May 1990
&EPA
United S tates Office of Water Regulations Of fir- of Policy, Planning
Environmental Protection and Standards and Evaluation
Agency
FEASIBILITY REPORT ON
ENVIRONMENTAL INDICATORS
FOR SURFACE WATER PROGRAMS
-------
CONTENTS
FIGURES i
TABLES iv
ABSTRACT v
ACKNOWLEDGMENTS vii
ACRONYMS ix
I. INTRODUCTION 1
II. DESIGNATED USE SUPPORT AND ATTAINMENT OF
"FISHABLE/SWIMMABLE" GOALS 11
III. SHELLFISH HARVEST AREA CLASSIFICATIONS 31
IV. TROPHIC STATUS OF LAKES 51
V. TOXICS IN FISH AND SHELLFISH 63
VI. BIOLOGICAL COMMUNITY MEASURES 81
VII. POLLUTANT LOADINGS FROM POINT SOURCES 105
VIII. SUMMARY AND CONCLUSIONS 119
-------
FIGURES
Page
Figure 1-1 Continuum of Indicators 4
Figure II-1 National Summary of CWA Fishable Goal 1988 21
Figure II-2 Nationwide Summary of CWA Swimmable Goal 1988 22
Figure II-3 Nationwide Designated Use Support 1988 23
Figure II-4 Designated Use Support - Rivers (EPA Region 4) 24
Figure II-5 Percentage of Assessed River Miles Per State Meeting
CWA Fishable Goal in 1988 25
Figure II-6 Percentage of Assessed River Miles Per State Meeting
CWA Swimmable Goal in 1988 26
Figure II-7 Percentage of Assessed River Miles Per State Fully
Supporting Designated Use in 1988 27
Figure II-8 Percentage of Total River Miles Assessed Per State in 1988 28
Figure II-9 Percentage of Assessed River Miles Fully Supporting
Designated Use in 1988 - EPA Region 1 29
Figure III-l Shellfish Harvest Area Affected by Pollution Sources -
Northeast Region (1988) 37
Figure III-2 Shellfish Harvest Area Affected by Pollution Sources -
Mid-Atlantic Region (1988) 38
Figure III-3 Shellfish Harvest Area Affected by Pollution Sources -
Southeast Region (1988) 39
Figure III-4 Shellfish Harvest Area Affected by Pollution Sources -
Gulf of Mexico (1988) 40
Figure III-5 Shellfish Harvest Area Affected by Pollution Sources -
West Coast Region (1988) 41
Figure III-6 1985 National Shellfish Harvest Area Classifications -
Subdivided by Regions 44
Figure III-7 Shellfish Harvest Area Reclassifications Due to Water
Quality Changes - Northeast Region (1971-1985) 46
Figure III-8 Shellfish Harvest Area Reclassifications Due to Water
Quality Changes - Mid-Atlantic Region (1971-1985) 47
-------
FIGURES (Continued)
Page
Figure HI-9 Shellfish Harvest Area Reclassifications Due to Water
Quality Changes - West Coast Region (1971-1985) 48
Figure IV-1 Trophic Status of Lakes By Number of Lakes Assessed (1988) 60
Figure IV-2 Trophic Status of Lakes By Lake Acreage Assessed (1988) 61
Figure IV-3 Percentage of Assessed Lakes Per State Classified
as Eutrophic (1988) 61
Figure IV-4 Percentage of Assessed Lake Acreage Per State Classified
as Eutrophic (1988) 62
Figure V-l National Contaminant Biomonitoring Program (NCBP) -
PCB Residues in Freshwater Fish (1980-81) 71
Figure V-2 National Contaminant Biomonitoring Program (NCBP) -
Geometric Mean Concentrations of p, p'-DDT Homologs in Fish
Samples (1969-81) 72
Figure V-3 Total DDT in Liver of Estuarine Fish Composites Collected
at 42 Sites in 1984 73
Figure V-4 Total PCBs in Mollusks - Top 20 Sites
for 3 Year Period (1986-1988) 75
Figure V-5 Example of a Translated STORET Retrieval 78
Figure VI-1 Approach Used in River Biosurvey Monitoring Programs (1989) 93
Figure VI-2 Communities Sampled in River Biosurveys (1989) 94
Figure VI-3 Illinois Streams Evaluated Using Index of Biotic
Integrity (ffil) . Color (1989) 97
Figure VI-4 Illinois Streams Evaluated Using Index of Biotic
Integrity (IBI) - Black and White (1989) 99
Figure VI-5 Illinois Streams Evaluated Using Macroinvertebrate
Biotic Index (MBI) - Color (1989) 101
Figure VI-6 Illinois Streams Evaluated Using Macroinvertebrate
Biotic Index (MBI) - Black and White (1989) 103
11
-------
FIGURES (Continued)
Figure VII-1 Availability of PCS Data for Making Point Source
Loading Estimates - Percent of DMR Forms from
Major Facilities Entered into PCS (4th Quarter of FY 1989) 111
Figure VII-2 PCS Example - Point Source Loading Estimates (1987-1989)
For Carbonaceous Biochemical Oxygen Demand (CBOD) 115
Figure VII-3 PCS Example - Three-Dimensional Map Displaying
Total Suspended Solids (TSS) Loadings Across Illinois (1988) 116
111
-------
TABLES
Table 1-1
Table H-l
Table ffl-1
Table m-2
Table IV-1
Table IV-2
Table V-l
Table VI-1
Table VI-2
Table VHI-1
Page
Workshop Results 9
Designated Use Support - Rivers 1986 & 1988 (37 States) 17-18
Shellfish Harvest Area Reclassifications Due to Water Quality
Changes - Northeast Region (1971-1985) 45
Shellfish Harvest Area Reclassifications Due to Water
Quality Changes - Subregions of Buzzards Bay, MA (1971-1985) 49
Trophic Status 1986 & 1988 (22 States) 55-56
States Using Different Trophic Status Reporting Methods 59
Total PCBs in Mdllusks - Top 20 Sites for
3-Year Period (1986-1988) 74
Comments Regarding the Utility of State Biocommunity Data
as a National Indicator - Results of TBS Interviews 90
Biocommunity Contacts by State 92
Summary Characteristics of Proposed Indicators 120
IV
-------
ABSTRACT
This report presents the results of a feasibility study on six environmental
indicators of the nation's surface waters: designated use support and attainment of
"fishable/swimmable" goals, shellfish harvest area classifications, trophic status of lakes,
toxics in fish and shellfish, biological community measures, and pollutant loadings from
point sources. The use of these and other environmental indicators is becoming an
increasingly important evaluation tool for State and federal environmental managers.
Important applications of environmental indicators include identifying trends over space
and time, evaluation program effectiveness, targeting resources to highest-risk problems,
targeting resources to activities achieving greatest risk reductions, and communicating
results to the public and legislators. This report represents the second phase of a three-
phase project conducted by the Office of Water Regulations and Standards (OWRS) and
the Office of Policy, Planning and Evaluation (OPPE) to identify, evaluate and determine
applications for environmental indicators of the nation's surface water resources.
Following the introduction, separate chapters are devoted to discussing the
feasibility of each of the six environmental indicators. Strengths, weaknesses and
possible improvements for each of the indicators are contained in these chapters and
summarized at the end of the report.Evaluation criteria included data availability, data
consistency/comparability, spatial and temporal representativeness of data, utility in trend
assessment, relationship to ultimate impact, scientific defensibility, sensitivity to change,
relationship to risk, cost of data collection and analysis, relationship to existing programs
and presentation value.
Important conclusions and recommendations from this study include: (1) The
shellfish harvest area classification data available from the National Oceanic and
Atmospheric Administration (NOAA) could be incorporated into an EPA indicator
reporting process in the near term. (2) It would be most efficient and logical for OW to
use the State 305(b) reports as the primary vehicle through which it develops data on
indicators. (3) The consistent use of the Waterbody System and of individual reach
numbers by the States should be encouraged. (4) In the long-term, there are some
additional monitoring and coordination activities that the Agency, other Federal agencies,
and the States might consider in order to develop more meaningful indicators. (5) EPA
should actively encourage State programs designed to implement measures of biological
community well-being.
v
-------
vi
-------
ACKNOWLEDGEMENTS
This document was prepared for the U.S. Environmental Protection Agency, Office
of Policy, Planning and Evaluation (OPPE) and the Office of Water Regulations and
Standards (OWRS), by Temple, Barker & Sloane, Incorporated (TBS). The information
in this document has been funded in part by the United States Environmental Protection
Agency under EPA Contracts 68-01-7288, 68-23-3548, 68-C8-0040, and 68-W9-0080.
This report presents the results of a feasibility study conducted by OPPE and OWRS at
EPA Headquarters. The work was accomplished by the time and effort of Kim
Devonald, Joe Abe, Kristina Groome, Tom Born and Eric Hyatt of OMSE, Bruce
Newton, Wayne Praskins and Chris Faulkner of OWRS, Dan Farrow of the National
Oceanic Atmospheric Administration (NOAA) and Andy Schwarz, Tom Flanigan and
Sarah Morrison of TBS.
vii
-------
viii
-------
ACRONYMS
BIOS A component of STORET which stores biological data
BMAN Benthic Macroinvertebrate Ambient Network
BOD Biochemical Oxygen Demand
BSC Biological Stream Classification system
CWA Clean Water Act
DDT Dichloro-Diphenyl-Trichloro-ethane
DMR Discharge Monitoring Report
EDS Effluent Data Statistics
EMAP Environmental Monitoring and Assessment Program
EPA Environmental Protection Agency
EPT total number of Ephemeroptera, Plecoptera and Trichoptera in a sample
FDA Food and Drug Administration
FWS Fish and Wildlife Service
GAO Government Accounting Office
GIS Geographic Information System
IBI Index of Biotic Integrity
ICI Invertebrate Community Index
IWB Index of Well-Being
MBI Macroinvertebrate Biotic Index
NASQAN National Stream Quality Monitoring Network
NAWQA National Water Quality Assessment program
NCBP National Contaminant Biomonitoring Program
NCC National Computer Center
NOAA National Oceanic and Atmospheric Administration
NPDES National Pollutant Discharge Elimination System
NSSP National Shellfish Sanitation Program
NST National Status and Trends program
OPPE Office of Policy, Planning and Evaluation
ORD Office of Research and Development
OW Office of Water
OWRS Office of Water Regulations and Standards
PAHs Polycyclic Aromatic Hydrocarbons
PC Personal Computer
IX
-------
ACRONYMS (Continued)
PCBs Polychlorinated Biphenyls
PCS Permit Compliance System database
PIBI Potential Index of Biotic Integrity
POTW Publicly Owned Wastewater Treatment plants
RTI Research Triangle Institute
SIC Standard Industrial Classification
STORET STOrage and RETrieval (EPA's computerized water data base)
STPs Sewage Treatment Plants
TBS Temple, Barker & Sloane, Inc.
TOC Total Organic Carbon
TOXNET TOXicology data NETwork
TRI Toxic Release Inventory
TSI Trophic Status Index
TSS Total Suspended Solids
USGS United States Geological Survey
WBS Waterbody System
WQI Water Quality Index
-------
I. INTRODUCTION
This report presents the results of a study on the feasibility of six measures
identified as potential environmental indicators of the quality of the nation's surface
waters. The feasibility analysis is the second portion of a three-phase project jointly
managed by the Office of Water Regulations and Standards (OWRS) and the Office of
Policy, Planning and Evaluation (OPPE) at EPA. The first phase consisted of identifying
and describing a series of potential indicators for freshwater, estuarine and coastal
environmental quality, and holding a workshop of federal and State personnel to review,
revise and narrow down the candidate indicator list. Three reports were completed in
Phase One: Resource Document for the Workshop on Environmental Indicators for the
Surface Water Program (March 28-29. 1989). Workshop on Environmental Indicators for
the Surface Water Program (March 28-29. 1989). and Results: Workshop on
Environmental Indicators for the Surface Water Program (July 1989). In the second
phase, contractors and EPA personnel assessed the feasibility of reporting on the set of
indicators selected at the workshop. These were selected as most meaningful and
practical for one or more of the following purposes: status and trend reporting; overall
water program evaluation, and evaluation of the effectiveness of individual program
components (e.g., point source regulation, or toxic chemical controls). The present report
addresses questions relating to data availability, and the degree to which the proposed
measures meet the criteria of a "good" indicator, and which of the possible "uses",
described further below, is met by the measure. In the third phase of the project, EPA
and State personnel will develop options and recommendations for specific applications
of the indicators by States, Regions, or EPA Headquarters.
Purpose of the Project
The use of environmental indicators is becoming an increasingly important
evaluation tool for federal and State environmental programs. Carefully chosen indicators
of surface water quality can help answer two fundamental questions:
• What is the quality of surface waters, and
• How are we doing in our efforts to improve it?
Environmental managers at EPA and elsewhere can use indicators for several
specific purposes related to these general questions, including:
• Identifying trends over time and space;
• Evaluating program effectiveness;
• Targeting resources to areas with greatest environmental impact;
• Targeting resources to areas of potential or developing problems; and
• Communicating results to the public and legislators.
-------
I. INTRODUCTION
A primary goal of this project is to identify existing data sets that can be used to
develop indicators of surface water quality. If Samuel Taylor Coleridge were describing
the plight of the modern day water quality manager, he might say that there are data,
data, everywhere, but not a trend to see. There is a large amount of water quality
information gathered by EPA, State Agencies and other Federal Agencies. However, as
EPA's Office of Water (OW) reported in its 1987 study; Surface Water Monitoring: A
Framework for Change, it is not collected or stored in a way that facilitates its use to
identify trends. Ward refers to this as a "data-rich but information-poor" syndrome
(Ward 1986), noting that there has not been a sufficient linkage between water quality
monitoring and the related management activities.
The historical problems that EPA has had in making use of data available from
all of its monitoring programs have been addressed in a number of reports (see for
example GAO, 1986). A number of Agency activities are underway to resolve these
problems. There are currently several workgroups active at OWRS to develop technical
guidance to aid States in their efforts to improve the collection of biological monitoring
data and to help State managers use this information. In addition, the Agency is actively
working to make data more readily available at the national level through scheduled
improvements to the STORET database and to BIOS, a component of the STORET
system which will be used to store biological data, and through implementation of the
Waterbody System (WBS), the computerized database recently developed to facilitate
State reporting of information in biennial 305 (b) reports to Congress.
The development and identification of useful indicators from this project could
complement these activities through the identification of specific measures for which data
collection should be encouraged. This information could provide a framework to help
steer future monitoring activities and allow EPA officials to reach a consensus on the
ways to report information in water quality reports. A similar approach is being used in
the development of the Environmental Monitoring and Assessment Program (EMAP), a
program that is being designed by the EPA's Office of Research and Development
(ORD) to periodically assess the country's ecological resources hi all ecosystems
(terrestrial and aquatic). EMAP program activities include the evaluation, development,
and testing of environmental indicators and the design and evaluation of indicator-based
monitoring programs for collecting status and trend data. Major differences in the
proposed EMAP program and the current Surface Water Indicators development project
are timing and the relationship of the programs to on-going monitoring activities. EMAP
is intended to be a national, rigorous statistically designed monitoring program to begin
several years from now. The Surface Water Indicators project intends to identify types
of information already available or available with relatively minor modifications, from
existing monitoring activities. Reporting on these indicators could begin very soon, and
in some cases will already be occurring. Eventually EMAP data would supplement these
indicators.
-------
I. INTRODUCTION
Background
OWRS has long been involved in the process of providing Congress and water
quality managers, at EPA and elsewhere, with information on the quality of aquatic
resources. In the 1987 Surface Water Monitoring study, OW recommended that EPA
and the States should take actions that would meet the following objectives:
• Enhance State and EPA capabilities to carry out characterization, problem
identification, and trend assessment in inland, estuarine, and marine waters;
• Increase ambient follow-up monitoring to evaluate the effectiveness of
management decisions; and
• Promote the use of available water-related data in EPA and State decision-
making.
The use of a select group of environmental indicators could help respond to these
concerns. Environmental indicators are appropriately applied to the first objective, of
supporting problem identification and trend assessment. Systematic monitoring for
indicators of ambient conditions will improve feedback to program managers regarding
the effectiveness of their decisions, thus fulfilling objective 2. Developing these
indicators will both promote and benefit from the third objective, taking advantage of
existing data sets.
The Office's interest in developing environmental indicators has been recently
enhanced by the Agency-wide emphasis to adopt true environmental measures of
progress. This new emphasis is part of EPA Administrator William Reilly's risk
reduction and strategic planning process designed to increase the focus of EPA programs
on environmental results. As the Administrator noted in the May 1989 issue of the EPA
Journal,"[the good news is] based on my years in the environmental movement, as well
as my first four months as EPA Administrator, I believe EPA has the most talented,
most dedicated, hardest-working professional staff in the federal government. What's
more, I think this Agency does an exemplary job of protecting the nation's public health
and the quality of our environment. Now the bad news: I can't prove it." In
developing its strategic plans, each EPA program area is being asked to identify
environmental goals and to select appropriate indicators that will allow the Office to
track progress in meeting those goals.
There is a continuum of types of information, ranging from administrative or
activity measures to the evaluation of true health or environmental effects, that can
provide a framework against which the utility of various measures can be evaluated.
The various levels of potential indicators are shown in Figure 1-1 below. Each type of
indicator is valuable for certain purposes. For example, indicators at level 1 and 2 may
be more appropriate for evaluating the efforts of governmental managers than say those
at level 3 or 4 because of the time necessary for any changes to be reflected hi
environmental results. Data are more likely to be readily obtained for Level 3 and 4
indicators than for Levels 5 and 6. Ideally, measures would be available at levels 5 and
-------
Figure 1-1
Continuum of Indicators
ADMINISTRATIVE
INDICATORS
Level 1
Level 2
ENVIRONMENTAL
INDICATORS
Level 3
Level 4
Level 5
Non-assimilative Changes
(e.g., Habitat Alteration)
Preferred Data For Measuring Environmental Results
Level 6
Changes in
Uptake
and/or
Assimilation
Actions by
EPA/State
Regulatory
Agencies
Changes in
Discharge/
Emission
Quantities
Responses
of the
Regulated
Community
Changes in
Ambient
Conditions
Health Effects
Ecological
Effects
J
Other Adverse
Effects
I
-------
I. INTRODUCTION
6 since the government and the public are most interested in seeing positive results at
these levels.
Barriers to the Use of Indicators
Despite widespread acceptance of the concept of using environmental indicators,
their actual application is limited. There are a number of reasons for this. States do not
have excess financial or personnel resources that can be dedicated specifically for the
purpose of developing and reporting on environmental indicators. In the absence of
sustained management emphasis and central Agency coordination for reporting and using
environmental indicators, it has often appeared that the costs of monitoring, analyzing,
and reporting on the measures outweigh the potential benefits. The paucity of
legislative mandates to collect and report on environmental results contributes to this
perception. Managers and scientists are often hesitant to use available data, or data from
other sources, recognizing the difficulties in simplifying and summarizing the information
for use as an indicator. Finally, managers may fear that they will be held accountable
for changes in environmental conditions that they cannot control, and they may therefore
resist developing indicators.
Through EPA's new strategic planning process, some of these barriers may be
overcome. In particular, top management emphasis on environmental results-based
planning will provide strong support for indicator reporting. And a process encouraging
the use of environmental indicators to understand program results, and to influence
program planning, without a negative component involving evaluating personal
performance, will hopefully diminish resistance due to fear of accountability.
Use of Indicators in EPA Surface Water Programs
The 305 (b) reporting mechanism provides OW with a State-driven information
system that already serves as a source of indicator data and can be improved upon for
use in the future. If desired by OW, and agreed upon by the States, changes in the
reporting system could allow for more uniform collection of information needed to
develop selected environmental indicators. One of the major problems facing various
Offices at EPA in developing indicator programs will be in finding ways to compile and
analyze the information available from various sources. OW, through the biennial 305(b)
reports and the computerized waterbody system (WBS), already has such a system in
place.
As part of the 305(b) reports, States assess and report on then* waterbodies
according to their ability to support the designated uses established for all waters in their
water quality standards. States determine whether the designated uses are supported by
compiling and interpreting data on a variety of physical, chemical and biological
measures. Chemical and physical measures, corresponding to properties for which water
quality criteria have been adopted in State standards, are the most common measures to
evaluate use support, while biological measures are becoming more common. In addition
to the use support information, EPA also encourages the States to report on the extent to
-------
I. INTRODUCTION
which their waters are meeting the fishable and swimmable goals of the Clean Water Act
and to provide other information including shellfish harvest area closures, and waters
affected by fish kills and incidents of waterborae disease. EPA aggregates and
summarizes the State reports into a single national assessment which it submits to the
Congress.
In reporting on designated use support, States set water quality goals and measure
progress in meeting them. In addition, information provided on the causes and sources
of pollution allows managers to identify emerging and existing problems so that they
could potentially target their resources more effectively. As well as its use in a national
report to Congress, information reported is used by the States in identifying problems,
monitoring compliance actions, and setting control priorities.
Unfortunately, the current value of the 305(b) reports as a source of environmental
indicator data is severely limited. There are very large inconsistencies among States in
how water quality data are generated, analyzed, and reported. States assess different
subsets of their waters from one year to the next. In some instances, States even change
their accounting of total waters from one year to the next. One problem in using this
information for national reporting purposes stems from the considerable discretion that
States have under the law in developing their own water quality standards. These
standards establish environmental goals for individual water segments, the designated use,
and numeric or narrative criteria designed to ensure protection of the use. As a result of
these differences among States and in the type of information they provide to EPA in
their 305(b) reports, making comparisons between States or trying to assess national
status and trends is essentially impossible. And the inconsistencies in sampling design
from year to year make it difficult to assess trends even within individual States.
EPA currently does not require that States use reach numbers or other geographic
locators to identify specific water body sections in the 305(b) reports. If more States
provided this information, it would allow for cross-referencing with other data sets, such
as STORET, and would allow for information on pollution stress factors, such as
population, to be correlated with the site-specific information.
OW is working with the States to improve the reporting process and as noted
above, the 305(b) reports can provide an excellent framework for the future reporting of
indicator information. First, EPA is working to improve the consistency in designated
use support information and in the way States assess their total waters. In WBS, States
must establish a consistent set of water segment boundaries and continue to use this
segmentation scheme in subsequent reports. This should allow accurate trend analyses
based on subsets of the data that are spatially consistent from year to year. The new
computerized WBS has the capacity to generate statewide summary reports of use
support status for waterbodies and also to provide more detailed reports on sources and
causes of impairment for waterbodies. The standardized use of this system by all States
would also provide EPA with more flexibility in providing summary documents on a
nationwide basis. For example, States are asked to include assessment dates and whether
the assessment is based on monitored (collected data) or evaluated (visual assessments or
opinions of water quality professionals) information. Trend data based only on systems
that had been monitored since the last report might prove to be a more valuable tool
-------
I. INTRODUCTION
than that which can be derived from the statewide summary reports, which include many
sites that haven't been investigated over the time period for which the trend is being
determined.
Characteristics of a Good Indicator
The identification of those criteria that comprise a "good" indicator will depend to
a large degree on the specific purposes for which the indicator might be used. For
example, an indicator designed to provide information on nationwide status and trends
may have different characteristics than one designed to evaluate the effectiveness of a
specific, localized program. The following list identifies many of the criteria by which
indicators can be evaluated. They are not in any specific order of importance, but in
differing circumstances some are clearly more important than others. In section H-VII of
this report we will evaluate the feasibility of the various indicators and will discuss each
one in terms of the characteristics noted below as well as other aspects of data quality
and availability.
• Indicator is a measure of environmental conditions rather than
administrative actions
• The indicator is national in scope
• The indicator is spatially and temporally representative
• Indicator is based on information that is relatively easy and inexpensive to
obtain (which often means it uses data that are already being collected, or
collected partially, or collected in some locations, etc.)
• Indicator is understandable to nontechnical users including Congress, the
news media, and the general public
• The indicator can show incremental changes
• The indicator is scientifically defensible, sufficiently consistent across its
areas of use, and is relevant to the system on which it is used.
This list represents the conclusions reached at a workshop held in March of this
year regarding appropriate criteria for indicators of surface water quality. Additional
information on that workshop and the conclusions reached there are presented in the next
section.
Activities Leading to the Feasibility Study
As noted earlier, this feasibility study is the end of the second phase of this three-
phase project.
-------
I. INTRODUCTION
The first phase of the project involved the identification of existing data bases, at
EPA and elsewhere, that could be utilized to develop indicators. Information on more
than two dozen potential indicators was compiled into a report entitled: Resource
Document for the Workshop on Environmental Indicators for the Surface Water Program
(March 28-29"). which provided the substantive basis for a workshop held on March 28-
29, 1989. This document contains brief descriptions of each measure and a discussion of
its major advantages and disadvantages, its applicability to different categories of
waterbodies, and the availability of data to support it. In addition, a workbook entitled:
Workshop on Environmental Indicators for the Surface Water Programs (March 28-29
1989) helped to focus participants in their discussions.
At the workshop, participants representing State governments, EPA Regional
Offices, EPA headquarters and other organizations shared their views on the strengths
and weaknesses of potential indicators. The participants were divided into three
workgroups that attempted to answer the same questions, focusing on their particular area
of concern. Each workgroup attempted to define the following, focusing on the use of
indicators for reporting status and trends, for evaluating program effectiveness or for
evaluating the impacts of specific sources of pollutants:
• The audience for whom the indicators were being developed
• The objectives of the type of indicator that was being addressed;
• Specific criteria that applied to indicators that would meet these objectives;
and
• An identification of the particular indicators that met the objectives and
responded to the concerns of the identified audience.
A complete summary of the results of the workshop is presented in a document
entitled Results: Workshop on Environmental Indicators for the Surface Water Program
(July 1989). Table II-l highlights the recommendations of each of the groups.
In the second phase, OWRS and OPPE reviewed the workshop recommendations
and selected the seven most highly rated indicators from the workshop for further
evaluation and inclusion in the feasibility study. The seven chosen measures are:
• Designated use support and attainment of fishable/swimmable" goals
• Shellfish harvest area classifications
• Trophic status of lakes
• Toxic contamination in fish and shellfish
• Biological community measures
• Pollutant loadings from point sources
8
-------
Table 1-1
WORKSHOP RESULTS
GROUP 1
National Status/Trends
GROUP2
Program Effectiveness
GROUPS
Sources Of Pollution
Audience
Government
Public/Media
Scientific Community
Environmental Groups
EPA, Congress and other Federal Agencies
The Regulated Community
Public/Media
Environmental Groups
States
Decision-makers (Federal and State)
Public
Regulated entities
Objectives
Program Mangement
Outreach/Education
Problem Identification
Agency Accountability
Evaluate overall water quality programmatic
success on national and stale basis
Targeting
Evaluate success of programs in all environmental
media that affect water quality
Evaluate specific programs
Data checks on permits
Measures program effectiveness
Correlate problems with sources
Targeting resources
Criteria
National in Scope
Accurate
Understandable
Relevant
Timely
1. Generally applicable across all objectives:
Timely
Reference (background) values available
Consistent across area of concern
Flexible
2. Applicable for some objectives:
Direct measure of environmental conditions
Understandable to public
Sensitive to incremental changes
Flexible: information from various
collecting agencies can be used
Provides quantitative correlation
between source and pollution
Understandable
Relevant
Predictive
Potential
Indicators
4 TIERED APPROACH
Tier: Representative Indicator
Integrated: Designated Use
Chemical/Physical: WQI
Biological: Measure of community
structure/function
Administrative: Beach Closures
3 CATEGORIES
Category: Representative Indicators
General Program Evaluation
(Ecological Protection):
Biological Community Index;
Toxics in Rsh; Designated Use Support
General Program Evaluation
(Human Health): Toxics in Fish
Tissue; Shellfish Closures; Contact
Recreation Closures
Program-Specific Indicators:
Chemical/Physical Properties; Loadings;
Ambient Water Toxicity
Did not identify an indicator
that would link pollution to a
specific source: more research is
necessary
-------
I. INTRODUCTION
• Water quality indices based on physical/chemical data
This report presents results of that study for six of the seven recommended
indicators. The seventh indicator, water quality indices based on physical/chemical data,
is being evaluated by OWRS in a separate study.
In the third and final phase of the project, the contents of this report will be
reviewed by OW, OPPE, EPA Regional Water Divisions, States, and other interested
parties with a particular emphasis on identifying specific applications for the various
indicators. A final report will then be prepared that describes options and
recommendations for the use of indicators by States, Regions or EPA Headquarters.
10
-------
II. DESIGNATED USE SUPPORT AND
ATTAINMENT OF "FISHABLE/SWIMMABLE" GOALS
SUMMARY
Description of the Indicator
This indicator uses information provided by the States to EPA on the status of
their waterbodies. The information, included in reports required under section 305(b)
of the Clean Water Act (CWA), describes to what degree individual waterbodies are
meeting their designated uses. State and local governments assign designated uses to
waterbodies and determine the criteria by which they will be evaluated as part of their
water quality standards.
States also report to EPA on the extent to which their waters meet the "fishable
and swimmable" goals of the CWA. This differs from designated use support in the types
of information each State considers adequate to allow a determination (e.g. monitored
versus professional judgement).
Possible Applications
The State data are currently used in the development of a national assessment of
water quality. This is the States' and EPA's primary vehicle for informing Congress of
the state of the nation's waters. At the March 1989 workshop on developing surface
water indicators, the workgroup on status and trends recommended that "designated use
support" be used as a "high visibility" indicator to generate attention and solicit
questions about the underlying results.
Strengths National data collection and reporting
system is already in place and used by
all States (Recently Computerized).
"Fishable/Swimmable" reporting provides
information on goals of primary interest to
the public and is easily understood by
general public.
Weaknesses Since State reporting is inconsistent and not
standardized, measures cannot be used to evaluate
trends. Within a State, inconsistencies exist from
one reporting cycle to the next.
Possible improvements Increase use of the Waterbody System
to allow EPA to aggregate results; Increase
use of reach numbers or other geographic
identifiers to facilitate trend assessments;
Increase systematic ambient monitoring by States
(e.g. returning to certain subsets of waters
regularly to establish trends).
11
-------
H. DESIGNATED USE
Data availability
Data consistency/comparability
Spatial representativeness
Temporal representativeness
Utility in trend assessment
Relation to ultimate impact
Scientific defensibilitv
Sensitivity to change
Relationship to risk
Cost to collect and analyze data
EVALUATION CRITERIA
Good-States currently collect information on a
biennial basis and report it to EPA through the
305(b)reports.
Poor-No consistent basis for monitoring sampling
design from State to State and water quality
standards vary.
Falr/Poor-Although the same waters are not
assessed from one cycle to the next, the identified
bodies are representative of those types of waters
within a State.
Fair/Poor-It is primarily based on physical-
chemical data which are transient.
Poor-Since the same waters are not assessed from
one cycle to the next, it is impossible to determine
trends. Waterbodies can be reported as evaluated,
meaning the data from the last cycle is sometimes
just repeated in the next report.
Good-For Fishable/Swimmable since the degree to
which these goals are met is the ultimate impact.
Fair-For designated use since the exact impact of
the chemical/physical data are not known; Better if
biological community data are included in the use
attainment determination.
Fair/Poor-Methodologies vary from State to
State, therefore, the degree to which designated use
and fishable/swimmable determinations are
defensible as true measures of water quality is also
variable.
Poor-Categories are too broad to detect
incremental changes.
Falr/Poor-The relationship is tenuous for both
ecological and human health risks. The link from
chemical/physical to actual environmental damage
is ill-defined. Changes in fishable/swimmable do
relate to potential human health risk, but the
connection with designated use is less strong.
Low-States already engaged in data collection;
Improved collecting/reporting would require
incremental cost increases.
12
-------
H. DESIGNATED USE
EVALUATION CRITERIA
Relationship to existing programs Good-Required as part of 305(b) CWA
requirements.
Presentation value Good-Fairly easily understood by the general
public; Can be presented using maps and graphs.
13
-------
14
-------
H. DESIGNATED USE
DISCUSSION
Background
At present, the primary mechanism by which EPA assembles and reports on the
nation's surface water quality is through the biennial State reports called for under
Section 305 (b) of the CWA. As part of these reports, States assess and report on a
variety of designated uses. Under each State's water quality standards program, the State
designates uses for waterbodies and establishes numeric and narrative water quality
criteria the State determines are needed to protect each use. The extent to which the
assessed waters meet or fail the established criteria determines whether or not, and to
what degree, the designated use is considered to be met. States also report on whether
the use support decision is based on actual monitoring data or more subjective
evaluations.
Surface waters may be designated for one or more uses including, but not limited
to: domestic water supply; aquatic fish and wildlife support; recreation; agriculture;
industrial use; navigation; and nondegradation waters. States report on the degree to
which assessed waters are: "fully supporting", "fully supporting but threatened", "partially
supporting", or "not supporting" their designated uses. EPA also encourages States to
report on the extent to which the waters meet the "fishable" and "swimmable" goals of
the CWA. This determination can take into account information different from that used
in use support assessments, for example fishery closures. A smaller set of waters would
be expected to support fishable and swimmable goals than to support their designated
uses, because some designated uses are less demanding than full support of CWA goals
(e.g., industrial use).
Uses of the Indicator
At the March workshop, the workgroup charged with recommending indicators
that could be used to report on status and trends in the nation's waters identified the
need for a "high visibility" indicator that would lead people to ask questions and would
generate attention. The workgroup felt it was not essential that such an indicator be able
in itself to provide answers to the questions. With some reservations concerning the lack
of consistency among States in the assignment and evaluation of designated uses, the
workgroup recommended that a "designated use" measure be used for this high visibility,
integrative indicator.
The primary audience for this nationwide measure includes the Congress, for
whom EPA must aggregate all the State 305(b) reports into a national assessment known
as the "National Water Quality Inventory", and the general public. The major benefit of
this nationwide measure is its ability to provide information in a similarly expressed and
relatively easily understood manner on the status of all of the nation's assessed surface
waters. In addition, States and EPA Regional Offices can use the individual State 305(b)
15
-------
H. DESIGNATED USE
reports to highlight potentially threatened waters and to provide additional information on
the causes and sources of "non-attainment" of uses. The 305(b) report thereby serves as
a management tool for identifying problems and setting control priorities.
Characteristics of the Indicator
As noted above, these nationally applicable measures can provide the Congress,
the press, and the general public with some indication of the quality of the nation's
surface waters and a measure of progress toward attainment of the fishable/swimmable
goals of the Clean Water Act. However, inconsistencies and the lack of standardization
among the States, and within States from one year to the next, in how water quality data
are collected, analyzed, and reported make it difficult to use the data as an indicator of
national status and trends.
States typically conduct surveys of different waters each year, creating uncertainty
when analyses of statewide or nationwide trends are attempted. As can be seen in Table
H-1, the total river miles assessed can vary from year to year and in some instances
decreases from one year to the next. Even the accounting of the total river miles
existing in a State is sometimes changed from year to year due to natural or human
modifications of river reaches or differences in reporting procedures.
States have considerable discretion to develop water quality criteria for waters in
various use categories, and to define when their waters are "fishable" and "swimmable".
Moreover, States vary in the quantity, type, and quality of the data they compile and the
methods they use to interpret the data in determining whether each waterbody is
supporting its uses or the CWA goals. Since appropriate assessment tools will, in fact,
vary from one situation to the next, the differing methodologies for determining the
extent of use support is in some cases appropriate. However, it does make comparisons
between States uncertain.
Data Availability
One of the major advantages of this indicator is the commitment of the EPA to
continuing and improving the 305(b) reporting process and the resultant surety that this
type of information will continue to be collected. In addition, EPA has developed a data
system for managing water quality information to support 305(b) reporting for specific
waterbodies, the Waterbody System (WBS). Both mainframe and PC versions of the
WBS exist. The WBS has the capability to generate statewide summary reports of "use
support" status for each of seven waterbody types, and once a State develops a
segmentation scheme for its waterbodies, it will be expected to use the same
segmentation in all future reporting. Beginning with the 1990 reporting cycle, all States
will be asked to use the WBS or submit their assessments in a WBS compatible format
and designated use data will therefore be available on a more consistent basis in the
future. The WBS will allow EPA and the States to generate individual summary reports
listing assessment types and dates, "use support" status, and sources and causes of
impairment for each waterbody.
16
-------
Table II-1
Designated Use Support - Rivers 1986 & 1988 (37 States)
I-XS.-'-X^ifc.'. s
-•«*} 'suttee -
Alabama '«>
AL-88
Changes
Arizona '86
AZ'88
Changes
Arkansas '86
AR-88
Changes
California '86
CA'88
Changes
Connecticut '86
CT'SS
Changes
beleware '86
DE-88
Changes
Florida '86
FL'SS
Changes
Georgia '86
OA'88
Changes
Illinois '86
IL-88
Changes
IowaT86
IA'88
Changes
Kansas '86
KS-88
Changes
Kentucky '86
KV88
Changes
Louisiana '86
LA -88
Changes
Maine '86
ME -88
Changes
Maryland '86
MD'88
Changes
Massachusetts '86
MA '88
MI -88
Changes
Mississippi '86
MS '88
Changes
Missouri '86
MO*88
Changes
'^%,:P¥H * ;• "•
'JlSIJ'fofyiy&fir <"
40,600
40,600
0
17,537
6,671
-10,866
11,438
11.508
70
26,959
26.970
11
8,400
8,400
0
579
500
-79
9,320
12,659
3,339
20,000
20,000
0
14,080
14,080
0
18,000
18.300
300
20,600
19.791
-809
40,000
18.465
-21.535
14,180
14.180
0
31,672
31,672
0
9,300
9,300
0
10,704
10.704
0
36.350
36,350
0
10,274
15,623
5,349
20.536
19.630
-906
'„ *- "~JH!
-•^jtowwwh-'
12,101
11.174
-927
1,412
2,279
867
11,438
4,107
-7,331
9.627
9.885
258
880
880
0
516
467
-49
6.575
7.943
1,368
17,000
20,000
3,000
3,395
12,970
9,575
4,365
8,235
3,870
4,495
6,888
2,393
5,683
8,653
2,970
2,500
8,483
5,983
31,672
31,672
0
7,440
9,300
1,860
1,676
1.646
-30
36,350
36,350
0
10.274
15.623
5,349
20,536
19.630
-906
•, 'fBOp ' -'"*
S^pH-OB*""
10,83$
10,118
-717
615
1.583
968
5,914
1.714
-4.200
6.163
6.578
415
597
582
-15
309
280
-29
4,448
5.287
839
16.185
19,443
3.258
1.861
5.783
3.922
72
69
-3
3,512
3.994
482
3,130
6,176
3,046
1,240
5,730
4,490
30,695
31.278
583
6,852
8,635
1.783
802
713
-89
35,696
35.567
-129
9.260
13.850
4,590
10,390
10.147
-243
~jf •wpm&%.
OJtoWNHtfftg ••>'..••,:
80i
625
-179
391
207
-184
5.524
29
-5.495
1,518
2.219
701
232
239
7
184
156
-28
1,670
2.021
351
458
383
-75
1,457
172
-1,285
3,077
6,503
3,426
359
760
401
1,877
878
-999
. 800
2,146
1.346
513
0
-513
449
504
55
572
598
26
0
0
0
1,014
1.331
317
10,075
9.445
-630
S'?'$$$P*' V
462
431
-31
406
489
83
0
2.364
2.364
335
1,088
753
51
59
8
23
31
8
457
635
178
357
174
-183
77
186
109
1.216
1.663
447
435
2,134
1,699
675
1.599
924
460
607
147
464
394
-70
139
161
22
302
335
33
497
783
286
0
442
442
71
38
-33
Sources: 1986 1988 305(b) Reports
17
-------
Table IM (Continued)
Designated Use Support - Rivers 1986 & 1988 (37 States)
'^JWife'''^'
Montana '86
MT-88
Changes
Nebraska '86
NE*88
Changes
New Hampshire '86
NH-88
Changes
New Mexico '86
MM '88
Changes
New York '86
NY 'SS
Changes
North Carolina '86
NC-88
Changes
Ohio '86
OH '88
Changes
Oregon '86
OR '88
Changes
Pennsylvania '86
PA '88
Changes
Rhode Island '86
RI-88
Changes
South Carolina '86
SC*88
Changes
Sooth Dakota '86
SD-88
Changes
Tennessee '86
TN'88
Changes
Texas '86
TX-88
Changes
Vermont '86
VT'88
Changes
Virginia '86
VASS
Changes
West Virginia '86
WV'88
Changes
Wyoming '86
WV88
Changes
Ti*wm.S' „.,.,-
jHULm-': '"
€ii&n$te8 ,"' ••.
^•"?^ >J0HU f
'>, ;CWW
\ \^,<^I34
-'.:" .'..-.JJMWI
' ^/^Wtff" ' '=•#•
- QffiWtto&x''
12,184
12061
77
2,717
3044
527
981
950
-31
3.140
576
-2,564
2,667
53.394
50.727
25.156
22.375
-2,781
4,048
2056
-1,792
9,665
12,546
2,881
3.332
9.642
6,310
655
489
-166
2,127
2,824
697
1.865
1.387
-478
3,794
5,976
2,182
14,966
12,169
-2,797
882
4,534
3,652
948
uio
262
10.225
2.862
-7,363
17,386
16,080
-1,306
< «'r, '^TjyWfi
;; - w$*u&*
.;.r".:...r'.«M«K
. ^'"^jgnua^sr^
'" '^'-iSttUiHtfflttR *:' :•
6,ii4
6.630
-304
1.135
1002
67
259
210
-49
360
554
194
246
8.087
7.841
10,171
9,152
-1.019
1,977
1,501
-476
1,915
8,497
6,582
1042
1,770
528
34
14
-20
194
395
201
1.130
1060
130
1,118
2,484
1,366
0
0
0
269
379
110
1,536
1.401
-135
6,631
10.107
3,476
297
3,350
3.053
r\l:'??x}.^3&®6
e&v&'&lf&Xi&yA , x
K~$$»«S!ttBfc-^
3^87
614
227
942
1044
302
80
171
91
0
22
22
487
8,507
8,020
1,867
1,748
-119
603
3088
2,685
275
6,695
6,420
1,651
1,830
179
35
78
43
121
576
455
992
1.103
111
847
968
121
976
1.829
853
16
249
233
2O32
921
-1311
1,388
1.332
-56
1.972
7
-1.965
'sl'/^^^ww
Sources: 1986 1988 305(b) Reports
18
-------
H. DESIGNATED USE
Improving the Indicator
As well as implementing the WBS, EPA is also working to improve consistency
in designated use reporting and in State definition of total waters. The lack of
consistency among the States is the greatest impediment to effective use of the national
designated use measure to give an accurate picture of the nation's water quality and to
show temporal trends in this evaluation. In addition, EPA is reviewing trend assessment
methodologies used by States and others and hopes to develop recommendations on
future trend reporting. Work on other measures that can be used by States to assess
their waters, such as the biological measures or water quality indices discussed elsewhere
in this report, could also help States make more meaningful assessments. Encouraging
the uniform use of reach numbers or other forms of geographic locator, to identify
waterbodies for trend analysis and to correlate information from other sources, would
greatly increase the efficiency with which State data could be analyzed and used.
Presentation of the Indicator
The following figures demonstrate different ways in which information on
designated use and the meeting of fishable/swimmable goals can be shown. Figures
II-1 and II-2 show the degree to which the nation's assessed waters were meeting the
fishable and swimmable goals expressed as a percentage of the total waters. Figure
II-3 shows similar data for rivers, lakes, and estuaries in 1988 for designated use support.
One could develop similar graphs for 1986 and earlier years for comparative purposes.
However, given the uncertainties noted earlier in comparing these data from one year to
another, this kind of graphical analysis may be misleading.
Designated use and fishable/swimmable information can also be presented in terms
of the actual river miles and number and area of lakes and estuaries assessed. Figure n-
4 shows the 1988 assessments for river miles assessed in Region 4. This type of bar
graph conveys the greatest amount of information, showing the amount of each water
type assessed as well as the results of the assessments. However, when there is a large
discrepancy in the total river miles assessed in various States within a region, the
resultant differences in the height of the bars may make the graphic difficult to read.
Another way to present the national data is with a shaded map. Figures H-5 and
n-6 show, for each State, the percentage of assessed river miles that are fishable and
swimmable respectively. The two maps can highlight differences among States. For
example, in New York, a larger percentage of waters are "swimmable" than "fishable";
while the reverse is true for Iowa. Figure n-7 shows, in 1988, for each State, the
percentage of assessed river miles that fully supported their designated use.
Since States do not .assess all of their waters, it is important to note what
percentage of their waters are included in the assessments. Figure II-8 provides this
information for 1988 and, by the use of arrows, indicates if this is an increase or
decrease in assessment activities from 1986 levels. Finally, one Could show trends in
designated use attainment, (again acknowledging the caveats preventing use of data prior
19
-------
H. DESIGNATED USE
to the present date for trend analyses) by placing arrows on a map to indicate whether a
later year's data represented an increase or a decrease in the percentage of waters
supporting designated use, when compared to an earlier year. While there are not
sufficient data to support this now, it may be appropriate for the future. Figure n-9
shows how this might look, using 1986 and 1988 data for rivers in Region I.
20
-------
Figure 11-1
Nationwide Summary of CWA Fishable Goal 1988
100 -i
Percentage of
Total Waters
D Not Assessed
• Not Meeting
Meeting
Rivers
Lakes
Estuaries
Source: 1988 305(b) Reports
21
-------
Figure 11-2
Nationwide Summary of CWA Swimmable Goal 1988
100 i
Percentage of
Total Waters
Not Assessed
Not Meeting
Meeting
Rivers
Lakes
Estuaries
Source: 1988 305(b) Reports
22
-------
Figure 11-3
Nationwide Designated Use Support 1988
Percentage of
Total Waters
100 T
80 -
60 -
40 •
20 -
D Not Assessed
• Not Supporting
H Partially Supporting
H Threatened
H Fully Supporting
Rivers
Lakes
Estuaries
Source: 1988 305(b) Reports
23
-------
Figure 11-4
Designated Use Support • Rivers (EPA Region 4)
35000 T
30000 •-
River Miles
Assessed
25000 •
20000 -
15000 --
10000 ••
5000 -•
Not Supporting
Partially Supporting
Threatened
Fully Supporting
ALFLGAKYMSNOSCTN
Source: 1988 305(b) Reports
24
-------
Figure 11-5
Percentage of Assessed River Miles Per State Meeting CWA
Fishable Goal in 1988
Source: 1988 305(b) Reports
-------
Figure 11-6
Percentage of Assessed River Miles Per State Meeting CWA
Swimmable Goal in 1988
I States not Reporting
Source: 1988 305(b) Reports
-------
Figure 11-7
Percentage of Assessed River Miles Per State Fully Supporting
Designated Use in 1988
D
m
60-79%
Less than 60%
States not Reporting
Source: 1988 305(b) Reports
-------
Figure 11-8
Percentage of Total River Miles Assessed Per State in 1988*
D
States not Reporting
•States without arrows or equal sign did not report in 1986
•Year-to-year comparisons are for Illustrative purposes only.
Source: 1988 & 1986 305(b)
Reports
-------
Figure 11-9
Percentage of Assessed River Miles
Fully Supporting Designated Use in 1988
EPA Region
90 to 100%
80 to 89%
[~] 70 to 79%
F73 60 to 69%
Less than 60%
Decreased
Source: 1988 & 1986 305(b) Reports
*Year-to-year comparisons are for Illustrative purposes only.
29
-------
30
-------
III. SHELLFISH HARVEST AREA CLASSIFICATIONS
SUMMARY
Description of the Indicator
The indicator identifies potential contamination of coastal waters by investigating
the degree to which State governments close off or limit access to shellfish harvesting
areas. The State agencies classify different areas based on water quality monitoring (not
shellfish tissue monitoring). The resulting information constitutes one of the nations
largest consistently collected water quality data bases. States report this information to
NOAA as part of the National Shellfish Sanitation Program (NSSP), which provides
well-defined guidelines to the States. States do, however, vary in their interpretation of
the guidelines. This measure is one of five indicators chosen by NOAA in its state of the
marine environment report.
Possible Applications
Shellfish harvest areas are the most commonly monitored feature of coastal
waters and data are readily available. On both a nationwide and regional level, data
provide a good indication of the general status of marine waters and, with continued
improvements, can be used to assess trends.
Strengths
Weaknesses
Possible Improvements
Data collection has been consistent for over 20
years on a nationwide basis using National Shellfish
Sanitation Program (NSSP) guidelines. National
standards developed by FDA are used by all States.
Indicator is easily understood by public and policy
makers.
Variations in State to State decision-making on
classifications limit nation-wide comparisons
somewhat; Reclassifications are not always due to
water quality changes, but in past have reflected
changes in areas monitored; Only fecal coliform
levels are monitored (which are not bacteria of
concern, but are indicators of pathogens); The
indicator is not well reported for open coastal (as
opposed to estuarine) waters.
Greater consistencies in classifications would allow
for nationwide comparisons; Correlating with
other data, such as sediment and shellfish tissue
contamination would provide a more complete
indicator of surface water quality.
31
-------
. SHELLFISH
EVALUATION CRITERIA
Data availability
Data consistency/comparability
Spatial representativeness
Temporal representativeness
Utility in trend assessment
Relation to ultimate impact
Related factors
Good-NOAA publishes the National Shellfish
Register covering all continental coastal States
approximately every five years. Some States
include shellfish information in their 305(b)
Reports.
Fair-Even with inconsistencies in classifications
the data are consistently presented. NOAA
personnel visit States and take into account State-
to-State differences to standardize the data for the
national report. Physical and administrative
differences limit comparability of different regions
(East Coast-West Coast).
Good-The majority of estuarine areas are
classified as shellfish growing waters
(approximately 95% of East Coast estuarine
areas), and are consequently covered by the
indicator. (Open coastal waters are not very well
represented.)
Good-Most stations sample at a minimum of 5
times annually (conditional classifications more
often), therefore, seasonal variation is taken into
account. Reasonably consistent data sets exist for
most areas since the 1970's.
Fair-Data are available to assess trends, however,
not all changes in classification are the result of
water quality changes. NOAA does distinguish
changes which are the result of water quality from
those due to changes in areas monitored for
Northeast, Mid-Atlantic, and West Coast. These
data will allow for trend assessments.
Fair-Relationship between shellfish harvest area
classification and ultimate impact is limited
because only fecal conform levels are monitored.
Coliform levels do not relate directly to human
health impacts, rather they are an indicator of the
possible presence of pathogens.
Important-To have a more useful indicator,
the ancillary data on pollutant sources should be
used. (NOAA began collecting these data in the mid-
1980's.)
32
-------
. SHELLFISH
Scientific extensibility
Sensitivity to change
Relationship to risk
Cost to collect and analyze data
EVALUATION CRITERIA
Fair-Not all classitications are the results of
monitoring; Coliform levels are only an
indicator of pathogens; Quality of sampling varies
among states.
Good-Can get immediate reading of change in
coliform levels.
Fair-Difficult to relate coliform levels directly to
health risk, but indirect qualitative relationship
definitely exists; Monitoring primarily coliform
levels excludes factors other than sewage pollutants
related to risk; Not relevant to ecological risk.
Moderate-State monitoring programs vary in size
and cost though reporting to NOAA is well
established and consistent; Due to high cost, States
only monitor fecal coliform levels, monitoring
actual pathogens would be prohibitively expensive.
Relationship to existing programs
Presentation value
Good-NOAA's National Shellfish Register presents
shellfish harvest area classifications and assesses
status, trends and pollution sources. States use
classifications to assess designated use support, and
to target sewage treatment plant and combined
sewer overflow upgrade activities.
Good-Shellfish harvest area classifications are
understandable to government decisionmakers and
the public. Status and trends can be easily
presented on graphs and charts.
33
-------
34
-------
ffl. SHELLFISH
DISCUSSION
Background
To ensure that shellfish are safe for human consumption, coastal State agencies
(health or water quality) monitor and classify their potential shellfishing areas, using
national criteria and standards developed under the interagency National Shellfish
Sanitation Program (NSSP). (NOAA, FDA and EPA have played varying roles in the
NSSP since the 1970's.) Under the NSSP, FDA establishes standards for the growing,
harvesting and processing of shellfish, and using these standards, States conduct sanitary
surveys of shellfish growing waters. The classifications have been conducted for over 15
years and States base them on water quality data, and not on shellfish tissue
contamination. While States do vary in their application of these standards, NOAA is
working hard to improve the consistency of the reporting. NOAA maintains a national
data base of the State data, the National Shellfish Register, and prepares periodic reports
presenting comprehensive data on shellfish harvest area classifications nationwide. The
information constitutes one of the nation's largest consistently collected water quality
data bases. NOAA's most recent report, published in 1985, included data from 20.6
million acres of shellfish growing waters in all 22 coastal States.
The main criterion used to evaluate the different areas is the presence of coliform
bacteria, which are normal to the digestive tracts of humans and all warm-blooded
animals. The presence of these coliforms may indicate the presence of human sewage
and act as a surrogate for other pathogens, which do cause disease in humans and which
are more difficult to detect than the coliforms. However, the presence of coliforms does
not prove the presence of pathogens. Rather, it simply means they might be present.
States classify shellfish growing waters into one of four categories: approved.
meaning water quality permits harvesting at all times; conditional, meaning that water
quality is sometimes degraded and harvesting is allowed only when conditions are safe
(e.g. when it has not rained for several days); restricted, meaning that water quality is
degraded, but fishing is allowed if safety measures are taken (such as placing the
harvested shellfish in bacteria free water for a sufficient period of time prior to
marketing); or prohibited, meaning that water quality is too degraded for harvesting at
any time.
States also report on the significant pollution sources that affect harvesting
potential, by acre, for harvest-limited areas (those classified as: conditional, restricted or
prohibited), and in many cases, multiple sources may be reported. Pollution sources
listed in various reports include: sewage treatment plants (STP's), combined sewer
overflows, raw sewage discharges, septic systems, agricultural/urban/suburban runoff, and
boating. This information can help States target regulatory activities to meet the most
pressing needs.
NOAA has also recognized the value of this measure as an indicator of marine
water quality. Shellfish harvest area classification is one of five indicators recently
identified by NOAA for inclusion in its "State of the Marine Environment" report.
35
-------
. SHELLFISH
Uses of the Indicator
As noted above, States collect data on potential shellfish contamination to reduce
the risk of human disease from consumption. These shellfish can bioaccumulate large
concentrations of pollutants from the water column, often without suffering significant
harm (depending on the types of pollutants). They do, however, pose risks to humans
eating the shellfish. By far the most common health impact from contaminated shellfish
is gastroenteritis. Other far more serious illnesses are transmitted by shellfish on
occasion, including hepatitis and cholera. Impacts from toxic chemicals may occur, but
exposure and effects data are not readily available.
The identification of the presence of pollutants in the water column also provides
valuable information on the general state of the water resource. Much of the potential
shellfishing areas occur in estuaries, among the most productive ecological areas hi the
world. The inclusion of information on the causes of water quality degradation is
extremely helpful to water quality managers. These data, highlighting the causes for
shellfish closures across the country in 1988, are shown hi figures IH-1 through ffl-5.
This information can help managers identify problem sources and priority rank regulatory
and monitoring activities.
Information on shellfish harvest area classifications and causes of impairment or
improvement will provide EPA with a valuable, though not complete, measure of status
and trends in coastal environments.
Characteristics of the Indicator
States base their shellfish harvest area classifications on uniform national
standards. However, administrative decisions on how to apply the classification scheme
vary from State to State. For example, west coast States only classify "productive"
shellfish waters, while half of the approved shellfish waters on the East coast are
"nonproductive". This might account for much of the difference between the East and
West Coasts in the percentage of growing areas classified as approved.
States may also classify harvest-limited areas as "prohibited" because of resource
restraints. The conditional classification is sometimes not used because the State must
allocate additional resources to develop an area management plan. The plan would
include procedures to monitor sources of pollution and to prevent illegal harvesting.
States might choose not to classify waters as restricted due to the high cost to the State
and the fishing industry of purifying the shellfish before marketing (NOAA 1985).
Since changes in classifications may occur due to improved monitoring efforts
rather than to changes in water quality it has been difficult in the past to assess trends in
the water quality of classified waters . Between 1971 and 1985, less than half of the
changes in classification in the Northeast and Mid-Atlantic subregions could be related to
changes in water quality (NOAA 1989). In addition, some States have added new
waters that raise questions about the validity of comparisons with prior surveys. The
presence of a baseline number of acres assessed per State would improve the ability to
develop trend data. Realizing these difficulties, NOAA now identifies changes hi
36
-------
Figure 111-1
Shellfish Harvest Area Affected by Pollution Sources
Northeast Region (1988)
Boating
Wildlife(Animal Waste)
Ag Runoff
Urban Runoff
Septic Systems
Combined Sewers
Industry
Sewage Treatment Plants
Total Harvest-Limited Area
100 200 300 400
Area (thousand acres)
500
600
Note: 'Total harvest-limited area includes: Conditional, Restricted and
Prohibited waters.
•Multiple pollution sources are often identified for a single harvest-
limited area, therefore the sum of the area affected by sources in an
estuary is usually greater than the amount of harvest-limited area.
Source: NOAA
37
-------
Figure 111-2
Shellfish Harvest Area Affected by Pollution Sources
Mid-Atlantic Region (1988)
Boating
Wildlife(Animal Waste)
Ag Runoff
Urban Runoff
Septic Systems
Combined Sewers
Industry
Sewage Treatment Plants
Total Harvest-Limited Area
'///////////////////////////SS////A
i i i i i i
50 100 150 200 250
Area (thousand acres)
300 350
Note: «Total harvest-limited area includes: Conditional, Restricted and
Prohibited waters.
•Multiple pollution sources are often identified for a single harvest-
limited area, therefore the sum of the area affected by sources in an
estuary is usually greater than the amount of harvest-limited area.
Source: NOAA
38
-------
Figure 111-3
Shellfish Harvest Area Affected by Pollution Sources
Southeast Region (1988)
Boating
Wildlife(Animal Waste)
Ag Runoff
Urban Runoff
Septic Systems
Combined Sewers
Industry
Sewage Treatment Plants
Total Harvest-Limited Area
100 200 300 400 500 600 700
Area (thousand acres)
Note: «Total harvest-limited area includes: Conditional, Restricted and
Prohibited waters.
•Multiple pollution sources are often identified for a single harvest-
limited area, therefore the sum of the area affected by sources in an
estuary is usually greater than the amount of harvest-limited area.
Source: NOAA
39
-------
Figure 111-4
Shellfish Harvest Area Affected by Pollution Sources
Gulf of Mexico (1988)
Boating
Wildlife(Animal Waste)
Ag Runoff/Feedlots
Urban Runoff
Septic Systems
Straight Pipes
Industry
Sewage Treatment Plants
Total Harvest-Limited Area
500 1000 1500 2000 2500 3000 3500
Area (thousand acres)
Note: 'Total harvest-limited area includes: Conditional, Restricted and
Prohibited waters.
•Multiple pollution sources are often identified for a single harvest-
limited area, therefore the sum of the area affected by sources in an
estuary is usually greater than the amount of harvest-limited area.
Source: NOAA
-------
Figure 111-5
Shellfish Harvest Area Affected by Pollution Sources
West Coast Region (1988)
Boating
Wildlife(Animal Waste)
Ag Runoff
Urban Runoff
Septic Systems
Combined Sewers
Industry
Sewage Treatment Plants
Total Harvest-Limited Area Y//////////////////////////////77771
i i i i i
50 100 150 200 250
Area (thousand acres)
Note: -Total harvest-limited area includes: Conditional, Restricted and
Prohibited waters.
•Multiple pollution sources are often identified for a single harvest-
limited area, therefore the sum of the area affected by sources in an
estuary is usually greater than the amount of harvest-limited area.
Source: NOAA
41
-------
m. SHELLFISH
classification due to water quality changes from purely administrative changes. This will
greatly improve the ability to develop trends.
Shellfish harvest area classifications are not a very exact measure of potential risk
to humans. Some harvest-limiting classifications are the result of potential rather than
actual pollution sources (for example, STP's have buffer zones in which waters are
prohibited even though the coliform levels may be acceptable). In other cases, coliform
bacteria from animal wastes (e.g., runoff from livestock and wildlife areas, waterfowl
nesting areas) can lead to restrictions where human pathogens are not present.
The NSSP provides data collection guidelines and standards that are well-defined.
If followed, the standards should account for temporal variations caused by rainfall or
seasonal population changes. Since the early 1980's, NOAA personnel have visited each
coastal State in preparing the periodic Shellfish Register reports, so the agency can now
ensure that State-to-State differences in monitoring and classifying are taken into account
in preparing the national report. As a result, it is expected that trend assessments will
be more defensible.
Data Availability
NOAA collects data on shellfish harvest area classifications nationally and
compiles and maintains the data on an agency computer system. Periodic reports are
released containing information on the East Coast, Gulf of Mexico, and West Coast, and
interim data may be available directly from the NOAA data base. Starting in 1990,
status and trend data will be available on a Geographic Information System (GIS).
In addition to NOAA's national report, many States report on shellfish harvest
area classifications in their 305 (b) reports. This reported State data is later compiled in
EPA's "National Water Quality Inventory." At the State and local level, maps of
shellfish harvest area classifications and coliform bacteria monitoring data are maintained
(not necessarily computerized) by the responsible organizations, usually State or local
public health agencies.
Improving the Indicator
Due to limited resources, shellfish sampling stations only monitor fecal coliform
bacteria levels as an indicator of pathogens. Pathogens, however, are not the only threat
to human health. This information by itself is not as valuable as it could be if
combined with data on toxic chemicals in shellfish. Perhaps these data could be
combined with information on shellfish contamination by toxics from NOAA's Status and
Trends reports, that are discussed in section V.
States often classify waters they do not monitor (due to resource constraints) as
prohibited. Therefore, it is possible that increased monitoring could lead to
reclassification of some waters. It is likely that States already monitor all highly
productive areas so that this change would lead to increased information on marginally
or non-productive areas.
-------
m. SHELLFISH
Presenting the Indicator
Figure III-6 presents the 1985 national shellfish harvest area classifications
subdivided by regions. The relatively low figure for the West Coast reflects the fact that
much of the region has highly exposed, deep coastal waters rather than estuaries, which
are the primary commercial shellfish habitat. As well as regionally, the classifications
could also be presented on a State level or for individual estuary systems.
As noted, NOAA now identifies reclassifications in shellfish growing acreage due
specifically to water quality changes. Table III-l shows losses, gains, net change and the
reasons for change for each major estuary listed in the Northeast The data used for this
table consist of reclassifications occurring between 1971 and 1985. Figure ffl-7 is a
graphical representation of Table III-l, with the left side representing losses due to
various pollution sources and the right side representing gains from controlling various
pollution sources. Figures III-8 and ffl-9 show the same thing for the Mid-Atlantic and
the West Coast.
Within each major estuary, data are available for smaller subregions. An example
of these subregions is shown in Table III-2 using Buzzards Bay, MA. Presenting the
data at this level provides useful information to State and local policy makers.
43
-------
Area Classified
(millions of
acres)
Figure 111-6
1985 National Shellfish Harvest Area Classifications
(Subdivided by Regions)
6
5
4 •
3
2
1 ••
0
Prohibited
Restricted
Conditional
Approved
North- Mid- South- Gulf of West
east Atlantic east Mexico Coast
Source: 1985 NOAA Data
44
-------
Table HI-1
Shellfish Harvest Area Reclassifications Due to Water Quality Changes
Northeast Region (1971-1985)
Source of RedassUkation:
Passamaquoddy Bay
Englishman Bay
NarraguagusR
Blue Hill Bay
PeoobscotBay
Muscongus Bay
Sheepscot Bay
CascoBay
SacoBay
Great Bay
MerrimackR
Massachusetts Bay
Cape Cod Bay
Buzzards Bay
Narragansett Bay
Long Island Sound
Gardiners Bay
Hudson/Ran tan
HlMj&- °" "
Jttewtw.
-290
-257
0
-48
-9,587
-43
-248
-2,600
-20
0
0
0
-118
-612
0
0
-152
0
i: ..$$#$
^'...SSCfc.^? ....&»«**"..
-22
0
0
0
-179
0
-120
0
-618
0
0
0
-136
-707
-210
-341
0
0
--" 4333
0
0
0
0
-123
-578
0
0
0
0
0
-1.097
0
0
0
0
0
0
'- - 4,-TW
f ''''
0
0
0
0
0
-202
0
0
0
0
0
0
0
-33
0
0
0
0
. *23$
i^illl^jS^
0
0
0
0
0
0
0
0
0
0
0
0
-1,568
-273
-368
-79
0
0
+%m
0
0
0
0
0
0
0
0
0
0
0
0
0
0
-3,315
-66
0
-4,448
*?£&
-312
-257
0
-48
-9,889
-823
-368
-2,600
-638
0
0
-1,097
-1^22
-1,625
-3,893
-486
-152
-4.448
48.45$
**m *****
0
142
0
510
280
504
148
2,048
1,807
0
216
251
0
0
0
310
0
0
«3I«
28
0
0
0
333
79
5,777
942
0
327
0
0
0
0
0
239
0
0
7,725
0
222
86
0
1,586
0
108
158
0
0
0
0
0
1,524
0
1,815
0
0
£.499
0
0
0
0
3.870
0
0
0
275
0
0
0
0
118
10
0
0
0
4573
28
364
86
510
6,069
583
6,033
3448
2,082
327
216
251
0
1,642
10
2^64
0
0
' ..2&7J3
'f^;i
-284
107
86
462
-3,820
-240
5,665
548
1,444
327
216
-846
-1,822
17
-3,883
1,878
-152
-4,448
. "*4345
SOURCES OF RECLASSIFICATIONS
Losses:
Develop.=Shore Development and population increases resulting in degraded water quality.
S_JE=Sewage Treatment Plants having discharges of inadequately treated effluent or a new STP buffer zone.
Source=Local point sources of water pollution (ex. fish processing-Massachusetts Bay)
Septics=Malfunctioning septic systems.
Boating=Increased boating activity.
WO Change=Declining water quality usually the result of non-point runoff from increasing development.
Gains:
STPMNew Sewage Treatment Plants or improved plants caused a reduction of inadequately treated effluent
SeptJcs=Abatement of Septics.
Source=Abatement of local point sources (ex. recovery from oil spill-Buzzard's Bay).
Sewage=Sewage abatement or area became sewered.
Source: NOAA
-------
Figure III-7
Shellfish Harvest Area Reclassifications Due to Water Quality Changes
Northeast Region (1971-1985)*
Hudson/Raritan
Gardiner's Bay
L. Island Sound
Narragansett
Buzzards Bay
Cape Cod Bay
Mass. Bay
Merrimack R
Great Bay
Saco Bay
Casco Bay
Sheepscot Bay
Muscongus Bay
Penobscot Bay
Blue Hill Bay
Narraguasus R
Englishman Bay
Passamaquoddy
=9889 Acres Total
(9587=Development)
D Development
0 STP
El Septics
M Source
@ Boating
ID W.Q. Declined
• Sewage
6000
4000 2000
Losses (Acres)
2000 4000
Gains (Acres)
6000
"This Figure is a graphical representation of Table IV-1.
SOURCES OF RECLASSIFICATIONS
Losses:
Dcvelopnient=Shore Development and population increases resulting in degraded water quality.
S3P=Sewage Treatment Plants having discharges of inadequately treated effluent or a new STP buffer zone.
Source=Local point sources of water pollution (ex. fish processing-Massachusetts Bay)
Septics=Malfunctioning septic systems.
Boating=Increased boating activity.
WO Change=Declining water quality usually the result of non-point runoff from increasing development.
Gains:
STP=New Sewage Treatment Plants or improved plants caused a reduction of inadequately treated effluent
Sepn'cs=Abatement of Septics.
Source=Abatement of local point sources (ex. recovery from oil spill-Buzzard's Bay).
Sewage=Sewage abatement or area became sewered.
Source: NOAA
-------
Figure III-8
Shellfish Harvest Area Reclassifications Due to Water Quality Changes
Mid-Atlantic Region (1971-1985)
Potomac
Chesepeake
D. Inland Bays
Deleware Bay
Reed/Absecon
L. Eggharbor
Barnegat Bay
D Development
0 STP
M Septics
• Source
B Boating
• Sewage
40000 30000 20000 10000
Losses (Acres)
SOURCES OF RECLASSIFICATIONS
10000 20000 30000
Gains (Acres)
40000
Losses:
Development=Shore Development and nonpoint sources resulting in degraded water quality.
S!P=Sewage Treatment Plants having discharges of inadequately treated effluent or a new STP buffer zone.
SQUj£S=Local point sources of water pollution (ex. seafood processing wastes-Delaware Inland Bays)
Scptics=Malfunctioning septic systems.
Boaring=Increased boating activity.
Gains:
SJJP=New Sewage Treatment Plants or improved plants caused a reduction of inadequately treated effluent
S£BJJCS=Abatement of Septics.
Sfiwjg£=Sewage abatement or area became sewered
Source: NOAA
-------
Figure III-9
Shellfish Harvest Area Rectifications Due to Water Quality Changes
West Coast Region (1971-1985)
Gray's Harbor
Willapa Bay
Winchester B.
Monterey Bay
Development
STP
60000 50000 40000 30000 20000 10000 0 10000 20000 30000 40000 50000 60000
Losses (Acres)
Gains (Acres)
* =139 acres, due to STP upgrade.
SOURCES OFRECLASSMCATIONS
Losses:
Devclopmenfc=Shore Development and nonpoint sources resulting in degraded water quality.
STP=Sewage Treatment Plants having discharges of inadequately treated effluent or a new STP buffer zone.
Gains:
STP=New Sewage Treatment Plants or improved plants caused a reduction of inadequately treated effluent
Source: NOAA
-------
Table III-2
Shellfish Harvest Area Reclassifications
Due to Water Quality Changes
Subregions of Buzzards Bay, MA (1971-1985)
BUZZARDS SAYMA,
W-'
West Falmouih
Cataumet
Back River
Buttermilk Bay
Wareham
Sippican Harbor
Sippican Harbor
Smiths Neck
Cuttyhunk Pond
Cuttvhunk Pond
P
a
a
a
P
a
P
a
a
a
c
P
P
a
c
a
P
P
0
-36
-74
-538
0
-166
0
-707
-33
1500
0
0
0
118
0
24
0
0
0
Development
Development
Area became sewered
Increased boating
Mercury Problem cleaned up
STP outfall
Increase in houses on septics
Increased hnarino
Total
p-prohibited
a-approved
c-conditional
Source: NOAA
49
-------
50
-------
IV. TROPHIC STATUS OF LAKES
SUMMARY
Description of the Indicator
Trophic status is the most commonly used measure of status and trends lake
water quality. Eutrophication is a process by which a waterbody becomes rich in
dissolved nutrients, filled with detritus, and seasonally deficient in dissolved oxygen to
the extent that aquatic life is impaired. Eutrophication can result from the slow, natural
aging of a lake or can be accelerated by excessive enrichment of nutrients (primarily
phosphorus) from pollution sources such as fertilizer, sewage and detergents. States
report the trophic status of publicly owned lakes in their 305(b) reports and some
States report into the Waterbody System (WBS). States vary in their methods of
determining trophic status with the majority using Carlson's Trophic Status Index
(TSI). As a result of the different reporting methods, EPA must slightly modify the data
of some States to achieve a nationwide comparison of trophic status.
Possible Applications
Eutrophication data provide a measure of the quality of lakes and support State
program managers in targeting specific monitoring or enforcement activities. The rapid
eutrophication of a lake signals a pollution problem and can serve as a warning system.
Although inconsistencies and data gaps in monitoring and reporting limit
nationwide evaluations, the indicator does present some useful qualitative information
regarding the nation's lake water quality.
Strengths Reporting the trophic status of publicly owned
lakes is required by Clean Water Act (CWA)
314(a); Provides a scientifically defensible
measure of the ecological health of lakes.
Weaknesses Regional geographic and hydrologic differences in
lakes may limit national comparisons. A eutrophic
condition does not necessarily mean that a lake does
not support its designated use(s). The number of
lakes evaluated by trophic status fluctuates, making
trend analysis difficult. Seasonal fluctuations in
trophic status are not always taken into account.
Possible Improvements Establish a baseline number of lakes to determine
trends; Record seasonal fluctuations; Report both
number of lakes and lake acres.
51
-------
IV. TROPHIC STATUS
Data availability
Data consistency/comparability
Spatial representativeness
Temporal representativeness
Utility in trend assessment
Scientific defensibilitv
Sensitivity to change
Relationship to risk
Cost to collect and analyze data
Relationship to existing programs
Presentation value
EVALUATION CRITERIA
Good/Fair-Reported in 305(b) reports, Clean
Lakes List and in the Waterbody System (WBS).
Currently six States put trophic data in the WBS,
although the number is expected to increase in the
future. EPA can compare results from States using
different methods.
Good/Falr-The majority of States use Carlson's
TSI. Others use methods suited to individual needs.
Some States only report trophic status by number
of lakes and not acreage.
Fair-States only assess a portion of their total
lakes.
Fair-There is no consistent accounting for
temporal fluctuations.
Fair-The number and frequency of lakes monitored
and assessed is not consistent enough to assess
trends. If a baseline number of lakes assessed were
established then utility in trend assessment would
be good.
Good-Carlson's TSI is a scientifically defensible
measure. Other methods such as professional
judgements may be not entirely consistent, but are
often technically sound in their own right. They
are not necessarily less valid for purposes of the
indicator program.
Good-TSI in particular has a range of 1 to 100 and
can show incremental changes.
Good-For ecological risk rapid eutrophication is
strongly related. This indicator is not directly
relevant to human health risks.
Low-Monitoring and analysis systems are already
in place and budgeted at the State level, and the
measurements are relatively inexpensive compared
to other water quality measures (e.g. toxics).
Good-Trophic status is reported in 305(b)
reports, Clean Lakes classification report, and used
in State management programs.
Good/Fair-Eutrophication is not as easily
understood by the public as other indicators but can
be explained. Presentations can consist of national
maps, accepting comparability of varying State
results.
52
-------
IV. TROPHIC STATUS
DISCUSSION
Background
The identification of trophic status is the most commonly used indicator of lake
water quality and provides a scientifically well understood, if not complete, measure of
the status of that water resource. Despite its well-sounding prefix, a eutrophic lake is
often one with poor or declining water quality. When a lake is eutrophic, the presence
of excessive quantities of nutrients leads to algal blooms which can, when decayed,
deplete the waterbody of oxygen, rendering it unsuitable for aquatic life. While
eutrophication is a natural aging process, it can be accelerated by nutrient enrichment
from sewage discharge and run-off from agricultural fertilizers, feedlots, detergents and
other sources. In most cases, phosphorus is the primary nutrient which affects algal
production. Of the total phosphorus discharged into the nation's lakes, approximately
one-half comes from agricultural runoff, one-fourth from detergents, and one-fourth from
all other sources.
States report on the trophic status of publicly owned lakes in their 305(b) reports,
and this information is also contained in Clean Lake Classification reports that States file
under Section 314 of the CWA. The trophic status of a waterbody is generally, though
not uniformly, reported in the following categories, in order of increasing eutrophication:
oligotrophic; mesotrophic; eutrophic; hypertrophic; or dystrophic (low in nutrients, but
colored with dissolved humic organic matter). EPA asks the States, in compiling their
305(b) reports, to describe how they determine trophic status, whether by applying
professional judgement or by employing a more quantitative measure such as Carlson's
TSI. Understanding how each State determines trophic status can help EPA evaluate the
comparability of information from different areas. As a general rule, and for purposes of
this indicator project, EPA is inclined to accept and rely on whatever methodology a
State employs.
Uses of the Indicator
Eutrophication data provide a measure of the quality of the lake resource and
State program managers can use this information to target specific lakes for regulatory
action. Although a eutrophic condition does not always represent a problem, rapid
eutrophication of a waterbody signals a pollution problem and can serve as a warning
system of water quality problems. A return to a less eutrophic state is typically taken to
indicate success in mitigating a lake's pollution problems.
To use the indicator on a national level, EPA will combine information from
individual States, according to their assigned trophic status. Some States, however, do
not report on trophic status or lake quality precisely in this manner. To include
information from these States in a national indicator, EPA will have to modify it slightly.
This issue is described in more detail in the later section entitled Presenting the
Indicator.
53
-------
IV. TROPHIC STATUS
Characteristics of the Indicator
Trophic status can be determined in a variety of ways. The majority of States
use Carlson's TSI, a technical measure based on the interrelationships of Secchi disk
transparency (the distance over which a white disk is visible), concentrations of
chlorophyll-a (an indicator of algal productivity), and total phosphorus. TSI is based on
the assumption that as phosphorus levels increase, chlorophyll-a also increases and as a
result, Secchi disk transparency will decrease. Increasing TSI values indicate increasing
eutrophication. For example, a lake with a score under 40 is generally considered to be
oligotrophic, between 40 and 50 is mesotrophic, between 50 and 70 is eutrophic and
over 70 is hypertrophic. When necessary, some States use additional parameters in
conjunction with TSI to more completely evaluate trophic status. For example, Indiana
measures seven additional variables (dissolved phosphorus, organic nitrogen, nitrate,
ammonia, dissolved oxygen, plankton and light transparency).
States use lake classification methods that best suit their particular environmental
requirements, since hydrological and ecological differences between geographic regions
necessitate different trophic evaluations. The use of Carlson's TSI, while generally
widespread, is not applicable to some lakes or some ecoregions. In areas with lakes that
are turbid from erosion or with lakes that have extensive weed problems, it is not a valid
measure of trophic status. In some regions, lakes would be classified as eutrophic using
standard classification systems, even though the lakes are healthy, so that having a high
percentage of lakes classified as eutrophic does not always mean that these lakes are not
meeting their designated uses.
Consequently, States have used alternative methods to determine trophic status or
lake quality, including best professional judgement, an assessment of lake uses, known
pollution sources, and other subjective information. Other quantitative indices have also
been developed in various ecoregions, such as Brezonik's index which specifically
matches the characteristics of Florida lakes and takes into account situations where
nitrogen rather than phosphorus may be driving algal growth (EPA 1988).
Due to inconsistencies in the frequency and extent of lake assessments, data on
trophic state may not presently support in-State trend assessments. For example data
from the 1986 and 1988 305(b) reports (see Table IV-1) show in many cases a large
discrepancy in the number of lakes assessed between those two years. States rarely
assess all their lakes and in some instances only rarely assess any lakes. In its 305(b)
report, for example, Georgia notes that most of the trophic status data comes from
studies conducted before 1981. When monitoring is done in response to a suspected
problem, there could be a bias in the results towards eutrophic lakes.
Further limiting the development of in-State trends, many lakes have been
sampled infrequently during the past 15 years and the range of seasonal and annual
fluctuations in key parameters has not been well documented Lakes experience higher
levels of eutrophication during the summer months, and in order to perform meaningful
status and trend analyses, detailed documentation of time of day, time of year, and depth
of sample are necessary. If not properly accounted for, these factors can limit the ability
to develop trends from these data.
54
-------
Table IV-1
Trophic Status of Lakes 1986 & 1988 (22 States)
"*?' " State
Connecticut 86
CT88
Chanees
Honda 86
FL88
Changes
Illinois 86
IL88
Changes
Indiana 86
IN 88
Qjanges
Iowa 86
IA88
Changes
Kansas 86
KS88
Changes
Kentucky 86
KY88
Changes
Massachusetts 86
MASS
ChanR6?
Michigan 86
MI 88
Chances
Minnesota 86
MN88
Changes
Mississippi 86
MS 88
Changes
Nebraska 86
NESS
Chanees
New Hampshire 86
NH88
Changes
New York 86
NY 88
Chun BBS
North Carolina 86
NC88
Changes
Pennsylvania 86
PA 88
ChfliiR?8
Rhode Island 86
RI88
Chances
70
160
90
135
91
-44
36
412
376
554
404
-150
107
114
7
154
193
39
92
92
0
462
478
16
160
682
•522
12,034
12,034
0
34
127
93
24
23
-1
418
415
-3
3,340
3,340
0
25
144
119
26
37
11
113
54
-59
Olterttfaphk! "
8
34
26
36
57
21
0
2
2
0
75
75
0
0
0
0
0
0
14
14
0
124
133
9
19
98
79
1,203
1,203
0
0
0
0
0
0
0
141
161
20
85
85
0
2
11
9
0
1
1
7
4
-3
M«6i*ephte
44
78
34
22
19
-3
4
25
21
55
144
89
0
0
0
29
68
39
28
27
-1
276
289
13
113
367
VA
3,009
3,009
0
5
0
-5
2
1
-1
158
172
14
132
132
0
13
21
8
3
29
26
52
41
-11
-'<•
Kutmfkte* '
18
17
-1
18
13
-5
32
385
353
499
67
-432
107
114
7
125
125
0
50
51
1
62
56
-6
28
217
189
7,822
7,822
0
29
33
4
22
22
0
76
82
6
84
84
0
10
34
24
23
7
-16
17
9
-8
1 - ', ** -
'f'Q&iet**'"
0
31
31
38
2
-36
0
0
0
0
118
118
0
0
0
0
38
38
0
0
0
0
2,381
2.381
0
0
0
0
0
0
0
94
94
0
0
0
43
0
-43
3,039
3,039
0
0
78
78
0
0
0
37
0
-37
Source: 1988 and 1986 305(b) Reports
55
-------
Table IV-1 (Continued)
Trophic Status of Lakes 1986 & 1988 (22 States)
^^"•Slafe ' '" ^'<
Tennessee 86
TN88
Qiar|ges
Vermont 86
VT88
Oianires
Virginia 86
VA88
duns?5
Washington 86
WA88
Changes
Wisconsin 86
WI88
Qjanpfts
TOTAL 86
TOTAL 88
NST^HANfSK
*V ' ' f' ''''''•
^
-------
IV. TROPHIC STATUS
Data Availability
EPA encourages States to report on the trophic status of all publicly owned lakes
in their 305(b) reports. However, as noted earlier, not all States do so. States may
provide trophic status data through the waterbody system (WBS). In 1988, only six
States (and the District of Columbia and Puerto Rico) used the WBS to report on trophic
status. However, this number is expected to increase significantly in 1990. Some States,
such as New Jersey, West Virginia, New Mexico, Arizona, South Dakota, and Missouri
do not present trophic status information in their reports. Others, including Ohio,
Wyoming, Georgia, Maine and Texas do provide information on the trophic evaluation of
some of their lakes, but not in a manner that is consistent with other States. For
inclusion as part of the nation-wide indicator, EPA will have to manipulate this data (see
Table IV-2 for more details).
Louisiana does not use standard trophic categories in assessing its lakes, since
many lakes would all be classified as eutrophic, even though they are quite productive.
Rather, the State uses a classification system based on the best professional judgement of
lake users, and assigns the lakes differing values based on this Lake Condition Index.
Improving the Indicator
States can report on trophic statue using number of lakes or by total acreage, and
the use of both allows a more complete view of lake conditions at the State level.
Figures IV-1 and IV-2 illustrate this point, using four States chosen at random. While it
appears, by looking at the number of lakes, that a significant portion of the lakes in
Vermont are of unknown status, the total acreage of these lakes is minimal. Although
the State's portion of mesotrophic lakes is small, mesotrophic acreage is the largest
classification due to the inclusion of 142,033 mesotrophic acres of Lake Champlain
(1988 VT 305 (b)). When examined together, the two measures provide a more
comprehensive picture of trophic status within a State.
Presenting the Indicator
As noted, Figures IV-1 and IV-2 illustrate usefulness of reporting both number of
lakes and lake acreage for each trophic classification. National data can also be
displayed. Maps shown in Figures IV-3 and IV-4 provide some national view on States'
trophic status, using information that comes largely from state 305 (b) reports. However,
as discussed above, some States did not report trophic status or did so differently other
States. In these cases, TBS has estimated the values for inclusion on these maps. For
example, Louisiana classified its lakes by assigning values relating to their productive
use, as determined by users perception. In compiling this map, TBS has taken these
classifications and assigned them to categories of various trophic states. Table V-2
provides more detailed information on the actual methods used by these States and
manipulations to their data done by TBS. Figures IV-3 and IV-4 also illustrate the
relatively few States that provide trophic status based on total acreage as opposed to
number of lakes.
57
-------
IV. TROPHIC STATUS
The identification of trophic status does provide a good indicator of lake quality.
However, inconsistencies in the frequency and extent to which lakes are assessed limit
EPA's ability to draw national or State-wide trends. As States report on trophic status in
a more consistent manner, and provide information on the acres assessed as well as the
number of lakes, this measure will become a more useful, nationally comparable
indicator.
58
-------
Table IV-2
States Using Different Trophic Status Reporting Methods
Georgia: Reporting: Lakes were classified in one of 3 categories: A, highest need
for restoration; B, moderate need for protection; and C, few water quality
problems.
Mapping Assumptions: All lakes in categories A and B were included
in the eutrophic category.
Louisiana: Reporting: As a result of geological and hydrologic processes, all
Louisiana lakes are considered eutrophic. Therefore, Louisiana has
developed a Lake Condition Index, based on users perception of lake
health. Using Total Organic Carbon(TOC) as the only measured parameter,
because of its correlation to user perception, lakes are categorized as
Excellent to Poor, Good to Acceptable; Acceptable to Marginal; Marginal to
Very Poor.
Mapping Assumptions: the number of Acceptable to Marginal lakes
were used to determine the percentage of eutrophic lakes.
Maine? Reporting: The State reports trends in trophic state rather than actual
trophic states. They categorize lakes according to water quality trends
(deteriorating, stable or improving) and whether or not the lake experiences
algal blooms (ex. "Those with deteriorating water quality and culturally-
induced algal blooms.")
Mapping Assumptions: Any lakes assessed as a priority problem due to
periodic algal blooms and lack of transparency were considered eutrophic
for use in the map.
Wyoming: Reporting: The vast majority of lakes were classified as unknown.
Mapping Assumptions: Only lakes with trophic status were considered
(oligotrophic=62, mesotrophic=12, and eutrophic=28).
States Not Reporting
New Jersey: Only very limited monitoring and assessment of lakes has been conducted in
the past 5 to 8 years.
W. Virginia: The trophic condition of the State's lakes has never been officially
documented or scientifically determined.
South Dakota: No current data available.
Arizona: No current data available.
New Mexico: Many lakes were surveyed prior to October 1,1982, no current information
available.
Missouri; No current data available.
59
-------
Figure IV-1
Trophic Status of Lakes By Number of Lakes Assessed
(1988)
Number of Lakes
Assessed
D Unknown
• Hypereutrophic
D Eutrophic
9 Mesotrophic
HI Oligotrophic
IL
UT
VT
Figure IV-2
Trophic Status of Lakes By Lake Acreage Assessed
(1988)
Acres Assessed
Unknown
Hypereutrophic
Eutrophic
Mesotrophic
Oligotrophic
VT
Source: 1988 305(b) Reports
60
-------
Figure IV-3
Percentage of Assessed Lakes Per State Classified as Eutrophic (1988)
D
0-24%
•Hypereutrophlc Lakes are Included In eutrophlc %.
•Only the total number of lakes wfthtraphfedasssiflcaticrovrere considered. Source: 1988 305(b) Reports
(Le.-Lakea wUi unknown trophto status were not taken Into account for the total number of assessed lakes.)
Slates not n»purt>n>
R Do not determine trophic status ki manner consistent with other states;
See Table V-2.
-------
Figure IV-4
Percentage of Assessed Lake Acreage Per State Classified as Eutrophic (1988)
<7i
CO
D
0-24%
States not Reporting
•Hypereutrcphic acreage Is Included in eutrcphfc %.
•Only the total acreage with trophic dasssffications were considered.
(I.e.-Lakes with unknown trophic status were not taken Mo account for the total number of assessed acreage.)
Source: 1988 305(b) Reports
Hr>> not determine trophic status In manner consistent wflh other states;
See Table V-Z.
-------
V. TOXICS IN FISH AND SHELLFISH
SUMMARY
Description of the Indicator
This proposed indicator measures the accumulation of pesticides and other toxic
chemicals in fish and macroinvertebrate tissue. Currently, several Federal Agencies
including the Fish and Wildlife Service (FWS), the National Oceanic and Atmospheric
Administration (NOAA) and the Environmental Protection Agency (EPA), as well as many
State Agencies collect these data. States commonly use the information to help develop
fish consumption advisories. FWS and NOAA studies are ongoing and can provide status
and trend information on a regional level. The State programs vary in the type of
animals tested, the amount and quality of the testing done, the chemicals that are
analyzed and the way the results are used. To use on a national level the State data will
require better data storage and accessibility. FWS and NOAA data are typically for
whole-fish or liver samples, and thus indicative of ecological risks but not human health
risks. EPA and State data are often for edible tissue fillets, and thus can be used to
estimate potential risks to human consumers.
Possible Applications
The FWS and NOAA data can be used directly to support regional status and trend
identification for river and marine systems. EPA will have to decide how it wants to use
the State data in a regional or national assessment program. The Agency might be able to
develop trend data for selected States and sites if it were to ask States to provide data
from locations that are part of ambient monitoring systems.
Strengths The accumulation of toxicants in fish and shellfish
tissue can provide an indication of the general
quality of the water resource, at least with regard
to specific chemicals and the implications of their
presence. Some nationally consistent data sets
exist. Many States collect tissue contamination
data. There is a lot of interest among the general
public, especially with regard to any human health
risks associated with toxic contamination of fish
and shellfish.
Weaknesses The actual ecological implications of fish and
shellfish contamination are a subject of
controversy. There is a lot of variability among
States in the chemicals tested for and in the quality
of the analysis. State data are hard to retrieve
if they are put into STORET.
Possible Improvements Increase in the use by States of BIOS as central
repository for storage and retrieval of data; Have
States include more information in their 305(b)
reports; Greater coordination among various
federal agencies (FWS, NOAA and EPA) in site
selection and identification of chemicals to be
tested; Possible extension of EPA Bioaccumulation
Study.
63
-------
V. TOXICS
Data availability
EVALUATION CRITERIA
Good-For FWS and NOAA data; Available in periodic
reports and from computerized databases; Also, data
should be readily available from EPA's
bioaccumulation study.
Fair-State data are maintained on State databases
which may or may not be automated. Some States
put data into STORET, but retrieval of information
from that system is difficult.
Data consistency/comparability
Spatial representativeness
Temporal representativeness
Good-For EPA Bioaccumulation Study and FWS and
NOAA studies, consistent analytical methods are
used throughout the country. Although different
fish species are, by necessity, used in different
regions, the differing samples can be used for
comparison purposes.
Fair/Poor-States have a lot of variability
in the type of chemicals tested for, species tested,
and the quality of the analysis.
Good-FWS sites were chosen to provide national
information on status and trends. Benthic
surveillance and mussel watch sites are located
near urbanized areas so as to be representative of
the general areas in which they are located.
Fair-Bioaccumulation study; Some sites were
randomly chosen, others were located in
undisturbed areas, areas of important fisheries and
at problem areas. Each type of site could be
representative of similar sites around the country.
Variable-State information is better in States
with ambient monitoring networks.
Good-The presence of contaminants in fish and
shellfish tissue is more temporally consistent than
measuring for the chemicals in the water column,
since they are less transient. In the National
Contaminant Biomonitoring program (NCBP),
sampling always occurs in the fall to increase
temporal comparability and similarly bivalves are
always tested in the fall in NOAA's status and trends
program.
Varlable-The State testing results vary not only
between States but may also change within a State
from one year to the next, affecting temporal
comparability.
64
-------
V. TOXICS
Utility in trend assessment
Scientific defensibilitv
Sensitivity to change
Relationship to risk
Cost to collect and analyze data
EVALUATION CRITERIA
Good/Fair-The FWS study is designed specifically
to measure trends for specific contaminants in fish
tissue. NOAA data will also support trends as
monitoring continues into the future. The state data
that is done at fixed monitoring stations would be
useful for trend monitoring. However, a lot of
monitoring is done for "special studies" and would
not support the development of trends.
Good-FWS and NOAA studies use well-defined and
accepted study protocols as does the EPA
Bioaccumulation study. State activities are much
more variable.
Fair-However, the utility of this measure to detect
changes in the toxic load would depend to a great
extent on the species involved and their tendency to
accumulate particular chemicals. The FWS study
demonstrated the decline in the use of certain
pesticides. The pesticides were still being found
long after their use was discontinued, which is
important in demonstrating continued impact, but
shows that natural response lags will slow down
the ability to demonstrate environmental results.
Non-linear relationships of loads to ambient
concentrations in tissues mean that modest
incremental changes in pollutant inputs will not
always be distinguishable from tissue monitoring.
Fair-For bivalves and where fish fillets are
analyzed the measure provides information
relevant to human health risks. The measure is
directly related to ecological risk. However,
quantitative information on the actual
environmental impacts of the accumulated toxicants
is usually not available.
Moderate/High-Costs for tissue toxicity tests
and their evaluation can be high. Accordingly,
studies such as EPA's Bioaccumulation study may
not be repeated.
Relationship to existing programs
Presentation value
Good-Several Federal Agencies are already
doing testing as are States in conjunction with FDA.
States can report findings in the 305(b) reports.
Good/Fair-The information is understood by
the public and can in certain instances be well
displayed on maps. However, given the wide range
of variables in different studies, it is difficult to
get a "nationwide" view from State studies.
65
-------
66
-------
V. TOXICS
DISCUSSION
Background
The accumulation of toxic chemicals, pesticides, or other industrial chemicals in
fish and shellfish tissue can serve as an indicator of the quality of the water resource in
terms of potential ecological and human health risks. This kind of data, showing status
and trends as well as information about impacts of specific pollution sources, is of great
interest to the general public as well as to government resource managers.
A number of federal agencies collect data on tissue contamination. Of particular
interest to the EPA are the efforts of FWS and NOAA. Since 1967, the FWS has been
collecting data on chemical residues in freshwater fish under the NCBP. There are 112
sampling sites in the fish monitoring network and the whole fish samples are tested for
organochlorine pesticides, PCBs, and seven metals. NOAA collects data on the
concentrations of a large suite of toxic chemicals in marine bottom fish and benthic
organisms at specific locations, as part of its National Status and Trends (NST) Program.
EPA is also conducting its own study (the National Bioaccumulation Study) to
determine the extent to which pollutants are bioaccumulating in fish and to identify
contaminant sources. In addition, many States include some level of fish tissue analysis
in their biological monitoring activities. According to a survey conducted in 1987 that
assessed biomonitoring activities at 65 State and Federal Agencies (EA Engineering
1987), 55 of the respondents indicated some level of tissue residue analyses. A more
recently completed review indicated that 46 States monitor tissue residues in fish in
rivers and streams, 25 in lakes, and 13 in coastal and estuarine areas (RTI, 1989). There
is, of course, a great deal of variability among the States in the frequency of the testing,
the species examined, and the chemical parameters examined. At present, some States
store their data in STORET, though the information is hard to retrieve easily. EPA is
planning to include tissue data storage and retrieval in improvements to BIOS, which
will improve the utility of the data.
Uses of the Indicator
FWS program managers and other State and federal agencies use the NCBP data
as a general measure of status and trends of certain pesticides and other toxicants in
waterways.
The FWS regularly issues reports on the results of their analyses. Reports on the
1984 data will be available in the fall of 1989. Results from the data collected in 1986
will be available in the fall of 1991 and by that time, the Agency will have established
trends at 80% of their stations. Analysis of the 1988 samples, which are currently
frozen, will then focus on the remaining 20% of the stations for which the trends have
not been established (Steffick 1989).
67
-------
V. TOXICS
The analyses have yielded useful information. Data from 1976 through 1981
show a statistically significant decline in DDT, PCBs, and dieldrin, reflecting the
effectiveness of the EPA's bans on these substances. The data also showed the
geographic spread of PCB's and toxaphene over the same period (LaRoe 1987). As well
as providing insights on status and trends, the data are used as a targeting tool, helping
to determine where additional research and clean up activities should be undertaken. The
results can also serve as a reference against which results at known contaminated sites
can be compared. At EPA, this information may be of particular utility to those dealing
with non-point sources of pollution, for which ambient monitoring data are sometimes
hard to come by.
The Benthic Surveillance component of NOAA's NST program measures the
concentrations of toxicants in bottom fish and sediments, and the same chemicals are
measured in bivalves under the Mussel Watch program. Benthic Surveillance and Mussel
Watch program data are maintained on a mainframe computer at NOAA and can be
down loaded onto PC databases and transferred to EPA. Results are periodically
presented in NOAA reports. NOAA uses the information to rank the most contaminated
areas and as a screen for areas that warrant more intensive monitoring. The information
will be very useful to assess national trends, but cannot be used to evaluate specific
regulatory actions since it would be difficult to identify any single activity or source
responsible for the observed contamination at a single monitoring station.
Characteristics of the Indicator:
The 112 sites in the NCBP program were originally sampled every year but
have recently been sampled every two years. At each site, samples are taken from
bottom dwelling fish and from a predator species such as a trout or black bass. Since
the Agency uses whole fish samples and not fillets, the results are not good indicators of
human health risk, because toxicants are typically concentrated in organs such as the
liver that are not generally consumed. One drawback of a national monitoring system is
the need to use different species at different locations around the country. However, the
FWS reports that as long as samples are adjusted for lipid content, valid comparison can
be drawn (LaRoe 1987). The sites were chosen to provide information on status and
trends at a regional and, when combined, at a national level. The stations were selected
with the USGS as part of the USGS National Stream Quality Monitoring Network
(NASQAN) to represent sub-basin characteristics.
Collection and analysis activities at the sites conform to strict technical standards.
Sampling occurs in the fall so the data are temporally as well as spatially comparable.
The results of analyses noted above demonstrate the ability of the indicator to track
changes as they occur.
At present, the NCBP sampling sites are located only in rivers or streams.
However, as the program is re-evaluated, the Service hopes to include estuaries and
wetlands in their monitoring scheme.
NOAA collects data for a number of chlorinated synthetic compounds, PAHs, and
trace elements for the Benthic Surveillance program at around 50 sites around the
country. Each site is visited every other year. As opposed to the NCBP program which
68
-------
V. TOXICS
analyzes whole fish tissue, the NOAA program analyzes individual liver tissue since the
liver is the organ that accumulates the greatest concentration of toxic chemicals. As with
the NCBP program, the data are not useful to determine human health risk.
NOAA collects information at about 150 sites around the country in the Mussel
Watch program and intends to study each site annually in the future. Thirty-seven of the
sites were chosen to allow comparison with data collected by EPA as part of a mussel
watch program it ran from 1976 through 1978. The sites for the Benthic Surveillance
and Mussel Watch programs are located near urbanized areas but not near specific
discharges, so as to be representative of the general areas in which they are located (with
a few exceptions due to early problems in sampling design). This differs from other
programs such as the California Mussel Watch program which collects data from
bivalves located or placed at specific discharges. Collection and analysis methods for the
NST data are subject to uniform quality assurance protocols. To ensure temporal
comparability, bivalves are examined only in the winter. While the data will clearly be
very useful for looking at temporal trends as time goes on, some questions remain as to
whether spatial comparisons are appropriate. Since some stations are closer to specific
pollutant sources than others, the data from these station may not represent the overall
water quality of the study area.
Data Availability
Results of the NCBP analyses are stored on a mainframe computer in Columbus,
Missouri and are available through the FWS by downloading onto a PC. The NOAA
data are maintained on a mainframe computer in Rockville, Maryland and can also be
downloaded onto a PC. The FWS and NOAA data are also made available periodically
in reports.
Improving the Indicator
The FWS is currently reevaluating the NCBP program. It is investigating the
possibility of including new types of bioassessment criteria and expanding the study to
include some new chemicals and types of waterbodies. Moreover, FWS is considering
including some of NOAA's mussel watch program in the NCBP and identifying new
sites with the USGS's National Water Quality Assessment (NAWQA) program
(Andraison 1989). The FWS is interested in coordinating its monitoring efforts with
those of other agencies. This would be a good time for EPA to get involved in the new
program design to ensure that chemicals or waterbody types that it felt were important
were included.
In a report that examined historical data on contamination in fish and shellfish,
NOAA has recommended that "a nationally centralized and easily accessible PCB and
pesticide database should be completed and used to receive and process new data from
State and federal programs (NOAA 1988)." The proposed modifications to BIOS might
help fill this need and allow for data from a number of agencies to be stored hi one
central location. However, the large amount of variation in methodologies, species
tested, chemicals examined, and the purposes for which the data are collected, dictate
69
-------
V. TOXICS
that a great deal of caution be exercised in drawing inferences from data collected by
different agencies.
Increased coordination among FWS, NOAA, EPA and other interested federal
agencies to help establish parameters for species to be examined, chemicals to be tested,
and areas to be investigated would greatly improve the value of this measure as an
indicator of water quality.
Presentation of the Indicator
EPA could try to take the raw data from the NCBP and NOAA programs and
derive their own presentations, or it could just use the presentation techniques used by
FWS and NOAA in their reports. In the short-term, it may be more appropriate to use
the data as it is analyzed and presented by FWS and NOAA. Among the ways that the
NCBP data can be displayed are shown in Figures V-l and V-2. In Figure V-l, the
regional distribution of PCBs in freshwater fish is shown graphically while in figure V-
2, the nationwide decline in the concentration of DDT and its homologs from 1969-1981
is shown.
The Benthic Surveillance data could be displayed on maps as shown in Figure
V-3 showing the level of DDT in livers of estuarine fish in 1984. These maps could be
modified to show trends over time.
Mussel Watch data might be shown as in Figure V-4, with levels of specific
chemicals (e.g., PCBs) in tissue at the same sites over time. This figure shows the 20
highest ranking sites for PCBs in bivalves, for the three-year period from 1986 through
1988. It is important to use care in looking at this data to try to ascribe trends. Since
three years of data are shown, a trend might be inferred when the movement has been
consistent for all 3 years. However, NOAA has indicated that a trend only occurs when
there is a significant difference among the years' data and the three-year change is in
one direction. As seen in Table V-l, only at three sites (Hudson River, Boston Harbor,
and Elliott Bay), could trends be determined according to this definition. In each of
those cases, major improvements have occurred, with PCB concentrations decreasing
markedly. However, additional data would be needed to establish any national trend.
EPA'S BIOACCUMULATION STUDY
The National Bioaccumulation Study is designed to test the accumulation of toxic
chemicals in fish and to relate this accumulation to specific sources. One of the
recommendations of the workshop group that looked at potential indicators that could be
related directly to specific sources was that new site specific studies be initiated.
Monitoring sites for the one-time EPA study, which could serve for such site-specific
studies, were selected with assistance from EPA Regional Offices and covered the
following:
• Problem sites with significant industrial, urban, or agricultural activity (with
special attention given to bleach kraft pulp mills)
70
-------
Figure V-1
National Contaminant Biomonitoring Program (NCBP)-PCB Residues
in Freshwater Fish (1980-81)
Source: FWS
-------
Figure V-2
National Contaminant Biomonitoring Program (NCBP) - Geometric Mean
Concentrations of p, p'-DDT Homologs
in Fish Samples (1969-81)
~ 1-2-
o
5 1.0-
| 0.8-
£ 0.8 H
g 0.4 H
O 0.2 H
iMt mo itri i»ra m* 1*74 itr«-rr
Collection Period
Source: FWS
-------
Figure V-3
Total DDT in Livers of Estuarine Fish Composites Collected
A at 42 Sites in 1984*
0.5 parts per million, dry weight
'Computed from original data for the 1984 NOAA Status and Trends Program.
Source: NOAA: A Historical
Assessment Report 1988
-------
Table V-1
Total PCBs in Mollusks -
Top 20 Sites for 3-Year Period
(1986-1988)
RANKING.
Code Location Stela So 1986 1997 1988
BBAR Buzzards Bay MA me 1 It 1
HRLB HudJRar. Est. NY me 2 3 3t
NYSR N.Y. Bight NJ - me 3 17t 11
HRUB HudTRar. Est. NY me 4 5t 2
SOHI San Diego Bay CA me 5 4 5t
GBYC GaJvestonBay TX cv 6 8t 16
NYSH N.Y. Bight NJ me 7 12 51
BHOB Boston Hrb. MA me 8 5t <
BBGN Buzzards Bay MA me 9 < <
BBRH Buzzards Bay MA me 10 1t 3t
BHDI Boston Hrb. MA me 11t 19 <
LJTM Long Is. Snd. NY me 11t 81 7t
EBFR Elliott Bay WA me 13 < <
BHHB Boston Hrb. MA me 14 13 <
HRJB Hud/Par. Est NY me 15 < 10
UNH Long Is. Snd. CT me 16 < 19t
UMR Long Is. Snd. NY me 17 < 14t
SAWB St. Andrew Bay FL cv 18 < I7t
USI Long Is. Snd. CT me 19 < <
NYLB N.Y. Bight NJ me 20 15 171
SFEM San Fran. Bay CA me 5t 7t
UHH Long Is. Snd. NY me < 8t <
UHR Long Is. Snd. CT me < 11 <
CBSP Choctawhat. Bay FL cv < 14 <
MOSJ Marina Del Rey CA me < 16 <
BHBI Boston Hrb. MA me < 171 <
PBIB PensacolaBay FL cv < 20 <
GBSC GalvestonBay TX cv • 9
APDB ApalachJcola Bay FL cv < < 12
ABWJ Anaheim Bay CA me < < 13
SFSM San Fran. Bay CA me < < 19t
LJCR Long Is. Snd. CT me < < 19t
Legend:
Sp.=Spectes:
me-Mytilus edulis, cv-Crassostrea virginica, mc-Mytilus califomianus
Rankings:
QMM9H oUni. Dul.
1
.2
6
3
5
10t
7t
10t
14
4
13
9
19t
18
15t
<
<
<
<
19t
7t
17
15t
<
<
<
<
12
<
<
<
<
- site was not sampled in the given year, t-sites tied in rankings, <-ranking was not among
Signif. Dirt.
yes-there was a significant difference among the three years of data
no-there was not a significant difference among the three years of data
- data was insufficient to make a determination
Trend
i-increasing concentrations of total PCBs, d-decreasing concentrations of total PCBs
no
yes
no
no
no
no
yes
no
no
no
yes
no
yes
no
yes
no
no
no
no
no
-
yes
yes
no
yes
yes
no
•
yes
no
no
no
the upper 20
I rfifln
no
d
no
no
no
no
no
no
no
no
d
ho
d
Tio
no
no
no
no
no
no
-
no
no
no
no
no
no
-
no
no
no
no
Source: NOAA
74
-------
Figure V-4
Total PCBs In Mollusks -
Top 10 Sites for 3-Year Period (1986-1988)
BBAR
HRLB
NYSR
HRUB
SDHI
GEPiC
NYSH
BHDB
BBGN
BBRH
1986 Mean
D 1987 Mean
• 1988 Mean
Note: Refer to Table V-1 for
an explanation of site codes.
(e.g., BBAR=Buzzards Bay).
1000 2000 3000 4000 5000 6000
ng/g-dry wt (Total PCBs in Mollusks)
7000 8000
Source: NOAA
75
-------
V. TOXICS
• Randomly located sites
• Relatively undisturbed areas, and
• Areas of important fisheries.
Uses of the Indicator
Since EPA does not currently intend to replicate this study, it is obviously of little
value for national trend analyses. There are two main objectives of the study:
• To identify sites with potential sources of contamination (point or nonpoint) in
need of further investigation and possibly stricter pollution controls; and
• To determine what potentially hazardous levels of accumulation in fish relate to
specific types of industries or land uses.
The data will be used primarily by water quality managers at the State and
regional level as a targeting tool.
Characteristics of the Indicator
The data collection and sampling methods follow well-defined and scientifically
defensible practices developed by EPA's Environmental Research Laboratory in Duluth,
Minnesota. The one-time investigation is not temporally representative and the non-
random selection of sites around specific areas of concern limits the spatial
representativeness of the data. As in the NCBP, whole fish samples (a composite of the
same species of fish captured at each site) are used. However, edible fillets were
analyzed as well, so that some preliminary analysis of human health implications at a
screening level is possible. Study results will be available during the fall of 1989.
Improving the Indicator
EPA now needs to consider whether portions or all of the monitoring done in the
Bioaccumulation Study should be repeated. This could allow for follow-up analysis of
those sites identified in this round as requiring additional pollution controls. Further
testing could demonstrate the impacts of regulatory actions on the levels of contaminants
in fish.
Some of the sites, such as the randomly chosen ones, could possibly be
incorporated into the FWS's revisions to the NCBP. Data collection might also be
continued by individual States as part of their water quality monitoring activities and
EPA could help provide guidance as to specific chemicals that would be most
worthwhile to monitor.
76
-------
V. TOXICS
Presentation of the Indicator
EPA is now producing a document to present the results of the study and the
report will include a number of graphical presentations. The results for a specific
chemical (such as PCBs or chlordane) could be displayed by site category (indicating the
concentrations found across the country around specific types of industrial outfalls) or
geographically.
STATE TISSUE RESIDUE TESTING
As noted earlier, a large number of States collect some information on
bioaccumulation in fish. On rivers and streams, 37 States conduct tissue residue
sampling through intensive surveys while 16 employ some fixed station monitoring (RTI
1989). These data are most often used to help develop fish consumption advisories,
usually in conjunction with State public health agencies. The data may also be used by
the States in determining designated use assessments for reporting purposes in the 305(b)
reports. As might be expected, there is a great deal of variability among the States with
regard to species investigated, the amount of testing done, how it is done, what
chemicals are analyzed, and how the results are used. From year to year, States can
change the goals of the monitoring and, therefore, the types of data collected. Many
States put their tissue data into STORET, as well as maintaining it on State computer
systems. As noted earlier, it is not easy to retrieve the information from STORET in a
meaningful form. It can be done, however, and Figure V-5 shows results of a retrieval
using a program specially designed by the State of Illinois.
Improving The Indicator
EPA is engaged in an effort to improve the BIOS component of STORET and
include in the upgrade a file for tissue toxicity data; which does not currently exist in
BIOS. The BIOS upgrade will, hopefully, allow for easier access and retrieval of tissue
data. In response to a needs assessment questionnaire, 32 States indicated an interest in
using BIOS for tissue residue data (EA Engineering 1987). Once the improved data
system is available, EPA might be able to use it to determine status and trends with
regard to tissue data. For example, the Agency might be able to see the levels of a
specific contaminant (such as PCBs) in certain types of fish (top feeders for example)
and track changes over time. For this to allow for trend or nationwide analysis, EPA
and the States would have to agree upon some criteria for the types of species and
chemicals to be analyzed and the types of sites to be included in the database.
A more useful reporting mechanism for purposes of the indicator project would be
for States to include more information in their 305(b) reports. If the EPA agrees that
information on tissue residues should be reported on separately, as an indication of water
quality, the Agency might ask the States to provide more information on the results of
their tissue residue analyses in their 305 (b)s. The current guidance, in the section on
public health/aquatic effects, asks for information on the number of exceedances of FDA
standards in tissue data, the pollution of concern, its source, and the size of the
waterbody affected. This can be useful information, and as part of a section on
"indicators" in the 305(b) reports, EPA could ask for more details including actual levels
77
-------
00
Figure V-5
Example of a Translated STORET Retrieval
ILLINOIS ENVIRONMENTAL PROTECTION AGENCY "«•« COOPERATIVE FISH CONTAMINANT MONITORING PROGRAM
EXCURSIONS OF FOA ACTION LEVELS NOTED BY •
STATION: K 96 HISS R E HANNIBAL RM 310 T4S R6M NEI6 COUNTY: PIKE BASIN: MISS NO CEN STATUS: PERMANENT
F CHLOR- DIEL- HEPT. TOTAL TOTAL MEH-
OR HEI6HT LIPIO DANE
DATE
660904
660904
660904
660904
660904
660904
660904
66MEAN
LAB FISH H
IEPA CARP H
IEPA CHCF H
IEPA MHCR M
IEPA CARP F
IEPA CARP F
IEPA CHCF F
ICPA MHCR F
-MHOLE
66HEAN-FILET
XEXCURSIONS-FILET
IBS '/. PPM
.64 13.0 0.046
.77 II. 0 0.110
.71 5.6 0.022
.64 A 3.3 0.036
.I6A II. 0 0.110
.74 A 3.4 0.0)6
.63 A O.I 0.020K
0.060
0.051
0.0
DRIN
PPM
.220
.250
.066
.069
.220
.072
.010K
.186
.093
0.0
EPOX.
PPM
.042
.0*9
.015
.010
.0)0
.012
.OIOK
.0)5
.017
0.0
DOT
PPM
.0)9
.070
.049
.033
.032
.021
.020K
.05)
.02*
0.0
PCBS CURV
PPM PPM
.20
.39
.2)
.14
.19
.15
. IOK
.27
.14
0.0
OTHER
PPM
T=
1 =
T»
1=
fe
T«
T»
As reprinted in "Preliminary
Requirement Statements For
Final
Toxicity Testing and Tissue Residue
Components of BIOS."
Washington Battelle Programs.
September 12, 1989.
-------
V. TOXICS
of specific contaminants found in the species tissue, the specific tissue impacted, and the
reach number or some geographic identifier of the area in which the contamination was
found. This would allow for some trend identification. Moreover, since some EPA
Regions and States now conduct health risk assessments for consumption of contaminated
fish (usually for a range of possible consumption rates, representing average and high-
consuming fish eaters), the Agency could ask that the results of those studies be
provided in addition to FDA action level exceedances. This would respond to two
problems not currently addressed using action levels: estimating high risks to consumers,
and evaluating the significance of contamination for the large number of chemicals for
which action levels have not been developed.
It would also be helpful if States would identify in their 305(b) reports monitoring
sites which were part of a special study (around a potential discharge site or as part of a
basin survey) and which were part of the State's ambient monitoring network. The
narrative information included with the data could contain information on whether the
site is being used to track non-point sources or a specific discharges. This tagging
would allow EPA to more appropriately develop trend information on routine monitoring
sites, and to note changes possibly related to program activities at special survey sites or
ambient sites located near discharges.
79
-------
80
-------
VI. MEASURES OF BIOLOGICAL COMMUNITY STRUCTURE
SUMMARY
Description of the Indicator
The basic indicator is the health of the biological communities of water body segments, as
measured by monitoring and abundance of species expected to inhabit that type of water (and in
some cases, of species considered to be sensitive to polluted conditions). States use a variety of
biological community measures, with the Index of Biotic Integrity (based on counts of fish
species), other fish community indices, and several types of macrobenthic invertebrate indices
being the most common.
To use this indicator at the national level, we will attempt to combine data on the health
of biological communities in all monitored water bodies according to their qualitative rankings
according to the various indices used for bioassessments. That is, we will use only the
information on whether the biological community of a given segment was considered to be in
excellent, good, fair, or poor condition, considering all segments with the same qualitative
ranking as equivalent for the national evaluation. This approach was recommended by several
State and regional biologists in the workgroup on program effectiveness evaluation at the March
1989 Surface Water Indicators Workshop, and was then voted one of the most highly rated
potential national indicators by the workgroup as a whole.
Possible Applications
Because the specific types of biological monitoring (i.e., which types of animals or
plants are counted) vary from State to State, only very qualitative reporting will be justified at
the national level. Such qualitative information could be useful for Federal planning and
targeting. It will not support site-to-site comparisons. However, the Agency believes that If
spatial data availability issues can be worked out, this indicator may support comparisons of
broad-scale trends in biological conditions among geographic regions, States, watersheds, etc.
At the State level, where consistent methods of monitoring and analysis can be assured,
sophisticated analyses of biological community data are already being done. Such analyses can be
used to support impact assessments for specific pollution sources, or to assess the results of
pollution control upgrades at particular facilities.
81
-------
VL BIOCOMMUNTTY
Strengths
Weaknesses
Possible Improvements
This indicator is the most direct measure possible of
support of a Clean Water Act (CWA) goal, because
maintaining biological integrity is one of the legislative
mandates. The information is scientifically defensible, and
also makes sense to the public and decision makers.
Availability of data to assess the status of waters
nationwide is moderately good, with data available for some
waters in almost all States, provided that a decision is
made to consider the various types of biological community
measures as comparable to one another in a broad,
qualitative sense.
The variety of approaches to assessing biological
community integrity means that no single approach will
likely ever be embraced by all States. This means that
only rough, qualitative comparisons of conditions from
place to place will ever be likely using this approach.
Also, biological monitoring is now often done in special
studies rather than as part of ambient network monitoring,
so that there are relatively few locations where temporal
trends can be assessed.
More consistent monitoring in space and time, i.e.,
establishing more monitoring network locations that will
be monitored repeatedly over time would allow trend
assessments for a larger portion of the nation's waters
than is currently possible.
82
-------
VI. BIOCOMMUNTTY
Data availability
Data consistencv/comnarabilitv
Ability to estimate
Spatial representativeness
EVALUATION CRITERIA
Fair - Most States use biological community monitoring
for at least a few critical water bodies, or for special one-
time studies. A number of States have incorporated such
monitoring into ambient monitoring networks, and
additional States may do so in the future. The proportion of
most States' assessed water bodies in which biological
community monitoring is done is currently quite low.
However, the positive response of most States to recent
recommendations by State and federal agencies concerning
the benefits of biological monitoring indicates that
biomonitoring will probably be carried out for larger
portions of many States' surface waters in coming years.
Fair - A variety of biological community measures are
commonly used, which differ greatly in the type of
organisms whose abundance is assessed, and in the
complexity of procedures for combining data on various
species' abundance. Some measures are formulated into
sophisticated mathematical indices to take into account
habitat features and other environmental factors (e.g., IBI
based on ecoregions) while others are relatively simple
weighted sums of a few key species. The great differences
in the types of measures used means that data from State to
State, and sometimes within States, will not support
sophisticated ecological comparisons among sites or
regions. This indicator development project is
investigating whether biomonitoring experts concur with
the sense of the March 1989 Surface Water Indicator
Development Workshop that the biological community
indices commonly used for water quality assessments are
sufficiently comparable to be aggregated for qualitative
national status and trend analyses.
Poor - The utility of biological community monitoring
derives from its direct nature. One is monitoring the
feature of the environment that water quality regulations
seek to protect, so that one cannot be fooled into falsely
believing the ecological protection goal of the CWA has been
met, as can occur when physical and chemical measures
are used. Attempting to infer what biological conditions
are from non-biological measures would thus be contrary
to the basic reason that biological community measures are
desirable in the first place.
Fair - Biological community assessments are done in most
States, but often in only a small portion of the assessed
waters. The major challenge in using these assessments to
evaluate national status and trends will be working out
spatial data availability issues. It will be necessary to
identify segments where monitoring could be expected to be
representative of the segment as a whole, rather than of
83
-------
VL BIOCOMMUNTTY
Temporal representativeness
Utility in trend assessment
Relation to ultimate impact
Related factors/
Ancillary information
Scientific defensibilitv
Sensitivity to change
EVALUATION CRITERIA
small, high impact areas subject to intensive study (such
as a point source discharge monitoring study).
While this issue of spatial representativeness is not a
trivial one, biological community measures in fact present
less of a problem than the chemical measures on which
State use-support assessments are traditionally based.
This is because biological communities tend to integrate
water quality conditions over space and time.
Good - Biological community health is more temporally
consistent in a given location than water quality itself is,
because water column conditions are transient while
organisms remain.
Fair - Biological community monitoring is sometimes
part of State monitoring networks designed for trend
assessment, in which case monitoring is repeated at fixed
locations over time. But other biomonitoring data are from
one-time studies that do not support temporal trend
assessments, and so would have to be excluded from a
national assessment intended to look systematically at
spatial and temporal trends.
Good - Damage to biological communities is one of the
ultimate impacts the CWA seeks to prevent. In relation to
the "fishable goal", if the particular measure used assesses
organisms other than fish (e.g., a benthic community
index), then the measure is still a very good indirect
indicator of impacts on fish, due to food web relationships.
Important - Data on habitat and water body type are
necessary to properly interpret biological community
data, because the types of organisms composing a healthy
community vary according to substrate, depth, flow,
climate, etc.
Good - Most biological community measures currently
used by States have been developed, reviewed, and refined
by academic and government scientists and are very sound
technically. The concept of considering different measures
or indices as qualitatively equivalent for national
assessment purposes is tentatively considered sound by a
selection of federal and State biologists, but requires
testing and further expert evaluation.
Good - Some particularly pollution-sensitive organisms
are typically included in each biological community index,
so that the biological community monitoring is an excellent
method to identify pollution impacts when they first
become ecologically significant.
84
-------
VI. BIOCOMMUNTTY
Relationship to risk
Cost to collect and analyze data
EVALUATION CRITERIA
Good - Degradation of biological community structure is a
direct measure of ecological impact.
Moderate - Biological monitoring can be more expensive
than basic chemical monitoring for conventional
pollutants, but it is often less expensive than toxic
chemical monitoring as presently done, and is much less
expensive than full-blown monitoring for all toxics of
potential concern.
Relationship to existing programs
Presentation value
Good - States and EPA already have made provision for
reporting and analyzing biological community data as part
of the process for assessing use support and preparing
305(b) reports. Guidance on how to incorporate biological
community parameters more explicitly into State
standards will be refined by EPA in the next few years.
Good - The concept of balanced biological communities is
well understood by decision makers and the public.
85
-------
86
-------
VI. BIOCOMMUNTTY
DISCUSSION
Background
One of the objectives of the CWA is the restoration and maintenance of the
physical, chemical, and biological integrity of surface waters. Measures of the biological
health of aquatic communities are an important component of any scheme designed to
evaluate the quality of surface water resources. Still until recently, most water quality
classification schemes have depended to a much greater degree on physical and chemical
parameters than biological ones. There are several possible reasons for this, including
the lack of easily defensible definitions of biological integrity, the perception that
biological data were too costly to acquire, and the presumption that physical and
chemical parameters were adequate surrogates of the health of the resource (Karr 1989).
Currently, these perceptions are viewed as misperceptions and there is a great deal
of effort in place by State and federal agencies to encourage the increased use of
biological community measures as part of States water quality evaluation activities. A
workshop held last March by EPA's Region IV on Biomonitoring and Biocriteria
evaluated current State efforts, reached a number of conclusions concerning the value of
biocriteria and made recommendations for ways to improve the use of biocriteria in the
Region. One of the recommendations of the workshop was that EPA should help
develop a menu of various scientifically acceptable methods that could be used
throughout the Region, as chosen by individual States (Region IV 1989).
At EPA, OW is actively engaged in a number of activities to help provide this
kind of guidance to the States. OW is working on a policy statement for establishing
biological criteria in water quality programs. One workgroup is working to develop
technical guidance on appropriate biological criteria for evaluating monitoring data, while
another is developing programmatic guidance that is directed more to State water quality
managers on how to use the information.
Characteristics of the Indicators
The true value of the use of biological criteria stems from the ability of this
information to integrate different physical, biological and chemical effects, thus providing
the investigator with a summary perspective on a number of factors involved in water
quality. A listing of some of the major advantages of biological community information
to assess environmental quality are set forth in a report from the Ohio EPA (1988) and
are quoted below:
• Some organism groups, particularly fish and many macroinvertebrates,
inhabit the receiving waters continuously and as such are a reflection of the
chemical, physical, and biological history of the receiving waters
• Resident biological communities are integrators of the prevailing and past
chemical, physical, and biological history of the receiving waters, i.e. they
87
-------
VI. BIOCOMMUNTTY
reflect the dynamic spatial and temporal interactions of stream flow,
pollutant loadings, toxicity, habitat, and chemical quality that are not
comprehensively measured by chemical or short-term bioassay results alone
• Many fish species and invertebrate taxa have life spans of several years (2-
10 yrs. and longer), thus the condition of the biota is an indication of past
and recent environmental conditions. Biological surveys need not be
conducted under absolute "worse case" conditions to provide a
comprehensive and meaningful evaluation of use attainment/non-attainment.
A finding that biological integrity is being achieved not only reflects the
current healthy condition, but also means that the community has withstood
and recovered from any short-term stresses that may have occurred prior to
the field sampling.
• Biological community condition portrays the results of water quality
management efforts in direct terms, i.e. increases and decreases in
community health (as reflected by structure and function abundance of
certain species, etc.) is a meaningful measure of regulatory program
progress and attainment/non-attainment of legislative goals
• Minimal manipulation of data using adjustment or uncertainty factors is
necessary
• Biological assessment techniques have progressed to the point that
incremental degrees and types of degradation can be determined and
presented as numerical evaluations (e.g. Index of Biotic Integrity,
Invertebrate Community Index, etc.) that have relative meaning to non-
biologists."
There are several different types of biological community measures ranging from
the simple observation of the well-being of a site to much more complex indices
incorporating a number of parameters (or metrics). Biological community measures also
differ in the types and numbers of the species analyzed. Classification schemes generally
measure either fish or macroinvertebrate species, and neither type of assemblage is
necessarily superior to the other. Each, however, has its advocates and potential
advantages. Fish assemblages provide direct information on one of the CWA goals
(fishability), and since they migrate, fish can integrate data from spatially separate areas.
Moreover, the general public is very interested in the status of fisheries.
Macroinvertebrates, some contend, are superior to fish since they are more easily
collected and analyzed and are often more sensitive to potential pollutants, providing a
better indicator of environmental stress and of water quality. James Karr, author of one
of the primary indices of fish assemblages known as the Index of Biotic Integrity, (IBI)
notes that biologists need to move beyond what he describes as taxa turf battles and
recognize that fish are not necessarily better than benthic organisms or visa versa.
Insights, he points out, come from both (Karr 1989).
Karr notes that the most important element of a biocriteria system is the
development of a conceptual framework involving a number of metrics (parameters).
The IBI consists of 12 different metrics which fall into three categories, species richness,
trophic composition, and fish abundance and condition, each of which can be given a
88
-------
VI. BIOCOMMUNTTY
score of 1, 3, or 5. The cumulative scope for a stream segment (up to 60) can then be
used to determine whether the stream conditions are excellent (58-60), good (48-52), fair
(40-44), or poor (28-34). It is extremely important that in valuing the individual metrics,
the results are assessed relative to regional expectations. These expectations would differ
across different parts of the State and the use of ecoregions can help define these
specific regional expectations. In Ohio, for example, an ecoregion approach is used with
differing expectations if the stream segment is in a headwater or is part of a large or
medium-sized river.
Indices also exist or can be adapted for macroinvertebrates such as the MBI
(Macroinvertebrate Biotic Index) of Ohio's ICI (Invertebrate Community Index), for
overall stream quality (Index of Well-Being) and for other types of waterbodies including
lakes and estuaries. The value of an integrated system that depends on individual
metrics is the ability it provides the scientist to look beyond a final number assigned to
a segment to further investigate the underlying reasons.
Uses Of Biological Community Measures As Indicators
Recognizing that biological community measures cannot, by themselves, substitute
for more comprehensive analyses of water quality that also include physical and chemical
parameter evaluation and habitat analysis, one objective of this project was to begin to
determine whether biological community measures, on their own, could serve as a
national indicator of the quality of the water resource. We wanted to answer the
question: "Can we take a State's determination of the biological well-being of a
waterbody (stream, river, lake or estuary) and compare it with other states evaluations
(based perhaps on completely different criteria) to develop a national picture of water
quality?" To develop a national indicator, EPA is considering combining the various
State qualitative rankings, considering all segments of the same qualitative rank (no
matter how the State determined it) as equivalent.
TBS conducted a survey of a number of State program managers to discover,
among other things, their views concerning this possibility of developing a national
indicator based on qualitative State assessments. There was general consensus
concerning the value of biocommunity data as an indicator of water quality. Some
respondents did raise concerns about developing a national indicator, including the
general paucity of State data, and the problems of comparing area beyond a regional
perspective. Bob Hughes, EPA's Environmental Research Laboratory in Corvallis,
questioned the use of State data in the absence of guidance from EPA regarding
appropriate measures, metrics, and some idea of how to determine health. There is, in
his view, too much variation in the quality of State programs at present to allow for
their use as a national indicator.
Still, these concerns would exist regarding the use of virtually any State collected
data as a national indicator and other respondents were very supportive of the qualitative
national indicator. On the following page, Table VI-1 lists the responses received from
several States to our informal survey.
89
-------
Table VI-1
Comments Regarding the Utility of State Biocommunity Data as a National Indicator
Results of TBS Interviews
Massachusetts: They feel benthic macroinvertebrates are more valid on a national
scale, fish are more variable geographically. Due to different methodologies, quantitative
comparisons between States aren't valid, but conclusions reached can be compared
Biological or chemical information alone is not sufficient to assess water quality,
therefore biological and chemical data should be used together.
New York: They view biocommunity measures as a valid national indicator.
Rapid bioassessments were cited as one area where most States use comparable rankings.
Vermont: A nationwide biocommunity evaluation would be valuable for certain
purposes, but poor interpretation can often lead to incorrect conclusions. Comparison of
macroinvertebrate data are rather limited due to inconsistencies among States.
Texas: Macroinvertebrate and fish data should be comparable, but biocommunity
data are not feasible as a nationwide indicator with the data currently available. They
were opposed to nation-wide trends, due to the frequency with which stream water
quality can change. In their opinion, it is not feasible to spend the resources necessary to
assess all the waters on a regular basis.
Indiana: It is difficult to determine water quality based solely on biocommunity
measures. They do not see how EPA could do it, because of the cost. Every State is
doing biomonitoring slightly differently, and it would be tough to standardize.
Maine: Macroinvertebrates are better on a national level than fish, because States
such as ME don't have many fish species.
Nebraska: A national biocommunity indicator is not possible (due to geographic
differences), but it would be worthwhile on a regional level.
Kansas: They had limited knowledge regarding other States biomeasures, but felt
a nationwide evaluation based on biocommunity measures would be worthwhile for
administrators and congress.
Pennsylvania: Using biocommunities as a nationwide indicator is extremely
worthwhile, although there would be some difficulties with standardization and
interpretation.
Kentucky: Biocriteria are a very good idea. Sets of data work well together, but
fish are sometimes a little misleading because streams don't cooperate.
South Carolina: Biomeasures should be utilized more. Their programs are fairly
comparable with other region IV States.
Ohio: A national biocommunity indicator would be appropriate. Biocommunity
data allow integration of a number of effects.
90
-------
VI. BIOCOMMUNTTY
Data Availability
Data for this measure are collected by individual States and there is, of course, a
lot of variation in the type and quality of the data collected. As an initial step in the
effort to determine the feasibility of this indicator, we contacted 20 States to determine
whether they collect biological community data, the uses to which the data are put and
the availability of the data for use in the indicator. A list of the States and individuals
contacted is included in Table VI-2. We were guided in our selection by discussion
with parties from the March workshop, including Jim Harrison of EPA and Ed Rankin of
the Ohio Environmental Protection Agency, as well as by a report recently completed for
the EPA by Research Triangle Institute (RTI) that provides a comprehensive review of
existing biomonitoring activities in various States (RTI 1989). Figures VI-1 and VI-2,
from the RTI report, show the type of monitoring activity in the different States and the
types of species evaluated. Our interviews were intended to complement rather than
duplicate the RTI effort. For example, in our interview we asked program managers to
identify how much of the river systems in their States could be characterized by
biocommunity data. Few States could provide an estimate, and when given, the
estimates were often small.
We wanted to determine the type of monitoring now being conducted in the State,
the availability of the data (is it automated?) and whether the data could presently be
mapped to provide an indication of State riverine quality.
Brief descriptions of the biological community activity in a number of the States
we contacted is included below. The list is not inclusive but is meant to provide a
representative picture of the types of activities that are being undertaken among the
States.
Florida: The State uses the Shannon-Weiner diversity index which is based on
macroinvertebrates, and the Beck's Biotic Index, a more qualitative index that is then
associated with a specific river.
For purposes of the 305 (b) reports, this information is used as part of an overall
WQI. Unfortunately, few data are available. Of the almost 1,500 reaches in their WBS,
there are biological data at about 100 of them ~ and a fair amount are old data sets.
North Carolina: The State currently uses a macrobenthic index which
characterizes the macroinvertebrate community at the site as excellent, good, fair, or
poor, based on the degree to which the benthic community compares to the benchmark
criteria established for its ecoregion. The monitoring system is a follow on to the Basic
Water Monitoring Program that was first designed at the national level in 1978. Thirty-
seven sites in the current system are part of an on-going ambient water quality
monitoring program. The data is used for trend monitoring and in designated use
determinations.
The State occasionally produces special reports and can map high quality waters
although this requires a lot of field work to test the limits to which the site-specific data
can be extrapolated along the stream. The Benthic Macroinvertebrate Ambient Network
(BMAN), report issued by the State includes maps which identify the sites and
characterize them as good, excellent, etc.
91
-------
Table VI-2
Biocommunity Contacts by State
State
Florida:
Indiana:
Kansas:
Kentuck:
Massachusetts:
Nebraska!
Nevada:
New York:
North Carolina:
Ohio:
Oklahoma:
Pennsylvania:
Rhode Island:
South Carolina:
Tennessee:
Texas:
Vermont;
Contact
John Geyse
Landon Ross
Joel Cross
Lee Bridges
Steve Cringan
Skip Call
Dave Courtemanche
Arthur Johnson
KenBazata
Marvin Burgoyne
Bob Bode
Diane Reed
EdRankin
Derrick Smithee
RodKime
Bob Richardson
Russell Sherer
DellRecter
Dave Buzan
Doug Burnham
Telephone Number
(501) 562-7444
(904) 487-2245
(217) 782-3362
(317)243-5030
(913) 296-5571
(502) 564-3410
(207) 289-3901
(508)366-9181
(402) 471-2186
(702) 789-0500
(518) 432-2624
(919) 733-6946
(614) 466-3700
(405) 271-2541
(717) 787-9633
(401) 277-6519
(803) 734-5300
(615) 262-6327
(512) 463-8471
(802) 244-5638
92
-------
Figure VI-1
Approach Used in River Biosurvey Monitoring Programs (1989)
VO
CO
Puerto Rico
Virgin Islands
I I NONE (8)
V77A FIXED STATION (9)
INTENSIVE SURVEY (12)
FIXED STATION AND INTENSIVE SURVEY (18)
FIXED STATION AND/OR INTENSIVE
SURVEY AND/OR ECOREGION (6) Source: RTI
-------
Figure VI-2
Communities Sampled in River Biosurveys (1989)
VO
Puerto Rico
Virgin Islands
NONE (8)
FISH (2)
OTHER BIOTA (18)
MACROINVERTEBRATES (9)
FISH AND MACROINVERTEBRATES (16)
Source: RTI
-------
VI. BIOCOMMUNTTY
The State is also working on the development of an IBI, but the program is still
in the development phase.
Maine: The State's water quality classification law identifies narrative water
quality standards for water quality classes, and designates a specific level of biological
integrity that each class must maintain. The levels of integrity range from the need to
maintain the structure and function of the aquatic community (Class C) to the need for
the aquatic life to be as naturally occurs (Class AA and A). A specific set of definitions
identifies ecological attributes (metrics) which may be tested to see if a given standard is
being achieved. The system goes beyond the use of a single index by relying on a
number of impact indices, "yes/no" decision that must be made on parameters.
The State uses the data in use attainment determinations and is in the process of
classifying all its waters using the community information. This process will not be
complete for a number of months.
Ohio: The State uses three biological indices, the Index of Well-Being (IWB),
the IBI and the Invertebrate Community Index. The IWB and the IBI are modified so
as to be appropriate to the specific ecoregion in which the sampling site is located. The
data are put to a wide range of uses including determination of attainment of CWA
goals, determination of designated uses, targeting, enforcement, and program evaluation.
Ohio is planning to use the WBS in its next 305 (b) reports and is now working to
identify reach numbers for its stream segments, which will allow State-wide maps to be
drawn in the future.
Illinois: The State uses both an IBI and a MBI and the data are available in the
305 (b) reports with the reach number of the appropriate stream segment. As well as the
IBI, they have developed a measure of the Potential IBI (PIBI) that is derived from
habitat data at the site. The PIBI identifies the potential IBI score at that site based on
the habitat information and can be used as an indicator of degradation.
The community data is used in designated use determinations and also as a
targeting tool. The State has developed a program to guide use of construction use
funds based on the community data.
The State has developed a stream classification system that integrates information
on the IBI, the PIBI and the MBI at various river sites. A report on this system, with
maps, is now at the printer and should be available in the next few weeks.
Arkansas: The State is in the process of developing a narrative fish community
measure that will primarily be used on a site-specific basis to help managers judge if the
community is being adequately protected. The IBI will be used as a standard against
which to measure this. The State used to have a large ambient monitoring system on
benthic data but they are moving away from this toward the use of fish. The State's
plan is to increase the use of rapid bioassessment techniques around expected sources of
toxicants.
Vermont: The State monitors both macroinvertebrate and fish communities, using
a modification of Hilsenhoff's biotic index and an IBI for fish populations. Currently,
95
-------
VI. BIOCOMMUNTTY
only 40-50 sites are monitored annually either as part of regular monitoring and as
special studies. The data are currently stored in State data bases and the State intends to
use WBS identification numbers in the near future.
Texas: The State assesses both invertebrate and fish communities, primarily in
rivers and streams. It is planning to begin use of an IBI for its fish community data.
The primary use of the data is for permit evaluations, and determining potential stream
uses. Currently, only about 5 percent of the State's stream miles are assessed for
biocommunity data.
New York: The State conducts biomonitoring on macroinvertebrates, using five
different parameters including: EPT values, species dominance and species diversity. A
composite value of the five indices is used to characterized the waterway. The State
currently monitors approximately 100 stations both ambient and special studies. The data
are automated on State computers and the State may use BIOS in the future.
The results of the TBS interviews demonstrated, as expected, a lot of variation in
the quality and type of biocommunity monitoring activity being undertaken by the States.
In most cases, data are available on State databases and would presumably be available
to EPA. As will be discussed later, EPA could more readily make use of the data, if it
were stored in a common database, such as BIOS, or included in State 305(b) reports.
Presentation Of The Indicator
Data from the different States could be presented in a number of ways to reflect
national status and, over time, trends. For example, bar charts could be developed
showing how many miles, or what percentage of assessed miles, are evaluated as
"excellent, good, or fair etc." Alternatively, a national map could be developed with the
individual States shaded to indicate the percentage of their waters or their assessed
waters that are in the various categories. The use of arrows on the State maps (similar
to that demonstrated in earlier versions of this report) could show trends in the State
data.
Another possible method of presentation for the indicator would be a series of
maps, showing the biological quality rating of rivers and streams in the different States.
In developing the maps, EPA could use the State's determination of whether the
particular segment has "excellent", "good", "fair" or "poor" biological characteristics (or a
similar 4-5 level rating scheme) without judging these determinations.
Figures VI-3 and VI-4 show how biological community data (IBI and MBI
respectively) could be presented using data from the State of Illinois 305 (b) report
Using data available from the State's own database file, TBS created a file that identified
an IBI or MBI measure for stream reach segments. TBS then assigned, using Illinois'
descriptions of excellent, good, etc; valves to the different scores. The qualitative
descriptions that go with the various scores are shown below.
Excellent Good Fair
IBI 60-51 50-41 40-31
MBI 0-5 5-6 6-7.5
96
-------
Figure VI-3
Illinois Streams Evaluated Using
Index of Biotic Integrity (IBI) (1989)
Note: This is a color map: If Report is photocopied,
use black and white map on next page for
distribution. Color maps are available from the
Environmental Results Branch, U.S. EPA (Phone
(FTS/202) 382-4900)
ENVIRONMENTAL PROTECTION flCENCY
STORET SYSTEM
ILLINOIS RERCHE3
BIO-COMMUNITY DATA
NOVEMBER 14 1989
'IBI2.PRN"
I***************-*
SCKLE OF MILES
a it <» ot tt
3=GOOO
- I=FH[R
2=POOR
********* f
PROJECTION - ftLBERS EOUflL
3CF)LE 1:1048312
-------
Figure VI-4
Illinois Streams Evaluated Using
Index of Biotic Integrity (IBI) (1989)
<
y
UJ% f
}<*: *
^
Note: This map was reproduced from a color map /
generated from STORET to allow photocopying in \
black and white. Color maps are available from the
Environmental Results Branch, U.S. EPA (Phone
(FTS/202) 382-4900)
.a,*
ENVIRONMENTAL PROTECTION AGENCY
3TORET SYSTEM
ILLINOIS REACHES
BIO-COMMUNITY OftTft
NOVEMBER 14 1989
'IBI2.PRN'
99
-------
100
-------
Figure VI-5
Illinois Streams Evaluated Using
Macroinvertebrate Biotic Index (MBI) (1989)
Note: This is a color map: If Report is photocopied,
use black and white map on next page for
distribution. Color maps are available from the
Environmental Results Branch, U.S. EPA (Phone
(FTS/202) 382-4900)
ENVIRONMENTAL PROTECTION ftGENCY
STORE! SYSTEM
ILLINOIS REACHES
BIO-COMMUNITY DftTft
NOVEMBER 14 1989
'MBI2.PRN*
4=EXCELLENT
• 3=GOOO
- l=FfllR
2=POOR
30M-E OF RILES
» zt 49 ra at
PROJECTION - ftLBERS EQUAL ftREft
3CBLE 1:1048312
-------
Figure VI-6
Illinois Streams Evaluated Using
Macroinvertebrate Biotic Index (MBI) (1989)
ENVIRONMENTAL PROTECTION AGENCY
3TORET SYSTEM
ILLINOIS REACHES
BIO-COMMUNITY DftTfi
NOVEMBER 14 1989
'MBI2.PRN"
Note: This map was reproduced from a coloring)
generated from STORET to allow photocopying
black and white. Color maps are available from the
Environmental Results Branch, U.S. EPA (Phone
(FTS/202) 382-4900)
/
103
-------
VI. BIOCOMMUNTTY
With the invaluable assistance of Tom Pandolfi of EPA's Office of Water, the
data was sent through STORET which created the color maps. Similar maps could be
developed for States that can provide reach numbers to correlate with their monitoring
sites. Black and white versions of these maps are also included in the original reports to
retain pertinent information after photocopying. Color maps may be ordered from the
Environmental Results Branch, EPA Headquarters (phone FTS/202-382-4900).
It is interesting to note the differences that exist between the IBI and MBI
evaluations of the stream's well-being. First, some stream reaches had only an IBI or an
MBI score, but not both. In those reaches that did have scores for both, the IBI and
MBI scores gave the same qualitative ranking only 32% of the time, however, the
rankings were within one level of each other 80% of the time. Joel Cross, of the
Illinois Department of Environmental Conservation pointed out that the different indices
(IBI and MBI) measure different attributes of water quality and this would contribute to
the differences. The discrepancies could be due either to the representativeness of the
sample or to the nature of the index used. This is where professional judgment comes
in. To address these differences, Illinois has developed a Biological Stream
Classification (BSC) System that incorporates both biocommunity measures and water
quality indices into one value. The State has recently mapped its waters using the BSC.
During our interviews, we asked the States about the feasibility of mapping then-
data. None of the States with whom we spoke had then- information correlated with
reach numbers that could be easily mapped through the use of STORET or BIOS,
although the State of Ohio is in the process of trying to correlate their data with reach
numbers. Without some type of geographical tag, mapping is not feasible. A number of
States, including Indiana, New York, Pennsylvania, and Nebraska have latitude/longitude
identifiers for their data and one might be able to map these. To do so, one would have
to be able to estimate how far from the monitoring station the qualitative evaluation
would apply.
Next Steps
EPA could get data from a number of State databases and incorporate it into a
national indicator. However, obtaining individual State records would require a lot of
time and effort. To develop an ongoing national indicator, the Agency could consider
the following.
The 305 (b) guidance document could include specific directions for the separate
reporting of stream, river, lake and estuarine areas where biological community
information is available (and what it is), and the degree to which it is being used in
designated use determinations. Again, the use of reach numbers in the reporting would
greatly enhance its utility. If this information were supplied, along with the assessment
of the biological community health at the reach, then maps could be generated directly
out of the waterbody system, and over time trends could be more easily demonstrated.
In addition, the identification of the extent to which biological data have been used in
use determination (measured in river miles for example) would provide an indication
(albeit an administrative one) of the success that EPA has had in encouraging States to
include this information in their water quality assessments.
104
-------
VII. LOADING ESTIMATES FROM POINT SOURCES
SUMMARY
Description of the Indicator
This indicator shows the location, magnitude, type and timing of pollutant discharges into
receiving surface waters. Although loading estimates do not directly reflect the quality of the
water resource, they are a useful measure of the pollutant stress placed on the system and
provide an indication of the effectiveness of regulatory programs in controlling pollutant
discharges.
Possible Applications
Pollutant loading estimates can be used for evaluating progress made in some point
source control programs, and targeting future point source regulation and enforcement
activities. If time series of loads are available, trends in discharges can be depicted and
correlated with changes in treatment technologies, land use management practices, service
areas and production levels. Pollutant loading estimates are often the primary measures for
evaluating enforcement programs.
Strengths
Weaknesses
Possible Improvements
Data collection and reporting system is in place in all
States. The data are spatially and temporally fairly
consistent, with good correlation between pollutant
loadings and water quality. Indicator is easily understood
by public and policy makers and is closely tied to a major
regulatory program.
Availability and quality of data is limited, especially for
toxic compounds. Current data management system
(Permit Compliance System (PCS)) cannot be used
reliably to compute loads. Additional data collection
requirements are costly.
Require permitted facilities to report seasonal and annual
loads; Modify PCS to allow computation of loads.
105
-------
VH. LOADINGS
Data availability
Data consistency/comparability
Ability to estimate
Spatial representativeness
Temporal representativeness
Utility in trend assessment
Relation to ultimate impact
Related factors
Scientific defensibilitv
EVALUATION CRITERIA
Fair - Discharge Monitoring Reports (DMR) data for
majors are reported to PCS for all States, with varying
levels of participation and quality control. Greatest
amount of monitored data is available for wastewater
volume, and conventional pollutants, with much less data
available for metals and toxic organics. Availability and
quality of data for minors is variable, but is generally
much poorer than for majors.
Good - Analysis methods are generally standardized for
permitted pollutants. Therefore, comparison of estimates
among States or regions is reasonable.
Fair - All load estimates are approximations. When
monitored data do not exist or are suspect, engineering
estimates can be substituted, with substantial reduction in
credibility.
Good - Because all major point sources are included in
PCS, data for majors is spatially representative. Data is
less reliable for minors.
Good - For major point sources, pollutant loadings are
monitored on a daily, weekly or monthly basis. Therefore,
monitoring data are reasonably representative of temporal
variation. Monitoring for minors can be less frequent and
therefore less representative.
Fair - It is impossible to assess trends on an individual
facility basis where time series data exist. However,
national trend assessment is difficult because PCS has not
been fully supported in the past. With proposed
improvements to the system, and better participation by
States, capability to assess trends will improve in the
future.
Fair - Loading estimates do not directly relate to ultimate
water quality impacts. However, correlations have been
observed between load reductions and improvements in
biological community structure and other measures of
water quality.
Important - For useful interpretation of loading trends,
it is critical to collect ancillary data characterizing levels
of industrial production, changes in treatment processes,
increases in service areas, etc.
Good - Load reductions have been correlated to
improvements in water quality.
106
-------
VH. LOADINGS
Sensitivity to change
Relationship to risk
Cost to collect and analyze data
EVALUATION CRITERIA
Good • Loads generally directly reflect changes in
industrial production levels, improvements or failures in
treatment, etc.
Poor - It is difficult to directly relate loads to risk.
Monitoring data to support loading estimates for toxic
compounds of greatest concern frequently are not available.
Moderate - Because self-monitoring and compliance
monitoring systems for the National Pollutant Discharge
Elimination System (NPDES) are already in place,
monitoring and analysis costs are already budgeted by
facilities and States. Increased toxics monitoring is
expensive. PCS would have to be modified to allow
computation of loadings.
Relationship to existing programs
Presentation value
Good - DMR reporting is well established, supported and
continually being improved. There is a clear connection
between load estimates and effectiveness of point source
control programs.
Good - Concept of pollutant loadings is generally accepted
and understandable to government decisionmakers and
public. Status and trends can be graphically portrayed on
charts and maps.
107
-------
108
-------
VH. LOADINGS
DISCUSSION
Background
Estimates of pollutant loads discharged to surface waters from point sources can
be used to track changes in discharges over time resulting from modifications in
treatment technologies, increases or decreases in populations served (in the case of
municipal wastewater treatment plants), or changes in the production levels of industrial
facih'ties. Although loading estimates are not a direct indicator of the quality of the
water resource, they are a useful measure of the pollutant stress being placed on the
system and provide an indication of the effectiveness of regulatory programs in
controlling pollutant discharges. By the scheme presented in Chapter I, they are a level
3 indicator, considered "environmental" rather than "administrative", but less direct than
other types of environmental indicators (see Figure 1-1).
On a national level, there are at least two sources of information that can be used
to make point source loading estimates. The first is EPA's PCS data base, which is
used to track each facility's compliance with the effluent limitations included in its
NPDES discharge permit. The monitoring data stored in PCS is taken from monthly or
quarterly Discharge Monitoring Reports (DMRs) submitted by each facility, and
represents averaged discharge values (usually based on a combination of daily, weekly,
and monthly self-monitoring) for the pollutants specified in the permit.
The second source is EPA's Toxic Release Inventory (TRI). This computerized
data base contains annual estimates of the amount of over 320 toxic chemicals released
directly to water, air, or land. For 1987, estimates had to be reported by all
manufacturing facilities that produced, imported, or processed 75,000 or more pounds of
any of the TRI chemicals, or used in any other manner 10,000 pounds or more of a TRI
chemical; engaged in general manufacturing activities (Standard Industrial Classification
(SIC) categories 20 to 39); and employed the equivalent of ten or more employees. In
subsequent years, the reporting requirements will include facilities handling smaller
amounts of TRI chemicals, thus expanding the inventory.
Information from the TRI is already being used to some extent as a national
environmental indicator. Results from the first year of data collection for TRI, released
in the spring of 1989, have been used in reports by environmental groups and in news
articles to show the national status of toxic chemical releases to all media, and highlight
areas of the country where reported toxic releases are the highest. The Office of Toxic
Substances, Economics and Technology Division has also published a report summarizing
the information collected in the TRI for 1987. The report contains many tables,
graphics, and national maps that present the release data aggregated by industrial activity,
chemical class, and receiving media.
Strengths and weaknesses both sources of point source loading data are evaluated
below.
109
-------
VH. LOADINGS
Uses of the Indicator
Loading estimates are a measure that can be used to understand the type and
magnitude of and change in pollutant inputs being discharged to surface waters. The
concept of loadings is technically simple and easily understood by the general public.
Trends associated with loadings, particularly if they can be presented along with a
discussion of the factors influencing the changes portrayed (for example, a reduction in
oxygen demanding materials discharged as a result of improved treatment), are
straightforward concepts to which the public can relate. While it is important when
presenting loading estimate not to simplify the relationship between discharges and water
quality (for example, a trend showing a reduction in loading does not necessarily
translate into measurably improved water quality), there exists a reasonable association
between the amount of pollutant discharged and the impact on the receiving water.
Data Access and Availability
Both PCS and TRI are national computer data bases. They thus have the benefit
for indicator development of being central storehouses of data that can be processed and
aggregated using either existing or specially written computer programs.
PCS is maintained on EPA's National Computer Center (NCC) mainframe system
in North Carolina. Access to PCS is generally restricted to selected EPA and State
personnel. A series of computer programs has been written to enter and retrieve
information from the data base.
TRI data is available from several sources. The official version of TRI is
maintained on the NCC mainframe. An interface has been developed that allows a user
to browse the data base and generate summary reports of the data. Although this system
can be accessed by users outside of EPA, it is intended primarily for internal EPA use.
A public version is available using the National Library of Medicine's TOXNET data
base system. The interface developed for this system is both more powerful and easier
to use than the NCC interface. Finally, the entire TRI data base can be placed in a
commercial database management system such as DBASE or SAS for customized
processing.
There are several issues related to data availability from PCS. The first concerns
State participation in the PCS data base. In the past, some States have not participated
in PCS, and others have participated only to a limited extent. Over the last few years,
participation has increased as a result of EPA's efforts to provide States with improved
telecommunications links and computer hardware. Figure Vn-1 shows the level of
participation in PCS reported for major facilities for the 4th quarter of FY 1989. During
this period, 35 of the 50 States (70 percent) have DMR data entered for at least 90
percent of majors in their jurisdictions, and only two States, Vermont and Rhode Island,
have no reports entered (Region 1 reports that within a year these two States will begin
participating in the system). To develop a comprehensive set of national loading
estimates, information from States with poor PCS participation would have to be obtained
from individual State offices.
110
-------
Figure VIM
Availability of PCS Data for Making Point Source
Loading Estimates
Percent of DMR Forms from Major Facilities Entered into PCS
(4th Quarter of FY 1989)
Source: Office of Water Enforcement Support Branch
Note: data for U.S. Territories not shown
111
DMR Forms Entered
90% to 100%
80% to 90%
50% to 80%
10% to 50%
| | Missing Data
-------
VH. LOADINGS
The second issue relates to the availability of information for major and minor
facilities. While PCS represents a comprehensive inventory for all permitted facilities
regardless of major/minor designation, entry of DMR data is only required for major
facilities. It is therefore not possible to make loading estimates for minor facilities. This
is not a severe limitation because major facilities are, by definition, more important
sources of pollutant discharge.
The third issue relates to the types of pollutant for which data are available.
Facilities are required to report on only those pollutants that are listed on their permits.
The pollutants for which monitoring is most frequently required (and hence for which the
greatest amount of monitoring data are available) are wastewater volume, biochemical
oxygen demand (BOD) and total suspended solids (TSS). Fewer facilities are required to
monitor for nutrients, and thus less data are available. Much less monitoring is required
for metals and toxic organics. Thus, PCS is most useful as a source of data for making
national loading estimates for flow and conventional pollutants; it has much more limited
utility as a source for loading estimates for nutrients, metals and toxic organics.
Data availability concerns also exist for the TRI. Currently there is about a 20 to
25 percent noncompliance rate for those facilities required to report releases to the
database. This percentage is expected to decrease in the future as more facilities become
aware of the reporting responsibilities and reporting requirements are enforced.
A potentially more important limitation is that data hi the TRI do not represent a
comprehensive inventory of point source discharges. This is because of the volume and
industrial activity criteria used to define the facilities that must report releases. The most
serious omission in this respect is for publicly owned wastewater treatment plants
(POTWs). Although releases to POTWs are reported in the TRI, the ultimate amount of
a substance released from a POTW, which is comprised of the portion of the TRI
releases not removed during treatment plus the amount contributed from other sources
(smaller industrial sources and residential/commercial sources), cannot be easily derived
from the database. FJ*A is currently reviewing the need to expand the list of discharging
activities required to report releases.
The availability of loading estimates for a large number of toxic pollutants is a
strength of the TRI. The 320 plus compounds for which release estimates are required
include metals, organic and inorganic acids, and many other toxic organic compounds.
The TRI does not require reporting for some of the conventional water quality
parameters such as BOD and TSS used to define acceptable levels of wastewater
treatment, but comparisons could be made between TRI releases and loads estimates from
PCS data for toxic compounds as a quality control check.
Data Quality
In general, the quality of data in PCS is reasonable. Depending on the State,
there can be problems associated with missing data values, incorrectly entered monitored
values and/or units, and uncertainty related to the methodology used for collection and
analysis of the wastewater sample. Some of these problems can be identified, and to
some extent corrected, by using various statistical techniques to screen the data.
However, resolving these problems completely is extremely time consuming and
112
-------
VH. LOADINGS
frequently unsuccessful. Therefore, the user must be willing to accept some error in the
loading estimates resulting from these data quality problems.
There are also potential data problems with the TRJ. To date, the quality control
efforts for TRI data have focused primarily on limiting data entry errors and correcting
data incorrectly entered on the reporting form. Only a limited amount of analysis has
been conducted to date to determine if the pollutants and volume of releases reported by
facilities were reasonable given the size and industrial activity of the plant. However, an
effort is underway to conduct facility inspections to verify reported releases. This effort,
along with more quality assurance work is needed to ensure that the release amounts in
the data base represent a realistic estimate for that pollutant and facility.
Data consistency/comparability
There are some consistency problems in PCS related to the units and form used to
report monitored values. Between States, and even within some States, permit values are
reported in different units. Additionally, some pollutants are reported as concentration
values, others as loadings. Finally, some States require that facilities report permit
average values only, some require permit maximums, and some require both. Before
using the data, estimates have to be transformed to standardized units, concentrations
have to be converted to loadings by multiplying by flow, and the type of value (permit
or maximum) has to be determined. However, if these conversions and standardizations
can be made, comparison of estimates among States or regions is reasonable because
chemical analysis methods are generally standardized for permitted pollutants.
Because the TRI estimates are all reported on the same form (Form R), using
standardized units, the consistency and comparability of these data are very good.
Ability to Estimate
National estimates of pollutant loadings from point sources have historically been
difficult to obtain from PCS for several reasons already mentioned. There have been
data availability and quality problems, and a lack of consistency in units and in the way
in which monitoring results are reported. In addition, access to PCS is restricted, and
any computation of loading estimates had to be performed by an external computer
program.
However, over the last year staff at Region 2 have been developing a program
utility called the Effluent Data Statistics (EDS) module that is part of the PCS data
retrieval system. EDS can be used to analyze and graph Discharge Monitoring Report
(DMR) data, generate loading estimates from this data, and aggregate the estimates for a
specified time period by outfall, facility, city, county, state, or river basin. These
estimates can be displayed on a three-dimensional map if latitude/longitude data are
available in PCS.
While the load estimation component of this program is still in the developmental
stages, it has the potential to be a very useful tool for compiling and presenting loading
113
-------
VH. LOADINGS
estimates on a national basis. Examples of the data and graphical output from this
program are shown in Figures VII-2 and VH-3.
In the TRI, there is no need to manipulate the release estimates because they are
already reported as annual loads. The capability to generate loading estimates from this
data base is therefore very good, within the data availability limitations noted above.
Temporal representativeness
For major point sources in PCS, pollutant loadings are monitored on a daily,
weekly or monthly basis. Therefore, monitoring data are reasonably representative of
temporal variation. Monitoring for minors can be less frequent, and therefore is much
less representative.
Data in TRI is only reported as annual releases. Because there is no information
on the timing of the releases during the year, the temporal representativeness of TRI data
must be considered poor.
Spatial representativeness
For States that participate in PCS, all major point sources are included. Therefore,
data for majors is spatially representative.
For the SIC categories reporting in TRI, the estimates are spatially representative.
However, as noted above, there are some serious omissions in types of facilities reporting
to TRI, and therefore the data base has limitations on its spatial coverage.
Utility for Trend Analysis
The capability to depict national trends using data from PCS is dependent on the
past level of State participation in the data base. As noted above, some States have not
participated in the past, though this has improved in the last few years. Therefore, the
present capability to generate trend data on a national level is limited. It should be
noted that trend analysis within regions may be possible because some regions, for
example, Region 5, have had close to one hundred percent participation for several years.
In the future the ability to generate trend information should improve as participation in
the data base increases.
For the TRI, 1987 release estimates are available. Estimates for 1988 are
currently being entered into the system, so that in the near future there will be a limited
capability to conduct some trend analysis. In the long term, there is good potential for
trend analysis if consistent estimation methods and reporting approaches are used from
year to year.
114
-------
Figure VII-2
Example - PCS Point Source Loading Estimates for
Carbonaceous Biochemical Oxygen Demand (CBOD)
(1987-1989)
GARY WASTEWATER TREATMENT PLAN 0N0022S77-001A)
UQMX EFFLUENT CROSS VALUE .
BOO. CARBONACEOUS 05 DAY. 2QC
QUANTITY MAXBAJU HUT 2996 KG/DAY
MOO r
2000
MOO
. Lflffl
JAN PEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
1987
woo r
4000
ZMO
UWI
JAN FEB MAR AW W JUN JUL AUG SEP OCT NOV DEC
1988
2000
MOO
JAN • FEB MAR APR MAY- JUN JUL AUO SEP OCT NOV DEC
1989 .
115
-------
Figure VII-3
PCS Example - Three-Dimensional Map Displaying
Total Suspended Solids (TSS) Loadings
Across Illinois (1988)
116
-------
VH. LOADINGS
Compounding factors
There are many external factors that can influence the amount and timing of
pollutant discharges. These include changes in treatment technologies and the volume
and composition of the wastestream treated as a result of increases or decreases in
production levels or populations served. In compiling and presenting estimates of
pollutant loadings, it is important to present this supporting information to help in the
interpretation of the information shown.
Program applications
The loading estimates that can be developed from PCS data are generated as part
of the self-monitoring requirement in the NPDES program. As such, they can be used
as a measure of program effectiveness by demonstrating the reduction in loadings
resulting from the implementation of more stringent control measures needed to meet
permit requirements.
The release estimates in TRI are compiled as part of a program conducted by the
Office of Toxic Substances, and thus do not have any direct bearing on the effectiveness
of Office of Water programs. Estimates from TRI could be used as a check against the
estimates made from PCS data. However, there is very little overlap between the two
data bases for pollutants for which there is useful data.
Improving the Indicator
There are several improvements that could be made to increase the capability to
make loading estimates from these data sources.
For estimates from PCS, EPA should take actions to ensure that DMR data for all
States are available in the data base. As an interim step, it would be useful to have an
assessment of the availability of pollutant data in the data base by State, region, year and
industrial activity (SIC code or major industrial category). This information could be
used to determine whether there were sufficient data to make it worthwhile using PCS to
make loading estimates or perform trend analysis for a particular pollutant or type of
discharge activity.
EPA should complete the development and testing of Region 2's EDS module of
the PCS software.
EPA could require, as part of the NPDES Discharge Monitoring Report process,
that facilities track and report their cumulative pollutant discharge for the year, and make
some assessment of trend, such as indicating if there is an increase or decrease in
discharge of the pollutant for the same time period in the previous year. As part of this
cumulative loading reporting, the facilities could submit an annual summary of the loads
discharged, which could be compiled by EPA and released as an annual report.
The TRI was designed to provide the public with information about toxic releases,
and, as noted, it is potentially a very useful source for national indicator data.
117
-------
VH. LOADINGS
Improvements could be made by expanding the types of facilities required to report
releases, by implementing a rigorous quality assurance program for the data, and by
requiring some additional information on the timing of the releases. One additional
suggestion to improve the interpretation of the data would be to development some
means to evaluate and portray the relative toxicity of the compounds. This type of
approach would give the public a better understanding of the relative risk caused by the
release of different substances.
118
-------
VIII. SUMMARY AND CONCLUSIONS
EPA's long history of working to develop environmental measures for the surface
water programs reflects the difficulties inherent in trying to define indicators that can
adequately reflect even portions of this complex environmental system. The indicators
that were identified at the March workshop and which are reviewed in this report can,
with some changes in data collection and reporting procedures, begin to provide the
Agency with the type of information needed to support reviews of our progress in
addressing pollution problems of surface waters.
The measures included in this review could be useful at the national level as
indicators of the status and trends of the nation's waters. At the State level, the same
type of information can also be used for targeting purposes or for evaluating of the
effectiveness of specific programs. This is usually not possible at the national level
unless a nationally consistent survey such as the Bioaccumulation study (which is
designed to look at some sites adjacent to suspected or known sources of contamination)
is repeated.
Table VIII-1 summarizes the major advantages and limitations of the proposed
indicators as well as identifying some areas in which the measures could be improved.
The shellfish harvest area classification data available from NOAA could be
incorporated into an EPA indicator reporting process in the near term. While the
indicator has some flaws nationwide (many areas are classified for reasons other than
water quality), it does provide a useful indication of the quality of the coastal waters.
Similarly, data available from the FWS National Contaminant Biomonitoring Program and
NOAA's National Status and Trends Program could be immediately brought into an EPA
indicator reporting program.
It would be most efficient and logical for OW to use the State 305(b) reports as
the primary vehicle through which it develops data on indicators. Improvements in
designated use, trophic status of lakes, toxics in fish and shellfish tissue, and biological
community measures could all be realized through more specific directions for reporting
hi the 305 (b) reports. These changes would place an increased burden on the States but
would greatly improve the ability of the Agency and the States to identify the results of
water quality protection programs. Much of the information is already being collected
by the States and the additional burden is in many cases a reporting one; asking the
States to make available through the 305(b) reports information that they already have,
and to make it available in more useful consistent formats.
The consistent use of the Waterbody System and of individual reach numbers by
the States should be encouraged. The computerized system provides EPA with the
ability to synthesize data and design maps or other graphic presentations easily. The use
of reach numbers would allow for comparisons of all types of data. Data in the Reach
File could be related to "pollution pressure" information contained in other databases that
track population densities or local land use practices. Using several environmental
quality databases linked through the reach file would provide a comprehensive set of
indicator information.
In the long-term, there is some additional monitoring activity that the Agency and
the States might consider in order to develop more meaningful indicators. For example,
119
-------
Table VIH-l
Summary Characteristics of Proposed Indicators
Imiieatars
Designated Use
and Attainment
ofFlshable/
iwimmable"
Evaluation
Goals
Shellfish
Harvest Area
Classifications
Trophic Status
of Lakes
Toxics in Fish
and Shellfish
USFWSNCBP
NOAANS&T
Benthic
Surveillance
NOAANS&T
Mussel Watch
EPA National
Bioaccumulan'on
Study
Tissue Residue
Data Collected
by States
Biological
Community
Measures
Pollutant Loading
Estimates from
Point Sources
''''^£&&&fa'^'^
•Data collection system already in place
•Computerized fnmework
•Low cost to Tuiir*^"1 «nH improve
•Of primary interest to majority of public
•Easily Understood
•Data collected for over 20 years
•Data collection has been fairly consistent
•Easily undentood by general public
•Data collection system already in place
•Report required by CWA 314ffii« it cites and fish
ibnonnalilies
•Spatial and Temporally representative
•Consistent nationwide sampling
•Can correlate info with sediment data
•Spatial and Temporally representative
•Located around known and suspected
sources
•Can draw conclusions on impacts
•Can develop information on
>ioaccumulation
•Being done at stale level
•Used in some use assessment
•Related to health advisories
•Fish integrate impacts over whole stream
•Bivalves are excellent indicator of
environmental stm*
•Heightened interest in states and regions
.rnmmimily A*t* k nrffftmjy
complement to Dhvsical/cbemical lestinB
•Different states f^n use systems that are
appiupiiate to their conditions
•Data collection and reporting in place
•Spatially and temporally representative
•Good correlation between pollutant loadtnffs
levels and water quality
f^*\?jS *? ^*?MttX££^£fc.'ltflLji ****** """ '^'^ 4^f^>
•At pteseul* can not be used to
evaluate trends
•Inconsistent within state from one reporting
cycle to me next
•Variations in state-to-state interpretation of
f*lsvsificatioos
**lstt*i1 to water quality changes
•Rtginnl diP"! **""*** limit national
compansions
•The tmfl*HT of lakes evaluated flwtTTHTif
•| •y*>« are monitored too •"fa»'p«»nt|y to
derive trends
Only for rivers and streams
•Ijtnitwi nufnlMT ^f rhfxnidli tested
•Actual ecokfgical 'ttlp^*<***ii*ns of fish •g**l
•nnllfifth contsmbiaiion an a subject of
iontroversy
•Different fish at different locations
•Not weful for human heallh
considerations
•Not point source specific
•Not point source specific
•One time only study
•No trends
•Not spatially or temporally rep.
•Varies from state to state in species.
amount of testing
•No data storage with easy retrieval
•Relatively limited amount of
part of water quality analyses
•Resource constraints
•No centralized database
•Limitations on availability and quality of dati
•Little *!•*• available for toxic ciiHiipff"11^*'
•Current data management system (PCS) can
not compute loads
•Additional data collection would be costly
^Si^^r^iiail
•Increase in the UM of WBS
•Omtrr cmfftftmry in clmifictiinra
•CofTolfttniA with fltfrffnv^pifonpfl diti
such as NOAA's Mussel Watch data)
•Establish baseline • order to be able to
evaluate mods
•Report seasonal fluctuations
•In 30SOX report both number of lakes and
ake acreage
•Plan to expand to estuaries
•Looking to inrlw^* new toxicants
•May include USDS sites and perhaps
NOAA's Mussel Watch sites
•Coordinate data collection with other
juuites
•Develop centralized database
•Develop protocols on monitoring to
ncrease comparablity with other
mutes data
•Same as above
•Repeat study to assess programs in
>lsce around sources and to further
•Identify sites to be re-tested as part
of nationwide trend study
•Change in 305(b) to provide
summarized results
•Use Bios data system
•Develop agreements on monitoring
Motocols to increase comparability
•Ttifjyff* guidance ID state in terms of
what to do with the inf onnatkm
•Make identification of e*»*"nimiiy
analyses part of the 305(b) reports
•Require permitted facilities to report
t*am»*«l and •twntal loads
•Modify PCS to allow computation of loads
120
-------
vm. CONCLUSIONS
EPA should consider whether it wants to repeat the testing conducted for the National
Bioaccumulation Study. This could provide the Agency with data on the effects of
specific programmatic actions (at the sites being surveyed) as well as information on
trends at the "randomly chosen" sites. Coordination with the Fish and Wildlife Service's
National Contaminant Biomonitoring Program or the new USGS National Ambient Water
Quality Assessment program might offer some opportunities for interagency cost sharing
or data sharing for such future work, although differences in data collection (e.g., whole
fish versus edible tissue sampling) would have to be resolved.
EPA should actively encourage State programs designed to implement measures of
biological community well-being. This information provides a better view of the quality
of the water resource than can be obtained simply from the chemical and physical
parameter testing that still forms the principal basis of most State water quality
monitoring programs, and in particular can give a better indication of problems associated
with non-point sources of pollution.
If it is desired that indicator data be used for national trend assessment, the Office
of Water (OW) and State water quality agencies might reconsider a monitoring strategy
in which each State returns to a subset of fixed monitoring stations on regular intervals
to permit trend reporting on a few indicators. Since only a percentage of total waters is
generally assessed, it is recommended that rotational monitoring be instituted so that the
same fixed stations be monitored at set intervals (e.g., every third year).
121
-------
122
-------
LIST OF REFERENCES
1. Andraison, Jim. Telephone Conversation, September 20, 1989.
2. EA Engineering, Science and Technology, Inc. Review of State and Selected
Federal Biological Monitoring Programs. Prepared for Barbara M. Lamborne,
BIOS Project Manager. September 19, 1987.
3. EPA, Criteria and Standards Division, "The Lake and Reservoir Restoration
Guidance Manual," 1988.
4. EPA, Office of Water, "Report to Congress: Water Quality of the Nation's
Lakes," 1988.
5. GAO, "The Nation's Water: Key Unanswered Questions About the Quality of
Rivers and Streams," 1986.
6. Illinois Environmental Protection Agency, "Phosphorus: A Summary of
Information Regarding Lake Water Quality," 1986.
7. Karr, James. Telephone Conversation, September 18, 1989.
8. LaRoe, Edward T. "Role of the Fish and Wildlife Service in Water Quality
Monitoring". As Reported in Status of the Nation's Water Quality Information
Activities. Proceedings of a joint meeting of the Advisory Committee on Water
Data for Public Use and the Interagency Advisory Committee on Water Data.
pp. 29-36, March 19-21, 1987.
9. Louisiana Department of Environmental Quality, Water Quality Inventory 1988.
305(b) Report, 1988, pp. 3-25.
10. National Oceanic and Atmospheric Administration (NOAA), "National Estuarine
Inventory: The Quality of Shellfish Growing Waters on the East Coast of the
United States," 1989.
11. NOAA, "PCB and Chlorinated Pesticide Contamination in U.S. Fish and Shellfish:
A Historical Assessment Report," Seattle, Washington, (1988).
12. Ohio, Environmental Protection Agency, Biological Criteria for the Protection of
Aquatic Life: Volume 1. The Role of Biological Data in Water Quality
Assessment. Division of Water Quality Monitoring and Assessment Updated
February 15, 1989, pp. 13, 14.
13. Region IV, U.S. Environmental Protection Agency, Workshop on Biomonitoring
and Biocriteria. Prepared by Killelly Environmental Associates, Inc. pp. 15, 16,
August 1989.
123
-------
LIST OF REFERENCES (Continued)
14. Research Triangle Institute (RTI). A Survey of the Status of Biomonitoring in
State NPDES and Nonpoint Source Monitoring Programs. P.A. Cunningham and
C. O. Whittaker. RTI/7839/02-03F. September 29, 1989.
15. Steffick, Don. U.S. Fish and Wildlife Service. Telephone Conversation on
September 15, 1989.
16. State of Vermont, "1988 Water Quality Assessment," 305(b), 1988, p. 32.
17. Vesilind, P., Peirce, J., Weiner, R., Environmental Engineering. 1988, p. 67.
18. Ward, Robert C. et al. "The 'Data-Rich' but Information-Poor Syndrome in Water
Quality Monitoring," Environmental Management. Volume 10, No. 3 (1986),
pp. 291-297.
124
------- |