United States       Office of Water       November 1982
            Environmental Protection   Program Operations (WH 546)  430/9-82-010
            Agency          Washington DC 20460
&EPA      Design of 301 (h)
            Monitoring Programs
            for Municipal
            Wastewater  Discharges
            to Marine Waters

-------
    DESIGN OF 301(h) MONITORING PROGRAMS FOR
MUNICIPAL WASTEWATER DISCHARGES TO MARINE WATERS
                 November,  1982
                       by

            Tetra Tech, Inc., Staff



           Contract Number 68-01-5906
                Project Officer

                Paul Pan, Ph.D.
        Environmental Protection Agency
            Washington, D.C.  20460
                Tetra Tech, Inc.
           1900 - 116th Avenue, N.E.
          Bellevue, Washington  98004

-------
                             EPA  REVIEW NOTICE
    This report was prepared  under the direction of the  Office of Marine
Discharge Evaluation (WH-546), Office of Water Program Operations, Office  of
Water, U.S.  Environmental Protection  Agency, 401 M Street,  S.W.,  Washington
D.C., 20460, (202)  755-9231.

    This report has been reviewed  by  the Office of Water and the Office  of
Research and Development, U.S. Environmental Protection Agency,  and approved
for publication.   Mention of  trade names  or commercial  products does not
constitute endorsement  or recommendation for use.

-------
                             TABLE OF CONTENTS
LIST OF FIGURES                                                         v
LIST OF TABLES                                                          v1

CHAPTER I                                                                1
OVERVIEW OF MONITORING REQUIREMENTS                                      1
     INTRODUCTION                                                        1
     OBJECTIVES OF MONITORING                                            3
          301(h) Requirements                                            4
          State Requirements                                             5
          NPDES Requirements                                             5
          Other Requirements                                             7

CHAPTER II                                                               8
TREATMENT PLANT AND EFFLUENT MONITORING                                  8
     OBJECTIVES                                                          8
     SPECIFICATIONS FOR TREATMENT PLANT AND EFFLUENT MONITORING          9
          Sampling Locations                                             9
          Parameters                                                    10
          Sample Collection and Frequency                               10
          Analytical Methods for Toxic Substances                       12
          Toxic Substance Data Reporting                                12

CHAPTER III                                                             22
RECEIVING WATER QUALITY AND SEDIMENT MONITORING                         22
     OBJECTIVES                                                         22
     SPECIFICATIONS FOR WATER AND SEDIMENT MONITORING                   22
          Station Locations                                             22
          Variables and Sampling Frequencies                            25
          Sampling and Analytical Methods                               28
          Oceanographic Measurements                                    30
          Data Analysis and Reporting                                   34

CHAPTER IV                                                              36
BIOLOGICAL MONITORING                                                   36
     OBJECTIVES                                                         36
     APPROACH AND RATIONALE                                             37
     SPECIFICATIONS FOR BIOLOGICAL MONITORING                           39
          Sampling of Biological Communities                            39
          Station Locations                                             41
          Sampling Frequency and Replication                            45
          Sample Collection and Processing                              46
          Analytical Techniques                                         75
          Data Reporting                                                90
                                    111

-------
CHAPTER V                                                               92
QUALITY CONTROL                                                         92
     APPROACH AND RATIONALE                                             92
     FIELD ACTIVITIES                                                   93
     QUANTITATIVE ERROR ANALYSIS                                        97
     TOXIC POLLUTANT ANALYSIS                                          108

APPENDIX A                                                             110
OCEANOGRAPHIC METHODS                                                  110
     CURRENT METERS                                                    110
          Uses of Current Meters                                       110
          Types of Current Meters                                      111
     DROGUES AND DRIFTERS                                              113
          Use of Drogues                                               113
          Types of Drogues                                             114
          Uses of Surface Drifters                                     115
          Types of Surface Drifters                                    116
          Use of Seabed Drifters                                       117
          Types of Seabed Drifters                                     118
     DYE STUDIES                                                       118
     SPECIFICATIONS FOR FIELD WORK                                     119
          Current Meters                                               119
          Drogues and Drifters                                         120
          Dye Studies                                                  120
          Navigation-Position Determination                            122

REFERENCES                                                             124

-------
                              LIST OF FIGURES


Number

   1   Representative sampling locations for two levels of
       biological monitoring                                            44

   2   Examples of graphical displays of biological  data from
       a marine sewage discharge site                                   78

   3   Oceanographlc surface observations log sheet                     96

   4   Example Precision Control Chart                                 100

   5   Example Accuracy Control Chart                                  102

-------
                               LIST OF TABLES
Number                                                                 Page
   1   Monitoring Required by California Ocean  Plan                       6
   2   Containers, Preservation,  and Holding Times for  Selected
       Groups of Toxic Chemicals                                         13
   3   List of Approved Analytical  Methods for  Selected Toxic
       Chemicals (Priority Pollutants)                                  15
   4   Example of Station Location  Descriptions for  301(h)
       Compliance Monitoring                                            26
   5   Recommended Sample Preservation and Storage Requirements
       for Water Quality                                                29
   6   Recommended Analytical Methods                                   31
   7   Selection of Appropriate Fish Sampling Gear                       55
   8   Examples of Some Nonparametric Statistical Tests                 81
   9   A List of Commonly-Used Indices of Diversity                      86
  10   A List of Some Common Polychaetes That Have Been
       Associated with Marine or  Estuarine Pollution                    89
  11   Recommended Procedures for Determination of Systematic
       Bias and Precision in Analytical  Methods                         98
  12   Summary of Expressions Necessary to Construct Control
       Charts                                                          104

-------
                                CHAPTER I

                    OVERVIEW OF MONITORING REQUIREMENTS
 INTRODUCTION

     Under Section 301(h) of the Clean Water Act of 1977 as  amended by the
 Municipal  Wastewater Treatment  Construction Grant Amendments  of  1981,
 publicly owned treatment works (POTWs) may apply  for  a variance from the
 secondary treatment  requirements for discharge into marine waters.   Each
 applicant  is  required to submit a detailed technical  evaluation of the
 discharge  and its effects on  the marine environment to demonstrate
 compliance  with the seven statutory criteria  listed under Section 301(h).
 If a variance  were granted,  monitoring would be  required  [Section 301(h)(3)]
 to assess  the impact of the  modified discharge on marine biota.   EPA
 regulations implementing Section  301(h) are set forth  in 40 CFR Part 125,
 Subpart G,  as  amended in November, 1982.

     The guidance provided in this document  has been developed to help meet
 the general monitoring requirements of  the 301(h)  program.  References to
 applicable  water quality standards and requirements are not intended to
 replace specific state requirements.  Applicants  must  also  check with the
 appropriate  state and  local agencies  for any  specific monitoring
 requirements applicable to their circumstances.

     This document was prepared  in order to provide guidance for designing
 monitoring  programs that will meet regulatory requirements (40 CFR 125.62)
 and allow  continuing assessment of  the  impact  of less-than-secondary
 discharges  on the  receiving  water  marine  environment.   It provides
 supplemental guidance on designing monitoring programs  to that included in
 the Revised Section  301(h)  Technical  Support  Document (Tetra  Tech  1982)
which is available by writing to the Office of Marine Discharge  Evaluation
 (WH-546),  U.S.  Environmental  Protection Agency,  401 M  Street, S.W.,
Washington, D.C., 20460.   The guidance provided  in these  documents  is

-------
advisory only;  its use is not required.   However,  EPA believes that Section
301(h) applicants will benefit substantially  by  following the guidance  and
procedures  provided in these documents.

     The amended 301(h)  regulations include  a  number of changes  to  the
monitoring  program requirements  contained in the original 301(h) regulations
promulgated in  1979.  While the basic objectives  of the overall  monitoring
requirements remain the  same, many of the  original  detailed requirements
were deleted from the amended regulations  so  that each applicant will have
the flexibility to design  a  cost-effective monitoring program to meet  its
individual  circumstances.  This  is especially true  for small applicants that
discharge  into depths greater than 10 meters  with  negligible  seabed
accumulation of suspended solids [40 CFR 125.62(b)(2)].

     Much of the guidance in this  document is  directed towards large
dischargers.   It covers  a  wide range of  possibilities that might  be
encountered when developing  301 (h) monitoring programs, including complex
waste streams and discharges into sensitive ecosystems.

     Users  of this guidance document  should keep  in mind that the level  of
effort for  each 301(h) monitoring program must be keyed to the individual
circumstances of each discharge  and corresponding  receiving water situation.
A monitoring program will  not have  to  be as  extensive  for smaller
dischargers  as  for large  dischargers.   A monitoring  program for a waste
discharge comprised primarily  of domestic wastes does not  have to be  as
comprehensive as a program to monitor the  impact  of a discharge with large
amounts  of  industrial and/or  toxic  wastes.  The frequency of sampling
required for a  resilient, high energy, or  otherwise nonsensitive receiving
water environment will be considerably less than  for a sensitive ecosystem.
A minimally  acceptable monitoring program, then,  will be based on a balance
of several  factors, including the size  of the discharge, the character of
the waste,  and the sensitivity and  variability  of the  receiving water
environment.   In addition,  a test of monitoring program practicability
should include consideration of the technical  feasibility of  available
measurement  procedures during a  variety of weather  and sea conditions.

     Those  EPA tentative decision  documents which  recommend Section 301(h)
variances will  highlight site-specific items which  must be addressed in  the

-------
applicant's proposed monitoring program.  The 301(h)  decision document
should,  therefore, be analyzed carefully and the  monitoring requirements
therein  reflected in the design  of the applicant's final monitoring program.
The Technical Evaluation or Technical  Review Reports on individual 301(h)
applications should also  be  used as a reference  in the design of final
monitoring  program proposals.

OBJECTIVES  OF MONITORING

     Monitoring programs  under 40 CFR 125.62 for dischargers receiving
modified NPDES permits under Section 301(h) of the Clean Water Act should be
designed to:

     •    Document short- and  long-term effects of  the  discharge on
          receiving water,  sediments, and biota; also, on beneficial
          uses of the receiving water

     t    Determine compliance with NPDES permit terms and conditions

     •    Assess  the effectiveness  of toxic control  programs.

While  divided into  general  biological, water  quality, and effluent
monitoring components, in general,  the monitoring program should focus  upon
demonstrating the  discharge's  compliance  with  applicable  standards and
permit  conditions,  and demonstrating  predictable  relationships  between
discharge characteristics and impacts upon  the  marine receiving water
quality and the marine  biota.   Although each general monitoring component
may  involve sampling at  different  locations for different variables and at
different times,  it should not be considered as a separate  and individual
activity, but as  an integrated study.   In this manner, the permittee should
be able to gain  the most meaningful data on an assessment of  the impacts of
the  discharge.   Further, once  an adequate background data  base  is
established and predictable relationships  among the   biological, water
quality, and effluent monitoring variables  are  demonstrated,  it should be
possible for many 301(h) permittees,  especially those with small  discharges,
to scale down the intensity of certain elements of their  field monitoring
studies.

-------
     Applicants may wish to  expand their monitoring programs  beyond the
 minimum required to further  demonstrate the impact  or  lack of impact of
 their discharge on  the environment.   In  addition,  they may wish to exploit
 unique opportunities provided by 301(h)  to  add  to  the body of knowledge on
 effects of marine discharges on the  receiving water environment and marine
 ecosystems.  The potential  benefits  to  the  municipality would be in
 assessing  long-range  wastewater  treatment  and disposal  needs  and
 alternatives.   Additionally, applicants  discharging in the same geographic
 proximity may  wish  to  develop an areawide  assessment of marine discharge
 environmental impacts.   Applicants  should  consider,  also, that  the
 monitoring data provided will  be  used by EPA to  assess  whether 301(h)
 variances  should  be renewed following expiration of  the  initial
 vari ances/permi ts.

 301(h) Requirements

     The monitoring requirements  specified by 40 CFR  Part 125.62 provide  for
 monitoring programs comprised  of  three elements:   (1) biological monitoring,
 (2) water quality monitoring, and (3) effluent monitoring.   In addition,
 applicants must  demonstrate  in their  monitoring program proposals that  they
 possess the economic, personnel,  technical, and other resources necessary to
 implement their proposed programs.  The  biological and water quality
 sampling must  be able to detect variations over time and space  as those
 changes relate to the permittee's discharge.  Monitoring must be conducted
 at the current  discharge  site before  and after any improvements  are
 implemented  and at the site  of new or relocated discharges.  Sampling times
 should include critical environmental periods and both typical  and unusual
meteorological or oceanographic conditions.   Biological  programs for large
permittees and some small  permittees  must include field surveys of  affected
or potentially affected biota,  bioaccumulation studies,  and an  assessment of
the condition  and productivity of  commercial and  recreational  fisheries.
Water quality  samples must be  from stations selected  to assess  compliance
with  water quality standards in the vicinity  of the zone of initial  dilution
 (ZID),  and beyond the ZID.  The toxics monitoring program must  determine the
effectiveness  of industrial  pretreatment and nonindustrial toxics control
programs.   An adequate  toxics monitoring  program will  also  aid  the
implementation of  toxics  control programs  and  the biological  monitoring
efforts.

-------
     The  applicants must provide EPA with  sufficient  information on quality
assurance3 and control  procedures to document  compliance with  accepted
scientific  practice.  The monitoring plan,  therefore,  should discuss quality
assurance in  general, and specific  data  generation  sections of the  plan
should reflect individual details of quality control.

State Requirements

     The  monitoring program  must  document compliance with all applicable
water quality standards.  Some states  have specific monitoring requirements
and/or recommendations on parameters to  be  sampled, station  locations,
sampling  frequencies, and analytical methods.   Table  1, for example,  shows
the parameters recommended by the California  Ocean Plan guidelines (CSWRCB
1972  and 1978).   The parameters  included  in  most  State standards  for
receiving waters are dissolved oxygen,  pH, coliform bacteria, and suspended
solids or a surrogate.   The standards have  been expressed  as  a maximum
allowable  pollutant concentration,  a maximum  allowable deviation  from
background concentrations, a  statistically significant difference between
stations, or as a  prescription against  harm to biota or degradation of
beneficial  uses of  the water body.

NPDES Requirements

     An NPDES permit is required, by Section 402 of the Clean Water Act, for
all discharges of  pollutants to navigable waters.   The permits specify
effluent limitations  plus effluent sample types (e.g., grab  or 24-hour
composite)  and sampling frequencies for assessing compliance.   In some  cases
the permits include receiving water  monitoring  requirements.  The  NPDES
requirements  should be  used  by 301(h) applicants as a  basis of decision
making on  influent and effluent monitoring.  NPDES  permit  sampling
a EPA policy  initiated by the Administrator in a memorandum, dated  May  30,
1979, stipulates that all  environmental  monitoring  and measurement efforts
mandated or supported  by EPA must  have quality assurance project plans (see
Chapter V, Quality Control).

-------
   TABLE  1.   MONITORING  REQUIRED  BY CALIFORNIA OCEAN  PLAN
                                       Location of Monltorinq
Parameter
Flow
Bacteriological
Grease and Oil
Floatinq Participates
Suspended Solids
Settleable Solids
Turbidity
pH
Arsenic
Cadmium
Total Chromium
Copper
Lead
Mercury
Nickel
Silver
Zinc
Cyanide
Phenolic Compounds
Total Chlorine Residual
Ammonia Nitrogen
Total Identifiable
Chlorinated Hydrocarbon
Toxicity
Radioactivity
Salinity
Temperature
Biochemical Oxygen Demand
Total Phosphate
Total Nitrogen
Dissolved Oxygen
Discoloration
Light Transmittance
Water
Supply







X
X
X
X
X
X
X
X
X
X

X



X

X



X
X



Untreated
Wastewater
X

X

X


X
X
X
X
X
X
X
X
X
X
X
X



X
X
X


X
X
X



Effluent
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X


X
X
X

X
X
X
X



Receiving
Water

X
X
x
X


X












X




X
X



X
X
X
Sediments








X
X
X
X
x
x
x
x
X
x
x



X

X


x





Fish and Macroinvertebrates

Sediment Sulphides
Particle Size Distribution
Benthic Biota
NOTE:   X means monitoring required.

Source:  California  State Water Resources Control Board (1972, 1978).

-------
specifications will be  supplemented (usually more  frequent monitoring or
addition  of parameters)  to  meet 301(h) objectives and/or state requirements.
Additional  requirements to  meet 301(h) objectives  will  consider 301(h)
related effluent limitations,  plant flow characteristics,  initial dilution
ratios, receiving water characteristics,  and biological communities and
beneficial uses to be protected.

Other Requirements

     For  each discharge,  an investigation  should determine if there are
other water quality standards applicable to the given water body or other
monitoring program requirements.  For example, the  Interstate  Sanitation
Commission has requirements  relating to wastewater  discharges  in the New
York Harbor area.  In California, basin  plans developed by regional boards
or agencies  contain requirements in  addition  to those found  in the
California Ocean Plan.

-------
                              CHAPTER II

                 TREATMENT PLANT AND EFFLUENT MONITORING
OBJECTIVES

     Plant  monitoring (influent and effluent) is primarily required to
determine  compliance with  NPDES permit conditions and water quality
standards.  In addition, influent and  effluent monitoring provides
indicators  for assessment of treatment plant performance.  High effluent
pollutant concentrations  may be due to  plant  malfunctions or overloads;
thus, plant monitoring  can  be used to identify problems and improve
performance.  Effluent monitoring  also  provides information on waste
characteristics and  flows  for use in  interpreting water quality and
biological  data.

     Monitoring programs for  toxic substances and pesticides are required as
part of  the 301{h)  regulations and should  be designed to:

     •   Determine the  potential  for toxicity to aquatic life and
         risk to  human  health from toxic  chemical  substances
         discharged to marine waters, and to

     •   Evaluate  the effectiveness of  industrial source  control
         pretreatment programs and nonindustrial  toxic  control
         programs.

The first objective can be attained by measuring toxic chemical  substances
1n the effluent  and in  selected samples taken from the receiving water
sediments and organisms  used  in  biomonitoring  protocols.  The second
objective can be  attained by  measuring toxic  substances in the treatment
plant influent and comparing  trends.   Sources  of  toxicants, whether
industrial or nonindustrial, are analyzed and  identified as part of the
toxics control  program.
                                  8

-------
SPECIFICATIONS FOR TREATMENT PLANT AND EFFLUENT MONITORING

     Major  considerations in the design of plant  sampling programs  include
the specification of:   sampling locations, parameters  to be measured,
collection  and analytical methods, and sampling frequencies.

Sampling Locations

     Specific locations  at  the  treatment plant for  influent and effluent
sampling may  be  included in the NPDES permit.   In  the 301(h) monitoring
program only  influent and effluent sampling  points are specified, although
sampling at  intermediate  points  within the plant may  be useful for
monitoring individual  treatment unit performance.   Sampling at  various
points in the collection system  may  also  be necessary to isolate sources of
toxic substances.

     For conventional  pollutants and  nutrients,  influent  samples should
generally  be collected just downstream of the coarse  screens or grit
chamber.  If multiple waste streams enter the plant and a representative
sample cannot be collected,  a flow-composite sample may be used for influent
analysis.   Effluent  samples  should be collected  downstream of any
chlorination  or disinfection units.  Samples should be taken as close to the
start of the  outfall as possible.  An example  of such  a sampling point would
be the effluent pumping station.  Separate samples  should be taken if two
outfalls are  used and the  effluent which enters the outfalls comes from
different parts of the treatment plant.  When  emergency bypasses are  made to
a different outfall  or  discharge point,  due  to high inflows or treatment
plant problems, separate samples of the bypassed flows should be taken.

     Sampling for  toxic pollutants should  include hourly grab samples
collected over a 24-hour period and composited in proportion to the flow.
Influent samples should be taken upstream of  the  plant intake works (prior
to the grit chamber, if possible)  and the  total (unfiltered) sample should
be analyzed.  Effluents should be sampled after treatment and just prior to
entering the outfall pipe.   If  the  effluent  is chlorinated,  samples should
be taken upstream and downstream of the chlorination unit.

-------
Parameters

     The treatment plant monitoring  parameters required by an NPDES permit
for a typical  large discharger  Include conventional pollutants, nutrients,
and toxicants.  Influent monitoring  normally includes volumetric flow rate,
BOD5, suspended solids,  pH,  and grease  and oil.   The suspended solids and
BODjj measurements are  used to determine removal efficiencies and to detect
changes in the character of the waste stream.   Measurements of pH and grease
and oil are used to determine the  need for and success of any pretreatment
programs.  Other influent  monitoring  parameters  which may be required are
total phosphorus, total nitrogen or  specific  forms of nitrogen, settleable
solids, COD, and  temperature.   These variables  may be needed to further
characterize the influent and to monitor treatment plant performance.

     The parameters to be measured in  the effluent include requirements of
the water quality  standards,  NPDES permit,  and  any additional variables
needed to  interpret water quality  and biological  data.   Plant  effluent
monitoring should normally include volumetric flow rate, dissolved oxygen,
BOD5, suspended  solids, settleable solids,  temperature,  total and fecal
coliform bacteria,  grease and oil,  and pH.   If the effluent is chlorinated,
total chlorine residual is  typically monitored.   To aid in evaluating the
Importance of chlorination  in forming  persistent, possibly hazardous,
chloro-organics,  a careful record of  the total mass  of chlorine  used per
unit of flow should be reported.   If other forms of disinfection are used,
type and dosage should be reported.   Other variables which may be  required
include floating particulates,  total  phosphorus, total nitrogen, ammonia or
other forms of nitrogen, and COD.

     In addition  to the above parameters,  Section 125.62(d) of the amended
301 (h) regulations requires  each applicant,  to  the extent practicable,  to
monitor toxic  substances and pesticides  [see 40 CFR 125.8(u) and (m)] in the
effluent.   A list of the 129 toxic  pollutants is provided later in Table 3.

Sample Collection and Frequency

     The type  of sampling equipment to be used and  sampling  frequencies
depend on the  size and nature of the  discharge.  Sampling frequency  and type
of sample should  be determined based on the  variability of the influent and
                                   10

-------
effluent characteristics and the flow  so  that the data  collected will be
representative of the discharge.  In  general,  volumetric  flow rates of the
influent  and effluent  should  be measured  continuously using  automatic
equipment.   Hourly and average daily  flow  rates should be recorded.  Daily
effluent and influent samples for BOD and suspended solids should be taken.
Twenty-four hour flow-composite  samples are  recommended.   Nutrient sampling
may be  done weekly or  monthly  using  grab  samples selected randomly or
24-hour flow-composite samples  collected on randomly selected days during
the sampling period.  Measurements of  effluent pH should be done on daily
grab samples taken at different times each day.   Daily grab samples are
typically taken for total and fecal  coliform  bacteria.

     Generally, a randomly selected date within a defined sampling period,
coordinated with other sampling  (i.e.,  for conventional  pollutants) during
wet and dry flow periods,  should  be chosen for influent and effluent
sampling for toxic pollutants.   Flows caused by bypass events  at the POTWs
should also be considered for sampling and  analysis.

     The sampling frequency for toxic pollutants depends on such  factors as
the size and location of the discharge, the types and quantities of toxic
pollutants present, and  the sensitivity and beneficial  uses of the receiving
water marine environment.  More frequent sampling should occur for POTWs
with large discharges, significant types and  quantities of toxic pollutants,
and sensitive receiving  waters.   All  large POTWs and those small  POTWs  that
cannot certify they have no known or suspected  sources of toxic  pollutants
or pesticides need to establish  baseline analyses for toxic pollutants and
pesticides present in their current discharge  (40 CFR Part 125.64).  Toxic
substance monitoring  is required of all  301(h) waiver  recipients  to  help
verify the type  and quantity of  the compounds identified in the  discharger's
301(h) application and to determine if significant  changes occur over time.

     It is  recommended that at least annual  representative  wet and dry
weather 24-hour composite sampling  and  analyses be undertaken.   More
frequent sampling and  analyses may  be required depending on  the  type of
substances found  in the  wastewater  and discharge  or the sensitivity of the
receiving waters.
                                   11

-------
      More  frequent plant monitoring may be necessary for both conventional
 and toxic  pollutants during  the first year of  the program to obtain reliable
 estimates  of when maximum waste load periods occur and the magnitude of peak
 concentrations.  In some cases the  relationship between concentrations of a
 variable (e.g., BOD or suspended solids)  and volumetric flow rates is not
 well  known.  Sampling at times of minimum, average, and maximum hourly flow
 rates  on  a monthly basis  for  the first  year  should  help define
 concentration-flow rate  relationships and  allow better interpretation of the
 receiving water quality  data.

 Analytical  Methods for Toxic Substances

      Regulations have  been  proposed  on allowable  holding  times  and
 analytical  procedures for  toxic  substances [45 Fed.  Reg. No.  231
 p.  79318-79379 (November 28,  1980)].   The final regulations  establishing
 test  procedures  for the analysis  of  toxic  substances  have  not yet been
 published.  Recommended holding  times, container  requirements,  and
 preservation methods  are  listed in Table 2.  Recommended analytical methods
 are shown in Table 3  for the priority pollutants.   Analytical methods for
 the six pesticides (methoxychlor,  mi rex, guthion, malathion, parathion,  and
 demeton) can be found  in Watts (1980) and U.S.  EPA (1978).

     For many dischargers it  will  be  necessary to contract  with outside
 laboratories for  the analytical work.  The laboratories  selected should be
 state certified  according  to U.S.  EPA approved procedures.  Sampling,
 holding and analysis procedures,  and equipment should  comply with state  and
 federally approved methods.

Toxic Substance Data Reporting

     Quality assurance information  should be  transferred quarterly  to  EPA
and should  include copies of  quality control  charts used in the laboratory
and the  results of replicate, split, spiked, and blank  sample analyses.   The
laboratory  results submitted  should  include the calibration standards  used,
copies of the calibration curves used,  and  the frequency  of calibration
runs.
                                  12

-------
                        TABLE 2.   CONTAINERS, PRESERVATION, AND HOLDING TIMES
                               FOR SELECTED GROUPS OF TOXIC CHEMICALS
Parameter Container3
Metals
Chromium VI P»G
Mercury P»G
Metals in Table 3 (except above) P,G
Asbestos p
Cyanide (total and amenable p,G
to chlorination)
Organic Compounds
Extractable (including phthalates, G, teflon-
nitrosamines, organochlorine lined cap
pesticides, PCBs, nitroaromatics,
isophorone, polynuclear aromatic
hydrocarbons, haloethers, chlori-
nated hydrocarbons and TCDD)
Extractables (phenols) G, teflon-
1 ined cap
Purqeables (halocarbons and G, teflon-
Preservation
Cool , 4°C
HN03 to pH > 2
0.05% K2Cr20?
HN03 to pH > 2
1 ml 2.71% HgCl2
Cool , 4°C
NaOH to pH > 12
0.008% Na2S203e

Cool, 4°C
0.008% Na2S203
Cool, 4°C
H?S04 to pH > 2
07008% Na2S203e
Cool , 4°C
Maximum
Holding Time
24 hours
28 days
6 months
5 days
14 days

7 days
(until extraction)
30 days
(after extraction)
7 days
(until extraction)
30 days
(after extraction)
14 days
and aromatics)
1ined septum
0.008%

-------
 TABLE 2.   (Continued)
Parameter
Purgeables (acrolein and
acrylonitrile)
Pesticides


Phenols

Container3
G, teflon-
1 ined septum
G, teflon-
lined cap


P,G

Preservation
Cool, 4°C
0.008% Na2$203
Cool, 4°C
0.008% Na2S203e


Cool, 4°C
H2S04 to pH > 2
Maximum
Holding Time
3 days
7 days
(until extraction)
30 days
(after extraction)
28 days

   Polyethylene (P) or Glass (G).

   Sample preservation should be performed  immediately upon sample collection.   For composite  samples each
alnquot should be preserved at the time  of collection.  When use of an automatic sampler makeliT Impossible

Sm??^6^!^04' ^^ S3mPleS ^ be PreS6rVed by m3intainin9 at 4ttC  until  compositing and sample
C  Samples should be analyzed as soon as possible after collection.  The times  listed are  the maximum times
that samples may be held before analysis and still considered valid.   Samples may  be held  for longer periods

 -6                                                                    "
Some samples may not be stable  for the maximum time period given in the table.  A permittee or
                                   sanple for a shorter "™ if knowud9e  exists ?° show ^
  Guidance applies to samples to be analyzed by GC, HPLC,  or GC/MS  for  specific organic compounds.
p
  Should only be used in  the presence of residual  chlorine.

NOTE:  If preservative is unavailable for organic  compounds, recommended holding time is .48 hours at 4°C.

Source:   U.S.  Environmental Protection Agency (1979a).

-------
            TABLE 3.   LIST OF APPROVED ANALYTICAL METHODS FOR SELECTED
                      TOXIC CHEMICALS (PRIORITY POLLUTANTS)
Parameter and Units

  1.  Acenaphthene, ug/1
  2.  Acrolein, ug/1
  3.  Acrylonitrile, ug/1
  4.  Benzene, ug/1
  5.  Benzidine, ug/1

  6.  Carbon tetrachloride (tetra
       chloromethane), ug/1
Methods (EPA Method Number)

GC or HPLC (610), GC/MS (625)
GC or HPLC (603), GC/MS (624)
GC or HPLC (603), GC/MS (624)
GC (602), GC/MS (624)
HPLC (605), Oxidation-
  col ormetric, GC/MS (625)
GC (601), GC/MS (624)
     Chlorinated Benzenes (other
       than dichlorobenzenes)

  7.  Chlorobenzene, ug/1
  8.  1,2,4-trichlorobenzene, ug/1
  9.  Hexachlorobenzene, ug/1
GC (601), (602), GC/MS (624)
GC (612), GC/MS (625)
GC (612), GC/MS (625)
    Chlorinated Ethanes

10.   1,2-dichloroethane, ug/1
11.   1,1,1, trichloroethane ug/1
12.   Hexachloroethane, ug/1
13.   1,1-dichloroethane, ug/1
14.   1,1,2-trichloroethane, ug/1
15.   1,1,2,2-tetrachloroethane, ug/1
16.   Chloroethane, ug/1

    Chloroalkylethers (chloromethyl,
      chloroethyl  and mixed ethers)

17.   Bis (chloromethyl) ether3
18.   Bis (2-chloroethyl) ether, ug/1
19.   2-chloroethyl vinyl ether
      (mixed),ug/1
                                                  GC (601),
                                                  GC (601),
                                                  GC (612),
                                                  GC (601),
                                                  GC (601),
                                                  GC (601),
                                                  GC (601),
          GC/MS (624)
          GC/MS (624)
          GC/MS (625)
          GC/MS (624)
          GC/MS (624)
          GC/MS (624)
          GC/MS (624)
                                                  GC (611), GC/MS (625)
                                                  GC (601), GC/MS (624)
     Chlorinated Naphthalene

 20.  2-chloronaphthalene
GC  (612), GC/MS  (625)
                                        15

-------
TABLE 3.  (Continued)
     Chlorinated Phenols (other than those
       listed elsewhere, includes trichloro-
       phenols and chlorinated cresols)

 21.   2,4,6-trichlorophenol, ug/1
 22.   Para-chloro meta-cresol, ug/1
 23.   Chloroform (trichloromethane), ug/1
 24.   2-chlorophenol,  ug/1
 GC (604), GC/MS (625)
 GC (604 , GC/MS  625)
 GC (601), GC/MS (624)
 GC (604), GC/MS (625)
     Dichlorobenzenes

 25.   1,2-dichlorobenzene,  ug/1
 26.   1,3-dichlorobenzene,  ug/1
 27.   1,4-dichlorobenzene,  ug/1
 GC  (601,  602,  612),  GC/MS (625)
 GC  (601,  602,  612),  GC/MS (625)
 GC  (601,  602,  612),  GC/MS (625
     Dichlorobenzidine

 28.   3,3-dichlorobenzidine,  ug/1
HPLC  (605),  GC/MS  (625)
    Dichloroethylenes

29.   1,1-dichloroethylene, ug/1
30.   1,2-trans-dichloroethylene, ug/1
31.   2,4-dichlorophenol, ug/1
GC  (601), GC/MS  (624)
GC  (601), GC/MS   624)
GC  (604), GC/MS  (625
    Dichloropropane and Dichloropropene

32.  1,2-dichloropropane, ug/1
33.  1,2-dichloropropylene (1,2-dichloro-
      propene), ug/1
34.  2,4-dimethylphenol
GC (601), GC/MS  (624)
GC (601), GC/MS  (624)

GC (604), GC/MS  (625)
    Dinitrotoluenes

35.  2,4-dinitrotoluene, ug/1
36.  2,6-dinitrotoluene, ug/1
37.  1,2-diphenylhydrazine, ug/1
38.  Ethyl benzene, ug/1
39.  Fluoranthene, ug/1
GC (609), GC/MS (625)
GC (609), GC/MS (625)
GC/MS (625)
GC (602), GC/MS (624)
GC or HPLC (610), GC/MS (625)
                                      16

-------
TABLE 3.  (Continued)
     Haloethers (other than those listed
       elsewhere)	

 40.  4-chlorophenyl phenyl ether, ug/1           GC (611), GC/MS  (625)
 41.  4-bromophenyl phenyl ether, ug/1            GC (611), GC/MS  (625)
 42.  Bis (2-chlorisopropyl) ether, ug/1          GC (611), GC/MS  (625)
 43.  Bis (2-chloroethoxy) methane, ug/1          GC (611), GC/MS  (625)


     Halomethanes  (other than those
       listed elsewhere)	

 44.  Methylene chloride  (dichloromethane), ug/1  GC (601), GC/MS  (624)
 45.  Methyl chloride (chloromethane), ug/1       GC (601), GC/MS  (624)
 46.  Methyl bromide (bromomethane), ug/1         GC (601), GC/MS  (624)
 47.  Bromoform (tribromomethane), ug/1           GC (601), GC/MS  (624)
 48.  Dichlorobromomethane, ug/1                  GC (601), GC/MS  (624)
 49.  Trichlorofluoromethaneb
 50.  Dichlorodifluoromethaneb
 51.  Chlorodibromomethane, ug/1                  GC (601), GC/MS  (624)
 52.  Hexachlorobutadiene, ug/1                   GC (612), GC/MS  (625)
 53.  Hexachlorocyclopentadiene, ug/1             GC (612), GC/MS  (625)
 54.  Isophorone,  ug/1                            GC (609), GC/MS  (625)
 55.  Naphthalene,  ug/1                           GC or  HPLC  (610),  GC/MS  (625)
 56.  Nitrobenzene, ug/1                          GC (609), GC/MS  (625)


     Nitrophenols

 57.  2-nitrophenol, ug/1                         GC (604), GC/MS  (625)
 58.  4-nitrophenol, ug/1                         GC (604), GC/MS  (625)
 59.  2,4-dinitrophenol,  ug/1                     GC (604), GC/MS  (625)
 60.  4,6-dinitro-o-cresol, ug/1                  GC (604), GC/MS  (625)


     Nitrosamines

 61.  N-nitrosodimethylamine,  ug/1                GC (607), GC/MS  (625)
 62.  N-nitrosodiphenylamine,  ug/1                GC (607), GC/MS  (625)
 63.  N-nitrosodi-n-propylamine, ug/1             GC (607), GC/MS  (625)
 64.  Pentachlorophenol,  ug/1                     GC (604), GC/MS  (625)
 65.  Phenol,  ug/1                                GC (604), GC/MS  (625)
                                       17

-------
TABLE 3.  (Continued)
     Phthalate Esters

 66.  Bis (2-ethylhexyl) phthalate,
 67.  Butyl  benzyl phthalate, ug/1
 68.  Di-n-butyl phthalate, ug/1
 69.  Di-n-octyl phthalate, ug/1
 70.  Diethyl  phthalate, ug/1
 71.  Dimethyl phthalate, ug/1
ug/1
GC (606)
GC (606)
GC (606
GC (606
GC (606
GC/MS (625)
GC/MS (625)
GC/MS (625)
GC/MS (625)
GC/MS (625)
              GC (606), GC/MS (625)
     Polynuclear Aromatic Hydrocarbons

 72.   Benzo (a) anthracene (1,2-benzy-
       anthracene), ug/1
 73.   Benzo (a) pyrene (3,4-benzopyrene), ug/1
 74.   3,4-benzofluoranthene, ug/1
 75.   Benzo (k) fluoranthene
       (11,12-benzofluoranthene), ug/1
 76.   Chrysene, ug/1
 77.   Acenaphthylene, ug/1
 78.   Anthracene, ug/1
 79.   Benzo (ghi) perylene (1,12 benzo-
       perylene), ug/1
 80.   Fluorene, ug/1
 81.   Phenanthrene, ug/1
 82.   Dibenzo (a,h) anthracene
       (1,2,5,6-dibenzanthracene),  ug/1
 83.   Indeno (1,2,3-cd)  pyrene,  ug/1
 84.   Pyrene,  ug/1
 85.   Tetrachloroethylene
       (tetrachloroethene),  ug/1
 86.   Toluene, ug/1
 87.   Trichloroethylene
       (trichloroethene), ug/1
 88.   Vinyl  chloride  (chloroethylene), ug/1
              GC or HPLC (610), GC/MS (625)

              GC or HPLC (610), GC/MS (625)
              GC or HPLC (610), GC/MS (625)
              GC or HPLC (610), GC/MS (625)

              GC or HPLC (610), GC/MS (625)
              GC or HPLC (610), GC/MS (625)
              GC or HPLC (610), GC/MS (625)
              GC or HPLC (610), GC/MS (625)

              GC or HPLC (610), GC/MS (625)
              GC or HPLC (610), GC/MS (625)
              GC or HPLC (610), GC/MS (625)

              GC or HPLC (610), GC/MS (625)
              GC or HPLC (610), GC/MS (625)
              GC (601), GC/MS (624)

              GC (602), GC/MS (624)
              GC (601), GC/MS (624)

              GC (601), GC/MS (624)
     Pesticides  and Metabolites

 89.   Aldrin,  ug/1
 90.   Dieldrin,  ug/1
 91.   Chlordane  (technical  mixture
       and  metabolites),  ug/1
              GC (608), GC/MS (625)
              GC (608), GC/MS (625)
              GC (608), GC/MS (625)
                                      18

-------
TABLE 3.  (Continued)
     DDT and Metabolites

 92.  4,4'-DDT, ug/1
 93.  4,4'-DDE (p,p-DDX), ug/1
 94.  4,4'-ODD (p.p-TDE), ug/1
                                            GC  (608),  6C/MS  (625)
                                            GC  (608),  GC/MS  (625)
                                            GC  (608),  GC/MS  (625)
     Endosulfan and Metabolities

 95.  a-endosulfan-Alpha, ug/1
 96.  B-endosulfan-Beta, ug/1
 97.  Endosulfan sulfate, ug/1
                                            GC  (608), GC/MS  (625)
                                            GC  (608), GC/MS  (625)
                                            GC  (608), GC/MS  (625)
     Endrin and Metabolites

 98.  Endrin, ug/1
 99.  Endrin aldehyde, ug/1
                                            GC  (608), GC/MS  (625)
                                            GC  (608), GC/MS  (625)
    Heptachlor and Metabolites

100.  Heptachlor, ug/1
101.  Heptachlor epoxide, ug/1
                                            GC (608), GC/MS  (625)
                                            GC (608), GC/MS  (625)
     Hexachlorocyclohexane (all isomers)

102.  a-BHC-Alpha, ug/1
103.  B-BHC-Beta, ug/1
104.  Y-BHC (lindane) -Gamma, ug/1
105.  6-BHC-Delta, ug/1
     Polychlorinated Biphenyls
106.
107.
108.
109.
110.
111.
112.
PCB-1242 (Aroclor
PCB-1254 (Aroclor
PCB-1221 (Aroclor
PCB-1232 (Aroclor
PCB-1248 (Aroclor
PCB-1260 (Aroclor
PCB-1016 (Aroclor
1242), ug/1
1254), ug/1
1221), ug/1
1232 , ug/1
1248), ug/1
1260), ug/1
1016), ug/1
                                            GC (608), GC/MS (625)
                                            GC (608), GC/MS (625)
                                            GC (608), GC/MS (625)
                                            GC (608), GC/MS (625)
GC (608),
GC (608),
GC (608),
GC (608),
GC (608),
GC (608),
GC (608),
GC/MS (625)
GC/MS (625)
GC/MS (625)
GC/MS (625)
GC/MS (625)
GC/MS (625)
GC/MS (625)
                                       19

-------
TABLE 3.  (Continued)
     Miscellaneous Substances (including
       metals, organic compounds not  listed
       elsewhere, and asbestos)	

113.  Toxaphene, ug/1                             GC  (608),  GC/MS  (625)
114.  Antimony (total), ug/1                       AA  (204.2)
115.  Arsenic (total), ug/1                        AA  (206.2)
116.  Asbestos (fibrous),  chrysotile               TEM
       fibers MFL (million fibers per liter)
117.  Beryllium (total), ug/1                     AA  (210.2)
118.  Cadmium (total), ug/1                        AA  (213.1)
119.  Chromium (total), ug/1                       AA  (218.1  or  .2  or  .3)
       (IV), ug/1                                 AA  (218.4)
120.  Copper (total), ug/1                        AA  (220.2)
121.  Cyanide (total), ug/1                        Titrimetric,  Spectro-
                                                   photometric (335.2)
122.  Lead (total), ug/1                           AA
123.  Mercury (total), ug/1                        AA
239.2)
245.1 or .2)
249.2)
124.  Nickel (total), ug/1                         AA
125.  Selenium (total), ug/1                       AA  (270.2)
126.  Silver (total),ug/1                          AA  (272.2)
127.  Thallium (total), ug/1                       AA  (279.2)
128.  Zinc  (total), ug/1                           AA  (289.2)
129.  2,3,7,8-tetrachlorodibenzo-                 GC/MS (613),  (625)
       p-dioxin (TCCD), ug/1


a Bis  (chloromethyl)  ether was removed  from  the  toxic pollutant list
(U.S. EPA 1981a).

b Dichlorodifluoromethane  and trichlorofluoromethane were removed from  the
toxic pollutant list (U.S.  EPA  1981b).

Note:     GC = Gas chrom  tography.
        HPLC = High performance liquid  chromatography.
       GC/MS = Gas chromatography coupled  with  mass spectrometry.

       Source:  U.S. EPA  (1979a).

          AA = Atomic absorption.

       Source:  U.S. EPA  (1979b).

         TEM = Transmission electron  microscopy.

       Source:  Anderson  and  Long (1980).
                                       20

-------
     The results of screening measurements for the  priority  pollutants in
the effluent  should be included in the  monitoring reports.  The results of
the chemical  analyses should, where possible, be reported as measured rather
than less-than-certain  values.   If  the analytical  results were below the
limit of detection, this should  be  noted on the data  sheet  and the value
given as less than  the  actual limit of detection (e.g., < 10 ug/1).  The
list of compounds identified in  previous  screenings should be compared to
the new results.  The estimated concentration after initial dilution should
be computed for each toxicant.  These values after initial dilution should
be compared to available  criteria for marine waters [45 Fed. Reg. No. 231
pp. 79318-79379  (November  28,  1980)].    Those  compounds which exceed the
criteria should be added to the list  of toxicants  subject  to  bioassay.
Toxicants found which can be bioaccumulated, but not previously included in
the sediment  monitoring program,  should  be added to that program.
                                   21

-------
                               CHAPTER III

              RECEIVING WATER QUALITY AND SEDIMENT MONITORING
OBJECTIVES

     To determine compliance with  water quality  standards and the 301(h)
criteria, the receiving water quality monitoring program  must document water
quality in the vicinity of  the  Zone of Initial  Dilution (ZID) boundary, at
control or reference stations,  and  at areas  beyond the  ZID where discharge
impacts might reasonably  be expected.   Monitoring  must reflect conditions
during all  critical  environmental  periods as identified in the  301{h)
applications.  If currently available  data  are  not adequate to predict when
critical periods will  occur, then greater monitoring effort may be necessary
to demonstrate that water quality data are collected under the appropriately
critical conditions.   Examples of  such  critical conditions are periods of
anadromous fish spawning  runs, juvenile  fish migrations or feedings, high
wastewater loadings,  high water temperature,  and low flushing rate.

     The applicant's historical   sampling  programs,  together with  new
requirements associated  with 301(h)  permit conditions, will  be the most
useful  guide for designing an adequate receiving  water  monitoring program.
Changes in station  locations, parameters, or frequencies may be required to
rectify deficiencies  in historical  programs.

SPECIFICATIONS FOR  WATER AND SEDIMENT MONITORING

Station Locations

     Section 125.61 of the  amended  301(h) regulations  requires that water
quality be maintained to assure the  protection of public water supplies,  the
protection and propagation of a balanced  indigenous population  (BIP) of
shellfish,  fish, and wildlife, and  to allow  recreational  activities.   Under
Section 125.62,  the establishment of a water quality monitoring program is
required which to the extent practicable:

                                   22

-------
     •    Provides adequate  data for evaluating compliance  with
          applicable water quality  standards

     •    Measures the presence  of toxic pollutants which have been
          identified or are reasonably expected  to be  present in the
          discharge.

In order to meet these water quality monitoring  requirements, receiving
water and sediment sample stations  need to be located in the vicinity of the
ZID boundary,  at control  sites,  and 1n Impact  areas  in such a way as to
allow adequate  correlations to be made between water quality, oceanographic
and sediment measurements, and toxic  substances and biological data.  Other
locations  which a state may wish  to  specify  include the shoreline in
swimming and shellfishery areas and within  the  ZID.  Placing stations so
that pollutant  concentration gradients  can be  detected between  the ZID
boundary and control stations may be  valuable for  larger discharges.

     When a discharge is into a saline estuary there is  a greater emphasis
on protecting benthic organisms  within the  ZID,  suggesting that  water
quality data near the seabed and  sediment quality  data may be necessary.  In
the case of oceanic  discharges there  are  general  requirements to prevent
extreme adverse  biological impacts  within the ZID  which  have adverse effects
beyond the ZID;  thus, again some  water  quality monitoring  within the ZID may
be required  for especially sensitive  ecosystems  and/or  large  or
industrialized wastewater systems.

     Criteria  for selection of specific  stations depend on the purpose of
the station.  ZID-boundary  stations  should  be placed  on the upcurrent and
downcurrent boundaries  of the ZID;  they will  not necessarily be at fixed
locations  but  more likely will  be  set on  the  day of sampling based on
observations of  current direction.   Since the objective is to intercept the
waste field drift flow  as it is carried across  the ZID boundary,  several
samples placed  at depth and across  the wastefield  need to  be obtained.   This
will  be necessary in order to compute  an average value and to show a range
of values if the waste field is not uniform.   To  demonstrate that the plume
has indeed been sampled, especially if this  is not evident by the water
quality values  themselves, data on currents, drogue tracks, and/or tracers
need to be provided.

                                  23

-------
     Stations  upcurrent of the ZID  may be the choice for water  quality
control stations,  although with  the gradient  concept in mind,  stations
sufficiently far downcurrent may be satisfactory.  Care should be taken in
selecting control  stations so that values presumably  representing control
conditions do not  include  diluted  wastes carried back  into the area of the
control station by  tidal currents.   Control  stations should be unaffected by
other pollutant sources  as well as  the applicant's discharge.  Also, control
stations should be  located in water of  similar depth as the discharge, with
similar bottom characteristics and similar distances from shore.  Additional
controls may be needed  when the applicant's proposal is  for a relocated
outfall and when the  discharge is into stressed waters.

     Impact area stations  vary from one  discharge site to another.  Areas
where monitoring may  be  required include recreational beaches, diving areas,
shellfish harvesting areas, kelp  beds,  coral  reefs,  commercial  and
recreational fishing grounds, and other distinctive  biological habitats.
The selection of impact  area stations should be based upon a thorough review
of the recreation  and biological  sections  in the  301 (h) application, the
Technical  Evaluation  or  Review Report,  the tentative decision document, and
the draft 301(h) permit.   State requirements  on station  locations need to be
met by the program  (e.g.,  both shore and nearshore stations must be sited to
protect beaches under the California Ocean Plan).

     Additional  stations may need to be located near other pollutant sources
to allow the effects  of  the subject discharge to  be distinguished from these
sources.  Examples  of other  pollutant  sources are areas off the mouths of
major rivers near the discharge,  sludge  disposal areas, and other municipal
and industrial  ocean  discharges.   The need for these stations is identified
by noting the extent  of  influence from  available water quality data, from
analysis of potential  impacts  based on volumetric flow and characteristics
of the discharge, and from analysis of the dispersion characteristics of the
receiving water body.

     All sited stations  should be plotted  on large-scale nautical charts or
15-min quadrangle  sheets  (USGS)  and then transferred to  more convenient
small  scale maps.   Latitude and longitude  should be  determined from the
large maps to the  nearest second.   The approximate  depth at each station
                                   24

-------
should be determined from previous sampling data  or estimated from soundings
shown on the large-scale charts.  Each  station location should be described
relative to the outfall/diffuser,  permanent navigation buoys, or distances
from known shoreline points as shown in  Table 4.  Historical designations
and/or the applicant's  designation should also be noted.  If ZID boundary
stations are occupied  using accurate  navigation methods  there should be
adequate assurance  that the resulting  water quality sampling data reflect
ZID boundary conditions.  To  document  station  locations,  the applicant's
program and periodic monitoring data  reports should describe navigation
(locating) methods  and  field conditions  during  sample collection.

Variables and Sampling  Frequencies

     The  variables to be sampled  in  the receiving water include those
specified in the 301(h) regulations, those required by the state, and those
necessary to evaluate other water quality data.   The variables which should
be included routinely are BODg  dissolved  oxygen,  pH, temperature, salinity,
suspended solids or its surrogates (e.g., light transmittance), total  and
fecal coliform bacteria, and settleable solids.   Light transmittance may be
specified in terms  of turbidity, Secchi disc depth, extinction coefficient,
or percent light transmittance.  The applicant should state the reason(s)
for the light transmittance method(s) selected.   Additional  variables which
may be required are:

     •    Total  nitrogen
          —nitrate
          —nitrite
          —total kjeldahl nitrogen
          —ammonia

     •    Total  phosphorus
          —reactive phosphorus

     t    Chlorophyll £

     •    Floating  particulates

     •    Color.
                                   25

-------
          TABLE  4.    EXAMPLE  OF  STATION  LOCATION  DESCRIPTIONS
                        FOR  301(h)  COMPLIANCE  MONITORING
Station                                                                   Historical  Approximate
 Number  N Latitude W Longitude          General Description               Designation   depth,  roa
1


2

3

4
21°15'10"


21°17'44"

21°17'37"

21°17'46"
157°49'53"


157°52'34"

157°53'22"

157°53'59"
3.14 km directly south of Waikiki
Beach, approx. 2.4 km west-southwest
of Diamond Head Beach Park.
0.79 km directly south of Sand Island
in Honolulu Channel, alonq west edge.
at end of old sewer outfall, 1.07 km
offshore from Sand Island.
along eastern edge of Kalihi Channel,
None


2b

3b

4b
70


12.5

12.fi

12.fi
                               1.0 km  directly southwest of coral  reef,
                               0.7 km  directly west of diffuser.

   5     21°17'44"  157°54'25"  2.74 km directly southeast of Ahua  Pt.,          5b          8.5
                               0.66 km directly west pf Kalihi  Channel,
                               0.60 km directly south of Keehi  Lagoon
                               coral reef.

   6     21°17'0r  157°54'24"  at  the  center of the zone of nixing            None         68
                               48  m north of diffuser within the ZID.

   7     21°17'01"  157°53'59"  ^0.7 km directly east of center of zone       None         70
                               of  mixing, at the eastern edge of ZID.

   8     21°16'56"  157°54'24"  130 meters directly south of center of         None         79
                               zone of mixing at ZID boundary.

   9     21°17'01"  157°55'00"  1.024 km directly west of center of zone       S1-9C       128
                               of  mixing, just outside west edge of
                               zone of mixing.

  10     21°17'16"  157°54'24"  at  north edge of zone of mixing, 0.48 km       S1-6C        31
                               directly north of center of zone of
                               mixing, 0.53 km north of diffuser.

  11     21°17'39" 157°54'57.5" 2.33 km south-southeast of Ahua  Pt.,           None         15
                               1.56 km directly west of Kalihi  Channel,
                               0.83 km directly south of Keehi  Lagoon
                               coral reef.

  12     21°17'08"  157°53'22"  0.91 km directly south of old sewer out-       None         70
                               fall, 1.77 km east of center of  zone of
                               mixing, 1.81 km directly south of coral
                               reef at south end of seaplane runway.

  13     21°17'22"  157°55'18"  2.68 km south of Ahua Pt., 1.69  km west-       None         70
                               northwest of center of zone of mixing,
                               1.54 km directly south of Keehi  Lagoon
                               coral reef.

  14     21°17'54"  157°55'46"  1.64 km directly south of Keehi  Lagoon         Hone         IS
                               Beach,  2.86 km directly northwest of
                               center  of zone of mixing.

  15     21°17'27"  157°52'26"  1.19 km directly south of southernmost         None         18
                               tip of  Sand Island, approx. 500  m south-
                               southeast of entrance to Honolulu Channel.

  16     21°15'H"  157°49'01"  0.76 km southeast of Diamond Head Beach          lb         18
                               Park near lighthouse and Coast Guard Res.

  17     21°17'01"  157°53'48"  1.0 km  directly east of center of zone         None         70
                               of  mixing, just outside eastern  edge of
                               zone of nixing.


a All  depths are  relative to MLLW.

b Applicant's  proposed monitoring station numbers.

c Historical sampling stations of applicant.
                                            26

-------
     Water column profiles  for salinity  and temperature  are  needed to
interpret  dissolved oxygen  data  and may be necessary to predict the location
of the drift flow unless profiling with  tracers or effluent constituents is
to be relied  upon  for finding  the plume.   In any event,  salinity and
temperature may be needed,  together with current speed and direction, to aid
in describing mass  movement of diluted wastes to farfield sites where
impacts must be assessed.   Current directions  are needed to determine where
in the horizontal plane one might expect to  find the drift plume crossing
the ZID boundary.  Current speeds are  important in assessing the vertical
height of  plume rise and,  hence, in establishing where in the vertical plane
samples should be  taken to  measure  plume constituents.   Salinity,
temperature, and currents,  furthermore, need to be  provided to assist in
evaluating the results of benthic  and other biological  responses.   In
estuaries, the amount  of freshwater  inflow  from rivers needs to be
documented as an  adjunct to evaluating  residence times  and  routes of
possible transport of diluted effluents and particulates.  Sampling may be
at generally accepted depths, e.g.,  1 m  below  the water surface,  mid-depth,
1 m above  the  sea bed and at  10-m intervals for depths greater than about 40
m; however, features of water masses  observed  in  profiling for salinity and
temperature (and possibly  light transmittance)  should take precedence in
establishing sample depths.

     Parameters to be measured in the sediments should include particle  size
distribution  and total volatile solids.   Other variables,  such as BOD5,
sulfides,  and  total  organic carbon,  may  be required by the states or may be
important  in analyzing discharge impacts on  benthic  biota.  In addition,
sediment  samples  should be analyzed  annually for toxic  substances and
pesticides identified  in  the plant effluent.  Sediment samples for toxic
substance  analysis should be taken  within the ZID, in the vicinity of the
ZID  boundary, at  representative impact area locations  outside  the ZID
boundary,  and  at control stations.

     Sampling  in estuaries should be conducted  at  slack water as recommended
in the Revised Section 301(h) Technical  Support Document (Tetra Tech 1982).
Where tidal effects are to  be  discriminated, sampling  should  be  done at
several times  over a tidal  cycle for both spring and neap tides. In order
to verify  continuing  compliance with 301(h) criteria,  the ZID boundary
                                  27

-------
 stations should be sampled during  those times of year when the discharge is
 least diluted.

     Sampling frequencies  must be selected  to meet state requirements and
 should  provide data during the critical environmental  period(s)  as
 identified in the 301(h) application.  For sites where available data do not
 define these periods adequately,  receiving  water sampling should be done
 monthly for at least the first year.  In most other cases, quarterly surveys
 which cover the critical period(s) should suffice.  More frequent sampling
 may  be  specified  by  the  states  in swimming and  shellfishery areas  to
 determine compliance with  bacteriological standards.

 Sampling and Analytical Methods

     The monitoring programs should specify sample  collection,  preservation,
 storage,  and analysis methods which  are approved by  the EPA and  state
 agencies and are appropriate for the site.   In  addition to specification  of
 analytical  methods, the minimum  accuracy, limits  of detection, and desired
 number of significant  digits to  be recorded should be  specified to help
 ensure that accurate and precise data are obtained.  Receiving  water samples
 should be taken with a Van Dorn,  Frauchy, or  comparable sampler and then
 transferred to the  proper  type of container.   When  variables are measured  by
 electronic  probe (e.g., temperature,  conductivity,  dissolved oxygen,  and pH)
 values should be measured  at 1- to 3-m  (3- to 10-ft) intervals.  Electronic
 probe systems often have severe accuracy  problems.  Frequent calibration  is
 essential.

     Sediment samples for  organic carbon should include only the upper 2  cm
 (0.8 in)  of the sediment to ensure that  the  sediment oxygen demand per unit
mass is  not diluted by underlying,  stabilized, or inorganic material.

     Table  5  lists  sample  preservation  and  storage requirements for some
variables,  showing  minimum sample volume, type of container,  preservative
required, and maximum storage time.   Any deviations in container type  or
preservative  from those specified in this table should be noted on  sample
bottles,  field data sheets,  and in the monitoring reports.   The  volumes
given are intended  as minimum amounts.   Sample volumes  should  be increased
depending on  the number of sample splits  to be made.
                                   28

-------
     TABLE 5.  RECOMMENDED SAMPLE  PRESERVATION AND
        STORAGE REQUIREMENTS FOR WATER QUALITY

Parameter
Oil and Grease
Total Suspended
Solids
Settleable Solids
Volume
Required (ml)
1.000
Sufficient aliquot to con-
tain residue of >_ 25 mg.
Successive aliquots of
sample may be added to
the same dish.
1,000
Extinction Coefficient 100
pH
Ainnonia Nitrogen
Salinity
Temperature
Biochemical Oxygen
Demand
Total Phosphorus
Reactive Phosphorus
Nitrate - N
Nitrite - N
25
400
240
1,000
1,000
50
50
100
50
Total Kjeldahl Nitrogen 50-500
Dissolved Oxygen
Color
Total and Fecal
Col i form
Total Organic Carbon
Chlorophyll a_
a If samples cannot
reported data should
300
50
100
25
200
be returned to the laboratory
Indicate the actual holding
Container
G only
G, P
P. G
P, G
P. G
P. G
G (with paraffined
corks)
P, G
P. G
P. G
P, G
P. G
P. G
P, G
G only
P. G
P, G, sterilized
P, G
P, not acid-washed,
keep in darkness
away from light
In less than 6 hours
time.
References: Standard Methods (American Public Health Association 1980)
Preservative
Holding Time
Analyze immediately or 24 hours
5 ml HC1 at time of „
collection. Cool, 4° C
None. Analyze
immediately.
None
Analyze same day or
cool, 4° C
Determine on site or
cool , 4° C ~
Cool . 4° C
H2S04 to pH <2
None
Det. on site
Cool . 4° C
Cool, 4° C
H2$04 to pH <2
Filter on site.
Cool , 4° C
Cool, 4° C
H2S04 to pH <2
Cool, 4° C
Cool, 4° C
H2S04 to pH <2
Determine on site
Cool , 4° C
Add Na thiosulfate
to effluent samples
Cool, 4° C
H2S04 to pH <2
Filter on site, add
MgC03 during
filtration
and holding time exceeds
and Methods for Chemical
No holding
24 hours
7 days
6 hours*
7 days
1 hour
(Longer If properly
sealed air tight)
No holding
24 hours
24 hours
24 hours
24 hours
>24 hours
24 hours
24 hours
No holding
24 hours
Analyze 1n field
or 6 hours
24 hours
Process Immedi-
ately or 2 weeks
if frozen and
kept in dark.
this limit, the final
Analysis of Water
and Hastes (U.S. EPA 1979b).
                          29

-------
     Chemical analysis procedures for sediments  are basically the  same as
for water  samples once the sediments have been digested.  The EPA/Corps of
Engineers  manual, Procedures  for Handling  and  Chemical Analysis of Sediment
and Water  Samples (Plumb 1981), should be consulted for detailed guidance on
sediment sample handling and  digestion  procedures.   Sediment samples can be
stored dried, frozen, or on ice for metals analyses.   If analysis of  organic
constituents  is to be performed,  sediment samples  should be stored on ice
only.

     Analytical methods should be selected based on  the water quality
standards  and the EPA-approved methods  listed in 40 CFR Part 136, with due
consideration of the extent  to which  interferences occur in the receiving
water and wastewater samples.   When several  methods are available, the
selection  should be made  by comparing the accuracy and  precision of the
candidate  methods for the concentration range expected at the site,  and the
adequacy of the detection limit of each  method  relative to the pertinent
water quality standard.   Table  6 shows recommended methods for the same
variables  listed in Table 5 along with acceptable minimum precision and
detection  limits for  each method.  The  table  also  shows the desired number
of  significant  digits to be reported for each parameter.

Oceanographic Measurements

     Oceanographic measurements  to  meet the 301(h)  objectives include  two
parts: data needed to detect plume and  sediment movement,  and observations
needed to interpret  the water quality  and  biological data.  This section
discusses acquisition  of current,  wind, and  tide  data.   Verification  of
initial  dilution  calculations  is  not required.  However, if plume
calculations are considered by the applicant to be unreliable  or inaccurate,
it  may be desirable  to  obtain supplemental monitoring  data  to improve the
models or to document field  validation  of  other models.

     The POTW's 301(h)  application and its Technical Evaluation or Review
Report should be reviewed  to  determine  if available current data  and
knowledge of wind  and tide effects  are  adequate to determine the direction
of  movement of the  wastefield beyond the ZIO  boundary, the  subsequent
dilution, and the  direction  of movement of  the sediment from the discharge.
                                   30

-------
TABLE 6.  RECOMMENDED ANALYTICAL METHODS

Parameter
Flow

Grease and Oil
Floating Participates
Total Suspended Solids

Settleable Solids



Extinction Coefficient
discoloration

oH
r ' '

Salinity

Temperature

Dissolved Oxygen









Biochemical Oxygen
Demand

Chemical Oxygen
Demand

Total Organic Carbon

Ammonia Nitrogen


Nitrate-Nitrogen0





Method
Continuous measurement, auto-
matic device
Gravimetric, Separatory Funnel
Extraction EPA Method 413.1
Flotation funnel extraction
Gravimetric, Dried at 103-105° C.
EPA Method 160.2 (SM, 14th ed..
p. 91, Section 206D)
Volumetric, Imhoff Cone.
EPA Method 160.5.
Gravimetric Method (SM 14th
ed. , pp. 95-96, Sec. 208F)
Light transmissometer
Presence or absence of color
at surface
Potentiometric. EPA Method
150.1 (SM 14th ed. , p. 460,
Sec. 424)
Induction Salinometer
or titration. (SM 14th ed.,
p. 107, Sec. 2G9C)
Bathythermograph or Thermo-
metric. EPA Method 170.1.
(Si: 14th ed., p. 125, Sec. 212)
Modified Winkler, Full Bottle
Technique, with azide modifica-
tion fur effluent sampler o_r
Membrane Electrode when cali-
brated with Modified Kinkier
with azide modification for
effluent samples. EPA Method
360.2 or EPA Method 360.1 when
calibrated with EPA Method 360.2
(SM 14th ed., pp. 441-447. Sec.
422 A and B for Winkler; SM 14th
ed., p. 450, Sec. 422F for probe)
5 day, 20° C
EPA Method 405.1
(SM 14th ed., p. 543, Sec. 507)

EPA Method 410.3
(SM 14th ed., p. 550,
Sec. 508)
Combustion-Infrared Method
EPA Method 236 (SM 14th ed.,
p. 532, Sec. 505)
Automated Phenate Method.
EPA Method 350.1 (SM 14th ed..
p. 616, Sec. 604 or Strickland
and Parsons, p. t>7)
Technician Auto Analyzer 11 px
Spectrophometric, manual,
Cadmium Reduction. EPA Method
3t>3.2 or EPA Method 353.3.
(SM 14th ed., p. 423, Sec.
419c, for manual method)


Significant
Minimum Digits
Precision Detection Desired
+ 8 percent 0.02 MGD

+_ 0.9 mg/1 5 mg/1
N/Aa mg/m2
N/A mg/m
approx. +_ 5 mg/1 10 mg/1

N/A 1 ml/l/h



N/A
N/A N/A

+0.1 standard unit pH = 12


titration: +_ 0.05 ppt 1 ppt

+_ 0.05° C

+ 0.05 mg/1 (Winkler) 0.1 mg/1
Probe: +0.1 mg/1 0.1 mg/1









+ 0.7 mg/1 BOD at 1 mg/1
2 mg/1 BOD
+ 26 mg/1 BOD at
175 mg/1 BOD
+ 13 mg/1 COD 1 mg/1


N/A N/A

+ 0.005 mg NHj-N/l 0.01 mg NHj-N/1


Automated, 0.05 mg/1
Cadmium:
+ 0.092 mg/1 at
0.35 mg/1
Spectrophometric,
Cadmium:
+ 0.004 mg/1 at
0.24 mg/1
3

2
2
2

2



2
N/A

3


4

3

3









2

2


2

2


2





                      31

-------
  TABLE  6.    (Continued)

Parameter
Nitrite-Nitrogen


Total Kjeldahl
Nitrogen

'Total Phosphorus






Method
Technician Auto Analyzer II or
Diazotization. EPA Method
304.1 (SM 14th ed.. p. 434.
Sec. 420)
Technician Auto Analyzer II or
Colorimetric, EPA Method 351.3.
(Sh 14th ed., p. 437. Sec. 421)
Technician Auto Analyzer II or
Digestion, Manual Ascorbic Acid
Technique. EPA Method 365.2
(SM 14th ed., p. 476, Sec.
42bC(III) and p. 481 Sec. 425F
for manual method)

Precision
N/A


Colorimetric
+ 1.056 at 4.10
mgN/1
Automated:
+ 0.130 mgP/1 at
0.8 mgP/1
Manual :
+ 0.033 mgP/1 at
0.11 ngP/1
Significant
Minimum Digits
Detection Desired
0.01 mg H02-N/1 2


Colorimetric 2
<1 mgN/1

0.005 mgP/1 2





 Fecal Coliform
Total Coliform
Chlorophyll a_
Particle Size
  Distribution"
Multiple Tube Fermentation
Technique, MPiJ Test.  (SM 14th
ed., p. 922, Sec.  908C)

Multiple Tube Fermentation
Technique, MPH Test (SM  14th
ed., p. 916. Sec.  906A)  or
for seawater only  Membrane
Filter (SM 14th ed..  p.  928,
Sec. 9CW)

Spectrophotometric (SM 14th
ed., p. 1029, Sec. 1002G
Strickland and Parsons,  SCOR/
UNESCO Equation, pp.  185-194)

Sieve Analysis (Buchanan)
                                                           MPN  with  95 percent    NA
                                                           confidence limit
                                                          HPN with 95            NA
                                                          percent confidence
                                                          limit
NA                     NA
8 N/A = not available.

  Minimum accuracy, when given as  a  range  about a specific value, has been taken from the below listed refer-
ences.  The associated  ranges are  for  1  standard deviation about the mean value.

  The cadmium reduction method determines  nitrate + nitrite-nitrogen.  The nitrate-nitrogen 1s calculated by
subtracting nitrite-nitrogen as determined by a separate diazotization test.

  Detailed procedure is given in the biological monitoring section.


Reference's:

Buchanan, O.B.  1971.  Sediments.   In:   International Biological Program (IBP) Handbook No  14
Blackwell Scientific Publ.,  Oxford,  pp.  30-52.

CLMBS:  Great Lakes Region  Comittee on  Analytical Methods.  1969.  Chemistry Laboratory manual
bottom sediments.   EPA, Federal Water  Quality Administration. Washington, D.C., 101 pp.

EPA Method:  U.S.  Environmental Protection Agency. 1979b.  Methods for Chemical Analysis of
Water and Wastes.   USEPA, Environmental  Support Laboratory, Cincinnati. OH.

SM 14th ed.:  American  Public Health Association.  1980.  Standard methods for the examination of
water and waste water.   14th ed.,  Washington, D.C.  1193 pp.

Strickland and Parsons:  Strickland, J.D.H.. and T.H. Parsons.  1972.  A practical handbook of
seawater analysis.  Dilution 167,  2nd  ed.  Fisheries Research Board of Canada.  Ottawa, Canada.
310 pp.
                                                    32

-------
The current data should be  reviewed to determine  whether surface, bottom,
and mid-depth  currents were measured.   The sampling times should be reviewed
to determine whether periods of minimum and maximum stratification and other
important conditions (e.g., onshore winds, upwelling periods)  were covered.

     Winds  generally  are an  important  influence on coastal currents.
Current  data  should therefore be checked  to see if historical  surface
currents  were  measured concurrent  with wind measurements and to what depth
wind effects are discernible.  The wind data should be reviewed to estimate
frequency of occurrence of onshore transport by season and location.  This
information is helpful  in identifying discharge impact areas  along the
shoreline.  Any deficiencies in available current data should be noted and a
determination  made as to whether intensive  current monitoring is needed or
if specific data gaps need to be filled.

     Field observation  methods selected for  oceanographic measurements
depend on the  kinds of information needed, the extent of potential discharge
impacts,  the  oceanographic  and  physical  conditions  at  the site,  and
resources available to the  applicant.  Drogues,  drifters, or dye released
from the  outfall site may  be used to  determine mean current velocities at
specified depths, and also to provide information on  the direction of
movement  and the dispersive properties of the velocity field.   These studies
may also  identify nearshore eddy patterns or "dead" circulation zones which
may be present.  Drogues set just above the bottom, or seabed drifters, can
be used to determine the direction of sediment movement.

     Oceanographic  data which should be  recorded at the  time  of  water
quality and biological sampling include:

     t    Wind speed and direction

     t    Sea  state (height of swell  and waves).

     A variety of field study methods are available to collect oceanographic
information.  The appropriate use and limitations of current meters, drogues
and drifters,  dye studies,  and field  positioning  methods are discussed in
Appendix  A to  this document.
                                   33

-------
Data Analysis and Reporting

     Data reporting procedures  include the preparation of field logs,  sample
container labels,  laboratory data  sheets,  and reporting forms.  The
reporting forms should  be  completed and sent  to  the appropriate U.S. EPA
Regional  Office on the schedule prescribed in  the 301(h) permit.   Chain of
custody  forms, showing  the  transfer of  data from the  field to the
laboratory,  and finally to the U.S. EPA, should  be  maintained along with
field logs and laboratory data  sheets.

     The  type of information recorded on the field logs and sample  labels
should ensure that samples are identified properly  and data are recorded
accurately.  Field logs should  include station location  and number, depth of
samples,  type(s)  of samples  taken,  date and time  of sampling, surface
observations as specified  in the  oceanographic section, depth of water at
the station where samples are  taken, all field water quality measurements,
and the  names  of all  individuals  who collected  the samples.   This
information should be entered on the field log at the time of sampling.  The
sample container labels should give sample  number,  station location, and
number; date, time, and depth  of  sample;  treatment of sample (e.g., ^$64
added); a code designating  what analyses are to be done on the sample, and
the name  of the individual(s) collecting the  sample.  Laboratory data  sheets
should include sample number, station location and number, sampling date and
time,  and name of the analyst.  For each individual analysis,  results  should
be reported along with the  unit of measurement, duration of sample storage,
date sample was analyzed,  and  any comments  on deviations from laboratory
procedures or unusual sample conditions.   The chain of  custody forms  should
show the  name of the person to  whom  the  form is being sent; and the name of
the person receiving the data and date received.

     Receiving water quality and sediment data should be compared with NPDES
requirements (when  applicable)  and applicable water quality standards.
Spatial  gradients should be examined  to determine  whether elevated
concentrations occur near the outfall and, if so,  where concentrations
return to background levels. Analysis  of temporal trends should be done to
identify  seasonal differences.  Appropriate  statistical tests (e.g.,  ANOVA)
can be used to  determine  if statistically  significant  differences exist
between  the ZID-boundary  and reference (control)  stations.  The water
                                  34

-------
quality  and sediment data can be used to define and  report on the  spatial
extent of the wastewater  plume  and sediment deposition area.  This
information should be used  in conjunction  with the biological  monitoring
data to  identify and interpret any changes detected  in  the biota.
                                   35

-------
                               CHAPTER IV

                          BIOLOGICAL MONITORING
OBJECTIVES

     Biological monitoring is necessary to evaluate  the overall impact of
the permittee's modified discharge.  The primary objective of  the biological
monitoring program is to provide evidence that:

     •   There  is  a continued attainment or maintenance of water
         quality  which assures  protection and  propagation of a
         balanced, indigenous population (BIP)  of shellfish, fish,
         and wildlife beyond  the  zone  of  initial dilution (ZID) and
         in the vicinity  of the ZID boundary

     •   Conditions within the ZID do not contribute  to  extreme
         biological  impacts,  such  as the  destruction of  distinctive
         habitats  of limited distribution (e.g.,  kelp beds and coral
         reefs),  the  presence  of disease epicenters,  or  the
         stimulation of  phytoplankton blooms which have  adverse
         effects beyond the ZID, etc.)

     t    For discharges into saline estuarine waters:   a)  benthic
         populations  within the ZID do  not differ substantially from
         balanced,  indigenous populations which exist  in the vicinity
         of the  ZID  boundary, b)  the  discharge  does not interfere
         with  estuarine migratory  pathways within  the  ZID, and c)  the
         discharge  does  not  result in an  accumulation of  toxic
         pollutants  or pesticides at  levels which exert  adverse
         effects on  the biota within the ZID

    t    There is  a continued attainment or maintenance of  water
         quality which allows  for  recreational activities (including
                                 36

-------
          fishing) beyond the ZID boundary and such  activities in the
          vicinity of the modified discharge are not  restricted unless
          such  restrictions are  routinely  imposed around  sewage
          outfalls discharging secondary effluent.

     Design of  the biological  monitoring  program  requires  careful
consideration  of potential impacts  specific to  the  301(h) permitee's
discharge(s).   Factors which are  important  to designing biological
monitoring programs and individual  sampling procedures are discussed below.

APPROACH  AND RATIONALE

     Section 125.62(b) of the amended 301 (h)  regulations requires  that the
biological monitoring programs for both small and  large 301(h)  discharges
must provide date adequate to evaluate the impact  of the modified discharge
on  the marine  biota.   This  generally necessitates  comparing  the
characteristics  of selected marine communities in the  vicinity of the
discharge with the characteristics of similar  communities  in reference
areas.   Therefore,  the same type of comparative  strategy required for
demonstrating a balanced,  indigenous population  (BIP) of shellfish, fish,
and wildlife in the application  should be  incorporated into the  biological
monitoring program.  [See  the  Revised Section  301(h)  Technical Support
Document  (Tetra Tech 1982) for guidance on demonstrating a BIP in  a 301{h)
application.]

     Under Section 125.62(b)(l)(i-iv) of the amended  regulations,  biological
monitoring programs must to  the extent practicable  include:

     •    Periodic  surveys of the biological  communities and
          populations most likely  affected by  the discharge, as  well
          as those at suitable  control  sites, to enable comparisons
          with baseline  conditions

     •    Periodic  determinations of the  accumulation of toxic
          pollutants and pesticides in  organisms and examination  of
          adverse effects  such  as disease,  growth abnormalities,
          physiological  stress,  or  death
                                  37

-------
     •    Sampling of sediments  in  the vicinity of the ZID boundary,
          in other areas of  expected sediment accumulation,  and  at
          appropriate reference sites to support water  quality and
          biological  surveys  and to measure  the accumulation of toxic
          pollutants and pesticides

     •    Periodic assessment of the  conditions  and productivity  of
          commercial  or recreational  fisheries where  the discharge
          would affect such fisheries.

Except for the periodic  survey requirement,  small permittees are not subject
to these  specific requirements if they discharge at depths greater than 10 m
and can demonstrate  through a  suspended solids  deposition analysis that
there will be  negligible sea  bed accumulation in the  vicinity of the
modified  discharage.   However, small  permittees still must provide  adequate
data to evaluate the impact of  the  modified discharge on the marine biota.
This should involve  the establishment of  a background  data  base and the
demonstration of predicted biological  impacts  of the small discharge.  In
all cases,  site-specific characteristics will  affect the selection of the
number of sampling sites, sampling  locations,  and the required  sampling
effort in each biological  category.

     Information available in the  discharger's 301(h) application and in
other investigations conducted  near the discharge should be utilized fully
to identify physical-chemical and  biological  characteristics of the
potentially  affected  receiving  waters.   Characterization  of  the
oceanographic and meteorological  setting of the discharge area  will be
necessary to make  decisions concerning  positioning of the discharge and
reference sampling stations.  Available biological data should be  reviewed
to define limits  of  natural  variability in biological populations.  The
number of sampling stations and  number of replicate samples at  each station
should be determined, in  part,  on  the basis of this information.   In these
respects, the historical  data may serve the same purposes as a pilot survey.
Decisions concerning taxonomic groups  to be sampled, station locations,
types of sampling equipment,   sample handling,  sorting  procedures,  and
ancillary measurements of  physical-chemical  parameters should be made on the
basis of  existing information to the extent practicable.   Where effects of
the proposed discharge on specific biological communities or important
                                  38

-------
species has  not been clearly resolved,  the monitoring program  should be
designed to  fill such data gaps.

     In designing the program specifications,  the complementary  nature of
the water quality and biological  monitoring programs should be recognized.
Concurrent collection of biological  and water  quality information  should be
emphasized in an effort to identify causal relationships.

SPECIFICATIONS  FOR BIOLOGICAL MONITORING

Sampling of  Biological Communities

     The 301(h)  regulations require  "periodic surveys of  the biological
communities  and populations which  are  most likely  affected by the  discharge
to  enable  comparisons with  baseline  conditions described  in the
application."   Emphasis should  generally  be placed on monitoring of  benthic
communities due to the  inherent  community characteristics,  sampling
considerations, and the importance of the  benthos in the marine ecosystem.
Benthic communities adjacent to  pollution sources can generally  provide
information  on  the area! extent of impact more  readily than other biological
communities because  many benthic organisms  are  sedentary or relatively
immobile and  are,  therefore, continually exposed to  pollution stress.
Benthic communities  are also  more easily  sampled than other biological
communities  and benthic sampling methods  are more  standardized than  methods
for other communities.  Existing information on benthic communities is  often
sufficiently extensive to provide  documentation  of both the magnitude and
direction of the community response to  specific  perturbations.  Finally,
benthic communities  are of a primary importance  in the food chain of the
nearshore marine environment.  For the  above reasons, monitoring of the
macrobenthos  should  normally  be a primary element of 301(h)  permit
biological monitoring programs.

     Another principal monitoring  requirement  defined in the regulations is
periodic assessment of the accumulation  of toxic  pollutants and pesticides
in  the  biota.   These assessments  are required  as  part of a  specific
monitoring effort for measuring the  impact of  elevated or increasing levels
of toxic pollutants and  pesticides  on susceptible  biological communities.
Bivalves (e.g., My til us californianus and M.  edulis), have been shown to
                                  39

-------
greatly concentrate most identified  marine pollutants relative to ambient
concentrations  in seawater.  Because water quality characteristics (e.g.,
temperature  and dissolved  oxygen)  affect biological  uptake,  the
concentrations  of toxic substances in the tissues of these organisms will
more accurately reflect the site-specific potential for bi©accumulation than
will the measurement  of  ambient concentrations of  toxic substances.  For
these reasons,  caged  bivalves  used in offshore biomonitoring  systems may
provide an early warning of excessive water column contamination,  and may be
used  to  monitor  the  potential  for transfer of  toxic  pollutants and
pesticides into and through the food  chain.  Such an in situ biomonitoring
system  also  provides a means of evaluating the effectiveness of  toxic
control  programs.

     Monitoring program requirements also include the periodic assessment of
commercially or recreationally  important fisheries that may be  affected by
the discharge.  The  objective of fisheries monitoring  is to assess the
condition and productivity of  those  fisheries.  Fisheries monitoring may
consist of the  periodic review of catch  data collected by state agencies,
interviews of  sport  fishermen to determine success  rates, or field and
market sampling of the fish or  shellfish populations.

     The biological  monitoring  program  regulations  specify that the
permittee monitor the biological  communities and  populations which are
likely to be affected by  the  discharge using comparisons with baseline
conditions described in the application.   In addition to  the benthos, such
communities may include phytoplankton, zooplankton, and fishes.

     Numerous  site-specific  characteristics  of the  environment may
necessitate additional biological  monitoring.  For example, hydrographic
characteristics  (current patterns,   water residence time)  and nutrient
concentrations  in an estuary or embayment  may  result in  the potential for
long-term biological  changes such as  eutrophication.   If such  changes are
determined to  be  a  potential impact  of  the  modified discharge, periodic
monitoring of the phytoplankton community may be required.  Furthermore, if
changes  in species composition of the  phytoplankton occur and thereby induce
changes  in the  species composition  of the herbivore community,  zooplankton
monitoring may  be required.
                                  40

-------
     In certain geographic settings,  some coordination  of  the monitoring
programs of adjacent  dischargers may  be required so that those dischargers
periodically sample the  same biotic  groups.  These conditions include, for
example, the situation where the  potential for eutrophication exists, or
where  further examination  of the  multiplicative  effects of  several
dischargers on widespread toxic dinoflagellate blooms is needed.

     Monitoring should be initiated  to ensure the continued existence of
distinctive habitats  of  limited distribution.   Such habitats include, but
are not limited to,  kelp beds, coral  communities, and  rocky  intertidal
communities.

Station Locations

     To meet the minimum requirements  of the biological monitoring program,
sampling of the selected biological communities in the  vicinity of the ZID,
in any other areas of expected impact,  and  at appropriate control sites will
generally be required.  Other  sampling locations  which may be specified
include nearfield areas where important habitats have  been identified, and
also at both new and old  discharge  sites in  the case of an improved
discharge involving outfall  relocation.   In the  case  of  large permittees,
additional  sampling is recommended  at  intermediate locations between the ZID
boundary and control  stations along a  gradient of effluent  concentrations to
help define the spatial extent of biological effects.

     Information  derived from the  water  quality  monitoring program will be
important  in interpreting the results of biological sampling.  Therefore,
the selection of water quality and  biological stations  must not be done
independently.  In  addition, the  station  selection process of  the
biomonitoring program should place emphasis on the  inclusion of historical
sampling sites.  This will maintain  sampling continuity,  and additional
information will  thus be made available  for  impact assessment.

     In instances where chemical  analysis  of the  effluent and/or sediments
in the  vicinity of the discharge has  identified  toxic  pollutants at levels
of concern, the required bioaccumulation  monitoring  of  toxic substances
should  be  undertaken  in the  vicinity  of the ZID boundary, at other areas of
expected impact,  and  at  control  site(s).   Test  organisms utilized for in
                                  41

-------
situ blomonitoring studies  should be placed at the depth of  the plume during
the exposure period.   If the plume surfaces, exposure should be conducted at
a sufficient distance below the surface to prevent damage to or loss of the
exposure  apparatus.  Additional  exposure depths may be used if the plume
depth is  uncertain  or  variable,  or if past discharges have  resulted 1n
substantial  sediment contamination.   In  cases  of  past  sediment
contamination,  near-bottom exposures can  be  used to evaluate  the
contribution of sediments to bioaccumulation levels.

     The  Technical Evaluation Report,  the tentative decision document, and
the NPDES [301(h)] permit should  be reviewed to determine  requirements for
sampling  other biotic groups.   If this review indicates the need to sample
phytoplankton, zooplankton, or  fish communities, sampling  locations should
be specified in the vicinity of  the ZID  boundary, other areas of expected
impact, and at one or more control  sites.   The  number of  control sampling
locations, as well  as the need  for and number of any sampling sites
specified in nearfield  areas,  should  be  determined based  on the nature of
the  potential  impacts of the discharge.   As  in the example previously
presented,  sampling depths should  be  specified for each of  the above
sampling  station  groups.  Determination of sampling depths must  be  made on
the basis of oceanographic  conditions  and behavioral characteristics of the
organisms; these  depths may vary  seasonally.  The monitoring  of habitats of
limited distribution may also be  required in the nearfield area at a single
site or at several locations  including control sites.

     There would  be additional  station  requirements for discharges  into
stressed waters  or in  situations where  other pollutant  sources could
potentially affect biological communities in the  vicinity of the applicant's
discharge.  In such  cases, it  is important to define the magnitude of the
discharge interaction(s)  and  describe any biological  response gradients
associated with the  applicant's discharge and other pollutant sources in the
study  area.  Therefore, several  additional stations may be required at
intermediate  positions between the applicant's discharge  and other
significant pollutant sources  in  the study area.

     For cases where there  is an improved discharge  involving  outfall
relocation, monitoring  is  required in the vicinity of the ZID boundary at
both the  relocated  discharge site  and at the existing discharge site  until
                                   42

-------
discharge at  the  latter site ceases.   Within-ZID stations, if necessary,
should be located  close to the midpoint  of the diffuser.  ZID-boundary
stations should be  oriented in the  direction of the predominant current.
Single within-ZID and ZID-boundary stations should be sufficient for small
discharges, while two or more stations of each type may be  needed for larger
discharges.

     Selection of control stations is one  of the more important aspects of
the design of the monitoring programs, since  all assessments  of  impacts will
rely on comparisons made with data from these locations.  Control stations
should be located outside the traceable waste field and not be affected by
the applicant's discharge.  Similarly,  the  selected locations should not be
influenced by other discharges.   Control  stations should be located in water
of similar depth to  that of the within-ZID, ZID-boundary, and gradient
stations.   Sediment characteristics  should be similar at all  sampling
stations except where sediment alterations are due  to an  outfall effect.
Control and other monitoring  stations should be located  approximately the
same distance from  shore.  Since it is often necessary  to locate control
sites a considerable  distance [5-10  km  (3-6 mi)] from the  outfall to escape
all waste field  influences,  candidate control  sites should be carefully
evaluated to  ensure that oceanographic conditions are not atypical.

     Example  layouts  of sampling  locations for  two  alternative biological
monitoring programs are presented in  Figure  1.  Included in  the example are
sampling stations that should be expected at  a relatively large  (x stations)
or small (o stations) discharge.  Some  of  the important features common to
both layouts  that should be noted are:   sampling stations  have been located
at the  same  depth  and at  approximately  the  same distance from  shore;
near-ZID  and nearfield  gradient  stations are  positioned in  the  same
direction from the diffuser as the predominant current direction;  and,
control stations are  located a considerable distance from the diffuser and
in the opposite direction of predominant currents.

     Although type(s)  of stations  are  not specified in the  figure,  the
layouts are typical  of expected  benthic  sampling designs.   Sampling of other
biotic groups, if required, might also be conducted at these stations.   The
number and location of stations  are  indicative  of  the most basic programs
that would be expected  of relatively small  and large discharges.  In the
                                  43

-------
                NEARFIELD   i« o  x.
                ZID
o  Small discharge station

x  Large discharge station
                  CONTROL
       Figure 1.   Representative sampling locations for two levels
                  of biological  monitoring.
                                  44

-------
case of a large  discharge where a high  potential for sediment accumulation
offshore has  been identified, a number of stations would  be required in the
deeper water  offshore of the diffuser and  at appropriate control stations.
These example layouts also do not  reflect the existence of areas of special
concern, such as important fish habitats where additional  sampling might be
required.

Sampling Frequency and Replication

     The 301(h)  regulations do not offer explicit guidance for either sample
frequency or  replication.   For those biological communities likely to be
affected by  the  discharge, sampling frequency will  be dictated by
community-specific characteristics.   For example, due to  the  rapid response
of phytoplankton to environmental  perturbations and seasonal fluctuations in
community structure, the most effective  sampling strategy  might be intensive
sampling for  relatively  short periods of time.   Similarly,  the ability to
sample juvenile  fish in nursery areas may  be limited to  certain seasons of
the year.  These examples point to the fact that sampling strategies should
be considered in the sampling design.  In the development  of the strategies,
data should be reviewed carefully to consider life history characteristics
of target species.

     Sample replication requirements  are  both site-  and species-specific.
Decisions on the level  of sample replication or sampling  effort should
include careful  consideration of  the minimum  detectable difference in
selected biological  parameters.  Field  experiments should be planned
carefully in order to define minimum detectable difference levels,  to
establish the number  of replicate samples  required, and to specify the
appropriate analytical approach.

     Prior to designing a sampling program,  the applicant should consider
two important criteria  associated with the sensitivity of the  study to
changes in biological  parameters.   These are the  probability of rejecting
the null hypothesis when it is true (commonly called the  probability, or
Type I error)  and the probability  of  accepting a null  hypothesis when it is
false (commonly called the   probability, or Type II error).  The complement
of e  (1  - 3) is referred to  as  the power  of  a  test and is especially
important since it  defines the probability of correctly detecting
experimental  effects (e.g.,  differences among sampling stations).

                                  45

-------
     For  a specified  variance associated with  a  biological parameter, the
probabilities of a,  3, and minimum detectable differences among sampling
areas can be expressed  as  a  function  of  sample  size.   The allocation of
sampling resources  (stations,  replication,  and frequency)  can then be
determined with regard to available resources, practicality of the design,
and  desired sensitivity of  the subsequent  analyses.   Discussions and
examples of this  approach are  included  in Cohen (1977), Winer (1971),
Scheffe (1959), Moore and Mclaughlin (1978), Gordon et al.  (1980),  and  Sail a
et al.  (1976).

Sample Collection and Processing

     The  following subsections provide a discussion  of  appropriate sample
collection, sample handling, and quality assurance/quality control methods
for  the  individual  biotic  groups which may be  included  in a 301(h)
monitoring program.

Benthos--

     Most biological  monitoring  programs  will  emphasize the  macrobenthos
since micro- or meiofaunal  benthic  samples are difficult  and expensive to
process  and also present interpretive difficulties  due to extreme
small-scale heterogeneity and lack of understanding of community
relationships.   Should  impact  upon micro- and meiofaunal benthos  prove
significant for  some outfalls  such that monitoring  of these infaunal
components is required,  the investigators should consult Fenchel (1969),
Wieser  (1960),  Mclntyre and Murison  (1973), and Hulings  and Gray (1971) for
information on  sampling methods and sample  handling.

     The  methods  and equipment for sampling  macrobenthic  infaunal
communities have  been the subject of several  publications [Holme and
Mclntyre  (1971),  Word (1976), Hedgpeth (1957),  and Swartz (1978)].  The
ideal bottom grab for sampling all  sediment grain sizes,  from sand to  silt,
has yet to be invented.   Word (1976) compares the sampling efficiency of
seven grab  samplers (Ponar,  corer,  Shipek,  van Veen, orange peel,
Smith-Mclntyre,  and a  chain-rigged  van Veen) in the  silty-sand to
clayey-silt sediment off  southern California.   The  results of Word's
                                 46

-------
investigation suggest that a 0.1-m2  chain-rigged van Veen grab is the most
reliable sampling device for these sediment  types.  Swartz (1978)  recommends
the use of a 0.1-m2 Smith-McIntyre grab due  to its essentially constant
"bite" area;  however, he  recognizes that the  depth of penetration of this
grab varies with sediment type.

     Regional consistency in infaunal monitoring is important to  the 301(h)
program objective.   In the southern  California  area, for example, sampling
should consist of replicate  sampling using a  0.1-m2 chain-rigged van Veen
grab.  In more sandy areas, this grab or the Smith-Mclntyre grab recommended
by Swartz (1978) may prove acceptable;  however, investigators studying sandy
infaunal communities should define the sampling efficiency of whichever grab
they chose to utilize.

     In areas  where visibility  and oceanographic conditions permit,
diver-operated coring or dredging  may be more desirable than grab sampling
from a surface vessel.   The  type and size  of  sampling device suitable  for
each kind of substrate may vary  from suction dredges (Brett 1964; Gale  and
Thompson 1975)  which cover large areas (for  substrates with a low  density of
organisms) to small  coring tubes or  small box corers (for substrates with a
fairly high density  of infauna).

     The number of replicate samples collected at each  station should be
sufficient to ensure statistical  reliability (see Sampling Frequency  and
Replication above and Effect of Sample Size  below).  At each station,  one or
more  additional,  separate  sediment sample(s) should  be  collected  for
analysis of  total organic carbon content, grain size  distribution,  and
percentage of gravel,  sand,  silt,  and clay.   Other physical-chemical
parameters discussed in the water quality monitoring section of this report
should be monitored at or near each benthic  station.

     Sample Handling—The monitoring  design should describe all  procedures
used in the benthic sampling program.  These descriptions should include  the
following requirements:

     1.   Each  replicate sample should be  screened  and preserved in
          the field on the day of collection.
                                   47

-------
2.   Each replicate  sample should be  screened, fixed, sorted,  and
     processed separately.

3.   Samples should be screened through a sieve having  1.0-mm
     mesh.   If smaller mesh size screens  are also used (e.g.,  0.5
     mm), the fraction  of the organisms  retained by the  1.0-mm
     sieve  and smaller sieves should  be processed separately.

4.   Organisms should  be  fixed  in a  buffered  10-percent
     formalin-seawater solution.    (Borax  is  suggested as a
     buffering agent.)  The specimens should be transferred to a
     70-percent ethanol  solution after an  initial  fixation  period
     of 24  hours  to 1 week.  Vital  staining techniques  may be
     used as  an aid  to  sorting (see Holme  and  Mclntyre 1971,
     Williams and Williams 1974).

5.   Permanent labels should accompany each sample throughout  all
     phases of sample handling,  processing,  and storage.  These
     labels should include the  date  and  time of collection,  the
     station and  replicate identification number,  the station
     location including at least latitude and longitude, and  the
     sample collection depth.   If available at the time of  sample
     collection, other  label   entries  should  include  water
     temperature, salinity, dissolved oxygen,  and bottom depth.

Sample Processing—

1.   The organisms should be  sorted and  identified to species,
     or, if unidentifiable, sorted into discrete taxa.

2.   The total  number of individuals  of  each species (or  lowest
     identified taxon) in each  replicate  should be determined.
     Counts should be expressed  as the number of individuals of
     each taxon in the samples and per m .

3.   The wet weight  of  organisms  in the six major taxonomic
     groups (polychaetes,  crustaceans, molluscs, coelenterates,
     ectoprocts, and echinoderms) and the total  biomass of each
     replicate  sample should be  determined.
                             48

-------
     Effect of Sample Size—The  accuracy and precision with which benthic
community parameters are estimated depend on the parameter in question and
on the size  of the sample.  It is  therefore appropriate to  discuss the
effects of sample  size on  estimates  of parameters  most frequently used to
describe benthic community structure  and function.

     The total  area sampled among the replicates at each  station should be
large enough  to estimate a given parameter within acceptable  limits of
accuracy and precision.  Within  a  study area, adequate sample size may be
determined by maximizing the  number  of species  collected or by minimizing
the error of estimation of the mean for the parameter in question (Gonor and
Kemp 1978).   Alternatively, the surface area sampled may be determined by a
review of sample sizes which in past  studies have been shown to yield data
with acceptable limits  of accuracy  and  precision.  If  the surface area
sampled  per station is too  small, the  data  will  poorly estimate  the
parameter in question because  the  ratio of  the  variance  to the mean for a
given  parameter  will be unacceptably  large  (Gonor  and  Kemp  1978).
Consequently,  within-habitat variability  (which  is  a function of nonrandom
distribution of the fauna) will obscure  differences in community structure
when stations are  compared.

     Holme and Mclntyre (1971)  and Swartz  (1978) recommend that an area of
0.5 nr (5.4  ft*) be sampled to assess species  composition in coastal  and
estuarine regions.  This  recommendation  is supported by the results of
benthic studies in  Puget Sound (Lie  1968).   From an analysis of ten 0.1-m2
(1.1-ft^) replicates  at  one  site, Lie concluded that a minimum  of five
replicates is  needed  to  accurately  assess  species composition,  while a
minimum of three replicates is required to  accurately estimate biomass  and
numerical  abundance.

     Word (1976) presented an  analysis of 10 replicate samples from southern
California (location  not  given).   He observed  that:  1) the cumulative
number of species does not appear to  approach an asymptote with increasing
number of samples, and 2) a second sample will include newly acquired
species which constitute  only  10  percent of the individuals in the first  and
second samples.  Word concludes that because numerical clustering strategies
are sensitive  to  species  which  contribute  90 - 95 percent of the total
                                  49

-------
 number of individuals, a  single  0.1-m2 (1.1-ft2)  sample is sufficient  to
 obtain "useful descriptive information" (Word  1976).

     Although Word et al.  (1980)  has shown that a single 0.1-m2 (1.1-ft2)
 sample is  appropriate to  describe the Infaunal  Trophic Index  (ITI  is a
 single number characterizing the trophic organization of soft bottom benthic
 communities) in southern California, the  single  sample limits the degree  of
 community characterization.  With  only a single sample, there is no direct
 estimate of within-group variance for  statistical analyses.  Because
 individuals are distributed logarithmically  among the species of a community
 (Preston 1948, Sanders 1968),  the species  collected in  the second and
 successive replicates most often will be  numerically "rare."   Note  that
 "rare" is not synonymous with  "unimportant."   Predators, for example, are
 usually "rare" because they are one  trophic level removed from their prey;
 yet,  predators are  usually a major  factor influencing  the diversity,
 structure,  and function of benthic communities  (e.g., Connell  1961, Paine
 1966, Bilyard 1974).  Hence,  it  should be acknowledged that one  0.1-m2
 (1.1-ft2) sample  is generally not  adequate  to characterize  benthic community
 structure and function.  Many uniformly  distributed "rare"  species which are
 important  in maintaining community structure  and function  will  not be
 captured  in a single  sample.   In general,  then, five replicate samples per
 station are recommended for determining benthic  community  structure and
 function, unless  evaluation of  site-specific data indicates  that  sufficient
 sensitivity could  be  obtained with  fewer samples or that a  greater number is
 required  due to extreme spatial heterogeneity.

     The  previous discussion  concerns the number  of  replicates  (or area
 sampled)  generally required to  adequately characterize infaunal  communities.
The other major aspect of sample size  concerns the statistical sensitivity
 or power  associated with the number  of  replicate samples.   A  discussion of
 the statistical  aspects  is  included in  a  previous section  (Sampling
Frequency and Replication).

     Quality Assurance and  Control  Procedures—To assure proper  handling and
processing  of benthic samples,  the  following procedures  are recommended:

     1.   At least 5  percent  of all  samples  should be resorted by
         individuals different from  those who conducted the original
                                  50

-------
          sorting.  Records  on  the results of  this  second sorting
          should  be  maintained and presented  in an appendix to the
          monitoring report.  This should be a double-blind  test.

     2.   Complete sorting, processing, and/or laboratory records for
          each replicate  sample should  be included  in a separate
          appendix volume to the annual report.   These record sheets
          should  present as a minimum the data specified in Item 5 of
          Sample  Handling (above).   The names and detailed  statements
          of the qualifications  of all  persons performing and
          confirming taxonomic identifications of organisms should be
          included in the appendix volume.

     3.   A voucher  collection consisting of specimens  representative
          of each species (or lowest taxonomic unit of  identification)
          collected  during this monioring program should be developed
          and maintained by the applicant.   This collection should be
          archived for  a  period of not less than 2 years  after the
          expiration date  of  the 301(h) modification and  should be
          housed at a  facility  where adequate curatorship can  be
          assured.

     4.   Taxonomic references used for the  identification  of
          organisms  should be  cited  in  the appendix to the report.

Bioaccumulation Studies—

     The amended  301(h)  regulations  require  periodic determinations (except
for small  applicants meeting certain depth  and  solids  deposition criteria)
of the accumulation  of toxic  pollutants and pesticides  in organisms  and
examination of other adverse effects of the  discharge such as  disease,
growth abnormalities, physiological stress, or  death.   At  discharges where
bioaccumulation  of toxic substances  is  known or  likely  to be a  problem,
tissue samples from  resident macroinvertebrates  and fish species should be
examined for abnormal  body burdens of toxic substances.   The identified
toxic pollutants  to be  monitored  are  the 129 priority  pollutants  plus  six
pesticides  listed in Table  3.
                                  51

-------
      The  primary method to be considered for determining  levels of toxic
 substance bioaccumulation  in  the  vicinity of the outfall  should be through
 the use of caged specimens of bivalve molluscs.  Recommended methods are
 provided  by EPA  (1982).   In  situ biomonitoring has been used to monitor
 levels of toxic  substances  in the  water  column (Young et al.,  1976).
 Generally, mussels  (Mytilus  californianus  or M.  edulis) or oysters
 (Crassostrea spp.) should  be utilized as  the test organisms.  These species
 are widely distributed  and easily collected in large numbers  in most coastal
 areas of  the U.S.   Also,  there is a considerable  amount of literature
 concerning the rate of  uptake of specific  substances,  experimental survival,
 and the selective uptake of the various groups  of toxic substances  by
 different  tissues (see, for example:   de Lappe et al.  1972; Young and Heesen
 1974; Clark and Finley  1973;  Alexander  and  Young 1976; and Eganhouse and
 Young 1976).  Other  filter-feeding molluscs have been used in a similar
 manner to monitor toxic substances in the marine  environment (Goldberg  et
 al., 1978), and these organisms could  be  substituted if conditions are not
 appropriate for survival of mussels or  oysters.

     Although  minimum  numbers of replicate samples and specimens are
 specified in EPA  (1982), the  investigation of site-specific environmental
 characteristics,  seasonal  variability  in background pollution concentration
 levels, and,  most importantly, variability in the uptake of different toxic
 pollutants  should  be examined during  the  initial  stages  of the
 bioaccumulation monitoring program.   The objective  of these  preliminary
 investigations  should be to determine the number of  replicate  samples and
 number of organisms included in composite samples which will result  in the
 optimal  sampling program, i.e.,  a program that will  provide  the basic
 assessment information at  a minimum cost for sample collection and analysis.

     The monitoring of toxic pollutant concentrations  in tissue samples  from
 resident macroinvertebrates and fish  species should  be conducted  in  cases
where the  bioaccumulation  of  toxic pollutants has been documented or the
potential  for accumulation in  the food chain is  considered to be  high.
Emphasis  should be placed  on the selection of  commercially or recreationally
important species  for  which  information is available on  the  uptake and
effects of elevated  toxic  pollutant  concentrations.   Composite  samples
consisting of at least six specimens  should  be  collected  at the specified
stations and sampling  periods  for tissue  analysis.  Since the objectives of
                                  52

-------
this phase of  the program are  to  investigate accumulation of  toxic
pollutants in the food chain  and  to  assess the suitability of commercially
or recreationally important  species  for human consumption, emphasis should
be placed on the determination  of toxic pollutant concentrations in muscle
tissue.

     At  the time of sample collection,  the  length, weight,  and mortality (in
the case of the caged bivalves)  should be recorded.  The physiological
condition  of all organsims  (e.g.,  the presence of external  lesions and
discoloration)  should also be  noted.   The observed concentrations  of
identified toxic pollutants  in tissue samples should be reported in both
tabular and graphic form.    Statistical  comparisons of the observed
concentrations of toxic substances  in tissue samples should be  made to
determine the  existence of significant differences among  stations  or
replicates.  Degree of fouling  of  cages  and presence of potential predators
(e.g., crabs) within cages should  be noted.

Fishes--

     Marine fish communities  are complex  and  dynamic in nature.   The
structure of these communities  changes seasonally as a result  of spawning,
migrations, and recruitment  of  juvenile fish to adult populations.   In the
short term, feeding activities (including diel movements) will influence
observed community composition.  Selective characteristics of various types
of fishing  gear tend to confound this problem (Hamley  1975)  since, for
example, different sized  (and aged)  individuals from the  same  species will
be selected differently by different  gear types,  while  individuals of the
same size from different  species  will also  exhibit differences  in
catchability.   Catchability also varies on a diel basis as  a result of
changes  in avoidance capabilities  under  different light conditions,  effects
of tidal currents on activity patterns, and other factors.

     Extreme spatial heterogeneity  is a characteristic aspect  of the
distribution of many species  of demersal and (especially) pelagic species;
sampling plans which fail to  take spatial  heterogeneity  into consideration
can result in biased conclusions.
                                  53

-------
     Spatial and temporal variation of this kind places significant demands
on the design of sampling  surveys.  For  example, Richkus  (1980)  reports
that, in a study conducted by Texas Instruments, Inc., in Chesapeake  Bay, it
was determined that  it would be necessary  to collect 252 samples  to  produce
an 80 percent chance of  detecting a 50 percent difference  in  density of
white perch (Morone  americana) among three locations.  Thus,  the  objectives
of a monitoring program for fishes,  together with  the  structure of the
target community, will exert a major influence on many aspects of  the design
of that program.

     Several  different types of  gear have  been used for  sampling  fish
communities.  The selection of gear which is appropriate to address  specific
survey objectives will depend upon  the  substrate type, the  communities to be
sampled, tidal  and other current conditions, depth, proximity to  the shore,
and survey vessel capabilities (Table 7).  A discussion of  the applicability
of various fish sampling techniques is included in Richkus  (1980).  Von
Brandt (1972)  provides a general  review of fish catching methods.  Uzmann et
al. (1977) present a comparison of  three survey techniques.

     When demersal fish populations are to be sampled in areas  of  sand or
mud bottom, use  of an  otter trawl  is  appropriate.   The  Marinovich 7.62-m
(25-ft)  headrope otter  trawl,  described as  the Coastal  Water  Project
Marinovich net in Table 3 of  Mearns and Stubbs  (1974), is recommended for
this purpose.   It is commonly  used for environmental survey work and is
easily handled from a small  boat (Mearns and Stubbs 1974).   The net  is towed
from a  single warp.  Gear specifications and  sampling procedures  are
critically important.  A discussion is provided by Mearns  and Allen  (1978),
but several  additional points  are important:

     •    Steel  (or stainless  steel) towing cable of 6.35 mm (0.25 in)
          minimum diameter is  recommended.

     •    A power winch is required;  gear  must be recovered while the
          vessel is moving forward.

     •    Effort must be reported in terms of distance  or area
          covered.  Fixed buoys  or  navigational  aids (e.g.,  the Mini
          Ranger) will  be useful  in this context.   Haul distances of
                                  54

-------
           TABLE  7.  SELECTION OF APPROPRIATE FISH SAMPLING GEAR
Demersal
Pelagic
Nearshore
Primary Approach

  Otter Trawl
   (Gillnet,
  Trammel  Net,
   or Trap)3

   Commercial
   Monitoring,
     Gillnet

  Beach Seine
                                              Habitat
Secondary Approach

      Diving
    Submersible
   Hook and Line
 Acoustic Transect
   Pelagic Trawl
Lampara/Purse Seine
a If not trawlable.
                                     55

-------
          700  - 1,000  m (2,297  to 3,281 ft)  are recommended.
          Information on  vessel  speed and haul duration  should be
          recorded but cannot substitute for distance estimates.

      •    A towing speed  (relative to the bottom)  of 1.3 m/sec (2.5
          knots) is recommended.

      •    The gear  should be  towed into  the  current and along  an
          isobath.   Current conditions should be recorded.

      A fundamental  problem of demersal  trawl  sampling concerns the manner
 with  which this  gear samples pelagic forms.  These  species (or life history
 stages) often exhibit schooling behavior.   Incidental encounter of pelagic
 forms occurs during  setting and  recovery of gear; this is inconsistent, but
 individual species  behavior is  not always  understood well enough to permit
 objective exclusion  of these data.   Species that are unquestionably pelagic
 in habit (such as the northern  anchovy Engraulis mordax) should be  excluded
 from  demersal trawl catch data.   This problem can also be addressed by
 separately analyzing data  for  one segment of  the  fish  community  which is
 known to  be demersal  (such  as the  flatfishes).  An  awareness of the
 selective characteristics of trawl  gear  and the behavior of the gear itself
 is also important (Wathne 1977, Harden Jones et  al.  1977).

     Gill nets  and  trammel  nets are often  utilized in areas where bottom
 conditions  preclude trawling or  where  improved  spatial resolution is
 required (such  as  within  the ZID  and  at  comparable stations).   Variable
 mesh, set gill nets are recommended  (Ricker 1971), although  trammel  nets
 (Becker et al.,  1975) may  be appropriate in some  situations.   Traps also
 provide high  spatial  resolution  (Becker et  al., 1975) but  are highly
 selective.

     In situations where  nearshore  fishes  need  to  be sampled  (such as when
an outfall is  located  in an area  of juvenile  salmonid migration),  beach
seine sampling should be  conducted  (Allen et al., 1959).

     Pelagic forms  are often  ignored during  survey sampling.   Pelagic
species are, however,  important  in many  of the  areas for which 301(h)
applications have  been  received.   In  a  situation where the  species  of
                                  56

-------
concern are  subject to consistent commercial or recreational  exploitation in
the vicinity of the outfall,  fisheries  surveys (see below)  can be utilized
to collect appropriate data.   When this  is not possible,  the  use of acoustic
(and sonar)  transects  and pelagic sampling nets (such as  pelagic trawls,
lamparas,  and purse  seines)  is recommended (Saville 1977; Becker et al.,
1975; Lemberg 1978; Fiedler 1978;  Richkus 1980).

     Specific data  can  be collected from  the immediate vicinity  of the
outfall structures by means of diving and submersible surveys.  Details are
provided by  Allen  et al. (1976),  Allen (1975), and Becker et al. (1975).
Line transect techniques often provide an appropriate method  for quantifying
diving observations; quantitative procedures are discussed by Seber (1973).
Some field  methods  are  presented by Fager et al.  (1966)  and  Walton and
Bartoo (1976).

     Hook-and-line surveys are especially  useful for sampling  in precise
locations and  for obtaining  larger  individuals which may avoid  other
sampling gear.   This type of  gear can be  used in areas which  may not be
accessible to other  sampling devices and  is frequently  employed  to provide
specimens for  bioaccumulation  analysis.  Allen et al. (1975)  describe
appropriate  techniques.   It  should be  noted that hook-and-line  techniques
are especially selective in nature; hook  size, bait type, and  other gear
specifications should be selected with this  consideration in  mind.

     Data collection requirements will depend on specific survey objectives.
At a minimum, all  fish catches should be identified to species, and counted
and weighed  by  species.   Taxonomic  procedures and authorities  should be
clearly  defined and a  procedure for seeking  expert advice  should be
included,  if specimens cannot be  identified  by employees  or consultants.

     For individual  species,  length-frequency and length-weight analyses may
be required  to allow consideration of population differences  between outfall
and reference sites.  Standard length is  the recommended measurement; if a
different length  measurement is  recorded, this should be stated, and a
regression relationship between this  measurement and standard length should
be provided  together with  the  data utilized in the analysis.  Appropriate
subsampling  procedures  must be defined for the collection of these data.
When individual observations are  recorded, life history stage should also be
                                   57

-------
 reported.   Unless  subsampling is to be  conducted, all individuals in the
 catches should be examined for external disease symptoms or abnormalities;  a
 standardized  technique for identifying  and defining abnormalities should be
 developed  and included  in the data reports.   Each individual observation
 should be  recorded; the use of computer data coding forms is recommended for
 this purpose.  All raw  data observations,  identified by sample  number,
 station location, date and time of collection,  and individuals responsible,
 should be  included as  an appendix to reports.

 Commercial  and Recreational  Fisheries—

      If commercial  or  recreational fisheries activities  are conducted in the
 vicinity of an outfall,  these  activities  must be monitored.   Commercial  and
 recreational fisheries catch  data are  generally reported as summary
 statistics  for statistical  blocks  defined  by the  state agencies concerned.
 In most cases, the  area  covered by these  statistical blocks  is too large to
 allow  resolution  of fishery  catch conditions close  to a  sewage outfall.

     Coordination  with state  agencies may  provide an effective  and
 inexpensive mechanism  for collecting data  which  can be used to assess  the
 condition  and productivity of specific  fisheries.  A voluntary logbook
 program  for fishermen  could  be designed which would allow  those fishing in
 areas  of concern  to  record data on catch and effort close to  the outfall  and
 at remote locations.   In  some cases,  vessels observed to  be fishing close to
 an outfall could be  identified  and the operators Interviewed when the
 vessels dock.  Similarly,  individuals  observed sportfishing could be
 identified and interviewed.

     An alternative approach to monitoring  recreational fishing  activities
would involve  sampling at the  outfall  and  reference area with  appropriate
sportfishing gear.  This  would allow direct comparison of species caught,
catch  rates,  disease prevalence, bioaccumulation,  and  other  relevant
aspects.

     Market or consumer acceptability of  fish caught in the  vicinity of a
POTW  outfall should  also be addressed  in all  commercial and recreational
fisheries surveys; a simple interview procedure would be  appropriate.
                                  58

-------
     Recreational harvesting of intertidal shellfish  resources occurs in the
vicinity  of  some POTW outfalls.   These shellfish  should  be sampled  with
respect  to  both public health  considerations and  possible population
responses to discharge effects.   While quadrat sampling (Seber 1973)  is  a
theoretically attractive  approach to  this  problem,  the time and effort
involved  would be prohibitive in most  cases.   Monitoring of recreational
catch and effort (in association  with  the appropriate  government agency)  is
an appropriate way to examine relative population abundances.

     When interviews, voluntary logbooks,  and field  observations of fishing
activity  are utilized for  data collection, a complete  log of all relevant
information should be maintained  by the  POTW.   Interviews should  be
conducted by means of questionnaire.  The  log should  contain all  interview
and logbook  returns and detailed  records  of  field  observations; details of
public health  analysis of  shellfish  will also be  included.  All entries
should be identified by time,  date,  and the  individuals who collected the
information. A  clear and comprehensive copy of this log should be included
as an appendix to monitoring reports.

Phytoplankton--

     Since  phytoplankton  are transient,  a  monitoring program  to sample
phytoplankton  should be  designed somewhat  differently  from monitoring
programs  for certain other biotic  groups.  Phytoplankton are carried about
by movements of  the water,  and consequently maximum  sewage effluent impacts
on phytoplankton may occur  well  away  from an outfall.  Stations should be
located at sufficient distance from an outfall to accommodate a lag time in
the response of  phytoplankton to  sewage  effluent inputs.  Due  to their short
turnover  times (on the order of  hours to days), phytoplankton communities
may respond  to perturbations much more rapidly than other  biotic groups;
therefore, samples must be collected more frequently.  Bimonthly samples are
probably  the least  frequent which could  be expected  to give meaningful
results,  although monthly or even biweekly samples would  be preferable.

     In  situations  where  phytoplankton communities display pronounced
seasonal  variations in standing  stock  or  production,  it may  be appropriate
to use a temporally  stratified sampling  approach.  For example,  if
phytoplankton growth is highest during the spring,  sampling may be conducted
                                   59

-------
 on a frequent basis  (e.g.,  weekly).   Similarly, during periods  of
 consistently low phytoplankton  growth,  a reduced sampling  frequency may  be
 used.

     Adequate  assessment of phytoplankton community response  to  sewage
 discharges will generally  involve  more sampling stations  than would  be
 required  for the benthos.  Thus, it is important to initially  assess whether
 or not  a phytoplankton  sampling program  is justified based on  a
 consideration  of discharge size,  sensitivity  of  receiving water, and
 evidence  of previous  impacts on phytoplankton.   Selection  of phytoplankton
 sampling  stations should  always involve a  thorough  consideration of water
 circulation patterns  to ensure  that putative waste  field stations are
 actually  being  exposed  to the diluted effluent and  that  control stations are
 not subject to  influence of the waste field.   If  evaluation of circulation
 patterns  indicates that the sewage waste  field  may be  transported to areas
 of limited flushing (e.g., embayments or eddies), special emphasis  should be
 placed  upon locating  sampling stations in these areas.   In all cases,
 phytoplankton  sampling stations  should be located  in  areas of maximum
 predicted effects, considering such  factors  as  response lag time, effluent
 dilution, and circulation patterns.

     The  most likely direct effect of  sewage effluent  on phytoplankton
 communities is  enhancement or inhibition  of primary production.  Enhancement
 may occur in areas where the  phytoplankton  are  naturally nutrient limited,
 since sewage effluent represents  a  significant source of nutrients.
 Inhibition may  occur if there  are  sufficient concentrations  of toxic or
 inhibitory substances in the  effluent.

     In  areas where phytoplankton production  is enhanced (or  inhibited), the
 standing  stock of phytoplankton  may be expected to be higher  (or lower) than
 in reference areas.  Measurement of  the  concentration of chlorophyll a in
the water is an indirect  method  of analyzing the standing  stock of
phytoplankton.   It  is recommended that  the two-dimensional  spatial
distribution of chlorophyll ^ concentrations  about the outfall  be  analyzed.
Samples should be collected  at  several  distances from the outfall in the
direction  of current  flow.  Samples should  also be collected  at a variety of
depths throughout the  euphotic zone (from  the  surface  to the  1-percent light
level, as  estimated from any  light  transmittance data collected as part of
the Water  Quality Monitoring Program).

                                  60

-------
     The  vertical distribution  of chlorophyll a^ concentrations at each
station may be examined through the collection of water samples with water
bottles at various depths (followed  by  fluorometric or spectrophotometric
determination of chlorophyll  a) or, if available, a pump system may be used
with a flow-through fluorometer (Lorenzen  1966) for a continuous profile of
chlorophyll ^concentrations  vs. depth.   The  advantages and disadvantages of
various water bottles  are discussed by  Venrick (1978a),  while the use of
pumps is  discussed by  Beers  (1978).   It is advisable  that a pump be used
only for  the determination of chlorophyll  a^ and that water bottles be used
for  the  collection  of phytoplankton for productivity measurements and
taxonomic analyses, since the inevitable agitation associated with pumping
may damage some cells.

     If it is determined that the vertical  distribution  of phytoplankton
biomass  (as mg  chlorophyll   a/nr*) is reasonably uniform  throughout the
euphotic  zone, water samples for  simulated in situ primary productivity
measurements (Ahlstrom 1969)  may  be collected with water bottles  at depths
corresponding to fixed percentages of  incident solar  radiation.   If,
however,  there is significant vertical  stratification of the phytoplankton
community,  sampling depths  should be adjusted so  that samples are also
collected within  subsurface  chlorophyll maxima or  minima. Phytoplankton
primary  productivity  should be  measured  by the 14c  light-dark  bottle
technique as  described by UNESCO  (1973),  and measurements at each
station-depth should be replicated to facilitate statistical  analysis.

     If the monitoring program described above  reveals perturbations of
chlorophyll £ concentrations  and/or  primary productivity within  or beyond
the ZID,  taxonomic analyses should  be conducted since phytoplankton species
vary in their responses to alterations in their nutrient source (Eppley et
a!., 1969) or in their responses  to  certain inhibitory substances (Thomas
and Seibert 1977).  Subsamples should be drawn from the water collected in
the water sampling bottles  and preserved for later microscopic analysis
onshore.  It is important that samples for taxonomic analysis be  collected
at various depths throughout  the  euphotic zone since different species may
have different depth distributions.  It  is also important that sampling be
conducted at  similar times  during  the  day (i.e., mid-morning  or
mid-afternoon) since some phytoplankton  are known to migrate vertically
(Stofan and Grant 1978).

                                  61

-------
      Choice of a fixative depends somewhat on  the dominant types of
 phytoplankton  known to inhabit a  given  area.   Buffered formaldehyde and
 Lugol's solution are two common fixatives.   The advantages  and  disadvantages
 of each are  discussed by Throndsen (1978a).

      Taxonomic analysis almost always  involves some form  of subsampling,
 which  is consequently a potential source of  bias or variability.   The
 statistical  implications of subsampling are  discussed by Venrick (1978b).
 Preserved phytoplankton samples normally must be concentrated  for
 quantitative microscopic analysis.  Although other methods  are available
 (Sukhanova  1978, Throndsen  1978b), the routine  method is  the Utermohl
 technique, which utilizes sedimentation  cylinders and an  inverted microscope
 (Hasle  1978).

      Taxonomic analysis should include identification  of the dominant
 phytoplankton  taxa  and counts of  individual  species whenever  possible.
 Numerous  taxonomic  references are  available [see Chapter 6.4  of  UNESCO
 (1978)].  The accuracy  and consistency  of phytoplankton  identifications  are
 of the  utmost importance  for  characterization of the BIP,  both  in  the
 reference area  and at stations in  the vicinity of the discharge.  Counts of
 individual species should be standardized to numbers per unit volume  of  the
 original water-bottle  sample, calculated with  consideration for whatever
 subsampling  technique was utilized.

      If replicate taxonomic samples  are available  for each station-depth,
 the  estimates of  abundance  of  individual species may be  analyzed
 statistically for differences  among  depths,  among stations,  or among  times.
 Particular attention should be given  to differences in community composition
 between  stations in  the vicinity of  the  outfall and  stations  in a  reference
 area.   Species diversity, richness (number of species), evenness,  or
 numerous other  parameters (Pielou 1970)  may  be utilized for  description  and
 comparison of the phytoplankton communities.

     If  available information  indicates a  potential for enhancement of
 individual phytoplankton groups (especially dinoflagellates),  the monitoring
program  should  include  an assessment of  the  magnitude,  duration,  and point
of initiation for phytoplankton  blooms.   Special emphasis should  be placed
                                  62

-------
upon species causing accumulation of  toxins  in other organisms, or blooms
which may  result in fish kills.

     The goal  of  the phytoplankton monitoring program should  be  to
demonstrate whether or not the discharge of  sewage effluent from the outfall
in question interferes with the protection and propagation or the natural
range of variation  of phytoplankton in areas beyond  the ZID.
Characteristics of phytoplankton which may be examined include the community
biomass (as  estimated through  the  measurement of chlorophyll £
concentrations),  community primary  productivity (as estimated  through
simulated  in situ incubations using the  light-dark bottle technique),  and
the various community composition  parameters.  The responses of biological
communities to pollutant stress appear to involve a continuum, as indicated
by the gradients in the biological variables of the benthos near sources of
organic pollutants (Pearson and  Rosenberg 1978).  Therefore, alteration to
the phytoplankton communities should be analyzed in relation to potential  or
determined (by  other community studies) impacts on other biological
communities which make up a balanced indigenous population.  This analysis
should include, although it should not be limited to, food web impacts,  the
occurrence of toxic or nuisance phytoplankton, eutrophication or blooms,  and
potential  second impacts on zooplankton or fish  communities.

Zooplankton—

     Zooplankton, like phytoplankton, are  transient,  and, consequently, a
monitoring program  designed to  sample zooplankton should be designed
somewhat differently  from monitoring programs for certain other biotic
groups that tend  to  be permanent residents  of an area.   Zooplankton  are
carried about by movements of  the  water;  therefore, the maximum sewage
effluent impacts on zooplankton may occur  at  some distance away from  the
outfall.  Unlike  phytoplankton, however,  zooplankton life spans are
typically on the order of a few months, so  their capacity  for responding  to
perturbations is much less than that  of  phytoplankton.   Bimonthly samples
are usually adequate for  analysis of  changes  in zooplankton communities.
Zooplankton possess varying degrees  of swimming ability and therefore have
the potential  for aggregating in patches or  in narrow depth strata,  which
introduces an additional complication in quantitative sampling.   In
addition,  the ability to swim means that  many zooplankton  can avoid certain
types of sampling gear.

                                  63

-------
      As  is the case for phytoplankton, the  design of zooplankton sampling
 programs  should  consider natural temporal  fluctuations  in  abundance and
 species  composition.  Zooplankton  assemblages display a high  degree of
 spatial  heterogeneity in  addition to pronounced diel vertical  migrations by
 many  groups.  These factors, combined with a  longer response time to  effects
 of sewage effluents,  would result in the  necessity of conducting relatively
 extensive programs (i.e., in  number  of sampling stations,  frequency of
 sampling, and replications) to adequately  assess responses of zooplankton to
 pollutant inputs.   Thus,  studies of  zooplankton assemblages should be
 conducted only when there is evidence of previous impact in zooplankton,
 when  phytoplankton communities  display significant effects,  or when large
 discharges are located in areas where  there is  a  high potential  impact on
 zooplankton (e.g., in  estuarine environments with important macroplanktom'c
 larvae of commercial  and  recreational  species).

     For zooplankton,  there is no easily measured functional  response to
 POTW effluent discharges  similar to primary  productivity for phytoplankton.
 Toxic effects of  effluent  on zooplankton  are possible  if there  are
 sufficient concentrations of toxic substances in the  effluent.   Alteration
 of zooplankton community composition is  a  distinct  possibility in areas
 where the phytoplankton community  composition has been affected,  since many
 zooplankton graze on phytoplankton.  Given the smaller proportion of their
 life spans spent  within the sphere of influence of  the outfall,  zooplankton
 are less  likely  to experience direct,  observable changes  in  community
 composition than  are phytoplankton.

     The  zooplankton encompasses a wide  range of organisms,  from microscopic
 protozoans to  large planktonic crustaceans and larval fish.   In general,  the
 smaller organisms have shorter life spans, so effluent impacts are more
 likely among the  smaller  organisms.   Sampling methods  vary depending upon
 the size  of the organisms.

     Microzooplankton (those which pass through the mesh  of  a 202-um net)
can be collected either with  water  bottles similar to those  used  for
phytoplankton collection [although  a  volume of at least 10  liters is
recommended  (Jacobs and Grant  1978)]  or with pumping systems (cf., Beers et
al.,  1967).  If water bottles  are  used, samples  should  be  collected from a
                                  64

-------
variety of depths  throughout the water column,  and the captured organisms
may  be  concentrated with  a fine-mesh (e.g.,  63-um) screen.  Replicate
samples should be  collected from each station-depth.  Pumping systems may
incorporate a single filter or  a mesh  of different size filters for
collection of the  organisms.  Flow rate should be in the range of 150 - 200
1/min (Jacobs and  Grant 1978).  Pumping systems have the advantage of being
able to take samples integrated  over  depth,  or of collecting samples while
the  ship is underway, but  they may damage soft-bodied organisms, and they
are more expensive and complicated than water bottles.

     For collection of  small mesozooplankton  (those retained on a 202-um
mesh), nets are generally used.  For  an excellent discussion of net design
and function, see  UNESCO (1968).   Nets with  small mouth diameters (20 - 40
cm) may introduce  error by underestimating abundance and diversity (McGowan
and Fraundorf 1966, Wiebe and Holland  1968).  A minimum mouth diameter of 60
cm is generally recommended  (Jacobs  and  Grant 1978).   With mesh sizes of
less than 202 urn,  clogging and  loss  of filtration efficiency are often a
problem,  so  a  202-um  mesh is the  smallest  which should  be  used  (UNESCO
1968).  Additional  tows may have to be made with larger nets (1.0-m mouth
diameter and 505-um mesh)  in order to collect  representative samples of
larger zooplankton and larval  fish.   All tows  should be  replicated; the
number of replicates necessary for the desired precision of  estimates should
be determined during a preliminary  or pilot  sampling  program (cf.,  Cochran
1963; Green 1979).   Paired bongo  nets (McGowan and  Brown 1966) are often
used because they  provide two replicate samples from  the same environment.
Alternatively, they can be rigged with two  di f ferent mesh nets  for
collection of different  size fractions.   Any  net used  should have a flow
meter attached to  the mouth  for  calculation of  volume filtered.  Prior to
selection of a sampling technique,  the spatial  and temporal distributional
characteristics of  the target zooplankton  assemblage  should be considered.
For  example, lobster (Homarus  americanus) larvae occur  near the water
surface and are appropriately sampled  by a neuston net.

     Oblique tows  are highly recommended,  with  sampling  extending from just
above the bottom  to the surface.   Avoidance  by  larger  zooplankton  is
significant at  slow tow speeds,  so  a ship's speed of 1.5  - 2 knots should be
maintained (Jacobs  and Grant 1978).   The  animals collected  should be washed
(from the outside  of the  net) into the cod  end, transferred to a  labeled
                                  65

-------
sample jar,  and preserved with buffered formalin in the proportion of nine
parts sample to one part formalin  (a  saturated solution containing 38- to
40-percent  formaldehyde).   Further discussion of  shipboard handling of
samples can  be found in Jacobs and Grant (1978).

     For quantitative  taxonomic analysis of the zooplankton  samples,
subsampling  will normally be  required.   Two methods which are  commonly used
are:  subsampling by Stempel  pipette, and splitting with  either a Folsom
splitter (McEwen et al., 1954) or the  newer Burrell et al. (1974) device.
In either case,  large  and/or  rare  organisms  should first be counted and
removed.   The use of the Stempel pipette  is discussed by Jacobs and Grant
(1978), who  point out that due to the small aliquot size, this  method should
only be used when rapid "ballpark" numbers are  needed.   The  use of plankton
sample splitters  is also discussed by  Jacobs  and Grant (1978), and they
indicate  that  this is  the   best  method for quantitative analysis  of
zooplankton.  The counts of individual  taxa should be transformed to numbers
per sample (considering the subsample  size) and then standarized to numbers
per unit volume of water filtered (calculated from the flowmeter reading).

     Taxonomic analysis of the samples should include identification of the
dominant  zooplankton  taxa  and counts of individual  species whenever
possible.  Particular attention should be given  to the meroplanktonic larvae
of commercially  and ecologically important species (e.g.,  fish, shrimp,
lobsters,  etc.).   The accuracy  and consistency of  zooplankton
identifications are of  the utmost  importance  for characterization of the
range of variation of  the zooplankton communities, both in  the reference
area and at  the various outfall stations.

     If replicate taxonomic  samples are available  for each  station, the
estimates  of abundance  of individual  species  may be analyzed  statistically
for differences among stations or among  times.   Particular attention should
be given  to  differences  in community  composition between  stations in the
vicinity  of the  outfall and stations in a  reference area.  Species
diversity,  richness  (number of species), evenness, or numerous  other
parameters (Pielou 1970)  may  be utilized for  description and  comparison of
the zooplankton communities.
                                  66

-------
     The goal of the  zooplankton monitoring program should be to demonstrate
 whether or not the discharge of sewage  effluent from the  outfall in question
 interferes with the  protection and propagation or the natural  range  of
 variation of  the zooplankton communities in  areas beyond the ZID.
 Characteristics of the zooplankton communities should include,  but not
 necessarily be limited  to,  species composition,  abundance, dominance, and
 diversity.  The  responses  of biological communities  to  pollutant stress
 appear to involve a continuum, as  indicated  by the gradients in biological
 variables of the benthos near sources  of  organic pollutants (Pearson and
 Rosenberg 1978).  Therefore, alteration  in the zooplankton communities
 should  be analyzed  in  relation  to  potential  or  determined (by other
 community studies)  impacts on other biiological communities which would make
 up a balanced indigenous population.  This analysis should include,  although
 it should not be limited to, the structure and function of both larval and
 adult zooplankton communities, as well as consideration of food web  impacts.

 Kelp Communities--

     Kelp beds  are distinctive habitats  of  limited  distribution whose
 protection is of special concern because of their ecological  significance
 and their economic  value to man.   The  kelp  plants themselves are largely
 responsible  for the spatial structure  of this community,  as they provide
 food, substrate,  and  shelter for a variety of organisms (Tegner 1980).   In
 some areas,  the kelp  itself  is  harvested,  and in many areas the kelp beds
 are the location  of valuable fisheries for abalone,  lobster,  fishes,  and sea
 urchins (Tegner 1980).  Kelp beds  may be particularly sensitive  to outfall
 discharges,  and adverse effects  of  sewage effluent  on  kelp  have been
 suggested by  Carlisle (1968), Mearns et al.  (1977),  and others.   If  kelp bed
 communities are  potentially affected  by a  sewage effluent discharge,  a
 monitoring program should be conducted  to evaluate  the health and extent of
 these communities.

     Kelp bed communities typically include a great variety  of plant and
 animal species,  but  since  the continued  existence of the community  is
 largely  dependent on the presence of  the  kelp plants themselves,  the
monitoring program should focus on  the  health and spatial  distribution of
these plants,  rather than  attempt a  detailed analysis of the entire
community.   The  location,  condition,  and  size of kelp  beds  along  the
                                  67

-------
southern California coast have been monitored for a  number of years through
the use of aerial  surveys (Wilson et al.,  1980).   This is a particularly
useful technique because changes  in the area!  extent  of  the beds can  be
monitored over  time, and areas of potential sewage impact can be identified.

     Aerial  surveys of kelp  beds utilize  infrared photographs taken from  an
aircraft flying at an altitude of 1.5 -  2.6 km (0.9  - 1.6 mi).  Photographs
taken around midday will  minimize reflected  glare,  and a polarizing filter
may be  used, if necessary.   Overlapping adjacent photographs (10 -  20
percent) assure full coverage and minimize  barrel  distortion at the film
edges.  Slides  of the kelp beds can  be  projected and drawn onto charts  of
the coast,  and  the surface area of  the kelp canopies can be calculated from
these charts using a polar planimeter or a measured  grid network (Wilson  et
al., 1980).

     If a decline in  nearby kelp  beds  occurs, the discharge of sewage
effluent may or may not be  the cause  or a cause.  Sewage effluent may
adversely  impact kelp  in  a  variety of ways.   Certain  constituents of the
effluent may be toxic to kelp sporophytes  and/or  gametophytes,  and the
plants  may  die in  areas where  destructive concentrations of these
constituents appear.   Turbidity within  the discharge field of the outfall
might be increased,  reducing  light  intensities or altering the spectral
distribution of the light  such  that kelp growth  is  adversely  affected.
Concentrations  of kelp enemies such as  grazers, pathogens,  or parasites
might in some way be enhanced by  the effluent,  and the increase in their
numbers might bring about a  decline in the kelp.   Finally, siltation effects
from  the settling of  suspended matter discharged by  the  outfall might
interfere with  the recruitment of young kelp plants (North 1964).

     There  have been suggestions (Wilson  et al.,  1980) that sewage effluent
discharge may have decreased  the maximum depth of  kelp growth in nearby kelp
beds.   If  the  light transmittance  data collected as  part of the water
quality monitoring program is  also collected on a  regular basis along the
seaward edge of nearby kelp  beds,  it should  be  possible,  given the known
photosynthetic  requirements  of kelp (Clendenning  1964;  Rosenthal  et al.,
1974), to determine whether  the sewage effluent  discharge may be adversely
impacting the kelp beds.   Comparison  of  light transmittance measurements  at
specific depths  along the  kelp bed  in question with  those at similar depths
                                  68

-------
 near other  kelp beds removed  from  anthropogenic sources  of  sedimentary
 material should indicate whether the kelp bed in question  could extend into
 deeper water in the  absence of the effluent discharge.

     The use of sediment traps to  quantify  the amount  of sedimentation
 occurring along the  margins of potentially impacted kelp beds should also be
 considered.   Comparison of the rates  of  sedimentation  there with those along
 kelp beds removed from anthropogenic  sources of sediments,  and consideration
 of the effects  of different amounts  of  sediment on the survival and growth
 of kelp germling stages (Devinny  and Volse  1978), may permit an evaluation
 of whether or not the sewage effluent discharge may be inhibiting expansion
 of the kelp  bed in question.

     While the  toxicity of certain effluent constituents on kelp has been
 studied (North  1964), it  is  probably unreasonable to expect that detailed
 studies of toxic effects  would be conducted  as part of  a kelp monitoring
 program,  given the  large  number  of  potentially  toxic  or inhibitory
 substances in most municipal  effluents  and  the  possibility for complicated
 synergistic  effects.

     Increased  abundances of certain animals  which graze on kelp [notably
 sea urchins  (see Lawrence 1975)], have  been  implicated  in the decline of
 certain kelp beds.  While there have  been suggestions that  the abundances of
 these grazers may  be enhanced by  the discharge of sewage effluent (Clark
 1969),  it is difficult  to establish cause-and-effect relationships.   The
 true cause  of  the increased  abundances of these organisms  may only  be
 revealed through detailed investigations of interspecific interactions  and
 predator-prey relationships (Tegner 1980),  which may be beyond the scope of
 individual 301(h)  monitoring programs.

     One promising method which may be  used  to infer causes  of observed
 changes in kelp  canopy size is regression analysis, utilizing such factors
 as suspended solids, mass emission  rates,  water  temperature, water
 transparency,  etc.   Wilson et  al.  (1980) used  this  method to examine
 potential  causes of the initial  decline and subsequent  recovery  of kelp beds
 along the  Palos  Verdes Peninsula,  in  close proximity to a  very  large sewage
outfall.   Nevertheless, identifying changes  which have  occurred in kelp
canopy size  is  somewhat easier than  deciding what factor(s) may  be
 responsible.

                                  69

-------
 Coral Communities--

     In  some areas,  it will  be  important to assess any impact the
 applicant's discharge  may have  on coral  communities.  Generally, an
 assessment of changes on living coral coverage which may include a study of
 reef fishes, will  be sufficient rather than a complete  study of the  reef
 community.

     A line transect method of sampling  should be used for  studies of  both
 the  living coral coverage  and reef  fishes.   All  stations should be
 comparable as far as the distance  from sand areas, rubble,  and base-rock
 relief.  At each  station, a 50-m (164.1-ft) length of electrical wire should
 be permanently attached to the reef, parallel to the shoreline.  Care should
 be taken to ensure that the line is  located within a reef area  of sufficient
 size so as to eliminate any patchiness in the data due  to sand area or  reef
 edge effects.

     Photographs  of  at least 0.67  m2 (7.24 ft2) of the bottom should be
 taken on the shoreward  side  of the line at  5-m  (16.4-ft)  intervals.  An
 underwater camera mounted on  a  rigid framing  device should be used.  Each
 photograph should contain a  small  slate indicating the station, date, and
 position of the photograph along the line.  Care shall  be taken in order to
 be certain of photographing the  same quadrat each quarter.   It is suggested
 that stakes be driven or cemented to the  reef  indicating  at least two
 corners of each frame.

     The photographs should be developed as  slides.  These slides should be
 projected onto a  grid having the dimensions of the original quadrat, and the
 percent  coverage of coral and  encrusting  algae by  species, and  of the
 noncoral  substrate,  should be estimated.

     Other transecting methods may  be used to  sample living coral  coverage
 if they are shown to be statistically valid sampling  techniques;  however,
 due to  the relative ease of sampling and data reduction, the photographic
method  described above is recommended.
                                  70

-------
      The investigation of reef  fishes should consist of SCUBA transects
 along the transect  lines  used for the reef benthos survey.  Beginning at
 least 1  day  after the lines  are  permanently  attached  to the  reef, a diver
 should swim  each line in order  to identify  and  count the fishes located
 within 3 m (9.8  ft) either side of the line.  The diver should take care to
 enter the water  away from the transect  area in order to  avoid disturbing the
 fishes.   One transect  at  each station should be completed during each of
 three consecutive days for each quarter.

      Consistency of technique is  important;  therefore,  the applicant should
 make every effort to ensure  that  transects are conducted in a similar manner
 by the same  diver-biologist  if possible.   Some investigators are known to
 look for cryptic species  more than others, or to  notice larger fish high in
 the water column more readily.  Such sources  of  variability should be kept
 to a minimum.

      Where applicable, those quality assurance/quality control  (QA/QC)
 procedures specified for  the  benthos  should be followed  for the coral  and
 fishes.   These procedures  include the  maintenance of a  voucher collection,
 verification of identifications,  inclusion  of raw  data sheets in reports,
 etc.   In  addition,  it is  important that the applicant maintain consistency
 between  sampling periods  in order  to reduce sampling  variability.

     The monitoring  report should  contain  a  tabulation of  the percentage of
 living coral, coralline  algae,   and  coral  rubble  for  each  photographic
 quadrat.   A tabulation  of the  number of individuals  of each species  of  fish
 identified on each transect should also be  included.

     A field  notebook, as discussed in  the benthos section above,  should be
maintained and  submitted along with the monitoring  report.

 Intertidal Communities—

     Due  to the offshore location  of most marine sewage outfalls, monitoring
of intertidal  communities will  usually not be specified as  part of  the
biological monitoring program.   Nevertheless, intertidal communities  are
sensitive assemblages  of organisms  which  may be  affected by  sewage
discharges (Dawson 1959, 1965).   In cases  where shoreward transport of  the
                                  71

-------
waste field  is predicted, monitoring  the  intertidal community  may be
required.

     The type of intertidal  community is largely determined  by the type of
substrate (tidal  flats,  sandy beaches, rocky shores,  gravel  and cobble
shores), and the appropriate sampling procedures vary considerably among the
various  communities  represented in these  environments.  Conor and Kemp
(1978) provide  a comprehensive review of the procedures used in quantitative
ecological  assessments in the various types of intertidal environments; it
is unnecessary  to  provide a detailed reiteration of the sampling procedures
described therein for each habitat and biotic  group.  Selected sampling
procedures  for  rocky intertidal  habitats will instead be described in order
to illustrate some of the basic  principles involved.

     A common attribute of many  intertidal communities is the  stratification
of the community with respect to tidal height, since many intertidal species
have discrete vertical limits  within the tidal range.  Since  both community
species  composition  and  density vary considerably  with slight changes of
vertical distance (on  the order of tenths of a meter), a  sampling plan
designed to monitor the entire intertidal community should be stratified by
vertical height.   Comparison of communities  between reference areas and
discharge impact areas  should be between  samples from similar elevation,
also.  If data  for an entire  transect across  the  intertidal  community are
pooled,  it is  unlikely that  differences between  reference and discharge
impact areas could be detected because the within-transect variance produced
by combining data  from different levels  would be enormous.   For an example
of the application of these principles, see  Batzli (1969).

     In  order  to  detect differences between  areas,  or changes  at one
location through time which  arise from anthropogenic perturbations, these
differences or  changes should be distinguished from the natural spatial and
temporal variations at a given location.   Field sampling should, therefore,
quantify the spatial and temporal  heterogeneity of each site.  Intertidal
sampling is typically  conducted along  a  transect.   Samples are usually
collected (or a census of the community  is  conducted) at various locations
along the transect.  These locations may be  spaced evenly along the transect
(systematic sampling) or randomly along the  transect (random sampling).  The
advantages and disadvantages  of  each are discussed by Cochran  (1977).
                                   72

-------
 Considerations for the selection  of  the number of transects, the length and
 width of  the transects, and the  number and size  of  the sampling units are
 discussed by Gonor and Kemp (1978).   Samples are normally collected (or the
 census of the community is conducted)  within quadrats along the transects
 (i.e., plots of constant area).  The size of the quadrat is a function of
 the nature of the species  to be examined,  their relative abundance, and the
 cost of collecting (or  conducting  a census of)  the  organisms.   Gonor and
 Kemp (1978) recommend that a preliminary  sampling program be conducted in
 order to  investigate  the variability of  a given area and to determine the
 relative  efficiency of various quadrat sizes and numbers of samples.

      Sampling intertidal communities  can  be either destructive (in which a
 quantitative sample  of the biota  is  removed for later  analysis), or
 nondestructive (the acquisition  of  similar data  using methods which do not
 disturb the communty).   Aside from  a  desire to minimize the damage done to
 an  area,  there exists  a  second reason  for  favoring nondestructive sampling,
 i.e.,  that repeated sampling of an area could be biased by the  effects of
 previous  sampling.   It is known,  for instance, that reduction in the
 abundance of certain  keystone species  (Paine 1969) may alter the rest of the
 community in such a manner that  change due  to other events may be
 undetectable.

     Nondestructive sampling can generally be used  if the species to be
 sampled are visible, measurable,  and unaffected  by the sampling procedure
 used (Gonor and Kemp 1978).  Both the macroalgae and the sessile macrofauna
 on  rocky  shores may be sampled  without  removal.  Individuals  within a
 quadrat can be counted  and measured in situ  for some  suitable dimension
 (e.g., percent cover)  which can be converted into an estimate of  biomass
 (Gonor and Kemp 1978).  Nondestructive sampling is inadequate  for both small
 and mobile species, however, so if the  entire community  is  to  be censused,
 some destructive sampling must  occur.   For most impact studies, however, it
may be sufficient to only examine effects on  the  plants and animals which
can be sampled nondestructively.   Nondestructive sampling  in the field may
be supplemented with  photography (Littler 1971), which  is  particularly
useful  when time in the field is  severely limited.

     Littler and Murray (1975) investigated the biological  effects of a
low-volume domestic sewage discharge on  the  intertidal  community on San
                                  73

-------
Clemente  Island, using sampling techniques which  illustrate well the
sampling design  considerations described  above.   Their study could be used
as a model for  monitoring programs to be  conducted in other  intertidal
communities.

     Littler  and Murray (1975)  determined the distributions and abundances
of macro-organisms with reference  to  tidal height and  distance from the
outfall.  They utilized  a  photographimetric  technique (Littler 1971) for
assessing standing  stocks (i.e., frequency  and  cover)  of species  popu-
lations.  Sampling was restricted  to macro-epibiota which could be discerned
with the unaided eye in the field  or in photographs.  Photographs were  taken
of ring quadrats 30 cm in diameter (providing 0.07-m2 stratified plots) at
1- or 2-m  intervals along transects both in the outfall area and at randomly
selected points  in  control  areas.   The  control  areas were sufficiently
remote from the  influence of  the outfall  and had morphometry similar to the
area in the immediate vicinity  of  the discharge.

     Cover was determined  (Littler  and Murray 1975)  from the photographs
using a point-intercept method.  If  species were  observed within a quadrat
but were absent  from the scores,  they  were  assigned a cover value of 0.05
percent.  In  some cases,  samples  contained multi-layered algal canopies;
thus, total cover was in excess of 100 percent.   In these cases, more than
one photograph had to be taken  per quadrat to measure stratification.   Field
notes  were taken using a tape recorder,  which then facilitated later
taxonomic  analysis of the samples.

     Vertical heights for each quadrat were  measured from fixed reference
points  using a  sighting  level,  a  stadia  rod,  and  standard surveying
techniques.   Relative tidal  heights  were  referenced  to  the level of
mean-lower-low-water (MLLW).

     Sampling in  different seasons  (Littler and  Murray  1975)  showed that
seasonal  changes  in standing  stock were minor,  especially  in the
sewage-affected  area, so  data for all sampling  periods were considered  to be
representative and were grouped as either outfall or control  samples.  If
seasonal changes had been  significant, comparisons of outfall  and control
samples would probably have had to be restricted to within a given season.
                                  74

-------
     The intertidal communities  were stratified  by 0.15-m intervals of tidal
 height.  The distribution of species  populations as a function of tidal
 height was then compared between the  outfall  and control areas.   The
 community  features of species  diversity,  stratification, and species
 assemblages were analyzed using  the  cover  data.  Cluster analysis  was
 utilized to objectively determine  natural  assemblages or groupings of
 organisms.

     The techniques utilized by Littler  and Murray  (1975) could easily be
 applied  to  monitoring  programs in other rocky intertidal  environments.
 While certain of  the principles  involved may  apply  to other environments,
 sampling techniques will probably differ, and  the more complete discussion
 of Conor and Kemp (1978) should  be consulted.

 Analytical  Techniques

 Introduction--

     The evaluation of compliance  with  BIP  maintenance requirements
 necessitates the  analysis  of biological  monitoring  data including
 comparisons of spatial and temporal  variability in the composition  and
 structure of the benthic macrofaunal  assemblages and other selected  biotic
 groups  among monitoring stations.    A  wide  variety of techniques  are
 available for such analysis of biological  data collected  as part of a  301(h)
 monitoring  program.  Analyses  may range from rather simple qualitative
 (tabular or graphical) comparisons  of species distributions to complex
 multivariate tests of the  relationships  of community  structure to
 environmental variables. The following sections provide  discussions  of  the
 applicability of several analytical techniques;  however,  no single approach
 is recommended for analyzing the monitoring program data.  In most  cases,
 the optimal  approach will  be to utilize  several techniques.   Selection of
 the number  and kinds of analytical techniques  employed in each  case will
 depend  upon  the  magnitude  of the monitoring program (e.g.,  number of
 stations and  sampling  frequency), the  type of data collected, and  the
 distributional characteristics of the  data.  As  an  example, a biomonitoring
program of moderate complexity may include the following analyses  of  the
benthic macrofauna:
                                  75

-------
     •    Calculation of community  parameters such as diversity and
          species richness

     •    Graphical or tabular  display of  abundances of dominant
          organisms or indicator  species

     •    Parametric or nonparametric  statistical analysis of total
          organism density, densities  of  individual groups,  species
          number, and community parameters

     t    Numerical classification  of  species abundance data (e.g.,
          dendrogram).

     Described below are analytical procedures recommended for use  by  the
applicant in  conducting the required comparisons.   Emphasis has been  placed
on demonstrating the applicability of each analytical approach to the type
of monitoring  required.   These descriptions are intended as concise
introductions to techniques, each  of which  is the  subject of numerous texts
and technical papers.  Many of the  more important references are included
under each description.  The reference lists are not exhaustive, but they do
provide a starting point for gaining access to the  literature.

     The formulation of plans  to  analyze results properly occurs during  the
development phase  of  the sampling design.   Some statement must be made
concerning the expectation  of the type of data that will be  developed  and
how that information will  be used  to address the issue of discharge impacts.
If preliminary  studies have  identified  substantial  site-specific
information,  sampling objectives  should be defined in detail.   As discussed
above  in  the section concerning sample frequency and  replication,   the
ability of selected analytical techniques  to detect differences in  target
parameters among monitoring stations must also be assessed.

     Limitations are inherent in each  of the analytical  methods described
below.   Therefore, the  inappropriateness of the singular use of  any
technique is  stressed.  Individual methods or statistical models may be of
considerable  utility in summarizing  biological parameters; however, because
important assumptions  are specific to each method,  a single  analytical
technique cannot accurately  guide the interpretation of the  monitoring
program results.

                                  76

-------
 Simple Graphic Displays—

     Although statistical  analyses of monitoring  data are generally
 necessary for demonstrating compliance with the 301(h)  biological criteria,
 graphical presentation  of results provides an additional, and generally very
 useful, means of making comparisons among sampling  sites.  In many cases,
 the  response of a biological parameter  may  be so pronounced that an effect
 is clearly evident in  a graphical presentation  and  a  detailed statistical
 analysis would not be  necessary.  Graphical  displays are also important in
 presenting summaries of large amounts  of data in a concise format.

     The following types of graphical  presentations  are  recommended for
 display of biomonitoring program results:

     1.   Community parameters (e.g., abundance,  species  richness,
          diversity) at sampling stations

     2.   Trellis diagrams or dendrograms of station similarities

     3.   Maps of faunal assemblages near  discharge  (generally  in
          cases  with  large   numbers  of  sampling  sites in  a
          heterogeneous environment).

Examples  of  appropriate graphical  displays of  biological  data collected  near
marine sewage discharges  are included in  Figure 2.  A  discussion of
techniques is  included in Green  (1979).

Parametric Techniques--

     Parametric  statistical  techniques such as Student's  t-tests and
analysis  of variance  (ANOVA)  are recommended for comparing  measures of
abundance and community  structure  among sampling stations.   When it is
hypothesized that outfall effects are evidenced  by measurable differences
between monitoring stations,   these  statistical  models can  be  used to
distinguish outfall-related impacts from  natural  variability  in community
structure.
                                  77

-------
r
  i.o
  OB
  Off
5
I
                        KEY-
           DW
                                 URFACE
                               -MIDWATER
                           STATION
                                              O£
OS
                          AW     AE
                            STATION
Figure  2.   Examples of  graphical  displays  of  biological data
            from a marine  sewage discharge  site.
                             78

-------
     As the name implies,  these  tests  involve null hypotheses concerning a
 statistical  parameter of  the  variable being measured (e.g., population
 means).  However,  as such, they have  specified assumptions concerning the
 distributional characteristics of the sample  data.  For ANOVA, it is assumed
 that the error  terms  of  the variates  in  each sample are  independent and
 normally  distributed and that the sample  variances are  not different.
 Independence of error terms is primarily associated with adequacy of
 experimental  design.   The  remaining two  assumptions can be tested following
 data collection.  If the data do not meet the assumptions,  transformations
 can sometimes be applied to correct deviations  from  the assumed
 distributions.  Discussions of  transformations prior to ANOVA are found in
 Sokal and Rohlf (1969), Downing (1979), and Green (1979).

     In many  cases,  a  single transformation  can correct both non-normal and
 heteroscedastic data.   It  is important to note, however,  that deviations
 from normality,  especially  in cases  of  large  sample size, will  generally not
 influence the overall  test  results  to the same  degree as heteroscedasticity.
 Correlation of variances with means is a frequently encountered problem in
 samples of organism abundances.   In  many cases, such  violations  of the
 variance assumption  can be corrected by a logarithmic transformation.

     ANOVA is used  to  test the hypothesis  that there are no differences in
 the biological  observations made at different sampling  stations.   In
 addition to evaluation of a single factor (e.g., stations), ANOVA models are
 especially appropriate for evaluation  of the importance of  multiple factor
 level effects (e.g., depth, times)  on the mean  value of the  dependent
 variable.

     The t-test  is statistically equivalent  to ANOVA  when only  two samples
are being  compared.   The t-test is  appropriate  for such  two-sample
comparisons;  however, it should be emphasized  that the  test cannot be used
to evaluate multi-sample hypotheses by testing all  possible  sample  pairs.
In such  cases  the probability of committing  a  Type I  error  is considerably
higher  than the  designated   level for each t test.

    Discussions of the applications of ANOVA  to  biological data are  found
in Zar  (1974), Sokal and Rohlf (1969), and Winer (1971).
                                  79

-------
      If a significant effect  is indicated  in  an  ANOVA, an a posteriori
 multiple  comparisons test should be used to  identify where differences are
 located  among  the  group means.   The most commonly  used a posteriori
 procedures are the Student-Newman-Keul s test (SNK), the least significant
 difference test (LSD),  and Scheffe's test.   Dunnett's test should be used if
 only  the  control mean is to be  compared with all  other group means rather
 than  all  possible comparisons.   The characteristics of multiple comparison
 tests are described in  Zar (1974).

 Nonparametric Techniques--

      If sample data  do  not meet the assumptions of parametric statistical
 tests, analogous  nonparametric tests may  be employed in  the analysis  of
 differences among  stations.    Nonparametric tests do  not  utilize  null
 hypotheses associated with statistical  parameters  and there are typically  no
 assumptions concerning the  distribution of  the variates.   An additional
 advantage of nonparametric tests is that they  can  be used to test ordinal  or
 nominal data in  addition  to numerical values.

      Nonparametric tests  have a  lower  power (i.e., 1-3)  than  the  analogous
 parametric  procedures.  For example,   a  nonparametric  ANOVA has a
 power-efficiency  of 95.5 percent  when  compared with the  F  test.  Thus,
 nonparametric tests should  not be applied  if  the sample data  meet the
 assumptions  of parametric techniques.   Nonparametric tests  are  also unable
 to test interactive  effects in the ANOVA  model.

     Examples  of some common  nonparametric  tests and their applications are
 shown in Table 8.  A comprehensive discussion of nonparametric  techniques is
 provided in  Siegel (1956) and  Hollander and Wolfe (1973).

     As for  parametric ANOVA,  an a posteriori  test should  be used following
 determination  of a significant overall effect in the nonparametric analog of
ANOVA.  A  multiple comparisons  test for equal  sample sizes analogous to the
SNK test described in Zar (1974)  and a procedure for unequal sample size is
presented  in Dunn  (1964).
                                  80

-------
        TABLE  8.  EXAMPLES OF  SOME NONPARAMETRIC  STATISTICAL TESTS
        Test
Mann-Whitney U-test
Kruskal-Wallis one-way ANOVA
Friedman  two-way ANOVA
X2 test  (or G-test)
Test of whether  two  independent  samples
are from same population (analogous to t
test)

Test of whether  K independent samples
are  from different  populations
(analogous to F test)

Test null hypothesis that  K matched
samples are from same population

Test of independence  of frequencies in K
samples.
                                  81

-------
 Multivariate Techniques—

     Multivariate  numerical  methods are used  to  reduce and order large
 matrices of data.   They effectively  summarize trends  or patterns in the  data
 that are not otherwise observed  from visual  examination or univariate
 analyses, and have  been used to explore the interrelationships between  sets
 of  biological and concomitant  physical-chemical observations.  Their  most
 common ecological  application involves a search  for patterns in measured
 biological   variables which can  be related to  patterns  in measured
 physical-chemical parameters.   The  goal  in these analyses is to explain the
 effect  of  environmental  variables  on both community  composition and
 structure.   Examples of multivariate methods are  discriminant analysis,
 multivariate ANOYA, ordination  techniques, and numerical  classification
 analysis.

     Most multivariate tests have distributional  assumptions analogous to
 the univariate case, the most  important of  which is equality of dispersion
 matrices.   It is  assumed that the vari ance-covari ance matrices are
 independent of group means and are not different  among  groups.   However,
 with increasing numbers  of variables (p)  in  the  multivariate case, the
 chance of detecting a significant difference  becomes relatively high since
 there  are  0.5 p  (p +  1) variances and  covariances.   Heterogenous
 variance-covariance matrices will result in an  increase in the probability
 of a Type I  error, i.e.,  that  a significant difference between groups  will
 be indicated  when one does  not actually  exist.   In general, the  potential
 for variance  heterogeneity can  be considerably reduced  by use of equal
 replication,  large sample  sizes,  and relatively  few variables.

     The increased probability of  a Type I  error is especially  important
when the overall power  of multivariate tests  is considered.   As  indicated by
Green (1980):   "When  formal  multivariate tests are made,  their power
 (especially with many variables) is so great  that a significant  result is
probable."  Thus, the investigator  should consider carefully  the  intended
purpose(s) of multivariate  analyses  before they are applied.   Specifically,
it should be  determined  if  the objectives  are to reduce large data  sets  into
a manageable  format,  to  evaluate general relationships to environmental
variables, or to test null  hypotheses that  no differences  exist  among sites.
                                  82

-------
     Numerical classification methods  are  used to distinguish groups of
entities  (e.g., sample sites)  according to  similarity of attributes  (e.g.,
species).  Similarity of group attributes may be expressed using a variety
of resemblance measures,  including commonly-used  similarity coefficients
such as Jaccard, Bray-Curtis,  Canberra metric, and  Euclidean distance.
Classification  begins with  the  compilation of  a matrix  of similarity
coefficients  (index scores)  between  all  possible pairs of entities.   One of
a variety of  available  clustering methods  is then used to form graphical
associations among  entities  to display groups of entities with similar
attributes.

     In most ecological  applications of classification methods,  sample
collection sites are designated  as  the entities, and the  relationship among
sites  is  defined  in terms of  similarity  of species  occurrence.   This
approach is  referred to as a normal  classification,  as  opposed to an inverse
classification in which species  are  selected as  entities  and their  presence
or abundance  at the sample sites serves as  the  attribute.  Analysis  of the
monitoring station  data set using both normal  and inverse classification
methods,  and  the subsequent examination of normal-inverse  coincidences using
a two-way table are recommended.

     A description  of  classification methodologies,  the use of numerical
classification,  and an  introduction to the literature concerning this
analytical  approach are  presented in the  EPA  report,  Application of
Numerical Classification in Ecological Investigations of Water Pollution
(Boesch 1977).  Reviews of important classification strategies are  given by
Clifford and  Stephenson  (1975),  Williams (1971),  Sneath and Sokol (1973),
and Goodall  (1973); and examples  of  ecological  applications can be  found in
Hughes and Thomas (1971), Boesch  (1973), and Grossman et al. (1974).

     Discriminant analysis summarizes multivariate  information by weighting
individual variables so  as  to maximize differences in groups of entities.
This  method  describes differences  between  relatively  homogeneous
species-assemblages (defined, for  example,  in  a  numerical classification
analysis) and facilitates  identification of environmental variables which
best separate these groups.  An introduction  to  discriminant analysis is
provided by  Cooley  and tonnes (1971).  Pertinent examples  of the use of this
method in the analysis of ecological  data include  Walker et al. (1979) and
Green and Vascotto  (1978).

                                   83

-------
     Ordination  refers to several multivariate  techniques which  are used to
 reduce the dimensionality of a data structure  and  to  relate biological
 characteristics  to environmental  factors.   The  dimensionality is reduced by
 one  of  several methods  which are  designed to minimize the  loss of
 information  resulting  from the  reduction.    In  its most basic  form,
 ordination  may be  used  to  group  similar sites  based on  biological
 characteristics  and to provide a graphical representation of between-group
 relationships.   Ordination in  this manner  is  analogous to the production of
 a dendrogram  using numerical  classification.  A  discussion of the use of
 reciprocal  averaging ordination  as  a classification technique is presented
 in Gulp and Davies (1980).

     Principal component analysis (PCA)  and factor analysis are techniques
 whereby  axes scores in  a  reduced  dimensional  space are examined for
 relationships  with abiotic  variables.   Variables displaying  high
 correlations  with component scores  are assumed to be responsible for group
 separation  based on biological characteristics.

     The relative merits of alternative  ordination methods are  compared by
 Gauch and Whittaker (1972), and  examples of environmental  applications are
 presented  in Smith  and Greene  (1976),  Sprules  (1977),  Gulp  and Davies
 (1980),  and Hughes and Thomas  (1971).

     Multivariate ANOVA  (MANOVA)  is  analogous to univariate  ANOVA, but
 includes measurement of more than one  biological  variable for each of
 several  samples and all  measured variables are tested simultaneously.  The
corresponding multivariate analysis  of two samples is Hotelling's T2 test.
The basic assumptions are essentially the  same as for the univariate case
 (i.e., normality  and independence  of  error terms and  homogeneity of
within-group  variance-covariance matrices).  Although multivariate tests for
variance  heterogeneity are available,  their application  is not  recommended
by Green  (1979) since they are  more sensitive to the variance assumption
than are  the MANOVA tests.   Transformations such as the  logarithmic may be
used to  correct  variance heterogeneity in the multivariate case so that
relatively minor violations of the  assumptions do not seriously affect the
test results  (Marriott 1974).
                                  84

-------
Biological Indices--

     Indices have commonly been employed in impact assessment of biological
communities because large amounts of multivariate data  (i.e., abundances of
individual  species) can be  reduced  to a  single number.   Indices  can be
useful  in this respect,  but definite problems and limitations are associated
with their use.

     A  primary problem  is  the failure of investigators  to recognize the
underlying  assumptions and  mathematical  relationships  of an  index.  By
overlooking such  considerations, an index may  be  selected,  applied, and
interpreted without a basic understanding  of the  properties  of the
biological  community which  are  actually  being measured.  Comprehensive
reviews of the assumptions and  uses  of diversity indices  are provided in
Green (1979), Pielou (1977),  Sanders (1968), and Peet(1974).

     An additional problem associated with indices is that they may be used
extensively at the exclusion of other  analytical  or comparative methods
which retain more of the available information.   Indices  may supplement
multivariate techniques or  analyses of individual  taxonomic groups (see
preceding sections), but field and laboratory studies have  demonstrated that
indices can be insensitive  to rather  intense  biological change (Godfrey
1978; Swartz et al., 1980; Smith  et al., 1979).  Moreover,  factors such as
sample  size,  collection method,  and time  of year may have a profound
influence on  the value of  an  index  (e.g.  Hughes  1978).  Therefore,
standardization  of  sampling procedures is a prerequisite to  conducting
comparisons among index values.

     Most indices  commonly used in  applied ecological  studies are
descriptions of community structure (e.g.,  species diversity, evenness, and
richness).  Other indices  (e.g.,  Infaunal  Trophic Index) incorporate
additional descriptive characteristics  for each species  and provide for  a
description of community function.  Several commonly-used  diversity indices
are  listed in Table 9.  Species richness  (S)  and Margalef's index (d)
emphasize the  number of species, are relatively simple,  may provide valuable
biological  information concerning impact assessment, and are  much less
ambiguous than  the information  theory  indices  of Brillouin  (H) and
Shannon-Wiener (H1).  H is  the diversity value for a sample, while H1 is an
                                  85

-------
           TABLE   9.  A  LIST  OF  COMMONLY-USED  INDICES  OF  DIVERSITY
     Index
                                  Symbol
                                                               Equation
Species Richness

Margalef

Shannon-Wiener

Brillouin

Simpson's Index
where:
S

d

H1

H

SI
      S = number of species.
      N = number of individuals.
     \i  = number of individuals  in  the
                                                          number of species
                                                               S-l/ln N
                                                          -In
                                                        1-1
                                                               N      N
                                                                   N!
                                                              N (N-l)
                                           species.
                                    86

-------
estimate for a random sample from a larger  population.  H1  is  used more
frequently due to  the  complexity  of  calculating large factorials in H.
However, for large numbers  of individuals (N),  the approximation, N (In
N-l), may be used for In N!, and computer programs are available  for exact
calculation of H  (Stauffer and Reish  1980).   Pielou  (1966)  has  suggested
that H is generally more appropriate than H'  since a truly random  sample is
required for estimation  of H1.  Simpson's  Index is a measure of  dominance
which is determined primarily by a few of the  most abundant species.

     Another index  available for impact assessment is the Infaunal  Trophic
Index (ITI) developed by Word  (1978).   The ITI is calculated as:
              ITI  =  100  -
33.33
where:

     n- is the number of  individuals in trophic  group i.

The ITI is based on the  relative proportions of individuals in four trophic
groups classified according to feeding  types:   suspended detritus feeders
(I), surface  detritus feeders  (II), surface  deposit feeders  (III),  and
sub-surface deposit feeders  (IV).   ITI  values correlate well with degree of
organic enrichment,  in  that decreasing ITI  values  indicate increasing
abundances of deposit-feeding organisms.  The ITI is  currently applicable to
benthic macroinvertebrate communities  at depths of 20  to 800 m (66 to 2,625
ft) in the Southern California Bight.  Research is  currently being conducted
to  determine applicability of the  index to the Puget  Sound area.

     Since different ecological  qualities are measured  by  the different
indices, it is recommended that  species abundance data be used to calculate
at  least three indices for each  study:   species number (S), combined number
of  species and  evenness  (H1 or H),  and dominance  (SI).   ITI would be an
important additional parameter for studies in the Southern California Bight.
                                   87

-------
  Indicator Species--

      Many species  display  characteristic distributional responses  to
  pollutant sources.  Analysis of the occurrence of such species, referred  to
  as "indicator species,"  may form a valuable component of  the analysis  of
  sewage  discharge impacts.   The primary  value  of  the indicator species
  concept is  that it allows  for  a  considerable reduction in analytical
  complexity,  since  individual abundances  of a few species are used  to
  evaluate response of the  community as  a whole.

      Indicator species  may be divided  into  two categories:

      •    Sensitive  organisms that display  severely reduced abundances
          near pollutant  sources

      •    Stress-tolerant or opportunistic  species that display
          greatly enhanced abundances  near pollutant  sources.

 In cases of organic  enrichment, the first  category  of indicator species is
 generally composed of suspension-feeding organisms such  as Ampelisca
 spp.  (Amphipoda)  and Amphiodia  spp.  (Ophiuroidea).   Reduced abundances of
 these species may also  indicate  high  sensitivity to toxic chemicals
 contained in  the  effluent.

     The second category of indicator  species, those  having high abundances
 in polluted areas, has received more intensive study  than the  former group.
 Such  species may have a high tolerance  to organic enrichment or toxic
 chemicals  in addition to an opportunistic life strategy (e.g., short
 generation time and/or lack of larval  dispersal). These attributes enable
 them  to  exploit  available  resources  in the  absence of nontolerant
 competitors or predators following habitat disruption or pollutant stress.
A list of some polychaetous  annelids  that have  been  associated with
pollutant sources  is  provided in Table  10.

     Although most of the species in  Table 10  have  been observed  in high
abundances in very polluted  areas,  indicator species may also be  used to
detect areas of moderate pollutant stress  or transitional  regions  between
polluted and  normal  areas.   Word  et al.  (1977) characterized  species

-------
  TABLE 10.   LIST OF SOME COMMON POLYCHAETES
    THAT HAVE BEEN ASSOCIATED WITH MARINE
            OR ESTUARINE POLLUTION
           Capitella capltata
           Polydora ligni
           Streblospio benedicti
           Scolelepis fuliqinosa
           Schistomeringos rudolphi
           Dorvillea articulata
           Heteromastus filiformis
           Mediomastus ambiseta
           M_. cal iforniensis
           Eteone longa
           Qphiodromus spp.
           Cirriformia tentaculata
           Neanthes succinea
           _N. caudata
Source:  Pearson and Rosenberg (1978)
                    89

-------
 indicative of polluted and  transitional areas  of  the Southern California
 Bight.  The clam  Parvilucina tenuisculpta.  the annelids  Tharyx spp.  and
 Mediomastus californiensis,  and  the ostracod Euphilomedes spp.  are present
 in  low abundances  in control  areas  and reach much  higher abundances (based
 on  absolute abundances  and proportion  of total  infauna) in areas of organic
 enrichment.

     The primary limitation of  the indicator  species concept  is that it
 should be used only  with a full  consideration of the normal distributional
 patterns and environmental associations  of  the species.  This is especially
 true in estuarine  environments where salinity fluctuations and high organic
 inputs may result  in natural  elevated  abundances of opportunistic species.
 In  the marine environment,  areas of  natural  organic accumulation (e.g.,
 submarine canyons, kelp beds) may also  have  high  abundances of opportunistic
 organisms.

     Before using an indicator  species as  part of a 301(h)  monitoring
 program,  several types of information should be developed:

     •    Natural  abundances of species in control areas

     •    Response of species to environmental  conditions other than
          pollutant  stress

     •    Observed response of species  to  pollutant sources in  the
          biogeographic zone.

     The primary  requirement is  that  species abundance  be  adequately
 described for  control conditions.   Proper  selection of control  sites will
 ensure that any observed differences in the abundances of  indicator species
 are due  to the  discharge  in  question  and  not  to  natural  or other
 anthropogenic  stresses.

Data Reporting

     The  information presented as part of a biological monitoring  program
 should consist of three general kinds:
                                  90

-------
     t    Methods

     •    Study  results and summary of  findings

     •    Data reports.

A discussion  of  study methods should be presented in each report, including
such aspects  as  station locations,  sampling procedures, sampling processing,
subsampling,  quality control,  and  analytical  methods.   Procedural details
should be provided unless a standardized technique is used,  in which case a
reference should be  included.  Citations  should also  be  provided for all
taxonomic references used for organism  identifications.

     The  presentation  of study results  should  include  a general
characterization of the biological  communities sampled.   Emphasis should be
placed upon descriptions  of both  spatial and  temporal  trends in community
structure and function.   Specific  comparisons should be conducted for all
biological  criteria contained in the 301(h)  regulations  (e.g., ZID boundary
vs. reference communities).  Where  statistical analyses are performed, the
report should include details of  the results.  For example,  in case of
ANOVA, the entire ANOVA  table should be presented, not just a statement
concerning the  significance  level of the F value.   Biological variables
(e.g., species  abundances, diversity,  richness) should  be presented in
graphical  or tabular format  as  means and their  95 percent confidence
intervals (X  +_ t(n_j)Sx) for each  sampling station.

     Each monitoring report should include  copies  of the  field collection
logs and laboratory sample counting forms.  The data  provided should include
the actual  numbers of each species  counted in each sample and the calculated
area!  or volumetric abundance  of  each taxon.   Sufficient  detail should be
provided to allow for verification of analyses  conducted  as part of the
monitoring program, or for reanalysis of the  submitted data.
                                   91

-------
                                 CHAPTER V

                              QUALITY CONTROL
      The U.S. EPA policy on  quality assurance and control  (EPA Administrator
 memoranda dated May 30, 1979, and  October 14,  1981)  stipulates that every
 monitoring and measurement project  must have a written and approved quality
 assurance  project plan.  This  requirement  applies to all environmental
 monitoring and measurement efforts  mandated or supported by EPA.  A quality
 assurance project plan will  specify the policies,  organization, objectives,
 and functional  activities  designed  to achieve  data  quality goals  of
 individual  projects or continuing  operations.  The monitoring programs  of
 301(h) permittees are covered by this policy.

      Several  EPA publications are available  from the Office of Monitoring
 Systems and Quality Assurance, ORD, USEPA, Washington, D.C., 20460,  on the
 subject of quality assurance.   EPA Publication QAMS-005/80,  for example,
 describes 16 elements which should  be included in all  quality assurance
 project plans.   That  publication establishes  criteria  for plan  preparation,
 including procedures  to  be used to document  and  report precision, accuracy,
 and completeness of environmental measurements.   In addition, the following
 paragraphs  provide guidance on quality control  procedures specific to  301(h)
 monitoring  programs.   The guidance provided below focuses primarily on water
 quality monitoring and toxics control  monitoring.  Additional  guidance  on
 quality  assurance and  control  procedures  is provided  in  Chapter IV,
 Biological  Monitoring, particularly  the  subsection on benthos.

 APPROACH AND  RATIONALE

     In 301(h) monitoring programs,  differences,  or the lack  of  differences,
 among samples must be  demonstrated.   When  the  differences  to be  defined are
 small  relative  to background  concentrations  (as is the  case for many
parameters), it  becomes imperative  to know and control the  uncertainty
associated with  sampling  and  analytical procedures.   This  is  necessary  to
                                   92

-------
 distinguish true field  variability from that  induced by sampling  or
 laboratory procedures.

     It is imperative to have qualified personnel  who  are  conscientious  and
 properly trained.  Such personnel,  using adequate equipment and procedures,
 will  help to ensure a clear  understanding of  sampling and analytical
 variability relative to  true  field variability.   A  well defined quality
 assurance/quality control  program should be designed as  an integral part of
 the  301(h) monitoring program.   The quality  assurance/quality control
 program should have as  its basis a simple, but  rigorous,  quantitative
 approach which  can be applied  consistently for the control of error.   The
 sampling and analytical procedures  selected should provide feedback so  that
 those performing  the  analyses can  promptly detect and correct procedural
 problems.  Thereby, the uncertainty regarding field measurement variability
 will be minimized.

     The  section which follows  describes  quality  control procedures
 associated with field activities.   A quantitative error analysis procedure
 is then presented  in  detail with  graphical  techniques  for  detection of
 increases in analytical error over  time.   Finally,  quality control
 procedures specific for toxic pollutant analyses  are briefly described.

 FIELD ACTIVITIES

     It is important that  field activities  be well planned in advance  and
 that as many decisions as  possible  be made before  field  sampling commences.
 Problems are encountered in coastal  work  which do  not  occur normally  during
 the sampling of inland waters.  Highly  conductive salt  mists frequently
 cause problems  with electronic  instrumentation.   There  is generally  a lack
 of nearby fixed points  (landmarks, permanent buoys)  from which to  locate
 sampling stations.   Waves and  swells  niake the sampling  vessel  unsteady,
 cause motion sickness,  and make work difficult.

     Field activity quality control  can be subdivided into  three categories:
 1) accurate station location,  2)  proper sampling  procedures,  and 3)  proper
documentation of sampling  efforts.   Station location can best be assured  by
use of redundant navigation systems.  For example,  the use of  portable
electronic  range  positioning systems plus the use  of a sextant in the
                                  93

-------
 horizontal mode would  form such a  redundant system.   The two separate
 procedures together help to minimize error.  The  installation onshore  of
 line of sight targets can  greatly aid  the  positioning  effort.  Sampling
 crews can line up the  vessel  from these  first and then go to the position
 fixing techniques for  improved resolution.  Finally, following sampling  at a
 station,  the bottom should  be sounded  using a manual  (non-electronic)
 system,  such as a lead  line.   The  known  depth (from  charts and prior
 measurements) and the lead line results can be  compared for a rough check  on
 position.

     There are two  general types of sampling:  1) in situ measurements, and
 2) the collection of water, sediment,  and biological samples for subsequent
 analysis.   In situ  measurements are  normally  made with electronic systems
 (in situ  observations by diver-biologists  are  discussed in Chapter IV,
 Biological Monitoring).  Calibration  of these  systems should be done before
 and after each series of field measurements.   Probe systems, except perhaps
 for the hydrogen-ion electrode,  are  often unreliable in terms of absolute
 accuracy and are  best used in a differential  sense (e.g., to measure changes
 in parameters with  depth while profiling  at a given station).

     Errors in sampling depth are caused  by  several factors including  ship
 motion and drag on  the underwater sampling equipment.   In high currents the
 drag on instrumentation and cables  may result  in significant  errors  if
 sampling depth is determined solely by the length  of cable underwater.   Drag
 can be mathematically corrected for only  if the current profile and the  drag
 coefficients for  the  instrumentation are known.  The  displacement of the
 sampling device(s)  from the desired position  is calculated by either summing
 the moments about the point where the  cable  enters the water or by making  a
 free body  analysis  of the forces on the sampling device and cable.

     Many multi-probe  in situ measurement  systems incorporate depth
measurement by use  of pressure transducers.   The accuracy and precision  of
 such systems must be periodically checked.  This is most easily done in  calm
 slack water.   Precision is determined by  multiple measurements at the  same
 depth.   Accuracy  is evaluated  by comparison  to  measurements made with  a
 heavily weighted  lead line.
                                   94

-------
     Dissolved  oxygen probe systems  should be calibrated using the modified
Winkler  titration technique.   Profile measurements of salinity  and
temperature  are used to determine water  column stability and are an aid in
the prediction  of  plume behavior.   Salinity probe systems offer moderate
accuracy  but should be cross-checked  by  discrete water samples analyzed by
induction type  laboratory salinometers.   Temperature probes may, at best, be
accurate to within  one-tenth  to three-tenths of  a  degree Celsius.
Temperature  probe  systems  are  rarely linear  over large temperature ranges
and must  be  checked against research grade laboratory thermometers.

     Water quality sample collection,  preservation,  and storage should be
performed in accordance with  the  procedures  discussed in Chapter  III.
Procedures for  taking biological  samples  are presented in Chapter IV of this
document.

     An important  aspect of receiving water  sampling is the order in which
procedures are  executed upon1 occupying a  station.  Vessel  positioning should
first be  completed.  (Vessel location  should be checked frequently.) Surface
observations should be made and recorded on  standard sampling sheets (see
for example  Figure 3).  Water column samples are then collected.  Following
that, the water column  is  profiled using  in situ measurement techniques.
Next, any benthic  samples  should be taken.   Only then should the depth be
sounded  using  a lead line  or  equivalent physical  technique.  Finally,
documentation should  be completed  before  proceeding  to  the next station.
The major purpose of the  above sequencing  of activities is to prevent
detritus, resuspended during  bottom  sampling  or depth sounding,  from
contaminating water column  samples or  in  situ measurements.

     Documentation should include surface  observations,  sample log sheets,
and data  sheets  with results from in  situ  sampling.   Waterproof preprinted
forms bound  as  log books have proven to be useful.

     It should  be  reemphasized  that  the sampling program design must be well
detailed  and that  as many decisions as possible  should be made before the
sampling  crew starts  their  efforts.   In coastal  and estuarine sampling,
unpredictable problems will  occur and will demand the immediate attention of
the sampling crew.  Time for this type of on-site decision making must be
available.
                                  95

-------
    FIELD  SAMPLING LOG     SURFACE  OBSERVATIONS
    OJte (e.g.  09 June  1976)
enter;  d.i^  (?  digits), enter month (3  letters), enter  year (4 digits)

Tide Stdorc pronounced.  Ctests hive
                                                                                                         • qldisy «(>|'(.'«r«nce 'OKl">»(wtlCiy I<0-^SO yiiTilst

    500-l.tKKJ (wtrrt («|>|>rujti'n4te1r iSO yjtdv-5/8 (t,n )
    1^ km (*(.|iru.t»4tfly S/8-I n m )
    Z-* km («t'l'Mi»(-i«-ly I-? B * |

    4-10 U (dj|>|>ri>«iiMtc-ly 7-t f> • )

    10-7U t« (*(.("-fi-i«.*tolx 6-1? n M )
    20-SD -- |JP|"<'. U,trl, 17OO n • }

    SO km ur mure (30 * "  ur iwirf)
                                                                                                                           rmsm wtAintu
                                                                                                                                    (no iloud at any lt«
                                                                                                                                   y [luudy (Kattrrrtt o
                                                                                                                            7  SPKIW. or rain and snow Hli

                                                                                                                            8  SlMwerfO
                                                                                                                            »  IhwnihTSlomfil
                                                                                                                                                l)
                                                                                                                                                 biok«n)
   •High Water  Sljck and Low Water Slack  as measured  at the southern point
    of Port Jefferson Harbor. at the Bayville Brid'jc1  (Oyster Hoy),  and at
    tlic Lloyd Harbor entrance (lluntington Bay Complex).
Figure  3.    Oceanographic  surface  observations  log   sheet.

-------
 QUANTITATIVE ERROR ANALYSIS

      This section presents techniques  for  error analysis.  The focus  is on
 those parameters for which discrete water samples are  taken in the field and
 analytical work  is  performed  in the laboratory.   The  error analysis
 procedures are also applicable to in situ measurement.   Analytical  precision
 is then determined by multiple measurements made over  a  short time period on
 a discrete water sample, a procedure analogous to the  splitting of  samples
 described in the following section.   Accuracy can be determined by  field
 (e.g., probe) and laboratory  analysis  of the  same sample,  and the  results
 analyzed in the  fashion  described below for  spiked samples (with the
 concentration added set to zero).

      Quantitative error analysis  is based  on the  premise that every
 measurement is subject to uncertainty.   Uncertainty  in water quality
 measurements arises from the  systematic bias  and  limited  precision  of
 sampling and  analytical  procedures and from  heterogeneity  in the water
 column at  sampling sites.   The  quantitative determination of  this
 uncertainty serves as the basis  of two essential steps in  the control and
 assurance of measurement quality:

     t   The routine monitoring of analytical precision and  accuracy

     •   The presentation of  results  in a  way that informs the
         reviewers of the  uncertainty of measurements and the
         confidence which may be placed in  conclusions drawn from the
         results.

     The first  step in  the quantitative  analysis of  errors  is the
verification  of a  discharger's  ability  to produce analytical  results which
are sufficiently accurate and  precise to meet  the specifications stated in
the table  of recommended analytical  methods in  Chapter  III.   This
determination is made prior to the initiation of routine monitoring for  each
water quality  parameter to be measured.  The  recommended  procedures for
determining precision and systematic bias are summarized in Table 11.
                                  97

-------
                                TABLE   11.   RECOMMENDED PROCEDURES  FOR DETERMINATION  OF  SYSTEMATIC
                                                  BIAS  AND PRECISION  IN  ANALYTICAL  METHODS
                                  Precision
                                                                                           Systematic Bias
            •    The precision  is determined at  four concentrations:  one
                 concentration  near the limit of detection of  the procedure,
                 two Intermediate concentrations, and one concentration near
                 the upper limit of application  of the method.

            •    Seven subsamples are analyzed at each of the  four concentra-
                 tions.

            •    For procedures using analytical Instruments:

                   1)  conduct  analyses over a two-hour period.
                   2)  run samples In the sequence high, low,
                       Intermediate, intermediate.  Repeat
                       sequence seven times.

            t    Report the mean, standard deviations (Sa) and number of
                 samples analyzed at each concentration.
                                                                         Add known amounts of the analyte to  the low and one of the Intermediate
                                                                         concentration  samples used to determine precision.   Enough additional
                                                                         material Is added to double the lower concentration and to raise the
                                                                         intermediate concentration to 75 percent of the analytical limit.

                                                                         Seven subsamples of each spiked sample are analyzed.

                                                                         Systematic bias Is reported as percent recovery at  the final concentra-
                                                                         tion of the spiked sample using the  mean of the seven analyses.
oo
It 1s strongly recommended that actual samples be used In these analyses 1n order to Include the effects of naturally occurring Interferences.

These procedures can be adapted to nearly all the analytical methods specified  In the monitoring program, except  those using gas chromatography -
mass spectometry.

These procedures for determination of precision and systematic bias are taken from U.S. EPA "Handbook for Analytical Quality Control  1n Water and
Wastewater Laboratories" (1972).

-------
      The  initial verification of  analytical competence  is  followed by the
 continuous monitoring of analytical  quality by  each discharger or their
 associated laboratory.  It is essential  that an  unsatisfactory analytical
 procedure be  quickly identified and corrected  in  order to prevent  the
 accumulation of inaccurate data.   This is accomplished primarily through the
 routine analysis of split  and  spiked  samples and the  use  of  quality control
 charts (as shown in Figure 4).

      Laboratory analytical  precision  should  be checked by splitting  a
 percentage of homogeneous field  samples  into replicates and analyzing all
 subsamples.   Sample  splits  should  be made  in  the field.  Laboratory
 personnel should be kept unaware  of which samples have been split.  Ideally,
 splits should not be  analyzed in succession.   For each  split sample,  the
 individual measurements and the  range are  reported for each parameter.   If
 possible, a double blind procedure should be used.

      The  accuracy,  which  includes precision and systematic bias,  of an
 analytical procedure is routinely  monitored by spiking a field sample with  a
 known amount of  the  analyte (a  standard addition).  As with the  split
 samples, personnel  performing  the  analyses ideally should remain unaware of
 which samples have been spiked.   The  deviation from stoichiometric behavior
 of spiked samples is calculated from the following  expression:

                            E = Xs - (xo  + A)                        (1)

 where:

      E = deviation  from stoichiometric behavior

     Xs = measured concentration of spiked samples

     XQ = measured concentration prior to  addition  of  spike

      A = increase in  concentration due to spike.

The deviation  from stoichiometric  behavior of each spiked sample  should be
 reported  along  with  the spiked and unspiked  sample  concentrations.
                                   99

-------
§
f-H
O
g_
o ""
^" o
^«» r^
I'd'
• •
UJ *O
t n * *"
^Z
X
° 0
O "1_
UJ o
>
"S_
•-. d
0
o
LU ro
^^ O
on
o
CM _
d
o
d ""
o
o
o
t
Analytical method operating
at specified precision
(standard deviation * .1)









•
* * •
^ •* 0 ™ 0 000
_ — ™ 00
Analytical method operating
below specified precision
(standard deviation • .2)



.


•
. . **
* • * *
Control limit
\ •
Warning limit » *
^B 0 A
' . w Mean ^
•• . ••**•**••!•. • " • * »
•*•••• • •• •
i i i i i i i i i i
) 10 20 30 40 50 60 70 80 90 100
                                SAMPLE NUMBER

                      SPECIFIED STANDARD DEVIATION - 0.1 mg/1
    NOTE:
    This  figure illustrates how a decrease  In measurement  precision would affect
    the distribution of data plotted on a precision control chart.  In actual
    practice, as soon as the loss of precision is detected—by about sample 62
    in this case—the cause of the decreased performance can be isolated and
    corrected.  (Precision control data were generated using a mean oxygen
    concentration of 5.0 mg/1.)
   Figure  4.   Example  Precision Control Chart.
                                     100

-------
      The frequency with which  split and spiked  samples are analyzed must
 represent a balance between reducing the likelihood  of generating inaccurate
 data and increasing laboratory costs  for analyzing  additional samples.   The
 U.S. EPA recommends that one  split and one spiked sample be analyzed for
 every 10 field samples  analyzed (U.S. EPA 1979b).  Those applicants who  wish
 to do so would have to justify  a  lower frequency  of  split samples.   This
 requirement can be applied to most parameters measured daily or weekly.   For
 parameters  which are measured only at longer intervals (e.g., every month or
 more),  at least one split and  one spiked  sample should be analyzed during
 each sampling  period.

      Project personnel other than the laboratory staff should be  careful  to
 distribute  the additional analyses  over time  and among sampling  locations.
 An even distribution of quality control  effort over time permits  continuous
 monitoring  of  laboratory performance  and ensures greater confidence in  the
 analytical  results.   Distributing quality control efforts among  sampling
 locations ensures  that interferences  present only at a few locations will be
 detected and enables  the  precision  in  measurement to  be  determined  for
 different locations.  It is  especially  important to be  able to distinguish
 analytical  precision associated with effluent and receiving water sample
 analyses.

     The  analysis of quality  control data is  facilitated  by the use of
 quality control charts.   Individual measurements of analytical measurement
 range and deviations from stoichiometric  behavior are plotted on a graph
 marked  with the  expected  limits  of  deviation  from the mean.   Standard
 practice is to draw "control limit"  lines at three standard  deviations  from
 the mean and "warning  limit" lines at two standard deviations from the mean.
 Since these correspond to 99.7  and 95.5 percent confidence limits,  the
 analytical procedure  is considered  out of control or  potentially out of
 control  when the limits  are exceeded more frequently  than one in 370 and  one
 in 20 samples,  respectively.  In some cases,  a developing analytical  problem
 appears  as a diverging  trend  on a control  chart even  before  the control
 limits are exceeded.  Examples  of control  charts illustrating cases when  the
 analytical  procedure  is under control  and when it is  not are  given  in
Figures  4 and 5.
                                   101

-------
g
o

o ~"
^3
. — > o
•^ d ~
3.
. O
§2-
HH
Q o
ct o
^~^ o
*— • • 1
0£
0 0
DC 0~
LU <
S 0
^^ PJ
al °~*
00 '
0
ro
O ~
O
O
i
O
IT)
O

Analytical method operating
without systematic bias
(Recovery • IOCS)
... *• •
* • • •
• • . v • •
• •
• **• *
• • *
• ^
• * m
• '
.
•
•
*








1 1 1 1

Analytical method operating
with systematic bias
(Recovery 902)
•


• • * t
. * .
• •• •'
• • • • •
•
•
. //. *
. •
0
• •
. •
	 	 , 	 A. — • 	
• • T>
•

• * *


1 1 1 1 1
1 0 10 20 30 40 50 60 70 80 90 100
                                 SAMPLE  NUMBER
                      SPECIFIED STANDARD DEVIATION * 0.1  w9/l
        NOTE:
        This figure illustrates how the presence of a  systematic bias would
        affect  the distribution of data plotted on an  accuracy control chart.
        In actual practice,as soon as the systematic bias was detected—
        before  sample 60 in this case—the cause of the  decreased performance
        would be  isolated and corrected. (Accuracy control data were generated
        using a mean cadmium concentration of 0.8 ijg/1 [analytical standard
        deviation - .1 pg/l] with 1.0 yg/1 added as spikes).
    Figure 5.    Example  Accuracy Control Chart.
                                     102

-------
      The construction of  quality control  charts is a simple process.   The
 expressions necessary to  calculate  the limit lines are summarized in Table
 12.   Note that in  many cases where the  analytical  variance shows a
 concentration dependence, the quality  control parameters can be adjusted to
 account for this dependence.   If most measurements  of  a parameter  remain
 within a small  range (e.g., +_ 15 percent  of  the  mean value), the assumption
 of constant variance is sufficiently  accurate  for control chart purposes
 even when the variance  is concentration dependent.

      For the purposes of verifying  compliance with the  precision  and
 accuracy specifications, quality control  data should be plotted on control
 charts derived  from  these same specifications.   If a laboratory's precision
 or accuracy is  substantially  better than  required,  additional  limit lines
 corresponding  to the  observed performance could be added  to assist in
 monitoring analytical  performance  more closely.   Methods  for generating
 control  charts  from  laboratory  data are described in U.S.  EPA  (1979b)  and
 American Society for Testing and Materials  (ASTM) (1951).

      Analysis of unknown standards for  all parameters  should  be performed at
 least once  a year in order to identify systematic error  not detectable by
 spiked  samples.  It is recommended  that the dischargers  or  their  selected
 laboratories participate in the  U.S.  EPA interlaboratory testing program.
 The  interlaboratory correlation  technique of Youden (1960) could  be used  for
 comparison  of results.

      As  a final  check, all submitted data should  be critically reviewed.   It
 may be requested that unusually high or low measurements be reanalyzed.

      The second function of quantitative error analysis is to facilitate  the
 presentation of data in  a way that  permits  open  inspection of  their
 certainty or uncertainty.  This requires  quantitative estimates of both  the
 variability introduced in measurements  by analytical imprecision and  field
 heterogeneity and the limits of sensitivity  of analytical methods.

     The variability  inherent in  the  analytical procedure is determined from
 the results of split  sample  analysis.   If the variance  is  relatively
constant over the range of measured concentrations, a pooled estimate of  the
analytical  variance can be made using the following expression:
                                   103

-------
                  TABLE  12.    SUMMARY  OF EXPRESSIONS  NECESSARY
                             TO  CONSTRUCT  CONTROL CHARTS

z
o
t/>
(_>
UJ
IX
0.
£1
>-
(_>

<_>

-------
                                1        NS
                     S a   "    2N        ^       i
                                s      i = l

 where:
      o
      S g = analytical  variance

      NS = number of split  samples pooled

      R^ = range of subsample measurements for sample  i split n ways.

 If  the variance is concentration dependent,  a different parameter, e.g.,  the
 coefficient of variation, can be averaged.

      The variability in measurement  caused by field  heterogeneity  is
 quantitatively determined by the analysis of  replicate field samples.   Two
 sampling strategies should  be considered:

      1.   Collect  replicate  field  samples  and analyze multiple
          subsamples of  each.   Analysis of  Variance (ANOVA) is then
          applied to determine the contribution of field heterogeneity
          to the variance.  The  drawback  of this approach is that it
          requires  a  large number  of analyses for  even minimum
          resolution power, e.g., 4 x 4  =  16.

     2.    Collect replicate field samples and  analyze each only  once.
          The field-induced variance is estimated using the principle
          of variance additivity.   This approach makes additional
          assumptions  concerning the analytical variance but requires
          many fewer analyses to be  made, e.g., 4 x 1  * 4.

Since the  application of the  first approach  is  well understood [see for
example  Sokal  and Rohlf (1969)],  only  the second will  be described in detail
here.
                                   105

-------
     The second  approach is  based on the assumption that  an  independent
estimate  of  the analytical  variance exists which  is  applicable  to  the
conditions under which  the  field replicates are analyzed.  If this is the
case, the field  variability can  be  estimated  from  the following expression
[American Chemical  Society  (1980)]:
where:

     S2f = variance  due to field heterogeneity

     S2  = variance  of  replicate field samples

     S2a = analytical  variance.
       a
Equation 3 can be applied using the pooled  estimate of analytical  variance
(Equation 2), if one the  following conditions is met:

      1.   The analytical  variance  is not strongly dependent on analyte
          concentration or  background interferences.

      2.   The concentration of analyte  in samples used to compute the
          pooled or initial  analytical  variance is similar to that in
          the field replicates.

      Replicate  sampling should be  conducted  at  all  field  stations  where
measurements are to be used  in comparisons.   Analysis of replicate sample
data is  necessary for assessing the reliability of such comparisons.   When
replicate sampling at all stations is not feasible, replication is required
for at least one station from each group  of  stations which may reasonably be
assumed to have similar amounts of  field heterogeneity.  As  a  minimum,
replicates  should be collected at  one ZID  boundary  station and one reference
 (control) station.

      The number of  replicates  to be collected at each  station depends on the
 use intended for the data.   For  example, more replicates  are required to
 make a meaningful comparison  of variances than are required  to compare mean
                                    106

-------
 values.   Appropriate  statistical  methods should  be  applied to  each case.
 Sevenfold  replication is currently recommended by U.S.  EPA  (American
 Chemical  Society 1980).   Depending upon circumstances,  overall project
 design, and the applicant's resources, fewer  replicate determinations may be
 accepted.

      If  possible, replicate  samples  should  be taken at  all  specified
 stations within the first year of the sampling program during a  period of
 maximi/m natural  variability.  This would  provide information  on  field
 heterogeneity to be utilized during  the  sampling program in-progress review
 (review of  first  annual  report).    At that  time,  the replication program
 should be  evaluated  and re-designed  if necessary.  In  addition,  the
 knowledge of field variability can  be applied to design a composite sampling
 scheme for  all stations.   By analyzing  one  sample formed  by combining and
 homogenizing a number of  replicate  samples,  the  uncertainty in  the
 measurement due to field heterogeneity can  be reduced without  performing any
 additional  analyses.  The following  expression defines  the number  of
 replicates  required  to  obtain a confidence  limit of E for  a  mean value:

                            Nr =  (t  • Sf/E)2                        (4)

 where:

      t = appropriate value of Student's  distribution

     Nr = number of replicates  required

     Sf =  field variability:   Sf » Sa.

When the  condition of negligible analytical  variance is not met,  composite
sampling  is  of little value.

     In addition to the analytical  and  field variances,  knowledge of the
limits of quantisation and detection is  essential  to the assessment of
measurements  at  trace levels.  The method recommended by  the  American
Chemical Society (1980)  is  based on the analytical variance determined using
a field blank as follows:
                                  107

-------
                               L  =   3  • S
                               LQ =  10  - S
where:

     LD = limit  of detection

     LQ = limit  of quantisation
     SB = ^St + sb      wnere st and sb  are the standard deviations
          of an instrument response to replicate instrument runs  of a
          single analyte - containing sample and a blank sample,
          respectively.

For a more detailed analysis  of limits  of  detection and quantisation,  see
Currie (1968).

TOXIC POLLUTANT ANALYSIS

     Assuring the  quality of  toxic pollutant  analyses requires  numerous
precautions beyond those necessary for other water quality parameters.   Most
of the additional quality control procedures are necessitated by the  greater
complexity of analytical instruments used for toxic pollutant analyses and
the risks of sample contamination.   Requiring all toxics  analyses  to be
conducted by U.S. EPA  certified laboratories helps assure the adequacy of
internal quality control  practices.   Each  laboratory  should be practicing
the  quality assurance  steps outlined  in  the source  documents for  each
procedure as well as appropriate methods in the quantitative error analysis
portion of this chapter.  Special quality control procedures for toxics not
dealt with in the  above sources are the concern of the remainder of this
section.

     Extra care in sample handling is  required  for  toxic pollutant analysis
since  these pollutants generally occur  in trace  concentrations  and
frequently are unstable.   Samples  should  be stored  in  the dark to avoid
photochemical decomposition.   Storage  at reduced  temperatures, as specified
in Chapter III, minimizes the  rate  of  other degradative chemical reactions.
                                   108

-------
 Exposure of the sample to the  atmosphere should be minimized in order to
 avoid loss of volatile compounds.

     Sample  bottles  must be clean and made  of materials which will  not
 contaminate the samples.   Plastic  or glass  bottles must be specified
 depending upon the  analyses to  be performed  on the sample.  Caps for all
 glass bottles used  to  store  toxics samples should  be  teflon lined.   Glass
 bottle and cap liners  should be cleaned with chromate cleaning solution and
 successively rinsed with distilled  water  and  several  portions of  the
 appropriate spectral  grade redistilled organic solvent.  Bottle caps  should
 be washed with detergent  and rinsed  using the  same  steps described  above.
 Plastic sample bottles should be cleaned with  detergent or concentrated
 hydrochloric acid and  rinsed with distilled water.

     Since  many  organic  substances  are  strongly  sorbed by particulate
 matter,  it is essential  that  effluent  samples  contain  a fraction  of
 suspended solids  representative of the entire waste  stream.  This should be
 considered in the selection of  sampling  devices and  locations.   The
 variability in measurements introduced  by sampling techniques, together with
 the variability caused by  effluent heterogeneity,  are  considered below.

     The assessment of the  significance  of  toxic pollutant measurements
 depends on a knowledge of the  variability in measurements  introduced by
 sampling technique  and site heterogenity.  This variability is determined by
 the analysis of replicate  samples.  The rationale used to specify the  number
 and frequency of  replicate samples is contained in  the Quantitative  Error
 Analysis portion  of this  chapter.  Replicate  samples  should be collected
 from the effluent stream and in  most  cases from the sediments  of at  least
 one field station.  Replication should be carried out at least once  a year
 during the period of highest  toxic pollutant levels.

     The qualitative and quantitative analytical  capabilities of a
 laboratory should be verified as  well.   One effluent sample spiked with  each
 of the routinely monitored  substances (except  dioxin)  should  be analyzed
with each group of  samples.   A  blank  (glass doubly-distilled water) should
 be analyzed  along  with  each  group   of  samples  screened  for priority
 pollutants.
                                  109

-------
                                APPENDIX A

                          OCEANOGRAPHIC METHODS
     This  appendix is intended  to provide background information  and
 guidance on collection  of oceanographic  data.   Types  of  current meters  and
 their proper use are  summarized.   The use of  drogues,  drifters, and  dye
 studies are also reviewed.   In addition,  specifications for field use of
 current meters,  drogues,  drifters, and  dye  studies are discussed.
 Positioning methods  are briefly reviewed.

 CURRENT METERS

 Uses of Current Meters

     Current meters  are used to measure the variation in current speed  and
 direction at a fixed location with time.  Since  each meter collects data
 only at a single point, several  current meters may be required to establish
 the  velocity field over depth and over a given area.   A  typical fixed
 current meter array may contain  a meter near the bottom,  one near the
 surface, and one at mid-depth.   Short-term current  measurements can be taken
 from a boat with a meter lowered over  the  side and connected  by cable to an
 on-deck recorder.   Longer  term in-situ measurements  may be obtained by
 installing  meters with self-contained recording devices (e.g.,  magnetic
 tape, film,  and  strip charts) on a  mooring system anchored  to the bottom.
 Current speed and direction may  be recorded either  continuously, or at
 specified time  intervals.   Vector  averaging  meters electronically average
 the measured velocity components  over specified sampling  intervals, and
 provide the  average current  over each interval as output.

     Different current meters are designed to work  in different operating
 environments.  Meters designed to operate  at greater depths must be able to
withstand higher pressures, and must  be  capable of accurately measuring
 relatively  low  current speeds.  Meters  designed to operate in  shallow
                                  110

-------
 nearshore coastal environments must be able  to accurately measure currents
 in  the presence  of wave  action.   The  effects of  waves include  both
 oscillatory water paYticle orbits  in  the  immediate  vicinity of the meter,
 and the pumping action of the mooring  lines where a surface or subsurface
 buoy  is subject to wave-induced motions.

 Types of Current Meters

     The current meters  presently in use  may be classified as mechanical,
 electromagnetic, or acoustic.   Mechanical  current meters include Savonious
 rotors, ducted impellers, drag  inclinometers,  and  propeller-type meters.
 Savonious rotor current  meters utilize  a  unidirectional  rotor on a vertical
 axis of rotation and a vane which senses  the  horizontal direction of flow.
 Although these meters have been  used for  many  years  in  oceanographic work,
 they are not suited for  operation  in  shallow water  environments which are
 exposed to wave and swell  activity.  In such waters the  oscillatory velocity
 components due to the orbital  motions  of  waves  are significant.   Since the
 rotor turns in only one  direction,  the oscillatory velocity components due
 to waves are recorded as a rectified velocity  input.  The rectified record
 cannot be adequately resolved into the oscillatory components since the
 directional  response of  the meter is slow  relative to the wave motions, and
 its response to deceleration is different  than  its response to acceleration
 (Horrer 1968).  As  a result,  the velocity  record is  distorted and never
 reaches zero velocity,  even  in the presence of wave  action  alone.   This
 makes it difficult to distinguish  the  steady component  of the current from
 the oscillatory component due to waves.   Savonious  rotors should,  therefore,
 only be used in deep waters below  the  influence of wave action or in areas
where the current speeds are high and the waves  very small.

     Two types of ducted impeller current meters  are available which attempt
 to eliminate the problems associated with  current measurements in  the
presence of  waves.   One  type manufactured  by Endeco  is a neutrally buoyant
meter with  a bidirectional impeller which is attached  to the  mooring line by
a tether several  feet in length.   The tether  and neutral buoyancy allow the
meter to move with wave-induced water  particle  orbits, so these  short-term
velocity oscillations  are not  recorded.  The meter is said to orient itself
in a wave  field so  that only the mean current  is measured.   A tilt
compensation mechanism  keeps the meter  horizontal  (Brainard and Lukens
                                   111

-------
 1975).  Another horizontal ducted impeller  meter manufactured by Bendix has
 a  long boom vane to keep  the meter  oriented in the direction of mean  flow,
 preventing  it from rotating  in  the presence of  short-term wave-induced
 oscillations (Horrer 1968).  The  ducted impeller  is bidirectional,  and the
 oscillatory components  of the  horizontal  current  velocity are recorded but
 can be separated from the record.

     Davis-Weilar propeller type meters are also designed to operate in the
 presence of waves.   These meters utilize two orthogonal  propeller assemblies
 designed  to respond linearly to the  current vector components  so  that
 averaging or filtering  of oscillatory  movements can be done properly  (Wald
 1979).  Electronic  circuits resolve  the signals from the propeller  sensors
 into N-S and E-W velocity components, and  electronically integrate them over
 a  specified sampling interval.   This gives  the vector averaged-current over
 the interval.   Oscillatory movements associated with wave action are  removed
 by electronic  filters in  the averaging  procedure.

     Drag inclinometer  type meters  consist of a cylinder with stabilizing
 fins which is  suspended from  a pivot point at one end.   The drag and lift
 forces due to  the current velocity  deflect the meter  at a vertical angle
 which is measured by an inclinometer.   The  horizontal  direction is measured
 by a compass.   Although this type  of meter is claimed to measure currents in
 the wave  zone accurately, it may  suffer some  limitations  under these
 conditions  since its response (tilt  angle versus current speed)  is not
 linear.   This  nonlinearity makes  it difficult to  separate the oscillatory
 velocity components  from the  mean  current (Daubin et al., 1977).  Also,
 vertical  wave-inducted  motions  will  deflect the meter  vertically, and
movement of the  mooring  line or significant turbulence in the flow may
 result in  measurement errors.

     Electromagnetic current meters  measure the instantaneous  x- and
y-velocity components at  a flow  sensor which  contains a wire  coil  and two
orthogonal  pairs  of  electrodes. The  coil  produces  a magnetic field,  and the
electrodes measure the  voltage  gradient across the coil  which is  induced by
the water  as it flows through the field.   The orthogonal  electrode pairs
measure  the x- and y-velocity components.   Since the voltage gradient is
proportional  to  the current speed  and there are no moving parts  in the
sensor,  the meter is capable  of a fast, linear,  highly sensitive component
                                   112

-------
 response (McCullough 1977).  These characteristics make  the meter suitable
 for use in  the presence of waves,  since the oscillatory components can be
 separated  either from the  records  or filtered  out by electronically
 integrating the electrical  signals  over a specified sampling interval.

      Acoustic current meters  determine current  velocities by measuring the
 relative travel  times  of  two simultaneous  acoustic signals  transmitted
 across  the  flow sensor.  The transmitted acoustic beams are focused on a
 reflecting  plate which returns  the  signals to the receiving transducers.   An
 acoustic  phase shift detection scheme correlates the travel  times with the
 current velocity component along the beam path.   Two  pairs of orthogonal
 transducers measure both  the x-  and y-velocity  components.  Voltages
 proportional to the velocity components  are resolved electronically using
 the output  signals from the  transducers  and compass.  Since acoustic flow
 sensors  have a fast, linear,  highly sensitive response (McCullough 1977),
 they are suitable  for use in the presence of waves.   The  wave-induced
 components  may be separated either  from the records  or filtered  out  by
 electronically integrating  the  records over  a  short sampling interval.

 DROGUES AND DRIFTERS

 Use  of Drogues

     Drogues are  used to trace the path of moving water near  the surface or
 at  fixed depths below  the  surface.  The drogues are released  at  a  given
 station and are subsequently  tracked  by  recording their positions at  short
 time intervals.   The  sequence  of positions  and travel times  between
 positions gives  information on the actual path a particle  may  travel in the
 currents, and the  mean velocities between points along  the path.  However,
 the actual paths  traced  by  two drogues simultaneously  released  at a  given
 point will rarely  be  the same due to the  random components of the velocity
 field.  Thus,  several drogues  should be released together.

     Drogue  studies  can  be  used to determine  the current patterns in the
vicinity of  an outfall and to  evaluate the  movement of  a waste plume which
either surfaces or remains submerged  at a given  depth.   If a sufficient
number of drogues are released  at appropriate depths and are monitored at
frequent intervals,  information can be obtained on the net transport and
                                  113

-------
horizontal  dispersion of a waste  field, and on the  variations in the  current
velocities along the path followed by  the plume.   If  the  drogues are
followed long  enough,  they may indicate where the plume will reach the
shoreline,  the  length of time  involved,  and the path followed between the
discharge point and the shoreline.

Types of Drogues

     Most  drogues  can be  classified  into  one  of  the following four
categories:  1)  parachute  drogues, 2) cruciform  drogues,  3)  window shade
drogues, and 4) cylindrical  drogues.   Of  these types,  parachute and
cruciform drogues are the most  widely  used.   Parachute drogues  consist of a
passenger parachute or smaller  pilot parachute which is usually attached to
a weighted  vertical pipe.  The  parachute is supposed to remain open and
oriented horizontally with  its  opening facing  the currents.   However, since
some parachutes  are  denser than water,  they tend  to  hang  downward in a
closed position  at low current  speeds.   This  problem can be reduced by using
a spreader bar or  ring to help keep the parachute open,  and by  adding
buoyance to the  parachute so that it becomes  neutrally buoyant.  Cruciform
drogues usually consist of  two  or sometimes three identical  slotted sheets
of plywood,  masonite,  plastic, or  metal  arranged in a  bi-planar (or
tri-planar) crossed vane.  The vanes  can  also be  constructed  of canvas or
cloth stretched  across  a  solid frame.  Window shade  drogues  consist of a
rectangular sheet  of plastic film,  canvas, or  cloth suspended  from a
spreader bar and bridle at the  top,  with a weighted spreader bar at the
bottom.  The rectangular drogue should hang  approximately  vertical with its
plane surface oriented perpendicular to  the  direction of  the horizontal
flow.  However, this  orientation,  which  maximizes  the drag forces of the
currents,  may  not  always  be  achieved  under  actual field  conditions.
Cylindrical  drogues include  various vertically oriented cylindrical objects
which have  been used to  trace currents (e.g.,  55 gallon drums,  drift poles,
wooden barrels).

     Most drogues areyreally  drogue-buoy systems consisting of  a small
marker buoy which is tracked at  the surface and a larger submerged drogue
portion with  is set at the desired depth  by a connecting line between the
two.   The drogue portion must be  weighted and ballasted so  that the drogue
assembly has sufficent negative  buoyancy to  keep both the drogue and
                                  114

-------
 connecting line in their  intended vertical  orientation, and to keep  the buoy
 mast upright.  The  connecting line must  remain close to vertical so the
 drogue will be measuring  currents at the desired  depth.

     Drogues are intended  to passively  drift with the  currents  at a
 specified  depth.  In reality,  some  errors are introduced  in the drogue
 trajectories by wind drag on the exposed portion of the marker buoy, by the
 relative surface current  drag on the submerged portion of the surface buoy,
 and, for deep drogues with long  lines,  by  the relative current drag on the
 connecting line.  Since surface  currents are  generally faster than the
 deeper currents,  it is important  to  design drogues  so  that the projected
 area of the  submerged  drogue is maximized  and  the projected  area of the
 surface  buoy is minimized.   Large drogues are much more  difficult and
 cumbersome to launch  and  retrieve from a boat  than  smaller  designs,
 especially when many drogues are involved.  However, since the accuracy of
 the measurements  generally increases with  the size of the submerged drogue,
 it is best to use the largest size practicable  in  a given situation.   The
 surface buoy should  be  the minimum size required to keep the  system buoyant
 and trackable at  the surface.  The  wind  forces on the exposed  portion can be
 minimized by using a buoy which is  almost completely submerged and which has
 only a thin radio antenna or small  radar  transponder protruding upward for
 tracking.   Buoys  with considerable  freeboard  or with larger tracking devices
 such as  flags, radar  reflectors, or  flashing  lights may  be  subject  to
 significant external wind  forces.    If  possible, it  is  best  to avoid
 performing drogue studies under high wind conditions, especially when trying
 to measure the lower current speeds in deeper waters.

 Uses of Surface Drifters

     Surface drifters are used  to  measure the average  path  of currents  at
 the surface.  Drift bottles  and drift cards are  the types most commonly
 used.   Vertical drift cards and drift bottles are useful  for  evaluating the
movement of effluents which  reach the surface.  Horizontal drift cards,
which  float horizontally on  the  surface,  may  be  used  to  determine the
potential  movement of surface slicks due to  oils or other floatables which
form a  surface film.  Because these movements are influenced  largely by the
wind,  horizontal  drift cards  should  be released under several different
meteorologic conditions.
                                   115

-------
      Surface drifters provide a rough estimate of the travel  times between
 the release and  recovery allowing  a net drift rate  to  be  computed.  No
 information  is given on  the  actual  flow path  or  the velocity variations
 along the path.   However, they do  provide  information as to whether or not a
 surface waste  field can  be  expected to reach  the shoreline, and, if so,
 where it will  make contact and approximately how long it may take.

 Types of Surface  Drifters

      Drift bottles are long-necked glass  bottles partially filled with sand
 ballast so that  only 0.25  to 1 in of the  bottle  neck remains  above the
 surface.   The  bottle size is  typically 4 to 6 ozs (Grace 1978).  Each bottle
 contains  a readily visible postcard with  instructions  requesting the return
 of the card with information  on the  location  and  time  of recovery  and
 generally  the  offering of a  reward.

      Drift cards  are  available in several  different  forms.   Most common
 types  are  either  drift envelopes,  which are plastic envelopes containing
 instructions on a return  card, or plastic  drift  cards, which are rectangular
 in shape with identification numbers and  instructions stamped into them.   As
 with  drift bottles,  the  information  requested  is  the  location and  time of
 recovery and a reward is  generally  offered for  their return.  Horizontal
 drift  cards  are designed to  float  horizontally and,  therefore, measure
 transport  in the surface  film or upper millimeter  of the water column.   As a
 result, they are strongly  influenced  by  the wind and do not really measure
 the surface currents.  Vertical drift cards are designed so that only  one
 edge remains at the  surface, with the card retaining a vertical  orientation
 in the water column.  This  is accomplished either by using  a negatively
 buoyant card with  a  foam flotation strip  on  the upper edge or a positively
 buoyant card with  a  weight strip on  the  lower edge.   Both  vertical drift
 cards and drift bottles measure the average horizontal transport in  the
 upper  few feet of the water  column, and thus  provide a measure  of the
 surface currents.  A good design will attempt to  minimize  the effects of
wind on the drifter by minimizing the  ratio  of the sail area  (area exposed
 to the wind)  to drogue area  (submerged area exposed to current).  Drift
bottles may be  better  in  this respect since the exposed  bottle necks  are
much narrower than the submerged part  of the  bottles.
                                  116

-------
     Since  the  recovery rate of  surface drifters  is  generally low, many
drifters  must be released at each station  in order to obtain  an adequate
amount of information.  The drifters  are usually deployed by boat, but may
also be released by airplane.  Because  of their compact size, it is easier
to release a large  quantity of  drift  cards than drift  bottles.   A good
drifter must be durable enough  to survive  at  sea,  reach  the shore through
the surf, and should attract attention once  it reaches the  shore.

Use of Seabed Drifters

     Seabed drifters  measure the average  path of currents  near the sea
floor.  They are useful for determining  the fate of waste materials subject
to transport by bottom  currents.   This  includes settleable solids and any
portion of the  effluent which remains  near the bottom.  The drifters provide
information on  the net movement of a waste field  along the bottom, including
where the waste field may reach the shoreline,  and a rough estimate  of how
long it may take.  If  a sufficient number  of  drifters are recovered, they
may indicate areas of possible shoreline contamination.

     The success of  a bottom drifter  study depends on reasonable  recovery of
the drifters.   The drifters generally  have labels attached which  request the
location and time of recovery and promise  a reward.  The condition  of the
drifter should also be recorded since,  if the rod becomes detached, the
saucer will float and therefore measure  the  surface currents rather than the
bottom currents.    The drifters may  be  recovered  either offshore by
commercial  fishermen using bottom trawls  or bottom gill nets,  or more  often
at the shoreline.  The  recovery rate  will depend on recreational access to
the beaches, the  intensity with which the beaches are  used,  and  on how
extensive the  commercial  fisheries are in the  area.   Recovery rates from
less than 5 percent  to  over 50 percent  have been reported.  Therefore, many
more drifters  should be released  than are expected to be recovered in  order
to  obtain  a  sufficient  amount of data.   The information obtained from
drifters recovered offshore by  commercial  fishermen  is probably more
accurate than  data from shore-recovered drifters, since in the  latter case
no information  is provided on the time  of  first contact with  the shore or
possible movement in the  surf zone.   In  either case,  the only  information
provided is  the point of release,  the point of recovery, and a rough
estimate of the travel  time between the  two  points.

                                   117

-------
Types of Seabed Drifters

     Woodhead  type drifters are  generally used.  These  devices resemble
umbrellas in shape, consisting of small  plastic dished saucers with a long
thin plastic rod attached at  the  center.  Typical  dimensions are 18 cm (7
in) for the  diameter of the  saucer and a 0.65  cm  (0.25  in) diameter rod
about 54 cm  (21 in)  long.   The rod terminates in a sharpened point and a
small, weighted collar [about 6 g  (0.2 oz)] is attached near  the end (Grace
1978).  The saucer  has  a  slight positive buoyancy so it tends  to hover
slightly above the sea floor and drift with the bottom  currents; the weight
collar causes  the  pointed end of the rod  to lightly drag along the bottom.

     Seabed  drifters may be deployed either at the surface by a boat or
low-flying small plane, or they may be  taken to the bottom and released by
divers.   Several  drifters may be  released  at  the same  location  on the
bottom, even if deployed at the surface,  by  attaching  them to a salt spool
which dissolves after the drifters reach the bottom.   However, because the
rate of descent of the drifters is fairly  slow, the actual release point on
the sea floor will differ from the known  release point at the  surface.  This
difference  increases with increasing  depth  and  subsurface current
velocities.

DYE STUDIES

     Dye studies,  using  fluorescent  tracers,  can  be  very useful  in
determining  the behavior of  waste  plumes,  as well  as indicating general
circulation  patterns in the vicinity of discharge sites. A suitable tracer
can be injected either directly into the waste stream before  it discharges,
or into the  receiving water  at some  point  near the  discharge site.  The
injection may  be a single dose or  a  continuous release.   The movement of the
waste plume  and the horizontal and vertical  dispersion of the effluent can
be determined  by measuring the concentration distribution of  the dye tracer
both temporally and  spatially after  the initial injection.  The dilution
rates and the  spatial and temporal distribution of contaminants at a given
distance from  the  discharge site  can  then  be evaluated. Three-dimensional
distributions  can  be obtained by sampling at various depths.
                                   118

-------
SPECIFICATIONS FOR FIELD WORK

     After  selection of the  appropriate  method(s) for obtaining the
necessary oceanographic information,  specifications  are  needed for the
general types of  equipment to be used, number of measuring devices, location
of measurement stations  (drogue/drifter release points or dye  injection
points), and frequency and times of measurements.  If a dye study is to be
done, type of dye, appropriate concentration, measurement technique, and
frequency should  be specified.   This section  discusses specifications for
current meters, drogues and drifters, and dye studies.

Current Meters

     The type of  fixed current meter selected  depends on the importance of
wave motion at the site and available equipment.  The wave motion expected
at the site  should  be evaluated  to determine if data  obtained could be
affected by wave-induced orbital  motion.   If so,  a  non-Savonius type of
curent meter should be used.   Calibration and visual inspection  should be
made of the current meters before  and after  use to detect other sources of
error such as fouling  of  the  rotor or the sensor probes.   In-situ current
meters set for at least  5 days at a time  are  preferred  to current meters
deployed for short periods  from a boat.  The current meter array should
include meters set near  the  surface, at mid-depth,  and 1.5 m (5  ft) above
the bottom.   The  exact depths are  determined after reviewing the  available
current data to locate depths where  different  currents exist, the expected
height of rise of the plume,  and the current measurement objectives (e.g.,
whether movement of  the waste field or  sediment movement is  of primary
interest).

     Current meter location  decisions also depend on the current measurement
objectives.   One  meter array  should be  located near the discharge site.  As
a minimum, currents should be measured continuously or  at least for 15 to 30
min during all 5-day  spring  and neap tide periods,  on a quarterly basis,
when receiving water samples  are collected.
                                   119

-------
Drogues and Drifters

     The resolution of information obtained  during a drogue or drifter study
depends on the  number of  drogues or drifters followed, the frequency with
which positions are fixed (recorded), and the  accuracy of the method used to
fix positions.   The use  of drogues is  recommended over  drifters because
drogues give information on the flow path  and current speeds along the path
and also because of the low recoverability  of  drifters and poor estimates of
travel times obtained  from drifters.   However, these  factors have to be
considered along with the cost of tracking  the drogues.

     If certain types of drogues are more  suitable  for the wind and current
conditions at the site,  then those  types should be  recommended  in the
monitoring program.  The drogues should be  released  over  the diffuser at the
approximate level  to which the plume  rises during the  time of year of the
study.  At least  five  to six drogues  should be released each time.  The
drogues' positions should  be  traced  at  half-hour intervals up to a distance
of 3.7 km  (2 nmi) from the outfall  or for a total of  10 hr.  Successive
vector plots  showing  drogue position  over time  should be provided on
large-scale nautical  charts.

     Drifters could  be  used if  the primary concern was  whether a  surfacing
waste  plume might reach  shore  or if only  an approximate travel  time were
needed.  The choice  of  drifter type would depend on  deployment method, cost,
and  distance from shore.   Vertical  drift cards  are  the easiest to  use and
are  more durable than glass bottles.  The drifters should be  released at the
point  of discharge or the middle  of  the diffuser.

Dye  Studies

      Specifications  for a fluorescent dye study  to  be done as  part of a
given  monitoring program  include the  time of year for the dye  study, the
type of dye to  be used, the approximate concentration to be  achieved in the
receiving  water, the length  of  time of dye  injection and measurement, and
suggestions on  the measurement technique to  be used.   Each of these  aspects
will  be discussed briefly.
                                    120

-------
     The purpose of a dye study  is to  determine the  horizontal and vertical
 extent of the plume and the  direction  of the plume's movement.  A dye study
 should be performed at a time  of minimum stratification, which occurs most
 often in the fall or winter, and  at  a  time  of maximum stratification, which
 occurs most often in the summer.  Density  profiles collected during water
 quality monitoring surveys should be checked  to identify or verify periods
 of maximum and minimum  stratification.  The  selection of the type of dye
 depends on the degree of adsorption which could occur at the site, the cost,
 and availability of chemicals.   The dye tracers most commonly used at the
 present time  are rhodamine B  and rhodamine  WT.   Rhodamine B  has  fairly
 strong adsorptive tendencies,  and may be adsorbed  onto suspended solids,
 sediments, plankton, aquatic plants, and sampling and injection equipment.
 Rhodamine WT is the preferred choice because it is much less susceptible to
 adsorption than rhodamine B, although  it is  about twice as expensive.  Both
 dyes  are  available in liquid form,  which  is  strongly  recommended  over
 powdered forms due to ease of handling.  The  solutions should be adjusted to
 the same density as sewage effluent,  if necessary, by mixing with methanol
 or by addition of salt.

     The effects of temperature,  degradation,  and  photodecomposition must
 also  be considered.  The  fluorescence  of  a sample will  vary with
 temperature,  although  this  can  be  adjusted easily with  a  temperature
 correction curve.   Chemical degradation  can  be a problem in  the presence  of
 strong oxidizing agents  such as  chlorine.   The  degradation  rate is usually
 low at the chlorine levels typically  encountered in the field,  but it  should
 be evaluated for dye  injected directly into  a chlorinated waste stream.
 Photodecomposition  rate can  be estimated from a control solution  of dye
 which is exposed to sunlight on  the  boat and monitored for  the duration  of
 the study.

     Concentrated  dye  solutions must  be diluted  to appropriate  levels  before
 being released.  Because  the  fluorometer  calibration curve reverses
 direction  for  dye  concentrations greater than  about 1.0 ppm  (Turner Designs
 1976), dye  studies should  be  designed so  that the dye concentrations
measured in  the  field  are below this  level.  Dilution techniques  can be  used
 for higher  levels where  discrete water  samples  are collected.   However,  this
 is not done  easily  when  an on-ship flow-through apparatus  is  used or when
 in-situ  measurements are taken  with  a towed submersible  fluorometer.  The
                                  121

-------
 dye should  be added to the effluent  just before  it enters the outfall  for a
 period of  about  6  hr,  begining 3  hr  before the current  reverses.   The
 concentration in the effluent  stream should be 0.2 ppm above the background
 fluorescence of the seawater.   The  background  fluorescence of moderately
 polluted water  may  be as high as 100  to 200 parts per trillion, and raw
 sewage  has  a background  fluorescence of about one part per billion (Turner
 Designs 1976).   The pigments present  in blue-green algae can  yield
 substantial background fluorescence if large concentrations are present.
 Although appropriate optical filters can be used to minimize this problem,
 background  fluorescence  should always  be  determined from  seawater samples
 before  releasing the dye.

     The dye can be  measured using a  submersible fluorometer  or a continuous
 pump and shipboard fluorometer.  Submersible fluorometers,  if used, should
 be  towed at a depth  approximately  equal  to the equilibrium  level of the
 plume.  Measurements  should be made  for  about 12  hr beginning 3 hr before
 the start of the dye  injection.  When dye  studies are conducted,  turbidity,
 temperature,  and salinity profiles  should  be taken at 3-hr intervals
 throughout the  tidal  cycle at  stations  near the outfall.   These data will
 help in locating  the  plume and in  interpreting dye study results.   More
 detailed  discussions of  dye  studies  and  fluorometers are  included  in
 Feuerstein and  Selleck (1963),   Smart and Laidlaw (1977), Turner Associates
 (1971),  and Wilson (1968).

 Navigation-Position Determination

     Position determination is important for  siting and returning  to
 sampling stations and for  tracking  drogues.  Although several different
methods  are  available, the most commonly used methods for nearshore coastal
 surveys  are Loran C,  electronic range-positioning systems  with  onshore
 transponders, horizontal  sextant readings  from aboard  ship, and theodolite
readings from shoreline stations.

     Loran C is  the government  sponsored radio navigation system selected
for use  in the coastal zone by  the  Department of  Transportation.  The Loran
C system has recently been implemented on the West Coast  and in Alaska.  The
previously existing chains on the East Coast, Gulf Coast, and the Hawaiian
Islands  have been expanded and  modernized  to provide a complete network of
                                  122

-------
Loran C  staions for the U.S. coastal zone.   Receivers automatically resolve
Loran C  signals and display position with claimed accuracies  as high as +_ 15
m (+_ 50  ft).

     Portable electronic range-positioning systems are  available  from
several  manufacturers and have claimed accuracies as high as  +_ 1 to 3 m (+_ 3
to  10 ft).   These  systems are the preferred  method  for  position
determinaton.   Electronic  range-positioning  systems  are limited to
line-of-sight measurements.  The maximum range varies with the manufacturer.
However, most types have ranges  sufficient  for nearshore applications.  It
should be  noted  that  at  very short  ranges  (<50 yd)  electronic
range-positioning system errors may be disproportionately large.

     Visual  methods of position determination such as sextant or  theodolite
readings are limited to use  under  conditions of adequate visibility.  In the
sextant method,  the position  of the observer on-board a boat is  fixed by
measuring two  horizontal sextant  angles between three charted objects, with
one  object being common to  both angular measurements.   In the  theodolite
method, the position of an object in the water (for example a drogue, or a
boat at a sampling station)  is determined by simultaneously  taking two
theodolite readings  from separate  shoreline stations.   By  knowing the
locations of the theodolite stations and  the horizontal  angles to the
observed object,  the position  is fixed.   The accuracy of  positions
determined with  a sextant or theodolite varies  with the precision of the
instruments,  the experience of the operators,  the horizontal  distances
involved, the magnitudes of the  angles, the strength of the three-point fix
(for sextant readings), the scale  of the charts, and  the accuracy  with  which
the  positions are plotted.
                                   123

-------
                                 REFERENCES


 Ahlstrom, E.  1969.   Recommended  procedures for measuring the productivity
 of plankton standing  stock and related  oceanic  properties.  Nat.  Acad.  Sci.
 59 pp.

 Alexander, G.Y.,  and  D.R. Young.   1976.   Trace metals in southern  California
 mussels.  Marine  Pollution Bulletin 7:7-9.

 Allen, G.H., A.C. DeLacy, and D.W.  Gotshall.  1959.  Quantitative  sampling
 ?Lmar,1ne flshen'es  - a Problem  in  fish behavior  and  fishing gear.   pp.
 448-511.  In:   Waste  disposal in  the marine environment.   Pergamon.

 All6!Ic U;J*   1975'    Alternat1ve  methods  for assessing fish  populations.
 pp.  95-98.   In:  Coastal Water Research Project Annual  Report.   SCCWRP, El
 Segundo, CA.

 Allen, M.J., J.B. Isaacs, and R.M. Voglin.   1975.   Hook-and-line survey of
 demersal fishes in Santa Monica Bay.   TM 223.  SCCWRP.   El  Segundo, CA.  23
 pp.

 Allen, M.J., H. Pecorelli, and J.Q.  Word.   1976.   Marine  organisms around
 ?QfoiaiJoiPl£«,s 1n Santa Monica  Bay-   J- Wat.  Poll.  Control Fed.
 48(8):1881-1893.

 American Chemical  Society.  1980.   Guidelines for data acquisition and data
 quality  evaluation in environmental chemistry. Anal. Chem.  52:2242-2249.

 American Public  Health Association.   1980.  Standard methods  for  the
 examination of water  and wastewater.   15th edition.  Washington, D.C.   1134
 pp. (update of 13th and 14th  editions).

 American Society for  Testing  and Materials.  1951.  ASTM Manual on quality
 control  of materials, special  technical  publication,  15-C.  ASTM,
 Philadelphia, PA.  922 pp.

 Anderson,  C.H.,  and  J.M. Long.   1980.   Interim methods for  determining
 asbestos in water.  Rep.  No. 600/4-80-005.  U.S. EPA Environmental Research
 Laboratory.

 Batzli, G.O.   1969.  Distribution  of biomass in rocky intertidal  communities
 on the Pacific  coast of  the United  States.  J. Anim.  Ecol. 38:531-546.

 Becker, C.D., J.A. Strand, and D.G.  Watson.   1975.  Environmental impact
monitoring of  nuclear power plants,   pp.  353-918.  In:   Source  book of
monitoring methods.  AIF/NESP-004.   Volume  II, Part 6, Aquatic  Ecology.
                                   124

-------
 Beers,  J.R.  1978.   Pump sampling.   pp. 41-49.   In:   Monographs  on
 oceanographic methodology.  6.   Phytoplankton  manual.   United Nations
 Educational, Scientific,  and Cultural Organization, Paris.

 Beers, J.R., G.I. Stewart,  and J.D.H. Strickland.  1967.  A pumping system
 for  sampling small  plankton.   J. Fish. Res. Board Can.  24:1811-1818.

 Bilyard, G.R.  1974.  The  feeding habits  and  ecology of Dentalium entale
 stimpsoni Henderson (Mollusca:   Scaphopoda).  Veliger 17(2):126-138.

 Boesch, D.F.  1973.  Classification  and  community structure of macrobenthos
 in the Hampton Roads  area,  Virginia.  Mar. Biol.  21:226-244.

 Boesch, D.F.  1977.  Application of numerical  classification in ecological
 investigations  of water pollution.   Rep. No. 600/3-77-033.   U.S.  EPA,
 Corvallis, OR.  115 pp.

 Brainard, E.C.,  and R.J.  Lukens.  1975.   A  comparison  of the accuracies  of
 various continuous recording current meters for  offshore  use.  Offshore
 Technology Conference, Paper No. OTC 2295.  8 p.

 Brett, C.E.  1964.   A portable  hydraulic diver-operated dredge-sieve for
 sampling subtidal macrofauna.  J. Mar. Res.  22(2):205-209.

 Buchanan, J.B.  1971.  Sediments.   In:   International Biological Program
 (IBP) Handbook No.  16.  Blackwell Scientific Publication, Oxford, pp.  30-52.

 Burrell, V.G., W.A. Van  Engel,  and S.G. Hummel.   1974.   A new device for
 subsampling plankton  species.  J. Cons.  Perm.  Int.  Explor. Mer.  35:364-366.

 California State  Water  Resources Control  Board.  1972.   Water quality
 control  plan  for ocean waters  of  California:   Guidelines for technical
 reports and monitoring programs.  Sacramento,  California.  65 pp.

 California State  Water  Resources Control  Board.  1978.   Water quality
 control  plan for  ocean waters  of California.   15  pp.

 Carlisle, J.G., Jr.   1968.  Red  tide  in  California.  Mar. Res.  Leaf.  No.  2,
 State of California,  Dept. Fish  and Game, Sacramento, CA.  6 pp.

 Clark,  M.E.   1969.  Dissolved free ami no acids  in  seawater and their
 contribution to the nutrition  of sea  urchins,   pp.  70-93.   In:   W.J. North
 (ed), Kelp Habitat  Improvement Project.   Annual  Report, 1968-1969.  Calif.
 Inst. Tech.,  Pasadena, CA.  (not seen).

Clark, R.C.,  Jr., and J.S. Finley.   1973.   Paraffin  hydrocarbon  patterns  in
petroleum-polluted  mussels.  Marine Pollution  Bulletin 4:172-176.

Clendenning,  K.A.   1964.  Photosynthesis  and  growth  in Macrocystis pyrifera.
Proc. Int.  Seaweed  Symp. 4:55-65.
                                   125

-------
Clifford,  H.T.,  and W. Stephenson.   1975.  An  introduction to numerical
classification.  Academic Press,  San Francisco,  CA.   229 pp.

Cochran, W.G.  1963.  Sampling  techniques.  2nd  edition.  John Wiley & Sons,
Inc.,  New  York, NY.  413 pp.

Cochran, W.G.  1977.  Sampling  techniques.  3rd  edition.  John Wiley 4 Sons,
Inc.   New  York, NY.  428 pp.

Cohen, J.   1977.  Statistical  power  analysis  for  the  behavioral sciences.
Academic Press, New York, NY.

Connell, J.H.  1961.  Effects  of competition,  predation by Thais lapillus,
and other  factors on natural populations of the  barnacle Balanus balanoides.
Ecol.  Monogr.  31(1):61-104.

Cooley, W.W.,  and P.R.  Lohnes.  1971.  Multivariate data analysis.  John
Wiley & Sons,  Inc., New York, NY.   364 pp.

Grossman,  J.S., R.I. Kaisler, and J.  Cairns, Jr.   1974.  The use of cluster
analysis  in the assessment of spills of  hazardous materials.   American
Midland Naturalist 92:94-114.

Culp,  J.M.,  and  R.W.  Davies.   1980.   Reciprocal  averaging and  polar
ordination as  techniques for analyzing lotic macroinvertebrate communities.
Canadian Journal Fisheries Aquatic  Science 37:1358-1364.

Currie, L.A.  1968.   Limits  for qualitative  detection and quantitative
determination.  Anal. Chem.  40:586.

Daubin, S.D., D.R. Scally,  and  R.F.  lusting.   1977.   Measurement of
deep-ocean currents  using  recording inclinometers.  Oceans '77, Marine  Tech.
Soc.  and  Inst. of Electrical and  Electronics Engineers.  6 pp.

Dawson, E.Y.   1959.  A primary report on the benthic  marine flora of
southern  California,   pp. 169-264.   In:  Oceanographic survey of  the
continental  shelf area of southern  California.   Calif.  St.  Water Poll.
Control Bd.  Publ. No. 20.

Dawson, E.Y.   1965.  Intertidal algae,   pp 220-231.  In:  An oceanographic
and biological survey of  the southern California  mainland shelf.  Calif.
St. Water  Poll. Control Bd.  Publ.  No. 27.

de Lappe,  B.W., R.W. Risebrough, J.T. Mendola, G.W. Bowes, and J.L. Monod.
1972.   Distribution of polychlorinated biphenyls on the Mediterranean Coast
of France,   pp.  43-45.   In:   Journees Etud.  Pollutions,  Athens, Greece:
CIESM.

Devinny,  J.S., and L.A.  Volse.   1978.   Effect of sediments  on  the
development of Macrocystis pyrifera gametophytes.   Mar.  Biol. 48:343-348.
                                   126

-------
 Downing, J.A.  1979.   Aggregation, transformation, and the design of  benthos
 sampling programs.  J.  Fish. Res. Bd. Can.  36:1454-1463.

 Dunn, O.J.   1964.   Multiple  comparisons using  rank  sums.   Technometric
 6(3):241-252.

 Eganhouse, R.P., and  D.R. Young.   1976.   Mercury in mussels,   pp. 105-109.
 In:  Coastal Water  Research Project Annual  Report.  SCCWRP.  El  Segundo, CA.

 Eppley, R.W., J.N.  Rogers,  and J.J. McCarthy.  1969.  Half saturation
 constants for uptake  of  nitrate and ammonium by marine phytoplankton.
 Limnol. Oceanogr. 14:912-920.

 Fager, E.W., A.O.  Flechsig,  R.F. Ford, R.I.  Clutter, and R.O. Ghelardi.
 1966.   Equipment  for  use  in  ecological  studies using  Scuba.   Limnol.
 Oceanogr. 11:503-509.

 Fenchel, T.  1969.  The ecology of marine  microbenthos.  IV.   Structure and
 function of the benthic ecosystem, its chemical  and physical  factors  and the
 microfauna communities with  special reference to the ciliated protozoa.
 Ophelia 6:1-182.

 Feuerstein, D.L., and R.E.  Selleck.    1963.  Tracers for dispersion
 measurements in surface waters.  SERL Report No. 63-1.  Sanitary Engineering
 Research Laboratory,  University of California, Berkeley.  69  pp.

 Fiedler,  P.C.  1978.  The  precision of simulated transect surveys of
 northern anchovy, Engraulis mordax, school  groups.  Fish.  Bull.  76:679-685.

 Gale, W.F., and J.D.  Thompson.   1975.  A suction sampler for quantitatively
 sampling  benthos  on rocky  substrates.   Trans.  American  Fish. Soc.
 1975(2):398-405.

 Gauch,  H.G.,  and  R.H. Whittaker.  1972.    Comparison  of  ordination
 techniques.  Ecology  53:868-875.

 Godfrey, P.J.   1978.  Diversity  as a measure of benthic  macroinvertebrate
 community response  to water pollution.  Hydrobiologia 57:111-122.

 Goldberg,  E.D.,  V.T.  Bowen,  J.W.  Farrington,  G. Harvey, J.H. Martin, P.L.
 Parker,  R.W.  Risebrough, W.  Robertson, E.  Schneider, and  E. Gamble.   1978.
The mussel  watch.   Environmental  Conservation 5:101-125.

Gonor,  J.J.,  and P.F.  Kemp.   1978.   Procedures  for quantitative ecological
assessments in  intertidal  environments.  Rep.  No. 600/3-78-078.  U.S. EPA,
Corvallis,  OR.   104 pp.

Goodall, W.D.   1973.   Numerical  methods of classification,   pp. 575-618.
In:  Ordination and Classification of Vegetation.   Part V., Handbook of
Vegetation  Science,  R.H. Whittaker,  (ed).  W.  Junk, The  Hague.
                                  127

-------
 Gordon,  M., G.A. Knauer,  and J.H. Martin.  1980.   Mytil us call form'anus as  a
 bioindicator  of  trace metal  pollution:   Variability  and statistical
 considerations.  Marine Pollution Bulletin 11:195-198.

 Grace,  R.A.    1978.  Marine  outfall  systems  planning, design, and
 construction.  Prentice-Hall, Inc., Englewood Cliffs, NJ.  600 pp.

 Great Lakes Regional Committee on Analytical  Methods.   1969.   Chemistry
 laboratory manual  bottom  sediments.  U.S. EPA, Washington, D.C.  101  pp.

 Green,  R.H.   1979.   Sampling  design  and  statistical  methods for
 environmental biologists.  John Wiley & Sons,  NY.  257 pp.

 Green, R.H.  1980.   Multivariate approaches in ecology:  The assessment of
 ecologic similarity.  Ann. Rev. Ecol.  Syst.  11:1-14.

 Green, R.H., and  G.L.  Vascotto.   1978.   A method for  the analysis of
 environmental  factors controlling patterns of  species composition  in  aquatic
 communities.  Water Res.  12:583-590.

 Hamley, J.M.  1975.  Review  of  gill net selectivity.   J.  Fish.  Res.  Board
 Can.  32:1943-1969.

 Harden Jones,  F.R.,  A.R. Margetts, M. Greer Walker, and G.P.  Arnold.   1977.
 The efficiency of  the  granton  otter trawl  determined by sector scanning
 sonar and acoustic  transponding tags.  Rapp. P-V. Reun.  Cons.  Int. Explor.
 Mer.  170:45-51.

 Hasle,  G.R.   1978.  The  inverted  microscope method,   pp. 88-96.
 In:  Monographs on  oceanographic methodology.   6.  Phytoplankton manual.
 United Nations  Educational, Scientific, and  Cultural  Organization,  Paris.

 Hedgpeth, J.W.   1957.   Obtaining ecological  data in  the sea.  Geol.  Soc.
 America Memoir  67(1):53-86.

 Hollander,  M.,  and D.A. Wolfe.   1973.  Nonparametric  statistical methods.
 John Wiley  & Sons, NY.  503 pp.

 Holme, N.A., and A.D. Mclntyre.   1971.  Methods  for the  study  of the marine
 benthos.   Blackwell  Scientific Publications, Oxford, Great Britain.  334 pp.

 Horrer, P.I.  1968.   Comments on  the measurement  of currents in the presence
 of waves.  Marinco,  Inc.  4 pp.

 Hughes, B.D.  1978.   The influence  of factors other than pollution on the
 value  of Shannon's  diversity  index  for  benthic macroinvertebrates in
 streams.  Water Res.  12:359-364.

 Hughes, R.N., and M.L.H. Thomas.   1971.   The  classification and ordination
 of shallow-water benthic samples from Prince  Edward Island, Canada.  J.  Exp.
Mar.  Biol.  Ecol. 7:1-39.
                                  128

-------
  Hulings, N.C., and J.S. Gray.   1971.   A manual for the study  of meiofauna.
  Smithsonian Contributions to Zoology No.  78, pp. vii+84.

  Jacobs, F., and G.C.  Grant.  1978.   Guidelines for zooplankton sampling in
  quantitative baseline and monitoring  programs.  Rep. No.  600/3-7-78-026
  U.S. EPA, Corvallis,  OR.  51 pp.

  Lawrence, J.M.  1975.  On the  relationship between marine plants and sea
  urchins.  Oceanogr. Mar. Biol.  Ann.  Rev.  13:213-286.

  Lemberg, N.A.   1978.  Hydroacoustic assessment of Puget Sound herring,  1972
  - 1978.  State of  Washington Department of Fisheries  Tech. Rept. 41.  43 pp.

  Lie, U.  1968.  A quantitative study  of benthic infauna  in  Puget Sound,
  Washington,  U.S.A.,  in 1963-1964.   Fiskeridirektorat,  Skrifter Serial
  Havunders 14:229-556.

  Littler, M.M.   1971.   Standing stock  measurements of crustose coralline
  algae (Rhodophyta) and other saxicolous organisms.  J. Exp. Mar. Biol. Ecol.
  O • j X ™ jj •

  Littler, M.M. ,  and S.N. Murray.   1975.    Impact of  sewage  on  the
  distribution, abundance and  community structure of  rocky intertidal
 macro-organisms.   Mar. Biol. 30:277-291.

 Lorenzen, C.J.  1966.   A method for  the  continuous measurement of in  vivo
 chlorophyll concentration.   Deep-Sea Res.  13:223-227.

 Marriott, F.H.C.   1974.   The  interpretation of multiple  observations.
 Academic Press, London.

 McCullough,  J.R.  1977.  Problems in  measuring currents  near the ocean
 surface.  Oceans   77,  Marine  Tech.  Soc.  and Inst. of Electrical  and
 Electronics Engineers. 7 pp.

 McEwen,  G.F., M.W.  Johnson, and T.R. Folsom.  1954.   A statistical  analysis
 of  the  performance  of  the  Folson plankton sample splitter  based upon test
 observations.   Arch. Meteorol .  Geophys.  Biolkimatol . , Ser. A. Meteorol.
 Geophys. 7:502-527.

 McGowan, J.A., and D.M.  Brown.   1966.   A new  opening-closing paired
 zooplankton net.  Univ. Calif.  Scripps Inst. Oceanogr.  (Ref. 66-23).

 McGowan, J.A.,  and  V.J. Fraundorf.  1966.  The relationship between size of
              est1mates of  zooplankton  diversity.   Limnol. Oceanogr.
Mclntyre,  A.D., and D.J. Murison.   1973.  The meifauna  of a flatfish nursery
ground.  J.  Mar. Bio. Assoc.  United Kingdom 53:93-118.

Mearns  A.J. , and M.J.  Allen.   1978.  The  use  of small  otter trawls  in
coastal  biological surveys.  Rep.  No.  600/3-78-083.  U.S. EPA, Corvallis,
UK •   j^r    •
                                  129

-------
Mearns, A.J., and H.H.  Stubbs.   1974.  Comparison of otter trawls used in
southern California  coastal surveys.  TM  213.  SCCWRP.  El  Segundo, CA.  15
pp.

Mearns, A.J., D.A. Hanan, and L. Harris.   1977.  Recovery of kelp  forest off
Palos Verdes.   pp.  99-108.  In:   Coastal  Water Research Project Annual
Report.  SCCWRP.   El Segundo, CA.

Medz, R.B.  19 April 1982.   Personal Communication (phone by  Larry Marx).
U.S. EPA Monitoring  Technology Division, Washington, D.C.

Moore, S.F.,  and  D.B.  Mclaughlin.   1978.  Design of  field experiments to
determine the ecological  effects of petroleum  in intertidal ecosystems.
Water Research 12:1091-1099.

North, W.J.  (ed.).   1964.   An  investigation of the  effects of  discharged
wastes on kelp.   California State Water Quality Control Board Publ. No. 26.
124 pp.

Paine, R.T.   1966.   Food web complexity and  species diversity.  Amer. Nat.
100:65-75.

Paine, R.T.   1969.  A  note on  trophic complexity and species  diversity.
Amer. Nat. 103:91-93.

Pearson, T.H.,  and R. Rosenberg.  1978.  Macrobenthic succession in relation
to organic enrichment and pollution of the marine environment.  Oceanogr.  Mar.
Biol. Ann. Rev. 16:229-311.

Peet, R.K.  1974.  The measurement  of species diversity.   Ann. Rev. Ecol.
Syst. 5:285-307.

Pielou, E.G.   1966.   The  measurement of diversity  in different types of
biological collections.  J. Theoret.  Biol. 13:131-144.

Pielou,  E.C.   1970.   An  introduction  to  mathematical  ecology.
Wiley-Interscience, New York.   286 pp.

Pielou, E.C.   1977.  Mathematical  ecology.   Wiley-Interscience, New York.
385 pp.

Plumb, R.H.,  Jr.  1981.  Procedures for handling  and chemical analysis of
sediment  and water samples.   Rept.  No.  EPA/CE-81-1.    Environmental
Laboratory,  U.S.  Army Engineer Waterways Experiment Station, Vicksburg,
Miss.

Preston, F.W.  1948.   The commonness, and  rarity, of species.   Ecology
29{3):254-283.

Richkus, W.A.   1980.  Problems  in  monitoring marine  and estuarine fishes.
In:  C.H.  Hocutt and J.R. Stauffer Jr. (eds),  Biological  Monitoring of Fish.
Lexington Books, Lexington, MA.
                                  130

-------
 Ricker, W.E.  1971.  Methods for the assessment of fish production in  fresh
 waters.  IBP Handbook No.  3.  Blackwell Oxford.  313 pp.

 Rosenthal, R.J., W.D. Clarke, and P.K. Dayton.  1974.  Ecology and natural
 history of a stand of giant kelp Macrocystis  pyrifera,  off Del  Mar,
 California.  Fish. Bull.  72:670-684.

 Saila,  S.B., R.A. Pikanowski, and D.S. Vaughan.  1976.  Optimum allocation
 strategies for  sampling benthos  in the New  York Bight.   Estuarine and
 Coastal Marine Science 4:119-128.

 Sanders, H.L.  1968.   Marine  benthic diversity:   A comparative study.  Amer.
 Nat. 102(925):243-282.

 Saville, A.  1977.  Survey methods  of apraising fishery resources.   Food and
 agriculture organization  of the  United Nations.  Rome.   76 pp.

 Scheffe, H.  1959.  The analysis  of variance.  John Wiley & Sons,  New  York,
 NY.

 Seber, G.A.F.   1973.  The  estimation  of animal abundance.   Griffin.   London.
 506 pp.

 Siegel, S.   1956.  Nonparametric statistics for the  behavioral  sciences.
 McGraw-Hill Book Co., New  York, NY.   312 pp.

 Smart, R.L., and I.M.S. Laidlaw.   1977.  An  evaluation of some fluorescent
 dyes for water tracing.  Water Resources Research 13(1):15-33.

 Smith, R.W., and C.S. Greene.  1976.   Biological communities near  submarine
 outfall.  J. Water Poll. Control Fed. 48(8):1894-1912.

 Smith,  W., V.R.  Gibson,  U.S.  Brown-Leger,  and J.F. Grassle.   1979.
 Diversity as an  indicator  of pollution:  cautionary results from  microcosm
 experiments,  pp. 269-277.   In:  Grassle et al. (eds).   Ecological  Diversity
 in Theory and  Practice, Internat. Coop. Publ.  House,  Fairland, MD.

 Sneath, P., and  R. Sokal.   1973.  Numerical  taxonomy.   W.H.  Freeman and Co.,
 San Francisco, CA.  573 pp.

 Sokal, R.R., and F.J. Rohlf.  1969.   Biometry.  W.H.  Freeman  and Co., San
 Francisco,  CA.   776 pp.

 Sprules, W.G.  1977.   Crustacean zooplankton communities  as indicators of
 limnological conditions:   an approach  using principal component  analysis.
J. Fish Res. Board Can. 34:962-975.

 Stauffer, J.R.,   and  R.L.  Reish.   1980.   FORTRAN Program for calculating
Brillouin's species diversity  index.  Prog.  Fish Cult.  42(3):185-187.
                                   131

-------
Stofan,  P.P., and G.C. Grant.   1978.  Phytoplankton sampling in quantitative
baseline  and monitoring programs.   Rep.  No. 600/3-78-025.   U.S.  EPA,
Con/all is,  OR.  83 pp.

Strickland, J.D.H.,  and T.H. Parsons.   1972.  A practical  handbook of
seawater analysis.  Bulletin 167.  2nd Edition.   Fisheries Research  Board of
Canada.   Ottawa, Canada.   310  pp.

Sukhanova,  I.N.  1978.  Settling without the inverted microscope,  p. 97.
In:  Monographs on Oceanographic Methodology.   6.   Phytoplankton Manual.
United Nations Educational,  Scientific, and Cultural Organization, Paris.

Swartz,  R.C.   1978.   Techniques for sampling and  analyzing  the marine
macrobenthos.  Rep. No. 600/3-78-030.   U.S. EPA,  Corvallis, OR.  27  pp.

Swartz,  R.C., W.A. DeBen,  F.A. Cole,  and  L.C.  Bentsen.  1980.  Recovery of
the macrobenthos  at a dredge  site in Yaquina Bay, Oregon,   pp.  391-408.
In:  R.  Baker (ed), Contaminants  and  Sediments,  Vol. 2, Ann Arbor  Science,
Ann Arbor,  MI.

Tegner,  M.   1980.  Multispecies  considerations  of  resource management in
southern  California  kelp beds.  pp.  125-143.  In:   Proceedings of the
Workshop on the Relationship between Sea Urchin Grazing and Commercial
Plant/Animal Harvesting.   J.D. Pringle,  G.J. Sharp, and J.F. Caddy (eds).
Canadian Tech. Rep. Fish,  and  Aquatic  Sci. No.  954.

Tetra Tech.  1982.   Revised  section 301(h)  technical  support document.
Contract No. 68-01-5906.  U.S. EPA,  Office  of Water Program  Operations,
Washington, D.C.

Thomas,  W.H., and D.L.R.  Seibert.  1977.   Effects  of copper on the dominance
and the  diversity of  algae:   controlled ecosystem pollution  experiment.
Bull. Mar.  Sci. 27:23-33.

Throndsen, J.   1978a.   Preservation  and  storage.   pp.  69-71.   In:
Monographs on Oceanographic  Methodology.   6.   Phytoplankton Manual.  United
Nations  Educational, Scientific, and Cultural  Organization, Paris.

Throndsen,  J.   1978b.  Centrifugation.   pp. 98-103.   In:   Monographs on
Oceanographic Methodology.   6.  Phytoplankton Manual.   United  Nations
Education!, Scientific, and  Cultural Organization, Paris.

Turner Designs, Inc.  1976.  Monograph entitled:   Flow measurements.

Turner,  G.K. and Associates.   1971.   Monograph  entitled:  Fluorometry in
studies  of pollution and  movement of fluids.   12  pp.

UNESCO.   1968.  Monographs on  Oceanographic  methodology.  2.  Zooplankton
sampling.  United  Nations  Educational,  Scientific,  and Cultural
Organization, Paris.  174 pp.
                                  132

-------
UNESCO.  1973.   Monographs on oceanographic methodology.   3.  A guide to the
measurement of marine  primary production under  some special conditions.
United Nations  Educational,  Scientific, and  Cultural Organization, Paris.
73 pp.

UNESCO.  1978.   Monographs on oceanographic methodology.   6.  Phytoplankton
manual.  United Nations Educational , Scientific, and Cultural Organization,
Paris.  337 pp.

U.S.  Environmental Protection Agency.  1972.   Handbook for analytical
quality control on  water and wastewater laboratories. U.S. EPA Analytical
Quality  Control Laboratory.   National Environmental  Research Center.
Cincinnati, OH.  95 pp.

U.S.  Environmental Protection Agency 1978.   Methods for benzidine,
chlorinated organic compounds,  pentachlorophenol ,  and pesticides in water
and wastewater.  U.S. EPA, Environmental Monitoring and Support Laboratory.
Cincinnati, OH.  155 pp.

U.S. Environmental  Protection Agency.  1979a.  Guidelines  establishing test
procedures for  the  analysis of pollutants; proposed regulations.  U.S.  EPA,
Washington, D.C.  Federal  Register, Vol. 44, No.  233, Part III.   pp.
69464-69575.

U.S. Environmental  Protection Agency.  1979b.   Methods for  chemical  analysis
of water and wastes.   Rept. No. 600/4-79-020.   U.S. EPA, Environmental
Support Laboratory.  Cincinnati, Ohio.  430 pp.

U.S. Environmental  Protection Agency.  1981a.  Removal of  bis(chloromethyl )
ether from the  toxic pollutant list  under Section 307(a)(l) of the Clean
Water Act.   U.S.  EPA, Washington,  D.C.  Federal Register, Vol. 46,  No.  23,
Part III, pp. 10723-10724.

U.S.   Environmental  Protection  Agency.    1981b.   Removal  of
dichlorodifluoromethane and trichlorofluoromethane from the toxic pollutant
list under Section  307(a)(l)  of the  Clean Water Act.   U.S. EPA, Washington,
D.C. Federal Register,  Vol. 46,  No.  5, p. 2266.

U.S. Environmental Protection Agency.  1982.   Method  for use of caged
mussels to  monitor  for  bioaccumulation and selected  biological  responses of
toxic  substances in municipal wastewater discharges to marine waters.
Draft.   U.S. EPA  Env. Support Lab, Cincinnati,  OH.

Uzmann, J.R., R.A.  Cooper,  R.B.  Theroux, and  R.L. Wigley.  1977.  Synoptic
comparison  of three sampling  techniques for estimating abundance of selected
megafauna:   submersible vs.  camera sled  vs. otter trawl.   Mar.  Fish. Rev.
Venrick, E.L.   1978a.   Water-bottles,   pp. 33-40.   In:   Monographs on
Oceanographic  Methodology.  6.  Phytoplankton Manual.   United Nations
Educational, Scientific,  and Cultural Organization, Paris.
                                  133

-------
 Venrick, E.L.   1978b.  The  implications of subsampl ing.  pp. 75-87.   In:
 Monographs on  Oceanographic Methodology.  6.  Phytoplankton Manual.  United
 Nations Educational, Scientific,  and Cultural Organization, Paris.

 von Brandt, A.   1972.   Fish catching methods  of the world.  Fishing News
 (Books) Ltd.,  London.  240 pp.

 Wald, S.  1979.   Development of the vector measuring  current meter.  Oceans
 '79, Marine Tech.  Soc., and Inst. of Electrical  and  Electronics Engineers.
 5 pp.

 Walker,  H.A., S.B. Saila,  and  E.L.  Anderson.  1979.   Exploring  data
 structure of New  York Bight  benthic  data  using post-collection
 stratification  of samples and linear  discriminant  analyses  for species
 composition comparisons.  Estuarine and Coastal  Mar. Sci. 9(2):101-120.

 Walton, J.M.,  and N.W. Bartoo.   1976.   Flatfish densities determined with a
 diver operated flounder sampler.   J. Fish. Res.  Board  Can. 33:2834-2836.

 Wathne, F.  1977.  Performance  of trawls used in resource assessment.  Mar.
 Fish. Rev. 39(6):16-23.

 Watts, R.R.   1980.   Analysis of  pesticide  residues  in  human and
 environmental  samples.   EPA-600/8-80-038.   U.S.  Environmental Protection
 Agency Health  Effects Research  Laboratory,  Research  Triangle Park, NC.  26
 pp.

 Wiebe, P.M.,  and W.R. Holland.   1968.   Plankton patchiness:   Effects  on
 repeated net tows.  Limnol.  Oceanogr. 13:315-321.

 Wieser, W.  1960.   Benthic  studies in Buzzards  Bay.  II.   The meiofauna.
 Limnol. Oceanogr.  5(2):121-137.

 Williams,  D.D., and N.E. Williams.  1974.   A counterstaining technique for
 use in sorting  benthic samples.   Limnol. Oceanogr.  19(1):152-154.

 Williams,  W.T.  1971.   Principles of clustering.  Ann.  Rev. Ecol. Syst.
 2:303-326.

 Wilson, J.R.   1968.  Fluorometric procedures for dye  tracing.  U.S.
 Geological Survey  techniques of water-resources  investigations.   Book  3,
 Chapter A12.   31  pp.

 Wilson, K.C.,  A.J. Mearns,  and  J.J. Grant.   1980.  Changes in kelp forests
 at Palos Verdes.   pp. 77-92.   In:  Coastal  Water Research Project  Biennial
 Report for the Years 1979-1980, W. Bascom, (ed).   SCCWRP.  Long Beach,  CA.

 Winer,  B.J.   1971.   Statistical  principles  in  experimental   design.
McGraw-Hill Book Company,  New York, NY.

 Word, J.Q. 1976.  Biological  comparison of grab  sampling devices,   pp.
 189-194.   In:  Coastal  Water Research  Project Annual Report.  SCCWRP.  El
 Segundo,  CA.
                                  134

-------
Word, J.Q.  1978.   The infaunal  trophic index,   pp.  19-39.  In:   Coastal
Water Research Project, Annual  Report for the Year  1978, W.  Bascom (ed).
SCCWRP.  El Segundo,  CA.

Word, J.Q., P.I.  Striplin,  and D. Tsukada.   1980.  Effects of screen  size
and replication on  the  Infaunal  Trophic Index,  pp. 123-130.  In:   Coastal
Water Research Project, Biennial Report  for the years 1979-1980,  W.  Bascom,
(ed).  SCCWRP.  Long  Beach, CA.

Word, J.Q., B.L. Myers, and A.J. Mearns.   1977.  Animals that are  indicators
of marine pollution,   pp.  199-206.   In:  Coastal Water Research  Project
Annual  Report.   SCCWRP.  El Segundo,  CA.

Youden, W.J.   1960.  The  sample, the procedure,  the laboratory.  Anal.
Chem. 32(13):23A-37A.

Young, D.R., and  T.C. Heesen.   1974.   Inputs and  distributions of
chlorinated hydrocarbons  in  three southern California harbors.  TM 214.
SCCWRP.  El Segundo,  CA.  27 pp.

Young, D.R., T.C.  Heesen,  and  D.J.  McDermott.   1976.   An offshore
biomonitoring system for  chlorinated hydrocarbons.   Marine Pollution
Bulletin  7(1):156-159.

Zar,  J.H.   1974.  Biostatistical analysis.  Prentice-Hall, Inc., Englewood
Cliffs, NJ.  620 pp.
                                  135


                                             •D.3. GOVERNMENT PRINTING OFFICE : 1982 0-393-769/278'

-------
United States
Environmental Protection
Agency
Washington DC 20460
Official Business                                                                        Fourth-Class
Penalty for Private Use $300

-------