United States
Environmental Protection
Agency
Office of
Research and Development
Washington, DC 20460
EPA/620/R-92/001
June 1992
Great Lakes
Monitoring and
Research Strategy
Environmental Monitoring and
Assessment Program
-------
EPA/620/R-92/001
June 1992
ENVIRONMENTAL MONITORING
AND ASSESSMENT PROGRAM
(EMAP)
GREAT LAKES MONITORING
AND RESEARCH STRATEGY
Environmental Research Laboratory
Office of Research and Development
United States Environmental Protection Agency
Duluth, Minnesota 55804
June 1992
^ฃp Printed on Recycled Paper
U.S. Environmental Protection Aserscy
Region 5, Library (PL-12J)
77 West Jackson Boulevard, 12th Floor
Chicago, IL 60604-3590
-------
EMAP-GREAT LAKES MONITORING
AND RESEARCH STRATEGY
BY:
Steven Hedtke1
Anne Pilli2
David Dolan3
Gil McRae2
Brian Goodno2
Russell Kreis5
Glenn Warren7
Deborah Swackhamer8
Mary Henry4
with contributions from:
Trefor Reynoldson9
Donald Stevens10
Nancy Leibowitz10
Janet Keough1
Stephen Lozano1
Jeffrey Rosen2
John Eaton1
Robert Hoke6
Technical Director: Steven Hedtke1
Associate Director: John Paul11
1 US Environmental Protection Agency, ERL-Duluth
2 Computer Sciences Corporation, Contract 68-WO-0043 (225-269)
3 International Joint Commission
4 US Fish and Wildlife Service
5 US Environmental Protection Agency, Large Lakes and Rivers Research Branch
6 AScI Corporation
7 US Environmental Protection Agency, Great Lakes National Program Office
8 University of Minnesota, School of Public Health
9 Environment Canada, National Water Research Institute, CCIW
10 Mantech International Corporation
11 US Environmental Protection Agency, ERL-Narragansett
Any mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
-------
Contents
Introduction 1-1
Content and Organization of Research Strategy 1-1
An Overview of EMAP 1-2
Legislative Mandate for Great Lakes Protection 1-3
Great Lakes Monitoring Programs 1-5
Societally Important Great Lakes Values 1-6
Specific Objectives of EMAP - GL 1-7
Focus and Purpose of the Research Strategy 1-7
Overview of Approach 2-1
Introduction 2-1
EMAP - GL Design Approach 2-1
EMAP - GL Indicator Approach 2-4
EMAP - GL Data Quality 2-7
EMAP - GL Expected Outputs 2-9
EMAP - GL Program Limitations 2-10
Questions to be Addressed 2-11
Questions to be Addressed in 1992 2-12
Implementation Plan 2-13
Monitoring Network and Field Sampling Design 3-1
Introduction 3-1
Physical boundaries 3-2
Regionalization 3-2
Primary Resource Classes 3-3
Offshore and Nearshore Areas 3-3
Harbors and Embayments 3-6
Coastal Wetlands 3-9
Frame Development 3-9
Offshore and Nearshore 3-10
Harbors and Embayments 3-10
Coastal Wetlands 3-10
Issues 3-10
Quality of Frame 3-11
Monitoring Network Design 3-11
Tier 1 Sampling 3-11
Tier 2 Sampling 3-12
Offshore Samples 3-12
Nearshore Samples 3-13
Harbors and Embayments 3-14
Coastal Wetlands 3-14
Tier 3 for EMAP - GL 3-14
Field Sampling Design 3-15
-------
Indicator Development and Evaluation 4-1
Introduction 4-1
Conceptual Framework for Indicators of Condition 4-1
Strategy for Indicator Development and Implementation 4-4
Establishing Nominal Condition 4-6
Response Indicators 4-7
Fish Indicators 4-8
Targeted Invertebrate Populations 4-14
Benthic Community Structure 4-15
Primary Producers/Lake Trophic Status 4-16
Trophic Status Index 4-16
Diatoms (Bacillariophyceae) 4-17
Exposure and Habitat Indicators 4-21
Stressor Indicators 4-26
Relationship of Indicators to Assessment Endpoints 4-27
Biotic Integrity 4-30
Trophic Status 4-33
Application of Indicators to Resource Classes 4-32
Sampling Index Period 4-32
Analyses of Existing Data 4-34
Estimation and Analysis 5-1
Introduction 5-1
Sampling Design 5-1
General Statistical Overview 5-2
Analysis for Biotic Integrity 5-5
Descriptive Statistics/Visualization 5-5
Classification/Cumulative Distribution Functions 5-5
Estimates for Combined Resource Classes 5-6
Analysis of Change and Trend 5-7
Linear Model Analysis of Trend 5-7
Non-Parametric Method for Trend Detection 5-8
Power to Detect Change and Trend 5-9
Associations 5-9
Great Lakes Ecological Condition Index: A Conceptual Proposal 5-10
Logistics Approach 6-1
Introduction 6-1
Logistics Implementation Components 6-1
Logistics Issues 6-4
Staffing 6-4
Access 6-4
Data Confidentiality 6-4
Field Operation Scenario 6-5
General Logistics Scenario 6-5
Organizational Structure 6-5
IV
-------
Quality Assurance Program 7-1
Introduction 7-1
The Data Quality Hierarchy 7-1
The Role of DQOs in EMAP 7-2
Data Quality Requirements 7-3
Precision and Bias 7-5
Comparability 7-5
Completeness 7-5
Representativeness 7-6
Tolerable Background Levels 7-6
Organization and Staffing Requirements 7-6
Quality Assurance Documentation 7-6
Quality Control Guidelines 7-7
Biological Measurements 7-8
Chemical Measurements 7-9
Habitat Quality and Site Characterization Measurements 7-9
Samples and Specimens 7-10
Data Review, Verification, and Validation 7-11
Assessment of Data Quality 7-12
Quality Assurance Reporting 7-13
Information Management 8-1
Overview of EMAP - GL Information Management 8-1
Objectives of EMAP - GL Information Management 8-2
Mission Needs Analysis 8-3
System Users and Information Use 8-3
Process Flow Diagram 8-6
Initial System Concept 8-6
EMAP - GL Processing Environment 8-6
Distributed EMAP - GL IMS . . 8-7
Developmental and Operational Personnel 8-7
Policies, Standards, and Standard Operating Procedures 8-9
Operational Components 8-10
Sample Collection System 8-10
Sample Tracking System 8-11
Field Logistics System 8-13
External Dataset Processing System 8-13
Indicator Development Data Management System 8-15
Data Dictionary System and Documentation 8-16
Data Archival System 8-17
Geographic Information System (GIS) Applications 8-18
Quality Assurance 8-19
Project Management 8-19
Resource Utilization 8-20
-------
Coordination 9-1
Introduction 9-1
Within EMAP 9-1
Other Federal Agencies 9-2
International Activities 9-2
Research Organizations 9-2
Conclusions 9-3
Fiscal Year 1992 Field and Analysis Activities 10-1
Introduction 10-1
Investigations Within the Offshore Resource Class 10-2
Application of the EMAP Offshore Design for Trophic Status 10-2
Application of EMAP Offshore Design to Benthic Community Structure . . 10-7
Sediment Indicators in the Nearshore Resource Class . 10-7
NWRI Study on Sediment Indicators 10-8
Site Selection 10-9
Field Methods 10-10
Laboratory Methods 10-13
Data Analysis 10-14
Reporting 10-14
NOAA-GLERL Project Description for Use of Sediment Traps for Indicator
Measurements 10-14
NOAA-GLERL Project Description for Benthic Survey and Methodology
Comparison (Southern Lake Michigan) 10-15
Development of the Harbors and Embayments Sampling Frame 10-16
Wetland Indicators 10-16
Fish Indicators 10-16
Index Period for Trophic Status Indicators 10-17
Investigations of Diatom Populations as Indicators of Trophic Status and Biotic
Integrity 10-17
References . Ref-1
Appendix A A-1
Appendix B B-1
VI
-------
Figures
Figure 1.1 Conceptual representation of the EMAP concept for a regional
characterization of ecological resources 1-8
Figure 2.1 Concept of a four-tiered approach in EMAP. Spatial coverage is
maximized in lower tiers while temporal coverage increases at the
higher tiers 2-3
Figure 2.2 Concept of the EMAP four-tiered approach as applied to EMAP -
Great Lakes 2-4
Figure 2.3 General approach for selection and development of indicators for
EMAP 2-6
Figure 3.1 Total phosphorus concentration (mg/L) (Lake Michigan) measured at
85 m, 50 m, and 30 m 3-6
Figure 3.2 Total chlorophyll-a concentration (u,g/L) (Lake Michigan) measured at
85 m, 50 m, and 30 m 3-6
Figure 3.3 Definition of embayments for EMAP - GL 3-7
Figure 3.4 EMAP base grid cells within the Great Lakes 3-11
Figure 3.5 Eighty-five meter depth contour for Lake Michigan with EMAP base
grid for offshore zone, 3-fold grid for nearshore zone 3-13
Figure 3.6 GLISP surveillance stations for Lake Michigan 3-13
Figure 4.1 Indicator selection, prioritization, and evaluation approach for EMAP. 4-5
Figure 4.2 Great Lakes Areas of Concern 4-22
Figure 4.3 Great Lakes land use distribution 4-28
Figure 4.4 Great Lakes population distribution 4-29
Figure 4.5 Pollution sources and trophic status of the Great Lakes 4-33
Figure 4.6 Cumulative Frequency Distribution of the Composite Trophic Index
(CTI) calculated for the nearshore zone of Lake Michigan, Spring
1976 4-34
Figure 5.1 Great Lakes ecological condition index: A conceptual proposal. . . . 5-11
Figure 8.1 EMAP user community 8-2
Figure 8.2 Relationship of classes of EMAP users and data 8-5
Figure 8.3 EMAP - GL node configuration and networking 8-8
Figure 10.1 1992 EMAP(D) and GLISPH sampling stations for Lakes Michigan
and Superior 10-3
VII
-------
Tables
Table 2.1 Indicators being considered for EMAP - GL 2-8
Table 2.2 EMAP - GL implementation schedule : 2-14
Table 3.1 Physical and geographical characteristics of the five Great Lakes .... 3-3
Table 3.2 Watershed characteristics of the Great Lakes connecting channels. . . 3-4
Table 3.3 Percent of Lake Michigan surface area for the nearshore resource
class as defined by depth contour 3-5
Table 3.4 Summary of sampling grid points for the nearshore zone using the
base grid, 3-fold, 4-fold, and 7-fold enhancement 3-12
Table 4.1 Indicators being considered for EMAP - GL 4-2
Table 4.2 Chronology of EMAP indicator development for the Great Lakes .... 4-8
Table 4.3 Chronology of first appearance of certain exotic species in the Great
Lakes 4-10
Table 4.4 Proposed oligotrophic fish community metrics for EMAP - GL 4-11
Table 4.5 Ecological properties of harmonic fish communities and astatic
assemblages in mesotrophic waters of the Great Lakes 4-14
Table 4.6 Apparent tolerances of Great Lakes diatoms to trophic conditions . . . 4-20
Table 4.7 IJC Areas of Concern: Summary of Use Impairment Identified by the
Jurisdictions in Areas of Concern and Whether or not Problem
Definition and Description of Causes is Complete 4-23
Table 4.8 IJC Critical Pollutant List1 (GLWQB 1987) and additional persistent
toxic substances in the Great Lakes2 : 4-25
Table 4.9 Proposed indicators, by resource class, of ecological condition for
EMAP - GL 4-33
Table 4.10 Summary of Lake Michigan Database Retrieved from STORET and Used
to Calculate Composite Trophic Indices (CTIs) 4-35
Table 6.1 EMAP Logistical Elements for Implementation of Great Lakes
Monitoring Programs 6-1
Table 7.1 Criteria for Selection of Appropriate Sampling and Analytical (or
Measurement) Methodology 7-4
Table 7.2 Quality Assurance Related Documentation of EMAP - GL 7-8
Table 7.3 Quality Control Activities Associated with Chemical Measurements. . 7-10
Table 10.1 1992 EMAP - GL Field Pilot Activities 10-4
Table 10.2 1992 EMAP - GL Design and Analysis Pilot Activities 10-5
Table 10.3 Great Lakes Historical Datasets Compiled by EMAP - GL 10-6
Table 10.4 Number of ecodistricts and sites in each Great Lake 10-10
Table 10.5 Geophysical parameters measured at each site 10-11
VIII
-------
1. Introduction
1.1. Content and Organization of Research Strategy
The research strategy for the Great Lakes component of the Environmental Monitoring
and Assessment Program (EMAP) is organized into ten chapters. A general intro-
duction to the EMAP - Great Lakes (EMAP - GL) research strategy components, their
focus and objectives, followed by a summary of the approach for meeting each of
these objectives, is described in this chapter. Each successive chapter provides
additional levels of detail about specific aspects of the approach such as design,
indicators, or information management, ending with planned 1992 field activities. The
sections of the research strategy are:
Chapter 1. Introduction - provides a general introduction to the focus and
objectives of subsequent chapters;
Chapter 2. Overview of Approach - provides an overview of all aspects of EMAP -
GL;
Chapter 3. Monitoring Network and Field Sampling Design - contains a detailed
description of the sampling design and site selection;
Chapter 4. Indicator Development and Evaluation - discusses selection of
indicators, their intended use, and the process of continual
improvement in achieving descriptions of ecological condition for the
Great Lakes;
Chapter 5. Estimation and Analysis - describes the proposed procedures for the
estimation and analysis of the current status, extent, changes, and
trends in the condition of the Great Lakes;
Chapter 6. Logistics Approach - outlines issues related to conducting a field
program;
Chapter 7. Quality Assurance Program - provides the approach and procedures
for ensuring that the quality of the data collected meets program
objectives;
Chapter 8. Information Management - describes data management procedures;
Chapter 9. Coordination - explains the approach for integration of information
within EMAP - GL, with other components of EMAP, and with
programs outside of EMAP;
1-1
-------
Chapter 10. Fiscal Year 1992 Field and Analysis Activities - outlines the
specific objectives for the field activities and data analysis activities
that have been planned;
Appendix A. Glossary of EMAP Terms - provides standard definitions for
terminology specific to EMAP - GL and the EMAP program; and
Appendix B. Indicator Fact Sheets - provides further detail and references for
the indicators examined for use in EMAP - GL.
1.2. An Overview of EMAP
Both the incidence and scale of reported environmental problems have changed over
the past two decades. The public is increasingly concerned that the resources upon
which they rely for recreation, quality of life, and economic livelihood remain
sustainable. Scientists are increasingly concerned that the impact of pollutants now
extends well beyond the local scale: global climate change, acidic deposition,
deposition of air toxics, ozone depletion, nonpoint source pollutant and sediment
discharges to waterways, and habitat alteration threaten our ecosystems on regional
and global scales. Unfortunately, the current status of our environment on regional
and global scales is often not well documented. While we believe that our policies
and programs are improving the quality of our environment, we often cannot
demonstrate improvement with available data.
In 1988, the US Environmental Protection Agency's (EPA) Science Advisory Board
recommended implementation of a program within the EPA to monitor ecological
status and trends, and to develop innovative methods for anticipating emerging
environmental problems before they reach crisis proportions. The Environmental
Monitoring and Assessment Program (EMAP) is part of the EPA Office of Research
and Development's response to the Science Advisory Board's recommendation. To
meet this need, EMAP was developed around the following objectives:
estimate current status, extent, changes, and trends in indicators of the
condition of the nation's ecological resources on a regional basis with known
confidence;
monitor indicators of pollutant exposure and habitat condition, and seek
associations between human-induced stresses and ecological condition; and
provide annual statistical summaries and periodic interpretive reports on
ecological status and trends to resource managers and the public.
These objectives pose a challenge that cannot be met without a commitment to
environmental monitoring, research, and assessment on long-term regional and
national scales. Furthermore, this challenge cannot be met efficiently without drawing
1-2
-------
on the experience and expertise within other organizations that share responsibility for
maintaining environmental quality or sustaining our resources.
EMAP is designed around six primary activities:
strategic evaluation, testing, and development of indicators of ecological
condition, pollutant exposure, and habitat condition; and protocols for collecting
data on these indicators;
design and evaluation of a comprehensive and versatile integrated monitoring
framework;
nationwide characterization of the extent and location of ecological resources;
demonstration studies and implementation of integrated sampling designs;
development of data handling and quality assurance, as well as spatial analysis
and statistical procedures for efficient analysis and reporting on status and
trends data; and
assessments of the probable causes of environmental conditions and trends.
To facilitate monitoring of the condition of the nation's ecological resources, the
program has been organized into seven basic resource groups: lakes and streams,
inland and coastal wetlands, forests, agroecosystems, estuaries, arid lands, and the
Great Lakes. In addition, there are groups responsible for coordination of indicators,
statistical design and analyses, landscape characterization, integration and
assessment, quality assurance, information management, geographical information
systems, and logistics. The Great Lakes resource group is the newest resource group
of EMAP, with planning initiated in 1990.
1.3. Legislative Mandate for Great Lakes Protection
The Great Lakes basin is one of the largest freshwater ecosystems in the world and
the most intensively used freshwater resource in North America. Great Lakes water
quality issues have been identified as: point and nonpoint source nutrient control,
toxic substance control, remedial programs, inadequate information, science policy
and allocation of research resources, and recommitment to an ecosystem approach
(Dworsky and Allee 1988). This ecosystem has been the focus of environmental
policies and legislation for almost thirty years. Four of the five Great Lakes are
regulated by legislation developed in Canada and the United States, whereas the fifth,
Lake Michigan, lies entirely within the US. The management, protection, and
development of the Great Lakes is under the jurisdiction of two federal governments,
the province of Ontario and eight US states, along with numerous municipal,
provincial, state, federal, regional, and international agencies. Current legislation relies
on water quality objectives which have been limited to specific pollutants. Initial efforts
1-3
-------
and activities, sponsored by the United States and Canada, for defining ecosystem
objectives are also underway. The ecosystem objectives endeavor to protect the
water quality while incorporating the effects of interactive ecosystem components.
The geographic, natural, and social components of the Great Lakes ecosystem play
important roles in the development of water quality management activities.
Prior to 1972, the United States operated under the statutes of the Federal Water
Pollution Control Act (FWPCA). In 1972, a state-oriented system of ambient water
quality standards based on water-use criteria was established (Findley and Farber
1988). Amendments to the FWPCA, in 1972, adopted goals of fishable and
swimmable waters by 1983 and total elimination of pollutant discharges to navigable
waters by 1985. The FWPCA was amended again in 1977 and henceforth is referred
to as the Clean Water Act (CWA). The focus on a water quality approach to
controlling pollutants was emphasized in the federal Water Quality Act of 1987
(Bascietto et al. 1990). This amendment to the CWA requires national assessments
of trophic status and trends, and nonpoint source controls, in addition to assessing
those waters not meeting established standards for priority toxic pollutants.
The United States and Canada initiated a combined major effort in the 1970's to
manage environmental quality in the Great Lakes through the Great Lakes Water
Quality Agreement (WQA). The Great Lakes WQA was signed in 1972 by the
governments of the United States and Canada, primarily in response to eutrophication
in Lakes Erie and Ontario which led to efforts to reduce the phosphorus load.
Revisions to the Great Lakes WQA, signed in 1978, included proposals for addressing
toxic chemical control in addition to a continued concern over phosphorus discharge.
The goal of the Great Lakes WQA is to "restore and maintain the chemical, physical,
and biological integrity of the waters of the Great Lakes Basin Ecosystem" (IJC 1989).
The International Joint Commission (IJC) was assigned the role of assisting in the
implementation of the Great Lakes WQA. Subsequent revisions, signed in 1987,
include annexes to address airborne toxic inputs, non-point source input, Remedial
Action Plan development to clean up severely polluted sites, and development of
ecosystem health objectives.
The Great Lakes Critical Programs Act of 1990 (US Public Law 101-596, 104 Stat.
3000) is the first US legislation which specifically refers to the Great Lakes WQA and
the associated role of the IJC (Chandler and Vechsler 1991). The legislation requires
the EPA to publish proposed water quality guidance for the Great Lakes system which
is at least as restrictive as the provisions of the Great Lakes WQA.
The Nonindigenous Aquatic Nuisance Prevention and Control Act of 1990 (Title I of
US Public Law 101-646, 104 Stat. 4761) addresses the prevention and control of
exotic species in the Great Lakes on the part of multiple US agencies and directs
consultation with the Government of Canada in developing an effective international
prevention program (Chandler and Vechsler 1991). This law was in response to
growing concerns over the introduction of nonindigenous species, particularly the
zebra mussel, into the Great Lakes and the potential for extensive damage to Great
Lakes ecosystems.
1-4
-------
Amendments to the Clean Air Act in 1990 (US Public Law 101-549, 104 Stat. 2399)
contain provisions which will affect the Great Lakes Basin. The EPA is charged with
assessment of the extent of atmospheric deposition of hazardous contaminants in the
Great Lakes, development of an atmospheric deposition network, and assessment of
pollutant loadings and their effects in accordance with the specific objectives of the
Great Lakes WQA (Chandler and Vechsler 1991).
1.4. Great Lakes Monitoring Programs
An understanding of agency involvement is critical to the water quality management
process. Multiple agencies need to work together to achieve common long-term
ecosystem goals while working towards diverse short-term goals. The dimensions of
this challenge can be appreciated with an understanding of the structure and function
of the institutional system. Colborn et al. (1990) provide a discussion of "the
institutional ecosystem" that wrestles with challenges as diverse as habitat
rehabilitation versus dredging and disposal in support of lake-shipping requirements, to
implementation of lifestyle changes in order to reduce stress on the ecosystem. At
each level of organization, ranging from federal to municipal, there are a multitude of
agencies that have responsibilities in regard to environmental legislation.
Prior to 1964, water quality studies on the Great Lakes were conducted largely by
individual universities and state and provincial pollution control and fisheries agencies
(Beeton and Chandler 1963). There was little coordination among these institutions or
their studies. As a result, much of the data was incompatible due to dissimilar
objectives, data collection methods and sampling designs, and analytical methods.
The 1972 Great Lakes WQA outlined a set of common goals for surveillance and
monitoring programs. By 1975 it was clear that coordinated efforts to gather data for
evaluating the condition of the Great Lakes were not in place (IJC 1973, 1975). The
IJC Water Quality Board recommended the development of the Great Lakes
International Surveillance Plan (GLISP) to provide the framework for coordinating, in a
bilaterally comprehensive and cost-effective manner, the various surveillance and
monitoring activities initiated under the Great Lakes WQA. GLISP was developed
over the period 1974-1980 to track phosphorus reductions and the subsequent
impacts. The 1978 Great Lakes WQA shifted the focus to toxic contaminants.
Subsequently the issues which GLISP attempted to address included eutrophication,
toxic contamination, microbiological contaminants, radiological elements, and biological
community and habitat. The common programs of GLISP have produced long-term
trend data for phosphorus and toxic contaminants based on selected, fixed stations in
the offshore areas of the lakes. An integrated program to assess overall ecological
condition has not yet been institutionalized in the Great Lakes.
1-5
-------
1.5. Societally Important Great Lakes Values
To be effective, the information from a Great Lakes monitoring and assessment
program must prompt action when it is required. This means that the information
produced must be related to perceptions regarding aquatic health and represent
issues of concern to the public, aquatic scientists, and decision makers. In ecological
risk assessment, these perceptions are called endpoints. Suter (1990) and Hunsaker
and Carpenter (1990) defined ecological endpoints as the environmental entity of
concern and the descriptor or quality of the entity. Such endpoints represent concepts
that are societally important but that tend to be nebulous or abstract and do not lend
themselves to direct measurement.
Historically, these endpoints for surface waters, including the Great Lakes, have been
expressed as designated uses, which have included habitat for aquatic life, fishability,
swimability, navigation, and drinking water supply. The attainment of some of these
uses depends directly on ecological condition. For others, it is less directly dependent
on ecological condition, although attainment may be indirectly associated with the
consequences of degraded ecological condition. As currently expressed in the CWA,
physical, chemical, and biological integrity embody a societal concern and desire that
the Great Lakes be unimpaired and healthy. In particular, increasing attention is
drawn to the biological condition of ecosystems, or biological integrity, defined by Karr
and Dudley (1981) as "a balanced, integrated, adaptive community of organisms
having a species composition, diversity, and functional organization comparable to that
of natural habitats in the region." This is similar to one current concept of biological
diversity defined as the diversity of life and its processes which have functional,
structural, and compositional attributes at the genetic, species, community, and
ecoregion levels of organization.
In addressing the ecological condition of the Great Lakes, the focus for EMAP - GL is
on biotic integrity. However, not all the societal concerns about the condition of the
Great Lakes fall neatly under the biotic integrity umbrella. For example, many Great
Lakes sports fisheries are managed through the introduction of non-native game
species. These introductions are intended to support sport and^commercial fishing
rather than to maintain or restore the biotic integrity of the lake ecosystems. Thus,
attaining the societal value of fishability may conflict with attaining the societal values
of biotic integrity.
There is also considerable public interest in the trophic condition of the Great Lakes.
The desire for lakes with clear water may conflict with the desire for lakes which
support a productive fishery or a diverse fish and wildlife fauna. Given multiple
societal concerns or values, EMAP - GL proposes to evaluate the condition of the
lakes with respect to trophic condition and biotic integrity.
1-6
-------
1.6. Specific Objectives of EMAP - GL
The objectives for EMAP - GL parallel the EMAP program objectives. The data
collected for EMAP is intended to describe current conditions (status) and to detect
trends using a set of ecological indicators at the lakewide scale of resolution.
Objectives of EMAP - GL are to:
estimate the current status and trends in indicators of the ecological condition
of each of the Great Lakes with known confidence;
monitor indicators of pollutant exposure and habitat conditions within the Great
Lakes and seek associations between stressors and condition that identify
probable causes of adverse effects;
evaluate the long-term change in condition of the Great Lakes as a result of
management and regulatory programs; and
publish annual statistical summaries on the extent and the status of indicators
of ecological condition of the Great Lakes and periodic interpretive reports on
the status and trends of indicators of ecological condition in the Great Lakes to
the EPA Administrator, decision makers, and the public.
Examples of the types of questions that are intended to be addressed by EMAP
include:
What proportion of the Great Lakes have degraded benthic communities?
What proportion of the sediments of the Great Lakes are contaminated by
toxics?
Is the trophic status of the Great Lakes changing over time?
What is the current extent and ecological condition of Great Lakes coastal
wetlands?
EMAP is not designed to be a site-specific, compliance-oriented, monitoring program
nor is it designed to provide information on specific, local scale issues. Questions at
the local scale can be addressed more effectively by existing or locally designed
monitoring networks. However, EMAP - GL seeks to determine the additive effects of
all management activities and environmental stresses at lakewide scales of resolution.
1.7. Focus and Purpose of the Research Strategy
This research strategy presents the rationale, objectives, approach, and plan for
establishing a monitoring and assessment program to document the status and trends
in ecological condition of the Great Lakes. The current document is not intended to
1-7
-------
EMAP - Great Lakes
Integration Concept
Figure 1.1 Conceptual representation of the EMAP concept for a regional characterization of
ecological resources.
be a final plan for establishing these monitoring programs but rather part of a
continually evolving process. The immediate objective is to inform potential EMAP
clients of the approach proposed for describing and monitoring the condition of the
Great Lakes, and to elicit input from the scientific, management, and regulatory
communities. It is recognized from the outset that no one agency or group has the
expertise to develop a program of this magnitude independently. The participants in
preparing this strategy include representatives of EPA's Environmental Research
Laboratory - Duluth (ERL-Duluth) and Great Lakes National Program Office (GLNPO),
the US Fish and Wildlife Service (USFWS), the International Joint Commission, the
University of Minnesota, Computer Sciences Corporation, and AScI Corporation.
Coordination with additional organizations including the National Oceanic and
Atmospheric Administration (NOAA) Great Lakes Environmental Research Laboratory
(GLERL), Environment Canada National Water Research Institute (NWRI), and
academic institutions around the Great Lakes is underway.
1-8
-------
2. Overview of Approach
2.1. Introduction
This chapter provides an overview of the EMAP approach in the Great Lakes and the
major questions that must be resolved as the program is implemented. Greater detail
is presented in succeeding chapters ending with a discussion of planned activities for
1992.
The strategy for implementing EMAP - GL requires the resolution of major questions
of statistical design and selection of indicators. Because long-term monitoring has
been conducted in the Great Lakes for a limited number of indicators, the following
approach is being taken:
identify the research and monitoring needed to meet EMAP objectives;
coordinate with ongoing research and monitoring programs to identify
monitoring gaps and issues;
conduct pilot studies to address the major outstanding questions;
implement demonstration monitoring networks to evaluate lakewide indicators,
design, and logistics; and
implement monitoring and assessment networks on a lakewide basis.
2.2. EMAP - GL Design Approach
Meeting the EMAP - GL objectives outlined in Chapter 1 requires a design that is
capable of:
estimating, with known certainty, the status of the ecological condition of the
Great Lakes at regional scales;
establishing baseline data leading to rigorous detection and description of
trends in the condition of the Great Lakes;
identifying associations among measured attributes to hypothesize possible
causes of impaired condition; and
responding quickly to new issues and questions.
For the Great Lakes, the individual lakes have been established as the regional scale
of resolution. These lakes are all individually large enough to be influenced by
regional events and each has unique physical, chemical, and biological characteristics
2-1
-------
that affect their response to stress. Additional important requirements and features of
the design include:
explicit definitions of the classes within the Lakes to be sampled (i.e., offshore,
nearshore, harbors and embayments, coastal wetlands);
identification or listing of all potential sampling units within each class (i.e.,
geographic descriptions of areas constituting offshore and nearshore classes,
lists of harbors and embayments and coastal wetlands);
probability sample site selection for each class;
flexibility to accommodate a variety of classes and problems, some of which
may not yet have been specified; and
a hierarchical structure that permits sampling at various levels of resolution.
The proposed EMAP design strategy is based on a tiered concept (Figure 2.1) that
begins with a permanent national sampling framework. This framework consists of a
hexagonal plate containing a triangular grid of approximately 12,600 points placed
randomly over the conterminous United States. Those points that fall on the Great
Lakes constitute the Tier 1 level of intensity for EMAP - GL. For much of EMAP, Tier
1 activities are intended to estimate the extent and geographic distribution of the
ecological resources being studied. However, within the Great Lakes resource, the
boundaries of the lakes and the locations of the harbors and embayments are known.
Great Lakes coastal wetlands, in contrast, have been little studied and poorly mapped.
Thus, Tier 1 activities for EMAP - GL will be focused on attempting to define the
extent and distribution of coastal wetlands associated with the Great Lakes.
Tier 1 grid locations will also be used by EMAP - GL to define the intensity of the base
grid for sampling and estimates of ecological condition. The activities associated with
estimates of condition are Tier 2 in the hierarchical scheme. A suite of biological,
chemical, and physical measurements will be obtained from each of these sites and
the information aggregated to make statements about the conditions of the lake or a
class within a lake. It is at this tier where most of the EMAP - GL monitoring and
assessment activities will be focused (Figure 2.2 and Chapter 10). An important
aspect of the EMAP design is the temporal and spatial interpenetrating nature of the
site characterization and field visits. Whereas the sampling grid consists of points
distributed across the Great Lakes, only one-fourth of these will be visited each year.
Thus, over a four-year cycle, each grid point will be visited.
Because each class (i.e., offshore, nearshore, harbors and embayments, coastal
wetlands) within a lake is likely to have a different level of homogeneity, sampling to
characterize the condition of classes may require different spatial densities. The
triangular nature of the EMAP grid design provides for the ability to increase or
decrease the density of the grid by three-, four-, or seven-fold as well as by multiples
2-2
-------
(0
O)
m
5
s
o
15
1
CO
IS
<0
Tier 4
Detailed Diagnostics \ Tier 3
Subpopulation Interests
Estimates of Condition
and Exposure
Status and Trends
Tier 2
Landscape Characterization \ jjer
Estimates, Extent, and Landuse
Figure 2.1 Concept of a four-tiered approach in EMAP. Spatial coverage is maximized in lower
tiers while temporal coverage increases at the higher tiers.
of these values. Further discussion of this concept applied to the resource classes of
Tier 3 of EMAP is intended to include activities related to the further diagnosis of
problems within a lake or to evaluate a subclass within a lake that is not adequately
covered using the existing sampling framework. In the Great Lakes, the designated
Areas of Concern and the associated Remedial Action Plans might be considered for
inclusion as Tier 3 EMAP - GL activities.
The final level (Tier 4) of activity potentially associated with EMAP - GL involves
research within the Great Lakes that compliments the activities at the other tiers. This
level is the link between EMAP and ecological research. At this time, EMAP - GL will
not be establishing new research sites but will rely on existing programs. This does
not mean that research will not be a component of EMAP - GL but rather that we will
continue to keep abreast of and utilize the ongoing research that various agencies and
institutions are conducting. The research that EMAP initially will be participating in
relates most directly to Tier 2 needs (see Chapter 10.)
2-3
-------
0)
O)
CO
CD
O
O
"co
NOAA
Sediment
Trap Program
Tier 4
Representative
Samples of the 43
Areas of Concern in
the Great Lakes
Tier 3
Each of the four resource classes
within each Great Lake, i.e.,
offshore, nearshore, harbors and
embayments, and coastal wetlands
Tier 2
Each of the five Great Lakes
i.e., Lakes Michigan, Suenor, Ontario,
Huron and Erie
Tierl
Figure 2.2 Concept of the EMAP four-tiered approach as applied to EMAP - Great Lakes.
2.3. EMAP - GL Indicator Approach
Traditionally, environmental monitoring programs have focused on individual chemical
or biological species on a local scale. The concentration or response of these
individual species was assumed to be related to a gradient of poor to good
environmental conditions. Individual chemicals and organisms, however, do not exist
in isolation but rather interact with other physical, chemical, and biological factors to
produce ecosystem response. In addition, we have learned over the last two decades
that continuing, persistent, and cumulative pollution is occurring not only at the local
scale but also on a regional, continental, and global scale. In some instances,
regulatory programs at the local scale have aggravated or contributed to problems on
a regional scale. The concept of regional and national scale impacts requires a new
approach to environmental monitoring, both in terms of what we measure and where
we measure it. Instead of focusing on problems that receive the greatest media or
public attention, we need to focus on problems that pose the greatest risk to the
environment. In "Unfinished Business: A Comparative Assessment of Environmental
Problems" (US EPA 1987), the EPA considered the highest potential risks to
ecosystems to be global warming and stratospheric ozone depletion followed by
regional problems of habitat alteration, nonpoint source pollution, and risks from
criteria air pollutants.
2-4
-------
Because environmental problems are becoming increasingly complex (i.e., cumulative
effects from multiple pollutants at multiple scales), traditional monitoring approaches
and indicators alone are insufficient to assess ecological condition. EMAP is being
designed with a "top down" approach in a risk assessment framework. This top-down
approach focuses on the endpoints of concern rather than the environmental
perturbations or stressors.
Indicators selected by EMAP - GL must relate to the assessment endpoints of biotic
integrity and trophic status and are characterized into four conceptual categories:
response, exposure, habitat, and stressors (Chapter 4). Definitions of these four
indicator types are:
Response Indicator: A characteristic of the environment measured to provide
evidence of the biological condition of a resource at the organism, population,
community, or ecosystem level of organization.
Exposure Indicator: A characteristic of the environment measured to provide
evidence of the occurrence or magnitude of a response indicators contact with
a chemical or biological stressor.
Habitat Indicator: A physical, chemical, or biological attribute measured to
characterize conditions necessary to support an organism, population, or
community in the absence of pollutants. Examples include salinity or substrate
type.
Stressor Indicator: A characteristic measured to quantify a natural process, an
environmental hazard, or a management action that affects changes in
exposure and habitat.
EMAP will collect data on response, exposure, and habitat indicators at its field
sampling sites. Stressor indicators will be primarily assembled from other sources.
Within EMAP, indicators are identified through the development of conceptual models
of ecosystems. These models may be based primarily on how current and anticipated
stresses affect ecosystems, or from a perspective of the structural, functional, and
recuperative features of "healthy" ecosystems.
Selection and development of indicators for EMAP - GL must be considered a long-
term process (Figure 2.3). Some indicators may be ready for implementation on a
regional basis, others require modification, and still others have been used only in
isolated instances and require demonstration of their regional applicability. A multi-
phase process has been identified to guide the selection of indicators and future
development:
Phase 1 - identification of issues (environmental values and apparent stressors)
and valued ecosystem attributes (assessment endpoints);
2-5
-------
Phase 2 - development of a set of candidate indicators linked to the identified
endpoints and responsive to expected stressors;
Phase 3 - screening of the candidate indicators based on a set of indicator
evaluation criteria, selecting as research indicators those that appear to fulfill key
requirements, rejecting those that clearly do not, and holding in a state of
evaluation those candidate indicators that may, in the near future, advance to
research status;
Phase 4 - quantitative testing and evaluation of the expected performance of
research indicators on regional scales, to identify the subset of developmental
indicators suitable for regional demonstration projects;
Phase 5 - regional scale demonstration of the sensitivity, reliability, and specificity
of response for development indicators; and
Phase 6 - implementation of a core set of indicators in a full EMAP program with
statistical summaries, and periodic reevaluation of indicators.
CANDIDATE INDICATORS
IDENTIFY AND PRIORITIZE
Expert Knowledge
Literature Review
Peer Review
RESEARCH INDICATORS
EVALUATE EXPECTED PERFORMANCE
Analysts of Existing Data
Simulations
Limited-Scale Field Tests
Peer Review
DEVELOPMENTAL INDICATORS
EVALUATE ACTUAL PERFORMANCE
Regional Demonstration Projects'
Peer Review
CORE INDICATORS
IMPLEMENT REGIONAL AND
NATIONAL MONITORING
PERIODIC REEEVALUATION
Figure 2.3 General approach for selection and development of indicators for EMAP.
2-6
-------
Selection of assessment endpoints and indicators requires consideration of public
values, policies, current threats to the Great Lakes, and an understanding of biological
and chemical processes. The selection of biotic indicators of condition is driven by our
understanding of biological communities, their interactions and the impact of various
stresses on the chemical and physical habitats in which the organisms exist. It is
equally influenced by the endpoints or values of concern. These endpoints, and thus
the indicators representing them, must be sufficient to motivate change in policy when
poor conditions are found.
In addition to determining the ecological condition of the Great Lakes, it is important to
determine the likely causes of poor condition. In EMAP, diagnosis will be achieved
through correlations between response indicators and the exposure, habitat, and
stressor indicators as defined above. In limited situations, the response indicators
themselves may shed some light on the likely causes of current conditions or trends.
Statistical associations between the occurrence of poor conditions as defined by
response indicators and values for the exposure, habitat, and stressor indicators will
be used to infer the most likely categories of probable cause. Correlative analyses of
these types cannot prove causality but will narrow the range of likely explanations for
observed regional patterns and lead to the generation of hypotheses which can be
investigated.
To meet the objectives of the EMAP - GL monitoring and assessment program, a
series of indicators are proposed (Table 2.1 and Chapter 4) which are representative
of the biological condition of the Great Lakes ecosystems, responsive and sensitive to
a broad array of potential stressors, produce results with an acceptable sample
variability, and are cost effective. The selection of indicators, however, is an evolving
process as data become available, as new indicators are developed, and as the joint
US-Canada workgroups complete their work in proposing indicators for the Great
Lakes ecological objectives.
2.4. EMAP - GL Data Quality
An essential complement to the careful design and indicator strategies outlined in this
section is consideration of data quality. Production and assurance of quality data
should be an integral part of any program that intends to produce useful information.
EMAP is committed to producing data for the intended purposes. To meet this
objective, it is necessary to have a process for identifying the quality of data required.
The following discussion represents a process that has just begun for EMAP - GL and
will be ongoing during the early stages of implementation.
Data quality objectives (DQOs) are statements of the level of uncertainty a decision
maker is willing to accept in results derived from environmental data. DQOs are
generated in a multistage process. Stage 1 involves defining the major questions or
problems of concern. At this stage, the focus is on the decision maker or data user.
Information needs are identified based on an understanding of how the environmental
data to be collected will be used. Also, resource and time constraints are identified
2-7
-------
and the consequences of Type 1 and Type 2 errors are examined. A Type 1 error
results in a false positive (e.g., data indicate adverse ecological effects when no such
effects exist), while a Type 2 error results in a false negative (e.g., data indicate no
adverse ecological effects when they actually exist).
Table 2.1 Indicators being considered for EMAP - GL.
Response Indicators
Fish pathology
Aquatic vegetation
Lake trout recruitment
Forage fish populations
Lake trophic status index
Lake trout/walleye populations
Diporeia/Hexagenia abundance
Chlorophyll composition in water
Diatom abundance in sediment cores and traps
Benthic macroinvertebrate community structure
Exposure Indicators
NP and Si:P Ratios
Contaminant residues in fish
Zebra mussel/exotics abundance
Sediment toxicity to Hyailela azteca
Water column toxicity to Ceriodaphnia
Contaminants in sediments from cores and traps
Habitat Indicators
Sediment physical characteristics
Water column optical characteristics
Temperature, pH, routine water chemistry
Stressor Indicators
Resource management
Human population densities
Atmospheric deposition rates
Landuse and landcover surveys
Agricultural chemical application rates
Point and nonpoint source pollutant loading
Stage 2 involves defining the information needed to answer the question or make the
decisions identified in Stage 1. This process includes developing pertinent
subordinate questions that may need to be answered in order to fully address the
problem. Also, at this stage, the population of interest should be clearly defined.
There should also be an identification of specific design constraints, i.e., the desired
2-8
-------
confidence in the results. Finally, Stage 2 should examine existing data and confirm
the need for new data, in cases where data do not exist, to provide the necessary
information.
Stage 3 involves determining a scientific approach to data collection and the data
quality requirements for that approach. This process includes considering as many
approaches to collecting the necessary data as possible, along with considering the
levels of data quality required to meet the constraints specified in Stage 1 and
Stage 2. Stage 3 should also serve to identify research activities needed to meet the
requirements of Stages 1 and 2.
Throughout the process, there is continuous feedback and communication at all levels
involving the decision makers, the scientists identifying specific information needs, and
those individuals collecting the data necessary to provide this information. EMAP is
committed to the DQO process as a means of optimizing the allocation of limited
resources and assuring that data collected in the program provides the information
needed to meet program goals.
2.5. EMAP - GL Expected Outputs
EMAP must effectively convey its scientific results to decision makers and the public.
If the original scientific objectives have been established in close consultation with
decision makers, then communicating useful information on a wide range of
environmental issues will be far easier and more meaningful. Although EMAP cannot
fully anticipate this interpretation and communication process until actual results are
available, Chapter 8 illustrates some of the potential users, questions, information
flows, and outputs.
In general, EMAP will produce four types of products:
Verified and/or aggregated data
Annual statistical summaries
Single-resource assessments
Multi-resource integrated assessments
Many clients desire access to the data being collected by EMAP either as the
individual, verified sample data or aggregated by assessment unit. There currently is
no information database that includes regional and national ecological data on multiple
ecological resources. Consequently, data are likely to be one of the early products
from EMAP.
2-9
-------
The current goal of each EMAP resource group is to have a statistical summary of the
response, exposure, and habitat indicators monitored by that group within nine months
of the last date of field-data collection. These summaries will contain descriptive
statistics such as means, medians, distributions, ranges, and standards deviations for
the various indicators monitored within the sampling frame or for selected indices
computed from these data.
EMAP will produce regional and multiple-region/national assessments that will address
either 1) the condition of particular resource or 2) the condition of all resources that
occur in a region (or megaregion). EMAP will produce two general types of
assessment reports: single-resource interpretive assessments and multiple-resource
integrated assessments.
Single-resource assessments will be produced by resource groups and task groups
composed of regional assessment specialists. Integrated assessments will be
produced by the Integration and Assessment group with assistance from resource and
task groups. Integrated monitoring data and interpretive reports on these data will be
provided regularly to environmental decision-makers, Congress, and the public. These
reports, or assessments, will be designed to contribute to informed decisions about
which risks should receive the greatest scrutiny. In producing these assessments,
EMAP scientists will incorporate relevant data from existing monitoring programs (of
EPA or other agencies).
EMAP must reach varied audiences: Congress, environmental groups, news media,
as well as the scientific community and other groups. These audiences will include
many who do not have the background in ecology, sampling statistics, and other areas
needed to fully understand EMAP's results. To improve communication with decision
makers and the public, EMAP will use focus.groups to critique proposed presentation
material for clarity, simplicity, and conciseness. The focus groups will be composed of
members representing the scientific community, environmental decision makers, policy
makers, and the public.
2.6. EMAP - GL Program Limitations
It is equally important in understanding EMAP - GL to describe not only what the
program will attempt to do but also its limitations. The program is not intended to
describe all components of an ecosystem or resource type. It will not describe how
systems function. It will, however, provide information about condition as measured
by specific indicators during an index period as a "snapshot" of the overall condition of
a system.
EMAP - GL has not been designed to directly address specific stressors such as
chemical contamination. EMAP is not intended to be compliance monitoring and will
not replace the need for such activities. In general, EMAP - GL is intended to provide
2-10
-------
a common sampling frame within which to assess the condition of the Great Lakes at
a broad scale so that the relative magnitude and geographical location of various
problems can be assessed and mitigation and research priorities can be made more
objectively. The monitoring program is not intended to be truly anticipatory, but rather
to provide an ongoing monitoring framework within which new variables can be added
or modifications be made so that the magnitude and extent of emerging issues can be
more quickly evaluated.
2.7. Questions to be Addressed
Throughout this strategy, there are questions which arise that should be addressed
before EMAP - GL can be fully implemented. It will not be possible to obtain answers
to all of these questions in one or even a few years. Because EMAP - GL is an
evolving and flexible program, the monitoring and assessment activities are intended
to change as new answers, indicators, and problems become apparent. At this time,
the major questions include the following:
Design Questions
Are the base and enhanced grid proposals adequate to describe condition of the offshore and
nearshore classes of the Lakes?
What boundary should be used to distinguish offshore from nearshore classes?
How should harbors and embayments be selected for sampling at Tier 2?
What is the number, location, and extent of Great Lakes coastal wetlands?
How should the connecting channels be incorporated into the EMAP - GL design?
How should wildlife populations be sampled?
Indicator Questions
What are the appropriate indicators for EMAP - GL wetlands?
Can nominal conditions for EMAP - GL indicators be determined?
Can the individual indicator measurements be combined to describe overall condition of the lakes
and their classes?
What indicators of fish populations are most efficient and effective in describing their ecological
condition?
Are diatoms representative of phytoplankton populations within the Great Lakes?
2-11
-------
How many sediment cores are needed to describe the historical status and trends of the
phytoplankton populations of the offshore class of the Lakes?
What is the variability of diatom population abundance collected with sediment traps as a function
of depth and spatial distribution?
Will the spatial distribution of samples at one time of year be sufficient to make statements about
indicators of water chemistry?
What contaminants should be measured in sediments and biota?
What protocol should be followed in sampling and measuring contaminants in zebra mussels?
Is there a method for measuring zooplankton populations that provides temporally integrated data
to account for the expected large temporal variations?
What indicators are available to measure wildlife?
Do process measures provide information that is necessary to make statements of ecological
condition that cannot be inferred from other indicators?
Logistical Questions
Who will actually conduct sampling once implementation is underway?
Is a lake-specific logistical plan the most efficient way to sample the lakes?
Are there sufficient numbers of trained Great Lakes scientists to undertake and maintain the
monitoring and assessment program over the long term?
Is there sufficient qualified analytical capability in the Great Lakes basin to conduct the needed
chemical analyses?
2.8 Questions to be Addressed in 1992
Some of the questions listed above will begin to be addressed during 1992. For a
variety of reasons, the initial sampling effort will focus on Lake Michigan. The
answers to some questions will be applicable to all the lakes, while others will need to
be addressed in each lake. The following are questions proposed as primary 1992
activities. These will be discussed in greater detail in Chapter 10.
Design Questions
Questions over the density of the base grid for offshore areas in Lake Michigan will be addressed
by evaluating existing data and collecting additional data for the trophic status and sediment related
indicators.
2-12
-------
Lists and areal extent of the harbors and embayments of Lakes Michigan and Superior will be
determined from USGS maps using the definitions described in Chapter 3.
Available information regarding the extent of coastal wetlands of Lake Michigan will be identified.
Indicator Questions
Recommendations for wetland indicators will be developed through a workshop of Great Lakes
wetland scientists.
Investigations into the definition of nominal conditions for sediment indicators in the nearshore of
Lake Michigan will be conducted in conjunction with Canadian studies on the remaining four Lakes.
The selection of appropriate indicators for fish will be explored through analysis of existing data and
consultations with Great Lakes experts.
Evaluation of existing data for Lakes Michigan and Superior, along with some sampling activities,
will investigate the use of diatoms as representatives of Great Lakes phytoplankton populations, the
use of sediment cores for historical trends analysis of diatom populations, and the exploration of
sediment traps as an integrative measure of annual diatom population abundance and distribution.
An evaluation of index periods for trophic status in the offshore resource class will be conducted by
comparing spring and summer data in Lake Michigan.
2.9. Implementation Plan
The proposed timelines for implementation of EMAP - GL are presented in Table 2.2.
Pilots are activities designed to test various aspects of the system, such as indicator
evaluation or sampling and logistic constraints. Site selection for pilot activities need
not be constrained to the EMAP - GL sampling grid. Demonstrations are activities in
which parts of the system continue to be tested. A key aspect of the demonstration is
the use of the sampling grid to provide policy relevant information for some subset of
the indicators. The demonstration may incorporate additional sampling to evaluate
indicator performance or examine components of variability for indicators of interest.
The timeline proposed in Table 2.2 represents an extremely optimistic time frame. It
assumes adequate and timely resources to acquire and train the needed staff for
program planning, implementation, and data analysis.
2-13
-------
0)
*z
111
o
3
X
CO
*-
c
O
_
1
o
,0)
Q.
E
._!
G
i
CL
<
LU
o
(0
(0
^
as
CO
-------
3. Monitoring Network and Field Sampling Design
3.1. Introduction
The broad EMAP objectives and the more specific Great Lakes components of those
broad objectives have been discussed in previous sections. Meeting these objectives
requires a design capable of:
estimating, with known certainty, the status and health of each Great Lake;
describing baseline data leading to rigorous detection and description of
trends in status and health of each Great Lake;
identifying associations among attributes, both within and among resources,
to establish possible causes of impaired condition; and
responding quickly to emerging issues and questions.
Important requirements and features of the design that are described in this section
include:
explicit definition of target populations and their sampling units;
explicit definition of a frame for listing or otherwise representing all the
potential sampling units within each target population;
use of a probabilistic design on well-defined sampling frames to rigorously
estimate population attributes through randomization and use of probabilistic
methods for sample unit selection;
flexibility to accommodate a variety of resource types and a variety of
problems, some of which have not yet been specified;
hierarchical structure that permits sampling at a coarser or finer level of
resolution than the general grid density, giving flexibility at global, national,
regional, or local scales; and
ability to focus on subpopulations of potentially greater interest (e.g., specific
resource classes within a lake).
A general overview of the EMAP design was introduced in Chapter 2; greater detail
about the design can be found in Overton et al. (1991). The remainder of this chapter
describes how the general EMAP design will be used in the Great Lakes component
of EMAP. A description of how sample units will be selected to meet the design
objectives of EMAP is provided. Some decisions have been made regarding pieces of
3-1
-------
the design puzzle; for other pieces, various options under consideration will be briefly
presented.
3.2. Physical boundaries
It is important to define the physical boundaries of EMAP - GL because the design is
based on the requirement to sample the lakes (i.e., the response, exposure, and
habitat indicators), not land associated with the lake or tributaries to the lake (i.e., the
stressor indicators). The program will be relying on additional data sources within the
Great Lakes to provide information on stressors. Therefore, the following is offered as
a working definition of the boundaries for EMAP - GL:
The waters of the Great Lakes and the sediments below them at high water including:
river mouths up to the maximum extent of lake influence;
wetlands contiguous to the lakes; and
the Connecting Channels, Lake St. Clair, and the upper portion of the St.
Lawrence Seaway.
The above is meant to be a natural, physical definition of the entire resource. It is
recognized that human-induced changes have altered and will continue to alter the
boundaries of the system. Thus, the presence of a dam makes it obvious where the
human-induced boundary for the river is located, while that same boundary defines the
maximum extent of lake influence. The same is true for filled areas and artificially
maintained shoreline.
Although EMAP has begun as a US program, the status of the Great Lakes cannot be
determined without sampling in Canadian waters with the cooperation of Canadian
agencies. The boundary between the two countries must not be a restriction to
determining the quality of the shared resource. Therefore, EMAP - GL intends to
include the nearshore zone, harbors and embayments, and coastal wetlands of the
entire Great Lakes and will engage the Canadians in discussions regarding access
and participation.
3.3. Regionalization
The EMAP - GL regionalization scheme consists of five regions within the Great Lakes
basin corresponding to the five Great Lakes: Superior, Michigan, Huron, Erie, and
Ontario.
Each of the lakes is physically distinct as is shown from the dimensions in Table 3.1.
The connecting channels tend to have some basic characteristics of their upstream
lakes but are physically unique (Table 3.2). In its initial phases, EMAP - GL will
monitor status and trends in the five lakes. Decisions on how to incorporate the
3-2
-------
connecting channels have not been made, but they may be treated as a separate
class for the lakes.
Table 3.1 Physical and geographical characteristics of the five Great Lakes
(adapted from MSU bulletins).
Michigan Ontario Superior Huron Erie
Length (km)
Breadth (km)
Depth (m): Average
Maximum
Volume (km3)
Surface area (km2)
Drainage basin area (km2)
Shoreline length1 (km)
Retention time (yrs)
Population: US
Canada
Landuse (% of total)
Agricultural
Residential/Industrial
Forest
Other
494
190
85
282
4920
57750
118100
2670
99
87099072
44
9
41
6
311
85
86
245
1640
18960
60600
1168
6
2657432
4616070
39
7
49
5
563
259
149
407
12230
82100
127700
4385
191
474150
155675
3
1
91
5
331
294
59
229
3540
59500
131300
6157
22
1606518
941300
27
2
68
3
338
92
19
64
483
25657
58800
1400
2.6
9183347
1742805
67
10
21
2
1 Includes islands;2 Does not include ~5 million residents of the Chicago metropolitan area who
depend on Lake Michigan for drinking water and domestic supplies but do not live in the Lake
Michigan drainage basin.
3.4. Primary Resource Classes
3.4.a. Offshore and Nearshore Areas
The primary resource classes for the Great Lakes include the offshore, nearshore,
harbor and embayment areas, and coastal wetlands. A clear definition for delineating
each of the resource classes is needed for sample allocation and for defining the
appropriate indicators to be used. For example, the variability of any indicator is
expected to be substantially greater in the nearshore waters than in the offshore
waters which tend to be homogenous. Some indicators, either by their nature or
because of the current status of the Great Lakes, are only appropriate for specific
resource classes (refer to Chapter 4).
3-3
-------
There are many potential definitions of the offshore and nearshore boundary in the
Great Lakes (Rathke 1984). In general, a simple physical definition is preferred,
especially one that can be applied to all the Great Lakes. Two definitions were
Table 3.2 Watershed characteristics of the Great Lakes connecting channels.
St. Mary's R. St. Clair R. L. St. Clair Detroit R. Niagara R."" St. Lawrence R.*"*
Inlet
Outlet
Length (Area)*
Elevation Fall (m)*
Flow m'/sec x 1000***
Minimum
Average
Maximum
Average flow vel. m/s*
Depth (m>*
Width (km)*
Retention Times
Controlled Flow
Land Drainage Area**
km2x 1000 (cum. total)
L. Superior
L. Huron
101-121 km
6.75
1.2
2.2
3.7
0.6-1.5
Shallow-30
0.3-6.4
~2 days
Y
49.3
L Huron
L. St. Clair
64km
1.5
3.0
5.2
6.7
0.6-1.8
9-21
0.25-1.2
21 hrs
N
146.6
St. Clair R.
Detroit R.
1115km2
-
-
0.02-0.08
3.4 avg.
8.2 max.
39
2-9 days
N
159.0
L. St. Clair
L Erie
51 km
1.0
3.2
5.3
7.1
0.3-0.6
6-15
0.66-3.0
21 hrs
N
160.9
L Erie
L. Ontario
59
-
5.8
-
-
--
-
-
-
L. Ontario
Atlantic Ocean
240
-
7.2
-
-
--
--
-
-
* Limno-Tech 1985, unpublished manuscript; ** David Cowgill, US Army Corps of Engineers; *** US EPA and
Environment Canada 1988; Limno-Tech. 1985; and **** Environment Canada. 1991.
considered in detail for the Great Lakes component, distance from shore and depth
contour. In describing the distance and depth relationship, Schelske (1980) states that
"physical processes in the nearshore are distinct from the offshore, currents are
stronger and effects of waves and currents, particularly relative to interactions with the
sediments, are greater." Schelske defines the nearshore zone "as that area lying
within the 30-meter contour line." He further notes that, except for Lake Erie, this
definition corresponds approximately to a nearshore strip 20 krn wide. Distance from
shore would be convenient if it were a good approximation of depth contours.
However, bathymetric data for Lake Michigan show that the distance from shore
corresponding to the 30-meter depth contour can vary from 3 - 22 km depending on
location in the lake. In terms of physical processes and sediment interactions and
because of variability of the distance parameter, depth contour is proposed as the
appropriate physical definition for delineating between the offshore and nearshore
areas. A number of depth contour proposals have been used for research and
monitoring efforts in the Great Lakes. The percent of lake surface area for the
nearshore resource class varies given different depth contour definitions for the
offshore and nearshore zones (see Table 3.3).
3-4
-------
A proposed strategy for an a priori definition of offshore and nearshore zones is to
select zones in a conservative manner to protect the integrity of zones. This strategy
provides a low probability that a given zone would influence another. Such influence
could impact data representation and statistical interpretation. The conservative
definition proposed is that the boundary between offshore and nearshore areas is the
Table 3.3 Percent of Lake Michigan surface area for the nearshore resource class as defined by depth
contour.
Depth Contour (m)
0-30
0-40
0-50
0-85
% Nearshore
Surface Area
22%
26%
28%
34%
Reference
Schelske 1980
White 1991
Surveillance Work Group 1986
Bennet 1974; Simons 1980
depth contour equal to the mean depth of the lake. As seen in Table 3.3, the
definitions of the nearshore zone for Lake Michigan can range from 30 - 85 m in depth
and from 22 - 34% of the lake area. The conservative properties of the 85 m division
may be illustrated by examining how cumulative distribution functions (cdf) change as
the division moves further inshore. In a relatively homogeneous region, such as the
offshore waters of Lake Michigan, one would expect the cdf to approximate that of a
normal distribution (i.e., a symmetrical "S" shape). Figures 3.1 and 3.2 present cdfs
for two parameters, total phosphorus and chlorophyll-a, from Lake Michigan measured
in regions delineated by 85 m, 50 m, and 30 m depth contours. For both parameters,
as more of the lake is included, the probability of picking up extreme values
associated with the nearshore increases, and, as a result, the upper tail of the cdf
becomes extended. The 85 m separation results in a cdf which closely approximates
a normal distribution and therefore, is indicative of the more homogeneous offshore
environment.
It should be noted that all of the current Lake Michigan offshore monitoring stations
from GLISP are in depths greater than 85 m (the shallowest station is approximately
100 m deep). The representativeness of the 0 - 85 m depth contour for nearshore
areas in Lake Michigan will be investigated during the 1992 pilot study (refer to
Chapter 10). This investigation will primarily involve historical data analysis but will
also include interpretation of 1992 field collected data.
Delineation of an additional, transition zone will be addressed during 1992 through
analysis of benthic community composition in Lake Michigan (White, unpublished
data). A transition zone has considerable merit because it would decrease the size of
the nearshore zone and consequently, the number of samples required for
characterization. Such a zone would reduce any potential diluting effect of the
nearshore results.
3-5
-------
85m
50m
30m
000? 0003 0005 0007 0009 0011 0013 0015 0017 0019 0021 0023 0025 0027 0029
Figure 3.1 Total phosphorus concentration (mg/L) (Lake
Michigan) measured at 85 m, 50 m, and
30m.
3.4.b. Harbors and Embayments
85m
50m
Q. 40
30m
00 05 10 15 20 25 30 35 40 45 50 55 60 65 70
Figure 3.2 Total chlorophyll-a concentration
(Lake Michigan) measured at 85 m, 50
m, and 30 m.
Harbors and embayments will be treated separately for two reasons. First, many
embayments are formed by tributaries which are a source of nutrients and
contaminants and thus, have differing water quality and higher variability. Second,
harbors are often the site of more intense human activity such as industry, shipping,
recreation and dredging, and thus are more likely to be sources of nutrients and
contaminants to the rest of the nearshore zone and to offshore waters.
It should be noted that many harbors and embayments often overlap. For example,
there are harbors at the mouths of tributaries in Lake Michigan such as Milwaukee
3-6
-------
Harbor, Benton Harbor, and Green Bay. There are also embayments that are not
harbors and harbors that do not have tributaries associated with them. Rather than try
to make distinctions between these two types of subclasses, we have elected to
consider them as one resource class. Definitions of these areas will be physical
(natural), based on structures such as breakwalls and docks or'easily recognizable
land features such as peninsulas or points. It is suggested that the "Geneva
Convention on the Territorial Sea and the Contiguous Zone" be used because it
provides a clear definition of bays in Article 7 (Hodgson and Alexander, manuscript).
The following paragraph is a summary of this definition (see also Figure 3.3).
Indentation
Semi-circle larger
' than water area.
Bay
Semi-circle
smaller than
water area.
Closure
Figure 3.3 Definition of embayments for EMAP - GL (Hodgson and Alexander, manuscript).
"A bay is a well-marked indentation whose penetration is in such proportion to the width of its mouth
as to contain landlocked waters and constitute more than a mere curvature of the coast. An
indentation shall not, however, be regarded as a bay unless its area is as large as, or larger than, that
of the semi-circle whose diameter is a line drawn across the mouth of that indentation.
For the purpose of measurement, the area of an indentation is that lying between the low-water mark
around the shore of the indentation and a line joining the low-water marks of its natural entrance
points. Where, because of the presence of islands, an indentation has more than one mouth, the
semi-circle shall be drawn on a line as long as the sum total of the lengths of the lines across the
different mouths. Islands within an indentation shall be included as if they were part of the water area
of the indentation."
3-7
-------
The lateral extent of the area will be the same as the general nearshore area (i.e.,
85 m depth contour). If the harbor or embayment includes a tributary, the area will
include the tributary mouth and upstream to the zone of lake influence as defined by
conductivity gradients. If the tributary is dammed, the area will include to the first dam
or zone of influence as applicable.
Those areas of the lakes that are often referred to as bays (e.g., Saginaw Bay, Green
Bay) and yet are themselves large enough to contain smaller bays, will be treated as
part of the general nearshore class.
Embayment minimum size will be determined from the GIS application during FY93.
Preliminary examination of maps for three areas (Green Bay, southern Lake Michigan,
and the Straits of Mackinac) was conducted to address this question. All of the map
scales were 1:24,000 or better. Preliminary work leads to the conclusion that use of a
minimum area, rather than a minimum distance across the mouth of a harbor or bay,
may be appropriate. This approach would be best defined with the aid of GIS.
Evaluation of nearshore, wetland, and embayment resource class overlap will be
further evaluated once these resources have been identified through the GIS
application. The idea that an embayment containing an embayment would be
considered in the nearshore class is just one possible outcome of our definition and
not a rigid rule. Other possibilities within our definition include:
1) An "embayment" could include all resources classes - i.e., offshore, nearshore,
harbors, and wetlands. Georgian Bay in Lake Huron may be such a case.
2) An "embayment" could include some nearshore area and several
harbor/embayment areas. Green Bay and Saginaw Bay are good examples of
this. For each of these, one of the specified harbor/embayment areas would be
the major tributary mouth that is located at the head of the embayment, e.g., Fox
River or Saginaw River.
3) An "embayment" could be defined as just a succession of two or more
harbors/embayments in the EMAP - GL sense. In other words, a large
"embayment" may be broken into several smaller harbors and/or embayments to
be included in our list frame. A good example of this would be the Bay of Quinte
in Lake Ontario, although Big Bay de Noc may qualify in Green Bay.
4) Some "embayments" would just be considered one harbor/embayment for our
purposes. These would probably include Hamilton Harbor and Sandusky Bay.
3-8
-------
3.4.C. Coastal Wetlands
The coastal wetland monitoring activities under EMAP - GL will be coordinated with
the EMAP - Wetlands group. Wetland classification will closely follow the protocols
used by EMAP - Wetlands. The general definition of a wetland provided by Cowardin
et al. (1979) will be used:
"Wetlands are lands transitional between terrestrial and aquatic systems where the water table is
usually at or near the surface or the land is covered by shallow water. Wetlands must have one or
more of the following three attributes: (1) at least periodically, the land supports predominantly
hydrophytes; (2) the substrate is predominantly undrained hydric soils; and (3) the substrate is nonsoil
and is saturated with water or covered by shallow water at some time during the growing season of the
year.'
EMAP - GL will adopt the same criteria for the target population of wetlands to be
monitored, that is wetlands with greater than 30% wetland vegetation cover that can
be identified using 1:40,000 aerial imagery. EMAP - Wetlands are a subset of the
jurisdictional wetlands as defined by the recent wetland identification criteria (Federal
Interagency Committee for Wetland Delineation 1989).
The EMAP - Wetlands program has adopted a hierarchical classification of wetlands
based on the Cowardin et al. (1979) system; the EMAP - Wetlands system
differentiates tidally influenced wetlands with significant salinity (estuarine) from non-
tidal, substantially freshwater wetlands (palustrine). Wetlands exhibiting salinity less
than 0.50 ppt during average annual flow and that are not affected by tides are placed
in the palustrine category and would include all Great Lakes wetlands. The palustrine
category includes palustrine, lacustrine, and riverine wetlands of Cowardin et al.
(1979). The EMAP - Wetlands classification system includes water source modifiers
(lacustrine, riverine, and basin) and water regime modifiers, as well as vegetation
classes (Leibowitz et al. 1991).
For purposes of defining the wetlands for monitoring under the EMAP - GL program,
wetlands contiguous to the lakeshore and wetlands that lie within 305 m of the
lakeshore will be included following Herdendorf et al. (1981). This definition
recognizes the dynamic relationship between coastal wetlands and the Great Lakes;
wetlands may be cut off from direct surface connection to the adjacent lake by bar or
dune formation or by low lake level, but subsurface connections may be maintained
and lake level may control aspects of the hydrology of nearby wetlands. Hydrologic
connections with a Great Lake may extend upstream along rivers; exchanges caused
by seiches and longer-period lake level fluctuations will influence riverine wetlands.
Wetlands under substantial hydrologic influence from Great Lakes waters will be
included in the EMAP - GL monitoring scheme.
3-9
-------
3.5. Frame Development
In order to develop the frame for EMAP - GL, it is necessary to list the sample units
for each of the primary resource categories. Depending on the spatial distribution of
the resource class, either a map or a list frame will be used.
3.5.a. Offshore and Nearshore
The offshore and nearshore areas of the lakes will be determined from bathymetric
maps from NOAA that have been digitized on a 2 km grid (except for Lake Superior
which is available only on a 4 km grid) and are available in a Geographic Information
System (GIS) format.
3.5.b. Harbors and Embayments
Harbors and embayments will be identified from United States Geographical Survey
(USGS) quad maps (and the Canadian equivalent) using the definition as described
previously.
3.5.c. Coastal Wetlands
A comprehensive inventory of Great Lakes wetlands is not currently available,
however, Herdendorf et al. (1981) provide a catalog of wetlands along the US
shoreline. Individual state and provincial efforts may have incorporated Great Lakes
coastal wetlands into regional inventories. Frame development efforts in this area will
continue during 1992-93 in cooperation with EMAP - Wetlands.
3.5.d. Issues
The previous discussion of approach points out issues that need to be resolved in
regard to frame development. A primary question involves whether harbors and
embayments should be sampled on a list frame or with an enhanced grid. Once this
question is resolved, determination of where the site should be located within the bay
or harbor needs to be considered. Considerable effort will need to be expended to
answer these questions and it is proposed that exploration of historical data, in
conjunction with GIS analysis, be conducted prior to any field work. Similarly,
sampling wetlands will require an understanding of their distribution. Investigations will
be conducted on the degree to which available inventories will be suitable for frame
development.
There are also issues for the offshore and nearshore classes that relate to sampling
for fish. Because fish populations are not fixed spatially, it may not be practical to
sample fish at the same locations as the rest of the offshore and nearshore indicators.
For example, if lake trout spawning success is found to be a necessary indicator of
ecological condition, sampling may be conducted only at known spawning reefs. A
detailed review of sampling options for fish as indicators of ecological condition will be
conducted during 1992.
3-10
-------
3.5.e. Quality of Frame
Generally, the quality of the frame is good to excellent, with the exception of the
coastal wetlands. The main variable for offshore and nearshore definitions is water
depth and, for the purposes of resource classification, it is obtained from bathymetric
maps. USGS maps can be used to identify harbors and embayments. Some small-
scale wetland inventories exist for states or provinces, but there is no current
comprehensive effort for the Great Lakes basin. Until such inventories are available,
the quality of the wetland frame will be uneven.
3.6. Monitoring Network Design
3.6.a. Tier 1 Sampling
The primary resource classes (offshore, nearshore, harbors and embayments, and
coastal wetlands) are the Tier 1 resources for EMAP - GL. Tier 1 samples are
intended to provide a frame for the Tier 2 samples as well as the scope, extent, and
spatial distribution of the primary resources.
40 u-
404 Cells
Alber Conic Equal-Area Projection
Figure 3.4 EMAP base grid cells within the Great Lakes.
3-11
-------
Because the resources of the Great Lakes are well defined, the extent and distribution
of the primary resource groups are well documented. The obvious exception is
coastal wetlands. The primary uses of the Tier 1 sampling units will be to provide the
frame for subsamples that are measured annually (Tier 2 samples).
3.6.b. Tier 2 Sampling
The Tier 2 sample is a subsample of the resource occurrences identified at Tier 1.
For each sample frame, a methodology for sample site selection must be specified.
As discussed under frame development, maps will be used to locate the appropriate
resource class. In addition, association rules to determine where actual samples will
be collected within the grid point will be needed for sediment and biota sampling.
Figure 3.4 presents the EMAP base grid for the Great Lakes. For wetland sampling,
site selection methodology will be developed in conjunction with EMAP - Wetlands.
The site selection methodology for sampling in the offshore and nearshore resource
classes is presented in the following sections. Sampling methods for harbors and
embayments and coastal wetlands have not yet been determined.
Table 3.4 Summary of sampling grid points for the nearshore zone using the base grid, 3-fold, 4-
fold, and 7-fold enhancement.
Number of nearshore grid points
Lake Michigan
Base Grid
3-Fold Enhanced Grid
4-Fold Enhanced Grid
7-Fold Enhanced Grid
45
127
145
272
3.6.c. Offshore Samples
Figure 3.5 is an overlay of the 85-m contour for Lake Michigan on the base grid.
Forty-three grid points fall outside of the 85-m contour and thus, lie in the offshore
resource class. Because of the rotating and interpenetrating nature of the EMAP
design, a Tier 1 grid point is sampled every four years. This corresponds to 10 or 11
samples per year. This degree of sampling intensity corresponds well with the
current GLISP offshore network which consists of eleven stations (Figure 3.6). As will
be discussed in Chapter 10, a comparison of data between the existing GLISP
stations and the proposed EMAP - GL offshore network will be initiated with studies
conducted in the spring and summer of 1992.
In addition, historical data on extensive limnological surveys conducted in the Great
Lakes will be investigated to determine if 10-11 samples are adequate to characterize
the offshore areas. Because it is expected that insufficient data will be available for all
3-12
-------
Base Grid
3 Fold Grid
Marauqu*
GfMn Sly
MilwiukM
A Roguiar Station
Maslปr Station
Chicago
Stnlon Harbor
Figure 3.5 Eighty-five meter depth contour
for Lake Michigan with EMAP
base grid for offshore zone, 3-
fold grid for nearshore zone.
Figure 3.6 GLISP surveillance stations for
Lake Michigan.
the lakes (and all indicators), more extensive sampling will probably be necessary to
determine whether the base grid and rotating design will be sufficient. These studies
would be initiated as part of the 1993 demonstration in Lake Michigan and the 1993
pilot in Lake Superior.
3.6.d. Nearshore Samples
The remaining base grid points (45), Table 3.4, are in water less than 85 m and would
be considered nearshore sampling locations. However, with the higher spatial
variability and the diverse nature of the nearshore zone, this sample size (11 per year)
would not be adequate for characterization of the resource. Therefore, grid
enhancement in the nearshore area is believed to be necessary. Using an enhanced
grid, either 3-fold or 4-fold, would result in 30 to 40 samples per year (Figure 3.5).
The next highest enhancement (7-fold) would probably result in too many sample sites
in the nearshore zone. To the extent possible, existing data will be used to
characterize the variability in nearshore indicators and to determine the appropriate
degree of enhancement necessary to characterize the nearshore portion of the lakes.
3-13
-------
The issue of adequate sampling intensity will continue to be investigated as EMAP -
GL extends to the rest of the Lakes.
3.6.e. Harbors and Embayments
Because harbors and embayments are unevenly and widely distributed, this resource
class is not sampled adequately by the base grid. There are two options being
considered for this class to obtain a reasonable sample size:
Option 1 - Develop an association rule for the enhanced grid that would select harbors
and embayments for Tier 2 sampling.
Option 2 - Develop a list frame of harbors and embayments and, by systematic
random sampling, select a subset as the Tier 2 sample.
3.6.f. Coastal Wetlands
Because wetlands are unevenly and widely distributed, this resource class is not
sampled adequately by the base grid. There are two options being considered for this
class to obtain a reasonable sample size:
Option 1 - Develop an association rule for the enhanced grid that would select
wetlands for Tier 2 sampling.
Option 2 - Develop a list frame of wetlands and, by systematic random sampling,
select a subset as the Tier 2 sample.
3.6.g. Tier 3 for EMAP - GL
As discussed in Chapter 2, most of the emphasis for the early stages of EMAP in the
Great Lakes will be placed on the Tier 2 level of sampling. Tier 3 is intended to
represent activities resulting in the detailed diagnostics of subpopulations of interest
(Figures 2.1 and 2.2). The Great Lakes region has, however, a subpopulation already
identified that would logically fall into the Tier 3 category. Through international
agreements, some 43 Areas of Concern (AOC) on the Great Lakes have been
identified as locations with special problems (see Figure 4.2). Federal, state,
provincial, and local groups are responsible for establishing Remedial Action Plans
(RAPs) for each of these areas. Establishing a monitoring program to determine the
improvements that result from implementing these RAPs is an essential component of
the plan. We propose that including RAP monitoring activities into EMAP Tier 3 would
be a logical umbrella to both put the RAP monitoring plans into a larger lakewide
context and to provide a component of continuity between RAP sites. Thus, EMAP
would assist other agencies in coordinating monitoring activities at AOC locations by
establishing a consistent design and some key indicators to be measured at all sites.
3-14
-------
3.7. Field Sampling Design
Because EMAP emphasizes regional coverage, the program has adopted an index
sampling regime. Index sampling targets one (or a very few) sampling times within
each lake, limiting temporal coverage to maximize spatial coverage with available
resources. The spatial coverage is important because EMAP is interested in the
regional pattern of indicators rather than conditions at individual sites. An ideal index
period occurs when the values of the indicators are relatively stable, when indicator
biota are present and measurable, and during a season of maximum stress. Because
multiple indicators will be used, it will not be possible to select an index time that is
optimal for all indicators and possible stresses. Indicators that integrate over annual
cycles (or some other critical period) would be most desirable.
Index sampling should not be confused with indicator indices which are calculations of
a metric from basic indicator measurements. Measurement of indicators and the use
of indicator indices will be discussed in Chapter 4.
3-15
-------
4. Indicator Development and Evaluation
4.1. Introduction
The process of indicator development and evaluation addresses the issues of
selecting and using a set of measurements to assess the ecological condition of each
of the Great Lakes. The process is defined by the goals of EMAP - GL which
emphasize regional assessments of the ecological condition of the Great Lakes. This
section encompasses a wide range of topics, including what organisms and chemical
or physical features to target for each resource class, when to sample, and how to
convert the measurements into meaningful assessments of ecological condition.
This overview of EMAP - GL indicators includes selection criteria, currently proposed
indicators, and a strategy for the continuing process of selecting, developing,
evaluating, and using indicators. The initial results of analyses of existing data which
begin to tie the design features to the indicators are presented. Before any field work
is performed, operations manuals (laboratory and field methods, quality assurance,
field implementation) will be prepared.
4.2. Conceptual Framework for Indicators of Condition
Assessments of the Great Lakes condition must relate to values and problems of
present or potential concern to society. These issues are referred to as endpoints of
concern, and are reflected in the ways the Lakes are managed. For EMAP - GL,
these values or endpoints are trophic state and biotic integrity.
The development and evaluation of indicators for EMAP - GL can take advantage of
several existing features of research and monitoring in the Great Lakes. First, a great
deal of ecological assessment work has been conducted on the Great Lakes in the
past, and the basic physical, chemical, and biological characteristics are fairly well
known, particularly in the relatively homogenous offshore waters of the upper Great
Lakes (Superior, Michigan, and Huron). For example, fish species diversity, which has
been used as an indicator in many inland lakes and streams, would be much less
useful in the offshore waters of the Great Lakes due to the low rates of emigration and
immigration and the resultant low diversity. Therefore, fish community indicators
which focus on individual species, particularly top predators, on certain trophic levels
(e.g., forage fish), or on discrete species assemblages, would be more directly
applicable to the Great Lakes.
Second, in the Great Lakes, much of the framework for assessing ecological condition
using indicators has been established. The concept of using indicators as "top-down"
measures of ecological condition or integrators of ecosystem health is not new to the
Great Lakes. Revisions to the Great Lakes WQA in 1978 emphasized a broad
ecosystem approach to managing the Great Lakes and mandated the development of
specific biological and chemical objectives which would aid in restoring and
4-1
-------
Table 4.1 Indicators being considered for EMAP - GL.
Response Indicators
Fish pathology
Aquatic vegetation
Lake trout recruitment
Forage fish populations
Lake trophic status index
Lake trout/walleye populations
Diporeia/Hexagenia abundance
Chlorophyll composition in water
Diatom abundance in sediment cores and traps
Benthic macroinvertebrate community structure
Exposure Indicators
N:P and Si:P Ratios
Contaminant residues in fish
Zebra mussel/exotics abundance
Sediment toxicity to Hyallela azteca
Water column toxicity to Ceriodaphnia
Contaminants in sediments from cores and traps
Habitat Indicators
Sediment physical characteristics
Water column optical characteristics
Temperature, pH, routine water chemistry
Stressor Indicators
Resource management
Human population densities
Atmospheric deposition rates
Landuse and landcover surveys
Agricultural chemical application rates
Point and nonpoint source pollutant loading
4-2
-------
maintaining balanced, stable oligotrophic and mesotrophic ecosystems (Cairns et al.
1991). Indicator development related to this objective has focused on the
development of surrogate organisms, that is, species which integrate physical,
chemical, and biological properties of the ecosystem (Ryder and Edwards 1985).
Surrogate organisms have been proposed for both oligotrophic (Edwards et al. 1990,
Ryder and Edwards 1985) and mesotrophic (Edwards and Ryder 1990) waters of the
Great Lakes. The indicator development and evaluation process for EMAP - GL has
relied heavily on this past work and embraces the tenet of the ecosystem approach
under which the work was conducted.
Nearshore waters have comparatively less information and coastal wetlands are poorly
known. It is recognized that only a few studies have examined the aquatic animal or
plant community composition, or chemical, physical, and biological functions of coastal
wetlands on the Great Lakes. Coastal wetlands are dynamic systems, changing in
area and configuration with long-term Lake levels. Aside from the areal dynamics of
these systems, very little is known about their general ecology or linkages to either
adjacent uplands or pelagic waters. The general lack of information on Great Lakes
coastal wetlands confounds the easy choice of indicators of system condition.
Indicator development for wetlands will necessarily lag behind that of other resource
classes.
EMAP has defined four types of indicators: response, exposure, habitat, and stressor.
It is important to understand that these categories provide a conceptual framework
which acts as a tool to guide the process of selecting, evaluating, and implementing
the actual measurements used to assess ecological condition. These categories are
not meant to be rigid boxes in which each measurement type may serve only one
function. The definitions are:
Response indicators are derived from measurements that describe the
biological condition of organisms, populations, communities, or other
components or processes of the aquatic ecosystem as they relate to the
endpoints of concern (trophic state, biotic integrity). Response indicators are
the focal points upon which the health of the Great Lakes will be assessed.
The measurements should be amenable to quantifying the integrated
response of the ecological resources to individual or multiple stressors.
Response indicators should clearly relate to aspects of the environment
valued by the public, including the scientific community. Response indicators
may also be chosen to identify problems, such as accelerating eutrophication
or decreasing recruitment in fish populations.
Exposure indicators are intended to serve a diagnostic function when
measured in conjunction with response indicators. They are used to identify
the likely causes of impaired conditions as detected by the response
indicators. The exposure and habitat indicators are characteristics of the
aquatic environment that give evidence of the occurrence and magnitude of a
response indicator's contact with a physical, chemical, or biological impact.
Historically, Great Lakes monitoring has primarily evaluated water quality, i.e.,.
4-3
-------
measures of physical and chemical characteristics of surface waters. These
assessments have largely relied on what we have called exposure indicators.
Habitat indicators are attributes that describe the physical conditions of the
environment that are necessary to support the integrity of the biological
community.
Stressor indicators quantify a natural process, environmental hazard, or
management activity causing change in exposure or habitat. They can be
thought of as characterizing the sources of exposure. Some examples
include landuse and landcover, pesticide application rates within the
watershed, human population densities, and export of pollutants. Sources of
information for stressor indicators proposed for EMAP - GL include: landuse
and landcover surveys, human popuiation density, atmospheric deposition
rates, agricultural chemical application rates, point source pollutant loading,
presence of introduced species, and stocking and harvest records.
The primary role of exposure, habitat, and stressor indicators within EMAP is to
identify the probable causes for impaired conditions. In addition, habitat indicators
may be used to normalize response indicators. For further discussion of the use of
ecological indicators in EMAP refer to Hunsaker and Carpenter (1990). Table 4.1
presents indicators currently being considered for use in EMAP - GL.
4.3 Strategy for Indicator Development and Implementation
Given the spatial, temporal, and ecological enormity and complexity of EMAP - GL, it
is essential to have a basic strategy that defines a process and a set of goals and
criteria by which to evaluate whether the steps have been successfully completed.
Each phase in the general strategy can be expanded to include as much detail as
needed to assure achievement of its goals. The following strategy has been modified
from the overall EMAP indicator strategy (Hunsaker and Carpenter 1990).
The indicator development strategy (Figure 4.1) does not propose a linear process.
Instead, there may be considerable branching and feedback to previous steps. Some
indicators may progress relatively quickly to full implementation, while others may
progress slowly for many years. There are six basic phases:
Identify environmental values and apparent stressors.
Develop a set of candidate indicators which are linked to endpoints of
concern and responsive to expected stressors.
4-4
-------
Phase 1
Phase 2
Phased
Phase 4
Phases
Phase 6
IDENTIFY
ISSUES/ASSESSMENT ENDPOINTS
Objectives
Develop indicators
linked to endpoints
Methods
Expert Knowledge
Literature Review
Conceptual Models
CANDIDATE INDICATORS
Prioritize based
on criteria
iปJ*ct, ปmpซnd, or
Expert Know* edge
Literature Review
RESEARCH INDICATORS
Evaluate expected
performance
Outntlktlv* fcrtlng
Analysis of Existing Data
Simulations
Pilot Teats
Exampl e Assessments
Conceptual Models
DEVELOPMENTAL INDICATORS
Evaluate actual
performance on a
regional scale
buU httMtrudur*
MMMlogktlci
Regional Demonstration
projects
Regional Statistical
Summaries
CORE INDICATORS
Implement regional
and
national monitoring
periodic itM
EMAP Data Analysis
Correlate Old Indicators with
Proposed Replacements
Assess Promising
Candidate Indicators
Evaluation
Workshops
Criteria
Criteria
Peer Review
Criteria
Peer Review
Criteria at
Regional Scale
Peer Review
Agency Review of
Summary
Feedback from
Peers and.Aflendee
Peer Review
Revisit Assessment Endpoints
Figure 4.1 Indicator selection, prioritization, and evaluation approach for EMAP (Hunsaker and
Carpenter 1990).
4-5
-------
Screen candidate indicators to select research indicators with reasonably well-
established databases, methods, and responsiveness.
Quantitatively evaluate expected regional scale performance of research
indicators to identify developmental indicators for regional demonstrations.
Demonstrate developmental indicators on a regional scale, using the sampling
frame, methods, and data analyses intended for full (core) implementation.
Implement core indicators with annual sampling and data analyses, and
periodic reevaluation of indicators.
The first two phases are meant to generate ideas for endpoints and indicators. These
two phases encourage broad-scale, lateral thinking, with a focus on breadth rather
than depth of coverage, and may be revisited at any time. Essentially, EMAP - GL
has completed the first pass through phases 1 and 2, using a series of meetings,
informal literature reviews, and exploratory data analyses.
Phases 3 to 5 are oriented toward critical evaluation and integrative filtering of the
candidate indicators down to a defensible, practical set of core indicators. Where
phases 1 and 2 are inclusive, these three phases focus on excluding indicators that
are currently not feasible within EMAP and on documenting the value of those
selected.
As noted earlier, the development of indicators for Great Lakes coastal wetlands will
lag behind other resource classes; the dynamic nature of coastal wetlands and
general lack of comprehensive information on the hydrologic, chemical, and biological
features of these systems require careful consideration. The EMAP - Wetlands
program has suggested candidate wetland indicators for measuring long-term
response and exposure. The general classes of indicators include: wetland extent,
landscape indicators, hydrologic indicators, sediment characteristics, community
composition and abundance of vegetation and animal biota, chemical contaminants,
bioaccumulation in tissues, and nutrients in sediment or plant tissues (Leibowitz et al.
1991). The list of EMAP - Wetlands candidate indicators will be used as a starting
point to identify appropriate indicators for Great Lakes coastal wetlands. A panel will
be convened to consider ways to tailor this list of candidate indicators and discuss
new indicators that are appropriate to the needs of the EMAP program for monitoring
Great Lakes coastal wetlands. The panel will be drawn from academic and
governmental scientists having experience with Great Lakes coastal wetlands,
including water quality, sediment, vegetation, wildlife, hydrodynamics, and landscape
ecology. The panel will also identify ways that Great Lakes coastal wetlands can be
partitioned by functional type, facilitating a cyclic sampling of representative wetlands.
4.3.a. Establishing Nominal Condition
The process of selecting and implementing indicators cannot be separated from the
problem of how the data will be used to make statements about the ecological
4-6
-------
condition of the lakes. The question is one of establishing the criteria that can be
used to separate nominal from subnominal. In other words, determining the value of
an index or metric score below which a region is considered to be in an unacceptable
condition, relative to a particular endpoint of concern. Some of the approaches to
establishing these criteria include selecting and assessing reference sites, using
ecological models, and basing the criteria on the empirical distribution of indicator
values.
EMAP - GL will interpret ecological condition based on a combination of these
approaches, relying on a series of regional reference sites (Hughes et al. 1986,
Hughes 1989) and historic condition as described through paleolimnological
reconstructions, benthic surveys, commercial catch records, and other applicable data.
This will allow integration of professional judgement, an understanding of historical or
pristine conditions, and knowledge of current ecosystem research to select the least
disturbed (most natural) but typical sites of a region. If reference sites can be
identified, the response indicator values, and the statistical variability found can be
used to develop the nominal conditions for the region. It will also be useful to
examine the range of stressors impinging on the Tier 2 sites and to purposely select
additional sites to ensure that the model of severely impaired systems is well defined.
Another option for determining whether a system is impaired is through one or more
ecological models. Models may be based on field data for a variable or set of
variables, complex ecosystem studies, or laboratory toxicity tests. Empirical models
that identify relationships are an effective tool for establishing criteria. Habitat
classification models should be useful for assessing exposure and habitat indicators.
As EMAP - GL progresses, models that are applicable to these systems will be
incorporated into the data interpretation process and used to refine criteria.
In many respects, the definition of acceptable conditions is a social, as well as a
scientific one. Our intent in EMAP - GL is not to defend any specific value as the one
and only value but rather to participate in the discussion on how to objectively
establish this criterion. EMAP will produce data that can provide the foundation for the
assessment and take part in the technical discussion on establishing scientifically and
socially desired boundaries.
4.4. Response Indicators
The process of selecting indicators has been complex, governed by EMAP's emphasis
on ecological response and on broad scale assessments. It has been guided by
consultations with other professionals in aquatic ecology, and examination of available
databases and the scientific and management literature. Table 4.2 lists a chronology
of EMAP indicator development for the Great Lakes. Much of the effort has been
placed on response indicators, which address the first of the EMAP objectives. The
discussions below are not intended to be a review of existing literature but rather a
summary of plans to use various taxa and indicators. Appendix B of this report
provides detailed indicator fact sheets for EMAP - GL.
4-7
-------
EMAP - GL has several general criteria for evaluating the usefulness of response
indicators. The response indicators should be biological and incorporate elements of
ecosystem structure and function; the selected indicators should correlate with
changes in other unmonitored biological components; they must have clear
connections with the endpoints of concern and must be responsive to a broad array of
potential stressors. An ideal indicator is applicable in lakes of similar productivity in
each of the Great Lakes; sensitive to varying levels of stress; cost effective, providing
considerable information in a limited amount of sampling time; easily implemented;
and provide reproducible results with low sampling variability.
Table 4.2 Chronology of EMAP indicator development for the Great Lakes.
April 1991. EMAP - GL Research Planning Committee formed from specialists in fields related
to indicator development.
April-May 1991. Weekly conference calls of the Research Planning Committee. Potential
indicators suggested and a list of potential indicators developed.
May 1991. First Indicator Development Workshop. List of potential EMAP - GL indicators
modified and discussed. Literature reviews and development of indicator fact sheets developed
out of this workshop.
May-August 1991. Weekly conference calls of the Research Planning Committee.
August 1991. Second Indicator Development Workshop. Following literature reviews and
discussion between Research Planning Committee members and experts in fields related to the
proposed indicators, selection of candidate indicators finalized.
August 1991 - Present. Weekly conference calls of the Research Planning Committee. List of
candidate indicators reviewed and updated following consultation with experts. Sampling
logistics and schedules for the pilots on Lakes Michigan and Superior in FY92 suggested and
discussed.
4.4.a. Fish Indicators
Fish indicators will be used as indicators of biotic integrity, providing information on
population abundance, recruitment, pathology, forage base, and average condition of
individuals.
This section presents a highly simplified outline of the changes in the composition of
some Great Lakes fish stocks during the last century. More detailed information for all
of the lakes can be found in Christie (1974), Hartman (1988), and Rathke and McRae
(1990); and for each lake individually in Christie (1973; Lake Ontario), Hartman (1973;
Lake Erie), Wells and McClain (1973; Lake Michigan), Berst and Spangler (1973; Lake
Huron), and Lawrie and Rahrer (1973; Lake Superior).
The Great Lakes contain the largest and most valuable assemblage of freshwater fish
resources in the world (Hartman 1988). Yet the fish communities in the Great Lakes
4-8
-------
today are drastically different than those found around 1800, before intensive
settlement began in the basin (Ryder 1972). Ryder and Edwards (1985) characterize
today's Great Lakes fish communities, relative to those of the distant past, in the
following manner:
"...as a generalization, the fish communities of the Great Lakes today are of
smaller mean size, comprised of species more dependent on the pelagic zone
and lacking large species that formerly were abundant in rivers and near-shore
zones, including large terminal predators and benthic feeders."
Historically, the upper Great Lakes supported a coolwater salmonid community,
including salmonines (salmon, trout, chars), a large benthic predator (the burbot), and
a coregonine complex (whitefish, ciscoes, chubs) that consisted of perhaps eleven
species in Lake Michigan (Koelz 1929) and slightly fewer species in Lakes Superior
and Huron. The species complex of prime importance was that of the lake trout.
Many different lake trout stocks, including several river-run stocks (Loftus 1958),
occupied much of the basins of these lakes where they preyed primarily on a complex
assortment of coregonine taxa, as well as on other fish species and invertebrates
(Eschmeyer 1957, Goodier 1981).
Lake Erie is distinct from the upper Great Lakes in that it has a shallower basin and
much higher levels of natural nutrient loading. Consequently, in its pristine state it
was probably mesotrophic rather than oligotrophic in nature (Beeton 1961, Ryder
1972). As a result, percid species including the walleye, blue pike, sauger, and yellow
perch dominated the Lake Erie fish community, particularly in the western and central
basins. Lake trout were moderately abundant at one time in Lake Erie, particularly in
the deeper eastern basin (Applegate and Van Meter 1970). However, it is likely that
only one stock of lake trout existed in Lake Erie due to a low level of environmental
heterogeneity (Ryder and Edwards 1985). Lake Erie has historically sustained the
most productive commercial fishery in the Great Lakes, with its yield generally
exceeding the combined yields of the other lakes (Baldwin et al. 1979).
Like Lake Erie, Lake Ontario also has a relatively high level of natural nutrient loading
relative to the upper Great Lakes (Beeton 1965). The indigenous fish community of
Lake Ontario somewhat resembled that of the upper Great Lakes, except for the
presence of euryhaline marine species, notably the Atlantic salmon and American eel,
which became established due to the lake's natural connection with the Atlantic Ocean
(Christie 1972).
The presettlement environments of many of the larger bays in the Great Lakes were
probably similar to that of the eastern basin of Lake Erie today (i.e., late oligotrophic or
early mesotrophic), with some being more oligotrophic (e.g., Thunder Bay and
Keweenaw Bay of Lake Superior) or more mesotrophic (e.g., Saginaw Bay in Lake
Huron, the Bay of Quinte in Lake Ontario). Fish communities in these large bays
ranged from a typical coldwater salmonid community (similar to that found in the
Upper Lakes) to a coolwater, mesotrophic percid community (Ryder and Edwards
1985). Naturally eutrophic conditions and their typical warmwater centrarchid-
4-9
-------
dominated communities were probably relatively rare in much of the Great Lakes
proper except for some tributaries to Lake Erie and their deltas, or along sheltered
shorelines in small bays.
The appearance of the sea lamprey in the upper Great Lakes in the 1940's and
1950's (Table 4.3) sequentially reduced the already over-exploited stocks of lake trout
to virtual extinction in Lakes Huron and Michigan and to near extinction in Lake
Superior. Lake trout stocks in Lakes Erie and Ontario were well on their way to
collapsing by this time, primarily due to uncontrolled fishing effort and overharvest.
Burbot stocks were also decimated by the sea lamprey. Elimination of these top
predators resulted in an explosion of various forage species, particularly those
comprising the coregonine complex, which radiated from their former predation refugia
and spread to areas of the upper Great Lakes where they did not traditionally occur in
high abundance (Ryder and Edwards 1985). In the absence of predation pressure,
many of these coregonine stocks began to interbreed, losing their identity and creating
an ever-changing assemblage of forage species. Further complications arose with the
introduction and invasion of two exotic species, the rainbow smelt and the alewife
(Table 4.3). These two species quickly became abundant in most areas of the Great
Lakes and eventually came to dominate the forage fish assemblages in all of the
lakes. Four species of Pacific salmon (coho, Chinook, pink, kokanee) were stocked
beginning in the mid-1960's, partly in response to the tremendous forage base
provided by the alewife and smelt. These stockings were in addition to numerous
introductions of two non-native trout species, the rainbow and brown trout, which had
been stocked over the years in the Great Lakes. All of these introduced salmonines,
except perhaps the kokanee, have reproduced to some extent in some of the Great
Lakes, but populations of Pacific salmon are still maintained through large stocking
programs. Other species of fishes continue to be introduced or invade the Great
Lakes (Table 4.3). The ruffe, a European percid, has been established in the St.
Louis River estuary (Lake Superior) since at least 1987 (Pratt 1988) and the white
perch, a European percicthyid, has become established in parts of Lakes Ontario,
Erie, Huron, and possibly Superior (Hartman 1988).
Table 4.3
Chronology of first appearance of certain exotic species in the Great Lakes.
Lake
Ontario
Erie
Huron
Michigan
Superior
Rainbow
Smelt
19292
19324
19252
19232
19302
Sea
Lamprey
18503
19211
19321
19361
19461
Alewife1
1860
1931
1933
1949
1954
White
Perch
1950
1953
1980
--
--
Ruffe5
--
--
--
--
1986
1 Christie (1974), 2 Van Oosten (1937), 3 Lark (1973),4 Trautman (1957), 5 Pratt (1988)
4-10
-------
It is apparent from the above discussion that the present fish communities in the Great
Lakes are highly dynamic, astatic assemblages (Ryder and Kerr 1978, Ryder et al.
1981). These characteristics create special problems in the development of fish
community indicators for EMAP - GL. A goal of restoring Great Lakes fish
communities to a pre-settlement state, for example, is not attainable because exotic
species such as the alewife, smelt, and Pacific salmon have come to dominate their
trophic levels in many areas, and extinct species, such as the blue pike in Lake Erie
and the Atlantic salmon in Lake Ontario, represent gene pools which can never be re-
established.
Table 4.4
Proposed oligotrophic fish community metrics for EMAP - GL.
* Number of fish of a given species or group per trawl tow
*
Catch per unit effort with various
sampling gear
* Target biomass for certain species or groups
*
*
*
*
*
*
*
*
Mean weight of individual fish at
Mean condition factor (W=weight
a. Standard Condition Factor (K):
b. Relative Condition Factor (Kn)
capture
; L=length; W'=weight predicted from historical regression):
K=W/L3
Kn=W/W'
Percentage of native juvenile lake trout caught in assessment gear
Percent occurrence of external tumors (gross external pathology)
Production/Biomass ratios
Species ratios
Commercial yields
Sport fishing yields
* Age/size structure of populations
The proposed oligotrophic fish community indicators for EMAP - GL (Table 4.4) focus
on the lake trout. An extensive evaluation of the lake trout as an indicator was
performed under the auspices of the IJC and the Great Lakes Fishery Commission by
the Work Group on Indicators of Ecosystem Quality (Ryder and Edwards 1985). The
rationale for using the lake trout as an indicator was cited by the work group, some of
which is summarized below.
1) During the early days of settlement in the Great Lakes Basin, lake trout were widely distributed in
Lakes Superior, Michigan, Huron, Ontario, and the eastern basin of Lake Erie; an area that
represented about 95% of the total surface water of the Great Lakes Basin.
4-11
-------
2) In general, lake trout move extensively throughout the aquatic system of the Great Lakes Basin
and consequently, are exposed to a multitude of natural stresses found in oligotrophic systems.
3) The lake trout is an excellent integrator of oligotrophic biota by virtue of its function as the major
terminal predator over most of the Great Lakes Basin.
4) The environmental requirements of lake trout (e.g., temperature, dissolved oxygen, habitat) are
widely documented in the Great Lakes.
5) Lake trout, as top predators, may be expected to bioconcentrate certain toxic substances that may
appear at levels too low to detect in lower trophic levels. Subdetectable concentrations of toxic
organic substances in the environment may eventually be back-calculated on the basis of lake trout
tissue analysis (Connolly and Thomann 1982, McNaught 1982, Rodgers and Swain 1983).
6) The reproductive and early life history stages of the lake trout are especially vulnerable to
environmental stresses. The inability of lake trout to produce or to sustain progeny to fishery
recruitment stages provides both an early-warning and a retrospective indicator of ecosystem
impairments.
7) Historic data series exist on the abundance of the lake trout in each of the Great Lakes in the
form of commercial catch statistics. Some of these data series extend well back into the 1800's.
Recent commercial data series combine catch and effort statistics as an index of abundance,
demographic characteristics of lake trout stocks and fecundity and food habits. Other studies detail
reproductive success, location and quality of historic spawning grounds, incidence of sea lamprey
attack, movements, stock diversity, genetic characteristics, and contaminant burdens.
8) There are ongoing programs of lake trout sampling by federal government agencies in both the
United States and Canada, by several states and the province of Ontario, which are coordinated by
the Great Lakes Fishery Commission and the IJC. These programs assess lake trout abundance,
recruitment, feeding habits, movements, contaminant burdens, spawning location and behavior, and
many other factors important to the maintenance of healthy stocks.
The proposed mesotrophic fish community indicators for EMAP - GL focus on the
walleye. The suitability of walleye as an indicator of mesotrophic ecosystem health
was evaluated by the Mesotrophic Indicators Work Group under the auspices of the
IJC (Edwards and Ryder 1990). It is within these geographic regions where the use of
the walleye as the principal fish community indicator is proposed.
The notion of harmonic fish communities was first developed from empirical data on
several unperturbed mesotrophic lakes in the boreal forest region (Ryder and Kerr
1978). In this region, in which the community structures were assumed to be similar
to those immediately following glaciation, the fishes were found to be highly similar in
both kind and proportion from lake to lake. The term "harmonic community" implies
high integration among species, high levels of stability, effective resilience to
exogenous stresses and appropriate complexity, as well as moderately constant
community composition and Production/Biomass ratios. The harmonic community
concept was used by the Mesotrophic Indicators Work Group to develop desirable
characteristics of harmonic components (walleye, yellow perch, northern pike, white
sucker) for use as indicators of ecological condition in mesotrophic regions of the
Great Lakes (Edwards and Ryder 1990). Table 4.5 presents those properties of
harmonic mesotrophic fish communities identified by the IJC workgroup.
4-12
-------
Tumor incidence may be an important indicator for harbors and embayments. These
areas (particularly AOCs) are considered to be among the most contaminated by
persistent chemicals, and tumors in fish are a highly visible indicator to the general
public. There are, however, a number of technical issues related to sampling and
interpretation of results. The difficulties of sampling fish have previously been
discussed. A sampling program specifically for tumors is particularly difficult because
incidence is almost always very low, even in contaminated areas. The problem
encountered when investigators are unable to collect fish from specified locations is
especially difficult to reconcile. If toxic chemicals are present at toxic concentrations,
exposed fish might die or otherwise avoid an area.
A negative response in a tumor survey is also difficult to interpret because the
exposure history of the sampled fishes cannot be determined. While fish movements
occur in generalized species-specific patterns, estimating the quality or quantity of an
exposure is impossible without the use of some other measure of exposure.
Nonetheless, the investigation of abnormal pathology has been included in the EMAP -
Estuaries program (Weisberg et at. 1991). While the overall incidence of abnormal
pathology was low, they have found greater incidence in small estuaries than in other
classes sampled particularly with bottom dwelling fishes. Thus, this indicator, coupled
with measures of sediment toxicity, was useful in interpreting the relative condition of
classes within EMAP - Estuaries as well as in suggesting chemical contamination as
the cause of degradation.
The use of acoustic techniques for enumerating fish populations is potentially useful in
an EMAP - GL context. Hydroacoustic techniques have been used, fairly successfully,
to enumerate pelagic planktivores in Lake Michigan (Brandt et al. 1991). However,
the selection of appropriate fish community indicators for EMAP - GL has not been
finalized, and until indicators are selected, it is difficult to assess the applicability of
acoustic enumeration techniques. In addition, there are many limitations associated
with present-day acoustic technology which may preclude its immediate use in EMAP -
GL. These include: little or no sampling capability near the bottom and surface, and
uncertainty of target strength values (Brandt et al. 1991; Thorne 1983). Present day
technology is such that all fishes in a mixed-species assemblage must be assumed to
have a particular fish length to acoustic-scattering relationship. Until these
relationships are established for individual species in the Great Lakes, the use of
hydroacoustics for species-specific enumeration will only be possible when considered
in concert with ground truth data obtained by traditional sampling methods (i.e., bottom
trawling).
Because of the extensive activity in the Great Lakes to measure fish populations, it is
proposed that decisions on the indicators to be selected for EMAP - GL be delayed
until a more comprehensive evaluation of existing data can be made during 1992
(refer to Chapter 10). The applicability of concepts such as the harmonic community
and the indicators listed in Table 4.1 (i.e., lake trout/walleye populations, lake trout
recruitment, forage fish populations, and gross pathology) should be considered very
tentative until this evaluation is completed.
4-13
-------
We will also actively coordinate our work on fish indicators with the expertise and
ongoing programs of the U.S. Fish and Wildlife Service and the Great Lakes Fisheries
Commission. These programs include surveys of non-game fish populations. The
ability to use existing information or modifications of existing approaches will be fully
considered before any additional sampling efforts are initiated by EMAP - GL. We do
not want to unnecessarily add new monitoring of fish populations in the Great Lakes.
Table 4.5 Ecological properties of harmonic fish communities and astatic assemblages in
mesotrophic waters of the Great Lakes (from Edwards and Ryder 1990).
Ecological Property
Integration
Stability
Resilience
Identity
(Persistence)
Species ratios
Production/Biomass
community ratio
Yields
Resistance to invasion
Size composition
Complexity
Resource utilization
Harmonic Community
High degree of integration among
indigenous species
Retains semblance of steady-
state
Rapid returns to steady-state
following topological distortion
Retains species identity following
topological distortion
Moderately constant
Circa 0.3
Predictable and constant
Moderately resistant under natural
regime
Slight overlap of niche
Optimal biotic complexity
Maximal
Astatic Assemblage
Random and loose linkage,
particularly when exotics
present
Highly variable
Descriptive identity not
possible over time
Descriptive identity not
possible over time
Changing
0.1-1.0 (variable)
Highly variable
Prone to invasion
High levels of space
contention along some niche
dimensions
Biolic complexity suboptimal
or uneven
Variable and unpredictable
4.4.b. Targeted Invertebrate Populations
In addition to fish community indicators, EMAP - GL will also use indicator organisms
occupying mid-levels in the food web. In oligotrophic regions, abundance of the
amphipod crustacean, Diporeia, is proposed as an indicator. Diporeia responds to a
variety of stresses and its level of response may be easily quantifiable in terms of
absolute or relative abundance (Ryder and Edwards 1985). For example, when
cultural eutrophication occurs in an ultra-oligotrophic system, Diporeia first increases in
abundance as a general response to increased nutrient levels and then decreases
when its nutrient optimum is surpassed. This observation is readily made through the
4-14
-------
use of standardized sampling methods for Diporeia, which assess standing stocks in
terms of numbers and biomass (e.g., Alley and Anderson 1968, Freitag et al. 1976,
Marzolf 1964).
In mesotrophic regions, abundance of the burrowing mayfly, Hexagenia, is proposed
as an indicator. The use of Hexagenia as a benthic indicator would complement
walleye as a pelagic indicator. This mayfly is an important food item for both subadult
and adult walleyes. It is strongly indicative of healthy surficial sediments with
adequate levels of dissolved oxygen in the overlying water column. Mayfly abundance
is easily quantified and historical datasets (e.g., Manny et al. 1988, Hiltunen and
Manny 1982) exist detailing past levels of abundance. Hexagenia occupies an
integrative node in mesotrophic ecosystems in that it tends to reflect the effects of
interactions at the sediment-water interface.
4.4.c. Benthic Community Structure
Sediments in lakes often contain elevated levels of nutrients, metals, organics, and
oxygen demanding substances as the result of anthropogenic input and deposition.
Sediments and the anthropogenic substances associated with particles are also
subject to resuspension, transport, and redeposition. In particular, many of the
problems in the Great Lakes are the result of sediment contamination. At the same
time, the invertebrate communities within and on these sediments are important in the
food web as intermediaries between decomposers, primary producers, and fish. Thus,
they are critical components of the biotic integrity of the lakes.
Benthic macroinvertebrate community structure has been used extensively as a
biomonitoring tool in the Great Lakes (Schneider et al. 1969, Nalepa 1987). These
communities generally form stable associations that integrate and reflect
environmental conditions over long periods of time. However, while extensive
sampling of benthic communities has occurred in some portions of the Great Lakes,
there have been no sustained long-term programs with a regional scope.
There is probably no best period to sample macroinvertebrates because the various
taxonomic groups mature and emerge at different times throughout the growing
season. Thus, any index period will miss some species present at other times. As
with fish, the complexity of possible substrate types and lake zones makes selecting
standardized sampling methods difficult. Sampling is proposed with a combination of
several methods including handpicking from natural substrates, Ekmanฎ or Ponarฎ
grabs, and box core analysis. Comparisons between various sampling methods exist
(Nalepa 1987) and will be utilized in decisions on the techniques to be used.
There are newly developed indices of biotic integrity for macroinvertebrate
assemblages in the Great Lakes but they require field validation (Edwards and Ryder
1990). Considerable research is needed to develop these indices, to modify
applications for different ecological zones of the Great Lakes, and to determine the
sensitivity of species to various stressors. Candidate indices and metrics include:
4-15
-------
taxa richness, number and proportion of indigenous families to the total, and diversity
indices compared to historic estimates.
4.4.d. Primary Producers/Lake Trophic Status
Primary producers have long been used to indicate the status and trends of
ecosystem condition. Primary producers reflect and respond directly to ambient water
quality, are sensitive to water quality changes, are the first or among the first trophic
levels to respond to changes, and form the basis of the food chain which impacts
each successively higher trophic level. In general, trophic status is synonymous with,
or associated with, primary producer assemblages and related exposure and habitat
indicators. There are five prominent methods for monitoring and assessing primary
producer assemblages or trophic condition: 1) algal species composition and
abundance; 2) chlorophyll-a concentrations; 3) Secchi disk transparency; 4) primary
productivity using radiolabelling; and 5) ambient phosphorus concentrations and
annual phosphorus loading.
Phytoplankton are the most important primary producers in large freshwater systems
such as the Great Lakes. Algal assemblages and particular species have often been
used as indicators of water quality and trophic status (Rawson 1956, Hutchinson 1967,
Palmer 1969, Stoermer 1978, Van Landingham 1982). The study of phytoplankton
was considerably accelerated during the 1960s and 1970s on a national basis due to
the eutrophication of waterways, reservoirs, and lakes. Symptoms associated with
eutrophication were taste, odor, and filter-clogging problems at municipal water
supplies, excessive oxygen depletion in lakes, and a deteriorated aesthetic condition.
Investigations were conducted on the role of point and nonpoint sources and
specifically the input of phosphorus as it related to phytoplankton biomass and
secondarily, to floristic composition changes during nitrate and silica depletion
episodes (Dillon and Rigler 1974, Schelske 1975, Rhee and Gotham 1980, Smith
1982, Bierman et al. 1984, Grover 1989). The implementation of phosphorus loading
strategies successfully improved water quality in regard to eutrophication in the Great
Lakes.
In many cases, primary producer-nutrient relationships have been established.
However, phytoplankton assemblages and chlorophyll-a typically exhibit a great deal of
seasonal variation due to physico-chemical changes. Within this seasonal timeframe,
populations may fluctuate dramatically on an hourly-, daily-, and weekly basis. These
factors present challenging problems in the assessment and monitoring of primary
producers, as they relate to the objectives of EMAP - GL.
4.4.d.1. Trophic Status Index
Lake trophic status indices directly address the trophic issues of the Great Lakes. The
classic indicator for lake trophic status is chlorophyll-a, as a reflection of phytoplankton
biomass. Chlorophyll-a has been extensively used as a trophic index in the Great
Lakes (Dobson et al. 1974, ULRG 1976) and elsewhere (Sakamoto 1966, USEPA
1974, Wetzel 1975, Brezonik 1976;1984, Rast and Lee 1978). Similar trophic
4-16
-------
classification systems have been developed for Secchi disk transparency (Dobson et
al. 1974, ULRG 1976, Carlson 1977, Past and Lee 1978), ambient phosphorus
concentrations (USEPA 1974, Vollenweider 1976, Brezonik 1976, Rast and Lee 1978),
primary productivity (Shannon and Brezonik 1972, Vollenweider et al. 1974, Likens
1975, ULRG 1976), and phosphorus loadings (Vollenweider et al. 1974, Rast and Lee
1978), due to their relationships with phytoplankton biomass. In many cases, these
variables are used interchangeably or used in a composite index.
It will be appropriate, however, to determine the most representative single- or
multi-variable index for the Great Lakes. Certain indices may only be appropriate for
certain lakes and may require further delineation for offshore and nearshore waters.
Comparison of the results of different indices must be conducted for the Great Lakes.
For EMAP - GL, Secchi disk transparency, chlorophyll-a, and total phosphorus will be
measured allowing an inspection of single- and multi-variable indices. These
parameters have been routinely measured in the Great Lakes for many years.
Nearshore (<85 m) trophic status will be assessed via the Composite Trophic Index
(CTI), a multi-variable index (Gregor and Rast 1979).
4.4.d.2. Diatoms (Bacillariophyceae)
Diatoms have been extensively used as indicators of a wide array of water quality
conditions. It has been recognized that species and assemblages exhibit sensitivities
and tolerances to different water quality variables (Kolkwitz and Marsson 1908,
Hustedt 1930, Patrick and Reimer 1966; 1975, Hutchinson 1967, Cholnoky 1968, Lowe
1974). Because the siliceous valves of diatoms are retained in lake sediments,
ecological relationships have been applied to the reconstruction of paleoecological
histories from the examination of diatom microfossils in sediment cores (Pennington
1943, Reid 1961, Stockner and Benson 1967, Stoermer and Yang 1968). During the
past decade, application of this technique has significantly accelerated in the
reconstruction of paleoecological/paleolimnological lake histories, particularly regarding
trophic state (e.g., Agebeti and Dickman 1989, Anderson et al. 1990, Whitmore 1991,
Wolin et al. 1991), lake acidification (e.g., Dixit et al. 1988, Charles et al. 1989, Birks
et al. 1990, Dixit et al. 1992), and other water quality variables (e.g., Tuchman et al.
1984, Bradbury 1986, Smol 1988, Kingston and Birks 1990). This approach also has
the resolution to detect the reversibility and recovery of lakes in response to
management strategies (Fritz and Carlson 1982, Battarbee et al. 1988, Anderson et al.
1990, Wolin et al. 1991).
Diatom populations are important floristic components of phytoplankton assemblages
in both marine and freshwater systems. In the Great Lakes, diatoms have historically
been the dominant phytoplanktonic group. Prior to 1955, every Lake Michigan
investigator indicated that diatoms were the dominant phytoplanktonic group.
Ahlstrom (1936) reported that diatoms dominated all samples during all seasons, but
deemed Lake Michigan as a "Dinobryon lake", as this chrysophyte was still regarded
as a protozoan. Davis (1966) suggested that the deep-water Great Lakes should be
more properly classified as "chrysophyte lakes" where diatoms taxonomically reside as
a class in the Division Chrysophyta. With improved optics, staining techniques, and a
4-17
-------
greater use of electron microscopy over the past two decades, soft-bodied flagellates
and picoplankton have received considerably more attention and their importance in
ecosystem productivity has been recognized (Munawar and Munawar 1975, Stoermer
and Sicko-Goad 1977, Fahnenstiel et al. 1986). On an annual basis, however, diatom
populations constitute the dominant portion (number and biovolume) of phytoplank-
tonic assemblages in the majority of areas in the Great Lakes.
Phytoplankton and particularly diatoms have been the subject of study in the Great
Lakes for 150 years. The earliest, traceable published account of Great Lakes
diatoms was by Bailey (1842) from collections made during the summer of 1839 in
northern Lake Huron. Within this timespan, lakewide studies and studies
encompassing large expanses of the Great Lakes have been conducted (Stoermer
1967, Vollenweider et al. 1974, Stoermer et al. 1975, Munawar and Munawar
1976; 1982, Stoermer and Kreis 1980, Kreis et al. 1985, Makarewicz 1988). Generally,
Lakes Michigan and Erie are associated with the greatest bodies of phytoplankton
literature, followed by Lake Ontario, and then Lakes Huron and Superior. The early
research was concentrated in Lake Erie, due to the onset of noticeable eutrophication
symptoms (e.g., Beeton 1961;1965), and some long-term, qualitative and quantitative
records at certain localities exist (Davis 1964; 1965, Hohn 1969, Verduin 1964, Nicholls
etal. 1980).
Phytoplankton investigations in Lake Michigan have a long history and contain a large
amount of diatom literature, many of which are classic accounts. Prior to 1960,
collections were primarily restricted to nearshore regions due to the inaccessibility of
offshore waters (Ehrenberg 1854-56, Briggs 1872, Forbes 1883, Kofoid 1896, Ward
1896, Chase 1904, Leighton 1907, Eddy 1927; 1934, Skvortzow 1937, Daily 1938,
Damman 1941, Lackey 1944, Griffith 1955, Vaughn 1961;1962), with the exception of
the deep-water study by Ahlstrom (1936). As accessibility to the offshore zone
increased, studies on larger expanses of the open waters ensued (Stoermer and
Kopzynska 1967, Stoermer 1968, Holland 1969; 1980, Stoermer et al. 1971; 1972,
Holland and Clafkin 1975, Schelske et al. 1976; 1980; 1983, Stoermer and Stevenson
1979, Stoermer and Tuchman 1979). Additionally, a number of studies in Lake
Michigan include historical examinations and long-term records (Thomas and Chase
1887, Damman 1945; 1960; 1966, Stoermer 1967, Stoermer and Yang 1969; 1970,
Williams 1972, Bowers et al. 1986, Makarewicz 1988). Unfortunately, most of these
monitoring programs have not been sustained for a variety of reasons. At present, the
longest, continuing monitoring programs on Lake Michigan are at the Chicago and
Milwaukee municipal water supplies; however, most recent data are not available in
the open literature.
A number of paleoecological studies of diatom microfossils have been conducted in
the Great Lakes: Lake Superior (Reid 1961, Thayer et al. 1983, Stoermer et al.
1985c), Lake Michigan (Parker and Edgington 1976, Glover 1982, Stoermer and
Wollin 1990, Stoermer et al. 1992), Lake Huron (Stoermer and Yang 1968, Wolin et al.
1988), Lake Erie (Peterson 1975, Frederick 1981, Harris and Vollenweider 1982,
Theriot and Stoermer 1984, Stoermer et al. 1987), and Lake Ontario (Duthie and
Sreenivasa 1971, Stoermer et al. 1985a;1985b;1985c;1989, Wolin et al. 1991). The
4-18
-------
greatest number of studies have been conducted on the lower Great Lakes (Ontario,
Erie) with the emphasis exclusively on the impact of eutrophication on diatom species
composition and abundance. Certain studies have examined the impact of trophic
state changes on diatom valve morphology. Wolin et al. (1991) suggests that the
recent recovery of Lake Ontario, as indicated by diatom assemblages, is in response
to phosphorus loading reductions.
Based on the extensive phytoplankton studies in the Great Lakes, several conclusions
can be drawn. Extant assemblages can be used to assess the ecological condition
and trophic status of the Great Lakes, in regard to nutrient enrichment. Certain
assemblages and their abundances indicate classic extremes in trophic status.
Seasonal and spatial variability, however, are great. As a result of changing
conditions and disturbance, shifts in abundance and species composition become
more intense and temporally would require more intensive monitoring.
The use of diatoms in sediment cores offers a distinct advantage in monitoring these
changes over time. This advantage is due to the integrated, unbroken record they
provide where trophic status can be assessed by species composition and the
abundance of microfossils. Diatoms in dated sediment cores have provided insights
which could only be inferred from phytoplankton collections. The onset of cultural
influences ensued at different times in the Great Lakes and progressed northward.
These initial changes were observed in diatom species composition and abundance,
and also indicate that cultural activities have affected each lake with differing
intensities. At the onset of nutrient enrichment, oligotrophic species are stimulated in
abundance; however, sustained nutrient input decreases their relative abundance and
increases species which are more tolerant of enriched conditions. In the extreme,
certain species are greatly reduced or possibly extirpated and replaced with tolerant
taxa (e.g., Cyclotella flora replaced by Stephanodiscus flora). Although the autecology
of many species and the diatom flora of the Great Lakes is moderately well-known
(Stoermer and Kreis 1978), a single, sentinel species indicative of ecological condition
cannot be applied universally to the Great Lakes. Although certain species can be
associated with different trophic states (Table 4.6), assemblages are more appropriate
in assessing conditions.
As described further in Chapter 10, EMAP - GL will be investigating the application of
paleoreconstruction of diatom populations from dated cores as a means of identifying
historical population composition and distribution within the lakes. However, the use of
cores is likely to be too inaccurate to measure conditions and trends on an annual
basis. An alternative approach to determining annual conditions is through the use of
sediment traps. These traps can be positioned in the lakes to collect sedimented
material over an entire annual cycle. Thus, quantifying the diatoms present in the
collected material would represent an integrated annual collection of the diatom
community. This technique has not been investigated sufficiently at this time and will
also be addressed in the pilot activities conducted during 1992 (refer to Chapter 10).
4-19
-------
Table 4.6 Apparent tolerances of Great Lakes diatoms to trophic conditions.
Oligotrophic: Eutrophic:
Cyclotella comta (Ehr.) Kutz. Actinocyclus normann/fo. subsalsa (Juhl.-Dannf.) Hust.
Cyclotella kutzingiana Thw. Diatoma tenue var. elongation Lyngb.
Cyclotella ocellata Pant. Fragilaria capucina Desm.
Cyclotella operculata (Ag.) Kutz. Melosira granulata (Ehr.) Ralfs
Melosira italica subsp. subartica O. Mull Stephanodiscus binderanus (Kutz.) Kreig.
Rhizosolenia eriensis H.L. Sm. Stephanodiscus tenuis Hust.
Meostrophic: Eurytopic:
Cyclotella comensis Grun. Asterionella formosa Hass.
Cyclotella michiganiana Skv. Fragilaria crotonensis Kitton
Cyclotella stelligera (Cl. & Grun.) V.H. Stephanodiscus niagarae Ehr.
Melosira islandica O. Mull. Tabellaria fenestrata (Lyngb.) Kutz.
Two Great Lakes studies have been identified in regard to sediment traps. Both
studies were conducted in Lake Michigan (NOAA and University of Wisconsin), and
neither study has published results. Additionally, these studies do not appear to be
described in data reports nor are they contained within a computerized database. A
study of this nature has also been conducted in Jellison Hill Pond in Maine (Sweets
1983), which examines diatoms in plankton, sediment traps, and surficial sediments.
There may be a limited number of studies conducted in inland waters and possibly in
the marine environment. Data from these studies will be pursued and examined as
part of the pilot study to determine applicability to EMAP - GL, and literature will be
searched for other similar studies in the Great Lakes and elsewhere.
The primary task for specific diatom studies is the development of a comprehensive
workplan based on literature, ongoing studies, and best available technology. The
workplan will be used for implementation of diatom studies for the entire EMAP - GL
program and will be coordinated with similar studies in the National EMAP Program.
Aspects regarding diatoms include: collection methods, sample preparation,
replication of cores, replication of intervals, replication of sediment traps, and variability
assessments from the above examinations. These will be used to develop spatial,
temporal, and sample number needs as well as to identify any weaknesses that will
require further examination. Other aspects in the workplan will include: quality
assurance, taxonomic and enumeration aspects, inter-laboratory comparisons,
documentation, verification, archiving, data analysis and reporting, and statistical
methods. Additional discussion of these factors will be incorporated in the 1992 pilot
study activities.
4-20
-------
The workplan will require the completion of several tasks conducted during the 1992
pilot and beyond. Tasks consist of: 1) development of the paleolimnological and
sediment trap approaches for diatoms, 2) examination of sediment trap samples from
the NOAA sediment trap recovery planned during fall, 1992, 3) examination of the
previously conducted Lake Michigan sediment trap diatom data, and 4) may require
the analysis of limited sediment cores for diatom populations. Please refer to Chapter
10 for additional information.
Questions to be addressed for diatom populations in sediment traps include:
- What is the temporal resolution required in sediment trap samples (yearly,
quarterly, etc.) to relate to the functional and operational resolution of diatoms
in sediment cores?
- What are the primary depths that should be targeted in sediment trap
placement as it relates to diatoms in sediment cores?
- Is there evidence of diatom dissolution when comparing trap and core
samples?
- How reproducible are diatom populations in replicate sediment trap samples?
4.5. Exposure and Habitat Indicators
Exposure and habitat indicators are intended to serve a diagnostic function when
measured in conjunction with response indicators. The exposure and habitat
indicators will be used in association with response indicators to develop hypotheses
of potential causes of impaired ecological condition. Analysis of the exposure and
habitat indicator data should help to characterize the physical, chemical, and biological
conditions that support healthy ecosystems.
As of 1985, the IJC had designated 42 Areas of Concern (AOCs) for the Great Lakes,
one more was recently added to the list (Figure 4.2). These AOCs primarily consist of
bays and harbors and often where tributaries discharge to the Great Lakes. The
variety of types of problems existing in these areas suggest that the Great Lakes are
susceptible to all the traditional categories of exposure and habitat alterations (Table
4.7). One of the significant unknowns at this time is the contribution of air deposition
to the Great Lakes. Based on data existing for Lake Superior and Green Bay (Lake
Michigan), the airshed above the lakes appears to be an important pathway for the
distribution and loading of air toxics into the lakes (IJC 1987). Research and
monitoring on air toxics deposition will be expanded as the result of the new Clean Air
Act of 1990 and should provide valuable data on potential exposure and sources of
persistent chemicals.
4-21
-------
ซ =
f
1C CC
ซ,
3
.2
.$
J2
n
3 5
I **.*|l
* iM*|i
? ฃ5-E 11
V)
03
^C
CO
CO
ฃ
CD
09
O
c=
CO
CO
03
03
8
o>
0)
0)
0)
ซ
"c
c
I
o
O
CO
(0
V)
0)
(0
CVJ
0)
O)
4-22
-------
Table 4.7 UC Areas of Concern: Summary of Use Impairment Identified by the Jurisdictions in Areas of
Concern and Whether or not Problem Definition and Description of Causes is Complete
Area of Concern
Penninsula Harbour
Jackfish Bay
Nipigon Bay
Thunder Bay
St. Louis Bay/River
Torch Lake
Deer Lake-Carp Creek/River
Manistique River
Menominee River
Fox River/Green Bay
Sheboygan River
Milwaukee Harbor
Waukegan Harbor
Grand Calumet River
Indiana Harbor Canal
Kalamazoo River
Muskegon Lake
White Lake
Saginaw River/Bay
Collingwood Harbour
Severn Sound
Spanish River
Clinton River
Rouge River
River Raisin
Maumee River
Black River
Cuyahoga River
Ashtabula River
Presque Isle Bay
Wheatley Harbour
Buffalo River
Eighteen Mile Creek
Rochester Hmbayment
Oswego River
Bay of Quinte
Port Hope
Metro Toronto
Hamilton Harbour
St. Mary's River
St. Clair River
Detroit River
Niagara River*
St. Lawrence River*
(Cornwall/Massena)
St. Lawrence River
(Cornwall/Massena)
Resrtictions on fish and
wildlife consumption
n
L
Tainting of fish and
wildlife flavor
O
L
O
n
Degradation of fish and
wildlife populations
L
L
D
L
D
L
Fish tumors or other
deformities
O
n
L
L
n
n
L
Bird or animal deformities
or reproduction problems
O
L
L
L
O
O
O
L
O
L
n
L
D
L
Degradation of benthos
n
n
L
L
Restrictions on dredging
activities
L
Eutrophication or
undesirable algae
n
L
n
L
L
Restriction on drinking
water consumption, or
tast and odor problems
D
L
Beach Closings
L
n
L
O
Degradation of aesthetics
Added costs to agriculture
or industry
L
n
Degradtion of
phytoplankton and
zooplankton populations
O
O
L
L
L
L
O
L
L
L
O
n
o
Loss of fish and wildlife
habitat
n
n
n
n
RAP reviewed
bylJC
N
N
N
N
N
Y
Y
Y
Y
Y
N
N
N
N
Y
Y
Y
Y
Y
N
Y
Y
Y
N
N
N
N
N
N
Y
N
N
Y
Y
Y
Y
Y
N
N
N
N
N
N
Based on LTC Review,
problem definition and
description of causes
is complete
N
N
N
Y
Y
Y
N
N
N
N
Y
N
N
N
N
N
Y
N
Y
Symbols Used:
Blank - Data confirm no use impairment
- Beneficial use impaired
O - No data available
* - Use impairments identified by Ontario
4-23
L
Y
N
- Under assessment
- Likely impaired
-Yes
-No
-------
Physical habitat quality characterizes physical conditions that may limit biological
components from reaching the full potential expected for an ecological zone within a
lake. In some cases, the physical habitat limitations are natural, in others, human-
induced. In either situation, physical habitat information is needed to fully interpret the
response indicator data. For all the resource classes, the structural characteristics of
the sediment will be determined as a physical habitat indicator. Additional features will
undoubtedly be included for wetlands as these indicators are developed.
The habitat most important for aquatic species is the surrounding water. Thus,
conventional water quality parameters can be considered as habitat indicators. In
EMAP - GL, these parameters include nutrient status, ionic strength, redox status, and
optical characteristics. Nutrient status addresses the supply of chemical compounds
that often limit the growth of algae and macrophytes. Total nitrogen, total phosphorus,
and silica will be determined. Ratios of these nutrients have been used to determine
unfavorable conditions for diatoms (Holm and Armstrong 1981, Smith 1983). Ionic
strength (e.g., Na, K, Mg, Ca, SO4, NO3, Cl,) indicates the association between water
quality and soil weathering processes, and some anthropogenic disturbances. The
redox status of waters (assessed by D.O., pH, temperature, Mn, and Fe) is a major
factor in the solubility, mobility, and toxicity of many chemicals including nitrogen and
toxic heavy metals. Specific conductance and Secchi disk transparency will also be
measured.
Indicators of exposure are intended to measure possible stress on the biota in
response to toxic chemical contaminants. Measurement of the concentrations of toxic
chemicals in the water column of the Great Lakes was not considered to be a practical
approach for EMAP - GL. Water concentrations of critical contaminants are usually
extremely low and may be quite variable due to intermittent inputs and the movement
of water masses within the lakes. Because the sediments are a sink and potential
source of contaminants in the lakes, exposure indicators for EMAP - GL are focused
on measures within the sediment component of the lakes.
The sediment exposure indicators currently proposed are sediment toxicity tests and
measures of critical contaminants in bulk sediment. Both will be conducted with
sediments collected at the same time and location as benthic community structure to
allow comparisons between these indicators. Sediment toxicity tests will address
possible biological exposure to toxic materials accumulated in sediments. EMAP is
not designed to pinpoint toxic "hotspots", but should be able to characterize, on a lake
basis, the degree to which contaminants are associated with biotic integrity.
Toxicity may be demonstrated by several adverse responses shown by test organisms
(e.g., mortality, impaired growth, reduced reproduction). In addition, there are several
possible methodologic choices, such as which test species to use, type of exposure
(solid-phase, pore water elutriate), acute vs. chronic, and test duration. Other
questions relate to the variability of laboratory operation, the representativeness of the
collected sediments, and how to define "reference" sediments. EMAP - GL will
conduct 10-day acute solid-phase bioassays with Hyalella azteca and also
Chironomus tentans solid-phase chronic assays (Giesy et al. 1990, Ingersoll and
4-24
-------
Nelson 1990, Borgman et al. 1989, Rosiu et al. 1989, Nebeker et al. 1984).
Overlying, clean water will be used in all tests. Mortality is a measurement endpoint
for both tests. Growth is an additional endpoint used only with the Chironomus test.
In addition to measurements on sediments, chemical contaminants in fish will be
determined to assess the exposure to chemicals that fish have experienced. This
indicator expresses a potential health hazard for humans, as well as the biota. As
such, this indicator may be considered to be both a response indicator and an
exposure indicator (as is true for trophic state and sediment toxicity). Because EMAP
emphasizes ecological response, chemical contaminants in fish fits best as an
exposure indicator. The contaminant monitoring effort will be closely coordinated with
the US FWS National Contaminant Biomonitoring program.
Currently, proposed analytes for fish or sediments include the IJC Critical Pollutant List
and additional persistent toxic chemicals identified in the Great Lakes (Table 4.8).
Appendix 1, Hazardous Polluting Substances, and Appendix 2, Potential Hazardous
Polluting Substances, of the Great Lakes WQA (IJC et al. 1989) contain information
on additional chemicals identified in the Great Lakes Basin. Because each of the
lakes differs in the types of contaminants present or expected, not all analytes would
be measured in each lake.
Table 4.8 IJC Critical Pollutant List1 (GLWQB 1987) and additional persistent toxic substances in the
Great Lakes2 (Minister of the Environment 1990).
Benzo(a)pyrene1
Chlordane2
Heptachlor2
Hexachlorobenzene1
PAHs1
PCBs1
EDDT1
Dieldrin1
Endrin2
Lindane2
Mirex1
a and y hexachlorocyclohexane1
DEHP2
Methoxychlor2
PCDDs1
PCDFs1
PCP2
Toxaphene1
As1 Cr2
Cd1 Cu2
Hg1 Se2
Pb1 Zn2
Although sediment exposure indicators will be the focus, harbors and embayments
may be a resource class where water column contaminants are sufficiently high in
concentration and occurrence to warrant measurements. Rather than measure
concentrations directly, a water column toxicity test may be utilized as a measure of
exposure. The test proposed is the standard Ceriodaphnia dubia/affinis seven-day
static renewal test (Weber et al. 1989, ASTM 1988, Mount and Norberg 1984). This
test uses survival and reproduction as endpoints and has been extensively used in
effluent toxicity evaluations.
An additional measure of exposure and habitat condition is the measurement of the
presence of the zebra mussel, Dreissena polymorpha. This invading species has
4-25
-------
become widespread in the Great Lakes basin and has the potential for serious
disruption of existing habitat. Because this is a new invader, with numerous agencies
and scientists conducting research, it is not apparent at this time whether existing
programs to monitor its spread are sufficient to meet EMAP goals. This will be
investigated further over the next year.
At this juncture, it appears that zebra mussels can only be examined in the nearshore
and harbor/embayment resource classes. EMAP - GL and NOAA Mussel Watch have
had preliminary discussions on the collection and analysis of zebra mussels for use in
ooth programs. The use of zebra mussels as an indicator of biotic integrity or trophic
status, however, presents difficulties due to our current state of knowledge. Because
the zebra mussel problem is new and potentially significant, a large number of
research projects have been initiated through other EPA programs as well as through
other agencies. Tracking increases and decreases in abundance may be useful as an
indicator of the status and trends of their distribution. There are several programs
actively developing methods for this purpose outside of EMAP. We are following
these developments through coordination with the zebra mussel program at the ERL-
Duluth laboratory as well as by other agencies through interagency coordination
committees.
Other potential uses as a stressor are unclear because of the limited knowledge of the
role of zebra mussels. For example, the rates of removal of abiotic and biotic solids,
nutrients, and contaminants from the water column as mediated by zebra mussels are
unknown and fluxes to the sediment of these parameters are unknown. More
importantly, the actual effects of the above processes are unknown but have many
implications, e.g., higher concentrations in the sediment, greater productivity, and
greater contamination of benthos and bottom-feeding fishes. These effects are
currently under study. The use of zebra mussels as an exposure indicator appears to
be one of the most likely candidates due to their widespread occurrence. Discussions
between EMAP - GL and NOAA Mussel Watch will continue and may lead to zebra
mussel collections in Lake Michigan at some point. Samples could potentially be
analyzed for contaminant body burden concentrations. In addition, appropriately
conducted studies may lead to a greater understanding of the processes and effects
which then may trigger the use of zebra mussel as an indicator.
4.6. Stressor Indicators
As described in Chapter 2, stressor indicators will be used (along with the exposure
and habitat indicators) to investigate associations with impaired conditions.
Presumably, the stressor indicators will address the "ultimate" cause of impairment
rather than the proximal causes. For example, industrial discharges are an ultimate
source of stress that show up as chemical contaminants in toxic sediments (exposure
indicators) and impact biota (response indicators). This section will describe stressor
indicators only in the broadest view because the highest EMAP priority for
development and implementation goes to the response indicators first and then to the
exposure and habitat indicators.
4-26
-------
Currently, none of the data needed to assess the stressors are planned for field
sampling. Instead, these data will be collected from a variety of existing data sources
including maps, and management and regulatory agency reports and databases.
Landuse and landcover characterizations will be the most general stressor indicators.
Landuse will describe the most prevalent types of human induced stress. For
example, agricultural landuse tends to be associated with increased nutrient and
sediment loading and increased pesticide levels. Urban landuse is associated with
increased toxic materials and nutrient loading. A representation of existing landuse on
a very coarse scale is presented in Figure 4.3.
Population density will describe the general level or intensity of human-induced
activity. Shifts in population density may change potential for stresses imposed on the
lakes. Figure 4.4 depicts the population distribution around the Great Lakes as of the
1980 census.
Pollutant loadings can be assessed by analyzing current pollution discharge permits,
both municipal and industrial. These records will further describe the kinds and
intensity of nutrient, chemical, and thermal stresses. On the US side of the lakes,
data sources include National Pollutant Discharge Elimination System (NPDES)
permits and compliance records which are available through state or regional offices
and through EPA databases stored at the National Computer Center.
Flow and channel modifications can be assessed from the landscape characteri-
zations, and regulatory agency records (e.g., Army Corps of Engineers). Normally,
channelization removes a great deal of the habitat diversity and also may create new
avenues for invasion by non-native species.
Stocking, harvesting, and species introduction records provide information about the
biological stresses in aquatic ecosystems. Management practices which are designed
to enhance fishability often result in a degradation of biotic integrity. Increased
stocking activities may indicate that the waterbody cannot currently maintain the
desired harvest levels. Introduced species (intentional or otherwise) have
considerable potential for decreasing biotic integrity, as is evidenced by the pervasive
influence of the sea lamprey, alewife, and rainbow smelt in the Great Lakes.
4.7. Relationship of Indicators to Assessment Endpoints
As discussed in Section 1.5 (Societally Important Great Lakes Values), the two
assessment endpoints proposed for EMAP - GL are biotic integrity and trophic status.
The purpose of selecting indicators is to allow the quantification of attributes that
provide descriptions of the assessment endpoints. For the most part, it is the
response indicators that will be used in this process.
4-27
-------
00
cx>
o>
re
O
ra
03
a
03
O
I
a
2
UJ
(A
O
'+3
X)
a
s
2
at
il
4-28
-------
CO
oo
O)
CO
o
8
3
o
Ul
a
UJ
.0
**3
J3
I
o
a.
S3
.*
CO
a
ฃ
5
O)
IE
4-29
-------
4.7.a. Biotic integrity
Biotic integrity has been defined by Karr and Dudley (1981) as "a balanced,
integrated, adaptive community of organisms having a species composition, diversity,
and functional organization comparable to that of natural habitats in the region". At
this time EMAP - GL has focused on structural components of ecosystems as its
measure of biotic integrity (Table 4.1). Quantitative measures of integrity under this
definition could include descriptions such as the number of species, population
abundances and age/size structure, the distribution and associations of populations,
and the presence and abundance of keystone or surrogate species. As previously
discussed, the comparison of indicators to expected nominal conditions is inherent in
judgements of biotic integrity. At the present time, we believe it is not possible to
quantify the biotic integrity of an entire Great Lakes ecosystem. While we believe this
is a goal to work toward, we propose to initially work with particular components of the
ecosystem and measures of their condition (status and trends). As our understanding
of ecosystem organization continues to expand, we will work toward improving our
ability to quantify what must now be considered the concept of overall biotic integrity.
Our approach has been to identify what we consider to be the major structural
components that make important contributions to the overall functioning of the Great
Lakes ecosystems. The following discussion outlines how we believe our proposed
indicators contribute to measures of integrity in the Great Lakes. We have addressed
major structural components and attempted to apply our previously described criteria
for indicator development.
Microbes. The microbial community is essential to the cycling and movement
of materials through aquatic ecosystems. Because their role in material
cycling is best described through process rates, we did not consider species
lists and population abundances as appropriate descriptors of the condition of
this group of organisms. Techniques to measure processes such as
nitrification and carbon utilization have been used to a limited extent in the
Great Lakes. The work that has been conducted has shown that some
estimates of nominal and subnominal conditions can be detected at a site-
specific level. The implication of subnominal processing rates of materials to
overall ecosystem condition, however, is not well understood. At the present
time, we have not proposed indicators for EMAP - GL that describe microbial
processes. However, we are aware of ongoing efforts in other programs within
the Agency to investigate these processes more fully in the Great Lakes. As
these efforts become better developed, the microbial community will be
evaluated as potential indicators for EMAP.
Phytoplankton. Phytoplankton represent the major carbon fixation pathway
within the Great Lakes proper. Because phytoplankton populations display
temporal variability, we have focused on diatom species distribution and
abundance determined from sediment cores and sediment traps as surrogates
for the entire phytoplankton community. In addition, diatoms are, in general,
4-30
-------
the most common type of phytoplankton under the historical oligotrophic and
mesotrophic conditions of the lakes.
Aquatic vegetation. Inclusion of macrophytes and periphytic community
organisms will be discussed with Research Planning Committee members and
invited scientists in this field. Preliminary discussions with Dr. Martin Auer
(Michigan Technological University), Dr. Jay Bloomfield (NY State Department
of Environmental Conservation), and Mr. John Madsen (U.S. Army Corps of
Engineers) have provided the following perspectives:
- Macrophytes as indicators of biotic integrity would perhaps work best based
on community structure (density and diversity) measurements. It seems
likely that presence/absence measurements would be variable due to
influences of light availability, wave action, nutrients, and general substrate-
related considerations. While reappearance of macrophyte communities
may signal an improvement in environmental quality, it has been difficult to
relate these observations to the environmental conditions which influence
their occurrence.
- Cladophora glomerata could be an excellent indicator of trophic condition
via evaluation of standing crop (abundance), distribution, or nutrient
(phosphorus) indicators. Each of these parameters have limitations which
are outlined below:
Standing crop varies dramatically over the growing season due to
stochastic, wind-driven, sloughing events. This detachment phenomenon
makes comparison of standing crop measurements among sites or among
years valueless from a status and trends perspective.
Distribution of Cladophora is highly sensitive to light availability and can
provide misleading information with respect to nutrient status, e.g., a
reduction in water level may expose shoal areas to light and foster growth.
Nutrient content, especially phosphorus, has routinely been observed at
elevated stored levels in Cladophora proximate to a point source of
nutrients. The analysis of stored phosphorus content itself is
straightforward, although some effort would be required to develop sampling
protocol which addresses the seasonality in stored phosphorus content.
Plans for using aquatic macrophytes as indicators of the condition of harbors,
embayments, and wetlands will also be discussed through a workshop on
"Indicators of Coastal Great Lakes Wetland Condition", to be held in early 1993.
Benthic invertebrate communities. Benthic invertebrate communities that
contribute to material cycling are a major source of food to predators and play
a significant role in the movement of contaminants through bioturbation of
sediments and food web transfer. Our descriptors of the condition of benthic
4-31
-------
communities depend on measures of the numbers and abundance of species,
the associations among species, and the abundance of key species (e.g.,
Diporeia) that have been proposed as surrogate indicators of ecological
condition. As described in Chapter 10, we propose to participate in ongoing
efforts to describe nominal conditions for benthic community structure in the
Great Lakes.
Zooplankton. Zooplankton function, in part, as an intermediary between
phytoplankton and fish within the Great Lakes. While their ecological
significance is unquestioned, the temporal dynamics of zooplankton
populations make them particularly difficult to characterize without extensive
time dependent sampling. Some monitoring of zooplankton on a spatial scale
has been conducted by GLNPO in their surveillance program. In addition,
research is underway by several investigators to evaluate the ability of
hydroacoustic sounding devices as a means of quantifying zooplankton
communities (as well as fish). As these research data become available, they
will be included as part of the continuing study to assess the application of
zooplankton monitoring to the EMAP program.
Forage fish. Forage fish are the food base for top predators (see below) that
existed historically or that have been stocked into the Great Lakes. As
discussed previously, forage fish populations have changed dramatically over
the past hundred years both as a result of invasions of new species and from
changing predation pressure due to stocking of Pacific salmonids. The
fisheries management practices related to sport fish (top predators) make
forage fish population fluctuations difficult to interpret. Monitoring of forage fish
populations is currently conducted in the Great Lakes but the intensity and
time period for which data is available is uneven across the lakes. This data
and its application to EMAP objectives will be investigated as part of the
overall fisheries analysis.
Top predators (fish). An extensive discussion of the rationale for selecting top
predators as representatives of the fish community and as integrators of the
ecosystem as a whole has been presented earlier in this chapter. As stated in
that discussion, we are not prepared at this time to select specific
measurements for top predators, A detailed analysis'of how EMAP can
complement existing monitoring of fish populations in the Great Lakes will be
conducted as part of the next year's activities.
Wildlife. We use this term as an aggregation of those mammals and birds that
depend on the Great Lakes for their survival. Due to their position in the food
web, they have been found to be sensitive to contamination from persistent
organics (e.g., DDT). The bald eagle has been proposed as an integrative
species to monitor overall condition of the lakes (Minister of the Environment
1990). In addition, herring gull populations have been monitored as part of
existing Great Lakes programs (Minister of the Environment 1990). As a
group, wildlife are integral to the organization and function of the Great Lakes
4-32
-------
ecosystems but pose particularly difficult problems for monitoring through the
EMAP framework.
Humans. The measurement and evaluation of the condition of human
populations around the Great Lakes is not an objective of EMAP. However,
human beings are an important component of the overall biological
organization of ecosystems and are undoubtedly the most significant source of
loadings to the lakes. The relationship between the Great Lakes and human
health can be included in the evaluation of contaminant residues in fish.
Residues will be included not only as part of the monitoring of fish populations
but also can be used to determine the risk associated with their consumption
by both wildlife and human beings. Residues in sport and commercial fish will
be compared to existing fish consumption advisories as well as data on
recommended "safe" levels for wildlife. In addition, the human population
density and distribution around the Great Lakes will be an important stressor
indicator included in the periodic reports that will attempt to interpret the
relationships between response, exposure, habitat, and stressor indicators.
Process rates. As discussed above, EMAP - GL is focusing on structural
measures of biotic integrity rather than functional characteristics. It is
recognized that processes are what make the ecosystem function and this
recognition has led to the selection of measures of structure representative of
major functional groups. The principal approach of EMAP in emphasizing
spatial distribution presents limitations on measures that require repeated visits
to the same site. It is possible, however, that some measures of process rates
can be incorporated into the program. For example, nitrification and
respiration rates of sediment microbes could be determined from sediment
samples collected during one site visit. Similarly, primary production estimates
have been made using spatially distributed surveys for collecting water
samples and then conducting onboard measures of carbon fixation. EMAP -
GL will continue to investigate the feasibility of incorporating such measures
into its sampling framework.
4.7.b. Trophic Status
Trophic status is an indication of the degree of nutrient enrichment and the resultant
degree of productivity. Nutrient concentrations within a lake are the result of loadings
from point sources such as municipal wastewater treatment discharges, nonpoint
runoff from landuse activities such as agriculture, atmospheric deposition,
resuspension of sediments, and other recycling processes. While lake trophic status
* jally a continuum of possible conditions, it is usually conceptualized as either
onyotrophic (low nutrients levels, low productivity), mesotrophic (moderate nutrient
levels, moderate productivity) or eutrophic (high nutrients, high productivity). Historical
data indicate that the Great Lakes could be characterized as predominantly
oligotrophic, with mesotrophic conditions occurring in the shallower bays and
nearshore areas prior to European settlement. Conditions within many portions of the
lower Great Lakes became much more eutrophic with the industrial revolution and
4-33
-------
accompanying growth of the human population around the Great Lakes. One of the
success stories of the environmental era of the 1970's and 1980's was the reduction
in nutrient inputs and the subsequent degree of recovery of many portions of the
lakes. There are still, however, areas that exhibit anthropogenic nutrient enrichment
and this is still a major concern within the Great Lakes basin (Figure 4.5).
There have been a variety of measurements that have been used to quantify lake
trophic status and that are often combined into a trophic status index. These indices
generally utilize a combination of ehlorophyll-a (an indication of phytoplankton
mass), total phosphorus or total nitrogen, and Secchi disk (as an indication of water
transparency). As discussed previously in this chapter, we are proposing to use the
Composite Trophic Index (CTI) of Gregor and Rast (1979) that combines chlorophyll-a,
total phosphorus, and Secchi disk measurements during the period of spring mixing.
The application of the CTI to EMAP objectives will be investigated, as well as the
individual components of the index. Estimates of phytoplankton populations
determined from sediment cores and sedimentation traps will also be evaluated.
4.8. Application of Indicators to Resource Classes
Because the four resource classes (offshore, nearshore, harbors and embayments,
coastal wetlands) have different physical, chemical, and biological characteristics,
some of the indicators proposed and eventually selected may not be appropriate for
each subclass. Table 4.9 presents a tentative listing of indicators proposed for
offshore, nearshore, and harbors and embayments. As has been stated several
times, this listing is tentative until further research and data analysis have been
conducted. No indicators for Great Lakes coastal wetlands are proposed at this time
but will be the subject of future workshops and discussion.
4.9. Sampling Index Period
Nutrient and major ion concentrations in the offshore waters appear to be vertically
homogeneous during spring isothermal conditions in all the lakes, except Lake Erie
(Rosa 1987, Nielson and Stevens 1987, Stevens et al. 1985, Bartone and Schelske
1982, Scavia and Bennett 1980, Shiomi and Chawla 1970). Open lake nutrient trends
based on GLISP sampling have been reported using spring isothermal data for 15
years. The other time period for which long-term data are available is mid to late
summer. However, due to uncertainties related to spatial and temporal
representativeness, trends based on this time period have not traditionally been
reported. Nearshore waters and harbors and embayments have not been studied as
extensively and there is no historically determined index period for nutrients and major
ions. Further analysis of existing data will be conducted during FY92 to determine
whether there is an optimal index period for the nearshore and harbor and embayment
waters.
4-34
-------
o
o o. =
S. o
Q.
O
DC
fz -M w -M *- i a
o o o o o t; o
5 5^ I S^! 5
00
oo
O)
CD
o
CD
(0
O
0)
c
o
T3
CO
C/i
w
(U
CO
0)
ei
0)
(0
4-*
CD
+J
w
o
z
Q.
O
in
O)
4-35
-------
Table 4.9 Proposed indicators, by resource class, of ecological condition for EMAP - GL.
Indicators
Response
Benthic macroinvertebrates
Dipomia/Hexagenia abundance
Forage fish population
Lake trout/walleye
Lake trout recruitment
Fish pathology
Diatom assemblages
Chlorophyll-a composition
Trophic status index
Aquatic vegetation
Exposure
Sediment toxicity
Sediment contamination
Fish contamination
N/P and Si/P ratios
Water column toxicity
Exotics abundance
Habitat
Sediment physical characteristics
Water column optical characteristics
Temperature, pH, etc.
Resource Classes*
Offshore
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
Nearshore
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
Harbors and
Embayments
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
XXX
* Wetland indicators to be evaluated during FY93.
Due to the mixing of the offshore waters during spring isothermal conditions, the
surface layer (0 - 1 m) of the lakes may be adequate for chemistry measurements. As
with the index period discussed above, open lake nutrient trends based on GLISP
sampling at 1 m depths have been reported for several years. Examples of various
statistical tests using historical data to verify the adequacy of this will be included in
the assessment of the Lake Michigan pilot (refer to Chapter 10). Historical data
available from summer sampling and data collected from EMAP - GL offshore sites
during the summer of 1992 will also be analyzed to evaluate the representativeness of
4-36
-------
samples at different depths. For nearshore zones, and harbors and embayments, the
surface value will probably not be as stable.
The proposed sediment sampling includes several indicators: sediment physical
characteristics, chemistry, toxicity, and benthic populations. In order to ensure that the
majority of benthic species with a non-aquatic life history phase will be present in the
aquatic community, sampling should occur in the fall (late August to early October).
Index periods and sampling locations for wetlands will follow EMAP - Wetlands
procedures, adjusted for conditions in the Great Lakes, if necessary.
100 -
_ 80 -
ง
0)
E
3
o
60 -
40 -
20 -
Oligotrophic
Eutrophic
2 3 4 5 6 7 8 9 10 11 12
Composite Trophic Index (CTI)
Figure 4.6 Cumulative Frequency Distribution of the Composite Trophic Index (CTI) calculated for the
nearshore zone of Lake Michigan, Spring 1976.
4.10. Analyses of Existing Data
This section will present results of preliminary analyses conducted on a portion of a
water chemistry database from Lake Michigan. The analyses are far from
comprehensive, but do illustrate some of the problems and concerns which often arise
when dealing with Great Lakes data.
4-37
-------
Secchi depth, chlorophyll-a, and total phosphorus data for Lake Michigan were
retrieved from STORE! for the period of record (1976-1990). The retrieval consists of
data collected by ten different agencies and groups, including federal, state and
academic institutions, over a generally discontinuous time interval (Table 4.10). This
same dataset was used in Chapter 3 to illustrate differences in cumulative distribution
functions of these three parameters using differing definitions of the offshore
boundary.
Table 4.10 Summary of Lake Michigan Database Retrieved from STORE! and Used to Calculate
Composite Trophic Indices (CTIs).
Parameters Sampled
Agency Sampling Years Total P Secchi Chlor a
ACOE1 84-86,89 x
GLNPO2 76,77,80,81,83,84-89 x x x
USGS3 76-79,82,84, 86-88 x x x
EPA Lake Survey 76,79,80 x x x
Illinois EPA 76,77,84-88, 90 x x
EPA Reg. V 76,77,79,81-90 x x x
IN Bd Health 80,81 x
Michigan DNR4 76-90 x x x
Wl DNR4 76-90 x x x
University of Michigan5 76-77 x x x
1 ACOE = Army Corps of Engineers, Chicago;2 GLNPO = Great Lakes National Program
Office, US EPA;3 USGS = United States Geological Survey;4 DNR = Department of Natural
Resources;5 Univ. Mich. = University of Michigan, Ann Arbor in cooperation with GLNPO.
The three parameters mentioned above will be used to assess trophic status in the
nearshore zones through use of the Composite Trophic Index (CTI) (Gregor and Past
1979). Figure 4.6 illustrates how these data will be used in an EMAP context. Certain
values of the CTI correspond roughly to trophic status. It is evident that the majority
of the nearshore stations in spring 1976 were oligotrophic and mesotrophic in nature
with only a few (< 5%) in the eutrophic category. Although the data used to generate
the CTI has not undergone extensive quality control procedures, the cumulative
distribution function of the CTI will eventually be used to determine subnominal
thresholds relating to the trophic status endpoint. Sampling stations in historically
mesotrophic areas of the Great Lakes will have a higher threshold CTI than historically
oligotrophic nearshore areas.
4-38
-------
5. Estimation and Analysis
5.1. Introduction
Measurements of the indicators for EMAP - GL will be made with known precision and
confidence. The next step is to provide analysis and interpretation of status and
trends. In general, status will be portrayed through descriptive statistics, visual
displays of spatial patterns and estimates of the proportion of the resource class in
various categories using classification and cumulative distribution functions (cdf).
Yearly statistical summaries will be combined to develop more comprehensive
statements and interpretations of the status of the Great Lakes. As the database
grows over the years, trends of indicators will be estimated by examining changes in
statistical descriptors including changes in proportions in various categories and the
cumulative distribution functions.
At each site selected for sampling, a series of indicators will be measured to give a
representation of the overall health of the Great Lakes primary resources. These
indicators were designed to address three major attributes of concern: 1) response
indicators to describe the biotic condition of the aquatic ecosystem; 2) habitat
indicators for describing the physical condition of the environment; and 3) exposure
indicators that characterize the impaired condition of the habitat.
5.2. Sampling Design
The use of probability-based sampling permits inferences about the condition of the
resource population, i.e., the Great Lakes resources. The sampling design allows the
flexibility to define and to refine classification schemes (e.g., nominal and subnominal)
through estimates of the proportion of the total area sampled. We can make these
predictions with measurable confidence and we can increase the level of confidence in
the estimate if necessary. The strict adherence to an overall sampling design allows
assessment of the spatial and temporal variation within each primary resource class.
We are also assured that information on status gathered over time can be used to
measure trends and change of the resource classes and that EMAP - GL information
can be interfaced with other EMAP programs (e.g., forests, lakes, wetlands, etc.) for
more inclusive statements on ecological health or condition over larger spatial scales.
Boundaries of two resource classes are well defined (refer to Chapter 3) and are not
expected to be changed during the next five years. A grid sample will be used for the
nearshore and offshore resource classes. Sampling design and frame have not been
established and are under development for the harbor and embayment and coastal
wetland resource classes.
Indicator selection for EMAP - GL has progressed to a tentative list of primary
measurements (see Table 4.1). Based on these primary measurements, we can
calculate secondary indices, e.g., Composite Trophic Index as described in Chapter 4.
5-1
-------
We may eventually attempt to combine all of our indicators and indices into an overall
assessment of the health of the Great Lakes as discussed below (refer to section 5.6).
5.3. General Statistical Overview
The analysis will follow the general model described in "Design Report for EMAP"
(Overton et al. 1991) and "EMAP Sampling Design Implementation, Perspectives and
Issues" (Stevens et al. in press). The general model is presented in terms of the
Horvitz-Thompson (HT) estimator (Horvitz and Thompson 1952). The HT estimator
can be used with any finite population probability sample, and requires only
specification of the inclusion probabilities. First order inclusion probabilities are the
probabilities with which the individual sampling units are included in the sample. The
HT estimator requires that every unit in the population have positive first order
inclusion probability; however, the actual values need only be calculated for each unit
that is selected for sampling. These are designated by the symbol K-, referring to the ith
sampling unit. Second order, or pairwise, inclusion probabilities are the probabilities
with which two specific sampling units are included in the sample. These are
designated as TC, , referring to the probability of simultaneously including units i and j.
Pairwise inclusion probabilities are necessary to calculate the variance of the HT
estimator. Design features such as stratification, methods of randomization, sample
selection methods, and sample size are required to determine it-tl.
The HT estimator of a total is given by.
f - ฃ y<
y ~
fe S 7C,.
where y is any attribute, Ty is the total of that attribute over any specific identified
population, and "A" denotes the estimator as opposed to the population parameter.
The summation is restricted to the specific set of units, S, in the sample or any subset
of the sample defined by a specific population. If the units in the sample are
numbered from 1 to n, where n is the sample size, (1) takes the form
t, - ฃ *
/=1 7C,
In some instances, interest will focus on totals; in other cases, there is more interest in
means or proportions. An estimate of the mean is obtained by dividing f by N, the
" 1
number of units in the population or its estimator N = E :
5-2
-------
y =
ฃ, population size known
N
(3)
ฃ, population size estimated
N
Estimators of proportions are obtained using indicator functions. For example, to
estimate the proportion of the population with attribute A, set y, = 1 if sample unit i has
attribute A, and set y-t - 0 otherwise, and apply (5.1). Then y
N
provides the desired
estimator. More generally, a comprehensive characterization of the populations is by
the estimated cdf for the variable of interest. The cdf of a variable Y, written FY(y),
represents the proportion of the population that has value of the variable Y less than
or equal to the number y. For example, if Y were the concentration of chlorophyll a
(c.a.), then Fca(2) = 0.30 would mean that 30% of the target population has a value of
c.a. less than or equal to 2.
The HT estimator of FY(y) is
FY (y) =
;=1
JC
(4)
where l(yy '
The Yates-Grundy estimator of variance for (1) is given by
n n
5" T
ฃ-1 ฃj
K
(5)
Basic population parameters are estimated by some form of (1), and so are strict HT
estimators. The variance estimator (5) is unbiased if all pain/vise inclusion probabilities
are positive and known exactly. However, systematic random designs, such as
EMAP's grid based design, have some joint inclusion probabilities that are zero.
Moreover, even the non-zero joint inclusion probabilities can be difficult to calculate
exactly. These design features will require approximations in calculating joint inclusion
5-3
-------
probabilities. The approximation developed by Overton (1987b) is used in most
instances. The approximation is derived under the assumption that the population is
randomized between draws. This assumption does not hold for most EMAP samples,
but simulation studies (Overton and Stehman 1987, Stehman and Overton 1987a;b)
have demonstrated that this approximation along with the variance estimator (5.2)
performs well in EMAP-like sampling circumstances. A convenient computational form
of the pairwrse inclusion probability approximation is given by
itjj = 2(n-1)7t|7Cj/(2n - n-, - K-):
The use of this approximation in conjunction with the Yates-Grundy variance estimator
in EMAP - GL amounts to assuming that the grid-based sample is nearly a simple
random sample. The grid-based sample should provide more precise estimates that
simple random sampling, so the (5) will provide conservative estimates of precision.
Variance estimates that account for the systematic design effect, such as
generalizations of the Yates successive difference estimators (Yates 1953), are being
investigated by the EMAP Statistics and Design Team.
The Yates-Grundy variance estimator is not an appropriate estimator for use with a
systematic design because of the failure of the assumption that ^ > 0 for all i,j. The
variance approximation we propose using is derived from the YG estimator by
replacing Tty with 2(n-1)7Cj7c/(2 n - n-t - KJ). The variance estimator then becomes strictly
a function of ic, and i^. Moreover, if in addition jq = n, a constant, for all i (as for a
uniform grid), and the HT estimate of the population size is used, then the variance
approximation for the mean collapses to (1 - 7t)s2/n, where s2 is the usual SRS based
estimate of the population variance. The factor (1 - TC) can be thought of as a finite
population correction factor.
We regard this as a working approximation that needs to be refined. Several
refinements are being investigated by the EMAP Statistics and Design Team. In
particular, most (all?) popular variance estimators for systematic samples are derived
for finite populations. The continuous population analogues need to be derived and
their properties investigated. Along the same lines, we are investigating variance
estimators that utilize the spatial structure of the population.
These general forms specified in the EMAP design protocol can be used with any
probability sampling design, including those with unequal probabilities of selection or
extensive stratification. They are presented here to demonstrate the ties between
EMAP - GL and the overall EMAP. However, under equal probability sampling as is
used for Great Lakes primary resource classes (open water and nearshore), the
general forms simplify to more familiar forms for estimates of the usual descriptive
statistics and their variance estimates.
5-4
-------
5.4. Analysis for Biotic Integrity
The assessment of status of the biological resources in the Great Lakes will require
the systematic and complete analysis of response, habitat, and exposure indicators.
For the strategy document, we will demonstrate our approach for statistical analysis of
a hypothetical dataset. We have assumed that chlorophyll-a was collected from 11
offshore sites around Lake Michigan. Chlorophyll-a is an indicator of algal biomass,
therefore it can be used as a measure of nutrient level and overall water quality. The
chlorophyll-a measurements are grouped so that cumulative distributions can easily be
calculated.
5.4.1. Descriptive Statistics/Visualization
The mean chlorophyll-a content in Lake Michigan open water was calculated for our
hypothetical dataset. The mean chlorophyll-a level as calculated from the dataset is
1.6 u,g/L, with a standard deviation of 0.22 u.g/L. The coefficient of variation (standard
deviation/mean) gives a relative measure of the variability of the different resource
indicators within the lake. For chlorophyll-a the coefficient of variation is 14%. A
small coefficient of variation would suggest that the resource is uniformly distributed.
The coefficient of variation for chlorophylt-a would suggest that the offshore regions of
Lake Michigan are homogeneous in regard to this indicator. Refer also to Chapter 3
for discussion of chlorophyll-a distributions.
5.4.2. Classification/Cumulative Distribution Functions
As discussed in Chapter 4, one of the desired goals of the assessment phase of
EMAP - GL would be to classify the responses into nominal and subnominal
conditions. This classification process will be based on a set of rules but the decision
points that separate each resource class into different groups will be flexible and
subject to modification as our expertise and knowledge grows. Determining
proportions of a class, e.g. offshore area, in nominal and subnominal groups can be
accomplished through the use of the cumulative distribution functions. However, the
decision points or indicator values that separate groups and therefore define the
proportions in the groups can be changed to accommodate new information or new
regulatory decisions. Because cumulative distribution functions represent the
complete distribution of values, the proportion of values that are above or below any
reference value can be estimated visually and the effect of changes in nominal and
subnominal boundary values on results can be evaluated without reanalysis of the
data. For example, the ecological objective workgroup to the IJC recommended an
objective of annual lake trout production greater than 0.38 kg/ha as determined using
mortality rates for Lake Superior. If we were to use lake trout production as an
indicator, we could initially use this a classification of nominal/subnominal for Lake
Superior. As the database on lake trout production in Lake Superior improved, some
other number may be determined to be a better delineation of the quality of the fish
community. This approach allows some initial classification but maintains the flexibility
for future descriptions.
5-5
-------
Assuming we can define degraded conditions based on single or multiple indices, we
can describe the status of each major resource class within the Great Lakes on an
area! basis. For instance, we could describe the trophic status of offshore areas either
by percent or actual area (km2). The degree to which these change over time would
be the measure of trends.
Using the hypothetical data for chlorophyll discussed above, we can assume that
offshore sites with chlorophyll values greater than 2 u.g/L are eutrophic. This would
mean that 27% of the offshore area is subnominal or eutrophic. If we assume that the
sampling grid is equivalent to a simple random sample, we can calculate 90%
confidence limits around this proportion (0), i.e.,
(6) = x/n
variance(9) = (9)*((1-(9))/n
90% Confidence Limits = 9 ฑ [t-value*S.D.(8)]
Where x = 3, the number of sample sites with chlorophyll-a greater than 2
n = 11, the number of sample sites
t-value = 1.80
Our 90% confidence limits are derived from the t-distribution, mean, and standard
deviation for 9, where n=11. The 90% confidence limits for the proportion 0.27 are
(0.04-0.50). The same procedure can be used to develop confidence limits around
the cdfs for the proportion of the population that is above or below a specified value of
chlorophyll-a. The confidence limits that were calculated for the cdf are based on an
assumption of a random sample. These confidence limits are conservative and, in the
future, the statistical support team will investigate alternative estimators of the
variance.
In addition to single variable (indicator or index) statistics, multivariate analysis, e.g.,
principal component or discriminant analysis will be used to find a linear combination
of indicators and indices that describe or discriminate between research sites for each
resource class. The multivariate approach will complement our other statistical
methods.
5.4.3. Estimates for Combined Resource Classes
The nearshore and offshore portions will be sampled with different density grids, so
that the inclusion probability densities will be different for the two resource classes. If
a combined estimate is desired for a common indicator, it could be obtained via the
general variable probability estimators. However, the general forms can again be
simplified by making them specific to the two resource classes. Let A^ be the area of
offshore waters and An8 be the area of nearshore waters. If y^ and /, denote
estimates of some quantity for the offshore and nearshore classes, respectively, then
a combined estimate is given by
5-6
-------
(6)
i.e., the composite estimate is given by combining the individual estimates with
weights proportional to their respective areas.
5.5. Analysis of Change and Trend
EMAP places emphasis on detection of change and trends in ecological condition.
We plan on extensive analyses of data using a variety of statistical techniques ranging
from classical techniques based linear model theory to non-parametric techniques to
techniques that are still under development that draw upon the theory of spatial
statistics and sampling theory. Some of these approaches are described in more
detail below.
5.5.1. Linear Model Analysis of Trend
Linear model techniques, such as analysis of variance and regression, are used with
various assumptions regarding spatial and temporal variance components, statistical
independence, explanatory variables, nature of trend, and nature of change. A linear
model to evaluate the effect of having some annual observations in the
interpenetrating design might begin with the model
ys = M, + etj + a, + bj + e,
where u. is the average value at time 0, 0 is the slope over time (fixed, linear trend), a,
is the effect of the im site, bj is the effect of the jlh year, and e^ is zero mean random
error with standard deviation ae. The site and year effects are random, with standard
deviations aa and ab, respectively. Under this model, aa describes inherent population
variation, ob describes the variation from year to year, and
-------
for a number of years. The basic arrangement for a rotating panel is for the total
sample to be split into the same number of equal-sized sets as the number of years a
set remains in the sample. Thus, for example, with a 4 year rotation, one fourth of the
sample is replaced every year. The serially alternating design again splits the total
sample into several equal-sized sets, but only one set is visited each year. A set is
not revisited until all other sets have been visited, and the serial revisiting is continued
indefinitely. The basic serially alternating design does not prescribe any replacement
or annual revisits. Both designs can be augmented by adding a set of sites that are
visited annually for the duration of the monitoring program.
Urquhart et al. (1991) used a general linear model to compare the relative efficiency of
these two designs. The linear model they used gave them sufficient flexibility to
consider the estimation of both status and trend, and to explore various levels of
population variation, measurement error, and inter-annual variation, and to incorporate
some correlation between years, and between sites measured at different times. Their
conclusions were that the serially alternating design is almost always more efficient
than the rotating panel, and, over the wide range of possibilities that they investigated,
was never less than 99% as efficient. Moreover, they concluded that the augmented
serially alternating design offered a substantial advantage in ability to make estimates
for subpopulations, since more sites are visited sooner than with the rotating panel.
The approach used by Urquhart et al. (1991) can also be used to gain insight into the
power of the design to detect change, since an output of their model is the precision
of a trend estimate. In order to assess that power, estimates of the several
components of variance and correlations are needed. For some indicators, these may
be available from existing data, or they may be obtained from pilot experiments.
5.5.2. Non-Parametric Method for Trend Detection
The Mann-Kendall (Mann 1945, Kendall 1975) test has been used to detect trends in
water quality (Hirsch et al. 1982) and has been recommended for use in more general
environmental applications (Gilbert 1987, Loftis et al. 1989). The Mann-Kendall test
n-1 n
statistic is S = ฃ ฃ sgn(Xj - xk) , where sgn(Xj - xk) =
/c=1 M+1
0, Xj-xk= 0 , and
-1, x,-x(f<0
{x,, X2, ..., xn} is a sequence of observations at a single fixed site. The applications
have been primarily concerned with evaluating trend at a single site, although both
Gilbert (1987) and Loftis et al. (1989) suggest approaches to the multiple site situation,
A chi-squared test was used in the Phase II Eastern Lake Survey to examine
population distributions for evidence of change (Overton I987a). The test can be
applied to both successive independent samples from a population, and to
remeasurements on successive occasions. In the independent sample case, the test
is carried out by using the cdf of the pooled samples to define classification criteria
(for example, quintiles of the combined population) and doing a chi-squared test of
5-8
-------
homogeneity. A pair of such tests, classifying first by the median and then by quintiles
of the pooled distribution, provide sensitivity to changes in shape as well as changes
in location of the distribution function. For the remeasurement or paired sample case,
the test is carried out by classifying according to the sign of the differences and to
percentiles of the cdf of the sum. Both versions can be extended to multiple years.
5.5.3. Power to Detect Change and Trend
An important consideration in the selection of techniques for change and trend
detection is the sensitivity of the tests, or, in statistical terms, the power of the tests.
Statistical power is the probability that a test of an hypothesis is rejected given that the
hypothesis is false. Generally, power increases with increasing departure from the
hypothesis, that is, a large change is more likely to be detected than a small change.
A statistical test of a particular hypothesis can be characterized by specifying the size
of the test (probability of rejecting a true hypothesis) and its power function (probability
of rejecting a false hypothesis as a function of degree of departure from the
hypothesis). Where choices among alternative methods of change and trend
detection exist, the more powerful test is preferred.
Power will be used to evaluate the adequacy of the design, and to determine the
sample sizes needed for specific subpopulations. The evaluation is in terms of
describing the probability of detecting a change or trend of a given magnitude. If the
magnitude of change that is detectable with, say 80% probability, is unacceptably
large for some subpopulation, then the sample size for that subpopulation may have to
be increased. Provisions for doing so are incorporated in the general EMAP design.
The EMAP-Statistics and Design Team is actively conducting studies of the power of
proposed tests for change and trend using simulation studies based on realistic data.
An important aspect of EMAP's interpenetrating design is that it will achieve its full
potential power to detect change only after repeat visits to all sites, that is, after two
complete cycles. Repeat visits to a site permit a paired analysis which essentially
eliminates the component of population variation. Thus, ability to detect change will
increase greatly in years 5 through 8 of the sampling. Similarly, the power to detect a
persistent trend will continue to increase as more years of data become available.
5.5.4. Associations
One of the objectives of EMAP - GL is to seek associations between ecological
condition, as determined from response indicators, and exposure and habitat
indicators. Associations among indicators will be evaluated using a suite of correlation
techniques including both parametric and non-parametric tests. Categorical and
logistic regressions are the techniques that will probably be more successful. The
statistical analysis to be selected will depend upon the characteristics of the data for
each indicator and the specific question to be answered by each analysis.
This approach will be applied to each of the resource classes to evaluate associations
that might be masked by combining all classes into one analysis. For example,
5-9
-------
harbors are likely to be subject to different stresses than the offshore class of the
lakes. The amount of data obtained for each class must be sufficiently large to ensure
that uncertainty associated with conclusions is not unacceptably large. Thus, this is
another factor to consider in evaluating the density of the proposed sampling network.
5.6. Great Lakes Ecological Condition Index: A Conceptual Proposal
Any program that professes the goal of describing the ecological condition of the
Great Lakes faces the difficulty of integrating a variety of indicators into an overall
statement of condition. One goal of EMAP is not only to describe the status and
trends of specific indicators but also to integrate these measurements into statements
that can be used by scientists, decision makers, and the general public. The scientific
community has attempted such integration through the use of indicators that integrate
overall ecosystem condition (Edwards et al. 1990) and through the use of indices.
Indices have been developed and used to describe various types of diversity (i.e.,
Shannon-Weaver index) or the biotic integrity of certain groups of organisms (i.e.,
Index of Biotic Integrity for freshwater fishes). There is a great deal of literature
devoted to both proposing such indices and to pointing out the difficulties, errors, or
limitations for their use. We do not intend to review all such literature in this report.
However, in the spirit of continuing to make attempts to provide an integrative
framework, the following discussion focuses on yet another approach. This approach,
still in its infancy, is derived primarily from some approaches being taken in Europe
and described in Rojanschi et al. (1991) and ten Brink et al. (1991).
The basic premise associated with this proposal is that indicators which are selected:
can be measured quantitatively and accurately;
are susceptible to human influence;
have some indicative value for the condition of the systems; and
have some social and political value.
It is our intention in the selection of EMAP - GL indicators (as previously discussed in
Chapter 4) to meet these criteria. In addition, there must be knowledge, information,
or decisions on what is the desired (nominal) state for each of the indicators. The
definition of nominal, of course, is not an a priori decision that can be made through
science alone. As also discussed in Chapter 4, some determination of nominal values
may be possible through the use of historical data, reference sites, and ecological
models. However, societal values are also a significant component.
Assuming indicators have been selected and nominal (desired) values for those
indicators are chosen, there remains the question of integration. The proposal is to
take a two-dimensional shape (we will use a circle for an example) with each indicator
represented as a radius of the circle. The point at which the indicator radius intersects
with the circle is defined as the nominal value. In other words, the length of the radius
is (by definition) quantified into units of the indicator with the circle connecting the
radius for each indicator (Figure 5.1). The center of the circle represents the zero
5-10
-------
point for the indicator, that is, the indicator is not present. In essence, the radius (in
the case of a circle) represents the axis of a quantified indicator measurement. An
alternative approach would be to characterize the nominal condition of an indicator as
100% of the length of the radius representing that indicator. In either case, the actual
measured value for each indicator is plotted against its radius. The measured points
are connected and a new interior shape is created (Figure 5.1). The index of overall
condition is thus described as a percent area of the nominal area (the whole circle).
Thus, the status of the ecosystem can be described, for example, as 50% of nominal
and trends in either direction can be reported and graphically displayed.
B
B
Example diagram with four
Indicators. "A-D" represent the
nominal condition for each
Indicator. Assumes equal
weighting and no correlation
between indicators.
Example diagram with "1-4*
representing actual measured
values for the four indicators.
The shaded area divided by
the total area would equal the
Index of overs! condition.
Figure 5.1 Great Lakes ecological condition index: A conceptual proposal.
As with any idea, there are a number of .difficulties that quickly arise from what seems
like a simple concept. The formation of a circle, for example, assumes that each
indicator has equal value. In contrast, if weighting indicators is desired, the
appropriate shape may not be a circle but rather some other geographic shape. This
would result from having the line representing the nominal condition of the indicators
consist of different lengths depending on their relative weight. For example, some
individuals might argue that top predator fish have more value to society and should
be weighted more heavily than a particular species of benthic invertebrate. Society or
regulations might dictate that endangered species should have more weight in such an
analysis than a common species. While weighting might result in different geographic
5-11
-------
shapes, the basic analysis of overall condition would follow the same process as
described above.
Another difficulty is that the basic shape of the nominal condition graph depends on
the distance separating the indicator axis. If all indicators are equally weighted and
have no correlation between them, the spacing of the indicator axis would be different
(and therefore forming a different shape and area) than if the indicators were
correlated. The degree of correlation would need to be known. Some techniques for
defining appropriate shapes based on the degree of autocorrelation have recently
been developed in the mathematics of graph theory and may be applicable to this
problem.
As described above, this approach to simplifying the analysis and display of complex
interactions is clearly in the preliminary phases of thought and discussion. The
reactions and ideas of many scientists will be solicited to determine if there is any
value added (or lost) in attempting to develop such an index and if so, to help develop
both the concept and the mechanics of its construction.
5-12
-------
6. Logistics Approach
6.1. Introduction
This section describes the logistics approach for implementing EMAP - GL. It includes
a summary of requirements for the logistics plans, a discussion of the major logistics
issues, a field operations scenario, and a proposed organizational structure.
6.2. Logistics Implementation Components
Implementing the EMAP - GL program will require detailed, comprehensive logistics
planning. Logistics considerations include coordination and oversight of all
implementation support activities and the actual data collection activities. A logistics
plan must be developed prior to start of implementation of field activities to assure that
the goals of the program are met. The logistics plan should include all elements given
in Table 6.1 as specified by US EPA (1990a).
Table 6.1 EMAP Logistical Elements for Implementation of Great Lakes
Monitoring Programs.
1. Overview of Logistical
Activities
2. Staffing
3. Communications
4. Sampling Schedule
5. Site Access
6. Reconnaissance
7. Waste Disposal Plan
8. Safety Plan
9. Procurement / Inventory
Control
10. Training
11. Field Operations
12. Laboratory Operations
13. Data Management Activities
14. Quality Assurance
15. Logistics Review/
Recommendations
Element 1. Overview of Logistical Activities-Summarize the types of activities
required to complete the project. Maintain a timeline showing all critical path
milestones, e.g., project design, indicator selection, site selection, reconnaissance,
procurement, methods selection, development of standard operating procedures, and
resolution of specific quality assurance issues.
Element 2. Staffing and Personnel Requirements-Describe the number of personnel
and the organizational structure necessary to accomplish project objectives. Define
6-1
-------
who is responsible for staffing and interagency and teaming mechanisms. Consider
work schedules to determine whether extra positions should be created or whether
existing personnel should work overtime. Create a contingency plan for replacing staff
members when necessary. Identify key personnel and provide plans for retaining
them.
Element 3. Communications-Address communications among field crews, laboratory
crews, and supervisory personnel, and between EMAP participants and any local
organizations who should be informed of EMAP field activities. Also include plans for
tracking samples, data, crews, equipment, and supplies. Discuss how field crews
should interact with the public or with the media. Explain how approved changes in
standard operating procedures will be documented and communicated for
implementation.
Element 4. Sampling Schedule-Based on project, indicator, and statistical design or
other program requirements, devise an efficient schedule for field activities. Consider
geographical sampling windows within geographical areas and other factors such as
climate and site access constraints.
Element 5. Site Access-Address issues related to gaining access to sampling sites
including scientific collection permits, if required. Develop a list of local contacts to
discern property ownership, jurisdiction, and the best site access methods. Address
plans to obtain appropriate access permission and applicable collection permits.
Consider how to coordinate activities in the same area of more than one resource task
group. Discuss ways to arrange long-term access rights, track changes in ownership
of private sites and management of public sites, notification of owners and managers
before revisiting the sites for future monitoring, and provide contingency plans in case
of future failure to obtain access permission.
Element 6. Reconnaissance-Define criteria for selecting base operation sites (take
into consideration personnel and technical support requirements), geographical
location with respect to sampling sites, and time constraints imposed by sampling
design or climate. Sampling sites identified as having potentially difficult physical or
legal access should be visited during field reconnaissance. Additional resources
needed for sampling should be identified if the access problem is due to physical
conditions.
Element 7. Waste Disposal Plan-Explain how chemical and biological wastes will be
stored, transported, and disposed of safely and legally. Address what permits will be
needed for storage, transport, and disposal of wastes.
Element 8. Safety Plan-Discuss how emergency situations will be evaluated and
handled. Determine which emergency services will be available in the field. Explain
what procedures will be used to initiate search and rescue operations. List the
training or other preventive measures required to conduct field operations safely.
Indicate how this field safety plan will be developed in conjunction with laboratory,
processing, and materials handling safety plans.
6-2
-------
Element 9. Procurement and Inventory Control-Identify equipment, supply, inventory
control and resupply, and service requirements of the field program. Consider the
shipping requirements for chemical and biological materials. Determine what
analytical or other services will be needed and the best mechanisms for acquiring
them. A procurement schedule should be provided for all items.
Element 10. Training Program-Describe who will prepare, review, and revise the field
training and operations manual and the procedures for field measurements, sampling,
sample handling shipment, data recording, quality control, safety, waste disposal, and
communications. Outline a schedule for the completion of these items. Describe
training needs and identify who will conduct and review training. Address how
personnel will be evaluated to ensure competency.
Element 11. Field Operations-Indicate the organizations that will perform each of the
daily field activities. Describe how and when the daily field activities will be performed.
Discuss and schedule the major events within field operations (i.e., mobilization,
demobilization, and phase changes in sampling activities). Consider contingencies
such as back-up personnel in the event of sickness. Require real time evaluation to
identify and resolve problems.
Element 12. Laboratory Operation-Indicate what organizations will be responsible for
each type of sample preparation or analysis and for formulating each laboratory
operations manual. If EPA conducts the activities directly, provide a development plan
for providing appropriate laboratory facilities.
Element 13. Information Management-Describe any data management activities that
might be affected directly by field operations. Establish guidelines for the timely and
responsive transferral of information from field personnel to data managers. Indicate
the groups that will be responsible for preparing and reviewing field data forms;
provide a schedule for the completion of these forms. Develop a schedule for
completion of the information management plan by the information management
group.
Element 14. Quality Assurance-Describe who will provide input to the QA plan on
field sampling, sample handling and preparation, sample shipment, sample disposition,
and data management. A schedule for completing the QA plan should be provided to
the logistics team and included in the logistics plan. QA activities should be
coordinated with other resource groups using similar methods. This effort should
identify common methods and standards when possible.
Element 15. Logistics Review and Recommendations-For each year of study within
each resource group, summarize logistics activities. Discuss how personnel will be
debriefed to identify and resolve problems. Discuss pilot studies and associated
methods evaluation experiments; present logistics data summaries within the full-scale
project.
6-3
-------
Field activities will start in 1992 with pilot programs in Lake Michigan and Lake
Superior. Additional Great Lakes will be phased into the program in each of the
following years.
6.3. Logistics Issues
The complexity of this program poses a number of logistics issues; overlooking or
ignoring apparently minor issues or details will eventually jeopardize the success of
the program. These issues will be addressed fully in each of the logistics plans prior
to the full implementation of field activities. A brief discussion of the major issues
(staffing, access, and data confidentiality) is provided in the following sections.
6.3.a. Staffing
Due to the type of field data needed for indicator evaluation (Chapter 4), field
personnel will require a nigh degree of expertise. They must have knowledge of fish,
macroinvertebrate, and diatom taxonomy, field sampling methods, and sample
handling. Various state resource agencies, EPA regional offices, other federal
agencies, and universities have large pools of experienced personnel. Long-term
agreements with these agencies and institutions to provide key personnel during the
field season may provide a solution to this issue. To accomplish this, EMAP will have
to demonstrate its utility to the other organizations by providing additional data and
information addressing their problems. A concerted effort to inform these
organizations of the goals and objectives of EMAP, and getting these organizations
involved in the early planning phases of the programs are initial steps being taken.
6.3.b. Access
Obtaining access information and permission to visit sampling sites involves public or
private authorization. If land is publicly owned, approval must be obtained from the
appropriate authority. If land is owned privately, each landowner will have to be
contacted and written access permission will have to be obtained. For the most part,
access is only a potential problem for the wetland class of EMAP - GL. The other
classes will most frequently be sampled by ships in the lakes themselves.
Gaining access permission and knowledge of access routes will require
reconnaissance. The amount of time devoted to sampling may be dependent upon
the physical access conditions.
6.3.c. Data Confidentiality
Data confidentiality is an issue of particular concern to EMAP. Many landowners may
be reluctant to permit access from their property because they fear regulatory and
enforcement actions. As with obtaining access, data confidentiality is only a potential
problem with the wetlands component of EMAP - GL. Access is not a design
constraint, and any denials by landowners could affect population estimates. EMAP
6-4
-------
data may have to be aggregated in such a way that individuals cannot be identified to
assure landowners and cooperating agencies that site-specific data will not be used
against their interests. Agreeing to withhold certain information, however, is in direct
conflict with the Freedom of Information Act and EPA's policy on data confidentiality.
This issue will have to be resolved in the near future. The EPA Office of General
Counsel is currently being consulted in this matter.
6.4. Field Operation Scenario
The following field operation scenario is presented to demonstrate that the proposed
field activities are logistically feasible within the allotted time frame. This scenario is
only one of many that could be developed at this time. It is strictly hypothetical and
does not necessarily include all proposed indicator parameters or the order in which
activities would take place when the program is actually implemented. Indicators are
being evaluated and developed for pilot programs; the actual protocols will be
solidified in the future.
6.4.a. General Logistics Scenario
(1) The index periods will be immediately after ice-out in the spring and late
July through September.
(2) The number of Tier 2 sites sampled per year will be approximately 10-20
per lake for the offshore zones and 40-80 per lake for the nearshore
zones. It will take four years to sample all Tier 2 sites.
(3) Site selection is completely random and does not consider site access.
(4) Distance between sites will be one-quarter the density of the Tier 2 sites,
or approximately 150 miles for the offshore sites.
(5) Research vessels will be the primary sampling platform.
(6) Samples requiring immediate laboratory analyses will be shipped to the
appropriate laboratory by overnight courier the day after collection or
analyzed on board.
6.5. Organizational Structure
The long-term success of EMAP is dependent on the development of an
interorganization program with common goals for the monitoring of the ecological
condition of the environment. Great Lakes monitoring will involve numerous agencies
and academic institutions. As EMAP evolves, agreements will have to be established
to define responsibilities. These organizations have highly experienced field
6-5
-------
personnel, and it is anticipated that personnel from these agencies will participate in
field activities, analyses, and interpretation.
6-6
-------
7. Quality Assurance Program
7.1. Introduction
The quality assurance (QA) program for EMAP - GL will be designed to ensure that
the type, amount, and quality of data collected will be in accordance with the data
quality objectives established for the program. Quality assurance programs are
mandated by the EPA for all data acquisition activities which it sponsors or in which it
participates (Stanley and Verner 1985). The EMAP - GL plan is based on ongoing
work and publications by the EMAP - Estuaries and EMAP - Surface Waters resource
groups.
Related to the spatial and temporal scales of implementation of data collection
activities is the number of participating groups. The participation of Federal and State
agencies, contractors, private consultants, analytical laboratories, and scientists from
universities or other research institutions is expected. Existing monitoring programs
considered for integration into the EMAP framework may have QA requirements that
are initially incompatible with those established for the Great Lakes component, or
EMAP as a whole. Differences in sampling and analytical methodology, whether
among participating groups, among regions, or as a result of new technologies over
the life of EMAP, must be monitored and assessed in order to quantify and minimize
their impact on the interpretation of the observed status and trends of ecological
condition.
In developing information of known quality consistent with data quality objectives,
emphasis will be on consistency in implementation, quality control, prompt corrective
action, and continuous improvement. This approach will identify and correct problems
as soon as possible, to minimize their impact on data quality. Appropriate guidance,
training, technical support, and tools (e.g., performance audit materials, quality
assurance documentation) will be provided to all participants to implement QA
programs that are consistent with the data quality requirements of the Great Lakes
component, and EMAP as a whole. While the emphasis of the QA program is to
provide guidance and support, there must also be the means to deal with instances of
poor performance, if necessary, to avoid compromising data quality.
The following subsections outline the general approaches, conceptual rationale, and
guidelines proposed for designing and implementing the overall QA program for
EMAP - GL.
7.1.a. The Data Quality Hierarchy
Data quality exists at several levels. Measurement data quality includes attributes
such as precision, accuracy, representativeness, comparability, and completeness
associated with the measurement of environmental variables. At the next level, it also
includes the uncertainty associated with the methods used to assimilate these
measurement data into an assessment (i.e., provide information from the data). For
7-1
-------
EMAP, this can be perceived as "indicator quality", since the indicators are those tools
used to provide information about specific aspects of ecosystem condition. Factors
affecting quality at this level comprise not only measurement data quality, but also
sampling design and statistical data analysis. At the next level, these indicators are
aggregated into an overall assessment of system condition. The uncertainty
associated with each indicator must be included in an estimate of the overall certainty
of this aggregate assessment. This uncertainty will then be compared to ecosystem-
level quality objectives to assure that data collection and interpretation activities are
consistent with program objectives in terms of the quality of the information provided.
Finally, EMAP intends to integrate information across ecosystems in order to make
regional-scale assessments of ecological condition. Again, the uncertainty associated
with each component of the evaluation (i.e., individual ecosystem assessments) must
be incorporated into an estimate of overall uncertainty for the assessment and
compared to cross-ecosystem quality objectives. The Data Quality Objective (DQO)
process requires that sources of variability be identified at each level and that all
relevant sources be considered in generating estimates of uncertainty at any level of
the hierarchy.
7.1.b. The Role of DQOs in EMAP
The EMAP mission provides the Stage 1 input to initiate the DQO process. The EPA
perceived a need within the Agency and in other client groups for information
regarding the current extent of various ecological resources (i.e., how much of each
resource exists), the current status or condition of those resources, and some
indication of trends in extent and condition over time. The need for environmental
information has been stated in very qualitative terms at this level.
At this point in the process, the tools necessary to measure extent and define
condition within each of these ecosystems are being developed. This, in turn, will
allow policy and decision makers to articulate the requirements for data quality in more
quantitative terms. The program is now in Stage 2, with extensive feedback to Stage
1, and the process may require several iterations.
In Stage 2, each resource group within EMAP must develop a series of indicators that,
in aggregate, allow for an overall assessment of ecosystem condition. Quantitative
"logic statements" must be developed describing the data to be collected for each
indicator and the way in which that data will be used to provide information on system
condition. These statements should include critical values above which the system is
in an acceptable or marginally impacted condition (i.e., nominal) and below which the
system is significantly impacted (i.e., sub-nominal). This critical value must be
scientifically defensible. Where possible, these statements should also relate each
indicator to endpoints of societal interest or concern, so the ramifications of changes in
system condition can be understood and appreciated by a variety of client groups.
In addition to developing these logic statements, a series of error constraints must be
developed that identify all known sources of error or uncertainty associated with the
indicator. These will include measurement error (the difference between sample
7-2
-------
values and in situ true values). Measurement error can be further divided into
analytical error (error associated with the measurement process) and error from other
sources such as sample design, collection, handling, storage, preservation, etc.
Sampling error is a function of natural spatial and temporal variability and sampling
design. Wherever possible, estimates should be provided for each source of
uncertainty. In this way, factors that contribute significantly to the overall variability of
the indicator are identified and the effectiveness of various options in resource
allocation can be evaluated. For example, if spatial variability is the major factor in the
overall uncertainty associated with an indicator and measurement error is small by
comparison, it may be judicious to use a less precise and less costly method of
analysis and invest more resources in increasing the sampling density within a region
to reduce the overall uncertainty in the data.
Early in the program, individual indicators will be used to make discreet assessments
of condition. Tools for making aggregate ecosystem assessments and cross-
ecosystem assessments will be developed over time. The DQO process should
provide the framework for this development, assuring that assessment tools at all
levels provide information of sufficient quality to meet program objectives.
7.2 Data Quality Requirements
In all data collection activities, data quality requirements will be specified in five areas:
precision, bias, comparability, completeness, and representativeness (Stanley and
Verner 1985; Smith et al. 1988). In addition (when appropriate), minimum tolerable
background levels of chemical constituents will be established. These levels represent
the maximum concentrations of constituents that can be contributed by the sample
collection and measurement process. In cases of trace-level analyses, limits of
detection will be monitored and assessed.
Ideally, data quality requirements will be developed based on the overall data quality
objectives. In some cases, the requirements established will be qualitative; in others,
quantitative. The requirements will also be reviewed periodically throughout the
program, and revised as necessary in response to improved capability, additional
knowledge, or technological or resource limitations.
Data quality requirements, constrained by sampling, measurement, or logistical
considerations, will determine the choice of appropriate methodology. Criteria required
for initial selection of appropriate chemical and biological methodologies are presented
in Table 7.1. Choice of these criteria is based primarily on the approach advocated by
Hunt and Wilson (1986), with the addition of criteria relevant to the design of a data
management system.
7-3
-------
Table 7.1 Criteria for Selection of Appropriate Sampling and Analytical (or Measurement) Methodology (based on Hunt
and Wilson 1986).
Criteria
Comments
Range of values of interest
Lowest value of interest (values below this will
probably have uncertainties > 100%); or, in case of
trace-level analyses, the required limit of detection
Maximum tolerable measurement error (random
and systematic)
Standard reference on which method is based
Required frequency of sampling or collection
Collection, analytical, or measurement constraints
Sample handling considerations or measurement
conditions
Reporting requirements
Data reporting time
May need to be tailored for different regions or different projects
Chemical: Used to determine appropriate limit of detection
Biological: May help define the minimum effort required in obtaining
data (e.g., at least 100 organisms in a benthic sample are needed for
determining species composition and relative abundances)
Defined on basis of sample collection and measurement only
Additional error components of interest can be defined in terms of
short-term or long-term (e.g., within-day or within-batch vs. among-day,
among-batch, or among-group)
Should be expressed in both absolute and relative terms:
Absolute: M equal to lowest value of interest
Relative: M some specified percentage of true or most probable value
"known value" = Absolute/Relative; Represents value at which absolute
error equals relative error
Describe any required modifications
Requirements based on sampling plan; used to estimate variance
components of interest
Site selection criteria, special equipment requirements; use of
hazardous reagents, etc.
Appropriate holding times
Operational: May be based on data reporting time requirement
Maximum: If greater than operational, point at which sample is no
longer considered representative of conditions at a time of collection
containers/preservation techniques
Type of variable (numeric coded, character, categorical, etc.)
Reporting units (mg/L, |ieq/L, number of individuals, etc.)
Number of significant figures desired, and maximum number of
decimal places
In addition to being used to select method of appropriate sensitivity,
information is needed to design database, also for collection forms
(hard copy or electronic)
Time period between collection and incorporation of validated data into
database for use in analysis and reporting
7-4
-------
7.2.a. Precision and Bias
Precision and bias are estimates of random and systematic error in a measurement
process (Kirchner 1983, Hunt and Wilson 1986). Collectively, they provide an
estimate of the total error or uncertainty associated with an individual measurement, or
set of measurements. In theory, random and systematic errors can be determined at
any point in a collection and measurement process. Estimates of the various error
components will be accomplished primarily through the use of replicate sampling; such
sampling can be modified to address and control major sources of variability. The
statistical design and sampling plan should act to minimize systematic errors in all
components except measurement error (corneas). Systematic errors in these
components will be minimized by using documented methodologies and standardized
procedures, and evaluated using samples of known composition that can be subjected
to the entire collection and measurement process. Variance components of the
collection and measurement process (e.g., among analytical laboratories or among
individuals identifying biological specimens) should be estimated periodically so that
quality assurance efforts can be allocated to control major sources of error.
The precision and bias requirements will be used to define criteria to monitor collection
and measurement activities and to maintain them in a state of statistical control (i.e.,
the distribution of individual measurements have a stable and predictable over time
(Taylor 1988). Estimates of precision and bias are also necessary to evaluate the
other three data quality indicators (comparability, completeness, and representa-
tiveness).
In general, data from one or more measurements of variables will be combined (and
possibly transformed or categorized) into metrics; one or more metrics will be
incorporated into an indicator, and one or more indicators will be used to provide an
estimate of the ecological health (as "nominal" or "subnominal") of a population.
7.2.b. Comparability
We also need to be cognizant of data comparability external to EMAP - GL,
i.e., 1) with other EMAP ecosystems group (e.g., surface waters, forests, and
wetlands); 2) with other environmental datasets, data from existing monitoring
programs being incorporated or integrated into EMAP - GL; and 3) the comparability of
data collected now to data that will be collected in the future, whether as part of EMAP
itself, or other monitoring efforts that may develop in the future. In these cases,
comparability would need to be evaluated with respect to the QA and QC data
available. The degree of comparability required will depend on the intended use of the
data (trend detection, associative analyses, etc.)
7.2.c. Completeness
For EMAP - GL, the requirements for completeness will be based on the amount of
data required to make conclusions pertinent to the program (or project-specific)
objectives.
7-5
-------
7.2.d. Representativeness
Representativeness is defined as "the degree to which the data accurately and
precisely represent a characteristic of a population parameter, variation of a property,
a process characteristic, or an operational condition" (Stanley and Verner 1985, Smith
et al. 1988). Representativeness can be affected by problems in any or all of all the
other indicators of data quality, as well as by issues such as the location of a sampling
site, the time of sampling, and the statistical selection of sampling sites. More specific
to a quality assurance program is the representativeness of samples or procedures
used to control and assess data quality as compared to the range of conditions being
sampled.
7.2.e. Tolerable Background Levels
Background is operationally defined as the amount of contamination due to collection,
handling, processing, and measurement. It is most relevant to those chemical
constituents present in the environment in very low concentrations. Requirements for
tolerable background limits will be determined based on the lowest concentration of
interest that is required to assure representativeness or completeness requirements
are not compromised. Careful adherence to sample collection, handling, and
processing protocols will minimize background levels. Blank samples of various types
will be used to provide estimates of background levels.
7.3. Organization and Staffing Requirements
Overall responsibility for implementing consistent and adequate quality assurance
programs within EMAP as a whole is the responsibility of EMAP QA coordinator. The
design and implementation of the QA program for the Great Lakes component is the
responsibility of the QA Officer. The QA Officer will be assisted by one or more
coordinators in implementing the large-scale annual sampling operations.
7.4. Quality Assurance Documentation
Prior to the implementation of field sampling operations, a number of different
documents will be prepared (or existing documents utilized) as part of the QA
program. These documents are described in Table 7.2.
Primary guidance for implementing the QA program will be provided by the EMAP
Quality Assurance Program Plan (QAPP) (US EPA 1990b). The policies, organization,
objectives, and functional activities that pertain specifically to the QA program for the
Great Lakes component will be detailed in a Quality Assurance Project Plan (QAPJP).
The Great Lakes QAPJP will be used as guidance in preparing QAPjPs for special
studies, be they regional or local in focus. In general, the QAPjP for any specific data
collection activity will detail the quality control and quality assessment activities
7-6
-------
(summarized in the following subsections) that will be used to ensure the data meet
the data quality requirements established for the project. Additional QA
documentation that may be appropriate in specific instances include guidance
documents or QA program plans for other federal or state agencies and facilities,
provided they meet or exceed the requirements set forth in the EMAP QA Program
Plan and Great Lakes QAPjP.
QA documentation pertaining to EMAP - GL will be reviewed periodically, and revised
as necessary to reflect changes based on previous performance, or other
modifications to either the QA program or to EMAP in general. Changes in various
aspects of the QA program should also be incorporated into revision of standard
operating procedures related to sample collection and measurement.
7.5. Quality Control Guidelines
Quality control is applicable to all stages of a data acquisition process, from design
through sampling and analysis, data management, and interpretative reporting. Each
stage in the process represents a point at which quality control measures can be
implemented (if necessary or desirable to monitor those aspects that are most subject
to error or inconsistency).
Those stages conducted after the commencement of field operations also represent
points where assessments of data quality can be made. In some cases, such
assessments are necessary to monitor sources of error to optimally allocate control
measures among points in the process where they are most needed.
General activities to maximize the success of quality control program include:
(1) documentation of procedures related to design, sampling, measurement,
information management, data analysis, reporting, and quality assurance;
(2) standardized training programs to ensures minimal level of competency in all
aspects of the project; (3) maintenance schedule for all sampling and analytical
equipment and instrumentation; and (4) periodic site visits by knowledgeable members
of the QA or management staffs to ensure that sampling and measurement activities
are being conducted appropriately, to recommend corrective actions as necessary,
and to assist on-site personnel with addressing QA-related issues.
Where appropriate, collection and measurement processes will be monitored through
the use of frequent quality control checks using samples of known composition, or
through replicate measurements. Control charts will be maintained whenever possible.
Use of these tools allows for rapid identifications and resolution of problems related to
sample collection or measurement, and provides documentation that the process is
being maintained in a state of statistical control. Specific examples of quality control
activities are provided in the following subsections.
7-7
-------
Table 7.2 Quality Assurance Related Documentation of EMAP - GL.
EMAP Quality Assurance Program Plan (QAPP)
Describes philosophy and QA policies of EMAP, and provides guidance for designing and
implementing QA programs within EMAP
Great Lakes Quality Assurance Project Plans (QAPjP)
Details the quality control and assessment activities that will be used in the QA program for
Great Lakes
Field Operations Manuals
Standard operating procedures for sample collection, handling, and processing, collection of field
data, and data management activities (including QA and QC procedures). Also describes other
logistical procedures (e.g., sample shipping, waste disposal, communications, safety, etc.)
conducted in the field.
Analytical Methods Manuals
Standard operating procedures for sample analysis (including QA/QC procedures).
Training Plan
Quality Assurance Project Plans from EMAP Support Groups
Information Management
Other QAPPs and appropriate QAPjPs for other participating groups (agencies, laboratories,
principal investigators, etc.)
It is important to recognize that the utility of quality control measurements will be
constrained (especially in the field) by the relatively brief index period each year
(approximately 2.5 months), by the turnaround time between collection of samples and
subsequent analysis (especially for complex analyses such as organic compounds or
fish tissue analyses).
7.5.a. Biological Measurements
When possible, some type of control criteria will be established to ensure an adequate
sampling effort has been conducted at each site to collect a representative index
sample. In cases where this is not feasible, some type of replicate sampling, repeated
measurement, or additional effort sampling program will be conducted at a subset of
sites to provide an estimate of sampling efficiency or precision. Such estimates can
subsequently be used to develop control criteria as the program will be conducted at a
subset of sites to provide an estimate of sampling efficiency or precision. Repeated
sampling and measurement strategies will be designed in conjunction with the sample
replication scenarios presented in Chapter 3. Repeated or independent checks on
sample processing and taxonomic identifications will be conducted on a subset of
7-8
-------
samples collected. Reference collections of biological specimens will be developed
and maintained by participating groups during the course of the program. Such
collections will be eventually archived at a permanent collection facility (e.g., museum)
for future use.
7.5.b. Chemical Measurements
Quality control activities for chemical measurements are well-documented (e.g., Hunt
and Wilson 1986, Taylor 1988). Table 7.3 summarizes these activities. Specialized
collection and handling procedures may be required for certain types of water samples
(e.g., those being analyzed for organic constituents) to minimize contamination and
prevent changes in composition between collection and analysis.
In the laboratory, appropriate types of control samples (and control charts) will be
used to monitor and evaluate statistical control of the analytical process. For inorganic
analyses, at least one check standard (at a concentration near the middle of the
calibration range) will be analyzed periodically with routine samples. Additional
standards may be necessary to determine detection limits for analytes present in low
concentrations. For organic analyses, internal standards may not be available; matrix
spikes or duplicate analyses on a subset of routine samples will be required to monitor
random and systematic errors.
When possible, Standard Reference Materials (SRMs) or Certified Reference Materials
(CRMs) will be used periodically as non-blind samples to assist laboratories in
maintaining statistical control. Such materials will be of most use for analyses of
sediment chemistry, and possibly for some organic analyses, and for analyses of
compounds in fish tissue. Such materials are currently being used by the Near
Coastal component of EMAP, and this program should provide information related to
feasibility, cost, and preliminary performance data that can be utilized in the QA
program for the Great Lakes. It would be advantageous to subject such reference
samples to the entire collection and measurement process, rather than just to the
analytical phase. This would assist in monitoring potential errors (random and
systematic) associated with sample collection and field processing, the feasibility of
implementing such an approach as a quality control tool will be investigated as part of
the QA program for the Great Lakes.
7.5.C. Habitat Quality and Site Characterization Measurements
Quality control activities associated with the landscape characterization measurements
being conducted in support of the EMAP Great Lakes component will be documented
in a separate quality assurance plan. For those measurements being collected as part
of the EMAP Great Lakes effort, the most critical quality control activities (once
standardized methods are implemented) are the development and use of standardized
codes and categories. For measurements collected from maps, an independent check
of the measurements conducted periodically by a second person (or group) would
serve to detect and correct errors on a timely basis. For data being collected during a
site visit, proper calibration of instruments (e.g., calibration of an electronic depth
7-9
-------
Table 7.3 Quality Control Activities Associated with Chemical Measurements.
Field:
Calibration checks of instrumentation using independent standards of similar composition to environmental samples. Use
of control charts to monitor performance.
Preventative maintenance program for all equipment and instrumentation.
Standardized procedures for collecting, handling, and processing samples. Procedures to minimize potential of
contamination during collection, handling, or processing.
Proper preservation and labeling of samples.
On-site review of recorded data and other information.
Periodic use of field blank samples and audit materia to check for effect of collection and processing.
Laboratory:
Standardized procedures for preparation, calibration, and analysis.
Preventative maintenance program for analytical instrumentation.
Routine use of control samples (blanks, check standards, matrix spikes, etc.) and control charts to monitor statistical
control of analytical process.
Periodic use of reference material (Standard Reference Materials, Certified Reference Materials) or other sample of
known composition as internal standards to check for systematic errors in analysis.
Review of analytical data immediately after analysis and before entry into database.
finder against a calibrated sounding line), and repeated measurements by a second
person on a subset of sites would be used as the primary means of minimizing errors.
Such repeated measurements also provide estimates of the magnitude of
measurement errors.
It would be desirable to implement methods to monitor for systematic errors in
collecting these data (whether they result from a particular method or from different
crews that utilize a method).
7.5.d. Samples and Specimens
Archival activities will involve samples for chemical analyses, the curation of biological
specimens (fish, invertebrates, and slides of diatoms from sediment cores), and
validated databases. For chemical samples, samples of water and sediment will
generally be archived during a particular index period in case some type of reanalysis
is warranted. Such samples (or a subset) may be preserved and archived for longer
periods to permit future analyses of constituents other than those initially determined.
An example might include more detailed analyses of samples when the results of
bioassay experiments indicate possible toxicity.
7-10
-------
Such long-term storage may be feasible for some inorganic constituents, but may not
be feasible for organic compounds. Samples of fish tissue may be placed in a
specimen bank, such as that established by the National Institute of Standards and
Technology (NIST) for possible analyses in the future.
Voucher specimens of fish and invertebrates will be collected as part of the routine
quality control program. Periodically, such specimens will be placed into a permanent
collection. Possible options for curation include the establishment of a specimen
banking and curation system specifically for EMAP, or to make arrangements with
regional facilities (e.g., national museums, university museums, or state biological
survey agencies) to incorporate specimens collected as part of EMAP into permanent
collections. The archival of specimens would be reported in the appropriate summary
or interpretative reports.
7.6. Data Review, Verification, and Validation
This aspect of the QA program overlaps with the QA program that will be established
for the information management program (Chapter 8). Operationally, QA for Great
Lakes information management can be considered as two separate elements. One
element, data review, verification, and validation ensures that all information which is
ultimately entered into a database is accurate. The second component, which is
addressed in Chapter 8, deals with maintaining the security and integrity of validated
databases once they have been archived and made available for use in data analysis,
reporting, or distribution to users outside of the Great Lakes component, or outside of
EMAP. Of primary concern here is the prevention of deletion, alteration, or
irretrievability of information stored in databases.
The general approach in minimizing data-related errors prior to archival will be to
emphasize the review of information at the point of collection or measurement as soon
as possible after the sample or measurement has been obtained. Where feasible,
data recording will be done electronically, with standardized recording forms being
used as backups. In the field, data logging devices (hand held computers that display
screens similar to manual field forms) are being tested by other EMAP task groups.
Use of data logging devices would reduce the time required for data entry, and will
automatically check for erroneous data as it is entered (e.g., range checks on numeric
data, misspellings, or invalid codes). Similar types of devices may also be utilized for
laboratory measurements.
The review process will be automated to the extent possible, but not to the exclusion
of a manual review by qualified and knowledgeable people. In order to complete the
review, verification, and validation process as quickly as possible, a substantial
investment in sources and personnel required for data entry and data review and
verification will be required.
Data review will initially involve a check of raw data (e.g., what is recorded on data
forms) before entry into an electronic database. In the field, forms should be reviewed
7-11
-------
by a second person before leaving a site. A subsequent review should occur
immediately after data entry by comparing the entered data to the raw data. Data
from quality control samples or measurements, and reviewing control charts, can be
used to determine if reanalysis (or remeasurement) is required before data are
entered.
The data verification process is largely an automated one, and serves to check that
entered values are correct and (when possible) internally consistent within a sample
(or set of measurements). Examples of verification procedures include range checks,
checks for duplicate entries, frequency checks of coded variables to identify
inappropriate codes, and format checks to ensure data has been entered in the
correct format. Quality control data from check samples, blacks, or replicate samples,
or use of redundant measurements for critical parameters (e.g., field versus laboratory
pH) can be used to determine if there are problems with sample collection or
measurement. Internal consistency checks include ion balance and conductivity
calculations for inorganic chemical constituents. For biological samples and
measurements, internal consistency checks include summing species proportions in
samples to ensure they do not total to more than 100 percent, checks for missing
taxa, evidence of "container effects" in bioassay experiments, and the taxonomic
accuracy of species identifications.
Once the entered data are verified as being accurate, they are validated by
examination against regional expectations to identify and explain outlier samples or
sites. Validation may involve comparison with historical data, or through the use of
association and multivariate analyses.
7.7. Assessment of Data Quality
The assessment of data quality for the Great Lakes component of EMAP will occur
within a lake and among sampling cycles. Qualitative assessments include
documenting methods, using a sampling design that ensures unbiased and
representative samples, and site visits to ensure consistency among participating
groups. Quantitative assessment will attempt to estimate errors associated with
sample collection and measurement that are important in either the interpretation of
indicators, or to optimize the QA program through time (to adjust the effort and
intensity of quality control to areas where it is needed the most).
The primary means of assessing error and uncertainty will be through carefully
designed performance evaluation studies. These studies will test hypotheses related
to data quality requirements for random and systematic errors.
The design will be based on consideration of Type I and Type II errors, and will
attempt to provide estimates of: (1) total measurement error for use in data
interpretation activities; and (2) important components of variance within measurement
error that can be used to determine which steps in the collection and measurement
process require more (or less) quality control emphasis. The sample sizes and
7-12
-------
frequency of measurement will be optimized to provide the necessary answers in the
required reporting period.
Performance evaluation studies could be conducted using performance audit samples
for chemical analyses, reference samples for biological measurements, and "round-
robin" studies using natural samples (either chemical or biological). For water column
chemistry, appropriate performance audit materials are available for most inorganic
constituents. Depending on the constituent, appropriate materials for organic analyses
may or may not be currently available. Materials for sediment chemistry and fish
tissue chemistry may be limited in their availability and appropriateness.
As mentioned previously, an important issue in the program is the impact that using
different methodologies, or modifying or changing methodologies over time, will have
on data interpretation, particularly in the detection of trends in ecological condition.
The QA program for the Great Lakes component will provide standard guidelines for
implementing a new methodology. Performance evaluation studies will provide some
information on methods comparability, but comparability studies should be a more
intensive effort designed to test specific hypotheses related to the comparability with a
previous methodology. Such a comparability study must be conducted, evaluated,
and approved before new or modified methods can be implemented.
7.8. Quality Assurance Reporting
In addition to the documentation described in Chapter 2, other types of reports will be
produced periodically as part of the QA program. These include: (1) summary reports
of site visits and audits; (2) performance evaluation (or method comparisons)
summaries; and (3) assessments of data quality. Summary reports of site visits will
serve to identify and track issues and subsequent corrective actions, and provide
information to update other QA documentation.
The results of performance evaluation studies will be reviewed and returned to
participants within a short time after submission. Evaluation summaries of QA-related
data and other appropriate information will be prepared and included in the appropriate
Great Lakes reports.
7-13
-------
8. Information Management
8.1. Overview of EMAP - GL Information Management
During the development and implementation of the EMAP - GL program a great
volume of raw data of diverse origins will be collected, analyzed, and reported. The
resulting information1 will be utilized to evaluate the current status, extent, changes,
and trends in the condition of Great Lakes ecological resources. The capability
provided by an EMAP- Information Management System (IMS) to manage and
disseminate such information in a timely and cost-effective manner will be a major
determinant of the success of the program. The IMS will provide the information
management functions of data collection, validation and verification, storage, analysis,
retrieval and reporting, and archiving. This section describes the information
management objectives of EMAP - GL and the operational approach to be used to
accomplish those objectives.
The EMAP - GL program will utilize historical datasets generated by other monitoring
programs currently or previously operational. Such datasets will contain usable
information in differing formats, and in various states of quality assurance. The
EMAP - GL IMS will have the capability to quality assure and incorporate historical
datasets into standard dataset formats.
EMAP - GL data will be summarized, compared, aggregated, and reported, using a
variety of statistical and spatial analysis techniques, to satisfy multiple levels of EMAP
users. Other EMAP resource groups, federal, state, and local agencies, academic
institutions, and other researchers will need timely access to EMAP - GL data and
information. The EMAP user community is illustrated in Figure 8.1. One established
goal of EMAP - GL is the production of statistical summaries within nine months after
the completion of data collection. The EMAP - GL IMS will have the capability to
process a variety of analyses and produce timely outputs.
The EMAP - GL IMS will integrate with and utilize the extensive datasets generated by
different monitoring programs within the Great Lakes region. The IMS must integrate
the data processing activities which simultaneously occur, within different sampling
programs, over broad geographic scales. An effective project management system is
a necessary component of the IMS.
1 The relationship of information to data may be summarized by the following 'processing'
function
information = f(data)
processing
which simply states that processing converts data into information in a defined,
consistent manner.
8-1
-------
The ultimate goal of EMAP - GL data management is the timely and cost-effective
production of pertinent quality-assured data readily accessible to a broad-based,
scientific user community.
EPA Administrator
Congress
Other Agencies
and
Scientific
Community
Environmental
Monitoring
and
Assessment
Program Results
EPA Regions
EPA Program
Offices and
States
Public, Environmental
Groups, and Industry
EPA Office of
Research and
Development
Figure 8.1 EMAP user community.
8.2. Objectives of EMAP - GL Information Management
The overall objectives of EMAP - GL information management are to:
provide state-of-technology IMS design and development within the
guidelines of the Office of Information and Resource Management (OIRM),
the EMAP Information Management Committee (IMC), and other applicable
guidelines;
incorporate recognized, and stated, agency, program, and task group
standards of Quality Assurance and Documentation (including a Data
Management Plan) into all phases of IMS design and development, and into
data management activities within the EMAP - GL program;
provide for the wide-spread availability of EMAP - GL data within the
EMAP - Great Lakes Resource Group user community and incorporate, into
IMS design, data interoperability on a program scale;
8-2
-------
produce and disseminate compilations of EMAP - GL information, both
periodic and special interest, in a timely manner and, specifically, produce
annual statistical summary reports within nine months after completion of
data collection;
ensure the statistically sound integration of all usable data from on-going
programs and historical data sources, from sources within, and external to,
the EMAP - GL program and from all appropriate levels of data aggregation;
incorporate an adaptability to evolving program needs, and to the state of
technology, into the IMS design, and into the EMAP - GL information
management perspective; and
develop and maintain a cost-effective and efficient IMS, responsive to
defined user requirements, utilizing to the extent possible, existing
resources.
8.3. Mission Needs Analysis
Information Resources Management policy specifies Mission Needs Analysis as the
first phase of the system development process. The IRM Mission Needs Analysis
results in the development of three products: a matrix of potential system users and
information uses, a process flow diagram, and an Initial System Concept.
8.3.a. System Users and Information Use
The definition of EMAP - GL system users is embedded within a program definition of
EMAP users and within an IMC definition of users groups, defined in terms of their
data requirements.
The EPA OMMSQA overview (EPA 1990b) defines the following user classes:
"The program will serve a wide spectrum of users:
decision-makers who require information to set environmental policy;
program managers who must assign priorities to research and monitoring
projects;
scientists who desire a broader understanding of ecosystems; and
managers and analysts who require an objective basis for evaluating the
effectiveness of the Nation's environmental policies."
8-3
-------
Also identified are 'Congress, the EPA Administrator, private environmental
organizations, The EPA Science Advisory Board, the public, other agencies,
Regional offices, states, and the international community'.
The IMC has produced a general classification of users based upon their function
within the EMAP program and the EMAP data category they need to access. The
following five classes are defined:
(1) EMAP - GL Core Research Group: Includes those individuals and groups
charged with designing and implementing the IMS, and interpreting the data
from the field sampling programs.
Requirements - The EMAP - GL Core Research Group will need access to
sampling measurement data, raw, verified, and validated, in all stages of
summarization and aggregation. This group will also need access to sampling
support data concerned with sample and shipment tracking, logistics, quality
assurance/quality control, and project management, in various media;
electronic, reports, maps.
This group also requires access to the data on as close to a real time basis as
possible. Raw data used by this group will not be quality assured. This group
needs access to all data described in the other categories.
(2) EMAP - GL Team - This group includes individuals and groups involved in
the EMAP - GL effort, but are not involved in the day-to-day field operations.
These participants include outside participating agencies, logistical support
personnel, GL QA/QC personnel, program reviewers, and EPA Headquarters
personnel.
Requirements - This group requires access to summary information regarding
logistics and project management as well as validated data. They will require
access to only those files that have validated and verified. They do not require
real time access, nor do they need to have access to a comprehensive data
set.
(3) EMAP Program - Includes all researchers directly involved in the design,
implementation, and analyses of the national EMAP program. These individuals
include members of other task groups, members of the Synthesis and
Integration Team, and personnel in other agencies directly involved in EMAP.
Requirements - This group requires final summaries regarding logistics and
project management. They will require access to only those files that have
been validated and verified. They do not require real time access, nor do they
need to have access to a comprehensive dataset. They need data in a context
which can be integrated with data from other disciplines. Document summaries
with interpretation and graphic outputs will be most useful.
8-4
-------
(4) Legislators and Environmental Managers -
Requirements - This group will need summarized and interpreted data. They
will not require any data regarding logistics or project management. This group
of users will be best served by published reports, maps and an on-line
summary system.
(5) The General Public -
Requirements - This group will need summarized and interpreted data. They
will not require any data regarding logistics or project management. This group
of users requires published reports, maps, and an on-line summary system.
The classes of EMAP users and data are illustrated in Figure 8.2.
Increasing
amount of
data
'EMAP '
Information
Center
Task Group
Information
Center
Regional
Information
Center
Public Officials, Environmental
Managers and the Public
Task Group Senior Scientists
EMAP Assessment,
and Environmental Decision
Officials
Regional Assessment
Regulators and
Scientists
Increasing
number
of users
Relationship of classes of EMAP users and data.
Figure 8.2 Relationship of classes of EMAP users and data.
8-5
-------
8.3.b. Process Flow Diagram
As utilized in IRM's Mission Needs Analysis, Process Flow Diagrams depict the
interconnected flow of information work processes from input origins to output
products, 'plotted' against a vertical axis of involved system organizations or
individuals. This document is currently being developed by EMAP - GL
8.3.c. Initial System Concept
The Initial System Concept is the end product of IRM's Mission Needs Analysis. It is
a concise depiction of inputs, processes, and outputs. This document is currently
being developed by EMAP - GL.
8.4. EMAP - GL Processing Environment
The EMAP - GL program is embedded within a national EMAP program which,
ultimately, will integrate the IMSs of all participating Task Groups, into a single
decentralized EMAP IMS, capable of providing complete national data interoperability.
The EMAP - GL program is also a member of an international community of Great
Lakes monitoring agencies. Finally, the EMAP - GL program will function
autonomously on a Task Group level, developing and maintaining an independent
IMS. The IMS will be a distributed information management system consisting of a
central EMAP processing node and remote information centers which will function as
regional data coordination centers for field collection and laboratory processing data.
Remote information center sites will be located at existing computing facilities proximal
to the region of data collection. The EMAP - GL information management program
will consist of the following components:
a distributed IMS resident, utilizing an Agency-standard hardware/software
configuration, on a central EMAP node and multiple remote processing
nodes;
developmental and operational personnel; and
policies and Standard Operating Procedures (SOPs).
The above components should not be considered static entities, capable of being
defined once for the lifetime of the Great Lakes EMAP program. An effective Great
Lakes EMAP program will result only if the above-listed program components are
allowed to remain in a continuing state of adjustment to the forces of evolving EMAP
user needs and state of IMS technology.
8-6
-------
8.4.a. Distributed EMAP - GL IMS
A hybrid multi-platform system has been proposed which links the EMAP - GL IMS as
a node in a Wide Area Network (WAN) of Task Group Information Centers and
Research Triangle Park (RTP). The proposed IMS will realize of the concepts of
system decentralization and data interoperability evolving in the EMAP program. The
EMAP - GL node will provide database management, IMS development, GIS
capabilities, and communication. To the extent possible, the proposed configuration
will utilize existing EPA computing resources. All additional hardware and software
resources needed in the course of IMS development will be acquired by approved
Agency procurement. The proposed node configuration and networking, local and
internode, is diagrammed in Figure 8.3.
The hardware configuration and suite of software utilized by EMAP - GL is determined
by agency and program standards. Hardware and software needs will be defined by
an IRM-required Essential Elements of Information document, EEI-2, " Preliminary
Design and Option Analysis". Currently, all database development and statistical
analysis is performed using the Statistical Analysis System (SAS). All GIS analysis
will be performed using ARC/INFO. The use of a Relational Database Management
System (RDBMS) to facilitate IMS development is proposed. Should a RDBMS be
adopted to serve IMS database functions, SAS will continue to be utilized as the
statistical analysis tool. The use of Computer Aided Software Engineering (CASE)
tools is also proposed as a means of standardizing development.
EMAP - GL will utilize existing compatible hardware configurations at remote node
processing sites, while software and peripherals will be Agency-approved.
8.4.b. Developmental and Operational Personnel
Developmental personnel requirements of an EMAP - GL information center can be
fairly well defined in consideration of immediate needs. The work efforts necessary to
develop and support an IMS integrated with multiple Great Lakes agencies, however,
can be only roughly estimated. The following personnel list is, accordingly,
preliminary.
Great Lakes Task Group Information Manager (TGIM): The TGIM is
responsible for planning, coordinating, and facilitating information
management activities within the task group. The TGIM serves as the
liaison to the EMAP Information Center and other EMAP resource and task
groups.
Great Lakes Database Manager (DBM): The DBM is responsible for the
design, development, and administration of a Great Lakes IMS that best
meets the evolving needs of EMAP users, is fully interoperable with other
EMAP IMSs, and adheres to applicable Information Management Committee
(IMC) and agency standards.
8-7
-------
CD
&
2%
O(0
งz-
^_
CD
C
0
O
c
0
cS
L_
CD
fc
0
CO
^>
c
o
rn ซ
C/J **
~~ CO
>ซ/ Jj^
o
5
i i V 4
i 1 V
* 1 A
i t /\
LU
(0
0
*
CO
O
CD
CO
O ^
9-0
ZJ CO
o- S
1 8
S*
Q
O
CO ฃ-
C 0
CD ป
E S
. Q.CO
5 ^
ง o
CD ง
Q >
UU_I
8-8
-------
Programmer/Analyst: The Programmer/Analyst provides specified systems
analysis, application programming, testing, documentation, and user
support.
Geographic Information System (GIS) Specialist: The GIS Specialist is
responsible for the development of descriptive and analytical GIS functions.
Data Clerk/Librarian: The responsibilities of the Data Clerk/Librarian include
data entry and maintenance, the documentation of Great Lakes datasets of
various origins, processing approved data requests and transfers, and
monitoring and reporting the "state of the database".
Remote information processing nodes will utilize computing facility staff.
8.4.C. Policies, Standards, and Standard Operating Procedures
Information management at EMAP - GL is embedded within a hierarchy of policy
levels, some well-defined, others in a nebulous state. The acquisition, development,
and management of all agency information technology is governed by an umbrella of
OIRM policies (US EPA 1987). Chapter 4 of the IRM Policy Manual, "Software
Management", establishes the Agency Software Management Program, one section
(4.5.h) of which states:
"The development of all application systems will conform to the Agency's
system development life cycle methodology."
The life cycle methodology was issued under a separate cover, in three volumes (US
EPA 1989). The objective of this document is "to provide guidance, assistance, and
only when necessary, controls" in the design, development, maintenance of application
systems. The EMAP - GL program is governed by IRM policy, and the IMS will be
developed in accordance with the requirements established by the guidance
document.
While an EMAP IMC information management policy document does not yet exist, the
prospect of a decentralized national EMAP system with data interoperability strongly
supports the establishment of, at least, some standards (e.g., data structure, variable
naming). EMAP - GL will fully cooperate in the establishment of inter-Task Group
standards and conventions that promote the national sharing of EMAP information.
Standards dealing with VAX system security and data confidentiality are an inherent
part of any ERL-Duluth laboratory system development. These standards are
consistent with those of other ORD laboratories. Also, ERL-Duluth is defining
Automated Data Processing standards and policies, beginning with Information System
Quality Assurance, to be applied to all IMS development.
Policies and SOPs will be developed to guide and govern all aspects of information
management for EMAP - GL and its remote information centers. The policies and
8-9
-------
SOPs formulated will be consistent, in intent and content, with those policies, outlined
previously, which apply to the EMAP - GL program.
Some areas of future policy development are:
data confidentiality issues involving the sometimes conflicting objectives of
promoting the widespread use of collected data and assuring the
confidentiality of proprietary data;
data accessibility issues arising from the need to assure EMAP - GL users
ready access to desired data and information, while maintaining system
security and data integrity;
data processing responsibilities and restrictions in remote information
centers;
QA/QC expectations and responsibilities of the various IMS functions within
the EMAP - GL IMS; and
issues involving the interagency and international cooperation in the
collection, processing, and dissemination of data.
The necessary SOP documents will be formulated to serve as a guide to the discrete
operations involved in the collection, verification, validation, analysis, and aggregation
of EMAP data and information. Each SOP should define the objectives of the SOP,
describe the data, operations, and methodologies involved, define pertinent criteria,
and establish contingencies. All EMAP - GL SOPs will be permanently stored in a
readily accessible file.
8.5. Operational Components
This section describes the EMAP - GL IMS at the operational component level. The
systems described are in a planning and design phase at the time of writing; more
detailed descriptions of the operational components of EMAP - GL IMS will be
published as the program develops. The following major operational components can
be identified at this time: Sample Collection System; Sample Tracking System;
Logistics Information System; Historical Data Processing; Indicator Development Data
Management; Data Retrieval System; Data Reporting System; Data Dictionary System
and Documentation, Data Archival, Geographic Information System (GIS) Applications,
Quality Assurance; and Project Management.
8.5.a. Sample Collection System
The EMAP - GL field sampling program will concurrently operate a computerized field
data entry system and a manual field data entry system in order to evaluate the cost
and effectiveness of both types of system. When EMAP - GL is fully operational,
8-10
-------
Portable Data Recorders (PDRs) will be used with the field computer system to enter
field data directly into field computer files, with computer entry screens duplicating the
manual entry field forms. PDR entry screens will have integral data verification
processing (e.g. numeric range checks, code field checks, duplicate entry verification).
Sampling measurements, as well as site information, will be stored in sample
measurements files, linked to sampling event and station files. A hierarchical (station,
sampling event, sample measurements) spatial and temporal indexing is created which
uniquely identifies any, and all, sample parameters. The use of a unique sampling
index assures non-duplication of sample records. Hierarchical indexed samples will be
correlated with an independent sample numbering system, which allows samples to be
tracked within the Sample Tracking System. If resources permit, bar code readers will
be used to improve the efficiency and accuracy of the sample number entry and
identification. Bar coding assures an objective anonymity of sampling station and
facilitates sample processing checks. The use of pre-numbered sample labels for
sample containers and data recording media will reduce incorrect sample identification.
Field data SOPs will provide a guide for the proper collection and entry of field data.
All field data will be archived on a periodic basis. Initially, a backup manual field
system will be retained should the computerized field system fail.
The Sample Collection System will be interfaced or, in some instances, integrated
with, the existing sample collection programs of those Great Lakes monitoring
programs cooperating with the Great Lakes EMAP program.
8.5.b. Sample Tracking System
A critical component of the IMS will be a Sample Tracking System which uniquely
identifies the three levels of field sampling structure (sample stations, sampling events,
and individual samples taken), interrelates data from these hierarchical levels, and
tracks individual samples from collection through all processing and analyses. The
Sample Tracking System will link Field Measurements Files, Lab Measurements Files
of Lab Analysis Systems, Station, Event, and Shipment files by means of a Sample
Status file. Simplistically, the Sample Tracking System is a set of files indexed by
variables cross-referenced to the Sample Status file. The Sample Status File is,
therefore, a file of cross references to other data files, presently including:
Station Id: Index of the Station file, links a sample to one station (may be
implicitly defined by the Event Id);
Event Id: Index of the Event file, links a sample to one sampling event, at
one station;
Sample Id: The index of the Sample Status file, this mnemonic variable
must contain ail necessary components to uniquely designate a given
sample. The necessary components vary, but normally includes spatial and
temporal components, as well as components to specify the type of sample
or member of a sample series;
8-11
-------
Date: A YYMMDD field that may or may not be a part of the Sample Id;
Ship Id: The index of the Shipment file, links a sample to one shipment
which may include a multiple events or stations;
Field Id: The index of a Field Data file (including a file identifier, if multiple
field data files exist), links a sample to field measurement(s);
Data Id: The index links sample data to a particular media location where
the data resides. Because various types of sample data exist (e.g., raw,
textual, QA), more than one type of Data Id may be necessary;
Team Id: The index links the sample to a Data Collection Team file,
identifying personnel, equipment, and other pertinent collection information;
and
Process Id: The index links the sample to a Processing Plan file, which
contains processing and analyses specifications. The index will operate with
an associated field containing the current processing status.
From the above descriptions it is obvious that the Sample Status file is the hub of
access to all sample information. This architecture provides for growth of the Sample
Tracking System. Additional types of sample information may be added to the system
simply by adding another spoke (index variable) linking the hub to the data file
containing the new type of sample data.
The Sample Tracking System should accomplish the following functions:
identify, uniquely, by Sample Id, all field samples, linking field samples to
sampling events, (and, thence, to sampling stations), and to sample
shipments, and to all subsequent sample processing steps;
report the status and location of any given field sample;
summarize the processing status of a specified group (e.g. station set, bar
code sequence, indicator dataset) of samples in report format;
display the processing status of groups of samples, by station, using an
interface to the GIS system;
automatically report incomplete sample sets at processing points (e.g.,
receipt of samples for lab analysis); and
Provide sample anonymity to all subsequent sample analyses, thereby
facilitating processing checks, using duplicate samples.
8-12
-------
Bar coding of sample collections will be used, if economically feasible, to facilitate use
of PDRs and a sample entry system by sampling crews. A bar code sequencing
scheme will be used; Sample Id values may be used directly or blocked ranges of
cross-referenced values to collection of data. Samples will bar coded prior to
sampling.
8.5.c. Field Logistics System
A Field Logistics System will be developed to provide support information for field
sampling. Indexed by Station Id, the system will supply critical resource information to
sampling crews. The location and means of contact for transportation facilities,
supplies and equipment locations, medical facilities, communication centers, fuel
suppliers, and other important resources proximal to sampling stations will be
referenced in the Logistics database. An Itinerary database will store specific planned
and actual sampling trip information. A Boat/Crew file will contain information on
personnel and equipment utilized; this file can be linked to the Itinerary database by a
Trip Index.
The Field Logistics System will provide, at least, the following:
an itinerary of sampling sites to be visited, by a specified boat and crew,
with a sampling event schedule for each station;
a log of actual sampling activities and sampling trip information, such as
resource utilization (and cost);
a report of critical resources (described previously) available, proximal to
each sampling station, for the entire sampling trip; and
summary reports, and displays using GIS, of sampling activities such as,
cost and percent success of sampling trips or types of sampling.
The Field Logistics System will be interfaced with the Sample Collection and Sample
Tracking Systems.
The EMAP - GL program will utilize sampling data collected by other on-going
monitoring programs, which have Logistic Systems supporting their sample collection.
Every effort will be made to utilize and interface with those Logistic Systems currently
functional.
8.5.d. External Dataset Processing System
Data useful to EMAP - GL will often reside in datasets other than those generated by
EMAP-initiated sampling programs. While important to the EMAP - GL program, such
data sources require special quality assurance and data confidentiality considerations,
and often special processing. External data sources that may be utilized by EMAP -
GL include:
8-13
-------
sampling programs from other agencies, at all levels of government (e.g.,
federal, state, county, municipal);
sampling programs conducted by institutions, private industry, and
environmental organizations; and
sampling programs carried out by functions in foreign governments, by
institutions, industry, or organizations in other countries, or by multinational
organizations.
The above-mentioned sampling programs may be currently active or they may have
concluded in the past.
Data contained in the datasets may be at various levels of aggregation and various
stages of quality assurance.
Beyond containing data valuable to Great Lakes resource assessment, external
datasets must satisfy criteria which determine their functionality within the EMAP - GL
IMS. The necessary SOP documents establishing criteria for inclusion of external
datasets into IMS will be formulated. Some of the fundamental questions to be
addressed by these SOPs are:
Is the dataset acquirable (or accessible), either through the dataset's owner
or administrator? Do data confidentiality issues restrict the acquisition of
subsets of interest of the dataset?
Is the dataset of EMAP-compatible (or of readily convertible or
massageable) format and level of aggregation?
Was the dataset produced by sampling design protocol comparable with that
of EMAP - GL sampling protocol?
What restrictions are placed on the use and distribution of data, in the
dataset, or on information generated from the data?
If the dataset is active, will EMAP - GL be assured of access to future
dataset updates?
While perhaps not a standard SOP criteria the cost-effectiveness of the
acquisition of external datasets and their future updates will be a primary
factor in their use.
To assure the efficient incorporation of applicable data into the IMS, the External
Dataset Processing System must perform the following functions:
conduct a comprehensive search of current and past Great Lakes sampling
programs, producing a list of datasets (and their administrators) whose data
8-14
-------
content is potentially applicable to the development of defined EMAP - GL
indicators;
evaluate the list of potentially usable datasets in terms of defined criteria;
appropriate efforts will be made to acquire the qualified datasets;
for each acquired dataset: formulate quality assurance, and if necessary,
data conversion SOP(s); archive original dataset; qualify original dataset to
produce a subset of potential EMAP - GL data; quality assure potential
EMAP - GL subset; and, if necessary, convert the quality-assured dataset to
EMAP - GL standard format datasets;
integrate the resulting converted dataset into the IMS. The processing
provides an interface with the Indicator Development Management System
and is highly variable; and
thoroughly document, for each dataset, all processing from acquisition to
conversion to IMS standard dataset format and produce archives of all
important stages in the dataset incorporation not readily reproducible. All
incorporated datasets will be automatically updated in the Data Inventory
System.
Segregated work areas for each stage of dataset conversion (e.g., original data,
qualified data, quality assured data) will be established for all datasets. Access to
external datasets resident in the work areas will be determined by the data
confidentiality policies of EMAP - GL and the dataset's administrator.
8.5.e. Indicator Development Data Management System
Initially, all processed raw data files, field and lab, as well as all historical data files will
be stored in SAS data sets. The acquisition and use of a Relational Data Base
Management System (RDBMS) is anticipated; at that time SAS data sets will be
converted to the RDBMS format. Statistical analyses of data will continue using SAS.
The field/lab and historical SAS data sets comprise the entry data sets of the Indicator
Development Data Management System (IDDMS). Both field/lab and historical data
sets will have been extensively quality assured using data verification and validation
SOPs. Indicator data sets will be generated from IDDMS entry data sets by analysis
and aggregation and further analysis of indicator data sets will result in overall
assessments of resource conditions. From the information perspective, this process,
the IDDMS, will be accomplished by SOP-mediated processing operating upon data
structures.
The IDDMS data sets will be segregated into indicator libraries and each data set
development thoroughly documented. Data archives of the stages of the quality
assurance process will be produced.
8-15
-------
8.5.f. Data Dictionary System and Documentation
An inventory system of data libraries, files, and elements provides important
information about the nature, location, and inter-relationships of data elements in the
EMAP - GL IMS and is essential to system maintenance and documentation.
The core of the DOS will consist of two data sets, developed and maintained in SAS;
a Data Set Index, and a Data Dictionary. The two data sets are hierarchical and are
linked by the data set index variable.
The Data Set Index will contain one record for each data set in the EMAP-GL IMS and
will include, at least, the following information:
a general description of the data set information content;
the purpose of the data set, the indicator(s) development with which it is
associated;
information on the location of the data set and how it may be accessed;
information on humans associated with the data set; who is responsible for
the data set, its origin, scientist(s) currently working with the data set. This
data set can be linked to an indexed Relevant Humans Data Set;
information on hierarchical relationships with other data sets; and
information on the quality of the data contents, restrictions, and data
confidentiality.
The Data Dictionary will contain one record per data element in the EMAP - GL IMS.
Each record will be related to a Data Set Index data set by a data set index variable.
The cardinality of the relationship of Data Set Index records to Data Dictionary records
is one-to-many. The Data Dictionary will contain, at least, the following information:
a description of the data variable (element);
the data type and format;
the range of valid values, if numeric, or the set of valid values, if
alphanumeric; and
the location (data set and library) and distribution of the variable (other data
sets in which it is located).
The DDS will have two main applications, which require it to be accessible to users of
a wide range of sophistication:
8-16
-------
System documentation, interactive with system configuration maintenance:
EMAP - GL core staff responsible for modifications to data structures in the
IMS must have current data inventory information.
IMS Information Content reports: A variety of potential EMAP-GL users will
need current listings, by categories of information, to construct retrievals
from EMAP - GL IMS.
The need for current, reliable data set index and data dictionary information requires
that any modifications to the GLIMS data structures generate an instantaneous system
update of the Data Inventory System. All structural modifications will be made within
an interactive system utility which automatically updates the data dictionary (a possible
CASE application).
The EMAP - GL IMS will ultimately be data interoperable with other Task Grouo IMSs
and may be part of a decentralized national EMAP system. It is, therefore, mandatory
that the current state of the DOS be available to all prospective users. Furthermore,
data interoperability implies at least some variable conventions and standards be
adopted. To that end, system configuration maintenance will be interactive with a
Standards file and review function.
The development of the EMAP - GL Data Dictionary System (DOS) must be
considered in context of EMAP IMC Dictionary/Catalog/Directory (D/C/D) Work
Group's efforts to develop a program-wide dictionary system. Also, the adoption of a
RDBMS may involve considerations of any integral or active data dictionary the
management system may contain.
Complete system design and development documentation is mandated by IRM's data
management policy, specifically the generation of the set of Essential Elements of
Information (EEI) documents over the life cycle of the EMAP - GL IMS. The EEI
documents address mission needs, design, procedural, managerial, and operational
documentation needs. Additional documentation (e.g. data and processing SOPs,
data set profiles, user logs) will be developed and maintained.
8.5.g. Data Archival System
The capabilities of the Data Archival System are determined by IMS policy (at all
levels) assuring data integrity. All data archival will be performed in accordance with
data archiving SOPs. All data archiving should be performed from within the IMS and
as automated as possible. For instance, if processing creates a new dataset, which
requires archiving, that processing should also automatically create an archive of the
new dataset.
The Data Archival System will be capable of archiving both data and procedural files.
Specific files to be archived will be established by Archival SOPs. Some examples of
files which will be archived are:
8-17
-------
raw data files from field sampling or lab analysis;
original historical datasets;
quality assured (verified and validated) datasets;
important data exception files to the quality assurance process;
any permanent dataset that represents the end product of an analysis of a
dataset;
any data file condition that is not readily recreatable from an existing static
file and existing processing; and
any data file to be transferred out of the IMS.
Initially, active files should be archived at all unrecreatable times of change. Problems
of redundant archiving can be analyzed and solved as they become recognized.
The frequency of archivals will also be established by SOPs. The frequency of
archival depends on the activity of the file; at the least, archive such that no file
activity will be lost or that no file condition can not be recreated.
A minimum of two on-site copies of all files should be archived, and at least one
archive copy stored off-site. Any file characterized by much update activity should
maintain a historical set of archives, the number of archives composing the set to be
determined by the amount of file activity.
While data archiving will be system maintained to the greatest degree possible,
human initiation of some data archives is inevitable. When human initiation is
necessary, memory aids ("tickler files") will be utilized in the system.
8.5.h. Geographic Information System (GIS) Applications
EMAP - GL will have access to the GIS system located at the ERL-Duluth. The
system currently uses a Data General AVIION Model 410 dual processor, with 32 MB
of memory and 2GB of disk storage. Accessories include a CalComp plotter and
CalComp Digitizer. The GIS system at ERL-Duluth uses the 6.0 version of
ARC/INFO, currently in Beta test.
The EMAP GIS Task Group is currently developing procedures and standards for the
use of all Task Groups. EMAP - GL will utilize the extensive spatial analyses and
display capabilities of ARC/INFO to produce base maps and overlays depicting Great
Lakes sampling and the results of indicator analyses.
8-18
-------
8.5.i. Quality Assurance
Data QA is not the responsibility of any single system component, instead it pervades
the entire IMS. While IMS QA is comprised of data verification and validation SOPs
and processing (applied SOPs), of procedures and practices that help ensure data
integrity, design perspective remains the motive force.
Data verification SOPs provide the basis for developing data entry and data review
verification. Several types of data verification processing will utilized:
Range checks on numeric data: Field and laboratory numeric data entry will
be checked against valid range definitions. Data that falls outside of the
acceptable range will be written to an exception report for review by QA
personnel.
Valid code checks on coded data: Coded field entries will be compared with
the set of valid code values established by the scientific personnel and
information management to assure compliance.
Duplicate record checks: Indexing of datasets establishes a means of
checking for duplicate (or missing) records.
All data entry validation will include error message display with contingency options.
8.5.J. Project Management
The integration of a number of complex database management systems, combined
with the timely, cost-effective production of processed data and the efficient utilization
of many varied resources makes the need for a Project Management System
imperative. Additionally, the EMAP - GL IMS will serve a program whose duration will
ensure change in outputs, technology, and methodologies. These changes will,
almost certainly, be manifested in structural and processing changes within IMS. A
Configuration Management System with the following components has been
suggested:
Configuration identification: definition and identification of items subject to
configuration control;
Configuration control: evaluation, coordination, and approval or disapproval
of proposed changes to controlled items;
Configuration status accounting: recording and monitoring of changes to
controlled items;
Data management: maintenance of official correspondence records,
configuration management records, and controlled documentation; and
8-19
-------
Configuration auditing: verification that controlled items are what their
documentation states they are and that they meet their assigned
requirements.
8.6. Resource Utilization
All information system development and management is accomplished by the
utilization of resources which may be acquired in two ways; the authorized use of or
access to appropriate existing capabilities, or the procurement, by approved Agency
procedure, of new resources. The EMAP - GL program will make use of existing and
planned resources compatible with the EMAP - GL program, to the extent possible,
and only acquire additional resources where needed. Agency resources which may
be accessible to the EMAP - GL program include current programs, data sources,
staff, and equipment. Similar resources may be available through other agencies,
institutions, contractors, and other governments.
Valuable guidance in the development of the IMS is provided through several sources.
The EPA OIRM within the Office of Administration and Resource Management
(OARM) provides a model which serves as guide to all Agency system design and
development. The IMC provides a forum for the cooperative resolution of problems
and the sharing of ideas and expertise, as well as guidance to the long term
integration of Task Group information systems. Most important, in practical terms,
other Task Groups have willingly shared their IMS development experiences, solutions
to problems, and design products. EMAP - GL will continue to utilize these available
resources to expedite IMS development.
The EMAP - GL program will be cooperating with several agencies that have
implemented broad-scale, long-term monitoring programs in the Great Lakes region.
The resources provided by this cooperative effort, e.g., equipment, human resources,
methodology, and experience, are essential to the success of Great Lakes EMAP
program.
The EMAP - GL program will make use of the ADP resources available at the ORD
Laboratories. The Information Manager will work with their ADP coordinators and
laboratory directors to plan and obtain approval for the utilization of these resources
by EMAP - GL
The EMAP - GL Information Management will, also, identify and utilize an ADP
Coordinator at each location within the program. Each ADP Coordinator will be an
existing local resource whose responsibilities include assisting the Information
Manager in development of the EMAP - GL ADP resources and plans.
8-20
-------
9. Coordination
9.1. Introduction
The focus of this section is coordination, both among components within EMAP and
between EMAP - GL and other state and federal programs.
In developing a complex program such as EMAP, a wide range of issues must be
acknowledged and addressed throughout the early phases. The coordination of
activities and analyses must occur at multiple levels both within the program and
across other programs. For example, within the Great Lakes component of EMAP,
there is the need to integrate water quality monitoring activities in the Great Lakes
Basin. Equally important is the need to integrate categories within EMAP. Outside of
EMAP itself are a host of programs which, although they cannot adequately address
the objectives of EMAP by themselves, can complement the information from EMAP to
provide a clearer view of current status trends in indicators of condition of surface
waters and diagnosis of conditions.
9.2 Within EMAP
An important aspect of EMAP is the inclusion of all ecological resources within the
program. From the National perspective, this provides the opportunity to evaluate the
relationships between conditions and problems in each resource category, their
impacts on one another and the potential to more effectively evaluate comprehensive
ecological resource management strategies. The achievement of these potentials
requires extensive coordination in the selection of indicators, methodologies, and
design.
Within EMAP, the focus to date has been the design and evaluation of programs to
identify status, trends, and probable cause of conditions within each ecological
resource group. Discussions to enhance the integrative aspect of the program have
been gradually increasing and will be the focus of the EMAP Integration and
Assessment Team. Coordination between ecological resource groups has been
facilitated by the coordination teams within EMAP (i.e., statistics and design, indicator
development, integration and assessment, information management, quality
assurance, and logistics). The activities sponsored by these teams currently provide
the framework to ensure our future ability to more fully integrate the information from
each aspect of the EMAP. Workshops have been held by the Indicator Development
coordinator to facilitate discussion between the ecological resource groups to ensure
that they take into account, in the development of their programs, information needs of
other groups that they might be able to supply or that might be supplied for them.
Coordination is also taking place in simply defining the specific resource categories
(e.g., wetlands from lakes) and in ensuring that all ecological resources are
considered.
9-1
-------
The Integration and Assessment team, in conjunction with the statistics and design
team, and the ecological resource groups began efforts during FY91 to evaluate
design alternatives that might maximize the ability to integrate information from various
ecological resource groups. These efforts can be coupled with the design and pilot
activities of each resource group to evaluate long-term options for the program.
9.3. Other Federal Agencies
Almost every federal agency that has a mandate or jurisdiction over some natural
resource has an interest in Great Lakes resources. We firmly believe that EMAP
should be a multi-agency program. It is the concept of the program that is important,
not the location within any particular agency. If aspects of the program objectives are
being met through activities in other agencies, then there is no need to duplicate that
aspect of the program.
While efforts will be needed to coordinate with each of the groups, we targeted our
early efforts toward those federal entities maintaining active Great Lakes monitoring
programs. This led to early interaction with the US Fish and Wildlife Service and the
National Oceanographic and Atmospheric Administration. Programs maintained by
each of these agencies were discussed in Chapter 1. Our intent here is to describe
potential interactions. Discussions have been taking place between EMAP - GL and
each of these agencies but no firm commitments have been achieved.
9.4. International Activities
Implementing monitoring activities within the Great Lakes will not likely be successful
without the cooperation of the Canadian government. They currently share
responsibilities with GLNPO for implementing GLISP. They also conduct a wide
variety of additional monitoring and research on the lakes. The initial contacts for joint
activities between EMAP - GL and the Canadians have been at the scientist level
within the Canadian Center for Inland Waters. As described in Chapter 10 which
addresses 1992 activities, a joint study on sediment indicators is currently being
proposed. In addition, there have been preliminary discussions of conducting
comparisons between offshore data on trophic status at the Canadian sites with data
obtained by them at proposed EMAP - GL offshore sites. Discussion at the
management level are underway with the goal of developing an integrated monitoring
program for ecological condition throughout the Great Lakes.
9.5. Research Organizations
As described in Chapter 1, research organizations such as universities and state and
federal research centers are important groups with which EMAP - GL needs to
interact. The research needed to allow EMAP to reach its potential will come from
these groups. Interaction with these groups during the developmental stages of
9-2
-------
EMAP will provide the mechanism to ensure that the program is grounded in sound
scientific principles. Existing field sites can provide locations for testing and
developing indicators of ecological condition and an understanding of the relationships
between our indicators of response and indicators of exposure, habitat and stress. In
addition, the feasibility of engaging university consortia to conduct portions of the long-
term program are being actively pursued.
9.6. Conclusions
The types of interactions and coordination which are needed to create a successful
program which best serves the public interest have been presented. During the next
year, EMAP - GL will be pursuing these coordination efforts and seeking to establish
mechanisms to facilitate this coordination. We envision the need for a scientific
advisory panel to ensure the sound footing of the program and an interagency
coordination panel to facilitate the interaction and coordination needed among
participating state and federal agencies.
9-3
-------
10. Fiscal Year 1992 Field and Analysis Activities
10.1. Introduction
As summarized in Chapter 2, and discussed throughout this strategy, there are a
number of questions that must be addressed before EMAP - GL can be fully
implemented. As a first step in accomplishing the EMAP - GL objectives, a series of
pilot activities will be conducted over the next year. The goals of these activities are
to begin answering the following questions. Not all questions, of course, can be
studied or answered in one year. Some will require extensive data collection,
sampling, research, and evaluation before answers will be apparent. Monetary
resources also restrict the number of questions that can be addressed in any given
year. Therefore, EMAP - GL will be continuing to evolve even after monitoring and
assessment activities have been started in all the Lakes. The following questions are
planned to be studied in 1992:
Questions on EMAP - GL Design
Questions over the density of the base grid for offshore areas in Lake
Michigan will be addressed by evaluating existing data and collecting
additional data for trophic status (Section 10.2.a.) and sediment related
indicators (Sections 10.2.b. and 10.3.C.).
Lists and areal extent of the harbors and embayments of Lakes Michigan
and Superior will be determined from USGS maps using the definitions
described in Chapter 3 (Section 10.4.).
Available information regarding the extent of coastal wetlands of Lake
Michigan will be identified (Section 10.5.).
Questions on EMAP - GL Indicators
Recommendations for wetland indicators will be developed through a
workshop of Great Lakes wetland scientists (Section 10.5.).
Investigations into the definition of nominal conditions for sediment indicators
in the nearshore of Lake Michigan will be conducted in conjunction with
Canadian studies on the remaining four Lakes (Section 10.3.).
The selection of appropriate indicators for fish will be explored through
analysis of existing data and consultations with Great Lakes experts
(Section 10.6.).
An evaluation of index periods for trophic status in the offshore resource
class will be conducted by comparing spring and summer data in Lake
Michigan (Section 10.7.).
10-1
-------
Evaluation of existing data for Lakes Michigan and Superior, along with
some sampling activities, will investigate the use of diatoms as
representatives of Great Lakes phytoplankton populations, the use of
sediment cores for historical trend analysis of diatom populations, and the
exploration of sediment traps as an integrative measure of annual diatom
population abundance and distribution (Sections 10.3.b. and 10.8.)
EMAP - GL pilot activities will begin on Lakes Michigan and Superior in 1992 (Tables
10.1 and 10.2). Spring cruises are planned for both lakes, while two additional
cruises, one summer and one fall, are planned for Lake Michigan. Assessment of
offshore trophic status and determining compatibility between data collected at EMAP
grid sites and that collected under other sampling programs (e.g., GLISP) will be the
primary focus of the spring cruises. The summer cruise in Lake Michigan will also
include a sediment sampling component which will be used to examine variability in
offshore benthic macroinvertebrate communities and the adequacy of the base grid
sampling intensity. The fall cruise is part of a cooperative nearshore study with
Environment Canada and NOAA which will collect sediment in unimpacted nearshore
areas. Benthic macroinvertebrates in these samples will be enumerated and used to
determine nominal conditions. The remainder of this chapter describes these field
activities in greater detail. Great Lakes historical datasets (Table 10.3) will be
compiled during the pilot project to evaluate design and analysis aspects of trophic
status and sediment indicators.
10.2. Investigations Within the Offshore Resource Class
10.2.a. Application of the EMAP Offshore Design for Trophic Status
Of all the indicators suggested for EMAP, those associated with estimates of offshore
trophic status are the closest to implementation (refer to Chapter 4). These
measurements have been utilized at the lake scale for almost two decades. As part of
the Great Lakes WQA, GLNPO has been monitoring offshore trophic status by
sampling at 11 sites throughout Lake Michigan (refer to Chapter 3). These sites were
selected to represent the offshore portion of the lake by investigating data from an
intensive survey conducted during the 1970's. However, the GLNPO site selection
process did not use a statistical approach that allows for statements of condition with
known levels of certainty. The EMAP base grid that will be used for offshore EMAP
monitoring has roughly the same density as the existing GLNPO monitoring sites
(Figure 10.1).
10-2
-------
Figure 10.1 1992 EMAP(D) and GLISP(A) sampling stations for Lakes Michigan and Superior.
10-3
-------
CO
92
;ซ
I
I
Q.
u.
CD
I
Q.
liJ
C\J
CD
5
CD
I
ง
Q.
F- UJ
UJ :r*
"5
0.
co
o
P
co
8
oc
a:
a
a!
AGENCY
0
1
O
UJ
O
cc
O
UJ
oc
V
3
O)
O
a.
CD
a.
UJ
ง
"w
g
i
.2"
o
5
Nominal Conditions for
Nearshore Benthic
Communities |
Cfl O
nt Communitie
and Geophysi
aract eristics
-It5
ID X
CO 0
o
CD
fc
J2
O
0
O
'c O
"0
2-g
> a
UJ a
n
o
.c
to
3
z
Trophic Status and
Historical Data Compatibility |
entional Water
istry; Nutrients
Is
CO
s-
o
O.
CD
o.
LU
Index Period, Indicator
Research, Trophic Status, and 1
Historical Data Compatibility |
entional Water
istry; Nutrients
> t
C 01
35
0)
1
Index Period, Indicator 1
Research, Trophic Status, and 1
Historical Data Compatibility |
entional Water
listry; Nutrients
11
0 0
en
to
ง,
c 0
0) o
1-8
> a
O
5
-------
.0)
I
x
w
T3
CO
c
O)
'ง
Q
O
111
O)
C\l
C)
Long-Term Objectives
g
:ง
Current Act
8
2
_c
5
s
Q/
<
T3
i
ฃ
o
0 ^
Determine required sampling intensity f
macroinvertebrates
Determine small-scale spatial variability
Address transition zone' between near
offshore
i i i
ซ
(/)
1
CO
?
Historical data acquisition ar
- Power analysis
- Components of variance
ฃ
u
^
53
ฃ
C
3
E
ง
o
0
'c
Q)
CO
ฃ 3
1 *
UJ 0
J5 M
Development of fish-related indicators fi
GL
Determine spatial/temporal variability as
with past sampling programs
1 1
o
CO
3
Historical data acquisition
Review of sampling techniqi
field/laboratory protocols
ง
ง
o
T>
.C
CO
Li.
5
. ฃ
0)
+;
s
Complete inventory of the number and
Great Lakes coastal wetlands
tป
o
*z
T3 Q-
C 0
CO .C
-^ w
ฃ ฃ
O o
c ฃ *
^~ Qi
w > w
c c g
CO
ist
0)
5
u
f'
* ฃ.ฃ ง
o 2 8 ซ
Determine required sampling intensity f
status indicators in offshore and nearsl"
Examine sensitivity of the CTI to changt
nearshore trophic status
Develop an index for offshore trophic st
1 1 1
Historical data analysis
- Power analysis
- Components of variance
ฃ
0
to
Q
'6
^
(A
i
w
o
ฃ
Q.
2
2
0) W
c s
~ 6
S>ฃ
Development of a bays/harbors samplir
be used in '93 pilot and demonstration
1
0
C ^
^ .2
ง ^ 2
2 w >
M S
5 5 S
"ฃง3 S
3SS i
Identification of Great Lakes
meeting the EMAP - GL defi
Incorporation of digitized Gr
into the EMAP - GL GIS
Comparison of frames (enhj
W
c
0)
E
>,
CO
.Q
E
LU
"O
CO
2
o
ฃ3
^
CO
I
10-5
-------
5
%
CD
|
01
5
1
'5.
o
O
CO
CO
O
"co
o
o
to
JC
5
to
2
O
CO
c>
13
CO
CO
f
(0
*
O
*
CO
o ^5
3 CtJ
w 0
IX
"CO
u
Is.
(0 O
^ o
c?co
3
0>
co G
CO 0>
o
0)
ป-
2
CO
o
co"
A
Chlorophyll-a Secchi
Total Phosphorus
3;
z"
O
c
I
3E
0>
3
O)
00
tO
fc
DC
O
CO
1
Q.
LU
CO
s
CM
c3
ง
Q.
Vir
O
Macrobenthos
Abundance and
Biomass
(Three Replicates each 1
and Peterson Grabs)
r
z"
o"
S
c o>
ฎ i
CO ฃ
3
CO
i
i
<
CD
.>
c
r>
CD
13
CO
w.
1
(O
CM
^
co
xf
LU
Sediment Chemistry, pH
Grain Size
X
z"
o"
co
.9?
o
Q)
3
ฃ
0)
"tu t
*^ -J
W ro
.52 .9
c o5
'=8
G
CO
,^
^*
Macrobenthos Communi
Structure
i
z"
b
1
CO
CD
CO
S
CD
C
O
CO
Q) Q) ป
9-^-.Q
5 co w
E
0
i:
fli
^ "d
3 ^
o to
Benthic Community Stru
Eight Sampling Devices
Multiple Box Cores
1
i
z"
i
LU
CO
CO
0)
00
oo
^
1 8
^ i
O CO
. ~O
LU ro
O
w
CO
T3 UJ
CO CO
ง1
- -
JO o
ฃf
> co
* CD
I2
CO
II
coS
CO OT
Q ZZ-
"O -
CD fO TT
fi GQ
"ti rt\
Q. ซ ฃ
C CO CO
3 Z O
II II II
T- OJ CO
CO (/)
ซ *=
g.0
CO
o
-------
The primary objective of the 1992 EMAP activities in Lake Michigan will be to assess
the compatibility of data collected at EMAP grid locations to the data collected as part
of GLISP. The results of this compatibility study will determine if 1) EMAP - GL can
use data collected under GLISP in an historical context (e.g., for trend analysis) and 2)
the EMAP base grid is of sufficient density to provide the ability to detect change. To
minimize potential between-ship variability the EPA surveillance vessel will conduct
sampling at the GLISP and EMAP sites, simply alternating its normal cruise pattern
between these two networks during the spring sampling program. Comparable data
will be collected at all stations and standard descriptions and operating procedures will
be utilized (Palmer and Warren 1992). The resulting data will be initially stored in the
shipboard computer system and subsequently transmitted electronically to the EMAP -
GL information management network.
In addition, GLNPO will be extending its normal spring cruise to Lake Superior for the
first time (Environment Canada has been responsible for surveillance in Lake Superior
since 1968). Because GLNPO has not previously conducted water quality sampling in
Lake Superior, the sample locations will be the EMAP base grid locations (Figure
10.1). The relative homogeneity of the offshore waters of Lake Superior should allow
the use of the Environment Canada dataset for trend analyses. To assess the degree
of offshore homogeneity, Environment Canada will add nine offshore EMAP grid
sampling sites to their regularly scheduled cruise in May 1992 (Figure 10.1.).
Although the sample size for comparison is small, differences detected between the
GLNPO and Environment Canada data may point to the need for more detailed
studies on between-ship variability.
10.2.b. Application of EMAP Offshore Design to Benthic Community Structure
Benthic community structure is one of the important response indicators for
EMAP - GL. The spatial variability of this indicator in offshore sediments will be a
significant factor in determining whether the proposed sampling density (EMAP base
grid) is adequate. As part of the GLNPO summer cruise in Lake Michigan, sediment
samples will be collected at the EMAP sites to determine 1) the variability of benthic
community structure in the offshore area at the EMAP sites using the proposed base
grid density and 2) the confidence intervals associated with the analyses. Methods of
sampling are the same as described in section 10.3.
10.3. Sediment Indicators in the Nearshore Resource Class
Sediments are both the repository of many chemical contaminants and the substrate
for many organisms critical to Great Lakes food webs. Many of the problems
identified in the Great Lakes, including most Areas of Concern, are associated with
sediment contamination. Because of the chemical binding properties of sediments
and the fact that particle associated contaminants settle out onto sediments, they have
often been considered as a sink for contaminants. We now realize that because
contaminants do concentrate in sediments, they are a critical source of contamination
often long after inputs of toxic materials have been reduced. Due to their critical role
10-7
-------
in Great Lakes ecosystems, many indicators tentatively selected for EMAP are related
to sediments. However, while there are varying amounts of data on benthic
community structure for the various lakes (e.g., Cook and Johnson 1974, Nalepa
1987, Nalepa 1991), the combination of benthic community structure, sediment toxicity
tests, contaminant analyses, and physical characterization has rarely been attempted.
The following is a joint study between Environment Canada's National Water Research
Institute (NWRI), NOAA's Great Lakes Environmental Research Laboratory (GLERL),
EPA's Great Lakes National Program Office (GLNPO), and EPA's Environmental
Research Laboratory - Duluth (ERL-D).
10.3.a. NWRI Study on Sediment Indicators
The NWRI has completed the first year of a three-year study to develop biological
sediment guidelines for the four Great Lakes with Canadian boundaries (Reynoldson
and Day 1991.) The NWRI program is based on the fact that unperturbed systems
support, for extensive periods of time, assemblages of species that are self-
maintaining and resilient to normal environmental fluctuations. Such communities can
be defined through the collection of physical, chemical, and biological data. The
strategy of the EMAP portion of the study is twofold: 1) to help establish a reference
database of clean sites and 2) to include Lake Michigan in the sampling program.
The overall objectives of the program are to:
Classify benthic invertebrate community assemblages and toxicity test
responses that represent different substrate condition (habitat).
Develop a model to predict the benthic community and toxicity test response
from habitat (sediment) characteristics.
Establish nominal conditions for benthic communities and test responses.
A series of EMAP objectives will also be addressed through this study:
Select key species and toxicity test endpoints that show the most robust
predictive response for defining nominal sediment conditions and propose a
framework for determining biologically significant direction changes in
sediment conditions based on the invertebrate fauna and response.
Examine annual and sampler variability in the approach by including a
subset of sites previously sampled over a ten-year period by NOAA.
Determine the relationship between community structure and bioassay
assessments of sediment conditions.
10-8
-------
Determine the efficiency of incorporating measures of sediment chemical
contaminants that might be used as a means of predicting future exposure
to organisms higher up in the food web.
10.3.a.1. Site Selection
Two sets of sites will be selected: 1) a reference set for inclusion in a Great Lakes
nearshore reference data matrix and 2) a subset for resolution of annual and operator
differences.
The first and major set of sites will be selected for incorporation into the NWRI
nearshore reference database. The reference sites to be sampled will represent, as
best as is practicable, pre-contamination conditions. Therefore, sites are ideally
required to be: less than 85 m deep, less than 3 k from shore, have more than 10 cm
of accumulated fine grained sediment, have more than 1 ha of contiguous fine grained
sediment, have an unexposed fetch, be away from outfalls, be away from
development, and be accessible. Sampling sites will be located using a stratified
approach with equal site distribution within a stratum. Sampling strata have been
selected on the assumption that the potential for differences in reference communities
will be maximal in varying physical habitats. For the Canadian sites, the sampling
strata were defined by seventeen ecodistricts (Wickware and Rubec 1989) that
encompass the Canadian Great Lakes shoreline; the ecodistricts are based on
differences in geomorphology, geography, climate, soil type, and vegetation. To
provide a data matrix of approximately 250 sites (required for model development), 15
sites were located in each ecodistrict. In the case of Lake Michigan, it is proposed to
use the five ecoregions identified by US EPA and Environment Canada (1988) to
locate 12 sites in each ecoregion. This would provide 60 sites that should be
sufficient and comparable with the other Great Lakes (Table 10.4). The exact
locations of sites within the ecoregions will be made from examination of land use
maps, topographic maps, hydrographic charts, nearshore sediment maps, and the
location of outfalls and intakes.
10-9
-------
Table 10.4 Number of ecodistricts and sites in each Great Lake.
LAKE
Superior
North Channel
Georgian Bay
Huron
Erie
Ontario
Michigan (proposed)
ECODISTRICT
4
3
3.5
2
2.5
2
5*
SITES
68
51
59
34
42
34
60
* US Ecoregions (US EPA and Environment Canada 1987)
The large number of sites in the reference site data matrix precludes sampling to
determine seasonal and annual variation. This variation is a concern particularly as it
may affect the community assemblages. In the Canadian dataset, these issues are
addressed by sampling a subset of stations (10%) over three years and at four
stations sampling monthly for two years. In Lake Michigan, NOAA has sampled 30
sites over a ten-year period (Nalepa 1987). These sites will be sampled during the
1992-93 field season. Nine of these sites meet the depth requirement and will be
included in the reference set. The remaining 21 sites will be used to test this
approach for EMAP offshore application. Finally, a subset (5-10) of sites will be
sampled by NWRI and NOAA to compare methods, the sensitivity of the collection and
analytical methods, and sampler variation (see Section 10.3.b.).
10.3.a.2. Field Methods
Geophysical Parameters (Table 10.5): When onsite, precise latitude and longitude will
be obtained from the Loran C system. The Loran C coordinates and the chain being
used will be noted. Water depth will be noted and air temperature will be measured
for later comparison with water temperatures. Also, comments on wind and weather
will be noted. The distance from the shore will be calculated from charts.
Limnoloqical Parameters (Table 10.5): A Van Dorn water sampler will be used to
obtain a water sample 0.5 m from the bottom. A one liter sample will be drained off
and stored for filtration. Dissolved oxygen, pH, and temperature will be taken on the
remainder of the sample. Half of the 1 L sample will be passed through the filtering
apparatus and divided as follows:
10-10
-------
Table 10.5 Geophysical parameters measured at each site.
GEOGRAPHICAL
Distance from shore
Latitude
Shoreline development
Slope
Depth
LIMNOLOGICAL
Temperature, surface and
bottom
Degree days
Thermocline depth
In bottom 0.5 m
Alkalinity, hardness
pH
Nutrients
Oxygen
SEDIMENTOLOGICAL
Water content
Particle size
Loss on ignition
TP, TOC, TON
AVS
Metals, major ions in pore water
Metals, major ions
Nutrients
i) For total phosphorous, filtered water will be placed in a square 125 ml_ glass
bottle containing 1 ml_ of sulfuric acid. The samples will be sealed, stored
at 4ฐC and shipped to CCIW for analysis.
ii) Filtered water for nutrients will be placed in a round 125 ml_ glass bottle.
The samples will be sealed, stored at 4ฐC and shipped to CCIW for analysis.
Samples (500 mL) will be filtered through a 0.45 Millepore ฎ Sartorius filter.
The first portion passed through the filter will be discarded. Clean
glassware will be used and the filters handled with forceps only.
The remaining water from the 1 L will be divided as follows:
i) For total phosphorous, water will be placed in a square 125 mL glass bottle
containing 1 mL of sulfuric acid. The samples will be sealed, stored at 4'C
and shipped to CCIW for analysis.
ii) For alkalinity, water will be placed in a round 125 mL Nalgeneฎ bottle. The
samples will be sealed, stored at 4ฐC and shipped to CCIW for analysis.
Sediment Characterization (Table 10.5): Sediment samples (600 mL) for
geochemistry will be taken from the upper 5 cm of the box core or from a Ponarฎ
sample. The sediment will then be homogenized in a glass dish with a Nalgeneฎ
spoon. The sample will then be divided as follows:
i) Organic contaminants (approximately 125 mL into a hexane-prewashed
glass bottle with hexane rinsed aluminum foil liner): Samples will be sealed
and stored frozen (or at 4ฐC in the field) and shipped to ERL-Duluth for
storage and analysis. Depending on other data, it is likely that only a subset
of these samples will be analyzed.
10-11
-------
ii) Particle size distribution (about 20 mL into a scintillation vial): Samples will
be stored at ambient temperature in the field and shipped to CCIW for
freeze drying and analysis.
iii) Metals and nutrients: samples for metals, LOI, AVS, TP, TOC and TN will
be stored together in a Whirlpakฎ plastic bag. Samples will be stored at
4"C in the field and shipped to CCIW for freeze drying and analysis.
Pore Water Sampling and Processing: From each box core, a 10-cm core will be
taken and sealed. Samples will be stored at 4ฐC in the field and shipped to CCIW for
extraction by squeezing in a nitrogen atmosphere and subsequent analysis.
Community Structure: Samples will be taken from a box corer, or mini-box corer
developed at CCIW for this project. As an alternative, a techops corer may be used;
this device has been calibrated against the box corer and mini-box corer. The
sampler to be used depends primarily on the research vessel used in the project. Box
cores will be sampled by inserting five 10 cm Plexiglassฎ tubes (i.d., 5.5 cm) into the
sample, an addition tube will be used for pore water characterization (below). Core
tubes for community structure will be removed and each replicate will be placed into a
Whirlpakฎ plastic bag and kept cool until sieved. Replicates will be sieved (250u.) in
the field as quickly as possible. If sieving cannot be done in the field, the replicate
samples will be stored in 4% formalin in the Whirlpakฎ bag and sieved as soon as
possible thereafter. When replicates have a high sand content, they will be placed in
a bucket and sieved (250u.) with water added. The replicates will be agitated and the
slurry poured through a 250|i sieve. The process will be repeated three times to
ensure that no invertebrates remain in the sediment. Sieved replicates will be placed
in scintillation vials and preserved with 4% formalin. Replicates with large amounts of
organic material will be placed in larger containers and preserved with 4% formalin.
Vials and containers will be properly labeled and stored at ambient temperature.
Formalin must be used as alcohol causes oligochaetes to deteriorate. After 26 h
formalin will be replaced by ethanol.
Sediment Bioassavs: If a box core (50 X 50 cm) is used for sampling, three
subsamples will be taken after coring for community structure and pore water.
Otherwise, three Ponarฎ grabs will be collected at each site and shipped to CCIW for
bioassays. Each sample will be placed in its own plastic lined bucket and sealed.
The buckets will be kept as cool as possible until shipped to CCIW.
Sample Labeling: Each sample will be labeled with a 12 digit number, e.g.,
01/19/03/090991. The first two digits represent the ecodistrict, the next two the site
number, the next two the replicate number, and the last six the date (month/day/year).
Additional labeling to meet the needs of EMAP tracking will be added if necessary.
10-12
-------
10.3.a.3. Laboratory Methods
Community Structure: All samples will be sorted and picked by hand using low power
on a stereo microscope. Small amounts of material will be placed in a petri dish
marked with a grid and scanned twice. Organisms will be identified to major group
(family, class) and placed in ethanol in labeled vials (20 mL plastic scintillation vials)
for identification. Oligochaetes and chironomids will be completed to the species level
where possible. This will provide greater discrimination, and provide information not
available if identifications will be only taken to lower levels. Taxonomic verification will
be conducted by referral to recently published keys, reference collections, and
consultation with recognized experts. A reference collection of all identified material
has been established and all identifications will be confirmed by an acknowledged
expert.
Bioassav: Four sediment bioassay organisms are being used at CCIW and will be
used in this project: Chironomus riparius, Hyalella azteca, Hexagenia limbata, and
Tubifex tubifex. Chronic tests will be conducted using reproduction and growth as
endpoints. All test methods are described in a Standard Operating Procedure (SOP)
document and will be referenced to internationally recognized test methods. Methods
will always specify acclimation periods, dose level selection (for reference toxicants),
dosing schedules, test duration, test endpoints, lifestage, age or size of test
organisms, and test acceptability criteria. Water change rates (if applicable), loading
rates (organism mass/L), and aeration rates will also be described. Mode of operation
will always be stated, i.e., static or semi-static (renewal). Exposure verification by
chemical analysis for reference compounds will be conducted.
The ability of laboratory personnel to obtain consistent, precise results will be
demonstrated with reference toxicants (positive controls) before they attempt to
conduct toxicity assessments with field samples. At least five toxicity tests with
reference toxicants will be conducted for each species of organism to establish
warning limits early in the program. Once consistent results (e.g., a coefficient of
variation in the endpoint of less than 30%) are achieved, the frequency of reference
toxicant tests will decrease to approximately once per month for each toxicity test.
Sediment samples for bioassays will be scheduled for analysis as soon as possible
with the schedule dependent on the permissible holding times according to the SOP.
All documentation and test results that will be stored electronically in computer files
will undergo human verification against input errors and inadvertent program changes.
There will be a back-up system to permit recovery of data lost due to hard disk
failures or operator errors.
As the objective is to determine background responses to a variety of conditions, test
organisms will not be fed. Because of the confounding effects of resident fauna,
primarily predation and competition for resources, sediments will be sieved through
250u, mesh before testing.
10-13
-------
10.3.a.4. Data Analysis
Data will be entered into one of three data matrices, these will be stored electronically
and on hard copy. Electronic storage will use Lotus 1-2-3 data files. Community
structure data will be entered into a two-way data matrix of columns (taxa) and rows
(individual replicate samples from a site). Environmental data and bioassay data will
be stored similarly with the columns being respectively: environmental variables and
individual assay endpoints. Bioassay response and benthic communities will be
classified using multivariate statistics. Assurances that these formats are compatible
with EMAP data files will be made and the information will be electronically transferred
to the EMAP information management system for storage and use.
Initial data examination will be by table arrangement. Data will be examined in its
original form, as percent abundance, and using a similarity index. Sites will be
classified using PCA (principal component analysis) or CCA (canonical
correspondence analysis). Two methods will be used to relate community structure
and bioassay response to environmental variables: CCA or MDA (multiple discriminant
analysis). Both will be done in a stepwise fashion. The computer software packages
TWINSPAN, DECORANA, and SYSTAT will be used.
10.3.a.5. Reporting
A report on the Lake Michigan portion of the project will be prepared within six months
of sample completion. Tentative determinations of nominal conditions, a test of their
sensitivity and, from the NOAA comparison, a test of temporal variability will be
included in this report. The report will be based on the Lake Michigan data and 150
Canadian stations that will have been sampled. A more detailed and comprehensive
report that will include detailed assessment of the importance of annual and seasonal
variability to the prediction process and more extensive testing of the recommen-
dations will be prepared. It will be based on the entire dataset on completion of the
Canadian study and will include more than 250 Canadian stations. This more
comprehensive report will be prepared for Environment Canada but will be coordinated
with EMAP - GL and GLERL
10.3.b. NOAA-GLERL Project Description for Use of Sediment Traps for Indicator
Measurements
Sediment traps have been proposed as the most efficient means of collecting
integrated diatom samples (see Chapter 4), but this approach has not been tested.
This project (coordinated through an IAG with NOAA-GLERL) will test the hypothesis
that diatom samples can be collected using long-term (one year) sediment traps in a
vertical mooring configuration.
The test site is an established GLERL long-term monitoring site in southern Lake
Michigan (100 m bottom depth, 26 km southwest of Grand Haven, Ml; 43.04ฐN,
86.64ฐW). Seven autosequencing sediment traps were deployed in October 1991 in a
single vertical mooring at 15, 35 (duplicate traps), 75, 90, and 95 (duplicate traps)
10-14
-------
meters depth. Each trap is programmed to collect 23 samples on a 15-day rotation,
thus providing 161 samples over the course of approximately one year. In order to
test the EMAP hypothesis, these traps will be retrieved between August and October
1992; the samples will be allowed to settle, and the overlying water will be siphoned
off. Samples will be freeze dried and weighed and then split to provide a portion of
each to the University of Michigan, responsible under a separate EMAP agreement for
the diatom analyses (a FY94 project).
In addition, it is apparent that sediment and sediment trap samples are matrices that
are believed to be important and useful to the goals of the EMAP - GL program.
However, the importance of these matrices as environmental/ecosystem indicators
(i.e., what do they indicate?) and the best methodology for application to EMAP's
goals have not been well thought out. Much has been learned over the past 20 years
about how to interpret sediment trap samples and the information stored in sediments.
Since similar analyses would be made on both sediments and sediment trap samples,
a unified or coordinated protocol should be developed during FY94. GLERL will
convene a small workshop of selected experts to assess the importance, utility, and
application of sediments and sediment trap materials to the goals of the EMAP - GL
program and to develop a recommended sampling protocol for these media that meets
the goals of EMAP, to develop a recommended sampling protocol for these media that
meets the goals of EMAP, and to be technically defensible.
10.3.C. NOAA-GLERL Project Description for Benthic Survey and Methodology
Comparison (Southern Lake Michigan)
The status and trends of the benthos in the Great Lakes have been identified as a
primary environmental/ecosystem indicator (Chapter 4). For southern Lake Michigan,
GLERL has compiled an extensive status and trends dataset for benthic organism
communities at some 40 stations. These data cover the periods 1964-1967, 1980-
1981, and 1986-1987. GLERL will resample these stations during the spring, summer,
and fall of 1992, collecting Ponarฎ grab samples in triplicate at each station.
As part of the EMAP Lake Michigan pilot project, a subset of ten of GLERL's long-
term trends stations will be selected, and two sets of triplicate samples will be
collected at each of these stations during each of the three sampling dates (i.e., a total
of 180 samples). One set of triplicates (90 samples) will be retained and processed
by GLERL; the duplicate set of triplicates will be provided to Environment Canada for
comparison between Ponarฎ grab sample methodology (used by GLERL) and the box
core sub-cores methodology (used by EC). GLERL will provide taxa abundances to
the lowest practical taxonomic level in each of the triplicate GLERL samples.
As part of the EMAP - GL Lake Michigan pilot project, the Great Lakes National
Program Office (GLNPO) will collect 11 benthic samples from the open lake region
during the summer of 1992. These samples will be processed (screened and
preserved) by GLNPO and sent to GLERL during FY94. GLERL will identify and
provide abundances of taxa in each of these samples.
10-15
-------
10.4. Development of the Harbors and Embayments Sampling Frame
USGS quad maps for the shoreline of Lake Michigan will be obtained and the number,
location, and area of the harbors and embayments (as defined in Chapter 3) will be
determined. If not available in digital format, these maps will be digitized and entered
into a GIS for future use and analysis.
10.5. Wetland Indicators
As discussed in Chapter 4, little research on indicators of Great Lakes wetland
condition has been conducted. As an initial activity, we intend to hold a workshop of
Great Lakes wetland scientists and EMAP - Wetland members to develop
recommendations for appropriate measurements and indices. Although not actually a
1992 fiscal year activity, (October 1, 1991 - September 30, 1992), this workshop will
probably be held in late fall 1992 or early winter 1993 before any further field work is
planned. Scientists with expertise in Great Lakes coastal wetland and nearcoastal
ecology (vegetation, benthos, fisheries, wildlife, hydrodynamics) and landscape
ecology will be drawn from academic institutions and state, provincial, federal and
tribal groups from the U.S. and Canada. The workshop will also bring in persons
experienced with EMAP design from other resource groups, notably EMAP-Wetlands
and EMAP - Near Coastal (Estuaries). Objectives of the workshop will include:
identifying existing databases on biota and wetland functions at landscape
and local scales;
identifying gaps in information that will be pertinent to developing indicators
specific for Great Lakes coastal wetlands;
identifying ways that Great Lakes coastal wetlands might be grouped for
comparative assessments;
identifying whether the EMAP grid or a list frame will be most appropriate for
selecting monitoring sites;
review of the indicators previously identified by EMAP - Wetlands and EMAP
- GL programs and developing recommendations for their use or
improvement; and
generation of a list of suggested reference sites for pilot and demonstration
studies on each of the Great Lakes.
10.6. Fish Indicators
As suggested throughout this report, there are many issues that require further
analysis of existing data before being addressed for EMAP - GL. Because of their
10-16
-------
high perceived social value and their position in the food web, evaluation of existing
data on fishes will be a high priority.
There are several agencies that have responsibility for gathering data on the fishes of
the Great Lakes including the measurement of chemical residues. Because this is
such a significant issue and one that can involve substantial cost, a thorough
evaluation of existing data and methods will be conducted. Specific concerns that will
be looked at include dealing with the high proportion of exotic versus native species,
variability of sampling techniques, the compatibility of existing sampling approaches to
the EMAP - GL objectives, community versus population versus individual based
indicators, inconsistencies in methodology and criteria for contaminants in fish tissue,
and how to treat the tremendous numbers of sport fish which continue to be stocked
into the Great Lakes.
10.7. Index Period for Trophic Status Indicators
The traditional season for collecting and reporting data on trophic status in the Great
Lakes is soon after ice-out (early spring). This is the period when the lakes are
generally well mixed and before biological activity has incorporated the nutrients
associated with enrichment. However, for many of the other indicators, late summer is
believed to be the most appropriate period to make measurements. GLNPO has been
conducting summer cruises for several years, and the data is in the process of being
evaluated regarding the relative information obtained at the two periods of time. As
part of this year's study, GLNPO will also be visiting the EMAP - GL offshore sites
during their summer cruise. This data, in addition to the existing data, will be used to
help evaluate the two sampling periods as they relate to EMAP - GL objectives.
10.8. Investigations of Diatom Populations as Indicators of Trophic Status and Biotic
Integrity
A major difficulty in monitoring and assessing primary producer assemblages is the
great temporal variation in species composition and abundance observed in the water
column. To meet the needs of EMAP - GL and reduce the amount of sampling that
would be required to characterize phytoplankton populations from water sampling,
diatom populations will be assessed using sediment cores and sediment traps. This
dual approach will:
provide the paleoecological history of the Great Lakes (sediment cores);
develop information on nominal or reference conditions, as they pertain to
diatom populations prior to European settlement in the Great Lakes Basin
(sediment cores);
10-17
-------
provide a temporally integrated assessment of diatom species composition
and abundance, both long-term (sediment cores) and annually (sediment
traps);
identify water quality trends and emerging trends based on sediment trap
results;
integrate population measurements from epilimnetic, hypolimnetic, or a
mixture of waters dependent upon sediment trap placement; and
offer various levels of temporal resolution, if desired, from both sediment
traps and sediment cores.
For FY92, further development of the sediment core/sediment trap approach strategy
will be necessary. The known literature regarding paleoecological investigations of
diatoms in the Great Lakes and elsewhere will be summarized. In addition, the
literature on diatom recovery from sediment traps will be summarized, as it
preliminarily appears that little information is available.
To address the sediment trap aspect of the approach, a limited amount of sampling
and analysis will be conducted in FY92 (refer to Section 10.3.b.) Sediment traps have
been deployed by NOAA-GLERL at GLERL's "100 meter station" in southern Lake
Michigan. Seven sequential sediment traps are positioned such that two are near the
surface, two are near the lake bottom, and the other three are staggered at various
depths. Upon retrieval, a quantitative portion of each set of deposited material will be
analyzed for diatom assemblages. Analysis of these samples will begin to address
issues regarding depth of trap placement and reproducibility of traps and samples.
Other factors that will have to be examined in the development of this strategy include:
an estimate of the number of sediment traps; the relationship of plankton in the water
column versus those in the sediment traps; and the number of diatom analyses
required to reduce the variability to an acceptable level.
Other factors which will be examined in strategy development will include: an estimate
of the number and distribution of cores required, the number of samples required to
reduce variability expected on spatially-correlated core intervals, resolution (time scale)
required for core intervals, potential of dissolution effects, relationship between extant
plankton and sediment cores (Battarbee 1981), and the relationship between diatoms
found in sediment cores and sediment traps.
The descriptive metrics associated with diatom populations will be examined and
metrics involving specific species, species diversity, evenness, and redundancy will be
evaluated. Abundance expressions of biovolume and number/cm2 will also be
evaluated. The development of composite and similarity indices and those using other
statistical methods, e.g., Canonical Correspondence Analysis, will also be evaluated
for the Great Lakes (ter Braake 1986; 1989, ter Braake and Barendregt 1986, ter
Braake and van Dam 1989).
10-18
-------
All methodological aspects of diatom studies will be coordinated with other national
EMAP programs. All diatom samples will be cleaned with hydrogen peroxide and
mounted on slides with a high-resolution, optical mounting resin. Preparation methods
used will allow a quantitative estimate of the flora. Samples will be examined with a
research quality, light microscope at a magnification of 1000X and 500 valves per slide
will be enumerated. Standard quality assurance protocol requires 1 of every 10 slides
to be replicated; within-slide variability will also be evaluated. Specimens will be
taxonomically identified using the recognized reference literature. Light and electron
micrographs will be used for documentation of the flora, where known taxa and
unknown entities will be maintained in a catalogue with pertinent information. Slides
will be ultimately stored in a repository such as the Philadelphia Academy of Sciences.
Raw data sheets (count sheets) will be retained at all times. Raw data will be
transferred to computer format and output methods will meet the different needs of
EMAP - GL for reporting purposes.
10-19
-------
References
Agebeti, M. and M. Dickman. 1989. Use of Lake Fossil Diatom Assemblages to
Determine Historical Changes in Trophic Status. Can. J. Fish. Aquat. Sci. 46:1013-
1021.
Ahlstrom, E.H. 1936. The Deep-water Plankton of Lake Michigan, Exclusive of the
Crustacea. Trans. Amer. Microsc. Soc. 55(3):286-289.
Alley, W.P. and R.F. Anderson. 1968. Small-scale Patterns of Spatial Distribution of
the Lake Michigan Macrobenthos. Proc. 11th Conf. Great Lakes Res., Internat. Nat.
Assoc. Great Lakes Res.:1-10.
American Society for Testing and Materials (ASTM). 1988. New Standard Guide for
Conducting Acute Three Brood, Renewal Toxicity Tests with Ceriodaphnia dubia.
American Society for Testing and Materials, Draft No. 6, ASTM E47.01, Philadelphia,
PA.
Anderson, N.J. 1986. Diatom Biostratigraphy and Comparative Core Correlation with
a Small Lake Basin. Hydrobiol. 143:105-112.
Anderson, N.J., B.Rippey, and A.C. Stevenson. 1990. Changes to a Diatom
Assemblage in a Eutrophic Lake Following Point Source Nutrient Re-direction: A
Paleoecological Approach. Freshwater Biol. 23:205-217.
Applegate, V.C. and H.D. Van Meter. 1970. A Brief History of Commercial Fishing in
Lake Erie. US Fish Wildl. Serv., Fish. Leaflet 630:1-28.
Bailey, J.W. 1842. A Sketch of the Infusoria of the Family Bacillaria, with some
Account of the Most Interesting Species Which have been Found in a Recent of Fossil
State in the United States. Amer. J. Sci. Arts. 42(1):88-105.
Baillargeon, W.S. (ed.) 1991. EMAP Glossary (Draft). Atmospheric Research and
Exposure Assessment Laboratory, U.S. Environmental Protection Agency, Research
Triangle Park, NC:15 p.
Baldwin, N.S., R.W. Saalfeld, M.A. Ross, and H.J. Buettner. 1979. Commercial Fish
Production in the Great Lakes, 1867-1977. Great Lakes Fish. Commission Tech. Rep.
No. 3.
Bartone, C.L. and C.L Schelske. 1982. Lake-Wide Seasonal Changes in limnological
Condition in Lake Michigan in 1976. J. Great Lakes Res. 8(3):413-427.
Ref-1
-------
Bascietto, J., D. Hinkley, J. Plafkin, and M. Slimak. 1990. Ecotoxicity and Ecological
Risk Assessment. Regulatory Applications at the Environmental Protection Agency.
Environ. Sci. Tech. 24(1):10-15.
Battarbee, R.W. 1981. Changes in the Diatom Microflora of a Eutrophic Lake Since
1900 from a Comparison of Old Algal Samples and the Sedimentary Record.
Holoarctic Ecol. 4:73-81.
Batterbee, R.W., R.J. Flower, A.C. Stevenson, V.C. Jones, R. Harriman, and P.G.
Appleby. 1988. Diatom and Chemical Evidence for Reversibility of Acidification of
Scottish Lochs. Nature 332:530-532.
Baylis, J.R. and H.H. Gerstein. 1929. Micro-organisms in the Lake Michigan Water at
Chicago. Munic. News Water Works. 76:291-296.
Beeton, A.M. 1965. Eutrophication of the St. Lawrence Great Lakes. Limnol.
Oceanogr. 10:240-254.
Beeton, A.M. 1961. Environmental Changes in Lake Erie. Trans. Amer. Fish. Soc.
90:153-159.
Beeton A.M. and D.C. Chandler. 1963. The St. Lawrence Great Lakes. D.G. Frey
(ed.), Limnology in North America, University of Wisconsin Press, Madison, Wl:535-
588.
Bennett, J.R. 1974. On the Dynamics of Wind-Driven Lake Currents. J. Physiol.
Oceanogr. 4:400-414.
Berst, A.M. and G.R. Spangler. 1973. Lake Huron: The Ecology of the Fish
Community and Man's Effects on It. Great Lakes Fish. Comm. Tech. Rep. No. 21.
Bierman, V.J. Jr., D.M. Dolan, and R. Kasprzyk. 1984. Retrospective Analysis of the
Response of Saginaw Bay, Lake Huron, to Reduction in Phosphorus Loadings.
Environ. Sci. Technol. 18:23-31.
Birks, H.J.B., J.M. Line, S. Juggins, A.C. Stevenson, and C.J.F. ter Braake. 1990.
Diatoms and pH Reconstructions. Phil. Trans. Royal Soc. (London), Series B.
327:263-278.
Borgmann, U., K.M. Ralph, and W.P. Norwood. 1989. Toxicity Test Procedures for
Hyalella azteca, and Chronic Toxicity of Cadmium and Pentachlorophenol to H.
azteca, Gammarus fasciatus, and Daphnia magna. Arch. Environ. Contam. Toxicol.
18:756-764.
Ref-2
-------
Bowers, J.A., R. Rossmann, J. Barres, and W.Y.B. Chang. 1986> Phytoplankton
Populations of Southeast Lake Michigan, 1974-1982. ]n: Impact of the Donald C. , "
Cook Nuclear Plant, R. Rossmann (ed.), Univ. Michigan, Great Lakes Res. Div. Publ.
No. 22, Ann Arbor, Ml: 141-168.
Bradbury, J.P. 1986. Effects of Forest Fire and Other Disturbances of Wilderness
Lakes in Northeastern Minnesota II: Paleolimnology. Arch. Hydrobiol. 106:203-217.
Brandt, S.B., D.M. Mason, E.V. Patrick, R.L. Argyle, L. Wells, P.A. Unger, and D.J.
Stewart. 1991. Acoustic measures of the abundance and size of pelagic planktivores
in Lake Michigan. Can. J. Fish. Aquat. Sci. 48:894-908.
Brezonik, P.L 1984. Trophic State Indices: Rationale for Multivariate Approaches.
In: Proc. 3rd Ann. Conf. North Amer. Lake Manag. Soc., Knoxville, TN. EPA 440/5-
84/001, U.S. Environmental Protection Agency, Washington, DC:441-445.
Brezonik, P.L. 1976. Trophic Classifications and Trophic State Indices: Rationale,
Progress, Prospects. Rept. No. ENV-07-76-01, Dept. Engin. Sci., Univ. Florida.
Briggs, S.A. 1972. The Diatomaceae of Lake Michigan. The Lens 1:41-44.
Cairns, J. Jr., P.V. McCormick, and B.R. Niederlehner. 1991. A Proposed Framework
for Developing Indicators of Ecosystem Health for the Great Lakes Region. VA
Polytechnic Institute and State Univ., Blacksburg, VA, A report submitted to the
Internat. Joint Comm. Windsor, Ont.:52 pp.
Carlson, R.E. 1977. A Trophic State Index for Lakes. Limnol. Oceanogr. 22:361-369.
Chandler, J. and M. Vechsler. 1991. United States Legislation Focuses on Great
Lakes Priorities. FOCUS on International Joint Commission Activities 16(1):10-11.
Charles, D.F., R.W. Battarbee, I. Renberg, H. van Dam, and J.P. Smol. 1989.
Paleoecological Studies of Lake Acidification Trends in North America and Europe
Using Diatoms and Chrysophytes. ]n: Soils, Aquatic Processes, and Lake
Acidification, S.A. Norton, S.E. Lindberg, and A.L. Page (Eds.), Springer-Verlag, New
York, NY:207-276.
Chase, H.H. 1904. Flora Michiganensis: Algae: Diatomaceae: A List of Michigan
Diatomaceae. Ann. Rept. Mich. Acad. Sci. 5:166-169.
Cholnoky, B.J. 1968. Die Okologie der Diatimeen in Binnengewasser. J. Cramer,
Lehre.:699 pp.
Ref-3
-------
Christie, W.J. 1974. Changes in the Fish Species Composition of the Great Lakes.
J. Fish. Res. Bd. Can. 31:827-854.
Christie, W.J. 1973. A Review of the Changes in the Fish Species Composition of
Lake Ontario. Great Lakes Fish. Comm. Tech. Rep. No. 23.
Christie, W.J. 1972. Lake Ontario: Effects of Exploitation, Introductions, and
Eutrophication on the Salmonid Community. J. Fish. Res. Bd. Can. 29:91-929.
Colborn, I.E., A. Davidson, S.N. Green, R.A. Hodge, C.I. Jackson, and R.A. Liroff.
1990. Great Lakes, Great Legacy? The Conservation Foundation, Washington, DC
and the Institute for Research on Public Policy, Ottawa, Ontario:301 pp.
Connolly, J.P. and R.V. Thomann. 1982. Calculated Contribution of Surface
Microlayer PCB to Contamination of Lake Michigan Lake Trout. J. Great Lakes Res.
8(2):367-375.
Cook, D.G. and M.G. Johnson. 1974. Benthic Macroinvertebrates of the St.
Lawrence Great Lakes. J. Fish. Res. Board Can. 31:763-782.
Cowardin, L.M., V. Carter, F.C. Golet, and E.T. LaRoe. 1979, Classification of
Wetlands and Deepwater Habitats of the United States. FWS/OBS-79/31, US Fish and
Wildlife Serv.:103 pp.
Daily, W.A. 1938. A Quantitative Study of the Phytoplankton of Lake Michigan
Collected in the Vicinity of Evanston, Illinois. Butler Univ. Bot. Stud. 4:65-83.
Damman, K.E. 1966. Plankton Studies of Lake Michigan. III. Seasonal Periodicity of
Total Plankton, jn.: Proc. 9th Conf. Great Lakes Res., Great Lakes Res. Div. Publ.
No. 15, Univ. Michigan, Ann Arbor, Ml:9-17.
Damman, K.E. 1960. Plankton Studies of Lake Michigan. II. Thirty-three Years of
Continuous Plankton and Coliform Bacteria Data Collected from Lake Michigan at
Chicago, Illinois. Trans. Amer. Microsc. Soc. 79:397-404.
Damman, K.E. 1945. Plankton Studies of Lake Michigan. I. Seventeen Years of
Plankton Data Collected at Chicago, Illinois. Amer. Midi. Nat. 34:769-796.
Damman, K.E. 1941. Quantitative Study of the Phytoplankton of Lake Michigan at
Evanston, Illinois. Butler Univ. Bot. Stud. 5:27-44.
Davis, C.C. 1966. Plankton Studies in the Largest Great Lakes of the World with
Special Reference to the St. Lawrence Great Lakes of North America. Univ. Michigan,
Great Lakes Res. Div. Publ. No. 14, Ann Arbor, Ml: 1-36.
Ref-4
-------
Davis, C.C. 1965. The Standing Stock of Phytoplankton in Lake Erie at Cleveland,
Ohio, 1964. Inform. Bull. Planktol. Japan 12:51-53.
Davis, C.C. 1964. Evidence for the Eutrophication of Lake Erie from Phytoplankton
Records. Limnol. Oceanogr. 9(3):275-283.
Dillon, PJ. and F.H. Rigler. 1974. The Phosphorus-Chlorophyll Relationship. Limnol.
Oceanogr. 19:767-773.
Dixit, S.S., A.S. Dixit, and R.D. Evans. 1988. Sedimentary Diatom Assemblages and
Their Utility in Computing Diatom-Inferred pH in Sudbury Ontario Lakes. Hydrobiol.
169:135-148.
Dixit, S.S., J.P. Smol, J.C. Kingston, and D.F. Charles. 1992. Diatoms: Power
Indicators of Environmental Change. Environ. Sci. Technol. 26(1):22-23.
Dobson, H.F., M. Gilbertson, and P.G. Sly. 1974. A Summary and Comparison of
Nutrients and Related Water Quality in Lakes Erie, Ontario, Huron, and Superior. J.
Fish. Res. Bd. Can. 31:731-738.
Duncan, G.J. and G. Kalton. 1987. Issues of Design and Analysis of Surveys Across
Time. International Statistical Review 55:97-117.
Duthie, H.C. and M.R. Sreenivasa. 1971. Evidence for the Eutrophication of Lake
Ontario from the Sedimentary Diatom Succession. |n: Proc. 14th Conf. Great Lakes
Res., Internat. Assoc. Great Lakes Res.: 1-13.
Dworsky, LB. and D.J. Allee. 1988. An Agenda for the Management of the Great
Lakes on a Long-Term Ecosystem Basis. ]n: D.H. Hickcox (ed.), Proceedings of the
Symposium on the Great Lakes: Living with North America's Inland Waters, Am. Wat.
Res. Assoc., Bethesda, MD:296 pp.
Eddy, S. 1934. A Study of Fresh-Water Plankton Communities. III. Biol. Monogr.
12:1-93.
Eddy, S. 1927. The Plankton of Lake Michigan. III. Nat. Hist. Surv. Bull. 17:302-232.
Edwards C.J. and R.A. Ryder (Eds). 1990. Biological Surrogates of Mesotrophic
Ecosystem Health in the Laurentian Great Lakes. Mesotrophic Indicators Work Group
Report to the Science Advisory Board of the International Joint Commission, Windsor,
Ont.
Ref-5
-------
Edwards, C.J., R.A. Ryder, and T.R. Marshall. 1990. Using Lake Trout as a
Surrogate of Ecosystem Health for Oligotrophic Waters of the Great Lakes. J. Great
Lakes Res. 16(4):591-608.
Ehrenberg, C.G. 1854-1856. Zur Mikrogeologie, das Erden und Felsen Schaffende
Wirken des Unsichtbar Kleien Selbststandigen Lebens auf der Erde. Texte.:374 pp.
(1854); Altas, 40 pis. (1854); Fortxetz.:88 pp. (1856). Leopold Voss, Leipzig.
Eschmeyer, P.M. 1957. The Near Extinction of the Lake Trout in Lake Michigan.
Trans. Am. Fish Soc. 85:102-119.
Fahnenstiel, G.L., L Sicko-Goad, D. Scavia, and E.F. Stoermer. 1986. Importance of
Picoplankton in Lake Superior. Can. J. Fish. Aquat. Sci. 43:235-240.
Federal Interagency Committee for Wetland Delineation. 1989. Federal Manual for
Identifying and Delineating Jurisdictional Wetlands. US Army Corps of Engineers, US
Environmental Protection Agency, US Fish and Wildlife Service, USDA Soil
Conservation Service, Washington, DC, Cooperative Technical Publication:76 p. plus
Appendices.
Findley, R.W. and D.A. Farber. 1988. Environmental Law in a Nutshell. Second
Edition, West Publ. Co., St. Paul, MN:367 pp.
Forbes, S.A. 1883. The Food of the Smaller Fresh-Water Fishes. Bull. III. State Lab.
Nat. Hist. 1:65-94.
Frederick, V.R. 1981. Preliminary Investigation of the Algal Flora in the Sediments of
Lake Erie. J. Great Lakes Res. 7(4):404-408.
Freitag, R., P. Fung, J.S. Mothersill, and G.K. Prouty. 1976. Distribution of Benthic
Macroinvertebrates in Canadian Waters of Northern Lake Superior. J. Great Lakes
Res. 2(1):177-192.
Fritz, S.C. and R.E. Carlson. 1982. Stratigraphic Diatom and Chemical Evidence for
Acid Strip-Mine Lake Recovery. Water, Air, Soil Poll. 17:151-163.
Gilbert, R.O. 1987. Statistical Methods for Environmental Pollution Monitoring. Van
Nostrand Reinhold Co., NY.
Giesy, J.P., C.J. Rosiu, R.L Graney, and M.G. Henry. 1990. Benthic Invertebrate
Bioassays with Toxic Sediment and Porewater. Environ. Toxicol. Chem. 9:233-248.
Ref-6
-------
Glover, P.M. 1982. Diatom Fragmentation in Grand Traverse Bay, Lake Michigan
and Its Implications for Silica Cycling. Ph.D. Dissertation, Univ. of Michigan, Ann
Arbor, M 1:204 pp.
Goodier, J.L 1981. Native Lake Trout (Salvelinus namaycush) Stocks in the
Canadian Waters of Lake Superior Prior to 1955. Can. J. Fish. Aquat. Sci. 38:1724-
1737.
Great Lakes Water Quality Board. 1987. Report on Great Lakes Water Quality.
Presented to the Internal. Joint Comm., Windsor, Ont.
Gregor, D.J. and W. Past. 1979. Trophic Characterization of the U.S. and Canadian
Nearshore Zones of the Great Lakes. Report to Pollution from Land Use Activities
Reference Group, Internat. Joint Comm., Windsor, Ont.:38 pp.
Griffith, R.E. 1955. Analysis of Phytoplankton Yields in Relation to Certain Physical
and Chemical Factors of Lake Michigan. Ecology 36:543-552.
Grover, J.P. 1989. Phosphorus Dependent Growth Kinetics of 11 Species of
Freshwater Algae. Limnol. Oceanogr. 34:341-348.
Hartman, W.L 1988. Historical Changes in the Major Fish Resources of the Great
Lakes. In: Toxic Contaminants and Ecosystem Health: A Great Lakes Focus. M.S.
Evans (ed.), John Wiley & Sons, Inc.: 103-131.
Hartman, W.L. 1973. Effects of Exploitation, Environmental Changes and New
Species on the Fish Habitats and Resources of Lake Erie. Great Lakes Fish. Comm.
Tech. Rep. 22:43 pp.
Harris, G.P. and R.A. Vollenweider. 1982. Paleolimnological Evidence of Early
Eutrophication in Lake Erie. Can. J. Fish. Aquat. Sci. 39:618-626.
Herdendorf, C.E., S.M. Hartley, and M.D. Barnes (Eds). 1981. Fish and Wildlife
Resources of the Great Lakes Coastal Wetlands Within the United States. US Fish
and Wildlife Service, FWS/OBS-81/02, Vol. 1-6.
Hiltunen, J.K. and B.A. Manny. 1982. Distribution and Abundance of Macrobenthos
in the Detroit River and Lake St. Clair, 1977. Administrative Report No. 82-2, U.S.
Fish Wildl. Sen/., Natl. Fish. Res. Cen. Great Lakes, Ann Arbor, Ml.
Hirsch, R.M., J.R. Slack, and R.A. Smith. 1982. Techniques of Trend Analysis for
Monthly Water Quality Data. Water Resources Research 18:107-121.
Ref-7
-------
Hodgson, R.D. and L.M. Alexander. Undated. Towards an Objective Analysis of
Special Circumstances: Bays, Rivers, Coastal and Oceanic Archipelagos and Atolls.
Occasional Paper No. 13, The Law of the Sea Institute, Univ. of Rhode Island,
Kingston, Rl:54 pp.
Hohn, M.H. 1969. Qualitative and Quantitative Analyses of Plankton Diatoms, Bass
Island Area, 1938-1965, Including Synoptic Surveys of 1960-1963. Ohio Biol. Surv.
Holland, R.E. 1980. Seasonal Fluctuations of Major Diatom Species at Five Stations
Across Lake Michigan, May 1970-October 1972. U.S. Environmental Protection
Agency, Office of Research and Development, EPA-600/3-80-006, Duluth, MN:85 pp.
Holland, R.E. 1969. Seasonal Fluctuations of Lake Michigan Diatoms. Limnol.
Oceanogr. 14:423-436.
Holland, R.E. and LW. Clafkin. 1975. Horizontal Distribution of Plankton Diatoms in
Green Bay, Mid-July 1970. Limnol. Oceanogr. 20:365-378.
Holm, N.P. and D.E. Armstrong. 1981. Role of Nutrient Limitation and Competition in
Controlling the Populations of Asterionella formosa and Microsystis aeruginosa in
Semicontinuous Culture. Limnol. Oceanogr. 26:622-634.
Horvitz, D.G. and D.J. Thompson. 1952. A Generalization of Sampling Without
Replacement from a Finite Universe. J. Amer. Statist. Assoc. 47:663-685.
Hunsaker, C.T. and D.E. Carpenter (Eds.). 1990. Ecological Indicators for the
Environmental Monitoring and Assessment Program. EPA-600/3-90/060, United States
Environmental Protection Agency, Office of Research and Development, Research
Triangle Park, NC:143 pp.
Hughes, R.M. 1989. Ecoregional Biological Criteria. Water Quality Standards for the
21st Century. United States Environmental Protection Agency, Office of Water,
Washington, DC:147-151.
Hughes, R.M., D.P. Larsen, and J.M. Omernik. 1986. Regional Reference Sites: A
Method for Assessing Stream Potentials. Envir. Manag. 10:629-635.
Hunt, D.T.E. and A.L. Wilson. 1986. The Chemical Analysis of Water. Second
Edition. Royal Society of Chemistry, London, England:683 pp.
Hustedt, F. 1930. Bacillariophyta (Diatomeae.) |n_: Die Susswasser-Flora
Mitteleuropas, A. Pacher (ed.), Gustav Fischer, Jena:466 pp.
Ref-8
-------
Hutchinson, G.E. 1967. A Treatise on Limnology. II. Introduction to Lake Biology
and the Limnoplankton. John Wiley and Sons, Inc., New York, NY:1115 pp.
Ingersoll, C.G. and M.K. Nelson. 1990. Testing Sediment Toxicity with Hyalella
azteca (Amphipoda) and Chironomus riparius (Diptera). In: Landis, W.G. and van der
Schalie, W.H. (eds.). Aquatic Toxicology and Risk Assessment: Thirteenth volume,
ASTM STP 109b., American Society for Testing and Materials, Philadelphia, PA:93-
109.
International Joint Commission. 1987. A Plan for Assessing Atmospheric Deposition
to the Great Lakes, Report and Scientific Background. The Report of the Atmospheric
Deposition Monitoring Task Force to the Surveillance Work Group, Water Quality
Board, IJC, Windsor, Ont:31 pp., 107 pp.
International Joint Commission. 1975. Report on Great Lakes Water Quality.
International Joint Commission, Windsor, Ontario.
International Joint Commission. 1973. Report on Great Lakes Water Quality.
International Joint Commission, Windsor, Ontario.
International Joint Commission, United States, and Canada. 1989. Revised Great
Lakes Water Quality Agreement of 1978, as amended by Protocol signed November
18, 1987. International Joint Commission, Windsor, Ont.:84 pp.
Karr, J.R. and D.R. Dudley. 1981. Ecological Perspective on Water Quality Goals.
Envir. Manag. 55:55-68.
Kendall, M.G. 1975. Rank Correlation Methods. 4th ed. Charles Griffin, London.
Kingston, J.C. and H.J.B. Birks. 1990. Dissolved Organic Carbon Reconstructions
from Diatom Assemblages in PIRLA Project Lakes, North America. Phil. Trans. Royal
Soc. (London), Series B. 327:279-288.
Kirchner, C.J. 1983. Quality Control in Water Analyses. Env. Sci. Tech. 17(4):174A-
181 A.
Koelz, W. 1929. Coregonid Fishes of the Great Lakes. Bull. U.S. Bur. Fish. 43(2): 1-
643.
Kofoid, C.A. 1896. Report on the Protozoa. Appendix II. Bull. Mich. Fish. Comm.
6:76-84.
Kolkwitz, R. and M. Marsson. 1908. Okologie der Pflanzlichen Saprobien. Ber. Bot.
Ges. (Stuttgart) 26:505-519.
Ref-9
-------
Kreis, R.G. Jr., E.F. Stoermer, and T.B. Ladewski. 1985. Phytoplankton Species
Composition, Abundance, and Distribution in Southern Lake Huron, 1980; Including a
Comparative Analysis with Conditions in 1974 Prior to Nutrient Loading Reductions.
Univ. Michigan, Great Lakes Res. Div. Spec. Rept. No. 107, Ann Arbor, Ml:377 pp.
Lackey, J.B. 1944. Quality and Quantity of Plankton in the South End of Lake
Michigan in 1942. J. Amer. Water Works Assoc. 36:669-674.
Lark, J.G.I. 1973. An Early Record of the Sea Lamprey (Petromyzon marinus) from
Lake Ontario. J. Fish. Res. Bd. Canada 30:131-133.
Lawrie, A.M. and J.F. Rahrer. 1973. Lake Superior: A Case History of the Lake and
Its Fisheries. Great Lakes Fish. Comm. Tech. Rep. No. 19.
Leibowitz, N.C., L. Squires, and J.P. Baker. 1991. Research Plan for Monitoring
Wetland Ecosystems. EPA-600/3-91/010, U.S. Environmental Protection Agency:
157pp.
Leighton, M.O. 1907. Pollution of Illinois and Mississippi Rivers by Chicago Sewage.
U.S. Geol. Surv. Water-Supply Pap. 194:1-369.
Likens, G.E. 1975. Primary Productivity of Inland Aquatic Ecosystems. ]n: Primary
Productivity of the Biosphere, H. Lieth and R.H. Whittaker (Eds.), Springer-Verlag:185-
202.
Limno-Tech. 1985. Summary of Existing Status of the Upper Great Lakes
Connecting Channels Data. Unpublished manuscript.
Loftis, J.C., R.C. Ward, R.D. Phillips, and C.H. Taylor. 1989. An Evaluation of trend
Detection Techniques for Use in Water Quality Monitoring Programs. EPA/600/3-
89/037, U.S. Environmental Protection Agency, Environmental Research Laboratory,
Corvallis, OR.
Loftus, K.H. 1958. Studies on River-Spawning Populations of Lake Trout in Eastern
Lake Superior. Trans. Am. Fish. Soc. 87:259-277.
Lowe, R.L. 1974. Environmental Requirements and Pollution Tolerance of
Freshwater Diatoms. U.S. Environmental Protection Agency, EPA-670/4-74-005,
NERC, Cincinnati, OH:275 pp.
Makarewicz, J.C. 1988. Phytoplankton and Zooplankton in Lakes Erie, Huron, and
Michigan: 1984. EPA-905/3-88-001, U.S. Environmental Protection Agency, Great
Lakes National Program Office, Chicago, IL:275 pp.
Ref-10
-------
Mann.H.B. 1945. Non-parametric'tests against trend. Econometrica 13:245-259.
Manny, B.A., J.K. Hiltunen, D.W. Schloesser, and J.H. Judd. 1988. Changes in the
Bottom Fauna of Western Lake Erie from 1961 to 1982. U.S. Fish Wildl. Serv., Natl.
Fish. Res. Cent. Great Lakes, Ann Arbor, Ml, unpublished manuscript.
Marzolf, G.R. 1964. Substrate Relations of Pontoporeia affinis (Amphipoda) in Lake
Michigan. Verh. Internal. Verein. Limnol. 15:864.
McNaught, D.C. 1982. Short Cycling of Contaminants by Zooplankton and Their
Impact on Great Lakes Ecosystems. J. Great Lakes Res. 8(2):360-366.
Michigan State University Bulletin Office, E-1866, 1867,1868, 1869, 1870, Extension
Bulletins, 10-B Agriculture Hall, Lansing, Ml.
Minister of the Environment. 1990. Second Report of Canada under the 1987
Protocol to the 1978 Great Lakes Water Quality Agreement. Technical Summary.
Prepared by the Government of Canada and the Province of Ontario, Cat. No. En40-
11/23-1990-2E, Minister of Supply and Services, Canada:180 pp.
Mount, D.I. and T.J. Norberg. 1984. A Seven-Day Life-Cycle Cladoceran Test.
Environ. Toxicol. Chem. 3:425-434.
Munawar, M. and I.F. Munawar. 1982. Phycological Studies in Lakes Ontario, Erie,
Huron, and Superior. Can. J. Bot. 60:1837-1858.
Munawar, M. and I.F. Munawar. 1976. A Lakewide Study of Phytoplankton Biomass
and Its Species Composition in Lake Erie, April-December, 1970. J. Fish. Res. Bd.
Can. 33(3):581-600.
Munawar, M. and I.F. Munawar. 1975. The Abundance and Significance of
Phytoflagellates and Nanoplankton in the St. Lawrence Great Lakes: I.
Phytoflagellates. Verh. Internat. Verein. Limnol. 19:705-723.
Nalepa, T.F. 1991. Status and Trends of the Lake Ontario Macrobenthos. Can. J.
Fish. Aquat. Sci. 48:1558-1567.
Nalepa, T.F. 1987. Long-Term Changes in the Macrobenthos of Southern Lake
Michigan. Can. J. Fish. Aquat. Sci. 44:515-524.
Nebeker, A.V., M.A. Cairns, J.H. Gakstatter, K.W. Malueg, G.S. Schuytema, and D.F.
Krawczyk. 1984. Biological Methods for Determining Toxicity of Contaminated
Freshwater Sediments to Invertebrates. Environ. Toxicol. Chem. 3:617-630.
Ref-11
-------
Neilson, M.A. and R.J.J. Stevens. 1987. Spatial Heterogeneity of Nutrients and
Organic Matter in Lake Ontario. Can. J. Fish. Aquat. Sci. 44:2192-2203.
Nicholls, K.H., D.W. Standen, and G.J. Hopkins. 1980. Recent Changes in the
Nearshore Phytoplankton of Lake Erie's Western Basin at Kingsville, Ontario. J. Great
Lakes Res. 6:146-153.
Overton, W.S. 1987a. Phase II Analysis Plan, National Lake Survey - Working Draft.
April 1987. Technical Report 115, Dept. of Statistics, Oregon State University,
Corvallis, OR.
Overton, W.S. 1987b. A Sampling and Analysis Plan for Streams, in the National
Surface Water Survey conducted by EPA. Technical Report 117, Dept. of Statistics,
Oregon State University, Corvallis, OR.
Overton, W.S. and S.V. Stehman. 1987. An Empirical Investigation of Sampling and
Other Errors in the National Stream Survey: Analysis of a Replicated Sample of
Streams. Technical Report 119, Dept. of Statistics, Oregon State University, Corvallis,
OR.
Overton, W.S., D. White, and D.L. Stevens. 1991. Design Report for EMAP, The
Environmental Monitoring and Assessment Program. EPA/600/3-91/053, U.S. EPA,
Washington, DC.
Palmer, C.M. 1969. A Composite Rating of Algae Tolerating Organic Pollution. J.
Phycol. 5:78-82.
Palmer, M. and G. Warren. 1992. Interim Quality Assurance Project Plan. Great
Lakes Survey Studies of Lakes Michigan, Huron, Erie, Ontario, and Superior. April
1992-February 1993. U.S. Environmental Protection Agency, Great Lakes National
Program Office, Chicago, IL.
Parker, J.I. and D.N. Edgington. 1976. Concentration of Diatom Frustules in Lake
Michigan Sediment Cores. Limnol. Oceanogr. 21:887-893.
Patrick, R. and C.W. Reimer. 1975. The Diatoms of the United States. Vol. 2, Part 1,
Acad. Nat. Sci., Philadelphia Monograph No. 13, Philadelphia, PA:213 pp.
Patrick, R. and C.W. Reimer. 1966. The Diatoms of the United States. Vol. 1, Acad.
Nat. Sci., Philadelphia Monograph No. 13, Philadelphia, PA:688 pp.
Pennington, W. 1943. Lake Sediments: The Bottom Deposits of the North Basin of
Windermere with Special Reference to the Diatom Succession. New Phytol. 43:1-27.
Ref-12
-------
Peterson, R.R. 1975. A Paleolimnological Study of the Eutrophication of Lake Erie.
Verh. Internat. Verein. Limnol. 19:2274-2283.
Pratt, D. 1988. Distribution and Population Status of the Ruffe (Gymnocephalus
cernua) in the St. Louis Estuary and Lake Superior. Research Completion Report
presented to the Great Lakes Fishery Commission, Ann Arbor, Ml.
Rast, W. and G.F. Lee. 1978. Summary Analysis of the North American OECD
Eutrophication Project: Nutrient Loading-Lake Response Relations and Trophic State
Indices. EPA-600/3-78-008, U.S. Environmental Protection Agency.
Rathke, D.E. 1984. Lake Erie Intensive Study 1978-1979. Center for Lake Erie Area
Research (CLEAR). The Ohio State University, Columbus, OH, CLEAR Tech. Rept.
No. 284.
Rathke, D.E. and G. McRae. 1990. 1987 Report on Great Lakes Water Quality.
Appendix B: Great Lakes Surveillance (3 volumes). International Joint Commission,
Windsor, Ontario.
Rawson, D.S. 1956. Algal Indicators of Trophic Lake Types. Limnol. Oceanogr.
1(1): 18-25.
Reid, J.R. 1961. Investigation of Bottom Cores from North and South-Central Lake
Superior, jn.: Proc. 4th Conf. Great Lakes Res., Univ. Michigan, Great Lakes Res.
Div. Pub. 7, Ann Arbor, Ml: 126-144.
Reynoldson, T.B. and K.E. Day. 1991. A Study Plan for the Development of
Biological Sediment Quality Guidelines. National Water Research Institute, Burlington,
Ont.:37 pp.
Rhee, G-Y. and I.J. Gotham. 1980. Optimum N:P Ratios and Coexistence of
Planktonic Algae. J. Phycol. 16(4):486-489.
Rodgers, P.W. and W.R. Swain. 1983. Analysis of Polychlorinated Biphenyl (PCB)
Loading Trends in Lake Michigan. J. Great Lakes Res. 9(4):548-558.
Rojanschi, V., L. Stefanescu, and I. lacob. 1991. Possibilities for a Global Evaluation
of the Ecosystems Quality. Seminar on Ecosystems Approach to Water Management,
Oslo, Norway, May 27-31, 1991.
Rosa, F. 1987. Lake Erie Central Basin Total Phosphorus Trend Analysis from 1968
to 1982. J. Great Lakes Res. 13:667-673.
Ref-13
-------
Rosiu, C.J., J.P. Giesy, and R.G. Kreis, Jr. 1989. Toxicity of Vertical Sediments in
the Trenton Channel, Detroit River, Michigan, to Chironomus tentans (Insecta:
Chironomidae). J. Great Lakes Res. 15(4):570-580.
Ryder, R.A. 1972. The Limnology and Fishes of Oligotrophic Glacial Lakes in North
America (about 1800 A.D.). J. Fish. Res. Bd. Canada 29:617-628.
Ryder, R.A. and C.J. Edwards. 1985. A Conceptual Approach for the Application of
Biological Indicators of Ecosystem Quality in the Great Lakes Basin. Report to the
Great Lakes Science Advisory Board of the International Joint Commission, Windsor,
Ont.
Ryder, R.A. and S.R. Kerr. 1978. "The Adult Walleye in the Percid Community--A
Niche Definition Based on Feeding Behavior and Food Specificity." hi: R.L. Kendall,
(ed.), Selected Coolwater Fishes of North America, Am. Fish. Soc. Spec. Publ.,
Bethesda, MD 11:39-51.
Ryder, R.A., S.R. Kerr, W.W. Taylor, and P.A. Larkin. 1981. Community
Consequences of Fish Stock Diversity. Can. J. Fish. Aquat. Sci. 38:1856-1866.
Sakamoto, M. 1966. Primary Production by Phytoplankton Community in some
Japanese Lakes and Its Dependence on Lake Depth. Arch. Hydrobiol. 62:1-28.
Scavia, D. and J.R. Bennet. 1980. Spring Transition Period in Lake Ontario~A
Numerical Study of the Causes of the Large Biological and Chemical Gradients. Can.
J. Fish. Aquat. Sci. 37:823-833.
Schelske, C.L. 1980. Dynamics of Nutrient Enrichment in Large Lakes: The Lake
Michigan Case. ]n,: Restoration of Lakes and Inland Waters. Internal. Symp. on
Inland Waters and Lake Restoration, EPA-440/5-81-010, U.S. EPA, Office of Water
Regulations and Standards, Washington, DC.
Schelske, C.L. 1975. Silica and Nitrate Depletion as Related to Rate of
Eutrophication in Lakes Michigan, Huron, and Superior. \rr. Coupling of Land and
Water Systems, A.D. Hasler (ed.), Ecological Series, Vol. 10, Springer-Verlag, Inc.,
New York, NY:277-298.
Schelske, C.L, LE. Feldt, and M.S. Simmons. 1980. Phytoplankton and Physical-
Chemical Conditions in Selected Rivers and the Coastal Zone of Lake Michigan, 1972.
Univ. Michigan, Great Lakes Res. Div. Publ. No. 19, Univ. Michigan, Ann Arbor,
Ml: 162 pp.
ReM4
-------
Schelske, C.L., R.A. Moll, T.D. Berry, and E.F. Stoermer. 1983. Limnological
Characteristics of Northern Lake Michigan, 1976. Univ. Michigan, Great Lakes Res.
Div. Spec. Rept. No. 95, Univ. Michigan, Ann Arbor, Ml:245 pp.
Schelske, C.L., E.F. Stoermer, J.E. Gannon, and M.S. Simmons. 1976. Biological,
Chemical, and Physical Relationships in the Straits of Mackinac. EPA-600/3-76-095,
U.S. Environmental Protection Agency, Office of Research and Development, Duluth,
MN:267 pp.
Schneider, J.C., F.F. Hooper, and A.M. Beeton. 1969. The Distribution and
Abundance of Benthic Fauna in Saginaw Bay, Lake Huron. Proc. 12th Conf. Great
Lakes Res.:80-90.
Shannon, E.E. and P.L. Brezonik. 1972. Eutrophication Analysis: A Multivariate
Approach. J. Sanit. Eng. Div., Amer. Soc. Civil Eng. 98:37-57.
Shiomi, M.T. and V.K. Chawla. 1970. Nutrients in Lake Ontario. Proc. Conf. Great
Lakes Res. 13:715-732.
Skvortzow, B.V. 1937. Diatoms from Lake Michigan. Amer. Midi. Nat. 18:652-658.
Simons, T.J. 1980. Circulation Models of Lakes and Inland Seas. Can. Bull. Fish.
Aquat. Sci. No. 203:146 pp.
Smith, F., S. Kulkarni, L.E. Myers, and M.J. Messner. 1988. Evaluating and
Presenting Quality Assurance Sampling Data, in: L.H. (ed.), Principles of
Environmental Sampling, Am. Chem. Soc., Washington, DC: 157-168.
Smith, V.H. 1983. Low Nitrogen to Phosphorus Ratios Favor Dominance by Blue-
Green Algae in Lake Phytoplankton. Science 221:669-671.
Smith, V.H. 1982. The Nitrogen and Phosphorus Dependence of Algal Biomass in
Lakes: An Empirical and Theoretical Analysis. Limnol. Oceanogr. 27:1101-1112.
Smol, J.P. 1988. Paleoclimate Proxy Data for Freshwater Arctic Diatoms. Verh.
Internat. Verein. Limnol. 23:837-844.
Stehman, S.V. and W.S. Overton. 1989a. Pairwise Inclusion Probability Formulas in
Random-order, Variable Probability, Systematic Sampling. Technical Report 131,
Dept. of Statistics, Oregon State University, Corvallis, OR.
Stehman, S.V. and W.S. Overton. 1989b. An Empirical, General Population
Assessment of the Properties of Variance Estimators of the Horvitz-Thompson
Ref-15
-------
Estimator under Random-order, Variable Probability Systematic Sampling. Technical
Report 132, Dept. of Statistics, Oregon State University, Corvallis, OR.
Stehman, S.V. and W.S. Overton. 1987a. Estimating the Variance of the
Horvitz-Thompson Estimator in Variable Probability, Systematic Samples.
Proceedings of the Section on Survey Research Methods of the American Statistical
Association.
Stehman, S.V. and W.S. Overton. 1987b. An Empirical Investigation of the Variance
Estimation Methodology Prescribed for the National Stream Survey: Simulated
Sampling from Stream Data Sets. Technical Report 118, Dept. of Statistics, Oregon
State University, Corvallis, OR.
Stevens, D.L, A.R. Olsen, and D. White. In Press. EMAP Sampling Design
Implementation, Perspectives and Issues. EPA/600/X-XX-XXX, U.S. EPA, Corvallis,
OR.
Stevens, R.J.J., M.A. Neilson, and N.D. Warry. 1985. Water Quality of the Lake
Huron - Georgian Bay System. Environment Canada, Inland Waters Directorate
Scientific Series No. 143, Ottawa, Ontario.
Stockner, J.G. and W.W. Benson. 1967. The Succession of Diatom Assemblages in
the Recent Sediments of Lake Washington. Limnol. Oceanogr. 12:513-532.
Stoermer, E.F. 1978. Phytoplankton Assemblages as Indicators of Water Quality in
the Laurentian Great Lakes. Trans. Amer. Micros. Soc. 97(1):2-16.
Stoermer, E.F. 1968. Nearshore Phytoplankton Populations in the Grand Haven,
Michigan, Vicinity During Thermal Bar Conditions, jn; Proc. 11th Conf. Great Lakes
Res., Internat. Assoc. Great Lakes Res.: 137-150.
Stoermer, E.F. 1967. A Historical Comparison of Offshore Phytoplankton Populations
in Lake Michigan, jn.: Studies on the Environment and Eutrophication of Lake
Michigan, J.C. Ayers and D.C. Chandler (Eds.), Univ. Michigan, Great Lakes Res. Div.
Spec. Rept. No. 30:47-77.
Stoermer, E.F., M.M. Bowman, J.C. Kingston, and A.L. Schaedel. 1975.
Phytoplankton Composition and Abundance in Lake Ontario During IFYGL. EPA-
600/3/75-004, U.S. Environmental Protection Agency, Corvallis, OR:373 pp.
Stoermer, E.F., G. Emmert, and C.L Schelske. 1989. Morphological Variation of
Stephanodiscus niagarae Ehrenb. (Bacillariophyta) in a Lake Ontario Sediment Core.
J. Paleolimnol. 2:227-236.
Ref-16
-------
Stoermer, E.F., J.P. Kociolek, C.L Schelske, and N.A. Andersen. 1992. Siliceous
Microfossil Succession in the Recent History of Green Bay, Lake Michigan. J.
Paleolimnol. 6:1-18.
Stoermer, E.F., J.P. Kociolek, C.L. Schelske, and D.J. Conley. 1987. Quantitative
Analysis of Siliceous Microfossils in the Sediments of Lake Erie's Central Basin.
Diatom. Res. 2:113-134.
Stoermer, E.F., J.P. Kociolek, C.L. Schelske, and D.J. Conley. 1985. Siliceous
Microfossil Succession in the Recent History of Lake Superior. Proc. Acad. Nat. Sci.
Philadelphia. 137:106-118.
Stoermer, E.F. and E. Kopczynska. 1967. Phytoplankton Populations in the Extreme
Southern End of Lake Michigan, 1962-63. ]n: Proc. 10th Conf. Great Lakes Res.,
Internat. Assoc. Great Lakes Res.:88-106.
Stoermer, E.F. and R.G. Kreis, Jr. 1980. Phytoplankton Composition and Abundance
in Southern Lake Huron. EPA-600/3-80-061, U.S. Environmental Protection Agency,
Duluth, MN:384 pp.
Stoermer, E.F. and R.G. Kreis, Jr. 1978. Preliminary Checklist of Diatoms
(Bacillariophyta) from the Laurentian Great Lakes. J. Great Lakes Res. 4(2): 149-169.
Stoermer, E.F., C.L. Schelske, and L.E. Feldt. 1971. Phytoplankton Assemblage
Difference at Inshore Versus Offshore Stations in Lake Michigan and Their Effects o'n
Nutrient Enrichment Experiments. ]n: Proc. 14th Conf. Great Lakes Res., Internat.
Assoc. Great Lakes Res.:114-118.
Stoermer, E.F., C.L. Schelske, M.A. Santiago, and L.E. Feldt. 1972. Spring
Phytoplankton Abundance and Productivity in Grand Traverse Bay, Lake Michigan,
1970. ]n: Proc. 15th Conf. Great Lakes Res., Internat. Assoc. Great Lakes Res.:181-
191.
Stoermer, E.F. and L. Sicko-Goad. 1977. A New Distribution Record for
Hymenomonas roseola Stein (Prymnesiophyceae, Coccolithophoraceae) and
Spiniferomonas trioralis Takahashi (Chrysophceae, Synuraceae) in the Laurentian
Great Lakes. Phycologia 16(4):355-358.
Stoermer, E.F. and R.J. Stevenson. 1979. Green Bay Phytoplankton Composition,
Abundance and Distribution. EPA-905/3-79-002, U.S. Environmental Protection
Agency, Great Lakes National Program Office, Chicago, IL:104 pp.
Ref-17
-------
Stoermer, E.F. and M.L Tuchman. 1979. Phytoplankton Assemblages of the
Nearshore Zone of Southern Lake Michigan. EPA-905/3-79-001, U.S. Environmental
Protection Agency, Great Lakes National Program Office, Chicago, IL89 pp.
Stoermer, E.F. and J.A. Wolin. 1990. Siliceous Microfossil Succession in Lake
Michigan. Limnol. Oceanogr. 35(4):959-967.
Stoermer, E.F., J.A. Wolin, C.L. Schelske, and D.J. Conley. 1985a. An Assessment
of Ecological Changes During the Recent History of Lake Ontario Based on Siliceous
Algal Microfossils Preserved in the Sediments. J. Phycol. 21:257-276.
Stoermer, E.F., J.A. Wolin, C.L Schelske, and D.J. Conley. 1985b. Post Settlement
Diatom Succession in the Bay of Quinte, Lake Ontario. Can. J. Fish. Aquat. Sci.
43:754-767.
Stoermer, E.F., J.A. Wolin, C.L Schelske, and D.J. Conley. 1985c. Variation in
Melosira islandica Valve Morphology Related to Eutrophication and Silica Depletion.
Limnol. Oceanogr. 30:414-418.
Stoermer, E.F. and J.J. Yang. 1970. Distribution and Relative Abundance of
Dominant Plankton Diatoms in Lake Michigan. Univ. Michigan, Great Lakes Res. Div.
Publ. No. 16, Univ. Michigan, Ann Arbor, Ml:64 pp.
Stoermer, E.F. and J.J. Yang. 1969. Plankton Diatom Assemblages in Lake
Michigan. Univ. Michigan, Great Lakes Res. Div. Spec. Rept. No. 47, Univ. Michigan,
Ann Arbor, Ml:268 pp.
Stoermer, E.F. and J.J. Yang. 1968. A Preliminary Report of the Fossil Diatom Flora
from Lake Huron Sediments. In: Proc. 11th Conf. Great Lakes Res., Internat. Assoc.
Great Lakes Res.:253-267.
Surveillance Work Group. 1986. Great Lakes International Surveillance Plan (GLISP)
Report to the Great Lakes Water Quality Board. Internat. Joint Comni., Windsor, Ont.
Vol. 1/2.
Suter, G.W. 1990. Endpoints for Regional Ecological Risk Assessments. Environ.
Manage. 14:9-23.
Sweets, P.R. 1983. Differential Deposition of Diatom Frustules in Jellison Hill Pond,
Maine. Masters Thesis, University of Maine, Orono, ME.
Taylor, J.K. 1988. Quality Assurance of Chemical Measurements. Lewis Publ.
Chelsea, Ml:328 pp.
Ref-18
-------
ten Brink, B.J.E., S.H. Hosper, and F. Colijn. 1991. A Quantitative Method for
Description and Assessment of Ecosystems: The Amoeba-Approach. Seminar on
Ecosystems Approach to Water Management, Oslo, Norway, May 1991.
ter Braake, E.J.F. 1989. CANACO - An Extensive of DECORANA to Analyze
Species-Environment Relationships. Hydrobiol. 184:169-170.
ter Braake, E.J.F. 1986. Canonical Correspondence Analysis: A New Eigenvestor
Technique for Multivariate Direct Gradient Analysis. Ecol. 67:1167-1179.
ter Braake, E.J.F. and LG. Barendregt. 1986. Weighted Averaging of Species
Indicator Values: Its Efficiency in Environmental Calibration. Mathemat. Biosci. 78:57-
72.
ter Braake, E.J.F. and H. van Dam. 1989. Inferring pH from Diatoms: A Comparison
of Old and New Calibration Methods. Hydrobiol. 178:209-223.
Thayer, V.L 1981. Diatoms in Lake Superior Sediments: Distribution, Stratigraphy
and Taxonomy. MS Thesis, University of Minnesota, Minneapolis, MN:140 pp.
Thayer, V.L., T.C. Johnson, and H.J. Schrader. 1983. A Preliminary Study of Recent.
Diatom Assemblages in Lake Superior Sediments. J. Great Lakes Res. 9(4):508-516.
Theriot, E.G. and E.F. Stoermer. 1984. Principal Components Analysis in Character
Variation in Stephanodiscus niagarae Ehrenb.: Morphological Variation Related to
Lake Trophic Status. ]n: Proc. Vlllth Internat. Diatom Symp. (D.G. Mann, ed.). Otto
Koeltz, Koenigstein:97-111.
Thomas, B.W. and H.H. Chase. 1887. Diatomaceae of Lake Michigan as Collected
During the Last 16 Years from the Water Supply of the City of Chicago. Notarisia
2(6):328-330.
Thompson, H.D. 1896. Report on the Plants. Bull. Mich. Fish. Comm. 6:72-75.
Thorne, R.E. 1983. Hydroacoustics. In: Fisheries Techniques, LA. Nielson and D.L.
Johnson (eds.), American Fisheries Society, Bethesda, MD:239-259.
Trautman, M.D. 1957. The Fishes of Ohio. Ohio State Univ. Press, Columbus, Ohio.
Tuchman, M.L., E.F. Stoermer, and H.J. Carney. 1984. Effects of Increased Salinity
on the Diatom Assemblages in Fonda Lake, Michigan. Hydrobiol. 109:179-188.
Ref-19
-------
Upper Lakes Reference Group. 1976. The Waters of Lake Huron and Lake Superior.
Vol. 1. Summary and Recommendations. Report to the Internal Joint Comm. by the
Upper Lakes Reference Group, Windsor, Ont.:236 pp.
United States Environmental Protection Agency. 1992. Environmental Monitoring and
Assessment Program. Quality Assurance Management Plan. EPA/600/XX-XX/ XXX,
OMMSQA, Office of Research and Development, US EPA, Washington, DC.
United States Environmental Protection Agency. 1991. Environmental Monitoring and
Assessment Program: Draft EMAP - Near Coastal 1991 Virginia Province Quality
Assurance Project Plan. EPA/600/x91/xxx, Office of Research and Development,
Washington, DC:84 pp.
United States Environmental Protection Agency. 1990a. Environmental Monitoring
and Assessment Program: Guidelines for Preparing Logistics Plans. EPA 600/x-
90/161, United States Environmental Protection Agency, Environmental Monitoring
Systems Laboratory, Las Vegas, NV:31 pp.
United States Environmental Protection Agency. 1990b. Environmental Monitoring
and Assessment Program (EMAP) Overview. EPA/608/9-90/001, US EPA Office of
Research and Development, Office of Modeling Monitoring Systems and Quality
Assurance (OMMSQA), Washington, DC.
United States Environmental Protection Agency. 1989. EPA System Design and
Development Guidance. US EPA Office of Information Resource Management,
Washington, DC.
United States Environmental Protection Agency. 1987. Office of Information
Resource Management Policy Manual (2100), US EPA Office of Resource
Management, Washington, DC.
United States Environmental Protection Agency. 1974. An Approach to a Relative
Trophic Index System for Classifying Lakes and Reservoirs. Working Paper No. 24,
National Eutrophication Survey, Pacific Northwest. Environmental Research
Laboratory, Corvallis, Oregon:44 pp.
United States Environmental Protection Agency and Environment Canada. 1988. The
Great Lakes. An Environmental Atlas and Resource Book. EPA-905/9-87-002,
GLNPO No. 2, Great Lakes National Program Office, U.S. Environmental Protection
Agency, Chicago, IL44 pp.
Urquhart, N.S, W.S. Overton, and D.S. Birkes. 1991. Comparing Sampling Designs
for Monitoring Ecological Status and Trends: Impact of Temporal Patterns. Technical
Report #49, Department of Statistics, Oregon State University, Corvallis, OR.
Ref-20
-------
Van Landingham, S.L. 1982. Guide to the Identification, Environmental
Requirements, and Pollution Tolerance of Freshwater Blue-Algae (Cyanophyta.) U.S.
Environmental Protection Agency, Office of Research and Development, Cincinnati,
OH.
Van Oosten, J. 1937. The Dispersal of Raiabow Smelt, Osmerus mordax (Mitchell),
in the Great Lakes Region. Trans. Am. Fish. Soc. 60:204-214.
Vaughn, J.C. 1962. Operation of the South District Filtration Plankton, 1961. Pure
Water 14(2):34-49.
Vaughn, J.C. 1961. Coagulation difficulties of the South District Filtration Plant. Pure
Water 13(3):45-49.
Verduin, J. 1964. Changes in Western Lake Erie During the Period 1948-1962. Verh.
Internat. Verin Limnol. 15:639-644.
Vollenweider, R.A. 1976. Advances in Defining Critical Loading Levels for
Phosphorus in Lake Eutrophication. Mem. 1st. Ital. Idrobiol. 33:53-83.
Vollenweider, R.A., M. Munawar, and P. Stadelmann. 1974. A Comparative Review
of Phytoplankton and Primary Production in the Laurentian Great Lakes. J. Fish. Res.
Board Can. 31:739-762.
Ward, H.B. 1896. A Biological Examination of Lake Michigan in the Traverse Bay
Region. Bull. Mich. Fish. Comm. 6:1-71.
Weber, C.I., W.H. Peltier, T.J. Norberg-King, W.B. Horning, F.A. Kessler, J.R.
Menkedick, T.A. Neiheisel, P.A. Lewis, D.J. Klemm, Q.H. Pickering, E.L. Robinson,
J.M. Lazorchak, L.J. Wymor, and R.W. Freyberg. 1989. Short-term methods for
estimating the chronic toxicity of effluents and receiving waters to freshwater
organisms. EPA/600/4-89/001, U.S. Environmental Protection Agency, Office of
Research and Development, EMSL-Cincinnati, OH:249 pp.
Weisberg, S.B., J.B. Frithsen, A.F. Holland, J.F. Paul, KJ. Scott, J.K. Summers, H.T.
Wilson, R. Valente, D.G. Heimbuch, J. Gerritsen, S.C. Schimmel, and R.W. Latimer.
1991. EMAP-Estuaries Virginian Province 1990 Demonstration Project Report. EPA
600/XXX/XXX. U.S. Environmental Protection Agency, Environmental Research
Laboratory, Narragansett, Rl 02882.
Wells, L. and A.L. McClain. 1973. Lake Michigan: Man's Effect on Native Fish
Stocks and Other Biota. Great Lakes Fish. Comm. Tech. Rep. No. 20.
Wetzel, R.G. 1975. Limnology. W.B. Saunders. 743 pp.
Ref-21
-------
White, D.S. 1991. Functional Structure of Large Lake Benthic Communities from the
Shoreline Outward. Paper Presented at 34th Conf. on Great Lakes Research, SUNY
Buffalo, June 2-6, 1991.
White, D.S. Unpublished data on benthic community composition from a 1975 survey
of Lake Michigan. Murray State University, Lexington, KY.
Whitmore, T.J. 1991. Sedimentary Diatom Concentrations arid Accumulation Rates
as Predictors of Lake Trophic State. Hydrobiol. 214:163-169.
Wickware, G.M. and C.D.A. Rubec. 1989. Ecoregions of Ontario. Environment
Canada, Ecological Land Classification Series No. 26:37 pp.
Williams, L.G. 1972. Plankton Diatom Species Biomasses and the Quality of
American Rivers and the Great Lakes. Ecology 53(6): 1038-1050.
Wolin, J.A., E.F. Stoermer, and C.L Schelske. 1991. Recent Changes in Lake
Ontario 1981-1987: Microfossil Evidence of Phosphorus Reduction. J. Great Lakes
Res. 17(2):229-240.
Wolin, J.A., E.F. Stoermer, C.L. Schelske, and D.J. Conley. 1988. Siliceous
Microfossil Succession in Recent Lake Huron Sediments. Arch. Hydrobiol. 114:175-
198.
Yates, F. 1953. Sampling Methods for Censuses and Surveys. 2nd ed. Charles
Griffin and Co., London.
Ref-22
-------
Appendix A.
EMAP Glossary (Draft)1
1 Based on Baillargeon 1991.
-------
EMAP - Great Lakes Glossary
accuracy: The degree of agreement between a measurement or set of measurements and the true
value or an accepted reference value. A set of measurements may be accurate without necessarily
being precise.
adaptive sampling strategy: A sampling approach which, by design, can be modified to meet
changing objectives or circumstances.
analysis: A detailed examination of a problem or entity made in order to understand its nature or to
determine its essential features or character.
Annual Statistical Summary: A document containing summaries of EMAP data collected on a single
EMAP resource type for a given year. Summaries may include cumulative frequency distributions,
estimates of the extent of nominal or subnominal condition, comparisons among regions, or
comparisons of data through time.
anthropogenic: Referring to the influence of human activities on nature.
area frame: A method of defining the sampling frame based on units of land area that in aggregate
comprise the total land area of a region of interest. The individual sampling units are defined with
maps or other cartographic materials. (See frame.)
area sample: A sample from an area frame.
assessment: Interpretation of information, or the assignment of significance or importance to data,
within the context of policy-relevant questions.
assessment endpoint: An explicit rule or set of rules that relates a specific resource sampling unit to
each element of a sampling frame.
attribute: Any property, quality, or characteristic of a sampling unit or site. The indicators and other
measures used to characterize a sampling unit or site are representations of the attributes of that unit
or site.
association rule: An explicit rule or set of rules that relates a specific resource sampling unit to each
element of a sampling frame.
auxiliary data: Data collected by a monitoring or sampling program other than EMAP. Such data may
be on-frame (but not the EMAP frame) or off-frame data. The sampling methods and quality assurance
protocols of auxiliary data must be evaluated before the data are used. The term is synonymous with
he term "non-EMAP data."
baseline condition: The status of a resource of resources as determined by initial sampling events.
Future samples are compared to baseline conditions to infer changes.
A-1
-------
EMAP - Great Lakes Glossary
baseline grid: The fixed position of the EMAP grid as established by the position of the hexagon
overlying the United States. This is distinguished from the sampling grid, which is shifted some random
direction and distance from the baseline grid.
best management practices: Application of techniques targeted at minimizing specific anthropogenic
disturbances, such as soil erosion, pollutant transport, storm water runoff, or similar land-use-related
disturbances.
bias: Systematic error manifested as a consistent deviation (positive or negative) from the known or
true value. It may be regarded as the difference between the conceptual weighted average value of an
estimator over all possible samples and the true value of the quantity being estimated. An estimator is
said to be "unbiased" if this difference is zero. Bias differs from random error, which shows no such
consistent deviation and for which this difference is always zero.
bioaccumulation. A process by which chemicals are taken up by organisms from environmental
media (air, water, soil, or food.)
bioassay: A laboratory or field test in which living organisms are used to detect the presence of or test
the effect of a particular substance, factor, or condition. Results are compared to a standard
preparation or control to determine the relative strength of the substance, factor, or condition.
biodiversity: The variety and variability among living organisms and the ecological complexes in which
they occur. Diversity can be quantified as the number of different items and their relative frequencies.
For biological diversity, these items are organized at many levels, ranging from complete ecosystem to
the biochemical structures that are the molecular basis of heredity. Thus, the term encompasses
expressions of the relative abundances of different ecosystems, species, and genes.
biogeographic province: Geographic areas characterized by specific plant formations and associated
fauna.
biological magnification (or biomagnification): The process by which the concentrations of certain
substances (e.g., radioactive materials and persistent pesticides) in the tissues of living organisms
become greater at higher trophic levels of a food web.
biomarker: A measurement of body fluids, cells, or tissues that indicates in biochemical or cellular
terms the presence and magnitude of toxicants or of host response.
biomass: The amount of organic matter contained in all living organisms per unit area or volume.
Because living organisms contain a large proportion of inorganic material (e.g., water and minerals),
biomass is usually measured in terms of carbon content, and it is usually expressed as a density (mass
per unit area or volume), such as kilograms of carbon per square kilometer or grams of carbon per
cubic meter. It is sometimes convenient to express the mass as its energetic equivalent, such as
kilocalories per square kilometer.
biome: A major class of ecosystem type possessing characteristic flora and fauna which have
developed under and adapted to characteristic climatic regimes. Examples of biomes include tropical
rain forest, tundra, northern boreal forest, or desert.
biosphere: That portion of the earth and its atmosphere that can support life.
bottom-up approach: A risk assessment methodology that uses first principles or experimental results
to assess environmental condition. This is a more traditional form of risk assessment that moves from
A-2
-------
EMAP - Great Lakes Glossary
source(s) and source assessment, through exposure assessment, to effects assessment. For example,
pollutant effects are related causally to pollutant sources by transport, fate, and dose-response models.
(See top-down approach.)
candidate indicator: A potential EMAP indicator proposed for a given resource category or class, or
as a cross-resource indicator, based on a combination of literature review, exchanges at expert
workshops, and interviews with scientists and environmental managers. An indicator with candidate
status is judged against specific EMAP criteria to determine its applicability/usefulness as a research
indicator.
classification: The process of assigning a resource unit to one of a set of groups or categories
defined by values of attributes measured on the resource units. For example, forest sites can be
classified into forest types, depending on the species composition of the forest. The process also may
be hierarchical - a stepwise partitioning of resource units into classes based on specified attributes of
those units. For example, lakes can be distinguished by size - large lakes versus small lakes; small
lakes can further be distinguished by hydrologic lake type - small seepage lakes versus small drainage
lakes.
client: An organization, administrative unit, or other entity with whom EMAP interacts to determine
policy-relevant assessment requirements, to conduct assessments, and to communicate assessment
results. Clients may include other EPA administrative units, other federal agencies, Congress, state
and local governments and their agencies, and private organizations.
cluster analysis: A statistical procedure for classifying data points by combining similar points for form
small classes, then combining small classes into larger classes, and so on.
community: In ecology, a group of interacting populations co-occurring in time and space.
Sometimes, a particular subgrouping may be specified, such as the fish community in lake or the soil
arthropod community in a forest.
confidence interval: An interval within which a particular parameter (e.g., the population mean) lies
with a specified degree of confidence. The interval is bounded by upper and lower confidence limits. A
confidence interval is so termed because the confidence that the specified parameter lies within the
interval can be expressly stated. For example, the confidence that the true population mean lies
between 19.82 and 21.78 is 95%. Such statements require an assumption about the form of the
underlying distribution (e.g., normal, binomial, multinomial) of the variable being measured.
coordination: A bringing into a common action, movement or condition; direction toward a common
goal. Coordinated activities in EMAP include Integration and Assessment, Indicator Development,
Information Management, Logistics, Total Quality Management, and Statistics and Design.
core indicator: EMAP indicator that Is selected for long-term, ecological monitoring as a result of its
acceptable performance in a regional demonstration project.
correlation coefficient: A measure of the degree to which two or more variables are related; its value
always lies between -1 (strong negative relationship) and +1 (strong positive relationship.)
cumulative frequency distribution: A curve or distribution generated by some function, F(x)
represents the proportion of the population, expressed as units or area (e.g., proportion of stream
A-3
-------
EMAP - Great Lakes Glossary
reaches or proportion of estuary area) in the target population having a value for that variable that is
less than or equal to x. An empirical cumulative frequency distribution, based on a probability sample,
is an estimate of the population cumulative distribution function, or CDF.
data quality objective: Quantitative and qualitative statement of the level of uncertainty one is willing
to accept with regard to a given variable being measured. A data quality objective may include goals
for accuracy, precision, and limits of detection. It may also include goals for completeness,
comparability, and representativeness. Data quality objectives are established before sampling is
begun and may influence the level of sampling effort required.
demonstration project: A regional monitoring study undertaken to test the EMAP approach as
implemented for a particular resource category. Demonstration projects are undertaken after a pilot
project is completed and data quality objectives are established. (See pilot project.)
descending analysis: A method of calculating the upper confidence bound of a cumulative frequency
distribution. A descending analysis provides the upper confidence bound on numbers of units
possessing a particular value or values above it. An alternate method is ascending analysis.
design-based: Referring to inferences using methodology based on the sampling design. Such
inferences derive their properties from the design protocols. (See model-based.)
developmental indicator: An EMAP indicator that has passed evaluation for expected performance
(existing data analyses, simulations, and limited-scale field tests or pilot projects) and, with the
concurrence of scientific peer reviewers, is deemed suitable for actual performance testing in a regional
demonstration project. The term "probationary core indictor" is preferred to "developmental indicator"
since the former is more descriptive of an indicator's status.
diagnosis: The process of associating exposure, habitat, and stressor indicators with indicators of
ecological condition (i.e., response indicators) in order to identify environmental problems.
diagnostic indicator: A characteristic of the environment measured for comparison with indicators of
ecological condition (i.e., response indicators) to determine possible explanations for subnominal, or
poor or unacceptable, conditions; a collective term for any exposure, habitat, or stressor indicator.
discrete resource: A resource category or class that can be regarded as being spatially discrete or
subdivided by boundaries. Examples of discrete resources might include small lakes or stream
reaches. Designation of a resource as discrete is scale-dependent. (See extensive resource.)
distribution function: A mathematical expression describing a random variable or a population. For
real-world finite populations, these distributions are knowable attributes (parameters) of the population
and may be determined exactly by a census, or estimated from a sample. The general form is the
proportion (or other measure, like number, length, or area) of the resource having a value of an
attribute equal to or less than a particular value. Proportions may also be ol the different possible
measures, like number (frequency distributions), area (areal distributions), length, or volume.
Distribution functions are a primary descriptive statistic of EMAP. (See cumulative frequency
distribution.)
A-4
-------
EMAP - Great Lakes Glossary
domain: In a spatial or geographic sense, domain refers to areal extent of a resource, that is to the
region(s) occupied by a resource. In mathematics and statistics, domain refers to the set of values to
which a variable is limited or the set on which a function is defined.
double sample: A sample of a sample. In EMAP, a larger (Tier 1) sample of resource attribute
measurements can be obtained via remote sensing or cartographic materials than can be obtained via
field sampling (Tier 2). Attributes measured on the Tier 1 sample can be used to guide selection of the
Tier 2 sample, and therefore, the Tier 2 sample constitutes a double sample of the Tier 1 sample.
ecological indicator: A characteristic of the environment that, when measured, quantifies magnitude
of stress, habitat characteristics, degree of exposure to a stressor, or ecological response to exposure.
The term is a collective term for response, exposure, habitat, and stressor indicators.
ecological risk assessment: The process that evaluates the likelihood that adverse ecological effects
may occur or are occurring as a result of exposure to one or more stressors.
ecoregion: A geographic area which is relatively homogeneous with respect to ecological systems.
ecosystem: The biotic community and its abiotic environment.
ecosystem function: Attribute related to rates of change of structural components of an ecosystem;
examples include primary productivity, denitrification rates, and species fecundity rates.
ecosystem structure: Attribute related to instantaneous physical state of an ecosystem; examples
include species population density, species richness or evenness, and standing crop biomass.
ecotone: A habitat created by the juxtaposition of distinctly different habitats; an edge habitat; an
ecological zone or boundary where two or more ecosystems meet.
effective sample size: A computed sample size for special circumstances in which the "normal" or
original sample size is determined to be insufficient. The effective sample size is computed from known
or preliminary statistical characteristics of the attribute(s) being sampled.
environmental indicator: A measurement, statistic or value that provides a proximate gauge or
evidence of the effects of environmental management programs or of the state of condition of the
environment.
environmental value: A characteristic of the environment that contributes to the quality of life provided
to an area's inhabitants; for example, the ability of an area to provide desired functions such as food,
clean water and air, aesthetic experience, recreation, and desired animal and plant species.
Biodiversity, sustainability, and aesthetics are examples of environmental values.
eutrophication: A developmental or aging process undergone by some water bodies in which
biological productivity increases through time. Eutrophic water bodies are characterized by high
nutrient (nitrogen and phosphorus) concentrations and, as a consequence, by heavy growth of aquatic
plants and algae. This in turn contributes to increasing production of organic matter and its subsequent
decomposition by microbes. Lowered dissolved oxygen in the water caused by microbial activity results
in increased release of nutrients from the sediments, further contributing to the eutrophication process.
A-5
-------
EMAP - Great Lakes Glossary
Eutrophication is accelerated by human activities (sewage inputs, crop and lawn fertilization, etc.) that
contribute nutrients to surface waters.
exposure: The co-occurrence of a stressor with one or more ecological components (ecosystems or
their biotic components.) Stressors can be chemical, biological, or physical in nature.
exposure indicator: A characteristic of the environment measured to provide evidence of the
occurrence or magnitude of a response indicator's contact with a chemical or biological stress.
extensive resource: A resource category or class covering a large geographic area and not
subdivided by natural boundaries. Examples of extensive resources include grasslands or large
marshes. Characterization of a resource as extensive is scale-dependent.
40-hex: A colloquial term for the landscape description hexagon or landscape sampling unit centered
on each of the grid points in the EMAP sampling grid. The area of each hexagon is approximately 40
km2. (The actual size of each hexagon is 39.7 km2.)
found data: Data not collected by EMAP. Examples may include historical monitoring data or
contemporary data collected by other monitoring programs. It is always necessary to establish the
population represented by such data.
frame: An explicit representation of all units or elements of a target population or universe that is to be
sampled. Examples include list frames and area frames. (See sampling frame.)
geographic information system (GIS): A computer system expressly designed for storing,
manipulating, analyzing, and displaying data in a geographic context. A geographic information system
is usually thought of as a software system, but specialized hardware, such as graphics display
terminals and plotters, is required to take full advantage of the software.
grid: A systematic network of sampling points superimposed on a region of interest. In EMAP, the grid
is a network of equilateral triangular structures with grid points 27.1 km apart. This network,
superimposed on the conterminous U.S., establishes a base grid of approximately 12,600 points with
approximately one grid point per 635 km2. (See hexagon.)
grid enhancement: The process of systematically increasing the number of points in a sampling grid.
Grid enhancement is one method of producing an augmented sample.
grid randomization: The process of randomly positioning the grid so that each (discrete) unit of area
of fixed size is equally likely to contain a grid point. This process is the basis for the probability-sample
designation for EMAP monitoring.
A-6
-------
EMAP - Great Lakes Glossary
habitat indicator: A physical, chemical, or biological attribute of the environment measured to
characterize conditions necessary to support an organism, population, or community in the absence of
pollutants. Examples include salinity of estuarine waters or substrate type in streams or lakes.
hazard: The intrinsic ability of a stressor to cause adverse effects under a particular set of
circumstances.
hazard indicator: A type of stressor indicator; a measure that reflects human activities that
inadvertently affect ecological resources. Examples include measures of pollutant release, number of
permits issued for construction activity, and rates of application of fertilizers to forests and crops that
influence nutrient concentrations in adjacent streams.
hexagon: A regular six-sided polygon. The EMAP sampling grid is designed around hierarchical
hexagonal shapes. A hexagonal face of a truncated icosahedron projected onto the United States is
the main hexagon on which the EMAP sampling grid is based. Other hexagons are formed by the
nature of the triangular grid: Each baseline grid point is bounded by six adjacent grid points that form a
hexagon. These so-called baseline tessellation hexagons have an area of 634.5 km2. Landscape
sampling units (40-hexes) are smaller hexagons (1/16 the area of a baseline tessellation hexagon,
634.5 km2/16 = 39.7 km2) centered on a sample grid point within each baseline tessellation hexagon.
icosahedron: Regular geometric solid with 20 equilateral triangular faces and 12 vertices. A modified
icosahedron, referred to as a truncated icosahedron, is the basis for the baseline EMAP sampling grid.
implicit sampling frame: A set of rules or criteria used to select sampling units that cannot, a priori,
be listed explicitly. In EMAP, the criteria are developed during landscape characterization activities to
identify resource sampling units within each landscape sampling unit.
inclusion probability: The probability of including a specific sampling unit within a sample. Inclusion
probabilities must be known in order to calculate sample statistics.
index: Mathematical aggregation of indicators or metrics; one example is the Index of Biotic Integrity,
which combines several metrics describing fish community structure, incidence of pathology, population
sizes, and other characteristics.
index period: The period of the year or sampling window when measurement of an indicator yields the
most meaningful information. The index period may vary from one indicator or resource class to
another.
index sample: A standardized form of judgment sample for which rules of selecting the sample are
formally prescribed. In EMAP, index samples usually are acquired to facilitate interpretation of a
probability sample.
indicator: A measurement that can be used to assess the status and trends of environmental quality,
that is, to assess the ability of the environment to support a desired human or ecological condition.
A-7
-------
EMAP - Great Lakes Glossary
integrated assessment: Periodic reports in which an evaluation is made of the significance of status
and trends in response indicators in the context of policy-relevant questions. Integrated assessments
focus on assessment of environmental condition in the context of multiple resource issues. They may
include correlative analyses among frame data from within and among resource categories and
auxiliary data, as well as relevant literature citations from outside EMAP that are used to support,
explain, or be refuted by EMAP results.
integration: The formation, coordination, or blending of units or components into a functioning or
unified whole. In EMAP, integration refers to the establishment of a cohesive approach to
environmental monitoring, the coordination of monitoring efforts in multiple resource categories and the
blending and analysis of ecological data in order to undertake interpretation and assessment.
Integration in EMAP also includes coordination of EMAP with other monitoring programs and
coordination of program activities to address the needs of constituent groups.
interpenetrating design: An aspect of monitoring design in which a new set of sampling units is
selected each year through a cycle of successive years. Each subsequent cycle uses the same set of
sampling units as the initial cycle. In a four-year cycle, for example, the set of sampling units sampled
in year 1 would be resampled in years 5,9,13 those sampled in year 2 would be resampled in years
6,10,14 and so on.
landscape: The set of traits, patterns, and structure of a specific geographic area, including its
biological composition, its physical environment, and its anthropogenic or social patterns. An area
where interacting ecosystems are grouped and repeated in similar form.
landscape characterization: Documentation of the traits and patterns of the essential elements of the
landscape, including attributes of the physical environment, biological composition, its physical
environment, and its anthropogenic or social patterns. An area where interacting ecosystems are
grouped and repeated in similar form.
landscape ecology: The study of the distribution patterns of communities and ecosystems, the
ecological processes that affect those patterns, and changes in pattern and process over time.
landscape indicator: A measurement of the landscape, calculated from mapped or remotely sensed
data, used to described spatial patterns of land use and land cover across a geographic area.
Landscape indicators may be useful as measures of certain kinds environmental degradation such as
forest fragmentation.
landscape pattern type: A geographic area throughout which a common set of ecological resources
and land uses form a consistent pattern. Landscape pattern types are visually interpreted, classified,
and delineated by analysis of remote imagery or maps. They are classified in terms of composition
(i.e., component land use/land cover classes such as forest, grassland, agriculture, or urban) and the
pattern in which the components are arranged (e.g., matrix, matrix/patch, mosaic.)
list frame: A method for defining the sampling frame based on a characteristic or attribute of the
sampling units. The individual sampling units are defined by tables or lists of units are defined by
tables or lists of units with common attributes. For example, all lakes greater than 4 ha in the
southeastern United States can be found and listed by using U.S. Geological Survey maps or remote
sensing (or some other standardizing method). Such a list constitutes a list frame.
A-8
-------
EMAP - Great Lakes Glossary
long-range plan: A document setting forth, in general, the milestones, activities, and tasks that must
occur to achieve some goal in a specified time frame. It is similar to an operating plan but is less
detailed and covers a longer time frame. The time frame for long-range planning in EMAP is five
years.
management indicator: Within EMAP, a type of stressor indicator; a measure reflecting human
activities that intentionally alter an ecological resource to meet some management objective; for
example, the dredging or filling of a wetland for the purpose of housing development.
management plan: A detailed program of action designed to achieve management objectives. A
management plan may include regulations (e.g., restricted fishing) and policies (e.g., permit
requirements.)
measurement endpoint: A measurable ecological characteristic that is related to the valued
characteristic chosen as the assessment endpoint. Measurement endpoints are often expressed as the
statistical or arithmetic summaries of the observations that comprise the measurement.
model: A description, analogy, or abstraction used to help visualize or conceptualize something that
cannot be directly observed or measured; a system of postulates, data, and inferences presented as a
mathematical description of an entity or a state of affairs.
model-based: Referring to inferences using methodology based on models. Such inferences derive
their properties from the model assumptions. In some instances model-based inference is
advantageous over design-based inference, given that certain assumptions hold. (See design-based.)
monitoring: The act of measuring indicators or characteristics or environmental condition through time.
monotonic: Characterized by a lack of reversals; that is, steadily increasing or decreasing without
changes in direction.
natural process indicator: In EMAP, a type of stressor indicator; a measure reflecting phenomena
that affect ecological condition, regardless of the presence of management actions or environmental
hazards; examples include natural climatic fluctuations, predator-prey cycles, and insect and disease
epidemics.
nearest resource unit: A step in the sampling process where a resource sampling unit is selected
because of its proximity to the center of a landscape sampling unit.
network: A generic term for a collection of sites (i.e., subset of population units in a target population)
where measurements are made "continuously" or periodically over time. The term usually is associated
with trend studies or long-term research studies. Network sites include those obtained either (1) as a
probability sample from a target population or (2) from an existing collection of sites, as "found" sites, or
by subjective selection. In the second case, estimates of population characteristics may not be
possible.
nominal: The state of having desirable or acceptable ecological condition.
A-9
-------
EMAP - Great Lakes Glossary
nominal value: A standard established for a response indicator to represent desirable or acceptable
(nominal) condition. Subnominal values are then those values that fall below the nominal value.
nonparametric statistics: Statistical approaches used when the sampled characteristics of
populations cannot be described by a normal distribution or when the distribution is unknown. Such
methods are also referred to as distribution-free methods.
off-frame data: Data acquired by a sampling approach that does not provide a probability sample.
For example, data acquired by the National Acid Deposition Program are off-frame data.
on-frame data: Data acquired by a sampling approach that provides a probability sample. For
example, the Surface Water Survey of the National Acid Precipitation Assessment Program produced
on-frame data.
operating plan: A detailed program of action describing the procedures, practices, and actions to be
undertaken to achieve specified ends in a specified time frame. An operating plan is similar to a long-
range plan but is more detailed and covers a narrower planning time frame. In EMAP, the time frame
for operational planning is three years.
parameter: An attribute or characteristic. The term is used in specific technical disciplines in various
ways: In statistics, parameter refers to attributes of models or populations, and variable refers to
attributes of sampling units. In chemistry, parameter often refers to the attributes of, for example, a
water sample. Sometimes it is appropriate to refer to a particular attribute as a parameter in one
context and variable in another; a parameter of the lake is a variable of the population of lakes.
pattern: The location, distribution, and composition of structural landscape components within a
particular geographic area or in a spatial context; for example, a mosiac of patches.
pilot project: A sampling effort conducted over a small area usually during a single index period. Pilot
projects are used to evaluate indicators, sampling design, methods, and logistics. (See demonstration
project.)
policy: A definite course or method of action selected, in light of given conditions, from among
alternatives to guide and determine present and future decisions.
policy analysis: The process of evaluating alternative prospective policies or the various components
of those policies.
population: In statistics and sampling design, the term population refers to the total universe
addressed in a sampling effort. An example of a population is the set of lakes between 10 and 2000
ha in the eastern United States. The term universe is often used interchangeably with population. In
ecology, the term population refers generally to a group of individuals of the same species residing in
close proximity to each other such that the individuals share a common gene pool.
population estimate: A statistical estimate or distribution of some characteristic that applies to an
explicitly defined target population (category, class, or subclass), for example, the median acid-
A-10
-------
EMAP - Great Lakes Glossary
neutralizing capacity (or the cumulative frequency distribution of acid-neutralizing capacity) for all small
lakes in the Northeast.
population units: The entities that make up a target population. The units can be defined in many
ways, depending on the survey objectives and the type of measurement to be made. Typically,
definitions of environmental units include (1) an explicit statement of the characteristics each population
unit must possess in order to be considered a member of the target population and a (2) specification
of location in space and time.
precision: The degree of agreement among replicate measurements of the same attribute. A set of
measurements may be precise without necessarily being accurate.
probability sample: A sample chosen so that the probability of including each selected unit in the
sample is known, and so that each population unit has a positive probability of selection. This implies
that the target population is represented by the sample and that the target population is explicitly
defined. Statistical estimates of characteristics of the population so sampled can then be made with
known precision.
probationary core indicator: An EMAP indicator that has passed evaluation for expected
performance (existing data analyses, simulations, and small-scale field tests) and with the concurrence
of scientific peer reviewers, is deemed suitable for actual performance testing in a demonstration
project.
program: An administrative entity (people operating in some management environment) created for
the purpose of achieving some stated goal or end.
quality assurance: The total integrated program for ensuring the reliability of environmental
measurements and analysis. Quality assurance consists of multiple steps taken to ensure that all data
quality objectives are achieved.
quality control: Specific steps taken during the data collection process to ensure that equipment and
procedures are operating as intended and that they will allow data quality objectives to be achieved.
quantile: The value of an attribute indexing a specified proportion of a population distribution or
distribution function. Quartiles (25th, 50th, and 75th percentiles), the median (50th percentile), and
other percentiles are special cases of quantiles.
reference site: One of a population of benchmark or control sampling locations that, taken collectively,
represent an ecoregion or other large biogeographic area; the sites, as a whole, represent the best
ecological conditions that can be reasonably attained, given the prevailing topography, soil, geology,
potential vegetation, and general land use of the region.
region: Any explicitly defined geographic area. Regions may be defined administratively (e.g., EPA
Region III), politically (e.g., Texas), geographically (e.g., the Southwest), biogeographically (e.g., short-
grass prairie), physiographically (e.g., Rocky Mountains), or by other means.
A-11
-------
EMAP - Great Lakes Glossary
remote sensing: The collection and interpretation of information about an object without physical
contact with the object. Satellite imaging and aerial photography are examples of remote sensing.
research indicator: A candidate indicator identified for an EMAP resource category which has been
prioritized on the basis of several criteria (e.g., regionally applicable, integrates effects, monotonic,
conducive to synoptic monitoring) and, following peer review, has been selected for further evaluation
for use in EMAP as a possible probationary core indicator. An evaluation of the expected performance
of a research indicator is made after existing data are analyzed, simulation studies are performed with
realistic scenarios and expected spatial and temporal variability, and limited field tests are conducted.
residual: In statistics, the deviation of a data point from the value predicted by a regression equation.
It is computed as the difference between the observed value and the predicted value.
resource: An ecological entity that is identified as a target of sampling, description, and analysis by
EMAP. Such an entity is ordinarily thought of and described as a statistical population. A resource can
be characterized as belonging to one of two types, discrete and extensive, that pose different problems
of sampling and representation.
resource assessment: Periodic reports in which an evaluation is made of the significance of status
and trends in response indicators in the context of policy-relevant questions. Resource assessments
focus on assessment of individual resource categories. They may include correlative analyses among
frame data from within a resource category and auxiliary data, as well as relevant literature citations
from outside EMAP that are used to support, explain, or be refuted by EMAP results.
resource category: A group of general, broad ecosystem types or ecological entities sharing certain
basic characteristics. Seven such categories currently are identified within EMAP: estuaries, Great
Lakes, inland surface waters, wetlands, forests, arid lands, and agroecosystems. These categories
define the organizational structure of monitoring groups in EMAP and are the resources addressed by
EMAP resource assessments.
resource class: A subdivision of a resource category; examples include small lakes, oak-hickory
forests, emergent estuarine wetlands, field cropland, small estuaries, and sagebrush dominated desert
scrub.
resource domain: The areal extent of a resource; the region occupied by a resource.
Resource Group: A group of scientific and administrative personnel, headed by a Technical Director,
responsible for monitoring a given EMAP resource category. There are seven such groups in EMAP:
Estuaries, Great Lakes, Inland Surface Waters, Wetlands, Forests, Arid Lands, and Agroecosystems.
resource sampling unit: An individual station or site of a particular ecological resource category or
class; an individual unit of a target population (e.g., a stream segment, a forest stand, a wetland, an
estuary) upon which indicator measurements are made. More than one resource sampling unit can
occur in a landscape sampling unit.
response indicator: A characteristic of the environment measured to provide evidence of the
biological condition of a resource at the organism, population, community, or ecosystem level of
organization.
risk: The probability or likelihood an adverse effect will occur.
A-12
-------
EMAP - Great Lakes Glossary
risk characterization: Determination of the nature of a given risk and quantification of the potential for
adverse change to the environment from that risk. Characterization is accompanied by a statement of
uncertainty.
risk communication: The exchange of information about environmental risks among risk assessors,
risk managers, the general public, news media, special interest groups, and others.
risk management: The process of evaluating alternative regulatory and non-regulatory responses to
risk and selecting among them. The selection process necessarily requires the consideration of
scientific, legal, economic, and social factors.
sample: A subset of the units from a sampling frame, for example, a subset of resource units from a
population or set of sampling sites.
sampling design: The set of procedures associated with the inspection of the target population,
population units, sampling frame, and measurements to be made on units for a specific survey (study)
objective.
sampling frame: A list or spatial representation of explicit, clearly defined, mutually exclusive,
ecological resource units containing all of the elements of a specified universe. (See frame.)
sampling strategy: A sampling design, together with a plan of analysis and estimation. The design
consists of a frame, either explicit or implicit, together with a protocol for selection of sampling units.
sampling unit: An entity that is subject to selection and characterization under a sampling design. A
sample consists of a set of sampling units that are characterized. Sampling units are defined by the
frame; they may correspond to resource units, or they may be artificial units constructed for the sole
purpose of the sampling design.
spatial statistics: Statistical methodology and theory that accounts for spatial structure in a dataset.
Conventional population estimation does not normally account for spatial attributes, except perhaps for
spatial identity of subpopulations.
strategic plan: A document setting forth the mission and objectives of a program, its goals, and the
products it will ultimately provide. A strategic plan must also provide an overarching vision of how to
achieve those ends.
stratified design: A statistical sampling design in which the target population is divided into groups
(strata) because of some distinguishing characteristic(s); a probability sample is selected from each
stratum.
stratum: A sampling structure that restricts sample randomizing/selection to a subset of the frame.
Inclusion probabilities may or may not differ among strata.
stressor indicator: A characteristic measured to quantify a natural process, an environmental hazard,
or a management action that can effect changes in exposure and habitat. Three types of stressor
indicators are considered in EMAP: hazard indicators, management indicators, and natural process
indicators. Examples include the incidence of fertilizer application, which can increase nutrient
A-13
-------
EMAP - Great Lakes Glossary
concentrations in lakes; incidence of dredging/filling, which can diminish availability of wetland habitat;
and climatic fluctuations, which can promote damage by pathogens.
subnominal: Having undesirable or unacceptable ecological condition.
subnominal threshold: A value selected for a response indicator below which a resource is
designated as subnominal with respect to a given assessment endpoint.
subpopulation: Any subset of population, usually having a specific attribute that distinguishes its
members from the rest of the population, for example, lakes from a specified population that are above
1000 m in elevation. Subpopulations are important entities in the EMAP plan. Any defined
subpopulation is subject to characterization via estimation of subpopulation attributes and comparison to
other subpopulations.
suite: A group or collection of related, complementary objects. In EMAP, the term usually is used in
the phrase "indicator suite," which refers to the complete array of indicators; measured on a given
resource category.
survey: The collection of data in order to examine, characterize, or analyze the condition or status of
some aspect or aspects of an area.
systematic sample: A sampling design that utilized regular spacing between sample points, in one
sense or another. The EMAP design selects samples via the triangular grid. Spatial arrangement of
the selected resource units is not always strictly systematic, but the systematic grid is an important
aspect of the design.
target population: The total entity of set of entities addressed in a sampling effort. An example of a
target population is the set of lakes between 10 and 2000 ha in the eastern United States. (See
population.)
Task Group: A group of scientific and administrative personnel headed by a Technical Coordinator
and charged with addressing specific cross-cutting, integrative issues in EMAP, such as Integration and
Assessment, Indicator Development, Information Management, logistics, Total Quality Management,
and Statistics and Design. (See Resource Group.)
tier structure: In EMAP, a set of hierarchically arranged levels or stages of monitoring/sampling effort.
Tier 1: The sampling level in which the landscape is described with information obtained through
remote sensing, maps, and other information. The principal aims of the Tier 1 sample are to describe
the extent of resource categories and classes and to describe landscape structure. Tier 1 efforts also
includes development of sampling frame information for each resource category.
Tier 2: A more intensive level of sampling than at Tier 1, in which a probability sample drawn from the
sampling frame information developed at Tier 1, that is, a double sample from the Tier 1 sample. Tier
2 is oriented to field monitoring of indicators to assess ecological condition of each resource category.
Tier 3: An augmented sampling effort within the same frame that Tier 2 sampling has been conducted,
but with increased spatial or temporal resolution. A Tier 3 sample is undertaken on a given resource
A-14
-------
EMAP - Great Lakes Glossary
when assessment of the routine Tier 2 sample indicates that enhanced spatial or temporal resolution
and/or a more complete suite of indicators is necessary to adequately evaluate that resource.
Tier 4: Specialized studies or experiments undertaken when Tier 2 or 3 samples indicate that a better
understanding of cause-effect relationships among indicators and/or stressors is necessary to evaluate
ecological condition.
top-down approach: A risk assessment methodology that uses ecoepidemiological analyses to
assess environmental condition. This is an effects-driven form of risk assessment; that is, ecological
effects or responses are first observed and then associated temporally or spatially with pollutant
exposures, habitat condition, and pollutant sources. (See bottom-up approach.)
total quality management (TQM): The quality assurance process adopted by EPA in which
management philosophy, planning, and operational methodology are completely committed to quality
improvement in all aspects of the organization or the program.
variable: An attribute or characteristic. In statistics, variable refers to attributes of sampling units, and
parameter refers to attributes of models or populations. Sometimes it is appropriate to refer to a
particular attribute as a parameter in one context and variable in another; a parameter of a lake is a
variable of the population of lakes.
variance: A measure of the variability or precision of a set of observations. It is calculated as the
mean of the sum of squared deviations from the mean value. Normally, a sampling program is
designed to achieve measurements that meet specified standards of precision.
watershed: The terrestrial area of the landscape contributing to flow at a given stream location.
A-15
-------
Appendix B. .
Indicator Fact Sheets
for
EMAP - Great Lakes
-------
EMAP - Great Lakes Indicator Fact Sheet
Indicator: Benthic invertebrate community structure
Category: Response indicator
Resource class: Offshore, nearshore, and harbor/embayment zones
Application: The abundance of benthic invertebrate species complements measures of pelagic biota
by representing the response of aquatic communities living in or on the sediments of the Great Lakes.
Benthic organisms can be impacted by changes in physical sediment characteristics, chemical
contamination in sediments, and by changes in the physical environment such as the frequency and
duration of anoxia. These organisms often serve as important food sources for fish and can contribute
to the transfer of contaminants through the food web. Benthic organisms often serve as the interface
between sediment conditions and the pelagic community. Benthic organisms are relatively stationary
and their abundance is therefore reflective of localized conditions. The life history characteristics of
many benthic organisms is documented and suggests that their population fluctuations are less
dramatic than is found with pelagic invertebrates such as zooplankton. A variety of metrics of benthic
community structure have been proposed for use in aquatic ecosystems. There is, however, no widely
accepted index of benthic community structure that provides quantification of healthy conditions.
Perhaps the most widely used comparative tool is simply the number of species present and
abundance of those species.
Index Period: Benthic community structure can be affected by the emergence of aquatic insects
during the spring, summer, and fall. However, winter and early spring sampling can be difficult due to
poor weather conditions and the presence of ice. Because rapidly changing environmental conditions
are often coupled with changes in community structure, late summer would be a preferred sampling
period. This would allow the development of newly hatched insects and more generally stable
environmental conditions. In any case, sampling an area, within which comparisons are to be made,
should occur over a relatively short time period.
Measurements: The measurements for benthic community structure consist of the number of
individuals of each species collected in a sediment sample. These samples can be obtained by a
variety of sampling devices. The most common are Eckmanฎ and Ponarฎ dredges and box cores.
The selection of devices depends in part on the substrate present. Eckmanฎ dredges are usually
lighter in weight and may have difficulty when large solid material (e.g., sticks, rocks) are present. After
collection, samples are usually sieved to remove sediment and retain the animals. Sieve mesh size
can also selectively affect the retention of organisms.
Variability: Variability of measures of community structure can be the result of variations in spatial
distribution and temporal conditions. Spatial distribution is affected by natural differences in habitat and
sediment characteristics (e.g., particle size distribution) that result from natural variations in water depth,
current speed and current direction. Temporal variation is the result of variations in environmental
conditions (season to season; year to year) and life history characteristics such as emergence and
reproductive patterns. These types of variability can be accounted for by careful characterization of
physical and environmental features and by selection and standardization of index periods for sampling.
Additional variation can result from the use of sampling gear and the degree of consistency among
scientists making identifications and counts. These also should be standardized to the extent possible.
Primary Problems: The primary difficulties associated with this indicator include the lack of an
integrative/interpretive metric and the spatial/temporal variability. Knowledge of the natural variability
will be necessary to make interpretive reports on the health of benthic communities in the Great Lakes.
B-1
-------
EMAP - Great Lakes Indicator Fact Sheet
Indicator: Hexagenia abundance
Category: Response indicator
Resource class: Nearshore and harbor/embayment zones
Application: The abundance of Hexagenia mayflies is used as an indicator of water and sediment
quality in the Great Lakes. Hexagenia were abundant in mesotrophic waters of the Great Lakes prior to
the 1950's (Schneider et al. 1969, Reynoldson 1989). Their populations are sensitive to eutrophication
(Rasmussen 1988), oil, metals, and other contaminants (Hiltunen and Schloesser 1983, Malueg et al.
1984, Specht et al. 1984). When water quality improves, Hexagenia populations often increase in
abundance (Fleming 1989, Reynoldson et al. 1989) and the existence of historical data makes
comparisons with previous population densities and ecosystem carrying capabilities possible.
Index Period: Hexagenia should be sampled monthly May through October to sample the seasonal
variability that is characteristic of mayflies (Schneider et al. 1969, Rasmussen 1988).
Measurements: Sampling stations are already established in some areas (Reynoldson et al. 1989);
others need to be chosen. Samples should be taken from sediments composed of mud or soft clay in
mesotrophic waters in which Hexagenia have been recorded, or in which it seems likely they would be
present if water and soil characteristics were suitable. Hexagenia are measured in number/m2 of
sediment. Sampling equipment varies; recent workers have used box core, Ponarฎ, and drag/dredge
devices (Reynoldson 1989). Standardization of methods and equipment is necessary. Identification to
species is desirable. Unfortunately, this sampling is time consuming and requires the use of
specialized equipment.
Variability: Hexagenia abundance is spatially and temporally variable because seasonal fluctuations in
weather and other abiotic factors affect the survival and developmental rate of the nymphs. One of the
most important abiotic factors to consider is sediment type. Sediment consistency should be adequate
enough to allow for burrow stability. Thus, long-term monitoring of a variety of sites is essential for the
use of Hexagenia as an indicator of water and sediment quality.
Primary Problems: Hexagenia is an appropriate indicator of water and sediment quality only in
sediments of appropriate substrata at depths up to 26 -m (Mozley and LaDronka 1988). The high
variability of Hexagenia abundance makes this an indicator that is useful for long-term monitoring and
does not provide short-term information about changes in environmental quality (Reynoldson 1989).
The great sensitivity of this genus to contamination and low oxygen concentrations means that
recolonization occurs only after large changes in environmental quality are attained, so small
improvements in water and sediment quality may not be indicated (Reynoldson 1989).
References:
Fremling, C.R. 1989. Hexagenia Mayflies: Biological monitors of water quality in the Upper
Mississippi River. Journal of the Minnesota Academy of Science 55:139-143.
Hiltunen, J.F. and D.W. Schloesser. 1983. The occurrence of oil and the distribution of Hexagenia
(Ephemeropteria: Ephemeridae) nymphs in the St. Marys River, Michigan and Ontario. Freshwat.
Invertebr. Biol. 2:199-203.
Malueg, K.W., G.S. Schuytema, J.H. Gakstatter, and D.F. Krawczyk. 1984. Toxicity of sediments from
three metal-contaminated areas. Environ. Toxicol. Chem. 3:279-291.
B-2
-------
EMAP - Great Lakes Indicator Fact Sheet
Mozley, S.C. and R.M. LaDronka. 1988. Ephemera and Hexagenia (Ephemeridae, Ephemeroptera) in
the Straits of Mackinac, 1955-56. J. Great Lakes Res. 14:171-177.
Rasmussen, J.B. 1988. Habitat requirements of burrowing mayflies (Ephemeridae: Hexagenia) in
iakes, with special reference to the effects of eutrophication. J.N. Am. Benthol. Soc. 7:51-64.
Reynoldson, T.B., D.W. Schloesser, and B.A. Manny. 1989. Development of a benthic invertebrate
objective for mesotrophic Great Lakes waters. J. Great Lakes Res. 15:669-686.
Schneider, J.C., F.F. Hooper, and A.M. Beeton. 1969. The distribution and abundance of benthic
fauna in Saginaw Bay, Lake Huron. Proc. 12th Conf. Great Lakes Res.:80-90.
Specht, W.L., D.S. Cherry, R.A. Lechleitner, and John Cairns, Jr. 1984. Structural, functional, and
recovery responses of stream invertebrates to fly ash effluent. Can. J. Fish. Aquat. Sci. 41:884-896.
B-3
-------
EMAP - Great Lakes Indicator Fact Sheet
Indicator: Diporeia abundance
Category: Response indicator
Resource class: Offshore and nearshore zones
Application: The abundance of the benthic amphipod Diporeia hoyi, formerly Pontoporeia hoyi,
(Bousfield 1989) has been suggested as an indicator of ecosystem health for oligotrophic waters in the
Great Lakes (Ryder and Edwards 1985). This species is found in all the Great Lakes (Balcer et al.
1984) and is a food item for a variety of fish species. The benthic habit of D. hoyi makes it a good
indicator of the environment at the water-substrate interface. Since many contaminants are found
associated with sediment particles, it is important to have an indicator that is sensitive to the
contaminants in this strata of the ecosystem. D. hoyi have been shown to bioconcentrate a variety of
pollutants found in the Great Lakes, including PCBs, toxaphene, DDE (Evans et al. 1991), PAHs (Eadie
et al. 1982). The abundance of the population is inversely correlated with contaminant concentrations
in some areas (Nalepa and Thomas 1976, Kraft 1979). The existence of historical data for abundance
in all of the Great Lakes makes comparisons with previous population densities and ecosystem carrying
capacities possible.
Index Period: Sampling should take place in mid-summer when abundance is generally greatest.
Measurements: The number of D. hoyi per meter is measured. A variety of methods for collecting
sediment and the associated organisms are used. D. hoyi is relatively easily distinguished from other
amphipods in the Great Lakes. Densities of up to 11,00 individuals/m2 have been recorded (Nalepa
and Thomas 1976, Evans et al. 1990). Minimum D. hoyi population density standards must be set up
for each of the Great Lakes.
Variability: There is little seasonal variability in the abundance of D. hoyi in deep water; some
variability may occur in shallow areas. Water depth and sediment type appear to influence D. hoyi
abundance. Depths of 40-65 m generally have the greatest densities (from 470 individuals/m2), and in
deeper water, the species is most common in sediments with a mean grain size of less than 0.5 mm
(Nalepa and Thomas 1976, Kraft 1979, Balcer et al. 1984). Sampling should be done at previously
established stations when possible, and methods of sediment and organism collection should be
standardized.
Primary Problems: The relationship between D. hoyi abundance and the condition of the whole
benthic community needs to be defined. Additional research is required to set standards for the
"healthy" abundance of D. hoyi at different sites in the Great Lakes.
References:
Balcer, M.D., N.L. Korda, and S.I. Dodson. 1984. Zooplankton of the Great Lakes. The University of
Wisconsin Press, Madison.
Bousfield, E.L. 1989. Revised morphological relationships within the amphipod genera Pontoporeia
and Gammaracanfbus and the "glacial relict" significance of their post-glacial distributions. Can. J.
Fish. Aquat. Sci. 46:1714-1725.
Eadie, B.J., P.F. Landrum, and W. Faust. 1982. Polycyclic aromatic hydrocarbons in sediments, pore
water and the amphipod Pontoporeia hoyi from Lake Michigan. Chemosphere 11 -.847-858.
B-4
-------
EMAP - Great Lakes Indicator Fact Sheet
Evans, M.S., G.E. Noguchi, and C.P. Rice. 1991. The biomagnification of polychlorinated biphenyls,
toxaphene, and DDT compounds in a Lake Michigan offshore food web. Arch. Environ. Contam.
Toxicol. 20:87-93.
Kraft, K.J. 1979. Pontoporeia distribution along the Keweenaw shore of Lake Superior affected by
copper tailings. J. Great Lakes Res. 5:28-35.
Nalepa, T.F. and N.A. Thomas. 1976. Distribution of macrobenthic species in Lake Ontario in relation
to sources of pollution and sediment parameters. J. Great Lakes Res. 2:150-163.
Ryder, R.A. and C.J. Edwards (Eds.) 1985. A conceptual approach for the application of biological
indicators of ecosystem quality in the Great Lakes basin. Report to the Great Lakes Science Advisory
Board, International Joint Commission, Windsor, Ontario, Canada.
B-5
-------
EMAP - Great Lakes Indicator Fact Sheet
Indicator: Fish pathology
Category: Response indicator
Resource class: Offshore, nearshore, and harbor/embayment zones
Application: When fish are exposed to contaminants, the effects observed often include morphological
change. Assessments of the morphology of fish collected from different habitats within the lake could
be used to indicate the proportion of the lake which is adversely affected by contaminants. Many of the
observable effects occur only after a dose threshold is achieved, however, beyond the threshold, all or
nearly all exposed organisms demonstrate the effect. Monitoring for structural pathology of this type
can be very cost effective because small sample sizes can be used as indicators of adverse
contaminant effects on lake ecosystems. Other types of structural pathology seem to be stochastic
processes with an associated low probability of occurrence within the population, e.g., cancer.
Detection of these processes requires large sample sizes.
Index Period: The sampling period can be variable, depending on the species selected. If
reproductive status is important, the sample should be collected both before and after spawning to
allow evaluation of the gonad during this critical event. Other considerations should include the
temporal migration patterns of the fish in the context of the overall assessment strategy, i.e., the relative
sizes of the fish home range, the sample site, and the area being assessed.
Measurements: Depending on the specific morphological indicators chosen, different measurement
techniques should be applied. For the threshold-type morphological responses, fish of the appropriate
segment of the year-class structure of the population should be collected. Relatively few specimens
are required (between 10 and 20 fish per site). These specimens should be thoroughly examined,
externally and internally, for gross lesions and anomalies. Following this examination, samples of the
various tissues should be sampled and examined in the lab for histopathology.
Non-threshold, stochastic lesion responses require large sample sizes from each collection site.
Because of this, detailed morphological analyses are very costly. Less costly but less robust
measurements of these types of lesions can be accomplished with gross examinations. If both
threshold and non-threshold lesions are selected for indicators, then a tiered measurement/sampling
strategy is recommended. The first tier would be performed during routine sampling of fish for other
purposes such as population estimates. This tier would consist of gross internal and external
examination of the appropriate year-class specimens for lesions and anomalies. Tier two would consist
of systematic subsampling of fish from the gross examination (for example, every 20th specimen) to be
examined for microscopic lesions within selected tissues such as the liver.
Variability: There is little field data to analyze for variance. Given that exposures in the field are
consistent, the variance in the response would be expected to be quite low as evidenced by lab
investigations. The simple threshold toxicant induced pathology measurements allow repeated
measures at several sites for determination of variance.
Primary problems: Little lab or field data exist to describe the variance of the low frequency non-
threshold pathological lesions. Even with large sample sizes, the ability to reliably detect the presence
of cancer within the population is difficult. Variance determinations would be even more difficult to
estimate.
B-6
-------
EMAP - Great Lakes Indicator Fact Sheet
References:
Johnson, R.D. and H.L. Bergman. 1984. Use of histopathology in aquatic toxicology: A critique. In:
Cairns, V.W., P.V. Hodson, and J.O. Nriagu (Eds.). Contaminant Effects on Fisheries. John Wiley and
Sons, New York, NY: 19-36.
Malins, D.C., B.B. McCain, J.T. Landahl, M.S. Myers, M.M. Krahn, D.W. Brown, S.L. Chang, and W.T.
Roubal. 1988. Neoplastic and other diseases in fish in relation to toxic chemicals: An overview.
Aquat. Toxicol. 11:43-67.
Mix, M.C. 1986. Cancerous diseases in aquatic animals and their association with environmental
pollutants: A critical literature review. Marine Environ. Res. 20:1-141.
Sindermann, C.J., F.B. Barg, N.O. Christionson, V. Dethlefsen, J.C. Harshbarger, J.R. Mitchell, and
M.F. Mulcahy. 1980. The role and value of pathobiology in pollution effects monitoring programs.
Rapp. P.-V. Re'un. Cons. Int. Explor. Mer. 179:135-151.
B-7
-------
EMAP - Great Lakes Indicator Fact Sheet
Indicator: Diatom assemblages in lake sediment cores and sediment traps
Category: Response indicator
Resource class: Offshore and nearshore zones
Application: The Bacillariophyceae (diatoms) are an important component of the plankton community
and of littoral substrata. Diatom taxa and assemblages have been used as indicators of environmental
conditions, and have been used to reconstruct past lake conditions (Lowe 1974, Kilham and Hecky
1984). They have also been used as indicators of anthropogenic influence in the Great Lakes
(Stoermer et al. 1985a, 1985b). Diatom taxa are known to be resistant or susceptible to toxicity from
heavy metals (Elner and Happy-Wood 1980, Stoermer et al. 1985c, Kingston and Birks 1990), and to
eutrophication (David 1964, Harris and Vollenweider 1982, Agbeti and Dickman 1989), temperature
(Stoemer and Ladewski 1976), and land use (Tuchman et al. 1984). Historical diatom communities
have been investigated in the Great Lakes (Duthie and Sreenivasa 1971, Frederick 1981, Kingston et
al. 1978, Stoermer et al. 1985c), and provide evidence for the usefulness of sub-fossil diatom remains
for providing historical information. Use of diatom assemblages in contemporary water column and
sediment trap samples will also provide information on the current status of the lakes (Stoermer and
Kreis 1980).
Index Period: The timing of collection of sediment cores for historical analysis is not crucial. Sediment
trap sampling also provides some flexibility in the time of collection. Since sediment traps collect
samples over relatively long periods, they integrate the diatom community over time. Deployment and
retrieval of sediment traps could be accomplished in one (summer) or two (spring and fall) trips/year,
depending on the trap's susceptibility to damage by ice.
Measurements: Sediment cores for historical data would be taken in depositional areas in both
offshore and nearshore waters. Cores would be sectioned and the sections dated using radioisotope
and/or paleotonlogical methods. Diatom remains in the sections would be analyzed by a taxonomist.
The resulting taxa counts would be analyzed using multivariate techniques which might include, but are
not limited to: similarity indices, canonical correspondence analysis, and clustering.
Sediment traps would be deployed in offshore and nearshore locations at hypolimnetic or near-bottom
depths. These traps would remain in place for a minimum of six months to collect an integrated sample
of the diatom community at each location. Sediment traps would be retrieved and the samples
preserved for taxonomic analysis. Data resulting from taxonomic analysis would be treated similarly to
that from derived from cores.
Variability: Because both sediment cores and sediment traps collect diatom remains over relatively
long periods of time, as compared with a grab sample, the temporal component of variability is
eliminated. Spatial variability of diatom remains in sediment cores and trap samples is expected to be
small in the offshore areas of the Great Lakes. The spatial variability in nearshore core and trap
samples is largely unknown. However, these samples will reflect local influences (e.g., tributary
loadings) and will be inherently more variable.
Primary Problems: Potentially, problems exist for this indicator in several areas. It has not been
determined whether enough highly trained taxonomists are available to analyze the large number of
samples which will be initially generated. Secondly, a quantitative index and/or a baseline "pristine"
diatom community must be established to make reporting of data straightforward. The question of
reporting of levels of individual indicator species must also be resolved.
B-8
-------
EMAP - Great Lakes Indicator Fact Sheet
References:
Agbeti, M. and M. Dickman. 1989. Use of lake fossil diatom assemblages to determine historical
changes in trophic status. Can. J. Fish. Aquat. Sci. 46:1013-1021.
Davis, C.C. 1964. Evidence for the eutrophication of Lake Erie from phytoplankton records. Limnol.
Oceanogr. 9:275-283.
Duthie, H.C. and M.R. Sreenivasa. 1971. Evidence for the eutrophication of Lake Ontario from
sedimentary diatom succession. Pages 1-13 jn Proceedings of the 14th Conference on Great Lakes
Research. International Association for Great Lakes Research.
Elner, J.K. and C.M. Happy-Wood. 1980. The history of two linked but contrasting lakes in North
Wales from a study of pollen, diatoms, and chemistry in sediment cores. J. Ecol. 68:95-121.
Frederick, V.R. 1981. Preliminary investigation of the algal flora in the sediments of Lake Erie. J.
Great Lakes Res. 7:404-408.
Harris, G.P. and R.A. Vollenweider. 1982. Paleolimnological evidence of early eutrophication in Lake
Erie. Can. J. Fish. Aquat. Sci. 39:618-626.
Kingston, J.C. and H.J.B. Birks. 1990. Dissolved organic carbon reconstructions from diatom
assemblages in PIRLA project lakes, North America. Phil. Trans. R. Soc. Lond. B 327:279-288.
Kingston, J.C., R.L. Lowe, E.F. Stoermer, and T.B. Ladewski. 1983. Spatial and temporal distribution
of benthic diatoms in northern Lake Michigan. Ecology 64:1566-1580.
Lowe, R.L. 1974. Environmental requirements and pollution tolerances of freshwater diatoms. EPA
670/4-74/005, U.S. Environmental Protection Agency, Cincinnati, OH.
Stoermer, E.F., J.A. Wolin, C.L. Schelske, and D.J. Conley. 1985a. An assessment of ecological
changes during the recent history of Lake Ontario based on siliceous algal microfossils preserved in the
sediments. J. Phycology 21:257-276.
Stoermer, E.F., J.A. Wolin, C.L. Schelske, and D.J. Conley. 1985b. Postsettlement diatom succession
in the Bay of Quinte, Lake Ontario. Can. J. Fish. Aquat. Sci. 42:754-767.
Stoermer, E.F., J.P. Kociolek, C.L. Schelske, and D.J. Conley. 1985c. Siliceous microfossil succession
in the recent history of Lake Superior. Proc. Acad. Nat. Sci. 137:106-118.
Stoermer, E.F. and T.B. Ladewski. 1976. Apparent optimal temperatures for the occurrence of some
common phytoplankton species in Lake Michigan. Univ. Mich., Great Lakes Res. Div. Special Pub. No.
18:48 pp.
Stoermer, E.F. and R.G. Kreis. 1980. Phytoplankton composition and abundance in southern Lake
Huron. EPA 600/3-80-061, U.S. Environmental Protection Agency, Duluth, MN.
Tuchman, M.L., E.F. Stoermer, and H.J. Carney. 1984. Effects of increased salinity on diatom
assemblages in Fonda Lake, Michigan. Hydrobiol. 109:179-188.
B-9
-------
EMAP - Great Lakes Indicator Fact Sheet
Indicator: Chlorophyll-a composition in water
Category: Response indicator
Resource class: Offshore, nearshore, and harbor/embayment zones
Application: This indicator assesses the trophic status endpoint. Chlorophyll-a has been used as a
measure of phytoplankton biomass and as an indicator of the productivity of aquatic systems. The
amount of chlorophyll in a water body corresponds most closely to the nutrient input to the lake
(Vollenweider 1968, Smith 1982). Chlorophyll-a is the primary photosynthetic pigment of algae and
provides an estimate of the total phytoplankton community. Pigment concentration does not respond
rapidly to stressors (Schindler 1987) other than nutrient input. It has been assumed to be an indicator
of food web changes (Carpenter and Kitchell 1984), but this may not be the case for the Great Lakes
(Lehman 1988). Chlorophyll-a as an indicator would best be interpreted by including measurement of
other water chemistry variables in the monitoring design.
Index Period: The index period will be late July through August. This is a period of (generally) low
chlorophyll in the Great Lakes. The main requisite is thai chlorophyll be sampled during the same time
period each year.
Measurements: Measurements would be made using an in situ fluorometer. The instrument of choice
is a Sea Tech, Inc. fluorometer, which is interfaced to a Sea Bird Electronics, Inc. CTD probe. For
consistency, surface water chlorophyll-a would be measured. Although this sampling scheme would
potentially miss any sub-thermocline chlorophyll-a peak (Brooks and Torke 1977), it would greatly
reduce variability among samples. The in situ fluorometer would be checked against filtered and
acetone extracted chlorophyll-a measurements.
Variability: Spatial variability in the offshore areas of the Great Lakes will be relatively low, with a
coefficient of variation in the 10% - 30% range. The variability in the nearshore area will be higher,
perhaps up to 100% in any lake, and several times that among lakes.
Primary Problems: The primary problem with the use of chlorophyll-a as a response indicator is that it
is unresponsive to many stressors. While it is a good indicator of eutrophication, there is some
uncertainty as to its response to food web changes. As there are constant food web manipulations
occurring in the Great Lakes, through stocking of salmonids and through the introductions of exotic
species (e.g., Dreissena polymoipha and Bythotrephes cederstroemi) this indicator requires information
on water chemistry and biology for interpretation.
References:
Brooks, A.S. and B.G. Torke. 1977. Vertical and seasonal distribution of chlorophyll-a in Lake
Michigan. J. Fish. Res. Board Can. 34:2280-2287.
Carpenter, S.R. and J.F. Kitchell. 1984. Plankton community structure and limnetic primary production.
Am. Nat. 124:159-172.
Lehman, J.T. 1988. Algal biomass unaltered by food-web changes in Lake Michigan. Nature 332:537-
538.
Schindler, D.W. 1987. Detecting ecosystem responses to anthropogenic stress. Can. J. Aquat. Sci.
44:6-25.
B-10
-------
EMAP - Great Lakes Indicator Fact Sheet
Smith, V.H. 1982. The nitrogen and phosphorus dependence of algal biomass in lakes: An empirical
and theoretical analysis. Limnol. Oceanogr. 27:1101-1112.
Vollenweider, R.A. 1968. Scientific fundamentals of the eutrophication of lakes and flowing waters with
particular reference to nitrogen and phosphorus as factors in eutrophication. DAS/CS168-27,
Organization for Economic Cooperation and Development, Paris.
B-11
-------
EMAP - Great Lakes Indicator Fact Sheet
Indicator: Trophic status index
Category: Response indicator
Resource class: Offshore, nearshore, and harbor/embayment zones
Application: These indicators come under the general class of TSIs (trophic state indices) as
described in the EMAP-Surface Waters Fact Sheets. One in particular, called the CTI (Composite
Trophic Index, Gregor and Rast 1979), has been developed for Great Lakes nearshore zones. For the
upper lakes, the index is calculated as:
=0.556 + 1.67 CHLa + 0.31 TP
C77-ฎ
except for highly turbid regions (harbors) where the index is calculated as:
f-^Sr - O-409! + 1-67 CHLa + 0.31 TP
C77=i^
where SD is the secchi depth in meters
CHLa is the chlorophyll-a concentration in uxj/l
and TP is the total phosphorus concentration in
The attached table indicates the trophic state corresponding to the value of the CTI.
Trophic State (Water Quality) CTI
Eutrophic (poor) >11.0
Eutrophic/Mesotrophic 9.0 - 11.0
Mesotrophic (fair) 4.6 - 8.9
Oligotrophic/mesotrophic 3.1 - 4.5
Oligotrophic (good) >3.1
Index Period: Summer (July)
Measurements: The CTI can be calculated from basic water chemistry measurements that would
normally be part of any core sampling effort. These data are available on all Great Lakes historically
going back to the early 70's.
Variability: The expected variability of the CTI is <10% from the mean value, but more variability is
expected for high values of CHLa and TP. The variability also depends on the coefficient of variations
of the individual measurements.
Primary Problems: The CTI was developed for average or steady-state nearshore conditions and
therefore, the index period may not always occur at the same time each year. Also, although data from
all five Great Lakes were used to develop the CTI, Lake Michigan had the least data available, and
therefore, would require further testing to ensure the reasonableness of the index values.
B-12
-------
EMAP - Great Lakes Indicator Fact Sheet
References.
Gregor, D.J. and W. Rast. 1979. Trophic characterization of the U.S. and Canadian nearshore zones
of the Great Lakes. Report to Pollution from Land Use Activities Reference Group, International Joint
Commission, Windsor, Ontario:38 pp.
B-13
-------
EMAP - Great Lakes Indicator Fact Sheet
Indicator: Sediment toxicity to Hyallela azteca
Category: Exposure indicator
Resource class: Nearshore and harbor/embayment zones
Application: The acute toxicity of Great Lakes sediments to the talitrid amphipod, Hyalella azteca, has
been proposed as an indicator of the toxicity of sediments to the broader array of Great Lakes benthic
macroinvertebrates. This species is an important food organism for many Great Lakes fish species and
is a representative of one of the dominant groups (Amphipoda) of Great Lakes benthic macro-
invertebrate fauna. Hyalella azteca is a clean water, benthic/epibenthic invertebrate which is found
primarily in the nearshore areas of the Great Lakes.
Index Period: not applicable
Measurements: Sediment samples for toxicity testing can be collected with a variety of techniques
depending on the exact purposes of the study. Core samples may be required to determine the vertical
extent of sediment contamination or to quantify the volume of toxic sediments at a given location. Grab
sampling may be sufficient if the sole purpose of the study is to determine the spatial distribution of
toxic surficial sediments or to compare the toxicity of surficial sediments between sites. If there is no
statistically significant difference in mortality between the field-collected and control or reference
sediments, the field-collected sediments are classified as not acutely toxic. If there is statistically
greater toxicity in the field-collected sediments, relative to controls or reference sediments, the field-
collected sediments are classified as acutely toxic.
Variability: The sources of variability for this indicator can be separated into sampling variability and
measurement variability. The representativeness of the sediment samples for testing and the health'of
the test species are the primary concerns. Test methods, conditions, and the source of test organisms
should be standardized to eliminate these factors as sources of variability.
Primary Problems: (1) How well do laboratory assays predict the toxicity of in-place sediments? (2)
How representative of the total benthic macroinvertebrate fauna of the Great Lakes are the sensitivities
of H. azteca to contaminants in sediments? (3) How many samples are necessary to adequately
characterize the toxicity of sediments in a given area of the Great Lakes and how frequently must
samples be collected and tested to adequately address the potential seasonal changes in sediment
toxicity.2
B-14
-------
EMAP - Great Lakes Indicator Fact Sheet
Indicator: Contaminant concentrations in sediments from cores and traps
Category: Exposure indicator
Resource class: Offshore, nearshore, and harbor/embayment zones
Application: Many contaminants in the Great Lakes associate with particulate matter and eventually
are delivered to the sediments where they are subsequently buried. Sediment concentrations and
accumulation rates provide a sensitive measure of the presence of these chemicals, provide a spatial
distribution of the contamination, and provide an historical record back in time of previous contaminant
inputs to sediments.
Index Period: Anytime during the stratified period, when resuspension is limited.
Measurements: This approach is most effective for hydrophobic contaminants and metals, which
preferentially associate with sedimenting material. A subset of possible chemicals should be
considered, to possibly include the IJC Critical Pollutant list. It is strongly recommended that initially
three cores per lake are sampled to obtain a detailed historical record of occurrence and input, as well
as obtain numerous surface samples for spatial coverage. The detailed cores should be dated by Pb-
210 and Cs-137, and normalized for focusing. This would not need to be repeated. Due to the slow
sedimentation rates in the Great Lakes, yearly surficial samples would not be used or cost-effective, as
little or no change may be detected. For yearly trends in accumulation and concentrations, sediment
traps should be deployed at master sites and time-integrated samples collected and analyzed. This will
provide a measure of shorter-term changes. The NOAA lab in Ann Arbor has an active sediment trap
program in several of the Great Lakes. Supporting measurements such as dry weight, porosity, organic
carbon, and particle size distribution also need to be made.
Variability: There will be a. certain analytical error associated with each chemical determination. In
addition, there is considerable spatial heterogeneity in bottom sediments. Some variability can be
reduced, such as normalizing the organic contaminant concentrations to sediment organic carbon, and
adjusting for sedimentation rate.
Primary Problems: Depending on the number of analytes, the analyses are expensive. Regarding
sediment traps, material needs to be collected or isolated frequently to limit degradation by natural
processes or grazing losses. Preserved material would likely need to be composited to have enough
material to analyze.
B-15
-------
EMAP - Great Lakes Indicator Fact Sheet
Indicator: Contaminant residue in fish
Category: Exposure indicator
Resource class: Offshore, nearshore, and harbor/embayment zones
Application: Fish can be used as bioindicators of the presence of persistent chemicals that
accumulate in organism lipids. Many of the chemicals of concern in the Great Lakes persist in the
environment because they are resistant to biological and chemical degradation processes. Often these
chemicals have low water solubilities and high octanol-water partition coefficients, resulting in a high
bioaccumulation potential. These same properties tend to give these chemicals toxic properties which
make consumption of contaminated fish a public health concern. Chemicals of concern in the Great
Lakes with such properties include PCBs, PCDDs, PCDFs, and chlorinated pesticides (DDT and
products, mirex, toxaphene, chlordane, dieldrin, nonachlor, heptachlor, heptachlor epoxide). Methyl
mercury is also of concern in fish; it accumulates in muscle tissue rather than lipids. Concentrations of
these chemicals in fish can be used to indicate the presence of contaminants in water at levels that
may be harmful to the environment, but at levels that cannot be easily measured directly in water.
They also can be used to directly indicate whether fish are safe for consumption, as indicated by health
advisories and FDA action levels. Fish contaminant concentrations are used to provide an early
warning indication of exposure for animals higher in the food chain, such as fish-eating birds and
mammals.
Index Period: The best collection period is late summer to early fall (August-September) prior to
spawning, when lipid stores and contaminants are at their peak in top predators.
Measurements: Whole-body contaminant burdens should be measured for PCBs, PCDDs, PCDFs,
and chlorinated pesticides. Top predators should be used as indicators, such as lake trout and walleye.
Young-of-the-year fish with narrow geographical range can be used to monitor local trends in
contaminants; choice of species would be site-specific. Lipid measurements should also be made, as
yearly variations in contaminant concentrations due to lipid differences need to be taken into account.
Variability: Precision and accuracy in data would depend on collection variability (spatial and
temporal) and on the variability associated with the analytical methodology used for a given chemical.
Contaminant concentrations are also a function of age and sex, and this variability should be limited as
much as possible in sample collection design. Some of the inherent variability in sampling can be
reduced by normalizing concentrations to lipid content.
Primary Problems: These analyses are expensive. Care should be taken to include emerging
problem chemicals in future monitoring.
B-16
-------
EMAP - Great Lakes Indicator Fact Sheet
Indicator: N/P/Si ratios
Category: Exposure indicator
Resource class: Offshore and nearshore zones
Application: The ratios of total nitrogen (TN) to total phosphorus (TP), and total reactive silicon (Si) to
total phosphorus (TP) have been proposed as indicators of conditions that favor blue-green algal
abundance (at the expense of more desirable algal groups such as diatoms). The ratio TN.TP is
indicative of favorable conditions for blue-green dominance when it is less than 29:1 (weight basis).
The ratio Si:P is indicative of unfavorable conditions for diatoms when it is less than 93:1 (molar basis).
Blue-green algae are the least desirable algal group not only because of their tendency to form mats
and scum on water surfaces and to cause taste and odor problems in potable water supplies, but also
because they are not preyed upon by zooplankton. If either of these ratios is below the threshold
value, eutrophic conditions are extremely likely to be present.
Index Period: After the spring diatom bloom (May) or the fall diatom bloom (October).
Measurements: These ratios may be calculated from basic water chemistry measurements that would
normally be part of any core sampling effort. There is sufficient data, especially for TN:TP, to establish
an historical baseline and to test this indicator.
Variability: The variability of this indicator depends on the variability of the measurements that are
used in the calculation. Usually the coefficient of variation (C.V.) for individual water chemistry
measurements in open waters will be less than 10%. In nearshore waters, this C.V. will be higher, and
in harbors and small embayments, it may be prohibitive.
Primary Problems: (1) Some lakes or areas of lakes may have other limiting factors besides
phosphorus and nitrogen that inhibit blue-greens; (2) If neither phosphorus nor nitrogen is limiting in a
lake, blue-greens may dominate at TN:TP ratios > 29.
References:
Smith, V.H. 1983. Low nitrogen to phosphorus ratios favor dominance by blue-green algae in lake
phytoplankton. Science 221:669-671.
Holm, N.P. and D.E. Armstrong. 1981. Role of nutrient limitation and competition in controlling the
populations of Asterionella Formosa and Microsystis aeruginosa in semicontinuous culture. Limnol.
Oceanogr. 26:622-634.
B-17
-------
EMAP - Great Lakes Indicator Fact Sheet
Indicator: Water column toxicity to Ceriodaphnia dubia/affinis
Category: Exposure indicator
Resource class: Offshore, nearshore, and harbor/embayment zones
Application: Water column toxicity tests are designed to indicate the integrated exposure of organisms
residing in ambient waters to toxic substances and suitability of the water column habitat for aquatic life
in tributaries, nearshore zones, and offshore waters. Acute and chronic toxicity tests are conducted to
reflect potential short-term and long-term impacts, respectively.
Water column toxicity is one indicator of water quality and is emphasized in the Water Quality Act and
Great Lakes Water Quality Agreement. Legislation provides the basis for a nontoxic environment,
striving toward zero discharge, protection of aquatic life, and risk reduction. Results of the water
column toxicity tests indicate the potential success or impairment of resident populations regarding
habitat condition. Similarly, these results can be used to assess the success of nonpoint and point
source control strategies on a regional basis. Selection of one or more water column toxicity tests will
be necessary to assess toxicity trends in ambient waters, exposure potential, habitat condition, and
control mechanisms.
The Ceriodaphnia dubia/affinis toxicity tests provide a surrogate for cladoceran populations and the
zooplankton community in the Great Lakes. Even though Ceriodaphnia does not occur in substantial
numbers in the Great Lakes basin, it appears to be indicative of zooplankton community exposure. The
zooplankton community is the link between primary producers and higher trophic levels, i.e.,
zooplanktivorous forage and immature predatory fishes. Zooplankton survival and reproduction is
critical to fish stock maintenance and production. The endpoints for the Ceriodaphnia toxicity tests are
lethality in the acute test and reproductive impairment in the chronic test.
Index Period: Acute and chronic effects can be measured during the late summer or early autumn,
when contaminant concentrations are likely to be at maxima during the low-flow period. This period
also approximates the period of young-of-the-year recruitment for many fish species when zooplankton
abundance is critical for survival.
Measurements: Standard methods for the Ceriodaphnia toxicity tests are available (ASTM 1988;
Weber et al. 1989). The Ceriodaphnia tests have been applied to various environmental samples in the
Great Lakes (Lien et al. 1986, McNaught and Mount 1986, White et al. 1989, Ankley et al. 1990) and
applied and evaluated nationally (Cowgill et al. 1984; 1985, Mount and Norberg 1984, Hamilton 1986,
DeGraeve and Cooney 1987, Takahashi et al. 1987, Winner 1988).
Whole water samples are collected using a Niskinฎ bottle and transported to the testing laboratory
within 24 to 48 h. Ceriodaphnia neonates (<24 h in age) are initially used in the acute and chronic
toxicity tests. Ten replicate test vessels in triplicate can be used in the assays (30 test vessels total).
Acute toxicity can be determined in 48- or 96-h tests. Chronic toxicity can be determined in the 7-day
test. Controls (10 test vessels) are conducted with every test series; 80% or greater survival and an
average of 15 or more young surviving per female in the controls and controls with at least 60% of
surviving females, producing a third brood is acceptable. Positive controls will also be conducted.
Aeration, daily static renewal, and feeding are required during the chronic test period. Daily monitoring
of physical and chemical factors is required. Lethal concentrations (LC 10:50) and effective
concentrations (EC10:50) for acute and chronic tests are possible using a dilutional series, however,
may not be necessary for the purposes of EMAP.
B-18
-------
EMAP - Great Lakes Indicator Fact Sheet
Lethality is expressed as a percentage with a standard deviation for acute test periods. Reproduction is
expressed as the number of neonates, broods, and percentages with standard deviations. Results of
both tests must be examined relative to controls. Proposed interpretation of data is that results with
greater than 10% mortality or reduction in neonate production is considered an effect. Less than 10%
is considered approximating a NOEL or within the potential variability of the control or mean of all
controls for a year. Ten to 30% impact is considered an effect, 30-50% a moderate effect, 50-75% a
great effect, and 75-100% a severe effect. Data treatment will result in indicating the percent of the
resource and habitat region which fall into the above toxic effect categories. Although some historical
data are available, the approach will primarily be comparisons to the controls.
Variability: It is expected that spatial variation in response within and among the Great Lakes (both
the lakes and habitat types) and temporal variation will far exceed the test variability.
Primary Problems: Specific delineation of test design for the purposes of EMAP and index of period
selection.
References:
American Society for Testing and Materials (ASTM). 1988. New standard guide for conducting acute
three brood, renewal toxicity tests with Ceriodaphnia dubia. American Society for Testing and
Materials, Draft No. 6, ASTM E47.01, Philadelphia, PA.
Ankley, G.T., A. Katko, and J. Arthur. 1990. Identification of ammonia as an important sediment-
associated toxicant in the lower Fox River and Green Bay, Wisconsin. Environ. Toxiciol. Chem. 9:313-
322.
Cowgill, U.M., IT. Takahaski, and S.L. Applegate. 1984. A comparison of the effect of four benchmark
chemical on Daphnia magna and Ceriodaphnia dubia-affinis tested at two different temperatures.
Environ. Toxicol. Chem. 4:414-422.
Cowgill, U.M., K.I. Keating, and IT. Takahaski. 1985. Fecundity and longevity of Ceriodaphnia dubia-
affinis in relation to diet at two different temperatures. J. Crust. Biol. 5(3):420-429.
DeGraeve, G.M. and J.D. Cooney. 1987. Ceriodaphnia: An update on effluent toxicity testing and
research needs. Environ. Toxicol. Chem. 6:331-333.
Hamilton, M.A. 1986. Statistical analysis of the cladoceran reproductivity test. Environ. Toxicol. Chem.
5:202-212.
Lien, G.J., K.E. Biesinger, L.E. Anderson, E.N. Leonard, and M.A. Gibbons. 1986. A toxicity evaluation
of lower Fox River water and sediments. EPA/600/3-86/008, U.S. Environmental Protection Agency,
ERL-Duluth:28 p.
McNaught, D.C. and D.I. Mount. 1986. Appropriate durations and measures for Ceriodaphnia dubia
toxicity tests, jn: Aquatic Toxicology and Hazard Assessment: Eighth Symposium, R.C. Bahner and
D.J. Hansen (Eds.), ASTM-STP 891, American Society for Testing and Materials, Philadelphia, PA:375-
381.
Mount, D.I. and T.J. Norberg. 1984. A seven-day life-cycle cladoceran test. Environ. Toxicol. Chem.
3:425-434.
B-19
-------
EMAP - Great Lakes Indicator Fact Sheet
Takahaski, IT., U.M. Cowgill, and P.G. Murphy. 1987. Comparison of ethanol toxicity to Daphnia
magna and Ceriodaphnia dubia tested at two different temperatures: static acute toxicity test results.
Bull. Environ. Contam. Toxicol. 39:236-299.
White, D.S., D.J. Jude, R.A. Moll, and J.A. Bowers. 1989. Exposure and biological effects of in-place
pollutants. U.S. Environmental Protection Agency, Office of Research and Development, ERL-Duluth,
MN, and LLRS-Grosse lie, Ml.
Weber, C.I., W.H. Peltier, T.J. Norberg-King, W.B. Homing, F.A. Kessler, J.R. Menkedick, T.A.
Neiheisel, P.A. Lewis, D.J. Klemm, Q.H. Pickering, E.L. Robinson, J.M. Lazorchak, L.J. Wymor, and
R.W. Freyberg. 1989. Short-term methods for estimating the chronic toxicity of effluents and receiving
waters to freshwater organisms. EPA/600/4-89/001, U.S. Environmental Protection Agency, Office of
Research and Development, EMSL-Cincinnati, OH:249 p.
Winner, R.W. 1988. Evaluation of the relative sensitivities of 7-d Daphnia magna and Ceriodaphnia
dubia toxicity tests for cadmium and sodium pentachlorophenate. Environ. Toxicol. Chem. 7:153-159.
B-20
-------
EMAP - Great Lakes Indicator Fact Sheet
Indicator: Water column and optical characteristics
Category: Habitat indicator
Resource class: Offshore, nearshore, and harbor/embayment zones
Application: This indicator provides exposure information for chemical and physical stressors of
aquatic biota. By using different subsets of the variables, as well as examining values for individual
variables, several potential stressors could be evaluated. For example, a subgroup of the variables
measured would be used to determine the state of eutrophication. Particularly, N, P, chlorophyll-a, and
dissolved oxygen measurements would provide an indicator of the state of eutrophication (Dillon and
Rigler 1974, Smith 1982, Vollenweider 1968). Dissolved oxygen measurements would also measure a
stressor of fish and other aquatic biota (Makarewicz and Bertram 1991). In nearshore areas,
conductivity, conservative variables (Na, Cl, SO4) and measures of particulates could be used to
investigate water masses and tributary inputs potentially associated with nutrient and toxic loadings.
Index Period: The index period would coincide with biological sampling in late July and August. This
period coincides with stable thermal stratification, and in the Great Lakes, as stable a biological period
as any except winter.
Measurements: Variables that would be measured are: Total P, Total N, NO2-NO3, Si, Na, Mg, Ca,
SO4, Cl, Dissolved oxygen, pH, Secchi depth, conductivity, transmissivity, temperature, in situ
chlorophyll, TSS, DOC, and POC. A subset of these variables (conductivity, temperature, pH, DO,
transmissivity, and chlorophyll) would be measured with a Sea Bird Electronics, Inc. CTD cast. Other
variables would be measured with discrete water samples taken in mid-epilimnion, metalimnion, and
hypolimnion.
Variability: The magnitude of spatial variability is a function of each variable. In the open waters of
the Great Lakes, variability will be considerably less than in nearshore or harbors and embayments. As
a rough estimate, one could expect a coefficient of variation (cv) of between 10 and 30% for offshore
areas and as much as 100% for nearshore areas. Between lake variability will be higher.
Primary Problems: The primary problem involved in measuring these variables is the problem of
sampling once per year. There will be some additional variability incorporated in the data because of
climatic conditions varying greatly from the average (e.g., a particularly cold or cloudy summer) which
may have the effect of changing the rates of ecosystem processes and community changes. This is a
problem for interannual comparisons as well as for interpretation of exposure in any year. A possible
solution to this problem is implementation of low cost monitoring, on a daily basis, of easily measured
physical and chemical parameters as is done at municipal water intake plants.
References:
Dillon, P.J. and F.H. Rigler. 1974. The phosphorus-chlorophyll relationship in lakes. Limnol.
Oceanogr. 19:767-773.
Makarewicz, J.C. and P. Bertram. 1991. Evidence for the restoration of the Lake Erie ecosystem.
BioScience 41(4):216-223.
Smith, V.H. 1982. The nitrogen and phosphorus dependence of algal biomass in lakes: An empirical
and theoretical analysis. Limnol. Oceanogr. 27:1101-1112.
B-21
-------
EMAP - Great Lakes Indicator Fact Sheet
Vollenweider, R.A. 1968. Scientific fundamentals of the eutrophication of lakes and flowing waters with
particular reference to nitrogen and phosphorus as factors in eutrophication. DAS/CS168-27,
Organization for Economic Cooperation and Development, Paris.
U.S. Environmental Protection Agency
Region 5, Library (PL-12J)
77 West Jackson Boulevard, 12th Ftanr
Chicago, IL 60604-3590
B-22 $ U.S. GOVERNMENT PRINTING OFFICE 1993 - 750-OQ2/3Q234
------- |