Puget Sound Estuary Program
PUGET SOUND
MONITORING PROGRAM
A PROPOSED PLAN
NOVEMBER 1986
DRAFT REPORT
PREPARED FOR:
U.S. ENVIRONMENTAL PROTECTION AGENCY
REGION X - OFFICE OF PUGET SOUND
SEATTLE, WASHINGTON
-------
Draft Report
TC 3338-05
PUGET SOUND MONITORING PROGRAM:
A PROPOSED PLAN
by
Tetra Tech, Inc.
and
E.V.S. Consultants
for
U.S. Environmental Protection Agency
Region X
Seattle, Washington
November, 1986
Tetra Tech, Inc.
11820 Northup Way, Suite 100
Bellevue, Washington 98005
-------
CONTENTS
Page
LIST OF FIGURES iv
LIST OF TABLES vii
EXECUTIVE SUMMARY ix
INTRODUCTION 1
GENERAL APPROACH TO THE DEVELOPMENT OF A MONITORING PROGRAM 5
MANAGEMENT GOALS AND MONITORING OBJECTIVES 5
KEY CONCEPTS 7
CRITERIA FOR PROBLEM IDENTIFICATION 8
IMPORTANT STATISTICAL CONSIDERATIONS 10
ONGOING MONITORING 11
MONITORING COMPONENTS IN RELATION TO GOALS AND OBJECTIVES 12
USES OF MONITORING DATA 18
AMBIENT MONITORING PROGRAM 22
OVERVIEW 22
SAMPLING STATIONS 24
ONGOING MONITORING 31
COMPLIANCE MONITORING PROGRAMS 33
INTENSIVE SURVEYS ' 37
NONPOINT SOURCE MONITORING 39
INTEGRATED WATERSHED MONITORING 39
BIOLOGICAL MONITORING OF NONPOINT SOURCES 42
ii
-------
DATABASE MANAGEMENT SYSTEM 46
DATA MANAGEMENT SYSTEM GOALS 46
KEY ISSUES 47
SYSTEM COMPARISONS 48
OPTIONS 49
MONITORING PROGRAM REPORTS 57
TECHNICAL REPORTS 57
MONITORING PROGRAM SUMMARY REPORT 58
REPORT CONTENT 59
INSTITUTIONAL MECHANISMS FOR PROGRAM MANAGEMENT AND IMPLEMENTATION 62
CONSIDERATIONS FOR ENHANCING COORDINATION 62
GENERAL INSTITUTIONAL STRUCTURE 63
APPROACHES TO PROGRAM MANAGEMENT 64
APPROACHES TO PROGRAM IMPLEMENTATION 69
PREFERRED APPROACH 74
PROGRAM MODIFICATION AND PHASED IMPLEMENTATION 83
ESTIMATED COSTS OF MONITORING PUGET SOUND 89
REFERENCES 92
APPENDIX A: POWER ANALYSES A-l
APPENDIX B: AGENCY-SPECIFIC USES OF MONITORING DATA B-l
APPENDIX C: DETAILED DESIGN CONSIDERATIONS FOR THE AMBIENT
MONITORING PROGRAM C-l
iii
-------
FIGURES
Following
Number Page
1 General organization and uses of Puget Sound monitoring
data 2
2 General approach to development of the Puget Sound
Monitoring Program 5
3 Locations of Puget Sound basins and bays 24
4 Proposed locations for sediment quality triad and water
quality components 25
5 Proposed locations of monitoring stations in Elliott Bay,
an example urban bay 26
6 Proposed locations of monitoring stations in Carr Inlet,
an example rural bay 26
7 Proposed locations of bottomfish sampling stations for
the ambient monitoring program 27
8 Proposed locations of bottomfish sampling stations for
Elliott Bay, an example urban bay 27
9 Proposed locations of bottomfish sampling stations for
Carr Inlet, an examle rural bay 27
10 Proposed locations of sampling stations for toxic chemicals
in fish from recreatinal harvest areas 28
11 Proposed locations of intertidal sampling stations for
shellfish monitoring components of the ambient monitoring
program 28
12 Primary downstream station locations on major rivers in
the Puget Sound drainage basin 30
13 Skagit River Basin 30
14 Green-Duwamish River Basin 30
15 Components of Integrated Watershed Monitoring Program 39
16 Summary of Puget Sound Monitoring Program reports 57
IV
-------
17 An example of a single variable mapped within a study
area: percent fines in sediments of Everett Harbor 60
18 An example of a simple bar chart 60
19 An example of a correlation plot 60
20 An example of concentrations of a variable profiled with
water depth at seven stations 60
21 An example box chart illustrating temporal trends 60
22 An example line graph of mean concentration or mean value
of a variable as a function of distance from a 301(h)
permittee's outfall 60
23 An example of spatial patterns plotted using two-dimensional
point symbols 60
24 An example of a scatterplot 60
25 An example of using shading density on a map to indicate
concentration, value, or priority 60
26 General structure for managing and implementing the Puget
Sound Monitoring Program 63
27 Agency ambient monitoring responsibilities under imple-
mentation Option 1 70
28 Agency ambient monitoring responsibilities under imple-
mentation Option 3 72
29 Cumulative costs of the ambient monitoring program as
components are added 89
Al Minimum detectable difference versus number of replicates
at selected levels of unexplained variance for 4 and 6
stations A-8
A2 Minimum detectable difference versus number of replicates
at selected levels of unexplained variance for 8 and 16
stations A-8
Cl Statistical sensitivity of the amphipod bioassay test as
a function of the number of replicates Oil
C2 Sample sizes required to detect one individual affected
with a lesion with 95 percent confidence, given various
population sizes and prevalences £-39
C3 Effects of sample size on the minimum detectable prevalence
at a test site relative to the prevalence at the reference
site c_39
-------
C4 Flow and nutrient loading in Skagit River C-66
C5 Nutrient loading in the Green-Duwamish River C-69
C6 Summary of copper, lead, and zinc concentrations in
Springbrook Creek C-69
C7 Hydrographs for mountain streams and lowland rivers in
Puget Sound Basin C-71
C8 Calculated estimates of suspended sediment loading compared
with actual measured loading C-78
C9 Comparisons of metals rating curves at two stations in
the Green-Duwamish River Basin C-81
CIO Suspended sediment rating curves for Snoqualmie River near
Carnation C-82
VI
-------
TABLES
Following
Number Page
1 Criteria proposed for components of the Puget Sound
Monitoring Program 10
2 Relationship of proposed monitoring components to
management goals and monitoring objectives 12
3 Summary of the ambient monitoring program 22
4 Proposed numbers of sampling stations to be assigned to
Puget Sound basins for monitoring of sediment quality,
water quality, and benthic macroinvertebrates 25
5 Proposed numbers of sampling stations to be assigned to
urban and industrialized bays for monitoring of sediment
quality, water quality, and benthic macroinvertebrates 25
6 Proposed numbers of sampling stations to be assigned to
rural bays for monitoring of sediment quality, water
quality, and benthic macroinvertebrates 25
7 Summary of Puget Sound river monitoring stations 30
8 Ongoing ambient monitoring programs in Puget Sound 31
9 Summary of scores from an evaluation of database
management approaches 48
10 Rough cost estimates (in thousands of dollars) for several
approaches to developing a data management system for
Puget Sound monitoring data 49
11 Matrix of relationships between pairs of monitoring
components 60
12 Estimated annual costs (in thousands of dollars) of the
Puget Sound monitoring program 89
13 Estimated annual field and laboratory costs for ambient
monitoring components 89
14 Estimated annual costs (in thousands of dollars) of ongoing
ambient monitoring programs for Puget Sound 90
VII
-------
Al Analysis of variance for one-way layout A-5
Cl List of target chemicals for sediment analyses C-2
C2 Locations of sampling stations for shellfish and pathogen
indicators in water C-27
C3 Water quality variables to be measured in the river
monitoring program C-57
C4 Flow ranking of rivers discharging into Puget Sound C-62
C5 Receiving water characteristics for principle Puget Sound
rivers C-62
C6 Skagit River Basin characteristics C-65
C7 Green-Duwamish River Basin characteristics C-68
C8 Puget Sound discharge characteristics C-70
C9 Suspended sediment loading estimation procedures evaluated
by Walling and Webb (1985) C-78
vm
-------
EXECUTIVE SUMMARY
INTRODUCTION
Information from monitoring programs is useful to regulatory agencies
for making a variety of decisions concerning resource management. However,
present efforts to monitor physical, chemical, and biological conditions
in Puget Sound are incomplete and uncoordinated. As a result, an interagency
effort was initiated by the U.S. Environmental Protection Agency (EPA),
the Washington Department of Ecology, and the Puget Sound Water Quality
Authority (PSWQA) to design and implement a comprehensive monitoring program
for Puget Sound. The focus of this working paper is on the design of the
proposed monitoring program, potential database management systems, and
institutional options and costs for implementation of the proposed program.
The structure of the proposed program and general uses of the data are
illustrated in the figure on the following page.
GENERAL APPROACH
The first step in designing the monitoring program was to establish
general goals for the program, define uses of monitoring data, and develop
focused monitoring program objectives to meet data needs. The three general
management goals of the monitoring program are:
t Ensure that Puget Sound is acceptable for human use
• Ensure that Puget Sound will adequately support productive
and diverse biological communities
• Improve the effectiveness of regulatory programs and management
decisions.
-------
USES OF
DATA
ASSESS
ENVIRONMENTAL
RESOURCES/PROBLEMS
ir
DEVELOP
MANAGEMENT
SOLUTIONS
-*•
EVALUATE
MANAGEMENT
EFFECTIVENESS
DATA
PUGET SOUND MONITORING DATABASE
DATA
COLLECTION
COMPLIANCE MONITORING
AMBIENT MONITORING
POINT
SOURCES
NONPOINT
SOURCES
MONITORING
DESIGN
INTENSIVE SURVEYS
SEDIMENT
QUALITY
SITE-SPECIFIC
DESIGNS
WATER
QUALITY
BIOLOGICAL
CONDITIONS
RIVERS
HABITATS
E
V
A
L
U
A
T
E
M
O
N
T
O
R
I
N
G
D
E
S
I
G
N
General organization and uses of Puget Sound monitoring data.
-------
The relationships between monitoring program components (i.e., groups of
related variables) and the goals and objectives of the program are summarized
in the text. A preliminary list of uses of monitoring data includes:
• Protection of human health
• Assessment of environmental impacts
• Evaluation of program effectiveness
• Design of remedial action
• Enforcement of criteria, standards, and permit limits
• Management of biological resources
• Characterization of baseline conditions
• Interpretation of data.
The remainder of the report is devoted to describing the proposed program.
AMBIENT MONITORING PROGRAM
The ambient monitoring program involves periodic sampling of the water,
sediment, and biota at specified stations, and collection of ancillary
data on human uses of Puget Sound resources. Major features of the ambient
program are summarized in the table on the following page. Much of the
recommended program is designed to fill gaps in existing monitoring efforts.
The monitoring program components are as follows:
0 Sediment quality, water quality, and benthic macroinverte-
brates
• Bottom fish
-------
SUMMARY OF THE AMBIENT MONITORING PROGRAM
Number of Stations
Components
Sediment Qual i ty
Sediment chemistry
Sediment toxicity bioassays
Conventional sediment variables
Water Quality
Hydrographic conditions
Dissolved oxygen
Turbidity/ transparency
Odor, floatables, etc. offshore
Odor, floatables, etc. - intertidal
Nutrient concentrations
Phytoplankton standing stock
Pathogen indicators in water
Biological Conditions
Benthic macroinvertebrate abundances
Toxic chemicals in fish - English sole
Toxic chemicals in fish - Cod and Salmon
Histopathological abnormalities in fish
Fish species abundances
Shellfish abundances
Toxic chemicals in shellfish
PSP in shellfish
Pathogen indicators in shellfish
Marine mammal abundances and reprod. success
Avian abundances and reproductive success
Urban and
Main Industrial
Basins Bays
31
31
31
11
11
11
11
16
11
11
16
16
31
16
3
16
16
16
16
16
16
16
16
-
- -
34
34
34
15
15
15
15
5
15
15
5
5
34
27
15
27
27
5
5
5
5
5
5
-Sound-wide-
-Sound-wide-
Rural
Bays
41
41
41
27
27
27
27
5
27
27
5
5
41
19
0
19
19
5
5
5
5
5
5
- - -
-
Sampl
Frequency
Annual
Annual
Annual
Monthly
Monthly
Monthly
Monthly
Weekly
Monthly
Monthly
Weekly
Monthly
Annual
Biennial
Annual
Biennial
Biennial
Annual
Annual
Weekly
Monthly
Weekly
Monthly
Annual
Monthly
ing Program
Number of
Timi ng Repl icates
March
March
March
Daytime
Daytime
Daytime
Daytime
May- July
Daytime
Daytime
May-July
Aug-April
March
July
Sep-Oct
July
July
May
May
May-July
Aug-April
May-July
Aug-April
June-Aug
Dec-Jan^
1 composite
3x5a
1 composite
1 prof i le
1 profile
5 readings
-b
1 composite
1 composite
3 composites
3 composites
5 grabs
3 composites
3 composites
60 specimens
4 trawls
15 cores
3 composites
3 composites
3 composites
3 composites
3 composites
1 survey0
1 survey0
River Monitoring
Habitat Types
8 primary* 12 times/yrf
- - 13 primary* 12 times/yrf
- - 11 secondary*— 12 times/yrf
Continuous 1 composite
3-yr period 1 composite
3-yr period 1 composite
— - -Sound-wide-
Every 5 yr
Summer
1 survey
Ancil lary Data
Climate/weather
Fisheries harvest
Waterfowl harvest
Aquaculture sites and yields
Demographic and socioeconomic conditions
Decision record-keeping
-9
-Sound-wide-
-Sound-wide- -
-Sound-wide
-Sound-wide- -
-Sound-wide —
Daily
Annual
Annual
Annual
Annual
Quarterly
Continuous
Continuous
Post-season
Continuous
Continuous
Continuous
a Three kinds of tests, each with five replicate analyses performed on subsamples of a single
composite sample.
b - - Not applicable.
c Ongoing programs are not replicated. It is recommended that existing data or data from a
pilot survey be evaluated to determine appropriate replication scheme.
d December to January for avian species abundances. July for reproductive success.
* All primary stations are located near mouths of rivers. All secondary stations are located
on upstream reaches or on tributaries to the major rivers.
f Bimonthly plus six high-flow events.
9 Four stations: Olympia, Sea-Tac Airport, Bellingham, and Port Angeles.
xii
-------
t Toxic chemicals in recreationally harvested fishes
t Shellfish
t Marine mammals and birds
• Rivers
• Habitat types
• Ancillary data.
Station locations and the rationale for their selection are discussed in
the text. The section on ambient monitoring also includes a summary of
ongoing monitoring programs in the Puget Sound Basin.
COMPLIANCE MONITORING PROGRAM
Compliance monitoring is undertaken by resource management agencies
in response to regulatory programs such as discharge permitting. Compliance
monitoring is undertaken to:
• Ensure compliance with discharge criteria or receiving water
standards
• Document the occurrence of unacceptable adverse impacts
• Evaluate the effectiveness of regulatory programs and management
decisions.
Compliance monitoring should be undertaken for all significant point sources
(including combined sewer overflows and storm drains) under the jurisdiction
of discharge permitting programs.
Compliance monitoring should include evaluations of effluent and sediment
in the receiving environment. Effluent monitoring is undertaken to directly
x i i 1
-------
characterize the effluent and its potential short-term effects. Sediment
monitoring is undertaken to characterize cumulative, long-term effects
of a discharge. Effluent monitoring should include:
• Chemical analyses
• Toxicity bioassays of organisms that represent (at least)
plants, invertebrates, and vertebrates.
Sediment analyses should include chemical analyses, sediment bioassays,
and benthic community structure.
INTENSIVE SURVEYS
Intensive surveys are designed to address specific problems and issues,
and are performed on an "as needed" basis. Intensive surveys should be
considered for the following areas: Sinclair Inlet, Inner Bellingham Bay,
Port Angeles Harbor, Oakland Bay, and Shilshole Bay.
NONPOINT SOURCE MONITORING
Elements of a conceptual framework for integrated watershed monitoring
are summarized in the text. These elements are:
• Integrated watershed modeling and analysis to predict runoff
and contaminant loading
• Land-use monitoring and identification of critical watershed
areas for nonpoint pollution
• Site stormwater monitoring, including improvement of the
stormwater component of National Pollution Discharge Elimination
System (NPDES) permits
• River and stream monitoring
xiv
-------
• Hazardous waste site assessment, especially ongoing RCRA
and Superfund investigations
• NPDES point-source monitoring.
The proposed ambient, compliance, and intensive monitoring programs
will contribute to a nonpoint source monitoring program. However, further
development of an integrated watershed monitoring approach depends on future
identification of critical areas for nonpoint pollution by county governments
(e.g., see PSWQA 1986c). Considerations concerning the design of biological
surveys for monitoring nonpoint sources are provided in the text.
DATA MANAGEMENT SYSTEM
This section outlines the goals and key issues that are important
for selecting a data management system for the Puget Sound Monitoring Program.
Three approaches to database management were evaluated:
• Existing databases (ODES and STORET)
• Database development products (SAS files, ARC/Info, and
other commercially available systems)
t Development of data transfer formats.
The evaluations were based on criteria such as the availability of quality
assurance procedures, ease of use, statistical analytical capabilities,
and cost advantages. Based on these and other criteria, ODES and database
development products, such as SAS files or ARC/Info, ranked as most appropriate
for use by the program.
The cost of developing a database system will largely depend on the
specific features of the system and the nature of the computer-user interface.
A data management system designed to be maintained by technical staff with
some programmer support can be expected to cost $150,000-$300,000.
xv
-------
Following the evaluation of data management systems, recommendations
are made for the frequency, format, and content of reports generated by
the monitoring program.
INSTITUTIONAL MECHANISMS FOR PROGRAM MANAGEMENT AND IMPLEMENTATION
Successful management and implementation of the Puget Sound Monitoring
Program will require a formal institutional arrangement for interagen&y
coordination. Three alternative structures for program management are
described (a committee, a^directorate, and an agency). Major advantages
and disadvantages are noted for each alternative. Jhree alternative approaches
to implementation of the ambient monitoring program are then described
(expand existing programs, centralize all existing and new programs into
a single agency, and expand existing programs while centralizing new ones).
The recommended approach for program management consists of one group
with semiformal interagency relationships for coordinating information
exchange, and another group with well-defined mechanisms for coordinating
management decision-making. This approach facilitates the input of technical
information while clearly assigning accountability for performance. A
distribution of responsibilities by agency is recommended for ambient moni-
toring, compliance monitoring, and intensive surveys. Increased ambient
monitoring, decreased ambient monitoring, and phased implementation are
discussed as alternatives to the proposed monitoring approach. Components
of the proposed program are ranked relative to priority for funding and
order of phased implementation.
ESTIMATED COSTS OF MONITORING PUGET SOUND
Preliminary cost estimates for the ambient program include estimates
of field and laboratory efforts, data entry, and data analysis and report
preparation. The cost estimate for intensive surveys is based on the estimated
annual cost of implementing a reasonable program of intensive surveys for
Puget Sound. Costs are also estimated for program management. The total
annual cost of the proposed Puget Sound Monitoring Program (excluding compliance
xvi
-------
monitoring) is about $4 million. This cost could be reduced to about $3 million
or less by applying monies from some ongoing monitoring programs to the
proposed program.
APPENDICES
Appendix A consists of a statistical power analysis for the proposed
monitoring program. In essence, power analysis is used to determine whether
a sampling plan is adequate to either:
• Reliably conclude that effects related to time and/or station
location are present (i.e., correctly reject the null hypothesis)
• Reliably conclude that no effects related to time and/or
station locations are present (i.e., correctly accept the
null hypothesis).
Sample replication is conducted so that statistical tests can be performed
for detection of spatial and temporal trends. The following monitoring
components and variables were selected for replication:
• Secchi depth
• Pathogen indicators in water
• Histopathological abnormalities in fish
• Shellfish abundance
t Paralytic shellfish poison in shellfish
• Benthic macroinvertebrate abundances
• Toxic chemicals in fish tissue
• Fish species abundances
xvil
-------
t Toxic chemicals in shellfish
• Pathogen indicators in shellfish.
Appendix B contains a summary of agency-specific uses of monitoring
data. This information was obtained by surveying 20 representatives of
participating organizations.
Appendix C contains detailed design considerations for the ambient
monitoring program. Sampling design considerations are presented for sediment
quality variables, water quality variables, biological variables, river
monitoring, and habitat variables, and for the collection of ancillary
data.
xviii
-------
INTRODUCTION
Information on physical, chemical, and biological conditions in Puget
Sound is needed to manage resources effectively and to maintain environmental
quality. Regulatory agencies require accurate technical data to make decisions
regarding resource harvests, waste discharges, aquaculture facilities, new
shoreline developments, and other uses of Puget Sound. In particular,
long-term (e.g., 10-20 yr) monitoring of key variables is valuable for
tracking trends in Puget Sound conditions and for evaluating management
effectiveness. This kind of information is especially important to determine
the cumulative impacts of many individual activities that may seem insignificant
when considered on a case-by-case basis. Monitoring is defined here as
the systematic collection of data to define spatial gradients and temporal
trends in water quality, sediment quality, or biological resources.
Present efforts to monitor physical, chemical, and biological conditions
in Puget Sound are incomplete and uncoordinated. Federal, state, and local
agencies monitor many variables, but methods for sample collection and
analysis, timing of monitoring, and area! coverage vary among programs
(e.g., see Chapman et al. 1985; Puget Sound Water Quality Authority 1986a).
Some important variables, such as concentrations of toxic chemicals in
sediments and in organisms, are not being monitored throughout the sound.
Data are not analyzed consistently and results of studies are often not
reported. Analyses of long-term trends in the status of Puget Sound are
limited (but see Dexter et al. 1985).
The present work proposes a coordinated, long-term monitoring program
for Puget Sound. The specific objectives of this report are to:
• Present goals, objectives, and components of the proposed
monitoring program
-------
• Describe proposed monitoring designs
• Define uses of monitoring data by regulatory and proprietary
agencies
• Evaluate available database management systems for potential
use in the Puget Sound Monitoring Program
• Outline the contents and formats of reports summarizing
the results of the monitoring program
• Evaluate and recommend institutional options for implementing
the monitoring program.
• Estimate approximate costs of monitoring.
The proposed monitoring program is intended to address management issues
and data needs described in the Draft Puget Sound Water Quality Management
Plan and Environmental Impact Statement (Puget Sound Water Quality Authority
1986c).
Under the Puget Sound Estuary Program, an effort is already underway
to standardize monitoring protocols (e.g., Tetra Tech 1986d-i; Tetra Tech
and E.V.S. Consultants 1986a,b). Thus, sampling and analytical protocols
are not described in detail in this report. Instead, references to appropriate
protocols are made.
The structure of the proposed monitoring program and general uses
of the data are illustrated in Figure 1. Three kinds of environmental
monitoring are addressed: ambient monitoring, compliance monitoring, and
periodic intensive surveys. The purpose of ambient monitoring is to assess
environmental conditions and determine system-wide trends, (e.g., on the
scale of a whole bay or the main basin of Puget Sound). Compliance monitoring
is conducted to determine compliance with National Pollutant Discharge
Elimination System (NPDES) discharge permits, and to assess the effectiveness
2
-------
USES OF
DATA
ASSESS
ENVIRONMENTAL
RESOURCES/PROBLEMS
DEVELOP
MANAGEMENT
SOLUTIONS
EVALUATE
MANAGEMENT
EFFECTIVENESS
DATA
PUGET SOUND MONITORING DATABASE
DATA
COLLECTION
COMPLIANCE MONITORING
AMBIENT MONITORING
POINT
SOURCES
NONPOINT
SOURCES
MONITORING
DESIGN
INTENSIVE SURVEYS
SEDIMENT
QUALITY
SITE-SPECIFIC
DESIGNS
WATER
QUALITY
BIOLOGICAL
CONDITIONS
RIVERS
HABITATS
E
V
A
L
U
A
T
E
M
O
N
I
T
O
R
I
N
G
0
E
S
I
G
N
Figure 1. General organization and uses of Puget Sound monitoring data:
-------
of controls on point discharges and stormwater sources of pollutants.
Intensive surveys are site-specific investigations designed to address
specific issues, such as identifying high priority problems, sources of
observed problems, relating land-use practices to water quality, and developing
wasteload allocations.
Ultimate responsibility for the design of a final Puget Sound Monitoring
Program will rest with an interagency management group chaired by a repre-
sentative of the Puget Sound Water Quality Authority (PSWQA). Participants
will include:
• Washington State Departments
Ecology
Game (WDG)
Fisheries (WDF)
Natural Resources (WDNR)
Social and Health Services (OSHS)
t Federal agencies
U.S. Environmental Protection Agency (EPA)
National Oceanic and Atmospheric Administration (NOAA)
U.S. Fish and Wildlife Service (FWS)
U.S. Geological Survey (GS)
U.S. Army Corps of Engineers (COE)
Soil Conservation Service
t Local organizations
Municipality of Metropolitan Seattle (Metro)
Washington Association of Sewer Districts
Washington Association of Water Districts
Washington Association of Local Public Health Officials
-------
• Northwest Indian Fisheries Commission
• Industrial discharger representatives (2)
• Shellfish industry representatives
• Scientists (2)
• Citizens (2)
• Canadian representatives.
The management group will use this report as the basis for development
of a final monitoring plan. Additional responsibilities of the management
group will be to define how the Puget Sound Atlas will be updated, managed
and integrated with ongoing monitoring activities. Eventually, the management
group may serve as an interagency directorate responsible for managing
the monitoring program by making needed refinements to ensure that program
objectives are being met in a cost-effective manner.
The management group will be supported by one or more technical working
committees. The technical working committee(s) will be responsible for
reviewing this report and developing the details of the final monitoring
plan.
-------
GENERAL APPROACH TO THE DEVELOPMENT
OF A MONITORING PROGRAM
The general approach taken to design the Puget Sound Monitoring Program
proposed herein is shown in Figure 2. First, general goals for managing
Puget Sound were established. Second, available information was compiled
on beneficial uses of Puget Sound, recognized water quality problems, and
kinds of physical and socioeconomic factors that influence conditions in
the sound. Because this information has been presented elsewhere (e.g.,
Chapman et al . 1985; PSWQA 1986b,c), it is not discussed in this report.
Results from a survey of regulatory and management agencies regarding monitoring
data they require were combined with the foregoing information to define
monitoring data needs. Third, monitoring objectives and potential components
of the monitoring program were developed, based on data needs and recommenda-
tions developed in a workshop. The initial identification of goals, objectives,
and potential monitoring components was based in part on Chapman et al. (1985).
Participants at a workshop commented on initial goals and objectives and
suggested additional monitoring components. The management goals and monitoring
objectives presented below reflect, in part, ideas suggested by workshop
participants.
MANAGEMENT GOALS AND MONITORING OBJECTIVES
MANAGEMENT GOAL 1. Ensure that Puget Sound is acceptable for human uses.
Monitoring Objective 1. Characterize spatial patterns and temporal
trends in factors that may endanger human health.
Monitoring Objective 2. Characterize spatial patterns and temporal
trends in factors that affect aesthetic conditions.
-------
ESTABLISH
MANAGEMENT GOALS
AVAILABLE INFORMATION
BENEFICIAL USES
WATER- QUALITY PROBLEMS
PHYSICAL AND SOCIOECONOMIC
FACTORS
DEFINE
DATA NEEDS
DATA NEEDS
SURVEY
ESTABLISH MONITORING
OBJECTIVES
DEVELOP
MONITORING DESIGNS
EVALUATE
INSTITUTIONAL OPTIONS
EVALUATE DATA BASE
MANAGEMENT SYSTEM
EVALUATE COSTS
OUTLINE
MONITORING REPORTS
Figure 2. General approach to development of the Puget
Sound Monitoring Program.
-------
Monitoring Objective 3. Characterize spatial patterns and temporal
trends in commercially and recreationally harvested resources.
MANAGEMENT GOAL 2. Ensure that Puget Sound will adequately support productive
and diverse biological communities.
Monitoring Objective 4. Characterize spatial patterns and long-term
trends in biological populations and communities.
Monitoring Objective 5. Determine spatial patterns and temporal trends
in physical, chemical, and biological factors that affect biological
populations.
MANAGEMENT GOAL 3. Improve the effectiveness of regulatory programs and
management decisions.
Monitoring Objective 6. Determine spatial and temporal trends in
contaminant sources and receiving-system properties that are critical
to the planning and design of economic uses and wasteload allocation.
Monitoring Objective 7- Report monitoring results at regular
intervals (e.g., annual) or after completion of individual surveys.
Monitoring Objective 8. Provide data to develop recommendations
for maintaining and improving the region's water quality consistent
with beneficial uses.
All the above goals and objectives apply to ambient and intensive
monitoring. Goals 1-3 and all objectives except Objective 3 apply to compliance
monitoring programs. Specific objectives for compliance monitoring are
provided below in the Compliance Monitoring Section. A subset of the objectives
may be emphasized in any given program.
Based on the above goals and objectives, monitoring components (i.e.,
variables or groups of variables) were selected, and specific designs were
-------
developed for sampling and analysis. Each component was evaluated with
respect to its sensitivity to anthropogenic impacts and with respect to
statistical precision, natural variability, and monitoring costs. Costs
and institutional options for implementing the proposed monitoring program
were evaluated. Refinements in monitoring designs were made as necessary
to maximize cost-effectiveness. Finally, potentially applicable database
systems were evaluated and data reporting requirements were determined.
Specific elements of the approach used to design the monitoring program
are described in the following sections.
KEY CONCEPTS
Certain key concepts formed the basis for the design of the monitoring
program:
• Monitoring variables should be chosen to be cost-effective,
sensitive indicators of changes in environmental conditions
in Puget Sound.
• Characterization of status and trends in the "health" of
Puget Sound is a primary objective of the ambient monitoring
program. Detection of episodes of poor water quality apart
from long-term trends is also a major concern.
• Sampling station distribution should be sound-wide and should
reflect spatial variation in human influences within and
among basins (or bays).
• Sampling frequency should be appropriate to each monitoring
variable, with the most rapidly changing variables receiving
the most frequent sampling. Groups of related variables
should be sampled at the same times and stations.
-------
t Ambient monitoring, compliance monitoring, and intensive
surveys should be complementary. Ambient monitoring should
track the cumulative effects of point and nonpoint sources
of contaminants. The compliance programs should monitor
contaminant sources directly, as well as their individual
effects. Intensive surveys should provide more detailed
data with adequate spatial coverage for defining problem
areas and linking sources with effects. Ambient, compliance,
and intensive programs should use the same protocols whenever
possible.
• Re-evaluation of monitoring designs every 3-5 yr (or more
often initially) will enhance program efficiency and flexi-
bility.
0 Major ongoing monitoring efforts should be incorporated
into the monitoring design when appropriate.
• Quality assurance and quality control are important throughout
all phases of program implementation (sampling, analysis,
data management, and reporting).
CRITERIA FOR PROBLEM IDENTIFICATION
Ambient monitoring data will be used primarily to identify environmental
problems or improvements in the status of Puget Sound. Criteria for definition
of environmental problems in Puget Sound need to be established during
development of the monitoring, program. If the criteria require a certain
precision or level of sample replication for measurement of a variable,
it is essential that these requirements be reflected in the monitoring
design.
-------
Several kinds of environmental criteria can be used to evaluate monitoring
data; for example:
• A fixed acceptable value
• A fixed degree of acceptable change
0 A statistically significant difference.
All three kinds of criteria are subjective to some degree. For example,
policy decisions must be made to determine what is "acceptable" or what
is adequate power for a statistical test.
Criteria based on a fixed acceptable value frequently have been used
to set environmental standards (e.g., State of Washington water quality
standards). A criterion based on a fixed degree of change could be more
sensitive than a fixed criterion because it could allow action to be taken
before the fixed-value criterion was reached. Criteria based on statistical
significance have not been used commonly to set environmental standards.
However, this kind of criterion is particularly useful for evaluating dif-
ferences from a reference value because it accounts for measurement varia-
bility. Unlike the other kinds of criteria, statistically based criteria
generally require replicate measurements. The number of replicates selected
depends on policy decisions regarding statistical power, levels of significance,
and minimum detectable difference.
Reference values can be used to detect spatial or temporal changes.
Spatial comparisons would be made between a potentially impacted site and
a site in relatively uncontami nated area (the reference site). Temporal
comparisons would be made between the values observed at a site during
one year and the values observed at that same site during the previous
year or during the initial year of sampling.
Each sampling scheme presented in this report was designed to reflect the
kind of criterion (i.e., fixed value, fixed degree of change, or statistical
-------
difference) that will likely be used to evaluate each component (Table I).
In most cases, the values of the criteria must still be established. The
minimum detectable difference (at power = 0.8 and significance level = 0.05)
for each statistically-based design is provided in Appendix C. Note that
application of criteria may require integration of data from multiple variables
[e.g., identification and ranking of toxic problem areas using sediment
chemistry, toxicity, and bioeffects data from urban bay action programs
(Tetra Tech 1985a)J.
IMPORTANT STATISTICAL CONSIDERATIONS
Changes in the values of monitoring variables over time may or may
not reflect true trends in the conditions of Puget Sound. Assuming that
sampling bias and measurement techniques remain constant, one major cause
of temporal change is random natural variability. Statistical techniques
are often needed to separate this random variability from anthropogenic
effects or to separate anthropogenic effects from natural trends of change.
Selection of an appropriate statistical model is an important step in designing
a monitoring program.
Once a statistical model is chosen, tradeoffs among elements of the
study design (e.g., level of sampling replication, statistical power, and
the minimum detectable difference between sample means) can be evaluated
by power analysis. Statistical power is defined as the probability of
detecting a real change in a monitoring variable over time or space. Cohen
(1977), Bernstein and Zalinski (1983), and Mar et al . (1986) address the
concept of statistical power as applied to monitoring program design (also
see Appendix A herein).
In this report, the generalized hypothesis selected for evaluating
study designs was the Null Hypothesis of no differences in mean values
of a monitoring variable among sampling times (or among stations). Power
analyses were generally performed using a one-way Analysis of Variance
(ANOVA) model (the exceptions are the fish histopathology design, where
a G-test was used for power analysis of data on lesion frequencies, ana
10
-------
TABLE 1. CRITERIA PROPOSED FOR COMPONENTS OF THE
PUGET SOUND MONITORING PROGRAM
Component
Kind of Criteria
Fixed* Statistical
Comments'*
Sediment Quality
Sediment chemistry
Sediment toxicity
bioassays
Conventional sediment
variables
Water Quality
Hydrographic conditions
Dissolved oxygen
Turbidity/transparency
Sediment quality values are being developed
by U.S. EPA and U.S. COE
Statistically detectable difference
from reference conditions
Sediment quality values are being evaluated
by U.S. EPA and U.S. COE
Natural phenomena; not applicable
Existing state standards or
new standards
New standards
Odor, floatables,
siicks, water color
Nutrient concentrations
Phytoplankton standing
stock
Pathogen indicators
in water
Biological Conditions
Benthic macroinvertebrate
abundances
Toxic chemicals in
fish tissue
X
X
X
X
Existing state and federal standards
or new standards
New standards
New standards
State standards, or statistically detectable
difference from reference conditions
Statistically detectable difference from
reference conditions
Existing U.S. FDA standards, U.S. EPA
risk assessment; statistical ly detectable
detectable difference from reference
conditions
-------
TABLE 1. (continued)
Histopathological
abnormalities in fish
Fish species abundances
Shellfish abundances
Toxic chemicals in
shellfish
PSP in shellfish
Pathogen indicators
in shellfish
Marine mammal abundances X
and reproductive success
Avian abundances and X
reproductive success
River Monitoring X
Habitat Types
Statistically detectable difference from
reference conditions
Statistically detectable difference from
reference conditions
Statistically detectable difference from
reference conditions
Existing U.S. FDA standards, U.S. EPA
risk assessment analysis; statistically
detectable difference from reference
conditions
Existing state or federal standards; or
statistically detectable difference from
reference conditions
Existing state or federal standards; or
statistically detectable difference from
reference conditions
Possible qualitative analysis and of
trends
Possible qualitative analysis of trends
Existing or new state standards; statistically
detectable difference from reference
conditions
Possible qualitative analysis of trends
a Fixed criteria include fixed values and/or fixed degrees of change.
b Values must be developed for all criteria designated as new standards or statistical differences
from reference conditions.
-------
the amphipod sediment toxicity design, where a power analysis was available
from the literature). The measure used to evaluate the statistical sensitivity
of a monitoring design was the minimum detectable difference between two
mean values. The statistical sensitivity corresponding to a selected level
of replication is given for each replicated sampling design in Appendix C.
ONGOING MONITORING
A number of monitoring programs have been, or are now being, conducted
in Puget Sound. Among the major ongoing monitoring efforts are those of
Washington Department of Ecology, Municipality of Metropolitan Seattle,
Washington Department of Fisheries, Washington Department of Game, National
Oceanic and Atmospheric Administration, U.S. Fish and Wildlife Service,
U.S. Geological Survey, and National Climate Data Center. Details of these
monitoring programs are given in Table 6 of Chapman et al. (1985). A summary
of major ongoing monitoring efforts is provided later in this report (see
below, Ambient Monitoring Program, Ongoing Monitoring).
11
-------
MONITORING COMPONENTS IN RELATION TO GOALS AND OBJECTIVES
The monitoring components selected for the Puget Sound Monitoring
Program are closely related to the program goals and objectives (Table 2).
The rationale for selecting each monitoring component identified below
is given in Appendix C. Table 2 encompasses ambient, compliance, and intensive
monitoring programs. Components that were selected for the proposed program
are:
Sediment Quality
Sediment chemistry
Sediment toxicity bioassays
Conventional sediment variables
Water Quality
Hydrographic conditions (salinity, temperature)
Dissolved oxygen
Turbidity/transparency
Odor, floatables, slicks, water color
Nutrient concentrations
Phytoplankton standing stock
Pathogen indicators in water
Biological Conditions
Benthic macroinvertebrate abundances
Toxic chemicals in fish tissue
Histopathological abnormalities in fish
Fish species abundances
Shellfish abundances
12
-------
TABLE 2. RELATIONSHIP OF PROPOSED MONITORING COMPONENTS
TO MANAGEMENT GOALS AND MONITORING OBJECTIVES3
MANAGEMENT GOAL 1. Ensure that Puget Sound is acceptable for human uses.
Monitoring Objective 1. Characterize spatial patterns and temporal
trends in factors that may endanger human health.
Monitoring Components:
Toxic chemicals in fish tissue
Toxic chemicals in shellfish
Pathogen indicators in shellfish
Pathogen indicators in water
PSP in shellfish
NPDES compliance and other point-source monitoring
Nonpoint sources^
Monitoring Objective 2. Characterize spatial patterns and temporal
trends in factors that affect aesthetic conditions.
Monitoring Components:
Odor, floatables, slicks, water color
Turbidity/transparency (Secchi depth)
Phytoplankton standing stock (chlorophyll a)
Habitat types (areal extent)
NPDES compliance and other point-source monitoring
Nonpoint sources
Monitoring Objective 3. Characterize spatial patterns and temporal
trends in commercially and recreationally harvested resources.
Monitoring Components:
Shellfish species abundances
Fish species abundances
Avian species abundances and reproductive success
Aquaculture sites and yields
MANAGEMENT GOAL 2. Ensure that Puget Sound will adequately support productive
and diverse biological communities.
Monitoring Objective 4. Characterize spatial patterns and long-term
trends in biological populations and communities.
Monitoring Components:
Phytoplankton standing stock (chlorophyll a)
Shellfish species abundances
Benthic macroinvertebrate species abundances
Fish species abundances
-------
TABLE 2. (Continued)
Avian species abundances and reproductive success
Marine mammal species abundances and reproductive success
NPDES compliance and other point-source monitoring
Monitoring Objective 5. Determine spatial patterns and temporal trends
in physical, chemical, and biological factors that affect biological
populations.
Monitoring Components:
a) Physical
Climate/weather
Hydrographic conditions (temperature, salinity)
River monitoring (e.g., flow, temperature)
Turbidity/transparency (Secchi depth)
Conventional sediment variables (grain size)
Habitat types and distributions (e.g., haulout sites,
macrophyte beds, human structures, and aquaculture)
including habitat gains and losses
b) Chemical
Dissolved oxygen
Nutrient concentrations
Sediment chemistry (chemical contaminants)
Conventional sediment variables (total organic carbon,
sulfides, salinity)
River monitoring (contaminant loading)
NPDES compliance and other point-source monitoring
Nonpoint sources
c) Biological
Habitat types (e.g., eelgrass, kelp)
d) Management
Fisheries harvest (plants and animals)
Waterfowl harvest
e) Indicator variables
Sediment toxicity bioassays
Histopathological abnormalities in fish
Marine mammal species abundances and reproductive success
Avian species abundances and reproductive success
-------
TABLE 2. (Continued)
MANAGEMENT GOAL 3. Improve the effectiveness of regulatory programs and
management decisions.
Monitoring Objective 6. Determine spatial and temporal trends in
contaminant sources and receiving-system properties that are critical
to the planning and design of economic uses and wasteload allocation.
Monitoring Components:
All components listed above, and especially:
Habitat types
Demographic and socioeconomic conditions
NPDES compliance and other point-source monitoring
Nonpoint sources
Decision record-keeping
Monitoring Objective 7. Report monitoring results at regular intervals
(e.g., annual) or after completion of individual surveys.
Monitoring Components:
All components listed above
Monitoring Objective 8. Provide data to develop recommendations for
maintaining and improving the region's water quality consistent with
beneficial uses.
Monitoring Components:
All components listed above
a Goals, objectives, and monitoring components are not listed in any order
of priority. See text for monitoring components that were evaluated but
not selected for the proposed program.
b Nonpoint source monitoring encompasses the river monitoring component
and the stormwater element of the compliance monitoring program. NPDES
regulations cover point discharges of stormwater, which represent a collection
of nonpoint sources.
-------
Toxic chemicals in shellfish
PSP in shellfish
Pathogen indicators in shellfish
Marine mammal species abundances and reproductive success
Avian species abundances and reproductive success
Nonpoint Sources
River Monitoring
Habitat Types
NPDES Compliance Monitoring
Ancillary Data
Climate/weather
Fisheries harvest
Waterfowl harvest
Aquaculture sites and yields
Demographic and socioeconomic conditions
Decision record-keeping.
The selected components represent a wide range of environmental indicators
at various trophic levels and in various media (water, sediment, and biota).
Some components (e.g., dissolved oxygen) are represented by single monitoring
variables. Other components (e.g., sediment chemistry) include a large
number of monitoring variables. Specific variables to be addressed as
part of each component of the ambient monitoring program are described
in Appendix C. Details of a recommended design for compliance monitoring
are being developed by Washington Department of Ecology and U.S. EPA.
General aspects of intensive surveys are discussed in a later section.
Intensive surveys may encompass a wide variety of potential monitoring
components.
13
-------
Monitoring of nonpoint sources encompasses portions of the Compliance
Monitoring Program (i.e., storm drains and combined sewer overflows) and
portions of the Habitat Types component. River Monitoring will also contribute
to nonpoint source assessment because rivers represent a collection of
nonpoint and point sources. Because of its importance and multi-program
status, an overview of a nonpoint source monitoring strategy is presented
in a separate chapter (see below, Nonpoint Source Monitoring).
The components that were not included in the proposed ambient program
are listed below, followed by the rationale for their exclusion. Note
that some of the excluded components listed below may be appropriate for
intensive surveys or compliance inspection surveys. The excluded components
are:
• Hydrographic conditions (currents)
• Plankton species abundances, including toxic dinoflagellate
abundances
t Primary productivity
t Macrophyte species abundances (community analysis)
• Toxic chemicals in macroalgae
• Toxic chemicals in birds and mammals
• Toxic chemicals in .the water column and microlayer
• Water toxicity bioassays.
Routine monitoring of water currents is not proposed as part of the
hydrographic conditions component. Further characterization of currents
in Puget Sound is needed for predicting transport and fate of contaminants,
but frequent or routine assessments may not be appropriate. Detailed patterns
14
-------
of water movement should be determined as part of special studies to develop
a Puget Sound circulation model. Locations of sampling stations and other
study design elements should be specified after a model is selected and
calibrated with existing data.
Plankton species abundances and primary productivity are not included
because of the high cost of monitoring these variables relative to the
gain in useful information. Because these components are highly variable
in space and time, an adequate monitoring program would be prohibitively
expensive. Monitoring of chlorophyll a and phaeophytin a pigments is proposed
as a measure of phytoplankton standing stock and condition. Although species
shifts in phytoplankton communities may sometimes occur without substantial
changes in chlorophyll a concentrations, total chlorophyll a trends will
often appear before water quality problems are revealed in species-specific
data. Moreover, some regulatory agencies, such as the California State
Water Resources Control Board, have proposed criteria for chlorophyll a
concentrations (Home and McCormick 1978).
Monitoring of macrophytic species abundances was excluded, with the
exception of kelp (Nereocystis 1 uetkeana) and eelgrass (Zostera marina).
Kelp and eelgrass beds are important habitat for many commercially and
recreationally important species. Monitoring of kelp and eelgrass is discussed
below under Habitat Monitoring Designs. Species-level monitoring of complete
macroalgae assembages would involve collection of ancillary data on water
quality, substrate quality, and grazer populations to allow interpretation
of trends in species abundances. Such extensive monitoring is considered
too costly for the value of the information gained.
Monitoring of toxic chemicals in macroalgae may be important relative
to concerns for human health. Because an initial survey is planned for
the near future, design of a monitoring program should await these results.
It may be concluded that information on toxic chemicals in macroalgae would
be redundant with data provided by other bioaccumulation monitoring components.
15
-------
Although excluded from the current proposed list of monitoring components,
toxic chemicals in birds and manmals could be monitored profitably to indicate
general conditions within an entire bay or a large portion of the sound
(Kendall, June 1986, personal communication; Poelker, 20 May 1986, personal
communication). Potential approaches include monitoring of body burdens
of toxic contaminants in animals such as harbor seals (or Great Blue Herons),
and in recreationally harvested waterfowl (Kendall, June 1986, personal
communication; Poelker, 20 May 1986, personal communication). Reasons
for monitoring these components include (Calambokidis, 2 June 1986, personal
communication): 1) these biotic groups represent higher trophic levels
that have a high potential for bioaccumul ation of toxic contaminants, 2)
resident harbor seals could serve as an early warning indicator or analog
for the effects of consumption of local seafood on humans, 3) mammals and
birds are highly mobile and thereby indicate trends in a broad geographic
region, and 4) mammals and birds are of recreational and aesthetic value
to humans.
Despite the possible value of including bioaccumulation components
for mammals and birds, such monitoring would be partly redundant with the
proposed monitoring of toxic chemicals in fish and shellfish. Moreover,
the proposal of bioaccumul ation monitoring designs for birds and mammals
would require additional work to define variance characteristics of represen-
tative data sets. Preliminary field surveys may be needed to obtain necessary
data. Use of available data on stranded animals is not recommended because
of potential degradation of organic contaminants.
Monitoring particulate materials caught in sediment traps represents
one possible way of collecting data on toxic chemicals in the water column.
Research on this potential monitoring technique has been undertaken by
the Pacific Marine Environmental Laboratory of the National Oceanic and
Atmospheric Administration (NOAA) (see Bates et.al. 1984). This research
is promising, especially for collecting data on seasonal and annual variations
in toxic chemicals that are bound to particles in the water column.
16
-------
At present, technical limitations and high costs preclude recommendation
of routine monitoring of toxic contaminants in the water column and the
microlayer, including those bound to suspended particles. For analyses
of whole-water samples, the analytical detection limits are not sufficiently
low to quantify concentrations of concern for many organic chemicals.
Use of sediment traps to monitor concentrations of contaminants on sediment
particles would provide partly redundant information given direct monitoring
of bottom sediments proposed here. Although sediment trap data would provide
more information on short-term variations (on the order of several months)
than would monitoring of bottom sediments, the higher cost associated with
sampling and analysis of sediment traps does not justify their use. Direct
monitoring of bottom sediments is needed to correlate chemical contamination
with toxicity measured by sediment bioassays and with effects on benthic
macroinvertebrate communities and histopathological abnormalities in fish.
Finally, contaminant flux data provided by sediment traps would be less
useful than direct monitoring of contaminant sources.
Water toxicity bioassays are also excluded from the monitoring program.
Without data on contaminant levels in the water column, interpretation
of the bioassay results would be severely limited. Further research on
microlayer toxicity bioassays and their interpretation is needed before
they can be recommended for a routine ambient monitoring program.
17
-------
USES OF MONITORING DATA
Monitoring information on Puget Sound may be used for many purposes
by regulatory agencies and other public and private entities (e.g., the
state legislature, Puget Sound Water Quality Authority, research scientists).
The purpose of this discussion is to indicate how data from the Puget Sound
Monitoring Program could be used by public agencies to satisfy their program
needs. Uses of data may be divided into two major categories: general
uses and agency-specific uses. General uses are described below, and agency-
specific uses are described in Appendix B.
Uses of the data were assembled, in part, from a survey of agency
representatives that was conducted to determine present data needs and
uses. Many of the uses of data discussed below are derived directly from
the monitoring program goals and objectives (Table 2 above). Others are
anticipated uses. The list of uses presented below is not comprehensive.
• Protection of human health. Components that could be used
to assess public health risks from contamination include
toxic chemicals in fish tissue, toxic chemicals in shellfish,
pathogen indicators in water and shellfish, and PSP in shellfish
(e.g., see Tetra Tech 1986b). Agencies performing risk
assessments could include U.S. EPA, DSHS, Washington Department
of Ecology, Metro, and County health departments. Agency
actions based on risk assessment results could include closing
fisheries or issuing public advisories against consumption
of seafood from problem areas.
• Assessment of environmental impacts. Impact assessments
may consist of intensive surveys of suspected problem areas.
For example, the three sediment quality components (i.e.,
sediment chemistry, sediment toxicity bioassays, and conventional
18
-------
sediment variables) plus benthic macroinvertebrate abundances,
bioaccumulation of toxic chemicals in fish, and histopathological
abnormalities in fish have been used effectively by U.S. EPA
and Washington Department of Ecology in the Urban Bay Action
Programs to assess the areal extent, magnitudes, and types
of impacts caused by municipal and industrial discharges.
Long-term trends determined from ambient monitoring data
may be used to assess cumulative impacts from multiple human
activities. Results of impact assessments could be used
to identify new contamination problems and establish cleanup
priorities.
Evaluation of program effectiveness. Data collected from
affected or managed sites are especially useful for determining
whether management decisions have yielded the desired effect.
For example, long-term data from compliance and ambient
monitoring could be used to assess the effectiveness of
industrial pretreatment programs. Intensive survey data
are often important for evaluating program effectiveness.
Data from the monitoring program could be used by each agency
to assess program effectiveness (and make changes, if warranted)
within ongoing or anticipated programs. A coordinated monitoring
database will be valuable for evaluating the effectiveness
of programs developed and PSWQA's Puget Sound Plan (PSWQA
1986C-).
Design of remedial action. Monitoring data could be used
for designing remedial action programs as well as for assessing
the effectiveness of those programs. Typical remedial actions
would include, for example, habitat restoration, removal
of contaminated sediments, and the separation of storm and
sewer drains. Historical data collected from an area of
interest could also be used to predict chances of recovery,
degree of recovery, and time required for recovery. Intensive
survey data on sediment quality, biological conditions,
19
-------
and habitat types would be most useful for designing a remedial
action program and assessing its effectiveness. Agencies
using data in this way include Washington Department of
Ecology, WDG, WDNR, WDF, U.S. EPA, U.S. FWS, Metro, counties,
and cities.
• Enforcement of criteria, standards, and permit limits.
The compliance monitoring program and intensive surveys
are essential to assure compliance with regulations and
permit specifications. The monitoring program could also
provide supplemental information for the development of
criteria and standards by Washington Department of Ecology,
DSHS, and U.S. EPA. For example, sediment "Triad" data
(chemistry, toxicity bioassays, benthic macroi nvertebrates)
could supplement databases being used to establish Sediment
Quality Values. Data from effluent chemistry analyses and
bioassays conducted as part of compliance monitoring programs
could be used by Washington Department of Ecology and U.S. EPA
to establish discharge limits for NPDES and state permits.
• Management of biological resources. The monitoring program
could provide resource-management agencies with general
information on trends throughout Puget Sound, and site-specific
information on applicable physical, chemical, and biological
variables. Monitoring data from each of the components
could be used in management of biological resources by Washington
Department of Ecology, WDG, WDF, Northwest Indian Fisheries
Commission, and U.S. FWS. For example, important fisheries
management actions include setting limits on harvest season
and daily catch per person.
• Collection of baseline information. Information on present
conditions throughout Puget Sound could be used in the future
as a baseline to describe spatial and temporal trends in
20
-------
specific variables, and to define reference and impacted
conditions. All monitoring components would contribute
valuable baseline information to a long-term database.
Interpretation of data. Monitoring data of one type may
be very useful for interpreting data of another type, regardless
of which programs are involved. For example, data on patho-
logical conditions in fish that were collected during an
intensive survey of an industrialized bay could be interpreted
more fully if long-term, historical information on sediment
quality were available. All monitoring components would
be expected to contribute to this general-use category.
Conventional sediment variables, hydrographic conditions,
and ancillary data components would most often be used for
interpretive purposes.
21
-------
AMBIENT MONITORING PROGRAM
The ambient monitoring program presented below involves periodic sampling
of the water, sediment, and biota at specified stations, and collection
of ancillary data on human uses of the resources within Puget Sound and
its watersheds. A summary of the ambient monitoring program and a description
of sampling stations are provided in the following sections (see Appendix C
for details and rationale). Major ongoing monitoring programs are summarized
in the last section.
The monitoring program described below should be re-evaluated after
data have been collected for 2-5 yr. Because baseline data are needed
for many areas of Puget Sound, the program involves allocation of more
effort to the proposed ambient program than to intensive surveys (see Insti-
tutional Mechanisms for Program Management and Implementation and Estimated
Costs of Monitoring Puget Sound). After an initial data set has been collected
and evaluated, the spatial and temporal coverage of the ambient program
may be reduced to allow greater allocation of effort and funds to intensive
surveys. Options for modifying the ambient program or implementing components
in a phased manner are addressed below (see Institutional Mechanisms for
Program Management and Implementation, Program Modification and Phased
Implementation).
OVERVIEW
The ambient monitoring program is summarized in Table 3. Because
of the importance of geographic coverage, sampling station locations are
described in detail below. Other details of the monitoring design (e.g.,
specific variables, statistical sensitivity) and the rationale for the
design are found in Appendix C.
22
-------
TABLE 3. SUMMARY OF THE AMBIENT MONITORING PROGRAM
Number of Stations ^m.
Components
Sediment Qual ity
Sediment chemistry
Sediment toxicity bioassays
Conventional sediment variables
Water Qual ity
Hydrographic conditions
Dissolved oxygen
Turbid ity/ transparency
Odor, floatables, etc. - offshore
Odor, floatables, etc. - intertidal
Nutrient concentrations
Phytoplankton standing stock
Pathogen indicators in water
Biological Conditions
Benthic macroinvertebrate abundances
Toxic chemicals in fish - English sole
Toxic chemicals in fish - Cod and Salmon
Histopathological abnormalities in fish
Fish species abundances
Shellfish abundances
Toxic chemicals in shellfish
PSP in shellfish
Pathogen indicators in shellfish
Marine mammal abundances and reprod. success
Avian abundances and reproductive success
River Monitoring
Habitat Types
Ancil lary Data
Cl imate/weather
Fisheries harvest
Waterfowl harvest
Aquaculture sites and yields
Demographic and socioeconomic conditions
Decision record-keeping
Urban and
Main Industrial
Basins Bays
31
31
31
11
11
11
11
16
11
11
16
16
31
16
3
16
16
16
16
16
16
16
16
-
-
-9
-
- -
- - -
-
34
34
34
15
15
15
15
5
15
15
5
5
34
27
15
27
27
5
5
5
5
5
5
-Sound -wide-
-Sound-wide-
-8 primary*- -
13 primary6 —
11 secondary6-
-Sound-wide —
-
-Sound-wide-
-Sound-wide-
-Sound-wide- -
-Sound-wide-
-Sound-wide- -
Rural
Bays Frequency
41
41
41
27
27
27
27
5
27
27
5
5
41
19
0
19
19
5
5
5
5
5
5
-
~
-
-
-
-
-
Annual
Annual
Annual
Monthly
Monthly
Monthly
Monthly
Weekly
Monthly
Monthly
Weekly
Monthly
Annual
Biennial
Annual
Biennial
Biennial
Annual
Annual
Weekly
Monthly
Weekly
Monthly
Annual
Monthly
12 times/yrf
12 times/yrf
12 times/yrf
Every 5 yr
Daily
Annual
Annual
Annual
Annual
Quarterly
pi inq Program
Number of
Timing Replicates
March
March
March
Daytime
Daytime
Daytime
Daytime
May- July
Daytime
Daytime
May-July
Aug-April
March
July
Sep-Oct
July
July
May
May
May-July
Aug-April
May-July
Aug-April
June-Aug
Dec-Jan"
Continuous
3-yr period
3-yr period
Summer
Conti nuous
Conti nuous
Post-season
Conti nuous
Cont i nuous
Continuous
1
composite
3x5a
1
1
1
5
1
1
3
3
5
3
3
60
4
composite
prof i le
prof i le
readings
-b
composite
composite
composites
composites
grabs
composites
composites
specimens
trawls
15 cores
3
3
3
3
3
1
1
1
1
1
1
composites
composites
composites
composites
composites
survey0
surveyc
composite
composite
composite
survey
a Three kinds of tests, each with five replicate analyses performed on subsamples of a single
composite sample.
b - = Not applicable.
c Ongoing programs are not replicated. It is recommended that existing data or data from
pilot survey be evaluated to determine appropriate replication scheme.
d December to January for avian species abundances. July for reproductive success.
« All primary stations are located near mouths of rivers. All secondary stations are located
on upstream reaches or on tributaries to the major rivers.
f Bimonthly plus six high-flow events.
9 Four stations: Olympia, Sea-Tac Airport, Bellingham, and Port Angeles.
-------
Much of the recommended program is designed to fill gaps in existing
monitoring efforts, e.g.:
• Toxic contamination. Assessment of toxic chemical concentrations
in sediments, sediment toxicity bioassays, and toxic chemicals
in fish and shellfish tissue is very limited in ongoing
monitoring programs.
t PSP and bacteria in shellfish. Present monitoring at beaches
used for recreational harvesting is irregular, and sampling
is unreplicated. The recommended shellfish monitoring designs
in Table 3 and Appendix C focus on recreational beaches,
thereby complementing ongoing monitoring by DSHS and county
health departments.
• Contaminant loading. Limited information on contaminant
loading from tributary rivers is available. The water quality
and river monitoring programs summarized in Table 3 replace
the present ambient monitoring program performed by Washington
Department of Ecology. Data being collected by U.S. GS
and Metro will supplement the river monitoring component
of the ambient program.
• Biological conditions. Certain key resources are monitored
at present. The recommended program focuses on higher trophic
levels (bottomfish, birds, and mammals), a major shellfish
resource (butter clams), and indicators of sediment quality
(benthic macroinvertebrate abundances, histopathological
abnormalities in fish, and bottomfish species abundances).
A summary of major ongoing monitoring programs is provided below (see
Ongoing Monitoring). Details of the relationship between ongoing monitoring
programs and the proposed Puget Sound Monitoring Program are provided in
a later section (Institutional Options and Estimated Costs of Monitoring).
23
-------
SAMPLING STATIONS
For the purpose of designing the ambient monitoring program, the waters
of Puget Sound (Figure 3) were divided into three categories:
• Main basins
t Rural bays
• Urban and industrialized bays.
Greatest emphasis is placed on the urban and industrialized bays because
the potential for anthropogenic impacts is greatest in those areas. The
next priority is placed on rural bays because of the potential for impacts
from agricultural pollutants, and their proximity to other human activities.
The main basins are given the least emphasis because of their greater distances
from pollutant sources and the greater potential for dilution of contaminants.
Because Puget Sound is very large, and because funds are limited,
the collection of detailed information on the spatial extent of problem
areas within all of Puget Sound is not a primary purpose of the ambient
monitoring program. Intensive surveys of smaller portions of the sound
will fulfill this need on the basis of priority (as discussed below).
Compliance monitoring programs for industrial and municipal discharges
and additional monitoring of combined sewer overflows (CSOs) and stormwater
drains by responsible parties will be needed to cover many nearshore areas.
Knowledge of broad-scale changes in conditions will be gained from the
ambient monitoring stations proposed below, however, because stations are
located throughout the basin.
The proposed ambient monitoring program encompasses a large geographic
area. It is strongly recommended that this broad geographic sampling plan
be implemented for 2-5 yr to collect baseline information throughout Puget
Sound. After 2-5 yr, the program should be reviewed and revised if necessary
to allocate resources more appropriately with respect to impacted areas
24
-------
Sfra/f of Juan de Fuca
Port Ang«l««
1. ADMIRALTY BAY
2. BUDO INLET
3. CARR INLET
4. CASE INLET
5. COMMENCEMENT BAY
6. DABOB BAY
7. DANA PASSAGE
8. DISCOVERY BAY
9. DUNGENESSBAY
10. DYES INLET
11. EAGLE HARBOR
12. ELD INLET
13. ELLIOTT BAY
14. GIG HARBOR
15. HALE PASSAGE
16. HAMMERSELY INLET
17. HENDERSON INLET
18. HOLMES HARBOR
19. KILJSUT HARBOR
20. LIBERTY BAY
21. MUTINY BAY
22. OAKLAND BAY
23. OAK BAY
24. PEALE PASSAGE
25. PICKERING PASSAGE
26. PORT ANGELES HARBOR
27. PORT GAMBLE
28. PORT GARDNER
29. PORT LUDLOW
30. PORT MADISON
31. PORT ORCHARD
32. PORTTOWNSEND
33. QUARTERMASTER HARBOR
34. QUILCENE BAY
35. SEQUIM BAY
36. SINCLAIR INLET
37. THE GREAT BEND
38. TOTTEN INLET
39. USELESS BAY
kilometers
nautical miles
Shtlten
Figure 3. Locations of Puget Sound basins and bays.
(continued on next page)
-------
>x, Strait of Georgia
VANCOUVER
ISLAND
S(ra'( of Juan de Fuca
BELLINGHAM BAY
BIRCH BAY
BURROWS BAY
CRESCENT HARBOR
DRAYTON HARBOR
FIDALGO BAY
GUEMES CHANNEL
HALE PASSAGE
LUMMI BAY
PADILLA BAY
PENN COVE
PORT SUSAN
SAMISH BAY
10 20 kilometers
10 nautical milea
Figure 3. Continued.
-------
non-impacted areas, and areas where a potential for impacts exists. This
approach is recommended because most areas of Puget Sound have never been
surveyed synoptical 1y. Therefore, any a priori judgments to delete areas
from the monitoring program would lack a strong technical basis, and could
result in the mi sal location of funds to lower priority areas. It would
be possible to rotate sampling efforts among areas from the beginning of
the program, but using a rotating-network approach might require 10-15 yr
of data before an important environmental trend would be discernible.
The more conservative approach proposed here is to establish baseline infor-
mation for 3-5 yr throughout Puget Sound, and then to consider rotating
sampling efforts among low-priority areas.
Station locations for the ambient monitoring program are described
in the following sections. Each section addresses a group of monitoring
components to be sampled at similar stations. Only general locations of
proposed sampling stations are given. A preliminary field survey will
be needed to select final station locations. The rationale for station
locations and criteria for final siting of stations are given in Appendix C.
Sediment Quality, Water Quality, and Benthic Macroinvertebrates
The components addressed in this section include all sediment quality
components, benthic macroinvertebrate species abundances, and all water
quality components with the exception of pathogen indicators. Sampling
of pathogen indicators in water is proposed for intertidal areas designated
for shellfish assessment (see below, Shellfish). The combination of sediment
chemistry, sediment bioassay, and benthic macroinvertebrate components
forms the "Triad" of monitoring variables recommended by Chapman and Long
(1983).
The basic station array for the ambient monitoring program consists
of 106 stations. These stations are apportioned among the main basins,
urban and industrialized bays, and rural bays as indicated in Figure 4
and Tables 4-6 (see Appendix C for rationale).
25
-------
BASJU. STATIONS
20-m
® CENTER OF CHANNEL
BAY STATIONS
RURAL BAY
URBAN BAY
NOTE:
Number of stations shown adjacent to symbol
Only general locations of bay stations are
shown. For examples of station layout in
bays, see Figures 5 and 6.
10 20 kilometers
10 nautical miles
Figure 4. Proposed locations for sediment quality triad and water
quality components, (continued on next page)
-------
Port Angel** >
NOTE:
Number of stations shown adjacent to symbol.
Only general locations of bay stations are
shown. For examples of station layout in
bays, see Figures 5 and 6.
20 kilometers
10 nautical miles
BASIN STATIONS
• 20-m
® CENTER OF CHANNEL
BAY STATIONS
RURAL BAY
URBAN BAY
Figure 4. Continued.
-------
TABLE 4. PROPOSED NUMBERS OF SAMPLING STATIONS TO BE ASSIGNED TO
PUGET SOUND BASINS FOR MONITORING OF SEDIMENT QUALITY WATER
QUALITY, AND BENTHIC MACROINVERTEBRATES
Area
Strait of Georgia
Boundary Pass
Haro Strait
Rosario Strait
Strait of Juan de Fuca
Admiralty Inlet
Hood Canal
Main Basin (incl. East Passage
and Colvos Passage)
Whidbey Basin
South Sound (incl. The Narrows,
Cormorant Passage, Drayton
Passage, Balch Passage and
Ni squally Reach)
TOTAL FOR ALL AREAS
Triad3
2
0
0
0
5
0
6
9
6
_3
31
Water Qual ity
1
0
0
0
2
0
2
3
2
_1
11
a Triad stations include sediment chemistry, toxicity bioassays, and benthic
macroinvertebrate components.
-------
TABLE 5. PROPOSED NUMBERS OF SAMPLING STATIONS TO BE ASSIGNED
TO URBAN AND INDUSTRIALIZED BAYS FOR MONITORING OF SEDIMENT
QUALITY, WATER QUALITY, AND BENTHIC MACROINVERTEBRATES
Area
Port Angeles Harbor
Drayton Harbor
Bell ingham Bay
Fidalgo Bay and
Guemes Channel
Port Townsend and
Kilisut Harbor
Port Ludlow
Port Gardner
Eagle Harbor
Elliott Bayb
Liberty Bay
Sinclair Inlet
Dyes Inlet
Commencement Bay
Budd Inlet
Oakland Bay and
Hammersely Inlet
TOTAL FOR ALL AREAS
Approx. Area
(sq mi)
5.26
3.35
44.21
7.18
14.91
2.40
22.86
0.63
12.86
3.43
4.29
7.14
9.15
8.17
5.43
Triad3
2
1
3
2
2
1
3
1
5
1
2
3
3
3
_2
34
Water Qual ity
1
1
1
1
1
1
1
1
1
1
1
1
1
1
_1
15
a Triad stations include sediment chemistry, toxicity bioassay, and benthic
macroinvertebrate components.
b Example Bay. See text for explanation.
-------
TABLE 6. PROPOSED NUMBERS OF SAMPLING STATIONS TO BE ASSIGNED TO
RURAL BAYS FOR MONITORING OF SEDIMENT QUALITY, WATER QUALITY,
AND BENTHIC MACROINVERTEBRATES
Area
Birch Bay
Lummi Bay and
Hale Passage
Samish Bay
Padilla Bay
Burrows Bay
San Juan Islands
(inclusive)
Skagit Bay (incl .
Similk Bay and
Dugualla Bay)
Crescent Harbor
Penn Cove
Port Susan
Holmes Harbor
Dungeness Bay
Sequim Bay
Discovery Bay
Admiralty Bay
Oak Bay
Mutiny Bay
Useless Bay
Port Gamble
Dabob Bay and
Quilcene Bay
Lynch Cove^
Port Madison
Port Orchard
Quartermaster Harbor
Gig Harbor
Carr Inlet (incl .
Hale Passage)0
Case Inlet
Henderson Inlet
Dana Passage, Peale
Passage and Pickering
Passage
Eld Inlet
Totten Inlet
TOTAL FOR ALL AREAS
Approx. Area
(sq mi )
5.93
14.07
43.64
32.15
4.31
126.89
56.08
8.97
5.71
52.51
10.11
3.71
5.49
14.40
11.71
5.49
1.71
11.03
2.29
34.29
12.89
8.29
12.29
5.14
0.46
46.70
21.14
2.34
16.97
6.46
9.43
Triad3
1
1
3
2
1
4
3
1
1
3
1
1
1
1
0
1
0
0
1
2
1
1
1
1
1
3
1
1
1
1
J^
41
Water Quality
1
1
1
1
1
1
1
1
1
1
1
1
1
0
1
0
0
1
1
1
1
1
1
1
1
1
1
1
1
JL
27
a Triad stations include sediment chemistry, toxicity bioassay, and benthic
macroinvertebrate components.
b Inner portion of inlet northwest of The Great Bend.
c Example Bay. See text for explanation.
-------
Thirty-one stations are proposed for the Puget Sound basins (Figure 4).
Basin boundaries conform to those used by Cokelet et al . (1984) in their
hydrodynamic model of Puget Sound. No stations are located in Boundary
Pass, Haro Strait, Rosario Strait, or Admiralty Inlet, because currents
in these areas are very swift, and bottom types typically range from gravel
to boulders. The basin stations are generally arranged in transects of
three stations each across the basins. One station at the center of each
basin is flanked by two 20-m stations in all but two transects (see Appendix C
for rationale for station locations). The transects in the Strait of Georgia
and the Strait of Juan de Fuca (near Port Angeles) consist of only'two
stations each because the third station would cross the Canadian border.
Bay stations were apportioned by the type of bay (i.e., urban or indus-
trialized bay or rural bay) and the size of the bay. Elliott Bay and Carr
Inlet were chosen as example urban and rural bays, respectively, and stations
were assigned to each (Figures 5 and 6). The number of stations per unit
2
area is greater in Elliott Bay (one station/2.6 mi ) than in Carr Inlet
2
(one station/15.6 mi ) because of the more numerous and diverse sources
of anthropogenic contaminants that occur there. Moreover, one station
in the Elliott Bay array is located in the lower Duwamish River to assess
large-scale gradients from the inner harbor to the outer bay. Based on
ratios of stations per unit area similar to the foregoing, numbers of stations
were determined for all the remaining urban or industrialized bays and
rural bays (as listed in Tables 5 and 6). A minimum of one station was
apportioned to each bay. The only exceptions were Admiralty, Mutiny, and
Useless Bays. These bays were not assigned any stations, because their
substrates are primarily gravel and cobble.
Following initial station assignments, the allocation of stations
among bays was examined to determine whether the numbers of station appeared
to be appropriate for the level of human uses and activities that occurs
in each. In the cases of four urban and industrialized bays (Bellingham
Bay, Fidalgo Bay and Guemes Channel, Port Townsend and Kilisut Harbor,
and Port Gardner) and two rural bays (San Juan Islands and Skagit Bay),
numbers of stations were reduced because adjacent population centers are
26
-------
DEPTH OF STATION 20m
DEPTH OF STATION <20m
Alkl Point v
Figure 5. Proposed locations of monitoring stations in Elliott Bay,
an example urban bay.
-------
DEPTH OF STATION 20m
01234 kilometers
1 2 nautical miles
Figure 6. Proposed locations of monitoring stations in Carr Inlet,
an example rural bay.
-------
relatively small and the potentials for anthropogenic impacts are low compared
to the sizes of the bays (Tables 5 and 6). The final apportionment includes
31 basin stations, 34 urban and industrialized bay stations, and 41 rural
bay stations. It may be possible to locate some of these bay stations
at sites now sampled by Washington Department of Ecology or other agencies
for various monitoring purposes.
Bottomfish
Samples for analysis of the following monitoring components will be
taken concurrently from otter trawls
• Toxic chemicals in fish (fillet of English Sole, Parophrys
vetulus)
• Histopathological abnormalities in fish (English sole liver)
• Fish species abundances (bottomfish catch per unit effort)
Data from these components combined with the sediment quality "Triad" (Chapman
and Long 1983) form the basis for ranking toxic areas of concern in the
U.S. EPA and Washington Department of Ecology Urban Bay Action Programs
(e.g., Tetra Tech 1985a). As discussed in a later section, samples for
analysis of toxic chemicals in Pacific cod (Gadus macrocephalus) and Chinook
salmon (Oncorhynchus tshawytscha) will be collected separately from bottomfish.
Trawl stations are located at 16, 27, and 19 locations in the main
basins, urban bays, and rural bays, respectively (Figure 7) (see Appendix C
for rationale). Examples of station layouts for bays are shown in Figures 8
and 9.
Toxic Chemicals in Recreational Fishes
Concentrations of toxic chemicals in muscle tissue of Pacific cod
(Gadus macrocehalus) and Chinook salmon (Oncorhynchus tshawytscha) will
27
-------
•] BASIN
• RURAL
A URBAN
NOTE:
All stations are located as close as possible
to 20-m sediment quality stations.
10 20 kilometers
10 nautical miles
Figure 7. Proposed locations of bottomfish sampling stations for
the ambient monitoring program.
(continued on next page)
-------
Port Angtl**
BASIN
RURAL
URBAN
20 kilometers
10 nautical miles
NOTE:
All stations are located as close as possible
to 20-m sediment quality stations.
Figure 7. Continued.
-------
FISH TRAWL LOCATION
Figure 8. Proposed locations of bottomfish sampling stations for
Elliott Bay, an example urban bay.
-------
I FISH TRAWL LOCATION
01234 kilometers
Figure 9. Proposed locations of bottomfish sampling stations for
Carr Inlet, an example rural bay.
-------
be analyzed to monitor risk to humans from consuming recreational fishes
(see Appendix C for rationale). This program complements the bottomfish
and shellfish monitoring designs.
Pacific cod will be sampled at 15 nearshore recreational fishing locations
in 5 urbanized areas of Puget Sound (Figure 10). Salmon will be sampled
at three offshore recreational fishing areas in central Puget Sound (Figure 10)
(see Appendix C for rationale).
Shellfish
Toxic chemicals, PSP, and bacteria in butter clams (Saxidomus giganteus)
will be monitored in conjunction with assessment of butter clam abundance
and pathogen indicators in water.
Shellfish will be sampled at 26 recreational harvesting areas throughout
Puget Sound (Figure 11). All of these areas (except Station No. 26) correspond
to those selected by DSHS for analysis of tissue contaminants in bivalves
during a recent survey (see Appendix C for additional rationale). Station
No. 26 was added to cover the area between the Tacoma Narrows and Olympia.
Marine Mammals and Birds
Wildlife components proposed for the Puget Sound Monitoring Program
include:
• Marine mammal abundance and reproductive success
Abundances of harbor seals (Phoca vitulina, adults
and pups)
Reproductive success of harbor seals (e.g., indices of
no. pups adult~lyr~l, no. live births, no. stillbirths;
prevalence of abnormalities)
28
-------
1st St. dock
Port Orchard
ShcHon
• PACIFIC COD
O SALMON
Everett Waterfront
'EVERETT
Mukilteo ferry dock
Possession Point
Edmonds fishing pier
Shllshole Bay
6 SITES
Pier 48
Pier 56
Pier 70
Myrtle Edwards fishing pier
Spokane St. Bridge
Armenl boat ramp
Pt. Defiance
4 SITES
Les Davis fishing pier
Old Tacoma
City Waterway
Browns Point
20 kilometers
10 nautical miles
Figure 10. Proposed locations of sampling stations for toxic
chemicals in fish from recreational harvest areas.
-------
SHELLFISH MONITORING COMPONENTS
TOXIC CHEMICALS IN SHELLFISH
PSP IN SHELLFISH
PATHOGEN INDICATORS IN SHELLFISH
SHELLFISH ABUNDANCES
TARGET SPECIES - BUTTER CLAMS
15 •SEATTLE
14
NOTE:
Exact locations and their descriptions are
presented in Appendix C, Table C-2.
0 10 20 kilometers
10 nautical miles
Figure 11. Proposed locations of intertidal sampling stations for
shellfish monitoring components of the ambient
monitoring program, (continued on next page)
-------
Port Ang*l*«
20 kilometers
10 nautical mile*
SHELLFISH MONITORING COMPONENTS
TOXIC CHEMICALS IN SHELLFISH
PSP IN SHELLFISH
PATHOGEN INDICATORS IN SHELLFISH
SHELLFISH ABUNDANCES
TARGET SPECIES - BUTTER CLAMS
NOTE:
Exact locations and their descriptions are
presented in Appendix C, Table C-2.
Figure 11. Continued.
-------
• Avian abundances and reproductive success
Abundances of selected avian species associated with
marine habitats (e.g., pigeon guillemot, Cepphus columba;
rhinoceros auklet, Cerorhi nca monocerata); selected
waterfowl (e.g., black brant, Branta bernicula); and
great blue heron (Ardea herodias).
Sound-wide surveys should be conducted with special emphasis on haul-
out areas and pupping areas for harbor seals, and on breeding and foraging
areas for avian species. See Appendix C for specific survey locations
and rationale.
Rivers
The river monitoring program will contribute to assessment of nonpoint
sources (see later chapter, Nonpoint Source Monitoring). The proposed
river water quality monitoring components include:
t Flow (daily records at gaging stations)
• Dissolved oxygen
• Alkalinity, pH
• Temperature
• Conductivity
• Contaminants in water
Suspended sediments
Toxic chemicals
Nutrients
Fecal coliform bacteria
29
-------
Enterococci bacteria.
Considerations for future design of a biological monitoring program for
rivers and other nonpoint sources are presented in a separate section (see
Nonpoint Source Monitoring).
Proposed station locations for river monitoring are summarized in
Table 7. Fixed stations are located near the mouths of the 8 largest rivers
that discharge directly into Puget Sound and rotating stations near the
mouths of the 13 remaining rivers (Figure 12). The 13 primary stations
on smaller rivers will be sampled on a 3-yr rotating schedule, with 7 stations
sampled during the first 3-yr period and 6 stations sampled during the
second 3-yr period. Additional monitoring stations in the upper drainage
basins are recommended for the eight primary rivers. These upstream monitoring
stations are also scheduled on a 3-yr rotating basis because it allows
more complete coverage of the entire Puget Sound Basin. Detailed design
of upstream sampling stations has been completed for the Skagit and Green-
Duwamish River Basins (Figures 13 and 14). The 11 upstream stations selected
in these two river basins will be sampled during the first 3-yr period.
The rationale for station locations is provided in Appendix C.
Selecting different river basins for detailed upstream monitoring
every 3 yr provides maximum coverage of the entire Puget Sound Basin.
For example, after the first 3-yr period, data will be available for 15
primary rivers and 11 upstream or secondary tributaries. After 6 yr, the
contaminant data will be available for all 21 primary rivers and 22 secondary
river basins.
Habitat Types
Shoreline habitats of the Puget Sound Basin will be classified and
mapped as part of this monitoring effort. The primary monitoring variable
is the areal extent of each kind of habitat in each region (basin or bay).
Puget Sound habitats to be addressed include salt marshes, eel grass beds,
30
-------
TABLE 7. SUMMARY OF PUGET SOUND
RIVER MONITORING STATIONS
Primary stations: sample near mouth
- 8 core stations—monitor every year
Skagit Rivera»b
Snohomish River
Stillaguamish Riverb
Nooksack Riverb
Puyallup River3
Green-Duwamish Riverb»c
Sammamish-Cedar Rivers (ship canal)b
Deschutes Riverb
- 13 rotating stations—monitor on 3-yr rotating basis
First 3-yr Period Second 3-yr Period
Nisqually Riverb Dosewallips River
Elwha River Duckabush River
Skokomish lb Hamma Hamma River
Skokomish 2^ Dungeness River
Samish Riverb Big Quilcene River
Whatcom River Little Quilcene River
Tahuya River
Secondary stations: 11 stations upstream in selected drainage
basins (3-yr rotating schedule)6
- Skagit River Basin
Skagit River at Marblemountb
Skagit River near Concreteb (below Baker River)
Skagit River near Sedro Wooley
Sauk River near Rockportb
- Green-Duwamish River Basin
Green River at Auburnc
Green River at Palmerc
Soos Creekc
Newaukum Creekc
Mill Creek above Western Processingb
Mill Creek below Western Processingb
Jenkins Creek
a Station currently monitored by U.S. Geological Survey.
b Station currently monitored by Washington Department of Ecology.
c Station currently monitored by Metro.
d Skokomish 2 is outlet from the Cushman powerhouse No. 2.
6 Only stations proposed for the first 3-yr period are shown.
-------
CANADA
B.C.
—•*— Drainage Basin Boundary
PRIMARY STATION LOCATIONS
® Core Stations - Monitor Every Year (12 times/year)
* Rotating Stations - Monitor on a 3 Year Rotating Schedule
Figure 12. Primary downstream station locations on major
rivers in Puget Sound drainage basin.
-------
o
USGS/Washington DOE
sampling stations
Sampling stations recommended
for the Puget Sound Program
B.C.
20 MILES
a
20 KILOMETERS
Figure 13. Skagit River Basin.
-------
• Metro/Washington DOE Sampling Stations
O Stations Recommended for Puget Sound River Monitoring Program
Figure 14. Green-Duwamish River Basin.
-------
kelp beds, and other selected habitats classified according to the Washington
Coastal Zone Atlas scheme (a modification of U.S. FWS classification system).
The habitat maps will cover all of the Puget Sound shoreline and riparian
habitat associated with major rivers.
Ancillary Data
Ancillary monitoring data will be collected for all of Puget Sound.
Information on demographic/socioeconomic conditions and management decisions
will also be collected for watersheds of the Puget Sound Basin. Sound-
wide coverage for ancillary variables is needed to support management activities
and to interpret other monitoring data. Specific stations for climate
and weather variables are listed in Table 3 and Appendix C.
ONGOING MONITORING
The major ongoing ambient monitoring programs in the Puget Sound Basin
are described in Table 8. These monitoring programs were selected because
they exhibit:
• Extensive spatial or temporal coverage of monitoring variables
relevant to the Puget Sound Monitoring Program
t Potential for contributing to the Puget Sound monitoring
database
• High likelihood of continuation.
For more detailed descriptions of these and other monitoring programs,
refer to Puget Sound Water Quality Authority (1986a) and Chapman et al. (1985).
The development of monitoring requirements as part of the Puget Sound
Dredged Disposal Analysis (PSDDA) program is timely because it coincides
with the development of the Puget Sound Monitoring Program. For the monitoring
31
-------
TABLE 8. ONGOING AMBIENT MONITORING PROGRAMS IN PUGET SOUND
Program and
Location of
Stations
No. Of
Stations
Variables3
Frequency
(Times/Yr)
Start of
Record QA/(
SEDIMENT QUALITY
NOAA National Status and Trends
Nisqually Reach (1984)
Budd Inlet (1986)
Commencement Bay (1984)
Elliott Bay (1984, 1986)
Bainbridge Is. (1986)
Whidbey Is. (1986)
Pt. Roberts (1986)
Bellingham Bay (1986)
Cape Flattery/Neah
Bay (1986)
33C
PSDDA
Disposal sites
Reference sites
Variable
PAHs, PCBs,
pesticides,
grain size,
carbonate,
chlorinated
coprostanol,
organic carbon,
17 trace elements
1984
Ves
Puget Sound contaminants of
concern
Every 2-4 yr 1987
Yes
WATER QUALITY
GENERAL SURVEYS
Metro
Offshore main basin 32
and bays
Ouwamish Head NAb
Ecology
South Puget Sound 44
Central Puget Sound
North Puget Sound
Strait of
Juan de Fuca
Haro Strait
Temperature, salinity, DO,
transparency, fecal coliforms,
total coliforms, tide,
suspended solids
NA
Temperature, salinity, DO,
pH, turbidity, transparency,
monthly fecal coliforms,
total fecal coliforms,
nutrients (NH3, N02, M
total and dissolved O-P
SWL, chlorophyll a, TOC.
Samples at 0 m, 10 m, 30 m
1965 Yes
26
(select
stations)
NA
12
1987 Yes
1967 Yes
for most
stations
-------
TABLE 8. (Continued)
U.S. DOO
Bangor, Hood Canal 1-20
Temperature, salinity, DO,
pH, nutrients (NH3, N02, N03,
total 0-P04
1974
No
PATHOGEN INDICATORS
Metro
Beach stations, 25
Central Puget Sound
DSHS
Station locations
vary
Temperature, fecal coliforms,
total coliforms (historically),
enterobacteria, tide
Variable Total and fecal coliforms
52
1970 .
for most
stations
1978
Yes
Yes
BIOLOGICAL CONDITIONS
BENTHIC MACROINVERTEBRATE ABUNDANCES
PSDOA
Disposal sites
Reference sites
Variable Total abundance, depth
profile of abundance
Every 2-4 yr 1987
Yes
TOXIC CHEMICALS IN FISH TISSUE
NOAA National Status and Trends
Nisqually Reach 9
Commencement Bay
Elliott Bay
Trace and major elements; PAHs,
PCBs, other organic HCs; PAH
metabolites; food organisms;
percent lipids in stomach
contents, liver, and bile
1984
Yes
HISTOPATHOLOGICAL ABNORMALITIES IN FISH
NOAA National Status and Trends
Nisqually Reach 9 Histopathology and visible lesions
Commencement Bay
Elliott Bay
1984
Yes
-------
TABLE 8. (Continued;
FISH SPECIES ABUNDANCES
WDF
Salmon run size Variable Returns and escapement
Juvenile Pink
and Chum Salmon
Herring Spawning
Ground
Surf Smelt
Spawning Ground
areas open to
harvest
Hake Abundance
Port Susan
5 Abundance
Variable Biomass of spawn
Variable Biomass of spawn
4 Biomass
1
March-July
Variable
Variable
1965 Data
for most entry
species only
1970
1965
1980
1975
No
Yes
Yes
No
SHELLFISH SPECIES ABUNDANCES
WDF
Shellfish abundance Variable
representative
sport beaches
Geoduck Stock Variable
Assessment
commercial
harvest areas
Oyster Larvae 7
Oabob Bay
Shrimp Abundance Variable
Hood Canal
Sea Urchin Variable
Abundance
Strait of
Juan de Fuca
commercial
harvest areas
Abundance
Abundance, size, and weight
composition
Abundance, age, population
structure
Abundance, size, distribution,
catch per unit effort
Abundance
1 1974
1 1972
varies by
site
16 or more 1949
July and
August
1 1977
April -May
Variable 1975
No
No
No
No
No
-------
TABLE 8. (Continued)
IQXJCLCHEMICALS IN SHELLFISH TISSUE
NOAA National Status and Trends
24
Budd Inlet
Commencement Bay
Elliott Bay
Bainbridge Is.
Whidbey Is.
Pt. Roberts
Bel 1ingham Bay
Cape Flattery
PSDDA
Disposal Sites
Reference Sites
Variable
Aromatic hydrocarbons, pesticides,
and PCBs in mussels
Puget Sound contaminants of
concern relevant to bioaccumu-
lation or human health concern**
1976 Yes
(3 sites)
(U.S. EPA)
1986
(NOAA)
Every 2-4 yr 1987
Yes
PSP IN SHELLFISH
DSHS
Commercial and
recreational
shellfish beds
Variable Toxin concentration in tissue
Variable 1978
Apri1-October
Yes
PATHOGEN INDICATORS IN SHELLFISH
DSHS and Counties
Commercial and Variable Total and fecal coliforms
recreational
shellfish beds
52
(commercial)
but location
may vary
1978
Yes
MARINE MAMMAL ABUNDANCE AND REPRODUCTIVE SUCCESS
WDG
Harbor Seals
Sound-wide
—e Abundance by species in
haulout areas
1980
No
-------
TABLE 8. (Continued)
AVIAN ABUNDANCE AND REPRODUCTIVE SUCCESS
WDG
Mid-winter
Waterfowl Survey
Sound-wide
Waterfowl Brood
Production
Sound-wide
Waterfowl Harvest
hunting areas
Sound-wide
USFVS
Waterfowl Survey
South Puget Sound
and North Puget Sound
Abundance by species
Estimated number of breeding pairs
and brood size by species
Estimated number of birds taken
by species, by area
Abundance by species
1
October-
March
1955
1974
1974
NA
NA
No
1978
No
WDOE/U.S.GS
Near mouths of
Puget Sound rivers
Metro
Green, Cedar,
Sammamish Rivers;
Lake Washington
Ship Canal; Lake
Washington Creeks
RIVER MONITORING
38 Flow, temperature, DO, turbidity, 12
suspended solids, conductivity,
pH, color, ammonia, nitrate,
nitrite, orthophosphate, total
phosphate, fecal coliforms, heavy
metals (8 stas.), COO (8 stas.),
hardness (10 stas.), organics
(2 stas.), priority pollutants
(selected stas.)
46 Temperature, DO, pH, alkalinity, 12
nutrients, fecal coliforms,
chlorophyll, metals, conductivity,
turbidity, suspended solids, benthic
communities (1/yr)
Variable Yes
1979 Yes
-------
TABLE 8. (Continued)
HABITAT
FISH HABITAT
WDF
Sound-wide
WDF
Salmon Catch
Groundfish Catch
Herring Catch
Shell fish harvest
20
NOAA (NWS)
Puget Sound Basin 23
Abundances of fish, seaweeds,
invertebrates, fish diet,
community structure; in arti-
ficial and natural reefs, mud
flats, tidegrass, kelpbeds, etc.
CLIMATE/WEATHER
Temperature (air), precipitation,
wind, sky cover (varies among
stations)
FISHERIES HARVEST
Sport and commercial catches
Commercial catches
Commercial catches
Commercial harvest
4
Varies
site
by
Variable
±2 yr
per site
1975
No
Hourly
to daily
Varies
by
station
(1890-
1980)
1910
1921
1980
1935
Yes
Data
entry
only
Data
entry
only
Data
entry
only
Data
entry
only
-------
TABLE 8. (Continued)
LANO_CO\/E^R_/LAND JJSE
U.S.GS
Land use and land
cover mapping,
nation-wide
DEMOGRAPHIC AND SOCIOECONOMIC CONDITIONS
9 general level categories,
37 specific categories;
available with political units,
hydrologic units; census county
subdivisions; federal land
ownership. 1:250,000 and
1:100,000 scales
To be
determined
1986
Yes
a 00 = Dissolved oxygen
0-P04 = Orthophosphate
NH3 = Ammonia
N02 = Nitrite
N03 = Nitrate
SWL = Sulphite waste liquor
TOC = Total organic carbon
PAHs = Polynuclear aromatic hydrocarbons
PCBs = Polychlorinated biphenyls.
b NA = Not available at time of writing.
c Three stations per site, except Elliott Bay and Commencement Bay (six stations per site).
For other components of the NOAA Status and Trends program, sampling stations are evenly divided
among sites.
d Target species may include benthic infauna other than shellfish.
e -- = Not applicable.
-------
and intensive survey data collected by PSDDA and other agencies to best
complement the programs discussed in this document, it is recommended that
sampling and analysis procedures be similar to those adopted for the Puget
Sound Monitoring Program (see Appendix C).
32
-------
COMPLIANCE MONITORING PROGRAM
Compliance monitoring involves the collection and analysis of data
from point-source discharges and nearby receiving environments. The major
objectives of compliance monitoring are to:
• Determine whether federal or state discharge criteria are
met. These criteria may be set forth as specifications
of discharge permits, as water quality standards, or as
sediment quality values.
t Determine whether unacceptable impacts are likely to occur
in the receiving water.
• Determine whether unacceptable biological impacts are present
in the receiving environment.
• Determine the effectiveness of regulatory programs and management
decisions.
Monitoring of combined sewer overflows and stormwater discharges should
be included in any proposed compliance monitoring program. Thus, industrial
nonpoint sources that eventually discharge to a receiving system via well-
defined channels or pipes should be covered by the program (U.S. EPA 1985a).
U.S. EPA and the Washington Department of Ecology are designing specific
monitoring programs to ensure compliance of NPDES discharges with permit
conditions. This process is concurrent with restructuring of the permit
system to incorporate a toxics control strategy (e.g., U.S. EPA 1985b)
and increased monitoring, especially biomonitoring of effluents. The remainder
of this section provides some general recommendations for the design of
compliance monitoring programs.
33
-------
The compliance monitoring program for major and significant minor
dischargers should include evaluations of both effluent and the sedimentary
receiving environment. Effluent analyses provide a direct assessment of
the characteristics of each discharge and the potential for biological
impacts. However, because the characteristics of many discharges vary
over time, periodic effluent analyses may not accurately quantify long-
term contaminant loadings. By contrast with effluent evaluations, measurements
of the sedimentary receiving environment represent cumulative, long-term
assessment of discharge effects. Detailed rationale for use of effluent
and environmental monitoring is provided by U.S. EPA (1985b).
The effluent evaluation should include chemical analyses and assessments
of toxicity. Toxicity assessments should employ at least three kinds of
bioassay that cover a range of organisms (i.e., plants, invertebrates,
vertebrates). Evaluations of the sedimentary receiving environment should
include chemical analyses, sediment bioassays, and assessments of benthic
invertebrate assemblages. The evaluation of the receiving environment
for each discharger should be conducted at a minimum of three stations.
These stations should represent conditions at the discharge point (worst
case), in a moderately impacted nearfield area, and in a reference area.
Larger dischargers may require additional stations.
To use all of the indicators mentioned above for compliance monitoring,
criteria must be developed that can be included in discharge permits.
These criteria should specify the critical values for each indicator that
determine an unacceptable adverse impact. Because few of the indicators
have accepted criteria at present, their use in compliance monitoring will
require an initial investment in criteria development.
To maximize the usefulness of data collected during compliance monitoring,
the program should be designed to complement the long-term ambient monitoring
program for Puget Sound. Complementarity between programs can be enhanced
by using similar indicators, sampling techniques, and analytical methods.
The ambient program is designed to assess integrated effects and cumulative
34
-------
long-term impacts of many sources, whereas the compliance program is designed
to assess effects of single sources. Although the specific objectives
of the two programs differ, data collected for one may augment interpretation
of data collected for the other.
The generalized strategy for compliance monitoring may be applied
at varying sampling frequencies, depending upon discharge-specific charac-
teristics (e.g., flow and composition of effluent). Some of the specifications
within the overall strategy are currently being developed. Major specifications
that require future resolution include:
0 What organization(s) will conduct the program; including
effluent measurements, monitoring of the receiving environments,
data management, and quality assurance (see below, Institutional
Options)?
• How will the program be integrated into the overall regulatory
process?
• What specific water and sediment quality criteria will be
used?
• How will dilution zones be defined?
The determination of which dischargers must undertake compliance monitoring
can be made using several approaches. The most straightforward approach
is a blanket requirement that all discharges of a specified kind (e.g.,
major, significant minor) be required to conduct compliance monitoring.
A more detailed approach m-ight include characterization of effluent to
determine the likelihood of adverse environmental impacts. The decision
as to whether a discharger must undertake compliance monitoring would be
based on the results of the effluent characterization.
U.S. EPA (1985a) has recommended a screening-level approach to effluent
characterization. The approach involves chemical analyses of effluent
35
-------
and toxicity assessments. Dischargers that pass the screening evaluation
would not be required to conduct compliance monitoring, whereas dischargers
that fail this evaluation would be required to conduct such monitoring.
For dischargers that fail the screening evaluation, additional site-specific
data collection and analysis may be required (e.g., critical flow, fate
modeling, mixing) to derive the detailed permit specifications upon which
the compliance monitoring would be based.
36
-------
INTENSIVE SURVEYS
Because intensive surveys address specific issues or problems, they
are usually performed on an "as needed" basis. Because the objectives
of intensive surveys are site-specific, their designs are developed individually
based on the type of problem and the characteristics of the receiving
environment. Most intensive surveys that will be performed as a part of
the Puget Sound Monitoring Program will address problems of the following
types:
0 Define the areal extents and magnitudes of existing impacts
to water quality, sediment quality, and biological conditions
• Determine the contributions of point and nonpoint sources
of pollutants to existing impacts
• Predict the general impacts of planned resource use or waste
disposal on water quality, sediment quality, and biological
conditions
• Develop wasteload allocations
• Assess the impacts of specific pollutants on individual
biological populations of concern (e.g., the effects of
PCB contamination on the reproductive success of harbor
seals)
• Determine the need for long-term ambient monitoring in areas
that are suspected of being contaminated, but have not been
sampled.
37
-------
Five features should be common to all intensive surveys (U.S. Environmental
Protection Agency 1976). First, data collected at intensive survey stations
should help explain the spatial significance of data collected at nearby
ambient monitoring stations. Second, if any mathematical models are to
be used or statistical hypothesis are to be tested, the design of the intensive
survey should be adequate to provide the type, quality, and quantity of
data needed to satisfy the requirements of the models or tests. Third,
all substantial point sources of pollutants should be addressed in the
design of the intensive survey. Substantial nonpoint sources should also
be addressed to the extent that their presence is known or suspected.
Fourth, the distributions, concentrations, and accumulations of toxic substances
in the receiving environment and biota should be determined where such
toxic substances are present. Finally, all intensive surveys should consider
historical data (when available) during the design, data analysis, and
data interpretation phases of the survey.
38
-------
NONPOINT SOURCE MONITORING
In the proposed Puget Sound Monitoring Program, NPDES compliance monitoring
of stormwater discharges and ambient monitoring of major rivers are included
as complementary components of a nonpoint source monitoring program. Neverthe-
less, further development of a monitoring design for nonpoint sources within
the Puget Sound Basin is needed. Elements of a conceptual framework for
nonpoint source monitoring are described below. Monitoring of selected
variables could be performed throughout the entire Puget Sound Basin.
Alternatively, intensive monitoring of selected sub-basins could be implemented
using the integrated watershed monitoring approach discussed below. Because
biological monitoring of nonpoint sources should be a key element of integrated
watershed monitoring, it is addressed further in the final section.
INTEGRATED WATERSHED MONITORING
The first step in developing an integrated watershed monitoring strategy
is to define the objectives of the program (Figure 15). Some of the general
objectives defined earlier in this report apply to nonpoint source monitoring.
Specific objectives analogous to those defined for regional 208 programs
(e.g., Metro 1978; Ward et al. 1986) and for maintenance of ecological
integrity (Karr and Dudley 1981) need to be developed if the Interagency
Management Group on monitoring decides to proceed with a watershed program.
The second step is to select a watershed modeling and analysis framework
(e.g., Skopp and Daniel 1978; Horner et al. 1986). The selected approach
could incorporate conceptual models applied to the Hubbard Brook ecosystem
(e.g., Likens et al. 1970), the Lake Washington drainage basin (e.g., Edmondson
and Lehman 1981; Wissmar et al. 1982), and agricultural watersheds (e.g.,
Karr and Schlosser 1978). For selected contaminants of concern, a multilevel
modeling approach (e.g., Horner et al. 1986) could be applied to watersheds
within the Puget Sound Basin.
39
-------
DEFINITION OF
OBJECTIVES
A
V
A
I
L
A
B
L
E
D
A
T
A
INTEGRATED WATERSHED
MODELING AND ANALYSIS
MONITORING PROGRAM
DESIGN AND EVALUATION
MONITORING PROGRAM
IMPLEMENTATION
LAND USE/HABITAT
MONITORING
SITE STORMWATER
MONITORING
RIVER/STREAM
MONITORING
HAZARDOUS WASTE
SITE MONITORING
MONITORING
DATA BASE
NPDES POINT
SOURCE MONITORING
Figure 15. Components of Integrated Watershed Monitoring
Program.
-------
Nonpoint source monitoring is incorporated into several components
of the integrated watershed monitoring approach (Figure 15), i.e.:
t Land-Use Monitoring—Elements of land-use monitoring could
include watershed land-use mapping based on record-keeping
by counties, remote sensing and mapping of habitat types,
and identification of critical areas. The proposed Puget
Sound Monitoring Program incorporates land-use monitoring
and habitat monitoring (see Apprendix C, Demographics and
Socioeconomic Conditions, Habitat Types). The proposed
Puget Sound Management Plan includes identification of critical
areas for nonpoint pollution (PSWQA 1986c). Guidelines
for identification of critical areas are given by Maas et
al. (1985) and Horner et al. (1986).
• Site Stormvater Monitoring—Monitoring of chemical constituents
of stormwater runoff at individual sites will eventually be
incorporated into NPDES permits (U.S. EPA 1985a). Monitoring at
additional sites may be needed (e.g., critical areas identified
by the land-use monitoring component, wetlands used as stormwater
filters). Moreover, biological monitoring may be applied to
runoff from specific sites as well as to larger streams and
rivers (see below, Biological Monitoring of Nonpoint Sources).
t River and Stream Monitoring—A proposed program for monitoring
water quality and biological conditions in rivers and streams
is described in Appendix C.
• Hazardous Waste Site Assessment—Information on upland waste
sites could be used to predict the effects of nonpoint source
pollution and to select control measures. Some information
will already be available from ongoing RCRA or Superfund
site investigations. Upland assessments could include chemical,
biological, and other analyses (e.g., Peterson et al . 1985;
Miller et al. 1985).
40
-------
In addition to nonpoint source assessment, monitoring of point sources
required through NPDES permits should be an element of an integrated watershed
monitoring program.
Some of the monitoring components listed above are already incorporated
into the proposed Puget Sound Monitoring Program. In other cases, monitoring
designs would have to be developed further. For example, monitoring of
land use, stormwater discharges, and biological effects at critical areas
within selected watersheds must await identification of those critical
areas. Some remaining issues that should be addressed by the interagency
management group and the technical working committee include:
• What are the objectives of a nonpoint source monitoring
program for watersheds within the Puget Sound Basin?
• What watersheds should receive a high priority for nonpoint
source pollution monitoring? What level of stream order
should be monitored?
• What model(s) and analyses should be applied to specific
nonpoint source problems in different kinds of watersheds?
• What monitoring variables and designs should be applied
to specific nonpoint source problems in different kinds
of watersheds?
t How should relative effort be allocated among potential
monitoring components?
Other issues in nonpoint source monitoring were discussed by Homer et
al. (1986). Considerations for biological monitoring of nonpoint sources
are provided in the next section.
41
-------
BIOLOGICAL MONITORING OF NONPOINT SOURCES
Biological monitoring of nonpoint sources may include assessment of
bioaccumulation of toxic substances, bioassays of water or sediment, and
community analysis (e.g., periphyton, macroinvertebrates, or fish). Incor-
poration of biological components into a monitoring program for nonpoint
sources would be valuable because:
• Monitoring of biological variables via bioassays or community
analysis can provide a direct assessment of contaminant
effects on living resources.
• Measurement of bioaccumulation provides information about uptake
and retention of bioavailable contaminants integrated over a
selected time period on the order of weeks to years. In
contrast, a single chemical analysis of water or suspended
sediment yields data for a given point in time (grab sample) or
a short time period (on the order of 1 h to 1 day for a composite
sample).
• Biological components may reflect physiological or ecological
responses to multiple chemicals as well as undetected chemicals.
Chemical analyses are limited to selected toxic substances.
Moreover, prediction of antagonistic or synergistic effects
from chemical data is difficult.
Bioaccumulation
Initiation of a freshwater bioaccumulation monitoring program is considered
a high priority. Concentrations of toxic contaminants in tissue of indigenous
or transplanted organisms could be determined on an annual basis. The
target species should be consistent within a drainage basin and among basins
to the extent practical. Use of freshwater bivalves (e.g., Margaritifera
spp. or Corbicula spp.) is recommended. Bivalves are preferable to fishes
because they are relatively stationary and because they are expected to
42
-------
bioaccumulate a variety of metals and organic compounds (especially aromatic
hydrocarbons) to a greater degree than do fishes. Bioaccumulation of metals
and PCBs by the clam Corbicula fluminea has been demonstrated in laboratory
experiments (Tatem 1986; Belanger et al. 1986). The State of California
has conducted a pilot "mussel watch" survey using Corbicula f1uminea.
Assessment of bivalve abundance and growth could complement the bioaccumulation
monitoring component. A pilot survey will be needed to establish the distribu-
tion and abundance of large bivalves in areas selected for monitoring.
Moreover, the options of using bivalves vs. fishes and indigenous organisms
vs. transplanted organisms need to be evaluated further.
Periphyton
Assessment of periphyton community composition and species abundances
is recommended as a primary tool for monitoring biological conditions in
rivers. Because indigenous populations are highly variable due to natural
disturbances, use of settling plates (e.g., a diatometer, Patrick et al. 1954)
can avoid problems of sampling and data interpretation. This technique
is inexpensive relative to most biological community analyses. Moreover,
interpretation of ecological effects of anthropogenic contaminants would
be facilitated by past research and past application of the technique in
monitoring programs throughout the world (Patrick 1971; Weber 1980).
Sediment Chemistry, Sediment Bioassays, and Benthic Macroinvertebrates
Development of a freshwater "triad" approach analogous to the technique
recommended by Chapman and Long (1983) for marine waters could be valuable
for selected depositional areas. However, sediments in most areas of rivers
and streams are unstable, making the application of this approach questionable.
Assessment of benthic macroinvertebrate communities in erosional areas
is not recommended because of their variability and sampling problems.
Further research is needed before a freshwater "triad" approach can be
applied to routine monitoring. Additional research on methods to trap
suspended sediments for chemical analyses and bioassays using freshwater
amphipods (Gammarus spp. or Hyallei a spp.) would be valuable.
43
-------
Water Bioassays
In situ toxicity tests and early warning systems (Cairns and van der
Schalie 1980) are promising techniques for monitoring nonpoint sources.
Juvenile stages of salmon or trout enclosed in field "live boxes" could
be placed in runoff streams to measure acute or chronic responses. However,
influences of temperature, suspended sediment concentration, and other
environmental factors would need to be evaluated to distinguish contaminant
effects from natural factors. Further development of in situ bioassays
may be needed before they are applied to routine monitoring. Laboratory
bioassays of water samples are not recommended for routine applications
because of their expected variability and limited interpretive value.
Fish Community Analysis
Analysis of fish communities in streams and rivers may be valuable
for biological monitoring programs with certain objectives (Karr 1981).
However, fish communities may be less sensitive to anthropogenic contaminants
than invertebrate or periphyton communities are. Interpretation of contaminant
effects on fishes is confounded by the impact of fisheries harvest. Stock
assessments of fishes are relatively expensive because fish are long-lived
and highly mobile. Assessment of fish communities is not recommended as
a primary component of a nonpoint source monitoring program. However,
ongoing salmon assessments conducted by the Department of Fisheries should
continue and be coordinated with other aspects of a biological monitoring
program. Intensive surveys by fisheries biologists of Northwest Indian
Tribes could also provide valuable information to complement a long-term
monitoring program. Finally, spawning habitat assessment and mapping should
be a high priority for ongoing and future monitoring.
Fish Histopathology
Monitoring of liver lesions in selected fish species could be a valuable
monitoring tool for rivers and streams. However, based on analyses of
44
-------
data for marine species, it is anticipated that a sample size of 60 specimens
is needed to provide adequate statistical power for detection of trends
in lesion prevalence over space or time (see Appendix C, Histopathological
Abnormalities in Fish). The logistic difficulties and costs involved in
collecting adequate sample sizes could preclude use of histopathology in
rivers and streams. Further work is needed to evaluate potential target
species. Results of the recent study of fish histopathology in the Yakima
River by the Washington Department of Ecology will be useful for further
evaluation of this component.
45
-------
DATABASE MANAGEMENT SYSTEM
A discussion and evaluation of options for management of data from
the Puget Sound Monitoring Program is presented in this section. The goals
of a data management system are stated, and the general advantages, disad-
vantages, and relative costs of several existing and potential database
systems are discussed.
An initial assessment of the data management needs of the Puget Sound
Estuary Program was conducted in 1985 (Tetra Tech 1985d). While that assessment
did not focus specifically on the development of a database to store data
from a Puget Sound Monitoring Program, Tetra Tech did recommend the development
of a central database for Puget Sound data. The central database would
be used in developing management scenarios, calculating sound-wide effects,
and developing sound-wide models.
DATA MANAGEMENT SYSTEM GOALS
The data management system for the Puget Sound Monitoring Program
should:
• Provide access to high-quality, pertinent data in as simple
and efficient a manner as possible
• Provide wide access to data by storing it in a centralized
database and/or facilitating data transfer among agencies
and groups collecting data
• Enable agencies and governmental entities involved in the
Program to maintain control over the capabilities and use
of the data management system
46
-------
• Characterize the quality of data being entered into the
system, thereby ensuring that information of appropriate
quality is used for management decisions
t Meet new and changing needs of the Puget Sound Monitoring
Program by being flexible and expandable
• Maximize compatibility with the data management system eventually
chosen for the U.S. EPA National Estuaries Program.
KEY ISSUES
Two key questions, as yet unanswered, will have a major impact on
the choice of a data management system for the Puget Sound Monitoring Program.
These questions are:
Who will be the direct users of the data management system?
Information collected in the Puget Sound Monitoring Program will ultimately
be used by managers and decision-makers for assessments of the condition
of the sound. It is assumed that the database itself will not be accessed
directly by these managers, but will be used by technical staff who are
more familiar with computers. Nevertheless, a remaining issue is whether
the database should be accessible to most technical staff or only to computer
specialists. To serve the former group, a database must be designed to
provide extensive menus and guidance for a user. This will substantially
increase the amount of programming required and the costs of system develop-
ment. However, if the latter group (i.e., computer specialists) are to
be the primary system users, then the database could be designed with less
extensive menus, less programming effort, and lower cost. If only computer
specialists have access to the system, then most technical personnel as
well as managers would be unable to access data and perform analyses on
an ad hoc basis.
47
-------
Win the data management system be used often to provide reports and tables
of data? Statistical analyses? Graphics? Maps?
Typical system outputs that may be needed fall into two general cate-
gories: specific technical reports with extensive data compilation and
analyses, and summary reports of overall monitoring program results. Agencies
that are responsible for collecting and analyzing monitoring data for specific
dischargers, areas, or resources are likely to use data reporting and formatting
capabilities to sort, list, and present data in various ways. Agencies
or other entities preparing sound-wide reports are likely to use graphs,
analyses, and maps to summarize trends and patterns in data.
The issue of what types of products or outputs will most often be
requested is crucial for determining what type of system will be most appro-
priate. For example, if 90 percent of the outputs needed are presentations
of various combinations, subsets, and types of data in tables, then a system
that provides both standard reports and flexible, ad-hoc, non-standard
data retrieval is likely to be most suitable. Graphs, statistical analyses
and maps that are required less frequently could be provided by linking
the database with other commercially available products for mapping or
statistics. Alternatively, if 90 percent of the needed outputs are maps,
then a database system incorporating mapping capabilities becomes more
desirable. The cost of developing a system with mapping capabilities is
likely to be substantially greater than costs for a system without such
built-in capabilities.
SYSTEM COMPARISONS
An evaluation of approaches to database management systems is summarized
in Table 9. The criteria presented in the table parallel the system goals
and key issues discussed above. Each approach is scored on a three-point
scale. The existing systems were evaluated for their current capabilities;
the potential systems were evaluated for their maximum potential capabilities.
The cost scores indicate the relative costs required to maximize the capa-
bilities of each system.
48
-------
TABLE 9. SUMMARY OF SCORES FROM AN EVALUATION OF
DATABASE MANAGEMENT APPROACHES3
Development of
Existing Databases Database Development Products1* Data Transfer
ODES Storet SAS Files ARC/Info DBMSC Formats
Criteria
Local control
Qual ity assurance
procedures
Ease of use
Reporting
capabilities
Statistical
capabilities
Graphics/mapping
Data in SAS files
Flexibil ity/
expandability
Cost advantage6
1
3
3
2
3
2
3
2
2
1
1
2
1
1
1
1
1
2
3
3
1 (3)d
2
3
2
3
2
3 (l)d
3
3
3
3
1
3
1
2
1
3
3
3
3
2
2
1
3
1
3
2
1
1
1
1
1
2
3
a Score: 1 = Low; 2 = Moderate; 3 = High relative advantage.
b Scores represent maximum potential score for database developed using each product.
c Other commercial database management systems (e.g., FOCUS).
d A score in parentheses is for SAS files with ODES-like user-interface.
e Cost scores reflect relative costs required to maximize capabilities of system
with a high score indicating low cost.
-------
Developing a data management system for Puget Sound monitoring data
could cost between $100,000 and $1.5 million, depending on the features
that such a system would have, and the nature of the computer-user interface.
For example, costs for development of the U.S. EPA Ocean Discharge Evaluation
System (ODES) have been approximately $1 million. Rough cost estimates
for the various approaches are summarized in Table 10. Two estimates are
provided for each approach: low-end and high-end. The low-end estimates
describe costs for a system that would minimally meet data management needs
of the Puget Sound Monitoring Program, and would be useable by an experienced
computer specialist. The high-end estimates describe costs for a usei—
friendly system with some advanced capabilities for data retrieval, presen-
tation, and analysis. Cost estimates for STORET were not available at
the time this document was produced.
These cost estimates should be used only for general comparison, and
do not represent comparisons of costs for implementing identical systems.
For example, the low-end cost figures for ARC/INFO include some costs for
developing programs that use geographic data; it was assumed that any ARC/INFO
system developed would include some type of mapping capability. Low-end
estimates for other systems do not include mapping capabilities. These
estimates do not include system expansion or maintenance costs, or user
costs. Cost estimates for adding monitoring data to a database are included
in total cost estimates for the monitoring program (see below, Estimated
Costs of Monitoring Puget Sound). Characteristics of existing and potential
database systems are evaluated in the next section.
OPTIONS
The alternative database systems for the Puget Sound Monitoring Program
include two existing databases and several commercially available database
development products. Information presented here is based largely on system
evaluations conducted by Tetra Tech (1985d). Only the general advantages,
disadvantages and costs of these approaches are outlined below. More detailed
analyses will be necessary to describe the exact nature, format, and costs
49
-------
TABLE 10. ROUGH COST ESTIMATES (IN THOUSANDS OF DOLLARS)
FOR SEVERAL APPROACHES TO DEVELOPING A DATA MANAGEMENT
SYSTEM FOR PUGET SOUND MONITORING DATA
Low-end
System
Cost Description
High-end
System
Cost
Description
ODES
SAS files
ARC/Info
15-30 Existing ODES with 60-120
new file types and
basic file-type
data retrieval
10-20 Basic data structures 300-750
for data analysis,
used by experienced
data processors
150-300 Basic data structures, 400-1,000
some user menus, some
mapping by experienced
users
New file types and
retrievals plus some
additional analyses
specific to Puget
Sound
Simple user interface
to do retrivals, some
statistics, graphics,
and mapping
Database with extensive
user i nterface, high
quality mapping and
link to stastical packages
Other DBMS3
100-200
Development of
standard data
transfer formats 10-20
Basic data structures, 300-600
some user menus, no
statistics or mapping
Standard formats 25-50
defined for limited
number of data types,
transfer programs
available for major
agency databases
Database with extensive
user interface and
link to statistical
and mapping packages
More extensive definition
of formats, and develop-
ment of transfer pro-
grams
a Database Management System.
-------
of the database once key questions have been resolved and a general decision
among these alternatives is made.
Existing Database Management Systems
Database system design involves the development of a number of components:
• Data formats and file structures for encoding and storing
data in computerized form
• Computer programs for managing data (i.e., sorting, entering,
editing, deleting)
t Computer programs to enable a database user to retrieve
and analyze the data
0 Menus, messages and forms that communicate with a user and
guide him or her through the process of entering, editing,
retrieving, or analyzing data
0 Procedures for the transfer, receipt, entry, and quality control
of data. These procedures should be based on scientific criteria
for the adequacy and acceptability of data, as well as technical
specifications for computerized data transfer.
One advantage of using an existing database system for storing Puget
Sound monitoring data is that many of these components already exist within
some of these systems. The costs of system development would be limited
to modifications of the system to meet specific needs of the Puget Sound
Monitoring Program.
Ocean Data Evaluation System (ODES) —
ODES is a database developed for the U.S. EPA Office of Marine and
Estuarine Protection (OMEP). ODES enables a user to analyze estuarine
50
-------
data stored on the U.S. EPA mainframe computer. The data are stored in
the Statistical Analysis System (SAS, copyright SAS Institute Inc.), a
set of commercially available statistical programs. Menus and messages,
written in a separate component called Wylbur enable a user to select data
and list it, analyze it, or graph it in a variety of ways.
There are two major advantages to the use of ODES. ODES already has
file structures, retrieval, analysis, and graphics tools, and data quality
control procedures for many of the estuarine data types to be collected
as part of the Puget Sound Monitoring Program. ODES also provides easy-
to-use, yet sophisticated, capabilities for data analysis, and will produce
graphs and some types of maps. The 1985 needs assessment recommended the
use of ODES as a central database for Puget Sound data.
The major disadvantage to the use of ODES for Puget Sound Monitoring
data is the lack of local control over the system. Priorities, budgets,
and schedules for ODES are established by U.S. EPA Headquarters. Funding
and management agreements would need to be established between U.S. EPA
Headquarters and Puget Sound entities regarding the use and modification
of ODES before it could be used for storing Puget Sound monitoring data.
Modifications to ODES would be required to add file structures, data
retrieval and analysis tools, and QA/QC procedures for those few data types
not currently available. If data reports and retrievals will be the major
products required of the database, then the ability of ODES to produce
formatted reports and ad-hoc retrievals will need to be enhanced. With
enhancement, ODES would provide excellent statistical analyses, good maps
and graphics, but limited flexibility in retrieving data.
ODES development has been funded to date by OMEP. The costs to modify
ODES to store Puget Sound monitoring data are estimated to be relatively
low ($15,000 - $120,000). It is not clear how these costs will be apportioned
among agencies, U.S. EPA regions, and ODES users.
51
-------
Water Quality Control Information System (STORE!)—
STORE! is a database system developed by the U.S. EPA to store and
manage water quality information. S!ORE! was one of the first environmental
databases; plans for an updated and upgraded S!ORE! are currently being
developed.
S!ORE! lacks file structures for storing many data types to be collected
by the Puget Sound Monitoring Program. !he "ability to produce formatted
reports and ad-hoc retrievals is limited. Some simple maps can be produced.
S!ORE! has no built-in statistical tools, but an experienced user can transfer
data to statistical programs (e.g., SAS) for analysis. Data quality control
procedures are currently insufficient to meet the needs of the Puget Sound
Monitoring Program.
S!ORE! is managed and administered by U.S. EPA Headquarters. Priorities,
schedules, and budgets for the inclusion of data, and the development of
needed data types and retrieval or analysis tools are not directly under
the control of Puget Sound users and may not meet their needs. If S!ORE!
could be used to store Puget Sound monitoring data, it is not clear how
costs of system modifications, expansion, and maintenance would be apportioned
among agencies, U.S. EPA regions, and S!ORE! users.
Database Development !oo1s
Instead of using an existing database system, a new database system
could be developed for storing Puget Sound Monitoring Program data. A
number of commercially available products could be used for developing
such a system. After one of these products was chosen, database development
would require the design and development of all of the components described
above: data formats; data management utilities; retrieval and analysis
programs; a computer-user interface; and data transfer and quality control
procedures.
52
-------
The Statistical Analysis System (SAS) —
SAS could be used for storing, managing, and analyzing Puget Sound
monitoring data. The National Estuaries Program is currently funding the
entry of estuarine data into SAS on an interim basis until a permanent
database is selected. SAS provides: sophisticated data analysis capabilities;
utilities for entering, storing and managing information; and limited data
reporting and mapping capabilities.
SAS itself provides no menus, and requires sufficient programming
expertise to make it accessible to a novice user. Use of SAS without a
menu-driven user interface would represent one of the least expensive options
for management of Puget Sound monitoring program data. However, a user
would have to be familiar with data structures and formats and the SAS
language to retrieve and analyse data.
A menu-driven user interface could be added to the SAS system. This
interface could resemble ODES in menus, messages, and retrieval options
and might provide an "ODES-like" system under local control. The feasibility
of adding this component depends on what programs similar to Wylbur are
available for local computer systems. Adding this component would substantially
increase the costs of using SAS files.
ARC/INFO and Other Fourth Generation Database Products—
So-called fourth generation database systems provide sophisticated
state-of-the-art utilities for data management, programming, and interaction
with users. They usually have an English-like language that allows users
to make ad-hoc, flexible retrieval requests and provide extensive capabilities
for formatting reports. Some of these tools also have built-in graphics,
statistics, and mapping capabilities. None provides the sophisticated
statistical analyses available in SAS. These systems are designed to be
flexible and allow later expansions and modifications to databases.
53
-------
The initial costs of database development using these tools is likely
to be relatively high. However, a database developed using .a fourth-generation
language is likely to have excellent capabilities. The costs of system
maintenance, expansion, and modification for fourth generation database
have generally been found to be substantially lower than those associated
with more conventional systems.
The 1985 data management needs assessment evaluated a database called
GEOMAPS. GEOMAPS is under development by the Washington Department of
Natural Resources to facilitate state land management and planning. The
assessment concluded that GEOMAPS could not store or analyse the types
of data of interest to the Puget Sound Estuary Program. However, the study
did note that a new database could be developed using ARC/INFO, the same
commercially available product that GEOMAPS uses.
ARC/INFO is composed of a database component and a mapping component.
These components are linked and enable a user to create complex maps and
geographic displays of data, and to obtain flexible data retrievals and
reports. ARC/INFO has limited statistical capabilities, but an experienced
programmer can transfer data to a statistical package for analysis.
The use of ARC/INFO would provide the Puget Sound Estuary Program
with a powerful database system with outstanding abilities to display geographic
information. ARC/INFO should be considered for use in developing a database
for Puget Sound monitoring data only if it is decided that production of
maps with state-of-the-art data display and overlays are to be a major
output required of the system. The storage, management, and manipulation
of geographic information in a database system such as ARC/INFO is complex
and expensive. ARC/INFO is likely to be the most expensive of database
options.
Other database management systems could be used for developing a database
for Puget Sound monitoring data. For example, Washington State owns FOCUS
(copyright Information Builders, Inc.), which is one of the most sophisticated
database development products on the market. FOCUS provides utilities
54
-------
for creating data structures, programs and menus, and extensive capabilities
for data reports, graphics, and statistical analysis. FOCUS does not produce
maps.
Data Transfer Formats
Rather than develop a central database, the Puget Sound monitoring
Program could consider developing standard formats for the storage and
transfer of monitoring data among agencies. Individual agencies would
maintain their own computer databases, and, upon request, transfer information
in an accepted format to other users. Data would be analyzed by individual
agencies or transferred to temporary files for overall analysis of trends
and relationships among data types.
Agencies currently share data upon request and to meet legal requirements.
For example, the Washington Department of Ecology transfers ambient water
quality monitoring data and NPDES monitoring data from its computers to
U.S. EPA's STORE! and Permit Compliance Systems. These data transfers
are done in a manner compatible with both agencys' databases; the formats
used for transfer vary.
The National Oceanographic Data Center's (NODC) data reporting formats
are currently used for transfer of data from specific studies funded by
NOAA to NODC archives, and for transfer of Section 301(h) monitoring data
into ODES. NODC publishes detailed specifications of the format and codes
used for each type of data, and ODES staff have expanded the formats and
codes to better meet their needs. Where available for a data type, NODC
format (as modified for ODES) could be used as a standard data transfer
format. However, the NODC format for a data type should not be used:
t If it omits variables or information needed for the Puget
Sound Monitoring Program
• If it would be simpler and less expensive to use other methods
to transfer data between specific databases.
55
-------
There would be costs involved in evaluating specific NODC code formats
and recommending modifications or new formats to meet Puget Sound Monitoring
Program needs. There would also be costs to agencies to develop programs
for the transfer of information from their numerous databases into standard
formats. Some of these costs would be incurred under any of the alternatives
discussed above, since data would probably be transferred in computerized
form from agency databases to a central database if one existed. Even
with these costs, the use of NODC or other formats to share data among
agencies would still be the least expensive data management approach for
Puget Sound. However, this approach might not meet the goals of data mangement
for the Monitoring Program since data would not be as easy to access or
analyze as it would be in a central database.
56
-------
MONITORING PROGRAM REPORTS
The purposes of the annual monitoring program reports will be to summarize
the monitoring studies conducted in Puget Sound and to identify temporal
trends, important spatial variation, and problem areas within the sound.
Information collected for each separate monitoring survey (e.g., river
monitoring) will be reported in individual technical reports produced by
the appropriate agencies. These technical reports will be summarized in
an annual monitoring program summary report prepared by the technical coordi-
nating group for the Puget Sound Monitoring Program (Figure 16).
TECHNICAL REPORTS
The Monitoring Program Technical Reports should include the following
information:
t Executive Summary
• Introduction
Problem
Objectives
• Methods
Sampling Design
Survey Methods
Laboratory Analyses
Quality Assurance/Quality Control Procedures
57
-------
MONITORING
PROGRAMS
MONITORING PROGRAM
TECHNICAL REPORTS
AMBIENT
COMPLIANCE
INTENSIVE/SPECIAL
SEDIMENT,WATER, AND
BIOLOGICAL MONITORING
RIVER MONITORING
HABITAT MONITORING
ETC.
DISCHARGER A
DISCHARGER B
ETC.
SPECIAL STUDY
INTENSIVE STUDY
ETC.
MONITORING
PROGRAM
SUMMARY
REPORT
Figure 16. Summary of Puget Sound Monitoring Program reports.
-------
• Results
Data Reports
Data Analyses
• Discussion (e.g., temporal trends, spatial variations, applicable
criteria, problem areas, potential cause-effect relationships)
t Conclusions
t Recommendations
• References.
MONITORING PROGRAM SUMMARY REPORT
The annual Monitoring Program Summary Report (approximately 100-200
pages) should include the following information:
I. Executive Summary
II. Introduction
A. Goals and objectives of the monitoring program
III. Annual Summary of Monitoring Designs
A. Summary table of the monitoring programs (ambient,
compliance, intensive/special) and their respective
survey methods (e.g., survey dates, number of stations,
number of samples, number of replicates, frequency,
and sampling gear)
B. Summary maps of station locations and variables sampled
58
-------
IV. Abstracts of Puget Sound Monitoring Program Technical Reports
A. Executive summaries from individual monitoring program
technical reports
B. References to selected technical reports
V. Temporal Trends and Spatial Variations in Environmental
Factors within Puget Sound
A. Temporal Trends - Analyses using ambient and compliance
monitoring program data
B. Spatial Variation - Analyses using ambient, compliance,
and intensive/special monitoring program data
C. Identification of problem areas
D. Summary of regulatory actions
VI. Summary of Other Monitoring Program Activities
A. Evaluation of monitoring program design and implementation
B. Recommendations for changes in monitoring programs
C. Publications, symposiums, and workshops.
REPORT CONTENT
The Monitoring Program Technical and Summary Reports should be clearly
and concisely written for technical persons within agencies as well as
informed laypersons. Complex and extensive data sets and tables, if necessary,
should be presented in appendices.
59
-------
Graphic Presentations
The documents should emphasize graphic presentations of the data.
For example, single variables may be plotted on a map of a study area
(Figure 17). Multiple variables may be presented using simple histogram
bar charts (Figure 18), correlation plots (Figure 19) or other graphic
presentations (Figure 20). Temporal trends may be displayed using box
charts (Figure 21), whereas spatial patterns may be plotted using line
graphs (Figure 22), two-dimensional point s-ymbols (Figure 23), and scatter
diagrams (Figure 24). Problem sites within a study area may be portrayed
using shading or hatching on a map (Figure 25).
Data Analyses and Interpretation
Interpretation of the monitoring data will require the use of various
analytical tools. Most analyses of monitoring data will be performed for
one or more of four purposes:
• Identify temporal or spatial trends in variables
• Determine potential cause and effect relationships
• Identify problems
• Recommend corrective actions.
In some cases, analyses of data on single variables will be sufficient to
meet these purposes. However, analyses of multiple variables will be required
in many cases. Analyses of multiple variables often adds greatly to the
understanding of a problem because complex interrelationships exist among
the monitoring components and variables (Table 11).
It is anticipated that identification of temporal and spatial trends
will be accomplished primarily through the use of ANOVA, time series analyses,
and gradient analyses. ANOVA is an especially useful tool for identification
60
-------
•"•••"^^^^ NAUTICAL MILES
KILOMETERS
2 CONTOURS IN FEET
D 0 . 250* 3 51 • 75%
Ei M • 50% I 76 • 100H
-"••. INTERTIDAL AREAS
Figure 17. An example of a single variable mapped within a study
area: Percent fines in sediments of Everett Harbor.
-------
40
6 T
5-
LU
I 3H
N
J3 2-
0.
1 -
TREATMENT PLANT
CSO
STORM DRAIN
Figure 18. An example of a simple bar chart.
-------
li
<8
OCQ.
£l
II
100^
75-
50-
25-
CM
i- GL
650
Q.
(0
1300
1950
in
OL
(0
D
2600
4-METHYLPHENOL ( yg/kg dry wt.)
• = % OYSTER ABNORMALITY
D = % AM PHI POD MORTALITY
Figure 19, An example of a correlation plot.
-------
CONCENTRATION
I I I I J I
CONCENTRATION
I I I I I I
CONCENTRATION
I I I I I I
CONCENTRATION
I I I I I I
I
Q.
UJ
O
cc
UJ
?
E"
Q.
UJ
Q ~
CC
UJ
<"
5
^
i "
Q.
UJ
Q ~
oc
nj
5
5
/
X s
< r
QL
UJ
Q ~
CC
UJ
H* ""
1
S
\
STATION 1
STATION 2
STATION 3
STATION 4
CONCENTRATION
I I I I J
Ul
o _
STATION 5
CONCENTRATION
I I I I I I
Q.
UJ
Q
CC
UJ
UJ
Q
CC
UJ
5
CONCENTRATION
| I I I I I
STATION 6
REFERENCE
MINIMUM CONCENTRATION (e.g. DISSOLVED OXYGEN) SPECIFIED IN PERMIT
Figure 20. An example of concentrations of a variable profiled with
water depth at seven stations.
-------
Duwomish River(Green River at Tukwila)
200n
o
4>
in
UJ
O
tr
<
i
o
V)
oc.
UJ
I50-
100-
50-
D Monthly Range (1968-1978)
0 Monthly Standard Deviation
• Monthly Me
7
/
/
/
/
>
^
7
/
•^
^
'
/
/
-
ar
7
/
/
£
/
/
/
^
1
_
r
/
f
/
/
£
1 i i l
-
7
/
'.
1
'
/
/
*
^
7
/
',
r
/
/
/
^
i i
i-
^
1
g |
" "
2
1
-
/
1
V
r
/
/
•,
^
-
«
>
^
/
>
i^,
^\
^
-
I I
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
Figure 21. An example box chart illustrating temporal trends.
-------
UJ
z>
UJ_
UJ
8
DISTANCE FROM OUTFALL
-O- DEPTH 1 -*- DEPTH 2
I
95 PERCENT CONFIDENCE INTERVAL
Figure 22. An example line graph of mean concentrations or mean value of a variable as a
function of distance from a 301 (h) permittee's outfall.
-------
NO SIGNIFICANT RESPONSE
AMPHIPOD OR OYSTER LARVAE
SIGNIFICANT RESPONSE
AMPHIPOD AND OYSTER LARVAE
SIGNIFICANT RESPONSE
COMMENCEMENT
BAY
air
WATERWAY
Figure 23. An example of spatial patterns plotted using two-
dimensional point symbols.
-------
Segment: 3 12 1
£ 800 -
o>
*0>
* 700 -
TJ
E 600 -
o.
o.
.£ 500 -
in
V
3
o 400 -
1
-6 300 -
t>
cn
•£ 200 -
o
c
0 100 -
0 -I
O
o
D D
O
O °
Q
^
rl
0
D
°
O
o °
o
o n
D
D DO
a a
O (3
V
0 °
1 1 1 1 1 I 1
Toxlclty
AET
Benthic
Effects
AET
02468
(Thousands)
Ft. from mouth of waterway
SURFICIAL SEDIMENT CONCENTRATIONS OF
L£An i^ rjTy uAjrpuAY
! G Tetra lech Investigation - Quantiuted value
0 Other Investigations - Quant Ha ted value
Figure 24. An example of a scatter plot.
-------
COMMENCEMENT
BAY
HIGHEST PRIORITY PROBLEM AREAS
SECOND PRIORITY PROBLEM AREAS
POTENTIAL PROBLEM AREAS
(NO CONFIRMING BIOLOGICAL
DATA AVAILABLE)
POTENTIAL PROBLEM AREA BY
HISTORICAL DATA ONLY
CHEMICALS EXCEED APPARENT
EFFECTS THRESHOLD
CHEMICALS BELOW APPARENT
EFFECTS THRESHOLD
CITY
WATERWAY
Figure 25. An example of using shading density on a map to indicate
concentration, value, or priority.
-------
TABLE 11: MATRIX OF RELATIONSHIPS BETWEEN PAIRS OF MONITORING COMPONENTS'
1
2
3
4
b
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
2tt
29
10
11
12
33
Components
SEDIMENT QUALITY
Sediment chemistry
Sediment toxicity bioassays
Conventional sediment variables
WATER QUALITY
Hydrographic conditions
Dissolved oxygen
Turbidity/ transparency
Odor, floatables, slicks, water color
Nutrient concentrations
Phytoplankton standing stock
Pathogen indicators in water
BIOLOGICAL CONDITIONS
Benthic macroinvertebrate abundances
Toxic chemicals in fish tissue
Histopathological abnormalities in fish
fish species abundances
Shellfish abundances
Toxic chemicals in shellfish
PSP in shellfish
Pathogen indicators in shellfish
Marine mammal abund. and reprod. success
Avian abundances and reproductive success
RIVER MONITORING
HABITAT TYPES
ANCILLARY DATA
Cl imate/weather
Fisheries harvest
Waterfowl harvest
Aquaculture sites and yields
Demographic and socioeconomic conditions
Decision record -keep ing
COMPLIANCE MONITORING
33
•
•
A
n
•
A
A
A
A
•
A
•
•
A
•
•
A
A
IJ
1 J
A
A
A
n
•
32
n
n
D
u
n
n
n
n
n
G
•
•
•
n
31
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
n
A
A
A
30
[ 1
1 1
n
n
n
n
n
u
a
n
n
n
n
u
n
29
D
•
n
LI
n
28
L)
n
n
n
n
D
n
•
•
ii
n
u
n
27
A
•
A
A
A
A
A
A
A
A
A
26
2b
•
•
A
•
•
A
•
•
*
•
•
I.)
24
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
23
n
•
•
22
•
*
•
•
21
n
n
A
20
n
19
n
r i
A
18
n
n
n
D
n
n
n
17
n
•
1 1
n
n
•
n
1 1
16
n
r i
n
ib
1 1
n
14
1 1
N
n
n
1 1
n
•
u
12
n
1 1
n
A
A
1 1
•
10
I 1
1 1
n
9
1 1
A
•
8
A
1 1
/
1 1
1 1
1 1
6
A
A
b
4
A
A
1
1 1
2
I
a For a given interaction between components: a "[]" indicates that a column component is dependent
upon a row component; a "A" indicates that a row component is dependent upon a column component;
and an "•" indicates mutual dependency.
-------
of temporal and spatial changes in environmental variables (Millard et
al. 1985; Millard and Lettenmaier 1986). Time series analyses may be used
to identify temporal trends if sufficient data are available. Gradient
analyses may be of several types, including ANOVA, graphic analyses, regression
analysis, and several multivariate techniques (see Green 1979; Gauch 1982).
Regression and correlation analyses will be the primary tools for
determining potential cause and effect relationships (e.g., Table 11).
Other analytical tools may also be used, however, including graphic analyses
and multivariate techniques (see Green 1979; Gauch 1982). Analyses of
monitoring data cannot prove cause-effect linkages, but they can help to
establish priorities for further research.
Problem identification will require careful selection of those variables
that may be relevant to the suspected problem. For example, analyses of
sediment chemistry; sediment toxicity bioassays, benthic macroinvertebrate
abundances, toxic chemicals in fish tissue, and fish histopathology may
be used to identify areas of contaminated sediment (Chapman and Long 1983;
Tetra Tech 1985a). Selection of the analytical tools that will be used
to identify the problem of interest is dependent on the type of problem
and the variables used in the analyses.
61
-------
INSTITUTIONAL MECHANISMS FOR PROGRAM MANAGEMENT
AND IMPLEMENTATION
Operation of the Puget Sound Monitoring Program will require two kinds
of coordination among the participating agencies. The first, program manage-
ment, involves coordination of policy-making and decision-making. The
second, program implementation, involves coordination of monitoring activities.
This section begins with a discussion of requirements for effective coordi-
nation. It then presents three management alternatives, three implementation
alternatives, and a preferred approach for program management and implemen-
tation. The final management structure and detailed mechanisms for implemen-
tation of the program will be developed by PSWQA's management committee.
CONSIDERATIONS FOR ENHANCING COORDINATION
A long-lasting program of coordination must be able to:
0 Meet the needs of all member agencies
• Change over time in response to changes in the natural and
socio-political environments
• Promote efficient decision-making
t Assign accountability for the program's accomplishments
and shortcomings. -
In order to achieve successful coordination, relationships among participating
organizations should be considered at the start of the planning process.
These relationships will also affect the design of the institutional mechanisms
for management, and include the following features (Wolfe 1982):
62
-------
t Information Exchange - Early planning should focus on improving
information exchange
t Interorganizational Linkages - Each organization in a program
must be able to interact with other organizations through
legal arrangements (e.g., memoranda of understanding or
interagency agreements) or administrative arrangements (e.g.,
personnel or funding transfers)
• Mutual Advantage - Because benefits of coordination are
seldom evenly distributed among participating organizations,
parties that accrue few benefits must be compensated in
some other way
• Organizational Power Bases - When power disparities hinder
coordination, intervening mechanisms (e.g., decision-making
process or interagency agreement) may be needed to restruc-
ture the distribution of power
t Good Will and Mutual Respect - Each organization must have
a legitimate respect for the interests, mandates, and goals
of the other organizations. This can be fostered by establishing
commonality of purpose early in the planning process.
GENERAL INSTITUTIONAL STRUCTURE
Figure 26 represents relationships among components of the Puget Sound
Monitoring Program. Primary components are connected by a solid line.
Technical support and complementary monitoring activities are connected
by a dotted line. Program implementation is divided into two functions.
One function consists of implementing monitoring design components. The
other consists of technical coordination, data management, and summary-
report preparation. These latter activites are assumed to be an integral
component of the program. It is recommended that the core group performing
this work be composed of at least:
63
-------
PUGET SOUND MONITORING
PROGRAM MANAGEMENT
TECHNICAL
ADVISORY COMMITTEE
COMPLEMENTARY
ONGOING MONITORING
PROGRAMS
TECHNICAL COORDINATION,
DATA MANAGEMENT. AND
SUMMARY-REPORT PREPARATION
IMPLEMENTATION OF
MONITORING DESIGN
COMPONENTS
MANAGEMENT OPTIONS
• Puget Sound Monitoring Committee
• Puget Sound Monitoring Agency
• Puget Sound Monitoring Directorate
IMPLEMENTATION OPTIONS
• Expansion of existing programs
• Single organization for monitoring
• Existing programs plus new monitoring entity
Figure 26. General structure for managing and implementing the Puget Sound Monitoring
Program.
-------
• An environmental chemist
t A marine biologist
• A data manager.
These first two individuals should have we!1-developed skills in design
and conduct of field investigations, quality assurance/quality control,
statistics, and program management. All three individuals should be supplied
with adequate administrative support to carry out their jobs efficiently.
In the next two sections, three institutional alternative for coordinating
program management activities and three alternative for program implementation
are presented. The advantages and disadvantages of each alternative are
listed in general terms. A section on the preferred approach is then presented.
APPROACHES TO PROGRAM MANAGEMENT
The institutional mechanisms for managing the monitoring program will
depend largely on how policy-making and decision-making are coordinated.
The mechanisms or interagency arrangements available for program coordination
range from ad-hoc, informal agreements to the creation of an autonomous
program or organization (e.g., a Super-Agency). Morris and Carney (1984)
describe four examples within this spectrum, as discussed below.
Information sharing is perhaps the simplest tool for coordinating
policy-making and decision-making. Information-sharing is usually a relatively
informal process whereby agencies learn about each other's programmatic
priorities, needs, methods, and results via the exchange of information
by reports, newsletters, meetings, or similar means. A more formalized
form of information-sharing is consultation. An example of consultation
is the interagency review process for environmental impact statements,
64
-------
which requires agencies to disseminate information and incorporate feedback
from other participants. Policy coordination can also be achieved through
shared decision-making whereby participating agencies express their positions
through voting arrangements. In the central control approach, decision-
making authority resides within one organization.
Sources of funding also affect institutional structures for a program
such as this. Because information on funding sources was incomplete at
the time of this writing, it was assumed that:
t Funding for monitoring program components will be allocated
based on the component's technical merits irrespective of
current monitoring activity levels
• Money will be available directly at the state level, regardless
of its ultimate source (e.g., federal grants, state legislative
appropriations, agency budget increases)
• Federal agencies will have little or no policy-making and
decision-making role in the formation and management of
the program.
The three institutional alternatives for program management are a
committee, a monitoring directorate, and an agency. These options were
selected because they represent three positions on the continuum for policy-
and decision-making coordination. These are (respectively): information
sharing, shared decision-making, and central control. The characteristics
of each alternative were selected to illustrate a given degree of organizational
complexity, and should not be interpreted as being the only possible alterna-
tives. The optimum institutional arrangement may well combine features
of all three alternatives presented here.
The description of each alternative below is followed by a list of
major strengths and weaknesses relative to the monitoring program. This
65
-------
discussion focuses on management options only, and thus does not differentiate
between ambient monitoring, compliance monitoring, and intensive surveys.
Management Alternative 1: A Puget Sound Monitoring Committee
Membership of the Puget Sound Monitoring Committee would consist of
mid-level program managers from participating organizations. Members of
the committee would not necessarily be bound by formal agreements, and
would not necessarily make policy-level decisions for the groups they
represent. A chairperson would most appropriately come from the Puget
Sound Water Quality Authority (PSWQA) or the Washington Department of Ecology.
The chairing agency would provide administrative support. Each member
agency would contribute staff time toward technical meetings and planning
activities. After program startup, meetings would consist of an annual
workshop (approximately 3 days) for program review and update. Decisions
would be made by consensus. Technical input for program design would come
from individual technical personnel within agencies or from a technical
advisory group. Membership of the technical advisory group, if one existed,
would include experts from non-agency organizations (e.g., universities
and consultants).
In this management alternative, no one participant has executive decision-
making or policy-making authority. Influence over policy and technical
issues would depend on each member agency's particular power base. However,
the chairing agency would have the advantage of presiding over planning
sessions and workshops.
Major advantages and disadvantages of Management Alternative 1 are
listed below:
Advantages Disadvantages
Participating agencies lose little Accountability is not clear because
or no autonomy over their existing of informal representation and informal
programs decision-making/policy-making structure
66
-------
Important differences may remain
unsettled due to the informal decision-
making process and lack of clear
hierarchy
Management Alternative 2: A Puget Sound Monitoring Directorate
A Puget Sound Monitoring Directorate would have a structure similar
to that of a committee but with formal procedures for enhancing interagency
coordination and decision-making. Directorate membership would include
upper-level program managers with decision-making and policy-making authority
from their respective agencies. The chairperson of the directorate would
be selected from the Washington Department of Ecology or the PSWQA. That
person and agency would have ultimate responsibility for the monitoring
program and for directorate activities. Other directorate members would
be able to commit their respective agencies to specific monitoring activities.
The directorate would operate on the principles of a charter, including:
• Mechanisms for providing incentives for participation
t Protocols for fostering interagency cooperation
t Procedures for arriving at decisions concerning policy issues
and technical matters
0 Procedures for mediating disputes
0 Accountability for certain aspects of program management
and monitoring activities
0 Procedures for reporting program results and assessing program
success
0 Procedures and a schedule for program review and update.
67
-------
The directorate would have its own technical staff and support services,
and would operate as an office of the chairing agency. After startup,
the directorate would meet at least quarterly. Technical input for program
design and review would come from a technical advisory committee consisting
of staff from member agencies, experts from nonmember organizations, or
both.
Major advantages and disadvantages of Management Alternative 2 are
listed below:
Advantages Disadvantages
Agencies lose little autonomy Relatively high costs are associated
with member participation and staffing
Accountability is clearly assigned due to the high degree of involvement
at all levels of program management in planning and decision-making
Management Alternative 3: A Puget Sound Monitoring Agency
A Puget Sound Monitoring Agency would be responsible for all technical
components of the monitoring program, plus program planning, implementation,
and review. The agency would assume all ambient monitoring responsibilities,
with the exception of federal programs and some local programs. The Puget
Sound Monitoring Agency would have its own technical and support staff
and could be supplemented by a technical advisory committee. The agency
would retain sole decision-making authority over all aspects of the monitoring
program. In this alternative, procedures for policy-making and decision-
making would be clearly defined. The Puget Sound Monitoring Agency would
either take the form of a new entity created by state legislation, or be
established as a discrete office within the Washington Department of Ecology
or PSWQA.
Major advantages and disadvantages of Management Alternative 3 are
listed below:
68
-------
Advantages Disadvantages
Accountability for overall program Most state agencies would lose autonomy
performance is clearly assigned over their existing monitoring programs
(this would pose formidable implemen-
Coordination among monitoring tation barriers)
activities is facilitated
Limited information exchange may
result in unfavorable changes in
original objectives
APPROACHES TO PROGRAM IMPLEMENTATION
Approaches to implementation of the proposed ambient, compliance,
and intensive monitoring programs are discussed in separate sections below.
Alternatives for implementing the ambient monitoring programs are presented,
and major advantages and disadvantages are summarized. Because existing
institutional mechanisms for compliance monitoring and intensive surveys
are well-established, they will probably remain intact in the future.
Therefore, alternative options are not proposed. Federal monitoring activities
are presumed to remain virtually unaffected by the program.
A management body is presumed to exist, but its structure (e.g., committee,
directorate, agency) is not important to the following discussion. This
body will be referred to as the Puget Sound Monitoring Office (PSMO).
Ambient Monitoring
Ambient monitoring components can be grouped into two categories:
• Components that overlap with an ongoing program, but represent
an increase in sampling or analysis effort (e.g., water
quality monitoring)
• Components that do not overlap with a major ongoing monitoring
effort (e.g., monitoring of benthic macroinvertebrates).
69
-------
Implementation Alternative 1: Existing Programs Expand-
In this implementation alternative, participating agencies would:
• Expand ongoing ambient monitoring programs to accommodate
the increased sampling and analysis intensity proposed by
the Puget Sound Monitoring Program
• Modify their monitoring programs (or create new ones) to accom-
modate new components proposed by the Puget Sound Monitoring
Program.
One possible arrangement for the allocation of major monitoring activities
among member agencies is illustrated in Figure 27. The allocation of components
was determined based on the nature of ongoing programs, probable responsi-
bilities under the Puget Sound Monitoring Program, agency missions, and
logistics of the program. This information was derived from existing reports
(e.g., Chapman et al. 1985) and contacts with agency personnel.
In this alternative, the Washington Department of Ecology would assume
responsibility for sediment quality components primarily because this activity
is consistent with its mission. As illustrated in Figure 27, the department
would also conduct the biological condition components (primarily because
these components are usually sampled concurrently with sediments). Roles
of other agencies were determined on a similar basis. Part of PSMO's
theoretical responsibilities would be to compile selected ancillary data
(already available from ongoing programs) needed to interpret other monitoring
data. It was assumed that volunteers would implement monitoring components
that could be performed accurately as a recreational activity or in conjunc-
tion with recreation.
The primary advantage of this alternative is that it fully capitalizes
on ongoing monitoring efforts, experience, and organizational support.
The major disadvantage is that is may require new programs. Planning,
budget allocation, intra- and interagency competition for funds, hiring,
70
-------
I
8
W 0) W X
i b zi uj
SEDIMENT QUALITY
SEDIMENT CHEMISTRY
SEDIMENT TOXICITY BIOASSAYS
CONVENTIONAL SEDIMENT VARIABLES
WATER QUALITY
HYDROGRAPHIC CONDITIONS
DISSOLVED OXYGEN
TURBIDITY/TRANSPARENCY
ODOR, FLOATABLES, SLICKS, WATER COLOR
NUTRIENT CONCENTRATIONS
PHYTOPLANKTON STANDING STOCK
PATHOGEN INDICATORS IN WATER
BIOLOGICAL CONDmONS
BENTHIC MACROINVERTEBRATE ABUNDANCES
TOXIC CHEMJCALS IN FISH TISSUE
HISTOPATHOLOGICAL ABNORMALfTIES IN FISH
RSH SPECIES ABUNDANCES - DEMERSAL
-OTHER
SHELLFISH ABUNDANCES
TOXIC CHEMICALS IN SHELLFISH
PSP IN SHELLFISH
PATHOGEN INDICATORS IN SHELLFISH
MARINE MAMMAL ABUNDANCES AND REPRODUCTIVE
SUCCESS
AVIAN ABUNDANCES AND REPRODUCTIVE SUCCESS
RIVER MONITORING
HABITAT TYPES
ANCILLARY DATA
CLIMATE/WEATHER
RSHERIES HARVEST
WATERFOWL HARVEST
AOUACULTURE SITES AND YIELDS
DEMOGRAPHIC AND SOCIOECONOMIC CONDITIONS
DECISION RECORD-KEEPING
O-O
-£*-
^>
-€»
-€J-
•<>
-(>
-()-
-€)-
-£)-
•O-
^^
-€>
£
a
-<)
-()
-()
• LEAD ROLE IN PUGET SOUND MONfTORING PROGRAM
O ONGOING MONITORING COMPLEMENTARY TO THE P.S. MONITORING PROGRAM
Figure 27. Agency ambient monitoring responsibilities under
implementation alternative 1.
-------
and procurement could hinder the efficient startup and operation of new
programs. The degree to which these problems manifest themselves in this
and other alternatives will depend on the government organization (e.g.,
size, mission, source of funding), and the number, magnitude, and complexity
of new programs.
Implementation Alternative 2: A Single Organization Performs all Monitoring—
In this alternative, a single organization would absorb all ongoing
monitoring activities, plus initiate programs for all new monitoring activi-
ties. Because of its extensive ongoing monitoring program, it is assumed
that the Washington Department of Ecology would absorb all pertinent ambient
monitoring and intensive survey activities of WDF, WDG, WDNR, DSHS, and
Metro. Selected monitoring activities would be delegated to volunteer
groups.
The primary advantage of this alternative is that all coordination
would occur within a single agency. The major disadvantages are:
• Possible inability of member agencies to terminate ongoing
programs (due to legal barriers), resulting in duplication
of effort
• Strong reluctance of agencies to sacrifice their autonomy
over monitoring programs
t The possible need to expand the Washington Department of
Ecology's legislative mandates for monitoring
• High costs (in time and money) of transferring programs
of several organizations to one agency.
These disadvantages preclude this scenario from further consideration.
71
-------
Implementation Alternative 3: Existing Programs Expand as Necessary with
All Remaining Monitoring Components Implemented by a Single Organization—
In this alternative, ongoing monitoring programs (e.g., water quality
monitoring) would expand to accommodate the increase in activity associated
with the core Puget Sound Monitoring Program (Figure 28). New components
(e.g., monitoring of benthic macroinvertebrates) would be performed by
a single organization. The PSMO would play the lead role in implementing
new components. Allocation of monitoring components in this alternative
was based on existing monitoring responsibilities and on the Puget Sound
Monitoring Program design for ambient monitoring (Figure 28).
The primary advantage of this alternative is that ongoing monitoring
efforts of member agencies absorb portions of the core monitoring program
without having to initiate new programs quickly. The primary disadvantage
is that the lead agency is burdened with implementing new monitoring components.
Compliance Monitoring
The Washington Department of Ecology currently manages a compliance
monitoring program under the National Pollutant Discharge Elimination System
(NPDES) and the State Waste Discharge permit system. The program consists
of self-monitoring by permittees and permit-related surveys by the Washington
Department of Ecology (or U.S. EPA for federal facilities). Commercial,
industrial, and municipal facilities are monitored as part of the compliance
program.
The institutional structures for compliance monitoring will likely
remain unaffected by the Puget Sound Monitoring Program. However, several
areas of expansion are anticipated:
• Increased activity by permittees for effluent biomonitoring
and receiving system monitoring (U.S. EPA 1985b)
72
-------
w *
828
SEDIMENT QUALITY
SEDIMENT CHEMISTRY
SEDIMENT TOXICITY BIOASSAYS
CONVENTKX4AL SEDIMENT VARIABLES
WATER QUALITY
HYDROGRAPHIC CONDITIONS
DISSOLVED OXYGEN
TURBIDITY/TRANSPARENCY
ODOR. FLOATABLES, SUCKS. WATER COLOR
NUTRIENT CONCENTRATIONS
PHYTOPLANKTON STANDING STOCK
PATHOGEN INDICATORS IN WATER
BIOLOGICAL CONDITIONS
BENTHIC MACROINVERTEBRATE ABUNDANCES
TOXIC CHEMICALS IN RSH TISSUE
mSTOPATHOLOGICAL ABNORMALITIES IN FISH
RSH SPECIES ABUNDANCES - DEMERSAL
-OTHER
SHELLFISH ABUNDANCES
TOXIC CHEMICALS IN SHELLRSH
PSP IN SHELLFISH
PATHOGEN INDICATORS IN SHELLFISH
MARINE MAMMAL ABUNDANCES AND REPRODUCTIVE
SUCCESS
AVIAN ABUNDANCES AND REPRODUCTIVE SUCCESS
RIVER MONITORING
HABITAT TYPES
ANCILLARY DATA
CLIMATE/WEATHER
RSHERIES HARVEST
WATERFOWL HARVEST
AOUACULTURE SITES AND YIELDS
DEMOGRAPHIC AND SOCIOECONOMIC CONDITIONS
DECISION RECORD-KEEPING
•
d
^
f\
(
\
r
f
(.
(
\
(]
! ?
Bf
\
\
t
f
•} f
N
;
•\
1
<
) 0
> :
\
>
"•>
s
,/
/•
k
(•
*,
(•
4 «
3 :
N
)
•)
f
\
(
1 I
j ,
j :
. f
k.
V
r
\
,
r
\
^
t
f
1 i
"\
J
|
d
^
i \
s
V.
(
\
i 1
c
S
J
i !
1
»(
>(
1 s
d
/*
i3
i3
*,
•s i-
^ I.
•s
J
"\
S
(
\
i i
•)
N
•\
j
"\
^
|
p)
|
— <
! 1
|
(
V.
^
»,
i_
• LEAD ROLE IN PUGET SOUND MONITORING PROGRAM
O ONGOING MONITORING COMPLEMENTARY TO THE P.S. MONITORING PROGRAM
Figure 28. Agency ambient monitoring responsibilities under
implementation alternative 3.
-------
t Development of monitoring programs and permits for stormwater
discharges (U.S. EPA 1985a)
• Increases in quality assurance/quality control (QA/QC) and
review of self-monitoring data by the Washington Department
of Ecology.
Although compliance monitoring is driven by legal mandates, the Puget
Sound Monitoring Program may influence the design or implementation of
compliance monitoring activities via technical input and review. For example,
technical input from the program may influence sampling and analysis protocols
or reporting practices.
Stormwater discharges and combined sewer overflows associated with
heavily urbanized or industrialized areas are recognized as important sources
of contaminants to Puget Sound (cf. Tetra Tech 1985a; PSWQA 1986a,b).
One option for monitoring these discharges is to have the responsible party
conduct the monitoring. This option is consistent with the present self-
monitoring structure of the compliance programs. Counties and municipalities
would play a key role in monitoring stormwater discharges and combined
sewer overflows in urban areas. Stormwater releases through private drains
associated with commercial and industrial facilities would be monitored
by property owners.
To be useful to the Puget Sound Monitoring Program, all monitoring
data must be evaluated in accordance with standard QA/QC procedures. In
light of the expansion described above, the Washington Department of Ecology
may implement a QA/QC program, a standard protocols agenda, and an approval
program for contract laboratories.
Intensive Surveys
Intensive surveys are performed by most participating agencies in
response to regulatory activities or other mission-oriented programs (e.g.,
fish and wildlife stock assessment, environmental problem identification,
73
-------
remedial action assessment). Examples of intensive survey programs are
given below:
t Study of bacterial contamination in Burley Lagoon/Minter
Bay, and Southern Puget Sound Water Quality Assessment by
Washington Department of Ecology (see PSWQA 1986a, Appendix)
• Intensive surveys by U.S. EPA as part of the Puget Sound
Estuary Program (e.g., urban bay toxics, shellfish contamination)
• Intensive surveys of fish histopathology by the National
Marine Fisheries Service (NOAA)
• Investigation of acute toxicity incidents (e.g., fish kills)
by Metro, Washington Department of Ecology, WDF, and WDG.
These and other intensive surveys will continue to be implemented by individual
agencies. Protocols for intensive surveys should be standardized to ensure
that data are comparable to those collected by the ambient and compliance
monitoring programs. Also, compatible data from intensive surveys must
receive adequate QA/QC verification to be incorporated into the Puget Sound
Monitoring Program.
PREFERRED APPROACH
Design of an institutional structure will be one of the responsibilities
of the interagency management group being organized by PSWQA. This section
describes general characteristics of a preferred approach for program management
and program implementation.
Program Management
To be successful, the monitoring program should facilitate efficient
and comprehensive exchange of technical information while assigning clear
74
-------
accountability for program implementation. These functions have implicit
and divergent institutional requirements for coordination.
No single alternative above can adequately fulfill these functions.
For example, to maximize information exchange, the program would best be
served by a committee-like group with members representing a broad range
of interests and disciplines. The group would adopt a consensus approach
to decision-making (e.g., Management Alternative 1). In contrast, efficient
policy-making and the assignment of accountability require more formal rela-
tionships among participating organizations (e.g., Management Alternative 2).
Because the monitoring program has these competing requirements, the approach
proposed below consists of institutional mechanisms for separating technical
input and decision-making from policy level decision-making. Such a fuctional
separation can in itself comprise an institutional mechanism for coordination
by establishing association among participants at more than one level of
activity (e.g., Evans et al. 1980).
This management approach could take many different forms. One possible
arrangement might be:
• A technical advisory group (as per above) chaired by the
PSWQA, which would be responsible for offering incentives
for participation, mediating disputes, producing reports,
and providing other administrative assistance
• A management group (as per above) with a flat hierarchical
structure, rotating meeting places, and shared administrative
responsibilities
• A director or directorate (e.g., chair of PSWQA and director
of Washington Department of Ecology) accountable for the
entire program, settling disputes, and ensuring consideration
of input from the technical advisory group.
75
-------
This arrangement, or one similar to it, would assign clear accountability,
allow monitoring agencies to retain autonomy over programs, and foster
essential information exchange. It would, however, be moderately expensive,
and administratively complex to establish.
The rationale for preferring an institutional structure that separates
policy level decision-making from technical decision-making is further
supported by a brief discussion of the role of information exchange in
the monitoring program.
Information exchange would occur at basically two levels in the monitoring
program. At the technical level, information would be used to design the
individual components of the program and to evaluate the program's results.
Information exchange at this level would best be served by a broad-based
group whose members provide feedback based primarily on their professional
backgrounds (as opposed to professional affiliations). Such a group should
have the following features:
t Membership composed of technical staff from agencies, educational
institutions, environmental groups, and the private sector
(e.g., industry, consulting firms)
• Participation encouraged by nonbinding agreements, compensation
for travel and meals (and possibly time spent at meetings),
and some mechanism to ensure that input from the group is
given consideration at the program level.
At a management level, information exchange would support evaluation
of program accomplishments, further development of interagency relations,
and program planning. This level of information exchange would best be
served by a group composed of accountable member agency representatives
bound by formal mechanisms for achieving coordination.
During the formation of a group for interagency coordination, several
process-related and administrative factors should be considered. Evans
76
-------
et al . (1980) and others contend the key to successful coordination is
gradual implementation of a program. Mechanisms should be developed to
keep participants from becoming "caught up" in the process and forced to
move in undesired directions. Such mechanisms can include periodic project
status checks to ensure that all participants are comfortable with the
progress being made. Along with pace, the sequence of program development
is also important. For example, Etzioni (1965) found that development
of cooperative relationships in the European Economic Community was fostered
when progress was made on an i ssue-by-i ss-ue basis. In a setting where
decision-making is by concensus (e.g., the technical advisory group of
the preferred approach), it is important for participants to have the ability
or willingness to negotiate, compromise, and exclude non-negotiable issues.
Administrative factors to consider in working toward interagency
cooperation include administrative support, and continuity and follow-through
(Wolfe 1982). Administrative support (including input of technical expertise)
can supply the coordination effort with a variety of otherwise time-consuming
services. Administrative support is most well-received when staff are
considered neutral and capable (Wolfe 1982). A successful coordination
effort also relies on continuity and follow-through, both for individual
decisions in the planning process and for long-term program continuity.
In the short term, continuity can be enhanced by supporting decision-making
with adequate research and by providing a feedback mechanism to ensure
that decisions have the desired outcomes. At the program level, continuity
can be enhanced by a periodic (e.g., yearly) evaluation of program performance
at both the technical level and policy coordination level. The evaluation
should be performed or audited independently (e.g., technical advisory
group, legislative committee, general public) and there should be a feedback
mechanism to ensure that recommendations resulting from the audit are taken
into account during the subsequent operational period.
Program Implementation
As mentioned above, and referred to in Figure 26, it is assumed that
a core technical group will be formed for coodinating the technical aspects
77
-------
of the program implementation, data management, and report preparation.
This group and its support staff will be responsible for the day-to-day
operation of the program. During implementation of the Puget Sound Monitoring
Program, ongoing monitoring efforts will be altered as needed and new monitoring
efforts will be initiated. The distribution of responsibilities among
agencies will depend on the sources and amount of funding for the Puget
Sound Monitoring Program (currently uncertain) and on a negotiation process
among participating organizations. The process by which this balance is
arrived at will be designed largely by the interagency management group
formed by the PSWQA. This section takes a step toward defining this balance
by describing the relationships between ongoing monitoring programs and
the core Puget Sound Program.
The following section describes the proposed relationship between
current monitoring and that associated with the new Puget Sound Program.
Although the discussion of institutional options has focused on ambient
monitoring, intensive surveys and compliance monitoring will also play
an important role in contributing to a comprehensive monitoring database.
Ambient monitoring is discussed first, followed by intensive surveys and
compliance monitoring.
Ambient Monitoring—
The major ongoing programs below are classified as either integral
or complementary to the Puget Sound Monitoring Program.
NOAA (Ocean Assessments Division. National Weather Service)—NOAA
monitoring programs would become integral components of the Puget Sound
Monitoring Program:
• National Status and Trends Program (Ocean Assessments Division) -
sediment chemistry, conventional sediment variables, fish
histopathology, and toxic chemicals in fish and mussels
78
-------
• National Weather Service - ancillary climate/weather data
for compilation by the Puget Sound Monitoring Program
• National Marine Fisheries Service (NMFS) - fish landing
data, marine mammal stock size and abundance estimates.
U.S. COE—U.S. COE monitoring programs would complement the Puget
Sound Monitoring Program. PSDDA is the only planned ambient monitoring
program managed by U.S. COE. PSDDA monitoring will produce data on sediment
chemistry, benthic macroinvertebrate abundances, and toxic chemicals in
fish and benthic macroinvertebrates.
U.S. FWS—Routine surveys of marine mammal and bird abundances and
reproductive success would become an integral part of the program. U.S. FWS,
Corvallis, Oregon monitors toxic pollutants in birds that winter in the
Puget Sound Basin. These data would complement the program.
U.S. GS—River monitoring data produced by U.S. GS would be an integral
part of the program.
Washington Department of Ecology—The department's present water quality
monitoring program would be modified and expanded to become an integral
element of the Puget Sound Monitoring Program. It would be most appropriate
for the department to assume responsibility for all water quality components
of the core program except possibly pathogen indicators (see DSHS below).
Ideally, Washington Department of Ecology would expand its existing water
quality monitoring program to match the sampling intensity prescribed by
the Puget Sound Program. The department's river monitoring would also
expand to become an integral component of the Puget Sound Program.
WDF—WDF monitoring of fisheries harvests and aquaculture sites and
yields would become an integral part of the Puget Sound Monitoring Program.
WDF monitoring activities related to fish and shellfish abundances would
complement the Puget Sound Monitoring Program.
79
-------
WPG—The following WDG monitoring activities would become integral
components of the Puget Sound program:
• Marine mammal abundances and reproductive success
• Avian abundances and reproductive success
• Waterfowl harvest surveys.
WDNR—Monitoring of eelgrass beds throughout Puget Sound is planned
by WDNR. Such a program would become an integral part of the Puget Sound
Monitoring Program.
DSHS—The following survey activities of DSHS would become integral
components of the core program:
• Pathogen indicators in water
t PSP in shellfish
• Pathogen indicators in shellfish.
Surveys of hydrographic conditions by DSHS would complement the core program.
Local Government—Although local government performs some ambient
monitoring (and some intensive surveys) these are generally small in scale,
and with few exceptions would not contribute significantly to the monitoring
database. Local government records for demographic and socioeconomic conditions
would become an integral part of the core program via transmittal of records
or summary reports to the managing body.
Metro (Seattle)—Metro currently conducts substantial monitoring of
water quality and pathogen indicators in water at several stations in Puget
Sound, and river monitoring in the Lake Washington and Green River basins.
River monitoring data collected by Metro would become an integral part
80
-------
of the new program. Metro's marine water quality effort may be modified
to complement the Puget Sound Monitoring Program.
Volunteers—Volunteers would be used for core monitoring efforts where
substantial cost savings could be realized without sacrificing data quality.
The best candidate components are the following water quality variables:
• Beach characteristics
• Turbidity/transparency
0 Odor, floatables, slicks, and water color
• Marine mammal sightings hotline
• Avian species abundances (e.g., Audubon Christmas Bird Count).
Additional volunteer efforts can be tapped via a telephone hotline. Information
gained through the hotline may include reports of spills or dumping, animal
abundances or unusual behavior, and a variety of qualitative information.
Compliance Monitoring—
Self monitoring by dischargers is the most relevant ongoing compliance
monitoring program. Self-monitoring is either performed in-house by the
permittee or under contract to a third party. Because of the program-specific
needs for discharge monitoring, it is likely that the Washington Department
of Ecology (and U.S. EPA) will remain solely responsible for managing the
program, and that implementation will remain the responsibility of permittees.
Management and implementation schemes of other compliance monitoring
programs would likewise remain unaffected. Other compliance monitoring
programs or surveys may include:
81
-------
• Dredging and dredged material disposal - monitoring required
by federal, state, and local governments (through U.S. COE)
at both the dredging sites and the disposal sites
t Aquaculture - monitoring required by a state agency for
impacts to the receiving environment of an aquaculture activity
(e.g., pen rearing of salmon)
• Environmental Impact Statement (EIS) -monitoring required
by a federal or state agency for any activity requiring
an EIS
t Local government permit activities - monitoring required
by local government for activities not requiring an EIS
(e.g., construction or grading permits).
Because of the multiple parties responsible for implementing these monitoring
efforts, it is unlikely that a common database could be developed. Never-
theless, information from these activities could be used on an ad-hoc basis
to supplement data from the Puget Sound Monitoring Program. Future efforts
toward standardization of protocols and coordination of EIS investigations
could substantially improve the database for assessment of cumulative impacts.
Long-term monitoring programs to be implemented by permittees should be
incorporated into permits for major projects.
Intensive Surveys—
The following discussion focuses only on existing or expected intensive
surveys that could generate data complementary to the monitoring program.
U.S. EPA—U.S. EPA is currently sponsoring intensive surveys in urban
embayments as part of PSEP. These surveys generate substantial volumes
of data for sediment quality variables and some biological variables.
U.S. EPA has also conducted surveys of shellfish at recreational beaches
for toxic chemicals analysis. It is very likely that U.S. EPA will continue
82
-------
to conduct surveys of this nature and other surveys that would contribute
to the monitoring database.
NOAA (NMFS)—The National Marine Fisheries Service has conducted surveys
of fish for toxic chemicals and histopathological conditions (and concurrent
sediment quality analyses) at various locations in Puget Sound. Data from
these efforts would supplement data from corresponding monitoring components
in the Puget Sound Monitoring Program.
NOAA (PAD)—The Ocean Assessment Division and its predecessors, Office
of Marine Pollution Assessment, and Marine Ecosystems Analysis supported
most of the early surveys in Puget Sound on toxicants and their effects.
They may continue to fund additional work in the future that satisfies
their programmatic needs. Data from such efforts would complement the
Puget Sound Monitoring Program.
Washington Department of Ecology—The Washington Department of Ecology
performs intensive surveys that focus on sediment quality, water quality,
and biological components throughout the sound. The department a 1 so conducts
water quality surveys in rivers discharging to Puget Sound. Surveys are
performed for a variety of reasons ranging from assessment of contamination
from hazardous waste releases (e.g., toxic chemicals in bottom fish in
Commencement Bay) to pollutant source investigations (e.g., Elliott Bay
Action Team).
PROGRAM MODIFICATION AND PHASED IMPLEMENTATION
Modification of the proposed ambient monitoring program and phasing
of program implementation are discussed in this section. Compliance monitoring
programs and their costs are not discussed.
Possible modifications of the proposed ambient monitoring program
are indicated below:
83
-------
t Increase effort and budget allocated to ambient monitoring,
with a corresponding increase in the total budget for the
Puget Sound Monitoring Program
t Decrease effort and budget allocated to ambient monitoring,
with a corresponding decrease in the total budget or a real lo-
cation of funds to increase the effort for intensive surveys
• Phase in groups of new monitoring components (e.g., sediment
quality) according to their relative priority.
Increased Ambient Monitoring
Any increased effort for ambient monitoring should be devoted to increasing
the spatial coverage and frequency of key monitoring components. The following
changes in the design are suggested:
• Increase number of sampling stations for sediment quality
components, especially in urban and industrial bays
• Increase the frequency of monitoring histopathological conditions
in English sole liver and toxic chemicals in English sole
fillet (to annual sampling)
• Increase the number of beach stations for shellfish monitoring
components
• Increase the frequency of sampling in rivers.
Increases in sample replication should also be considered for each monitoring
component with a replicated sampling design. The value of increased replication
should be judged relative to 1) the minimum detectable difference among
stations (or among times) that is desired and 2) marginal cost of increased
sampling effort. The statistical power analysis framework presented in
Appendix A can be used to evaluate further sample replication.
84
-------
Decreased Ambient Monitoring
Any major reduction of effort for ambient monitoring should be achieved
by decreasing the number of monitoring components (or variables). The
proposed design for each monitoring component (Table 3 above) represents
the minimum adequate level of effort. Substantial reduction of spatial
coverage, sampling frequency, or level of replication for an individual
monitoring component would greatly weaken the design, without corresponding
benefits from cost savings. Loss of the ability to detect differences
among sampling stations and times would jeopardize achievement of the monitoring
objectives (Table 2 above). Segar and Stamman (1986) offer similar cautions
against specifying inadequate effort for individual monitoring variables
because of the reluctance to drop variables. The effort devoted to each
monitoring component need not remain stable forever. Reevaluation of the
monitoring program after data have been collected for 3-5 yr is recommended.
At that time, analyses of the data may indicate that reductions in spatial
coverage, sampling frequency, or level of replication for individual components
(or variables) are warranted.
A priority ranking of monitoring components is needed if some components
are to be dropped from this program. An initial priority list was developed
by considering the following factors:
• Sensitivity to anthropogenic impacts of concern
• Natural variability
t Statistical precision achieved for a given level of replication
• Use in previous monitoring programs or projected use in
decision-making by an agency
• Ability to complement other monitoring components or ongoing
monitoring programs.
85
-------
A preliminary list of components, in approximate order from highest to
lowest priority for inclusion in the ambient monitoring program, is given
below:
HIGH
Pathogen indicators in shellfish
Toxic chemicals in shellfish
PSP in shellfish
Sediment chemistry
Conventional sediment variables
Sediment toxicity bioassays
Benthic macroinvertebrate species abundances
Hydrographic conditions
Dissolved oxygen
Histopathological abnormalities in fish
Toxic chemicals in fish
NPDES compliance monitoring
Nonpoint source monitoring
River monitoring
Habitat types
Ancillary data
MEDIUM
Pathogen indicators in water
Marine mammal abundances and reproductive success
Avian abundances and reproductive success
LOW
Shellfish abundances
Fish species abundances
Nutrient concentrations
86
-------
Phytoplankton standing stock
Turbidity/transparency
Odor, floatables, slicks, water color.
Phased Implementation of Components
If initial funding is not sufficient to implement the entire ambient
monitoring program, then phased implementation should be considered. The
priority list recommended above should be used as a guide. However, groups
of related monitoring variables should be implemented simultaneously.
The recommended order of priority, from highest to lowest, is shown below:
t Contaminant source monitoring
NPDES compliance monitoring
River monitoring
Nonpoint sources
• Sediment contamination monitoring
Sediment chemistry
Conventional sediment variables
Sediment toxicity bioassays
Benthic macroinvertebrate abundances
Histopathological conditions in fish
Toxic chemicals in fish
• Shellfish monitoring
Pathogen indicators in shellfish
Toxic chemicals in shellfish
PSP in shellfish
Shellfish abundances
Pathogen indicators in water
87
-------
• Habitat monitoring
• Biological resource monitoring
Marine mammal abundances and reproductive success
Avian abundances and reproductive success
Shellfish abundances
Fish abundances
• Water quality monitoring
Hydrographic conditions
Dissolved oxygen
Turbidity/transparency
Nutrient concentrations
Phytoplankton standing stock
bdor, floatables, slicks, water color
• Ancillary data.
Further development of the contaminant source monitoring program is the highest
priority. A relatively high ranking was assigned to sediment contamination
monitoring because of the value of the data and the lack of a comprehensive
monitoring effort at present. Because some shellfish monitoring programs
are ongoing, implementation of the proposed program was assigned a moderate
priority despite the importance of protecting public health. Water quality
monitoring was assigned a relatively low priority for implementation because
an ongoing program would provide some data until the new program begins.
88
-------
ESTIMATED COSTS OF MONITORING PUGET SOUND
Estimated costs for the first year of the ambient and intensive elements
of the Puget Sound Monitoring Program are given in Table 12. Data entry
cost estimates include quality assurance (QA) checks, and were based on
the anticipated volume of records. Cost estimates for data analysis, report
writing, and program management cover all monitoring components combined.
The cost estimate for intensive surveys is not based on a specific survey
design, since intensive surveys will likely be performed on an "as needed"
basis and will vary greatly in size and objectives. Rather, it reflects
the estimated annual cost of implementing a reasonable program of intensive
surveys throughout the sound. Compliance monitoring costs will not be
addressed here. Cost estimates for compliance monitoring are being prepared
by the Washington Department of Ecology and PSWQA.
Estimated annual field and laboratory costs for individual components
of the proposed ambient monitoring program are given in Table 13. Costs
for ancillary data components (e.g., climate/weather) are not shown in
Table 13 because data compilation and management effort are included under
"Data Entry," "Data Analysis and Report Preparation," and "Management"
categories in Table 12. Furthermore, it is assumed that any additional
costs for collection or formatting of ancillary data would be borne by
the agencies responsible for data generation.
The cumulative cost of implementing groups of monitoring components
in sequence is shown in Figure 29. The order of components in Figure 29
corresponds to the recommended priority for inclusion in the program (see
above, Institutional Mechanisms for Program Management and Implementation,
Program Modification and Phased Implementation). To obtain the total cost
of each component shown in Figure 29, the estimated total cost of data
entry, data analysis, report preparation, and management from Table 12
was divided among components and added to each total cost for field and
89
-------
TABLE 12. ESTIMATED ANNUAL COSTS (IN THOUSANDS OF DOLLARS)
OF THE PUGET SOUND MONITORING PROGRAM
Program Element Cost/Yr
Ambient Monitoring Program
Station Siting Survey3 50
Field and Laboratory Effortb 2,480
Data Entry and QA 200
Data Analysis and Report Preparation 400
Total Ambient 3,130
Intensive Surveys 900
Management 70
Grand Total0 4,100
Net Costd 3,100b
3 Preliminary field surveys to select final station
locations.
b See Table 13 for field and laboratory costs of individual
monitoring components.
c Grand total includes cost for station siting survey
($50,000), pilot river program ($40,000), installation
of flow-gaging stations ($100,000), and a portion of
equipment costs ($38,000), all of which would be incurred
only during the first year. Equipment and replacement
costs will be incurred in each year of the program.
d Net cost equals grand total cost minus potential
amount reallocated from ongoing monitoring programs
(see text).
-------
TABLE 13. ESTIMATED ANNUAL FIELD AND LABORATORY COSTS FOR AMBIENT MONITORING COMPONENTS
Number of
Field Number of Field Cost per
Component Program ($) Code Stations Frequency Replicates Replicate ($)
Sediment Quality
Sediment chemistry
Sediment toxicity bioassays
Conventional sediment variables
Water Quality d
Hydrographic conditions
Dissolved oxygen
Turbidity/transparency
Odor, floatables, slicks, water color
Nutrient concentrations
Phytoplankton standing stock
Pathogen indicators in water
Biological Conditions d
Benthic macroinvertebrate abundances
Toxic chemicals in fish tissue:
Pacific Cod
Salmon
Engfish Sole
Histopathological abnormalities in fish
Fish species abundances
Shellfish abundances
Toxic chemicals in shellfish
PSP in shellfish
Pathogen indicators in shellfish
Marine mammal abundances and
reproductive success
Avian abundances and reproductive
success
River Monitoring
Toxic Chemicals
Conventional Variables
Habitat Types
Landsat
COE overflights
Field program (groundtruthing)
55,000
1,500
500
500
1 70,000
3.600
25,000
100
0
2,000
5,000
0
3,000
4,200
1,400
0
34,100
0
2,500
0
1,000
21,800
9,000
48,000
374,500
0
15,000
a
b
b
b
c
b
b
b
a
b
b
f
b
g
b
h
g
h
g
i
1
g
j
k
1
m
n
106
106
106
53
53
53
53
53
26
106
15
3
31
31
31
26
26
26
26
2
1
26
26
1
1
1
12
12
12
12
12
20
1
1
1
1
1
1
1
1
20
20
1
12
12
12
1
1
1
1
1
5
2 (depths)
1
3
5
3
3
3
1
4
15
3
3
3
1
1
1
1
1,430
1,400
200
12
10
0
85
30
40
400
1.200
1.200
1,200
1,500
0
0
1.320
50
60
500
400
1,250
230
500
1,400
Component
Cost ($)
55,000
153,080
148,900
21,700
170,000
11,232
31,360
100
0
110,120
24,080
62,400
215,000
0
58,200
12,200
111,600
80,600
0
2.500
102.960
79,000
115,400
10,000
52,800
764.500
71,760
500
1,400
15,000
Grand Total =
Total ($)
379,000
409,000
840,000
836,000
17,000
2,481,000
-------
TABLE 13. (Continued)
a
Field program casts for sediment quality plus benthic macroinvertebrates is included in this estimate. Ship time and crew only are accounted for.
Gear necessary for individual components are listed separately.
b
Cost of gear only.
c
Includes ship time and crew for all water quality sampling. Gear necessary for individual components are listed separately.
d
Requires major capital expenditures: CTO System (Hydrography) $60,000
Boxcorer (Benlhc macroinvertebrates) $20,000
e
Observations will be made routinely at all non-remote monitoring stations. Records of spills will also be obtained from the U.S. Coast Guard.
Because observations will be made in conjunction with other programs, costs are negligible.
f
Cost of field work included in pathogen indicators in shellfish component.
g
Gear plus field time.
h
Field costs included in fish pathology component
i
Gear only. Field time costed under pathogen indicators in shellfish.
i
Flight costs plus land-based surveys.
k
Flight costs for waterfowl censusing by U. S. Fish and Wildlife Service and Washington Department of Game, plus annual spring waterfowl
brood survey by Washington Department of Game.
Includes cost to install 10 continuously monitored, flow-gaging stations ($100,000), annual maintenance cost for 10 flow-gaging
stations ($50,000), initial purchase of two continuous-flow centrifuges ($8,000), and pilot program ($40,000).
m
Cost of field work included in toxic chemical component of river monitoring.
n
Photographic analysis would be conducted every 5 yrs. Groundtruthing surveys would cover one-fifth of the Puget Sound region each year.
-------
MONITORING COMPONENT
Pathogen indicators in shellfish
Toxic chemicals in shellfish*
PSP in shellfish*
Sediment chemistry
Conventional sediment variables
Sediment toxicity bktassays
Benthic macroinvertebrate species abundances'*
Hydrographic conditions
Dissolved oxygen
Hislopalholooical abnormalities in fish
Toxic chemicals in fish
River monitoring
Habitat types
Pathogen indicators in water*
Marine mammal abundances and reproductive success
Avian abundances and reproductive success
Shellfish abundances
Rsh species abundances6
Nutrient concentrations
Phytoplankton standing slock
Turbidity/transparency
Odor, Hoatabtos, slicks, water color
^v^
\,t f f 7%'##/f,#/,/' ?
Vxy.wS^'s#4»Sv** V,"*?'*+ **{$4»' 4*%8t ', '
{<<,%',*^:>,}Ti'-*' ^ r-i"f;f;/%r/''
^-'^4^%r* /< -; 'v":%f*^^x'/;;
•4,v,r*^^;jv«/s'^\ ^-'^n <
«$^£"*'V'' -" -s / V-* ^r^: "-
'"^.''s v-'-t s , '-'< ' -A' "?-',tsS-"J „ '-L-
, \ {v&tfcb&*& > A , v s\ ' *., *» ,'ty/
. ff f j.fJL\ i ¥5 % W* ^SSv ^ * ' X. . Lf f. .. ATA Vrf'.. ^ ' < A. .
. *.*,
500 1000 1500 2000 2500 3000
CUMULATIVE COST IN THOUSANDS OF DOLLARS
3500
Note: Capitol costs are included in each component where applicable (see text).
River toxics component includes $40,000 lor pilot program.
Cost of preliminary field survey ($50,000) lor selection oi final locations is not shown.
Figure 29. Cumulative costs of the ambient monitoring program as components are added.
-------
laboratory work from Table 13. For each component, this "added cost" was
weighted according to the contribution of the component to the total cost
for field and laboratory work. The total cost of river monitoring in Figure 29
includes the cost of a pilot program ($40,000). The cost of a survey to
select final station locations ($50,000, Table 12) was not included in
Figure 29 because the added cost per component is negligible.
All cost estimates in Tables 12 and 13 are preliminary. Actual costs
will depend on the study designs for the ambient monitoring program and
intensive surveys as implemented, and on the relationship between the Puget
Sound Monitoring Program and ongoing monitoring efforts.
Estimated costs of ongoing ambient monitoring programs are shown in
Table 14. Some of the present funds for monitoring by state agencies could
support related components of the Puget Sound Monitoring Program. Potential
implementation mechanisms are:
• An ongoing monitoring program is modified to become a recommended
component of the Puget Sound Monitoring Program (e.g., Washington
Department of Ecology's ambient water quality programs for
Puget Sound and rivers)
• An ongoing monitoring program, possibly with minor modifications,
is incorporated directly into the Puget Sound Monitoring
Program (e.g., WDG's surveys of marine mammals and waterfowl
U.S. FWS's surveys of avian abundances)
• A portion of an ongoing program is modified to become a
recommended component (e.g., DSHS's monitoring of PSP and
bacteria in shellfish and of bacteria in water).
This strategy could make available as much as $500,000 for the recommended
ambient monitoring program. In addition, $500,000 that are presented allocated
for intensive monitoring by Washington Department of Ecology could be applied
to intensive surveys under the Puget Sound Monitoring Program. Thus, a
90
-------
TABLE 14. ESTIMATED ANNUAL COSTS (IN THOUSANDS OF DOLLARS)
OF ONGOING AMBIENT MONITORING PROGRAMS FOR PUGET SOUND
Agency/Program Cost/Yr
Ecology
Water quality (Puget Sound) 50
Rivers 150
Metro
Water quality (offshore) 40
Pathogen indicators (beaches) 3
Duwamish head (to be determined) NAa
Rivers 200-225
DOD
Water quality 3
DSHS
Fecal coliforms in water 48
Fecal coliforms in shellfish 25
PSP in shellfish 100
NOAA
Status and trends 75
PSDOA
Disposal site management (to be determined) NA
WDF
Salmon run NA
Shellfish abundance NA
Geoduck stock assessment NA
Oyster larvae abundance NA
Juvenile and Pink salmon 25
Herring spawning ground 50
Surf smelt spawning ground 10
Hake abundance 20
Shrimp abundance NA
Sea urchin abundance NA
Salmon, ground fish, herring, and
shellfish harvest 40
Fish habitat surveys 110
WDG
Sea population estimates 8
Mid-winter waterfowl survey 5
Waterfowl brood production 11
Waterfowl harvest 5
USFWS
Avian abundance 15
U.S.GS
Rivers 100
a NA = Not available.
-------
total of at least $1,000,000 from ongoing programs could be made available
to support the proposed programs. This would reduce the total cost of
the proposed ambient and intensive programs from $4,010,000 to $3,010,000.
91
-------
REFERENCES
American Public Health Association. 1985. Standard methods for examination
of waste and wastewater. A.E. Greenberg, R. Trusseil, and L.S. Clersceri
(eds). APHA, Washington, DC. 1268 pp.
Baker, E.T. 1982. Suspended particulate matter in Elliott Bay. NOAA
Tech. Rept. ERL 417-PMEL 35.
Bates, T.S., S.E. Hamilton, and J.D. Cline. 1984. Vertical transport
and sedimentation of hydrocarbons in the central main basin of Puget Sound,
Washington. Environ. Sci. Technol. 18:299-305.
Beach, R.J., A.C. Geigar, S.J. Jeffries, S.D. Treacy, and B.L. Troutman.
1985. Marine mammals and their reaction with fisheries of the Columbia
River and adjacent waters, 1980-1982. Final Report to National Oceanic
and Atmospheric Administration, National Marine Fisheries Service Grant
No. 80-ADB-0012. Northwest Alaska Fisheries Center, Seattle, WA. 316 pp.
Belanger, S.E., J.L. Farris, D.S. Cherry, and J. Cairns, Jr. 1986. Growth
of Asiatic clams (Corbicula sp.) during and after long-term zinc exposure
in field-located and laboratory artificial streams. Arch. Environ. Contam.
Toxicol. 15:427-434.
Bernstein, B.B., and J. Zalinski. 1983. An optimum sampling design and
power tests for environmental biologists. J. Environ. Manage. 16:35-
43.
Box, G.E.P,, and G.M. Jenkins. 1976. Time series analysis: forecasting
and control. Second edition. Hoi den-Day, San Francisco, CA.
Buchanan, J.B. 1984. Sediment analysis, pp. 41-65. In: Methods for
the Study of Marine Benthos. IBP Handbook No. 16, Second Edition. N.A. Holme
and A.D. Mclntyre (eds). Blackwell Scientific Publications, Oxford, UK.
Cairns, J., Jr., and W.H. van der Schalie. 1980. Biological monitoring
Part I - early warning systems. Water Res. 14:1179-1196.
Calambokidis, J. 2 June 1986. Personal Communication (letter to Dr. Robert
A. Pastorok). Cascadia Research Collective, Olympia, WA.
Campbell, S.A., W.K. Peterson, and J.R. Postel. 1977. Phytoplankton production
and standing stock in the main basin of Puget Sound. Final Report. Prepared
for the Municipality of Metropolitan Seattle by the University of Washington
Department of Oceanography. 132 pp.
92
-------
Chapman, P.M., and E.R. Long. 1983. The use of bioassays as part of a
comprehensive approach to marine pollution assessment. Mar. Pollut. Bull.
14:81-84.
Chapman, P.M., R.N. Dexter, L.S. Goldstein, and E.A. Quinlan. 1985. Develop-
ment of effective regional environmental monitoring for Puget Sound. NOAA
Tech. Memo. NOS OMA 22. National Oceanic and Atmospheric Administration,
Rockville, MD.
Cohen, J. 1977. Statistical power analysis for the behavioral sciences.
Academic Press, New York, NY.
Cokelet, E.D., R.J. Stewart, and C.C. Ebbesmeyer. 1984. The exchange
of water in fjords: a sample model of two-layer advective reaches separated
by mixing zones, pp. 3124-3133. In: Proceedings of the 19th Coastal
Engineering Conference, Houston, TX.
Collias, E.E., and J.H. Lincoln. 1977. A study of the nutrients in the
main basin of Puget Sound. Unpublished report prepared for the Municipality
of Metropolitan Seattle, Seattle, WA. 151 pp.
Comiskey, C.E., T.A. Farmer, C.C. Brandt, and G.P. Romberg. 1984. Toxicant
Pretreatment Planning Study. Technical Report C2: Puget Sound Benthic
Studies and Ecological Implications. Municipality of Metropolitan Seattle,
Seattle, WA.
Conover, R.J., R. Durvasula, S. Roy, and R. Wang. 1986. Probable loss
of chlorophyll-derived pigments during passage through the gut of zooplankton,
and some of the consequences. Limnol. Oceanogr. 31:878-886.
Cross, J.N. 1982. Evaluation of otter trawl data. pp. 91-97. In: Coastal
Water Research Project, Biennial Report for the Years 1981-1982. W. Bascom
(ed). SCCWRP, Long Beach, CA.
Dexter, R.N., D.E. Anderson, E.A. Quinlan, L.S. Goldstein, R.M. Strickland,
R.M. Kocan, M. Landolt, J.P. Pavlou, and J.R. Clayton, Jr. 1981. A summary
of knowledge of Puget Sound. NOAA Tech. Memo. OMPA-13. 435 pp.
Dolan, D.M., A1K. Yui, and R.D. Geist. 1981. Evaluation of river load
estimation methods for total phosphorus. J. Great Lakes Res. 7:207-214.
Ebbesmeyer, C.C., C.A. Coomes,-J.M. Cox, and J.M. Helseth. 1982. Historical
oceanographic data in East Passage and approaches. Prepared for the Munici-
pality of Metropolitan Seattle, Seattle, WA. 514 pp.
Edmondson, W., and J. Lehman. 1981. The effect of changes in the nutrient
income on the condition of Lake Washington. Limol. Oceanogr. 26:1-29.
Eleftheriou, A., and N.A. Holme. 1984. Macrofauna techniques, pp. 140-
216. In: Methods for the Study of Marine Benthos. IBP Handbook No. 16,
Second Edition. N.A. Holme and A.D. Mclntyre (eds). Blackwell Scientific
Publications, Oxford, UK.
93
-------
Entrance Engineers. 1986. The state of the Sound. 1986. Draft Report.
Prepared for the Puget Sound Water Quality Authority, Kirkland, WA. 146 pp.
plus appendices.
Etzioni, A. 1965. Dual leadership in complex organizations. American
Sociological Review 30: 688-698.
Evans, N., M.J. Hershman, G.V. Blomberg, and W.B. Lawrence. 1980. The
search for predictability: planning and conflict resolution in Grays Harbor,
Washington. Washington Sea Grant Division of Marine Resources. University
of Washington, Seattle, Washington. 117 pp.
Evans-Hamilton, Inc., and D.R. Systems, Inc. 1986. Puget Sound environmental
atlas. Prepared for U.S. Environmental Protection Agency, U.S. Army Corps
of Engineers, and Puget Sound Water Quality Authority. Seattle, WA.
Ferguson, F.I. 1986. River loads underestimated by rating curves. Water
Resour. Res. 22:74-76.
Folk, R.L. 1968. Petrology of sedimentary rocks. University of Texas,
Austin, TX. 172 pp.
Gauch, H.G. 1982. Multivariate analysis in community ecology. Cambridge
Studies in Ecology:!. Cambridge University Press, Cambridge. 298 pp.
Glass, G.V., P.O. Peckham, and J.R. Sanders. 1972. Consequences of failure
to meet assumption underlying the analysis of variance and covariance.
Rev. Educ. Res. 42(3):237-288.
Gray, J.S., D. Boesch, C. Heys, A.M. Jones, J. Lassig, R. Vanderhorst,
and D. Wolfe. 1980. The role of ecology in marine pollution monitoring.
Rapp. P-V. Reun. Cons. Int. Explor. Mer. 179:237-252.
Green, R.H. 1979. Sampling design and statistical methods for environmental
biologists. John Wiley and Sons, Inc., New York, NY. 257 pp.
Greenberg, A.E., and D.A. Hunt (eds). 1984. Laboratory procedures for
the examination of seawater and shellfish. Fifth Edition. American Public
Health Association, Washington, DC.
Grieb, T. 1984. Robustness of the ANOVA model in environmental monitoring
applications. Final Report. Prepared for Electric Power Research Institute,
Palo Alto, California, by Tetra Tech, Inc., Lafayette, CA.
Guy, H.P., and V.W. Norman. 1970. Field methods for measurement of fluvial
sediment. Chapter 2. p. 59. In: Techniques of Water Resources Investi-
gations. U.S. Geological Survey.
Hamilton, S.E. 1980. Hydrocarbons associated with suspended matter in
the Green River, Washington. Masters thesis. University of Wash. Seattle,
WA.
94
-------
Harper-Owes. 1983. Water quality assessment of the Duwamish Estuary,
Washington. Final Report Prepared for Municipality of Metropolitan Seattle,
Harper-Owes, Seattle, WA.
Hirsch, R.M. 1986. Conceptual design of the surface water component of
the National Water Quality Assessment (NAWQA) Program, pp. 779-784. In:
Oceans 86 Proceedings. Vol. 3. Monitoring Strategies Symposium. Marine
Techology Society, Washington, DC.
Home, A.O., and S. McCormick. 1978. An assessment of eutrophication
in San Francisco Bay. Prepared for the Association of Bay Area Governments,
Berkeley, CA. 136 pp.
Horner, R.R., B.W. Mar, L.E. Reinelt, J.S. Richey, and J.M. Lee. 1986.
Design of monitoring programs for determination of ecological change resulting
from nonpoint source water pollution in Washington State. Prepared for
Washington State Department of Ecology, Olympia, WA.
Karr, J. 1981. Assessment of biotic integrity using fish communities.
Fisheries 6:21-27.
Karr, J., and D. Dudley. 1981. Ecological perspective on water quality
goals. Environ. Manag. 5:55-68.
Karr, J., and I. Schlosser. 1978. Water resources and the land-water
interface. Science 201:229-234.
Kendall, R. June 1986. Personal Communication (phone by Dr. Robert A.
Pastorok). Western Washington University, Bellingham, WA.
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 the occurrence of hepatic lesions in English Sole (Parophyrys
vetulus) from Puget Sound, Washington. Arch. Environ. Contam. Toxicol.
15:61-67.
Krumbein, W.C., and F.J. Pettijohn. 1938. Manual of sedimentary petrography.
Appleton-Centry Crafts, NY. 549 pp.
Kuntz, K., C.H. Chan, A.W. Clignett, R. Boucher. 1982. Water quality
sampling methods at Niagara on the lake, Inland Waters Directorate. Environment
Canada, Ontario, Canada.
Landolt, M.L., F.R. Hafer, A. Nevissi, G. Van Belle, K. Van Ness, and C.
Rockwell. 1985. Potential toxicant exposure among consumers of recreationally
caught fish from urban embayments of Puget Sound. NOAA Tech. Memo. NOS-
OMA-23. National Oceanographic and Atmospheric Administration, Rockville, MD.
Lettenmaier, D.P. 1976. Detection of trends in water quality data from
records with dependent observations. Water Res. Bull. 12:1037-1046.
95
-------
Lettenmaier, D.P., L.L. Conquest, and J.P. Hughes. 1982. Routine streams
and rivers water quality trend monitoring review. Tech. Rep. No. 75,
Charles W. Harris. Hydraulics Laboratory, University of Washington, Seattle,
WA. 223 pp.
Likens, G., F. Bormann, N. Johnson, D. Fisher, and R. Pierce. 1970. Effect
of forest cutting and herbicide treatment or nutrient budgets in the Hubbard
Brook watershed-ecosystem. Ecological Monographs 40:23-47.
Malins, D.C., B.B. McCain, D.W. Brown, A.K. Sparks, H.O. Hodgins, and S.L.
Chan. 1982. Chemical contaminants and abnormalities in fish and invertebrates
from Puget Sound. NOAA Tech. Memo. OMPA-19. National Oceanic and Atmospheric
Administration, Boulder, CO.
Malins, D.C., 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.O. Hodgins. 1984. Chemical pollutants in sediments and diseases
of bottom-dwelling fish in Puget Sound, Washington. Environ. Sci. Technol .
18:705-713.
Malins, D.C., M.M. Krahn, D.W. Brown, L.D. Rhodes, M.S. Myers, B.B. McCain,
and S.L. Chan. 1985a. Toxic chemicals in marine sediments and biota from
Mukilteo, Washington: relationships with hepatic neoplasms and other hepatic
lesions in English Sole (Parophyrys vetulus). J. Nat. Cancer Inst 74:487-
494.
Malins, D.C., M.M. Krahn, M.S. Myers, L.D. Rhodes, D.W. Brown, C.A. Krone,
B.B. McCain, S.L. Chan. 1985b. Toxic chemicals in marine sediments and
biota from Mukilteo, Washington: relationships with hepatic neoplasms
and other hepatic lesions in English Sole (Parophyrys vetulus). Carcino-
genesis 6:1463-1469.
Mar, B.W., R.R. Homer, O.S. Richey, R.N. Palmer, and D.P. Lettenmaier.
1986. Data acquisition. Environ. Sci. Technol. 20:545-551.
Mass, R.P., M.D. Smolen, S.A. Dresisng, C.A. Jamieson, and J. Spooner.
1985. Practical guidelines for selecting critical areas for controlling
nonpoint source pesticide contamination of aquatic systems, pp. 363-367.
In: Perspectives on Nonpoint Pollution. Proceedings of a conference held
in Kansas City, Missouri May 19-22, 1985. EPA 440/5-85/001. U.S. Environmental
Protection Agency, Washington, DC.
McCallum, M. 1985. Recreational and subsistence catch and consumption
of seafood from three urban industrial bays of Puget Sound: Port Gardner,
Elliott Bay, and Sinclair Inlet. State of Washington. Department of Social
and Health Services, Olympia, WA.
McCrea, R.C., and J.D. Fischer. 1984. Construction and operation of a
continuous sampling system for monitoring organochlorine contaminants in
natural waters, Inland Waters Directorate. Environment Canada, Ontario,
Canada.
96
-------
Mclntyre, A.O., J.M. Elliott, and D.V. Ellis. 1984. Design of sampling
programmes, pp. 1-26. In: Methods for the Study of Marine Benthos.
JBP Handbook No. 16, Second Edition. N.A. Holme and A.D. Mclntyre (eds).
Blackwell Scientific Publications, Oxford, UK.
Mearns, A.J., and J.M. Allen. 1978. Use of small otter trawls in coastal
biological surveys. EPA-600/3-78-083. U.S. Environmental Protection Agency,
Corvallis, OR. 33 pp.
Metro. 1978. Areawide water quality plan. Municipality of Metropolitan
Seattle. Seattle, WA. 139 pp.
Mil lard, S.P., and D.P. Lettenmaier. 1986. Optimal design of biological
sampling programs using the analysis of variance. Estuarine, Coastal and
Shelf Science 22:637-656.
Millard, S.P., J.R. Yearsley, and D.P. Lettenmaier. 1985. Space-time
correlation and its effects on methods for detecting aquatic ecological
change. Can. J. Fish. Aquat. Sci. 42:1391-1400.
Miller, W., S. Peterson, J. Greene, and C. Callahan. 1985. Comparative
toxicology of laboratory organisms for assessing hazardous waste sites.
J. Environ. Qual. 14:569-574.
Montgomery, R.H., and K.H. Reckhow. 1984. Techniques for detecting trends
in lake water quality. Water Res. Bull. 20:43-52.
Myers, M.S., L.D. Rhodes, and B.B. McCain. (In review). Pathologic anatomy
and patterns of hepatic neoplasms, putative preneoplastic lesions and other
idiopathic hepatic conditions in English Sole (Parophyrys vetulus) from
Puget Sound, Washington. Submitted to the J. Nat. Cancer Inst.
Nichols, F.H. 1985. Abundance fluctuations among benthic invertebrates
in two Pacific estuaries. Estuaries 8:136-144.
Nielsen, L.A., and D.L. Johnson. 1983. Fisheries techniques. American
Fisheries Society, Bethesda, MD. 468 pp.
Ongley, E.D. 1973. Sediment discharge from Canadian basins into Lake
Ontario. Can. J. Earth Sci. 10:146-156.
Ongley, E.D., and D.P. Blackford. 1982. Application of continuous flow
centrifigation to contaminant analyses of suspended sediment in fluvial
systems. Environmental Technology Letters 3:219-228.
Paine, R.T. 1986. Benthic community-water column coupling during the
1982-1983 El Nino. Are community changes at high latitudes attributable
to cause or coincidence? Limnol. Oceanogr. 31(2):351-360.
97
-------
Patrick, R. 1971. Diatom communities, pp. 151-164. In: The Structure
and Function of Fresh-water Microbial Communities. J. Cairns, Jr. (ed).
Research Division Monograph 3, Virginia Polytechnic Institute and State
University, Blacksburg, VA.
Patrick, R., M.H. Hohn, and J.H. Wallace. 1954. A new method for determining
the pattern of the diatom flora. Natul. Nat. 259:1-12.
Peterson, S.A., W.E. Miller, J.C. Greene, and C.A. Callahan. 1985. Use
of bioassays to determine potential toxicity effects of environmental pollu-
tants, pp. 38-45. In: Perspectives on Nonpoint Source Pollution. Proceedings
of a conference held in Kansas City, Missouri. May 19-22, 1985. EPA 440/
5-85-001. U.S. Environmental Protection Agency. Washington, DC.
Plumb, R.H., Jr. 1981. Procedures for handling and chemical analysis
of sediment and water samples. Tech. Rep. EPA/CE-81-1. U.S. Environmental
Protection Agency/Corps of Engineers Technical Committee on Criteria for
Dredged and Fill Material, U.S. Army Waterways Experiments Station, Vicksburg,
MS. 471 pp.
Poelker, R. 20 May 1986. Personal Comunication (meeting with Dr. Robert
A. Pastorok). Washington Department of Game, Olympia, WA.
Puget Sound Task Force of the Pacific Northwest River Basins Commission.
1970. Comprehensive study of water and related land resources. Puget
Sound and adjacent waters. Appendix IV. Water-related Land Resources.
Pacific Northwest River Basins Commission. Seattle, WA.
Puget Sound Water Quality Authority. 1986a. Comprehensive Monitoring
of Puget Sound. Seattle, WA.
Puget Sound Water Quality Authority. 1986b. The state of the Sound, 1986.
Seattle, WA.
Puget Sound Water Quality Authority. 1986c. 1987 Puget Sound water quality
management plan and environmental impact statement. Seattle, WA.
Richey, J.E., R.H. Meade, E. Salati, A. Devol, C.F. Nordin, and U. dos
Santos. 1986. Water discharge and suspended sediment concentrations in
the Amazon River: 1982-1984. Water Resour. Res. 22:756-764.
Rodda, J.C., and G.N. Jones. '1983. Preliminary estimates of loads carried
by rivers to estuaries and coastal waters around Great Britian derived
from the Harmonized Monitoring Scheme. J. Inst. Wat. Eng. Sci. 37:529-539.
Russek, E., and R.R. Colwell. 1983. Computation of most probable numbers.
Appl. Environ. Microbiol. 45:1646-1650.
Sanders, T.G., R.C. Ward, J.C. Loftis, T.D. Steele, D.D. Adrian, V. Yevjevich.
1983. Design of networks for monitoring water quality. Water Resource
Publications, Littleton, CO.
98
-------
Scheffe, H. 1959. The analysis of variance. John Wiley and Sons, New
York, NY. 477 pp.
Schmitt, C.J. 1981. Analysis of variance as a method for examining contaminant
residues in fish: National Pesticide Monitoring Program, pp. 270-298.
In: Aquatic Toxicology and Hazard Assessment, Fourth Conference. D.R. Branson
and K.L. Dickson (eds). ASTM STP 737. American Society for Testing and
Materials, Philadelphia, PA.
Segar, D.A., and E. Stamman. 1986. Fundamentals of marine pollution monitoring
programme design. Mar. Pollut. Bull. 17:194-200.
Shepard, F.P. 1954. Nomenclature based on sand-silt-clay ratios. J. Sedi-
ment. Petrol. 24:151-158.
Skopp, J., and T.C. Daniel. 1978. A review of sediment predictive techniques
as viewed from the perspective of nonpoint pollution management. Environ.
Manage. 2:39-53.
Stevens, H.H., G.A. Lutz, and D.W. Hubbell. 1980. Collapsible-bag suspended-
sediment sampler, pp. 11-16. Proceedings of the American Society of Civil
Engineers. Volume 106.
Soil Conservation Service. 1965. Mean Annual Precipitation 1930-1957
State of Washington. M-4430 Portland, OR.
Sokal, R.R., and F.J. Rohlf. 1981. Biometry. Second Edition. W.H. Freeman
& Co., San Francisco, CA. 859 pp.
Stober, Q.J., and K.K. Chew (eds). 1984. Renton sewage treatment plant
project: Duwamish Head. Report for the Municipality of Metropolitan Seattle,
Seattle, WA. 370 pp.
Stofan, P.E., and G.C. Grant. 1978. Phytoplankton sampling in quantitative
baseline and monitoring programs. EPA-600/3-78-025. U.S. Environmental
Protection Agency, Corvallis, OR. 27 pp.
Strange, R.J. 1983. Field examination of fish. pp. 337-347. In: Fisheries
Techniques. L.A. Nielsen, D.L. Johnson, and S.S. Lampton (eds). American
Fisheries Society, Bethesda, MD.
Still, D.A., and S.F. Shih. 1985. Using LANDSAT data to classify land
use for assessing the basinwide runoff index. Water Resour. Bull. 21:931-940.
Strickland, J.D.H., and T.R. Parsons. 1972. A practical handbook of seawater
analysis. Bulletin 167 (2nd edition). Fisheries Research Board of Canada,
Ottawa, Canada. 310 pp.
Sullivan, J.J., J. Jonas-Davis, and L.L. Kentala. No date. The determination
of PSP toxins by HPLC and autoanalyzer. Manuscript. U.S. Food and Drug
Administration, Seattle, WA.
99
-------
Sullivan, J.J., and M.M. Wekel1. 1984. Determination of paralytic shellfish
poisoning toxins by high pressure liquid chromatography. American Chemical
Society Symposium Series, No. 262, Seafood Toxins.
Swartz, R.C. 1978. Techniques for sampling and analyzing the marine macro-
benthos. EPA-600/3-78-030. U.S. Environmental Protection Agency, Corvallis,
OR. 27 pp.
Swartz, R.C., W.A. DeBen, J.K. Phillips, 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 Sym-
posium. ASTM STP 854. R.D. Cardwel1, R. Purdy, and R.C. Bahrer (eds).
Tatem, H.E. Bioaccumul ation of pol ychlorinated biphenyls and metal from
contaminated sediment by freshwater prawns, Macrobrachium rosenbergi i and
clams, Corbicula fluminea. Arch. Environ. Contam. Toxicol. 15:171-183.
Tetra Tech. 1985a. Commencement Bay nearshore/tideflats remedial investiga-
tion. Final Report. 2 vols. Prepared for Washington State Department
of Ecology and U.S. Environmental Protection Agency. Tetra Tech, Inc.,
Bellevue, WA.
Tetra Tech. 1985b. Elliott Bay Toxics Action Plan: Initial data summaries
and problem identification. Draft Report Prepared for U.S. Environmental
Protection Agency. Tetra Tech, Inc., Bellevue, WA.
Tetra Tech. 1985c. Everett Harbor Toxics Action Plan: Initial data summaries
and problem identification. Draft Report Prepared for U.S. Environmental
Protection Agency. Tetra Tech, Inc., Bellevue, WA.
Tetra Tech. 1985d. Puget Sound data management system development. Prepared
for the U.S. Environmental Protection Agency, Office of Puget Sound, Seattle,
WA. 71 pp. plus appendices.
Tetra Tech. 1986a. Bioaccumul ation monitoring guidance: 5. strategies
for sample replication and compositing. Prepared for the Marine Operations
Division, Office of Marine and Estuarine Protection, U.S. Environmental
Protection Agency, Washington, DC. Tetra Tech, Inc., Bellevue, WA. 46 pp.
Tetra Tech. 1986b. Guidance manual for health risk assessment of chemically
contaminated seafood. Prepared for the U.S. Environmental Protection Agency,
Region X, Seattle, Washington. Tetra Tech, Inc., Bellevue, WA.
Tetra Tech. 1986c. Quality assurance and quality control (QA/QC) for
301(h) monitoring programs guidance on field and laboratory methods. Prepared
for Marine Operations Division, Office of Marine and Estuarine Protection,
U.S. Environmental Protection Agency, Washington, DC. Tetra Tech, Inc.,
Bellevue, WA. 267 pp. plus appendices.
100
-------
Tetra Tech. 1986d. Recommended protocols for conducting fish pathology
studies in Puget Sound. Draft Report. Prepared for U.S. Environmental
Protection Agency- Region X, Office of Puget Sound. Tetra Tech, Inc.,
Bellevue, WA.
Tetra Tech. 1986e. Recommended protocols for measuring conventional sediment
variables in Puget Sound. Prepared for U.S. Environmental Protection Agency,
Region X, Office of Puget Sound. Tetra Tech, Inc., Bellevue, WA. 46 pp.
Tetra Tech. 1986f. Recommended protocols for measuring metals in Puget
Sound sediment and tissue samples. Draft Report. Prepared for U.S. Environ-
mental Protection Agency, Region X, Office of Puget Sound. Tetra Tech,
Inc., Bellevue, WA.
Tetra Tech. 1986g. Recommended protocols for measuring organic compounds
in Puget Sound sediment and tissue samples. Draft Report. Prepared for
U.S. Environmental Protection Agency, Region X, Office of Puget Sound.
Tetra Tech, Inc., Bellevue, WA. 55 pp.
Tetra Tech. 1986h. Recommended protocols for sampling and analyzing subtidal
benthic macroinvertebrate assemblages in Puget Sound. Draft Report. Prepared
for U.S. Environmental Protection Agency, Region X, Office of Puget Sound.
Tetra Tech, Inc., Bellevue, WA. 37 pp.
Tetra Tech. 1986i. User's manual for pollutant of concern matrix. Draft
Report. Prepared for U.S. Environmental Protection Agency, Region X, Office
of Puget Sound. Tetra Tech, Inc., Bellevue, WA. 34 pp. plus appendices.
Tetra Tech. 1986J. ODES user's guide: Supplement A. Description and
use of Ocean Data Evaluation System (ODES) tools. Draft Report. Prepared
for U.S. Environmental Protection Agency, Office of Marine and Estuarine
Protection. Tetra Tech, Inc., Bellevue, WA. 198 pp.
Tetra Tech. 1986k. Laboratory analytical protocol for storm drain samples.
Commencement Bay Nearshore/Tideflats Feasiblity Study. Prepared for Washington
Department of Ecology and U.S. Environmental Protection Agency. Tetra
Tech, Inc., Bellevue, WA.
Tetra Tech. 19861. Analytical methods for U.S. Environmental Protection
Agency priority pollutants and particulate matter from discharges and receiving
water. Tetra Tech, Inc., Bellevue, WA.
Tetra Tech and E.V.S. Consultants. 1986a. Recommended protocols for conducting
laboratory bioassays on Puget Sound sediments. Final Report. Prepared
for U.S. Environmental Protection Agency, Region X, Office of Puget Sound.
Tetra Tech, Inc., Bellevue, WA. 55 pp.
Tetra Tech and E.V.S. Consultants. 1986b. Recommended protocols for micro-
biological studies in Puget Sound. Draft Report. Prepared for U.S. Environ-
mental Protection Agency, Region X, Office of Puget Sound. Tetra Tech,
Inc., Bellevue, WA. 16 pp.
101
-------
Thomas, R.B. 1985. Estimating total suspended sediment yield with probability
sampling. Water Resources Res. 21:1381-1388.
URS Company- (unpublished manuscript). Toxic chemicals and biological
effects on Puget Sound: status and scenarios for the future. Draft Report.
Prepared for National Oceanic and Atmospheric Administration, Seattle,
WA.
U.S. Department of Commerce. 1970. Census of population. Vol. I. Charac-
teristics of the population. Chapter A. Number of Inhabitants. Part
49. Washington PC70-1-A49. Bureau of the Census. Washington, DC.
U.S. Department of Commerce. 1980. Census of population. Vol. I. Charac-
teristics of the population. Chapter A. Number of Inhabitants. Part
49. Washington PC80-1-A49. Bureau of the Census. Washington, DC.
U.S. Department of the Navy. 1985. Final environmental impact statement.
Carrier battle group. Puget Sound region ship homeporting project. Technical
Appendices. Volume 2, Appendix S. U.S. Department of the Navy, Western
Division, Naval Facilities Engineering Command, San Bruno, CA. (Raw data
sheets obtained from Parametrix, Inc., Bellevue, WA.)
U.S. Environmental Protection Agency. 1976. Basic water monitoring program.
EPA 440/9-76-025. 51 pp.
U.S. Environmental Protection Agency. 1979 (revised March, 1983). Methods
for chemical analysis of water and wastes. EPA 600/4-79-020. U.S. EPA,
Environmental Monitoring and Support Laboratory, Cincinnati, OH.
U.S. Environmental Protection Agency. 1982. Methods for organic chemical
analysis of municipal and industrial wastewater. EPA 600/4-79-057. U.S. EPA,
Environmental Monitoring and Support Laboratory, Cincinnati, OH.
U.S. Environmental Protection Agency. 1983. Methods for chemical analysis
of water and wastes. EPA 600/4-79-020. (Revised 1983). U.S. EPA, Environ-
mental Monitoring and Support Laboratory, Cincinnati, OH.
U.S. Environmental Protection Agency. 1985a. Natonal Pollutant Discharge
Elimination System permit regulations; modification of application deadline
for stormwater point sources. U.S. EPA, Washington, D.C. Federal Register,
Vol. 50, No. 168. pp. 35200-35203.
U.S. Environmental Protection Agency. 1985b. Technical support document
for water quality-based toxics control. EPA-440/4-85-032. U.S. EPA, Office
of Water, Washington, D.C. 74 pp. plus appendices.
U.S. Environmental Protection Agency. 1985c. Water quality criteria;
availability of documents; notice. U.S. EPA, Washington, D.C. Federal
Register, Vol. 50 No. 145 Part II. pp. 30,784-30,796.
102
-------
U.S. Food and Drug Administration. 1985. Chemical Contaminants, Chapter 4.
Pesticides and industrial chemicals in domestic foods (FY 86). In: Food
and Drug Administration Compliance Program Guidance Manual. Program 7304.004.
Verhoff, F.H., S.M. Yaksich, and D.A. Melfi. 1980. River nutrient and
chemical transport estimation. J. Environ. Eng. Div., Amer. Soc. Civil
Eng. 106:591-608.
Walling, D.E., and B.W. Webb. 1981. The reliabi1ity of suspended load
data. pp. 177-194. In: Erosion and Sediment Transport Measurement.
IAHS Publ. No. 133. (As cited by Walling and Webb 1985.)
Walling, D.E., and B.W. Webb. 1985. Estimating the discharge of contaminants
to coastal waters by rivers: some cautionary comments. Mar. Pollut. Bull.
16:488-492.
Ward, R.C., J.C. Loftis, and G.B. McBride. 1986. The "data-rich but infor-
mation-poor" syndrome in water quality monitoring. Environ. Manage. 10:291-297.
Washington Administrative Code. 2 June 1982. Water quality standards
for waters of the State of Washington. Chapter 173-201 WAC. Olympia,
WA.
Weber, C.I. 1980. Federal and state biomonitoring programs, pp. 25-52.
In: Biological Monitoring for Environmental Effects. D.L. Worf (ed).
Lexington Borks, Lexington, MA.
Williams, J.R., H.E. Pearson, J.D. Wilson. 1985a. Streamflow statistics
and drainage - basin characteristics for the Puget Sound region, Washington.
Volume II: Eastern Puget Sound from Seattle to the Canadian border. U.S.
Geological Survey. Open File Report 84-144-B. Tacoma, WA.
Williams, J.R., H.E. Pearson, J.D. Wilson. 1985b. Streamflow statistics
and drainage - basin characteristics for the Puget Sound region, Washington.
Volume I: Western and Southern Puget Sound. U.S. Geological Survey.
Open File Report 84-144-A. Tacoma, WA.
Wissman, R., J. Richey, A. Devol, and D. Eggers. 1982. pp. 333-385.
In: Analysis of coniferous forest ecosystems in the western United States.
R. Edmonds (ed). Hutchinson Ross, Sundbury, PA.
Wolfe, L.D.S. 1982. Fraser River Estuary Study: Organizational Options
for linked management technical background report, phase II. Prepared
by L.D.S. Wolfe for the Management Systems Sub-committee. Fraser River
Estuary Study. Vancouver, British Columbia. 79 pp.
Word, J.Q., P.L. Striplin, K. Keeley, J. Ward, P. Sparks-McConkey, L. Bentler,
S. Hulsman, K. Li, and J. Schroder. 1984. Renton sewage treatment plant
project: Seahurst baseline study. Vol. V. Subtidal benthic ecology.
Report for the Municipality of Metropolitan Seattle. Prepared by University
of Washington Fisheries Research Institute, Seattle, WA. 461 pp.
103
-------
Yake, W.E. 1985. Impact of Western Processing on water quality in Mill
Creek (Kent, WA). Washington Department of Ecology, Olympia, WA.
104
-------
APPENDICES
-------
APPENDIX A
POWER ANALYSES
-------
POWER ANALYSES
INTRODUCTION
The probability of correctly rejecting a false null hypothesis is
referred to as the power of a statistical test. Since two of the objectives
of the monitoring program are to correctly detect the effects with respect
to time and station location, the power of a statistical test serves as
a basis for evaluating the performance of the monitoring program. When
existing data are available for the selected monitoring variables, power
calculations can be made to provide a quantitative comparison of alternative
sampling plans. For example, the probability of correctly detecting effects
with respect to time or station location can be determined for a specified
level of sampling effort. These methods can also be used to evaluate and
interpret statistical analyses in which the null hypothesis has been accepted.
In this case, the probability of detecting specific levels of differences
between stations or effects associated with different treatments in the
analysis can be determined for the fixed parameters of the experimental
design, and a judgement can be made as to whether the sampling plan was
adequate to reliably conclude that no effects were present.
The topics addressed in the remainder of this introduction are: 1)
monitoring components selected for replicated sampling, 2) selection of
the statistical model for design evaluation, and 3) the measure of performance
of the chosen design. Details of the methods and results of the power
analyses are given in later sections.
Components of the Monitoring Program Selected for Replicated Sampling
Replicated measurements of monitoring variables are needed for statistical
tests of spatial and temporal trends. Two kinds of error are associated
with the estimated magnitude of a monitoring variable: sampling error
A-l
-------
and analytical error. As pointed out by Mar et al. (1986) and others,
the choice of a replication scheme depends on the relative importance of
these errors. If analytical error is very large and sampling error is
very small, a single sample with replicate analyses may be appropriate.
If, on the other hand, the sampling error is very large and the analytical
error is very small, a single analysis of each of several replicate samples
may be appropriate. Sampling error may be minimized by stratified sampling
and composite sampling. These latter strategies have been incorporated
into the proposed monitoring designs whenever possible. Nevertheless,
sampling error is expected to be substantially larger than analytical error
for most monitoring variables.
Replicated sampling is specified for the following monitoring components:
• Secchi depth
t Benthic macroinvertebrate abundance
t Pathogen indicators in water
• Toxic chemicals in fish tissue
• Histopathological abnormalities in fish
t Fish species abundance
• Shellfish abundance
• Toxic chemicals in shellfish
• Paralytic shellfish poison (PSP) in shellfish
• Pathogen indicators in shellfish.
A-2
-------
These components are key elements of the monitoring program. Moreover,
some are expected to show substantial variation in time and space as a
result of natural environmental factors and human influence. For the other
components, which were not assigned replicated sampling designs, sampling
error can be minimized by stratified sampling, composite sampling, and
frequent observation (e.g., monthly for water quality variables). Analytical
replication of selected samples would be specified for most components
as part of a QA/QC program. For sediment toxicity bioassays, analytical
replication is specified for all samples.
General Statistical Model
A variety of approaches are available for trend detection (Box and
Jenkins 1976; Lettenmaier 1976; Sokal and Rohlf 1981; Montgomery and Reckhow
1984). The choice of a specific statistical model depends on objectives,
hypotheses, and characteristics of the data set. Because of the complexity
of the proposed monitoring program, a single statistical approach is not
appropriate for all data sets and all possible hypotheses. The approach
taken here is to postulate a generalized hypothesis based on the objectives
stated earlier (Table 2 in text). The generalized hypothesis selected
for the statistical design is the Null Hypothesis of no differences in
mean values of monitoring variable x among sampling times (or among sampling
stations). A statistical model is then selected to examine performance
of the monitoring design in relation to replication level for each component.
A two-way Analysis of Variance (ANOVA) is appropriate to test simulta-
neously for temporal and spatial effects, assuming the data set meets the
assumptions underlying the model (Green 1979; Sokal and Rohlf 1981). For
simplicity^ however, assume that either spatial or temporal trends are
analyzed individually. In this case, a one-way ANOVA fixed-effects model
(see below, "POWER ANALYSES," "Analytical Methods") is appropriate for
evaluating monitoring designs. This model has been applied extensively
to environmental monitoring and impact assessment (Green 1979; Sokal and
Rohlf 1981). For limitations of the ANOVA model as applied to monitoring
data refer to Glass et al. (1972) and Millard et al. (1985).
A-3
-------
In practice, more complex models may be used in future analyses of
monitoring data [e.g., two way or higher order ANOVA or nested designs
(Schmitt 1981), or paired station design (reference site vs. study site;
Bernstein and Zalinski 1983)]. For non-replicated monitoring variables
with extensive historical data (e.g., some conventional water quality variables)
formal time series analysis may be used to test for trends (Box and Jenkins
1976; Montgomery and Reckhow 1983). Power analysis of such models for
program design purposes was beyond the scope of this task.
Performance Evaluation of Particular Monitoring Designs
The measure used to evaluate the statistical sensitivity of the monitoring
design was the minimum detectable difference between two mean values.
To generalize the results of the power analysis, the minimum detectable
difference was expressed as a percentage of the grand mean among treatments.
The power of the test was fixed at 0.80. Type I error ( a) was fixed at
0.05. Minimum detectable difference was plotted versus number of replicate
samples for the following cases:
9 Number of stations (or sampling times) equal to 4, 6, 8,
and 16 stations (or times)
• Coefficient of Variation (across treatments) equal to 30,
50, 70, and 90 percent. The Coefficient of Variation is
equal to the standard deviation expressed as a percentage
of the mean.
Available data used to estimate the Coefficient of Variation are cited
in Appendix B under individual monitoring designs.
POWER ANALYSES
In statistical power analysis, relationships among the following study
design parameters are evaluated:
A-4
-------
o Power - Probability of detecting a real change
o Type I error (a) - Probability of not detecting a real change
o Minimum detectable difference - Magnitude of the smallest
change that can be detected for given power and Type I error
o Residual error - Natural variability
o Number of stations
o Number of replicate samples.
The analyses presented in this appendix were conducted with the objective of
providing guidance in selecting levels of sampling replication. This objective
was addressed by determining the magnitudes of difference among variables
that can be reliably detected with varying levels of sampling effort.
Analytical Methods
The results of a one-way ANOVA are usually summarized in a manner
similar to that shown in Table Al. The test statistic is the F ratio,
which is the ratio of the between-groups mean square (BMS) to the within-groups
mean square (WMS). As indicated in Table Al, the WMS is an unbiased estimate
of the population variance (5^), while the expected value of the BMS is
represented by the sum of the population variance and another term representing
the actual fixed effects. This added quantity is
(I-l)-1 *Ji( *i-*)2 (1)
where:
I = The number of sampling stations
Ji = The number of replicates at the i^n station
A-5
-------
TABLE Al. ANALYSIS OF VARIANCE TABLE FOR ONE-WAY LAYOUT
Source _ Sum of Squares d.f. Mean Square _ E(MS)
Between groups ZJ.(y.-y)2 1-1 SSB/(I-1) o2 + (I-l)~l
Within groups ZZ(y..-y.)2 n-I SSu/(n-I) o2
ij J
Total about ZZ(v..-y)2 n-1
grand mean ij ^
where:
y.. = observation at group (station) i and replicate j
y. = i group mean
y = overall mean of all i, j observations
n = total number of observations
SSg = between groups sum of squares
SSy = within groups sum of squares
E(MS) = expected mean square
I = number of groups
-------
*i = The true value of the ith effect
* = The mean of the treatment effects.
Under the null hypothesis, the value of the actual fixed effects term
is 0, and the F ratio is equal to 1. When fixed effects are observed in
the monitoring program, the value of this term increases and results in
an increase in the value of the numerator of the F ratio. Large effects
will result in an increase in the power of the test (i.e., the probability
of rejecting a false null hypothesis).
In performing power analyses, a set of effects is assumed and the
performance of the sampling design is evaluated as if these assumed effects
actually occurred. However, when a sample design involves several station
locations, many sets of effects can be assumed. For example, alternative
hypotheses can be constructed under which actual station effects of a certain
magnitude occur at one, two, three, or more of the total number of sampling
locations. The magnitude of the effects could also be varied among stations.
It can be seen that a very large number of alternative hypotheses can be
constructed for evaluation in power analyses.
The power analyses presented in this report were conducted in order
to provide a conservative evaluation of monitoring program performance.
Alternative hypotheses were constructed which assumed that the effects
occur in the combination that is most difficult to detect. Scheffe (1959)
showed that this conservative set of effects is defined by:
I *~ *l =A J * = ^ -1— J-L , for all k * i or j (2)
where:
A= is the maximum difference in actual effects
*l< = the true value of the kth effect.
A-6
-------
Equation (2) states that the two effects associated with the hypothesis
of interest differ by A while all other effects are equal to the mean of
these two. For the maximum difference in effects equal to A, this arrangement
gives the lowest test power.
The power analyses presented in this report were conducted on the
Ocean Data Evaluation System (ODES). The power analysis tool available
on ODES is described in a user-guidance document (Tetra Tech 1986k). Statis-
tical power analyses and methods of calculation are also described by Scheffe
(1959) and Cohen (1977).
Recent evidence has indicated that the ANOVA model is very robust
with respect to deviations from assumptions of normality and equal variances,
and is thus appropriate in environmental monitoring applications (Grieb
1984). However, nonparametric statistical methods, such as the Kruskal-Wallis
one-way analysis of variance by ranks, could also be used for the analysis
of monitoring data. While the statistical analysis results in this report
apply to the parametric ANOVA model, these results can also be used to
evaluate the corresponding performance of alternative, nonparametric statistical
methods by computing the power-efficiency of the nonparametric analog.
The power-efficiency of the nonparametric test provides a comparison of
the sample size required to achieve the same level of power associated
with the corresponding parametric tests. For example, the power-efficiency
of statistical Test B relative to Test A is given by:
NA
(100) percent
where:
Ng = The number of samples required in Test B to achieve the same
level of power obtained in Test A with a sample size of N/\.
A-7
-------
Calculation of the power-efficiency ratio for the Kruskal-Wal 1 is test is
described in Andrews (1954) and Lehmann (1975).
Results
The power analyses presented in Figures Al and A2 were conducted to
demonstrate the importance of the level of unexplained variability, represented
by the residual error variance design parameter, in determining the expected
performance of a monitoring program. Specifically, these analyses demonstrate
the effect of increased levels of unexplained variance on: 1) the ability
to detect a specified difference between stations and 2) the relative effect
of an increase in the numbers of stations on the minimum detection level.
These analyses were conducted for four levels of unexplained variability.
Coefficients of variation were set at 30, 50, 70, and 90. The number of
sampling stations was set at 4, 6, 8, and 16. All calculations were conducted
for fixed levels of power (0.8) and statistical significance (0.05).
Results of these analyses have general applicability, since the minimum
detectable difference is expressed as a percentage of the mean (grand mean
among treatment). The presentation of the minimum detectable difference
as a percent of the observed mean value, rather than as an absolute value
of the variable, provides a basis for comparing the results obtained for
different data sets. For example, this makes it possible to readily evaluate
the effect of increased sample variability, expressed as an increase in
the coefficient of variation, on the ability to detect statistically significant
differences among sampling locations or times. In general, the results
can be applied to any data set exhibiting a similar coefficient of variation.
However, as discussed below, it is important to evaluate individual monitoring
programs in terms of the value of the monitoring variable that can be detected
among sampling locations. Since the power curves are presented for coefficients
of variation representing a wide range of unexplained variability in the
sampling environment, these curves can be used to evaluate monitoring program
performance for any sampling designs utilizing 4 to 16 stations and for
sampling data exhibiting coefficients of variation between 30 and 90.
A-8
-------
LLJ
2
U_
O
CD
f
O
HI
LLJ
Q
550
500
450
400
350
300-
ULJ
O
HI
DC
HI
U_
5 25°-
HI
200-
150-
100-
50 -
COEFFICIENT
OF VARIATION
90
70
50
30
NUMBER OF STATIONS
4 6
4 6 8 10 12
NUMBER OF REPLICATES
14
16
Figure A1. Minimum detectable difference versus number of replicates
at selected levels of unexplained variance for 4 and 6 sta-
tions. Power of test=0.80, significance level=0,05.
-------
<
ULJ
U_
O
LU
O
Z
LU
CC
LU
LL
UJ
CD
fS
O
UJ
HI
O
550
500
450
400-
350-
300-
200-
150-
100-
50-
COEFFICIENT
OF VARIATION
90
70
50
30
NUMBER OF STATIONS
8 16
4 6 8 10 12
NUMBER OF REPLICATES
14
16
Figure A2. Minimum detectable difference versus number of replicates
at selected levels of unexplained variance for 8 and 16 sta-
tions. Power of test=0.80, significance level=0.05.
-------
These results show that as the level of unexplained variance increases
the minimum detectable difference between sampling stations increases.
For example, in Figure Al it can be seen that with five replicates at four
stations the minimum detectable difference between stations ranges from
approximately 70 percent of the mean (for a coefficient of variation of
30) to 212 percent of the mean (for a coefficient of variation of 90).
Correspondingly, both figures show that as the level of unexplained variance
increases, more sample replication is required to detect a specified level
of difference. For example, in a sample design with four sampling stations
(Figure Al), the number of replicate samples required to detect a difference
between stations equal to the mean is 3, 7, 12, and 15, respectively, for
coefficients of variation of 30, 50, 70, and 90.
Results of these analyses also demonstrate that for a fixed level
of sample variability, the minimum detectable difference between stations
increases as the number of stations increases. This increase is small,
however, compared to the effect of increased variability in the sampling
environment. For example, in Figure A2 for a coefficient of variation
equal to 30, the minimum difference detectable with five replicates is
approximately 80 and 90 percent of the mean for 8 and 16 stations, respec-
tively. In general, monitoring program performance, measured by the ability
to detect specified differences among stations, is increased for a fixed
level of sampling effort by the collection of more replicates at fewer
stations. However, the effect of number of stations on program performance
is small relative to that of the number of replicate samples.
In Figures Al and A2 the minimum detectable difference was expressed
in terms of a percentage of the overall mean. As indicated, this provides
a direct basis for the comparison of results between data sets, and the
results allow a quick evaluation of the expected performance of a large
number of study designs. However, in many monitoring programs, there may
be an interest not only in the relative change in variable values among
sampling locations, but also in the minimum value of the variable that
can be detected. In fact, results of power analyses used to evaluate individual
monitoring program design are generally expressed in terms of the measured
A-9
-------
units. Such presentations for all individual components are beyond the
scope of this task. (See discussions of specific components in the ambient
monitoring program sections, and in Appendix B).
A-10
-------
APPENDIX B
AGENCY-SPECIFIC USES OF MONITORING DATA
-------
AGENCY-SPECIFIC USES OF MONITORING DATA
Actual and potential uses of Puget Sound monitoring data by regulatory
and management agencies are given below. A list of persons contacted to
obtain information on uses of monitoring data is provided at the end of
this section.
STATE AGENCIES
Department of Ecology
• Use information from compliance and river monitoring programs
(especially contaminant loading data) to determine sources
and quantities of contaminants entering Puget Sound. Also
use those data for regulatory and management decisions regarding
dischargers, including wasteload allocations.
• Use data on water quality to determine compliance with or
violation of state water quality statutes.
• Use information on demographic and socioeconomic conditions,
and on decision record-keeping for land-use management and
siting decisions.
• Use data from river monitoring program to evaluate water
and sediment quality, and to determine compliance or violation
of state water quality standards.
B-l
-------
Department of Fisheries
t Use information on fisheries harvests (i.e., fish, shellfish,
algae) and fish and shellfish abundances to assess status
of population, and to support resource management decisions.
t Use data on water quality and habitat types for interpretation
of fish stock and harvest data.
• Use data on tissue contamination in fish and shellfish and
data on sediment chemistry to assess possible reproductive
impairment in harvested species.
Department of Game
• Use data on habitat types to evaluate species status, identify
land for possible aquisition, evaluate the importance of
remaining habitat for a given species, and design possible
mitigation procedures.
• Use information on waterfowl harvest, avian abundances,
and reproductive success to make management decisions regarding
hunting seasons and bag limits.
• Use information on toxic contaminants in biota (e.g., anadromous
fishes) to assess possible risks to wildlife populations
and humans.
• Use information on river hydrographic characteristics, contam-
inant loadings, and habitat types to assess status of habitats
for anadronous fishes and waterfowl.
B-2
-------
Department of Natural Resources
• Use information on shellfish abundances, pathogens in shellfish,
and PSP in shellfish as a basis for management decisions
regarding shellfish harvest areas and seasons.
t Use habitat data in support of decisions relating to shellfish
and kelp resource management.
• Use data on sediment chemistry- sediment toxicity (i.e.,
bioassays), hydrography, and biological conditions for imple-
mentation of PSDDA program.
• Use data on water quality, biological conditions, habitat
types, NPDES monitoring, contaminant loadings from rivers,
demographic and socioeconomic conditions, fisheries harvest,
waterfowl harvest, and aquaculture sites and yields to designate
and manage aquaculture sites.
Department of Social and Health Sciences
• Use data on pathogen indicators in water and pathogen indicators
and PSP and shellfish to determine compliance with existing
standards, and to support management decisions (e.g., shellfish
harvest closures, beach closures).
• Use data on toxic chemicals in fish and shellfish, and pathogens
and PSP shellfish to assess possible sources and causes
of problems, to determine possible long term spatial and
temporal trends in those problems, and to help locate and
eliminate sources and causes.
B-3
-------
Puget Sound Water Quality Authority
• Use all Puget Sound monitoring program data to assess the
state of the Sound, and to judge improvements due to implemen-
tation of the Puget Sound plan.
LOCAL AGENCIES, TRIBAL GOVERNMENTS, AND OTHER ENTITIES
City and County Health Departments
• Use contaminant loading data and hydrographic data from
rivers to assess pollutant inputs and to locate sources.
t Use data on sediment chemistry and water quality to interpret
compliance monitoring data collected by the agency.
• Use compliance monitoring data to supplement ongoing industrial
discharge monitoring, for the purpose of minimizing pollutant
inputs into Puget Sound.
• Use information on toxic chemicals in fish and shellfish
tissue to evaluate possible public health risks and to institute
closures.
• Use data on pathogens and PSP in water and shellfish to
determine compliance with existing standards, and in support
of management decisions (e.g., shellfish harvest closures,
beach closures).
Municipality of Metropolitan Seattle (Metro)
t Use data on water quality, sediment quality, and biological
conditions to supplement NPDES compliance monitoring, and
to facilitate interpretation of those data.
B-4
-------
t Use data on water quality, sediment quality, and biological
conditions to determine compliance with criteria for water
quality, sediment quality, and biological conditions, for
planning of treatment facilities, and for control of combined
sewer overflows.
Northwest Indian Fisheries Commission
0 Use habitat data and river monitoring data (i.e., T, S,
DO, TSS, contaminant loading) for assessment of salmon spawning
and rearing habitat.
• Use habitat data to assess problems with beach erosion or
alteration.
t Use shellfish abundance data to assess stock condition.
FEDERAL AGENCIES
U.S. Army Corps of Engineers
• Use information on habitat types, water quality (especially
hydrography) and sediment quality to assess potential dredge
spoil disposal sites.
• Use data on sediment quality, water quality, and biological
conditions to define reference conditions and evaluate possible
far-field impacts of dredge spoil disposal.
• Use information on habitat types to assess feasibility and
impacts of specific dredge and fill operations.
0 Use habitat and stock abundance data to assess possible
impacts of shoreline development activities.
B-5
-------
U.S. Environmental Protection Agency
• Use compliance monitoring data for compliance tracking,
and for evaluating issuance of new permits and reissuance
of old permits.
• Use compliance monitoring data to evaluate effectiveness
of discharger's pollution control program.
• Use information on water quality, sediment quality, and
biological conditions to assess impacts of point and non-
point pollutant sources.
• Use information on water quality, sediment quality, and
biological conditions, plus conclusions regarding impacts,
to support urban bay action programs.
U.S. Fish and Wildlife Service
• Use information on habitat types, marine mammals and birds
to assess habitat use, stock abundances, and migration patterns
• Use information on biological conditions and habitats as
data source for reviewing other agencies' proposals and
activities (e.g., COE, PSDDA, PSEP, PSWQA).
U.S. Geological Survey
• Supplement ongoing U.S.G.S. river hydrographic data collection
with Puget Sound river monitoring data (e.g., T, S, DO,
SS, contaminant loading).
• Use habitat data to assess changes in soil types, habitats,
erosional patterns, agricultural practices, and sedimentation.
B-6
-------
National Oceanic and Atmospheric Administration
t Use hydrographic and sediment quality data to supplement
data collected during intensive surveys and the status and
trends program, and to assist in the interpretation of those
data.
• Use data on sediment chemistry, toxic chemicals in fish
and shellfish, and on histopathological abnormalities in
fish to supplement and assist in the interpretation of mussel
watch data (part of the status and trends program).
• Use all monitoring program data as background information
for the design of intensive surveys.
t Use data on fisheries harvest, fish species abundances,
histopathological abnormalities in fish, and toxic chemicals
in fish tissue as a basis for assessing management decisions
regarding fisheries stocks and harvests, pollution, and
habitat alterations.
0 Use data on contaminant loading in rivers, sediment quality,
and toxic chemicals in fish tissue to define potential sources
of contaminants in fish populations.
• Use data on habitats to assess the condition of spawning
and rearing areas for oceanic fish species.
j>j,
• Use data on marine mammal abundances and reproductive success
to assess the health and status of populations.
t Use hydrographic data as inputs to hydrographic models of
Puget Sound
B-7
-------
Use river contaminant loading data and sediment quality
data to trace water mass movements and the dispersion of
pollutants. Link inferences on contaminant transport and
fate with information derived from hydrographic models.
CONTACTS
Following is a list of individuals who were contacted to obtain information
about agency uses of monitoring data.
Name
Dr. William Aaron
Dr. John Armstrong
Dr. Eddie Bernard
Mr. Russ Cahill
Mr. Chuck Dunn
Dr. Howard Harris
Mr. Dwaine Hogan
Mr. Dave Jamison
Ms. Carol Jolly
Mr. Charles Kleeburg
Mr. Ronald Kriezenbeck
Mr. Dennis McDonald
Mr. Kenneth Merry
Mr. Douglas Pierce
Mr. Richard Poelker
Mr. Earl Skinner
Affiliation
Northwest and Alaska Fisheries Center,
NOAA
U.S. Environmental Protection Agency
Pacific Marine Environmental Laboratory,
NOAA
Washington Department of Fisheries
U.S. Fish and Wildlife Service
Ocean Assessment Division, NOAA
U.S. Army Corps of Engineers
Washington Department of Natural Resources
Washington Department of Ecology
Seattle-King County Health Department
U.S. Environmental Protection Agency
Northwest Indian Fisheries Commission
Washington Department of Social and
Health Services
Tacoma-Pierce County Health Department
Washington Department of Game
U.S. Geological Survey
B-8
-------
Mr. Robert Stott U.S. Food and Drug Administration
Mr. Robert Swartz Municipality of Metropolitan Seattle
Mr. Jan Tvetan Washington Parks and Recreation Commission
Mr. John Underwood U.S. Environmental Protection Agency
B-9
-------
APPENDIX C
DETAILED DESIGN CONSIDERATIONS FOR THE AMBIENT MONITORING PROGRAM
-------
CONTENTS
Page
SEDIMENT QUALITY MONITORING DESIGNS C-l
SEDIMENT CHEMISTRY C-l
SEDIMENT TOXICITY BIOASSAYS C-8
CONVENTIONAL SEDIMENT VARIABLES C-ll
WATER QUALITY MONITORING DESIGNS C-14
HYDROGRAPHIC CONDITIONS C-14
DISSOLVED OXYGEN (DO) C-17
TURBIDITY/TRANSPARENCY C-19
ODORS, FLOATABLES, SLICKS, WATER COLOR C-20
NUTRIENT CONCENTRATIONS C-22
PHYTOPLANKTON STANDING STOCK C-24
PATHOGEN INDICATORS IN WATER C-26
BIOLOGICAL MONITORING DESIGNS C-29
BENTHIC MACROINVERTEBRATE ABUNDANCES C-29
TOXIC CHEMICALS IN FISH C-32
HISTOPATHOLOGICAL ABNORMALITIES IN FISH C-37
FISH SPECIES ABUNDANCES C-40
SHELLFISH ABUNDANCES C-42
TOXIC CHEMICALS IN SHELLFISH C-44
PSP IN SHELLFISH C-46
PATHOGEN INDICATORS IN SHELLFISH C-48
MARINE MAMMAL ABUNDANCES AND REPRODUCTIVE SUCCESS C-50
AVIAN ABUNDANCES AND REPRODUCTIVE SUCCESS C-53
RIVER MONITORING DESIGN C-57
HABITAT MONITORING DESIGN C-84
HABITAT TYPES C-84
COLLECTION OF ANCILLARY DATA C-87
CLIMATE/WEATHER C-87
FISHERIES HARVEST C-88
WATERFOWL HARVEST C-89
AQUACULTURE SITES AND YIELDS C-91
DEMOGRAPHIC AND SOCIOECONOMIC CONDITIONS C-92
DECISION RECORD-KEEPING C-94
-------
SEDIMENT QUALITY MONITORING DESIGNS
Each of the sediment quality components is to be monitored annually
at all stations. Emphasis is placed on the sediment quality components
in the ambient monitoring program because:
t They are excellent "integrators" of conditions in the receiving
environment
• Field and laboratory methods are well developed for the
variables proposed for monitoring within each component
t The combination of benthic macroinvertebrate monitoring
and sediment quality components forms the "Triad" of monitoring
variables recommended by Chapman and Long (1983).
Sediments are the sink for a mixture of materials, including organic and
inorganic particulate matter, and toxic metals and organic substances.
Organic and inorganic particulate matter determine the physical structure
of the sediments, and influence the biological communities associated with
the sediments. Organic particulate matter has the additional capacity
to enrich the sediments and thereby impact the biota. Many toxic metals
and organic substances released into the environment adsorb to particulate
matter that accumulates as sediments in Puget Sound. Potential effects
of these substances on biota depend in part on interactions governed by
the specific composition and structure of the sediments.
SEDIMENT CHEMISTRY
Variables and Rationale—Samples for sediment chemistry will be collected
from the upper 2 cm of sediment. Variables to be monitored will include
selected U.S. EPA priority pollutant metals and selected U.S. EPA priority
C-l
-------
pollutant organic compounds (Table Cl). Concentrations of additional compounds
of concern in Puget Sound will also be determined (Table Cl; also see below,
Data to be Reported). Miscellaneous organic acids and volatile organic
compounds will be analyzed for as part of intensive surveys or ambient
monitoring only where a problem is expected. Periodic spot checks are
also recommended. Costs of such analyses have not been included in cost
estimates for the ambient monitoring program.
Most of the toxic chemicals of concern in Table Cl have been demonstrated
to accumulate to relatively high concentrations (on a mass basis) in marine
and estuarine sediments, compared with the water column. These chemical
data will provide information that can be used to:
t Assess potential for toxicity to resident biota
• Identify areas of Puget Sound that have been, or are, accumu-
lating substantial amounts of toxic chemicals
• Evaluate temporal changes in the historical record of toxic
chemicals accumulating in sediments partly as a result of
anthropogenic discharges
t Interpret biological and sediment toxicity bioassay data.
A 0.06-m2 box corer or a 0.1-m2 van Veen grab will be used for sediment
sampling (including sediment toxicity bioassays, conventional sediment
variables, and benthic macroinvertebrate abundances). Each sampling device
has advantages and disadvantages. Although a box corer takes a deeper,
and possibly less disturbed, sample than does a van Veen grab, the former
is more difficult and more expensive to use. An evaluation of benthic
sampling equipment for use in the Puget Sound Monitoring Program is in
progress.
C-2
-------
TABLE Cl. LIST OF TARGET CHEMICALS FOR SEDIMENT ANALYSIS
Metals and Metalloids
Antimony Lead Zinc
Arsenic Mercury Aluminum3
Cadmium Nickel Iron3
Chromium Silver Manganese3
Copper
Phenols (organic acids)
phenol
2-methylphenol
4-methylphenol
2,4-dimethylphenol
Substituted Phenols (organic acids)
2-chlorophenol 2,4,5-trichlorophenol
2,4-di chlorophenol pentachlorophenol
4-chloro-3-methylphenol 2-ni trophenol
2,4,6-trichlorophenol 2,4-di ntrophenol
4,6-dinitro-o-cresol
Miscellaneous Organic Acids (selected samples only)
2-methoxyphenolb
3,4,5-trichloroguaiacolb
4,5,6-trichloroguaiacolb
tetrachloroguai acolb
mono- and di- chlorodehydroabietic acidsb
Low Molecular Weight Aromatic Hydrocarbons (neutrals)
naphthalene fluorene
acenaphthylene phenanthrene
acehaphthene anthracene
High Molecular Weight PAH (neutrals)
fluoranthene benzo(k)fluoranthene
pyrene benzo(a)pyrene
benzo(a)anthracene indeno(l,2,3-c,d)pyrene
chrysene dibenzo(a,h)anthracene
benzo(b)fluranthene benzo(g,h,i jperylene
-------
TABLE Cl. (Continued)
Chlorinated Aromatic Hydrocarbons (neutrals)
1,3-dichlorobenzene
1,4-di chlorobenzene
1,2-di chlorobenzene
1,2,4-trichlorobenzene
2-chloronaphthalene
hexachlorobenzene (HCB)
Total PCBs (mono- through decachlorobiphenyls)
Chlorinated Aliphatic Hydrocarbons (neutrals)
hexachlorobutadi ene
hexachloroethene
trichlorobutadiene isomersc
tetrachlorobutadiene isomersc
pentachlorobutadiene isomers0
Phthalate Esters (neutrals)
dimethyl phthalate
diethyl phthalate
di-n-butyl phthalate
butyl benzyl phthalate
bis(2-ethylhexyl)phthalate
di-n-octyl phthalate
Miscellaneous oxygenated compounds (neutrals)
isophorone
benzyl alcohol
benzoic acid
dibenzofuran
polychlorodi benzofurans^
polychlorodibenzodioxinsd
coprostanol3
Organonitrogen Compounds (baes and neutrals)
N-ni trosodi phenylami ne
9(H)-carbazole
Pesticides (neutrals)
p,p'-DDE
p,p'-DDD
p,p'-DDT
aldrin
dieldrin
alpha-chlordane
alpha-endosulfane
beta-endosulfane
endosulfan sulfatee
endrin
endrin aldehyde6
heptachlor
heptachlor epoxide6
alpha-HCH
beta-HCH
delta-HCH
gamma-HCH (lindane)
-------
TABLE Cl. (Continued)
Volatile Halogenated Alkanes (neutrals)
chloromethane carbon tetrachloride6
bromomethane bromodichloromethane6
chloroethane6 1,2-di chloropropane
di chloromethane chlorodi bromomethane6
l,l'-dichloroethane 1,1,1-trichloroethane
c h1o ro fo rm 'b romo fo rme
1,2-di chloroethane6 1,1,2,2-tetrachloroethane6
1,1,1-trichloroethane6
Volatile Halogenated Alkenes (neutrals)
vinyl chloride cis-l,3-dichloropropene
1,1'-di chloroethene trans-1,3-di chloropropene
trans-1,2-di chloroethene tri chloroethene
tetrachloroethene
Volatile Aromatic and Chlorinated Aromatic Hydrocarbons (neutrals)
benzene styrene (ethenylbenzene)
toluene total xylenes
ethyl benzene chlorobenzene
a Not of concern as pollutants, but to be analyzed as ancillary variables
used in interpretation of data.
b Recommmended for analysis only near pulp mill facilities (chlorinated
guaiacols are only of concern near kraft mills)
c Recommended for analysis only where chlorinated butadienes are suspected
to have a major source.
d Chlorinated dibenzofurans and dioxins are recommended as special analyses
only, as determined by specific project goals.
6 Compound is seldom or not reported, but can be easily analysed for with
other recommended analytes.
-------
Stations—106 stations total (Figures 4-6 and Tables 4-6 in text):
—31 in basins
—34 in urban/industrial bays
—41 in rural bays
Efforts were made during the design of the program to locate stations
and transects in areas where the monitoring information would be of greatest
utility. The criteria for selecting station locations were:
• Most stations should be located close enough to shore to
assess integrated effects of multiple sources of contamination
before an entire bay or basin is affected.
• Some stations should be located at the centers of the main basins
to assess cumulative, long-term effects on entire basins.
• Stations should not be located adjacent to major anthropogenic
sources of contaminants. Site-specific monitoring of the
receiving environments near contaminant sources will be
covered by the compliance monitoring programs.
• Most stations should be located at relatively shallow depths
(<30 m) to be in biologically productive areas.
• Whenever possible, stations and transects should be located
in areas that have been sampled during earlier surveys,
thereby facilitating comparisons with historical data.
Additional criteria should be applied when specific stations are established
during field implementation of the program:
t Stations in rural bays should be located near the mouths
of tributaries whenever possible, to detect contamination
and effects asociated with nonpoint sources (e.g., bacteria,
pesticides)
C-3
-------
• Stations in urban bays should be located down-current from
major sources of contaminants (e.g., major industrialized
areas in inner harbors or industrial waterways such as the
Duwamish River)
t Stations should be stratified by sediment characteristics
(e.g., grain size) to allow comparisons of benthic infauna
among areas
• Stations should be located in depositional areas.
Three types of stations are proposed for the collection of integrated
water quality and sediment quality data. They are stratified primarily
by depth, and include:
• Stations located at the centers of the Puget Sound basins
t Stations located at 20-m depth
t Stations located at water depths as close to 20 m as possible
(water depth in some bays does not reach 20 m).
Stations at the centers of the basins are furthest from anthropogenic dis-
turbances. However, Nichols (1985) has shown that major changes in benthic
macroi nvertebrate communities can occur at such stations over relatively
short periods of time (about 5-10 yr). The 20-m depth was chosen because:
• Monitoring of productive nearshore areas is a high priority,
and total abundances of benthic organisms are often greatest
at approximately this depth (Word et al. 1984)
t Stations will be close enough to shore to exhibit any integrated
effects of contaminant sources on shore
C-4
-------
• Depth and distance offshore are usually sufficient for a
given station not to be unduly influenced by any single
point source of contaminants.
It will not always be possible to sample at 20-m depth because some bays
are shallow. However, adherence to the 20-m contour whenever possible
will facilitate among-station comparisons where substrates are comparable.
The transects west of Port Angeles arid in the Strait of Georgia are
located to provide information on conditions at the boundaries of the study
area. The transect just northwest of Admiralty Inlet will provide valuable
hydrographic information for the interpretation of data from Hood Canal,
the main basin, and South Sound. In combination with the transects near
Port Angeles and in the Strait of Georgia it will also provide information
on bottom conditions in the more oceanic portion of the study area. The
transesct in Saratoga passage is recommended to provide information on
hydrographic, chemical, and biological conditions in the Whidbey Basin,
which is a major channel for mass transport. The Saratoga Passage is also
influenced by freshwater inflows from the Skagit River. The transect in
Posession Sound is designed to assess hydrographic interactions between
Whidbey Basin and the main basin.
The main basin of Puget Sound was assigned more stations than any
other basin (nine stations arranged along three transects) because of its
size and the intensity of human use that occurs there. Monitoring at these
three transects will provide information that is adequate to characterize
water quality, sediment quality, and biological conditions in this intensively
used portion of Puget Sound. From north to south, these three transects
in the main basin correspond to: 1) the area sampled annually by Nichols
(1985); 2) a transect between Elliott Bay and Bainbridge Island to assess
chemical and biological conditions in an area potentially influenced by
outflows from Elliott Bay and Sinclair Inlet; 3) Seahurst Transect J.5
sampled as part of Metro Seahurst studies (Word et al. 1984).
C-5
-------
The transect in the Nisqually Reach portion of South Sound is the
only transect in that basin. Its location corresponds with that of a water
quality station that is part of the ongoing Ecology ambient monitoring
program. Similarly, the southern transect in Hood Canal corresponds to
an area now sampled by Ecology. It is needed to assess conditions in lower
Hood Canal. The northern transect in Hood Canal is located just north
of the military base at Bangor for the purpose of detecting possible impacts
of military activities.
Frequency—Annual.
Initially^ annual sampling will provide important baseline data (and
possibly trend data) for many areas of Puget Sound. Sediment chemistry
data will also be needed to interpret data from bioassays and benthic macro-
invertebrate analyses. Nichols (1985) has shown that major changes in
benthic biological conditions in the main basin of Puget Sound occurred
over a 5-10 yr period. Thus, annual or biennial data will probably be
required to track benthic conditions. The initial design incorporating
annual sampling should be re-evaluated after 3-5 yr of data have been collected.
Timing—Spring (March).
Sampling of all benthic variables will be conducted during March to
coincide with a time of relative stability of benthic macroinvertebrate
assemblages.
Data to be Reported—
• U.S. EPA Priority pollutant metals concentrations as mg/kg
to a maximum of three significant figures on a blank-corrected,
dry weight basis (Table Cl)
• U.S. EPA Priority pollutant organic compound concentrations
as ug/kg to a maximum of two significant figures on a blank-
corrected, dry weight basis (Table Cl)
C-6
-------
t Additional pollutants of concern listed in Table Cl
• Station location, date, time, and water depth
• All field and laboratory QA/QC procedures and results.
Data Analyses—
• Comparisons of temporal changes in the spatial distributions
of pollutant concentrations
• Comparisons, by station, of temporal trends in pollutant
concentrations
• Comparisons of patterns in data on pollutant concentrations
with patterns in data on conventional sediment variables,
benthic macroinvertebrate and sediment toxicity bioassays
0 Station location, date, time, and water depth
t Sediment horizon
t All field and laboratory QA/QC procedures and results.
Replication and Statistical Sensitivity—Replicate samples will not
be collected (except for QA/QC analyses), thereby precluding statistical
analyses among individual stations within a survey- Replicate data on
the chemical composition of sediments within the Commencement Bay waterways
indicated that coefficients of variation for several groups of organic
chemicals ranged from 17-61 percent (Tetra Tech 1985a). Given a coefficient
of variation of 30 percent and three to four replicates (in space or time),
the minimum detectable difference in mean chemical concentration among
stations would be equal to about 100 percent of the overall mean among
stations (see Figures Al and A2 in Appendix A). Replicated sampling at
C-7
-------
all stations was precluded because of the high cost of laboratory analysis
and has not been recommended by PSEP sampling and analysis protocols.
Nevertheless, statistical analyses may be performed for related groups
(clusters) of stations within a survey, or for selected stations over time.
Protocols--
0 Field and Laboratory References: Tetra Tech (1986c,f,g)
t Supporting literature: U.S. EPA (1979), Plumb (1981), U.S. EPA
(1982).
SEDIMENT TOXICITY BIOASSAYS
Variables and Rationale—As recommended by Chapman and Long (1983),
separate toxicity tests should be conducted to assess acute lethality,
sublethal effects, and genotoxicity. Moreover, a range of methods and
test organisms should be used. The methods recommended are all detailed
in the Puget Sound protocols and are: the amphipod, Rhepoxynius abronius,
acute lethality (survival and emergence) test; the bivalve larvae (survival
and abnormalities) test or the Microtox (bacterial luminescence) test;
and the anaphase aberration (mitotic abnormalities) test. All three tests
will be conducted on subsamples of a single composite sample that is also
used for chemistry analyses.
The previous section on Sediment Chemistry should be consulted for
the rationale for locations of stations and the frequency and timing of
sampling.
Stations—106 stations total (Figures 4-6 and Tables 4-6 in text):
—31 in basins
—34 in urban/industrial bays
—41 in rural bays.
Frequency—Annual.
C-8
-------
Timing—Spring (March).
Data to be Reported—Data to be reported for all sediment toxicity
bioassay tests include:
• Station location, date, time, and water depth
• All field and laboratory QA/QC procedures and results.
Data to be reported for the amphipod acute lethality test include:
• Water quality measurements during testing (i.e., dissolved
oxygen, temperature, salinity, pH)
• Daily emergence for each beaker and the 10-day mean and
standard deviation for each treatment
• Interstitial salinity values of test sediments
• 96-h LC50 values with reference toxicants
• Any problems that may have influenced data quality.
Data to be reported for the bivalve larvae test include:
0 Water quality measurements at the beginning and end of testing
(e.g., dissolved oxygen, temperature, salinity, pH)
• Individual replicate and mean and standard deviation data
for larval survival after 48 h
0 Individual replicate and mean and standard deviation data
for larval abnormalities after 48 h
C-9
-------
• 48-h LCrr> and EC<-n values with reference toxicants
bU =>U
• Data on larval presence in the sediment
• Any problems that may have influenced data quality.
Data to be reported for the Microtox test include:
• Percent decrease in luminescence for each concentrations
of supernatant (e.g., saline sediment extract) tested
• Determination of a significant dose-response relationship
by least-squares regression of percent decrease in luminescence
on the logarithm of sample dilution
• Determination of EC5Q values and 95-percent confidence limits
for the reference toxicant.
Data to be reported for the anaphase aberration test include:
• Initial screening data for the determination of extract
concentrations
• Individual replicate and mean and standard deviation data
for numbers of normal and abnormal anaphases observed
t Types of anaphase aberrations observed
0 Frequency of anaphase aberrations observed with the positive
control
0 Any problems that may have influenced data quality.
C-10
-------
Data Analyses—
• Comparisons, by site and by area, of temporal trends in
sediment toxicity
t Comparisons through time of spatial differences in sediment
toxicity
t Identification of problem areas
t Comparison of sediment chemistry, infauna, and bioassay
data.
Replication and Statistical Sensitivity—Five laboratory replicates
are recommended in the Puget Sound protocols (Tetra Tech and E.V.S. Con-
sultants 1986a), and are required to achieve an acceptable level of statistical
sensitivity using any of the aforementioned sediment toxicity bioassay
tests. For the amphipod bioassay test using 20 organisms per replicate,
five replicates are capable of detecting a difference between two survival
means of 2.8 amphipods (Figure Cl). This is about a 15 percent reduction
in survival, and is considered adequate for most applications (Swartz et
al. 1985).
Protocols—
• Field and Laboratory Reference: Tetra Tech (1986c), Tetra
Tech and E.V.S. Consultants (1986a).
• Supporting literature: Chapman and Long (1983).
CONVENTIONAL SEDIMENT VARIABLES
Variables and Rationale—Particle size distribution, total organic
carbon, and sulfides will be determined on bulk sediment samples. The
salinity of pore water will also be measured. These variables are needed
C-ll
-------
10 -,
CO
IU 8
oc
en 6
o
I
i
Ul
o
4-
Ul
oc
u.
5
2 -
i
2
r
4
3456
NUMBER OF REPLICATES
~
7
I
10
NOTE: SIGNIFICANCE LEVEL - 0.05
STATISTICAL POWER - 0.75
Reference: Swartz et aJ. (1985)
Figure C1. Statistical sensitivity of the amphipod bioassay test as a
function of the number of replicates.
-------
to interpret data on the distributions and abundances of sediment-associated
biota. Moreover, many chemicals tend to be sorbed to higher concentrations
on finer-grained materials, so that the concentrations of many chemicals
in the sediments are high correlated with the proportions of silt- and
clay-sized particulates. Particle size distribution and total organic carbon
also provide a qualitative indication of hydrographic conditions at a given
site. Samples for conventional sediment variables will be subsamples of
the composite sample used for analyses of priority pollutant concentrations
and bioassays.
The previous section on Sediment Chemistry should be consulted for
the rationale for locations of stations and the frequency and timing of
sampling.
Stations—106 stations total (Figures 4-6 and Tables 4-6 in text):
—31 in basins
—34 in urban/industrial bays
—41 in rural bays.
Frequency—Annual.
Timing—Spring (March).
Data to be Reported—
• Proportions of grain size categories by phi sizes (as percents)
• Percent total organic carbon and percent total nitrogen,
each by dry weight, to nearest 0.01 percent
• Free sulfide (i.e., water-soluble) concentration by dry
weight, to two significant digits
• Salinity of pore water
C-12
-------
• Station location, date, time, and water depth
• All field and laboratory QA/QC procedures and results.
Data Analyses—
• Maps of sediment texture and total organic carbon
t Grain size distributions summarized on Shepard diagrams
(Shepard 1954)
• Mean grain size plus plots of skewness and kurtosis
• Comparisons of grain size and total organic carbon data
with macroinvertebrate, sediment chemistry, and sediment
toxicity bioassay data.
Replication and Statistical Sensitivity—Single estimates of grain
size distribution and total organic carbon will be collected for each station.
These data are intended to provide "snapshots" of conditions in the sound
and to provide data on how these conditions vary through time. Their primary
purpose, however, is to provide information necessary for the interpretation
of other data on sediment quality. For this reason, routine statistical
characterizations of sampling error are not recommended.
Protocols—
0 Field and Laboratory Reference: Tetra Tech (1986c,e)
0 Supporting literature: Buchanan (1984), Folk (1968), Krumbein
and Pettijohn (1938), Plumb (1981), and U.S. EPA (1983).
C-13
-------
WATER QUALITY MONITORING DESIGNS
Water quality components will be monitored monthly at a subset of
the basic array stations used for sampling of sediment quality components.
That subset includes the center station from each basin transect, plus
one station from each urban/industrialized-and rural bay where sediment
quality samples are collected.
Water quality components are proposed for inclusion in this monitoring
program for two reasons:
• Knowledge of water quality conditions throughout Puget Sound
is essential to meet all three program goals (Table 2 in
text)
• Water quality information is needed to interpret spatial
patterns and temporal trends in many other monitoring com-
ponents.
All water quality samples will be collected concurrently at each station.
Due to logistical constraints, sampling at the 53 water quality stations
will likely occur over 2 wk or more each month. However, sampling at each
station will be performed on approximately the same date each month to
ensure an approximately 4-wk interval between each sampling event. All
samples will be collected during daylight hours.
HYDROGRAPHIC CONDITIONS
Variables and Rationale—Temperature and salinity will be measured
because they define habitat characteristics that are essential to Puget
Sound biota. The normal seasonal cycles of these variables are important
for determining biological growth cycles in the sound. In addition, unusual
C-14
-------
changes in temperature and salinity in response to abnormal climatic events,
such as drought or flooding, may cause substantial alteration of many biological
communities.
Because of recent advances in instrumentation, particularly automated
data logging, it is recommended that the temperature and salinity data
be collected with a high-quality conductivity-temperature-depth (CTD) sensor
system, similar to that Metro has developed for their monitoring purposes.
Such a system provides continuous profiles of good quality data. Continuous
profiles are especially important for determining the internal structure
of the water column. For example, depths of significant pycnoclines can
be defined clearly.
Stations—53 stations total (Figures 4-6 and Tables 4-6 in text):
—11 in basins
—15 in urban/industrial bays
—27 in rural bays.
Sampling stations are located to coincide with those used for the
sediment quality triad and bottomfish sampling. A single station in the
center of each basin transect and a single station in each bay will provide
the minimal adequate data to characterize large-scale spatial and temporal
trends. Consult the previous section on Sediment Chemistry for the rationale
for transect locations.
Frequency—Monthly.
Because water quality varies greatly through time, monthly sampling
is the minimum frequency that will yield interpretable data. Sampling
should be conducted during all 12 mo to ensure that all major hydrographic
trends are observed, and to provide a complete data set for analysis of
temporal trends (including possibly time series analysis).
C-15
-------
Timing—Entire year, in conjunction with other water quality variables.
Year-round sampling is needed to characterize seasonal variations
and provide support data for other monitoring components.
Data to be Reported—
• Temperature and salinity with depth
• All field and laboratory QA/QC procedures and results, including
database codes (i.e., alphanumeric surrogates suitable for
inclusion with the data in a computerized database)
t Station location, date, time, and water depth.
Data Analyses—
• Temporal variations from plots of monthly values
t Variations from long-term average trends
• Temporal trends as a function of climatic or other variables
• Spatial relationships
• Areas of anomalous behavior/standards violation
• Average annual flushing and refluxing for major basins.
Replication and Statistical Sensitivity—Single vertical profiles
are to be collected at each station. These measurements are intended to
provide "snapshots" of conditions in the sound and to provide data on how
these conditions vary through time. Statistical definitions of sampling
error are unnecessary.
C-16
-------
Protocols--
• Field and Laboratory Reference: Metro (in preparation).
DISSOLVED OXYGEN (DO)
Variables and Rationale—Dissolved oxygen is a critical variable for
characterizing Puget Sound marine habitats. An adequate DO concentration
is essential for all higher life forms. DO levels in the sound are largely
controlled by natural processes, and show strong seasonal changes in response
to varying concentrations of DO in incoming ocean water and to production
and decay processes in the sound. DO levels are also sensitive to anthropogenic
perturbations, and can change in response to direct or indirect loadings
of nutrients and BOD. DO can be used as a surrogate for estimating overall
impacts of conventional pollutants.
Given recent improvements in instrumentation, continuous profiles
using a DO probe with an automated datalogger will be collected (i.e.,
instrumentation will be similar to that now used by Metro). This DO probe
can be included with the CTD package used for the temperature and salinity
measurements.
The previous sections on Hydrographic Conditions and Sediment Chemistry
should be consulted for the rationale for station locations, frequency,
and timing given below.
Stations—53 stations total (Figures 4-6 and Tables 4-6 in text):
—11 in basins .
—15 in urban/industrial bays
—27 in rural bays.
Frequency—Monthly.
Timing—Entire year, in conjunction with other water quality variables.
C-17
-------
Data to be Reported--
t Dissolved oxygen concentrations with depth
• All field and laboratory QA/QC procedures and results, including
database codes
• Station location, date, time, and water depth.
Data Analyses—
• Temporal trends of DO concentrations on an annual basis
t Variations from long-term average trends
0 Spatial differences among stations
• Areas or regions showing significant DO depletion including
standard violations
• Multi-year trends by region.
Replication and Statistical Sensitivity—Single DO profiles will be
collected at each station, thereby precluding statistical analyses among
individual stations within a survey. DO concentrations respond to a number
of physical-chemical variables. Hence, the natural range and short-term
variations in DO concentrations tend to be large. Between-month changes
at any one site and depth are generally much smaller and have a standard
deviation on the order of +20 percent of the monthly mean DO concentration
(Ebbesmeyer et al. 1982). At present, eutrophication and other factors
that would be reflected in DO changes are major problems only in very restricted
areas of the sound. Given the monthly determination of vertical profiles
recommended herein, replicated measurements are unnecessary. If oxygen-
C-18
-------
depletion problems are indicated at one or more sites, then twice weekly
or more frequent sampling in those areas could be performed to provide
more detailed data.
Protocols—
• Field and Laboratory Reference: Metro (in preparation).
TURBIDITY/TRANSPARENCY
Variables and Rationale—Secchi depths will be determined at each
station. Secchi depth data are inexpensive to collect and act as a surrogate
for turbidity measurements. Direct monitoring of turbidity using an electronic
meter is not recommended because high measurement error makes detection
of trends difficult (Lettenmaier et al. 1982).
Turbidity (i.e., particulates including plankton) affects the ecosystem
directly by decreasing light penetration and hence plant productivity.
Excess turbidity can also present visual/aesthetic problems. Most turbidity
in the sound is due to natural causes, particularly river discharges, and
is not considered to be a major water quality problem (Baker 1982; Dexter
et al. 1981). Human disturbance can increase turbidity by increasing upland
and shoreline erosion.
The previous sections on Hydrographic Conditions and Sediment Chemistry
should be consulted for the rationale for station locations, frequency,
and timing given below.
Stations—SB stations total (Figures 4-6 and Tables 4-6 in text):
—11 in basins
—15 in urban/industrial bays
—27 in rural bays.
Frequency—Monthly.
C-19
-------
Timing—Entire year, in conjunction with other water quality variables.
Data to be Reported--
• Secchi depth
t Station location, date, time, and water depth
• Cloud cover
• All field and laboratory QA/QC procedures.
Data Analyses—
• Temporal and spatial trends on a monthly basis
t Major deviations from long-term average conditions.
Replication and Statistical Sensitivity—Relatively large natural
variations in turbidity occur, especially over a seasonal cycle. For this
reason, five or more measurements of Secchi depth should be made and the
results should be averaged. This measurement depends greatly on the person
taking the reading. As a result, it is recommended that all Secchi depths
be taken by the same person, or a group of persons who have worked together
to achieve consistent readings.
Protocols—
• Field and Laboratory Reference: Tetra Tech (1986c).
ODORS, FLOATABLES, SLICKS, WATER COLOR—
Variables and Rationale—Unsightly water conditions and objectionable
odors can develop from the presence of floatable materials, slicks, discolored
water, and excessive turbidity. These conditions can impair the aesthetic
C-20
-------
qualities of the sound and the use of the sound for recreation. Notes
on these conditions will be taken as part of the routine water quality
survey and the shellfish surveys.
At present, citizens may report oil spills, floatables, or other aesthetic
problems in Puget Sound or its watersheds to the U.S. Coast Guard (Puget
Sound only) or the Washington Department of Ecology. These reports should
also be accessed quarterly for information relative to aesthetic conditions
in Puget Sound or its watersheds. The Co'ast Guard data system should,
however, be upgraded. At present spills and other problems are entered
into a computer file that records location only to the nearest minute of
latitude and longitude. More precise locations (i.e., to the second or
less) are highly desirable for anticipating and documenting the impacts
of such events on Puget Sound resources.
Stations—53 water quality stations (Figures 4-6, Tables 4-6 in text)
—25 intertidal shellfish stations (discussed below).
Frequency—Weekly during May-July at shellfish stations (discussed
below); monthly at water quality stations.
Timi ng — Entire year, in conjunction with other water quality variables
and with shellfish monitoring.
Data to be Reported—
• Location affected
• Type of incident (e.g., floatables, slick, malodorous condition,
etc.)
t Size of affected area
• Severity of incident.
C-21
-------
It is recommended that a public notification system be maintained (possibly
within an existing agency) to accept, record, and ensure appropriate responses
to objectionable events.
Data Analyses—
• Annual compilation of numbers, locations, and severity of
different types of incidences
• Evaluation of temporal trends by region
• Evaluation of spatial differences.
Replication and Statistical Sensitivity—The data that will be collected
will consist of subjective observations of sporadic occurrences. They
will not be replicated or quantified, thereby precluding rigorous statistical
analyses. Consistent observations should be obtained by limiting observation
responsibilities to a few persons who have worked together to achieve similar
value systems.
Protocol--
• Field and Laboratory Reference: None. (Separate blanks
will be provided on the field log sheets for the purpose
of recording information on odors, floatables, slicks, and
watercolor.)
NUTRIENT CONCENTRATIONS
Variables and Rationale—Concentrations of the following nutrients
in the water column will be determined: NO.,, N02, NH,, total nitrogen,
C-22
-------
PCL, total phosphorus, and SiCh. Two samples will be used for determination
of nutrient concentrations.
t Composite sample of the 0-30-m layer, which will also be
used for chlorophyll a determinations. Where the euphotic
zone is less than 30 m deep, as determined by a photometer,
a composite sample of the only euphotic layer will be taken.
• Grab sample taken 1 m off the bottom.
Nutrients are essential growth factors for attached algae and phyto-
plankton. Nutrient enrichment can constitute a water quality problem,
but only when: 1) it stimulates such excessive plant growth that the resulting
decay causes DO depletion, or 2) when changes in absolute or relative nutrient
availability shift the composition of the plant community to non-edible
or noxious species. In Puget Sound, nutrients apparently do not play a
major role in controlling plant growth except in a few poorly flushed embay-
ments, such as Budd Inlet (Collias and Lincoln 1977). Nutrient enrichment
may have had some role in the apparent increased incidence of PSP organisms
in the sound (Saunders et al . 1982), but this relationship has not been
conclusively established. The need for direct nutrient monitoring in the
sound appears to be limited at present, but continued monitoring of this
parameter is recommended to provide a direct measure of overall anthropogenic
nutrient influences.
The rationale for station locations, frequency, and timing is given
in the previous sections on Hydrographic/Conditions and Sediment Chemistry.
Stations—53 stations total (Figures 4-6 and Tables 4-6 in text):
—11 in basins
—15 in urban/industrial bays
—27 in rural bays.
Frequency—Monthly.
C-23
-------
Timing—Entire year, in conjunction with other water quality variables.
Data to be Reported—
• Nutrient concentrations at surface layer (0-30 m) and 1 m
off bottom
t Station location, date, time, and water depth
• All field and laboratory QA/QC procedures and results.
Data Analyses—
t Relationships between phytoplankton standing stock and nutrient
concentrations in 0-30-m layer
• Long-term trends analysis comparing monthly means for sampled
months.
Replication and Statistical Sensitivity—Replicated sampling is not
recommended (with the exception of routine QA/QC replication), thereby
precluding statistical comparisons among individual stations within a survey.
Examination of long-term trends will be based on qualitative analysis of
graphic data, supported by time-series analysis.
Protocols--
• Field and Laboratory Reference: Tetra Tech (1986c), Strickland
and Parsons (1972).
PHYTOPLANKTON STANDING STOCK
Variables and Rationale—The concentration of chlorophyll a in surface
water will be measured because it provides a convenient, quantitative measure
of phytoplankton standing stock. As primary producers, phytoplankton are
C-24
-------
critical to ecosystem function. Under conditions of nutrient enrichment
in enclosed bays, phytoplankton blooms may cause depresions of dissolved
oxygen leading to fish kills.
Total chlorophyll a will be determined on the same composite sample
analyzed for nutrient concentrations. Total chlorophyll a is to be measured
by spectrophotometric methods. The rationale for station locations, frequency,
and timing is given in the previous section on Hydrographic/Conditions
and Sediment Chemistry.
Stations—53 stations total (Figures 4-6 and Tables 4-6 in text):
—11 in basins
—15 in urban/industrial bays
—27 in rural bays.
Frequency—Monthly.
Timing—Entire year, in conjunction with other water quality variables.
Data to be Reported—
• Chlorophyll a as mg/m seawater to the nearest 0.01 unit
• Station location, water depth, euphhotic depth, date, and
time
• All field and laboratory QA/QC procedures and results.
Data Analyses—
0 Comparisons of chlorophyll a concentrations among sites
for each survey, and within sites through time
• Comparisons of chlorophyll a concentrations with other water
quality variables to determine possible interrelationships.
C-25
-------
Replication and Statistical Sensitivity—Phytoplankton populations
are highly variable in space and time, and often require very frequent
sampling to permit statistical resolution of differences in population
abundances among areas. Replicate samples will not be collected, thereby
precluding statistical analysis among individual stations within a survey.
However, statistical time-series analyses may be performed for one or more
stations.
Protocols--
t Field and Laboratory Reference: Tetra Tech (1986c), Strickland
and Parsons (1972).
• Supporting literature: Stofan and Grant (1978), American
Public Health Association (1985), Tetra Tech (1986c), and
Conover et al. (1986).
PATHOGEN INDICATORS IN WATER
Variables and Rationale—Pathogen indicators in water will be measured
at the same time that they are measured in shellfish. At present, only
analyses for pathogen surrogates (i.e., fecal coliform bacteria; enterococci)
are recommended. Further analyses for pathogenic viruses (e.g., Vi brio
spp.) may be appropriate as part of future intensive surveys in response
to specific problems identified from ambient monitoring data.
Stations—26 stations (Figure 11 in text):
—16 stations in basins
—5 stations in urban/industrial bays
—5 stations in rural bays.
Stations will coincide with those established for intertidal shellfish
monitoring. Stations are located in recreational harvest areas where the
target species (butter clam, Saxidomus giganteus) for shellfish monitoring
C-26
-------
components is known to be adequately abundant. These areas therefore represent
the locations at which contaminated shellfish would pose a threat to a
large number of people. Some of these areas are within the influence of
known point or nonpoint sources of contaminants (Table C2). Final locations
of sites will be determined following completion of the ongoing U.S. EPA/Depart-
ment of Social and Health Services (DSHS) study of toxic chemicals and
bacteria in shellfish.
Frequency—Weekly (May-July)
—Monthly (August-April).
Because bacterial counts are highly variable, weekly sampling is needed
to provide adequate data to characterize the average count and range.
Because frequent sampling is required for at least part of the year, a
seasonally stratified design is recommended. Samples will be collected
weekly during May-July, a period of intensive harvesting of butter clams
and other shellfish. Samples will be collected monthly throughout the
rest of the year to provide a "spot-check" for persistent problems.
Timing—Entire year (low tide).
Data to be Reported—
0 Fecal coliform bacteria (most probably number)
• Enterococci (most probable number).
Data Analyses—
• Comparisons, by area, of temporal trends in coliform bacteria
and enterococci
0 Comparisons through time of spatial differences in coliforms
and enterococci
C-27
-------
TABLE C2. LOCATIONS OF SAMPLING STATIONS
FOR SHELLFISH AND PATHOGEN INDICATORS IN WATER
Location
Description
1. Oakland Bay -
Munson Point
2. Budd Inlet -
Priest Point Park
3. Commencement Bay -
Browns Point
4. Carr Inlet -
mouth of Burley Lagoon
5. Quartermaster Harbor -
Burton County Park
6. Saltwater State Park -
King County
7. Three Tree Point -
King County
8. Sinclair Inlet -
Ross Point
9. Vashon Island -
Point Vashon
10. Lincoln Park -
King County
11. Alki Point -
south side
12. Hood Canal -
Seabeck Bay
13. Port Orchard -
Brownsville
14. Bainbridge Island -
Skiff Point
Rural, residential; near major STP outfall.
Near major urban point and nonpoint pollution
sources.
Near major urban, industrial area; numerous
point and nonpoint pollution sources.
Rural, residential; removed from major urban,
industrial areas; near PCB source.
Rural, residential area; sheltered bay, non-
point sources.
Residential, commercial area; near major
point source discharges.
Urban, residential area; numerous point and
nonpoint sources.
Urban, industrial area; numerous point and
nonpoint pollution sources.
Rural, residential; removed from major point
source discharges.
Urban, residential area; numerous point
and nonpoint sources.
Residential, commercial area; numerous urban
point and nonpoint pollution sources.
Small residential, commercial; possible control
site.
Small residential, commercial; near major
marina.
Rural, residential; far from major point
source discharges.
-------
TABLE C2. (Continued)
15. Golden Gardens Park -
King County
16. Hood Canal -
Hood Head
17. Picnic Point -
Snohomish County
18. Port Gardner - 1 mile
east of Mukilteo
19. Marrowstone Island -
Fort Flagler
20. Camano Island - Camano
Island State Park
21. Padilla Bay -
Bay View State Park
22. Lopez Island -
Spencer Spit
23. Portage Bay -
Brandt Point
24. Bellingham Bay -
Post Point
25. Birch Bay -
State Park
Urban, commercial area; numerous point and
nonpoint sources.
Rural; removed from major point source
discharges.
Rural, residential; removed from major point
source discharges.
Urban, industrial area; numerous point and
nonpoint sources.
Rural; removed from major point source
discharges.
Rural; near major industrial area.
Rural; near oil refineries.
Rural; removed from major point source
discharges.
Rural, residential; near major urban, in-
dustrial area.
Urban, industrial area; numerous point and
nonpoint pollution sources.
Residential, commercial; removed from major
urban, industrial areas.
-------
• Identification of problem areas
• Comparisons of pathogen levels in clams with the above.
Replication and Statistical Sensitivity—Statistical sensitivity is
unknown. Composite samples are typically collected for ongoing compliance
monitoring. Tests on composite samples are used to determine whether public
health criteria have been exceeded, not to test for differences in levels
within or among stations. Although replicate field data on bacterial counts
in Puget Sound are unavailable, American Public Health Association (1985)
provides 95 percent confidence intervals for counts (most probable number)
of fecal coliforms based on analytical replication.
Analyses of three composite water samples are recommended. Concen-
trations of pathogen indicators are known to be highly variable in space
and time. Three replicates are the minimum recommended for statistical
comparisons among areas. To determine statistical power of this preliminary
design would require replicated daily samples over a period of 10-20 days.
It is recommended that a special study be conducted to determine the statistical
sensitivity of the present monitoring design, and to recommend an alternative
(if needed) to achieve reasonable statistical sensitivity.
Protocols—
t Field and Laboratory References: Tetra Tech and E.V.S. Con-
sultants (1986b).
• Supporting literature: American Public Health Association
(1985), Russek and Colwell (1983).
C-28
-------
BIOLOGICAL MONITORING DESIGNS
Biological populations of concern are distributed among all habitats
of Puget Sound. Sampling stations, frequency of sampling, and other aspects
of monitoring designs are discussed below for each biological population
addressed in the monitoring program.
BENTHIC MACROINVERTEBRATE ABUNDANCES
Variables and Rationale—Subtidal benthic macroinvertebrates will
be monitored because they are important biological components of the Puget
Sound ecosystem. Benthic invertebrates (especially infauna) are sensitive
indicators of both the intensity and areal extent of environmental pertur-
bations. They are also important mediators of nutrient recycling from
the detrital food web, thereby providing nutrients for primary production
in the water column. Infauna and epi fauna are important prey of species
at higher trophic levels, especially large epifaunal invertebrates and
fishes, many of which are harvested commercially or recreationally. Intertidal
macroi nvertebrates are often highly variable in species composition and
abundance due to natural extremes in physical, chemical, and biological
factors, and are therefore too variable to be used as a monitoring component
(Gray et al. 1980; Paine 1986).
All taxa should be identified to the lowest practical taxonomic level
and enumerated. Because species are the basic ecological units, changes
in species composition and abundance may be used to document temporal trends
in community composition [see, for example, Nichols (1985)], and to interpret
those trends with respect to anthropogenic impacts (Gray et al. 1980; Tetra
Tech 1985a). Abundances of pollution sensitive, pollution tolerant, and
opportunistic taxa are especially useful for this purpose. Benthic macro-
invertebrates are to be collected with either a 0.06-m2 box corer or a 0.1-m2
van Veen grab. An evaluation of sampling devices is in progress. Regardless
C-29
-------
of sampling device, samples should be washed on screens having 1.0-mm mesh
openings.
Stations—106 stations total (Figures 4-6 and Tables 4-6 in text):
—31 in basins
—34 in urban/industrial bays
—27 in rural bays.
Consult the previous section on Sediment Chemistry for the rationale
for station locations.
Frequency—Annual.
Annual sampling will be sufficient to document long-term trends in
species composition and abundance. It will also permit detailed comparisons
of results with those obtained from concurrent sampling of the sediments
(i.e., sediment chemistry, sediment toxicity bioassays, conventional sediment
variables). This group of components has been shown to be especially useful
for documenting anthropogenic impacts on benthic marine habitats (Chapman
and Long 1983; Tetra Tech 1985a).
Timing—Spring (March).
Sampling of benthic macroinvertebrates will coincide with sampling
of sediments for chemical analyses and toxicity bioassays. Sampling during
March is recommended because most of the sampled organisms will have been
exposed to the sediments for at least the previous winter. Benthic macroin-
vertebrate assemblages are also at a relatively stable point in their seasonal
cycles during March.
Data to be Reported--
• Species abundances for each replicate
• Numbers of taxa per replicate and per station
C-30
-------
Numbers of individuals for each replicate and numbers of
2
individuals/m for each station
• Station location, date, time, and water depth
• All field and laboratory QA/QC procedures and results.
Data Analyses—
• Graphical and statistical comparisons, by site, of temporal
trends in benthic community structure
• Comparisons through time (e.g., correlations, numerical
classification analyses) of spatial differences in benthic
community structure
• Identification of problem areas, as defined by changes in
species composition and abundance
• Correlations of patterns in benthic macroinvertebrates data
with patterns in sediment chemistry, bioassay, and fish
data.
Replication and Statistical Sensitivity—Replicate data are not available
for all types of benthic infaunal habitats within Puget Sound. However,
coefficients of variation were determined from representative data sets
from deep water stations in the main basin and from nearshore stations
in Port Gardner [respectively; Metro Seahurst Transects D-250E, E-750,
and H-640 (Word et al. 1984); Port Gardner Stations PG-1, PG-2,- and E-2
(U.S. Department of the Navy 1985)]. Power analyses of these data indicated
2
that the use of five replicate 0.1-m van Veen grab samples washed on 1.0-mm
mesh enable detection of differences of 75-125 percent (of the overall
mean among stations) in total numbers of species and 75-250 percent or
more in numbers of individuals, based on coefficients of variation of 9-45
C-31
-------
percent and 11-134 percent, respectively (see Appendix A, Figures Al and
A2). More replication would greatly increase statistical sensitivity,
but cost becomes prohibitive. The use of five O.OS-m^ box core samples
per station recommended herein may provide greater precision and therefore
greater statistical power.
Protocols--
t Field and Laboratory Reference: Te.tra Tech (1986c,h)
t Supporting literature: Swartz (1978), Mclntyre et al. (1984),
Eleftheriou and Holme (1984).
TOXIC CHEMICALS IN FISH
Variables and Rationale—Concentrations of toxic chemicals in fish
muscle tissue will be determined in Pacific cod (Gadus macrocephalus) caught
at recreational fishing piers, in salmon (Oncorhynchus tshawytscha), and
in English sole (Parophrys vetulus) used for histopathological analysis
(discussed in the subsequent component). Target chemicals include mercury
and Puget Sound organic contaminants of concern (Table Cl above), except
acid extractable and volatile organic compounds. The latter organic compounds
and metals other than mercury are not expected to accumulate to high levels
in fish muscle tissue. Periodic checks of these chemicals during intensive
surveys are recommended. These data can be used to identify problem areas
based on estimates of human health risk (e.g., Tetra Tech 1985a, 1986b)
and to manage recreational fisheries. Species recommendations and rationale
for sampling are given below for each of the three elements.
Based on the results of Landolt et al. (1985), bioaccumulation will
be evaluated for Pacific cod caught at recreational fishing piers. Pacific
cod was found by Landolt et al . (1985) to be among the most sought after
and most frequently consumed species. In addition, that species ranked
fifth and sixth in terms of numbers and weight in the recreational catch
evaluated by Landolt et al. (1985). Pacific cod was also found to have
C-32
-------
relatively high levels of PCBs in muscle tissue. Four of the five most
abundant species of the recreational catch (including Pacific cod) belonged
to the family Gadidae. Thus, Pacific cod was representative of the major
kinds of fishes that dominated the recreational catch.
Chemical contamination of muscle tissue from chinook salmon will be
monitored at three major fishing areas: Possession Point, Shilshole Bay,
and Point Defiance. Salmon were chosen for evaluation because of their
importance as a recreational and commercial resource. Moreover, many chinook
salmon are resident in Puget Sound for the entire marine portion of their
lifetimes.
In addition to bioaccumul ation analyses of recreationally harvested
species, concentrations of contaminants in English sole muscle will be
determined at all stations at which this species will be evaluated for
liver lesions (discussed below). These data will provide a sound-wide
picture of bioaccumulation patterns in a representative bottom-feeding
fish. The bioaccumulation data can also be related to the patterns of
lesion prevalences in English sole and characteristics of entire fish assem-
blages. Testing of English sole is consistent with the U.S. Food and Drug
Administration recommendation that non-migratory bottom-feeding fish are
the species of choice for pesticide and PCB testing (U.S. Food and Drug
Administration 1985).
Stations—
• Recreational pier fisheries - 15 major recreational fishing
areas (Figure 10 in text):
—Elliott Bay - 6 stations
—Commencement Bay - 4 stations
—Sinclair Inlet - 2 stations
—Edmonds - 1 station
—Everett Harbor - 2 stations.
C-33
-------
Fishes will be collected from major fishing locations near urbanized
areas to provide worst-case assessments of risk to public health from
consumption of chemically contaminated fishes. The 15 nearshore sampling
locations for Pacific cod are based mainly on those evaluated by Landolt
et al. (1985) and McCallum et al. (1985). A sampling location in
the heavily-contaminated City Waterway was substituted for the Point
Defiance site sampled by Landolt et al . (1985). Additional data on
fishing activity in other urbanized areas (e.g., Bellingham, Port
Angeles, Olympia, and Shelton) are needed to evaluate whether additional
stations are required.
t Salmon fisheries—3 major salmon fishing areas:
—Possession Point
—Shilshole Bay
—Point Defiance
t English sole—62 stations total (Figures 7-9 in text):
—16 in basis
—27 in urban/industrial bays
—19 in rural bays.
Samples of English sole to be analyzed for chemical contamination will
be taken from the specimens collected for histopathological analyses (discussed
below). Most trawl stations were selected to coincide with one or more
sediment quality stations. Thus, information collected from both kinds
of sampling can be interrelated. Trawl stations were not placed in most
rural bays with areas <10 mi^.
Frequency—
• Recreational pier species: annual for Pacific cod only
at recreational stations
C-34
-------
t Salmon: annual at salmon stations
t English sole: biennial at pathology stations.
Annual monitoring of Pacific cod and salmon is recommended because
of the importance of this information to assessment of human health risks.
The primary objective of the English sole monitoring will be to evaluate
bioaccumulation of toxic chemicals in a high trophic level organism. Because
assessing health risks based on English sole data is a secondary objective,
less frequent monitoring (biannual) is proposed.
Timing—
0 Fall (September-October) for recreational and salmon stations
• Summer (July) for English sole.
The fall sampling of cod and salmon is timed to coincide with a period
of intensive recreational harvest activity. English sole sampling will
be conducted at the time of their greatest abundance in nearshore areas.
Data to be Reported—
• Length, sex, reproductive status, and gross pathological
observations for each fish used for bioaccumulation analysis
• Lipid content of each composite tissue sample
• Chemical concentrations (wet weight) in each composite tissue
sample [priority pollutant metals and organic compounds
plus additional contaminants of concern (see above, Sediment
Chemistry, Table Cl). Volatile organic compounds and acid-
extractable organic compounds should be analyzed only if
data from sediment analyses indicate a potential for excessive
bioaccumulation.
C-35
-------
t Station location, date, time, and water depth
• All field and laboratory QA/QC procedures and results.
Data Analyses--
• Comparison of temporal trends in bioaccumulation at each
station
• Comparison of spatial trends in bioaccumulation among stations
t Comparison of tissue concentrations of chemicals with human
consumption guidelines
• Comparison of bioaccumulation results for English sole with
lesion prevalences and characteristics of entire fish assem-
blages.
Replication and Statistical Sensitivity—Because of the presumed high
level of variability, three composite samples will be analyzed for each
species at each station. Each composite should contain equal weights of
tissue from five individual fish. Little is known about statistical variability
of toxic chemical concentrations in any of the target species (but see
Landolt et al. 1985 and Tetra Tech 1986a). Preliminary data suggest that
coefficients of variation for concentrations of some chemical contaminants
in individual samples of selected species of flatfish are on the order
of 40-60 percent (Tetra Tech 1985b). The composite sampling strategy recom-
mended here will substantially increase statistical precision. For example,
the variance of the mean estimated by the sampling design using five individuals
per composite sample will be one-fifth of the underlying population variance
(Tetra Tech 1986a). If the coefficient of variation of the mean of individual
samples is 50 percent, then that of the composite samples will be 10 percent.
In this case, the minimum detectable difference in the mean concentration
of a chemical among stations would be about 35 percent of the overall mean
C-36
-------
among stations (extrapolated from Figures Al and A2 of Appendix A). More
precise estimates of statistical sensitivity will be obtained after data
are available for the composite samples recommended here.
Protocols—
• Field and Laboratory Reference: Tetra Tech (1986c,f,g).
HISTOPATHOLOGICAL ABNORMALITIES IN FISH
Variables and Rationale—All analyses will be conducted on English
sole (Parophrys vetulus). This species is likely to be found in adequate
abundances at all monitoring stations throughout the sound, and is known
to be affected by liver lesions in contaminated areas of the sound (e.g.,
Malins et al. 1984, 1985a,b; Tetra Tech 1985a; Krahn et al. 1986). Analyses
will be conducted on fish >3 yr old because they are the individuals most
likely to be affected with liver lesions (Malins et al. 1982). Otoliths
will be collected from each fish selected for hi stopathol ogical analysis
so that age can be determined. The age distribution of each sample must
be determined because prevalences of several liver lesions correlate positively
with age of English sole (Tetra Tech 1985a). In the field, only specimens
>23 cm total length (TL) will be sampled, to ensure that each fish is >3
yr old. Each specimen selected for histopathological analysis will be
measured (nearest mm TL) and weighed (nearest g wet weight) prior to necropsy
so that comparisons of fish condition (i.e., weight-at-length) and fish
growth (i.e., length-at-age) can be made. Grossly visible internal abnoi—
malities, sex, and reproductive state will also be determined for each
individual selected for histopathological analysis.
The primary liver lesions that will be evaluated microscopically are
neoplasms (i.e., tumors), foci of cellular alteration (i.e., putative pren-
eoplasms) and megalocytic hepatosis (i.e., a specific degenerative condition).
All three of these lesions are found in elevated prevalences in fish from
C-37
-------
contaminated areas of Puget Sound (e.g., Malins et al. 1984; Tetra Tech
1985a). The nomenclature used for describing these lesions should be consistent
with that described by Myers et al. (in prep.).
Consult the preceeding section on Sediment Chemistry for rationale
for station locations and the section on Toxic Chemicals in Fish for rationale
for timing and frequency.
Stations—62 stations total (Figure 7-9 in text):
—16 in basins
—27 in urban/industrial bays
—19 in rural bays.
These stations (trawl transects) will be generally located near stations
for monitoring of sediment quality and water quality components (see Sediment
Chemistry for rationale).
Frequency—Biennial at all stations.
Timing—Summer (July).
Data to be Reported—
• Length, weight, age, sex, reproductive state, and gross
pathological observations for each fish selected for histo-
pathological analysis
• Presence or absence of neoplasms, preneoplasms, and megalocytic
hepatosis in the liver of each fish
t Presence or absence of other abnormalities (including parasites)
in the liver of each fish
• Station location, date, time, and water depth
C-38
-------
• Transect length
• All field and laboratory QA/QC procedures and results.
Data Analyses—
• Evaluation of age effects on lesion prevalences
• Evaluation of sex effects on lesion prevalences
0 Comparisons of temporal trends in lesion prevalences at
each station
• Comparisons of spatial trends in lesion prevalences among
sites
• Comparison of changes in lesion prevalence with patterns
related to bioaccumulation in English sole and characteristics
of entire fish assemblages.
Replication and Statistical Sensitivity—A sample size of 60 English
sole is recommended for histopathological analyses at each station. A
sample size of this size will provide a confidence level of 95 percent
that at least one fish having a particular kind of lesion will be sampled
if the prevalence of that lesion in the population is >5 percent (Figure C2).
In addition, this sample size will also allow a 10 percent elevation in
prevalence to be discriminated statistically from a reference prevalence
of 0 percent (Figure C3). These estimates of statistical precision were
based on the results of a G-test for independence (Sokal and Rohlf 1981).
Protocols—
t Field and Laboratory References: Tetra Tech (1986c,d).
C-39
-------
UJ
N
CO
Ul
_J
Q.
2
<
CO
0
Ul
DC
5
o
ui
cc
300 -i
280-
Pravalence
In
Population
• 10%
I
0 1
I
2
1
3
1
4
1
5
1
6
1
7
1
8
1
9
1
10
POPULATION SIZE ( x 1,000)
Figure C2. Sample sizes required to detect one individual affected
with a lesion with 95% confidence, given various pop-
ulation sizes and prevalences.
-------
u
UJ
c
Q.
m
u
Q
D
Sample Size
-a- U-2Q
-•- N-40
-o- N-60
-o- N.100
-•*• N-200
2.5 5.0 7.5 10.0 125 15.0 175 20.0
PREVALENCE (%) AT REFERENCE SITE
Note: Significance level > 0.05
Statistical power - 0.80
Figure C3. Effects of sample size on the minimum detectable pre-
valence at a test site relative to the prevalence at the
reference site.
-------
FISH SPECIES ABUNDANCES
Variables and Rationale—Fish abundances will be determined at all
fish histopathology stations. These data can be used to evaluate potential
changes in demersal fish assemblages and to relate any observed changes
to patterns of liver pathology and bioaccumulation in English sole. Changes
in species assemblages can also be related to sediment chemistry, sediment
toxicity, and benthic macroinvertebrate assemblages. Each species should
be identified, measured (nearest mm total length), and examined for grossly
visible external abnormalities.
The rationale for station locations, frequency, and timing is given
in the previous section on Toxic Chemical in Fish
Changes in the abundances of demersal and pelagic fishes should also
be estimated using the catch and stock assessment information collected
routinely by the Washington Department of Fisheries.
Stations—62 stations total (Figures 7-9 in text):
—16 in basins
—27 in urban/industrial bays
—19 in rural bays.
Frequency—Biennial at all stations.
Periods—Summer (July).
Data to be Reported—
• Station location, date, time
t Depth
t Transect length
C-40
-------
• Species-level identification of each individual
• Length and gross pathological observation for each individual
t All field and laboratory QA/QC procedures and results.
Data Analyses—
• Comparisons of temporal trends at each site
• Spatial comparisons of differences among sites located at
similar depths
• Comparisons of assemblage changes with patterns of lesion
prevalence and bioaccumulation in English sole.
Replication and Statistical Sensitivity—Fish abundance data should
be collected from a minimum of four trawls per station. Statistical sensitivity
is unknown for Puget Sound waters. However, limited data from the Southern
California Bight indicate that coefficients of variation for numbers of
fish and numbers of fish species are on the order of 37-111 percent and
17-49 percent, respectively (Cross 1982). Given these estimates of coefficients
of variation, data from four trawls would permit the discrimination of
differences equal to 80-400 percent (of the grand mean among stations)
in numbers of fish and, 50-175 percent in numbers of fish species (see
Figures Al and A2 of Appendix A).
Protocols—
• Field and Laboratory References: Strange (1983), Tetra
Tech (1986c)
• Supporting literature: Nielsen and Johnson (1983), Mearns
and Allen (1978).
C-41
-------
SHELLFISH ABUNDANCES
Variables and Rationale—Shellfish (in particular oysters, clams,
crabs, and shrimp) are of ecological, economic, and recreational importance
in Puget Sound. Commercial stock assessments for oysters and catch statistics
of other shellfish stocks are monitored by Washington Department of Fisheries
(WDF). These monitoring efforts should be continued. As noted by Chapman
et al. (1985), there is also a need for regular stock assessments of crabs
and shrimp. This will be done for crabs, as one indicator, by assessing
bottom trawl catches made during bottomfish histopathology studies. The
use of butter clams (Saxidomus giganteus) as indicators of the status of
shellfish stocks is detailed below.
Shellfish abundances will be monitored in conjunction with other shellfish
components (pathogen indicators in water and shellfish, PSP in shellfish,
toxic chemicals in shellfish). The rationale for station locations is
given in the previous section on Pathogen Indicators in Water.
Stations—26 stations total (Figure 11 in text and Table C2 above):
—16 stations in basins
—5 stations in urban/industrial bays
—5 stations in rural bays.
Frequency—Annual.
Because butter clam populations have a slow turnover rate, annual
monitoring is adequate for assessment of abundance.
Timing—Spring (May).
Spring sampling is recommended to coincide with the analyses of toxic
chemicals in shellfish.
C-42
-------
Data to be Reported--
• Relative abundance by area
t Population age distribution
• Relative sizes of clams
• Station location, date, and time of sampling
• All field and laboratory QA/QC procedures and results.
Data Analyses—
t Comparisons, by area, of temporal trends in abundance, age,
and size of clams
• Comparisons of temporal change in spatial patterns in the
above
• Identification of problem areas
0 Comparisons of patterns in body burden, PSP, and pathogen
data with the above.
Replication and Statistical Sensitivity—Based on data obtained by
Stober and Chew (1984) for Puget Sound intertidal communities, 15 replicate
2
625 cm core samples, taken, to a depth of 30 cm, are generally required
to adequately assess intertidal butter clam populations.
Protocols—
0 Field and Laboratory References: Stober and Chew (1984).
C-43
-------
TOXIC CHEMICALS IN SHELLFISH
Variables and Rationale—Shellfish (in particular oysters, clams,
crabs, and shrimp) are of ecological, economic, and recreational importance
in Puget Sound. They are also relatively stationary, and have the potential
to bioaccumulate toxic substances from the water and sediments. Toxic
chemicals in butter clams (Saxidomus giganteus) will be monitored at the
same time that shellfish abundances are being determined. Specimens collected
for testing will be within the size range harvested recreationally and
commercially. Target chemicals are listed in Table Cl above. Volatile
organic compounds and acid extractable compounds should be analyzed for
only if a problem is suspected based on data from analyses of sediments
or nearby sources.
Toxic chemicals in butter clams will be monitored in conjunction with
other shellfish components. The rationale for station locations is given
in the previous section on Pathogen Indicators in Water.
Stations—26 stations total (Figure 11 in text and Table C2 above):
—16 stations in basins
—5 stations in urban/industrial bays
—5 stations in rural bays.
Frequency—Annual.
Annual monitoring will be conducted because of the importance of this
data to human health risk assessment. More frequent monitoring is precluded
by cost constraints.
Timing—Spring (May).
Sampling will be performed during spring when lipid content (and hence
contamination) of shellfish is expected to be highest.
C-44
-------
Data to be Reported--
• Concentrations of toxic chemicals in tissue (chemicals analyzed
for will be the same as those analyzed for in fish tissue
except all metals in Table Cl above will be included in
shellfish analyses)
• Shell length of each individual
• Lipid content of each composite sample
0 Station location, date, and time of sampling
t All field and laboratory QA/QC procedures and results.
Data Analyses—
t Comparisons, by area, of temporal trends in concentrations
of toxic chemicals in tissue
• Comparisons of concentrations of toxic chemicals in tissue
among areas
• Comparisons of tissue concentrations of chemicals with human
consumption guidelines (human health risk assessment)
• Identification of problem areas.
Replication and Statistical Sensitivity—Analyses of three composite
tissue samples (subsamples of which will also be used for pathogen and
other analyses) are recommended. Each composite should contain equal weights
of tissue from five individuals. Statistical sensitivity of the composite-
sampling strategy is unknown but can be estimated theoretically. Preliminary
data suggest that coefficients of variation for some chemical contaminants
in selected species of shellfish are on the order of 40-60 percent based
C-45
-------
on a grab sampling strategy (Tetra Tech 1986a). The composite sampling
strategy recommended here will substantially increase statistical precision.
For example, the variance of the mean estimated by the sampling design
using five individuals per composite sample will be one-fifth of the underlying
population variance (Tetra Tech 1986a). If the coefficient of variation
of the mean of individual samples is 50 percent, then that of the composite
samples will be 10 percent. In this case, the minimum detectable difference
would be about 35 percent of the overall mean among stations (extrapolated
from Figures Al and A2 of Appendix A).
Protocols—
• Field and Laboratory References: Stober and Chew (1984),Tetra
Tech (1986c, 1986f,g).
PSP IN SHELLFISH
Variables and Rationale—Shellfish (in particular oysters, clams,
crabs, and shrimp) are of ecological, economic, and recreational importance
in Puget Sound. PSP in butter clams (Saxidomus giganteus) will be monitored
at the same time that pathogen indicators in water and shellfish are being
determined. Specimens collected for testing should be within the size
range harvested recreationally and commercially. Analyses will be conducted
using High Pressure Liquid Chromotography (Sullivan and Wekel1 1984; Sullivan
et al. undated). The ongoing DSHS Monitoring Program, whi-ch covers various
shellfish species in both commercial and recreational areas complements
this component of the Puget Sound Monitoring Program.
Stations—26 stations total (Figure 11 in text and Table C2 above):
—16 stations in basins
—5 stations in urban/industrial bays
—5 stations in rural bays.
The rationale for station locations is given in the previous section
on Pathogen Indicators in Water. After 1-2 yr of baseline data are collected
C-46
-------
for South Puget Sound and other regions where PSP concentrations in shellfish
are undetected, selected stations may be deleted from the program or sampled
less frequently.
Frequency—Weekly (May-Ouly)
—Monthly (August-April).
Frequent sampling is needed to characterize PSP concentrations in
shellfish because of their high variability.' Because intensive sampling
is required, a seasonally stratified sampling design is recommended. The
most frequent sampling will occur during May-Ouly, a time of intensive
harvesting of butter clams and other shellfish. Monthly sampling is considered
adequate for the remainder of the year.
Timing—Entire year (low tide).
Data to be Reported—
• PSP concentration in tissue
• All field and laboratory QA/QC procedures and results.
Data Analyses—
t Comparisons, by area, of temporal trends of PSP in tissue
• Comparisons of temporal change in spatial patterns in the
distribution of PSP
• Identification of problem areas.
Replication and Statistical Sensitivity—Statistical sensitivity is
unknown. Analyses of three composite tissue samples (subsamples of which
will also be used for analyses of body burden of toxic chemicals and pathogen
indicators) are recommended.
C-47
-------
Protocols—
t Field and Laboratory References: Stober and Chew (1984),
Greenberg and Hunt (1984), Sullivan and Wekel1 (1984), and
Sullivan et al. (undated).
PATHOGEN INDICATORS IN SHELLFISH
Variables and Rationale—Shellfish (in particular oysters, clams,
crabs, and shrimp) are of ecological, economic, and recreational importance
in Puget Sound. Pathogens in butter clams (Saxidomus giganteus) will be
monitored at the same time that pathogen indicators in water are being
determined. At present, only analyses for pathogen surrogates (i.e., total
and fecal coliform bacteria; enterococci) are planned. Further analyses
for viruses (e.g., Vibrio spp.) may be appropriate if specific problems
are identified. Specimens collected for testing should be within the size
range harvested recreationally and commercially. The ongoing DSHS Monitoring
Program, which covers various shellfish species in commercial harvest areas,
complements this component of the Puget Sound Monitoring Program. It is
recommended that complementary monitoring programs be implemented by county
health districts to provide wider spatial coverage of Puget Sound beaches.
The rationale for station locations, frequency, and timing of sampling
is given in the previous sections on Pathogen Indicators in Water and PSP
in Shellfish.
Stations—26 stations total (Figure 11 in text and Table C2 above):
—16 stations in basins
—5 stations in urban/industrial bays
—5 stations in rural bays.
Frequency—Weekly (May-July)
—Monthly (August-April).
C-48
-------
Timing—Entire year (low tide).
Data to be Reported—
t Fecal coliform bacteria in tissue (most probable number)
• Enterococci in tissue (most probable number)
• All field and laboratory QA/QC procedures and results.
Data Analyses—
• Comparisons, by area, of temporal trends in coliform bacteria
and enterococci
• Comparisons of temporal change in spatial patterns in coliform
bacteria and enterococci
t Identification of problem areas.
Replication and Statistical Sensitivity—Analyses of three composite
tissue samples (subsamples of which will also be used for body burden analyses)
are recommended. Statistical sensitivity of this design is unknown. At
present, composite samples are routinely tested for determining levels
of pathogen indicators in shellfish. With a few exceptions, one composite
sample is used to perform the test. These tests are conducted solely for
the purpose of determining whether or not public health criteria are exceeded.
Hence, there has been no effort to determine the statistical sensitivity
of this sampling design or alternate designs. It is recommended that a
special study be conducted to determine the statistical sensitivity of
the present monitoring design, and to recommend an alternative (if needed)
to achieve reasonable statistical sensitivity.
C-49
-------
Protocols--
t Reference: Stober and Chew (1984), Chapman et al. (1985),
Tetra Tech (1986h).
MARINE MAMMAL ABUNDANCES AND REPRODUCTIVE SUCCESS
Variables and Rationale—Marine mammals constitute an important ecological
and aesthetic component of the Puget Sound ecosystem and are protected
by the Marine Mammal Protection Act of 1972. Many are near the top of
the food chain, and any effects of pollution on this group of animals may
serve as an analog for possible effects on humans.
Of the variety of marine mammals found in Puget Sound, the harbor
seal (Phoca vitulina) is most appropriate for monitoring for a number of
reasons. It is the only common marine mammal that is a resident and breeds
in Puget Sound. Unlike more migratory species it is not potentially exposed
to contaminants in other areas. Harbor seals feed in waters influenced
by industrial activities, and feed on many of the same fish species consumed
by humans. Studies have shown that harbor seals in the sound can have
high levels of toxic chemicals, and they may be vulnerable to the effects
of pollution. Data are available on concentrations of contaminants in
Puget Sound harbor seals, and experimental research on harbor seals is
in progress in Europe to investigate the effects of eating contaminated
fish. Finally, there is an ongoing program to monitor harbor seal populations.
Data from this program will be incorporated into the Puget Sound database.
The abundances of harbor seals (adults and pups) are determined by
the Washington Department of Game using aerial survey counts. Annual counts
are made at the peak of the 4-6 wk pupping period, when the highest percentage
of harbor seals are hauled-out on shore. This ensures that the maximum
possible number of individuals is observed. To determine the peak of the
pupping period, aerial and/or land-based surveys will be conducted in each
region of Puget Sound. Land-based surveys in selected areas will provide
additional information on the number of young produced and pup mortality.
C-50
-------
Permits from the Department of Commerce are required for all studies
and surveys, following regulations outlined in the Marine Mammal Protection
Act.
Survey Area—Aerial surveys will be conducted in areas known to be
used for breeding and non-breeding activities (primarily haul-out areas)
that are currently being monitored:
0 North Puget Sound: San Juan Islands, Smith Island, Protection
Island, Dungeness Spit, eastern Puget Sound bays (e.g.,
Skagit, Padilla, Samish, Boundary)
• South Puget Sound: Gertrude Island, Henderson Inlet, McMicken
Island, Cutts Island, Budd Inlet, Eld Inlet, plus less frequently
used sites such as Nisqually Delta and Eagle Island.
Land-based surveys will be conducted at selected sites from the above
list.
Frequency—Annual for aerial surveys to estimate population size and
reproductive success.
—Monthly for land-based surveys, June to August
—Additional surveys as needed to determine the peak of the pupping
period.
Annual surveys are adequate to track population trends in these long-
lived, slowly reproducing species. Additional monitoring may be necessary
to determine the peak of the pupping season.
Timing—Annual 2-3 day aerial surveys should be conducted during the
peak of the pupping season in each region:
C-51
-------
• North Puget Sound: Typically during the second or third
weeks of August
t South Puget Sound: Typically during the second or third
weeks of September.
Aerial survey flights in each region will be conducted during low tides
and, if possible, on the same or consecutive days.
Monthly land-based surveys will be conducted from June to August in
North Puget Sound and from August to October in south Puget Sound.
Data to be Reported—
• Abundances of harbor seal adults and pups in the north and
south Puget Sound regions
• Pup productivity indices (numbers of pups/yr and pups/female)
• Number of live births, still-births, and abnormalities (from
land-based surveys)
• Field procedures, including training for survey personnel,
and QA/QC documentation.
Data Analyses--
• Comparisons of temporal and spatial trends in harbor seal
population estimates using the pup productivity index and
abundance estimates
• Comparisons between north and south Puget Sound harbor seal
populations
C-52
-------
• Number of pups and percent survival, percent stillborn,
and percent abnormalities of pups.
Replication and Statistical Sensitivity—
0 Replicated data will not be collected because of cost constraints
• Statistical sensitivity unknown, but should be determined.
Protocols—Beach et al. (1985).
AVIAN ABUNDANCES AND REPRODUCTIVE SUCCESS
Variables and Rationale—Birds are important ecological and aesthetic
components of the Puget Sound ecosystem. Because many species are near
the top of the food chain, they are vulnerable to bioaccumulation of various
potentially toxic chemicals. Many species are vulnerable, as well, to
the loss, deterioration, or disturbance of breeding and foraging habitats.
The abundance of selected avian species will be monitored to identify any
significant changes that may be related to pollution, habitat loss, or
disturbance. Data from existing monitoring programs will be incorporated
into the Puget Sound database. In the future, additional monitoring may
be needed to provide information not provided through existing programs.
Abundances of many species are presently monitored monthly from October
through March using aerial survey techniques. Surveys are conducted concur-
rently by the U.S. Fish and Wildlife Service and the Washington Department
of Game (WDG). The primary .purpose of the aerial surveys is to census
migrant waterfowl (e.g., snow geese, Chen caerulescens; black brant, Branta
bernicula; ducks). Other birds are censused, but the data are not currently
being analyzed. The WDG also conducts waterfowl brood surveys statewide,
including the Puget Sound area. The U.S. Fish and Wildlife Service censuses
are more extensive. Both agencies survey from Bellingham to Olympia, exclusive
of the San Juan Islands. Ground surveys of seabird colonies have also
been conducted by the U.S. Fish and Wildlife Service and other investigators
C-53
-------
in the San Juan Islands, on Protection Island, and elsewhere in the sound.
Many of these data are included in the Catalog of Washington Seabird Colonies
currently being prepared. The Audubon Society also conducts an annual
"Christmas Bird Count" of all species in various areas of Puget Sound each
year.
To date, investigations of reproductive variables (e.g., nesting success,
clutch size, fledging success, thinning of eggshells) for marine birds
within Puget Sound have been only preliminary. NOAA and the U.S. Fish
and Wildlife Service are now funding studies of reproductive success and
its relationship to contaminants in tissues. Species presently considered
at possible risk in Puget Sound, and for which data on contamination are
available, include the pigeon guillemot, grebes, cormorants, scoters, great
blue heron, and rhinoceros auklet. Glaucous-winged gulls and bald eagles
are also potential candidates for monitoring because both breed and feed
in the marine waters of Puget Sound (Calambokidis, J., 2 June 1986, personal
communication). A monitoring program for reproductive success will not
be recommended until ongoing preliminary studies have been completed, and
the extent of possible problems has been defined.
Additional studies may be needed to identify species that may be most
appropriate for focused monitoring. Future monitoring should focus not
only on the problems of marine pollution but also on breeding and foraging
habitat loss and disturbance, particularly for colonially breeding seabirds
and species that breed widely throughout the sound. Selection of species
for focused monitoring should also involve determining the degree of risk
for particular marine species that may be vulnerable to oil spills.
Survey Area—Areas currently being surveyed in north Puget Sound and
south Puget Sound from Bellingham to Olympia (exclusive of the San Juan
Islands). Known seabird breeding colonies should also be surveyed during
the nesting period.
Frequency—Monthly aerial surveys, plus seasonal waterfowl brood surveys
and surveys of seabird colonies.
C-54
-------
Timing—Monthly surveys October through March, brood surveys and surveys
of seabird colonies during summer.
Data to be Reported—
• Date, area, and time of surveys
t Weather and tidal conditions during surveys
• Survey methods, transect length, number of observers, method
of transportation
• Species identifications
• Numbers of individuals per species
• Proportions of young to adults during brood surveys and
as applicable during other surveys.
Data Analyses—
• Temporal and spatial trends in species abundances
• Temporal and spatial trends in habitat use by area
0 Temporal and spatial trends in indicators of reproductive
success.
Replication and Statistical Sensitivity—Statistical sensitivity is
unknown. There are no current plans for replication. However, avian abundances
can be quite variable. To determine the extent to which replication may
be necessary, existing survey data for birds in the sound should be analyzed.
If data prove inadequate for making a judgment on replication, more surveys
should be conducted to determine coefficients of variation and to establish
C-55
-------
the necessary level of sampling replication. It is expected that efforts
will be focused on species that exhibit the least unexplained variance
in abundance, and that are judged to be the best indicators of environmental
change.
Protocols—None available at present.
C-56
-------
RIVER MONITORING DESIGN
Most freshwater inflows to Puget Sound consist of discharges from
natural river systems that include both point and nonpoint sources of contam-
inants. The river basins within the scope of the Puget Sound monitoring
program include the Nooksack-Sumas Basins, Skagit-Samish Basins, Stillaguamish
Basin, Snohomish Basin, Cedar-Green Basins, Puyallup Basin, Nisqually-Deschutes
Basins, West Sound Basins, and Elwha-Dungeness Basins (Figure 12 in text).
The design of a water quality monitoring program for these river basins
is discussed in the next section. Some considerations for future development
of a biological monitoring program for rivers and other nonpoint sources
are presented in the main text (see above, Nonpoint Source Monitoring).
The primary objectives of the ambient river monitoring program are to:
• Quantify annual mass loadings of contaminants from freshwater
inflow to Puget Sound
• Determine compliance with established ambient water quality
standards and criteria
0 Estabish baseline water quality conditions and detect trends
in water quality.
These objectives have conflicting requirements for both sample station
location and sampling frequency. For example, because contaminant mass
loading is often positively correlated with discharge, a monitoring program
designed to determine average annual contaminant load in a river system
would focus the sampling effort on the point of discharge (mouth of the
river) under high-flow conditions. Samples would be collected more frequently
during spring runoff and storm events to characterize peak mass loading
episodes. However, to determine compliance with water quality criteria,
C-57
-------
the emphasis should be on characterizing critical water quality conditions
(e.g., worst-case). The concentrations of dissolved chemical constituents
in streams influenced by point sources or groundwater infiltration are
often inversely related to flow. Therefore, critical periods when water
quality criteria are typically exceeded may occur during low-flow periods
in late summer and early fall. To satisfy the requirements of both program
objectives (quantifying contaminant loadings and detecting trends in water
quality), a balance must be achieved between alternative monitoring designs
for both sample station location and sampling frequency.
Although the monitoring program for rivers discharging into Puget
Sound has been designed to obtain information on both contaminant loading
and water quality conditions, the major emphasis has been placed on quantifying
contaminant loading. Consequently, the first priority is on sampling the
mouths of the major rivers that discharge directly into Puget Sound. Additional
stations have been placed upstream in selected drainage basins to isolate
the contaminant input from individual sub-basins and to track changes in
water quality along the length of the river. The ambient monitoring program
consists of a core set of fixed sampling stations plus a series of stations
that will be sampled on a rotating schedule with each station sampled for
3 consecutive years. The following sections describe the rationale for
the selection of water quality parameters to be analyzed, sample station
locations, and sampling frequency. Because of the importance of mass loading
calculations, approaches to estimation of contaminant mass loading are
evaluated in a later section (see below, Data Analyses, Approaches to Quanti-
fication of Contaminant Loadings).
One of the objectives of .the monitoring program is to establish baseline
conditions in the major areas discharging into Puget Sound. Consequently,
the proposed program includes complete analysis of all chemicals of concern
at each monitoring station. However, after the results of the baseline
sampling are available it is recommended that the data be reviewed to evaluate
water quality variables and station location. A tiered approach to sampling
design was taken because little is known about contaminant concentration
and loadings in many of the rivers in the Puget Sound Basin. Therefore,
C-58
-------
it is necessary to first conduct a comprehensive monitoring program to
determine baseline conditions and identify specific problems. Once the
current conditions are understood, the program can be modified to address
specific problems in individual drainage basins and a routine monitoring
program established. Potential modifications to the present design are
described in the section below (see Program Evaluation Strategy).
Variables and Rationale—Water quality variables that will be measured
include selected U.S. EPA priority pollutant metals and organic compounds,
additional metals, additional pesticides of concern, and conventional water
quality variables (Table C3). The rationale for the selection of these
variables is discussed below.
Flow is one of the most important indicators of river conditions.
Accurate discharge records are needed to quantify contaminant loads and
to establish critical water quality periods (e.g., low flow). For this
reason, continuous flow gaging is recommended at each monitoring station.
Analyses of several conventional water quality constituents are recom-
mended. Specific conductance, which is a measure of the ability of water
to conduct electricity, serves as an indicator of total ionic concentration.
The pH of water is one of the primary indicators used to evaluate surface
waters for their suitability for specific beneficial uses (Washington Admini-
strative Code 1982). Also related to pH is alkalinity, a measure of buffering
capacity or ability to neutralize acids or bases. Dissolved oxygen is
necessary to maintain aerobic conditions in surface waters and is considered
a primary indicator of the capability of surface water to support aquatic
life. Fecal coliform bacteria and enterococci bacteria are used as indicators
of contamination by human and animal wastes.
Nutrients are essential elements that control aquatic plant growth.
Both nitrogen and phosphorus are common nonpoint pollution problems, associated
with runoff from agricultural and urban lands. Nitrogen and phosphorus
are also present in domestic wastewater. Identification of specific nitrogen
compounds present in surface waters can be used to evaluate contaminant
C-59
-------
TABLE C3. WATER QUALITY VARIABLES TO BE MEASURED
IN THE RIVER MONITORING PROGRAM
Conventionals (bulk water)
flow
temperature
dissolved oxygen
alkalinity
conductivity
total suspended solids
PH
fecal coliform bacteria
enterococci bacteria
total organic carbon3
•total nitrogen3
nitrite and nitrate
ammonia
total phosphorus
Metals (bulk water and particulate fraction)^
arsenic
cadmium
chromium
copper
lead
mercury
nickel
silver
zinc
aluminum
iron
manganese
Organic Compounds (bulk water and particulate fraction)^
4,4'-DDE
4,4'-DDD
4,4'-DDT
aldrin
dieldrin
alpha-chlordane
gamma-chlordane
alpha-endosulfan
beta-endosulfan
PCBs
endosulfan sulfate
endrin
endrin aldehyde
heptachlor
heptachlor epoxide
alpha-HCH
beta-HCH
delta-HCH
gamma-HCH (lindane)
other pesticidesd
3 By elemental analyzer on particulate fraction only.
b Data to be reported as total recoverable (mg/L for
metals; ug/L for organic compounds) and as particulate
fraction (mg/kg for metals; ug/kg for organic compounds;
dry weight).
c Individual polynuclear aromatic hydrocarbons (PAH)
classified as U.S. EPA priority pollutants.
d Additional pesticides of concern may vary among sampling
stations.
-------
sources. An excessive amount of ammonia usually indicates a recent discharge
of organic wastes. Waters containing mostly nitrate nitrogen are considered
to be stabilized, suggesting a less recent discharge of waste (e.g., upstream
discharge). Because nutrients undergo transformation within the estuarine
environment prior to entering the sound, the analyses have been limited
to total phosphorus, total particulate nitrogen, nitrite/nitrate nitrogen,
and ammonia nitrogen.
Total organic carbon (TOC) provides a measure of plant detritus, decay
products, and organic chemicals, as well as other organic substances.
It is a gross indicator of organic pollution. In addition, waters with
relatively high TOC concentrations tend to exhibit correspondingly high
levels of nutrients (Sanders et al. 1983).
Suspended solids are contributed by erosion processes and wastewater
discharges. Because many of the toxic contaminants are associated with
particulate material, measurement of total suspended solids also aids in
quantifying contaminant loads.
The metals and organic compounds recommended for analysis in the river
monitoring program have been identified as potential problem chemicals
in Puget Sound marine and estuarine sediments (Tetra Tech 1985a,b,c).
The metals include nine priority pollutants that have been consistently
elevated above reference in Puget Sound sediments and three additional
metals (iron, aluminum, and manganese) that are used to evaluate natural
inputs, particularly from nonpoint sources. The organic compounds include
U.S. EPA priority pollutant pesticides, PCBs, and individual PAH. Other
organic compounds have been identified as problem chemicals in select areas
of Puget Sound, but have not been included in the target variables because
most compounds are not likely to be problems on a regional scale. PCBs
and PAH are ubiquitous in sediments throughout Puget Sound and are typically
found in nonpoint sources in urban, industrial areas. Furthermore, both
PCBs and PAH can be analyzed with relatively low cost, sensistive analytical
techniques (i.e., PCB analyzed by gas chromatography/electron capture detection,
and PAH analyzed by gas chromatography/flame ionization detection) making
C-60
-------
them good indicator compounds for a regional monitoring program. Pesticides
are also analyzed with PCBs at little additional cost. Additional chemicals
of concern in specific watersheds, such as certain pesticides applied to
agricultural areas, should be evaluated further for possible inclusion
in the program.
Metals and organic compounds tend to associate with particulate matter
because of their low solubility in water. The majority of the contaminant
loading to Puget Sound from rivers could be contaminants associated with
the particulate matter. In addition, lower detection limits can be achieved
by separate analyses of the particulate fraction. Thus, both particulate
fraction and whole water samples will be analyzed at all stations. Analysis
of whole water samples is necessary to account for dissolved forms of contam-
inants and to obtain estimates of total concentrations and total loading.
Stations—32 stations total (Figures 12-14; Table 7 in text):
—8 fixed stations at mouths of major rivers
—13 rotating stations at mouths of minor rivers (7 stations
sampled during first 3-yr period and 6 sampled during second
3-yr period)
—11 rotating stations upstream in the Green-Duwamish (7 stations)
and Skagit (4 stations) River Basins.
The selection of monitoring station locations was based on a combina-
tion of drainage basin characteristics, available water quality data, and
expected vulnerability of the receiving environment. The initial selection
of the major river systems to be monitored was based primarily on mean
annual streamflow because contaminant loading is largely a function of
discharge. Rivers that discharge into receiving waters with low flushing
potential, known contaminant problems, and extensive biological resources
were also given a high priority for monitoring.
Primary Stations—The following eight rivers were selected for the
core of the monitoring program: Skagit, Snohomish, Stillaguamish, Nooksack,
Puyallup, Sammamish-Cedar, Green-Duwamish and Deschutes. With the exception
C-61
-------
of the Deschutes River, the core rivers rank as the largest rivers in the
Puget Sound Basin (Table C4). The Deschutes River contributes only about
1 percent of the total annual riverine inflow to the sound. Nevertheless,
it was selected as a core station because of the large population in the
basin (441/mi2) (especially in the lower basin) and because it discharges
into Budd Inlet, a shallow, confined bay with documented water quality
problems (Table C5). Most of the larger rivers discharge along the eastern
border of the sound near major population centers. With the exception
of the Nisqually River, the eight largest rivers discharge into confined
bays with moderate to extensive development of sensitive habitats or bays
with known contamination problems. Therefore, the Nisqually River was
assigned a lower priority (see below).
The remaining 13 rivers that discharge to Puget Sound will be sampled
on a 3-yr rotating schedule, with seven stations sampled during each year
of the first 3-yr period and six stations sampled during each year of the
second 3-yr period. U.S. GS has applied a similar approach in their National
Water Quality Assessment Program (see Hirsch 1986 for rationale). Although
these 13 rivers are expected to account for less contaminant loading to
the sound than the 8 major rivers, they may have significant local influences,
especially when discharging to small, confined bays. The 3-yr rotating
sampling scheme will also be used as an initial screening tool to rate
the importance (relative to Puget Sound contaminant loading) of the smaller
Puget Sound tributaries (see below, Program Evaluation Strategy).
The locations of the 8 core stations and the 13 rotating stations
are shown in Figure 12 and Table 7 of the main text. These stations are
located near the mouths of the rivers but upstream of tidal influence.
Sampling above tidal influence is necessary to ensure that water quality
measurements are independent of marine and estuarine influence. Also,
accurate flow-gaging is not possible in river sections influenced by tides.
Contaminant loading from lands adjacent to riverine tide zones can
be accounted for by establishing sampling stations on the tributaries that
discharge into the estuarine portion of the river. In many areas, the
C-62
-------
TABLE C4. FLOW RANKING OF RIVERS
DISCHARGING INTO PUGET SOUND
Skagit
Snohomish
Stillaguamish
Nooksack
Puyallup
Nisqual ly
Green-Duwamish
Sammamish-Cedar
Elwha
Skokomish 1
Skokomish 2a
Deschutes
Dosewallips
Duckabush
Hamma Hamma
Dungeness
Samish
Big Quilcene
Tahuya
Whatcom
Little Quilcene
Other
Total
Basin
Area
(mi2)
3,130
1,780
684
822
972
712
483
607
321
116
128
162
116
77
85
198
106
101
47
65
25
13,210
Average
Annual
Flow
(ft3/sec)
16,700
10,400
4,600
3,900
3,400
2,500
1,700
1,800
1,500
1,300
1,100
810
800
540
500
460
190
180
140
120
74
2,180
56,690
Percent
Flow
30
18
8
7
6
5
3
3
2
2
2
1
1
1
1
1
<1
<1
<1
<1
<1
3
100
Cumulative
Percent Flow
30
48
56
63
69
74
77
80
82
85
87
88
89
90
91
92
92
93
93
93
94
97
100
a Skokomish 2 is outlet from the Cushman powerhouse No. 2.
-------
TABLE C5. RECEIVING WATER CHARACTERISTICS FOR
PRINCIPAL PUGET SOUND RIVERS
River
Nooksack
Whatcom
Samish
Skagit
Stillaguamish
Snohomi sh
Cedar-Sammamish
Duwami sh-€reen
Puyallup
Nisqually
Deschutes
Elwha/Dungeness
Quilcene
Dosewallips/
Ouckabush/
Hama Hama
Skokomi sh/Tahuya
Receiving Water
Lummi Bay
Bel lingham Bay
Samish Bay
Skagit Bay
Port Susan
Everett Harbor
Shilshole Bay
Elliott Bay
Commencement Bay
Ni squally Reach
Budd Inlet
Dungeness Bay
Quilcene Bay
North/Central
Hood Canal
Great Bend
Hood Canal
Receiving
Water Morphology
Shallow,
semi -enclosed
Moderately deep,
semi -enclosed
Shal low,
semi -enclosed
Shallow, enclosed
Shallow to deep,
enclosed
Shallow to deep,
open
Shallow, very open
Deep, very open
Deep, open
Shallow, very open
Shallow, enclosed
Shallow,
semi-enclosed
Shallow,
semi -enclosed
Shallow to deep,
enclosed
Shallow to deep,
very enclosed
Chemical
Contamination3
No data
Med i urn
Low
Low
Low
Moderate
Moderate
Moderate-High
Low-High
No data
Low-Moderate
No data
No data
Low
Low
Bacterial
Contamination''
Certified
Not certified
Certified
Certified
Certified*1
Not certified
Not certified
Not certified
Not certified
Certified
Not certified
Certified
Certified
Certified
Certified
Estuarine/Riparian
Habitat Availability0
Moderate
Moderate
Extensive
Extensive
Extensive
Extensive
Limited
Limited
Limited
Extensive
Limited
Extensive
Extensive
Moderate
Moderate
a Based on average elevations above reference (EAR) for PAH, PCS, Hg, and As:
High = EARMOO
Medium = 1025 percent
d North Port Susan only.
Reference: Evans-Hamilton and D.R. Systems, Inc. (1968).
-------
most highly developed agricultural and urban lands are located in the lower
part of the drainage basin. By sampling only above tidal influence on
the mainstem of the river, a potentially significant portion of the drainage
basin is not taken into account in the loading determination. In urbanized
watersheds, there are few natural tributaries in these areas. Instead,
much of the surface water runoff is conveyed by municipal storm drain systems.
Discharges from storm drain systems should be monitored to determine the
contaminant contribution from urban runoff. Storm drain discharges are
currently regulated under the NPDES program. However, little information
is available on pollutant loadings. Monitoring of storm drain discharges
will eventually be required under the NPDES program, but until the program
is established, interim monitoring by responsible parties may be required
to evaluate contaminant loadings from urban runoff.
Secondary Stations—Additional sampling stations located upstream
in the drainage basin are needed to define water quality patterns along
the length of the rivers, as well as to identify contaminant input from
individual sub-basins. Sampling in the upper drainage basins will be conducted
on a 3-yr rotating schedule for the Deschutes River and the eight major
rivers in the Puget Sound Basin: Skagit, Snohomish, Stillaguamish, Nooksack,
Puyallup, Nisqually, Sammamish-Cedar, and Green-Duwamish.
Selection of ambient water quality monitoring stations within individual
drainage basins has been completed for two example basins: the Skagit
and Green-Duwamish drainage basins. The Skagit River was selected because
it is the largest river discharging into Puget Sound. The Green-Duwamish
River was selected because of the heavy urban and industrial development
in the lower portion of the drainage basin. Cadmium, copper, lead, and
mercury concentrations at several stations in the lower Green-Duwamish
River exceed U.S. EPA criteria for aquatic life (Harper-Owes 1983). The
selection of sampling station locations was based on drainage basin charac-
teristics and available water quality and loading data. The basin parameters
used to select stations were:
C-63
-------
t Mean annual discharge
• Percentage of basin in major land use categories
• 1980 population density
• Change in population between 1960 and 1970
0 Number and distribution of NPDES-permitted discharges.
A detailed description of the station selection sampling for the Skagit
River and the Green-Duwamish River is presented below.
Four monitoring stations (in addition to the station at the mouth
of the river) have been selected in the Skagit River Basin to obtain information
on contaminant loadings from individual sub-basins and to trace water quality
and loading patterns along the mainstem of the Skagit River (Figure 13
in text). The four upstream monitoring stations are:
• Skagit River at Marblemount
t Sauk River near Rockport
• Skagit River below Concrete
• Skagit River near Sedro Wooley.
Where possible, these stations have been located at sites where there are
existing U.S. Geological Survey stream discharge gaging stations. The
monitoring stations on the Skagit River at Marblemount and Sauk River at
Rockport are designed to determine background water quality conditions
in the basin and provide contaminant loading data for the two largest sub-basins
of the Skagit River. The two additional mainstem stations located below
Concrete, combined with the station near the mouth of the Skagit River,
will aid in tracing changes in water quality conditions along the length
C-64
-------
of the river. In addition, these stations will help to identify the areas
in the lower river basin that contribute the largest mass loadings of contami-
nants to Puget Sound. Source control activities can then be targeted to
these specific areas. The lower river basin is suspected of contributing
a significant portion of the total basin contaminant load because of the
higher percentage of urban and agricultural development and because it
has supported a larger population growth (1960-1970) than has the upper
river basin. Additional rationale for station locations is given below.
The Skagit River drains an area of about 3,130 mi^ in Snohomish, Skagit,
and Whatcom Counties (Figure 13 in text). The Skagit River discharges
into Puget Sound below Mount Vernon. With an average annual flow of 16,700
ft-Vsec, the Skagit River accounts for about 30 percent of the total freshwater
input to Puget Sound. The Sauk River and Baker River, two major tributaries of
the Skagit River, account for a total mean annual flow of about 7,010 ft3/sec,
or about 40 percent of the total flow in the Skagit River.
Flow in the upper Skagit River is controlled by a series of dams.
All three dams (Gorge, Diablo, and Ross) are located on the mainstem of
the river upstream of Marblemount and are operated for hydroelectric power
generation. The Baker River, a major tributary of the Skagit River, is
also impounded for hydroelectric power generation. Flow is usually controlled
by releases from Baker Dam located near the mouth of the river just north
of Concrete.
The lower Skagit River Basin below Concrete may be a more significant
source of contaminant loading to Puget Sound than is the upper basin above
Concrete (Table C6). Population density in the lower basin (95/mi^) greatly
exceeds population density in the upper basin (2/mi^). in addition, the
population in the lower basin increased by about 11 percent between 1960
and 1970, while the population in the upper basin declined by 40 percent.
The upper Skagit River Basin, above Rockport, consists primarily of forest
lands in the Mount Baker-Snoqualmie National Forest and the North Cascades
National Park. Most of the agricultural and urban development has been
confined to the lower river basin below Concrete.
C-65
-------
TABLE C6. SKAGIT RIVER BASIN CHARACTERISTICS
Average Average
River Basin
Skagit River Basin
Upper Basin
Cascade River
Baker River
Sauk River
South Fork
Sulattle River
White Chuck River
Lower Basin
Nookachamps Creek
Basin
Area
(ml2)
3130
1520
185
297
732
50
346
86
400
72
Annual
Flowa
(ft3/sec)
16700
5900
1030
2630
4380
300
1750
890
2700
160
Land Use0
Forest
94
97
98
96
97
90
98
98
76
71
Agric.
2
<1
<1
0
<1
<1
<1
<1
14
15
Range
1
<1
0
<1
<1
10
<1
1
3
5
Dev.
1
<1
1
<1
<1
<1
<1
<1
6
6
(X)
Storage
2
2
1
4
1
<1
1
<1
2
3
Precip.c
(In)
97
77
131
134
125
160
130
140
78
53
Spec.
Dlsch.6
1980. («3/sec
Pop.d per ml2)
42000 5
3000 4
100 6
150 9
1700 6
2000 6
100 5
<100 10
38000 7
3000 2
Pop.
Density'
(per ml2)
13.4
2.0
0.5
0.5
2.3
40.0
0.3
1.2
95.0
41.7
NPDES
Major
1
0
0
0
0
0
0
0
1
0
Permit
Minor
14
3
1
0
0
0
0
0
9
0
1960 to Sed.
1970 Pop. Loadh
Chanqe9 (1000
(X) mt/yr)
0.3 1240
-40
-16
-16
-10
-32
-10
-5
11
13
aWilliams et al. (1985a).
bPuget Sound Task Force (1970).
cSoil Conservation Service (1965).
dU.S. Department of Commerce (1980).
eSpecific discharge = average annual discharge/basin area.
fPopulation density = total population/discharge basin area.
9U.S. Department of Commerce (1970).
hURS (unpublished).
-------
Water quality data for 1980-1985 were obtained from the STORE! database
at four stations in the Skagit River Basin. Station locations are shown
in Figure 13 of the main text. Data are generally unavailable for many
of the variables of concern described in the previous section. Existing
data are limited to analyses of conventional contaminants and nutrients,
with sporadic metals data available for the Skagit River near Mount Vernon
and the Baker River at Concrete.
The available data were used to estimate average mass loading of nutrients
at the four stations in the basin. Total nitrogen and total phosphorus
loading summarized in Figure C4 show that the upper Skagit River Basin
(above Concrete) contributes more than 50 percent of the nutrient load
measured in the Skagit River near Mount Vernon. This indicates that although
most developed areas are located in the lower river basin below Concrete,
the relatively undeveloped upper portion of the basin contributes a significant
part of the total nutrient load in the Skagit River.
The available metals data are not sufficient to conduct a sub-basin
loading analysis in the Skagit River Basin. Instead, the existing metals
data for the Skagit River near Mount Vernon and the Baker River at Concrete
were compared to ambient water quality criteria. The chronic toxicity
criteria are exceeded in all samples for lead, in 4 of 14 samples for copper,
and in 2 of 14 samples for zinc at the station located on the Skagit River
near Mount Vernon. The acute toxicity criterion for copper is exceeded
in a single sample at the Mount Vernon station. In the Baker River (at
Concrete), lead is the only metal that exceeds ambient water quality criteria.
Lead concentrations exceed the criteria for chronic toxicity in 4 of 18
samples.
Seven upstream monitoring stations are recommended in the Green-Duwamish
River Basin (Figure 14 in text). These upstream monitoring stations are:
C-66
-------
TOTAL NITROGEN LOAD (Ib/day)
TOTAL PHOSPHOROUS LOAD (Ib/day)
FLOW(ft3/sec)
PUGET
SOUND
MOUKTVERNON
"3" -s; 15,000
•C I 10,000
0 5 10 15
MILES
II
CONCRETE j MARBLEMOUNT
ROCKPORT
Note: Dashed line on Cascade River indicates no nutrient loading data are available.
Figure C4. Flow and nutrient loading in Skagit River.
-------
• Green River at Auburn
• Green River at Palmer
• Mill Creek above Western Processing site
• Mill Creek below Western Processing site
t Soos Creek near the mouth
• Jenkins Creek near the mouth
• Newaukum Creek near the mouth.
The majority of the upstream sampling stations are located on tributaries of
the Green River and along the mainstem in the lower river basin below Black
Diamond. Because most agricultural and urban development has occurred in the
lower basin, this area is expected to contribute a significant portion of the
total contaminant loading from the Green-Duwamish River system to Puget Sound.
The two stations on the mainstem of the Green River have been selected to
provide the necessary information to determine water quality and contaminant
loading patterns along the length of the river. Including the station near
the mouth of the river, these three mainstem stations will also provide
sufficient coverage to identify individual reaches along the Green River
where significant increases in contaminant loading occur. There are no
river monitoring stations proposed for the estuary downstream from Tukwila.
Most of the surface water runoff from this approximately 40-mi2 area discharges
to the Green-Duwamish River via municipal and private storm drain systems.
There are also numerous combined sewer overflows (CSO) that periodically
discharge into the estuary from the City of Seattle's combined sewer system.
Effluent from the Renton wastewater treatment plant also discharges into
the Green-Duwamish River below Tukwila. Additional monitoring of the major
storm drain systems and CSOs combined with a review of NPDES monitoring
reports from the Renton treatment plant will be required to quantify contaminant
loading from this portion of the Green-Duwamish River Basin.
C-67
-------
The two Mill Creek monitoring stations are located upstream and downstream
of the Western Processing site and can be used to evaluate the effectiveness
of remedial activities, as well as to estimate contaminant loading from
the Mill Creek Basin. The monitoring stations located at the mouths of
Soos Creek and Newaukum Creek are designed to provide information on contaminant
load and water quality conditions in the two largest tributaries of the
lower Green River. The station on Jenkins Creek, a tributary of Soos Creek,
was selected because available data suggest that Jenkins Creek contributes
a significant portion of the total load in Soos Creek. Additional rationale
for station locations is given below.
The Green-Duwamish River drains an area of about 483 mi2 in King County
(Figure 14 in text). The headwaters of the Green River originate in the
Cascade Mountains and flow about 82 mi before discharging into Elliott
Bay via the Duwamish Waterway system. The estuary extends from the mouth
of the Duwamish Waterway upstream to the confluence with the Black River
at river mile 12.0. Annual flow measured just above the estuary at Tukwila
averages about 1,700 ft-Vsec. Flow in the lower river is regulated by
releases from Howard Hanson Dam located upstream of Black Diamond at river
mile 65. Howard Hanson Dam is operated by the U.S. COE for flood control
and low-flow augmentation.
Land use characteristics in the Green-Duwamish River Basin are summarized
in Table C7. Over 70 percent of the lower river basin (below Black Diamond)
is developed for agricultural or urban use, while about 22 percent of the
area is forested or undeveloped. Approximately 97 percent of the total
basin population resides in the area downstream of Black Diamond. Population
density in the lower basin is about 1,700/mi2 compared to 24/mi2 in the
upper river basin. Additionally, 37 of the 40 NPDES-permi tted facilities
in the river basin, including the only major facility, discharge to the
Green-Duwamish River below Black Diamond. Most of the upper drainage basin
lies within the Mount Baker-Snoqualmie National Forest. About 92 percent
of the land in the upper basin is forested, with only about 7 percent in
the agricultural or developed-land category.
C-68
-------
TABLE C7. GREEN-DUWAMISH RIVER BASIN CHARACTERISTICS
River Basin
Green-Duwamish Basin
Upper Green River
Lower Green River
Soos Creek
Newaukum Creek
Hill Creek
Springbrook Creek
Average Average
Basin Annual
Area Flowa
(mtf) (ft3/sec)
483 1700
317 1000
170 700
67 130
28 62
9 15
23 40
Land Useb (X)
Forest
73
92
22
70
37
22
15
Agric.
8
5
39
11
56
39
10
Range Dev.
1 14
<1 2
<1 35
0 18
<1 7
<1 35
<1 75
Storage (in)
2 67
1 80
3 42
1 47
<1 48
3 47
<1 47
1980
Pop.d
300000
7500
292000
16900
13400
4000
24100
Spec.
Disch.e Pop.
(ft3/sec Density/
per mi^) (per miz)
4
3
4
2
2
2
2
621.1
23.7
1717.6
252.2
478.6
444.4
603.5
NPDES
Major
1
0
1
0
0
0
0
Permit 1
Minor
40
1
36
1
0
2
0
1960 to Sed..
970 Pop. Load"
Change? (1000
(») mt/yr)
30 123
47
29
19
47
18
132
aWilliams et al. (1985a).
bPuget Sound Task Force (1970).
cSoil Conservation Service (1965).
dU.S. Department of Commerce (1980).
eSpec1fic discharge = average annual discharge/basin area.
fPopulation density = total population/discharge basin area.
9U.S. Department of Commerce (1970).
(unpublished).
-------
Water quality data are available at 15 stations in the Green-Duwamish
River Basin (Figure 14 in text). Data for the 1980-1985 period of record,
obtained from the STORE! database, were used to estimate average daily
nutrient and metals loading at each station. In addition, available data
were compared to the ambient water quality criteria to determine where
frequent violations of water quality criteria occur.
Total nitrogen and total phosphorus loading estimates are summarized
in Figure C5. The upper Green River above Black Diamond contributes nearly
40 percent (1,450 Ib/day) of the total nutrient load in the river. Other
major nutrient sources in the Green-Duwamish River Basin include Newaukum
Creek (620 Ib/day), Soos Creek (650 Ib/day), Springbrook Creek (250 Ib/day).
and Jenkins Creek (240 Ib/day). Ammonia nitrogen and total phosphorus
concentrations are similar at most stations in the basin. Total phosphorus
concentrations (as phosphorus) range between 0.01 and 0.2 mg/L, and NH4
+ NH3 concentrations range between 0.001 and 0.5 mg/L. However, N02 + N03
concentrations in most of the tributaries that discharge into the lower
river basin are consistently 2-10 times greater than the concentrations
observed in the mainstem of the Green-Duwamish River (0.02-0.4 mg/L).
Higher N02 + N03 concentrations in the lower basin tributaries presumably
result from the increased urban and agricultural development in the lower
basin. One other exception is noted in Hill Creek, where total phosphorus
concentrations are frequently 5-10 times greater than elsewhere in the
basin.
With the exception of the Springbrook Creek station, the metals concen-
trations at most stations in the basin are frequently below analytical
detection limits. Springbrook Creek consistently exhibits elevated zinc
concentrations (Figure C6). Zinc concentrations exceed the ambient water
quality criteria for chronic toxicity (47 ug/L) in 39 out of the 42 samples
and exceed the acute toxicity criteria (320 ug/L) in 10 of the samples
(U.S. Environmental Protection Agency 1985c).
C-69
-------
TOTAL
NITROGEN
LOAD(lb/day)
Big Soos Creek —4 \
West Branch
' 3000
12000
! 1000
0246
MLES
TOTAL
PHOSPHOROUS
LOAD(lb/day)
FLOW
(ft 3/sec)
ir
600
400
200
0246
MLES
60,
40
H
0 246
MILES
TUKW1LA
AUBURN
PALMER
BLACK
DIAMOND
Figure C5. Nutrient loading in the Green-Duwamish River Basin.
-------
I
o
PC
o
o
IU
u
8
.O
0.
chronic = 3.2jig/L
YEAR
Note: Water quality criteria are shown by horizontal lines.
Figure C6. Summary of copper, lead, and zinc concentrations
in Springbrook Creek.
-------
The available data indicate that discharges from Mill Creek are responsible
for most of the metals and about half of the total nitrogen load in Springbrook
Creek. Mill Creek, a tributary of Springbrook Creek, drains the predominantly
commercial and light industrial areas north of Kent. The Western Processing
Superfund site is also located in the Mill Creek drainage basin. According
to the 1984 investigation by the Washington Department of Ecology, concen-
trations of metals (cadmium, copper, nickel, and zinc), total nitrogen,
and several organic compounds (chloroform, trichloroethane, trichloroethene,
and tetrachloroethene) in Mill Creek increase significantly downstream
of the Western Processing site (Yake 1985). Copper and zinc concentrations
in Mill Creek samples collected below the Western Processing site exceed
the ambient water quality criteria for acute toxicity (18 ug/L copper and
320 ug/L zinc) in all samples analyzed. In addition, lead concentrations
exceed the chronic toxicity level (3.2 ug/L) in 5 of 10 samples. None
of the organic contaminants exceed ambient water quality criteria.
Upstream sampling stations may be established in the remaining seven
river basins of the Puget Sound region after evaluating data from primary
stations. The seven basins are: Snohomish, Stillaguamish, Nooksack, Puyallup,
Nisqually, Sammamish-Cedar, and Deschutes. Locating monitoring stations on
upstream tributaries would provide information on contaminant loading contribu-
tions from individual sub-basins. It would also enable regulatory agencies
to target source identification and control activities to specific problem
areas. The selection of upstream sampling stations should be based on drainage
basin characteristics (e.g., population density, average annual flow, number
of NPDES-permitted facilities in the basin, land use characteristics, and
population growth potential) and available water quality data. Drainage
basin characteristics for the remaining rivers that discharge into Puget
Sound are summarized in Table C8.
Sampling Frequency—Bimonthly
—Six additional sampling periods during high-flow
(storm) events.
C-70
-------
TABLE C8. PUGET SOUND DISCHARGE CHARACTERISTICS
Average Average3
River Basin Basin Annual
Area Flow
(mi 2) (ft3/sec) F
Snohomish River Basin
Skykomish River
Sultan River
Woods Creek
Snoqualmie River
North Fork
Tolt River
Middle Fork
South Fork
Raging River
Lower Basin
Pilchuck River
Marshland Canal
French Creek
Stillagaumish River Basin
North Fork
South Fork
Jim Creek
Lower Stillagaumish
Pilchuck Creek
Nooksack River Basin
North Fork
Middle Fork
South Fork
Lower Nooksack
Anderson Creek
Bertrand/Fishtrap Creeks
Tenmile/Wiser Lakes
Puyallup River Basin
White River
Carbon River
South Prairie Creek
Hylebos Creek
Wapato Creek
Nisqually River Basin
Muck Creek
Mashel River
Horn-Tanwax Creek
Ohop Creek
1820
844
110
65
693
100
85
170
82
31
274
125
24
28
684
284
255
46
62
76
758
285
101
192
180
14
74
41
972
494
138
88
25
10
712
92
84
54
44
10400
7400
800
160
3900
700
600
1180
330
140
450
210
30
50
4600
1900
1900
206
100
300
3900
1600
600
1100
360
40
140
8
3400
1100
500
230
40
15
2500
64
230
90
100
Land Useb (*)
i
orest
88
94
96
87
91
92
88
74
81
99
57
80
1
41
88
94
95
92
42
89
80
97
99
94
30
69
13
21
88
89
96
93
45
13
82
59
96
71
88
Agric.
6
2
<1
9
5
<1
10
4
8
<1
24
9
89
46
8
4
3
4
39
6
16
1
<1
3
60
26
77
71
5
5
2
4
8
26
6
14
1
13
7
Range
60 to9 Sed.h
'0 Pop. Load
lange ( 1000
(1) rnt/yr)
50
47
48
38
52
32
68
30
53
46
69
70
70
58
33
39
57
59
37
11
14
-8
-8
3
25
-8
27
28
27
48
8
-13
10
4
55
24
12
12
0
462
16
41
527
113
-------
TABLE C8. (Continued)
Average Average3
River Basin Basin Annual
Area Flow
(ml2) (ft3/sec) F
Sammamish- Cedar Basin
Cedar River
Sammamish River
Issaquah Creek
Swamp/Bear/North Creeks
Lake Washington
Elwha River Basin
Skokomish River Basin
South Fork
North Fork
Deschutes River Basin
Dosewallips River Basin
Duckabush River Basin
Hamma Hamma Basin
Dungeness River Basin
Samish River Basin
Quilcene River Basin
607
188
167
60
70
164
321
240
116
128
170
116
77
85
198
106
101
1800
700
330
140
90
200
1500
2400
730
115
810
800
540
500
460
190
180
:orest
56
88
66
70
44
15
97
95
97
93
78
98
98
99
86
58
96
Agrlc.
5
2
11
24
9
2
<1
1
2
<1
7
<1
<1
0
11
31
1
Land Useb (5
Range Dev.
<1 30
<1 6
<1 17
<1 10
<1 47
0 62
<1 <1
<1 1
<1 <1
<1 1
7 7
<1 1
<1 1
0 1
1 2
5 4
<1 2
0
Storage
8
3
5
4
<1
22
1
3
<1
6
0.7
<1
<1
<1
<1
1
°
Spec.6
Disch. Pop.'
'rec1p.c 1980d («3/sec Density.
(In) Pop. per ml2) (per mi^ )
54
83
49
53
38
37
112
131
145
150
57
97
113
110
62
49
51
728000
30000
164000
18400
85000
449000
2300
1800
<100
400
75000
500
500
500
12200
5000
500
3
4
2
2
1
1
5
10
6
1
5
7
7
6
2
2
2
1199.3
159.6
982.0
306.7
1214.3
2737.8
7.2
7.5
0.9
3.1
441.2
4.3
6.5
5.9
61.6
47.2
5.0
1
NPDES Permit 19
Major Minor
1
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
12
2
2
1
7
1
1
0
1
0
0
0
0
1
0
1
960 to9 Sed.h
70 Pop. Load
hange (1000
(X) mt/yr)
27
53
79
69
230
-35
32
20
47
36
6
6
21
36
8
6
30
144
5.5
27.5
14.1
10.8
5.5
aWilliams et al. (1985a,b).
bPuget Sound Task Force (1970).
cSoil Conservation Service (1965).
dU.S. Department of Commerce (1980).
eSpecific discharge = average annual discharge/basin area.
fPopulation density = total population/discharge basin area.
9U.S. Department of Commerce (1970).
''URS (unpublished).
HUH. bkagit River and Green-Duwamish River Basin characteristics shown in Tables C6 and C7.
-------
This sampling scheme has been selected as a practical compromise to
achieve long-term monitoring objectives within the constraints of program
management and logistics. The rationale for the recommended sampling frequency
is provided below.
Mass loading is typically directly related to streamflow, with the
largest loadings occuring during periods of peak flow. Therefore, to quantify
the annual mass loading of contaminants to Puget Sound from river discharges,
more frequent sampling is required during high-flow periods. Rivers whose
drainage basins extend into the mountainous regions experience two separate
peak flows per year (two upper panels of Figure C7). The first peak occurs
during the winter months between November and February due to heavy runoff
from precipitation in the lower basin. The second peak occurs in the spring
between May and July and is caused by snowmelt runoff from the mountains.
In contrast, river discharge from the lowland areas bordering Puget Sound
does not reflect spring snowmelt runoff. Therefore, lowland rivers experience
a single peak discharge caused by runoff from heavy winter rainfall.
Samples should also be collected during baseflow periods to determine
compliance with established water quality criteria. Low-flow conditions
in Puget Sound rivers typically occur during the late summer and early
fall (Figure C7).
The sampling schedule has been designed to satisfy the requirements
for evaluation of both average water quality during low-flow periods and
contaminant loading. Bimonthly sampling will provide the information needed
to establish baseline conditions and to detect trends in water quality.
Six additional samples will be collected during high-flow conditions to
characterize peak contaminant loading and to capture the bulk of the annual
contaminant loading to the sound. Emphasis should be placed on collecting
samples during extreme runoff events.
Sampling time requirements have been estimated assuming that a continuous-
flow centrifuge is available to separate the particulate material. Approxi-
mately 5 g of particulate materials is required to complete the organic
C-71
-------
30,000-
j
600-
400-
200-
Duckabush River
, Hamma
' Hamma River
Oungeness River
J FMAMJJASOND
MONTH
20,000-
10,000 -
Skagit River
Snohomish River
Nooksack River
Puyallup River
Ehwrta River
JFMAMJJASOND
MONTH
3,000-,
2,000-
1,000
JFMAMJJASOND
Nisqually River
Gmen-Ouwamisfi River
SkokorroM River
Cedar River
Dwchutes River
Sammamisti River
Figure C7. Flow hydrographs for mountain streams and
lowland rivers in Puget Sound Basin.
-------
analyses (Tetra Tech 1986k). Assuming an average centrifuge flow capacity
of 2-6 L/min (Tetra Tech 19861) and an average suspended solids concentration
in the water sample of 10-50 mg/L, the time required to collect the organic
particulate sample is 1-5 h. Therefore, between 5 and 10 stations could
be sampled each week (per centrifuge). One centrifuge would be sufficient
to conduct the bimonthly sampling at the 25-26 monitoring stations sampled
each year. However, at least one additional centrifuge would probably
be required to conduct the high-flow event sampling at all the river stations.
A pilot study is recommended to test the centrifugation procedures
(i.e., optimum flow, suspended solids recovery, and contaminant recovery)
and to better determine the sampling time requirements. Of practical concern
is the ease in recovering particulate samples from the continuous-flow
centrifuge and the time required to clean the equipment between stations.
Estimates of the down-time between stations for individual centrifuges
ranges from 20 min to 2 h (Tetra Tech 19861). It would also be useful
to review the work of other organizations that have used continuous-flow
systems to conduct field monitoring programs. For example, Environment
Canada has been using continuous-flow centri fugation techniques to sample
contaminants in the Niagara River (Kuntz et al. 1982; Ongley and Blackford
1982; McCrea and Fischer 1984).
Timing—Every year at eight primary stations
--Three consecutive years on a rotating basis at 13 primary
stations and 11 secondary (upstream) stations
—Same time of day at a given station
—During at least six high-flow events each year.
Program Evaluation Strategy—The proposed river monitoring program
will be implemented in two phases: the baseline phase and the routine
ambient monitoring phase. In the first phase, the design described above
will be implemented to provide baseline data on water quality and contaminant
loading at a variety of sampling stations. The second phase will involve
implementation of a routine program, with a specific list of target variables
for each basin and possibly a modified list of sampling stations. The
C-72
-------
second phase design will represent a modification of the initial design
based on evaluation of baseline data. The first (baseline) phase will
require a year or less of sampling for evaluation of target variables,
and 3 yr of sampling for evaluation of station locations. To identify
any changes in the water quality variables of concern, the baseline design
proposed above should be repeated periodically, e.g., every 5-10 yr.
A tiered approach to sampling design helps tailor the program to specific
problem areas and reduce costs of the routine monitoring program. The
major program components that may require adjustment are selection of target
variables and station location. Evaluation strategies for both components
are discussed below.
Selection of Target Variables—The selection of target variables may
be modified to address specific problem chemicals within individual drainage
basins. The original list of target chemicals to be analyzed in the river
monitoring program has been developed during the design phase to cover
a broad range of chemical contaminants in addition to conventional water
quality parameters. The analysis of conventional water parameters at each
station will likely continue throughout the routine monitoring programs.
At individual river stations, concentrations of organic contaminants and
metals may be low enough to conclude that the river is not a significant
source of the contaminant to Puget Sound. Such contaminant analyses should
be eliminated from the program to maximize effective use of the analytical
budget.
Two separate approaches could be taken to fine-tune the list of target
variables at each monitoring station. The recommended approach involves
initiating the monitoring program as described above. Sampling results
from the first-year baseline study would be evaluated to determine the
relative importance of individual contaminants. Under this program, the
sampling would begin in the summer months to evaluate low-flow water quality
conditions. Contaminants that are undetected during three successive low-
flow events would then be analyzed in a single high-flow event, and if
C-73
-------
not detected, would be dropped from the program. This analysis would be
conducted on a station by station basis.
An alternative approach would involve conducting a 1-yr screening
study before initiating the monitoring program. Under this approach, complete
analyses of all target variables would be conducted at each monitoring
station, but at a lower sampling frequency (e.g., during two separate low-
flow and high-flow events). Contaminants that are consistently below analytical
detection limits and do not result in high loadings to Puget Sound could
be dropped from the list of variables to be analyzed at a station.
In both screening approaches, contaminants that are detected infrequently
should undergo further review before they are eliminated from the routine
monitoring program. Also, it is recommended that chemicals that have been
dropped from the analysis during the initial screening be checked periodically
(every 5 yr) to confirm that they are not present. Finally, target variables
should be eliminated only if it is cost-effective to do so. In many cases,
whole groups of related compounds (e.g., pesticides, PAH) would have to
b.e eliminated from the analytical scheme to realize a cost savings.
Station Locations—The data generated by the routine monitoring program
should be reviewed periodically to evaluate the station network. For example,
contaminant loadings measured at primary river stations should be compared
to available data on the receiving water environment to evaluate the relative
importance of individual river systems. If the data show that a particular
river does not contribute a significant loading to Puget Sound, then the
sampling frequency at that station should be reduced. If the results from
one of the eight core monitoring stations show that contaminant loading
is relatively minor, the station could be removed from the core program
and placed on the 3-yr rotating sampling schedule. Primary river stations
already on the 3-yr rotating schedule that are not found to contribute
significant contaminant loadings to the sound could be dropped from the
monitoring program. This strategy would make more stations available for
intensive surveys in the problem drainage basins. However, abandoned stations
should be periodically checked (e.g., sample for 1-yr period every 6 yr)
C-74
-------
to ensure that contaminant loadings have not increased as a result of develop-
ment in the basin.
Program evaluation should also focus on identifying river basins that
require additional sampling at upstream stations to locate specific contaminant
sources. If primary station data indicate that the river is contributing
a significant contaminant load to the sound, then more intensive sampling
in the basin is recommended. Selection of rivers requiring intensive sampling
should also take into account the conditions in the receiving environment.
Data to be Reported—
t Flow
t Concentrations of conventional constituents and toxic chemicals
(see Table C3 above)
• Alkalinity, pH, conductivity, temperature
• Station location, date, time
t All field and laboratory QA/QC procedures and results.
Data Analyses--
• Comparisons of temporal changes in water quality at individual
river stations
t Determination of compliance with existing water quality
criteria
t Comparisons of contaminant concentrations and loading along
the length of each river
C-75
-------
• Comparisons of annual mass loading of each contaminant among
stations
t Estimation of total riverine contaminant loading to Puget
Sound
• Determination of relative contaminant loading from individual
sub-basins, and identification of the major contributors
to contaminant loading for source identification and control
activities.
Approaches to Quantification of Contaminant Loadings—Possible approaches
to estimating contaminant mass loading from rivers to Puget Sound include:
t Range of annual loading—Estimate only upper and lower bounds
on annual mass loading for each contaminant during each
year.
• Annual loading—Estimate mean or total annual loading by
one of the following approaches:
Direct calculation—Direct use of measured values of
flow and contaminant concentrations using one of five
models evaluated by Walling and Webb (1985), including
models of Verhoff et al. (1980), Rodda and Jones (1983),
Walling and Webb (1981), and Ongley (1973).
Rating functions—Use of log-linear regressions of
contaminant concentration (or loading) versus flow
to estimate daily concentration (or daily loading)
combined with daily flow estimates to generate annual
loading estimates (e.g., Ferguson 1986).
C-76
-------
Ratio estimator—Use of a ratio of the mean of measured
loadings to the mean of corresponding flows, combined
with the mean of all daily flows to,obtain an estimate
of mean loading (Do!an et al. 1981).
Probability sampling—Sampling with probability proportional
to size of load (Thomas 1985).
• Monthly loading—Estimate mean or total monthly loading
by selected approaches presented above.
Each of these approaches has advantages and disadvantages. Loading estimation
methods are discussed below with respect to monitoring program objectives,
relative amount of data required, relative bias, and relative precision.
Although no single author has evaluated all approaches, Oolan et al. (1981)
and Walling and Webb (1985) provide relatively comprehensive evaluations.
The objectives of the river monitoring program require that annual
mass loading of contaminants from freshwater inflow to Puget Sound be
determined. Because estimation of the approximate range of loading would
provide relatively little information for detecting trends and for mass
balance modeling, reliance on only range estimates would be insufficient.
On the other hand, estimates of monthly loading may not be required, although
they would possibly be useful for identification of discrete nonpoint sources
through temporal analysis of river quality matched with modeling or direct
measurement of source loading.
Data requirements would generally be highest for estimation of monthly
loading, somewhat lower for estimation of annual loading, and lowest for
estimation of the approximate range of annual loading. Estimation of monthly
loading by probability sampling (cf. Thomas 1985) or direct calculation
(cf. Walling and Webb 1985) would require substantial amounts of data,
perhaps as many as 30 data points per month. Use of rating functions or
the ratio estimator to develop monthly loading estimates might be feasible.
Data requirements for application of the rating function approach would
C-77
-------
be lower than for use of the ratio estimator. However, derivation of rating
curves would still require monthly sampling at a minimum. Even with monthly
sampling, data from at least 2-3 yr of sampling might be required before
relatively accurate estimates of annual mass loading could be obtained.
Detection of long-term trends in contaminant loading would require periodic
updating of rating curves.
Relatively large bias could result from estimating the approximate
range of annual loading or from estimating monthly loading. In each case,
the amount of data required for accurate estimates is large, and cost con-
straints could preclude sufficient sampling. For example, if sampling
occurred only twice per year for range estimation (e.g., low-flow and high-
flow periods), this approach would have a large inherent bias.
Relative bias of methods for estimating annual mass loading of suspended
sediment by direct calculation is shown in Figure C8 (Walling and Webb
1985). Based on an extensive data set comprising hourly estimates of suspended
sediment concentration and river flow over a 2-yr period, the actual measured
annual loading is shown by the arrow in Figure C8. By abstracting replicate
subsets of data from the entire data set, Walling and Webb (1985) investigated
the bias and precision of monitoring programs using relatively infrequent
sampling. The frequency distributions of annual loading estimates for
the subsets of data resulting from 7-day, 14-day, and 28-day sampling fre-
quencies are illustrated by the histograms in Figure C8. The relative
bias, as represented by the horizontal distance between the arrow and the
mode of the histogram, is generally least for Methods 2 and 5, which incorporate
a discharge-weighted mean concentration (Table C9). However, these methods
provide relatively low precision, as illustrated by the broad spread of
their histograms (Figure C8). Methods 1 and 4 provide relatively precise
estimates of annual loading, but the latter is substantially underestimated.
Using daily measurements of flow and total phosphorus concentration
from the Grand River, Michigan, Dolan et al. (1981) selected replicate
subsets of data to investigate precision and bias of mean annual loading
estimates. The actual measured load was calculated from the entire year-
C-78
-------
60-i
30-
SAMPLING FREQUENCY
7 DAY 14 DAY
ACTUAL LOAD ACTUAL LOAD
1 '
METHOD 1
8
UJ
O
UJ
c
oc
3
O
a
o
u.
O
O
Ul
o
Ul
oc
60-
30-
S
'
60-
30-
60-
30-
_p^
i i
i rrfl
ap
METHOD 2
METHOD 3
METHOD 4
METHOD 5
28 DAY
ACTUAL LOAD
n-n-n
50 100 150 0 50 100 150 0 50 100 150
SUSPENDED SEDIMENT LOAD 1978-1980 (tonnes x 103)
Reference: Walling and Webb (1 985).
Figure C8. Calculated estimates of suspended sediment loading
compared with actual measured loading.
-------
TABLE C9. SUSPENDED SEDIMENT LOADING ESTIMATION PROCEDURES
EVALUATED BY WALLING AND WEBB (1985)
Method Numerical Procedure Examples of Use
Total load = K
n Ct\ / n 0\
s n I I * n I Verhoff et al. (1980)
i=l n / \i=l n/
n /CQ \
Total load = K 21 -±-- ) Rodda and Jones (1983)
i=l \ /
n / - \
Total load = K I IC.Q J Walling and Webb (1981)
- / n Ci\
Total load = KQn I I -1) Ongley (1973)
r \i=l n /
Total load H LJ: |Q Verhoff et al. (1980)
A^A
I " I r
V AQi /
K = Conversion factor to take account of period of record.
C-j = Instantaneous concentration associated with individual samples.
Qi = Instantaneous discharge at time of sampling.
Qr = Mean discharge for period of record.
Qp = Mean discharge for interval between samples.
n = Number of samples.
-------
long data set. For each replicate subset of data, 25 values (sampling
dates) were randomly selected from the full data set. The loading estimation
methods investigated by Dolan et al. (1981) were of three types:
t Direct calculation—Simple methods of direct calculation
that combined daily flow measurements and concentration
values that were either daily concentrations [(same as Method 2
of Walling and Webb (1985) shown in Table C9], monthly average
concentrations, quarterly average concentrations, semiannual
average concentrations, or annual average concentration.
• Rating function—Log-linear regressions of concentration
against flow (i.e., a rating function approach) to estimate
daily concentrations, which were combined with daily flow
using Method 2 of Table C9 to estimate mean annual loading.
Alternative rating functions were used by deriving either
a quarterly, semiannual, annual, or flow-stratified (high-
flow, low-flow) regression.
• Ratio estimator—Stratified ratio estimator as follows:
mx
where:
My = Estimated load
MX = Mean daily flow for the year
my = Mean daily loading for the days on which concentrations were
determined
C-79
-------
mx = Mean daily flow for the days on which concentrations were determined
n = Number of days on which concentrations were determined
1 ^n
S = . ,. 5. x.y. - n m m
c 2 _ 1 " 2 2
x ~ FfvTT .s xi ~ n mx
x-j = Individual measured flows
y-j = Daily loading for each day on which concentration was determined.
This approach was termed the "stratified" ratio estimator because the data
were stratified by high-flow and low-flow periods before applying the model.
Dolan et al. (1981) found that the stratified ratio estimator exhibited
the least bias and the most precision (least variance) of all methods.
Among the simple methods of direct calculation, bias was generally positively
correlated with precision. Moreover, for the simple methods, bias increasd
with the increase in the time period over which the concentration was averaged.
Among rating function approaches, semiannual or annual regressions gave
the best results.
Thomas (1985) showed that the probability sampling approach could
provide an unbiased estimate of total annual suspended sediment loading.
Precision of the technique relative to other methods is unknown.
Based on this evaluation, it is recommended that either a rating function
approach or stratified ratio estimator be used to estimate contaminant
mass loading. When data from the Puget Sound rivers monitoring program
become available, methods of determining annual mass loading of contaminants
should be evaluated further. Because rating functions show promise for
C-80
-------
further application, rating curves derived from data on rivers of the Puget
Sound Basin are discussed below.
Rating Curves—Rating curves are generated as an aid in predicting
chemical concentration and contaminant loading at individual river stations
under various flow conditions. Many natural streams exhibit distinctive
relationships between certain water quality variables and flow. The concen-
tration of a dissolved chemical is frequently inversely related to discharge.
The concentrations of total suspended solids and contaminants associated
with particles often increase with flows. Regardless of the form of the
relationship between concentration and flow, contaminant mass loading is
usually positively correlated with discharge. In either case, the relationship
is unique at each river station.
Examples of rating curves are shown in Figure C9 where metals concen-
trations are plotted versus discharge for two stations in the Green-Duwamish
River Basin (e.g., mouth of Springbrook Creek and Green River at Renton).
The Springbrook Creek station shows a possible inverse relationship between
metals concentration and flow at the Green River station. Springbrook
Creek drains an area of about 23 mi^. There are no impoundments in the
basin that could affect the concentration-discharge relationship. A different
situation exists at the Green River station. Howard Hanson Dam, located
on the upper Green River, completely regulates flow to the lower basin
which interferes with the concentration-discharge relationship at mainstem
stations below the dam.
Typical rating curves for metals loading are also shown in Figure C9.
The curves were generated using copper, lead, and zinc data and discharge
records from 1980 to 1985 at several stations in the Green-Duwamish River
Basin. Although metals concentration was highly variable and did not exhibit
any correlation with discharge, metals loading consistently showed a direct
relationship to flow. This indicates that metals loading in the Green-
Duwamish Basin is primarily a function of river discharge. In addition,
the rating curves developed for metals loading did not appear to be affected
by upstream impoundments. A direct relationship between loading and discharge
C-81
-------
Standard error - 2S0 - 0.388 log units
O
UJ
u
O
UJ
5
UJ
CO
o
UJ
o
UJ
o.
co
CO
2,000
1.000
500
200
100
50
20
10
5
2
1
0.5
1
£
o
3
O
Ul
CO
Q
UJ
O
IU
O.
CO
CO
200.000
100,000
50,000
20,000
10,000
5,000
2.000
1.000
500
200
100
50
FLOW (n3/sec)
FLOW(R/9ec)
Reference: Nelson (1977).
Figure C9. Suspended sediment rating curves for Snoqualmie River near Carnation.
-------
was apparent at monitoring stations located on unregulated streams, as
well as those below major impoundments.
Unlike many chemical constituents, suspended sediment concentrations
are typically directly proportional to discharge. An example of a suspended
sediment rating curve for the Snoqualmie River near Carnation is shown
in Figure CIO. In this case, both suspended sediment concentration and
load vary directly with flow.
Assuming characteristic rating curves can be developed for the river
monitoring stations, then the sampling frequency at these stations could
be reduced, as long as daily flow records are maintained. Once a rating
curve is generated, then loading estimates, and in some cases contaminant
concentration estimates, can be made based entirely on flow. However,
periodic sampling (e.g., every third year) would still be required to monitor
the accuracy of the rating curve.
Replication and Statistical Sensitivity—Past monitoring programs for
rivers have generally used a single grab or composite sample to characterize
water quality. Precise measurements of most conventional water quality varia-
bles is possible. Homer et al. (1986) evaluated various sampling designs for
conventional variables in several watersheds and generally concluded that
collection of a single sample was the most cost-effective approach. For toxic
chemicals, sample replication at every station and sampling period is not
cost-effective. To minimize variance, a single composite sample should be
collected by the Equal Transit Rate method following protocols given below.
For sampling, the river is divided transversely into equal width-increments.
The sampler is lowered from the surface to the bottom and back at a uniform
speed (generally less than 20 percent of the mean current velocity).
Protocols—
t Field references: Guy and Normal (1970), Stevens et al. (1980),
Richey et al. (1986), Tetra Tech (1986c), Ongly and Blackford
(1982), Kuntz et al. (1982), McCrea and Fischer (1984)
C-82
-------
SPRINGBROOK CREEK
GREEN RIVER AT RENTON
a
I
o
800
600-
400
200-
o o
B 0
O " O
00
a
o o
0 20 40 60 80 100 120 140
FLOW(ft3/sec)
i
8
•o
JQ
2
£
100
80
60-
40
20
a o
a
a a
6000
2000 4000 6000 8000 10,000
FLOW(ft3/sec)
Figure C10. Comparison of metals rating curves at two stations in the Green-Duwamish
River Basin.
-------
• Laboratory: Strickland and Parsons (1972), Tetra Tech
(1986c,f,g), Tetra Tech and EVS Consultants (1986b).
C-83
-------
HABITAT MONITORING DESIGN
HABITAT TYPES
Variables and Rationale—The distribution of habitat types within Puget
Sound is of fundamental importance to the structure and function of the Puget
Sound ecosystem. Information on changes in habitat distributions may be
important for assessing possible anthropogenic impacts on habitats and their
resident biota and for interpreting other monitoring data. Habitats at greatest
risk from the effects of development and pollution are those in intertidal
and shallow subtidal zones, and riparian zone habitats associated with river
systems. Prior to mapping, habitats should be defined and classified to be
consistent with the Washington Cosatal Zone Atlas to the extent possible.
Specific habitats that should be monitored include wetlands (e.g.,
salt marshes, freshwater marshes), kelp beds, eelgrass beds, and major
intertidal substrate types (e.g., mud flats, rocky intertidal areas, cobble
and sand beaches). Other habitats of concern may be selected by the inter-
agency management group responsible for the final monitoring design. Criteria
for selecting specific specific habitats/locations may include: 1) areal
extent in Puget Sound, 2) sensitivity to disturbance, 3) species richness,
4) importance of use by fish, birds, and mammals, and 5) relative use by
humans for commercial or recreational purposes.
Habitat types should be monitored with satellite images and aerial
photographs, coordinated with ground surveys (ground truthing) to determine
and improve the accuracy of image data. Satellite images should have a
minimum resolution of 30 m. Aerial photographs of the Puget Sound shoreline
are taken annually by the U.S. COE. However, these photographs are not
necessarily taken at low tide, and may not be optimal for surveying marine
habitats. The Department of Natural Resources aerial photos should be
examined to supplement the U.S. COE photos.
C-84
-------
King County is currently developing a wildlife habitat management program
that involves classifying, defining, mapping, and monitoring all wildlife
habitat types occurring in the county. Saltwater, freshwater, and riparian
habitats are included. The Landsat imagery Thematic Mapping system is being
used with ground truthing for mapping and will be used for monitoring changes
in habitat distribution in the future. Information from the King County
project should be incorporated into the Puget Sound database.
Survey Area—Because of the importance of habitat information for
land-use planning, management of aquatic resources, and impact assessment,
a sound-wide effort is warranted. Satellite images and aerial photographs
will be obtained for all of the shoreline of Puget Sound and selected habitats
associated with major rivers of the Puget Sound Basin.
Frequency—Every 5 yr.
Surveys should be conducted on a rotational basis, such that about
one-fifth of the sound is analyzed each year. Ground surveys should also
be made on a rotational basis for a variety of sites, and should correspond
with those parts of the sound that are being analyzed that year.
Timing—Summer (low tide).
Satellite and aerial images will be collected on a low tide during the
summer, so as much intertidal habitat as possible is exposed. In particular,
the resolution of the extent of eelgrass and kelp beds is best at low tide.
Late July is the preferred time because the development of the kelp Nereocystis
is maximal at that time.
Data to be Reported—
• Areal extent of habitat by habitat type for major areas
(e.g., Padilla Bay, Nisqually Delta)
• Total areal extent of habitat by habitat type by county.
C-85
-------
Data Analysis—
• Changes in area! extent of habitat over time
• Correlation of significant changes with other events (e.g.,
El Nino event, oil spills, changes in contaminant loading).
Replication and Statistical Sensitivity—There will be no replication
for satellite or aerial photos of acceptable quality. Areal resolution
and accuracy of habitat identifications from the photos will be determined
by ground truthing. Studies are presently being undertaken by Thomas Mumford
of the Washington Department of Natural Resources (WDNR) to determine the
level of ground truthing required for the collection of accurate data.
Reasonably accurate estimates of resources needed for ground truthing LANDSAT
and aerial photo data will be available following completion of his studies.
Protocols—No standardized protocols for Puget Sound studies are available.
Protocols should be developed based on results of the ongoing projects
by WDNR and King County, and a review of relevant literature (e.g., Still
and Shih 1985).
C-86
-------
COLLECTION OF ANCILLARY DATA
Ancillary data comprise a variety of information necessary for inter-
pretation of the ambient monitoring data, compliance monitoring data, and
intensive survey data. All ancillary data discussed below are presently
collected by various agencies as part of ongoing programs. Hence, these
data will be assembled and summarized as part of the Puget Sound Monitoring
Program. The categories of ancillary data are:
t Climate/weather
• Fisheries harvest
• Waterfowl harvest
t Aquaculture
t Demographic and socioeconomic conditions
• Decision record-keeping.
CLIMATE/WEATHER
Variables and Rationale—Monthly and annual summaries of available
climatological data (i.e., precipitation, wind speed and direction, temperature,
hours of daylight and percent cloud cover) will be reviewed to discern
trends in local and regional weather patterns. Climatological trends are
important because they may affect the interpretation of data collected
within other monitoring components. For example, El Nino events alter
the characteristics of oceanic water entering Puget Sound. This, in turn,
could have an effect on Puget Sound biota. Abnormally long spells of dry
or wet weather may also have implications for monitoring because of changes
in freshwater input to the Sound. This would be especially true in estuaries.
C-87
-------
Stations—Olympia
—Port Angeles
—Bel 1ingham
—Sea-Tac Airport.
Frequency—Dai 1y.
Data to be Reported—An annual report will be prepared that contains
monthly and annual climatological data for the Puget Sound Basin. The
report will contain a section summarizing climatological anomalies (e.g.,
drought, floods, and El Nino events), and major storm events. Summaries
of appropriate statistics will be made for the following variables:
• Wind speed (and direction)
• Precipitation
0 Hours of daylight
• Air temperature
• Percent cloud cover.
Data Analyses—Quarterly reviews will be performed of:
• The National Weather Service's monthly weather summaries
• The State Climatologist's data.
FISHERIES HARVEST
Variables and Rationale—Catch statistics will be assembled for all
commercially and recreationally important species, including (at a minimum):
C-88
-------
Salmonids
Chinook salmon
Coho salmon
Pink salmon
Chum salmon
Sockeye salmon
Steel head trout
Rainbow trout
Dolly Varden
Shellfish
Butter clams
Littleneck clams
Horse clams
Cockles
Manila clams
Dungeness crab
Red rock crab
Shrimp
Oysters
Abalone
Geoduck
Bottom fish Bait fish
Halibut Herring
English sole Surf smelt
Rock sole Anchovy
Petrale sole Sand lance
Dover sole
Sand sole
Rex sole
Butter sole
Starry flounder
Arrowtooth flounder
Sable fish
Surfperch
Dogfish
Rockfish
Lingcod
Pacific cod
Pollock
Whiting
Other species to be considered include sea urchin and octopus.
Harvest information is important to meet agency obligations and for
the interpretation of the other monitoring data collected during this program.
Harvest information may also be important for assessing the biological
impacts of contamination. For example, commercial shellfish harvest may
vary considerably among various areas of Puget Sound, depending on the
degree of bacterial contamination of the water column in those areas.
Similarly, the sizes of wild stocks of salmonids are closely related to
the availability of adequate spawning habitat.
Record Review—Trends in fisheries harvests will be summarized by
areas of Puget Sound. Data sources for preparing the summary include the
Washington Department of Fisheries annual statistical report (and other
relevant reports or data), the Washington Department of Game Steel head
C-89
-------
report, and areal surveys. A mechanism will be established to ensure that
catch data are reported in a manner useful for evaluating monitoring data.
WATERFOWL HARVEST
Variables and Rationale—Waterfowl harvest statistics are reported
by hunting area by the Washington Department of Game for the following
species:
Ducks
Mallard Shoveler Ring-necked duck
Gadwall Pintail Goldeneye
Wigeon Wood duck Bufflehead
Green-winged teal Redhead Ruddy duck
Blue-winged teal Canvasback Scoters
Cinnamon teal Scaup Mergansers
• Geese
Large Canada Lesser Canada
Dusky Canada Greater White-fronted
Cackling Canada Snow
Harvest statistics are derived from hunter surveys. These data are usefu
to the monitoring program for three primary reasons:
t They provide an indication of the value of the Puget Sound
estuaries as waterfowl habitat and as an associated recreational
resource
C-90
-------
t A downward trend (e.g., 3-4 yr or more) in harvest may indicate
the possibility of adverse impacts from contamination or
habitat disturbance.
Record Review—Temporal trends in waterfowl harvests will be summarized
to provide information useful for monitoring program review. Where an
unexpected downward trend is evidenced, more detailed at data analyses
may be warranted (e.g., a review of individual survey forms by area).
AQUACULTURE SITES AND YIELDS
Variables and Rationale—As part of a new program, the Washington
Department of Fisheries will be collecting a variety of production-related
data for all aquaculture activities having to do with the following organisms:
• Crayfish
• Trout and other freshwater finfish species
• Salmon
• Oysters
• Clams
• Geoducks
• Mussels
• Marine algae.
Variables that will be reported include the age and type of operation,
quarterly yields, and species composition of aqualculture operations.
This information is valuable to the monitoring program because it:
C-91
-------
• Provides information about Puget Sound's value to industries
dependent on water
• Provides an additional monitoring tool for detecting possible
decreases in water quality.
The Washington Department of Natural Resources is the agency responsible
for leasing subtidal lands in the public domain. However, it does not
routinely collect information on aquaculture sites and yields.
Record Review and Reporting—Yearly statistical summary reports will
be compiled and reviewed to discern any trends within aquaculture areas
of Puget Sound. The results of this review will be reported to provide
information for evaluating water quality.
DEMOGRAPHIC AND SOCIOECONOMIC CONDITIONS
Variables and Rationale—Records will be compiled and tracked through
time for several variables, including:
• Land use and zoning, especially in the coastal zone and
along rivers and streams. The program will be designed
for early recognition of negative trends such as encroachment
on wetlands, displacement of water-dependent uses of shorelines,
and reduction of public access.
• Construction permit awards, especially those for substantial
development in shoreline areas and in rivers and streams.
The program will be designed to ensure that permit awards
are consistent with the intent of the Shoreline Management
Act and local Shoreline Master Plans (see items under land
use and zoning, above).
• Population by census tract, especially in coastal areas
and along rivers and streams. The program will be designed
C-92
-------
so that potentially negative effects of demographic trends
can be identified before they become serious. For example,
rapid residential growth in a rural area may foreshadow
a potential increase in bacterial contamination.
• Employment by economic sector, as an indicator of contaminant
loading. For example, an increase in employment by electronics
manufacturer may result in an increased discharges of solvents
to municipal sewage treatment systems.
Changes in demography and economic activity are important because anthropogenic
impacts to Puget Sound are strongly influenced by the distribution of people
and their economic activities. For example, areas not served by sewers
and agricultural areas may significantly impact downstream commercial shellfish
rearing areas because of bacterial loading. Toxic contamination of the
sediments and biota may be associated with heavily populated or industrialized
areas.
Record Review and Reporting—A yearly review of demographic and socio-
economic records will be undertaken to define trends in demography and
economic activity. An annual survey will be conducted to inventory changes
to land use master plans and local codes and ordinances that could have
significant environmental consequences (e.g., revisions to accomodate previously
unallowed uses, or revisions to a grading and drainage ordinance) Shoreline
use records of the Washington Department of Ecology will be reviewed annually.
These records include the following information on shoreline permits:
location, type of project, land use and environmental designation, and
type of permit (e.g., substantial development, conditional, or variable).
Projects with potential significant environmental impacts, and potential
cumulative or far-field impacts of projects will be noted in the annual
review report. Available census summary data will be reviewed by census
tract to identify trends in population distribution and density, especially
in coastal zone areas. Trends on employment and unemployment by economic
sector can be derived annually from reports generated by the Puget Sound
Council of Governments.
C-93
-------
DECISION RECORD-KEEPING
Variables and Rationale—A record will be kept of all relevant regulatory
and resource management decisions that could potentially affect the Puget
Sound environment, beneficial uses of its resources, or the design of the
monitoring program and interpretation of monitoring program data. Relevant
decisions include those related to:
• Environmentally Significant Projects. The Environmental
Impact Statement process provides good records for tracking
all projects that are deemed to have significant environmental
impacts. Also, permits are issued for a variety of activities,
including aquatic disposal of effluent (NPDES permits),
dredging and filling (Clean Water Act, Section 404 U.S. COE
permits), storage and handling of hazardous or toxic substances
and acquaculture (WDNR submerged lands leases). Records
will be kept of trends in the number and location of significant
projects by type of activity so that potentially significant
trends can be identified.
• Planning activities. Planning is undertaken by all levels
of government and can indicate trends that are significant
to the monitoring program. Records will be kept on planning
activities such as land use planning (see section on demographic
and socioeconomic conditions), public utility planning
(especially wastewater treatment and disposal) and water
quality planning. Records will include annual summaries
of plan content, activities (e.g., regulation, resource
management), and decisions carried out under the plan.
• Institutional and legal activities. A variety of institutional
and legal activities will be recorded so that relevant infor-
mation can be channeled to key organizations and key managers
of the monitoring program. Changes in local ordinances
C-94
-------
(e.g., implementation of grading and drainage codes) and
state statutes (e.g., Shoreline Management Act amendments),
and reorganization of government bodies (e.g., foundation
of a new agency or reorganization of a branch of government),
may have far reaching consequences to the monitoring program,
and environmental management in Puget Sound. An annual
summary of institutional changes will be analyzed for relevant
implications.
Record Review and Reporting—Regulatory and resource management agencies,
and branches of federal, state, and local government now keep detailed
records of permit-related decisions, planning activities, institutional
changes, and legal changes within the Puget Sound area. Record-keeping
of these activities will consist of reviewing available reports (e.g.,
Washington Department of Ecology NPDES actions) from appropriate offices
and personnel at least quarterly. However, direct contacts with agency
personnel will also be necessary. A summary report that identifies major
decisions and activities will be produced quarterly.
C-95
------- |