UNIT-ED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C 20460
May 27, 1987
SAB-EC~87™Q31
O F F I C t Of
r H fe, AQMINfST^AFQH
Honorable Lee M. Thomas
Administrator
U. S, Environmental Protection Agency
401 M Street, S. W.
Washington, D. C. 20460
tear Mr. Thomas:
The Science Advisory Board's Integrated Environmental Management
Subcommittee has completed its review of EPA's Draft Kanawha Valley Toxics
Screening Study and is pleased to transmit its tinal report to you. The
Subcommittee met in public session on March 16, 1987 in Philadelphia,
Pa., to review the study. During March 11-13, 1987, three representatives
of the Subcommittee visited the Kanawha Valley to become more faroilar
with its environmental problems.
The Subcommittee unanimously concludes that the Kanawha Valley study!"
represents an important component of EPA's overall effort to develop
methodologies to define public health and environmental priorities.
Studies such as this provide valuable technical challenges and experiences
to EPA staff, particularly those working in regional offices. And,
finally, they provide a valuable means for developing closer working
relationships with state and local officials and the general public.
This letter is the Subcormittee's second cortrajn icat ion to you. On
July 30, 1986 it "many concerns about the ability of the current
study to satisfy a number of technical issues. A chief concern is "the
incongruity between [the study's] .... objectives and the fact that the
study design itself is not an integrated multimedia effort, nor a response
to Bhopal."
Since the transunittal of that letter, EPA staff have modified the
study's objectives and technical design, and have conducted supplementary
analyses to support the revised objectives and design. In general, the
Subcommittee believes that the staff have made appropriate responses to its
major concerns. The study reaches a number of scientifically supportable
conclusions about health risks from cancer in the Kanawha Valley. The
study also points EPA and other interested parties in a direction for
conducting further analyses of problems related to accidental releases of
pollutants and acute health effects.
Specific issues addressed during the Subcommittee's review include:
the study's objectives and scope; pollution sources} pollution transport
and fate by media; health effects,- risk communication; and recommendations
for additional follow-up efforts. Attachment A presents additional, more-
detailed recommendations for modifying the current study and future
activities in the Kanawha Valley. Attachment B lists the Subcommittee
members.
-------
_ 2 -
In generaj, the Subcommittee views the Draft Kanawha Valley Toxics
Screening Study as one step of a continuing process to risks. The
current study addresses chronic health exposures to carcinogens which
represent one of many public health concerns in the Valley. As a follow-up
to the current study* the Subcommittee recommends two additional steps
that include:
o Expanded monitoring of air toxics, and use of monitored values
to obtain more precise estimates of exposure* and health risks.
o Greater focus on accidental releases and fugitive emissions as
areas of public health concern.
The Subcommittee appreciates the opportunity to conduct an independent
scientific review of these important public health issues in the Kanawha
Valley. We request that EPA formally respond to our scientific advice.
Sincerely,
Ronald Wyzga, Chairman
Integrated Environmental
Management Subcommittee
Science Advisory
Norton Nelson, Chairman
Executive Conmittee
Science Advisory Board
-------
SAB-EC-87-031
MAJOR FINDINGS AND RECOMMENDATIONS
OF 'THE
INTEGRATED ENVIRONMENTAL MANGEMENT SUBCOMMITTEE
ON THE
U. S. ENVIRONMENTAL PROTECTION AGENCY'S DRAFT
RANAWHA VALLEY TOXICS SCREENING STUDY
INTEGRATED
SCIENCE ADVISORY
U. S. ENVIRONMENTAL PROTECTION AGENCY
May, 1987
-------
u. s.
NOTICE
This report has been written as a part of the activities of
the Science Advisory Board, a public advisory group providing
extramural scientific information and advice to the Administrator
and other officials of the Environmental Protection Agency. The
Board is structured to provide a balanced expert assessment of
scientific matters related to problems facing the Agency, This
report has not been reviewed for approval by the Agency, and
hence the contents of this report do not necessarily represent
the views and policies of the Environmental Protection Agency,
nor of other agencies in the Executive Branch of the Federal
government, nor does mention of trade or ccfirosrcial products
constitute endorsement of recommendation for use.
-------
The objectives of the Kanawha Valley study are limited, but reasonably
well-defined. In most instances, the study seeks to derive an upper bound
for the health risks associated with airborne carcinogens for which
EPA's Cancer Assessment Group has derived potency estimates. Other
potential carcinogens are minimally considered, and the health risks of
non-carcinogens, including those risks associated with the accidential
release ot chemicals such as occurred at Bhopal, are not considered.
Hence, the health assessment of airborne toxics is far from complete, but
this is clearly articulated in the study report. Available resources did
not allow a more comprehensive assessment.
The study attempts "to develop a of potential public health
concerns" associated with carcinogens in drinking water, surface water
and hazardous wastes. The efforts are not multimedia efforts, but medium-
specific efforts based upon ver^_J,ijni,t.ed data; thus, conclusions from
efforts are subject to considerable uncertainty.
Sources
The air analyses depend very heavily upon an emissions inventory of?"
some 450 substances developed by the Wast Virginia Air Pollution Control •
Ccmrdssion (APCC). The inventory is as extensive and comprehensive as
any other available information. Nevertheless, there exist
uncertainties in the inventory, particularly with respect to fugitive
emissions, which the study identifies as a major source of health risk in
some Kanawha Valley communities. The possibility that the inventory is
incomplete is also suggested by the fact that ethylene oxide not included
for either the Belle or Nitro connunities despite limited monitoring
evidence that it may be present. If a compound was not in the- inventory
it was not included in subsequent EPA modeling. This discrepancy underlines
the need for including ethylene oxide in future monitoring programs.
The drinking water and surface water analyses depend upon monitored
levels of toxics in water supply systems and fish fillets, respectively,
Data limited to a subset of all public water suppliers, with no private
well samples, and to a very small number of fish sampled from only one
location for a very limited number of toxic substances. The hazardous
waste inventory is based upon a priority pollutant screening of inventories
for a subset of RCRA and potential CERCLA sites. No information
available on the total quantity and overall composition of toxic wastes
that may be entering surface or ground water. For this reason alone, the
results of this part of the study are, at best, suggestive.
The transport models used in the studies generally appear to be
congruent with the study objectives. The air transport modeling addresses
the concerns of the Subcommittee in its July 30, 1986 letter, although better
-------
- 2 -
docunentation of this modeling is needed. There is a factor of two
uncertainty on point source air emissions and another uncertainty of a
factor of two in dispersion modeling. The current modeling efforts do
not address these potential uncertainties, although "worst case" scenarios
should recognize their existence. Drinking water exposure was estimated
by assuming that individuals consume two liters of the water delivered to
their neighborhoods. Similar assumptions are often made in risk assessments.
The surface water and hazardous waste studies are greatly hanpered by a
lack of data, making large assumptions necessary to estimate exposure to
toxics.
Health_Effects
The study evaluated 20 known or suspected cancer causing chanicals
from the West Virginia APCC inventory of more than 450 compounds. The
Subcommittee concludes that the current study provides useful information
on health effects from cancer and environmental loadings of these 20
compounds. After finalizing the current study, EPA should conduct additional
efforts that include:
o Using the APCC inventory and information on toxicity to evaluate^
the potential health effects of of the remaining ccnpounds.v
Of the remaining 430 or so compounds, relatively few merit further
attention,, but EPA and APCC should work together to identify
compounds that need additional evaluation. These should be
identified by defining the set of those conpounds to which some
exposure may be likely at known toxic levels.
o Broadening the health endpoints of concern to include non-cancer
and acute effects. Concern about the potential effects from
acute releases is strong within the ccfimunity; hence, some
priority should be given to addressing this issue. The
methodologies used to these endpoints require further
development, particularly in estimating the effects of accidential
releases. fault-tree or alternative analysis should be
designed to this possibility. Experts fron other groups
within the EPA should be enlisted in this effort.
o Incorporating frequency plots of pollutant concentrations versus
time? in addition to stating average pollutant concentrations.
o Assessing the conversion of reference doses from the ingestion to
the inhalation pathway, where reference dose information for the
inhalation pathway is not available,
o Evaluating whether to develop or use biological markers for health
assessment.
o Comparing risks from high emissions of pollutants with low
toxicity, with low emissions ot pollutants with high toxicity
as a means to identify priority risk management needs.
-------
_ 3 —
o Exploring other potentially useful sources of data for compounds
of concern? including monographs prepared by the International
Agency for Research on Cancer, Health Effects Profiles developed
by the Office of Research and Development, Reportable Quantities
for hazardous compounds and gaining access to information through
the ccnmmity right-to-know provision of Superfund.
The Subcommittee encourages EPA to continue its- efforts of working
with officials and citizens of the Kanawha Valley to update them on the
sources and magnitude of risks they experience. In particular, EPA should
seek to further improve its presentation of technical Information to
better enable lay persons to understand the results of technical analyses
and to ensure it is understood that the risk numbers reflect upper bound
estimates. Clarification of the latter issue is also in the
executive summary of the study,
It is important for citizens, scientists and public officials to
understand that the principal value of the Kanawha Valley study Is as a
screening study of airborne carcinogens. As the study acknowledges,
a screening study should strive to ensure that all potential risks are
Identified even at the expense of calling attention to risks that subsequent
analysis may not confirm, or will be less than indicated In the screening
study. Accordingly, assumptions in screening studies are conservative in
nature; assumptions should be avoided that might cause potential risks to
be ignored. Within the stated scope of the study, conservative assunptions
are made; for example, individuals are assumed to be exposed continuously
to ambient outdoor levels of industrially emitted toxics and upper bound
risk estimates are given. There are a few instances, however, where the
study did not rigorously pursue conservative assumptions. These include
potential uncertainties or omissions In the emissions inventory. The
study suggests that point estimates could be too small (or too large) by
a factor of two. For fugitive emissions it could be greater. It is
important that these uncertainties and their likely direction be clearly
articulated in the report along with a discussion about whether additional
scenarios are necessary to consider these uncertainties.
In addition, the air quality models are equally likely to under-and-
over predict ambient concentrations. The biases of the models are fairly
predictable. Exposures are likely underestimated at the peaks of ridges
where the river turns and when overlapping models were not used. On the
other hand, the use of the Box model probably overpredicts exposure in
some neighborhoods on the Valley floor, which are not adjacent to emissions
sources. Although it is to the study's credit to have implemented two
different modeling approaches to estimate exposure, further discussion
in the report is merited on the potential model biases and on their
implications for the risk estimates.
-------
Attachment A
ADDITIONAL RECOMMENDATIONS TO THE AGENCY
A.
1. The technical assumptions for the underlying transport models
should be documented and made accessible to readers of the
report.
2. Given the comprehensive nature of the airborne toxic risk
assessment in contrast to the rudimentary nature of the other
three studies, it may be desirable to more cle'arly separate the
air toxic studies from the others; moreover, the various studies
are undertaken for differently defined geographic areas.
3. The risk estimate bounds are probably more clearly defined than
in roost similar documents; nevertheless , further clarification
raay be necessary. Cases could be presented as
-------
- 2 -
3. The hazardous waste data considered are very limited. CERCLA
requirements can perhaps provide useful information. Other
parts of EPA should be enlisted to improve the source inventory
for these data. Analysis of historical operations and land use
may also be useful to characterize the types of chemicals in
waste sites. The fundamental approach to consider risk frcm
hazardous waste should be replaced, however, by one that examines
specific waste sites,
4. Increased monitoring data can aid the analysis of drinking water,
surface water, and ground water. For chemicals of concern in the
Valley? such efforts should be instituted to help ensure that no
major problems are overlooked.
5. Health surveys and measurement of biological markers could provide
validation of the estimated health profile of the Valley.
Such efforts will not, however, be useful when incremental risk
estimates are small.
-------
Attachment B
U. S. ENVIRONMENTAL PROTECTION AGENCY
SCIENCE ADVISORY
' Integrated Envirormental Management Subcommittee
Dr. Ronald Wyzga, Chairman
Electric Power Research Institute
3412 Hillview Avenue
Post Office Box 1041
Palo Alto, California 94303
Dr. Stephen L, Brown
Environ Corporation
1000 Potomac Street, N.
Washington, D. C. 20007
w.
Dr. Thomas Clarkson
University of Rochester Medical Center
Box RBB
575 Elrrwood Avenue
Rochester, York 14642
Dr. Herbert H, Cornish
830 West Clark Road
Ypsilanti, Michigan 48198
Dr. Robert Frank
Johns Hopkins University
Department of Environmental
Health Sciences - JHSHPH
615 North Wolfe Street
Baltimore, Maryland 21205
Dr. Rolf Hartung
School of Public Health
University of Michigan
Ann Arbor, Michigan 48109
Terry F. Yosie, Director
Science Advisory Board
U." S. Environmental Protection
Agency
401 M Street, S. W.
Washington, D. C. 20460
Dr. Thomas Burke
New Jersey Department of Health
Div. of Occupational & Environ.
Health CN 360
Trenton, New Jersey 08625
Dr. Yoram Cohen
Chemical Engineering Department
U. C, L, A. 5531,
Boelter Hall
Los Angeles, California 90024
Dr. Terry Davies
World Wildlife Fund
The Conservation Foundation
1255 23rd Street, N* W.
Washington, D. C. 20037
Dr. Janes Gruhl
7610 N. Christie Drive
Tucson, Arizona 85718
Dr. Paul Lioy
University of Medicine and
Dentistry of New Jersey
675 Hoes Lane
Robert Wood Johnson
Medical School
Piscataway, New Jersey 08854
-------
- 2 -
Dr. Francis
The of
Environmental Engineering
Department of Civil Engineering
Carnegie Mellon University
5000 Forbes Avenue
Hall 123A
Pittsburgh, Pennsylvania 15213
Dr. Ellen Silbergeld
Chief Toxics Scientist
Environmental Defense Fund
1525 18th Street, N. W.
Washington, D. C. 20036
Dr. North
Principal, Decision Focus, Inc.
Altos Office Center
Suite 200
El
Los Altos, California 94022
------- |