GREAT LAKES BASIN RISK
CHARACTERIZATION STUDY
Lake Superior
Huron
CHICAGO/GARY
Pennsytvani*
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Page I - 3
IV. PRIMARY AIR AND WASTE PROBLEM AREAS
M. Sulfur and Nitrogen Oxide Emissions IV - 1
N. Hazardous/Toxic Air Pollutants -11
O. Active RCRA Hazardous Waste Management Facilities -16
P. Abandoned Hazardous Waste Sites - 28
Q. Municipal Solid Waste Facilities - 36
R. Industrial Solid Waste Facilities - 45
V. SECONDARY PROBLEM AREAS
S. Aggregated Drinking Water V • 1
T. Aggregated Ground Water - 20
U. Ozone and Caibon Monoxide - 25
V. Particulate Emissions • 31
W. Underground Storage Tanks - 35
X. Pesticides: Human Health - 43
VI. APPENDICES
A. Data Sources, Uncertainties, and Gaps VI - 2
B. Bibliography - 12
C. Acronyms - 20
D. Annex I List of Great Lakes Substances - (insert)
E. Descriptions of United States Areas of Concern - 23
F. Accidental Nuclear Releases - 33
G. Mercury Trends - 37
-------
Page I - 4
A. INTRODUCTION
1. EXECUTIVE SUMMARY
PURPOSE OF THE STUDY
The Great Lakes, the largest freshwater ecosystem in the world, has received a variety of chemical and physical
impacts from human activities, particularly those involving industry, agriculture and urbanization. The eight states
which cover the Great Lake watershed (Minnesota, Wisconsin, Illinois, Michigan, Indiana, Ohio, Pennsylvania and
New York) contain diverse ecosystems including inland lakes, sand dunes, wetlands, old growth and secoad growth
forests, agricultural lands and urban areas. Approximately 30 million people live and work in the Great Lakes Basin.
While progress has been made to reduce chemical contamination in the Great Lakes, the ecosystem and its
inhabitants still face significant environmental risks. Ibis study is the first attempt to apply the methodology of
EPA's comparative risk process to assess and rank such risks in the United States portion of the Great Lakes Basin.
The study provides EPA Regions 2,3, and 5 and the Great Lakes States with a risk-based approach which can assist
in protecting and restoring this vast yet sensitive ecosystem and reducing risks to its human cohabitants. The Great
Lakes National Program Office (GLNPO) and the three EPA Regions will use the results of this study to better target
their activities. Study results are integrated into the Great Lakes Five Year Strategic Plan.
The study will also be used in conjunction with restoration efforts, such as implementing Remedial Action Plans for
Great Lakes Areas of Concern and Lakewide Management Plans, to help evaluate the relative environmental risks
being addressed by these and other activities.
PROBLEM AREA DESCRIPTIONS
The study is organized into twenty-three separate topics (problem areas). These problem areas generally reflect EPA
programmatic priorities and therefore represent mixed sources, receptors, media and regulatory obligations. However,
three problem areas reflect issues of particular importance to the Great Lakes: Introduction of Exotic Species; Toxic
Sediments; and Changing Lake Levels. Data is sketchy for some of these problem areas (e.g., Underground Storage
Tanks). Further information could cause the evaluations of problem areas to change.
The study groups problem areas into four categories:
• Emerging Basin-wide Issues are potentially catastrophic risks to human and ecological populations:
• Physical Degradation of Water and Wetland Habitats,
• Physical Degradation of Terrestrial Ecosystems,
. Introduction of Exotic Species,
• Spills and Other Accidental Releases,
• Global Climate Change, and
. Changing Lake Levels.
• Primary Water Problem Areas are sources of pollutant loadings which pose significant risks to aquatic
ecosystems and to human health through fish consumption: Industrial Point Source Discharges, Municipal
Point Source Discharges, Nonpoint Source Loadings, Atmospheric Loadings, Toxic Sediments, and
Pesticides Discharges. Due to the difficulty in apportioning the exact share of effects to specific types of
-------
Page I - 5
sources, human health and environmental risks from these sources are discussed in the aggregate. An
informed, but speculative apportionment of risks among types sources has been attempted to estimate their
relative contributions.
Primary Air and Waste Problem Areas are sources which are also considered to pose potentially
significant risks: Hazardous/Toxic Air Pollutants, Abandoned Hazardous Waste Facilities, Sulfur and
Nitrogen Oxide Emissions, Active RCRA Hazardous Waste Management Facilities, Municipal Solid Waste
Facilities, and Industrial Solid Waste Facilities.
Secondary Problem Areas are believed to present lower risks: Aggregated Drinking Water, Aggregated
Ground Water, Ozone and Carbon Monoxide, Particulate Emissions, Underground Storage Tanks, Pesticides
(Applications and Food Residues).
METHODOLOGY, DATA GAPS AND UNCERTAINTY
The general methodology used in EPA's 1987 risk project, Unfinished Business, and in Regional comparative risk
studies was used to characterize risks found in the Basin. The study uses data on a county-wide basis within the
watershed of the Great Lakes system. Each problem area analyzes the magnitude of the problem, toxicity of
contaminants (where applicable), and exposure levels. Three general categories of impacts are considered: cancer
human health risks, non-cancer human health risks, and ecological risks. In some cases, welfare impacts (i.e.,
economic damages) were also considered. Risks are those that are believed to be in effect today, after the
implementation of remedial programs.
Risk estimates are made in a semi-quantitative manner whenever possible. However, there are numerous
uncertainties and gaps in available data, leaving significant uncertainties in the risk characterization and underscoring
the need for further environmental data and integration of data. Data gaps are addressed in Appendix A.
Problems associated with Indoor Air, Indoor Radon, and Stratospheric Ozone were not addressed because this study
focuses on environmental issues specific to the Great Lakes. Discussion of these issues are available in the Regional
analyses.
RESULTS
Not surprisingly, both human health and ecological risks were found to be most significant where human population/
are greatest. Large urban and industrial areas such as Chicago/Gary, Cleveland, Detroit and Buffalo/Niagara Falls
have multiple significant sources of environmental contaminants. A spatial relationship between high ecological risks
and risks to human health was also found. High ecological risks should be viewed as sentinels of potential adverse
human health effects.
It is clear that the Great Lakes, because of their unique properties, are particularly vulnerable to the effects of toxic
substances, regardless of the pollution source. When persistent toxic chemicals, such as PCBs and mercury, enter
the Great Lakes, they are not readily removed. Unlike rivers or coastal areas, where pollutants are borne off to do
damage elsewhere, the retention time of the lakes vary from 3 years (Erie) to 200 years (Superior).
The most significant ecological and human health risks in the Great Lakes were identified by following the
approaches used in the Region 5 Comparative Risk Project. Significant risks which are directly associated with the
Great Lakes are summarized below. Other risks, which are significant in the Basin, but not directly limited to the
Great Lakes, are also briefly described.
-------
Page I - 6
A. ECOLOGICAL RISKS
Significant ecological risks directly associated with the Great Lakes themselves were found to be:
• Persistent toxic contaminants, from all sources, bioaocumulate in fish and wildlife causing birth
deformities, tumors, and reproductive problems for species in a number of areas in the Great I -nVr-t
Basin.
• Nonpoint source loadings to surface waters are currently a significant source of ecological risk to
the Basin.
• Toxic sediments are a significant reservoir of past contamination that contributes to poisoning the
Great Lakes food web.
• Industrial and Municipal water discharges in many cases cause substantial impacts to aquatic
ecosystems.
• Spflis and other accidental releases could cause devastating and widespread damage to aquatic and
terrestrial ecosystems.
• Exotic species, especially the zebra mussel, could impose catastrophic effects on the great I
food web.
Other significant ecological risks include:
• Agricultural, residential and other development activities, destroy wildlife habitats (e.g. fish
spawning areas, wetlands, prairies, and old-growth forests) at disturbing rates.
• Global climate change presents potentially significant impacts in the Great Lakes Basin on fish and
wildlife, forestry and agriculture, although estimates of impacts are highly uncertain.
• Atmospheric sources of sulfur and nitrogen oxides, and mercury were found to affect a large
number of inland lakes within the watershed.
B. HUMAN HEALTH RISKS
Significant human health risks directly associated with the Great Lakes are:
• Risks from consumption of sport fish were linked to industrial and municipal point source
discharges, nonpoint sources, atmospheric deposition and toxic sediments. The majority of the risk
is presented by exposure to PCBs. Other contaminants of high concern include chlordane,
mercury, dioxins, mirex and DDT.
• Potential spills and other accidental releases present risks to human health.
Other significant risks to human health in the watershed include:
• Risks from exposure to hazardous/toxic air pollutants and to sulfur and nitrogen oxides.
• Risks from consumption of sport fish from inland lakes due to mercury contamination.
-------
Page I - 7
RECOMMENDATIONS
The importance of several measures which are now being pursued by EPA and States are supported by this study.
This study recommends that EPA:
• Institute a Lake wide Management Process that identifies environmental objectives for each Lake
and load reductions necessary to achieve those objectives. Management priorities should be based
upon risk reduction, including prioritization of Areas of Concern for remediation.
• Give priority to reducing chemical releases from locations where human and ecological risks are
high such as Chicago/Gary, Detroit, Cleveland and Buffalo/Niagara Falls (see figure).
CHICAGO/0ARY
Pmnnmyivania
HUnoh
• Focus future risk analyses on a geographical basis (e.g., through Remedial Action Plans) to
characterize the type and magnitude of risks and to help decision-making.
• Increase integration of efforts to reduce ecological risks, within EPA and among other agencies to
provide for maximum protection of wetlands, spawning areas, and other wildlife habitats.
• Increase coordination of efforts to reduce human health risks, including the development of risk-
based fish advisories which are consistent across the lakes.
• Develop environmental indicators to judge the health of the environment and to monitor the
efficacy of programs. Indicators would be chemical and biological, including: achievement of
water quality standards, the restoration of beneficial uses, reduction in fish consumption advisories,
reduction in toxic chemical loadings/emissions, and measurements of contaminant concentrations
in toxicological and environmental endpoints, and achievement of stable populations of species that
are indicators of ecosystem health such as bald eagles and lake trout.
These recommendations are consistent with the ten recommendations of the SAB's 1990 report, Reducing Risk:
Setting Priorities and Strategies for Environmental Protection.
-------
Page 1-8
2. Purpose of the Study
This study is designed to support ongoing Great Lakes protection and planning efforts conducted by EPA Regions
2, 3, and 5 and the Great Lakes National Program Office (GLNPO). Such efforts include, but are not limited to,
identifying Great Lakes Areas of Concern (AOCs), preparing and implementing Remedial Action Plans to restore
beneficial uses to AOCs, developing and implementing Lakewide Management Plans to eliminate effects of key
pollutants, and designing and implementing strategic plans for the Great Lakes National Program Office. This study
characterizes the relative environmental risks to be addressed by these and other ongoing activities.
The study attempts to characterize, quantitatively where possible, the nature and extent of current health and
environmental risks within the United States portion of the Great Lakes Basin. It follows the same general approach
developed for previous comparative risk studies by EPA Headquarters and the Regions. This report describes the
current risks and provides, where available, a semi-quantitative measure of the health and environmental risks
associated with each problem area.
The Great Lakes Basin is defined hydrologically, including all watersheds within the Great Lakes system. Because
the Basin includes portions of both Canada and the United States, each country is home to sources of risk that impact
both sides of the border. Therefore, both U.S. and Canadian populations face potentially similar risks. However,
depending upon the problem area, risks to OwHinn populations and ecosystems could be higher or lower than those
to U.S. populations. While similar risks are found in both portions of the Basin, Canadian human health and
ecological impacts are not considered in this study. Therefore, references to the Basin contained in this report refer
solely to the United States portion.
In order to relate more directly to currently significant Great Lakes Basin issues, this study addresses several problem
areas that either were not considered or were only minimally discussed in previous EPA risk analyses. These
problem areas include impacts from exotic species, contaminated sediments, and changing lake levels. In addition,
some problem areas that are discussed in other Agency or State comparative risk studies are not included in this
analysis because they were considered to have only a minimal impact on the Basin.
3. Organization of the Report
The study is organized into several sections. Section I includes this introduction, a description of the approach "srd
and data issues that arose in the study, an overview of the Great Lakes Basin and its population, and a brief
characterization of the current health of Great Lakes and its ecosystems. Risk characterizations, the primary
emphasis of the study, are found in Section II, Emerging Basin-wide Issues, Section HI, Primary Water Problem
Areas, Section IV, Primary Air and Waste Problem Areas, and Section V, Secondary Problem Areas. The problem
areas examined are presented in a preliminary order to reflect risks which are believed to be most significant and
requiring the particular attention of Great Lakes planners.
-------
Page I - 9
B. GREAT LAKES BASIN PROBLEM AREAS: APPROACH AND DATA ISSUES
1. Overview of the Study Approach
This study is intended to build upon the comparative risk studies that have been completed in Regions 2, 3, and 5.
Many of the analytical and risk assessment approaches described in those studies were adopted as guidelines for this
analysis. In particular, the following guidelines were followed:
¦ When available, health risks were derived from environmental data and measurements of exposed
populations rather than on modeled estimations.
¦ Average human health risks were derived by weighting the risks for populations exposed to ranges in
chemical contaminants. Risks were not computed based upon the maximally exposed individual.
¦ Chemicals or case studies were used which were considered most representative of the risks posed by the
problem areas.
¦ A risk assessment procedure consisting of toxicity assessment, exposure assessment, and risk
characterization was applied. EPA cancer potency factors and reference doses for non-cancer effects were
used.
¦ Ecological risk was assessed using EPA's approach that analyzes area affected, severity, and reversibility.
¦ Risk analyses were based solely on current risks with the exception of Spills and Other Accidental Releases,
Global Climate Change, and Introduction of Exotic Species.
¦ Uncertainties in the risk analyses, including data gaps, were identified.
However, this study differs from the previous work primarily in its focus on the Great Lakes Basin. This focus
means that the human health and environmental risks as well as their sources are assessed in terms of the populations
residing in the United States Great Lakes Basin oounties. Where data are available, the number and type of problem
sources or activities in each of the Great Lakes Basin counties are depicted. If county-specific information is not
available, Basin-wide estimates are noted for each problem area. This reflects a new attempt to quantify the actual
level of potential or residual human health and environmental risk found within the United States portion of the Great
Lakes Basin.
While the Great Lakes Basin is defined by watershed, most Basin data was collected at the county level. Therefore,
data was collected on all counties which fall within or contain a portion of the Basin watershed. The population of
the Great Lakes Basin as presented in this study was determined by summing the populations of the counties within
the Great Lakes' watershed. This yields a United States Great Lakes Basin population of approximately 30,050,000
people in 1988. This figure is used as the "Basin population" in the problem area risk assessments.
Most of the problem areas used in this were defined by EPA Headquarters. They correspond generally with existing
environmental programs, statutes, or budget categories, such as RCRA hazardous waste management facilities,
industrial point source discharges regulated under the Clean Water Act, and toxic air emissions regulated under the
Clean Air Act. These programs are reducing substantial risks. However, the problem areas included here
characterize current, or residual risks, or those environmental and health risks that remain beyond those presently
controlled by those environmental programs or statutes.
Because of the correspondence with EPA programs or activities, some problem areas have been defined as sources
of contaminants, while other as pathways or receptors of those contaminants. This leads to some inconsistencies in
-------
Page I - 10
methodology. For example, portions of some problem areas may overlap with several others, as is the case with a
problem area such as pesticides, which enters the Basin through several pathways, including atmospheric deposition,
sediments, and nonpoint source discharges to surface waters. Sources may contribute to environmental or human
health risks more than pathways. Future risk characterizations will benefit greatly from a redefinition of problem
areas by endpoints for more consistency in the risk analysis.
The problem area discussions separate human health and ecological risks among the various sources of environmental
contamination and, where possible, attempt to avoid double counting of risks. For example, risks in the Great Lakes
Basin surface waters are divided by contaminant source including industrial point sources, municipal point sources,
nonpoint sources, and atmospheric deposition. Furthermore, toxic sediments and contaminated ground-water
discharge are also recognized as sources of contamination. However, their relative contribution to health and
ecological risks from surface water was difficult to quantify. As a result, these problem areas are addressed
separately. Where possible, the difficulties inherent in differentiating among human health and ecological risk across
the problem areas are also discussed.
Under the problem areas, four different types of risk are considered, including cancer and non-cancer human health
risk, ecological impacts, and welfare effects. These risks are expressed in terms of potential cancer cases that are
estimated to be caused by the problem area. In several of the problem areas, human health or environmental risks
are described as possible impacts rather than as documented current risks.
Sections II through V characterize the magnitude of risks posed by twenty-three problem areas relative to the Great
Lakes Basin. Under each section, the activities addressed by the problem area, the relative magnitude of the problem
area sources or activities, and estimates of human health, environmental, and welfare impacts are outlined in each
subsection A through X.
Indoor Air, Indoor Radon, Radiation and Stratospheric Ozone problem areas were addressed by the risk analyses
conducted Regions 2,3 and 5. Discussions of these problem areas are available in the Regional risk project reports,
and consequently, are not addressed in this report.
Other risks to the Great Lakes Basin and its populations have not been addressed due to time constraints. These risks
include:
¦ Risks from stratospheric ozone to plankton and other links within the Great Lakes food web;
¦ Risks associated with consumption of wild rice, cranberries, and other foods gathered by Native
Americans in the Great Lakes Basin;
¦ Risks stemming from mobile sources of toxic air emissions within the Rasin
2. Problem Area Organization
Problem areas in Section II, Emerging Basin-wide Issues, are very important to Great I*'Basin human health,
ecological populations, and economic welfare: forms of habitat destruction and their causes, exotic species, and other
problem areas that reflect potentially catastrophic risks.
Sections HI and IV, Primary Problem Areas, present water, waste and air problem areas that represent primary
sources of risks to the Basin and its populations. The diverse and complex nature of atmospheric loadings, toxic
sediments, and other nonpoint sources, their effects, and probable remediation methods within the Great Lakes Basin
have justified separate discussions. It is very difficult at this time to accurately apportion risks from contaminants
found in surface waters to specific sources or to industrial and municipal point source discharges to surface water.
A rough estimate has been attempted to show possible proportions. However, due the difficulties in accurately
attributing health, ecological and welfare effects to specific sources, this study groups the three nonpoint source and
two point source problem areas together as primary water concerns, and discusses the effects only once, in the first
-------
Page I -11
of the five, Industrial Point Source Discharges to Surface Water.
Section V, Secondary Problem Areas, presents water, air and waste problem areas which are believed to be of
secondary concern to the Basin. This belief is partly predicated on professional judgment, but was influenced by
the lack of data. With more information, some of these risks may actually shift in priority for the Basin.
3. Data Uncertainties
Until this study, no generalized attempt has been made to use EPA's comparative risk approach to quantitatively
characterize all the major human health, ecological, and welfare risks in the Great Lakes Basin. The ground-breaking
nature of this work has, however, meant several uncertainties and data gaps were inevitable. The risks which are
derived for several of these problem areas are subject to considerable uncertainty.
Treatment of many risks in the study was restricted to qualitative discussions. Often the data needed to conduct an
ideal quantitative risk analysis simply does not exist. In addition, the data which does exist was generally created
for other purposes, and is often ill-suited for risk analysis. These factors create significant levels of uncertainty in
many of the preliminary results shown here. To help clarify these data issues, the databases used, their weaknesses,
and major data gaps that exist are discussed in Appendix A.
-------
Exhibit 1
UNITED STATES
GREAT LAKES
Lake Superior
AREA
St Marys
ver
Michigan
St Laurence
River
U.S.
Great Lakes
Basin
Boundary
Michigan \ Lake Huron
Lake Ontario
fiagata * ROCHESTER
UFFALO J New York
Buff ah RlvBr
Wisconsin
GRAND
RAPIDS
Minnesota
SAGINAW Vito*
Lake Jwwr
stciair /Lake
DETROIT y—
Dsfro# jkW JkERIE
River
MILWAUKEE Lake
Michigan
CHICAGO
Illinois
Pennsylvania
TOLEDO CLEVELAND
Ohio
^GARY
Indiana
0 30 60 90 120
INDEX
MAP
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Page I - 12
C GREAT LAKES BASIN AND POPULATION
1. A Description of The Great Lakes Basin1
The Great Lakes Basin encompasses all or parts of eight
states: Michigan, Minnesota, Wisconsin, Illinois, Indiana,
Ohio, Pennsylvania, and New York (see Exhibit 1). The
Lakes have a total shore line of over 4,000 miles within the
United States and cover an area of 94,250 square miles.
This makes the Great Lakes the largest freshwater system
in the world, with about 18 percent of the world's supply of
fresh surface water.
In addition to expansive urban areas, the Great Lakes Basin
is home to numerous ecosystems including lakes, wetlands,
dunes, old growth and second growth forests, prairies, and
agricultural lands. Much of the Basin is covered with
glacial-remnant freshwater lakes. There are four general
types of wetlands in the Basin: fens, marshes, bogs, and
swamps. Fens and bogs both develop in shallow stagnant
water. Fens are characterized by grasses and shrubs while
bogs are generally mossy. Fens and bogs usually occur in the cooler areas of the Basin. Swamps and marshes
contain trees and shrubs which are rooted in the sediment below the water. Marshes often occur in ponds or bays.
Swamps and marshes occur in the warmer areas and are seen mostly in the southern and eastern sections of the
Basin.
Boreal, mixed and deciduous forests cover much of the northern half of the Basin. Exhibit 2 shows the relative '
amounts of forest acreage by sections of each Great Lake
watershed. However, much of the original forests in the
Basin have been clear-cut for logging, agricultural and
residential purposes, and this decline in forest land is
continuing today.
Much land has been cleared throughout the Basin for
agriculture. Exhibit 3 illustrates the geographic distribution
of agricultural acreage in the Basin by county. Exhibit 4
shows the relative amounts of cropland by section of each
Great Lake watershed. Exhibit 5 depicts trends in Basin
agricultural land from 1984 to 1989. Basin soils support a
variety of crops. However, in addition to habitat
destruction, intensive agricultural land use has been a major
contributor to the degradation of the Basin through soil
erosion, fertilizer applications, and the use of pesticides
which has lead in some instances to both surface water and
ground water contamination.
1 Much of the information presented in this section hss been taken from The Great Lakes: An Environmental Atlas and Resource Book.
U.S. EPA and Environment Canada, 1988.
GREAT LAKES BASIN LAND USE
W Lata 8uparior
8 Latka Superior
iMfchfcan
8WI *- " —
8E I
NEI
NWLataHuon
8W Lata Huron
8tCUr-DMolRV
W Lata Eria
8 Lata Eria
E Lata* Eria
8W Lata Ontario
8 E Lata Ontario
NE Lata Ontario
SOURCE; HPt, 1887
8 3
ACRES
Exhibit 2
CULTIVATED CROPLAND
GREAT LAKES BASIN
W Lata Superior
8 Laka Superior
NW Lata Michigan
8W Lata Michigan
CC 1 - ' - »»¦ ' 1
5>t La mi Mcrugan
NE Lata Michigan
m
UN Lata Huron
i
8W Lata Huron
¦¦¦¦¦¦
8t Ctar-Oatrofe
1AI 1 Ed.
« 1 —too
S Lata Eria
E Lata Eria
¦
8W Lata Ontario
¦n
8E Lata Ortario
¦¦1
NE Lata Ontario
1
1 2 3 4 8 8 7
ACRES (MtMiont)
SOURCE: NRI, 1907
Exhibit 4
-------
Exhibit 3 TOTAL AGRICULTURAL LAND
«
Lake Superior
Lake Huron
Lake >
Michigan
Great Lakes Basin Counties
ACRES
0- 10,000
10,001 - 50,000
50,001 - 100,000
100,001 - 150,000
150,001 -250,000
>250,000
ERA
W
Source: National Resources Inventory, 1982
Soil Conservation Service
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Page I - 13
The urban ecosystem is perhaps the most evident setting in
the Basin. Uiban areas are prevalent throughout the
southern end of Lake Michigan, Lake St Clair, and
surround Lakes Erie and Ontario. The urban ecosystem
also contributes greatly to the degradation of the Basin
through manufacturing and industrial pollution, urban
runoff, and municipal wastewater discharge.
2. The Five Great Lakes
Lake Superior
The largest Great Lake in both surface area and volume is
Lake Superior. Lake Superior has an area of 31,700 square
miles, an average depth of 483 feet, and a volume of 2,900
cubic miles. Superior has a long retention time of about
190 years. About 90% of Lake Superior's basin is forested with only 3% under cultivation due to a cool climate
and poor soils. The surrounding forests and the sparse population of the lake's basin (approximately three-quarters
of a million people in the United States) help insure that relatively few pollutants enter Lake Superior, except through
significant atmospheric loading. Lake Superior flows into Lake Huron by way of the St. Mary's River. The
population of the Lake Superior basin is economically dependent on its rich natural resources, including mineral
extraction, and forests. The area also receives a significant amount of its income from recreation and tourism.
Lake Michigan
Lake Michigan is the second largest of the Great Lakes by volume with 1,180 cubic miles, and is the only Great
Lake that does not share a shore with Canada. It has an area of 22,300 square miles, an average depth of 280 feet,
and a retention time of about 100 years. The northern part of the lake is in the less developed upper Great Lakes
region. This region is sparsely populated, except for Green Bay. While the Bay has one of the most productive
fisheries in the Great Lakes, it also receives wastes from the world's largest concentration of pulp and paper mills.
The southern part of the Lake is much more urbanized, with cities such as Chicago, Milwaukee, and Gary. These
population centers contribute much to the pollution of the Lake. Lake Michigan also has the world's largest
freshwater dunes. The basin contains the most farmland of all the Great Lakes Basin, with 44% of its drainage basin
devoted to agriculture, while 41% is forested and 9% is residential.
Lake Huron
Lake Huron has the longest shoreline of the Great Lakes and includes thousands of islands. It is the second largest
Great Lake with a surface area of 23,000 square miles. It has an average depth of 195 feet, a volume of 850 cubic
miles, and a retention time of about 20 years. Approximately 68% of Lake Huron's basin is forested, especially in
the north, and 27% is devoted to agriculture. The southern portion of the Lake leads to Detroit via the St Gair
River. The Saginaw River watershed is intensively fanned and flows through Flint, Saginaw and Bay City
metropolitan areas into Saginaw Bay. This bay, like Green Bay, contains a very productive fishery. The Lake Huron
basin has the second smallest population in the Great Lakes, 2.4 million people.
Lake Erie
Lake Erie has the smallest volume of the Great Lakes with 116 cubic miles. It has an area of about 9,900 square
miles, and an average depth of only 62 feet Almost all of Lake Erie's water comes from the Northern Lakes via
the Detroit River. Seventeen metropolitan areas of over 50,000 people are located within Erie's basin, with a total
of 11 million people. It is the shallowest Great Lakes and therefore warms rapidly in the spring and summer and
GREAT LAKES BASIN
AGRICULTURAL LAND*
Millions of Ac rat
Exhibit 5
-------
Page I -14
frequently freezes over in the winter. It also has the shortest retention time of the lakes, 2.6 yean. The western fifth
of the basin is very shallow with an average depth of only 24 feet Nonpoint source and municipal discharges, in
conjunction with the shallowness of Lake Erie, put the lake in the greatest danger of eutrophication. Land use in
the lake's drainage basin is highly agricultural, with 67% of it used for fanning. Foresting accounts for 21% of die
basin's use, and 10% is residential.
Lake Ontario
The Niagara River Qows from Lake Erie into Lake Ontario, which then empties into the St Lawrence River.
Ontario has an area of 7,340 square miles, an average depth of 283 feet, and a volume of about 390 cubic miles.
Although the smallest in surface area, Lake Ontario contains more than three times the water volume of Erie. The
lake has a retention time of about 6 years. The Lake Ontario basin is largely rural, with a few large urban areas,
particularly on the Canadian shores. Land use in the drainage basin includes 49% forestry, 39% agriculture, and 7%
residential. The population of Lake Ontario's drainage basin is approximately 13 million people.
3. Basin Populations
Population declined in a number of the Great Lakes Basin Counties from 1980 to 1986. Overall Basin county
population declined by about 37,000 during that period. Exhibit 6 depicts the population changes of the Great Lakes
Basin counties by state. Much of this decline can be attributed to industrial relocation. However, not all Basin states
have had population declines. Counties in Illinois and Wisconsin have shown growth. This may be attributed to
the increased manufacturing and service industries in the Kenosha, Milwaukee and Green Bay metropolitan areas.
Population characteristics of the Great Lakes Basin States are further summarized in Exhibit 6. The geographic
distribution of the 1988 Basin population is depicted in Exhibit 7. Population density patterns, in terms of population
per square mile, mirror the general distribution of population by county as shown in Exhibit 8. The percent change
in county population in the Basin is depicted in Exhibit 9.
There are several Indian Tribes currently located upon the United States shoreline of Great Lakes: Grand Portage,
MN, Bad River, WI, Red Cliff, WI, Keweenaw Bay, MI, Bay Mills, MI, and Sault Ste. Marie, MI are located on
Lake Superior, The Grand Traverse Band of Ottawa and Chippewa and the Oneida Tribe near Green Bay, WI, are
located on Lake Michigan; The St Regis Mohawk Tribe is located on the St Lawrence River, NY; the Seneca
Tribe near Irving, NY, is located on Cattaragus Creek, which flows into Lake Erie, and the Seneca and Tuscarora
Tribes are located within die Great Lakes watershed, near Buffalo, NY.
Heavy industry and manufacturing are not the only prominent industries in the Basin. Agricultural activities are the
mainstay in much of the Basin. Because many agricultural practices rely heavily on pesticide use to control insects,
weeds, and fungi, contamination of ground and surface waters from pesticides is a growing concern in the Basin.
This problem is particularly noticeable on the Southern end of Lake Huron and in the Lake Erie basin.
In addition to agricultural activities, the lakes are commercially fished for whitefish and smelt Historically, the
fisheries of the Great Lakes have held enormous significance for the Basin economies. In the 1960s and 1970s,
commercial fishing declined due to over-fishing and the bioaccumulation of unacceptable levels of pesticides, PCBs
and other contaminants in the fish. Since then, commercial ffoh'"g activity has failed to regain its former share of
the Basin economy.
-------
Page 1-15
EXHIBIT €
STATE NO. SHORELINE POPULATION % BASIN POPULATION * POP POPULATION
COUNTIES IN MILES IN BASIN POPULATION GROWTH GROWTH DENSITY
IN BASIN COUNTIES (19M-19M) PEOPLE/MI1
(mllllnni)
(MM)
ILLINOIS
2
71
5.78
19j0
+85.584
+150
3293.95
INDIANA
14
31
1.70
5.6
414,570
+0.86
245.49
MICHIGAN
78
2552
9.24
305
-22,244
-030
17334
MINNESOTA
7
175
032
1.0
•24,944
-730
1630
NEW YORK
32
561
438
14.5
-63,038
-150
180.74
OHIO
34
294
5.23
173
•88,925
-1.70
346.72
PENNSYLVANIA
3
46
038
13
-5375
-1.40
150.17
WISCONSIN
37
768
3.21
10.6
+67,452
+2.10
218.40
One of die most significant industries in the Basin is the recreation and sports industry. The Great Lakes Fishery
Commission estimates that sport fishermen spent about $2 billion in 1985 on sport fishing in the Great Lakes for
perch, trout, bass, coho and Chinook salmon.2 However, it has been limited because of fish contamination and fish
consumption advisories within the Basin. Michigan, Minnesota and Wisconsin maintain large recreation areas.
Additionally, there are many national parks, recreation areas, and underwater preserves scattered throughout the
Basin.
1 Grot Likcr Grot Legacy?. He Conservation Foundation, 1990; page 150.
-------
Exhibit 7
~
~
:::::::::
iiiii:;;:
Hi:::!::
ililllli;
:::::::::
LAKE SUPERIOR
HE
POPULATION
< 100000
100000- 499999
500000- 999999
1000000 - 2000000
> 2000000
GREAT LAKES BASIN
COUNTIES
1988 POPULATION1
LAKE
HURON
^LAKE
ONTARIO
LAKE
ERIE
UNITED STATES
ENVIIONMENTAL MOTECTION ACENCT
GHAT IAKES NATIONAL PtOGIAM OFFICE
&EPA
1 County Population Estimates:
July 1,1988,1987,1986
BUREAU OF THE CENSUS
-------
POPULATION
PER
SQUARE MILE
0- 100
101 - 500
501 - 750
751 -1,000
1,001 -3,000
> 3,000
1
I
¦
Lake Huron
Lake Ontario
1986
POPULATION DENSITY
Great Lakes Basin Counties
Source: Bureau of the Census
County Population Estimates
July 1,1988,1987,1986
& EPAfcr
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Exhibit 9
GREAT LAKES STATES
1980 -1988 PERCENT CHANGE
IN COUNTY POPULATION
LAKE SUPERIOR
LAKE
HURON
LAKE
MICHIGAN
LAK
ONTARIO
WMm
LAKE
ERIE
PERCENT CHANGE
-10- 0
0-+10
+10-+20
>+20
SOURCE: Bureau of the Census
County Population Estimates
July 1, 1988, 1987, 1986
UNITED STATES
| ENVItONMEWTAl PtOTECTION ACENCV
CIIAT LAKES NATIONAL MOCIAM OFFICE
-------
Page 1-16
4. Future Trends
Future trends for many factors will likely influence the risks addressed in this study. Factors such as changes in
population distributions, land use patterns, industrial activity, and recreational patterns serve as indicators of where
future environmental impacts are likely to be felt
The Basin's population distribution influences many human health risks because of exposure to the problem areas.
For example, air emission of a toxic contaminant in Detroit is likely to be a greater threat to human health in terms
of the total number of potentially impacted persons than a similar emission in Saginaw, Michigan; while the risk
to an individual may be similar in the two areas, the differences in population density signify a greater danger in
terms of numbers exposed. As discussed above, population trends throughout the Basin indicate an overall decrease
in Basin inhabitants. While the total human population which may be exposed to problem areas sources and
activities is declining, areas of localized growth and new activities could witness increased residual risks.
Land use patterns also impact residual risks in the Basin. For example, changes in agricultural land use will affect
those areas where pesticides are intensively applied. From 1949 to 1987 agricultural acreage declined over one-third
in the United States portion of the Basin. This is depicted in Exhibit 10. In spite of less acreage, however, dramatic
increases in crop production have occurred This was
accomplished with the use of great quantities of both
pesticides and fertilizers.
Presently Basin land use can be generally divided among
agricultural, residential, industrial, commercial, recreational
and forest areas. These uses are changing. Industry along
the Lakes' shores is being replaced by recreational and
leisure-based development. According to a mid-1980s study
by Francies Dcmoy of Rochester Institute of Technology's
Center of Management Study, "...the next 20 years will see
the growth of a water-based recreation economy, steeped in
sport-fishing, supported by private investment., and played
against the backdrop of improved water quality."9
This trend of recreational development is also occurring
along Lake Michigan shores. The increase of certain
stressors on the ecosystems in these areas result in increased
habitat destruction. Areas such as Traverse Bay have seen
increased condominium and resort development and are witnessing a high degree of permit applications to Gil in
wetlands.
The extent of industrial activity in the Basin has a direct impact on the intensity of industrial water discharges, air
emission, and land disposal of wastes. Long term decline in worker-intensive industry in the Basin is evident in the
forecasts for manufacturing employment The Erie County Pennsylvania Task Force on Economic Adjustment
reported in 1989 that, "Despite the high concentration of manufacturing employment in Erie, the sector is forecast
to decline from 35,200 in 1989 to 33,800 by 1992."4 Where such changes represent a decrease in industrial activity,
they could result in a lowered importance in industrial point wastewater discharges and toxic air emissions, although
it may indicate a potential increase in other risks, such as abandoned waste sites.
' 1985-86 Annual Report, Monroe County Water Quality Management Agency, page 19.
4 Economic Adjustment Strategy for Erie County. Pennsylvania. August, 1989, The Economic Development Corporation of Erie County,
Pennsylvania Task Force on Economic Adjustment.
Total Farmland in U.S. Great Lakes Basin
1949-1987
Million Acres
1948 1054 1060 1064 1060 1074 1078 1062 1067
Exhibit 10
-------
Page I - 17
Finally, changes in environmental statutes may also affect many of the risks described under the problem areas. For
example, pending amendments to the Clean Air Act will reduce risks from sulfur and nitrogen oxides, hazardous and
toxic air emissions.
-------
Page 1-18
D.
THE HEALTH OF THE GREAT LAKES ECOSYSTEMS
1.
Stressors
As the problem area discussions indicate, anthropogenic releases of contaminants to the Great Lakes Basin occur
through all media, and from a myriad of sources. The following exhibits provide a flavor of these stressors and some
of their sources.
Exhibit 11 depicts the geographic distribution of estimated
total releases for all media in Great Lakes Basin counties in
pounds per year. Exhibit 12 shows releases of known,
probable and suspected human carcinogens to the Basin.
The information for these exhibits was taken from 1988
Toxic Chemical Release Inventory (TRI) which estimates
releases for a limited number of chemicals (slightly higher
than 300) from the heavy manufacturing sector in the Basin.
Therefore, this TRI data represents only a portion of the
many chemicals released to the Basin. Exhibits 13 and 14
show the proportions by media for these emissions, and
Exhibit 15 shows the distribution of facilities monitored in
TRI. Exhibit 16 depicts the distribution of total air releases
to all media for the Basin counties.
TOXIC RELEASES
IN GREAT LAKES COUNTIES
TOXIC CHEMICAL RELEASES INVENTORY, 1988
FUGITIVE AJR
TOTAL
100 200 300 400
MMian Poundi/VMr
Exhibit 13
TOXIC RELEASES IN GREAT LAKES
TOXIC RELEASE INVENTORY, 1988
FUGITIVE AJR
STACK AIR
WATER
INJECTION
POTW
OFF-SITE
Exhibit 14
10 IS 20
MMicns ol Poundk/Vattr
Environmental distribution of TRI emissions/discharges are
placed into the following seven groups: fugitive air
comprises unidentifiable points of origin; stack air is an
intentional release from a discrete source (point source);
discharge to surface water, disposal into underground
injection wells; land disposal is also regarded as on-site;
transfers/discharges to public sewage systems or POTW's
(publicly owned treatment works); and off-site transfers are
to locations other than POTW's.
Chemicals known to be present and toxic in the Great Lakes
have been placed on List 1 of Annex 1. Exhibit 17
illustrates the geographic distribution of releases for these
Annex 1 chemicals for all media. This list, which is
currently under development as required by the Great Lakes Water Quality Agreement between the United States
and Canada, is included as Appendix D. The list, presently in draft form, represents the toxic substances that are
known to exist in water, sediments, fish and other biota of the Great Lakes. Most of these substances are not in TRI,
and so are not depicted here. Exhibit 18 depicts Annex I releases by pathway of release.
-------
Exhibit 11
TOTAL RELEASES
FOR ALL MEDIA
Lake Superior
Lake Huron
Lake Ontario•
Lake *
Michigan
w
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
POUNDS
PER
YEAR
0
1 - 100,000
Great Lakes Basin Counties
100,001 - 1,000,000
1,000,001 - 10,000,000
10,000,001 -100,000,000
>100,000,000
Source: Toxic Chemical Release
Inventory, 1988.
-------
Exhibit 12 RELEASES OF CARCINOGENS
FOR ALL MEDIA
Lake Superior
Lake Huron
Lake Ontarli
Lake \
Michigan
Great Lakes Basin Counties
POUNDS
PER
YEAR
0
1 - 100,000
100,001 - 1,000,000
1,000,001 - 10,000,000
10,000,001 -100,000,000
>100,000,000
Source: Toxic Chemical Release
Inventory, 1988.
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Lake Superior
Lake Huron
Lake Ontai
Lake 1
Michigan
Exhibit 15
TOTAL
TOXIC RELEASE INVENTORY
FACILITIES
Great Lakes Basin Counties
FACILITIES
0
1 - 3
4- 6
7- 27
28- 81
82 - 243
Source: Toxic Chemical Release
Inventory, 1988
ERA
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Exhibit 16
TOTAL AIR RELEASES
FOR ALL MEDIA
Lake Superior
Lake Huron
Lake Ontario.
Lake ^
Michigan
Great Lakes Basin Counties
POUNDS
PER
YEAR
0
1 - 100,000
100,001 - 1,000,000
1,000,001 - 5,000,000
5,000,001 - 10,000,000
> 10,000,000
Source: Toxic Chemical Release
Inventory, 1988.
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Exhi"it 17 RELEASES OF ANNEX 1 CHEMICALS
FOR ALL MEDIA
1
I
Lake Superior
Lake Huron
Lake Ontario
Lake \
Michigan
Great Lakes Basin Counties
POUNDS
PER
YEAR
0
1 - 100,000
100,001 - 1,000,000
1,000,001 - 10,000,000
10,000,001 -100,000,000
>100,000,000
Source: Toxic Chemical Release
Inventory, 1988.
s&EPAl
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Page I - 19
Exhibits 195 through 24 provide information on the type of industrial processes, major companies, and facilities that
are included in TRI and that release Annex 1 and carcinogenic substances.
A significant amount of toxic contaminants in the Great Lakes Basin may be emitted from mobile sources. For
example, Exhibit 25 shows the relative contribution of mobile, point, and other sources to emissions of benzene, lead,
and trichloroethylene in the Detroit/Windsor area.6 Mobile sources were also shown to be a significant source of
cancer risk from air pollution in southeast Chicago.7 Gasoline vapors are primarily released from mobile sources
as well. While mobile sources may have potentially significant impacts on the Lakes, especially through atmospheric
loadings, time constraints have prevented these sources from being fully analyzed here.
Exhibit 26 shows the distribution and magnitude of TRI PCB releases in the Great Lakes Basin. PCBs have been
shown to be a major source of risk to wildlife and to human health through fish consumption. In addition to
industrial releases, they enter the water and the food chain through the resuspension of toxic sediments.
TOXIC RELEASES IN GREAT LAKES
TOXIC RELEASES INVENTORY, 1988
FUGITIVE AIR
STACK AIR
WATER
INJECTION
LAND
POTW
OFF-SITE
ANNEX 1
20 40 60 ao
Million* of Powdc/Yaa
100 120
Exhibit 19
GREAT LAKES PARENT COMPANIES
RELEASES OF ANNEX 1 SUBSTANCES
GENERAL MOTORS
FORD
UWOHN
RR DONNELLEY
CHRYSLER
JESUP
OUTBOARD MARINE
AVERY
MAXWELL
USX
INLAND t
INTERNATIONAL P/
DOW
SOURCE' TRI. 1MB
STEELCASE
¦v CHEMICAL
WHIRLPOOL
CORNING CLASS
BRISTOL-MYERS
4 « a 10 12
M*on(*P«ndtfYwr
Exhibit 21
GREAT LAKES INDUSTRIAL PROCESSES
RELEASES OF CARCINOGENIC SUBSTANCES*
PRIMARY METAL*
CHEMICAL* AMD ALLIED
TRANSPORTATION EQUIP
RUBBER 4 MISC PLAIT
FABRICATED METALS
ELECTRONIC
MO I COMM MACHINERY
MIBC MANUFACTURING
RfcPER 4 ALL CO PRO Da
FOOO
LEATHER
PETROLEUM REFINING
PMOTOORAPMIC
PRINTING 4 PUBLISH
•TONE. CLAY. BLABS
OTHERS
FURNITURE
APPAREL/FABRICS
TEXTS.E MILL PRODS
LUMBER 4 WOOD PRODS
TOBACOO
•SOURCE; TIM. 1BS4
• 10 1« «0
Millions of PouodB/YsBr
Exhibit 20
GREAT LAKES PARENT COMPANIES
RELEASES OF CARCINOGENIC SUBSTANCES
U8 REDUCTION
UPJOHN
GENERAL MOTORS
OUTBOARD MARINE
U8X
KNOLL
DOW CHEMICAL
NATIONAL STEa
BRISTOL-MYERS
REXHAM
LTVB^ao
CORNING GLASS
HASTINGS
oooovIar
EAGLE-RICHER
SOURCE: TRI. 1*M
2 3 4 8
MeomofPoundeWw
Exhibit 22
9 Exhibit 18 his been deleted.
' Preliminary information from Draft U.S. EPA Transboundary Air Emissions Project, 1990.
' U.S. EPA. "Estimation of Cancer Risks Attributed to Air Pollution in Southeast Chicago." Region 5, Air and Radiation Division,
September, 19S9.
-------
P«gc I - 20
GREAT LAKES FACILITIES
RELEASES OF ANNEX 1 SUBSTANCES
0 10 40 *0 10
MMona of Pounda/Ymt
SOURCE: TRJ. 1008
Exhibit 23
2. Water Quality
As evident from the previous exhibits, numerous
contaminants have been released to the Great Lakes through
surface water discbarges, air deposition, and ground water
migration. The contaminants that are examined in this risk
characterization study have striking effects on the health of
the Basin's ecosystems, populations and their activities.
Exhibit 27 shows concentrations of six toxics found in the
water of the Great Lakes. While these concentrations are
within EPA ambient water quality criteria, they are
lakewide averages. Concentrations are likely to be higher
in locations closer to large populations centers, such as the
Areas of Concern designated for remedial action by the
International Joint Commission (IJC).
GREAT LAKES FACILITIES
RELEASES OF CARCINOGENIC SUBSTANCES
SMUNAWQI
DOW
AK&fe&R
SOURCE: TO, 1S06
2 3 4 8 1
MUora or Pounda/VMr
Exhibit 24
TRANSBOUNDARY PROJECT
SELECTED EMISSIONS FROM POINT,
AREA, AND MOBILE SOURCES
MOTOR VEHtCl£ MANU.
STEEL Mil AND COKE
MOBILE SOURCES
LmLTTTES
OTHER INOUSTOAL
OTHER AREA SOURCES
12 3 4
ThouiWKti of Tora/Vaar
Exhibit 25
-------
Exhibit 2
PCB's
POUNDS per YEAR
200.001-600,000
100,001-200,000
10,001 - 100,000
1,001 . 10,000
1 - 1,000
TOTAL PCB RELEASES
U.S. GREAT LAKES
BASIN
1 DYNEX INDUSTRIES INC.
2 WHIRLPOOL CORP.
3 ROUGE STEEL CO.
4 WHIRLPOOL CORP.
5 G. E. MEDICAL SYSTEMS
6 TRANSFORMER INSPECTION
7 SAGINAW GREY IRON PLANT
8 NEWTON FALLS PAPER MILL
9 MANOIR ELECTRO ALLOYS
10 HEMLOCK SEMICONDUCTOR
11 NATIONAL STEEL
12 CHICAGO HEIGHTS STEEL
13 RACO INC.
14 JAMES RIVER CORP.
15 GMC
16 GENERAL ELECTRIC-ERIE
17 EATON CORP.
18 BP OIL CO.
19 TRW INC.
20 TELEDYNE PENN-UNION
21 GERITY SCHULTZ CORP
22 PACKAGING CORP. OF AMER.
Source : Toxic Release Inventory, 1988.
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Page I - 21
Exhibit 27
Concentrations of Toxics in the Water of the Great Lakes*
(lakewide averages, ng/L)
Contaminant
Lake Superior
Lake Huron
Lake Erie
Lake Ontario
alpha-BHC
7.84
5.19
3.82
3.95
Lindane
1.07
0.79
1.02
1.25
Dieldrin
0.28
0.36
037
031
pp-DDE
—
0.01
0.03
0.05
PCBs
032
0.57
1.16
1.20
HCB
0.03
0.03
0.05
0.05
Source: 1989 Report on Great Lakes Water Quality. Great Lakes Water Quality Board, Report to the
International Joint Commission, presented at Hamilton Ontario, October 1989.
Furthermore, the six contaminants listed in Exhibit 27 are persistent, allowing for their ingestion by plankton and
other species. This creates a tendency for these contaminants to bioaccumulate in the food chain, and opportunities
to impact fish and wildlife in several areas of the Basin. While Great Lakes water quality is generally good, when
judged by EPA drinking water standards, die phenomenon of bioaccumulation makes these contaminants a health
concern for human populations. Elevated concentrations of PCBs, for example, are found in several species of fish
that are consumed by Basin populations.
3. Ecological Endpoints
Anthropogenic stresses have had noticeable effects on many ecological endpoints in the Great Lakes Basin.
Researchers have found that since the 1950s, species at the top of the Great Lakes food web have shown reproductive
problems and population declines. Some of the affected species include the bald eagle, black crowned night heron
Caspian tern, common tern, double-crested cormorant, Forster's tern, herring gull, «ninlc, otter, wolf, and snapping
turtle. While population declines can be attributed in part to habitat destruction, reproductive abnormalities and birth
defects are the result of toxic contaminants such as DDT, diddrin, PCBs, mirex and others entering the food chain.
Effects on wildlife are often the most visible and well-reported examples. However, toxic effects are also found in
more common species which are more ecologically important to die structure of Great Lakes ecosystems. For
example, a zooplanlcton can bioaccumulate 500 times the ambient concentration of PCBs during its lifetime.
Furthermore, several fish species which have direct impacts on human health through consumption, have shown
tumors and reproductive problems. For example, lake trout, coho and chinook salmon are popular sport fish which
have been unable to produce viable self-sustaining populations, and so are forced to be stocked by state and federal
agencies.
' Data were not available Cor Lake Michigan.
-------
Page I - 22
The ecological effects in these and other fish sometimes imply health risks to human populations as well. These and
other species of fish are currently listed under health advisories which warn populations to reduce or, in some cases,
eliminate consumption due to die accumulation of toxic contaminants. For example, Michigan's 1989 fish
consumption advisory recommended that consumption of walleye, northern pike, muskie, and several other fish from
Michigan's inland lakes, which are all within the Great Lakes Basin, be restricted because of tests finding some fish
with mercury exceeding the State's level of public health concern.® The advisory recommends that pregnant women,
nursing mothers, or women who intend to have children, and children age 15 and under avoid eating any of these
fish.
This study also points to one of the most striking effects that human activity has had on the Great Lakes Basin, that
of habitat destruction. Degradation of water and terrestrial ecosystems has occurred at an alarming rate. Not only
have habitats such as wetlands, prairies, and dunes been degraded by pollution, but more striking, is the wholesale
elimination of ecosystem due to urban, agricultural and recreational development In the State of Michigan, for
example, wetlands acreage has decreased by 71% since presettlement This is especially sobering when considering
that total land area in Michigan was initially one-third wetlands.10
The impacts of both toxic releases and habitat destruction on fish and wildlife are discussed in more detail in the
problem areas.
4. Impaired Beneficial Uses
In addition to human health impacts and wildlife impairments, toxic contaminants and habitat destruction have
impaired many other uses deemed beneficial to Basin populations. For example, in some harbors, such as Indiana
Harbor and the lower Maumee River, transportation has been impaired because the existence of toxic sediments has
halted dredging activities. Indiana Harbor has not been dredged since 1972, and as a result of the sediments
deposited since then, ships have been forced to reduce their drafts by up to ten feet
These and other impacts are listed in Annex I of the Great Lakes Water Quality Agreement as beneficial uses that
have been impaired in the Basin. The uses on this list are addressed in the study's risk characterizations. The list
contains the following 14 impaired uses:
a.
Restrictions on fish and wildlife consumption;
b.
Tainting of fish and wildlife flavor;
c.
Degradation of fish and wildlife populations;
d.
Fish tumors or other deformities;
e.
Bird or animal deformities or reproductive problems;
f.
Degradation of benthos;
g-
Restrictions on dredging activities;
h.
Eutrophication or undesirable algae;
i.
Restrictions on drinking water consumption, or taste and odor problems;
j-
Beach closings;
k.
Degradation of aesthetics;
1.
Added costs to agriculture or industry;
m.
Degradation of phytoplankton and zooplanlcton populations; and
n.
Loss of fish and wildlife habitat
* Michigan Department of Public Health, 'Public Health Newt,' December 14,1988.
10 Great Laic*' flreat Legacy?, pp. 141-144.
-------
Page I - 23
5. Ecosystem Approach
The problems mentioned above stem from human activities in the Basin that have taken place over a long period
of time. While a pristine natural setting cannot be restored, many of the ecostress ore can be corrected thereby
reducing human health and ecological risks.
The contaminants that have caused damage come through seven) pathways and from many sources. Furthermore,
the Great Lakes Basin is a unique and complex collection of ecosystems encompassing many interactions. These
two facts signify that the problems presented here cannot be treated in isolation from each other. For example, as
the problem area discussions show, numerous pesticides have many sources, both rural and urban. They enter the
Basin by using several pathways: urban and rural runoff, ground water, atmospheric loading, and through
resuspension of contaminated sediments. Many pesticides can directly impact the health of human applicators. And
many have persisted in the environment thereby indirectly affecting the health of all organisms through
biomagnification in the food chain.
Such complexity is best addressed by using a holistic, ecosystem approach that targets specific risks and geographic
areas for action. An ecosystem approach to risk reduction involves a comprehensive analysis of the interrelated
activities that affect human health and ecological risks in the Great Lakes Basin, aspects which include: the physical,
chemical, and biological health of the Great Lakes; geographic aspects of air, water, and land use problems; and
the social, economic, technical, and political variables which must also be considered.
This report points to several problems which create the highest risks for human populations and ecosystems in the
Basin. The issues which demand more attention than others are exotic species, habitat destruction, toxic sediments
spills, toxic loadings and their bioaccumulative impact in the food chain. On many issues, important action has
already been taken. Two of the most important initiatives in the Great Lakes Basin are the development of Lakewide
Management Plans for each Great Lake, and Remedial Action Plans to remediate degraded locations known as Great
Lakes Areas of Concern. Through these efforts, binational, federal, State, Tribal, local governments and the public
have the mM-henfoni to address the worst problems and locations in the Great Lakes Basin.
6. Great Lakes Lakewide Management Plans11
Annex 2 of the Great Lakes Water Quality Agreement, as amended in 1987, and the Great Lakes Critical Programs
Act of 1990, direct the United States and Canada to develop Lakewide Management Plans (LaMPs) for each of the
Great Lakes. LaMPs are meant to reduce loadings of Critical Pollutants in order to restore the beneficial uses of
open lake waters, as defined by the Agreement, and to move toward the goal of "virtual elimination of persistent
toxic substances and toward restoring and maintaining the chemical, physical and biological integrity of the Great
Lakes Basin Ecosystem."12
The first LaMP being developed is the Lake Michigan LaMP, for which the United States has sole responsibility.
EPA is preparing a draft Baseline Report for the Lake Michigan LaMP which is meant to describe current knowledge
of Lake Michigan as a basis for the further development of a comprehensive plan. LaMPs will be submitted as they
satisfy four stages as laid out by the IJC: 1) when a definition of the problem has been completed; 2) when the
schedule of load reductions is determined; 3) when remedial measures are selected; and 4) when monitoring
indicates that the contribution of the Critical Pollutants to impairments of identified beneficial uses has been
eliminated.
11 Much of Ibis lection it wmmarized from EPA'» draft "Bweliae Report: Lake Michigan Lakewide Management Plan,' October, 1990.
u Annex 2, Subsection 2(b), Great Lakes Water Quality Agreement of 1978. as amended by Protocol signed November 18,1987.
-------
Page I - 24
The draft Lake Michigan LaMP baseline report describes current knowledge regarding:
1. Lakewide beneficial use impairments and the Critical Pollutants associated with those impairments, including
identification of the highest priority candidate Critical Pollutants for immediate action and a process for
updating the priority list;
2. The sources of the highest priority candidate Critical Pollutants and, where possible, load estimates from
these sources;
3. An inventory of ongoing pollution control activities being conducted by Federal, State, local and private
organizations;
4. Recommendations for specific "fast track" actions to be undertaken to maximize current authorities in order
to reduce, and where possible, eliminate pollutant loads in order to move towards the environmental
objectives described earlier.
A LaMP is cunently under development for Lake Ontario as well. This LaMP will build upon the work done to
develop a Lake Ontario Toxics Management Plan. The Toxics Management Plan (TMP) was developed in 1989 and
is to be periodically updated to reflect new developments. A1990 update is scheduled for distribution and discussion
by the TMP's coordinating committee and the public in December, 1990. While die TMP will form the bulk of the
LaMP, the LaMP will cover two additional issues not addressed by the TMP: 1) the development of a committee
to ensure public participation, and 2) attention to habitat losses.
EPA does not believe that the LaMP process should overshadow other efforts underway to address issues such as
habitat destruction, exotic species introduction, and other human activities which impair beneficial uses of the Lakes.
The emphasis on Critical Pollutants for open waters is meant to focus the efforts of the LaMPs to achieve clear
results in each Lake. LaMPs will also address contaminant problems that occur in near shore areas, and where
possible, will work in concert with Remedial Action Plans for the Areas of Conoern described below.
LaMPs can also take full advantage of current and future risk characterizations when targeting specific chemicals
and remedial actions. Risk analysis is a necessary tool for LaMP success.
7. Great Lakes Areas of Concern
The Great Lakes Areas of Concern (AOCs) are depicted in Exhibit 28. Brief descriptions of the AOCs in the United
States along with their problem contaminants are provided in Appendix E.u
a. Definition and History
For more than twenty years, the Water Quality Board of the UC has identified specific polluted areas in the Great
Lakes. Over time, as more environmental data have become available, environmental conditions have changed, and
as a result, the number of areas has fluctuated. In 1985, the IJC identified forty-two AOCs and the United States
designated the forty-third AOC in 1990. These are areas where water use is impaired and objectives of the Great
Lakes Water Quality Agreement or local environmental standards are not being achieved. Of the forty-three AOCs,
twenty-six of them are in the United States alone, and five are shared with Canada. The most significant issue for
u For detailed descriptions of the AOCs, «ee Appendix A. 1989 Report on Great Lakea Water Quality, Water Quality Board, IJC.
-------
U.S.
GREAT LAKES
AREAS OF CONCERN
LAKE SUPERIOR
LAKE
HURON
LAKE
ONTARIO
LAKE
mMICHIGAN
LAKE
ST CIA*
LAKE
ERIE
LAKE SUPERIOR
(1) St Louts River
(2) Torch Lake
(3) Deer Lake-Carp Creek
-Carp River
LAKE MICHIGAN
(4) Manistique River
(5) Menominee River
(6) Fox River/Southern
Green Bay
(7) Sheboygan
(8) Milwaukee Estuary
(9) Waukegan Harbor
(10) Grand Calumet River
/Indiana Harbor Canal
(11) Kalamazoo River
(12) Muskegon River
(13) White Lake
LAKE HURON
(14) Saginaw River
/Saginaw Bay
LAKE ERIE
(15) Clinton River
(16) Rouge River
(17) Raisin River
(18) Maumee River
(19) Black River
(20) Cuyahoga River
(21) Ashtabula River
(22) Buffalo River
(43) Presque Isle Bay
LAKE ONTARIO
(23) Eighteen Mile Creek
(24) Rochester Embayment
(25) Oswego River
CONNECTING CHANNELS
(26) St Mary's River
(27) St Clair River
(28) Detroit River
(29) Niagara River
(30) St Lawrence River
L
so
_L
100
_L_
150
_L
200
_J
KILOMETERS
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE I
-------
Page I • 25
the vast majority of the AOCs is the presence of chemical toxic contamination, with hundreds of toxic substances
identified in the Lakes. Many AOCs are so saturated with contaminants that the areas themselves have became a
major source of pollutant loadings to the Lakes.
b. Remedial Action Plans
The IJC identified the AOCs to encourage jurisdictions to rehabilitate these areas. The federal, State, Tribal and
provincial governments around the Great Lakes have agreed to cooperate with local governments to dean up AOCs
through the development of Remedial Action Plans (RAPs). RAPs are being developed by federal, State, Tribal,
local, and provincial governments around the Great Lakes, with the respective States functioning as lead coordinators
for AOCs within their borders. A RAP is being developed for each AOC. They will identify the degree of
contamination at each area, and the level to which the area must be restored in order to make die area safe for
fishing, swimming, drinking, and other beneficial uses. They will also recommend which remedial measures are
appropriate for each problem, and a time table for their execution.
An important objective in developing the RAPs is to use an ecosystem approach to remediation. Another important
aspect of developing the RAPs is public involvement The public, particularly the communities adjacent to the
AOCs, must be involved with RAP planning and implementation. The insights and priorities of local residents is
vital to the success of each RAP, and ongoing public involvement and support will be important to sustain the
lengthy process.
As part of the Great Lakes Water Quality Agreement, each RAP will be submitted to the IJC for review at three
different stages of the RAP's development The first stage is the definition of the contamination problems in the
AOC; the second stage comes after appropriate plans for addressing the contamination problems have been
developed; and the third stage is after monitoring activities have been conducted and the monitoring data indicates
that beneficial uses have been restored.
By the end of the fiscal year 1990, 14 United States RAPs bad been submitted to the IJC for Stage 1, the
identification of problems, sources, and impaired beneficial uses, and 9 of those for Stage 2, the selection and
implementation of remedial options. However, initial submission of RAPs is not a completion of the plan. Rather,
RAPs are an iterative procedure where sections of some Stages will be refined and resubmitted, based on new
information or advice from the IJC. The Great Lakes Critical Programs Act of 1990 requires that all U.S. RAPs be
submitted to the UC by January 1, 1992.
8. Relationship Between AOCs and the Risk Characterization Problem Areas
The AOC approach to remediation has been a vital beginning to remediation of the Great Lakes. But while the worst
cases of pollution must be cleaned up in order to achieve clear progress, a risk-based approach whidi deals with the
whole ecosystem is also necessary in order to ensure overall reductions in toxics. AOCs often represent remediation.
This must also be accompanied by source reductions guided by risk assessments. A risk-based approach focusing
on aggregate risks as a means to prioritize AOCs for action and to identify other areas or sites for special attention
should also be used. The broadened approach would draw in new sites for attention based upon terrestrial and other
human health risks not considered in the general AOC approach to environmental problems.
Many of the environmental problems addressed in this report exist within the AOCs. While discharges or releases
to surface water predominate, waste management and air emission problem areas are present at many. Several of
the environmental problems, such as toxic sediments and point source discharges to surface waters, are found in
virtually all of the AOCs. Nonpoint source loadings are also common problems. Nonetheless, the number of
environmental problem areas associated with an AOC does not necessarily provide an adequate measure of the
relative level of residual risk at that site. The current risk at each AOC can only be measured by assessing the level
of overall environmental contamination, constituents involved, likely exposure routes, and environmental receptors
-------
Page I - 26
at the site.
As apparent, each AOC may not encompass all sources of environmental risk described in this study. Many risks
are not addressed by each RAP, such as those associated with atmospheric loading, climate change, and introduction
of exotic species. But many of the risks which are addressed by RAPs are in fact problems outside the AOCs as
well. Risks from environmental problems such as sediments, active and abandoned hazardous waste sites, point and
nonpoint source discharges to surface water could be just as significant outside AOC boundaries.
The apparent inconsistency between what problems are addressed by this study and what is addressed by RAPs is
due, in part, to the definitions of an Area of Concern, a Remedial Action Plan, and the risk analysis problem areas.
The AOC approach inherently focuses on impairments to specific beneficial uses at a particular location.
The problem areas, which were defined by EPA Headquarters, are not always compatible with the RAP approach
for specific actions directed at risk reduction. For example, the air problem areas focus on risks associated with
inhalation of constituents within the air emissions, an exposure pathway not considered in the RAPs.
Risk-based analyses and actions should not replace the important work of the RAP process. In fact, in-depth risk
characterization studies can complement the work being achieved in AOCs. For example, they can fit into this
process by helping prioritize the tasks to be done, as well as by giving a general indication of which AOCs may need
more immediate attention. This approach can help to point out other locations with risk that are not part of the
formal AOC process. It may also help to guide the overall direction of AOC clean-up from an ecosystem approach.
The interrelationships between the AOCs, their surrounding communities, and sources of degradation should be tied
together. Such connections between AOCs and broader risks can be addressed largely with LAMP development
9. Where to Go
1. Future Study and Refinement of Risk Analysis in the Great Lakes
Reassessment of problem area definitions. To better assess risks, environmental endpoints, appropriate indicators,
and stressors must be characterized. Problem areas should be redefined to reflect this approach, rather than
programmatic divisions or budget categories in EPA. Future emphasis will be on areas of risk which reflect the
uniqueness and interaction of the Great Lakes.
Characterization of risks on a lake-bv-lake basis. Average concentration data may underestimate real risks for
humans and ecosystems in nearsbore areas. Therefore, the use of Great Lakes-wide contaminant data is not
necessarily the best approach for taigeting risk-based activities. A better approach would be to assess risk on a lake-
by-lake basis, and eventually even a segment-by-segment basis. Contamination of water and fish varies significantly
from one place to another, and there are a number of geographic areas, particularly in the Connecting Channels and
embayments, that have toxic chemical gradients. The development of LaMPs and RAPs provides the opportunity
to develop such refined characterizations of risk.
Mass Balance data for each Lake. In general, this report assumed that concentrations in the lakes, sediment and fish
are in equilibrium with loads. For sodium equilibrium, concentrations may well be too low. For otheis, they may
be too high. The way to determine this and the materials balance in the ecosystem for all contaminants of concern
is to develop mathematical mass balance models and measures with a good degree of confidence in the loadings.
EPA's Large Lakes Research Station has begun modeling contaminants using this approach in Green Bay. A
balance input-output approach is a necessary component of an ecosystem approach to define contamination of each
Great Lake, or of the Basin as a whole. LaMPs provide the institutional structure to conduct such Mass Balance
analysis.
Data development, coordination, and improved access. Because EPA Programs generally do not collect information
on the success of the problems they address in the environment, there is a need for more data collection. In addition,
-------
Page I - 27
data which docs exist in other federal or state agencies could contribute much to understanding contamination in the
Great Lakes. The development of new data sets, coordination and improved access to these databases need to
become priorities in future risk characterizations.
2. Future Action to Clean Up and Manage the Great Lakes
Environmental Indicators. Future actions must include a focus on environmental indicators to judge the success of
new and continuing efforts. Measurements of success, such as good monitoring data on loadings, emissions, fish
advisories, designated uses, and measurements of contaminant concentrations in endpoints, can all help determine
if actions are having an effect In addition to successes measured in this way, successes in the prevention of new
problems must also be measured.
Ecological Risks. Ecological risks need to receive more attention than they have in the past. New inter-agency
approaches and activities must address the destruction of wetlands, spawning areas, and other wildlife habitats.
Pollution Prevention and Source Reduction. In addition to conventional forms of dealing with pollution in the Great
Lakes, this study encourages several new approaches to clean up and manage the Great Lakes. The complex
interrelationships between problem areas, for both sources and pathways of pollutants, encourage further refinement
of ecosystem approaches. These strategies stress the importance of reducing contaminants at their sources. La MPs
and the dean-up in Remedial Action Plans must be complemented with pollution prevention activities.
Risk-based Planning and Strategies. Future actions should be based on targeting the most critical problems first from
a risk perspective, while remaining aware of financial and institutional feasibility. Public education must also occur
to help direct the attention of States, EPA Regions, other federal agencies and general populations to those issues
that have high residual risks. In addition, enforcement and prevention priorities should be set for geographically
targeted areas where the suspected estimated human and ecological risks presented here are thought to be high, such
as, Chicago/Gary, Detroit, Cleveland and Buffalo/Niagara Falls (see Exhibit 29).
CHICAGO/GAI
Exhibit 29
HIGH HUMAN HEALTH RISK AREAS
PARTICULATES » 25,000 TONS/VR SPILLS > 20/YR
THI RELEASES > 100M LBS/YR
NPOES MAJORS > 40 POPULATION » 1,000,000
-------
Page II -1
H. EMERGING BASIN-WIDE ISSUES
A. PHYSICAL DEGRADATION OF OPEN WATER AND WETLANDS HABITAT
Problem Area Description
This section presents information on the physical degradation of aquatic habitats and wetlands in the Great Lakes
Basin. Chemical effects from toxic contaminants and degraded water quality are discussed in Primary Water
Problem Areas. Over many decades, development along the shores of the Great Lakes has significantly affected
aquatic habitats, including wetlands. Habitat for many species has become impaired by shoreline development and
water quality degradation in open water areas and tributaries.
In die United States, there are about 4400 miles of Great Lakes shoreline permanent loss. Much of it has undergone
significant urban and agricultural development Depletion of shoreline along the Great Lakes threatens the many
species that spend part or all of their lifecydes along the Great Lakes shores. The following table summarizes 1985
shoreline use by lake1:
SHORELINE USE
SUPERIOR
%
MICHIGAN
%
HURON
%
ERIE
%
ONTARIO
%
RESIDENTIAL
N/A*
39
42
45
40
RECREATIONAL
N/A
24
4
13
12
AGRICULTURAL
N/A
20
15
14
33
COMMERCIAL
N/A
12
32
12
8
OTHER
N/A
5
7
16
7
not available.
Wetlands, once abundant in the Great Lakes Basin, have been threatened over the years by agricultural, uiban, and
industrial development Wetlands play a significant role in die life cycle of many fish and wildlife species in the
Great Lakes. They provide spawning habitat for several fish species, as well as nesting places for large numbers
of migratory birds using the Atlantic and Mississippi flyways. Development activities have created shoreline where
once wetland was far more common.
Basin wetlands have been devastated by conversion to increase agricultural production in the area and to facilitate
uiban and industrial growth. Human intervention has resulted in the 1ms of approximately two-thirds of the original
wetlands in the Basin3 and this loss continues at a rate of about 20,000 acres per year3. Currently there are about
500,000 acres of wetland left scattered throughout the Basin4. However, much of the remaining wetland areas
' The Great Utoi An EnrirtmmtnUl Atlat rnd ft—caret Book, p. Ml
' Great Ulw! Great Iwcrf. page 190.
'Intern*tiooal Joint CwnrnMin 1989. "Living with tbc Likw Cfcallaf— and Opportanitm" pM,
*IaUr&alional Joint Commlsslea. June 1989. with lb* Lakau CkaDtngat aad Oppartunibtt . Annex B" p.B-90.
-------
Page n - 2
functions and values arc further reduced due to changes in shoreline, proximity of human intervention, and reduced
size or access of the lakes. Nevertheless, the remaining wetlands are of extreme value to the system.
Ecological Impacts
The ecological impacts of the physical degradation of aquatic habitats and wetlands are severe. Wetlands are often
used as places to deposit sediment dredged from navigation channels in lakes or rivers, which increases in times of
low water levels. Dredging can disrupt wetland ecosystems and may contaminate them with pollutants in the
sediment Deposition of contaminants may also impact wetland fish and wildlife habitat and thus increase the
likelihood of contaminant uptake in the food chain (See ecological effects addressed in Primary Water Problem
Areas).
The effect of dredging sediments on aquatic habitats is a concern in the Great Lakes Basin due to the large amount
of dredging the U.S. Army Corps of Engineers initiates annually to support the Great Lakes shipping industry.
Throughout the Great Lakes, there are 26 confined disposal facilities that are used to store sediments dredged from
Great Lakes ports. Their locations are depicted in Exhibit A-l3. Although no data exists regarding the effect of
dredging sediments on aquatic habitat, the amount of dredging can be put into perspective by looking at the capacity
of the combined disposal facilities. According to a USEPA report, 48 million cubic yards of sediments have been
dredged and stored in confined disposal facilities in the Great Lakes since 1960. According to the same report, the
remaining capacity of the facilities in the Great Lakes is about 30 million cubic yards6.
Section 404 of the Clean Water Act regulates dredge and Gil activities. Exhibit A-2 depicts the number of dredge
and fill permit applications for 1989 by the Basin county. This information suggests that areas such as Traverse Bay
and the St Clair River, which contain ecologically significant wetlands, are witnessing serious dredge and fill
activities.
Destruction of wetlands in the Great Lakes Basin has had severe ecological impacts. It is estimated that one-third
of the land area of Michigan was once wetland. Wetlands also covered about 60% of southwestern Ontario. At the
west end of Lake Erie, the Black Swamp covered about 1,500 square miles from Sandusky Bay to the Indiana
border. Once wetlands are developed or drained, they cannot be restored to their natural state. Further, there are
several different types of wetlands that serve different functions. Swamps, marshes, bogs, and fens all provide
different habitats for different plant and animal species that depend on these wetlands for their survival.
The Natural Heritage Program data base has identified the most biologically significant wetlands in the Great Lakes
Basin, and while there are several occurrences of high ranking wetlands, the occurrences are only a handful compared
to the number of wetlands in the region. The data base identifies the last remaining wetlands that are considered
biologically significant and a priority for protection. Biological significance, according to The Nature Conservancy,
is the contribution a locality or site makes to sustain biological diversity. Biological diversity is the variety and
process of life, comprised of species and communities of species, their interactions, and the functioning ecosystems
they make up. Sites are biologically significant if they: first, represent ecosystem remnants that are little affected
by European-American development compared to the regional landscape, or second, harbor concentrations of
imperiled species under threat throughout the region. The following table indicates the types of biologically
significant occurrences of wetlands remaining in the Great Lakes Basin. The sites described were selected on the
basis of the presence of exemplary plant communities and/or rare and endangered species. They are located
throughout the Basin.
' US. Environmental Protection Afeacy, 1919. Report op Gwt LAb CmBbmI Dtspoul FnflMta. b.11
tJA EavirotunMUl Protection A$mey. 19P. Report on Greet likw Confined Dtoo—1 FecflWee. TiM» 3.
-------
EXHIBIT A-l
LOCATIONS OF CONFINED DISPOSAL FACILITIES
U.S. GREAT LAKES
Erie Pier
Crooked R.
Kawkawlin,
Saginaw
Bayport
Kidney I.
ebewaing
anilac
Dickenson
Harsen's
Frankfort
Kewaune
Manitowo
Harbor
Verplank
Times Beach
Small Boat
Dike 4
Riverview
Windmill
Clinton
Grassy I.
Bolles
Milwauke
Kenosh
Whirlpool
. , .^Cleveland 12
Lorain\ Cleveland 14
Chicag
Michigan City
Mouillee
Huron
Monroe
(Sterling)
(Edison)
Toledo 4
Island 18
-------
Exhibit A-2
APPLICATIONS
NO DATA
0
1 - 5
6-10
11-20
21 -40
>40
REGION 5
DREDGE & FILL
PERMIT APPLICATIONS
IN GREAT LAKES BASIN
Source: Region 5,2/89 - 2/90.
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Rinlrtgirally Significant Wetlands Remaining in the Great Lakes Basin"7
Page H - 3
State
Wisconsin
Ohio
New York
Minnesota
Michigan
open bog, calcareous fen, coastal plain manh, intcrdunal wetland, southern hardwood swamp, slough,
bog relict
leatberleaf bog, tell Arab bog, mixed swamp, submerged marsh, cm que foil sedge fen, sphagnum
bog, tamarack fen, tall ihnib bog, tamarack-hardwood bog, mixed swamp, mixed emergent manh,
mixed shrub swamp, miple-uh swamp.
rich hem lock-hard wood peat swamp, rich sloping fen, red maple-tamarack peal swamp, rich
graminoid fen, north em white cedar iwamp, marl fen, perched white oak swamp, rich shrub fen,
black spruce-tamarack bog, patterned peatland, iwtif shrub bog, island poor bog, rich shrub fas, marl
pond shore, island salt manh, deep emergent manh, spnice-dir mrsmp, rich sloping fen, sinkhole
wetland, northern white cedar swamp.
shrub swamp
rich conifer swamp, Great Likes marsh, bog, aoutheni swamp, bardwood-coaifer swamp, poor conifer
swamp, muskeg, relict conifer swamp, prairie fen, patterned fea, coastal plain mars, northern fen,
emergent marsh, aoutheni wet swamp, interdunal wetland, rich conifer swamp, southern swamp, relict
conifer swamp, inland salt manh, emergent mars, intermittent wetland.
beach marl, bog, forested fen, tea, flat muck, man, panne, meadow sedge, swamp shrub, bog, acid
bog circumaeutral, flat sand
Welfare Impacts
The environmental costs of physical degradation of aquatic habitats and wetlands can be illustrated by the number
of wetlands acres lost. The following table summarizes the amount of wetlands lost in the Great Lakes region since
presettlement times.
Location/Area
Lake St Clair wetlands
Minnesota
Wisconsin
Illinois
Indiana
Michigan
Ohio
Wetlands Losses in the Great Lakes Region '
Wetland Assessment
85% loss since presettlement
76% loss since 1953
50% loss since presettlement
90% loss since presettlement
86% loss in areas studied
71% loss in northern Indiana
71% loss since presettlement
loss of almost the entire 1,500 square miles of the Black Swamp
*n« Nalart Conservancy. 1990, Report to EPA on Significant Bloloaical Areas of the Grant Lakes WntanhedL
The Conservation Foundation awl the Institute for Research at Public PaUcy. 199ft Grant Lakes Great Leaner?. p.144. Its
source cM did not pin art an onrd assessment for Pennsylvania or New York.
-------
Page II - 4
Material and Property Damage
It is difficult to assign a value to the material and property damage resulting from physical degradation of aquatic
habitat and wetlands. Studies have been conducted to quantify the economic value of wetlands. A study conducted
by the Region 2 office of the Environmental Protection Agency concludes that the economic benefit generated per
acre of coastal wetland ranges from $9,000 to $17,000 using a discount rate of 3%, or from $2,400 to $6,400 using
a discount rate of 8%. The report also looked at the costs associated with replacement of wetlands and the functions
that they provide. The analysis did not look at Great Lakes wetlands9.
It is important to understand the various functions that wetlands play, not only in providing spawning areas for fish
and habitat for a variety of other species, but as an influence on water quality and ground water conditions. For
example, wetland losses may have a devastating effect on the multi-billion dollar Great Lakes sport fisheries.
*U.S. EavirauMaUl Protection Agwcjr. 1990.
Valuta. nJ.
Redan 1 Ctrortht RMc Protect Probi— Area Analwii. Erttmti— W»i1.»h
-------
Page H- 5
B. PHYSICAL DEGRADATION OF TERRESTRIAL ECOSYSTEMS
Problem Area Description
This section describes the stressors on terrestrial ecosystems in the Great Lakes Basin and presents information on
the biologically significant10 occurrences that remain in the Basin. The primary data sources used in this section
include reports from the Natural Heritage Inventory Program and the Region 5 report on physical degradation of
terrestrial ecosystems."
The fact that so few Basin terrestrial ecosystems can be found in a diverse, self-sustaining, and natural state indicates
that degradation of these ecosystems has been significant since European settlement The reversibility of terrestrial
ecosystem damage due to destruction of habitat is very limited. Some work is being done in the area of natural area
restoration, but the work generally focuses on restoring degraded natural areas still in existence, not on the
establishment of new habitat
Within the Great Lakes Basin the main types of native terrestrial ecosystems include old and second growth forests,
wetlands, prairies and other grasslands, and dune communities. Two types of Basin ecosystems, prairies and dunes,
are especially rare. Prairies are complex natural grassland communities whose habitat structure is vulnerable to
changes in wildlife and plant populations as well as changes in the proportions in which they exist
Dunes result from wind and water deposits of lake bottom sand over a sustained period. The dune ecosystem
consists of a natural succession of plant life and accompanying wildlife. Tracts of dune land occur, in varying
widths, intermittently along the Lake Michigan shoreline from northern Wisconsin, around the southern tip of the
Lake, and up the Michigan shoreline to the northernmost area of the lower peninsula. Michigan dune lands are by
far the largest tracts in the Basin.
Stressors
The conversion of natural terrestrial habitat to agricultural
and urban environments has had a considerable impact on
numerous plant and animal species in the Basin. The
remaining habitat continues to be threatened by a variety of
stressors, including agricultural practices, urbanization,
silviculture, mining, road construction, and a variety of
recreational activities. Today, land in the Great Lakes
Basin is used for several purposes, all of which have
impacts on terrestrial ecosystems. Exhibit B-l depicts the
percentages of agricultural, residential, forest and other land
uses for each Lake's basin.12
The impacts associated with agriculture in some
circumstances can be reversible, but the cessation of
farming activities is not likely. Over several hundred years,
" Biological significance is defined in the Physical Degradation at Water Habitat* and Wetland*.
11 One of the moat apparent human health impact* from the detraction of natural terrestrial ecosystems 1* the increase in noise
associated with the anthropogenic changes and their resultant activities. A* noise pollution is not peculiar to the Great Lakes Basin,
it is not discussed here. The reader is referred to the Regional comparative risk projects for more information.
u Environment Canada, Brock University, Northwestern University, United States Environmental Protection Agency. 1988. The
r.Mi I jIm. An Environmental AUas and Resource Book, p. IS.
LAND USE IN THE GREAT LAKES BASIN
Exhibit B-l
-------
Page II - 6
clearing woodlands, plowing under prairies, and draining of wetlands to accommodate fanning practices has served
to degrade all these terrestrial ecosystems. In addition, through farming activities, food and cover for wildlife is
reduced, soil fertility is decreasing, and soil is more susceptible to erosion by wind or runoff. Soil is currently
eroding at a rate that is exceeding the top soil creation process. This results from a variety of farming practices.
For example:
¦ Planting patterns that do not match the natural contour of the land;
¦ Reducing hedge rows to accommodate laiger equipment or to increase acreage;
¦ Plowing harvested fields;
¦ Plowing stream banks or roadsides for weed control;
¦ Planting fewer crop strains, leading to monocultural agriculture which increases the chance of diseases and
requires more intensive use of pesticides and fertilizers.
The major influences on the region's forests include forestry practices (silviculture), farming, mining, recreation, and
urbanization. While selective cutting and replanting activities do occur, many forested ecosystems cannot withstand
the effects of silviculture due to the lack of contiguous habitat needed to sustain them. Only a small fraction of
original old growth forests remain in the Basin. The majority of the forests are second-growth and are less
biologically diverse than the original old growth forests.
Stresses that result from silviculture have very long term effects. Degrading practices include:
¦ Short rotation management and monotypic reforestation for future timber supplies;
¦ Gear cutting forests to harvest timber;
¦ Logging support activities, including construction of maintenance roads, camps and storage areas.
Mining activities can have a devastating effect on both water and terrestrial ecosystems, and the localized impacts
can be seen throughout the Basin. The space needed to support mining activities is significant, especially in areas
which support strip mining activities such as northeastern Minnesota, and the noise and vibration from blasting and
other mining operations can be significant
Mining occurs in various forms throughout the Basin. Iron we mining has been extensive and still exists adjacent
to Lake Superior in Minnesota. Sand and gravel pits exist throughout the Basin. Other resources mined in the Basin
include peat, lime, oil, natural gas and coal. Impacts from mining activities are often irreversible.
Rural development begins with road construction which provides access to previously undisturbed or less developed
tracts of land, opening up the area for agriculture, bousing, business, or recreation. The siting of new roads
determines the relative severity and reversibility of the habitat impacts caused by both the road construction itself
and the subsequent development of the surrounding area. The new land uses that are spawned by road development
can snowball as further development occurs to support them, resulting in wbanization. In addition, increases in
roadside litter create hazards to many animal species.
Ecological Impacts
Impacts of the stressors described above on terrestrial ecosystems are numerous. For example:
¦ Loss of soil fertility;
¦ soil erosion or compaction;
¦ ground or surface water contamination from tailings;
¦ Loss of biodiversity;
-------
Page II - 7
¦ Reduction in species populations;
¦ Increase in noise pollution;
¦ Loss of vegetation with the capacity to convert urban pollutants such as carbon dioxide;
¦ Long-term or permanent loss of habitat
The most obvious ecological impact of terrestrial ecosystem degradation is the loss of habitat which directly affects
species populations. Many species in the Great Lakes Basin are witnessing population declines. Some of these
species include the bald eagle, black crowned night heron, Caspian tem, common tern, double-crested cormorant,
Foistcr's tern, herring gull, mink, otter, wolf, and snapping turtle. The number of plant species that have become
threatened is another sign of the physical degradation of terrestrial ecosystems.13
Prairies provide a good example of how ecosystems are impacted by stressors such as agriculture and development
In addition to the primary destruction that results from these stressors, the unintentional introduction of exotic species
of plants and animals upset the delicate habitat structure. Opportunistic weeds are hardier than many prairie plant
species and compete effectively for nutrients and soil space. Likewise, exotic animal species upset the food web
by altering the predator/prey balance.
The shrinking geographic scope of prairies exacerbates declining habitat quality as the community becomes too small
to buffer non-prairie species and as reproduction opportunities decrease due to reduced population size. The
shrinking range of continuous prairies becomes insufficient to enable the long-term survival of species such as prairie
chicken, marbled godwit, badger and bison.
Dune lands are likewise fragile and therefore face possible extinction in their natural form due to the human
encroachments described above. In particular, construction and recreation are significant stressors to the dune
ecosystem. These lands are often stabilized for development purposes, thereby altering the natural evolution of the
terrain. Unnatural stabilization is accomplished through the introduction of exotic plant species, and by primary
physical alterations including driving beams into the ground and constructing wind breakers to enable development
of homes and industry. The impact is potentially reversible, but unlikely because the nearby development usually
occurs specifically for the aesthetic amenities associated with the lakeshore or dune property.
A major impact from recreational stressors such as trampling is the displacement of sand and the exposure of root
systems. Recently, severe impacts have resulted from off-road vehicle recreation leaving nits and scars on the dune
tracts. The unnatural destabilization of the plant cover delays the natural succession through the loss of vegetation,
as well as increasing the area's vulnerability to noxious weeds such as purple loosestrife.
Remaining old growth forests in the Basin have greater biological diversity than newer second growth forests. The
fragile wildlife-rich ecosystem is more severely impacted by habitat fragmentation than the new forest ecosystem.
Clearing results in localized and peripheral reductions of plant and animal species. The loss of diversity in forest
ecosystems causes alterations in native animal populations. As pioneer tree and undergrowth populations decline
through the reduction of old growth forest, the local gene pool declines as well, which potentially leaves an entire
species in a localized area more vulnerable to disease. Snags and rotting logs, crucial life bearers in the old growth
ecosystem, are notably absent in young forests.
u Th* Coatervaacj Foundation, 1W butitolc far Research «a PabUc Policy. 1990. |^i Great Lwacr? p. 131.
-------
Page II - 8
Biologically Significant Sites in the Great Lakes Basin
The Natural Heritage Inventory Program was established to identify the biologically significant14 areas across the
country in an effort to set priorities for the protection of biotic diversity by private or government oiganizations.
In a report compiled by The Nature Conservancy of Natural Heritage Inventory data, biologically significant sites
within the Great Lakes watershed are identified. The sites described in the report were selected on the basis of the
presence of exemplary plant communities and/or rare and endangered species. The following table summarizes the
significant sites found in the Great Lakes Basin.13
Suie
Wisconsin
Biologically Significant Sites in the Great Lakes States"
No. of Site* Type
130
Ohio
66
New York
62
Graasland:
Pond:
Foittt:
Prairie:
Sand:
Other
Forest:
Sand:
Prairie:
Grassland:
Fond:
Fond:
Grassland:
Sand:
Other
aonhern aedge meadow, auullieiu sedge meadow, bracken grassland, acid
bedrock glade
spring pond, ephemeral pond
north era mesic forest, pine barrens, boreal, southern dry-mesic forest, northern
-------
Page II • 9
Minoesou
Michigan
Indiana
Illinoii
Peanaytv.:
27
140
56
Forest: hardwoods, Great Lake* pine forest, norther hardwood conifer forest, upland
while cedar Cored
Sand: open dune* Great Lake* barren*, bedrock beach, wooded dune and iwale
complex, cobble beach
Forest: Oak barrens, mesic northern forest, dry-mesic northern forest, dry northern
{bred, mesic southern, dry-mesic southern, boreal, southern plain
Prairie: wet, lake plain wet-meaic, dry sand, hillside
Other Moist non-acid cliU bedrock glade, southern wet meadow, dry non-acid clift,
northern bald
Forest: Flatwoodi boreal, upland dry-mesic, flood plain wet, flood plain mesic, upland
-------
Page H - 10
Hie location! of the following biologically significant sites as registered in the Heritage Data Baae are shown in Exhibits B-2 through B-9.
niiNOls
1.
Bumham Prairie
2.
Calumet City Prairie
3.
Dolton Avenue Prairie
4.
Illinois Beach Nature Preserve
5.
Illinois Dunes North
6.
Powder horn Lake and Prairie
INDIANA
1.
Barker Woods Nature Preserve
39.
Moraine Nature Preserve
2.
Basin Lake
40.
MuddLake
3.
Beaver Dam Lake
41.
North Chain Lake Woods
4.
Beech wood Nature Preserve
42.
Olin Lake Nature Preserve
5.
Bell Lake
43.
Paul Thomas Memorial Bog
6.
Bender (Lloyd W.) Nature Preserve
44.
Pipewort Pond Nature Preserve
7.
Big Swamp
45.
Plato Oak Opening
8.
Binkley Bog
46.
Pottawatomie Lake
9.
Boot Lake
47.
Rodenbeck (Albeit D.) Nature Preserve
10.
Brunswick Center Savanna
48.
Ropchan Wildlife Refuge Nature Preserve
11.
Clark and Pine Dune and Swale
49.
Springfield Fen Nsture Preserve
12.
Clarke Junction Prairie
50.
Spurgeon (Edna W.) Nature Preserve
13.
Cline Lake Fen
51.
Sum an Fen
14.
Croteau Till Hill Prairie
52.
Tamarack Bog Nature Preserve
15.
Douglas Wood!
53
Tolleston Ridges Nature Preserve
16.
Dutch Street Bog
54.
Umbrella Sedge Bog
17.
East Lake
55.
Walters Lake Wetlands
18.
Elkhart River Floodplain
56.
Wood Lake Dune Savanna
19.
Engle Lake Marsh
20.
Fawn River Fen
21.
Fish Creek Fen Nature Preserve
MICHIGAN
22.
Gary Dune Forest (Lakswood)
23.
Gibaon Woods Nature Preserve
1.
Agate Harbor
24.
Gritier's Woods Nature Preserve
2.
Allegan Oak Barrena
25.
Hildebraad Lake
3.
ArquiUa Creek
26.
Hoosier Prairie Nature Preserve
4.
BaieDe Wasai
27.
Howard Lake
5.
Bakertown Fen
28.
Indiana Dunes National Lakeshore
6.
Barfield Lakes
IDNL • Cowles Bog
7.
Bates Tower
IDNL - Dune Acres Unit
8.
Bay Park
IDNL - Heron Rookery Unit
9.
Bear Creek (Kalanazoo)
IDNL • Miller Unit
10.
Bear Swamp
IDNL - Pinhook Bog Unit
11.
Beavertown Lakes
IDNL - Tamarack Unit
12.
Big Sable Point
IDNL - Tolleston Unit
13.
Boardwalk
IDNL ¦ West Beach Unit
14.
Brandy Creek Wetlands
29.
Indiana Dunes State Park
15.
Bridge Valley
30.
Ivmboe Dune and Swale
16.
Cadillac Bog
31.
Lime Lake
17.
Captain Jenks Homestead
32.
Long Swamp Woods
18.
Carr Lake
33.
Loon Lake
19.
Cathead Bay
34.
Lost Forty
20.
Cedar bland (Schoolcraft)
35.
Malaxis Bog
21.
Charles
36.
Marsh Lake Wetland Conservation Area
22.
Cheboygan Point
37.
Merry Lea Nature Preserve / Environmental Center
38.
Mongoquiaong Nature Preserve
-------
Page n • 11
Michigan foont)
80.
Mud Lake Swamp (Oakland)
81.
Mulligan Creek
23.
ChilfOD Fen
82.
Munuscong/Little Munuacong Riven
24.
Clinton River Headwaters
83.
Muskegon State Park
25.
Creighton Mush
84.
Noble Lake
26.
Crooked Lake
8S.
Nordhouse Dunes
27.
Crow River Mouth
86.
Northern Hardwoods
28.
Dan'a Point
87.
NorthAore RNA
29.
Dayton Wet Prairie
88.
Norway Oak Barrens
30.
Detroit River
89.
Ogontz Lakeplain
31.
Devil'a Waihtub
90.
Pennefield Bog
32.
Dukes RNA
91.
Pidcford Point
33.
Elizabeth Lake
92.
Pine Hill
34.
Empire Blub
93.
Pine Island Lake (Kalamazoo)
35.
Eacanibs River Gorge
94.
Pine Lake (Van Buren)
36.
Essexville Prairie
95.
Pine River - Frith Road
37.
Eativaat Pines
96.
Pine River - Gratiot Road
38.
Fairchild Lake
97.
Platte Bay
39.
Finger Prairie
98.
Platte River Point
40.
Gogomain River
99.
Pointe Aux Chenea
41.
Good Harbor Bay Dunes
100.
Porcupine Mountains
42.
Grand Beach
101.
Potuwattomie Bayou
43.
Grand Sable Dune*
102.
Presser Bog
44.
Grass Bay
103.
RadrickFen
45.
Hagennan Swamp (Oakland)
104.
Rsttalee Lake Fen
46.
Harlow Lake (Marquette)
IDS.
Roach Point
47.
Harris Cedar Swamp
106.
Roaoommon Red Pines
48.
Hartwick Pines
107.
Ross Property
49.
Harwood Lake
108.
Rush Pond
50.
Haven Hill
109.
Runs Forest
51.
High Rollway Prairie
110.
Sand Island
52.
Highland Cemetery
111.
Sand Portage Falls
53.
Hog bland
112.
Saugatuck Dunes
54.
Horseshoe Bay
113.
Sebewaing Bay South
55.
Horseshoe Harbor
114.
Sebewaing Railroad
56.
Huron Mountain
115.
Section 27 Bowl Prairie
57.
Huron Mountains
116.
Shakey Lakes
58.
Indian Bowl
117.
Sharkey Lake
59.
Indian Mound Savanna
118.
Sheldon Forest
60.
Inland Harbor
119.
Shingle Bay
61.
Isle Royale National Park
120.
Silver Creek Hemlocks
a.
Ives Road Fen
121.
Silver Lake Dunea
63.
Jack-Pine Swamp
122.
Skeel Creek Prairie
64.
Jeaney Woods
123.
Sleeping Bear Dunes
65.
Kates Lake
124.
Smith's Woods
66.
Lake Gogebic State Park
125.
Snake Island (Mackinac)
67.
Lakeville Swamp
126.
South Fox Island
68.
Laughing WhiteGsh Falls
127.
South Manitou Ialand
69.
Liberty Fen
128.
St. Clair Woods
70.
Loon Lake (Gogebic)
129.
St. John's Mvsh
71.
Lost Lake Bog
130.
Sturgeon Bay
72.
Maple River Salt Marsh
131.
Summerby F«
73.
Marsh Creek
132.
Sylvania
74.
Mnxton Plains
133.
Tbompeon Road Prairie
75.
McCormick Tract
134.
Thompsons Harbor
76.
McMahon Lake
135.
Union Lake
77.
Moaette Street
136.
Walt Creek
78.
Moss Lake (Oakland)
137.
Warren Woods
79.
Mt. Lookout
138.
Wangosbance Point
-------
Page II -12
MICHIGAN (coat.)
139. Well* Stale Park
140. Wildfowl Bay Prairie
MINNESOTA
1..
Bogus Lake Hardwoods
2.
Cascade River SP
3.
Cathedral Grove
4.
Cherry Portage
5.
Clearwater Lake
6.
Cloquet Turtle Preserve
7.
Cross River Hardwoods
8.
Daniels Lake
9.
Fourmile Creek White Cedar
10.
Gaakin Lake
11.
Gooseberry Falls SP
12.
Hovland Woods
13.
Jay Cooke SP
14.
Lake Agnes Hardwoods NHR
15.
Marble Lake Lookout Tower
16.
Mineral Center Maple Ridge
17.
Moose Mountain SNA
18.
Mt. Leveaux
19.
Palisade Head
20.
Pigeon Point
21.
Royal River
22.
Schroeder RNA
23.
Split Rock Lighthouae SP
24.
Spring Beauty Hardwoods SNA
25.
Susie Islands
26.
Tettegouche SP
27.
Yellow Birch Natural Area
NEW YORK
1.
Bear Swamp Sempronhu
2.
Beaver Brook Fen Cortlandville
3.
Beaver Brook Springs
4.
Betgen Swamp
5.
Blueberry Patch Swamp
6.
Bonaparte Swamp
7.
Boreal Heritage Macrosite
8.
Brandy Brook Swamp
9.
Brennan Beach Fen
10.
Brewers Comers Bog
11.
Burnt Rock Barrens
12.
Caledonia Swamp
13.
Camctoss Salt Pond
14.
Chaumont Barrens
15.
Chippewa Bay
16.
Chittenango Falls
17.
Clarence Escarpment
18.
Colton Ponds
19. Dsnsville Fen
20. Dead Creek Flow
21. East Malloryville Tamarack Swamp
22. El Dorado Beach
23. Five Ponds Old-Growth
24. Fox Ridge Salt Marsh
25. Gadway Road Flatrock
26. Goat Island
27. Great Gully
28. Groton City Fen
29. Johnny Cake Road Sinkhole Wetlands
30. Junius Ponds
31. Letch worth SP Big Bead
Letch worth SP High banks Entrance Site
Letch worth SP Highbanks Recreation Area Site
Letch worth SP Kisil Point
Letch worth SP Kisil Point East Bank
Letch worth SP Lower Fills
Letch worth SP Mt Morris Road Site
32. Limerick Cedars
33. Limerick Game Farm Road Site
34. Little Gallop Island
35. Lucky Star Alvar
36. Malloryville Fen
37. MallotyviUe Swamp
38. Massawepie Mire Macrocite
39. McLean Fen
40. McLean Marl Meadows
41. Mendon Ponds
42. Middaugh Road Woods
43. Mud Creek Swamp
44. Mud Pond Fen
45. Nelson Swamp
46. Oatka Creek Gulf
47. Rainbow Shores Wetlands
48. Rome Sand Plains
49. Route 222 Fen
50. Rush Oak Opening
51. Sam Adams Road Woods
52. Smith Woods
53. South Hill Swamp
54. Star School Road Barrens
55. Three Mile Creek Road Barrens
56. Tonawanda Creek Millenport
57. Tonaqwanda Creek Rapids
58. Twomile Creek
59. Upper and Lower Lakes
60. White Lake Swamp
61. Wycfcoff Swamp
62. Zurich Bog
OHIO
1. Barnacle Bog Wetlwds
2. Bird Bog
3. Black Swamp Woods
4. Bradley Woods Reservation
-------
Page n - 13
OHIO (coat.)
62.
Triangle Lake Bog
63.
Walnut Beach
5.
Camp Calvary
64.
West Mkrblehead Quarry
6.
Camp Timaberlane
65.
White Pine Bog Forest
7.
Cedar Point National Wildlife Refuge
66.
Yondota Road Orchid Site
8.
Cemetery Lane
9.
Central Marblehead Quarry
10.
Crane Creek State Park
PENNSYLVANIA
11.
Crystal Lake
12.
Cuyahoga River Aconitum Site
1.
Eight Mile Creek Gorge
13.
Cuyahoga Wetlands-Fried'! Bog
2.
Lake Erie Bio reserve
14.
Dupont Marsh State Nature Preserve
3.
Little Elk Creek Slumps
15.
East Marblehead Quarry
4.
Northeast Slump
16.
Erie Sand Barrens State Nature Preserve
5.
Pennline Swamp
17.
Evans R. Beck Memorial Nature Preserve
6.
Presque Isle Macrosite
18.
Fish Creek
7.
Wintergreeo Gorge
19.
Fish Ridge
20.
Flit iron Lake Bog
WISCONSIN
21.
Fowler Woods Nature Preserve
22.
Frame Lake Fen-Hemck Nature Preserve
1.
Alvin Greek Headwaters
23.
Freedom Twp. Sand Pit
2.
Argonne Esker
24.
Girdhsm Road Openings
3.
Bark Bay Slough
25.
Gol! Woods Nature Preserve
4.
Bass Lake Fen
26.
Gott State Nature Preserve
5.
Bastile Lake
27.
Grand River Terraces
6.
Bear Island Hemlocks
28.
Headlands Dunes State Nature Preserve
7.
Belden Swamp
29.
Irwin Prairie Nature Preserve
8.
Berlin Fen
30.
Kenestrick Woods
9.
Bibon Marsh
31.
Kent Bog Addition (DNAP)
10.
Bibon Swamp
32.
Killdeer Plains Wildlife Area
11.
Big Bay Sand Spit and Bog
33.
Kitty Todd Preserve
12.
Black Lake Bog
34.
Lakeside Daisy Prairie
13.
Bloch Tr»rt
35.
Little Mountain & Crile-s Knob
14.
Bose Lake Hemlock Hardwoods
36.
Louis W. Campbell Nature Preserve
15.
Brooks Bluff Prairie
37.
Mantua Swamp
16.
Brule River Cliffs
38.
Marie Delanne Creek
17.
Budsin Mudflats
39.
Maumee State Forest
18.
Burney Lake
40.
Monclova Woods and Dunes
19.
Camp Two Creek Swamp
41.
Morgsn Swamp
20.
Cheemon Maskik
42.
Morgan Woods
21.
Cherry-Maple Ridge
43.
Mud Lake Nature Preserve
22.
Chiwaukec Prairie
44.
Oak Openings Metropolitan Park
23.
Coleman Lake Meadows
45.
Old Stale Line Road Dune
24.
Comstock Bog-Meadow
46.
Ottokee Cemetery
25.
Crooked Lake Shrub-Carr
47.
Padanaram Swamp
26.
Crowell Lake
48.
Pennline Bog
27.
Dalles Creek Maskik
49.
Perry Nuclear Plant Woodi
28.
Deadman Kapo
50.
Pickerel Creek State Wildlife Area
29.
Devil's Island
51.
Reed Road Dunes
30.
Dunbar Barrens
52.
Resthaven Wildlife Area
31.
Dunn Lake Pines
53.
Rocky River Reservation
32.
Fox River Crane Marsh
54.
Russell Park
33.
French Creek Fea
55.
Sandy Point
34.
Giant While Pine Grove
56.
Schwamberger Prairie
35.
Gtocke Lake
57.
Sheldon's Marsh
36.
Goodman Wild Lakes
58.
South Beach
37.
Grandma Lake Wetlands
59.
Stebbins Gulch
38.
Green Bay West Shore
60.
Temky Bog
39.
Hemlock Lake
61.
Tontogany Prairie South
40.
Highway 70 Virgin Maple Woods
-------
Page II - 14
WISCONSIN (cont.)
98.
Raspberry Island
99.
Rat Lake
41.
HortonvilU Bog
100.
Renak-Polak Maple-Beech Woods
42.
Island Lake Stale Trust Land
101.
Rhine Center Bog
43.
Jackson Harbor Ridge*
102.
Ridges Sanctuary
44.
Jackson March Swamp
103.
Riley Lake and Conifer Swamp
45.
Johnson Springs
104.
Riveredge Creek
46.
Jone* Swamp
105.
Rush Lake
47.
Kakagon Sloughs
106.
Rushes Lake Cedar Swamp
48.
Kettle Moraine Red Oaks
107.
Sadjak Springs
49.
Kewaikum Wood*
108.
Sandldand
50.
Kohler Park Dunes
109.
Scout Lake
51.
La litem) an Lake (South)
110.
Seidel Lake
52.
Lawrence Creek
111.
Shivering Sands Wetland Complex
53.
Lewiston Sedge Meadow
112.
Shoe Lake
54.
Long Island • Chequamegon Point
113.
Silver Lake - Shields
55.
Loct Creek Embayment
114.
Skunk-Foster Lake
56.
Lower Baas Lake
115.
Snow Falls Creek
57.
Lower Hiwanka Lake
116.
Spider Lake Black Asb Swamp
58.
Lunch Creek Wetland*
117.
Spread Eagle Barrens
59.
Manger Boreal For eat
118.
Springvaie Wet Prairie
60.
Marinette County Beech Forest
119.
Stockton Island Tombolo
61.
Marshall's Point
120.
SummenoD Bog
62.
Match Cedar Swamp
121.
Swader Tamarack Swamp
63.
McCall Lake*
122.
Thompson Hill Maskik
64.
McDougal Springs and Cedar*
123.
Three Springs Creek
65.
McPhiUips Mod Lake
124.
Toft Point
66.
Menominee Creek Maskik
125.
Van Zile Cedars
67.
Millhome Dry-Mesic Forest
126.
West Alaska Lake
68.
Mink River Skiugh
127.
Westphal Woods
69.
Miacauno Cedar Swamp
128.
White River Marsh
70.
Moquah Barrens
129.
White fish Dunes
71.
Mud Lake
130.
Wisconsin Paint
72.
Mud Lake - Springwster
73.
Mud Lake Wildlife Am
74.
Mullet Lake
75.
Murphy Lake
76.
Myklebust Lake
77.
New Hope Pine*
78.
Newport State Park
79.
North Twin Island
80.
Oak Island
81.
Oak Lake
82.
Ordwty Pines
83.
Osamiwaeket Island
84.
Otter Island Cliff*
85.
Outer Island Hemlocks
86.
Outer Island Sand Spit
87.
88.
Peshtigo Harbor Wildlife Area
Piel Creek
89.
Pine Bluff
90.
Pine Lake
91.
Point Beach Ridge*
92.
Pope Lake
93.
Popple Lakes
94.
Port Wing Boreal Forest
95.
Puchyan Prairie
96.
Quarto Lake
97.
Raspberry Bay
-------
EXHIBIT B-2
ILLINOIS
j,. - 2 • l- *rji
10 20 30 40 80 eo 70
0 }Q 40 <0 tO 100
Oeartype*
County Outline
ILLINOIS
¦®A.MC
*•«#«
w#m m
-------
EXHIBIT B-3
JS^5RftriaKtjBKl6i~- >o-.-ku^»i*-
INDIANA
LaV^F"
r##h>'25 21 /
&0 2° 38!' 55r
°35 c,i,u
-------
EXHIBIT B-4
!7V
100
J120
66
.70'
132
SO 75 MMi
0 25 90 75 100 K*om*«n
Oeartype
County Outline
MICHIGAN
-------
EXHIBIT B-5
MINNtbUU
jo-» .-|.tftfw.i>glA^ffl^ai^«M>»,*| -%-m ;v.l *u
Qeartype*
County Outline
MINNESOTA
MMC
10 5 4N5»oj
1 24 2S
kaSoySh
F
-wanes*jwawfc^»g«6
gv*?rfc>y-- f^Tr-- -M'-rvry'
-------
\
A
» I » I » I « I » I « I » I « I
11 1 11 I U I H
It
17
"35
H 54
32 33
OTSEGO
YATES
CAYUGA
CATTARAUGUS
SCHUYLER
CHAUTAUQUA
DUTCHESS
37 39 40 A3 49
"CSV
IcmesTT*
ST-
HESTER
QUEENS
KINGS
NASSAU
CLEAR. TYPE
COUNTY OUTLINE
NEW YORK
Sctk of Mite.
MAP NO. MO
COPYRIGHT
AMERICAN MAP COMPANY, INC.
MHdflNi mmp m*r ta m
r«(t«w4 mnj Im ky *1 ——
hvmc m *c«,
n ITW* tK#ttK M«# Cm . Inc.
I! /
A
/ 1
i—r
¦ « i » ' «' 7 i
II I n I ~
it I iT
is _M
-------
OHIO
EXHIBIT B-7
JWU1
30"® 21
h
Gecrtype*
County Outline
OHIO
«AMC
0 10 20 30 40 S0M4W
20 40 60 Ktoffwtsrt
/ ¦ -•1 L^|.uj:Ua. ¦ I v^to 13 -¦) .-. H --^
D*^» 233
-------
2 Lake Erie Bioreserve = Lake Erie Watershed
SUSQUEHANNA
BRADFORD
WAYNE
• CRAWFORD
LACKA-
WANNA
WYOMING
VENANGO
LYCOMING
lUZlW*
CICAftffCLO
COLUMBIA
CENTRE
LAWRENCE
CARBON
NORTHAMPTON
SNtDIR
ItHlGM
MONTGOMERY
CUMBERLANO
CHESTER
LANCASTER
/CLEARTYPE
1 COUNTY OUTLINE
PENNSYLVANIA
-------
WISCONSIN
EXHIBIT R-Q
County Outline
WISCONSIN
v^AMC
aoM.iti
47
> ASHLAND
(to Kik>rn«l«rs
, WASHBURN I SAWYER
|17»12S
40 56
81 ig3
TAVLO*
6187110
52 17 ?"
32 113 95
¦*»••• XI
-------
Page n - 15
C. EXOTIC SPECIES INTRODUCTION TO THE GREAT LAKES
Problem Area Description
A number of species of fish and invertebrates not native to the Great Lakes have been introduced to the Lakes over
the last several decades. Since these species are exotic and therefore do not have any naturally occurring predators,
their populations are virtually unchecked. As the populations of exotic species increase they impact native species
through displacement, competition for food supplies, and predation.
The magnitude of the exotic species problem in the Great Lakes can be measured by a population estimates of the
species, the extent of their distribution, the likelihood that they will expand their ranges, and their impacts. The
environmental and economic impacts from each of the exotic species will be discussed in detail in later sections.
The environmental costs of the exotic species in the Great Lakes stem largely from actual and estimated losses of
sport fishes. Commercial and subsistence fishing may face similar costs. Unfortunately, even estimated values of
lost fish are not available for til exotic species.
The list of exotic species in the Great Lakes is extensive. It includes species which were introduced both
accidentally or purposefully, and species which are considered both beneficial and detrimental. This paper will focus
on the following exotic species that have the potential to have a significantly negative impact on the Great Lakes
ecosystem as well as its regional economy:
¦ Sea lamprey;
¦ Zebn Mussel;
at Alewife;
¦ Ruffe; and
¦ Spiny Water Flea.
There are a number of other exotic species causing researchers concern, including the three spine stickleback, and
the Asian dam. Unfortunately, information was not available on these exotic species for this report
In addition, although exotic to the Great Lakes, the stocked salmonoids, such as the ciiinook and coho salmon, and
other non-native sport fishes, will not be discussed here. These species have impacted the Great Lakes aquatic
ecosystem in a variety of ways that are often either unnoticed or little understood. For example, sport fishes provide
a pathway for human exposure to heavy metals and synthetic organic chemicals. Yet, while they are a part of forces
that have permanently changed the ecosystem, they are deemed beneficial by the general public. They are widely
considered desirable species and are intensively managed by state and federal authorities to ensure their continued
survival in the lakes and therefore will not be discussed in detail.
Sea Lamprey
The sea lamprey is a parasitic eel that feeds on and can ultimately kill desirable sport fish. Native to Ihe Atlantic
Ocean, the St Lawrence River and Lake Ontario, the sea lamprey entered the remaining Great Lakes via die Welland
Canal in the early 1930s. The sea lamprey devastated native lake trout population causing the populations of the
spoit fish, and other fish species, to collapse in the late 1950s.
Magnitude of the Problem
Due to widespread use of applications of the chemical lampricide TFM, lamprey populations are significantly lower
now than they were during the late 1940s through early 1960s when populations in various lakes were at their peak.
-------
Page II -16
One 1987 study estimates the number of adult lamprey in the most heavily populated areas of the upper Great Lakes
at approximately 390,000 individuals.17 By contrast, in 1949 lamprey populations in Lake Huron were at their peak
as measured by the approximately 25,000 lamprey counted at one sampling weir. Lamprey numbers reached their
peak in Lake Superior in 1961."
Today, the sea lamprey population is distributed widely over the three upper Great Lakes of Superior, Michigan, and
Huron and their tributaries where the lamprey spawns. The largest populations are concentrated in the northern
section of Lake Huron, northern Lake Michigan, and various sections of Lake Superior. The single largest population
in is the northern section of Lake Huron neaT the St. Marys River. The laige size and swift currents of the SL Marys
River have severely limited the success of lampricide applications in the river. As a result a relatively laige
population of lamprey survive in the river and nearby portions of Lake Huron.19
Ironically, improvements in water quality are believed to have allowed the sea lamprey to repopulate some lake
tributaries that were effectively treated with lampricides in the past The high costs of lampricides and the need for
periodic applications in lamprey preferred tributaries will likely allow sea lamprey to at least maintain their present
geographic distribution in the upper Great Lakes.
Ecological Impacts: Severity
To survive, the sea lamprey must feed on the bodily fluids of fish. Upon arriving to the Great Lakes, the lamprey
found a large supply of food in a number of native fish species, particularly the lake trout. The ecological impacts
of the infestation of the sea lamprey were particularly devastating during the late 1930s through the 1960s. At that
time lake trout production in each of the upper Lakes fell by as much as 97 per cent over IS years, eventually
collapsing completely in Lake Huron in 1959.30 Lamprey then turned to other native fish species such as the turbot,
lake whitefish, walleye, and rainbow trout
The ecological impact of the lamprey's rapid depletion of predator species was obvious soon afterwards. Populations
of smaller baitfish, namely the cisco and the exotic alewife, were unchecked and exploded. The lamprey turned their
attention to the larger of these baitGsh as the populations of their preferred hosts fell. Two of the largest cisco
species were hit first and were reduced to near extinction in the early 1960s.21 The flourishing alewife population
also forced serious reductions in a number of native species such as the lake herring, yellow perch, and the emerald
shiner.
The application of lampricides and intensively fish stocking programs allowed lake trout populations to rebound.
However, the remaining lamprey population is still impacting game fish in the lakes. In 1981 and 1985,60 per cent
of the chinook salmon sampled in the St Marys River were wounded by lamprey with a wound rate of 130 wounds
per 100 fish.®
"Water Quality Board, International Joiat CemaUsskw. 1987. "1987 Report oa Gnat Lakaa Walar Quality," p. 2J • 65.
"Smith, B. K, and J. J. Tibbie*. 19M. "Sea Lamprey fa Lake* Haroa, Michigan ud Ssptrion hictory of favaakw aad control,
1936-78," fa Canadian Journal af Fiaherte* aad Aquatic Sdeac* 37t pp. 1780-1801.
"Water Quality Board, International Joint Cnmaihrioa. 1987. "1987 Report oa Gnat Lakaa Water Quality," p. 13 > 65.
"Smith, B. It, aad J. J. Tibbies. 1980. "Sea Lamprey fa Lakaa Huroa, MicUgaa aad Sapatoi history of favaatoa aad control,
1936-78," fa Canadian Jonrnai of Fisheries and Aquatic Sdeaces 37) pp. 1780-1801.
"Smith, B. It, aad J. J. Tibbie*. 1980. "Sea Lamprey fa Lakes Huron, MidMgaa aad Superior! history of invasion aad control,
1936-78," in Caaadiaa Joaraal at Fisheries aad Aquatic Sdeacec 37i pp. 1780-1801.
"Water Quality Board, International Joiat Coaimhsioa. 1987. "1987 Report oa Gnat Lake* Water Quality," p. 2.3 - 65.
-------
Page H - 17
The invasion of the sea lamprey in the Great Lakes, and more specifically, the resulting lamprey control program
has caused secondary ecological impacts in the Great Lakes' ecosystem. The use of lampricides, applied in
tributaries to kill larval lamprey, has killed numerous non-target species. Total numbers of non-target species killed
by TFM are not known but several incidents are noted. In approximately 12 separate incidents, lampricide
applications killed over:
¦ 75,000 white and longnose suckers;
¦ 12,500 northern pike;
¦ 6000 walleye;
¦ 5000 carp; and
¦ 2000 brown trout33
Given the thousands of applications of lampricides since the lamprey control program began, these fish kills likely
account for only a small percentage of species impacted by die control program. Lampricides are also known to
adversely affect aquatic invertebrates, especially mayfly nymphs. Lamprey barriers have also caused non-taiget fish
kills by trapping fish.
Ecological Impacts: Reversibility
The reversibility of the present populations of sea lamprey in the Great Lakes and their impacts on the ecosystem
appears to be low. The lamprey control efforts during the past few decades have reduced sea lamprey populations
in the Lakes but the rates of decrease have slowed significantly in the past few years suggesting that sea lamprey
populations may be stabilizing at current levels. The lamprey population in the St Marys River is very difficult to
control given the limited effects of lampricide treatments due to die size and swift currents in the river. The rising
costs of the lampricides and limited budgets will also likely result in less intensive control treatments in the future
making it more difficult to reduce present lamprey populations.
Welfare Impacts: Environmental Costs
The sea lamprey caused enormous losses to the commercial and sport fish industries during the late 1940s through
the early 1960s when lamprey populations were at their peak in various lakes. The sea lamprey devastated
populations of many important sport fishes, especially the lake trout, during these years as well as afterwards. By
1950 the lake trout almost disappeared with populations at only 10 percent of their average during the previous
decade. Lamprey also preyed heavily on lake whitefish, another important commercial fish species.
There is little information on the value of lost fish harvests due to lamprey predation. In 1985 the Great Lakes
Fishery Commission (GLFC), however, estimated the annual gross economic value (consumer expenditures plus
willingness to pay) of the U.S. Great Lakes fishery at $33 billion.14 The GLFC further estimated the economic
value of total losses in U.S. fishing opportunity without a sea lamprey control program at approximately $260 million
in 1985. This estimate was made at a time when the sea lamprey population was lower than previously as a result
of an extensive lamprey control program over the last few decades. The annual economic loss from lamprey
predation during the years of heaviest impact on sport fisheries was likely to be much greater than the estimate.
"Dahl, F. 11, and R. B. McDonald. I960. "EDidi cf Coatrol of the Sta Lamprey on Migratory ud RcsMoat Flch Population*,"
in Canadian Journal of Fisheries and Aquatic Sciences 37i pp. 1S86-1894.
"Great Lakes Fbhery Commiuioa. 19S9. "TW Report of the Evalaatioa cf lb Graat Lalua Flahery Coaunbsioa by the Bl-
Natkmal Evaluation Team: an aaalyris of the wwk coatribotioB of the Gnat Lata Ma lamprey program," p. 6.
-------
Page D- 18
Welfare Impacts: Material and Property Damage
Part of Ihe social cost of the sea lamprey introduction into the Great Lakes is the cost of controlling this parasite.
The GLFC estimated that the sea lamprey control program in the U.S. cost taxpayers nearly $6.6 million in 1988.23
Since this estimate covers only one year, it does not include the hundreds of millions of dollars likely spent in the
past on such items as researching various lamprey control strategies, the expenditures for previous applications of
expensive lampricides especially during the peak years of infestations, the construction and maintenance of lamprey
barriers, and other lamprey control expenditures. No figures were available on the total costs of the lamprey control
program to date.
Zebra Mussel
The zebra mussel is small (1-2 inches long) freshwater shellfish native to areas of the Blade and Caspian seas in the
southern region of the Soviet Union. Zebra mussels were apparently introduced in the Great Lakes in 1985 or 1986
when one or more transoceanic ships discharged ballast water into Lake St Clair. Zebra mussels are extremely
prolific with females capable of producing 30,000 to 40,000 eggs each year. With few predators, zebra mussels are
spreading quickly and causing various environmental and economic damages in the Great Lakes.
Magnitude of the Problem
Zebra mussels, which first appeared in Lake St Clair in 1985 or 1986, are now found in each of the Great T
Exhibit C-l shows the extent of the mussel's geographic distribution as of July, 1990. Since then, its range ha
continued to expand, and has now gone beyond the Great Lakes Basin. This rapid spread is due in part to thei
prolific reproduction rates. Females can produce between 30,000 and 40,000 eggs cadi year. Water temperatures
in the lakes are also very conducive to the mussels reproductive cycles. Researchers believe that water temperature
allow females to produce eggs from late April to October, a few months longer than in the mussels' own natiy5
habitat.*5,27 Once eggs hatch, lake currents spread the bee-swimming larvae far and wide. e
The increase in concentrations of zebra mussels is as astounding as the speed with which the mollusk has spread
across the Lakes. The zebra mussel, which clusters in colonies and must attach to a firm surface or die, have
increased in concentrations from 50 per square meter to over 700,000 per square meter in one year in Lake Erie near
Monroe, Michigan.21 In addition to the fertile females and the relatively long reproductive season, the wide
availability of plankton in the lakes, especially Lake Erie, allows the zebra mussel to thrive. Compounding the
problem, the zebra mussel population is virtually unchecked by currently available natural controls such as predators
Diving ducks such as bluegills and the lesser scalp appear to feed on zebn mussels but researchers do not feel they
will significantly reduce zebra mussel populations due to their migration periods and limited numbers.29*10
Most authorities are virtually certain that the zebra mussel will spread quickly to inland lakes and rivers across North
"Great Lain* Fishery CoooWm. 1989. "The Report of the Evaluation of the Gnat Lake* Fishery Cntnmiwion by the B»-
National Evaluation Teami aa analysis of the economic cautribation of the Gnat Lakes w lamprey program," p. 14
"Snyder, F. L, Ohio Saa Gnat College Program. 1990. "Zebra Maxell k> the Great Llluu The invasion and ill
implication*," p. L
"Mackie, G. L, W. N. Gibbons, B. W. Mnncaster, and L M, Gray, Environment OaUrio. 1989. "The Zebra Maud, Prrinena
Polvmorphai A Synthesis of European Experiences aad a Preview for North America," pp. 6-7.
"Green, Larry, Loe Angeles Times. January 15,1990. "Invasion of Zebra Miuaeb Threatens US. Waterway*."
"Wbac, John, The Vindicator. December 24,1989. "Trash' Fkb, Dub Could Sare Lake Erie."
"EPA Region V. 19S9. "Zebra Mauris Fact Sheet," p. L
-------
Exhibit C-1
REPORTED GREAT LAKES
ZEBRA MUSSEL DISTRIBUTION
APRIL 26,1991
LAKE SUPERIOR
LAKE
HURON
DNTO
Irl
Ml
MUSKEGON
HOLLAND
BUFFALO
LAKE
DETROIT
KENOSHA _
CHICAGO ~
IL
i IN
GARY
SOURCES: Sen. John Glenn's Office
U.S. Environmental Protection Agency
U.S. Fish and Wildlife Service
Wisconsin Sea Grant
ZEBRA MUSSEL
AREA OF INTRUSION
&EPA
UNITED ITATE1
ENVItONMEMTAl PIOTECTION ACENCY
GHAT LAKES NATIONAL PIOCIAM OFFICI
-------
Page n - 19
America. The only questions are how quickly this infestation will occur, and what control methods can be used
successfully.31 The mussels can spread in several ways, including:
¦ Great Lakes water diversions such as the Erie and the Chicago Sanitary Canals offer zebra mussels the
easiest route to inland lakes and river systems;
¦ Fishermen and boaters can unknowingly transport zebra mussel larvae, invisible to the naked eye, to other
waters in bait containers, cooling systems in boat engines, or live wells;
¦ Water fowl and other animals can transport larvae in wet feathers or fur, and
¦ Adult zebra mussels can survive out of the water for several days and thus can be transported to inland
lakes and rivers attached to boat hulls or a number of other objects.32
Ecological Impacts: Severity
Zebra mussel food requirements and their feeding methods lead to a growing concern over the exotic mollusk's
potential to disrupt the Great Lakes ecosystem. Each adult mussel filters approximately one liter of water per dav
feeding on particulate matter including plankton.33 The growing population of zebra mussels thus is capable of
removing considerable quantities of plankton from the water which is likely to have a devastating impact on n
entire food chain. The diminished supply of plankton has a significant impact on the lower levels of the food ch«*
such as microscopic crustaceans which are a source of food for larval and baitfish which in turn support larger fish
and other animal species.
Zebra mussel habitat preferences and their ability to cluster in large colonies is another source of concern over th
mussels likely ecological impacts. The zebra mussel colonizes hard surfaces such as rocky reefs forming layers th ^
are often several inches thick. Such colonies of zebra mussel displace walleye, yellow perch, white bass, and
smallmouth bass that spawn on these rocky reefs.34 Researchers are also investigating the potential of zebra mussel
feces to deplete dissolved oxygen levels and raise the pH of the water further impacting native sport fishes.33
The combined effects of impacts to the food chain and habitat displacement are leading researchers to portend
devastating impacts to the sport fisheries, especially walleyes, in the Lakes. Although researchers have not been able
to analyze all the data yet, many do believe that slight declines in fish populations in Lake Erie are a result of the
" Green, Liny, Lo* Angela Tim**. January IS, 1990. "Invasion of Zebra Mussels Threatens VS. Witewiji)' Haorwilz,
Ralph, The Pittsburgh Prtu. February 26,199a "Clogging by Mussel Predicted in River*)" Sayder, F. L, Ohio Sea Grant College
Program. 199a "Zebra Mustek in the Gnat Lakes) The invasion and its implications," p. 4; Gerdes, WyBe, Detroit Free Pro,
April 5,199a "Zebra Mussels May Leapfrog to Inland Lakes by Way ef Boats."
"Snyder, F. L, Ohio Sea Grant College PrograuL 199a "Zebra Massels hi the Great Lakes: The invasion and its
implications," pp. 3-4.
" Snyder, F. L Ohio Sea Grant College Program. 199a "Zebra Mussels in the Great Laktsi The invasion and Its
implications," p. 3,
^Gerdes, WyBe, Detroit Free Press. April 5,199a "Zebra Mussels May Leapfrog to Iaiaud Lakes by Way of Boats)"
McMeekin, Carolyn, Niagara Gaiette. 1990. "Molhulu Endanger Fisbi" aad Pollick, Steve, He Toledo Blade. February 4,199a
"Waning Flags are Waving over Lake Erie Walleye Fishery."
"Snyder, F. L, Ohio Sea Grant College Program. 199a "Zebra Mussels in the Great Lakes: The invasion aad its
implications," p. 2.
-------
Page II - 20
laige population of zebra mussels in tbat lake.36
Ecological Impacts: Reversibility
The reversibility of the present zebra mussel population in the Great Lakes as well as their impacts on the ecosystem
is very low. The favorable water temperatures in most of the lakes, the readily available supply of food, and the
mussel's prolific rate of reproduction all suggest tbat the distribution and population of zebra mussels in the Great
Lakes will expand rather than shrink in the future. In addition, the mussels ability to survive out of the water for
several hours, the microscopic size of its larvae, and the fact that fishermen often remove their boats from the Great
Lakes and launch them soon afterwards in inland lakes and rivers suggests that zebra mussel populations will likely
expand beyond the Great Lakes.
Welfare Impacts: Environmental Costs
The environmental costs of the zebra mussel invasion in the Great Lakes are largely unknown but officials have
made a number of gross estimates. One official of the U.S. Fish and Wildlife Service estimated that the exotic
mussel could cut die value of the commercial and sport fisheries in die Lakes in half over the next 10 years. As
stated above, the Great Lakes Fishery Commission has estimated the economic value of the Great Lakes fishery at
$3.3 billion dollars.
The zebra mussels will also negatively impact recreational activities at beaches along the Great Lakes. As the zebra
mussels die, their shells wash up on beaches. The razor-sharp edges of the shells can easily cut beachgoers' feet
In the fall of 1989 extensive deposits of zebra mussel shells were evident on several Lake Erie beaches.17 Dead
zebra mussels can also produce offensive quantities of methane gas giving the water a foul taste and smell.1*
Great Lakes legislators and other officials in the Basin have made estimates of the total economic damages in the
Great Lakes from the zebra mussel. These estimates range from $4 to $7 billion dollars over the next 10 years.19
These estimates go beyond environmental costs alone, and include estimated damages to materials and property, and
costs of controlling the exotic mussel.
Welfare Impacts: Material and Property Damage
Over the few years that the zebra mussel has invaded the Great Lakes, it has caused a significant amount of material
and property damage. The extent of this damage is likely to only increase as zebra mussel populations expand. The
zebra mussels ability to form large colonies on firm surfaces is the main cause of this damage. Documented cases
of zebra mussel damages are replete with incidents of mussel colonies clogging industrial and municipal water intake
pipes.
¦ Zebra mussels clogged intake screens at Detroit Edison's Power Plant in Monroe, Michigan in the summer
"Levy, Michael, The Buffalo News. Decmher 5,1M9. "Onal oT Ztbn Muul fa Spreading from Lib Eric to Oatario."
"Snyder, F. L, Okie S*a Grant College Program. 1990. "Zebra Mawk b the Great I alreet The iavatkw and Its
implications," p. 3.
"New York Haes. March 12,1990. "U.S. Help fa Sought to Cut Maseeh."
"New York Times. March 12,1990. "U.S. Help fa Sought to Curb Mouebj" Cronan, Jan, Sounding*. March 1990.
"Multiplying Molluik a $7 Billion Problem for Gnat Lake* Areaj" aad Green, Larry, Los Angela* Timet. January 15,1990.
"Invasion of Zebra Mustek Threatens US. Waterways."
-------
Page n - 21
of 1989. Remedial activities were estimated to cost $100,OOO.40
¦ A water intake pipe at the Monroe Water Treatment Plant in Monroe, Michigan was clogged by zebra
mussel colonies on two separate occasions in January and December 1989. The plant operator estimated
it would cost about $100,000 to mechanically dean the intake. The reduced flows caused a number of bars,
restaurants, and manufacturing plants to close for a day or longer. One restauranteur estimated he lost
$5,000 in business. A manager at one steel plant that shut down operations for one day estimated that lost
production cost the plant as much as $40,000."
Several hundred other manufacturing plants located on the shores of the Great Lakes have water intake pipes that
are at risk of being clogged by zebra mussels.
Control of zebra mussel populations can be expensive. For example, in October, 1990, Bethlehem Steel and Tn».nij
Steel spent $41,000 and $49,000 respectively on molluscicides to control zebra mussels at their water intakes in
Southern Lake Michigan. The use of molluscicides show not only a direct economic damage from zebra mussels
but a potential for further deterioration of water quality and new toxic threats from the techniques chosen to fight
this exotic species. The consequences of the introduction of such chemicals should be weighed carefully
those other forms of treatment EPA Region 5 is presently encouraging industries to control zebra mussel infestations
with CO, and chemicals of raw toxics by.
In addition to water intake problems, zebra mussels are likely to cause significant economic losses to materials «nd
property. Examples of such impacts include damages to:
¦ boats and ships where zebra mussels dog engines and attach to hulls;42
¦ buoys and floating docks that can be colonized by zebra mussels to the point of submergence;43
Unfortunately dollar estimates of these damages are not available.
As in the case of the sea lamprey, the total cost of the zebra mussel infestation of the Great Lakes must indude
expenditures on control policies. Members of Congress from the Great Lakes States proposed spending $40 million
dollars to research and apply measures to control the zebra mussel.44 At an EPA workshop on zebra mussels in
September, 1990, several current and potential control measures were discussed.43 Alternatives would have
economic and environmental costs. Methods indude:
¦ Chemical actions: oxidizing chemicals such as hydrogen peroxide, and non-oxidizing chemicals such as
carbon dioxide or molluscicides;
"Woaalalc, Mary, Niagara Gautt*. 1990. 'MmwI bound Flub Soak Safe Way to Oaat lavadm."
"Woiaiak, Mary, Niagara Gaiette. 1990. "Ma«el Maaaca has Akaady Hit Otkar Water Ptaat"
"Maddo, G. L, W. N. Gibbons, B. W. Maacaatar, aad L M. Gray, EfcrirauMrt Ontario. 19t9. "The Zabrm M Basel, Drtiu«n»
Pohrnornhat A Svthaah of Europeaa Expcricmca* aad a Prrrfaw for North AmHcb," p. 22| aad Jahaaoa, Aaa, Eric Daily Timas.
Mink 28,1990. "Zebra Maaaals Expected to Plague BoaUrv"
"EPA Region V. 1989. "Zebra Mussels Fact Sheot," p. X
4*T«raerI Douglas, Buffalo News. Mardi 8,1990. "$40 MOlioa Sought to Fight Zebra MaaaaL"
*Ik« Baal report for this woriuhop I* doc la aiid-Noveaber, 199a
-------
Page H - 22
¦ Mechanical actions: removal by scraping or other means, and engineering/design such as deep water intakes
or laiger pipes;
¦ Physical actions: the use of screens or filters such as infiltration beds, strainers or biofilters, and other
physical actions such as heat, acoustics, magnetic fields, irradiation, and others;
¦ Biological actions: such as the use of parasites, natural or bioengineered disease, human or wildlife
piedation.
Potential economic benefits from the harvesting of zebra mussels for commercial, industrial, or other uses should
also be examined. Such uses of the exotic species, if found, may provide further economic incentives to control their
numbers.
Alewife
The alewife is a small baitfish that was introduced into the Great Lakes in the late 1800s. Alewife populations
frequently surge and then crash creating massive die-offs that foul beaches, reduce supplies of dissolved oxygen in
the lakes, and lead to obnoxious odor problems.
Magnitude of the Problem
Alewives were first identified in die Great Lakes in 1873 when an established colony was found in Lake Ontario.46
Researchers, however, are unsure whether they migrated themselves through the Eric baige canal system or were
inadvertently introduced during stocking efforts of the American Shad. Regardless of their origin, alewife
populations are seen as both a curse and a Messing. Massive die-offs foul beaches, but the small baitfish is also a
critical source of forage for the native Lake Trout as well as stocked salmon populations in the Great Lakes.47
Due to the favorable habitat in the Great Lakes, alewives quickly expanded their range in some lakes as well as
spread to other lakes. They circumnavigated the Niagara Falls through the Wdland Canal and were first reported
in Lake Erie in 1931. From there, however, their expansion to die other lakes was much more rapid.
The size of the alewife population in the Great Lakes has fluctuated widely over the years. Alewife numbers rose
dramatically, especially in Lake Michigan, after the precipitous decline in native Lake Trout population in the late
1950s and early 1960s due to predation by the sea lamprey. Estimates of biomass of alewives in Lake Michigan
during the late 1960s through the late 1970s showed less radical change but still fluctuated. Alewife biomass in Lake
Michigan rose gradually from 46,000 metric tons in 1967 to 114,000 in 1973, then declined to 45,000 in 1977 only
to rise again to 77,000 metric tons in 1978.41
Alewife biomass surveys in Lake Huron revealed lower total alewife biomass in that lake, although numbers still
fluctuated. Alewife biomass estimates in Lake Huron increased sharply from 15,000 metric tons in the spring of
1973 to 71,000 in 1975, then dropped as sharply to 14,000 metric tons one year later. Alewife biomass estimates
Lee, VS. Elah ami Wildlife Service. IMS. "Review «T Fkk Specie* iBtrodaced Mo tk« Gnat Lalcee, 1819-1974," p. 4.
"Arfyle, R. L* US. Flih and WUdlift Service. 1982. "Akwivw and Rainbow Smelt Id Lake Huron: Mid water and Bottom
Agmiliau and ErtimaU* of Standing Stocks" in Tmuadiou of the Aaierican FWieriee Social; 111:, p. 267.
•Hatch, R. W„ Haadc, P. M, and Brown, Jr, E. H. "EetimatfoD of Alewife Bioman ia Lake Michigan, 1967-1978" fa
Traauctioaj of the American FUierie* Society, U0i, p. 575.
-------
Page n • 23
climbed back to 41,000 metric tons in 1978, dropping again to 28,000 in 1980.49
Ecological Impacts: Severity
Increases in alewife populations over the yean have severely impacted populations of several fish species native to
the Great Lakes including the bloater, lake herring, emerald shiner, yellow perch, and deepwater sculpin. The
populations of these species in Lake Michigan all declined when alewife populations were high.* The lake herring
and emerald shiner were particularly affected as their numbers did not rebound when alewife populations declined.
Researchers believe that alewives impact native species in a number of ways. First, it is believed that alewives
adversely affect other species through the competition for food. Zooplankton populations in Lake Michigan were
subjected to severe changes between 1954 and 1966 which corresponded to periods of a tremendous abundance of
alewives in the lake.31 Other researchers have found evidence that alewives impact other fish «p»"ir« by preying
on their larvae or eggs."
Rainbow Smelt and Alewife53
The ecological impacts of the rainbow smelt and the alewife in the Great Lakes are similar. The rainbow smelt
the alewife serve as both prey and predator. As prey, these exotic fish to the Great Lakes, have become an importa***
source of food for the native lake trout and the introduced coho and chinook salmon. This new food source has ***
however, been a clear benefit for these sport fishes. Several researchers have found that the accumulation of mercu^
in the lake trout accelerates with increases in the consumption of rainbow smelt.
It is as a predator, however, that the rainbow smelt and alewife have perhaps their greatest ecological impact in th
Great Lakes. Together the rainbow smelt and alewife have adversely impacted a number of native species of f ^
The rainbow smelt is known to feed on lake trout eggs and larvae, but researchers feel this predation is of li«i
consequence. It is also believed that the rainbow smelt has impacted recruitment of the bloater chub, but researche 6
have not confirmed this. Smelt and alewives, however, are known to have had widespread impacts on those fi v!
species that have diets similar to theirs. After the establishment of the rainbow smelt and alewife in Lake Michigan
the abundance of a number of species that have diets similar to these exotic species' diets, such as emerald shiners'
lake herring, and deepwater ciscoes, declined. '
Perhaps the greatest impact of the rainbow smelt, however, is on lake hening populations. Researchers speculate
that rainbow smelt predation on young lake herring adversely affects this native species more than any adverse
impact this exotic species has on lake herring through competition. Interestingly, the rainbow smelt and alewife also
adversely affect each other as well as native fish species. However, the full impact of this interaction is not clear
"Argy)*» & L Ui. Fish aad WUdlifc Service. 1982. "Alewives aad Rainbow SbiK ta Lake Huron! Midwatcr and Bottom
Aggregations and Estimate* of Standing Stocks" k Transactions of the Aawrkaa Fish tries Society lilt, p. 267.
"Eclt, G. W. aad Wells, L. 1987, "Recast Changes hi Lake Michigan's Fish Connnalty aad Their Probable Causes, with
l&nphasb ea the Role of the Alewife" h Canadian Journal ef Fisheries and Aquatic Sdeaces 44>, p. 53.
"Colby, P. J. 1971. "Alewife Dieoflb Why do they eccar?" in Lhnaoe 4i, p. 26.
nEck, G. W. aad Wells, L. 1917. "Recast Changes fa Lake Mkhigaa's Fish Community and Their Probable Causes, with
Emphasis on the Role of the Alewife" fa Caaadian Journal of Fisheries aad Aquatic Sdeaces 44s, p. 53.
" Information for Ike raiabow sadt aad alewife from Evaat, David O. aad Loftas, David H 1987. "Colonization of Inlaad
Lakes in the Great Lakes Region by Rainbow Smelt, Osroerus mordac Their Freshwater Niche and Effects oa Indigeaou Fishes,"
Canadian Journal of Fisheries and Aquatic Sdeaces 44s pp. 249-266. At the thae of this writing, at least one other report on the
rainbow smelt and alewife, which aiay yield further information, has not been received.
-------
Page D - 24
Some researchers have repotted that smelt adversely ¦fleet recruitment of the alewife while other researchers have
found that the abundance of rainbow smelt have declined during periods of alewife population increases.
Ecological Impacts: Reversibility
Due to its presence in the Great Lakes for over 100 years and its importance to maintaining native Lake Trout and
stocked salmon populations, the likely reversibility of the alewife population is very low.
Welfare Impacts: Environmental Costs
No information was available at the time of this report
Welfare Impacts: Material and Property Damage
The alewife has caused a variety of material and property damages principally as a result of massive die-offs in
specific lakes. Decaying alewife carcasses from large scale die-ofls foul recreational beaches creating obnoxious
odor problems, dog municipal and industrial water intakes, and deplete dissolved oxygen levels in the lakes.
The most infamous alewife die-off occurred in Lake Michigan in 1967. This nationally publicized incident involved
an alewife die-off of astounding proportions, with one estimate placing the amount of dead alewives at several
hundred millions pounds. Tremendous numbers of dead alewives washed up on beaches in Chicago. The resulting
odor problems from the decaying fish forced city officials to bulldoze beaches to bury the dead fish. Large mats
of floating, dead alewives covered harbors in Chicago, and drinking water intakes collapsed from the weight of
alewives, requiring officials to temporarily shut down the plant for repairs. One researcher estimated the costs and
economic losses to industry, municipalities, and recreational interests from this alewife die-off to be in excess of
$100 million. A similar die-off in the southern end of Lake Huron and in St Clair River resulted in an estimated
200 million dead alewives.*4
Die-ofls similar to these have occurred sporadically since die late 1890s when the alewife was introduced in the
Great Lakes. Researchers believe that the die-ofls occur as a result laige alewife population coupled with severe
stress from low water temperatures during cold winters. A predicted large alewife die-off during the winter of 1989-
1990 and, although it did not materialize, fear that one may occur in the near future.19
Ruffe
The ruffe is a small fish related to the indigenous yellow perch. It is native to Europe and is believed to have been
introduced in the Great Lakes sometime between 1986 and 1987 via the discharge of ballast water. Officials are
concerned that the ruffe will impact native fish populations. The ruffe has already been found to eat whitefish eggs.
Magnitude of the Problem
The Ruffe, like the zebra mussel is very prolific with females capable of producing up to 100,000 eggs per year.
As such, the Ruffe population is growing larger each year. Introduced in 1986 or 1987 from the ballast water in
grain ships, the 1989 spring spawning population expanded at a dramatic rate to a total of 300,000 individuals. One
"Colby, P. J. 1971. "Almtfi DfeoCbt Why do they tent?" ia Ikm
-------
Page II - 25
year later, the population was estimated to be at 500,000 to 700,000.*
The Ruffe's range in the Great Lakes, however, is currently confined to the St Louis River estuary and Duluth
harbor in western Lake Superior. Although the Ruffe has not expanded beyond this portion of Lake Superior at this
time, researchers do not feel that the temperature, food supply, or other natural features of Lake Superior will act
as a barrier to the spread of this exotic specie. In fact conditions found in the near shore areas of Lake Superior are
consistent with the Ruffe's preferred habitats. Researchers believe die Ruffe will eventually find its way to the other
lakes, but feel this expansion will not take place rapidly.57
U.S. Fish and Wildlife Service (USFWS) researchers in Duluth, Minnesota have been intensively studying the
population of exotic Ruffe in the Great Lakes. To date, however, these researchers do not have any citable data on
the ecological impacts of the Ruffe. What they do know is that (1) through 1990, the abundance of Ruffe has
continued to increase dramatically although it has not expanded its range, and (2) from 1989 through 1990, every
other prey-type species (e.g., the several species of shiners and minnows) has declined in abundance. As a result
of the increase in the Ruffe population and the decline in other prey-type species, die Ruffe is the second most
abundant prey-type species in its range*
Ecological Impacts: Severity
Although there is little hard data, researchers widely believe that die Ruffe will seriously impact some native
populations. As a member of the perch family, the Ruffe is known to be an aggressive competitor of die native
yellow perch. Scientists studying exotic populations of Ruffe in Europe found that yellow perch and whitefish suffer
due to competitive displacement by the more prolific Ruffe.19
What USFWS researchers do not yet know is how, if at all, die Ruffe affects these or other species. It is possible
that efforts by the States of Wisconsin and Michigan to increase the populations of certain predator species may be
responsible for the decrease in the populations of shiners and minnows rather than the Ruffe. For the last two years
these two States have undertaken a massive program to increase the density of walleye, northern pike, and
muskellunge in an attempt to control die abundance of Ruffe through predation.
Since the population of Ruffe has continued to increase despite efforts lo control it through changes in the
management of predator species, USFWS researchers say they have started to examine the possible use of other
methods of control such as the release of sterile male Ruffe and chemical controls. Research ere said it was too early
to assess potential ecological impacts from these alternative control measures.
Researchers at USFWS say they do not yet know if the decrease in native prey species is a result of greater predation
from the increase in the density of predator sport fishes, or due to competition from the Ruffe. More generally
researchers say they are several yean away from knowing basic information regarding how the Ruffe affects other
species and the ecology of the Lakes. The USFWS researchers guaranteed that no other researchers can speak of
the ecological impacts of the Ruffe in the Great Lakes either. Additionally, the USFWS researchers said they would
be the first to know of any impacts of the Ruffe since they are the only researchers in North America undertaking
M P«r cawuliM with Jim ScQfcby, VS. Fish ud Wildlife Strict. Aagarf 27,1990.
"flbUL
* laformatfon per ca?«Mlfaw with Or. Jh Srig^bjr, VS. Fbh awd WiUUf* Scrrica, aad Or. T«a Skua, UJ. finlmMau
Science* DivUm, Rtgioo 5, m March 28,1991.
"Par cmnmIIm with Jiai Scflfthy, Vi. 91«h aid Wildlife Santo. Aafwt 27,1999.
-------
Page n - 26
a comprehensive study of the Ruffe population in the Great Lakes."9
Spiny Water Flea®
The spiny water flea (Bythotrephes Cederstroemii) is an exotic aquatic animal, or zooplankter, that was first reported
in the Great Lakes in late 1984. The spiny water flea gets its name from its very long spiny tail. The spiny water
flea is very large for a zooplankter. It is about three-eighths to one-half inch long including its taO, and is easily
seen with the naked eye. It is not known exactly how die spiny water flea found its way to the Great Lakes, but
researchers suspect it was transported from its native territories in Europe and western Soviet Union via ballast water
from ships. The spiny water flea was first found in the Great Lakes in southern Lake Huron in December, 1984.
It spread quickly to the other Great Lakes, moving east to Lakes Erie and Ontario in 1985, west into Lake Michigan
in the late summer of 1986, and northwest into Lake Superior in 1987.
The information on the ecological impacts of the spiny water flea in the Great Lakes comes primarily from over four
years of research on the effects of this exotic species on the food web of Lake Michigan. Researchers have found
that the spiny water has caused changes at different levels in the food web. In Lake Michigan, the spiny water has
selectively fed on a few species of Daphnia. the dominant herbivorous plankton in the system. Populations of certain
species of Daphnia have been severely reduced since the invasion of the spiny water flea.
The spiny water flea's impacts on Daphnia may also be having serious consequences for the bloater chub, an
important prey species for the Lakes' salmon populations. The young-of-the-year bloater chubs also feed on
Daphnia. but it appears that the spiny water flea is out-competing the bloater for this food source. The decline in
the abundance of Daphnia has also caused an apparent increase in the algae that Daphnia feed on. These effects of
the spiny water flea arc not limited to the Great Lakes. The spiny water flea has also been found in an inland lake
north of Dulutb, Minnesota.6 Officials are not certain how the spiny water flea made the trip to inland waters, but
like the zebra mussel, the spiny water flea can travel inland via minnow buckets, and bilge and live wells from
recreational boats.
Researchers theorize that the combined effects of the spiny water flea may ultimately affect sport fishes at the top
of the food web in the Great Lakes as well as inland lakes in the Basin. Researchers have written, "All these
changes [caused by the spiny water flea] may be influencing the flux of carbon and energy through the food web
in such a manner that the future may require fundamental changes in fisheries management practices and a
reevaluation of realistic sport fishing expectations. An assessment of the economic impact of the [spiny water flea]
must wait until a clear fisheries response to the fundamental and complex food web alterations of the last few years
can be ascertained."0
" Information per caawnatiow with Dr. Jim Selgtby, U5. Fbh and Wildlife Service, ami Dr. Toa» Shaon, U£ EPA Gnat
Lake* National Program OtBce, on March 2S, 1991.
*' Information on the Spiny Water Flu priaMrily from Sandgren, Craig D. and John T.1 **""r StpUmbw 1990, "Ecology of
the Cladoceran Bythotr«ph«* CatltiUiuchU (Spiny Water Flea or "BC") fa Ik* Laorentian Great Lakee", nMitod for the
Introduced Specie* Workshop. At tht time af thia writing, at but m other raport which may jUd further information oa the
Spiny Water Flea ha* not been rtcdved.
8 Tke Seidx. Fall 1990. "B.C. Jnaap to Island Lake".
° Sandgren aad I ihwiia.
-------
D.
SPILLS AND OTHER ACCIDENTAL RELEASES
Page II - 27
Problem Area Description
This section presents information about hazardous materials that are transported or spilled in the Great Lakes Rarcin
Data from a spills report by the Canadian Coast Guard and U.S. Coast Guard data are used in this section
Information on accidental releases to all media in the Great Lakes Basin is also presented here. For an
of accidental nuclear releases in the Basin, see Appendix G.
Magnitude of the Problem
Exhibits D-l through D-4 portray the geographic distributions of spills through water, air, land and ground water by
Basin county. This information was obtained from EPA's 1990 Emergency Response Network System (ERNS) fQr
accidental chemical releases.
Spills and accidental releases of oil, or other hazardous substances to the Great Lakes and its tributaries, ue »
common occurrence in the Basin. For example, reports to ERNS for 1989 indicate that 298 spills to surface waters
occurred in Michigan counties alone. Furthermore, research done by the International Joint Commission indicate
that the spills recorded in various data bases underestimates the actual number of spill occurrences. The data »l1tl
suggest that the toxic loading from spills greatly exceeds the amount of toxics released through regulated
pathways.44
Data compiled by the United States Coast Guard regarding spills in the Great Lakes and connecting
indicates that from January 1,1980 through September 30,1989 approximately 5,000 spills of oil and toxic materials
occurred. Eighty percent of those spills came from land-based facilities such as storage tanks and pipelines. For
example, one facility located an the shores of the Great Lakes stores over 1-5 million gallons of the canceled
pesticide dinoseb. Given this relatively large quantity of a canceled pesticide stored in one area, the severity of
accidental release could be devastating to the Great Lakes ecosystem in that area.
Only twenty percent of the spills in ihe Coast Guard data
came from vessels. However, given the amount of
hazardous bulk material that is transported by vessel
through the Great Lakes, there is a potential for significant
spills.
Most spills on the Great Lakes occur in port areas. This
tendency can be beneficial because of proximity to on-shore
facilities that can help to prevent the spread of the spill.
However, it could be detrimental if these spills tended to
occur near drinking water intakes of the many large centers
of population around the Lakes. Because port areas are
population centers, spills to these areas not only threaten the
local drinking water, they can be a source of airborne toxics
if the chemical spilled is volatile. In fact, most of the
volume of hazardous chemicals shipped on the Great Lakes
U.S. GREAT LAKES AREAS
WITH NUMBER OF OIL AND CHEMICAL 8PILLS
JAN. 1980 - SEPT. 1089
ocTftorr
TOLEDO
MILWAUKEE IMIIIHUMIIIIIIIIII «ot
CLEVELAND IIHIIIUHHHIIIIIH
NIAQARA RIVER IIIIIIUHUIIIIIIH mo
CHICAGO SSjSSSStM
DULUTH BSQBi"
8AULT STE. MARIE B9m
GRAND HWEM-MIMKEOON IS «i
US SENATE OOM SUBCOMMITTEE
tram data on US Oraat LahM water*
provided by US CoMt Quart, April 1M0
Exhibit D-S
* brtcnutional Joint Commlaiiaa. Jut 1M& "SpUbt the Homu-Macklac blirfict".
-------
EXHIBIT D-l
SURFACE WATER SPILLS
Great Lakes Basin Counties
Lake Superior
c
Lake Huron
0 Ontario
Lake
Michigan
OCCURRENCES
OF
SPILL
1
I
6-10
11-20
21 -40
it ERA
Source: Emergency Response Notification
System (ERNS), D.O.T.
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
EXHIBIT D-2
AIR RELEASES
Great Lakes Basin Counties
Lake Superior
Lake Huron
Lake Ontario
D=
Lake
¦:\\\ Michigan
OCCURRENCES
OF
RELEASES
6-10
11 -15
SzEFyV
16-20
Source: Emergency Response Notification us environmental protection agency
System (ERNS) D.O.T. GREAT LAKES NATI0NAL PROGRAM omcB
-------
EXHIBIT D-3
LAND SPILLS
Lake Superior
Lake Huron
Lake Ontario
Lake M
Michigan]
Great Lakes Basin Counties
OCCURRENCES
OF
SPILL
0
1 - 5
6-10
11 -15
16-20
>20
Source: Emergency Response Notification
System (ERNS), D.O.T.
jSzERAJ
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
EXHIBIT D-A
GROUND WATER SPILLS
Lake Superior
Lake Huron
Lake Ontarii
Lake >
jMichigan
Great Lakes Basin Counties
OCCURRENCES
OF
SPILL
Source: Emergency Response Notification
System (ERNS), D.O.T.
EPAtftec
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Page II - 28
are volatile to some extent Exhibit D-5 shows the frequency of spills in some Great Lakes haibors from 1980 to
1989.
Maritime Traffic on the Lakes presents a serious potential for spills through grounding, collision, leakage, and
unregulated emissions. Vessels carrying hazardous substances, however, account for a relatively small amount of
the overall cargo traffic on the Lakes. Other sources of spills to the Great Lakes include sewer systems, power
plants, and industry.
Severe environmental and health problems can result from such spills in the lake, where the relatively closed
freshwater ecosystem is fragile and highly sensitive to contamination. The Great Lakes supply millions of residents
with drinking water, and hundreds of drinking water supply intakes ire located in heavily navigated areas. Fisheries
and freshwater wetland bordering industrialized and trafficked regions of the lake risk exposure from spill incidents.
Some of the reported ofl and chemical spills into the lake have involved oil, acid, paint, phenol, styiene, sodium
hydroxide, vinyl chloride, xylene, and benzene. Near-shore facilities, as well as industries located along tributaries
in the Basin contribute to the potential treat of hazardous materials spills. The composite effects of near-shore spills
have a larger impact on the lake and longer-term ramifications, than do spills to open waters where there is more
opportunity for dispersion and cleansing of the system.
Transformer oil was also released, presenting a source of PCB pollution. Total industry-related accidental discharges
and spills may represent up over 30 percent of all spills. Of the 492 spills reported by the U.S. and Canadian Coast
Guards in 1987, 85% were oil spills, and the remainder toxics. In particular, the great variety of chemicals
manufactured in the Basin pose a threat of spills and make containment difficult0
In 1989, SI chemical spills that impacted ground water were reported in the ERNS data base of chemical spills in
counties within the Great Lakes Basin. Twenty of these spills occurred in Michigan, nine occurred in Illinois, seven
occurred in both New York and Ohio, five occurred in Indiana, and three occurred in Wisconsin. Exhibit D-6
presents the occurrences of spills in the Basin to different media, summed for the Basin counties.
Exhibit D-6
Occurrences for Medium of Spill
Air Land Water Ground Water
Basin Totals 358 741 677 51
Exhibit D-7 presents tons of bulk shipments of some of the potentially hazardous materials that travel on the
Great Lakes.66 This list represents the tonnage for materials transported annually on the St Clair River."
Each of these materials can have significant effects on various ecosystems and on drinking water supplies,
depending upon where and in what quantity they may be released. The exhibit depicts materials shipped in
"Region 2 summary of problem ana impacts an Lake Ontario.
"Hum malarial* art bdadod bacam* of lUr potential impact on Ik likw. M^Jer naofcazardoas materials nch as grata and
soybeans, coal and iron or*, art not inchuM in this list
"Tfce Canadian Coast Guard, "Counter Measures for Mark* Spills of Hazardous Materials." Told commo-da] Teasel
movement* on Ike St. Clair, aot in do ding car and nBcar terries, an sanaaOy between <500 and 5,000.
-------
Page II - 29
quantities of greater than 1000 tons per year. Other hazardous materials shipped on the St. Clair River, but
under 1000 tons per year, include: Acetate Monomer, Ammonium Sulfate, Calcium, Calcium Fluoride, Calcium
Sulfonate, Carbon Tetrachloride, Caustic Potash, Chloroform, Dichloromethane, Dowfax, Dowtherm, Dowvanol
EPH, Ethylene, Ethylhexanol, Formic Acid, Genklene A, Hexane, Isodecanol, Methacrylate Monomer, Methyl
Methacrylate, MMM (HQ 60), MMM (PMP SO), Monopropylene Glycol, Multanol-7249, Nonanol, Palm Oleine,
Perchloroethylene, Perklone A, Propionic Acid, Triklone NFS, Vinyl Acettte, and Zinc Bromide.
Exhibit D-7
Hazardous Material Movement on the Great Lakes: St Qair River
Material
Tons/Yr.
Number
^thousands')
Benzene
25-40
4-6
Brine
30-40
4-5
BIX Mix
20-30
4-6
C9-200
0-5
1-2
Calcium Chloride
60-80
8-12
Caustic Soda
150-200
20-25
Distillate
30-40
6-8
Ethyl Benzene
5-10
2-3
Gasoline
500-700
80-110
Isopropyl Alcohol
30-40
6-8
Jet Fuel
150-200
30-40
Light Oil
10% Benzene
25-35
7-9
Mixed Xylene
10-20
2-4
Naphtha
15-25
3-6
Nitrogenous Chemical
Fertilizers
30-40
4-5
Nonene
5-10
2-3
Ortho Xylene
10-20
2-4
Pentane
10-20
2-4
Phosphate Rock
20-30
2-3
Potash
900-1100
50-60
Propylene Tetramer
5-10
2-3
Rock Salt
700-900
40-50
Shell Sol 16
5-10
2-3
Shell Sol 2
0-5
1-2
Shell Sol 7
5-10
2-3
Styrene Monomer
80-100
8-14
Tolulene
70-90
10-13
Tolulene-Xylene Mix
20-30
4-6
Urea
50-75
5-7
Xylene
80-100
40-60
Other Chemicals
80-|00
40-60
Totals
3120-4115
360-497
-------
Page II - 30
The Canadian Report determined the annual probability of a spill of hazardous material per vessel shipment by
assessing the number of vessels carrying hazardous cargo in the Great Lakes for the years 1985 and 1986, the
frequency with which these vessels are involved in "incidents" that can breach containment, and the relative
ability of different vessel types to inhibit cargo releases. Based on these factors, the report determined the
relative annual probability in each of eleven areas in the Canadian Great Lakes. The aggregate probability that
any one vessel shipment in the Canadian Great Lakes will result in a hazardous material spill is slightly more
than 1 out of 10,000. If one assumes that similar characteristics apply to U.S. shipping on the Great Lakes, then
this spill probability may be applied to all Great Lakes shipping. The probabilities are:
LOCATION PROBABILITY EST. MOVEMENTS FREQUENCY PER
PER YEAR YEAR
Thunder Bay 8/100,000 2500 0.2
N. Shore Lake 47/100,000 90 <0.1
Superior
St Marys River 3/100,000 3750 0.1
L Huron North 13/100,000 700 0.1
Channel
L. Huron East Coast 163/100,000 2S0 0.4
St Clair River 1/10,000 5000 OS
Detroit River 4/100,000 7500 03
N. Shore Lake Erie 5/100,000 5000 03
Welland Canal 44/100,000 4500 2.0
N. Shore L. Ontario 17/100,000 3250 0.6
St. Lawrence - L. 42/100,000 3500 1.5
Ontario
Ecological Impacts
Many of the hazardous chemicals that are shipped on the Great Lakes are light and volatile, a combination
posing special threats to the Great Lakes ecosystem. These chemicals include carbon tetrachloride, toluene, and
gasoline. Once spilled, these toxic chemicals are likely to spread rapidly because of fast lake currents and
changing winds. While many of these chemicals evaporate quickly thereby lessening impact to the water
column, they may also be highly combustible. Furthermore, hazardous chemical spills which occur during the
winter months may slide underneath shore ice and make remediation more difficult Because these chemicals
may tend to spread more rapidly than oil and are, therefore, more difficult to contain, cleanup costs are often
higher for hazardous chemicals as compared to oil cleanups.
-------
Page II - 31
Nonetheless, most reported releases in the Great Lakes have involved oil. For example, 218 verified pollution
incidents in the Great Lakes were reported to the U.S. Coast Guard in 1989, and all but nine of those incidents
involved releases of oil. Although oil spills have occurred on the Lakes, very little is known about the
cumulative ecological impacts of these spills. Even less is understood about the consequences of chemicals or
oil that enter the aquatic environment in smaller amounts from ship sources other than tanker accidents. It is
very likely that more oil and chemicals enter the Lakes from non-spOl shipping incidents than from actual spill
events.
According to an official with the U.S. National Oceanic and Atmospheric Administration41, the greatest
potential for a disastrous spill, is along the connecting channels. The heavy flow of vessel traffic and extremely
fast currents in these channels (e.g., the St Lawrence River, St Clair-Detroit river system, and St Mary's River)
will further complicate any cleanup efforts in these areas.
Most information on potential ecological impacts in the Great Lakes and connecting channels is derived from
observations of past spill incidents. One of the best documented of these incidents occurred in 1976 in the St
Lawrence River. During the evening of June 23, the Tank Barge NEPCO spilled an estimated 1,167,000 liters of
oil into the St Lawrence River after grounding on Wellesley Island in the American Narrows Region.**
Following the spill a two-year study of environmental impacts in the River and Lake Ontario was conducted
under the sponsorship of the U.S. Environmental Protection Agency. The immediate impact of the spil] W|s
found to be extensive on fish and wildlife, with some species such as the great blue heron becoming endan»
and many other species becoming threatened. Although the spill occurred after the reproductive season for
species, the collection of the spill in bays and marshes impeded several speties' ability to raise their young
addition, human activities associated with the cleanup process also disturbed habitat in the marshes and bav
Overall, however, the high flow rate of the lake and river at the time of the spill may have increased flush
the oil and mitigated some of the potential adverse impacts in the bay and marsh habitats.10 °f
Long-term effects on resident species was difficult to quantify due to a lack of baseline data. Nonetheless j.
available data did suggest that losses went beyond the initial direct mortality of individuals, and that ren A
and survival was reduced for some species. For example, the number of golden and spottail shiners rem™
low two years after the spill, while the yellow perch population increased. This may indicate a higher de&T
tolerance to the oil for the perch. The number of fish species remained the same in the study area, with th °f
pumpkin seed being the dominant species. ' C
Waterfowl populations may also have been affected, with a decline in the success of breeding pairs to produce
broods in the heavily oiled areas. Pairs outside the impacted areas experienced an increase in breeding success
although this pattern may also have been influenced by high water levels during the study period. Finally, the '
study found no relationship between the abundance of oil and populations of muskrats, songbirds, or reptiles.
Overall, the study concluded as follows:
It is not known how long the oil residuals will remain in the marshes and influence fish and
wildlife populations. An oil spill catastrophe compares in its ecological effects with those of
"ReporUd hi the "Gnat Lilw Reporter," JaMury/Fabruty 1991, VoL t, Nuaber L Cater for Iki Gnat
" VS. EPA. 1979. "Damage Amamt Stadia* faHowiBg the NEPCO 140 Oil SpQ) on the St Lawreacc River. EPA-600T7-79-
256.
" Ibid, p. S.
-------
Page H - 32
natural catastrophes such as fire and flood. In each, the losses are great and the recovery is
slow. However, fish and wildlife can overcome these adversities. The question becomes one of
evaluating the loss of a resource to the using public and the local economy for a period of one
to several years.71
As apparent, the environmental, social and economic consequences of spills to a relatively enclosed fresh water
body can be devastating. Oil or chemical spills can destroy the microscopic organisms that anchor the food
chain, causing extensive damage to aquatic ecosystems. Spills can seriously reduce fish stocks, either directly or
by damaging spawning areas. Waterfowl are frequent casualties of spills, and are especially vulnerable to direct
contamination as well as to habitat/food source losses. In some cases on the Great Lakes entire populations of a
species could be put at risk by either the spill itself or the subsequent cleanup operation. Spills affecting
wetlands and ecologically sensitive areas can have severe long-term impacts. Tourism, and therefore local
economies, is known to be severely impacted by contaminated shorelines and drinking water supplies or by
reduced fish and waterfowl stocks. Major spills can therefore jeopardize the environmental health of all
members of the ecological community, in addition to threatening the health of all human residents in the Basin.
Preparedness
The need for well-planned emergency response preparedness in the Great Lakes has emerged as an area of
concern following the Exxon Valdez accident and other recent releases to U.S. coastal waters. In 1989, a special
Teview board convened by the Canadian federal Cabinet, concluded that "the capability to respond effectively to
a spill of any significant magnitude does not presently exist anywhere in Canada."72 In addition, the Great
Lakes Water Quality Board of the International Joint Commission concluded that unplanned and illicit releases
from vessels and from onshore and offshore facilities constitute a significant source of contamination to the
Great Lakes. Furthermore, the Water Quality Board encouraged the thorough training of personnel who operate
complex spill cleanup machinery.71
Despite these concerns, the United States has several mechanisms in place to respond to releases in the Great
Lakes Basin. On June 24,1974, a joint U.S./Canada Marine Pollution Contingency plan was adopted. This
agreement resulted in regular updates of the plan to ensure that a coordinated and integrated response to spills in
the Great Lakes would involve U.S. Federal, Canadian, State, and local agencies.
Continued preparedness of U.S. agencies is led by the U.S. Coast Guard, which has developed contingency plans
for addressing spills within the jurisdictions of each of its Great Lakes Marine Safety Offices. These
contingency plans inventory the facilities or other sources from which spills can occur, specify the names and
numbers of available cleanup contractors, and identify the locations of equipment needed to contain and cleanup
accidental releases. The ability to implement these contingency plans is further tested through joint spill
response exercises conducted in cooperation with the Canadian government Although the Coast Guard
acknowledges that impacts will occur as a result of future releases, they believe that they are adequately prepared
to respond to spills and thereby lessen the potential environmental damage.74
" Ibid, p.9.
"Ibid.
" WiUr Quality Board, International Joint Cnmmhrion 1989. "1969 lUpott on Gnat Lakti Water Quality," pp. 39-41.
14 Personal communication, Commands' P. Chubb, U.S. Coast Guard.
-------
Page H - 33
In addition to the above activities, die United States is continuing to develop new spill response capabilities. For
example, a spill response law was passed in August, 1990. Among other things, the law will phase in mandatory
double-hull requirements by the year 2015; establish special pilotage requirements for operation in U.S. waters;
and provide the Great Lakes region with Federal emergency response centers during the next few years. The
Canadian government is examining similar plans. State and local agencies and private organizations are also
working to increase response capabilities.
Prevention activities are also being initiated. The Council of Great Lakes Governors is studying several
preventive strategies. Additionally, the Coast Guard and EPA are increasing inspection activities, especially at
industrial sites in die Basin, in an effort to reduce the number of spills and accidental releases. Testimony before
a U.S. Senate subcommittee in 1990 indicated that as inspections increase, spills decrease.73
wHHd.
-------
Page II - 34
E. GLOBAL CLIMATE CHANGE
Problem Area Definition
Gases in the atmosphere allow sunlight to pass through the air and heat the Earth's surface. The Earth's surface
absorbs the sunlight and emits thermal radiation back to the atmosphere. Several gases in the atmosphere, often
referred to as greenhouse gases, absorb some of the outgoing thermal radiation and heat the atmosphere. The
atmosphere emits thermal radiation, both upward to outerspace and downward to the Earth's surface, further
wanning the surface. This is known as the "greenhouse effect"
Since the beginning of the Industrial Revolution, human activities have led to increased concentrations of
greenhouse gases in the atmosphere. Scientists have concluded that the increase in greenhouse gases will
eventually change global climate. In 1979, the National Academy of Sciences estimated that doubling carbon
dioxide concentrations over preindustrial levels would lead to an increase of 1.5 to 4.5 degrees Celsius (2 to 8
degrees Fahrenheit) in global air temperatures. In 1985, the World Meteorological Organization, the United
Nations Environment Programme, and the International Council of Scientific Unions reaffirmed these estimates.
Magnitude of the Problem
While the global climate is continuously changing as a result of natural processes, current greenhouse warming is
different from past climate changes. Not only will temperatures be higher than they have been in the last
125,000 years, but the rate of temperature change will be unprecedented. Past climate changes of comparable
magnitude have generally occurred over tens of thousands of years. The change in temperature resulting from
the greenhouse effect, however, is estimated to take less than a century.
Carbon dioxide, the most abundant greenhouse gas, is responsible for approximately half of the total
anthropogenically caused greenhouse effect Since the industrial revolution, the concentration of carbon dioxide
in the atmosphere has increased 25 percent and continues to increase at a rate of .4 percent each year. Fossil-
fuel combustion and deforestation are the primary sources of this increase in atmospheric carbon dioxide.
Methane in the atmosphere has more than doubled in the past 300 years and is currently responsible for about 18
percent of anthropogenically caused greenhouse effect Currently, total methane emissions are increasing at a
rate of 1 percent per year. Agricultural sources, particularly rice cultivation and livestock, seem to be the most
significant contributors to recent increases in methane concentrations. Other important sources of methane
emissions include landfills, coal seams, melting permafrost, natural gas exploration and pipeline leakage, and
biomass burning associated with deforestation.
Chlorofluotocarbons (CFCs) currently account for about 14 percent of the anthropogenically caused greenhouse
effect CFCs are used in refrigerants, aerosol propellants, foam-blowing agents, and solvents. In the "Montreal
Protocol" of 1987, international cooperation resulted in commitments by many nations to phase out their use of
CFCs. While efforts like these will probably result in a reduction of CFCs in the future, the total impact of
CFCs on the greenhouse effect will most likely increase for some time because of the long lifetime of these
gases.
Nitrous oxide has increased in concentration by 5 to 10 percent in the past 200 years and is currently increasing
at a rate of 0.25 percent per year. The cause of this increase is uncertain, but nitrogen-based fertilizers, land
clearing, biomass burning, and fossil fuel combustion are all contributors. Oceans are a significant natural source
-------
Page H - 35
of nitrous oxide. Including both natural and anthropogenic sources, nitrous oxide contributes about 6 percent to
the enhanced greenhouse effect
The contribution of ozone to global wanning was not estimated. However, it should be noted that both ozone
increases in the troposphere and lower stratosphere and ozone decreases in the upper stratosphere tend to warm
the Earth's surface.
It should also be noted that water vapor is an important natural greenhouse gas. When the climate warms, more
water will evaporate into the atmosphere from the wanned surface and thus enhance the greenhouse effect This
in turn, will result in the production of still more water vapor through evaporation.
Human Health Risk
Human illness and mortality are linked to weather patterns. A variety of human illnesses show sensitivity to
changes in temperature or humidity which accompany changes in aeason. Stroke and heart attacks, for example
increase with very cold or very warm weather and allergic disease such as asthma and hay fever incr<^Sf ^
spring and summer when pollens are released. Mortality rates, particularly for die elderly and very ill, lrc
influenced by the frequency and severity of extreme temperatures.
Indirectly, the incidence or severity of respiratory disease such as emphysema and asthma are likely to inert*
due to incre*«es in air pollution which are frequently associated with climate change. Increases in ozone SC
concentrations, in the lower atmosphere in particular, are associated with increasing temperature. Higher
temperatures will speed reaction rates among chemicals in the atmosphere, causing higher concentrations of
ozone. Also, longer summers will result in longer ozone seasons and thus a greater potential for ozone
problems.
Ecological Risk Assessment
Toxicity Assessment
If current trends continue, the rate of climate change could be much quicker than rates of natural migration and
adaptation. Gimatc zones may shift hundreds of miles northward, and animals and especially plants, may have
difficulty migrating northward. The presence of uiban areas, agricultural lands, and roads would restrict habitats
and block migratory pathways. Inhabited ranges and populations of many species are likely to decrease, and in
many cases species may become extinct The effects could last for centuries and would be virtually irreversible
Climate change may significantly alter forest composition and reduce the land area of healthy forests. Higher
temperatures may lead to drier soils in many parts of the United States. Consequently, trees that need wetter
soils may die. Studies of the potential effects of climate change on forests predict northward shifts in ranges and
significant changes in composition, although specific results vary depending on sites and scenarios used.
Higher temperatures may lead to more aquatic growth, such as algal blooms, and decreased mixing of 1»Vnr
This would deplete oxygen levels in shallow areas of the Great Lakes and make diem less habitable for fish.
Fish in small lakes and streams may be unable to escape temperatures beyond their tolerances, or their habitats
may simply disappear.
In many regions of the Basin, climate change alone could reduce dryland yields of corn, wheat, and soybeans
with site-to-site losses ranging from negligible amounts to 80 percent These decreases would be primarily the
-------
Page II - 36
result of higher temperatures, which would shorten a crop's life cycle. Even under the more extreme climate
change scenarios, the production capacity of U.S. agriculture was estimated to meet domestic needs. However, a
decline in crop production of this magnitude would reduce exports, which could significantly impact food-
importing nations.
Higher temperatures will speed reaction rates among chemicals in the atmosphere, causing higher concentrations
of ozone in many urban areas. In addition, the length of the summer season would be increased, which implies
an increase in the ozone season. Although the impact of higher temperatures on acid rain have not been
analyzed, it is likely that sulfur and nitrogen would oxidize more rapidly under higher temperatures.
Exposure Assessment
Due to the nature of global wanning, the entire Basin may be affected by climate change, with the agricultural
areas in the southern portion of the Basin experiencing crop yield declines and the uiban areas experiencing
increased ozone production.
Ecological Risk Characterization
In "The Potential Effects of Global Gimate Change on the United States," the effects of global warming on
various regions of the country were studied. In the Basin, impacts on three ecosystems were studied: lakes and
aquatic ecosystems, forests and agricultural areas. All scenarios show dramatic increases in temperature for the
Basin. The combination of significant increases in temperature and relatively small increases in precipitation
make the Geophysical Fluid Dynamics Laboratory model the most severe scenario of the three.
Lakes and Aquatic Ecosystems
Global climate change could affect the Great Lakes by lowering lake levels, reducing ice cover, and degrading
water quality in rivers and shallow areas of the lakes. It is estimated that higher temperatures may cause lake
levels to fall by 0.5 to 25 meters, a decrease within the normal range of long-term fluctuations. Even if
precipitation increases, lake levels would continue to fall because higher temperatures would reduce the
snowpack and accelerate evaporation. It should be noted that estimates of lake level drop are sensitive to
assumptions about evaporation and that under certain limited conditions, lake levels could rise.
Wanner winters are expected to reduce ice cover on the Great Lakes, with ice formation generally limited to
near shore and shallow areas. In addition, the duration of ice cover on the lakes would be reduced by 1 to 3
months. While less ice could extend the Basin's shipping season, this reduction in ice cover could have negative
impacts because the ice protects some aquatic life, such as whitefish, and protects shorelines against the erosive
impact of high-energy waves.
Higher temperatures may lead to more aquatic growth, such as algal blooms, and increase stratification. The
increased turnover time of lake waters could disrupt mixing of oxygen and nutrients, possibly affecting the
abundance of life in the lower and upper layers of the lakes. In addition, fish in small lakes and streams may be
unable to escape temperatures beyond their tolerances, or their habitats may simply disappear. Shifts in
community structures toward warm water fisheries would occur and be accompanied by a decrease in diversity.
-------
Page H - 37
Forests
The composition and abundance of forests could change significantly. Higher temperatures and lower soil
moisture could reduce forest biomass in dry sites in central Michigan by about 80 to 100 percent These
hardwood and oak forest could become oak savannas or grasslands. In northern areas like Minnesota, boreal and
cedar bog forests could become treeless bogs and mixed northern hardwoods. It is anticipated that productivity
may decrease on dry sites and bogland sites, but may increase on some well-drained wet sites. Softwood species
may be eliminated and replaced by hardwoods, such as oak and maple. It is uncertain whether forests in the
southern part of the Basin will die bade leaving grasslands or whether new species will be able to migrate. Such
forest could also be transplanted in these areas. In addition, the rate of forest migration, where it is still possible
is likely to be slower than the climate change. Many tracts of forest are cut off by geography or by human
development Consequently, the total range of many species would be reduced.
Agriculture
Studies indicate that temperature and precipitation changes could reduce crop yields throughout the Basin, with
the exception of the northernmost latitudes where yields could increase depending on rainfall availability. The
reduction in yields in the southern portion of the Basin would primarily result from the shortened growing sea so
caused by extreme summer heat Production in the north would increase largely due to a longer frost-Crec season
which would result in increased yields.
Cora yields throughout most of the Basin could decrease from 3 to 60 percent depending on climate and Wate
regime (dryland or irrigated). It should be noted that yields in areas near Duluth, for example, may increase
much as 49 to 86 percent While current com yields are lower in this northern area than in more southern sit *
increases in yield of this magnitude could result in a dryland yield equal to other sites and an irrigated yje]d
exceeding other locations.
Dryland soybean yields are expected to decrease by 3 to 65 percent throughout all but the northernmost part
the Basin. Irrigated yields in the north are estimated to increase by about 100 to 140 percent Even with ne
increases in yield of this magnitude for northern areas, northern lands may still have yields lower than in area***1*
the south. s *°
-------
Page II • 38
F. CHANGING LAKE LEVELS
Problem Area Description
The changing lake levels problem area addresses the risks to humans and the environment from fluctuations in
lake water levels of the Great Lakes Basin. The Great Lakes Basin consists of an area approximately of 297,000
square miles1*, A profile of the Basin is shown in Exhibit F-l.77 An estimated 5,400 cubic miles of water
held in the system71. Maximum water depths range from 21 feet in Lake St Clair to over 1,300 feet in Lake
Superior"'*.
Fluctuations in water levels are considered a natural occurrence; they have occurred since the Basin's formation,
and will continue. It is difficult to generalize about the magnitude of the problem lake level fluctuations present
because the Basin covers such a large area. Also, what affects one lake may be of an entirely different scale
from the effect on another lake. In addition, what may be considered an adverse effect for some shoreline
interests, may not be considered a problem for others (i.e. hydroelectric facilities prefer high water levels and
beach front property owners often prefer lower levels). However, it is typically the extreme events of record
high and low lakes levels (i.e. flooding and drought) that generate the most concern. These extreme events ate
particularly difficult to deal with because they usually take people by surprise.
EXHIBIT F-l
St. Marys River
Lakes
R'Ver^\" Huron"*"] r61' ^ R,w
IQO.Ott. AITHII isTl.TM.
Lake
or I
923ft.
Lake
St. Clair
Detroit
River
750ft.
Michigan Laka Huron
210M.-J
Niagara Rlvar -J
Laka St. Lawrence
Niagara Falla r-Lake St Francis
Lake Ontario
Lake St. Loula
Montreal Harbour
Gulf of
St. Lawrence
Lawrence River
-802ft.
Atlantic
Ocean
35 29 33
379 , 60. 223 . 99 ¦ 236 ¦ } .1 SO. 77 .J.52. ^ .
Q
950
Distance In Miles
2,200
PROFILE OF THE GREAT LAKES - ST. LAWRENCE RIVER SYSTEM
M International Joint CwwnMiw, 1999. "Living with the Lake*: Cfcallaagw ud OpportaaMw" p.14.
" Exhibits F-I through F-5 arc taken frw "Living with the Lakaa." op. dt
"Ibid.
"Ibid.
-------
Page n - 39
Tbe Great Lakes are of such large size and their individual outlets so limited, that when extremely high or low
lake levels occur, they will persist for a considerable length of time, even after those factors which caused tbe
extremes have changed. However, predicting future water levels is extremely difficult because lake level
measurements began only in 1860. The intervening 130 years is considered too short a period of record from a
hydrologic and geologic standpoint Accurate information would be required for many of the factors that affect
lake levels.
Factors influencing lake levels
A variety of natural factors affect Great Lakes water levels including precipitation, runoff, temperature,
evapotranspiration, meteorological events, crustal movement, and flooding and erosion". Although all these
factors have some significance in terms of lake level changes, precipitation is tbe main cause of long-term wate
level fluctuations on the Great Lakes.
Tbe average annual precipitation for the Great Lakes Basin ranges from approximately 25 inches in northern
Minnesota, to about 50 inches in New York." Over tbe course of the century two distinct precipitation pattern
have been determined for tbe region. From 1900-1939 the Great Lakes experienced below average precipjtar*
However, during the period of 1940-1984 the Great Lakes received above average precipitation. On averaz IO,X"
Basin received 2 additional inches of precipitation per year representing a 6% increase overall. Exhibit F-2 ***'
indicates tbe degree of variation in precipitation for tbe Great Lakes Basin as a whole. Precipitation usual]
leads lake levels by about one year.*2
EXHIBIT F-2
u
i*
¦t.
as
09
•1J
VEM
GREAT LAKES BASIN ANNUAL PRECIPITATION
(1965 - 1868)
" The Great Lakes; An Environmental Atlas, op. dL
"Ibid
-------
Page 11-40
Other factors that influence lake levels include dredging for navigation improvements, regulation, and control
structures. In addition, the actual amount of water withdrawn for consumptive uses and diversions from the
Basin impact water levels. Water withdrawn for consumptive use is that amount assumed to be lost or otherwise
not returned to the Great Lakes Basin due to evaporation, incorporation into products or other processes. Inter-
basin diversions refer to tbe amount of water transferred from the Great Lakes Basin into another watershed.
Artificial modifications that attempt to regulate water levels do so only to a small degree. It is the natural
factors, particularly climatic changes, that influence lake levels tbe most.
Lake level fluctuations
Water levels in each of tbe lakes will rise or fall according to the amount of water entering or leaving the lake.
Great Lakes lake levels fluctuate on a very short-term basis due to strong winds and storms. On a seasonal basis
and on a longer-term year-to-year basis, lake level alterations are due to changes in the balance between
precipitation, inflow, and runoff, on one hand, and evaporation and drainage, on the other.
Short-term changes can be dramatic and can cause considerable amounts of damage along exposed shoreline
areas. An example would be tbe storm of December 2, 1985 that caused a 16 foot difference in water levels on
Lake Erie between Toledo (eight feet lower) and Buffalo (eight feet higher) (Exhibit F-3). Although this is an
extreme example, all the lakes are affected by harsh weather, the severity of which is dependent on the size and
depth of the lake, and tbe orientation and shape of the lake as well as the magnitude of the storm0.
EXHIBIT F-3
581
579
BUFF A LOj
577
575
&
£ 573
z"
9 571
569
Ui
567
TOll .Off
565
11:00 PM
DECEMBER J, 1»*»
12:00 PM
DECEMBER 1, 1ISS
12:00 PU
DECEMBER 1, 1*0S
WIND TIDE AND SEICHE EFFECT ON LAKE ERIE
DURING STORM OF DECEMBER 2, 1985
"loUnutloatl Joint Coaimiuion. 1989. "Lirin| wilk Ik* Ltluw Ch«Uc
-------
Page II - 41
Seasonal fluctuations occur over the course of a year and account for average changes of about 12 to 18 inches
with lows normally occurring in January or February and highs in June through September*4 (Exhibit F-4).
Long-term fluctuations in lake levels as shown in Exhibit F-5 during the period of 1950 - 1988, are on the order
of five to six feet from record lows to highs. Significant changes in lake levels will only occur as a result of
significant climatic changes, such as a period of higher than normal precipitation. For example, Great Lakes
water level records indicate that the record high lake levels set in 1985 and 1986 were the result of a 5-year
period of higher than normal precipitation19.
It is the short-term changes, similar to those found in agricultural runoff, that cause the most damage because of
their severity and usually unanticipated occurrence. The seasonal and long-term fluctuations do impact certain
interests within the Great Lakes Basin, however, these impacts are not as severe and are usually expected. For
the most part, higher water levels (i.e. levels above the average normal, up to the point of flooding) are preferred
over low water levels.
Ecological Impacts
Changes in Great Lakes water levels have a variety of ecological impacts, including impacts associated with
wetlands and other water resources, which in turn affect aquatic plant and animal life, terrestrial plant and animal
life and humans.
Wetlands of the Great Lakes are exposed to variations in water levels caused by long-term climatic cycles, Sh
term climatic occurrences, and annual distribution of water, and wave actions. The extent and quality of
wetlands found along the shores of the Great Lakes during any time period, are essentially the product of
seasonal and long-term environmental conditions. Wetland communities react to the pattern and magnitude of
water-level variations according to the tolerance of the biotic community to them. Also, it has been determined
that the highest quality wetlands are those that experience water level fluctuations.
Wetland vegetation is dependent on the ground water table that is on or close to the surface. During periods of
reduced lake levels, the ground-water table may be lowered, thus changing the wetland. Some wetland
vegetation will change their growth form to accommodate drier conditions, but the overall vegetation will usuali
change dramatically as species intolerant of drying die and are replaced by species emerging from reserves of
buried seed16.
The optimal wetland wildlife habitat has been described as a semi-marsh, i.e. 50 percent open water and 50
percent wetland vegetation*7. High water conditions (Le., levels above the historical long-term mean) produce
habitat approaching the semi-marsh which benefit wildlife such as waterfowl, muskrats, black terns and
herons'*. These conditions increase wildlife species diveisity. High water may facilitate the interchange
between the lake and wetland, and thus permit fish spawning (i.e., northern pike). Low water conditions (i.e.
water levels below the historical long-term mean) encourage the predominance of the sedge-meadow and d^nse
emergent zones. During extended low water years, wildlife species diveisity decreases with habitat conditions
"Gnat Lakes Coamiuioa. 1990. "A Guidebook to Droaght Planning, Management and WaUr Level Change* in the Great
Lake*" p.49.
"Ibid.
"Ibid.
"Ibi«L,p.B-9S.
"Ibid.
-------
EXHIBIT F-4
in
in
o>
v—
Q
_J
a
ui
UJ
601
600-
579-
578-
577-
576-
575-
574-
573-
572-
571-
570-
569H
246-
2AS-.
244-
_n|^_
1/
^i\
1*
Wh II -&¦
ake Vlichi
La te
ak
La
; Sup
jai-Ij
St.
ke
Ei
en
air
le
tano
or
ur
on
JAN FEB IIAR APR HAY JUN JUL AUG SEP OCT NOV DEC
MONTH
AVERAGE •
monthly
MEAN
LONQ. —
TERM
AVERAGE
SEASONAL FLUCTUATIONS
GREAT LAKES WATER LEVELS
(1900-1988)
EXHIBIT F-5
FEET
(IOLD 1I6S)
578-
576-
674-
572-
570-
PRE-REC
ULATION
YEAR
1M0
i»as
LAKE
SUPERIOR
LAKES
MICHIGAN-
HURON
LAKE
ST. CLAIR
LAKE
ERIE
LAKE
ONTARIO
ANNUAL AVERAGE GREAT LAKES WATER LEVELS
(1950-1988)
-------
Page II - 42
favoring red-winged blackbirds, short-billed marsh wrens, rails, white-tailed deer, cottontail rabbits and small
rodents16. When a wetland area is reduced, many important functions and values, such as ground-water
discharge and recharge; absorption of toxic pollutants and nutrients entering the lakes; habitat for fish, waterfowl
and other wildlife; and recreation such as hunting, fishing, and canoeing, are reduced also.
The aquatic environment of the lakes is less influenced by fluctuating lake levels than are wetlands and
shorelines. A rich variety of aquatic life is found in the waters of the Great Lakes. Water temperature, levels of
oxygen, the quantity of nutrients available for food, the amount of sunlight penetrating the shallower depths, and
the amount of contaminants in the water or sediments determine the species present and their relative
abundance*9.
High lake levels tend to be beneficial to aquatic habitat and water quality because low levels result in raised
water temperatures, reduced dissolved oxygen, increased concentrations of pollutants, particularly sewage
effluent, and increased need for dredging of contaminated sediments. Nevertheless, water quality degradation
can result from flooding of septic systems,reduced treatment plant efficiency and submergence of shoreline
vegetation and nutrient-rich soil90. However, the impacts of extreme levels on the resuspension of pollutants
and on the volume of discharge from sewage treatment plants and septic systems will require future study in
order to better establish the relationship between water quantity and water quality*.
Low lake levels also reduce the amount of "edge" habitat for fish, particularly fish spawning areas, and other
aquatic organisms. High flows, however, move larval fish and other small organisms more rapidly through th
system, improving prospects for growth and survival.
Reductions in lake water levels affect other water resources within the Basin such as rivers, smaller lakes
reservoirs, and ground water by decreasing the availability of these resources for use. The most limited of th
resources are affected first by increased demand for water, and decreased runoff into streams and lakes that ' ^
turn recharge ground-water supplies. Continued low water levels commonly result in water use conflicts, XV* n
interference complaints typically arise when a user withdraws large quantities of water from a high-capacity
and thereby restricts the withdrawal capability of a domestic user's smaller well. WcU
Welfare Impacts
Changes in lake levels can have major implications for a variety of interests. Typically, it is the extreme eve
of flooding and drought that cause the most severe problems, and it is these extreme events that tend to persist
overtime due to the size of the lakes and their limited individual outlets. This section will address the impacts
low and high lake levels on humans and the environment
Environmental costs
The environmental costs associated with changing lake levels include the economic losses due to the loss of
wetlands (see Physical Degradation of Wetlands and Aquatic Ecosystems). Other associated environmental costs
due to changing lake levels include the economic losses of fish, wildlife, recreation, and other aesthetic values
and the costs to developed areas of maintaining drainage and erosion control.
•ibid.
"InUrnatloosl Joint Commission. 1989. "Living with the Lakes: Challenges and Opportunities" p.29.
"IbiiL,p.30.
-------
Page D • 43
Materials and property damage
Navigation on the Great Lakes and in particular its harbors and connecting channels, is significantly affected
during periods of extended low water levels and flows. During low water periods, cargo volume is reduced and
trip frequency is adjusted. These changes compromise efficiency by increasing transportation costs and
decreasing revenue*1. Also, recreational harbors pose navigation problems during low water periods, with the
impacts of drought compounded by substantial sediment loads from nearby shorelands and riven.
Channel and harbor dredging may be needed during low water periods. However, dredging of commercial
harbors, canals, and navigation channels can remove a large volume of polluted sediments, though a small
percentage may be resuspended. Disposal of this material can be difficult and expensive if it does not meet state
and provincial environmental requirements for open water disposal. Qn-land or in-water confined disposal
areas/facilities are difficult to establish and access (particularly under emergency conditions) and pose their own
set of environmental concerns.
Hydroelectric facilities are the greatest users of Great Lakes water and represent more than 94% of the total
amount of water used per day. Hydroelectric withdrawals are, however, considered an in-stream use, not a
withdrawal in the sense that water is lost A major concern of power companies during periods of low water
levels is meeting environmental standards for the discharge of cooling water from thermal generating facilities
into rivers or lakes. Water temperatures tend to rise during low water levels, reducing dissolved oxygen levels
which can be hazardous to fish and other aquatic animals. Power company discharges are typically limited
during drought conditions, resulting in the power company having to reduce its energy production, thereby
limiting revenues and possibly disrupting electricity supply. Also, hydroelectric power generation may be
reduced due to decreased streamflows. Usually, the costs incurred on the utilities of having to purchase higher
priced electricity from other suppliers are passed onto the consumer.
Commercial and industrial water users are also impacted by changing lake levels. Major industries located along
the shores of the Great Lakes use the lakes for both water supply and waste disposal. As with all structures and
facilities along the shores, periodic property damage occurs due to storms and flooding. However, in general
higher water levels are preferable, up to the point of flooding, for water supply, greater dilution of waste
discharges, access to water for boats, and clearances for commercial navigation deliveries to industrial users.
Persisting low water levels can result in difficult access to high quality water if near-shore water quality is
reduced due to loweT lake levels. In addition, changes in water temperatures due to low levels may affect
industrial cooling processes. Emergency or alternative water supplies may be needed to accommodate declining
lake levels, reduced streamflows, or lowered ground water levels. Most commercial and industrial businesses
accept the fluctuating water levels as a part of the cost of doing business and consider location on the shore to be
a greater advantage than the disadvantage of changing water levels.
Many public water utilities obtain their water directly from the lakes. Even though intake systems are generally
located in deep water, low lake levels can be an issue for public water suppliers who may be forced to do
additional treatment if the concentration of pollutants has increased due to reductions in water levels. In
addition, when an extended period of reduced precipitation occurs, lake levels are reduced, as well as surface
water flows, ground-water flows, and reservoir levels. Those public water suppliers who are not using Great
Lakes water as their primary source, may need emergency or alternative water supplies which can be obtained by
deepening existing wells, drilling new wells, building temporary dams, changing supply from groundwater to
"IbM.
-------
Page n - 44
surface water, or extending pipelines to deeper water or to different sources. Conflicts between competing users
(particularly domestic users and irrigators) are frequent
The degree of risk to shoreline property depends an the location. The most serious impacts on riparian property
are due to erosion and flooding which occur during severe storms. The impacts could result in loss of land tad
trees, and damages to shore protection structures and buildings and their contents". Economic impacts h^lwde
the cost of alternate accommodations, costs of maintaining septic systems and costs of repairing or replacing
damaged property. At this time there is no data available which indicates costs incurred by property owners due
to damages from high lake levels. However, a large census and survey is underway in order to gain a better
understanding of the magnitude of these impacts on shoreline properties94. In general, shoreline property will
benefit from lower lake levels. Shore erosion is reduced and low lake levels may expose enough beach to enihl
on-shore winds to form dunes which can protect shoreline property during storms. Finally, beaches on the lake*6
will become wider, accommodating recreationists.
Relatively little of the Basin's extensive agricultural land is on die shore of the lakes. However, persistent lent,
water levels do impact those areas as well. Low soil moisture may result in loss of arable land and reduced
productivity, causing farm revenues to drop. In addition, if feed crop productivity is affected, feed shortages CT°®>
result in problems for dairy farmers, cattle ranchers, and other livestock operations. Furthermore, shallow weH**
ponds, and wetlands may dry up forcing fanners and rural residents to drill new wells, deepen old wells or *'
transport water from other sources. Many farmers must increase their irrigation activities during these times
thereby reducing stream flows, reservoirs and ground water levels. Conflicts between irrigators and domes*'
water users are common. In addition, as topsoll loses moisture, it becomes more susceptible to wind eiogi
Topsofl erosion can reduce agricultural productivity by as much as 30% - 40%*. There are some farmers ^
however, who benefit from low water levels in the Great Lakes Basin, particularly those in less affected re *
They may experience substantial increases in total income due to higher unit prices. For those agricultural^0114*
located close to the water, particularly on former floodplain or wetland areas, high lake levels make those 1 "*!***
vulnerable to flooding and erosion.
Many of the impacts already discussed contribute to the overall impact of changing lake levels cm residents '
general. During periods of low lake levels homeowners may experience reductions in available water sunni10
high costs for electrical power, high food prices due to reduced crop yields, and damage to landscaping
periods of high lake levels, particularly flooding, property damage is a major concern. na®
In sum, water level changes can affect a variety of interests including property owners, water-use industries,
recreational and commercial users of die lakes, those whose source of drinking water is the lakes or ground
water fed sources, and agriculturalists. Changes in the lake levels also affect non-human populations includino
terrestrial, shoreline, and aquatic plants and animals, particularly those associated with wetlands.
The issue of changing lake levels is a low priority for those potentially affected. Because these changes are
natural and have been occurring since the Great lakes were formed, people are accustomed to them, expect them
and have adjusted accordingly. In addition, plant and animal life have adapted as well. For many, "'•-rginr ul
levels are more an inconvenience than a risk. However, the extreme events of flooding and drought are a majo*
concern throughout the Basin and have the greatest potential catastrophic impact
"Cr*t 1 »\m CmubMm. 1990. "A Guidebook to Dmtbt FUaaiq, Mwfwut ui Witar Lrrtl rhMign ia tk« Gnat
Lakw" *15.
"Ultra*liana! Jotat n i—<— 1989. "Uriag with (be Laluu Cbalkaf** and Opportaaitim" p.42.
"lbM^p.43.
-------
Page in - 1
ID. PRIMARY WATER PROBLEM AREAS
The water quality of the Great Lakes and their connecting channels and tributaries are of great importance to the
health of the Basin. In Section m, we discuss a number of primary water problem areas (generally defined by
pollutant source) related to the Basin's water quality. The following problem areas are discussed in this section:
¦ Industrial point source sicharges to surface waters;
¦ Municipal point source discharges to surface waters;
¦ Nonpoint surface loadings to surface waters;
¦ Atmospheric loadings to surface waters;
¦ Toxic sediments; and
¦ Pesticide discharges.
These problem areas all contribute to the load of pollutants that degrade water quality and increase the body burden
of contaminants in fish and other aquatic organisms.
G. INDUSTRIAL POINT SOURCE DISCHARGES TO SURFACE WATER
Problem Area Description
This problem area addresses contaminant load and surface water quality impairment resulting from industrial point
source discharges. This problem area does not include point source discbaiges from publicly and privately owned
municipal wastewater discharges. Typical sources of industrial point source discharge permitted under the National
Pollutant Discharge Elimination System (NPDES) include electric utilities, metal finishing, pulp and paper processing,
and iron and steel production. Within the Great Lakes Basin, there are 40 utilities classified as major dischargers.
While electric utility cooling water discharges principally contribute to thermal load in receiving waters, other
pollutants discharged by industrial sources include total suspended solids, biochemical oxygen demand (BOD), toxic
oiganics such as pesticides and phenols, and toxic inorganics such as heavy metals.
Most health and ecological risks that result from surface water impacts, such as toxics bioconcentration in sport fish,
cannot be differentiated from among industrial, municipal, and nonpoint source discharges (including atmospheric
loadings and toxic sediments). As a result, the aggregate impacts of these problem sources are discussed only once,
in this section, but are nevertheless applicable to the following surface water problem areas as well.
Magnitude of the problem
There are approximately 570 major (i.e., facilities discharging over one million gallons per day or those identified
as having a potential to adversely impact receiving waters) and 2,750 minor industrial dischargers in the Basin.1
The distribution of major and minor point source dischargers by lake watershed is presented in Exhibit G-l. Exhibit
G-2 further identifies the major and minor dischaigers by EPA Region. Exhibit G-3 identifies the major dischaigers
by facility type (industrial and municipal) within the Basins. Exhibit G-4 distinguishes the Basin industrial and
municipal dischargers by the EPA Region in which they are found. Finally, the location of all major dischargers
within the Basin are presented in Exhibit G-5.
1 1991 Permit CaaipUaocc System (PCS) data.
-------
Page III - 2
MAJOR AND MINOR DISCHARGERS
IN THE GREAT LAKES BASIN
(NUMBER OF DISCHARGERS IN PCS)
LAKE BASIN
Exhibit O-l
MAJOR AND MINOR DISCHARGERS
IN THE GREAT LAKES BASIN
daily flow from major electric utility dischargers is approximately 20 billion gallons per day
itely 4 bgd from other major industrial dischargers. The majority of the discharges from electric i *nd
are "non-consumptive" and involve the pass through of cooling water. The volume of discbarge from the othcr''''^^
sources is largely attributed to the iron and steel industry, followed by the pulp and paper, nonferrous m r, mai°r
chemical manufacturing industries. The human health risk assessment analysis presented below focuses on th
- • — t t- ji « ^CSC non.
In general, industrial and municipal facilities have made significant progress in reducing their loadings of
priority pollutants to the Great Lakes. In the Niagara River, organics, metals, cyanides, and phenols have be^
n
FACILITY TYPES FOR MAJOR DISCHARGERS
IN EACH OF THE GREAT LAKES BASINS
LAKE BASIN
MAJOfl OOCHARQERB TO THE OREA'TlAKES
Exhibit (j-J
FACILITY TYPES FOR NPDES MAJOR
DISCHARGERS IN THE GREAT LAKES BASIN
<¦
MAJOR DISCHARGERS TO THE GREAT LAKES
-------
Exhibit G-5
Lake Superior
FACILITIES
PER COUNTY
9 - 16
GREAT LAKES STATES
MAJOR NPDES
DISCHARGERS
:i::
Lake trie
IwrriD HATH
tNYIIONMIMTAl riOKCTtON ACWCT
fillAT UIM NATIONAL MOCIAU OfMCf
-------
Psge m - 3
reduced, on average, at least 80%.2 The current loadings from industrial facilities in the Niagara River Basin total
around 14 lbs/day for organics, 104 lbs/day for metals, 2 lbs/day for cyanides, and 14 lbs/day for phenols.
Human Health Impacts
. l. a inoiot risks from industrial point source discharges to the Great Lakes are addressed
As stated above, health and ecological ns nonpoim source discbarges, atmospheric loadings, and toxic
together with those of municipal point ' cxposure routes: (1) consumption of fish containing bio-
sediments. This risk cha^cterizaUon focus« ^ drinking water exposure. The risks expressed do not
concentrated toxic constituents, and (2) ® , municipal, or nonpoint source. Therefore, this
differentiate among the source of under this section. At the end of this problem area discussion, a
characterization discusses these risks w«c»^ buman facaUb and environmental impacts from Industrial
preliminary attempt is made to weign ^ d contribution to total contaminant load in the Basin.
Jotot sources .g.iwt other so.ro* by H.M
, {n rTTMt T „i^ fish, health risks related from Basin-wide exposure to Great
Although other contaminants are P*"* primarily by PCB exposures.1 With regard to Great Lakes fish
Lakes fisheries and ambient water qua ity exposures account for 85 percent of cancer risks. Therefore,
consumption, chemical risk analyses
the risk calculations concentrate on PCB expos
. .. Tins have , more localized effect, creating problems of fish contamination
Other contaminants, such as mirex, ^ , ^tinc problem only in Lake Ontario/
in specific locales. For example, mirex levels
Toxicity assessment
Cancer Risks
To assess the human health impacts from these sources, we used environmental concentration data (water and r
standard exposure and toxicity assumptions, and estimates of exposed populations to estimate overall human h ^'
risks from toxics in the Lakes and then apportioned these risks by contaminant source. Water concentrati e*1^1
were available foT alpha-BHC, lindane, dieldrin, pp-DDE (a metabolite of DDT), PCBs, and HCB°n p*'*
concentration data were available for dieldrin, DDT/DDE, PCBs, and mercury.5 Consequently, the human h ^
impact estimates are based on risks from only these contaminants®. ' eaith
1 RctiM n saauaary of problem ami rriatiag to the Lake Ontario Basin.
' See DeVaalt, D£„ et aL CootaaOaant trade in take feroat (SahrtUwu aamaycaafa) tram the apper Great Lakes. Ank. gnrij
Taxied. 15, pp. 349-356,1986; DeVanh, DS. Contaminant* hi Fish froaa Great Uka Harbon aad Tributary Mouth*. Arek.
Emtio*. CoHtami*. Taxied. 14, pp. 587-594, 1985.
4 DeVaalt, D£n J. Miltoo Clark, aad Garet Labels. CoataaUaaats sad Trad* in Fall Rn Cobo Sahoa. J. Gnat '-±rr
14(1), 1918. p. 29.
' Diektria aad DDT went saspeaded aad ceaceiled ia the eariy 1970*, aad arc ao ioager asad or auaafactarod in the
States Gnat Lakes Bad*. How«v«r, IJC daU oa water fuK; aad Bak samples iadicate tkat aadi pesticides, while ilimalu ^
coaceatratiaa, dill exist ia the Baria. See 1989 Report oa Great lakae Water Quit}, DC, Water Qaahty Board, 1989. Ukdy
Marcel of these pesticide* ia sarface waters asay therefore be atmospkcric depositioa aad I ¦iiniiMilua of toxic lodhaeat*. (Sea tk^
problem area dlsrwrioas far farther treetmeat) Howerer, soarcas ere aot dearly aadentood. Am h ao NPDES BMailoriag af
tkeee rwitemlaonts, Residual discharges from fonur auaafsctiirers or die nmoff auy also be soarcas of thee* coataadmaats.
' Pesticides risks to human health from fish conswnptioa are addressed here. For ecological effects of pesticides, see pi nlilsaa
area L. Far other health effects af pesticides, see problem area X.
-------
Page III - 4
Toxicologic^ data on the chemicals listed above were obtained from two EPA data sources: the Integrated Risk
Information System (IRIS) and the Health Effects Assessment Summary Tables (HEAST). Exhibit G-6 presents
information relating cancer and noncancer health effects with exposure levels for this list of constituents. Among
the constituents for which data are available, six of the seven are carcinogens (all are Group B2 carcinogens) and
four of the seven are noncarcinogens. Mercury is listed as a Group D carcinogen which means that it is not
classifiable in terms of human carcinogenicity.
The cancer slope factors and the chronic reference doses listed in Exhibit G-6 were used in conjunction with
estimates of exposure levels to calculate risks posed by toxics in the Great Lakes.
Exhibit G-6
Toxicological Information for Contaminants
Discharged to Surface Waters
Contaminant
Carcinogenicity
Classification
Cancer Potency Slope
Factor
(mg/kg/day)'1
Chronic Oral Reference
Dose
(mg/kg/day)
alpha-BHC
B2
63
na*
Lindane
B2
1.3
0.0003
Dieldrin
B2
16.0
0.00005
DDE/DDT
B2
0.3
0.0005 (DDT)
PCBs
B2
7.7
0.0001/0.000017
HCB
B2
1.7
0.0008
Mercury
D
-
0.0003
- not available
Source: U.S. Food and Drug Adminstration, Total Diet Survey 1987-1989 (Commerical Fish) and the IJC
(Great Lakes sport fish)
Noncancer Risks
Noncancer risks are assessed in this problem area in terms of hazard ranking index associated with the reference
doses and the estimated exposure levels for chlordane, dieldrin, and mercury, as used by EPA Region S. The Hazard
Index quantifies the extent to which a population is above the Reference Dose for chronic, noncancer health effects
such as reproductive impairment, neural effects or kidney damage. Noncancer effects are assumed to be additive.
A Hazard Index of 1.0 indicates that environmental exposures are not greater than a dosage considered to be
acceptable for long term exposures without a significant probability of an adverse noncancer risk. As a Hazard Index
increases above 1.0, the probability of adverse impacts increases.
' R*ftract Dom for ArocUor 1016 it 0.0001; 0.00001 kac km ntimiiH for pu-paaa ol Ibii paper.
-------
Page m - 5
In addition to the hazard ranking, however, additional evidence has recently been developed which indicates that
prenatal exposure to PCBs may affect childhood development Evidence of PCB teratogenicity came from studies
of pregnant women in Japan and Taiwan. In the children bom to these women, exposure to PCBs and related
contaminants was associated with low birth weight, neonatal behavioral anomalies, and poorer recognition memory
in infants.' A recent study evaluated 236 children whose mothers had consumed Lake Michigan sport fish during
their pregnancy. Prenatal exposure was found to affect the children by reducing short-term memory function on both
verbal and quantitative tests in a dose-dependent fashion.9 In this study, PCB levels in breast milk were not
predictors of short-term memory function deficits. Therefore, intra-uterine exposure may pose a greater threat to
infants than postnatal lactatinal exposure. This may be due to the protective barriers and metabolizing capacities
which are lacking in the fetus but which develop postnatally.
Exposure Assessment
Exposures to toxics in the surface waters of the Great Lakes Basin were estimated from two routes: ingestion of
drinking water and consumption of fish. Other possible exposure routes include dermal contact through recreation
in surface waters and dermal exposure through domestic use of water. These other routes were not considered
because because drinking water and fish consumption are considered to be the two major routes.
Data on water quality in the Great Lakes were used to calculate drinking water exposure levels. Data from the 1989
Report on Great likes Water Quality (Great Lakes Water Quality Board, Report to the International
Commission) were used to characterize water quality in the lakes. The water quality data used in this analysis are
presented in Exhibit G-7. Data on concentrations of contaminants in fish were used to calculate fish consumption
exposure levels. These data are presented in Exhibit G-8. Concentrations in sport fish are not provided for etcw
lake, but represent averages for the five Great Lakes. However, there can be several orders of magnitude different*,
between species and lakes.
Exhibit G-7
Approximate Concentrations of Toxics in the Water of the Great Lakes10
(lakewide averages, ngTL)
Contaminant
Lake Superior
Lake Huron
Lake Erie
Lake Ontario
alpha-BHC
7.84
5.91
3.82
3.95
Lindane
1.08
0.79
1.02
1.25
Dieldrin
0.29
036
037
0.31
pp-DDE
-
0.01
0.03
0.05
PCBs
0.33
057
1.16
1.20
HCB
0.03
0.03
0.0S
0.06
' Wo*, K.C. Ud M.Y. Hwang. 1*1. "CU*w ban to PCB iaamd ¦otkn' CHn. MM. (TaipaQ. 7tS347.
* Jacobiaa, J.L, S.W. Jacobaos, and H.ELB. Haaopkray. 1990. "Effada of is «tarv aapoaara to polydrioriaatad Mpbcayli and
nlitcrf f(BtinilM«l» oat copUin Auctioning ia young cUidrra.* J. Mhlr. 1163MS.
N Data wan aol available for Lalt* Michigan.
-------
Page HI - 6
Source: 1989 Report on Great Lakes Water Quality. Great Lakes Water Quality Board, Report to the
International Joint Commission, presented at Hamilton Ontario, October 1989.
Exhibit G-8
Approximate Concentrations of Toxics in Fish
(ppm)
Contaminant
Commercial Fish
(FDA)
Great Lakes Sport Fish
(UC)
DDT/DDE
0.009
0.37
Dieldrin
0.005
0.09
PCBs
0.001
1.32
Mercury
-
0.17
Source: U.S Food and Drug Administration, Total Diet Survey 1987-1989 (commercial fish) and the IJC
(Great Lakes sport fish).
For drinking water ingestion, we assume that the residents of the Great Lake Basin who derive their drinking water
from surface water sources ingest 2.0 liters of contaminated water per day. The estimated duration of exposure is
70 years (lifetime) and the estimated average body weight is 70 kg.
Surveys indicate that about 11% of the population in the Great Lakes basin are licensed sportfishennen.11
Therefore, 3.36 million persons are considered sportfisbermen. The remaining population (27.15 million) are
assumed to be non-fishermen.12 Based upon surveys of sportfishennen and their families, the average
sportfishennen consumes an average of 19 grams of fish per day.11 For the purposes of assessing exposure it will
be assumed that 3.4 million people consume 19 grams of Great Lakes basin fish per day," although it should be
recognized that the exposed population may be underestimated as only fish consumption by licensed fishermen and
not their families has been considered.
U.S. Department of Agriculture national statistics indicate the average person consumes about 19 grams of
commercial fish (primarily saltwater fish) per day.14 The estimates of commercial fish consumption for the Great
Lakes States indicate a lower consumption rate of 15 g/day. For the purposes of this assessment it is assumed that
the non-fishermen subpopulation consumes 15 grams of commercial fish per day and from zero to four grams per
"Wort, Patrick, «t «L Michigan Sport Anglers Fish Consumptive Sway. University of Michigan, 1989.
UA third category, anbriatence Unin, alio exists. Far example, Tribal mambara who practice snbaiateace fishing oa
raaarratioM and other ceded bttiloriaa. Indian* tend to consume man traditional food* Ik* fisb and game than either apart
Unita or son-fishermen populations, and so aaay be at higher risk. The risks to populations of subsistence fishermen need to be
addreaaed in fa tar* risk analyse*.
"For upper intake sport fisherman or subsistence fishermen, daily fish intakes would be approximately 100 • 150 g/day, aad
resultant risks would be approximately five tinea pair. See Donnan, Mkhael L., aad J. Milton Clark. 1990. "Fish
Consumption Advisories: Toward a Unified, Scientifically Credible Approach." Regulatory Toxicology and Pharmacology. 12, pp.
161-178.
"UJS. Department of Agriculture, Consumption aad Family Living, Agricultural Statistics, Hyattsvilk, MD, 1969
-------
Page in - 7
day of Great Lakes sport fish. An exposure duration of 70 years and an average body weight of 70 kg have been
assumed.
Human Health Risk Characterization
Cancer Risks
Cancer risks are determined separately for drinking water exposure and contaminated fish consumption.
1. Drinking water consumption.
Cancer risks are calculated below for the additive risk posed
by the contaminants evaluated in the water of the Great
Lakes. We assume that the concentrations of toxics in the
waters of the Great lakes is representative of all surface
water sources of drinking water in die Basin. The estimated
cancer incidence is based upon a Great Lakes Basin
population of approximately 12,700,000 relying on drinking
water from surface water supplied systems within Great
Lakes counties.11 Exhibit G-9 shows the distribution of
this population. Exhibit G-10 compares Basin populations
who use surface waters with overall Basin populations.
Based on the contaminant concentrations and exposure
assumptions outlined above, we predict an upperbound
incremental lifetime cancer risk of 5 x 10"6 for exposure to
toxics from surface water-derived drinking water supplies.
The potential number of excess cancer cases related to
ingestion of drinking water across the Basin totals
approximately 66 over a 70 year span.
These calculations are based on exposure to drinking water prior to water treatment Disinfection by rhlnrimt^
results in the formation of trihalomethanes (THM). A study by Purdue University of small systems (serving less
than 10,000 people) in Indiana found an average level of THM, as chloroform, of 60 ppb. The maximum average
exposure, obtained from FRDS violation reports, was 101 ppb. These levels would give rise to lifetime excess r»'">rr
risks ranging from 1-2 x 10"5. Assuming exposure to the total Great Lakes Basin population, the potential number
of excess cancer cases relative to ingestion of treated drinking water is between 130 and 230. Assuming no reduction
of contaminants (aBHC, Lindane, dieldrin, ppDDE, PCB, and MCB) the potential number of total excess
cases would range between 196 and 296. However, because much of die contaminants would be bound to
which will be removed during treatment, these numbers are upper bound estimates of risk.
u 1990 FRDS data retrieval
POPULATION vs SURFACE WATER
USE IN GREAT LAKES COUNTIES
TOTAL POP Hi SURFACE WATW
•ounce: CB«UB. morn
Exhibit G-10
-------
~
~
I
Lake Superior
Lake Huron
Lake Ontario
Lake ^
Michigan
Source: Federal Reporting Data
System - II, 1990.
Exhibit G-9
POPULATION SERVED
BY
SURFACE WATER SYSTEMS
Great Lakes Basin Counties
POPULATION
SERVED
0
1- 1,000
1,001 - 10,000
10,001 - 100,000
100,001 - 1,000,000
> 1,000,000
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
s&EPA
-------
Page IE - 8
2. Fish consumption.
Cancer risks arc calculated below for the additive risk posed by the contaminants evaluated in Great Lakes fish. The
number of potential cancer cases is based upon a Great Lakes Basin population of 30,050,000.
Potential Cancer
Population Lifetime Risk Cases/Year
sport fishermen (average) 3.36 million 3 x 10'1 160
non-sport fishermen 27.15 million 2 x 10* to 7 x 10*4 <1 to 270
The total numbeT of potential excess cancer cases related to consumption of contaminated fish totals 30,000 cases
over a 70 year span. These data reveal that fish consumption poses several orders of magnitude of excess cancer
risk when compared with the risks posed by drinking water consumption.
Noncanccr Risks
1. Drinking water consumption.
We estimate that drinking contaminated surface water would not pose a risk of noncancer health effects based on
the constituents monitored in the water of the Great Lakes. We assume that noncancer effects are additive and that
concentrations toxics in the waters of the Great lakes are representative of all surface water sources of drinking water
in the Basin, but the maximum hazard index calculated is only about 0.003 (PCBs), well below the threshold of l.O,
Data on the concentration of mercury in the Great Lakes was not available and, therefore, a hazard index could not
be calculated. However, due to bioaccumulation of mercury in fish, fish consumption, not drinking water, is judged
to be the dominant pathway.
2. Fish consumption.
Noncancer risks are quantified below using hazard index values for the constituents monitored in Great Lakes fish.
The noncancer effects are assumed to be additive.
Hazard Index Population
spoit fishermen (average) 4.0 - 40 336 million
non-sport fishermen 0.02 - 8 27.15 million
As shown above, 3.36 million sport fisherman in the Basin are at risk of noncancer health effects from «Hisniqjng
sportfish. The hazard index is greater than one, and may be as great as 40 for the average sport fisherman using
a reference dose for higher chlorinated PCBs of 1.0 x 10"3 mg/kg/day. If die reference dose (1.0 x 10"4 mg/kg/day)
for Arochlor 1016 is assumed, then the hazard index would be about four for the average sport fisherman.
The hazard indexes assume an average fish consumption intake and average contaminant levels. Other evaluations
have also indicated high hazard indices from frequent consumption of sport fish. If a higher rate of fish consumption
or consumption of fish from particular locations in the Basin is assumed, resultant hazard indices may be
-------
Page IH - 9
considerably higher.1' Non-fisherman who occasionally consume Great Lakes fish may also have some probability
of risk of noncancer effects, because the hazard index for these individuals may be greater than one.
As discussed previously, noncancer effects of particular concern include reproductive impacts, such as low infant
birth weight and neonatal behavioral anomalies, which have been observed in women who frequently consume
particular sport fish from the Great Lakes.
Ecological Impacts
Thirty years ago, warning signs of serious problems in the health of the Great Lakes ecosystem began appearing.
These early signals included fishery stock losses, gross defects and population declines in birds, and the death of
ranch mink when fed Lake Michigan salmon. Today, symptoms of a damaged ecosystem are found throughout most
of the Great Lakes Basin: bald eagles along the shores of the lakes do not reproduce as well as those that reside
inland and, in some areas, do not reproduce at all; other fish-eating birds and mammals at the top of the food chain
suffer marked developmental abnormalities; many species of fish can no longer survive in the waters of the Lakes,
and many of the survivors can no longer reproduce in situ.11
The Relationship of Cause and Effect
The presence of toxic chemicals in the Great Lakes coincides with poor health and reproductive problems in many
species. Studies have identified many cases in which the testing of individual animals (or their eggs) that belong
to troubled populations, showed relatively high concentrations of at least one contaminant Also, animals that live
near the shoreline, such as the bald eagle and mink, have been found to accumulate more toxic substances than do
members of the same species living inland. However, the association between health effects and concentrations of
toxicants is not enough to prove causation. The problem of linking cause with effect is illustrated by a 1983 study
In which reproductive impairment in Forster's terns could not be indisputably linked to exposure to PCBs and
dioxins; the weight of evidence suggested that the problems of the Forster's terns were caused by these chemicals,
but otter causes could not be completely ruled out11
Efforts have been made to prove cause-and-effect relationships through controlled laboratory experiments. A series
of laboratory tests have shown that the chemicals introduced into the Great Lakes environment over the past half-
oentury disrupt normal functioning in cells of animals. These chemicals are associated with a number of
reproductive, neurological, immunological, and carcinogenic effects and with abnormal sexual development The
utility and applicability of laboratory tests are limited, however, as they do not simulate the complex species
interactions and environmental influences of natural ecosystems nor can they account for such factors as adaptability
or pre-exposure to a toxicant. In addition, the effects of a toxicant in the environment depend upon concentration
and chemical form. The effect of a toxicant on an organism in the field is determined by a complex suite of abiotic
and biotic factors such as pH, temperature, alkalinity, suspended solids, dissolved oxygen, and the presence or
absence of specific types of bacteria.
The analysis of how various toxicants affect wildlife is further confounded by the fact that different chemicals
produce different types of toxic effects. Some metals produce similar toxic effects among many different groups
M Dobtmb, Michael L., and J. Milton Clark op. dt
" Great Lafcwi Great Legacy?. pp. u and 131.
* Great I akft: Great Legacy?, pp. 139 -135.
-------
Pigc m -10
of animals (e.g., mercury is neurotoxic in fish, birds, and mammals), while other inorganic stressors produce different
toxic effects in different animal groups (e.g., selenium is teratogenic in birds but does not appear to cause Rimii^r
abnormalities in mammals).19 Although it has been difficult to prove that a specific chemical causes a specific
effect on a given species, scientists have been able to combine laboratory tests and field observations to make some
strong associations.
Toxicity Assessment
Five general categories of impact may be observed in the Great Lakes Basin resulting from point and nonpont source
discharges: oxygen-demand from BOD discharges, eutrophication from nutrient discharges, microbiological
contamination from viral and bacterial discharges, toxic impacts from organic and inorganic constituent discharges,
and beat impacts from thermal discharges.
1. Oxygen Demand:
Oxygen demanding materials are usually described by biochemical or chemical oxygen demand (BOD and COD,
respectively). They act on aquatic systems by removing oxygen from the water column either through chemical
oxidation (COD) or, most likely, by biological and chemical means (as measured by BOD). Lack of oxygen leads
to decreased survival of fish, zooplankton, and macroinvertebrates. As with culturally induced eutrophication a more
or less permanent shift in the structure of the ecosystem occurs until the causal agents are removed and the system
has an opportunity to return to its natural state.
2. Eutrophication:
Nutrients typically refer to various compounds of nitrogen and phosphorous. Their effect is a stimulation of growth
in algae. Such increased growth allows the dominance of less desirable forms of algae, leads to increased turbidity
in die water column, and unless controlled, leads to culturally induced eutrophication. Such eutrophication is a
vicious cycle of intense algal production, progressive deoxygenation of bottom sediments and then the water column,
and ultimately, severe shifts in the natural ecosystem to one dominated by a few species of pollution tolerant
organisms.
3. Microbiological contamination:
Microbial pathogens are associated with municipal sewer systems and nonpoint runoff that has been contaminated
with septic system waste or waste from animal feeding operations.
5. Heat:
The discharge of heated water from power plants and primary metals manufacturing can cause significant local
effects. Aquatic life can be absent from the area because it is simply too hot for them to survive. Shifts in the
composition of fish and insect communities can occur so that the natural system is replaced by heat tolerant species.
For example, gizzard shad has been observed to overwinter in cold water areas due to the localized beat supplied
by such discharges.
5. Toxics:
Inorganic toxicants include heavy metals such as lead, cadmium, mercury, copper and zinc; and other inorganic
compounds such as cyanide or dissolved solids. There are numerous toxic effects of these compounds, indurffag
" ICF 1b corpora tad, IMS. Ecoloricd Writ AuwimcBt DocnmtaU Pwdnctd for Pmmhmh CoBMwtiw Ri«k ftaittt
CfcapUr ILG„ p. 3.
-------
Page m - 11
death, impairment of reproductive processes, and changes in plankton, fish and macroinvertebrates. The particular
effect depends on the concentration of the toxicant and can be affected by the chemistry of the water bodies: some
metals are less toxic in harder water. Ecosystem effects may include minor impacts such as a small change in
species diversity (as measured by standard indices that consider the number and richness of species as well as the
number or individuals) or major impacts such as loss of certain types of species previously suited to the habitat
Once released into the environment, these compounds partition in bottom sediments, the water column or biological
organisms.
Ecosystem-level effects of toxics include:
¦ reduced primary and secondary productivity;
¦ loss of top carnivores;
¦ changes in community composition; and
¦ modification of nutrient cycling.
The effects of heavy metals in the environment depend on their concentrations and chemical form(s). The chemical
form(s) of metals are determined by complex suites of both abiotic and biotic factors including pH, electrical-
chemical potential, salinity, alkalinity, the presence of other metals and ligands, dissolved oxygen, awl the presence
or absence of specific types of bacteria. Toxic compounds that are persistent, such as metals, and that bioaccumulate
can have serious adverse effects on species at high trophic levels (e.g., trout) despite low environmental
concentrations. Exhibit G-ll lists EPA acute and chronic Ambient Water Quality Criteria (AWQC) and
bioooncentration factors for arsenic, cadmium, lead, and mercury. These toxicity values indicate that cadmium, lead,
and mercury can produce adverse effects at low environmental concentrations.
Exhibit G-ll
Toxicity Reference Values for Aquatic Species for Metals
Metal
EPA Acute Freshwater
AWQC (ug/L)
EPA Chronic Freshwater
AWQC (ug/L)
Bioconcentration Factor
Log(BCF)
Arsenic HI
360*
190*
12
Arsenic V
850
48
1.2
Cadmium
3.9*
1.1*
3.6
Lead
82'
3.2'
3.4
Mercury
2.4
0.01
3.4
* Data insufficient to develop an AWQC. Value listed is a lowest observed effect level.
* Criterion is hardness dependent Value listed is based on 100 mg/L.
Sources: U.S. EPA. 1986. Quality Criteria for Water, and U.S. EPA. 1979. Water-Related Fate of 129 Priority
Pollutants.*
Concentrations of toxics in Great Lakes fish are well documented. Exhibit G-12 shows 1987 samples of lake trout
show trends in PCB concentrations for Lake Michigan Lake Trout over a twelve year period. PCB concentrations
-------
Page m - 12
for Lakes Superior, Huron, Erie and Ontario fiar exceed acceptable levels outlined by tbe Great Lakes Water Quality
Agreement objective of 0.1 ppm. Organic toxicants, such as PCBs, dioxins, pesticides20, solvents, chlorination
products bom municipal sewage treatment plants, and other synthetic organic compounds, have effects similar to
the inorganic toxicants discussed above. In addition, many of these compounds can bioconcentrate in an organism
and biomagnify through tbe food web. For example, a zooplankton can bioconcentrate the ambient concentration
of PCBs 500 times during its life.21
LAKE MICHIGAN LAKE TROUT
There is growing evidence of the impacts of toxics on
wildlife. Piscivorous wateibirds, such as herring gulls and
cormorants, have proven to be a reliable, sensitive,
indicator for detecting toxic effects and ecosystem
changes.22 The eggs of a herring gull, a pi&civore in the
Great Lakes ecosystem, show a biomagnification of 25
million times the ambient concentration of PCBs.23 Thus
these compounds have the ability to exert effects in
predators even when the ambient water concentration is
very low. Predators, even those that are extensively
terrestrial, can suffer reproductive disorders, behavioral
abnormalities and other ill effects induced by these
compounds.
Exliibit G-12
One of the most demonstrable toxic effects is from the
Lake Ontario ecosystem where Gilbertson reported on the
sever* reproductive Mure of Scotch Bonnet bind hening gull colonies." GaberBon "ported - low breeding
sucm/viiue of 012 fledged young per td.lt nuting p«ii. This fe .bout one-tenth the success iMe to. bemng gulls
forlorn, the New &gfsndco»t. On He ssm« islsnd in W3, Gilbemo, .nd Hsle found tbst, on sversge, ody
16 p«»n?rf the eggstaucbed. Tbe man bleeding sua*» wss 0.06 Hedged young p«sdult pnJWbertson found
LfeS of Scotctf Bonnet islsnd bentag gull egg. thin .nd highly conttn.to.ted (PCBs over 800 ppm .nd DDE
2qa These vslues were the highest found for gull eggs on Ihe Greit Likes ind, when compared to more
St uJL (Id ppm DDE) .nd Ore B., of Fuudy (32 ppm DDE,, empbssfce the
magnitude of contamination in specific areas.
Teeple assessed the breeding failure of herring gulls on Brothers Island in eastern Lake Ontario.2* On average, only
23 percent of the eggs laid eventually hatched, and the breeding success ranged from 0.06 to 0.18 fledged young per
adult pair. Further study by Fox et al * found reproductive failure of herring gulls in the Great lakes was mostly
" Tke ecological effects of pesticide* art iddnnd hi proMeai am L
" Draft Probieai Ana Paper, "Moaidpai, ladostrial, aad Noopoiaf Sobw Discharges to Surface WaUn," page 7, Region 5
Coai para live Ride Pntfect, 199ft
" Regk* II Mouuary of probkat area kapecta oa tk« Lake OaUrio Bad*.
"fcld.
^GBbartMk, M. 1974. "PalioUaU ia Breeding Henfeg Galk la the Lower Gnat Lake*." Caa. FMd-NataraHat Si273>3M.
" Taeplc, S.M. 1977. " Reproductive Success «r Herring Gab Nesting cm Broghtn Island, Lake Ontario fa 1973." Caa. FMd-
NaMraliaL 91(148.157.
" Fox, ME* Cany, J.fL aad D.G. Oliver. 19S3. "ConiparbneaU] Distribution of OiiuockMw CoaUiMiaeaU ia ike Niagara
River aad the Western Basin of Lake Ontario.* J. Great Lakes Research. 9:7X1-294.
-------
Page m - 13
restricted to Like Ontario. To a degree, the situation has improved. By 1977-78, Weselob reported the breeding
success of the Scotch Bonnet Island colonies to have improved from 1.01 to 1.1 fledged young per adult pair.27
Birth defects have been found in doublc-crested cormorant populations in Green Bay and Saginaw Bay. Crossed
beaks and yolk sacs hanging outside of unborn embryos have occurred in this and other piscivorous species.28
Since 1986, researchers have found deformities in more than 1 of each 100 cormorant chicks bom on islands in
Green Bay.
Exposure Assessment
The Great Lakes contain approximately 20 percent of the fresh water on our planet There are nearly 88,000 square
miles of surface area associated with Lake Superior (31,700 square miles), Lake Huron (23,000 square miles), Lake
Michigan (23,300 square miles), and Lake Erie (9,900 square miles). Hydraulic retention times in the individual
lakes range from a high of 200 years in Lake Superior, to a low of 2.6 years in Lake Erie. The ability of the lakes
to purge themselves of pollution, therefore, varies dramatically. The Great Lakes contain a vigorous sport fishery,
although it is supported by an extensive program of introducing nonresident species, such as Coho salmon in certain
areas. The lakes and their shores are especially sensitive to nutrient enrichment, impacts from toxic compounds and
oxygen depletion. Bioconcentration and biomagnification of toxic organics present a threat to non-aquatic species
sucb as mink, gulls, and bald eagles.
Some improvements in ecological impacts have been noticed in the Basin. There is dear evidence that the levels
of some problem toxics in Lake Ontario biota have been reduced over the past two decades. Levels of PCBs, mirex,
DDT and metabolites, dieldrin and hexachlorobenzene in herring gull eggs taken from colonies on Lake Ontario
during the period from 1974 to 1986 show significant declines. Levels of PCBs in lake trout, brown trout, and coho
salmon collected since 1975 show significant declines.
There is evidence that mercury loadings from industrial sources may be decreasing, while concentrations of mercury
at many locations within the Basin are actually increasing, possibly due to atmospheric deposition. This potentially
significant trend should be considered a high priority. Recent trends in mercury are addressed in Appendix H.
By contrast, the trends in levels of mirex in lake sportfish are not clear. A major concern remains, however, that
problem toxics, such as mirex, in Lake Ontario biota may be stabilizing at unacceptably high concentrations.
Although toxics concentrations in Lake Ontario appear to have declined over the past decade, reductions are no
longer occurring at the rate they have in the past Current reductions in toxic loading may not be sufficient to sustain
corresponding reduction in toxics concentrations. Bioaccumulation, atmospheric deposition, runoff, and resuspension
of toxics from contaminated sediments are continuing sources of concern.19
Ecological Risk Characterization
There are approximately 4600 miles of Great Lakes shoreline in the Basin. 320 miles (7%) are identified as fully
supporting their designated uses in the 1990 305(b) reports, 990 miles (22%) are identified as partially supporting
** Wmttk, D. et aL 1979. "OrginocUorine CoaUmiaanU ud Treads h Riprwhictioa Id Gnat Lakaa Herring Galls, 1974-
197&" Truft. 440> N.A. Wildlif. Nat. Resources Coot 543-557.
" Beta, Dob, "Bird defects warn human*," page 7 to "The Great Lakesi A Toxic Tragedy," The MDwanltte J~i»-n»lr 1989.
" Region II ranmiry on problem area impacts in the Lake Ontario Basin.
-------
Page in - 14
uses, and 3290 miles (71%) are identified as not supporting designated uses.*1 This unique ecosystem is severely
disrupted in those areas where the designated uses are not being attained. Native species are often absent, or those
that are present are subject to unacceptable levels of other stresses such as reproductive or behavioral abnormalities.
Where uses are being partially supported, the long term viability of the system is threatened. In both cases, the
timeframes for recovery is extended. Decades to centuries may be required for the systems to recover once the
stressors are removed or controlled.
Nutrients are identified as the second most widespread influence on the shoreline areas. Eutrophication has also been
a concern since algae blooms began to choke the lakes in the mid 60s and Lake Erie was pronounced dead in the
70s. Presently all of the lakes which receive discharges from the Basin meet the in lake target for phosphorous, the
material that effectively controls the production of algae. Lake Erie is the wannest, shallowest and most productive
lake and still is threatened by culturally induced eutrophication. At present, that threat is of low severity, as controls
are in place that ensure that effects will be marginal, if all other factors such as water levels and atmospheric
temperature remain more or less constant Reversibility of the damage is medium to high based on the experience
of phosphorous control in the past and the relatively low hydraulic residence time.
An increase in sodium levels has also been observed over the past 15 to 20 years in the Lake Michigan basin.
Cultural uses of sodium, such as the use of road salt for clearing roads in the winter, are major sources. Such
increases represent additional ecological risks to the Basin. Increased sodium concentrations favor the growth of
blue-green algae, and may enhance eutrophication by stimulating phosphorus transport into algal cells.11
While the health of the organisms at the upper end of the food chain depends on the stability of populations at the
lower end, little is known about the potentially significant effects of toxic contamination on organisms such as
phytoplankton and zooplankton. Although it is known that these organisms are generally sensitive to toxicants, the
ability to predict the effects of a toxicant on natural communities is confounded by a variety of physical and
biological factors.
Generalizations about the health of animals at the top of the food chain are based mainly on individual studies of
specific species or populations and on data from programs that monitor particular species of fish and herring gulls.
An analysis of the combined results of these studies summarized several trends and convergences. The table shown
below, which was adapted from Great Lakes: Great Legacy?, by the Conservation Foundation and the Institute for
Research on Public Policy, summarizes reports in the scientific literature on the state of wildlife populations, focusing
on identifiable health effects.
Population, Organism, and Tissue Effects Found
in Great Lakes Organisms
SpadH
repMet
Dadta*
¦Uprat
Bkb
Wartta*
¦M
nam
tea*
onm
fc a j 1
¦Mrami
fTw.it.Mt
UIkU
Mlatk
X
X
X
X
X
bma
X
X
X
X
" TImk anbtn ahouM b* conaidvad approjdautc. 30$ (b) rtporU caatain acvcrml discrapaaciM. Baca dm Umm nport* cover
atatawida water conditions, they may sot accurately nfWct Grot Lakaa caadittoM,
" Goem, Bnma, Paul Horvatin, Vacyi Sanlyt, aad David Data, "Lake MUfcifaa Sadism Maaa Balaac* and Pntftctiow for
Future In-Lakt Sodium Conceelmlioes," US. EPA, Anft, 19tS, p. 2.
-------
Page m - IS
CifiaM
X
X
X
X
X
X
X
Chisook/oofco aaksm
NA
X
NA
X
X
Hm in
X
X
X
X
X
DrnUv tw—irt
X
X
X
X
X
X
X
X
In
X
X
X
X
X
X
X
X
HoriagpU
X
X
X
X
X
X
X
X
X
Lake Iron
X
X
NA
X
X
X
Huk
X
X
NA
X
X
OmfKf
X
X
X
Oatr
x
NA
Ris f-billed yill
X
X
X
X
SMITHS arte
X
X
X
X
X
X
X * Observed effects that have been reported in the literature. Cells not marked do not mean there is not effect; only that no citation was found.
NA * Not applicable.
Summary of Effects:
Population Decline and Reproductive Effects. Scientists have found that animal species at the upper end of the Great
Lakes food chain have shown occasional and, in some cases, long-term reproductive problems and/or population
declines since the 1950s. The affected species, including birds, mammals, and fish, all derive sustenance from
animals in the Great Lakes food chain.
Fggshell Thinning. It is well-documented that oiganochlorines cause eggshell thinning in many fish-eating birds.
In fact, the only specific chemicals to which a cause-and-effect relationship in wildlife has been assigned
unequivocally are DDT and its decay product, DDE which have been shown to cause eggshell thinning. Elevated
concentrations of DDE and other residues have been found in the eggs of several Great Lakes fish-eating bird
species.
'Wasting". Several species showed changes in fit metabolism and disposition of body fat A condition termed
"wasting" was reported in extreme circumstances. Animals in this condition appear lethargic, lose their appetites,
and eventually die. In some cases, emaciated and dead birds with low body fat and high concentrations of
oiganochlorines have been found.
Birth Defects. Species of mammals, reptiles, birds, and fish in the Great Lakes basin have also shown significantly
more birth defects than control populations with less exposure to toxic contamination. For example, some chicks
in colonies of double-crested cormorants were born with deformed bills such that they were not able to feed
themselves. Other birth defects seen in double-crested cormorants and other bird species include club feet, shortened
appendages, missing eyes, absence of brains, major organs developing outside the body, and edema.
Tumors. An unusual number of facial tumors have been reported in various fish species that dwell in the Great
Lakes, such as the brown bullhead, a bottom feeding catfish. Extensive research has shown that liver tumors also
-------
Page m - 16
develop in fish in the presence of carcinogens and carcinogen-containing sediments. In addition, tumors have been
found in whales that reside in the estuary of the St Lawrence River.
Target Organ Damage. Target organ damage, as opposed to systemic damage, occurs in specific organs such as the
thyroid, liver, kidney, or brain. One study, for example, examined thyroid damage to herring gulls in the Great
Lakes and postulated that the cause was environmental contaminants. Another study found that porphyria, a liver
condition, is aggravated by the presence of organochlorine chemicals.
Immune Suppression. Immune suppression, or the lowering of an organism's ability to withstand disease, has been
found to occur in both the common tern and the herring gull.
Behavioral Effects. The literature also suggests that behavioral changes in species may be attributed to toxic
chemicals. Abnormal behavior such as female/female pairing of herring gulls, lake fry swimming upside down, and
nocturnal abandonment of nests by terns have been associated with exposure to toxicants.
Generational Effects. The literature implies that, for some species, stress on parents, such as exposure to chemical
contaminants, affects the health of an embryo or newborn. These effects include metabolic, hormonal, and target-
organ changes that are manifested by "wasting", abnormal development, and immune suppression.
Welfare Impacts
Economic damage due to dischaiges to surface waters include costs from human health effects, losses in commercial
and recreational fisheries, and other costs from impaired uses. Because of the lade of quantitative data on many of
these costs, quantified economic damages are based on estimated annual cancer cases.
Risks from point sources result from release of both conventional and hazardous materials into surface waters
Economic damages can occur at levels permitted under the NPDES system, but are more likely when permitted levels
are exceeded. Risks are greatest where receiving waters provide the least dilution and dischargers are clustered
enhancing cumulative "down stream* impacts.
Health Costs
Human health risks and related economic damages are associated with direct and indirect contact with pollute
receiving waters. In the preceding analysis, we estimated 30,000 potential cancer cases over 70 years resulting
primarily from ingestion of contaminated Great Lakes Basin fish. To estimate costs associated with these cancer
cases, we multiplied the estimated number of cancer cases by the direct medical cost and foregone earnings per
cancer case:
(Annual Cancer Cases)(Dircct Costs & Forgone Earnings) * Health Costs
These estimated direct and indirect medical cancer costs ate based on a range of cost per case estimates. The lower
bound estimate, based on Hartunian, et al., is $80,000, while the upper bound estimate developed by the American
Cancer Society is $137,000. These estimates provide differing values for foregone earnings and medical costs
associated with cancer treatment Both estimates ate weighted average costs associated with all types of cancers.
Neither of these estimates, however, take into account the value of individual pain and suffering associated with
cancer and should, therefore, be considered partial economic damage estimates. They also depict costs over a 70
year time period.
-------
Page ffl - 17
Lower bound estimate
HC=(30,000X$80,000)=$2.4 billion (1988 $)
Upper bound estimate
HC<30,000X$137,000)=S4.11 billion (1988 S).
Non-Health Costs
Estimating non-health economic damage due to surface water pollution entails describing the effects of water
pollution on numerous services that society receives from surface water. Quantitative data are not currently available
to support estimates of the damage caused by surface water pollution to these services. The services include:
* freshwater recreation;
* water storage;
* navigation;
* irrigation;
* freshwater recreational and commercial fishing;
* drinking water supply;
* industrial water use; and,
» steam power cooling.
In lieu of quantitative damage information, fish advisories can succinctly summarize water quality problems in the
Basin, and the effects those problems have had on commercial and recreational fisheries. Exhibit G-13 provides a
summary of fish advisories issued by State agencies in Wisconsin, Michigan, and Illinois. Fish advisories fall into
three groups, depending upon the amount of health risk involved with eating the fish. Advice to the public differs
by group:
¦ Group 1: Contaminant levels in 10% or less of fish tested are higher than one or more health
standards. Eating fish from this group poses the lowest health risk. Advise is to trim fat and skin before
cooking and eating.
¦ Group 2: Contaminant levels in more than 10% but less than 50% of the fish tested are higher than
one or more health standards. Nursing motheis, pregnant women, and women who intend to have children
should not eat Group 2 fish. Overall consumption of Group 2 fish should be limited. Trim skin and fat
from these fish before eating them.
¦ Group 3: Contaminant levels in 50% or more of tested fish are higher than one or more health
standards. No one should eat Group 3 fish.12
* Descriptions of health advice ii takes from Wisconsin Department of Natural Resources, Wisconsin Division of Health'* "Health
Advisory Ear People Who Eat Sport Fish from Wisconsin's Waters,* October, 1989.
-------
Page m - 18
The exhibit shows that there are numerous fish advisories in these states. In fact, all Great Lakes have health
advisories on the consumption of some Great Lakes fish. Recent reports indicate that these fish advisories are
insufficient protection to human health for those who frequently consume sport fish.35 However, there are no
quantitative data that fully describe the true economic impact of these advisories of recreational fishing and fish
consumption.
While these advisories, if followed, may limit the human health risks discussed above, they indicate a considerable
welfare impact from contamination of the Lakes. Fishing advisories began on Lake Ontario in 1970 with the
discovery of bioaccumulated mercury and DDT. Later, additional advisories were imposed with the discovery of
bioaccumulated PCBs, dioxin, and mirex. In 1976, a very unpopular ban was placed on the possession and stocking
of seven Lake Ontario species — it was lifted in 1978. There are no data on the impact these bans had on
consumption of Lake Ontario fish.
The decline in valuable commercial fisheries exemplified by the fish advisories has affected a number of
communities on the Great Lakes. By the 1960s, New York's reported commercial harvest was only 7 percent of
the maximum recorded in 1879.14 As a result, recreational fishing now dominates use of Lake Ontario fisheries
resources. Both recreational and commercial fishing currently make a significant contribution to local and state-wide
economies. In 1982, recreational fishing expenditures were valued at approximately $47 million; commercial fish
values for 1983 were reported at $184,000. The total value of direct and indirect economic activities of the U.S.
Great Lakes fishery has been increasing since die 1970s, and as of 1989, is valued at $3.3 billion." This major
economic activity is at obvious risk.
Several sources report a decline in the Great Lakes fishery. However, this decline cannot be presently attributed to
any single factor. Those factors which do affect the fishery may include loss of spawning habitat, overfishing,
competition between species, introduction of exotic species, and toxic impacts. No information was available to
allow a determination of the effect of toxic biomagnification and resulting fish advisories on the fishing industry.
At various times, however, several States have insitituted bans on certain fish consumption.
Exhibit G-13
Fiih Adviaoriea in Wiaoonain. Illinois and Michtgv
L WiacoMfa
Fiih Specie*
Group 1
Group 2
Group 3
Like Michigan
Lake trout up to 20"
Co bo salmon up to 26"
Chinook Mlmon
up to 21"
Brook trout
Rainbow Bout
Lake trout 20-23'
Coho salmon over 26*
Lake trout over 23"
Qiiaook aalmon over 32'
Brows trout over 23"
Chinook salmon 21 to 32"
Brown trout up to 23"
Carp
Catfish
Pink salmon
" Michael L Douraon and J. Milton Clark, op dt.
M Region D summary of problem area impacts in the Lake Ontario Basin.
" Great Lake* Fishery Commission. 1989. "The Report of the Evaluation of the Great Lakes Fishery Commission by the Bi-National
Evaluation Team: an analysis of the economic contribution of the Great Lakes aea lamprey program." p. 6.
-------
P»ge in -
Smell
Pen*
Green Bay (ouch of
Mariaette and iut tributaries
Rainbow trout to 22"
Chinook salmon up
lo2S*
Brook trout up to IS"
Smatkaouih bus
Northern Pike up v> 28"
Perch
Walleye up to 20"
Biown trout up lo 12"
Bullhead
While tucker
Sptake up to 16*
Rainbow trout over 22"
Chinook salmon over 25"
Brown trout over 12"
Brook bout over 15"
Splakc over 16"
North an Pike over 28*
Walleye over 20"
White baas
Carp
Lake Superior
Lake trout up lo 30"
Lake trout over 30"
IL
m»ois
Fish Specie*
Group 1
Group 2
Group 3
Lake Michigan
Lake trout up to 20*
Cobo salmon up to 26"
Chinook salmon
up to 21'
Brook trout
Rainbow trout
Pink salmon
Smell
*» «
cores
Lake trout 20-23"
Cobo salmon over 26*
Chinook salmon 21-32"
Brown trout up to 23"
Lake trout over 23"
Chinook salmon over 32"
Brown trout over 23*
Carp
Catfish
IIL Michigan
Lake Huron
Saginaw Bay
Lake Superior
Lake Michigan
Green Bay
Fish Scecias
Restrict Consumption
Brown trout up to 21*
Like ¦ rout
Rainbow trout
Rainbow troul
Brown trout
Lake trout 20-30*
Like trout 20-23"
Cobo Salmon over 26"
Chinook salmon 21-32*
Btown tout up to 23*
Spiake ap lo 16"
No Consumption
Brown trout over 21"
Carp
Catfish
Lake trout over 30*
Rainbow trout over 23"
Chinook over 32*
Brown trout over 23*
Carp and Catfish
Brook trout ever 15"
Rainbow mat over 22'
Chinook salmon over 23*
Btown trout over 12*
Spiake over 16*
Northern pike over 28"
Walleye over 20"
-------
Page in - 20
Lake Erie
Wbii« base
Carp
Carp
Catfish
Lake St. Clair
Walleye over 20"
White Ban over 12*
Smouth. Bast over 14*
Yellow Perch over 12*
Carp over 22"
Rode Bus over 8*
Black Crappie over 10*
Lmouth. Baas 12-14"
BhiegiU and Pump-
kinseed over 8*
Freshwater Dram
over 14'
Carpsucker over 18*
Brown Bullhead over 10"
Northern Pike over 22*
Largemouth bass over 14"
Muskie
Sturgeon
Catfish over 22*
St. Clair River
Detroit River
Gizzard Shad over 10*
Carp
Carp
Apportionment of Risk
As mentioned above, most risks resulting from surface water contamination cannot be differentiated from among
industrial, municipal, atmospheric or other nonpoint sources. Differences in source proportions from lake to lake,
and from harbor to harbor, further obfuscates any attempt at lakewide apportionment of risks. However, while
current techniques for the apportionment of risk among these problem areas face significant uncertainties, it
nevertheless helps to conceptualize source proportions in order to analyze the wide-ranging methods necessary for
reduction of those risks. This study uses a rough apportionment technique and hypothetical results to underscore
not only the difficulty in determining sources, but also the need for accurate knowledge of those sources, where real
reductions of contaminant loadings begin.
The principal sources of PCB contamination in the Great Lakes will be associated with the greatest human health
risks. Sources of PCBs in the environment are varied. While sediments are a known sink for PCBs throughout the
Great Lakes, the contribution of PCBs from these sources to fish concentration levels is unknown. This potential
source is considered in a following problem area. In many other areas of the Great Lakes, atmospheric deposition
appears to be a main PCB source. The percent contribution of PCBs to the Lakes from atmospheric deposition is:
Super. Mich. Huron Erie Ontario
percent PCBs from
atmospheric dep.16 90 58 78 13 7
M LIC. 1987. "Summary Report of the Workshop on Great Lakes Atmospheric Deposition." presented at Windsor, Ontario. October
1987. p. 14. See footnote 43 for discussion of accuracy.
-------
Page m - 21
Weighting these atmospheric contributions by the populations of the individual lake basins yields a Great Lakes
Basin-wide PCB population exposure proportion from air deposition of 38 percent
Few data are available to differentiate among the remaining contaminant load sources to the Great Lakes Basin. As
presented in Exhibit G-14 below, those data that are available tend to focus on phosphorus and other nutrient load.
Because of the higher nutrient content in agricultural nonpoint source and municipal point source discharge, these
data may not adequately represent many organic and inorganic toxic constituent loads. Limited data estimating load
by source for these other constituents are available for selected areas in the lakes.
Exhibit G-1417
1986 Approximate Phosphorus Loadings to the Great Lakes
tons/vear
percent
industrial
250
1
municipal
3800
16
atmospheric
2260
9
nonpoint
18,120
74
In terms of PCB load to the Great Lakes, data collected for the St Clair River1* in Michigan provide some
indication of the relative importance of the various source terms. The data in Exhibit 15 illustrates that although
nonpoint sources still predominate PCB load in die St Marys River, municipal and industrial sources are more
important sources than estimated for phosphorus load.
Exhibit G-15
Approximate PCB Load by Source for the St Clair River
PCB load Percent
(g/davl Contribution
Industrial point source 0.3 5
Municipal point source 25 41
Urban nonpoint source39 3.2 54
Exhibit G-16 presents data describing polycydic aromatic hydrocarbon (PAH) load to St Marys River. These data
illustrate that the contribution of toxic organic constituents from industrial point sources may still be a significant
source in certain Great Lakes watersheds. "While the relatively high concentration of industrial sources on the St
" llC. 1989. "Report on Great Lakcc Water Quality."
* "Upper Great Lake* Connecting Channel* Study." 1988. p. 2S8.
" Nonpoint source load estimate derived from ratio of point aource to urban nonpoint source load of 0.86 from Mtrialelt, J. and H.Y.F.
Ng. 1987. Contaminants in Urban Runoff in the Upper Great Lakes Connecting Chancels Area. Environment Canada, National Water
Research Institute. NWR1 #87-112. As reported in "Upper Great Lakes Connecting Channels Study * 1988.
-------
Page m - 22
Marys River signify that this load distribution across sources may be somewhat anomalous,40 these data are reported
to illustrate the diversity in load distributions across the Basin.
Exhibit G-1641
Approximate PAH Load to the St Mary's River
Point Source kg/dav percent41
industrial 1-25 48
municipal 0.42 16
tributary 0.05 2
Nonpoint Source
urban 0.65 26
atmospheric 0.25 8
rural NA
In sum, the data reported in the above three exhibits illustrate the difficulties in apportioning contaminant load across
the various sources within the Basin. Nevertheless, for the purposes of this analysis and based on load data reported
for the other connecting channels, the Niagara River, and other Great Lakes tributaries, it is apparent that nonpoint
sources and municipal point sources are significant sources of toxic organic and inorganic constituent load to the
lakes. While this conclusion may not hold for all toxic constituents, the source load proportions described for the
SL Clair River and the St Marys River serve as the basis for a preliminary allocation of the non-atmospheric human
health risk estimates to the industrial, municipal, and nonpoint (rural and uiban) source problem areas. The toxic
constituent load allocations are as follows:
Source Load percentage
industrial point source 16
municipal point source 18
nonpoint source 28
atmospheric load 38"
40 The Si. Mary's River has about seven point sources on the Canadian side. These sources arc not subject to the same level of control
thai they would be in the U.S. portion of the Basin. This increases the uncertainty involved is using such distributions as overall estimates of
loadings. The U.S. side of the river is largely rural.
41 "Upper Great Lakes Connecting Channels Study." 1988.
4 Percentage of reported sources.
" He Baain-wide percent contribution of PCBs from atmospheric depostion was calculated by taking a population-weighted average of
the five values. Specifically, each lake's percent contribution we* multiplied by the fraction of the total Basin population that resides in that
lake's basin. The product of percent contribution and population fraction was summed for each of the five lakes to derive the Basin-wide
estimate.
-------
Page m - 23
This distribution across source terms will differ for nutrient or total suspended solid load to the lakes by placing
greater emphasis on nonpoint sources. Furthermore, this allocation does not account for contaminated sediments or
spills as significant sources of toxic constituents. Based on this distribution, Exhibit G-17 provides an apportionment
of potential cancer cases (drinking water/prior to treatment* and fish consumption) and welfare costs for industrial
and municipal point source discharges, nonpoint source loadings, and atmospheric deposition.
• See preceding section on Drinking Water Consumption under heading! of Risk Characterization, Cancer Risks.
Exhibit G-17
Hypothetical Apportionment of Risk to Surface Water Loadings
Source
Potential Cancer Cases/Year
Drinking Water
Prior to Treatment
(Upper Bound Estimates)
Potential Cancer Casec/Year
Fish Consumption
(Upper Bound Estimates)
Welfare Impacts/Year
Fish Consumption
(Upper Bound
Estimates)
Industrial Point
0.16
69
$ 9 Million
Municipal Point
0.17
77
$10 Million
Nonpoint
0.27
120
$16 Million
Atmospheric
034
163
$22 Million
Totals'*
0.96
430
$59 Million
** Reflects 95% Ci_ for cancer potency slopes
-------
Page m - 24
H. MUNICIPAL POINT SOURCE DISCHARGES TO SURFACE WATER
Problem Area Description
This problem area addresses contaminant load and surface water quality impairment resulting from discharges from
municipal point sources. These point sources include discharges from publicly-owned treatment works (POTWs)
and privately-owned wastewater treatment facilities that manage domestic waste. A significant portion of the
wastewater stream entering these treatment facilities may come from industrial indirect dischargers, such as
electroplaters, chemical blenders, or pulp and paper manufacturers. As a result, the status of industrial pretreatment
programs in the Basin are reviewed as part of this analysis. Finally, this section includes a discussion of impacts
from combined sewer overflows (CSOs), which have been recognized as a significant source of urban stormwater
and municipal wastewater overflow discharge.
Because many surface water impacts, such as toxics biocoaccotration in sport fish, cannot be differentiated from
among industrial, municipal, and nonpoint source discharges, human health impacts from fish consumption and oral
water intake are discussed in the previous Section G. In that section, the risk derived from this problem area is
differentiated by estimating the proportion of total contaminant load discharge to the Basin accounted for by
municipal sources.
Magnitude of the Problem
The risks to human health and the environment posed by municipal point sources are a function of the pollutants
discharged by these sources, the number of sources in the Basin, and the total pollutant load attributable to this
problem area. In addition, the States have recognized CSOs as an additional source of concern in many of their
surface waters. These CSO-rclated impacts are discussed separately.
Pollutants of Concern
A variety of pollutants are discharged from municipal point sources. These substances are generally categorized in
two broad classes: conventional and non-conventional or toxic pollutants. The conventional pollutants include those
substances that are associated with domestic waste, such as biochemical oxygen demand (BOD), total suspended
solids, phosphorus, and nitrogen compounds.
Non-conventional pollutants include toxic substances that may be indirectly discharged from industrial facilities or,
to a lesser extent, as part of household hazardous waste disposal These toxic substances include metals, volatile
organic compounds, and other halogenated organics. The International Joint Commission (IIC) reports that, as of
1985, the predominant toxic substances being discharged to the Great Lakes from municipal point sources included
zinc, lead, chromium, copper, nickel, and cyanide. In addition, the LIC reports that greater than 50 tonnes/year of
several organic compounds were discharged in 1985, including phenol, ethylbenzene, tetrachloroethane, and
tetrachloroethylene.4* Furthermore, although the IJC estimates that less than one too of PCBs is discharged from
municipal point sources per year, this substance poses a substantially greater risk than other compounds (see
discussion of PCB risks under Sections G and K). Therefore, PCS discharges may often pose a greater risk than
" Maalclpil FretrciUseil Tuk Force/litcmtloul loiat Conniuiot. 1969. A Review of Pietieatmeat Program* U the Gicat Lake*
Bui*.
Note: Tke IJC report! dui Ike di*ckirfe coaceaaalioa vilies ikoald be regarded at "order-of-migmitade" cttiaiaiet.
-------
Page m- 25
larger volume discharges of other toxic substances.
Controlling these toxic substance discbarges is primarily accomplished through pretrcatment programs which require
indirect dischargers to reduce toxic concentrations in their effluent. Pretrcatment programs were designed to address
four concerns within the sewage treatment system: interference with the effective operation of the system; pass
through of toxic contaminants to receiving waters; contamination of municipal sludge by toxic substances; and
exposure of treatment plant workers to chemical hazards. As of 1989, the LJC reported that over 95 percent of the
pretreatment programs at individual United States Great Lakes municipalities were approved by U.S. EPA.
Nonetheless, the IJC found that although the framework; for an adequate pretreatment program is in place, further
resources at the municipal and State level are necessary to adequately trade and enforce pretreatment programs. In
atktiH"", the failure of some municipal governments to prosecute pretreatment program violators should be rectified,
and multijurisdictional variances in requirements and procedures among municipalities resolved.49
Number of Municipal Sources and Pretreatment Programs
There are 314 major municipal dischargers in the Basin who discharge approximately 6.91 billion gallons of treated
wastewater per day. In addition, approximately 2729 minor municipal dischargers are found within the Basin.44
Under 40 CFR Fart 403, all POTWs or combination of POTWs operated under the same authority with total design
flows of greater than five million gallons per day must establish a pretreatment program as a condition of the
National Pollutant Discharge Elimination System permit for the municipal point source. As of 1988,192 approved
programs were in place for dischargers within the Basin. The number of indirect dischargers covered under the
programs varied. For example, New York reported that, within the entire State, pretreatment programs in 56
municipalities were in place which affected more than 1,500 indirect dischargers. Wisconsin reports approximately
450 indirect dischargers are regulated under its program.45
Pollutant Loadings in the Basin
The load of conventional pollutants in the VS. portion of the Great Lakes Basin is tracked through regular reporting
to EPA's Permit Compliance System (PCS). The average annual discharge flow and mass loading of BOD and
nitrogen-ammonia from municipal point sources in each of the five lake basins for 1990 is as follows:
° Grcit Lake* W»«er Qulity Boant/Iaienafioaat Joiat Coamiaaioa. 1989. 1989 Report oa Great Lake* Water Q*aUty.
44 U.S. EPA Permit ComplUacc System dab.
" Msalcipal Pietreatmeat Task Foicc/Iatenatloaal Joist Commluioi. 1969. A Review of Pretreatmeat Program* la die Great lakes
Basis.
-------
Page m - 26
Exhibit B-l • 1990 Municipal Mass Loading
Annual Flow
(million
gallons)
BOD4*
(kilograms)
Total
Phosphorus
(kilograms)
Nitrogen-
Ammonia
(kilograms)
Total
Mercury
(kilograms)
Lake Erie
1,008,127
35,877,087*
10361,012
25,757,489
116
Lake Huron
79,623
3,211,858
193,711
379,030
S
Lake Michigan
376,467
12,859317
717,284
4,044,107
259
Lake Ontario
59,248
15,045,421
570,841
27,637
16
Lake Superior
23,247
763,894
52,433
48,232
0
Basin Total
1,546,713
67,757,579
11,895,283
30,256,497
396
The above summary (Exhibit H-l) addresses loadings to each of the Lakes from municipal point sources. The data
indicate that lake Erie receives both the greatest volume of municipal point source discharge, and the largest BOD
phosphorus and nitrogen-ammonia mass load. This is largely because the Detroit WWTP, the single largest
municipal discharger in the Great Lakes basin, discharges to the Lake Erie basin. Lake Michigan receives the Cf>/>^nd
greatest discharge volume and mass load of BOD, phosphorus and nitrogen, as well as the greatest mercury load.
Lake Superior appears to receive the lowest level of discharges.
The annual loading of selected toxic pollutants to the Great Lakes from Canadian and U.S. municipal point sources
was estimated by the LJC based on data drawn from a sample of POTWs. These data were used in conjunction with
a predictive model and estimates of contaminant removal efficiencies achievable in different treatment plant
to estimate concentrations of selected contaminants in raw sewage, treated effluents, and municipal facility sludge
These Basin-wide estimates are as follows in Exhibit H-2.
Overall, the volume of conventional pollutant loadings to the Basin exceed the volume of non-conventional or tr>rie
pollutant loadings. Nonetheless, loadings of toxic constituents may foster more severe impacts oo human heait^ ^
ecological receptors within the Basin.
" repraeati load* Cram facilities reporting either carboaaceoas BOD or BOD 5-day.
" The Nltrogea-ammoaia loadiag totals for Lake Erie iacladc a reported »aaa load of 21,654,155 kilogram boa the Detroit lUvor
-------
Page III - 27
Exhibit H-2"
% of Total
Release via
Effluent
% of Total
Release via
Atmosphere
% of Total
Release via
Sludge Disp.
Total Release
(tonnes/yr)
Arsenic
66%
NS
34%
19
Cadmium
76%
NS
24%
26
Chromium
49%
NS
51%
640
Copper
43%
NS
57%
30
Cyanide
73%
NS
27%
89
Lead
59%
NS
41%
580
Mercury
44%
NS
56%
3
Nickel
79%
NS
21%
130
Zinc
51%
NS
49%
1300
Benzene
61%
39%
NS
2
Toluene
45%
55%
NS
42
Etbylbenzene
31%
69%
NS
55
Chloroform
70%
30%
NS
34
Tetrachloro-ethyteoe
36%
64%
NS
76
Ttichloro-
ethylene
58%
42%
NS
26
1,1,1-
tricfaloroetbane
49%
51%
NS
76
Hcxachloro-
beozene
NS
25%
75%
<1.0
PCBs (total)
50%
NS
50%
<1.0
Phenol (total)
94%
NS
6%
85
N!> ='not significant
* Mnicipal Pietreatmeat Tuk Force/Uter»«tlouI Jo lit Commiuioi. 1989. A Review of Pietnt&neit Programs U Ike Gteat Lakes
B»»ii, p. 3.
-------
Page III - 28
Sipnificanoc of Combined Sewer Overflows in the Basin
An additional source of municipal effluent in the Basin is derived from CSOs. Combined sewer overflows may
discharge a variety of «wtqminnntR collected in urban storm water runoff, as well as conventional and toxic pollutants
that by-pass municipal wastewater treatment during periods of peak storm flow. The precise impact of CSO
discharges on Basin-wide water quality is unknown at this time. However, in a 1981 study, the IJC Task Force oo
Urban Drainage and Combined Sewer Overflows concluded that the loadings of BOD and phosphorus from urban
drainage and overflows in a municipality were about an order of magnitude less than loadings from a wastewater
treatment plant serving the same municipal area. With the decline in municipal point source loadings brought about
by improving treatment processes throughout the 1980's, the loads from CSOs may have become more significant.4*
Despite this lack of precise information on Basin-wide impacts, there is evidence that CSOs may have a
effect oo Great Lakes water quality. For example, in the 1990 Water Quality Inventory Reports prepared by the
Basin States under Section 305(b) of the Clean Water Act, the States report that at least 39 miles of Great Lakes
shoreline have experienced major impairment and at least 368 miles have experienced moderate or minor impairment
from CSO discharges.*0 In total, the eight Basin States reported that 1,834 river and stream miles and 153,102
acres of inland intrrg have undergone major impairment from CSO discharges."
The local significance of CSO impacts may also vary regionally. For example, the City of Sault Stc. Marie,
Michigan has a combined storm and sanitary sewer system with ten CSOs. A study completed in 1978 indicated
that there were oo adverse impacts from these CSOs on river water quality, although the presence of the CSOs
suggests the likelihood of sporadic storm water loadings." In contrast, CSO loadings to the Detroit and Rouge
Rivers have been found to be significant. There are 188 CSO outfalls on the Rouge River and 243 CSOs discharging
to the Detroit River either through direct discharge or through tributaries.0 In the 1970s, Detroit CSOs accounted
far 13 percent of the total phosphorus, 15 percent of the suspended solids, 21 percent of the oil and grease, 25
percent of the cadmium, 29 percent of the chromium, 20 percent of the copper, 32 percent of the lead, 96 percent
of the mercury, and 34 percent of the total PCS loading to the Detroit River.54 CSO loadings to the Niagara River
watershed have also been found to be significant The 16,000 foot long Falls River tunnel was constructed in the
early 1900s as a combined sewer for Niagara Falls, New York. It later became a major contributor of toxic load
to the Niagara River. In 1985, the City eliminated dry-weather sanitary and industrial flows to the tunnel; however,
because the tunnel is unlined, storm water continues to be discharged directly to the Niagara River.
Human Health Impacts, Ecological Risks, and Welfare Effects
For a discussion of health, ecological and coooomic risks see Section G.
* Great likes Water Qeallty Board/Iateraatioaal Jolat Comniuioi. 1989.1989 Report oi Gnat lakes Water Qaality, p. 29.
* The reader shoald aoie that these Ogares may sigsificaatly asdeiestimate Great Lakes shoreUae leaps Irmeat, as oaly fotr Biaia
Stttes • Mickigsa, New York, Ohio, aad Peaasylvaala - reported Oml Likes lapaimeat data.
51 These total* iadade all areas of the State, both 1 aside aad oatside the Great Lakes Bssia.
" US. Baviroameatil Protectioi Ageacy aad Eaviroaaeat Caaada. 1988. Upper Gnat lakes Coaaectiag Chaaaeb Stady, p. 170.
* Peraoaal commiaicatloa, Joha HtagUad, VS. EPA/Reglos V-, aad U.S. Eaviroameatal Protectioa Ageacy aad Eaviroaaeai Caaada.
1988. Upper Gnat Lakes Coaaeetiag Chaaaele Stady, p. 513.
" Ibid.
-------
FLOW FROM MAJOR MUMCIPALAHD
INDUSTRIAL DISCHARGERS
SELECT COUNTIES
BBlOOl—910228
H—ajSL
SO'
,o^®'
p®'
««*
Co***
SI
-------
LOADS FROM MAJOR MUNICIPAL AND
INDUSTRIAL DISCHARGERS
SELECT COUNTIES
8810O1-91Q2I)
i
-------
,VP^V
p!&
,. 1*$^
-* t3S0$P&
«s8g***
V$%»*
So>
»o<*e
. ?®
OQ^
»\\«^
I.C®
-------
I. NONFOINT SOURCE LOADING TO SURFACE WATERS
Page m - 29
Problem Area Description
This problem area addresses contaminant load and surface water quality impairment from oonpoint sources. Where
possible, environmental degradation resulting specifically from such sources is addressed However, as noted
previously, many surface water impacts, such as pesticide bioconcentration in sport fish, cannot be differentiated from
among industrial, municipal, nonpoint, and atmospheric load sources. As a result, human health impacts from fish
consumption and drinking water intake are discussed in Problem Area G, Industrial Point Source Discharges to
Surface Waters. The risk specifically accounted for by nonpoint sources is differentiated there by weighting the risfc
by the total contaminant load attributed to nonpoint sources in the Basin.
As defined in this section, nonpoint sources include agricultural runoff and urban runoff which is not discharged
through a Combined Sewer Overflow or POTW outfall. Loading from atmospheric deposition is addressed in
Problem Area J. In addition, surface water contamination through ground water discharge is discussed in Aggregated
Ground Water, Problem Area T.
Magnitude of the problem
Under Section 319 of the Clean Water Act, the States were required to prepare assessments of the magnitude and
impact of nonpoint source pollution in their waters. Overall, nonpoint sources were found to be a significant source
of surface water contamination in the Basin States. For example, New York reports that 80 percent of the surface
waters classified as impaired are affected by nonpoint sources.*5 The following discussion outlines the 319 findings
for portions of the Great Lakes Basin found in three States: New York, Michigan, and Minnesota.
New York assessed nonpoint source impacts for the six main drainage basins within the Great Lakes watershed.5*
The results for each of these basins are summarized below:
¦ The Lake Erie-Niagara River drainage basin drains 2300 square miles of eastern New York, including the
metropolitan areas of Buffalo and Niagara Falls. This basin contains the largest concentration of heavy
industry in the State and over SO percent of the Basin contains agricultural land. Eighty-eight stream miles
and about 180 acres of surface waters are impaired by nonpoint sources in this basin, with the majority of
impairment regarded as moderate. The significant nonpoint sources identified in this basin include on-site
wastewater systems, urban runoff, and streambank erosion.
¦ The Lake Ontario drainage basin drains approximately 2450 square miles and includes the Syracuse and
portions of the Rochester metropolitan areas. Six stream miles, 33 acres of inland lake, and about 5130
square miles of Great Lakes bays are impacted by nonpoint sources. In addition, 3560 square miles of Lake
Ontario are moderately impaired by contaminated sediments. The major nonpoint sources in this are
agricultural runoff, on-site wastewater systems, and inactive landfills. landfills are addressed in the waste
problem areas.
" Note: The aoapolat *olicet Ideadfled by New Yoifc Sttie lactate Him addressed la other problem area* ia this Report, each at
atmaapkerlc depot!tiom aid coatamiaated aediaaeata.
M New York Stale Depaitmeat of Eavlroaneaul Coaaervatioa. Fcbnaiy 1989. Noapoiat Soarce Asaeasmeat Report.
-------
Page III - 30
¦ The Genessee River Basin flows across the width of western New York and drains into Lake Ontario. Ibis
Basin contains portions of the Rochester metropolitan area, but the majority of the Basin is characterized
by rural-agricultural land use. Sixty-five stream miles and 670 acres of inland lakes are stressed, threatened,
or impaired. Hie predominant nonpoint sources in this basin are agriculture, on-site wastewater systems,
and landfills.
¦ The Seneca-Oneida-Oswego drainage basin drains about 5120 square miles in central New York, and
includes some 200 square miles of inland lakes. Agriculture is the major land use in the basin fallowed
by woodlands. Approximately 91,900 acres of inland lakes are impaired by nonpoint sources, with the
predominant source being agriculture. In addition, 54 stream miles are impaired by agriculture, urban
runoff, and landfills. Much of this impairment is judged to be severe.
¦ The Black River basin drains an area of about 1900 square miles, with much of the basin being heavily
forested and sparsely populated. Acid rain is regarded as the dominant cause of impairments in this basin,
with over 5,000 acres of lakes severely impacted. The impairments range from poor fishing to low fish
propagation and survival rates. On-site wastewater systems and agricultural runoff are also identified as
sources in this basin.
¦ The St. Lawrence River Basin drains almost 5540 square miles of forested, sparsely populated land. Acid
rain is judged to be severely impairing approximately 160 inland lakes in this basin, with impairments
tanging from low fish propagation to poor survival rates. On-site wastewater systems are also regarded as
a problem in this basin.
The State of Michigan's Nonpoint Source Assessment Report indicates that over 80 percent of the State's 297
watersheds are impacted by nonpoint sources.57 The predominant nonpoint sources of concern include septic
systems, streambank erosion, agricultural erosion, construction, animal wastes, agricultural fertilizers, and urban
runoff. With regard to impacts in the Great Lakes, Michigan reports that persistent toxic substance loading from
nonpoint sources is a growing concern. Toxics loading from stormwater discharges has been documented in the
State, but the impact on the open lakes is unclear. Michigan also regards phosphorus loading as an important
concern in the Basin and identifies conservation tillage as a proven technique to control phosphorus loading from
agricultural
Minnesota's Nonpoint Pollution Assessment Report identifies agricultural runoff, animal feedlots, pesticide and
fertilizer application, urban runoff/infiltration, construction, oo-site sewage systems, hydrologic modifications, forestry
mining runoff, highway runoff, and special erosion problems as important nonpoint sources of pollution.1* The
northern lakes and forest ecaregkm of the State contains much of Minnesota's Great Lakes drainage basin. This
region contains approximately 5550 lakes, most of which are 100 to 550 acres in size. Although the great majority
of the lakes in the region either fully support or partially support swimmablc uses, 15 lakes were regarded as not
supporting their uses. Ten of these lakes are reportedly impacted by pockets of agriculture, feedlots, pasturing near
the waterbodies, on-site septic systems, and ditching which may affect the quantity and quality of runoff to the lake.
In addition to these findings, studies conducted at Heidelberg College, Ohio, show that rural nonpoint source
pollution in the Lake Erie basin delivers not only large sediment loads through area rivers, but also both particulate
and soluble phosphorus derived from agricultural sources. Also, nitrate concentrations from agricultural runoff
" Micktgw Depeitneit of N*t*r*l Reaoefcee. November 1988. Mickigu'i 1968 Noipoial Poll«tk>i AmcsmdmI Report.
* Miuesou Pollstioi CoBtiel Afeiey. September 1988. Miiietou Noapoial Soiree PoUitioa Aaaemaeet Report
-------
Page m - 31
frequently exceed drinking water standards in basin rivers."
In sum, the impact of nonpoint sources appears to vary across the Great Lakes Basin with greater impacts associated
with areas encompassing more intense agricultural land use, urban development, and residential land use relying on
on-site wastewater disposal systems. As demonstrated in the above States, agricultural runoff is recognized as a
major nonpoint source of nutrient loading to surface waters. Exhibits 1-1 and 1-2 show the relative distribution of
cultivated cropland and of cropland erosion by section of the Basin. Exhibits 1-3 and 1-4 show the relative
Distribution of pastureland and pasture erosion by those same watersheds. I-S depicts the geographical distribution
of highly erodable land.
GREAT LAKES BASIN LAND USE
I CUUWKfSDCROFUY®
W Lata I
BLataSuaatar
NWUauSgn
aWLataMfcttw
BE Lata Mkhoan
NELataMkNgv
NWLataHuran
BWLMlim
Si Cta-Oam* Mv
WIMEn
8 Lata Eta
ELataEn*
SW Lata Ontario
8E I
NEL
SOURCE. NR. 1M7
Exhibit 1-1
GREAT LAKES BASIN LAND USE
w Lata Smarter
|
B LataSi^atar
1
MW Lata Michigan
tall I ¦
¦¦¦
wv lam uongan
8E Lata* Mkahigan
NE Lata* Mtahigan
¦¦
NWLtaahuran
1
9N Lata* Huron
¦
•iCttrOatfoftR*
¦¦
W Lata* Eft*
HHHHHHHi
8 Lata* Eft*
¦¦m
E Lata* Eft*
9N Lata* Onlaho
*Q0«MB
8£ Lata* Ontario
NE Lata* Oraaho
1
5 10 15 20 2fl
CE: NR. 1M7
Exhibit 1-2
GREAT LAKES BASIN LAND USE
IftWUflE
iiwiaiu wm
SMTLataUtMpn
BELitaMkttgan
NE LataMkMgan
NWUtoHnn
8W Lata Huron
Si Cta-Oata* Rtv
W Lata Eta
S Lata Eta
ELataEta
SW Lata Ontario
SOURCE: NR 1H7
Exhibit 1-3
GREAT LAKES BASIN LAND USE
I wtuhectowqn
Si CWr-Dauo* Mv
W Lata Eta
8 Lata Eta
ELataEta
SOURCE: NR, 1M7
Exhibit M
* Baker, D.B., "Overview of Rani Noipoiat Poiiatioa ia the Lake Erie Basia," Log*a, Terry J., et aL, eda., Effect* of Coa»eividc>a
Tillage oa Groaadwaler Quality: Nitrate* aad Peaticlde*. chapter 4, Lewis Pabliakera, lac., Chelae*, Ml, 1987, p. 8L
-------
Exhibit 1-5
TOTAL HIGHLY ERODIBLE LAND
Great Lakes Basin Counties
I
Lake Superior
Lake Huron
Lake t
Michigan
ACRES
0- 1,000
1,001 - 5,000
5,001 - 15,000
15,001 -30,000
30,001 - 60,000
> 60,000
A
EPA
Source: National Resources Inventory, 1982
Soil Conservation Service
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Page HI - 32
Conservation tillage practices are currently being promoted as a way of reducing agricultural runoff in the Great
Lakes Basin. There are four main types of conservation tillage practices: no-till, mulch-till, ridge-tin, and strip-
till.'0 However, conventional tillage practices are currently used far more within the Basin than nationally, with
conservation tillage practices being used on approximately 27 percent of the Basin cropland. Exhibit 1-6 pn nh mi
the percentage of total land in each of the Great Lakes Basin counties currently being used for cropland. p»hibit
1-7 presents the percentage of the cropland within the counties that is currently in conservation tillage. Mulch-tin
is the most frequently used of the conservation tillage practices, accounting 68 percent of all conservation nii~y
acres.41 No-till is used on 24 percent of all conservation tillage acres, and ridge-till and strip-till have only minor
applications.
The impact of these conservation-tillage practices on Great Lakes Basin water-quality is difficult to judge; however
recent research supported by U.S. EPA in the Lake Erie Basin found no significant differences in runoff, tile flow*
and pesticide losses between no-till and fall plowing on test plots.*2 As depicted in Exhibit 1-8, the corn-producdng
areas in the Basin that could benefit from these conservation tillage practices include northern Ohio, western
Michigan, and western Wisconsin.
Human Health Impacts
The likely human health impacts from nonpoint source loadings to the Great Lakes Basin are also based upon the
analysis developed under Problem Area G, above. As with the previous sections, the risk associated to contaminants
exposure is apportioned there to the various sources of loading to surface water in the Basin based on the relative
magnitude of that loading for selected toxic substances.
Ecological Risks
The impact from these pollution sources is difficult to quantify, although various studies in the Basin have
documented localized impacts from nonpoint sources. In general, many of the impacts from these sources ace
identical to those discussed in the previous two problem areas. Nonpoint sources contribute to nutrient, BOD,
organic, and toxic inorganic load. Therefore, eutrophication, dissolved-oxygen depletion, and toxic impacts, such
as bioaccumulation and reduced species propagation and diversity are all associated with nonpoint sources.
* Pnickeviciu, P., K. Schroer, ud B. Maaac-1989. U.S. Africa Haul Tillage Fraction (a die Great Lakes Basil, 1988. Great Lakes
Natknul Program Office.
* Ibid.
c Lops et al. 1989. u reported It Praackeviclaa, F, el al. 1989.
-------
EXHIBIT 1-6
Lake Superior
Lake
Michigan
PERCENTAGE
~ « 15
16-29
30 - 44
45 - 59
PERCENTAGE CROPLAND OF
TOTAL COUNTY AREA
Lake Huron
Lake Ontario
Source: Conservation Technology
Information Center, 1988
-------
r.
PERCENTAGE
~ 0-20
20 - 40
40 - 80
60 - 80
80 -100
TOTAL CROPLAND IN
CONSERVATION TILLAGE
Lake Huron
Lake Ontario
Source: Conservation Technology
Information Center, 1988
-------
J. ATMOSPHERIC LOADINGS TO SURFACE WATER
Page III - 33
Problem Area Description
Atmospheric loading to surface water occurs when air emissions of toxics settle onto surface water or become
incorporated in weather systems and are then precipitated out Atmospheric deposition of particles and gases occurs
in three ways:
¦ precipitation scavenging via rain or snow;
¦ dry particle deposition, which generally occurs closer to the point of emission; and
¦ vapor transfer of gases between the water and air phases.
In addition, atmospheric deposition in an upstream lake (e.g., Superior) can lead to indirect atmospheric loading in
downstream lakes (i.e, Huron, Erie and Ontario).
Sources
Although we have no data on the specific sources of toxic emissions that reach the Great Lakes through atmospheric
deposition, the following types of emission sources contribute to atmospheric deposition:
¦ transportation sources, especially cars, trucks, airplanes, and trains (lead and B(a)P);
¦ industrial sources (mercury, PCBs, HCB, and mirex); and
¦ agricultural sources, including volatilization and suspension of agricultural chemicals (organochlorine
pesticides).
Two studies of toxic air pollution in local areas within the Great Lakes Basin have identified steel/coke
manufacturing, highway vehicles, other manufacturing, and fuel combustion sources as the primary sources of toxic
air pollutants that pose cancer risks from inhalation of ambient air." The contaminants from these sources might
also contribute to risks through deposition into the waters of the Great Lakes. Airborne toxics can be transported
over very king distances in the atmosphere. Therefore, sources of pollutants that pose risks in the Basin may be very
far upwind of the Basin and sources within the Great Lakes Basin may contribute to problems in distant downwind
locations (i.e., in the northeast U.S. and southeast Canada).
Pollutants
Available monitoring and emissions modeling data suggest that a number of types of pollutants are introduced to the
waters of the Great Lakes through atmospheric deposition. These pollutants include PCBs, metals (lead, cadmium,
zinc, mercury," arsenic, and others), polycyclic aromatic hydrocarbons (PAHs), pesticides and their byproducts
(DDT, dicldrin, cblordane), non-metal inorganics (nitrate, sulfate, chloride, and phosphorus), and other toxic air
0 EPA. 1989. "Eatimitioa imd Ev»l»»tioa of Caaccr Risk* Alirfbitod to Air PoUtttoi ia Soatheut Chicago." Air tad Radiitioa
DivUioa, US. EPA Regioa V; a ad Eagiaeeriag-Scieace, lac. 1990. "Traaaboaadaiy Air Toxica Stady.* Fiaal Saamaiy Report Prepared
tor All aad Radiatioa DivWoa, US. EPA Regioa V.
" Receat atadic* ahow that mercary to aa iacieaalagly aigaiflcaat chemical of coacera for Ac apper Gnat Lakea Baaia, moat Ukely
tkroagk atnoapkeric depoaiUoa. For more iaforaatioa, ace Appeadix E
-------
Page m - 34
pollutants (gasoline vapors, coke oven emissions, volatile organic compounds, and others). These toxins can pose
human health risks via ingestion of drinking water and contaminated fish and can cause disease or death in ffeh
other aquatic organisms.
The magnitude of atmospheric loadings to surface water is dependent upon the amount of pollutants released
the air, the direction and speed of atmospheric transport, and the Me of toxins in the atmosphere. A 1988 report
on the Great Lakes Atmospheric Deposition network provides the following information on atmospheric fin^ca of
cadmium and lead to the lakes.*5
¦ Monitoring data for 1983 indicate highest deposition rates for cadmium occurring along the central portion
of the southern shore of Lake Ontario (near Rochester, NY) and at the far eastern and western ends of T
Erie (near Buffalo, NY and Toledo, OH, respectively). Except for an area of high cadmium flux in the
center of Lake Michigan, cadmium fluxes to the upper Great Lakes are low relative to the fluxes to t -i^
Erie and Ontario.
¦ Monitoring data for 1983 indicate highest deposition rates for lead at the southern cod of Lake Michigan
(near Chicago, IL and Gary, IN), at the central portion of the southern shore of Lalce Ontario (near
Rochester, NY), and at the eastern end of Lake Erie (near Erie, PA and Buffalo, NY). Lakes Superior
Huron and the northern end of Lake Michigan receive low fluxes of lead relative to the other areas of the
Great Lakes.
A recent modeling study of emissions, atmospheric transport, and deposition in the Southeast Michigan/Windsor area
indicates that a number of air pollutants emitted in this area are deposited to Great Lakes waterways in signify^
quantities.4* Contaminants deposited in the largest quantities include gasoline vapors, benzene, coke oven
emissions, styrene, and perchloroethylene.
In addition to these absolute estimates of contaminant loadings, researchers have estimated the relative important
of atmospheric deposition as a source of contaminants to the Great Lakes. The following table shows the relative
magnitude of (1) direct atmospheric deposition of PCBs, lead, and benzo(a)pyrene, (2) indirect atmospheric
deposition that is from the outflow flux of upstream Great Lakes, and (3) the remainder of the flux into the lai^,
that is assumed to be from non-atmospberic sources. The table illustrates the following points about the relative tofe
of atmospheric deposition of toxics in each of the five lakes.
¦ The upper Great Lakes of Superior, Michigan, and Huron receive almost all of their influx of PCBs fioia
direct and indirect atmospheric deposition. In contrast, Lakes Erie and Ontario, receive the majority of their
PCB inputs from non-atmospheric sources. This difference is due to the importance of industrial sources
on the lower lakes and the connecting channels (ix., the Detroit and Niagara rivets).
¦ Atmospheric deposition is the dominant input pathway for lead and B(a)P into all the except for lead
inputs into Lake Erie. (Atmospheric lead loadings have markedly decreased over recent years, however
because of government controls resulting in the declined use of leaded gasoline in automobiles.)
• Great Lalua Nitioaal Program Office. 1988. Great Laket Atmotofceric Pcpo»ltk>» «jLAD\ Networtj. 1982 OA EPA,
Chicago, IL EPA 90S 4 88 002.
** Op. di Esgtacertag-Sckace, Uc. 1990.
-------
Page III - 35
Percent Contribution of Atmospheric Deposition to Contaminant Loadings in the Great Lakes
Lake
PCB Loading by Pathway
(Perceat)
Lead Loadiag by Pathway
(Perceat)
B(a)P Loadiag by Pathway
(Perceat)
Atmospheric
Noa-
Atmospheric
Atmospheric
Nob-
Atmotpheric
Atmospheric
Noa-
Atmoipheric
Direct
la direct
Direct
la direct
Direct
Iadirect
Superior
90
0
10
97
0
3
96
0
4
Michigan
58
0
42
99 S
0
0.5
86
0
14
Htroa
63
15
22
94
4
2
63
17
20
Erie
7
6
87
39
7
54
66
13
21
Ontario
6
1
93
50
23
27
40
32
28
Source: Summary Report of the Workshop on Great Lakes Atmospheric Deposition. International Joint
Commission, Science Advisory Board/Water Quality Board/International Air Quality Advisory
Board, October 1987, Windsor, Ontario. (Noo-atmospberic loading by difference.)
Most of the toxins of concern are resistant to degradation in the atmosphere, and we do not discuss loss through
degradation in the atmosphere. The magnitude of the problem is also affected by the water flux among the Great
Lakes, allowing the water contaminated from atmospheric inputs to spread throughout the five lakes.
WET DEPOSITION LOADINGS TO GREAT LAKES WET DEPOSITION LOADINGS TO GREAT LAKES
LEAD NITRATE
METRIC TONS (THOUSANDS) METRIC TONS (THOUSANDS)
4, — —— 3601
1882 1983 1984 1985 1986 1982 1983 1984 1985 1986
8QURCE: QLA0 NETWORK
Ei hi bit J-l Exhibit J-2
Monitoring data from the Great Lakes Atmospheric Deposition (GLAD) Network provide estimates of the rates of
atmospheric loading of metals and some other inorganic pollutants to the waters of the Great Lakes over the years
1982 through 1986. These data for lead, nitrate, sulfate, chloride, and total phosphorus arc summarized in Exhibits
J-l through J-S. Other monitoring data provide estimates of the deposition rates of lead, zinc, and chromium into
-------
Page HI - 36
WET DEPOSITION LOADINGS TO GREAT LAKES
SULFATE
THOUSANDS OF METRIC TON8
° 1982 1983 1984 1885 1086
SOURCE: QLAO NETWORC
the waters of each of the five Great Lakes.47 The findings
of this study are summarized in Exhibit J-6. This study
indicates that total lead loadings from the atmosphere in
1986 range from a low of 1S3 metric tons in Lake Ontario,
to a high of 263 tons per year in Lake Michigan. 7ifK-
loadings (1986) from the atmosphere are estimated to range
from 184 netric tons per year in Lake Ontario to 460
metric tons per year in Lake Michigan. This study
estimates that cadmium loadings from the atmosphere in
1986 ranged from 2.7 metric tons per year in Lake Ontario
to 11.1 metric tons per year in Lake Superior. A 1988
report on the GLAD Network provides the following
information on the spatial distribution of atmospheric fluxes
of and lead to the lakes.**
Exhibit Jf-3
WET DEPOSITION LOADINGS TO GREAT LAKES
CHLORIDE
METOC TONS (THOJBAN08)
200
ibo
100
.
—
1
—
,
| ¦:•:•••. ¦
1984
1985
1986
SOURCE: QLAD NETWORK
lifcibiuT
WET DEPOSITION LOADINGS TO GREAT LAKES
TOTAL P
METT1C TOWS (THOUSANDS)
1982 1983
SOURCE: QLAD NETWOfK
Exhibit J-5
1984
1985
1988
" Re Id, KS„ Tug AJS., Lasis, M.A., Klappeabach, EW. 1989. "The Depositioa of Meuls ia OmUrio." Symposia* oi Metab
Deposition, Air Poll¦ boa Coatrol Asaociatioa, Research Triaagie Park, Spriag 1989.
• Great Lake* Nitioaal Program Office. 1988. Great Lakes Atmospheric Depositioa (GLAD1 Network, 1982 a»d 1983. U.S. EPA,
Chicago, IL. EPA 905 4
-------
Page in - 37
Exhibit J-6
Estimated Rates of Metal Deposition to the Great Lakes
(1986 data, metric tons)
Lead
Zinc
Cadmium
Ontario
153
184
2.7
Erie
214
199
3.8
Michigan
263
460
10.0
Huron
244
344
6.8
Superior
204
274
11.1
Source: Reid, N.W., Tang AJ.S., Lusis, M.A., Klappcobach, E.W. 1989. "The Deposition of Metals in
Ontario."
For Human health impacts, ecological impacts, welfare effects, see Problem Area G, Industrial Point Source
Discbarges to Surfaoc Waters.
-------
Page m - 38
K. TOXIC SEDIMENTS
Problem Area Description
This problem area is usually considered a component of nonpoint source loadings to surface waters. However, to
emphasize the complexity and significance of sediments as both a source and pathway for contamination in the Great
Lakes Basin, this section deals separately with this issue. Bottom sediments in a number of areas of the Great Lakes
Basin arc contaminated with toxic metals, ammonia, and organics. Sediments have been identified as "the major
reservoir of pollutants in the Great Lakes system."*' The toxic sediments problem area addresses the risks to human
health and the environment generated by this contamination.
Pollutants that have accumulated in bottom sediments include: lead, mercury, iron, cadmium, chromium, nickel, Tir^
barium, copper, phosphorus, arsenic, cyanide, PCBs, polynuclear aromatic hydrocarbons (PAHs), hexachlorobcnzeoe
(HCB), chlordane, phenols, DDT and metabolites, phthalate esters, oil and grease.10 The magnitude of the tnyfr
sediments problem varies with location. In some areas contamination is so severe that sediments have been ci»«jf)F
-------
Page m - 39
supplies.72 Other risks may be created by the inadequate loogtenn CDF performance, resulting in further
bioaccumulation of contaminants through plants and animals in or near CDFs.73 The physical degradation of water
habitats caused by such dredging activities is discussed in Problem Area A, Physical Degradation of Water Habitats
and Wetlands.
The U.S. EPA's Great Lakes National Program Office (GLNPO) is responsible for a multi-organization program
entitled Assessment and Remediation of Contaminated Sediments (ARCS), scheduled for completion in 1992. The
ARCS Program will characterize, sample and map sediment deposits; evaluate sediment toxicity; assess abnormalities
and bioaccumulation in fish; assess/model risks and hazards to human health, aquatic life, and wildlife; and evaluate
remediation technologies in its five study areas. The ARCS Program will concentrate on five geographic locations:
Saginaw Bay, Sheboygan Harbor, the Grand Calumet River, the Ashtabula River, and the Buffalo River.74 (See
Exhibit K-l.) While the results of the ARCS Program should provide much clarification as to the extent of the
contaminated sediment problem in the five study areas, data is yet to be released.
Human Health Impacts
This section presents an overview of the potential human health risks posed by contaminated sediments within the
Great Lakes Basin. The risk assessment approach described in this paper is drawn primarily from work done by U.S.
EPA in a baseline human health risk assessment for the Buffalo River, New York.75 Study areas for this section
were selected to represent sites located across the Great Lakes Basin. In addition, these sites were determined to
have sufficient data to at least partially support the risk assessment. All of these study areas have been designated
as Areas of Concern. For brief descriptions of these and other AOCs, the reader is referred to Appendix E.
The study areas have all been investigated for the presence of a wide variety of organic and inorganic contaminants
in resident fish populations. Nonetheless, previous studies have determined that human health ride from exposure
to contaminated sediments fish populations are predominately a function of PCB, and to a lesser extend, DDT, and
mercury concentrations. For example, the Buffalo River Baseline Human Health Risk Assessment found that PCB
levels accounted for more than 75 percent of the non-carcinogenic risk and more than 95 percent of the carcinogenic
risk. The predominance of PCB related risk was also confirmed in the analysis of surface water implants in EPA's
Great Lakes Basin Risk characterization study. As a result, this analysis was focused on potential impacts from
levels of PCB, DDT, and mercury in the study areas.
The ambient sediment PCB, DDT, and mercury levels in the study areas were determined for this analysis based on
data gathered as part of Remedial Action Plan investigation and other ambient water quality studies. In general, the
mean sediment concentrations presented in Exhibit K-2 were determined by averaging the most-recent year of
available sediment data for all sites sampled within the study areas.
These summary data vary in terms of the number of sites included in the average, the year in which the data were
collected, and the reliability of the data as described in the subject reports. Furthermore, do attempt was made to
compare the sampling or analytical procedures used to generate the data. As a result, the reliability of the data used
n VS. EPA, Gnat Lake* Nadoaal Pro grim Office. 1990. "Aaaeaameat aid Remediatioa of Coatamiaaied SedimeaU (ARCS) Work
PUa". p. 24.
" Cheat Lakes Commisaioa. "Great Lakes Commercial aid Recieatioaal Harbor Dredgiag: Iaaica a ad Recommeidatioma".
* US. EPA, Gnat Lake* Natioaal Program Office. 1990. "Aasesameat a ad Remediatioa of Coatamiaaied Sedimeata (ARCS) Woik
Pitt".
71 US. EPA, 1990 (Draft). Baaeiiae Hamaa Healtk Risk Aaacsameat: Baffalo River, New Y«t.
-------
Exhibit K-1
ARCS1 PRIORITY
AREAS OF CONCERN
LAKE SUPERIOR
LAKE
HURON
LAKE
ONTARIO
LAKE
ICHIGAN
5T CLAW
LAKE
ERIE
GREAT LAKES AREAS OF CONCERN
1. SHEBOYGAN HARBOR
2. GRAND CALUMET / INDIANA HARBOR
3. SAGINAW RIVER/BAY
4. ASHTABULA RIVER
5. BUFFALO RIVER
' Assessment and Remediation of Contaminated Sediments
M
0 SO 100 ISO 200
1 I I I I
KILOMETERS
&EM
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Page m - 40
for the analysis cannot be confirmed.
Exhibit K-2
Representative Sediment Concentrations for Study Anas
Concentration in Sediment, ppm
PCB DDT Hg
Buffalo River*'' 3.8 0.013 1.71
Duluth/St. Louis Rivet*' 0.1S ND 029
Green Bayc) 3.8 0.020 1.64
Rochester4 0.11 0.044 022
Saginaw*1 15 ND 0.082
Sheboygan0 9.4 ND 0.21
* Reported as 90th percentile sediment concentrations
Sources of Concentration figures
a) U.S. EPA, 1990 (Draft). Baseline Human Health Risk Assessment:
Buffalo River, New York.
b) Data report, personal communication, Mr. Frank Snitz, U.S. Army Corps of Engineers.
c) STORET data retrieval, personal communication, VJS. EPA Great Lakes National Program Office.
d) STORET data retrieval, personal communication, U.S. EPA Great Lakes National Program Office.
e) U.S. Army Corps of Engineers, 1988. Report oo Saginaw River, Detroit, Michigan.
f) U.S. Army Corps of Engineers, 1988. Remedial Investigation Advanced Screening Report: Sheboygan
River and Harbor.
-------
Page m - 41
Data summarizing PCB, DDT, and mercury concentrations in carp were also drawn from several different sources.
As cited following Exhibit K-3, these sources varied in the number of fish sampled and the year of analysis.
Nonetheless, these data represent the best available information concerning contaminant concentrations in fi«h
Exhibit K-3
Fish Contaminant Concentrations in Study Areas*
(ppm)
PCB DDT Hg
Buffalo River* 33 0.42 0.12
Duluth/St. Louis River*' 0.95
Green Bay1' 11.42 0.9
Rochester** 0.70 0.16 0.47
Saginaw** 14.0 1.5 0.04
Sheboygan1' 66.7
*Whole body concentrations for carp
Sources of Concentrations
g) U.S. EPA, 1990 (Draft). Baseline Human Risk Assessment: Buffalo River, New York.
h) Minnesota Pollution Control Agency and Wisconsin Department of Natural Resources, September 1990
Stage I, Preliminary Draft). The St Louis River System Remedial Action Plan.
i) Wisconsin Department of Natural Resources. Toxic Substances Management Technical Advisory Committee
Report, July 1987. Lower Green Bay Remedial Action Plan.
Source Unknown. "Fish Contaminants • Lake Michigan."
j) U.S. Army Corps of Engineers, 1987. Genesee River/Rochester Embayment Remedial Action Plan,
k) U.S. Army Corp6 of Engineers, 1990. Information summary, Area of Concern: Saginaw River and
Saginaw Bay.
1) U.S. Army Corps of Engineers, 1988. Remedial Investigation Advanced Screening Report: Sheboygan
River and Harbor.
-------
Page IE - 42
Human Health Risk Assessment
This section describes the likely exposure scenarios, exposure concentrations, close-related toxicities, and human
health risks associated with the contaminants of concern in the study areas.
Exposure Pathways
Exposure to the hazardous constituents found in contaminated sediments may involve a variety of direct and indirect
routes. These direct routes of exposure include dermal contact with sediments and direct ingestion of suspended
sediments. Indirect routes of exposure include dermal contact with constituents released from the sediments to the
water column, ingestion of such released constituents, and ingestion of fish that have bioconcentrated the constituents.
The Buffalo River Baseline Human Health Risk Assessment found that four exposure route were most important to
the determination of risk:
1) ingestion of surface water while swimming;
2) ingestion of fish;
3) dermal contact with surface water while swimming; and
4) dermal contact with sediments while swimming.
These four pathways are assessed in this analysis.
Exposure Scenarios
Determining the concentration of a constituent to which an individual may be exposed is a function of the following:
1) the constituent concentration at the point of contact; and 2) the frequency and duration of contract for different
age groups.
For the scenarios involving either direct contact with sediments or ingestion of fish, the contaminant concentrations
at the point of exposure have already been reported. However, determining exposure to contaminants in the water
column requires an estimate of the concentrations of constituents in the water body while at equilibrium with the
sediment contaminant "pool." Under this relationship, the concentration of organic contaminants in the water body
can be estimated using the following relationship:
Q-CyKrf + C. *TSS
where C, is the sediment concentration (mg/kg), TSS is the concentration of suspended solids (kg/1), and K* is a
partition coefficient. K* can be estimated using the following relationship:'*
K.-0.79 -K_ -C
where K^, is the octanoL-water partition coefficient and is the fraction of organic carbon.77 For the organic
contaminants of concern for this study, the K^, values are as follows:
* Valaea for K« tke aedlmeat-waier partitioa coetDcieai, are deUnnlaed for a petfcalar eoatamlaait aad atady aile at t faactloa of Ike
orgaalc carina fractioa la ike aedlmeat, Ike octaaol:water partitioa cocfQcicaC for Ike coatamiaaa! of coaceia, aad aa empirical valae
deiermiaed tkroagk obiervatioa. The empirical valae of 0.79 aaed la tkia aaalyab waa Ike empirical valte aaed to ealcilate tke water colimi
coaceatratioas waa takea from tke BaacUae Hamaa Healtk Riak Aaaeaameat for tke Baffato River, New York, op. cit.
n For a more detailed deacripdoa of tkia metkod, ace US. EPA, 1990 (Draft). BaacUae Hamaa Healtk Riak Aaaeaameat, Appeadiz D.
-------
Page LI - 43
for PCB: 1.1 x 10s
for DDT: 155 * 10*
Id addition, the TSS and values estimated far the study areas ate presented in Exhibit K-4. It is important to Wrrp
in mind that these values are generally means within wide ranges for specific locations only, and as such, they are
not necessarily representative of larger areas.
Exhibit K-4
Total Suspended Solid and Sediment Organic Carbon
Fractions for Study Areas
TSSfnwffl
Buffalo River 18 0.024*
Duluth/St. Louis River (18) 0.041**
Green Bay 12 0.026**
Rochester (18) 0.001"
Saginaw 30 0.020*
Sheboygan (18) 0.036*
() Values in parentheses indicate that no site-specific data were available for the study areas. As a result, values
for the Buffalo River were used as default values.
* Preliminary values from EPA Assessment and Remediation of Contaminated Sediments program.
** U.S. Army Corps of Engineers.
Based on the above relationship, the water column concentrations listed in Exhibit K-5 were derived. Exhibit K-6
was compiled to depict the mercury concentrations in the water column of there study areas. However, at all of the
study areas except the Buffalo River, no mercury concentrations were detected in the water or no data were reported.
The second component of the exposure scenario estimate focuses on such factors as the age of the population
exposed to the various pathways, the amount of time the individual typically remains exposed in a lifetime, the
anticipated intake rate for exposed parts erf the body, and the body weight of the typical exposed individual within
each age group. Based on these relationships a weighing factor can be determined to estimate the lifetime exposure
for each of the four pathways. These weighing factors are presented in Exhibit K-7*
* Ibid.
-------
Page III - 44
Exhibit K-5
Estimated Water Column Concentrations for the Organic Contaminants of Concern
(mg/l)
PCB DDT
2J5 x Iff4 6.8 x la7
6.9 x 10* ND
2.1 x 1
-------
Page HI - 45
Exhibit K-7
Lifetime Exposure Weighting Factors79
Weighting Factor
Ingestion of water while swimming
Water dermal contact while swimming
Sediment dermal contact while swimming
Ingestion of contaminated fish
1.58 x Iff'
3.79 x Iff5
725 x Iff*
928 x Iff5
Hie weighting factors used in this analysis encompass a variety of assumptions regarding the patterns of lifetime
exposure for each of the four pathways. Determining the wieghting factors essentially involves estimating three sets
of variables: 1) the chemical concentration at the point of exposure; 2) the characteristics of the exposed
population (e.g., frequency of exposure, duration, age and body weight of exposed individuals); and 3) the time
period over which the exposure is averaged. The general form of the equation used to determine the weighting
is:
The weighting factors used in this analysis were calculated by assessing values for each of the above values for age
classes of 0-6, 7-18, and 19-70.
Contaminant Intake/Dose Estimates
The anticipated lifetime exposure for each pathway, study setting, and contaminant can now be estimated by
multiplying tbe estimated exposure concentrations by the weighting factors controlling uptake. Hie lifetime intni^
dose for each study area and pathway are outlined in Exhibits K-£ through K-ll.
Intake ¦ (C*CR*EFD)/(BW*AT) where:
Intake - amount of chemical ingested or absorbed at tbe point of exposure;
C - chemical contraction, the avarage chemical concentration contacted over tbe exposure period;
CR - contact rate, tbe amount of contaminated medium contacted per unit time or event;
EFD - exposure frequency and duration, bow long and bow often exposure occurs;
BW - body weight, tbe average body weight of exposed individuals; and
AT • averaging time, period over which exposure is averaged.
* US. BP A, 1990 (Drift). Bwetiae Hnui Health Risk Aucstmeit: Baffalo River, New Ywi.
-------
Page m - 46
Exhibit K-8
Study Area Intake/Dose Summaries
Exposure Pathway: Ingestion of water while swimming
(mg/kg/day)
PCB
DDT
Hg_
Buffalo River
4.0 x icr*
1.1 x Iff10
2.7 x 10"*
Duluth/St. Louis River
1.1 x 10*
-
-
Green Bay
33 x Iff*
1.4 x Iff"
Rochester
1.9 x 10*
5.8 x 10 *
Saginaw
2.1 x 10*
-
-
Sheboygan
7.4 x Iff*
Exhibit K-9
Study Area Intalce/Dose Summaries
Exposure Pathway: Dermal contact with water while swimming
(mg/lcg/day)
PCB
DDT
_HS_
Buffalo River
9.5 x 1 ff*
2.6 x Iff"
6.4 x Iff*
Duluth/St. Louis River
2.6 x icrw
-
-
Green Bay
8.0 x Iff*
33 x Iff"
-
Rochester
4.6 x 1(T*
1.4 x Iff*
Saginaw
4.9 x 10"*
-
-
Sheboygan
1.8 x Iff*
•
•
-------
Page m - 47
Exhibit K-10
Study Area Intake/Dose Summaries
Exposure Pathway: Dermal contact with sediment while swimming
(mg/kg/day)
PCS
DDT
He
Buffalo River
2.8 x
lO"*
9.4 x Iff4
12 x
10"'
Duluth/St. Louis River
1.1 X
Iff*
-
2.1 x
Iff*
Green Bay
2J& x
Iff5
1J x 10-7
12 x
i(r5
Rochester
8.0 jc
Iff7
32 x Iff7
1.6 x
104
Saginaw
1.1 X
10s
-
6.0 x
ltr1
Sheboygan
6.8 x
Iff5
-
1.5 x
iff*
Exhibit K-ll
Study Area Intake/Dose Summaries
Exposure Pathway: Ingestion of contaminated fish
(mg/kg/day)
PCB
DDT
JJgL
Buffalo River
3.1 x Iff"
35 x Iff5
1.1 x Iff5
Duluth/St. Louis River
8£ x Iff5
-
.
Green Bay
1.1 x Iff*
8.4 x Iff5
-
Rochester
63 x Iff5
13 x Iff5
4,4 x Iff1
Saginaw
13 x Iff*
1/4 x Iff*
3.7 x Iff4
Sheboygan
&2 x Iff*
-
-
Human Health Impact Assessment
TWo measures of potential human health impact are used in this analysis. The first measures the risk of cancer
from exposure to each contaminant and is estimated using a cancer potency factor, which represents the risk of
contracting cancer from daily lifetime exposure, The second measure is known as a reference dose (RFD). The
RFD is the measure of chronic daily intake that will not result in adverse, oon-carcinogenic health effects. The
ratio of the exposure intake level and the reference is the measure of potential non-carcinogenic health
A ratio greater than ooe indicates the potential for adverse, non-carcinogenic health effects. Exhibit K-12
presents representative RFD and cancer potency factor values for the contaminants of concern.
-------
Page HI - 48
Exhibit K-12
Toxicity Values for Contaminants of Concern
Cancer Potency
RFD fmg/kg/davl factotfmg/kg/dav"1')
PCB
l^)x 10^
7.70
DDT
5.0 x 10-"
034
3.0 x ltr*
-
Estimating Human Health Impact
The potential for adverse human health impacts from exposure to the contaminants of concern in the study areas
is reflected in Exhibits K-13 through K-16. These exhibits present the reference dose retire and lifetime cancer
risks for the exposure pathways, study areas, and contaminants investigated in this analysis. Absent values in the
summary exhibits reflect data gaps in the source documents. No cancer potency values are reported for exposure
to mercury, because it is regarded as Class D, not classifiable as to human carcinogenicity.
Exhibit K-13
Lifetime Human Health Risk Summary
Exposure Pathway: Ingestion of water while swimming
ISfi £21 Ik
ktake/ Ctacer IiUke/ Ctacer IaUke/ Caiccr
RFD Riik RFD Risk RFD Risk
B*0ilo River
7.0 x 10*
3.1 x W
22 x 10'
3.7 x 10^'
9.0 x W
N/A
DilitV
St. Lo«i* River
LI * Iff*
MxXO4
-
-
-
N/A
Gteei Biy
3.3 x 10*
Z5110 '
Z8 x 10"T
4.8 x 10J1
-
N/A
Rocketler
1.9 x lO4
L5 x 107
L2x lC®
ZOx 10*
-
N/A
S«gii»w
il x 10-1
1.6 x 101
•
-
-
N/A
Sheboygu
7.4 x W
5.7 x 10'
-
•
-
N/A
N/A-aol ippli cable
-------
Page m - 49
Exhibit K-14
Lifetime Human Health Risk Summary
Exposure Pathway: Dermal contact with water while swimming
PCS PPT Hg
Intake/ Cuter Utakc/ Caacer lattke/ Caacer
RFD Rick RFD Rlik RFD Rfak
Bafialo River
Mul/
St. Lotii River
Creei Bay
Rochester
Sigiaaw
Skeboygii
N/A-aot applicable
Lifetime Human Health Risk Summary
Exposure Pathway: Dermal contact with sediment while swimming
PCB DDT at
la take/ Caacer la take/ Caacer latake/ Caacer
RFD Risk RFD Risk RFD Risk
BaSilo River
Dalaik/
Sl Loaii River
Gteea Bigr
Rochester
Sagiaaw
Skeboypa
K/A—101 applicable
9.51 10J
73 x 10*
5.2 x IO*1
8.8 x 10 u
21 x 10*
N/A
16 s 10*
zo x 10*
-
-
-
N/A
8.0 x 10'
ui io*
6.6 x 10*
LI x 10*'
-
N/A
4.6 s 10J
UxW4
2.8 x 10*
4.7 x 10*
-
N/A
4.9 x 10*
3£x 10*
-
-
-
N/A
L8x 10*
L4 x 10-T
-
-
-
N/A
Exhibit K-15
2.8 x 104
12 x 10*
L9x 10*
12 x 10*
4.0 x 10*
N/A
LI x 10J
Ui 10*
-
-
1.0 x 10*
N/A
28 x 10"*
22x10*
3.0 x 10*
5.1 x 10*
4.0 x 10*
N/A
8.0 x 10*
&2x 10*
6.4 x 10*
LI x 10*
5.3x10*
N/A
LI x 10 '
8^x10*
-
-
20 x 10*
N/A
0.8 x 10*
5.2x10*
-
-
5.0 x 10*
N/A
-------
Page III - SO
Exhibit K-16
Lifetime Human Health Risk Summary
Exposure Pathway: Ingestion of contaminated fish
PCB DDT Hg
Iittke/ Ciacer Uuke/ Cueer Iatakt! Caacer
RFD Riak RFD Riak RFD Riak
Bafblo River
3.1
24 x 10-*
7.8 x 10*
1.3 x 10*
3.7 x 10*
N/A
Dalitfc/
St LoaU River
8.8 x 101
6.8 x 10-"
-
-
-
N/A
Greea Bay
LI x 10 '
8.5 x 10"
2.8 x 10'
29 x 10*
-
N/A
Rochester
6.5 * 101
5.0 x 10"
3.0 x 10*
5.1 x 10"
1.5 x 10*
N/A
Sagiaaw
13
1.0 x 10*
4.7 x 104
4.8 x 10*
1.2 x 10*
N/A
Sfceboygu
62
4.8 x 10*
-
-
-
N/A
N/A-aol applicable
Across all of the study areas, ingestion of contaminated fish served as the greatest risk to human health. All five
areas have sport fish consumption advisories in place. The risks were greatest in the Sheboygan, Saginaw, and
Buffalo study areas. Risks from PCB exposure also far outweighed risks from DDT or mercury, although this may
also result from an absence of DDT and mercury data. Finally, RFD ratio values greater than one were found only
for the fish ingestion exposure pathway in Sheboygan, Saginaw, and Buffalo. This indicates that for these three areas,
lifetime non-carcinogenic risk is limited to a single exposure route. In contrast, cancer risks of greater than 1 x Iff4
are found in all of the study areas for one or more exposure pathways.
Establishing a direct link between the risk from fish consumption and contaminated sediments is difficult at best.
Since this analysis used carp in coming up with the risk, and carp are bottom feeders, we are probably getting the
most direct link to the sediments. However, it is probably incorrect to assume that all the risk from fish ties back to
the sediments. Sediments are not the sole source of PCBs to fish, but considering that carp were used to calculate
risk, the sediments were probably the source for a very large portion of the total PCBs, and thus in turn are a large
factor in the overall risk to the specific human populations who consume these fish.
Ecological Impacts
Study of the ecological effects of contaminated sediments has been very limited. Therefore the following discussion
of the ecological impacts of toxic sediments will be qualitative in nature.
Severity
There are documented cases of ecological damage in the Great Lakes Basin which are suspected to be due, at least in
part, to toxic sediments. For example, sediments in several areas have been shown to be toxic to biota, and some
areas of contaminated sediments are reported to be completely devoid of life. In Lake Ontario, poor reproductive
success and declines in colony size of herring gulls due to elevated levels of PCBs and mirex were documented in the
-------
Page in - 51
early to mid 1970s.*0 More recent data indicate a decline is organochlorine residue levels in herring gull eggs from
1974 to 1983, a decline in dioxin and dibenzofuran levels from 1981 to 1986, and an increase in the reproductive
rates of gulls to normal levels.*1
EPA has developed preliminary interim sediment quality criteria or screening level criteria values for 13 non-ionic
organic chemicals (see Exhibit K-5).c Maximum sediment concentrations of PAHs, PCBs, DDT, dicldrin, and/or
cblordane exceed sediment criteria in several of the AOGs in Lake Erie, several Lake Erie tributaries and harbora
outside the AOGs, and Lake Ontario sediments. In some Lake Eric AOCs, average sediment concentrations of thmr
compounds also exceed sediment criteria. Concentrations of PAHs, PCBs, and cblordane exceed sediment criteria,
even assuming 20% organic carbon content of the sediments. With the exception of Lake Ontario sediments,
concentrations of DDT and dieldrin exceed sediment criteria only when 2% organic content is assumed. These
indicate that contaminated sediments in the Great Lakes Basin are likely to result in adverse chronic effects aod/or
bioaccumulation in aquatic ecosystems.
While a cause-effect relationship is strongly indicated between contaminated sediments and neoplastic disease in fish,
no standardized assessment of carcinogenic potential in fish is available." The following is a list of organisms
known to have been affected by contaminated sediments: carp, benthic invertebrates, fathead minnows, aquatic
insects, bexagenia, and hydallella. It should be noted that these impacts include only effects due to direct contact
with sediments. Even less study has been performed on the impacts of bioaocumulation and possible releases from or
contact with CDFs.
Reversibility of Damage
Just as the extent of ecosystem damage is uncertain, so is the speed of ecosystem recovery (assuming the toxic
sediments were removed). However, it is known that in some areas the ongoing natural process of sedimentation win
eventually cause the burial of contaminated sediments with new material. These new sediments, while still
contaminated, tend to be less toxic because of recent reductions in contaminant loadings into the Great Lakes Basin.
Evidence of this process is seen in sediment core samples. In several areas, contaminant concentrations are highest at
a sediment depth corresponding to the 1960's." In other areas weather and internal lake processes have served to
distribute sediments in broad depositional Basins. In these areas (Lake Ontario for example) and in areas where the
natural rate of sedimentation is low, contaminant concentrations show little change sinoe the early 1970's." Except
for areas in which sedimentation rates are very high and lake currents and turbulence are low, it is unlikely that
natural processes would bury or reduce concentrations in toxic sediments in the short term even if loadings were
eliminated completely.
" Gilnaa, A.P. et iL 1977. "Reproductive ptnaelm ud egg ooatamlaaat levels of gnat lake* kerrUg galls". J. Wild. Maaage. 41:
438-468.
¦ New York State Department of Eavlroameatal Coaservadoa, Divttoa of Water, Banes of Moaitoriag aid Aw—meat. 1990. "New
York Stole Water Qsallty 1990" (New Yotk 1990 305(b) Report].
" US. EPA, Office of Water RegalatiaM aad Staadanb, Criteria aad Staadank Diviaioa. 1989. "Brieflag Report to tke EPA Sdeace
Advisory Board oa ike Eqiilibriam PartMoaJag Approack Co Geaetadoa Sediaieat Qaality Criteria* (EPA 440/5-89-002, April, 1989).
" Iatenatioaal Joist Commbaioa. 1989. "Report oa Gnat Lake* Water Qsallty". p. 53.
" Great Lakes Water Qiality Board. 1987. "Report oi Great Lakes Water Qulitjr". pp. £5-2.14.
¦ Coaaetvadoa Foaadattoa. "Great Lakes, Great Legacy?", p. 103.
-------
Page IH - 52
EXHIBIT K-17
INTERIM SEDIMENT QUALITY CRITERIA FOR NON-IONIC ORGANIC COMPOUNDS
Sedimeat eoaceatntioa
Criterioa
(ppfa) based oa % OC'
Conpoaad
(ppm OC')
Type1
2% OC
20% OC
PAH*:
Aceaaptkeae
730.0
CE
14,600.0
146,000.0
Aailiae
0.0662
CE
1.32
13.2
Pkeaaatkreae
139.0
CE
2,780.0
27^00.0
Otker
10.1
SC
202.0
2,020.0
Pestiddes:
Cklordiae
0.098
SC
1.96
19.6
Cklorpyrifot
3.22
CE
64.40
644.0
DDT
0.828
RB
1636
165.6
Dieldita
0.130
RB
2.60
26.0
Eadria
0.0532
RB
1.06
10.6
Etkyl Paiatkioa
0.081
CE
1.62
16.2
Heptacklor
0.110
RB
Z20
22.0
He puck lor Epoxide
0.008
SC
0.16
1.6
Liadaae
0.157
CE
3.14
31.4
Otker:
PCBe
195
RB
390.0
3,900.0
Soiree: EPA 440/5-89-002 (April 1989)
'OC m oigaaic carboa coaleal of aediaeat
"CE « criterioa developed oa « ckroaic effect bull
RB ¦ criterion developed oa i itaidae bub
SC ¦ scteeaiig level citieitoa
A number of technologies for remediating toxic sediments are currently in the developmental stages. The
Engineering/Technology Work Group of tbe ARCS program will be evaluating the effectiveness, technical feasibility,
and cost of various technologies as well as estimating tbe risks associated with each method. Technologies to be
evaluated include: solidification/stabilization, inoiganic treatment/recovery, biorcmediation, potassium-polyethylene
glycol (KPEG) nucleophilic substitution, best extraction sludge technology (BEST) extraction process, critical fluids
(CF) systems solvent extraction, incineration, low temperature thermal stripping, wet air oxidation, tow energy
extraction, ecological destruction process, in-situ stabilization, acetone extraction, aqueous surfactant extraction, taciuk
thermal extraction, and sediment dewatering methods.** None of these technologies has yet been tested even on a
pilot-scale.
The only method for tbe remediation of contaminated sediments which has been used extensively up to this time is
dredging and removal to CDFs. Because tbe impacts of resuspension of sediments during dredging and the risks
associated with CDFs are largely unknown, it is uncertain how effective this method of remediation is at reducing
ecological risk. Tbe risk created by dredging possibly might be greater than tbe risk posed by leaving sediments in
place. Leaving sediments, dredging tbem, and disposing of them all create risks which are currently being assessed in
order to weigh remediation options.
* VS. EPA, Gnat Lake* Nadoaal Program Office. 1990. "Aueumeat aad Rtmediitioi of Coatamiaited Scdimeats (ARCS) Work
Pita." p. 36.
-------
Page m - 53
Based on the information above, reversibility of damage can be characterized as low. Although the rate at which the
ecosystem would recover to its original state is not known, it is known that the elimination of contamination through
natural processes is not likely in the short term, and that the reduction of risk through currently available remediation
technologies is not yet certain. The effectiveness and costs of other remediation technologies has yet to be fully
evaluated.
Welfare Impacts
Health Care Costs
It is not possible to estimate health care costs, because no information is available about the human health impacts
that treats toxic sediments separately from other source of surface water contamination. The reader is referred to the
welfare discussions in Problem Area G for a general assessment.
Environmental Costs
The environmental costs of toxic sediments include the values of lost fish and other aquatic organisms, as well as
economic losses from commercial and sport fishing activities. The extent of damage caused to aquatic life in the
Great Lakes by contaminated sediments is not wholly quantifiable at this time. Therefore, it is difficult to assess the
total environmental cost of this damage. However, for a general assessment of such impacts, the reader is referred to
Problem Area G.
Economic losses are similarly hard to estimate. The following data from the 1990 Statistical Abstract of the United
States provide some insight into the size of the effected industries. The total fish catch from the Great Lakes in 1988
was 40 million pounds (representing 0.6 percent of the total domestic fish catch,) and had a value of $19 million
dollars (representing 0.5 percent of the value of the total domestic catch.) The total catch from the Great Lakes has
declined both in absolute terms and as a percentage of the domestic catch since 1980. 3,766,000 Americans (2
percent of the population) engaged in sport fishing in the Great Lakes in 1988, spending a total of $1,560,000 on
food, lodging, transportation, licenses, and equipment. The Great Lakes Fishery Commission estimated the total
annual regional economic impact of the Great Lakes fisheries (commercial and sport) to be between $2 billion and $4
billion fx 1985." However, without an estimate of the reduction in fish population or edibility, it is difficult to
assign a dollar value under the beading of environmental costs even though toxic sediments are known to have a
negative impact on fisheries and on the commercial and sport fishing industries.
Toxic sediments also cause economic damages to the shipping industry because of restrictions on dredging. For
example, in some harbors, such as Indiana Harbor and the lower Maumee River, transportation has been impaired
because contaminated sediments have halted dredging activities. Indiana Harbor has not been dredged since 1972,
and as a result of the sediments deposited since then, ships have been forced to reduce their drafts by up to ten feet.
" Glut Lake* Commiuloa. "Great Lake* ComaeicUl aid Recreatioaal Hiibor Dicdgiag: Isaac* aad RcconmeadatioM". p. 28.
-------
Page IK - 54
L. PESTICIDE DISCHARGES AND ENVIRONMENTAL RISK
Problem Area Description
The Great Lakes Basin is one of the major agriculturally cultivated areas in the U.S. As a result, it receives large
applications of pesticides. This section will address discharges and the environmental risks resulting from the use of
21 herbicides and insecticides in the Great Lakes Basin. Human health impacts from pesticide applications, food
residues, and other uses of pesticides are discussed in Problem Area X, Pesticides: Human Health. Human health
impacts from pesticide accumulation in fish are addressed in the discussion of health impacts from point and nonpoint
sources, and is found in Problem Area G, Industrial Point Source Discharges to Surface Waters. For a discussion of
pesticide loading to surface water from atmospheric deposition, see Problem Area J, Atmospheric Loadings to Surface
Water.
There are a number of primary sources of pesticide exposure in the Basin. These sources include:
¦ agricultural applications;
¦ lawn care and household uses;
¦ manufacture and blending of pesticides;
¦ storage of pesticides; and
¦ accidental releases and spills.
The Great Lakes Basin produces a number of major crops such as com and soybeans to which farm workers apply
large volumes of herbicides and insecticides. Suburban spraying of property, often with high pressure systems, can
result in contamination of neighboring property, residents, pets, and livestock. The Basin also bouses thousands of
pesticide producing and custom blending establishments. There is also a number of storage facilities that bouse the
canceled pesticide, dinoseb.
Data are most available on pesticides applied for agricultural purposes. Therefore, the discussion of this problem area
will focus mainly on risks posed by this major source. Some summary data on levels of exposure to pesticides from
the other sources are also available and will be presented. At risk from exposure to pesticides are terrestrial animals,
plants, and insects, as well as fish and other aquatic »nimni«t and aquatic plants and insects.
Magnitude of the Problem
Agricultural Applications
Cropland in the Great Lakes Basin comprises 1&5 million acres or 18 percent of the total area of the Basin. Major
cropland areas of the Basin include northwest Ohio in the Lake Erie drainage Basin, the Saginaw River and Bay area
in Michigan in the Lake Huron drainage Basin, and east-central Wisconsin in the Lake Michigan drainage Basin."
Exhibit L-l compares the total cropland and major crops in the Basin and in Region 5. The Basin represents 17
percent of the cropland in Region 5 States. Proportions of cropland are similar in both Basin and Region, with corn
representing the major crop in both, and soybeans the next major crop.
" U5. EPA, Gntt Lake* Natfoul Program Office, US. Agricilt»r»l Tillage Practice* U the Oitil L»k«» Birim. 1988.
-------
Page in - ss
Exhibit L-l
Major Crops in Basin
(millions of acres)
Crop
Basin %
Com
Soybeans
Small Grains
7.8 42
4J5 24
3.1 17
Total
l&S 100
Ullage practices in the Basin were examined in 1988 by the Great Lakes National Program Office (GLNPO) and the
Conservation Technology Information Center (CTIC). Based on tbe draft report, agricultural landowners are applying
conventional tillage methods more extensively than conservation tillage methods, which are used on 27 percent of
cropland in tbe Rasin Conservation tillage is used for 38 percent of the com crop but only 25 percent of tbe soybc«Q
crop. It is interesting to note that two counties in northwest Indiana have 95 and 89 percent, respectively, of tbe acres
used for com production in conservation tillage. Tbe counties showing tbe highest proportion of acres in cropland
found in Ohio, also show the lowest rates of conservation tillage in tbe Basin.
Exhibit L-2 shows tbe geographic distribution of total conservation tillage use by acreage in Basin counties. Tbe
GLNPO and CTIC report indicates it would be difficult to evaluate tbe effect of conservation tillage on
runoff, including pesticides, on a regional or lake-wide scale. Research done by tbe U.S. EPA in tbe Lake Erie B^in
found no significant differences in runoff, tile flow, and pesticide losses between conservation tillage and conventional
test plots. However, tbe report underscores tbe difficulty in gathering accurate information on soil type, tillage
practices, and runoff in the Great Lakes Basin.
More research and information on tillage practices on all agricultural acres can be expected over the next few yean «
best management practices improve and are implemented into farm operations in the Basin. Developing and
implementing methods which reduce erosion, such as conservation tillage practices, is important for many reasons, foj
example, soil erosion from agricultural activities can transport contaminants such as pesticides into the Lakes. Once
the contaminants reach the sediments of the Great Lakes they can leach into the water degrading water quality and
impacting aquatic life. It is probably not coincidental that some of tbe same areas of the Basin that are most
intensively farmed - northwest Ohio and tbe Saginaw River and Bay area in Michigan - are plagued with some of
the highest rates of erosion in the Basin.
Agricultural landowners in the Basin apply large volumes of pesticides on their crops to control a wide variety of
pests. Exhibit L-3 presents tbe total pounds of active ingredients applied in tbe Basin for 8 different pesticides. The
data were gathered from 213 counties in tbe Basin for a typical year in tbe mid-1980's. Data are weighted by tbe
area of land within tbe Basin. Applicators utilize large quantities of tbe herbicides alachlor, atrazioc, and cyaoazine
and tbe insecticide carbofuran in the Basin. With over 9,500,000 pounds of alachlor applied per year, it exceeds other
pesticides applied in the Basin. Following next are atrazdne and metolachlor with over 7,000,000 and 6,000,000
pounds respectively applied per year in the Basin. Perhaps tbe most widely used chemical in the Basin is 2,4-D,
which is used in agriculture, turf treatment, industrial vegetation control, and by homeowners, although, by volume
2,4-D does not equal the major agricultural herbicides.
-------
Exhibit L-2
TOTAL CONSERVATION TILLAGE USE
Great Lakes Basin Counties
Lake Supencr
Lake Huron
Lake Onta
Lake
ichigan
ACRES
CONSERVATION
TILLAGE
1 - 1,000
1,001 - 5,000
5,001 - 15,000
H 15,001 -30,000
> 30,000
&EPA
Source: National Resources Inventory, 1982 us environmental protection agency
Soil Conservation Service great lakes national program office
-------
Page III - 56
Exhibit L-3
Total Founds of Pesticides Applied Per Year in Basin by State
(approximate, for typical year in late-19805 in millions of pounds)
Pesticide
IL
IN
MI
MN
NY
OH
PA
WI
PESTIC3DB
TOTAL
2,4-D
3385
29,245
212,548
27,669
59,666
81,018
2,769
35,879
452,179
Aciflaoifei
784
3,871
0
3
0
53/591
316
549
59,214
Alacklor
28,796
1313,967
3345,989
19,744
705,269
2377,806
43,519
1334,894
9,669,984
Auaziae
31,925
899,083
1,945,597
10,814
951,309
1,775,784
55,698
13U.748
7,481,958
CiiboFarai
862
64,261
577,187
119
197,988
179,352
7,891
87,949
1,115,609
CjfMBiK
8^53
336,766
1,035,687
5,032
527,168
600,165
27,609
657,211
3,197,891
Metobchlor
34,041
671,801
2,726,006
5,115
770,836
1,941,632
59,110
545,354
6,753,895
Triflmralia
11,173
171,400
183,956
601
55306
263,532
179
48352
734,999
STATE TOTAL
119,000
3,490,000
njmjm
3,268,100
7,773,000
197,000
4^27,000
29300,000
-------
Page III - 57
Once applied, pesticides may volatilize to the air, degrade in tbe soil, leacb into ground water or be transported
through runoff into surface water. Tbe environmental fate of these pesticides depends on several factors, including
persistence, mobility, volatility, water solubility, ability to adhere to soil particles, formulation, as well as the amount
of precipitation. One source estimates that 4 to 20 percent of tbe pesticides applied is taken up by plants and perhaps
Jess than 0.1 percent readies tbe target pest*. Many pesticides applied in significant quantities in the Basin are
considered by EPA to be very mobile thus having a high potential to leach through the unsaturated soil zone and
contaminate ground-water supplies. Indeed, a number of studies have found several pesticides in drinking water wells
as well as surface waters in tbe Basin.
Consideration of a pesticide's ability to degrade is an important part of the risk/benefit analysis that takes place in
pesticide registration. The risk/benefit analysis also considers tbe metabolites and degradation products of pesticides
in tbe environmental fate portion of tbe review, which includes aerobic and anaerobic metabolism and the products of
these processes. Each metabolite of toxicologic concern enters into the risk analysis.
Exhibit L-4 presents the acreage in tbe Basin treated with 7 pesticides that data was available for. Atrazine was
applied on more acres in the Basin than any other pesticide followed by Alachlor and Metolachlor.
Exhibit L-4
Acreage Per Year in Basin Treated
with Selected Pesticides
(typical year in late 1980s in millions of acres)
Pesticide Acres
2,4-D
1 JO
AtiHuorfeo
02
Alachlor
4.7
Atrazine
5.7
Cyanazine
1.9
Metolachlor
33
Thfluralin
0.9
Exhibit L~5 presents tbe application rate of these 7 pesticides in tbe Waain Exhibits L-6 through L-12 show tbe
geographic distribution of these applications. Alachlor was applied at tbe highest rate in the Basin than any other
pesticide, followed closely by metoiacblor and cyanazine. The rates of application of these pesticides are similar to
tbe rates of application across the nation. Some pesticides axe applied at higher rates across tbe nation than in the
Basin, while tbe application rate of other pesticides, such as cyaoaztae and metolachlor, are higher in tbe Basin than
across tbe nation.
** Waddell, T£, B.T. Baser, Md K. Cox. 1988. "MuugUg AgrfcaUanl Chemicals it Ike Eaviroaneat: A Cue for t MilH-Mdli
Approack," Coaaerratioa Foaadatioi, p. 42.
-------
Page m - 58
Exhibit L-5
Selected Pesticide Application Rates"
in the Great Lakes Basin for 1987
(lbs Al/acre/ycar)
Pesticide Basin Application Rate
Cyanazine
Metolachlor
Trifluralin
Acifluorfen
Atrazine
2,4-D
Alachlor
0.5
2.1
03
12
1J6
19
0.8
An analysis of total pounds of pesticides applied for agricultural purposes as well as their rates of application sbows a
general high usage of a number of pesticides in various counties within the Basin.
Noo-agricultural Uses
Several of the pesticides used for agricultural purposes are also used in the care of lawns, turf, and golf courses. On aa
area basis, EPA estimated that homeowners alone apply from 5 to 11 pounds of active ingredients per acre.*1 Compered
to agricultural application rates, lawns and gardens receive greater rates of pesticide applications than agricultural land.
The General Accounting Office, using a 1988 EPA estimate, found that 67 millions pounds of active ingredients are
applied on lawns by lawn care operators and homeowners each year, across the U.S. Although lawn pests are not
distributed evenly across the country, an approximation of the amount of lawn care pesticides applied in the Basin can
be determined using the proportion of the U.S. population residing in the Basin (30,050,000/243300,000), or 12 percent.
Using this approximation, over 8 million pounds of active ingredients are used annually on lawns in the Great T
Basin.
Ecological Impacts
Although the fate of applied pesticides in not well known, two phenomena have been described in the scientific literature*
(1) the bioaccumulation of pesticides in fish, and (2) the effects of increased nutrient and pesticide loadings on the balance
of aquatic life forms. Pesticides and nutrients can enter the Great Lakes ecosystem through sediment transport produced
" AppUcatk* ntea depead oa the crop aid oa peat problems. la tome caaca, aaziaam label nle of application ahoald be aaaamed.
* Water Qaality Board, Ialeraatioaal Joiat Conmiaaioa. 1967. "1987 Report oa Gnat Lake* Water Qaaltty,* p. 181.
Severity
-------
Page III - 59
by soil erosion from agricultural activities. Once these substances reach the sediments of the Great Lakes, they can
gradually leach into the water and enter the food chain or upset the balance of aquatic life in the lake. Several studies
bave rinrnmi'nteri the bioaccumulation of chemicals including such suspended and canceled pesticides as aldrin, dieldrin,
cblordane, heptacfalor, DDT, and mirex in large sport fish species in the Great Lakes.
Applications of the insecticide caibofuran presents a risk to non-target species, which include many avian species. More
th«n 1,100,000 pounds of caibofuran were applied primarily on com and soybean oops in the Basin in one recent year.
Tbcre are a number of documented cases in the United States of bird poisonings after use of the granular formulation of
the pesticide.*2 Although documented cases of carbofuran poisonings in the Basin are relatively few, it is reasonable
to assume the potential exists for more bird poisoning incidents in the Great Lakes Basin, even when the pesticides are
used in accordance with existing regulations.
}
A potentially severe ecological impact of pesticide usage in the Basin involves the deleterious effects of these chemicals
on endangered species. However, EPA's comparative risk project found impacts to endangered species in Region 5 from
pesticides as "low". To project future needs for endangered species in Region 5, Temple, Barter and Sloane used U.S.
Fish and Wildlife Service listings and the Nature Conservancy listings to determine potential candidates for protection
under the OPP endangered species protection program.
There are a number endangered species in the Basin including, but not limited to:
¦ white cat's paw pearly mussel in Indiana and Ohio;
¦ piping plover in Michigan;
¦ Kirtland's warbler in Michigan; and
¦ prairie bush-clover in Illinois.
The preferred habitat for the prairie bush-clover is northern facing slopes, often bordering on agricultural fields or
pasturelands. These areas present a high risk of exposure from spray drift application of pesticides.
Unfortunately, the impacts of pesticides on wetlands are largely unknown. The major agricultural impact in the Basin
have been due to the drainage of wetlands to make cropland. This has resulted in reduced critical habitat in the Basin
and could potentially impact the survival of critical species if left unchecked. A thorough of wetlands appears
in Problem Area A, Physical Degradation of Water Habitats and Wetlands.
Other species-level effects of oonoern from pesticide contamination include, but are not limited to:
¦ insecticide-induced mortality among the aquatic juvenile life stages of many insect species (e.g., dragooflies,
mayflies);
¦ herbicide-induced reduction of submerged aquatic vegetation;
¦ impaired reproduction of fish; and
¦ fish kills.
Ecosystem-level effects include:
¦ reduced biomass and diversity of invertebrate and vertebrate communities;
¦ herbicide-induced reduction of submerged aquatic vegetation; and
¦ concomitant loss of invertebrate communities as well as fish that utilize plant cover for spawning or protection
« Pestidde * Toxic Chemical Newt: Aigut 8,1990, "Bald Eagle Valaed at $40,000, FWS Notes ia Catbofaiaa-Pndaior Cue,"
pp. 9-10; May 2, 1990, "Birds ii Va. Ciibofina Kill of Coacen, USDI Official Says," p. 27; Jaae 6,1990, "Caibofuaa Uaked to Two
Bird Kill* ia Delaware; No Miaaac Seca," p. 23.
-------
Exhibit L-6
2,4 - D Application, 1986
Great Lakes Basin Counties
~
i
Lake Superior
Lake Huron
Lake Ontario
Lake
Michigan
POUNDS PER YEAR
NO DATA
0- 1,000
1,001 - 5,000
5,001 - 10,000
H 10,001 - 15,000
I > 15,000
Source: Resources lor the Future, 1990
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Exhibit L-7
ACIFLUOR Application, 1986
Great Lakes Basin Counties
Lake Huron
Lake Ontario A !_
Lake \m---
Michigan
POUNDS PER YEAR
~
NO DATA
i
1 - 100
101 - 1,000
1,001 -2,000
> 2,000
&ERJV
Source: Resources for the Future, 1990
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Exhibit L-8
ALACHLOR Application, 1986
Great Lakes Basin Counties
MN
POUNDS PER YEAR
NO DATA
0
1 - 10,000
10,001 - 100,000
100,001 -200,000
Wl
Lake Superior
Lake Huron
Lake
I Michigan^
IL
IN
OH
Source: Resources for the Future, 1990
Lake Ontario.
NY
PA
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Exhibit L-9
ATRAZINE Application, 1986
Great Lakes Basin Counties
Lake Superior
Lake Huron
Lake Ontario
Lake
Michigan
POUNDS PER YEAR
NO DATA
0- 1,000
1,001 - 10,000
10,001 - 50,000
50,001 - 100,000
> 100,000
Source: Resources (or the Future. 1990
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Ehibit L-10
CYANAZINE Application, 1986
Great Lakes Basin Counties
Lake Superior
Lake Huron
Lake Ontario
Lake
Michigan
POUNDS PER YEAR
NO DATA
0 - 1,000
1,001 - 10,000
10,001 - 20,000
I 20,001 - 30,000
I > 30,000
Source: Resources lor the Future, 1990
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Exhibit L-12
TRIFLURALIN Application, 1986
Great Lakes Basin Counties
MN
POUNDS PER YEAR
i
I
NO DATA
0
1 - 1,000
1,001 -15,000
15,001 -30,000
> 30,000
Wl
Lake Superior
Lake Huron
I'^'s
b
I!
s Ha Hi*
£1
i
Lake
\Mlchlgan'
Lake
Eriel
IL
IN
OH
Source: Resources for the Future, 1990
Lake Ontario i
NY
PA
!§zERA
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Exhibit L-11
METOLACHLOR Application, 1986
Great Lakes Basin Counties
i
V
I
Lake Superior
Lake Huron
Lake Ontario
Lake
Michigan
POUNDS PER YEAR
NO DATA
0- 1,000
1,001 - 10,000
10,001 - 50,000
50,001 - 100,000
> 100,000
Source: Resources for the Future, 1990
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Page IH • 60
of juvenile stages.
Historically, two pesticide characteristics have proven key to the potential for a pesticide to cause adverse aquatic effects:
toxicity and persistence in the environment through bioaccumulation. Exhibit L-13 presents information on the
bioaccumulation potential and aquatic toxicity of DDT, dieldrin, toxaphene, bexachlorobcnzcne, and lindane.
Exhibit L-13
Approximate Aquatic Toxicity for Selected Pesticides
Pestidde
BioaccamaUtioi
PoteitUI
Log(BCF)
EPA Acate
AWQC (ig/L)
EPAChioaic
AWQC (ig/L)
Lowesl Acate LC»
for Fill
(ag/L)
Lowed Acate
LCa for
la vertebra la
(¦g/L)
DDT
4.7
1.1
0.001
2
0.2
Dieldrii
3.7
2.5
0.002
1
0.5
Toxapheae
4.1
0.72
<0.001
2
1.3
HCB
3.9
-
-
12,000
-
liadaae
2.7
-
-
2
3.2
Soircet: EPA Office of Solid Waste tad Emergeicy Reipouc Directive 9285.4-1. October 1986.
Venckierei, K, 1966. Haadbook of EaviroamcaUl San oa Oigtaic Ckemiede.
EPA. 1986. Qtality Ctilerii for Water.
U.S. Fiih lid Wildlife Service. 1980. Hildbook of Acile Toxicity of Ckemictli to Fish aad Aqutic Iivertebnte*.
p. 137.
As mentioned in Section G, concentrations of pesticides in Great Lakes fish are weU documented. Exhibits L-14 through
L-15 show trends in DDT and dieldrin concentrations for Lake Michigan Lake Trout over a twelve year period. Although
banned in the 1970s, dieldrin and DDT are still found in samples.
LAKE MICHIGAN LAKE TROUT
DDT
mgAg
70 71 73 73 74 7B 76 77 7B7BB0 81 82 63M
YEAR
LAKE MICHIGAN LAKE TROUT
DIELDRIN
mgAB
70 71 7273 74 757077787B80 81 8283#4
YEAR
Exhibit L14
Exhibit L-15
-------
Page in - 61
Reversibility
The reversibility assessment for the bioaccumulation of pesticides in fish is assumed to be low. Many of the P^rtlrkV 11
found in fish tissues are very persistent. DDT has been banned for a number of years but is still found in fish and bircte.
The concentration of these pesticides in sediments in the Basin are still significant, and therefore will continue to present
risks of'exposure through the food chain. Recent fish monitoring data does suggest, however, that [raticide- residues in
fish may be declining.w
Welfare Impacts
Due to a lack of information the assessment of environmental costs from pesticide exposure in the Basin will be primarily
based on a qualitative discussiaa of such costs.
The bioaccumulation of pesticides in sport fishes in the Great Lakes has probably resulted in fewer fish caught due to
the necessity of fish advisories, fewer viable fish species, and a decrease in the aesthetic appeal to fisherman of
contaminated fish. Hie Great Lakes Sport fishery has been estimated to be worth $1.56 billion(US).
The cost of replacing one adult bald eagle was estimated to be approximately $40,000. The Fish and Wildlife Service
estimated the value of a bald eagle based on the need to "back" at least eight eagles to gain one adult bird. The estimate
was made in conjunction with a Idll of 7 eagles that ingested and died from the pesticide carbofuran." However
estimates such as these should not be interpreted as suggestions that the replacement of impacted species is even
or that replacement can be considered in the formulation of policy.
Other environmental costs include lost habitat from pesticide contamination. These costs are particularly severe if
endangered species habitat is involved.
* Sckmltt, Zajlcek, aid Petennu, "Nitkmil Coatamlaatioa Biomoaitoriag Program: Rctidies of Oipaockloriae Chemical* U US.
Fieahwafer Fiafc 1976-1984. Arefc. Envlroa. Coattala. Toxicol- 19,1990, pp. 746-781.
* Petttdde A Toxic Chemical News. Aagast 8,1990. "Bald Eagle Valaed at WO,000, FWS Notes ia Cartxrftrai -Proditor Cue," p. 9.
-------
IV- 1
IV. PRIMARY AIR AND WASTE PROBLEM AREAS
M. SULFUR AND NITROGEN OXIDE EMISSIONS
Problem Area Description
Sulfur and nitrogen oxides cause a variety of primary and secondary effects. Primary effects include health,
visibility, and welfare impacts. A major secondary effect is acid deposition, which results from the chemical
transformations of oxides of sulfur and nitrogen to sulfates and nitrates. This process results in the production of
acid rain, snow, and fog, as well as dry deposition of acidic particulate matter. Acid deposition alters the chemistry
of affected aquatic and terrestrial ecosystems, damaging plant and animal life. Sources of sulfur and nitrogen oxides
include a variety of industrial, commercial, and residential fuel and related combustion sources.
Sulfur dioxide (SO,) is emitted primarily from the combustion of fossil fuels, particularly coal. The use of coal-fired
power plants by the electric utility industry accounts for approximately 63 percent of nationwide emissions of SO,.
Other sulfur dioxide sources include industrial processes such as petroleum refining, pulp and paper manufacturing,
iron and steel production, and industrial boilers. Nitrogen oxide (NOJ emissions are also the result of fossil fuel
combustion, namely the combustion of coal, natural gas, oil, and gasoline. Major emitters include the utility industry,
industrial sources, and motor vehicles.
Sulfur and nitrogen oxides and acid aerosols can be transported over very long distances in die atmosphere.
Therefore, problems in the Great Lakes Basin resulting from these pollutants ate probably due to sources both inside
and outside the Basin. Similarly, sources within the Basin probably contribute to acid deposition problems outside
the Basin (i.e., in the northeast U.S. and southeast Canada). Nevertheless, in order to provide information on the
potential sources of sulfur and nitrogen oxide pollution problems, we present information on the location of major
concentrations of emitters within the Basin. Keep in mind, however, that these and other sources are responsible
for problems in the Basin, and some emissions from the Basin produce effects far downwind of the Basin.
EPA's Aerometric Information Retrieval System (AIRS) data management system lists 1,434 stationary sources of
sulfur and nitrogen oxide emissions in the Great Lakes Basin. Exhibit M-l illustrates the geographic distribution
of these sources by estimated SO, and NO, emissions in the Basin counties. Exhibit M-2 illustrates the geographic
distribution of SO, and NO, monitoring sites within Basin counties. These sources are spread throughout 140 of the
213 Great Lake Basin counties. The counties with the greatest numbers of sources include:
¦ Milwaukee county, Wisconsin;
ai Cook county, Illinois (Chicago);
¦ Lake county, Indiana (East Chicago, Gary);
¦ Wayne county, Michigan (Detroit);
ai and numerous other counties in Wisconsin and Indiana.
Stationary sources within the Great Lake Basin counties emit nearly 3 million tons of sulfur and nitrogen oxides
annually. The following counties emit the largest quantities of these pollutants:
¦ Monroe, Ottawa, St Clair, and Wayne counties in Michigan (757,000 tons/yr);
¦ Columbia and Milwaukee counties in Wisconsin (307,000 tons/yr);
¦ Lake, Porter, and La Porte counties in Indiana (475,000 tons/yr);
¦ Monroe and Erie counties in New York (260,000 tons/yr); and
¦ Itasca county, Minnesota (99,000 tons/yr).
-------
Exhibit M-1
Estimated S0X and N0X Emissions
Great Lakes Basin Counties
Lake Superior
Lake Huron
Lake Ontario
Lake
Michigan
TONS PER YEAR
1 - 1,000
1,001 - 10,000
10,001 - 100,000
H 100,001 -200,000
I > 200,000
Source: Aerometric Information Retrieval System, 1990
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Exhibit M-2
SO and NOy Emissions Monitoring Sites
Great Lakes Basin Counties
Lake Superior
Lake Huron
Lake Onta
Lake
Michigan
MONITORING
SITES
Source: Aerometrlc Information Retrieval System, 1990
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
IV - 2
Human Health Impacts
Toxicity Assessment
Exposure to sulfur dioxide can significantly increase the incidence of acute and chronic respiratory diseases as well
as cause damage to lung tissue. The primary National Ambient Air Quality Standards (NAAQS) for sulfur dioxide,
arc designed to protect public health with an adequate margin of safety. The NAAQS for sulfur dioxide is an
arithmetic mean of 80 ug/m® and a maximum 24-hour concentration (not to be exceeded more than once per year)
of 365 ug/m1.
Exposure to nitrogen dioxide is also associated with respiratory illness and lung damage. The NAAQS for nitrogen
dioxide is an annual arithmetic mean concentration of 0.0S3 parts per million (ppm).
Despite evidence of health effects resulting directly from exposure to these pollutants, this risk evaluation focuses
on exposure to sulfite (as a surrogate for acid aerosols). This approach is adopted directly from the Region S
comparative risk project Acid aerosols irritate the lungs, causing constricted breathing. This is of particular concern
to asthmatics, who are especially vulnerable. Further, evidence suggests that exposure to acid aerosols is
with respiratory symptoms, such as cough, both in healthy individuals and in individuals with chronic respiratory
diseases. Researchers have also observed associations between sulfate levels and both mortality rates and hospital
admissions due to respiratory illness. Laboratory work with animals suggests that acid aerosols could be related to
the development of chronic respiratory diseases, however this has not been confirmed with human epidemiological
studies. For this risk characterization, we apply the approach used in the Region 5 comparative risk work, which
estimates annual number deaths and respiratory symptoms resulting from sulfate concentrations above 10 ugAn1 and
hospital admissions resulting from sulfate concentrations above 7 ug/m1.
Exposure Assessment
Parts of 10 counties in the Great Lakes Basin are classified in the Code of Federal Regulations (CFR) as not attaining
die primary or secondary NAAQS for sulfur dioxide. Further, a review of 1985 through 1989 monitoring data
indicates that nine counties in the Great Lakes Basin have recorded values exceeding the primary NAAQS.
No areas in the Great Lakes Basin are classified in the CFR as nonattainment for nitrogen dioxide and no areas had
monitored violations of a NAAQS during the period 1985 through 1989 (based on a review of AIRS reports).
We rely on sulfate concentrations in the ambient air of the Great Lakes Basin, used as a surrogate for die
concentration acid aerosols, to characterize the human health risk from this problem area. Based on the limited
available monitoring data (primarily 1989 annual concentrations) that are available and Region 5's assumptions about
sulfate concentrations in areas without sulfate monitoring, we estimate that 5 million people in the Great l-«frfs
are exposed to sulfate levels above 7 ug/mJ and approximately 2 million people in the Basin are exposed to levels
in excess of 10 ug/m'.
Risk Characterization
To characterize the human health risk posed by sulfur and nitrogen oxidea in the Great Lakes Basin, we have adopted
the method developed by RCG/Hagler, Bailly, Inc. under contract to EPA and used in the Region 5 comparative risk
study. This method uses ambient sulfate levels to estimate the annual number of deaths, hospital admissions, and
-------
IV- 3
episodes of respiratory symptoms among children and adults Faulting from exposure to acid aerosols. In this
approach, only sulfate levels above 7 ug/m3 were used to calculate annual hospital admissions, and only levels above
10 ug/m1 were used to estimate the number of annual deaths and episodes of respiratory symptoms. Hie following
equations were used to estimate the number of people affected annually by add aerosol pollution:
Annual Deaths - (3.7 x 10 s) (S, - 10) (POP,),
Annual Hospital
Admissions * (8.6 x 10"5) (S/9) (POPj),
Children with
Respiratory Symptoms * (0.035) (C), and
Adults with
Respiratory Symptoms « (0.05) (M) + (0.02) (W),
where, j ¦ location of monitoring site,
S, * annual average sulfate concentration recorded at monitoring site j (if > 10 ug/m1),
POPj ¦ exposed population at monitoring site j,
C * number of children in areas where the annual average sulfate level is greater than or
equal to 10 ugAn',
M ¦ number of men in areas where the annual average sulfite level is greater than or
equal to 10 ugAn\ and
W * number of women in areas where the annual average sulfate level is greater than or
equal to 10 ug/m1.
Monitoring data and professional judgment were used to identify locations with sulfate concentrations above the
critical levels (i.e., 7 and 10 ug/m1). The equations listed above were used with relevant population data to estimate
the health impacts of sulfur and nitrogen oxides pollution. We estimate that exposure to sulfate in the Great Lakes
Basin results in 148 premature deaths annually, and approximately 470 hospital admissions annually. In addition,
we estimate that over 33,000 children and nearly 110,000 adults in the Great Lakes Basin experience respiratory
symptoms as a result of exposure to sulfates.
Ecological Impacts
In the states east of the Mississippi, rain is almost always acidic In the Great Lakes Basin, the pH of rainfall
decreases (i.e., acidity increases) steadily from west to east. Contour maps of annual precipitation throughout the
United States indicate that rain in the Great Lakes Basin ranges from a pH of approximately 5.1 in northeast
Minnesota to a pH of 4,2 in eastern Ohio, Pennsylvania, and New York. Much of the central portion of the Great
-------
IV-4
Lakes Basin receives precipitation with an average pH of approximately 4.3 to 4.7.1
A number of scientific studies have clearly linked adverse effects on aquatic ecosystems to acid deposition. For
example, considerable evidence suggests that numerous lakes and headwater streams in the Appalachian mountains
have been acidified as a result of atmospheric deposition, either directly as deposition to the water surface or
indirectly by interaction with soils in the watershed to enhance transport of hydrogen and aluminum ions to surface
waters. Fish and other aquatic organisms are adversely affected by acidification and the increased concentrations
of aluminum that frequently accompany it
In contrast, there is no direct evidence linking observed adverse effects on terrestrial (forest) ecosystems to add
deposition and/or its precursors. Researchers have demonstrated a variety of potential adverse effects of acid
deposition on foliar integrity, foliar leaching, root growth, soil properties, microbial activity in soil, resistance to pests
and pathogens, germination of seeds, and establishment of seedlings. Adverse effects also might result from the
combination of acid deposition, other gaseous pollutants such as ozone, natural biotic stresses (e.g., pests and
pathogens), and natural abiotic stresses (e.g., sever climatic conditions). While there are many ways in which add
deposition might affect forests, no proposed linkage between deposition and effects on trees in North America axe
adequately supported by current data.2
In this section we first describe the severity of the effects of acid deposition on aquatic and terrestrial ecosystem);
and then discuss the reversibility of these effects.
Severity
Effects of acid deposition on aquatic ecosystems include changes in population density, community diversity, »n
-------
IV - 5
In addition to acidifying surface waters, acid rain can leach toxic metals, including mercury, from the soil and deposit
them in waterways. This problem is particularly evident in Michigan, Minnesota, and Wisconsin, where biologists
have found unsafe levels of mercury in inland lakes.3
Studies have also indicated that the nitrogen compounds present in acid rain may act as a fertilizer, causing excessive
growth of algae. When the algae die, the decaying algae depress oxygen levels, thus contributing to eutrophication.
This, in turn, can threaten fish populations.
Exhibit M-3
Notes on Effects of Acidification on Freshwater Aquatic Ecosystems
PH
Effect
7.0
Reduced reproduction in most sensitive species (Daphnia)
6.6
First species lost (snails)
6.1
Reduction in zooplankton species
6.0
Loss of pa^uiia, reduction in crayfish species
5.9 - 5.8
First fish species lost
5.8 - 5.7
Major crustaceans lost
5.6
First substantial change in phytoplankton density, crayfish lost
5 JS
Most species of rotifers lost
5 J - 5.0
Large reduction in decomposition rates
5 J - 4.7
Most fish species lost
4.7
Major reduction in phytoplankton density
4.7 - 4.6
Major reduction in zooplankton density
4.2
Last fish species lost
Source: EPA (1986)'
The effect of acid deposition on lakes in the Great Lakes Basin has been evaluated in EPA's Eastern Lake Survey
conducted by the National Acid Precipitation Assessment Program. In this study, the effects of acid deposition on
5 Region 5 comparative risk report.
* For a more detailed summary over a broader hydrogen ion concentration range, ice p. 141, Water Quality Criteria 1972. A Report of
the Committee on Water Quality Criteria. Prepared by the National Academy of Science, National Academy of Engineering for EPA.
March 1973. EPA-R3-73-003.
-------
IV - 6
lakes In the following five areas in the Great Lakes Basin are evaluated:
¦ northeast Minnesota,
¦ the upper peninsula of Michigan,
¦ north central Wisconsin,
¦ "the upper Great Lakes" (i.e., land areas of Minnesota, Wisconsin, and Michigan), and
¦ the Adirondack*.
Exhibit M-4 summarizes the survey's findings regarding the pH of lakes in these five study areas. Note that NAPAP
surveyed only lakes between 4 hectares (10 acres) and 2,000 hectares (4,940 acres). The results of the survey are
applicable only to lakes foiling within this size range. Consequently, the results shown in Table 2 underestimate the
extent of acidification. This discrepancy is evident for the Adirondacks where the NAPAP study found that 11
percent of the lakes were already acidic, while a New Yoric State study found that over 25 percent of the lakes were
already acidic.1
As shown in Exhibit M-4, NAPAP's findings indicate that approximately 3,100 hectares of lake area in the Great
Lakes Basin have been acidified (i.e., pH is less than or equal to 5). Another source of information, the state's
reports to EPA under section 305(b) of the Gean Water Act, indicates that as many as 6,200 hectares of lakes and
42 river miles in the Great Lakes Basin are affected by acid deposition.
Exhibits M-5 and M-6 illustrate some of the observed effects on fish population associated with acidification of
Adirondack lakes.
' Hie lakes that were studied in this report are not ill in the Great Lake* Basin. However, because a large portion of the Adirondacks
lies within the Basin, it is believed that many of these findings ire germane ss well to Basin lakes. Furthermore, virtually all of the region
experiencing the highest hydrogen ion deposition in the NAPAP study vea does foil within the Great Lakes Basin.
-------
IV- 7
Exhibit M-4
pH Levels in NAPAP Eastern Lake Survey Lakes in the Great Lakes Basin
Northeast
Minnesota
Upper
Peninsula of
Michigan
North Central
Wisconsin
Upper
Great
Lakes
Adirondacks
pH
*
Number of
Lakes
0
95
30
0
129
5.0
Percent of
Lake Area
0 %
2 %
0.5 %
0 %
2 %
Area in
Hectares
(Acres)
0
680
(1,680)
0
0
2,380
(5,880)
PH
s
Number of
Lakes
0
137
163
0
258
5-5
Percent of
Lake Area
0 % 1
3 %
2 %
0 %
8 %
Area in
Hectares
(Acres)
0
1,200
(2420)
1,960
(4,843)
0
9,520
(23,500)
pH
£
6.0
Number of
Lakes
14
189
414
180
348
Percent of
Lake Area
0 %l
5 %
6 %
1 %
10 %
Area in
Hectares
(Acres)
0
1,700
(4,200)
5,880
(14,530) .
2,270
(5,610)
11,900
(29,400)
Source: NAPAP. 1988. NAPAP Interim Assessment: The Causes and Effects of Acidic Deposition.
Volume IV: Effects of Acidic Deposition.
-------
IV- 8
Exhibit M-5
Water Chemistry and Fish Population Status for 289 Adirondack Lakes
Fish Population Status
pH < 5.0
5.0 < pH < 6.0
pH > 6.0
Total
Waters with trout/salmonid species
15
77
122
214
Waters with only nontrout species
5
11
35
51
Waters without fish
13
5
6
24
Tout
33
93
163
289
Source: Colquhoun, J., Kretzer, W, and Pfeiffer, M. 1984. Acidity status update of lakes and streams in New York State. New
York State Department of Environmental Conservation Report WM (P-83(6/64)). Albany, N.Y.
Exhibit M-6
Effects of Acidification on Fish Populations in 604 Adirondack Lakes
Species
No Evidence for
Loss of Population
Loss Apparently
Unrelated to
Acidification
Lou Apparently the Result of
Acidification
Marginal
Evidence
Adequate
Evidence
Brook trout
409
86
114'
98
Lake trout
68
25
10
8
Rainbow trout
54
34
2
Lake whitefuh
28
24
1
Smallmouth bass
92
42
2
Largemouth bass
49
12
1
~tain pickerel
29
12
0
White sucker
336
66
10
Brown bullhead
420
74
13
13
Pumpkin seed sunfish
240
104
7
Yellow perch
185
80
2
Golden shiner
213
83
18
12
Creek chub
147
100
7
Lake chub
6
12
0
1 Includes 71 lakes considered typical brook trout habitat, with only recent survey data available and no fish caught.
Source: Baker, J, and Harvey, H. 1985. Critique of acid lakes and Ssh population status in the Adirondack region of New York
State. Draft final report for NAPAP Project E3-25. U.S. EPA, Washington, D.C
-------
IV- 9
Forests are also threatened by acid deposition. The Region 5 comparative risk report states that exposure to highly
acidified rain or fog can injure leaves and needles. Further, tree growth may be stunted by alteration in soil
chemistry caused by leaching of nutrients, such as magnesium and calcium, from the soil and by contamination of
the soil with heavy metals, such as aluminum, that ate released from soil particles. In addition, acid deposition can
destroy flora and kill nitrogen-fixing microorganisms that nourish plants.
A paper in Penn State Agriculture summarizes the major hypotheses regarding the potential adverse effects of acid
deposition (and other factors) on tenestrial ecosystems.6 Acid deposition and ozone together might be leachiqg
nutrients out of tree foliage and soil, weakening the tree, and stunting its growth. This, in turn, can make the tree
more susceptible to biotic stresses from insects to fungi and abiotic stresses such as drought or nutrient-poor soil.
An increasingly acidic soil can inhibit germination of seeds and establishment of seedlings, thus slowing the rate of
replacement of dead trees. Acidic soil can have a direct effect on soil microorganisms, insects, and competing plants.
At the cellular level, ozone causes reactions that move nutrients such as magnesium and potassium from inside the
cell to the cell and leaf surface. When it rains, those nutrients wash off the leaf and into the soil. On steep slopes
and in certain soils, these nutrients can be earned away from the tree instead of remaining where the root system
could recapture them. The result can be a deficiency of nutrients to the tree.
Reversibility of Damage
The panel of scientists convened by the Cornell Ecosystems Research Center to assist EPA's national comparative
risk study judged the impacts of acid deposition on aquatic and terrestrial ecosystems to be reversible in terms of
decades.
Welfare Impacts
As in the Region 5 comparative risk study, we quantify the following economic damages from nitrogen and sulfur
oxides pollution, including acid deposition:
¦ health effects due to sulfate levels;
¦ materials damage attributable to nitrogen oxides; and
¦ aesthetic damages associated with reductions in visual range.
Health Effects
EPA's Region 5 comparative risk study presented information on die costs of health effects associated with exposure
to sulfate aerosols. We have applied the same approach to our findings for health effects in the Great Lakes Basin.
Each premature death is valued at approximately $77,000 (in 1988 dollars) and hospital admissions are valued at
$4,400 each (also in 1988 dollars). Consequently, the costs for the 148 predicted premature deaths is $11 million
and the costs for the 470 hospital admissions is $2.1 million.
Materials Damage (Nitrogen Oxides)
As presented in the Region 5 comparative risk study, the primary materials damage associated with nitrogen dioxide
' Erb, C. 1987. "The jury is (till out." Penn State Agriculture, Fail 1967: 2-11.
-------
IV. 10
results from the fading of dyes exposed to the pollutant The EPA assumed the national cost of dye fading due to
nitrogen dioxide is approximately $280 million per year. On a per capita basis, this equals roughly $1.20. To
estimate the economic value of dye fading due to nitrogen oxide in the Great Lakes Basin, we multiplied the R»cjn
population, 21 million people, by $1.20 for a total cost of $24.8 million.
Visibility Effects
Evaluation of the value of visibility effects also uses the approach presented in the Region 5 comparative risk study
Review of visual range data indicates that the only portions of the Basin with visibility reduced to approximately
IS km or less is northeastern Indiana, far southeastern Michigan, the entire portion of Ohio that is in the Basin, and
Erie and Crawford counties in Pennsylvania.7 The affected population in these areas is estimated to be 6.6 million
people, which corresponds to 2JS million households. Applying the value of $220 per household to this number of
affected households in the Basin gives a cost of visibility effects of $560 million.
11iopleth map from EPA (1979) provided by Kathleen D'Agoctino, Region V. Source unknown. Figure 2-5: Median Sumner Viiual
Range and laoplefhs (or SuburbanAfonurban Area* 1974 - 1976.
-------
IV- 11
N. HAZARDOUS/TOXIC AIR POLLUTANTS
Problem Area Description
This problem area covers outdoor exposure to airborne hazardous air pollutants from routine or continuous point and
non-point source emissions. This problem area covers exposure through both inhalation and air deposition of these
pollutants to land areas. Runoff of deposited pollutants to surface waters and direct deposit to surface water is
addressed in Section J (Atmospheric Loading to Surface Water). This category excludes, to the extent possible, risks
from pesticides, radioactive substance, chlorofluorocarbons, emissions from waste treatment, storage and disposal
facilities, storage tanks, and indoor air pollutants.
Two recent EPA studies on the risks of hazardous or toxic air pollutants in local areas of the Great Lakes Basin
(Southeast Chicago, and Detroit/Windsor) have found the following pollutants to be the most significant contributors
to cancer risk: coke oven emissions, formaldehyde, chromium, 1,3-butadiene, carbon tetrachloride, polynudear
aromatic hydrocarbons (PAHs), dioxin, benzene, gasoline vapors, arsenic, beryllium, and asbestos.' Toxic air
pollutants with noncancer health effects include lead, arsenic, beryllium, and a number or organic compounds (such
as acrolein, carbon tetrachloride, and methyl ethyl ketone).
Major sources of hazardous and toxic air pollutants include large industrial facilities, motor vehicles, chemical plants,
commercial solvent users, and combustion sources. The Southeast Chicago and Windsor studies mentioned above
found the following types of activities to be the most significant sources of these air pollutants: steel and coke
manufacturing, motor vehicles, other manufacturing, fuel combustion (E.G., home heating).*
Lead gasoline additives, nonfenous smelters, and battery plants are the most significant contributors to atmospheric
lead emissions. Transportation sources in 1988 contributed 34 percent of the annual emissions. This contribution
has decreased substantially from approximately 83 percent in 1975, due to regulations issued in the early 1970's and
mid 1980's which required gradual reduction of the lead content in all gasoline.
Human Health Impacts
Data presented in the Office of Air Quality, Planning and Standards' (OAQPS) "Cancer Risk from Outdoor Exposure
to Air Toxics" (September 1989) was used in this report to estimate cancer risk. This approach parallels that used
in the Region 5 comparative risk study. Noncancer effects, including risks from airborne lead, were also calculated
in a manner that parallels the Region 5 comparative risk effort
Toxicity Assessment
Cancer Risk
The carcinogens selected for evaluation include acrylonitrile, arsenic, asbestos, benzene, 1,3-butadiene, cadmium,
' SPA. 1989. "Eatimation tad Evaluation of Cancer Risks Attributed to Air Pollution in Southend Chicago." Air and Radiation
Division, U.S. EPA Region V; and Engineering-Science, lac. 1990. "Tmiabouadary Air Toxica Study." Final Summary Report. Prepared
for Air and Radiation Division, U.S. EPA Region V.
' Op. Cit, EPA, 1989, and Engineering-Science, Inc. 1990.
-------
IV- 12
carbon tetrachloride, chloroform, chromium (hexavalent), coke oven emissions, dioxin, ethylene dibromide, ethylene
dichloride, ethylene oxide, formaldehyde, gasoline vapor, hexachlorobutadiene, hydrazine, methylene chloride
perchloroethylene, products of incomplete combustion, trichloroethylene, vinyl chloride, and vinylidene chloride.'
As necessary, OAQPS updated the cancer rates presented in the studies used in its report based on unit risk factors
used by EPA. Although the pollutants studied varied from proven human carcinogens to probable human carcinogens
to possible human carcinogens, all were treated equally in the analyses.
Noncancer Risk
Pollutants selected for evaluation of noncancer effects include acetaldehyde, acrolein, arsenic, benzene, beryllium
caibon disulfide, carbon tetrachloride, chloroform, ethylene oxide, formaldehyde, hydrogen sulfide, lead, methyl ethyl
ketone, methyl methacrylate, methyl isocyanate, nitrobenzene, perchloroethylene, phenol, phthalic anhydride, styrene
tetraethyl lead, toluene diisocyanate, and vinyl chloride.
Exposure to these pollutants in ambient air can result in noncancer health effects ranging from subtle biochemical
physiological, or pathological effects to the pulmonary, nervous, gastrointestinal, cardiovascular, and hematopoietic
systems. In addition, hepatic, renal, reproductive, and developmental toxicity have been observed, although not at
levels present in ambient air. Although air levels of lead have greatly declined, atmospheric lead can contribute to
the overall body burden of lead. Children six years old or younger are generally considered to be most vulnerable
to the adverse health effects of lead. The NAAQS for lead is an arithmetic average over a calendar year of 13
ug/m3.
Exposure Assessment
Cancer Risk
The OAQPS study used as the basis for the characterization of cancer risks for this problem area is based on
information contained in 10 area specific or national air quality based risk-related reports on air toxics, 14 EPA
source category and pollutant specific studies, risk assessments performed for the development of National Emission
Standards for Hazardous Air Pollutants, and source specific risk data contained in the EPA Air Toxic Exposure and
Risk Information System data base. National data was apportioned to the Great Lakes Basin by population using
the following equation:
Great Lakes Basin Data = (National Data) (Population of Great Lakes Basin)/(Population of Nation)
In the national study, cancer risk estimates were derived giving equal consideration to measured and m"dried data
provided that one estimate was not clearly preferable. Cancer rates for a pollutant and source category were
extrapolated to total nationwide estimates based on the geographic scope of each study examined. Direct
extrapolation to total nationwide estimates was possible for most pollutants due to their inclusion in at least one study
of nationwide scope. In instances where a pollutant was included in a study of limited geographic scope, the
concentration of the pollutant/source category in the area studied relative to the national concentration was
considered. This information was then utilized to extrapolate to nationwide estimates.
It is specifically noted that Region 5 conducted a risk assessment in Southeast Chicago for a variety of air toxicants
Those results were not independently utilized, however, they were incorporated in the national study that was
as the basis population risk estimate for this problem area.
-------
IV- 13
The national population utilized in the OAQPS report is 243,400,000. The population assumed in this analysis for
the Great Lakes Basin is 30,050,000 or approximately 12 percent of the national population. Further, information
from the Toxic Release Inventory indicates that the Great Lakes Basin emits about 350,000,000 pounds of air toxics
per year while approximately 2,420,000,000 pounds per year are emitted nationally.10 This corresponds to a Basin
contribution of approximately 14 percent Therefore, while this comparison is reasonably similar, the use of
population may somewhat underestimate the cancer effects of air toxics.
Noncancer Risk
The preliminary results of an OAQPS Broad Screening study of non-carcinogens indicate that- 1) approximately 48
percent of the chemicals studied exceeded the health reference levels for chronic exposure; 2) long-term (annual)
exposures were estimated for the Lowest Observed Adverse Effect Level (LQAEL) for 3-5 percent of the chemicals
studied; 3) more frequently, 58 percent of the exposures exceeded health reference levels for short term (24-hour)
exposures; and 4) in hundreds of U.S. cities, exposure to multiple pollutants was of concern, with concentrations in
260 cities exceeding the hazard index.
The preliminary results of OAQPS's Urban County study indicate that: 1) using long-term modeling of both average
and maximum emissions, a substantial number of facilities were estimated to cause exceedances of health levels, with
31 percent of 131 facilities exceeding chronic health effect levels for 9 chemicals; 2) using short term modeling,
more pollutants and facilities were associated with exceedances of LOAELs with and without uncertainty factors
applied, with 75 percent of 131 facilities exceeding acute health effects levels for 442 chemicals; and 3) for
chemicals of concern, substantial numbers of facilities were associated with exceedances of the health reference level,
and a small percentage of facilities emitted pollutants in quantities exceeding the LOAELs.
Based on a review of 1985 to 1989 AIRS monitoring data, only 1 of 33 Great Lakes Basin counties that were
monitored for lead reported exceedances of the lead standard. The exceedances occurred in Lake county, Indiana,
where a total five monitoring stations reported exceedances including two monitors located in East Chicago and three
in the city of Hammond. Seven violations occurred in Lake county in 1985, two in 1987, and one in 1989.
Risk Characterization
Cancer Risk
In the studies used in OAQPS's "Cancer Risk from Outdoor Exposure to Air Toxics", aggregated maximum lifetime
individual risks exceeding 10*4 were reported in almost every case. Risks of 10'J or greater from individual pollutants
were reported adjacent to many types of sources. Average lifetime individual risks in urban areas from exposure
to many pollutants were generally between 10*4 and 10"J but ranged from 10'3 or iff*. These levels were the result
of combined exposure to mobile and stationary sources. The two EPA studies on local effects estimated overall
individual risks of 2x10"* in Southeast Chicago11 and 9xl0"5 in the Southeast Michigan/Windsor area11
10 US. EPA. 1990. Draft of "Second Annul Report oa U.S. Progress in Implementing the Great Lakes Water Quality Agreement with
Canada."
u Op. Cit„EPA. 1989.
12 Op. Cit, Engineering-Science, Inc. 1990.
-------
IV- 14
Ib the OAQPS study, estimates of annual cancer incidence were initially derived by estimating the annua] cancer case
per million population for each pollutant source category combination reported in the data sources. These estimates
were then modified as necessary to reflect updated unit risk and emission factors. Total nationwide annual inri^mT
was then estimated by summing across all pollutant/source categories.
The procedure outlined above results in an estimate of 1,580 to 2,540 cancer cases per year, nationally, caused by
exposure to the pollutants listed previously.13 With the Great Lakes Basin comprising approximately 12 percent
of the Nation's population, apportioning national data to the region by population results in a projection of 190 to
310 cancer cases per year in the Great Lakes Basin. The two local area risk assessments discussed above (i.e.,
Southeast Chicago and the Transboundary area) estimated approximately 1 case per year in Southeast Chicao and
5 cases per year in the Southeast Michigan/Windsor area.
Noncancer Risk
Noncancer risks from exposure to toxic pollutants that are routinely emitted to the air by industrial or commercial
sources are being evaluated by the OAQPS in a Broad Screening Urban County Study. Based on analysis of the
preliminary data available from the study, it is reasonable to conclude that acute and chronic environmental exposures
to toxic air pollutants have the potential to adversely impact public health, although the exact magnitude of tbe
increased risks identified in this project is unclear. It is expected that short-term intermittent releases may be
expected to affect greater numbers of individuals than long-term emissions.
There is some uncertainty associated with characterizing noncancer health risks at exposures greater than tbe
reference dose and less than the LOAEL. Nevertheless, using the Broad Screening portion of the study, it is
estimated that 58 million people and 38 million people nationally arc exposed to levels of pollutants that are greater
than health reference levels for acute effects and chronic effects, respectively.
Provided that Basin characteristics are similar to national characteristics, an estimate of tbe Great Lakes Basin values
can be detennined using population as an indicator. National data was therefore apportioned to the Basin population
resulting in an estimate of 7.2 million people and 4.7 million people in die Great Lakes Basin exposed to levels of
pollutants which are greater than health reference levels for acute and chronic effects, respectively.
All of the five monitors registering violations of tbe lead standard in the Great Lakes Basin were located in two cities
in Lake County, Indiana (i.e., East Chicago and Hammond). Tbe populations of these cities were therefore
considered to be reasonable representation of tbe number of people potentially at risk of experiencing adverse health
effects due to exposure to airborne lead. Using 1980 population data, the total number of people considered to be
at risk in the Great Lakes Basin is approximately 134,000.
Ecological Impacts
With respect to terrestrial ecosystems, air toxics may have similar effects on animal populations as they do on
humans. However, studies on wildlife cannot readily be performed either through epidemiological or laboratory
approaches. Further, sufficient information could not be found to characterize die effects of air toxics on other flora
and fauna.
Atmospheric deposition is generally believed to be the major pathway for many toxicants reaching tbe Great Lakes,
11 These national numbers are different than those prtMMed in the OAQPS report since the risks due to waste treatment, storage and
disposal facilities, radionuclides, and radon have been subtracted. This is consistent with the problem area definition and description.
-------
IV- 15
this pathway is discussed in Section J (Atmospheric Loading to Surface Water).
Welfare Impacts
Quantified economic damage due to hazardous and toxic air pollutants are based on estimates of health effect costs
associated with the number of annual cancer cases presented above. This method was developed by RCG/Hagler,
Bailly, Inc. under contract to EPA and was used in the Region 5 comparative risk study. Economic effects associated
with noncancer health endpoints were not estimated, but based on the laige population exposed to pollutant levels
greater than health reference levels, these effects could be significant in the Great Lakes Basin.
To estimate costs, the estimated annual cancer cases in the Great Lakes Basin from air toxics were multiplied by the
direct medical cost and foregone earnings per cancer case according to the following equation:
HC * (Annual Cancer Cases) (Direct Costs and Foregone Earnings)
where:
HC = health costs
Estimated direct and indirect medical cancer costs are based on a range of cost per estimates. The lower bound
estimate, based on Hartunian, et al., is $80,000, while the upper bound estimate developed by the American Cancer
Society is $137,000. These estimates provide differing values for foregone earnings and medical costs. Both
estimates are weighted average costs associated with all types of cancer. Loss of earnings estimates ranged from
a lower bound of $11 million to an upper bound of $32 million.
-------
o.
ACTIVE HAZARDOUS WASTE MANAGEMENT FACILITIES
IV- 16
Problem Area Definition
The Active Hazardous Waste Management Facilities problem area evaluates the risks to human health and die
environment posed by facilities that generate, store, treat, and/or dispose of hazardous wastes. Specific facilities and
activities covered in this problem area are:
¦ hazardous waste generating sites, including industrial plants and other facilities producing and accumulating
hazardous wastes;
¦ hazardous waste storage and treatment facilities;
¦ land disposal units including landfills, land treatment units, waste piles, and surface impoundments;
¦ hazardous waste incinerators;
¦ boilers and industrial furnaces using hazardous waste as fuel;
¦ hazardous waste recycling units currently exempted from RCRA; and
¦ inactive solid waste management units at active hazardous waste facilities;
The types of risks to human health and the environment resulting from active hazardous waste facilities include those
resulting from accidental or non-accidental releases of hazardous waste and waste constituents to air, soDs, surface
water, and ground water. This analysis focuses on releases of hazardous waste constituents to ground water and the
resulting risks to human health and the environment posed by these releases. The analysis does not address risks
from air and soil contamination, nor does it address the risk posed by the trans portion of hazardous waste to and
from RCRA hazardous waste facilities.
There are approximately 450 hazardous wastes including solvents, process wastes, and discarded commercial
chemical products generated, stored, treated, or disposed at RCRA Subtitle C facilities and which pose a potential
threat to human health and the environment14
The total number of estimated operating and corrective action units for the Great Lakes Basin is seen below in
Exhibit O-l. Exhibit 0-2 depicts the land disposal facilities throughout the eight Great Lakes States. Exhibits 0-3
through 0-9 provide the geographic distribution for RCRA facilities of different sizes and types.
14 Hie Region 5 problem area paper was examined to determine whether jt could be applied to Ike Great Lakes Basin. The Region 5
analysis evaluated a sample of 49 RCRA facilities and used the results to characterize the remainder of Region 5 RCRA facilities. Brra'm a
majority of the Region 5 sample sites (52%) as well as a majority of all Region 5 RCRA sites are located in the Great' Basin, these
same facilities were determined to be representative of the Great Lakes Basin RCRA sites. Thus the Region 5 methodology was Ttf aa a
basis for the Great Lakes Basin risk analysis.
-------
IV- 17
EXHIBIT O-l
RCRA Units in the Great Lakes Basin Region
Land Disposal Units
(LDUs)
Storage and Treatment
Units (STUs)
Incinerators
Operating*
183
407
18
Corrective Action*
284
977
NA
SOURCE: HWDMS, 1991, for Great Lakes Basin.
Estimated using ratio of corrective action units to operating units in Region 5.
Human Health Risk Assessment
Toxicity Assessment
Active hazardous waste management facilities manage a wide range of wastes containing hazardous constituents.
The random sample of 49 hazardous waste units used in the Region 5 comparative risk analysis was used to evaluate
potential risks from hazardous waste facilities in the Great Lakes Basin. The broad categories of waste and
hazardous constituents reported by these facilities include:
0 • chlorinated and non-chlorinated solvents;
m electroplating solutions and wastewater treatment sludges containing heavy metals and cyanides;
B acids and bases from electroplating, metal finishing and other processes;
a process wastes and sludges from petroleum refining, chemicals manufacturing, and wood treating containing
heavy metals and toxic organics; and
¦ characteristic wastes which are ignitible, reactive, corrosive and toxic.
In order to evaluate risks from these hazardous waste facilities, several hazardous constituents that are representative
of common waste streams reported for these facilities were selected. These wastes include halogenated and
nonhalogenated solvents, heavy metal bearing sludges, and metal finishing wastes. These constituents were also
found to be present as environment*} contaminants at six or more sites in a sample of 19 corrective action sites. The
constituents selected are benzene, trichloroethylene, tetracbloroethylenc, lead, cadmium, and chromium. Exposure
to the constituents may result in cancels of the liver, kidney, lung, and respiratory tract; leukemia; hepatotoxicity
(liver damage); and kidney damage. Cancer slope factors for the Region S analysis for LDUs ranged from 9.5 x 10"1
(mg/kg/day)'1 (tetrachloroethylene inhalation) to 5.1 x 10*1 (mg/kg/day)"1 (tetrachloroethylene ingestion). Reference
doses (for noncancer effects) ranged from 0.01 mg/kg/day (tetrachloroethylene, chronic effects) to 10 mg/kg/day
(chromium, oral exposure) subchronic effects. Toxicity for tanks, incinerators, and waste fuel burners were
incorporated into the risk numbers obtained from other sources (see Risk Characterization).
-------
Exhibit 0-2
Lake Superior
HP Lake
Michigan
SITES PER
COUNTY
~
II 12
¦
RCRA FACILITIES
LAND DISPOSAL
Lake Huron
LakaOmario
Lake
Erie
Source: Hazardous Waste Data Management System, 1990
ebu
UNITfD STATES
1 ENVItONMINTAL PIOTfCTION ACENCT
1 CIEAT lAfEJ NATIONAL MOCIAM OFFICE
-------
Ehibit 0-3
!!!!!!!! !!
Lake Superior
Lake
Michigan F
SITES PER
COUNTY
1 - 49
50- 99
100-999
>999
RCRA GENERATORS
LARGE QUANTITY
Lake Huron
Lake
Ontario
Lake
Erie
Source: Hazardous Waste Data Management System, 1990
X
UNITED STATES
ENVItONMENTAl PIOTECTION AGENCY
CHAT IAKES NATIONAL PtOGtAM OfFICE
-------
Exhibit 0-4
iiiiiiiiiii
SITES PER
COUNTY
~
1 - 99
100- 499
500- 1499
Lake Superior
Lake
m
::j:i
> 1500
RCRA GENERATORS
SMALL QUANTITY
Lake Huron
Lake Ontario
Lake
Erie
Source. Hazardous Waste Data Management System. 1990
N
Si ERA
1
mm mm
WVII0NMMTA1 MOTfCTION ACINCT
CHAT I AXIS MATIOMAt H06IAM OHICI
-------
Exhibit 0-5
V: I:::
SITES PER
COUNTY
~ 0
1 - 19
20- 99
100-299
>300
RCRA GENERATORS
VERY SMALL QUANTITY
Lake Ontario
Hazardous Waste Data Management System, 1990
N
saERAI
UNITED STATES
ENVIIONMEKTAl PIOTECTION ACWCT
CIEAT LAKES NATIONAL PIOCtAM OFFICE
-------
Exhibit 0-6
Lake Superior
~
rn
u
Lake
Michigan
SITES PER
COUNTY
RCRA FACILITIES
INCINERATION
Lake Huron
i Ontario
Lake
Erie
Source: Hazardous Waste Data Management System, 1990
X
\
UNITED HATES
I WVIIONMNTAl HOTECTION ACENCT
HUT I ACES NATIONAL PIOCIAM OFFICE
-------
Exhibit 0-7
Lake Superior
35
~
Michigan
SITES PER
COUNTY
~ 0
6-15
16-35
RCRA FACILITIES
STORAGE AND
TREATMENT
Lake Huron
Lake Ontario
'Lake
Erie.
Source: Hazardous Waste Data Management System, 1990
N
ERA
UNITED STATES
ENVIIONMEWTAL PIOTECTION ACENCT
CIEAT LAKES NATIONAL PIOGIAM OFFICE
-------
Exhibit 0-8
Lake Superior
3
E
SITES PER
COUNTY
~
RCRA FACILITIES
CLEAN - CLOSED
Lake Huron
Lake Ontario
Lake
Erie
Source: Hazardous Waste Data Management System. 1990
N
EFtt
UNITED STATES
ENVIIONMfNTAl HOTECTION ACENCT
CIEAT LAKES NATIONAL PIOCIAM OFFICE
-------
Exhibit 0-9
HAZARDOUS WASTE
FUEL BURNERS
Lake Superior
Lake Huron
Lake Ontarit
LakeX.
Michigan
HAZARDOUS WASTE
FUELBURNERS
Great Lakes Basin Counties
Source: Hazardous Waste Data
Management System, 1990.
GREAT LAKES NATIONAL PROGRAM OFFICE
&ERAI
US ENVIRONMENTAL PROTECTION AGENCY
-------
rv- 18
Exposure Assessment
Potential routes of exposure to releases of hazardous constituents from hazardous waste sites include exposure
through ground-water contamination, surface water contamination, air emissions, and direct contact with hazardous
wastes. In general, exposure through ingestion of drinking water has been assumed for this analysis as the most
important route of exposure to hazardous constituents from operating hazardous waste facilities and corrective action
sites. Therefore, this assessment is focused on exposures through ground water for evaluating risks from operating
LDUs, storage and treatment units (STUs), and for evaluating risks from all units projected to undergo corrective
action. It is also focused on exposures through air for evaluating risks of operating incinerators.
Surface water has not been included as a significant route for human exposure to contaminants from active hazardous
waste facilities. Public water supply systems generally are used when the water supply is surface water, and
systems are highly regulated. Hence, the probability of significant exposure to hazardous constituents from drinking
water from these systems is assumed to be quite low, since any contamination of these public drinking water supplies
would be greatly reduced as a result of dilution and contaminant reduction through treatment.
From the Region 5 analysis, concentrations in water for LDUs and corrective action sites are assumed to range from
0.01 mg/I for benzene, to 0.02 mg/1 for trichloroethylene, to .35 mg/1 for tetrachloroethylene. These concentrations
then were adjusted by a factor of 10 for the 1 km distance and 1000 for the 5 km distance to reflect dilution and
attenuation between the facility boundary and tbe wells (see below). Exposure concentrations for tanks, incinerators,
and waste fuel burners were incorporated into the risk numbers obtained from other sources.
The potentially exposed population from RCRA units described above was estimated for the Great Lakes Basin
through a process of extrapolating the data used in the comparative risk analysis perforated for Region 5. The
methodology used in that analysis also was applied. The Region 5 comparative risk analysis estimated the potentially
exposed populations using the following assumptions:
¦ Exposures through ground water would occur through private residential wells, and not through public wells
supplying public systems due to Safe Drinking Water Act requirements for public system safety.
¦ Areas with residential populations of less than 1,000 within 1 km of the hazardous waste unit, or less than
25,000 within 5 km are primarily niral/subuiban in nature and drinking water for these residents would be
provided by private wells.
¦ About 25 percent of the people residing within 5 km of facilities located in a rural/suburban area (i.e., using
private wells) could be potentially exposed to ground-water contamination based on potential plume size:
¦ Those residing within 1 km would be exposed at contaminant concentrations equal to the fenceline
concentration divided by a dilution/attenuation factor of 10.
¦ Those residing within 5 km but greater than 1 km would be exposed at contaminant concentration
equal to tbe fenceline concentration divided by a dilution/attenuation factor of 1000.
Using these assumptions, the potentially exposed population was calculated by:
¦ determining tbe percentage of units in the sample with populations of less than 1,000 within 1 km, and
25,000 within S km;
¦ assuming an average population of 500 within 1 km, and 12,500 within 5 km, of facilities within these
-------
IV- 19
categories; and
* scaling up the universe of operating units and units projected to undergo corrective action by using the
resulting percentages.
75 percent of LDUs were assumed to eventually result in ground-water contamination (based on a sample of Region
5 Environmental Priority Initiative [EPI] scoring results). This is a conservative assumption in that:
¦ Operating units meeting requirements for liners and leachate collection systems, and land disposal
restrictions, are not expected to result in environmentally significant ground-water contamination; and
¦ Confirmed ground-water contamination exists at a much lower number of LDUs than that estimated using
the 75 percent assumption.
One percent of operating STUs and 100 percent of corrective action STUs were assumed to eventually result in
ground-water contamination.
This methodology was applied to the universe of operating and corrective action units in the Great Lakes Basin with
certain modifications due to data gaps and the fact that the Great Lakes Basin is more urban than Region 5 and
therefore has greater population density. That is, the Region 5 analysis assumed that 66% of the LDUs are located
in rural areas where private wells are used. Because the population density of Region 5 is about 80% of the Great
Lakes Basin (140 people per square mile in Region 5 versus 180 people per square mile in Great Lakes Basin), and
therefore the Great Lakes Basin is assumed to have fewer of its LDUs in rural areas where private wells are used,
the percentage of LDUs located in rural Great Lakes Basin areas was estimated to be 0.8 x 0.66 ¦ 53 percent A
similar adjustment was made for STUs and incinerators. The estimated potentially exposed populations within 1 km
and 5 km of operating or corrective action units is shown in Exhibit O-10.
Human Health Risk Characterization
This section addresses cancer and noncancer risks from RCRA hazardous waste management facilities.13 The
following subsections discuss land disposal units, storage and treatment units, incinerators, and hazardous waste fuel
burners.
Operating and Corrective Action Land Disposal Units
For operating corrective action and LDUs, cancer risks were calculated for the Region 5 analysis based on
contaminant levels observed in ground water at a sample of LDUs requiring corrective action. The concentrations
used are pre-cleanup levels, which exceed applicable ground-water protection standards for operating units and for
corrective action units. Under the current RCRA program, these levels of contaminants would not be allowed to
occur over a long time period without remediation. Therefore, the estimate prepared is an upper bound risk.
The estimated individual incremental lifetime cancer risk resulting from these units and corrective action sites ranges
from 2 x 10"4 to 3 x 10"6 (see Exhibit O-ll). Individual risk estimates for operating and corrective action LDUs were
apportioned by risk level, based on the percentages determined in the sample (i.e., some sites are represented by
tetrachloroethylene and some by both benzene and trichloroethylene; see Exhibit 0-11).
" Due to lime constraints, noncancer health effects were not addressed.
-------
IV - 20
Calculations of individual cancer risks for ground-water contamination from these operating units are shown in
Exhibit 0-12.
When applied to the exposed population, these risk levels result in an estimate of 0.77 additional cancer cases from
operating units over a 70-year period (0.01 cases per yeaT) and 1.2 additional cancer cases from corrective action
units over a 70-year period (0.02 cases per year).
EXHIBIT 0-10
Approximate Potentially Exposed Populations*
Land Disposal Units
Storage & Treatment Units
Incineratorsb
Operating
Corrective
Action
Operating
Corrective
Action
Operating
Exposed Population,
1 km
9,089
14,090
224
5,374
NA
Exposed Population,
5 km
227,293
352,424
5,482
1,343,116
1,200,000
* Estimates of populations are based on a sample of 49 operating RCRA Subtitle C facilities in Region S.
Population characteristics of these facilities were applied to all RCRA facilities located in the Great Lakes Basin.
b Estimate of exposed population was derived using per unit exposure in Region 5 and adjusting for greater
urbanization of Great Lakes Basin.
-------
IV. 21
EXHIBIT Oil
Calculation of Individual Carcinogenic Risks: LDUs
Constituent
Slope Factor
(ppm/d)'1
Concentration
in water
Chemical Specific
Risk
Oral
Inh.
(mg/I)1
Onlb
Inh.
Tetrachloio-
ethytene
5.1 x 1CT1
95 x 10 '
035
2x1 (T4
2x10*
Benzene
2.9 x lO*1
83 x 10*
0.01
4x10*
5 x 10l#
TricWoro-
ethylene
1.1 x 10"J
1.7 x 10*
0.02
3 x 10*
2 x lO"10
Total Pathway
Risk
2x1
3 x 10*
Total Exposure
Risk
2 x 10^
* Concentrations are based oa avenge of maximum concentrations observed in monitoring wells at correcrive action sites in Region 5. Concentrations have
been adjusted by a factor of 10 lo reflect dilutkm/attenution between the facility boundary and residential welb for populations within 1 km.
' Assumes oral exposure through drinking water.
-------
IV - 22
EXHIBIT 0-12*
Cancer Risk Levels From LDUs
Percent of LDUs
Distance to Private Well (D)
Risk
32%
D< 1 km
2x 10-4
68%
D < 1 km
7x 10*
32%
1 km < D < 5 km
2 x 10*
68%
1 km
-------
IV-23
Cancer Risk Summary
Exhibit 0-13 shows the calculations of cancer risks for operating units and corrective action sites. The total
estimated incremental annual cancer rates associated with this problem area is about 0.16 cases per year, with a total
of about 10.9 cases attributable to this area over a 70-year period. Seventy-two percent of these cases are attributable
to sites requiring corrective action, rather than to operating units.
Certain caveats need to be pointed out in order to put these estimates in proper perspective. First, virtually all the
estimated additional cancer cases are attributable to ground water contamination at "pre-remedial" concentrations,
and the risk calculation assumes continual exposure to contaminated ground water from operating units and corrective
action sites over a long period. However, die RCRA program is designed to be a virtual "zero discharge* program
with respect to ground water. That is, the program is designed to prevent any future discharges of hazardous
constituents from operating facilities to ground water, and to require the deanup of any past environmentally
&ignificant discharges of constituents to ground water.
Therefore, continued permitting, inspection, and enforcement of current standards into the future should drive actual
risks from both operating land disposal units and corrective action sites towards very low levels; however, without
this activity to inspect facilities and enforce standards, risks could be significant
Second, risk from discharges to the environment allowed under the program, such as emissions from incinerators
»nd hazardous waste fuel burners, are relatively low even under conservative assumptions if emissions meet
applicable requirements. Incinerators and hazardous waste fuel burners account for an estimated additional 3 cancer
cases in the Great Lakes Basin. Third, the estimate does not include risks arising from a variety of activities
generally thought to pose relatively low levels of risk, such as hazardous waste generation, accumulation at generator
facilities, and storage in containers. The number of generators greatly exceeds the number of TSDFs, and generators
are frequently located in densely populated cities. Generally, these generators accumulate wastes in containers or
tanks before shipping them off-site; in fret, the number of tanks estimated to exist at generator sites in the Region
for waste accumulation exceeds the number estimated to exist at TSDFs. Although the facility-specific risks from
accumulation in containers and tanks is probably very small, the cumulative risk may be found to be significant if
data are developed to better characterize potential hazards.
Exhibit 0-13
Estimated Potential Cancer Cases From Operating Sites
unit
Life-Time
Cancer Risk
Level
Estimate
Population
Exposure (Oper)
Potential Addit '
Cancer Cases
(Oper)
Estimate
Population
Exposure (CA)
Potential
Addit Cancer
Cases (CA)
Land Disposal
Units (1 km)
2x10-
2,908
si x io*j
4,509
0.906
Land Disposal
Units (1 km)
7X 10'
6,181
4.3 X lir
9,581
0.067
Land Disposal
Units (5 km)
2x10*
72,734
1.4 X UP
112,858
0.22
Land Disposal
Units (5 km)
7 x 10"*
154^159
1.1 X 10'
23&JS66
0.016
Storage and
Treatment
1 x 10*
22.4
12 x lO-1
' sjm
5.4
-------
JV-2*
Unit
Life-Time
Cancer Risk
Level
Estimate
Population
Exposure (Oper)
Potential Addit.
Cancer Cases
(Oper)
Estimate
Population
Exposure (CA)
Potential
Addit Cancer
Cases (CA) _
Land Disposal
Units (1 km)
2 x 10"*
2,908
5.8 x 10"
4,509
0.90d
Storage and
Treatment
i x 10*
214
2.2 x 10 *
5374
0J4
Storage and
Treatment
1 x 10"'
44.8
4.8 x 10J
107.5
1 x
Incinerators
1 x 10*
2.0
NA
NA „
Haz. Waste Fuel
Burners
NA
NA
T.0
"NA
0.00 ~
Total Additional
Cancers
3.80
7.15 "
—==
Non-Cancer Effects
A
Chemicals causing non-cancer health effects are presumed to act via a different mechanism than carcinogen* Jy
threshold below which advene effects do not occur can generally be demonstrated to exist for such compounds.
lifetime daily dose which is likely to be without appreciable risk of deleterious effects to the human populate0*
defined as the Reference Dose (RED). Potential non-cancer health effects of constituents are evaluated through ft*
use of hazard quotients. The hazard quotient is defined as the ratio of the exposure dose to the RID. The sun*0
hazaid quotients for all constituents is the hazard index.
Potential non-cancer health effects from RCRA sites in the basin were extrapolated from the Regional analysis usi"S
the same assumptions described above. Of the six selected constituents, only the potential estimated concentration
for tetrachloroetbylene exceeded its RfD for chronic exposure. Assuming a concentration in water of 3SsagHt ®e
hazard quotient for chronic exposure to tetrachloroetbylene was calculated to be 4.29. The health effect that
occur due to exposure to tetrachlofoethylene in excess of the Rfd is liver damage. Of the remaining five select
compounds, lead deserves special mention. The well documented association between lead exposure and learni*^
disabilities in children makes lead a compound of great concent, but the lack of a Rfd for this metal makes it diffi®1"
to assess the problem quantitatively.
Ecological Impact17
The extent of ecological threats posed by RCRA sites has been identified in many regulatory impact analyses. to®*
has not been quantitatively discussed. Most ecological threats have been caused by unlined landfills and surfW®
impoundments, storage in waste piles, improper disposal, and waste water discharges. All of these can produce to*'c
leachate which can then contaminate both ground and surface waters.
Severity and Toxicity
" USEPA Office of Policy Analysis and Office of Policy, Pluming, tad Evaluation's "Summary of Ecological Risks, Asseeneat
Methods, sad risk Management Decisions in Superfund and RCRA," lane 1989, pH-5.
-------
IV - 25
both die severity and Toxicity of RCRA sites on the environment are dependent upon:
a environmental setting;
B facility size;
¦ toxicity of released constituents; and,
¦ the nature and extent of the release.
These releases can impact surface water, wetlands, and special habitats.
Ecological Risk Assessment
The effects on ecosystems can include:
¦ fish kills;
¦ diseased benthic invertebrates;
11 chronic or behavioral effects on aquatic plants and animals;
¦ reduced floral and faunal species diversity;
¦ sediment contamination;
¦ decreased productivity in the wetlands; and,
¦ impaired health and fertility of plants and animals.
Perhaps the most fragile ecosystem in the Basin is the wetland habitat Wetlands are heavily impacted by leachate
waters, and often times the habitat is destroyed.
Reversibility
Destruction of the wetlands and special habitats is a serious concern because many ecosystems cannot be restored.
Consequently, these effects are virtually irreversible. Contaminated ground water can be treated, although this
process is very expensive.
Welfare Assessment
Economic damages associated with active hazardous wastes include damages resulting from diminished property
values adjacent to these sites and the costs associated with human health effects attributable to these sites.
Property Value Damage
To estimate the value of losses associated with active hazardous waste facilities in the Great Lakes Basin, the
following equation was used:
$ Damages" ¦ (households residing within 1 km of active sites) ($ damage per household)
Empirical research suggests that values of residential properties increase as die distance from active hazardous waste
" Not corrected for 1991 U.S. dollars.
-------
IV - 26
facilities increases. These studies use econometric analyses to estimate the difference in value of residences near
a site compared to homes further from the site. The study results indicate the following:
¦ Homeowners in the Boston area are willing to pay from $300 to S500 per mile for a location more
distant from a hazardous waste landfill (Smith and Desvouges, 1986.)
¦ The mean willingness to pay of homeowners in the Boston area for an increase of 0.8 miles in
distance from a Superfund site was $69 in the average housing market (Michaels et al., 1987).
¦ McClelland, et aL (1989) determined that, on average, housing prices were about $4,800 lower for
residents concerned about the proximity of a Superfund site than they would have been if residents
were not concerned about the site.
To estimate a range in potential damage to property values resulting from active hazardous waste facilities in the
Great Lakes Basin, a lower bound cost of $69 and an upper bound cost of $500 per home residiqg within a kilometer
of a site, was assumed. A worst case of $4,800 per home was assumed. These costs were taken from the previously
mentioned studies. Exhibit 0-14 summarizes the property value damages associated with these costs.
Health Care Costs
To estimate the health care costs (HCC)19, the annual cancer cases previously determined were multiplied by the
direct medical cost and foregone earnings per cancer case.
HCC s (Annual Cancer Cases) (Direct Costs and Foregone Earnings)
Lower bound estimate:
HC « (10.3) ($80,000) - $824,000
Upper bound estimate:
HC = (103) ($137,000) ¦ $1,411,100
"ICF Incorporated. 1988. 'Office of Solid Waste Emergency Response (OSWER) Gomparitive Risk Project: Ground Water Valuation
Task Force Report." Prepared for U.S. EPA, Office of Underground Storage Tanks.
-------
IV-27
EXHIBIT 0-14
Estimated Property Value Damages ia the Great Lakes Basin
Facility Type
Number of Sites
Population*
Households'
Lower Bound
Damage
Upper Bound Damage
Maximum Damage*
Land Disposal Unit
191
347,700
130,200
$8,986,000
$65,116,000
$625,108,800
Storage and Treatment
Tanks
296
712,200
266,800
$18,406,000
$133376,000
$1,280,409,600
Incinerator
17
34,800
13,000
$900,000
$6,525,000
$62,635,200
' Great Lakes Basis population within 1 km of all facilities = (Region 5 population within 1 km of Region 5 facilities/number of facilities in Region 5) (number of facilities in
the Great Lakes Basin)
1 Number of households = population/2.67 people per household. Avenge household size estimated by the U.S. Department of the Census.
' Maximum estimated property damages is $4,800 per boose ho Id.
-------
IV. 28
P. ABANDONED HAZARDOUS WASTE SITES
Problem Area Description
Abandoned Hazardous Waste Sites covers the risks posed to human health and the environment by abandoned waste
treatment, storage, disposal, and recycling facilities, illegal dumpsites, and other abandoned waste sites. Specific
categories of sites covered here included:
¦ Super fund National Priority List (NPL) sites (including Federal facilities);
¦ State Superfund sites;
¦ other sites reported to EPA and listed in CERCLIS (including Federal facilities, sites scored using the
Hazard Ranking System [HRS], and sites remaining to be scored); and
¦ other unmanaged hazardous waste sites that are no longer in operation.30
Magnitude of the Problem
There are currently about 4,100 sites21 within the Great Lakes Basin that have either been assessed for inclusion
on the NPL or that have been targeted for assessment in the future.2 One hundred and forty (140) of these sites
are on the NPL and 23 have been proposed for the NPL. Exhibit P-l presents the distribution of National Priorities
List sites across the Basin, and Exhibit P-2 depicts the distribution of all sites in the CERCLIS system.
Human Health Risk Assessment*
Toxicity Assessment
Potential carcinogens found at significant concentrations at a majority of the 51 Region 5 sites included
polychlorinated biphenyls (PCBs), polycydic aromatic hydrocarbons (PAHs), benzene, nickel, methyl chloride, vinyl
chloride, and trichloroethene. Hie cancer slope factors for these chemicals range from 0.011 (ppm/day)'1 for
trichloroethene to 7.7 (ppm/day)'1 for PCBs.
For noncarcinogenic effects, arsenic, chromium, cadmium, lead, nickel, toluene and zinc were studied. The reference
doses (RfDs) for these chemicals range from 0.01 ppm/day for nickel to 5.0 ppm/day for chromium.
" The Region 5 draft problem area paper wit examined to determine whether that methodology could be used as a basis for the Great
Lakes Basin analysis. Region 5 evaluated a sample of 51 Regional site* (49 on the NPL and two non-NPL) and used them as the h««;T fa,
characterizing the remainder ot the Region's sites. Because Region 5 contains most of the Great Lakes Basin sites (over 75 percent), these
same 51 sites were determined to be representative of the Great Lakes Basin sites. Therefore, the Region 5 analysis was determined to be
applicable as a basis for this Great Lakes Basin analysis.
" 4,109 sites listed in January 1990 CERCLIS report.
® Draft 1990 Great Lakes Basin Report to Congress.
29 Hie disunion of human health effects is a summary of the analysis conducted by Region 5.
-------
Exhibit P- 1
FINAL AND PROPOSED NPL SITES
Lake Superior
Lake Huron
Lake Ontario
Lake
Michiganl
Source; CERCLIS January, 1990
-------
Exhibit P-2
CERCLIS SITES
Lake Superior
Lake Huron
® Ontario
Source: CERCLIS January, 1990
-------
IV- 29
Exposure Assessment
Ground-water contamination was tbe exposure pathway of concern at the overwhelming majority of the sample of
sites (91.3 percent). Ground water was withdrawn for drinking water within three miles of 86 percent of the sites,
and there were operable wells within one mile of 96 percent of the sites. Of those drinking water wells within three
miles, the population served was greater than 10,000 in 45 percent of the cases, from 3,000 to 10,000 in 26 percent
of the cases, from 1,000 to 3,000 in 21 percent of the cases, and less than 1,000 in the remaining (eight percent) of
the cases. Tbe next most frequent pathway was air (4.7 percent), and then surface water (4.5 percent). The estimates
of population exposed at the 51 sites ranged from 38 persons drinking ground water up to 34,000 with a threatened
water supply. The types of exposures assessed within this problem area include those via air, surface water, ground
water, the food chain, and direct contact
Risk Characterization
Cancer Risks
Cancer risks were calculated based on the most probable cases, not the maximum. The values selected were based
on reasonable exposure pathways now and in the future, but not on such worst cases as a residential well in the
griddle of a site, or landfill leachate being used for drinking water. Wherever appropriate, a reasonable aggregate
risk number was selected.
The results of the analysis of 51 sites were extrapolated to the universe of Region 5 sites by assigning them risk
values in proportion to their perceived potential for human health effects, as follows:
0 The 267 NPL and proposed NPL sites were assigned risks in the same proportion as that found for the 51
reviewed sites. That is, excess lifetime cancer cases per site were 0.97 for sites with HRS scores > 45,0.58
for sites with scores in the range of 35 to 45, and 0.16 for sites with scores < 35. The resulting total cancer
cases were 100.
m 15 percent of the remaining sites (900) were determined to be high priority for further action. These sites
were assigned risks equal to the average risk level of the 51 sites that had HRS scores of 35 to 45. The
resulting total cancer cases were 520.
¦ 35 percent of tbe non-NPL or proposed NPL sites (2070) were determined to be medium priority for further
action. These sites were assigned risks equal to the average risk level of the 51 sites with HRS scores <
35. The resulting total cancer cases were 330.
¦ 50 percent of the non-NPL or proposed NPL sites (2970) were determined to require no further action.
These sites were assigned risks equal to 25 percent of the average risk level of the 51 sites with HRS scores
< 35. The resulting total cancer cases were 120.
The Region 5 analysis estimated 1,070 excess cancer cases for the approximately 6,150 sites in the Region. This
results in »bout 0.17 cases per site. Because the breakdown of Great Lakes Basin sites (i.e., NPL, high priority,
-------
IV - 30
medium priority, no further action; see discussion below) is similar to that of Region 5,14 a direct extrapolation
Region 5 to the Great Lakes Basin is considered reasonable. However, because site selection for the Region 5
analysis was not random, scaling from the Region to the Basin limits the statistical validity of the results.
The average cases of 0.17 per site is multiplied by the 4,100 sites in the Great Lakes Basin to result in about 700
excess cancer cases. On an annual basis, this represents approximately 10 additional Cahnw background)
cases per year.3* —
Noncancer Effects
Using the sample of 51 sites in basically the same way as with assessing cancer risk, the total universe of Region
5 sites were assessed for noncancer effects by assigning His in the same proportions as in the sample of sites.
¦ An HI of 1.1 was applied to a certain proportion of NPL and proposed NPL sites, with the resulting
population exposed at this HI estimated at about 10,110.
¦ An HI of 1.7 was applied to a proportion of high priority sites, with the resulting population exposed it this
HI estimated at about 27,480.
¦ All other sites were assumed to have an HI of less than 1.
The Region 5 analysis estimated that a population of 10,110 is potentially exposed to a hazard index (HI) of 1 l
and that a population of 27,480 are potentially exposed to an HI of 1.7. Extrapolating these populations to the Great
Likes Basin (by multiplying these populations by the ratio of the number of sites in the Great Lakes Basin f4 1091
to the number of sites in Region 5 [6,150]) results in approximately 6.750 people potentially exposed to an in 0f
1.1. and 18.360 potentially exposed to an HI of 1.7.
Ecological Risk Assessment
• . v ^ v- assessment, this Great Lakes Basin analysis relied on the Region 5 study of 51 sites.
As with the human health n 17 33 ^ of all sitcs (not counting those sites requiring no
H, analysis "J**CoMmuing with the Region 5 myology, i,
£UrUler Likes Basin sites ate mpousible for up to 16 stream miles of contaminated surface waten,
22?rJm mtelS. Wis, 1,090 s««n mile, of contaminated sediments, 8,330 acm of coniaminaKd soil, 2,950
acres of defoliation, and 2,950 acres of contaminated wetlands.
" For example, the percentages of Region S aad Great Lake* Basin file* that are not on the NFL or proposed for the NPL and that we
determined to Deed no further action ii approximately 50 percent.
° An attempt wac made to adjust the number of people exposed based on the higher density of people within the Great I n..j„
compared lo Region 5. This adjustment was not done, however, bwiusf of the complexity of effects on exposure from this population
density difference. For example, although the population exposed via air is expected to increase with increasing population density, the
population exposed vis ground water is expected to decrease because of fewer users of untreated ground water in urban areas.
* These results are a direct extrapolation from the Region 5 analysis. However, it is not clear exactly how the Region 5 analysis used
the numbers in the EPA 1989 report. The methodology, therefore, is being checked.
-------
IV-31
Toxicity Assessment
Three general types of stressors are released from abandoned hazardous waste sites: metals, pesticides, and organic
chemicals. Specific hazards associated with each type of stressor to aquatic and terrestrial ecosystems arc
summarized below.
Species level effects of metals on fish include neurotoxicity, impaired reproduction, reduced growth, damage to gill
surfaces and impaired respiration, mortality, and other effects. Ecosystem-level effects in aquatic systems include
reduced primary and secondary productivity, loss of top carnivores, changes in community composition, and
modification of nutrient cycling. In terrestrial systems, high concentrations of heavy metals are known to impede
root growth, nutrient uptake, and seed germination in plants. Some metals are known to cause eggshell thinning and
teratogenesis in birds. Airborne toxic metals can lead to death of almost all plant life growing immediately near an
emission source. Metals in high concentration put intensive selective pressure in favor of resistant biotypes,
eventually resulting in colonization by resistant species or subspecies.
In aquatic systems, species-level effects associated with pesticides include mortality among the aquatic juvenile life
stages of many insect species, herbicide-induced reduction of submerged aquatic vegetation, teratogenicity in Qsh,
impaired reproduction of fish, and fish kills. Ecosystem-level effects include reduced biomass and diversity of
invertebrate and vertebrate communities, herbicide-induced reduction of submerged aquatic vegetation, and
concomitant loss of invertebrate communities as well as fish that utilize plant cover for spawning or protection of
juvenile stages. In terrestrial systems, species-level effects associated with pesticides include bioaccumulation and
egg-shell thinning in birds, neurotoxicity, oncogenesis, teratogenesis, acute mortality (i.e., bird kills), mortality of
insect pollinators (e.g., honey bees), and non-target plant injury. Ecosystem-level effects include destruction of non-
target species key to ecosystem structure and function (e.g., parasites, pollinators), release of secondary pests from
the constraints of natural enemies, and loss of plant communities dependent upon pollinators.
In both aquatic and terrestrial systems, the possible species-level and ecosystem-level effects of toxic organic
compounds are diverse. In addition to producing adverse effects on reproduction, growth, and/or survivorship or
inducing organ dysfunctions, many organic compounds are carcinogenic. Organic compounds that are persistent and
bioaccumulate (e.g., PCBs, phenols) generally are of greatest concern.
Exposure Assessment
In the study of ecological effects from Region 5 sites, limited quantitative information was available. Over half of
the 51 study sites had no detailed ecological assessment, although some of these did list species and surrounding land
uses. Generally, the endangermeat assessments focused on ground water because of its human health implications.
Ground water, however, typically is a less important pathway for ecological effects compared to other pathways such
as surface water, unless of course the ground water discharges to surface water.
Several endangerment assessments evaluated the ingestion of fish as a human health exposure pathway because fish
near the sites had elevated levels of contaminants in their fat tissues. Such contamination is a persistent problem
in the Great Lakes. Other sites were associated with contamination of livestock and milk as far as a mile or more
away. These observations are more of a concern for human health, although they do indicate exposure to ecological
systems is occurring.
-------
IV. 32
Ecological Risk Characterization
Roughly 33 percent of those sites that bad an ecological assessment indicated signiGcant ecological impacts or
potential for impacts. For sites with no ecological assessments, lower and upper limits of the number of sites with
significant impacts were estimated as follows:
¦ For a lower limit for the sites with no assessment, assume no signiGcant ecological impacts ate expected.
¦ For an upper limit, assume 33 percent of the sites with no assessments are expected to have significant
ecological impacts (as with sites that have ecological assessments).
¦ Combine the lower and upper limits for sites with no assessments (0 to 33 percent, respectively) with the
33 percent estimate for sites with assessments to obtain a range of 17 to 33 percent
Applying the range of 17 to 33 percent of sites expected to have signiGcant ecological impacts to the 3,226 Region
5 sites (not including those needing no further action) resulted in 550 to 1,100 sites in Region 5 having the potential
for signiGcant ecological effects. Applying this to the 2,137 Great Lakes Basin sites (not counting those sites
requiring no further action), results in 363 to 705 Great Lakes sites having the potential for signiGcant eco1ooi/».i
effects.
The effects were quantiGed further by projecting the above results to a nation-wide Superfund study.77 Using the
upper limit on the number of Region 5 sites that have the potential for significant ecological effects (1,100), it was
estimated that Region 5 sites are responsible for up to 25 stream miles of contaminated surface waters, 350 stream
miles of fish kills, 1,700 stream miles of contaminated sediments, 13,000 acres of contaminated soil, 4,600 acres of
defoliation, and 4,600 acres of contaminates wetlands.
Some sites in Region 5 were found to exceed several of the Hazard Ranking system (HRS) measures of risk: the
acute acceptable concentrations (AACs), die chronic acceptable concentrations (CACs), lethal concentration 50s
(LC50s), and lowest-observed-adversc-effect-level (LOAEL). Nine of the 51 sites listed ground water or surface
water concentrations that exceeded at least one of the standards. At one site, discharges to a river from site leachate
violated AACs or CACs for six different contaminants. Two violated CACs even after dilution in the river, in
virtually all of these cases, the contaminants were metals - copper, zinc, chromium, and lead. PCBs exceeded a
standard at one site, and the endangerment assessment for that site noted that the immediate population of several
small animals were facing reproductive failure due to the PCBs.
Severity
The Great Lakes Basin sites have been ranked for severity of ecological effects as potentially high for local
and low to medium when the highly localized nature of the effects are factored in. Some sites appear to threaten
the long-term viability and stability of the local ecosystems, and reduce the population of one or more critical
species. Other sites may result in the loss of an entire ecosystem or the loss of several critical species.
Reversibility of Damage
The reversibility of damage to ecosystems, once source control measures are implemented, is a function of two key
v EPA, 1989, Summary of Ecological Risks. Assessment Methods. »nd Risk Management Decisions in SupcTfund »nd RCRA Office of
Policy, Planning, and Evaluation.
-------
IV-33
factors: the tim<- required for the existing contamination to abate, and the time required for the biological community
to be reestablished once the contamination levels subside. Some types of contamination (e.g., metal contamination
of soils or sediments) leave significant residual contamination that can require decades to decline to no-effect levels.
Once sources of new snd residual contamination have been eliminated, the time required for an aquatic community
to recover will depend upon many additional factors.
There are several iactors that influence the speed with which a healthy community can become reestablished once
contamination is removed (i.e., all sources including sediments). One key factor is the severity of the effects
incurred. If entire trophic levels were eliminated, a healthy biotic community would take longer to develop than if
only a few species had been eliminated or removed. The time required for full recovery also would depend upon
the areal extent of the contamination and the distance that would have to be travelled by recolonizing individuals
bom uncontaminated areas. Fish, birds, flying insects, and some plants can recolonize rapidly from long distances
provided that the intervening habitat is suitable. Most invertebrates and some plants recolonize more slowly.
Another factor is the level of residual contamination in biota.
Decades or centuries would be required for ecosystems to recover from metal contamination of soils or sediments
because these media continue to be a source of contamination of biota. Moreover, episodic acute exposures to
soil/sediment contaminants can occur when storms or other infrequent events cause soil erosion or resuspension of
contaminated sediments. Animals that have been exposed to heavy metals and that have developed significant body
burdens will remain contaminated for years or for life. Thus, biotic recovery after residual metal contamination
subsides will generally be slow because body burdens of the metals can contribute to lifetime depression of
reproductive success and other continuing sublethal effects in the longer-lived organisms. In the absence of
remediation, therefore, it can be assumed that reversibility is low.
For pesticides that bind to soils or sediments, reversibility of ecological damage can be slow following the institution
of control measures at the source because the soil/sediment can serve as a continuing source of contamination for
decades. Animals that have been exposed to persistent pesticides and that have developed significant body burdens
will remain contaminated for years or for life. In the absence of remediation, therefore, it can be assumed that
reversibility is low for moderate to highly persistent pesticides (i.e., DDT and other organochlorines). For less
persistent pesticides, the time required to reverse ecological damage is reduced. In the case of non-persistent
pesticides, recovery of an ecosystem can begin soon after the source of contamination is eliminated.
The potential reversibility of ecosystem damage resulting from releases of toxic oiganics to surface waters is less
for the persisitent, bioaccumulating compounds than for the nonpersistent compounds. Similar considerations to those
discussed above for pesticides apply.
Welfare Risks
Welfare risks associated with abandoned hazardous waste sites are estimated on the basis of damages resulting from
human health effects (i.e., health care costs), the costs of replacing contaminated drinking water supplies (i.c.,
environmental costs), and diminished property values (i.e., materials and property damage).
Health Care Costs
To estimate damages resulting from human health effects, the 10 annual cancer cases estimated previously are
multiplied by the direct medical cost and forgone earnings per cancer case. Estimated direct and indirect medical
-------
IV-34
cancer costs are based on a range of costs per case estimates. The lower bound estimate, based on Hartunian, et
al.M, is $80,000 (1988 $), while the upper bound estimate developed by the American Cancer Society is $137,000
(1988 $). Both estimates are weighted average costs associated with all types of cancers. When combined with the
number of annual cancer cases, the result ranges from $800.000 to $1.370.000 annually.
Environmental Costs
These costs are estimated by assessing the costs of replacing contaminated drinking water supplies. Given die
number of private well replacements and public water supply hook ups associated with ground-water contamination
due to abandoned hazardous waste sites, the cost of replacing contaminated drinking water supplies can be estimated
by multiplying this number by the costs of such replacements and hook ups.
The capital costs are assumed to be $3,500 for replacing a private well by digging a new well, and $300,000 for
replacing a public supply well by extending a hook up from another public supply. These were the upper bound
costs associated with replacing these two types of supplies in Region 34. These costs do not include the annual
operating costs for private and public systems.
Using data from the Region 5 report that indicates 86 percent of the sites are within three miles of operable wells
and that in 71 percent of these cases the population served is greater than 3,000 (probably indicating a public well
system), the following assumptions are made.
Assuming as an upper bound that of the NPL, proposed NPL, and high priority sites (a total of approximately 755),
86 percent would result in the need for replacing either one public well system (71 percent of the cases) or 100
private systems (the remaining 29 percent of the cases), the following equation is used to estimate environmental
costs:
EC * F, x S x [(F„ x Pu x C„) + (F* x Pr x C„)]
where
EC = environmental costs
Fj « 0.86 (ratio of sites within 3mi. of wells)
S = 755 sites (NPL, proposed NPL, and high priority)
Fh s 0.71 (ratio of sites with public wells)
Pu s 1 public well/site
CN = $300,000/public well
F* = 0.29 (ratio of sites with private wells)
Pr = 100 private wells/site
Cf, = $3,500/private well.
Solving for EC results in approximately $204 million (1988 $).
This result was compared to results derived from the OSWER comparative risk study.29 In that study, an estimate
" Referenced, a* is, in a memorandum from Don Peterson to Rosalie Day, Augutt 2, 1940.
" ICF Incorporated, OSWER Comparative Risk Project: Ground-water Valuation Task Force Report (Draftt. prepared for EPA, Office of
Underground Storage Tanks, February 4,1988.
-------
IV- 35
*•" - °~
Lkes Basin is estimated at from $?M mfllion to over Sl.6 billion.
Materials and Property Damage
, . nmnertv value losses associated with abandoned hazardous waste sites
TOe Region 5 m it m Rcgion 5, the lower bound estimate of total damages was estimated at
«¦*»¦¦«; *» .ppn,^-!, ?260,°00 „ 52,000,000 to
Sw a^Lte SI .!¦«• The Mtori* « ' «—»» of the Reg«» S Odin.
The equation used to estimate total damages was as follows:
DC « DH x HS x S
where
DC s total damage cost
DH ¦ damage cost per home
HS ¦ number of homes per site
S c number of sites.
Results of a growing body of empirical research suggest that values of midential properties increase as the distance
offcc*epASLreases from active hazardous waste facilities and other solid waste disposal sites. Usuig (1)
estimates ranging from $69 to $500 in damages per exposed home, (2) data from toe heahh «
populations potentially e*po*d, and (3) an estimate of 0.375 residences per individual potentully exposed, the
following was obtained:
¦ No farther rmedi.1 .cton .to »*•"»«'»*»<"' * » 10 ** raito<* C0"1 -
$0).
. Medium priority sites were assumed to result in $0 to $69 in damages per residence, with 2,347 residences
potentially exposed (total damages * $0 to $161,943).
. Hish priority sites were assumed to result in $69 to $500 in damages per residence, with 2,937 residences
potentially exposed (total damages = $202,653 to $1,468,500).
> Proposed NPL and NPL sites were assumed to result in $69 to $500 in damages per residence, with 2,740
residences potentially exposed (total damages - $189,060 to 1,370,000).
-------
IV - 36
Q. MUNICIPAL SOLID WASTE FACILITIES
Problem Area Definition
The Municipal Solid Waste Facilities problem area covers the risks to human health and the environment posed by
open or closed land disposal facilities, landfills, and open dumps used to dispose of municipal refuse; municipal
water and sewage treatment sludges; and municipal incinerator ash. This problem area also includes the risks from
municipal waste incinerators and municipal surface impoundments.
Hiese facilities can threaten human health and the environment through air emissions of volatile toxic chemicals and
methane gas, subsurface methane migration resulting in potentially explosive conditions in structures, and ground
and surface water contamination by landfill leachate containing organic and inorganic toxic contaminants and
pathogenic substances. Contamination may occur through subsurface migration, runoff, evaporation or wind erosion.
The types of substances disposed at these facilities include mixed municipal and commercial refuse, household
hazardous waste, hazardous waste from conditionally exempt small quantity generators, sewage treatment sludge,
incinerator ash, and water treatment sludge. These wastes may include some toxic contaminants, including solvents
and heavy metals.
Industrial waste disposal sites are covered in a separate problem area. Additionally, this problem area excludes
municipal solid waste sites that have been included on the National Priorities List
Magnitude of the Problem
There are approximately 6,034 operating municipal solid waste landfills in the United States and its 5 territories.30
These data are not organized by county; therefore, a precise census of the number of municipal solid waste facilities
to the Basin counties is not available. As a result, population size serves as a proxy indicator of the number of the
municipal solid waste facilities in the Basin. The Great Lakes Basin contains approximately 30 million persons or
12 percent of the U.S. population resulting in an estimate of 724 municipal solid waste landfills in the Basin. This
estimation may be low since Wisconsin contains 12.2 percent of the national operating municipal solid waste
landfills.
In addition to these operating landfills, there are approximately 1919 dosed municipal landfills in the Basin, based
on reductions in the population of operating municipal facilities reported by the states between 1981 and 1989.
Although there are many small landfills, national statistics show that over 40% of the waste is handled by very large
landfills averaging over 1,125 tons of waste per day.
Although there are many municipal solid waste facilities in the Great Lakes Basin, EPA has taken action to control
both active landfills and the siting and design of new landfills. EPA guidelines for municipal solid waste facilities
include that they:
(1) cannot be located in flood plains;
(2) cannot impact endangered species;
(3) must comply with the Clean Water Act;
(4) cannot contaminate ground water,
" Subtitle D Regulatory Impact Analysis
-------
IV-37
(5) cannot be near food chain crops;
(6) must apply a cover material; and,
(7) cannot bum waste openly.
Additionally, states have also instituted requirements which include, but are not limited to:
(1) administrative licensing;
(2) design standards;
(3) operation standards;
(4) location standards;
(5) monitoring;
(6) closure requirements; and,
(7) financial responsibility.
These emerging controls may limit the environmental risks posed by the municipal facilities. Nonetheless, the
following discussion focuses on the baseline risks present at facilities active during the mid-to-late 1980s.
Human Health Impacts
The municipal solid waste facilities problem area for the Great Lakes Basin are estimated at 724. Human health
risks have been estimated quantitatively only for operating municipal solid waste landfills, due to limitations in
available data. Human health risks for closed municipal solid waste landfills in the Basin were not estimated due
to a lack of available data on these facilities. These risks are discussed qualitatively below.
Toxicity Assessment
Documented contaminants from municipal landfills encompass approximately 200 constituents, including:
¦ Conventional pollutants, such as biochemical oxygen demand (BOD), chemical oxygen demand (COD), iron,
chloride, and ammonia.
¦ Heavy metals, such as arsenic, antimony, cadmium, and chromium.
¦ Organic chemicals, including vinyl chloride, tetrachlorocthylene, dichloromethane, carbon tetrachloride, and
phenol.
Risks to human health and the environment by selected hazardous constituents found in municipal landfill leachates
were based on:
(1) their prevalence in samples of leachate taken from multiple facilities;
(2) average concentrations found in leachate;
(3) toxicity;
(4) mobility; and
(5) persistence.
The constituents selected for risk modeling were vinyl chloride, arsenic, iron, tetrachloroethane, dichloromethane,
carbon tetrachloride, antimony, and phenol. Five of the contaminants are carcinogens. The noncancer effects of
these contaminants are neurotoxicity, cardiovascular changes, and kidney and liver damage. Information on the
carcinogenicity and the toxicity of these compounds is presented in Exhibits Q-l and Q-2.
-------
IV-38
CARCINOGENIC EFFECTS OF MSW LEACHATE CONSTITUENTS
CONSTITUENT
CARCINOGEN
ROUTE
Exhibit Q-l
CARCINOGENIC
WEIGHT OF
EVIDENCE
SLOPE FACTOR
ORAL (ppm/d)-l
SLOPE FACTOR
INHALATION
(ppm/d)-l
TUMOR SITE-
ORAL
TUMOR SITE -
INHALATION
1.1,2,2- Y
Tetncbloroe thine
Carbon Tetrachloride Y
Vinyl Chloride Y
Arsenic Y
Oral, lnh
Oral, lnh
Oral, lnh
Oral, lnh
B2
A
A
02
0.1
13
NA
050
0.13
0.295
SO
Liver
Liver
Liver
Skin
Liver
Liver
Lung
Resp. Tract
NA = not applicable.
Source: 'Health Effects Assessment Summary Tables - Third Quarter, FY 1990*. U.S. EPA OSWER, July, 1990l
-------
IV-39
Exhibit Q-2
SUBCHRONIC AND CHRONIC EFFECTS OF MSW LEACHATE CONSTITUENTS
Route: Oral
CONSTITUENT
Carbon Tetrachloride
Subchronic
Chronic
Antimony
Subchronic
Chronic
RFD
(ppm/d)
7E-03
7E-04
4E-04
4E-04
EFFECT OF CONCERN UNCERTAINTY FACTOR
Liver Lesions
Liver Lesions
Blood
Blood
100
1000
1000
1000
Arsenic
Phenol
Subchronic
Chronic
Subchronic
Chronic
1E-03
1E-03
0.6
0.6
Keratosis
Keratosis
Reduced Fetal
Body Weight
Reduced Fetal
Body Weight
100
100
Source: "Health Effects Assessment Summary Tables - Third Quarter, FY 1990", U.S. EPA, OSWER, July,
1990.
Two principle routes of exposure or risk were considered in this analysis: exposure due to migration of
contaminants from municipal solid waste landfills through ground water, and risk arising through subsurface
methane gas migration. Although exposure risk can be due to evaporation or wind erosion, these routes were
selected based on data availability and previous risk assessments. Other studies indicate that migrations of
contaminants and subsurface methane migration are the principle routes through which risks to human health
may occur.
Risk Characterization
Cancer Risks
Cancer risks were estimated using the cancer risk ranges developed in the Regulatory Impact Analysis (RIA) for
municipal solid waste landfills. Risk estimates from the RIA were used for facilities located in a "Migration
Potential (MP) IV" category location. This category represents facilities located in areas of high net precipitation
and short ground-water travel times due to shallow ground water and relatively permeable subsurface conditions.
Cancer risk estimates presented in the RIA represent average individual lifetime cancer risks due to exposure to
-------
IV- 40
leachate conttminated ground water. The analysis calculates ground water amcentrafcon averages based on a
300 year modeling period, in order to provide time for risks to be manifested (e.g., for cap failure, leachate
rdeasl, and conumtaant migration) under a variety of different design and location scenanos.
Be ria apportions facilities by risk range. For MP IV facilities, the distribution of facility risk is as follows:
¦Approximately 10 percent of facilities have risks greater than 1 x 10 3 (high)
¦17 percent have risks in the range of 1 x 10" to 1 x 10 (medium)
¦9 percent have risks in the range of 1 x 10 to 1 x 10 (low)
¦10 percent have risks lower than 1 x 10"* (very low) . ...
¦54 percent had zero risk because no wells were in the vicinity of the facility
For calculating the risks for this assessment, facilities and exjwsed populations were apportioned to risk rwges
based on these percentages, adjusted to eliminate the zero risk category since the nsk ranges were applied to an
.?¦; Of potentially «posed population. Wsto ™ then c.lculated by applying the rn.dp.int or the
risk range to the exposed population.
In ~ .be esttaute indiates .pp.oxim.tely 0.02 .dditional once* over. 70 ye„ operating^**, or 13
p., ye., in the potentially expoKd population. litis emanate can increase if new well,
drilled near landfills or in leachate paths.
Closed municipal solid waste landfills p»b.Uy po« ¦ *»"¦' orgreater threat of enviraunental contamination
.nd resulting risks thsn operating landfills. He populaoon .t nsk from ctaed faciliMs was not estnuted doe to
data limiutkms. However, as . group, dosed Imdfflta in tie Great Ute Bum .re likely to be smsller .nd
more rural tl»n cunently open ting tondfllls, doe to increases in l.ndQI costs Roasted wilh . higher level of
enviroiunenul protect^ .nd resulting economies of rale. For ex.mple,» 198 th«e were appro*,mately th™
times the number of currently operation munlcip.1 solid w«te landfill Ck»«d landfill, m., hive .n average
exposed population simil.r to the sample facilities, which were predominantly looted m rural .tea.
Noncancer Risks
Noncancer health effects were initially evaluated in developing the RIA criteria for municipal solid waste
UnS ESXZL* were found to be subst.nti.lly lower than the Reference Dos. (R®s) for the
due to the low concentrations of the constituents in leachate. Since noncancer health effects
SSS £££- h",th effectt >re oWed % f"*- 2*levds'or
NOAELs), and RfDs are based on these levels, no cases of noncancer health impacts were projected.
Risk from Explosions
There is a risk of explosion due to methane gas migration from municipal waste landfills to nearby smictarts
Methane may collect in confined spaces within structures at levels exceeding the Lower Explosive Limit (LEL)
»d result in an explosion if provided a source of ignition. THere are documented cases of such explosions.
In a,n«,l horizontal methane migration in the subsurface environment is limited because methane vents through
I8'ou"d surface. Typically, only structures located in proximity*, .landfill or on a filled area, are at risk.
hX™ where ventt® is prevented by pavement, locally saturated surface wtls, or frozen surface sotls, or
-------
IV- 41
where sewers or underground utilities provide a highly permeable pathway, methane may migrate substantial
distances (up to approximately 1000 yards). Great Lakes Basin states typically will have frozen ground
conditions for a portion of the year. This climatic condition will place the Basin at greater risk for explosions
than more temperate areas.
Based on the sample of facilities reported in Region 5, it is estimated that approximately 17,500 structures are
within 500 meters of municipal solid waste landfills in Region 5 and may be susceptible to methane gas
collection. Assuming these are residences with an average of four inhabitants, approximately 70,000 people may
be at risk of injury due to an explosion from methane gas. We assume the Region 5 data to be comparable to
situations found in the Basin.
Ecological Impact
The major routes of contaminant releases from municipal solid waste landfills are surface runoff of leachate
(e g., discharged from surface seeps in above-ground landfills) and leachate discbarge to ground water. In both
cases the ecological damages associated with these facilities would occur in surface waters or wetlands through
surface runoff or discbarge of contaminated ground water from hydraulically connected surficial aquifers.
There is very limited documentation available on environmental damages resulting from municipal solid waste
landfills. One documented damage case reported by EPA identified damages to wetlands and nearby lakes from
a 300 acre landfill leaching into these waters. Projected ecological damages from these releases included toxic
effects on freshwater and estuarine fish and macroinvertebrates.
prevention of ground-water and fresh water contamination is primarily dependent upon the thickness of the
unsaturated zone and the type of soil. To minimize the potentials for environmental contamination, a landfill
should have an unsaturated zone of more than 20 feet and clay substrate. Sinoe clay enhances surface runoff,
leaching would be minimal and would have a long distance to travel to reach the ground-water supply.
In response to the ecological impact on some areas, the EPA has imposed location restrictions on active seismic
zones, unstable areas where conditions prevent year round monitoring, wetlands, and 5 and 10 year flood plains.
Additionally, the EPA may close down active facilities in these areas.
Toxicity Assessment
Municipal solid waste landfills may release a large variety of contaminants to surface or ground waters.
Constituents released include BOD, COD, chlorides, ammonia, aluminum, arsenic, barium, vinyl chlorides, heavy
metals, pesticides, and volatile organics. Biochemical oxygen demand (BOD) and chemical oxygen demand
(COD) affect aquatic environments by depleting oxygen which may affect fish and other aquatic organisms.
Other constituents, including ammonia and aluminum, are toxic to marine organisms.
The effects of any release of leachate to surface waters will largely depend on the ultimate concentration of the
contaminants in the receiving body of water. In general, tbe volume of leachate flowing from an average landfill
is low, but the relative concentrations of constituents are high compared to other sources of surface water
contamination.
Exposure Assessment
-------
IV. 42
The number of surface waters or wetlands at potential risk of damage due to municipal solid waste sites in
Region 5 were estimated based on tbe sample of 31 operating facilities. Ninety three percent of tbe study set of
31 facilities, or an extrapolated total of 991 sites were determined to be located within one kilometer of surface
water such as lakes, ponds, rivers, creeks, or intermittent streams. Thirty one percent of the sample sites, or an
extrapolated total of 335 sites were determined to be located within one kilometer of wetlands. Adjusting for a,c
Basin, 620 sites were determined to be located within 1 kilometer of surface water and 211 sites were
determined to be within 1 km of a wetland. These extrapolations may be underestimated since the Basin bas a
considerably higher percentage of wetland and surface water areas than Region 5.
Ecological Risk Characterization
Ecological risks from municipal solid waste sites are, in general, poorly documented. Limited case study
information exists for relatively large landfills, but most landfills are substantially smaller than these sites. For
example the case cited above was for a 300 acre landfill; the average landfill in Region 5, and presumably the
Great Lakes Basin, is 20 acres.
To obtain a relative gauge of tbe ecological hazard from municipal solid waste sites, an estimate was developed
of the average daily and annual pollutant loadings through leachate discharge from an average landfill. Tbe
assumptions on which this estimate is based are:
¦Average landfill size is 20 acres;
¦Average net filtration for the Basin is 2.5 inches;
¦The landfill is unlined and does not have a leachate collection system;
¦All leachate generated flows to surface water through surface runoff or ground water without attenuation or
transformation in route;
¦Concentrations of constituents are mean values obtained from a large sample of leachates.
The average daily flow of leachate from the landfill, assuming the landfill is at moisture equilibrium
(outQowsinflow), is about 37,500 gallons per day.
In general the calculated pollutant loadings from leachate would not be anticipated to pose a serious water
quality problem if discharged to even a relatively small sized stream (e.g., 5,000,000 gal/day, or 8 tf/s) due to
dilution. For example, tbe discharge would only raise the BOD level in a 5 million gallon per day stream by
about 2 ppm, and the chloride level by about 0.6 ppm. However, discharges to relatively static water bodies
such as wetlands or ponds may significantly degrade quality due to the relatively concentrated nature of the
leachate, poor mixing, and lack of dilution.
An estimated 93 percent of the landfills in the Region 5 area are within one kilometer of surface water bodies.
Thus, about 2800 open and closed municipal landfills are estimated to be within one kilometer of surface waters
through surface runoff and ground-water discharges. Potentially more important, is an estimate of 31 percent of
open and closed landfills, or about 925, which appear to be located within one kilometer of a wetland. This
estimate is assumed to be lower than the situations found in die Basin, since the Basin contains a higher
percentage of surface water and wetlands than Region 5.
Reversibility
No information on the reversibility of ecological damages is available; however, experience shows us that some
-------
IV-43
practices are irreversible. Landfilling of the wetlands is one such practice. Once the wetlands are destroyed, we
have no way of restoring the land to its original state. This has become a problem in Dlinois and Wisconsin
where the wetland areas are disappearing. When the wetlands are landfilled, we also lose the habitats that the
wetland supported, and many species may cease to exist Another crisis in landfilling the wetlands is the
contamination of ground water. Since the wetlands are so close to other wetlands and the freshwater supply,
leachate and surface water runoff can become contaminated and endanger the habitats of other wetlands as well
as the fresh water creatures.
Other problems such as ground-water contamination are reversible, though the welfare impact is high. Ground-
water and sediment treatment is expensive and may not be implemented in time to save an ecosystem.
Furthermore, BOD loading to surface waters is reversible depending upon the waste assimilation capacity at the
receiving stream.
Welfare Impacts
Quantified welfare risks associated with the Municipal Solid Waste Sites problem area address damages resulting
from diminished property values adjacent to these sites and the costs associated with human health effects
attributable to these sites. Ecological welfare risks were not quantified due to lack of data.
Health Care Costs
To estimate health effects costs, the annual cancer cases estimated in the Human Health Risks section of this
problem area were multiplied by the direct medical cost and forgone earnings per cancer case:
(annual cancer casesXdirect costs plus forgone earnings)* {health costs
Estimated direct and indirect medical cancer costs are based on a range of case estimates. The lower bound
estimate is $80,000, while the upper bound estimate developed by the American Cancer Society31 is $137,000.
These estimates provide differing values for forgone earnings and medical costs. Both estimates are averages of
costs associated with all types of cancer.
Annually, the lower bound estimate is $120,000.00, and the upper bound limit is $205,500.00.
Environmental Costs
To estimate property value losses associated with municipal solid waste sites, the following equation was used:
(damage/homeX#MSWF)(homes/MSWF) * $ damages
Results of a growing body of empirical research suggest that values of residential properties increase as the
distance of those properties increases from active hazardous waste facilities and other solid waste disposal sites.
These studies use econometric analyses to estimate the difference in value of residences near a site and
"ICF Incorporated. 1988. 'OfBce of Solid Waste Emergency Response (OSWER) Comparifive Risk Project: Ground Water Valuation
Task Force Report." Prepared for U.S. EPA, Office of Underground Storage Tanks.
-------
IV- 44
comparable homes further from the site.
Results from these studies indicate that homeowners in the Boston area are willing to pay from $300 to $500 per
mile for a location more distant from a hazardous waste landfill, and that this amount was significantly higher
when considering a Superfund site.
Results of some other recent studies do not consistently indicate that there is an impact on nearby property
values. As a result a lower bound of property value damage due to municipal solid waste facilities was set at
zero for both active and dosed facilities. Upper bound damages were estimated to be $500 for residences
proximate to open facilities and $69 for those neti closed facilities.
To estimate the distribution of rural and urban municipal solid waste facilities, we assumed that the geographic
distribution of municipal solid waste sites is proportional to the population distribution of the states. As noted
before, this assumption may be in error since Wisconsin holds 2 percent of the nation's population and over 12
percent of its municipal solid waste landfills.
-------
IV - 45
rjndustrial solid waste facilities
Problem Area Description
The industrial solid waste facilities problem area includes all waste sites such as landfills, surface impoundments,
land application units and waste piles which are used for the disposal of industrial waste.
Magnitude of the Problem
Industrial solid waste facilities generally fall into four groups: landfills, surface impoundments, land application
units and waste piles. These four types of facilities handle a variety of waste such as: organic chemicals;
primary iron and steel byproducts; fertilizer and agricultural chemicals; plastics and resins manufacturing waste;
inorganic chemicals; stone, day, glass and concrete; pulp and paper waste; primary nonferrous metals; food and
kindred product waste; water treatment sludge; petroleum refining waste; rubber and rubber related byproducts;
transportation equipment waste; selected chemicals and allied products; textile manufacturing waste; and leather
and leather related waste.
In 1985 there were nationally:
¦2760 landfills managing 78.4 million metric tons of waste
¦15250 surface impoundments managing 6.7 billion metric tons of waste
¦4300 land application units managing 90.1 million metric tons of waste
¦5330 waste piles storing 69.9 million metric tons of waste.
Weighting this for the Great Lakes Basin population gives the following estimates:
¦331 landfills managing 9.4 million metric tons of waste.
¦1830 surface impoundments managing 804 million metric tons of waste.
¦516 land application units managing 10.8 million metric tons of waste.
¦640 waste piles managing 8.4 million metric tons of waste.
These figures may be underestimated due to the high concentration of industry around the southern lakes.
Human Health Risk Characterization
Cancer Risks
Toxicity Assessment
The Screening Analysis methodology for assessing human canocr risk is presented below. The Screening
analysis reviewed 87 chemicals found at industrial solid waste facilities and classified each constituent as
belonging to one or mote of the four facility types. The highest cancer risks were found associated with:
acrylonitrile, aniline, ethylene oxide, vinyl chloride, arsenic, benzene, chloroform and tetrachloroethylene.
Exposure Assessment
-------
IV- 46
Concentrations of the selected substances were calculated for the boundary of the facility taking into account
attenuation and dilution, and a constant was calculated.
To relate these results to the Great Lakes Basin, we depended upon the following assumptions:
¦The average landfill area is 20 acres. The average surface impoundment area is 1.6 acres. The average land
application unit area is 50 acres. The average waste pile area is 1 acre.
¦2S.S percent of the Great Lakes Basin industrial solid waste facilities pose an exposure risk.
¦1.6 people drink from wells per acre.
¦The exposure risk is confined to 400 meters from the site of the facility.
¦The exposure risk at 400 meters is the same as the exposure risk at the boundary of the facility.
Risk Characterization
From the data provided by the Screening analysis and our assumptions, we were able to compile excess cancer
estimates for each of the four types of industrial solid waste facilities. The total excess cancer in the Basin due
to industrial solid waste facilities is estimated to be 1.687 for a 70 year period. Exhibit R-l shows the
distribution of cancers over the various types of facilities is well as the acreage at risk, people at risk, cancer
risk ratios, and waste accumulated, all adjusted for the Great Lakes Basin.
This risk assessment is tentative since:
¦The nation wide data was adjusted for the Basin. The Basin may have more industrial waste sites than the data
would indicate since the Basin is known for the extent of heavy industry along the lakes' shores.
¦We assumed that the exposure risk at 400 meters is the same as the exposure risk at the boundary of the
facility. The exposure risk at 400 meters would probably be less than the exposure risk at the boundary of the
facility due to attenuation and dilution.
¦The average areas of the facilities were estimated. We do not know how Basin facilities would compare to the
national average.
Ecological Impacts
Severity
There is very limited documentation available on environmental damages resulting from industrial solid waste
facilities. As with municipal solid waste facilities, tbe main ecological impact of industrial solid waste facilities
results from leaching. Surface water runoff from leachate and leachate discharge to ground water from the
facilities can contaminate surface water and wetlands.
The severity of the impact is dependent upon tbe types of chemicals contained in the leachate, the amount of
leachate reaching the surface water or ground water, tbe concentrations of the constituents in the leachate, and
-------
IV- 47
the duration of the discharges. The component of the ecosystem adversely affected will depend on the media
Into which the chemicals are released. Aquatic life will be impacted by releases to surface waters, whereas
pi»nrc and avian species will be affected by releases to air and water.
Both acute and chronic effects may be observed. Ecological impacts such as fish kills, impairment of health and
ceproductive capabilities, and bioaccumulation in the food chain may result from dischaiges to the environment
The severity of their effects will depend on the resiliency of the ecological system impacted and the presence of
any fragile or endangered species in that system. Ecological impacts can be evaluated only if information on the
type and population of the exposed species, its stability, chemicals they are exposed to and duration of exposures
js available. Such data relevant to industrial solid waste facilities were not available. As a result, no attempt
was made to quantitatively evaluate risk assessments for ecological exposures
It is assumed that larger ecological impacts will result from surface impoundments handling mining waste
because of their size relative to other industrial solid waste facilities. At this point in time, such impacts cannot
be quantitatively evaluated.
Reversibility
No information on the reversibility of ecological damages is available, although some environmental damage
appears to be irreversible. Wetland destruction is one such irreversible impact Landfilling or contamination
destroys the wetland habitat and the species that live in the wetlands. Wetlands are prevalent in the Basin, and
much of the Basin's industry was built in landfilled wetlands, so wetland destruction poses an immediate
concent.
Ground and surface water contamination is another ecological impact Contaminated water can be treated and
this condition is reversible, although die welfare impacts are costly. BOD loading to surface waters can also
impact the ecosystem. BOD loading is reversible depending upon the waste assimilation capacity of die
receiving stream.
Welfare Impacts
Quantitative estimates of welfare risks due to Industrial Solid Waste Facilities in Region 5 are severely
constrained by the absence of data concerning the number and distribution of these sites. Quantitative risk
estimates have been developed only for estimated annual cancer cases. Limiting the welfare analysis to this one
quantitative estimate results in a downward bias of die "true* damage magnitude.
Health Effects
To estimate health effects costs, the annual cancer cases estimated were multiplied by the direct medical cost and
foregone earnings per cancer case:
(annual cancer casesXdirect costs and foregone earnings)*$health costs
Estimated direct and indirect medical cancer costs are based on a range of cost per case estimates. The Iowa
bound estimate, based on Hartunian, et al, is $80,000, while the upper bound estimate developed by the
-------
IV-
American Cancer Society is12 $137,000. these estimates provide differing values for foregone earnings and
medical costs. Both estimates are average costs associated with all types of cancer.
Over a 70 year period, the lower bound estimate is: (1.687X580,000)= $134,960.00 for the Basin
Over a 70 year period, the upper bound estimate is: (1.687X5137,000)= 5231,119.00 for the Basin
"ICF Incorporated. 1988. "Office of Solid Waste Emergency Reipoiise (OSWER) Compariitve Risk Project: Ground Water Valuation
Tuk Force Report." Prepared for U.S, EPA, Office of Underground Storage Tub.
-------
Facility Facilities in
Basin
LandGlls 331
Surface 1,830
Impoundments
Land Application 516
Units
Waste Piles 640
Waste in Million
Metric tons in
Basin
9.4
804
10.08
8.4
IV.49
Exhibit R-l
Average Area of Cancer Risk Acreage at Risk Population at Excess
Facility in Acres Ratio in Basin Risk in Basin Cancers
in Basin
20 .000053 2,169.2 3,470.7 .180
1.6 .000085 3,406.5 5,450.4 .463
50 .000012 5,394.8 8,631.7 1.036
1
.000053
979.2
1,566.7
-------
V - 1
V. SECONDARY PROBLEM AREAS
S. AGGREGATED DRINKING WATER
Problem Area Description
Recognition that a reliable and safe water supply is necessary for a healthy, thriving comm unity was not a special insight
for the United States or any other country. Instead, disease and epidemics, tbe results of unsafe drinking water supplies,
prompted communities to provide clean drinking water. Most States have operated drinking water programs since the
early 1900s that have improved public health by providing safe and reliable drinking water in major communities.
Whenever possible, consumers are encouraged to obtain water from a public water system in order to enjoy the
advantages of supervised water supply under the control of a responsible public agency. This philosophy, along with
advancem ents in water supply engineering and treatment, have significantly improved the quality, safety, and reliability
of drinking water over the past century.
Magnitude of the Problem
The federal Safe Drinking Water Act (SDWA) of 1974, which was amended in 1987, recognized the importance of
drinking water quality. The original Act encouraged all States to adopt the minimum federal drinking water standards
that EPA considered at that time to provide minimal protection of public health. The amendments of 1987 expanded
the scope of the original Act to account for the risks identified by an improved understanding of hydiogeology, water
chemistry, environmental degradation, and health effects. While the risk of a drinking water-related epidemic may be
regarded as unlikely in most large public water systems in the Great Lakes Basin, some threat continues to exist in
smaller communities. Rural and private water supply systems continue to present a significant risk for several reasons,
including the following.
Many small water supplies are not regulated under an enforceable well construction code, resulting in the
construction of systems that fail to provide even minimal protection of the water supply. Unenforceable well
construction codes may foster
• Substandard well construction, which may allow contaminated surface water to enter the well;
• Improper well abandonment, which may allow unfiltered surface water to enter the aquifer;
• Siting of septic tanks, drainage fields, and dry wells within the zone of influence of water supply
wells;
• Siting of agricultural chemical mixing sites at or near well heads;
• Siting of small production wells that tap the upper portion of shallow aquifers where nitrates and other
surface contaminants are concentrated; or
• Makeshift repairs using inadequate materials and procedures that can result in the infiltration of
contaminants and pressure loss, allowing holding tanks to flow back into distribution lines and
threaten consumers when water pressure is restored.
Inadequate or infrequent flushing of distribution lines along with a lack of bacterial disinfection (e.g., residual
chlorine) are threats to public health. Also, inadequate line flushing following routine repairs to distribution
lines can have similar effects.
-------
V -2
The majority of small water supplies arc without benefit of a competent, certified water supply operator. All
miblic and drinking water facilities are subject to mechanical failure. Larger water utilities constantly aspect
m components and are prepared to maintain those that fail. Smaller utilities, having at least minimal
,afeKua£s like proper well construction, may lack incentive to provide surveillance, and most of the smallest
svstems lack the finances necessary to fund surveillance. Hiese systems usually discover problems that present
sLous health risks only through regulatory procedures that most water supply experts consider inadequate.
Ac a result all States in the Basin support operator recruiting, training, and certification and outreach to those
communities unable to procure technical expertise. All States, including Indiana, which is only now adopting
a federally recognized drinking water program, continue to devote a substantial portion of their resources to
p^aintaining a field staff and providing outreach.
As stated above, States identify problems associated with public drinking water supplies through compliance
monitorine programs. The fact that States are addressing acute incidents of noncompliance should not suggest
thauhose violations considered to be relatively insignificant should be ignored. Intermittent mooting
violations may present a small risk that contamination goes undetected in a single water system but the sheer
number of violations occurring in unrelated systems is cause for concern Tie Region 5 office issues, on
average more than 250 quarterly and 1,000 annual violation letters to small public water supplies in Indiana
The majority of these violations arise fiom failure to monitor water supplies, which can be further
ZLedto igiraL of SD WA regulations. Enforcement can alleviate some ignorance but actions must be
22y approbate, and consistent. While the lack of monitoring data does not in itself constitute a public
bealth'problerru it is necessary to insure that public health is not compromised.
Small water systems often avoid the issue of non-compliance by stating that their facilities serve only a minor
portion of the population. TOs statement is largely true but most non-community water supphes sc™
transient populations (travelers), which may potentially affect a significant portion of an area s population.
Other non-community water supplies serve schools and factories, locations where a person may consume most
of his drinking water during a 24-hour period. When relative risks within the angle arena of dnntang water
are weighed, EPA finds that it is more important to address violations affecting the torgest populations and
direct available resources toward this goal; however, this focus should not downplay the significance of
non-compliance by small utilities. Given the potential for health-related problems at these systems, the overall
risk presented by non-compliance is potentially large.
The above discussions identify what EPA believes are the most significant risks to public health by drinking water.
Additional risks arise from inadequate response to these situations. Perhaps the most visible threat is the increasingly
common use of home water treatment devices. Many consumers are aware of the findings of environmental groups, and
the lack of timely and effective State response to recognized situations serves to fuel public concern. The home water
treatment devices and bottled water industries are growing at an unprecedented rate that far exceeds the development
of appropriate health regulations. Home treatment devices have an undeniable place in public health protection, but they
are not without risk. In almost every case, the risk presented by reliance upon a home treatment system far exceeds that
of a public water supply violation. Lack of consumer education encourages inappropriate use of these devices (as well
as a lack of confidence with the local public water supply), often leading to a false sense of security. The primary issue
is the availability of sufficient resources to operate a comprehensive public water supply program.
Another source of risk to drinking water supplies is the increase in sodium concentration in the Great Lakes. In the Lake
Michigan basin, for example, sodium from anthropogenic sources, including road salt (whose winter use began in the
late 1940s), has been increasing and now far exceeds sodium from natural weathering sources. Sodium concentrations
-------
V - 3
for Lake Michigan could eventually approach those of drinking water standards under present growth trends.1
SOURCES, CONTAMINANTS, EXPOSURE PATHWAYS, AND EFFECTS
Selection of Representative Contaminants
EPA has promulgated primary regulations for 30 drinking water contain inaats that are expected to have adverse impacts
on human health and secondary regulations for contaminants that affect the aesthetics of drinking water. EPA has also
established monitoring requirements for other contaminants. Under the 1986 amendments to the SDWA, Congress
stipulated that EPA set monitoring and MCL requirements for 83 additional contaminants by June 1989.
Since it is not possible to quantitatively assess the risks for all drinking water contaminants or even a subset of those that
are regulated, EPA planned to select representative drinking water contaminants by reviewing quantitative information
from its Federal Reporting Data System (FRDS) national data base for public drinking water supplies within the Great
Lakes Basin. EPA retrieved the following general information for the Great Lakes Basin from FRDS:
number of public water systems by county for each of the eight states within the Great Lakes Basin and the
populations they serve and their source of water (e.g., ground water or surface water);
number of public water systems by county with at least one MCL violation for each of the past five years;
distribution of these violations among ground water-based and surface water-based combined systems for each
of the 213 counties within the Great Lakes Basin;2
specific contaminants associated with the violations, as well as the monitored concentrations; and
sizes of the populations affected by each contaminant violation.
Ideally, contaminants of concern for this analysis would depend upon those constituents for which MCL violations were
noted in the FRDS retrieval; however, due to time constraints, EPA was unable to fully examine the data retrieved from
FRDS. Instead, EPA reviewed the Region 5 comparative risk study and, where appropriate, scaled risk estimates down
to reflect populations potentially at risk within the Basin. Where possible, EPA supplemented this analysis with data
retrieved from FRDS for the Basin. (We provide as an addendum to this chapter a list of the numbers of MCL violations
from 1985 through 1989 for each county in the Basin as well as the population impacted by violations in 1989.) A
review of the Region 5 com parative risk study indicated the following potential contaminants of concern within the Great
Lakes Basin.
Total Coliform/Microbiological Contaminants
There are many historical records with reported cases of disease transmission from microbial agents in drinking water
1 Gaerra, Bruna, Paul Horvatin, Vacy* Saalys, and David Do Ian, 'Lake Michigan Sodium Mats Balance and Projection for
Fit tare In-Lake Sodium Concentration*,* U.S. EPA, Aagast, 1985.
1 Some pablk water inppljr ijiUmi utilize both ground water and sarface water, however, FRDS does not acknowledge these
combined systems but designates them as surface water-based systems. FRDS also does not specify the percent contributions of
groand water or sarface water to a combined system; therefore, a system that is largely ground water-baaed would be designated a
sarface water-based system in FRDS. Only systems that re If solely en groand water are designated as ground water-based.
-------
V - 4
supplies. Pathogenic and nonpathogenic bacteria, viruses, protozoa, and cysts can all be transmitted via ingestion of
contaminated ground water. Cases of waterbome disease in the Basin have involved such diverse contaminants as
Giardia lamblia. campbylobacter, and a multitude of other pathogenic coliforms and viruses. Since there are literally
hundreds of possible microbial agents, agent-specific monitoring of water supplies is not practiced. Instead, the intiir-atry
organisms of the coliform bacteria group are used to evaluate the bacteriological quality of water. Turbidity is also an
important parameter since it is believed to shield bacteria during the disinfection process, thus making disinfection less
effective; however, available information seems to indicate that there may be no direct relationship between turbidity
levels and coliform in finished drinking water and the incidence of waterbome disease.3 In addition to violation data
for total coliform, regulatory agencies need data to relate disease incidence to drinking water. This additional data would
consist of the following:
Disease incidence reports by State;
Surveys of water systems that experienced a contamination problem;
Special investigations by State, local, or federal health agencies that involved data collected from drinking
water supplies.
Nitrate as Nitrogen
Nitrate present in drinking water in excess of 45 mg/L (10 mg/L as nitrogen) is associated with the incidence of
methemoglobinemia, which can cause "blue baby" syndrome. This problem is confined mainly to infants less than 6
months of age and in agricultural areas. According to the Region 5 comparative risk study, there have been no cases
of methemoglobinemia reported in the Region in recent years; because the Great Lakes Basin is largely within Region
5, it is assumed that no cases of methemoglobinemia have been reported in the Great Lakes Basin in recent years.
Lead
EPAset the MCL for lead at 0.05 mg/L in 1976. Since then, toxicological information indicates that lead concentrations
in drinking water that are far lower than the MCL are capable of producing adverse health effects, particularly in younger
children, infants, and fetuses. Lead causes damage to the nervous system, blood-forming processes (hemopoietic), the
gastro-intestinal system, and the kidneys. More recent studies show that lead also causes cognitive damage retards
growth, and can raise blood pressure in adult males, even at low exposure levels. Health effects range from relatively
subtle biochemical changes at low doses to severe retardation or death at higher levels.
The MCL for lead is currently being lowered to reflect the increasing knowledge of the adverse effects of lead in
drinking water to human health and its widespread occurrence in water supply distribution systems. According to the
Region 5 comparative risk study, there have been few violations of the lead MCL in Region 5, primarily because the
presence of lead in sources of drinking water is relatively rare. Most lead in drinking water results from corrosion of
distribution systems. Since most distribution systems are flushed before sampling, elevated lead concentrations are often
not measured through compliance monitoring and go unnoticed in the FRDS data base.
' Turbidity k an Important indicator of the effectiveness of iirtice water treatment and is currently regulated for larface-water
supplies. Becauae its significance thro ag boat the Basin is not known, It win not be evaluated in this report.
-------
V - 5
Trihalom ethanes
Trihalomethanes (THM) are volatile organic compounds that are formed when chlorine used in the disinfection process
comes in contact with organic bumic and fulvic acids. While four different by-products are formed (chloroform,
bromoform, dichlorobromomethane, and dibromochloromethane), chloroform is the species found in the highest
concentration. THMs are regulated collectively as total THMs. The MCL is set at 1.0 mg/L and applies for the total
concentration of any combination of THMs present. Projections regarding THMs could be based on the mean
chloroform exposure levels for chlorinated surface-water supplies. The population exposed to THMs would be those
surface-water systems serving populations of 10,000 or more customers (as per the regulations these systems are
required to chlorinate) and any other surface supplies that chlorinate.
The Office of Drinking Water (ODW) is preparing a mandatory disinfection treatment rule for ground water. The
anticipated proposal date is January 1991, with promulgation approximately one year later. ODW is also preparing a
rule that will limit the levels of disinfectants and their by-products in finished drinking water. EPA expects that the rule
will be proposed by September 1991 and promulgated in 1992.
Radionuclides
EPA regulated the radionuclides radium-226 and radium-228 under a combined MCL of S picocuries per liter (pCi/L)
under the SDWA of 1976. This MCL used gross alpha as a screen (IS pCi/L) for these regulated alpha emitters. ODW
is developing new regulations and is expected to publish a Notice of Proposed Rulemaking in January 1991. In this
Notice, EPA will propose Maximum Contaminant Level Goals (MCLG); MCLs; Best Available Technologies (BAT)
for monitoring requirements for radon-222, radium-226, radium-228, gross alpha, natural uranium, and beta particle and
photon emitters. Because all radionuclides under consideration are known human carcinogens (i.e., classified as Group
A), MCLGs for these radionuclides will be proposed as zero. EPA is also considering proposing a separate MCL for
each radium isotope, possibly at 5 pCi/L.
Data on the average occurrence of radium in public water supplies in the Great Lakes Basin is not available in FRDS
since only MCL violations and associated concentrations are reported. EPA's Office of Radiation Programs conducted
a nationwide survey in 1980-81 of2,500 public, ground water-based drinking water supplies in 27 States that represent
45% of all drinking water consumed nationally.4 Hie survey estimated that the population-weighted average value for
radium-226 and radium-228ranged betweenO.3 pCi/L and 0.8 pCi/Land between 0.4 pCi/L and 1.0 pCi/L, respectively.
Radon
The Region 5 comparative risk study estimates that radon levels in indoor air are responsible for between 5,000 and
20,000cases of lung cancer annually in the United States. Approximately one to seven percent of these cases result from
domestic uses of drinking water, such as showering, bathing, cooking, and washing clothes and dishes. The study further
estimates that in an average lifetime of 70 years, between 2,000 and 40,000 deaths by lung cancer will occur in the
United States due to radon levels in public water supplies that are primarily served by ground water. The risk assessment
estimates that the nationwide population-weighted average concentration for radon in public drinking water supplies
served by ground water is about 420 pCi/L For systems served by both ground and surface water, the concentration
is estimated to range between 50 pCi/L and 300 pCi/L.
ODW plans to propose an MCL for radon in the Notice of Proposed Rulemaking anticipated in January 1991.
4 Federal RtglsUr. Volume 51, Number 189, September 30, 1986.
-------
V - 6
Trichloroethvlene
EPA regulated trichloroetbylene as a primary drinking water contaminant at 0.005 mg/L in July 1987. It is classified
as a Class B2 probable human carcinogen based on the weight of toxicological evidence. Acute oral exposures (15-2S
ml) to TCE in humans has resulted in vomiting, abdominal pain, and transient unconsciousness, while longer-term
occupational exposures suggest damage to the liver. The major source of TCE released to the environment is from its
use as a metal degreaser. Since TCE is not spent during its use, the majority of all TCE produced is released to the
environment TCE released to the air is degraded within a few days. TCE released to surface water volatilizes within
a few days or weeks and degrades. However, TCE that released to land migrates readily to ground water where it
remains for months to years. TCE does not easily degrade in ground water, but under certain conditions, it may degrade
to dichloroethylene and vinyl chloride. EPA selected TCE as a constituent of concern because it is a common
contaminant in ground and surface water, with higher levels found in ground water. National surveys of drinking water
supplies have shown that three percent of all public water systems using ground water contain TCE at levels of 0 J g/L
or higher. Approximately 0.4% have levels greater than 100 g/L. In Region 5, about 3.5% of the wells contained some
measurable level of TCE, with 0.5% of all wells exceeding a 10"5 lifetime cancer risk. Unlike other chlorinated
compounds, TCE does not bioaccumulate in animals or food chains.
Tctrachlorocthvlcne
Tetrachloroethylcne (PCE) is also classified as a VOC. After slating PCE for promulgation along with TCE in July
1987, EPA withdrew PCE from the group due to available bioassay data. This data created a controversy surrounding
its weight of evidence classification as a B2 or C carcinogen. Currently, PCE is classified as a B2 carcinogen and is
scheduled for promulgation of a MCL and MCLG in December 1990. The MCL and MCLG will be 0.005 mg/L and
zero, respectively. PCE also has many industrial uses and, like TCE, is not spent during use but is released directly back
into the atmosphere. PCE is usually discharged directly to land and surface water and degrades slowly. PCE released
to the atmosphere degrades within days or weeks. It is very mobile in soil and travels easily to ground water where it
remains for months or years. Under certain conditions, PCE degrades to TCE and then to dichloroethylene and vinyl
chloride. National surveys of drinking water supplies have shown that 3% of all public water systems using ground
water contain PCE at levels of 0.5 g/L or higher. About 0.7% have PCE levels above 5 g/L In Region 5, about 3%
of all public water systems supplied by ground water have measurable levels of PCE, with one percent of all wells
exceeding a 10"5 lifetime cancer risk.
1.1.1 -Trichloroethane
EPA regulated trichloroethane (TCA) as a primary drinking water contaminant at 02 mg/L in July 1987. It has been
placed in the category of class D carcinogens that have not been evaluated as to their human carcinogenic potential due
to insufficient data. The major source of TCA released to the environment is from its use as a metal degreaser. As with
the other two previously discussed VOCs, TCA is not consumed during metal degreasing but is released to the
environment. TCA released to air degrades slowly with an estimated half life of one to eight days. TCA release to
surface water volatilizes in a few days or weeks. TCA released to land does not sorb onto soil and travels quickly to
ground water. It slowly hydrolyses in ground water with an estimated half life greater thai 6 months. As with TCE,
TCA does not bioaccumulate in animals or food chains. TCA is a good representative chemical because it occurs widely
in the environment. It is a common contaminant in ground water and surface water, with higher levels measured in
ground water. National drinking water surveys have found that 3% of all public water systems using ground water
contain TCA at levels of 05 g/L or higher. Approximately 0.1% have TCA levels above 100 g/L. In Region 5, about
two percent of the public water wells showed measurable levels of TCA.
-------
V - 7
Alachlor
The MCL and MCLG for alachlor are scheduled for promulgation in December, 1990. The proposed MCL is 0.002
mg/L, and the MCLG is zero. Alachlor has been classified as a B2 carcinogen due to the weight-of-evidence of human
carcinogen properties. Alachlor had one of the largest production volumes of any pesticide. It is applied to the soil
either before or just after the crop has emerged and is rapidly metabolized by crops after application. It is widely used
for corn and soybean crops. In the soil, alachlor is degraded by bacteria under both anaerobic and aerobic conditions.
Alachlor is not photodegradable and does not hydrolyze under environmental conditions. Alachlor is moderately mobile
in sandy and silly soil and has been shown to migrate to ground water. On a national basis, alachlor has been measured
in both surface and ground waters. Federal and State surveys of surface water have reported occurrences of alachlor
at levels of one part per billion. Alachlor is used extensively in Region 5 and the Great Lakes Basin due to high com
and soybean production. In Region 5 (and most likely the Great Lakes Basin), available data indicates that alachlor has
the potential to contaminate both ground and surface water widely.
Atrazine
The MCL and MCLG for atrazine are scheduled for promulgation in December 1990. The proposed MCL and MCLG
are both 0.003 mg/L. EPA has classified atrazine as a C carcinogen (i.e., a possible human carcinogen) because of a
lack of evidence of potential carcinogenicity to humans and incomplete evidence of potential carcinogenicity to animals.
According to the STORET1988 national data base, atrazine has been detected in 4,123 of 10,942surface-water samples
and in 343 of3,208 ground-water samples. These samples were collected at 1,659 surface-water locations and at 2,510
ground-water locations, respectively, across the nation. Atrazine was present at the 85th percentile of all non-zero
samples at concentrations of 2.3 g/L in surface water and 1.9 g/L in ground water. s Atrazine is moderately to highly
mobile in soils that range in texture from clay to gravelly sand. Atrazine degrades in soils by photolysis and microbial
processes. It is widely used in the Basin for corn.
Other Contaminants
The Region 5 comparative risk study does not include several inorganic contaminants, such as fluoride, in its analysis
because FRDS data and national monitoring data show that MCLs for these contaminants are violated infrequently and
are remedied quickly when they are violated.
EPA is scheduled to regulate several synthetic organic contaminants other than pesticides, including PCBs and
nonvolatile organic solvents, but these contaminants have not been included in this analysis. National monitoring data
indicate that these contaminants do not occur frequently in public water supplies and would, therefore, present a small
risk in comparison to other contaminants. Numerous contaminants that are currently regulated, such as inorganics and
pesticides, are also not included. The Region 5 comparative risk study indicates that FRDS reports very few MCL
violations for most regulated inorganics, pesticides, and radionuclides. Therefore, EPA has chosen not to include these
parameters in this analysis, and these assumptions are applied to the Great Lakes Basin as well
Exposure Pathways
Three possible routes of exposure from drinking water as a source of contaminants exist: ingestion, inhalation, and
* This Inform*lion comes from the Health Advisor; for Atrazine, p. 44. The section warns thai tk Individual data "-.have not
been confirmed as to their validity. STORET data are often not valid when individual numbers are aied oat of the contest of the
entire sampling regbne.Therefore, this Information can only be aaed to form an impression of the Intensity and location of sampling
for a partknhr c Ik mica L"
-------
V- 8
dermal contact. The major route of exposure from most drinking water contaminants is through ingestion; however,
depending on the physical properties of a contaminant (e.g., volatility, octanol-water partitioning), inhalation of volatile
organic compounds (VOCs) such as trichloroethylene can occur as the compound is released to the air from use of hot
water in showering, bathing, or dishwashing. The Region 5 comparative risk study indicates that some studies have
shown that exposure to VOCs via inhalation of indoor air can be as significant as exposure via ingestion.
Inhalation is also a possible exposure pathway for radon that has been liberated from drinking water. Radon is released
from drinking water in the same manner as from VOCs. Radon has the potential to cause lung cancer through inhalation
and radiotoxicity through ingestion. EPA's Office of Drinking Water is developing maximum contaminant levels
(MCLs) and monitoring requirements for radon, and it will be a regulated parameter under the Public Water Supply
program; however, EPA estimates that the regulatory levels being considered for radon in drinking water (200 - 500
Pci/L) would contribute less than one percent to the total amount of indoor radon. EPA did not have adequate
information to assess risks from ingestion and inhaling radon in the Great Lakes Basin, but nationwide risk data has been
scaled down to estimate risk within the Basin.
Human Health Risk Characterization
In analyzing the public health risk posed by drinking water, we examined cancer and noncancer risks for both individual
and population risks using Region 5's exposure assumptions described in the comparative risk study. We define
population risk as the risk posed to all persons in the Great Lakes Basin exposed to each of the above contaminants at
the assumed exposure point concentrations. Individual risk is the risk posed to an individual potentially exposed to the
chemicals at theseaverage concentrations. In both the individual and population risk analyses, we considered reasonable
average exposures instead of worst case or most exposed individual scenarios. In addition, we examined typical
exposures for both chronic and sub-chronic adverse health effects.
The methodology outlined in the Region 5 comparative risk study essentially encompasses the following five major
steps:
(1) determine the environmental concentrations for selected contaminants in the Basin water supply;
(2) evaluate the hazard or toxicity for the selected chemicals;
(3) determine the population exposed assessment to the identified concentrations;
(4) characterize the chemical's potency or dose-response relationship; and
(5) quantify the risk associated with exposure levels.
In addition, uncertainties and results are discussed. The individual purposes, methods, assumptions, and limitations are
discussed under each section of the risk analysis.
Toxicity Characterization
In this section, we present both cancer and noncancer toxicity information for the contaminants of concern described
in the previous section. Of the contaminants examined, radionuclides are human carcinogens (Class A). Chloroform
is the predominant compound among the trihalomethanes (THMs) formed by the chlorination of drinking water.
-------
V - 9
Chloroform is regarded as a probable human carcinogen (Class B2). The Cancer Potency Factor (CPF) or unit slope
factor that will be used for this analysis was assumed to be the same as that for chloroform (6.1 E-3 ppm/day-1). Other
contaminants that are B2 carcinogens are lead, trichloroethene, tetrachloroethene, and alachlor. Atrazine and alachlor
were evaluated for both carcinogenic and noncarcenogic risks. Coliform and nitrate are noncarcinogens and were
evaluated with respect to noncancer endpoints. Trichloroethane has not yet been evaluated with respect to evidence of
human carcinogenicity (Class D) and was assessed only for noncarcinogenic risks.
For the carcinogens, the purpose of this step is to determine the relationship between the dose and the probability of
developing cancer. This relationship predicts the level of risk associated with a certain exposure. The information
presented Exhibit S-l includes EPA's weight-of-evidence carcinogenicity classification, the cancer potency factors, the
Maximum Contaminant Level (MCL) for drinking water, and the acute and chronic toxicological endpoints. All values
presented in Exhibit S-l are taken from the Region 5 comparative risk study.
For noncarcinogens, the purpose of this step was to determine the relationship between doses of a contaminant and the
probability of developing an adverse health effect. Unlike the dose-response relationship for carcinogens, these
relationships are thought to involve a threshold below which an adverse health effect will not likely occur. This is
because the human body is able to detoxify and repair systemic damages up to a certain point. Toxicity information for
noncarcinogens is also listed in Exhibit S-l and includes the oral reference doses (RfDo) for chronic exposures, the
MCLs, Health Advisories (exposure doses considered by the Agency to be acceptable for acute and less than lifetime
exposures), and the acute and chronic toxicological endpoints.
-------
V- 10
EXHIBIT S-l
Carclaogealc aid Toxlcological Pinmclcn of Driikiig Water Coatamlaaats
Coatamlaaat
Collform aad
microbiological
ageats
Lead
Nitrate
Total
Trifcalomethaaes
(expressed as
chlorofonn)
Trichloroetheae
Tetrachloroethene
Trichloroethaae
Carciaogea
Class
NA
B2 (teatative
classification)
NA
B2
B2
B2
Oral CPF
(slope factor)
(ppm/d)'1
NA
1.41 lO"1
NA
6.1 x 10'
1.1 x 10'
5.1 x 101
NA
Iahal. CPF
(slope factor)
(ppm/d)1
NA
Not Determlaed
Oral RID
(ppm/d)
NA
Not
Determlaed
NA
8.1 x 101
1.7 x 102
3.3 x 10'
NA
1.0
0.010
0.007
0.010
0.090
MCL
(mg/L)
varies
0.005 soiree,
treatment
technology at tap
10.0
0.100
0.005
HA, Longer
Term (7-yr)
(mg/L)
NA
NA
NA
NA
NA
0.005 (proposed) 5.0
HA, Level 1-day
& 10-day
(mg/L)
NA
NA
10.0
NA
0.200
100.0
NA
2.00 & 2.00
100 & 40
Toxlcologlcal
Eadpoints
(Acate/Ckroaic)
GI liritatioa
neurological:
snbtle biochem.
changes, impaired
mental
performance,
circilatory
system effects
methemoglobine
mia in infants less
than 6
months/death
hepa to toxicity/
hepi to toxicity/
cancer
hepa to toxicity,
nephrotoxicity,
CNS effects/
cancer
CNS effects,
hepatotoxicity,
-------
V- 11
Atrazlne
Alack lor
B2
Not Determined Not Determiied 0.005
Not Determiaed Not Determiaed 0.010
0.003 (proposed) 0.200
0.002 (proposed) NA
0.100 A 0.100
0.100 & 0.100
Radiant 226-228
3.6 x ia'
Not Determined NA
5 pCi/L
NA
NA
kidaey, liver
damage/possible
reprod., devei.,
matag. effects
kepatotoxlcity,
ocalar
effects/cancer
bone, mastoid
cancer
Rjdoa 222
Not Determiaed 1.8 * 104
NA
Not Determiaed NA
NA
ling cancer
MCL varies based oa analytical method, sample volame, aad number of samples collected per moatk. Two types of MCLs exist for coliform: (1) for a monthly average, and (2) for a single sample. After
December 12,1990, no more tfcaa Ave percent of simples may be positive.
-------
V- 12
Exposure Characterization
As described above, our approach was to scale the results of the Region 5's exposure assessment to a level that
approximates risks posed to the Great Lakes Basin population. The purpose of the exposure assessment step was to
estimate the concentrations in drinking water at which exposure occurs, identify and estimate the sizes of the exposed
populations, and identify the pathways of exposure and delivered doses. The assessments for carcinogens and
noncarcinogens for both individuals and for populations were based on realistic maximum and average exposures
utilizing data from the Region 5 comparative risk study. Exposures were assumed to be constant over a 70-year lifetime.
The risk assessment included the derivation of average risks by weighting to the exposed populations. We made the
following additional assumptions in estimating the risks posed by both carcinogens and noncarcinogens:
The estimates embrace Region 5's exposure assumptions for its comparative risk study;6
All Basin population values used in the analysis were extrapolated from Region 5 values by the ratio
of overall Basin population to overall Regional population; and
The concentrations used in the analysis were taken directly from the Region 5 comparative risk study.
Exhibit S-2 presents the populations in the Basin served by public and private drinking water supplies Human
population numbers for the Great Lakes Basin used in the assessment are based on the 1988 estimate from the Bureau
of Census. Rather than using population estimates retrieved from FRDS, which were found to reflect populations
impacted by MCL violations only and not total populations regardless of quality of water, estimates of populations
served by each of the source categories were used as presented in the USGS report, National Water Summary 1987 ~
Hvdroloric Events Water Sunnlv and Use. Water-Supply Paper 2350, which describes 1985 water use. This report
presented, for each state, percentages of the population served by public water supply systems and by private wells.
These percentages are assumed to apply to the portion of the Basin in each state and scaled the total population in each
State-portion of the Basin by these percentages. It is possible that this approach underestimates the population served
by public water supply systems because populations within the Basin are more likely to receive drinking water from
treated withdrawals from one of the Great Lakes. The USGS report did not provide percentages of populations served
by surface- and ground-water based systems for the Basin states (with the exception of New York), so percentages for
the quantities of surface water and ground water used by PWS systems in the states were applied to calculate the
populations served by each source. This approach may underestimate the population served by surface-water based
systems, because systems in the Basin are likely to use Great Lakes water.
Exhibit S-3 presents a summary of the exposure assessments used in this analysis. Exposure assumptions are consistent
with those made in the Region 5 comparative risk study.
Risk Characterization
Exhibit S-4 presents the approach and results of the carcinogenic risk characterization. We combined the estimated
drinking water concentrations and exposure assumptions (encompassed in the Oral Chronic Intake Factors) to produce
a range of chronic daily intakes (dosages). We then multiplied the dose by the cancer slope factor to calculate the
individual chemical-specific lifetime cancer risks presented by each chemical. The individual risk was then multiplied
by the estimated exposed population to obtain an estimate of the average number of cancer cases expected over a 70-year
lifetime. We also calculated chemical-specific average population-weighted cancer risks. We estimate the following
' For the noncarcinogens, we ued an averaging time equal to the exposure deration. Assumptions used in the assessment are
questionable for lead, because its carcinogenicity has not been determined.
-------
total numbers of cancer cases presented by all of the contaminants of concern (with the exception of lead because its
carcinogenicity has not been determined):
Total Cancer Cases
70-year lifetime Annual Cancer Cases
Minimum 45 0.64
Maximum 114 1.63
Average 80 1.14
-------
V - 14
EXHIBIT S-2
Population on Public and Private Drinking Water Supplies in the Great Lakes Basin
State
Total Population
in Basin1
Population Served
by Public Water
Supply Systems2
Population
Served by
Surface Water
Systems3
Population
Served by
Ground-Water
Systems4
Population
Served by Priv
Wells5
Illinois
5,779,600
4,912,700
2,996,700
1,916,000
866,900
Indiana
1,699,300
1,138,500
603,400
535,100
560,800
Michigan
9,239,800
7,576,600
6,212,800
1,363,800
1,663,200
Minnesota
320,700
205,200
90,300
114,900
115,500
New York
4,377,800
3,896,200
2,883,200
1,013,000
481,600
Ohio
5,227,500
4338,800
3,123,900
1,214,900
888,700
Pennsylvania
381,000
262,900
220,800
42,100
118,100
Wisconsin
3,206,900
2,084,500
1,083,900
1,000,600
1,122,400
Great Lakes
Basin
30,232,600
24,415,400
17,215,000
7,200,400
5,817,200
1 Total population ia all counties with any land area la the Great Lakes Basin. 1988 estimate from "U.S. Great Lakes: Drainage Basin Coanty Snmmary," U.S. EPA, Great Lakes National Program Office,
July 1990.
1 Estimated Basin population served by public water sapply systems, calculated by scaling populations In Basin by factors obtained from National Water Summary 1987 — Hvdrologlc Events and Water
Supply and Use. US. Geological Survey Water-Supply Paper 2350.
' Estimated Basin population that is served by surface-water based public water supply systems, calculated by scaling populations served by public systems by factors obtained from USGS report.
4 Estimated Basin population that is served by ground-water based public water supply systems, calculated by scaling populations served by public systems by factors obtained from USGS report.
' Estimated Basio population that is served by self-snpplled sonrces, the majority of which are assumed to be private wells, calculated by scaling total populations in Basin by factors obtained from USGS
report.
-------
V- 15
EXHIBIT S-3
Exposure Assessment Summary
Contaminant
Source
Type
Population Characteristics
Potential
Exposed
Population'
Avenge
Exposure
Concentration
(mg/L)J
Exposure
Route
Total Colifonn/
Microbiological
Agents
Ground
and
Surface
Users of PWS in violation' of MCL and users of private
wells
7,100,000
NA
Ingestion
Nitrate as Nitrogen
Ground
and
Surface
Fetuses and infants less than 6 months using PWS is
violation of MCL and private wells in areas likely to be
affected by agriculture
167,000*
15.75
Ingestion
Total
Trihalomethanes as
Chloroform
Surface
Users of PWS serving >10,000 with MCL violation
Usen of PWS serving <10,000 currently not regulated
15,600*
1,540,000*
0.101
0.060
Ingestion
Inhalation
Lead
Ground
and
Surface
Fetuses, infants, and children using PWS with MCL
violations
PWS usen with positive detections below MCL
PWS usen and private well users with no or minimal
detections
1,950*
370,000*
30,232,600
0.05
0.03
0.003
IngestioB
Radium
226-228
Ground
Users of PWS in violation of MCL
Usen of PWS serving >1,000 persons
Usen of PWS serving <1,000 people and private well usen
37,500*
7,200,000
5,817,200
9.62 pCi/L
0.70 pCi/L
1.1 pCi/L
Ingestion
Radon
Groand
Usen of PWS serving >1,000 people
Usen of PWS serving <1,000 people and private well uaen
7,200,000
5,817,200
165 pCi/L
245 pCi/L
Ingestion
Inhalation
Volatile Organic*:
Trichloroethene
Tetrachloroe these
Trichlorocthane
Ground
aud
Surface
Usen of PWS in violation of MCLs
Usen of PWS serving <3,300 currently not regulated
Usen of private wells subject to commercial,industrial, and
active hazardous snd municipal waste sites, etc.
49,800*
130,000
310,000*
TCE: 0.011
PCE- 0.014
TCA: 0.008
TCE: 0.03
PCE: 0.007
TCA: 0.20
Ingestion
Inhalation
Alachlor
Ground
and
Surface
Usen of PWS with levels over the pMCL in highly
vulnerable areas
Usen of private wells near agricultural areas
26,700*
37,100*
0.002
0.002
Ingestion
Inhalation
Atrazine
Ground
and
Surface
Usen of PWS with levels over the pMCL in highly
vulnerable areas
Usen of private wells near agricultural areas
26,700*
37,100*
0.003
0.003
Ingestion
' Exposure assumption! ire consistent with those made is the Region V comparative risk study. For those populations identified by an asterisk, we assumed
that the Basin potentially exposed population is similar to that in Region V and estimated the affected population in the Basin by applying a factor of 0.651 (i.e.,
the ratio of Basin population to Region V population). The remaining populations are estimated from Exhibit S-2 and from the FRDS retrieval.
2 Avenge exposure concentrations are those determined to be appropriate for Region V and are assumed to also be appropriate for the Basin.
' "Violation" in the case of colifonn refers to both types - MCL and monitoring/reporting.
-------
V- 16
EXHIBIT S-4
CakaUlton «f CarctMgeaic RMo
Contaminant
Cancer Slope
Drinking Water
Chronic Dally Intake
Chemical-
Estimated Exposed
Potential Cancer
Chemical-specific Avg
Factor (mg/kg/d)'
Concentration (tng/L)
(mg/kg/day)
sped He Lifetime
Popilatk»
Cases (70-yr
Popilatloa-weighted
Risk
lifetime)
Cancer Risk
Lead
1.4 * 10"
0.051
3.1 x 10'
4.3 x 101
1,950
8x10"
2A x 10 *
0.029
1.7 x 10'
2.4 x 10'
370,000
0.09
0.003
1.8 x 10"
2.5 x 10*
30,232,600
0.76
Total Trihalomethanes
6.1 x 10'
0.101
2.9x10"
1.7 x 10"
15,600
0.03
1.0 x 10'
0.060
1.7 x 10'
1.0 x 10'
1,540,000
16
Trichloioethene
l.l x 10 '
0.011
3.1 x 10'
3.5 x 101
49,800
0.02
8.4 x 10*
0.030
8.6 x 10"
9.4 x 10"
130,000
1.2
0.030
8.6 x 10"
9.4 x 10*
310,000
2.9
Tetrachloroethene
5.11101
0.014
4.0 x 10'
2.0 x 10"
49,800
0.10
9.4 x 10"
0.007
2.0 x 10"
1.0 x 10'
130,000
13
0.007
2.0x10"
1.0x10'
310,000
3.2
Alachlor
8.0* 10*
0.002
5.7x10'
4.6 x 10"
26,700
0.12
4.6 x 10*
0.002
5.7 x 10'
4.6 x 10"
37,100
0.17
Atrazine
2.2* 101
0.003
8.6 x 10'
1.9 x 10'
26,700
0.50
1.9 x 10'
0.003
8.6 x 10'
1.9 x 10'
37,100
0.70
Radium
3.6 x 105
9.62 pCI/L
2.7 x 10'
9.9 x 107
37,500
0.04
8.9 x 10'
226-228
0.70 pCi/L
2.0 x 102
7.2 x 10'
7,200,000
5.2
1.06 pCl/L
3.1 x 102
1.1 x 10"
5,871,200
6.5
Radon1
ND
165 pCi/L
ND
2x 10*
NA
3.7
3.0 x 10*
245 pCi/L
to
to
6.0 x 10"
73
'The lifetime cancer risk aid potential number of cancer cases from radon exposure were not calculated from the exposure data provided above. Tke range of estimates of the cancer risk was obtained from the Federal
Register (51(189):34845, September 30, 1986) and inclndes inhalation exposure. The anmber of nationwide cancer cases was also provided in this soarce, and we scaled these nnmbeis to the Basin by applying a
maldpiier of 12.2 percent.
-------
V - 17
As shown in Exhibit S-4, the contaminant that contributes the greatest number of potential cancer cases over a 70-year
period is radon, followed successively by total trihalomethanes, radium, tetrachloroethene, and trichloroethene.
Exhibit S-5 presents the approach and results of the noncarcinogenic risk characterization. As for the carcinogenic
estimates, we combined the estimated drinking water concentrations and exposure assumptions to produce a range of
chronic daily intakes. To calculate the Hazard Index, we divided the chronic daily intake by the oral reference dose.
We also include in Exhibit S-4 the population exposed to each contaminant. As shown by the Hazard Indices, nitrate
presents the most significant risk, which is manifested as methemoglobinemia in infants of less than six months of age.
It is also expected that lead would pose a significant individual and population noncancer risk through subtle biochemical
changes in infants and small children; however, neither an oral RfD nor Health Advisory levels have been developed
for subchronic exposure to lead through drinking water. Although not included in the noncarcinogenic risk estimates,
it is also estimated that microbiological agents may cause between 5,500 and 27,500 cases of waterbome disease.
Welfare Impacts
In this section, we present estimates for health care and environmental costs due to contaminated drinking water within
the Great Lakes Basin. We base our health care cost estimates on the number of annual cancer cases associated with
public sources of drinking water estimated in the aggregated drinking water risk assessment. These damages do not
account for the costs associated with replacement of contaminated drinking water supplies; instead, we present these
costs separately. In addition, we did not estimate costs of installing residential water purification systems and the cost
of bottled water.
-------
V - 18
EXHIBIT S-5
Calculation of Noncardnogenic Risks
Contaminant
Oral Reference Dose
Drinking Water
Daily Intake
Hazard Index
Population Exposed
(mg/kg/day)
Concentration (mg/L)
(mg/kg/day)
Nitrate as Nitrogen
1.0
15.75
2.25
2.25
167,000
Trichloroethanc
0.09
0.008
2.3 x Iff4
0.0025
49,800
0.2
5.7 x 10°
0.063
440,000
Atrazine
0.005
0.003
8.6 x 10"5
0.017
63,800
Lead1
ND
0.05
3.1 x Iff3
ND
1,950
0.03
1.7 x Iff3
ND
370,000
0.003
1.8 x Iff4
ND
30,232,600
Total Trihalomethanes as
0.01
0.10
2.9 x Iff3
0.29
15,600
Chloroform
0.06
1.7 x Iff3
0.17
1,540,000
Trichloroethene
0.007
0.01
3.1 x Iff4
0.044
49,800
0.03
8.6 x Iff4
0.12
440,000
Tetrachloroethane
0.01
0.014
4.0 x Iff4
0.04
49,800
0.007
2.0 x Iff4
0.02
440,000
Alachlor
0.01
0.002
5.7 x Iff5
0.0057
63,800
ND = Not determined
1 Neither an oral RfD nor Health Advisory levels have been developed for subchronic exposure to lead through drinking water. Such a level could be developed
by converting lead concentrations in drinking water to a child's blood-lead level (PbB). The difference in child's PbB level from drinking water contributions alone
could then be estimated.
1 The lifetime cancer risk and potential number of cancer cases from radon exposure were not calculated from the exposure data provided above. The range of
estimates of the cancer risk was obtained from the Federal Register (51(189):34845, September 30, 1986) and includes inhalation exposure. The number of
nationwide cancer cases was also provided in this source, and we scaled these numbers to the Basin by applying a multiplier of 12.2 percent.
-------
V- 19
Health Care Costs
Given the lower and upper bound cost per case estimates, we calculate the lower and upper bound health costs as
follows:
Lower bound cost per case estimate = HCL = (1.14)($80,000) = $91,200
Upper bound cost per case estimate = HQ; = (1.14)($137,000) = $156,000
Environmental Costs
Replacement Costs
Using RCG/Hagler, Bailly, Inc. 's approach, we can estimate the cost of replacing contaminated drinking water supplies
given the number of annual private well replacements and public water supply hookups associated with ground-water
contamination. We use RCG's estimated capital costs of $3,500 to replace a private well by digging a new well and
$300,000 to replace a public water supply well by extending a hookup from another public supply. These costs do not
include annual operating costs.
EPA Region 5's Safe Drinking Water Branch estimates that about 4 percent of households served by private wells have
replaced these wells. To estimate the Basin-wide cost of replacing private water supplies, we will assume that 4 percent
of households in the Basin served by private wells have also replaced their wells. To estimate the total number of
households in the Basin served by private wells, we will subtract the number of people served by public water systems
in the Basin (obtained from FRDS) from the total Basin population and divide this difference by four, the number of
people per household, to estimate the number of households. Then, the Basin-wide cost of replacing private water
supplies will be
(Number of private wells replaced)($3,500) = Basin-wide cost of replacing private wells
To estimate the Basin-wide cost of replacing contaminated public water supplies, we can utilize violations data retrieved
from FRDS. First, for each MCL violation recorded in FRDS, the number of months the system was out of compliance
is also recorded. For our upper bound, we could assume that all systems in violation would need to be connected to an
alternative system; this estimate would most likely be very high and, possibly, highly inaccurate. For our lower bound,
we could select a cut-off for the number of months of noncompliance for which an alternative system would be needed.
We will investigate this proposed method by speaking with our FRDS contact and reviewing the FRDS data element
dictionary.
Treatment Costs
In general, superintendents for sample drinking water supply systems reported no significant changes in treatment costs
between 1987 and 1990.
-------
V- 20
T. AGGREGATED GROUNDWATER
Problem Area Description
Aggrcgated ground-water contaminant source risks have been analyzed on a qualitative basis to assess the cumulative
impact on ground-water resources within the Great Lakes Basin and to provide an estimate of the relative contaminant
contribution of the major sources to human health, ecological impacts, and welfare costs. The major sources of
contamination include waste disposal facilities (e.g., landfills, septic systems, land spreading), direct discharge to ground
water (i.e., injection wells), storage facilities, spills, and pesticide and fertilizer applications.
Magnitude of the Problem
Because of the high level of industrial activity, and agricultural activity to a lesser extent, there are a variety of sources
of ground-water contamination within the Basin. Ground water is used intensively in the Basin as a source of drinking
water: Approximately 43 percent of the population of the Basin relies on ground water for potable water supplies.7
Although ground-water resources are generally high in quality, numerous incidents of contamination and the large
number of potential sources of contamination pose a chronic risk to the integrity of ground water within the Basin.
Hie Office of Technology Assessment (OTA) conducted a survey of scientific literature regarding sources of
ground-water contamination and compiled a list of thirty-three potential sources of ground-water contamination. The
following list applies to the nation as a whole and is not meant to suggest that each source is present in the Great l.airrx
Basin; however, several of the sources presented in the list may exist in the Basin:
(1)
subsurface percolation from septic tanks and cesspools,
(2)
injection wells, (Exhibit T-l)
(3)
land application of wastes,
(«)
landfills,
(5)
open dumps,
(6)
residential (local) disposal,
<7)
surface impoundments,
(8)
waste tailings,
(9)
waste piles,
(10)
materials stockpiles,
(11)
graveyards,
(12)
animal burial,
(13)
aboveground storage tanks,
(14)
underground storage tanks,
(15)
containers,
(16)
open burning and detonation sites,
7 Tkk percentage was derived from ur estimates of the Basin-wide populations relying on pablic and private drinking water
supplies presented In Exhibit S-2 af this memo. This percentage acceanls for populations that receive drinking water from either
groand-wster based public water supply systems or from private wells. Thb percentage may be low becaase a portion of the sarface-
water baaed pablic water supply systems that serve the remaining 57 percent of the Basin population may receive part of their raw
water from ground-water sources.
-------
V - 21
(17)
(18)
(19)
(20)
(21)
(22)
(23)
(24).
(25)
(26)
(27)
(28)
(29)
(30)
(31)
(32)
(33)
radioactive disposal sites,
pipelines,
material transport and transfer operations,
irrigation practices,
pesticide applications,
fertilizer applications,
animal feeding operations,
de-icing salts applications,
urban runoff,
percolation of atmospheric pollutants,
mining and mine drainage,
production wells,
other wells,
construction excavation,
ground-water/surface-water interaction,
natural leaching, and
salt-water intrusion/brackish-water upconing.
Chemical spills may also pose a hazard to ground water. In 1989,51 chemical spills that impacted ground water were
reported in EPA's Emergency Response Network System (ERNS) data base of chemical spills in counties within the
Great Lakes Basin. Twenty of these spills occurred in Michigan, nine oocuned in Illinois, seven occurred in both New
York and Ohio, five occurred in Indiana, and three occurred in Wisconsin*
The types of chemicals associated with these sources of ground-water contamination include volatile organic compounds
(VOCs), non-aqueous phase liquids (NAPLs), pesticides, heavy metals, inorganic salts, and nutrients. In its Region V
comparative ride study, EPA used the following representative chemicals to evaluate residual and potential risk to
populations and ecosystems due to ground-water contamination; we include them here as examples of the types of
chemicals that may pose risk to those populations in the Basin utilizing ground water:
tricboloroethene,
toluene,
chromium,
nitrates,
alachlor,
atrazine, and
fecal coliform.
Geographic Analysis of Contaminant Sources
Region 5, in its comparative risk study, concluded after preliminary assessment of existing data that the Great Lakes
Basin would have lower overall residual public health risks due to ground-water contamination than would Region 5.
EPA's conclusion was based on the following factors:
Major population centers within the Basin use lake water, rather than ground water, for their potable
* Spill* and other accidental releaie* arc addressed more falty In Section D.
-------
Exhibit T-1
Lake Superior
Michigan
SITES PER
COUNTY
~ 0
RCRA FACILITIES
UNDERGROUND
INJECTION
Lake Huron
9
Ontario
Source: Hazardous Waste Data Management System. 1990
EFftiffa
UNITED STATES
ENVIIONMENTAl HOTECTION ACENCT
CHAT lAKiS NATIONAL PtOCIAM OFFICE
-------
V- 22
water supplies;
Much of the vulnerable and productive farmland in Region V lies outside the Great Lakes Basin, thus
reducing the impact to ground water within the Basin to contamination by agricultural chemicals; and
Most CERCLA sites in Region 5 lie outside the Basin.
In terms of ecological impacts, however, the Great Lakes Basin has several harbors and associated rivers that are
severely polluted; contaminated ground water contributes to pollution of surface water as well as sediments in many of
these cases. The percentage of the total impact is difficult to estimate, but it appears to be as high as 20 percent of the
problem in some areas, such as the Grand Calumet Basin and connecting channels, according to Region 5's comparative
risk study.
Human Health Impacts
Although ground water is the source of drinking water for as much as 43 percent of the Basin population, major
population centers within the Basin use lake water for their potable water supplies. Contaminated ground water
principally affects human health when it impacts drinking water supplies. In this section, both residual and potential
risk will be discussed. In general, residual carcinogenic and noncarcinogenic risks due to contaminated ground water
are negligible for the population at large. In Section C, Aggregated Drinking Water Risk Assessment, we estimated the
average annual number of cancer cases presented by all contaminants of concern in drinking water supplies within the
Basin to be 1.14 and the average total cancer cases over a 70-year lifetime to be 80. There are two factors responsible
for this low incidence of excess cancers due to drinking water.
According to analysis of FRDS data for Region 5, less than four percent of public
water supplies within Region V are contaminated with carcinogens above the MCL
level; it is assumed here that this percentage applies also to the Great Lakes Basin;
and
Usually, it takes less than three years to install treatment systems or replace wells, therefore the time
of the exposure is minimized.
However, there are subpopulations, namely small community water systems and residential wells where the risks are
greater due to longer exposure times and the potential for higher concentrations of contaminants. This is due primarily
to the lack of resources to replace contaminated water supplies, use of shallow wells, and, in the case of residential wells,
a lack of periodic monitoring that can lead to longer term exposures to contamination. These subpopulations are
significant, as Exhibit T-2 demonstrates.
-------
V - 23
EXHIBIT T-2
Great Lakes Basin Population Relying on Ground Water
Population Type
Percent of Total Population (30,232,600)
Population relying on ground water
Population using small (<3,300 people served)
community PWS systems
Population served by private wells
43%
15%*
19%
This percentage is based on estimates for Region 5; we assume that a similar population within the Great Lakes Basin
also receive drinking water from small community PWS systems.
Given the percentages presented in Exhibit T-2, we estimate that as much as 52 peroent of the Basin populntion that uses
ground water relies on small community and private wells. Since the numbers of MCL violations are generally greater
for these subpopulations according to the Region 5 comparative risk study, there are small communities and single
residents exposed to a significantly higher risk than the general population.
As an example, exposure to pesticides is greater in residential wells and small community systems. From EPA's analysis
of available pesticides contamination data, atrazine and alachlor are representative of the most widely used and toxic
pesticides in Region 5 and possibly within the Basin. Because of their extensive use they are the most commonly
detected pesticides in ground water. The greatest frequency of detections of pesticides and MCL violations occur in
areas rated highly vulnerable to ground water contamination and where wells less than 100 feet deep are commonly used.
For example, 38 percent of such wells tested in Minnesota had at least trace levels of atrazine (2 percent above MCL).
However, less than two percent of the vulnerable public water supply wells in Illinois had detectable levels of pesticides.
It is important to remember, however, that much of the vulnerable and productive farmland in Region 5 lies outside the
Great Lakes Basin, and, thus, the risks posed to small community systems and private wells by pesticides is likely to
be lower in the Basin than in the Region as a whole.
A detailed discussion of cancer and non-cancer risks can be found in the aggregated drinking water risk assessment
discussion. Both the cancer and non-cancer residual risks are not considered large due to the stringent standards and
regulatory infrastructure. The potential health risks would be much greater if no action was taken to mitigate and prevent
ground water contamination.
-------
V - 24
Ecological Impacts
Ground water impacts ecological systems via the discbarge of contaminated ground water to surface water bodies and
wetland areas and, in some cases, underlying sediments. As described above, the Great Lakes Basin has several harbors
and associated rivers that are severely polluted; it is likely that contaminated ground water contributes to pollution of
surface water as well as sediments in many of these cases. In its comparative risk study, EPA Region 5 estimates that
as much as 20% of overall contaminant sources to the Grand Calumet Basin and connecting channels may be atributed
to contaminated ground water.
From current analyses it is evident that ground water can have a significant impact in localized areas but is not likely
to be the major contributor of nonpoint source contamination within the Basin. Most of these impacts would be
reversible, provided the sediments could be remediated. Region 5 estimated in its comparative risk study that ground
water could be responsible for up to 15 percent of the total contaminant contribution to surface water in the Region; it
is possible that this percentage would be less for the Great Lakes Basin, given the lesser degree of agricultural activity
in the Basin.
In addition, chemical spills on land could adversely impact local habitat directly or via contaminated ground water that
leaks away from the spill and discharges into wetlands. Effective response to spills is essential to minimize ecological
damage.
Welfare Impacts
Estimates of health care and environmental costs of contaminated drinking water as well the costs of treating drinking
water are presented in Section C, Aggregated Drinking Water Risk Assessment.
Summary and Uncertainties
As for the drinking water risk assessment, we relied on the Region 5 comparative risk study for much of our analysis
of risk posed to ground water within the Basin; therefore, the uncertainties associated with the use of the FRDS data base
and the USGS data apply also to this analysis.
-------
V - 25
U. OZONE AND CARBON MONOXIDE
Problem Area Description
Pollution of ambient air by ozone and carbon monoxide (CO) is significant in many areas of the Great Lakes Basin.
Ozone is formed in the ambient air as a result of a series of complex chemical reactions involving volatile organic
compounds (VOCs) and nitrogen oxides emitted from both stationary and mobile sources, atmospheric oxygen, and
sunlight. Ozone can adversely affect human health, agricultural crops, forests, ecosystems, and materials. Interactions
of ozone with nitrogen oxides and sulfur oxides can also contribute to the formation of acid precipitation. CO is formed
as a result of incomplete combustion and is emitted from both stationary and mobile sources. CO adversely affects
human health and the health of other organisms because of its ability to reduce oxygen absorption capacity, transport,
and utilization.
Ozone and CO are relatively short-lived pollutants. Consequently, ozone and CO pollution problems in the Great Lakes
Basin generally arise from pollution sources in the Basin.9
We do not discuss the geographical distribution of ozone because this pollutant is not emitted but is the product of a
series of chemical reactions dependent on emissions of ozone precursors (i.e., VOCs, CO, and nitrogen oxides). VOCs
are emitted from sources including automobiles, dry cleaners, bakeries, auto body paint shops, household cleaning
products, and any sources using solvents. Nitrogen oxides are emitted in the combustion of fossil fuels; motor vehicles
are the predominant source. Consequently, ozone precursor emission and ozone pollution problems are likely to be
concentrated in densely populated and highly commercialized areas of the Great Lakes Basin.
CO pollution arises from mobile and stationary sources. Tbe major source of CO is motor vehicle exhaust, particularly
who] engines are burning fuel inefficiently as they do when engines are started, idling, or moving slowly. Other sources
include incinerators and industrial processes. Mobile sources of CO are likely to be concentrated in densely populated
and highly commercialized areas of the Great Lakes Basin. According to EPA's Aerometric Information Retrieval
System (AIRS), stationary sources of CO in tbe Great Lakes Basin are concentrated in the following areas:
Milwaukee county, Wisconsin;
Cook county, Illinois (Chicago);
Lake county, Indiana (East Chicago, Gary);
Wayne county, Michigan (Detroit); and
numerous other counties in Wisconsin (i.e., Waukesha, Brown, Winnebago, Sheboygan, Outagamie,
Dodge, Manitowoc, and Washington) and Indiana (i.e., Allen, Elkhart, St. Joseph, and La Porte).
Exhibit U-l illustrates the geographic distribution of CO emissions monitoring sites by county throughout the Basin.
Exhibit U-2 shows the geographic distribution of estimated CO emissions by Basin county. AIRS reports 1,511
stationary sources of CO emissions in 139 of the 213 Great Lakes Basin counties. Great Lake Basin counties with the
greatest quantities of CO emissions from stationary sources include the following:
Porter and Lake counties in Indiana;
Cook county, Illinois;
Gratiot, Macomb, Wayne, Saginaw, Muskegon, Delta, and Charlevoix counties in
* Sources from beyond (be Basin are also important, however, in areas such as northeastern Illinois. Much of Chicago lies
outside tbe Basin, but Is nevertheless a source of these pollutants to tbe Basin. Such loadings arc not addressed in thfe analysis.
-------
Exhibit U-1
CO Emissons Monitoring Sites
Great Lakes Basin Counties
Lake Superior
Lake Huron
I. :
::: : :i::::
Lake Ontario
Michigan
MONITORING
SITES
~
26- 50
S&ERA
76-100
> 100
US ENVIRONMENTAL PROTECTION AGENCY
Source: Aerometric Information Retrieval System, 1990 GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Exhibit U-2
CO Emissions
Great Lakes Basin Counties
Lake Superior
Lake Huron
Lake Ontario
¦UK0|
Michigan
TONS PER YEAR
101 - 1,000
1,001 - 10,000
10,001 - 100,000
> 100,000
Source: Aerometrlc Information Retrieval System, 1990
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
V - 26
Michigan;
Niagara county, New York; and
Waukesha, Milwaukee, Outagamie, Columbia, and Manitowoc counties in
Wisconsin.
Data from AIRS indicates that an estimated 610,000tons of CO are emitted annually from stationary sources in the Great
Lakes Basin.
Human Health Impacts
Toxicity Assessment
Ozone. Exposure to unacceptable levels of ozone can cause various health effects including alterations in pulmonary
functions, symptomatic effects, aggravation of preexisting respiratory disease, altered host defenses systems, and extra-
pulmonary effects. Studies indicate that healthy individuals exposed to ozone may experience chest pain, coughing,
wheezing, pulmonary and nasal congestion, labored breathing, sore throat, nausea, increased respiratory rate, and loss
of lung function. Health problems are exacerbated for individuals who exercise, because of the increased intake of air.
There is growing concern that long-term exposure to ozone at current levels may lead to chronic effects. Preliminary
data indicates that these effects take the form of irreversible lung injury and/or lung disease such as lesions in the lung.
Children and outdoor workers are considered to be most susceptible because, on average, they spend more time outdoors.
Further, children may be particularly sensitive because their lungs are still developing.
Primary National Ambient Air Quality Standards (NAAQS) were established to define levels of air quality that are
necessary to protect public health with an adequate margin of safety. The NAAQS for ozone is 0.12 parts per million
(ppm). The standard is attained when the average expected number of days per calendar year with maximum hourly
average concentrations above 0.12 ppm is equal to or less than one.
Carbon Monoxide. Exposure to CO can impair breathing, visioa, alertness, and mental function; aggravate preexisting
conditions such as angina; and, under acute conditions, cause nausea, dizziness, unconsciousness, and death. The
primary NAAQS for CO are 9 ppm for an 8-hour average concentration and 35 ppm for a 1 -hour average concentration,
neither of which is to be exceeded more than once per year.
Exposure Assessment
Ozone. Thirty-nine counties in the Great Lakes Basin are classified in the Code of Federal Regulations as not attaining
the NAAQS for ozone. Further, of the 60 Great Lakes Basin counties that were monitored in 1985 through 1989, 48
recorded hourly average values above 0.12 ppm. This corresponds to over 14 million people in the Great Lakes Basin
who are living in areas where ambient concentrations exceed the primary NAAQS for ozone.
Carbon Monoxide. Parts of eight counties in the Great Lakes Basin are listed in the CFR as not attaining the primary
NAAQS for CO. Review of 1985 through 1989 monitoring data indicates that portions of the population of eight Great
Lakes Basin cities are exposed to CO levels violating the NAAQS of 9 ppm (i.e., concentrations are s 9.5 ppm), and
portions of the population in two of those cities are exposed to levels exceeding 15 ppm (i.e., 215.5 ppm). Combining
these monitoring data with the estimated population associated with each monitoring location (using the method applied
for the Region 5 Comparative Risk project), it is estimated that approximately 1.1 million people were potentially
exposed to levels exoeeding 9 ppm and 220,000 of these people were potentially exposed to levels exceeding 15 ppm.
-------
V- 27
Risk Characterization
Ozone. To characterize the human health effects resulting from exposure to ozone, we applied a method developed by
RCG/Hagler, Bailly, Inc. under contract to EPA. (This method was used in the Region 5 Comparative Risk project.)
In this approach, the annual number of asthma attacks and the annual number of people days of respiratory restricted
activity are calculated as follows:
Annual Number of Asthma = (0.04) (POPj) (EXDAYj) (0.00037)
Attacks at Location j
Annual Number of People Days = (0.96) (POP,) (EXDAYj) (0.00116)
of Respiratory Restricted
Activity at Location j
where, j = location of monitoring site,
POPj = exposed population at monitoring site j, and
EXDAY, = number of days where hourly average ozone concentration exceeded 0.12 ppm
at monitoring site j.
Using 1989 monitoring data with 1988 county populations (as estimated by the U.S. Bureau of the Census), we estimate
that 1300 annual asthma attacks and 98,200annual people days of respiratory restricted activity in the Great Lakes Basin
result from exposure to ozone.
Carbon Monoxide. Methods developed by RCG/Hagler, Bailly, Inc. under contract to EPA were also used to
characterize public health effects resulting from exposure to CO. Id this approach, it is assumed that at CO
concentrations greater than 9 ppm but less than IS ppm, 10 percent of the exposed population are at moderate risk of
increased angina pain and 90 percent of the exposed population are at low risk of mild symptoms. At concentrations
above 15 ppm, 10 percent of the exposed population are at high risk of increased angina pain and the remaining 90
percent of the population are at moderate risk of mild symptoms. Results based on review of the 1987 through 1989
monitoring data indicate that 22,000 people in the Great Lakes Basin are estimated to be at high risk of experiencing
increased angina pain and an additional 115,000 people are at moderate risk of experiencing increased angina pain. In
addition, we estimate that 200,000 people in the Basin are at moderate risk of experiencing mild symptoms, while 1
million people are at low risk.
Ecological Impacts
Data on the ecological impacts of CO are not readily available. Consequently, the following discussion focuses on the
ecological impacts of ozone pollution.
Ozone is generally considered to be the most phytotoxic air pollutant adversely affecting terrestrial vegetation. Damage
often begins with degeneration of feeder roots and mycorrhizae and later progresses to the onset of above ground
symptoms.10 The following subsections discuss the severity of the effects of ozone exposure and the reversibility of
" Manlon (1981) it died in Review of the National Ambtent Air Quality Standards for Ozont. Preliminary Assessment of
Scientific and Technical Information (EPA, Office of Air Quality Planning and Standards, Research Triangle Park, N.C., 1986).
-------
V- 28
these effects.
Severity
The following species level effects of ozone oo herbaceous and woody vegetation observable above ground have been
well documented:
injury to foliage;
reduction in growth;
reduction in yield;
increased susceptibility to pests and pathogens; and
alterations in reproductive capacity.11
Several tree species are particularly susceptible to the effects of ozone. Observed species effects include:
reductions in the size of annual rings in Ponderosa, Jeffrey, and white pine stands in southern
California, attributed to exposure of trees to ozone over a period of 10 to 20 years; and
reduced growth rates of the eastern white pine, red spruce, balsam fir, and Fraser fir in West Virginia.
Ecosystem-level effects observed from exposure to ozone include
overall declines in primary productivity of ooniferous and mixed deciduous forests; and
significant changes in community structure and composition due to increased mortality of sensitive
species.
In general, the response of vegetation to phytotoxic gaseous inorganics such as ozone is a function of both the ambient
chemical concentration and the exposure duration. The higher the ambient concentration, the shorter the exposure
required to produce a specified level of damage. (Although mathematical formulations have been developed to relate
damage to both duration and intensity of exposure, they are not appropriate for the level of effort associated with this
analysis.) In a recent assessment of the scientific and technical information relevant to the NAAQS for ozone, EPA
categorized selected tree species according to three ranges of sensitivity to ozone exposure: sensitive (i.e., significant
growth reduction at 0.03 to 0.12 ppm); intermediate (i.e., growth reduction at 0.12 to 020 ppm); and resistant (i.e.,
growth reduction at greater than 0.20 ppm).11 Sensitive species include American sycamore, loblolly pine, pitch pine,
eastern white pine, and green ash. Intermediate species include sweet gum, white ash, and sugar maple. Resistant
species include black cherry and white spruce.
The Region 5 comparative risk project states that damage to white pine in the eastern United States and Canada has been
associated with repeated exposure to peak ozone concentrations of 0.08 ppm or greater. In addition, it is believed that
ozone is a major contributor to the decline in growth rates of red spruce at numerous high-elevation sites throughout the
Appalachian mountains.
" Unfinished Business: A Comparative AutgnnBl of Environmental Problems. Appendix III. EPA, Ecological Rkk
Workgroup, Office of Policy, Planning, and Evaluation. Washington, D.C. 1987.
" Review of the National Ambient Air Quality Standard for Owm. Preliminary Assessment of Scientific ind Technical
Information. EPA, Office of Air Quality Planning and Standards, Research Triangle, NC, 1986.
-------
V - 29
Reversibility of Damage
The panel of scientists convened by the Cornell Ecosystems Research Center to assist EPA's national comparative risk
study j udged the impacts of ozone on forests and grasslands to be reversible in terms of decades. However, full recovery
from severe damage (i.e., total elimination of many species) to a mixed species climax deciduous forest could require
more than a century.
Welfare Impacts
Welfare effects associated with ozone and CO pollution include health care costs and damages to agricultural crops and
materials. In the subsections below we present costs of health care, agricultural crop losses, and costs of material
damages resulting from ozone pollution. Welfare effects from CO pollution arc not quantified.
Health Care Costs of Ozone Exposure
Ozone pollution in the Great Lakes Basin presents a risk to human health. The costs associated with health care
necessitated by the adverse effects of exposure to ozone are estimated as one measure of the welfare impacts of ozone
pollution. The method used to estimate health care costs was developed by RCG/Hagler, Bailly, Inc. under contract to
EPA. In general, their approach applies unit costs for the two health effects predicted from ozone exposure:
annual asthma attacks, valued at $44 in 1988 dollars; and
annual respiratory restricted activity days, valued at $11 in 1988 dollars.
Multiplying these unit costs by the frequency of occurrence of these effects, we estimate that the total annual cost of
health care resulting from ozone exposure is $1.2 million. $1.1 million of this total results from costs Wjt^
restricted activity days, and the remaining $0.1 million results from costs associated with asthma attacks.
Effects of Ozone of Agricultural Crops
The Region 5 comparative risk project estimated the effects of ozone exposure on agricultural crops. The following
discussion borrows heavily from the Region 5 report and consists of relatively little information developed specifically
for the Great Lakes Basin.
The majority of the evidence available on the effects of ozone exposure on agricultural crops focuses on reduction in
growth and yield resulting from long-term exposure to various ozone concentrations. Although the actual amount of
yield loss due to decreased aesthetic value or appearance is important for some crops (such as tobacco, spinach, and
ornamentals), it is difficult to quantify. Consequently, ozone induced yield loss is primarily quantified in terms of
reduction in weight or volume. It is noted, however, that plant appearance can be affected by exposure to concentrations
as low as 0.041 ppm for several weeks or 0.10 ppm for two hours.
In the Region S comparative risk report, data presented in a report from the staff at EPA's Corvalis, Oregon laboratory
and exposure response relationships developed by EPA's Office of Air Quality Planning and Standards (OAQPS) were
used to predict the impact of current ozone exposure on crop yield, for those crops that are most susceptible to ozone
and are grown in significant quantities in Region 5. Monitored ambient ozone concentrations were combined with
composite predicted relative yield loss functions to estimate soybean and wheat loss at specific geographic sites. Using
these point estimates, crop losses were projected for all areas where the crop is cultivated in Region 5. The area-
weighted yield loss was determined to be approximately 7 percent for soybean and 12 percent for wheat.
-------
V -30
To estimate ozone effects on soybean and wheat production in the Great Lakes Basin we applied the Region 5 findings
to the Great Lakes Basin acreage planted in soybeans and wheat. In 1988,4,519,000 acres of soybeans and 3,140,000
acres of wheat were planted in the counties in the Great Lakes Basin.13 This corresponds to an annual yield of
approximately 168 million bushels of soybeans and 153 million bushels of wheat. Applying the Region 5 area-weighted
yield loss estimates to these quantities results in an estimated loss of 12 million bushels of soybeans, with a value of
approximately $90 million, and 18 million bushels of wheat, with a value of approximately $64 million, in the Great
Lakes Basin.
Materials Damage
The final type of welfare impact from ozone pollution that can be easily quantified is the cost of materials damage due
to this pollution. To estimate the cost of material damage in Region 5, RCG/Hagler, Bailly, Inc. provided EPA with an
estimated cost of $0.94 per capita for anti-ozonant additions to materials that could be damaged by ozone. Applying
this per capita cost to the population of the Great Lakes Basin, we find that the estimated cost of materials damage from
ozone pollution is approximately $19 million per year.
0 Baaed on data for 1988 from tht Conservation Technology Information Center (CTIC).
-------
V - 31
V. PARTICULATE EMISSIONS
Problem Area Description
Airborne particulate matter smaller than 10 microns in diameter (PM10) causes adverse health, welfare, and ecological
effects. Major sources of PM10 include a variety of industrial and commercial processes, motor vehicles, wind blown
dust, and residential heating. La this problem area we address particulate matter generally, excluding sulfate aerosols
(see Problem Area M, Sulfur and Nitrogen Oxide Emissions) and air toxics that may be in particulate form (see Problem
Area N, Hazardous and Toxic Air Pollutants) from this discussion. This problem area parallels the Region 5 discussion,
and some of the analysis presented in this section is borrowed directly from the Region 5 work.
The following are significant sources of PM,0 in the Great Lakes Basin:
steel mills, including coke batteries, blast furnaces, sinter plants, steel furnaces, coke byproduct plants,
and slag handling;
industrial open dust sources, including unpaved roads, paved roads, and storage piles;
iron and aluminum foundries;
asphalt and asphaltic concrete plants;
Portland cement plants;
lime plants;
construction materials plants, particular conveyors, quarrying, and other materials handling;
construction and demolition;
grain terminals, including loading and unloading, particularly ship loading and terminal loading;
landfills;
brick kilns;
boilers; and
surface mining.
Analysis of ambient particles (aerosols) collected in Chicago in 1971 indicated that crustal dust (18 percent), coal (6.4
percent), steel processing (3.9 percent), limestone (3.2 percent), motor vehicles (2.8 percent), and residual oil (1.4
percent), as well as sulfates and nitrates from unspecified sources (16.8 percent), accounted for significant portions of
observed particulate matter pollution.14
Pollution of the ambient air by particulate matter is generally concentrated in areas in the local vicinity of sources.
Therefore, particulate matter pollution problems in the Great Lakes Basin are likely to be caused by anthropogenic and
natural emissions in the Basin. According to EPA's Aerometric Information Retrieval System (AIRS), stationary
sources of particulate matter emissions in the Great Lakes Basin are concentrated in the following areas:
Milwaukee county, Wisconsin;
Cook county, Illinois (Chicago);
Lake county, Indiana (East Chicago, Gary);
Wayne county, Michigan (Detroit);
numerous other counties in Wisconsin (i.e., Waukesha, Brown, Winnebago, Sheboygan, Outagamie,
M Gate (1975) cited in Controlling Airborne Particle* (National Research Council, Committee on Particalale Control Technology,
Washington, D.C, 1980).
-------
V -32
Dodge, Manitowoc) and Indiana (i.c., Allen, Elkhart, St. Joseph, La Porte).
Exhibit V-l illustrates the geographic distribution of monitoring sites for particulate emissions in the Basin counties.
Exhibit V-2 shows the geographic distribution of estimated particular emissions by Basin county. AIRS reports 1,698
stationary sources of particulate matter emissions in 146 of the 213 Great Lake Basin counties. Great Lake Basin
counties with the greatest quantities of particulate matter emissions from stationary sources include the following:
Lake, Porter, and Allen counties in Indiana (62,000 tons/yr);
Cook and Lake counties in Illinois (33,000 tons/yr);
St. Louis and Itasca counties in Minnesota (31,000 tons/yr);
Wayne and Presque Isle counties in Michigan (23,000 tons/yr); and
Monroe, Oswego, Erie, St. Lawrence, and Niagara counties in New York (37,000 tons/yr).
Data from AIRS indicates that approximately 253,500 tons of particulate matter are emitted annually from stationary
sources in the Great Lakes Basin. (The 14 counties listed above account for 73 percent of the particulate matter
emissions in the Great Lakes Basin.)
Human Health Risk
Toxicity Assessment
Particulate matter less than 10 microns in diameter causes a number of respiratory problems. These small particles may
be toxic in themselves or may transport toxic elements and compounds such as lead, cadmium, and other heavy metals.
As summarized in a 1980 National Research Council report on particulate matter pollution:
the potential risk posed by particulate matter depends on the mass of particulate
matter to which people are exposed and the size and Chemical characteristics of the
particles;
only particles less than 10 microns reach the lower respiratory tract during normal
nasal breathing and only particles less than 2 to 3 microns in diameter reach the
alveoli (where gases are exchanged with the circulatory system); and
particles pose risks by dissolving and releasing contaminants to the digestive or
lymphatic system or by physically carrying contaminants (e.g., lead, cadmium)
directly to the alveoli where they can enter the circulatory system.15
Respiratory effects caused by exposure to particulate matter include increased incidence of respiratory disease, especially
in children; aggravation of existing respiratory diseases, particularly bronchitis; reduced resistance to infection; increased
respiratory symptoms; and reductions in lung function. Epidemiological studies demonstrate that airborne particulate
matter can cause premature mortality, particularly in elderly and ill persons. Particulate matter also causes various lesser
effects such as irritation of the eyes and throat.
Primary National Ambient Air Quality Standards (NAAQS) were established to define levels of air quality that are
necessary to protect public health with an adequate margin of safety. The NAAQS for PMW are a 24-hour average
" National Rcicarch Council, rnntrnlHug Airborne Ftrtlcfct. Commit Ice an Particalale Control Techno tog;, Washington, D.C,
mo.
-------
Exhibit V-1
Particulate Emissons Monitoring Sites
Great Lakes Basin Counties
Lake Superior
Lake Huron
Lake Ontari
Lake
Michigan
MONITORING
SITES
26- 50
76-100
> 100
sSzERA
US ENVIRONMENTAL PROTECTION AGENCY
GREAT LAKES NATIONAL PROGRAM OFFICE
-------
Exhibit V-2
Estimated Particulate
Emissions
Great Lakes Basin Counties
MN
TONS PER YEAR
1 - 100
101- 1,000
1,001 - 10,000
10,001 -25,000
> 25,000
Wl
Lake Superior
Lake Huron
Lake Ontario
Lake
Michigan1
NY
Erie
PA
IL
IN
OH
$8zEPA
US ENVIRONMENTAL PROTECTION AGENCY
Source: Aerometric Information Retrieval System, 1990 GREAT LAKES national program office
-------
V - 33
concentration of ISO ug/m3 and an annual arithmetic mean of SO ug/m3. The standard is violated if the 24-hour standard
is expected to be exceeded more than one day per year or if any annual average concentrations exceed that standard.
For the purpose of human health risk assessment, we use the exposure-response functions outlined in the Region 5
comparative risk report. In that report, functions were presented relating exposure to two human health endpoints: death
and restricted activity days. The risk of death from particulate matter exposure is directly related to frequency and
magnitude of exposures to levels above 150 ug/m3. In addition, the risk of restricted activity days is directly related to
the frequency and magnitude of exposures to levels above 38 ug/m3. The specific quantitative relationships between
exposure and health effect endpoints are described in the section below on risk characterization.
Exposure Assessment
EPA has recorded considerable data on ambient concentrations of PM10. These data, retrieved from AIRS and derived
directly from the Region S comparative risk analysis when possible, form the basis of this exposure assessment.
In the Great Lakes Basin, the areas with the highest PM10 concentrations are also among the highest populated areas (i.e.,
Chicago; Detroit; Lake County, IN; and Syracuse, NY). In general, the highest concentrations are found in heavily
industrialized areas, which are found in highly urbanized areas. Therefore, particulate matter causes more adverse
human health impacts than more evenly distributed (or more rural-oriented) environmental contaminants.
The EPA Region 5 comparative risk project identified nine Great Lakes Basin localities that exceed the 24-hour standard
for PMI0 or that have annual PM10 concentrations above 38 ug/m3. Evaluation of monitoring data for the years 1985
through 1989, indicates that in the Great Lakes Basin portions of Pennsylvania and New York only one additional
locality (i.e., Syracuse, NY) exceed one of these levels. By associating populations with these areas of high PM,0
exposure, we estimate that 2.5 million people in the Great Lakes Basin are exposed to 24-hour average PM10
concentrations in excess of ISO ug/m3 and 2.7 million people in the Basin are exposed to average annual concentrations
in excess of 38 ug/m3.
Risk Characterization
The following equations, developed by RCG/Hagler, Bailly, Inc. under contract to EPA, are used to calculate the annual
number of premature deaths and the annual number of restricted activity days caused by particulate matter pollution in
the Great Lakes Basin.
365 j
Annual Deaths = 2 2 (1.6 x 10"8) (PM,Wj - 150) (POP,); and
d=l j=l
Annual Restricted j
Activity Days = 2 (0.046) (PM10j - 38) (POPj)
j=l
where, d = day of the year;
j = location of the monitoring site;
POPj = exposed population at monitoring site j;
-------
V- 34
PM,Wj = measured particulate concentration (ug/m3) on day d at location j, provided it exceeds 150;
PM,q = measured annnnl average particulate concentration (ug/m3) at location j, provided it exceeds
38.
Using the Region 5 comparative risk assessment results to characterize risks in the Region 5 portion of the Great Lakes
Basin and 1987 through 1989 monitoring data along with 1980 census data to characterize the additional risk in
Pennsylvania and New York, we estimate that exposure to PM]D in the Great Lakes Basin results in 24 annual deaths
and over 1.9 million annual restricted activity days.
Ecological Impacts
As explained in the Region 5 comparative risk report, particulate matter has similar effects on animal populations as it
does on the human population. However, studies on wildlife cannot readily be performed through epidemiological or
laboratory approaches. The Region 5 report states that sufficient information could not be found to characterize the
effects of particulate matter on other flora and fauna. Consequently, we do not characterize the ecological effects of
particulate matter.
Welfare Impacts
Welfare effects associated with particulate matter pollution include value of lives prematurely lost, value of lost days
(i.e., restricted activity days), and value of household soiling damage. The methods used to estimate these costs were
developed by RCG/Hagler, Bailly, Inc. under contract to EPA. Their approach applies unit costs for the two health
effects predicted from particulate matter exposure:
annual deaths, valued at $77,300 per death in 1988 dollars; and
annual restricted activity days, valued at $39 in 1988 dollars.
In addition, RCG/Hagler, Bailly, Inc. provides the following equation for calculating the value of household soiling
damage:
Household j
Soiling = I (0.82) (HHj) (PM10j -10)
j=l
where the terms are as described above and HHj is the number of households associated with monitoring point j.
RCG/Hagler, Bailly, Inc. point out that this equation was developed for use with total suspended particles data and that
use with PMt0 data will underestimate the welfare effects of household soiling damage.
Calculating the values described above, we estimate that a total annual value of welfare effects of Si 12 million. $75
million results from restricted activity days, $35 million of this total results from costs associated with household soiling,
and the remaining $2 million results from costs associated with premature aeaths.
The EPA Region 5 comparative risk assessment also calculates the value of particulate matter damages in the
manufacturing sector. In this analysis Region 5 develops a per capita cost of $15.46 (in 1988 dollars). Applying this
factor to the population of the Great Lakes Basin, 30,050,000, we determine that the value of these damages is an
additional $460 million.
-------
V - 35
W. UNDERGROUND STORAGE TANKS
Problem Area Definition
There are three types of storage tanks that can potentially affect human health and the environment: underground storage
tanks (USTs), ground level or on ground storage tanks, and above ground storage tanks. The risks to human health and
the environment from these tanks appear to result primarily from the routine or continuous release of petroleum products
or regulated substances to the soils, surface water, ground water, and air. In a national survey of release incidents from
USTs, 68 percent of releases were to soil; 45 percent to ground water; 22 percent to surface water; and 15 percent to
air.1*
The stored substances that are of concern include gasoline, diesel fuel, motor oils, beating oils, solvents, lubricants, and
inorganic acids and bases. The toxic constituents of these substances include benzene, toluene, xylene, ethylbenzene,
chlorinated solvents, petroleum hydrocarbons, and heavy metals.
There are an estimated 31334617 registered USTs containing petroleum products or regulated substances in the Great
Lakes Basin (Great Lakes Basin). The estimated universe of releases is approximated at 10 to 30 percent of the total
registered tanks.18 This would give an actual count of 31335 to 94,004 releases.
Although there are three basic types of storage tanks (i.e., USTs, ground level, and above ground) that can potentially
leak and thus threaten human health and the environment, this problem area focuses only on USTs. The reason for this
is two fold. First, there is little pertinent or available data on other types of tanks. Second, USTs are the most common
of the three storage tanks.
Even within this narrower focus on USTs, several assumptions were made in order to complete the analysis. They
include the following:
(1) the majority of USTs in the Great Lakes Basin contain petroleum products, particularly gasoline;1'
(2) the majority of leaks from USTs are gasoline;20
(3) USTs are primarily used by retailers (e.g., gas stations); and
(4) there is a high correlation between population density and the concentration of USTs.
The following types of facilities are not analyzed here:
14 EPA, Anilwb of the Nitloml Data Bw of Undergronnd Storage Tank Release Incidents. Office of Solid Waste, Jane 13,1986.
17 The number of registered USTs wis obuiied from each state's Notificatioa Database which lisu the number of USTs within that state.
The total number of USTs in each stale was multiplied by the percentage of that state's population falling within the GLB, and then summed
across the eight GLB states. This is based on the Region 5 observation that there is a high correlation between the presence of USTs and
population density.
" Range stated by the Region 5 comparative risk analysis to be derived from a survey of technical and regulatory literature.
" George Halloran, EPA, Region 5, Office of UST/LUST, memorandum to Loraa Jereta dated June 18, 1990. Information current
through May 31, 1990.
* EPA, 1986, o£ dt.
-------
V - 36
Tanks used to store hazardous waste. These facilities are covered in the active hazardous waste
facility problem area.
Acute accidents or releases from storage tanks, including tank collapses or explosions (estimated in
Region 5 to result in 3 to 4 deaths per year). These incidents are covered in Problem Area D, Spills
and Other Accidental Releases.
Human Health Risk Assessment
Toxicity Assessment
U.S. EPA examined possible cancer risk from exposure to gasoline in the 1987 Document "Evaluation of the
Carcinogenicity of Unleaded Gasoline." This document reviewed several animal studies including two lifetime
inhalation bioassays of unleaded gasoline in rats and mice which demonstrated statistically significant increases in renal
carcinomas and hepatocellular carcinomas in male rats and female mice respectively. On the basis of these results, U.S.
EPA concluded the evidence is sufficient to establish that gasoline is carcinogenic in animals. Carcinogenic potency
estimates from these experiments were similar: .0021 (mg/kg/day)'1 from the mouse data and .0035 (mg/kg/day)'1 from
the rat data. In accordance with EPA's Guidelines on Carcinogen Assessment, the more conservative value from the
rat study is the recommended cancer potency factor. The presence of benzene (EPA Group A, known human
carcinogen) in unleaded gasoline could contribute to the carcinogenic response, but the magnitude of this potential
contribution is unknown. It should be pointed out that the types of tumors observed in the rat and mice studies on
gasoline are not the types associated with benzene exposure.
The epidemiological literature is replete with studies suggesting a relationship between exposure to gasoline or its
constituents, principally benzene, and the incidence of certain types of cancer. Yet to date, only benzene has been
causally linked to cancer in humans. Fifty-five epidemiologic studies were reviewed by U.S. EPA and were found to
provide inadequate evidence of carcinogenicity in humans. Based on sufficient evidence of carcinogenicity in animals
and inadequate evidence in humans, unleaded gasoline is classified as an EPA B2 carcinogen (probable human
carcinogen). NESCAUM's (Northeast States for Coordinated Air Use Management) recent study21 evaluating the
health effects from exposure to gasoline offers the following summary of findings from the literature:
Based on an assessment of the available literature, gasoline is presumed to be carcinogenic to human
beings. This finding is based largely on the fact that benzene, a volatile component of gasoline, is an
established human carcinogen. Any exposure to gasoline would also involve exposure to benzene.
The epidemiological evidence regarding benzene carcinogenicity is widely accepted. The evidence
regarding the human carcinogenicity of the gasoline mixture, however, is subject to significant
uncertainties. These uncertainties stem from the limited sensitivity of epidemiological studies to
identify carcinogens (particularly weak carcinogens), the complexity of the chemical exposures
associated with the petroleum industry, and the lack of any clearly defined target organ which
gasoline may specifically affect.
" Northeast States for Coordinated Air Use Management (NESCAUM), Evaluation of the Health Effects from Exposure to Gasoline
Vapors. Final Report, August 1989.
-------
V - 37
Use of the cancer slope factor of .0035 (mg/kg/day)1 from the previously mentioned EPA document, and the Agency
verified cancer slope factor for benzene obtained from the Integrated Risk Information System, combined with projected
exposures to gasoline due to underground storage tanks, allows quantification of potential carcinogenic risk incurred
by exposed populations. Use of the cancer potency factor for gasoline22 results in a more conservative estimate than
the value for benzene since gasoline concentrations in the drinking water are projected to be substantially higher than
benzene concentrations.
A multitude of noncancer health effects have been attributed to ingestion, inhalation, or dermal exposure to gasoline or
its constituents. As there is currently no EPA reference dose for either gasoline or benzene, NESCAUM values will be
used as a default. According to NESCAUM, the most sensitive endpoint for gasoline is kidney toxicity. The most
sensitive endpoints for principal constituents of concern include hematotoxicity associated with benzene exposure and
ncurobehavioral, hematological, and immunological effects associated with toluene exposure. Reproductive and
fetotoxic effects have been associated with exposure to xylene. To protect against these endpoints, reference doses and
subchronic reference doses were established by U.S. EPA and NESCAUM. A reference dose is an estimate of a lifetime
dose which is likely to be without significant risk to human populations. A subchronic reference dose is an estimate of
a dose which is likely to be without significant risk to human populations for an exposure period of up to seven years.
Subchronic reference dose values were used where available because it is considered unlikely that someone would be
exposed to ground water contamination for a period of more than seven years over a 70-year average lifetime. The
EPA23 subchronic reference dose values are 2.0 (mg/kg/day) for toluene and 4.0 (mg/kg/day) for xylene. EPA
subchronic reference dose values are not available for benzene or gasoline so NESCAUM values of 0.1 (mg/kg/day)
for benzene and 0.1 (mg/kg/day) for gasoline were used.
Exposure Assessment
The primary exposure pathways of concern in regard to leaking storage tanks are via contaminated ground water and
soil. Human health exposure may occur through ingestion of contaminated water, inhalation of vapors and dermal
exposure from contaminated water during household use, or through inhalation of vapors penetrating the basement or
foundation walls. (In unusual cases, vapor concentrations may reach levels conducive to fire or explosion.) Vapor
inhalation may also occur outdoors in areas adjacent to a leaking tank, but this is generally viewed as a less significant
exposure pathway. The intake estimates used in this study for gasoline and benzene, however, are based on ingestion
of two liters of water per day only, and are shown in Exhibit W-l.
The population potentially at risk in the Great Lakes Basin is defined for this problem area as those whose primary
source of drinking water is ground water that is not routinely monitored and treated for volatile organic compounds
(VOCs). Routine monitoring and treatment, it is assumed, would prevent ingestion of petroleum-contaminated water
and eliminate any risk. There are two types of systems that fall into this unmonitored and untreated category: (1) ground
water-based public water systems serving populations of 3,300 or less, and (2) private well water systems.
The population potentially at risk constitutes approximately 24 percent of the total Great Lakes Basin population. This
figure was extrapolated in part from the Region 5 comparative risk assessment. It is the percentage of the Region 5
population who use ground-water systems serving 3,300 people or less (20 percent), plus the percentage of the
population who use private wells (10 percent), times an "urban correction factor" of 0.8 to account for the higher
population density in the Great Lakes Basin as compared to Region 5. The higher density in the Great Lakes Basin is
a These {actors for gasoline are based on inhalation data.
"US. EPA, Environmental Criteria and Assessment Office (ORD) and listed in the Health Effects Assessment Summary Table.
-------
V - 38
assumed to result in fewer users of small ground-water systems and private wells than in Region S, and therefore in a
smaller percentage of the population at risk. The correction factor was obtained by dividing the population density of
Region 5 (140 people/mi2) by tbe density of the Great Lakes Basin (180 people/mi2). The percent population at risk (i.e.,
24 percent) was multiplied by the total population in tbe Great Lakes Basin (20,672,600)24, then multiplied by the
fraction of tanks estimated to be leaking (0.2), and then multiplied by tbe fraction of leaking tanks that are known to or
have the potential to coataminate drinking water (033), in order to determine the population exposed. See Exhibit W-1
for the population exposure data and additional details on how they were obtained.
Risk calculations were preformed using drinking water concentrations of gasoline, benzene, toluene and xylene just
below the taste and odor threshold, based on NESCAUM's study.
Exhibit w-i
EXPOSURE DATA
Gasoline
Benzene
Toluene
Xylene
Great Lakes Basin
Concentration In Water*
(mg/1)
5.7
0.46
0.27
0.29
Population at Risk'
4,960,000
Estimated Mean Exposure*
(mg/kg/day)
0.17
0.014
0.008
0.009
Population Exposed*
327,500
' Data taken from the NESCAUM study of health effects. The data is based on ingestion only and on concentrations that are just below the taste and
odor threshold.
k Estimated mean exposure is based on 2 L/day ingestion by a 70-kg adult
* Population at risk « 20,672,600 (the population of the Great Lakes Basin; estimated by multiplying the population of those counties lying within
the Great Lakes Basin by the percentage of that coanty's land that lies within the Gnat Lakes Basin) x 0 J (the fraction of the population who use
private wells or public wells serving less than 3,300 people ia Region 5; see text) x 0.8 (the popnlation density of Region 5 divided by the density
of the Great Lakes Basin - this factor was included because it is assumed that there are fewer private and small public well users in more densely
populated areas).
'Population exposed » 4,960,000 (popnlation at risk) x 0.2 (midpoint of the fraction of tanks nationally estimated to be leaking) x 0.33 (fraction of
leaking tanks that are known to, or have the potential to contaminate drinking water - otherwise knows ss high priority leaking tanks; this number
was determined by Region S using best professional judgment).
Human Health Risk Characterization
The toxicity and exposure information have been used to derive estimated annual cancer cases and noncanoer hazard
v Here, the population of the Great Lakes Basin is estimated by mnltiplying the population of those counties lying within the Basin by
the percentage of that county's laid thst Ues within the Basin.
-------
V - 39
indices for the population at risk in the Great Lakes Basin (see Exhibits W-2 and W-3). Potential cancer cases are
estimated at 20 over a 70-year period, or 0.28 per year. The estimated hazard index for systemic toxicity is 1.7 or
slightly above the reference dose. This hazard index was obtained using a reference dose which is not verified by U.S.
EPA. A hazard index above one raises concern about the possibility of non cancer health effects in some members of
the population at risk, but since there is perhaps an order of magnitude uncertainty in non-canoer evaluations, it is
difficult to quantify the number of individuals who may be effected.
Exhibit W-2
CANCER RISK ASSESSMENT
Estimated Men Exposure' Cancer slope factor Estimated Individual
(mg/kg/day) (mg/kg/day)'' Lifetime Cancer Risk*
Gasoline 0.017 0.0035 6 x 10J
Benzene 0.001 0.029 4 x 10'
Popnlition Exposed Estimated Cancer Cases' Estimated Annual Cancer
(70-yr life) Cases'
Great Lakes Basin 327,500 20 0.28
' Estimated mean exposure from Table 1 has been multiplied by 0.1 to average the snbchronic exposure over a lifetime (7 yean/70 years).
' Individual cancer risk « (mean exposure) (ssbchronic cancer slope factor). As the data indicates, although the cancer slope factor for benzene is
higher than for gasoline, the estimated individaal cancer risk is lower die to less exposure. The gasoline risk value, therefore, is used to estimate
cancer deaths since it is more conservative.
* Estimated cancer cases - (estimated individaal cancer risk from gasoline) (population exposed).
4 Estimated annual cancer cases « estimated cancer cases (70-yr Ufey70 yis.
Exhibit W-3
NONCANCER RISK ASSESSMENT
Exposure
(mg/kg/day)
Snbchronic
Reference Dose
(mg/kg/day)
Hazard Index
ExfVRfD
Population ExnnsoH
Gasoline
Benzene
Toluene
Xylene
0.17
0.014
0.008
0.009
0.1
0.1
2.0
4.0
1.70
0.14
0.004
0.002
327,454
327,454
327,454
327,454
Two recent activities may tend to decrease risks from USTs. First, effective January 1,1991 all systems serving 25
-------
V - 40
people or more are required to sam pie each water supply source, initially on a quarterly basis to determine the presence
and extent of VOC contamination.25 Assuming regulations are being observed by smaller systems and effectively
enforced by the Agency, this requirement will reduce the estimated population at risk (those who are exposed to the
contaminated ground water) by up to two-thirds (two-thirds being the fraction of the population at risk who use public
water systems serving populations of 3,300 or less*). Second, in conjunction with this new law, extensive federal and
state leak prevention efforts, such as new containment and leak detection systems, will lower the incidence of releases
and therefore the risk to human health.
Ecological Risk Assessment
The locations of USTs in the Great Lakes Basin were not readily available for this analysis. Therefore, this analysis does
not involve site-specific information.
Toxicity Assessment
In the event of a release of gasoline or other petroleum product from an UST, a number of scenarios may result.
Depending on the extent of release and proximity to sensitive receptors, a release could result in potentially significant
adverse ecological effects. The extent of damage to an ecosystem would depend on a number of factors including the
nature of the released material, type of habitat, ecosystem stability, and affected species.
Exposure Assessment
The Great Lakes Basin obviously has a significant amount of surface water. Because of the long water retention times
of the individual Great Lakes, most noticeable in Lake Superior and Lake Michigan, pollutants are not quickly flushed
from the system. Leaking USTs, particularly into surface waters, can therefore levy a significant toll on the aquatic
environment. The factors that influence the incidence of surface water contamination from these leaking USTs include:
proximity of tank to body of water; materials leaking from tank; amount of released material; ground water flow; and
local soils and geology. While available data concentrates on effects on human health rather than on the environment,
it is clear that pollutants, specifically petroleum torn leaking USTs, can easily find their way into the aquatic food chain.
Historically, leaks from USTs do not appear to have had an overwhelming impact on surface water, probably due to the
fact that soil and ground water between the leak and the surface water body acts as an attenuator. If such conditions do
exist, however, to allow the flow of released material to surface water, the negative ecological impacts can be extensive.
A national survey of tank releases27 indicates that 22 percent of releases are to surface water. If this information is
applied to the estimated number of releases, 62,700 (estimated by multiplying the number of USTs by the fraction of
releases, or 313,346 times 0.2), then approximately 13,800 releases in the Great Lakes Basin are estimated to
15 EPA, "N»tionil Primary Drinking Water Regulations; Syntketic Organic Chemicals; Monitoring for Unregulated Contaminants," Final
Rale, 52 Federal Register 25690, July 8, 1987.
"Previously, only primaiy water supply systems serving 3,300 customers or less were required to lest for VOC's.
"EPA, 1986, ogclt.
-------
V- 41
contaminate surface water. These releases are expected to be small, however.28
Leaking USTs can also affect ecosystems via soil contamination and by releasing to ground water, both of which can
ultimately contaminate aquifers and surface water. Soil contamination may adversely affect terrestrial organisms at the
lowest level of the food chain. This may expose higher-order organisms to adverse health consequences due to ingestion
of contaminated food sources. Ground-water contamination may result in adverse health effects upon plant and animal
life at the surface through the consumption of contaminated water from springs or wetland areas.
Ecological Risk Characterization
Severity
Ecological damage resulting from UST releases generally can be considered low in severity. Water quality data used
for the Region 5 comparative risk study indicates some contamination, but none specifically linked to benzene (a
constituent of gasoline) or to USTs.
Reversibility of Damage
It is difficult to determine the time frame in which the affected ecosystems will recover from contamination due to
leaking USTs. Oil and gas spills to surface water evaporate quickly (depending on the extent of the release). If
sediments are contaminated the period of recovery will be extended accordingly. Because the Great Lakes are relatively
clear the natural burial of contaminated sediment is slow. Since this is the primary method of sediment removal from
the system, the ecosystem recovery period could be long. Water temperature and turbidity will also affect the rate of
recovery. Ground-water contamination is more complex, and it is difficult to determine the rate of recovery. If
contaminated, ground water may affect the whole water supply, significant and irreversible damage to plant and animal
life could result. Overall, the reversibility of damage from USTs is believed to be medium.
Welfare Damage Assessment
Health Care Costs
In order to estimate health care costs (HCC), the annual cancer cases estimated in the human health risk characterization
section were multiplied by the direct medical cost and foregone earnings per cancer case used in the Region 5
comparative risk assessment. Cancer costs were based the Hartunian ct al., lower-bound estimate, ($80,000,1988 S)
and the American Cancer Society, upper-bound estimate ($137,000, 1988 $) discussed in previous sections.
Lower-Bound Estimate:
HCC = (0.28) ($80,000) = $22,400 annually
Upper-Bound Estimate:
HCC = (0.28) ($137,000) = $38,400 annually
Environmental Costs
" EPA, OSWER Comwrative Risk Study: Ecological Risk Chincterization Methodology. OSW, April 1988.
-------
V - 42
Remediation Costs
Remediation costs associated with leaking USTs were estimated using the following methodology taken from the Region
5 comparative risk study, which assumes a remediation cost of $150,000 per corrective action. The number of corrective
actions was determined by multiplying 313,346 (the total number of tanks) by 0.2 (midpoint of the fraction of tanks
nationally estimated to be leaking) and by 0.33 (fraction of leaking tanks that are known to, or have the potential to
contaminate drinking water). The result is approximately $3 billion.
Slightly different assumptions were used in the Region 5 comparative risk assessment to determine the remediation costs
associated with leaking USTs; these assumptions resulted in a lower estimate of costs. The following formula was used
to apply the Region 5 methodology to the Great Lakes Basin: cumulative cost of corrective actions = 313346 (number
of tanks in the Great Lakes Basin) x 0.026 (fraction of tanks in Region 5 with confirmed releases) x $150,000. The total
cost comes to approximately Si.2 billion.
Replacement Of Contaminated Drinking Water Supplies
Although the analysis of remediation costs should, in effect, account for well replacements and water supply hook ups,
the Region 5 comparative risk assessment places these activities in a separate category. We have used this same
distinction and applied the Region 5 methodology to the Great Lakes Basin.
Given the number of private well replacements and public water supply book ups associated with leaking USTs, the cost
of replacing contaminated drinking water supplies can be estimated following the method used in the Region 5 risk
assessment. This method assumes that the capital costs are $3,500 for replacing a private well by digging a new well,
and $300,000 for replacing a public supply well by extending a hook up from another public supply well. These were
upper-bound cost estimates, yet they do not include the annual operating costs for these private and public systems.
Because data describing the number of private well replacements and public water supply hook ups were not available,
best estimates were used. It was assumed that 10 percent of releases to ground water (assuming 20 percent of tanks leak,
and 33 percent of these potentially contaminate drinking water) would result in private well replacements, and 2 percent
of releases would result in public well replacements. Hie reason there are fewer replacements for public wells is due
to the larger size of the aquifer, and the more comprehensive treatment and monitoring that is associated with a public
water supply system as opposed to a private one. The cumulative costs associated with these replacements are $7.2
million and $124 million, respectively, for a total of $131.4 million. This does not correspond closely with an estimate
of ground-water valuation (using replacement costs as a proxy) derived from an OSWER comparative risk study.29
That study estimated an average of $10,700would be spent for every UST, and when multiplied by the number of USTs
in the Great Lakes Basin, 313,346, the total cost is approximately $3.4 billion. One of the remediation cost estimates
above (i.e., $3 billion) is much closer to this amount.
" ICF Incorporated, OSWER Compnrative Risk Project: Ground-Water Valuation Task Force Report (PraflV prepared for EPA, Office
of Underground Storage Tanks, February 4, 1988.
-------
V- 43
X. PESTICIDES: HUMAN HEALTH
Human Health Impacts
The magnitude of the Pesticides problem is characterized above, under Problem Area L, Pesticides Discharges and
Environmental Risk. To characterize the human health impacts posed by pesticide applications in the Great Lakes Basin,
we have adopted the analytical methodology presented in the Pesticides section of the Region 5 comparative risk study.
Pesticide risks to human health occur through various pathways, including
direct contact;
inhalation of short and long range drift;
runoff into surface water sources including drinking water sources;
leaching into ground-water drinking water supplies; and
residues on food and bioaccumulatioa.
A number of subgroups of the people residing in the Basin are at risk from pesticide exposure. The population most at
risk is pesticide applicators, particularly farm workers who mix, load, and apply pesticides. People residing in
agricultural areas are at risk through pesticide contaminated drinking water as well as short range drift. Employees at
pesticide manufacturing and storage facilities are at risk from direct contact with pesticides through accidents and spills
as well as short range drift. Likewise, persons residing near such facilities are also at risk. Finally, the general public
is also at risk from exposure particularly through pesticide residues on food.
Toxicity Characterization
Each pesticide was examined for evidence of human toxicity, cancer and noncancer, and for ecological effects. For
cancer risk, both risks from food residues and to applicators were evaluated according to the 1987 "Unfinished Business"
Report For noncancer risk from pesticide residues in food, data did not permit a separate analysis. Where no
carcinogen category has been assigned, the Reference Dose for noncancer health effects was used to analyze risk for
applicators, along with available exposure assessments from EPA documents, such as Special Review Position
Documents, and Registration Standards (now called Reregistration Documents).
Exhibit X-l presents the toxicity factors available for 7 pesticides used in the Basin. With over 600 active ingredients
registered by the U.S. EPA, it was not possible to analyze all pesticides for the Great Lakes Basin. Toxicity factors are
given for those pesticides which had appropriate information available.
Exposure Characterization
As discussed earlier, humans are exposed to pesticides through a variety of pathways, including during application-
related tasks, through drift, runoff, leaching into ground-water supplies, residues in food, and other exposure routes.
To estimate the risk from each route of exposure, we must first determine the population exposed.
Since 12 percent of the U.S. population (30 million people) resides in the Great Lakes Basin, this is the number assumed
to be exposed to pesticide residues in food. The level of exposure is assumed to be minimal since most residues are
negligible in foods consumed after preparation (washing, peeling, and cooking).
The next major population group exposed to risks from pesticides that is considered here is certified applicators.
Certified applicators include (1) private applicators, primarily farmers, and (2) commercial applicators, which are
-------
V - 44
categorized by agricultural pests, ornamental and turf pests, structural pests, forest pests, and rights-of-way maintenance,
among others. There are approximately 100,500 and 29,200 private and commercial pesticide applicators in the Great
Lakes Basin in 1989. These figures were derived from the number of certified applicators in Region 5 using the
proportion of people in the Basin to the number of people residing in Region 5 (65 percent).
The population in the Basin estimated to be exposed to widely used chemicals, such as homeowner use of lawn care and
garden pesticides, will be the total number of people residing in the Basin (30 million people).
Risk Characterization
To analyze human health risk from consumption of pesticide residues in food and from pesticide use in the Great Lakes
Basin, the highest volume pesticides were examined for evidence of cancer and noncancer effects in animal testing, as
described in the Toxicity Assessment above.
Exhibit X-l
Toxicity Factors for Selected
Pesticides Used in the Great Lakes Basin
Pesticide Reference Dose
(ppm/day)
Atrazine
0.005
Alachlor
0.01
Metolachlor
0.15
Cyanazine
0.002
Trifluralin
0.003
Carbofuran
0.005
2,4-D
0.01
Pesticide Residues in Food: Cancer Risk
Ibe "Unfinished Business: Report (1987) considered 7 oncogenic pesticides and determined the number of cancer cases
in the U.S. population making several assumptions. For the purposes of estimating cancer risk in the Basin, a proportion
of the national cancers was taken using the percentage of the U.S. population residing in the Basin (12 percent). The
Office of Pesticide Programs (OPP) estimated that the total annual population risk from dietary exposure to oncogenic
pesticides was 6000 people/year. This means that for the Great Lakes Basin there would be an estimated population risk
of720 people/year.
Another issue of concern is the cons urn ption of fish contaminated with such suspended and canceled pesticides as aldrin,
dieldrin, chlordane, heptachlor, DDT, and mirex. Several studies of fish consumption have been undertaken in the Basin.
For example, a recent study by the Michigan Department of Health found that average fish consumption is 16.1 grams
per day for a sample of 2600 holders of sport fishing licenses. Of the 16.1 grams per day consumed, 56 percent or 9
grams per day represents commercially purchased fish and 44 percent or 7 grams per day represents locally caught sport
-------
V - 45
fish. The population eating sportfish in Michigan was estimated to be about 2.5 million people.
Pesticide Residues in Food: Noncancer Risk
Available data did not permit separate analysis of the noncancer risk in the Basin from residues in food. The assessment
in "Unfinished Business" and that provided by OPPE on July 26,1990, analyzed dietary risk using only pesticides with
oncogenic risk numbers.
Risks to Pesticide Applicators: Cancer Risk
As stated above, the "Unfinished Business" Report used extrapolations from long term animal studies to estimate human
risk on a nationwide basis. Since oncogenicity studies are based on lifetime daily oral exposure, several adjustment were
made, as follows:
Yearly exposure is used for risk calculation and average daily exposure is calculated by dividing
yearly exposure by 365.
Workers are exposed for 40 years of a lifetime of 70 years.
Dermal and inhalation absorption versus oral absorption, if known, was factored in.
OPP estimated the average lifetime population risk to be 35 persons per lifetime per chemical. The yearly risk per
chemical would then be 0.5 persons per year per chemical (35/70 years in a lifetime). Since approximately 200
pesticides were estimated to be oncogens, the yearly risk was estimated to be 0.5 X 200 or 100 persons per year.
To determine the number of certified applicators in the Basin we used the proportion of the Basin population to the
Region 5 population (30,050,000/46,428,000 or 65 percent) multiplied by the number of certified applicators in Region
5 (200,000) to arrive at an estimated 130,000 applicators in the Basin (200,000 X 0.65). We estimated the number of
applicators in the Basin at risk by multiplying the population risk of 100 persons per year for the nation by the percent
of national applicators that practice in the Basin (130,000/1,249,000 or 10 percent) to arrive at 10 Basin applicators at
risk per year (100 X 0.10).
Risk to Pesticide Applicators: Noncancer risk
The reference dose was used to calculate risk to applicators for pesticides without quantitative cancer risk assessments.
We used available EPA documents, which included Special Review Position Documents, and Reregistration Documents
(formerly called Registration Standards) to obtain exposure assessments for pesticides used in the Basin. Exposure
populations were estimated from numbers of certified applicators in categories known to use particular pesticides and
other available information. Using methods prescribed by the Comparative Risk Technical Steering Committee for
Region 5, final risk evaluations were made. The two pesticides used in the Basin were in the medium to low category,
as follows:
Pesticide Final Risk Score
Cyanazine
2,4-D
Medium-Low
Medium-Low
-------
V - 46
Other pesticides were not included because either they did not appear to be used in the Basin as evidenced by their
failure to appear in the RFF data base, or because applicator exposure information was not available. Most risks were
medium-low using this method of analysis. The analysis assumes that protective clothing is worn during application
and applicators exercise appropriate precautions when applying pesticides.
OPPE provided a summary of non-dietary risks from pesticide use in their July 26,1990report. Six pesticides were used
to calculate noncancer risks to applicators and farm workers. The Temple, Barker, and Sloane report was used to obtain
rough estimates of the population of farm workers and applicators for Region 5 which were then scaled down using a
Basin percentage figure. These estimates are as follows:
Total farm workers: 13 million (includes unpaid workers, hired workers, farm operators, and ground applicators;
migrant workers were not added).
Total off-site workers: 350,000 (includes all hired workers and commercial ground applicators).
With higher numbers of people exposed, the overall risk ranking for this category in Region 5 was medium-high. Given
the lower numbers of people exposed in the Basin as compared to Region 5, we assume the overall ranking to be medium
to medium-high.
Other Impacts on Human Health from Pesticides in the Basin
Several other potential impacts to human health exist in the Basin from such sources as the use of pesticides in the lawn
care, turf, and golf course industries, the use of insect repellents such as DEET due to the concern about the spread of
the Deer tick and Lyme disease within the Basin, the impacts of having approximately 2600 pesticide producing and
custom blending establishments in the Basin, and non-occupational exposure to pesticides. These analyses are
qualitative and are described below.
Lawn Care and Turf Industries
Use of pesticides in the lawn care and turf industries, has been estimated by EPA to be 67 million pounds of active
ingredient per year. Using a population proportion for the Basin, this amount scales down to approximately 8 million
pounds of active ingredient per year applied to lawns by professional lawn care operators and homeowners in the Basin.
Using a population proportion for the Basin again, there are approximately 11,200 certified applicators in the
turf/omamental category in the Basin who apply pesticides for about 3-4 hours per day during the late Spring to early
Fall season. Estimates by industry technical representatives indicate the solutions applied are rather dilute, with less than
1 percent active ingredient with 6-7 percent fertilizer. The major pesticides used in the industry are 2,4-D, MCPP,
pcndimcthalin, diazinon, chlorpyrifos, dicamba, and isofenphos. The RFF data base shows use of some but not all of
these pesticides in the Basin. One company in Region 5 indicated that no pesticides requiring respirators were used in
their operations. In addition, the company requires protective clothing to be worn when hauling concentrate material
and when filling tanks. During application, boots and clean uniforms are worn and gloves are encouraged. Over the
next few years, as the Agency looks into the lawn care industry and evaluates the risks from this use of pesticides, more
information on exposure will become available and this can be used to make a quantitative evaluation of risk.
Golf Courses
Extrapolating from Region S data, there are approximately 2.4 million acres of golf courses in the Basin and are treated
-------
V- 47
with 2.2 million pounds of pesticides annually 30 In late 1986, EPA called for the withdrawal of diazinon on golf
courses and sod farms because of cases of severe allergic reactions to some golfers and impacts to birds. Approximately
399,900 pounds of diazinon were applied in the Basin in one year in the mid 1980s. Starting with the 1990 season,
Wisconsin is requiring that all commercial applicators keep records of all pesticides used, including restricted use and
general use pesticides. Most golf course superintendents are interested in gathering information which reflects the true
practices in their industry. In the future, more information will develop to use in the quantitative evaluation of risks from
the use of pesticides on golf courses.
Insect Repellents/DEET
Use of insect repellents including the active ingredient DEET has increased greatly over the last few years. USA Today
found that sales of "OFF" insect repellent had increased by 50 percent during the summer of 1989. With the high degree
of outdoor recreational opportunities in the Basin, it can be reasonably estimated that most of the Basin population at
one time or another has used insect repellents. With increasing concern over Lyme disease and the spread of the Deer
Tick, we expect the use of repellents will most likely increase. In Michigan, the Centers for Disease Control reported
an incident rate for Lyme disease of 41 cases per 100,000 people between 1987 and 1988. DEET, an especially effective
active ingredient in repellents, is currently under reassessment, and when more information is known on the toxicity and
exposure associated with its use, the Agency will provide this information to States and the general public in order to
reduce the risks even further. In the interim, EPA recommends using repellents with no higher than IS percent DEET
for children and infants, and to apply such products sparingly to the outside of clothing.
Pesticide Producing and Custom Blending Facilities
Extrapolating from Region 5 data, there are approximately 2600 pesticide producing and custom blending facilities
located in the Great Lakes Basin. In preparing the Region 5 comparative risk report on pesticides, an attempt was made
to look at the activities associated with the synthesis of pesticides, the combination of active ingredients with inert
ingredients and the risks associated with these activities. A FIFRA/TSCA Tracking System report was obtained with
pesticide producing establishments and products made at each location. The Toxic Release Inventory System was also
accessed to look for use of toxic chemicals at locations where pesticides are synthesized and formulated. About 25-30
locations in the Region were found to have reported threshold quantities of toxic chemicals and to be associated with
the production of agricultural chemicals, as determined by Standard Industrial Codes. The search provided only a
general idea of association between these activities and exposure of risks and no quantitative risk assessment was
passible. In the future, as reporting becomes more sophisticated, such an assessment may become more possible.
Non-Agricultural Pesticide Use
The Non-Occupational Pesticide Exposure Study (NOPES) considered the human exposure to 32 household pesticides.
The objective of the NOPES was to estimate the levels of non-occupational exposure to selected household pesticides,
primarily through indoor air, but also through food, drinking water, and dermal contact. The study was conducted in
two locations, (1) Jacksonville, Florida, and (2) Springfield and Chioopee, Massachusetts in order to consider areas of,
respectively, high and low to moderate non-agricultural pesticide use. The Basin could be considered in the low to
moderate non-agricultural pesticide use category.
Results of the NOPES study showed a number of pesticides, some already canceled, were detected at least once in
minute quantities in the majority of households including chlordane, beptachlor, chlorpyrifos, orthophenylphenol, and
x Water Quality Board, Intenatiooal Jo ill Commisaioa. 1987. "1987 Report oo Great Lakes Water Quality," p. 183.
-------
V - 48
propoxur. Other pesticides detected include aldrin and diazinon.
Other conclusions of the NOPES include:
diet was a more important source than air for exposures to 14 of the 32 household pesticides;
airborne concentrations of pesticide residues were higher indoors than outdoors;
there were seasonal variations in ambient residue levels;
short term variations in residue levels were influenced by recent pesticide applications, indoor
ventilation, and ambient temperatures;
usage categories did not correlate with airborne residue concentrations; and
residue levels could be reduced if labels are read closely and followed, and with improved guidance
on how to safely dispose of unused pesticides.
EPA concluded that the concern for human health risk from exposure to pesticides found in this study is low to
negligible.
-------
APPENDICES
APPENDIX A: DATA SOURCES, UNCERTAINTIES AND GAPS
APPENDIX B: BIBLIOGRAPHY
APPENDIX C: ACRONYMS
APPENDIX D: ANNEX I LIST OF GREAT LAKES SUBSTANCES
APPENDIX E: DESCRIPTIONS OF U.S. GREAT LAKES AREAS OF CONCERN
APPENDIX F: NUCLEAR RELEASES
APPENDIX G: MERCURY TRENDS
-------
Page VI - 2
APPENDIX A; DATA SOURCES. UNCERTAINTIES AND GAPS
Physical Degradation of Water Habitat and Wetlands
Sources .-The primary sources of information drawn upon include the 1990 water quality reports prepared by each
of the Great Lakes states, and Natural Heritage Inventory data for the Basin counties which was supplied by the
midwest regional office of The Nature Conservancy.
Uncertainties and Gaps:Thcrc is no hard data that can be used to accurately quantify the physical degradation of open
water and wetlands habitat. Data provided by the Nature Conservancy only identifies the examples of wetlands that
are biotically diverse, sensitive, or that hold a rare or endangered plant or animal species.
No data could be used to calculate risks posed to water and wetlands habitats. More specifically, no quantitative
data was found addressing the residual risks posed to human health due to wetlands loss, the environmental hazards
and toxicity of pollutants found in wetlands, or human and monetary losses due to flooding.
Physical Degradation of Terrestrial Ecosystems
Sources :This section employed tracking data of biologically significant sites in Basin counties from the National
Heritage Program, and information from the Region 5 Problem Area Paper "Physical Degradation of Terrestrial
Ecosystems."
Uncertainties and Gaps: The Heritage data only identifies examples of biotically diverse or sensitive areas, or those
that bold rare or endangered species. It does not address more common ecosystems, or the degree of impact on those
ecosystems from anthropogenic stressors. Quantitative data oo these impacts is unavailable.
Exotic Species Introduction to the Great Lakes
Sources: The primary sources of data were articles from academic journals, agency reports, and conversations
with agency experts. The U.S. Fish and Wildlife Service supplied a bibliography documenting journal articles on
a number of different exotic species. Itae Great Lakes Fishery Commission supplied articles and reports on the sea
lamprey and alewife as well as other exotic species. The Great Lakes Commission supplied newspaper articles and
reports on the zebra mussel. The U.S. Fish and Wildlife Service provided summary information on Ruffe. Taken
together, these sources provide a general characterization of the problems in the Great Lakes due to the presence of
exotic species.
Uncertainties and Gans: Data is largely anecdotal Quantitative data is unavailable on the problems or their
impacts. Population figures for the exotic species are estimates whose statistical validity is not known.
Welfare impacts are gross estimates. They have been provided to give a range of costs that officials have ascribed
to the problem.
Data on the ecological and welfare impacts of the spiny water flea, asian clam, alewife and the smelt were
unavailable for this draft.
-------
Page VI - 3
Spills and Other Accidental Releases
Sources: 1969 ERNS data on Basin counties was used as well as United States and ranflrijnq Coast Guard
reports and data. The Canadian Coast Guard commissioned a report entitled Counter Measures for Marine Spills
of Hazardous Materials that looked at the amount of hazardous materials that travel on the Great Lakes, and
calculates the probability of spills in specific areas of the Great Lakes. This section draws heavily on data from that
report
Uncertainties and Gaps: There are several nuclear power plants within the Great Lakes Basin. These risks are of
low probability but high catastrophic impact. Due to time constraints, risks of accidental releases from these sources
were not included.
No information was available oa exposures to accidental releases. No information was used on ecological impacts
in the Basin. US Fish and Wildlife as well as State natural resource trustees have identified ecologically sensitive
areas and information that could be used. Some data on general ecological effects could also be extrapolated to
identify some risk to ecological impact A correlation might be made between the areas of greatest ecological
sensitivity and the possibility of a spill in those areas.
Differing jurisdictions and inconsistencies in the data reported mean inaccurate information on spills in the Great
Lakes. There is no precise spill inventory far the Great Lakes Basin.
Global Climate Change
Sources: The information presented in this section is primarily based on the EPA Office of Policy, Planning,
and Evaluation's report entitled "The Potential Effects of Global riimate Change on the United States." In this
study, regional outputs, including one for the Great Lakes Basin, were used from three General Circulation Models.
All three estimate climate change caused by a doubling of carbon dioxide in the atmosphere. Expert judgment by
the report's authors regarding potential effects was used to supplement the The results were not
predictions, but rather indications of the impacts that could occur as a result of anthropogenic climate change. The
analytic approaches described above were used as tools to determine the potential sensitivities and vulnerabilities of
systems such as the Great Lakes to climate change.
Uncertainties and Gaps: The precise impact of global climate change in the Great t aimt Pasiq is impossible to
determine due to the vast uncertainties involved in this emerging area of investigation. The information presented
above represents the current best professional judgment concerning likely impacts of such change in the Fn^n. As
modeling resolution and data indicating the trends improve, these estimates will be refined.
Changing Lake Levels
Sources: This problem area used information from the IJC's 1985 report, "Living with the Lakes:
Challenges and Opportunities," and the Great Lakes Commission 1986 report, "Water Level Pinnges, Factors
influencing the Great Lakes."
Uncertainties and Gaps: This analysis is primarily qualitative in nature, as sufficient data are not yet available to
quantify specific impacts. Where available, data are presented to assess overall losses in terms of recreational and
infrastructure impacts. Quantifiable data for the changing lake levels problem area was unavailable for:
¦ the extent of property damage from severe weather and flooding,
¦ the ecaoomic losses incurred due to property damage,
-------
Page VI - 4
¦ (be extent of wetland damage from severe weather including drought and flooding,
¦ human health impacts and costs.
Industrial Point Source Discharges to Surface Waters
Sources : The characteristics of the industrial dischargers in the Great Lakes Basin counties were drawn from
EPA Permit Compliance System (PCS) data1. Ambient surface water quality information for the Basin was taken
from the 1989 305(b) Reports repotted by the Great Lakes States. Ibis ambient surface water information and toxic
constituent concentrations in fish tissue were supplemented by the 1987 and 1989IJC Water Quality Board Reports1
and the Upper Great Lakes Onn^ing Channels Study.3 Fish consumption data and exposure information were
taken from U.S. Department of Agriculture studies and EPA guidance.4
Uncertainties and Gans: The level of toxic or conventional pollutant loading to the Basin from industrial point
sources can only be estimated. The apportionment of load to industrial point sources in this study was made based
on limited *afa from two of the r*hannr.is in the Basin. As those data demonstrate, the relative
importance of industrial, municipal, and nonpoint source load varies considerably across the Basin.
PCS data does not include data on the many minor dischargers in the Basin.
Precise data on fish consumption and human exposure to contaminants in Basin fish is currently unavailable.
Municipal Point Source Discharges to Surface Waters
Sources and Uncertainties: See Industrial Point Source Discharges to Surface Waters.
Nonpoint Source Loading to Surface Water
Sources: The major data sources investigated for this analysis included 1989 Section 319 State Nonpoint
Source Assessment Reports, Conservation Tillage Information Center data, the Upper Great Lakes Connecting
Channels Study, and the 1989 DC Report on Great Lakes Water Quality. Studies conducted at Heidelberg College,
Ohio, were also reviewed for information on nonpoint source pollution in the Lake Erie basin.5
Uncertainties and Gaps: There is a great deal of uncertainty cooceming the total load of nonpoint source
contaminants to the Great Lakes Basin. Unlike point sources, which can be measured at the end of the pipe,
nonpoint source load can only be estimated based on land use practices and observed contaminant levels in the
surface water environment. As a result, this analysis relied on the apportionment of contaminant source load
1 Note: PCS doe* >ot tack all nil or diackaigen ia Ike Baaia.
1 Great Lake* Water Qaallty Board. 1987. *1987 Report oa Gnat Lake* Water Qmallty," Report to Ike Ialermatioial Joiit Commit*!oa;
Great Lake* Water Qaallty Board. 1989. "1989 Report oa Gnat Lake* Water Qaality," Report to tke Iatermatioaal Mat Commiuioa.
' US. EPA, et »l. 1988. "Upper Great Lake* Coiaectiag Ckaaael* Stady."
4 US. Depaitmeat of Agricaltare. 1989. "Coiaimptioi aid Family Liviag," Aaricaltatal Statistic*: U.S. EPA/OWRS. 1989. "Aaieatlag
Himaa Healtk Risks boat CkemtcaUy Coaumliated Fiak aad Skellflak: A Galdaace Maaaal," BPA-503/8-89-002.
5 Baker, D3., "Overview of Rani Noapoiat Pollatioa ia ike Lake Erie Baaia," Lopa, Terry J., et aL, edk, Effect* of Coaaervadoa
Tillage oa Groaadwiter Quality: Nitrate* aid Pe*tfclde*. chapter 4, Lewi* PiMiaker*, lac., Ckebea, Ml, 1987.
-------
Page VI - 5
described in Industrial Point Source Discharges to Surface Waters. Additional data to better characterize the true
nature of nonpoint source load in the Basin would improve the risk assessment.
Atmospheric Loading to Surface Water
Sources: Sources of data for this section include: The Great Lakes Atmospheric Deposition Network, 1982
and 1983, GLNPO, EPA; IJC's 1987 "Summary Report of the Workshop on Great Lakes Atmospheric Deposition;"
and the IJC's 1969 Report on Great Lakes Water Quality.
Uncertainties and Gaps: Evaluation of this problem area involves a number of assumptions necessary to overcome
the high uncertainties associated with the available data and the simplified analytical methodology applied here.
Major sources of uncertainty include:
¦ the relative role of atmospheric deposition to levels of all of the contaminants of concern in each of the
Great Lakes,
¦ the geographic distribution of fish contamination,
¦ the effect of fishing advisories on human health risks from fish consumption,
¦ the effect of water treatment on human health risks from drinking water,
¦ the severity of ecological effects attributable to atmospheric deposition of toxic contaminants into the Lakes,
and
¦ the degree to which the Lakes are sources of PCBs to the atmosphere.
Toxic Sediments
At this time there are major data gaps over the amounts, pathways, and risks from contaminated sediments. The U.S.
EPA's Great Lakes National Program Office (GLNPO) is responsible for a multi-agency program entitled Assessment
and Remediation of Contaminated Sediments (ARCS), scheduled for completion in 1992. While the results of the
ARCS Program should provide clarification as to the extent of the contaminated sediment problem, not much data
is currently available. Therefore, the sediment analysis summarizes the problem qualitatively untiring information
available from sources such as the IJC Reports on Great Lakes Water Quality, GLNPO, New York State 305(b)
report, Great Lakes: Great Legacy? and others.
Pesticides Discharge*
Sources: All data on amounts and rates of active ingredients applied to agricultural lands in the Basin were
obtained from Resources for the Future (RFF). RFF data are presented in two separate data bases:
¦ RFF published data for 21 pesticides used in the Basin for a typical year in the mid 1980s. These data are
digitized and are available by county for 181 counties in the
¦ RFF is also currently updating its National Pesticide Usage Data Base which includes data oo application
rates and total agricultural acres treated with 35 active ingredients for a typical year in the late 1960s.
These data are also automated and are available by county for 213 counties in the Basin.
This discussion uses these more recent data for the 7 active ingredients applied in the Basin that are also in the
earlier, Basin-exclusive data base. RFF has not updated the more recent data base for the 21 active ingredients used
in the Basin.
-------
Page VI - 6
This section also uses data from the 1988 U.S. EPA, GLNPO Report oo Agricultural Tillage Practices in the Basin,
and the 1990 Region 5 comparative risk project. Finally, this section incorporated summary data from the IJC's 1987
Report on Great Lakes Water Quality.
Uncertainties and Gaps: There are several caveats to the use estimates of the RFF pesticide data. These caveats
are:
¦ Not all pesticides used in the Basin were included;
¦ Non-cropland uses of pesticides were not included;
¦ Two different data sets were used, as noted, and are for a typical year. It is not known to what extent these
estimates reflect current use patterns;
¦ The estimates assume uniformity for all counties within states for the same pesticide/crop use.
¦ It is not known to what extent the actual use pattern varies within each state.
¦ The accuracy of state-level use estimates cannot be assessed. Many of the use estimates represent the
opinions of weed scientists and other extension personnel. There is no independent set of estimates to verify
these estimates; and
¦ Hie older and more recent data sets are taken for 181 and 213 counties, respectively, in the Basin. This
difference will result in differences in the total amount of pesticides applied. Any conclusion based on this
data will have to be carefully drawn.
Risk estimates were made by extrapolating data generated from Region 5. It is not known to what extent these
derived risk estimates are statistically valid.
Sulfur and Nitrogen Oxide Emissions
Sources: For this section, the AIRS Facility Subsystem (1990 retrieval) provided data for the Rnsin counties
on total, permitted, and fugitive emissions as well as data for ambient levels.
Uncertainties and Gans: The Region 5 comparative risk study applied the same method to estimate health impacts
from sulfur and nitrogen oxide pollution as was used in this study. The Region S paper points out three important
limitations of this approach. First, the quantitative relationship between estimated health effects and sulfate exposure
is based on epidemiological study results that demonstrate association but not causation. Therefore, Region 5 pninfc
out that some uncertainty exists regarding the magnitude, and in some cases even the existence, of the presumed
causal relationship presented between the pollutants and human health.
Second, although sulfate aerosols are the best available indicator for acid aerosol exposure, acidic sulfate aerosols
would be a preferred, and mote valid indicator.
Finally, estimates of the number of people exposed to various levels of sulfate are quite uncertain. Problems
associated with determining the exposed population include Basing judgments on limited monitoring data, estimating
the area that each monitor represents, and estimating the number of people in this area who arc actually exposed to
the pollutant.
Uncertainties related to the ecological impacts section are most significant with regard to terrestrial effects. The
National Acid Precipitation Assessment Program study of lake acidification provides a good lower bound estimate
of the acreage of acidified lakes. Studies of fish populations in aririifVri and non-acidified lakes in the Adirondacks
provide information on the impacts of lake acidification. Consequently, there is a low degree of uncertainty in the
effects of acid deposition on aquatic ecosystems.
In contrast, the terrestrial impacts of acid deposition must be considered to be hypothetical. There is a high degree
of uncertainty regarding effects of acid deposition on terrestrial ecosystems.
-------
Page VI - 7
Hazardous/Toxic Air Pollutants
Sources: This section and the airborne lead problem rely oc the Region 5 comparative risk study. Data to
estimate cancer risks was taken from EPA Office of Air Quality, Planning and Standards report: "Cancer Risk from
Outdoor Exposure to Air Toxics," September, 1990. National data from the EPA Air Toxic Exposure and Risk
Information System data base was apportioned to the Great Lake Basin by population.
Uncertainties and Gaps: The cancer risk values used in this study are based on many assumptions and are therefore
somewhat uncertain. Further, the fraction of the total risk attributable to pollutants and source categories not covered
in the study is unknown. Nevertheless, the OAQPS study is valuable as a reasonable indication of the magnitude
of potential canocr risk caused by this specific group of pollutants and is therefore, in general, considered to be
moderately to highly certain.
For noncancer risk, data was insufficient to predict ambient concentrations of most air pollutants. In addition, data
pertaining to various noncancer endpoints for many pollutants was minimal. Further, inhalation studies are scarce.
The degree of certainty associated with this study is considered to be low to moderate.
It should be noted that the monitoring network for lead is somewhat limited. It is likely that we underestimated the
number of people potentially at risk from exposure to airborne lead. In addition, the current NAAQS for lead was
promulgated in 1978 to protect children from blood levels of 30 micrograms per deciliter (ug/dl). Since that time,
new information has become available which has prompted the Clean Air Scientific Advisory Committee to set the
new maximum safe blood lead level at 10 ug/dL Because of this, using the present NAAQS as the only basis for
evaluating health impacts is likely to result in significant underestimation of risk.
The analysis of health cost for cancer cases assumes that economic damages with cancer cases can be
adequately described without taking into account the pain and suffering caused by these cancers. This assumption
leads to an underestimate of the actual economic damages.
Active RCRA Hazardous Waste Management Facilities
Sources: The Hazardous Waste Data Management System (HWDMS), September 1990, was used to obtain
the number of operating land disposal units (LDUs), storage and treatment units (STUs), and incinerators in the Great
Lakes Basin. Information from the Database on Marketers and Burners of Harartiniiy Waste Fuel and from the
Corrective Action Reporting System was also used.
Uncertainties and Gaps: No Basin data was available on accidents resulting in injuries or at RCRA sites, risks
from recycling units, ecological risks associated with facilities, potential toxic loadings from RCRA sites to the Great
Lakes, air exposures to emissions from operating units other than incinerators and burners, or size distribution of
corrective action sites.
Abandoned Hazardous Waste Sites
Sources: This problem area used data from a 1990 CERCLIS retrieval for the Basin counties.
Uncertainties and Gaps: The health risk analysis was believed to be a reasonable estimate of the risks from
abandoned waste sites. However, some significant uncertainties remain:
First, the risks presented in the endangerment assessments for the NFL sites used in the Region 5 Comparative Risk
Project are the risks at the sites without any remedial actions. Presumably, a significant number of the sites that pose
a significant risk to human health, if not all of these sites, will be undergoing some degree of cleanup either by the
federal government, the state, or private parties.
-------
Page VI - 8
Second, Region 5's data on populations exposed were highly uncertain. Only a few endangerment a&
-------
Page VI - 9
¦ To collect basic data on each type of waste management unit;
¦ To identify which establishments managed industrial Subtitle D wastes using methods other than the four
land-based waste management units.
The second source is a 1988 EPA report entitled Industrial Subtitle D Risk Screening Analysis" (Screening
Analysis). The objective of this study was to identify industrial waste management scenarios that may pose
significant human health and environmental risks. This study was also conducted on a national basis, so the national
data was weighted in terms of Great Lakes Basin population. Most of these results will appear in the human health
risk section of this problem area.
Uncertainties and Gaps: For this study, the national data were weighted in terms of Basin population. The figures
obtained may not accurately represent the actual industrial solid waste data for the Basin due to the following
reasons:
¦ The Great Lakes Basin contains a substantial amount of industry. Consequently, weighting of the national
data may not be representative of the Basin since it may underestimate the proportion of industry.
¦ The data from the Screening Survey was collected in 198S. During the past five yeans, the Basin has had
a significant population decline. The affect of this change further adds to uncertainty in the analysis.
Additionally, 198S industrial solid waste site data was weighted against the 1988 population data, rather than
the 198S population data. This would distort the human risk analysis because the population at risk would
appear to be lower.
Readily available data on the population of industrial solid waste facilities are limited, and, when available, are often
aggregated at the national leveL While states are in a better position to generate data on specific facilities, this
information is not easily accessible.
Although the toxic constituents in various types of industrial wastes are known and can be evaluated, the data needed
to translate known potential health effects of these toxic constituents into risks posed to populations or ecological
systems are currently unavailable. Basic data needs include the concentrations of toxic constituents in the solid
wastes managed at industrial facilities, estimates of the potential for uncontrolled releases of contaminants from these
facilities, and information on the potentially exposed populations residing near these facilities.
Aggregated Drinking Water
Sources: 1990 FRDS data for public water systems, their sources, the number of public systems with at least
one MCL violation, the specific contaminants associated with the violations by county, and the populations
affected was collected and analyzed for this problem ana.'
Uncertainties and Gaps: Because we followed the general approach used in the Region 5 comparative risk study,
many of the uncertainties realized in the Region 5 study were implicit in this analysis. The most important of these
is the use of the FRDS database as the primary data source. First, there is great potential for inaccurate data in the
system, particularly with regard to populations and chemical concentrations. Second, because FRDS only contains
violations and associated concentrations, these concentrations may represent worst-case exposures instead of average
exposures. Third, because FRDS classifies all systems that receive unfinished water from both ground and surface
waters as surface-water based systems, a tally of the number of ground-water based systems can potentially be
grossly underestimated; conversely, a tally of the number of surface-water based systems may be an overestimate.
' FirUer aulytU of FRDS retrieval will be Ucorponled iato »ib»cq«e«t dnftL
-------
Page VI - 10
Aggregated Ground Water
Sources: As described in Aggregated Drinking Water, EPA planned to select representative drinking water
contaminants by reviewing quantitative information from its 1990 FRDS data base for public drinking water supplies
within the Great Lakes Basin. Ideally, contaminants of concern for this analysis would depend upon those
constituents for which MCL violations in ground-water based public water supply systems were noted in the FRDS
retrieval. This analysis was based largely on the Region 5 comparative risk study.7
Uncertainties and Gaps: As for the drinking water risk analysis, we relied on the Region 5 comparative risk study
for much of our analysis of risk posed to ground water within the Basin; therefore, the uncertainties associated with
the use of the FRDS data base apply also to this analysis. Of particular note is recognition that the total population
served by ground water may be underestimated because FRDS classifies all systems that receive unfinished water
from both ground and surface waters as surface-water based systems; thus, a percentage of the population that relies
of ground water may go unnoticed when data from FRDS are utilized.
No data was available on ground water impacts on ecosystems, or on percentages of runoff versus ground water
contributions to baseflow for streams and lakes.
Ozone and Carbon Monoxide
Sources: The AIRS Facility subsystem (1990 retrieval) provided data for Basin counties on total, permitted,
and fugitive emissions, and for ambient levels.
Uncertainties and Gans: It is highly certain that same people in the Great Lakes Basin are experiencing adverse
health effects due to the significant ozone pollution problem. It is also highly certain that people in various areas
of the Basin are experiencing adverse health effects due to exposure to CO. However, only a moderate degree of
certainty can be associated with the exact number of people experiencing the specific health effects that are presented
above. The greatest difficulty in determining exact figures arises in assessing the population exposed.
First, it is not certain that all exceedances of NAAQS were monitored and thereby factored into the estimates
presented above. In particular, we are confident that there are numerous areas in the Basin that are not monitored
and that have ambient ozone concentrations exceeding the cut-off of 0.12 ppm. Since only monitoring data were
used, the figures presented above are likely to underestimate the risk.
Second, the area that each monitor represents must be determined. Finally, the number of people in this area who
are actually exposed to the pollutant must be determined. For ozone we made the assumption that an exceedance
at a monitoring location in any given county indicates exposure to the entire Great Lakes Basin population in that
county, even if exceedances were not predicted for all monitors in the county. This process leads to a moderate
degree of uncertainty.
As indicated by the qualitative nature of the discussion of the ecological impacts of ozone and CO pollution, there
is a high degree of uncertainty in the severity of the effect of the these pollutants. It is generally accepted that ozone
adversely impacts vegetation, but its severity in the Great Lakes Basin is not known.
Likewise, the estimated welfare impacts of ozone and CO pollution are highly uncertain. First, the analysis of these
impacts did not consider any impacts of CO. Second, the estimated impacts caused by ozone are quite uncertain.
Health care cost estimates are based on moderately uncertain health effects estimates and average health care unit
7 FRDS aiiiyiis will be iaclided U a later draft
-------
Page VI -11
costs that may introduce additional uncertainty. Estimates of agricultural damage from ozone are highly uncertain
because they are based on Region 5 exposure-response levels that may be uncertain and may not be accurate for the
Great Lakes Basin. In addition, we assumed the same yield per acre as was used for Region 5, which may not be
correct Material damage costs are similarly uncertain because they rely cm average costs.
Particulate Emissions
Sources: The AIRS Facility Subsystem (1990 retrieval) provided Basin county data on total, permitted, and
fugitive emissions, and on ambient levels.
Uncertainties and Gaps: The are considerable uncertainties in this evaluation of the risks posed by
particulate matter. First, monitoring for PM10 does not encompass the entire areal extent of the Basin and high levels
of particulate matter outside the monitored areas have not been incorporated into the estimated effects. Second, the
population exposed to levels of concern is not known with certainty. Therefore, this analysis uses best professional
judgment to estimate the exposed population.
In addition, ecological effects associated with particulate matter pollution are not known with any certainty. No data
was available concerning the effects of PM10 on flora and fauna.
Finally, the welfare effects are limited by their use of relatively uncertain human health effects estimates and the
application of a model developed for use with total suspended particles data while only PMt( data are available.
Underground Storage Tanks
Sources: Summary State UST data was obtained from State notification databases, and the National Database
of UST Release Incidences was employed in this section.
Uncertainties and Gans: Due to limitations inherent in the data on which these estimates are based, conservative
(but not worst-case) assumptions have been introduced whenever necessary to fill primary data gaps or inadequacies.
Thus, the resulting estimates could tend to overstate real-work! risk levels to some extent. Data was unavailable for
other than petroleum products, and for vapor release from ground water or soil.
Ibis analysis is confined to cancer and noncancer risks associated with daily ingestion of two liters of drinking water
contaminated by low levels of petroleum or its by-products. Higher exposure levels are assumed to trigger a taste
or odor response which effectively serves to limit subsequent exposure. It is further assumed that few, if any, would
be subjected to a lifetime of exposure. Available evidence points to much shorter exposure periods. The subchrooic
level of seven years provides a suitably conservative assumption for purposes of this analysis.
A potentially significant problem with this analysis, one that suggests the results are underestimated, is that it does
not incorporate risk calculations for inhalation and dermal exposure. The exposure figures are solely based on
ingestion of ground water. While dermal exposure appears to result in the least exposure, there is much controversy
over the relative importance of inhalation vs. ingestion. Estimates of the relative magnitude of inhalation compared
to ingestion exposure of VOCs range from approximately 1 to 6 times.*
Pesticide* • Health See Pesticide Discharges.
' Tkomu E. McRoae, "Himai Expouie to Volatile Orguic Compoiadi it Hoiachold Tap Water Tke la door bhalatioa Pathway,'
Table VD, Eivlroimcital Scieace lad Tectaolow. December 1987.
-------
Page VI - 12
APPENDIX B: BIBLIOGRAPHY
Agency for Toxic Substances and Disease Registry, 1989. Toxicological Profile for Mercury.
Argyle, R.L., U.S. Fish and Wildlife Service. 1982. "Alewives and Rainbow Smelt in Lake Huron: Midwater and
Bottom Aggregations and Estimates of Standing Stocks," in Transaction of the American Fisheries Society 111:,
p.267.
Baker, D.B. 1987. "Overview of Rural Nonpoint Pollution in the Lake Erie Basin," Logan, Terry J., et al., eds,
Effects of Conservation Tillage on Groundwater Quality: Nitrates and Pesticides. Ch. 4, Lewis Publishers, Inc.,
Chelsea MI, 1987.
Baker, J. and H. Harvey. 1985. "Critique of Acid Lakes and Fish Population Status in the Adirondack Region of
New York State. Draft Final Report for NAPA Project E3-25. U.S. EPA, Washington, D.C
Bishop, C. and Weseloh, D. 1990. "Contaminants in Herring Gull Eggs from the Great Lakes." State of the
Environment Fact Sheet No. 90-2. Environment Canada
The Canadian Coast Guard. "Counter Measures for Marine Spills of Hazardous Materials."
Chubb, Cmdr. P U.S. Coast Guard. Personal Communication.
Colby, PJ. 1971. "Alewife Dicoffs: Why do they occur?" in Limnos 4: p.26.
The Center for the Great Lakes. Great Lakes Area of Concern Fact Sheets.
Colquhoun, J. W. Kretzer, and M. Pfciffcr. 1984. "Acidity Status Update of Lakes and Streams in New York State.:
New York State Department of Environmental Conservation Report WM(P-83(6/64)). Albany, NY.
The Conservation Foundation. 1990. Great Lakes: Great Legacy?
Cronan, Jan, Soundings. March 1990. "Multiplying Mollusk a $7 Billion Problem for Great Lakes Area."
Dahl, F. H. and R.B. Mac Donald. 1980. "Effects of Control of the Sea Lamprey on Migratory and Resident Fish
Populations," In the Canadian Journal of Fisheries and Aquatic Sciences 37:pp.l886-1894.
DcVault, D. 1985. "Contaminants in Fish from Great Lake Harbors and Tributary Mouths." Archives of
Environmental Contamination and Toxicology 14:587-594.
Dourson, ML. and J. Milton Clark. 1990. "Fish Consumption Advisories: Toward a Unified, Scientifically
Credible Approach." Regulatory Toxicology and Pharmacology. 12, pp.161-178.
Eck, G.W., and L. Wells. 1987. "Recent Changes in Lake Michigan's Fish Community and Their Probable Causes,
with Emphasis on the Role of the Akwife," in ranadian Journal of Fisheries and Aquatic Sciences 44:, p.53.
The Economic Development Corporation of Erie County, Pennsylvania Task Force on Economic Adjustment
August, 1989. Economic Adjustment Strategy for Erie County. Pennsylvania.
Eisler, R. 1987. Mercury Hazards to Fish, Wildlife, and Invertebrates: A Synaptic Review. U.S. Fish and Wildlife
Service Biology Report No. 85(1.10).
-------
Page VI- 13
Eisler, R. 1988. "Lead Hazards to Fish, Wildlife, and Invertebrates: A Synoptic Review." Fish and Wildlife service,
U.S. Department of the Interior. Contaminant Hazard Reviews Report No. 14.
Emery, Lee, U.S. Fish and Wildlife Service. 1985. "Review of Fish Species Introduced into the Great Lakes, 1819-
1974," p.4.
Erb, C. 1987. "The Jury is Still Out." Penn State Agriculture, Fall, 1987: 2-11.
Evans, David 0. and David H. Loftus. 1987. "Colonization of Inland Lakes in the Great Lakes Region by Rainbow
Smelt, Osmerus mordax: Their Freshwater Niche and Effects on Indigenous Fishes." Canadian Journal of Fisheries
and Aquatic Sciences 44: 249-266.
Federal Register. Volume SI, Number 189, September 30,1986.
Fox, M.E., J.H. Carey, and D.G. Oliver. 1982. "Compartmental Distribution of Organochlocine Contaminants in
the Niagara River and the Western Basin of Lake Ontario." J. Great Lakes Research. 9:287-294.
Gatz. 197S. Controlling Airborne Particles. National Research Council, Committee on Particulate Control
Technology.
Gavis, J., and J.F. Ferguson, 1972. The Cycling of Mercury Through the Environment, Water Res.. 6:986-1008.
Gerdes, Wylie, Detroit Free Press. April 5,1990. "Zebra Mussels May Leapfrog to Inland Lakes by Way of Boats."
Great Lakes Commission. 1990. "A Guidebook to Drought Planning, Management and Water Level Changes in
the Great Lakes," p.49.
Gilbertson, M. 1974. "Pollutants in Breeding Herring Gulls in the Lower Great Lakes." Can. Field-Naturalist
8:273-280.
Gilman, A.P. et aL 1977. "Reproductive Parameters and Egg Contaminant Levels of Great Lakes Herring Gulls."
J. Wild. Manage. 41: 458-468.
Great Lake Fishery Commission. 1989. "The Report of the Evaluation of the Great Lakes Fishery Commission by
the Bi-National Evaluation Team: An Analysis of the Economic Contribution of the Great Tata* Sea Lamprey
Program," p.6.
Great Lakes Commission. "Great Lake Commercial and Recreational Harbor Dredging: Issues and
Recommendations."
"Great Lakes Reporter" Jan/Feb 1991. Center for the Great Lakes. 8:1.
Green, Larry, Los Angeles Times. January 15,1990. "Invasion of Zebra Mussels Threatens U.S. Waterways."
Guerra, D. and S. Horvatin. 1985. "Lake Michigan Sodium Mass Balance and Projections for Future IN-Lake
Sodium Concentrations."
Hatch, R.W., P.M. Haack, and E.H. Brawn, Jr. "Estimation of Alewife Biomass in Lake Michigan, 1967-1978" in
Transaction of the American Fisheries Society, 110:, p.575.
Haurwitz, Ralph, The Pittsburgh Press. February 26, 1990. "Clogging by Mussel Predicted In Rivers."
-------
Page VI -14
Henning, et aL 1989. Historical and Areal Deposition of Mercury in N.E. Minnesota and Northern Wisconsin Lakes.
Minnesota Pollution Control Agency.
Hoffman, D J. 1985. Morphological and Physiological Responses to Environmental Contaminants in Bird Embryos
and Nestlings. SETAC abstracts, U.S. Fish and Wildlife Research Service.
ICF Incorporated. 1988. Ecological Risk Assessment Documents Produced for Pennsylvania Comparative Risk
Project.
ICF Incorporated. 1988. OSWER Comparative Risk Project: Ground-water Valuation Task Force Report (PraftY
Prepared for EPA Office of Underground Storage Tanks, February 4, 1988.
International Joint Commission. 1985. "Living with the Lakes: Challenges and Opportunities."
International Joint Commission, Water Quality Board. 1987. "1987 Report on Great Lakes Water Quality," p. 23-65.
International Joint Commission. 1987. "Summary Report for the Workshop on Great Lakes Atmospheric Deposition."
presented at Windsor, Ontario October, 1987. p.14.
International Joint Commission. June, 1988. "Spills: the Human-Machine Interface."
International Joint Commission. 1989. "Living with the Lakes: Challenges and Opportunities."
International Joint Commission. 1989. "Report on Great Lakes Water Quality: Appendix A."
f
Jacobson, JX., S.W. Jacobson, and ILE.B. Humphrey. 1990. "Effects of in Utero Exposure to Polychlorinated
Biphenyls and Related Contaminants on Cognitive Functioning in Young Children." J. Pcdiatr. 11638-45.
Johnson, Ann, Erie Daily Times. March 28,1990. "Zebra Mussels Expected to Plague Boaters."
Johnson, A.H. and McLaughlin, S.B. 1986. "The Nature and Timing of the Deterioration of Red Spruce in the
Northern Appalachian Mountains." p. 200-230. in Acid Deposition - Long-Term Trends. Natural Research Council,
Washington, D.C.
Kubialc, T.J., et al. 1989. "Microoontaminants and Reproductive Impairment of the Forster's tern on Green Bay,
Lake Michigan-1983." Archives of Environmental Contamination and Toxicology. 18: 706-727.
Levey, Michael, The Buffalo News. December 5,1989. "Threat of Zebra Mussel is Spreading from Lake Erie to
Ontario."
Lindberg, SJE., 1984. Emission and Deposition of Atmospheric Mercury. Oak Ridge National Laboratory.
DE95006304.
Ludwig, J., in Behm Don, "Bird Defects Warn Humans," of "The Great Lakes: A Toxic Tragedy," The Milwaukee
Journal 1969, p.7.
Maclrie, G.L, W.N. Gibbons, B.W. Muocaster, and LM. Gray, Environment Ontario. 1989. "The Zebra Mussel,
Drcisscna Polymorphs: A Synthesis of European Experiences and a Preview for North America," pp. 6-7.
Manioc. 1981. Review of the National Ambient Air Quality Standards for Ozone. Preliminary Assessment of
Scientific and Technical Information. (EPA, Office of Air Quality Planning and Standards, Research Triangle Park,
N.C., 1986).
-------
Page VI - 15
Marsale, J. and H.Y.F. Ng. 1987. "Contaminants in Urban Runoff in the Upper Great Lakes Connecting Channels
Area." Environment Canada, National Water Research Institute.
McKone, T.E. December, 1987. "Human Exposure to Volatile Organic Compounds in Household Tap Water: The
Indoor Inhalation Pathway." Table VII, Environmental Science and Technology.
McMcekin, Carolyn, Niagara Gazette. 1990. "Mollusks Endanger Fish."
Michigan Department of Natural Resources. September 1988. Remedial Action Plan for Saginaw River Area of
Concern.
Michigan Department of Natural Resources. November 1988. Michigan's 1988 Nonpoint Pollution Assessment
Report.
Michigan Department of Public Health. December 14,1988. "Public Health News."
)
Minnesota Pollution Control Agency. September, 1988. Minnesota Nonpoint Source Pollution Assessment Report.
Minnesota Pollution Control Agency and Wisconsin Department of Natural Resources. September 1990. Stage I
Preliminary Draft The st. Louis River System Remedial Action Plan.
Monroe County Water Quality Management Agency. 1986. 1985-86 Annual Repent p.19.
NAPAP. 1988. "NAPA Interim Assessment: The Causes and Effects of Acidic Deposition. Volume IV: Effects
of Acidic Deposition."
National Academy of Science, National Academy for Engineering. March, 1973. "A Report of the Committee on
Water Quality Criteria." Water Quality Criteria 1972. p. 141.
National Atmospheric Deposition Program. 1987. "NADP/NTN Annual Data Summary: Precipitation Chemistry
in the U.S." Ft. Collins, CO.
The Nature Conservancy. 1990. Report of Significant Biological Areas of the Great Lakes Watershed.
New York State Department of Environmental Conservation. November, 1989. Buffalo River Remedial Action Plan.
New York State Department of Environmental Conservation. February, 1989. "Nonpoint Source Assessment
Report."
New York State Department of Environmental Conservation, Division of Water, Bureau of Monitoring and
Assessment. 1990. "New York State Water Quality 1990." {New York 1990 305(b) Report}.
New York Times. March 12,1990. "U.S. Help is Sought to Curb Mussels."
Northeast States for Coordinated Air Use Management (NESCAUM). August, 1989. Evaluation of the Health Effects
from Exposure to Gasoline Vapors. Final Report
Pesticide and Toxic Chemical News. May 2,1990. "Birds in Va. Carbofuran Kill of Concern, USDI Official Says,"
p.27.
Pesticide and Toxic Chemical News. June 6,1990. "Carbofuran Linked to Two Bird Kills in Delaware; No Misuse
Seen," p.23.
-------
Page VI - 16
Pesticide and Toxic Chemical News. August 8,1990. "Bald Eagle Valued at $40,000, FWS Notes in Carbofuran-
Predator Case." pp.9-10.
Pollick, Steve, The Toledo Blade. February 4,1990. "Warning Flags are Waving over Lake Erie Walleye Fishery."
Prancekevitius, P., K. Schrocr, and B. Manne. 1989. "U.S. Agricultural Tillage Practices in the Great Lake Basin,
1968." Great Lakes National Program Office.
Reid, N.W., Tang, AJ.S., Lusis, MA, and Klappenback, E.W. 1989. "The Deposition of Metals in Ontario."
Sandgren, Craig D. and John T. Lehman. September 1990. "Ecology of the Cladoceran Bythotrephes Cederstroemii
in the Laurentian Great Lakes.
The Seiche. 1990. "B.C. Jump to Inland Lake."
Sellgeby, J. August 27,1990. U.S. Fish and Wildlife Service, Personal Conversation.
Smith, B. R. and JJ. Tibbies. 1980. "Sea Lamprey in Lakes Huron, Michigan, and Superior. History of Invasion
and Control, 1936-1978," in the Canadian Journal of Fisheries and Aquatic Sciences 37: pp.1780-1801.
Snyder, F.L., Ohio Sea Grant College Program. 1990. "Zebra Mussels in the Great Lakes: The Invasion and its
Implications." p.l.
Sorensen, et aL 1969. Airborne Mercury Deposition and Watershed Characteristics in Relationship to Mercury
Concentrations in Water, Sediments, Plankton and fish of Eighty Northern Minnesota Lakes. Minnesota Pollution
Cootrol Agency.
Spiro, T.G. and W.M. Sligliani. 1980. Environmental Science in Persoecitve. p.204
Swackhamer, D.L. and D.E. Armstrong. 1986. "Estimation of the Atmospheric and Nonatmospheric Contributions
and Losses of Polycfalocinated Bipbenyls for Lake Michigan on the Basis of Sediment Records of Remote Lakes."
Environmental Science and Technology VoL 20: 9, 1986. p.879-883.
Swain, E.B., and D.D. Helwig. 1989. Mercury in Fish from Northeastern Minnesota Lakes. Minnesota Pollution
Control Agency.
Teeple, S. M. 1977. "Reproductive Suocess of Herring Gulls Nesting on Brogbers Island, Lake Ontario in 1973."
Can. Field-Naturalist 91:148-157.
Turner, Douglas, Buffalo News. March 8,1990. "$40 Million Sought to Fight Zebra Mussel."
VS. Department of Agriculture. 1969. "Consumption and Family Living." Agricultural Statistics
U.S. Department of Energy. 1969. "Chemical Data Bases for the Multimedia Environmental Pollutant Assessment
System (MEPAS): Version 1."
U.S. Environmental Protection Agency. Annex 2, Subsection 2(b), Great Lakes Water Quality Agreement of 1978.
as amended by Protocol signed November 18,1987.
U.S. Environmental Protection Agency. 1979. "Damage Assessment Studies following the NEPCO 140 Oils Spill
on the St. Lawrence River." EPA-600/7-79-256.
-------
Page VI - 17
U-S. Environmental Protection Agency, Office of Radiation Programs. October, 1984. Radionuclides: Background
Information for Final Rules-Volume n.
UJS. Environmental Protection Agency Office of Water Regualtions and Standards. 1984. Ambient Water Quality
Criteria for Mercury. EPA440/5-84-026, PB85-227452.
U.S. Environmental Protection Agency, Office of Solid Waste. June 13,1986. Analysis of the National Data Base
of Underground Storage Tank Release Incidents.
U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards, Research Triangle, NC. 1986.
Review of the National Ambient Air Quality Standard for Ozone. Preliminary Assessment of Scientific and Technical
Information.
U.S. Environmental Protection Agency. July 8, 1987. "National Primary Drinking Water Regulations; Synthetic
Organic Chemicals; Monitoring for Unregulated Contaminants." Final Rule 52. Federal Register 25690.
U.S. Environmental Protection Agency Great Lakes National Program Office. November 18, 1967. Genesee
River/Rochester Embayment Area of Concern Remedial Action Plan.
U.S. Environmental Protection Agency, Ecological Risk Workgroup, Office of Policy, Planning and Evaluation.
1987. Unfinished Business: A Comparative Assessment of Environmental Problems. Appendix in.
U.S. Environmental Protection Agency, 1987. Toxic Air Pollution Source Crosswalk. EPA 450/4-87-023a.
VS. Environmental Protection Agency and Environment Canada. 1988. The Great Lakes: An Environmental Atlas
and Resource Book.
U.S. Environmental Protection Agency, Great Lakes National Program Office. 1988. Great Lakes Atmospheric
Deposition /'GLAD') Network. 1982 and 1983. U.S. EPA, Chicago, II EPA 905 4 88 002.
U.S. Environmental Protection Agency, Office of Solid Waste. April, 1988. OSWER Comparative Risk Studv:
Ecological Risk Characterization Methodology.
U.S. Environmental Protection Agency. 1988. "Upper Great Lakes Connecting Channels Study." p. 258.
U.S. Environmental Protection Agency, Great Lakes Nation Program Office. 1988. U.S. Agricultural Tillage
Practices in the Great Lakes Basin. 1988.
U.S. Environmental Protection Agency, OWRS. 1989. "Assessing Human Health Risks from Chemically
Contaminated Fish and Shellfish: A Guidance Manual." EPA-503/8-89-002.
U.S. Environmental Protection Agency, Office of Water Regulations and Standards, Criteria and Standards Division.
1989. "Briefing Report to the EPA Science Advisory Board on the Equilibrium Partitioning Approach to Generation
Sediment Quality Criteria." (EPA 440/5-89-002, April, 1989).
U.S. Environmental Protection Agency, National Effluent Toxicity Assessment Center, Environmental Research
Laboratory - Duluth; Office of Water Enforcement and Permits; Office of Water Regulations and Standards; Office
of Health Effects Assessment - Cincinnati. 1989. "Draft Guidance oo Assessment, Criteria Development, and
Control of Biocon centra table Contaminants in Surface Waters."
-------
Page VI - 18
U.S. Environmental Protection Agency, Office of Radiation Programs. September, 1989. Environmental Impact
Statement—NESHAPS for Radionuclides: Background Information Document-Volume I: Risk Assessment
Methodology.
U.S. Environmental Protection Agency, Office of Radiation Programs. September, 1989. Environmental Impact
Statement—NESHAPS for Radionuclides: Background Information Document-Volume II: Risk Assessments.
U.S. Environmental Protection Agency, OSWER. October 1989. "Health Effects Assessment Summary Tables -
Fourth Quarter, FY 1989."
U.S. Environmental Protection Agency. 1989. Report on Great Lakes Confined Disposal Facilities.
U.S. Environmental Protection Agency, Office of Policy Analysis and Office of Policy, Planning, and Evaluation.
June, 1989. "Summary of Ecological Risks, Assessment Methods, and Risk Management Decisions in Superfund
and RCRA," p. 0-5.
U.S. Environmental Protection Agency Region S. 1989. "Zebra Mussels Fact Sheet," p.l.
U.S. Environmental Protection Agency. 1990. Draft Baseline Human Health Risk Assessment: Buffalo River, New
York.
U.S. Environmental Protection Agency Region 5. 1989. "Estimation and Evaluation of Cancer Risks Attributed to
Air Pollution in Southeast Chicago."
U.S. Environmental Protection Agency, Great Lakes National Program Office. 1990. "Assessment and Remediation
of Contaminated Sediments (ARCS) Work Plan." p24.
US. Environmental Protection Agency. October, 1990. Draft Paper "Baseline Report: Lake Michigan Lakewide
Management Plan."
U.S. Environmental Protection Agency. 1990. Draft 1990 Great Lakes Basin Report to Congress.
U.S. Environmental Protection Agency, Office of Policy, Planning and Evaluation. 1990. Pollutant Loadinst to the
Great Lakes and Chesapeake Bay.
U.S. Environmental Protection Agency. 1990. Draft U.S. EPA Transboundary Air Toxics Study.
U.S. Environmental Protection Agency. April, 1990. Draft Air Toxics Emission Inventories for the Lake Michigan
Region.
U.S. Environmental Protection Agency. 1990. Region 2 Comparative Risk Project Problem Area Analysis-
Estimating Wetland Values.
U.S. Environmental Protection Agency. 1990. Draft of "Second Annual Report on U.S. Progress in Implementing
the Great Lakes Water Quality Agreement with Canada."
U.S. Environmental Protection Agency. Draft Problem Area Paper Region 5 Comparative Risk Project. "Municipal,
Industrial and Nonpoint Source Discharges to Surface Waters," p. 7.
U.S. Environmental Protection Agency. Region 2 Summary of Problem Area Impacts on Lake Ontario.
-------
Page VI - 19
U.S. Environmental Protection Agency. "Subtitle D Regulatory Impact Analysis."
U.S. Geological Survey. 1987. National Water Summary 1987 - Hvdrologic Events and Water Supply and Use.
Water Supply Paper 2350.
U.S. Nuclear Regulatory Commission, Office of Nuclear Regulatory Research. "Sever Accident Risks: An
Assessment for Five Nuclear Power Plants."
Walsh, D. August 27,1990. Assistant Director, U.S. Fish and Wildlife Service, Personal Conversation.
Weseloh, D. et al 1979. "Organocfalorine Contaminants and Trends in Reproduction in Great Lakes Herring Gulls,
1974-1978." Trans 44th NA. Wildlif. Nat. Resources Coaf. 543-557.
Wisconsin Department of Natural Resources. September 1988. Remedial Investigation/Enhanced Screening Report,
Sheboygan River and Harbor, Volume 1 of 2.
Wisconsin Department of Natural Resources. July 1987. Toxic Substances Management Technical Advisory
Committee Report Lower Green Bay Remedial Action Plan.
Wisconsin Department of Natural Resources, Wisconsin Division of Health. October, 1989. "Health Advisory for
People Who Eat Sport Fish from Wisconsin's Waters,"
Wisse, John, The Vindicator. December 24,1989. "Trash' Fish, Dudes Could Save Lake Erie."
Wong, K.C. and M.Y. Hwang. 1981. "Children Born to PCB Poisoned Mothers." Clin. Med. (Taipei). 7:83-87.
Wozniak, Mary, Niagara Gazette. 1990. "Mussel-bound Plants Seek Safe Way to Oust Invaders."
Wozniak, Mary, Niagara Gazette. 1990. "Mussel Menace has Already Hit Other Water Plant
-------
Page VI-20
APPENDIX C; ACRONYM TTST
ACRONYMS
LISTING
(AIRS)
Acromctric Information Retrieval System
(AOC)
Areas of Concern
(ARC)
Assessment and Remediation of Contaminated Sediments
(AWQC)
Ambient Water Quality Criteria
(BAT)
Best Available Technologies
(BEST)
Best Extraction Sludge Technology
(CDF)
Confined Disposal Facilities
(CERCLA)
Comprehensive Environmental Response, Compensation, and Liability Act
(CERCLIS)
Comprehensive Environmental Response, Compensation and liability
Information System
(CF)
Critical Fluids
(CFR)
Code of Federal Regulation
(CSO)
Combined Sewer Overflow
(cnc)
Conservation Technology Information Center
(EPA)
Environmental Protection Agency
(ERNS)
Emergency Response Notification System
(FRDS)
Federal Reporting Data System
(GLAD)
Great Lakes Atmospheric Deposition
(GLFC)
Great Lakes Fishery Commission
(GLNPO)
Great Lakes National Program Office
(HCC)
Health Care Costs
(HDWMS)
Hazardous Waste Data Management System
(HEAST)
Health Effects Assessment Summary Tables
(HI)
Hazard Index
(HRS)
Hazard Ranking System
(DC)
International Joint Commission
(IRIS)
Integrated Risk Information System
(LAMP)
Lakewide Action Management Plan
(LDU)
Land Disposal Unit
(LEL)
Lower Explosive Limit
(LOAEL)
Lowest Observed Adverse Effect Level
(MCL)
Maximum Contaminant Level
(MCLG)
Maximum Contaminant Level Goals
(MEPAS)
Multimedia Environmental Pollutant Assessment System
(MP)
Migration Potential
(NAAQS)
National Ambient Air Quality Standards
(NAPAP)
National Acid Precipitation Assessment Program
(NAPL)
Non-aqueous phase liquids
(NESCAUM)
Northeast States for Coordinated Air Use Management
(NOAEL)
No Observed Adverse Effect Levels
(NOPES)
Non-Occupational Pesticide Exposure Study
(NPDES)
National Pollutant Discharge Elimination System
(NPL)
National Priority List
(OAQPS)
Office of Air Quality, Planning and Standards
(ODW)
Office of Drinking Water
(OPP)
Office of Pesticide Programs
(OPPE)
Office of Policy, Planning and Evaluation
-------
(OSWER)
(OTA)
(OWD)
(PCS)
(PWS)
(RAP)
(RCRA)
(RFF)
(SDWA)
(STU)
(TMP)
(TRI)
(TSDF)
(USCG)
(US)
(UST)
ACRONYMS
(AAC)
(ANC)
(AOX)
(BCF)
(BHC)
(BOD)
(CAC)
(CFC)
(CO)
(COD)
(CPF)
(DDE)
(DDT)
(HCB)
(kg)
(km)
(KPEG)
(mg)
(mg/L)
(NOJ
(OSC)
(PCB)
(PCE)
(PAH)
(PPb)
(ppm)
(RfD)
(RIA)
(SO*)
(TCA)
(TCE)
(TCDD)
Page VI - 21
Office of Solid Waste and Emergency Response
Office of Technology Assessment
Office of Drinking Water
Permit Compliance System
Public Water System
Remedial Action Plan
Resource Conservation and Recovery Act
Resources for the Future
Safe Drinking Water Act
Storage and Treatment Unit
Toxic Management Plan
Toxic Chemical Release Inventory
Treatment, Storage, and Disposal Facility
United States Coast Guard
United States
Underground Storage Tank
CHEMICAL/MEASUREMENT LISTING
Acute acceptable concentrations
Acid neutralizing capacity
Adsoibablc organic halogen
Bioooncentration factor
Benzenhexachloride
Biochemical oxygen demand
Chronic acceptable concentrations
Chlorafluorocarbons
Carbon monoxide
Chemical oxidation
Cancer Potency Factor
Dichlorodiphenyl dictaloriethylene
Dichlorodiphenyltrichloroethane
Hexacfalorobeozeoe
kilogram
kilometer
Potassium-polyethylene
milligram
milligram per liter • part per million
Nitrogen oxide
Octachlorostyrene
Polychlorinated Biphenyls
Tetrachloroethylene
Polycyclic aromatic hydrocarbon
parts per billion
parts per million
Reference Dose
Regulatory Impact Analysis
Sulfur dioxide
Tricfaloroe thane
Ttichloroetbykne
Tctracfalorodibenzo-p-dioxin
-------
(TCDF) Tetracblorodibenzofuran
(THM) Trihalom ethane
(TOC) Total organic chlorine
(TSS) Total suspended solids
(TTO) Total toxic organics
(ug/L) microgram per litre
(VOC) Volatile organic compound
-------
APPENDIX D
ANNEX 1 LIST OF GREAT LAKES SUBSTANCES
PRELIMINARY
CHEMICAL
CAS No.
LISTING
TRIGGER
SYNONYM
Acenaphthene
83-32-9
1
AlachJor
15972-60-8
5
Aldrin
309-00-2
5
Antimony
7440-36-0
5
Arsenic
7440-38-2
8
Atrazine
1912-24-9
5
Barium
7440-39-3
5
Benzene
71-43-2
8
Benz(a)anthracene
56-55-3
8
Benzo(a)pyrene
50-32-8
8
Benzo(b)fluoranthene
205-99-2
8
Benzo(k)fluoranthene
207-08-9
8
Benzo(g,h,i)pe xylene
191-24-2
11
Beryllium
7440-41-7
1
Bis(2-ethylhexyl)phthalate
117-81-7
8
Bromodichloromethane
75-27-4
8
Butylate
2008-41-5
5
Butylbenzylphthalate
85-68-7
4
Cadmium
7440-43-9
1
Carbofuran
1563-66-2
5
Carbon disulfide
75-15-0
7
Carbon tetrachloride
56-23-5
8
Chlordane
54_74_9
5
cis-chlordane
5103-71-9
*
trans-chlordane
5103-74-2
*
oxychlordane
27304-13-8
•
Chlorobenzene
108-90-7
5
Chlorodibromome thane
124-48-1
8
Chloroform
67-66-3
8
Chloromethane
74-87-3
8
1 -Chloronaphthalene
90-13-1
4
Chlorpyrifos '
2921-88-2
5
Chromium
7440-47-3
1
Cobalt
7440-48-4
5
Copper
7440-50-8
1
Cyanazine
21725-46-2
5
Cyanide (an ion)
57-12-5
1
DDT and metabolites
NA
8
o.p'-DDT
789-02-6
8
p,p'-DDT
50-29-3
8
o,p'-DDD
53-19-0
8
p,p'-DDD
74-54-8
8
o,p'-DDE
3424-82-6
8
p,p'-DDE
72-55-9
8
p,p'-DDMU
1022-22-6
8
Dehydroabietic Acid
1740-19-8
1
Diazinon
333-41-5
1
3,4-benzopyrene
11,12-benzofluoroanthene
alpha-chlordane
beta-chlordane
bladex
-------
Di bcnzo(a ,h)antiiracene 53-70-3 8
Di-n-butylphthalatc 84-74-2 1
Dicamba 1918-00-9 5
Dichloromethane 75-09-2 8
1,2-Dichlorobenzene 95-50-1 5
1,4-Dichlorobenzene 106-46-7 8
1.1-Dichloroethane 75-34-3 8
1.2-Dichloroethane 107-06-2 8
1,2-DichIoropropane 78-87-5 8
2,4-Dichlorophenoxy-
acetic acid 94-75-7 5
Dieldrin 60-57-1 8
N,N-Dimethylaniline 121-69-7 5
Dimethyidisulfide 624-92-0 1
Di-n-octylphthalate 117-84-0 1
2,6-Di-t-butyl-p-cresol (BHT) 128-37-0 10
Endosulfan 115-29-7 5
alpha-EndosuIfan 959-98-8 *
beta-Endosulfan 33213-65-9 *
Endosulfan sulfate 1031-07-8 *
Endrin 72-20-8 5
Ethion 563-12-2 5
Ethylbenzene 100-41-4 1
S-Ethyl dipropylthiocarbamate 759-94-4 5
Fonophos 944-22-9 5
Furan 110-00-9 5
Heptachlor 76-44-8 5
HepCachlor epoxide 1024-57-3 8
Heptachlorinated dibenzo-
p-dioxins 38871-00-4 8
Heptachlorinated dibenzofurans 38998-75-3 8
Hcxachlorinaled dibenzo-
p-dioxins 34465-46-8 8
Hexachlorinated dibenzofurans 55684-94-1 8
Hexacbiorobeozene 118-74-1 8
Hexachloro-l,3-butadiene 87-68-3 5
Hexacblorocyclopentadiene 77-47-4 1
Hexacbloroethane 67-72-1 5
Indeno(l>2l3-cd)pyrene 193-39-5 11
Lead 7439-92-1 8
Lindane (gamma-HCH) 58-89-9 8
Hexachlorocyclobexane 608-73-1 *
alpha-Hexachlorocyclobexane 319-84-6 8
beta-Hexachlorocyclohexane 319-85-7 *
Linuron 330-55-2 5
Malathion 121-75-5 5
Mercury 7439-97-6 I
p,p-Melhoxychlor 72-43-5 1
2-Methyl-4-chlorophenoxy-
acetic acid 94-74-6 5
2-(2-Methyl-4-cblorophenoxy)-
propionic acid 7085-19-0 5
2,4-D
methyldisulfide
EPTC
furfuran; divinylene oxide
gamma-bexacblorocyclobexane
alpha-HCH
MCPA
MCPP; mecoprop
-------
Metolachlor
Metribuzin
Mirex
10-monoH-mircx
2,8-diH-mirex
photomirex
Naphthalene
Nickel
n-Nitrosodiphenylamine
Octachlordibenzo-
p-dioxins
Octachlorodibenzofcrans
Octachlorostyrene
Pendimethalin
Pentachlorobeazene
Pentachlorodibenzo-
p-dioxins
Pentachlorodibenzofurans
Pentachlorophenol
Perylene
Phenanthrene
Phenol
Polychlorinated biphenyls
Aroclor-1232
Aroclor-1242
Aroclor-1248
Aroclor-1254
Aroclor-1260
Aroclor-1262
2,2' ,3,4,5,5' ,6-hepta-
chloro-1,1 '-biphenyl
hexachloro-1,1 '-biphenyls
2,2' ,3,3 \4,4'-hexachloro-
1,1'-biphenyl
2,2',3,4,4',5 '-hexachloro-
1,1'-biphenyl
2,2',3,4,5,5'-hexachloro-
l.l'-biphenyl
2,2',4,4',5,5'-hexachloro-
l.l'-biphenyl
2,2',4,4',6,6'-hexachloro-
l.l'-biphenyl
pentachloro-1,1 '-biphenyls
2,2',3,4,5'-penUchloro-
l,l'-biphenyl
2,2',3,5',6-pentachloro-
l,r-biphenyl
2,2',4,5,5'-pentachloro-
1,1 '-biphenyl
2,3' ,4,4' ,5-pentachloro-
1,1*-biphenyl
2,2',3,3,,4,4,,5,5'-octa-
51218-45-2 5
21087-64-9 5
2385-85-5 5
NA •
NA ~
39801-14-4 •
91-20-3 5
7440-02-0 8
86-30-6 8
3268-87-9 8
39001-02-0 8
29082-74-4 11
40487-42-1 5 penoxalin; prowl
608-93-5 11
36088-22-9 8
30402-15-4 8
87-86-5 5
198-55-0 10
85-01-8 1
108-95-2 9
NA 8
11141-16-5 8
53469-21-9 8
12672-29-6 8
11097-69-1 8
11096-82-5 8
37324-23-5 8
52712-05-7 8
26601-64-9 8
38380-07-3 8
35065-28-2 8
52712-04-6 8
35065-27-1 8
33979-03-2 8
25429-29-2 8
38380-02-8 8
38379-99-6 8
37680-73-2 8
31508-00-6 8
-------
chloro-1, l'-biphenyl
Tetrachloro-1,1 '-biphenyls
2,2',3,5'-tetrach]oro-
1, l'-biphenyl
212',4,5'-tetrachloro-
1, l'-biphenyl
2,2' ,6,6'-tetrachloro-
1, l'-biphenyl
2,3',4,4'-tetrachloro-
1, l'-biphenyl
2,3,4,5-tetrachloro-
1, l'-biphenyl
2,3' ,4* ,5-tetrachloro-
1, l'-biphenyl
2' ,3,4-trichloro-1,1 '-
biphenyl
Pyridine
Selenium
Silver
Silvex (2,4,5-TP)
Styrene
1.2.3.4-Tetrachlorobenzene
1.2.3.5-Tetrachlorobenzene
1,2,4,5-Tetrachlorobenzene
T etrachlorodibenzo-
p-dioxins
2,3,7,8-Tetrachlorodibenzo-
p-dioxin
Tetrachlorodibenzofurans
1,1,2,2,-Tetrachloroethane
T etrachloroethylene
Thallium
Toluene
Toxaphene
T ribromomethane
1,2,3-T richlorobenzene
1,1,1 -T richloroethane
1,1,2-Trichloroethane
T richloroethylene
2.4.5-T richlorophenol
2.4.6-T richlorophenol
2,4,5-T richlorophenoxyacetic
acid (2,4,5-T)
2,4,5-T richlorotoluene
Trifluralin
Uranium
Xylenes
o-Xylene
m-Xylene
p-Xylene
35694-08-7 8
26914-33-0 8
41464-39-5 8
41464-40-8 8
15968-05-5 8
32598-10-0 8
33284-53-6 8
32598-11-1 8
38444-86-9 8
110-86-1 5
7782-49-2 1
7440-22-4 5
93-72-1 5
100-42-5 8
634-66-2 1
634-90-2 1
95-94-3 5
41903-57-5 8
1746-01-6 8
55722-27-5 8
79-34-5 8
127-18-4 8
7440-28-0 4
108-88-3 7
8001-35-2 8
75-25-2 5
87-61-6 1
71-55-6 8
79-00-5 8
79-01-6 8
95-95-4 1
88-06-2 8
93-76-5 5
6639-30-1 1
1582-09-8 5
7440-61-1 5
1330-20-7 9
95-47-6 9
108-38-3 9
106-42-3 9
fenoprop; 2,4,5-TP
2378-TCDD, "dioxin"
bromoform
2,4,5-T
-------
GREAT LAKES WATER QUALITY AGREEMENT ANNEX 1 LIST NO. 1; 1989 ISSUE
KEY TO SYMBOLS
NA ¦ Not Available
LISTING TRIGGER: This refers to the category within the Standard Methods
which resulted in the chemical being placed on List No. 1. A chemical
may be eligible for placement due to one or more triggers; the one
listed is not necessarily the most important toxicological effect of
the chemical. See the Standard Methods (draft 11/30/89) for the
toxicity trigger level associated with each of the below categories:
1. Acute aquatic toxicity
2. Acute mammalian toxicity
3. Chronic aquatic toxicity
4. Chronic plant toxicity
5. Chronic mammalian toxicity
6. Chronic terrestrial non-mammalian toxicity
7. Teratogenicity
8. Carcinogenicity
9. Advene reproductive effects
10. Mutagenicity
U. Bioaccumulation
* Placed on the List based upon parent chemical toxicity, for the
purposes of grouping
-------
Page VI - 23
APPENDIX E: DESCRIPTIONS OF U.S. GREAT LAKES AREAS OF CONCERN*
Lake Superior
(1) St. Louis River. The St Louis River is 171 miles long, with its headwaters in Minnesota, emptying into
Lake Superior, forming the boundary between Wisconsin and Minnesota for the lower 23 miles. The AOC includes
the near-shore waters of southwestern T aire Superior, all of the Duluth-Superior Harbor, and upstream past the City
of Qoquet Land use is residential, recreational, and industrial, including shipping and pulp and paper mills.
The following contaminants are found in bottom sediments and have impaired fish consumption, as well as
swimming and other water contact sports:
Problems: Sources:
¦
Organic Chemicals
¦
Past and present industrial
¦
Heavy Metals
discharges
¦
Cyanide
¦
Municipal combined sewer
overflows
¦
Municipal treatment plant waste water
¦
Stormwater and agricultural runoff
¦
Contaminated ground water, surface water, and
sediments
¦
Atmospheric deposition
(2) Torch Lake. Torch Lake is located on the southern shore of Lake Superior in the State of Michigan's
upper peninsula. The entire lake is the AOC, including the shoreline. The land around the AOC is primarily
forested now, and copper mining and processing were prevalent in the area until 1968.
Lake sediments are contaminated with the following contaminants, resulting in impaired fish consumption and
degraded benthic organisms.
Problems: Sources:
¦ Heavy metals ¦ Tailings Copper mine tailings
¦ Organic chemicals deposited in the lake
(3) Deer Lake, Carp Creek, Carp River. Located in the upper peninsula of Michigan northwest of Ishpeming,
Deer Lake is near the shore of Lake Superior. The lake is connected to Lake Superior by Carp River, and fed by
Carp Creek. The AOC includes Deer Lake, the Carp River and Carp Creek. Land use around the area is primarily
forestry, however, inn ore and gold mining activities took place in the past. Sediments in the surface water are
contaminated, impairing fish consumption, water, and aquatic wildlife.
Problems: Sources:
¦ Heavy metals ¦ Past mining industry
¦ Ishpeming wastewater treatment plants
* Iaformatios presealed ii tkU sectioa was drawa from Ike 1969 1)C Report oa Great Likes Water Qaality: Appeadix A, aad from Great
Lake* Areas of Coacera Fact Skeels prepared by the Cealer for the Great Lakes.
-------
Page VI - 24
Lake Huron
(1) Saginaw Bay, Saginaw River. On the western shore of Lake Huron, Saginaw River is 52 miles long and
between 13-26 miles wide. Saginaw River accounts for about 75% of the water which drains into Saginaw Bay.
The AOC includes the entire River and Bay. Over half the region's land use is agricultural. The constituents below
have ffintamirmted sediments, and increased the level of nutrients which leads to a growth in undesirable algae, a
threat to fish consumption, and adverse effects on wildlife and their habitat.
Problems: Sources:
¦
PCBs
¦
Contaminated sediments
¦
Nutrients
¦
Agricultural runoff
¦
Combined sewer overflows
¦
Urban runoff
¦
Atmospheric deposition
¦
Past industrial discharges
¦
Municipal wastewater treatment plants
¦
Industrial discharges
Lake Michigan
(1) Manistique River. The Manistique River is located in Upper Peninsula Michigan, and flows northeast into
Lake Michigan at the City of Manistique. The AOC includes 1.7 miles of the river from the dam in Manistique to
the harbor. Land use within the AOC is primarily residential and industrial,
The following contaminants have contaminated sediment and ground water and surface runoff into the river,
impairing fish consumption.
Problems: Sources:
¦ Heavy metals ¦ Paper products
¦ PCBs manufacturing
(2) Menominee River. The Menominee River borders Upper Peninsula Michigan and northeastern Wisconsin.
The main portion flows between the cities of Menominee and Marionette and into Green Bay. The AOC includes
the lower three miles of the river up to the second Scott Paper Company dam, the cities of Marionette and
Menominee, and the nearshore area of Green Bay extending three miles from the river mouth. Land use is primarily
industrial and residential.
Hie contaminants below have been found in sediments and ground water beneath the AOC, advene affects
on fish consumption, navigation, aesthetics, and water recreational activities.
Problems: Sources:
¦ PCBs ¦ Combined sewer overflows
¦ Arsenic and other and sewage bypassing
heavy metals ¦ TnHitctrifli discharge such
¦ Oil and grease as chemical companies
-------
Page VI - 25
¦ Urban runoff
¦ Coal and salt pile runoff
¦ Contaminated ground water
(3) Fox River/Southern Green Bay. The Fox River is located in northeastern Wisconsin at the southern end
of Green Bay. The AOC covets approximately 20 square miles and includes: the City of Green Bay, the lower 7
miles of the Fox River, and the southern end of Green Bay extending north to Long Tail Point and Point au Sable.
Land use is highly industrial and agricultural, with the largest concentration of pulp and paper mills in the world.
A significant number of contaminants have been found in sediments, surface water, and ground water, causing
adverse affects on wetlands and other critical habitats; water and terrestrial wildlife; underwater plants; shore and
witter use; water recreational activities; and navigation. In addition, these contaminants have caused an imbalance
in the food chain around the AOC.
Problems: Sources:
¦ PCBs and other organic ¦ Pulp and paper mills and
compounds other industrial
¦ Pesticides discharges
¦ Heavy metals ¦ Municipal sewage
¦ Phosphorus ¦ Agricultural and urban
runoff
¦ Contaminated sediments
(4) Sheboygan River. The Sheboygan River feeds into Lake Michigan at the City of Sheboygan, Wisconsin.
The AOC consists of the lower 14 miles of the river from the Sheboygan Falls and includes Sheboygan Harbor.
Land use is industrial and residential, but the headwaters and tributaries of the Sheboygan River flow from
agricultural areas west of the city.
The following contaminants have contaminated wildlife, sediments, ground water and surface water, impairing fish
consumption, water recreational activities, and wildlife habitat.
Problems: Sources:
¦ PCBs • Industrial discharge
¦ Heavy metals ¦ Agricultural and urban
¦ Phosphorus and organic runoff
compounds ¦ Contaminated sediment
¦ Fecal coliform
(5) Milwaukee Estuary. The Milwaukee Estuary AOC, also referred to as the Milwaukee Harbor AOC, is
located near Milwaukee, Wisconsin. The AOC includes the lower 32 miles of the river, the lower 2 J miles of the
Menominee River, and the lower 25 miles of the Kinnickinnic River. Land use is primarily industrial, and contains
large shipping and port facilities. Other uses include residential and commercial districts along the river. Sediment
and water quality problems have been identified in the AOC, as a result of the following contaminants. These
contaminants have impaired fish and wildlife consumption, recreational activities, navigation, and aesthetics.
Problems: Sources:
¦ PCBs ¦ Industrial discharge
¦ Heavy metals ¦ Municipal wastewater
-------
Page VI - 26
¦ Phosphorus and organic
compounds
¦ Contaminated sediments
¦ Low levels of oxygen
¦ Bacteria
¦ Nutrients
treatment plants
¦ Combined sewer overflows
¦ Urban and storm sewer
runoff
¦ Impoundment area for
dredged sediment
(6) Waukegan Harbor. Waukegan Harbor is located approximately 37 miles north of Chicago. The AOC
consists of the entire harbor, approximately 37 acres in size. Land use around the harbor is primarily industrial and
commercial.
The following contaminants found in soil and harbor sediments have impaired fish consumption and navigation.
Problems: Sources:
¦ PCBs ¦ Outboard Motor Company
plant
(7) Grand Calumet River, Indiana Harbor Canal. The Grand Calumet River is 13 miles long, and flows
through the heavily industrialized cities of Gary, East Chicago, and Hammond in northwest Indiana. The AOC
includes the east branch of the river, a small portion of the west branch, and the entire Indiana Harbor Canal. Land
use is primarily industrial and residential
Sediments and water are contaminated with the following contaminants, impairing water recreational activities,
industrial water supplies, fish consumption, and navigation.
Problems: Sources:
PCBs and organic
compounds
Heavy metals
Fecal ooliform bacteria
Industrial and municipal
wastewater discharge
Combined sewer overflows
Contaminated sediment
Highway and industrial runoff
Leachate and ground water from waste sites
Atmospheric deposits
Illegal dumping
(8). Kalamazoo River. The Kalamazoo River is located in southwestern Michigan, and discharges into Lake
Michigan near the City of Saugatuck. The AOC includes the lower 28 miles of the river. Land upstream of the
AOC is heavily industrialized, including several large paper companies.
Fish, water, and sediment oontains the following contaminants, impairing fish mnaumptimv
Problems: Sources:
¦ PCBs ¦ Contaminated sediment
¦ Impoundment area for dredged sediment
(9). Muskegon River. Muskegon Lake is on the eastern shore of Lake Michigan near the City of Muskegon,
separated from Lake Michigan by large sand dunes. The entire lake is the AOC. Land use in its watershed is
primarily residential and industrial, with chemical companies, and foundries.
-------
Page VI - 27
Sediment and water contain the following contaminants, impairing fish consumption.
Problems: Sources:
PCBs
Heavy metals
Contaminated sediments
Ground water
contamination
Industrial and municipal discharge
Storm sewer drainage
(10). White Lake. White Lake is located near the eastern shore of Lake Michigan just north of the City of
Muskegon. The AOC includes the entire lake and a 25 mile-wide zone around the lake. Land use is primarily
recreational and agricultural, however, there is some residential and industrial activity.
The primary problem is contaminated ground-water containing the following contaminants. This contamination is
impairing fish consumption and drinking water.
Problems:
Sources:
PCBs
Heavy metals
Phosphorus
Organic chemicals
Chloroform
Municipal wastewater
treatment discharge
Industrial and chemical
company discharge
Contaminated ground water
Urban runoff and combined sewer overflows
Contaminated sediment
Lake Erie
(1). Clinton River. The Clinton River is located just north of Detroit, Michigan and flows into Lake St. Clair.
The AOC includes a spillway at the mouth and the main branch of the river, downstream of Red Run. Land use
within the AOC is entirely urban, and on the north branch is agricultural
Sediment and water contains the following contaminants, impairing fish and river bottom aquatic wildlife
communities.
Problems: Sources:
¦ PCBs ¦ Municipal and industrial
¦ Heavy metals discharge
¦ Pesticides ¦ Urban and agricultural
¦ Phosphorus and organic runoff
matter ¦ Contaminated sediments
¦ Contaminated ground water
(2). Rouge River. Rouge River flows into the Detroit River at Zug Island in southeastern Michigan. Over 400
lakes and ponds and many tributaries flow into the river's four main branches. The AOC includes the entire river
basin. Land use in the river basin is used for housing, commerce, or industry.
Sediment and water contains the following contaminants, impairing fish consumption, water recreational activities,
navigation, and industrial and agricultural water supplies.
-------
Page VI - 28
Problems: Sources:
PCBs
Heavy metals
Organic chemicals
Bacteria
Habitat loss
Combined sewer overflows
Urban runoff
Hazardous waste sites
Contaminated sediment
Municipal and industrial discharge
(3). Raisin River. The Raisin River is located southeastern Michigan, and discharges into Lake Erie at Monroe
Harbor. The AOC includes the lower 2.6 miles of the river, from the dam No. 6 through the harbor and one-half
mile into Lake Erie, Plum Creek which discharges to Lake Erie through a canal, and from the north end of Sterling
State Park to one-half mile south of Dunbar Road on the south bank of Plum Creek. Land use is residential and
industrial, inrimting metal, steel, paper, chemical, and food processing industries.
The following contaminants have been found in sediment, water, and fish, impairing fish consumption, drinking
water, and navigation.
Problems: Sources:
¦ PCBs ¦ Municipal and industrial
¦ Heavy metals landfills and waste piles
¦ Oil and grease ¦ Contaminated ground- and
surface-water
¦ Contaminated sediment
(4). Maumee River. The Maumee River, the largest tributary to Lake Erie, is located in northwestern Ohio.
The AOC includes the lower 20J5 miles of the river, Maumee Bay, the area near the shore of Lake Erie southeast
of the mouth, and the lower segments of several tributaries to the river and bay. Land use is industrial, agricultural
and indudes the City of Toledo.
The following contaminants have been found in water and sediment, impairing habitats for fish, plants, and other
organisms, navigation, aesthetics, industrial and agricultural supplies, and water recreational activities.
Problems: Sources:
¦ PCBs ¦ Contaminated sediments
¦ Heavy metals ¦ Municipal sewage
¦ Phosphorus and organic treatment plants
¦ Fecal coliform bacteria ¦ Agricultural and urban runoff
¦ Combined sewer overflow
¦ Municipal and industrial discharge
¦ Municipal and industrial landfills and
impoundments
(5). Black River. The Black River flows into Lake Erie between the cities of Cleveland and Sandusky at the
City of Lorain. The AOC includes Lorain Harbor and the portion of the river flowing between Lorain Harbor and
the city of Elyria, which lies approximately 15.5 miles upstream.
Sediment and water contain the following contaminants, impairing fish consumption, navigate, and water recreation
activities.
-------
Page VI - 29
Problems:
¦ Heavy metals
¦ Organic chemicals
¦ Decreased oxygen levels
¦ Increased water
temperature
Sources:
¦ Municipal wastewater
treatment plants
¦ Industrial discbarge
¦ Municipal landfills
¦ Agricultural runoff
¦ Combined sewer overflow
(6). Cuyahoga River. The Cuyahoga River is located in northeastern Ohio. It enters Lake Erie through the
Cleveland Harbor. The AOC includes the portion of the river below the Akron Wastewater Treatment Plant, the ship
channel and the shore area of Lake Erie inside the Cleveland Harbor breakwater.
The following contaminants have been found in water and sediment, impairing fish consumption, recreational
activities, navigation, and aesthetics.
Problems: Sources:
¦ PCBs ¦ Combined sewer overflow
¦ Organic chemicals ¦ Municipal wastewater
¦ Heavy metals treatment plants
¦ Decreased oxygen levels ¦ Industrial discharge
¦ Fecal colifonn bacteria ¦ Dredging of ship canal
¦ Agricultural and urban runoff
¦ Erosion
¦ Contaminated ground water
¦ Shoreline development
(7). Ashtabula River. The Ashtabula River is located in the northeastern comer of Ohio and enters Lake Erie
at the City of Ashtabula. The AOC includes the lower 2 miles of the river, Fields Brook which empties into the
river approximately 1.5 miles upstream river's mouth, and nearshore Lake Erie.
The following contaminants are in sediment and water, impairing fish consumption, navigation, and water recreational
activities.
Problems: Sources:
¦ PCBs ¦ Municipal wastewater
¦ Heavy metals treatment plants
¦ Organic chemicals ¦ Industrial discharge
¦ Contaminated sediments
¦ Municipal and private landfills
¦ Atmospheric deposits
¦ Combined sewer overflows
(8). Buffalo River. In the southern part of the City of Buffalo, New York, Buffalo River empties into eastern
Lake Erie. The AOC includes the river and the three major tributaries that feed it: Cayuga, Cazenovia, and Buffalo
Creek. Land use was heavily industrial, however, steel production and oil refining stopped in the early 1980s. In
addition, sediments were found to contain dyes from chemical and dye industrial dischargers. In the early 1980s,
seventeen major industries were then diverted into Buffalo's municipal treatment systems.
The following substances have contaminated sediments, impairing fish consumption; navigation and water
recreational activities; causing loss of fish and wildlife habitat; and degrading fish and wildlife populations.
-------
Page VI - 30
Problems:
Sources:
PCBs
Heavy metals
Industrial organic
chemicals
Pesticides
Municipal and industrial
wastewater discharges
Combined sewer overflows
Contaminated sediment
Inactive hazardous waste sites
Lake Ontario
(1). Eighteen Mile Creek. Located in northwestern New York, the AOC is just north of the city of Lockport.
The AOC includes the creek, Olcott Harbor, and the waters near the shore of Lake Ontario near Olcott.
Sediment and water has been found to contain the following contaminants, impairing fish consumption and
navigation.
Problems: Sources:
¦ PCBs ¦ Municipal wastewater
¦ Heavy metals treatment plants
¦ Organic chemicals ¦ Industrial discharge
¦ Pesticides
(2). Rochester EmbaymenL Rochester Embayment is located on the southern shore of Lake Ontario near the
city of Rochester, New York. The AOC includes nearshore areas of Ontario from the town of Greece to the
Nine Mile Point area of the town of Webster, and the Genessee River, which flows into the Rochester EmbaymenL
Land use is primarily industrial and residential.
Sediment and water contains the following contaminants, impairing fish consumption, water recreational activities,
and aesthetics.
Problems: Sources:
¦ PCBs ¦ Municipal wastewater
¦ Heavy metals treatment plants
¦ Organic chemicals ¦ Industrial discharge
¦ Combined sewer overflows
¦ Agricultural and urban runoff
¦ Contaminated sediment
¦ Inactive disposal sites
(3). Oswego River. The Oswego River empties into Lake Ontario near the City of Oswego, New York. The
AOC includes the harbor within the breakwaters and upstream to the first barrier dam. Land use is urban and
residential
The following contaminants have been found in sediment and water, resulting in impaired fish consumption and
degradation of aquatic wildlife and ecosystems.
-------
Page VI - 31
Problems:
Sources:
PCBs
Heavy metals
Organic chemicals
Nutrients
Municipal and industrial
discbarge
Municipal wastewater
treatment plants
Combined sewer overflows
Ground- and surface-water runoff from
landfills
Contaminated sediments
Connecting Channeb
The five connecting ehannrig of the Great Lakes are designated as international AOCs.
(1). St. Mary's River. The St. Mary's River flows southeast for 68 miles, connecting Lake Superior to Lake
Huron. The entire river is within the AOC Land use along the river is mostly agricultural, forestry, and residential
Contamination problems are mostly confined to the Canadian shore, causing restrictions on water recreational
activities, impairing fish consumption, and aesthetics.
Problems:
PCBs
Heavy metals
Organic chemicals
Bacterial contamination
Sources:
Municipal and industrial
Municipal wastewater
treatment plants
Abandoned waste sites
Contaminated sediment
(2). St Clair River. The St Clair River flows south, connecting Lake Huron to Lake St. Clair. The entire river
is within the AOC Land use differs between the Canadian side and the Michigan side. On the Ontario shoreline
it is primarily industrial with petrochemical industry and refineries. Along the Michigan shoreline it is primarily
agricultural.
The following contaminants are found in sediment, aquatic life, and water, impairing water recreational activities,
and fish consumption.
Problems:
Sources:
PCBs
Heavy metals
Organic chemicals
Bacterial contaminants
Oil and grease
Industrial discharge and spills
Storm sewers
Industrial landfills and
other waste disposal sites
Municipal sewage treatment plants
Combined sewer and overflows
Agricultural and urban runoff
Inactive deep well disposal areas
Contaminated sediment
-------
Page VI - 32
(3). Detroit River. The Detroit River is 31 miles long, connecting Lake St. Clair and Lake Erie. The AOC
is found between Windmill Point and tbe Detroit Light. The largest tributary to the Detroit River is Rouge River,
which is an AOC itself. Land use is residential, commercial and industrial, with a population of approximately four
million people.
Sediment and water have been found to contain the following contaminants, impairing fish and wildlife consumption,
water recreational activities, wildlife habitat, and river bottom aquatic life.
Problems: Sources:
¦ Heavy metals ¦ Industrial discharge,
¦ Organic chemicals storage and spills
¦ Bacterial contaminants ¦ Industrial and municipal
wastewater discharge
¦ Waste disposal sites
¦ Agricultural and urban runoff
¦ Sewage treatment plants
(4). Niagara River. Tbe Niagara River flows 37 miles from Lake Erie to Lake Ontario. The entire river is
within the AOC. Land use is agricultural, residential, and industrial, with textile production, paper product
production, and metal fabrication.
The problems identified below have impaired fish consumption. Additional contaminants, sources, and impairments
will be determined as a cleanup plan is developed.
Problems:
Sources:
PCBs
Heavy metals
Organic chemicals
Pesticides
Cyanide
Industrial discharge
Municipal wastewater
treatment plants
Industrial and municipal
landfills
Urban runoff
(5). SL Lawrence River. The St. Lawrence River leads to the St. Lawrence Seaway which extends to tbe
Atlantic Ocean. The AOC centers around Massena, New York, but includes areas on tbe Ontario shore near
Cornwall and Maitland, and includes Lake St. Francis immediately downstream of Cornwall. Land use in tbe AOC
is primarily residential, commercial, and industrial Tbe following contaminants have been found in sediment and
water, impairing fish consumption and water recreational activities.
Problems: Sources:
¦ Heavy metals ¦ Industrial discharge
¦ Organic chemicals ¦ Municipal wastewater
treatment plants
¦ Combined sewer overflows
¦ Hazardous waste sites
-------
APPENDIX F: NUCLEAR RELEASES
Page VI - 33
This discussion focuses on human health and ecological risk posed by the potential for releases from the IS nuclear
power plants in the Great Lakes Basin.1' Nuclear power plant accidental releases are a high impact, low probability
event. These releases can be caused by a single catastrophic event or a combination of several minor events. For
example, six core meltdown accident types have been identified for the Zioo nuclear power plant by the Nuclear
Regulatory Commission (NRC) including:11
¦ Station blackout;
¦ Loss of coolant accident;
¦ Component cooling water and service water induced reactor coolant pump seal loss of coolant
accident;
¦ Anticipated transients without scram accident;
¦ Interfacing system loss of coolant accident and steam generator tube rupture; and,
¦ Transients other than station blackout
The most likely event is a malfunction of the cooling system, which accounts for more than 75 percent of the
probability of core damage. The extent of damage from a nuclear power plant release is dependent upon:
¦ Hie extent and magnitude of the radiation emission;
¦ The number of people within range of the radiation;
¦ The vulnerability of the ecosystem affected by radiation; and
¦ The amount of radiation taken internally by humans, animals,and plants through ingestion of
radioactive food and Water.
Human Health Risk Assessment
Ionizing radiation refers to radiation that strips electrons from atoms in the medium through which it passes. The
adverse effects of exposure to ionizing radiation, and hence of radioactive materials, are carcinogenicity,
mutagenicity, and teratogenicity. From the perspective of total societal risk, cancer inducing and genetic mutations
are believed to be stochastic effects; i.e., the probability of these effects and the risk of the occurrence increases with
dose, but the severity of the effect is independent of the dose. Furthermore, there is no convincing evidence of a
threshold of exposure below which the risks are zero. Evidence of the deleterious effects of exposure to ionizing
radiation comes from both human epidemiology and animal studies.
The unit used in radiation dose assessment is the tad (radiation absorbed dose). One rad is the dose corresponding
to the absorption of 100 ergs per gram of tissue. Since not all forms of ionizing radiation produce the same effect
per rad, the rem is used as the unit of dose equivalence. For materials taken into the body, the dose will be delivered
over the period that the material remains in the body. Thus, the convention has been established to integrate the dose
over the entire period that the material will remain in the body and assign the total dose to the year of exposure,
resulting in the committed dose equivalent (rem). Finally, since irradiation of the organs and tissues of the body may
1 "Nuclear power pUat ladiatioa data were lot available for tfcia atady for Ike New York aid Peaaaylvaaia portion of Ike Great Lakes
Basia. Coaseqaeatly, actaal risk Caclon as estimated by Ike Naclear Regalalory Commiuioi are prcseated for ladiaaa, IlUaois, Michigaa,
Miaaesota, Okio, aid Wiscoasia, wkile beat profeaaioaal jadgmeat estimates of expotare levels are provided for aaclear power plaatt ia New
York aad Peaaaylvaaia.
"U.S. Naclear Regalaloty Commissioa, Office of Naclear Regalatoiy Reseaick. Severe Accident Risks: Aa Assessmeat for Five U.S.
Naclear Power Plain. Stmmaiy Report Volune 1.
-------
Page VI - 34
not be uniform, the radiation protection community has introduced the concept of the effective dose equivalent (rem
EDE). Hie EDE is calculated by weighting tbe doses received by tbe various organs by risk based factors and then
summing the weighted organ doses to derive the EDE. The collective population exposure is given in person-rem
EDE and is derived by simply summing the exposures of tbe individuals in the population In this report, the doses
are given in rem or millirem EDE for individuals and person-rem EDE for populations. The quantification of
radiation exposures and resulting cancer risks are based an tbe following estimates:12
Lifetime exposure to 3 mrem/year EDE « 1 x 10"3 lifetime fatal cancer risk;
1 x 105 person-rem/year EDE * 400 fatal cancers/year, and
Total Cancer Incidenoc/Fatal Cancer Incidence * 2, a 50 percent mortality rate once a cancer has been expressed.
Tbe risk factors used in this report are consistent with those used by tbe EPA in the recent radionuclide NESHAPS
rulemaking.13 They are based on a linear extrapolation of tbe dose response exhibited by tbe Japanese A-bomb
survivors (rather than from a nuclear accident) and other epidemiologic data using tbe relative risk projection model
and assuming that there is no risk threshold. Tbe EPA believes that the estimated fatal cancer risk of 400 per 1 x
10s person-rem EDE represents a best estimate, and that tbe actual risks likely lie within the range of 120 to 1,200
fatal cancers per 1 x 10s person-rem EDE. For radiation exposure to the whole body, as is thought to occur in a
nuclear accident, the total incidence of cancer does not exceed the incidence of fatal cancer by more than a factor
of two. Tbe risk coefficient has been extrapolated from high dose and high dose rates. At tbe lower doses, tbe
current consensus of scientific opinion is that no zero risk threshold exists.
The EPA's Office of Radiation Programs has estimated the exposures to both nearby individuals and the populations
within 80 km of nuclear power plants under likely nuclear contaminant failure scenarios.14 Tbe estimates that are
presented in Table 1 are derived from EPA's estimates and only include exposure to emissions released to air.
Exposure to radioactive materials via liquid pathways is not estimated, but is roughly comparable to exposures to
radioactive materials released to air from industrial sources. Additionally, these ****matt* only represent commercial
nuclear power plants in tbe Basin and do not include Department of Energy or Department of Defense facilities
which also release radiation or other manmarte enhanced sources.
Hie estimates for exposure of the general population to industrial sources are based on both site-specific assessments
and extrapolations from reference facilities. Where reference facilities provide tbe basis, the site name is marked
with an asterisk. For actual facilities, tbe exposure of tbe maximally exposed individual reflects either an actual off-
site residence, or tbe fenccpo6t exposure reference facilities where enhanced radiation is used. Tbe maximum
exposure is based on an individual assumed at a close location in tbe predominant wind direction. Where the original
assessment used a reference facility, collective populations are using the generic population distributions
that were assessed and tbe number of facilities in the Basin. If the projected population obtained in this manner
exoeeded the Basin population, the population at risk was constrained to tbe regional population.
"U.S. Eaviroaaeatil Protection Ageaey, Eavtro«me«t»l Imract Stitemett-NESHAPS for R»dk>«»cllde»: Bjckgromd Itfonnttiot
Pocameat Volame I: RUt Attettmeat Methodology. EPA 520/1-89-005. Office of Radialkw Program*, September, 1990.
"Ibid.
'*Thi» MtbMie to ia dinci coauadlctioa to the NRC't lately goala which estimate tuiy hullty ri*k withia oae Bile «(tbe tile tad Uteat
ctacer CiUUty ride withia 10 mile* of the tile.
-------
TABLE 1
INDUSTRY/
SITE
AVERAGE
EXPOSURE
in rem/year
POPULATION
AT RISK
EXPOSURE
in person-rem 1
year
FATAL
CANCERS
PER YEAR
TOTAL
CANCERS
PER YEAR*
Zion 1
0.0002
5898100
1179.62
4.72
9.44
Zion 2
0.0002
5898100
1179.62
4.72
9.44
Cook 1
0.0002
318200
63.64
0.25
0.51
Cook 2
0.0002
318200
63.64
0.25
0.51
Fermi 2
0.0002
4098200
819.64
3.28
6.56
Palisades
0.00009
329800
29.68
0.12
0.24
Davis-Besse 1
0.00009
579100
52.12
0.21
0.42
Perry 1
0.0002
1928500
385.70
1.54
3.09
Point Beach 1
0.0002
310700
62.14
0.25
0.50
Point Beach 2
0.0002
310700
62.14
0.25
0.50
Big Rock Point
0.0002
91000
18.20
0.07
0.15
** Kewaunee
0.0002
26900
5.38
0.02
0.04
* * Fitzpatrick
0.0002
216900
43.38
0.17
0.35
**Nine Mile Point 1
0.0002
216900
43.38
0.17
0.35
**Ginna
0.0002
216900
43.38
0.17
0.35
TOTAL
V :
16.21
32.41
* Does not include occupational exposures for power plant workers.
* * Best professional judgment estimates for average exposure at these plants.
-------
Page VI - 35
Ecological Risk Assessment
The adverse effects associated with low-levels of radioactivity in the environment include cancer, genetic effects and
birth defects. Such effects, even if extremely rate or undetectable, are of concern to humans. However, for
organisms other than man, the concern is not with individual organisms, but with the viability of the species and the
function and structure of the ecosystem as a whole. The following briefly summarizes current research which has
demonstrated that low level radiation may be considered inconsequential in terms of its potential ecological effects.
However, a short term high level dose may have considerable impact on certain ecosystems.
During the 1960s and 1970s, a vast amount of radiobiological research was performed to assess the impacts of
radiation on plant and animal communities. The research included a large number of comprehensive laboratory and
field studies motivated primarily by concern over fallout from weapons tests. A literature review was prepared by
the Office of Radiation Programs in 1986.u
Overall, it appears that prolonged exposures to ecosystems below a few rad per day, results in no detectable adverse
ecological impacts. Though the community interactions to prolonged exposures to ionizing radiation are complex
and difficult to predict, doses on the order of several hundred tads per year would be needed to cause extinction of
a species. Such exposures can occur following a major nuclear accident, but are not associated with the production
and use of radioactive materials. Nor are they associated with uncontrolled sites where previous activities have
resulted in contamination of the site. It should also be noted that welfare impacts are high for ground-water clean
up as well as species preservation, should a mid-level accident occur.
Welfare Assessment
The potential welfare effects associated with radiation exposure can be divided into two broad categories:
¦ costs associated with effects on human health, and
¦ costs associated with commercial damage.
The costs associated with health effects include direct medical costs and low productivity due to the inability to
conduct normal work activities. The 1968 Pannpy ffarts and Figures, published by the American Cancer Society,
estimates that for 1985 the total economic cost of cancer was $71.5 billion. This includes direct medical costs and
indirect costs associated with lost productivity. The American Canoer Society estimates that there were 985,000 new
cases of cancer in the United States in 1988. Over the past 30 year period, the per capita age adjusted cancer death
rate has increased at a rate of less than one percent per year, therefore, the estimate of 985,000 cancers can be used
to estimate the approximate cost per cancer. Escalating the cost by 7.5 percent per year for health care service costs
and assuming the cancer incidence remains virtually unchanged, results in an economic cost per cancer case in 1990
dollars of approximately $100,000 on average. The lower bound estimate based on Hartunian, et al is $80,000.00,
while the upper bound estimate given by the American Cancer Society is $137,000.00.
Health care costs estimate can be obtained using the following formula:
(Annual Cancer Cases)(Dircct Costs and Foregone Earnings) « SHealtb Costs.
As described above, the total number of cancers in the Basin that result from nuclear power plant operations and
likely release scenarios is 32.41. The upper and lower bound costs resulting from these releases are w follows:
"U.S. Eavirouneattl Pfolectkw Aacscv. Effect* of Ridlidoi o» Aomtc Oimlm r j Ridlobiolorictl Mettodoloda far Effect!
AwcMwest. EPA S20/1-8S-016. Office of R*di*tioa Progrisu, Wafkiagloa, D.C, Fcbruiy, 1986.
-------
Page VI - 36
The lower bound estimate is:
(32.41)(S80,000) = $2492,800.00
The upper bound estimate is:
(32.41X5137,000) = $4,401,700.00
The costs associated with commercial damage caused directly by radiation are variable. The contamination of
facilities and sites where radioactive materials have been or are produced and used, can result in considerable ripgm'p
costs. For commercial facilities, the costs of decontaminating and decommissioning the facilities and the sites are
reflected in the costs of the products or services. For sites owned by government agencies, the costs will be borne
by the taxpayers. Restoration of the sites operated for the Department of Energy has been intimated. Current
estimates place these restoration costs in the hundreds of billions of dollars. Whether or not such costs will actually
be incurred is uncertain at this time, and no estimate is made of the costs on a Basinwide basis.
-------
Page VI - 37
APPENDIX G: MERCURY TRENDS IN THE GREAT LAKES BASIN
This section addresses the treads in mercury concentrations in the Great Lakes Basin and the resulting potential
effects of mercury contamination. Historical trends indicate that mercury concentrations have been increasing in fish
and sediments from inland lakes in Michigan, Minnesota, and Wisconsin. These increases have been attributed to
atmospheric deposition of mercury.
Problem Area Description
Mercury can cause adverse effects on human health and ecological receptors at low concentrations, although the
toxicity of mercury is dependent on its chemical state. Exhibit AH-1" presents the chemical forms of mercury and
their principal characteristics.
EXHIBIT AH-1
CHEMICAL FORMS OF MERCURY
Molecular Form
Chemical and Biochemical Properties
Hg
Pigmental mercury. The vapor is highly toxic when inhaled;
less toxic via direct ingestion.
Hg+
Mercurous ion. Insoluble as the chloride; low toxicity.
Hg*
Mercuric ion. Highly toxic via ingestion and inhalation routes;
may be converted to methyl mercury.
w
Organotnercurials. Extremely trade, especially CHjHg*
(methyl mercury); causes irreversible nerve and brain damage;
easily transported across biological membranes; stored in fat
tissue.
R*Hg
Diorganomercurials. Moderately toxic bid often converted to
RHg\
HgS
Mercuric sulfide. Highly insoluble and nontoxic. Mercury is
trapped in soil in this form.
As presented in the exhibit, some forms of mercury may pose lower toxicity because they are insoluble and therefore
are not readily absorbed. However, in aquatic environments, the most toxic form of mercury, methyl mercury, if
often produced from elemental mercury.17 For instance, elemental mercury can be oxidized to mercuric ion.
Bacteria found in sediments then transform this inorganic mercury into methyl and dimethyl mercury. These
compounds are highly toxic and bioconcentrate readily in the aquatic food chain to a greater degree than do their
inorganic precursors.
Historically, principal wastewater sources of mercury pollution in the Great Lakes Basin have been paper mills and
" Spiio, T.O. a»d W.M. SttglUmi, 1980. Bavlreameittl Scitace fa fennecrivt. p. 204.
17 EPA, 1984, Anbieit Water Qulitjr Criteria for Meicaiy. Office of Water Ragalatioaa aad Suidiixk. EPA440/S-84-026, PB85-
227452.
-------
Page VI-38
chloralkali plants. The combustion of fossil fuels, particularly coal, is considered to be a primary source of
atmospheric mercury.1* Other atmospheric sources are mining and smelting operations, involving mercury and
copper ore, and solid waste incineration.1' Same analyses have reported that mercury from paint was a major
atmospheric source. In 1990 EPA banned the use of mercury in indoor paints, and is phasing out all mercury in
outdoor paints. The weathering of rocks containing mercury also provides a background level of mercury in aquatic
systems."
Magnitude of the Problem
The existence of high concentrations of mercury poses a number of threats to human health and the environment in
the Great Lakes Basin. Among these threats arc:21
¦ human nerve and brain damage from exposure to bioconcentrated levels of mercury in fish and
other game species;
¦ restrictions on fish and wildlife consumption;
¦ restrictions on dredging;
¦ degradation of fish and wildlife populations;
¦ deformities and other reproductive problems in birds or animals feeding on fish (bald eagle, osprey,
loon, mink and otter)
¦ lower species diversity.
The typical pathway of exposure to methyl mercury is from mercury deposited to waters, from either atmospheric
or point sources, which then settles to sediments where the methylation process described above occurs. The
resulting methyl mercury is then bioaccumulated and passed up the food chain in ever increasing concentrations.
The biocoocentration factor has been estimated to be about 80,000.B As top predators, large game fish and humans
are ultimately exposed to the highest mercury concentrations.
Point source discharges of mercury in the Basin are tracked through EPA's Permit Compliance System (PCS).
Although the available data do not support a detailed analysis, the lower Great Lakes appear to receive a higher load
of mercury releases from municipal sources (See Exhibit H-l). Mercury discharged from municipal sources to Tair^
Michigan and Erie represents 95 percent of the mercury discharged to the entire Great Lakes Basin.
A second indicator of current and past mercury loadings is the sediment concentrations of mercury observed in the
Basin. STORET 405b data summarizing sediment concentrations were obtained from sampling stations located in
32 harbors throughout the Basin. The average mercury concentration for all locations is 0.63 ppm. The maximum
value, 24.0 ppm, was reported in Erie County, New York. Wayne County, Michigan, which includes the Detroit
area, recorded several mercury concentration readings of 3.00 ppm to 3.60 ppm. Brown County, Wisconsin, which
includes Green Bay, reported concentrations ranging from 0.40 ppm to 3.00 ppm.
As expected, those areas reporting higher point discbarges of mercury also exhibit higher sediment concentrations
of mercury, with urban areas around Tab! Erie and its connecting rhanneu, such as Buffalo, Cleveland and Detroit
" Liadbeig, S.E., 1984. Emiwioa aid DepoiMoB of Atnwpkcric Mereuy. Oak Ridge Natkmal Laboratory. DE95006304.
" EPA. 1987. Toxic Air PoUrioa/Soucc Croaswalk, EPA 450/447423a.
" Gavis, J., Ferguoi, J.F. 1972. The Cycliag of Mercuy Tkroigi tke Eaviroameat, WaterJRxs., 6.-986-1008.
" Ageacy for Toxic Sabsttaces tad Disease Registry, Toxico logical Profile for Marcaiy, Dec., 1989.
» Op CiL, EPA 1984.
-------
Page VI-39
exhibiting high sediment concentrations, along with the Green Bay area cm Lake Michigan.
The significance of these sediment concentrations can best be assessed by observing the resulting bioaccum illation
impacts on Great Lakes Basin fish. Mercury contamination was found in virtually all areas sampled for fish, with
levels ranging from 0.1 ppm to over 0.4 ppm.8 This is not unexpected given the high sediment mercury levels
reported in most of the sampled areas. The area with the lowest mercury levels was Isle Royale, located in a remote
area of Lake Superior, which had fish samples with mercury levels below 0.1 ppm. In contrast, fish samples taken
in the Green Bay area showed higher concentrations with mercury levels greater than 0.4 ppm. These figures
generally coincide with the patterns observed in mercury discharge levels from point sources and sediment
concentrations.
Most recently, there have been reports suggesting an upward tread for mercury in fish. Studies conducted on fish
from lakes in upper Minnesota show mercury concentrations increasing 5% per year since the 1970s.14 Similarly,
sediment concentrations have shown a three fold increase since the early 1800s." The lakes chosen for study have
no known point sources of mercury. A survey of 80 lakes revealed mercury concentrations averaging 15 ppt
mercury.* In contrast, fish collected from Minnesota rivers during the 1970s were reported to show a decline,
attributed to increased regulation of point sources.27
These findings indicate that the atmosphere is a significant source of mercury. Research conducted in northern
Wisconsin Lakes has also shown that mercury concentrations are greatest in the uppermost portion of sediment
cores.1* It is not clear if atmospheric input of mercury has been increasing over the past decade. Atmospheric
inputs in Minnesota are reported to be in the range of 18-26 ug/mJ yr* Lakes with highest levels of mercury
contamination tend to be those of lower pH (less than 7.0) and lower buffering capacity* A possible relationship
with increasing mercury levels in fish and increases in the acidity of rain could exist Further research, establishing
trends in atmospheric loadings and lake chemistry, is needed.
Birds and mammals which consume fish, in particular, bald eagles, looos, ospreys, herons, mergansers, minks and
otters, have also been shown to have elevated levels of mercury in Wisconsin." This indicates that mercury is
passed upward through the food chain. Research is also being conducted on fish-eating birds from Lakes in
Northeastern Minnesota.
" Op. OC, EPA, 1990.
" Smli, B.B. ud DD. He]wig. 1969. Meicaiy ia Flsk bom Northerner* MiMNk Like*. Miaaeeon Poliitioa Coatrol Ageacy.
" Heeaiag, el el. 1989. HiXorical ead Aieel DepWtioa of Mercery ie RE Mieaeeol* aid Northers Wbcoaali Lake*. Report to Ike
Miaaeeon Poliitioa Coatrol Ageacy.
* Soieaiea, et iL 1989. Aiibone Mercery Depoeitioe ead Weteiafced ChiiarteriHc* ia Relatioaihip to Mercery Coaceatntioei la
Wiier, Sedlneaft, PUikioa aad Fish of Eighty Northen Miaaeeott Lake*. Report to (he Mlaaeeote PoUattoe coatrol Ageacy.
" Op at, Swili ead Helwig, 1989.
" Beeedey, CD., Penoaal Letter of ConBialcatioB to WJL Reilly, April 27,1990.
* Op. ClL, Heeaiag, et el., 1989.
" Op. ClL, Soraiea, et eL, 1989.
u Op. ClL, Beeeday, 1990.
-------
Page VI - 40
Human Health Impacts
Due to bioaccumulation of mercury in fish, fish consumption is considered the primary threat to human health (i.e.,
a greater threat than drinking water consumption). The chronic oral Reference Dose for mercury is 0.0003 mg/kg-
bw-day.32 Therefore, adults may have an intake of 21 ug/per day over their lifetime without a significant potential
for adverse health impacts. Risk assessments by Great Lakes State health departments have established sportfish
consumption advisories from 0.16 to 0.5 ppm to assure adequate protection for heavy fish consumers, pregnant
women, and children.13 The Food and Drug Administration has established an action level of 1.0 ppm mercury
for fish entering interstate commerce.34
Following extensive surveys of mercury in sportfish from hundreds of lakes, starting in 1989, the States of Michigan,
Minnesota, and Wisconsin began expanding the number of lakes under fish consumption advisories for mercury.M
The May, 1991 advisory for Minnesota, lists 310 (95%) of 327 inland lakes on fish advisories due to mercury. In
1990 Wisconsin placed 170 (29%) of 600 lakes on a fish consumption advisory due to elevated levels of mercury.
Three out of four lakes (a total of 170) tested in Michigan were found to have mercury levels exceeding 0.50 ppm
in at least one fish species, prompting the State to issue an advisory for women and children not to consume more
than one meal per month of fish from all inland lakes.
For an individual consuming 150g/day of fish having 05 ppm of mercury, over several days or weeks, the average
mercury dosage would be 0.001 mg/kg-bw-day. When oompared to 14 day health advisory values of 0.001 mg/kg-
bw-day derived from animal studies, no adverse impacts would be expected. For long-term exposures, and those
consuming 100 g/day (upper 90% intake far spoitfishermen), dosages would be 0.0007 mg/kg-bw-day. When
compared to EPA's Reference Dose of 0.0001 mg/kg-bw-day, a Hazard Index of 7.0 results. This indicates increased
risk for the heavy consumer of sportfish. For an average sportfishermen in the Great Lakes consuming 19 g of fish
per day a Hazard Index of 13 results, indicating that the average sportfishermen are not at an advene risk, as long
as they follow fish advisories.
Ecological Impacts
As discussed in the introduction above, mercury contamination can cause serious ecological impacts. These impacts
range from reproductive impairment and deformities in offspring to lower species diversity and other adverse
ecosystem impacts. Direct aquatic toxicity from mercury appears unlikely in inland lakes as water concentrations
of about 1 ppb are required before chronic effects on zooplankton and fish occur.3* The highest values reported
in Minnesota waters was 7 ppt.
However, with biotnagnification, low part per trillion levels in water, result in fish concentrations of 0.1 to 1 ppm.
Studies on loons and other waterfowl in Minnesota and Wisconsin indicate elevated levels of mercury, with
concentrations of 20-80 ppm found in livers of diseased loons.37 Noo-diseased loongs were found to have
n US. EPA. 1991. Utegnled Risk Ufonnatioi System.
" Sate Ft4 Advisories for Miaiesota, Mickigu, Wlsoisii, 1989-1991.
M Op. dL, Ageacy for Toxic SsbsUices tad Dheue Registry, 1989.
" Op. ClL, Fish advisories.
* Op. at, EPA, 1984.
97 Pcnoul commiiicjtio* with Kirei Lsnti of the Misiesot* PoUstioi Cottvol Afeicy, 4/19^1.
-------
Page VI - 41
significantly lower concentrations of mercury. Threshold effects for impacts on mink occur at about 1.1 ppm in
fish.3* Hoffman* found that for quail, a concentration dose of 0.2 to 8 ppm of mercury in feed, caused eggshell
thinning to occur. Therefore, it is dear that fish-eating wildlife are currently exposed to mercury at levels of
concern.
The mercury toxicity reference values for aquatic species have been set at:
¦ EPA Acute Freshwater Ambient Water Quality Criteria (ug/L) * 2.4.
¦ EPA Chronic Freshwater AWQC (ug/L) = 0.012
(for bioaccumulation in fish and human consumption)
Summary of Human Health and Ecological Impart*
Given the high concentrations of mercury at many locations in the Great Lakes Basin, and the fact that mercury may
be increasing due to atmospheric deposition, pervasive mercury levels are a current and future threat to the health
of the Great Lakes Basin ecosystem and indigenous wildlife. The extensive State fish advisories for mercury in
Minnesota, Michigan, and Wisconsin, affecting over 500 lakes, indicates the magnitude and significance of this
emerging problem. Persons consuming large amounts of fish from inland Lakes, in particular those not following
State issued fish advisories, are at increased risk of reproductive and neurological impacts. Similarly, fish eating
wildlife are also subjected to these adverse health impacts from mercury. Continued research on mercury sources
and trends should be considered a high priority.
» Op. at, EPA, 1984.
- Hoffota, D J. 1965. Morphological aid pkjniologfcal napoaaca to aaviioaaeatil eoaiamiaiak la Mid tabryos tad ---"'-g-
Novenbar 1985, SETAC Abaftacto, VS. FU aad Wildlife Rcaeaick Service, Liard, MD.
------- |