from one class to
       another.
       The Panel notes differences between national and EPA regional data for land use
       in the Report. The discussion regarding the land use indicator is inconsistent
       concerning the inability to obtain data for land cover and the data available for
       land use. The land use data sets imply that there are trend data available for land
       cover.  If this is  not the case because the land use and land cover data are different
       in nature, this will need further explanation. For future Reports on the
       Environment the agency should work to adopt standard approaches for land use
       and land cover analyses. Standardized land use and land cover analysis and
       reporting at national and regional scales ultimately will benefit from a higher level
       of classification and mapping consistency across all federal agencies.
       For future Reports on the Environment, the Panel recommends that EPA consider
       adding road density (which can be measured directly just like stream density) as
       an indicator for land use. Accurate road density data are readily available in
       electronic, Geographic Information System (GIS) format (e.g., as  Topographically
       Integrated Geographically Encoded Referencing System [TIGER] files). Since
       TIGER is a U.S. Census Bureau product, its limits and accuracy are well
       documented  (http://tiger.census.gov/).  Nationwide data are available, and
       changes over time can be mapped and measured.  Density can be determined for
       specified regions of interest.  There are other sources of road network data, but a
       1990-2000 TIGER would be a good start.
Urbanization and Population Change
   -   The Panel finds that the urbanization and population change indicator in the land
       chapter presents much good information regarding the relationship between these
       factors.  However, the chapter provides limited and indirect examination of the
       relationship between the available information and the resulting affect on human
       and environmental health.
       The Panel finds that the urbanization and population change indicator actually
       measures a stressor of land use in addition to one type of land use (developed
       land). The Panel questions whether population is a land use indicator or part of a
       group of indicators considered to be major drivers (stressors) of most indicators.
                                       A-9

-------
   -   In the final Report, the developed land data set that is used in the land chapter to
       represent urbanization and population change needs to be clearly described in the
       introductory text for this indicator.  EPA should evaluate whether a more useful
       indicator might be "population density by land use type," not by EPA region.
       Exhibit 4-11 on page 4-33 shows population density in the U.S. by EPA Region
       but the EPA Regional averages do not capture the aggregation of population
       density.  Data aggregation is a major issue and EPA needs to be cautious that this
       does not misrepresent the extent and intensity  of environmental impact.
   -   The discussion for the land use indicator addresses human residential and
       commercial uses.  The Panel suggests that more could be said in the final Report
       about other land changes (e.g., changes in agricultural land and associated
       fertilizer and pesticide use), beyond just identifying them as gaps.

Quantity of Municipal Solid Waste Generated and Managed
       The data used to represent this indicator are well defined and consistently
       collected. However, the connection to human health and the environment is
       missing and should be discussed in the final Report.
       In the discussion of indicator limitations it is stated that the available information
       is model driven. The Panel recommends that more information be provided in the
       final Report about sources of uncertainty associated with the modeled estimates.
       If the estimated waste generation is based on a model that uses materials utilized,
       these changes and thus the quality and quantity of the waste is not "consistent
       from year to year" as stated.  The Panel also notes that this indicator does not
       appear to meet EPA's indicator acceptance criteria. The Panel  does not
       recommend omission of the indicator, but more discussion of the quality of the
       estimate is needed in the final Report.
       In the discussion of indicator limitations, a gap concerning landfill capacity is
       identified.  The Panel notes that landfill capacity is not a nationally limited
       resource (only cheap landfill space near some very large cities is in short supply).
       Therefore, landfill capacity should probably not be listed as a gap in the final
       Report.
       Some discussion regarding the influence of programs on trends in this indicator is
       not well linked to the data and should be avoided in the final Report. For
       example, on page 4-46 it is stated that, "Recycling efforts related to municipal
       solid waste have increased over the four decades showing the steepest increases
       between 1980 and 2000, most likely due to the increased awareness about the
       benefits of recycling and the implementation of policies by state and local
       governments tying waste generation directly to the cost of waste services."

Quantity of RCRA Hazardous Waste Generated and Managed
   -   The data used to represent this indicator are well defined and consistently
       collected. However, the connection to changing levels of exposure and the
       resulting impact to human health and the environment is missing and should be
       discussed the final Report. This is an indirect  land use issue, especially when
       deep well injection is a major method of getting rid of the RCRA waste.
                                      A-10

-------
Fertilizer Applied for Agricultural Purposes
       There is good information presented in the land chapter discussion of this
       indicator, but it is again not related to human and environmental health. The
       "delta" between fertilizers applied, that taken up by the crops, and that which is
       released to the environment is the most relevant indicator.
       The Panel notes that this indicator is limited to three crops and questions how well
       it represents fertilizer application in cropping across the U.S.  The Panel also
       notes that separation of data for this indicator by EPA regions could be helpful
       since nitrogen and phosphorus drain into rivers, and large watershed regions
       might be more appropriate.
       The Panel recommends that a pesticide use indicator be added to the land chapter
       of the final Report. This could be done by renaming the indicator as "Fertilizer
       and Pesticide Applied."  In this regard, one possible indicator that could be used
       is pesticide sales, which could likely be parsed into agricultural and
       residential/commercial landscape  applications. The latter would provide a
       suburban/urban indicator, which is important from the standpoint of human
       exposure.

Toxic Chemicals in Production-Related Wastes Released, Treated, Recycled, or
Recovered for Energy Use
   -   The Panel notes that the title for the indicator should perhaps be modified in the
       final Report so that it does not appear that only toxic chemicals related to energy
       use are being considered.
   -   Reductions in toxic chemical releases can benefit human health and ecological
       condition. The indicator limitations section clearly points out the gaps in our
       knowledge and reporting base.  In the final Report, this indicator might be more
       appropriately placed in a section dealing with toxic and harmful chemicals.
       In the final Report it would be helpful to weight the amounts of toxic chemicals
       by toxicity (e.g., the un-normalized weights given in Exhibit 4-18  on page 4-55),
       but this is addressed under limitations.
       The Panel recommends that in the final Report indicator data (e.g., Toxics
       Release Inventory [TRI] derived) be included for persistent bioaccumulative
       toxics (PBTs) and mining wastes, even if the available data are limited, such as is
       apparently the case for PBTs. The Panel notes that PBT data are available for
       some aquatic ecosystems such as the Great Lakes. This indicator therefore
       provides an opportunity for integrating land the land and water chapters.  The
       Panel  suggests that EPA consult PBT data available in the draft 2007 State of the
       Great  Lakes report
       (http://www.solecregistration.ca/documents/4201%20Contaminants%20in%20Sp
       ort%20fish%20(SOLEC%202006).pdf).

Pesticide Residues in Food
       The Panel notes that reductions in pesticide residues in food may have net
       positive health benefits. However, the linkage of this indicator to land use is
       weak and the Panel recommends that in the final Report the indicator be moved to
       Chapter 5 (Human Health).
                                      A-ll

-------
    -   The indicator limitations section clearly points out that we should be monitoring
       the detections that exceed established tolerance levels in addition to what our
       instruments are able to detect.

Reported Pesticide Incidents
       The Panel finds that the decline in reported pesticide incidents has a direct
       relationship with human health. However, the link between reported pesticide
       incidents and the human health impacts of land management practices is tenuous.
       Reported pesticide incidents cover all sorts of uses of pesticides, and are based on
       calls to poison control centers. Many of these incidents are related to misuse of
       household products and activities far removed from land management. The Panel
       recommends that in the final Report the indicator be moved to Chapter 5 (Human
       Health).

High Priority Cleanup Sites with No Human Contact to Contamination in Excess of
Health-Based Standards
    -   The Panel finds that this indicator has a direct connection to human health and
       addresses whether people are being kept away from hazardous sites.  It may be
       useful to include some RCRA Corrective Action sites in the analyses in the final
       Report.  In addition, it may be useful to provide an indicator that would address
       the number of sites that have been taken off the high priority site list.
       The Panel recommends that EPA consider including in the final Report an
       indicator for the number and associated land area of sites of this type that have
       been cleaned up.

High Priority Cleanup Sites where Contaminated Groundwater is Not Continuing
to Spread Above Levels of Concern
    -   The Panel finds that this indicator also has a direct connection to human health as
       it addresses whether contaminated waters are being contained. It would appear
       that there are many additional CERCLIS listed  sites and other RCRA Corrective
       Action sites that could also be included in these analyses. It may also be useful in
       this case to provide an indicator in the final Report that would address the number
       of sites that have been taken off the high priority list. EPA should also consider
       including an indicator of the number and associated land area of sites of this  type
       that have been cleaned up.

Human Health Chapter

Health Effects of Air Pollutants
    -   In describing health effects associated with air pollutants, authors should be
       careful to include in the final Report those effects associated with low-level
       exposure as occurs in the ambient environment.  For example, ambient carbon
       monoxide is described as having effects including cardiovascular, neurological,
       visual impairment, reduced work capacity, reduced manual dexterity, poor
       learning ability, and difficulty performing complex tasks. The Panel questions
       whether these effects are associated  with low level exposures.
                                      A-12

-------
Ecological Condition Chapter

Need for Additional Indicators
    -   Indicators are provided in the ecological condition chapter to answer the question:
       "What are the trends in the diversity and biological balance of the nation's
       ecological systems?" The Panel notes that the concept of biological balance
       includes complex interrelationships for which clear indicators are not easily
       selected.  Those indicators selected are either population states or events which
       are difficult to translate into "balance." Very few biological taxa indicators are
       included.  At present, it is a good start but inadequate. Far more indicators of
       floral and faunal groups as well as biological communities should be included in
       future Reports on the Environment.

Ecosystems are Missing
       Western continental issues.  In the ROE 2007 Science Report there is little  or no
       attention paid to the arid ecosystems in the Great Basin and the desert southwest.
       Grassland/prairie, shrublands, rangelands, and chaparral are important ecosystems
       in terms of biodiversity.  It is important to include information on these
       ecosystems in future Reports on the Environment.
    -   Coral reefs.  Coral reefs have been in  serious decline due to eutrophication,
       overfishing, siltation, disease, and climate, among other factors. Many of the
       factors affecting  coral reefs are germane to EPA regulatory programs. Much
       monitoring data are available on these ecosystems. The Panel notes that earlier
       reviews recommended that coral reef  cover,  which had been proposed as an
       indicator, not be  included in the 2007  ROE Science Report because it lacked
       calibration between methods, lacked an explanation of how sites were selected,
       and lacked a consistent analytical framework to adjust for bias in geographic
       distribution and sampling method. We think that, because of their ecological,
       economic, and recreational value, the  benefits of including corals in the Report
       outweigh these problems. Many coral reef monitoring programs use transects,
       and data from these monitoring programs could be used in the Report. A regional
       coral reef indicator could be developed, using only those that reefs that were
       sampled appropriately. Problems with the data could be described in the
       limitations and gaps section.  The Panel recommends that coral reef information
       could be  added to future Reports on the Environment.
       Soil ecosystems. Soils are one of the  key drivers that cut across all terrestrial
       ecosystems. Soil is a fragile  and finite resource that plays a unique role in
       maintaining air and water quality.  Use and management of native, agricultural,
       forested,  range, and urban lands play an integral part in influencing soil  and water
       quality within a watershed. Protecting soil quality is important for ecosystem
       productivity and water quality.  Soil morphological, physical, chemical, and
       biological properties can serve as indicators.  Spatial data in various ecoregions
       are currently available on range of soil properties and should be included in future
       Reports on the Environment.
                                       A-13

-------
Populations are Missing
       Marine/estuarine fish. The Panel recommends that in future Reports on the
       Environment, the ecological condition chapter include considerations of
       marine/estuarine fish populations.  There are numerous long-term data on these
       populations available from NOAA Fisheries.  Many species are in decline due to
       overfishing; this has received considerable attention. The depletion of predatory
       fish can have ramifications through the food web via trophic cascades that can
       result in reduced numbers of grazers, and subsequent algae blooms, that can
       exacerbate eutrophication. The depletion of filter feeders such as oysters can also
       lead to reduced water quality.  While fisheries are not EPA's responsibility, the
       depletion of upper trophic level species can have major effects on the ecosystem
       and environmental quality.
       Amphibians.  The Panel recommends that in future Reports on the Environment
       EPA include in the ecological condition chapter an indicator dealing with
       amphibians.  There have been many studies documenting the precipitous decline
       and loss of populations of amphibians, and some of those could be used to
       construct an indicator.  While the reasons for the disappearance of amphibians are
       not all understood, some factors involved appear to be climate change, ultraviolet
       radiation, and pesticides, all  of which are relevant to EPA. If development of a
       national indicator is not possible, a regional one could be developed.
       Invasive species. The Panel recommends that in future Reports on the
       Environment EPA include data on non-indigenous invasive species in a variety of
       terrestrial and aquatic ecosystems.  There are numerous data sets that could be
       used to develop indicators, at least for some regions. For example, SERC
       (Smithsonian Environmental Research Center) has  data sets for marine/estuarine
       invasive species. Additional sources of information are the Global Invasive
       Species Database of the Global Invasive Species Information Network
       (http://www.invasivespecies.net/), and the National Invasive Species Information
       Center hosted by the U.S. Department of Agriculture
       (http://www.invasivespeciesinfo.gov/).
       Taxa containing massive diversity. The Panel recommends that in future Reports
       on the Environment indicators be developed for taxa such as microflora and
       microfauna, and non-vascular and vascular plants, which have very high
       biodiversity. Ecosystems host complex microbial communities, including
       bacteria, fungi, protozoa, and viruses. The size and diversity of microbial
       communities are directly related to quality and quantity of resources available.
       Microbial processes  and populations  have more rapid turnover than higher trophic
       levels and are often more  responsive  to environmental change.  These
       characteristics make microbes good indicators of ecosystem condition because
       they are potentially very sensitive to perturbations such as nutrient loading,
       hydrologic alterations, and fire. New information is now emerging about these
       indicators and the Panel hopes this information will be added in future reports.

Processes are Missing
       Denitrification.  The Panel recommends that in future Reports on the Environment
       the ecological condition chapter include an indicator of the natural denitrification
                                      A-14

-------
       process which is important for nutrient balance in ecosystems, for example, the
       denitrification of nitrate from atmospheric deposition. Ecological processes in low
       order streams are important in processing excess nutrients (e.g., denitrification of
       N from atmospheric deposition).
       Soil processes.  Another issue of importance is the trend in the extent and
       condition of the nation's soil resources. As noted above, soils are one of the key
       drivers that cut across all terrestrial ecosystems.  Soil quality and associated
       processes can have major influences on ecosystem productivity and nutrient
       cycling. Loss of topsoil due to erosion and other processes can influence
       ecosystem productivity and long-term assimilative capacity as well as stream
       water quality. Assimilative capacity is important as ecosystems have finite
       capacity to provide services before they are drastically altered.  For example,
       long-term  application of nutrients via fertilizers or organic wastes may ultimately
       saturate a system. This is evident through accumulation of phosphorus in soils
       and increased levels of nitrate in ground waters.  Salination of irrigated farmland
       soil is an urgent issue in the arid Southwest. Potential soil quality indicators
       include: carbon storage, organic matter, nutrient inventory, phosphorus index,
       extent and soil type, soil quality, salinity, soil erosion. The Panel recommends
       that future Reports on the Environment consider these indicators.
   -   Acidification. The Panel notes that there are long-term data sets available on
       responses to acidification and its reduction (National Acid Precipitation
       Assessment Program [NAPAP]) that should be included in future Reports on the
       Environment.
   -   Disturbance.  Disturbance is a critical process in all ecosystems and should be
       included in future Reports on the Environment. The Report discusses its
       importance but has no indicator of disturbance or response to it (e.g., resilience).
       Disturbance processes can be used as indicators of anthropogenic effects on the
       environment.  For example, maps showing how fire cycles have changed in
       relation to the health of forests can provide important information on a critical
       issue

Trends in Diversity  and Biological Balance of the Nation's Ecological Systems
   -   On page 6-29, the final Report should acknowledge that some systems inherently
       have different numbers and variety of species, making it inappropriate  to make
       comparisons among systems.

Choice of Forests, Wetlands, and Land  Use  as Indicators in Chapter 6
       While there is nothing wrong with these categories, it is unclear to readers why
       these were chosen and not other equally appropriate categories.  A conceptual
       framework would be very helpful in the final Report to place these categories and
       indicators  into some type of context.

Forest Extent and Type
       This indicator is limited to "timberlands" which is misleading.  This is nearly
       equivalent to using corn and wheat fields in order to describe the extent of
       grasslands. The Panel notes that this indicator is based on productive capacity,
                                       A-15

-------
       and therefore a statement in the Report concerning the limits of indicators that
       have excluded production does not apply. However, the discussion of indicator
       limitations does recognize some of the limits of using timberland data.
   -   6-16/15: What percent of forest land is not being captured in this analysis? In the
       final Report this percent should be explicitly noted as part of the uncertainty.

Forest Fragmentation
   -   The Panel understands the value of using forest fragmentation as an indicator but
       questions why a fragmentation indicator is not equally important for the other
       ecosystems. The Panel questions whether this is because of the availability of
       data. The Panel finds that in the final Report, a schematic diagram graphically
       showing the four degrees of forest cover would be helpful to complement the
       narrative, as would a presentation of the absolute area of forested lands identified
       for each region.

Wetland Extent, Change, and Source of Change
   -   Development of artificial wetlands, ponds etc. may skew data for this indicator.
Ecological Connectivity (Region 4)
   -   The Panel notes that development of this indicator is an exercise demonstrating
       how to show connectivity, but since it is regional it does not tell much about
       connectivity either nationally or in major ecoregions.  The distinction between
       hub and corridor should be better defined and shown in the map in the final
       Report.  If the methodology is relatively simple and uses just National Land
       Cover Data Set (NLCD) data, then a major effort should be made to see if it is
       applicable to non-forested regions.

Relative Ecological Condition of Undeveloped Land (Region 5)
       The Panel finds that this is a case where a tool has been developed for one EPA
       region but it does not tell the story about the landscape in general or its
       usefulness.  The indices used have the potential to  display a lot of information, but
       it is not stated what exact data layers are included in each index. This tool used
       only NLCD data to generate three indices, two  of which use species diversity or
       rarity.  The Panel questions whether it is possible to go to species level with
       NLCD satellite data. If models were used for the diversity and rarity indices, they
       should be explained. In the discussion in the final  Report it should be noted that
       increases in developed land affect habitat and impact physical and chemical
       processes such as runoff from impervious surfaces, reduced groundwater
       recharge, and increased stream temperatures.
       Shades of green are extremely difficult to distinguish in Exhibit 6-8.
       Undeveloped is a relative term and appears to be confounded with population
       density, making it inappropriate to draw conclusions or causative associations (as
       on page 6-27, "The potential for future land use changes with increasing
       urbanization is the major determinant for judging potential fragmentation of
       ecological systems in EPA Region 5...")
                                      A-16

-------
   -   In the final Report EPA should clarify the interpretation or importance of the
       cover types mentioned: maple-beech-birch, spruce and pine.  Is this simply a
       descriptive statement or should the reader be able to infer something about a trend
       of ecological significance?

6.2.3 Discussion
       It is unclear why forests, wetlands, and land development, of all available
       indicators, are the three worth highlighting for the nation's ecological  condition.
       This should be clarified in the final Report.
       6-27/1-3: It may also be worth noting that these increases in developed land affect
       not only habitat loss for biota, but also impact physical and chemical factors, such
       as more runoff from impervious surfaces, leading to greater loading of nutrients
       and contaminants, a more unstable hydrology, reduced groundwater inputs, and
       increased stream temperatures.

6.3 Discussion
   -   6-30/25-31: It is helpful to know about the absence of a systematic biodiversity
       initiative in the U.S., but there is still a need to explain in the final Report the
       rationale behind including those indicators that are found in the Report.

Bird Populations
       The limitations on the data set should not detract from the usefulness of this
       indicator.  It is one of the most consistent, long-term sets of ecological measures
       in the whole Report.
   -   In the final Report, EPA may want to note that the significant increases or
       decreases are of observations, not population size.

Fish Faunal Intactness
   -   The discussion in the ecological condition chapter states a concern over the
       inability to show magnitude of loss. The Panel notes that this could be remedied
       by using a map of number of species lost.  With such a small number of species to
       begin with, the percent decline figure can be misleading.  The Panel questions
       whether using 1970 as a reference year potentially confounds comparisons from
       regions that were heavily polluted at that time. The Panel recommends that data
       from estuarine fish should be included in future Reports on the Environment.

Non-indigenous Species in the Estuaries of the Pacific Northwest
       The limitation one area implies that non-indigenous species are less important in
       other estuaries.  The Panel notes that the restriction to species captured in a grab
       sample suggests that this is how most invasive species can be sampled. However,
       this is not true.  More estuarine invasive species tend to be epibionts that attach to
       surfaces.  Some invasive species cause greater disruption of ecosystems than
       others, so it may make sense to use indicators that address those species that are
       most ecologically or economically problematic. The Panel finds the preliminary
       classification of estuaries as "exposed" or "background" depending upon the
       assumed amount of ballast water or aquaculture releases is naive, since estuarine
                                      A-17

-------
       biota disperse, and currents aid their spread, particularly in the planktonic stages.
       There is no need to pre-classify estuaries.  Once the data on non-indigenous
       species are collected, then estuaries can be classified according to their percentage
       of non-indigenous species. The Panel recommends that in future Reports on the
       Environment this indicator be expanded to other estuaries as well as other aquatic
       and terrestrial ecosystems.

6.3.3 Discussion
   -   6-40/18: Chesapeake Bay SAV may not be a representative example for wider-
       spread phenomena.
       6-40/24: It may not be possible to statistically defend this claim with the  available
       data sets.
   -   Good regional long-term data sets may be available to address above ground plant
       richness and diversity (e.g., Long-term Ecological Research Programs, Harvard
       Forest data)
   -   6-42/25: Perhaps a useful template for the type of exhibit on this page would be a
       map of the U.S. subdivided into regions, with more detailed maps of each region
       showing data for different representative species. For invasive species, this may
       be an autotroph or a heterotroph, or aquatic vs. terrestrial, depending upon which
       species provides the best information for the region.

6.4 Discussion
       Perhaps the Millennium Ecosystem Assessment (MEA) model could be  used for
       identifying ecological processes that sustain the nation's ecological systems (i.e.,
       provisioning, regulating, cultural, and supporting).
       There is a fundamental problem in the indicator chosen for this question. The
       question deals with processes, but the indicator deals with a state variable, not a
       process.  This can be resolved by changing the question or choosing an indicator
       that answers the question, such as primary productivity, decomposition rates, or
       nutrient uptake/cycling rates. Long Term Ecological Research Program sites
       should provide a rich source of data for these types of information.

Carbon Storage in Forests
       The Panel notes that carbon storage in forests is not an ecological process per se,
       but a condition representing the net balance between the processes of
       photosynthesis and decomposition. This indicator can show trends. However,
       many more processes need to be covered in future Reports on the Environment.
       The use of several geographic regions is more logical here than the use of EPA
       regions elsewhere.  Unfortunately, the data in the chapter represent only
       "timberlands" which include many highly managed forests and this should be
       pointed out in the limitations section. Use of this indicator should be expanded in
       future Reports on the Environment to carbon storage reservoirs, such as
       grasslands, especially below ground (soil) storage which holds a significant
       portion of the total carbon.

Photosynthesis and Decomposition
                                       A-18

-------
   -   Photosynthesis and decomposition are the two most important ecological
       processes. Carbon storage is described as an indicator representing the net
       balance between these two processes. Restricting the indicator to forests and
       excluding grasslands greatly weakens this indicator. In the final Report this needs
       to be discussed in the limitations section on page 6-46.

6.4.3 Discussion
    -   6-48/4: The indicator does not provide data on trends in primary production; this
       process is a rate.  The indicator provides data on a stock, which is different.
       6-49/1: Another limitation, assuming carbon storage is used as the proxy
       indicator for this question, is that carbon storage from many other important
       terrestrial ecosystems is missed.

U.S. Temperature and Precipitation, Sea Surface Temperature, Sea Level
       These are very good time series data. They are all physical attributes that have
       impacts on biota and on ecological processes. These indicators, and their links to
       greenhouse gas emissions discussed in the air chapter, should be included in the
       ecological condition chapter discussion in the final Report.

Sea Surface Temperature
       6-59/4: Why not include statistical information?

Sea Level
   -   Although not technically "sea" level, one limitation is the lack of data reported for
       Great Lakes levels.  These data are available from the U.S. Army Corps of
       Engineers (Detroit District), and should be considered for inclusion in future
       Reports on the Environment.
                                       A-19

-------
Appendix B: Editorial Comments

General

 -  Throughout the Report there is generous use of acronyms, which may be confusing,
   but perhaps unavoidable. Thus, including a list of acronyms and abbreviations (e.g.,
   units of measurements) would be an improvement.
 -  As this version of the Report is intended for scientists rather than the general public, it
   would be helpful to use metric system units throughout for measurements (e.g. °C
   rather than °F for temperature).

Introduction

1-4: Identify explicitly the philosophy behind choosing indicators.

Water Chapter

3-7/14: Replace "like" with "such as".
3-7/42: Move "only" to after "meet".
3-9: Should N and P discharge be load?
3-9: Delete "wetland extent..." from the coastal waters box.
3-11/37: Seems that NFS paragraph also should include affects of land cover, such as
impervious surfaces.
3-12/1-3: Air deposition should include nutrients, as well (N and P).
3-12/12-18: It is not just extent of the fresh waters, but also their configuration in the
landscape that matter. This should be noted.
3-15: Exhibit 3-1: Clarify caption: "Relative percentages of rivers and streams in terms of
their changes of high and low flow ..."
3-17/3: Are any estimates available of the percentage increase of dammed rivers between
1949 and 1970?
3-26/10: Suggestion—briefly  explain the weighting scheme used.
3-28/24: Change to "have a broad geographic distribution".
3-36: Needs enumeration of rows 1-45.
3-42/19: Add: Groundwater accretions in agricultural watersheds may also increase
contaminant loads of rivers and streams.
3-50/17: This is not  a location classification—it is salinity of media.
3-53/4: Insert "and other types of coastal" after "Estuarine".
3-53/7: Insert', chemical'  after "biological".
3-53/14: Insert "These conversions reduce the area of the relatively unique systems such
as forested swamps and bogs and increase the area of the ubiquitous ponds and marshes."
after "pond."
3-54/19: Insert "although still much less in  absolute terms than the other wetland types"
before "Panel D".
3-54/40; 3-55: Exhibit 3-19—Please clarify  the meaning of describe the process of
"deepwater conversion".
3-57: Insert "and continue  to be lost" after "1990s".
                                       B-l

-------
3-57/29: Insert "and some wetland types such as forested swamp and bogs are difficult or
even impossible to create or restore." after "lost".
3-57/32: Insert "using a logistically plausible" after "estimate" and Remove "without an
impractical".
3-58/3: Insert "function and" before "condition".
3-63/9; Exhibit 3-21: Include quantitative information for nitrogen concentrations.
3-64: Exhibit 3-32: Include quantitative information for phosphorus and chlorophyll.
3-67/3:  May want to define what is meant by "adverse".
3-71: Exhibit 3-28 is missing letters and a dash in the label within the text box.
3-79: Exhibit 3-32 is missing data in Panel A.
3-71: Exhibit 3-38—Fix caption number.
3-72/19: Address the possible effect of the weighting scheme and methodology on the
results.
3-79: Exhibit 3-32— Include missing data in graphic.
3-82/6: Limitations—the temporal trend is limited by the short time span (only 5 years of
data).
3-84/30: Comment—How much different?  Many times higher or less?
3-87/5: Suggestions—Include brief definitions of surface water and ground water.
3-103/17:  Suggestion—Include brief description of health risk basis of guideline.
3-104: Exhibit 3-38—Indicate that the values are percentages. Add: "Percent" to caption.
3-108/10:  Briefly explain toxic equivalents (TEQ). MDLs have no direct relations to
health risk.
3-109/2: Comment—Imported seafood accounts for 70% of consumption. Perhaps it is
also an FDA issue.

Land Chapter

4-61: Consider an outline to the bars in Exhibit 4-6, 4-7 (pp. 4-26, 4-27) to make them
more visible, such as in Exhibit 4-23.
4-31: legend is incomplete in Exhibit 4-8.
4-50: NPK are identified as pounds per acre; are these the desired units?
Are these  values devoid  of inert ingredients (i.e., just element)?

Human Health Chapter

5-7: Figure 5-1  can be enhanced by depicting susceptibility factors including genetics,
diet, etc. described in the paragraph starting on line 28 of page 5-6.  Furthermore, this
figure might more effectively appear within the introduction as a way to provide both a
conceptual framework for the Report on the Environment as well as the organizing
principle.  The figure and text would need to be modified to include ecological effects
and to show increased uncertainty as indicators move from left to right.
5-7 and 5-8: In the introduction, terms such as definitive proof and "conclusive
evidence" are used.  It might be better to omit the adjectives.

Ecological Condition Chapter
                                       B-2

-------
6-14: In Exhibit 6-2 add black outline bars to make it easier to see light colors.
6-14: In Exhibit 6-2, indicate the percentage changes rather than absolute changes
because forest coverage and sizes differ from region to region.
6-15: Clarify that emphasis in Exhibits 6-3 and 6-4 is on economically important species.
6-15: In Exhibits 6-3 and 6-4, indicate percentage changes rather than absolute changes
because covering and sizes differ from region to region.
6-18: On line 13 explain "degree of connectivity." Can a quantitative definition be used?
6-21: In Exhibit 6-6, a different color scheme should be used. The map does not show
clearly the difference in the greens.
6-22: On lines 22-24, please specify the twelve layers and the four layers if possible. Are
any weighting factors used?
6-25: On line 15 clarify "decreases in Regions 6 and 9" and "increases in Regions 3 and
5." The data in Exhibit 6-2 show discrepancies from the general  statement in the text.
Region 9 has increased during 1977-2002.
6-30: Insert acknowledgement that nutrient enrichment can also be considered a
"pollutant" and be responsible for community shifts toward invasive species.
6-30: On line 1 the following suggested change in the wording is provided: ".. .by global
events  such as large meteor impacts..." or .. ."bolide collisions..."
6-32: With regard to bird populations, delete the following debatable statement, "are
among the most visible and important biological components of ecological systems  and"
6-32: Note whether abundances in Exhibit 6-9 are standardized by numbers of observers.
6-32: On line 22 discuss the possible causes for the decrease in grassland species.
6-34: With regard to fish faunal intactness, explain why 1970 is chosen as the reference.
6-35: Expand the legend in Exhibit 6-10 to explain the  pie chart (i.e., reduction areas
expressed as % total land area).
6-37: On line 12 replace ">=" with ">".
6-38: In Exhibit 6-12, illustrate where the "exposed" and "minimally exposed" estuaries
are located on the map and provide an idea of the sampling intensity.
6-40: On line 24, the following statement needs supporting data and justification:
".. .fewer blooms in recent years as compared to 1996..."
6-45: The key in Exhibit 6-13 is missing the color codes.  Letters are missing in the  title
of the exhibit.
6-45/32: The word "somewhat" understates the trend.  Inspection of the data indicates a
decline in the 1990's of approximately 33%, which is more substantial than "somewhat."
6-46: In Exhibit 6-14, indicate in the captions and on the labels that the values are net
changes of storage, not total storage.
6-46: On line 31, can an estimate of carbon storage (e.g., % of total) in soils be provided?
How significant is this omission?
6-46: In Exhibit 6-14, add  outlines to fill in order to increase the visibility and
acknowledge that the net carbon storage is affected by climate and soils.
6-48: On line 9, the dates in the discussion do not correspond to the dates presented  in the
indicator.
6-49: On line 2 include estimates of carbon storage in soils.
6-53: On line 41, what is the confidence level or statistical significance of the regression?
6-54: In Exhibit 6-16, add  negative signs on the temperature scale.
6-55: On line 14, include the names of the three climate regions.

-------
6-56: In Exhibit 6-18, the graphs as presented do not clearly show support for the
discussion. Please modify the graphic data to show statistical significance.
6-56: In Exhibit 6-18, the Y axis scales should be changed to appropriate values to better
show trend data.
6-57: On line 2, a limitation should be added indicating that the empirical debiasing
models used to adjust the data may themselves introduce non-climatic biases.
6-61: On line 33 the following change in wording is suggested: "subsidence or uplift
caused by tectonic movements of landmasses." Delete "changes in natural land
accretion."
6-64: On line 6, the following change in wording is suggested: ".. .due to changes in sea
level or land elevation caused by tectonic movements."
                                       B-4

-------
Appendix C: Example Conceptual Framework

   A synthesis chapter is needed to pull together the findings of EPA's Report on the
Environment. The findings reported through the questions and indicators show status and
trends of many different environmental parameters. These parameters have been placed
in two types of chapters: 1) media (i.e., air, water and land), and 2) health and ecological
condition. However, it is important to indicate that parameters are linked through
attributes and processes that control the parameters as well as attributes and processes
that the parameters influence.  Consequently, to  demonstrate the integration and synthesis
of the Report, three components need to be added 1) a conceptual  framework, 2) a
synthesis discussion, and 3) a simple and clear description of each indicator, with a
discussion providing a rationale of why it was selected, and what it should tell.  It is
recommended that the conceptual framework and the description of each indicator be
added to the final Report, and the synthesis chapter be added to future Reports.

   Conceptual Framework Component: We suggest that this be included as part of the
suggested introductory chapter and possibly part of the introductions to the human health
and ecological condition chapters.  An example conceptual diagram is shown in the
accompanying figure, which has the compartments of air, water, and land, and shows
their connection to each other and their impacts and linkages to human health and
ecological condition through physical, chemical  and biological processes. Any such
diagram that EPA develops for this purpose should show the flows, influences or
connections among the many parameters selected to be used in the ROE 2007 (see
figure).  The figure is provided as an example. A more specific and detailed conceptual
framework could be developed to represent relationships among indicators and  media.

   Synthesis Discussion: A discussion of the interactions among the many parameters
selected to assess the state of the environment should include a commentary on the
importance of status and trends of selected parameters within a particular medium.  The
commentary should draw upon the observations  of status and trends within each indicator
discussion. The synthesis should indicate how status and trends in one parameter might
influence other parameters within the same and other media. This  cross media discussion
then should demonstrate how status and trends in media parameters might influence
status and trends of parameters within the integrative chapters (human and ecological
health and/or condition). The status and trends within the human health and ecological
condition chapters should each be discussed to describe the importance of the findings
relative to "well being" and/or "sustainability" of each attribute (i.e., humans and
ecosystems).  The synthesis discussion should not put values (e.g., good, bad,
inconsequential) on status and trend data but rather focus the discussion on the
importance of the  magnitude or "direction" of the status or trend and its implications for
other factors.

   Importance of Indicator: Each indicator should have a clear explanation of why it is
important based upon a scientific (perhaps conceptual) fundamental understanding of
drivers and processes that if changing will alter the status of the indicator over time. An
example indicator description is provided in Appendix D.
                                       C-l

-------
            SAB 3/24/08 Deliberative Draft. DO NOT CITE OR QUOTE.
This draft SAB panel report has been prepared for final review and approval of the chartered SAB.
                 This draft report does not represent EPA policy.
            Example Conceptual Framework Diagram
                                                  HUMAN HEALTH
                                                  Exposure
                                                 •Air
                                                 • Drinking Water
                                                 • Food
                                               ECOLOGICAL CONDITION
                                                Ecosystem Processes &
                                                        Services
                                                 Ecosystem components
                                                \
                                                    Ecosystem Types
                               C-2

-------
Appendix D: Example Indicator Description

   The indicators should all have a consistent description and presentation to the reader.
We recommend that each indicator have a description and/or definition of the indicator,
and a clear justification and explanation for its selection. Secondly, we recommend that
each of the metrics used to report the indicator be described. The following example is
based on information contained in the SAB report, a Framework for Assessing and
Reporting on Ecological Condition (U.S. EPA Science Advisory Board, 2002)

Example indicator: Landscape Condition

Landscape is an area composed of a mosaic of interacting ecosystems or habitat patches.
A change in the size and number of natural habitat patches, or a change in the
connectivity between habitat patches, affects the probability of local extinction and loss
of diversity of native species, and can affect regional species persistence.  At the
landscape scale, the extent of broad land cover classes (e.g., forest, agriculture,
urban/suburban, surface waters) can serve as surrogates of habitat extent for broad classes
of species.

Landscape indicators are reported in the following three categories: extent of ecological
system type, landscape composition, and landscape pattern/structure. The extent of
ecological system type is an important indicator because it is correlated with species
decline.  Landscape composition information provides insight into long-term population
viability  because populations are unlikely to persist in landscapes where the largest patch
of habitat type is smaller than that species'  home range. Landscape pattern and structure
provides a measure of habitat fragmentation that may isolate vulnerable species restricted
to specific habitat types.

Metrics used to measure the indicator:
•  Extent of the ecological system/habitat type: (e.g., habitat area, perimeter-to-area
   ratio, core area, elongation, etc.)
•  Landscape composition: (e.g., number of habitat types, number of patches of each
   habitat, size of the largest patch, presence/absence of native plant communities,
   measures of topographic relief, slope, and aspect, etc.)
•  Landscape pattern/structure: (e.g., dominance, distance between patches, longitudinal
   and lateral connectivity, juxtaposition of patch types, width of habitat adjacent to
   wetlands, etc.).
                                       D-l

-------
Appendix E:  The Use of Ecoregionally Derived Indicator Information for Action
and Decision Making at the EPA Regional Offices
   All indicators need to be developed, analyzed and reported within an appropriate
ecoregional context.  This context includes important dimensions of scale and boundary,
and should be driven by the intended uses for the indicator information. The formulation
of the indicator parameters will be constructed from knowledge regarding their
relationship to regional-scale ecological processes. The mapping of indicator values will
be dictated by the amount of relevant data available for spatial analysis.

   Environmental protection and resource management agencies are administered
through hierarchical regional structures.  These regionalizations are agency specific, and
were developed through a complicated historical set of administrative and mission driven
factors.  Knowing that many environmental management and protection actions will be
implemented by the regional offices, it is a common error to use these administrative
regions as a surrogate for ecoregions for all ecological indicators. These administrative
regions often do not represent the boundaries of resources that are being protected and
managed by the Agency. These administrative regions should not be used as a
framework for indicator analysis.

   A two-step process is required to use environmental indicators to inform priority
management  and protection actions within an administrative region. First, the indicators
should be developed within an appropriate ecoregionalization framework, and the
analytical results should be generated for each indicator within each ecoregion in that
framework. Second, the indicator results should be spatially parsed to provide relevant
management  directives to the regional offices.

   As an illustrative example, let us consider the use of freshwater mussels as an
indicator of water quality factors and biological intactness. The current distribution of
mussel species represents a combination of hydrological connectivity, geochemical, land
use and pollution factors. The logical  ecoregional context for this indicator would be a
hydrologic watershed framework that is represented at an appropriate scale that captures
the relevant geochemical regimes and  associated land-use patterns. Time series data on
the composition and distribution of these species within these ecoregions relative to land
use, exotic introductions and pollution sources would provide the desired indicators of
ecological health.

   Any administrative region could contain either entire watersheds or parts of
watersheds. When a watershed is completely contained in one regional jurisdiction, that
region would take responsibility to respond when the indicator demonstrates the need for
intervention.  When a watershed is shared by multiple jurisdictions, a decision must be
made to lay out a formula for a) responsibility sharing, or b) designating full
responsibility for management and protection decisions across the entire watershed to a
particular administrative region.  The designation of full responsibility could either be
based on the relative percentage of geographic intersection between watersheds and
                                       E-l

-------
administrative regions, or by capacity and expertise factors within the different
administrative regions. The critical point is that responsibility must be assigned and
accepted to ensure the appropriate management response.
                                        E-2

-------
Appendix F:  Table of Recommendations to be Considered Before Finalizing the
2007 Draft Report

Table F-l in this appendix presents a summary of recommendations to be considered
before finalizing the 2007 Report.  The second column of the table provides page
numbers where recommendations may be found this advisory report. Additional
comments and suggestions are provided in the text of the report.  Detailed comments
pertaining to specific indicators are included in Appendix A.

Table F-l. Recommended Changes in the Final Report
Chapter
Page
Recommendation
Overarching
                   The introduction should be revised to clearly indicate
                   that the first three chapters are intended to address
                   status and trends using specific indicators for the
                   individual "media" of air, water, and land, and that
                   the next two chapters are intended to provide
                   integrated assessments of status and trends in human
                   health and ecosystem condition.
Overarching
                   The introduction should clearly state its purpose for
                   intended audiences and EPA. The introduction
                   should discuss how the Agency plans to use the
                   Report and its analyses and how the Agency wants
                   the Report to be used by the broader public. In this
                   regard, the final Report should state that the current
                   version of the Report provides status information to
                   establish baselines for reporting future trends, but
                   does not yet include long-term trend information for
                   many indicators..
Overarching
                  EPA should incorporate a conceptual framework into
                  the introduction to illustrate the connectedness
                  between the media, human health, and ecological
                  condition chapters.
Overarching
                  In appropriate places of the final Report,
                  interconnections between the indicators should be
                  established by cross-referencing the discussion of
                  indicators in different chapters.
                                       F-l

-------
Table F-l. Recommended Changes in the Final Report
Chapter
Page
Recommendation
Overarching
                   EPA should clarify whether the document refers to
                   qualitative or quantitative trends, or both (i.e., "trend"
                   as used in the document needs to be defined). The
                   word "trend" (used in the questions) has a specific
                   meaning in statistical science.
Overarching
                   All questions should be broadened to ask "What are
                   the status and trends..." rather than focusing only on
                   trends.
Overarching
                   EPA should explicitly state how each question in the
                   Report is related to a conceptual framework.
Overarching
                   EPA should provide a clear description of why each
                   indicator is important, the rationale for selecting the
                   indicator, what it tells, and the documented
                   relationship between the indicator and human health
                   and ecological condition
Overarching
                   Additional indicators (identified various sections of
                   this report) should be included to show the response
                   of more integrated components of the system or
                   address missing issues. For example indicators
                   should be included to capture the status of and trends
                   in ecosystem services.
Overarching
                   EPA should identify the status of the monitoring
                   programs (e.g., extant, "on hold," or expired) that
                   have provided indicator data used in the Report.
Overarching
                   Additional trend data (classified as either qualitative
                   or quantitative) should be included for as many
                   indicators as possible.
Overarching
10
EPA should clarify whether specific bullets in the
indicator limitations sections refer to indicator
limitations or data gaps.
                                       F-2

-------
Table F-l. Recommended Changes in the Final Report
Chapter
Page
Recommendation
Overarching
10
The discussion of gaps and limitations should be
expanded to identify some of the more prominent
available data sets that were excluded and the reasons
for their exclusion (e.g., technical concerns, lack of
statistical power, or other specific reasons).
Overarching
10
The discussion of data gaps and limitations should be
strengthened by adding or expanding existing
information in several areas. These include: 1)
Discussion of the need for a transparent set of
indicator metrics that can be well justified. The
current choices of metrics and benchmarks are not
well justified. 2) The need to provide additional
information on emerging issues such as chemicals of
emerging concern, exotic wildlife diseases or
invasive species (the emerging issues should be
discussed at the end of each individual chapter).  The
Panel specifically notes that perfluorinated chemicals
should be added to the list of emerging contaminants
of importance in Chapter 7 of the ROE 2007. 3)
Further justification and discussion of limitations
associated with the intervals of time used to establish
trends.
Overarching
11
The implications of each indicator limitation should
be discussed, and the uncertainties associated with
each limitation should be quantified to the extent
feasible.
Overarching
12
If EPA administrative regions continue to be used as
the basis for regionalizing data, the Panel
recommends that this process be better explained.
Air
16
A science framework should be incorporated into the
air chapter of the final Report to show the interaction
within, between and among media as well as between
and among indicators.
Air
16
A short historical section should be added to the air
chapter to provide background information on the
criteria pollutants.
                                       F-3

-------
Table F-l. Recommended Changes in the Final Report

Chapter
Air
Air
Air
Air
Air
Air
Air
Air
Page
16
16
17
17
17
17
18
18
Recommendation
SC>2 concentration should be added to the air chapter
as an indicator.
The air toxics indicator should be expanded in the air
chapter.
A broader explanation of what is in the National
Emissions Inventory (NEI) should be added to the air
chapter.
Further analysis of the trends in air indicators should
be added to the air chapter. While it is important to
know whether air indicator trends are increasing, it is
important for the reader to understand the reason for
the direction of indicator trends. The Report should
state where have we been, where we are now, and
where we are going.
An indicator should be added to the air chapter to
focus on the clear reduction of primary pollutants
(CO, SO2, and Pb) but much flatter trends in
secondary pollutants (63 and PM2.5), reflecting the
growing importance of secondary air pollutants.
A small section should be added to the air chapter to
discuss how climate change is affecting aerosols.
EPA should acknowledge and discuss the limitations
of a single pollutant, local source approach to
pollution control in the context of the marked
reductions in individual pollutants documented by the
indicators, and as exemplified by continuing
challenges with regard to ozone and PM2.5.
EPA should view the PM speciation network as the
vehicle to provide the needed information on PM
composition.
F-4

-------
Table F-l.  Recommended Changes in the Final Report
Chapter
Page
Recommendation
Air
18
The bias that may result from the choice of base year
for trends for a given air indicator should be
discussed, as this has implications in the
interpretation of the air indicator data.
Air
18
The effects of trends in ambient concentrations of air
pollutant indicators on exposure and dose should be
discussed.
Water
20
The questions in the water chapter should be
expanded to focus on the interconnectedness of
different systems (both within the different water
types and across media).
Water
20
Additional questions should be included in the water
chapter to incorporate missing information on
availability and usage of water for human activities,
especially with respect to both ground water and
surface water withdrawals (see data in Roy et al.,
2005 and Solley et al., 1995).
Water
20
EPA should examine the relevance of measures of
"Extent and Condition" across all aquatic ecosystem
types.  In this regard, the Panel finds that the question
on the "extent" of coastal waters is not meaningful
because for coastal waters, the issue of importance is
their condition not their extent.
Water
23
Data for the indicator "pesticides in agricultural
streams" should not be compared to human health
benchmarks.
                                       F-5

-------
Table F-l. Recommended Changes in the Final Report

Chapter
Land
Land
Land
Land
Land
Land
Page
30
31
31
31
31
31
Recommendation
EPA should consider the following suggested
revisions of the land chapter questions in order to
improve their clarity. 1) The Panel suggests that
trend information be developed wherever possible,
and that EPA use both qualitative as well as
quantitative data to generate trend information for all
indicators. 2) The waste deposition addressed in
Question 3 (wastes) could be considered a "land use"
issue and included as a subtopic of Question 2 (land
use). 3) The topic encompassed by Question 3 has
overlap with the fundamental question regarding
groundwater in Chapter 3, and there is a need for an
explanation of integration among components of the
Report in the introduction. 4) The agency may wish
to list agriculture explicitly as the focus in Question
4. An alternative would be to include agricultural
land indicators under Question 2 (addressing land
use), considering agriculture as a specific land use. 5)
The factors distinguishing Question 5 (addressing
contaminated land) from Questions 3 and 4 should be
explained more fully.
EPA should include more direct indicators of effects
in the land chapter.
EPA should consider adding indicators for mining
wastes as well as animal and other wastes applied on
agricultural land.
EPA should add an indicator based on the generation
and disposal of civilian radioactive waste.
A pesticide use indicator should be added to the land
chapter.
The pesticide residues in food and reported pesticide
incident indicators should be moved to the human
health chapter.
F-6

-------
Table F-l. Recommended Changes in the Final Report
Chapter
Page
Recommendation
Land
32
The discussions of the data gaps in the land chapter
should be modified to make it clear that the gaps
mentioned are the highest priority gaps determined by
the agency, and that the list is not intended to be
comprehensive.
Human Health
34
The questions within the human health chapter
should be reordered to be consistent with event
sequence in the environmental health paradigm as
depicted in Figure 5.1 of the Report (i.e., exposure
precedes  the health effect).
Human Health
34
The human health chapter should be more
descriptively renamed as "Human Exposures and
Health."  This change is needed because the questions
contained within the chapter encompass both human
health and exposure.
Human Health
35
If credible quantitative impact estimates are available
(e.g., estimates of the mortality or morbidity impacts
of paniculate air pollution in selected locations in the
U.S.), they should be included.
Human Health
36
The discussion of gaps and limitations should be
expanded to include a more quantitative description
of the indicator's relevance by relying on the
epidemiologic literature.
Human Health
36
The concept statements in the indicator limitations
sections such as "the measurement of mercury or any
other environmental chemical in a person's blood or
urine does not by itself mean that the chemical has
caused or will cause harmful effects in that person"
should be removed from each discussion of indicator
gap and instead be placed in the conceptual
framework section of the chapter.
                                      F-7

-------
Table F-l. Recommended Changes in the Final Report
Chapter
Page
Recommendation
Human Health
37
EPA should build on the higher geographic resolution
theme by presenting individual or multiple state data
which could inform the gross national estimates
presented and point toward the future. This should be
done if possible, given the time constraints of
revising this version of the ROE.
Human Health
38
Bullet #2 on page 5-5 should be rewritten to include
biological agents. The following sentence should be
added: "Infectious diseases associated with
environmental exposures or conditions are also
addressed."
Human Health
38
Expanded health indicator information should be
provided for sensitive populations. Expanded
information is needed because these populations are
important in considerations of environmental health.
Ecological
Condition
39
The climate indicator trends in the ecological
condition chapter should be placed in a paleoclimatic
context to distinguish between human induced
changes and other long-term changes.  References to
the Report of the Intergovernmental Panel on Climate
Change (IPCC, 2007a,b) should be included.
Ecological
Condition
39
A question should refer to trends in exposure and
effects of contaminants in organisms rather than
focusing on trends in biomarkers.
Ecological
Condition
40
EPA should reorganize the ecological condition
chapter to focus on three major indicator categories:
Ecosystems, Ecological Processes and Services, and
Ecosystem Components.
Ecological
Condition
40
Appropriate indicators should be included in the
ecological condition chapter to provide information
on the ecosystem extent (e.g., land cover, land use,
urbanization) and quality /condition (e.g., landscape
integrity, connectedness, fragmentation, and
contamination) of major ecosystem types.
                                       F-8

-------
Table F-l.  Recommended Changes in the Final Report
Chapter
Page
Recommendation
Ecological
Condition
40
Indicators should be included in the ecological
condition chapter to represent important ecosystem
processes and services such as: provisioning (e.g.,
timber, fuel, minerals, and other services); regulating
(e.g., disease, climate, and flood processes); cultural
(e.g., spiritual and aesthetic services); and supporting
(e.g., soil formation, primary productivity,
pollination,  decomposition, disturbance, nutrient
cycling, hydrological/chemical cycling, carbon
sequestration processes, and services such as clean
air, clean water, and net production).
Ecological
Condition
40
Indicators should be included in the ecological
condition chapter to represent physico-chemical
components of ecosystems (e.g., soils, water,
chemicals, snow pack, and physical habitats).
Ecological
Condition
41
Indicators should be included in the ecological
condition chapter to represent biological components
of ecosystems ranging from the genome to the
community level of organization.  Such components
include biodiversity, endangered species, invasive
species, keystone species, and communities.
Ecological
Condition
42
The discussion of "trends in diversity and biological
balance of the nation's ecological systems," (on page
6-29) should acknowledge that some systems
inherently have different numbers and variety of
species, making it inappropriate to make comparisons
between systems.
Ecological
Condition
42
The discussion offish faunal intactness," should
explain why 1970 is chosen as the reference year.
                                       F-9

-------
Table F-l. Recommended Changes in the Final Report

Chapter
Ecological
Condition
Ecological
Condition
Page
42
43
Recommendation
Trend data should be adjusted to account for
methodological inconsistencies. For example, in the
discussion of "SAV in the Chesapeake" which shows
trends since 1978, the Report on the Environment
states that "methods changed over the course of this
study. However, data have been adjusted to account
for any methodological inconsistencies." The same
should have been done with other parameters that are
presented as a snapshot at one time that could have
shown trends.
It should be clearly stated that specific case studies in
the Report may not be representative of a general or
national situation.
F-10

-------
Appendix G:  Table of Recommended Improvements for Future Reports on the
Environment

Table G-l presents a summary of recommendations that should be considered to improve
future Reports.  The second column of the table provides page numbers where
recommendations may be found this advisory report.  Additional comments and
suggestions are provided in the text of this advisory report. Detailed comments
pertaining to specific indicators are included in Appendix A.
Table G-l.  Recommended Improvements for Future Reports on the Environment
Chapter
Page
Recommendation
Overarching
                   A synthesis chapter should be included to fully
                   integrate the Reports and to provide an overall
                   assessment of health and ecosystem status, trends and
                   effects. The synthesis chapter in future reports could
                   also analyze and discuss in more detail the
                   connections among various related indicators.
Overarching
                   A summary section should be included after each
                   media chapter to summarize information presented in
                   the chapter and identify relevant emerging issues that
                   could potentially affect human health and the
                   environment.
Overarching
                  EPA should incorporate statistical analysis and
                  interpretation in the reporting of all indicators. This
                  should be part of the results presentation for each
                  indicator. In some cases, this may involve formal
                  statistical analyses, whereas in other cases it may
                  involve the inclusion of additional information such
                  as error bars around mean values.
Overarching
                   The indicators selected should be clearly related to
                   the "big picture" fundamental questions, and not
                   chosen just because of data availability or compliance
                   with indicator criteria (i.e., they are the only
                   indicators left after others have been eliminated).
Overarching
                  EPA should consider relaxing the restrictive indicator
                  selection criteria so that additional indicators can be
                  included.  This will enable EPA to better evaluate
                  trends and answer questions in the Report.
                                       G-l

-------
Table G-l.  Recommended Improvements for Future Reports on the Environment
Chapter
Page
Recommendation
Overarching
11
Each of the sections that address data gaps and
limitations should be separated into clear discussions
of types of limitations (e.g., geographic, statistical,
data coverage, etc.)
Overarching
12
EPA should analyze the air, water, land, human
health, and ecological condition indicators using
appropriate airshed, watershed, and ecoregional units.
Overarching
13
EPA should identify and use, with appropriate
caveats, more regional indicators and data bases to
illustrate trends when national data sets are not
available. The Panel notes, however, that such
regional data are not a substitute for national or even
representative national data and can be misleading if
not carefully presented.
Overarching
13
EPA should develop clear and transparent criteria that
are uniformly used for the selection of regional
indicators and case studies, with the recognition that
not all data will meet the criteria for these regional
indicators. For example, regional indicators should
have long-term well supported data sets, be of
particular national or local significance, or represent
an assessment approach that that could be replicated.
Air
14
The discussion provided in the response to the indoor
air quality question should be expanded. The
discussion of indoor air and related indicators is too
limited considering the importance of the indoor
environment and the amount of time spent by the
population indoors. While indoor environments do
not fall within the statutory mandate of EPA,
exclusion of available and relevant data makes the
Report incomplete.
                                       G-2

-------
Table G-l.  Recommended Improvements for Future Reports on the Environment
Chapter
Page
Recommendation
Water
20
Additional questions should be included in the water
chapter to incorporate missing information on critical
habitats or thematic elements such as: 1) Extent and
condition of coral reefs; 2) Wastewater management
information (it is recommended that EPA review
available National Pollution Discharge Elimination
System data for possible useful indicators); 3) Extent
and condition of, and trends in, riparian zones and
lake shoreline (i.e., land-water interface, where much
of the biological activity occurs), and their effects on
human health and the environment; and 4) More
national indicators and analyses providing data and
information on non-indigenous invasive species.
Water
20
Some key model aquatic systems should be identified
in several ecoregions of the U.S. and data collected
from these systems should be mined and analyzed in
the context of questions presented in the Report.
Water
20
EPA should examine the 2004 National Research
Council Report on national and global water
resources and water infrastructure problems, and the
importance of research in addressing them (National
Research Council, 2004).
Water
23
EPA should include appropriate indicators of
condition of lakes, ponds, and reservoirs.
Water
23
EPA should consider including the following
important specific indicators:!) Snow pack (extent,
condition, and volume); 2) Pathogens (coliforms,
enteric viruses, toxins, etc.); 3) Storm water and
wastewater (contaminant effects); 4) Drinking water
primary contaminants (e.g., microbial indicators and
pathogens: bacterial, viral or protozoan); 5) Emerging
contaminants such as pharmaceutical and personal
care products, nanoparticles, and others.
Water
23
Additional wetland data should be used. In many
areas, wetlands will indicate more efficiently the
ecological integrity of the entire watershed than will
any other portion of the landscape.
                                       G-3

-------
Table G-l.  Recommended Improvements for Future Reports on the Environment
Chapter
Page
Recommendation
Water
23
EPA should evaluate whether nutrient indicators
based on bioavailable nitrogen and phosphorus or
nitrogen:phosphorus ratios may be more useful.
Water
23
EPA should develop drinking water indicators based
on the available data from the Agency's own
databases and the consumer confidence reports
released to the public annually by community water
systems.  Based on these data, EPA could formulate
indicators that can delineate trends in drinking water
quality. The water chapter should include source
water monitoring data in addition to treated water
quality data.
Water
24
Pathogen monitoring should be investigated as a
primary indicator for water quality trends and human
health effects across various water sources.
Water
24
Composite or multi-metric indicators should be
complemented with single metric indicators that are
easier to understand and require fewer caveats and
assumptions.
Water
24
EPA should incorporate more information on specific
toxic industrial chemicals for which the Agency has
statutory responsibility under the Clean Water Act.
Water
24
EPA should analyze fish tissue contaminant data by
different species, or at least conduct separate analyses
offish from different trophic levels or different
habitats (as was done for the "lake fish tissue"
indicator) to see which species (e.g., piscivores) are
more likely to have higher levels of contaminants
than others.
Water
 26
EPA should visit (or revisit) the Agency's guidance
manuals for lakes, rivers, coastal waters, and
wetlands for potential data sets to fill identified data
gaps.
                                       G-4

-------
Table G-l.  Recommended Improvements for Future Reports on the Environment
Chapter
Page
Recommendation
Water
26
Long-term monitoring programs of EPA (e.g.,
Environmental Monitoring and Assessment Program
- EMAP) and other Federal Agencies (e.g., the U.S.
Geological Survey's National Water Quality
Assessment Program, and the National Oceanic and
Atmospheric Administration's Status and Trends and
Mussel Watch Programs), and of states or universities
should be examined. Indicator criteria should be
relaxed (within reason) to enable use of important
trend data.
Water
27
EPA should utilize and build on existing databases
that have been collected and existing local expertise
that has been developed at benchmark sites in various
ecoregions.
Water
28
EPA should give state data sets much closer scrutiny
for possible inclusion. Some states have a wealth of
area-specific data.
Water
28
EPA should consider the following as an example
potential local/regional indicator for use in the water
chapter. The State Water Resources Control Board of
California is funding USGS to lead and conduct a
Ground-Water Ambient Monitoring and Assessment
(G.A.M.A.) program (U.S. Geological Survey, 2008)
under which groundwater samples from public and
private water supply wells from California are
analyzed for water quality.
Water
28
In addition to the Gulf of Mexico and Long Island
Sound, other places where hypoxic conditions tend to
occur and are well monitored (such as Chesapeake
Bay, the coastal waters off Oregon, and parts of Lake
Erie) should be added to the hypoxia indicator.
Water
29
EPA should develop indicators for arid regions.  In
this regard the Agency should draw upon the
numerous studies and data collection efforts
conducted by various federal and state agencies in the
western states where the climate is arid.
                                      G-5

-------
Table G-l.  Recommended Improvements for Future Reports on the Environment
Chapter
Page
Recommendation
Water
29
EPA should consider identifying the following
emerging issues in the summary section of the water
chapter: 1) Effect of climate change on water quantity
and quality, 2) Emerging pathogens associated with
climate change, 3) Chemicals of emerging concern,
4) Nanoparticle waste products, 5) Water availability
and sustainability, 6) Invasive species, and 7) Algal
toxins.
Land
30
EPA should consider adding a fundamental question
on soil quality and conservation to the land chapter.
The structure of the question could be parallel to the
others in the chapter.
Land
32
EPA should: 1) consider a range of land cover
classification schemes with different levels of
resolution.  This is necessary because the resolution
of the data in the current Report is too coarse to
completely answer the questions; 2) characterize land
cover of all major ecosystem types, not just the forest
land cover characterized the current draft of the
Report; 3) adopt standard, established approaches for
land use and land cover analysis to evaluate
information and document trends across a range of
available data sets.
Land
33
Different types of regional groupings could be used
to show the location and extent of features in various
chapters of the Report.  For example, ecologically
relevant units such as watersheds, climatic provinces,
and major coastal realms could be used to regionalize
data.
Human Health
35
EPA should consider using an expanded suite of
human health indicators (discussed in Section 9.0).
Human Health
36
EPA should adopt the suites of indicators that other
agencies have developed, but present them in relation
to environmental factors.
                                       G-6

-------
Table G-l.  Recommended Improvements for Future Reports on the Environment
Chapter
Page
Recommendation
Human Health
37
EPA should consider making use of county-level data
available from the states.  All of the vital statistic data
presented and used for the EPA Regional indicators
can and have been scaled to the county level and
excellent maps have been generated and already
published in books.
Ecological
Condition
42
EPA should use available information from the
Agency's water quality criteria guidance manuals.
The Panel notes that EPA has previously conducted a
detailed review of current information and developed
water quality criteria guidance manuals for lakes,
rivers, and coastal waters.
Ecological
Condition
44
Specific case studies using regional indicators should
be selected for their ability to demonstrate the long-
term trends that cannot be accomplished at the
national level.  It would be useful to pick well-studied
sites (e.g., Lake Mendota, Lake Tahoe) where there
are long-term data sets available for each region.
Ecological
Condition
44
Some of the regional indicators should be expanded
to become national indicators (e.g., SAV, invasive
species, and harmful algal blooms).
                                       G-7

-------