Risk Assessment Pilot Study. Phase 3
Naval Construction Battalion Center, Davisville, Rhode Island
Science Applications International Corp., Narragansett, RI
Prepared for:
Environmental Research Lab., Narragansett, RI
Dec 93
Y I
U.S. DEPARTMENT OF COMMERCE
National Technical Information Service
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO. 2.
EPA/600/R-94/046
3. RECIPI EN"
4. title AND subtitle RISK ASSESSMENT PILOT STUDY-PHASE III
NAVAL CONSTRUCTION BATTALION CENTER DAVISVILLE, RI
DRAFT FINAL REPORT_FEBRUARY 1994
5. REPORT DATE
December 1993
6. performing organization code
7. authoris) y pj.Munns, Jr., C. Mueller, B. Rogers, S. Benyi,
S.Cayula, T. Gleason, W.G. Nelson, R.K. Johnson
8. performing organization report no.
ERLN-1428
9. performing organization name and address
US EPA Environmental Research Laboratory
27 Tarzwell Drive
Narragansett, RI 02882
10. program element no.
11. contract/grant no.
68-C1-0005
12. sponsoring agency name and address
13. TYPE OF REPORT AND PERIOD COVEREO
Report
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
Naval Command, Control and Ocean Surveillance Center San Diego, CA 92152
16. ABSTRACT
ฆTo undertake a marine ecological risk assessment at the Naval Construction Battalion
Center (NCBC) Davisville, Rhode Island to determine the effect of hazardous waste disposal
on Allen Harbor and Narragansett Bay. Allen Harbor, located in Nanagansett Bay at NCBC
Davisville, was closed to shellfishing by the Rhode Island Department of Environmental
Management because of suspected hazardous waste contamination from a landfill and disposal
area adjacent to the haibor. NCBC Davisville was added to the National Priority List in
November 1989. Between 1946 and 1972, the 15-acne landfill received a wide variety of
wastes, including sewage sludge, solvents, paints, chromic acid, PCB-contamLiated waste
oils, preservatives, blasting grit, and other municipal and industrial wastes generated at
NCBC Davisville and at the Naval Air Station Quonset Point Another site, also adjacent to.
Allen Harbor on Calf Pasture Point, was used for disposal of calcium hypochlorite
decontaminating solution and chlorides. -
17. KEY WORDS AND DOCUMENT ANALYSIS
3. DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Risk Assessment
Naval Battalion Center Davisville, RI
marine ecology
hazardous waste disposal
landfill
Allen Harbor
shellfish closinq
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21. NO. OF PAGES
134
20. SECURITY CLASS (TIlis page)
UNCLASSIFIED"
22. PRICE
EPA Form 2220-1 (Rซv. 4-77) previous edition is obsolete
-------
EPA/600/R-94/046
December 1993
DRAFT FINAL REPORT
Risk Assessment Pilot Study - Phase in
Naval Construction Battalion Center
Davisville, Rhode Island
Prepared by:
Wayne R. Munns, Jr., Cornelia Mueller,
Betty Anne Rogers, Sandra Benyi, Stephanie Cayula, Timothy Gleason
c/o U.S. Environmental Protection Agency
Science Applications International Corporation
27 Tarzwell Drive
Narragansett, Rhode Island 02882
William G. Nelson
U.S. Environmental Protection Agency
Environmental Research Laboratory
27 Tarzwell Drive
Narragansett, RI 02882
and
Robert K. Johnston
Naval Command, Control and Oceans Surveillance Center
Code 522
San Diego, California 92152
Submitted
December 1993
ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
NARRAGANSETT, RHODE ISLAND 02882
ERLN CONTRIBUTION NO. 1428
-------
Disclaimer
This report has been reviewed by the Environmental Research Laboratory, U.S.
Environmental Protection Agency, Narragansett, Rhode Island, and approved for publication.
Approval does not signify that the contents necessarily reflect the views and policies of the
U.S. Environmental Protection Agency, nor does mention of trade-names or commercial
products constitute endorsement or recommendation for use.
ii
-------
Acknowledgements
The information and analyses presented in this document are the result of the efforts
of numerous individuals. We especially thank Warren Boothman, David Hansen, Kay Ho,
Darryl Keith, George Morrison, Richard Pruell, and Henry Walker of EPA-ERLN, Walter
Berry, Maxine Boncovage, Robert Burgess, Dan Carder, Donald Cobb, Pamela Comeleo,
Randy Comeleo, Jeffery Corbin, Ruth Gutjahr-Gobell, Saroja Jayaraman, Glen Modica,
Timothy Montmarquet, Diane Nacci, Sara Nelson, Claudia Olivieri, Deborah Robson, and
Mark Tagliabue of SAIC, and David Bengston of URI for providing field sampling, chemical
and biological testing and analysis, and data interpretation support throughout this project.
Nancy Lockwood and Diane Sheehan of Computer Sciences Corporation administered and
maintained computerized data bases. Special thanks to Sara Nelson for preparation of
graphics, and to Darryl Keith, George Morrison, Richard Pruell, Henry Walker for
providing comments on this report.
We also thank Commander Robert Buchholz, Leo Tomasettii, and Louis Fayan of
NCBC Davisville; Marilyn Powers, Todd Boder, Simeon Hahn, Franco LeGreca, and A!
Haring of Northern Division, Naval Facilities Engineering.Command; and die other members
of the NCBC Technical Review Committee for their continuing input, interest, and support.
Managerial and administrative support for the effort was provided by Jeff Grovhoug, Head,
Marine Environmental Quality Branch, and Peter F. Seligman, Head, Environmental
Sciences Division, NCCOSC RDTE DIV. This work was sponsored by the Assistant
Commander for Environment, Safety, and Health, Naval Facilities Engineering Command,
Alexandria, VA, with Joe Kaminski, Director, Environmental Programs Division, and
Theodore Zabrobelny, Director, Environmental Restoration Division providing programmatic
support. This work was conducted as part of the U.S. Navy's Installation Restoration
Program. Cooperative research and monitoring activities were jointly funded through an
interagency memorandum of agreement between NCCOSC and ERLN by Naval Facilities
Engineering Command and U.S. EPA Office of Research and Development, Washington,
ill
-------
D.C. W. Munns, C. Mueller, B. Rogers, S. Benyi, S. Cayula, and T. Gleason were
supported under U.S. EPA Contract No. 68-03-3529, Richard Latimer, Project Officer, and
No. 68-C1-0005, Patricia Gant, Barbara Brown, and Brian Melzian, Project Officers, to
Science Applications International Corporation. This is Contribution No. 1428 of the
Environmental Research Laboratory - Narragansett.
iv
-------
Contents
Disclaimer ii
Acknowledgements i i i
Figures vii
Tables ix
Abbreviations and Acronyms x
Executive Summary . 1
Introduction 3
Background 3
Phase I Summary 6
Phase II Summary 11
Phase III Approach 16
Methods 21
Quality Assurance/Quality Control 21
Exposure-Response Models 22
Exposure Media Collection and Preparation 23
Laboratory Assays 24
Modei Development 27
Mya Laboratory Exposures 28
Chemical Analyses 29
Results 31
Exposure-Response Models 31
Laboratory Assays 31
Model Development 46
Mya Laboratory Experiment , 54
Quantification of Ecological Risk 57
Risks of Toxicological Impact 57
Risk Quantified by Toxic Unit Exposure-Response Models ... 57
Risk Quantified as Joint Probabilities 69
Risk of Neoplastic Disease Development in Mya 78
References 79
Appendices
A. Trace Metal Concentrations A-l
B. Pesticide Concentrations B-l
C. Polychlorinated Biphenyl Congener Concentrations C-l
D. Total Polychlorinated Biphenyl Concentrations D-l
E. Polycyclic Aromatic Hydrocarbon Concentrations . E-l
v
-------
F. Acid Volatile Sulfide Concentrations in Sediments F-l
G. Total Organic Carbon Concentrations in Sediments G-l
H. Seep Water Bioassay Results H-l
I. Sediment Bioassay Results 1-1
J. Sediment Extract Bioassay Results J-l
K. Mya Neoplasia Experiment Results K-l
L. Grain Size Analyses of Sediments H-l
vi
-------
Figures
Number Page
1 Allen Harbor and the locations of the landfill and Calf Pasture Point 4
2 Fertilization response of Arbacia to seep water exposure 32
3 Larval mortality and development responses of Arbacia to seep water
exposure 33
4 Acute mortality response of Mysidopsis to seep water exposure 34
5 Sexual reproduction response of Champia to seep water exposure 35
6 Embryo/larval toxicity response of Mulinia to seep water exposure 37
7 Larval mortality response of Memdia to seep water exposure 38
8 Seven-day mortality response of Mulinia to sediment exposure 40
9 Seven-day growth response of Mulinia to sediment exposure 41
10 Fertilization response of Arbacia to sediment extract exposure 42
11 Larval mortality and development responses of Arbacia to sediment
extract exposure . 43
12 Embryo/larval toxicity response of Mulinia to sediment extract exposure ... 44
13 Bioluminescence response of Photobacterium to sediment extract
exposure . 45
14 Whole media exposure-response model for fertilization response of Arbacia
to seep water exposure 47
15 Whole media exposure-response model for larval development response
of Arbacia to seep water exposure 48
16 Whole media exposure-response model for sexual reproduction response
of Champia to seep water exposure 49
vi i
-------
17
18
19
20
21
22
23
24
25
26
27
28
29
50
51
52
53
60
61
62
63
64
65
66
68
71
Whole media exposure-response model for larval survivorship response
of Mulinia to seep water exposure ,
Whole media exposure-response model for fertilization response of
Arbacia to sediment extract exposure
Whole media exposure-response model for larval development response
of Arbacia to sediment extract exposure
Whole media exposure-response model for bioluminescence response
of Photobacterium to sediment extract exposure
ETU exposure-response model for fertilization response of Arbacia
to seep water exposure
ETU exposure-response model for larval development response
of Arbacia to seep water exposure
ETU exposure-response model for sexual reproduction response
of Champia to seep water exposure
ETU exposure-response model for larval survivorship response
of Mulinia to seep water exposure
ETU exposure-response model for fertilization response of
Arbacia to sediment extract exposure
ETU exposure-response model for larval development response
of Arbacia to sediment extract exposure
ETU exposure-response model for bioluminescence response
of Photobacterium to sediment extract exposure
Comparison of predicted and observed Arbacia fertilization success
Characterization of ecological risk as a joint probability
vm
-------
Tables
Number Page
1 Species and endpoints used in development of exposure-response models ... 19
2 Characteristics of sediments used in bioassays 23
3 Whole-media exposure-response model parameter estimates and coefficients
of determination 54
4 Mean responses of Mya endpoints in the 90-d neoplasia experiment 55
5 Range of bioaccumulation factors of sediment contaminants in the 90-d
neoplasia experiment 55
6 Federal marine chronic Water Quality Criteria used in ETU
exposure-response model development 58
7 Contaminant-specific TUs for undiluted exposure media used in
model development 59
8 ITU exposure-response model parameter estimates and coefficients
of determination 67
9 Mean probabilities of maximum risk to pelagic systems 73
10 Mean probabilities of maximum risk to benthic systems 75
11 Partition coefficients used in pore water contaminant concentration
calculations 76
12 Mean probabilities of risk to benthic systems incorporating
contaminant bioavailability . 77
ix
-------
Abbreviations and Acronyms
AA atomic absorption
ANOVA analysis of variance
BAF bioaccumulation factor
BHC hexachlorocyclohexane
CERCLA Comprehensive Environmental Response Compensation and Liability Act of
1980
CLIS Central Long Island Sound
DANC Decontaminating Agent Non-Corrosive (l,3-dichloro-5,5-dimethylhydantoin)
DDD dichlorodiphenyldichloroethane
DDE dichlorodiphenyldichloroethene
DDT dichlorodiphenyltrichloroethene
DMSO - dimethylsulfoxide
ECD electron capture detection
EPA United States Environmental Protection Agency
ERA environmental risk assessment
ERLN Environmental Research Laboratory, Narragansett, RI
GC gas chromatograph
HCB hexachlorobenzene
H-DANC 5,5-dimethylhydantoin
HGA heated graphite atomization
Hn -- hematopoietic neoplasia
ICP inductively coupled plasma spectrometer
MBT ~ monobutyltin
MOA Memorandum of Agreement
MS - mass spectrometer
NAS ~ Naval Air Station
NCCOSC Naval Command, Control and Ocean Surveillance Center, San Diego, CA
NCBC Naval Construction Battalion Center
ND not detected
NM not measured
NPL National Priorities List
x
-------
NSW
Narragansett Bay seawater
PAH
polycyclic aromatic hydrocarbon
PCA
principal component analysis
PCB
polychlorinated biphenyl
PPb
parts per billion
ppm
parts per million
pptr
parts per trillion
RAPS
Allen Harbor Risk Assessment Pilot Study
RCRA
~ Resource Conservation and Recovery Act of 1976
RIDEM
Rhode Island Department of Environmental Management
RI/FS
Remedial Investigation/Feasibility Study
RQ
risk quotient
QA/QC
quality assurance/quality control
SAIC
~ Science Applications International Corporation
SARA
Superfund Amendment and Reauthorization Act of 1986
SFG
Scope for Growth
SOP
Standard Operating Procedure
ctu
sum of toxic units
TBT
tributyltin
TOC
total organic carbon
TU
toxic unit
voc
volatile organic compound
XI
-------
Risk Assessment Pilot Study - Phase III
Naval Construction Battalion Center
Davisville, Rhode Island
DRAFT FINAL REPORT
February 1994
ERLN Contribution Number 1428
RISK AS A JOINT PROBABILITY
DISTRIBUTION OF EXPOSURES EXPOSURE-RESPONSE MODEL
LOG(CONC)
LOG(CONC)
LOG(CONCENTRATION)
xii
-------
Executive Summary
OBJECTIVE
To undertake a marine ecological risk assessment at the Naval Construction Battalion
Center (NCBC) Davisville, Rhode Island to determine the effect of hazardous waste disposal
on Allen Harbor and Narragansett Bay. Allen Harbor, located in Narragansett Bay at NCBC
Davisville, was closed to shellfishing by the Rhode Island Department of Environmental
Management because of suspected hazardous waste contamination from a landfill and disposal
area adjacent to the harbor. NCBC Davisville was added to the National Priority List in
November 1989. Between 1946 and 1972, the 15-acre landfill received a wide variety of
wastes, including sewage sludge, solvents, paints, chromic acid, PCB-contaminated waste
oils, preservatives, blasting grit, and other municipal and industrial wastes generated at
NCBC Davisville aid at the Naval Air Station Quonset Point. Another site, also adjacent to
Allen Harbor on Calf Pasture Point, was used for disposal of calcium hypochlorite
decontaminating solution and chlorides.
APPROACH
A phased approach was developed to assess the ecological risks to Allen Harbor and
Narragansett Bay posed by these hazardous waste sites. This report covers Phase in, a
quantification of biological effects and ecological risks directly associated with the NCBC
landfill. Exposure-response assays were conducted of landfill seep water, sediments, and
sediment extracts using a variety of marine species and endpoints. Resulting data were used
to develop models describing biological response as a function of exposure concentration.
These models were used to quantify risks posed by the landfill to pelagic and benthic
ecological systems in Allen Harbor using a joint probability method. A laboratory evaluation
of the relationship between landfill contaminants and neoplasia development in the soft-shell
clam also was conducted.
1
-------
RESULTS
This Phase III study provided quantitative information useful in describing the
ecological risks to Allen Harbor. Exposure-response models were developed successfully for
landfill seep water and sediment extracts using data obtained for a number of species and
short-term toxicological endpoints. Using a joint probability method, upper-bound
probabilities of risk ranging between 0,24 and 0.69 were estimated for landfill seep water,
with similar values calculated for storm runoff sources. Whole landfill sediments were not
toxic to organisms tested in the laboratory, but sediment extract models suggested risks up to
0.75 to benthic organisms with contaminant bioavailability taken into account. No statistical
relationships were observed between landfill exposure media and soft-shell clam neoplasia,
although the experiment was not conclusive because conditions may have compromised
treatment effects.
2
-------
Introduction
BACKGROUND
In 1988, the U.S. Environmental Protection Agency (U.S. EPA) Environmental
Research Laboratory at Narragansett, Rhode Island (ERLN), and the U.S. Navy Naval
Command, Control and Ocean Surveillance Center (NCCOSC; formerly the Naval Ocean
Systems Center), entered into a Memorandum of Agreement (MOA) to develop cooperative
research and monitoring activities for conducting ecological risk assessments (ERAs). Under
this agreement case studies were developed to characterize the risk of Navy hazardous waste
disposal sites which could potentially impact aquatic ecosystems. This joint research
supports the Navy's response to the requirements of the Comprehensive Environmental
Response Compensation and Liability Act of 1980 (CERCLA), as amended by the Superfund
Amendment and Reauthorization Act of 1986 (SARA), and the Resource Conservation and
Recovery Act of 1976 (RCRA). Additionally, the agreement afforded the opportunity for
ERLN to develop and refine methodologies for examining ecological risks associated with
anthropogenic wastes in the marine environment through their application in specific case
studies.
The first case study developed under the MOA was the Allen Harbor Risk Assessment
Pilot Study (RAPS) conducted at the Naval Construction Battalion Center (NCBC)
Davisville, Rhode Island. Allen Harbor is a small embayment of Narragansett Bay located
adjacent to NCBC Davisville, a facility added to the National Priorities List (NPL) in 1989.
Two sites at NCBC Davisville were of particular concern with respect to potential negative
impacts on Allen Harbor: a 15-acre landfill situated next to Allen Harbor, and Calf Pasture
Point, which separates Allen Harbor from the West Passage of Narragansett Bay (Figure 1).
The primary objective of the RAPS was to determine the presence and extent of
adverse ecological impacts in Allen Harbor and Narragansett Bay potentially related to
NCBC Davisville. A phased approach was developed for this study. These phases, modified
somewhat from those reported in earlier documents (Munns et ah 1991) are:
3
-------
Figure 1. Allen Harbor and the locations of the landfill and Calf Pasture Point.
4
-------
Phase I - Information Gathering To determine the existence, nature, and extent of
adverse impact in Allen Harbor and Narragansett Bay resulting from
contaminants originating from NCBC Davisville. The specific activities
involved in this step included identification and collation of existing data and
information relevant to the ecology of Allen Harbor, characterization of
sediments and the water column in the harbor and nearby areas of Narragansett
Bay, evaluation of the natural resources of Allen Harbor relative to nearby
areas of Narragansett Bay, and development of a preliminary ERA of Allen
Harbor.
Phase II - Verification and Quantification of Toxicological Effects -- To verify the lack
of adverse environmental impact (Option I), or to determine the nature and
extent of contaminant impact on the marine system (Option II). Option I was
indicated, so studies were to be conducted to confirm the lack of negative
impact. Information obtained from this phase was used to further evaluate
marine risks at NCBC Davisville. If Option II had been indicated,
characterizations of contaminant source and movement were to occur.
Further, criteria were to be identified for evaluation of remedial alternatives,
and development of a monitoring plan capable of evaluating remedial activities
would be developed.
Phase in - Quantification of Ecological Risks ~ To quantify ecological risks to Allen
Harbor associated with waste sites of concern. TTie primary activities of this
phase included conduct of laboratory assays and experiments to characterize
toxicological impacts to biota, representative of those living in Allen Harbor,
in the form of exposure-response models. Together with the information
collected in Phases I and II, these models formed the basis of the Final Marine
Ecological Risk Assessment for NCBC Davisville. This information could be
used to develop a monitoring program for continuous verification of
environmental safety.
Detailed descriptions of the activities and findings of Phases I, completed in 1990, and Phase
n, completed in 1991, are given in Munns et al. (1991, 1993). Summaries of Phases I and
n also are presented below. The approach taken to address the objectives of Phase HI, and
the results of those activities, are the primary subject of this report.
5
-------
PHASE I SUMMARY
Phase I involved the collection and collation of empirical environmental data
characterizing contaminant input to Allen Harbor (Waste Characterization), the resulting
exposure field relative to contaminant levels in Narragansett Bay (Exposure Assessment), and
the status and responses of biota residing in the harbor (Ecosystem Characterization and
Effects Assessment). This information was synthesized into a preliminary characterization of
ecological risk to Allen Harbor (Munns et ah 1991). The specific findings of these
characterization and assessment activities are described briefly below.
Waste Site Characterization
Information regarding contaminants associated with the NCBC Davisville landfill and
Calf Pasture Point was obtained from past reports and from chemical analyses of seep and
ground water, and of sediments. Historic information indicated a range of waste materials to
have been disposed in the landfill, including complex organic and inorganic wastes such as
jet fiiel, waste oils, and coal ash, as well as organic solvents, asbestos, and sewage sludge.
Chemical analysis of seep, test pit, and well water samples indicated high levels of several
chlorinated solvents, including cis- and trans-\,2-dichIoroethene, 1,2-dichloroethane,
chlorobenzene, and benzene. High levels (up to 1.49 ppb) of total polychlorinated biphenyls
(PCBs) were measured in seep samples from the south face of the landfill, and trace metal
concentrations were high enough to violate the U.S. EPA Water Quality Criteria (WQC) for
Cu, Cd, and Pb. Petroleum hydrocarbons were also present (up to 100-200 ppb) in some
samples. Pesticides were typically not detected in the ground water.
Exposure Assessment
Contaminant exposure conditions in Allen Harbor were assessed through chemical
analyses of sediments, large volume water samples, and of indigenous and deployed biota
obtained at a total of 29 intertidal and subtidal stations in Allen Harbor and Narragansett
Bay. Allen Harbor displayed some of the highest chemical concentrations observed among
stations. Highest concentrations within the harbor generally occurred at a station at the
6
-------
southern end of the harbor, removed from the landfill. Little evidence of contaminant
migration from the landfill or from Calf Pasture Point was observed.
Tissue residue levels in benthie organisms provided an additional measure of sediment
exposure conditions. Significant differences were often observed among stations in quahog
(Mercenaria mercenaria) tissue residues, and Allen Harbor clams often grouped with those
exhibiting the highest mean concentrations. In concordance with sediment chemistry levels,
tissue residues of quahogs within Allen Harbor were typically highest at the southern end of
the harbor, away from the landfill and Calf Pasture Point. Tissue concentrations in Mya
aremria were elevated relative to other stations for the pesticide 7-hexachlorocyclohexane
(BHC) and for butyltin species. Tributyltin (TBT) residues were extremely high (8,800 ppb
dry weight) in Allen Harbor Mya. Again, contaminant residues were generally higher at the
south end of the landfill than at other stations within Allen Harbor.
Characterization of water column exposure conditions through direct water sampling
and analysis was limited to organic compounds. Concentrations of pesticides in both
dissolved and particulate phases were generally below detection both within and outside of
the harbor. PCBs were observed In the particulate phase at concentrations in the 1-2 parts
per trillion (pptr) range, with somewhat higher levels in Allen Harbor than in Narragansett
Bay. Generally, however, water-bourne contaminant levels were similar to background
levels observed in relatively clean areas. Tissue residues of selected contaminants in
deployed Mytilus edulis, used as a surrogate for water chemistry, were somewhat elevated
relative to mussels from reference areas in Narragansett Bay, but were fairly typical of clean
areas elsewhere in the northeast United States.
Effects Assessment
The ecological impacts of contaminants within Allen Harbor were evaluated through a
combination of field sampling, field experimentation, and laboratory assays. These activities
involved evaluation of a number of biological endpoints which have been shown to be
sensitive to contaminant insult, and whose relationship with ecological status are fairly well
established. Native Mercenaria mercenaria, Mya arenaria, and Crassostrea virginka were
7
-------
sampled for population abundance, individual condition, and histopathological effects. The
blue mussel, Mytilus edulis, was deployed at several stations to address the effects of water
quality on physiological condition and growth. Finally, the toxicity of sediments within
Allen Harbor and at stations in Narragansett Bay was determined in the laboratory using both
standard amphipod (Ampelisca abdita) bioassays and biomarker tests under development at
ERLN, These later tests utilized field exposed organisms or laboratory exposed cell cultures
to investigate the modes and mechanisms of contaminant impact on cellular and subcellular
biological processes.
Despite the closure of Allen Harbor to shellfishing in 1984, Mercenaria in Allen
Harbor were significantly smaller than those found at Narragansett Bay stations. Reduced
shell size may reflect some impact of sediment or water quality, or may simply be the result
of a lack of fishing pressure (thereby increasing intraspecific competition) due to the harbor
shellfish closure. Condition Index followed a pattern among stations similar to that of shell
length. Proximity to the landfill had no discernible effect on Mercenaria length or condition.
No significant pathologies were observed in Allen Harbor animals. Densities of Mya were
higher in Allen Harbor than at Narragansett Bay stations, likely reflecting the lack of
recreational clamming in the harbor. No clear pattern in Mya shell length emerged in
station-wise comparisons, although Allen Harbor animals were typically larger than those
collected outside of the harbor. A number of pathological conditions were observed in Mya
from all stations throughout this study. These included pathologies commonly associated
with soft shell clams, such as atypical cell hyperplasia in the gills and kidney, and
inflammatory responses. Neoplastic lesions associated with the heart and hematopoietic
system (Hn) were found in clams collected in Allen Harbor and at nearby Marsh Point.
Within Allen Harbor proper, the highest prevalence of Hn was found near Calf Pasture
Point. Crassostrea in Allen Harbor were both larger and in better condition than those at a
reference station in Narragansett Bay. Differences in shellfishing pressure may explain these
differences. Histological examination of oysters revealed no pathology. All organisms were
in good to excellent health.
Allen Harbor subtidal sediments exhibited uniformly low toxicity, as measured by the
8
-------
acute mortality response of the benthic amphipod Ampelisca abdita, indicating little impact
from the landfill or Calf Pasture Point. In contrast, extreme mortality was associated with
material collected from the north and middle faces of the landfill. Although suggestive of
landfill-associated contaminant effects, material from these sites was composed primarily of
very course-grained material, thereby confounding toxicological analyses through a grain size
effect. Extracts of Allen Harbor sediment produced significant mutagenic effects under
certain conditions as determined by the biomarker Sister Chromatid Exchange assay. No
significant response was observed in the V79/Metabolic Cooperation biomarker assay for the
presence of tumor promoters. Sediment extracts also affected fertilization, growth (length)
and survival of the sea urchin Arbacia, although equivalent responses were observed with
extracts of sediments obtained from a reference site.
Myrilus edulis deployed in Allen Harbor in May-June 1989 showed both lower
clearance and higher respiration rates than did mussels deployed at the Narragansett Bay
stations- When integrated into the Scope for Growth (SFG) index, these rates indicated
significantly reduced physiological condition for Allen Harbor mussels. Chemical analysis of
the soft tissues of these animals were equivocal with respect to the causes of the observed
differences in physiological response. Mussels exposed during a fall deployment in
September-October of that year also exhibited differences in clearance and respiration rates
with respect to station. These differences translate into SFG estimates which were depressed
in Allen Harbor and immediately outside the harbor, relative to that in lower Narragansett
Bay. The consistently low clearance rate and SFG integration observed in Allen Harbor
mussels indicates a harbor water quality problem. No differences were observed in the in
vivo immunological response of Myrilus deployed in Allen Harbor and in lower Narragansett
Bay, nor was pathology observed in animals deployed at any site.
Preliminary Characterization of Ecological Risk
Information collected during the Waste Site Characterization, and Exposure and
Effects Assessments were synthesized into a preliminary characterization of ecological risk to
Allen Harbor. Two approaches were used to assess risk. The first involved calculation of
9
-------
risk quotients as the ratio of contaminant-specific exposure concentrations to benchmark
effect concentrations for single contaminants. In this process, field measurements of
sediment and water column contaminant concentrations were compared with published
measures of sediment and water quality. The second approach compared the results of all
biological and chemical assessments conducted for Allen Harbor with those obtained for
stations in Narragansett Bay pปoper. The intent behind this latter approach was to evaluate
conditions in Allen Harbor within the context of the larger bay system as a whole. Such an
evaluation might identify potential influences of the land-based hazardous waste sites on the
ecology of Allen Harbor.
Risk quotients (RQs) calculated for Allen Harbor sediments ranged in magnitude from
much less than 0.1, to as high as 47 for the maximum level observed of the pesticide DDT.
Classes of contaminants were identified as falling into three levels of concern: those with
quotients less than 0.1 (no risk presumed), those with RQs greater than 0.1 but less than 1
(moderate risk presumed), and those with RQs greater than 1 (risk presumed). Although the
actual quotient values for specific contaminants varied with the particular ecological
benchmark used, the major risk to benthic systems derived primarily from pesticides, PCBs,
selected metals, and polycyclic aromatic hydrocarbons (PAHs). There was, however, no
clear association of this risk with the land-based hazardous waste sites located at NCBC
Davisville.
Based upon the small number of RQs calculated for Allen Harbor surface waters, the
ecological risks associated with water-borne contaminants appeared to be minimal. This
contrasted with the Mytilus SFG results observed during Effects Assessment activities. It
may be that contaminants for which toxicological benchmarks do not exist, or which were
not quantified in this study, played some role in reducing harbor water quality.
A more subjective, but equally useful approach to assessing ecological risks associated
with the landfill and Calf Pasture Point was to compare the results of all assessment activities
in Allen Harbor with those obtained for the bay stations. At a gross level, differences
observed in such a comparison might reasonably be attributed to the unique association of
Allen Harbor with the hazardous waste sites. Confounding this assessment were the other
unique attributes of the harbor, such as its enclosed nature, and the high level of boating
10
-------
activity present therein. At this level of analysis, there was a fairly strong indication that
both sediment and water quality were impacted in Allen Harbor relative to the bay proper.
However, other sites within the bay also appeared to be impacted to some degree. The
causes of these suggested risks are not at all clear, as none of the sites exhibited untoward
contamination.
Results obtained during Phase I suggested no major environmental problems unique to
Allen Harbor, but did call into question some aspects of the quality of water column and
sediment conditions. Most notably, mussels deployed in the harbor consistently exhibited
reduced physiological condition relative to those exposed at other stations in Narragansett
Bay. Impacts were observed on sea urchin early life stage processes and in biomarkers
assays, and an increased incidence of hematopoietic neoplasia in Mya was associated with
proximity to Allen Harbor. Appreciation of the meaning of the observed responses within an
ecological context is confounded by the general lack of impact observed at higher levels of
biological organization: in situ populations of benthic organisms seemed reasonably healthy
with respect to those in other areas of the Bay.
PHASE H SUMMARY
The findings of Phase I were equivocal with respect to the degree of impact in Allen
Harbor, and to the extent of ecological risks posed by the land-based waste sites associated
with NCBC Davisville. Because some impact was observed, a modified version of Option
n, as described above, was followed in developing the objectives and activities of Phase n
(see Munns et al. 1993).
Although the results of Waste Characterization activities indicated the landfill to be a
potential source of toxicologically important contaminants to the harbor, there was no clear
association of observed impacts with proximity to the landfill. Of particular interest were the
observations that contaminant exposures and biological effects were often most severe at the
southern end of the harbor, farthest removed from the landfill and Calf Pasture Point. Other
potential sources of contamination of the harbor were known to exist. For instance, the area
immediately south of Allen Harbor currently is used as a staging area for automobiles off-
11
-------
loaded from transport ships. Large areas of land are paved over with asphalt, serving as
holding lots in preparation for over-land distribution. The community of Mount View, RI is
located to the immediate north of the harbor, and contains a golf course in addition to
residential housing. Surface water runoff from these areas was viewed as a potential
contaminant source. Additionally, Allen Harbor supports an active marina for the Town of
North Kingstown, RI on its eastern shore, and is a popular anchoring spot for day trips by
local boaters. A second marina, serving the Quonset Davisville Yacht Club, is located on
the southwest shore of the harbor. Fuel leakage, dispersion of hull antifoulant paints, and
septic wastes resulting from this intense boating and marina activity were suspected
potentially to impact harbor quality. To clarify the role of NCBC-associated waste sites on
the observed impacts to Allen Harbor, activities in Phase n focused on partitioning
contamination and toxicity among these three potential sources: the NCBC landfill, surface
runoff from the surrounding land, and from boating and marina activities conducted within
the harbor. The approach taken involved implementation of a temporal and spatial sampling
plan which took advantage of the seasonal nature of boating activities in Allen Harbor. This
was accomplished through collection of water, sediment, and biota samples, and subsequent
quantification of contaminant levels and biological effects.
A second component of Phase II involved further examination of the hemopoietic
neoplasia observed in Allen Harbor Mya arenaria. Because harbor Mya displayed high rates
of Hn relative to Narragansett Bay stations, the possibility exists that Allen Harbor may be
acting as a source of the disease. To address this question, a one-time survey of Mya
neoplasia was conducted throughout the West Passage of Narragansett Bay. Samples of Mya
were collected at 20 stations, and were scored in the laboratory for rate of infliction within
each subpopulation. Additionally, research was conducted to identify chemical compounds
which could potentially be used to identify and quantify sources of contaminant input to
Allen Harbor. This effort involved a survey of existing inventories of chemicals disposed in
the landfill, and analyses of selected sediment samples to evaluate potential input from
sewage, runoff, atmospheric, and petroleum sources in addition to the landfill. The specific
findings of these activities are described briefly below.
12
-------
Exposure and Effects Partitioning
Environmental samples representing the three potential source inputs to Allen Harbor
were collected prior to the onset of major boating activity in Spring, at the height of the
boating season in Summer, and at the conclusion of the season in Fall. Replicate sampling
stations were established to characterize each source: three at active seeps along the face of
the Landfill, three in association with surface water Runoff (two at the mouths of major
creeks and one at a storm drain), and two within the areas of significant boating and Marina
activities in the harbor. Two additional stations in the West Passage of Narragansett Bay
(one located at mid-bay, one in the southern part of the bay) provided Reference information.
Samples were obtained of input water, sediments (as proximal receptors of water-bourne
contaminant input), and biota to assess exposure conditions through chemical and microbial
analyses, and to evaluate potential biological effects through performance of laboratory
bioassays and biomonitoring activities. Data analyses focused upon comparisons among
sources and seasons to address the central questions of Phase H, and among individual
stations to identify targets of potential remediation.
Chemical analyses of water and receiving sediments indicated significant contributions
of PCB, PAHs, pesticides, and trace metals by the NCBC Davisville landfill and surface
water runoff sources. Although large variation in water-oourne contaminant levels
confounded analyses of relative source strengths, the Spink Neck (SN) storm drain (located at
the southern end of the harbor) and a seep located along the middle face of the landfill
(LANDM) were identified as the major contributors. Volatile organic compound (VOC)
concentrations were ubiquitously low throughout the study. Tissue chemistry of deployed
Mytilus edulis, measured to evaluate Marina water column exposure conditions, indicated
higher PCB and PAH levels in the harbor relative to Narragansett Bay reference stations, but
lower concentrations of metals such as Cr and Pb. Typically, residues-in both areas were
highest in Spring, prior to the onset of intense boating activities. Patterns of tissue
chemistry of native Mercenaria, another water column suspension feeder, reflected those
observed in Mytilus.
Highest densities in water samples of fecal coliforms, indicators of sewage
contamination and exposure to pathogens, were observed at the Runoff station North Creek
13
-------
(NC). Levels also were elevated at West Creek (WC), but were relatively low at Marina and
Reference stations. Peak concentrations typically occurred in Summer at all stations. These
patterns were generally reflected in the concentrations of coliforms in indigenous and
deployed bivalves: highest levels were observed in ribbed mussels (Modiolus demissus)
collected from the Runoff stations NC and WC in Summer and Fall, and levels of indicator
bacteria in Mercenaria and deployed Mytilus tended to be lower than those in ribbed mussels.
Statistical comparisons of receiving sediment chemistry indicated Runoff and Landfill
sources to be the largest contributors of contaminant input, whereas concentrations associated
with Marina stations were indistinguishable from Reference. SN and LANDM were again
implicated as the major contributors of chemical stressors: PAHs and trace metals were
highest at SN, while PCBs and the pesticide DDE were highest at LANDM. Sediment
chemistry was correlated with that of the source water input for metals, but not for organic
compounds. Tissue chemistry of indigenous Modiolus indicated overall higher levels of
PCBs, DDT, and Cr at Landfill stations relative to animals from Runoff stations, although
PAH concentrations were highest at SN. Contaminant residues were correlated with organic
chemistry in sediments, but were statistically unrelated to trace metal sediment chemistry or
to source water chemistry.
No overall differences in water toxicity, as measured by sea urchin fertilization
success, were observed between Landfill and Runoff sources or among seasons. However,
toxicity of water collected from SN was statistically higher than those of all other stations.
LANDM samples also displayed some toxicity. Fairly strong negative correlations were
observed between fertilization success and the concentrations of PAHs and metals in source
waters. Marina waters caused statistically higher 7-day mortality in mysid shrimp
(Mysidopsis bahia) laboratory exposures, while differences between sources in mysid
reproduction and growth were lacking. Mortality was highest in Early Summer (an
additional sampling event conducted immediately after onset of boating activities), but
reproduction and growth were highest in Summer and Fall, respectively. As evaluated by
these endpoints, Spring water quality was relatively poor: mortality was high, and
reproduction and growth were low. No differences were observed in endpoints among
14
-------
stations. Experimental field deployments of Mysidopsis were attempted in during all four
seasons, but were successful in Summer and Fall only. No source or seasonal differences
were observed in mortality, but both reproduction and growth were higher at Marina stations
than at Reference stations. In all seasons except Early Summer (when no differences
occurred), the SFG index of Mytilus was lower for mussels deployed at Marina stations.
Reduced SFG in Spring suggested deterioration of harbor water quality prior to the boating
season. SFG was negatively correlated with Mytilus PCB, PAH, DDT, and Cu tissue
residues.
Statistically significant source and seasonal effects were absent in the acute mortality
response of Ampelisca, although Spring mortality tended to be lower at most stations.
Biologically significant rates of mortality were observed at LANDM, WC, and SN in
Summer and Fall. Toxicity was correlated with sediment concentrations of DDE, CU, Ni,
Pb, and Zn.
The assessments conducted during Phase II strongly implicated the NCBC landfill and
surface water runoff as important contributors to contaminant exposure and biological effects
in Allen Harbor. The design utilized in that phase did not permit evaluation of the absolute
magnitude of these contributions, nor of their relative importance. However, seasonal and
spatial patterns in exposure and effects associated with boating and marina activities
suggested this source to be relatively unimportant to the environmental quality of Allen
Harbor.
Mya Neoplasia Survey
A semi-synoptic survey of neoplastic disease in Mya aremric was conducted at 20
stations in Allen Harbor and the West Passage of Narragansett Bay during the spring of 1990
to provide correlative information regarding the role of NCBC contaminant sources in Hn
etiology. Of 820 animals examined, 91 contained neoplasms. Average incidences of Hn
varied among stations from 0% (at several stations) to 37.9% at a station within Allen
Harbor. Overall, the rate of Hn affliction was related to proximity to the NCBC landfill,
indicating a possible association between Mya neoplasia and release of contaminants from this
15
-------
site.
Chemical Markers Research
A review of past disposal practices and of the types of materials disposed at Calf
Pasture Point indicated the hydrolyzation product of l,3-dichloro-5,5-dimethylhydatoin to be
a potential chemical marker useful in assessing the contributions of Calf Pasture Point to
environmental contamination in Allen Harbor. The parent compound, referred to as
Decontaminating Agent Non-Corrosive (DANC), was used by Naval Air Station (NAS)
Quonset Point while the base was in operation repairing helicopters. H-DANC, produced
when DANC is exposed to water, was quantified along with chemical markers of sewage,
runoff, atmosphere, and petroleum source inputs in five selected sediments from Allen
Harbor. The results of these analyses were used to evaluate the utility of the chemical
approach to partitioning source contributions of chemical stressors.
The levels of marker compounds quantified in sediments indicated several potential
origins of contamination to the harbor, including sewage, petroleum, and atmospheric
sources. The pattern of sewage markers suggested direct input of fecal material, rather than
input from municipal treatment facilities. Petroleum marker relationships implicated both
direct introduction of high molecular weight petroleum mixtures (such as crankcase oil), and
indirect input from pyrogenic sources. Atmospheric sources were implicated by the ratios of
specific PAHs. H-DANC was not detected in any of the five samples analyzed, nor was a
marker of roadway runoff. The absence of these two compounds should not be interpreted to
suggest that Calf Pasture Point and surface runoff are insignificant contaminant sources, as
chemical and/or biological degradation may have reduced their concentrations to levels below
detection.
PHASE HI APPROACH
The results obtained in Phases I and II indicated some degree of risk to ecological
systems in Allen Harbor from chemical contaminants in the NCBC landfill. Phase III
activities focused upon the direct quantification of this risk. Additional activities were
16
-------
conducted to define the role of landfill contaminants in the etiology of hematopoietic
neoplasia in Allen Harbor Mya arenaria. In conjunction with Phases I and II, this work
provides the information necessary to support characterization of ecological risks associated
with NCBC Davisville.
The approach established for quantification of ecological risk relied upon
characterization of the responses of a number of benthic and water column species to direct
exposure to landfill material. This effort consisted of the performance of laboratory
exposure-response bioassays involving landfill seep water, sediments, and sediment extracts
collected in close association with the landfill. The bioassays examined a variety of acute
and chronic endpoints of survival, growth, and reproduction. Assay results were
summarized into exposure-response models describing biological impact at any level of
contamination. This approach is outlined in Norton et al. (1988).
The rationale for selecting seep water, whole sediment, and sediment extracts as
surrogates for the landfill itself involved the feasibility of obtaining appropriate exposure
media, as well as the validity of assays of landfill material with respect to marine systems.
Previous reconnaissance activities conducted by TRC Environmental Consultants, the on-site
contractor for the NCBC Davisville RI/FS process, indicated a low potential for collection of
ground water within the landfill. Seeps were therefore selected as sources of water to obtain
volumes sufficient for bioassay conduct. This medium likely represents the most immediate
route of contaminant transport into Allen Harbor. Whereas EPA's Superfund Program has
cautioned against the use of exposure media inappropriate to the species used in toxicity
bioassays (Charters 1990), sediments were selected over landfill soils to examine risks to the
marine system of the harbor. Utilization of sediment extracts permitted characterization of
the effects of contaminants at all concentrations potentially available to biological systems.
Assays were selected for inclusion in Phase III based upon the following (unordered)
criteria:
* involvement of species representative of the Allen Harbor benthic and water
column systems,
involvement of a range of taxonomic groups,
17
-------
involvement of endpoints addressing a range of ecological organization,
potential for extrapolation to higher-level endpoints,
ecological relevance of resulting data,
suitability to quantifying effects to the selected exposure media,
suitability to quantifying effects across exposure media,
relevance of resulting data to results obtained in the previous two phases of the
study, and
ป feasibility of and familiarity with exposure-response assay protocols.
The species and endpoints of assays" meeting these criteria and therefore utilized in Phase HI
are indicated in Table 1. For the most part, these assays were developed at ERLN to
evaluate toxicity and effects of exposure media from marine and estuarine settings, and their
utility has been validated in a variety of laboratory and field programs. ERLN Standard
Operating Procedures have been developed for each assay and can be found in Mueller et al.
(1992).
The second component of Phase HI involved laboratory investigation into the etiology
of soft-shell clam {Mya arenaria) hematopoietic neoplasia. High incidences of this disease, a
disseminated sarcoma occurring in bivalves, were first observed by Farley (1969) in eastern
and Pacific oysters. Since this time these malignant neoplasms, similar in nature to
vertebrate leukemia, have been documented worldwide in 15 species of oysters, clams,
cockles, and mussels (Peters 1988). During Hn, normal circulating hemocytes are replaced
by round, non-aggregating, anaplastic cells which have lost their ability to adhere to glass, to
form pseudopods, and to phagocytize and neutralize foreign particles (Beckmann 1989).
They have a large nuclear to cytoplasmic ratio, a distinct nucleolus, and a high mitotic index
with abnormal figures. As Hn progresses the number of aberrant cells increases, invading
and destroying the soft tissues of the clam and leading eventually to death.
While more recent field surveys such as those conducted by Farley et al. (1986) have
18
-------
Table 1. Species and Midpoints used in development of exposure-
response models.
Exposure
Medium
Species
Eodpoints
Seep water
indicated an increase in the
prevalence of Hn,
researchers have been unable
to determine the etiology of
this disease. It has been
attributed to infectious
agents, such as viruses, to
pollution, and to
transmission of neoplastic
cells from introduced or
transplanted affected
organisms to healthy
indigenous populations.
Brown (1980), Appeldoom
et al. (1984), and Farley
(1989) have successfully
conducted transmission
studies by holding healthy
animals under head tanks
containing neoplastic clams
and through the inoculation
of neoplastic cells into
healthy animals. However,
attempts to confirm the
presence of a virus,
originally isolated by
Oprandy et al. (1981), have failed, as have several attempts to correlate Hn with
environmental pollution (Mix 1979, 1986).
Hn was reported in Rhode Island soft-shell clams as long ago as 1976 (Brown et al.
1976, 1977) and again by Cooper et al. (1982a, 1982b). In Allen Harbor, incidences of Hn
Sediment
Sediment extract
Arbacia
(sea urchin)
Mysidopsis
(mysid shrimp)
Champia
(red alga)
Mulinia
(coot clam)
Menidia
(siiverside minnow)
Ampelisca
(benthic amphipod)
Mulinia
(coot clam)
Arbacia
(sea urchin)
Mulinia
(coot clam)
Photobacterium
(bacterium)
fertilization
larval development
larval mortality
mortality
reproduction
larval development
mortality
mortality
growth
mortality
fertilization
larval development
larval mortality
mortality
mortality
19
-------
as high as 23% have been documented (Munns et al. 1993). A single long-term laboratory
study was conducted during this phase to determine the potential role of landfill contaminants
in Hn etiology. Hn-free Mya were injected with hemolymph from non-affected and affected
animals and exposed in the laboratory to either landfill or reference sediments for 90 days.
Animals were examined for mortality, growth, and for the presence of Hn sarcoma cells.
This experiment would provide Information sufficient to assess risks to Allen Harbor clams
of neoplasia development.
Phase III of the Risk Assessment Pilot Study began in October 1990. ERLN was the
lead laboratory in this study with the cooperation and participation of NCCOSC. The
remainder of this document describes ihe activities performed to address Phase III objectives
and the results obtained through their conduct.
20
-------
Methods
The material in this section provides a description of methods used to address the
objectives of Phase HI. The two major activities in this phase, the establishment of
exposure-response relationships and the evaluation of the role of landfill contaminants in
development of Mya Hn are described separately. Because quality assurance/quality control
procedures are common to all activities, their general description is given under a separate
heading.
QUALITY ASSURANCE/QUALITY CONTROL
This project has been conducted in accordance with all ERLN quality assurance and
quality control procedures outlined in the Work/Quality Assurance Plan for Marine
Ecological Risk Assessment Pilot Study (ERLN/NOSC 1991), the ERLN Standard Operating
Procedures Manual (ERLN 1991), the Standard Operating Procedures and Field Methods
Used for Conducting Ecological Risk Assessment Case Studies at: Naval Construction
Battalion Center Davisville, RI and Naval Shipyard Portsmouth, Kittery, ME (Mueller et al.
1992), and the Quality Assurance Program Plan for the Environmental Research Laboratory -
- Narragansett and Newport (ERLN 1992). The first document addresses quality assurance
steps undertaken for the specific activities of this project. The two Standard Operating
Procedures (SOP) manuals describe the methods used to perform the biological, physical, and
chemical assessments of this project. The last document describes general quality assurance
requirements for research activities at ERLN. A copy of these documents have ban added
to the administrative record for NCBC Davisville, and may be obtained by contacting ERLN
or NCBC Davisville.
All data generated during sample collection, preparation (e.g., dry weight, wet
weight, volume, etc.), and analysis were entered into computerized data bases for use in
subsequent data reduction and statistical analysis. A description of the data management plan
for this project is given in Rosen et al. (1988). In addition to describing quality
21
-------
assurance/quality control (QA/QC) considerations with respect to data storage, transfer, and
manipulation, this document provides a description of data base design and its relationship to
the interdisciplinary data management strategy of ERLN. This document is also part of the
administrative record, and may be obtained by contacting ERLN or NCBC Davisville.
A large portion of the QA/QC procedures used for this study were specific to each
type of activity. For example, calibration of specific instrumentation is relevant only to the
operation of that instrument. These procedures are described in the two SOP manuals cited
above. However, important QA/QC descriptions are given where appropriate throughout the
remainder of this Methods Section.
EXPOSURE-RESPONSE MODELS
Bioassays were performed to evaluate the effects of landfill exposure media on marine
organisms and to establish exposure-response relationships for development of an ecological
risk assessment model. Water emanating from landfill seeps, sediments associated with the
landfill, and extracts of sediments associated with the landfill were used in the laboratory as
exposure media. Prior to sample collection, data from Phases I and 13 were reviewed to
select a site of maximum contamination. This was necessary to ensure the fiill range of
exposure concentrations required to adequately describe the responses of test organisms to
landfill contaminants. LANDM, located at a seep in the middle of the face of the landfill
(see Munns et al. 1991, 1993 for descriptions of stations associated with the NCBC landfill),
was chosen as the site for exposure media collection.
The basic protocol for the laboratory assays required serial dilutions of the landfill
material with appropriate (i.e., relatively uncontaminated) reference materials. Contaminant
bioavailability in sediments has been shown to be influenced by such sediment attributes as
total organic carbon (TOC) (e.g., Di Toro el al. 1991), acid volatile sulfide (AVS) (e.g., Di
Toro et al. 1990), and grain size, which is related to the particle surface area available for
contaminant sorption. Thus the diluting sediment for solid phase exposures needed to match
LANDM in these attributes. Several sediments from relatively clean areas of Narragansett
Bay and from Central Long Island Sound (CLIS) were evaluated for TOC, AVS, and grain
22
-------
size. ASTM methods (1988) were adapted by Huffman Laboratories, Inc. (Golden, CO) for
sediment TOC determinations. Total carbon
concentrations were quantified by high
temperature combustion of 0.1 to 1.0 g of
sediment in a model CR12 Analyzer.
Carbonate carbon was measured with a
Coulometric 14D instrument as carbon
dioxide. AVS in LANDM and potential
diluting sediments were evaluated according
to methods described in Boothman and
Helmstetter (1993) and Johnston (1993).
Volatile hydrogen sulfide released by HC1
from aliquots of homogenized sediment
samples was trapped with a sulfide anti-oxidant buffer solution and evolved S2" was quantified
with a sulfide ion-specific electrode. Sediment grain size was determined by a sieve and
centrifuge method as described in ERLN SOP 1.01.005 (Mueller et al. 1992). A hydrogen
peroxide solution was added to 12 g of dried sediment. The sample was sonicated, washed
through a sieve, and centrifuged. The supernatant was decanted, distilled water was added,
the silt plus clay fraction was resuspended and centrifuged several times to remove all clay
particles. The remaining silt fraction was dried and weighed. The proportion of clay was
determined by subtracting the weights of the sand and silt fractions from the weight of the
total sample. Based on these analyses (Table 2), sediment from Potowomut Cove (POTO), a
small riverine inlet to the north of Allen Harbor, was selected as the diluting sediment.
Exposure Media Collection and Preparation
Seep samples were collected from LANDM directly into Nalgeneฎ bottles, transported
on ice in insulated coolers, and refrigerated at 4ฐC until used. Serial dilutions of 5.8, 11.5,
23, 46, and 92% LANDM seep water were constructed with Narragansett Bay brine and
deionized water prepared according to procedures described in EkLN SOP 1.01.004
Table 2. Characteristics of sediments used in
bioassays.
LANDM POTO
AVS 0iM/g)
51.00
53.64
TOC (%)
3.74
3.16
Sand (%)
92.7
47.7
Silt (%)
4.0
40.4
Clay (%)
3.3
11.9
23
-------
(Mueller et al. 1992). Surficial sediments were collected at LANDM and at POTO with a
large teflon coated spatula, transported in acid-stripped 2-g glass jars on ice, and refrigerated
at 4ฐC until used. Serial dilutions of 12.5, 25, 50, and 100% LANDM sediments were
constructed with POTO sediment on a volume basis. Following homogenization, these
mixtures were allowed to "age" for 30 d under refrigeration at 4ฐC to permit equilibration of
contaminants among sediment surfaces.
Extracts of LANDM sediment for use in laboratory exposure-response assays were
prepared by sonicating samples (approximately 300 g wet weight) in acetonitrile and
centrifuging three times. The supematants were combined in pentane-extracted deionized
water and back-extracted three times with pentane. The extracts were combined, dried over
sodium sulfate, reduced twice in pentane, brought to dryness, and dissolved in
dimethylsulfoxide (DMSO). Detailed methods for preparing sediment extracts are described
in Munns et al. (1991) and Mueller et al. (1992). Final exposure concentrations of 0.001,
0.003, 0.006, 0.010, 0.013, 0.025, 0.05, 0.10, 0.15, 0.20, and 0.25% were created by
diluting the original extract with reconstituted Narragansett seawater.
Laboratory Assays
The exposure media (prepared as described above) were evaluated over a range of
concentrations in the laboratory bioassays indicated in Table 1. Detailed descriptions of the
standard methods employed in each of these are given In Mueller et al. (1992). These tests
are described briefly below, with references made to the corresponding SOP.
Seep water The effects of contaminants currently migrating from the landfill in seep water
on sea urchin (Arbacia punctulata) fertilization were evaluated following ERLN SOP
1.03.0^5. In this test, gametes obtained from adults collected at reference field sites were
artificially released in response to electrical stimulation and collected using a syringe with a
blunted needle. One milliliter of eggs (2000/ml ฑ 200) was added for 20 minutes to 100 (il
of a 5 X 107 cell/ml suspension of sperm which had previously been exposed to 5 ml of seep
water dilutions in scintillation vials for 1 h at 20ฐC. Two ml of 10% formalin in seawater
24
-------
were added at the conclusion of the test, and 100 eggs were observed by compound
microscope (100X) in a Sedgwick-Rafter counting chamber for the presence of a membrane
surrounding the egg, an indication of successful fertilization.
Seep water effects on Arbacia larval development and survival followed procedures
given in ERLN SOP 1.03.007. Gametes were obtained and diluted as described above, and
eggs and sperm were co-exposed for 48 h in 10 ml of seep water dilutions in scintillation
vials at 20ฐC. Larvae were preserved by the addition of 2 ml of 10% formalin in seawater
and were stained for microscopic observation with Rose Bengal in a 10% buffered
formalin/seawater solution. Two hundred larvae were examined for development and
mortality.
During the Mysidopsis bahia acute toxicity test (ERLN SOP 1.03.003), ten 1-5 day-
old mysid shrimp (cultured according to methods described in ERLN SOP 1.01.003) were
exposed at 20ฐC in two replicates of 200 ml of seep water dilutions for 48 h. Non-motile,
opaque organisms were recorded at assay termination.
The effects of seep water on red algae (Champia parvula) reproduction were
evaluated following ERLN SOP 1.03.001. One male and 5 female Champia branches,
cultured in the laboratory as described in ERLN SOP 1.03.001, were exposed in 250 ml
Erlenmeyer flasks to 100 ml seep water dilutions for 2 days at 23 ฐC. After a 5-7 day
recovery period, the number of mature cystocarps produced by each female was enumerated
by stereomieroscopy.
During the Mulinia embryo-larval toxicity test, field-collected adults maintained in the
laboratory were cooled to 4ฐC for 0.5-2 h and warmed to 25-28ฐC to induce spawning
(ERLN SOP 1.03.008). Seven hundred and fifty embryos were exposed to 3 replicates of 10
ml of seep water dilutions in sciutillation vials no more than 2 h after fertilization for 48 h at
22ฐC. Formalin preserved larvae were examined in Sedgwick-Rafter counting chambers by
compound microscope (100X) for the presence of shells, an indication of normal
development and therefore survival.
Survival effects on silverside minnow (Menidia beryllina) were determined for ten 7-
11 day old Menidia larvae cultured according to methods described in ERLN SOP 1.01.003,
25
-------
exposed at 20ฐC in 2 replicate 250 ml chambers for 96 h containing 200 ml seep water
dilutions. Non-motile, opaque animals were recorded as dead (ERLN SOP 1.03.004).
Sediment Toxicity of sediments associated with the landfill was assessed to evaluate
contaminant migration which may have occurred in the recent p* The 10-day amphipod
(Ampelisca abdita) acute mortality tests was performed following ERLN SOP 1.03.002.
Briefly, twenty immature amphipods collected from a nearby reference site were exposed in
1-quart jars to 200 ml dilutions of LANDM sediment for 10 days at 20ฐC. Each jar was
monitored daily and at test termination for dead or moribund organisms.
Evaluation of sediment effects on Mulinia growth and mortality followed procedures
described in Burgess and Morrison (in prep.). Survival and weight was measured in 10 1-
mm laboratory-cultured Mulinia juveniles obtained from artificially spawned field-collected
adults. Juveniles were exposed to 50 g of the landfill sediment dilutions in 150 ml beakers
for seven days at 22 ฐC. Mortality was quantified by the presence of open shells and through
microscopic observation for bacteria infestation. Whole animal dry weight was calculated
and compared to control or reference whole animal dry weight.
Sediment extracts To ensure a full range of contaminant concentrations for development of
the exposure-response models, extracts of LANDM sediment were evaluated using four
standard methods. In addition to effects on Arbacia fertilization, Arbacia larval development
and survival, and Mulinia mortality (described above), assessments of sediment extract were
conducted using Microtoxฎ (Photobacterium phosphoreum) mortality as an endpoint.
Microtox methods are described in detail in ERLN SOP 1.03.009. Briefly, suspensions
containing 106 colony-forming units of the luminescent bacterium Photobacterium
phosphoreum were exposed to dilutions of sediment extracts for 15 min at 15ฐC.
Bioluminescence was monitored using a model 2055 Microtox Toxicity Analyzer (Beckman
Instruments).
26
-------
Model Development
Data obtained from laboratory assays were used to develop quantitative models of
biological response to landfill media exposure. These exposure-response models utilize
whole-waste concentrations as independent variables determining the level of endpoint
response for each test species.
The model identified to describe the exposure-response relationships is based upon the
assumption that thresholds exist in the sensitivities of organisms to contaminant
concentration, ar.d that due in part to inter-individual variability, these thresholds are log-
normally distributed within the test population as a function of exposure concentration. Thus
individuals may respond at different exposure concentrations and the overall test population's
response to varying exposure concentrations can be modeled as a log-normal distribution (or
as a Gaussian (normal) distribution if concentrations are log-transformed). This model
describes an S-shaped logistic curve, the pattern classically observed in dose-response
relationships. This pattern of response was observed in several of the assays described above
(see Results). The approach taken here offers advantages over other logistic curve fitting
procedures (e.g., Bamthouse et al. 1987, Munns and Comeleo 1991), including its
mechanistic theoretical basis, and that data obtained from reference (background or control)
treatments can be incorporated directly into the model fitting procedure.
A nonlinear least-squares regression procedure described by Bruce and Versteeg
(1992) was used to estimate parameters of the model:
where R is the predicted biological response at exposure concentration C, R<, is the biological
response observed in reference or control treatments, # is the cumulative area under the
standard Gaussian distribution, EC, is the Xth percentile effects concentration, Z, is the
normal deviate above which x percent of the Gaussian distribution lies, and a is the standard
deviation of the Gaussian distribution. EC, and a can be thought of as parameters describing
the position and slope of the cumulative Gaussian distribution, whereas Ro describes the level
C>0
c=o
27
-------
of response expected in the absence of contaminant exposure (i.e., the intercept).
Assay data sets needed to meet two criteria of acceptance to be successfully modeled.
The first of these was that the responses had to increase or decrease reasonably
monotonically with increasing exposure concentration. Large deviations from this
requirement would suggest the lack of cause-and-effect relationship between the two
variables, and further would yield a poorly fitting model of little value for quantifying risk.
The second criterion was that S50% of the full response range needed to be present in the
data set. A realized range of response less than this would not only suggest the lack of
sufficiently high exposure concentrations, but also would yield unreliable estimates of the
parameters ECX and a. Data sets failing this criterion also would yield a poorly fitting model
of little value for quantifying risk.
The NLIN Procedure in SASฎ (SAS Institute 1989) was used to determine estimates
of R< EC*, and a resulting in the best description of each of the data sets meeting the above
acceptance criteria. The resulting models could then be used to predict the toxicological
responses of test organisms to whole-waste exposure.
MYA LABORATORY EXPOSURES
The experimental design employed in the laboratory assessment of the role of the
NCBC Davisville landfill in neoplasia development involved a two-way design incorporating
Mya exposed to landfill sediment or reference sediment, and the presence or absence of a
suspected transmissible component. This design required injection of previously unaffected
animals with hemolymph obtained from animals displaying the disease. Results obtained
during Phases I and II indicated the FDA station in Allen Harbor to be a source of affected
Mya. Clams were obtained from intertidal zone of this site at low tide with clam forks and
garden rakes. One tenth of one milliliter of hemolymph was drawn from the posterior
adductor muscle of each animal into a 1-cc tuberculin syringe with a 26-gauge needle and
diagnosed by phase-contrast microscopy of hemolymph hemocytometer preparations.
Affected hemolymph was injected into the siphon (Farley 1989) of disease-free 1-3 year old
clams (evaluated as above) obtained from a commercial supplier for treatments involving the
28
-------
presence of a transmissible component. The remaining clams received a placebo injection of
disease-free hemolymph.
Inoculated Mya were planted in 10-gal aquaria containing 4 gal of sediment obtained
from either the landfill (AH) or at a reference station in Narrow River (PR). Neoplasia was
not observed at PR during the Phase II Mya survey (Munns et al 1993), nor has it ever been
reported at that site. Surficial sediment (top 2 cm) had been collected previously from the
intertidal zones of these stations at low tide with a teflon-coated scoop, and was kept under
refrigeration at 4ฐC until used. Each of the four treatments was replicated twice, for a total
of eight aquaria. Following introduction of the inoculated animals (60/aquarium), aquaria
were supplied with ambient temperature seawater flowing at 0.1 L/min. Animals were
monitored for mortality and fed a suspension of the alga Isochrysis cultured as described in
ERLN SOP 1.03.013 (Mueller et al 1992) daily.
Fixed hemolymph cells were examined by bright-field microscopy every 30 days for 3
months according to the histocytological methods described by Farley et al (1986) with
several modifications. Briefly, cells were allowed to settle on 1 % poly-L-lysine coated
standard microscope slides for 30 min. Excess fluid was removed, slides were fixed in a 1%
glutaraldehyde and 4% formaldehyde seawater solution, and stained with a standard
histological Pap preparation. Pre- and post-exposure weights were measured to evaluate
growth. Two-way analysis of variance (ANOVA) was conducted to determine the effects of
sarcoma cell inoculations and contaminated sediments on Hn, mortality, and growth.
CHEMICAL ANALYSES
Analyses were performed to characterize organic and inorganic contaminants in
sediment, water, and Mya tissue samples. These procedures are described here briefly and
in detail in Munns et al. (1991). Seep water was collected in solvent-rinsed 1-L amber
bottles for organic chemical analysis and in acid-washed 250-ml polyethylene bottles for trace
metal analysis. This material was transported to ERLN on ice and refrigerated at 4ฐC until
used. Organic contaminants were extracted from water samples with methylene chloride,
29
-------
dried over anhydrous sodium sulfate, volume reduced, and separated into two fractions by
silicic acid column chromatography. The f, fraction was analyzed for PCBs and pesticides
using capillary gas chromatography (GC), and the f2 fraction was analyzed for pesticides and
PAHs using GC with electron capture detection (ECD) and gas chromatograph-mass
spectrometry (GC-MS). For inorganic analyses, water samples were acidified with nitric
acid, shaken, allowed to settle, subsampled into concentrated nitric acid, and analyzed by
heated graphite atomization (HGA) atomic absorption (AA).
Sediments were homogenized, sonicated with acetonitrile, and centrifuged three times
for organic analyses. The supernatants were combined and extracted with pentane. The
extract was volume reduced and separated into the three fractions for quantification as
described above. Homogenized sediments were freeze-dried, acidified with nitric acid,
sonicated, and centrifuged for trace metal analyses, lire supernatant was decanted, sonicated
with nitric acid, centrifuged, and decanted. The resulting supernatant was analyzed on an
inductively coupled plasma (ICP) emission spectrometer. Samples which required low
detection limits were analyzed by AA as described above for water samples.
Mya were collected as described above, and were frozen at -20ฐC in plastic bags for
chemical analyses. Homogenized tissue samples were rehomogenized with acetonitrile for
organic analyses. The sample was centrifuged and decanted into deionized water three times.
The supernatants were combined, extracted with pentane, and volume reduced. Silica gel
chromatography was used to obtain three fractions for quantification as described above for
water and sediment samples. Homogenized, freeze-dried, heated tissue samples were
microwave digested in nitric acid, cooled, and vacuum-filtered for inorganic analyses. The
filtrates were diluted with deionized water and analyzed by ICP (or AA, if necessary).
30
-------
Results
EXPOSURE-RESPONSE MODELS
Laboratory Assays
The presentation given below focuses upon the sensitivities of individual endpoint
responses evaluated over the full range of exposure media concentrations used in each
bioassay, with attention given to the utility of individual data sets in constructing exposure-
response models. Comparisons of species* sensitivities are made where appropriate. Results
are organized by exposure medium (seep water, sediment, or sediment extract). All bioassay
data are given in Appendices H-J.
Seep Water The fertilization response of Arbacia exposed to landfill seep water decreased
monotonically from near 100% in reconstituted Narragansett seawater to approximately 47%
at the maximum concentration of 92% LANDM seep water (Figure 2), indicating that levels
of contaminants in near full strength seep water were not sufficient to totally prevent
successful fertilization. Exposure concentrations below 23% seep water had little effect on
this endpoint.
Relatively little response to seep water was observed in Arbacia larval mortality:
survival was near 80% at the maximum seep water concentration tested (Figure 3).
However, developmental effects in this assay were measured at very low concentrations. The
combination of larval mortality and abnormal development resulted in the near absence of
normally developing larvae at concentrations of 46% seep water and above.
Seep water exposures were not acutely toxic to Mysidopsis, as measured by mortality
(Figure 4). With exceptions of minor mortality (< 10%) at concentrations of 0 and 23%, no
survival effects were measured through the full range of seep water concentrations.
Sexual reproduction in Champia was affected by seep water concentrations as low as
5.8%, and the number of cystocarps produced by each female decreased monotonically
thereafter (Figure 5). Little to no reproduction was observed at concentrations of 46% and
31
-------
Arbacia
FERTILIZATION SUCCESS
SEEP CONCENTRATION (%)
Figure 2. Fertilization Response of Arbacia to seep water exposure.
32
-------
Arbacia
48-HOUR LARVAL DEVELOPMENT
o Normal a Abnormal a Mortality
SEEP CONCENTRATION (%)
Figure 3. Larval mortality and development responses of Arbacia to seep water exposure.
33
-------
Mysidopsis
48-HR MORTALITY
100 i
80
n 60
$
O
CO S 40
20
0 4
20
40
r
60
SEEP CONCENTRATION
r
80
ฆฉ-
100
Figure 4. Acute mortality response of Mysidopsis to seep water exposure.
34
-------
Champia
Sexual Reproduction
SEEP CONCENTRATION (%)
Figure 5. Sexual reproduction response of Champia to seep water exposure.
-------
above.
Development of larval Mulinia shells, indicative of normal development and therefore
larval survival, was impacted by seep water concentrations of 23% and greater (Figure 6).
Apparent effects below this level (Figure 6) were not different from those experienced in the
reference Narragansett seawater treatment in this test (see Appendix H).
As with the mortality response of mysid shrimp, survival of Menidla larvae was
unaffected by seep water concentration (Figure 7). The apparent reduction in survival at low
concentrations was evidently unrelated to contaminant concentrations in exposure medium, as
no mortality was observed at concentrations higher than 11.5%.
In a comparison among species and endpoints, Arbacia development, Champia
reproduction, and Mulinia larval shell development were most sensitive to seep water
exposures. Acute mortality of Mysidopsis and Menidia larvae were not affected, showing no
response at the maximum exposure concentration and suggesting that seep water was not
acutely toxic to either species. Under the conditions of exposure used in these latter two
assays, insufficient information was obtained for development of exposure-response models.
Sensitivities of the Arbacia larval mortality and fertilization responses fell between these two
extremes, with less-than-total effects being observed at maximum seep water concentrations.
Although models of response to seep water exposure could be constructed for Arbacia
mortality and fertilization, they would be incomplete with respect to risks associated with
higher constitute contaminant levels than those existing in the LANDM sample used in these
assays.
Sediment An initial 10-day Ampelisca acute mortality assay was repeated after the first test
failed established QA criteria (reference mortality of greater than 10%). However, mortality
in this first test was equivalent between the 0 and 100% LANDM treatments. A second test
of these extreme concentrations was conducted in conjunction with an assay being performed
as part of a separate project, in which mortality in the 100% POTO treatment and in the
100% LANDM treatment were both less than 10%. Thus, amphipod mortality was
insensitive to exposure to landfill sediments in this assay, suggesting the lack of acute
36
-------
Mulinia
LARVAL MORTALITY
100
-------
Menidia
96-HR LARVAL MORTALITY
SEEP CONCENTRATION (%)
Figure 7. Larval mortality response of Menidia to seep water exposure.
-------
toxicity to Ampelisca. These data are not displayed graphically, but can be found in
Appendix I.
Mulinia 7-day mortality and growth tests also proved to be insensitive to solid phase
sediment exposure (Figure 8 and 9). Insignificant mortality and no trends in growth were
observed through the range of landfill sediment concentrations for development of exposure-
response models. Thus none of the Ampelisca or Mulinia endpoints measured in sediment
assays indicated this exposure medium to be toxic.
Sediment Extract The sea urchin sperm cell test was initially performed using extract
dilutions in excess of 0.05%. These concentrations proved to be too high, so a second test
was conducted using lower dilutions. Success of Arbacia fertilization in the second assay
displayed a graded response to increasing concentrations of sediment extract (Figure 10),
with near-maximum effects being observed at a sediment extract concentrations of 0.025 %
and above. Concentrations below 0.006% had little effect on fertilization success.
Abnormal development of Arbacia larvae increased dramatically at sediment extract
concentrations greater than 0.05%, whereas larval mortality was affected at concentrations of
sediment extract greater than 0.15% (Figure 11). As a result of these effects, normally
developing larvae were absent in treatments above the 0.10% dilution.
As with the Arbacia discussed above, initial dilutions of sediment extract used in the
Mulinia embryo/larval toxicity assay proved to be too high to adequately describe response,
and a second test was conducted. Extract exposure conditions in this second test again
resulted in minimal Mulinia larval shell development at all concentrations higher than 0%
(Figure 12). However, examination of responses in the DMSO chemical control treatment
indicated QA problems associated with the use of this compound as a carrier solvent: 88% of
the animals did not develop shells in the DMSO-only treatment (Appendix J). The observed
responses therefore were due to a solvent effect rather than to exposure to sediment extract
contaminants, and the entire test was invalidated. Insufficient resources were available to
explore use of other carrier solvents for this assay.
39
-------
Mulinia
7-DAY MORTALITY
100 i
80
-f*
O
~ 60
a
tc
o
s 40
20 -
20 40 60 80
SEDIMENT CONCENTRATION (%)
100
Figure 8. Seven-day mortality response of Mulinia to sediment exposure.
40
-------
Mulinia
7-DAY GROWTH
I " I " I ' I l 11 1 I
0 20 40 60 80 100
SEDIMENT CONCENTRATION (%)
Figure 9. Seven-day growth response of Mulinia to sediment exposure.
41
-------
Arbacia
FERTILIZATION SUCCESS
SEDIMENT EXTRACT CONCENTRATION (%)
Figure 10. Fertilization response of Arbacia to sediment extract exposure.
42
-------
Arbacia
48-HOUR LARVAL DEVELOPMENT
SEDIMENT EXTRACT CONCENTRATION (%)
Figure 11. Larval mortality and development responses of Arbacia to sediment extract exposure.
43
-------
Mulinia
LARVAL MORTALITY
SEDIMENT EXTRACT CONCENTRATION (%)
Figure 12. Embryo/larval toxicity response of Mulinia to sediment extract exposure.
44
-------
Photobacterium
MICROTOX
100
80
60
40
20 -
0.001
0.01
i
0.1
SEDIMENT EXTRACT CONCENTRATION (%)
Figure 13. Bioluminescence response of Photobacterium to sediment extract exposure.
45
-------
Photobacterium bioluminescence decreased geometrically with increasing extract
concentration in the Microtox test (Figure 13), indicating a strong response between
bacterium survival and contaminant exposure. This test was conducted using a dilution series
different from the other extract assays to follow the standard operating protocols suggested
by the manufacturer.
Overall, the Arbacia larval development assay was most sensitive to the suite of
environmental contaminants present in landfill sediment extracts, although Arbacia
fertilization and Microtox also were fairly sensitive. These three assays provided data
sufficient to develop exposure-response models.
Model Development
Of the 15 species/endpoint/exposure medium data sets obtained in this study, six
failed the criteria for exposure-response model development established earlier. Landfill
sediments elicited no observable response in Ampelisca (mortality) or Mulinia (mortality,
growth), whereas maximum exposure concentrations of seep water were insufficient to elicit
Arbacia, Mysidopsis, or Menidia mortality responses suitably high for the model fitting
procedure to be successful. Additionally, characterization of Mulinia growth in response to
sediment extracts was confounded by solvent carrier effects, thus invalidating this assay.
Although they contributed to the overall assessment of risks, these seven data sets were
eliminated from further quantitative analysis.
Least-squares estimates of EC20, a, and R<, for the remaining 8 assay data sets are
provided in Table 3, along with model coefficients of determination (describing the
percentage of total variance explained by the model). EQ0 is presented to indicate exposure
media concentrations at which small, but perhaps ecologically significant responses are
expected to occur. For the most part, models were generated which fit the assay data
reasonably well, as illustrated in Figures 14-20 and the R2s listed in Table 3.
46
-------
Arbacia
FERTILIZATION SUCCESS
SEEP CONCENTRATION (%)
Figure 14. Whole media exposure-response model for fertilization response of Arbacia to seep water exposure.
47
-------
Arbacia
48-HOUR LARVAL DEVELOPMENT
SEEP CONCENTRATION (%)
Figure 15, "Whole media exposure-response model for larval development response of Arbacia to seep water exposure.
48
-------
60
40
Champia
Sexual Reproduction
20 1!
wQm
40 60
SEEP CONCENTRATION (%)
80
100
Figure 16. Whole media exposure-response model for sexual reproduction response of Champia to seep water exposure.
49
-------
Mulinia
LARVAL MORTALITY
|
3
CO
cn
o ซ*p
6s
100
80
60
40
20
100
SEEP CONCENTRATION (%)
figure 17. Whole media exposure-response model for larval survivorship response of Mulinia to seep water exposure.
50
-------
Arbacia
FERTILIZATION SUCCESS
SEDIMENT EXTRACT CONCENTRATION (%)
figure 18. Whole media exposure-response model for fertilization response of Arbacia to sediment extract exposure.
51
-------
Arbacia
48-HOUR LARVAL DEVELOPMENT
SEDIMENT EXTRACT CONCENTRATION (%)
figure 19. Whole media exposure-response model for larval development response of Arbacia to sediment extract exposure.
52
-------
Photobacterium
MICRQTOX
SEDIMENT EXTRACT CONCENTRATION (%)
figure 20. Whole media exposure-response model for bioluminescence response of Photobacterium to sediment extract
exposure.
53
-------
Table 3. Whole-media exposure-response model parameter estimates and coefficients of determination.
Exposure
Medium Species/Endpoint EC^,* tr Rob RJ
Seep water
/4r&aria/fertilization
60.3
0.21
99.2
999
Arbacia/nonasl development
13.5
0.17
92.6
994
Champia/mptoductioa
9.20
0.27
23.2
883
Mulinia/mortality
19.1
0.21
79.5
931
Sediment extract
Arbacia/kitiliiation
0.01
0.14
97.9
972
Arbacialnormal development
0.06
0.09
89.6
Arbacialmortality
0.19
0.05
89.1
995
PhotobacteriumfmortsilHy
0.004
1.0
100
985
* In units of % seep water or % extract.
b Cystocarps/female for Champia, percent for all other.
MYA LABORATORY EXPERIMENT
Suggestively higher rates of neoplasia development and mortality in Mya inoculated
with affected hemolymph and exposed to AH sediment were observed at the end of the 90-d
laboratory exposure (Table 4). (In this table and the following text, references to inocula
containing sarcoma cells are indicated by a appended to the sediment treatment
acronym, whereas inocula of unaffected hemolymph is referenced by a However, two-
way ANOVA indicated no significant sediment or inoculation effects in any of the three
biological endpoints measured. Because concentrations of contaminants in PR were
substantially lower than those in AH (Appendices A-E), these results support observations
made during Phase II (Munns et al. 1993) and other studies (e.g., Mix 1979, 1986; but see
54
-------
Walker et al. 1981) regarding the
general lack of correlation between
Hn and sediment chemistry.
It is possible that
contaminants in PR sediments were
more available to Mya than were
those in AH treatment. In contrast
to the initial efforts undertaken for
the exposure-response assays to
match sediment attributes which
Table 4. Mean responses of Mya endpoints in the 90-d
neoplasia experiment.
Endpoint
PR-
Treatment
PR+ AH-
AH+
Neoplasia (%)
28.1
32.5
39.2
47.5
Mortality (%)
13.3
10.9
8.3
16.7
Growth (g)
-1.0
-1.2
-1.0
-1.6
control contaminant bioavailability, the most important characteristics for the reference
sediment used in the Mya laboratory exposure were low contaminant concentrations and the
absence of neoplastic disease at the collection site. The diluting sediment used in the
exposure-response bioassays (POTO) was
unavailable for the Mya experiment because high incidences of neoplasia at been observed at
Marsh Point (MP), located at the mouth of Potowomut Cove, during Phase I and n
investigations (Munns et al. 1991, 1993). Bioaccumulation factors (BAFs) for most
contaminants (with the exception of
Table 5. Range ofbioaccumulation factor* of sediment pAH Fg md Dm% were
contaminants in the 90-d neoplasia experiment.
higher in clams exposed to PR (Table 5).
While this implies that contaminants in
PR sediments were more available for
accumulation by the organism than
contaminants in AH sediments, tissue
concentrations in PR clams were lower
than those of AH clams due to the overall
higher levels of contamination found in
the latter sediment.
Despite these general
Contaminant
Class
AH
PR
PCBs
0 - 1.27
36.5 - 74.5
PAHs
0 - 0.23
0 - 4.64
Pesticides
0.03 - 1.39
7.33 - 35.5
Metals
0.03 - 1.59
0.27 - 850
55
-------
observations, the ubiquitous presence of Hn in all treatments renders suspect the assumptions
of the experiment that no Hn should have occurred in the PR- treatment. Hn was not
observed in Mya collected at PR during the field survey conducted in Phase n (Munns et ah
1993), so it is unlikely that some characteristic of that sediment initiated or promoted
neoplasia development. Because of the lack of Hn in the large number of test animals
screened at test initiation, it also is unlikely that these animals were affected prior to the
laboratory exposure. One possibility is that hemolymph from animals diagnosed as being
unaffected and used to inoculate the PR- and AH- animals was in the initial stages of Hn
development and were contaminated with small numbers of sarcoma cells which went
undetected in the initial screening. Cooper et al. (1982b) have demonstrated that diagnoses
of mild cases of Hn by the histocytological technique used here to be accurate only 66 to
71% of the time. Perhaps the use of more sophisticated diagnostic techniques, such as those
involving monoclonal antibodies developed by Reinisch et al. (1983) and Smolowitz and
Reinisch (1986), would have minimized the potential for misdiagnosis of Hn.
56
-------
Quantification of Ecological Risk
RISKS OF TOXICOLOGICAL IMPACT
The exposure-response models developed through performance of laboratory bioassays
provided insight to the potential effects of landfill-associated contaminants on a range of
species and endpoints on a whole-medium basis. For example, assays involving sediment
exposures indicated little to no acute toxicity to Ampelisca or Mulinia, whereas seep waters
diluted down to 20% affected Arbacia larval development. This information is valuable in
examining the risks to ecological systems in immediate association with the landfill, and
could be used to evaluate remediation options. However, each exposure medium was a
complex mixture of many environmental contaminants. This multiplicity of chemical
stressors renders decisions regarding ecological risks posed by the landfill to the greater
Allen Harbor system somewhat difficult, because the behaviors of individual contaminants
differ in response to geochemical and biological processes acting in the harbor. The
concentrations of individual contaminants in any given environmental sample likely do not
represent simple dilutions of landfill media.
Risk Quantified by Toxic Unit Exposure-Response Models
To address this problem, a normalization procedure was employed involving the
concept of a toxic unit (TU) which allowed more direct comparison of exposures associated
with environmental samples with those of landfill media. EPA utilizes the TU approach in
its water quality-based toxics control program (U.S. EPA 1991). In a general sense, a toxic
unit is simply the ratio of a contaminant concentration to some biological benchmark
concentration for that chemical (such as an LCM or EC50), and is often expressed as a
percentage. As such, it is arithmetically similar to the risk quotients developed in Phase I
(Munns et ah 1991). In the current analysis, however, contaminant-specific TUs were
summed to derive a single, aggregated metric (CTU) of chemical contamination for each
unique environmental sample. This approach assumes additivity in the toxic actions of
57
-------
contaminants in complex mixtures, a conjecture of some debate (see Alabaster and Lloyd
1982, U.S. EPA 1991).
As presented earlier, insufficient toxicological response was observed in laboratory
exposures to derive exposure-response models for whole sediments. However, these results
and those obtained during Phases I and II (Munns et al. 1991, 1993) suggest ecological risks
associated with whole sediment exposures to be small. Efforts therefore were directed
towards development of TU-based models for seep water and sediment extracts only. For
this analysis, TUs were calculated using established federal marine Water Quality Criteria
(WQC) and the concentrations of contaminants measured in each exposure medium dilution
series.
EPA has suggested that appropriate biological benchmarks be used for each specific
endpoint (U.S. EPA 1991). This means that acute benchmarks be used with acute endpoints
and chronic benchmarks be used with
chronic endpoints. In this analysis,
however, chronic WQC were used in
all modeling efforts to yield
conservative predictions of ecological
response. WQC available (U.S. EPA
1987) for this exercise are indicated
in Table 6. (The criterion shown for
copper is actually the acute value, as
no chronic value is given). The
general paucity of available marine
WQC can be viewed as a limitation to
this analysis, although several of the contaminants listed in Table 6 were identified as
potential contaminants of concern in Munns et al. (1991). The nature of chemical data
available for TU calculations also restricted model development in that organic contaminants
for which WQC are available were generally not detected in seep water media, whereas the
method employed to develop sediment extracts isolated organic contaminants only. (In
retrospect, a sediment elutriate procedure may have been more appropriate for isolating
58
Table 6. Federal marine chronic Water Qualify
Criteria used in ETU exposure-response model
development.
Chronic
Contaminant WQC Oig/L)
cadmium 9.3
zinc 86
copper 2.9
lead 8.5
nickel 8.3
PCB 0.03
Chlordane 0.004
DDT 0.001
-------
sediment contaminants.) Despite these limitations, efforts to develop TU-based models were
generally successful. Contaminant-specific TUs for the two undiluted exposure media for
which exposure-response models were developed are shown in Table 7.
Table 7. Contaminant-specific TUs for undiluted exposure media used in model development.
Exposure TU*
Medium Cd Zn Cu Pb Ni PCB Chlordane DDT ETU
seep water 1.5 2.9 15.2' 0.5 2.3 nd2 nd nd 22.3
sediment extract np2 np np np np 2.2 0.2 0.9 3.3
f Table entries are the number of toxic units, for each contaminant, quantified in the exposure medium, TUs
for whole sediment were not calculated (see text for explanation). TUs for sediment extract are based on
an extract concentration of 0.25%, the highest concentration used in sediment extract assays.
1 Nominal values were used for copper concentrations in seep water in calculating contaminant-specific TUs,
because measured concentrations above the 46 % exposure media concentration did not agree with expected
nominal levels (see Appendix A).
2 nd = not detected
np = not present in medium
The TU models were constructed in a manner similar to those for whole exposure
media, using nonlinear regression techniques, with the metric ETU replacing whole media
concentration as the exposure term. Not surprisingly (as ITU is a direct function of the
individual contaminant concentrations measured in each dilution series), fair success was
achieved in obtaining models which described response data reasonably well (Table 8,
Figures 21-27).
To some extent, both the whole-media and ETU exposure-response models developed
in this study are assay-dependent: model parameter estimates depend not only upon the
general responses of endpoints to exposure levels, but also upon the exact values measured in
replicates of each exposure treatment. The replicate measures reflect some degree of
inherent variability within the test population, as well as error (uncertainty) in part
59
-------
Arbacia
FERTILIZATION SUCCESS
SUM TU IN SEEP
Figure 21. ETU exposure-response model for fertilization response of Arbacia to seep water exposure.
60
-------
Arbacia
48-HOUR LARVAL DEVELOPMENT
SUM TU IN SEEP
Figure 22. ETU exposure-response model for larval development response of Arbacia to seep water exposure,
61
-------
60 t
Champia
Sexual Reproduction
<
S 40
LLl
U-
20 i
1 M "1
10 15
SUM TU IN SEEP
Figure 23. 57TU exposure-response model for sexual reproduction response of Champia to seep water exposure.
62
-------
Mulinia
LARVAL MORTALITY
SUM TU IN SEEP
Figure 24, ETU exposure-response model for larval survival response of Mulinia to seep water exposure.
-------
Arbacia
FERTILIZATION SUCCESS
SUM TU IN SEDIMENT EXTRACT
Figure 25. ETU exposure-response model for fertilization response of Arbacia to sediment extract exposure.
64
-------
Arbacia
48-HOUR LARVAL DEVELOPMENT
SUM TU IN SEDIMENT EXTRACT
Figure 26. ETU exposure-response model for larval mortality and development response of Arbacia to sediment extract
exposure.
65
-------
Photobacterium
MICROTOX
LU
O
Z
LU
o
w
ui
100
80
60
40
20
0.001
0.01 0.1 1 10
SUM TU IN SEDIMENT EXTRACT
100
Figure 27, ETU exposure-response model for bioluminescence response of Photobacterium to sediment extract exposure.
66
-------
Table 8. ETU exposure-response model parameter estimates and coefficients of determination.
Exposure
Medium Species/Endpoint EC,,* a Raซ'a/fertilization
0.16
0.13
97.0
98.7
Arbaciataarmal development
0.73
0.09
89.6
ArbacialmoTtality
2.55
0.05
88.7
99.9
F'iotabacteriumlmortality
0.049
1.0
100
98.7
In units of ETU.
k Cystocarps/female for Champia, percent for all other.
determined by experimental design. Thus it is reasonable to question the adequacy of the
models to describe or predict the responses of test populations (of the same species) different
from those used the exposure-response assays, or the responses of organisms in Allen
Harbor. To evaluate model validity, toxicity data were collated from Phases I and II (Munns
et al. 1991, 1993) which could be compared with the predictions of the ETU models.
Obviously, to be useful in this exercise, information concerning both exposure and the
response of the appropriate species/endpoint needed to be available.
Of all biological measures obtained over the course of this project, only Arbacia
fertilization data from water exposures met these criteria. Twelve exposure-response data
pairs had been obtained as part of Phase n exercises (Munns et al. 1993). These were
associated with both landfill seep (LANDN, LANDM, and LANDS), and runoff (SN, WC,
NC) stations in the summer and fall of 1990. Validation of the Arbacia fertilization model
proceeded by calculating ETU for each sample, and using the model in Table 8 (see Figure
67
-------
100 1
~
LU
N
dE
HI
11
80
60 -
Q
LL1
>
QC
ง HI
W
CO
o
40 -
20 -
OBSERVED = PREDICTED
0
o
I I 1 1
40 60
PREDICTED % FERTILIZED
r-
80
0
20
100
Figure 28. Comparison of predicted and observed Arbacia fertilization success.
68
-------
21) to predict an associated fertilization response. These predictions were then compared
with the response actually observed during Phase 13. Of the 12 predictions, 10 (83%)
accurately reflected the Phase n data, one overpredicted response, and one underpredicted
response (Figure 28; five data points are obscured in this figure due to overlap with other
points). The latter of the two mispredictions is the more troubling (risk managers likely
would prefer to error on the conservative side), but is not surprising given the paucity of
chemistry data utilked in model development. Thus the seep water-Arbacia fertilization
model, at a minimum, appears to be useful in predicting risk within Allen Harbor. It should
be noted, however, that insufficient data were available to adequately evaluate the model
within the range of predicted partial (less than 90% but greater than 10%) effects.
Assuming the ETU exposure-response models to be reasonable predictors of
biological response, the 8 models developed here should be valuable in quantifying risks
directly associated with the Allen Harbor landfill. In this form, they are useful to
environmental managers in at least two ways. With knowledge of the contaminant makeup
of any particular site within the harbor, they can be used to establish associated ecological
risk by summing the TU equivalents of chemical stressors, and estimating the probability of a
particular ecological response using the appropriate model. Further, levels of remediation
required to reduce risk (if deemed to be too high at a particular site) to some acceptable level
can be established by restating the models in terms of an acceptable level of response and
solving for the exposure term. (Defining acceptable levels of risk is beyond the scope of this
project, and is best left up to the environmental and resource managers involved with the
site.) Reductions in risk could theoretically be achieved by a remediation plan which
decreases the concentrations of contaminants contributing the majority of risk (i.e., those
associated with the largest contributions to ETU).
Risk Quantified as Joint Probabilities
The ETU exposure-response models provide a means of estimating the degree of
biological response expected given an understanding of contamination at a given site (say in
immediate association with a landfill seep). Perhaps a more holistic approach to quantifying
ecological risk in Allen Harbor would be to evaluate environmental conditions in the harbor
69
-------
in its entirety, rather than on a site-by-site basis. To accomplish this, a probabilistic
approach was applied which utilized the statistical distributions of exposure and expected
biological effect to provide a direct, quantitative measure of ecological risk.
Conceptually, this procedure involves estimation of the joint probabilities of exposure
and effects distributions, as illustrated in Figure 29. The graphic in the upper left-hand side
of this figure represents the distribution of stressor concentration measured or modeled in
space or time. Assumed to be log-normal, this distribution can be characterized by its
associated geometric mean and standard deviation, and describes the probability of observing
any particular stressor concentration within the bounds (again, spatial or temporal) of the risk
assessment. The upper right-hand graphic illustrates an exposure-response model like those
developed here. The ECm and a associated with this model also describe a statistical
distribution, this time of the response thresholds to stressor concentration of the endpoint it
models (as described previously). In moving from this depiction to the bottom graphic, this
model, or cumulative distribution function, has been restated as a probability density function
to correspond in form with the exposure distribution.
The area of overlap of these two distributions, shown at the bottom of Figure 29,
defines the degree of risk expected within the system. It is within this region of existing
stressor concentration thai ecological effects are expected to occur. The probability that they
will occur depends upon the probability of experiencing an exposure high enough to elicit an
ecological response, or the joint probability of exposure and effects. The degree of overlap
is therefore a quantitative measure of ecological risk.
In this type of analysis, overlap can be calculated directly by solving for the
intersection area of the two distributions, or can be estimated using Monte Carlo simulation
techniques. In this latter approach, the two distributions are artificially sampled (using a
computer) in pair-wise fashion, and the value of the resulting exposure concentration is
compared with that of the sampled effects threshold concentration. A biological response is
expected to occur (and is scored as such) if the former is greater or equal to the latter. At
the end of several such sampling events or iterations, the proportion of total iterations during
which the biological threshold concentration was exceeded is calculated. Following standard
sampling theory, this proportion estimates the probability of biological response, thus
70
-------
RISK AS A JOINT PROBABILITY
DISTRIBUTION OF EXPOSURES
EXPOSURE-RESPONSE MODEL
LOtXCOMCEHTKATION)
LOO( CONCENTRATION)
m
m
g
Qu
EXPOSURE
i EFFECTS
k
..lllllrra-
LOG(CONCENTRATION)
Figure 29. Characterization of ecological risk as a joint probability.
71
-------
providing a probabilistic estimate of risk. As described, this method can be used to estimate
probabilities associated with observing an ecological response, or with exceedence of some
degree of response, such as an LCM or EC*,.
To estimate risks to Allen Harbor pelagic and benthic systems, Monte Carlo
simulations were conducted employing exposure data obtained throughout all three phases of
this study, and the ETU exposure-response models derived for seep water and sediment
extracts. Analyses of risks associated with these two systems differed somewhat due to the
availability of data, and are described separately below.
Risk to pelagic system -- Due to reasons given elsewhere (Munns et al. 1991, 1993), little
emphasis was placed in this study on characterizing pelagic exposure conditions through
direct measurement of water column chemistry. However, chemistry data were available for
water entering the harbor through landfill seeps and from runoff sources. These two data
sets were used separately to quantify upper-bound risks (because they represent exposure
media undiluted by harbor water) associated with each source.
Exposure distributions were derived for each source independently by calculating ETU
for each sample (17 seep samples, 6 runoff samples), and determining the geometric means
and standard deviations representing each source. Monte Carlo simulations were conducted
with Crystal Ball* software using these distributions and the appropriate ITU exposure-
response models, as described above. Five simulations, each involving 1,000 sampling
iterations, were conducted to estimate joint probabilities. The five estimates of overlap were
then averaged to yield an estimate of risk to each biological endpoint. Standard errors of
these means were also calculated as a method of describing (minimal) uncertainty associated
with each risk estimate.
Risks from landfill seep waters to the four endpoints ranged from 0.24 for Arbacia
fertilization, to as high as 0.69 for Champia reproduction (Table 9). With the exception of
Arbacia fertilization, risks to measured endpoints associated with landfill seeps were
statistically higher than those from runoff sources (t-test, P<0.05). Although these estimates
indicate the potential for negative ecological impact associated with both landfill and runoff
sources, actual risks to the harbor pelagic system would be expected to be lower than these
72
-------
Table 9. Mean probabilities of maximum risk to pelagic
systems.
Endpoint
Seep
Runoff
estimates suggest, because both
seep and runoff water would be
diluted substantially upon mixing.
Detailed water column
measurements and/or transport
studies would be necessary to
fully characterize this risk.
Because these simulations
were conducted to estimate total
overlap between the exposure and
effects distributions, the resulting
risk estimates should be
interpreted as describing the
probability that any negative
effect will occur. That is, in the case of Arbacia fertilization, there is a 24% chance that
reduced fertilization success would be observed for sea urchin gametes in immediate
proximity to the landfill. The magnitudes of such effects were not evaluated in this analysis.
Arbacia fertilization
0.243
0.252
(0.004)*
(0.004)
Arbacia development
0.644
0.688
(0.012)
(0.005)
Champia reproduction
0.688
0.498
(0.011)
(0.003)
Mulinia survival
0.529
0.403
(0.007)
(0.006)
Standard error of the mean.
Risk to beiuhic system ~ While full effects distributions could not be generated due to the
lack of response observed in whole sediment assays (and thus an incomplete exposure-
response description), it is straightforward to conclude minimal risks to Allen Harbor benthic
system, at least as evaluated by Ampelisca and Mulinia responses, without actual calculation
of joint probabilities. Such a conclusion is supported by Ampelisca assays results and general
descriptions of the conditions of benthic populations obtained during the initial phase of this
study (Munns et al. 1991). However, the sediment extract models provided an additional
means of evaluating benthic risk. These models were used to calculate both upper-bound,
maximum probabilities of risk, as well as more likely estimates which incorporated an
understanding of contaminant bioavailability.
The extraction procedure was employed to isolate the entire quantity of nonionic
73
-------
organic contaminants associated with the whole sediment. It was thus possible to relate
contaminant concentrations in the whole sediment to those measured in the extract, and to
use this relationship to derive expectations of the concentrations of contaminants in extract
exposure media from harbor samples, had they been extracted. These sediment extract
equivalents, quantified in terms of ETU, represent the maximum exposure potential of harbor
samples assuming all contaminants to be bioavailable. Joint probability calculations
employing sediment extract equivalents to describe the exposure distribution therefore yielded
upper-bound probabilities of risk.
Sediment extract equivalents were calculated for 42 sediment samples collected in
Allen Harbor throughout the course of this project (5 during Phase I, 37 during Phase II) as
follows. Concentrations of relevant contaminants were normalized to those of the LANDM
whole sediment sample used to generate the sediment extract exposure medium. These ratios
were then multiplied by the contaminant-specific TUs determined for the LANDM extract
sample to produce a TU estimate for each contaminant in each field sample. TUs were
summed, and geometric means and standard deviations calculated to describe the statistical
distribution of exposure. This distribution was then used in Monte Carlo simulations with
sediment extract ETU exposure-response models. As described for simulations involving
waters, five simulations involving 1,000 iterations each were conducted to characterize risk
to modeled endpoints.
As indicated in Table 10, Allen Harbor sediments have the substantial potential to
impact the four biological endpoints. Probabilities of maximum risk to Arbacia fertilization
and development were estimated as unity, and maximum risks to Arbacia survival and
Photobacterium survival were greater than 80%. These predictions are borne out to some
extent by Arbacia fertilization and development endpoint data obtained during Phase I: Allen
Harbor sediment extract concentrations as low as 0.2 and 0.05% were observed to impact
these endpoints (Munns el al, 1991). Similarly, Allen Harbor sediment interstitial waters
also impacted these endpoints at concentrations as low as 12.5% (Munns et al. 1991).
To aid in the interpretation of maximum risks calculated for Allen Harbor sediments,
a similar analysis was conducted using sediment data obtained from Mount View (MV),
74
-------
identified as a mid-Narragansett
Bay reference station in Phases I
and n (Munns et al. 1991,
1993). Results similar to those
for Allen Harbor were obtained
with one exception (Table 10):
risk to Arbacia survival was
calculated to be 0 at MV.
Lower contaminant levels and
the steepness of the exposure-
response curve for this endpoint
likely contributed to this result.
The above calculations
assume all contaminants to be
available to benthic organisms. Typically, however, some fraction of organic contaminants
associated with sediments is bound in dynamic equilibrium to the organic carbon of those
sediments, and is therefore unavailable to biota. Evidence suggests that for nonionic
organics, exposure conditions are better represented by concentrations of contaminants
measured in pore water than by bulk sediment chemistry (see Di Toro et al. 1991). Since
the sediment extract equivalents employed above reflect harbor sample bulk chemistry, a
more reasonable approach to evaluating risks would incorporate the bioavailability of organic
compounds in estimating exposure distributions. Following the equilibrium partitioning
approach described in Di Toro el al (1991), concentrations of organic contaminants expected
in harbor sediment pore water were calculated using the relationship;
C< = C./^xKJ
where CA is the concentration (jig/L) in pore water, C, is the bulk concentration (ng/g) in the
sediment, fx is the fraction of organic carbon in the sediment (see Appendix G), and is
the partition coefficient for sediment organic carbon, expressed in terms of (ng chemical/g
75
Table 10. Mean probabilities of maximum risk to benthic
systems.
Endpoint
AH
MV
Arbacia fertilization
1.0
(0.0)*
1.0
(0.0)
Arbacia development
1.0
(0.012)
1.0
(0.005)
Arbacia survival
0.814
(0.009)
0.0
(0.0)
Phosobacterium survival
0.844
(0.007)
0.769
(0.003)
* Standard error of the mean.
-------
organic carbon)/(^g chemical/L pore water). When is not known, it can be estimated
from the chemical's octanol-water partition coefficient (K^) using the regression relationship
(Di Toro ei al. 1991):
= an tilog{0.00028 + 0.983Pog(Kow)]}.
K^s (from MacKay et al. 1992 and U.S. EPA 1984) and K^s for the nonionic organic
contaminants used in predicting pore water concentrations are given in Table 11.
Using the pore water concentrations predicted in this manner, TUs were calculated,
summed, and geometric means and standard deviations calculated to describe the new
statistical distribution of exposure from AH sediments. This distribution was then used in
Monte Carlo simulations with sediment extract ETU exposure-response models. As with
earlier risk estimation procedures, five simulations involving 1,000 iterations each were
conducted to characterize risk to modeled endpoints.
With contaminant bioavailability taken into account, estimates of risks to benthic
systems in Allen Harbor (Table 12) were substantially reduced from those calculated using
sediment extract equivalents, particularly with respect to Arbacia development and survival.
Some degree of risk was still indicated for Arbacia fertilization and the Microtox endpoint,
however.
Thus, the discrepancy between the degrees of risk concluded from evaluations
involving whole sediment and sediment extracts most likely reflects differences in the
availability of organic contaminants in the two media. The extraction procedure is, by
design, highly efficient at isolating
Table 11. Partition coefficients used in pore water
organic contaminants from binding mntammant concentration calculations.
matrices in the sediment. Such
factors as organic carbon reduce
the availability of organic
Contaminant
PCB (as Aroclor 1254) 6.47 6.36
Chlordane 5.54 5.45
DDT 5.75 5.65
76
-------
contaminants in natural sediments,
such that bulk chemistry
measurements tend to overestimate
actual exposure conditions. These
contaminants are bioavailable in the
sediment extract, however. Thus
an analysis of sediment risks may
be overly conservative when
involving sediment extracts. It also
is possible, however, that
differences in sensitivity exist
between endpoints of species used
in whole sediment exposures and
those of species involved in
sediment extract model
development.
A final word concerning uncertainties associated with the estimates of risk developed
here is cogent. While attempts were made to quantify the uncertainties of the final risk
calculations themselves through performance of multiple simulations, little regard was given
to uncertainties introduced through exposure media sampling error, assay performance, and
chemical analysis. The lack of quantification of these sources of error (and others) weakens
the confidence with which conclusions can be drawn regarding this ecological risks to Allen
Harbor. It also should be noted that the effects endpoints evaluated in this study generally
are short-term in nature; the effects of some contaminants in the harbor may require longer
time periods to manifest. Despite these caveats, the probabilities estimated above should
serve to identify the degree to which ecological systems are at risk from landfill
contaminants. A long-term monitoring program would provide information to verify this risk
assessment and to assist in management of Allen Harbor.
Table 12. Mean probabilities of risk to beothic systems
incorporating contaminant bioavailability.
Endpoint AH
Arbacia fertilization .755
(0.006)*
Arbacia development 0.317
(0.009)
Arbacia survival 0.083
(0.007)
Photobacterium survival 0.550
(0.010)
* Standard error of the mean.
77
-------
RISK OF NEOPLASTIC DISEASE DEVELOPMENT IN MYA
Although a possible relationship between NCBC Davisville and hematopoietic
neoplasia in Mya arenaria has been a focus of attention throughout the entire Risk
Assessment Pilot Study, efforts undertaken in all three phases have failed to conclusively link
disease etiology with chemical contamination at the facility. In a final evaluation of the risk
of neoplastic disease development in relation to sediment contamination, attempts were made
to correlate rates of Hn in Mya observed by histocytologieal techniques during the Phase II
survey with corresponding exposure concentrations measured during Phases I, n, and HI.
Chemical information was available for nine stations visited during the survey: LANDM,
LANDN, AH10, PR, SN, NC, CC, FDA, and MP (see Munns et al. 1991, 1993 for original
station descriptions).
These analyses indicated no relationship between the incidence of Hn and sediment
contamination. As an illustration of this, 23% of the clams examined at FDA during Phase
II (Munns et al. 1993) were afflicted with Hn while contaminant levels were between 10 and
100 times lower than at SN, a station at which no Hn was diagnosed. Similar efforts by
Brown (1980) and Mix (1986), among others, also have failed to link Mya neoplasia with
sediment contamination. Additionally, Appeldoorn et al. (1984) were unable to relate Hn to
sediment contamination in laboratory exposures. Recently, Chang et al. (1993) conclusively
demonstrated Hn to be caused by a retrovirus. The general lack of observed association
between environmental contamination and disease etiology in laboratory experiments and
field studies strongly suggests the risk of Mya hematopoietic neoplasia development relative
to chemical contamination at NCBC Davisville to be minimal.
78
-------
Beferences
Alabaster, I. and R. Lloyd (Editors). 1982. Water Quality Criteria for Fish. Butterworths,
London, England.
Appeldoorn, R.S., C.W. Brown, P.W. Chang, K.R. Cooper, E. Lorda, S.B. Saila, H.
Walker, R.E. Wolke, 1984. Field and laboratory studies to define the occurrence of
neoplasia in the soft-shell clam, Mya arenaria. American Petroleum Institute
Publication No. 4345, Washington, D.C.
ASTM. 1988. Test methods for total and dissolved carbon dioxide in water. ASTM D 513
Method G. Vol 11.01 on Water. American Society for Testing and Materials,
Philadelphia, PA.
Barnthouse, L.W., G.W. Suter, A.E. Rosen, and J.J. Beauchamp. 1987. Estimating
responses of fish populations to toxic contaminants. Environmental Toxicology and
Chemistry 6: 811-824.
Beckmann, N.S. 1989. Comparative study of phagocytosis in diseased and normal
hemocytes of the bivalve mollusc, Mya arenaria. Master's Thesis. Northeastern
University, Boston, MA.
Boothman, W. and A. Helmstetter. 1993. Vertical and seasonal variability of acid volatile
sulfides in marine sediments. EMAP Research Project, Final Report. ERLN
Contribution No. 1389
Brown, R.S. 1980. The value of the multidisciplinary approach to research on marine
pollution effects as evidenced in a three-year study to determine the etiology and
pathogenesis of neoplasia in the soft-shell clam, Mya arenaria. Rapports Et Proces-
verbaux Des Reunions (Conseil Permanent International Pour L'Exploration De La
Mer) 179: 125-128,
Brown, R.S., R.E. Wolke and S.B. Saila. 1976. A preliminary report on neoplasia in feral
populations of the soft-shell clam, Mya arenaria, prevalence, histopathology and
diagnosis. Proceedings of the International Colloquium on Invertebrate Pathology J:
151-158.
Brown, R.S., R.E. Wolke, S.B. Saila and C. Brown. 1977. Prevalence of neoplasia in 10
New England populations of the soft-shelled clam (Mya arenaria). Annals of the
New York Academy of Sciences 298: 522-534.
Bruce, R.D. and D.J. Versteeg. 1992. A statistical procedure for modeling continuous
79
-------
toxicity data. Environmental Toxicology and Chemistry 11: 1485-1494.
Burgess, R. and G. Morrison. In preparation. A seven-day rapid exposure sediment toxicity
test with the marine bivalve Mulinia lateralis.
Charters, D.W. 1990. The use of toxicity tests in environmental assessments at Superfund
sites. Paper presented at the 11th Annual Meeting of the Society of Environmental
Toxicology and Chemistry, Arlington, VA.
Cooper, K.R., R.S. Brown and P.W. Chang. 1982a. The course and mortality of a
hematopoietic neoplasm in the soft-shell clam, Mya arenaria. Journal of Invertebrate
Pathology 39: 149-157.
Cooper, K.R., R.S. Brown and P.W. Chang. 1982b. Accuracy of blood cytological
screening techniques for the diagnosis of a possible hematopoietic neoplasm in the
bivalve mollusc, Mya arenaria. Journal of Invertebrate Pathology 39: 281-289.
Di Toro, D.M., J.D. Mahony, D.J. Hansen, K.J. Scott, M.B. Hicks, S.M. Mayr, and M.S.
Redmond. 1990. The toxicity of cadmium in sediments: The role of acid volatile
sulfide. Environmental Toxicology and Chemistry 9: 1487-1502.
Di Toro, D.M., C.S. Zarba, D.J. Hansen, W.J. Berry, R.C. Swartz, C.E. Cowan, S.P.
Pavlou, H.E. Allen, N.A. Thomas, and P.R. Paquin. 1991. Technical basis for
establishing sediment quality criteria for nonionic organic chemicals using equilibrium
partitioning. Environmental Toxicology and Chemistry 10: 1541-1583.
ERLN. 1991. ERLN Standard Operating Procedures Manual. U.S. EPA Environmental
Research Laboratory, Narragansett, RI.
ERLN. 1992. Quality Assurance Program Plan for the Environmental Research Laboratory
Narragansett and Newport. U.S. EPA Environmental Research Laboratory,
Narragansett, RI.
ERLN/NOSC. 1991. Work/Quality Assurance Plan for Marine Ecological Risk Assessment
Pilot Study. U.S. EPA Environmental Research Laboratory, Narragansett, RI.
Farley, C.A. 1969. Probable neoplastic disease of the hemopoietic systems in oysters
{Crassostrea virginica and Crassostrea gigas). National Cancer Institute Monograph
31: 541-555.
Farley, C.A. 1989. Selected aspects of neoplastic progression in mollusks, pp. 24-31. In:
H.E. Kaiser (Ed.), Comparative Aspects of Tumor Development. Vol. 5. Cancer
Growth and Progression. Kluwer Academic Publishing, The Netherlands.
80
-------
Farley, C.A., S.V, Otto and C.L. Reinisch. 1986, New occurrence of epizootic sarcoma in
Chesapeake Bay soft-shell clams (Mya arenaria). Fishery Bulletin 84: 851-857.
Johnston, R.K. 1993. Acid volatile sulfides and simultaneously extracted copper, lead, and
zinc in sediment of Sinclair Inlet, Washington. NCCOSC Research, Development,
Test and Evaluation Division Technical Report 1552, San Diego, CA,
Mackay, D., S. Su, and M. Li. 1992. Illustrated Handbook of Physical Properties of
Organic Chemicals, Vol. 1. Lewis Publishers, Boca Raton, FL, 326 pp.
Mix, M.C. 1979. Chemical carcinogens in bivalve mollusks from Oregon estuaries. EPA-
600/3-79-34, U.S. Environmental Protection Agency, Environmental Research
Laboratory, Gulf Breeze, FL.
Mix, M.C. 1986. Cancerous diseases in aquatic animals and their association with
environmental pollutants: a critical review. Marine Environmental Research 20. 1-
141.
Mueller, C., W.R. Munns, Jr., D.J. Cobb, E.A. Petrocelli, G.G. Pesch, D.M. Burdick,
F.T. Short and R,K. Johnston (Editors). 1992. Standard Operating Procedures and
Field Methods Used for Conducting Ecological Risk Assessment Case Studies at;
Naval Construction Battalion Center Davisville, RI and Naval Shipyard Portsmouth,
Kittery, ME. Naval Command, Control and Ocean Surveillance Center Technical
Document 2296, Research, Development, Test and Evaluation Division, San Diego,
CA.
Munns, W.R., Jr. and R. Comeleo. 1991. Interim report on evaluation of marine
ecosystems associated with Superfund sites. Report prepared for the EPA Office of
Emergency and Remedial Response, ERLN Contribution No. 8316A.
Munns, W.R., Jr., C.A. Mueller, D.J. Cobb, T.R. Gleason, G.G. Pesch and R.K. Johnston.
1991. Marine Ecological Risk Assessment at Naval Construction Battalion Center,
Davisville, Rhode Island Phase I. Naval Ocean Systems Center Technical Report
1437, Naval Ocean Systems Center, San Diego, CA, 129 pp. + Appendices.
Munns, W.R., Jr., C.A. Mueller, R. Comeleo, D.J. Cobb, S. Anderson, G.G. Pesch, W.G.
Nelson, and R.K. Johnston. 1993. Marine Ecological Risk Assessment at Naval
Construction Battalion Center, Davisville, Rhode Island ~ Phase II. Report prepared
for the Naval Command, Control and Ocean Surveillance Center, ERLN Contribution
No. 1427.
Norton, S., M. McVey, J. Colt, J. Durda, and R. Hegner. 1988. Review of ecological risk
assessment methods. EPA/230-10-88-041, U.S. Environmental Protection Agency
Office of Policy, Planning and Evaluation, Washington, DC.
81
-------
Oprandy, J.J., P.W. Chang, A.D. Provost, K.R. Cooper, R.S. Brown, and V.J. Yates.
1981. Isolation of a viral agent causing hematopoietic neoplasia in the soft-shell clam,
Mya arenaria. Journal Invertebrate Pathology 38: 45-51.
Peters, E.G. 1988. Recent investigations on the disseminated sarcomas of marine bivalve
molluscs. American Fisheries Society Special Publications 18: 74-92.
Reinisch, C.L., A.M. Charles and A.M. Stone. 1983. Unique antigens on neoplastic cells
of the softshell clam, Mya arenaria. Developmental and Comparative Immunology 7:
33-39.
Rosen, J.S., D. Sheehan and R. Petrocelli. 1988. Data management plan [for the] Risk
Assessment Pilot Study, Naval Construction Battalion Center, Davisville, R.I. Report
prepared for the U.S. Environmental Protection Agency Environmental Research
Laboratory, Narragansett, RI. Computer Sciences Corporation, Narragansett, RI.
SAS Institute. 1989. SAS*/STAT User's Guide, Version 6 Edition. SAS Institute, Cary,
NC.
Smolowitz, R.M. and C.L. Reinisch. 1986. Indirect peroxidase staining using monoclonal
antibodies specific for Mya arenaria neoplastic cells. Journal of Invertebrate
Pathology 48: 139-145.
U.S. EPA. 1984. AQUIRE: Aquatic Information Retrieval Toxicity Data Base. EPA-
600/8-84/021, U.S. Environmental Protection Agency, Environmental Research
Laboratory, Duluth, MN.
U.S. EPA. 1987. Quality Criteria for Water. EPA/505/2-90-001, U.S. Environmental
Protection Agency, Office of Water Regulations and Standards, Washington, DC.
U.S. EPA. 1991. Technical Support Document for Water Quality-based Toxics Control.
EPA/505/2-90-001, U.S. Environmental Protection Agency, Office of Water,
Washington, DC.
Walker, H.A., E. Lorda and S.B. Saila. 1981. A comparison of the incidence of five
pathological conditions in soft-shell clams, Mya arenaria, from environments with
various pollution histories. Marine Environmental Research 5: 109-123,
82
-------
APPENDIX A
TRACE METAL CONCENTRATIONS
A- 1
-------
TABLE A-l. TRACE METAL CONCENTRATIONS QiglL) IN SEEP WATERS
SAMPNUM
STATION
CONC(5S)
Cu
Za
Cr
Pb
Ni
Cd
Mn
Fe
As
798904
BRINE
0
ND
ND
5
ND
ND
ND
I
3
ND
798909
LANDM
5.8
3
29
ND
ND
1
ND
1
19
ND
798908
LANDM
11.5
6
62
ND
ND
2
ND
9
124
ND
798907
LANDM
23
11
90
ND
3
5
2
18
303
ND
798906
LANDM
46
16
152
ND
4
9
5
26
563
ND
798905
LANDM
92
16
249
3
4
19
14
13
252
ND
798902
LANDM
100
16
322
ND
5
28
21
59
ND
ND
TABLE A-2. TRACE METAL CONCENTRATIONS (#ซg/g) IN SEDIMENTS
SAMPNUM
STATION
CONC(%)
Cu
Za
Cr
Pb
Ni
Cd
Mn
Fe
As
798900
AH
100
455
1110
109
515
58.5
6.52
186
49600
7.74
798910
PR
100
2.51
17.6
5.31
2.76
3.42
0.06
45.8
4400
0.67
798912
POTO
100
29.4
98.5
24.6
20.5
13.4
0.89
128
18200
4.83
798939
CUS
100
53.2
155
59.9
34.1
26.2
0.05
499
27300
5.21
798915
LANDM
12.5
86.5
219
31.7
73.9
18.8
1.59
134
23800
5.29
798914
LANDM
25
120
349
40.0
118
24.4
2.07
145
25500
6.20
798913
LANDM
50
196
489
50.6
190
31.7
3.21
160
30800
6.15
798903
LANDM
100
290
708
63.9
275
39.4
4.26
169
37000
8.25
TABLE A-3. TRACE METAL CONCENTRATIONS Oig/g DRY WT) IN Mya arenaria
SAMPNUM
STATION
Cu
Za
Cr
Pb
Ni
Cd
Mn
Fc
As
798901
SALTPOND
14.5
64.8
0.690
3.70
0.00
0.130
54.9
445
4.87
798918
AH
35.2
97.8
3.72
61.4
1.99
0.61
26.0
18700
12.3
798922
PR
11.0
43.1
2.10
5.39
0.93
0.49
21.6
1350
5.70
A-2
-------
APPENDIX B
PESTICIDE CONCENTRATIONS
B- 1
-------
TABLE B-l, PESTICIDE CONCENTRATIONS (pglL) IN SEEP WATERS
alpba-
gamma-
alpba-
gamma-
SAMPNUM
STATION
CONC(S)
HCB
BHC
BHC
Chlordane
Chlordane
p,p'-DDE
p,p"-DDD
p.p'-DDT
798904
BRINE
0
ND
ND
ND
ND
ND
ND
ND
ND
798909
LANDM
5.8
NM
NM
NM
NM
NM
NM
NM
NM
798908
LANDM
11.5
NM
NM
NM
NM
NM
NM
NM
NM
798907
LANDM
23
NM
NM
NM
NM
NM
NM
NM
NM
798906
LANDM
46
NM
NM
NM
NM
NM
NM
NM
NM
798905
LANDM
92
NM
NM
NM
NM
NM
NM
NM
NM
798902
LANDM
100
ND
ND
ND
ND
ND
ND
ND
ND
TABLE B-2, PESTICIDE CONCENTRATIONS (ng/g DRY WT) IN SEDIMENTS
SAMPNUM
STATION
CONC(%)
HCB
alpha*
BHC
gamma-
BHC
alpha-
Cblordane
gamma-
Chlordaoe
p,p'-DDE
p.p'-DDD
p,p'-DDT
798900
AH
100
3.80
0.33
0.24
2.10
3.55
8.03
58.5
43.6
798910
PR
100
0.05
0.02
0.06
0.06
0.05
0.05
0.17
0.08
798912
POTO
100
0.04
0.04
0.13
0.41
0.56
0.39
0.91
0.30
798939
CLIS
100
0.07
0.16
0.06
0.27
1.00
0.89
1.19
0.49
798915
LANDM
12.5
0.23
0.06
0.21
0.82
1.22
1.75
4.74
1.27
798914
LANDM
25
0.35
0.07
0.35
1.09
1.59
2.36
8.12
2.21
798913
LANDM
50
0.56
0.10
0.22
1.14
1.81
2.61
11.9
3.17
798903
LANDM
100
0.72
0.15
0.53
2.13
3.31
4.08
26.4
14.1
B-2
-------
TABLE B-3. PESTICIDE CONCENTRATIONS (ng/g DRY WT) IN Mya arenaria
alpha-
gamma-
alpha-
gamma-
SAMPNUM
STATION
HCB
BHC
BHC
Chlordane
Chlordane
p.p'-DDE p.p'-DDD
p.p'-DDT
798901
-a ALT POND
0.54
0.S5
0.55
0.66
0.74
1.18 0.75
0.64
798918
AH
1,03
0.46
ND
0.73
1.14
1.35 2.18
1.23
798922
PR
0.66
0.71
0.64
0.44
1.39
1.46 0.88
0.81
B-3
-------
APPENDIX C
nAT \7/""*TTT ATiTXT A THA DTTItinXTVT r^AXTPEXTCn AAXTACXTTTI A TTAXTP
POLYCHLORINATED JBIPHENYL CONGENER CONCENTRATIONS
C- 1
-------
TABLE C-l. PCB CONGENER CONCENTRATIONS (/tg/L) IN SEEP WATERS
SAMPNUM
STATION
CONC<*) CB052
CB047
CB101
CB151
CBU8
CB153
CB138
CB128
CB180
CB195
CB194
CB206
CB209
798904
BRINE
0
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
798909
LANDM
5,8
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
798908
LAN DM
11.5
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
798907
LANDM
23
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
798906
LANDM
46
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
798905
LANDM
92
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
798902
LANDM
100
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
TABLE C-2. PCB CONGENER CONCENTRATIONS (ng/g DRY WT) IN SEDIMENTS
SAMPNUM
STATION
CONC(%)
CB052
CB047
CB101
CB151
CB118
CB153
CB138
CB128
CB180
CB195
CB194
CB206
CB209
798900
AH
100
11.8
2.44
23.1
8.74
30.2
34.7
42.0
2.05
26.1
2.60
5.74
3.52
2.40
798910
PR
100
0.35
0.19
0.22
ND
0.20
0.20
0.20
ND
0.20
ND
ND
ND
ND
798912
POTO
100
0.76
0.30
0.99
0.20
1.30
1.00
1.30
ND
0.60
0.20
0.10
0.45
0.40
798939
CLIS
100
2.23
1.43
3.60
1.20
4.37
5.83
4.50
0.20
2.53
0.50
0.87
1.33
1.37
798915
LANDM
12.5
2.57
0.90
4.18
1.80
5.10
7.00
7.60
ND
6.00
0.50
1.30
1.30
1.00
798914
LANDM
25
3.77
1.40
6.62
2.80
7.70
11.0
11.6
ND
9.10
0.80
2.10
1.70
1.20
798913
LANDM
50
4.39
1.40
8.93
4.50
11.1
17.0
17.6
ND
14.9
1.50
3.30
2.50
2.80
798903
LANDM
100
6.20
1.83
13.4
6.43
18.5
25.9
26.7
1.13
22.7
2.17
5.30
2.97
1.73
TABLE C-3. PCB CONGENER CONCENTRATIONS {ng/g DRY WT) IN Afya amnaria
SAMPNUM
STATION
CB052
CB047
CB101
CB151
CB118
CB153
CB138
CB128
CB180
CB195
CB194
CB206
CB209
798901
SALT POND
9.05
2.13
7.41
1.40
5.50
6.43
4.43
0.23
2.13
ND
ND
ND
ND
798918
AH
6.54
3.11
11.1
3.80
7.40
14.1
10.3
ND
6.00
0.20
0.60
0.30
ND
798922
PR
20.9
9.92
16.4
2.10
13.9
11.7
7.30
0.40
9.20
ND
ND
ND
0.100
C-2
-------
APPENDIX D
TOTAL POLY CHLORINATED BIPHENYL CONCENTRATIONS
D - 1
-------
TABLE D-l. TOTAL PCB CONCENTRATIONS (fig/L) IN SEEP WATERS
SAMPNUM
STATION
CONC<*)
Aroclor 1242
Aroclor 1254
TOTAL PCB
798904
BRINE
0
ND
ND
ND
798909
LANDM
5.8
NM
NM
NM
798908
LANDM
11.5
NM
NM
NM
798907
LANDM
23
NM
NM
NM
798906
LANDM
46
NM
NM
NM
798905
LANDM
92
NM
NM
NM
798902
LANDM
100
ND
ND
ND
TABLE D-2.
TOTAL PCB CONCENTRATIONS (ng/g DRY WT) IN SEDIMENTS
SAMPNUM
STATION
CONC{%)
Aroclor 1242
Aroclor 1254
TOTAL PCB
798900
AH
100
39.0
1400
1440
798910
PR
100
ND
7.70
7.70
798912
POTO
100
ND
34.7
34.7
798939
CLIS
100
6.49
117
123
798915
LANDM
12.5
ND
243
240
798914
LANDM
25
ND
378
380
798913
LANDM
50
ND
608
610
798903
LANDM
100
5.92
1030
1040
TABLE D-3.
TOTAL PCB CONCENTRATIONS (ng/g DRY WT) IN Mya aremria
SAMPNUM
STATION
Aroclor 1242
Aroclor 1254
TOTAL PCB
798901
SALT POND
ND
76.3
76.3
798918
AH
ND
226
230
798922
PR
99.6
247
350
D-2
-------
APPENDIX E
POLYCYCLIC AROMATIC HYDROCARBON CONCENTRATIONS
E- 1
-------
TABLE E-l. PAH CONCENTRATIONS {p&lh) IN SEEP WATERS
SAMPNUM 798904 798909 798908 798907 798906 798905 798902
STATION BRINE LANDM LANDM LANDM LANDM LANDM LANDM
CONC (%) 0 5,8 11.5 23 46 92 100
Fluorene
ND
NM
NM
NM
NM
NM
ND
Fheoanthrene
ND
NM
NM
NM
NM
NM
0.011
Anthracene
ND
NM
NM
NM
NM
NM
ND
Sum MW178-CI
ND
NM
NM
NM
NM
NM
ND
Sum MW178-C2
ND
NM
NM
NM
NM
NM
ND
Sum MW178-C3
ND
NM
NM
NM
NM
NM
ND
Sum MW178-C4
ND
NM
NM
NM
NM
NM
ND
Fluoranthene
ND
NM
NM
NM
NM
NM
0.034
Pyrene
ND
NM
NM
NM
NM
NM
0.041
Ben2{a]anthraceae
ND
NM
NM
NM
NM
NM
0.015
Chiysene
ND
NM
NM
NM
NM
NM
0.021
Sum MW228
ND
NM
NM
NM
NM
NM
0.044
Tinuvin 328
ND
NM
NM
NM
NM
NM
ND
Tinuvin 327
ND
NM
NM
NM
NM
NM
ND
Sum Beazofluoranthenes
ND
NM
NM
NM
NM
NM
0.060
Benzo[e]pyrene
ND
NM
NM
NM
NM
NM
0.029
Benzo[s]pyrene
ND
NM
NM
NM
NM
NM
0.022
Peryleae
ND
NM
NM
NM
NM
NM
ND
Indeno[ 123-cdJpy rene
ND
NM
NM
NM
NM
NM
0.017
Dibenzfah]aiithnicene
ND
NM
NM
NM
NM
NM
ND
Beazo[ghl]perylene
ND
NM
NM
NM
NM
NM
0.025
Sum MW276
ND
NM
NM
NM
NM
NM
0.044
Sum MW278
ND
NM
NM
NM
NM
NM
0.012
Sum MW302
ND
NM
NM
NM
NM
NM
0.015
Coronene
ND
NM
NM
NM
NM
NM
ND
LOD
0.007
0.007
E - 2
-------
TABLE E-2, PAH CONCENTRATIONS (ng/g DRY WT) IN SEDIMENTS
SAMPNUM 798910 798912 798900 798903 798913 798914 798915 798939
STATION PR POTO AH LANDM LANDM LANDM LANDM CLIS
CONC (%) 100 100 100 100 50 25 12% 100
Fluorene
15
4.3
144
153
34.1
35.1
19.7
13.2
Phenaathrene
132
40.7
346
955
394
398
229
155
Anthracene
19.7
9.09
265
346
109
106
62.5
36.9
Sum MW178-C1
52
40.5
595
507
198
186
95.8
130
Sum MW178-C2
32.4
37.2
351
300
135
122
67.9
118
Sum MW178-C3
13.1
22.1
102
140
77.8
59.9
37.5
61.6
Sum MW178-C4
3.24
12.8
40.3
51.5
27.8
21.1
14.2
25.2
Fluonmthene
170
118
561
1350
845
887
534
413
Pyrene
143
104
815
1240
778
808
445
448
Benz[a]anthracene
47.6
38.5
1350
963
422
477
201
180
Chrysene
54.1
51.9
1190
1120
484
444
225
243
Sum MW228
116
108
3110
2390
1040
1060
488
487
Tinuvin 328
6.28
302
388
495
387
326
334
2.67
Tinuvin 327
1.44
56.1
60.7
81.2
68.5
60.8
57.6
ND
Sum Benzonuoranthenes
91.9
105
2670
1890
853
866
413
537
Benzo[e]pyrene
33.2
39.2
1100
677
310
303
149
216
Benzo[a]pyrene
47.3
42.8
1330
859
368
384
173
250
Perylene
13.4
29.5
432
286
129
132
69.8
83.6
Indeno[ 123-cd]pyrene
28
33.7
544
473 .
216
203
109
191
Dibenz|ah]anthracaie
8.76
10.6
187
186
83.9
78.9
40.1
55.9
Benzo[ghi]perylene
29.9
34.6
540
477
222
207
118
213
Sum MW276
80.1
91.2
1440
1290
588
552
304
525
Sum MW278
32.1
38.6
547
637
270
249
142
203
Sum MW302
61
70.3
800
907
421
377
234
440
Coroneno
9.65
13
90
103
54.4
49.6
35.3
74.7
LOD
0.608
1.16
1.15
1.77
1.32
1.32
1.45
1.83
E - 3
-------
TABLE E-3. PAH CONCENTRATIONS (ng/g DRY WT) IN Mya arenaria
SAMPNUM 798901 798918 798922
STATION SALT POND AH PR
CONG (%) 100 100 100
Fluoreae
ND
ND
ND
Phenanthreae
13.6
23.2
8.56
Anthracene
ND
ND
ND
Sum MW178-C1
13.1
19.2
14.1
Sum MW178-C2
12.5
29.6
25.2
Sum MW178-C3
ND
13
14.8
Sum MW178-C4
ND
ND
ND
Fluoranthcne
33.4
68.4
26.6
Pyrene
33.9
72.9
34.4
Beaz(a]antliracene
11
26.4
7.67
Chrysene
23.4
47.1
16.5
Sum MW228
35.8
79.9
28.9
Tinuvin 328
14.1
36.2
13.3
Tinuvin 327
7.93
14.1
6.68
Sum Benzofluorantbenes
29.5
91
25.7
Beozo[e]pyrene
23.1
53.6
22.4
Beazo[a]pyrene
10.3
33.4
9.79
Perylene
5.48
17.1
ND
Indeno[ 123-cd]pyrene
ND
19.7
ND
Diben2(ah]anthrBcene
ND
6.79
ND
Benzo[ghi]peiyIene
11.2
32.3
12.1
Sum MW276
16.3
52.3
16.8
Sum MW278
ND
7.2
ND
Sum MW302
ND
ND
ND
Corcnene
ND
ND
ND
LOD
6.21
6.67
6.15
E - 4
-------
APPENDS F
ACID VOLATILE SULFIDE CONCENTRATIONS IN SEDIMENTS
F- 1
-------
ABLE F-l. AVS CONCENTRATIONS 0*MOL/g DRY WT) IN SEDIMENTS
iAMPNUM STATION CONC(%) AVS
'98912 POTO 100 53.64
'98903 LANDM 100 51.0
F - 2
-------
APPENDIX G
TOTAL ORGANIC CARBON CONCENTRATIONS IN SEDIMENTS
G- 1
-------
TABLE G-l. TOG CONCENTRATIONS IN SEDIMENTS
SAMPNUM
STATION
CONC (96) Dry Loss (*)
Carbonate (%)
Total C (55)
Organic C (%)
798912
POTO
100 55.83
<0.02
3.16
3.16
79E903
LANDM
100 44,41
<0.02
3.74
3.74
798400
LANDM
100
<0.02
1.86
1.86
798406
LANDS
100
0.08
1.57
1.49
798412
WC
100
0.03
5.38
5.35
798414
SN
100
<0.02
1.07
1.07
798416
LANDN
100
0.04
3.13
3.09
798419
BI
100
0.03
1.50
1.47
798435
MV1
100
0.12
3.99
3.87
798436
MV1
100
0.06
3.71
3.65
798437
MV1
100
0.46
0.56
0.10
798438
MV1
100
0.12
3.75
3.63
798444
AH7
100
0.04
2.81
2.77
798445
AH7
100
0.04
3.12
3.08
798446
AH7
100
0.07
3.76
3.69
798447
AH7
100
0.17
3.18
3.01
798453
AH2
100
0.04
3.20
3.16
798454
AH2
100
<0.02
4.26
4.26
798455
AH2
100
0.02
2.37
2.35
798465
LAB
100
0.11
1.65
1.54
798466
LAB
100
0.09
1.09
1.00
798467
LAB
100
0.15
1.23
1.08
798468
LAB
100
0.12
1.35
1.23
798635
AH7
100
0.03
3.21
3.18
798636
AH7
100
0.06
3.33
3.27
798637
AH7
100
0.04
3.43
3.39
798644
AH2
100
0.04
3.26
3.22
798645
AH2
100
0.02
2.88
2.86
798646
AH2
100
0.03
3.36
3.33
798653
AH2
100
0.53
3.17
2.64
798654
MV1
100
0.21
4.06
3.85
798655
MV1
100
0.88
2.42
1.54
798665
LAB
100
0.18
1.30
1.12
798666
LAB
100
0.08
1.23
1.15
798667
LAB
100
0.08
1.24
1.16
798760
100
<0.02
1.97
1.97
798731
WC
100
<0.02
1.71
1.71
798732
SN
100
<0.02
0.77
0.77
798794
MV1
100
0.47
3.21
2.74
798805
NC
100
<0.02
3.50
3.50
798806
WC
100
<0.02
2.53
2.53
798807
SN
100
<0.02
0.79
0.79
798824
LAB
100
0.11
1.68
1.57
798825
LAB
100
0.30
2.01
1.71
798826
LAB
100
0.11
1.21
1.10
G - 2
-------
TABLE G-l (cant). TOC CONCENTRATIONS IN SEDIMENTS
SAMFNUM STATION CONG (%) Dry Loss (%) Carbonate <%) Total C (ซ) Organic C (56)
798828
AH7
100
0.02
3.04
3.02
798829
AH7
100
0.02
2.73
2.71
798830
AH7
100
0.05
3.59
3.53
798832
AH2
100
0.02
2.79
2.77
798833
AH2
100
0.02
3.17
3.15
798834
AH2
100
0.03
3.48
3.45
798836
MV1
100
0.36
3.70
3.34
798837
MV1
100
0.32
3.50
3.18
798838
MV1
100
0.36
3.05
2.69
798900
AH
100
0.08
4.67
4.59
798903
LANDM
100
0.07
3.77
3.70
798910
PR
100
<0.02
0.56
0.56
798912
POTO
100
0.10
2.98
2.88
798913
LANDM
100
0.05
2.63
2.58
798914
LANDM
100
0.04
2.75
2.71
798915
LANDM
100
0.15
2.66
2.51
798939
CLIS
100
0.17
1.81
1.64
798945
AH7
100
0.02
1.55
1.53
798946
AH7
100
0.15
1.47
1.32
G - 3
-------
APPENDIX H
SEEP WATER BIOASSAY RESULTS
H- 1
-------
TABLE H-l. Arbada pisnctulata FERTILIZATION TEST ON SEEP SAMPLES
SAMPNUM
STATION
CONG(%)
REPLICATE
% FERTILIZATION
798904
BRINE+DI
0
1
100
798904
BRINE+DI
0
2
100
798904
BRINE+DI
0
3
99
56886
NSW
0
I
99
56886
NSW
0
2
100
56886
NSW
0
3
100
798909
LANDM
5,8
1
99
798909
LANDM
5.8
2
100
798909
LANDM
5.8
3
100
798908
LANDM
11.5
1
100
798908
LANDM
11.5
2
98
798908
LANDM
11.5
3
99
798907
LANDM
23
1
98
798907
LANDM
23
2
100
798907
LANDM
23
3
96
798906
LANDM
46
1
89
798906
LANDM
46
2
94
798906
LANDM
46
3
90
798905
LANDM
92
1
42
798905
LANDM
92
2
53
798905
LANDM
92
3
47
H - 2
-------
TABLE H-2. Arbacia punctulata 48 HOUR LARVAL DEVELOPMENT TEST ON SEEP SAMPLES
SAMPNUM
STATION
CONC (%)
% NORMAL
% MORTALITY
%ABNORMAL
NSW
0
96.3
3.3
0.4
NSW
0
93.3
5.7
1.0
NSW
0
93.3
5.9
0.8
NSW
0
94.7
4.8
0.5
NSW
0
92.7
6.1
1.2
NSW
0
96.2
3.3
0.5
798909
LANDM
5.8
93.5
3.2
3.3
798909
LANDM
5.8
83.1
8.7
8.2
798909
LANDM
5.8
92.5
3.1
4.4
789908
LANDM
11.5
82.8
5.1
12.1
789908
LANDM
11.5
86.1
2.7
11.2
789907
LANDM
23
29.6
6.2
64.2
789907
LANDM
23
23.8
6.4
69.8
789907
LANDM
23
27.1
6.7
66.2
798906
LANDM
46
0.0
15.4
84.6
798906
LANDM
46
2.4
16.5
81.1
798906
LANDM
46
1.0
17.4
81.6
798905
LANDM
92
0.0
34.2
65.8
79S905
LANDM
92
0.0
22.0
78.0
798905
LANDM
92
0.0
18.7
81.3
H - 3
-------
TABLE H-3. Myddopsis bahia 48 HOUR ACUTE TEST ON SEEP SAMPLES
SAMPNUM
STATION
CONG (96) REPLICATE
% MORTALITY
NSW
0
1
0.0
NSW
0
2
0.0
NSW
0
3
0.0
NSW
0
4
20.0
798904
BRINE+DI
0
1
0.0
798904
BRINE+DI
0
2
0.0
798904
BRINE+DI
0
3
0.0
798904
BRINE+DI
0
4
0.0
798909
LANDM
5.8
1
0.0
798909
LANDM
5.8
2
0.0
798909
LANDM
5.8
3
0.0
798909
LANDM
5.8
4
0.0
789908
LANDM
11.5
1
0.0
789908
LANDM
11.5
2
0.0
789908
LANDM
11.5
3
0.0
789908
LANDM
11.5
4
0.0
789907
LANDM
23
1
20.0
789907
LANDM
23
2
20.0
789907
LANDM
23
3
0.0
789907
LANDM
23
4
0.0
798906
LANDM
46
1
0.0
798906
LANDM
46
2
0.0
798906
LANDM
46
3
0.0
798906
LANDM
46
4
0.0
798905
LANDM
92
1
0.0
798905
LANDM
92
2
0.0
798905
LANDM
92
3
0.0
798905
LANDM
92
4
0.0
H - 4
-------
TABLE H-4. Champla parvula SEXUAL REPRODUCTION TEST ON SEEP SAMPLES
SAMPNUM STATION CONC (%) REPLICATE CYSTOCARPS
NSW
0
A1
22
NSW
0
A2
27
NSW
0
A3
30
NSW
0
A4
22
NSW
0
AS
38
NSW
0
Bl
28
NSW
0
B2
20
NSW
0
B3
14
NSW
0
B4
19
NSW
0
B5
17
NSW
0
CI
18
NSW
0
C2
12
NSW
0
C3
24
NSW
0
C4
10
NSW
0
C5
16
798904
BRINE+DI
0
A1
18
798904
BRINE+DI
0
A2
17
798904
BRINE+DI
0
A3
17
798904
BRINE+DI
0
A4
30
798904
BRINE+DI
0
A5
19
798904
BRINE+DI
0
Bl
12
798904
BRINE+DI
0
B2
19
798904
BRINE+DI
0
B3
21
798904
BRINE+DI
0
B4
11
798904
BRINE+DI
0
B5
31
798904
BRINE+DI
0
CI
15
798904
BRINE+DI
0
C2
29
798904
BRINE+DI
0
C3
16
798904
BRINE+DI
0
C4
26
798904
BRINE+DI
0
C5
31
798909
LAN DM
5.8
A1
36
798909
LANDM
5.8
A2
42
798909
LANDM
5.8
A3
31
798909
LANDM
5.8
A4
42
798909
LANDM
5.8
A5
50
798909
LANDM
5.8
Bl
16
798909
LANDM
5.8
B2
14
798909
LANDM
5.8
B3
14
798909
LANDM
5.8
B4
17
798909
LANDM
5.8
B5
25
798909
LANDM
5.8
CI
17
798909
LANDM
5.8
C2
20
798909
LANDM
5.8
C3
22
798909
LANDM
5.8
C4
16
798909
LANDM
5.8
C5
17
H - 5
-------
TABLE H-4(cont). Champla parvula SEXUAL REPRODUCTION TEST ON SEEP SAMPLES
SAMPNUM STATION CONC (%) REPLICATE CYSTOCARPS
789908
LANDM
11.5
A1
12
789908
LANDM
11.5
A2
11
789908
LANDM
11.5
A3
21
789908
LANDM
11.5
A4
9
789908
LANPM
11.5
A5
14
789908
LANDM
11.5
B1
6
789908
LANDM
11.5
B2
14
789908
LANDM
11.5
B3
7
789908
LANDM
11.5
B4
14
789908
LANDM
11.5
B5
13
789908
LANDM
11.5
CI
22
789908
LANDM
11.5
C2
29
789908
LANDM
11.5
C3
19
789908
LANDM
11.5
C4
20
789908
LANDM
11.5
C5
16
789907
LANDM
23
A1
5
789907
LANDM
23
A2
1
789907
LANDM
23
A3
0
789907
LANDM
23
A4
7
789907
LANDM
23
AS
3
789907
LANDM
23
B1
3
789907
LANDM
23
B2
9
789907
LANDM
23
B3
1
789907
LANDM
23
B4
4
789907
LANDM
23
B5
8
789907
LANDM
23
CI
2
789907
LANDM
23
C2
. 12
789907
LANDM
23
C3
9
789907
LANDM
23
C4
13
789907
LANDM
23
C5
7
798906
LANDM
46
A1
2
798906
LANDM
46
A2
0
798906
LANDM
46
A3
1
798906
LANDM
46
A4
5
798906
LANDM
46
A5
0
798906
LANDM
46
B1
0
798906
LANDM
46
B2
0
798906
LANDM
46
B3
0
798906
LANDM
46
B4
0
798906
LANDM
46
B5
0
798906
LANDM
46
CI
4
798906
LANDM
46
C2
1
798906
LANDM
46
C3
2
798906
LANDM
46
C4
2
798906
LANDM
46
C5
1
798905
LANDM
92
A1
0
H - 6
-------
TABLE H-4(cont). Champla parvula SEXUAL REPRODUCTION TEST ON SEEP SAMPLES
SAMPNUM STATION CONC (%) REPLICATE CYSTOCARPS
798905
LANDM
92
A2
0
798905
LANDM
92
A3
0
798905
LANDM
92
A4
0
798905
LANDM
92
A5
0
798905
LANDM
92
B1
0
798905
LANDM
92
B2
0
798905
LANDM
92
B3
0
798905
LANDM
92
B4
0
798905
LANDM
92
B5
0
798905
LANDM
92
CI
0
798905
LANDM
92
C2
0
798905
LANDM
92
C3
0
798905
LANDM
92
C4
0
798905
LANDM
92
C5
0
H-7
-------
TABLE H-5. Mtdtnia lateralis EMBRYO / LARVAL TOXICITY TEST ON SEEP SAMPLES
SAMPNUM STATION CONC {%) REPLICATE % NO SHELL
NSW
0
1
14.7
NSW
0
2
18.0
NSW
0
3
25.3
798904
BRINE+DI
0
1
7.8
798904
BRINE+DI
0
2
16.5
798904
BRINE+DI
0
3
31.1
798909
LANDM
5.8
1
28.3
798909
LANDM
5.8
2
26.0
798909
LANDM
5.8
3
21.4
789908
LANDM
11.5
1
25.0
789908
LANDM
11.5
2
24.0
789908
LANDM
11.5
3
21.9
789907
LANDM
23
1
39.6
789907
LANDM
23
2
44.3
789907
LANDM
23
3
35.0
798906
LANDM
46
1
82.9
798906
LANDM
46
2
95.7
798906
LANDM
46
3
97.6
798905
LANDM
92
1
94.6
798905
LANDM
92
2
100.0
798905
LANDM
92
3
100.0
H- 8
-------
TABLE H-6. Menidia beryUna 96 HOUR LARVAL SURVIVAL TEST ON SEEP SAMPLES
SAMPNUM STATION CONC (56) REPLICATE % MORTALITY
NSW
0
1
10.0
NSW
0
2
10.0
798909
LANDM
5.8
1
30.0
798909
LANDM
5.8
2
20.0
789908
LANDM
11.5
1
10.0
789908
LANDM
11.5
2
0.0
789907
LANDM
23
1
0.0
789907
LANDM
23
2
0.0
798906
LANDM
46
I
0.0
798906
LANDM
46
2
0.0
798905
LANDM
92
1
0.0
798905
LANDM
92
2
0.0
H - 9
-------
APPENDIX I
SEDIMENT BIOASSAY RESULTS
I- 1
-------
TABLE 1-1. Ampelisca abdita SEDIMENT BIOASSAY ON SEDIMENT SAMPLES
SAMPNUM STATION CONC (%) RHPUCATE % MORTALITY
798912
POTO
100
1
S.O
798912
POTO
100
2
0.0
798903
LANDM
100
1
15.0
798903
LANDM
100
2
5.0
798903
LANDM
100
3
10.0
1-2
-------
TABLE 1-2. Mulima latemlis 7 DAY SEDIMENT TOXICITY TEST
SAMPNUM STATION CONC (56) REPLICATE % MORTALITY GROWTH A GROWTH
und
1
0.0
1.4
0.9
ซ*nd
2
0.0
13
0.9
WTVf
3
0.0
1.4
0.9
ฆand
4
0.0
1.7
1.1
land
5
0.0
13
0.9
(tod
6
17.0
13
1.0
798910
PR
1
0.0
1.4
0.9
798910
PR
2
0.0
1.8
1.2
798910
PR
3
33.0
1.4
0.9
798910
PR
4
0.0
1.4
0.9
798910
PR
5
0.0
1.5
1.0
798910
PR
6
0.0
1.4
0.9
NJT
1
0.0
1.6
1.1
NIT
2
17.0
13
0.9
NIT
3
17.0
13
1.0
NJT
4
0.0
1.6
1.1
NJT
5
0.0
1.4
0.9
NJT
6
0.0
1.6
1.1
798939
cus
1
0.0
1.1
0.7
798939
cus
2
0.0
13
0.9
798939
cus
3
17.0
1.0
0.7
798939
CLIS
4
0.0
1.4
0.9
798939
cus
5
17.0
1.6
1.1
798939
CLIS
6
17.0
13
1.0
798912
POTO
0
1
0.0
1.0
0.7
798912
POTO
0
2
0.0
1.2
0.8
798912
POTO
0
3
0.0
1.2
0.8
798912
POTO
0
4
0.0
1.1
0.7
798912
POTO
0
5
0.0
1.2
0.8
798912
POTO
0
6
0.0
1.2
0.8
798915
LANDM
12.5
1
17.0
13
1.0
798915
LANDM
123
2
0.0
1.2
0.8
798915
LANDM
12.5
3
0.0
13
0.9
798915
LANDM
12.5
4
17.0
1.6
1.1
798915
LANDM
123
5
0.0
1.4
0.9
798915
LANDM
12.5
6
17.0
13
0.9
798914
LANDM
25
1
17.0
1.4
0.9
798914
LANDM
25
2
17.0
1.4
0.9
798914
LANDM
25
3
0.0
1.8
1.2
798914
LANDM
25
4
0.0
13
0.9
798914
LANDM
25
5
17.0
1.2
0.8
798914
LANDM
25
6
0.0
13
0.9
798913
LANDM
50
1
0.0
1.6
1.1
798913
LANDM
50
2
0.0
1.6
1.1
798913
LANDM
50
3
17.0
1.2
0.8
798913
LANDM
50
4
0.0
1.2
0.8
798913
LANDM
50
5
0.0
1.1
0.7
798913
LANDM
50
6
0.0
1.4
0.9
798903
LANDM
100
1
0.0
1.5
1.0
798903
LANDM
100
2
17.0
1.2
0.8
798903
LANDM
100
3
17.0
1.4
0.9
798903
LANDM
100
4
17.0
1.6
1.1
7989(0
LANDM
100
S
0.0
13
0.9
798903
LANDM
100
6
0.0
13
1.0
1-3
-------
APPENDIX J
SEDIMENT EXTRACT BIOASSAY RESULTS
J- 1
-------
TABLE J-l. Arbacia punctulata FERTILIZATION TEST ON SEDIMENT EXTRACTS
SAMPNUM STATION CONC<%) REPLICATE % FERTILIZATION
798947
NSW
0.000
1
98.0
798947
NSW
0.000
2
99.0
798947
NSW
0.000
3
98.0
798946
DMSO
(0.05)
1
100.0
798946
DMSO
(0.05)
2
99.0
798946
DMSO
(0.05)
3
97.0
798945
BLANK
(0.05)
1
98.0
798945
BLANK
(0.05)
2
95.0
798945
BLANK
(0.05)
3
96.0
798944
LANDM
0.003
1
99.0
798944
LANDM
0.003
2
98.0
798944
LANDM
0.003
3
98.0
798943
LANDM
0.003
1
98.0
798943
LANDM
0.006
2
94.0
798943
LANDM
0.006
3
93.0
798942
LANDM
0.013
1
66.0
798942
LANDM
0.013
2
81.0
798942
LANDM
0.013
3
79.0
798941
LANDM
0.025
1
0.0
798941
LANDM
0.025
2
0.0
798941
LANDM
0.025
3
3.0
798940
LANDM
0.050
1
0.0
798940
LANDM
0.050
2
0.0
798940
LANDM
0.050
3
0.0
3-2
-------
TABLE J-2. Arbacia punctulata 48-HOUR LARVAL DEVELOPMENT TEST ON SEDIMENT
EXTRACTS RUN 1
SAMPNUM STATION CONC (%) REPLICATE % NORMAL % MORTALITY % ABNORMAL
798950
NSW
0
1
82.5
11.0
6.5
798950
NSW
0
2
87.5
9.0
3.5
798950
NSW
0
3
89.0
8.5
2.5
798948
BLANK
(0.25)
1
81.0
13.5
5.5
798948
BLANK
(0.25)
2
84.5
13.0
2.5
798948
BLANK
(0.25)
3
84.0
12.5
3.5
798949
DMSO
(0.25)
1
82.5
13.0
4.5
798949
DMSO
(0.25)
2
89.5
9.0
1.5
798949
DMSO
(0.25)
3
87.0
12.0
1.0
798938
LANDM
0.025
1
86.5
12.0
1.5
798938
LANDM
0.025
2
88.5
9.5
4.0
798938
LANDM
0.025
3
86.5
7.5
4.0
798937
LANDM
0.05
1
86.5
12.0
1.5
798937
LANDM
0.05
2
83.5
12.5
4.0
798937
LANDM
0.05
3
85.0
9.5
5.5
798936
LANDM
0.1
1
1.5
10.5
88.0
798936
LANDM
0.1
2
1.0
14.0
85.0
798936
LANDM
0.1
3
0.0
15.5
84.5
798935
LANDM
0.15
1
0.5
12.5
87.0
798935
LANDM
0.15
2
0.0
11.5
88.5
798935
LANDM
0.15
3
0.0
14.0
86.0
798934
LANDM
0.2
1
0.5
32.0
67.5
798934
LANDM
0.2
2
0.0
36.0
64.0
798934
LANDM
0.2
3
0.0
43.0
57.0
798933
LANDM
0.25
1
0.0
91.5
8.5
798933
LANDM
0.25
2
0.0
90.0
10.0
798933
LANDM
0.25
3
0.0
91.5
8.5
J - 3
-------
TABLE 1-2 (coni). Arimcia punctulata 48-HOUR LARVAL DEVELOPMENT TEST ON
SEDIMENT EXTRACTS RUN 2
SAMPNUM STATION CONG (ป) REPLICATE % NORMAL % MORTALITY % ABNORMAL
798947
NSW
0
1
77.5
13.0
9.5
798947
NSW
0
2
74.0
16.0
9.5
798947
NSW
0
3
75.5
13.0
11.5
798945
BLANK
(0.05)
1
74.0
14.0
12.0
798945
BLANK
(0.05)
2
78.0
12.0
10.0
798945
BLANK
(0.05)
3
*1.5
12.5
16.0
798946
DMSO
(0.05)
1
85.0
8.5
6.5
798946
DMSO
(0.05)
2
78.5
10.5
11.0
798946
DMSO
(0.05)
3
85.0
6.0
9.0
798944
LANDM
0.003
1
72.5
20.5
7.0
798944
LANDM
0.003
2
87.0
5.5
7.5
798944
LANDM
0.003
3
89.5
3.0
7.5
798943
LANDM
0.006
1
79.0
11.0
10.0
798943
LANDM
0.006
2
71.0
19.0
10.0
798943
LANDM
0.006
3
84.0
5.5
10.5
798942
LANDM
0.013
1
- 83.5
11,5
5.0
798942
LANDM
0.013
2
88.5
8.5
6.0
798942
LANDM
0.013
3
83.5
6.5
10.0
798941
LANDM
0.025
1
78.5
6.5
15,0
798941
LANDM
0.025
2
77.5
11.5
11.0
798941
LANDM
0.025
3
85.0
9.5
5.5
798940
LANDM
0.05
1
86.0
7.5
6.5
798940
LANDM
0.05
2
79.0
11.5
C.5
798940
LANDM
0.05
3
76.0
9.5
14.5
1-4
-------
TABLE J-3. Mulinia lateralis EMBRYO / LARVAL TOXICITY TEST ON SEDIMENT
EXTRACTS
SAMPNUM
STATION
CONC (%>
REPLICATE
% NO SHELL
NSW
0
1
26.0
NSW
0
2
30.0
NSW
0
3
34.0
BLANK
(0.05)
1
84.0
BLANK
(0.05)
2
93.0
BLANK
(0-05)
3
87.0
DMSO
(0.05)
1
93.0
DMSO
(0.05)
2
92.0
DMSO
(0.05)
3
93.0
798944
LANDM
0.003
1
92.0
798944
LANDM
0.003
2
89.0
798944
LANDM
0.003
3
88.0
798943
LANDM
0.006
1
90.0
798943
LANDM
0.006
2
92.0
798943
LANDM
0.006
3
89.0
798942
LANDM
0.013
1
" 99.0
798942
LANDM
0.013
2
99.0
798942
LANDM
0.013
3
93.0
798941
LANDM
0.025
1
100.0
798941
LANDM
0.025
2
100.0
798941
LANDM
0.025
3
100.0
798940
LANDM
0.05
1
100.0
798940
LANDM
0.05
2
100.0
798940
LANDM
0.05
3
100.0
J - 5
-------
TABLE J-4. MICROTOX ASSAY ON SEDIMENT EXTRACTS
SAMPNUM
STATION
CONC <*)
REPLICATE
BIOLUMINESCENCE
798932
BLANK
(0.001)
1
100.0
798931
LANDM
0.001
1
97.0
798931
LANDM
0.001
2
98.0
798930
BLANK
(0.01)
1
98.0
798930
BLANK
(0.01)
2
101.0
798929
LANDM
0.01
1
73.0
798929
LANDM
0.01
2
71.0
798928
BLANK
(0.1)
1
100.0
798928
BLANK
(0.1)
2
99.0
798927
LANDM
0.1
1
20.0
798927
LANDM
0.1
2
20.0
798926
BLANK
(1)
1
102.0
798926
BLANK
(1)
2
99.0
798925
LANDM
1
1
8.0
798925
LANDM
1
2
8.0
J - 6
-------
APPENDIX K
MYA NEOPLASIA EXPERIMENT RESULTS
K- 1
-------
TABLE K-l. LABORATORY EXPOSURE OF Mya artmria.
SAMPNUM
STATION
REPLICATE INJECTED
95 HN
MORTALITY GROWTH
SE
798922
PR
1
+
28.3
15
-0.71
0.237
798922
PR
2
+
36.7
6.7
-1.61
0.282
798922
PR
1
-
25
16.7
-1.29
0.230
798922
PR
2
-
31.1
9.8
-0.74
0.359
798918
AH
1
+
60
18.3
-0.64
0.210
798918
AH
2
+
35
15
-2.55
0.313
798918
AH
1
ฆ
39
6.8
-0.34
0.395
798918
AH
2
-
39.3
9.8
-1.76
0.226
K - 2
-------
APPENDIX L
GRAIN SIZE ANALYSES OF SEDIMENTS
L- 1
-------
TABLE L-l. GRAIN SIZE ANALYSIS.
SAMPNUM STATION CONC (X) % SAND % SILT % CLAY
798912 POTO 100 47.7 40.4 11.9
798903 LANDM 100 92.69 4.05 3.26
L - 2
------- |