EPA 910/9-88-213
uget Sound Estuary Program
ELLIOTT BAY ACTION PROGRAM:
Analysis of Toxic Problem Areas
FINAL REPORT
July 1988
Prepared for
U.S. Environmental Protection Agency
Region X - Office of Puget Sound
Seattle, Washington
-------
TC 3338-23
Final Report
U.5. £PA Contract No. 68-02-4341
ELLIOTT BAY ACTION PROGRAM:
ANALYSIS OF TOXIC PROBLEM AREAS
Prepared by:
PTI Environmental Services
3625 132nd Avenue SE, Suite 301
Bellevue, WA 98006
and
Tetra Tech, Inc.
11820 Northup Way N.E.
Bellevue, WA 98005
For:
U.S. Environmental Protection Agency
Region X, Office of Puget Sound
Seattle, WA 98101
July 1988
-------
CONTENTS
Page
LIST OF FIGURES vii
LIST OF TABLES xiii
ACKNOWLEDGMENTS xvi
EXECUTIVE SUMMARY xx
PROJECT AREA xxi
DECISION MAKING APPROACH xxi
REFERENCE AREAS xxiii
SEDIMENT CHEMISTRY xxiii
BIOACCUMULATION xxvi
SEDIMENT TOXICITY BIOASSAYS xxvi
BENTHIC MACROINVERTEBRATE COMMUNITIES xxvi
FISH PATHOLOGY xxvii
CONTAMINANT, TOXICITY, AND BIOLOGICAL EFFECTS RELATIONSHIPS xxvii
PRIORITIZATION OF PROBLEM AREAS AND CONTAMINANTS xxviii
INTRODUCTION 1
SITE DESCRIPTION 4
DRAINAGE PATTERNS 4
STUDY AREAS 6
METHODS 8
DECISION-MAKING FRAMEWORK 8
Overview 8
Chemical, Biological, and Toxicological Indices 10
Action Assessment Matrix 15
Action-Level Guidelines 17
Ranking of Problem Areas 19
Quantitative Relationships 21
i i
-------
Spatial Resolution of Effects 23
Source Evaluation 23
OVERVIEW, OF FIELD STUDY DESIGN 23
Station Locations 28
Data Analysis Methods 31
SEDIMENT CHEMISTRY 35
Field Sampling 35
Laboratory Analysis for Metals 36
Laboratory Analysis for Semivolatile Organic Compounds
(Acid/Neutral and Pesticide/PCB) 37
Laboratory Analysis for Volatile Organic Compounds 38
Ancillary Analyses 38
Quality Assurance/Quality Control Results 40
BIOACCUMULATION 44
Field Sampling 44
Laboratory Analysis for Mercury 45
Laboratory Analysis for PCBs/Pesticides 45
Quality Assurance/Quality Control Results 46
SEDIMENT BIOASSAY 47
Field Sampling 47
Laboratory Analysis 47
Quality Assurance/Quality Control Results 48
BENTHIC MACROINVERTEBRATES 48
Field Sampling 48
Laboratory Analysis 49
Quality Assurance/Quality Control Results 50
FISH ECOLOGY AND HISTOPATHOLOGY 50
Field Sampling 50
Laboratory Analysis 52
Quality Assurance/Quality Control Results 53
DATA MANAGEMENT 53
Data Organization 53
Data Analysis 54
Data Entry and Quality Control 54
iii
-------
RESULTS
55
SEDIMENT CHEMISTRY 55
Normalization of Chemical Concentrations 55
Evaluation of the Reference Area 56
Conventional Sediment Characteristics 57
Sediment.Chemistry: Metals 72
Sediment Chemistry: Organic Compounds 88
Spatial Correlations Among Chemicals 122
Comparison with Recent Historical Data 123
Summary 140
BIOACCUMULATION 141
Normalization of Chemical Concentrations 141
Evaluation of the Reference Area 143
Mercury in Fish Tissue 143
PCBs/Pesticides in Fish Tissue 145
Comparison with Recent Historical Data 148
Summary 149
SEDIMENT BIOASSAYS 149
Evaluation of the Reference Area 150
General Patterns of Amphipod Mortality 150
Comparison of the Elliott Bay System with Port Susan 153
Comparison with Recent Historical Data 153
Summary 160
BENTHIC MACROINVERTEBRATES 160
Evaluation of the Reference Area 162
Characteristics of Benthic Communities in Elliott Bay
and Port Susan 169
Comparisons Between Reference and Potentially Impacted Areas 171
Species-Level Comparisons Among Elliott Bay and Port Susan
Stations 177
Classification Analyses 186
Comparison of Species and Major Taxa Level Analyses 189
Indices for Decision Criteria 190
Comparison with Recent Historical Data 194
Summary 194
FISH ECOLOGY 194
Demersal Fish Assemblages 195
English Sole Populations 199
Summary 199
FISH HIST0PATH0L0GY 199
Evaluation of the Reference Area 201
Age and Sex Characteristics of Fish Populations 201
iv
-------
General Patterns of Lesion Prevalences 201
Comparisons of Lesion Prevalences Between Study Areas 205
Length-at-Age Comparisons 209
Comparison with Recent Historical Data 215
Summary 217
CONTAMINANT, TOXICITY, AND BIOLOGICAL EFFECTS RELATIONSHIPS 219
RELATIONSHIPS AMONG CONTAMINANTS, TOXICITY, AND BENTHIC
EFFECTS 219
General Correlation of Indicators 219
Comparison of Contamination and Significant Biological
Effects 227
COMPARISON OF BIOASSAY RESPONSES WITH BENTHIC INVERTEBRATE
ASSEMBLAGES 229
Correlation of Indicators 230
Comparison of Bioassay Responses with Benthic Groupings 232
Comparison of Significant Responses 232
RELATIONSHIP BETWEEN BIOACCUMULATION AND SEDIMENT CONTAMINATION 235
SUMMARY 237
PRIORITIZATION OF PROBLEM AREAS AND CONTAMINANTS 239
IDENTIFICATION OF PROBLEM AREAS 239
RANKING OF PROBLEM AREAS 239
CHEMICAL CHARACTERIZATION OF PROBLEM AREAS 248
Problem Area SS 248
Problem Area NHI 252
Problem Area NHII 255
Problem Area WWI 256
Problem Area WWII 256
East Waterway (EW) Problem Area 259
Problem Area KGII 261
Problem Area NSII 264
Problem Area NSI 264
Upper Duwamish River Problem Areas 264
Problem Area KGI (Slip 1) 268
Problem Area KG-03 268
Problem Area AB-01 268
Problem Area NH-10 269
SUMMARY 269
Identification of Problem Areas 269
Ranking of Problem Areas 269
Characterization of Problem Areas 269
v
-------
REFERENCES
APPENDICES (BOUND SEPARATELY)
APPENDIX A: CHEMICAL AND PHYSICAL DATA COLLECTED DURING THE
ELLIOTT BAY ACTION PROGRAM
APPENDIX
B:
ELLIOTT BAY STATION LOCATIONS
APPENDIX
C:
CORRELATION MATRICES FOR SELECTED CHEMICALS
APPENDIX
D:
SUMMARY OF SEDIMENT GRAIN SIZE AND CONVENTIONAL DATA
APPENDIX
E:
AMPHIPOD BIOASSAY AND BENTHIC INFAUNA DATA
APPENDIX
F:
APPLICATION OF LAET AND HAET TO ELLIOTT BAY DATA
APPENDIX
G:
RANKING OF TIER II PROBLEM AREAS
vi
-------
FIGURES
Number Page
ES-1 Project location and study areas within Elliott Bay
and the lower Duwamish River xxii
ES-2 Locations of sampling stations for sediment chemistry,
amphipod bioassay, and benthic macroinvertebrates xxiv
ES-3 Highest priority problem sites xxix
1 Elements of the Elliott Bay Toxics Action Program 2
2 Project location and study areas within Elliott Bay
and the lower Duwamish River 5
3 Decision-making approach for the Elliott Bay Action Program 9
4 Theoretical example of relationship between sediment
contamination and an effects index 22
5 Locations of sampling stations for sediment chemistry,
amphipod bioassay, and benthic macroinvertebrates 29
6 Locations of trawl transects in Elliott Bay 30
7 Cumulative numbers of species collected over successive
replicate benthic grab samples at four stations in
Port Susan and at two stations in Elliott Bay in 1985 51
8 Average percent fine-grained material (silt plus clay)
in sediments of Elliott Bay, the Duwamish River,
and Port Susan 58
9 Percent fine-grained material in sediments of Elliott Bay
and the Duwamish River 59
10 Percent fine-grained material in sediments of the
East Waterway 60
11 Percent fine-grained material in sediments of North
Harbor Island 61
12 Percent fine-grained material in sediments of the
West Waterway 62
13 Grain size and total organic carbon content of Port Susan
sediments 63
VM
-------
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
64
66
67
68
69
70
71
78
79
81
82
84
85
86
87
89
90
108
Average oil and grease concentrations in sediments of
Elliott Bay, the Duwamish River, and Port Susan
Average total organic carbon content in sediments of
Elliott Bay, the Duwamish River, and Port Susan
Total organic carbon content in sediments of Elliott Bay
and the Duwamish River (individual stations)
Total organic carbon content in sediments of the
East Waterway
Total organic carbon content in sediments of North
Harbor Island
Total organic carbon content in sediments of the
West Waterway
Average sulfide concentrations in sediments of Elliott Bay,
the Duwamish River, and Port Susan
Mean elevations above reference (EAR) of antimony in
sediments from all study areas
Elevations above reference (EAR) of antimony at individual
stations in the most contaminated study areas
Mean elevations above reference (EAR) of copper in
sediments from all study areas
Elevations above reference (EAR) of copper at individual
stations in the most contaminated study areas
Mean elevations above reference (EAR) of lead in sediments
from all study areas
Elevations above reference (EAR) of lead at individual
stations in the most contaminated study areas
Mean elevations above reference (EAR) of mercury in
sediments from all study areas
Elevations above reference (EAR) of mercury at individual
stations in the most contaminated study areas
Mean elevations above reference (EAR) of zinc in sediments
from all study areas
Elevations above reference (EAR) of zinc at individual
stations in the most contaminated study areas
Mean elevations above reference (EAR) of LPAH and HPAH
in sediments from all study areas
vi i i
-------
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
Elevations above reference (EAR) of LPAH and HPAH at
individual stations in the most contaminated study areas 109
Elevations above reference (EAR) of LPAH and HPAH at
individual stations in highly contaminated study areas 110
Mean elevations above reference (EAR) of PCBs in sediments
from all study areas 112
Elevations above reference (EAR) of PCBs in the Duwamish
River 113
Elevations above reference (EAR) of PCBs along the
Seattle South Waterfront 115
Elevations above reference (EAR) of PCBs in the East
Waterway 116
Elevations above reference (EAR) of PCBs in North Harbor
Island and the West Waterway 117
Locations of sampling stations from historical studies of
sediment chemistry in Elliott Bay and the Duwamish River 127
Locations of sampling stations from historical studies of
sediment chemistry in the East Waterway 128
Locations of sampling stations from.historical studies of
sediment chemistry in the North Harbor Island area 129
Locations of sampling stations from historical studies of
sediment chemistry in the West Waterway 130
Contours of copper concentrations in study area sediments 131
Contours of lead concentrations in study area sediments 133
Contours of mercury concentrations in study area sediments 134
Contours of zinc concentrations in study area sediments 135
Contours of LPAH concentrations in study area sediments 137
Contours of HPAH concentrations in study area sediments 139
Contours of PCB concentrations in study area sediments 142
PCB concentrations in muscle tissue of English sole from
Elliott Bay, the Duwamish River, and Point Pully 146
Amphipod bioassay data from Puget Sound reference areas 151
Mean and range of amphipod bioassay responses within
study areas 152
ix
-------
53 Significant amphipod bioassay mortalities compared to the
Port Susan reference area 158
54 Comparison of amphipod bioassay results with previous data
from the Puget Sound atlas 161
55 Mean total abundances and mean abundances of major taxonomic
groups of benthic invertebrates in Puget Sound reference
areas 165
56 Mean number of individuals/m^ in each study area segment 170
57 Summary of significant enhancements and depressions of
polychaete abundances among the Elliott Bay stations
compared with reference conditions 172
58 Summary of significant enhancements and depressions of
crustacean abundances among the Elliott Bay stations
compared with reference conditions 173
59 Summary of significant enhancements and depressions of
pelecypod abundances among the Elliott Bay stations
compared with reference conditions 174
60 Summary of significant enhancements and depressions of
gastropod abundances .among the Elliott Bay stations
compared with reference conditions 175
61 Summary of significant enhancements and depressions of
total macrofaunal abundances among the Elliott Bay
stations compared with reference conditions 176
62 Mean number of benthic species per station collected from
the 20 Elliott Bay and Port Susan stations where complete
identification of benthic samples was performed 178
63 Mean abundance of total benthic infauna per station
collected from the 20 Elliott Bay and Port Susan stations
where complete identification of benthic samples was
performed 179
64 Abundances of the five most numerically dominant species at
completely identified stations in Elliott Bay and Port Susan 180
65 Relative abundances of the numerically dominant species at
completely identified stations in Elliott Bay and Port Susan
and the proportions of total infaunal abundance for which
they account 181
66 Abundances (as percent of fauna) of opportunistic and
pollution-tolerant taxa at completely identified stations
in Elliott Bay and Port Susan 184
x
-------
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
Results of a Q-mode classification analysis using log-
transformed abundances of the benthic infauna at the
completely identified stations in Elliott Bay and Port Susan 187
Summary of spatial patterns of significant benthic
depressions among the Elliott Bay stations 192
Comparisons of major characteristics of fish assemblages
between Elliott Bay transects and Point Pully 19ff
Comparisons of abundances and relative abundances of
English sole between Elliott Bay transects and Point Pully 200
Comparisons of prevalences of hepatic lesions between
Elliott Bay and Point Pully 204
Comparisons of prevalences of hepatic lesions with age
of English sole from Elliott Bay using Spearman's
coefficient of rank correlation 206
Comparisons of age distributions between Elliott Bay
transects and Point Pully 207
Spatial distribution of prevalences of neoplasms in
English sole from Elliott Bay 211
Spatial distribution of prevalences of foci of cellular
alteration in English sole from Elliott Bay 212
Spatial distribution of prevalences of megalocytic hepatosis
in English sole from Elliott Bay 213
Comparisons of length-at-age between English sole with and
without hepatic lesions using the Mann-Whitney U-test 214
Comparison of prevalences of hepatic lesions in English sole
sampled in the present study and in Malins et al. (1984) 216
Relationships between abundances of selected benthic taxa
and PCB concentration in sediments 221
Relationships between abundances of selected benthic taxa
and concentration of high molecular weight aromatic
hydrocarbons in sediments 222
Relationships between abundances of selected benthic taxa
and copper concentration in sediments 223
Relationships between abundances of selected benthic taxa
and sulfide concentration in sediments 224
Relationships between abundances of selected benthic taxa
and sediment grain size 225
xi
-------
84 Relationship between amphipod bioassay response and sulfides,
copper, grain size, and total organic carbon in sediments 226
85 Relationships between abundances of selected benthic taxa
and amphipod bioassay response 231
86 Amphipod bioassay responses in relation to station groupings
based on classification analysis of benthic assemblages 233
87 Relationship between PCB bioaccumulation and PCB sediment
contamination 236
88 Boundaries of priority problem areas 241
89 Ranking of priority problem areas based on average
conditions within each area 244
90 Ranking of single stations classified as problem sites 245
91 Highest priority problem sites 247
xii
-------
TABLES
Number Page
1 List of contaminants and conventional variables for
analysis in Elliott Bay project 12
2 Theoretical example of action assessment matrix 16
3 Action-level guidelines 18
4 Summary of scoring criteria for sediment contamination,
toxicity, and biological effects indicators 20
5 Puget Sound AET 24
6 Summary of field study design 27
7 Comparison of analytical methods for selected metals 42
8 Concentrations of metals in surface sediments of Elliott Bay,
Duwamish River, and Port Susan 73
9 Summary of metal concentrations in sediments from Puget Sound
reference areas 74
10 Range in (EAR) for inorganic contaminants of concern in
sediments of Elliott Bay and the Duwamish River 76
11 Summary of metals with EAR between 100-1,000 and >1,000
in sediments of Elliott Bay and the Duwamish River 77
12 Concentrations of detected semivolatile organic compounds
in surface sediments of Elliott Bay, Duwamish River, and
Port Susan 91
13 Infrequently detected acid/neutral compounds with high
detection limits 93
14 Concentrations of volatile organic compounds in surface
sediments of Elliott Bay and the Duwamish River 95
15 Summary of organic compound concentrations in sediments
from Puget Sound reference areas 96
16 Range in EAR for selected organic contaminants of concern
in sediments of Elliott Bay and the Duwamish River 100
xm
-------
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Summary of semi volatile organic chemicals with EAR between
100-1,000 in sediments of Elliott Bay and the Duwamish River
(using area means) 101
Summary of semi volatile organic chemicals with EAR between
100-1,000 in sediments of Elliott Bay and the Duwamish River
(using individual stations) 102
Summary of semivolatile organic chemicals with EAR >1,000
in sediments of Elliott Bay and the Duwamish River
(using area means) 105
Summary of semivolatile organic chemicals with EAR >1,000
in sediments of Elliott Bay and the Duwamish River
(using individual station) 106
Correlations between PAH and miscellaneous organic compounds 121
Strong chemical correlations 124
Mercury concentrations in fish collected from Elliott Bay,
the Duwamish River, and Point Pully 144
PCB concentrations, lipid contents, and ages of Elliott Bay
fish 147
Summary of amphipod bioassay results 154
Surface sediment characteristics at benthic infauna stations
in Port Susan compared with other reference areas in
Puget Sound 164
Numerically dominant taxa at Port Susan stations sampled
in 1985 and 1986, and at Carr Inlet station sampled in 1984 166
Comparison of abundances (as percent of fauna) of oppor-
tunistic and pollution-tolerant taxa at stations in
Port Susan and Carr Inlet 167
Key for Figures 63 and 64 182
Abundances (as percent of fauna) of opportunistic and
pollution-tolerant taxa at stations in Port Susan and
Elliott Bay 185
Relative abundances of fishes captured in Elliott Bay and
at Point Pully 196
Comparisons of age and male proportion between English sole
from Point Pully and Elliott Bay 202
Prevalences of hepatic lesions in English sole from
Elliott Bay and Point Pully 203
xiv
-------
34 Comparisons of lesion prevalences between male and female
English sole from Elliott Bay 208
35 Comparisons of prevalences of hepatic lesions in English sole
from Elliott Bay and Point Pully 210
36 Correspondence between stations having significant bioassay
responses and stations having significant benthic depressions 234
37 Action assessment matrix of sediment contamination, toxicity,
and biological effect indices for Elliott Bay problem areas 242
38 Potential problem chemicals 249
39 Seattle South Waterfront problem areas 250
40 North Harbor Island problem areas 253
41 West Waterway problem areas 257
42 East Waterway problem area 260
43 Kellogg Island problem areas 262
44 Seattle North Waterfront problem areas 265
45 Upper Duwamish River problem areas 266
xv
-------
ACKNOWLEDGMENTS
This document was prepared by Tetra Tech and PTI Environmental Services
under the direction of Dr. Jean Jacoby of Tetra Tech, for the U.S. En-
vironmental Protection Agency (EPA) in partial fulfillment of Contract
No. 68-02-4341. This project was funded through the National Estuary
Program under the authorities of the Clean Water Act as amended. Funding
was approved by the U.S. EPA Office of Marine and Estuarine Protection. Ms.
Clare Ryan, Ms. Martha Burke, and Dr. John Armstrong served as technical
monitors for EPA Region X throughout various phases of the project. Mr.
Barry Towns and Mr. Raleigh Farlow were Quality Assurance Officers for EPA
Region X.
The primary authors of this report are Mr. Harry Beller, Dr. Robert
Pastorok, and Dr. Scott Becker of PTI Environmental Services; Mr. Gary Braun
and Dr. Gordon Bilyard of Tetra Tech; and Dr. Peter Chapman of E.V.S. Con-
sultants. Technical staff that contributed to the sampling, data analysis,
and report writing efforts are listed below.
Tetra Tech (Tt) and PTI Environmental Services (PTI)
Mr. Robert Barrick (PTI)
Dr. Scott Becker (PTI)
Mr. Harry Beller (PTI)
Dr. Gordon Bilyard (Tt)
Mr. Gary Braun (Tt)
Ms. Roberta Feins (Tt)
Dr. Thomas Ginn (PTI)
Dr. Everett Hogue (PTI)
Dr. Jean Jacoby (Tt)
Ms. Karen Keeley (Tt)
Ms. Lynn Ki1patrick-Howard (Tt)
Mr. Dreas Nielsen (PTI)
Dr. Robert Pastorok (PTI)
Ms. Beth Schmoyer (Tt)
Mr. George Schupp (Tt)
Mr. Jeffrey Stern (Tt)
Mr. Andy Torres (PTI)
Study Design, Data Analysis,
Chemistry Quality Assurance,
Decision-Making Approach
Study Design, Field Sampling, Data
Analysis, Fish Ecology and Pathology
Chemistry Quality Assurance, Data
Analysis, Interdisciplinary Analysis
Study Design, Data Analysis, Benthic
Infauna
Data Analysis, Benthic Infauna
Database Management
Study Design, Decision-Making
Approach
Data Analysis, Bioassays
Project Management, Technical Review
Study Design, Field Coordination
and Sampling, Benthic Infauna
Chemistry Quality Assurance
Database Management
Technical Coordination, Study
Design, Interdisciplinary Analysis
Study Design, Field Sampling, Source
Evaluation
Chemistry Quality Assurance
Field Sampling
Chemistry Data Compilation and
Mapping
xvi
-------
Ms. Stacey Vineberg (Tt)
Mr. John Virgin (Tt)
Ms. Carolyn Walden (Tt)
Ms. Kim Wiers (Tt)
Ms. Julia Wilcox (Tt)
Dr. Les Williams (Tt)
E.V.S. Consultants
Dr. Peter Chapman
Dr. Robert Dexter
Staff
Fish and Wildlife Health Consultants
Database Management, Mapping
Chemistry Quality Assurance Coordina-
tion
Database Management
Field Sampling
Chemistry Quality Assurance Coordina-
tion, Sample Tracking
Technical Review
Bioassays, Data Analysis
Field Supervision, Station Position-
ing
Infauna Sorting
Dr. Richard Kocan
Dr. Marsha Landolt
Mr. David Powell
Invert-Aid
Ms. Dianne Robbins
Independent Taxonomists
Mr. Howard Jones
Mr. Ralph Brinkhurst
Mr. Robert G.B. Reid
Mr. William C. Austin
Mr. Peter Stripl.in
Analytical Chemistry Laboratories
Battelle Northwest Marine Res. Lab.
Science Applications Int'l. Corp.
Battelle Columbus Laboratory
Am Test
Fish Pathology Sampling and Analysis
Fish Pathology Sampling and Analysis
Fish Pathology Sampling and Analysis
Field Sampling
Metals in sediments and tissue,
PCB/pesticides in tissue, total
organic carbon, total nitrogen,
total volatile solids, oil and
grease, and grain size distribution
Organic compounds in sediments
except volatile compounds
Volatile organic compounds
Sulfides
Polychaeta
Oligochaeta
Mollusca
Arthropoda
Echinodermata/Other Phyla
The Elliott Bay Toxics Action Program has benefited from the partici-
pation of an Interagency Work Group (IAWG) and a Citizen's Advisory Committee
(CAC). Duties of the IAWG and CAC members included: 1) reviewing program
documents, agency policies, and proposed actions; 2) providing data reports
xv i i
-------
and other technical information to EPA; and 3) disseminating action program
information to respective interest groups or constituencies. We thank the
IAWG and CAC members for their past and continuing efforts. We are espe-
cially grateful to Ms. Joan Thomas, Mr. Gary Brugger, and Mr. Dan Cargill
for chairing the IAWG, and to Mr. David Schneidler and Ms. Janet Anderson
for co-chairing the CAC. Members of the IAWG and CAC are listed below.
Elliott Bay Interagency Work Group
Name
Mr. Dave Aggerholm
Dr. John Armstrong
Mr. Morgan Bradley
Mr. Gary Brugger
Mr. Dan Cargill
Lori Cohen
Lee Dorigan
Nancy Ellison
Richard A. Feeley
Ms.
Ms.
Ms.
Dr.
Mr. Jay Field
Mr. Skip Fox
Ms. Annette Frahm
Dr. Jack Gakstatter
Douglas Hotchkiss
Tom Hubbard
Mr. David Jamison
Mr. Richard Koch
Mr. Michael Kuntz
Ms. Jane Lee
Mr. Edward Long
Mr.
Mr.
Mr. Cliff Marks
Mr. Keith Rose
Ms. Clare Ryan
Mr. Carl Sagerser
Mr. Tim Sample
Mr. Martin Seybold
Mr. Dave Smith
Ms. MaryLou Soscia
Mr. Joe Talbot
Mr. Niel Thibert
Ms. Joan Thomas
Mr. John Underwood
Mr. Frank Urabeck
Mr. Fred Weinmann
Affiliation
Port of Seattle
U.S. Environmental Protection Agency
Muckleshoot Tribe
Washington Department of Ecology
Washington Department of Ecology
U.S. Environmental Protection Agency
Washington Department of Ecology
Washington Department of Ecology
National Oceanic and Atmospheric Adminis-
tration
National Oceanic and Atmospheric Adminis-
tration
National Oceanic and Atmospheric Adminis-
tration
Puget Sound Water Quality Authority
U.S. Environmental Protection Agency
Port of Seattle
Municipality of Metropolitan Seattle
Washington Department of Natural Resources
Washington Department of Ecology
Washington Department of Ecology
Seattle-King County Dept. of Public Health
National Oceanic and Atmospheric Adminis-
tration
City of Seattle
U.S. Environmental Protection Agency
U.S. Environmental Protection Agency
Washington Dept. of Social and Health
Services
Municipality of Metropolitan Seattle
King County
Washington Department of Ecology
U.S. Environmental Protection Agency
City of Seattle
City of Seattle
Washington Department of Ecology
U.S. Environmental Protection Agency
U.S. Army Corps of Engineers
U.S. Army Corps of Engineers
xv i i i
-------
Elliott Bay Citizen4s Advisory Committee
Name
Affi1i ati on
Ms. Janet Anderson
Mr. Douglas Briggs
Ms. Harriett Bullitt
Ms. Virginia Van Engelgen
Mr. Donald Hamilton
Mr. James Heil
Mr. Paul Hickey
Ms. Dee Ann Kirkpatrick
Mr. Minor Li 1e
Mr. Charles Muller
Mr. James Puckett
Mr. Tom Putman
Ms. Annette Ramsour
Mr. David Schneidler
Ms. Diana Swain
Mr. Terry Thomas
Mr. Mike White
Mr. Robert Williscroft
Alternates and Other Participants
Mr. Chris Luboff
Mr. Richard Rutz
Magnolia Community Club
Puget Sound Industrial Council
Friends of the Duwamish
League of Women Voters
Seattle Poggie Club
Puget Sound A11 i ance
Muckelshoot Indian Tribe
Suquamish Indian Tribe
Greater Seattle Chamber of Commerce
Sierra Club
Puget Sound Alliance
Seattle Audubon Society
Washington State Sports
Seattle Marine Business
Port Watch
Northwest Steelhead and Salmon Council
Northwest Marine Trade Association
Washington State Sports Diving Council
Western Washington Toxics Coalition
Seattle Audubon Sbciety
Diving Council
Coalition
xix
-------
EXECUTIVE SUMMARY
The goal of the Elliott Bay Toxics Action Program is to identify and
control toxic contamination in Elliott Bay and the lower Duwamish River
system. Under this program, federal, state, and local agencies cooperate to
respond to toxic contamination problems. Response actions may include
regulatory control of point and nonpoint sources of contaminants, and
treatment, capping, or removal of highly contaminated sediments.
In this report, the results of a field investigation (Tetra Tech 1985e)
of the nearshore region of the Elliott Bay/Duwamish River system are
evaluated to achieve the following objectives:
¦ Define spatial patterns and quantitative relationships of
sediment contamination, toxicity, and biological effects
¦ Identify problem areas of sediment contamination and asso-
ciated biological effects
¦ Rank problem areas relative to priority for evaluation of
potential contaminant sources and possible remedial action
¦ Identify potential problem chemicals (i.e.., chemicals that
display high concentrations in association with biological
effects).
A standardized assessment technique used in several other urban bay projects
(e.g., Tetra Tech 1985a; PTI and Tetra Tech 1988) was used to identify
problem sediments. Information presented in this report will be used in
conjunction with the results of a field survey of contaminant sources to
evaluate potential sources and develop recommended source controls in the
Elliott Bay/Duwamish River system. A revised action plan will be developed
based in part on an update of an interim work plan developed earlier in this
program (Tetra Tech 1985b). Although these data may not be sufficient alone
for initiating enforcement action against specific entities, they provide a
basis for demonstrating potential environmental effects of contaminant
sources. As such, they may be especially valuable for targeting priority
drainage areas for more focused investigations of sources and ultimately
initiation of source controls.
The use of these data in controlling contaminant sources or restoring
areas of contaminated sediments depends in part on coordination of the
activities of the Elliott Bay Action Team with other sediment programs. The
evaluation of problem areas as part of the Elliott Bay Toxics Action Program
was performed in the context of several other major sediment programs.
These include the following programs which are described in the main text:
the Puget Sound Dredged Disposal Analysis, the Harbor Island Superfund
Project, and the Sediment Criteria Element (P-2) of the Comprehensive
Management Plan for Puget Sound. The future use of the data presented in
xx
-------
this report to implement corrective actions will depend in part on ongoing
activities of the latter two programs.
The project area is described in the next section. The decision-making
approach for problem evaluation, the spatial distribution of contaminants
and effects in the.Elliott Bay system, and the ranking of problem areas are
described in the following sections.
PROJECT AREA
Elliott Bay is an embayment of approximately 21 km2 in central Puget
Sound, Washington. The project area includes the shallow areas (<20 m deep)
of the bay east of a line from Alki Point to West Point and the lower
Duwamish River upstream to the Head of Navigation (Figure ES-1). The inner
bay receives fresh water from the Duwamish River and most of the stormwater
runoff from about 67 km2 of highly developed land in metropolitan Seattle.
The lower Duwamish River is a salt-wedge estuary, influenced by tidal
action over a 16-km downstream reach (including all of the riverine habitat
within the project area). Saltwater intrusion occurs in the portions of the
Duwamish River within the project area for all river flow rates and tides.
Near its mouth, the river is divided by Harbor Island into the East and West
Waterways. In this area, and upstream to several kilometers beyond the Head
of Navigation, the river passes through heavily industrialized areas. The
entire Duwamish River drainage basin presently covers about 1,250 km2,
including large expanses of agricultural and forested land.
DECISION-MAKING APPROACH
The approach to identification and ranking of problem areas relies on
empirical measurements of the environmental hazard of contaminated areas.
The primary information used in the decision process includes:
¦ Sediment characteristics
Contaminant concentrations
Conventional physical/chemical characteristics (e.g.,
grain size distribution, sulfides, total organic carbon
content).
¦ Biological effects
Benthic invertebrate community structure
Sediment toxicity bioassays using amphipods
Concentrations of PCBs, chlorinated pesticides, and
mercury in muscle tissue of English sole
Prevalence of liver lesions in English sole.
xx i
-------
WEST POINT
©\\
Smith'
Cove
(D
¦Her 90/91
Denny Way
-Pier 70
(CDJ
LEGEND
1. MAGNOLIA
2. SEATTLE NORTH WATERPRONT
3. SEATTLE SOUTH WATERPRONT
4. NORTH HARBOR ISLAND
5. EAST WATERWAY
6. WEST WATERWAY
7. KEU00G ISLAND •
8. UPPER OUWAMtSH ESTUARY
9. OUWAM&H HEAO/ALM BEACH
SEATTLE
Terminal 37
ALKI POINT
'DUWAMISHV
HEAO
Fairmoum .
Avenue
West'
Waterway
• [>£asf
Waterway
Spokane Street
Washington
KeUogg island-
/ Slip 2
'Sip 3
/S0P4
Figure ES-1. Project location and study areas within Elliott Bay
and the lower Duwamish River.
'Slip 6
xxi i
-------
Tetra Tech (1984, 1985a, 1986e) described the rationale for selecting these
five major kinds of data to characterize the environmental hazard of con-
taminated sediments.
The environmental data on sediment contamination and biological effects
were organized into a matrix of biological and toxicological indices used to
compare study areas. This Action Assessment Matrix uses multiple independent
indices termed "elevations above reference" (EAR) to indicate the magnitudes
of contaminant levels and biological effects relative to reference condi-
tions. A decision to proceed with source evaluation and ranking of problem
areas is limited to sites that exceed a minimum action level defined by one
or more significant EAR. The action-level guidelines provide a consistent,
framework for defining toxic problem areas based on the weight of evidence
from evaluation of the selected hazard indicators. In the case of a single
significantly elevated index, the magnitude of the elevation must provide
sufficient evidence of a problem to outweigh the absence of significant
elevations in multiple indicators.
REFERENCE AREAS
Reference areas included nine reference embayments in Puget Sound
(including Port Susan) for sediment chemistry, Point Pully for fish pathology
and bioaccumulation, and Port Susan for sediment bioassays and benthic
infauna. Sediment chemical concentrations in Port Susan generally fell
within the range of other Puget Sound reference areas. Point Pully was
selected because of its proximity to Elliott Bay and the previously demon-
strated low prevalences of major liver lesions in English sole. Port Susan
was chosen to represent a relatively uncontaminated bay in east-central
Puget Sound with a major riverine input (i.e., the Stillaguamish River).
Data collected during this study generally confirmed the adequacy of Port
Susan and Point Pully as reference areas. Nevertheless, at Station PS-02 in
Port Susan, an observation of 24 percent mean mortality of amphipods
suggested that sediments at this site may be marginally toxic to amphipods
based on the criterion of >25 percent mortality proposed by Mearns et al.
(1986).
SEDIMENT CHEMISTRY
Sediment samples from over 100 subtidal and intertidal stations
(Figure ES-2) were analyzed for over 80 metals and organic compounds and
conventional variables. In general, chemical contamination was spatially
heterogeneous. The areas of the most severe contamination were localized,
suggesting the importance of local contaminant sources. The Seattle South
waterfront, the North Harbor Island area, and the West Waterway were among
the most severely contaminated study areas. Stations within these areas
accounted for many of the highest concentrations of metals and organic
compounds observed in the study (e.g., Stations SS-09, SS-08, SS-03, NH-03,
NH-04, NH-06, NH-08, WW-12, WW-14, and WW-19 in Figure ES-2). Other
stations with relatively high concentrations included EW-05 and EW-14 in the
East Waterway, Station AB-01 along Alki Beach, and Stations DR-08 and DR-12
in the upper Duwamish River. In contrast, stations in outer Elliott Bay
(i.e., along Magnolia Bluff and Alki Beach, excluding inner bay Station
AB-01) were the least contaminated overall.
xxiii
-------
STATIONS sampled in
PORT SUSAN DURING THE
ELLIOTT BAY SURVEY
)
MG-04
MG-G3
M&02
MG-01
port . %
¦ SUSAN
' notes
NS-07
SS-11
SS-10
ew-16
NHC2
EW-1S
NH04
5W-1Q
EW-11
EW»12< t
WW-17
WW-1S
WW-16
WW-13
WW-10
"EW-09
EWW7
WW-12
: ww-09
WVV-11
EW05
EW-04
EW-03
KO-IO
KG40
KO03
KO02
K041
0*17
OR-12
STITCHlOCAW
NOT 6MCT CU£ *C
IMPSC«L£
n
6000
a '•••
S me ton
2000
0*49-
-oaoe
-0*07
0JV06-
DR-OS -
Figure ES-2. Locations of sampling stations of sediment chemistry,
amphipod bioassay, and benthic macroinvertebrates.
DR-2S-
xxiv
-------
Concentrations of copper, lead, mercury, and zinc were among the most
elevated of the metals detected in the study area. Maximum EAR values for
these chemicals ranged from 320 (for copper and zinc) to 7,700 (lead),
whereas median EAR values were between 8 and 15. Concentrations of other
metals that were somewhat less elevated but nonetheless of concern included
arsenic (maximum EAR = 170), silver (maximum EAR =92), cadmium, chromium,
and nickel. Maximum copper concentrations [up to 2,050 mg/kg dry weight
(DW)] occurred near the mouth of the West Waterway (Areas NH and WW) and
along the Seattle South waterfront. Maximum lead concentrations (up to
71,100 mg/kg DW) occurred along the Seattle South waterfront and on the east
side of the West Waterway. Maximum zinc concentrations (up to 6,010 mg/kg
DW) occurred along the Seattle South waterfront. The highest mercury
concentrations occurred at relatively isolated stations [AB-01 (28.8 mg/kg
DW), NH-03, and EW-05] with generally elevated concentrations along the
Seattle South waterfront.
Polycyclic aromatic hydrocarbons (PAH) and polychlorinated biphenyls
(PCBs) occurred at the highest concentration and were the most frequently
detected of the organic contaminants. Other organic compounds occurred at
high concentrations at isolated stations. Maximum EAR values for PAH and
PCBs ranged from nearly 1,000 (for PCBs) to 15,000 (for low molecular weight
PAH, or LPAH) to over 40,000 (for high molecular weight PAH, or HPAH).
Median EAR for these compound classes ranged between 32 and 80. Other
organic compounds that were infrequently found at elevated concentrations
(maximum EAR >100) included 1,4-dichlorobenzene (maximum EAR = 8,900), benzyl
alcohol (maximum EAR = 880), 4-methylphenol, pentachlorophenol, butyl benzyl
phthalate, and retene. PAH concentrations were most elevated along the
Seattle South waterfront (up to 3,800,000 ug/kg DW total PAH), but were aTso
highly elevated in the North Harbor Island study area, the East and West
Waterways, and at isolated stations in other areas. PCB concentrations were
most elevated in the upper Duwamish River (up to 5,800 ug/kg DW) and along
the Seattle South waterfront, but high concentrations occurred throughout
the Duwamish River and in the North Harbor Island study area.
Pairwise Pearson correlations were performed to examine covariance in
the distribution of selected contaminants. Except among PAH and related
compounds, few correlations among chemicals were observed on a study-wide
basis. When correlations were performed on a smaller scale for individual
study areas, strong correlations were observed, particularly among metals.
These results are consistent with the presumed importance of sources that
predominate in localized areas.
Based on a quality assurance/quality control (QA/QC) review of the
chemical data, a substantial portion of the data set was qualified for
reasons of varying severity. The QA/QC review included control actions that
are not included in U.S. EPA Contract Laboratory Program or Puget Sound
Estuary Program guidelines for data review (e.g., in accordance with
guidance from U.S. EPA Region X, all organic compounds detected at
<1,000 ug/kg DW were qualified as estimates based on analyses of sediment
reference materials). The data quality issue that most affected data
analysis was high detection limits (i.e., low analytical sensitivity) for
certain organic compounds (most notably chlorinated phenols, chlorinated
benzenes, hexachlorobutadiene, benzyl alcohol, and benzoic acid). These
high detection limits resulted from low surrogate recoveries, as detection
xxv
-------
limits of acid/neutral compounds were adjusted on a chemical-specific basis
to reflect surrogate recoveries. Analysis of identified problem areas,
which had relatively high chemical concentrations, was largely unaffected by
the issue of detection limits and analytical sensitivity.
BIOACCUMULATION
Chemical analyses for PCBs, selected chlorinated pesticides, and
mercury were performed for 60 English sole muscle tissue samples (5 indi-
vidual fish were analyzed for each of 12 trawl stations, including the Point
Pully reference area). Tissue PCB concentrations were significantly
elevated above reference levels over much of the study area, with the
highest mean concentrations (390-470 ug/kg wet weight) observed in the
Duwamish River (especially in the East and West Waterways). PCB concentra-
tions tended to decrease with distance from the mouth of the Duwamish River.
Pesticide bioaccumulation was not of concern with the exception of p,p'-DDE,
which was detected in a single fish collected along the Seattle South
waterfront (410 ug/kg wet weight). Mercury bioaccumulation was not signifi-
cantly elevated above Point Pully reference levels in fish from any of the
Elliott Bay/Duwamish River trawl stations.
SEDIMENT TOXICITY BIOASSAYS
The acute toxicity of field-collected sediments was determined using
the amphipod (Rhepoxvnius abronius) bioassay. Sediments from 17 of the 102
project area sites tested caused significant mortality (P<0.001) in the
amphipod bioassay when compared with the Port Susan reference area. The two
most toxic areas in Elliott Bay were North Harbor Island (mean mortality = 60
percent) and the East Waterway (mean mortality = 43 percent). The range of
mean mortality within each study area was large, indicating considerable
spatial heterogeneity. For example, the mean mortality at a station ranged
from 9 to 100 percent within the North Harbor Island study area, and from
3 to 100 percent within the East Waterway. Sediment toxicity in the Magnolia
and Alki Beach areas was low (<7 percent mean mortality at each of seven
stations). One site (Station AB-01) just east of Duwamish Head displayed a
mean amphipod mortality of 47 percent. Overall, there was good agreement
between the present study and previous bioassay studies in Elliott Bay.
BENTHIC MACROINVERTEBRATE COMMUNITIES
Benthic macroinvertebrate samples from 78 stations in Elliott Bay and
Port Susan were collected, and specimens were identified to major taxonomic
levels and enumerated. Samples from 20 of the stations were further
identified to the lowest possible taxonomic level. Abundances of major taxa
at the Elliott Bay stations differed significantly (P<0.001) from the
abundances in the Port Susan reference area. Significantly enhanced
abundances (P<0.001) were detected in 78 comparisons, and significantly
depressed abundances were detected in 73 comparisons.
Of the 74 stations sampled for benthic infauna, one or more significant
depressions (P<0.001) in abundances of major taxa selected for problem
definition (i.e., Polychaeta, Crustacea, Pelecypoda, Gastropoda) were
detected at 40 of the stations. The most impacted areas in the Elliott Bay
study area were North Harbor Island, West Waterway, and the Kellogg Island
xxv i
-------
segment. Stations within those segments accounted for 73 percent of the
depressions detected among all Elliott Bay stations. Pelecypods and
crustaceans appeared to be the most sensitive indicators among the major
taxonomic groups.
The lowest numbers of taxa (at stations where organisms were identified
to the species level) occurred at Stations NH-03, EW-05, NS-08, and KG-01,
all of which exhibited two or more significant depressions (P<0.001) in the
abundances of major taxa. A normal classification analysis of the species-
level data separated stations into groups. Those groups corresponded to the
groups based on numbers of major taxa that exhibited significantly depressed
abundances (P<0.001) recorded among the stations.
FISH PATHOLOGY
Three major kinds of hepatic lesions (neoplasms, foci of cellular
alteration, and megalocytic hepatosis) found in livers of English sole were
used to characterize environmental conditions at 11 transect stations in the
study area. Although the exact causes of these lesions are unknown,
previous studies have demonstrated correlations between high lesion preva-
lences and toxic contamination of sediments. The spatial distributions of
the three kinds of hepatic lesions indicate that most abnormalities were
confined to inner Elliott Bay and the Duwamish River. The only study areas
without significant (P<0.001) elevation of the prevalence of at least one
kind of lesion were Alki Beach and Magnolia. Within inner Elliott Bay, most
serious abnormalities were confined largely to the. areas in or near the
Duwamish River. Comparisons of length-at-age between fish with and without
hepatic lesions suggested that the presence pf lesions may be associated
with reductions in fish growth only for females at ages greater than 5 years
old. Prevalences of neoplasms and foci of cellular alteration were corre-
lated positively (P<0.05) with fish age. Spatial distributions of lesion
prevalences found in the present study were consistent with historical data
collected by Mai ins et al. (1984).
CONTAMINANT, TOXICITY, AND BIOLOGICAL EFFECTS RELATIONSHIPS
Biological effects as measured by the amphipod toxicity bioassay and
significant reductions in abundances of benthic infauna taxa were generally
associated with higher concentrations of contaminants in sediments. Scatter-
plots of biological effects vs. concentrations of selected physical/chemical
variables were examined. A strong relationship with an apparent threshold
in the biological response of most benthic taxa evaluated was found for
selected chemicals: LPAH, HPAH, PCBs, copper, arsenic, cadmium, and
sulfides. Moreover, consistently low abundances of selected benthic taxa
were observed above the concentration of these chemicals generally corres-
ponding to Puget Sound Apparent Effect Thresholds (AET). These AET were
developed from independent data sets. For other organic compounds evaluated,
the number of stations with detected values was too small or the distribution
of the data too skewed to show a strong association with biological varia-
bles. The abundance of the pollution-tolerant polychaete Capitella capitata
displayed a linear increase with sulfides concentration in sediments and no
apparent relationship with copper, PCBs, or grain size. The magnitude of
amphipod mortality was generally high at higher concentrations of copper and
xxv i "i
-------
sulfides, but was not clearly correlated to total organic carbon (TOC)
content, grain size, or the other chemical variables evaluated.
Biological effects at several stations in Area SS along the Seattle
South waterfront were less than expected based on the severity of chemical
contamination at these sites (especially Stations SS-03, SS-08, and SS-09).
Other highly contaminated sites where statistically significant biological
effects were not found (P>0.001) included Stations EW-14 and AB-01.
However, at Station AB-01 the mean amphipod mortality was 47 percent; this
test result was not statistically significant (P<0.001) because of substan-
tial variability among replicates (a range of 25 to 80 percent mortality).
The abundances of several benthic taxa evaluated were consistently low
at stations with >50 percent amphipod mortality in the toxicity bioassay. In
contrast, £. capitata reached high abundances at stations with high levels of
amphipod mortality, but was relatively rare at stations with low amphipod
mortality. Concordance between statistically significant responses in the
toxicity bioassay and depressions of infaunal taxa among the 74 stations
where both were measured was not greater than that expected by chance alone.
However, lack of concordance at low to moderate levels of contamination is
not surprising given the wide range and levels of contaminants in the Elliott
Bay system and the different endpoints measured by these two indicators.
High concordance between these biological indicators was observed at highly
contaminated stations.
Linear correlations between PCB bioaccumulation and PCB sediment
contamination (i.e., wet weight bioaccumulation data vs. dry weight sediment
data and lipid-normalized .bioaccumulation data vs. TOC-normalized sediment
data) were not significant (P>0.05).
PRIORITIZATION OF PROBLEM AREAS AND CONTAMINANTS
The selected data for indicators of sediment contamination, toxicity,
and biological effects were integrated to evaluate toxic contamination
problems in the project area. In the first phase of problem identification,
all of the nearshore region inside a line from Pier 91 to Duwamish Head was
defined as a problem area (Tier I) based on significant elevations of
sediment contaminants and liver lesions in English sole. Seventy-two
stations were designated as higher priority problem stations (Tier II).
Tier I problem areas that were not also identified as Tier II areas are
considered lower priority problems that may require further evaluation of
potential sources and corrective actions at a future date. Sixty-one of the
Tier II stations were grouped into the following multi-station problem
areas: DRI (Slip 4), EW (East Waterway), KGI and KGII (near Kellogg
Island), NHI and NHII (North Harbor Island, the mouth of the West Waterway,
and west to just beyond Longfellow Slough outlet), NSI (Denny Way CSO), NSII
(Pier 90/91), SS (Seattle South waterfront), WWI (southern segment of West
Waterway), and WWII (northern segment of West Waterway).
Ranking of Tier II problem areas and stations identified five areas (SS,
NHI, NHII, WWI, and WWII) and 33 single stations as the highest priority
sites (Figure ES-3). Of the latter, the following stations were outside the
five highest priority areas: Stations NS-01, EW-05, AB-01, KG-01, KG-05,
xxvi ii
-------
NS-01 (B)
AMI 1<0
- 0R>16(8)
Note: Problem areas and single stations with
labels exhibited ranks >75% for chemistry (C),
biology (B), or both (C, 8). NS-01 and WW-02
were internal sites, which were ranked for
biology based on the amphipod bio assay only.
- DR-12(C)
Figure ES-3. Highest priority problem sites.
XX1 X
-------
KG-06, DR-12, DR-15, and DR-16. Each of the major problem areas is charac-
terized in the following sections.
Problem Area SS
This area is highly contaminated overall (especially by PAH), with
localized patches of extreme contamination. Stations SS-08 and SS-09 in this
area account for the highest concentrations of several metals (e.g., lead,
cadmium, zinc) and organic compounds (e.g., PAH, 1,4-dichlorobenzene)
observed in this study; maximum concentrations of other metals occurred at
other stations in this problem area (e.g., arsenic at Station SS-03).
Sediments in this area were enriched in organic matter, with TOC concen-
trations typically greater than 5 percent and as high as 27 percent.
Benthic infaunal depressions were found at Stations SS-04, SS-09, and SS-08,
with greater than 80 percent depressions of Pelecypoda at the first two
sites (PcO.QOl). Significant sediment toxicity to amphipods was found at
Station SS-06, where mortality was 45 percent (PcO.OOl). Overall, biological
effects in this area were confined to selected stations and did not exhibit
the widespread occurrence that might be expected based on consideration of
the chemistry data.
Problem Area NHI
Problem Area NHI, which encompasses two shipyard facilities at the
mouth of the West Waterway, was heavily contaminated with several metals
(most notably copper, but also mercury, lead, zinc, and arsenic), PAH, and
PCBs. The most elevated concentrations of organic compounds and metals were
observed near Station NH-03 and two historical stations located east of that
station. On the west side of the West Waterway mouth, Station NH-04
contained high concentrations of copper, pentachlorophenol, and PAH, among
other chemicals. Problem Area NHI exhibited severe depressions in the
abundances of major taxa of benthic infauna (especially Pelecypoda) at all
stations. Abundances of all four major taxa evaluated were severely
depressed at Station NH-03. Significant amphipod mortality above 85 percent
was found at Stations NH-03 and NH-04.
Problem Area NHII
PAH and related compounds (including alkylated PAH, carbazole, and
biphenyl) were the predominant contaminants in Problem Area NHII. The most
extreme contamination was observed at Stations NH-06 and NH-08, located near
the Wyckoff creosote facility and the outflow of Longfellow Slough, res-
pectively. These two stations were very similar in terms of PAH composition
and concentration. Benthic infaunal effects in Problem Area NHII were less
than in most other problem areas, although Station NH-08 exhibited severe
depressions of pelecypods and crustaceans. Overall sediment toxicity was
highest in this area, with Station NH-08 displaying 100 percent mortality.
Problem Area WWI
Chemical contamination was severe but patchy in this area, and included
a relatively isolated but high historical concentration of PCBs in the
southwest corner of the waterway, a high benzyl alcohol concentration at
Station WW-02, and high concentrations of PAH at Station WW-04. Effects on
-------
benthic infauna were moderate overall for this area. However, Station WW-03
was one of two stations in this study where severe (>80 percent) depressions
in abundances of all four major taxa were observed. Sediment toxicity in
Area WWI was generally low, except at Station WW-02, where 82 percent
mortality was observed.
Problem Area WWII
The most distinctive chemical feature of this area was extreme lead
contamination along the east side of the West Waterway (near the SW Lander
Street discharges). PAH concentrations were generally elevated, but
gradients or pronounced maxima were not apparent. PCB contamination was
moderate and patchy in this area. Severe effects on benthic infauna were
restricted to depressions in the abundances of pelecypods and crustaceans at
several stations within this area. Sediment toxicity was relatively low in
this area overall. Nevertheless, mean amphipod mortality above 40 percent
was observed at Stations WW-09 and WW-11.
r
xxx i
-------
INTRODUCTION
The U.S. Environmental Protection Agency (EPA) and the Washington
Department of Ecology initiated the Elliott Bay Toxics Action Program to
correct problems associated with toxic contamination of Elliott Bay and the
lower Duwamish River (Figure 1). Previous studies (e.g., Malins et al.
1980, 1982; Romberg et al. 1984) have demonstrated widespread contamination
of the nearshore region of the bay and river by a variety of toxic chemicals
associated with discharges of industrial and municipal wastewater, storm-
water, combined sewer overflows, and uncontrolled spills. A preliminary
evaluation of toxic problem areas (Tetra Tech 1986e) and a review of
existing plans for corrective actions (Tetra Tech 1985c) formed the basis
for development of an interim work plan for the Elliott Bay Toxics Action
Program (Tetra Tech 1985b). Under this program, remedial actions may
include source controls designed to reduce emissions of toxic chemicals and
possibly future cleanup of contaminated sediments.
An assessment of chemical contamination and adverse biological effects
is provided in this report, including a ranking of problem areas in terms of
priority for evaluation of potential contaminant sources. The field surveys
described in this report were designed to fill data gaps and to provide a
detailed assessment of environmental hazards. By identifying the worst
problem areas based on multiple indicators of chemical contamination and
bioeffects, this assessment allows resources for pollutant source investiga-
tions and remedial actions to be assigned to the highest priority -areas.
Nevertheless, the ranking of problem areas presented in this report may be
modified by consideration of the economic and technical feasibility of
potential remedial actions (e.g., Tetra Tech 1988c).
Development of a remedial action plan requires that the following kinds
of questions be answered for areas within the bay/river system:
1. What is the magnitude of sediment contamination?
2. What is the extent of contamination of aquatic organisms?
3. Has the contamination resulted in adverse biological effects?
4. Can the contaminant sources be identified?
5. Would remedial action reduce the threat to the environment?
Answering Questions 1-5 involves development of an information base, that
includes data on sources, fates, and effects of contaminants. Data on fates
and environmental effects of contaminants are presented in this report. The
evaluation of pollutant sources and an updated summary of remedial actions
developed by an interagency work group will- be presented in separate
documents. An assessment of human health risks associated with chemical
contamination of fisheries resources in the Elliott Bay system will be
performed as part of a Puget Sound-wide study funded by EPA.
1
-------
ENVIRONMENTAL
MONITORING PROGRAM
REVISED
ACTION PLAN
ACTION PROGRAM
EVALUATION
SEDIMENT
MANAGEMENT
INTERIM
ACTION PLAN
INITIAL PROBLEM
IDENTIFICATION
SEDIMENT REMEDIAL
ACTION
INSPECTIONS
ENFORCEMENT
SOURCE CONTROL
PERMITTING
POTENTIAL
ACTION TEAM
ACTIVITIES
NOTE: Analyses of environmental data used to define problem areas are presented in this report.
Figure 1. Elements of the Elliott Bay Toxics Action Program.
2
-------
The evaluation of problem areas as part of the Elliott Bay Toxics
Action Program was performed in context with the following major sediment
manaxement programs:
¦ Puget Sound Dredged Disposal Analysis (PSDDA)—A multiagency
program to develop evaluation procedures for disposal of
dredged material and to. identify disposal sites. A recent
baseline survey of the designated disposal site in inner
Elliott Bay analyzed sediment chemistry, benthic infauna
abundances, sediment toxicity, and bioaccumulation of toxic
chemicals by benthic infauna. PSDDA is not intended to
identify contaminated problem areas or proposed cleanup
actions.
¦ Harbor Island Superfund Study—As part of this U.S. EPA
Superfund project, contaminated sites on Harbor Island and
potential sources of contaminants to the East and West
Waterways and Elliott Bay are being identified. Stabilization
or cleanup of upland sites may be required to prevent offsite
transport of contaminants. During the first phase of the
Harbor Island Superfund study, U.S. EPA will also review
information generated by the Elliott Bay Action Program with
regard to contaminated sediments and potential sources of
contaminants. Future phases of the study may entail further
evaluation of contaminated sediments and, as appropriate,
plans for remedial action.
¦ Sediment Criteria Element (P-2) of the Comprehensive Manage-
ment Plan for Puget Sound—As directed under the plan
developed by the Puget Sound Water Quality Authority, the
Washington Department of Ecology is developing regulatory
criteria for identifying and designating sediments that have
observable acute or chronic effects on biological resources
or that pose a significant health hazard to humans.
The latter program is designed to identify problem sediments in the environ-
ment throughout Puget Sound. Interim sediment criteria will be proposed by
the Washington Department of Ecology by 31 July 1988, with adoption of final
regulations by 30 June 1989. Development of the sediment criteria will rely
partly on previous efforts to develop sediment quality values such as the
U.S. EPA urban bay action programs (including the present project), the
Commencement Bay Superfund Program, and PSDDA. A review of other plans and
activities related to water quality management in Elliott Bay is provided in
Tetra Tech (1985c).
The project area and its physical setting are described below. The
decision-making approach used to identify and prioritize contamination
problems is presented in the next major section. The second section also
provides a summary of the study design and methods used to collect field
data. Results of individual study components are presented in the third
section, including analyses of 1) sediment contamination, 2) benthic macro-
invertebrates, 3) sediment toxicity as measured by the amphipod bioassay
using Rhepoxvnius abronius. 4) bottomfish ecology, 5) histopathology of
3
-------
English sole (Paroohrvs vetulus) livers, and 6) bioaccumulation of toxic
substances in English sole muscle tissue. In the fourth section, relation-
ships among contamination, toxicity, and biological effects are evaluated.
Data for selected indicators are then integrated and evaluated within the
decision-making framework to develop a prioritization of problem areas and
contaminants.
SITE DESCRIPTION
Elliott Bay is an embayment of approximately 21 km2 in central Puget
Sound, Washington (Figure 2). The bay opens toward the main basin of Puget
Sound to the west, with Seattle situated on the surrounding shore. The inner
bay receives fresh water from the Duwamish River, which is channelized in the
last 10 km of its downstream reach. Near its mouth, the river is divided by
Harbor Island into the East and West Waterways. In this area, and upstream
to several kilometers beyond the Head of Navigation, the river passes through
the heavily industrialized portions of Seattle. The Duwamish River drainage
basin presently covers about 1,250 km2 (Santos and Stoner 1977). The major
commercial district of Seattle is located on the northeastern shore of
Elliott Bay. Much of the remainder of the shoreline is covered by residen-
tial development or recreational areas.
Drainage patterns and study areas within the bay/river system are
described in the following sections. Background information on physical
oceanography and beneficial uses of the bay is provided in Tetra Tech
(1986e).
DRAINAGE PATTERNS
The immediate drainage basin of Elliott Bay and the lower Duwamish
River consists of about 67 km2 of highly developed land in metropolitan
Seattle. The wider Duwamish/Green River basin includes large expanses of
agricultural land and undeveloped forests. Boundaries of the immediate
drainage basin considered in this study are roughly defined by Beacon
Avenue on the east side and 35th Avenue S.W. on the west side. The basin
includes residential areas in the southern portions of the Queen Anne and
Magnolia neighborhoods, and most of West Seattle; the industrial areas along
the Duwamish Waterway; the 1-5 corridor from James Street to about S. Dawson
Street; and the downtown business district.
The residential areas are generally served by partially separated storm
and sanitary systems. Much of the surface runoff from streets and surround-
ing land surfaces is collected in storm sewers and discharged directly to
Elliott Bay or the lower Duwamish River. Runoff from rooftops is generally
discharged into the combined sewer system and treated at the three area
wastewater plants. Runoff from the business district is served mainly by
combined sewers, and is transported to the West Point plant via Municipality
of Metropolitan Seattle's (Metro) interceptor system. Runoff from 1-5 is
collected in two large storm drains and discharged to the Duwamish Waterway
at Slip 4 and Diagonal Way. The remaining industrial areas, excluding
Harbor Island, are served by combined sewers, and private and municipal
storm drains. Presently, several storm sewer separation projects are
underway or are planned for business/industrial areas. Harbor Island has
its own storm sewer system which discharges to the East and West Waterways.
4
-------
WEST POINT
MAGNOLIA
SEATTLE NORTH WATERFRONT
SEATTLE SOUTH WATERFRONT
NORTH HARBOR ISLAND
EAST WATERWAY
WEST WATERWAY
KELLOGG ISLANO
UPPER DUWAMISH ESTUARY
OUWAMISM HEAO/AUO BEACH
Smith*
Cove
Pier 90/91
Denny Way
Pier 70
SEATTLE
Terminal 37
Fairmount
Avenue
. ALKI POINT
Spokane Street
Kellogg Island
Slip 2
-Slip 3
.Slip 4
SOOO
¦Slip 6
Figure 2. Project location and study areas within Elliott Bay
and the lower Duwamish River.
5
-------
The Duwamish River discharges at an annual average rate of 47 m^/sec
and contributes much of the suspended particulate matter entering Elliott
Bay. The Duwamish River is a salt-wedge estuary, influenced by tidal action
over its lower 16 km (including all of the riverine study area which extends
to the Head of Navigation located approximately 10 km upstream from the
mouth of the West Waterway). Saltwater intrusion occurs in the portions of
the Duwamish River within the project area for all river flow rates and
tides. The leading edge of the saltwater intrusion, called the wedge toe,
is defined as the farthest point upstream where salinity of the wedge water
is at least 25 ppt. Stoner (1967) found that when river discharge was less
than 28 m3/sec, the toe of the salt wedge did not intrude past 12.6 km (East
Marginal Way Bridge), but it intruded at least that far on most flood tides
when discharge was less than 18 m3/sec. During some periods of low discharge
and high tides, salt has been observed as far upstream as 16.4 km, and on
rare occasions 21 km (Stoner et al. 1975). The salt wedge and overlying
river water were fairly discrete (highly stratified) at river discharge
rates above 28 nr/sec, but at lower rates, the first 6 km of the estuary
were partly mixed.
STUDY AREAS
A major objective of this report is to identify spatial patterns in the
distribution of contaminants, sediment toxicity, and biological responses in
the nearshore region of Elliott Bay and the lower Duwamish River. Sediment
sampling was limited to areas of approximately 50 m water depth or less
because previous studies (e.g., Mai ins et al. 1980, 1982; Romberg et al.
1984, 1987) have demonstrated that the most contaminated sediments are
typically in these relatively shallow areas. To facilitate spatial analysis,
the nearshore region (i.e., less than about 50 m deep) has been divided into
nine smaller areas based on geographic features and locations of major
sources of contaminants [i.e., storm drains and combined sewer overflows
(CSOs) (Figure 2)]. Area boundaries and major features are as follows:
1. Magnolia (M6) - West Point, south to Smith Cove
2. Seattle Waterfront North (NS) - Terminal 90/91 to Pier 70;
Interbay CS0 at Terminal 90/91, Denny Way CS0, Myrtle Edwards
public fishing pier
3. Seattle Waterfront South (SS) - Pier 70 to Terminal 37; main
Seattle waterfront, ferry terminals, King Street and Connec-
ticut Street CSOs, Seattle Aquarium, public fishing pier
4. North Harbor Island (NH) - Southern end of Elliott Bay from
T37 west to Fairmount Avenue; northern Harbor Island, outflow
of Longfellow Slough
5. East Waterway (EW) - Mouth to Spokane Street bridge; Hanford
and Lander CSOs
6. West Waterway (WW) - Mouth to Spokane Street bridge; S.W. Lan-
der and S.W. Florida CSOs
6
-------
7. Kellogg Island (KG) - Spokane Street bridge to Kellogg Island/
Slip 1; Hanford-1 CSO
8. Upper Duwamish Estuary (DR) - Kellogg Island/Slip 1 to head of
navigation; Michigan Street CSO, Georgetown flume
9. Duwamish Head/Alki Beach (AB) - Eastern shoreline of Duwamish
Head, north of Fair-mount Avenue S.W., to Alki Point
In this report, the phrase "Elliott Bay system" refers to the entire project
area defined above.
-------
METHODS
Descriptions of the field, laboratory, and data management methods are
provided in the following sections. Before these methods are presented, the
general approach used to define and rank toxic problem areas is summarized.
The decision-making approach provides perspective on the uses of the various
kinds of data that were collected and the rationale for specific variables.
DECISION-MAKING FRAMEWORK
Information on the extent of toxic contamination and adverse environ-
mental effects formed the basis for prioritization of areas for cleanup or
source control. The decision-making framework developed for the Elliott Bay
Toxics Action Program incorporates a preponderance-of-evidence approach to
problem identification. Study areas that exhibit high values for indices of
contamination and adverse effects relative to a reference site receive a
ranking of "high priority" for evaluation of pollutant sources and remedial
action.
Overview
The decision-making process to evaluate toxic contamination problems is
shown in Figure 3. This process ultimately focuses regulatory effort and
remedial actions on localized areas of severe contamination and biological
effects. Major steps in the process to identify and rank toxic chemical
problems are as follows:
¦ Characterize sediment contamination, sediment toxicity, and
biological effects
¦ Quantify relationships among sediment contamination, sediment
toxicity, and biological effects
¦ Apply action levels to determine problem areas
¦ Determine problem chemicals in problem areas
¦ Define spatial extent of problem areas
Evaluate sources contributing to problem areas
Evaluate, prioritize, and recommend problem areas and sources
for potential remedial action.
Four major premises underlie this approach. First, it was determined
that no single measure of environmental conditions could be used in all
cases to define adequately the requirements for potential remedial action.
Therefore, recommendations for remedial action investigations are based on
several measures of sediment contamination and biological effects. When
results of these independent measures corroborate one another (i.e., there
8
-------
CHARACTERIZE:
CHEMICAL CONTAMINATION
SEDIMENT TOXICITY
BIOLOGICAL EFFECTS
*
DETERMINE ELEVATION ABOVE
REFERENCE (EAR)
4
ASSEMBLE ACTION ASSESSMENT
MATRICES
;
APPLY ACTION LEVEL GUIDELINES
r
i
Q
U
A
N
IDENTIFY STUDY AREAS AND
SEGMENTS OF CONCERN
3
i
!
I
T
i
T
DEFINE EXTENT OF PROBLEM
A
T
AREAS WITHIN STUDY AREAS
AND SEGMENTS
s
V
*
|
b
n
RANK PROBLEM AREAS
(WORST CONDITIONS)
s
|
B
L
i
:
i
A
T
IDENTIFY POTENTIAL PROBLEM
1
CHEMICALS IN PROBLEM AREAS
0
N
S
H
RANK PROBLEM CHEMICALS
P
I
S
NHINI
CONDUCT SOURCE EVALUATIONS
J*
*
FINAL PRIORITIZATION OF
PROBLEM AREAS FOR
REMEDIAL ACTION
i s
! f
Figure 3. Decision-making approach for the Elliott Bay Action
Program.
9
-------
is a preponderance of evidence), a problem area is defined. There may be
special circumstances where corroboration is not needed because a single
indicator provides an exceptional basis for recommending that the site be
evaluated for potential source control or remedial action.
Multiple contamination and effects variables were used in the preponder-
ance-of-evidence approach for several reasons. First, chemical data are
needed to characterize the extent of contamination and to relate contaminant
distributions to potential sources. Second, direct measures of biological
effects are needed because some chemicals present at high concentrations may
not be bioavailable. Conversely, biological effects may result from
chemicals that are not analyzed for. Several biological variables (e.g.,
mortality of amphipods exposed to sediments in the laboratory, abundances of
indigenous benthic infauna organisms, and liver lesions in English sole)
were used to account for effects at various levels of biological organization
(i.e., cellular/tissue, whole organism, community). Also, different
biological variables may demonstrate different sensitivities to specific
chemicals or combinations of chemicals.
Second, action-level guidelines provide a consistent means for defining
problem areas based on significant contamination and effects. These
guidelines are developed in consultation with the Elliott Bay Work Group, an
interagency task force consisting of representatives from federal, state,
and local agencies.
Third, it is assumed that adverse biological effects are linked to
environmental conditions that result from source emissions and that these
links may be characterized empirically. Relationships between sources, and
effects will be quantified where possible, for example, by correlations of
specific contaminant concentrations and distributions with the occurrence of
adverse biological effects (see Tetra Tech 1988b). However, proof of
specific causal agents is not provided by these studies. Direct cause-effect
relationships in the sense of laboratory verification studies are not within
the scope of the Elliott Bay investigation. Nevertheless, analysis of a
wide range of contaminants should increase the probability of measuring
either the causative substances, or related substances from the same source
and with the same distribution in the environment.
Finally, a fourth premise is that the recommended remedial actions may
vary from location to location. For example, only removal of contaminated
sediments may be recommended where contamination originated only from past
sources and biological effects are apparent. In contrast, source control
may be recommended where contamination originates from an ongoing source even
though biological effects may not be apparent. In other cases, both sediment
removal and source control may be recommended. To prevent recontamination
of newly cleaned areas, sediment remedial actions should be implemented only
after sources have been controlled.
Chemical. Biological, and Toxicolooical Indices
The primary kinds of data used in the decision-making process are listed
below:
10
-------
¦ Sediment Quality
Contaminant concentrations
Apparent effects thresholds (AET), relating sediment
contamination and predicted biological effects
¦ Bioaccumulation
Contaminant concentrations in muscle tissue of English
sole
¦ Sediment Toxicity
Amphipod mortality (10-day bioassay)
¦ Benthic Infauna
Polychaete abundance
Crustacean abundance
Pelecypod abundance
Gastropod abundance
¦ Fish Pathology
Lesion prevalence in livers of English sole.
The rationale for using the five general kinds of data is provided in Tetra
Tech (1985a, 1986e). Puget Sound AET were used as sediment quality values to
evaluate chemical data relative to predicted biological effects (see
PRIORITIZATION OF PROBLEM AREAS AND CONTAMINANTS). Although many other
variables were evaluated throughout the decision-making process, those shown
above formed the basis for problem identification and priority ranking.
Justification for specific variables is provided below and in later sections
(see RESULTS for individual data types).
Target Chemicals--
A list of chemical contaminants analyzed for in sediments collected
during the Elliott Bay studies is given in Table 1. Most of the substances
on this list have at least one of the following two properties: they can
bioaccumulate, possibly with adverse biological effects in the food chain if
bioaccumulated, or they can produce adverse biological effects even when not
bioaccumulated. EPA priority pollutants that may be currently or histor-
ically discharged into the study area are included on the list. Compounds
not on the EPA list of priority pollutants also have been considered on the
basis of their local significance. Several conventional sediment quality
variables were measured [e.g., total organic carbon (TOC) content, grain
size]. These conventional variables provide a means of comparing areas with
different bulk chemical or physical properties. Also, observed biological
effects could result from a characteristic of the system unrelated to the
selected organic compounds or metals of concern (e.g., the deleterious
effects of sediment anoxia on benthic communities).
11
-------
TABLE 1. LIST OF CONTAMINANTS AND CONVENTIONAL
VARIABLES FOR ANALYSIS IN ELLIOTT BAY PROJECT
Low molecular weight PAH
naphthalene
acenaphthylene
acenaphthene
fluorene
phenanthrene
anthracene
High molecular weight PAH
fluoranthene
pyrene
benz(a)anthracene
chrysene
benzof1uoranthenes
benzo(a)pyrene
i ndeno(1,2,3-c,d)pyrene
dibenzo(a,h)anthracene
benzo(g,h,i jperylene
Total PCBs
Neutral halogenated compounds
1.2-dichlorobenzene
1.3-dichlorobenzene
1.4-dichlorobenzene
1,2,4-trichlorobenzene
hexachlorobenzene (HCB)
2-chloronaphthalene
trichlorobutadienes3'®
tetrachlorobutadi enes3'°
pentachlorobutadienes3'®
hexachlorobutadi ene
Phthalate esters
dimethyl phthalate
diethyl phthalate
di-n-butyl phthalate
butyl benzyl phthalate
bis('2-ethylhexyl )phthalate
di-n-octyl phthalate
Pesticides
p,p'-DDE
p,p'-DDD
p,p'-DDT
aldrin
chlordane
dieldrin
endrin
endrin aldehyde
heptachlor
alpha-HCH
beta-HCH
delta-HCH
gamma-HCH (lindane)
Phenols
phenol
2-methylphenol3
4-methyl phenol4
2,4-dimethylphenol
2-chlorophenol
2,4-dichlorophenol
4-chloro-3-methylphenol
2,4,6-trichlorophenol
2,4,5-trichlorophenola.
pentachlorophenol (PCP)
Miscellaneous extractable
compounds
1-methylphenanthrene3'°
2-methylphenanthrene3'®
3-methylphenanthrene3'"
2-methylnaphthalene3
biphenyl3'"
di benzoth i ophene3'®
dibenzofuran3
benzyl alcohol3
benzoic acid3
carbazole3»b
coprostanol3*"
alpha-tocopheryl acetate3'"
retene3*®
isophorone
Volatile organic compounds
acrolein
acrylonitrile
benzene
bromodichloromethane
bromoform
bromomethane
carbon tetrachloride
chlorobenzene
12
-------
TABLE 1. (Continued)
chloroethane
2-chloroethylvinyl ether
chloroform
chloromethane
d i bromoch1oromethane
dichloromethane
1.1-dichloroethane
1.2-dichloroethane
1.1-dichloroethene
trans-l,2-dichloroethene
1.2-dichloropropane
cis-l,3-dichloropropene
trans-1,3-di chloropropene
ethyl benzene
1,1,2,2-tetrach1oroethane
tetrachloroethene
1.1.1-trichloroethane
1.1.2-trichloroethane
trichloroethene
toluene
total xylenes3
vinyl chloride
Metals
antimony
arsenic
cadmi um
chromium
copper
i rona
lead
manganese®
mercury
nickel
selenium
silver
zinc
Conventional variables3
total organic carbon
total solids
percent fine-grained material
total nitrogen
water-soluble sulfides
total volatile solids
oil and grease
a Not a U.S. EPA priority pollutant. Chemicals without footnotes
are U.S. EPA priority pollutants; no sediment conventional variables
are priority pollutants.
b Tentatively identified organic compounds.
13
-------
The target contaminants measured during the Elliott Bay project have the
potential to cause observed sediment toxicity or biological effects.
However, the ability to identify poorly-understood chemical interactions
(e.g., synergism and antagonism) is limited. Although interactive effects
may not be distinguishable from other kinds of effects, they may be measured
through the use of biological indicators explained below.
Biological Variables-
Selection of individual biological and toxicological variables was
based on the following considerations:
¦ Analysis of several levels of potential biological effects
Bioaccumulation at the tissue level
Pathology at the tissue level
Acute lethality of test sediments to whole-organisms
Chronic effects at the community level in benthic infauna
¦ Use of each variable in past Puget Sound studies
¦ Documented sensitivity of each variable to contaminants
¦ Ability to quantify each variable within the resource and time
constraints of the program.
Response variables were selected to characterize several important toxic
effects in resident organisms of Elliott Bay. Although a study of effects
on fish populations was beyond the scope of the current project, a study of
effects on individual fishes is possible through an assessment of liver
lesion prevalence. Benthic macroinvertebrates were selected because of
their sensitivity to sediment contamination, their importance in local
trophic relationships, arid their ability to establish site-specific response
gradients relative to sediment contamination.
The use of Rhepoxvnius abronius to determine the acute lethality of
field-collected sediments has been documented by numerous authors (e.g.,
Swartz et al. 1982, 1985; Chapman et al. 1982a,b; Mearns et al. 1986). This
amphipod species is a sensitive indicator of contaminated areas both by its
absence, from some natural populations in such areas (Swartz et al. 1982;
Comiskey et al. 1984), and by its response to contaminated sediments in
laboratory studies (Swartz et al. 1985).
As described below, the chemical, biological, and toxicological data
were used to develop indices of environmental quality. These indices
allowed evaluation of the data relative to relatively uncontaminated
reference areas.
14
-------
Elevation Above Reference
Indices
Environmental quality indices were developed to rank areas based on
observed contamination and biological effects. The indices have the general
form of a ratio between the average value of a variable at a site in the
Elliott Bay system and the value of the same variable at a reference site.
The ratios are structured so that the value of the index increases as the
deviation from reference conditions increases. Thus, each ratio is termed
an Elevation Above Reference (EAR) index. For most variables, the measured
average value at the study site is divided by the value at the reference
area to obtain the EAR. For benthic infauna, EAR are derived as the inverse
ratio of values (i.e., reference divided by Elliott Bay site) to reflect the
magnitude of adverse effects on benthic populations. Chemical effects on
infauna are expected to be manifested as decreases in taxa abundance
relative to reference. An increase in the EAR for infauna would therefore
reflect a decrease in absolute value of the variable but an increase in
adverse effect relative to reference conditions.
It should be noted that these indices were not used in lieu of the
original data (e.g., contaminant concentrations), but in addition to these
data. The original data were used to identify statistically detectable
increases in sediment contamination, sediment toxicity, or biological
variables, and to determine quantitative relationships among these variables.
The EAR indices were used to reduce large data sets into interpretable
numbers that reflect the magnitudes of the different variables among areas.
EAR were used to rank sites based on synoptic biological and chemical
data from the 1985 Elliott Bay Action Program survey. In the absence of
matched biological and chemical data (i.e., for historical sediment chemistry
data), existing Puget Sound AET were used to evaluate sediment chemistry
data and identify problem chemicals. Historical chemistry data were also
used to establish boundaries of problem areas. The use of sediment chemistry
EAR (rather than existing Puget Sound AET) to rank problem areas ensured
that high chemical contamination independent of observed biological effects
was taken into account in ranking sites for evaluation of potential sources.
The use of site-specific biological effects data for Elliott Bay provided
complementary information for identifying and ranking problem sediments.
Action Assessment Matrix
The environmental contamination and effects indices (i.e., EAR) were
organized into an Action Assessment Matrix used to compare study areas or
stations. A simplified hypothetical example of an Action Assessment Matrix
is shown in Table 2. This example matrix is presented to demonstrate how
information from multiple indicators can be integrated for an overall
evaluation and prioritization of different study areas. For this example,
only general indices such as "sediment contamination", or "benthic macro-
invertebrates" are used. In the actual application of the approach, multiple
indices for specific types of sediment contamination were evaluated,
including separate measures for organic compounds and metals (see PRIORITIZA-
TION OF PROBLEM AREAS AND CONTAMINANTS). Similarly, the benthic macroinver-
tebrates category was replaced by more specific measures of benthic community
structure.
15
-------
TABLE 2. THEORETICAL EXAMPLE OF ACTION ASSESSMENT MATRIX®
EAR Values for Study Sites
Indicator
B
Reference Value
Sediment contamination
Toxicity
Bioaccumulation
Pathology
Benthic macroinvertebrates
1,300
5.2
4.0
45
2.6
1.2
800
8.0
5.0
75
875] 2.0 | 10.01 [4~5l
900l r~20] 11,1001 [200]
2.8
1.3
8 1,000 ppb
2.2 10% mortality
13 10 ppb
2.0 5% prevalence
1.1 60 individuals/m^
a EAR values for indicator variables are shown for Sites A-E. Benthic macroinvertebrate
factors represent the reduction in numbers of individuals at the study site relative
to the reference site. Factors for all of the other indices represent increases
relative to the reference site values shown.
- Indicates indicator value for the specified area is significantly different from
reference value.
16
-------
Evaluation of infprmation in this format enables the decision-maker to
answer the following questions:
¦ Is there a significant increase in sediment contamination,
sediment toxicity, or biological effects at any study site? .
¦ What combination of indicators is significant?
¦ What are the relative magnitudes of the elevated indices
(i.e., which represent the greatest relative hazard)?
The term "significant" is generally used in this report to mean statis-
tically significant at the 99.9 percent confidence level (alpha = 0.001).
Because replicate data for sediment chemistry were not collected at every
station, an alternative criterion for significance was developed. Following
the approach used earlier in Commencement Bay and Elliott Bay (Tetra Tech
1985a, 1986e), a significant elevation of a chemical concentration in
sediments was defined as exceedance of the maximum concentration of that
chemical in all Puget Sound reference areas.
Action-Level Guidelines
The decision to evaluate potential sources of contamination and the need
for possible remedial alternatives applies only to those sites that exceed a
minimum action level. An "action level" is a level of contamination or
effects that defines a problem area. It is assumed that an area requires no
action unless at least one of the indicators of contamination, toxicity, or
biological effects is significantly elevated above reference levels.
The action levels used to define problem areas in the Elliott Bay system
are shown in Table 3. The action-level guidelines are summarized as follows:
¦ Significant EAR for THREE OR MORE INDICES identifies a
problem area requiring evaluation of sources and potential
remedial action
¦ For ANY TWO INDICES showing significant elevations, the
decision to proceed with source and remedial action evalua-
tions depends on the actual combination of indices and the
degree to which they are site-specific
¦ When only a SINGLE INDEX is significantly elevated, a problem
area may be defined when additional criteria are met (i.e.,
the magnitude of the index is sufficiently above the signifi-
cance threshold to warrant further evaluation).
It is conceivable (but not likely) that significant sediment toxicity or
biological effects occur in areas without apparent contamination by toxic
substances. In such cases, it would be important to evaluate the possibility
that the observed conditions result from variables not measured in the field
studies. An attempt would be made to distinguish the biological problem area
from surrounding areas using chemical characteristics, and to identify
sources based on these distinguishing chemical characteristics.
-------
TABLE 3. ACTION-LEVEL GUIDELINES
Condition Observed
Threshold Required for Action
I. Any THREE OR MORE significantly
elevated indices3
II. TWO significantly elevated indices
1. Sediments contaminated, but below
HAET and 90th percentile PLUS:
Bioaccumulation elevated relative
to that at the reference area, OR
Sediment toxicity with no more
than 40% mortality, OR
Benthic community structure
indicates altered assemblage,
but less than 80% depression
2. Sediments contaminated but
below HAET and 90th percentile
PLUS elevated fish pathology
III. SINGLE significantly elevated index
1. Sediment contamination
2. Sediment toxicity
3. Benthic community structure
4. Fish pathology OR bioaccumulati on
Threshold exceeded, continue with source
and remedial action evaluation.
No immediate action. Recommend site for
future monitoring.
Threshold for source evaluation exceeded
if elevated contaminants are considered
to be.biologically available. If not,
recommend site for future monitoring.
If the magnitude of contamination exceeds
the 90th percentile for all study areas
or the HAET, proceed with source and
remedial action evaluation.
Greater than 40% response (mortality).
80% depression or greater.
Insufficient as a . sole indicator.
Recommend site for future monitoring.
Check adjacent areas for significant
contamination, toxicity, and/or bio-
logical effects.
a Combinations of significant indices are from independent data types (i.e., sediment
chemistry, bioaccumulation, sediment toxicity, benthic infauna, fish pathology).
Significant indices are defined as follows:
Sediment Chemistry = Chemical concentration at study site exceeds highest value
observed at all Puget Sound reference areas.
Sediment Toxicity, Benthic Infauna, Bioaccumulation, and Pathology = Statistically
significant difference between study area and reference area (PcO.OOl) at one
or more stations within area.
18
-------
Ranking of Problem Areas
Ranking of problem areas was based on a systematic method of assigning
scores to sampling sites based on the significance and severity (i.e., EAR)
of the various chemical and biological variables. Criteria for scoring
problem areas in terms of priority for evaluation of sources and remedial
actions are shown in Table 4. Based on these criteria, higher priority
would be assigned to an area with many elevated indices (i.e., EAR) than to
an area with few. Because the values of the individual indices arie assumed
to represent relative environmental hazards, areas with higher values of the
indices are scored higher. Two ranking schemes were used. The first used
sediment chemistry indicators only, primarily to characterize the extent and
magnitude of contamination. The second used all biological indicators to
measure the response to chemical contamination. For ranking based on
biological variables, scores for bioaccumulation and pathology were assigned
to each subtidal station based on trawl data for the corresponding area in
which the station was located. Biological scores for intertidal stations
were based only on the results of the amphipod bioassay since other biologi-
cal variables were not measured at intertidal stations. Scores assigned to
a station for individual biological indicators (i.e., bioassay, infauna,
bioaccumulation, pathology) were summed to obtain an overall biological-
effects score for the station. The total biological-effects score for a
station was normalized to the maximum possible score attainable with the
available data. This normalization step was necessary to avoid biasing
ranks for some study areas towards lower values just because certain data
were missing. The maximum possible score for biological effects when all
variables were measured was 16 (= sum of 4 for amphipod bioassay, 4 for
benthic macroinvertebrates, 4 for bioaccumulation, and 4 for pathology).
The range of possible normalized scores for biology is from 0-1 (or 0-100
expressed as a percentage). The various areas were then ranked according to
the magnitudes of their overall biological-effects scores.
Similarly, scores assigned to the sediment chemistry indicators (i.e.,
metals and organic compounds) were summed and normalized to the maximum
possible score to obtain an overall chemical-contamination score for each
station. The maximum possible score for sediment chemistry was 8 (= sum
of 4 for metals and 4 for organic compounds). A ranking of problem stations
was then developed based on their relative chemical-contamination scores.
Total chemical and biological scores for each multi-station problem area
were calculated as the averages of the corresponding total scores for
individual stations within the area. Separate rankings of multi-station
problem areas were developed based on the total chemical and biological
scores. If the final ranking based on biological effects for a single
station or a multi-station problem area differed substantially from that
based on sediment chemistry, then the higher-ranking score was given
precedence. Thus, some high priority sites were designated strictly on the
basis of chemical contamination (i.e., no corresponding biological problems
apparent) or strictly on the basis of biological conditions (i.e., no
chemical contamination apparent).
19
-------
TABLE 4. SUMMARY OF SCORING CRITERIA FOR SEDIMENT
CONTAMINATION, TOXICITY, AND BIOLOGICAL EFFECTS INDICATORS
Indicator
Criteria
Score
Metals (one or more)
Concentration not significant
0
Significant; EAR <10
1
Significant; EAR 10-<50
2
Significant; EAR 50-<100
3
Significant; EAR >100
4
Organic Compounds
Concentration not significant
0
(one or more)
Significant; EAR <10
1
Significant; EAR 10-<100
2
Significant; EAR 100-<1,000
3
Significant; EAR >1,000
4
Toxicity3
No significant bioassay response
0
Amphipod bioassay significant
2
>40% response in bioassay
4
Macroi nvertebrates^
No significant depressions
0
1 significant depression
1
2 significant depressions
2
>3 significant depressions
3
>95% depression for >1 variable
4
Bioaccumulation
No significant chemicals
0
(fish muscle)
1 significant chemical
1
2 significant chemicals
2
>3 significant chemicals
3
EAR >50 for >1 chemical
4
Fish Pathology^
No significant lesion types
0
1 significant lesion type
1
2 significant lesion types
2
>3 significant lesion types
3
>5% prevalence of hepatic neoplasms
4
a Toxicity based on amphipod mortality bioassay.
b Variables considered were polychaete abundance, crustacean abundance, gastropod
abundance, and pelecypod abundance.
c Lesions considered were hepatic neoplasms, foci of cellular alteration
(preneoplastic nodules), and megalocytic hepatosis.
20
-------
Quantitative Relationships
The development of quantitative relationships among possible causative
factors, sediment toxicity, and benthic effects identifies threshold
concentrations above which changes in the biological indicators are detect-
able. The basic concept of increased biological effects or sediment
toxicity resulting from increased concentrations of a single chemical in
sediments is depicted in Figure 4. Four study areas that have statistically
elevated effects are shown in the figure. Although there is an elevation in
contamination relative to reference conditions at four of the remaining five
study areas, there are no statistically detectable increases in the effect
indicator above background conditions. Thus, the level of sediment contami-
nation corresponding to Area X (arrow) represents an apparent threshold
above which significant effects occur. The contamination of sediments by
multiple chemicals may result in a more complex relationship than the
example in Figure 4. Such relationships are discussed in detail later (see
CONTAMINANT, TOXICITY, AND BIOLOGICAL EFFECTS RELATIONSHIPS).
AET have been identified using synoptic chemical and biological data
sets from throughout Puget Sound (Tetra Tech 1986c, 1987). Puget Sound AET
are used in this study to identify potential problem areas based on chemical
data collected in other studies [e.g., Metro, National Oceanic and Atmos-
pheric Administration (NOAA)] where appropriate biological data were
unavailable. The concept of AET and the data sets used in deriving AET are
explained below. The Elliott Bay data collected during this study are being
compiled along with other recent data from Puget Sound into the EPA Sediment,
Quality Values database (3EDQUAL). This database will be used as part of an
ongoing study to update Puget Sound AET.
The focus of the AET approach is to identify concentrations of con-
taminants that are associated exclusively with sediments having statistically
significant biological effects (relative to appropriate reference sediments).
Thus, to generate AET values, chemical data are classified according to the
absence or presence of significant biological effects to determine concentra-
tions of contaminants above which statistically significant biological
effects would always be expected to occur. AET were originally developed to
identify problem sediments in the Commencement Bay Nearshore/Tideflats
Remedial Investigation (Tetra Tech 1985a). AET have been subsequently
revised with an expanded database (200 stations) and their accuracy has been
evaluated using biological and chemical data for geographically diverse
areas of Puget Sound (Tetra Tech 1986c, 1987). The AET method and accuracy
tests in Puget Sound are described in detail in those documents.
AET have been established for 64 organic and inorganic toxic chemicals
using matched chemical and biological data for several biological indicators
and embayments in Puget Sound. Because of patchy biological and chemical
conditions in the environment, it was important that chemical analyses be
performed on the same or nearly the same sediment that was used in bioassays
and benthic infaunal analyses. AET are available for predicting significant
effects based on the following biological indicators:
¦ Depressions in abundances of major taxonomic groups of benthic
infauna (i.e., Crustacea, Mollusca, Polychaeta, and total
abundance)
21
-------
^Area Z
(
<
t
c
0
**
u
•
^Area X
at
«
3
¦o
>
^ Area Y
^AreaW
A A
A
° A
Average Reference Index
0 0 ^
Sediment Concentration
of Contaminant '
0 Reference
A Elliott Bay, not statistically significant
A Elliott Bay, statistically significant at
the 99.9% confidence level ( a - 0.001)
Figure 4. Theoretical example of relationship between sediment
contamination and an effects index.
22
-------
¦ Amphipod mortality bioassay using Rhepoxvnius abronius
¦ Oyster larvae abnormality bioassay using Crassostrea giqas
¦ Microtox bioluminescence bioassay using Photobacterium phos-
phoreum.
For each chemical, a separate AET was developed for each biological indi-
cator, resulting in four sets of AET. A list of the different AET generated
thus far for Puget Sound is provided in Table 5, along with the lowest AET
(LAET) and highest AET (HAET) among' these four indicators. The derivation
of these AET is described in more detail in Tetra Tech (1986c).
The AET method has been shown to be sensitive in correctly predicting
impacted stations in Puget Sound, but in doing so the approach also predicts
impacts at some stations that do not exhibit adverse effects (i.e., the
approach is not completely efficient in only identifying impacted stations).
Because the objective for using AET in this study was to identify potential
problem chemicals and problem areas (in conjunction with established action
level guidelines), the ability to correctly predict all impacted stations
(sensitivity) is more important than the ability to predict only impacted
stations (efficiency).
Spatial Resolution of Effects
Using the Action Assessment Matrix, contamination and effects were
analyzed at several levels of spatial resolution (e.g., study areas within
the project area or individual stations). Detailed examination of each
sampling station was necessary because spatial heterogeneity of sediment
contamination was relatively high.
Quantitative relationships among sediment contaminants, sediment
toxicity, and benthic macroinvertebrates were examined to evaluate small-
scale response gradients. AET were used to predict the occurrence of
biological problems at stations where chemistry data were available but
biological data were not.
Source Evaluation
The objective of source evaluation is to identify sources of contamina-
tion, and in turn to guide remedial activities. A limited evaluation of
sources is presented in this report based upon the spatial distribution of
contamination, the geochemical properties of observed contaminants, and
characteristics of known or potential sources. A more complete evaluation
of sources will be presented in a separate report (Tetra Tech 1988b).
OVERVIEW OF FIELD STUDY DESIGN
The general design of the field study is described in this section. A
summary of data types and samples collected in the Elliott Bay system and the
reference area is shown in Table 6. All of the Elliott Bay and initial Port
Susan data were collected during September-October 1985. A second sampling
of Port Susan occurred during September-October 1986 as part of the Everett
23
-------
TABLE 5. PUGET SOUND AET (DRY WEIGHT)a>b
(ug/kg dry weight for organic compounds; mg/kg dry weight for metals)
Chemical
Amphipod
AETC
Oyster
AET®
Benthic
AETe
Microtox
AETf
LAET
HAET
Low molecular weight PAH
• 5,5009**?*1
5,200
6,1001
5,200
5,200
6,100
naphthalene
2,400h'1
2,100
2,100.
2,100
2,100
2,400
acenaphthylene
560 .
>560
6401
>560
560
640
acenaphthene
980h»1
500
500.
500
500
980
fluorene
l,800f'i
540
640]
540
540
1,800
phenanthrene
5,400"'," .
1,500
3,200]
1,500
1,500
5,400
anthracene
l,9009'h'1
960
1,300!
960
960
1,900
High molecular weight PAH
38, OOOj? 1
17,000
>51,000]
12,000
12,000
38,000
fluoranthene
9,800^1
2,500
6,300]
1,700
1,700
9,800
pyrene
11,000"*]
3,300
>7,300!
2,600
2,600
11,000
benz(a)anthracene
3,000"']
1,600
4,500!
1,300
1,300
4,500
chrysene
5,oooh»!
2,800
6,700!
1,400
1,400
6,700
benzof1uoranthenes
3,700
3,600
8,000!
3,200
3,200
8,000
benzo(a)pyrene
2,400 .
1,600
6,800!
1,600
1,600
6,800
indeno(l,2,3-c,d)pyrene
880"'1
690
>5,200!
600
600
880
d i benzri(a, h)anthracene
510h'i
230
1,200!
230
230
1,200
benzo(g,h,ijperylene
860h'!
720
5,400!
670
670
5,400
Total PCBs
2,500!
1,100
1,100
130
130
2,500
Total chlorinated benzenes
6801
400
400
170
170
680
1,3-dichlorobenzene
>170
>170
>170
>170
1,4-dichlorobenzene
260
120
120
110
110
260
1,2-dichlorobenzene
>350
50
50
35
35
50
1,2,4-trichlorobenzene
51
64
64
31
31
64
hexachlorobenzene (HCB)
130
230
230
70
70
230
Total phthalates
>5,200,1 .
3,400
>70,000!
3,300
3,300
3,400
dimethyl phthalate
>700"»1
160
160 .
71
71
160
diethyl phthalate
>l,200h'1
>73
200"'1
>48
200
200
di-n-butyl phthalate
>5,100
1,400
>5,100
1,400
1,400
1,400
butyl benzyl phthalate
>470
>470
470
63
63
470
bis(2-ethylhexyl)phthalate
>3,100.
1,900
1,900
1,900
1,900
1,900
di-n-octyl phthalate
>5901
>420
>68,000!
--
Pesticides
p,p*-DDE
15
¦
9
9
15
p,p'-DDD
43
2.
2
43
p,p'-DDT
3.9
>6
II1
3.9
11
24
-------
TABLE 5. (Continued)
Chemical
Amphipod
AETfc
Oyster
AET"
Benthic
AETe
Microtox
AETf
LAET
HAET
Phenols
670h»i
phenol
420
1,200
1,200
420
1,200
2-methylphenol
63
63
>72
>72
63
63
4-methylphenol
l'200h •
670
670
670
670
1,200
2,4-dimethylphenol
>72h»1
29
29
29
29
29
pentachlorophenol
>140
>140
>140
>140
2-methoxyphenol
930
930
930
930
930
930
Miscellaneous extractables
hexachlorobutadiene
290
270
270
120
120
290
1-methy1phenanthrene
310
370
370
370
310
370
2-methylnaphthalene
670
670
670
670
670
670
biphenyl
260
260
270
270
260
270
dibenzothiophene
240
240
250
250
240
250
dibenzofuran
540
540
540
540
540
540
benzyl alcohol
73
73
73
57
57
73
benzoic acid
>690
650
650
650
650
650
N-ni trosodi phenyl ami ne
220
130
75
40
40
220
Volatile organics
tetrach1oroethene
>210
140
140
140
140
140
ethyl benzene
>50
37
37
33
33
37
total xylenes
>160
120
120
100
100
120
Metals
antimony
5.3
26
3.2
26
3.2
26
arsenic
93
700
85
700
85
700
cadmium
6.7
9.6
5.8
9.6
5.8
9.6
chromium
>130.
>37
59
27
27
59
copper
800]
390
310
390
310
800
lead
700]
660
300
530
300
700
mercury
2.1!
0.59
0.88
0.41
0.41
2.1
nickel
>120]
39
49
28
28
49
silver
>3.7]
>0.56
5.2
>0.56
5.2
5.2
zinc
8701
1,600
260
1,600
260
1,600
a ">" indicates that a definite AET could not be established because the highest concentration
occurred at a station without biological effects (hence, it is not clear from available data
if biological effects always occur above this concentration, as specified in the definition
of AET). For the purposes of problem identification in Elliott Bay, these values were excluded
when LAET (low AET) and HAET (high AET) were generated.
b The following data sets were used to generate the AET in this table:
1. Battelle (1986)
2. Chan et al. (1985, unpublished)
3. Comiskey et al. (1984)
4. Osborn et al. (1985)
25
-------
TABLE 5. (Continued)
5. Romberg et al. (1984)
6. Tetra Tech (1985a)
7. Tetra Tech (1986d)
8. Trial and Michaud (1985)
9. U.S. Department of the Navy (1985).
c Based on 160 stations.
d Based on 56 stations (all from Commencement Bay Remedial Investigation).
e Based on 104 stations.
f Based on 50 stations (all from Commencement Bay Remedial Investigation).
9 A higher AET (24,000 ug/kg for low molecular weight PAH and 13,000 ug/kg for anthracene)
could be established based on data from an Eagle Harbor station. However, the low molecular
weight PAH composition at this station is considered atypical of Puget Sound sediments because
of the unusually high relative proportion of anthracene. Thus, the low molecular weight PAH
and anthracene AET shown are based on the next highest station in the data set.
h The value shown exceeds the Puget Sound AET established in Tetra Tech (1986c) and results
from the addition of Eagle Harbor Prelimi nary Investigation data (Tetra Tech 1986d).
1 The value shown exceeds AET established from Commencement Bay Remedial Investigation data (Tetra
Tech 1985a) and results from the addition of Puget Sound data presented in Tetra Tech (1986c).
26
-------
TABLE 6. SUMMIT OF FIELD STUDY DESIGN
No.
Stations
Indicator
Variables
Sample Type
No.
Replicates
Elliott
Bay
Reference®
Sediment chemistry
selected chemicals
and conventional
variables
composite 0-2 cmc
1
102
5
Toxicity bioassay
X mortality
composite 0-2 cmc
5
102
4
Benthic infauna
major taxa abundances
species abundances
0.1-m? grab
0.1-nr grab
5
5
74
16
4
4
Bioaccumulation
PCBs, pesticides,
mercury
English sole
(>22 cm) fillet
5
11
1
Fish histopathology
lesion prevalences
English sole liver
1*
11
1
a Reference area for sediment chemistry, bioassay, and benthic macroinvertebrates was Port Susan.
Reference area for. bioaccunulation and fish pathology was Point Pully.
b See Table 1.
c Chemistry and bioassay samples were aliquots of the same composite sample.
** 60 fish per sample.
27
-------
Harbor Action Program. Results of the 1986 sampling were compared with the
1985 results and used to evaluate Port Susan as a reference area. Similar
methods were used during both years for Port Susan variables discussed in
later sections of this report.
Station Locations
The locations of stations sampled during the Elliott Bay project are
presented in Figures 5 and 6. Subtidal stations for sampling of sediments
were located in shallow water [generally 5-15 m (corrected to mean lower low
water; Appendix B)] near shore. Detailed information on station locations
is provided in Appendix B, and large-scale maps are included in the source
evaluation report (Tetra Tech 1988b). Trawl stations for sampling English
sole (Parophrvs vetulus) were located near areas sampled for sediments at
water depths generally <33 m.
The rationale for station locations is provided in the samplinq and
analysis plan for the Elliott Bay Action Program (Tetra Tech 1985e) and
sections below. Briefly, stations were selected to:
¦ Fill data gaps from previous studies
¦ Define known areas of contamination more precisely
¦ Determine large-scale gradients of contamination and
bioeffects in relation to known sources
¦ Detect localized areas of contamination and bioeffects near
potential sources.
Note that the triad of sediment chemistry, amphipod bioassay, and benthic
infauna was sampled at all sediment stations except those in the upper
Duwamish River (Stations DR-01 to DR-17 and DR-25 in Figure 5). The
Duwamish River stations were not sampled for benthos because of the likeli-
hood of estuarine gradients in community composition that would confound
interpretation of the data (Tetra Tech 1985e). Based on the availability of
an extensive database for the area offshore of the Denny Way CSO (e.g.,
Romberg et al. 1984), U.S. EPA limited the sampling effort for the present
study to a single intertidal station (i.e., NS-01 in Figure 5). At the time
of the interim action plan (Tetra Tech 1985b), U.S. EPA concluded that
sufficient data were available to classify the Denny Way CSO area as a high-
priority problem area.
Port Susan was selected as a reference area for sediment chemistry, bio-
assays, and benthic variables. Port Susan was chosen to represent a bay in
east-central Puget Sound with a relatively unurbanized watershed and a major
riverine inflow (i.e., the Stillaguamish River) (Tetra Tech 1985e).
Previous studies (Malins et al. 1980) indicated that sediments of Port Susan
were relatively uncontaminated compared to Elliott Bay and other urbanized
embayments of Puget Sound. Point Pully was used as a reference area for
bioaccumulation and histopathology. Point Pully was chosen because of its
proximity to Elliott Bay and because previous studies (Landolt et al. 1984)
demonstrated low prevalences of major liver lesions in English sole from this
area.
28
-------
STATIONS SAMPLED IN
PORT SUSAN OURING THE
ELLIOTT BAY SURVEY
)
MG-04
MG-03
M&02
U&01
' PORT
SUSAN
*»SS-i
\
SnTCNLDCATCNS .
•or gjucroosrs
IMP SCALE
NM7
-SS03
Nmlt
NHC2
NH-03 —
EW-15
WW-20
EW-U
ew-t3
NH-10
WW-19
EW-12
WW*1B
WWM4
EW-07
ew-10
ew-n
WW-17
ww-is
WW»16
WW-1J
ww-to
NK-08
W+07
KG-oe
K&07
KG-tO
KO-09
• PH
\ v».«"
KQ-03
KO-02
KGWl
» 0fVt6
-DR-15
I—OR-13
•0R-12
-OR-11
ii
0M6-
0R-0S -
6000
a imi
Figure 5. Locations of sampling stations for sediment chemistry,
amphipod bioassay, and benthic macroinvertebrates.
Ofi-25 -
29
-------
MAGNOLIA
BLUFF
NS-91
SS-92
TACOMA
NH-91
NH-92
DUWAMISH .
HEAD
EW-91
HARBOR
ISLAND
WW-91
ALKI POINT
KG-91
POINT PULLY
(PP-91)
DR-91
kilometers
Figure 6. Locations of trawl transects in Elliott Bay.
30
-------
In riearshore areas of Elliott Bay and in the Duwamish River, station
locations were determined by line-of-sight fixes on stationary shoreline
features. In offshore areas of Elliott Bay and in Port Susan, LORAN C navi-
gational coordinates were recorded for each station. Wherever possible, a
variable range marker (VRM) was used with LORAN C to determine ranges
between two reference points, or to determine distances to shore objects. In-
addition, photographic records of al-1 position alignments and ranges were
made at all stations, and depth soundings were recorded. Station positioning
methods were sufficiently accurate to define locations within a 15-m radius
at most stations and within a 8-m radius at stations that were located in
areas where the vessel could be tied to a stationary object. The research
vessel was anchored at the stations whenever possible during the survey, and
station locations were verified before each sample was collected.
Data Analysis Methods
Chemical Contamination—
The magnitude and spatial extent of contamination of sediments was
determined by comparisons of chemical concentrations among Elliott Bay study
areas and with Puget Sound reference conditions. EAR values were calculated
using Carr Inlet data, and the "significance" of EAR values was assessed by
comparison to the maximum values reported in Puget Sound reference areas
(including Carr Inlet and Port Susan).
Averaged data from six Carr Inlet stations sampled in 1984 were used to
calculate EAR for Elliott Bay and Duwamish River sediments. Recent Carr
Inlet data were used as the basis for calculating EAR for the following
reasons:
¦ The most complete reference data set is available for Carr
Inlet, including synoptic data for metals, a wide range of
organic compounds, grain size, organic carbon content, and
other conventional variables
¦ The lowest detection limits for most substances of concern in
Puget Sound embayments are available for Carr Inlet
¦ EAR for chemical variables measured in other urban embayments
(e.g., Commencement Bay) have been calculated with these
data, and therefore, will be directly comparable with the
values calculated for this study
¦ In almost all cases where chemicals were detected in multiple
reference areas, the Carr Inlet samples had comparable or
lower values. Therefore, Carr Inlet is considered repre-
sentative of Puget Sound reference conditions.
Because replicate data for sediment chemistry were not collected at
every station, tests for statistically significant differences between
Elliott Bay/Duwamish River samples and reference area samples could not be
conducted. Instead, data from a wide range of Puget Sound reference areas
(collected from 1976 to 1986., including the Port Susan data from the present
31
-------
study) were used as the criteria for determining whether EAR were "signifi-
cant" (i.e., whether the contamination exceeded all Puget Sound reference
conditions). Port Susan data from this study were added to reference data
compiled previously (Tetra Tech 1985a). If a chemical was not detected in a
reference area sample, detection limits were used to define reference
conditions for that chemical. Detection limits greater than 50 ug/kg dry
weight (DW) for organic compounds in Port Susan sediments were excluded from
the reference area concentration ranges to minimize the bias resulting from
less sensitive chemical analyses. Such detection limits observed in other
studies were previously excluded from Puget Sound reference area data (Tetra
Tech 1985a).
Pairwise Pearson correlations (Sokal and Rohlf 1981) were performed to
examine covariance in the distribution of selected contaminants. These
correlations were performed on a study-wide basis and for most individual
study areas. Only those contaminants that were detected four or more times
in at least one study area were selected for analysis. Evaluation of
correlations focused on chemical pairs that had correlation coefficients
>0.7. Scatterplots (i.e., graphs of the concentration of one chemical vs.
that of another chemical) were examined, and correlations that were apparent-
ly driven by anomalously high chemical concentrations at single stations
were not reported.
The relatively high detection limits reported for a number of organic
compounds in this study necessitated several special measures during data
analysis. Low molecular weight polynuclear aromatic hydrocarbon (LPAH) and
high molecular weight polynuclear aromatic hydrocarbon (HPAH) sums in thi.is
study do not include detection limits except for reference area.(Port Susan)
samples. For most samples, detection limits would not affect the PAH sums,
but certain samples had detection limits sufficiently high to produce sums
that would be misleading. PAH detection limits for Port Susan samples (all
less than 50 ug/kg DW) were retained in PAH sums to maximize consistency
with data treatment for other reference areas.
High detection limits observed for a number of semivolatile [acid/neu-
tral] target compounds in this study would have resulted in 50th percentile
(median) and 90th-percentile concentrations that were likely not representa-
tive of chemical conditions in the study area. For this reason, detection
limits greater than 100 ug/kg DW for semivolatile organic compounds and
PCBs, or greater than 25 ug/kg DW for single component pesticides, were not
used in determining percentile concentrations. However, detection limits
were not excluded from the analysis of spatial patterns of contamination.
Biological Effects-
Selected biological variables within each of the four categories of
data (i.e., amphipod bioassay, benthic infauna, bioaccumulation, and liver
histopathology) were used to test for statistical differences between study
area stations and the reference area (Port Susan or Point Pully). Use of
statistical criteria ensured that between-site differences were judged
objectively. The statistical design used to test for significant differences
between control and test stations adjusted the individual error rate for
multiple comparisons. This reduced the probability of a Type I error (i.e.,
the probability of defining a significant difference from reference when
32
-------
none actually exists). The null hypothesis was that the mean value of a
variable at the test station was equal to the mean value of that variable at
the reference station. This null hypothesis was tested versus several
alternative hypotheses, depending on the biological variable being tested.
In environmental studies, control of the Type I error rate becomes
increasingly important as the regulatory and legal consequences of incorrect-
ly identifying a difference between mean values become important.
Correction of the error rate for multiple comparisons was necessary
because the repetitive use of data collected at a control station results in
non-independence among the pairwise comparisons (Winer 1971). If the
individual error rate for each comparison is not corrected, then the
probability of falsely identifying at least one significant difference
between the test and reference stations within the entire data set increases
with the number of pairwise comparisons made. For example, if the selected
pairwise significance level is P=0.05, and 20 hypotheses are tested (each
using the same reference data), then the probability that all of the
significant differences identified in the 20 comparisons are correct is
(0.95)20, or 35.8 percent. This probability decreases as the number of
tests increases.
In these studies, the number of pairwise comparisons varied from 11 to
102 among the four biological variables. Selecting an experimentwise error
rate of P=0.05, and dividing that rate by the number of comparisons would
have yielded individual error rates between P=0.005 to P=0.0005. However, a
pairwise comparison error rate of P=0.001 was selected for all four vari-
ables,* for two reasons. .First, a significance level of P=0.001 was suffi-
ciently conservative to assure with 90.3 to 98.9. percent probability that
all identified significant differences were true differences. Second, use
of the same error rate for tests of all four biological variables ensured
comparability among the test results.
For the benthic macroinvertebrate data, an unpaired two-sample t-test
was used to test for a statistically significant difference (P<0.001) between
each study site {n-S) and the pooled reference stations (n=20 for the four
stations combined). Before applying the parametric tests, the Fmax test was
used to determine whether the sample variances were homogeneous for each
paired comparison. Because the variances were heterogeneous in most of the
pairwise comparisons, the abundances of infaunal taxa (i.e., polychaetes,
crustaceans, pelecypods, and gastropods) used to test for between-site
differences were log-transformed [logio (x+l)]» If the variances of the log-
transformed data were heterogeneous between the study site and the reference
site (Fmax test; P<0.05), an approximate t-test was applied to the data
(Sokal and Rohlf 1981). Comparisons of benthic infaunal assemblages among
stations in Elliott Bay and among Elliott Bay, Port Susan, and other
reference areas within Puget Sound are also described. Based on species-
level data for 20 stations within the Elliott Bay system, faunal similarities
among stations were determined using a normal classification analysis. All
data were log transformed [logjo(x+l)] prior to analysis. The classification
analysis involved two steps. First, similarity values were generated for
all possible pairs of stations included in the analysis using the Bray-
Curtis Similarity Index (see Boesch 1977). This index uses both species
composition and the abundances of the individual species to estimate
between-site similarity. The group average clustering strategy was then
33
-------
applied to the matrix of similarity values to generate a dendrogram of
stations, from which groups of stations (i.e., stations that are most
similar in species composition and abundance) were determined.
For the amphipod (Rhepoxvnius abronius) bioassay data, a two-sample
analysis of variance (ANOVA), which is statistically equivalent to a t-test,
was used to test for a significant difference (P<0.001) between each study
site station (n=5) and the pooled reference stations (n=20 for the four
stations combined). Following the aipproach of Meams et al. (1986), the data
were transformed using an arcsine transformation only when the variances were
heterogeneous. Homogeneity of variances was tested using the Fmav test
(P<0.05). The mean mortality and 95 percent confidence limits for reference
area samples from Port Susan were compared with similar statistics for other
reference areas used during previous studies. Amphipod mortality in Elliott
Bay sediments was compared with reference-area mortality rather than with
native-sediment controls (i.e., West Beach, Whidbey Island) to account for
possible effects of simply removing amphipods from their native sediments.
There is evidence that the fi. abronius bioassay is subject to limited
grain-size effects and attempts have been made to quantify and compensate for
these effects (DeWitt et al. in press). However, the present data showed no
predictive relationship between amphipod mortality and grain-size (percent
fines) within Elliott Bay, Port Susan, or the combined data set (see RESULTS,
Sediment Bioassays). Accordingly, the data were analyzed without any
correction for possible sediment grain size effects.
Lesion prevalence and male proportion (i.e.,-an index of fish sex ratio)
were compared between Elliott Bay transects and Point Pully using the G-test
of independence (P<0.001) with Williams' correction factor (Sokal and Rohlf
1981). Fish age was compared between Elliott Bay transects and Point Pully
using the Mann-Whitney U-test. The similarity among the prevalences of all
three lesions across all transects was determined using Kendall's coefficient
concordance (W). Within Elliott Bay, the proportion of males having each
kind of hepatic lesion was compared with the corresponding proportion of
females using the G-test of independence with Williams' correction factor.
Positive association between the prevalence of each kind of lesion and fish
age was tested using Spearman's coefficient of rank correlation (rs).
Length-at-age was compared between fish with and without hepatic lesions
using the Mann-Whitney U-test.
For bioaccumulation, pairwise t-tests were used to test for statistical-
ly significant (P<0.001) elevations of PCB and mercury tissue concentrations
at each Elliott Bay/Duwamish River trawl station (n=5) relative to the Point
Pully reference station (n=5). Before applying the t-tests, the Fmax test
was used to determine whether sample variances were homogenous for each
paired comparison. Because variances were heterogeneous for PCBs in most of
the pairwise comparisons, the PCB data were logip-transformed. Variances for
mercury data were homogenous and no transformation was required. The effect
of lipid normalization on the data distributions was examined for mercury
and PCB bioaccumulation.
34
-------
SEDIMENT CHEMISTRY
Field Sampling
Field collection procedures followed the recommendations of the Puget
Sound Estuary Program (PSEP) (Tetra Tech 1986f). Full details of the
sampling design and techniques are provided in Tetra Tech (1985d,e).
Sediment samples were collected during September-October 1985 using a chaio-
rigged van Veen grab sampler with a cross-sectional area of 0.1 m2.
Following deployment, the closed grab was retrieved and placed in a sampling
tray. The hinged lids of the van Veen sampler were opened to inspect the
sample.
Care was taken to ensure recovery of an intact surface sediment layer,
with four major criteria used for rejection of a sample;
¦ Overflowing sediments, with sediment touching the top of the
closed cover
¦ Water leaking from the sides or bottom (i.e., indicating that
the interstitial water in the sample was being flushed with
overlying seawater), or visible scour of the sediment surface
near the edges of the sampler
¦ Turbid water overlying the sediments
¦ Insufficient penetration depth.
After the field supervisor determined maximum penetration depth and
sample acceptability, qualitative observations were recorded on field log
sheets for sediment color, odor, texture, and the presence of recognizable
organisms. An HNu photoionization detector was used to monitor all sediment
samples for harmful vapors.
When it was determined that the grab triggered incorrectly, that the
sample was disturbed, or that some of the sample was lost, a new sample was
taken. More than one grab at the same station sometimes was necessary to
obtain an acceptable depth of penetration. In medium to coarse sand, a
minimum penetration depth of 5 cm was considered acceptable. In fine sand
and sandy silt, a penetration depth of 7 cm was the minimum acceptable
depth. When attempts to sample a station were unsuccessful, another nearby
station was selected and documented. Standardized collection data (i.e.,
collection date and time, station location, depth, and replicate number)
were recorded for each sample.
Once onboard, samples were held in a vertical position by blocks and the
overlying water was carefully drained off by an aspirator hooked to the
ship's hose. Subsamples for volatile organic analyses were taken by placing
40-cm3 glass vials (duplicates) at the undisturbed sediment surface and
filling them using a stainless steel spatula. Subsamples for sulfide
analysis were immediately removed from the sampler and placed in a weighed
plastic container with 50 mL of sulfur antioxidant buffer (SAOB).
35
-------
The remaining subsamples were taken from a homogenized sample. Each
sampling horizon from the upper 2 cm of sediment away from the edge of the
van Veen sampler was carefully removed with a stainless steel spatula, trans-
ferred to a stainless steel bowl, and homogenized by stirring with a
stainless steel spoon (sampling equipment was rinsed with site water and
methylene chloride between sampling events). Samples were stirred until
uniform color and textural homogeneity were observed. Homogenized subsamples
were collected as follows:
¦ 500 cm^ was transferred to a precleaned glass jar with a
tetrafluoroethene (TFE) cap liner for organic analysis
¦ 125 cm^ was transferred to a precleaned glass jar for metals
analysis
¦ 100 cm-* was transferred to a freon-rinsed glass jar for oil
and grease analysis, total organic carbon and nitrogen
¦ 100 cm^ was transferred to a Ziploc bag for grain size
analysis
¦ 1,500 cm3 was transferred to a polyethylene bag for bioassays.
Intertidal samples were collected from shore using a stainless steel
spatula. Otherwise, intertidal and subtidal sediment samples were processed
and analyzed in similar fashion. All sediment samples for chemical analysis
were stored on ice on the ship and were transported directly to the labora-
tory (for sulfide and volatile organic compound analyses) or were stored
frozen before delivery to the laboratory (for the remaining chemical
analyses). Chain-of-custody procedures were followed for all samples.
Laboratory Analysis for Metals
The following 11 of the 13 EPA priority pollutant metals were analyzed
for all 107 sediment samples for this study: antimony, arsenic, cadmium,
chromium, copper, lead, mercury, nickel, selenium, silver, and zinc. The
remaining two priority pollutant metals, beryllium and thallium, were not
analyzed because historical data did not suggest that these metals were of
concern in the study area. Iron and manganese were also analyzed. Analyses
were performed at Battelle Northwest Marine Research Laboratory in Sequim,
Washington.
Samples were prepared by thawing the frozen sediment, homogenizing,
freeze-drying, and grinding each sample. The sample was then either
subjected to a total acid digestion for atomic absorption (AA) analysis, or
pressed into a pellet for x-ray fluorescence (XRF) analysis. Total acid
digestion was performed by combining 1 mL of 4:1 nitric acidrperchloric acid
(HNO3:HC1O4) with a 0.200-gram sample in a polytetrafluoroethylene (PTFE)
bomb at 130° C for 4 hours. After the sample was cooled, 3 mL of hydro-
fluoric acid (HF) was added and the bomb was heated overnight at 130° C.
After cooling, 20 mL of 2.5 percent boric acid (H3BO3) was added and the
bomb was heated again at 130° C for 8 hours. After the weight and volume of
the digestate were determined, the solution was analyzed for silver,
cadmium, antimony, and selenium by Zeeman graphite furnace atomic absorption
36
-------
(GFAA) using the method of standard addition for calibration. Mercury was
determined on aliquots by cold vapor atomic absorption (CVAA). The mercury
detector was calibrated with standard solutions.
XRF was used to quantify arsenic, chromium, copper, iron, manganese,
nickel, lead, and zinc. Thin film standards (Neilson 1977) were used to
calibrate the XRF analyzer. Although the analysis of these metals by XRF
differs from the methods specified in the quality assurance project plan
(Tetra Tech 1985d), it is stated in PSEP protocols (Tetra Tech I986f) that
XRF may be used if accuracy and precision can be demonstrated to the levels
specified by the program.
Laboratory Analysis for Semi volatile Organic Compounds (Acid/Neutral and
Pesticide/PCBl
Analyses for 37 of the 57 EPA priority pollutant semivolatile
(acid/neutral) organic compounds, six additional Hazardous Substance List
compounds, PCBs, and 13 priority pollutant chlorinated pesticides were
performed on all 107 sediment samples collected for this study. Priority
pollutant bases and halogenated ethers were not included in the analyses, in
accordance with recommendations of PSEP (Tetra Tech 1986f). The excluded
compounds have been detected infrequently in Puget Sound studies. Gas
chromatograph/mass spectrometer (GC/MS) searches for additional non-priority
pollutant compounds or compound classes (i.e., tentatively identified organic
compounds) were also performed on all sediment extracts; these compounds
were, specified to the laboratory prior to analysis. Analyses were performed
by Science Applications International Corporation in La Jolla, California.
The procedure used to analyze acid/neutral compounds was derived from
Tetra Tech (1986a). Briefly, the extraction and cleanup procedure for
acid/neutral compounds included addition of 38 stable isotope-labeled
surrogate compounds to sediment samples, 24-hour Soxhlet extraction with
methylene chloride/ methanol (2:1, vol/vol), separatory funnel extraction,
extract drying (Na2S04) and concentration (Kuderna-Danish apparatus),
removal of elemental sulfur with metallic mercury (extracts were shaken with
mercury for >4 hours and then filtered), automated gel permeation chroma-
tography (GPC), reverse-phase column chromatography (Cjg phase eluted with
methanol), and analysis by isotope dilution GC/MS. Compounds without
labeled analogs (identified below) were quantified using the nearest
eluting, most chemically similar labeled compound as a recovery standard.
Compound Without Labeled Compound Used
Labeled Analog as a Recovery Standard
2-methylphenol d4-2-chlorophenol
4-methylphenol d4-2-chlorophenol
benzoic acid d3-l,2,4-trichlorobenzene
benzyl alcohol d4-l,4-dichlorobenzene
2-methylnaphthalene d7-2-chloronaphthalene
di benzo(a,h)anthracene dl2-benzo(g,h,i)pery1ene
i ndeno(1,2,3-cd)pyrene dl2-benzo(g,h,i)pery1ene
37
-------
Recovery corrections were applied to detection limits as well as
detected values to account for the effect of analytical losses on overall
analytical sensitivity. Poor surrogate recoveries for a number of compounds
(especially chlorinated benzenes and chlorinated phenols) resulted in high
detection limits for these compounds.
Extracts used for gas chromatography/electron capture detection (GC/ECD)
analysis of PCBs and pesticides were subsamples of the acid/neutral extracts
that were taken after GPC cleanup. The portion of the extract for
PCBs/pesticides was exchanged into hexane, subjected to alumina column
cleanup [per the EPA Contract Laboratory Program (CLP) procedure; (U.S. EPA
1986)], and analyzed by dual capillary column (0.25-mm i.d. x 30 m) GC/ECD
(Ourabond DB-5 as the analytical column and DB-1701 as the confirmation
column).
Laboratory Analysis for Volatile Organic Compounds
The analytical laboratory (Battel!e-Columbus, Columbus, Ohio) gave the
following report: "Just prior to analysis, the samples were mixed and 10.0
grams of the homogenous sediment was transferred to a clean 40-mL vial. Ten
milliliters of reagent water with 250 ng each of d4-l,2-dichloroethane, dg-
benzene, d5-ethylbenzene and bromofluorobenzene was then added to facilitate
the purging of the sample. The vial was placed in a 110° C heated sand bath
and purged for analysis in accordance with the Speis (1980) Method" (Battelle
1985).
No conclusive documentation is available to explain mixing of sediments
prior to volatile organic analysis (V0A). PSEP protocols warn that many of.
the volatile compounds of interest could be lost during mixing and composit-
ing. The analytical laboratory confirmed that vigorous mixing of volatile
organic samples is not a standard or recommended procedure. However,
samples may have been gently stirred once to homogenize the sediment prior
to analysis (Gebhart, J., 30 December 1987, personal communication).
The analyses were performed using a Tekmar LSC-I liquid sample con-
centrator interfaced to a Finnigan 3200 GC/MS and Finnigan/INCOS Model 2300
data system. Quantification was performed by the internal standard method
using dg-benzene. The MS was operated in the electron impact (EI) ionization
mode with an electron energy of 70 eV and set to scan from 35 to 350 amu/2
sec. A 6-ft x 2-mm ID glass. column, packed with 1 percent SP-1000 on
Carbopack B and interfaced to the ion source by a standard glass jet
separator, was used in the GC.
Ancillarv Analyses
Methods used for the analysis of conventional variables are briefly
described below:
¦ Grain size - Approximately 25 grams of homogenized wet
sediment was treated with hydrogen peroxide to remove organic
material. Sediment was then wet-sieved through a 0.0625-mm
screen and the fines were collected in a cylinder. Sand and
gravel fractions (0.0625 to >2 mm) were wet-sieved and then
dried to constant weight at 90° C. The silt-clay fraction
38
-------
(<0.004 to 0.0625 mm) was treated with a preweighed dispersant
and analyzed by pipetting 20 mL at 30-second intervals for
10-inch depth, and 59-minute intervals for 5-cm depth.
Pipetted samples were dried at 90° C to constant weight and
then corrected for dispersant weight. Results were calculated
based on total weight of the eight fractions. The following
size fractions were evaluated:
gravel - >2 mm
very coarse sand - 1-2 mm
coarse sand - 0.5-1 mm
medium sand - 0.25-0.5 mm
fine sand - 0.125-0.25 mm
very fine sand 0.0625-0.125 mm
silt - 0.004-0.0625 mm
clay - <0.004 mm.
¦ Total solids - Frozen sediment was thawed, homogenized,
weighed, and dried to a constant weight at 103° C. The ratio
of dry sediment weight to wet sediment weight was reported as
a percent.
¦ Total Volatile Solids - Samples were thawed and dried to a
constant weight at 103° C, and then combusted at 550° C. The
percent change in weight between the dried and combusted
sample was determined.
¦ Total Organic Carbon (T0C) and Total Nitrogen (TN) - Samples
were freeze-dried, ground to a fine particle size, treated
with hydrochloric acid, and analyzed using a Carlo-Erba
instrument. T0C and TN results were reported as the percent
of unacidified sample DW.
¦ Oil and Grease - Approximately 25 grams of thawed and
homogenized wet sediment was mixed with 25 grams of magnesium
sulfate monohydrate and extracted twice with carbon tetra-
chloride. Absorbance was measured at 2,930 cm"* on an
infrared spectrophotometer, and compared to the standard curve
prepared from the oil reference standard described in EPA
Method 413.2. Concentrations were reported as mg/kg DW.
¦ Water-Soluble Sulfides - Water-soluble sulfides were measured
according to the method described in Green and Schnitker
(1974), with one exception: 0.005 M Pb(C104)2 was used for
sample titration rather than 0.001 M Cd(N03)2-
The irethods used to analyze conventional variables, except those used
for analysis of water-soluble sulfides and grain size, followed the methods
recommended in PSEP protocols (Tetra Tech 1986f). A specific method for
water-soluble sulfides is not presented in PSEP protocols, but analysis of
water-soluble sulfides is recommended for estimation of biologically
available sulfides. The PSEP method for grain size analysis specifies that
the sieves used in wet-sieving particles greater than 65 um are to be dried
after each use. This process will allow particles less than 65 um that
39
-------
adhere to the sieve surface to be removed and added to the sample fraction
analyzed for percent silt and clay. However, only wet sieving was conducted
for this sample set (Crecelius, E., 2 March 1987, personal communication).
The consequences of not following the sieve drying step and particle
collection step cannot be evaluated from the data. The initial weights of
samples were not provided by the laboratory and therefore sample recoveries
following sieving could not be calculated.
Battelle Northwest Marine Research Laboratory (Sequim, Washington)
performed the above analyses with the exception of water-soluble sulfides,
which were analyzed by Am Test (Redmond, Washington).
Quality Assurance/Qua!itv Control Results
Reviews of sediment chemistry data were performed in accordance with
PSEP guidelines (Tetra Tech 1986f). Quality assurance/quality control
(QA/QC) reviews of chemistry included assessments of accuracy [using
standard reference materials (SRM), matrix spikes, and surrogate recoveries,
when applicable], precision (using analytical replicates), initial and
ongoing calibration and tuning, blank results, sample holding times, and
initial performance tests or validation data for certain non-CLP procedures.
Detailed QA reports were prepared for chemical analyses and were
compiled in a single document (Tetra Tech 1988a). These reports will not be
reproduced in this section, but are summarized below.
Metals--
The data are considered acceptable as qualified. SRM-1646, certified
by the National Bureau of Standards (NBS), was analyzed with each batch of
20 samples. Spike recoveries were determined for antimony and selenium
because these elements were not certified in the SRM analyzed. Silver was
also not certified in the SRM. However, the analytical laboratory had
confidence in the silver results based on intercalibration comparisons with
other laboratories, and comparison with neutron activation analysis
(Crecelius, E., 5 December 1986, personal communication), and therefore did
not perform spike recoveries for silver. It was recommended in a U.S. EPA
Region X policy decision (Gakstatter, J., 24 November 1987, personal
communication) that the data for compounds not certified in the SRM be
considered estimates and qualified in the database. Under this recommenda-
tion, antimony, selenium, and silver would be qualified with an "E" (esti-
mated) in the database.
The matrix spike recovery data for selenium and antimony were within
the control limit criteria (75-125 percent) established by PSEP (Tetra Tech
1986f). Therefore, qualifying all data for these compounds did not appear
warranted and data qualifiers were not assigned to sample results for these
compounds. However, all sample results for silver were assigned an "E"
qualifier because matrix spike data were not available to evaluate accuracy.
In some samples, precision data for silver and antimony exceeded the control
limit established by PSEP. In these cases, an "E" qualifier was assigned to
silver and antimony data. Reanalyses of antimony by XRF in several samples
from Elliott Bay and Carr Inlet indicated that the antimony data generated
40
-------
by GFA'A in this study are not suitable for use in identifying or ranking
problem areas (see next section).
Detected values for manganese, zinc, and chromium were qualified as
estimates based on low recoveries for the SRM during accuracy analysis. In
addition, because the 28-day maximum sample holding time recommended by PSEP
was exceeded for all samples, positive sample results for mercury were
assigned an "E" qualifier.
Assessment of the Effect of Analytical Procedures on Metals Results--
The analytical methods used to determine metals in this study were
designed to measure the total concentrations of metals in sediments (includ-
ing mineral-bound components), in contrast to methods that rely on partial
digestion. A small study was conducted to examine the implications of using
the "total metals" methods, especially when comparing results to historical
reference area data generated by "strong acid" methods (e.g., reference area
data from Carr Inlet).
Two archived Carr Inlet samples collected during the Commencement Bay
Remedial Investigation (Tetra Tech 1985a) were analyzed in triplicate by
"total metals" methods used in the present study and by the "strong acid"
method (per EPA CLP) used during the Commencement Bay study. In addition,
selected samples collected during the present study were reanalyzed by the
"strong acid" method for comparison purposes. The results are presented in
Table 7.
Although differences were observed for a number of metals analyzed by
both methods, the consistently largest differences were observed for
chromium (Table 7). In both Carr Inlet samples, mean chromium concentrations
by "total metals" methods (in this case, XRF) were over 4 times the mean
concentrations determined by the "strong acid" technique. Samples with
higher overall chromium concentrations from Port Susan and Elliott Bay/
Duwamish River tended to have approximately a factor of 2 difference between
"total metals" and "strong acid" results.
Differences between antimony results by "total metals" vs. "strong acid"
methods could not be determined for Carr Inlet samples because antimony was
consistently undetected by the "strong acid" procedure (Table 7). However,
data reported for Samples CR-11 and CR-13 during the Commencement Bay
Remedial Investigation were roughly 15 times lower than the "total metals"
values in Table 7. Similarly, for Duwamish River Samples WW-12 and EW-15,
the "total metals" procedure (including HF digestion and analysis by GFAA)
resulted in concentrations roughly 10-20 times higher than concentrations
determined by the "strong acid" procedure. This marked discrepancy prompted
reanalysis by an independent and more reliable technique for antimony (i.e.,
XRF). Comparisons between XRF and the "total metals" procedure used in this
study were confounding (Table 7), but suggest that antimony concentrations
observed during this study could be considerable overestimates (e.g., by a
factor of approximately 5).
The XRF analysis of antimony aroused sufficient uncertainty about the
"total metals" antimony concentrations reported in this study that antimony
data were not used to define or rank problem areas. However, antimony
41
-------
TABLE 7. OOWARISON OF ANALYTICAL METHODS FOR SELECTS) METALS
Sample
Technique
Antimony
Arsenic
Cadmiun
Chromium
Copper
Lead
Nickel
Silver
Zinc
Reference •
Area Samples
CR-11
CR-11
Cn=3)
(n=3)
Total metal8
Strong acidc
1.91 + 0.16b
U0.92 + 0
4.47.+ 0.79
2.07 + 0.09
0.09 + 0.01
0.13 + 0.01
99+26
19 + 1.5
9.8 + 1.6
6.1 + 0
4.4 + 1.2
3.6 + 0.1
17.7 + 3.1
13.8 + 1.3
0.043 + 0
0.027 + 0.002
28.2 + 1.7
19.3 + 0.58
CR-13
CR-13
(n=3)
(n=3)
Total metal
Strong acid
1.40 + 0.18
U0.92 + 0
3.69 +1.1
2.48 ~ 0.52
0.19 ~ 0.02
0.22 + 0.01
84+10
19 + 1.5
12.6 +0.36
6.73 + 1.4
8.1 + 0.55
2.2 + 0.60
20.6 + 0.85
15.1 +1.3
0.076 + 0.029
0.022 + 0.007
34 + 2
19 + 0.6
PS-01
PS-01
(n=1)
(n=3)
Total metal
Strong acid
2.66
U0.92 + 0
236
133 + 2
49.8
43.3 + 1.1
10.4
5.6 + 2.4
139
130 + 2
Elliott Bay/
Duwamish
River Samples
WW-12
WW-12
WW-12
(n=1)
(n=2)
(n=1)
Total metal
Strong acid
XRF
1,200
59 + 6.6
240
555
266 + 12
618
920 + 33
1,180
1,510 + 69
100
87 + 3.6
EW-15
EW-15
EU-15
(n=1)
(n=2)
(n=1)
Total metal
Strong acid
XRF
150
14 + 1.2
32
223
115 + 2.8
176
305 + 33
210
330 + 40
63.8
45.9 + 2.7
NH-04
NH-04
(n=1)
(n=1)
Total metal
XRF
504
120
KG-06
KG-06
(n=1)
(n=1)
Total metal
XRF
192
27
WW-14
WW-14
(n=1)
(n=1)
Total metal
XRF
1,370
217
SS-09
SS-09
(n=1)
(n=1)
Total metal
XRF
680
547
a Methods used in the present study: digestion with nitric, perchloric, and hydrofluoric acids and analysis by AA (for antimony, cactaifun, and silver) or x-ray fluorescence
(for arsenic, chromium, copper, lead, nickel, and zinc).
''Mean + standard deviation. All concentrations are in mg/kg DU.
c EPA CLP procedure involving digestion with nitric acid and hydrogen peroxide.
** X-ray fluorescence.
-------
distributions are described in the RESULTS section because of their potential
value in assessing relative antimony contamination in Elliott Bay. The XRF
results for antimony were sufficiently eleyated in some samples to establish
that antimony may be of concern in Elliott Bay and may warrant further
investigation.
Because analytical procedures can clearly have an effect on chromium
results, AET generated from historical data (i.e., by strong acid digestion)
were not used in this study. Nevertheless, the results of the analytical
comparisons in Table 7 do not provide a basis for excluding chromium data.
Therefore, these data were included in problem area identification and
ranking.
Semivolatile Organic Compounds—
The data are considered acceptable as qualified. Qualifiers were
assigned to sample results for reasons that vary in severity. Acid/neutral
compound data from 68 samples and pesticide/PCB data from 26 samples were
assigned "E" qualifiers because the extracts were held longer than the PSEP-
recommended maximum holding time (Tetra Tech 1986f). During the initial and
continuing 6C/MS and GC/ECD calibrations, the relative percent difference
(RPD) between response factors for several acid/neutral and pesticide
compounds exceeded the control limit criteria. Sample data associated with
calibrations out of the control limits were therefore assigned an "E"
qualifier and are considered .estimates. In some cases, precision data
exceeded the established control limits. Sample results for compounds that
did not meet the required level of precision outlined in PSEP protocols
(Tetra Tech 1986f) were assigned an "E" qualifier. All sample data for
endrin aldehyde and total PCBs are considered estimates and were assigned an
"E" qualifier due to low recoveries of matrix spikes (recoveries for all six
PCB matrix spikes ranged from 0-39 percent). Because of excessive laboratory
contamination for diethylphthalate, di-n-butylphthalate, and bis(2-ethyl*
hexyl)phthalate, all sample data for these phthalates were rejected and do
not appear in the database.
An EPA policy decision (Gakstatter, J., 10 December 1987, personal com-
munication) required the qualification, as estimates, of all acid/neutral
compounds detected at <1,000 ug/kg DW. This qualification was based upon
results of analyses of two sediment reference materials. One reference
material had moderate contaminant concentrations (e.g., roughly 100 to 150
ug/kg DW for PAH), whereas the other reference material was more contaminated
(roughly 300 to 4,000 ug/kg DW for PAH). For compounds detected in either
reference material, recovery was typically >75 percent (an acceptable level
of accuracy under PSEP guidelines). However, the laboratory failed to
detect a large proportion of the compounds known to be present in the
reference materials at concentrations between roughly 100 and 500 ug/kg DW;
in these cases, detection limits were typically at or above the known
concentrations of target compounds. Thus, the accuracy for detected
compounds was acceptable across a range of concentrations, but the inability
to consistently detect compounds known to be present at <1,000 ug/kg DW
resulted in the qualification of acid/neutral compounds detected at <1,000
ug/kg DW.
43
-------
Volatile Organic Compounds--
Detected volatile organic data were considered estimates and assigned
an "E" qualifier in the database, because it is not known whether a multi-
point calibration was performed by the analytical laboratory. Information
provided by the laboratory indicated that a single point calibration was
performed at least two times a day. Data for three compounds (i.e.,
methylene chloride, trichlorofluoromethane, and 1,1,1-trichloroethane) were
rejected due to excessive blank contamination.
The following practices occurred during the analysis of volatile organic
compounds in Elliott Bay sediment samples that may have compromised data
quality:
¦ The PSEP-recommended holding time of 14 days was exceeded for
31 of the 50 samples analyzed.
¦ An evaluation of accuracy indicated percent recovery of
surrogates outside PSEP control limits (Tetra Tech 1986f) at
three stations.
¦ Failure of the laboratory computer resulted in loss of
surrogate recovery data at three stations.
¦ An insufficient number of replicate samples were analyzed
(i.e., 2 percent frequency of analysis) for the determination
of precision.
The above factors necessitated qualification of all detected volatile
organic data as estimates.
Ancillary Analyses-
Quality assurance review was performed for the following conventional
variables: total solids (TS), total volatile solids (TVS), total organic
carbon (TOC), total nitrogen (TN), oil and grease (0G), water-soluble
sulfides, and grain size. Given the nature of the variables evaluated and
the methods of analysis, data are considered acceptable as qualified. All
TOC data were qualified with an "E" and are considered estimates based on
SRM results. The mean TOC value determined for the SRM was 2.86+0.07
percent (n=6), which was slightly below the certified range (2.99+0.09
percent). No other data were qualified.
BI0ACCUMULATI0N
Field Sampling
English sole (Parophrvs vetulus^ were sampled at 11 transects in Elliott
Bay and at 1 transect at Point Pully, a nonurban reference area (see Figure
6). Point Pully was used as a reference area because previous studies have
found that the area is relatively uncontaminated (Nevissi et al. 1984). In
addition, Pierce et al. (1978) and Landolt et al. (1984) found no serious
hepatic lesions in English sole (n=18 and 163, respectively) collected from
that general area.
44
-------
Most sampling was conducted between 16 and 25 September 1985. However,
Transect SS-91 was revisited on 17 October 1985 to collect additional fish.
Fish were collected using a 7.6-m (headrope) Marinovich otter trawl having a
body mesh size of 3.2 cm (stretched) and cod-end liner mesh size of 0.8 cm
(stretched). Trawling was conducted along each transect at a constant
vessel speed of approximately 2.5 kn during daylight hours (0730-1730 h).
Five of 60 English sole (>220 mm) that were collected for histopatho-
logical analysis (see FISH ECOLOGY AND HIST0PATH0L0GY) were selected for
analysis of PCBs, EPA priority pollutant pesticides, and mercury in raw
muscle tissue. After removal of liver and otoliths, each fish was wrapped in
aluminum foil and stored on ice. In the laboratory, fillets of dorsal muscle
tissue were excised and skinned with a stainless steel spatula in preparation
for analysis.
Laboratory Analysis for Mercury
Mercury was the only EPA priority pollutant metal analyzed in fish
tissue because of its high potential for bioaccumulation. Muscle tissue was
homogenized and subjected to nitric acid/perchloric acid digestion at
Battelle Northwest Marine Research Laboratory (Sequim, Washington). The
digestate was analyzed by cold vapor atomic absorption spectrophotometry.
Laboratory Analysis for PCBs/Pesticides
Extraction and Cleanup--
The analytical procedure used by Battelie Northwest Marine Research
Laboratory was derived from Tetra Tech (1986b), "Analytical Methods for U.S.
EPA Priority Pollutants and 301(h) Pesticides in Tissues from Estuarine and
Marine Organisms." Only the sections relevant to analysis of PCBs and
pesticides were followed. The procedure involves Soxhlet extraction with
CH2CWMe0H (2:1, vol/vol), extract cleanup by. GPC (Biobeads S-X3; elution
with CH2C12) and alumina column chromatography, and capillary column GC/ECD.
Several notable exceptions to the Tetra Tech (1986b) procedure were cited in
the laboratory's cover letter:
¦ Soxhlet extraction was carried out for 12 hours rather than
the specified 24 hours
¦ Rotary evaporation was used for extract concentration rather
than the specified Kuderna-Danish (K-D) apparatus.
These two modifications did not appear to affect laboratory performance based
on results of initial laboratory performance tests with standard reference
materials (SRM) and spiked blanks.
GC/ECD Analysis and Quantification--
Pesticides and PCBs were analyzed by capillary column GC/ECD with a
DB-5 quantification column (0.25-mm i.d. x 30 m, J & W Scientific) and a
SP-608 (Supelco, Inc.) confirmation column. The 80-minute temperature
program used for these samples allowed for a high degree of chromatographic
45
-------
resolution [roughly 76 peaks were resolved in a PCB standard consisting of
Aroclor 1242:1254:1260 (1:1:1, wt/wt/wt)].
The quantification procedure used for PCBs is described in Tetra Tech
(1986b). The procedure involves peak-by-peak quantification using an
internal standard. Relative response factors for resolved peaks in a PCB
standard containing Aroclors 1242, 1254, and 1260 (the range of PCB congeners
expected in environmental samples) were determined by measuring the indi-
vidual peak concentrations by GC/MS and then using dilutions of the same
standard for initial and ongoing GC/ECD calibrations.
Extracts containing pesticides identified on the DB-5 column were rerun
on the SP-608 confirmation column. The sole p,p'-QDE identification on the
DB-5 column (Sample SS-91-643) was confirmed on the SP-608 column and by
GC/MS.
Quality Assurance/Qua!itv Control Results
Reviews of bioaccumulation data were performed in accordance with PSEP
guidelines (Tetra Tech 1986f). QA/QC reviews included assessments of
accuracy (using standard reference materials, matrix spikes, and surrogate
recoveries, when applicable), precision (using analytical replicates),
initial and ongoing calibration and tuning, blank results, sample holding
times, and initial performance tests or validation data for PCB bioaccumula-
tion.
Detailed QA reports were prepared for chemical analyses and were
compiled in a single document (Tetra Tech 1988a). These reports will not be
reproduced in this section, but are summarized below.
Mercury Bioaccumulation--
Mercury data exhibited accuracy and precision within the guidelines
established by PSEP (Tetra Tech 1986f). However, because the 28-day maximum
sample holding time recommended by PSEP was exceeded for all of the tissue
samples analyzed, sample results for mercury are considered to be estimates
and were assigned an "E" qualifier. No raw data accompanied the results;
therefore, this data set was not reviewed for calculation algorithm or
transcription errors.
PCB/Pesticide Bioaccumulation--
Sample holding times exceeded PSEP guidelines; however, qualification
was not considered necessary because the target analytes (particularly PCBs)
are not very susceptible to microbial or chemical alteration. The relatively
high concentrations of PCBs in samples may have precluded detection of
pesticides at low to moderate levels. An "E" qualifier was applied to
samples associated with ongoing calibrations that were outside control
limits. Matrix spike recoveries were low; however, samples were not
qualified based upon low matrix spike recoveries for three reasons:
1. Surrogate recoveries strongly suggest that the sample batch in
which the matrix spikes were included had considerably lower
recovery than the majority of the samples
46
-------
2. Average 4,4'-dibromooctafluorobiphenyl (DBOFB) recoveries for
trawl stations were typically above 50 percent despite low
recoveries in the latter batches
3. Isodrin recoveries were above 50 percent in roughly 95 percent
of all individual samples.
Overall, these data are considered acceptable.
SEDIMENT BIOASSAY
Field Samplino
Sediment toxicity tests with Rhepoxvnius abronius were performed using
sediments from 102 stations in the Elliott Bay system and 4 stations in the
reference area (Port Susan). A subsample of the composite sediment sample
collected for chemical analyses was tested for toxicity using the amphipod
bioassay. Field collection methods for sediment samples are described above
(see SEDIMENT CHEMISTRY, Field Sampling).
The infaunal amphipod fi. abronius was collected subtidally from West
Beach on Whidbey Island (Washington) using a bottom dredge. Amphipods were
maintained and transported in clean coolers with ice, and were returned to
the laboratory within 18 hours of collection.
Laboratory Analysis
Following their arrival in the laboratory, amphipods were kept in
holding containers filled with fresh seawater (28 ppt salinity) and clean
sediment and maintained at 15+1° C under continuous light until used in
testing. Cultures were aerated but not fed during acclimation and were held
for not more than 10 days. Prior to testing amphipods were sorted by hand
from sediments and identifications were confirmed using a Wild M5 dissecting
microscope. Damaged, dead, or unhealthy individuals were discarded.
Acute lethality of amphipods exposed for 10 days to whole, fresh
(unfrozen) sediments was measured using the methodology of Swartz et al.
(1982, 1985) as modified in the PSEP protocols (Tetra Tech 1986f). A 2-cm
layer of test sediment was placed in 1-L glass jars and covered with 800 mL
of clean seawater (28 ppt salinity). Each beaker was seeded (randomly and
blindly) with 20 amphipods and aerated. Six replicates (20 amphipods each)
were run per test sediment. Five beakers served to determine toxicity,
while the sixth beaker served as a reference for daily measurement of water
chemistry (i.e., pH, dissolved oxygen, salinity, and temperature). Testing
was conducted' at 15+1° C under constant light. Test containers were checked
daily to establish early trends in mortality and sediment avoidance, and
also to gently sink any amphipods that had left the sediments overnight and
become trapped by surface tension at the air/water interface. A negative
(clean) control sediment from the amphipod collection site at West Beach,
Whidbey Island was run concurrently with each series of test sediments. In
addition, clean seawater sediment spiked with CdClo was used as a positive
control to ensure that response criteria (lethality) were operative.
47
-------
Amphipod bioassays were initiated on all sediments within a 2-week period
following field collection of sediments.
Bioassay tests were terminated after 10 days, when sediments were sieved
(0.5-mm screen), and live and dead amphipods were removed and counted.
Amphipods were considered dead when there was no response to physical
stimulation and microscopic examination revealed no evidence of pleopod or
other movement. Missing amphipods were assumed to have died and decomposed
prior to the termination of the bioassay (Swartz et al. 1982, 1985).
Quality Assurance/Qua!itv Control Results
Mean mortality ranged from 0 to 10 percent in the clean sediment
(Whidbey Island) controls. A mean mortality of 10 percent is considered
acceptable for amphipod sediment bioassay controls (Swartz et al. 1985).
ANOVA indicated no significant differences (P>0.05) in mean mortality values
among the clean sediment controls. Mortality in cadmium-spiked sediments was
100 percent, which is consistent with the expected mortality rate. Inter-
stitial salinities in the sediment samples were acceptable according to
Puget Sound Protocols (Tetra Tech and E.V.S. Consultants 1986), and none
required adjustment. Dissolved oxygen concentrations in water overlying the
sediments in the bioassay chambers were acceptably high.
Because of a labeling error in the field, the bioassay results from the
intertidal Station PS-05 and replicates at Station PS-01 in Port Susan could
not be.distinguished. Therefore, these data were deleted from the database.
The amphipod bioassay results are considered acceptable for use in
problem area identification. However, it should be noted that the data for
the following stations showed high variance due to an extreme outlier repli-
cate: NS-02, NS-06, KG-07, DR-05, EW-03, EW-14, NH-01, WW-12 (Figure 5).
BENTHIC MACROINVERTEBRATES
Field Sampling
Benthic macroinvertebrates were collected at 74 subtidal stations in the
Elliott Bay study area and at four subtidal stations in the Port Susan
reference area between 25 September and 16 October 1985 (see Figure 5).
Among the 78 stations, water depths ranged from 4.3 m to 21.4 m (corrected
to mean lower low water; Appendix B). A majority of the stations were
located at depths of 8.8-12.2 m, but it was not possible to sample within
this depth range at all stations because of differences in shoreline
bathymetry.
Five replicate grab samples were collected at each station, for a total
of 390 samples. All grab samples were collected using a 0.1-nr modified van
Veen grab sampler. In the field, samples were washed on a sieve with 1.0-mm
mesh openings and fixed with a 10-percent solution of buffered formalin.
Sample tracking records followed each sample through all stages of sample
collection and laboratory processing.
The field sampling methods used to collect benthic macroinvertebrate
samples during the Elliott Bay survey are outlined in the Puget Sound
48
-------
Protocols (Tetra Tech 1986g) and the Elliott Bay Quality Assurance Project
Plan (Tetra Tech 1985d). The following discussion summarizes those proce-
dures. Following deployment and retrieval of the van Veen grab, it was
placed in a sieve stand and the sediment sample was inspected carefully to
determine the acceptability of the sample. Samples were rejected if
excessive leakage or surface disturbance occurred. Samples were also
rejected if they did not meet or exceed the following minimum penetration
depths:
¦ Medium to coarse sand and gravel - 4 to 5 cm
¦ Fine sand and sandy silt - 7 to 10 cm
¦ Silt - 10 cm.
When a sample was judged to be acceptable, the following qualitative
sediment characteristics were recorded:
¦ Penetration depth
n Sediment texture
¦ Sediment color
¦ Presence and strength of odors
¦ Degree of leakage and/or surface disturbance
¦ Presence of debris or shell fragments.
After the foregoing observations were recorded, the sampler was opened
and the sediment was released into the top section of the sieving stand. The
sediment was then washed from above with a gentle spray of seawater, and the
larger masses of sediment were broken apart. Sediment was rinsed into a
sieve box located in the lower level of the sieving stand. The sediment in
the sieve box was then completely washed until materials no longer passed
through the 1.0-mm mesh screen. That portion retained on the screen was
placed in a plastic sample bag having external and internal labels. Samples
were then fixed in the field with a 10-percent solution of Borax-buffered
formalin.
Laboratory Analysis
In the laboratory, benthic macroinvertebrate samples were washed on a
0.5-mm sieve and transferred to a 70-percent solution of isopropyl alcohol.
Organisms from each sample were sorted into the following major taxonomic
groups and enumerated: Porifera, Hydrozoa, Anthozoa, Platyhelminthes,
Nematoda, Nemertea, Sipuncula, Polychaeta, Oligochaeta, Pelecypoda, Acma-
eidae, Gastropoda, Opisthobranchia, Acarina, Pycnogonida, Cirripedia,
Nebaliacea, Amphipoda, Ostracoda, Mysidacea, Decapoda, Isopoda, Cumacea,
Tanaidacea, other Crustacea, Priapulida, Phoronida, Bryzoa, Brachiopoda,
Echinodermata, and Holothuroidea. Planktonic organisms that occurred in the
samples were not enumerated. Colonial organisms that occurred in the
samples were noted as "present" but also were not enumerated.
49
-------
Quality control checks of sample sorting were performed by resorting
20 percent of each sample. If the 20-percent resort indicated a calculated
difference of 5.0 percent or greater in total sample abundance for all taxa
combined, the entire sample was resorted. Independent quality control checks
of sorting procedures (including checks of 20-percent aliquots) and checks of
abundance counts for the major taxonomic groups were also performed.
Samples that failed either quality control check were resorted completely.
Of the 390 samples (i.e,. 78 stations) scheduled for analysis, organisms
in 290 samples (i.e., 58 stations) were sorted into major taxonomic groups
and enumerated by the sorters. Organisms in the remaining 100 samples
(i.e., 20 stations) were identified to the lowest possible taxonomic level
and enumerated. The 20 stations selected for detailed taxonomy were chosen
to represent widespread geographic locations within the project area. Other
criteria used to select these 20 stations included proximity to suspected
contaminant sources and high values of mean mortality in the amphipod
bioassay. Specimens of each species (or lowest possible taxon) that occurred
in the Elliott Bay study area were placed in reference museums prepared by
the taxonomists.
Quality ftssurance/Oualitv Control Results
QA/QC procedures resulted in an acceptable data set without qualifica-
tion. As part of QA/QC, the adequacy of collecting five 0.1-ra' grab samples
was assessed by plotting the. cumulative number of species collected per
replicate at the Port Susan reference stations and at two randomly-selected
stations from Elliott Bay (Figure 7). Reference area stations were used in
this analysis because they were considered more typical of benthic copnuni-
ties than the potentially impacted stations in Elliott Bay. In general, the
cumulative number of new taxa collected in successive grab samples declines
and appears to approach an asymptote by the fifth replicate at most stations.
(Station PS-04 is a possible exception.) This condition indicates that a
reasonably full complement of taxa were collected at both the Port Susan and
Elliott Bay stations using five replicate 0.1-nr samples.
FISH ECOLOGY AND HISTGPATHOLOGY
Field Sampling
Field sampling methods were described earlier (see BI0ACCUMULATI0N,
Field Sampling). English sole larger than 220-imt total length (TL) were
selected for histopathological analysis. This size limit was used to ensure
that most fish were greater than 2 years old. A selection criterion based
indirectly on age was used because English sole younger than 2 years old
have substantially lower prevalences of hepatic lesions than older fish
(Malins et al. 1982). The present study therefore focused on those fish
most likely to be afflicted with hepatic lesions.
Sixty English sole of appropriate length were collected at every
transect except Transect SS-9I, where 54 fish were sampled. A total of 714
fish was collected during the overall study. Immediately after collection,
each selected fish was sacrificed by a blow to the head, measured to the
nearest millimeter (TL), examined for grossly visible external abnormalities
50
-------
130
(0
ID
O
Ui
Q.
F
<
3
s
=)
o
120 -
110 -
100 -
60 -
30 -
20 -
REPLICATES
Figure 7. Cumulative numbers of species collected over successive replicate benthjc
grab samples at four stations in Port Susan and at two stations in Elliott Bay
in 1985.
-------
(e.g., fin erosion, skin tumors, scoliosis, parasites), and transferred to
the shipboard laboratory for liver removal.
In the shipboard laboratory, the liver of each fish was removed in its
entirety, cut into multiple sections, and examined for the presence of
grossly visible lesions. If lesions or discontinuities were noted, a
subsample was taken from the affected area-for histopathological analysis.
If the liver appeared to be normal, a subsample was taken from the center of
the organ at its broadest point. Each subsample was fixed in 10 percent
neutral-buffered formalin. After the liver was removed from each individual,
the sex of the fish was noted and the otoliths (sagittae) were removed for
subsequent age determination.
All fishes in the remainder of the catch at each transect were identic
fied to species and counted. All English sole not selected for histopatho-
logical analysis were measured (nearest mm TL) and counted.
Laboratory Analysis
Each formalin-fixed liver was dehydrated in a graded series of ethanol,
cleared in xylene, and embedded in paraffin. Embedded livers were sectioned
at 4 um using a rotary microtome and stained using hematoxylin and eosin
(H&E). Prepared slides were examined using conventional light microscopy.
Each slide was coded, so the pathologist did not know where the corresponding
fish was captured. Lesion identifications were confirmed by M.S. Myers
(Chief Pathologist, Northwest and Alaska Fisheries Center) to ensure their
consistency with the identifications made by Malins et al. (1980, 1982,
1984). '
Three major kinds of idiopathic hepatic lesion were evaluated:
neoplasms, foci of cellular alteration, and megalocytic hepatosis. Briefly,
neoplasms include both benign and malignant tumors. Foci of cellular
alteration are discrete clusters of altered cells that have specific
staining characteristics and are suspected of being preneoplastic. Megalo-
cytic hepatosis is a specific degenerative condition characterized by a
marked increase in both nuclear and cellular diameters in the absence of
cellular inflammatory responses.
Prevalences of all three major lesions have been found to be elevated in
English sole from urban embayments of Puget Sound (e.g., Malins et al. 1984;
Becker et al. 1987). In addition, Myers et al. (1987) found consistent
patterns of co-occurrence of these lesions in English sole from Eagle
Harbor. Based on those patterns of co-occurrence and comparisons with
similar lesions induced in rodents following laboratory exposure to chemi-
cals, Myers et al. (1987) concluded that megalocytic hepatosis, foci of
cellular alteration, and neoplasms may be related sequentially in the
progression towards hepatic neoplasia in English sole. In that scenario,
the following steps are thought to occur:
¦ Megalocytic hepatosis and associated degenerative lesions are
induced as the initial, subchronic to chronic manifestations
of the cytotoxic effects of hepatocarcinogens. These lesions
form the proper stimulus for a proliferative response.
52
-------
¦ In the above environment favoring proliferation, foci of
cellular alteration develop. Because these lesions are
selectively resistant to the cytotoxic effects of hepatocar-
cinogens, they have a growth advantage over normal hepato-
cytes.
a Autonomous, neoplastic hepatocytes arise from the non-
autonomous foci of cellular alteration to form neoplasms.
This final transformation is probably a complex, multistep
process of mutation followed by selection.
Quality Assurance/Qua!itv Control Results
Lesion identifications were confirmed by Mark Myers of the National
Marine Fisheries Service. To ensure consistent identification of lesions
between the two pathologists for this project, each examined one-half the
slides from each station. For all three major kinds of lesions, the numbers
of each lesion identified by the two pathologist were very similar, implying
consistent diagnostic criteria. In addition, the relative prevalences of
lesions among stations and among lesion types identified during this study
were similar to results from previous studies by National Marine Fisheries
Service (Malins et al. 1980). The final histopathology data were considered
acceptable without qualification.
DATA MANAGEMENT
To facilitate data storage, quality control, and analysis, recent and
historical data from the Urban Bay Action Programs have been incorporated
into a DBase Ill-compatible microcomputer database. This database system is
based on the same software, data formats, and files as the SEDQUAL system.
The database software performs a wide variety of retrievals, reports, and
analyses. It also allows data to be transferred directly to other software
(e.g., SPSS/PC+ and Lotus 1-2-3) for statistical analyses and graphic
displays. A library system is incorporated into the database to document
data sources, changes to data, and other information to be linked to sample
measurements.
Data Organization
Data are linked so that related kinds of information can be retrieved
together for interdisciplinary analyses. For example, sediment chemistry,
infauna abundances, and bioassay data can be retrieved into a single table,
based on common samples or stations. During data retrievals, data can be
summarized across laboratory replicates and field replicates as requested by
the user.
The database design requires that only actual measurements be recorded.
For example, if cadmium was measured at all stations but one during a
survey, no value for cadmium need be stored for that single station. This
reduces ambiguity and complexity of the database as well as storage require-
ments and retrieval speed. For biological effects data, a distinction is
made between "not significant" (as compared to a reference area) and "not
evaluated."
53
-------
Each data value is associated with a single survey and station. The
survey identifies the sampling program responsible for data collection. The
station coordinates identify a unique geographic position sampled during that
survey. Stations are described by an identifier, latitude and longitude, and
basin and subbasin codes. Samples collected at each station are further
identified by a unique sample ID, the date of collection, and field replicate
number. Data of any type can be retrieved by:
¦ Date - all dates or a specified range
¦ Survey - any survey or only data from one or more specific
surveys
¦ Station - any station or only data from one or more specific
stations
¦ Basin and subbasin - any basin/subbasin or only data from one
or more specific basin/subbasin.
These criteria allow any subset of the data to be retrieved or combined for
analysis.
Data Analysis
Procedures for summarization of data are programmed into the database,
providing consistent treatment and formatting of the data for >na-lysis and
interpretation. These procedures include, for example, the ability to rank
observations by station or chemical; create new variables (such as sums of
HPAH and LPAH); construct species lists by replicate, sample, or station; and
compare data to AET or other sediment quality values. Statistical analyses
were carried out using SPSS/PC+, and Lotus 1-2-3 was used for other analyses
and data manipulations.
Data Entry and Quality Control
Quality control of the data was based upon technical evaluation of the
data, automated error-checking procedures in the database, and consistent and
reliable automated procedures for retrieving and sunmiarizing the data. All
additions of data and modifications to data are documented, and the appro-
priate document reference code linked to the affected samples. The date is
automatically inserted in all permanent output from the database. Technical
review of the data is carried out before data entry and during analysis and
interpretation. Automated error-checking procedures are used to screen data
to preclude erroneous codes, duplicate data, and insufficiently or incor-
rectly identified data (e.g., measurements that are not assigned to a
previously defined station). All access to the database is carried out
through a series of menus and prompts, ensuring that all summaries and
analyses are carried out in a consistent and replicable manner.
54
-------
RESULTS
Results of field investigations in the Elliott Bay system and Port
Susan are presented in the following sections on sediment chemistry,
bioaccumulation, sediment bioassays, benthic macroinvertebrates, and fish
ecology and histopathology. An evaluation of data collected during the
present study and comparisons with recent data from previous studies is
provided in each section.
SEDIMENT CHEMISTRY
The following section provides a summary of chemical results for over
100 subtidal and intertidal sediment samples collected in Elliott Bay and
the Duwamish River. An additional five samples collected from the Port
Susan reference area were also analyzed. Chemical data were collected for
11 EPA priority pollutant metals, 37 acid/neutral EPA priority pollutant
organic compounds, PCBs, 13 EPA priority pollutant chlorinated pesticides,
6 additional Hazardous Substance List compounds, and selected tentatively
identified compounds (TIOs) (see Table 1). Twenty-eight of the samples were
also analyzed for EPA priority pollutant volatile organic compounds. In
addition, sediment conventional variables (e.g., grain size distribution and
total organic carbon content) were analyzed and are discussed in 'this
section. The objectives of this section are to:
¦ Provide a chemical perspective of the Elliott Bay/Duwamish
River study area, including the general distributions,- concen-
tration ranges, and frequencies of detection of chemical
contaminants
¦ Determine the magnitude of contamination relative to reference
area conditions and to determine the significance of this
contamination relative to Puget Sound reference areas
¦ Summarize spatial correlations among frequently detected
chemicals and to define groups of chemicals with similar
distribution patterns
¦ Compile historical data to supplement the sediment chemistry
results of the present study.
All chemical data and sediment conventional data are presented in Appendix A.
Normalization of Chemical Concentrations
Sediment concentrations presented in this report are typically expressed
as the weight of contaminant per dry weight of sediment (e.g., ug/kg dry
weight or DW). Normalization of sediment concentrations to other variables
[e.g., percent organic carbon or percent of fine-grained material (silt plus
clay)] can aid in the interpretation of contaminant distributions by
focusing on the most contaminated fractions of sediment, thus reducing the
55
-------
significance of variations in less important components of sediment texture
and composition. The following is a brief description of each type of
normalization.
Dry Weight Normalization-
Most sedimentary contaminants are associated primarily with the solid
material in bulk sediments, not with the interstitial water. Thus, dry
weight contaminant concentrations are preferred to wet-weight concentrations.
Use of dry weight concentrations precludes the possibility that variations
in sedimentary moisture content will obscure informative trends in chemical
data.
Total Organic Carbon Normalization-
Chemical concentration gradients, particularly of nonpolar, nonionic
compounds, have been observed to correlate well with sedimentary organic
carbon content (e.g., Choi and Chen 1976). This observation is commonly
interpreted in one of two ways:
1. Organic matter is the "active fraction" of sediment and serves
as a sorptive sink for neutral, and possibly polar or
metallic, compounds
2. Carbon-rich particles may be an important transport medium for
contaminants [e.g., HPA'H may be associated with soot particles
(Prahl and Carpenter 1983)].
The occurrence of multiple contaminant sources in a localized area can
obscure gradients of concentrations normalized to TOC content.
Normalization to Percent Fine-Grained (<63 um) Particles--
On a limited spatial basis, contaminant concentrations are often
inversely correlated with particle size (e.g., Lee 1985). Thus, contaminants
(especially metals) may be concentrated in the fine-grained particles of
bulk sediments. This observation is often explained in terms of surface
area, in that finer particles have greater specific surface area, and thus
greater sorption capacity, than larger particles. However, organic carbon
content also tends to vary inversely with particle size in natural sediments
(Choi and Chen 1976). Thus, normalizing to percent fines may be effectively
equivalent to normalizing to organic carbon content in natural sediments.
Grain size relationships can be confounded by anthropogenic processes such
as sandblasting, which can result in coarse-grained sediments with high
metals concentrations.
Evaluation of the Reference Area
Port Susan appeared to be a suitable reference area based on available
chemical data. The Port Susan sediment chemistry data are compared to data
from Elliott Bay and other Puget Sound reference areas later in this section
(Sediment Metals of Concern and Organic Compounds of Concern). As discussed
later, Port Susan Station PS-05 was excluded from reference area comparisons
because of anomalously high concentrations of several metals and organic
56
-------
compounds. Exclusion of this intertidal station does not affect interpreta-
tion of reference area biological data because amphipod bioassay and benthic
infaunal data were not collected at Station PS-05.
Conventional Sediment Characteristics
Conventional sediment variables measured for this study included grain
size distribution (as percent sand, silt, and clay), "oil and grease"
concentrations, total organic carbon, nitrogen, and sulfides.
Grain Size-
Average percentages of fine-grained material in the nine study areas
and Port Susan are presented in Figure 8. Grain size distributions generally
did not follow spatial trends and areas are not easily discriminated based
on the summary information presented in Figure 8. Ranges of grain size
distributions were extremely wide in all areas except Alki Beach (Area AB),
which had sediments with markedly coarser particle size distributions
(ranging from roughly 5 to 40 percent fine-grained material). The lack of
apparent gradients is not surprising, as the sampling design was oriented
toward characterization of contaminants near potential sources rather than
characterization of sediment transport; for example, samples in the Duwamish
River were not collected from a consistent channel position.
A more detailed ' summary of grain size distributions is presented in
Figures 9, 10, 11, and 12. One general trend apparent from Figure 9 is that
intertidal sediments tended to be coarse-grained (e.g., <20 percent fines in
virtually all cases). In contrast, sediment samples collected in slips and
other relatively quiescent areas (e.g., near piers) tended to be predomi-
nantly fine-grained (e.g., >70 percent fine-grained material).
Grain size distributions and organic carbon content of the reference
(Port Susan) sediments are presented in Figure 13. Stations PS-01 through
PS-04 form a transect away from the mouth of the Stillaguamish River
(Station PS-05 is intertidal and is not a part of this transect). Based on
the information in Figure 13, it is apparent that finer particles relatively
high in organic carbon content settled near the mouth of the river whereas
coarser particles with lower TOC content were prevalent farther from the
river mouth.
Oil and Grease-
Average "oil and grease" concentrations (mg/kg DW) for the nine study
areas and Port Susan are presented in Figure 14. These concentrations are a
broad indicator of the amount of solvent-extractable organic matter in sedi-
ments, but do not provide information on chemical-specific contamination.
Mean concentrations were highest along the Seattle South waterfront
(Area SS), although the highest individual station concentrations were in the
Duwamish River (Area DR) and East Waterway (Area EW). Within Area SS, the
maximum concentrations were at Stations SS-08 and SS-09 (5,200 to 5,470 mg/kg
DW), with generally decreasing concentrations moving away from these
stations in either direction. Concentrations in the East Waterway were
patchy; the maximum, 7,800 mg/kg DW, was observed at Station EW-05.
Concentrations in the Duwamish River (Area DR) were also patchy, with a
57
-------
100 -1
80 -
CO
UJ
z
UL
z
Ui
o
DC
UJ
Q.
60 -
40
20 -
ui
O
z
UI
cc
UJ
u_
UI
DC
LEGEND
PS
MG NS SS NH EW WW KG DR AB
STUDY AREA
MEAN
I
II
PS Port Susan (n = 4)
MO Magnoia Bluff (n = 4)
NS Seattle North Waterfront (n = 8)
SS Seattle South Waterfront (n = 11)
NH North Haibor Island (n-11)
EW EaslWatenway(n» 16)
WW West Waterway (n= 19)
KQ Kellogg Island (n = 11)
DR Duwamfsh River (n = 18)
AB AIM Beach (n » 4)
Figure 8. Average percent fine-grained material (silt plus clay) in sediments of Elliott Bay,
the Duwamish River, and Port Susan.
-------
PERCENT RNES
CD 0-20%
CD > 20-40%
¦D> 40-60%
¦0 > 60*80%
WM >80-100%
.— NM6CD
Ohm
UG41
'"s^D
-MMSC
-NS-oaTl
D*As/
Cam _
CNM7
ss13
wmo
EW-11
K&OI
KMT
KMt'G
won
wfroe'ri
¦»T7|
-oats|
-0IV15 |
•OM2M1
¦D»iin
n
6000 .
B rati
? meters
j DUct-
-DHMl
¦ 0R4J |
• . LOCATION AND DESIGNATION
NS Ol Of WTERTOA1. STATION
• LOCATION AND DESIGNATION
«*»-10 Of SUBTIDAL STATION
~ iw-
¦ l Df^oa -
J OFVos d
Figure 9. Percent fine-grained material in sediments of Elliott
Bay and the Duwamish River.
ID M-2S —
ItMlC
59
-------
EW-16
EW-14 EW-15
EW-13
EW-10
EW-12
EW-11
EW-09
Ew-oe
EW-07 »
EW-04
EW-03
EW-02
EW-01 1 I
PERCENT FINES
CH 0-20%
ED > 20-40%
WD >40-60%
> 60-80%
>80-100%
EW-06 [~
• « LOCATION AND DESIGNATION
NS-01 OF INTERHDAL STATION
• LOCATION ANO DESIGNATION
WW-10 OF SUBTIDAL STATION
Figure 10. Percent fine-grained material in sediments of the
East Waterway.
60
-------
NH10 ¦ I
NH-091 I
NH06
NH05«
NH-04
NB-07 g 1
• t LOCATION AND DESIGNATION
NS 01 OF INTERTIDAL STATION
• LOCATION AND DESIGNATION
WW-10 OF SU0TIDAL STATION
meters
NH42
PERCENT FINES
IZ1 0-20%
>20-40%
>40-60%
>60-80%
>80-100%
Figure 11. Percent fine-grained material in sediments of North Harbor Island.
-------
I I WW-17
WW-15*f
WW-13 0
PERCENT FINES
CD 0-20%
d > 20-40%
>40-60%
> 60-80%
>80-100%
• # LOCATION AND DESIGNATION
NS-01 OF INTERTIDAL STATION
• LOCATION AND DESIGNATION
WW-10 OF SUBTIDAL STATION
meters
300
WW-16
Ww-14
WW-12
WW-10
WW-09
WW-11
WW-06M1
ww-os
WW-03
I 1WW-02*
WW-01;
Figure 12. Percent fine-grained material in sediments of the
West Waterway.
sz
-------
4.5
z
4.0 -
UJ
O
cc
3.5 -
UJ
a.
z
3.0 -
o
a
cc
2.5 -
<
o
o
2.0-
z
<
o
1.5 -
cc
o
_J
1.0 -
<
1-
o
1-
0.5 -
LEGEND
PS-01
EZ3 TOC
\ | Percent Fines
, \
PS-02
PS-03
STATION
PS-04
r-90
80
<
£
70 p
<
- 60
Q
UJ
SO 3
<
£C
40 9
UJ
- 30
Z
20 U
O
K
111
10 o.
PS-05
Figure 13. Grain size and total organic carbon content of Port Susan sediments.
-------
7800
6000 -i
-p>
O
"55
>»
w
¦o
O)
O)
£
uj
w
<
UJ
DC
O
Q
Z
<
4500 -
3000 -
1500
'I
7600
UJ
O
z
Ul
cc
III
u.
Ul
EC
Hfc-i
PS
'4
MG NS SS NH EW WW KG
STUDY AREA
AB
DR
LEGEND
I MEAN
I
z ^
Is
P 3 Port Susan (n = 4)
MQ Magnolia Bkifl (n = 4)
N S Seattle North Waterfront (n = 8)
SS Seattle South Waterfront (n = 11)
NH North Harbor Island (n = 11)
EW East Waterway (n = 16)
WW West Waterway (n = 19)
KO Kelogg Island (n = 11)
DR Duwamish River (n = IB)
AB AIM Beach (n = 4)
Figure 14. Average oil and grease concentrations (mg/kg dry wt.) in sediments of Elliott
Bay, the Duwamish River, and Port Susan.
-------
pronounced maximum at Station DR-10 (7,600 mg/kg DW). Maximum concen-
trations in Areas NH and WW were roughly 4,000 mg/kg DW. Magnolia bluff
(Area MG) was the only area in Elliott Bay with "oil and grease" concentra-
tions within the range of the reference area (Port Susan).
Total Organic Carbon—
T0C concentrations (summarized in Figure 15) demonstrated similar
trends as observed for oil and grease concentrations (Figure 14). The
correlation between T0C and oil and grease concentrations for all stations
was moderate (r=0.59, n=107, P<0.05). Area SS (the Seattle South waterfront)
had the highest mean T0C concentration (7.3 percent) and the highest
individual; concentration (26.6 percent at Station SS-08). The next highest
TOC concentration in the study (10.3 percent), was found at adjacent Station
SS-09. Average TOC concentrations for other areas were less than 3 percent.
TOC concentrations along Magnolia Bluff were very low; all individual
MG stations had lower TOC concentrations than those found in the reference
area (Port Susan). TOC concentrations are presented in more detail in
Figures 16, 17, 18, and 19. Gradients in TOC concentrations were not
readily apparent in study areas other than Port Susan.
An inverse relationship between particle size and TOC content has been
reported in other studies (e.g., Choi and Chen 1976) and is expected when
hydrodynamic sorting is a predominant process. However, the correlation
between TOC and percent fine-grained material was poor overall (r=0.31,
n=107, P<0.05) and in most study areas. In part, the poor correlation may be
attributed to the presence of relatively coarse-grained sediments located
near discharges of material rich in organic matter (e.g., CSOs). The North
Harbor Island (NH) area was an exception, in that a moderately strong
correlation was observed between TOC and percent fine-grained material
(r=0.79, n=ll, P<0.05). Because this area is at the mouth of the Duwamish
River, hydrodynamic sorting may be responsible for the better correlation
relative to other study areas.
TOC concentrations correlated well with total volatile solids (TVS)
concentrations (r=0.93, n=107, P<0.05), based on a regression that included
all sediment stations. The following regression equation related TOC to
TVS:
TVS = 1.54(TOC) + 3.11.
Water-Soluble Sulfides--
Distributions of water-soluble sulfides in sediments are summarized in
Figure 20. Sulfide concentrations correlated moderately well with total
organic carbon concentrations (Figure 15) (r=0.65, n=107, P<0.05). Because
sulfides are indicative of sulfate-reducing (oxygen-poor) conditions, it is
not unusual for sulfide concentrations to be relatively high in areas that
are rich in organic matter and high in oxygen demand. The highest mean and
individual sulfide concentrations were found along the Seattle South
waterfront (Area SS), an area with consistently high sulfide concentrations
that maximized at Station SS-06 (1,500 mg/kg DW). High values were also
observed in the East Waterway [particularly Stations EW-05 and EW-06 (1,100
and 740 mg/kg DW, respectively)], in the North Harbor Island area [especially
65
-------
LEGEND
26.6
en
CT)
z
o
to
oc
o£
o.5>
V
Of
OC-O
o _
o
H
15 -|
10 -
ui
o
z
Ui
(C
UI
u.
UI
DC
ih
A.
ill
PS
MG NS SS NH EW WW KG DR AB
I MEAN
I
v> a
41
P 8 Port Susan (n « 4)
Ma Magnolia BKiff (n = 4)
NS Seattle North Waterfront (n = 8)
S S Seattle South Waterfront (n = 11)
NH North Haibor Island (n-s 11)
E W East Waterway (n = 16)
WW Wast Waterway (n = 19)
KG KeBogg island (n = 11)
DR Duwamlsh River (n = 18)
AB AIM Beach (n a 4)
STUDY AREA
Figure 15. Average total organic carbon (TOC) content (as percent dry wt.) in sediments of
Elliott Bay, the Duwamish River, and Port Susan.
-------
PERCENTTQC
~ DM)
~ KHI
-NMtC
-NSOtO
~ UfrOI
''CD
¦~1N&07
-NS03f~l
-NM2 B~l
-NMl'O
-SS-I2 CD
-SS.11 I
-SS-lOl
I I 0-1%
d >1-3%
T1 >3-5%
1 >5-10%
M >10%
-ssttl
-SMI
-S&04M
-SS^3«~I
CU*M3/
NH01 -
NH4S-
\wvmc
NH4M
-J/
WW-l7i
WWM5#^
. 7 WW-16-
WWMO-
WYM#
WWOB
WW4S
WMMB-
_ WW4BV=1.
CjKsrr'
.cass®
5KO-10"
KO09-
CNMe-
- EW-14
-0AM3
-EW-10
-EWM1
-EYWt
-EW4S
-EWMIS
-SWM*
-EW43
-EVW8
-©AMI*
-K04S|Q
-KM7C
-K004*B3
-K»02*O
-oai7«~l
Cm-m
¦"iPB-m
CD DM*
-whs n
¦OR-13 CD
-w-ia
-0»11
E
ii
6000
B tot
! meters
2000
• •
NS-01
LOCATION AND 06SCNATION
OP NTERT10AL STATION
O0MJ-
Coims-
• LOCATION AND 06SCNATON
WW-10 OF SUSTIOM. STATION
-0MB
-0H07 « I
16. Total organic carbon (TOC) content in sediments of
Elliott Bay and the Duwamish River (inrtivoual stations).
Figure
r2o»«n-
CD wws —
-omxg'l
I 0R42 Ml
> Daoi CD
67
-------
N_
EW-12 i»
EW-07
EW-04-
EW-16 I
•
MD _
tEW-14 EW-15
I
EW-13
EW-10
EW-08
EW-11
• EW-09
EW-05
EW-03
EW-02
EW-01
EW-06
PERCENTTOC
0-1%
>1-3%
> 3-5%
>5-10%
>10%
• , LOCATION AND DESIGNATION
NS-01 OF INTERTOAL STATION
• LOCATION AND DESIGNATION
WW-10 OF SUOTIDAL STATION
meters
Figure 17. Total organic carbon (TOC) content in sediments of
the East Waterway.
68
-------
NH10 ¦I
• NH-09 ¦ I
"l06 NH-05
/SJI
B INH-08
7*1 1
L3
NH-04
WO,
• t LOCATION AND DESIGNATION
NS01 OF INTER!IDAL STATION
• LOCATION AND DESIGNATION
WW 10 OF SUBTIDAL STATION
meters
NH111 I
NH 01 •
Nf+02
NH-03
CD 0-1%
>1-3%
>3-5%
>5-10%
>10%
Figure 18. Total organic carbon (TOC) content in sediments of North Harbor Island.
-------
IWW-17-
I I WW15* #
I WW-13
PERCENT TOG
IZJ 0-1%
MD >1-3%
>3-5%
>5-10%
>10%
• „ LOCATION AND DESIGNATION
NS-01 OF INTERTIDAL STATION
• LOCATION AND DESIGNATION
WW-10 OF SUBTIDAL STATION
WW-14 M\
VyW-12
WW-10
WW-09B I
WW-11'
meters
WW-OB
WW-05
U I WW-02*
WW-01 g I
Figure 19. Total organic carbon (TOC) content in sediments of
the West Waterway.
70
-------
o>
5
>»
T3
O)
O)
E
V)
ui
Q
UL
_l
3
(0
1200 ~t
1000 -
800 -
600 -
400-
200 -
500
r^~i
r^i
LEGEND
I MEAN
PS Port Susan (n = 4)
MO Magnolia Bluff (n = 4)
N S Seattle North Waterfront (n = 8)
S S Seattle South Waterfront (n = 11)
NH North Haifoor Island (n = 11)
E W East Waterway (n = 16)
WW West Waterway (n = 19)
KQ Kelogg Island (n = 11)
DR Duwamish River (n = 18)
AB AIM Beach (n = 4)
PS
MG NS SS NH EW WW KG DR AB
STUDY AREA
Figure 20.
Average sulfide concentrations (mg/kg dry wt.) in sediments of Elliott Bay, the
Duwamish River, and Port Susan.
-------
Station NH-03 (870 mg/kg DW)], and along the east side of the West Waterway
[particularly Station WW-09 (650 mg/kg DW)]. Sulfides concentrations along
Magnolia Bluff, Alki Beach, and in the reference area were relatively low
and consistent.
Sediment Chemistry: Metals
Concentration ranges and detection frequencies for the 13 elements
analyzed in this study are presented in Table 8. Eleven of the 13 EPA
priority pollutant metals are included in this group; beryllium and thallium
were not analyzed because historical data did not suggest that these metals
were of concern in the study area.
Most metals were detected in all samples (Table 8). Concentration
ranges were very broad for many metals, ranging from reference area con-
centrations to some of the highest concentrations reported in Puget Sound
sediments. For example, concentration ranges of antimony, lead, and mercury
spanned three to four orders of magnitude. Maximum values occurred most
often along the Seattle South waterfront (Area SS; arsenic, cadmium,
chromium, lead, nickel, selenium, and zinc), although other areas had
maximum concentrations of certain metals (Table 8). Station SS-09 was
noteworthy in that three metals, cadmium, lead, and zinc, maximized there.
The maximum mercury concentration was found in the Alki Beach area (Station
AB-01), the maximum copper concentration was found in Area NH (Station
NH-03), and the maximum antimony concentration was found in the West Waterway
(Station WW-14). High levels of contamination by most metals occurred in
more than one area. The distributions of the metals of greatest concern are
described in the following section.
Sediment Metals of Concern--
Metals of concern are defined as those metals that occur at concentra-
tions exceeding Puget Sound reference areas. It is assumed that the range
of reference concentrations provides a reasonable measure of the possible
variability of concentrations in relatively uncontaminated sediments.
The range of trace metal concentrations in Puget Sound reference areas
are presented in Table 9. Metals concentrations from Port Susan sediments
collected for this study (Stations PS-01 to PS-04) are included among the
reference values summarized in Table 9. Station PS-05 was excluded from
compilations of reference stations because it had anomalously high concentra-
tions of several metals (as well as organic compounds). Chromium was the
most notable metal at Station PS-05; the chromium concentration of 886 mg/kg
DW at Station PS-05 was over 3 times that of any other Port Susan station and
was uncharacteristically high for natural mineral assemblages. The exclusion
of Station PS-05 from reference area chemical characterization does not have
serious consequences for data interpretation, because this intertidal
station was not included in the bioassay comparisons or in the benthic
infaunal comparisons. Other Port Susan stations had metals concentrations
that were generally characteristic of other Puget Sound reference areas
(Table 9), although chromium and nickel concentrations were 2-3 times
existing reference values. This was expected as the chemical analyses used
in this study yielded a more complete measurement of mineral-bound crustal
elements (such as chromium and nickel) than analyses used in previous
72
-------
TABLE 8. CONCENTRATIONS OF METALS
IN SURFACE SEDIMENTS OF ELLIOTT BAY,
DUWAMISH RIVER, AND PORT SUSAN
Range
Detection
Location of
Chemical
(mg/kg dry wt)
Frequency
Maximum
Antimony
U0.293
- 1,370
106/107
WW-14
Arsenic
2.37
- 584
107/107
SS-03
Cadmium
0.07
- 17.2
107/107
SS-09
Chromium
E32a
- El,080
107/107
SS-10
Copper
9.3
- 2,050
107/107
NH-03
Iron
15,900
- 112,000
107/107
WW-19
Lead
7.2
- 71,100
107/107
SS-09
Manganese
E258
- E3,390
107/107
NS-04
Mercury
E0.012
- E28.8
107/107
AB-01
Nickel
18.2
- 366 t
107/107
SS-10
Selenium
U0.11
- 0.93b
78/107
SS-06
Silver
U0.02
- E8.27
102/107
NS-01
Zinc
E32.7
- E6,010
107/107
SS-09
a Qualifiers:
U = Undetected at the detection limit shown.
E = Estimated value.
b Maximum observed value does not exceed range of values for
Puget Sound reference areas.
73
-------
TABLE 9. SUM1ARY OF METAL CONCENTRATIONS IN SEDIMENTS
FROM PUGET SOUND REFERENCE AREAS
Detection
Reference
Sites°
Chemical
Range (mg/kg dry wt)a
Frequency
Antimony
U0.1c-2.76 (0.92-2.76)
19/39
1,2,3,4,7,8,9,10,11
Arsenic
1.9-17 (6.41-11)
41/41
1,2,3,4,7,8,9,10,11
Cadmium
0.047-1.9 (0.07-0.15)
31/31
1,2,3,4,6,9,10,11.
Chromium
9.6-E255 (E132-E255)
45/45
1-11
Copper
5-74 (16.2-49)
35/35
1,2,3,4,5,6,9,10,11
Lead
U0.1-24 (7.2-11)
28/35
1,2,3,4,5,6,9,10,11
Mercury
0.01-0.28 (E0.015-E0.110)
45/45
1-11
Nickel
4-140 (63.2-140)
33/33
1,2,3,4,5,9,10,11
Selenium
UO.1-1.0 (UO.14-0.22)
21/31
1,2,3,4,6,9,10,11
Silver
U0.02-3.3 (U0.02-E0.1)
31/33
1,2,3,4,5,9,10,11
Zinc
15-E101 (E57.2-E101)
33/33
1,2,3,4,5,9,10,11.
a The range of Port Susan concentrations from this study is shown in parentheses.
Port Susan concentrations from this study were the highest of all reference area
sites for antimony, chromium, nickel, and zinc. The highest values for reference
areas without Port Susan were 1.7 (antimony), 130 (chromium), 47 (nickel), and-100
(zinc). Station PS-05 (this study) was excluded from this table for reasons
discussed in text.
Reference sites:
1. Carr Inlet 5. Port Madison 9.
2. Samish Bay 6. Port Susan 10.
3. Dabob Bay 7. Nisqually Delta 11.
4. Case Inlet 8. Hood Canal
Sequim Bay
Port Susan (this study)
Port Susan (1986)
c U = Undetected at the method detection limit shown.
References:
(Site
(Site
(Site
(Site
(Site
(Site
(Site
(Site
(Site
(Site
(Site
1) Tetra Tech (1985a); Crecelius et al. (1975)
2) Battelle (1986)
3) Battelle (1986)
4) Crecelius et al. (1975); Mai ins et al. (1980)
5) Mai ins et al. (1980)
6) Mai ins et al. (1982)
7) Crecelius et al. (1975)
8) Crecelius et al. (1975)
9) Battelle (1985)
10) This study
11) PTI and Tetra Tech (1988).
74
-------
studies. In fact, the "total metals" technique (used in this study) and the
"strong acid" technique (common in past studies) were recently compared with
Carr Inlet, Port Susan, and Elliott Bay sediments. The results indicate
that the different analytical procedures can account for the relatively
higher chromium concentrations observed in Port Susan sediments during this
study but not the higher nickel concentrations (see METHODS, Sediment
Chemistry, Assessment of the Effect of Analytical Procedures on Metals
Results).
Ten of the eleven EPA priority pollutant metals analyzed in this study
were found at concentrations exceeding the highest Puget Sound reference
concentrations (Tables 8 and 9). Among these ten metals of concern, nickel
will not be considered in detail because only one station (SS-10) had a
nickel concentration that exceeded the range of Puget Sound reference
concentrations (Table 9). Selenium concentrations did not exceed Puget Sound
reference area values and are thus not considered of concern in this study.
A summary of the distributions of the metals of concern is presented in
Table 10. In this table, EAR are used to describe chemical distributions.
Each EAR is the ratio of the dry weight concentration of a chemical divided
by the average concentration determined for six Carr Inlet stations (Tetra
Tech 1985a). Table 11 summarizes the most contaminated study areas and
stations in terms of EAR values; areas with EAR between 100 and 1,000 are
presented in Table 11 as well as areas with EAR values of greater than
1,000. Descriptions of the distributions of metals of concern are described
below for the most contaminated study areas. Concentrations normalized to
fine-grained material were examined in the most contaminated study areas,
but will not be discussed unless they suggest gradients not apparent for dry-
weight concentrations.
Antimonv--Absolute antimony concentrations reported for samples in this
study are probably overestimates, as determined by subsequent XRF analyses
(see METHODS, Sediment Chemistry, Assessment of the Effect of Analytical
Procedures on Metals Results). Nonetheless, the data are presented in this
section to yield information on the areas with the highest relative concen-
trations. Antimony data were not used for problem identification or
ranking.
Antimony EAR values for all study areas are presented in Figure 21, and
EAR values for individual stations in the most contaminated areas are
presented in Figure 22. The two highest concentrations in the study were
found in the West Waterway (1,370 mg/kg DW at Station WW-14 and 1160 mg/kg
DW at Station WW-12; Figure 22). Other stations on the east side of the West
Waterway (e.g., WW-09, WW-11, WW-16, WW-18, WW-19) were also elevated. High
concentrations were also found at NH stations adjacent to the mouth of the
West Waterway (Station NH-04 = 504 mg/kg DW and NH-03 = 249 mg/kg DW) with
rapidlv decreasing concentrations moving away from these stations (Fig-
ure 22). Other high concentrations were found along the Seattle South water-
front, at Stations SS-03 (690 mg/kg DW) and SS-09 (680 mg/kg DW). These two
stations were not adjacent, and intervening stations had order-of-magnitude
lower concentrations that did not appear to follow a concentration gradient
on a dry-weight or percent fines normalized basis. Other stations with high
concentrations were EW-15 (the remainder of the East Waterwav stations were
75
-------
TABLE 10. RANGE IN EAR FOR INORGANIC CONTAMINANTS
OF CONCERN IN SEDIMENTS OF ELLIOTT BAY
AND THE DUWAMISH RIVER
EAR®
Areas wnere Threshold
Exceeded by 10 Timesc
Chemical
Range
Medi an
Threshold^
Antimony
2.6 - 12,000
240
25
AB ,DR)EW,KG,NH,NS,SS,WW
Arsenic
0.7 - 170
4.6
5.0
— DR,NH,SS,WW ~~
Cadmi um
0.07 - 18
0.73
2.0
Chromium
2.2 - 72
6.6
17
—
Copper
1.5 - 320
15
12
NH.SS.WW
ab/6r>fit,kg,nh,ss,ww
Lead
0.78 - 7,700
15
2.6
Mercury
0.30 - 720
8.1
7.0
--AB.EW-iNH.SS~
Nickel
1.1 - 21
2.3
8.1
—
Si 1ver
0.24 - 92
7.0
37
—
Zinc
1.7 - 320
12
5.4
NH,SS,WW
a Dry-weight concentration in study area sediments divided by the average concentra-
tion measured in six Carr Inlet sediments (Tetra Tech 1985a).
b The threshold EAR is defined as the ratio of the maximum reference sediment concen-
tration in Puget Sound divided by the average for six Carr Inlet reference sediments.
Above the threshold EAR, the dry-weight concentration of a study area sediment
contaminant would exceed the maximum concentration reported for any Puget Sound
reference site listed in Table 9.
c The contaminant EAR in sediments from at least one station in each listed area
exceeded the threshold level by at least one order of magnitude. The factor of 10 is
arbitrary, but is useful for indicating the areas of greatest contamination. It was
not used in problem area identification or ranking. Sediments in the underlined
areas had the highest observed concentrations.
76
-------
TABLE 11. SUMMARY OF METALS WITH
EAR BETWEEN 100-1,000 AND >1,000
IN SEDIMENTS OF ELLIOTT BAY
AND THE DUWAMISH RIVER*
Chemical Station
EAR = 100-1,000
Arsenic DR-12) SS-03
Copper NH-03, NH-04, SS-03, WW-19
Lead WW-12, WW-14
Mercury AB-01, NH-03
Zinc SS-03, SS-09
EAR >1,000
Lead SS-09
Chemical Areab
EAR = 100-1,000
Lead SS
Mercury AB
a Antimony has been excluded from this table.
b Concentrations averaged over all stations in the
area indicated.
77
-------
2000
CO
Lii .O)
Q jjj 1000
LEGEND
PS Port Susan (n = 4)
MQ Magnolia Bluff (n = 4)
NS Seattle North Waterfront (n => 8)
S3 Seattle South Waterfront (no 11)
NH North Harbor Island (n = 11)
EW East Waterway (n = 16)
WW West Waterway (n = 19)
KG KeBogg Island (n= 11)
OR Ouwamtsh River (n = 18)
AB AIM Beach (n = 4)
THRESHOLD - equivalent to the highest
concentration in Puget Sound reference
areas.
THRESHOLD o
STATIONS
Carr Met Reference = 0.11 mgftg DW.
-------
2000
1900
1000
900
THRESHOLD*/)
9300
1
I
7a
.'4.
I
i
v
'2a
#
I
1
SS-OI 8&03 88-0* 8843
o>
flj
3
>.
2,
UJ
o
z
UJ
oc
UJ
U.
Ui
cc
UJ
>
o
CD
<
<
>
UJ
-I
UJ
SS-06 8847 88-08
STATIONS
SM SS-10 8S-11 88-12
1S00 -
500
THRESHOLD'!
V7\ ?Z\
I
I.
I
m
NH.10 NH40 NKO# NK
-------
considerably less contaminated) and Stations KG-06 and KG-07 near Kellogg
Island.
Arsenic—Arsenic concentrations were not highly elevated in the study
area; the areas of highest contamination appeared to be geographically
isolated. The highest concentration in the study was found at Station SS-03,
along the Seattle South waterfront (584 mg/kg DW; EAR = 173). The adjacent
station, SS-04, had a far lower concentration (28.5 mg/kg DW). The next
highest concentration was observed in the Duwamish River, at Station DR-12
(449 mg/kg DW). Concentrations of arsenic at stations adjacent to Station
DR-12 were near reference values. The West Waterway had somewhat elevated
concentrations. For example, Station WW-12 had an arsenic concentration of
239 mg/kg DW. Stations NH-04 and NH-03, at the mouth of the West Waterway,
had concentrations of 174 and 119 mg/kg DW, respectively.
Cadmium—The highest cadmium concentration was observed at Station
SS-09 (17.2 mg/kg DW; EAR = 18), although adjacent stations had concentra-
tions within the range of Puget Sound reference areas (i.e., <2 mg/kg DW;
Table 9). Station SS-03, far removed from Station SS-09, had a concentra-
tion of 7.16 mg/kg DW. The next highest concentrations were found at
Stations EW-05 (12.4 mg/kg DW) and EW-06 (6.32 mg/kg DW) in the East
Waterway. The nearest East Waterway stations were within Puget Sound
reference levels.
. Chromium—Mean EAR values for study areas in Elliott Bay and the
Duwamish River were very consistent and ranged from 5 to 10, with the
exception of Area SS, which had an EAR of 16. These mean EAR values were
all within the range of Puget Sound reference areas. The highest chromium
concentration, 1,080 mg/kg DW, occurred at Station SS-10. Nearby Station
SS-09 had a concentration of 304 mg/kg DW. Other SS stations had concentra-
tions within the range, of Puget Sound reference areas; concentration
gradients were not readily apparent in this area.
A chromium concentration of 555 mg/kg DW was reported at Station WW-12
in the West Waterway, with concentrations at adjacent stations below
200 mg/kg DW. All other concentrations in the study were within the Puget
Sound reference range.
Copper—Copper concentrations were most elevated in and near the mouth
of the West Waterway (Areas NH and WW) and along the Seattle South waterfront
(Area SS) (Figure 23). The highest concentrations were observed at Stations
NH-03 (2,050 mg/kg DW; EAR = 320) and NH-04 (1,770 mg/kg DW; EAR = 280) at
the mouth of the West Waterway (Figure 24). As was the case for antimony
and arsenic, concentrations at NH stations decreased with distance from the
mouth of the waterway. Station WW-19, also near the mouth of the waterway,
had the next highest copper concentration (1,300 mg/kg DW). Other stations
along the east side of the West Waterway also had elevated concentrations
(Figure 24), although a smooth gradient was not apparent. Station WW-19
clearly had the highest concentration in the waterway based on dry weight or
percent fines normalization.
Along the Seattle South waterfront, Station SS-03 had a concentration of
1,040 mg/kg DW, relatively coarse-grained Station SS-07 had a concentration
of 525 mg/kg DW, and Station SS-09 had a concentration of 350 mg/kg DW.
80
-------
ULl o
¦* THRESHOLD
\7/\
STATIONS
LEGEND
PS Port Susan (n c 4)
HQ Magnolia BluH (n = 4)
NS Seattle North Waterfront (n = 8)
3 S Seattle South Waterfront (n = 11)
NH North Haifoor Island (n= 11)
EW East Waterway (n = 16)
WW West Waterway (n = 19)
KG Kelogg Island (n = 11)
OR Duwamish River (n = 18)
AB Alki Beach (n = 4)
THRESHOLD - equivalent to the highest
concentration in Puget Sound reference
areas.
Carr Inlet Reference = 6.4 mg/kg DW.
Figure 23. Mean elevations above reference (EAR) of copper in sediments from all
study areas.
-------
70 -I
* -
THRESHOLD*
10 -
m
i
i"
i
i
i
O)
>•
h.
V
Ui
u
z
UI
oc
lii
U.
Ui
QC
UI
>
o
ffi
<
z
o
F
<
>
UJ
_l
UI
SS-01 SS-03 SMI SS-03 SS-oe SS47 SfrM SMt 88-10 86-11 88-12
STATIONS
20s aw
threshold*
10
1
Ul
i
7/
1
77-
1
sail
NH.10 Mitt NH-Od NH-07 NH06 NH05 NHO* NHfla NH4B NK01 NK11
STATIONS WEST WATERWAY MOUTH
SO
THRESHOLD*
vz
1H
*
M6lV
SEATTLE
SOUTH
WATERFRONT
NORTH
HARBOR
ISLAND
WEST
WATERWAY
WW-01 I WW-031 WW-051 WW-08 I WW-10 IWW-1! i WW-14 I WW-161 WW-H I WW-»
WW-02 WW-04 WW-06 WVi-09 WW-11 WW-13 WW-15 WVM7 WM-lt
STATIONS
Carr Inlet Reference = 6 4 mg/kg DW.
Figure 24. Elevations above reference of copper at individual
stations in the most contaminated study areas.
82
-------
However, these stations were not adjacent and copper concentrations along the
waterfront were, in general, patchy (Figure 24). Based on normalization to
percent fine-grained material, Station SS-07 had a higher copper concentra-
tion than Station SS-03, but no gradients were apparent with this normaliza-
tion. Other areas were less contaminated, but EAR values of greater than 50
were observed at Alki Beach (Station AB-01, 440 mg/kg DW) and in the Duwamish
River (Station DR-12, 386 mg/kg DW). Concentrations at stations in the
vicinity of these samples were typically within or near the range of Puget
Sound reference stations.
Lead—An extreme range of lead concentrations was observed in this
study. The Seattle South waterfront and the West Waterway were the most
contaminated areas (Figure 25). The highest concentration, observed at
Station SS-09, was over 7 percent lead (71,000 mg/kg DW; EAR = 7,700). Lead
concentrations at adjacent SS stations were over 200 times lower (Figure 26)
and were relatively constant throughout the rest of the area. Station SS-03,
far removed from Station SS-09, had the next highest lead concentration in
this area (646 mg/kg DW).
Very high lead concentrations were reported on the east side of the West
Waterway. Concentrations appeared to decrease with distance from Station
WW-14 (8,730 mg/kg DW; EAR = 950). Moving south from Station WW-14,
concentrations decreased from 1,180 mg/kg DW (WW-12) to roughly 700 mg/kg DW
(WW-11 and WW-09). Station NH-03, at the mouth of the West Waterway, had a
lead concentration of 550 mg/kg DW (Figure 26). Patchy areas of moderate
.concentrations (e.g., EAR >30) were found in other areas, such' as the East
Waterway (Station EW-05), Kellogg Island (Stations KG-04 and K6-05), and
other areas in the Duwamish River (Station DR-12).
Mercury—Mercury concentrations in this study ranged over three orders
of magnitude (see Table 8) but were patchy in the areas of highest contamina-
tion. The highest concentration was observed in the Alki Beach area, at
Station AB-01 (28.8 mg/kg DW; EAR = 720) (Figures 27 and 28). Adjacent
stations had concentrations within the range of Puget Sound reference areas
(Figure 28). A mercury concentration of 10.5 (EAR = 260) was observed at
Station NH-03, near the mouth of the West Waterway. Adjacent stations had
concentrations that were roughly 10 times lower on a dry weight or fines
normalized basis (Figure 28). An isolated high concentration also occurred
in the East Waterway (EW-05; 3.82 mg/kg DW). Concentrations along the
Seattle South waterfront were generally elevated, except at Station SS-01
(Figure 28); SS station concentrations ranged from 0.905 to 3.89 mg/kg DW
(excluding Station SS-01), with a mean EAR of 41.
Silver—Silver concentrations were most elevated along the Seattle
waterfront. The highest concentration was located at intertidal Station
NS-01 (8.27 mg/kg DW; EAR = 92) even though the sediment at this station was
very coarse-grained, with less than 4 percent fine-grained material.
Concentrations decreased moving south along the waterfront, from Station
SS-12 (4.98 mg/kg DW) to Station SS-08 (1.99 mg/kg DW), and then increased at
Stations SS-07, SS-06, and SS-05 (between 4.72 and 5.85 mg/kg DW). Other
concentrations were typically within the range of Puget Sound reference
areas (i.e., <3.3 mg/kg DW; see Table 9).
83
-------
730
THRESHOLD
LEGEND
P S Port Susan (n = 4)
MO Magnofia Blutf (n = 4)
NS Seattle North Waterfront (n = 8)
S3 Seattle South Waterfront (n s 11)
HH North Harbor Island (n= 11)
E W East Waterway (n = 16)
WW West Waterway (n = 19)
KG Kellogg Island (n = 11)
DR Duwamish River (n= 16)
AB Alkl Beach (n = 4)
* THRESHOLD - equivalent to the highest
concentration in Puget Sound reference
areas.
SS NH EW
STATIONS
Canr Inlet Reference = 9.2 mg/kg DW.
Figure 25. Mean elevations above reference (EAR) of lead in sediments from all study areas.
-------
1X
90
ao
70 -
80
SO
40
ao
20 -
10
THRESHOLDS
I
7700
I
l
o>
"S
5
>»
33
ui
o
z
LU
0C
Ui
U.
LU
CC
UI
>
o
m
<
z
o
H
<
>
UJ
_i
Ui
8&01 S849 SS-M SMS SfrM S&OT SS-Oi 8S-0S 8S-10 88-11 88-12
STATIONS
100
*0
80
70
60
50
40
X-
20
10
TnresHour -E3.
1
1
1
NH.10 NH-00 met mar NK0$ NHO5 NH04 NH49 NH4B NK01 NH-11
STATIONS ^ ^ST WATERWAY MOUTH
100
90
ao
70
60
90
40
30
20
10
THRESHOLD^ «4
130
960
& E
K\
0
i1-11
iwwos 1
V
?
1-4'
-121
¥-14 I
I
SEATTLE
SOUTH
WATERFRONT
NORTH
HARBOR
ISLAND
WEST
WATERWAY
WW01 IWW-00 i WW06 i WW-081 WW-10 lWW-12 \ WW-14 \ WW-16 I WW-181 WW-20
WW-02 WW-04 WW-OS WW-09 WW-11 WW-13 WW-15 WW-17 WW-19
STATIONS
Cart Inlet Reference = 9.2 mg/Kg DW.
Figure 26. Elevations above reference (EAR) of lead at individual
stations in the most contaminated study areas.
85
-------
200
Ui
o
z
Ui
DC
UI
u.
UI
DC ~
•c
LU £>
> ®
o 5
m .
< £
¦o
z ^
o
F
<
>
UJ
_j
UI
150
100
50
i THRESHOLD*
T—
PS
XZZL
MG
~r
NS
STy
A
ZZA.
EW WW
~r
KG
r7T71
T
DR
LEGEND
P S Port Susan (n = 4)
HQ Magnolia Bluff (n = 4)
NS Seattle North Waterfront (n = 8)
SS Seattle South Waterfront (n = 11)
NH North Harbor Island (n = 11)
E W East Waterway (n = 16)
WW West Waterway (n = 19)
KO Kellogg Island (n = 11)
DR Duwamish River (n = IB)
AB AIM Beach (n = 4)
* THRESHOLD - equivalent to the highest
concentration in Puget Sound reference
areas.
STATIONS
Carr Mel Refersnoe = 0.04 m^kg DW.
Figure 27. Mean elevations above reference (EAR) of mercury in sediments from all
study areas.
-------
200
130
THRESHOIOJ »
M
I
1
¦
1
1
ill
o>
a>
5
>.
w
2,
UJ
o
z
UJ
CE
UJ
U.
UJ
oc
UJ
>
o
CD
<
h-
<
>
UJ
¦J
UJ
ISO
I
THRESHOLD* ¦
0
T77\
NH-10 NH09 NH-00 NH07 NH06 NH4S NH44 NH03 NH-02 N*01 NH-11
STATIONS WEST WATERWAY MOUTH
200 -I
threshold; i
0 —
IZZL
-v:
AB-04
A8-03 AB-02
STATIONS
A&01
SEATTLE
SOUTH
WATERFRONT
ss-oi sms sfro* ss-os ssoe ss-07 ss-oe s&oa ss-io ss-n ss-ia
STATIONS
NORTH
HARBOR
ISLAND
rrr\
ALKI
BEACH
Carr Inlet Reference = 0.04 mg/kg DW.
Figure 28. Elevations above reference (EAR) of mercury at
individual stations in the most contaminated study areas.
87
-------
Zinc--Zinc concentrations were elevated above Puget Sound reference
conditions in much of the study area, although the most highly elevated
concentrations were along the Seattle South waterfront (Figure 29). The
maximum concentration, 6,010 mg/kg DW (EAR = 320), occurred at Station
SS-09. The next highest concentration in the study was observed at Station
SS-03 (4,830 mg/kg DW). Zinc concentrations at intervening stations were
relatively constant and were more than 10 times lower than observed at
Stations SS-09 and SS-03 (Figure 30).
The area near the mouth of the West Waterway aiso naa nighiv elevated
zinc concentrations, particularly Stations NH-03 and NH-04 (1,300 and
994 mg/kg DW, respectively) at the mouth, and Stations WW-12 and WW-19
(1,170 and 705 mg/kg DW, respectively) on the east side of the waterway
(Figure 30). West of the mouth of the waterway, concentrations were greater
than 600 mg/kg at Stations NH-06 and NH-08. However, these two stations
were not adjacent and were separated by a less contaminated, coarse-grained
sediment station (Figure 30).
High zinc concentrations were observed in isolated areas of contamina-
tion in the Duwamish River (e.q., near 1,000 mg/kg DW at KG-01 and DR-15).
Somewhat lower concentrations (between 400 and 800 mg/kg DW) were observed
in the East Waterway (Stations EW-05, EW-06 and EW-15).
Sediment Chemistry; Organic Compounds
The ranges in concentration (ug/kg DW) and detection frequencies of
semivolatile compounds detected at least once in the study area are presented
in Table 12.
Acid and neutral priority pollutants that were searched for but not
found in any samples were 2-chlorophenol, 2,4-dichlorophenol, 2,4,5- and
2,4,6-trichlorophenol, 4-chloro-3-methylphenol, 2-chloronaphthalene,
1,2- and 1,3-dichlorobenzene, hexachlorobenzene, isophorone, and hexachloro-
butadiene. Priority pollutant pesticides that were searched for but not
detected in any samples included alpha-, beta-, and gamma-hexachlorocyclo-
hexane (HCH). TIOs searched for but not found were tri-, tetra-, and penta-
chlorobutadienes, coprostanol, and alpha-tocopheryl acetate.
Relatively high detection limits may have precluded the detection of
certain semivolatile organic compounds at moderate to high concentrations.
Among the undetected or infrequently detected compounds that had high
detection limits were all chlorinated benzenes, hexachlorobutadiene, several
substituted phenols (e.g., pentachlorophenol and other chlorinated phenols),
benzoic acid, and benzyl alcohol. Acid/neutral organic compounds for which
detection limits exceeded existing Puget Sound AET (when available) for more
than roughly half the stations are presented in Table 13. For the compounds
in Table 13, characterization of spatial concentration gradients is incon-
clusive in most areas. The implications of detection limits for data
analysis will be discussed on a chemical-by-chemical basis in the following
section, when appropriate.
The 16 EPA priority pollutant PAH (2-6 rings) were the most frequently
detected organic compounds in the study and occurred at the highest concen-
trations of all organic compounds (Table 12). Wide concentration ranges
88
-------
Ui
o
z
Ui
oc
Ui
u.
Ui
DC ~
JZ
Ui .J?
> ®
O 5
m .
< b
¦o
z
o
K
<
>
Ui
J
Ui
70
60
50
40
30
20
10
-« THRESHOLD
2
i
PS
v.
rz7i
vl
MG NS
LEGEND
PS Port Susan (n = 4)
MO Magnolia Bluff (n = 4)
NS Seattle North Waterfront (n = 8)
SS Seattle South Waterfront (n = 11)
NH North Haibor Island (n = 11)
E W East Waterway (n = 16)
WW West Waterway (n = 19)
KQ Kelogg Wand (n = 11)
DR Duwamish River (n = 18)
AB AIM Beach (n = 4)
* THRESHOLD - equivalent to the highest
concentration in Puget Sound reference
areas.
STATIONS
Carr Inlel Reference » 19 m^kg DW.
-------
60 -
20
10
THRESHOLD* •
0
290
^7
i
380
ti,
UJ
o
z
UJ
DC
UJ
u.
ULt
oc
•UJ
>
o
m
<
H
<
>
Ui
-I
UJ
SS-01 SS-03 SS-04 SS-OS SS-OS SS07 SS-06 SS49 SS-tQ 88-11 88-12
STATIONS
eo
50
10
THRESHOLD'
0
'/<
//
'//.
V/
y/,
1
I
4*
V/,
m
NH-10 NH40 NH08 NH47 NH46 NKOS NHO* f**09 NH4B NH61 NH>11
STATIONS ^WESTVMTERMAYMOaTH
70
,0 y?
THRESHOLD* » ' /
\
t
STATIONS
SEATTLE
SOUTH
WATERFRONT
NORTH
HARBOR
ISLAND
WEST
WATERWAY
WW411 MMMBl WMMSI W*W#I WW-101WW-121 WW-1*I MAM-IS I MMMI1WW40
WW-02 WW-04 WW-08 WW-08 WW-11 WW-13 WW-13 WW-17 WW-11
Carr Inlet Reference = 19 mgftg 0W.
Figure 30. Elevations above reference (EAR) of zinc at individual
stations in the most contaminated study areas.
90
-------
TABLE 12. CONCENTRATIONS OF DETECTED SEMIVOLATILE ORGANIC
COMPOUNDS IN SURFACE SEDIMENTS OF ELLIOTT BAY,
DUUAMISH RIVER, AND PORT SUSANa
Chemical
Range
(ug/kg dry wt)b
Detection
Frequency
Location of
Maximums
Low molecular weight PAH
E7
_
630,000
105/107
SS-08
naphthalene
XE2
-
15,000
66/107
NH-06
acenaphthylene
El.7
-
X37,000
62/107
SS-08
acenaphthene
U6
-
33,000
62/107
SS-08
fluorene
U3
-
37,000
65/107
SS-08
phenanthrene
E5.2
-
330,000
105/107
SS-08
anthracene
E2
-
190,000
93/107
SS-08
High molecular weight PAH
E24
E3,200,000
107/107
SS-08
fluoranthene
E9.4
-
1,300,000
107/107
SS-08
pyrene
E8.0
-
740,000
107/107
SS-08
benz(a)anthracene
U2
-
300,000
101/107
SS-08
chrysette
U2.4
-
350,000
102/107
SS-08
benzofluoranthenes
E9.3
-
300,000
94/107
SS-08
benzo(a)pyrene
U3
-
E100,000
92/107
SS-08
indeno(1,2,3-c,d)pyrene
El .0
-
E40,000
85/107
SS-08
dibenzo(a,h)anthracene
El .0
-
El2,000
63/107
SS-08.
benzo(g,h,i jperylene
~El.2
-
E32,000
83/107
SS-08
Total PCBs
E59
-
E5,800
79/106
DR-08
Chlorinated benzenes
1,4-di chlorobenzene
U1.0
-
X31,000
7/107
SS-09
1,2,4-trichlorobenzene
U39
-
E15c
1/107
CQR-0S>
Phthalates
dimethyl phthalate
E0.70
-
XI,400
41/107
SS-11
butyl benzyl phthalate
U1
-
1,800
60/107
EW-07, EW-08,
EW-09
di-n-octyl phthalate
U1
-
Z9,900
40/106
NS-06
Pesticides
p,p'-DDE
Ul.l
-
62
4/107
DR-1QV
p,p'-DDD
U1.7
-
E140
9/107
SS-09
p,p'-DDT
U1.4
-
E270
6/107
KG-06
aldrin
U1.0
-
E90
2/107
NH-10
delta-HCH
U0.8
-
L17
1/107
WW-06
dieldrin
U1.3
-
E51
1/106
NH-10
endrin
U1.4
-
E8C
1/107
NH-10
endrin aldehyde
U1.9
-
68
2/107
heptachlor
U0.9
-
E6C
1/107
NH-10
chlordane
U6.4
-
L200
1/106
EW-05
91
-------
TABLE 12. (Continued)
Chemical
Range
(ug/kg dry wt)b
Detection
Frequency
Location
Maximum®
Phenols
phenol
2-methylphenol
4-methylphenol
2,4-dimethylphenol
pentachlorophenol
Miscellaneous extractables and
tentatively identified compounds
benzyl alcohol
benzoic acid
dibenzofuran
2-methylnaphthal ene
1-methylphenanthrene
2-methylphenanthrene
3-methylphenanthrene
biphenyl
retene
carbazole
XE0.9
U3
E2
U2.6
U38
E5.2
U25
U2.9
E0.3
Urf
UH
u j
UH
UH
ud
XI,200
E280
2,600
E210
X6,000
E8,800
6,300
7,100
3,300
E100,000
El 10,000
E97,000
El,800
E10,000
El,200
36/107
2/107
32/107
2/107
8/107
5/107
1/106
80/107
66/107
70/107
71/107
69/107
27/107
56/107
45/107
DR-25
DR-25
WW-19
SS-07
NH-04-
WW-02
KG-10
SS-08
SS-08
SS-08
SS-08
SS-08
NH-06
KG-10
NH-06
a Qualifiers:
U = Substance undetected at the detection limit shown.
E = Estimated value.
X = The surrogate recovery for this compound was low (<10 percent). Hence, the
recovery correction was at least a factor of 10.
L = The reported concentration is the mean of a detected value and a detection 1 imit.
Z = Value corrected for blank contributions; the resulting value still exceeds the
detection limit.
b Maximum is the highest detected value even if maximum detection limits were higher.
c Maximum detected concentration does not exceed the maximum concentration in Puget
Sound reference areas.
^ This tentatively identified organic compound was not found during a mass spectral
search of a sample extract. Actual detection limits for tentatively identified
compounds were not assigned in these cases.
92
-------
TABLE 13. INFREQUENTLY DEtECTED ACID/NEUTRAL
COMPOUNDS WITH HIGH DETECTION LIMITS3
No. Samples with
LAETb
(ug/kg DW)
Detection
Limits
Range of
Detection Limit'
Chemical
>100
>1,000
>10,000
1,2-dichlorobenzene
35
65
24
0
U2
- U7,300
1,3-dichlorobenzene
>170
83
27
0
U2
- U7,300
1,4-dichlorobenzene
110
71
27
0
U2
- U7,300
1,2,4-trichlorobenzene
31
73
35
14
U39
- U73,000
hexachlorobenzene
70
66
25
3
U18
- U99,000
hexachlorobutadiene
120
105
56
22
U34
- U150,000
2,4-dichlorophenol
NAC
45
26
12
U9
- U20,000
2,4,5-trichlorophenol
NA
41
24
5
U8
- U14,000
2,4,6-trichlorophenol
NA
39
24
6
U7
- U19,000
pentachlorophenol
>140
94
62
17
U3 8
- U84,000
2,4-dimethylphenol *
29
44
17
0
U3
- U7,300
benzoic acid
650
56
28
11
U25
- U46,000
benzyl alcohol
57
81
15
4
U33
- 1122,000
a Criteria for inclusion in this table: compound is undetected in more than roughly
half the samples in the study at detection limits above the lowest Puget Sound AET
(Tetra Tech 1986c, 1987) or above 100 ug/kg DW if no AET value is available.
b Lowest of AET values based on benthic infaunal abundances and amphipod, oyster
larvae, and Microtox bioassays; taken from Tetra Tech (1986c, 1987). The ">" values
shown in this table are considered less reliable than unqualified AET and were not used
during problem identification.
c Not available.
93
-------
(e.g., over five orders of magnitude) were observed for individual PAH.
Maximum concentrations of individual PAH (except naphthalene) were observed
along the Seattle South waterfront (Station SS-08); naphthalene maximized at
Station NH-06 (west of the mouth of the West Waterway) (Table 12). The
total PAH concentration at Station SS-08 was roughly 4,000,000 ug/kg DW
(i.e., 0.4 percent DW). Among the TIOs typically associated with PAH,
alkylated PAH (e.g., methyl naphthalene and methyl phenanthrenes), and
dibenzofuran also maximized at Station SS-08. Maximum concentrations of
biphenyl and carbazole were found at Station NH-06.
The Seattle South waterfront had maximum concentrations of chemicals
other than PAH. For example, 1,4-dichlorobenzene and p,p'-DDD maximized at
Station SS-09 (as did cadmium, lead, and zinc) (see Tables 8 and 12).
Maximum concentrations of a number of chemicals were found in various
sections of the Duwamish River. The maximum PCB concentration was observed
in the Duwamish River (Station DR-08), as were the maximum concentrations of
several chlorinated pesticides (e.g., p.p'-DDE at Station DR-10 and endrio
aldehyde at Station DR-08). Phenol and 2-methylphenol concentrations
maximized at intertidal Station DR-25.
Compounds detected infrequently (i.e., one or two times) in the study
included 1,2,4-trichlorobenzene, 2-methylphenol, 2,4-dimethylphenol, benzoic
acid, and seven chlorinated pesticides (aldrin, delta-HCH, dieldrin, endrin,
endrin aldehyde, heptachlor, and chlordane) (see Table 12). These compounds
will not be considered for describing spatial correlations among chemicals,
but will be included in the chemical characterization of problem areas (see
PRIORITIZATION OF PROBLEM AREAS AND CONTAMINANTS).
Volatile organic compounds were measured at 28 of the 107 sediment
stations. The concentration ranges and detection frequencies of compounds
detected at least once are presented in Table 14. In general, concentration
ranges were narrow and compounds were detected infrequently. Total xylenes
(i.e., ortho-, meta-, and para- isomers) were most frequently detected among
volatile organic compounds; the maximum concentration occurred in the West
Waterway (Station WW-11). Trichloroethene was the next most frequently
detected compound, but never exceeded 4 ug/kg DW (near detection limits).
Other volatile organic compounds were detected only once or twice, and at
relatively low concentrations. The maximum values were found near Kellogg
Island; Station KG-05 had maximum concentrations of five of the eight
volatile pollutants detected in this study. Data for three volatile
compounds were rejected because of potentially high laboratory contamination
(see METHODS, Sediment Chemistry, Quality Assurance/Quality Control Results).
Organic Compounds of Concern-
As described for metals, chemicals of concern are those chemicals that
occur at concentrations exceeding the range of Puget Sound reference areas.
The range of concentrations of organic compounds in Puget Sound reference
areas are summarized in Table 15. Concentrations in Port Susan sediments
analyzed for this study (PS-01 to PS-04) are included in Table 15. Station
PS-05 (this study) was. excluded from this table because it had PAH concentra-
tions that were far above those characteristic of reference areas. In fact,
the high molecular weight PAH concentration at Station PS-05, 28,000 ug/kg
DW, exceeded existing AET based on amphipod mortality, oyster larvae
94
-------
TABLE 14. CONCENTRATIONS OF VOLATILE ORGANIC
COMPOUNDS IN SURFACE SEDIMENTS OF ELLIOTT BAY
AND THE DUWAMISH RIVERa
Chemi cal
Range
(ug/kg dry wt)b
Detection
Frequency
Location of
Maxi mumb
Total xylenes
E0.1
- E64
24/28
WW-11
Trichloroethene
E0.5
- E3.4
6/28
EW-09
Trans-1,2-dichloroethene
E0.8
- E8.9
2/28
KG-05
Toluene
U0.6
- Z54
3/28
KG-05
Ethyl benzene
U0.7
- Z10
1/28
NH-08
Chlorobenzene
U0.2
- Z26
2/28
KG-05
Tetrachloroethene
U0.1
- Z8.3
1/28
KG-05
1,1-Dichloroethane
U0.6
- El .1
1/28
KG-05
a On.ly compounds that were detected one or more times are listed.
Qualifiers:
U = Substance undetected at the-detection limit shown.
Z = Value corrected for blank contributions; the resulting value still-
exceeds the detection limit.
E = Estimated value.
b Maximum is highest detected value even if maximum detection limit was
higher.
95
-------
TABLE 15. SUMMARY OF ORGANIC COMPOUND CONCENTRATIONS
IN SEDIMENTS FROM PUGET SOUND REFERENCE AREAS3
Chemical
Range (ug/kg dry wt)&
Detection
Frequency
Reference
Sitesc
Low molecular weight PAH
4-L71
(L45-L71)
13/13
i,a,9-
naphthalene
U0.5-U40
(X2-U26)
12/27
1,2,3,4,5,6,8,9
acenaphthylene
U0.1-U40
(U4.1-U9.3)
2/27
1,2,3,4,5,6,8,9
acenaphthene
U0.1-U40
(U6.7-U14)
4/27
1,2,3,4,5,6,8,9
f1uorene
UO.1-40
(U6.7-U13)
7/28
All
phenanthrene
4-170
(E5.2-E17)
18/24
1,2,3,6,7,8,9
anthracene
U0.5-U40
(E2-U20)
11/24
1,2,3,6,7,8,9
High molecular weight PAH
34-L100
(E48-L100)
13/13
1,8,9
fluoranthene
5-100
(E9.4-E19)
24/29
All
pyrene
5-120
(E8.0-E19)
23/29
All
benz(a)anthracene
2-U40
(E2.8-E6.5)
15/24
1,2,3,6,7,8,9
chrysene
4-U40
(E5.0-E13)
15/24
1,2,3,6,7,8,9
benzo(b)fluoranthene
U5-94
(U5.9-E8.3)
15/25
1,2,3,4,5,6,7,8
benzo(k)fluoranthene
E4.8-94
(E4.8-U7.5)
15/25
1,2,3,4,5,6,7,8
benzo(a)pyrene
UO.37-40
(U3.8-E15)
16/21
1,3,4,5,6,7,8,9
indeno(l,2,3-c,d)pyrene
UO.37-30
(E2.2-U8.2)
10/19
1,4,5,6,7,8,9
dibenzo(a,h)anthracene
U0.4-E10
(E1-U13)
3/12-
1,8,9
benzo(g,h,i)perylene-
El.2-20
(E1.2-U7.6)
8/13
1,7,8,9
Total PCBs
3.1-U50d
(U120-U170)
7/22
1,2,3,4,6,7,9
Chlorinated benzenes
U0.06-U40<|
1,4-dichlorobenzene
(U20-U120)
1/23
1,2,3,4,5,8,9
1,2,4-tri chlorobenzene
U0.5-U16d
(U75-U190)
0/9
1,9
Phthalate esters
dimethyl phthalate
U0.5-U50
(E0.7-U7.5)
1/12
1,8,9
butyl benzyl phthalate
U0.5-U25
(U3.7-E14)
3/12
1,8,9
di-n-octyl phthalate
U0.5-E56
(E3.0-E56)
4/12
1,8,9
Pesticidese
p,p'-DDE
U1.6-U10
(U1.6-U2.1)
0/8
1,8,9
p,p'-DDD
U1.9-U10
(U1.9-U2.6)
0/9
1,8,9
p.p.'-DDT
U1.0-U10
(U1.7-U2.3)
0/8
1,8,9
aldrin
U0.5-U10
(U1.4-U1.9)
0/9
1,8,9
chlordane
U5-U50
(U23-U31)
0/13
1,8,9-
endrin aldehyde
U2.3-U10
(U2.3-U3.2)
0/5
1,8
delta-HCH
U0.5-U10
(U1.0-U1.4)
0/9
1,8,9
dieldrin
U1-U10
(U1.3-U1.8)
0/9
1,8,9
endrin
U1-U10
(U1.8-U2.4)
0/9
1,8,9
heptachlor
U0.5-U10
(U1.3-U1.8)
0/9
1,8,9
96
-------
TABLE 15. (Continued)
Chemical
Range (ug/kg dry wt)b
Detection
Frequency
Reference
Sites0
Phenols
phenol
2-methylphen9l
4-methylphenol
2,4-dimethylphenol
pentachlorophenol
U0.5-62f (U3.3-E33)
U0.7-U50 (U3.3-U11)
U0.8-290 (E2-U8.3)
U1-U14 (U4.7-U14)
0.1-U50d (U38-U220)
5/17
0/11
7/11
0/13
1/10
1,2,3,8
1,8,9
1,8,9
1,8,9
Miscellaneous extractables
benzyl alcohol
benzoic acid
dibenzofuran
2-methylnaphthalene
1-methylphenanthrene9
2-methylphenanthrene9
3-methylphenanthrene9
biphenyl9
retene9
carbazole9
U3.4-U20
-------
TABLE 15. (Continued)
^ An anomalously high phenol value of 1,800 ug/kg dry wt was found at one Carr Inlet
station (Tetra Tech 1985a). For the purpose of reference area comparison, this value
has been excluded. Data from Site 9 were excluded because laboratory contamination
of phenol was observed during analysis of these reference area samples
9 Tentatively identified compound.
^ U - This tentatively identified compound was not found during a mass spectral
search of reference sample extracts. Actual detection limits for tentatively identified
compounds were not assigned in these cases.
References:
(Site 1
(Site 2
(Site 3
(Site 4
(Site 5
(Site 6
(Site 7
(Site 8
(Site 9
Tetra Tech (1985a); Mowrer et al. (1977)
Battelle (1986)
Battelle (1986); Prahl and Carpenter (1979)
Mai ins et al. (1980); Mowrer et al. (1977)
Mai ins et al. (1980)
Mai ins et al. (1982)
Barrick and Prahl (1987); Mowrer et al. (1977)
This study; Port Susan Stations PS-01 through PS-04
PTI and Tetra Tech (1988).
98
-------
abnormality, and Microtox bioassays (Tetra Tech 1987). The remaining four
Port Susan stations from this study had concentrations of PAH that were
roughly two orders of magnitude lower and concentrations of other organic
compounds that were typically within the existing range of Puget Sound
reference areas (Table 15).
The three semivolatile organic compounds that did not exceed the range
of reference concentrations were 1,2,4-trichlorobenzene, endrin, and
heptachlor (see Tables 12 and 15); these three compounds were detected only
once in the study (see Table 12). These compounds are thus of relatively
minor concern and will not be discussed further.
Puget Sound reference area data were not available for most volatile
organic compounds. In the absence of reference area data, volatile contami-
nants were considered of concern only if they exceeded a concentration of
10 ug/kg DW at one or more stations. The concentration of 10 ug/kg was
chosen as a reasonable detection limit for relatively uncontaminated
sediments (e.g., Tetra Tech 1986f). Based on this assigned concentration,
the volatile organic compounds of concern are total xylenes, toluene, and
chlorobenzene. Of these compounds, only total xylenes were detected more
than three times in the study. The highest concentrations of chlorobenzene
and toluene were qualified with a Z, indicating that these compounds were
also found in blanks. Although concentrations exceeded those of corres-
ponding blanks, the blank contamination of volatile compounds is often
variable. Hence, these data are considered less reliable than total xylene
data. Volatile organic compounds will not be discussed further in this
section but were considered during problem area identification.
Distributions of selected contaminants of concern are summarized in
terms of EAR values in Table 16. Summaries of distributions of organic
chemicals with EAR between 100 and 1,000 are presented in Tables 17 and 18;
distributions of chemicals with EAR greater than 1,000 are presented in
Tables 19 and 20. The distributions of the organic chemicals of concern
that were highly elevated and fairly widespread in the study area are
described below.
Polvcvclic Aromatic Hydrocarbons—To facilitate data analysis and to
maximize comparability with data analyses performed for other Puget Sound
studies, the 16 individual EPA priority pollutants were treated as two
groups: LPAH and HPAH. This grouping was considered to be a reasonable data
reduction method because the concentrations of individual PAH within each
group tended to correlate well. It is also a reasonable grouping for
potential source correlations, as relatively high concentrations of LPAH are
characteristic of petroleum-derived materials whereas relatively high
concentrations of HPAH are more characteristic of combustion-derived
materials (e.g., Readman et al. 1982; Prahl and Carpenter 1983: Tetra Tech
1985a).
A Pearson correlation analysis including all individual PAH data was
performed (see Appendix C) and scatterplots of the correlations were
analyzed. Correlation coefficients (r) among the six LPAH (using dry weight
concentrations) ranged from a relatively poor 0.29 to a very strong >0.99
(P<0.05). However, of all 16 unique correlations among these six chemicals,
11 had r values of greater than 0.95. The correlations of naphthalene with
99
-------
TABLE 16. RANGE IN EAR FOR SELECTED ORGANIC CONTAMINANTS OF
CONCERN IN SEDIMENTS OF ELLIOTT BAY AND THE DUWAMISH RIVER3
EARb
Chemical
Range
Median0
Threshold**
Areas where Threshold
Exceeded by 10 Timese
LPAH
0.2 ¦
¦ 15,000
32
1.7
AB,DR,EW,KG,NH,NS,SS,WW
HPAH
0.3 ¦
¦ 41,000
75
1.3
AB,DR,EW,KG,NH,SS,WW
Total PCBs
9.8 -
- 970
80
8.3
DR,EW,KG,NH,SS,WW
1,4-Dichlorobenzene
0.3 ¦
- 8,900
17
11
SS
Dimethyl phthalate
0.02 •
- 35
0.5
1.3
SS,WW
Butyl benzyl phthalate
0.06 •
¦ 110
1.2
1.5
DR,EW,KG,SS
Di-n-octyl phthalate
0.05 ¦
•490
0.9
2.8
DR,NS,SS
p,p'-DDD
0.17 ¦
¦ 14
0.97
1.0
NH, SS
p,p'-DDT
0.14 ¦
- 27
0.87
1.0
EW, KG, SS
Phenol
0.03 ¦
- 36
0.67
1.9
DE
4-Methylphenol
0.15 •
- 200
2.0
22
--
Pentachlorophenol
1.1 ¦
- 180
2.8
1.5
EW,NH
Benzyl alcohol
0.5 •
¦ 880
8.3
2.0
EW,SS,WW
Dibenzofuran
0.8 ¦
- 1,900
24
3.8
AB,DR,EW,KG,NH,NS,SS,WW
2-Methyl naphthalene
0.1 ¦
¦ 900
12
6.0
AB,EW,KG,NH,SS,WW
Retene
U*
- 370
3.3
4.8
EW,KG,SS
a This table includes only compounds that were detected five times or more in the study. Certain
TIOs were not included because of the unavailability of reference concentrations.
b Dry-weight concentration in study area sediments divided by the average concentration measured
in six Carr Inlet sediments (taken from Tetra Tech 1985a).
c Medians are based on data after exclusion of detection limits >100 ug/kg DW for acid/neutral
compounds and PCBs, and >25 ug/kg DW for single component pesticides.
d The threshold EAR is defined as the ratio of the maximum reference sediment concentration in
Puget Sound divided by the average for six Carr Inlet reference sediments. Above the threshold
EAR, the dry-weight concentration of a study area sediment contaminant would exceed the maximum
concentration (or detection limit) reported for any Puget Sound reference site listed in Table
15.
e The contaminant EAR in sediments from at least one station in each listed area exceeded the
threshold level by at least one order of magnitude. The factor of 10 is arbitrary, but is useful
for indicating the areas of greatest contamination. It was not used in problem area identification
or ranking. Sediments in the underlined areas had the highest observed concentrations.
f Retene is a tentatively identified compound for which detected limits are not reported. Hence,
the lower end of the range is unknown.
100
-------
TABLE 17. SUMMARY OF SEMIVOLATILE ORGANIC CHEMICALS
WITH EAR BETWEEN 100-1,000 IN SEDIMENTS
OF ELLIOTT BAY AND THE DUHAMISH RIVER
(USING AREA MEANS)3
Chemical Area
Low molecular weight PAH EW, NH
naphthalene AB, NH, SS
acenaphthene EW, NH
fluorene EW, NH
phenanthrene AB, EW, NH, WW
anthracene AB, EW, NH
High molecular weight PAH AB, EW, NH, NS, WW
fluoranthene AB, EW, NH, NS, WW
pyrene AB, EW, NH, NS, WW
benz(a)anthracene AB, EW, NH, NS, WW
chrysene EW, NH, WW
benzofluoranthenes AB, EW, NH, NS, WW
benzo(a)pyrene AB, EW, NH, NS, WW
indeno(l,2,3-c,d)pyrene AB, EW, NH, NS, WW
dibenzo(a,h)anthracene NH, SS
benzo(g,h,i)perylene AB, EW, NH
Total PCBs DR, EW, KG, NH, SS
2-Methylnaphthalene KG, NH, SS
Dibenzofuran EW, NH, SS
Benzyl alcohol WW
1,4-Dichlorobenzene SS
a Concentrations averaged over all stations in the area
indicated.
Detection limits >100 ug/kg DW have been excluded from these
means.
101
-------
TABLE 18. SUMMARY OF SEMIVOLATILE ORGANIC CHEMICALS
WITH EAR BETWEEN 100-1,000 IN SEDIMENTS OF ELLIOTT BAY
AND THE DUWAMISH RIVER (USING INDIVIDUAL STATIONS)
Chemi cal
Station
Low molecular weight PAH
naphthalene
acenaphthylene
acenaphthene
fluorene
phenanthrene
anthracene
High molecular weight PAH
fluoranthene
pyrene
AB-01, EW-02, EW-04, EW-06, EW-14, NH-03,
NH-04, NH-05, NH-08, NS-07, SS-03 to SS-07, SS-09,
SS-10, SS-11, WW-04, WW-09, WW-12
AB-01, AB-02, NH-03, NH-05, SS-04 to SS-11
EW-06, EW-14, NH-04, NH-06, NH-08, SS-04 to SS-
07, SS-09, SS-10
AB-01, BR-16, EW-12, EW-14, NH-03, NH-04, NH-05,
NS-08, SS-03 to SS-07, SS-09, WW-04, WW-09
AB-01, OR-16, EW-06, EW-14, NH-03 to NH-05, NS-07,
SS-03 to SS-07, SS-09, WW-04, WW-09
AB-01, AB-02, DR-12, DR-16, EW-04, EW-06, EW-11,
EW-12, EW-14, KG-01, KG-07, NH-02 to NH-05, NS-02,
NS-07, NS-08, SS-03 to SS-07, SS-09 to SS-11, WW-
04, WW-06, WW-09 to WW-14, WW-16 to WW-18
AB-01, DR-16, EW-02, EW-04, EW-06, EW-11, EW-14,
NH-03 to NH-05, NH-08, NS-07, NS-08, SS-03 to SS-
05, SS-07, SS-09 to SS-11, WW-04, WW-09, WW-11,
WW-12, WW-14, WW-16
AB-01, AB-02, DR-16, EW-02 to EW-04, EW-06, EW-07,
EW-09, EW-11, EW-14, EW-15, KG-01, KG-07, KG-09,
NH-01 to NH-05, NS-07, NS-08, SS-03 to SS-07, SS-
10, SS-11, WW-04, WW-06, WW-09 to WW-14, WW-16 to
WW-19
AB-01, DR-02, DR-12, DR-16, EW-02, EW-04, EW-06,
EW-07, EW-09, EW-11, KG-01, KG-07, NH-01 to NH-05,
NH-08, NS-07, NS-08, SS-03 to SS-07, SS-09 to SS-
11, WW-06, WW-09 to WW-14, WW-16 to WW-19
AB-01, AB-02, DR-12, DR-16, EW-02, EW-04, EW-06,
EW-07, EW-09, EW-11, EW-12, EW-15, KG-01, KG-07,
KG-09, NH-01 to NH-05, NS-07, NS-08, SS-03 to SS-
05, SS-07, SS-10 to SS-12, WW-04, WW-06, WW-09 to
WW-14, WW-16 to WW-19
102
-------
TABLE 18. (Continued)
Chemical Station
benz(a)anthracene
AB-01, AB-02, DR-08, DR-16, EW-02 to EW-04, EW-06,
EW-07, EW-09 to EW-12, EW-14, EW-15, KG-01, KG-07,
KG-09, NH-01 to NH-05, NS-07, NS-08, SS-03 to SS-
07, SS-10 to SS-12, WW-04, WW-06 to WW-14, UW-16
to WW-19
chrysene
AB-01, EW-02, EW-04, EW-06 to EW-12, EW-14, EW-15,
KG-01, KG-07, KG-09, NH-01 to NH-05, NS-07, NS-08,
SS-03 to SS-07, SS-10 to SS-12, WW-04, WW-06, WW-
09 to WW-14, WW-16 to WW-19
benzof1uoranthenes
AB-01, AB-02, DR-16, EW-02 to EW-04, EW-06 to EW-
09, EW-11, EW-12, EW-14, EW-15, KG-01, KG-09, NH-
01 to NH-05, NS-07, NS-08, SS-03 to SS-05, SS-07,
SS-10, SS-11, WW-04, WW-06 to WW-14, WW-16 to WW-20
benzo(a)pyrene
AB-01, AB-02, DR-08, DR-16, EW-02 to EW-04, EW-06
to EW-09, EW-11, EW-14, EW-15, KG-01, KG-09, NH-01
to NH-05, NH-08, NS-07, NS-08, SS-03 to SS-07, SS-
10, SS-11, WW-04, WW-06, WW-09 to WW-14, WW-16 tO
WW-19
indeno(1,2,3-cd)pyrene
AB-01, DR-08, DR-12, DR-16, EW-02, EW-04, EW-06,
EW-07, EW-11, EW-14, EW-15, KG-01, NH-01, NH-04,
NH-05, NH-08, NS-07, NS-08, SS-03 to SS-07, SS-09
to SS-11, WW-04, WW-09, WW-11, WW-14, WW-17 to WW-
20
dibenzo(a,h)anthracene
AB-01, DR-16, EW-06, EW-14, NH-03, NH-04, NH-06,
NH-08, NS-07, SS-03 to SS-07, SS-09, WW-04, WW-17
benzo(g,h,i)perylene
AB-01, DR-08, DR-12, DR-16, EW-04, EW-06, EW-11,
EW-14, EW-15, KG-01, NH-01, NH-04 to NH-06, NH-08,
NS-07, NS-08, SS-03 to SS-07, SS-09 to SS-11, WW-
04, WW-09, WW-11, WW-17, WW-18
Total PCBs
DR-03, DR-08, DR-10 to DR-14, DR-16, DR-17, EW-02,
EW-03, EW-05, EW-06, EW-11, KG-06, NH-03, NH-06,
NH-08, SS-04, SS-09, WW-05, WW-06, WW-08, WW-09,
WW-16
2-Methylnaphthalene
AB-01, KG-06, KG-10, NH-05, NH-06, NH-08, SS-06 to
SS-08
103
-------
TABLE 18. (Continued)
Chemical Station
Dibenzofuran AB-01, DR-16, EW-02, KG-10, NH-03, NH-05, NH-06,
NH-08, SS-03 to SS-07, SS-09, WW-04
Retene KG-10
1,4-Dichlorobenzene SS-03
4-Methylphenol KG-09, WW-19
PCP NH-04
Benzyl alcohol SS-03, WW-02
Butyl benzyl phthalate EW-07 to EW-09
Di-n-octyl phthalate NS-06
104
-------
TABLE 19. SUMMARY OF SEMIVOLATILE ORGANIC
CHEMICALS WITH EAR >1,000 IN SEDIMENTS
OF ELLIOTT BAY AND THE DUHAMISH RIVER
(USING AREA MEANS)
Chemical Area*
Low molecular weight PAH SS
acenaphthylene SS
acenaphthene SS
fluorene SS
phenanthrene SS
anthracene SS
High molecular weight PAH SS
fluoranthene SS
pyrene SS
benz(a)anthracene SS
chrysene SS
benzofluoranthenes SS
benzo'( a) pyrene SS
indeno(l,2,3-c,d)pyrene SS
benzo(g,h,i)perylene SS
a Concentrations averaged over all stations in the
area indicated.
Detection limits >100 ug/kg DW have been excluded
from these means.
105
-------
TABLE 20. SUVMRY OF SEMI VOLATILE ORGANIC
CHEMICALS WITH EAR >1,000 IN SEDIMENTS
OF ELLIOTT BAY AND THE DUWAMISH RIVER
(USING INDIVIDUAL STATIONS)
Chemical
Station
Low molecular weight PAH
naphthalene
acenaphthylenie
acenaphthene
fluorene
phenanthrene
anthracene
High molecular weight PAH
fluoranthene
pyrene
benz(a)anthracene
chry'sene
benzofluoranthenes
benzo(a)pyrene
i ndeno(1,2,3-cd)pyrene
dibenzo(a,h)anthracene
benzo(g,h,i)perylene
Dibenzofuran
NH-06, SS-08
NH-06, NH-08
SS-08
EW-02, NH-06, NH-08, SS-08
EW-02, NH-06, NH-08, SS-08
EW-02, NH-06, NH-08, SS-08
NH-06, SS-06, SS-08
NH-06, NH-08, SS-08, SS-09
EW-14, NH-06, SS-08, WW-04
EW-14, NH-06, NH-08, SS-06,
SS-08, SS-09
NH-06, NH-08, SS-08, SS-09
NH-06, NH-08, SS-08, SS-09
NH-06, NH-08, SS-06, SS-08, SS-09
NH-06, SS-08, SS-09
NH-03, NH-06, SS-08
SS-08
NH-03, SS-08
SS-08
1,4-Di chlorobenzene
SS-09
106
-------
other LPAH were consistently the poorest, and had correlation coefficients
ranging from 0.29 to 0.44. When naphthalene was correlated with total LPAH
in individual study areas with more than five naphthalene detections,
regressions were stronger and r values ranged from 0.75 to 0.98. Special
attention will be given to naphthalene in data analysis to account for its
poorer covariation with other LPAH. Concentrations normalized to organic
carbon content did not generally yield improved correlations, and in many
cases, correlations were worse. Organic carbon normalized concentrations of
PAH were examined but will not be discussed because they did not indicate
spatial gradients that were not apparent based on dry weight concentrations.
Correlations among the 10 individual HPAH were uniformly strong, and all
had r values greater than 0.90 (P<0.05). Although the strength of these
correlations was enhanced by extreme PAH concentrations at Station SS-08,
correlations were strong even in the absence of that outstanding station.
Correlation coefficients without Station SS-08 ranged from 0.56 to 0.99 for
individual HPAH.
LPAH concentrations were generally elevated (e.g., EAR >30) over most of
the study area. The areas of most severe contamination were the Seattle
South waterfront and NH stations (west of the mouth of the West Waterway)
(Tables 17-20) (Figure 31). The highest LPAH concentration occurred at
Station SS-08 (630,000 ug/kg DW; EAR = 15,500). The mean LPAH concentration
of SS stations was 65,000 ug/kg DW (EAR = 1,600). A clear concentration
gradient was apparent at stations surrounding Station SS-08 (Figure 32); the
concentration gradient was not linear, as Station SS-08 was over 40 times
more contaminated than adjacent stations. Nonetheless, Stations. SS-03 to
SS-11 all had EAR values greater than 100.
Stations west of the mouth of the West Waterway (Area NH) also had very
high LPAH concentrations, although concentration gradients along shore were
not continuous. Highly contaminated Stations NH-06 (57,000 ug/kg DW;
EAR = 1,400) and NH-08 (37,000 ug/kg DW; EAR = 910) were separated by a far
lower concentration at intertidal Station NH-07 (230 ug/kg DW) (Figure 32).
The sediments at Station NH-07 were far more coarse-grained (over 90 percent
sand) than Stations NH-06 and NH-08 (roughly 53 percent silt and clay).
Concentrations from 5,000 to 10,000 ug/kg (EAR >100) occurred at Stations
NH-04, NH-05, and Station AB-01 located along Alki Beach. A similarly high
concentration was observed at Station NH-03 (8,200 ug/kg DW) on the east side
of the mouth of the West Waterway. Elevated LPAH concentrations were
observed in the West Waterway (EAR >100 at Stations WW-04, WW-09, and
WW-12), and sites of isolated, severe contamination (EAR >100) were observed
in the East Waterway (Stations EW-02, EW-04, EW-06, and EW-14), the Duwamish
River (Station DR-16), and along the Seattle North waterfront (Station
NS-07) (Figure 33). Only a few stations in the study had detected concentra-
tions within the range of reference sediments (Stations AB-03, MG-01, MG-04,
and DR-06), and all individual LPAH were undetected at Station MG-03 and
intertidal Station WW-15.
Naphthalene distributions differed somewhat from overall LPAH distribu-
tions. The maximum naphthalene concentration (15,000 ug/kg DW; EAR = 2,600)
was located at Station NH-06, west of the mouth of West Waterway; the
maximum LPAH concentration occurred at Station SS-08. However, the relative
107
-------
o
00
UJ
O
4000
3000
UJ
oc
LU
U.
UJ
OC ~
x:
III o>
> Q>
o 5
m 2000
< £
"O
z w
o
I-
<
>
LU
-J
UJ
1000
THRESHOLD 0»"
LEGEND
~i—
PS
r^-pn l>f\
\
0 LPAH
53 HpAH
nfl
P S Port Susan (n = 4)
M Q Magnolia Blutt (n - 4)
N S Seattle North Waterfront (n = 8)
S S Seattle South Waterfront (n = 11)
NH North Haifoor Island (n = 11)
EW East Waterway (n = 16)
WW West Waterway (n = 19)
K Q Kellogg Island (n = 11)
DR Duwamish River (n = 18)
AB Alki Beach (n = 4)
* THRESHOLD - equivalent to the highest
concentration in Puget Sound reference
areas.
—r
MQ NS SS NH EW WW
STATIONS
.. |~TI nr-jT-Tl \f |\l
KG DR AB
Carr Inlet Relerence = 41 ng/kg OW (LPAH) and 79 pg/kg DW (HPAH).
Figure 31. Mean elevations above reference (EAR) of LPAH and HPAH in sediments
from all study areas.
-------
o>
"55
3
>»
b.
2,
UJ
o
z
UJ
oc
LU
LL
UJ
QC
UJ
>
o
ffi
<
H
<
>
lii
-I
UJ
1000
900
800 -
700 -
600
500
400 -
300 -
200
100-
J
£
A
£
J
\C>
1
a I
iri o
i
/\
&
/N
/\
/N
/S
/.S
&
8
«
&
SEATTLE
SOUTH
WATERFRONT
I
43
SS01 SS-03 SS-04 SS-05 SS06 SS-07 S&08 SS-09 SS-10 SS-11 SS-12
STATIONS
1000
900 -
800 -
700 -
600 -
500
400 -
300 -
200
ioo H
0
/;ni
r>-pn
I
£
1
I
I
s
I
/\
/\
/S
j
&
NORTH
HARBOR
ISLAND
J
n-fsl
NH-10 NH-0B NH-08 NH-07 NH-06 NH-05 NH-04 NH-03 NK02 NHOI NH-11
'—WEST WATERWAY MOUTH
STATIONS
LP AH
HPAH
Carr Inlet Reference » 41 |ig/kg DW (LPAH) and 79 )ig/kg DW (HPAH).
Figure 32. Elevations above reference (EAR) of LPAH and HPAH at
individual stations in the most contaminated study areas.
109
-------
o>
®
5
>»
33
IXJ
o
z
UJ
oc
UJ
Li.
Uj
cc
UJ
>
o
00
<
<
>
UJ
-J
U1
1000-1
900 -
800 -
700 -
600 -
500 -
400 -
300 -
200 -
100
311V
41
zU.
£
£
b
E
WEST
WATERWAY
a
WW-011WW-03 WW-05 WW-08 WW-10 WW-12 WW-HlWW-18 WW-18 WW-2D
WW-02 WWW WW-06 WW-09 WW-11 WW-13 WW-15 WW-17 WW-19
STATIONS
1000 -i
900 -
800 -
700 -
600 -
500
400 -
300
200
100
7\
/
/
/
/
/
/
/
/
/
/
~ .
/s
/s
/\
/s
/\
EW-01
k
\
\
s
\
s
\
\
EW-03
£0
n
s
\
\
s
s
7^
/\
/\
/\
/\
EW-05
EW-07
EW-09
EW-11
EW-13
\
\
\
\
\
\
\
\
\
s
\
s
\
P\
~ \
/\
/s
/\
EAST
WATERWAY
N
\
EW-15
rr
EW-02 EW-0* EW-06 EW-08 EW-10 EW-12 EW-14 EW-16
STATIONS
0LPAH £3HPAH
Caw Inlet Reference • 41 |ig/kg OW (LPAH) and 79 |ig/kg OW (HPAH).
Figure 33. Elevations above reference (EAR) of LPAH and HPAH at
individual stations in highly contaminated study areas.
110
-------
distributions of naphthalene and LPAH were otherwise similar in the most
contaminated areas (i.e., Areas SS and NH).
High molecular weight PAH had similar distributions to LPAH, but HPAH
typically occurred at higher concentrations and were more elevated above Carr
Inlet reference concentrations than LPAH (Figure 31). The correlation
between LPAH and HPAH was very strong on a study-wide basis (r=0.997, n=105,
P<0.05; dry weight concentrations). Although correlations between LPAH and
HPAH were strong overall, scatterplots of LPAH vs. HPAH in each of the study
areas revealed several noteworthy exceptions within areas. An anomalously
high ratio of LPAH/HPAH was apparent at Station EW-02 (Figure 33). An
anomalously low LPAH/HPAH ratio was noted at Station NS-07.
As was the case for LPAH contamination, HPAH contamination was most
severe along the Seattle South waterfront and at NH stations (see Tables
17-20). EAR values along the Seattle South waterfront (Stations SS-03 to
SS-12) ranged from 78 to 41,000 (maximum at Station SS-08). The mean EAR in
this area was 4,100. Concentrations decreased sharply moving away in either
direction from Station SS-08, as was the case for LPAH (Figure 32). EAR
values greater than 1,000 were observed at Stations SS-08 and SS-09, and
greater than 500 at Stations SS-06, SS-07, and SS-10.
In Area NH, EAR values of near or greater than 1,000 for HPAH were
observed at Stations NH-06 (EAR = 1,600) and NH-08 (Figure 32). As was the
case for LPAH, these stations were separated by a coarse-grained intertidal
station with a much lower concentration (NH-07; EAR = 18). Nearby Station
AB-01 had an EAR of 440, but was flanked by less, contaminated stations.
Station NH-03, on the opposite side of the mouth of West Waterway, had an
EAR of roughly 50.
Other areas with mean HPAH concentrations that exceeded an EAR of
100 were the East and West Waterways and the Seattle North waterfront (see
Table 17).; Within these areas, individual stations with EAR greater than
500 were EW-14 (69,000 ug/kg DW), WW-04, WW-09, and NS-07. Although the EAR
plotted for West Waterway stations in Figure 33 appear to follow a gradient
that decreases in concentration toward the mouth, this is an artifact of the
linear presentation of the figure. Stations WW-01 to WW-20 were not
actually positioned along a straight line in the waterway and PAH concentra-
tion gradients were not apparent when concentrations were plotted on a map,
although concentrations were generally elevated. All individual stations
with EAR between 100 and 1,000 are listed in Table 18, and those with EAR
greater than 1,000 are listed in Table 20.
Polvchlorinated Biohenvls—PCBs were detected in roughly 75 percent of
the samples in the study. However, detection limits were generally high
(>100 ug/kg DW) and may have precluded detection of low-level contamination.
Roughly 24 percent of the stations in this study had EAR values of over 100
for PCBs (see Table 17) and the median EAR was 80 (see Table 16). PCB
distributions were generally patchy but were most elevated in the Duwamish
River and along the Seattle South waterfront. The mean EAR for PCBs in the
Duwamish River (Area DR) was 160 (using detected values only) (Figure 34).
The highest concentration at an individual station was observed in the
Duwamish River at Station DR-08 (5,800 ug/kg DW; EAR = 970) (Figure 35). On
an organic carbon normalized basis, concentrations tended to decrease moving
111
-------
ISO
200
Ui
o
z
w
DC
Ui
150
Ui
cc c*
•C
UI .J?
> 0>
O 5 too
CQ .
< ^
TJ
50 -
o
»-
<
>
UI
UJ
THRESHOLD1=1
r *
—7
~\
PS
V/
r —i— i
MG NS SS NH EW
STATIONS
WW KG
r
DR
LEGEND
—r~
AB
; Including Detection Limits
Detected Values Only
7
PS Pbrt Susan (n = 4)
M Q Magnolia Blull (n = 4)
MS Seattle North Waterfront (n = 8)
SS Seattle South Waterfront (n = 11)
NH North Harbor Island (n = 11)
E W East Waterway (n = 16)
WW West Waterway (n - 19}
KG Kellogg Island (n = 11)
OR Duwamish River (n = 18)
AB Alki Beach (n = 4)
* THRESHOLD - equivalent to the highest
concentration in Puget Sound reference
areas.
Carr Inlet Reference = 6 |i0/kg OW.
Figure 34. Mean elevations above reference (EAR) of PCBs in sediments from alt study areas
-------
JO
o>
*55
13
U1
o
z
Ui
DC
Ui
U.
Hi
E
Ui
>
o
m
<
O
H
<
>
Ui
_J
Ui
600
500 -
400 -
300 -
200
100
71
I DA02 I DRO* I DFW6 I I
970
?
I I (
I
1
\ I DR-16 I
DR-25 I 0A02 I DfV04 I 0R46 I DR-08 I 0R-10 I DR-12 I DR-14 I DR-16
DR-01 DR-03 Ofi-05 DFM7 DR49 0R-11 OR-13 OR-15 DR-17
STATIONS
"O
0)
N
"5
E
k.
o
c
so
40
30 -
20 -
O
o
H 10
86
¦E
vx
v\
2
0
1
2
F3
f
f
2
s
I I c
96
ft
2
f
f
f
DR-25 I DR02 I DR-04 I DR46 I DR-08 I DR-10 I DR-12 I DR-14 I DR-16
DR-01 DR-03 DR-05 DR-07 DR-09 DR-11 DR-13 0R-15 DR-17
STATIONS
Figure 35. Elevations above reference (EAR) of PCBs in the
Duwamish River (Area DR).
113
-------
upstream from Station DR-08 (i.e., from Stations DR-07 to DR-03) (Figure 35),
possibly suggesting upriver transport related to salt-wedge intrusion.
Along this gradient, EAR on a dry weight basis were near or above 100 at
Stations DR-03, DR-04, and DR-05. Another area of relatively heavy con-
tamination (i.e., EAR >100) in Area DR consisted of a group of contiguous
stations including DR-10 (2100 ug/kg DW), DR-11, DR-12, DR-13, DR-14, DR-16,
and DR-17 (Figure 35).
The Seattle South waterfront had a mean EAR of 160 (detected values
only), including several stations with EAR greater than 100 (Figure 36).
Station SS-09 had a concentration of 3,300 ug/kg 0W and Station SS-04 had a
concentration of 1,600 ug/kg DW. Although a gradient was not clear on a dry
weight basis, total organic carbon normalized concentrations tended to
decrease moving north from Station SS-03 to SS-07 (Figure 36). A maximum
occurred at Station SS-09.
EAR values greater than 100 were also found in the East Waterway, which
had a mean EAR of 115 (PCBs detected at all stations), a maximum concentra-
tion of 2,500 ug/kg DW (Station EW-05), and a concentration of 1,500 ug/kg
DW at nearby Station EW-06 (Figure 37). Other stations with EAR values
greater than 100 in the East Waterway were EW-02, EW-03, and EW-11 (all
roughly 1,000 ug/kg DW; Table 18). On an organic carbon normalized basis,
concentrations were more uniform than on a dry weight basis, and appeared to
decrease moving toward the mouth of the East Waterway (Figure 37). The
nearest upriver stations (e.g., KG-08) do not appear to continue this trend
upriver. The correlation between T0C and PCB concentrations in the East
Waterway was unusually strong relative to other study areas with elevated
PCB contamination (r=0.92, n-16, P<0.05).
The West Waterway tended to have lower concentrations than the East
Waterway, with a mean EAR of 88 (using only detected values) and a maximum
concentration of 1,500 ug/kg DW (Station WW-09). PCB concentrations at
stations on the east side of the West Waterway ranged from 270 to 1,500 ug/kg
DW, with no clear gradient in concentration on a dry weight or organic carbon
normalized basis (Figure 38). Station WW-05, located in another section of
the West Waterway, had a PCB concentration of 1,200 ug/kg DW. In another
area of the Duwamish River, an isolated high concentration was found near
Kellogg Island (Station KG-06; 3,100 ug/kg DW).
The mean EAR for NH stations at which PCBs were detected was 120. Only
three individual stations had EAR of 100 or greater: Stations NH-03
(3,300 ug/kg DW; EAR = 550), NH-08 (1,300 ug/kg DW; EAR = 220), and NH-06
(600 ug/kg DW; EAR = 100) (Figure 38).
Lower PCB concentrations were reported along the Seattle North water-
front and along Alki Beach (Figure 34). PCBs were undetected at all four
Magnolia Bluff stations at moderate to high detection limits (100-200 ug/kg
DW) .
Pheno1s--Analvses of phenolic compound distributions were impeded
because of high detection limits (see Table 13). Although high detection
limits resulting from poor surrogate recoveries were obtained for all target
chlorinated phenols and 2,4-dimethylphenol, the most problematic detection
114
-------
600
500 -
LU
O
z
HI
DC
UJ
LL
UJ
GE
UJ
>
o
CD
<
z
o
H
<
>
111
-I
LU
jZ 400
O)
0)
£ 300 H
>»
¦£; 200 H
100 -
'A
ill
i
VAvn
S&01 SS-03 SS-04 SS-05 SS06 SS-07 SS-08 SS-09 SS-10 SS-11 SS-12
STATIONS
so
40 -
TD
0
N
a 30
o
O
H 10 A
i
1
i
_E2_E3
! 1 I
SS01 SS-03 SS-04 SS-05 SS-06 SS07 SS-08 S&09 SS-10 SS-11 SS-12
STATIONS
Figure 36. Elevations above reference (EAR) of PCBs along the
Seattle South Waterfront.
115
-------
UJ
o
z
Ui
DC
Ui
LL
Ui
AC
Ui
>
o
CD
¦<'
z
o
H
<
>
Ui
-i
Ui
EW-01 I EW-03 I EW-05 I EW-07 I EW-09 I EW-11 I EW-13 I EW-15
EW-02 EW-04 EW-06 EW-08 EW-10 EW-12 EW-14 EW-16
STATIONS
50
40 -
"O
0)
« 30 -
ha
O
C
20 -
o
o
H 10 H
T,
'4
V}
r
¦4
t
t
i
EW-01 I EW-03 I EW-05 I EW-07 I EW-09 I EW-11 I EW-13 I EW-15 I
EW-02 EW-04 EW-06 EW-08 EW-10 EW-12 EW-14 EW-16
STATIONS
Figure 37. Elevations above reference (EAR) of PCBs in the
East Waterway.
116
-------
600
500
400
300 "
o>
©
3
>»
2,
Ui
o
z
Ui
cc
UI
u.
UI
cc
Ui
>
o
m
<
z
o
H
<
>
Ui
Ui
200 -
100 -
YZ7\ W/
&
il
P7^\
1
NORTH
HARBOR
ISLAND
77?[ 77* V7;
NH-10 NH-09 NH-08 NH-07 NH-06 NH«5 NH-04 NH-03 NH-02 NH01 NH-11
STATIONS
600 -i
500 -
400
300 -
200 -
100-
2
f-rJV
lWW-03 Iv
i
WEST
WATERWAY
* Not reported
0
2?
WW-011WW-03 IWW-051WW-081WW-10IWW-121WW-141WW-161WW-181WW-20
WW-02 WW-04 WW-06 WW-00 WW-11 WW-13 WW-15 WW-17 WW-19
STATIONS
Carr Inlet Reference = 6 ng/kg DW.
Figure 38. Elevations above reference (EAR) of PCBs in
North Harbor Island and the West Waterway.
117
-------
limits were for pentachlorophenol (over half the samples had detection
limits over 1,000 ug/kg for pentachlorophenol; see Table 13).
Pentachlorophenol (PCP) was detected eight times in the study with a
maximum concentration near the mouth of the West Waterway (Station NH-04;
6,000 ug/kg DW; EAR ¦ 180) (see Table 12). A nearby station in the West
Waterway (Station WW-17) had a concentration of 360 ug/kg DW. High detection
limits in this area do not allow for further analysis. Another area of PCP
contamination was at the mouth of the East Waterway. Station EW-16 had a
concentration of 690 .ug/kg DW and EW-15 had a concentration of 95 ug/kg DW.
Station NH-01, east of the mouth, had a concentration of 100 ug/kg DW. PCP
was undetected at other stations at the mouth of the East Waterway (Stations
SS-01 and NH-11) at reasonable detection limits (56 and 75 ug/kg DW). Three
relatively isolated detections of PCP occurred at Station AB-04 (110 ug/kg
DW), coarse-grained intertidal Station NS-04 (330 ug/kg DW), and Station
SS-05 (47 ug/kg DW). PCP was undetected at relatively high detection limits
(e.g., 160 to 4,600 ug/kg DW) at adjacent stations in these areas (i.e.,
Alki Beach and the Seattle waterfront).
Phenol was detected at 36 stations but had a relatively narrow concen-
tration range (see Table 12). The maximum concentration (1,200 ug/kg DW;
EAR = 36) occurred at intertidal Station DR-25 in the Duwamish River.
Phenol was undetected at adjacent stations at low detection limits (e.g.,
<50 ug/kg DW). Other isolated stations of moderate phenol concentration
(roughly 400 ug/kg DW; EAR = 12) were Station SS-09, intertidal Station
KG-10,'and Station PS-05 (Port Susan), the latter of which had been excluded
from reference area comparisons because of elevated chemistry. Other phenol
concentrations that exceeded reference conditions were relatively isolated.
Concentrations near 100 ug/kg DW were detected in or near the West Waterway
(Station NH-04 and intertidal Station WW-15) and along the Seattle North
waterfront (Station NS-07).
4-Methylphenol was detected 32 times in the study, although the con-
centration range was relatively narrow. The maximum concentration occurred
at Station WW-19, on the east side of the West Waterway (2,600 ug/kg DW,
EAR = 200). Moderate concentrations occurred at intertidal Station WW-15 on
the west side of the waterway (540 ug/kg DW) and at NH stations near the
mouth (Stations NH-04 and NH-06, 1,000 and 170 ug/kg DW, respectively).
Concentrations of greater than 1,000 ug/kg DW were observed at intertidal
Station NS-04, intertidal Station DR-25, and at Station KG-09 near Kellogg
Island. Adjacent Station KG-11 had a concentration of 610 ug/kg DW. All
other detected concentrations were equivalent to EAR less than 30.
Chlorinated Benzenes--!.4-Dichlorobenzene. detected seven times, was
the only chlorinated benzene detected more than once and the only detected
dichlorobenzene. High detection limits for chlorinated benzenes (see
Table 13) precluded comprehensive data analysis for these compounds. More
than half of the samples had detection limits greater than 100 ug/kg DW for
each of the chlorinated benzenes.
1,4-Dichlorobenzene had a wide concentration range in this study, but
only one extremely high concentration was observed (Station SS-09, 31,000
ug/kg DW, EAR - 8,900). The next highest concentration, 380 ug/kg at
Station SS-03, also occurred along the Seattle South waterfront but was not
118
-------
located near Station SS-09. 1,4-Dichlorobenzene was undetected at other SS
stations over a wide range of detection limits (2 to 7,300 ug/kg DW).
Isolated areas of contamination were observed in the Duwamish River at
adjacent Stations WW-04 and WW-05 (90 and 59 ug/kg DW, respectively) and much
further upriver at Station DR-15 (100 ug/kg DW). Detection limits at
adjacent stations in these areas were relatively low (i.e., below the
detected concentrations). All other detected concentrations were less than
40 ug/kg DW.
Phthalate Esters—Data for three phthalate esters were rejected because
of high blank contamination (diethyl phthalate, di-n-butyl phthalate, and
bis(2-ethylhexyl)phthalate; see METHODS, Sediment Chemistry, Quality
Assurance/Quality Control Results). The remaining phthalates were dimethyl
phthalate, butyl benzyl phthalate, and di-n-octyl phthalate. These compounds
did not correlate well enough in the overall study to justify their treatment
as one group during data analysis (using detected values only, dimethyl vs.
butyl benzyl, r=0.06, n=26, P>0.05; di-n-octyl vs. butyl benzyl, r=0.51,
n=27, P<0.05; dimethyl vs. di-n-octyl, r=0.20, n=18, P>0.05).
Butyl benzyl phthalate was the most commonly detected phthalate ester
(detection frequency = 60/107; Table 12). Areas of relatively high concen-
tration (EAR >50) were observed in the East Waterway and along the Seattle
South waterfront. Three adjacent stations in the East Waterway (EW-07,
EW-08, " i- EW-09) had the maximum concentration of 1,800 ug/kg DW
(EAR = lli>/. Lower concentrations (760 to 1,300 ug/kg DW) were observed in
this area at Stations EW-05 and EW-11. Notably, butyl benzyl phthalate and
di-n-octyl phthalate correlated well in the East Waterway (r=0.98, n=7,
P<0.05). Elevated concentrations also occurred along the Seattle South
waterfront at Stations SS-04 and SS-05 (500 to 1,000 ug/kg DW). Adjacent
stations had concentrations below 50 ug/kg DW. Except for an isolated high
concentration at Station KG-06 (690 ug/kg DW; EAR = 41), other concentrations
were relatively low (EAR <20).
Distributions of dimethyl phthalate were patchy. The highest observed
concentration was found along the Seattle South waterfront (Station SS-11,
1,400 ug/kg DW; EAR = 35). Dimethyl phthalate was undetected at adjacent
stations at high detection limits (600 to 2,200 ug/kg DW), but was detected
at Station SS-09 (160 ug/kg DW) and at SS-03 (300 ug/kg DW) with low
detection limits adjacent to Station SS-03. Similarly elevated concen-
trations were found in the West Waterway at Stations WW-09 (440 ug/kg DW)
and WW-13 (1,000 ug/kg DW); these stations were located on opposite sides of
the waterway. All other detected values in the study were less than
200 ug/kg DW (EAR = 5).
Di-n-octyl phthalate was detected 40 times in this study, but an EAR of
25 (500 ug/kg DW) was exceeded only 4 times. The highest concentration in
the study was observed at Station NS-06 (9,900 ug/kg DW, blank-corrected;
EAR = 490). Observations at other stations in the Seattle North waterfront
area were typically detected concentrations or detection limits of less than
50 ug/kg DW. Intertidal Station DR-25 had the next highest concentration in
the study (1,300 ug/kg DW; EAR = 65). Concentrations of di-n-octyl phthalate
at adjacent stations were blank corrected down to low detection limits.
Station SS-05, along the Seattle South waterfront, had a concentration of
1,100 ug/kg DW, although di-n-octyl phthalate was undetected at 5 ug/kg DW
119
-------
at adjacent stations. Station KG-05 had a moderate concentration of
400 ug/kg DW. All other detections in the study were equivalent to an EAR of
15 or less'.
Pesticides—Only three of the target pesticides were detected more than
two times in this study: p(p1-DDT (six detections), p,p'-DDE (four detec-
tions), and p.p'-DDD (nine detections) (see Table 12). Elevated concentra-
tions tended to be geographically isolated, and the pesticides did not tend
to occur at the same stations. Only Station DR-10 had all three pesticides,
whereas Stations DR-08 and SS-09 had p,p'-DDE and p,p'-DDD, Station EW-05
had p,p'-DDE and p#p'-DDT, and Station SS-04 had p.p'-DDD and p.p'-DDT.
Relatively high and isolated p,p1-DDT concentrations were observed in
the East Waterway (Station EW-05, 84 ug/kg DW), along the Seattle South
waterfront (Station SS-04, 180 ug/kg DW), and in the Duwamish River (Station
KG-06, 270 ug/kg DW). DDT was undetected at adjacent stations at detection
limits lower than 20 ug/kg DW. The highest p.p-DDE concentration was found
at Station DR-10 (62 ug/kg DW). Relatively high p.p'-DDD concentrations
were observed at Stations NH-03 (120 ug/kg DW) and SS-09 (140 ug/kg DW).
It is noteworthy that all the stations with high pesticide concentra-
tions also, had high PCB concentrations (>1,000 ug/kg PCBs at all stations
discussed in this section). It is possible that PCBs could have acted as
interferences during 6C/ECD analysis and may have artificially increased
pesticide concentrations. Confirmation of pesticide concentrations by GC/MS
is recommended if future analyses of pesticides are conducted in these areas.
Miscellaneous Organic Compounds—Compounds assigned to this class
include 2-methylnaphthalene, dibenzofuran, benzyl alcohol, benzoic acid, and
the following TIOs detected in this study: alkylated phenanthrenes (1-, 2-,
and 3-methylphenanthrene), biphenyl, carbazole, and retene. Detection
frequencies^ for these compounds are presented in Table 12. High detection
limits for benzoic acid and benzyl alcohol (see Table 13) impeded data
analysis for these compounds.
Biphenyl and the alkylated PAH and heterocycles detected in this study
(2-methylnaphthalene, methylphenanthrenes, dibenzofuran, and carbazole)
tended to co-vary with PAH; therefore, their distributions will not be
described in detail. Such covariation was expected, as these compounds are
known to co-occur in fossil fuel products. Retene and benzyl alcohol have a
wider range of potential sources and will be discussed separately.
y i
Correlations of miscellaneous organic compounds vs. PAH are presented in
Table 21 (correlations excluding anomalous Station SS-08 are also presented
in Table 21). 2-Methylnaphthalene correlated well with the most structurally
related PAH, naphthalene (Table 21). Similarly, 1-, 2-, and 3-methylphenan-
threne correlated well with phenanthrene (Table 21). Although Station SS-08
was an outlier in these three phenanthrene correlations and resulted in
deceptively high correlation coefficients, correlations run without SS-08
were still relatively strong (Table 21). Correlations of dibenzofuran,
carbazole, land biphenyl vs. LPAH and HPAH were examined because these
organic compounds co-occur with PAH (especially LPAH) in fossil fuel
products such as petroleum and creosote (Clark and Brown 1977; Nestler
1974). Carbazole, an important creosote component (e.g., Nestler 1974; Krone
120
-------
TABLE 21. CORRELATIONS BETWEEN PAH AND
MISCELLANEOUS ORGANIC COMPOUNDS8
Compounds
All Detected
Stations
r n
Without
SS-08®
naphthalene:
vs. 2-methylnaphthalene
phenanthrene:
vs. 1-methylphenanthrene
vs. 2-methylphenanthrene
vs. 3-methylphenanthrene
LPAH:
vs. dibenzofuran
vs. biphenyl
vs. carbazole
vs. retene
HPAH:
vs. dibenzofuran
vs. biphenyl
vs. carbazole
vs. retene
0.87
0.998
0.998
0.998
0.84
0.94
0.83
0.08
0.80
0.87
0.90
0.06
61
69
70
68
79
27
45
55
80
27
45
56
0.92
0.75
0.68
0.70
60
68
69
67
0.91 78
(same)c
(same)c
(same)c
0.65 79
(same)c
(sa«ne)c
(same)c
a Based on dry weight concentrations; detected values only.
b Station SS-08 excluded from correlation.
c Same as "all detected stations" - not detected at Station SS-08.
121
-------
et al. 1986) and biphenyl correlated well with LPAH and HPAH (Table 21).
Neither compound was detected at Station SS-08. Both compounds maximized at
Station NH-06, as did naphthalene. Dibenzofuran also correlated well with
LPAH and HPAH, even when the maximum concentration at Station SS-08 was
excluded.
Retene is a useful geochemical marker for sub-bituminous coal and
lignite found in the Green River area (Barrick et al. 1984), but is not.
prevalent in higher grade coals (e.g., anthracite). Unlike other compounds
described in this section, retene correlated very poorly with PAH on a
study-wide basis, even when outlier stations were eliminated (Table 21;
P>0.05). Retene was not detected at the station with the highest PAH
concentrations (SS-08); also, the highest retene concentration (10,000 ug/kg
DW) was reported at Station KG-10, which had low PAH concentrations (total
PAH = 1,900 ug/kg DW). Retene was undetected or detected at low concentra-
tions (<25 ug/kg DW) at stations near KG-10. Other than Station KG-10,
retene was detected at concentrations greater than 1,000 ug/kg DW at
Stations KG-06, EW-05, and SS-09. Notably, detected retene concentrations
correlated well with LPAH and HPAH in Area SS (r=0.93 and 0.91, respectively;
n=5, P<0.05). However, as noted above, retene was not detected at the
station in this area with the highest PAH concentration.
Benzyl alcohol was detected only five times, but was reported at
relatively high concentrations in most cases. The highest concentration was
found in the southern portion of the West Waterway at intertidal Station
WW-02 (8,800 ug/kg DW). Another station in the West Waterway, WW-08, had a
far lower concentration (140 ug/kg DW). An isolated concentration of
870 ug/kg DW occurred at Station EW-12 in the East Waterway. Along the
Seattle South waterfront, Station SS-03 had a concentration of 1,300 ug/kg
DW.
Spatial Correlations Among Chemicals
Pearson linear correlations were performed on a study-wide basis and for
most individual study areas to examine the covariance of chemical distribu-
tions. Only detected data were used in this analysis. The range of
correlations examined in detail was reduced by application of the following
criteria:
¦ Chemicals had to be detected at least four times in at least
one study area
¦ Correlation coefficients (r) had to be at least 0.7 (i.e., r2
was at least 0.5)
¦ Based on examination of scatterplots, correlations were not
apparently driven by stations with anomalously high concentra-
tions.
Although these criteria are subjective, they provided a reasonable means for
producing a reliable, if not comprehensive, summary of spatial covariance.
The variables included in the correlation analysis that satisfied the first
criterion specified above were all metals of concern, HPAH and LPAH (as
group sums), PCBs, p.p'-DDD, 4-methylphenol, phenol, dimethyl phthalate,
122
-------
butyl benzyl phthalate, di-n-octyl phthalate, and retene. TOC content and
percent fine-grained material were also included in the analyses to examine
trends between chemistry and sediment texture. Study-wide correlations
(particularly among PAH and related alkylated PAH and heterocycles) that
were discussed in the previous section will not be recounted here. Three
individual study areas (Areas AB, MG, and NS) were not analyzed on an area-
specific basis because contamination was not highly elevated or widespread
in these areas.
On a study-wide basis, most correlations were poor or driven by
outliers. An exception to this generalization was the correlation between
chromium and nickel on a study-wide basis (r=0.83, n»107, PO.05). Because
concentrations of these two metals seldom exceeded reference levels, it is
possible that the relatively consistent chromium/nickel ratio is primarily
related to mineralogical composition rather than anthropogenic contamination.
Many strong correlations were observed when regression analyses were
performed on an area basis rather than a study-wide basis. Apparently, the
localized, heterogeneous contamination observed in this study required small-
scale analyses of spatial covariance to resolve trends that were confounded
during the study-wide analysis. Strong correlations may be useful during
source evaluation for defining the chemical nature of sources. A summary of
significant correlations that were relatively strong (i.e., r>0«7 without
apparent outliers) in at least two study areas is presented fn Table 22.
Strong correlations among metals were clearly far more prevalent than strong
correlations among organic compounds. Correlations among copper, lead, and
zinc were relatively strong in Areas DR, NH, and EW (Table 22). As expected,
correlations between LPAH and HPAH were strong in most study areas. Strong
correlations observed between retene and mercury may be related to particu-
late material derived from the Green River. The Green River passes through
both mercury and coal mines, which are potential sources of these chemicals.
Strong correlations between organic chemicals and TOC content or between
metals and percent fine-grained material, which support normalization to
these variables, were not typically observed even on a small scale basis.
However, PAH correlated well with TOC in Areas NH and SS (Table 22), and a
number of metals correlated well with percent fine-grained material in Area
NH, and to a lesser extent, in Area KG.
Certain relatively strong correlations were observed for only single
areas and are not listed in Table 22. These are described briefly below.
Along the Seattle South waterfront, relatively strong correlations were
observed between copper and arsenic, PAH (both LPAH and HPAH) and retene,
and PCBs and mercury. In the North Harbor Island area, very strong correla-
tions were observed between PAH (both LPAH and HPAH) and cadmium. Among the
many strong correlations observed in the East Waterway were PCBs with
several metals (lead, cadmium, and silver), and silver with cadmium,
mercury, and copper. In the West Waterway, zinc correlated well with nickel
and chromium. Lead and mercury correlated well in Area KG, and butyl benzyl
phthalate and cadmium correlated well in Area DR.
Comparison with Recent Historical Data
Data from previous studies of Elliott Bay and the Duwamish River were
compiled to confirm contaminant distributions found in this study and to
123
-------
TABLE 22. STRONG CHEMICAL CORRELATIONS3
Area^
Chemi cal s
DR
EW
KG
NH
SS
WW
Copper - Lead
0.84c
(18)d
0.76
(16)
0.91
(11)
Copper - Zinc
0.98
(18)
0.79
(16)
0.90
(11)
0.72
(19)
Lead - Zinc
0.84
(18)
0.77
(16)
0.96
(11)
Chromium - Nickel
0.85
(16)
0.79
(11)
0.98
(11)
0.93
(19)
Cadmium - Zinc
0.77
(16)
0.89
(11)
0.92
(11)
Cadmium - Lead
0.92
(16)
0.86e
(10)
0.91
(19)
Silver - Lead
0.78
(18)
0.89
(16)
0.87
(11)
Silver - Zinc
0.91
(11)
0.91
(19)
HPAH - LPAH
0.98
(18)
0.86
(11)
0.97
(11)
0.85e
(10)
0.88
(18)
Retene - Mercury
0.95
(10)
0.93
(5)
0.88
(7)
LPAH - TOC
0.74
(11)
0.83e
(10)
HPAH - TOC
0.81
(11)
0.89e
(10)
a Correlations included in this table meet the following criteria:
1. r value >0.7 (based on n >4)
2. scatterpTot confirming strong correlation (i.e., not driven by a
station with anomalously high concentrations)
3. apparently strong correlations in at least two study areas.
124
-------
TABLE 22. (Conti nued)
b The most contaminated areas were the focus of spatial characterization,
so Areas AB, MG, and NS are not included.
c Correlation coefficient (r).
d Number of observations (n); only detected values were included.
e Correlation excludes one anomalous station; inclusion of the station would
result in a misleading higher correlation coefficient.
125
-------
provide a more comprehensive assessment of contamination (especially in areas
not sampled in this study, such as central Elliott Bay and the area near
Denny Way CSO). Data from the following historical studies were compiled:
¦ Gamponia et al. (1986), a Metro report of sampling performed
in the West Waterway and North Harbor Island area
¦ Metro (1987), a report on the baseline investigations
performed near Duwamish Head in 1985-1986
¦ Romberg et al. (1984), a report on Metro's extensive environ-
mental sampling as part of their Toxicant Pretreatment
Planning Study (TPPS)
¦ Stober and Chew (1984), a report on the baseline investiga-
tions performed for Metro near Duwamish Head
¦ U.S. EPA (1982, 1983), two unpublished surveys performed by
EPA in the Duwamish River in 1982 and 1983
¦ Mai ins et al. (1980, 1982), two reports presenting the
results of sampling performed in Elliott Bay in support of
pathology studies of resident organisms.
Sampling stations from these studies are plotted in Figures 39-42. In
general, historical data tended to confirm the distributions reported irv the
present study. In the sections below, notable historical findings are
discussed for chemicals or chemical groups of concern that were most
elevated in the present study (phthalate ester concentrations were not
compared because historical data were not corrected for potential laboratory
contamination). The maps presented in this section focus on nearshore
chemical distributions; not all available historical data from deeper
Elliott Bay stations have been plotted in these maps. In a limited number
of cases, concentrations for a station sampled more than once differed
considerably. In such cases, the highest concentration was plotted. It was
assumed that analytical variability (especially among different studies) and
small-scale spatial heterogeneity were more likely responsible for such
discrepancies than temporal trends.
Several historical stations had very elevated concentrations of a number
of chemicals, in some cases exceeding the maximum concentrations observed in
the present study. Several of these stations occurred in the North Harbor
Island area: Station E42 (U.S. EPA 1982, 1983), and EPA Station E4 and
Station 3 of Gamponia et al. (1986), both located east of Station NH-03.
Other notable historical stations were Gamponia Station 9 in the West
Waterway, and TPPS Station S0090, located near Station SS-05 along the
Seattle South waterfront.
Copper-
Copper data for the present study and historical studies are presented
in a contour map (Figure 43). Four historical stations near the mouth of
the West Waterway had concentrations greater than 1,000 mg/kg DW. One
126
-------
LEGEND
• TEmATSCHfTHiSSTUDY)
* ' M6T*0<195n
~ GAMPQNIAETAU19M)
O R0M8£AGETAU(iM«)
v STOBB4ANOCH&tf(iOM|
~ us. 6M.ritas.itaa)
• MMJMSSTALiltW.tta®
latuft
QSEOSJ
on**
U1J4*
*VJlW ttMO
cv-ai
U»30V *U1»S
~ L-13
A0542
VUnO
*OW-ttl
A0W-C1
FC-Ufc
wwot
GP4I4 ' * tW1
GG4t i
ui»*
ET-01*
luiiv
AFM1
IJIU
FV«H
JMJ
»Ul2f
GWU
*HSW
OMS1 GV4I
»8B»i
AHLKff
A *
kui *
W® 3V0T
CT4I
utoav
KMH 4
-------
U124
110016
NH-01*
E34C-
EW-12
EW-16 I - ¦ v - ¦
• E1A EW-15
~ #eib ~ *
E1CS0064 »V-EW-14
I
EW-13
E34B
% <
EW-10
EW-11*
E2
~
EW-07
-E34A
EW-09
S0039
EW-OfljO
• 10039
C062
O
B062
EW-05* O
A062O
•EW-04
EW-03
LEGEND
EW-06
• TETRA TECH (THIS STUDY)
A METRO (1987)
~ GAMPONIA ET AL (1986)
O ROMBERG ET AL (1984)
~ STOBER AND CHEW (1984)
~ U.S. EPA (1982,1983)
¦ MAUNS ET AL (1980.1982)
-EW-02
EW-01
meters
Figure 40. Locations of sampling stations from historical studies
of sediment chemistry in the East Waterway.
128
-------
110016
•NH-11
r
ro
U3
NH-01
S00630
AB-01
~ NH-0
E5 WW-20
NH-03
¦ 10045
OS0034
NH-09
U120
~
~07
NH-069 E43 •
NH-04
WW-19
NH-08#
E44
WW-18#
WW 17
WW15» WW 16
,, • •-t-WW-14
g 10^ 9^
E6C E6B E6
meters
300
TETRA TECH (TH® STUDY)
METRO (1987)
GAM PON IA ET AL (1986)
ROMBERG ETAL. (1984)
STOBER AND CHEW (1984)
U.S. EPA (1982.1983)
MALINS ET AL. (1980.1982)
Figure 41. Locations of sampling stations from historical studies of sediment chemistry in the
North Harbor Island area.
-------
110016
NH-05
u
NH-02
NH-03
Is WW-20
NH-04
WW-19
WW-18#
WW-17
WW-15< ~
WW-16
E6C E6B E6,
Jii i S
10 9
WW-13
LEGEND
• TETRA TECH (THIS STUDY)
A METRO (1987)
~ GAMPONIA ETAL (1986)
O ROMBERG ET AL (1984)
» STOBER AND CHEW (1984)
~ U.S. EPA (1982,1983)
¦ MAIINS ET AL. (1980,1982)
14
WW-12
WW-10
10028
WW-09*
WW-11
WW-14
1
E7
~ E7A
13 «0V12
E7B£P •ww-oa
D# S0036
E7C
^WW-06
meters
300
WW-03
WW-02
Figure 42. Locations of sampling stations from historical studies
of sediment chemistry in the West Waterway.
130
-------
COPPER
Designation
Number
1
2
3
4
5
mg/kg
(dry weight)
>1000
>500-1000
>250-500
>100-250
undetected-100
n
6000
Si <«•(
5H9 meters
Figure 43. Contours of copper concentrations in study area
sediments.
131
-------
station was at the western side of the mouth (EPA Station E42, 1,050 mg/kg)
and three other stations were on the east side of the mouth near Station
NH-03 (1,220 to 2,800 mg/kg DW) (Gamponia et al. 1986; U.S. EPA 1982, 1983).
Station 9 of Gamponia et al. (1986), located on the east side of the West
Waterway, had a copper concentration of 860 mg/kg DW. Concentrations of a
number of other metals (e.g., lead and zinc) were also highly elevated at
this station and at Station E42.
Lead-
Historical lead concentrations (summarized in Figure 44) were generally
consistent with those reported in this study. Among the noteworthy histori-
cal findings is Gamponia Station 9 on the east side of the West Waterway,
which had a lead concentration of nearly 11,000 mg/kg DW (Station WW-14 in
that area had a comparable concentration). Historical EPA Station E42 near
the mouth of the West Waterway had a concentration of greater than 2,000
mg/kg DW (U.S. EPA 1982, 1983). Two historical stations east of Station
NH-03 had lead concentrations between 500 and 1,000 mg/kg DW (Gamponia et al.
1986; U.S. EPA 1982,1983).
Several: high historical concentrations were reported along the Seattle
waterfront. TPPS Station 1603 near Denny Way CS0 had a lead concentration
of 670 mg/kg DW. Along the Seattle South waterfront, a concentration of
1,700 mg/kg DW was found at TPPS Station S0090. This concentration was over
5 times higher than the concentration at nearby Station SS-05.
Mercury—
Mercury distributions are summarized in Figure 45. Historical data
tended to confirm the findings of this study. A high mercury concentration
(12 mg/kg DW) was reported in the slip immediately to the east of Station
NH-03 (Gamponia Station 3). Four other historical stations in the North
Harbor Island area had mercury concentrations of greater than 1 mg/kg DW
(Malins Station ml0016 and Gamponia Stations 2, 4, and 6). A number of
historical stations in the Denny Way CSO area (not sampled subtidally during
the present study) had mercury concentrations of greater than 1 mg/kg DW,
with a maximum of 3.6 mg/kg DW (TPPS Station 1612) during early studies
(Romberg et al. 1984, Malins et al. 1980, 1982) and a maximum of 2.2 mg/kg
DW during recent studies (Romberg et al. 1987). Few historical stations
were taken along the Seattle South waterfront, but the two historical
stations located between Stations SS-05 and SS-08 (present study) confirmed
the high concentrations reported in this study (0.92 and 3.3 mg/kg DW; TPPS
Stations S0065 and S0090). Data from U.S. EPA (1982, 1983) were reported as
wet weight without additional information on total solids content; hence,
these data were unusable.
Zinc —
A number of relatively high zinc concentrations were observed in
historical studies (Figure 46). EPA Station E42 near the west side of the
mouth of the West Waterway had a zinc concentration of 4,810 mg/kg DW (U.S.
EPA 1982, 1983). Station 9 of Gamponia et al. (1986), located on the east
side of the West Waterway, had a zinc concentration of greater than
1,500 mg/kg DW. Concentrations greater than 1,700 mg/kg DW east of Station
132
-------
LEAD
Designation
Number
mgfcg
[dry weight)
>1000
>500-1000
>250-500
>100-250
undetected-100
F.gure 44. Contours of lead concentrations in stuOy area
sediments
-------
MERCURY
Designation mg/kg
Number (dry weigM)
Ot.
>2.0
>1.0-2.0
>0.50-1.0
>0.25-0.50
undetected-0.25
3 5
»3
50
-2
jU
5 4
Figure 45. Contours of mercury concentrations in study area
sediments.
134
-------
ZINC
40
40
30 40
40
5«
40
Designation
Number
1
2
3
4
5
mg/kg
(dry weight)
>1000
>500-1000
>250-500
>100-250
undetected-100
Figure 46. Contours of zinc concentrations in study area sediments
-------
NH-03 confirmed the high zinc concentrations at that station (Gamponia et al.
1986; U.S. EPA 1982, 1983). Other relatively high concentrations (>1,000
mg/kg DW) were observed near Kellogg Island (Romberg et al. 1984) and along
the Seattle South waterfront, offshore of Stations SS-10 and SS-12 (TPPS
Stations A061, B061, and C061) and near Station SS-05 (TPPS Station S0090,
4,700 mg/kg DW). The latter zinc concentration was anomalously high
relative to nearby stations (by roughly an order of magnitude) but was
within the range of other SS stations.
Other Metals--
A number of relatively high silver concentrations (up to 6 mg/kg DW)
were reported in the area near the Denny Way CS0 (Romberg et al. 1984). The
only station taken in that area during the present study, intertidal Station
NS-01, had the highest silver concentration observed in this study (over
8 mg/kg DW).
Cadmium concentrations reported by Mai ins et al. (1980, 1982) were
considerably higher (e.g., in some cases, by an order of magnitude or
greater) than concentrations at nearby stations from other studies. These
apparently anomalous results did not significantly affect problem area
identification or ranking. The highest cadmium concentration observed in
all studies compiled was found at TPPS Station S0090 (27 mg/kg DW), located
near Station SS-05 (2.39 mg/kg DW). Although this historical concentration
was considerably higher than the concentration observed at Station SS-05,
other SS stations had highly elevated cadmium concentrations (e»g., Station
SS-09, 17 mg/kg DW).
Arsenic concentrations reported in historical studies were typically
consistent with those of the present study; however, notably high values were
reported in the North Harbor Island area. EPA Station E42, on the west side
of the mouth of the West Waterway, had the highest arsenic concentration of
any study compiled (1,420 mg/kg DW). Concentrations between 250 and
600 mg/kg DW were reported at historical stations east of Station NH-03.
LPAH—
Concentrations of LPAH from this study and historical studies are sum-
marized in Figure 47. Detection limits were not available for PAH in all
historical studies and therefore could not be included in LPAH and HPAH sums
for those studies.
Historical LPAH concentrations were generally consistent with those of
the present study. However, historical EPA Station E12, near Kellogg
Island, had an extremely high LPAH concentration (42,000 ug/kg DW), whereas
nearby Station KG-01 (this study) had a considerably lower concentration
(2,100 ug/kg DW). A number of historical stations in the Denny Way area
(not sampled subtidally in the present study) had LPAH concentrations
ranging from 4,200 to 21,000 ug/kg DW in early studies (Romberg et al. 1984)
and from 240 to 185,000 ug/kg DW in more recent studies (Romberg et al.
1987). a
i
High LPAH concentrations were observed in Area NH in the present study
and in historical studies. A number of stations east of the mouth of the
136
-------
LPAH
3»
5«
*5
S«
40
5»
40
IS
MOO
Figure 47. Contours of LPAH concentrations in study area sediments.
137
-------
West Waterway had concentrations exceeding 15,000 ug/kg DW. the most extreme
example is a station located near NH-03 with reported concentrations of
150,000 and 25,000 ug/kg DW (U.S. EPA 1982, 1983; Gamponia et al. 1986). On
the other side of the West Waterway, high LPAH concentrations observed in
the present study (Stations NH-06 and NH-08) were confirmed by historical
data (U.S.; EPA 1982, 1983).
HPAH—
Although HPAH concentrations were higher than LPAH concentrations, the
trends observed for historical HPAH data were similar to those for LPAH data
(Figure 48). In the North Harbor Island area, historical HPAH concentrations
were reported over 450,000 ug/kg DW near Station NH-03, and over 150,000
ug/kg DW near NH-06. Concentrations in the Denny Way CSO area were reported
as high as 130,000 ug/kg DW (Romberg et al. 1984) and 159,000 ug/kg DW
(Romberg et al. 1987).
PCBs—
Interpretation of historical PCB data was impeded to some extent by
high detection limits, especially for data reported in U.S. EPA (1982,
1983). The highest concentration in the present study (Station DR-08;
5,800 ug/kg DW) was confirmed by historical EPA Station E19 (5,600 ug/kg DW)
(U.S. EPA 1982, 1983). Similar precision was observed between PCB concen-
trations for Station DR-10 (this study; 2,100 ug/kg DW) and historical EPA
Station E17 (2,400 ug/kg DW) (U.S. EPA 1982, 1983). The data are presented
in Figure 49.
The highest reported value from historical data was in the West Waterway
(24,000 ug/kg DW; Gamponia Station 14). PCB concentrations at nearby
stations from the present study (WW-06) and from historical studies were
considerably lower (typically from 600 to 1,200 ug/kg DW). Furthermore, PCB
concentrations reported by Gamponia et al. (1986) were higher than those
reported by other studies in other areas of the West Waterway and in
Area NH. Gamponia et al. (1986) reported a number of PCB concentrations
greater than 1,000 ug/kg DW in Area NH, including a concentration of
14,000 ug/kg in a slip east of Station NH-03 (Station 3; adjacent EPA
Station 4 had a concentration of 3,800 ug/kg DW). Elsewhere in Area NH,
Gamponia et al. (1986) reported concentrations greater than 5,000 ug/kg DW
at Station 1 (near NH-02, present study), Station 6, and Station 7 (near
Station NH-06, present study). TPPS Station S0034 had a relatively isolated
high PCB concentration of 3,100 ug/kg DW.
A number of stations with PCB concentrations between 1,000 and 4,000
ug/kg DW were reported in the Denny Way CSO area (Romberg et al. 1984). A
concentration of 2,600 ug/kg was reported at TPPS Station S0090 (near Station
SS-05), located along the Seattle South waterfront.
Other Organic Compounds--
A notable discrepancy between historical data and data from the present
study was observed for 2,4-dimethylphenol. Gamponia et al. (1986) reported
six observations in the general concentration range of 500-800 ug/kg DW in
the North Harbor Island area. In contrast, in the present study, 2,4-di-
138
-------
HPAH
>50,CCD
>15.000-50,000
>5.000-15,000
>500-5,000
untfstectwJ-500
50
40
3a
04
•5
23
3000
2000
Figure 48. Contours 01 HPAH concentrations in study area sediments.
139
\ Designation pg/kg
] Nuwbei tdry weight)
( 1 >50,CCD
\ 2 >15.000-50,000
50 3 >5.000-15,000
<34 X. 4 >500-5.000
5o 5 ufl(tetect«
-------
methyl phenol was undetected at most stations in the same area at detection
limits typically below 20 ug/kg DW.
Summary
¦ Contamination in the study area was spatially heterogeneous.
The most severe sediment contamination was localized,
suggesting the importance of local contaminant sources.
Relatively small-scale gradients were not prevalent, although
this may in part be a function of sample locations (i.e.,
samples were typically collected along shorelines rather than
in offshore or cross-channel transects).
h The Seattle South waterfront, the North Harbor Island area,
and the West Waterway were among the most severely con-
taminated study areas, and contained stations that accounted
for many of the highest concentrations observed in the study
(e.g., Stations SS-09, SS-08, SS-03, NH-03, NH-04, NH-06,
NH-08, WW-12, WW-14, and WW-19). Highly-contaminated
historical stations in these areas included Station 3 of
Gamponia et al. (1986) and EPA Stations 4 and 42 in the North
Harbor Island area, Stations 9 and 14 of Gamponia et al.
(1986) in the West Waterway, and TPPS Station S0090 along the
Seattle South waterfront. Other stations with relatively
high concentrations included EW-05 and EW-14 in the East
Waterway, Station AB-01 along Alki Beach, and Stations DR-08
and DR-12 in the upper Duwamish River. In contrast, stations
in outer Elliott Bay (i.e., Areas MG and AB, excluding inner
bay Station AB-01) were the least contaminated areas overall.
¦ Concentrations of copper, lead, mercury, and zinc were among
the most elevated of the metals detected in the study area.
Maximum EAR values for these chemicals ranged from 320 (for
copper and zinc) to 7,700 (lead), whereas median EAR values
were between 8 and 15. Concentrations of other metals that
were somewhat less elevated but nonetheless of concern
included arsenic (maximum EAR = 170), silver (maximum
EAR = 92), cadmium, chromium, and nickel. Maximum copper
concentrations (up to 2,050 mg/kg dry weight or DW) occurred
near the mouth of the West Waterway (Areas NH and WW) and
along the Seattle South waterfront (see Figure 43). Maximum
lead concentrations (up to 71,100 mg/kg DW) occurred along the
Seattle South waterfront and on the east side of the West
Waterway (see Figure 44). Maximum zinc concentrations (up to
6,010 mg/kg DW) occurred along the Seattle South waterfront
(see Figure 46). The highest mercury concentrations occurred
at relatively isolated stations [AB-01 (28.8 mg/kg DW),
NH-03, and EW-05] with generally elevated concentrations along
the Seattle South waterfront (see Figure 45).
¦ PAH and PCBs occurred at the highest concentration and were
the most frequently detected of the organic contaminants.
Other organic compounds occurred at high concentrations at
isolated stations. Maximum EAR values for PAH and PCBs
140
-------
ranged from nearly 1,000 (for PCBs) to 15,000 (for LPAH) to
over 40,000 (for HPAH). Median EAR for these compound
classes ranged between 32 and 80. Other organic compounds
that were infrequently found at elevated concentrations
(maximum EAR >100) included 1,4-dichlorobenzene (maximum
EAR = 8,900), benzyl alcohol (maximum EAR ¦ 880), 4-methyl-
phenol, pentachlorophenol, butyl benzyl phthalate, and
retene. PAH concentrations were most elevated along the
Seattle South waterfront (up to 3,800,000 ug/kg DW total
PAH), but were also highly elevated in the North Harbor
Island study area, the East and West Waterways, and at
isolated stations in other areas (see Figures 47 and 48).
PCB concentrations were most elevated in the upper Duwamish
River (up to 5,800 ug/kg DW) and along the Seattle South
waterfront, but high concentrations occurred throughout the
Duwamish River and in the North Harbor Island study area
(Figure 49). Relatively high detection limits reported for a
number of organic compounds (most notably chlorinated
phenols, chlorinated benzenes, hexachlorobutadiene, benzyl
alcohol, and benzoic acid) impeded data analysis for these
compounds.
¦ Except among PAH and related compounds, few correlations
among chemicals were observed on a study-wide basis. When
correlations were performed on a smaller scale for individual
study areas, strong correlations were observed, particularly
among metals. These results are consistent with the"presumed
importance of sources that predominate in localized areas.
BI0ACCUMULATI0N
Bioaccumulation studies were conducted to determine if selected con-
taminants in sediment or water were accumulated in the tissues of indigenous
organisms. This section summarizes the bioaccumulation of mercury, selected
chlorinated pesticides, and PGBs in edible muscle tissue of English sole.
The objectives of this section are to describe geographic trends in bioaccum-
ulation and to determine whether tissue contaminants in the Elliott Bay
study area were significantly elevated above concentrations observed at the
reference area (Point Pully).
Normalization of Chemical Concentrations
A number of interrelated variables have been observed to influence the
concentrations of lipophilic compounds in fish tissue, most notably tissue
lipid content, sex, age, and season of collection (Phillips 1980). Of these
variables, lipid content (estimated as total extractable organic matter) is
often used for normalization, based on empirical relationships between
pollutant concentrations and lipid content (e.g., Sloan et al. 1985), as
well as theoretical (partitioning) considerations (e.g., Mackay 1982; Chiou
1985). Strong relationships between bioaccumulation factors and partition
coefficients [e.g., octanol-water partition coefficients (kpw) or trigly-
ceride-water partition coefficients (k^)] in controlled studies support the
theory that hydrophobic chemicals partition into organism lipids from water
as they would partition into an organic solvent. Therefore, lipid normaliza-
141
-------
Designation
' Humtear
ug/kg
(«Y wsigwi
>2500
>1000-2500
>500-1000
>150-500
undaiected-150
Figure 49. Ccntot rs at PCS conceritrarors in study area SBdiments
-------
tion was examined in this study to examine trends that may have been
obscured on a wet weight basis. One important factor that can reduce the
applicability of lipid-bioaccumulation relationships is deviation from
equilibrium conditions in the environment (e.g., as related to the kinetics
of uptake and depuration).
Evaluation of the Reference Area
Concentrations of mercury, PCBs, and pesticides in muscle tissue of
English sole collected from the Point Pully reference area were within the
range of values expected for areas of Puget Sound that are remote from
contaminant sources. The mean concentration of mercury in Point Pully
samples was 0.089 mg/kg wet weight, with a range of 0.056 to 0.118 mg/kg wet
weight. The concentration of PCBs in Point Pully samples was 5.4 ug/kg wet
weight, with a range of 2-19 ug/kg wet weight. EPA priority pollutant
pesticides were not detected in muscle tissue of English sole from Point
Pully.
By comparison, the mean concentration of mercury found in muscle tissue
of English sole from other reference areas of Puget Sound was 0.04 mg/kg wet
weight in Discovery Bay (Gahler et al. 1982) and <0.06 mg/kg wet weight in
Carr Inlet (Tetra Tech 1985a). The mean concentration of PCBs in muscle
tissue of English sole was <13 ug/kg wet weight in Discovery Bay (Gahler et
al. 1982) and 36 ug/kg wet weight in Carr Inlet (Tetra Tech 1985a). EPA
priority pollutant pesticides were not detected in the samples from Carr
Inlet (Tetra Tech 1985a)'and from Discovery Bay (Gahler et al. 1982), with
two exceptions. p,p'-DDT and p,p'-DDE were detected in English sole muscle
from Discovery Bay at low concentrations (<1 ug/kg wet weight and 3 ug/kg
wet weight, respectively).
In summary, Point Pully appears to be an adequate reference area for
evaluation of bioaccumulation in English sole.
Mercury in Fish Tissue
The results of the mercury analyses are presented in Table 23. The
means and ranges of each station are based upon data for five individual fish
(not including analytical replicates). The data do not suggest that mercury
in tissue is of concern, as the reference station from Point Pully (PP-91)
had the highest mean mercury concentration of all stations sampled
(Table 23). Although maximum concentrations of mercury at 3 of 11 stations
equalled or exceeded the maximum value observed at the reference station,
the ranges were not broad and never exceeded 0.15 mg/kg wet weight. No
Elliott Bay/Duwamish River test station had a mercury concentration that was
statistically different from that at the reference station (t-test; P>0.001).
^e ^max test revealed homogenous variances for all comparisons,
suggesting tnat wet weight normalization was appropriate for statistical
analysis. However, several additional analyses were conducted to examine
whether normalizations other than wet weight might be appropriate for these
data. Pearson correlation analyses were performed for mercury vs. lipid
content and mercury vs. fish age. For both variables, correlations were^very
poor (mercury vs. lipids: r=-0.08, P>0.05, n=60; mercury vs. fish age:
r=0.32, P<0.05, n=60). It is possible (but speculative) that such relation-
143
-------
TABLE 23. MERCURY CONCENTRATIONS IN
FISH COLLECTED FROM ELLIOTT BAY,
DUHAMISH RIVER, AND POINT PULLY
(¦g/kg wet weight)
Station
Mean*
Range
AB-91
0.064
0.013 - 0.125
DR-91
0.071
0.053 - 0.099
EW-91
0.036
0.018 - 0.055
KG-91
0.068
0.048 - 0.097
MG-91
0.084
0.061 - 0.114
NH-91
0.059
0.022 - 0.102
NH-92
0.077
0.045 - 0.118
NS-91
0.074
0.058 - 0.099
pp_9lb
0.089
0.056 - 0.118
SS-91
0.037
0.011 - 0.071
SS-92
0.060
0.023 - 0.099
WW-91
0.085
0.029 - 0.149
a Arithmetic mean.
b Reference station; no test stations were significantly
different from the reference station (P>0.001).
144
-------
ships could have existed but were not discernible because of the narrow
range of observed mercury concentrations. Mercury data for all stations
were evaluated for normality with the Kolmogorov-Smirnov (K-S) test (Sokal
and Rohlf 1981). The data were normally distributed (P>0.05). Lipid
normalization of the data skewed the data set toward the lower end of the
concentration range and resulted in a distribution that was not normal (K-S
test; P<0.05). These results indicate that lipid normalization was not
preferable to wet weight normalization to satisfy the assumptions of the
statistical tests used for comparisons with the reference area.
PCBs/Pesticides in Fish Tissue
This section focuses on PCB concentrations, as EPA priority pollutant
pesticides were seldom detected in this study. Detection limits for single-
component pesticides ranged from 1 to 50 ug/kg wet weight, but typically were
less than 5 ug/kg wet weight. The only pesticide detected at greater than
10 ug/kg wet weight was p,p'-DDE (410 ug/kg wet weight) for a single fish at
trawl Station SS-91 (along the Seattle South waterfront). For other fish in
trawl Station SS-91, p,p'-DDE was undetected at a detection limit of 2 ug/kg
wet weight. The 410 ug/kg concentration was confirmed by GC/MS.
Statistical comparisons between test stations and the reference station
were conducted as for mercury bioaccumulation data. However, for PCB
bioaccumulation based on wet weight concentrations, the Fmax test revealed
extremely heterogeneous variances for all trawl stations except AB-91
(alpha = 0.05). Observed F values exceeded the critical F value by a factor
of 35 to >1,100. Comparable results were obtained for lipid normalized PCB
data. Therefore, PCB concentrations (wet weight) were logio-transformed to
better satisfy the assumption of homogeneous variances for the t-test. Log
transformation resulted in homogenous variances for all comparisons (alpha »
0.05). Cumulative frequency plots of all individual fish (n=60) revealed
that, for the overall data set, log transformation resulted in a more normal
distribution than was observed for the untransformed wet weight PCB concen-
trations. However, neither the untransformed nor the transformed PCB data
were normally distributed (K-S test, P<0.05).
Results of the t-tests using log-transformed PCB data (wet weight) are
presented in Figure 50. Eight of the eleven trawl stations had mean PCB
concentrations that were significantly greater than the reference area mean
(P<0.001). Only trawl Stations AB-91, MG-91, and NH-92 were not significant-
ly different than the reference station (PP-91).
Geometric mean concentrations (ug/kg wet weight) are presented in
Figure 50 and in Table 24, along with along with data for lipid content and
ages of fish in all trawls. Concentrations were highest in the Duwamish
River and tended to decrease toward outer Elliott Bay. Specifically, PCB
concentrations were highest in the East and West Waterways of the Duwamish
River (460-470 ug/kg wet weight) and decreased slightly moving upriver
(390-400 ug/kg wet weight at trawl Stations KG-91 and DR-91). Trawl Station
NH-91, located north of Harbor Island, had a geometric mean concentration of
350 ug/kg wet weight; concentrations decreased rapidly moving west from
Harbor Island (53 ug/kg wet weight at trawl Station NH-92 and 2.4 ug/kg wet
weight at Station AB-91). Mean concentrations at three trawl stations along
the Seattle waterfront ranged from 190 to 250 ug/kg wet weight, whereas trawl
145
-------
1200T 1200T
LEGEND
<71
700 -i
600 -
P 500 -
<
DC 5
t — 400 -I
z a)
111 $
o *
Z O)
o*
O O)
n
0)'-'
o
0.
300-
200 -
100
ui
o
z
UI
DC
UI
U.
UI
K
:-r»:
I MEAN
I
9 z
5 o
II
H- UJ
m a
+i
PP Poinl Pully
IIG Magnolia Bluff
NS Seattle North Waterfront
S S Seattle South Waterfront
NH North Haifoor Island
EW East Waterway
WW West Waterway
KG Kellogg Island
DR Duwamish River
AB Alki Beach
* Mean concentration was
signilicanlly different lhan refer-
ence (PP-91) at PcO.001.
n e S for each station.
PP-91 MG-91 SS-92 NH-91 WW-91 DR-91 AB-91
NS-91 SS-91 EW-91 KG-91 NH-92
TRAWL STATION
Figure 50. PCB concentrations (ng/kg wet wt.) in muscle tissue of English sole from Elliott Bay,
the Duwamish River, and Point Pully. Means and standard deviations are based upon
log10-transformed data.
-------
TABLE 24. PCB CONCENTRATIONS, LIPID CONTENTS,
AND AGES OF ELLIOTT BAY FISH
PCB Concentration Lipid Content
(ug/kg wet weight) (percent of wet weight) Age (yr)
Station
Meana
Range
Meanb
Range
Mean^
Range
AB-91
E2.4C
E1-E10
0.42
0.20-0.63
11
7-16
DR-91
E390
E280-E630
1.9
0.94-3.1
8.6
7-11
EW-91
470
230-2,060
1.3
0.59-1.8
5.0
3-9
KG-91
E400
E190-E730
1.9
0.94-2.9
9.0
7-10
MG-91
E63
E13-1,490
0.78
0.50-1.2
8.2
6-10
NH-91
E350
E150-690
2.9
1.2-4.2
6.0
3-9
NH-92
E53
E12-480
1.3
0.44-2.8
8.8
6-12
NS-91
E190
79-E420
1.2
0.25-1.9
6.6
4-8
PP-91
E5.4
E2-E19
0.85
0.64-1.0
8.8
8-10
SS-91
El 90
E42-670
1.5
0.62-3.1
8.2
4-15
SS-92
E250
E53-530
2.1
1.5-3.2
5.0
4-7
WW-91
E460
100-1,030
0.68
0.38-1.0
7.2
3-11
a Geometric mean (based on the arithmetic mean of log-transformed samples). Log-
transformed data were used to satisfy the assumptions of statistical tests.
b Arithmetic mean; appropriate because the data were normally distributed.
c E = Estimated concentration.
147
-------
Station MG-91 had a considerably lower mean concentration (63 ug/kg wet
weight). An anomalously high concentration was reported for an individual
fish in the MG-91 trawl (1,490 ug/kg wet weight; Table 24). Although bio-
accumulation data are often variable because fish are mobile and can have
different exposure histories, the anomalous PCB concentration in one Magnolia
fish was extreme (roughly 30 times higher than any other fish in the trawl).
To examine whether alternative normalization would be appropriate for
these data, PCB concentrations were correlated against lipid content and fish
age. The correlation coefficient (r) for PCB concentration (wet weight) vs.
lipid content was 0.27; the correlation coefficient for PCB concentration vs.
age was 0.08. To some extent, this lack of correlation is indicated in
Table 24 [e.g., the fish trawl from Alki Beach (AB-91) had the lowest mean
and range for PCB concentration but had the greatest mean and range for fish
age]. Scatterplots revealed generally poor correlations and indicated that
the low correlation coefficients did not result from a small number of
anomalous cases.
Based on these data, geographic location of fish (and presumably, the
PCB exposure associated with different geographic areas) is a more important
factor in PCB bioaccumulation than lipid content or age (for the age classes
sampled). Whereas PCB concentrations appeared to correspond to proximity
to the Duwamish River, PCB concentrations did not correlate well with lipid
content or age of fish over the entire data set.
¦ |
Comparison with Recent Historical Data
The bioaccumulation data for English sole collected during this study
were compared with data from recent studies by Malins et al. (1982), Romberg
et al. (1984), and Landolt et al. (1985). Because of the limited number of
observations of contaminant levels in muscle tissue of English sole from
Elliott Bay, only qualitative comparisons of the results of this study with
previous studies could be made.
Mercury, concentrations in four composite samples of muscle tissue from
trawl-caught English sole were analyzed during the Metro TPPS at three
locations within the present study area [Alki Point (two samples), West
Point, Denny Way CS0]. Numbers of individual fish represented in each
sample were not given by the authors. The concentrations of mercury in the
four samples ranged from 0.061 mg/kg (wet weight) at Alki Point to 0.125
mg/kg (wet weight) at the Denny Way CS0, with a mean concentration of
0.08 mg/kg (wet weight). The mean concentrations of mercury in English sole
muscle tissue observed during the present study ranged from 0.036 mg/kg (wet
weight) in the East Waterway to 0.089 mg/kg (wet weight) at Point Pully.
These concentrations are comparable to those found by Romberg et al. (1984).
EPA priority pollutant pesticides were generally not detected in the
English sole samples collected by Romberg et al. (1984) [mean detection
limits from 0.01 to 3 ug/kg (wet weight) were reported]. The pesticides DDD
and DDE were measured in several of the samples, with a maximum total
concentration of 7.0 ug/kg (wet weight) in fish collected offshore of the
Denny Way CS0. Because these were composite samples, concentrations of DDD
and DDE in individual fish could be much greater than the reported values.
Nevertheless, their results are consistent with those of the present study
148
-------
[i.e., pesticides were generally not detected; the only high level observed
was 410 ug/kg (wet weight) of p,p'-DDE in a single fish at Station SS-91].
In historical studies, mean concentrations of PCBs in muscle tissue of
English sole ranged from 11 ug/kg (wet weight) to 2,100 ug/kg (wet weight),
with a mean of about 400 ug/kg (wet weight; n=14 samples, some of which were
composites). Romberg et al. (1984) found an average of 24 ug/kg (wet
weight) of PCBs in English sole muscle at Alki Point, 290 ug/kg (wet weight)
at Denny Way CS0, and 486 ug/kg (wet weight) at West Point. Mai ins et al.
(1982) found a range of 270-2,100 ug/kg (wet weight) in four samples of
English sole muscle from Elliott Bay (specific sampling locations not
reported). PCB concentrations in muscle tissue samples from five individual
English sole collected by Landolt et al. (1985) from Smith Cove ranged from
20 ug/kg to 47 ug/kg (wet weight), with a mean of 28 ug/kg (wet weight). A
clear spatial pattern was not evident in the historical PCB data. The
overall mean PCB concentration for historical studies is of the same order
of magnitude as several of the highest mean concentrations found in the
present investigation. The low concentrations reported from Smith Cove by
Landolt et al. (1985) are consistent with the low values observed at the
Magnolia area in the present study. The PCB value at West Point found by
Romberg et al. (1984) is substantially higher than the mean PCB concentration
reported here for the Magnolia area. The apparent discrepancy could be due
to differences in specific sampling locations. The station sampled by
Romberg et al. (1984) was north of West Point and could represent a local
population of English sole that is somewhat distinct from that in the.
Magnolia area sampled during this study.
Summary
¦ Mercury bioaccumulation in English sole in the Elliott
Bay/Duwamish River area was not significant relative to that
near Point Pully; in fact, the reference area had the highest
mean mercury concentration observed in the study.
¦ Pesticide bioaccumulation was not important in the study area
with the exception of p,p1-DDE, which was detected in a single
fish collected along the Seattle South waterfront (410 ug/kg
wet weight).
¦ PCB bioaccumulation was significant over much of the study
area, with the highest concentrations observed in the Duwamish
River (specifically in the East and West Waterways). Concen-
trations tended to decrease with distance from the mouth of
the Duwamish River.
SEDIMENT BI0ASSAYS
The results of sediment toxicity tests using the amphipod Rhepoxvnius
abronius are presented in this section. First, amphipod bioassay results for
Port Susan are compared with results from other reference areas used during
previous studies. The amphipod mortality values for each station in the
Elliott Bay system are then presented and compared statistically with the
Port Susan values. Finally, results of the present study are compared with
149
-------
those of previous studies on the toxicity of sediments in Elliott Bay to
R. abronius.
Evaluation of the Reference Area
Mean values of amphipod survival and their 95 percent confidence limits
are shown in Figure 51 for individual stations in Port Susan and other
reference areas of Puget Sound. Data for the 1986 survey of Port Susan,
which were collected as part of the Everett Harbor Toxics Action Program, are
also shown in the figure. Mean amphipod survival for several of the Port
Susan observations was low (<80 percent) relative to data for most other
reference areas. Mean survival was also low (<75 percent) at one station in
Carr Inlet, where a single replicate was an extreme outlier, and at one
station in Sequim Bay. The relatively low survival at some Port Susan
stations can not be explained by a response of the amphipods to fine-grained
sediments. The product-moment correlation between mean amphipod survival
and percent fine-grained material was not significant (r=0.38, R>0.05, n=7).
The range of percent fine-grained material in samples collected during
1985-1986 ih Port Susan was 7.4-88 percent. Only a single sample contained
more than 24 percent fine-grained material. Moreover, mean amphipod
mortality for that sample (Station PS-01, 1985) was relatively low (13 per-
cent) .
The 1985 data for amphipod response to Port Susan sediments are
considered to be adequate for use as reference data. At one of the Port
Susan stations (PS-02), mean amphipod survival was only 76 percent, a value
that indicates marginal toxicity compared to the criterion of <75.4 percent
survival for toxic sediments in Mearns et al. (1986). However, the mean
amphipod survival among all four stations sampled during 1985 was 84 percent.
Moreover, contaminant concentrations in sediments of Port Susan were
typically within the range of those observed at other reference areas in
Puget Sound;(see SEDIMENT CHEMISTRY). Thus, all of the 1985 bioassay data
for Port Susan were pooled for statistical comparisons with data from sites
in the Elliott Bay system.
The Elliott Bay bioassay data could have been compared with native
sediment controls (i.e., West Beach, Whidbey Island) rather than with Port
Susan to establish statistically significant elevations in mortality.
Because of the relatively low levels of mean mortality (i.e., <10 percent)
in native sediments, comparisons of Elliott Bay sediments to native-sediment
controls would have resulted in identification of more stations with
significantly elevated amphipod mortality. However, such comparisons may
confound thei effects of simply removing amphipods from native sediment with
toxic effects or other site-specific factors in Elliott Bay. Comparisons
with a separate reference area (i.e., Port Susan) were used to account for
possible factors related to removal of amphipods from native sediments.
General Patterns of Amphipod Mortality
The mean amphipod mortality and the range of station-specific means for
each study area within the Elliott Bay system and the reference area are
shown in Figure 52. The highest overall mortalities were found in the North
Harbor Is1 and area (mean mortality = 60 percent) and in East Waterway (mean
mortality = 43 percent). The maximum mortality (100 percent) was observed at
150
-------
(J1
20
18
0)
dc
o
>
>
DC
3
(0
u.
0
DC
UJ
01
S
3
Z
16-
14
12-
10-
8-
LEGEND
MEAN
II
96%
CONFIDENCE
INTERVAL
1984
1985
1984
I
1985
1985
1985
1985
1986
CARft
INLET
BLAKELY
HARBOR
SEQUIM
BAY
PORT
SUSAN
SAMPLING YEAR/REFERENCE AREA
Figure 51. Amphipod bioassay data from Puget Sound reference areas.
-------
LEGEND
MEAN
100
<
K-
tr
o
<
LU
80-
60-
UJ
O 40
DC
Hi
O.
20-
l!el
6.0
5.0
-4.0
111
m iii
«o
3.0 g
UJ
j- 111
u-
•2.0 < 111
> IE
UJ
i o iij
P S Port Susan (n = 4)
MG Magnolia BluH (n - 4)
N S Seattle North Waterfront (n = 8)
SS Seattle South Waterlront (n = 11)
NH North Harbor Island (n = 11)
E W East Waterway (n = 16)
WW West Waterway (n = 19)
KG Kellogg Island (n = 11)
DR Duwamish River (n = 18)
AB Alki Beach (n = 4)
PS
MQ NS SS NH EW WW KG DR AB
STUDY AREA
Figure 52. Mean and range of amphipod bioassay responses within study areas.
-------
Stations NH-08 and EW-05. The range of mean mortality at stations within
most study areas was large, indicating considerable spatial heterogeneity.
Although relatively high mean mortalities (>35 percent) were observed within
all study areas (except Magnolia), some stations within each area exhibited
mean mortality values as low or lower than the lowest value (i.e., 10 per-
cent) in Port Susan.
Comparison of the Elliott Bav System with Port Susan
Results of the amphipod bioassay tests for all stations sampled during
the Elliott Bay investigation are summarized in Table 25. Statistical,
comparisons between Elliott Bay sites and the reference area (Port Susan
1985 data pooled) indicated that mortalities in 17 test sediments were
significantly different from the Port Susan samples (PO.OOl). Seven of the
nine study areas tested contained one or more sampling sites with statisti-
cally significant amphipod mortality (Figure 53).
In 14 of the 28 samples that exhibited a mean amphipod mortality
>40 percent, mean mortality was not statistically different from the mean
reference value at PO.OOl. However, all 28 of these samples did exhibit a
significant difference from the reference area at P<0.05. The lack of
significance at PO.OOl for mean mortality values over 40 percent can be
explained by low statistical power (<0.6 at PO.OOl), partly due to the
relatively high mean mortality in the reference area. Also, the variance of
the bioassay test is typically higher at intermediate mortality values
(35-65 percent) compared with the extremes of the mortality range. The
variance of the mean in the Elliott Bay amphipod bioassays was very high
[standard deviation >28, corresponding to a standard error (SE) >12] at four
stations: Stations SS-03, NH-02, DR-13, and DR-16. The relative influence
of good reference survival vs. low variability on the statistical power of
the amphipod bioassay is being investigated in a separate EPA project on
refinement of sediment quality values.
Comparison with Recent Historical Data
Previous R. abronius sediment bioassay tests in Elliott Bay, which used
acceptable protocols (cf. Swartz et al. 1985) and fresh sediments, are
summarized by Evans-Hamilton and D.R. Systems (1987). Other kinds of
bioassay tests performed on fresh sediments from the Elliott Bay system are
summarized by Long (unpublished). The majority of Elliott Bay has not been
tested for sediment toxicity to amphipods (see Figure 50).
The Magnolia stations (MG-01 to MG-04) were all non-toxic. Previous
sediment bioassays in this area have shown a similar lack of toxicity to
R. abronius.
Previous sampling along the Seattle Waterfront North (Stations NS-01 to
NS-08) has been concentrated near the Denny Way CS0, where sediments have
consistently been shown to be toxic. Evans-Hamilton and D.R. Systems (1987)
determined that the north end of Pier 91 and the Denny Way CS0 were sites of
most concern based on the amphipod bioassay. In the present study, these
two sites (represented by Stations NS-01 and NS-08, respectively) were the
only stations that displayed significant toxicity (PO.OOl) in the Seattle
Waterfront North area.
153
-------
TABLE 25. SUMMARY OF AMPHIPOD BIOASSAY RESULTS
Range of Mean
Mortality Mortality3
Station (percent) (percent)
AB-01
25-80
47
( 9.3
AB-02
0-15
6
( 2.9
AB-03
0-10
3
( 2.0
AB-04
0-5
3
( 1.2
DR-01
5-25
13
( 3.7
DR-02
20-80
38
(11.0
DR-03
2.5-35
14.5
( 6.5
DR-04
10-30
18
( 4.1
DR-05
0-100
29
(18.3
DR-06
0-15
8
( 3.0
DR-07
0-20
8
( 3.7
DR-08
0-45
25
( 7.6
DR-09
15-20
17
( 1.2
DR-10
0-10
7
( 2.0
DR-11
5-35
23
( 5.1
DR-12
.5-25
17
( 3.4
DR-13
15-90
57
(16.3
DR-14
15-60
32
( 9.6
DR-15
85-100
89
( 2.9
*
DR-16
5-90
45
(14.6
DR-17
0-20
. 6
( 3.7
DR-25
15-65
40
( 9.7
EW-01
0-10
3
( 2.0
EW-02
20-65
39
( 8.3
EW-03
5-100
29
(18.1
EW-04
50-75
58
( 5.1
*
EW-05
100b
100
( 0)*
EW-06
30-55
39
( 4.3
~
EW-07
50-80
63
( 5.8
~
EW-08
35-90
65
(11.3
EW-09
20-85
59
(10.9
EW-10
30-90
58
(12.1
EW-11
40-80
62
( 7.2
*
EW-12
10-25
16
( 2.4
EW-13
10-45
24
( 6.8
EW-14
5-70
22
(12.1
EW-15
10-20
16
( 1.9
EW-16
20-40
31
( 3.7
154
-------
TABLE 25. (Continued)
Range of Mean
Mortality Mortality3
Station (percent) (percent)
KG-01
10-35
22
( 4.1
KG-02
10-60
37
(10.2
KG-03
10-40
27
( 5.4
KG-04
5-35
18
( 4.9
KG-05
25-45
32
( 3.7
KG-06
5-15
9
( 2.4
KG-07
5-75
23
(13.3
KG-08
5-35
16
( 5.1
KG-09
25-40
33
( 2.5
*
KG-10
15-45
31
( 5.3
KG-11
25-40
32
( 2.5
*
MG-01
5-10
7
( 1.2
MG-02
0-10
3
( 2.0
MG-03
0-10
7
( 2.0
MG-04
0-15
6
( 2.9
NH-01
0-65
20
(12.0
NH-02
20-90
45
(12.6
NH-03
85-100
94
( 3..7
~
NH-04
85-95
87
( 2.0
*
NH-05
70-100
80
( 6.3
~
NH-06
75-100
83
( 4.6
*
NH-07
5-15
9
( 2.4
NH-08
100b
100
( 0)*
NH-09
30-90
58
(10.2
NH-10
10-45
24
( 5.8
NH-11
30-95
55
(11.5
NS-01
45-75
58
( 6.2
*
NS-02
0-65
16
(12.4
NS-03
0-25
13
( 4.1
NS-04
0-70
33
(13.7
NS-05
0-15
10
( 2.7
NS-06
0-65
15
(12.6
NS-07
10-65
36
( 9.7
NS-08
65-100
82
( 6.8
*
PS-01
5-20
13
( 2.5
PS-02
15-35
24
( 3.7
PS-03
0-20
10
( 3.5
PS-04
0-30
17
( 5.4
155
-------
TABLE 25. (Continued)
Range of
Mean
Mortality
Mortalitya
Station
(percent)
(percent)
SS-01
0-20
9
( 3.7)
SS-03
5-80
57
(14.5)
SS-04
0-45
13
( 8.3)
SS-05
5-40
18
( 6.0)
SS-06
30-65
45
( 5.7)*
SS-07
15-35
30
( 3.9)
SS-08
20-65
44
( 8.3)
SS-09
15-45
29
( 5.1)
SS-10
5-20
14
( 2.9)
SS-11
0-15
10
( 2.7)
SS-12
5-35
19
( 4.8)
WW-01
5-25
15
( 3.5)
WW-02
65-100
82
( 6.6)*
WW-03
0-20
9
( 3.3)
WW-04
0-25
11
( 4.0)
WW-05
5-15
9
( 2.4)
WW-06
10-40
19
( 5.3)
WW-08
15-70
41
(11.3)
WW-09
30-90
60
(11.7)
WW-10
5-25
12
( 3.4)
WW -11
30-70
41
( 7.5)
WW -12
15-75
33
(10.8)
WW-13
5-40
15
( 6.3)
WW-14
5-30
18
( 5.4)
WW-15
5-20
13
( 2.5)
WW-16
10-35
17
( 4.6)
WW-17
10-25
16
( 2.9)
WW-18
5-35
14
( 5.3)
WW-19
10-25
18
( 3.0)
WW-20
5-50
22
( 7.8)
Controls - clean
1
0-10
1
( 1.0)c
2
0-20
10
( 3.5)
3
0-10
5
( 1-6)
4
0-5
1
( 1.0)
5
0-15
2.2
( 0.8)d
Control-Cd-spi ked
O
o
cr
100
( 0)
156
-------
TABLE 25. (Continued)
a Mean mortality is based on five replicate samples per station,
unless otherwise indicated. Standard error of each mean is given
jn parentheses.
b A mortality level of 100 percent was observed for each of the
five replicates.
c Mean mortality is based on ten replicate samples per station.
d Mean mortality is based on twenty replicate samples per station.
* - Denotes that mean mortality differed significantly (P<0.001)
from the mean mortality of pooled replicates from the four Port
Susan stations.
157
-------
MS-01
NH-Ofl
NH-06
EW-06
EW-05
EW-04
KQ-11
OR-15
6000
2000
Figure 53. Significant amphipod bioassay mortalities compared
to the Port Susan reference area.
158
-------
The Seattle Waterfront South (Stations SS-03 to SS-12) has shown patchy
toxicity, and has been less toxic than expected (Long 1984). Evans-Hamilton
and D.R. Systems (1987) delineated three sites of most concern based on the
amphipod bioassay. In the present study only one of these (SS-06) showed
statistically significant toxicity (PcO.OOl). Nevertheless, two stations
(SS-03 and SS-08) located at sites where high toxicity was found previously
exhibited high values of mean mortality (>40 percent). Despite the high
mortality levels, these stations were not statistically different (PcO.OOl)
from the reference area. High variability (defined as SE >10) was observed
at Station SS-03.
North Harbor Island (Stations NH-01 to NH-11) has been shown to have
high toxicity, particularly around four stations which were significantly
toxic (PO.OOl) in the present study: NH-03, NH-04, NH-05, and NH-06.
Three additional stations (NH-02, NH-09, and NH-11) had high toxicity in
previous studies. In the present study, these three stations displayed high
mean mortality (>40 percent) and high variability. All three sites exhibited
statistical differences from the reference area at P<0.05, but not at
PcO.OOl. Station NH-08 was significantly toxic (PO.OOl) in the present
study, whereas previous studies at this site had shown very low amphipod
mortalities (Evans-Hamilton and D.R. Systems 1987).
East Waterway (Stations EW-01 to EW-16) showed high toxicity along the
transect from EW-04 to EW-11. Although only five stations (EW-04, EW-05,
EW-06, EW-09, EW-11) were statistically different from the reference area
(P<0.001), three additional stations (EW-08, EW-09, EW-10) had mean
mortalities of 58-65 perpent and high variability. This transect has shown
similar previous evidence of high toxicity (Evans-Hamilton and D.R.- Systems
1987) that extended through two stations which showed low toxicity in the
present study: Stations EW-12 and EW-13.
West Waterway (Stations WW-01 to WW-19) had only one area of significant
toxicity (PcO.OOl) in the present study: Station WW-02. Previous studies
have shown high toxicity at this site, and also at sites WW-08, WW-09 and
WW-11 (Evans-Hamilton and D.R. Systems 1987). These three latter stations
all had greater than 40 percent mortality in the present study, but were not
statistically different (PcO.OOl) from the reference area. The amphipod
tests at Stations WW-08 and WW-09 displayed high variability.
Kellogg Island (Stations KG-01 to KG-11) has shown previous high
toxicity at the tip and immediately south of Harbor Island (Long unpublished;
Evans-Hamilton and D.R. Systems 1987). The same pattern was shown in the
present study, with significant toxicity at Stations KG-09 and KG-11
(PcO.OOl).
The Upper Duwamish Estuary (Stations DR-01 to DR-17) had. three stations
with mean mortalities greater than 40 percent, one of which (DR-15) was
significantly more toxic than the reference area, while the other two (DR-13
and DR-16) were not significant due to high variability. Toxicity at all
three stations is as expected from previous studies in which greater than 40
percent mortality was recorded (Evans-Hamilton and D.R. Systems 1987).
159
-------
The final area, Duwamish Head/ATki Beach (Stations AB-01 to AB-04), had
one station (AB-01) with greater than 40 percent mortality. This station
was significantly different from reference at P<0.05, but not at PcO.OOl.
The other three stations had low toxicity. These results are in accord with
previous studies (Evans-Hamilton and D.R. Systems 1987).
In conclusion, the results of the present study are generally in accord
with previous work. The only major differences were unexpectedly high
toxicity of Station NH-08 and the low toxicity at Stations EW-12 and EW-13.
It is possible that sediment toxicity at these three sites has changed due to
dredging, sedimentation, changes in contaminant inputs, or other causes.
However, it is also possible that these differences are simply a function of
the patchy distribution of sediment toxicity in the waterways (Swartz et al.
1982). Results of the replicate tests conducted in the present study should
be considered representative of present conditions at the stations tested.
Summary
¦ Sediments from 17 of the 102 Elliott Bay sites tested
displayed significant toxicities (P<0.001) in the amphipod
bioassay when compared with the Port Susan reference area
(Figure 54; see Table 25)
¦ Overall, there was good agreement between the present study
and previous bioassay studies in Elliott Bay (Figure 54)
n The two most toxic areas in Elliott Bay were North Harbor
Island and East Waterway.
BENTHIC MACROINVERTEBRATES
The purposes of this section are to describe the general characteris-
tics of benthic communities within the segments of Elliott Bay and in the
Port Susan reference area, and to identify possible areas where benthic
communities are impacted. Characterization of the benthic communities
within Elliott Bay is based primarily on abundances of the major infaunal
taxonomic groups (i.e., polychaetes, molluscs, and crustaceans). Complete
species-level analyses of benthic communities were conducted at 16 selected
stations in Elliott Bay and at all 4 stations in Port Susan to supplement
the characterizations based on abundances of major taxa. Identifications of
possible impacted areas in the Elliott Bay study area were based on statis-
tical comparisons between the abundances of major taxa groups in Elliott Bay
and those in the Port Susan reference area.
The following discussion is organized into several major topics. First,
the adequacy of the reference area (i.e., Port Susan) is evaluated. Second,
the general characteristics of infaunal communities within the segments of
Elliott Bay and in Port Susan, are described, and abundances of major
taxonomic groups are compared statistically to identify possible areas of
impact. Third, species-level characteristics of the completely identified
stations are described. Individual stations and segments within Elliott Bay
that appear to be degraded and the degree of apparent degradation are then
discussed.
160
-------
•*MS-OB
¦ AN&01
¦ ASS-06
¦ *SS<03
¦A N*02
¦AAB01
¦4NH4fl
AEW-13
¦ EW-tO
¦A EW.1t
¦* EW-09
WUM»
aAVWMI
KG»11
LEGEND
SIGNIFICANT OR CREATED THAN 40% MORTALITY
• PR6SSNT STUDY
PREVIOUS STUDY
(Svant>HwnUtonandO. ft Sy«UMt«
-------
Evaluation of the Reference Area
The ideal reference area for any investigation of anthropogenic impacts
would be identical to the potentially impacted area, but would lack all
anthropogenic influences. This condition is, of course, unachievable
because no two areas are exactly alike, and because nearly all areas exhibit
some evidence of human activities. Nevertheless, the reference area should
exhibit physical, chemical, and biological characteristics that are as
similar as!possible to the study area (anthropogenic stresses excepted), so
as not to unduly bias comparisons between the two areas. In this investiga-
tion, Port Susan was selected as the reference area for benthic communities
in Elliott iBay because:
¦ It is one of only a few areas in northern Puget Sound with a
major riverine input (i.e., the Stillaguamish River), similar
to that of the Duwamish River in Elliott Bay
¦ In comparison with other bays in northern Puget Sound, Port
Slusan does not have any obvious sources of contamination
(except slight organic enrichment from the Stillaguamish
River) and it exhibits low concentrations of sediment con-
taminants
¦ It exhibits a range of sediment grain sizes.
A major difference between Port Susan and Elliott Bay is the wind and wave
exposure regimes. Port Susan is a fairly narrow bay and is protected from
the larger waves generated in the main Puget Sound basin, whereas, a large
portion of the study area in Elliott Bay (e.g., Magnolia, Duwamish Head/Alki
Beach) is exposed to these types of wind and wave conditions.
Another difference between the two areas is in the sampling design used
for each area. Four stations were sampled in Port Susan in 1985. Stations
PS-01 to PS-04 were located on the western side of Port Susan, along the
12-m isobath. (Stations PS-02, PS-03, and PS-04 were also sampled in 1986 in
conjunction with the Everett Harbor investigation.) None of these stations
was located in the upper reaches of the Stillaguamish estuary, whereas
benthic infaunal stations in Elliott Bay were located in eight of the nine
segments (delineated in this report) and extended to Kellogg Island in the
Duwamish River (see Figure 5).
Port Susan is known to be an extremely productive estuary, which
indicates that ample nutrients are available to support the flora and fauna.
Indeed, a slight degree of organic enrichment may be occurring in Port Susan,
possibly as a consequence of organic materials advected into the bay by the
Stillaguamish River. Much of the watershed of the Stillaguamish River is
agricultural land. However, no evidence of organic enrichment is evident in
any of the data collected during this survey or in the 1986 Everett Harbor
investigation, where Port Susan was also used as a reference area.
i "
A range of grain-size characteristics were exhibited at the Port Susan
stations, from sandy sediments at Station PS-04 (furthest from the Stil-
laguamish River mouth) to fine Sands at Stations PS-02 and PS-03, to mixed
silty-clays at Station PS-01 (closest to the river mouth) (see Appendix D).
162
-------
Other conventional sediment variables (i.e., percent nitrogen, percent total
organic carbon, and percent total solids) also exhibited fairly strong
gradients in relation to distance from the river (see Appendix D). Sediments
at stations sampled in other selected reference areas of Puget Sound are
also predominantly sandy (Table 26). Mean concentrations of total organic
carbon, sulfides, and total solids at the Port Susan stations were similar
to the mean concentrations of those variables among the other reference
areas listed in Table 26, and appear to be typical of unimpacted areas.
Abundances of the major taxonomic groups of benthic invertebrates at the
Port Susan stations were also similar to those in other reference areas
within Puget Sound. Mean total abundances and mean abundances of poly-
chaetes, molluscs, and crustaceans were generally similar to mean abundances
observed in Carr Inlet, in Blakely Harbor, at 15-22 m depth stations in
Central Puget Sound, and at stations in Port Susan sampled in 1986 (Fig-
ure 55).
Comparisons of mean total abundances and mean abundances of the major
taxonomic groups of benthic macroinvertebrates between the 1985 and the 1986
Port Susan stations revealed interannual differences (Figure 55). Mean total
abundances were significantly lower in the samples collected in 1986 than
those collected in 1985. Mean abundances among the major taxonomic groups
appeared to be lower in 1986, but no significant differences (P>0.05) in mean
abundances of the major taxonomic groups were detected between 1986 and 1985
samples. The consistent interannual differences in mean total abundances
document the importance of using simultaneously collected data from reference
and impact areas for impact assessment, as was done in this investigation.
Comparisons of the five most abundant taxa at each of the Port Susan
stations indicate that species composition was fairly similar both within and
between years (Table 27). Two to three species at each station were also
among the abundant taxa at the other stations. This high degree of simi-
larity documents that structurally similar assemblages of benthic macro-
invertebrates were sampled at all four stations in Port Susan in 1985, and
suggests that those assemblages were temporally stable. Comparisons of the
five most abundant species at theY1985 Port Susan stations with those at the
Carr Inlet stations reveals little similarity between these two areas
(Table 27). This is not unexpected, however, because Port Susan and Carr
Inlet are located in different regions of Puget Sound and exhibit different
habitat characteristics (e.g., exposure, freshwater input).
Opportunistic and pollution-tolerant taxa (as defined by Pearson and
Rosenberg 1978) constituted an average 17.4 percent of the fauna at stations
in Port Susan (Table 28). This value was similar to the 21.1 percent of the
Carr Inlet fauna represented by those same taxa. However, Prionospio
steenstrupi and Macoma nasuta constituted most of the opportunistic and
pollution-tolerant organisms in Carr Inlet, whereas Euohi1omedes carcharodon-
ta and Euphilomedes oroducta accounted for nearly all of those organisms in
Port Susan. Euphilomedes spp. are known to increase in abundance only in
areas where organic enrichment of the ecosystem is moderate (Word et al.
1977). If abundances of Euohi1omedes spp. are not considered in the
foregoing calculations, opportunistic and pollution-tolerant organisms would
have constituted less than 5.0 percent of the benthic macroinvertebrates at
all stations in Port Susan.
163
-------
TABLE 26. SURFACE SEDIMENT CHARACTERISTICS AT BENTHIC INFAUNA STATIONS
IN PORT SUSAN COMPARED WITH OTHER REFERENCE AREAS IN PUGET SOUND
Reference Area
Sediment
Type3
Mean TOC
(%)
Mean Sulfide
(mg/L)
Mean Total
Solids (%)
Depth
Range (m)
Port Susan (1985,
this study)
sand-
clayey silts
0.78
22.8
66.4
10-12
Port Susan (1986)^
sand
0.34
Uc
77.4
11-12
Carr Inlet**
sand
0.41
2.3
70.4
2-26
Blakely Harbore
sand
1.65
"15
67.8
10-18
Central Puget^
Sound (Seahurst)
sand
1.51
—
* m
15-22
Samish Bay9
silty sand/
clayey silt
1.65
--
—
10-30
Case InletS
sandy silt
2.2
— ¦
—
21-41
Dabob Bay9
silty sand/
sandy silt
1.88
-- ¦
—
88-113
Sequim Bay9
sandy silt
2.35
—
~
19-26
a Sediment type designations after Shephard (1954).
b Data from PTI and Tetra Tech (1988).
c Undetected at a detection limit of 5 ppm.
d Data from Tetra Tech (1985a).
e Data from Tetra Tech (1986d).
f Data from Word et al. (1984).
9 Data from Battel!e (1986).
164
-------
0)
en
M
E
o
z
CO
_l
<
3
9»
51
O (0
Z W
— 3
U. O
Of
DC "
Ul
m
2
3
Z
<
Ul
2
3
2
2
PORT
SUSAN
<1985)
(4 Stations)
PORT
SUSAN
(1986)
(3 Stations)
BLAKELY CARR SEAHURST
HARBOR INLET (58 Stations)
(2 Stations) (4 Stations)
LEGEND
l MEAN
o z
50
if
, 111
I (/) Q
+1
~ TOTAL ABUNDANCE
POLYCHAETES
MOLLUSCS
CRUSTACEANS
Figure 55. Mean total abundances (no./m2) and mean abundances of major taxonomic
groups of benthic invertebrates in Puget Sound reference areas.
-------
TABLE 27. NUMERICALLY DOMINANT TAXA AT PORT SUSAN
STATIONS SAMPLED IN 1985 AND 1986, AND AT
CARR INLET STATIONS SAMPLED IN 1984
Port Susan 1985 Port Susan 1986 Carr Inlet 1984
Taxon PS1 PS2 PS3 PS4 PS2 PS3 PS4 CR11 CR12 CR13 CR14
Protomedia prudens o
Psephidia Tordi oooo ooo
Terebellides stroemi o o
Euphilomedes producta o o o o o
Lumbrineris spp. o o o
Axinopsida serricata oooo o o
Lumbrineris luti o o
Euphilomedes carcharodonta o o ooo o
Ampharete acutifrons o
CIinocardium nuttali o
Leitoscoloplos ouoetensis o
Pectinaria qranulata o o
Macoma baltica o
Pista spp. o
Leptochelia dubia o
Phvl1ochaetopterus prolifica o
Prionospio steenstrupi ooo
Odostomia spp. o o
Platvnereis bicanaliculata o
Amphioda urtica o
Sealibreqoma inflatum o o
Mitrella gauldi o
Macoma nasuta o
Caprellidae o
Caprella mendax o
Spiophanes berkelvorum o
166
-------
TABLE 28. COMPARISON OF ABUNDANCES (AS PERENCT OF FAUNA)
OF OPPORTUNISTS AND POLLUTION-TOLERANT TAXAa
AT STATIONS IN PORT SUSAN AND CARR INLET®
Stations
Carr
Taxon
PS1
PS2
PS3
PS4
Inlet0
Eteone lonaa (Po)^
0.19
0.53
0.04
0.07
0.05
EuDhilomedes carcharodonta (Os)
0.0
3.80
9.90
16.10
2.97
EuDhilomedes producta (Os)
5.80
3.20
7.60
9.50
0.14
Glvcinde picta (Po)
0.08
0.19
0.0
0.07
0.54
LeitoscoloDlos Duaettensis (Po)
0.0
0.0
2.9
1.20
1.64
Macoma nasuta (Pe)
0.0
0.0
0.0
0.0
3.20
Mediomastus californiensis (Po)
0.08
0.19
0.08
0.24
1.61
NeDhtvs cornuta franciscana (Po)
0.0
0.0
0.0
0.0
0.68
ParaorionosDio Dinnata (Po)
0.04
0.0
0.0
0.0
0.71.
PrionosDio cirrifera (Po)
0.04
1.99
0.16
0.03
0.03
PrionosDio steenstruDi (Po)
0.23
0.30
0.20
0.38
8.63
Tharvx SDp. (Po)
0.70
0.70
0.30
0.20
0.45
Others
0.36
0.72
0.52
0.88
0.46
(No. of taxa)
(5)
(4)
(4)
(6)
(5)
TOTAL
7.52
11.62
21.70
28.67
21.11
a As defined by Word et. al. (1977), Pearson and Rosenberg (1978), and Word
(1980).
b Data from Tetra Tech (1985a).
c Mean value of four stations.
d Po = Polycheata, Os = Ostracoda, Pe = Pelecypoda
167
-------
Overall, the foregoing comparisons of conditions in Port Susan in 1985
with conditions in other reference areas in Puget Sound, and with Port Susan
in 1986, affirm the adequacy of Port Susan as a reference area for benthic
macroinvertebrate communities in Elliott Bay. They demonstrate that sediment
grain-size characteristics, the values of other conventional sediment
variables, and abundances of major taxonomic groups of benthic invertebrates
at the four Port Susan stations were similar to those in other reference
areas. Abundances of opportunistic and pollution-tolerant taxa were found
to be low, and comparable to the Carr Inlet reference area. Finally, the
similarity1 between the numerically dominant taxa collected at the three Port
Susan stations sampled in both 1985 and 1986 indicates that the structure of
the benthic assemblages in Port Susan at a given time of year may be
temporally stable over the long term. In this study, impacts to the benthos
were inferred using statistical criteria to identify changes in the abun-
dances of, the major taxonomic groups of benthic invertebrates. When
comparing 'benthic communities in potentially impacted areas with those in
reference areas, it is often advisable to stratify between-station com-
parisons by habitat characteristics. Such comparisons may be stratified by
sediment grain-size characteristics because sediment grain size is often an
important determinant of benthic community structure (Sanders 1960; Johnson
1971; Gray 1974; Fresi et al. 1983).
As in;Port Susan, sediments in Elliott Bay exhibited a wide variety of
textural characteristics, which ranged from silty-clays to coarse sands.
But in most areas of Elliott. Bay, the sediments have been highly modified by
anthropogenic inputs, and are no longer representative of "natural" condi-
tions. Typically, the sediments smelled of hydrogen sulfide or petroleum,
or both. Wood chips, scrap metal, oil droplets, and other debris were
common in the grab samples. Field notes indicate that at least 73 percent
of the stations sampled in Elliott Bay exhibited evidence of sediment
modification (e.g., hydrogen sulfide, petroleum, foreign objects). Largely
because of these anthropogenic modifications, sediment grain sizes at Port
Susan stations do not cover the entire spectrum of sediment grain sizes in
Elliott Bay, and stratification of the between station statistical compari-
sons would, in many cases, be artificial.
Results of species-level analyses of the benthic macroinvertebrate data
also indicate that stratification of between-station comparisons by sediment
characteristics would be inconsistent because it would not be possible to
match grain size characteristics in many of the tests. Species composition
of the benthic macroinvertebrate assemblages was fairly similar among the
Port Susan stations (Table 27), despite the gradient of sediment grain size
composition that occurred moving away from the river mouth (see Appendix D).
In summary, an a priori examination of the data on sediment characteris-
tics and benthic community structure in Port Susan and Elliott Bay indicated
that:
¦ Total abundances and abundances of the major taxonomic groups
of benthic invertebrates in Port susan were comparable to
abundances in other reference areas in Puget Sound, despite
differences in species composition
168
-------
¦ Species composition among Port Susan stations was temporally
stable between 1985 and 1986
¦ The sediments in Elliott Bay are highly modified in many
cases, to the extent that stratification of statistical tests
based on sediment characteristics would be largely artificial
¦ The range of sediment characteristics in Port Susan was
similar to the range of sediment characteristics in Elliott
Bay, but the data set from Port Susan was not adequate to
stratify by grain size for individual stations.
For these reasons, data on abundances of the major taxonomic groups of
benthic infauna at the four stations in Port Susan were not stratified by
habitat characteristics prior to statistical testing. Instead, the data were
pooled, such that mean values of variables at each station in Elliott Bay
were compared with mean values of variables across all four stations in Port
Susan. Pooling the Port Susan data increases the number of replicate
reference values used in each statistical test from 5 to 20.
Characteristics of Benthic Communities in Elliott Bav and Port Susan
During this study, 364,446 individuals were collected from 78 sampling
stations in Port Susan and Elliott Bay (see Figure 5). Total abundances per
station ranged from 322 to 25,046/m2, and averaged 9,345/m2. This range of
total abundances-is much greater than the range of total abundances in Carr
Inlet (2,767 to 5,946/m2) (Tetra Tech 1985a) and Eagle and Blakely Harbors
(3,134 to 9,406/m2) (Tetra Tech 1986d). However, it is similar to the range
of total abundances that was observed in 50 benthic samples from Commencement
Bay (13 to 33,887/m2) (Tetra Tech 1985a). Total abundances at the Port
Susan stations ranged from 5,040 to 7,232/m2, x=5f799/m2, and were comparable
with those in Carr Inlet and Blakely Harbor (see Figure 55). Hence, total
infaunal abundances at Elliott Bay stations were more variable than those of
Port Susan and other reference areas in Puget Sound. This situation is
illustrated graphically in Figure 56, where the mean number of individuals/m2
collected in each segment of Elliott Bay is plotted. Abundances were
variable among the segments, tending to be higher in the Magnolia, East
Waterway, Kellogg Island, and Duwamish Head/Alki Beach segments than
elsewhere (Figure 56).
Polychaetous annelids were the most abundant mdor taxonomic group
among the Elliott Bay and Port Susan stations (x=5f202/m2), followed by total
crustaceans (x=2,103/m2). [Crustaceans, exclusive of amphipods (x=l,731/m2),
were also the second most abundant major taxonomic group.] Bivalve molluscs
ranked third in mean abundance amona all stations (x=l,524/m2). Amphipods
(x=372/m2) and gastropods (x=280/m2) were ranked fourth and fifth in mean
total abundance among all stations, respectively. Miscellaneous taxa and
echinoderms ranked sixth and seventh in mean abundances and were generally
very minor contributors to community composition at most stations. However,
nematodes and oligochaetes were abundant at some stations in the Elliott Bay
segments.
169
-------
20 -i
CM
E
o
5. ^
(0
-j
<
=> _
S»
Q a
Z » io H
— 3
u. o
Of
(c w
IU
0
2
3 5 H
Z
Z
<
Ul
2
22,724
NS
SS
NM
LEGEND
¦41 MEAN
I
a z
<2
ll
H- Ui
v) a
~i
P S Port Susan (n = 4)
UG Magnolia Blull (n = 4)
Seattle North Waterfront (n = 6)
Seattle South Watertrort In = 1
North Haibor Island (n = 9)
E W East Waterway (n = 15)
WW West Waterway (n = 17}
KG Kellogg Island (n = fi)
AB AIM Bead) (n = 4)
PS MG NS SS NH EW WW KG AB
STUDY AREA
Figure 56. Mean number of individuals/m^ in each study area segment.
-------
Comparisons Between Reference and Potentially Impacted Areas
As discussed earlier, 74 stations in Elliott Bay (divided into eight
segments) and 4 stations in Port Susan (reference area) were sampled for
benthic infauna. Species-level identifications were available for only 20
of the 78 stations. The absence of identification of all organisms to the
lowest possible taxonomic level at all stations precluded in-depth analyses
(e.g., calculation of diversity indices) at all stations because basic
measures of community structure (e.g., numbers of species, dominance) that
are needed for these calculations could not be estimated for 58 of the
78 stations. In addition, information on pollution-sensitive, pollution-
tolerant, and opportunistic taxa also were not available for 58 of the
78 stations. Thus, identification of potentially impacted areas in the
Elliott Bay study area were based on comparisons of abundances of major taxa
groups between stations in Port Susan and stations in Elliott Bay. Species-
level data were used to further characterize and interpret the structure of
the benthic assemblages at those stations where it was available.
As discussed above, total mean abundances among the Port Susan stations
did not vary greatly (i.e., 5,040-7,232/m2). Mean abundances of the major
taxonomic groups (i.e., polychaetes, total crustaceans, amphipods, crusta-
ceans other than amphipods, pelecypods, gastropods, echinoderms, and mis-
cellaneous taxa) also did not vary greatly among the Port Susan stations
(see Appendix E). In contrast, mean abundances of the major taxa and total
infaunal abundances differed greatly among stations in Elliott Bay (see
Appendix E). Statistical analyses of infaunal abundances were conducted
during this study to determine whether any of the differences between
abundances'in Port Susan and at stations in Elliott Bay were significant.
Results of the t-tests are summarized in Figures 57-61 and Appendix E.
Among the 370 paired comparisons that were performed (i.e., 74 stations x 5
taxa), 219 were not statistically significant, 78 indicated enhanced
abundances, and 73 indicated depressed abundances (Figures 57-61). Concen-
trations of toxic substances in sediments have been correlated with reduced
abundances of sensitive taxa (Wolfe et al. 1982; Rygg 1985, 1986) and may
affect all taxa (Bilyard 1987). Extreme organic enrichment may also result
in reduced abundances of infaunal taxa (Pearson and Rosenberg 1978). At
lower levels of organic enrichment, numbers of infaunal organisms may become
moderately to extremely abundant (Pearson and Rosenberg 1978). Thus,
depressed abundances may be indicative of sediments with high levels of
toxic substances, and enhanced abundances may be indicative of organically
enriched conditions.
Results of the paired comparisons are addressed on a segment by segment
basis in Appendix E, beginning with Magnolia (Segment 1) and ending with
Duwamish Head/Alki Beach (Segment 9). The relative degree of impact at each
test station was estimated by ranking stations according to the number of
significant depressions in the abundance of the following major taxonomic
groups: polychaetes, crustaceans, pelecypods, and gastropods (see Indices
for Decision Criteria, for discussions of the rationale).
171
-------
•AB-03O
• SIGNIFICANT DEPRESSIONS
* SIGNIFICANT ENHANCEMENTS
Figure 57. Summary of significant (P<0.001) enhancements and
depressions of polychaete abundances among the
Elliott Bay stations compared with reference conditions.
172
-------
¦ NH43 .
¦ WW-20
¦NH-04 N
¦ NH05- .
3000
• SIGNIFICANT DEPRESSIONS
* SIGNIFICANT ENHANCEMENTS
tooo
2000
Figure 58. Summary of significant (P<0.001) enhancements and
depressions of crustacean abundances among the
Elliott Bay stations compared with reference conditions.
173
-------
¦ KOOI
¦ SIGNIFICANT DEPRESSIONS
« SIGNIFICANT ENHANCEMENTS
2000
Figure 59. Summary of significant (P<0.001) enhancements and
depressions of pelecypod abundances among the
Elliott Bay stations compared with reference conditions.
174
-------
3000
6000
¦ SIGNIFICANT DEPRESSIONS
» SIGNIFICANT ENHANCEMENTS
1000
Figure 60. Summary of significant (P<0.001) enhancements and
depressions of gastropod abundances among the
Elliott Bay stations compared with reference conditions.
175
-------
J?
AAB-04*
•EVWS
AKO-11
ii
6000
Sleet
J meter*
2000
SIGNIFICANT DEPRESSIONS
SIGNIFICANT ENHANCEMENTS
Figure 61. Summary of significant (P<0.001) enhancements and
depressions ol total macrofaunal abundances among the
Elliott Bay stations compared with reference conditions.
176
-------
Species-Level Comparisons Among Elliott Bay and Port Susan Stations
As discussed above, benthic macroinvertebrates at only 20 of the
78 stations were identified to the lowest possible taxonomic level. Those
20 stations included 4 stations from the Port Susan reference area and
16 stations from Elliott Bay. The 16 Elliott Bay stations were selected
based on observed high mortality in the amphipod bioassay tests and by the
proximity of stations to known sources of contamination among the nine
segments.
Numbers of Taxa--
Mean numbers of species (in some cases higher taxa) per O.l-m^ grab
sample varied considerably among the 20 stations (Figure 62). In Port
Susan, the mean number of species per grab varied from 44 to 60 and averaged
52.5 species among stations. In Elliott Bay, the mean number of species per
station varied from 7 to 80. The highest mean numbers of species occurred
at Stations AB-01, NS-03, and NH-02 (80, 65, and 61, respectively), where
zero (AB-01 and NS-03) or one (NH-02) significant depression in major taxa
abundance (relative to reference conditions) was detected. The lowest mean
numbers of species occurred at Stations NH-03, EW-05, NS-08, and KG-01 (7,
11, 22, and 25, respectively), where two or more significant depressions in
the abundances of major taxa (relative to reference conditions) were
detected. Mean numbers of species at the remaining 13 stations were
relatively similar to those in Port Susan, ranging from 32 to 59 species per
grab sample (Figure 62). One or no significant depressions (P<0.05) in
major taxa abundances occurred at these stations, except at Station. NH-08
where two significant depressions .occurred. In general, the mean numbers of
species among these stations varied in a manner similar to total abundances
and abundances of the major taxonomic groups (Figure 63; see Appendix E),
Mean numbers of species tend to indicate similar potential problem stations
as those identified by significant depressions of major taxonomic groups.
Station NH-08 is an exception to this trend because two depressions were
detected, yet, the mean number of species (41) is only slightly lower than
the mean number of species at Port Susan stations.
Numerically Abundant Taxa—
Absolute and relative abundances of the five most abundant taxa (i.e.,
the numerically dominant taxa) at each station are summarized in Figures 64
and 65 and in Table 29. Among the Port Susan stations, abundances and
relative abundances of the dominant taxa were fairly consistent (3,448-
4,154/m2 and 57.5-69.8 percent of the fauna). Species composition of the
dominant taxa was also similar among these stations. Each station had at
least two, and as many as four taxa in common with each of the other
stations. No single taxon dominated the community to the exclusion or near
exclusion of other taxa at any of these stations. The highest relative
abundance was exhibited by Psephidia lordi at Station PS-03 where it
represented 39.9 percent of the fauna.
Among the Elliott Bay stations, relative abundances of the dominant taxa
ranged from 43 to 94 percent of the total infaunal abundance. At all but
Stations NS-03, SS-11, NH-01, NH-02, and AB-01 (where only one or zero
significant depressions were detected), the five numerically abundant taxa
177
-------
REFERENCE
AREA
PS-2 PS-4 NS-8 SS-11 ' NH-2 NH^4 EW-5 WW-9 WW-14 AB-1
STATIONS
NOTE: Numbers at top* of bars indicate numbers ol significant depressions detected in major taxa abundances relative to the Port Susan reference area
Figure 62. Mean number of benthic species per station collected from the 20 Elliott Bay
and Port Susan stations where complete identification of benthic samples
was performed.
-------
CM
E
c> «T
z -o
w c
as
(0
LU
y 3
5 °
< JZ
Q I-
Z w
3
m
<
24
22 -
20 -
18
16
14
12 -
10 -
8 -
6 -
4 -
2 -
0
REFERENCE
AREA
T
PS-1
r
/
/
/
~
/
/
~
/
/
~
/
/
/
/
/
~
T
PS-3
_L
NS-3
SS-4
4
-T-
NH-3
PS-2
NS-8
t
NH-8 I EW-11 | WW-11 | KG-1
EW-5 WW-9 WW-14 AB-1
SS-11 NH2 NH-4
STATIONS
NOTE: Numbers at lope Of bars indicate numbers o( significant depressions detected in major taxa abundances relative to the Port Susan reference area.
Figure 63. Mean abundance (no./m2) of total benthic infauna per station collected from the
20 Elliott Bay and Port Susan stations where complete identification of
benthic samples was performed.
-------
NOTES:
• Numbers inside bare refer to species listed in Table GMB 6.
* Numters at tops ol bars indicate numbers of significant
depressions detected in major tax a abundances relative
to the Port Susan reference area.
10 000 -I
- REFERENCE -
AREA
N
E
CO
<
3
Q
>
Q
8000
= 6000 -
O 4000
DC
111
DO
s
3 2000 -
PS-1 | PS-3 | NS-3
PS-2 PS-4
13
12
10
13
16
NS-8
16
1
14
11
8
22
6 —
?1
12-i
10-1
13
eM
I
1
NH-3
2 v24
1^-21
14
23
25
13
18.046
13.79S
10
27—1
28
fl
SS-4 | NH-1 | NH-3 | NH-8
SS-11 NH-2 NH-4
STATIONS
14
13
13.676
11.698
6284
1
16
13
EW
EW-5
29
13
16
16.350
18.574
1
27
20
23
WW-11
27
13
11,476
9054
19
KG-1
WW-9 WW-14
AB-1
Figure 64. Abundances (no./m2) of the five most numerically dominant species at
completely identified stations in Elliott Bay and Port Susan.
-------
UJ
o
DC
UJ
a.
100-1
3
<
80-
60-
40-
20 -
NOTES:
• Numbers inside bars relar to species listed in Table GMB 6.
• Numbers al tops of bars indicate numbers ol significant
depressions detected in major taxa abundances relative
to the Port Susan reference area
-REFERENCE -
AREA
11
10
10
IS
14
13
12
13
16
18
16
13
19
16
10
12^
6-
14
21
13
22
14
14
23
-J
25
13
13
10
28
27
14
13
-10
29
21
16
13
11
29
13
16
29
30
27
20
23
10
13
PS^I I PS-3 I NS-3 I SS-4 I NH 1 | NH-3 | NH-8 | EW-11 | WW-11 | KG-1
PS-2 PS^J NS-8 SS-11 NH-2 NH-4 EW-5 WW-9 WW-14 AB-1
STATIONS
Figure 65. Relative abundances of the numerically dominant species at completely
identified stations in Elliott Bay and Port Susan and the proportions of total
infaunal abundance for which they account.
-------
TABLE 29. KEY FOR FIGURES 63 AND 64
1. Protomedia prudens
2. Psephidia lordi
3. Terebellides stroemi
4. Euphilomedes producta
5. Lumbrineris spp.
6. Axinopsida serricata
7. Lumbrineris luti
8. Euphilomedes carchardonta
9. Ampharete acutifrons
10. Maconia carlottensis
11. Odostomia spp.
12. Nematoda
13. Tharvx spp.
14. Capitella capitata
15. Platvlnereis bicanaculata
16. Leptochelia lourei
17. Exooone dubia
18. Heterophoxus oculatus
19. Notomastus tenuis
20. Mediomastus taliforniensis
21. Prionospio steenstrupi
22. Cirratulis soectabilis
23. Cirratulis cirratus
24. Caulleriella hamata
25. Lumbrineris cruzensis
26. Polvcirrus spp.
27. Euchone 1imnicola
28. Spiochaetopterus costarum
29. Photis brevipes
30. Corophium acherusicum
182
-------
represented over 70 percent of the total fauna per station (Figure 65). High
relative abundances of dominant taxa often indicate stressed conditions. As
less tolerant species are eliminated from the habitat, opportunistic species
fill the vacant niches, often achieving high abundances (Pearson and
Rosenberg 1978; Gray 1982). Abumlances of the five numerically dominant
taxa averaged 3,659 individuals/m2 among the Port Susan stations. Most
abundances of the numerically dominant taxa at Elliott Bay stations were
above this average, and Stations EW-11, KG-01, WW-11, and WW-14 greatly
exceeded this value (Figure 64; Table 29). Abundances of polychaetes and
total infauna at these stations exhibited significant enhancements. In
this case, species-specific data indicates that Stations EW-11, WW-11, and
WW-14 may be more stressed (in terms of organic enrichment) than indicated
by t-tests performed at the major taxa level. Among the major taxa groups
tested at these stations only one significant depression was detected.
Abundances at Stations EW-05 and NH-03 were greatly depressed, below the Port
Susan value. Greatly depressed abundances often indicate excessively
enriched sediments or the presence of contaminants (Pearson and Rosenberg
1978; Carriker et al. 1982; Wolfe et al. 1982; Dillon 1984). The highest
number of significant depressions detected among the stations identified to
the species level, occurred at Stations EW-05 and NH-03, where three and four
depressions were detected respectively.
Taxonomic composition of the dominant species differed considerably
within Elliott Bay and between Elliott Bay and Port Susan (Figure 64;
Table 29). For example, Tharvx sp. was among the dominant taxa at 11 of the
16 Elliott Bay stations but was not among the dominant species in Port
Susan. This surface deposit-feeding polychaete (Fauchauld and Oumars 1979)
dominated many of the benthic communities in contaminated Commencement Bay
waterways, and may be indicative of stressed conditions (Tetra Tech 1985a).
Other members of the genus have been identified as opportunistic (Dorsey et
al. 1983; Raman and Ganapati 1983). Subsurface deposit-feeding nematodes
and the polychaetes Caoitella caoitata and Mediomastus califomiensis were
among the numerically dominant taxa at two, three, and five stations, re-
spectively, in Elliott Bay. Capitella caoitata was the numerically dominant
species at Stations EW-05 and NH-03, and nematodes were the dominant taxon
at Station NS-08. These stations exhibited the greatest number of depres-
sions in abundances. The molluscs Macoma carlottensis and Axinoosida
serri cata. the ostracod Euohilomedes carcharodonta. and the tanaid Lep-
tochelia dubia were also abundant at five or more stations in Elliott Bay.
Among the foregoing seven taxa, only £. carcharodonta and serricata were
among the numerically dominant taxa at stations in Port Susan. Nematodes
and polychaetes are known to reach very high abundances in organically
enriched sediments, often to the exclusion of other taxa (Nichols 1977;
Pearson and Rosenberg 1978; Van Es et al. 1980). The dominance of these
taxa and the low total abundances observed at Stations EW-05 and NH-03
(Figures 64 and 65; Table 29) suggest that the sediments at these stations
are highly organically enriched, contaminated with toxic substances, or both.
Relative abundances of opportunistic and pollution-tolerant taxa (as
defined by Word et al. 1977; Pearson and Rosenberg 1978; Word 1980) at the 16
Elliott Bay stations and the 4 Port Susan stations further supports the
conclusion that benthic infaunal communities at most Elliott Bay stations are
stressed (Figure 66; Table 30). Total relative abundances of opportunistic
and pollution-tolerant taxa among Port Susan stations averaged 17.2 percent
183
-------
CO
-f*
<
Z
D
<
U.
U.
O
H
Z
Ui
o
oc
UJ
Q.
90
80 H
70
60
50-
40-
30 -
20 -
10 -
0
REFERENCE .
AREA
I
PI
~
~
/
~
/
~
/
/
/
~
/
I PS-3 I
PS-2 PS-4
NS-3 | SS-4 | NH-1 | NH-3 | NH-8 | EW-11 | WW-11 | KG-1
NS-8 SS-11 NH 2 NH-4 EW-5 WW-9 WW-14 AB-1
STATIONS
NOTE: Numbers at lops ol barn indicate numbers of significant depressions detected In major taxa abundances relative to the Port Susan reference area.
Figure 66. Abundances (as percent of fauna) of opportunistic ?md pollution-tolerant taxa
(as defined by Word et al., 1977; Pearson and Rosenburg, 1978; and
Word, 1980) at completely identified stations in Elliott Bay and Port Susan.
-------
TABLE 30. ABUNDANCES (AS PERCENT OF FAUNA) OF OPPORTUNISTIC AND
POLLUTION-TOLERANT TAXAa AT STATIONS IN PORT SUSAN AND ELLIOTT BAY
oo
cn
¦ -
...
—
¦
¦
Stations
-=: ¦
Taxun
PS1
PS2
PS3
PS4
NS3
NS8
SS4
SS11
Nil 1
Nil?
NII3
NH4
NH8
EW5
EW1I
WW9
WWII
WWI4
KG 1
AB1
Cauitel la caoitata (Poj'>
<0.1
<0.1
13.6
0.6
0.1
3.3
5.5
78.3
0.8
1.2
58.5
0.4
1.7
0.7
0.8
0.8
0.4
Coroohium asherusicum (Am)
0.1
<0 1
<0. 1
0 5
1.2
0.1
<0.1
0.1
0.2
1.3
0.2
Ftgone lonqa (Po)
0.2
0.5
<0.1
0.1
0.1
0.1
0.4
0.1
0.5
0.7
0.1
1.3
0.4
<0.1
0.1
<0.1
0.1
0.2
EuohUomedes carcharodonta (Os)
3.8
9.9
16.1
20.6
0.3
2.5
12.4
6.1
8.8
0.6
0.3
2.0
0.6
1.0
0.4
0.4
0.1
0.3
13.9
typh i]omedes product a (Os)
5 8
3.2
7.6
9.5
6.2
0.1
0.2
0.6
0.4
1.5
0.1
0.1
0.2
<0.1
0.4
Glvcinda oicta (Po)
0.1
0.2
0.1
0.1
0.2
0.1
0.4
0.6
<0.1
0.1
0.1
<0.1
<0.1
0.3
Leitoscoplos ouaettensis (Po)
2.9
1.2
0.4
0.2
0.5
Mediomastus californiensis (Po)
0.1
0.2
0.1
0.2
0.3
0.3
0.6
8.6
5.3
5.7
1.0
0.1
1.6
0.7
6.1
<0.1
5.3
Oligochaeta (01)
0.3
<0.1
0.3
4 4
0.2
ParaDrionoSDio oinnata (Po)
<0.1
<0.1
<0.1
0 1
0 1
1.2
0.2
<0.1
<0.1
<0.1
Prionosoio cirrifera (Pol
<0.1
2.0
0.2
<0.1
0.1
0.3
0.7
1.4
4.7
4 0
0.1
<0.1
0.2
2.9
0.9
0.3
<0.1
2.6
Prtonospio steenstruui (Po)
0.2
0.3
0.2
0.4
1.6
0.1
1.1
4.3
5.0
9.6
1.4
0.4
0.2
0.5
8.7
1.8
0.6
<0.1
3.0
Tharvx sdd. (Po)
0.7
0.7
0.3
0.3
1.3
30.7
15.0
14.1
13
13.1
45.9
5.4
69.5
36.6
33.7
2.0
74.2
3.4
Others
0.3
0.4
0.5
0.8
0.3
0.2
0.1
0.3
0 3
0.1
0.5
0.2
0.3
0.2
0.1
0.1
1.0
(No. of taxa)
(4)
(3)
(4)
(6)
(4)
(2)
(4)
(6)
(2)
(5)
(4)
(4)
(4)
(2)
(3)
(6)
Iota)
7.5
11.6
21.7
28.7
31.0
45.5
21.1
34.3
30.6
54.4
81.3
8.7
52.5
65.3
72.1
52.5
38.6
11.5
75.5
31.2
a As defined by Word et. al. (1977), Pearson and Rosenberg (1978), and Word (1980).
b Po = Polychaeta, Am = Atnphipoda, Os - Ostracoda, 01 » Oligorchaeta.
-------
af the total fauna. The ostracods Euphilomedes carcharodonta and £. producta
constituted the largest proportion of the opportunistic and pollution-
tolerant taxa among these stations. Euphilomedes spp. are known to increase
in abundance in areas where only slight organic enrichment of the ecosystem
may be occurring. Among the Elliott Bay stations, 13 of the 16 stations had
relative abundances of opportunistic and pollution-tolerant taxa higher than
any of the Port Susan stations (Figure 66; Table 30), ranging from 30.7 to
81.4 percent of the total fauna. Only at Stations SS-04, NH-04, and WW-14
did relative abundances of opportunistic and pollution-tolerant taxa* not
exceed the relative abundances of those taxa observed at the Port Susan
stations. Stations NH-03, EW-05, EW-11, and KG-01 had the highest relative
abundances] of opportunistic and pollution-tolerant taxa, ranging from 65.3
to 81.4 percent. At the major taxa level, these four stations exhibited the
greatest number of significant depressions. Capitella capitata constituted
the largest proportion of opportunistic and pollution-tolerant taxa at
Stations NH-03 and EW-05, while Tharvx sp. constituted the largest proportion
at Stations EW-11 and KG-01. C. capitata is a known indicator of organically
enriched conditions and typically is the only species found in the most
stressed areas. Tharvx sp. has also been cited as an indicator of organical-
ly enriched conditions. The remaining stations had relative abundances of
opportunistic and pollution-tolerant taxa ranging from 30.7 to 54.4 percent;
Tharvx spi also represented a large proportion of the opportunistic and
pollution-tolerant taxa at these stations (Figure 66; Table 30).
Classification Analyses
Similarities between station pairs and groups of stations based on the
Bray-Curtis Similarity Index and normal classification analysis are shown in
Figure 67. Station groups were determined by selecting a 55 percent
similarity value. This level of similarity was selected because it separated
the stations into three interpretable groups with four outliers. Results of
the normal classification analysis indicated a high degree of similarity
among the four Port Susan stations (65 percent). Much lower degrees of
similarity1 were apparent between the four stations in Port Susan and.the
16 stations in Elliott Bay (25 percent). These results suggest that despite
the gradient in sediment grain size composition that exists among the four
stations in Port Susan, the benthic macroinvertebrate assemblages were very
similar. They also indicate that species composition of the benthic
communities in Port Susan differs somewhat from that in Elliott Bay.
Reasons for the apparent differences in species composition cannot be
confirmed, but likely include habitat and biogeographic differences between
the two aifeas, effects of anthropogenic modifications of Elliott Bay, and
stresses resulting from inputs of pollutants from point and nonpoint sources
into Elliott Bay.
Included in Group 1 were all four Port Susan stations. Stations PS-03
and PS-04 were very similar to one another, probably due to the similar
abundances of Pseohidia lordi. Euphilomedes carcharodonta. and E. producta
that occurred at each. These taxa were the three most abundant species.
They occurred in the same rank order at both stations. P. lordi was common
to all four stations, and Axinopsida serricata and E. producta were common
to three jof the stations. Sediments at the Port Susan stations were
primarily fine sands, but spanned a range from medium sands to silt-clays.
TOC was low among the Port Susan stations, ranging from 0.39 to 1.49 percent.
186
-------
SIMILARITY (Percent)
too
90
ao
70
60
50
| i i i i | i i i i | i i r i | i i i i j i i i i j
40
30
20
10
<
H
CO
-2
WW 9
WW-11
WW-14
NH 1
NH-2
SS-4
NS-3
SS11
W AB 1
Z
O EW
KG-1 h 3
NH-a
NH-4 - O
N&a - o
PS-3
PS4
PS-1
PS-2
EVW-5 - O
NH-3 - O
"i
:r
-i
i ii 11111 111 11 111 11 11 ii 11
1111 11
NOTE: Stationgroups deBneaiad by a similarity level of 49 percent are enumerated. Outliers are designated as" O"
Figure 67. Results of a Q-mode classification analysis (Bray-Curtis similarity index, group
average clustering strategy) using log-transformed [log10(x+1)] abundances of the
benthic infauna at the completely identified stations in Elliott Bay and Port Susan.
-------
Group 2 comprised nine stations (i.e., Stations AB-01, NS-03, SS-04,
SS-11, NH-Ql, NH-02, WW-09, WW-11, and WW-14) from five of the nine segments.
Within this group, the West Waterway stations, the North Harbor Island
stations, and Stations AB-01, NS-03, and SS-11 formed groups at higher
similarity levels. West Waterway Stations WW-09 and WW-11 were the most
similar to each other because of the shared dominance of Tharvx sp. and
Leotochelia dubia. In addition to these two species, they also had four of
the five most abundant taxa in common. Leptochelia dubia and Tharvx sp.
were the two most abundant taxa at Station SS-04.
Within Group 2, Stations AB-01, NS-03, SS-11, NH-01, and NH-02 exhibited
a dominance structure similar to that at the Port Susan stations. The most
abundant taxa represented 43-63 percent of the fauna, and no individual
taxon was highly dominant at any of the stations (see Figures 64 and 65;
Table 29). | As discussed above, numerically dominant taxa at the Port Susan
stations represented 57.5-69.8 percent of the fauna, and were not dominated
by a single! taxon. The numerically abundant species were also fairly similar
among Stations AB-01, NS-03, SS-11, NH-01, and NH-02. £. carcharodonta was
among the dominant taxa at all five stations, while A. serricata and
Qdostomia spp. were common to three of the five stations.
! • •
The nine stations in Group 2 had few significant depressions. Among the
stations in this group, no significant depressions were detected at Stations
AB-01, NS-03, and SS-11 and only one significant depression was detected at
the remaining stations when compared to the Port Susan stations. All
depressions were attributed to low pelecypod abundances. No consistent
trends in sediment characteristics were evident among the stations in
Group 2. Grain size varied from medium sands to si 1ty sand, and TOC varied
from 0.49 to 6.83 percent.
i
Group 3 consisted of Stations EW-11, KG-01, and NH-08. All three
stations were characterized by extremely high abundances of the polychaete
Tharvx sp. Tharvx sp. represented 69.5, 74.2, and 45.9 percent of the total
fauna at Stations EW-11, KG-01, and NH-08, respectively. These high
abundances were reflected in the t-tests for Stations EW-11 and KG-01, which
showed significant enhancements of polychaetes and total infaunal abundances.
Crustacean abundances were significantly depressed at all three stations
(see Figure 58), and Stations KG-01 and NH-08 exhibited significantly
depressed abundances of pelecypods. Sediment characteristics were similar
among these stations. Stations EW-11 and KG-01 exhibited similar TOC values
(3.43 and 3.13 percent, respectively) and total solids values (47.23 and
43.58 percent, respectively). Sediments at Station NH-08 differed slightly,
having 1.96 percent TOC and 57.63 percent total solids. All three stations
had relatively high percentages (37.4-79.4 percent) of silt (see Appendix D).
Stations NH-04 and NS-08 were both outliers in the classification
analysis. Station NH-04 was an outlier because Cirratulus cirratus was
extremely dominant, representing 84.5 percent of the fauna (see Figures 64,
65 and Table 29). Although sediment characteristics (i.e., percent TOC and
percent total solids) were similar to most other test stations, polychaete
abundances were significantly enhanced and pelecypod abundances were
significantly depressed at this station (see Figures 57 and 59). Station
NS-08 was located at the north end of the Navy docks, and was co-dominated by
188
-------
nematodes and Tharvx sp. at relatively low abundances. The sediments at
Station NS-08 were characterized as silts, and sulfide concentrations were
elevated (i.e., 167 mg/L). High abundances of nematodes are indicative of
highly stressed environments (Nichols 1977; Van Es et al. 1980; Bouwman
et al. 1984; Lambshead 1986).
Stations EW-05 and NH-03 had the greatest number of significant
differences compared with the reference area and were also outliers in the
classification analysis. Both exhibited very low total abundances, and had
significant depressions in pelecypod; gastropod, and crustacean abundances,
and in total infaunal abundances (see Figures 57-61). Both stations were
dominated by the polychaete, Capitella capitata. which represented 58.5
percent and 78.3 percent of the fauna at Stations EW-05 and NH-03, respec-
tively. C. capitata is an opportunistic species that is a known indicator of
highly stressed conditions (Pearson and Rosenberg 1978).
Comparison of Species and Maior Taxa Level Analyses
Species data at the 20 stations in Port Susan and Elliott Bay provide
detailed information about benthic communities at particular sites. In
general, the species data tend to identify similar patterns of potential
problem areas as the significance tests using only major taxa abundances.
For example, mean numbers of taxa per station (see Figure 62) were very low
at stations where two, three, or four significant depressions were detected
(i.e., Stations NS-08, NH-03, NH-04, EW-05, and KG-01). Mean numbers of
taxa were slightly lower than reference stations at stations where only one
or two depressions were detected, and were equal to or higher than reference
conditions at stations where no significant depressions were detected at the
major taxa level. Thus, the reductions in the mean number of species
generally coincided with depressed abundances of major taxa and were not
random occurrences.
Species-level data also provides important information beyond that
provided by the major taxa. A major application of species data is for
interpreting significant differences in abundances at the major taxa level.
For example, polychaete abundances were enhanced at Stations EW-11, WW-09,
WW-11, WW-14, NH-04, and KG-01. Only one significant depression was
detected at Stations EW-11, WW-09, WW-11, and WW-14. Using the major taxa
approach, these stations are not considered to be highly stressed. However,
species data (e.g., numerically dominant taxa, dominance structure,
opportunistic and pollution-tolerant taxa) at these stations indicate that
the high abundances are due to the high abundances of only a few taxa (i.e.,
Tharvx sp., Leotochelia dubia. Cirratulus cirratus). most of which are known
to be indicative of organically enriched conditions. Thus, these stations
are potentially more stressed than indicated by the significance test of
major taxa abundances.
Abundances of polychaetes and total infauna at stations in the East
Waterway (e.g., Stations EW-03 to EW-12) were significantly enhanced. In
contrast, Station EW-05, located very near the Hanford CSO, exhibited very
low numbers of individuals. The high abundances at nearby stations and the
low abundances at Station EW-05 seemed to indicate a perfect example of the
organic enrichment model as proposed by Pearson and Rosenberg (1978).
Examination of the species data at Station EW-05 indicated that, although the
total abundance was low, this station was dominated by Capitella capitata. a
189
-------
known opportunistic species. Without species data for Stations EW-05 to
EW-11, the; possible effects of organic enrichment would not be apparent.
Thus, it is possible that extreme organic enrichment of the sediments at
Station EW-05 is also contributing to depressed abundances of benthic
infauna. Species data from nearby stations are needed to confirm this
hypothesis.
Among the 16 Elliott Bay stations for which species data are available,
one or more significant depressions in abundance at the major taxa level
were detected at 13 stations. Of the 13 stations with one or more depres-
sions, abundances at all but two stations (Stations NH-01 and NH-02) were
dominated by one or two species (see Figures 64 and 65). At each of these
11 stations, the percentage of the fauna represented by the five numerically
dominant tixa exceeded the highest value of the Port Susan stations (i.e.,
69.8 percent of the fauna). At Stations NH-03 and EW-05 (where four and
three depressions, respectively, were detected) only one species, the
opportunistic polychaete Caoitel1 a capitata. was dominant. Stations NH-04,
NH-08, and !KG-01, where two depressions were detected, and Stations SS-04,
EW-11, and WW-14, where only one depression was detected, were also dominated
by one taxa. Tharvx sp. (Stations NH-08, EW-11, and KG-01), Cirratulus
cirratus (Stations NH-04 and WW-14), and Leotochelia dubia (Station SS-04)
were the dominant taxa at these stations. Five of the seven stations (71
percent) wh'ere only one significant depression was detected were dominated
by one or two taxa and had a dominance structure that suggests a stressed
condition, i Thus, although only one significant depression was detected at
the major taxa level, species level indicators suggest a somewhat higher
stressed condition. One. significant, depression was detected at 25 of the
74 stations sampled in Elliott Bay. One depression at stations without
species data may or may not be important but cannot be evaluated. Where no
significant depressions occur (i.e., Stations NS-03, SS-11, AB-01), species-
level indicators provided additional information about the benthic community
but did not!change the conclusions reached by comparing major taxa abundan-
ces.
Relative abundances of opportunistic and pollution-tolerant taxa were
more variable than relative abundances of numerically dominant taxa (see
Figure 66). In general, relative abundances of opportunistic and pollution-
tolerant taxa at stations where two or more significant depressions were
detected, represented a higher percentage of the total fauna than any Port
Susan station. Of the seven stations where one depression was detected, five
had relative abundances of opportunistic and pollution-tolerant taxa that
were greater than Port Susan stations. As with the numerically dominant
taxa, relative abundances of opportunistic and pollution-tolerant taxa
indicate that a more stressed condition may be occurring at stations where
only one significant depression was detected.
Indices for Decision Criteria
Concentrations of toxic substances in the sediments may cause reductions
in the abundances of sensitive taxa (Wolfe et al. 1982; Rygg 1985, 1!J86).
Similarly, a high degree of organic enrichment may also result in the demise
of infaunal species (Pearson and Rosenberg 1978), In this study, the
locations ana magnitudes of impacts were determined by the results of the
statistical tests discussed earlier (see Figures 57-61). Where available,
190
-------
species level data were used to infer whether the presence of organic
materials, toxic substances, or both were causing the observed depressions.
The relative degree of impact at each test station was estimated
by ranking stations according to the number of statistically significant
depressions in the abundances of the following major taxonomic groups:
Polychaeta, Crustacea, Pelecypoda, and Gastropoda. The major taxonomic
groups comprised many different species with varying degrees of sensitivities
to organic enrichment and toxic chemicals. Most pelecypods and crustaceans
appear to be fairly sensitive to organic enrichment and toxic substances and
tend to exhibit reduced abundances under these conditions. Because of this
sensitivity, they tend to be good indicators of stressed conditions at the
major taxonomic level. Some species of polychaetes are sensitive to organic
enrichment and toxic chemicals, while some species are tolerant of organic
enrichment, or respond positively to low to moderate levels of organic
materials. Some species of polychaetes may also be relatively tolerant of
low levels of toxic substances. Hence, different species of polychaetes may
exhibit either reduced or enhanced abundances in the presence of organic
materials and toxic substances. Because of this variability among species,
polychaetes do not tend to be very sensitive indicators of toxic substances
at the major taxonomic level. Total infaunal abundance was not used in the
criteria because the patterns of significance virtually mirrored the
polychaetes. Because abundances of polychaetes were often enhanced, and
often represented more than 50 percent of the fauna, the depressions in
abundances of the other taxa were masked by these abundances at the total
infaunal level. Although individual statistical comparisons were performed
for amphipods and the other crustaceans, the results of tests on total
crustaceans were very similar to both of these, taxa.
Using the foregoing statistical criterion, Stations NH-03 and WW-03
appeared to be the most impacted: four depressions were recorded at each
(Figure 68). Abundances of all major taxa at these two stations were
depressed 33.3-94.5 percent compared with abundances at Port Susan. Poly-
chaetes were depressed the least at both stations (37.3 and 33.3 percent,
respectively). The moderately depressed abundances of polychaetes at these
stations probably reflect the variable, species-specific responses of
polychaetes to pollutant stresses. Many polychaete species are enhanced
due to organic enrichment, while other species are reduced in the presence
of organic materials and toxic substances. The polychaete Capitella
capitata dominated the benthic assemblages at Station NH-03. As discussed
above, assemblages dominated by £. capitata are indicative of highly stressed
conditions. No species-specific information was available at Station WW-03.
Gastropods exhibited the most severely depressed abundances among the major
taxa (94.5 and 91.3 percent), although abundances of pelecypods (87.3 and
59.7 percent) and crustaceans (76.6 and 64.8 percent) were also severely
depressed at both stations.
Stations EW-05, WW-20, KG-05, and KG-06 also appeared to be highly
impacted; three significant depressions were recorded at each of these
stations. Abundances of pelecypods, gastropods, and crustaceans at these
stations were depressed by 32.5-94.5 percent when compared with Port Susan.
Pelecypods and gastropods were completely absent from the samples at Station
KG-05, and were also considered depressed. Abundances of polychaetes were
191
-------
MG-01
NS-07
CHmG-04
NS48B
NS-06B2
ss^c
SS-03B—I
SS-07
$&Oft
1 1AE&3 J
i"lNW01
r"iwiog-
^>NH03
HDww-ao—» \
\ "
[ t-J WWM7—* \
IAS-02 \\
AB-01 A* \\\
NH-10 \ \s
NHO9 ^ •
Cnms
NH07
M^vm.\%
WW-13
gnwv».tQ—'
r~iwvM».
Ewwmjs
WW4S
I lww•
EW-1Q
Ew-11 n
I EWOfllI
ew-oa
EW-06 ffj
EW-04
€W>03
EvwaCJ
JKQ4S-
-K&oaMH
-KO07
-KO04
-K04$n
-K&03C3
:K&oa~^
-KO01
-on-ie
-OR-18
-0M2
- 0M1
DM0-
OFV09-
«
6000
aiMt
I meters
2000
. 1 SGNIRCANT DEPRESSION
. 2 SIGNIFICANT DEPRESSIONS
. 3 SIGNIFICANT DEPRESSIONS
• 4 SIGNIFICANT DEPRESSIONS
DR06-
0MS-
Figure 68. Summary of spatial patterns of significant (P<0.001)
benthic depressions among the Elliott Bay stations.
192
-------
not significantly depressed among any of these four stations, but Capitell a
capitata dominated the assemblage at Station EW-05.
An intermediate level of impact (two depressions detected at each
station) was indicated at Stations NS-06, NS-08, SS-09, NH-04, NH-05, NH-08,
WW-08, WW-10, WW-12, WW-16, WW-19, KG-01, and KG-08. Four of the five most
abundant taxa at Station NS-08 were opportunistic or pollution-tolerant.
Nematodes and Tharvx sp. co-dominated the assemblage at this station.
Station NH-04 was dominated by Cirratulus cirratus. also an indicator of
stressed conditions. Tharvx sp. also was the dominant species at Stations
NH-08 and KG-01. No species-specific information was available for the other
stations. One significant depression was detected at 21 other stations.
When the number of benthic depressions per station was examined by
segment, numbers of depressions were highest in those segments closest to the
mouth of the Duwamish River. Stations within the Magnolia and Duwamish
Head/Alki Beach segments were virtually unimpacted, only one depression per
segment was detected, and these depressions were most likely attributable to
low variability among the replicate samples. The benthic communities within
both these segments were similar to each other, but the community composition
appeared to be different from that of Port Susan. Benthic communities
within both these segments tended to have much higher abundances of crus-
taceans (many of which are typically pollution sensitive) than did the Port
Susan stations.
With five exceptions, depressed abundances of benthic infauna were not
detected at stations within the Seattle Waterfront North and South segments.
Significant depressions were only detected at Seattle Waterfront North.
Abundances of pelecypods and crustaceans were depressed at both stations.
Pelecypods and crustaceans exhibited depressed abundances at Station SS-09.
Depressions were detected at only two other stations (Stations SS-04 and
SS-08) and only pelecypods abundances were depressed. No other stations
within the Seattle Waterfront South segment had any significant depressions.
Seven significant depressions were detected among 4 of the 15 stations
in East Waterway. One depression was detected at Stations EW-02, EW-06,
EW-09, and EW-16, and three depressions were detected at Station EW-05.
Polychaetes did not account for any of the depressions within this segment,
but they did exhibit significantly enhanced abundances at eight of the
15 stations, indicating that this area is probably organically enriched.
Polychaete abundances at Station EW-05 were low compared with Port Susan,
and very low compared with the other stations in East Waterway. The
pollution-tolerant species Caoitella capitata dominated this community, and
represented 58.8 percent of the total infaunal abundance at this station.
The low abundances of polychaete and the domination of this community by
C. capitata indicates that this station is highly stressed by organic inputs
and may be beyond the "peak of opportunists" defined by Pearson and Rosenberg
(1978).
North Harbor Island, West Waterway, and Kellogg Island appeared, to be
the most severely impacted (see Figure 68). Twelve depressions were
detected among 9 stations in North Harbor Island, 25 depressions were
detected among 17 stations in West Waterway, and 12 depressions were
detected among 8 stations in Kellogg Island. Stations NH-03, WW-03, WW-20,
193
-------
KG-05, and KG-06 were the stations most severely impacted with three to four
depressions each. Two depressions each were detected at 10 of the other
stations. Stations within these three segments, all of which were located
in the vicinity of the Upper Duwamish Estuary, accounted for 73 percent of
the depressions among all Elliott Bay stations.
Comparison with Recent Historical Data
Comparisons of the benthic macroinvertebrate data collected in this
study with benthic data collected in previous studies were not conducted.
The available historical data were from studies collected in areas that did
not correspond to the study area segments in this study, or were collected at
locations further from shore and in deeper water (Tetra Tech 1985c). Thus,
no direct comparisons were appropriate.
Summary
¦ Of the 370 paired comparisons conducted among the Elliott Bay
stations where benthic infauna were collected, 219 of the
comparisons did not detect a significant difference in the
abundances of the major taxa. Significantly enhanced
abundances (P<0.001) were detected in 78 comparisons, and
significantly depressed abundances were detected in 73
comparisons (see Figures 57-61).
i
¦ Of the 74 stations sampled for benthic infauna, one or more
significant depressions (P<0.001) of four major taxa selected
for problem definition (i.e., Polychaeta, Crustacea, Pele-
cypoda, and Gastropoda) were detected at 40 of the stations
(see Figure 68).
¦ The most impacted areas in the Elliott Bay study area were
North Harbor Island, West Waterway, and Kellogg Island.
Stations within these segments accounted for 73 percent of
th£ depressions detected in all Elliott Bay stations.
Pe;lecypods and crustaceans appeared to be the most sensitive
indicators among the major taxonomic groups.
¦ The lowest number of taxa (at stations where organisms were
identified to the species level) occurred at Stations NH-03,
EW-05, NS-08, and KG-01. Two or more significant depressions
(P<0.001) in the abundances of major taxa were detected at
these stations (see Figure 68).
¦ Classification analysis of the species-level data (see
Figure 67) generated station groups that correspond fairly
well to the numbers of significant depressions (P<0.001)
recorded for abundances of major taxa among the stations.
FISH ECOLOGY
This section provides a description of the general characteristics of
the demersal fish assemblages and English sole populations sampled at the
194
-------
11 transects in Elliott Bay and the single transect at Point Pully (see
Figure 6). Demersal fish assemblages are compared between Elliott Bay and
Point Pully with respect to species composition, total abundance, total
number of species, and diversity. English sole populations are compared
between the two areas with respect to abundance and relative abundance.
Demersal Fish Assemblages
Species Composition—
A total of 10,751 fishes, representing 17 families and 37 species, was
sampled in this study (Table 31). Elliott Bay yielded 10,399 individuals
and 36 species, whereas 352 individuals and 18 species were captured at
Point Pully. Much of the observed differences in catches between the two
study areas likely resulted from the larger sampling effort expended in
Elliott Bay, but may also have been partly the result of increased habitat
complexity (e.g., pilings, rocks, debris) in Elliott Bay.
The most abundant family of fishes sampled in both Elliott Bay and Point
Pully was Pleuronectidae (35.8 and 61.1 percent, respectively). The most
abundant pleuronectids in both areas were rock sole (Lepidopsetta bilineata)
(15.6 and 36.4 percent, respectively) and English sole (14.8 and 20.5
percent, respectively).
Assemblage Characteristics-
Demersal fish assemblages at individual transects in Elliott Bay were
compared qualitatively with the assemblage at Point Pully on*the basis of
three major characteristics: total abundance, total number of species, and
diversity (Figure 69). The latter variable was estimated using the Shannon-
Wiener index (H1) (Shannon and Weaver 1949).
Total abundance at 9 of the 11 Elliott Bay transects exceeded the value
at Point Pully (i.e., 5.6 individuals/100 m) by a factor of 1.5 or greater.
Total abundances at the remaining two Elliott Bay transects (i.e., AB-91,
SS-91) were similar to the value at Point Pully.
The total number of species at Elliott Bay transects (i.e., range =
13-25) exceeded the value observed at Point Pully (i.e., 18) in five cases
and was lower than that value in six cases. The six transects having fewer
species than Point Pully included the four transects within or directly
influenced by the Duwamish River (DR-91, KG-91, EW-91, WW-91) and the two
transects in outer Elliott Bay (AB-91, MG-91).
The diversity at Elliott Bay transects (i.e., range = 1.38-2.35)
exceeded the value observed at Point Pully (i.e., 2.03) in six cases and was
lower than that value in five cases. The five transects having lower
diversity than that at Point Pully included three of the four transects in
or directly influenced by the Duwamish River (DR-91, KG-91, EW-91) and the
two transects in outer Elliott Bay (AB-91, MG-91).
In summary, fish assemblages at most Elliott Bay transects were more
abundant than those at Point Pully. By contrast, the numbers of species and
diversities of assemblages from most transects in or near the Duwamish River
195
-------
TABLE 31. RELATIVE ABUNDANCES OF FISHES CAPTURED
IN ELLIOTT BAY AND AT POINT PULLY
Relative Abundance (%)
Family
Species
Common Name
Elliott Bay
Point Pully
Squal idae
Squalus acanthias
spiny dogfish
<0.1
0.6
Rajidae
Raja binoculata
big skate
<0.1
Chimeridae
Hydro!agus colliei
ratfish
0.1
2.6
Clupeidae
Clupea harengus pal 1 asi
Pacific herring
2.9
Osmeridae
Hypomesus pretiosus
pretiosus
surf smelt
0.8
Batrachoididae
Porichthys notatus
plainfin midshipman
0.4
1.4
Gadidae
Gadus macrocephalus
Merluccius productus
Microgadus proximus
Pacific cod
Pacific hake
Pacific tomcod
<0.1
0.1
11.0
Gasterosteidae
Aulorhynchus flavidus
tube-snout
<0.1
Embiotocidae
Cymatogaster aggregata
Embiotoca lateral is
Rhacochilus vacca
shiner perch
striped seaperch
pile perch
14.6
0.8
2.7
0.9
2.8
6.8
Bathymasteridae
Ronquilus jordan
northern ronquil
0.4
Stichaeidae
Lumpenus sagitta
snake prickleback
22.2
Scorpaenidae
Sebastes auriculatus
Sebastes caurinus
brown rockfish
copper rockfish
0.5
2.4
4.0
11.4
Hexagrammidae
Hexagrammos stelleri
Ophiodon elongatus
Oxylebius pictus
whitespotted greenling
lingcod
painted greenling
0.2
<0.1
<0.1
0.3
Cottidae
Chitonotus pugetensis
Enophrys bison
Leptocottus armatus
Rhamphocottus richardsoni
Scorpaenichthys marmoratus
roughback sculpin
buffalo sculpin
Pacific staghorn sculpin
grunt sculpin
cabezon
0.4
<0.1
2.3
<0.1
0.3
0.3
Agonidae
Agonus acipenserinus
sturgeon poacher
<0.1
196
-------
TABLE 31. (Continued)
Relative Abundance (%\
Family
Species
Common Name
Elliott Bay
Point. Pully
Bothidae
Citharichthys sordidus
Pacific sanddab
0.5
2.0
Citharichthys stigmaeus
speckled sanddab
1.7
5.7
Pleuronectidae
Glyptocephalus zachirus
rex sole
0.7
Hippoglossoides elassodon
flathead sole
1.8
Lepidopsetta bilineata
rock sole
15.6
36,4
Lyopsetta exilis
slender sole
1.0
0.3
Microstomus pacificus
Dover sole
1.6
Parophrys vetulus
English sole
14.8
20.5
Platichthys stellatus
starry flounder
0.1
Pleuronichthys coenosus
C-0 sole
<0.1
2.8
Psettichthys melanostictus
sand sole
0.2
1.1
Total Catch
10,399
352
197
-------
i
i
i
I
I
i
i
511 '
1
1
I
UJ
UJ
o
UJ
0.
LL
o
cr
UJ
03
25-i
20-
1 5-
1 0-
I
I
2.5-1
I
il
i
i
i
i
!
i
I
i
i
i
i
i
i
i
i
I
i
PP-91 MG-91 NS-91 SS-92 SS-91 NH-91 EW-91 WW-91 KG-91 DR-91 NH-92 AB-91
STUDY AREA
Figure 69. Comparisons of major charactersistics of fish assemblages
between Elliott Bay transects and Point Pully (i.e., PP-91).
198
-------
and in outer Elliott Bay were lower than the respective values observed at
Point Pully. Although these comparisons are largely descriptive, they
suggest that the gross characteristics of fish assemblages in Elliott Bay
differed substantially from those at Point Pully only in the Duwamish River
and in outer Elliott Bay. The reasons for these differences are uncertain.
They could be related to several environmental factors, including chemical
contamination, low salinity in the Duwamish River, or low habitat complexity
and reduced benthic food sources in, outer Elliott Bay.
English Sole Populations
The abundance of English sole at 9 of the 11 Elliott Bay transects
exceeded that at Point Pully (i.e., 1.1 individuals/100 m) by a factor of 2
or greater (Figure 70). Abundances at the remaining two transects (i.e.,
SS-91, NH-92) were similar to the value at Point Pully. The reason English
sole are generally more abundant in Elliott Bay than at Point Pully is
unknown, but may be related partly to enhanced abundances of benthic
invertebrates (i.e., English sole prey) in the former embayment. The
relative abundance of English sole at Elliott Bay transects (i.e., range =
6.2-51.7 percent) exceeded the value observed at Point Pully (i.e.,
20.5 percent) in six cases and was lower than that value in five cases
(Figure 70). Relative abundances of English sole showed no apparent spatial
relationships within Elliott Bay.
Summary
¦ • The most abundant family of fishes in both Elliott Bay and
Point Pully was Pleuronectidae
¦ The most abundant pleuronectids in both study areas were rock
sole and English sole, respectively
¦ The abundances of demersal fishes at all Elliott Bay transects
were similar to or greater than the abundance at Point Pully
¦ The total numbers of species and diversities of fish assem-
blages at most transects in or near the Duwamish River and in
outer Elliott Bay were lower than the respective values at
Point Pully
¦ The abundance of English sole at most Elliott Bay transects
exceeded the abundance of that species at Point Pully
¦ The relative abundances of English sole exhibited no apparent
spatial relationships within Elliott Bay.
FISH HIST0PATH0L0GY
This section presents the results of histopathological analyses
conducted on the livers of English sole collected at 11 trawl transects in
Elliott Bay and at one transect in a reference area off Point Pully (see
Figure 6). Three major kinds of lesion were evaluated: neoplasms, foci of
cellular alteration, and megalocytic hepatosis (see METHODS section).
Before the lesions are considered, the age and sex characteristics of the
199
-------
E
o
o
Ui
o
z
<
a
z
3
CD
<
6-1
5
4
3 -
2 -
1
0
M
i
1
31.5
1
&
i
51.7
Z
UJ
o
tr
Ui
&
UJ
o
z
<
Q
Z
3
03
<
Ui
>
Ui
QC
i
I
i
i
II
i
i
I
i
I
i
i
PP-91 MG-91 NS-91 SS-92 SS-91 NH-91 EW-91 WW-91 KG-91 OR-91 NH-92 AB-91
STUDY AREA
Figure 70. Comparisons of abundances and relative abundances of
English sole between Elliott Bay transects and Point Fully
(i.e., PP-91).
200
-------
English sole sample from each Elliott Bay transect is compared with the
respective characteristics of the sample from Point Pully. The overall
prevalences of the lesions in Elliott Bay and at Point Pully are then
presented, and the relationships between the prevalence of each lesion and
fish sex and age are determined. Next, comparisons of lesion prevalences
between each Elliott Bay transect and Point Pulley are made. Comparisons of
length-at-age are then made between fish with and without hepatic lesions in
Elliott Bay. Finally, results of the present study are compared with
historical information on the prevalence of hepatic lesions in English sole
from Elliott Bay.
Evaluation of the Reference Area
The prevalences of major lesions in livers of English sole collected
from the Point Pully reference area during the present study were as
follows: 3.3 percent megalocytic hepatosis, 6.7 percent foci of cellular
alteration, and 0 percent neoplasms. Landolt et al. (1984) previously
reported a prevalence of 0 percent for each of these three lesions in
English sole at Point Pully. Among eight reference areas reviewed by Tetra
Tech (1986e), the range of lesion prevalence was 0-2.6 for megalocytic
hepatosis and 0-5.8 percent for foci of cellular alteration. The prevalence
of neoplasms in livers of English sole was 0 percent at all eight reference
areas. Based on these data, Point Pully appears to be representative of
reference area conditions for major liver lesions in English sole of Puget
Sound.
Aae and Sex Characteristics of Fish Populations
Otoliths were available for age determination for 702 of the 714
(98 percent) English sole sampled in Elliott Bay and at Point Pully. Ages
of the 12 fish not having corresponding otoliths were estimated from age-
length keys based on the 702 fish of known age, stratified by sex (Ricker
1975). Once age determinations were made, fish younger than 3 years old
(n=17) were excluded from subsequent analyses. As noted previously, the
goal of this study was to focus on the individuals most likely to be
afflicted with serious idiopathic hepatic lesions.
Ages of English sole at four transects from Elliott Bay differed sig-
nificantly (P<0.05) from ages of fish captured at Point Pully (Table 32).
In all four cases, median fish age was less than that observed at Point
Pully. Male proportion of English sole at nine transects from Elliott Bay
differed significantly (P<0.05) from the proportion at Point Pully, being
greater in all instances (Table" 32).
General Patterns of Lesion Prevalences
A total of 291 of the 697 (41.8 percent) English sole (age >3 years)
sampled from Elliott Bay and Point Pully had one or more of the three kinds
of hepatic lesion considered in this study (Table 33). Of this total, 185
(63.6 percent) had only a single kind of lesion, 92 (31.6 percent) had two
kinds of lesion, and 14 (4.8 percent) had all three kinds of lesion.
Prevalences of every kind of hepatic lesion evaluated in this study were
greater in Elliott Bay than at Point Pully (Table 33; Figure 71). In most
201
-------
TABLE 32. COMPARISONS OF AGE AND MALE
PROPORTION BETWEEN ENGLISH SOLE FROM
POINT PULLY AND ELLIOTT BAY
Transect
Sample
Size3
Median
Age (yr)b
Male
Proportion0
Point Pully
PP-91
60
6.2
0.13
Elliott Bay
MG-91
58
6.2 ns
0.05 ns
NS-91
60
5.0*
0.58***
SS-92
60
4.9*
0.80***
SS-91
48
4.8*
0.63***
NH-91
57
4.6*
0.61***
EW-91
58
"5.6 ns
0,81***
WW-91
60
6.1 ns
0.67***
KG-91
59
6.8 ns
0.20 ns
DR-91
59
6.2 ns
0.47***
NH - 92
58
6.3 ns
0.47***
AB-91
60
5.1 ns
0.38***
a All fish were >3 years old.
b Comparisons were made using the Mann-Whitney U-test.
* = P<0.05; ns = P>0.05.
c Comparisons were made using the G-test of independence.
*** = P<0.001; ns = P>0.05.
202
-------
TABLE 33. PREVALENCES OF HEPATIC LESIONS
IN ENGLISH SOLE FROM ELLIOTT BAY
AND POINT PULLY
Prevalence 1%)
Elliott Bay Point Pully
Hepatic Lesion (n=637a) (n=60a)
Neoplasms
Liver cell adenoma 2.4 0
Hepatocellular carcinoma 3.6 0
Choi angioma 0.3 0
One or more kinds of neoplasm** 5.8 0
Foci of celluar alteration
Eosinophilic foci 14.8 1.7
Basophilic foci 14.9 3.3
Clear cell foci 6.6 1.7
One or more kinds of fpci of
cellular alteration" 25.0 6.7
Megalocytic hepatosis 33.0 3.3
One or more of the three major
kinds of hepatic lesion" 44.9 8.3
a All fish were >3 years old.
b Some fish had more than one kind of hepatic lesion.
203
-------
~
POINT PULLY
n = 60
ELLIOTT BAY
n = 637
40-
HI
o
cc
UJ
a
Ui
o
z
uu
-J
<
>
UJ
GE
a.
30-
20-
10-
(0)
NEOPLASMS
FOCI OF CELLULAR
ALTERATION
MEGALOCYTIC
HEPATOSIS
HEPATIC LESION
Figure 71. Comparisons of prevalences of hepatic lesions between
Elliott Bay and Point Pully.
204
-------
cases, the differences between these two areas were substantial. Megalocytic
hepatosis was the lesion found most frequently in Elliott Bay (33.0 percent),
whereas foci of cellular alteration was the lesion observed most commonly at
Point Pully (6.7 percent). Neoplasms were not found at Point Pully.
Within Elliott Bay, prevalences of neoplasms and foci of cellular
alteration were correlated positively (P<0.05) with increasing age of fish
(Figure 72). These patterns are consistent with the results of past studies
(Mai ins et al. 1982; McCain et al. 1982; Becker et al. 1987; Rhodes et al.
1987). Prevalence of neoplasms increased from 0 percent in 2-year-old fish
to 15.4 percent in fish aged >10 years old. Prevalence of foci of cellular
alteration increased from 5.9~percent in 2-year-old fish to 50.0 percent in
fish aged >10 years old. Prevalence of megalocytic hepatosis was not
correlated with fish age (P>0.05) (Figure 72), but exhibited a steadily
increasing trend between age 2 (11.8 percent) and age 7 (42.1 percent).
Prevalences of megalocytic hepatosis at ages 8, 9, and >10 (26.8, 30.8, and
23.1 percent, respectively) were considerably lower than the prevalence
observed for age 7.
Prevalence of neoplasms exhibited a significant difference (P<0.05)
between sexes, whereas prevalences of foci of cellular alteration and megalo-
cytic hepatosis did not differ significantly (P>0.05) between sexes. Of the
three kinds of neoplasm observed in this study (Table 34), only hepato-
cellular carcinomas exhibited a substantial difference between sexes (i.e.,
26 percent for males, 74 percent for females). No past study has found a
relationship between fish sex and prevalence of hepatic neoplasms in English
sole (McCain et al. 1977, 1982; Malins et al. 1982; Krahn et al. 1986;
Becker et al. 1987; Rhodes et al. 1987). Several of these studies have
considered fish collected from Elliott Bay.
Comparisons of Lesion Prevalences Between Study Areas
Because prevalences of neoplasms and foci of cellular alteration in
Elliott Bay correlated with fish age (Figure 72), age distributions at those
transects that differed from Point Pully with respect to fish age (NS-91,
SS-92, SS-91, and NH-91; see Table 32) were adjusted before comparisons with
the reference area were made. Adjustments were made by sequentially removing
the youngest fish from each of the four Elliott Bay transects until the
remaining age distribution did not differ significantly (P>0.05) from the
age distribution at Point Pully. In making these adjustments, 7, 7, 3, and
14 fish were removed from Transects NS-91, SS-92, SS-91, and NH-91, respec-
tively. A graphical comparison between the age distribution at each Elliott
Bay transect (including the four adjusted distributions) and the age
distribution at Point Pully is presented in Figure 73.
Although prevalence of neoplasms differed between sexes (Table 34), the
sex distributions at those transects that differed from Point Pully with
respect to male proportion were not adjusted before comparisons with the
reference area were made. Adjustments were not made primarily because the
relationship observed in this study is not consistent with results of past
studies of hepatic lesions in English sole. Thus, the relationship observed
in this study does not appear to be a general pattern.
205
-------
H
Z
IU
o
DC
IU
Q.
Ui
O
z
111
_l
<
>
IU
DC
a.
25 -
20 -
15 -
10 -
5 -
0
Neoplasms
rs- 0.92"
/a12LelEL
2 3
&.I
i
i
8 9 >10
50
40 -
30 -
20 -
10
0
Foci of Cellular Alteration
YA
rs- 1.00*
of
I
50
40 -
30 -
20 -
10 -
0
2 3 4
Megalocytic Hepatosis
rs- 0.25 ns
8 9 >'
>10
y/«
I
I
i
8 9 >10
FISH AGE (yr)
AGE
SAMPLE
GROUP
SIZE
2
17
3
68
4
154
S
100
6
92
7
76
8
56
9
39
210 .
52
Figure 72. Comparisons of prevalences of hepatic lesions with age of
English sole from Elliott Bay using Spearman's coefficient of
rank correlation (rs). " P £ 0.01, P <, 0.001, ns = P > 0.05.
206
-------
LU
O
K
Ui
Q.
>
O
z
LU
3
o
UJ
(C
MG-91
n - 58
SS-92
i ¦ 53.
NH-91
n -43
40-,
30
20-
10-
0
WW-91
n-60
40 -|
30-
20-
10
0
DR-91
n -59
t—n—i—i—i—i
3 4 5 6 7 fl 9 10 >10
NS-91
n-53
SS-91
n > 45
40-|
30
20
10-)
EW-91
n -58
40-|
30-
20-
10
0
KG-91
n • 59
40
30-j
20
10
0
NH-92
fl — 58
t-i—i—n—i—i—n
3 4 5 6 7 3 9 10 >10
AB-91
n - 60
—i—i—i—n—t—i
3 4 5 8 7 8 9 10 >10
FISH AGE (yr)
Figure 73. Comparisons of age distributions between Elliott Bay
transects (solid lines) and Point Pully (dashed lines).
207
-------
TABLE 34. COMPARISONS OF LESION PREVALENCES BETWEEN MALE
AND FEMALE ENGLISH SOLE FROM ELLIOTT BAY
Hepatic Lesion
Percent having Each
Kind of Lesion®
Males Females
(n=328) (n=309)
Significance^
Neoplasms
3.7
8.1
*
Foci of cellular alteration
23.2
26.9
ns
Megalocytic hepatosis
35.4
30.4
ns
a All fish were >3 years old.
b Comparisons were made using the G-test of independence. * = P<0.05; ns =
P>0.05.
208
-------
In most cases, age-adjusted prevalences of each kind of hepatic lesion
at each transect in Elliott Bay exceeded the corresponding value from Point
Pully (Table 35). Concordance among the prevalences of the three kinds of
lesions across all 12 transects was significant (W=0.70, P<0.05).
Prevalence of neoplasms in fish from the 11 Elliott Bay transects
ranged from 0 to 12.1 percent, and did not differ significantly (P>0.001)
from the prevalence of 0 percent observed at Point Pully at any Elliott Bay
transect. The highest prevalence of neoplasms in fish from Elliott Bay was
found at Transect NH-^92.
Prevalence of foci of cellular alteration in fish from Elliott Bay
transects ranged from 6.7 to 44.1 percent, and differed significantly
(P<0.001) from the prevalence of 6.7 percent observed at Point Pully at four
transects (NH-91, WW-91, KG-91, and NH-92). The highest prevalence of foci
of cellular alteration in fish from Elliott Bay was found at Transect KG-91.
Prevalence of megalocytic hepatosis in fish from Elliott Bay transects
ranged from 1.7 to 58.1 percent, and differed significantly (P<0.001) from
the prevalence of 3.3 percent observed at Point Pully at eight transects
(NS-91, SS-92, SS-91, NH-91, EW-91, WW-91, KG-91, and NH-92). The highest
prevalence of megalocytic hepatosis in fish from Elliott Bay was found at
Transect NH-91.
Prevalence of one or more of the three kinds of hepatic lesion in fish
from Elliott Bay transects ranged from 10.3 to 65.0 percent, and differed
(P<0.001) from the prevalence of 8.3 percent observed at Point Pully at nine
transects (NS-91, SS-92,. SS-91, NH-91, EW-91, WW-91, KG-91, DR-91, and
NH-92). The highest prevalence of one or more of the three lesions in
Elliott Bay was found at Transect WW-91.
The spatial distributions of the three kinds of hepatic lesion are
presented in Figures 74-76. The highest prevalences of both neoplasms and
foci of cellular alteration were found in fish in or near the Duwamish
River. Prevalences of both of these lesions were moderate along the Seattle
waterfront and low (similar to the prevalences at Point Pully) in outer
Elliott Bay. The highest prevalences of megalocytic hepatosis were found in
fish from the lower Duwamish River and along the Seattle waterfront.
Prevalences of this lesion were moderate in fish from the upper Duwamish
River and similar to the prevalences at Point Pully in outer Elliott Bay.
The spatial distributions of the three kinds of hepatic lesion indicate
that most abnormalities were confined to inner Elliott Bay. Within inner
Elliott Bay, the most serious abnormalities were confined largely to the
areas in or near the Duwamish River. These patterns suggest that fish
throughout inner Elliott Bay are stressed, and that individuals near the
Duwamish River are stressed to the greatest extent.
Lenoth-at-Aoe Comparisons
Length-at-age of male English sole with hepatic lesions differed sig-
nificantly (P<0.05) from that of fish without lesions only for 10-year-old
individuals, being greater for fish without lesions (Figure 77). Median
length-at-age for all other age groups was similar between males with and
209
-------
TABLE 35. COMPARISONS OF PREVALENCES OF HEPATIC LESIONS
IN ENGLISH SOLE FROM ELLIOTT BAY AND POINT PULLY
Prevalence [%)*
Foci of
Cellular
One or
Transect
Size**
Neoplasms
A1 teration
Hepatosis
Lesions
Point Pully
PP-91
60
0
6.7
3.3
8.3
Elliott Bay
MG-91
58
0 ns
8.6 ns
1.7 ns
10.3 ns
NS-91
53
3.8 ns
15.1 ns
37.7***
41.5***
SS-92
53
0 ns
22.6*
45.3***
54.7***
SS-91
45
4.4 ns
11.1 ns
40.. 0***
44.4***
NH-91
43
11.6**
41.9***
58.1***
60.5***
EW-91
58
1.7 ns
27.6**
50.0***
53.5***
WW-91
60
10.0**
35.0***
51.7***
65.0***
KG-91
59
10.2**
44.1***
27.1***
62.7***
DR-91
59
11.9**
33.9**
20.3**
49.2***
NH-92
58
12.1**
36.2***
39.7***
53.5***
AB-91
60
1.7 ns
6.7 ns
5.0 ns
11.7 ns
a Comparisons were made using the G-test of independence. * = P<0.05; ** =
P<0.01; *** = P<0.001; ns = P>0.05.~ ^
b All fish were >3 years old, and the age distribution at each transect from
Elliott Bay does not differ significantly (P>0.05) from the age distribution
at Point Pully.
210
-------
MAGNOLIA
BLUFF
> 0 - 5%
>5-10%
r. OUWAMISH
HEAO
HARBOR
ISLAND
ALKI POINT
REFERENCE AREA
POINT PULLY
~
miles
kilometers
LEGEND
> 10%
Significant Elevation Above
Reference:
* P < 0.05
** P<0.01
*** P< 0.001
Figure 74. Spatial distribution of prevalences of neoplasms in
English sole from Elliott Bay.
211
-------
MAGNOLIA
BLUFF
E
D
/ OUWAWISH
HEAO
ALKI POINT
HARBOR
: ISLANO
LEGEND
[ | 0-10%
EH > 10-20%
H > 20 - 30%
¦¦ >30%
Significant Elevation Above
Reference:
* P < 0.05
** P S 0.01
~** Ps 0.001
REFERENCE AREA
POINT PULLY
E
miles
kilometers
Figure 75. Spatial distribution of prevalences of foci of cellular
alteration in English sole from Elliott Bay.
212
-------
MAGNOLIA
BLUFF
c
o
/ OUWAMISH
/ HEAD
ALKI POINT
i.lSLANO
LEGEND
I I «•<«*
EH > 15 - 30%
gg >30-45%
¦¦ >45%
Significant Elevation Above
Reference:
* P S 0.05
** PS 0.01
~** Ps 0.001
f. SEATTLE
~~~
REFERENCE AREA
» 1/
POINT PULLY
C
miles
kilometers
L***
Figure 76. Spatial distribution of prevalences of megaiocytic
hepatosis in English sole from Elliott Bay.
213
-------
350
_ MALES
~ LESIONS ABSENT
LESIONS PRESENT
E
E,
z
H"
o
z
uu
300-
250-
200-
ns
13
1
ns
29 28 44
ns
19
I
34
ns
28
ns
18
18
ns
14 13
s
ns
9 9
!
6
I
<
y-
o
<
Q
UJ
350-1
300 -
250-
200-
FEMALES
6
I
20
ns
26 56
I
ns
24 24
!
20
18
i
ns
17
24
I
ns
14
IS
ns
15
I
ns
10
FISH AGE (yr)
Figure 77. Comparisons of length-at-age between English sole-with
and without hepatic lesions using the Mann-Whitney LMest.
U £ 0.05, ns = P > 0.05. Sample size is presented above
each bar.
214
-------
without lesions. Length-at-age of female English sole with hepatic lesions
differed significantly (P<0.05) from that of fish without lesions only for
3- and 6-year-old individuals, being greater for 3-year-old fish with
lesions and 6-year-old fish without lesions. Median length-at-age for 4-,
5-, and 10-year-old females was similar between fish with and without
lesions. Although not significant (P>0.05), length-at-age of females aged
7-9 years old was substantially greater for fish without lesions.
The comparisons of length-at-age between fish with and without hepatic
lesions suggest that the presence of lesions may be associated with reduc-
tions in fish growth only for female English sole. In that case, differences
in length-at-age first became evident when fish reached an age of 6 years
old. McCain et al. (1982) evaluated English sole from the Duwamish River
and found that length-at-age (both sexes pooled) did not differ significantly
(P>0.05) between fish with and without hepatic lesions. However, the
authors noted that a slight reduction (approximately 2-3 percent) in length-
at-age was evident for fish with lesions.
Comparison with Recent Historical Data
Results of the present study were compared with those of a survey
conducted throughout Elliott Bay between 1979 and 1982 by Mai ins et al.
(1984). Comparisons were limited to descriptive evaluations because
different age distributions of English sole were examined in the two studies
and because, in some cases, fish were collected at different locations and
during different seasons. In making these comparisons, prevalences were
averaged within four areas: the upper Duwamish River (Transects KG-91, and
DR-91), the Harbor Island area (Transects NH-91, EW-91, WW-91, and NH-92),
the Seattle waterfront (Transects NS-91, SS-92, and SS-91), and outer
Elliott Bay (Transects MG-91 and AB-91). Because Malins et al. (1984) did
not sample near Point Pully, results from that area in the present study
were compared with the averaged results of the four nonurban Puget Sound
embayments sampled by Malins et al. (1984): Case Inlet, Discovery Bay, Port
Madison, and Port Susan.
The relative spatial patterns of lesion prevalences in the present study
and in that of Malins et al. (1984) were nearly identical (Figure 78). The
prevalences of both neoplasms and foci of cellular alteration in Elliott Bay
steadily declined with increasing distance from the upper Duwamish River.
Prevalences were highest in the upper Duwamish River, second in magnitude
near Harbor Island, third in magnitude along the Seattle waterfront, and
lowest in outer Elliott Bay. Prevalences in outer Elliott Bay were similar
to those in the reference areas. By contrast with the above pattern, the
prevalences of megalocytic hepatosis did not exhibit a gradient with respect
to distance from the upper Duwamish River. Prevalences generally were high
throughout the upper Duwamish River, the Harbor Island area, and the Seattle
waterfront. Prevalences of all three lesions in outer Elliott Bay were
similar to those found in reference areas.
The similarity in the relative spatial patterns of lesion prevalences
between the present study and the study conducted 3-6 years earlier by Malins
et al. (1984) indicate that these patterns are real (i.e., they are not
artifacts of the design of either study), and that they are quite stable
over time. The temporal stability of these patterns suggests that the
215
-------
NEOPLASMS
20
15
10 -
5
0
F3 <°> ™ f0> <°>
40
z
UJ
a
E
UJ 20
CL
30 -
UJ
o
z
UJ
-I
<
>
UJ
CE
CL
10 -
so
40 -
30 -
20 -
10-
DUWAMSH
RIVER
(136-118)
FOCI OF CELLULAR ALTERATION
0
n
MEGALOCYTIC HEPATOSIS
n T7i
HARBOR
ISLAND
(399-219)
SEATTLE
WATERFRONT
(161-151)
OUTER
BAY
(116-118)
JS
JS
REFERENCE
AREA
(156-60)
LOCATION
(NOTE: SAMPLE SIZES ARE GIVEN BELOW LOCATION NAMES.)
Figure 78. Comparison of prevalences of hepatic lesions in English
sole sampled in the present study (stippled bars) and in
Malins et al (1984; open bare)
216
-------
causes of the lesions are localized within Elliott Bay, and that the causes
have not been reduced substantially between 1979 and 1985.
Differences between the results of the present study and Mai ins et al.
(1984) were related primarily to the absolute prevalence of each kind of
lesion at each location. The prevalences of foci of cellular alteration and
megalocytic hepatosis found in the present study were higher than the
prevalences observed by Mai ins et al. (1984) at most locations. By contrast,
the prevalences of neoplasms observed by Malins et al. (1984) were higher
than the prevalences found during the present study at three of the five
locations.
Several other previous studies have documented high prevalences of
hepatic lesions in English sole from the Duwamish River. Pierce et al.
(1978) examined 62 fish and found hepatic neoplasms in 32 percent of the
sample. McCain et al. (1982) examined 551 fish and found prevalences of
neoplasms, foci of cellular alteration, and megalocytic hepatosis to be
12.9, 9.0, and 18.5 percent, respectively. Krahn et al. (1986) evaluated
the same three lesions in 58 fish and found prevalences of 20.7, 32.8, and
44.8 percent, respectively.
Summary
¦ Three kinds of hepatic lesion were considered in this study:
neoplasms, foci of cellular alteration, and megalocytic
hepatosis.
¦ Prevalences of neoplasms and foci of cellular alteration-were
correlated positively (P<0.05) with fish age.
¦ Prevalence of neoplasms, primarily hepatocellular carcinomas,
was higher (P<0.05) in females than in males.
¦ Prevalence of neoplasms was not elevated significantly
(P>0-001) at any of the Elliott Bay transects.
¦ Prevalence of foci of cellular alteration was elevated
significantly (P<0.001) at four Elliott Bay transects (NH-91,
WW-91, KG-91, and NH-92).
¦ Prevalence of megalocytic hepatosis was elevated significantly
(P<0.001) at eight Elliott Bay transects (NS-91, SS-92, SS-91,
NH-91, EW-91, WW-91, KG-91, and NH-92).
¦ The spatial distributions of the three kinds of hepatic lesion
indicate that most abnormalities were confined to inner
Elliott Bay. Within inner Elliott Bay, most serious abnor-
malities were confined largely to the areas in or near the
Duwamish River.
¦ Comparisons of length-at-age between fish with and without
hepatic lesions suggested that the presence of lesions may be
associated with reductions in fish growth only for females at
ages greater than 5 years old.
217
-------
d Results of the present study were compared with historical
data collected by Mai ins et al. (1984). The relative
magnitudes of lesion prevalences were similar between
studies. The prevalences of both neoplasms and foci of
cellular alteration steadily declined with increasing
distance from the "upper Duwamish River. Prevalences were
Highest in the upper Duwamish River, second in magnitude near
Harbor Island, third in magnitude along the Seattle water-
front, and lowest in outer Elliott Bay. Prevalences of
megalocytic hepatosis generally were high throughout the
upper Duwamish River, the Harbor Island area, and the Seattle
Waterfront.
218
-------
Comparison of Bioassav Responses with Benthic Groupings
Values of amphipod mortality were compared with>the groupings of
stations determined by classification analysis of benthic invertebrate
assemblages (see RESULTS, Benthic Macroinvertebrates). The results of this
comparison show that the mean level of amphipod mortality differed among
groups of stations that differed in benthic infaunal characteristics
(Figure 86). The mean bioassay mortality among Port Susan stations (Group I
of benthic infauna in Figure 86) was low relative to other benthic infaunal
groups. Amphipod mortalities at stations in benthic infaunal Group II
ranged from 10 to 60 percent, with a mean of 30 percent. Generally, the
stations in Group II exhibited one or no significant depressions of major
taxa. At the six stations where depressions were observed (i.e., Stations
NH-01, NH-02, SS-04, WW-09, WW-11, and WW-14), they could all be attributed
to reductions of pelecypod abundance. Group III stations (EW-11, KG-01, and
NH-08) displayed a wide range of amphipod mortality values, with a mean of
61 percent among stations. All three stations were characterized by
extremely high abundances of Tharvx sp. Crustacean abundances were
significantly depressed at all three stations in Group III. Pelecypods were
also significantly depressed at Stations KG-01 and NH-08. The four stations
that were ungrouped in the classification analysis (i.e., Stations EW-05,
NH-03, NH-04, and NS-08) showed high amphipod mortalities, ranging from
82 to 100 percent, with a mean of 91 percent. These stations exhibited
several characteristics indicative of extremely stressed conditions,
including depressions in the abundances of major taxa and dominance by
pollution tolerant species (see RESULTS, Benthic Macroinvertebrates).
In conclusion, stations that exhibited the most severe benthic effects
also displayed the highest toxicity levels found in the study. These
stations were outliers (ungrouped stations) in the classification analysis.
A similar finding was reported by Tetra Tech (1985a). The least contaminated
stations (i.e., Group I) displayed the lowest toxicity values. At stations
with intermediate effects on benthic infauna (e.g., Groups II and III in
Figure 86), a wide range of toxicity values was observed, and mean toxicity
among stations was generally moderate.
Comparison of Significant Responses
The relationship of significant and nonsignificant amphipod bioassay
results to the presence or absence of at least one significant depression of
major infaunal taxa is presented in Table 36. Overall, the concordance in
the responses of the bioassay and infauna was not substantially better than
that expected by chance alone (i.e., the percentage of stations showing
consistent responses was not substantially higher than 50 percent).
Nevertheless, concordance between the bioassay and benthos was found at
highly contaminated sites, especially in the North Harbor Island study area
(see RELATIONSHIPS AMONG CONTAMINANTS, TOXICITY, AND BENTHIC EFFECTS,
General Correlation of Indicators). The sensitivity of the benthic effects
indicators relative to the amphipod bioassay was high in Elliott Bay. In
the East Waterway, West Waterway, and Kellogg Island areas, many stations
displayed significant benthic effects (P0.001), but were not classified as
toxic in the amphipod bioassay (P>0.001). There are several reasons why the
amphipod bioassay appears insensitive relative to benthic effects indicators.
232
-------
fSJ
CO
u>
100 -
r NH8
- EW5
- NHS
<
~
- NH4
80 -
- NS8
>-
H .
LEGEND
-J
<
i
r EW11
- STATION MEAN
tr 60 -
r WW9 '
f
Ua
o
• MEAN AMONG STATIONS
s
- Afll
H
- NH2
U 40 -
- WWII
O
cc
UJ
<
~
0.
I-PS2
20 -
- Wt
I^Ul
1
FPS4
- WW14
- PSt
- NS3/SS4
LPS3
L ssii
o ¦
I II III UNGROUPED
STATION GROUP
Figure 86. Amphipod bioassay responses in relation to station groupings based on
classification analysis of benthic assemblages.
-------
TABLE 36. CORRESPONDENCE BETWEEN STATIONS HAVING
SIGNIFICANT (P<0.001) BIOASSAY RESPONSES
AND STATIONS HAVING SIGNIFICANT (P<0.001)
BENTHIC DEPRESSIONS
Benthic Depression
Bioassay Response
Yes
No
Yes
13%
7%
No
40%
40%
NOTE: Total no. stations = 74.
234
-------
First, the| statistical power of the bioassay tests in the Elliott Bay
project was low relative to benthic taxa tests because of the relatively
high bioassay mortality in Port Susan and because log-transformation of the
benthic data greatly increased the power of the test. Second, the benthic
effects tests incorporate four indicators (i.e., major taxa), whereas the
bioassay is a single variable. Finally, the benthic effects tests are
indicators of both chronic and acute effects whereas the bioassay measures
only the latter.
RELATIONSHIP BETWEEN BIOACCUMULATION AND SEDIMENT CONTAMINATION
This section addresses the relationship between sediment contamination
and bioaccumulation in English sole muscle tissue. Of the chemicals analyzed
for bioaccumulation (PCBs, EPA priority pollutant pesticides, and mercury),
only PCBs will be examined. Mean concentrations of mercury at study sites
were relatively low and in all cases were below the mean concentration at the
reference area (Point Pully). Pesticides were detected infrequently and
occurred at concentrations well above detection limits in only one fish.
The following approach was used to examine sediment/fish PCB bioac-
cumulation relationships:
¦ Sediment PCB concentrations were averaged across regions
corresponding to trawl transects (using geometric means).
Because detection limits were relatively high for PCBs in
sediments (>100 ug/kg DW), detection limits were not used in
tliese analyses. This exclusion resulted- in the elimination of
Magnolia Beach data (PCBs were undetected at all MG stations
at: 100-200 ug/kg DW). Sediment PCB data were not collected
for Point Pully, so these reference data were not available
for inclusion.
¦ Geometric mean PCB concentrations in muscle tissue were
generated for each trawl transect (detection limits were not
of concern, as PCBs were detected in all tissue samples).
¦ Scatterplots were generated and Pearson correlation coef-
ficients were calculated. Scatterplots were generated for
wet weight bioaccumulation data vs. dry weight sediment data
and for lipid-normalized bioaccumulation data vs. T0C-
nqrmalized sediment data.
The use of I'ipid-normal ized bioaccumulation data and TOC-normalized sediment
data derives from theoretical models based upon equilibrium partitioning of
nonpolar organic compounds, such as PCBs (e.g., McFarland 1984). For sedi-
ment/fish systems at equilibrium in the environment, theoretical models
predict a linear relationship between lipid- and TOC-normalized data.
Scatterplots of sediment/fish bioaccumulation relationships are
presented in Figure 87. The correlations for these scatterplots were not
significant (r=0.33 to 0.36; P>0.05) and did not appear to conform to
theoretical predictions. Possible explanations for the lack of significant
linear correlations are that: (1) the system is not at equilibrium,
(2) fish were not confined to the areas designated for this analysis,
235
-------
80 -i
WW-91
a
a
60 -
5*
Q
3
lii
3
CO
Ui
p
40 -
EW-flt
a
z
(ft
CD
a
&
20 -
0 -
SS-tt
a
A8-91
1 • n I
SS-91
a
K09t
a
NS-01
a
HMl
a
i i
NH*91
a
i
OR-91
a
i
0 10 20 30 40
PCBs IN SEDIMENT (ug/g organic carbon)
500 -
ew-91
a
a
WW-91
£
a»
*5
S
9
3
400-
KtWl
a
on-91
a
NH-91
a
Q»
300-
-a*
3
SS-92
a
1U
3
CO
2
20O-
NS*1
a
SS-91
a
P
z
a
a
o
&
s
1 1 ft
A0-91
A
i i
UUM
iwnft
a
0 200 400 600 BOO
PC8s IN SEDIMENT (ug/kg dry weight)
Figure 87. Relationship between PCB bioaccumulation and PCB
sediment contamination.
236
-------
(3) sampling intensity was not sufficient to characterize fish contamination
and/or sediment contamination, particularly after detection limits were
excluded from sediment data. With regard to the latter explanation, fewer
than five stations were used to characterize sediment contamination in 4 of
the 10 stutjy areas. All trawl transects were represented by five fish.
SUMMARY
¦ Biological effects as measured by the amphipod toxicity
b'ioassay and significant reductions in abundances of benthic
infauna taxa were generally associated with higher concentra-
tions of contaminants in sediments.
¦ The relationship between sediment contamination and abundances
of several selected benthic taxa was nonlinear. A clear
rjelationship with an apparent threshold in the biological
response was found for selected chemicals: PAH, PCBs, copper,
arsenic, cadmium, and sulfides. For the other organic
compounds evaluated, the number of stations with detected
vlalues was too small or the distribution of the data was too
skewed to show a strong association with biological variables.
¦ The abundance of the pollution-tolerant polychaete Caoitella
capitata displayed a roughly linear increase with sulfide
concentration in sediments and no apparent relationship with
copper, PCBs, or,grain size.
¦ Amphipod mortality was generally high at higher concentrations
ojf copper and sulfides, but was not clearly related to TOC
cbntent, grain size, or the other chemical variables evalu-
ated.
¦ Both high (>40 percent) amphipod mortality and severe
(>80 percent) depression of at least one major taxon of
infauna were observed at the following highly contaminated
sites: Stations EW-05, NH-03, NH-04, NH-05, NH-08, and
NS-08. Severe biological effects were found at only one
station (i.e., depressions of all four major taxa at Station
WW-03) where chemical contamination was relatively low (i.e.,
all concentrations were below LAET).
¦ Biological effects at several stations in Area SS along the
Seattle South waterfront were less than expected based on the
severity of chemical contamination at these sites (especially
Stations SS-03, SS-08, and SS-09). Other highly contaminated
sites where severe biological effects were not found included
Stations EW-14, AB-01, and intertidal Station KG-10.
¦ Tlpe abundances of most of the benthic taxa evaluated were low
at stations with >50 percent amphipod mortality in the
toxicity bioassay. In contrast, £. caoitata displayed low
abundances at low levels of amphipod mortality and a wide
range of abundances at high amphipod mortality.
237
-------
¦ Concordance between statistically significant responses in the
toxicity bioassay and depressions of infaunal taxa was not
greater than that expected by chance alone. However, this is
not surprising given the wide range and levels of contaminants
in the Elliott Bay system and the different endpoints measured
by these two indicators (i.e., acute mortality of adults of a
single species in the bioassay and chronic effects on all life
stages of an assemblagfe of species in the benthic infaunal
indices).
¦ Linear correlations between PCB bioaccumulation and PCB
sediment contamination (i.e., wet weight bioaccumulation data
vs. dry weight sediment data and lipid-normalized bioaccumula-
tion data vs. TOC-normalized sediment data) were not signifi-
cant (P>0.05).
238
-------
PRIORITIZATION OF PROBLEM AREAS AND CONTAMINANTS
In this section, the selected data for indicators of sediment contamina-
tion, toxicity, and biological effects are integrated to evaluate toxic con-
tamination problems in the Elliott Bay system. The approach for the ranking
of problem areas was described earlier (see METHODS, Decision-Making Frame-
work) and is summarized in Figure 3. Based on the significance and magnitude
of EAR compiled in the Action Assessment Matrix format, analysis of problem
areas and their priority ranking was performed in the following phases:
¦ Tier I Problem Definition—Identification of broad areas that
exceeded Action-Level Guidelines (see Table 3) for combined
significant elevations of sediment chemistry, fish pathology
and bioaccumulation
¦ Tier II Problem Definition—Identification of problem stations
that triggered Action-Level Guidelines based on significant
EAR and exceedance of a) the 90th percentile concentration or
HAET of chemicals in sediments, b) 80 percent depression of
any one of four major benthic taxa (Polychaeta, Crustacea,
Pelecypoda, or Gastropoda), c) 40 percent mortality in the
amphipod bioassay, or d) any combination of the preceding.
Grouping of problem stations .into problem areas was based on chemical
distributions (including data from recent historical studies), the nature and
proximity of potential sources, and geographic and hydrographic boundaries.
¦ Ranking of Problem Sites—Scoring of problem stations
following the criteria in Table 4, and ranking of each
problem area based on the average of the scores for individual
stations within the area.
Finally, potential problem chemicals that exceeded AET were identified.
IDENTIFICATION OF PROBLEM AREAS
Broad areas of the nearshore Elliott Bay system displayed significant
elevations of PCB concentrations in muscle tissue and of lesion prevalences
in liver of English sole. Only the Magnolia and Alki Beach study areas (as
delineated in Figure 2) did not exhibit statistically significant elevation
of liver lesions. Because every sediment chemistry station in areas other
than Magnolia and Alki Beach showed a significant elevation of at least one
chemical indicator, all of the Elliott Bay system inside of a line from Pier
90/91 to Duwamish Head exceeded an action level for problem area definition.
Information on the significance of EAR for all indicators at each
station was compiled in an Action Assessment matrix. Because of the large
number of sediment stations (102) in Elliott Bay, only stations identified
as Tier II problem sites were considered for further priority ranking.
Stations identified as part of Tier II problem definition are listed in
239
-------
Appendix G. The indicators that exceeded action levels for severe contami-
nation and effects (i.e., those that potentially cause a single indicator to
trigger problem definition) are also shown in Appendix 6.
The Tier II problem stations were grouped into problem areas based on
consideration of the following factors:
¦ Chemical distributions (including data from recent historical
studies)
¦ Nature and proximity of potential sources
¦ Geographic and hydrographic boundaries.
The following problem areas containing multiple stations (including histor-
ical stations) were identified (Figure 88):
¦ DRI (in the upper Duwamish River)
¦ EW (in the East Waterway)
¦ KGI and KGII (near Kellogg Island)
¦ NHI and NHII (in the North Harbor Island area)
¦ NSI and NSII (along the Seattle North waterfront)
¦ SS (along the Seattle South waterfront)
¦ WWI and WWII (in the West Waterway).
In addition, the following single stations were identified as problem areas:
AB-01, DR-05, DR-10, DR-12, DR-15, DR-16, DR-25, KG-03, and NH-10
(Figure 88).
RANKING OF PROBLEM AREAS
Ranking of problem areas within the Elliott Bay system was performed
using the Action Assessment Matrix. Arithmetic mean EAR values compiled for
each data type and each multi-station problem area (Tier II) are shown in
Table 37. Reference values are shown on the right-hand side of the table.
For each indicator, mean reference values across all stations within the
reference area are shown for comparison. The original value for an indicator
can be obtained by multiplying the EAR reported in the table by the ap-
propriate reference value. Only the original data for the prevalence of
liver neoplasms is shown because the reference area prevalence was zero,
resulting in infinite elevations at the study sites. Note that benthic
infauna EAR are calculated as the ratio of the reference value to the study
site value because a toxic effect is expected to produce a depression in
abundance. Refer to Appendix G for information on sample sizes (number of
stations) for each indicator.
For perspective in interpreting Table 37, each of the following
represent a severe effect that is sufficient for definition of a problem
240
-------
rco-oa
OR-te
DR-15
LEGEND
HS Seattle North Waterfront
$ S Seattle South Waterfront
NH North Harbor Island
£ W Easi Waterway
WW West Wate/way
K Q Kellogg Island
OROuwamish Aiver
6000
53 feet
S rn*tar%
2000
NOTE: Historical stations (not shown)
were used to define problem
areas in addition to the stations
shown. For example. Problem
Ares NSI was defined predomn
nantly by data from Romberg
et al
Figure 88. Boundaries of priority problem areas
241
-------
TABLE 37. ACTION ASSESSMENT MATRIX OF SEDIMENT CONTAMINATION, TOXICITY,
AND BIOLOGICAL EFFECT INDICES FOR ELLIOTT BAY PROBLEM AREAS
Problem Area tievations*
Variable
East
Waterway
North
Kellog Harbor
Island Is. I
North
Harber
Is. II
Seattle
Water-
front-N.
Seattle
Water-
front-S.
West
Waterway
I
Uest
Waterway
II
. Reference
Valueb
Sediment Chemistry
Ks
Cd
Cu
Pb
Hg
Ag
Zn
LP AH
HPAH
PCBs
Phenol
2-methylphenol
4-me thy1 phenol
Dimethyl phthalate
Butyl benzyl phthalate
1,4-Dichlorobenzene
p.p'-DDE
p,p'-ODD
p.p'-ODT
Benzyl alcohol
Benzoic acid
Sediment Toxicity
Amphlpod mortality
Infauna
Polychaetes
Gastropods
Pelycypods
Crustaceans
Fish Pathology
HeopTasms*
Foci
Heg. hep.
Bioaccumulation
PCBs
I 1.6 I
2,900
3.7
0.S4
"15
18
QT
230
I S.5 | I 2.6 I | 1.71
1.7
0.32
0.45
"TO"
4.7
1.7
4.1
I 15 I
rss i
JO.
6.6
8.2
12
12
6.3
18
5.4
12
60
56
17
0.87
1.0
0.6
0.71
5.9
6.4
3.1
1.0
3.8
2.3
2.2
2.6
11 1
13 I
29 1
44 1
2.0
1.5
0.8
2.0
4.5
3.2
5.2"
16
81
81
3.37 ppm
0.95 ppm
6.37 ppm
9.2 ppm
0.04 ppm
<0.09 ppm'
19 ppm
<41 ppb
<79 ppb
<6 ppb
<33 ppb
U 7 ppb
<13 ppb
U 40 ppb
U 17 ppb
tl 3.5 ppb
U 10 ppb
U 10 ppb
U 10 ppb
U 10 ppb
<150 ppb
161
l,958/m2
146/m2
2,307/m2
1,349/m2
01
6.7*
3.3*
5.4 ppb
¦ Boxed numbers represent elevations or chemical concentrations that exceed ail Puget sound reference area values, and
statistically significant toxicity and biological effects at the P<0.001 significance level compared with reference conditions.
The "U" qualifier indicates the chemical was undetected and the detection limit is shown. The "<" qualifier indicates the
chemical was undetected at one or more stations. The detection limit is used in the calculations. Infauna EAR are based on
the elevation in biological effects represented by reductions in infaunal abundances relative to reference conditions. EAR
for all other variables reflect an increase in the value of the variable at Elliott Bay compared with reference conditions.
Blank spaces in sediment chemistry columns indicate that the chemical was undetected throughout the problem area.
b EAR values shown for each area are based on Carr Inlet reference values for sediment chemistry, on Point Pully reference
values for fish pathology and bioaccumulation, and on Port Susan (1985) reference values for sediment toxicity and infauna.
c Prevalence of neoplasms at each problem area is shown in table Instead of EAR because the reference value was Q%.
d Data for trawl Station WW-91 were assigned to both problem areas (I and II) in the Nest Waterway.
-------
area:
¦ >40 percent amphipod mortality, which corresponds to an EAR
of >2.5
¦ >80 percent depression in abundance of one or more benthic
taxa, which corresponds to an EAR of >5
¦ Exceedance of the 90th percentile or the high AET for
sediment chemistry
¦ Significant elevation of any three indicators.
At least one of the four primary conditions just listed are met by each area
shown in Table 37. Significant EAR for sediment chemistry and fish pathology
were found in all of the problem areas shown in Table 37. Nearly all
stations within these areas exceeded the 90th percentile for one or more
chemicals in sediments (see Appendix G). Chemicals that exceeded AET at each
station are discussed in the next section (see CHEMICAL CHARACTERIZATION OF
PROBLEM AREAS).
The ranking criteria presented in Table 4 were applied to the Action
Assessment Matrix for single stations (see Appendix G). Total scores for
sediment chemistry and biological effects were determined separately for
each station. The score for each multi-station area was calculated as the
average of the scores for individual stations within the area (for details,
see METHODS). Normalized scores for the Tier II problem areas and single
stations are presented in Figures 89 and 90, respectively. The highest
priority problem areas, which were defined as those areas with scorei
>60 percent are:
I
a SS—South Shore of downtown Seattle
l
n NHI and NHII—Stations immediately north of Harbor Island, at
the mouth of the West Waterway, and west to Station NH-08
near the outflow of Longfellow Creek
¦ WWI and WWII—Segments of the West Waterway.
Areas SS, NHII, and WWII ranked as highest-priority areas based on sediment
chemistry only. Area WWI ranked above 60 percent based on biology only.
Area NHI scored among the highest priority areas for both chemistry and
biology.
Twenty-nine stations scored >75 percent based on either sediment
chemistry or biological effects (Figures 90 and 91). Of the 29 highest
priority stations, 5 scored >75 percent for both sediment chemistry and
biological effects: Stations EW-05, NH-03, NH-04, WW-09, and WW-11.
Station NH-03 received high scores for both chemistry and biology. Station
SS-09 received the highest possible score for sediment chemistry, but ranked
relatively low for biological effects. Stations DR-15, DR-16, and WW-02
received the maximum possible score for biological effects, but displayed
only moderate scores for sediment chemistry.
243
-------
100 —i
90 -
80 -
NHI
ss
NHI
w.v.
NHII
70 -
n
WWI
WWII
60-
•XvX
44+H
NHII
H+M
1**¥
KGII
WWI KGII
X-Xv
50-
.w.v
EW
-
WWII
40 -
NSII
30 -
Iv'.v'.*
SS
•AW!
NSII
20 -
CHEMISTRY
BIOLOGY
AVERAGE RANK SCORE
LEGEND
NS Seattle North Waterfront
S S Seattle South Waterfront
N H North Harbor Island
EW East Waterway
WW West Waterway
KG Kellogg Island
Figure 89. Ranking of priority problem areas based on average
conditions within each area.
244
-------
ro
cn
Seattle North Waterfront and
Seattle South Waterfront Stations
Alki Beach and
North Harbor Island Stations
SS-09
SS-03
NS-01 SS-04
-07 -10
•11
NS-06
100-J
90
60
70-
60 -
50-
40-
30 -j
20
*20-
NS-01
SS-09
I NS-06
I SS-04 SS-10 NS-04
-05 11 -07
-07 -12
NH-03
04
AB01
NH-06
-00
NH-01
•02
-05
NK-10
m
100
90-
80
70 -
60 -
50
40
30
20
c20 -1
NH 03
I
I
•M
W.
NH-02
NH-04
NH-0S
-08
NH-01
NH-06
A8-01
NH-10
East Waterway Stations
too
90
EW-QS
EW-02 EW11
«4 14
<16 15
•07
•09
EWI0
12
•13
EW-06
EW-16
60
70 -
60 -
50 *
40
30 -
20
m
EW-G5
EW09
11
EW-04 EW-08
07 -10
EW06
EW-02
16
EW 12
13
14
-t5
CHEMISTRY
BIOLOGY
CHEMISTRY
BIOLOGY
CHEMISTRY
BIOLOGY
AVERAGE RANK SCORE
NOTE: Sea Figure 5 lor nation locallons. Intertldal site*, which were ranked toi biology based on the
amphipod bioassay only. Inducted DR-25, KQ-10, NH10, NS-01, NS-04, WW-02. and WW-15.
Figure 90. Ranking of single stations classified as problem sites.
-------
West Waterway Stations
ro
a>
WW-12
14
•19
WW-09
•10
•11
WW-02 WW-16
•04 -17
-06 18
-08
-13
WW-01
-05
-IS
20
WW-09
100
90-
80
70
60 -
50-
40-
30
20 -
*20-
M
WW-02
WW-09
II
WW-03
WW-20
WW-IO WW-16
•12 19
ww4i ww u
45 17
WW-04 -06 18
•13
Kellogg Island Stations
100 -i
90-
•80
KG 01
70-
KG-OS
1
06
-09
>:&:
60-
10
8ft:
KG-03
50-
-11
i
v.v
Iw
40-
M
30 -
KG 08
•
WW-15
20
<20
KG05
06
KG-11
KG-01
08
09
KG-03
Duwamish River Stations
100
OR 12
OR-08
16
Ofl-10
15
25
OR-06
ESSE
90 -
80 -
70
60
50 -
40 -
30
20
¦20
&
ORIS
16
DR-05
-08
10
12
CHEMISTRY
BIOLOGY
CHEMISTRY
BIOLOGY
CHEMISTRY
BIOLOGY
AVERAGE RANK SCORE
NOTE; Sea Figure 5 lor station locations. Intertidal sites, which were ranked lor biology based on ihe
amphlpod bloassay only. Included DR-25, KQ-10. NH-10, NS-01, NS-04, WW-02, and WW-15.
Figure 90. (Continued).
-------
,ss
\NHI
AB-Ot (C)
NHII
WWI
Note: Problem areas and single stations with
labels exhibited ranks >75% for chemistry (C).
biology (0). or both (C, B). NS-01 and WW-02
wore intertidal sites, which were ranked for
biology based on the amphipod bioassay only.
2000
Figure 91. Highest priority problem sites.
247
-------
CHEMICAL CHARACTERIZATION OF PROBLEM AREAS
In this section, the multiple-station and single-station problem areas
listed above are characterized with regard to the distributions of selected
problem chemicals (i.e., those chemicals with concentrations that exceed
AET). To facilitate analysis by the Elliott Bay Work Group, a description
is provided of the chemicals at notable stations within each problem area.
Problem chemicals for all Tier II problem stations are summarized in Table
38. Detailed tables of AET exceedances (including the factor by which AET
were exceeded) are included in Appendix F. The following points should be
considered regarding the application of AET to chemical data in this study:
¦ AET values have not been established for all chemicals
measured in the present study (e.g., pentachlorophenol).
However, AET have been developed for the most frequently
detected and most elevated chemicals.
¦ AET for antimony, chromium, nickel, beryllium, and thallium
were not used to determine problem stations in this study.
The AET for antimony and chromium were not used because of
the likelihood that analytical methods used to generate AET
for these chemicals are not directly comparable to the methods
used in the present study (see Table 7). The AET for nickel
was not used because the range of nickel concentrations in the
database used to generate Puget Sound AET is relatively
limited. Beryllium and thallium were excluded for similar
reasons; the exclusion of these two chemicals applies only to
historical data, as they were not measured in the present
study.
The 90th percentile concentrations of nickel and chromium and chemicals
without AET were used to designate problem stations (if the concentrations
were above the range of Puget Sound reference concentrations).
Problem Area SS
The sediments in the problem area along the Seattle South waterfront
were highly contaminated throughout, and had more chemicals exceeding HAET
than in any other problem area (Table 39). Superimposed upon the high
levels of certain problem chemicals throughout this area (e.g., PAH and
mercury) were maximum concentrations of different problem chemicals occurring
at a number of non-adjacent stations. These patterns suggested that multiple
local sources were present, perhaps in conjunction with a more diffuse source
for compounds such as PAH.
PAH were the most commonly occurring problem chemicals and exceeded the
HAET at 14 of the 15 stations that defined the problem area (Table 39).
HAET for other organic compounds (e.g., PCBs, 1,4-dichlorobenzene, and
chlorinated pesticides) and a number of metals (including cadmium, copper,
lead, zinc, mercury, and silver) were exceeded at least once in the problem
area. The highest concentrations of many problem chemicals in the problem
area (and in the entire study) occurred at Stations SS-08 and SS-09 (see
Tables 8 and 12). Historical TPPS Station S0090 was similar to Station
SS-09 in its assemblage of problem chemicals exceeding HAET, but the two
248
-------
TABLE 38. POTENTIAL PROBLEM CHEMICALS
LPAHj*
HPAHb
1-methylphenanthrene
2-methyl naphthalene
biphenyl
dibenzofuran
PCBs
p,p'-DDE
p,p'-DDD
p,p'-DDT
phenol0
2-methylphenol
4-methylphenol
2,4-dimethylphenol
1,4-dichlorobenzene
dimethyl phthalate
butyl benzyl phthalate
benzoic acid
benzyl alcohol
arsenic0
cadmium
copper
lead
mercury
silver
zinc
a The term LPAH represents the following chemicals: naph-
thalene, acenaphthylene, acenaphthene, fluorene, phenan-
threne, and anthracene.
b The term HPAH represents the following chemicals:
fluoranthene, pyrene, benzo(a)anthracene, chrysene, total
benzofluoranthenes, benzo(a)pyrene, indeno(l,2,3-c,d)py-
rene, dibenzo(a,h)anthracene, and benzo(g,h,i)perylene.
c This chemical exceeded the LAET but never exceeded the
HAET. All other chemicals exceeded the HAET at least
one time.
Z49
-------
TABLE 39. SEATTLE SOUTH UATERFRONT PROBLEM AREAS3
HAET Exceedances'1
Station
LPAH HPAH PCBs
CU
PB
ZN
HG
CD
AG DDT/DDE/DDD PHNL 2.4MEPHN BNZOH
14DICLBNZ
BUTBNZPH DMP
TOC
LAET Exceedancesc
SS-03
SS-04
SS-05
r\)
B061
SS-12
m10015d
X
X
S0090e
X
X
SS-06
X
X
SS-07
X
X
S0065e
X
SS-08
X
X
SS-09
X
X
SS-10
X
X
, C061e
X
X
I SS-11
A
X
X X X X
X X X X
HG, LPAH, AS, PB, CD, PCBs,
DDD [AG, DMP1
HG, PB, PCBs, ZN [AG]
HG, PB, PCBs, ODD, ZN
[RETENE, CARBAZOLE, DINOCT]
HG, PCBs
HG, PB, PCBs, ZN [CO, AG]
PB, CU, PCBs, ZN
HG, PCBs
HG
CU, DMP, PHNL [ALDRIN,
RETENE, CARBAZOLE, AS, CR]
HG, ZN [CR, Nl]
PCBs, ZN
HG, LPAH, PCBs, ZN [AG]
PCBs, DDE, ZN
[AG]
LPAH - Signifies AET exceedances for the sun of naphthalene, acenaphthylene, acenaphthene, fluorene, phenanthrene, and anthracene, or any of these compounds individually. To
simplify the presentation of AET exceedances in this table, exceedances of AET for 1-methylphenanthrene, 2-methylnaphthalene, biphenyl, and dibenzofuran are included under
LPAH. These compounds covaried with LPAH but are not included in LPAH sons.
HPAH - Signifies AET exceedances for the sum of fluoranthene, pyrene, benzo(a)anthracene, chrysene, total benzofluoranthenes, benzo(a)pyrene, indeno(1,2,3-cd)pyrene, dibenzo(a,h)-
anthracene, and benzo(g,h,i)perylene, or of any of these compounds individually.
CU - copper
PB - lead
ZN - zinc
HG - mercury
CD - cactnium
AG - silver
AS - arsenic
BNZOH - benzyl alcohol
BNZACID - benzoic acid
1,4DICLBNZ- 1,4-dichlorobenzene
BUTBNZPH - butyl benzyl phthalate
DMP - dimethyl phthalate
TOC - total organic carbon
DINOCT - di-n-octyl phthalate
PHNL - phenol
4MEPHNL - 4-methylpher>ol
2MEPHNL - 2-methylphenol
2.4MEPHNL - 2,4-dimethylphenol
TOTXYLENE - total xylenes
CR - chromiun
ENDALD - endrin aldehyde
PCP - pentachlorophenol
^ Chemicals exceeding HAET for Puget Sound. More detailed information on exceedances is presented in Appendix F.
c Chemicals exceeding LAET for Puget Sound. More detailed information on exceedances is presented in Appendix F. Chemicals shown in brackets exceeded 90th-percentile concentra-
tions but did not exceed any AET. Chemicals exceeding both AET and 90th-percentile concentrations are not bracketed.
Mai ins et al. (1982).
e Romberg et al. (1984).
-------
stations were not adjacent (Table 39). Although contaminated stations such
as SS-08 arid SS-09 were riot Rotated near obvious potential sources, a number
of stations in this problem area are located near CSOs: Station SS-03 (King
Street CSO), Station SS-04 [Washington Street CSO (072)], Station SS-05
[Madison Street CSOs (071,164)], Station SS-06 [University CSO (070)], and
Station SSj-11 [Vine Street CSO (069)]. Sediments throughout the problem
area tended to be fine-grained (with the exception of Station SS-07) and
rich in organic matter (see Figures 8, 9, 15, and 16). For example,
sediments at Stations SS-08 and SS-09 were composed of over 60 percent fine-
grained material and had 26.6 and 10 percent T0C, respectively.
Concentrations of PAH, the prevalent organic contaminants in the problem
area, decreased in either direction from the extremely high concentrations at
Station SS^08 (roughly 0.38 percent DW of the EPA priority pollutant hydro-
carbons) (see Figure 32), and tended to correlate well with T0C content (see
Table 22). Detection limits for most individual LPAH were very high
(>1,000 ug/jkg DW) at Station SS-12, which may explain why HAET for PAH were
not exceeded at this station.
The most elevated metals in this problem area had similar overall
distribution patterns. For the metals of highest concentration in the area,
concentrations were relatively constant and elevated throughout the area,
with pronounced maxima at non-adjacent stations (typically SS-03, SS-09, and
TPPS Station S0090). Examples of these distributions include mercury [with
pronounced imaxima at Station SS-09 (see Figure 28) and TPPS Station S0090)],
zinc [with pronounced maxima at Stations SS-03 and SS-09 (Figure 30) and
TPPS Station S0090], lead [with an extreme concentration of roughly 7 percent
DW at Statio'n SS-09 (see Figure 26) and a high concentration at TPPS
Station S0090], cadmium [with maxima at Station SS-09 (17.2 mg/kg DW), TPPS
Station S0(D90 (27 mg/kg DW), and Station SS-03 (7.16 mg/kg DW)], and
arsenic, which had a maximum of 584 mg/kg DW at Station SS-03; other
SS station^ were typically an order of magnitude lower in arsenic concen-
trations. Copper distributions were somewhat more variable but maximized at
Stations SS-03 and SS-07 (see Figure 24). The high concentration at
Station SS^07 is unusual because this station is relatively coarse-grained
(roughly 15 percent fine-grained material). Silver concentrations, which
were elevated in the problem area near Denny Way CSO, were also elevated
above HAET in the SS problem area. Concentrations between 4.3 and 6 mg/kg DW
occurred at Stations SS-12, SS-11, and SS-05 to SS-07. Notably, chromium
and nickel! concentrations maximized at Station SS-10 and were the highest
values observed in the study (chromium = 1,080 mg/kg DW; nickel =366 mg/kg
DW). Chromium and nickel were near or below reference levels at other
stations in the problem area.
PCB concentrations were generally elevated but patchy in Problem Area
SS (see Figure 36), especially considering historical data (e.g., 2,600 ug/kg
DW at TPP^ Station S0090 in contrast to 600 ug/kg DW at nearby Station
SS-05). Ai high detection limit reported at Station SS-08 (2,400 ug/kg DW)
impeded data interpretation in that area. Pesticides exceeded HAET at
stations with relatively high PCB concentrations (e.g., Stations SS-04 and
SS-09) and may have been overestimated because of GC/ECD interferences.
1,4-Dichlorobenzene concentrations exceeded the HAET at Stations SS-09
(31,000 ug/kg DW) and SS-03 (380 ug/kg DW), but high detection limits
preclude detailed evaluation of its distribution in the problem area.
251
-------
Benzyl alcohol (1,300 ug/kg DW) exceeded the HAET at Station SS-03 (near the
King Street CSO), and was undetected at a relatively high detection limit
(690 ug/kg DW) at the nearest station.
The Port of Seattle recently provided further characterization of the
area by resampling sediments near Stations SS-08 and SS-09 (Aggerholm, D.A.,
22 February 1988, personal communication). Analyses for three metals
(cadmium, lead, and zinc) and a range of semivolatile organic compounds
revealed that contamination in these slips is heterogeneous but highly
elevated. PAH concentrations at resampled Station SS-08 and the lead
concentrations at resampled Station SS-09 were over 200 times lower than
were reported in the present study (but were still at or near the LAET
range). Cadmium and zinc concentrations at resampled Station SS-09 were
roughly an order of magnitude less than the original concentrations.
However, the PAH concentrations at resampled Station SS-09 were similar to
but higher than the high concentrations reported in the present study, and
the lead concentration at resampled Station SS-08 was 4-5 times the original
value. The PAH and lead concentrations at these resampled stations exceeded
HAET. Also, PCP was detected at roughly 1,000 ug/kg DW at resampled
Station SS-09. These results do not confirm the extreme concentrations
originally reported at Stations SS-08 and SS-09. Nevertheless, they do
support the characterization of this area as highly contaminated overall
(especially by PAH), with localized patches of extreme contamination.
Problem Area NHI
Problem Area NHI, which encompasses two shipyard facilities at the
mouth of the West Waterway, is characterized by very high.concentrations of
PAH, PCBs, and several metals (including copper, mercury, lead, zinc, and
arsenic). Chemicals that exceeded HAET in at least one station in this area
include LPAH, HPAH, PCBs, copper, lead, zinc, mercury, arsenic, p.p'-DDD,
2-methylphenol, 4-methylphenol, and 2,4-dimethylphenol (Table 40). Although
high concentrations were found in this problem area during the present study,
data from historical studies contributed significantly to the characteri-
zation of this problem area and included the maximum concentrations for most
of the problem chemicals.
Stations adjacent to Todd Shipyard [including Station NH-03, Station 4
(U.S. EPA 1982, 1983), and Station 3 (Gamponia et al. 1986)] had the highest
concentrations of several high priority chemicals in this problem area. At
these three stations, concentrations of mercury ranged from 10.5 to 12 mg/kg
DW, concentrations of copper ranged from 1,700 to 2,800 mg/kg DW, concentra-
tions of PAH (as the combined sum of LPAH and HPAH) ranged from 59,000 to
over 600,000 ug/kg DW, and concentrations of PCBs ranged from 3,300 to
14,000 ug/kg DW. Concentrations at a historical station seaward of
Station NH-03 exceeded HAET values for copper and PAH. Contamination
generally decreased moving east from Todd Shipyard, but HAET for PAH were
exceeded as far east as Station NH-01. PCB contamination was high but
patchy along this area, with a concentration of 7,500 ug/kg DW at Station 1
of Gamponia et al. (1986). The highest lead concentration on the east side
of the West Waterway mouth was not adjacent to Todd Shipyard (roughly
1,000 mg/kg DW at EPA Station 37) (see U.S. EPA 1982( 1983). Arsenic
concentrations were not highly elevated over most of the problem area, but a
252
-------
TABLE
40.
NORTH HARBOR ISLAND PROBLEM AREAS*
HAET Exceedances™
Area
Station
LPAH
HPAH
PCBs
2MEPHNL/
CU PB ZN HG AS ODD 4MEPHNL 2.4MEPHNL
LAET Exceedancesc
NHI
NH-Ol
NH-02
HPAH, PCBs
HPAH, PCBs, HG
GAMPld
X
X
HPAH, HG
E36e
X
X
GAMP2d
X
AS. HPAH, CU, PB, HG,
ZN, PCBs
E37e
X
X
X
ZN
GAMP3d
X
X
X
XXX X
AS
E4e
X
X
X
X X
AS
NH-03
X
X
X
X XX
AS. BUTBNZPH, PB, ZN
[TOTXYL]
E39e
X
X
LPAH. PCBs, ZN
WW-19
X X
ZN, HG, PCBs. HPAH
E40e
X
HPAH. ZN
(West
Waterway
Mouth)
GAMP4d
WW-20
GAMP5d
E41e
X
X
AS, HG, ZN
HPAH, HG, PCBs
LPAH, HPAH, PCBs, HG
HPAH
GAMP6d
X
X
AS, HPAH, LPAH, PCBs,
HG, ZN
E42e
X
XXX X
HPAH, LPAH
NH-04
X
X
X X
AS, PB. HG, ZN, 4MEPHNL,
PCBs [PCP, PHNL]
NHI I
NH-05
X
HPAH, HG. PCBs. ZN
GAMP7d^
X
X
X
HG
£43*
X
X
ZN
U120f
X
PCBs, ZN
HH-06
X
X
HG, PCBs, ZN [CARBA-
ZOLE, CD]
NH-08
X
X
PCBs. ZN
GAMP8d
X
LPAH, HPAH, PCBs
E44e
X
X
ZN
S00349
X
X
HG. LPAH
NH-10 PCBs
[ALDRIN, D1EL0RIN]
-------
TABLE 40. (Continued)
a LPAH - Signifies AET exceedances for the sum of naphthalene, acenaphthylene, acenaphthene, fluorene, phenanthrene, and anthracene,
or any of these compounds individually. To simplify the presentation of AET exceedances In this table, exceedances of
AET for 1-methylphenanthrene, 2-methylnaphthalene, biphenyl, and dibenzofuran are Included under LPAH. These compounds
covaried with LPAH but are not included In LPAH sums.
HPAH - Signifies AET exceedances for the sum of fluoranthene, pyrene, benzol a)anthracene, chrysene, total benzOfluoranthenes,
benzo(a)pyrene, 1ndeno(1,2,3-cd)pyrene, dibenzo(a,h)anthracene, and benzo(g,h,1)pery1ene, or of any of these compounds
individually.
b Chemicals exceeding HAET for Puget Sound. More detailed information on exceedances is presented in Appendix F.
c Chemicals exceeding LAET for Puget Sound. More detailed information on exceedances is presented In Appendix F. Chemicals
shown in brackets exceeded 90th-percent1le concentrations but did not exceed any AET. Chemicals exceeding both AET and 90th-
percentile concentrations are not bracketed.
d Gamponia et al. (1986).
e U.S. EPA (1982, 1983).
f Stober and Chew (1984).
4MEPHNL - 4-methyl phenol
2MEPHNL - 2-methylphenol
2.4MEPHNL - 2,4-dimethylphenol
HG - mercury
CD - cadmium
AG - silver
AS - arsenic
PHNL - phenol
CU - copper
PB - lead
ZN - zinc
BNZOH - benzyl alcohol
BNZACID - benzoic acid
1.4DICLBNZ - 1,4-dichlorobenzene
BUTBNZPH - butyl benzyl phthalate
DMP - dimethyl phthalate
TOC - total organic carbon
DINOCT - dl-n-octyl phthalate
TOTXYLENE - total xylenes
CR - chromium
ENDALD - endrln aldehyde
PCP - pentachlorophenol
9 Romberg et al. (1984).
-------
concentration of 560 mg/kg DW was observed at a historical station near Todd
Shipyard.
Concentrations of most problem chemicals decreased markedly moving west
across the| mouth of the West Waterway. A historical station on the west
side of the waterway near Lockheed Shipyard (EPA Station 42) (U.S. EPA 1982,
1983) had concentrations of a number of metals that far exceeded HAET,
including copper (1,050 mg/kg DW), lead (2,180 mg/kg DW), zinc (4,810 mg/kg
DW), and arsenic (1,420 mg/kg DW). Station NH-04, located adjacent to
Lockheed Shipyard on the east side of the West Waterway mouth, had LPAH,
HPAH, and copper concentrations that exceeded HAET. Notably, this station
also had a pentachlorophenol concentration of 6,000 ug/kg DW, which was the
highest detected concentration in this study. 4-Methylphenol and 2-methyl-
phenol were also elevated at this station (1,000 ug/kg DW and 240 ug/kg DW).
The southern boundary of the problem area was extended to include
Station WWrl9, which is located within the West Waterway adjacent to Todd
Shipyard. 'This station was included because it had a copper concentration
that was anomalously high relative to other West Waterway stations in the
area. The HAET for 4-methylphenol was also exceeded at this station.
Problem Area NHII
Problem Area NHII is predominated by PAH contamination, with less
widespread contamination by PCBs and less severe contamination by mercury
and zinc (Table 40). The most severe PAH contamination occurred near
Stations NH-06 and NH-08, located near the Wyckoff creosote facility and the
outflow of Longfellow Slough, respectively. Problem Areas NHI and NHII are
contiguous,, but are distinguished based upon differences in the nature of
contamination (metals contamination is far more severe in Problem Area NHI)
and based upon the nature of potential sources.
LPAH concentrations at Stations NH-06 and NH-08 were 57,000 and
37,000 ug/kg DW, respectively, and HPAH concentrations were 130,000 and
79,000 ug/kg DW (among the highest PAH concentrations observed in the present
study). Station NH-06 had the highest concentrations of naphthalene,
biphenyl, and carbazole in the present study, suggesting creosote as a
possible source material (Nestler 1974). Data for historical stations in
this area iconfirmed the high PAH concentrations (U.S. EPA 1982, 1983;
Gamponia etj al. 1986), although the concentrations reported by Gamponia et
alj. (1986) were somewhat lower than those reported by other studies at
similar locations. PAH concentrations decreased with distance moving east
and west from Stations NH-06 and NH-08 but nonetheless exceeded AET at
Stations NHj-05 and S0034 (Romberg et al. 1984). Notably, contamination was
very similar at Stations NH-06 and NH-08 in terms of composition and
absolute concentration (based upon consideration of PAH, alkylated PAH,
biphenyl, dibenzofuran, and carbazole). The major difference between the
two stations was a higher fluoranthene concentration at Station NH-06 (the
concentration was higher by a factor of roughly 6). Neglecting fluoranthene,
the precision between the concentrations of EPA priority pollutant PAH at
these two stations ranged from 4 to 78 RPD (relative percent difference)
with a mean RPD of 36 percent. As an indication of the high degree of
similarity ^observed between Stations NH-06 and NH-08, these two samples
255
-------
would easily pass the precision criterion for analytical replicates specified
under the PSEP program.
PCB concentrations (ranging from 3,100 to 6,600 ug/kg DW) exceeded the
HAET at two historical stations in the problem area. PCB concentrations in
this problem area did not follow readily apparent gradients; however, it was
apparent that PCB contamination did not covary with PAH contamination.
Zinc and mercury exceeded LAET at a number of stations in the problem
area. Zinc distributions were consistent with PAH distributions (concentra-
tions were nearly identical at Stations NH-06 and NH-08), whereas mercury
was more elevated in the region near Stations NH-06 and NH-05 than in the
region near Station NH-08. An isolated concentration of 1.2 mg/kg occurred
at historical station S0034 (Romberg et al. 1984).
Problem Area WWI
This problem area encompasses several areas of isolated contamination by
organic compounds (particularly PAH and PCBs; Table 41) and several stations
with severe biological effects but only moderate measured contamination.
Isolated but extreme PAH contamination was observed at Station WW-04,
located adjacent to the 16th Avenue SW CSO/SD (104). This station was
unique not only because of its high PAH concentrations (LPAH ¦ 15,000 ug/kg
DW and HPAH = 53,000 ug/kg DW) but also because of its anomalously low
HPAH/LPAH ratio (an indication of petroleum-related source material).
Concentrations at adjacent stations were over an order of magnitude lower
than at Station WW-04 despite comparable grain size and T0C content at these
stations. Relatively high PAH concentrations were also observed at
Station WW-06 [adjacent to the SW Hinds CSO/SD (099)] and at stations north
of this area.
A very high PCB concentration (24,000 ug/kg DW) was observed at
historical Station 14 of Gamponia et al. (1986); this was the highest PCB
concentration observed among all data reviewed for this study. PCB concen-
trations decreased sharply at stations on a cross-waterway gradient moving
away from Station 14 (mid-waterway concentrations were roughly 1,000 ug/kg
DW). Although the SW Hinds CSO/SD (099) is in the vicinity of Gamponia
Station 14, the PCB concentration at a station closer to the outfall
(Station WW-06, this study) was roughly 40 times lower.
This problem area also contained the highest concentration of benzyl
alcohol in the study (8,800 ug/kg; intertidal Station WW-02), but the
compound was undetected at detection limits below 200 ug/kg DW at nearby
stations and no potential source was apparent. Station WW-08, separated
from Station WW-02 by a number of stations without detected benzyl alcohol,
had a benzyl alcohol concentration exceeding the HAET (140 ug/kg DW).
Problem Area WWII
Among the variety of chemicals exceeding HAET in this area (Table 41),
lead contamination most clearly distinguishes this area from Problem Area
WWI. The HAET for lead was exceeded along the east side of the waterway from
Stations WW-11 to WW-14. Station WW-14, located adjacent to the SW Lander
CSO/SD (105) and the SW Lander SD (21 inches), had an extremely high lead
256
-------
TABLE 41. WEST WATERWAY PROBLEM AREAS3
HAET Exceedances'1
Area
Station
LPAH
HPAH
PCBs
CU
PB DDT/ODD 2.4MEPHNL
BNZOH
DMP
LAET Exceedancesc
wwi
WW-01
WW-02
WM-03
WW-05
WW-04
WM-06
GAMP14d
S0036e
WW-08
GAMP12d
PCBs
PCBs, ZN
PCBs
PCBs
PCBs, PB, HG, ZN,
BUTBNZPH, HPAH
HPAH, HG
PCBs, PB, HG
PCBs, HG, BUTBNZPH
[4MEPHNL]
HPAH, LPAH, PCBs
ro
on
¦vj
WWII
WW-11
WW-09
WW-12
WU-li)
WW-13
GAMP9d
GAMP101'
GAMPlld
WW-14
-WW-18
WW-16
WW-15
WW-17
X X
HPAH, LPAH, HG. PCBs.
ZN [TOTXYLENE]
CU, HG, PCBs, ZN
AS, CU, HG. ZN,
PCBs [CR]
HPAH, PB. HG, ZN. PCBs
LPAH. HPAH. HG
HPAH. AS. HG. ZN
HPAH. LPAH. HG. ZN
HPAH. PCBs. HG
PCBs. LPAH, HG. ZN [CD]
PB. HG. ZN, PCBs, DDD
LPAH. HPAH. PCBs. HG. ZN
[PHNL]
LPAH. PCBs, DMP, HG. ZN
a LPAH - Signifies AET exceedances for the sum of naphthalene, acenaphthylene. acenaphthene, fluorene, phenanthrene, and anthracene.
or any of these compounds Individually. To simplify the presentation of AET exceedances In this table, exceedances of
AET for 1-methylphenanthrene, 2-methylnaphthalene, biphenyl, and dlbenzofuran are Included under LPAH. These compounds
covaried with LPAH but are not Included in LPAH sums.
HPAH - Signifies AET exceedances for the sum of fluoranthene, pyrene, benzo(a)anthracene, chrysene, total benzofluoranthenes,
benzo(a)pyrene, 1ndeno(l,2,3-cd)pyrene, dibenzo(a,h)anthracene, and benzo(g,h,iiperylene, pr of any of these compounds
individually.
-------
TABLE 41. (Continued)
8NZ0H - benzyl alcohol
0NZAC1O - benzoic acid
1.4DICLBNZ - 1,4-dichlorobenzene
BUTBNZPH - butyl benzyl phthalate
DMP - dimethyl phthalate
TOC - total organic carbon
OINOCT - dl-n-octyl phthalate
TOTXYLENE - total xylenes
CR - chromium
ENDALD - endrln aldehyde
PCP - pentachlorophenol
More detailed Information on exceedances Is presented In Appendix F.
CU - copper
PB - lead
ZN - zinc
HG - mercury
CD - cadmium
AG - silver
AS - arsenic
PHNL - phenol
4MEPHNL - 4-methylphenol
2MEPHNL - 2-methylphenol
2.4MEPHNL - 2,4-dimethylphenol
b Chemicals exceeding HAET for Puget Sound.
c Chemicals exceeding LAET for Puget Sound. More detailed information on exceedances is presented in Appendix F. Chemicals
shown In brackets exceeded 90th-percentt1e concentrations but did not exceed any AET. Chemicals exceeding both AET and 90th-
percentile concentrations are not bracketed.
d Gaiqponla et al. (1986).
e Romberg et al. (1984).
-------
concentration of 8,730 mg/kg DW. A nearby historical station had a similarly
high concentration (10,600 mg/kg DW) (Gamponia et al. 1986), as well as
copper and zinc concentrations near or above HAET values. Lead concentra-
tions decreased moving down the east side of the waterway toward
Station WW-11 (721 mg/kg DW). Outside of this area, however, concentrations
appeared to decrease sharply in all directions. Stations located between
WW-11 and WW-14 were predominantly fine-grained (>60 percent fine-grained
material in al1 cases).
PAH contamination was more widespread than lead contamination and had a
less evident relationship to potential sources. HPAH concentrations
typically | ranged from 20,000 to 40,000 ug/kg along the east side of the
waterway and remained at a relatively high level in transects of historical
stations across the waterway. Potential sources of PAH on the east side of
the waterway include the ARCO and Lockheed facilities and the Lander Street
discharges discussed previously. The SW Florida CS0/SD (098) near Station
WW-17 on the west side of the waterway is a potential source of PAH for that
area. Overall, however, PAH distributions are generally elevated in the
problem area and do not strongly suggest specific localized sources.
Among the other contaminants exceeding HAET in this problem area, PCBs
decreased from 3,700 to 2,300 going from east to west in a cross-waterway
transect (Gamponia et al. 1986). However, concentrations at the same
stations reported in another study were roughly 5 times lower (U.S. EPA
1982, 1983). Overall, PCB concentrations did not show any strong gradients,
although a relatively high concentration was observed at Station WW-09
(1,500 ug/kg DW). Another problem chemical in this area, p,p'-DDD, occurred
at 80 ug/kg at Station WW-09, but was undetected at nearby Stations WW-10
and WW-11] at detection limits below 20 ug/kg DW. Although AET were not
applied to chromium for reasons discussed earlier, it is notable that the
chromium concentration at Station WW-12 (555 mg/kg DW) exceeded the 90th
percentile concentration for the study.
East Waterway (EW) Problem Area
!
The East Waterway problem area appeared to be dominated by contamina-
tion with diverse metals and organic chemicals at Station EW-05 and had other
relatively isolated areas of local contamination (primarily by PAH). Station
EW-05, located in the middle of the waterway near the discharge point of the
Hanford CS0 (W032), had concentrations exceeding HAET for PCBs, cadmium,
mercury, butyl benzyl phthalate, p,p1-DDE, p,p1-DDT, and LPAH (represented by
1-methyl phenanthrene) (Table 42). Sediment at Station EW-05 was also
enriched in organic matter (7.39 percent T0C content) and had "oil and
grease" and water-soluble sulfide concentrations that were among the highest
observed in the study. Station EW-06 was geographically isolated from
Station EW-05 but had some similar patterns of chemical contamination (e.g.,
relatively high concentrations of PCBs, cadmium, and mercury).
1
Certain chemical evidence suggests that contamination apparently
originating near Station EW-05 is dispersed throughout the waterway.
Correlation analyses revealed a relatively large number of strong cor-
relations ' among chemicals in this area (see RESULTS, Spatial Correlations
Among Chemicals), suggesting that these contaminants could have derived from
the same source and were subject to similar transport processes. For
259
-------
TABLE 42. EAST WATERWAY PROBLEM AREA*
HAET Exceedances1
Station
LPAH
HPAH PCBs H6 CD DOT/DDE/ODD
BNZOH BUTBNZPH
LAET Exceedances®
EW-02
EW-04
EW-05
EW-06
A062d
C062d
EW-07
EW-08
ml0039e
S0039d
EW-09
EW-11
EW-10
EW-12
EW-13
EW-14
S0064d
EW-15
ElAf
EW-16
XXX
PCBs. HPAH. HQ
PCBs, LPAH. HG
HPAH, PB, ZN [CHLOROANE,
RETENE, AG. OINOCT]
PCBs. BUTBNZPH, CD. HG. ZN
PCBs. CU. BUTBNZPH
PCBs, PB, ZN, ODD
PCBs, HPAH, HG, ZN
PCBs, HG
PCBs. AS
PCBs. BUTBNZPH. HPAH
PCBs. HPAH, HG, ZN
PCBs. LPAH. HPAH, H6, ZN
[BUTBNZPH. DINOCT]
PCBs. BUTBNZPH, HG [DINOCT]
PCBs, HPAH
PCBs. BUTBNZPH
PCBs. BUTBNZPH. HG
PCBs. HG
BUTBNZPH, HPAH, HG. ZN [AS]
HPAH
BUTBNZPH
LPAH - Signifies AET exceedances for the sum of naphthalene, acenaphthylene. acenaphthene, fluorene, phenanthrene, and
anthracene, or any of these compounds individually. To simplify the presentation of AET exceedances in this table,
exceedances of AET for 1-methylphenanthrene. 2-methylnaphthalene, biphenyl, and dlbenzofuran are included under
LPAH. These compounds covaried with LPAH but are not included in LPAH sums.
HPAH - Signifies AET exceedances for the sum of fluoranthene, pyrene, benzo(a)anthracene, chrysene, total benzofluoranthenes,
benzo(a)pyrene, indeno(l,2,3-cd)pyrene, dibenzo(a,h)anthracene, and benzo(g,h,i )perylene, or of any of these compounds
individually.
CU - copper
PB - lead
ZN - zinc
HG - mercury
CD - cadmium
AG - silver
AS - arsenic
PHNL - phenol
4MEPHNL - 4-methylphenol
2MEPHNL - 2-methylphenol
2.4MEPHNL - 2,4-dimethyl phenol
BNZOH - benzyl alcohol
BNZACID - benzoic acid
1.4DICLBNZ - 1,4-dlchlorobenzene
BUTBNZPH - butyl benzyl phthalate
DMP - dimethyl phthalate
TOC - total organic carbon
DINOCT - di-n-octyl phthalate
TOTXYLENE - total xylenes
CR - chromium
ENDALD - endrin aldehyde
PCP - pentachlorophenol
D Chemicals exceeding HAET for Puget Sound. More detailed information on exceedances is presented in Appendix F.
c Chemicals exceeding LAET for Puget Sound. More detailed information on exceedances is presented in Appendix P. Chemicals
shown in brackets exceeded 90th-percentile concentrations but did not exceed any AET. Chemicals exceeding both AET and 90th-
percentile concentrations are not bracketed.
d Romberg et al. (1984).
e Mai ins et al. (1982).
f U.S. EPA (1982, 1983).
260
-------
example, PCBs and cadmium correlated well in the waterway (as did PCBs and
lead, and PCBs and silver), and may have dispersed from the area of their
maximum concentrations (Station EW-05). Also, PCB contamination was
relatively constant on an organic carbon normalized basis (see Figure 37),
again suggesting a uniform source. The apparent decrease in TOC normalized
PCB concentrations toward the mouth of the waterway (see Figure 37) suggests
that the most contaminated particles are not reaching the mouth of the
waterway or are being diluted considerably there. In addition, metals that
were most concentrated at Station EW-05 (e.g., cadmium and mercury)
correlated well with TOC content (e.g., r>0.8) and were likely transported
with the organic material originating in that area. Notably, there metals
correlated poorly with percent fine-grained material (e.g., r<0.4).
Pesticide contamination by p,p'-DDT and related compounds appeared to
be localized in the area between TPPS Stations A062 and S0039. Relatively
low detection limits were reported even within this area. Chlordane has no
established AET but exceeded its 90th percentile concentration at Station
EW-05. Interference from high PCB concentrations may have contributed to the
high pesticide concentrations at this station.
Butyl benzyl phthalate concentrations exceeded the HAET in an area
encompassing Stations EW-05 to EW-11. Station EW-05 is not strongly indi-
cated as the source region for this compound, as concentrations were higher
at Stations EW-07, EW-08, and EW-09.
Unlike most problem chemicals in the East Waterway problem area, PAH
did not appear to originate in the area near Station EW-05. Instead, PAH
were elevated at various relatively -isolated regions throughout the study
area, in some cases near storm drain (SD) or CSO discharges. An area-wide
gradient in PAH concentrations was not apparent. Station EW-02, located near
the SW Hinds CSO/SD (107), had an LPAH concentration (32,000 ug/kg DW) that
exceeded the HAET. This station had a distinctly low ratio of HPAH/LPAH
(<1; see Figure 33) indicative of a petroleum source. Station EW-04, on the
west side of the waterway near the SW Hanford CSO/SD (162) and other storm
drains, had a relatively high concentration of HPAH (34,000 ug/kg DW).
Station EW-06 had HAET exceedances for both LPAH and HPAH. Station EW-14,
near the mouth of the waterway, had very elevated concentrations of LPAH
(13,000 ug/kg DW) and HPAH (69,000 ug/kg DW - the highest HPAH sum in the
East Waterway). Stations EW-07 and EW-08 had high detection limits for LPAH
(particularly naphthalene) that impeded data interpretation in that area.
Benzyl;alcohol exceeded the HAET at Station EW-12 (870 ug/kg DW), which
is located hear the SW Florida SD. However, detection limits at comparable
levels at nearby stations did not allow for further characterization.
Problem Area KG11
Distributions of problem chemicals (predominantly polar and nonpolar
organic compounds) within this area were heterogeneous. The stations in this
area were combined because they are relatively closely spaced together and
because the transport of certain compounds throughout the area (PCBs in
particular) cannot be discounted. Problem chemical distributions are
summarized in Table 43. Three general areas of contamination were observed:
Station KG-06, which had the most HAET exceedances of any station in the
261
-------
TABLE 43. KELLOGG ISLAND PROBLEM AREAS3
Area
Station
•
HAET Exceedances'*
LAET
Exceedances6
LPAH
HPAH
PCBS DOT PHNL 4MEPHNI
BNZACID
BUTBNZPH
KGI
KG-01
E12*
X
X
ZN, HPAH, PCBs,
BUTBNZPH, H6
[CO. DINOCT]
ZN, PCBs
(CGI I
KG-05
PB, HG, ZN, PCBs
[CO]
KG-06
X
X X
X
HG, ZN
(2METHNAP)*
[RETENE, AS]
KG-08
PCBs
KG-09
X
PCBs
S0037f
X
PCBs, HPAH
U1339
X
PCBs, HG, ZN
KG-10
X
X
[RETENE]
(2HETHNAP)e
KG-11
PCBs [4MEPHNL]
KG-03
BUTBNZPH,
PCBs, ZM
LPAH - Signifies AET exceedances for the sum of naphthalene, acenaphthylene, acenaphthene, fluorene, phenanthrene, and
anthracene, or any of these compounds individually. To simplify the presentation of AET exceedances in this table,
exceedances of AET for 1-methylphenanthrene, 2-methy1 naphthalene, blphenyl, and dlbenzofuran are Included under
LPAH. These compounds covaried with LPAH but are not Included in LPAH sums.
HPAH - Signifies AET exceedances for the sum of fluoranthene, pyrene, benzo(a)anthracene, chr-ysene, total benzofluoranthenes,
benzo(a)pyrene, 1ndeno< 1,2,3-cd)pyrene, dibenzo< a,h)anthracene, and benzo(g,h,1 )perylene, or of any of these compounds
individually.
CU - copper
PB - lead
ZN - zinc
HG - mercury
CO - cadmium
AG - silver
AS - arsenic
PHNL - phenol
4MEPHNL - 4-methylphenol
2MEPHNL - 2-methyl phenol
2.4MEPHNL - 2,4-dlmethylphenol
BNZQH - benzyl alcohol
BNZACID - benzoic acid
1.40ICLBNZ - 1,4-dlchlorobenzene
BUTBNZPH - butyl benzyl phthalate
OMP - dimethyl phthalate
TOC - total organic carbon
DINOCT - di-n-octyl phthalate
TOTXYLENE - total xylenes
CR - chromium
ENDALD - endrlnaldehyde
PCP - pentachlorophenol
b Chemicals exceeding HAET for Puget Sound. Hore detailed Information on exceedances Is presented In Appendix F.
c Chemicals exceeding LAET for Puget Sound. More detailed Information on exceedances Is presented 1n Appendix F. Chemicals
shown in brackets exceeded 90th-percent11e concentrations but did not exceed any ACT. Chemicals exceeding both AET and
90th-percentile concentrations are not bracketed.
d U.S. EPA (1982, 1983).
e 2-Methyl naphthalene (2METHNAP) was the only PAH-related compound to exceed HAET at this station.
f Romberg et al. 1984).
9 Stober and Chew (1984).
262
-------
area; Statfion KG-05, which was characterized by LAET exceedances of a number
of metal s;j and six stations between Stations KG-08 and KG-11, which were
contaminated with a variety of oxygenated organic compounds including phenol,
4-methylphenol, and benzoic acid.
Station KG-06 had the highest PCB concentration in the problem area
(3,100 ug/kg DW). Although no PCB concentration gradient was readily
apparent, a number of historical stations in the problem area had concentra-
tions between 500 and 800 ug/kg DW. The p,p'-DDT contamination at
Station KG-j-06 (270 ug/kg DW) appeared relatively isolated, as the pesticide
was undetected at most stations within the problem area at detection limits
of less than 10 ug/kg DW. It is possible that GC/ECD interferences from PCB
contamination at Station KG-06 resulted in an overestimation of p,p'-DDT.
Butyl benzyl phthalate at Station KG-06 was over an order of magnitude
higher than at any other station in the problem area; however, interpretation
is impeded! because several samples were blank-corrected down to detection
limits. 2-Methylnaphthalene, at 1,900 ug/kg DW, was the only PAH that
exceeded a! HAET at Station KG-06; this concentration was more than double
the sum of LPAH at Station KG-06. High detection limits for 2-methyl-
naphthalene (e.g., the extreme detection limit of 8,900 ug/kg at
Station KG-05) and for LPAH in general impede interpretation of their
distribution in the area around Stations KG-04, KG-05, KG-06, and KG-07.
Station KG-05, located adjacent to the Diagonal Way CS0/SD (111), had
no HAET exceedances but had 90th percentile exceedances of cadmium (4.1 mg/kg
DW) and lead (500 mg/kg DW). Mercury (1.63 mg/kg DW) exceeded the LAET at
Station KG-05 and was over 5 times higher than at adjacent intertidal
Station KG-j04; both sediment samples (i.e.,-KG-04 and KG-05) were coarse-
grained.
The northern portion of the problem area was characterized mostly by
contamination by polar organic compounds, although different problem
chemicals vjrere observed at different problem stations. Station KG-09,
located next to the SW Dakota SD, had a 4-methyl phenol concentration
exceeding the HAET (1,500 ug/kg DW). This compound was undetected at nearby
stations but was detected above the 90th percentile concentration at
Station KG-11 (610 ug/kg DW). Historical Station U133 (Stober and Chew
1984), located between Station KG-09 and intertidal Station KG-10, had a
phenol concentration exceeding the HAET (2,200 ug/kg DW); phenol was
detected at 400 ug/kg DW at Station KG-10.
Station KG-10 was unique in that it had the highest retene concentration
in the study (10,000 ug/kg DW), the only detection of benzoic acid
(6,300 ug/kg DW), and a very high organic carbon content (10 percent)
despite a relatively coarse grain size distribution (17 percent fine-grained
material). These attributes suggest that this sample may contain coal
particles. , Transport of coal particles in the lower Green-Duwamish River
system has Been suggested by Hamilton et al. (1984); the coal likely derives
from exposed coal seams along the river (Barrick et al. 1984). Retene is a
useful geochemical marker for sub-bituminous and lignite coal found in the
Green River[area (Barrick et al. 1984). Retene was reported at 1,700 ug/kg
DW at Statidn KG-06, but was undetected or detected at low concentrations at
stations between Station KG-10 and KG-06.
263
-------
Problem Area NSII
This group of four stations in the vicinity of Piers 90 and 91 was
characterized by a relatively small number of problem chemicals, largely
consisting of PAH and other organic compounds (Table 44). However, con-
taminant composition at the four stations was not uniform. The most severe
PAH contamination occurred at Station NS-07, which had a HPAH concentration
of 40,000 ug/kg DW and an anomalously high HPAH/LPAH ratio (8.7, as compared
to ratios typically less than 4 in this area). PAH [specifically 1-methyl-
phenanthrene and indeno(l,2,3-cd)pyrene] also exceeded HAET at Station NS-08,
although overall PAH concentrations were considerably lower than at Station
NS-07 (e.g., HPAH = 12,000 ug/kg DW). Station NS-06 did not exceed AET for
PAH; detection limits for LPAH were very high at this station (e.g., >1,000
ug/kg DW for most individual LPAH) but were acceptable for HPAH. Di-n-octyl
phthalate exceeded its 90th percentile concentration at Station NS-06.
Intertidal Station NS-04 had a 4-methylphenol concentration that exceeded
the HAET (1,300 ug/kg DW), although detected concentrations and detection
limits at nearby stations did not exceed 10 ug/kg DW. Pentachlorophenol,
which has no established AET, was detected at 330 ug/kg DW at this station.
Problem Area NSI
Only one sample was collected in this area for the present study (inter-
tidal Station NS-01). The problem area boundary is otherwise defined by
historical TPPS stations. The sole problem chemical at Station NS-01 was
silver (8.27 mg/kg DW), which also exceeded HAET at a number of historical
stations (Table 44). Other chemicals exceeding HAET at historical TPPS
stations were PAH, PCBs, several chlorinated pesticides, and mercury
(Table 44). The area near Denny Way CS0 has been characterized previously
with these data (Romberg et al. 1984).
A recent study (Romberg et al. 1987) provides results for a more
intensive sampling of sediments near the Denny Way CS0 outfall. Although
these data were not included in the present document, the data presented by
Romberg et al. (1987) are generally consistent with the data used for
problem area identification. However, silver was not measured by Romberg
et al. (1987).
Upper Duwamish River Problem Areas
The upper Duwamish River (defined as Area DR in this study) did not
demonstrate gradients for high priority contaminants that were clearly
indicative of large-scale transport processes throughout the area; instead,
heavy contamination appeared to be localized and, in some cases, in close
proximity to potential sources. For this reason, seven separate problem
areas containing one or two stations were identified; these areas were
separated from one another by less contaminated stations. Most of the
problem stations were located in relatively quiescent areas (i.e., three
areas were within slips and one area consisted of an intertidal station).
Problem chemicals occurring at the seven problem areas are summarized in
Table 45. In most cases, the highest priority problem chemicals (i.e.,
those based on HAET exceedances) differ among problem areas in the upper
Duwamish River area, even for the problem areas nearest one another.
264
-------
TABLE 44. SEATTLE NORTH WATERFRONT PROBLEM AREAS3
HAET Exceedances"
Area
Station
LPAH
HPAH
PCBs HG
AG
DDT/00E/00D
4MEPHNL
LAET Exceedancesc
NSII
NS-04
X
NS-07
X
LPAH, PCBs, HG [PHNL]
NS-06
[DINOCT]
NS-08
X
X
BUTBNZPH
MSI
NS-01
X
S0031d
X
X
X
X
PCBs
S0032d
X
X
X
PCBs
1406d
X
X
PCBs, HG
1603d
X
X
X
X
PB, HG, ZN
1606*®
X
X
X
X
HG, HPAH
1706d
X ,
PCBs, HG, LPAH, HPAH
1810d
X
X
PCBs, HG, LPAH
1612d
X
X
X
PCBs, LPAH
1512d
X
X
PCBs, HG
1830d
X
PCBs, HG
LPAH
HPAH
- Signifies AET exceedances for the- sun of naphthalene, acenaphthylene. acenaphthene, fluorene, phenanthrene,
and anthracene, or any of these compounds individually. To simplify the presentation of AET exceedances In this
table, exceedances of AET for 1-methylphenanthrene, 2-methylnaphthalene, blphenyl, and dlbenzofuran are included
under LPAH. These compounds covarled with LPAH but are not included 1n LPAH sums.
- Signifies AET exceedances for the sum.of fluoranthene, pyrene, benzo(a)anthracene, chrysene, total benzofluoran-
thenes, benzo(a)pyrene, 1ndeno(l,2,3-cd)pyrene, d1benzo(a,h)anthracene, and benzo(g,h,1Iperylene, or of any of
these compounds individually.
CU - copper
PB - lead
ZN - zlnic
HG - mercury
CD - cadmium
AG - silver
AS - arsenic
PHNL - phenol
4MEPHNL -4-methylphenol
2MEPHNL - 2-methyl phenol
2.4MEPHNL • 2,4-dimethylphenol
BNZOH • benzyl alcohol
BNZACID • benzoic acid
1.40ICLBNZ - 1,4-dlchlorobenzene
BUTBNZPH - butyl benzyl phthalate
DW> • dimethyl phthalate
TOC - total organic carbon
DINOCT - dl-n-octyl phthalate
TOTXYLENE - total xylenes
CR • chromium
ENOALD - endrln aldehyde
PCP - pentachlorophenol
6 Chemicals exceeding HAET for Puget Sound. More detailed information on exceedances Is presented In Appendix F.
c Chemicals exceeding LAET for Puget Sound. More detailed information on exceedances is presented In Appendix F.
Chemicals shown in brackets exceeded 90th-percent11e concentrations but did not exceed any AET. Chem1c»1s exceeding
both AET and 90th-percentile concentrations are not bracketed.
d Romberg et al. (1984).
265
-------
TABLE 45. UPPER DUWAMISH RIVER PROBLEM AREAS*
HAET Exceedances**
Area
Station
LPAH
HPAH
PCBs HG DDT/DDE
2MEPHNL
DMP
LAET Exceedances0
DR-25
X
X
4MEPHNL, BUTBNZPH, PHNL, ZN
[DINOCT]
DR-05
X
PCBs
DRI
DR-08
X
X X
BUTBNZPH, HG, ODD, ZN
[CO, ENOALD]
DR-IO
OR-12
DR-15
DR-16
E19d
HP AH, ZN
PCBs, ODD
AS, CU, PB, 2N, PCBs, HPAH
PCBs [OINOCT]
LPAH, BUTBNZPH, PCBs,
ZN [AS]
HPAH
LPAH - Signifies AET exceedances for the sum of naphthalene, acenaphthylene, acenaphthene, fluorene, phenanthrene, and
anthracene, or any of these compounds Individually. To simplify the presentation of AET exceedances In this table,
exceedances of AET for 1-methylphenanthrene, 2-methylnaphthalene, blphenyl, and dlbenzofuran are included under
LPAH. These compounds covarled with LPAH but are not included In LPAH sums.
Signifies AET exceedances for the sum of fluoranthene, pyrene, benzo(a)anthracene, chrysene, total
benzofluoranthenes, benzo(a)pyrene, indeno(l,2,3-cd)pyrene, dibenzo(a,h)anthracene, and benzo(g,h,1tperylene,
or of any of these compounds Individually.
BNZOH - benzyl alcohol
BNZACID - benzoic acid
1.4DICLBNZ - 1,4-dlchlorobenzene
BUTBNZPH - butyl benzyl phthalate
BMP - dimethyl phthalate
TOC - total organic carbon
OINOCT - di-n-octyl phthalate
TOTXYLENE - total xylenes
CR - chromium
ENOALD - endrln aldehyde
PCP - pentachlorophenol
More detailed infortaation on exceedances Is presented In Appendix F.
c Chemicals exceeding LAET for Puget Sound. More detailed information on exceedances Is presented 1n Appendix F. Chemicals
shown in brackets exceeded 90th-percent11e concentrations but did not exceed any AET. Chemicals exceeding both AET and
90th-percent11e concentrations are not bracketed.
d U.S. EPA (1982, 1983).
CU -
PB -
ZN -
HG -
CO -
AG -
AS -
PHNL
4MEPHNL
2MEPHNL
copper
lead
zinc
mercury
cadmium
silver
arsenic
- phenol
- 4-methylphenol
- 2-methyl phenol
2.4MEPHNL - 2,4-dlmethyl phenol
b Chemicals exceeding HAET for Puget Sound.
-------
Problem Ar[ea DR-25--
Most |problem chemicals at this intertidal station were polar organic
compounds j (phenols). 2-Methylphenol exceeded its HAET, and phenol and
4-methylphenol exceeded their LAET (concentrations for the former two com-
pounds were the highest observed in the study). All three phenols were
undetected at low detection limits (<6 ug/kg DW) at the nearest stations
upriver arid downriver from this station. Station DR-25 is located near the
S. 96th Street SD.
Problem Area DR-05--
The only compound to exceed HAET at this station was p,p'-DDT (33 ug/kg
DW). This compound was undetected at reasonable detection limits (less than
7 ug/kg ' DW) at stations downriver (intertidal Station DR-06 and
Station DR-07). PCBs were reported at 570 ug/kg DW at this station.
Station DR-05 is located adjacent to the 16th Avenue S. SD.
Problem Area DRI (Slip 4)--
A problem area was defined by Station DR-08 and historical EPA
Station 19; (U.S. EPA 1982, 1983) based primarily on PCB contamination
(5,600 to 5,800 ug/kg DW). PCB concentrations decreased toward the mouth of
Slip 4 to 490 ug/kg DW and concentrations at the nearest stations within the
river were lower. PAH were also problem chemicals in Slip 4 and appeared
highest all the head and lowest at the mouth. Mercury and, zinc followed
similar patterns. In addition, p,p1-DDE was detected only at Station DR-08.
A number of potential .sources occur at the head of Slip 4 including; the
Georgetown| Flume, the Slip 4 CS0/SD (117), the Slip 4 SD, the 1-5 SD, and
the East Marginal PS GS0.
Problem Area DR-10--
Station DR-10, located near the 2nd Avenue S. SD, had p,p'-DDT and
p,p1-DDE concentrations exceeding HAET values (both pesticides occurred at
roughly 65 ug/kg DW). A related pesticide, p,p'-DDD, exceeded the LAET at
this station, as did PCBs (2,100 ug/kg DW). All three pesticides were
undetected at detection limits of 10 or less at stations upriver and
downriver from DR-10. PCBs were generally elevated in this area of the
river (see; Figure 35). It is notable that Station DR-10, which had the
highest "oil and grease" concentration in the study, had low overall PAH
concentrations (LPAH + HPAH = 4,500 ug/kg DW).
Problem Area DR-12--
Station DR-12, at the head of Slip 3, was defined as a problem area
because three metals (arsenic, zinc, and copper) exceeded their
90th percentile concentrations at this station. Arsenic was detected at
449 mg/kg DW, zinc at 969 mg/kg DW, and copper at 386 mg/kg DW. Lower
concentrations of all three metals were reported at a nearby historical EPA
Station E15 (U.S. EPA 1982, 1983). The arsenic concentration, 449 mg/kg DW,
was the second highest observed in the present study. Concentrations at the
nearest stations in the river were at least several times lower than at
267
-------
Station DR-12. LAET were exceeded for HPAH and PCBs at this station. There
are no known discharges at the head of Slip 3.
Problem Area DR-15--
Station DR-15, located in Slip 2, was included as a problem area
because of a 90th percentile exceedance for di-n-octyl phthalate (310 ug/kg
DW; this is a blank corrected value). PCBs, which exceeded the LAET at this
station, were reported at 230 ug/kg DW.
Problem Area DR-16--
Station DR-16, located near the SW Graham SD, exceeded the HAET for one
HPAH compound [indeno(l,2,3-cd)pyrene]. LPAH and HPAH concentrations at
this station were higher than concentrations at historical stations upriver
and were roughly 4-7 times the concentrations at the nearest station
downriver (DR-17),
Problem Area KGI (Slip 1)
High priority problem chemicals in this area were LPAH and HPAH,
although several metals were also elevated (see Table 43). The two adjacent
samples that established this problem area were not in complete agreement in
terms of contaminant concentrations: historical EPA Station 12 (U.S. EPA
1982, 1983) was heavily contaminated with LPAH (42,000 ug/kg DW) and HPAH
(94,000 ug/kg DW), whereas concentrations at nearby Station KG-01 were
considerably lower (LPAH = 2,400 ug/kg DW; HPAH = 15,000 ug/kg DW). PAH
concentrations at Station KG-01 were consistent with concentrations observed
at the mouth of the slip. Station KG-01 and EPA Station 12 had comparable
concentrations of zinc and cadmium, both of which exceeded their 90th percen-
tile concentrations at Station KG-01 and decreased sharply toward the mouth
of the slip. A potential source is not readily apparent in this area.
Problem Area KG-Q3
Station KG-03 was included among the Tier II problem areas because of
exceedance of action level guidelines for biological variables. Contamina-
tion at this station was not extreme, although several chemicals exceeded
LAET [PCBs (300 ug/kg DW), zinc (275 mg/kg DW), and butyl benzyl phthalate
(66 ug/kg DW)]. Because of the lack of a clear association of the contami-
nation at this station with contamination in the other KG problem areas, and
because there is a lack of available data for the area between KG-03 and
Problem Area KGII, this station has not been combined with Problem Area KGII.
Problem Area AB-01
This problem area is defined by a single station (AB-01) with anomalous-
ly high concentrations of mercury and PAH (particularly LPAH) that exceeded
HAET. In addition, copper and zinc concentrations exceeded LAET at this
station. The mercury concentration at this station, 28.8 mg/kg DW-, was the
highest observed at any station in the study area (including historical
stations). Mercury concentrations at the nearest stations (including
intertidal Station NH-10) were over 100 times lower. PAH concentrations
were also lower at nearby stations, but the decreases were not as sharp as
268
-------
those for mercury (LPAH decreased from 10,000 ug/kg at Station AB-01 to
3,400 at Station AB-02). Station AB-01 is located near the discharge point
of the Fairmount Avenue SW CS0 (078).
Problem Area NH-10
This i|ntertida1 station was classified among the Tier II problem areas
because of 90th percentile exceedances of two chlorinated pesticides
(aldrin, at 90 ug/kg DW, and dieldrin, at 51 ug/kg DW). These pesticides;do
not have established AET values. The LAET for PCBs was also slightly
exceeded this station (160 ug/kg DW). Station NH-10 was not combined in
the problem area for nearby subtidal Station AB-01 because the two stations
differed considerably with regard to chemical contamination (e.g., the
concentrations of the predominant problem chemicals at Station AB-01 were
one to two orders of magnitude lower at Station NH-10).
SUMMARY
Identification of Problem Areas
¦ The nearshore region of the Elliott Bay system inside a line
from Pier 91 to Duwamish Head displayed significant elevations
of both sediment contaminants and liver lesions in English
sole. This entire area was designated as a Tier I problem
area.
¦ Seventy-two stations were designated as higher priority
problem areas (Tier II). Sixty-one of these stations were
grouped into the following multi-station problem areas: DRI
(Slip 4), EW (East Waterway), KGI and KGII (near Kellogg
Island), NHI and NHII (North Harbor Island, the mouth of the
West Waterway, and west to just beyond Longfellow Slough
outlet), NSI (Denny Way CS0), NSII (Pier 90/91), SS (Seattle
South waterfront), WWI (southern segment of West Waterway),
and WWII (northern segment of West Waterway).
Ranking of Problem Areas
¦ Ranking of Tier II problem areas and stations identified five
areas (SS, NHI, NHII, HUI, and HHII) and 29 single stations as
the highest priority sites. Of the latter, the following
stations were outside the five highest priority areas:
Stations NS-01, EW-05, AB-01, KG-01, KG-05, KG-06, DR-12,
DR-15, and DR-16.
Characterization of Problem Areas
¦ Problem Area SS~This area is highly contaminated overall
(especially by PAH), with localized patches of extreme
contamination. Stations SS-08 and SS-09 in this area account
for the highest concentrations of several metals (e.g., lead,
cadmium, zinc) and organic compounds (e.g., PAH, 1,4-dichloro-
benzene) observed in this study; maximum concentrations of
other metals occurred at other stations in this problem area
269
-------
(e.g., arsenic at Station SS-03). Sediments in this area were
enriched in organic matter, with TOC concentrations typically
greater than 5 percent and as high as 27 percent. Benthic
infaunal depressions were found at Stations SS-04, SS-09, and
SS-08, with greater than 80 percent depressions of Pelecypoda
at the first two sites (P<0.001). Significant sediment
toxicity to amphipods was found at Station SS-06, where
mortality was 45 percent (P<0.001). Overall, the SS area did
not exhibit the widespread biological effects that might be
expected based on consideration of the chemistry data.
¦ Problem Area NHI—Problem Area NHI, which encompasses two
shipyard facilities at the mouth of the West Waterway, was
heavily contaminated with several metals (most notably copper,
but also mercury, lead, zinc, and arsenic), PAH, and PCBs.
The most elevated assemblages of organic compounds and metals
were observed near Station NH-03 and two historical stations
located east of that station. On the west side of the West
Waterway mouth, Station NH-04 contained high concentrations
of copper, PCP, and PAH, among other chemicals. Problem Area
NHI exhibited severe depressions in the abundances of major
taxa of benthic infauna (especially Pelecypoda) at all
stations. Abundances of all four major taxa evaluated were
severely depressed at Station NH-03. Significant amphipod
mortality above 85 percent was found at Stations NH-03 and
NH-04.
¦ Problem Area NHII—PAH and related compounds (including
alkylated PAH, carbazole, and biphenyl) were the predominant
contaminants in Problem Area NHII. The most extreme contami-
nation was observed at Stations NH-06 and NH-08, located near
the Wyckoff facility and the outflow of Longfellow Slough,
respectively. These two stations were very similar in terms
of PAH composition and concentration. The high concentrations
of carbazole (particularly at Station NH-06) and PAH in this
area are consistent with a creosote source. Benthic infaunal
effects in Problem Area NHI I were less than in most other
problem areas, although Station NH-08 exhibited severe
depressions of pelecypods and crustaceans. Overall sediment
toxicity was highest in this area, with Station NH-08
displaying 100 percent mortality.
¦ Problem Area WWI—Chemical contamination was severe but
patchy in this area, and included a relatively isolated but
high historical concentration of PCBs in the southwest corner
of the waterway, a high benzyl alcohol concentration at
Station WW-02, and a high concentration of PAH at
Station WW-04. Effects on benthic infauna were moderate
overall for this area. However, Station WW-03 was one of two
stations in this study where severe (>80 percent) depressions
in abundances of all four major taxa were observed. Sediment
toxicity in area WWI was generally low, except at
Station WW-02, where 82 percent mortality was observed.
270
-------
¦ Problem Area HWII~The most distinctive chemical feature of
this area was extreme lead contamination along the east side
of the West Waterway (near the SW Lander Street discharges).
PAH concentrations were generally elevated, but gradients or
pronounced maxima were not apparent. PCB contamination was
moderate and patchy in this area. Severe effects on benthic
infauna were restricted to depressions in the abundances of
pelecypods and crustaceans at several stations within this
area. Sediment toxicity was relatively low in this area
overall. Nevertheless, mean amphipod mortality above
40 percent was observed at Stations WW-09 and WW-11.
271
-------
REFERENCES
Aggerholm, D.A. 22 February 1988. Personal Communication (letter with
enclosed data to Ms. Clare Ryan, U.S. Environmental Protection Agency
Region X, Seattle, WA). Port of Seattle, Seattle, WA.
Barrick, R.C., and F.G. Prahl. 1987. Hydrocarbon geochemistry of the Puget
Sound region. III. Polycyclic aromatic hydrocarbons in sediments. Estuar.
Coastal Shelf Sci. 25:175-191.
Barrick, R.C., E.T. Furlong, and R. Carpenter. 1984. Hydrocarbon and
azaarene markers of coal transport to aquatic sediments. Environ. Sci.
Technol. 18:846-854.
Battelle. 1985. Data report on Elliott Bay action plan: Puget Sound.
Appendix 1. Prepared for the U.S. Environmental Protection Agency Region X.
Battelle Washington Environmental Program Office, Washington, DC.
Battelle. 1986. Reconnaissance survey of eight bays in Puget Sound.
Final Report. Vol. I and II. Prepared for the U.S. Environmental Protection
Agency, Region X. Battelle Pacific Northwest Division, Marine Research
Laboratory, Sequim, WA.
Becker, D.S., T.C. Ginn, M.L. Landolt, and D.B. Powell. 1987. Hepatic
lesions in English sole (Paroohrvs vetulus) from Commencement Bay, Washington
(USA). Mar. Environ. Res. 23:153-173.
Bilyard, G.R. 1987. The value of benthic infauna in marine pollution
monitoring studies. Mar. Pollut. Bull. 18:581-585.
Boesch, D.F. 1977. Applications of numerical classification in ecological
investigations of water pollution. EPA-600/3-77-033. U.S. Environmental
Protection Agency, Environmental Research Laboratory, Cprvallis, OR. 125 pp.
Bouwman, L.A., K. Romeijn, and W. Admiraal. 1984. On the ecology of
meiofauna in an organically polluted estuarine mudflat. Estuar. Coast. Shelf
Sci. 19:633-655.
Carriker, M.R., J.W. Anderson, W.P. Davis, D.R. Franz, G.F, Mayer, J.B.
Pearce, T.K. Sawyer, J.H. Tietjen, J.F. Timoney, and D.R. Young. 1982.
Effects of pollutants on benthos, pp. 3-21. In: Ecological Stress and the
New York Bight: Science and Management. G. F. Mayer (ed). Estuarine
Research Federation, Columbia, SC.
Chan, S., M. Schiewe, and D. Brown. (Unpublished). Analyses of sediment
samples for U.S. Army Corps of Engineers East, West, and Duwamish Waterways
navigation improvement project, operations and maintenance sampling and
testing of Duwamish River sediments.
111
-------
Chan, S., M. Schiewe, and D. Brown. 1985. Analysis of sediment samples for
U.S. ArmyJ Corps of Engineers Seattle Harbor navigation project, operations
and maintenance sampling and testing of Duwamish River sediments. Draft
Report. 15 pp. + appendices.
Chapman, P.M., M.A. Farrell, R.M. Kocan, and M.L. Landolt. 1982a. Marine
sediment toxicity tests in connection with toxicant pretreatment planning
studies, Metro Seattle. E.V.S. Consultants, Vancouver, B.C. 15 pp.
Chapman, jP.M. G.A. Vigers, M.A. Farrell, R.N. Dexter, E.A. Quinlan, R.M.
Kocan, and M.L. Landolt. 1982b. Survey of biological effects of toxicants
upon Puget Sound biota. I. Broad scale toxicity survey. NOAA Technical
Memorandum OMPA-25. National Oceanic and Atmospheric Administration,
Boulder, CO. 98 pp.
Chiou, C.t. 1985. Partition coefficients of organic compounds in lipid-
water systems and correlations with fish bioconcentration factors. Environ.
Sci. Technol. 19:57-62.
Choi, W.W., and K.Y. Chen. 1976. Associations of chlorinated hydrocarbons
with fine| particles and humic substances in nearshore surficial sediments.
Environ. Sci. Technol. 10:782-786.
Clark, R.C., and D.W. Brown. 1977. Petroleum: properties and analyses in
biotic an'd abiotic systems. In: Effects of Petroleum on Arctic and
Subarctic i Marine Environments and Organisms. Vol. 1. D.C. Malins (ed).
Academic Ifress, New York, NY.
Comiskey,. C.E., T.A. Farmer, C.C. Brandt, and G.P. Romberg. 1984. Puget
Sound benthic studies and ecological implications. Technical Report C2,
Toxicant Pretreatment Planning Study. Municipality of Metropolitan Seattle,
Seattle, WA. 373 pp.
Crecelius,; E.A. 5 December 1986. Personal Communication (letter to Tetra
Tech, Incj., Bellevue, WA). Battelle Pacific Northwest Division, Marine
Research Laboratory, Sequim, WA.
Crecelius, E.A. 2 March 1987. Personal Communication (phone by Ms.
Deborah Coffey, Tetra Tech, Inc.). Battelle Pacific Northwest Division,
Marine Research Laboratory, Sequim, WA.
Creceliusr E.A., M.H. Bothner, and R. Carpenter. 1975. Geochemistry of
arsenic, antimony, mercury, and related elements in sediments of Puget
Sound. Enlviron. Sci. Technol. 9:325-33.
DeWitt, TiH., G.R. Ditsworth and R.C. Swartz. (In Press). Effects of
natural sediment features on survival of the phoxocephalid amphipod,
Rhepoxvnius abronius. Submitted to Marine Environmental Research.
Dillon, T:M. 1984. Long-term effects of dredging operations program,
biological consequences of bioaccumulation in aquatic animals: an assessment
of the current literature. Final Report. Technical Report D-82-2.
National Technical Information Service, Springfield, VA.
273
-------
Dorsey, J.H., K.D. Green, and R.C. Rowe. 1983. Effects of sewage disposal
on the polychaetous annelids at San Clemente Island, California, pp. 209-
233. In: Waste Disposal in Oceans. D.F. Soule and 0. Walsh (eds).
Westview Press, Boulder, CO.
Evans-Hamilton and D.R. Systems. 1987. Puget Sound environmental atlas.
Prepared for U.S. Environmental Protection Agency, Puget Sound Water Quality
Authority, and U.S. Army Corps of Engineers, Seattle, WA. Evans-Hamilton.
Seattle, WA.
Fauchald, K., and P. Jumars.- 1979. The diet of worms: a study of poly-
chaete feeding guilds. Oceanogr. Mar. Biol. Annu. Rev. 17:193-284.
Fresi, E., M.C. Gambi, S. Focardi, R. Bargagli, F. Baldi, and L. Falciai.
1983. Benthic community and sediment types: a structural analysis. Mar.
Ecol. 4:101-121.
Gahler, A.R., R.L. Arp, and J.M. Cummins. 1982. Chemical contaminants in
edible non-salmonid fish and crabs from Commencement Bay, Washington. U.S.
Environmental Protection Agency Region X, Environmental Services Division,
Seattle, WA 117 pp.
Gakstatter, J. 24 November 1987. Personal Communication (letter to Dr.
William Brownlie, Tetra Tech, Inc.). U.S. Environmental Protection Agency
Region X, Seattle, WA.
Gakstatter, J. 10 December 1987. Personal Communication (letter to Dr.
William Brownlie, Tetra Tech, Inc.). U.S. Environmental Protection Agency
Region X, Seattle, WA.
Gamponia, V., T. Hubbard, P. Romberg, T. Sample, and R. Swartz. 1986.
Identifying hot spots in the lower Duwamish River using sediment chemistry
and distribution patterns. Municipality of Metropolitan Seattle, Seattle,
WA.
Gebhart, J. 30 December 1987. Personal Communication (phone by Dr. Jean
Jacoby, Tetra Tech, Inc.). Battelle Columbus Division, Columbus
Laboratories, Columbus, OH.
Gray, J.S. 1974. Animal-sediment relationships. Oceanogr. Mar. Biol. Ann.
Rev. 12:223-261.
Gray, J.S. 1982. Effects of pollutants on marine ecosystems. Neth. J. Sea
Res. 16:424-443.
Green, E.J., and D. Schnitker. 1974. The direct titration of water-soluble
sulfide in estuarine muds of Montsweag Bay, Maine. Mar. Chem. 2:111-124.
Hamilton, S.E., T.S. Bates, and J.D. Cline. 1984. Sources and transport of
hydrocarbons in the Green-Duwamish River, Washington. Environ. Sci. Technol.
18:72-79.
Johnson, R.G. 1971. Animal-sediment relations in shallow water benthic
communities. Marine Geol. 11:93-104.
274
-------
Krahn, M.M., L.D. Rhodes, M.S. Myers, L.K. Moore, W.D. MacLeod, and D.C.
Malins. 1986. Associations between metabolites of aromatic compounds in
bile and Sthe occurrence of hepatic lesions in English sole (Parophrvs
vetul_us) from Puget Sound, Washington. Arch. Environ. Contam. Toxicol.
15:61-67.
Krone, C.Aj, O.G. Burrows, D.W. Brown, P.A. Robisch, A.J. Friedman, and D.C.
Malins. 1986. Nitrogen-containing aromatic compounds in sediments from a
polluted harbor in Puget Sound. Environ. Sci. Technol. 20:1144-1150.
Lambshead, P.J.D. 1986. Sub-catastrophic sewage and industrial waste
contamination as revealed by marine nematode faunal analysis. Mar. Ecol.
Prog. Ser. 29:247-260.
Landolt, M.L., D.B. Powell, and R.M. Kocan. 1984. Renton sewage treatment
plant project. Seahurst baseline study. Volume VI'I, Section 8. Fish
health. University of Washington Fisheries Research Institute, Seattle, WA.
160 pp.
Landolt, MiL.r F.R. Hafer, A. Nevissi, G. van Belle, K. Van Ness, and C.
Rockwell. 1985. Potential exposure among consumers of recreationally
caught fish from urban embayments of Puget Sound. NOAA Technical Memorandum
N0S-0MA-23. National Oceanic and Atmospheric Administration, RockviTle, MD.
104 pp.
Lee, C.B. 1985. Sedimentary processes of fine sediments and the behavior of
associated metals in the Keum Estuary, Korea, pp.. 209-225. In: Marine and
Estuarine Geochemistry. A.C. Sigleo and A. Hattori (eds). Lewis Publishers,
Inc., Chelsba, MI.
Long, E.R. (Unpublished). Sediment bioassays: a summary of their use in
Puget Sound!. Ocean Assessments Division, National Oceanic and Atmospheric
Administration, Seattle, WA.
Mackay, D. 1982. Correlation of bioconcentration factors. Environ. Sci.
Technol. 16:274-278.
Malins, D.C:, B.B. McCain, D.W. Brown, A.K. Sparks, and H.0. Hodgins. 1980.
Chemical contaminants and biological abnormalities in central and southern
Puget Sound. NOAA Technical Memorandum OMPA-2. National Oceanic and
Atmospheric[Administration, Rockville, MD. 295pp.
Malins, D.Ci, B.B. McCain, D.W. Brown, A.K. Sparks, H.0. Hodgins, and S.L.
Chan. 1982. Chemical contaminants and abnormalities in fish and inver-
tebrates from Puget Sound. NOAA Technical Memorandum OMPA-19. National
Oceanic and [Atmospheric Administration, Boulder, CO.
Malins, D.C.j, B.B. McCain, D.W. Brown, S.L. Chan, M.S. Myers, J.T. Landahl,
P.G. Prohaska, A.J. Friedman, L.D. Rhodes, D.G. Burrows, W.D. Gronlund, and
H.0. Hodgins. 1984. Chemical pollutants in sediments and diseases of
bottom-dwelling fish in Puget Sound, WA. Environ. Sci. Technol. 18:705-713.
275
-------
McCain, B.B., K.V. Pierce, S.R. Wellings, and B.S. Miller. 1977. Hepatomas
in marine fish from an urban estuary. Bull. Environ. Contain. Toxicol.
18:1-2.
McCain, B.B., M.S. Myers, U. Varanasi, D.W. Brown, L.D. Rhodes, W.D.
Gronlund, D.G. Elliott, W.A. Palsson, H.O. Hodgins, and D.C. Malins. 1982.
Pathology of two species of flatfish from urban estuaries in Puget Sound.
EPA-600/7-82-001. National Oceanic and Atmospheric Administration and U.S.
Environmental Protection Agency, Washington, DC. 100 pp.
McFarland, V.A. 1984. Activity-based evaluation of potential bioaccumula-
tion from sediments. pp. 461-466. In: Dredging and Dredged MateriaT
Disposal, Vol. 1. R.L. Montgomery and J.W. Leach (eds). American Society
of Civil Engineers, New York, NY.
Mearns, A., R. Swartz, J. Cummins, D. Dinnell, P. Plesha and P. Chapman.
1986. Inter-laboratory comparison of a sediment toxicity test using the
marine amphipod Rhepoxvnius abronius. Mar. Environ. Res. 19:13-37.
Municipality of Metropolitan Seattle. 1987. Duwamish Head Baseline Study.
Metro, Seattle, WA.
Mowrer, J., J. Calambokidis, N. Musgrove, B. Drager, M.W. Beug, and S.G.
Herman. 1977. Polychlorinated biphenyls in cottids, mussels, and sediment
in southern Puget Sound, Washington. Bull. Environ. Contam. Toxicol.
18:588-594.
Myers, M.S., L.D. Rhodes, and B.B. McCain. 1987. Pathologic anatomy and
patterns of occurrence of hepatic neoplasms, putative preneoplastic lesions
and other idiopathic hepatic conditions in English sole (Paroohrvs vetulus)
from Puget Sound, Washington, USA. J. Natl. Cancer Inst. 78:333-363.
Neilson, K.K. 1977. Matrix corrections for energy dispersive x-ray
fluorescence analyses of environmental samples with coherent/incoherent
scattered x-rays. Anal. Chem. 49:641-648.
Nestler, F.H.M. 1974. The characterization of wood-preserving creosote by
physical and chemical methods of analysis. USDA Forest Service Research
Paper FPL 195. U.S. Department of Agriculture, Forest Products Laboratory,
Madison, WI.
Nevissi, A., S. Felton, and R. McClain. 1984. Renton sewage treatment plant
project. Seahurst baseline study. Volume IX, Section 11. Marine chemistry
- trace metals in marine water, biota and sediments. University of Washing-
ton Fisheries Research Institute, Seattle, WA. 125 pp.
Nichols, J.A. 1977. Benthic community structure near the Woods Hole sewage
outfall. Int. Revue Ges. Hydrobiol. 62:235-244.
Osborn, J.G., D.E. Weitkamp, and T.H. Schadt. 1985. Alki wastewater
treatment plant outfall improvements predesign study. Technical Report No.
6.0. Marine Biology. Municipality of Metropolitan Seattle, Seattle, WA.
50 pp.
276
-------
Pearson, t• H., and R. Rosenberg. 1978. Macrobenthic succession in relation
to organic enrichment and pollution of the marine environment. Oceanogr.
Mar. Biol.I Annu. Rev. 16:229-311.
Phillips, |D.J.H. 1980. Quantitative aquatic biological indicators: their
use to monitor trace metal and organochlorine pollution. Applied Science
Publishers, Ltd., London, UK. 488 pp.
Pierce, KIV.,'B.B. McCain, and S.R. Wellings. 1978. The pathology of
hepatomas land other liver abnormalities in English sole (Paroohrvs vetulus)
from the D'uwamish River Estuary, Seattle, Washington. J. Natl. Cancer Inst.
60:1445-14149.
Prahl, F.G., and R. Carpenter. 1979. Role of zooplankton fecal pellets in
the sedimentation of polycyclic aromatic hydrocarbons in Dabob Bay, Washing-
ton. Geochim. Cosmochim. Acta 43:1959-1972.
Prahl, F.G|., and R. Carpenter. 1983. PAH-phase associations in Washington
coastal sediment. Geochim. Cosmochim. Acta 47:1013-1023.
PTI Environmental Services and Tetra Tech. 1988. Everett Harbor action
program: analysis of toxic -problem areas. Draft Report. Prepared for the
U.S. Environmental Protection Agency Region X, Office of Puget Sound. Tetra
Tech, Inc.i, Bellevue, WA.
Raman, A.V., and P.N. Ganapati. 1983. Pollution effects on ecobiology of
benthic polychaetes in Visakhapatnam Harbor (Bay of Bengal). Mar. Poll.
Bull'. 14:46-52.
Readman, J.W., R.F.C. Mantoura, M.M. Rhead, and L. Brown. 1982. Aquatic
distribution and heterotrophic degradation of polycyclic aromatic hydrocar-
bons (PAH) in the Tamar Estuary. Estuar. Coast. Shelf Sci. 14:369-389.
Reish, D.J|., and J.L. Barnard. 1979. Amphipods (Arthopoda: Crustacea:
Amphipoda): pp. 345-370. In: Pollution Ecology of Estuarine Invertebrates.
C.W. Hart,;Jr., and L.H. Fuller, (eds). Academic Press, New York, NY.
Rhodes, L.D., M.S. Myers, W.D. Gronlund, and B.B. McCain. 1987. Epizootic
characteristics of hepatic and renal lesions in English sole, Paroohrvs
vetulus, from Puget Sound. J. Fish Biol. 31:395-407.
Ricker, W.E. 1975. Computation and interpretation of biological statistics
of fish populations. Bull. Fish. Res. Board Can. No. 191.
Romberg, G.P., S.P. Pavlou, R.F. Shokes, W. Horn, E.A. Crecelius, P.
Hamilton, jj.T. Gunn, R.D. Muaench, and J. Vinelli. 1984. Presence,
distribution, and fate of toxicants in Puget Sound and Lake Washington.
Metro Toxicant Program Report No. 6A. Toxicant Pretreatment Planning Study
Technical Report CI. Municipality of Metropolitan Seattle, Seattle, WA.
231 pp.
Romberg, GiP., D. Healy, and K. Lund. 1987. Toxicant reduction in the
Denny Way jcombined sewer system. Municipality of Metropolitan Seattle,
Seattle, WA;.
277
-------
Rygg, B. 1985. Effect of sediment copper on benthic fauna. Mar. Ecol.
Prog. Ser. 25:83-89.
Rygg, B. 1986. Heavy-metal pollution and log-normal distribution of
individuals among species in benthic communities. Mar. Pollut. Bull. 17:31-
36-
Sanders, H.L. 1960. Benthic studies in Buzzards Bay. III. The structure
of the soft bottom community. Limnol. Oceanogr. 5:138-153.
Santos, J.F., and J.D. Stoner. 1977. Physical, chemical, and biological
aspects of the Duwamish River Estuary, King County, Washington, 1963-67.
Environmental quality. Geological Survey Water Supply Paper 1873-C. U.S.
Geological Survey, Reston, VA. 74 pp.
Shannon, C.E., and W. Weaver. 1949. The mathematical theory of communica-
tion.' University of Illinois Press, Urbana, IL. 117 pp.
Shepard, F.P. 1954. Nomenclature based on sand-silt-clay ratios.
J. Sediment. Petrol. 24:151-158.
Sloan, R., M. Brown, R. Brandt, and C. Barnes. 1985. Hudson River PCB
relationships between resident fish, water and sediment. Northeastern
Environ. Sci. 3:137-151.
Sokal, R.R., and F.J. Rohlf. 1981. • Biometry. 2nd Edition. W.H. Freeman
and Co., San Francisco, CA. 859 pp.
Speis, D.N. 1980. Determination of purgeable organics in sediment using a
modified purge-and-trap technique. U.S. Environmental Protection Agency
Region II, Edison, NJ.
Stober, Q.J., and K.K. Chew. 1984. Renton sewage treatment plant project.
Duwamish Head. Final report for the period 1 July to 31 December 1984.
University of Washington Fisheries Research Institute, Seattle, WA. 370 pp.
Stoner, J.D. 1967. Prediction of salt-wedge intrusion in the Duwamish River
Estuary, King County, Washington. Geological Survey Professional Paper
575D. U.S. Geological Survey, Reston, VA. 3 pp.
Stoner, J.D., W.L. Hauschild, and J.B. McConnell. 1975. A numerical model
of material transport in salt-wedge estuaries. Part II. Model computation
of salinity and salt-wedge dissolved oxygen in the Duwamish River Estuary,
King County, Washington. Geological Survey Professional Paper 917. U.S.
Geological Survey, Reston, VA. 36 pp.
Swartz, R.C., W.A. D?Ben, K.A. Sercu, and J.O. Lamberson. 1982. Sediment
toxicity and the distribution of amphipods in Commencement Bay, Washington,
USA. Mar. Pollut. Bull. 13:359-364.
278
-------
Swartz, R.C., W.A. DeBen, J.K.P. Jones, J.O. Lamberson, and F.A. Cole.
1985. Phoxocephalid amphipod bioassay for marine sediment toxicity, pp.
284-307. In: Aquatic Toxicology and Hazard Assessment: Seventh Symposium.
R.D. Cardwell, R. Purdy, and R.C. Bahner (eds). ASTM STP 854. American
Society of Testing and Materials, Philadelphia, PA.
Tetra Tech. 1984. A decision-making approach for the Commencement Bay
Nearshore/ Tideflats Superfund Project. Prepared for the Washington State
Department of Ecology. Tetra Tech, Inc., Bellevue, WA.
Tetra Tech. 1985a. Commencement Bay nearshore/tideflats remedial investi-
gation. Final Report. EPA-910/9-85-134b. Prepared for the Washington State
Department |of Ecology and U.S. Environmental Protection Agency Region X,
Office of Puget Sound. Tetra Tech, Inc., Bellevue, WA. 2 volumes +
appendices.
Tetra Tech. 1985b. Elliott Bay toxics action program: interim work plan.
Final Report. Prepared for the U.S. Environmental Protection Agency Region
X and Washington State Department of Ecology. Tetra Tech, Inc., Bellevue,
WA. 35 pp.
Tetra Tech. 1985c. Elliott Bay toxics action program: review of existing
action plans. Final Report. Prepared for the U.S. Environmental Protection
Agency Region X, Office of Puget Sound. Tetra Tech, Inc., Bellevue, WA.
Tetra Tech. ; 1985d. Quality assurance project plan for field investigations
to support development of the Elliott Bay toxics action plan. Draft Report.
Prepared fof U.S. Environmental Protection Agency Region X, Seattle, WA.
Tetra Tech, [Bellevue, WA.
Tetra Tech. , 1985e. Sampling and analysis design for development of Elliott
Bay toxics action plan. Final Report. Prepared for the U.S. Environmental
Protection ^9ency Region X, Office of Puget Sound. Tetra Tech, Inc.,
Bellevue, WA. 69 pp.
Tetra Tech. 1986a. Analytical methods for U.S. EPA priority pollutants and
301(h) pesticides in estuarine and marine sediments. Final Report. Prepared
for U.S. Environmental Protection Agency, Office of Marine and Estuary
Protection, Washington, DC.
Tetra Tech. 1986b. Bioaccumulation monitoring guidance: 4. Analytical
methods for U.S. EPA priority pollutants and 301(h) pesticides in tissues
from estuarine and marine organisms. Final Report. Prepared for U.S.
Environmental Protection Agency, Office of Marine and Estuarine Protection,
Washington, DC.
Tetra Tech. 1986c. Development of sediment quality values for Puget Sound.
Final Report). Prepared for Resource Planning Associates under U.S. Army
Corps of Engineers, Seattle District, for the Puget Sound Dredged Disposal
Analysis and|Puget Sound Estuary Programs. Tetra Tech, Inc., Bellevue, WA.
128 pp. + appendices.
279
-------
Tetra Tech. 1986d. Eagle Harbor preliminary investigation. Final Report.
Prepared for Black & Veatch, Engineers-Architects and the Washington State
Department of Ecology. Tetra Tech, Inc., Bellevue, WA. 192 pp. + appen-
dices.
Tetra Tech. 1986e. Elliott Bay toxics action program: initial data
summaries and problem identification. Final Report. Prepared for U.S.
Environmental Protection Agency Region X, Seattle, WA. Tetra Tech, Bellevue,
WA.
Tetra Tech. 1986f. Recommended protocols for measuring selected environmen-
tal variables in Puget Sound. Final Report. Prepared for the U.S. Environ-
mental Protection Agency and U.S. Army Corps of Engineers. Tetra Tech,
Inc., Bellevue, WA.
Tetra Tech. 1986g. Recommended protocols for sampling and analyzing
subtidal benthic macroinvertebrate assemblages in Puget Sound. Final
Report. Prepared for the U.S. Environmental Protection Agency Region X,
Office of Puget Sound. Tetra Tech, Inc., Bellevue, WA. 34 pp. + appendices.
Tetra Tech. 1987. Commencement Bay nearshore/tideflats feasibility study:
development of sediment criteria. Final Draft Report. Prepared for the
Washington State Department of Ecology and U.S. Environmental Protection
Agency. Tetra Tech, Inc., Bellevue, WA.
Tetra Tech. 1988a. Elliott Bay action program: data quality assurance
assessment. Final Report. Prepared for the U.S. Environmental Protection
Agency Region X, Office of Puget Sound. Tetra Tech, Inc., Bellevue, WA.
Tetra Tech. 1988b. Elliott Bay action program: evaluation of potential
contaminant sources. Final Report. Prepared for the U.S. Environmental
Protection Agency Region. X, Office of Puget Sound. Tetra Tech, Inc.,
Bellevue, WA.
Tetra Tech. 1988c. Elliott Bay action program: evaluation of sediment
remedial alternatives. Draft Report. Prepared for U.S. Environmental
Protection Agency Region X, Office of Puget Sound. Tetra Tech, Inc.,
Bellevue, WA.
Tetra Tech and E.V.S. Consultants. 1986. Recommended protocols for
conducting laboratory bioassays on Puget Sound sediments. Final Report.
Prepared for the U.S. Environmental Protection Agency Region X, Office of
Puget Sound. Tetra Tech, Inc., Bellevue, WA. 55 pp.
Trial, W., and J. Michaud. 1985. Alki wastewater treatment plant outfall
improvements predesign study. Technical Report No. 8.3. Water Quality.
Municipality of Metropolitan Seattle, Seattle, WA. 89 pp.
U.S. Department of the Navy. 1985. Final environmental impact statement.
Carrier Battle Group Puget Sound region ship homeporting project. Volume 2.
Technical Appendix. Prepared for U.S. Department of the Navy, Western
Division, Naval Facilities Engineering Command, San Bruno, CA.
280
-------
U.S. Environmental Protection Agency. 1982. Organic analyses for Duwamish
River surveys, September 1982. Unpublished data. U.S. EPA, Region X,
Seattle, WA. 10 pp.
U.S. Environmental Protection Agency. 1983. Organic analyses for Duwamish
River surveys, July 1983. Unpublished data. U.S. EPA, Region X, Seattle,
WA. 10 ppJ
U.S. Environmental Protection Agency. 1986. U.S. EPA Contract Laboratory
Program statement of work for organic analysis, multi-media, multi-concentra-
tion. U.S., EPA, Washington, DC.
Van Es, F.B., M.A. Van Arkel, L.A. Bouwman, and H.G.I. Schroeder. 1980.
Influence of organic pollution on bacterial, macrobenthic, and meiobenthic
populations1 in the intertidal flats, of the Dollard. Neth. J. Sea. Res.
14:288-304.
Winer, B.J.; 1971. Statistical principles in experimental design. McGraw-
Hill Book Co., New York, NY. 907 pp.
Wolfe, D.A., D.F. Boesch, A. Calabrese, J.J. Lee, C.D. Litchfield, R.J.
Livingston, A.D. Michael, J.M. O'Connor, M. Pilson, and L.V. Sick. 1982.
Effect of toxic substances on communities and ecosystems, pp. 67-86. In:
Ecological stress and the New York Bight: Science and Management. G.F.
Mayer (ed). Estuarine Research Federation, Columbia, SC.
Word,. J.Q. , 1980. Classification of benthic invertebrates into infaunal
trophic index feeding groups, pp. 103-121. In: Coastal Water Research
Project, Biennial Report for the Years 1979.-1980. W. Bascom (ed). Southern
California Coastal Water Research Project, Long Beach, CA.
Word, J.Q., B.L. Myers, and A.J. Mearns. 1977. Animals that are indicators
of marine pollution, pp. 199-206. In: Coastal Water Research Project
Annual Reporjt. W. Bascom (ed). Southern California Coastal Water Research
Project, El Segundo, CA.
Word, J.Q., P.L. Striplin, K. Keeley, J. Ward, P. Sparks-McConkey, L.
Bentler, S. Hulsman, K. Li, J. Schroeder, and K. Chew. 1984. Renton sewage
treatment plant project. Seahurst baseline study. Volume V, Section 6.
Subtidal benthic ecology. University of Washington Fisheries Research
Institute, Seattle, WA. 461 pp.
281
------- |