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 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

-------