United States
Environmental Protection
Agency
Research and
Development
(8101R)
EPA 600-R-99-106
December 1999
www.epa.gov
&EPA
Research and Development
Fiscal Years 1997-1998
Research Accomplishments
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
The mission of the U.S. Environmental Protection Agency (EPA) is to protect human health and
safeguard the natural environment air, water, and land upon which life depends. To meet this
challenge, EPA has developed a strategy that combines strengthening the current system of
environmental regulations and designing approaches to provide better environmental protection at
less cost. The success of this strategy depends in large part on making credible environmental
decisions based on sound science. Working in partnership with EPA's program and regional offices,
EPA's Office of Research and Development (ORD) is dedicated to developing the scientific
knowledge and innovative technological solutions needed to ensure the success of this strategy to
achieve EPA's mission.
As EPA's Assistant Administrator for Research and Development, I am proud of the expertise and
dedication ORD's scientists, engineers, and other personnel bring to addressing the environmental
challenges of today and tomorrow. ORD's researchers are expanding our nation's scientific
knowledge about the environment, developing guidance for assessing both human health and
ecological risks, devising new technologies and risk management approaches to both prevent and
mitigate pollution, and providing technical assistance to those working to protect our environment.
All of our work is guided by sound scientific principles, including independent peer review, to ensure
that our contributions are consistently of the highest quality.
This report communicates ORD's most significant research accomplishments during 1997 and 1998
in support of EPA's mission. A few example highlights include:
> Development of improved methods for detecting drinking water contaminants to reduce
outbreaks of illness.
^ Creation of computer-based models of long-range transport of air pollution that can be used
by states in designing pollution control programs.
> Demonstration of strategies to reduce children's exposure to lead in the home, a risk factor
for impaired nervous system development.
^ Evaluation of the potential of chemicals to interfere with the endocrine system of humans
and wildlife.
> Development of advanced methods for monitoring the condition of the environment, using
data sources such as satellite imagery and ground-based ecological measurements.
I hope you will take the time to read about the significant ways ORD's research contributes to our
understanding of environmental issues, especially in the areas of particulate matter, drinking water,
risk assessment, ecological assessment, and endocrine disrupting chemicals. We in ORD
remain committed to bringing our creativity and technical expertise to meeting the highest priority
science needs in support of EPA's mission.
9,
Norine E. Noonan, Ph.D.
Assistant Administrator
for Research and Development
-------
Contents
Introduction...
1.
Health Effects of Airborne Paniculate Matter
....5
2. Drinking Water: Micronial Pathogens and Disinfection Dy-Products 13
3. Advances in Risk Assessment 21
4. Mid-Atlantic Integrated Assessment [MAIA) 29
5. Endocrine Disrupters 35
6. Environmental Risks to Children 43
7. Harmful Algal Dlooms 45
8. Pollution Prevention 47
9. U.S.-Mexico Dorder Environmental Health 49
10. Monitored Natural Attenuation 51
11. Global Change 53
12. Arsenic in Drinking Water 5 5
13. Economic and Decision-Making Research 57
14. Ecological Indicators 59
-------
-------
Introduction
As we get up each morning and get ready for
another day, we often take for granted that the
water from the faucet we use to make our coffee is
drinkable, the food we feed our children is safe, and
the air we breathe is healthful. Not until we need to
take unusual steps, such as boiling our water to
eliminate microbial contamination or reducing
physical activity on ozone action days, do we
realize how fragile our environment can be and how
important it is to protect it. The mission of the U.S.
Environmental Protection Agency (EPA) is to
protect human health and to safeguard the natural
environment - air, water, and land - upon which life
depends. Although EPA has made substantial
progress in both cleaning up and protecting these
natural resources, many challenges remain, and new
human health and environmental problems continu-
ally confront the Agency.
Responding to these challenges requires research
to understand these problems and to develop
technologies to solve them. The Office of Research
and Development (ORD) is the primary arm of EPA
responsible for carrying out this work. ORD
supports EPA's mission by conducting state-of-the-
art research in all environmental media (air, water,
and land) to address unanswered scientific
questions. As part of its research efforts, ORD
develops innovative methods and approaches for
solving problems that range from broad scientific
issues such as global climate change, to specific
problems such as removing microbes from drinking
water systems. To make the most efficient use of
resources and research dollars, ORD relies on
extensive collaboration among ORD Laboratories
and Centers, EPA partners in the program and
regional offices, and the external scientific commu-
nity. In addition, ORD has made independent peer
review an integral part of its programs to ensure
ORD's research is of the highest quality.
This Research Accomplishments Report highlights
selected accomplishments completed by ORD
during Fiscal Years 1997 and 1998. We hope this
report will help you better understand the environ-
mental problems all of us face, and the role of EPA's
research in helping to solve them. The report
features significant accomplishments that have
advanced our scientific and technical knowledge
and capabilities, as well as ongoing activities that
will generate future accomplishments. Accomplish-
ments in five broad research areas are described in
depth in the report (main topics), and accomplish-
ments in nine other research areas representing
either emerging environmental issues or issues
narrower in scope are described more briefly
(shorter topics). In each topic area, the major
scientific problems are defined, the scope of ORD's
research program to address these problems is
outlined, selected accomplishments are highlighted,
and the direction of future research is described.
The topics are:
Main Topic Areas:
1. Health Effects of Airborne Paniculate Matter
2. Drinking Water: Microbial Pathogens and
Disinfection By-Products
3. Advances in Risk Assessment
4. Mid-Atlantic Integrated Assessment (MAIA)
5. Endocrine Disrupters
Shorter Topic Areas:
6. Environmental Risks to Children
7. Harmful Algal Blooms
8. Pollution Prevention
9. U. S .-Mexico Border Environmental Health
10. Monitored Natural Attenuation
11. Global Climate Change
12. Arsenic in Drinking Water
13. Economic and Decision-Making Research
14. Ecological Indicators
The research accomplishments presented in this
report represent only a snapshot of ORD's dy-
namic, evolving research portfolio. Nonetheless,
these accomplishments describe some of the key
research that supports important EPA decisions,
and the report presents ORD research in a compre-
hensive, integrated format that is an alternative to
the more focused reporting of research results in
the scientific literature and other publications.
-------
Introduction
Developing and evaluating
approaches to prevent or
reduce risks from
environmental stressors
Describing and estimating
the risks to humans or
ecosystems from exposure
to stressors
Credible
Environmental
Research
For any research organization
to have credibility within the
scientific community, its
research must be able to
withstand the rigors of
scientific scrutiny. ORD has
made the peer review process
an integral part of its research
program to ensure that its
research is based on sound
methodologies and generates
credible data. Peer review is
an independent evaluation of
a work product by experts
who have not participated in developing the work
product. Peer review can be internal (evaluation by
experts within EPA), or external (evaluationby
independent experts outside of EPA, such as EPA's
Science Advisory Board). ORD recognizes the
importance of its research to both the Agency (for
use in regulatory decisions) and the scientific
community (for application to specific environmen-
tal problems), and strives to conduct the best
science possible.
The Risk Paradigm
To understand ORD's research program, it helps to
be familiar with the "risk paradigm," an important
Agency organizing principle. The risk paradigm
consists of two interrelated phases, risk assessment
and risk management. Risk assessment is the
process used to evaluate the degree and probability
of harm to human health and the environment from
such stressors as pollution or habitat loss. The risk
assessment process, as proposed by the National
Academy of Sciences (NAS) in 1983, consists of:
Exposure Assessment - describing the popula-
tions or ecosystems exposed to stressors and
the magnitude, duration, and spatial extent of
exposure
Hazard Identification - identifying adverse
effects (e.g., short-term illness, cancer) that may
occur from exposure to environmental stressors
Dose-Response Assessment - determining the
toxicity or potency of stressors
Understanding how populations and
ecosystems are exposed to
environmental stressors
Identifying adverse effects
from exposure to stressors
(hazard identification)
Determining the toxicity or potency
of stressors (dose-response)
Figure 1. The risk assessment-risk management framework used by
ORD to organize its research and development activities.
Risk Characterization - using the data collected
in the first three steps to estimate and describe
the effects of human or ecological exposure to
stressors
Risk management entails determining whether and
how risks should be managed or reduced. It is
based on the results of the risk assessment as well
as other factors (e.g., public health, social, and
economic factors). Risk management options
include pollution prevention or control technolo-
gies to reduce or eliminate the pollutant or other
stressor on the environment. The environmental or
public health impacts resulting from risk manage-
ment decisions must then be monitored so that any
necessary adjustments can be made. A simple
diagram of this cycle of risk assessment and risk
management is shown in Figure 1 (with the steps of
hazard identification and dose-response assess-
ment combined into a category entitled Effects
Assessment).
ORD has aligned its organizational structure to
comport with this risk paradigm and has made the
principles central to its strategy for determining
priorities for environmental research. Several topics
in the report are presented in terms of the risk
paradigm, with accomplishments linked to discrete
components of the paradigm. For other topics the
risk paradigm was not used, either because the
accomplishments could not be cleanly divided
among the different risk steps or because most of
the accomplishments fell within a single step of the
paradigm.
-------
Introduction
ORD Laboratories and Centers
ORD comprises three National Laboratories and
two National Centers. Most of the National
Laboratories and Centers have multiple research
facilities (Figure 2). The brief descriptions of the
ORD Laboratories and Centers listed below include
which aspect of the risk paradigm they support
(italics).
National Health and Environmental Effects
Research Laboratory (NHEERL) (www.epa.gov/
nheerl/) conducts research on the effects of
contaminants and environmental stressors on
human health and the environment. Hazard
identification and dose-response assessment
National Exposure Research Laboratory
(NERL) (www.epa.gov/nerl/) conducts research
to improve the scientific bases for human and
ecosystem exposure assessment. Exposure
assessment
National Center for Environmental Assessment
(NCEA) (www.epa.gov/ncea/) conducts research
in risk assessment methods, and serves as a
national resource for human health and ecologi-
cal risk assessment by conducting assessments
and developing new methods and tools for risk
management. Risk characterization
National Risk Management Research Labora-
tory (NRMRL) (www.epa.gov/ORD/NRMRL/)
conducts research and technology transfer to
prevent, mitigate, and control pollution. Risk
management
National Center for Environmental Research
and Quality Assurance (NCERQA)
(www.epa.gov/ncerqa/) manages an extramural
research program (grants, fellowships, and
national centers of excellence) known as Science
to Achieve Results (STAR) to complement
ORD's internal research program and expand
EPA's science and technology base. NCERQA
also develops EPA-wide quality assurance
policies and manages EPA's peer review process.
All phases of risk assessment and risk manage-
ment
With this approach and organizational structure,
ORD can assure that science resources are directed
to the most pressing environmental problems
posing the greatest risks to people and the environ-
ment. We will continue to bring our creativity and
technical expertise to meet the environmental
science needs of today while positioning ourselves
to identify and aid in resolving the environmental
problems of tomorrow.
Newport, OR
NHEERL
Coruallis, OR
NHEERL
DuluthMN Grosselle,MI Cincinnati, OH
NHEERL \ NHEERL \ NERL
NRMRL
NCEA
Las Vegas, NV
NERL
Ada, OK
NRMRL
Gulf Breeze, Ft
NHEERL
Narragansett, Rl
\ NHEERL
Edison, NJ
NRMRL
Washington, DC
NCEA
NCERQA
NRMRL
Research Triangle Park, NC
NERL
NHEERL
NRMRL
NCEA
Athens, GA
NERL
NCEA National Center tor Environmental Assessment
NCERQA National Center lor Environmental Research and Quality Assurance
NERL National Exposure Research Laboratory
NHEERL National Health and Environmental Effects Research Laboratory
NRMRL National Risk Management Research Laboratory
Figure 2. ORD Locations.
-------
-------
1. Health Effects of Airborne
Paniculate Matter
In December 1952, a choking black fog enveloped
London, darkening the city during the daytime.
Unusual weather conditions trapped a mass of
stagnant air that filled with the smoke from tons of
burning coal. By the time the air cleared, an
estimated four thousand residents of the metro-
politan London area had died, with the elderly and
people with heart and respiratory ailments espe-
cially affected. Studies that have looked back on
the event have concluded that tiny, inhalable
particles found in the smoke - also known as
paniculate matter (PM) (Figure 1-1) - played a
crucial role in the deaths. Similar events, though
less catastrophic, occurred in the United States and
other parts of Europe during the middle decades of
the century.
In the nearly 50 years since the London episode, air
quality has improved dramatically in the U.S. and
Western Europe. One of the major reasons for the
improvement seen in the United States is the Clean
Air Act, which had its beginnings in the enactment
of this nation's first federal air pollution law, the Air
Pollution Control Act of 1955. Since then, the Clean
Air Act has been amended many times, leading to
strengthened regulatory programs to reduce particle
emissions from power plants, motor vehicles, and
other combustion sources. Clean Air Act require-
ments have also contributed to a vastly improved
scientific ability to detect and understand health
effects from particles and other air pollutants.
A key section of the Clean Air Act Amendments
requires EPA to review the risks to public health
and welfare from PM (and other maj or pollutants)
every five years to determine whether to revise air
quality requirements known as the National
Ambient Air Quality Standards (NAAQS). This
continual cycle of NAAQS review assures that the
Agency considers the most recent scientific
research as it decides whether to revise a current
standard. To implement this provision, ORD is
responsible for periodically preparing comprehen-
sive Air Quality Criteria Documents that present the
analysis, review, and assessment of the latest
available scientific information on maj or air pollut-
ants. These documents serve as written consulta-
tions on the current state of the science for use in
environmental decision-making by risk managers in
EPA's Office of Air and Radiation.
Figure 1-1. Examples of fine particles. Airborne
particles appear in a variety of sizes and shapes
and differ in their composition. They also can grow
in size in the atmosphere as additional materials
condense on or collide with the particles. These
particles were captured in samples of outdoor air
from Washington, DC.
EPA's most recent review of the PM NAAQS was
supported by ORD's 1996 document entitled Air
Quality Criteria for Particulate Matter (EPA 600/P-
95/00 laF). This document explained that particles
can be comprised of many chemicals, such as
organic material, acids, metals, and oxides. It also
noted that PM consists of particles in a range of
sizes, commonly differentiated by fine PM or PM2 5
-------
1. Health Effects of Airborne Paniculate Matter
(particles less than 2.5 micrometers in diameter), and
coarse PM (between2.5 and 10 micrometers in
diameter). The report concluded that fine particles
were more consistently associated with health
effects than coarse particles. Based on the ORD
report and other findings, EPA published a Staff
Paper estimating that current
PM levels may cause tens of
thousands of premature
deaths each year as well as
hundreds of thousands of
cases of hospital admissions,
aggravation of asthma, and
other health effects.
efforts to reduce emissions can target these
sources. As described in the Introduction, ORD
uses the risk assessment/risk management frame-
work to organize its research approach in solving
scientific and technological problems. This
framework is visually portrayed in the inner ring in
Developing and evaluating
approaches to prevent or
reduce PM emissions
After considering ORD's
review and extensive public
comments, EPA revised the
NAAQSforPMinl997. The
most significant change was
the addition of a new
standard to protect the public
from PM2 5, which can
penetrate deeply into the
lungs. The prior standards
were based on PM10, which
encompassed all particles
less than 10 micrometers in
diameter. (As of 1999, the new ^^^^^^^^^_
standards were the subject of
a lawsuit that may result in changes to the PM
NAAQS.)
Given the potential magnitude of health risks, the
need for better understanding of PM, and Clean Air
Act review requirements, ORD is carrying out a
major PM research program. This research will be
essential for EPA's reviews of the PM NAAQS
scheduled for 2002 and 2007. Ultimately, the goal of
EPA's research program is to provide the scientific
and technological basis for developing and
implementing the Clean Air Act standards so that
all Americans enjoy clean, safe air.
ORD's Research Program to
Answer Outstanding Scientific
Questions
Although ORD has made considerable progress in
understanding the health impacts of PM, substan-
tial uncertainty still exists. With each review of the
NAAQS, research fills gaps in our knowledge and
refines unanswered questions (or raises new ones).
Finding answers to outstanding questions is
important because of the implications for protecting
the public. If certain emission sources are found to
be most responsible for health effects, for example,
Understanding the sources of
particulate matter
Understanding the long-
distance transport of PM
Assessing particle doses in
individuals with respiratory
disease
Understanding the loxicity of PM
Understanding the health
effects of PM, especially for
sensitive groups
Figure 1-2. Some of the major research and assessment activities
conducted by ORD in examining the health effects of airborne particulate
matter.
Figure 1-2, with ORD's major PM research and
assessment activities depicted by the outer ring.
ORD has focused on the following types of
questions in these research areas over the past
several years:
Exposure: Who is exposed to PM? What are the
characteristics of the particles people are
exposed to? How much PM are they exposed to?
Health effects: What are the health effects of
exposure to PM? Who is affected by exposure
to PM? What levels cause adverse effects?
What are the characteristics of particles respon-
sible for adverse health effects?
Risk characterization: What are the overall
risks to the public given exposure and potential
health effects? What uncertainties remain in the
research data?
Risk management: What are the major sources
of PM in the atmosphere? What are the most
cost-effective ways to reduce or prevent the
risks associated with exposure to PM?
The scientific products of ORD's PM research, as in
the rest of EPA's research program, are closely
scrutinized through independent peer review to
ensure they are of high quality, credible, and
-------
1. Health Effects of Airborne Paniculate Matter
advance the state-of-the-science for PM. In
addition, ORD strengthens its PM program and
reduces duplicative efforts through partnering/
collaborating with many of the public and private
organizations that are also conducting PM re-
search. A few of ORD's partners include the
National Institute for Environmental Health
Sciences, Department of Energy, NOAA, Health
Effects Institute, NARSTO (a public/private
partnership focused on atmospheric science), and
the World Health Organization. The goal of ORD's
PM research program is to reduce the uncertainties
in understanding the risks from PM to humans,
which directly supports each evaluation of the PM
NAAQS, and to provide a scientific and technical
basis for implementing effective control measures.
Recent Accomplishments
Understanding the Health Effects of
Paniculate Matter, Especially for Sensitive
Groups
How do the airborne particles we breathe lead to
hospitalization and even death? ORD is seeking
the answer to this question through studies
involving people, laboratory animals, and cell
cultures. ORD's study with the University of North
Carolina of a group of elderly persons in Baltimore,
Maryland over the past two years has yielded
insights into the changes in heart and lung
functioning that may indicate more serious illness
following exposure to PM.
In the first year of the study, ORD researchers
chose 26 volunteers from a retirement community
and studied them for three weeks. A retirement
Figure 1-3. A volunteer in the Baltimore study
exhales into a peak expiratory flow meter, which
measures lung function. These results were
correlated with indoor and outdoor PM levels to
evaluate lung effects from PM exposure.
community was chosen for study because the
elderly have been found to be more susceptible to
PM effects. PM levels outdoors and indoors were
monitored at the same time that various physio-
logical measures, such as lung function and heart
rate, were measured for each participant (Figure 1 -3).
A surprising finding was that as PM2 5 levels rose,
people with pre-existing heart problems experienced
lower heart rate variability. Lower heart rate
variability is a well-established factor in sudden
death from heart attack and, consequently, may
represent an important link in the sequence from
breathing particles to adverse effects. In 1998, ORD
expanded the study to include 60 volunteers and
more in-depth air pollution monitoring over four
weeks, and results are now being analyzed.
ORD has also performed laboratory studies that
complement the Baltimore work. In one set of
studies, ORD scientists found that fine particles
(PMj) were deposited at a rate 50% higher or more
in the lungs of people with pre-existing respiratory
disease than those with healthy lungs. ORD
researchers also exposed laboratory rats and mice
with respiratory diseases (similar to human cardio-
pulmonary disease and asthma) to PM. These
experiments identified abnormalities and inflamma-
tory changes in lungs and cardiac changes that are
in agreement with the adverse effects observed in
people. An example of adverse effects in a rat lung
from PM is shown in Figure 1-4. From these
findings, a picture of how particles may be causing
effects is beginning to emerge. Individuals with pre-
existing respiratory disease seem to face greater
risks not only because of their underlying disease
but also because more particles penetrate into their
lungs. Once deposited on the surfaces of the lungs,
particles may cause damage that further impairs the
functioning of the heart and lungs. If sufficiently
severe, the damage can induce serious and even
fatal illness. But what aspects of the particles
render them toxic?
Understanding the Toxicity of Paniculate
Matter
The biological, chemical and physical characteris-
tics of PM, and their relationship to mechanisms
underlying the toxic effects of PM are not well
understood. Major hypotheses state that lung
inflammation and cardiopulmonary stress produced
by PM are related to particle size and to the
chemical nature of PM, such as acidity, organic
chemicals on particle surfaces, or metal content.
ORD research, using both human and animal
studies, is providing critical insight into how these
factors may cause adverse effects in humans. The
-------
1. Health Effects of Airborne Paniculate Matter
Airways
Normal Cell*
Lesion
^BSM 7^
^w^SvA£^>
. Normal Lung Tissue Structure
EL Particulate Matter Exposed Lung
Figure 1-4. Comparison of the lung of a rat exposed to saline solution (A), with no effects seen, to the
lung of a rat exposed to ambient air particulate matter (B), where a lesion has developed.
supposition that metals co-existing with particles is
one important factor in the events leading to PM
health effects has captured particular attention
among several scientific investigators at EPA
(Figure 1-5), and the following studies were
designed to test this theory.
ORD toxicology studies have examined a source of
PM with high metal content, residual oil fly ash,
emitted by power plants. A study published in 1997
reported that different metals each can cause injury
to rat lungs, and in combination the metals appear
to account for much of the toxicity of the particles
(Figure 1-6).
ORD studies on residual oil fly ash and other
emissions set the stage for investigations into the
role of metals in PM toxicity in a specific geo-
graphic area, the Utah Valley. A labor strike that
temporarily closed a major polluting industry in the
Utah Valley provided a unique opportunity to
compare the effects of exposure to PM in air when
the industry was in operation versus when it was
not. Epidemiology studies had reported that
particle-related mortality declined during the year
workers were on strike. ORD researchers studied
samples of ambient PM collected over a three-year
period in the Utah Valley before, during, and after
the strike to elucidate its toxic properties. In
addition, tests were conducted in cultured lung
cells, in laboratory animals, and in human volun-
teers. Tests with cultured lung cells showed that
these cells initiated a chain of events that would
trigger lung inflammation when exposed to air
collected during industry operations. Inflammation
can indicate cellular damage. ORD lexicologists
discovered that metals associated with the PM were
strongly correlated with the cell response that
triggered inflammation.
Ambient PM collected from the same location (Utah
Valley) also was instilled into the lungs of animals
and human volunteers to understand how metals
injure lungs within an organism. The research showed
that metals associated with PM, such as iron, copper,
and zinc, caused lung damage. Although some metals
were more toxic than others, the toxicity appeared to
be related to the type of lung cell exposed in the
experiments. ORD also discovered that even though
different metals can cause similar toxic effects in lungs,
they may do so in different ways.
Other research examining the metal hypothesis
includes ORD's groundbreaking efforts with free
radicals (highly reactive atoms or groups of atoms
with an unpaired electron). ORD scientists were the
first to show that free radicals occurred in the lungs of
rats exposed to metal-bearing particles, and that
metals capable of producing free radicals also caused
lung inflammation in humans. This research suggests
that toxicity appears to be related, in part, to the
formation of free radicals in the lung caused by metals.
University researchers funded under EPA's STAR
Program have also examined how PM causes
toxicity. Investigators found brief exposures of one
hour to residual oil fly ash evoked stress response
from human airway cells sufficient to release
interleukin-8, a factor associated with airway
-------
1. Health Effects of Airborne Paniculate Matter
Inhaled
contaiining particle*
Direct activation, of o°-.,
genes by nvetals
Eiprsisicm of
products leading lo
inflammation, cell
ami risjBE injury
Mtenngtf ,
(«lular
tnnduain
Figure 1-5. The metals hypothesis of particle toxicity. ORD researchers are investigating the role that
metals contained in participate matter may play in toxicity. It has been proposed that metals in particles
can act on cells in several different ways that result in toxicity, including by directly interacting with genes
(through metal response elements), by acting through cellular signaling pathways, and by generating
highly reactive oxygen species.
inflammation. Other STAR researchers examining
PM toxicity in the lungs of rats and monkeys
confirmed that cell injury may be due to a cascade
of events in the lung following short-term expo-
sures to PM. Cell injury was
independent of ozone exposure and
occurred in different parts of the
lung.
Understanding the Sources of
Paniculate Matter
We now have a better understand-
ing of the adverse health effects of
PM, and potential causes of toxicity,
but what are the physical, chemical
and lexicological characteristics of
PM emitted by various sources? To
answer this question for sources
that burn heavy fuel oil, OPJ)
engineers and health scientists have
evaluated PM from different fuels
(of two grades and three sulfur
contents) burned in EPA research
boilers that are representative of
small industrial, commercial and
institutional applications. The
composition of the PM varied with
particle size, withPM25 composed of
significantly higher levels of metals
and sulfates than the larger particles (see Figure 1-7).
Animal lexicological testing revealed greater
pulmonary damage by Ihe fine particles.
£. 100
Saline ROFA
(control) particles
Fe
Modified from K.L. Dreher, etal., J.Toxicol. Environ. Hlth., 1997.
All 3
metals
Dose: 2.5mg/rat
Figure 1-6. Role of metals in particle-induced lung injury.
Lung injury in rats from a saline solution control is compared
to injury from a suspension of residual oil fly ash (ROFA)
particles and to solutions containing iron (Fe+3), vanadium (V+2)
and nickel (Ni+2) at the same concentration as found in ROFA.
Each metal causes toxicity, and in combination the metals
appear to account for much of the toxicity of the ROFA particles.
-------
1. Health Effects of Airborne Paniculate Matter
High suirur
ruei on
Medium
i.i i. i
Fuel Oil
Low Sulfur
Fuel Oil
copper
iron
Nickel
Vanadium
Zinc
Pint Coarse Fine Coarse Pint Coarse
Figure 1-7. The metals content of PM from combustion of heavy fuel oil is highly dependent on particle
size, with fine particles (PM25) containing higher concentrations than larger particles. Data from C.A. Miller
et al., Combustion Science and Technology, 1998.
These tests showed that the mechanisms by which
fine particles are formed are significantly different
than those governing the formation of larger
particles. These different mechanisms lead to
substantial differences in chemical composition,
especially the higher content of metals and sulfates
observed in the fine PM. The mechanisms apply to
combustion systems in general, including coal and
other fuels. This work has therefore been important
in identifying toxic constituents of combustion PM
and in setting research priorities for control of
combustion emissions.
Additional studies have characterized particle
emissions from large diesel trucks as they are
traveling along the highway. ORD researchers
equipped a tractor-trailer rig for real-time emission
monitoring and made first-of-a-kind measurements
of fine PM (less than 1 micrometer in diameter)
emissions under highway conditions. Simultaneous
measurements were made from the exhaust pipe and
in the plume left by the truck. These data will enable
air quality regulators to make better estimates of the
impact of large diesel trucks on ambient air quality.
Another significant source of fine PM in certain
parts of the country is residential wood combus-
tion. ORD researchers have made field and labora-
tory measurements that have shown that emissions
from wood stoves are strongly affected by moisture
content of the wood and how a homeowner
operates the stove, as well as stove type and
condition. Even stoves that had been certified as
clean burning when purchased did not maintain
their high level of emission control for very long.
Most had significant loss of control effectiveness
after a few years of use. These data will influence
control strategies to meet PM standards in areas
with high usage of wood stoves.
Understanding the Long-Distance
Transport of Paniculate Matter
An air sample from a typical American city can
contain particles from an astonishing variety of
sources, ranging from nearby buses and cars, to
power plants or forest fires hundreds of kilometers
away, to even the Sahara Desert on the other side
of the globe. Scientists look for unique "signa-
tures" or characteristics of the particles that may
reveal their source and use computer models to
show how particles move in air currents to decipher
where airborne particles come from and predict
where they will go.
In 1998, ORD reached a major milestone in its
efforts to understand the movement of particles
and other air pollutants when it publicly released
iheModels-3 Community Multi-Scale Air Quality
model. This computer model, available free of
10
-------
1. Health Effects of Airborne Paniculate Matter
Figure 1-8. Graphic of PM25 mass (24-hour average) across the eastern U.S. generated by Models-3, an
ORD air quality computer model.
charge on the Internet (www.epa.gov/asmdnerl/
modelsS/), is the first to simulate the concentra-
tions of multiple air pollutants simultaneously and
show their movement across entire regions or
subcontinents. An example of the type of data that
can be generated by Models-3 for PM is shown in
Figure 1-8. By manipulating simulated emissions of
particles and particle precursors, users of the model
can gain valuable insights into the likely effects of
various strategies to control PM, ozone, and other
air pollutants. As a result, EPA expects that
Models-3 will be used extensively by environmental
managers in state and local governments and by
scientists who develop air quality modeling
systems. ORD plans additional refinements of
Models-3 as more extensive PM25 monitoring data
become available.
Several studies funded under the STAR Program
also have focused on understanding and modeling
particle transport. STAR investigators have
developed a new chemical model that describes
how particles are formed in the atmosphere. It
incorporates up to 335 chemical reactions simulat-
ing the formation, growth, and removal of particles
in the atmosphere. This will improve the ability of
air quality models to accurately assess the contri-
butions of both emissions related to human activity
and biogenic (naturally formed) emissions to
ambient PM levels. Other studies have included a
pilot field study that characterized the formation,
fate and transport of fine particles and ozone using
vertical profile measurements. This effort demon-
strated the importance of transported pollutants in
the initiation of PM and ozone events in Philadel-
phia. Finally, researchers have documented a sharp
11
-------
1. Health Effects of Airborne Paniculate Matter
decrease in sulfate particle concentrations in the
northern United States that may be related to the
reduction of sulfur dioxide (SO2) emissions in the
Midwest. The researchers reached their conclu-
sions by obtaining and analyzing measurements of
SO2 and sulfate from 1979 through 1996 inNew York
State.
Where Do We Go From HereP
The paniculate matter models and studies de-
scribed in this report have helped answer important
questions about the sources of PM, how people are
exposed, and the health effects that PM can cause.
ORD is using these findings, along with the
discoveries of other research organizations, to
refine the direction of its comprehensive research
program for paniculate matter. The program is
designed to reflect the near and long-term research
priorities presented to EPA by the Committee on
Research Priorities for Paniculate Matter of the
National Research Council. As ORD carries out this
program, it continues to use the risk assessment/risk
management framework depicted in Figure 1-2 to
organize and integrate its research.
Within the general category of exposure assessment,
for example, ORD is conducting atmospheric sciences
research under the aegis of NARSTO, a consortium of
public and private research organizations of which
EPA is a member. NARSTO was originally created to
coordinate North American atmospheric research on
tropospheric ozone in support of air quality manage-
ment. It has recently expanded its mission to encom-
pass PM research following findings by ORD and
others that have improved understanding of the
linkages between ozone and paniculate matter in the
atmosphere. ORD scientists are conducting PM
research in areas such as monitoring, emissions,
atmospheric chemistry and processes, and modeling
underNARSTO.
ORD is also conducting studies that will contribute
to exposure assessment and effects assessment in
such areas as particle dosimetry, particle toxicity,
and the role of PM and associated air pollutants in
adverse health effects. For example, ORD is
building on its research on the effects of PM in
susceptible individuals by carrying out "panel
studies" that follow small groups of individuals
over time through intensive personal exposure
monitoring and activity diaries. ORD is also
extending the initial findings of the study of elderly
Baltimore residents by completing more thorough
analyses and conducting a similar study in Fresno,
California. Health effects assessment is also being
advanced through targeted toxicology studies in
both laboratory animals and clinical studies by both
EPA laboratories and through the STAR program.
In the area of risk characterization, ORD will review
and summarize the exposure and health effects work
by ORD and others in the next Air Quality Criteria
Documents forPM, currently scheduled for 2000
and 2005. These documents will present the state-
of-the science for paniculate matter and will be
critical in EPA's review of the national standards for
PM levels in the ambient air.
Finally, ORD is conducting risk management
research to evaluate the cost and effectiveness of
options for reducing emissions of both particles
and gaseous precursors that develop into particles
in the atmosphere. Examples include working with
electrical utilities to design and test electrostatic
filters and fluid bed scrubbers to control emissions
at power plants. This research will help EPA, states,
and industry develop cost-effective strategies to
reduce exposure to paniculate matter, thereby
protecting public health.
12
-------
2. Drinking Water: Micronial Pathogens and
Disinfection By-Products
/ptosporidiu
parvum
In 1993, contaminated drinking water in Milwaukee,
Wisconsin, brought about an outbreak of
cryptosporidiosis (caused by the microorganism
Cryptosporidium parvum) that resulted in approxi-
mately 400,000 cases of acute gastroenteritis and
about 100 deaths. This outbreak represents the
largest documented occurrence of disease associ-
ated with contamination of a water supply in the
United States. According to EPA and the Centers
for Disease Control and Prevention, during 1991-
1996 more than 40 out-
breaks of waterborne
disease occurred due to
contamination by a variety
of bacteria (e.g., E. coli),
viruses (e.g., Norwalk
virus), and parasites (e.g.,
Crypto sporidium and
Giardia lamblia) (Figure
2-1).
The continued occurrence
of outbreaks of water-
borne disease each year
demonstrates that the
quality and safety of
drinking water can still be
compromised by patho-
gens when it is not adequately treated. It is also
likely that many other outbreaks occur but are
either unrecognized or unreported.
Cryptosporidium is a particular concern because it
poses a risk to groups more susceptible to infection
than the general population, such as those with
weakened immune systems or preexisting diseases.
Even in healthy individuals, ingestion of a small
number of Cryptosporidium oocysts (a phase in
Cryptosporidium's life cycle) may cause illness.
This threat makes it critical to have sampling
methods able to accurately and quickly detect the
presence of Cryptosporidium in drinking water, and
for all water treatment systems to be able to
eliminate or inactivate the microbe.
To combat these threats, systems to treat drinking
water have been in place for many years, taking
advantage of both physical (sedimentation) and
chemical (various disinfectants) treatment. A
typical drinking water treatment system is shown in
Figure 2-2. The use of chlorine to disinfect drinking
water has been standard practice over the past 100
f raatg
f
Figure 2-1. Photomicrograph of
Cryptosporidium parvum oocysts and Giardia
lamblia cysts. White Bar =10 microns.
years, and has been a major factor in the dramatic
decline of waterborne disease worldwide.
Although disinfectants have been highly success-
ful in reducing the incidence of waterborne disease
in humans, other concerns have been raised about
the safety of disinfected water. For example,
chlorine reacts with natural organic substances in
source (untreated) water during treatment to form a
number of potentially harmful chemical by-products
termed disinfection by-
products, or DBFs. It is
now known that chlorine
and alternative disinfec-
tants such as chloram-
ines, ozone, and chlorine
dioxide produce hundreds
of DBFs that end up in
the drinking water supply
at relatively low concen-
trations. Many of these
substances have been
shown to cause cancer
and other effects in
laboratory animals at very
high levels of exposures.
In addition, some epidemi-
ology studies have
reported slightly increased risks of cancer and
adverse reproductive outcomes in populations
exposed to disinfected drinking water, but a causal
relationship has not been established.
The Safe Drinking Water Act (SDWA) Amend-
ments of 1996 respond to these contamination
problems by mandating that EPA appropriately
address microbes and DBFs as well as other known
or anticipated water contaminants. The Agency is
developing new rules to establish additional
standards that limit public exposure to microbial
contaminants and DBFs. The overall goal and
challenge of these new rules is to establish cost-
effective approaches that minimize potential risks
associated with DBFs without compromising the
critically important need to control pathogenic
microorganisms.
Given the uncertainty surrounding DBF-associated
health risks, the severity of effects from certain
microbes, and the potentially high costs of further
regulation, EPA has designated drinking water as
13
-------
2. Drinking Water: Microbial Pathogens and Disinfection By-Products
one if its highest research priorities. ORD 's
Research Plan for Microbial Pathogens and
Disinfection By-Products in Drinking Water
(www.epa.gov/ORD/WebPubs/fmal) describes the
research needed to support the regulatory program
of EPA's Office of Water and serves as the founda-
1) Dtiurnical
A j-d'. -"
3) FlDcojIahcn .jammed r«it*is
Illl IIIHMII
6) D
-------
2. Drinking Water: Microbial Pathogens and Disinfection Dy-Products
Understanding the types and sources of
disinfection by-products (DBFs)
Evaluating new small system
treatment technologies
Developing improved risk
assessment guidance for
disinfectants
Developing improved methods
to detect drinking water
pathogens
Identifying new DBFs from
alternative disinfectants used
to treat drinking water
Understanding the cancer and
non-cancer effects of DBFs
Understanding the comparative
risks posed by exposure to pathogens
and DBFs
Understanding the nature
and magnitude of water-
borne disease
Figure 2-3. Some of the major research and assessment activities
conducted by ORD in examining the health effects of microbial pathogens
and DBFs in drinking water.
Recent Accomplishments
Developing Improved Methods to Detect
Drinking Water Pathogens
The 1993 outbreak of Cryptosporidium in Milwau-
kee highlighted the importance of having sensitive
and rapid methods to detect waterborne pathogens.
In response to this need, ORD scientists, in
consultation with other scientists in the field,
developed and evaluated Method 1622, an im-
proved method for detecting Cryptosporidium
oocysts in source and drinking water. The previous
method typically detected approximately 11% of
Cryptosporidium oocysts added to a sample,
whereas Method 1622 has an average detection rate
of 38%. Method 1622 can be accessed on a website
(www.epa.gov/nerlcwww/1622ja99.pdf).
An important limitation of available detection
methods for Cryptosporidium is an inability to
determine if the oocysts are viable (live) or infec-
tious. This information is critical in assessing the
public health significance of finding evidence of
this pathogen in drinking water. Applying recent
advances in the cell culture of Cryptosporidium,
scientists at the Metropolitan Water District of
Southern California under a STAR grant improved a
method for determining oocyst viability and
infectivity. The method involves: (1) recovery and
purification of Cryptosporidium oocysts from the
water samples using Method 1622; (2) inoculation
of the oocysts onto human cells grown on slides;
and (3) after incubation, detection of infected cells
using molecular techniques. This method, com-
bined with innovative molecular techniques, is a
promising indirect approach
for detection and quantifica-
tion of infectious oocysts in
drinking water.
Identifying New
Disinfection
Dy-Products from the
Use of Ozone
Many drinking water treat-
ment plants in the United
States use disinfectants other
than chlorine (particularly
chlorine dioxide, ozone, or
chloramines) to help control
the risks from waterborne
pathogens. However,
uncertainty exists over the
types and amounts of by-
^^^^^^^^^n products produced from
these alternative disinfec-
tants, and their potential health risks. Of particular
concern is the possibility that switching to alterna-
tive disinfectants to control microbial risks may
actually lead to an increase in the risk associated
with exposure to a new set of poorly characterized
DBFs. Using specialized analytical techniques,
ORD scientists have made considerable progress in
identifying more DBFs that form when these
alternative disinfectants are used (Figure 2-4).
Figure 2-4. ORD scientist prepares to inject a
sample into the high resolution mass
spectrometer to identify DBFs in drinking water.
-------
2. Drinking Water: Microbial Pathogens and Disinfection By-Products
Ozone can be an effective alternative disinfectant
because fewer chlorinated by-products are pro-
duced. Ozone must be used in combination with a
secondary disinfectant such as chlorine or chlora-
mine, since residual concentrations of ozone do not
remain in the treated water to provide continuous
protection as does chlorine. Special concerns arise
when ozone is used to treat water containing high
levels of bromide, a naturally occurring substance
present in source water in various parts of the
United States. Ozonation of water containing
bromide leads to the formation of bromate, which
has been shown to cause cancer in laboratory
animals. ORD scientists are conducting research to
investigate the possible formation of other bromi-
nated by-products, which tend to be more toxic
than related chlorinated by-products. These studies
have found that elevated bromide levels contrib-
uted to an increase in brominated by-products
following ozonation combined with chlorine or
chloramine. Many of these DBFs had not been
previously identified (Table 2-1), and they are now
being prioritized for possible future health effects
research to determine if any may be a concern to
public health.
Understanding the Cancer Effects of
Disinfection By-Products
Bromate. Bromate is one of several by-products
regulated under the EPA's new Stage I Disinfec-
tants/Disinfection By-Products (D/DBP) Rule,
which established new standards that limit public
exposure to disinfectants and certain DBFs in
Disinfection By-Products
1,1- Dibromopropanone
1,1- Dibromo - 3 - chloropropanone
1, 1, 1 - Tribromopropanone
1, 1, 3 - Tribromopropanone
1,1- Dibromo - 3, 3 - dichloropropanone
1,3- Dibromo - 1, 3 - dichloropropanone
1,1,3- Tribromo - 3 - chloropropanone
1, 1, 3, 3 - Tetrabromopropanone
1,1,1,3- Tetrabromo - 3 - chloropropanone
1, 1, 1, 3, 3 - Pentabromo - 3 - chloropropanone
Table 2-1. Family of recently identified disinfection
by-products (Halopropanones).
drinking water. Studies in one Japanese laboratory
in the mid-1980s showed that bromate caused
cancer in rodents. To more fully explore the public
health risks of bromate, ORD scientists conducted a
chronic exposure study in which rats and mice were
exposed to various doses of bromate in their
drinking water for up to two years. The animals
were examined for tumors and other evidence of
cancer at various intervals over the two year period.
Bromate was found to cause cancer in the male rat
at three different organ sites and to cause kidney
tumors in the male mouse. EPA used the results of
this study in its decision to establish a Maximum
Contaminant Level Goal (MCLG) of zero for
bromate. The MCLG is the level of a contaminant at
which there would be no risk to human health. This
value is used in the development of the Maximum
Contaminant Level (MCL), a legally enforceable
standard that represents the highest level of a
contaminant that EPA allows in drinking water.
Brominated trihalomethanes. The trihalomethanes
(chloroform, bromodichloromethane,
dibromochloromethane andbromoform) are among
the most prevalent DBFs in chlorinated drinking
water. These substances are currently regulated by
EPA as a class of compounds under the new Stage I
D/DBP Rule. The trihalomethanes have been shown
to cause cancer at high doses in laboratory animals
and have also been associated with low risks of
cancer in several epidemiology studies. One of the
most important avenues of investigation for
trihalomethanes is the potential mechanism(s) by
which they cause cancer, since this will help clarify if
and how these substances pose a risk to humans.
Previous research suggests that brominated
trihalomethanes may be more carcinogenic than similar
by-products containing chlorine (e.g., chloroform), but
the reasons for these differences are unknown.
Some insights into this issue are being provided by
new ORD findings on how the brominated
trihalomethanes are metabolized. In animals and
people, typically one or more metabolic pathways
exist that transform environmental toxicants into
metabolites that can be eliminated from the body,
but in some cases the metabolites are more harmful
than the parent chemical. ORD scientists have
identified one such potentially harmful pathway
that is active for brominated trihalomethanes, but
not chloroform. This pathway involves an enzyme,
glutathione S-transferase theta, and may result in
damage to cell DNA, increasing the likelihood that
the affected cell may become cancerous (Figure 2-5).
An analogous enzyme is found in humans, and
therefore a similar pathway is likely to be active in
people. Moreover, the production of glutathione S-
transferase theta is known to vary greatly among
16
-------
2. Drinking Water: Microbial Pathogens and Disinfection By-Products
0r-C-H
IMUIT HI
BlUflRMjtJtfoflHnil>UIM llT
$ -h
LJI
(JUUItWVI
*'t'i: ti -p >. In n :» ,1 i !;! !': r= i-:i I .m^ ^ ":nn- : .1
-mi
n vw«- t^nim m-iiir
.in «i-T r^i' rnt-".in rl * n-i . %w
Uirxn-
Figure 2-5. Metabolic pathway by which the DBP
bromodichloromethane may cause DMA damage.
humans, which could mean that individuals differ in
their susceptibility to cancer from exposure to the
brominated trihalomethanes in drinking water.
Understanding these metabolic processes in both
animals and humans will provide a better scientific
foundation for the risk assessments of these
contaminants, and could ultimately influence risk
management strategies to limit exposures to the
DBFs that pose the highest concern.
Investigating Noncancer Health Effects of
Disinfection By-Products
Recent publications in the scientific literature on
DBF exposures and adverse reproductive effects
have prompted a concerted effort by EPA and
others to study this issue. In 1998, a landmark
study that evaluated the relationship between
exposure to trihalomethanes and spontaneous
abortion in women was published by scientists
from the State of California. This investigation,
which was partially supported by ORD through a
grant to the California Public Health Foundation,
examined health and exposure data collected for
in OKI*. over 5 000 pregnant women living in three
-_;.-. , . , , . , - ,
areas of California. An increased nsk of early
term miscarriage was observed in women who
consumed large amounts of water containing high
levels of total trihalomethanes and
bromodichloromethane. Although not conclusive,
this study provided important new information to
better characterize the potential health risks
associated with exposure to drinking water contami-
nants. It also highlighted the need for additional
research to replicate the findings as well
as refine how exposure is characterized in
these types of studies.
Developing Improved Risk Assessment
Guidance for Disinfectants
To support EPA's Stage I D/DBP Rule, ORD
developed comprehensive risk assessment criteria
documents for chlorine and chloramine. Each
criteria document provided guidance for estimating
the risks of cancer and other health effects (e.g.,
cardiovascular disease) associated with exposure to
these disinfectants. The criteria documents were
peer reviewed by EPA's Science Advisory Board.
The risk assessments concluded there was no
significant evidence that exposure to chlorine or
chloramine in drinking water caused cancer in
humans, produced long-term toxic effects at levels
found in drinking water, or caused long-term
noncancer health risks such as reproductive failure,
cardiovascular disease, liver toxicity, ordevelop-
mental toxicity These results served as the
scientific basis for establishing the maximum
residual disinfectant goals for chlorine and chloram-
me m ^e stage I D/DBP Rule
17
-------
2. Drinking Water: Microbial Pathogens and Disinfection By-Products
Evaluating New and More
Cost-Effective Small System Treatment
Technologies
Some issues of concern for small community
drinking water systems in the United States are lack
of appropriate, readily available technologies at
affordable operation and maintenance costs.
Nearly 90% of the water quality violations under
the SDWA occur in small systems and are related to
microbiological contamination of the water. ORD is
evaluating a variety of alternatives to conventional
water treatment systems that are effective, simpler,
and less expensive to operate and maintain in order
to mitigate many of these problems.
ORD investigated several types of filtration
systems to determine their effectiveness in remov-
ing harmful microorganisms from drinking water.
ORD scientists found that the ability of the
filtration systems to remove Cryptosporidium was
highly variable. Bag filtration systems demon-
strated a wide range of effectiveness, whereas
removal with cartridge filtration was more reliable.
Ultrafiltration (Figure 2-6) removed most of the
Cryptosporidium, although imperfections in
manufacturing and design seriously reduced
effectiveness.
ORD scientists also demonstrated that small, inert,
plastic beads called microspheres are the most
accurate and precise Cryptosporidium surrogates
for safer, quicker, and less-expensive pilot testing of
the different types of filters. These surrogates will
make it easier to evaluate filtration technologies,
and help ensure that the alternative technologies
are operated properly by local entities, particularly
in small communities.
Where Do We Go From HereP
The challenges that lie ahead for the drinking water
research program are broad, covering a wide range
of issues and contaminants that are the focus of
either current or future regulatory decision making
under the Safe Drinking Water Act Amendments of
1996. As ORD carries out its drinking water
research program, it will continue to use the risk
assessment/risk management framework depicted in
Figure 2-3 to organize and integrate its research.
Research on DBFs and microbial pathogens
currently underway builds on past research and
accomplishments.
For example, in the area of exposure assessment,
development of more reliable and sensitive methods
to measure very low levels of bromate and aldehyde
by-products (important ozone DBFs) in treated
Figure 2-6(a): ORD scientist testing Ultrafiltration membrane package plant. 2-6(b) Inset is a cut-away
drawing of the spiral-wound membrane, which is placed in the horizontal white tube at the top of the
stainless steel structure. Three 8-inch by 40-inch membranes are loaded into the white tube like
batteries in a flashlight. Water enters the tube from the right, passes through the membranes, and exits
the tube on the left.
IS
-------
2. Drinking Water: Microbial Pathogens and Disinfection Dy-Products
water is nearing completion. These methods can
then be used for regulatory compliance monitoring
of bromate and aldehydes. Studies are also being
conducted to develop improved detection methods
for viruses (e.g., echovirus, coxsachievirus) that
may be responsible for waterborne outbreaks of
acute nonbacterial gastroenteritis.
In addition to improving detection methods, ORD
researchers are evaluating laboratory tests to
accurately determine the characteristics of certain
bacteria known to grow on filtration materials used
for treating water and collect on the surfaces of
pipes in the drinking water distribution systems
(known as biofilms). Although these bacteria are
not known to affect healthy individuals, they may
pose a risk to people whose resistance to infection
is impaired. Research is also being conducted to
determine the most effective means for controlling
the growth of bacteria in the distribution system by
using alternative disinfectants and pH control.
Health effects research is continuing to investigate
the potential reproductive and developmental
effects of priority DBFs. Follow-up research is
being conducted on by-products from selected
chemical classes (e.g., haloacetic acids) that may be
of greater concern for these kinds of effects. ORD
scientists are also reevaluating the California
populations included in a spontaneous abortion
study using an improved exposure assessment to
address some of the weaknesses of the previous
study.
In addition, ORD is collaborating with the Centers
for Disease Control and Prevention to analyze data
for 1997-1998 on the occurrence and causes of
waterborne disease outbreaks in the United States.
Goals include characterizing the epidemiology of
waterborne disease for this period, and identifying
water treatment system deficiencies and specific
contaminants responsible for the outbreaks. ORD
is also conducting studies in several parts of the
United States to evaluate the occurrence of
endemic waterborne disease (i.e., a "non-epidemic"
level of disease that may be attributable to drinking
water but not reported as an outbreak).
In the area of risk management research, ORD
scientists are continuing to examine the effective-
ness of alternative disinfectants such as ozone to
inactivate Cryptosporidium and reduce harmful
DBFs without creating biofilm growth problems in
the distribution system. A manual for state
drinking water agencies will be prepared summariz-
ing the most recent developments. In addition to
research on drinking water distribution systems,
ORD scientists are developing improved methods
for estimating the vulnerability of ground water
systems to viral contamination. Research includes
evaluating important hydrogeological, geochemical,
and microbiological factors affecting the transport
and survivability of viruses in the subsurface to
more accurately characterize and predict viral
contamination of ground water.
19
-------
20
-------
3. Advances in Risk Assessment
RiskAssessment Overview
Introduction to Risks and Risk
Assessment
Each day we face a multitude of risks that vary both
in their likelihood of occurrence and in how much
we can control them. For example, depending on
our jobs and hobbies, we may be at risk from
accidents and injuries. Each of us also faces
environmental risks as we are exposed to pollutants
in the air we breathe, the water we drink, and the
food we eat. Although we may not realize it, each of
us acts to evaluate and control many of the risks we
face. Before we set out in our car on a snowy day,
for instance, we may assess the road conditions and
take actions to manage the risks by driving slowly
and wearing a seatbelt. Government agencies and
other institutions also endeavor to understand and
control risks, but they often use more formal and
structured processes known as risk assessment and
risk management.
Risk can be defined as a measure of the likelihood
that a given hazard will cause harmful events to
occur, such as illness and death in people and
wildlife or damage to ecosystems and property.
Risk assessment is a tool for gathering and organiz-
ing the best available information so that risks can
be understood. As described in the introduction to
this report, risk assessment is commonly divided
into four phases following an approach developed
by the National Academy of Sciences. These are
hazard identification, dose-response assessment,
exposure assessment, and risk characterization.
Risk management is the process of evaluating
information from a risk assessmentas well as
factors such as economic and social consider-
ationsto decide what should be done about the
risks. Figure 3-1 depicts the risk assessment and
risk management processes.
When applied to chemicals, risk assessment
examines the types of adverse health effects that
might occur in humans and wildlife following
chemical exposure (hazard identification), how they
vary with the degree of exposure (dose-response),
and the degree to which exposure actually occurs
(exposure assessment). Combining this information
enables the overall risk to be described for decision
makers (risk characterization). Risk management
involves deciding what actions, if any, are needed
to prevent or reduce the risk, such as limiting
pollutant emissions.
RiskAssessment
Risk Management
Control
Options
Risk
Characterization
Figure 3-1. As commonly described, risk assessment consists of four phases, culminating in risk
characterization. The findings of the risk assessment are considered along with several other factors in
making risk management decisions.
21
-------
3. Advances in Risk Assessment
Use of Risk Assessment at EPA
Many of the laws that govern EPA programs
require the use of risk assessment. EPA applies risk
assessment to a variety of regulatory issues
including toxic chemical control, pesticide registra-
tion, hazardous waste cleanups, and the setting of
air, water, and soil standards. Given the costs that
are required for activities such as cleaning up
hazardous waste sites, it is important that EPA's risk
assessments provide the information needed for
setting priorities for cost-effective responses to
environmental problems. Even though EPA's
various mandates may require different risk
management approaches, EPA strives to use
consistent, publicly-reviewed methods for conduct-
ing risk assessment across the Agency. Consis-
tency in risk assessments contributes to their more
efficient development, less confusion on behalf of
affected parties, and better regulatory decisions.
In applying risk assessment to regulatory decision
making, EPA analyzes all available scientific
evidence in order to evaluate the relationship
between exposure to environmental agents and the
potential to cause harm. The concerns addressed
by a risk assessment may include health effects
such as reproductive and developmental abnormali-
ties, cancer, and neurological effects, as well as
ecological effects such as species extinction, loss
of habitat and other forms of ecosystem damage.
Challenges in Risk Assessment
Although risk assessment provides a structured
framework for rational regulatory decision making, it
is not without controversy. One of the most difficult
- and frequent - risk assessment challenges faced
by EPA is how to handle uncertainties that arise
when environmental exposures of concern differ
greatly from the situations in which risks have been
scientifically studied. For some kinds of hazards,
risks can be estimated directly from readily available
sources, such as risks of dying in a plane crash or
from lightning. However, for many of the risks EPA
deals with, such as risks of dying of cancer from
chemical exposure, risks are much more difficult to
estimate. Much of the information about a
chemical's potential toxicity may be restricted to
laboratory animal studies. Even if health effects
have been confirmed in people, the information
typically is for highly exposed subgroups such as
industrial workers. In the case of new chemicals
under consideration for approval by EPA, little or
no human data may exist at all.
As a result, risk assessors must extrapolate from the
conditions under which risk information has been
collected to the actual conditions of human
exposure. To bridge information gaps, EPA and
other regulatory agencies use what are known as
default assumptions. Default assumptions are
inferences based on general scientific knowledge of
the phenomena in question and are also matters of
policy concerning the appropriate way to bridge
uncertainties. Examples include standard ap-
proaches for extrapolation from laboratory animals
to humans and from the high exposures of labora-
tory and occupational studies to the lower expo-
sures experienced under environmental conditions.
For instance, in the absence of data to the contrary,
EPA assumes that humans are more sensitive to
chemicals of concern than laboratory animals.
While necessary, default assumptions come under
frequent criticism for either being overly protective
or insufficiently protective of human health and the
environment.
Several advisory panels have examined the risk
assessment approaches used within regulatory
agencies and made a number of recommendations
for improving the process. Most recently, the
National Research Council (Science and Judgment
in Risk Assessment, 1994) and the Commission on
Risk Assessment and Risk Management (Risk
Assessment and Risk Management in Regulatory
Decision-Making, 1997) provided recommenda-
tions to EPA. In summary, these advisory groups
concluded that the scientific foundation that forms
the basis for EPA's approach to risk assessment
should be further strengthened through a long-term
research program to reduce EPA's reliance on
default assumptions and by developing information
on cumulative risk.
"The quality of risk analysis will improve as
the quality of input improves. As we learn
more about biology, chemistry, physics, and
demography, we can make progressively
better assessments of the risks involved.
Risk assessment evolves continually, with
reevaluation as new models and data
become available."
Science and Judgment in Risk Assessment
(National Research Council, 1994)
ORD's Program to Improve Risk
Assessment
In response to these continuing challenges and the
recommendations from scientific organizations,
ORE) is carrying out a major program to improve
both human health and ecological risk assessment.
22
-------
3. Advances in Risk Assessment
In carrying out this program, considerable emphasis
is placed on scientific peer involvement and peer
review. In addition, ORD is collaborating with other
federal agencies, states, international organizations
and professional societies to harmonize risk
assessment approaches on a national and interna-
tional scale.
One major component of ORD's risk assessment
program is the development of risk assessment
guidance, in partnership with EPA's regulatory
offices, to foster consistency across EPA. To date,
risk assessment guidelines have been developed
addressing cancer, mutagenicity, developmental
and reproductive effects, neurotoxicity, exposure,
chemical mixtures, and ecological effects. Periodi-
cally, these guidelines are updated to reflect current
knowledge and emerging issues. Another aspect of
ORD's program is the continual refinement of
existing exposure and effects data, models and
methods. As the scope and quality of data are
improved, ORD is able to improve the various
models and methods used in risk assessment,
which in turn improves the overall accuracy of a
risk assessment. Through this research, ORD can
reduce the uncertainty in risk assessments and the
need to rely on default assumptions. Finally, ORD
also conducts risk assessments. These assessments
generally either serve as prototypes demonstrating
new scientific approaches or address Agency needs
that span many different programs or are particularly
contentious and precedent-setting.
Recent Accomplishments
Development of Risk Assessment
Guidance
EPA risk assessment guidance is prepared by the
Risk Assessment Forum, which is staffed by ORD
and brings together panels of experts from across
the Agency. Not to be confused with EPA's testing
guidelines (which provide specific guidance on
how to conduct toxicity and other tests), the risk
assessment guidelines set forth Agency-wide
approaches for assessing risks. During 1997 to
1998, the following guidelines and guidance were
published by the Forum.
Guidelines for Ecological Risk Assessment
Ecological risk assessment evaluates whether
adverse ecological effects may occur or are
occurring from exposure to one or more chemical,
biological, or physical stressors (such as pollutants
or habitat loss). In 1992, EPA proposed its first
principles and terminology for the ecological risk
assessment process in the Framework for Ecologi-
cal Risk Assessment. Other materials followed, such
as ecological assessment case studies. The
Guidelines for Ecological Risk Assessment
(www.epa.gov/ncea/ecorsk.htmj, published in May
1998, build on the earlier efforts. Developed to
increase consistency and improve the quality of
ecological risk assessments within EPA, the
Guidelines describe EPA's current scientific
thinking and approaches for conducting ecological
risk assessments.
The Guidelines provide examples of how ecological
risk assessment can be applied to a wide range of
situations, such as hazardous waste clean-up, new
chemical and pesticide registration, and watershed
management. Rather than specifying techniques
and methods for ecological risk assessment, they
describe general approaches that can be used and
their strengths and weaknesses. Figure 3-2 depicts
the overall framework for ecological risk assessment
presented in the Guidelines. Although the Guide-
lines are primarily intended for use within EPA,
other government agencies and the interested
public will benefit from understanding EPA's
approach to ecological risk assessment.
'" limn. -ii::.ilir.-| Rrk.ill:-,
to Iht Rnh MaroQcr
la
Figure 3-2. The framework for ecological risk
assessment.
23
-------
3. Advances in Risk Assessment
Guidelines for Neurotoxicity Risk Assessment
The nervous system (composed of the brain, spinal
cord, and nerves) regulates the flow of information
in the body and controls bodily functions. Neuro-
toxicity refers to a toxic effect on the nervous
system after exposure to a chemical, physical, or
biological substance. Toxic effects that occur, such
as changes in muscle coordination, paralysis,
seizures, or behavioral changes, vary depending on
the amount of exposure and the region of the
nervous system affected (Figure 3-3). Incidents of
severe and irreversible nervous system damage in
people following exposure to chemicals such as
mercury have underscored the health risks that
neurotoxic agents can pose. EPA developed the
Guidelines for Neurotoxicity Risk Assessment
fwww.epa.gov/ncea/nurotox.htmj to provide a
Neuron
Neurotransmitters
Nerve Terminal
Axon
Cell Body
Dendrite -^
Example Neurotoxins
Organophosphate and carbamate pesticides:
block breakdown of neurotransmitters
Botulinum toxin: blocks
release of neurotransmitters
MPTP: toxic to nerve terminals
Acrylamide: toxic to axons
Methylmercury: toxic to cell bodies
Figure 3-3. Examples of neurotoxicity. Nerves are
made up of cells called neurons. Neurons receive
impulses from other neurons through dendrites,
transmitting signals along axons to nerve terminals,
where neurotransmitters are released.
Neurotransmitters stimulate adjacent neurons or, in
the case of motor neurons, stimulate muscles.
Neurons are vulnerable to toxic agents that can act
on various stages of the signal pathway; a few
examples are shown here.
sound scientific basis and promote consistency in
conducting neurotoxicity risk assessments. The
Guidelines:
Outline the scientific basis for evaluating effects
from exposure to neurotoxicants and discuss
how to evaluate data from human and animal
studies.
Note the special vulnerability of the nervous
systems of infants and children to chemicals and
provide guidance on interpreting developmental
and reproductive studies that involve the
nervous system.
Characterize the health-related data base for
neurotoxicity risk assessment.
Describe calculations and approaches for some
specific elements of neurotoxicity risk assess-
ment, such as estimating the exposure level
below which adverse effects will not occur (the
reference dose).
Guiding Principles for Monte Carlo Analysis
Though risk assessments are indispensable tools
for EPA, they have limitations in how accurately
they reflect the true risks that people or ecosystems
face. Two major reasons for their limitations are
uncertainty and variability. Uncertainty refers to
lack of knowledge about particular factors that are
important in risk assessment. For example, any
given technique for measuring pollution levels has
inherent uncertainties and the amount of a pollutant
required to cause a specific health effect usually is
also uncertain. Variability refers to true diversity
among individuals or properties assessed. For
instance, people vary in their body weights, age,
and susceptibilities to toxic chemicals. Also, the
amount of pollution emitted and weather conditions
that affect pollutants in the environment vary over
time.
A recent trend in risk assessments has been to
enhance their usefulness and validity by more
thoroughly describing the variability and uncer-
tainty in the exposure or risk estimates. One
approach to accomplishing this is known as
probabilistic analysis, which includes a technique
known as Monte Carlo Analysis. In probabilistic
analysis, statistical techniques and computer
simulations are used to generate a range of risk
values, rather than a single "point estimate" of the
average or "worst-case" risk (Figure 3-4). Express-
ing risk as a range is much more informative than a
single number, and can help identify groups of
individuals who may be most at risk from a particu-
lar chemical.
In 1997, EPA prepared the first Agency-wide
principles for the use of probabilistic analysis in
risk assessments, the Guiding Principles for Monte
Carlo Analysis (www.epa. gov/ncea/mcpolicy. htm).
The principles represent an important scientific
advance in the way EPA conducts risk assess-
24
-------
3. Advances in Risk Assessment
Figure 3-4. Simplified example of the use of Monte Carlo Analysis to
estimate the exposure of a population to an air pollutant. Expressing
exposure (or risk) in a range is more informative than generating a
single value to estimate exposure for an entire population.
ments. Highlights of the principles include recom-
mendations that analyses should:
Clearly state the purpose and scope of the
assessment and the methods used.
Discuss any highly exposed or sensitive groups,
such as children.
Describe what went into the risk calculations
and results.
Compare distributions of exposures to health-
based values, such as drinking water standards.
Assessment of Thyroid Tumors
In 1998, EPA published the Assessment of Thyroid
Follicular Cell Tumors (www.epa.gov/ncea/
thyroid.htm), which describes the procedures the
Agency will use in assessing the risks to people
from chemicals found to cause thyroid cancer in
laboratory animals. The guidance is noteworthy not
only for its contribution to assessing thyroid
cancer risks, but also because it presents an official
EPA position on when it is appropriate to deviate
from an important default assumption about cancer
risks. Historically, EPA assumed that as exposure to
a carcinogen decreases, the risk also decreases but
does not disappear entirely until the exposure
ceases, which is known as a linear extrapolation
approach to estimating risks (Figure 3-5). Under the
guidance, EPA will continue to use this approach if
a chemical causes thyroid cancer by damaging
DNA (i.e., it is mutagenic) or if no data are available
about how it causes cancer. The guidance also
clearly explains the criteria that must be met before
a nonlinear extrapolation
approach can be considered.
Such an approach presumes a
threshold exists below which
cancer is unlikely to occur.
The guidance bases its
conclusions on improved
scientific understanding of how
chemicals cause thyroid cancer.
Though the only verified
thyroid carcinogen in humans
is ionizing radiation, thyroid
tumors are fairly common in
long-term studies of chemicals
in rodents. The thyroid gland
sets the metabolic rate of the
body's cells, based on a
hormone feedback loop
regulated by the pituitary
gland. Long-term exposure to
some chemicals can disrupt this
feedback loop, leading to
thyroid tumors. EPA's guidance
concludes that brief, low-level exposures to
chemicals, however, would not be likely to cause
sustained disruption of the thyroid-pituitary loop
and thus would not pose cancer risks in people.
The guidance applies only to thyroid tumors, not
cancer of other organs.
Improvement in Risk Assessment
Methods and Models
In addition to preparing guidance on how to
conduct risk assessments, ORD develops methods,
models, and data that can be adapted for individual
risk assessments. Two recent accomplishments
include the following:
Exposure Factors Handbook
Assessing exposure is one of the major steps in
performing a human health risk assessment. To
accurately assess exposure, not only must the
concentrations of the chemical of concern be
ascertained, but the activities that lead individuals
to come in contact with the chemical must be
understood as well. Examples include the amount of
water people drink, how much air they breathe,
where they work, and the amount of time they
spend outdoors. Collectively, these activities and
characteristics are termed exposure factors. In
August 1997, ORD advanced the state-of-the-
science for exposure assessment by publishing a
revised and expanded Exposure Factors Handbook
fwww.epa.gov/nceawwwl/exposure.htmj, originally
issued by ORD in 1989.
25
-------
3. Advances in Risk Assessment
Risk of Cancer
(response)
Exposures where
^ risks have been
observed
Exposures where
risks are unknown
> (estimates of risk
must be extrapolated)
Linear extrapolation
Exposure (dose) _ _ _ Nonlinear extrapolation (threshold approach)
Figure 3-5. Linear and nonlinear extrapolation of cancer risks.
Studies of risks from exposure to chemicals often find that cancer risks
decrease as exposure drops, but risks at low levels of exposure are
difficult to measure and frequently unknown. Historically, risk
assessors have usually used a linear extrapolation for carcinogens to
estimate low exposure risks. EPA's thyroid tumor guidance discusses
when a nonlinear approach, which assumes a threshold of exposure
below which cancer risks do not exist, may be appropriate.
Among the many new sets of information in the
revised handbook are data on consumer product
use, drinking water rates for children, and daily
intake of food by region and age.
The Handbook recommends exposure factors to be
used in estimating chemical exposure for different
age and gender groups, including national means
and ranges of values for water and food ingestion,
body weights, and inhalation rates. Factors
important in determining contaminant exposure for
potentially sensitive groups, such as pregnant
women and children, are included. For example, the
Handbook estimates how much drinking water is
ingested by pregnant women, children, and others
each day (Table 3-1). As the only authoritative
source for peer-reviewed exposure factors, the
Handbook has become an indispensable resource
for risk assessors within and outside of EPA.
National Human Activity Pattern Survey
Daily activities, such as smoking, driving, and time
spent in different locations, make up an important
set of exposure factors discussed in ORD's
Exposure Factors Handbook. They are vital to
exposure assessment because they impact how
often, how long, and how many pollutants people
are exposed to. Mathematical exposure models have
been developed to incorporate daily activity
information into risk assessments, but have been
limited by the lack of sufficient data on the wide
range of activities people engage in. The ORD-
sponsored National Human Activity Pattern Survey
(NHAPS) fills this vital human exposure research
need by providing comprehen-
sive human activity and
location information on a
national level that can be used
in risk assessments. NHAPS
findings have been incorpo-
rated into the Exposure
Factors Handbook and into
ORD's Consolidated Human
Activity Database, a compen-
dium often human activity
databases available on the
Internet at www.epa.gov/
chadnetl/index.html.
NHAPS consisted of a two-
year telephone survey of more
than 9,000 individuals carried
out by the University of
Maryland Survey Research
Center to collect detailed
information on the time,
location, and nature of activi-
ties relevant to estimating pollutant exposure.
Technical reports and journal articles describing the
findings of the Survey were completed in 1998.
Participants reported all activities for specific
locations (e.g., at home vs. in a car) for the prior 24-
hours, as well as information on activities that can
increase chemical exposure, such as smoking,
cooking, washing, or cleaning. Key findings of the
Survey (see also Figure 3-6) included:
The largest amount of time was spent indoors in
homes (69%).
Almost 6% of time was spent in vehicles.
Nearly 8% of time was spent outdoors.
Children, particularly those of school-age, spent
more time outdoors than adults. Children and the
elderly spent less time in vehicles than younger
and middle-age adults.
Precedent-Setting Risk Assessments:
Mercury Report to Congress
In addition to preparing guidance on how to
conduct risk assessments and generating underly-
ing data and methods for risk assessment, ORD
conducts selected risk assessments that set a
precedent for EPA by testing innovative ap-
proaches or have national implications that cut
across many Agency programs. A noteworthy
example is the eight-volume Mercury Report to
Congress mandated by the Clean Air Act that EPA
issued in December 1997. This report was a multi-
year effort, involving scientists from across EPA, to
evaluate the impact of air emissions of mercury on
26
-------
3. Advances in Risk Assessment
Summary of Drinking Water Intake Rates
Age Group/Population
<1 year
1 - 10 years
11 - 19 years
Adults
Pregnant Women
Average Amount
Consumed Per Day
(Liters/ Day]
0.30
0.74
0.97
1.4
1.2
Amount Consumed Per Day
Per Unit of Dody Weight
(Milliliters/Kilogram- Day]
44
35
18
21
18
Adults in high activity/ 0.21 to 0.65 liters/hour, depending on temperature and]
hot climate conditions activity level
Table 3-1. When assessing risks from contaminants found in drinking water, risk assessors need
information on how much water people consume. This table presents the drinking water intake values
that the Exposure Factors Handbook recommends be used in risk assessment (excerpted from Table 3-
30 of the Handbook). Drinking water intake is one of many exposure factors presented in the Handbook.
human health and the environment and review
available control technologies. ORD contributed a
series of innovative models used in the report that
examine atmospheric and water transport,
bioaccumulation, exposure, and health effects of
mercury.
Mercury is considered a serious concern, because
it persists and accumulates in the environment and
can damage the nervous system of humans and
wildlife, especially during development. Atmo-
spheric emissions can reach waterways as a result
of rainfall or runoff and then build up as methy Imer-
cury in the tissues of predatory fish that feed on
contaminated smaller fish (Figure 3-7). Contami-
nated fish also can be eaten by people and wildlife.
The report estimated that about 158 tons of mercury
were emitted into the air in 1995 from all U.S.
industrial sources. Major emission sources include
electric utilities, incinerators, industrial boilers, and
chloralkali plants. Consumption of contaminated
fish is the greatest source of human exposure to
mercury. The report noted that, given the mercury
levels generally found in commercial fish, it is safe
to eat fish and other seafood in moderation from
grocery stores and restaurants. Pregnant women
should heed state and federal fish advisories for
mercury due to risks to the developing fetus.
Where Do We Go From HereP
ORD is continuing work in the three areas de-
scribed above: developing risk assessment
guidance, improving methods and data for risk
assessment, and preparing precedent-setting risk
assessments.
One important project involves updating the
Guidelines for Carcinogen Risk Assessment, which
EPA originally published in 1986. Since that time,
significant scientific progress has been made in
understanding how cancer develops, and EPA's
experience with the 1986 Guidelines has revealed
several limitations in their approach to cancer risk
assessment. EPA proposed revised guidelines in
April 1996 and is now preparing final guidelines.
Historically, EPA has concentrated on determining
how many tumors a chemical induces at certain
doses. The proposed revisions give greater weight
to accounting for the molecular events leading up
to tumor formation - information known as mode-
of-action data. Expanding the breadth of informa-
tion considered will strengthen the biological
foundation of cancer risk assessments and reduce
uncertainties. Among the other proposed changes
are providing additional guidance for assessing
risks to potentially susceptible populations, such
as children.
Another risk assessment guidance project under-
way is the development of guidance for Cumulative
Risk Assessment. To date, most risk assessments
have evaluated one chemical at a time, or at most a
few related chemicals. However, several recent
reports from the National Academy of Science and
others, as well as the 1996 Food Quality Protection
27
-------
3. Advances in Risk Assessment
Act, have called on federal
agencies to do more to account
for the cumulative exposure to
multiple chemicals. In the real
world, human populations are
exposed to many chemicals
and other stressors simulta-
neously, which could result in
health risks that differ from
what would be expected
considering only single
chemical exposures. Chemicals
that cause similar forms of
toxicity, for example, might lead
to risks that should be added
together in estimating the
likelihood of an adverse health
effect. The guidance under
development is intended to
provide EPA with a consistent
and scientifically credible
approach to assessing
cumulative risk both for
multiple stressors and for
multiple routes of exposure
(e.g., inhalation, ingestion,
exposure through the skin).
A notable ORD project that is
expected to improve the ability
of the Agency to carry out
cumulative risk assessments is
the National Human Exposure
Assessment Survey, or
NHEXAS. Begun in 1993, the
effort is examining the feasibil-
ity of conducting human
exposure studies for multiple
chemicals for multiple routes of
exposure at a regional scale.
The study is focusing on
actual exposures of individuals
to contaminants in their daily
lives. Human exposures to
metals, pesticides, and other
chemicals were measured in the
air, food and beverages, and in
residential soil and dust. Levels of chemicals were
measured in blood and urine samples, and volun-
teers completed questionnaires to help identify
their activity patterns and possible sources of
exposure. Hundreds of volunteers participated in
three interrelated studies located in Arizona, the
Midwest, and Maryland. As the results are
analyzed over the next several years, they should
prove useful in understanding chemical exposures
of concern as well as aiding in planning and
conducting future exposure studies.
Time cH Day
Figure 3-6. Plot showing the percentage of respondents reported being
in one often different locations for different times during the 24-hour day,
from the National Human Activity Pattern Survey.
Figure 3-7. Transport of mercury in the environment. Mercury is
emitted into the air from coal-fired utility boilers, municipal waste
combustors, and a variety of other sources. It subsequently is
deposited on the land and water. After reaching bodies of water, it can
accumulate in the tissues of predatory fish, which become the principal
source of human mercury exposure.
Finally, ORD continues to prepare precedent-
setting and cross-cutting risk assessments. Projects
underway include a comprehensive reassessment
of the risks of dioxin, Air Quality Criteria Docu-
ments for pollutants such as paniculate matter and
carbon monoxide, a review of the health risks of
diesel engine emissions, and a number of Agency
consensus positions on other chemicals that are
being compiled on the Internet-accessible Inte-
grated Risk Information System (www.epa.gov/iris).
28
-------
4. Mid-Atlantic Integrated Assessment
(NIAIA)
Since EPA was founded in 1970, the nation has
devoted enormous effort to cleaning up pollution
and protecting the environment. Businesses and
government agencies have spent millions of dollars
to carry out and comply with laws designed to
protect the environment such as the Clean Air Act,
Safe Drinking Water Act, Superfund, and the
Endangered Species Act, not to mention state and
local laws and initiatives. Over this same time
period, the U.S. population has grown by one-third
to more than 270 million people and economic
activity has accelerated. Rising consumption of
resources such as water, forests, and fossil fuels
has the potential to increase pollution and deplete
future supplies if not they are not managed
sustainably. Overall, is the environment getting
better or worse? Why?
These questions may sound simple, but answering
them is not, in part because "the environment"
encompasses such a wide range of natural compo-
nents. Over the past few decades most environmen-
tal monitoring programs have focused on relatively
limited geographic areas or narrow measures of
environmental quality, such as commercial fish
stocks or concentrations of major air pollutants.
Programs to monitor the status of the environment
more broadly face sizable scientific and logistical
challenges related to what, where, and how often to
measure. As a result, concerted efforts to integrate
many kinds of ecological components across large
geographic areas have been much rarer, despite
recommendations by the National Research
Council, the U.S. General Accounting Office, EPA's
Science Advisory Board, and others beginning in
the late 1970s.
In 1989, ORD responded to these recommendations
by creating the Environmental Monitoring and
Assessment Program (EMAP) to estimate the
status and trends of the condition of the nation's
ecological resources. The goals of this effort
included evaluating the cumulative effect of
programs designed to protect the environment and
detecting emerging environmental problems before
they became widespread or irreversible. During its
early years, EMAP focused on developing scientifi-
cally sound "ecological indicators" (see chapter on
Ecological Indicators for more information) and
designs for monitoring major classes of natural
resources at a national scale.
In 1995, ORD decided to pursue a major new
direction for EMAP as a result of a review of
national monitoring programs by a committee of
federal agencies (the Committee on Environment
and Natural Resources of the National Science and
Technology Council). A component of the new
approach, known as the Mid-Atlantic Integrated
Assessment (MAIA), involves conducting
intensive assessments of the environment on a
regional scale as a model that can be transferred to
other areas of the country. The Mid-Atlantic region
of the United States was selected as the area to be
studied for two reasons: first, the strong interest by
both EPA Region 3 (the regional office that guides
Agency activities in the Mid-Atlantic) and the Mid-
Atlantic states; and second, the extensive environ-
mental data already available for the region.
Approach of the Mid-Atlantic
Integrated Assessment
The Mid-Atlantic region being assessed by MAIA
encompasses the states of Pennsylvania, West Virginia,
Maryland, Delaware, and Virginia (Figure 4-1), with
adjacent portions of New York, New Jersey, and
North Carolina also included in some assessments.
The Mid-Atlantic contains a mosaic of ecological
systems - lakes, streams, forests, agricultural areas,
wetlands, and estuaries. It encompasses the
Chesapeake Bay, the largest estuary in the world,
and an uplands area that is among the most
biologically diverse regions in the country. It is also
home to over 35 million people and has experienced
some of the most rapid population growth, indus-
trial development, and intensive agriculture in the
country. A wide range of environmental problems
has accompanied this growth and development.
Beginning as a partnership between ORD and EPA
Region 3 to assess these environmental problems,
MAIA has grown to include other federal, state,
tribal, and private environmental organizations. The
overall goal of MAIA is to improve environmental
decision making by incorporating the best available
information on the condition of resources along
with an improved understanding of the relative
29
-------
4. Mid-Atlantic Integrated Assessment (MAIA)
-------
4. Mid-Atlantic Integrated Assessment (MAIA)
Recent Accomplishments
MAIA research has provided "proof of concept"
for large-scale monitoring, emphasizing regional-
scale rather than site-specific assessments. It sets a
standard for integrating multiple environmental
measures and for analyzing and presenting
environmental data to the public. In 1998, MAIA
completed its first two major MAIA products, An
Ecological Assessment of the United States Mid-
Atlantic Region: A Landscape Atlas and Condi-
tion of the Mid-Atlantic Estuaries. While both
reports assess the Mid-Atlantic, the atlas focuses
on conditions on land and the estuary report
focuses on the waters that make up the region's
extensive estuaries. These and other MAIA
products are available through the MAIA website
(www.epa.gov/maia).
The Landscape Atlas
The Landscape Atlas is a report that analyzes and
interprets environmental conditions in 125 water-
sheds of the Mid-Atlantic, with a watershed
consisting of an area of land drained by a single
river or other body of water. It compares water-
sheds using 33 indicators of landscape condition,
which were derived from satellite imagery and
databases of features such as soils, elevation, and
human population patterns. For some of the
indicators, the Atlas compares watersheds in the
Mid-Atlantic to watersheds in the other lower 48
states. An example of a landscape indicator used in
the Atlas is the proportion of a watershed that has
agriculture or urban land cover (with more land of
this type considered less ecologically desirable)
(Figure 4-2). The Atlas identifies patterns of land
cover and land use across the region with respect
to potential human impacts, water resources,
forests, and landscape change. The level of detail
and comparability seen in this report has never
before been achieved across such a large area.
The report concludes that mountainous watersheds
in the Mid-Atlantic contain the least amount of land
in agricultural or urban use (and thus the greatest
proportion of forests), while coastal areas have the
lowest proportion of forests (Figure 4-3). Compared
to many parts of the country, Mid-Atlantic water-
sheds have relatively favorable values for forest
conditions, including forests along streams. An
overall ranking based on factors such as popula-
tion, road density, and amount of forests finds that
the watersheds with the most desirable landscape
conditions are in southern West Virginia and
Figure 4-2. Proportion of watershed area with agriculture or urban land cover, from the Mid-Atlantic
Landscape Atlas. In the atlas, each watershed is coded in one of five colors ranging from green (more
desirable ecologically) to red (less desirable) for each condition that is evaluated. Each quintile
(represented by one color) contains one-fifth of the watersheds.
31
-------
4. Mid-Atlantic Integrated Assessment (MAIA)
Figure 4-3. Proportion of
watershed area that is forested
in the Mid-Atlantic region, from
the Mid-Atlantic Landscape
Atlas. Watersheds with relatively
low proportions of forests are
clustered around major urban
centers and Chesapeake Bay.
Virginia and in north-central
Pennsylvania. Most of the
watersheds with the least desirable
conditions are clustered around
the major metropolitan areas of
Baltimore, MD, Washington, DC,
Pittsburgh, PA, and Norfolk, VA.
The ecological snapshot provided
by MAIA landscape work can be
applied to current environmental
and regional economic develop-
ment decision making and will
allow future trends in the region to
be examined. For example, EPA is
using MAIA land cover analyses
in evaluating the potential impacts
of a new, large-scale coal mining
practice in the region that involves
removing mountain tops and filling in valleys.
Landscape analyses, along with other kinds of
assessments, have indicated that mountain-top
forests in areas proposed for mining are in relatively
pristine condition and contain high quality streams.
These findings are useful to EPA Region 3 in
making regulatory decisions about mountain-top
mining.
Condition of the Mid-Atlantic Estuaries
Estuaries are transitional zones where sea water
mixes with fresh water flowing off the land. They
provide habitats for many birds, mammals, fish and
other aquatic life, and are important assets that
people use in a variety of ways. The Delaware
Estuary, Chesapeake Bay, and the coastal bays
along the Delmarva (Delaware-Maryland-Virginia)
Peninsula comprise the Mid-Atlantic estuaries and
are the subject of a second major MAIA report,
Condition of the Mid-Atlantic Estuaries. This
report synthesizes information gathered from
various state and federal programs on the condition
of the Mid-Atlantic estuaries from the early- to mid-
1990s and describes how these estuaries have
changed. The MAIA effort for estuaries reflects
important scientific advances in large-scale
assessment. Because of carefully designed
programs to make sure sampling data were thor-
ough and representative, conditions across all of
the estuaries can be compared for the first time.
The report identifies the location of problem areas,
and provides estimates of the percentage of
estuaries in good, moderate, or poor condition
based on specific environmental indicators.
Indicators included water quality, sediment
contamination, habitat change, and condition of
living resources such as shellfish, fish, and
waterfowl.
For example, the Chesapeake Bay suffers from over-
enrichment with nutrients, which can lead to algal
blooms and subsequent depletion of oxygen that
threaten plant and animal life. As shown in Figure
4-4, about one third of Chesapeake Bay bottom
waters are considered moderately or severely
affected by low oxygen levels in the summer.
Nutrient levels are declining, however, in response
to measures such as improved wastewater manage-
ment. The Delaware Estuary is impacted by the lack
of water clarity and by toxic contaminants associ-
ated with urbanization and industrialization. The
coastal bays are the least degraded estuaries in the
Mid-Atlantic, but are threatened by encroaching
urbanization. Across the region, oyster harvests
have declined drastically over the past 100 years
due to disease and other factors (Figure 4-5).
32
-------
4. Mid-Atlantic Integrated Assessment (MAIA)
* a
Delaware River
0 « 12
Dissolved Oxygen Conditions
Graph Sources: Strobel e/o/., 1995; Paul e/o/., 1997
Map Sources: Chaillou etal., 1996; CBP, 1997; USEPA, 1995; Magnien
ad., 1993; Strobel aal., 1995; Paul
etal., 1997
Figure 4-4. Distribution of summer-time dissolved oxygen within one meter of bottom sediments across
estuaries in the Mid-Atlantic region, from Condition of the Mid-Atlantic Estuaries. Conditions of low levels
of dissolved oxygen (hypoxia) can harm bottom-dwelling organisms and are most widespread in the
middle portions of the Chesapeake Bay and the lower Potomac River.
The report is proving useful to environmental
managers. For example, Maryland has established a
National Estuary Program to further protect
Maryland's coastal bays based on the report's
findings.
Where Do We Go From HereP
MAIA efforts are proceeding on several fronts.
First, MAIA participants have built upon the work
of the estuaries report by developing a comprehen-
sive, integrated monitoring design that consists of
more than 700 stations throughout the Mid-Atlantic
estuaries. For the first time, a common set of key
indicators of overall environmental quality are
being adopted by the various state and federal
organizations studying these estuaries. A number
of other "State of the Region" MAIA reports are
also underway, including reports on streams,
groundwater, forests, and agricultural lands.
MAIA is also beginning to develop methods to
integrate the individual resource assessments
prepared for the State of the Region reports. One
effort focuses on the use of ecological indicators
that provide information across multiple categories
of resources. Two promising examples are indica-
tors based on monitoring of bird and amphibian
populations. MAIA is also exploring ways to
connect indicators of environmental condition with
assessment of public health. The goal is to prepare
integrated assessments that bring together
information about ecological and human health
effects, social goals, economics, politics, and law in
a way that is useful to policy makers. Such efforts
will be complemented by MAIA efforts to develop
cost-effective and reliable approaches for managing
or restoring ecosystem components, such as
habitats along streams and rivers (riparian zones).
Additionally, MAIA plans to move from assessing
current environmental conditions to predicting
33
-------
4. Mid-Atlantic Integrated Assessment (MAIA)
140
Figure 4-5. Annual oyster harvest for Mid-Atlantic estuaries, from Condition of
the Mid-Atlantic Estuaries. Gaps in the late 1800s and early 1900s represent
missing data. Harvests have plummeted due to disease, pollution, and over-
harvesting.
future conditions. ORD's Regional Vulnerability
Assessment (ReVA), which, like EMAP, will initially
focus on the Mid-Atlantic, will be a major compo-
nent of this effort. ReVA will develop the next
generation of measurements and tools to assess the
simultaneous impact of stressors such
as urbanization, industrial and
agricultural pollution, and climate
change to make regional predictions
of environmental conditions over the
next five to 25 years. This will
improve the ability of decision
makers to evaluate the consequences
of various economic, land use, and
environmental choices before they
are made. The first phase of ReVA is
the Mid-Atlantic Stressor Atlas,
currently in draft, which looks at
stressors such as mining, agro-
chemicals, ground level ozone, and
land use change.
In its short existence, MAIA has
forged alliances among federal and
state agencies to cooperatively
answer questions about the
condition of the environment and
whether it is improving or declin-
ing. The success of the program
has prompted ORD to begin a new
intensive regional assessment, this
one for the western United
States (Figure 4-6). The
EMAP Western Pilot will
test the approach used
by MAIA on a much
larger scale in a region
that contains ecosystems
not present in the Mid-
Atlantic, such as
mountainous and arid
zones. The Western Pilot
is planned to be a five-
year cooperative effort
between EPA, the states,
and tribal nations
concentrating on
estuaries, inland waters
(e.g., streams), and
landscape characteristics.
The program will assess
problems of critical
importance to resource
managers and environ-
mental decision makers
throughout this region.
^^^^^^^^^^ Together, MAIA, ReVA,
and the EMAP Western
Pilot are advancing the scientific basis for evaluat-
ing the condition of the environment in ways that
can be applied throughout the United States.
.
Figure 4-6. The geographic scope of the EMAP Western Pilot.
34
-------
5. Endocrine Disrupters
In the 1950s, scientists became concerned by
dramatic declines in the reproductive success in
bird populations, which were experiencing
problems such as hatching failures due to eggshell
thinning. These effects were found to be caused
by exposure to a class of pesticides known as
organochlorine pesticides, particularly DDT.
Scientist Rachel Carson opened the eyes of the
American public to these and other effects of
pesticide use with her 1962 book Silent Spring. In
the 1970s, doctors traced reproductive tract
cancers in women to DBS, a drug similar to the
hormone estrogen, which was used by their
mothers in the 1950s to prevent miscarriage.
Although these two events may outwardly appear
dissimilar and unrelated, they both are cases of
chemicals interfering with the functioning of the
endocrine (hormone) system. Overtime, scientists
began piecing together this and other evidence of
adverse effects in wildlife and humans to develop
a hypothesis that chemicals could be causing
widespread harm by disrupting the endocrine
system.
The endocrine system plays a key role in the
development, growth, reproduction, and behavior
of humans and wildlife. Endocrine glands (Figure
5-1) produce hormones that act as chemical
messengers, traveling through blood to tissues
and organs where they can bind to specific cell
sites called receptors. By binding to receptors,
hormones trigger numerous responses, such as
the release of eggs from ovaries. Hormones are
tightly regulated by the body, and exposure to
chemicals that alter their function may result in
abnormal growth and development. The conse-
quences of hormonal disruption during an animal's
development can be profound and long-lasting,
and developing organisms are therefore especially
at risk. Chemicals that interfere with any aspect of
hormone production, activity, or elimination in the
endocrine system are referred to as endocrine
disrupting chemicals (EDCs). They are also
sometimes referred to as hormonally-active
agents. Suspected EDCs include chemicals among
the following classes of compounds: pesticides;
polyhalogenated aromatic hydrocarbons; plasticiz-
ers; industrial surfactants; pharmaceutical agents;
and substances naturally found in some plants
(phytoestrogens).
Potential Effects on Wildlife and Humans
Evidence of potential effects from EDCs has been
collected primarily through laboratory animal
experiments and documentation of effects in wildlife
in specific contaminated ecosystems such as the
Great Lakes. To date, problems have predominantly
been identified in wildlife species with relatively
high exposures to specific compounds (e.g., DDT,
PCBs, and dioxins), or in domestic animals consum-
ing plants with high levels of phytoestrogens.
Pituitary
Parathroids
Thyroid
Thymus
- Adrenals
Pancreas
"*\- Ovaries
Tostis
i i-i H,il,:-1
Figure 5-1. The human endocrine system.
35
-------
5. Endocrine Disrupters
Examples of effects on wildlife potentially due to
EDCs (Figure 5-2) include the following:
Birth defects in a Lake Michigan bird population
(cormorants) exposed to PCBs and other
compounds.
Nearly complete mortality of young Lake Ontario
lake trout, presumably resulting from exposure to
dioxin-like compounds.
Abnormal reproductive development in alliga-
tors in Lake Apopka, Florida, following a
pesticide spill.
Simultaneous presence of both male and female
reproductive organs (imposex) in mollusks
exposed to chemicals (alkyltins) used to prevent
the growth of organisms such as barnacles and
algae on ship hulls.
Synthesis in male fish living near sewage
outfalls of a hormonally-regulated protein
(vitellogenin) normally found only in female fish.
In humans, in addition to the well-documented
effects of DBS, studies have indicated that PCBs
and related chemicals may cause developmental
neurological problems in exposed children. Further-
more, scientists have speculated that EDCs could
be responsible for such effects as reported declines
in the quality and quantity of sperm production
over the last four decades and increases in certain
cancers (breast, prostate, testicular) that may have
an endocrine-related basis.
Despite documented cases of endocrine disruption,
the scientific community has not reached a consen-
sus on the extent of the problem. Information about
how, at the cellular level, apparent EDCs are
causing their effects is almost always lacking.
Knowledge about the effects of EDCs at low doses
and the levels at which exposure to EDCs occurs is
limited. To answer the many questions surrounding
the endocrine disruption hypothesis, concerted
research programs are needed. OKD is contributing
to this research. To understand ORD's role, it is
Figure 5-2. Chemical
contamination has been found
to interfere with reproductive
functioning in a number of
wildlife species. Populations of
several species of raptors (left)
dropped precipitously before
DDT was banned in the United
States. Alligators (right)
developed abnormal
reproductive organs following
pesticide spills that reached
Lake Apopka, Florida.
helpful to understand the context of national and
international research and recent legislative
developments affecting EPA.
A Coordinated Federal Program
Given EPA's mandate to protect both public health
and the environment, the Agency has for several
years taken a leadership role in investigating
endocrine disruption. Other federal agencies are
involved as well. To coordinate research across the
federal government, the Committee on Environment
and Natural Resources (CENR) of the National
Science and Technology Council (NSTC) convened
an endocrine disrupter Working Group, chaired by
EPA, in 1996. The NSTC advises the President and
federal agencies on directions for national research
and development efforts. The Working Group
developed an inventory of federal research
programs, identified high priority research gaps in
the federal portfolio, and developed a national
research framework (www.epa.gov/endocrine).
These efforts have helped both ensure cooperation
on endocrine disruption research within the federal
government and refine the research areas on which
EPA is concentrating.
An Environmental Issue That Transcends
National Boundaries
Many suspected EDCs are long-lived and mobile in
the environment, meaning they can readily be
transported across national boundaries. This has
contributed to shared international concern about
their impacts. Additionally, the breadth of the
scientific uncertainties about the causes, effects, and
solutions for this issue necessitate international
cooperation and communication. The international
community has responded on both the research and
the policy fronts. Internationally, at least 25 major
scientific workshops have been held over the past
five years to assess the scope and magnitude of the
potential EDC problem, most of which have involved
ORD participation. Nations have also begun
36
-------
5. Endocrine Disrupters
Understanding the sources, fates and
transport of EDCs
Developing mitigation and
pollution prevention strategies
Understanding the risks from
exposure to environmental
levels of EDCs
Und
jrstanding exposure to EDCs
Developing methods to screen for
endocrine effects
Assessing the current state of the
science for EDCs
Understanding the potential
health and ecological
effects of EDCs
Figure 5-3. Some of the major research and assessment activities
conducted by ORD in examining potential disruption of the endocrine
system in humans and wildlife due to environmental contaminants.
international initiatives such as developing screen-
ing and testing guidelines for EDCs through the
Organization of Economic Cooperation and Develop-
ment (OECD) and other international venues.
Legislation
Concerns about endocrine disruption in the
environment have impacted national legislation as
well. In 1996, the Safe Drinking Water Act Amend-
ments and Food Quality Protection Act (FQPA)
were enacted. Aspects of these two laws mandate
the development of a screening and testing
program to evaluate the potential of chemicals
found in drinking water and food to have hormonal
activity. ORD supports EPA's Office of Water and
Office of Prevention, Pesticides, and Toxic Sub-
stances in meeting the requirements of these laws.
EPA is developing an Endocrine Disrupters
Screening and Testing Program (EDSTP), taking
into consideration recommendations received from
an advisory committee in which ORD was an active
member. The scientific questions that must be
answered to create a successful testing program
have placed additional demands on ORD research.
ORD's Research Program to
Answer Outstanding Scientific
Questions
Based upon recognition of the potential scope of
the problem, the possibility of serious effects on
the health of populations, and the persistence of
some endocrine-disrupting agents in the environ-
ment, ORD identified research
on endocrine disruption as
one of the six high-priority
topics in its 1996 Strategic
Plan. In 1998, ORD published
an endocrine disruption
research plan (www.epa/gov/
ORD/WebPubs/final/) that
presented ORD's research
priorities for this topic. ORD
research reflects the national
and international efforts to
prioritize research needs and
the legislative mandates
described previously. As
described in the Introduction,
ORD uses the risk assess-
ment/risk management
framework to organize its
research approach in solving
scientific and technological
problems. This framework is
visually portrayed in the inner ring in Figure 5-3,
with ORD's major endocrine disruption activities
depicted by the outer ring. Major uncertainties exist
in virtually every aspect of assessing the impact of
endocrine disruption. Key questions for ORD
identified in the research plan include the following:
Health and ecological effects: What effects are
occurring in exposed human and wildlife
populations? What are the chemical classes of
interest? What are the potencies of these
chemicals at low doses? Do testing guidelines
adequately evaluate potential endocrine effects?
How can experimental findings be extrapolated
from one system to another, such as from tissue
cultures to whole organisms? What are the
effects of exposure to multiple EDCs?
Exposure: How and to what degree are human
and wildlife populations exposed to EDCs?
What are the major sources and environmental
fates of EDCs?
Risk management: How can unreasonable risks
be managed? Are new technologies needed?
Recent Accomplishments
Assessing the Current State of the Science
In 1997, EPA's Risk Assessment Forum published a
report entitled Special Report on Environmental
Endocrine Disruption: An Effects Assessment and
Analysis (www.epa.gov/ORDAVebPubs/endocrine/).
The Risk Assessment Forum is staffed by ORD and
brings together expert panels from across EPA. The
report assessed the current state of the science for
37
-------
5. Endocrine Disrupters
endocrine disruption in humans and wildlife. For
human health, the report concluded that with few
exceptions (e.g., DBS, PCBs), a causal relationship
has not been established between exposure to a
specific chemical and an endocrine-mediated
adverse effect. For ecological effects, the report
noted that although a number of compounds can
affect development in invertebrates, fish, and
wildlife via the endocrine system, few examples
established the extent to which these effects have
had impacts on populations of organisms.
This report was significant for a number of reasons.
First, it represented a cross-Agency assessment of
the state of the science. Second, it resulted in the
development of an interim Agency position on
EDCs that has served to guide Agency decisions
since. Third, its release allowed the scientific and
regulated communities to know EPA's position on
EDCs. The conclusions of EPA's report have since
been largely supported by the endocrine disruption
report of the National Academy of Sciences,
released in August 1999.
Developing Methods to Screen for
Endocrine Effects
Literally thousands of different chemicals can be
found in commerce, most of which have never been
tested for their effects on the endocrine system.
The 1996 Food Quality Protection Act (FQPA)
recognized this by requiring EPA to develop a
screening and testing program for endocrine
disrupters. At the time the law passed, a standard-
ized and validated battery of tests for endocrine
disruption did not exist, and ORD is playing a
critical role in helping develop needed assays in
support of EPA program offices. These screening
assays will be used to determine whether more in-
depth, long-term test procedures are needed to
characterize potential risks to the endocrine system.
Because of concerns of potential adverse effects to
fish populations from EDCs, EPA has decided to
include within the battery of endocrine tests a
hormonal screening assay involving fish reproduc-
tion. ORD scientists have completed the initial
stages in developing such a screening test using
fathead minnows (Figure 5-4). In developing this
test, investigators designed techniques to measure
baseline or "normal" values of sex hormones
(estrogen and androgen) and vitellogenin (a protein
found in egg yolk that is regulated by estrogen)
that guide reproduction in this species. The assay
allows scientists to compare values obtained
following exposure to chemicals with unknown
effects with baseline values to see if the chemicals
may be impacting the endocrine system. Further
Figure 5-4. Using the fathead minnow, ORD has
developed a screen for potential adverse effects to
fish from hormonal effects of chemicals.
work is focusing on standardizing the procedures
for this 21-day reproductive function assay.
Given that testing the thousands of chemicals for
which little endocrine data exist could take substan-
tial funds and time, ways to prioritize the testing are
needed. ORD researchers made promising progress
in an approach to prioritizing testing by making
inferences about a compound's potential effects
based on its chemical structure. These inferences
draw upon what are known as quantitative struc-
ture activity relationships (QSARs). ORD research-
ers contributed PC-based modeling techniques for
predicting the ability of a chemical to bind to
receptors for estrogen and androgens, which could
indicate potential for disrupting endocrine func-
tions that involve these sex hormones. It is hoped
that when these techniques are fully validated, they
will be used to rapidly prioritize chemical databases
for testing.
Understanding the Potential Health and
Ecological Effects of EDCs
Given the many uncertainties regarding the
potential health effects of EDCs in humans and
wildlife, most of ORD's research program has been
focused on addressing these data gaps. The
research efforts have covered a broad set of
hormones and their pathways, in a variety of
species, that may be disrupted by numerous
environmental pollutants.
One aspect of endocrine functioning that ORD has
studied involves the thyroid gland. ORD found that
in laboratory rats, exposure to PCBs depressed
thyroid hormones during a critical period of
38
-------
5. Endocrine Disrupters
3
0
Q.
X
LU
00
o
Q.
S
n
o
a.
Low-frequency
Hearing loss
<
5
E
o
T
O
-------
5. Endocrine Disrupters
Figure 5-6. Testosterone and anti-androgens. The top left structure depicts testosterone, a natural male
hormone (androgen). The bottom right structure depicts a pharmaceutical drug, flutamide, considered
an anti-androgen because it binds to the same receptors as testosterone, thereby interfering with its
functioning. The fungicide vinclozolin (bottom left) is not a direct anti-androgen, but it breaks down into M2
(top right), which like flutamide, is anti-androgenic.
sensitive enough to detect them in water, soil, and
other environmental media. Therefore, amajorfocus of
ORD 's exposure program has been the development
of analytical methods for determining the extent of
EDCs in the environment. Once these tools are
developed, ORD scientists publish their findings in
the peer-reviewed literature so other scientists can use
them.
During 1997 to 1998, ORD chemists developed or
improved analytical tools for measuring a number of
suspected EDCs, including organochlorines, PCBs,
and polycyclic
aromatic
hydrocarbons
(PAHs).ORD
scientists then
tested these
tools in the Neuse River Basin of North Carolina
(Figure 5-8) as part of abroader effort to develop a risk
assessment of EDCs within this area. ORD scientists
also tested a variety of screening tools, such as
commercially available immunoassay kits, in the basin.
The U. S Geological Survey, Duke University, North
Carolina State University, the University of North
Carolina, and others collaborated with ORD in this
work. Suspected EDCs were analyzed in water, soil,
sediment, fish, and other selected plants and animals
at a low impact agriculture area, two high impact
agriculture sites, and a coastal site. Examples of
analytical tools developed and tested through this
effort were gas chromatography-mass spectrometry
(GC-MS) methods for detecting suspected EDCs at
low levels in white-tailed deer, fish tissue, natural
waters, and sediments. ORD chemists also developed
Figure 5-7. The effects of methoprene on leopard frog tadpoles
(Rana pipiens) at day six of development. The top panel depicts
normal development in a control organism. The bottom panel shows
the typical effects of methoprene on development. High levels of
methoprene exposure (458 ppb) resulted in complete mortality.
40
-------
5. Endocrine Disrupters
a high performance liquid chromatography (HPLC)
method for measuring different isomers (forms) of
PCBs in fish. Finally, ORD scientists designed and
tested a DNA-probe to screen for exposure offish to
EDCs. The probe takes advantage of the fact that
hormonally-active contaminants can switch on certain
genes in fish tissue, which are then detectable.
ORD used the results from the application of the
analytical tools to samples collected in the Neuse to
develop a computer model simulating the movement
of a suspected EDC, benzo-a-pyrene, through the
Middle Neuse Basin. Known as MEND-TOX, the
model predicts the movement of this compound
through different compartments of the environment,
such as water, sediment, and fish tissue. With future
data, this multi-media compartmental model will be
expanded and tested with other EDCs. When fully
developed, it will be a useful tool for assessing
possible exposure in other ecologically similar river
basins.
Demonstrating Leadership Nationally and
Internationally
As described in the introduction to this chapter, EPA
is leading an interagency Working Group, under the
auspices of the CENR, to coordinate endocrine
disruption research activities of the federal govern-
ment. As part of these efforts, ORD led the 1998
revisions to the inventory of EDC projects funded by
the federal government that ORD originally created in
1996. The inventory is an Internet-accessible database
(available atwww.epa.gov/endocrine) that can be
searched by topics such as chemical, species, or
hormonal effect. The 1998 update added projects from
Canada, Japan, and Europe to the U.S. component,
bringing the inventory to more than 700 projects. The
inventory has been used by the CENR Working
Group to assess how well the critical research needs it
identified were being addressed by government
research and to develop a multi-agency grants
program to fill the biggest gaps.
Where Do We Go From HereP
The results of the EDCs research program high-
lighted in this report are providing valuable insights
into understanding the potential effects of EDCs
and their patterns of exposure for humans and
wildlife. Following the risk assessment/risk manage-
ment framework depicted inFigure 5-3, ORD
continues to focus on the most critical uncertainties
in determining whether humans and wildlife
populations are being impacted by levels of
endocrine disrupters in the environment and in
identifying the sources of those exposures. As
these issues become further resolved, ORD will
place greater attention on the development of an
North Carolina River Basins
Neuse
Figure 5-8. The Neuse River Basin of North Carolina has been the focus of ORD studies on how to
measure environmental concentrations of chemicals that may disrupt the endocrine system. Map
prepared by the North Carolina Center for Geographic Information & Analysis.
41
-------
5. Endocrine Disrupters
integrated risk assessment framework for endocrine
disruption and mitigation strategies to reduce risks.
ORD also will continue to lead national and
international efforts to coordinate endocrine
disruption research programs.
ORD is committed to continued support for the
development of methods and their standardization
and validation for the screening assays required by
the Food Quality Protection Act. Particular atten-
tion will be focused on refining the mammalian tests
for estrogen, androgen and thyroid action, and in
developing and standardizing the amphibian and
fishbioassays. ORD will continue to refine and
validate the fish bioassay and PC-based modeling
techniques described above. These efforts are
being organized under an interagency Endocrine
Disrupter Screening Program Taskforce that is also
working to standardize test methods internationally
in conjunction with the OECD. The goal of the
screening programs is to efficiently identify
chemicals that may pose risks because of their
effects on the endocrine system.
ORD also plans a number of projects to investigate
the potential health and ecological effects of EDCs.
For example, ORD will study the effects of atrazine
on the endocrine system to help determine how this
herbicide causes mammary tumors in rats. ORD will
also examine the impact of phthalate esters on the
development of the male reproductive tract in light
of observations that some of these plasticizers can
act as anti-androgens. Another important research
area will compare endocrine disruption in different
kinds of animals so that results observed in one
class of animals can be more readily applied to
other classes.
In the area of exposure assessment, ORD will
continue laboratory and field research to develop
and verify techniques for measuring EDC concen-
trations and exposures in the environment. The
Neuse River Basin of North Carolina will remain a
focal point of the research, with efforts expanded to
include broader classes of chemicals and wider
coverage of the environment. ORD will also
continue to develop and refine models of the
movement of EDCs through the environment and
the resulting exposures experienced by people.
ORD is working with organizations such as the
National Institute of Environmental Health Sciences
(NIEHS) to identify the critical risk assessment
issues for EDCs. EDC risk assessment remains a
challenge because effects may be a consequence of
the cumulative impact of a wide range of contami-
nants that act in similar ways. Moreover, exposure
to substances other than contaminants, such as
compounds naturally found in some plants, may
confound interpretation of the endocrine effects of
contaminants. A case study involving integration
of human and ecological assessments for classes of
EDCs will be prepared as a central component of
this effort.
In the area of risk management, ORD is pursuing
efforts in several areas. One area is preventing or
controlling the release of suspected EDCs into the
environment, such as alkylphenols from sewage
treatment systems, airborne dioxins from industrial
sources, and other EDCs in chemical production
plants. ORD is also developing and evaluating
tools for destroying or containing EDCs in soils at
Superfund sites or in contaminated sediments.
Finally, in an effort to engage the best scientists in
the academic community in addressing the scien-
tific uncertainties regarding EDCs, ORD has carried
out a STAR grants effort devoted to endocrine
disruption since 1996. ORD has awarded grants
across several areas of the risk assessment
paradigm, such as developing methods to monitor
and model exposures, developing biomarkers of
exposure and effects, constructing short term
screening assays, studying the effects of endocrine
disrupters in wildlife populations, and investigating
human health endpoints that may be related to
endocrine disruption. In 1998, ORD linked its STAR
grants program with a broader interagency set of
grants for endocrine disruption issued under the
umbrella of the CENR and currently continues to
participate in this partnership.
42
-------
6. Environmental Risks to Children
To young children, the world is full of objects to be
grabbed, tasted, and chewed - edible or not. As they
explore their surroundings, children gain valuable
skills, but they also come into contact with chemicals
found in or on carpets, toys, furniture, lawns, and
many other items. These activities can lead to greater
exposure to chemicals than adults who share the
same environment. Because their bodies process
chemicals differently than adults and they experience
windows of vulnerability as their bodies develop,
children can also be more sensitive to toxic effects.
For example, mercury and lead appear to harm the
developing nervous systems of children at levels
that do not harm adults. Infants and young children
also eat different types of food and consume more
food and fluids per unit of body weight than adults.
These differences between children and adults have
gained heightened attention over the past decade. In
1993, the National Academy of Sciences published a
report entitled Pesticides in the Diets of Infants and
Children. In 1995, EPA announced a policy to
explicitly consider children when assessing environ-
mental risks. The Food Quality Protection Act of
1996 and the Safe Drinking Water Act Amendments
of 1996 also require EPA to give special attention to
children. In 1997, the President signed an Executive
Order directing federal agencies to give a high
priority to protecting children from environmental
health and safety risks.
In implementing these directives, EPA has realized
that scientific knowledge is not yet sufficiently
advanced to accurately assess environmental risks
to children. In response, ORD is carrying out a
research program to fill in gaps in understanding
about how children are exposed to environmental
contaminants, what health effects these contami-
nants might cause, how to assess risks to children,
and how to prevent or reduce exposure to these
contaminants. The research also involves studying
known health problems of children, such as asthma,
to see what role environmental exposures may play
(Figure 6-1).
Recent Accomplishments
Exposure of Children to Environmental
Contaminants
Though most people think of pollution as existing
outdoors, much of children's exposure to chemicals
actually occurs indoors. Children spend most of
their time indoors, where chemicals can be released
(e .g., from consumer products or building materials)
or infiltrate the home from outdoors. Chemicals can
also be tracked-in with contaminated soil or
brought inside on workplace clothing. ORD is
studying how children are exposed to potentially
harmful chemicals that occur indoors and are slow
to degrade. Pilot studies completed by ORD
scientists in 1997 evaluated the exposure of
children to contaminants in their homes and nine
day care centers in the Durham, North Carolina
area. The chemicals studied included PCBs,
polycyclic aromatic hydrocarbons (PAHs), phenols,
and various pesticides. The studies found that for
some chemicals, children received most of their
exposure through the air, while for others most
exposure occurred through the diet. Generally,
differences in exposure among the day care centers
and the homes were small.
Figure 6-1. ORD is investigating the environmental
risks faced by children.
43
-------
6. Environmental Risks to Children
Health Effects of Environmental
Contaminants
In the field of toxicology, ORD is studying the
effects of chemicals on young animals both before
and after birth. Findings on how young animals are
affected can be used along with exposure informa-
tion to predict whether children may face similar
risks and, ultimately, in deciding whether additional
steps to protect children's health are needed. For
example, ORD lexicologists have studied
chlorpyrifos, a pesticide widely used to control
insects. They found that young laboratory rats were
five to seven times more sensitive to nervous system
effects than adult rats. Subsequent studies found
that enzymes produced by the body to detoxify the
pesticide were not as effective in young animals as
adults, resulting in the greater sensitivity observed.
Other research found that if pregnant rats were
exposed to the pesticide, fetuses could be exposed
as well. The development of the visual system also
appears to be affected by chlorpyrifos early in life,
based on experiments using birds, which are more
similar to humans than are rodents in the structure
and development of their eyes.
Other studies by ORD toxicologists found that the
developing fetus is at special risk to the adverse
effects of dioxin and related chemicals. Exposing
rodents to dioxin prior to birth (prenatally) caused
permanent alterations in the developing reproduc-
tive tract of both female and male offspring. Not
only can these alterations affect future reproduc-
tion, but they might lead to enhanced susceptibility
to cancer of the reproductive organs. Many of the
effects are not detectable until puberty or even later
in life. Another study found that prenatal exposure
to dioxin resulted in long-term suppression of one
form of immune response (delayed-type hypersen-
sitivity) important in defending against certain
bacterial and viral infections.
Reducing Exposure to Environmental
Contaminants
ORD scientists collaborated with the Robert Wood
Johnson Medical School and the Centers for
Disease Control and Prevention (CDC) to conduct
the Children's Lead Exposure and Reduction Study.
If children are exposed to excessive levels of lead
from ingestion of lead paint dust (or other sources),
they may suffer impaired nervous system develop-
ment. More than one million children under the age
of six have blood lead levels that may place them at
risk, and children living in poverty are dispropor-
tionately affected. Public health strategies have
been stymied by a lack of practical, proven preven-
tion methods. This study, carried out in Jersey City,
New Jersey, demonstrated an effective strategy for
reducing lead exposure. In families that received
education about reducing exposure and whose
homes were treated biweekly (by vacuuming with a
high efficiency vacuum cleaner and by mopping
uncarpeted surfaces), children's blood lead levels
decreased 17 percent over one to two years.
Where Do We Go From HereP
ORD is currently conducting or planning many
other activities to understand and lessen the
environmental risks children face. Examples include
the following:
Based on lessons in study design and implemen-
tation learned from the two children's exposure
pilot studies described previously, ORD is
planning a large-scale extension of the pilots
that will allow for more definitive conclusions to
be drawn about children's exposure to persistent
chemicals.
ORD researchers are collaborating with the
Minnesota Department of Health and the
University of Minnesota (under the STAR
program) to study children's exposure to
pesticides through their diet, the air, and skin
contact. The STAR program is also sponsoring
research on pesticide exposure among children
in Arizona, California, and Washington state.
ORD is supporting the National Health and
Nutrition Examination Survey (NHANES-IV),
conducted by the CDC, so that information on
children's exposure to certain pesticides and
other environmental contaminants is collected.
ORD scientists are studying the mold
Stachybotras as a model for assessing and
managing risks from indoor molds. Stachybotras
has been implicated in the onset of sometimes
fatal pulmonary hemorrhaging in infants and, like
other molds, may play a role in childhood
asthma.
In partnership with the National Institute of
Environmental Health Sciences, ORD selected
eight academic institutions in 1998 to establish
Centers of Excellence in Children's Environmen-
tal Health and Disease Prevention. Three
institutions are focusing on pesticide risks, and
the other five are focusing on the role of
pollutants in inducing or exacerbating childhood
asthma.
Through EPA's Risk Assessment Forum, ORD is
contributing to the development of guidance for
assessing environmental risks to children.
44
-------
7. Harmful Algal Blooms
If you visit coastal areas, you may have seen a sign
like this one:
DANGER: Area Closed: Shellfish in this area
contain toxins and are not safe for use as food.
Warnings like these have become more common in
coastal areas around the United States. You may
ask, "What is poisoning coastal aquatic life?"
Often, the culprits are certain types of algae. While
most algae are not harmful and form the base of the
ocean's food web, under certain conditions some
species can proliferate or "bloom," causing what is
termed a harmful algal bloom (HAB). HABscan
include some red tides, brown tides, and other
potentially toxic organisms such as Pfiesteria.
Recently, the incidence and types of HABs have
been increasing in U.S. coastal waters, potentially
threatening humans, marine life, and
economic resources in almost every
coastal state. While the factors contrib-
uting to the proliferation of HABs are
poorly understood, they may include
nutrient enrichment or input of other
pollutants by humans; species
transport via ship ballast water; long-
term climatic shifts; natural species
dispersal; or even our increased
abilities to detect new toxins and toxic
events. In addition, blue-green algae
(i.e., cyanobacteria) blooms are thought
to be a growing problem in ground
water, potentially threatening
drinking water supplies. v
Many of the species that
form HABs produce
potent neurotoxins that " *
are known to cause serious
human illnesses (e.g., amnesic
shellfish poisoning) through ingestion of
contaminated seafood, and in some cases
through direct contact with seawater and inhalation
of aerosolized toxin. Figure 7-1 shows the different
types of human illnesses caused by HABs and
where they have occurred in coastal areas around
the United States. HAB toxins have been impli-
cated in large-scale mortalities offish, birds,
manatees and other aquatic animals. Pfiesteria has
been linked to fish kills, diseased fish, and human
health problems from Delaware to North Carolina.
Other blooms do not produce toxins, but they can
reduce the amount of light or oxygen in the water.
This shading or depletion of oxygen can damage
coral reefs and sea grass beds. Fish kills, beach
cleanup, closures of contaminated commercial
shellfish beds, toxin monitoring programs, and loss
of tourism caused by HABs have caused large
economic losses.
Focusing ORD's Current
Research and Implementing the
National HUB Strategy
In response to the growing concerns over HABs,
EPA and other federal agencies implemented a
jointly-funded interagency research program, the
O Neurotonic Shellfish Poisoning
Paralytic Shellfish Poisoning
Amnesic Shellfish Poisoning
Gigualgra Fish Poisoning
Pf eE'erca. complex
iM: KV i I :li;
proliferation
Fish, bird, mammal & submerged
aquatic vegetation kills
Figure 7-1. Cumulative map showing locations of
major HAB-related events along the coastal
United States. Events include human illnesses
(shellfish and fish poisoning), microorganism
outbreaks, and loss of fish, birds, mammals, and
vegetation. Source: NOAA COP/National HAB
Office-WHOI.
45
-------
7. Harmful Algal Blooms
Ecology and Oceanography of Harmful Algal
Blooms (ECOHAB). This program, which is funded
under the STAR Program within EPA, is designed to
increase the understanding of all aspects of the
ecology and oceanography of HABs.
The ECOHAB effort complements ORD 's intramural
research program devoted to HABs. Together these
efforts are contributing to addressing critical
research gaps and uncertainties in the causes and
impacts of HABs on ecosystems and human health.
ORD's HAD research strategy is the outcome of an
ORD-sponsored workshop held in Florida in
October 1997, in which leading experts from federal
and state agencies and academic institutes met with
EPA personnel. Goals of the strategy are to: 1)
improve EPA's capability to provide a unified
assessment of the risks of HABs by understanding
effects on ecological condition and human health;
2) predict the occurrence of HABs by better
understanding the causes of blooms and the
relationships with coastal nutrient enrichment; 3)
facilitate rapid response to HAD events by devel-
oping techniques for real-time detection of blooms
and related potential impacts; and 4) evaluate
potential management strategies to mitigate,
control, and prevent HAD occurrences and their
impacts. ORD has built a state-of-the-art laboratory
culture facility for HABs in Florida to achieve these
research goals and objectives.
Advancing the Science
To further support ORD's HAD research strategy,
ORD has conducted research to better understand
the potential adverse health effects of harmful
algae. In one recent study, ORD investigated
whether contact with estuarine waters where
Pfiesteria is killing fish could adversely affect
vision. A visual assessment performed on a group
of fishermen found preliminary evidence that
exposure to these waters adversely affected their
ability to detect visual patterns. Because the
fishermen had not been recently exposed to fish
kills, the data suggest that the effects may be
persistent. Another series of studies at Duke
University, in collaboration with scientists in ORD,
reinforced this finding when it showed learning
impairments in lab rats exposed to Pfiesteria similar
to problems experienced by people who have
experienced significant exposure to Pfiesteria.
Where Do We Go From HereP
Highlights of ORD HAB research to achieve the
objectives of its research strategy include:
Predict occurrence of HABs. FY97 STAR grants
are investigating the causes of specific types of
blooms (e.g., blue-green algae blooms, at the
University of Guam), and learning more about
specific problem algal species (e.g., Pfiesteria, at
North Carolina State University). ORD scientists
have been working on the HAB problem in
affected areas, including research and modeling,
to determine some of the environmental factors
critical in regulating population growth and toxin
production in the Gulf of Mexico and on the
Neuse River of North Carolina. This research is
contributing to our ability to predict outbreaks
and to control them when they occur.
Effects on ecological condition and human
health. ORD is cooperating with several federal
and state agencies to monitor the ecological
condition of Mid-Atlantic estuaries, including
the Chesapeake and Delaware Bays. ORD
scientists are helping several states identify
potential adverse human health effects from
exposure to toxins possibly produced by
Pfisteria-like organisms. Other studies continue
to characterize effects, including perception and
memory effects, using animal models, as well as
to verify effects on human vision.
Facilitate rapid response to HAB events. ORD
scientists are cooperating with the Florida
Department of Environmental Protection under a
STAR-funded grant to determine the factors
leading to the formation and movement of toxic
Gymnodinium breve blooms in the Gulf of
Mexico. This could lead to potential HAB
control and prevention strategies. ORD
scientists are also collaborating with NOAA and
the Naval Research Laboratory to identify and
apply unique signatures of red tide species to
monitor and track blooms in the Gulf of Mexico
using remote sensing.
Control and prevent HABs. FY98 STAR grants
included funding methods to determine the
feasibility of controlling HABs (e.g., work by
the Woods Hole Oceanographic Institution on
using clays to remove cells from infected waters)
and to improve detection of HABs (e.g., work by
the University of Maryland to develop a DNA-
based assay for detecting Pfiesteria).
46
-------
8. Pollution Prevention
The last time you opened a bag of potato chips, you
probably didn't give much thought to how the bag
was made. But preparing food packaging involves a
number of steps and chemicals that, given the
millions of packages produced each year, can add up
to appreciable environmental impacts. For example,
printing labels for foods and other consumer goods
often involves the use of chemical solvents that
contribute to ozone pollution when they evaporate
and interact in the atmosphere with sunlight. Some
solvents can be toxic when inhaled as well.
Research by Sigma Technologies International, Inc.,
with funding from ORD under the Small Business
Innovative Research (SBIR) program, has recently
resulted in technology with potential to reduce the
environmental impacts of food packaging by
eliminating inks that depend on organic solvents
such as toluene. Sigma Technologies has developed
inexpensive, high-speed technology and equipment
for treating plastic film surfaces so that water-based
inks or inks that employ no solvents at all can be
used in printing packages. The technology has now
been adopted by a major snack-food processor.
This kind of research is part of an approach to
environmental protection known as pollution
prevention. Pollution prevention has been embraced
by EPA as the preferred way to address human health
and environmental risks since passage of the
Pollution Prevention Act of 1990. Pollution
prevention differs from traditional, "end-of-the-pipe"
approaches to controlling environmental pollution
that address wastes after they have already been
created, such as sending them to waste disposal
facilities. Instead, it involves carefully analyzing
pollution sources and seeking creative ways to avoid
generating wastes in the first place. Replacing a toxic
chemical with a less toxic one during manufacture is
one example of pollution prevention. Another
example is modifying an industrial practice so that it is
more efficient, resulting in less production of waste
materials. "Green Chemistry" - the development and
introduction of new, less toxic chemicals into industry
- is a rapidly growing field of pollution prevention.
Over the past decade, ORD has supported hundreds
of pollution prevention projects and studies. Recent
areas of focus of ORD's laboratories have been green
chemistry, development of membrane technologies
for purifying waste streams, cleaner industrial process
and design, and benign solvents. Another important
component of ORD's pollution prevention efforts is
the Technology for a Sustainable Environment
program (TSE), conducted through ORD's STAR
program in concert with the National Science
Foundation. Competitive grants are awarded to
universities and other nonprofit organizations to
support fundamental and applied research related to
pollution prevention in industrial processes, method-
ologies, and technologies.
Recent Accomplishments
Use of More Environmentally Benign
Solvents
Several recent ORD projects have examined
solvents. The use of solvents goes well beyond
food packaging; more than 30 billion pounds of
organic solvents and solvents containing halogens
(chlorine and related elements) are used worldwide
each year in a myriad of industrial processes and as
cleaning agents. Funded in part by a TSE grant, a
researcher at the University of North Carolina has
developed a new technique that allows the use of
liquid carbon dioxide (CO2) to replace solvents such
as perchloroethylene in parts cleaning for industry
and electronics and in dry cleaning. A researcher at
the University of Notre Dame has used a TSE grant
to study practical applications of solvents such as
CO2 when in the form of supercritical fluids (liquids
heated above their boiling points). ORD has also
developed a software tool to enable manufacturers
to design more benign solvents or solvent mixtures
for their operations. Called PARIS II, the "Program
for Assisting in the Replacement of Industrial
Solvents," it will soon be available to businesses.
Green Chemistry
In the area of green chemistry, ORD has successfully
demonstrated a novel method for synthesizing
chemicals using ultraviolet light and a special titanium
dioxide catalyst developed by the University of
Cincinnati. This process has potential as a clean
method for manufacturing many oxychemicals, which
constitute a large class of commercially significant
chemicals currently made by less clean technologies.
Through a TSE grant, researchers at the University of
Kansas examined a class of reactions known as
47
-------
8. Pollution Prevention
"alkylation" used in formulating gasoline and other
industrial processes. These reactions traditionally use
liquid hydrofluoric acid or sulfuric acid, which pose
environmental and safety concerns because of the
dangers from spills and the need for disposal of toxic
wastes. Solid acid catalysts are potentially safer
alternatives, but they tend to deactivate rapidly due
to the formation of 'coke' deposits. The University of
Kansas researchers have demonstrated a method for
extending the life of solid acid catalysts by using
supercritical fluids to remove coke precursors. This
method may eliminate amajortechnologicalbarrierto
the use of safer catalysts in alkylation.
Membrane Technology to Reduce Pollution
Synthetic membranes are becoming increasingly
important in wastewater treatment and a variety of
other applications. ORD has worked with the
University of Kentucky to develop a microfiltration
membrane to remove toxic metals from waste
streams, which may prove useful at both contami-
nated waste sites and in manufacturing. The
approach involves incorporating low-cost materials
into the membrane to capture metals, thereby
preventing pollution from entering the environment.
ORD has also developed a membrane technology
(patent recently awarded) that will have significant
potential in recovering volatile organic compounds
for recycling from mixtures with surfactants.
Cleaner Process and Design
ORD is working to help industry reduce emissions
from the manufacturing process, which can be a
major source of pollution. For example, ORD has
validated a technology that greatly reduces
chromium emissions from hard chrome metal plating
baths by using a class of fluorinated fume
suppressants. Already commercially adopted by
several companies, EPA will recommend this
technology for complying with the Clean Air Act.
ORD has also developed a software tool known as
the Waste Reduction (WAR) algorithm, which can
be used in analyzing the pollution impacts of
various manufacturing process options to design
one with the least adverse environmental impact
(Figure 8-1). WAR will soonbe available to
businesses.
Where Do We Go From HereP
In September 1998, ORD published the Pollution
Prevention Research Strategy (www.epa.gov/ORD/
WebPubs/final/) that provides the framework for
implementing a program of systematic research and
development activities to carry pollution prevention
well into the future. ORD's pollution prevention
research program will have four main objectives: (1)
delivering broadly applicable tools and methodolo-
gies; (2) developing and transferring pollution
prevention technologies and approaches; (3)
verifying selected pollution prevention technolo-
gies; and (4) conducting research to address
economic, social, and behavioral aspects of
pollution prevention.
Light gases used for fuel
REACTANTS
FLASH - to separate
the light gases from
A the product
Product
REACTOR
DISTILLATION COLUMN
Unused
Reactant
Recycle Unused Reactant
Figure 8-1. The Waste Reduction (WAR) algorithm uses process information to evaluate the
environmental friendliness of a process design and to identify areas for pollution prevention, as shown in
this simplified process flow diagram.
48
-------
9. U.S.-Mexico Border Environmental
Health
Arizona
N*w Mnaico
PaciHc Qce-an
Guff of Vlcxico
i
Figure 9-1. The border zone where the United States and Mexico are working together to solve
environmental problems (as defined by the 1983 La Paz agreement). Source: modified from the report
United States-Mexico Environmental Indicators 1997.
Mexico and the United States share a border that
stretches more than 3,000 kilometers (nearly 2,000
miles) from the Pacific Ocean to the Gulf of Mexico.
Within the zone that extends 100 km (62 miles) on
each side of the border (Figure 9-1), the distinct
social and economic features of the two countries
converge and blend. Through trade, migration, and
shared natural resources, the destinies of communi-
ties on both sides of the border are tightly interwo-
ven. Rapid population growth over the past
hundred years is one shared feature of the border
area, with more than six million people now inhabit-
ing the U. S. side of the border and more than four
million on the Mexican side.
Rising commerce between the United States and
Mexico has contributed to rapid industrialization of
the border area, especially in the form of
"maquiladoras" (manufacturing plants) in Mexico.
More than 1,700 maquiladoras employ over 700,000
workers in the border area. While economic
opportunities have improved for residents on both
sides of the border, industrialization and population
growth have placed a great strain on infrastructure
and resources such as adequate housing, safe
drinking water, clean air, and effective wastewater
treatment. Until recently, for example, millions of
gallons of raw sewage entered the Tijuana River
each day. Such environmental stresses have caused
or are suspected of having played a role in a variety
of reported health problems such as elevated lead
exposure in children, hepatitis, pesticide poison-
ings, childhood asthma, and various infectious
gastrointestinal diseases.
With the interrelated nature of border communities
and the fact that pollution does not respect political
boundaries, the governments of Mexico and the
United States have recognized that they must work
together to solve environmental problems that
affect communities on both sides of the border. The
La Paz Agreement, signed by the Presidents of the
United States and Mexico in 1983, intensified
efforts to protect the environment along the U.S.-
Mexico border. Mutual cooperation has led to
projects such as upgrading wastewater and solid
waste treatment capabilities. A new phase of
addressing shared problems commenced in 1996
with the announcement of the Border XXI Program
to build on the work of the La Paz agreement.
ORD'sRole
ORD is involved in U. S. -Mexico border issues
through the Environmental Health Workgroup
(EHWG), which is co-chaired by ORD, the U.S.
Department of Health and Human Services (HHS),
and the Mexico Secretariat of Health (SSA). The
EHWG is one of three new workgroups created by
Border XXI to augment a set of six workgroups
49
-------
9. U.S.-Mexico Border Environmental Health
50
already established to address areas such as air and
water. It evolved from an Interagency Coordinating
Committee established in 1992 by EPA, HHS, and
agencies of the U.S. border states. The EHWG
seeks to improve the capacity of state, tribal, and
local agencies on both sides of the border to assess
and respond to health and environmental threats;
improve opportunities for stakeholders in the
border area to participate in environmental health
initiatives; and prevent health and environmental
problems through public education.
Recent Accomplishments
Under the auspices of the EHWG and its predeces-
sor, ORD has funded and participated in numerous
projects. One important accomplishment involved
studying children's exposure to lead. Excessive lead
exposure can impair nervous system development.
Children in the border region may be exposed to
lead through sources not commonly found in other
parts of the United States, such as lead-glazed
pottery used for cooking and storing food. Begin-
ning in 1997-1998, the EHWG screened children's
blood lead levels in the Arizona-Sonora, Tijuana,
and New Mexico-Chihuahua border regions. The
program introduced a new technology to the border
area, a hand-held device for determining blood lead
levels in the field. Results to date suggest that
elevated blood lead levels are not pervasive among
children living in the regions, though problems
associated with specific contaminated sites exist.
Another effort has involved identifying areas where
infants and young children may be at risk from
agricultural pesticide exposure. To identify such
areas, information on agricultural areas, pesticide
use locations, and locations where children are
present must be combined. ORD and other EHWG
members established a Geographical Information
Systems (GIS) workgroup to accomplish this task
and similar projects on environmental health issues
of concern. GIS is a computer-based approach for
layering different types of information onto a single
map so that patterns and relationships can be
examined. The workgroup has developed standard-
ized maps for both sides of the border and an
inventory of existing environmental, population,
and health datasets for the region. Using the maps,
pilot screening studies to identify potential high
priority areas for children's pesticide exposure are
underway. The effort, which also included prepar-
ing a report for environmental health practitioners
on how they can use GIS, has built the capacity for
border states to continue their own GIS work.
Another ORD-supported effort to strengthen the
capabilities of individuals and institutions in the
border region to respond to environmental health
issues is the "Advanced Training" program. This
binational program focuses on training and
education in the areas of environmental and
occupational toxicology, epidemiology, and
engineering, and risk communication. During 1997-
98, scholarships were awarded to public health
workers to obtain advanced degrees in epidemiol-
ogy and six short courses were carried out in
Mexico on a variety of environmental health topics.
ORD also conducted a study to find out if air
pollutants were moving across the border from
Mexico into the Lower Rio Grande Valley of Texas.
Overall, transport of air pollution across the border
did not appear to adversely affect air quality on the
Texas side of the Lower Rio Grande Valley. Levels of
air pollution were similar to or lower than other
urban and rural areas in Texas and elsewhere. In
addition to providing a better understanding of
current air quality, the findings may be useful to
compare to future levels of air pollution to see how
conditions of the Valley are changing.
Where Do We Go From HereP
The EHWG has also pursued a number of other
projects that are continuing beyond 1998, including:
Assessing the risks to children from exposure to
multiple pesticides from a variety of pathways,
including food, soil, and dust.
Completing the Texas Border Health Survey, a
study of the health and environmental
conditions along the Texas border.
Developing an Environmental Health Yellow
Pages to facilitate access to quality health and
environmental information in border communi-
ties.
Assessing the relationship between air quality
and respiratory health in children in the El Paso-
Ciudad Juarez area.
Developing a small research grants and training
program and an environmental health education
program for health care providers to further
strengthen institutional capacity.
The Workgroup is now implementing a new vision
that expands its original focus by strengthening
cross linkages with other Border XXI workgroups.
For example, the EHWG is conducting joint
planning exercises with the Air and Water
Workgroups. The goal is to identify and test actual
measures of health outcomes that can be used to
assess the benefit of improvements in air and water
quality, such as a decline in gastrointestinal illness
after water treatment plants are upgraded. More
information is available on their website
(www.epa.gov/orsearth/).
-------
10. Monitored Natural Attenuation
Long before humans began to recycle wastes and
turn them into useful products, nature was already
doing it. Organisms such as earthworms and
bacteria have been recycling waste organic
materials (e.g., dead leaves and other plant material)
for millions of years. They do this by breaking
them down into simpler compounds such as water,
carbon dioxide, and minerals, which are used by
other plants and animals. This process is called
biodegradation. Natural attenuation takes advan-
tage of biodegradation and other natural processes
to reduce the toxicity, mobility, volume, or concen-
tration of toxic materials and reduce the risk from
these materials to humans and the environment.
Examples of different natural attenuation processes
are showninFigure 10-1.
Ground water contamination by organic chemicals
is a major national problem, with costs to clean up
(remediate) contaminated water ranging into the
hundreds of billions of dollars. Because fewer
remediation options exist for ground water than
other media such as soil, and it is technically
difficult and extraordinarily expensive to cleanup
ground water, there is wide interest in using natural
attenuation for ground water contamination
problems. Although organisms in ground water
can degrade many kinds of chemicals, they can be
overwhelmed if the load of waste material they
receive is too great. This happens when wastes,
such as solvents, are improperly disposed or when
fuels leak out of underground storage tanks, acting
as a continuing source of contamination. Even in
these cases, biodegradation may occur at the edge
of the contaminated area due to dilution, limiting
further movement of contaminants in the environ-
ment.
If a contamination source can be controlled, natural
attenuation processes, combined with a carefully
designed continuous monitoring program, may be a
cost-effective approach for cleaning up contami-
Figure 10-1. Natural attenuation processes may include: biodegradation (breakdown of waste organic
materials by microorganisms into simpler compounds such as water and carbon dioxide); sorption
(adherence of a contaminant to a solid); volatilization (evaporation); chemical reactions; and dispersion
and dilution. This figure depicts these processes on dissolved chlorinated solvents leaking from an
underground storage tank.
-------
10. Monitored Natural Attenuation
nated ground water. This approach is called
monitored natural attenuation (MNA). Monitoring
the natural attenuation processes at a contaminated
site is key to successfully using this approach,
because it determines whether natural attenuation
is actually occurring and, if so, at what rate. Natural
attenuation is monitored by collecting water
samples from wells strategically placed in and
around the contaminated area.
ORD's Research Program
ORD's extensive research program dates to the mid-
1980s and is considered a leader in researching
MNA of organic chemicals. During the 1990s, ORD
scientists and other researchers found that natural
attenuation of organic chemicals in ground water
was much more extensive than previously believed.
As a result, many site owners and others affected
by contaminated ground water became interested in
natural attenuation and began submitting proposals
for meeting site cleanup standards by relying on
natural attenuation to regulatory agencies. How-
ever, technical and regulatory guidance on the
consistent evaluation and use of MNA did not yet
exist, so potential cost savings from using MNA
were not being realized.
Recent Accomplishments
ORD helped provide critically needed guidance in
three ways. First, ORD worked with EPA's Office of
Solid Waste and Emergency Response (OSWER) to
develop an interim directive in 1998 to clarify EPA's
policy on the use of natural attenuation in the clean-
up of contaminated sites administered by EPA.
Second, ORD developed aprotocol in 1998 for MNA
of fuels and chlorinated solvents in ground water to
support the policy directive. The protocol focused
on these contaminants because they were recog-
nized as national problems in ground water and
could be addressed by natural attenuation. The
protocol provides guidance on collecting and
evaluating data at a site to determine the extent to
which natural attenuation is occurring, and whether
it might be used as part or all of the site clean-up.
Finally, ORD has been providing technical assistance
and training to EPA Regional Offices and other
federal organizations to implement MNA.
Key research findings on MNA that formed the
basis for the protocol include:
In general, sites should be screened for biodeg-
radation potential before investing in a detailed
data collection effort to evaluate natural
attenuation
Site subsurface conditions must be carefully
evaluated
For some plumes (the region containing the
ground water contamination), MNA may be an
option in only part of the plume
Ground water contaminant transport and fate
modeling can be important for evaluating MNA
at a site but requires extensive data gathering for
the model to be meaningful
MNA is not likely to be appropriate if the plume
is expanding
Key to any MNA evaluation is an analysis of
whether and to what extent humans and
ecosystems are exposed to site contaminants
The protocol was a major step in implementing
MNA at sites with ground water contaminated by
chlorinated solvents and fuels in a way that is
protective of human health and the environment.
By following the protocol, tens to hundreds of
millions of dollars in site cleanup costs can
potentially be saved at Superfund, RCRA, and
other waste sites. ORD collaborated with research-
ers from the U.S. Air Force, U.S. Geological Survey,
and National Research Council Resident Research
Associates in conducting this research.
Where Do We Go From HereP
MNA is potentially applicable to other ground
water contamination problems, but more research is
needed before it can be used in a manner that is
protective of the environment. Current ORD
research on MNA is focused on the fuel additive
methyl tertiary-butyl ether (MTBE) and metals,
particularly chromium and arsenic. ORD chose
these chemicals because they are common ground
water contaminants for which conventional
remediation technologies are either not available or
extremely expensive. MTBE is used in automotive
fuels in most regions of the country, and appears to
resist biodegradation. ORD is conducting exten-
sive field and laboratory research to evaluate the
natural attenuation of MTBE and to provide
guidance to EPA's Office of Underground Storage
Tanks.
Unlike organic compounds, metals cannot be
broken down into other components. However,
natural processes may immobilize metals in the
subsurface, preventing further exposure to humans
and ecosystems. ORD is conducting research to
determine whether these processes may, under
appropriate circumstances, be used to clean-up
ground water contaminated with metals such as
chromium or arsenic.
52
-------
11. Global Change
Over the past two centuries, the Earth's atmosphere
has changed appreciably. Largely due to combus-
tion of fossil fuels, elimination of forests, and other
human activities, compounds known as "green-
house gases" have been accumulating in the
atmosphere. Greenhouse gases include carbon
dioxide (CO2), methane, and chlorofluorocarbons
(CFCs). CO2 levels have increased by 30% over the
past 200 years, and are thought to be higher now
than at any time in more than 100,000 years.
Greenhouse gases trap heat radiating from the
Earth's surface, raising concerns that the planet
may be warming.
The best available evidence indicates that world-
wide temperatures have in fact risen. 1998 was the
warmest year since widespread temperature records
began in the late nineteenth century, according to
the National Oceanic and Atmospheric Administra-
tion (NOAA), and seven of the ten warmest years
on record have occurred in the 1990s (Figure 11-1).
Rising sea levels and retreating glaciers also
provide evidence that global temperatures are
increasing.
While scientific consensus has emerged that
human activities are affecting global climate, the
causes and consequences of climate changes are
far from completely understood. Climate is linked to
a myriad of interconnected factors, including solar
radiation, atmospheric gases and particles, cloud
cover, global ocean currents, geological features,
and polar ice sheets. Moreover, living organisms
both respond to climate changes (e.g., by changing
growth rates or moving to different areas) and
modify them (e.g., by emitting gases and influenc-
ing water cycles). Human societies are no excep-
tion, reacting to and affecting climate in compli-
cated ways.
Because of the vast scope and complexity of global
changes, research must be integrated across
scientific disciplines and national borders for
scientific knowledge to advance effectively. Within
the United States, the Global Change Research Act
of 1990 led to the creation of a national research
framework under the auspices of the U.S. Global
Change Research Program (USGCRP), of which EPA
is a member. Research findings from the United
States and other countries are used by the Inter-
governmental Panel on Climate Change (IPCC) to
develop global assessments.
Annual Global Surface Mean Temperature Anomalies
0.75
'.330
2000
-D.9D
Figure 11-1. Combined global land and ocean temperature anomalies (deviations from the average)
1880-1998. "Zero degrees" represents the overall average during that time period. Seven of the ten
warmest years have occurred in the 1990s. Source: National Climate Data Center/NESDIS/NOAA.
53
-------
11. Global Change
A New Direction for EPA Global
Change Research
In 1998, ORD began a major redirection of its global
change research program in response to changes in
emphasis of the USGCRP and recommendations of
external peer reviewers. Previously, ORD research
largely pursued goals of understanding environmen-
tal processes (such as the carbon cycle) and
developing technologies to reduce emissions of
gases that contribute to global warming. Under the
new program, ORD emphasizes assessing the human
health, ecosystem, and socioeconomic conse-
quences of global change. ORD's global change
research encompasses not only climate change and
climate variability, but also the effects of increasing
ultraviolet radiation (due to stratospheric ozone
depletion) and land use changes.
More specifically, ORD is examining the potential
effects of global change on: (1) human health,
including heat-related mortality and illness and the
spread of infectious diseases, (2) air quality, (3) water
quality and quantity, and (4) ecosystem health,
including effects on biodiversity and important
ecosystem services. These categories, which are
interrelated, are being assessed in the context of
other stressors not necessarily related to global
change. A focal point of ORD's research efforts is
leading public-private partnerships to complete
assessments that will be part of the USGCRP's first
National Assessment Report in 2000. ORD is
directing the Great Lakes, Gulf Coast, and Mid-
Atlantic Regional Assessments (three of the 19
regional assessments) and the Health Sector
Assessment (one of six sectoral assessments). Other
ongoing projects include the operation of a network
of ultraviolet radiation monitors inNational Parks in
cooperation with the National Park Service and
contribution to a report of the IPCC on land use
change and forestry.
Recent Accomplishments
Because ORD recently redirected its global change
program, most accomplishments will occur after the
time frame of this report. Several recent achievements
are worth noting, however. ORD and the U.S.
Geological Survey jointly developed the North
American Landscape Characterization database,
which uses NASA Landsat satellite imagery to
depict land cover and land use in the lower 48 states
and Mexico from the 1970s through the 1990s.
Researchers can analyze these data to see how land
cover has changed over this time period and assess
restoration opportunities. Additionally, ORD
sponsored several public workshops during 1997
and 1998 for the health and regional assessments
that will be part of the 2000 national assessment.
ORD's STAR grant to the Johns Hopkins Univer-
sity School of Public Health also yielded several
noteworthy accomplishments. For example,
researchers demonstrated that satellite maps of
vegetative cover can be used to predict (and
prevent) outbreaks of hantavirus more than nine
months in advance. Hantavirus is a potentially
deadly virus transmitted to people by rodents in the
southwestern United States. Outbreaks are most
likely when rodent populations flourish, which
occurs during years with above-normal rainfall and
resulting increased growth of vegetation. Changes
in patterns of disease like this are examples of the
possible consequences of changing climate. Johns
Hopkins researchers also described connections
between changes in climate and Lyme Disease in
the U.S. mid-Atlantic; mosquito-borne dengue fever
in the Brownsville, Texas area; and cholera in Peru.
ORD also completed research projects under its
prior, more process-oriented global program. One
noteworthy experiment, conducted over four years,
examined the effects of elevated CO2 and tempera-
ture on Douglas fir trees. The Douglas fir is an
important species of Pacific Northwest forests
valued for its timber. ORD researchers working in
Corvallis, Oregon grew tree seedlings in a set of
highly sophisticated growth chambers that con-
tained the components of a forested ecosystem in
miniature. The scientists were able to independently
manipulate temperature and CO2 concentration to
examine the effects on soil characteristics and
seedling growth. Both elevated temperature and CO2
stimulated the release of CO2 from the soil, suggest-
ing that global warming might, over the long-term,
reduce soil fertility and water holding capacity.
Elevated temperatures (but not CO2 alone) also
altered bud growth and reduced shoot growth,
resulting in deformed, shorter trees. These findings
will be useful in predicting the impacts of global
warming on Douglas fir trees.
54
-------
12. Arsenic in Drinking Water
In 1996 Bangladesh suffered a national crisis when
it was discovered that millions of its citizens were
drinking water contaminated with arsenic. The
source: wells constructed in the 1970s and 1980s to
replace surface water supplies contaminated with
disease-causing microorganisms. This and other
events, including other large-scale arsenic poison-
ings from contaminated drinking water throughout
the world (e.g., Taiwan, China, India, Mexico, and
Chile) have heightened the need for addressing this
health issue. Contamination of drinking water in the
United States has been reported in several states,
although the levels are generally much lower than
those observed in Bangladesh. In some locations,
however, concentrations in individual wells have been
found to be extremely high.
Arsenic occurs naturally in the earth's crust and is a
natural water contaminant in some areas. Human
activities such as mining may also contribute to
elevated levels in water. In addition to water, arsenic
also occurs in foods. Arsenic can take different forms
or species, which differ considerably in their ability to
cause adverse health effects. Potential human health
effects of ingested arsenic include skin and internal
cancers (bladder, lung, liver, kidney, and prostate),
cardiovascular disease, cerebrovascular disease,
diabetes, and reproductive and developmental toxicity.
EPA has classified arsenic as a human carcinogen
through both ingestion and inhalation routes of
exposure.
The 1986 Amendments to the Safe Drinking Water
Act (SDWA) established a maximum contaminant
level (MCL) of 50 ug/liter for arsenic. An MCL is the
maximum allowed level of a contaminant in water
delivered to any user of a public system serving 10
or more people. This arsenic MCL was adopted from
a U.S. Public Health Service drinking water standard
set in 1942, before modern cancer and other health
related data on arsenic became available. As
required by the SDWA Amendments of 1996, EPA is
now reevaluating the MCL and will propose a new
standard in 2000. Given this regulatory mandate, the
potential health risks, and the high costs of treatment
to remove arsenic from drinking water, research on
the health effects of arsenic and cost-effective
treatment technologies is a high priority for EPA.
ORD's Research Program
ORD 's arsenic research program is contributing to
the scientific and technical basis for the new
arsenic standard. The 1998 Research Plan for
Arsenic in Drinking Water describes a prioritized
research agenda that has served as a guide to
research conducted inside and outside of EPA
(www.epa.gov/ORDAVebPubs/final/) High priority
areas include measurement of different forms of
arsenic in water, foods, and biological materials;
research on internal cancers; development of tools
and models to predict uptake of arsenic in humans;
and evaluation of drinking water treatment tech-
nologies. ORD coordinates with organizations
such as the American Water Works Association
Research Foundation and the Association of
California Water Agencies on arsenic research.
Recent Accomplishments
Understanding Human Exposure to Arsenic
The predominant inorganic arsenic species found in
drinking waters are the trivalent (As+3) and the
pentavalent (As+5) forms, or arsenite and arsen-
ate, respectively. Because As+5 is easier to remove
from drinking water, it is necessary to determine
how much of each species is present before
devising a treatment strategy. ORD researchers
have successfully developed a sensitive method
to measure both of these forms at very low
concentrations (low parts-per-billion range).
To estimate the amount of arsenic in tissues in
humans chronically exposed to arsenic, its metabo-
lism and its elimination from the body must be
understood. ORD research has shown that arsenic
can be metabolized in the human gut, and that
various factors, such as the amount of selenium in
the diet, influence how arsenic is metabolized and
eliminated. These insights are being used to
develop a model of arsenic behavior in humans.
Insight Into How Arsenic Causes Cancer
Several ORD studies have focused on how arsenic
causes cancer and other toxic effects. For
example, ORD scientists discovered that when
cells were exposed to arsenic, changes occurred to
55
-------
12. Arsenic in Drinking Water
a specific gene, p53, that may reduce production
of a protein that suppresses the growth of tumors.
This in turn may predispose arsenic-exposed cells
to transform into a cancerous state. Other ORD
studies have shown that methyl arsenic, a form of
arsenic that can be created within the body by
methylation of inorganic arsenic, is an extremely
potent enzyme inhibitor. These studies are
significant because they differ from the current
scientific view that methylation is simply a
detoxification step.
Feasibility of Population Studies in the
United States
In addition to research at the cellular level, ORD
scientists have studied the health effects of
arsenic in human populations in the United States
and elsewhere in the world. ORD conducted a
study in several Utah communities that histori-
cally have had long-term exposure to arsenic in
drinking water. The contamination of groundwater
in Utah is believed to have resulted from deposi-
tion of arsenic-contaminated ash and dust from
formerly active volcanoes (Figure 12-1). This
study showed that it is feasible to conduct an
epidemic logic study of waterborne exposure to
arsenic in the United States in which effects seen
in studies of non-U.S. populations (e.g., skin and
bladder cancer) can be evaluated.
Figure 12-1. Formerly active volcano in Utah, a source of
arsenic contamination. The inset provides an interior
view of the volcano.
Removing Arsenic From Drinking Water
ORD scientists evaluated two innovative treatment
processes to remove arsenic from drinking water:
ion exchange with brine (salt solution) recycle and
iron coagulation with microfiltration. The ion
exchange process uses a large amount of salt
solution that ends up as waste. ORD-sponsored
research at the University of Houston demon-
strated that the waste solution can be reused over
20 times before it must be disposed, which saves
costs and increases the efficiency of the process.
The second treatment method involves the use of
iron coagulant to absorb arsenic, which is then
filtered from the water. Both processes were very
cost effective for removing arsenic and practical for
small systems.
Where Do We Go From HereP
In the area of health effects, the work in Utah has
been expanded to a full-scale study that includes an
evaluation of the use of urinary arsenic concentra-
tions as a biomarker for exposure. ORD is also
working in Chile with researchers from the University
of Kentucky to evaluate the relationship between
exposure to arsenic and potential effects in infants
(mortality, birthweight, and prematurity) and in
mothers (preeclampsia and gestational diabetes).
Scientists from ORD are collaborating with the
Chinese Inner Mongolia Institute of Public Health
to study the health effects of arsenic in drinking
water and to compare inter-cultural differences that
may impact its toxicity. EPA and other agencies
sponsored the development of an International
Tissue and Tumor Repository on Chronic
Arsenosis at the Armed Forces Institute of
Pathology in Washington, D.C. This repository will
be useful in the study of chronic toxicity of arsenic,
especially at the cellular and biochemical level.
In the area of exposure assessment, ORD
scientists are developing analytical methods to
more accurately measure arsenic in human
biological samples and foods. These methods
will provide information for assessing exposure
and risk.
In the area of risk management, ORD scientists
are evaluating the effectiveness of seven
different oxidants (e.g., chlorine and ozone) in
converting As+3 to the more readily removed
As+5 as a pretreatment step for drinking water.
ORD is also evaluating the performance of nine
existing drinking water treatment plants in
removing arsenic using conventional treatment
methods, such as coagulation/filtration and ion
exchange.
56
-------
13. Economic and Decision-Making
Research
On a clear day, the treeless summit of Mt. Washing-
ton in New Hampshire boasts a spectacular view of
the surrounding White Mountains (Figure 13-1).
Featuring many of the highest peaks in the North-
east, and forests ablaze with color during the fall
foliage season, the scenic vistas of the White
Mountains attract thousands of visitors each year.
However, air pollution from countless sources
some nearby and some hundreds of miles away
can obscure these views and lessen the enjoyment
of visitors. What is the economic impact of air
pollution in this region? How much are these scenic
vistas worth to people?
These kinds of questions are being examined
through an ORD grant to the University of New
Hampshire, University of Massachusetts, and the
Appalachian Mountain Club. This grant is one of
25 STAR research grants underway or awarded
during 1997-1998 to study methods for valuing
environmental benefits. Unlike commodities like
shoes or candy bars, which have well-established
markets, placing a dollar value on environmental
commodities such as clean air and water supplies is
often controversial and difficult. Nevertheless,
regulatory agencies like EPA need some way of
gauging benefits in making decisions like whether
further pollution controls are worth the costs
placed on industry and consumers. In response to
this need, ORD is sponsoring research on a variety
of techniques for estimating the value of environ-
mental benefits.
Research into valuing environmental benefits is one
important component of ORD's economic and
decision-making grants program, which is con-
ducted in partnership with the National Science
Foundation. Research in this field is an important
complement to EPA's other research in disciplines
such as toxicology and ecology. The findings from
economic and decision-making research can help
make EPA's programs to safeguard public health
and the environment more effective and less costly
to society.
Figure 13-1. View of the White Mountains from Mt.
Washington, New Hampshire.
Recent Accomplishments in
Valuing Environmental Benefits
One of the principal methods for estimating the
value of environmental benefits involves surveying
people on how much they are willing to pay, or
what they are willing to give up, to obtain an
environmental improvement. To be reliable esti-
mates of real values, survey instruments must be
carefully designed and tested. Research under
STAR grants has led to more reliable methods for
designing and conducting valuation surveys. For
example, research by Cornell University demon-
strated that low-cost phone surveys could yield
results at least as valid as more expensive face-to-
face interviews. Duke University researchers found
that surveyed respondents tend to be more willing
to accept a public good (such as a new park) than
cash as compensation for environmental harm.
Researchers at the University of California at
Berkeley demonstrated that the accuracy of the
context of a question is more important than the
57
-------
13. Economic and Decision-Making Research
kind of question used in surveys, based on
comparisons to actual payments for environmental
amenities. University of Georgia investigators
evaluated methods for estimating the value of
environmental benefits that included focus group
discussions, interviews, and experimental payments
combined with educational materials, and, like the
Cornell researchers, found that context and
education were of paramount importance in
obtaining accurate values.
Other Economics Research
Accomplishments
Some observers have criticized conventional
measures of economic performance, such as a
nation's gross domestic product (GDP), for
counting practices that cause environmental harm
as positive economic activity while failing to
account for the environmental depletion that
ultimately may lower standards of living. In
response to these concerns, ORD has supported
research through the STAR grants program on how
to incorporate the value of natural resources and
environmental quality into these aggregate
economic measures to give a more accurate
portrayal of the nation's quality of life. For example,
the Colorado School of Mines developed more
accurate estimates of the value of depletable energy
and mineral resources to use in measures like the
GDR Other STAR grants have examined areas such
as corporate decision making and policy alterna-
tives. For instance, Resources for the Future
examined different approaches for reducing
pollution emissions. The research demonstrated
methods for improving the efficiency of market-
based approaches to pollution control, such as
those that involve the buying and selling of
emissions quotas.
Where Do We Go From HereP
A cornerstone of EPA's efforts to reinvent environ-
mental policy is to find ways to accomplish
environmental goals at lower costs, and experience
has shown that using economic incentives and
other market mechanisms is one way of doing so.
Traditional, "command and control" regulations
often require all pollution sources to meet the same
emissions standards. This can be inefficient
because some companies may find it extremely
costly to meet the new standards, while others who
are already in compliance have little incentive to
improve. An alternative approach is to set overall
emissions goals, and then let industries decide
through the market how to meet them, using
economic instruments such as tradable permits,
fees, and taxes. ORD recently issued a grant
solicitation for research into environmental
management using these kinds of market ap-
proaches to implement more cost-effective ap-
proaches for controlling pollution.
Additionally, a recent EPA initiative focuses on the
effects of environmental factors on children's
health (see chapter on Children's Health for more
information). Children may be more susceptible to
pollutants and disease than adults, and they also
differ from adults from an economic standpoint in
that they would not be expected to pay to avoid
health problems. As a result, estimating the
economic costs and benefits of measures to protect
children's health differs from making these esti-
mates for adults. ORD is working with EPA's Office
of Children's Health Protection to develop grants
that will encourage research into the economic
valuation of children's health.
Finally, ORD is initiating a joint solicitation with the
Department of Justice (DOJ) to investigate determi-
nants of corporate environmental performance and
how various government interventions may impact
this performance. Results of this research should
help EPA, DOJ and the states better understand
what kinds of incentives and deterrents are most
effective in helping regulated entities achieve
compliance with environmental laws.
58
-------
14. Ecological Indicators
When you go to the doctor for a check-up, often
one of the first things he or she will do is check
your blood pressure. Your blood pressure reading
is just one "indicator" of your health. By evaluating
a combination of relevant indicators, a doctor can
make a general assessment of your overall health.
In much the same way, ecological indicators help
ecologists assess the health or integrity of ecosys-
tems. The extent of forests in a region or the types
offish able to live in a stream are two examples of
ecological indicators. A natural ecosystem is so
complex and dynamic that it is not feasible to
monitor the status of all of the organisms, and all
their interactions with their environment, that occur
within the ecosystem. Ecological indicators are a
way of sifting through this complexity, so that
important signals about the condition of an
ecosystem can be discerned. Consequently,
ecological indicators are vital to scientists and
resource managers in detecting changes in our
environment and in pointing to the possible causes
of these changes.
In some cases it is obvious what should comprise
an ecological indicator, but often it takes a great
deal of investigation and field testing to produce
valid, reliable indicators. What might be the best
indicators, for instance, of the condition of a stream
the number offish? The types offish? The types
of stream insects? The vegetation growing along
the stream banks? A combination of factors? Fully
documented indicators exist for only a few types of
North American ecosystems. For the past two
decades, ORD has been performing research to
develop needed ecological indicators. Currently,
ORD is emphasizing the development of two types
of indicators. The first is landscape indicators on
local, regional, and national scales. The second is
ecological indicators for aquatic systems, compris-
ing streams, rivers, lakes, estuaries, and wetlands,
including special or threatened ecosystems such as
coral reefs. As a result of the efforts of ORD and
other organizations, we are now better able to
assess the state of our environment and meet the
information needs of policymakers designing
solutions to environmental problems.
Recent Accomplishments
Guidance Manual for Indicator Development.
Recognizing the need for an overall framework for
how to develop, test, and document indicators for a
given ecosystem, an ORD-wide working group
developed the Evaluation Guidelines for Ecologi-
cal Indicators. The guidelines describe a four
phase progression for developing an indicator: 1)
developing a sound conceptual model; 2) evaluat-
ing the feasibility of gathering the data that
comprise the indicator; 3) understanding the
variability in the components of the indicator (such
as measurement error and variability across time
and space); and 4) ensuring that indicator results
are clearly understood and useful to scientists and
policy makers. Indicator development experts
outside of EPA have cited the guidelines as a
crucial step in assessing ecological condition. The
guidelines are currently being used by ORD
scientists and a technical manual is in press.
Landscape Indicators. Working in a number of
locations around the country, ORD scientists have
developed indicators of ecological condition from a
broad, landscape scale. These landscape indicators
range from conceptually simple (such as the
percentage of forested land in an area) to more
complex indicators made of several components
(such as the percentage of crops on steep slopes
with highly erodible soils), as shown in Figure 14-1.
ORD has collaborated with other federal agencies
to develop these indicators using satellite imagery
as well as data collected on the ground. An example
of the application of these indicators is ORD's 1998
report, An Ecological Assessment of the United
States Mid-Atlantic Region: A Landscape Atlas,
which is described more fully in the chapter on the
Mid-Atlantic Integrated Assessment. Some of the
more than 30 indicators used in the report to assess
watershed conditions include population density,
road density, proportion of watershed with suitable
interior forest habitat, and patterns of vegetation
change from 1975 to 1990.
Under a STAR grant, a team of investigators led by
Iowa State University has also developed ecologi-
cal indicators that take into account landscape
characteristics. These indicators were based on
data gathered from satellite imagery on the spatial
-------
14. Ecological Indicators
pattern and condition of mountain meadows in the
Yellowstone National Park and Grand Teton
National Park regions of Wyoming and Montana.
The indicators can be used to predict the diversity
of plants, birds, and butterflies in these meadows.
Because meadow communities are highly sensitive
to variations in precipitation and temperature, the
indicators are a promising tool for quickly detecting
ecological changes, such as those that may be
brought on by climate change.
Estuary Indicators. Estuaries are vital resources
that are home to many kinds of plants and animals,
including commercially valuable fish and shellfish.
Bottom-dwelling animals, orbenthic organisms,
help maintain water clarity in estuaries by filtering
algae and sediment from the water and are a critical
component of the estuarine food web. ORD applied
and demonstrated an approach - known as the
estuarine benthic index - for using benthic
organisms as indicators of estuarine condition in a
number of regions including the Mid-Atlantic, the
Gulf of Mexico, and California. Just as the Dow
Jones average is used to track the "condition" of
the stock market, the index combines individual
pieces of information into a single measure of the
condition of benthic communities. ORD's 1998
report Condition of the Mid-Atlantic Estuaries
(described in more detail in the chapter on the Mid-
Atlantic Integrated Assessment) applied the index
as one indicator of the ecological status of the
Chesapeake Bay and other estuaries of the Mid-
Atlantic.
Coral Reef Indicators. Under a STAR grant,
researchers at the University of Guam developed
indicators to assess impacts on coral reefs from
sewage, sediment runoff, and pesticides. The
indicators were based on measuring reproductive
failure and disturbances of larval settling behavior
in coral. Using these indicators, the scientists were
able to recommend ways to prevent or reduce
human impacts on reefs. Guam will implement one
critical recommendation to stop particular sources
of pollution such as sewage discharge and con-
struction activity during the extremely brief period
of time when coral reef fertilization occurs.
Where Do We Go From HereP
Future ORD ecological indicator research will
continue to focus on developing landscape
indicators and indicators for aquatic resources.
Projects include:
Landscape indicator development for the San
Pedro Watershed of Arizona, the Tensas River
Basin of Louisiana, and for the EMAP Western
Pilot (see the Mid-Atlantic Integrated Assess-
ment chapter for more information about the
Western Pilot).
Nationwide development and implementation of
ecological indicators by EPA and other federal
agencies to establish the current condition of all
U.S. estuaries, provide a listing of stressors
associated with impaired conditions, and inform
environmental decisions to protect these critical
resources.
Continued work by academic researchers under
more than 40 STAR grants for ecological
indicators awarded.
Over time, as indicators are developed for indi-
vidual ecological resources such as streams and
estuaries, ORD will assess the extent to which the
indicators can be used to document and assess
ecological stressors that impact multiple resources.
One such stressor is global climate change, which
could involve changes in weather patterns and
resulting shifts in distributions of organisms.
Ultimately, the goal of ORD's research is to develop
indicators that can be used to assess not only
individual resources but also larger, interlinked
ecological systems on continental and global
scales.
Figure 14-1. An example of an ecological indicator. Overlaying a land cover map (agriculture in red) with
topography produces a new map showing agriculture on steep slopes. Agriculture on steep slopes
indicates a risk to streams from soil erosion. Source: An Ecological Assessment of the United States Mid-
Atlantic Region: A Landscape Atlas.
60
-------
For more information, please access the
Office of Research and Development website at:
www.epa.gov/ORD
This document is available on ORD's website at:
www.epa.gov/ORD/publications
Paper copies of this and many other EPA publications may be
ordered from the National Service Center for Environmental
Publications (NSCEP) by telephone at 1-800490-9198
or online through:
www.epa.gov/epahome/publications.htm
-------
Q)
I P S 3 I
£ 5 ai co
o i.,,1 ST
03 8 5L
O 5 Q) -n
I2 I
(3) o =2:
O 4^ O
CQ
0
I
m
------- |