TC 3338-23
Task 5
Final Report
ELLIOTT BAY REVISED ACTION PROGRAM:
A STORM DRAIN MONITORING APPROACH
by
Tetra Tech, Inc.
for
U.S. Environmental Protection Agency
Region X - Office of Puget Sound
Seattle, WA
June 1988
Tetra Tech, Inc.
11820 Northup Way, Suite 100
Bellevue, Washington 98005
-------
EPA 910/9-88-207
uget Sound Estuary Program
ELLIOTT BAY
ACTION PROGRAM:
Storm Drain Monitoring
Approach
TC-3338-23
FINAL REPORT
June 1988
Prepared for
U.S. Environmental Protection Agency
Region X - Office of Puget Sound
Seattle, Washington
-------
PREFACE
This document was prepared by Tetra Tech, Inc. for the U.S. Environ-
mental Protection Agency (EPA) Region X, Office of Puget Sound under the
Elliott Bay Action Program work assignment of U.S. EPA Contract
No. 68-02-4341. The primary objective of the Elliott Bay Action Program is
to identify toxic contamination and appropriate corrective actions in
Elliott Bay and the lower Duwamish River. Corrective actions include source
controls and sediment remedial actions. An Interagency Work Group (IAWG),
comprising representatives from the U.S. EPA, Washington Department of
Ecology (Ecology), and other resource management agencies, provides technical
oversight for all work conducted under this work assignment.
A four-phased approach for identifying sources of toxic contaminants in
storm drains in the Puget Sound area is described in this report. The
evaluation of potential contaminant sources in the Elliott Bay project area
incorporated the first two phases of this monitoring approach (i.e., the
preliminary site investigation and the initial screening of sediment
contamination near the mouth of storm drains) (Tetra Tech 1988a).
The following reports are in preparation or have been drafted under the
Elliott Bay Action Program:
Analysis of toxic problem areas (PTI and Tetra Tech 1988)
Evaluation of potential contaminant sources (Tetra Tech 1988a)
Development of a revised action plan (PTI and Tetra Tech in
preparation)
Evaluation of the relationship between source control and
recovery of contaminated sediments (Tetra Tech 1988c)
ii
-------
Development of a storm drain monitoring approach (this report)
Evaluation of sediment remedial alternatives (Tetra Tech
1988b)
Development of an environmental monitoring approach (EVS and
Tetra Tech in preparation).
m
-------
CONTENTS
Page
PREFACE i 1
LIST OF FIGURES vi
LIST OF TABLES vii
ACKNOWLEDGMENTS ix
EXECUTIVE SUMMARY xi
SECTION 1.0 INTRODUCTION 1
SECTION 2.0 BACKGROUND 5
2.1 MONITORING STORMWATER RUNOFF 7
SECTION 3.0 PRELIMINARY INVESTIGATION 11
SECTION 4.0 PHASE I - INITIAL SCREENING 14
4.1 SELECTION OF STORM DRAINS 14
4.2 SAMPLE COLLECTION 17
4.2.1 Sampling Equipment and Procedures 17
4.2.2 Documentation 23
4.2.3 Sample Packaging and Shipping 26
4.2.4 Decontamination 28
4.2.5 Chemical and Physical Analyses 28
4.2.6 Quality Assurance/Quality Control 35
4.3 IDENTIFYING AND PROBLEM DRAINS 38
4.3.1 Evaluating Sediment Data 39
4.3.2 Ranking Procedure 49
SECTION 5.0 PHASE II - CONTAMINANT TRACING 62
5.1 SELECTION OF SAMPLING STATIONS 62
5.2 INTERPRETATION OF SEDIMENT CHEMISTRY DATA 66
5.3 ADDITIONAL INVESTIGATIONS 69
w
-------
5.4 SAMPLE COLLECTION 71
5,4,1 Chemical and Physical Analyses 71
5.4.2 Quality Assurance/Quality Control 72
SECTION 6.0 PHASE III - CONFIRMATION 73
6.1 DISCHARGE MONITORING TECHNIQUES 75
6.1.1 Bulk Water vs. Particulate Analysis 75
6.2 SAMPLE COLLECTION 77
6.2.1 Bulk Water Sampling 77
6.2.2 Particulate Fraction Sampling 80
6.2.3 Chemical Analyses 82
6.2.4 Quality Assurance/Quality Control 83
6.2.5 Data Interpretation 87
7.0 CONCLUSIONS 93
REFERENCES 95
APPENDIX A. STORM DRAIN MONITORING APPROACH COSTS A-l
APPENDIX B. SUMMARY OF PREVIOUS STORM DRAIN INVESTIGATIONS B-l
APPENDIX C. POLLUTANTS OF CONCERN Ctl
-------
FIGURES
Number Page
1 Overview of storm drain monitoring approach 3
2 Decision criteria to select storm drains for Phase I
initial screening 15
3 Example of station location and sample log form 21
4 Example of summary sampling log 22
5 Example of chain-of-custody record 24
6 Example of sample analysis request form 25
7 Decision criteria for selecting problem chemicals and
ranking problem storm drains 40
8 Schematic of a hypothetical storm drain system 64
B-l Metals concentration in sediments collected from the
Lander Street drains B-2
B-2 In-line sediment data for stations on SW Florida Street CSO/SD B-4
B-3 Summary of PCB data for Slip 4 drains B-6
B-4 Metro sampling stations on Fox Street CSO/SD (#116) B-ll
B-5 Sampling stations in Denny Way CSO source toxicant
investigation B-14
VI
-------
TABLES
Number Page
1 Traffic-related sources of roadway pollution 6
2 List of equipment needed for storm drain sediment sampling 19
3 Extractable organic compounds recommended for analysis
during Phase I screening 30
4 Limits of detection for metals in sediment 32
5 Sample containers, preservation, and recommended holding
times for sediment samples 34
6 Puget Sound AET values 42
7 Freshwater sediment criteria 45
8 Summary of metals measured in street dust samples collected
from Seattle and Bellevue 46
9 Summary of organic compounds measured in street dust samples
collected from Seattle and Bellevue 47
10 Summary of metal concentrations in sediments from
Puget Sound reference areas 51
11 Summary of organic compound concentrations in sediments from
Puget Sound reference areas 52
12 Summary of metal concentrations in sediments from Carr Inlet
reference area 56
13 Summary of organic compound concentrations in sediments from
Carr Inlet reference area 57
14 List of equipment needed for storm drain discharge sampling 79
15 Recommended methods, sample containers, preservation, and
holding times for water sample analysis 84
16 Volatile organic compounds recommended for analysis in
discharge samples 85
17 Summary of U.S. EPA water quality criteria 89
vn
-------
A-l Summary of analytical costs A-3
A-2 Approximate personnel costs for field sampling - sediment A-6
A-3 Approximate costs for sampling equipment - sediment A-7
A-4 Approximate personnel costs for field sampling - discharge A-9
A-5 Approximate costs for sampling equipment - discharge A-10
B-l Description of drains discharging into Slip 4 B-7
B-2 Summary of metals concentrations in sediment samples from
Fox Street CSO/SD #116 and surrounding area (mg/kg) B-10
B-3 Summary of metal concentrations in sediments collected
from storm drains discharging into Lake Union B-17
C-l Inorganic contaminants of potential concern in Puget Sound C-l
C-2 Organic contaminants of potential concern in Puget Sound C-2
C-3 Pollutants of concern list for Puget Sound C-7
vm
-------
ACKNOWLEDGMENTS
This document was prepared by Tetra Tech, Inc. under the direction of
Dr. Jean M. Jacoby, for U'.S. EPA in partial fulfillment of Contract
No. 68-0204341. This project was funded through the National Estuary
Program under the authorities of the Clean Water Act as amended. Funding
was approved by the U.S. EPA Office of Marine and Estuarine Protection. Dr.
Jack Gakstatter and Ms. Claire Ryan served as U.S. EPA Project Officers. Dr.
Don Wilson served as the Tetra Tech Program Manager.
The primary authors of this report are Ms. Beth Schmoyer, Dr. Jean
Jacoby and Mr. John Virgin of Tetra Tech, and Ms. Ann Bailey of EeoChem,
Inc. Ms. Jane Dewell and Ms. Sue Trevathan of Tetra Tech performed technical
editing and supervised report production. Ms. Karen Keeley of Tetra Tech
provided technical assistance in the revisions of this document.
Technical guidance and information was provided by Mr. Tim Sample and
Mr. Tom Hubbard of Municipality of Metropolitan Seattle (Metro), Ms. Lori
Geissinger of Seattle City Light, Mr. W.T. Clendaniel of City of Seattle
Sewer Utility, and Ms. Lee Oorigan of Ecology.
The Elliott Bay Action Program has benefited from the participation of
an IAWG and a Citizen's Advisory Committee (CAC). Duties of the IAWG and
CAC included 1) reviewing program documents, agency policies, and proposed
actions (including selection of problem areas for further study); 2)
providing data reports and other technical information to U.S. EPA; and 3)
disseminating action program information to respective interest groups or
constituencies. We thank the 'IAWG and CAC members for their past and
continuing efforts. We are especially grateful to Ms. Joan Thomas, Mr. Gary
Brugger, and Mr. Dan Cargill for chairing the IAWG, and to Mr. David
Schneidler and Ms. Janet Anderson for co-chairing the CAC.
IX
-------
The Tetra Tech production staff are also acknowledged for their
efforts: Ms. Vivia Boe (word processing), Ms. Pamela Charlesworth
(graphics), Ms. Betty Dowd (graphics), Ms. Lisa Fosse (word processing), Ms,
Joanne Graden (word processing), Ms. Rosemarie Jackson (report reproduction),
Ms. Kim Reading (graphics), Ms. Debra Shlosser (word processing), Ms. Gail
Singer (word processing), and Ms. Gestin Suttle (word processing).
-------
EXECUTIVE SUMMARY
One component of the Elliott Bay Action Program entails the development
of a generic approach to identify and prioritize potential sources of toxic
contaminants. A four-phased monitoring approach to trace contaminants and
identify sources of toxic contaminants in storm drains is presented in this
report. Although specifically developed for Puget Sound embayments receiving
toxic contaminants, the storm drain monitoring approach can be adapted for
other types of pollutants in other areas.
The monitoring approach comprises the following four phases:
Preliminary Investigation: Compile available information to
define the storm drain system, drainage basin characteris-
tics, and conditions in the receiving environment
Phase I - Initial Screening: Collect in-line sediment
samples near the mouths of storm drains to identify con-
taminated drainage systems
Phase II - Contaminant Tracing; Select problem drains for
further intensive inspection and conduct sampling activities
to trace contaminants and identify the ultimate sources of
contamination
Phase III - Confirmation: Confirm contaminant contributions
from individual sources and identify sources by collecting
water samples from side drain connections that discharge into
the storm drain.
This approach is an expansion of the in-line sediment sampling technique
used by Metro during the Duwamish industrial nonpoint source investigation
(Lampe, J., 21 January 1985, personal communication). The storm drain
XI
-------
monitoring program presented in this document is a sequential approach,
where the results of one phase determine the necessity of each successive
phase. For larger basins, Phase I and Phase II efforts are designed to
limit the size of the area investigated by excluding noncontaminated
sections of the drainage system in further monitoring efforts. It is
expected that the entire process may not be required to identify and trace
contaminants in all storm drain systems. For example, smaller drainage
basins that serve a limited number of potential sources may not require
additional Phase II contaminant-tracing procedures. In certain situations,
discharge sampling during Phase III may not be required to confirm con-
taminant contributions from specific sources if sources of contaminants are
identified during the preliminary investigation and sediment sampling
efforts.
BACKGROUND
Stormwater runoff that is collected and routed to storm drains is
difficult to monitor because of its intermittent and highly variable nature.
An alternate method of sampling storm drains has been used to avoid the
complications associated with stormwater (i.e., discharge) monitoring. This
alternate sampling approach uses in-line sediment samples collected from low
energy sections of the drainage system (i.e., manholes and shallow sloped
lines) to screen drainage systems for contamination.
Sediment sampling has several advantages over stormwater discharge
monitoring. First, sediment samples are collected from the storm drain
during dry (i.e., nonrainfall) conditions, thereby eliminating the need to
coordinate sampling efforts with rainfall events. Therefore, sediment
sampling is easier and less expensive than discharge monitoring. Second,
storm drain sediments serve as a sink for contaminants associated with the
particulate component of stormwater runoff. Sediments accumulate in low
energy areas of the storm drain system and generally provide a composite of
multiple storm events. However, storm drain sediment samples may be biased
toward larger-grained particles due to sedimentation processes within the
storm drain lines, and therefore, may not be representative of sediments
xii
-------
discharged to the receiving environment. In addition, sediment data cannot
be used to calculate pollutant loadings from the storm drain system.
Despite the above limitations, storm drain sediment samples can be used
to screen a large number of storm drains so that more intensive studies can
be focused on those basins whose drains are associated with toxic contamina-
tion. The four phases of the storm drain monitoring approach are described
in the following sections.
PRELIMINARY INVESTIGATION
A preliminary investigation is recommended as the first step in
conducting a storm drain investigation. This task will involve compiling
existing information to define storm drain systems, drainage basin charac-
teristics, and conditions in the receiving environment. The following are
major activities to be conducted during the preliminary investigation:
Review city utility plans to determine location and layout of
storm drain systems
Contact private property owners to obtain storm drain maps
Conduct shoreline reconnaissance to verify outfall locations
and to identify unmapped outfalls
Trace drainage basin boundaries for each storm drain system,
determine land-use characteristics, and determine potential
contaminant sources in each drainage basin
Compile and review available flow data, pollutant loading
data, and offshore sediment chemistry data for each storm
drain.
This information will be used to select storm drain systems that should be
screened for contaminants during Phase I.
xiii
-------
Phase I - INITIAL SCREENING
Phase I is designed to initially screen storm drains in the study area
for chemical contamination. The initial screening will involve collecting
and analyzing sediment samples from manholes located near the mouth of each
storm drain. Samples collected at the downstream end of the pipe will
provide an indication of contaminants in the entire system. This phase is
expected to minimize the amount of sampling required by eliminating
noncontaminated storm drains early in the investigation. Phase I screening
can be conducted in several steps with high-priority storm drains sampled
first, and lower-priority storm drains sampled at a later date. The
following components are included in Phase I screening:
Selection of storm drains based on information compiled
during the preliminary investigation
Sample collection, selection of appropriate variables for
chemical and physical analyses, and application of quality
assurance/quality control (QA/QC) procedures
Identification and prioritization of problem storm drains
using available sediment criteria, chemical concentrations in
reference area sediments, and loading indices.
The results of the initial screening are used to focus more intensive storm
drain investigations on problem (i.e., contaminated) drains during subsequent
sampling efforts.
Phase II - CONTAMINANT TRACING
The contaminant tracing phase of the investigation is an extension of
the initial screening phase. The objective of this phase is to isolate
contaminated sections of storm drain lines and associated drainage subbasins
in problem storm drains identified during Phase I screening. Source
identification efforts can then be focused on contaminated sections of storm
drain lines while uncontaminated sections can be eliminated from further
xiv
-------
study. To trace contaminants to the sources, additional field sampling and
continued investigation of land use in the drainage basin will be required.
Phase II will entail collecting additional sediment samples from manholes
throughout the storm drain system to trace contaminants in the problem storm
drains. The Phase II effort will focus on problem chemicals and associated
source categories identified during Phase I and the preliminary investiga-
tion. The Phase II sampling procedure is expected to be an interactive
process because it may take several rounds of sampling to isolate con-
taminated sections of the storm drain system and identify ultimate sources
of contaminants. Information obtained during the preliminary investigation
will be used to select sampling station locations. In addition, a detailed
investigation of industrial and commercial facilities operating in each
drainage basin will be required to support the sampling program.
Phase III - CONFIRMATION
The information obtained from Phase I screening and Phase II con-
taminant tracing, combined with the supporting evidence from the site
inspections, is expected to provide sufficient evidence to identify
contaminant sources for many problem drains. However, in some cases,
additional sampling efforts may be required to confirm contaminant contribu-
tions from specific sources. Source confirmation sampling performed during
Phase III entails the collection of water samples discharging to the storm
drain rather than the collection of sediment deposits in the drain. The
following situations may warrant discharge sampling:
To distinguish between historical and ongoing source
contributions
To confirm sources where volatile organic compounds are
suspected as the major toxic contaminant
To determine contributions from NPOES-permitted sources
To document source contaminant loading conditions for
possible enforcement actions.
xv
-------
SECTION 1.0. INTRODUCTION
The 1987 Puget Sound Water Quality Management Plan, prepared by the
Puget Sound Water Quality Authority (1987), included the recommendation that
urban storm water control programs be developed and implemented to reduce
pollutant loadings to Puget Sound. Under this plan, each city or urban area
will be required to develop storm water control programs. As part of these
programs, cities will be required to conduct storm drain investigations to
determine the location of existing storm drain systems, determine land use
characteristics in each drainage basin, and identify and monitor problem
storm drains. The primary objective of this report is to provide an
approach for identifying sources of toxic contaminants in storm drains in
the Puget Sound area. Although specifically developed for the Puget Sound
area, the storm drain monitoring approach can be adapted to other areas.
The monitoring approach presented in this report focuses on toxic chemical
contamination rather than conventional pollutants, such as nutrients and
microbial pathogens. A slight modification of this approach would allow the
identification of sources of conventional pollutants, which was not performed
as part of this study.
In this report, the following four-phased approach to conducting storm
drain investigations is recommended:
Preliminary Investigation: Compile available information to
define the storm drain system, drainage basin characteristics,
and conditions in the receiving environment
Phase I - Initial Screening: Collect in-line sediment
samples from near the mouths of storm drains to identify
contaminated drainage systems
-------
Phase II - Contaminant Tracing: Select problem drains for
further intensive inspection and conduct sampling activities
to trace contaminants and identify the ultimate sources of
contamination
Phase III - Confirmation: Confirm contaminant contributions
from individual sources and identify sources by collecting
water samples from side connections that discharge into the
storm drain.
This approach is based on a sampling technique that was developed and
used by Metro as part of their Duwamish industrial nonpoint source investiga-
tions (Lampe, J., 21 January 1985, personal communication). During that
study, Metro collected and analyzed in-line sediment samples from storm
drains and adjacent catch basins to identify contaminant sources. The
approach presented in this report is an expansion of Metro's techniques.
The procedures recommended for conducting storm drain investigations
(Figure 1) are applicable to any storm drain system, however, it is expected
that the entire process will not be applied in every case. As is shown in
Figure 1, smaller drainage basins may not require Phase II procedures.
Study of these small basins could possibly bypass Phase II and directly
implement Phase III if needed. This situation will likely occur in simple
drainage networks that serve a limited number of potential sources. For
larger basins, Phase I and Phase II efforts are designed to limit the size
of the area investigated by eliminating noncontaminated sections of the
drainage system from further analysis. This approach is intended to reduce
the amount of sampling required for the storm drain investigations by
focusing source identification activities only on contaminated areas.
The history of urban storm water pollution and rationale for recommend-
ing sediment sampling is presented in Section 2.0. In Section 3.0, the
process for preliminary investigations is explained. In Section 4.0,
initial screening of major storm drain systems is outlined. The process of
contaminant tracing in high-priority storm drain systems identified during
initial screening is presented in Section 5.0. In Section 6.0, confirmation
-------
PREUMINARV INVESTIGATION
(ALL STORM DRAINS)
SELECT DRAINS FOR
INfTIAL SCREENING
PHASE ONE-
INITIAL SCREENING
FURTHER
INVESTIQA.
ARE SEDIMENTS IN DRAIN
CONTAMINATED?
IS STORM DRAIN SYSTEM
COMPLEX?
COLLECT SEDIMENT SAMPLES
FROM STOflM DRAIN SIDE
CONNECTIONS
CONTAMINANT TRACING
IN DRAIN SYSTEM
DO CONCENTRATION GRADIENTS
IN STORM DRAIN SEDIMENTS
INDICATE SOURCE(S)?
IS CONTAMINATION FOUND
IN SOI CONNECTIONS?
PHASE THREE-
CONFIRMATION
DOES SITE INVESTIGATION
CONFIRM SOURCES?
INITIATE
SOURCE CONTROL
ACTIVITIES AND
REMOVE
CONTAMINATED
STORM DRAIN
SEDIMENTS
DOES DISCHARGE
MONITORING CONFIRM ONOOING
CONTAMINANT CONTRIBUTIONS?
NO ONGOING
SOURCE-REMOVE
CONTAMINATED
SEDIMENTS
moM DRAIN
Figure 1. Overview of storm drain monitoring approach.
-------
of contaminant sources to storm drain systems by further sampling is
explained. Report conclusions are presented in Section 7.0, Potential
costs of the storm drain monitoring approach are outlined in Appendix A of
this report. Appendix B contains a summary of previous storm drain investi-
gations and Appendix C is a list of pollutants of concern for Puget Sound.
-------
SECTION 2.0. BACKGROUND
Stormwater runoff is typically considered a nonpoint source of pollu-
tion, even though it is usually collected and routed to nearby surface
waters for disposal via ditches and pipes (i.e., point source discharges).
Nonpoint surface water pollution is generated when storm water comes into
contact with pollutants that have accumulated on land. The contamination of
stormwater runoff is generally related to land use in a drainage basin
(e.g., industrial, commercial, and residential uses in urban areasj agri-
cultural and silvacultural uses in rural areas). Sources of pollutants in
urban stormwater runoff can be categorized as follows:
Atmospheric deposition
Traffic-related sources (see Table 1)
Chemical spills
Waste and chemical storage and handling practices
Refuse deposition in streets
Urban erosion
Road de-icing.
Storwwater runoff, particularly runoff from urban areas, has long been
recognized as the source of a wide variety of pollutants to surface waters.
Early investigations of urban runoff pollution focused on conventional
pollutants (i.e., biochemical oxygen demand, total suspended solids,
coliform bacteria, and nutrients). Recently, however, the concern has
shifted toward toxic contaminants in urban runoff (i.e., metals and organic
compounds). In response to these concerns, U.S. EPA initiated the National
-------
TABLE 1. TRAFFIC-RELATED SOURCES OF ROADWAY POLLUTION
Pollutant Traffic-Related Source
Asbestos Clutch plates, brake linings
Copper Thrust bearings, brushings, and brake
1i n i ngs
Chromium Metal plating, rocker arms, crankshafts,
rings, brake linings, and pavement
materials
Lead Leaded gasoline, motor oil transmission
fluid, Babbitt metal bearings
Nickel Brake linings and pavement material
Phosphorous Motor oil
Zinc Motor oil and tires
Reference: Krenkel and Novotny 1980.
-------
Urban Runoff Program (NURP) in 1978 (U.S. EPA 1983c). The program was
developed to characterize water quality of urban runoff, determine the
effects of different land uses on composition and volume of runoff, and to
evaluate the effectiveness of management programs for controlling pollutant
loads in runoff. The study concluded that metals, especially copper, lead,
and zinc, are the most prevalent contaminants found in urban runoff.
Organic compounds, although detected much less frequently than the metals,
were also identified as a potential problem, but were considered site-
specific rather than widespread (U.S. EPA 1983c).
As a component of NURP, Metro measured contaminants in local urban
stormwater runoff (Galvin and Moore 1982). Six metals (i.e., arsenic,
cadmium, chromium, copper, lead, and zinc) were detected in all stormwater
runoff samples. Nineteen of 111 U.S. EPA priority pollutant organic compounds
were detected in stormwater runoff samples. The most frequently detected
organic compounds were pesticides and polynuclear aromatic hydrocarbons
(PAHs). In comparison with stormwater runoff, nearly twice as many organic
compounds were detected in samples of local street dust.
In another study, the sources of petroleum hydrocarbons to Lake
Washington were investigated (Wakeham 1977). The concentrations of
petroleum-type (i.e., aliphatic) hydrocarbons were measured in urban
stormwater runoff, runoff from bridges, rivers and creeks, rainfall, and
dustfall. Stormwater and bridge runoff were found to have the highest
concentrations of hydrocarbons.
2.1 MONITORING STORMWATER RUNOFF
Urban stormwater runoff is difficult to monitor because of its intermit-
tent and highly variable nature. Volume and pollutant loadings associated
with stormwater runoff are a function of many different variables, including
precipitation conditions, land use and cover in the basin, antecedent
moisture conditions, and illegal discharges (i.e., midnight dumpers). Peak
runoff periods, and therefore the bulk of the contaminant loading, generally
occur during intense rainfall events. Many pollutants adsorb onto soil
particles and are transported by surface runoff as particulates. Under
-------
high flow conditions, sediments are scoured from drainage ditches and pipes,
increasing the total loading to the receiving water body. As a result, it
is difficult to obtain representative samples of discharge from any one
drain. Therefore, stormwater monitoring typically requires that samples be
collected during several different storm events to characterize storm drain
loading. Even then, it is nearly impossible to sample at the exact time
when illegal discharges are occurring, so documentation of extreme cases of
pollutant loading is rare.
Tidal influences must also be considered in monitoring stormwater
runoff in the Puget Sound area because many storm drains serving metropolitan
areas along the sound are tidally influenced. Consequently, sampling must
be scheduled during periods of low tide to reduce saltwater intrusion to the
storm drain lines. Because rainfall events can occur at any tidal stage, it
is often difficult to catch a low tide storm event for discharge sampling.
Collecting representative storm drain samples above the tidally influenced
portion of the drain line is generally not an option because heavily
developed areas are frequently located along the waterfront. As a result,
samples collected upstream of the tidal area may exclude a significant
portion of contaminant loading to the drain. However, the outfalls and
downstream sections of some storm drains may be tidally influenced, even at
low tide. In these drains, it will only be possible to collect discharge
samples from stations located above the tidally influenced section of the
drain system.
An alternate method of sampling storm drains has been developed to
avoid the complications of stormwater monitoring. This alternate sampling
approach uses in-line sediment samples collected from low energy sections of
the drainage system (i.e., manholes and shallow sloped lines) to screen
drainage systems for contamination. This approach has been used locally by
Metro and the City of Seattle (see Appendix B) and nationally (e.g., Wilber
and Hunter 1979). Sediment sampling has several advantages over stormwater
monitoring. First, sediment samples are simply collected from the storm
drain system during dry (i.e., nonrainfall) conditions. No coordination
with rainfall events is required, making sediment sampling easier and
therefore less expensive to collect than water samples. Second, storm drain
-------
sediments act as a natural sink for contaminants associated with the
participate component of stormwater runoff. Sediments deposit in low energy
areas of the storm drain system, accumulating through successive storms.
Therefore, they generally provide a composite of several storm events and
can be used to characterize contamination in storm drain lines. Sediment
sampling is scheduled during low tide to enable entry to the manhole or
drain line for sample collection.
There are disadvantages to sediment sampling. First, sediment data
cannot be used to calculate pollutant loadings (measured in Ib/day) from the
storm drain system. Information on pollutant loadings is often used to
prioritize pollutant sources by indicating the degree of potential effects
on the receiving environment. Second, no specific criteria exist to aid in
interpreting potential effects of storm drain sediment data, while criteria
do exist for water quality data. However, the recently developed Apparent
Effects Threshold (AET) (Tetra Tech 1986b) approach for sediments can be
used to assess toxicity of marine sediments. In addition, sediment data can
be compared with data collected from receiving environment reference areas
and with data from normal urban street dust (e.g., Salvin and Moore 1982;
Wilber and Hunter 1979). Third, sediment sampling suffers from inherent
difficulties in obtaining representative samples. For example, runoff tidal
action may disturb sediment deposits in the drain and affect contaminant
distribution patterns. Fourth, storm drain sediment samples may be biased
toward larger-grained particles due to sedimentation processes within the
storm drain lines, and therefore, may not be representative of sediments
discharged to the receiving environment.
It should be emphasized, however, that storm drain sediment sampling is
intended as a screening tool and has been used by Metro and the City of
Seattle to trace contaminants in storm drain lines (see Appendix B).
Sediment data alone will probably not be sufficient to confirm contaminant
sources, and other supporting evidence (e.g., documented spills and
discharges, evidence of improper chemical storage at facilities, discharge
monitoring) will be required. The storm drain sediment sampling approach
outlined in this report should be used primarily for initial screening of
-------
large numbers of storm drains so that future, more intensive studies can be
focused on major problem storm drain systems.
10
-------
SECTION 3.0. PRELIMINARY INVESTIGATION
A preliminary investigation is recommended as the first step in
conducting a storm drain investigation. This task will involve compiling
existing information to define storm drain systems, drainage basin charac-
teristics, and conditions in the receiving environment. This information
will be used to select storm drain systems that should be screened during
Phase I for contaminants. The following are major activities to be conducted
during the preliminary investigation:
Review city utility plans to determine location and layout of
storm drain systems
Contact private property owners to obtain storm drain maps
Conduct shoreline reconnaissance to verify outfall locations
and to identify unmapped outfalls
Trace drainage basin boundaries for each storm drain system,
determine land use characteristics, and determine potential
pollutant sources in each drainage basin
Compile and review available flow data, pollutant loading
data, and offshore sediment chemistry data for each storm
drain.
Detailed maps of the storm drain system are needed to determine the
location of existing drain lines, access points to the drain lines (i.e.,
manholes), and outfalls. Most cities maintain utility plans of their storm
drain systems that are periodically updated to reflect changes and modifica-
tions in the system. These plans typically show the general layout of the
system, manhole locations, and occasionally topographic information.
Engineering plans may include detailed design information such as profiles
of the storm drain system.
11
-------
All storm drain system plans should be verified in the field. Field
verification is required because many cities do not have as-built drawings
for their storm drain system, and the system actually constructed may vary
considerably from design plans. Field verification will involve inspecting
drain lines, manholes, and outfall locations. In addition, a shoreline
reconnaissance should be conducted to determine locations of outfalls not
marked on existing utility plans. Shoreline inspections should be conducted
at low tide when most outfalls will be exposed. In waterfront areas where
the beach is not exposed, a small boat should be used for inspection of
bulkheads and underneath piers.
Most cities require that private property owners inform them of any
side connections to the storm drain system so the city can inspect and map
these connections. Some cities maintain information (i.e., side sewer
cards) that show locations of side connections and catch basins within the
storm drain system. These detailed plans are useful in defining drainage
basin boundaries. However, private property owners often tie into the city
storm drain system without reporting to the city. Therefore, the side sewer
cards may not be accurate. This is frequently a problem along the waterfront
where many industrial facilities are located and may tie into city storm
drain systems without the city's knowledge. In addition, many large
industrial complexes maintain their own storm drain systems that discharge
directly to area waterways. To ensure that major storm drain systems are
identified, it is recommended that private property owners, especially along
the waterfront, be contacted to obtain storm drain system plans for their
property. These plans should be field checked to verify the location of
storm drain outfalls. Detailed inspection of the drain lines and manhole
locations will probably not be needed until Phase I screening.
Storm drain plans should be used to trace drainage basin boundaries for
each storm drain system. In addition, contributing areas should be
calculated, land use characteristics assessed, and potential pollutant
sources in each basin mapped. Pollutant source information is generally
available from local, state, or federal agencies. The U.S. EPA regional
program offices maintain lists of permitted facilities and potential
12
-------
hazardous waste sites in their region. CERCLIS, a list of Superfund sites
currently under investigation by U.S. EPA Region X, can be obtained from the
Comprehensive Environmental Response, Compensation, and Liability Act
(CERCLA) program office (U.S. EPA, 22 October 1987, personal communication).
The Resource Conservation and Recovery Act (RCRA) program office keeps a
list of RCRA-permitted facilities and facilities that are in the process of
applying for a RCRA permit. Ecology keeps records of all dischargers and
daily monitoring reports for National Pollutant Discharge and Elimination
System (NPDES)-permitted facilities. Other lists of specific problem sites
may be available from individual program offices within Ecology. In
addition, the Washington Department of Revenue, Division of Information
Systems maintains a computerized list of all businesses by address and
Standard Industrial Code (SIC) for tax purposes.
The final activity during the preliminary investigation is to compile
available storm drain pollutant loading data and offshore sediment chemistry
data. U.S. EPA has only recently included storm drains in the NPDES permit
program, therefore, little information is probably available on storm drain
pollutant loadings. The best sources for storm drain information are
Ecology, U.S. EPA, local universities, and Metro. In addition to these four
sources, National Oceanic and Atmospheric Administration (NOAA) would have
information on offshore sediment chemistry.
13
-------
SECTION 4.0. PHASE I - INITIAL SCREENING
Phase I is designed to initially screen storm drains in the study area
for chemical contamination. The initial screening will involve collecting
and analyzing sediment samples from manholes located near the mouth of each
storm drain. Samples collected at the downstream end of the pipe will
provide an indication of contaminants in the entire system. The results of
this initial screening are used to focus future, more intensive storm drain
investigations only on problem (i.e., contaminated) drains. This procedure
is expected to minimize the amount of sampling required by eliminating
noncontaminated storm drains early in the investigation. Phase I screening
can be conducted in several steps with high-priority storm drain systems
sampled first, and lower-priority storm drains sampled at a later date.
4.1 SELECTION OF STORM DRAINS
Selection of storm drains to be sampled during initial screening should
be based on information compiled during the preliminary investigation. The
first two points to consider are whether problem areas exist in offshore
sediments or whether contamination problems exist in the drainage basin. If
either of these situations exist, the storm drain system immediately
qualifies for Phase II contaminant tracing (Figure 2). If neither of these
situations exist, the following criteria should be considered:
Average annual discharge from the storm drain
Land-use characteristics in the drainage basin
Sensitivity of offshore environment and/or recreational uses.
If available data reveal contamination in offshore sediments that cannot
be attributed to a specific point source (e.g., chemical spill, industrial
14
-------
KNOWN CONTAMINATION
IN OFFSHORE SEDIMENTS?
YES
NO
KNOWN CONTAMINANT
DISCHARGES IN DRAINAGE BASIN?
YES
NO
YES
DOES DRAIN DISCHARGE
LARGE VOLUME OF RUNOFF
TO RECEIVING ENVIRONMENT?
NO
i
IS DRAINAGE BASIN HIGHLY
DEVELOPED/INDUSTRIALIZED?
YES
DO SENSITIVE HABITATS ANCHOR
RECREATIONAL USES EXIST
IN RECEIVING ENVIRONMENT?
NO
LOW PRIORITY STORM DRAINS
TARGET DRAIN FOR FUTURE
SCREENING BASED ON
AVAILABILITY OF FUNDING
NO
YES
HIGH PRIORITY STORM DRAINS
SELECT DRAIN FOR CHEMICAL
CONTAMINANT SCREENING
Figure 2. Decision criteria to select storm drains for phase
initial screening.
15
-------
discharge), then the storm drains that discharge nearby should automat-
ically be selected for Phase I screening. In addition, if the preliminary
investigation identifies potential problem sites within the drainage basin
(e.g., uncontrolled hazardous waste sites, industrial discharges to the
storm drain, or improper chemical storage and handling practices), the storm
drain serving the basin should be targeted for chemical screening.
Storm drain systems not associated with documented contaminated
offshore sediments or contaminant sources in the drainage basin should be
selected for chemical screening based on a priority ranking system. The
criteria recommended for ranking include estimated average annual storm drain
discharge, land use characteristics in the drainage basin, and existence of
sensitive habitats in the receiving environment. These criteria provide an
indication of potential loading from the drain and possible effects on the
receiving environment. A schematic of the decision criteria involved in
selecting storm drains for chemical screening is presented in Figure 2.
Average annual discharge is important because it can indicate the
loading potential for each storm drain (Figure 2). Where data are available,
average annual discharge should be estimated based on existing records. If
no data are available, annual discharge can be approximated based on the
drainage basin area, land use and cover in the basin, and average annual
precipitation. Storm drains with the highest estimated annual discharge
should be given high priority for chemical screening because these drains
have a high probability of impacting the receiving environment. Impacts
from smaller storm drains are expected to be less extensive and more
localized. However, before ranking smaller storm drains as low in priority,
it is recommended that conditions in the receiving environment be examined
to determine existence of sensitive habitats (e.g., shellfish beds, eel
grass and kelp beds, commercial/recreational fisheries, and nursery or
spawning grounds) and potential of high contact recreational uses (i.e.,
boating and swimming). If sensitive habitats or heavy recreational areas
exist in the offshore environment, the small drains should be targeted for
chemical screening. Storm drains with low annual discharge and little
potential for impacting a sensitive receiving environment can be given a low
priority for chemical screening. Chemical screening of these low priority
16
-------
storm drains could be postponed, depending on the availability of funds, so
that intensive sampling can be carried out as soon as possible in the high
priority storm drains.
Land-use characteristics in each drainage basin are recommended as a
criterion in selecting storm drains for Phase I (Figure 2). Land use and
zoning maps of the area should be reviewed to determine distribution of
industrial, commercial, residential, and undeveloped property in the
drainage basin. Industrialized areas are suspected as a major source of
contaminants to surface water runoff because of industrial plant emissions,
possible improper storage and disposal of industrial chemicals, and chemical
spills. Storm drains serving highly industrialized and commercial areas
should be given a high priority for chemical screening. Nonindustrialized,
heavily developed (i.e., residential) and undeveloped areas should have
conditions in the receiving environment investigated before being assigned a
low priority. If sensitive habitats or recreational areas exist in the
offshore environment, a high priority should be assigned to storm drains
serving residential and undeveloped areas.
4.2 SAMPLE COLLECTION
It is recommended that Phase I screening of storm drains be conducted
during a dry period when rainfall will not greatly affect sediment accumu-
lations in the storm drains. Access to manholes on tidally influenced storm
drains will only be possible during low tide. Therefore, in many cases the
scheduling of the sampling program must be based on tide schedule, as well
as weather conditions. Sampling should also be coordinated with local
stormwater drainage utilities to avoid potential interference from routine
maintenance operations such as catch basin cleaning activities.
4.2.1 Sampling Equipment and Procedures
In-line sediment samples should be collected from manholes located near
the mouths of each of the high priority storm drains. Tidally influenced
drains must be sampled during low tides to enable access to the manholes for
collection of sediment samples. All sampling activities should be co-
17
-------
ordinated with local drainage utilities. Coordination with local utilities
is especially important in areas where catch basin and storm drain cleaning
programs are conducted because these activities may interfere with sample
collection.
If insufficient sediment is found at the proposed sampling station, an
alternate station farther upstream should be selected in the drain line. If
an alternate station is not identified in that drain line, sediments should
be collected in adjacent catch basins. If an adequate sampling station is
still not identified, and the drain lines have been checked on several
occasions, it may be necessary to defer sediment collection efforts and
initiate discharge monitoring.
A list of the equipment needed for storm drain sediment sampling
activities is provided in Table 2. The following safety precautions and
methods are recommended for manhole entry and sediment sampling (Conklin
1986) :
When necessary, erect traffic barricades and markers around
the area before the manhole is opened. If the manhole is
located along a busy street or intersection, flaggers must be
provided to divert traffic around the area.
Prior to entry, measure the depth of water in the manhole to
determine whether manhole entry and sediment collection will
be feasible. Test the atmosphere in the manhole to measure
oxygen content, combustible gas, hydrogen sulfide, and
organic vapor concentrations. This information will be used
to determine the level-of respiratory protection required.
In all cases, individuals entering the storm drain should
wear at least Level C protective equipment (i.e., respirator,
coveralls, gloves, boots, safety harness, and line). In
addition, one rescue person at the surface should be dressed
in similar protective clothing. If the atmosphere measure-
ments indicate that conditions warrant Level B respiratory
18
-------
TABLE 2. LIST OF EQUIPMENT NEEDED FOR STORM DRAIN
SEDIMENT SAMPLING
Hard hats
Lights
Maps
Camera and film
Manhole cover hook
Manhole depth and water level mea-
suring device
Sledge hammer
Methanol
Squirt bottles
Waste solvent bottle and funnel
Bags - garbage, small plastic
Rope
Barricades, traffic cones, traffic
signs
Sampling equipment:
Stainless steel bucket
Extension pole
(2) Stainless steel scoops
Stainless steel spoons (long-
handled and teaspoons)
Aluminum foil
Sample containers (organic compounds,
metals, total organic carbon, oil
and grease, grain size)
Coolers
Ice
Custody seals
Chain-of-custody forms
Analysis request forms
Field data log forms
Field logbook
Sample tags
Clear tape
Marking pens
Knife
Sample tray
Kimwipes or equivalent
Gloves (leather and chemical
resistant)
Coveralls (cotton and chemical
resistant)
Respirators
(including extra filters)
Waders (two pair)
Duct tape
Op/combustible gas meter and tubing
Pnotoionization detector (PID)
meter and tubing
Drager tubes/bellows
Decontamination sprayer
Brushes (for decontamination)
Alconox or equivalent
First aid kit
Safety harness and rope
Clipboard
Tide tables
Self-contained breathing apparatus
(SCBA) equipment
19
-------
equipment [self-contained breathing apparatus (SCBA)], a
decision should be made whether to enter the manhole or
select an alternative manhole for sampling. Sampling of a
different manhole may be necessary if manhole dimensions
preclude entry with Level B SCBA and equipment.
Collect samples from the sediment deposits in the drain
system using stainless steel sampling equipment. A sufficient
quantity of sediment for the chemical and physical analyses
(see Section 4.2.5) should be placed in a pre-eleaned
stainless steel bucket and brought to the surface. Document
sampling locations with a map showing where the sediment
sample was collected (e.g., near discharge pipe in manhole or
influent line to manhole).
If insufficient sediment is found at the proposed sampling
station, select an alternate station farther upstream in the
drain line.
Thoroughly homogenize the sediment sample in a bucket prior
to filling the sample bottles. Because storm drain sediment
samples are not recommended for volatile organic analyses, it
is not necessary to retain any unhomogenized sediments.
Fill all sample containers with homogenized sediments. Label
each sample container with sampling station location, date
sample was collected, sampler's initials, and preservative
used. Place a custody seal on both the glass and lid so that
the custody seal must be broken to open the sample container.
Immediately place the sample containers in a cooler and pack
with ice. Complete a sample log form (Figure 3) and record
samples on the summary sampling log (Figure 4).
20
-------
STORM DRAIN SAMPLING
Station Location and Sample Log
DATE
TIME
STATION
LOCATION
METER
READINGS
02-
COMB, GAS -
HNu/OVA.
H2S.
PERSON SAMPLING.
SAMPLE
NUMBER
WATER:
DEPTH
FLOW
SEDIMENT:
TYPE
DEPTH
COLOR
ODOR
COMMENTS
SKETCH OF MANHOLE SAMPLING LOCATION
RECORDER
Figure 3. Example of station location and sample log form.
21
-------
SUMMARY SAMPLING LOG
PAGE OF.
SURVEY;
SAMPLING
DATE
STATION
SAMPLER
HOfilZON
SAMPLE
NUMBER
SAMPLES COLLECTED
§
&
&
§
1
^
1
Q
d
c
RECQRnPR! QflG.GQnF- DATF:
Figure 4. Example of summary sampling log.
22
-------
Wash all sampling equipment with water and methanol to prevent
cross-contamination of the samples between sampling stations.
Cover the clean sampling containers with aluminum foil to
prevent atmospheric contamination by dust and soot particles.
At the end of each day, complete a chain-of-custody record
(Figure 5) and the sample analysis request form (Figure 6)
for all samples.
4.2.2 Documentation
All pertinent field survey and sampling information should be recorded
in a bound logbook. Sufficient information should be provided for each
day's activities so that someone can reconstruct the field activity without
relying on the memory of the field crew. Entries should be made in indelible
ink. At a minimum, entries in the logbook should include the following:
Date and time of starting work
Names of field task leader and team members
Purpose of proposed work effort
Description of sampling station locations, including map
reference
Details of work effort, particularly any deviation from the
proposed procedures
Field observations
Field measurements (e.g., oxygen, combustible gas, organic
vapor meter readings, hydrogen sulfide measurements).
Photographs should be taken to document sampling station locations,
because they provide the most accurate record of the field worker's
23
-------
FIELD SAMPLE DATA AND CHAIN OF CUSTODY SHEET
*»* »^» 10
Protect Cod*:
N»m«/lOC»tk
Project Officw
u
DO
OO
I/1U
-
-
m
Account:
MArmx
-
1
1
..
1
3
i mtsEitv
1
|
1
LA*
MUMMN
Vr
Mt
*«
D PMfcfe Tosk/H«afdou
D Data ConM«ntW
D Dttafar StorM
STATION
NUMMH
1 Males:
S*
R<
DATE
Vr
Ma
Ov
Urn.
COMFOSITt ONLY
ENOrttfi 6ATI
Ma
Or
Tknt
1
Fm,
mpt«n
cord*r
i|HHII linn ll«
STATVX
LAB
NUMBER
Ti
-
-
«M
S*q
-
DEPTH
I
!
COL
HTC
CO
COM
Tttlf
OK
C
CNOCTVTV
MISCaxAMtOUS
CHAM Of CUSTODY NfCOMO
DHDHm
nR;=
nmttumut *»».
am
am
uew
W*5S= HOT
VliOBUUI Mil
1MT
twr
mar
-"" """sr< ~°"
Figure 5. Example of chain-of-custody record.
-------
SAMPLE ANALYSIS REQUEST
PACKING LIST
PROJECT:
SAMPLING CONTACT:
(name)
(phone)
SAMPLING DATE(S):
DATE SHIPPED:
TASK NAME/CODE:
SHIP TO:
ATTN:
FOR LAB USE ONLY
DATE SAMPLES RECEIVED:
RECEIVED BY:
1..
2..
3..
4..
5..
6,.
7..
8..
9..
10,.
11,.
12..
13..
14..
15..
16..
17..
18..
19..
20..
SAMPLE
NUMBERS
SAMPLE DESCRIPTION
(ANALYSIS/MATRIX/CONCENTRATION/PRESERVATIVE)
Figure 6. Example of sample analysis request form.
25
-------
observations. Each photograph should be documented with the following
information:
Date and time
Name of photographer
Description of station location
General magnetic direction and description of the subject
Sequential number of the photograph and roll number.
Once a roll of film is developed, the slides or prints should be placed in
the project file.
4.2.3 Sample Packaging and Shipping
Samples should be packed securely to prevent spills and breaking during
sample shipment. Recommendations for packaging nonhazardous samples are
presented below (49 CFR 173):
Place sample container in a 2-mil thick (or thicker) polyethy-
lene bag, one sample per bag. Position identification tag so
it can be read through the bag. Seal the bag.
Place sealed bags inside a strong outside container, such as
a lined metal picnic cooler or a Department of Transportation
(DOT)-approved fiberboard box. The outside container should
be lined with a polyethylene bag. Surround the sample
containers with noncombustible cushioning material for
stability during transport. Seal the large polyethylene
liner bag.
26
-------
To keep samples cool packaged blue ice or polyethylene
containers of frozen water should be placed in the shipping
container.
Place the laboratory and sampling paperwork in a large
envelope and tape it to the inside lid of the shipping
container.
Close and seal the outside container with fiberglass tape.
An additional packaging requirement is necessary for samples that are
suspected of containing hazardous materials based on observations made
during the field sampling or from information obtained during the preliminary
investigation. For hazardous materials, it is recommended that each sealed
bag containing a sample container be placed inside a metal can prior to
packaging in a lined metal cooler or DOT-approved fiberboard box. The
metal can should be lined with enough noncombustible, absorbent material
(e.g., vermiculite or diatomaceous earth) between the bottom and sides of
the can and the sample bag to prevent breaking and to absorb any leakage.
Pack only one bag per can using clips or tape to hold the can lid securely
and tightly.
The outside of the shipping container should be marked with the
laboratory name and address, and the return name and address of the sender.
A "Cargo Aircraft Only" DOT label and the following descriptive information
should be clearly printed on each shipping containers "Laboratory Samples,"
"This End Up," and "Inside packages comply with prescribed regulations,"
Hazardous materials should additionally be labelled with the DOT "Flammable
Liquid n.o.s." label. Arrows pointing upward should be placed on all four
sides of the shipping container.
Shipping documents must accompany the sample shipment and should be
taped to the inside lid of the outside sample container. These documents
are the chain-of-custody form (see Figure 5) and sample analysis request form
(see Figure 6).
27
-------
4.2.4 Decontamination
Decontamination of sampling equipment and personal protective gear is
required to prevent sample cross-contamination and to assist in maintaining
health and safety of field personnel. The following general procedures are
recommended for decontamination:
After sampling is completed at each station, remove sediment
residues remaining on boots and sampling equipment with a
high pressure sprayer filled with water. Sediment residues
can be returned to the manhole.
Wash sampling equipment (e.g., spoons, buckets, shovels) with
laboratory-grade detergent solution (e.g., Alconox) and rinse
with water. Detergent and rinse water can be disposed in a
nearby sanitary sewer.
All sampling equipment should be rinsed with methanol.
Solvents used for decontamination must be collected, placed
in an approved waste container, and transported to a licensed
waste recycling facility at the end of the project.
A final rinse with distilled water is also recommended.
Outer gloves worn by field personnel should be changed
between each station to prevent cross-contamination of
samples.
4.2.5 Chemical and Physical Analyses
Analysis of storm drain sediments should be performed using methods
recommended by Puget Sound Estuary Program (PSEP). In the past, collection
and analysis of Puget Sound environmental samples in different studies were
performed using nonstandardized protocols. The data generated using these
nonstandardized protocols were acceptable for individual project objectives,
but the differences in protocols limited comparability of data between
28
-------
studies. PSEP formulated a compendium of recommended methods (Tetra Tech
1986d) to overcome these problems in future Puget Sound studies. The
majority of commercial laboratories in the Puget Sound area are familiar
with the PSEP methods and their application. The use of PSEP protocols is
strongly recommended for storm drain sediment analysis to provide data that
will be directly comparable on a regional basis.
Selection of appropriate variables for chemical and physical analyses
is essential during the initial screening of storm drains for toxic pollu-
tants. Because phase one is intended to screen storm drains for chemical
contamination, it is recommended that a broad range of chemicals be analyzed.
U.S. EPA has developed the Hazardous Substance List (HSL) [also known as
the Target Compound List (TCL)] which contains all 126 priority pollutants
and additional compounds targeted for Superfund site investigations.
Analysis of storm drain sediments in the initial screening phase should be
performed using PSEP protocols for the following classes of chemicals on the
TCL:
Extractable organic compounds (Table 3)
Metals (Table 4).
In addition, the following conventional variables are recommended for
analysis:
Total solids
Total organic carbon
Oil and grease
Particle size.
The PSEP protocols provide two levels of analysis for extractable
organicsj screening and low level. The differences in the level of
analysis are most evident in the detection limits achieved. Detection
29
-------
TABLE 3. EXTRACTABLE ORGANIC COMPOUNDS RECOMMENDED
FOR ANALYSIS DURING PHASE ONE SCREENING
Acid Extractable Compounds
Phenols Substituted Phenols
Phenol 2-Chlorophenol
2-Methy1 phenol 2,4-Di ch1orophenol
4-Methy1 phenol 4-Chloro-3-Methy1 phenol
2,4-Dimethylphenol 2,4,6-Triehlorophenol
2,4,5-Trichlorophenol
Pentachlorophenol
2-Nitrophenol
2,4-Dinitrophenol
4-Nitrophenol
4f6-Dinitro-2-methylphenol
Base/Neutral Extractable Compounds
LowMolecular Weight
Polvnuclear Aromatic Hydrocarbons Hal.oaenat.ed Ethers
Naphthalene Bis
Acenaphthylene Bis
Acenaphthene Bis
2-chloroethyl)ether
2-chloroisopropyl)ether
2-chloroethoxy)methane
Fluorene 4-Chlorophenyl phenyl ether
Phenanthrene 4-Bromophenyl phenyl ether
Anthracene
High Molecular Weight
Polvnuclear Aromatic Hydrocarbons Phthalates
Fluoranthene Dimethyl phthalate
Pyrene Diethyl phthalate
Benzo(a)anthracene Di-n-butyl phthalate
Chrysene Butyl benzyl phthalate
Benzo(b
Benzoik
Benzo(a
fluoranthene Bis(2-ethylhexyl) phthalate
fluoranthene Di-n-oetyl phthalate
pyrene
Indeno(l,2,3-e,d)pyrene
Dibenzo(a,h)anthracene
Benzo(g,h,i)pery1ene
Chlorinated aromatic hydrocarbons Miscellaneous oxygenated compounds
1,3-Dichlorobenzene Isophorone
1,4-Dichlorobenzene Benzyl alcohol
1,2-Dichlorobenzene Benzoic acid
1,2,4-Tri ch1orobenzene Di benzofuran
2-Chloronaphthalene
Hexach1orobenzene
30
-------
TABLE 3. (Continued)
Base/Neutral Extractable Compounds (continued)
Oroanonltrogen Compounds Chlorinated Aliphatic Hydrocarbons
Aniline Hexachloroethane
Nitrobenzene Hexachlorobutadiene
n-Nitroso-di-n-propylamine Hexachlorocyclopentadiene
4-Chloroaniline
2-Nitroaniline Substituted Aromatics
3-Ni troani1i ne 2-Methylnaphtha!ene
4-Nitroaniline
2,6-Di ni trotoluene
2,4-Dinitrotoluene
n-Nitrosodiphenylamine
Benzidine
3,3'-Dichlorobenzidine
Pesticides PCBs
p,p'-DDE Aroclor 1016
p.p'-DDD Aroclor 1221
p,p'-DDT Aroclor 1242
Aldrin Aroclor 1248
Dieldrin Aroclor 1254
Chlordane Aroclor 1260
alpha-Endosulfan
beta-Endosulfan
Endosulfan sulfate
Endrin
Endrin aldehyde
Heptachlor
Heptachlor epoxide
al pha-Hexachlorocyclohexane (HCH)
beta-HCH
delta-HCH
gamma-HCH (Lindane)
Toxaphene
31
-------
TABLE 4. LIMITS OF DETECTION FOR METALS IN SEDIMENT
Analytical Instrument3*^ Recommended
ICP AA Limits of Detection0
Antimony
Arsenic
Cadmi um
Copper
Iron
Lead
Mercury
Manganese
Nickel
Silver
Zinc
3.2
4.0
0.6
0.7
4.2
-
2.0
1.5
0.7
0.2
0,ld
O.ld
O.ld
O.ld
O.ld
0.01e
O.ld
O.ld
0.2d
0.1
0.1
0.1
0.1
0.7
0.1
0.01
2.0
0.1
0.1
0.2
a ICP » Inductively coupled plasma atomic emission spectroscopy.
b ICP data are from Tetra Tech (1984) SFAA and CVAA data are detection
limits that can be reasonably attained by various laboratories. Under
strict conditions these limits can be lowered (Battelle 1985). Values are
mg/kg dry weight for 5 g (wet) sediment in a 100 ml digestate. These values
are provided as examples of typical attainable levels.
c Values are mg/kg dry weight. Limits of detection were selected by
considering attainable recommended limits for different instruments and
reported environmental levels.
d Graphite furnace atomic absorption.
e Cold vapor atomic absorption.
References Tetra Tech (1986d).
32
-------
limits for the screening level analysis are 500-1,000 ppb (dry weight) for
acid/neutral compounds, and 15-300 ppb (dry weight) for pesticides and PCBs.
Detection limits for the low level analysis are 1-50 ppb (dry weight) for
acid/neutral compounds, and 0.1-15 ppb (dry weight) for pesticides and PCBs,
For Phase I screening of storm drain sediments, low level analysis detection
limits are recommended because screening level analysis detection limits for
some compounds are higher than available sediment criteria. Metals
recommended for analysis under PSEP protocols and their detection limits are
presented in Table 4.
Total solids are determined so that sediment chemical concentrations
can be converted from a wet-weight to a dry-weight basis. Total solids
concentrations are normally determined as part of the extractable organic
compounds and metals analyses, and should be specified for determination by
the laboratory. Total organic carbon is a measure of the organic matter in
a sample. Oil and grease tests measure all materials that are soluble in a
nonpolar solvent (e.g., Freon) under acidic conditions. Hydrocarbons,
vegetable oils, animal fats, waxes, soaps, greases, and related industrial
compounds are included in the oil and grease concentrations. Particle size
(i.e., grain size distribution) is analyzed so that contaminant concentra-
tions can be normalized to the percent fines (percent fines = percent clay +
percent silt). In general, samples containing higher percentages of percent
fines and/or organic carbon will have higher contaminant concentrations
because of the greater sorption capacity of fine particulates. To account
for these sample characteristics, data can be normalized to total percent
fines or organic carbon content.
The analytical methods for PSEP specify the minimum sample volumes
required for chemical analysis, appropriate sample containers and preserva-
tives for each chemical analysis, and recommended holding times for samples
prior to analysis (Table 5). Based on the minimum sample volume require-
ments, approximately 500 g (minimum 1.75 L) will be required for analyses
recommended for Phase I. Specified containers are adequate for collection
of the sediment sample plus an amount of sample sufficient for QA/QC
samples. Prior to collecting any samples, the laboratory performing the
analyses should be consulted to confirm that the volumes of sediment
33
-------
TABLE 5. SAMPLE CONTAINERS, PRESERVATION, AND
RECOMMENDED HOLDING TIMES FOR SEDIMENT SAMPLES
Variables
Semi volatile
organ 1cs
Metal s
Minimum Sample Preservation
Sample S1zea Container and Handling
50-100 g 16-oz glass jar. Cool (4° C),
[20g] PTFEc-lined lid or Freeze
50 g 8-oz linear poly- Cool (4° C),
[lOg] ethyl ene or boro- or Freeze
silicate glass,
PTFE-llned I1d
Holding
Times"
7 days/4Q days
1 yra
6 mo (Hg 28 days)
6 mo (Hg 28 days)d
Total sol Ids,
Total organic
carbon
011 and grease
75 g 8-oz glass or
[20g] polyethylene jar
100 g 4-oz glass Jar,
[50g] PTFE-llned lid
Freeze
Cool (4° C),
HC1, or
Freeze
6ma°
28 daysd
6 mo0
Particle size
100-150 ge
[lOOg]
SI ass or plastic Cool (4° C)
jar, or scalable
plastic bag
(approx. 16-oz)
8 mod
a The minimum sample size (wet-weight) presented is for one laboratory analysis. If
additional laboratory analyses are required (e.g., replicates), the sample size should
be adjusted accordingly. Because it may be difficult to correct an adequate amount of
sediment from storm drains, the absolute minimum analytical sample size 1s provided In
brackets.
" Where two times are given, the first refers to the maximum time prior to extraction,
the second to the maximum time prior to Instrumental analysis. U.S. EPA has not
established holding times for sediment samples, however, the holding times for water
samples should be met to help ensure the sample integrity,
0 PTFE - Polytetrafluoroethylene.
" This 1s a suggested holding time.
of this variable.
No U.S. EPA criteria exist for the preservation
8 Large-grain size samples (I.e., sand) require a larger sample size than sllty
sampl es.
34
-------
collected will be sufficient for the requested analysis and any QA/QC
samples.
Prepared sample containers can be obtained through commercial sources
or from the laboratory performing the analyses. Sample containers should be
documented as to cleanliness by the supplier, or container blanks should be
analyzed to provide necessary documentation. The preservation and handling
procedures can be met for the majority of variables by placing samples on
ice following collection, and then transferring the samples to a freezer as
soon as possible. Freezing of samples will require that sample containers
have adequate headspaee for the expansion of pore water. Because pore
water expands, containers for samples that will be frozen should only be
filled three quarters full. If oil and grease samples cannot be analyzed
within 24 h, concentrated hydrochloric acid should be added at approximately
1 mL/80 g of sediment. The container should be sealed and inverted several
times to mix the acid and sediment.
Holding times for sediment samples have not been established by
U.S. EPA. The holding times cited for frozen samples are those recommended
under PSEP protocols (Tetra Tech 1986d). The recommended holding times for
unfrozen sediment cited in PSEP were based on U.S. EPA holding times for
water samples (U.S. EPA 1987). Extract holding times (i.e., the time from
extraction of a sample until analysis) of 40 days have been established for
water samples and have also been recommended for extractable organic
compounds in sediment (Tetra Tech 1986d).
4.2.6 Quality Assurance/Quality Control
Quality assurance (QA) is the program for assuring reliability of
sampling procedures and analytical measurements. Quality control (QC) is the
routine application of procedures by the analytical lab, such as periodic
instrument calibration, to obtain prescribed standards of performance in
monitoring and measurement. The integration of QA/QC into sample collection,
analysis, and data reporting procedures is important for generating reliable
data. When QA/QC procedures are defined at the inception of a project and
35
-------
adhered to during performance of the project, comparison of the procedures
and results with QA/QC goals can be made to determine data reliability.
Sampling programs regulated by U.S. EPA and Ecology require preparation
of a Quality Assurance Project Plan (QAPP). The QAPP details sampling and
analysis procedures, data quality objectives (i.e., precision, accuracy, and
completeness), and other procedures necessary for obtaining reliable data.
Guidelines have been published (U.S. EPA 1983a) that describe the required
elements of a QAPP. Additional guidelines on field QA/QC can be found in
the PSEP protocols (Tetra Tech 1986d) and from U.S. EPA (1986b). Guidelines
on laboratory QA/QC procedures can be found in the method references (U.S.
EPA 1984, 1987) and in the PSEP protocols (Tetra Tech 1986d).
QA/QC for sediment samples collected in the field include the following:
Field replicates and blind analytical replicates
Field rinsate blanks (i.e., field decontamination blank
plus field transport blank)
Standard reference materials (SRMs).
Field replicate samples are used to determine total variability (i.e.,
field variability plus analytical variability). To collect field replicate
samples two separate sets of samples are collected at a single station and
sediments from each set are not homogenized together. Collection of field
replicate samples may not be feasible in some manholes because of insuffi-
cient sediment deposits. To collect blind analytical replicates, a volume
of sediment sufficient for two or more sets of samples is collected,
thoroughly homogenized, and individual aliquots are placed in separate
sample containers. Field and analytical replicates should be labeled
consistently with other samples and submitted blind to the laboratory (i.e.,
the laboratory should not know the samples are replicates of each other).
One set of blind analytical replicate samples can also be analyzed by a
different laboratory to evaluate analytical variability between laboratories.
36
-------
Field rinsate blanks are used to assess potential contamination during
equipment decontamination procedures and sample collection, shipping, and
storage. A field rinsate blank is collected by pouring analyte-free (i.e.,
deionized and distilled) water through the appropriate sampling device and
collecting the rinsate. The field rinsate blank should be collected
following sample collection and decontamination of sampling equipment. The
field rinsate blank serves to check effectiveness of decontamination
procedures. After collection of the field rinsate blank, the sample is
shipped and stored in the same manner as all other field samples. Therefore,
the field rinsate blank also serves to identify whether sample contamination
occurred from field sources, or during shipping or storage. The analyte-
free water should be stored in a sample container that accompanies sample
containers and samples until final delivery to the laboratory.
Analytical results from field rinsate samples will not allow for
differences between contamination that occurred because of ineffective
decontamination procedures, and contamination that occurred during collec-
tion, shipping, or storage procedures. To differentiate between those
possible contaminant sources, separate field decontamination blanks and
field transport blanks would need to be collected and analyzed. Because of
the additional associated costs, the decision to collect both field
decontamination and transport blanks should be made by the project officer
and will be dependent on project objectives. One option would be to collect
separate decontamination and transport blanks, and archive those samples for
analysis only if contaminants were detected in the field rinsate sample.
The frequency for collecting field rinsate blanks should be determined
by the project manager before beginning the project. For the majority of
field sampling efforts, one field rinsate blank per day should be collected.
Collecting and analyzing field rinsate blanks can add considerable cost to
projects. To help minimize costs, collection and analysis of one field
rinsate blank and archiving subsequent field rinsate blanks is recommended.
If problems with contamination are noted in the initial field rinsate blank,
additional field rinsate blank analyses should be conducted on archived
samples. Maximum holding times for analyses of archived samples should not
be exceeded.
37
-------
SRMs are used to assess the accuracy of analysis. SRMs, usually
provided through a government agency, have been sufficiently characterized
for one or more analytes such that certified values are provided. SRMs are
submitted as a sample to the laboratory for analysis.
Northwest National Oceanographic and Atmospheric Administration/National
Marine Fisheries Service (NOAA/NMFS) has prepared a marine sediment sample
SRM with PCBs, polynuclear aromatic hydrocarbons (PAH), and selected
pesticides. The NOAA/NMFS SRMs are available from the U.S. EPA Office of
Puget Sound. SRMs are also available from the National Bureau of Standards
(NBS) and the Marine Analytical Chemistry Standards Program of the National
Research Council of Canada. An estuarine sediment sample containing trace
metals is currently available, and SRMs with PCBs and organic compounds in
marine sediments are currently in preparation. The recommended frequency
and evaluation procedures for SRM analysis are discussed in the PSEP
protocols (Tetra Tech 1986d).
Laboratory QA/QC is performed by the analytical laboratory. A
discussion of laboratory QA/QC requirements and the required minimum
frequency of analysis is presented in the PSEP protocols (Tetra Tech 1986d),
and the U.S. EPA Contract Laboratory Program (CLP) statement of work (U.S.
EPA 1987). Prior to initiation of sampling efforts, the project manager
should specify the frequency of analysis for laboratory QA/QC samples (i.e.,
method blanks, matrix spikes, method spikes, and analytical replicates).
Technical evaluation of the data should be performed by an expert, and the
results of all QA/QC analyses should be reported with the sample data.
4.3 IDENTIFYING AND RANKING PROBLEM STORM DRAINS
Problem (i.e., contaminated) storm drains will be identified based on
the in-line sediment chemistry measured during the initial screening. All
problem drains will be included in Phase II of the sampling program. More
intensive Phase II sampling is recommended for the high priority storm
drains to trace contaminants so that ultimate sources can be identified. In
addition, a ranking procedure has been developed to prioritize individual
38
-------
problem storm drains to aid in scheduling Phase II. A schematic of the
decision criteria recommended for identifying high priority storm drains is
presented in Figure 7.
4.3.1 Evaluating Sediment Data
Although there are no specific criteria for storm drain sediments,
several approaches can be used to identify and eventually rank problem storm
drains based on contaminant levels. One approach to identifying problem
storm drains involves comparing storm drain sediment data with available
criteria for sediments in the receiving environment. Criteria for receiving
environment sediments have recently been proposed for freshwater and marine
sediments. Apparent Effects Threshold (AET) values are applicable to marine
sediments (Tetra Tech 1986b) in Puget Sound, and screening level concentra-
tions (Neff et al. 1986) and interim criteria (Wisconsin Department of
Natural Resources 1985) are applicable to freshwater sediments. When no
sediment criteria are available for specific contaminants, problem storm
drains can be selected based on sediment contamination ranking in the 90th
percentile of contaminant concentration measured for all storm drain
sediment data. These criteria help in identification of problem storm drain
systems based on contamination of storm drain sediments.
AET values have been proposed for the Puget Sound Dredged Disposal
Analysis Program (Tetra Tech 1986b) and have recently been updated to include
new environmental data sets (Tetra Tech 1987). The focus of the AET
approach is to identify concentrations of chemical contaminants in sediments
that are associated with statistically significant biological effects
(relative to reference conditions). Biological indicators used to develop
AET values include:
Depression in abundances of major taxonomic groups of benthic
infauna (e.g., Crustacea, Mollusca, Polychaeta)
Amphipod mortality bioassay using Rhepoxvnius abronius
39
-------
IDENTIFY PROBLEM CHEMICALS
IN STORM DRAINS
ARiAETVALUESOR OTHER
SEDIMENT CRITERIA AVAILABLE?
(NEFFetal. 1986;
WISCONSIN DNR1985)
YES
NO
DO CONCENTRATIONS IN STORM
DRAIN SEDIMENTS EXCEED AET
OR OTHER SEDIMENT CRITERIA?
DO CONCENTRATIONS IN STORM
DRAIN SEDIMENTS RANK IN
90th PERCENTILE?
YES
DO CONCENTBATIONS IN STORM
DRAIN SEDIMENTS EXCEED
STREET DUST LEVELS?
YES
YES
YES
ARECHEMCALSON
POLLUTANT-OF-CONCERN LIST?
NO
RANK PROBLEM DRAINS FOR
CHEMICALS EXCEEDING AET
AND 90lh PERCENTILE
CALCULATE EAR VALUES
CALCULATE LOADING INDEX
I
SELECT HIGHEST PRIORITY
DRAINS FOR CONTAMINANT
TRACING SAMPLING
Figure 7. Decision criteria for selecting problem chemicals and
ranking problem storm drains.
40
-------
Oyster larvae abnormality bioassay using Crassostrea qiqas
Microtox bioluminescence bioassay using Photobacterium
phosphoreum.
For a given chemical and a specific biological indicator, the AET is the
concentration above which statistically significant biological effects
occurred in all samples of sediments analyzed.
AET values have been proposed for 64 organic and inorganic toxic
chemicals using synoptic chemical and biological data from 200 stations in
Puget Sound (Tetra Tech 1987). For each chemical, a separate AET was
developed for each biological indicator listed above, resulting in four sets
of AET values. A list of the highest (HAET) and lowest AET (LAET) for each
chemical is provided in Table 6.
Because the AET approach was originally developed for marine sediments,
it is not directly applicable to storm drain sediments. However, because
there are no specific criteria yet available for storm drains, the AET
approach is recommended as a conservative approach for evaluating contamina-
tion in storm drain sediments. The range of available AET values (Tetra
Tech 1987) are listed in Table 6. It is expected that many chemicals
present in storm drain sediments would exceed the lowest AET value.
Therefore, the highest AET value will be used to identify problem concentra-
tions of chemicals in storm drain sediments because it represents a less
stringent and more practical criteria for evaluating contamination in storm
drains. A problem storm drain, in this case, is defined as having at least
one chemical in the in-line sediments with a measured concentration exceeding
the highest AET value.
Because the AET values have been developed specifically for marine
sediments, alternate sediment criteria are needed for storm drains that
discharge into freshwater environments. Few criteria have been developed
for freshwater sediments. Interim criteria have been proposed for PCBs and
certain metals (i.e., arsenic, cadmium, chromium, copper, lead, mercury,
nickel, and zinc) by the Wisconsin Department of Natural Resources (1985).
41
-------
TABLE 6. PUGET SOUND AET VALUES
(ug/kg dry weight = ppb for organic compounds;
mg/kg dry weight = ppm for metals)
LPAHa
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
HPAHb
Fluoranthene
Pyrene
Benzo(a) anthracene
Chrysene
Benzof 1 uoranthenes
Benzo(a)pyrene
IndenoCl.ZjS-CjdJpyrene
Dibenzo(a,h) anthracene
Benzo (g , h , i ) pery 1 ene
Total PCBs
Total Chlorinated Benzenes
1 , 3-Di chl orobenzene
1,4-Di chl orobenzene
1 , 2-Di chl orobenzene
1 , 2 , 4-Tri ch 1 orobenzene
Hexachl orobenzene
Total Phthalates
Dimethyl phthalate
Di ethyl phthalate
Di-n-butyl phthalate
Butyl benzyl phthalate
Bis(2-ethylhexyl) phthalate
Pesticides
4,4'-DDE
4,4' -ODD
4,4'-DDT
Lowest AET
5,200
2,100
560
500
540
1,500
960
12,000
1,700
2,600
1,300
1,400
3,200
1,600
600
230
670
130
170
..
110
35
31
70
3,300
71
--
1,400
63
1,900
9
2
3.9
Highest AET
6,100
2,400
640
980
1,800
5,400
1,900
38,000
9,800
11,000
4,500
6,700
8,000
6,800
880
1,200
5,400
2,500
680
..
260
50
64
230
3,400
160
200
1,400
470
1,900
15
43
11
42
-------
TABLE 6. (Continued)
Lowest AET Highest AET
Phenols
Phenol 420 1,200
2-Methylphenol 63 63
4-Methylphenol 670 1,200
2,4-Dimethyl phenol 29 29
Pentachlorophenol
2-Methoxyphenol 930 930
Miseellaneous Extractables
Hexachlorobutadiene
1-Methylphenanthrene
2-Methyl naphtha! ene
Biphenyl
Di benzothi ophene
Dibenzofuran
Benzyl alcohol
Benzoic acid
n-Nitrosodiphenylamine
Volatile Oraanic Compounds
Tetrach 1 oroethene
Ethyl benzene
Total xylenes
Metals
Antimony
Arsenic
Cadmium
Copper
Lead
Mercury
Nickel
Silver
Zinc
120
310
670
260
240
540
57
650
40
140
33
100
3.2
85
5.8
310
300
0.41
28
5.2
260
290
370
670
270
250
540
73
650
220
140
37
120
26
700
9.6
800
700
2.1
49
5.2
1,600
a LPAH = Low molecular weight polynuclear aromatic hydrocarbons.
b HPAH = High molecular weight polynuclear aromatic hydrocarbons.
Reference: Tetra Tech (1987).
43
-------
These criteria were developed to assess the suitability for open-water
disposal of sediments dredged from the Great Lakes. The criteria are based
on comparisons of current and historical sediment toxicant concentrations
measured in the Great Lakes. In addition to the interim criteria, screening
level concentrations (SLC) have been proposed by U.S. EPA for PCBs, chlor-
dane, dieldrin, heptachlor epoxide, and DDT in freshwater sediment (Neff et
al. 1986). The SLC approach uses synoptic field data on co-occurrence in
sediments of benthic infaunal invertebrates and different concentrations of
each organic contaminant. Each SLC is a conservative estimate of the highest
organic carbon normalized concentration of a specific contaminant in
sediment that can be tolerated by approximately 95 percent of benthic
infauna. Available freshwater sediment criteria are summarized in Table 7.
It is recommended that these values be used to evaluate storm drain sediment
contaminant levels for drains discharging into a freshwater environment.
AET values and freshwater criteria have not been developed for many
contaminants associated with storm drain sediments and discharges.
Therefore, the 90th percentile concentration of a chemical (i.e., the
concentration above which 10 percent of the observations fall) is recommended
to evaluate contamination levels in storm drain sediments. Using this
method, storm drains with sediments having a chemical concentration exceeding
the 90th percentile concentration in all storm drain sediments measured will
be identified as a problem drain.
Results from the initial screening should also be compared with the
data for normal urban street dust. Street dust has been identified as the
primary source of suspended particulates in urban runoff (Galvin and Moore
1982), and therefore is directly associated with storm drain sediment
accumulations. Street dust values may be more representative of general
background contaminant levels in storm drain sediments than AET values or
90th percent!le ranking. Representative street dust contaminant levels for
urban areas are presented in Tables 8 and 9. Phthalates and PAH are the
only contaminants whose average concentration in urban street dust samples
from Seattle and Bellevue exceeded the highest AET values. This suggests
that, under normal background conditions, storm drain sediments can be
expected to exceed AET criteria. This exceedance indicates that AET values
44
-------
TABLE 7. FRESHWATER SEDIMENT CRITERIA
Metals (mg/kg dry weight)3
Arsenic
Cadmi urn
Chromium
Copper
Lead
Mercury
Nickel
Zinc
10
1.0
100
100
50
0.10
100
100
Organic Compounds (ug/kg dry weight)
Heptachlor epoxide 8
Chlordane 9.8
Dieldrin 21
PCBs 290
DDT 190
a Interim criteria for open-water disposal of dredged
materials (Wisconsin Department of Natural Resources
1985). If concentration in dredged materials exceeds
125 percent of the interim criteria value, then sediment
cannot be disposed of in open water.
b Estimated highest concentration in the sediment that
can be tolerated by approximately 95 percent of benthic
infauna (Neff et al. 1986). Values are based on the
organic carbon normalized concentration in the sediment.
45
-------
TABLE 8. SUMMARY OF METALS MEASURED IN STREET DUST SAMPLES
COLLECTED FROM SEATTLE AND BELLEVUEa
Chemical
Antimony
Arsenic
Beryllium
Cadmium
Chromium
Copper
Lead
Mercury
Nickel
Selenium
Silver
Thallium
Zinc
Range
(mg/kg dry wt)
<1-2.0
11-39
0.17-0.34
0.6-2.0
20-230
31-260
90-1300
0.02-0.18
20-44
<0.6-<3
0.01-0.5
<0.2-0.34
110-970
Meanb
(mg/kg dry wt)
1.1
25
0.26
1.0
97
93
520
0.07
32
2
0.32
0.6
310
Detection
Frequency
8/12
12/12
12/12
12/12
12/12
12/12
12/12
9/12
12/12
0/12
6/12
3/12
12/12
a Street dust sampled collected from five residential areas and three
suburban arterials in Bellevue; two industrial and two commercial areas in
Seattle.
b Mean calculated using the reported detection limit for undetected values.
Reference: Galvin and Moore (1982).
46
-------
TABLE 9. SUMMARY OF ORGANIC COMPOUNDS
MEASURED IN STREET DUST SAMPLES COLLECTED FROM
SEATTLE AND BELLEVUE3
Chemical
Pesticides
al pha-Hexach 1 orocycl ohexane
gamma-Hexach 1 orocycl ohexane
DDD
Heptachlor
Hal oqenated Al Iphat i cs
Trichloromethane
Tetrachloroethane
1,1, 1-Trichloroethane
4-Chlorophenyl phenyl ether
Monocvclic Aromatic Hydrocarbons
Benzene
Hexachl orobenzene
Ethyl benzene
Toluene
Nitrosodimethylamine
Phenols
Phenol
Pentachlorophenol
2, 4-Dimethyl phenol
4-Nitrophenol
Phthalates
Dimethyl phthalate
Di ethyl phthalate
Di-n-butyl phthalate
Di-n-octyl phthalate
Butyl benzyl phthalate
Bis-2-ethylhexyl phthalate
Meanb
(mg/kg)
0.014
0.025
0.005
0.048
0.007
0.024
0.013
0.24
0.021
2.0
0.021
0.009
0.76
0.21
1.76
0.02
0.11
0.78
0.41
0.70
0.54
6.2
38
Range*3
(rag/kg)
0.010-0.018
0.006-0.043
0.005
0.048
0.004-0.015
0.016-0.032
0.012-0.016
0.24
0.01-0.032
2.0
0.005-0.025
0.004-0.019
0.76
0.08-0.47
0.12-3.4
0.01-0.03
0.11
0.78
0.16-0.89
0.22-2.4
0.23-0.97
0.22-0.35
2.4-90
Detection
Frequency
2/14
2/14
1/14
1/14
4/14
2/14
3/14
1/14
2/14
1/14
3/14
4/14
1/14
4/14
2/14
2/14
1/14
1/14
3/14
7/14
4/14
7/14
9/14
47
-------
TABLE 9. (Continued)
Chemical
Meanb
(mg/kg)
Range*1
(mg/kg)
Detection
Frequency
LPAHC
Acenaphthylene 0.21 0.16-0.25 2/14
Anthracene 0.35 0.1-0.6 5/14
Fluorene 0.23 0.2-0.25 2/14
Phenanthrene 1.5 0.18-2.4 14/14
HPAHd
Fluoranthene
Pyrene
Chrysene
Benzo(a
Benzofk
Benzol a
pyrene
fluoranthene
anthracene
1.7
1.7
1.04
0.63
1.1
0.63
0.36-2.6
0.32-2.5
0.11-2.0
0.08-0.90
0.27-1.5
0.20-0.85
14/14
13/14
11/14
7/14
8/14
7/14
a Street dust samples were collected from five residential areas and three
suburban arterials in Bellevue, and from two industrial and two commercial
areas in Seattle.
b Calculation based on detected values only.
c LPAH = Low molecular weight polynuclear aromatic hydrocarbons.
d HPAH = High molecular weight polynuclear aromatic hydrocarbons.
Reference: Galvin and Moore (1982).
48
-------
may be too stringent for these contaminants. Therefore, it is recommended
that the average urban street dust concentrations, rather than the AET
values, be used to assess phthalates and PAH during the initial screening
phase.
Storm drains with chemical concentrations measured in the in-line
sediments that exceed the AET value or that rank in the 90th percentile
should be considered for additional sampling during Phase II contaminant
tracing (see Figure 7). If chemicals in these drains are on the pollutant of
concern list (Appendix C), additional sampling under Phase II is recom-
mended. Pollutants of concern are chemicals that have been identified as
potential problems in the Puget Sound receiving environment based on
consideration of chemical toxicity, environmental persistence, bioaccumula-
tion potential, high concentration in the water column, existence of known
sources, high concentration in offshore sediments relative to reference
area conditions, or widespread distribution in Puget Sound. If the chemical
is not on the pollutant of concern list but exceeds normal urban street dust
values, it should be considered in the ranking process. Additional sampling
will consist of collecting in-line sediment samples from selected manholes
to trace contaminants throughout the system and to isolate specific
contaminated sections of the storm drain lines.
4.3.2 Ranking Procedure
A ranking procedure is provided to help prioritize problem storm drains
(Figure 7). It is expected that cities may be unable to conduct intensive
contaminant tracing sampling activities in all problem drains at once
because of limitations in available funding. Therefore, the ranking
procedure is provided to aid in scheduling the Phase II contaminant tracing
program so that the highest priority drains can be investigated as soon as
possible. Two methods, elevation above reference (EAR) and loading indices,
are recommended for ranking problem drains.
The elevation above reference (EAR) technique is a comparison of storm
drain sediment data with data available for offshore receiving environment
sediments. Sediment quality data are available for 10 reference areas in
49
-------
Puget Sound. These data (Tables 10 and 11) are assumed to provide a
reasonable measure of the variability in contaminant concentrations for
relatively uncontauiinated sediments, but are expected to represent fairly
conservative levels of contaminant concentrations for storm drain sediments.
In previous Puget Sound studies (Tetra Tech 1985a,c,d), EAR values were
calculated based only on six Carr Inlet reference stations (Tables 12 and
13). Only the Carr Inlet data, rather than the full range of Puget Sound
reference area data, are recommended for ranking storm drain data for the
following reasons:
The most complete reference data set is available for Carr
Inlet and includes synoptic data for metals, organic com-
pounds, grain size, organic carbon, and other conventional
variables
The lowest reference detection limits for most substances of
concern in Puget Sound embayments are available for Carr Inlet
EAR values for many urban embayments in Puget Sound (e.g.
Commencement Bay, Elliott Bay, and Everett Harbor) have been
calculated with these data, so direct comparisons with
previous investigations are possible
Where chemicals were detected in more than one reference
area, the Carr Inlet samples usually had comparable or lower
values and on this basis appear to be reasonably representa-
tive of Puget Sound reference conditions.
EAR values for each problem storm drain are calculated by dividing the
measured concentration of a contaminant by the reference concentration. It
is recognized that concentrations of chemical contaminants in storm drain
sediments will generally exceed reference concentrations. Therefore, storm
drains will be ranked based on the magnitude of exceedance of reference
conditions for each problem chemical.
50
-------
TABLE 10. SUMMARY OF METAL CONCENTRATIONS IN SEDIMENTS
FROM PUGET SOUND REFERENCE AREAS
Chemical
Antimony
Arsenic
Cadmi urn
Chromium
Copper
Lead
Mercury
Nickel
Selenium
Silver
Zinc
Range
(mg/kg dry wt)
UQ.lb-2.79
1.9-17
0.1-1.9
9.6-255
5-74
UO.1-24
0.01-0.28
4-140
UO. 1-1.0
UO.02-3.3
15-102
Detection
Frequency
16/36
38/38
28/28
42/42
32/32
25/32
42/42
30/30
18/28
28/30
30/30
Reference
Sites3
1,2,3,4,7,8,9,10
1,2,3,4,7,8,9,10
1,2,3,4,6,9,10
1-10
1,2,3,4,5,6,9,10
1,2,3,4,5,6,9,10
1-10
1,2,3,4,5,9,10
1,2,3,4,6,9,10
1,2,3,4,5,9,10
1,2,3,4,5,9,10
a Reference sites:
1. Carr Inlet
2. Samish Bay
3. Dabob Bay
4. Case Inlet 7. Nisqually Delta
5. Port Madison 8. Hood Canal
6. Port Susan 9. Sequim Bay
10. Port Susan.
b U = Undetected at the method detection limit shown.
References:
Site
Site
Site
Site
Site
Site
Site
Site 8
(Site 9)
Tetra Tech (1985b); Crecelius et al. (1975).
Battelle (1985).
Battelle (1985).
Crecelius et al. (1975); Mai ins et al. (1980).
(1980).
(1982).
al. (1975).
al. (1975),
Mai ins et al
Mai ins et al
Crecelius et
Crecelius et
Battelle (1985).
(Site 10) Tetra Tech (unpublished).
51
-------
TABLE 11. SUMMARY OF ORGANIC COMPOUND CONCENTRATIONS
IN SEDIMENTS FROM PUSET SOUND REFERENCE AREAS
Substance3
Phenols
65 Phenol
HSL 2 -Methyl phenol
HSL 4-Methyl phenol
34 2, 4-Dimethyl phenol
Substituted Phenols
24 2-Chlorophenol
31 2,4-Dichlorophenol
22 4-Chloro-3-methylphenol
21 2,4,6-Trichlorophenol
HSL 2,4,5-Trichlorophenol
64 Pentachlorophenol
57 2-Nitrophenol
59 2,4-Dinitrophenol
60 4,6-Dinitro-o-cresol
58 4-Nitrophenol
Low Molecular Weiaht Polvnuclear
55 Naphthalene
77 Acenaphthylene
1 Acenaphthene
80 Fluorene
81 Phenanthrene
78 Anthracene
HSL 2-Methyl naphthalene
Hiah Molecular Weiaht Polvnuclear
39 Fluroanthene
84 Pyrene
72 Benzo (a) anthracene
76 Chrysene
74 Benzo(b)fluoranthene
75 Benzo (k)fluoranthene
Range
(ug/kg
dry wt)
U3,3-62c«d
U10
U2-290
U1-U14
U0.5-U500
UO.5-50
UO.5-50
U0.5-U100
U10-U10Q
0.1-U1000
0.1-U50
U0.5-U50
UQ.5-U1QO
U0.5-U100
Detection
Frequency
8/20
0/11
7/11
0/13
0/13
0/13
0/13
0/13
0/11
0/13
1/9
0/9
0/9
0/9
Reference
Sitesb
1,2,3,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
Aromatic Hydrocarbons
U0.5-U40
U0.1-U40
U0.1-U40
U0.1-U40
4-170
U0.5-U40
0.3-U22
14/27
2/27
4/27
7/28
18/24
11/24
10/17
1-6,10
1-6,10
1-6,10
1-7,10
1,2,3,6,7,10
1,2,3,6,7,10
1,4,5,6,10
Aromatic Hydrocarbons
5-100
5-120
2-U40
4-U40
U5-94
4.8-94
24/29
23/29
15/22
15/22
15/25
15/25
1-7,10
1-7,10
1,2,3,6,7,10
1,2,3,6,7,10
1-7,10
1-7,10
52
-------
TABLE 11. (Continued)
Substance4
Hiah Molecular Weiaht Polvnuclear
73 Benzo(a)pyrene
83 Indeno(l,2,3-c,d)pyrene
82 Dibenzo(a,h) anthracene
79 BenzoCgjhJjperylene
Chlorinated Aromatic Hydrocarbons
26 1,3-Dichlorobenzene
27 1,4-Dichlorobenzene
25 1,2-Dichlorobenzene
8 1,2, 4-TH chlorobenzene
20 2-Chloronaphthalene
9 Hexachlorobenzene (HCB)
Chlorinated Aliphatic Hydrocarbons
12 Hexachloroethane
xx Trichlorobutadiene
xx Tetrachlorobutadiene isoiers
xx Pentachlorobutadiene isomers
52 Hexachlorobutadiene
53 Hexachlorocyclopentadiene
Halogenated Ethers
18 Bis(2-chloroethyl) ether
42 Bis(2-chloroisopropyl) ether
43 Bis(2-chloroethoxy)methane
40 4-Chlorophenyl phenyl ether
41 4-Bromophenyl phenyl ether
Phthalates
71 Dimethyl phthalate
70 Di ethyl phthalate
68 Di-n-butyl phthalate
67 Butyl benzyl phthalate
66 Bis(2-ethylhexyl) phthalate
69 Di-n-octyl phthalate
Range
(ug/kg
dry wt)
Detection
Frequency
Reference
Sitesb
Aromatic Hydrocarbons (Continued)
UO.37-40
UO. 37-30
0.4-U13
1.2-20
U0.06-U160
U0.06-U120
UO. 06-65
U0.5-U190
U0.5-U50
0.01-U100
U0.5-U.50
U0.03-U25
U0.03-U25
U0.03-U25
U0.03-U25
U200
0.3-U20
U0.5-U10
U10
U0.5-U10
U0.5-U10
U0.5-U50
9.0-11
U20-760
U0.5-U25
UO.5-58
U0.5-U56
16/21
10/19
3/12
8/13
1/25
1/25
1/25
0/13
0/13
6/19
0/9
5/12
5/12
5/19
0.07-8.5
0/3
1/9
0/9
0/9
0/9
0/9
1/12
4/8
6/8
3/12
3/8
5/12
1,3,4,5,6,7,10
1,4,5,6,7,10
1,10
1,7,10
1,2,3,4,5,10
1,2,3,4,5,10
1,2,3,4,5,10
1,10
1,10
1,4,5,6,10
1,10
1,4,5,6
1,4,5,6
1,4,5,6,10
1,4,5,6
10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
53
-------
TABLE 11. (Continued)
Range
(ug/kq Detection Reference
Substance3 dry wt) Frequency Sites11
Miscellaneous Oxygenated Compounds
54 Isophorone U0.5-U130 0/12 1,10
HSL Benzyl alcohol U10-U340 0/11 1,10
HSL Benzoic acid U-430 4/11 1,10
129 2,3,7,8-Tetrachlorodibenzo-p-
dioxin Not Analyzed
HSL Dibenzofuran U5-14 4/11 1,10
Orqanom'trooen Compounds
HSL Aniline
56 Nitrobenzene
63 n-Nitroso-di-n-propylamine
HSL 4-Chloroaniline
HSL 2-Nitroaniline
HSL 3-Nitroaniline
HSL 4-Nitroaniline
36 2,6-Dinitrotoluene
35 2,4-Dinitrotoluene
62 n-Nitrosodiphenylamine
37 1,2-Diphenylhydrazine
5 Benzidine (4l4'-diaminobiphenyl)
28 3,3'-Dich1orobenzidine
Pesticides
93 p,p'-DDE
94 p,p'-DDD
92 p,p'-DDT
89 Aldrin
90 Dieldrin
91 Chlordane
95 alpha-Endosulfan
96 beta-Endosulfan
97 Endosulfan sulfate
98 Endrin
99 Endrin aldehyde
100 Heptachlor
101 Heptachlor epoxide
102 alpha-HCH
103 beta-HCH
U1.0-U20
U0.5-U10
U0.5-U10
U10-U50
U10-U50
U50
U50
U0.5-U10
U.05-U10
U0.5-U10
U0.5-U5
U0.5
UG.5-U1QQ
U1.0-U25
U1.0-U25
U1.0-U25
U0.5-U25
U1.0-U25
U5.0-U50
U.5-U25
U1.0-U25
U1.0-U25
U1.0-U25
U2.3-U25
U0.5-U50
U0.5-U25
U0.5-U50
U0.5-U5Q
0/6
0/8
0/8
0/7
0/7
0/7
0/7
0/8
0/8
0/8
0/6
0/2
0/9
0/12
0/13
0/12
0/13
0/13
0/13
0/8
0/8
0/8
0/13
0/9
0/13
0/9
0/13
0/13
1
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
. 1
1
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
1,10
54
-------
TABLE 11. (Continued)
Substance3
Range
(ug/kg
dry wt)
Detection
Frequency
Reference
Sitesb
Pesticides (Continued)
104 delta-HCH
105 gamma-HCH (Lindane)
113 Toxaphene
PCBs
xx Total PCBs (primarily
1254/1260)
Volatile Organic Compounds
85 Tetrachloroethene
38 Ethyl benzene
U0.5-U25
U0.5-U50
U10-U100
3.1-U170
U3-U16
U3-U16
0/13
0/13
0/5
7/26
0/11
0/11
1,10
1,10
1,10
1,2,3,
4,6,7,10
2,3,10
2,3,10
a Number indicates U.S.
Hazardous Substance List.
EPA priority pollutant number. HSL indicates
Reference sites:
1.
2.
3.
Carr Inlet
Samish Bay
Dabob Bay
4. Case Inlet ' 7.
5. Port Madison 10.
6. Port Susan
Nisqually Delta
Port Susan.
c An anomalously high phenol value of 1,800 ug/kg dry weight was found at one
station. For the purposes of developing reference area concentrations, the
value has been excluded.
d U = Undetected at the method detection limit shown.
Reference:
Site 1
Site 2
Site 3
Site 4
Site 5
Site 6
Site 7
Tetra Tech (1985b); Mowrer et al. (1977).
Battelle (1985).
Battelle (1985); Prahl and Carpenter (1979).
Mai ins et al.
Mai ins et al.
Mai ins et al.
1980); Mowrer et al. (1977).
1980 .
1982).
Barrick and Prahl (1987); Mowrer et al. (1977).
Site 10) Tetra Tech (unpublished).
55
-------
TABLE 12. SUMMARY OF METAL CONCENTRATIONS IN
SEDIMENTS FROM CARR INLET REFERENCE AREA
Chemical
Antimony
Arsenic
Cadmi urn
Chromium
Copper
Lead
Mercury
Nickel
Selenium
Silver
Zinc
Range
(mg/kg dry wt)
UO.1-0.14
2.4-3.8
0.29-1.5
9.6-24.4
4.9-8.0
4.4-13
0.01-0.098
11-27.6
U0.1-U1
0.02-0.12
15-24.1
Mean3
(mg/kg dry wt)
0.11
3.4
0.95
15
6.4
9.2
0.04
17
0.7
0.09
19
Detection
Frequency
4/6
6/6
6/6
6/6
6/6
6/6
6/6
6/6
0/6
2/6
6/6
a Mean calculated using the reported detection limit for undetected values,
Reference: Tetra Tech (1985b).
56
-------
TABLE 13. SUMMARY OF ORGANIC COMPOUND CONCENTRATIONS
IN SEDIMENTS FROM CARR INLET REFERENCE AREA
Substance3
Phenols
65 Phenol
HSL 2-Methyl phenol
HSL 4-Methyl phenol
34 2,4-Dimethylphenol
Substituted Phenols
24 2-Chlorophenol
31 2,4-Dichlorophenol
22 4-Chloro-3 -methyl phenol
21 2,4,6-Trichlorophenol
HSL 2,4,5-Trichlorophenol
64 Pentachlorophenol
57 2-Nitrophenol
59 2,4-Dinitrophenol
60 4,6-Dinitro-o-cresol
58 4-Nitrophenol
Low Molecular Weiaht Polvnuclear
55 Naphthalene
77 Acenaphthylene
1 Acenaphthene
80 Fluorene
81 Phenanthrene
78 Anthracene
HSL 2-Methyl naphthalene
Hi ah Molecular Weiaht Polvnuclear
39 Fluroanthene
84 Pyrene
72 Benzo (a) anthracene
76 Chrysene
74 Benzo(b)fluoranthene
75 Benzo (k)fluoranthene
Range
(ug/kg
dry wt)
U10-62c'd
U1-U10
U10-32
U1-U10
U0.5-U5
U0.5-U10
U0.5-U10
U0.5-U10
U10
0.1-U50
0.1-U10
U0.5
U0.5-U100
U0.5-U100
Meanb
(ug/kg
dry wt)
33
7,0
13
6.8
3.5
6.8
6.8
6.8
10
33
6.8
0.5
67
67
Detection
Frequency
3/13
0/6
2/6
0/6
0/6
0/6
0/6
0/6
0/4
1/6
1/6
0/2
0/6
0/6
Aromatic Hydrocarbons
1-13
U0.5-U5
U0.5-U5
U0.5-U5
5-16
3-22
U1-U5
6.8
4.1
4.1
4.1
13
9.1
4.2
3/5
0/5
0/5
0/5
5/5
4/5
0/5
Aromatic Hydrocarbons
11-20
11-18
U5-8
U5-19
3
5
15.4
14.4
8.0
10.8
3
5
5/5
5/5
4/5
4/5
1/1
1/1
57
-------
TABLE 13. (Continued)
Substance3
Hiah Molecular Weiaht Polvnuelear
73 Benzo(a)pyrene
83 Indeno(l(2,3-c,d)pyrene
82 Dibenzo(a,h)anthracene
79 Benzo(g,h,i)perylene
Chlorinated Aromatic Hydrocarbons
26 1,3-Dichlorobenzene
27 1,4-Diehlorobenzene
25 1,2-Dichlorobenzene
8 1,2,4-Trichlorobenzene
20 2-Chloronaphthalene
9 Hexachlorobenzene (HCB)
Chlorinated Aliphatic Hydrocarbons
12 Hexachloroethane
xx Trichlorobutadiene
xx Tetraehlorobutadiene isomers
xx Pentachlorobutadiene isomers
52 Hexachlorobutadiene
53 Hexachlorocyclopentadiene
Halpqenated Ethers
18 Bis(2-chloroethyl)ether
42 Bis(2-chloroisopropy1) ether
43 Bis(2-chloroethoxy) methane
40 4-Chlorophenyl phenyl ether
41 4-Bromophenyl phenyl ether
Phthalates
71 Dimethyl phthalate
70 Di ethyl phthalate
68 Di-n-butyl phthalate
67 Butyl benzyl phthalate
66 Bis(2-ethylhexyl) phthalate
69 Di-n-octyl phthalate
Range
{ug/kg
dry wt)
Meanb
(ug/kg
dry wt)
Detection
Frequency
Aromatic Hydrocarbons (Continued)
3-7.1
4-U5
0.4-U5
3-U5
U0.5-U5
U0.5-U5
U0.5-U5
U0.5-U5
U0.5-U5
U0.5-U10
U0.5-U50
U0.5-U25
U0.5-U25
U0.5-U25
U0.5-U25
U0.5
0.3-U10
U0.5-U10
U10
U0.5-U5
U0.5-U5
U0.5-U50
9.0-11
U20-760
U0.5-U25
U0.5-U25
U0.5-U25
5.7
4.8
4.1
4.6
3.5
3.5
3.5
3.5
3.5
6.8
34
15
15
15
17
0.5
6.8
6.8
10
3.5
3.5
40
11
170
17
17
20
3/5
1/5
1/5
1/5
0/6
0/6
0/6
0/6
0/6
0/6
0/6
0/6
0/6
0/6
0/6
0/1
1/6
0/6
0/6
0/6
0/6
0/5
4/5
3/5
0/5
0/5
0/5
58
-------
TABLE 13. (Continued)
Substance3
Miscellaneous Oxvaenated Compounds
54 Isophorone
HSL Benzyl alcohol
HSL Benzole acid
129 2,3,7, 8-Tetrach 1 orodi benzo-p-
dioxin
HSL Dibenzofuran
Oroanonitrogen Compounds
HSL Aniline
56 Nitrobenzene
63 n-Nitroso-di-n-propylatnine
HSL 4-ChloroaniHne
HSL 2-Nitroaniline
HSL 3-Nitroaniline
HSL 4-Nitroaniline
36 2,6-Dinitrotoluene
35 2,4-Dinitrotoluene
62 n-Nitrosodiphenylamine
37 1,2-Diphenylhydrazine
5 Benzidine (4,4'-diaminobiphenyl)
28 3,3'-Dichlorobenzidine
Pesticides
93 p,p'-DDE
94 p,p'-DDD
92 p,p'-DDT
89 Aldrin
90 Dieldrin
91 Chlordane
95 alpha-Endosulfan
96 beta-Endosulfan
97 Endosulfan sulfate
98 Endrln
99 Endrin aldehyde
100 Heptachlor
101 Heptachlor epoxide
102 alpha-HCH
103 beta-HCH
Range
(ug/kg
dry wt)
U0.5-U25
U10
U25-430
U5
U5
U1.0-U20
U0.5-U5
U0.5-U10
U50
U50
U50
U50
U0.5-U10
U0.5-U5
U0.5-U5
U0.5-U5
U0.5
U0.5-U100
U10-U25
U10-U25
U10-U25
U10-U25
U10-U25
U10-U25
U10-U25
U10-U25
U10-U25
U10-U25
U10-U25
U10-U25
U10-U25
U10-U25
U10-U25
Meanb
(ug/kg
dry wt)
20
10
140
5
3.7
14
4,1
8.1
50
50
50
50
8.1
4.1
4.1
4.1
0.5
67
10e
10e
10e
10e
10e
10e
10e
10e
10e
10e
10e
10e
10e
10e
10e
Detection
Frequency
0/5
0/4
3/4
0/2
0/4
0/6
0/5
0/5
0/4
0/4
0/4
0/4
0/5
0/5
0/5
0/6
0/2
0/6
0/5
0/6
0/5
0/6
0/6
0/6
0/5
0/5
0/5
0/6
0/5
0/6
0/6
0/6
0/6
59
-------
TABLE 13. (Continued)
Substance3
Pesticides (Continued)
104 delta-HCH
105 ganuna-HCH (lindane)
113 Toxaphene
Range
(ug/kq
dry wt)
U10-U25
U10-U25
U10
Mean5
(ug/kg
dry wt)
10e
10e
10e
Detection
Frequency
0/6
0/6
0/2
PCBs
xx Total PCBs
Volati1eOrganic Compounds
85 Tetrachloroethene
38 Ethyl benzene
L4.3f-U7
2/6
a Number indicates U.S. EPA priority pollutant number. HSL indicates
Hazardous Substance List.
b Mean calculated using the reported detection limit for undetected values.
c An anomalously high phenol value of 1,800 ug/kg dry weight was found at one
station. For the purposes of developing reference area concentrations, the
value has been excluded.
d U = Undetected at the method detection limit shown.
e The lower detection limit was used for the mean because it is probably
more representative of reference area conditions.
f L = The value is less than the maximum shown.
9 -- = Not analyzed.
Reference: Tetra Tech (1985b).
60
-------
Loading indices are the second method for ranking problem storm drains.
It will not be possible to calculate true discharge loading values for most
storm drains because the necessary flow and water quality data are generally
not available. However, sediment data collected during Phase I screening
can be used to calculate an index of contaminant loading. The loading index
is defined as the product of the contaminant concentration measured in the
storm drain sediment and the estimated average annual flow (see Section 4.1)
for each storm drain. Loading indices should be calculated for each of the
problem chemicals in each problem storm drain.
Problem storm drains should be prioritized based on the two ranking
procedures; the EAR and the loading index. Problem storm drains ranking the
highest using both procedures are recommended for immediate contaminant
tracing activities performed during Phase II. Lower priority drains can be
sampled as funding allows. Highest priority should be given to storm drains
with the greatest number of problem chemicals identified as pollutants of
concern for the Puget Sound area.
61
-------
SECTION 5.0. PHASE II - CONTAMINANT TRACING
The contaminant-tracing phase of the investigation is an extension of
the initial screening program. The objective during this phase is to
isolate contaminated sections of storm drain line and associated drainage
subbasins in problem storm drains identified during Phase I screening.
Source identification efforts can then be focused on contaminated sections
of storm drain lines while uncontaminated sections can be eliminated from
further study. To trace contaminants to the sources, additional field
sampling and continued investigation of land use in the drainage basin will
be required. Phase II will entail collecting additional sediment samples
from manholes throughout the storm drain system to trace contaminants in the
problem storm drains. The Phase II sampling effort will focus on problem
chemicals and associated source categories identified during Phase I and
the preliminary investigation. The Phase II sampling procedure is expected
to be an iterative process because it may take several rounds of sampling
to isolate contaminated sections of the storm drain system and identify the
ultimate sources of contaminants. In addition to supporting source
investigation, the contaminant-tracing procedure will identify sections of
the storm drain system where contaminated sediments have accumulated and
should be removed to prevent contamination of receiving waterways.
Information obtained during the preliminary investigation will be used to
help select sampling station locations. In addition, a detailed investiga-
tion of industrial and commercial facilities operating in each drainage
basin will be required to support the sampling program. The following
sections provide recommendations on conducting a detailed contaminant-
tracing program in problem storm drain systems.
5.1 SELECTION OF SAMPLING STATIONS
Contaminant-tracing sampling will have to be tailored to each problem
storm drain so the unique characteristics of each drain, its service area,
and specific problem chemicals are considered. This section of the report
62
-------
provides general guidance on selecting sampling station locations. In
general, the complexity of the storm drain system and the number of sampling
stations required will increase as a function of drainage basin area. Large
storm drain systems will serve several subbasins and have numerous branches
in the storm drain network. Therefore, it will be important to carefully
select sampling stations to minimize the number of samples required and to
allow identification of contaminant sources. Sampling stations will be
ultimately selected through a process of elimination. As noncontaminated
sections of the storm drain system are identified, they will be eliminated
from further investigation. Information obtained during the preliminary
investigation will provide a basis for selecting sampling station locations.
However, further detailed investigation of the storm water collection system
and the facilities operating in the drainage basin will be required during
this phase for accurate identification of sources.
It is recommended that sampling stations be selected in problem storm
drains to satisfy the following three objectives (arranged in order of
increasing level of detail):
Isolate subbasins with different land-use characteristics
Determine contaminant gradients along major trunk lines, if
possible
Isolate specific contaminant sources.
As the first step, sampling stations should be selected to isolate
specific branches and subbasins within the problem drainage basins. The
selection should be based on the layout of the storm drain system and the
land-use characteristics within each drainage subbasin. Sampling stations
should be located at manholes on major junctions with in the storm drain
system. The intent is to isolate subbasins with a high potential of
contributing to the contamination in the system from those with low
contaminant potential. For example, in the hypothetical storm drain system
shown in Figure 8, the service area can be divided into the following four
major subbasins:
63
-------
17
15
> »
A. RESIDENTIAL
COMMUNITY i i
i> ".- - * fV;»
-
RECBVING WATER &,
D. HIGHWAY RUNOFF
£3 dm
19
LEGEND
MANHOLE
< ROW DIRECTION
> I I 10
C. INDUSTRIAL COMPLEX
MiM
(e.g., abandoned landfiD.
chemkal storage area,
maintenance shop, etc.)
(DRAWING NOT TO SCALE)
Figure 8. Schematic of a hypothetical storm drain system.
-------
Subbasin A - Residential Community
Subbasin B - Commercial District
Subbasin C - Industrial Complex
Subbasin D - Highway Drainage.
For this system, approximately four sampling stations (i.e., Manholes 2, 7,
15, and 17) would be required to isolate the major subbasins connected to
the trunk line.
The presence of concentration gradients in the storm drain system can
be used to identify sources because contaminant concentrations in the
sediment will generally decrease in the storm drain line upstream and
downstream of the source input. Therefore, it is recommended that additional
stations be sited along the major trunk line to identify potential con-
taminant concentration gradients. In the hypothetical storm drain system
(Figure 8), additional stations at Manholes 5, 14, and 20 would be sufficient
to determine if there are any discernable gradients in contaminant concentra-
tions in the main trunk line.
Sampling stations should also be located at manholes upstream and
downstream of suspected contaminant sources to determine if the suspected
source has contributed significant amounts of contaminants to the storm
drain. A specific source would be identified as a problem if contaminant
concentrations in the storm drain sediments increase in the manhole below
the source. In Figure 8, additional sampling stations are recommended at
Manholes 12 and 13 to document contaminant contributions from the suspected
source in Subbasin C,
The contaminant tracing program should focus on the specific problem
chemicals identified during Phase I. A recent study conducted as part of
PSEP (Tetra Tech 1986c) identified pollutants of concern for the Puget Sound
region. For a select subset of the pollutant of concern list, a matrix was
65
-------
developed for the report that linked chemicals with specific source
categories and industry types. This information, provided in Appendix C,
can be used in selecting sampling stations to focus on facilities in each
subbasin that may have contributed specific problem chemicals to the storm
drain system.
Multiple rounds of sampling will likely be required to trace contami-
nants through the storm drain system to the ultimate sources and the
procedures above should be used to design subsequent sampling plans.
Sampling activities should continually move upstream in the storm drain
system toward the ultimate sources. As sampling progresses, uncontaminated
sections of the storm drain system are eliminated from further investigation,
and problem side connections and branch lines are identified. The general
progression in the contaminant tracing approach is as follows:
outfall-*-trunk 11 ne-»- branch line-*-side connection-*- catch basin -"-source.
5.2 INTERPRETATION OF SEDIMENT CHEMISTRY DATA
The decision to eliminate a portion of a storm drain system or a
drainage subbasin from further sampling must include review of data QA/QC
procedures and sediment characteristics. Review of contaminant data for a
storm drain system must be performed to ensure that analytical results are
properly interpreted, and detection of potential contaminant sources has
not been missed due to field or laboratory constraints.
Data validation procedures should be specified in the Quality Assurance
Project Plan (QAPP), and should include a QA summary report. In the QA
summary report, results from the QA/QC checks performed in the field and
laboratory should be compared against criteria established for the sampling
program in the QAPP. QA review of data should include, as a minimum,
assessing the following:
Method detection limits
Holding times for analyses
66
-------
Documentation and ehain-of-custody procedures
Frequency of QA/QC sample checks
Contamination of field and laboratory blanks by problem
chemicals
Control limits for laboratory replicate and matrix spike
results
Control limits for blind field replicate results
Control limits for SRM results.
If QA review indicates that any of the above QC checks do not meet data
quality objectives, then data must be qualified. Guidelines for performing
data review and qualification have been established for the U.S. EPA CLP
(U.S. EPA 1985a,b), and can be of assistance when performing the data
evaluation. Qualified data can be used in the decision process for tracking
contaminant sources. However, data qualifiers must be taken into considera-
tion when performing data comparisons. In some cases, high data variability
or semiquantitative results may require that resampling or reanalysis be
performed to allow determination of contaminant concentration gradients.
For example, if results from the blind field replicates are outside the
control limits for data variability, then this high variability must be
taken into consideration when comparing results from upstream and downstream
sample points. The resulting wide confidence limits may not allow determin-
ation of significant contaminant concentration differences.
In addition to evaluating QA/QC procedures, relative concentrations of
organic carbon and fine particulate matter in the samples should be assessed.
In general, contaminant loading will be higher in samples containing higher
concentrations of organic carbon and/or silt and clay (i.e., percent fines)
because of the greater sorption capacity of organic matter and fine
67
-------
particulate. To account for these sample characteristics, data can be
normalized to organic carbon content or total percent fines.
After the data QA review and characterization of sediment are completed,
results from the storm drain sample analyses can be compared to determine
which drainage subbasins require additional contaminant-tracing and which
can be eliminated from further investigation. The data should first be
reviewed to determine whether the TOC and percent fines content of sediments
within each storm drain line are comparable (i.e., within the variability of
the test method). If TOC and/or percent fines content of the sediment
samples collected froi each drain are not comparable, then data should be
normalized prior to the contaminant concentration comparisons. Further Phase
II contaminant tracing activities will be required in a specific drainage
subbasin if the concentration of the problem chemical in the upstream
station is equal to or greater than the concentration measured in the
downstream station.
A subbasin can automatically be eliminated from further investigation
if 1) the problem chemicals identified in the downstream station are
undetected in the sediment from the upstream station, 2) and the detection
limit for the problem chemical is significantly lower than the action
criteria (i.e., AET values, proposed freshwater sediment criteria, 90th
percentile, or street dust levels).
Elimination of drainage subbasins from further investigations where the
problem chemicals are detected (i.e., quantified) in sediments from the
upstream station, but at lower concentrations than the sediments from the
downstream station, will require careful data interpretation. For these
cases, it is recommended that data first be evaluated to determine whether
the differences in concentration between upstream and downstream stations
are significant. The following two steps are recommended:
The concentration of the problem chemical in the sediments
from the upstream and downstream station must be at least five
times greater than the method detection limit to ensure that
concentrations are in the quantifiable range of the method
68
-------
The concentration of the problem chemical in the sediments
from the upstream station must be significantly lower (based
on the QA review) than the concentration reported in the
downstream station.
As an example of the latter point, assume that the analytical variability
for the downstream station is ± 20 percent and the measured concentration
of the problem chemical in the downstream station is 100 ug/kg. Concentra-
tion of the problem chemical in the upstream station must be <80 ug/kg or
>120 ug/kg for a significant difference in contaminant concentration to exist
between the two stations. If the data are not qualified, then the data
variability detailed in the QA assessment can be used for the data com-
parisons. However, if the data are qualified, confidence limits must be
established on a case by case basis by reviewing the QA assessment.
Once it has been determined that significant differences in concentra-
tion between two stations exist, an examination should be made of con-
taminant gradients along the main trunk line. In most cases, it is expected
that concentration gradients will point in the direction of a particular
source. However, if a concentration gradient cannot be established for a
problem chemical after sampling in the upper reaches of a subbasin, then the
possibility of a nonpoint source of contaminants should be considered.
Additional data evaluation can include comparisons of the overall
chemical composition of the upstream and downstream sediment samples.
Ratios of chemical concentration within sediment stations can be compared
to determine if the relative contaminant composition (i.e., chemical
signature) changes between sampling stations. A change in chemical signature
between two stations may indicate multiple sources.
5.3 ADDITIONAL INVESTIGATIONS
In some cases, additional investigative activities will be required to
complete Phase II and the source identification process. The following
69
-------
additional activities are recommended to support the contaminant-tracing
program:
Distribute questionnaires to businesses in the problem
drainage basin to obtain information on current operations
Conduct inspections at key industries in the problem drainage
basin to locate, identify, and characterize wastes and
pretreatment processes and to provide information on proper
waste handling and disposal practices at the facilities
Review Ecology files and inspection reports on business and
industries
Conduct dye and/or smoke tests to verify side connections to
the storm drain system and to identify improper connections.
Questionnaire surveys are an effective way of obtaining information on
operations, waste discharges, and waste handling procedures for the busi-
nesses operating in the problem drainage basins. Questionnaires have been
used in recent contaminant source investigations in the Puget Sound area
(Romberg et al. 1987), and mailing lists can be obtained from the state tax
records (see Section 3.0). Questionnaires can be designed to target
particular industry types and the information obtained can be used to select
which businesses should be inspected. The following are suggestions for the
type of information that should be requested:
Type of business (e.g., product manufacture or service).
Water use and volume (e.g., restroom, rinsing, cooling,
product manufacturing, floor cleaning, washdown)
Types of connections to the storm drain system (e.g., catch
basins, floor drains, sumps)
Types of chemicals used or stored onsite.
70
-------
Inspections of industries will provide detailed information on possible
contaminant sources in the drainage basin. In addition, inspections can be
used to inform the facility of recommended waste handling practices to
reduce contaminant loading to the storm drain system. During these inspec-
tions, dye or smoke tests can be used if identification and verification of
individual side connections to the storm drain system are necessary.
5.4 SAMPLE COLLECTION
Recommended storm drain sediment sampling procedures, decontamination
procedures, documentation, and sample packaging and shipping requirement are
described in Section 4.2. It is recommended that chemicals analyzed during
the contaminant tracing program be the same chemical identified during Phase
I. Chemical and physical analyses and quality assurance/quality control
(QA/QC) recommendations for Phase II are presented in the following sections.
5.4.1 Chemicaland Physical Analyses
Analysis of sediment samples for Phase II contaminant-tracing should
include, at a minimum, the problem chemicals identified during Phase I
screening and the preliminary investigation. If a particular compound or
class of compounds was not detected during Phase I screening, or is not
indicated as important during the preliminary investigation, a smaller
number of variables may be analyzed than in Phase I screening. A technical
expert should be consulted prior to contaminant tracing to select appropriate
variables for this phase.
Analysis of a smaller number of target analytes may reduce costs
incurred from the laboratory. However, cost is determined by the required
analytical procedures, and the difference between analyzing a few target
compounds and a broader range of compounds may not be substantial. Relative
costs and analytical methods should be discussed with the analytical
laboratory prior to sample collection.
71
-------
Protocols developed under PSEP (Tetra Tech 1986d) should be used to
collect and analyze sediment samples for extractable organic compounds,
metals, and conventional variables. Analysis of conventional variables
(e.g., total solids, total organic carbon, and grain size) is recommended
during the contaminant-tracing effort to permit comparison with sediment
samples collected during Phase I screening.
5.4.2 Quality Assurance/Quality Control
Collection of field QA/QC samples specified for the Phase I screening
is also appropriate for Phase II contaminant tracing. A detailed discussion
of field QA/QC samples and collection procedures is presented in Section
4.2.6, and should be followed during Phase II sampling efforts.
Laboratory QA/QC requirements are described in the PSEP protocols (Tetra
Tech 1986d) and the U.S. EPA CLP statement of work (U.S. EPA 1987). Prior to
collection of sediment samples during Phase II, the project manager should
specify the frequency of analysis for laboratory QA/QC samples (i.e., method
blanks, matrix blanks, method spikes, and analytical replicates). The
assessment of data quality should be performed by a QA/QC expert and
reported with the sample data.
72
-------
SECTION 6.0. PHASE III - CONFIRMATION
The information obtained from Phase I screening and Phase II contaminant
tracing, combined with the supporting evidence from the site inspections, is
expected to provide sufficient evidence to identify contaminant sources for
many of the problem drains. However, in some cases, additional sampling
efforts may be required to confirm contaminant contributions from specific
sources. Source confirmation sampling performed during Phase III will
require that samples be collected from the actual discharges to the storm
drain rather than from sediment deposits in the drain. This section
provides general recommendations on how to collect and interpret discharge
monitoring data.
The following situations may warrant discharge sampling:
To distinguish between historical and ongoing source contri-
butions
To confirm sources where volatile organic compounds are
suspected as the major toxic contaminant
To determine contributions from NPDES-permitted sources
To document source contaminant loading conditions for
possible enforcement actions.
Storm drain sediments may represent historical rather than ongoing
source contributions. For example, when contaminants are present in storm
drain sediments, but cannot be associated with current activities in the
drainage basin, it may be necessary to monitor stormwater discharges to
determine whether there are ongoing sources in the basin. If no ongoing
sources are identified in the problem drainage basins, and historical
landfills or waste pits are not currently contributing contaminants, the
73
-------
adverse effects on the receiving environment may be reduced by simply
cleaning the storm drain system. Storm drain cleaning includes removal and
disposal of contaminated sediments from the drain lines and catch basins,
However, if ongoing contaminant discharges to the storm drain system are
identified, then source control efforts and storm drain cleaning will be
required. Drains should be resampled after cleaning to determine the
effectiveness of source control. Resampling should be conducted a sufficient
time period after cleaning to allow several storm events to contribute
runoff (and sediments) to the drain system. Samples should be collected
from the drain system adjacent to the contaminant sources (i.e., in catch
basins).
Volatile organic compounds have not been recommended for analysis
during Phase I screening and Phase II contaminant tracing because available
data indicate that volatile organic compounds are not frequently detected in
storm drain sediments. As part of PSEP, volatile organic compounds were
analyzed in sediment samples from 20 storm drains discharging into Elliott
Bay and the lower Duwamish River (Tetra Tech unpublished). These drains
were suspected of having a high potential for contamination based on the
visual appearance of the sediment and odors reported during sample collect-
ion. Detection frequencies for the volatile organic compounds in these
samples ranged from 0 to 40 percent. Compounds detected most frequently
included trans-l,2-dichloroethene (40 percent), trichloroethene (35 percent),
and ethyl benzene (35 percent). The remaining volatile organic compounds
were detected in 20 percent or less of the sediment samples analyzed.
However, storm drain monitoring conducted by Galvin and Moore (1982) and
U.S. EPA (1983c) indicate that volatile organic compounds are one of the
most frequently detected class of organic compounds found in stormwater
runoff. Consequently, analysis of volatile organic compounds is recommended
for discharge samples rather than storm drain sediments in drains where
potential sources of volatile organic compounds exist.
NPDES permits typically limit the concentration and loading of contamin-
ants in a facility's effluent, and do not set limits for sediments.
Although NPDES-permitted facilities are required to monitor their effluent,
toxic contaminants are not usually included in the variables measured.
74
-------
Therefore, additional discharge monitoring will be required to confirm the
contaminant contributions from these potential sources.
Enforcement typically requires detailed information on the sources,
including the type of discharge (i.e., stormwater runoff from property,
process water, illegal discharge), type of contaminants and their concentra-
tion in the discharges to the storm drain system, contaminant loading, and
the effects on the receiving environment. The presence of contaminants in
sediments collected from catch basins at the facility suspected of con-
taminating storm drains should be sufficient to document contaminant
problems associated with stormwater runoff. However, confirmation of
contaminant contributions from process water and other plant discharges to
the storm drain system will require collecting water samples and monitoring
flow in the discharge to the storm drain.
6.1 DISCHARGE MONITORING TECHNIQUES
In general, discharge monitoring is more complex than storm drain
sediment sampling because it is typically event-oriented and must consider
rainfall conditions and variability in flow and water quality conditions of
the discharge. Discharges to the storm drain system may consist of storm
water runoff or industrial effluent, such as noncontact cooling water or
process water. Stormwater monitoring must be carefully coordinated with
rainfall conditions. For other types of discharges to the storm drain,
timing of the sampling event will not be as critical. In addition, sampling
events will have to be scheduled during periods of low tide to avoid tidal
interferences in tidally influenced drains.
6.1.1 Bulk Water vs. Particulgte Fraction Analysis
An important issue in designing a discharge sampling plan is whether to
use bulk water or the particulate fraction of the discharge for chemical
analyses. Separate analysis of bulk water and the particulate fraction is
used to obtain lower detection limits for problem chemicals (Tetra Tech
1986a). Separate analyses are often required because many contaminants
associated with the particulate fraction may not be detected in analyses of
75
-------
bulk water only. Separate collection and analysis of the particulate
fraction concentrates contaminants adsorbed to solids, which improves
quantification of the contaminants.
Particulate fraction analysis of a discharge is recommended only if the
problem chemicals are difficult to detect due to low concentrations in the
bulk water discharge. If high concentrations of contaminants are expected
in the bulk water samples (i.e., greater than five times the method detection
limit), then particulate fraction analysis is not required. In cases where
relative pollutant loadings from drainage subbasins are difficult to assess
because concentrations of the contaminants are low in the bulk water samples,
particulate fraction analysis can improve the ability to quantify relative
contaminant concentrations. However, collection of an adequate particulate
fraction for laboratory analysis requires specialized equipment (e.g.,
continuous centrifuge or filtration apparatus) and can be labor-intensive.
Therefore, particulate fraction analysis should be considered only if no
other means of contaminant tracing are available. Alternatives to particu-
late fraction analysis include modifying analytical techniques to improve
detection limits, tracing the contaminant source further upstream to
minimize dilution, or diverting a potential source of dilution water during
sample collection.
If particulate fraction analysis is to be performed, the total suspended
solids content of the bulk water sample should be determined prior to
particulate sample collection. The quantity of particulates in a discharge
can vary widely. Discharges consisting primarily of noncontact cooling
water may contain less than 5 mg/L of suspended material, while stormwater
runoff may contain greater than 1,000 mg/L of suspended material during an
intense rainfall event. If a discharge consists primarily of cooling water
containing minimal particulate matter, then collection of an adequate
quantity of the discharge for particulate analysis may not be practical.
For example, if the suspended solids content of the discharge is 10 mg/L,
approximately 2,000 L of water would be required to obtain the sediment
necessary for analysis of extractable organic compounds (i.e., approximately
20 g). Processing 2,000 L of sample would require 8 h using a continuous
centrifuge that processes approximately 4 L of sample/rain. Manual collection
76
-------
of a sufficient volume of discharge water for analysis of the particulate
fraction would not be feasible in this situation.
The decision whether to analyze bulk water or the particulate fraction
will depend upon project objectives and funds, expected total suspended
solids content in the discharge, availability of a continuous centrifuge,
and hydrogeographic characteristics of the drainage subbasin. General
guidance on bulk water and particulate sampling is provided in the following
sections.
6.2 SAMPLE COLLECTION
Scheduling requirements for Phase III sampling activities will depend
on the type of source sampled. For example, stormwater runoff samples must
be collected during a rainfall event. Therefore, weather forecasts should
be monitored to aid in predicting rainfall conditions so field crews can be
mobilized in time to sample the event. Sampling of industrial discharges
(i.e., process waste, noncontact cooling water) can be scheduled to coincide
with a particular plant operation suspected as a potential contaminant
source. Automatic samplers can sometimes be used to monitor illegal
discharges. For this, samplers are placed in-line and programmed to collect
samples during a period when illegal discharges are suspected.
Sampling conducted during Phase III confirmation should follow the
same equipment decontamination, documentation, sample packaging, and
shipping procedures recommended for phase one screening (see Section 4.2).
6.2.1 Bulk Water Sampling
It is recommended that continuous composite samples be collected for
bulk water chemical analysis to provide representative samples of the storm
drain discharges. Samples should be collected with an automatic sampler
that composites samples in proportion to flow. If continuous, flow-
proportioned samples cannot be collected, manually composited samples can be
substituted. If samples are manually composited, the individual grab
samples should be collected no longer than 30 min apart if feasible.
77
-------
The length of time for compositing samples will be dependent on the
type of discharge sampled because flow characteristics will vary depending
on the source type (e.g., storm water, process water). Samples of industrial
process discharges should be composited over a 12-h period. Storm water
samples should be composited over the duration of the storm event or 12 h,
whichever is shorter. It will probably not be possible to achieve these
compositing periods for tidally influenced drains. However, samples from
tidally influenced drains should, at a minimum, be composited for the
duration of the low tide. In addition, rainfall must be recorded for all
stormwater runoff sampling events to enable comparisons with other storm
events.
If automatic samplers are used, the sampler should have a capacity at
least as large as the total volume required for the chemical analysis (see
Section 6.2.5) to avoid changing collection bottles during sampling. In
addition, access to samplers located inside storm drain manholes may be
infeasible during storm events. Equipment needed to collect discharge
samples is summarized in Table 14. General guidance on collecting discharge
samples for bulk water chemical analysis is provided below:
Automatic samplers and meters can be installed inside the
manhole on side connections to the problem drain. Sampling
equipment should be installed above the mean high tide
elevation in tidally influenced storm drains. Recommended
manhole entry procedures are described in Section 4.2.1. If
manhole installation is not feasible, and the equipment
cannot be installed in a secure area, provisions will have to
be made to protect the equipment from vandalism during the
sampling period. Consult the manufacturer's instruction for
proper installation and operation of the equipment.
Set equipment to collect samples for the appropriate time
interval (e.g., 12 h for process wastestreams). Insure that
sample collection bottles in the automatic sampler contain the
appropriate preservatives (see Section 6.2.4). Beginning and
78
-------
TABLE 14. LIST OF EQUIPMENT NEEDED FOR STORM DRAIN
DISCHARGE SAMPLING
Hard hats
Calculator
Lights
Maps
Camera and film
Manhole cover hook
Manhole depth and water level mea-
suring device
Sledge hammer
Methanol
Squirt bottles
Waste solvent bottle and funnel
Bags - garbage, small plastic
Rope
Barricades, traffic cones, traffic
signs
Sampling equipment:
Extension pole
Automatic sampler/flow meter
1-gal glass container
Aluminum foil
Sample containers (organic compounds,
metals, total suspended solids,
volatile organic compounds)
Coolers
Ice
Custody seals
Chain-of-custody forms
Analysis request forms
Field data log forms
Field logbook
Sample tags
Clear tape
Marking pens
Knife
Sample tray
Kimwipes or equivalent
Gloves (leather and chemical
resistant)
Coveralls (cotton and chemical
resistant)
Respiratorsa
(including extra filters)
Waders (two pair)a
Duct tape3
Oo/combustible gas meter and tubing*
Pnotoionization detector (PID)a
meter and tubing9
Drager tubes/bellows3
Decontamination sprayer3
Brushes (for decontamination)3
Self-contained breathing apparatus
(SCBA) equipment3
Safety harness and ropea
Alconox or equivalent
First aid kit
Clipboard
Tide tables
pH meter
Flow meter
Continuous flow centrifuge"
Pump/tubing"
Filtration equipment"
Generator"
a Required if personnel must enter manhole to install sampling equipment.
b Required for collection of particulate material.
79
-------
ending times should be recorded. Sampling equipment should
be checked periodically during the sampling period to ensure
that it is functioning properly.
A separate grab sample must be collected for sources requiring
analysis of volatile organic compounds because composite
samples are not suitable. Completely fill the sample bottle
to eliminate air bubbles and prevent loss of compounds.
If sampling stormwater runoff, the sampling equipment should
be set up prior to the event. Set the equipment to begin
sampling at the start of the rising limb of the runoff
hydrograph, and to stop when flow returns to pre-storm
conditions or 12 h later, whichever is shorter.
For manual compositing, grab samples can be collected from
the side connection to the storm drain. Manhole entry may
not be required. In some cases, samples may be collected by
attaching the sampling container to the end of an extension
rod that reaches into the manhole. Samples should be
collected in 1-gal glass containers at 30-min intervals. A
minimum container size of 1 gal is recommended to ensure
that sufficient sample volume is collected for compositing.
Each sample bottle should be fixed with preservative, sealed,
and placed on ice in a cooler. In addition, flow measurements
must be recorded each time a grab sample is collected so
samples can be composited in proportion to flow and to
determine when the stormwater runoff has subsided. At the
end of the sampling period, a single flow-proportioned
composite sample should be prepared by removing aliquots from
each grab sample and combining them in a single container.
6.2.2 Particulate Fraction Sampling
As explained earlier, separate analysis of bulk water and the partic-
ulate fraction of a discharge sample would only be recommended under special
80
-------
conditions due to cost and difficulty in collecting samples. The following
discussion presents general guidance for collecting separate bulk water and
particulate fraction samples.
Filtration and centrifugation techniques are commonly used to separate
the particulate fraction from the bulk water sample for analysis. Filtration
is recommended for most routine analyses because it requires less expensive
equipment and provides a sample suitable for direct chemical analysis (i.e.,
residue on a filter that can be extracted or digested). Centrifugation
techniques can yield comparable results, but require careful and complete
transfer of the sample from the centrifuge tubes prior to analysis. The
amount of material required for chemical analysis and the concentration of
suspended solids in the wastestream are the major factors affecting the
choice between filtration and centrifugation techniques. These two factors
determine the volume of sample that must be processed, and therefore, the
time required to collect each sample.
Most metals of interest are found at much higher concentrations than
the organic compounds and are more easily analyzed using a small sample
size. Generally, a minimum of 4 L of composited sample is sufficient for
analysis of metals in the particulate fraction (Tetra Tech 1986a). An
additional 2 L of sample is required for mercury analysis. These volumes
can yield sufficient amounts of particulate matter by filtration without
special techniques or extremely long filtration times. Because of the
potential for contamination in the field, and the time required to process
the samples, it is recommended that filtration procedures be conducted in a
field laboratory. Filtration procedures are summarized in "Analytical
Methods for U.S. EPA Priority Pollutants and Particulate Matter from
Discharges and Receiving Waters" (Tetra Tech 1986a). General equipment
requirements include a filter apparatus capable of efficiently handling the
required sample volume (i.e., 4-6 L), glass fiber filters, distilled water,
and appropriate glassware. Samples would be collected in the field using
the same procedures described in Section 6.2.1. However, a larger volume of
sample would have to be collected to meet the requirements for both the bulk
water and particulate fraction chemical analyses.
81
-------
Much larger sample volumes are generally required to obtain low
detection limits for organic compounds. As explained in the example
presented in Section 6.1.1, as much as 2,000 L of sample may have to be
processed to obtain a sufficient amount of particulate material for organic
compound analyses. Filtration of this volume of sample would be impractical.
Therefore, centrifugation techniques are typically used to process samples
for organic analysis on particulate fraction. Because of the large volumes
required, samples are typically processed in the field using a continuous-
flow centrifuge (Ongley 1982). Several field models are available that are
capable of processing between 4 and 8 L/min of sample (Tetra Tech 1986a).
The specialized equipment required for field centrifugation of particulate
samples includes a portable (i.e., truck-mounted), continuous flow centrifuge
pump and chemically inert tubing to collect the sample and route it through
the centrifuge, and a generator.
6.2.3 Chemical Analyses
Chemical analyses of discharge samples for Phase III confirmation
should include problem chemicals identified in the storm drain sediment
samples collected during Phase I and Phase II. In addition, other chemical
compounds (e.g., volatile organic compounds) identified as potential
contaminants in process wastestreams or stormwater runoff during the
preliminary investigation should be included in the analyses. A technical
expert should be consulted prior to discharge sampling to determine
appropriate variables. Groups of chemicals that may be included in the
analysis of samples for Phase III confirmation are listed below:
Metals
Extractable organic compounds
Volatile organic compounds
Conventionals (i.e., pH, total suspended solids, total
dissolved solids),
82
-------
Metals analyses can be conducted using PSEP protocols (Tetra Tech
1986d). A list of metals of concern and their recommended detection limits
in water is provided in Table 4. However, PSEP does not make recommendations
for the analysis of volatile organic compounds, extractable organic
compounds, and conventional variables in water samples. Therefore, for
these analyses, it is recommended that analytical procedures approved under
the Clean Water Act be used (U.S. EPA 1984). The analytical methods, sample
containers, preservation, and holding times for water samples collected
during Phase III confirmation are presented in Table 15.
The analysis of extractable organic compounds and pesticides/PCBs can be
performed on the same sample extract, so the collection of separate samples
is not required. Detection limits of 10-50 ug/L for acid/neutral compounds
and 0.05-1.0 ug/L for pesticides/PCBs are required under U.S. EPA CLP (U.S.
EPA 1987). These detection limits will provide adequate sensitivity for
source tracing. Methods for the preparation and analysis of water samples
are discussed in the U.S. EPA CLP statement of work (U.S. EPA 1987).
Analysis of discharge samples for volatile organic compounds (see Table
16) requires that detection limits of 0.5-1 ug/L be attained. These
detection limits are necessary for determining the trace levels of volatile
organic compounds which may be present in the system. The detection limits
specified for the CLP analysis of volatile organic compounds (5-10 ug/L|
U.S. EPA 1987) may not prove adequate in some instances for tracing
contaminants. Low-level detection limits for volatile organic compounds
should be specified when arranging laboratory analyses.
6.2.4 Quality Assurance/Quality Control
Field QA/QC samples that should be collected and analyzed during
discharge sampling are summarized below:
Field replicates and blind analytical replicates
Field decontamination blanks
83
-------
TABLE 15. RECOMMENDED METHODS, SAMPLE CONTAINERS, PRESERVATION,
AND HOLDING TIMES FOR WATER SAMPLE ANALYSIS
Variable
Semi volatile
organlcs
Pest1c1des/PCBs
Volatile organ 1 os
Metals
(total)
Total dissolved
sol Ids, total
suspended solids
011 and grease
Sample
Container
2-L glass bottle;
PTFEc-l1ned cap
2-L glass bottle;
PTFE-l1ned cap
Two 40-mL glass
vials; PTFE-l1ned
silicon septum caps
1-L glass or linear
polyethylene bottle,
PTFE-l1ned cap
2-L glass or plastic,
PTFE-Hned cap
2-L glass,
PTFE-Hned cap
Preservation
and Handling
Keep on Ice
(4°C)
Keep on ice
(4°C)
Fill, leaving
no air space,
keep In dark
on 1ce (4° C)
HN03 to pH<2
Cool (4° C)
Cool (4° C),
H2S04 to pH<2
Holding T1mea
7 days/40 days
7 days/40 days
14 days
8 mo
(Hg 28 days)
7 days
28 days
Method1*
Extraction,
SC/MS
Extraction,
6C/ECO
Purge and trap,
QC/HS
ICP, FLAA
6FAA, CVAA
Methods 160.1,
160.2
Method 413.1
or 413.2
Referencs
U.S. EPA 1984
U.S. EPA 1984
U.S. EPA 1984
Tetra Tech
1986c
U.S. EPA 1983b
U.S. EPA 1983b
a Where two times are given, the first refers to the maximum time prior to extraction, the second to the maximum
time prior to Instrumental analysis.
GC/MS « Gas chromatography/mass spectroscopy.
GC/ECD « fias chronatography/electron capture detection.
ICP » Inductively coupled plasma atomic emission spectroscopy.
FLAA = Flame atomic absorption.
GFAA = Graphite furnace atomic absorption.
CVAA » Cold vapor atomic absorption.
0 PTFE » Polytetrafluoroathylene.
84
-------
TABLE 16. VOLATILE ORGANIC COMPOUNDS RECOMMENDED
FOR ANALYSIS IN DISCHARGE SAMPLES
Halogenated Alkanes
Chloromethane
Bromomethane
Chloroethane
Methylene chloride
1,1'-Dichloroethane
Chloroform
1,2-Di ehloroethane
1,1,1-TH chl oroethane
Carbon tetrachloride
Bromodichloromethane
1,2-Diehloropropane
Chlorodibromomethane
1,1i2-Trichloroethane
Bromoform
1,1,2,2-Tetrach1oroethane
Chlorinated Aromatic Hydrocarbons
Chlorobenzene
Unsaturated Carbonyl Compounds
Acrolein
Acrylonitrile
Ketones
Acetone
2-Butanone
2-Hexanone
4-Methyl-2-pentanone
Halogenated Alkenes
Vinyl chloride
1,1-Dichloroethene
trans-1,2-Di chloroethene
cis- and trans-
1(3-Ti chloropropene
Trichloroethene
Tetrach1oroethene
Aromatic Hydrocarbons
Benzene
Toluene
Ethyl benzene
Styrene
Total xylenes
Ethers
2-Chloroethy1vi nylether
Miscellaneous
Carbon disulfide
Vinyl acetate
85
-------
Transport blanks
Trip blanks
SRMs.
Field replicate samples are used to determine total (i.e., analytical
plus field) sample variability. To collect field replicate samples, two
separate sets of samples are collected at a single station, and each set is
submitted separately to the laboratory. To collect blind analytical
replicate samples to evaluate analytical variability, samples are collected
at a single station from a completely mixed discharge composite sample.
This composited sample will then be split and for each different analysis,
the water will be placed into two separate sampling containers. To prepare
field and blind analytical replicates for volatile samples, separate samples
will be collected at the sampling station, and placed into separate sample
containers. The order of collection of replicates for volatile samples
should be noted on the summary sampling log (see Figure 4) and in the field
logbook. One set of blind analytical replicate samples collected from
composited samples could be analyzed by a different laboratory to evaluate
analytical variability between laboratories. All field and analytical
replicates should be labeled consistently with other samples and submitted
blind to the laboratory.
Field decontamination blanks, transport blanks, and trip blanks should
be collected during discharge sampling to assess potential contamination of
samples from ineffective decontamination procedures or during sample
collection, shipping, or storage procedures. The frequency of collection of
field blanks should be determined by the project manager prior to initiation
of the sampling effort. The overall frequency of field blanks is generally
5-20 percent of the total number of field samples.
Techniques for collecting field decontamination blanks are discussed in
Section 4.2.6. To prepare transport blanks, empty sample containers should
be filled in the field with analyte-free water. The transport blank is
opened in the field concurrently with the collection of a sample, and serves
86
-------
to check contamination that results from field sources, shipping, or storage
procedures. Any preservatives used for samples should also be added to the
transport blank to assess the potential contamination from this source.
Trip blanks are used when samples for volatile organic compounds are
collected to check cross-contamination among samples. To prepare trip
blanks, sample containers that were filled in a laboratory with analyte-free
water are not opened in the field. The trip blank accompanies all other
sample containers through field collection, shipping, and storage procedures.
All field blanks should be appropriately labeled and submitted blind to the
laboratory. Field blanks should be clearly identified on the sample log
form.
A SRM with trace metals in water is available from the National Bureau
of Standards. A certified SRM with organic constituents in water is
presently unavailable. Holding times for organic compounds in water (7 days
until extraction) preclude the availability of a prepared SRM. Organic
compound SRMs are available in ampules that can be added to a specified
volume of water. The recommended minimum frequency of submittal and
analysis of SRMs is 1 per 50 samples. The results of SRM analysis should be
evaluated according to procedures outlined in the PSEP protocols (Tetra Tech
1986d) to provide an estimate of the accuracy of sample analysis.
Laboratory QA/QC is performed by the analytical laboratory. A discuss-
ion of laboratory QA/QC requirements and the recommended minimum frequency
of analysis is presented in the PSEP protocols (Tetra Tech 1986d), and the
U.S. EPA CLP statement of work (U.S. EPA 1987). Prior to initiation of the
sampling efforts, the project manager should specify the frequency of
analysis of laboratory QA/QC samples (i.e., method blanks, matrix spikes,
method spikes, and analytical replicates). Technical evaluation of the data
should be performed by an expert, and results of all QA/QC analyses should
be reported with sample data.
6.2.5 Data Interpretation
Contaminant concentrations measured in discharge samples collected
during Phase III can be compared with available water quality criteria to
87
-------
evaluate potential impacts on the receiving environment. Available
freshwater and saltwater criteria (U.S. EPA 1986a) are summarized in
Table 17. These values are based on acute and chronic toxicity to aquatic
life. Although these ambient water quality criteria are not enforceable
standards, they are commonly used general guidelines for interpreting water
quality data. In January 1988, a subset of these U.S. EPA water quality
criteria (i.e., 20 inorganics and organic compounds) not listed in Table 17
was adopted by Washington State (WAC 173-201-047). These Washington State
water quality standards are enforceable in freshwater and saltwater outside
any mixing zones.
A discharge sample that exceeds ambient water quality criteria or
standards for a problem chemical may indicate that the storm drain system
warrants further consideration to determine if source control actions are
needed. However, because large variations may occur in contaminant
concentrations and loading from many potential sources, nonexceedance of
criteria for a single sampling event does not confirm the lack of a potential
source of contaminants. If results of the discharge sampling conflict with
available information from site investigations, further sampling may be
warranted.
Contaminant loadings for problem chemicals should be calculated for
each stormwater discharge based on the contaminant concentration and flow
data. These loadings can be used to compare different sources. Relative
contaminant contributions from individual sources are often used to rank and
select major contaminant sources for remedial action.
In addition to ambient water quality criteria and contaminant loading
data, data that are collected under the NPDES program should be reviewed.
As part of that program, contaminant concentrations measured in permitted
discharges are compared with permit limitations. This comparison will help
determine whether a facility is in compliance with its permit requirements,
and whether it could be a potential source of contaminants to a storm drain
system.
-------
TABLE 17. SUMMARY OF U.S. EPA (1986a) WATER QUALITY CRITERIA (US/L)
Freshwater Aquatic Life3
Acute Chronic
Tox i c i ty Tox i c i ty
Metal s
Antimony
Arsenic
Beryllium
Cadmi urn
Chromium
Copper
Lead
Mercury
Nickel
Selenium
Silver
Thallium
Zinc
Cyanide
LPAH
Naphthalene
Acenaphthyl ene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
HPAH
Fluoranthene
Pyrene
Benzo (a) anthracene
Chrysene
(9,000)
360
(130)
1.8d .
980c/16d
9.2C
34C
2.4
790C
260
1.2C
(1,400)
65C
22
(2,300)
U.JOO)
b
b
b
(3,980)
b
SL_
Total benzofluoranthenes ?
Benzo(a)pyrene
Indeno(l,2,3-c,d)pyrene
Dibenzo (a ,h) anthracene
Benzo (g , h , i )peryl ene
PAH Total
b
b
b
(1,600)
190
(5.3)
0.66°..
120c/lld
6.5C
1.3C
0.012
88C
35
(0.12)
(40)
59d
5.2
(620)
(5|0)
b
b
b
b
b
b
b
Saltwater Aauatic Life3
Acute
Taxi city
b
₯
43
1,100
2.9
140
2.1
75
410
2.3
(2,130)
95
1
(2.J50)
(9JO)
b
b
b
(40)
b
b
b
b
(300)
Chronic
Toxicity
b
₯
9.3
50
2.9
5.6
0.025
8.3
54
D
b
86
1
b
b
(7JO)
b
b
b
(16)
b
b
b
b
89
-------
TABLE 17. (Continued)
Freshwater Aquatic Lifea
Acute Chronic
Toxicity Toxicity
Phenols
Phenol
2,4-Dichlorophenol
4-Chloro-3-methyl
phenol
2 ,4-Dimethyl phenol
Pentachlorophenol
2,3,5, 6-Tetrachloro-
phenol
2,4,5-Trichlorophenol
2,4,6-Trichlorophenol
Nitrophenols
2-Chlorophenol
4-Chlorophenol
Phthalate esters
Pesticides
Aldrin
DDT
DDE
TDE
Demeton
Dieldrin
Endosulfan
Endrin
Suthion
Heptachlor
Hexachl orocycl ohexane
(Lindane)
Ma lath ion
Methoxychlor
Mi rex
Pa rat hi on
Toxaphene
PCBs
(10,200)
(2,020)
(30)
(2,120)
13e
b
(230)
(4,|80)
(940)
3.0
1.1
(1,050)
(0,06)
D
2.5
0,22
048
Tb
0.52
2,0
D
b
0.065
0.73
2.0
(2,560)
(365)
7.9e
b
(970)
(150)
(2, |00)
(3)
b
0.001
D
0.1
0.0019
0.056
0.0023
0.01
0.0038
0.06
0.1
0.03
0.001
0.013
0.0002
0.014
Saltwater Aquatic Life5
Acute Chronic
Toxicity Toxicity
(5,800)
«
13
b
(4,850)
5
(29,700)
(2,944)
1.3
0.13
(14)
(3,6)
b
0.71
0.034
0.037
b
0.053
0V16
b
b
b
b
0.21
10
b
(7.9)
(440)
b
b
b
(3.4)
b
0.001
b
0.1
0.0019
0.0087
0.0023
0.01
0.0036
0.1
0.03
0.001
B
0.0002
0.03
90
-------
TABLE 17. (Continued)
Freshwater Aquatic Life3
Volatiles
Acrylonitrile
Acrolein
Benzene
Trichloromethane
(chloroform)
Tetrachl oromethane
(carbon tetra-
chloride)
1,2-dichloroethane
Dichloroethylenes
Di ch 1 oropropanes
Dichloropropenes
Ethyl benzene
Halomethanes
Pentachlorinated
ethanes
Tetrachl oroethanes
1,1,2,2-Tetrachloro-
ethane
Tetrach 1 oroethy 1 ene
Toluene
Trichl oroethanes
1,1, 1-Trichloroethane
1 , 1 ,2-Triehloroethane
Trichloroethylene
Acute
Toxicity
(7,550)
(68)
(5,300)
(28,900)
(35,200)
(118,000)
(11,600)
(23,000)
(6,060)
(32,000)
(11,000)
(7,240)
(9,320)
b
(5,280)
(17,500)
18,000)
b
b
(45,000)
Chronic
Toxicity
(2,600)
(21)
D
(1,240)
b
(20.000)
b
(5,700)
(244)
b
(l.JOO)
(2,400)
(840) '
b
b
(9,400)
(21,900)
Saltwater Aquatic Life3
Acute
Toxicity
b
(55)
(5,100)
b
(50,000)
(113,000)
(224,000)
(10,300)
(790)
(430)
(12,000)
(3|0)
(9,020)
(10,200)
(6,300)
b
(31,200)
b
(2,000)
Chronic
Toxicity
b
(700)
b
.
(3,040)
b
(6,400)
(2gl)
b
(450)
(5,000)
b
b
Miscellaneous Oxvaenated Comoounds
2,3,7,8-Tetrachlorodi-
benzo-p-dioxin (TCDD)
Isophorone
Oraanonitrogen Compounds
Benzidine
Di nitre-toluene
Nitrobenzene
Nitrosamines
1,2-Diphenylhydrazine
(0.01)
(117,000)
(2,500)
(330)
(27,000)
(5,850)
(270)
(0.00001)
b
(2|0)
b
(12,900)
b
(590)
(6,680)
(3,300,000)
b
b
b
b
(3JO)
91
-------
TABLE 17, (Continued)
Freshwater Aquatic Lifea Saltwater Aquatic L1fea
Acute Chronic Acute Chronic
Toxicity Toxicity Toxicity Toxicity
Chlorinated Aliphatic Hydrocarbons
Hexachloroethane (980) (540) (940) b
Hexachlorobutadiene (90) (9.3) (32) °
Hexachlorocyclopenta-
diene (7) (5.2) (7) b
Ethers
Chloroalkyl ethers (238,000) b 55
Haloethers (360) (122) b b
Chlorinated Aromatic Hydrocarbons
Chlorinated benzenes
Chlorinated naphtha-
lenes
Dichlorobenzenes
(250)
(1,600)
(1,120)
(50)
(763)
(160)
(7.5)
(1970)
(129)
b
a ( ) = Where insufficient data are available to derive criteria, concentra-
tions representative of apparent threshold levels for acute and/or chronic
toxic effects are described in the U.S. EPA criteria documents. These
concentrations, along with associated narrative descriptions, are intended
to convey some information about the degree of toxicity of a pollutant in
the absence of established criteria. In some instances, the documents
provide separate toxicity concentrations for algae. These have not been
included in this table.
b No criteria or toxicity thresholds are presented in the water quality
criteria documents.
c Freshwater quality criteria for some chemicals are a function of hardness.
The relationship is not linear and the equations specific to each chemical
are found in the criteria documents. For this table, a criteria concentra-
tion based on a hardness value of 50 mg/L calcium carbonate is provided.
Exact criteria values must be calculated from the equations.
d The first value is for trivalent chromium (III) and the second value is for
hexavalent chromium (VI).
e Freshwater quality criteria for some chemicals are a function of pH. The
relationship is not linear and the equations specific to each chemical are
found in the criteria documents. For this table, a criteria concentration
based on a pH value of 6.5 is provided. Exact criteria values must be
calculated from the equations.
References U.S. EPA (1986a).
92
-------
SECTION 7.0 CONCLUSIONS
The storm drain monitoring program presented in this document provides
a sequential approach to identify and trace toxic contaminants in storm
drain systems. The four phases of the monitoring approach are implemented
in a sequential manner with the results of one phase determining the
necessity of each successive phase. The preliminary site investigation is
required to initiate all storm drain studies. Phase I initial screening of
in-line sediment samples is designed to screen a large number of drains to
eliminate uncontaminated drains from further consideration. In Phase II,
problem (i.e., contaminated) drains are selected for further intensive
inspection and sampling activities to trace contaminants to the ultimate
source. It is possible that smaller, less complex drainage basins that
serve a limited number of potential sources may not require additional
Phase II contaminant tracing procedures. Collection of water samples in
storm drains is recommended in Phase III if confirmation of contaminant
contributions from individual sources is required. Stormwater discharge
sampling may not be required in all cases if sources of contaminants can be
identified during the preliminary investigation and sediment sampling
efforts.
In-line sediment sampling during Phases I and II eliminates much of the
time and costs associated with collection of stormwater samples during
rainfall events. Because collection of sediment samples from drains is
relatively easy, a larger number of drains can be screened for contamination
compared those screened during discharge sampling activities. Sediment
sampling suffers from inherent difficulties in obtaining representative
samples. For example, runoff tidal action may disturb sediment deposits in
the drain thereby altering contaminant distribution patterns. However, it
is also difficult to obtain representative samples of stormwater discharge
due to the intermittent and highly variable nature of stormwater runoff.
Other limitations of the storm drain sediment sampling technique include 1}
the lack of specific criteria to assess the toxicity of many toxic eon-
93
-------
taminants, 2) the inability to calculate pollutant loading (measured in
Ib/day), and 3) the bias toward larger grain size deposition of particles
and their associated contaminants.
Despite the above limitations, the sediment sampling technique provides
a useful and cost-effective screening tool to identify toxic contamination
in storm drains. When used in conjunction with the preliminary investigation
phase and, if required, Phase III discharge sampling, toxic contamination
in storm drain systems can be traced to the ultimate source.
94
-------
REFERENCES
Barrick, R.C., and F.G. Prahl. 1987. Hydrocarbon geochemistry of the Puget
Sound Region III, polycyclic aromatic hydrocarbons in sediments. Estuarine
Coastal Shelf Sci. 25:175-191.
Battelle. 1985. Detailed chemical and biological analyses of selected
sediments from Puget Sound. Draft Final Report. Prepared for U.S.
Environmental Protection Agency Region X, Seattle, WA. Battelle Northwest
Laboratories, Sequim, WA. 300 pp.
Cargill, D. 25 February 1988. Personal Communication (conversation with
Ms. Beth Schmoyer). Washington Department of Ecology, Redmond, WA.
Clendaniel, W.T. 25 January 1985. Personal Communication (memorandum to
Mr. T. Lorenz, City of Seattle Sewer Utility, Seattle, WA). City of Seattle
Sewer Utility, Seattle, WA.
Clendaniel, W.T. 1 July 1985. Personal Communication (memorandum to file,
City of Seattle Sewer Utility, Seattle, WA). City of Seattle Sewer Utility,
Seattle, WA.
Crecelius, E.A., M.H. Bothner, and R. Carpenter. 1975. Geochemistry of
arsenic, antimony, mercury, and related elements in sediments of Puget
Sound. Environ. Sci. Technol. 9:325-333.
Conklin, J.T. 1986. The practicalities of manhole entry. Water Engineering
and Management, pp. 34-35.
EVS and Tetra Tech. (In preparation). Elliot Bay action program: a
monitoring approach. EVS Consultants, Vancouver, B.C. and Tetra Tech, Inc.,
Bellevue, WA.
Galyin, D.V., and R.K. Moore. 1982. Toxicants in urban runoff. Metro
Toxicant Program Report No. 2. Municipality of Metropolitan Seattle,
Seattle, WA. 160 pp.
Geissinger, L. 9 December 1987. Personal Communication (phone by Ms. Beth
Schmoyer). Seattle City Light, Seattle, WA.
Hubbard, T. 15 March 1988. Personal Communication (phone by Ms. Beth
Schmoyer). Municipality of Metropolitan Seattle, Seattle, WA.
Hubbard, T., and T. Sample. 1988. Sources of toxicants in storm drains,
control measures, and remedial actions. Paper presented at the First Annual
Meeting on Puget Sound Research held 18 and 19 March 1988, co-sponsored by
Puget Sound Water Quality Management Authority. Municipality of Metropolitan
Seattle, Seattle, WA.
95
-------
Jones & Stokes. 1983. Water quality management program for Puget Sound.
Prepared for U.S. Environmental Protection Agency, Region X, Seattle, WA.
Jones & Stokes Associates, Inc., Sacramento, CA.
Kennedy/Jenks/Chilton. 1987. Lake Union and ship canal sampling and
analysis program. Prepared for Seattle Engineering Department, Office for
Planning, Seattle, WA. Kennedy/Jenks/Chilton, Federal Way, WA.
Konasewich, D.E., P.M. Chapman, E. Gerencher, G. Vigers, and N. Treloar.
1982. Effects, pathways, processes, and transformation of Puget Sound
contaminants of concern. NOAA Technical Memorandum OMPA-20. National
Oceanic and Atmospheric Administration, Boulder, CO. 357 pp.
Krenkel, P.A., and V. Novotny. 1980. Water quality management. Academic
Press, NY. p. 225.
Lampe, J. 21 January 1985. Personal Communication (letter to Mr. C.
Carselli, Washington Department of Ecology concerning the Duwamish industrial
nonpoint source investigation). Municipality of Metropolitan Seattle,
Seattle, WA.
Malins, D.C., B.B. McCain, D.W. Brown, A.K. Sparks, and H.O. Hodgins, 1980.
Chemical contaminants and biological abnormalities in central and southern
Puget Sound. NOAA Technical Memorandum OMPA-2. National Oceanic and
Atmospheric Administration, Boulder, CO. 295 pp.
Malins, D.C., B.B. McCain, D.W. Brown, A.K. Sparks, H.O. Hodgins, and S.-L.
Chan. 1982. Chemical contaminants and abnormalities in fish and inverte-
brates from Puget Sound. NOAA Technical Memorandum OMPA-19. National
Oceanic and Atmospheric Administration, Office of Marine Pollution Assess-
ment, Boulder, CO.
Mowrer, J., J. Calambokidis, N. Musgrove, B. Drager, M.W. Beug, and S.G.
Herman. 1977. Polychlorinated biphenyls in cottids, mussels, and sediment
in southern Puget Sound, Washington. Bull. Environ. Contain. Toxicol.
18:588-594.
Municipality of Metropolitan Seattle. 1983. Duwamish clean water plan.
Metro, Seattle, WA.
Municipality of Metropolitan Seattle. 1987. The Duwamish River: running
toward a brighter future. Metro, Seattle, WA.
Neff, J.M., D.I. Bean, B.W. Cornaby, R.M. Vaga, T.C. Gulbraunsen, and J.A.
Scanlon. 1986. Sediment quality criteria methodology validation: calcula-
tion of screening level concentrations from field data. Prepared for U.S.
Environmental Protection Agency. Battelle, Washington, DC.
Ongley, E.D. 1982. Application of continuous flow centrifugation to
contaminant analyses of suspended sediment in fluvial systems. Environ.
Technol. Lett. 3:219-228.
96
-------
Prahl, F.G., and R. Carpenter. 1979. Role of zooplankton fecal pellets in
the sedimentation of polycyclic aromatic hydrocarbons in Dabob Bay, Washing-
ton. Geochim. Cosmochim. Acta 43:1959-1972.
PTI and Tetra Tech. (In preparation). Elliott Bay action program; revised
action plan. PTI Environmental Services, Bellevue, WA and Tetra Tech, Inc.,
Bellevue, WA.
PTI and Tetra Tech. 1988. Elliott Bay action program: analysis of toxic
problem areas. Draft Report. Prepared for the U.S. Environmental Protection
Agency Region X, Office of Puget Sound, Seattle, WA. Tetra Tech, Inc.,
Bellevue, WA. 330 pp. + appendices.
Puget Sound Water Quality Authority, 1987. 1987 Puget Sound water quality
management plan. Puget Sound Water Quality Authority, Seattle, WA.
Quinlan, E.A., P.M. Chapman, R.N. Dexter, D.E. Konasewich, C.C. Ebbesmeyer,
G.A. Erickson, B.R. Kowalski, and T.A. Silver. 1985. Toxic chemicals and
biological effects in Puget Sound; states and scenarios for the future.
Draft. NOAA Technical Memorandum. National Oceanic and Atmospheric
Administration, Boulder, CO. 334 pp.
Raven Systems & Research, Inc. 1988. Analysis of historic sampling results
from Georgetown steam plant and environs. Prepared for Seattle City Light.
Raven Systems & Research, Inc., Seattle, WA.
Romberg, P., and A. Sumeri. 1988. Sediment capping at the Denny Way CSO.
Paper presented at the First Annual Meeting on Puget Sound Research, held on
18 and 19 March 1988, co-sponsored by the Puget Sound Water Quality
Management Authority, Municipality of Metropolitan Seattle, WA.
Romberg, P., D. Healy, and K. Lund. 1987. Toxicant reduction in the Denny
Way combined sewer system. Municipality of Metropolitan Seattle, Seattle,
WA.
Sample, T, 27 March 1987. Personal Communication (letter to Mr. Ned
Bergman, U.S. Attorney General's Office, concerning catch basin on Wyckoff
property). Municipality of Metropolitan Seattle, Seattle, WA.
Sample, T. 23 October 1987. Personal Communication (data from Duwamish
industrial nonpoint source investigation provided to Tetra Tech, Inc.,
Bellevue, WA). Municipality of Metropolitan Seattle, Seattle, WA.
Schwartz, L. 1 August 1985. Personal Communication (memorandum to Mr. M.L.
LaBay, City of Seattle Sewer Utility, Seattle, WA). City of Seattle,
Engineering Department, Seattle, WA.
Smukowski, D. 14 December 1987. Personal Communication (phone by Ms. Beth
Schmoyer). Boeing Company, Seattle, WA.
Standifer, J.B. 17 May 1985. Personal Communication (memorandum to Mr.
W.T. Clendaniel, City of Seattle Sewer Utility, Seattle, WA). City of
Seattle Engineering Department, Seattle, WA.
97
-------
Tetra Tech. 1984. Laboratory analytical protocol for the Anaconda smelter
RI/FS. Tetra Tech, Inc., Bellevue, WA.
Tetra Tech. 1985a. Commencement Bay nearshore/tideflats remedial investi-
gation. Volume 1. Prepared for Washington Department of Ecology and U.S.
Environmental Protection Agency. Tetra Tech, Inc., Bellevue, WA.
Tetra Tech. 1985b. Commencement Bay nearshore/tideflats remedial investi-
gation. Volume 3, Appendices I-V. Prepared for Washington Department of
Ecology and U.S. Environmental Protection Agency. Tetra Tech, Inc.,
Bellevue, WA.
Tetra Tech. 1985c. Elliott Bay Toxics action program, initial data
summaries and problem identification. Prepared for U.S. Environmental
Protection Agency and Washington Department of Ecology. Tetra Tech, Inc.,
Bellevue, WA.
Tetra Tech. 1985d. Everett Harbor Toxics action program, initial data
summaries and problem identification. Prepared for U.S. Environmental
Protection Agency and Washington Department of Ecology. Tetra Tech, Inc.,
Bellevue, WA.
Tetra Tech. 1986a. Analytical methods for U.S. EPA priority pollutants and
particulate matter from discharges and receiving waters. Prepared for U.S.
Environmental Protection Agency and Washington Department of Ecology. Tetra
Tech, Inc., Bellevue, WA.
Tetra Tech. 1986b. Development of sediment quality values for Puget Sound.
Prepared for Resource Planning Associates for Puget Sound Dredged Disposal
Analysis, and Puget Sound Estuary Program. Tetra Tech, Inc., Bellevue, WA.
Tetra Tech. 1986c. Puget Sound Estuary Program. Users manual for the
pollutants of concern matrix. Prepared for U.S. Environmental Protection
Agency Region X, Office of Puget Sound, Seattle, WA. Tetra Tech, Inc.,
Bellevue, WA.
Tetra Tech. 1986d. Recommended protocols for measuring selected environ-
mental variables in Puget Sound. Prepared for Puget Sound Estuary Program.
Tetra Tech, Inc., Bellevue, WA.
Tetra Tech. 1987. Commencement Bay nearshore/tideflats feasibility study:
development of sediment criteria. Prepared for Washington Department of
Ecology and U.S. Environmental Protection Agency. Tetra Tech, Inc.,
Bellevue, WA.
Tetra Tech. 1988a. Elliott Bay action program: evaluation of potential
contaminant sources. Draft Report. Prepared for the U.S. Environmental
Protection Agency Region X, Office of Puget Sound, Seattle, WA. Tetra Tech,
Inc., Bellevue, WA.
Tetra Tech. 1988b. Elliott Bay action program: evaluation of sediment
remedial alternatives. Draft Report. Prepared for U.S. Environmental
Protection Agency Region X, Office of Puget Sound, Seattle, WA. Tetra Tech,
Inc., Bellevue, WA. 130 pp. + appendices.
98
-------
Tetra Tech. 1988c. Elliott Bay action program: the relationship between
source control and recovery of contaminated sediments in two problem areas.
Draft Report. Prepared for the U.S. Environmental Protection Agency Region
X, Office of Puget Sound, Seattle, WA. Tetra Tech, Inc., Bellevue, WA.
66 pp. + appendices.
U.S. Environmental Protection Agency. 1982-1983. Organic analyses for
Duwamish River survey. Unpublished data.
U.S. Environmental Protection Agency. 1983a. Interim guidelines and
specifications for preparing quality assurance project plans. U.S. EPA,
Office of Exploration Research, Office of Research and Development, Washing-
ton, DC.
U.S. Environmental Protection Agency. 1983b. Methods for chemical analysis
of water and wastes. EPA-600/4-79-020. U.S. EPA, Environmental Monitoring
and Support Laboratory, Cincinnati, OH.
U.S. Environmental Protection Agency. 1983c. Results of the nationwide
urban runoff program. Vol. I - Final Report. U.S. EPA, Water Planning
Division, Washington, DC.
U.S. Environmental Protection Agency. 1984. Guidelines establishing test
procedures for the analysis of pollutants under the Clean Water Act.
U.S. EPA, Washington, DC. Federal Register Vol. 49, No. 209, pp. 43234-
43436.
U.S. Environmental Protection Agency. 1985a. Laboratory data validation
functional guidelines for evaluating organics analyses. TDD IHQ-8410-01.
U.S. EPA, Hazardous Site Control Division, Washington, DC.
U.S. Environmental Protection Agency. 1985b. Laboratory data validation
functional guidelines for evaluating inorganics analyses. U.S. EPA, Office
of Emergency and Remedial Response, Washington, DC.
U.S. Environmental Protection Agency. 1986a. Quality criteria for water
1986. Update #1 and #2. EPA 440/5-86-001. U.S. EPA, Office of Water,
Washington, DC.
U.S. Environmental Protection Agency. 1986b. Test Methods for evaluating
solid waste. SW-846. U.S. EPA, Office of Solid Waste and Emergency
Response, Washington, DC.
U.S. Environmental Protection Agency. 22 October 1987. Personal Communica-
tion (computer printout of CERCLIS V.2.0. List 1 - site location listing
for King County retrieved from U.S. EPA Region X Superfund database). U.S.
EPA Region X, Seattle, WA. 21 pp.
U.S. Environmental Protection Agency. 1987. Contract laboratory program
statement of work for organics analysis, multimedia, multiconcentration.
Revised July 1987. IFB WA-87K236, 237, 288. U.S. EPA, Washington, DC.
99
-------
Wakeham, S.G. 1977. A characterization of the sources of petroleum
hydrocarbons in Lake Washington. Journal WPCF 49(7)J1680-1687.
Washington Department of Ecology. 1987. National Pollutant Discharge
Elimination System waste discharge permit for Marine Power and Equipment
Company, Inc. Permit #WA-003089-9. Final. Washington Department of
Ecology, Olympia, WA.
Wilber, W.6. and J.V. Hunter. 1979. Distribution of metals in street
sweepings, stormwater solids, and urban aquatic sediments. Journal WPCF
51:2810-2822.
Wisconsin Department of Natural Resources. 1985. Report of the technical
subcommittee on determination of dredge material suitability for in-water
disposal. Wisconsin Department of Natural Resources, Madison, WI.
100
-------
APPENDIX A
STORM DRAIN MONITORING
APPROACH COSTS
-------
APPENDIX A: STORM DRAIN MONITORING 'APPROACH COSTS
Costs that would be incurred during a storm drain investigation include
sampling equipment purchase or rental costs, personnel, and sample analysis.
Additional costs are incurred if the investigation indicates that corrective
actions or cleanup procedures are required. It was not possible to provide
a total cost for the storm drain monitoring approach presented in this
document because costs vary widely depending upon the following factors.
Availability and amount of background information required
for the preliminary investigation
Type of chemical and physical analyses and the level of
analytical services required
Number of drains sampled during each successive monitoring
phase
Variability in hourly rates for field personnel
Type of equipment required to collect the sediment or
stormwater samples.
The above factors will vary on a site-specific basis. In the following
sections, costs are presented on a per-unit-basis for sample analyses and
sampling equipment. In addition, approximate personnel costs (in total
person hours) are presented for a typical storm drain sediment sampling
effort. Using the costs and level-of-effort provided in the following
sections, total costs for implementing the recommended phased monitoring
program could be estimated for specific cases and drainage basins.
A-l
-------
ANALYTICAL COSTS
A summary of the costs for analytical procedures recommended in the
storm drain monitoring approach is presented in Table A-l. The costs
presented for each procedure can vary depending on the following factors:
Number of samples submitted for analysis
Sample characteristics
Level of services provided
Sample matrix (soil/sediment vs. water)
Turnaround time
Identification of additional organic compounds.
Most laboratories will negotiate a price break for samples submitted in
groups, reducing the per sample price as the number of samples submitted
rises. Price reductions of up to 20 percent can often be negotiated with a
laboratory when submitting large groups of samples (i.e., 20 or more).
Sample characteristics, such as high concentrations of target analytes
or interferences, may require that sample preparation and analysis procedures
be modified. A sample that contains oil or other interferences often
requires some form of sample cleanup (e.g., gel permeation chrontatography)
before analysis. Samples that contain high concentrations (i.e., >1 percent)
of target analytes often require cleanup and must often undergo one or more
dilutions before satisfactory results and detection limits can be obtained.
The analytical laboratories will perform the necessary dilutions, however, an
additional cost is often incurred for sample cleanup.
Tabulated analytical results are often the only data a laboratory will
provide without payment of an additional fee. QA/QC information is often
necessary to perform data review and validation. Obtaining QA/QC information
A-2
-------
TABLE A-l. SUMMARY OF ANALYTICAL COSTS
Variable
Approximate Cost
Per Sample ($)
Methoda
Target Compound List
Volatile organic
compounds
Extractable ABNC organic
compounds
Pesticides/PCBs
Priority pollutant metals
Sediment: 200-275
Total solids
Total volatile solids
Total organic carbon
Oil and grease
Particle size
Water: 200-250
Sediment: 250-300
Water: 375-750
Sediment: 475-800
Water: 135-160
Sediment: 160-200
Water: 150-210
10-20b
25-40
Water: 30-50
Sediment: 45-65
Water: 40-70
Sediment: 45-65
45-125
Purge & trap GC/MS
GC/MS
GC/ECD
AAS, CVAA, ICP, GFAA
Gravimetric
Gravimetric
Elemental analysis
Gravimetric, spectro-
photometric
Sieve and pi pet
a GC/MS = Gas chromatography/mass spectroscopy.
GC/ECD = Gas chromatography/electron capture detection.
AAS = Atomic absorption spectroscopy.
CVAA ~ Cold vapor atomic absorption.
ICP = Inductively coupled plasma atomic emission spectroscopy.
GFAA = Graphite furnace atomic absorption.
b Total solids measurements are normally included with other analyses at no
additional cost,
c ABN = Acid and base/neutral.
References: Tetra Tech (1986a)( U.S. EPA (1983br 1984, 1987).
A-3
-------
necessary for a thorough data validation review (see U.S. EPA 1987) can
raise the cost of sample analysis by 60 percent, depending on laboratory and
procedure.
The analysis of sediment samples costs more than the same analysis of
water samples due to sample characteristics. Matrix interferences often
necessitate the use of sample cleanup procedures to achieve the required
detection limits for sediment samples, while water samples tend to have
fewer matrix effects.
Sample turnaround time is usually from 14 to 40 days. When a shorter
turnaround time is requested, an additional fee is often charged.
The analysis requested from the laboratory may be for a particular
compound, class of compounds (e.g. pesticides), or a full scan of priority
pollutants. Conventional variables (i.e., oil and grease, total solids,
particle size, total organic carbon) should be analyzed to allow for
comparison with other data. In some cases, the initial full scan of
priority pollutants may detect only certain compounds or classes of compounds
in a discharge. Additional analyses of samples from a drainage basin, where
only a limited suite of toxic pollutants have been detected, can be tailored
to measure only the variables of interest.
The identification of organic compounds other than priority pollutants
and Hazardous Substances List (HSL) compounds may be requested for volatile
and extractable organic compounds, A library search can be performed that
compares mass spectra of standards with mass spectra generated during sample
analysis. Costs for the library search and reporting of additional organic
compounds can increase analytical costs up to 75 dollars per sample,
depending on the method and number of additional compounds requested.
FIELD SAMPLING COSTS
Field costs are divided into labor and equipment charges. Because of
variability in hourly rates for field personnel, labor cost estimates are
presented as total person hour requirements rather than as a dollar value.
A-4
-------
Purchase price and/or rental fees are presented for sampling equipment,
protective clothing, protective gear, and meters that are unique to storm
drain monitoring. Costs for sampling materials such as plastic bags, tape,
and ice have not been provided, because these items are considered standard
sampling materials and are not necessarily unique to the sampling effort
described in this report.
Field costs will vary depending on the type of samples collected (i.e.,
sediment or storm water). In general, it will cost more per station to
collect water samples than sediment samples due to sample compositing.
Estimated costs for conducting sediment vs. storm water sampling programs
are discussed below.
Storm Drain Sediment Sampling
Approximate personnel costs for a typical storm drain sediment sampling
program are summarized in Table A-2. Labor costs have been determined
based on a four-person field crew consisting of the sampler, a safety/rescue
person, a field note taker, and a traffic control person (needed for
manholes located in busy intersections). Based on experience from PSEP
sampling efforts, it is estimated that approximately 1 h will be spent at
each station to complete samprle collection, equipment decontamination,
documentation, and sample packaging and shipping procedures. A travel time
estimate of about 15 min between each station has been included in costs.
Because it is not cost-effective to mobilize an entire crew for a single
sampling station, the costs have been estimated based on a total of 20
sampling stations. In addition, it has been assumed that, due to tidal
interferences, sampling will only be possible for a 4-h period each day
(i.e., 5-day sampling event). Based on these assumptions, approximately 150
person hours will be required to complete a sediment sampling program for 20
sampling stations.
Approximate rental or purchase costs for the major field sampling
equipment are summarized in Table A-3. Protective gear and clothing are
the most expensive items. Protective clothing is expendable, and therefore,
will have to be purchased for each sampling effort. However, protective
A-5
-------
TABLE A-2. APPROXIMATE PERSONNEL COSTS FOR FIELD SAMPLING - SEDIMENT
Estimated Time Requirements/ Total Person Hour
Sampling Event/Station Requirements3
Equipment Mobilization 10 h 10
Sample Collection (1 h/station)'1
Sampler (1}C
Safety and rescue (1)
Field note taker, (1)
Traffic controller (1)
(if needed)
Subtotal 4 h/station 80
Travel Timed (4) 2 h/day 40
Documentation (1) 2 h/day 10
Equipment Demobilization 10 h 10
TOTAL 150
a Based on the following assumptions: 20 sampling stations, 4 h/day
sampling period due to tidal interferences, 5-day sampling event.
b Time requirements per station = 1 h.
c Indicates number of people.
d Includes travel time between sampling stations and travel to and from
site.
A-6
-------
TABLE A-3. APPROXIMATE COSTS FOR SAMPLING EQUIPMENT - SEDIMENT3
Approximate
Purchase
Cost ($)
Approximate
Weekly Rental
Cost ($)
Sampling Equipment
1 Stainless steel bucket
2 Stainless steel scoops
1 Large stainless steel spoon
Small stainless steel spoons
1 Telescoping extension plate
Coolers
Protective Clothing
Chemical resistant gloves
Inner gloves
Outer gloves
Chemical resistant coveralls
Hip waders, 2 pair
Protective Sear
2 Full-face Respirators
Filter cartridges
1 Safety harness/rope
2 SCBA
Meters
40
50
5
5
50
60 each
2.80/pair
3.50/pair
3.50/pair
50.00/pair
240
3.33 each
150
2,600
NAb
NA
NA
NA
NA
NA
NA
NA
NA
NA
32
c
12
210
Og/combustible gas
PTD meter
Draeger bellows
H£S tubes
1,500
6,000
200
3.50/tube
120
300
20
NA
a Costs may vary depending on supplier.
b NA = Not applicable.
c Cost included in respirator rental fee.
A-7
-------
gear is considered nonexpendable and could be rented to reduce costs of the
sampling effort.
Discharge Sampling
Approximate personnel costs for a typical discharge monitoring program
are summarized in Table A-4. Labor costs have been determined based on a
two-person field crew. Based on the sample compositing requirements of 12-h
intervals, an estimated 13 h will be required at each station to complete
sample collection, equipment decontamination, documentation, and sample
packaging and shipping procedures. For comparison with personnel require-
ments for the sediment sampling program, costs have been estimated based on
a total of 20 sampling stations.
Approximate rental or purchase costs for the major field sampling
equipment are summarized in Table A-5. The automatic sampler and continuous
flow centrifuge are the most expensive items. Therefore, these items would
probably be rented, particularly for small sampling projects.
REMOVAL COSTS
The cost for removing contaminated sediment deposits from storm drains
will be determined by the following major factors;
Diameter of the storm drain
Length of the drain lines that need to be cleaned
Amount of sediment accumulation in the storm drain.
Other factors will also indirectly affect the cost of removal opera-
tions as follows:
A-8
-------
TABLE A-4, APPROXIMATE PERSONNEL COSTS FOR FIELD SAMPLING - DISCHARGE
Estimated Time Requirements/
Hour
Sampling Event/Station
quirementsa
Equipment Mobilization
Sample Collection (13 h/station)^
Samplers (2)c
Documentation
Equipment Demobilization
TOTAL
10 h
26 h/station
1 h/day
10 h
Total Person
R e
10
520
20
10
560
a Based on the following assuptions: 20 sampling stations, 1 station/day.
k Time requirements per station = 13 h. This includes equipment set up and
decontamination.
c Indicates number of people.
A-9
-------
TABLE A-5. APPROXIMATE COSTS FOR SAMPLING EQUIPMENT - DISCHARGE3
Sampling Equipment
1 Telescoping extension rod
Coolers
Automatic sampler/flow meter
pH meter
Flow meter
Continuous flow centrifuge
Pump/tubing
Generator
Filtration equipment
Protective Clothing
Chemical resistant gloves
Inner gloves
Outer gloves
Chemical resistant coveralls
Hip waders, 2 pair
Protective Gear
2 Respirators
Filter cartridges
1 Safety harness/rope
2 SCBA
Meters
Oo/combustible gas
PiD meter
Draeger bellows
H£S tubes
Approximate
Purchase
Cost ($)
50
60 each
5,000
200
2,000
27,000
500
300
400
2.80/pair
3. 50/pair
3.50/pair
, 50.00/pair
240
3.33 each
150
2,600
1,500
6,000
200
3.50/tube
Approximate
Weekly Rental
Cost ($)
NAb
NA
800
25
300
2,500
75
45
60
NA
NA
NA
NA
32
c
12
210
120
300
20
NA
a Costs may vary depending on supplier.
b NA = Not applicable.
c Cost included in respirator rental fee.
A-10
-------
Tidal interferences
Season that cleanup activities are conducted
Medical monitoring requirements for personnel.
In tidally influenced drains, cleanup will only be feasible during low
tides. Tidal interferences will limit the number of hours during the day
when cleanup can occur. Tidal interferences may force cleanup crews to work
long or irregular shifts resulting in potential overtime charges. The
season the cleanup is conducted will affect how long cleanup crews can work.
During hot summer months, crews will have to take frequent breaks to avoid
heat stress. Heat stress is a particular problem due to amount of safety
equipment and clothing that must be worn in the potentially hazardous
environment of the storm drain. A medical monitoring program consisting of
a baseline medical examination, and a follow-up examination at the completion
of the project is recommended to ensure the health and safety of cleanup
personnel.
Another cost to be considered when budgeting a sediment removal
operation is disposal of the contaminated sediments after removal from the
storm drain. Although disposal costs are not considered during removal
operations, they may significantly affect the overall costs of cleanup.
Sediments that classify as a hazardous substance will have to be disposed of
at a licensed facility.
Because there are many variables involved in determining costs, it
will not be possible to develop accurate cost prediction procedures
applicable to all storm drains cleanup operations. However, cost figures
are available for several storm drain cleanup operations recently conducted
in the Puget Sound area. These costs, and a general description of the
cleanup operations, are presented below to provide a reference for overall
costs of cleanup activities.
A-ll
-------
Lander Street CSO/SD
Sediments in the Lander Street drain [combined sewer overflow/storm
drain (CSO/SD) 1105] contained lead at concentrations as high as 35 percent
(Hubbard and Sample 1988). The lead contamination was traced to atmospheric
deposition and surface runoff from the area surrounding a secondary lead
smelter (see Appendix B). In October 1984, the City of Seattle removed
approximately 20 yd^ of contaminated sediments from 1,600 ft of 36-in and
42-in lines in the SW Lander Street drain system. Sediments were dislodged
from the pipes using a high-pressure jet water wash and were collected at
the downstream end of the system. Weirs were installed at two locations in
the drain using sandbags to retain wash water. Water and sediments were
removed at each of the weirs by hydraulic jet-cleaning equipment. All
materials removed from the drain were transported to the smelter to recover
lead prior to sediment disposal. The cost of removing contaminated sediments
from Lander Street CSO/SD #105 were as follows (Clendaniel, B.( 25 January
1985, personal communication); $8,090.27 for labor and $5,661.00 for
equipment. The total cost of this cleanup project was $13,751.27.
SW Florida Street CSO/SD
Metro sampled the SW Florida Street (CSO/SD #098) drain system in 1984
and reported elevated concentrations of PCBs, pentachlorophenol, arsenic,
copper, and PAH (Hubbard and Sample 1988). Approximately 30 yd^ of
contaminated sediments were removed from the SW Florida Street drain in 1985
by the City of Seattle. Sediments were removed by bucket and dragline in a
400-ft section of a 36-in line that had the largest accumulations of
sediment. After dragline operations were completed, the line was flushed
with a high-pressure jet wash. Sandbag weirs were constructed in the
downstream end of the line to retain all wash water. Debris was collected
at the downstream end using hydraulic jet cleaner. The remaining 1,449 ft of
the 36-in to 48-in line was cleaned with a high-pressure jet wash, and the
debris was removed. In addition, all catch basins connected to the
contaminated section of the storm drain were cleaned using a hydraulic jet
cleaner. All material removed was placed in three lined settling ponds.
Decant liquids from the ponds were discharged into the City of Seattle
A-12
-------
sanitary sewer system. Solids were removed from the ponds and temporarily
stored on the nearby Purdy Company property. This material was transferred
to a licensed hazardous waste facility in Oregon for disposal (Standifer,
J., 17 May 1985, personal communication; Schwartz, L., 1 August 1985,
personal communication; Clendaniel, W., 1 July 1985, personal communication).
Approximately 30 yd^ of contaminated sediments were removed from the SW
Florida Street drain, and total cost of the removal operations was
$38,656.09. These costs include all charges to labor (approximately 60
percent) and equipment (approximately 40 percent).
Georgetown Flume
In 1984, Metro discovered that sediments in the Georgetown Flume, which
discharges into the head of Slip 4, were contaminated with PCBs (Metro 1987)
(see Appendix B). In November 1985, a contractor hired by Seattle City Light
removed the contaminated sediments from the flume (Ravens Systems Research
1988). Sediments were removed, treated at a treatment/storage/disposal
facility, and shipped to a licensed landfill for disposal. Removal
operations were similar to those described for the SW Florida Street drain.
In addition, debris was removed from the downstream end of the flume and
placed in storage tanks. Decant water in the storage tanks was tested
periodically. When the PCB concentration decreased to below 0.001 mg/L, the
decant water was discharged to the sanitary sewer system.
Removal costs (includes labor and equipment costs) for each section of
the storm drain were as follows; $10,500.00 for 547 ft of 6-in to 8-in pipe;
$12,500.00 for 240 ft of 15-in pipe; and $40,200.00 for 2,000 ft of open
flume. An additional $9,600.00 was spent to collect 50,500 gal of storm
water from a large rainstorm that occurred during cleanup operations. The
total cost of this project was $72,800.00.
A-13
-------
APPENDIX B
SUMMARY OF PREVIOUS
STORM DRAIN INVESTIGATIONS
-------
APPENDIX B: SUMMARY OF PREVIOUS
STORM DRAIN INVESTIGATIONS
Metro and the City of Seattle have successfully used the storm drain
sediment sampling approach to investigate contamination problems in several
storm drain systems in the Seattle area.
Metro developed the Duwamish Clean Water Plan in 1983 using funds from
a Clean Water Act 208 grant (Metro 1983). The plan was designed to identify
and control pollution problems in the Duwamish River and was adopted by the
Metro Council in 1983. Metro received a 2Q5(j) grant to implement part of
the plan that focused on studying industrial sites in the lower Duwamish
River and sampling of the major storm drain systems discharging into the
river. As part of the program, sediment samples were collected at key
junctions in 12 storm drain systems along the Duwamish River. The results
were compared with offshore sediment chemistry data and available data for
urban street dust in the Seattle area (Galvin'and Moore 1982). Significant
problem areas were identified in four of the 12 combined sewer outfall/storm
drains (CSO/SD) (Lander Street, Florida Street, Slip 4, and Fox Street).
LANDER STREET CSO/SD
The Lander Street drain (CSO/SD #105) serves a 54-ac area on the
interior of Harbor Island between 16th Avenue SW and 13th Avenue SW. In
March 1984, Metro collected sediment samples from the city CSO/SD #105 and
from a 21-in private drain located on the north side of Lander Street
(Hubbard and Sample 1988). Samples were analyzed for metals. Results,
summarized in Figure B-l, showed that the city drain was contaminated with
lead at concentrations as high as 370,000 mg/kg (37 percent). These values
are 800 times greater than the levels measured in typical urban street dust
(460 mg/kg; Galvin and Moore 1982). Lead concentrations in sediments
offshore of the Lander Street drains were measured at 18,000 mg/kg (1.8 per-
cent of total dry weight).
B-l
-------
Pb a 6300
As » 55
Cu a 73
Pb « 370,000
As * 2300
Cu » 690
(O
-------
The source of lead was traced to stack emissions from a former lead
smelter (Hubbard and Sample 1988) located in the city CSO/SD #105 drainage
basin. The smelter operated lead smelting, refining, and battery-recycling
facilities from 1937 to 1984. Lead concentrations as high as 180,000 mg/kg
(18.0 percent) were reported in soil samples collected near the smelter
property by the Puget Sound Air Pollution Control Agency in 1979 (Hubbard
and Sample 1988). Consequently, it was recommended that parking lots near
the smelter be paved to reduce contamination of surface water runoff.
Paving was completed in 1983.
In 1984, the City of Seattle removed approximately 20 yd-* of contami-
nated sediments from the Lander Street drain. The sediments were shipped to
a lead smelter for recovery (Hubbard and Sample 1988). When U.S. EPA
resampled the Lander Street drain in 1985 as part of the Elliott Bay Toxics
Action Program, they found that new sediment deposits in the drain were
again contaminated with lead up to concentrations of 52,800 mg/kg. Metro
also reported elevated lead concentrations (150,000 mg/kg) in sediments
collected from the Lander Street drain during 1986 (Sample, T., 23 October
1987, personal communication). These data indicate that residual contam-
ination from the lead smelter is an ongoing source of lead in the Lander
Street drain.
SW FLORIDA STREET CSO/SD
The SW Florida Street drain (CSO/SD #098) serves an approximately 25-ac
area between Harbor Avenue SW and 26th Avenue SW, and discharges into the
West Waterway. Metro collected sediment samples from 10 stations in the
drainage system, including 6 stations on the main trunk line (SW Florida
Street line), 1 station on the 26th Avenue SW line, 1 station at the sewer
overflow point, and 2 stations in catch basins connected to the SW Florida
Street trunk line (Hubbard and Sample 1988). Major contaminants found in
the drainage system are summarized in Figure B-2. With the exception of
PCBs, all contaminants found in the Florida Street drain system are used in
the wood treatment process. Arsenic, pentachlorophenol, and high molecular
weight polycyclic aromatic hydrocarbons (HPAH), a component of creosote, are
J-3
-------
PLAN VIEW
<0
0 200 400 800 800
ifMt
n*Mm
100 200
WYCKOFF
IA
WYCKOFF
LOCKHEED
SHIPBUILDING
IC-01
IC-02
r*i
06
f
TE
IC-04
OvwHtw
RMINAL
S
j
1
I
" MW
'
.
1
'
-IC-03
n
I
1
I
CONTAMINANT CONCENTRATION PROFILES
800,000-1
600,000-
I
400,000-
0)
O
200,000
oi
r-300,000
IA-01 IB-01 18-04 18-03 IC-02
STATION LOCATION
NOTE: Samples collected 23 October 1984,
Reference; S*mpl», T, (23 October 1987,
IC-01
personal communication).
Figure B-2, In-line sadiment data for stations on SW Florida Street CSO/SO,
B-4
-------
found in wood preservatives. Profiles of arsenic, pentachlorophenol, and
HPAH concentrations along the SW Florida Street trunk line show a distinct
peak approximately 1,700 ft upstream of the outfall adjacent to the Wyckoff
wood treating facility. These results match information obtained by U.S. EPA
during their 1983 investigation of the Wyckoff facility. U.S. EPA, Ecology,
and Metro determined that Wyckoff was illegally discharging hazardous wastes
containing arsenic, creosote, and pentachlorophenol into a catch basin
connected to the SW Florida Street drain (Ecology Northwest Region Office
facility file). As a result, the company was convicted on criminal charges,
fined, and placed on probation.
The PCB contamination in the SW Florida Street drain exhibited a
distinctly different pattern than the arsenic, HPAH, and pentachlorophenol
contamination. PCB concentrations in storm drain sediments were highest
(810,000 ug/kg) at the station upstream of the Wyckoff facility (Figure B-2).
Metro investigated properties in the vicinity of this station, and found
that the Purdy scrap yard had recycled old transformers containing PCBs.
However, there is some discrepancy in data from soil sampling conducted at
the Purdy property and the exact location of the PCB-contaminated soils has
not been determined (Cargill, D., 25 February 1988, personal communication).
The City of Seattle removed about 30 yd^ of contaminated sediments from
the SW Florida Street trunk line in 1985. Sediments were shipped to a
licensed hazardous waste facility in Oregon for disposal. Even so, subsequ-
ent sampling of a catch basin on the Wyckoff property has shown continued
contamination of surface water runoff from contaminated soil at the Wyckoff
facility (Sample, T., 27 March 1987, personal communication).
SLIP 4 DRAINS
Elevated concentrations of PCBs have been measured in the surficial
sediments in Slip 4 (Figure B-3). Samples collected by U.S. EPA from the
head of Slip 4 in 1982 and 1983 exhibited PCB concentrations between 1,600
and 5,600 ug/kg (U.S. EPA 1982-1983). Five drains discharge into Slip 4 (15
SD, Slip 4 CSO/SD 1117, Slip 4 SD, Georgetown Flume, and East Marginal Pump
Station CSO W043). Descriptions of each drain are presented in Table B-l.
5-5
-------
LEQEND
A SAMPIM STATION
MANHOLE
____ SANITARY SEWEK
__^_ STORUORAN
PCB DATA (ng/kg dry weight)
STATION U.S. EPA U.S. EPA METRO METRO
NUMBER 1982* 1983* 4/18/84° 4/18/84°
1
2
3
4
5
8
7
8
9
10
1 1
1 2
.
17,900
160.000
.
.
.
600
103,000
19,500
CITY LIGHT
15,000
17,000
19,000
79,000
19.000
12.000
340,000
462,000
1 ,800,000
.
TETRA TECH
9/850
.
.
-
260.000E
390E
13 1,600 5,600 6,000
14
15
18
1 .590 4,000
3,000
.
27SE
a U.S. EPA (1982-1983)
b Sample. T. (23 October 1987, peraonal communication).
o TatraT«ch(1988»)
E Estimated value
Figure B-3. Summary of
PCB data for Slip 4 drains.
B-6
-------
TABLE B-l. DESCRIPTION OF DRAINS
DISCHARGING INTO SLIP 4
1-5
Name
SDa
Outf al 1
Diameter (in)
66
Drainage
Basin Area
(ac)
30
Description of
Service Area
Drains approximately
1.5 mi
Georgetown Flume
60
Slip 4 CSOb/SD #117 24
Slip 4 SD 60
East Marginal Pump 36
Station CSO W043
150 (SD)
74.6 (CSO)
170
318
of 1-5 between S. Dawson and
S. Myrtle Streets and part of
Georgetown area.
Open wood flume originally
installed to discharge
cooling water from Seattle
City Light's Georgetown
Steam Plant. Exact service
area unknown. Numerous
other side connections have
been identified. All side
connections have been
plugged by Seattle City
Light.
Drain for the north end of
King County Airport.
Serves portions of King
County Airport.
Emergency sewer overflow for
Metro pump station.
a SD = Storm drain.
b CSO = Combined sewer overflow.
-------
In 1984, Metro collected sediment samples from the four storm drains
discharging into Slip 4 to determine the source of the PCB contamination in
offshore sediments (Metro 1987). The results indicated that three of the
four storm drains (i.e., Georgetown Flume, Slip 4 CSO/SD #117, and Slip 4
SD) were contaminated with PCBs (Figure B-3). PCB levels were measured at
17,900-160,000 ug/kg in the Georetown Flume, 103,000 ug/kg in Slip 4 CSO/SD,
and 19,500 ug/kg in Slip 4 SD (Sample, 23 October 1987, personal communi-
cation). These concentrations exceed the average level reported for urban
street dust from eight cities in the U.S. (770 ug/kg; Galvin and Moore 1982)
by 2-3 orders of magnitude. PCB concentrations in the sediments collected
from the 15 SD were 2-3 orders of magnitude lower than the concentrations
measured for the other three storm drains, and did not exceed levels
reported for urban street dust. Therefore, 15 SD has not been considered a
source of PCBs to Slip 4.
Seattle City Light (City Light) collected sediment samples in 1984 from
various locations along the Georgetown Flume to trace contamination
(Figure B-3). The highest PCB concentration (1,800,000 ug/kg) was found in
sediments collected from the downstream side of the tunnel in the flume
(Figure B-3). PCB contamination was subsequently traced to a City Light
property at the head of the flume where soil contained PCBs in concentrations
as high as 91,000,000 ug/kg. These soils were excavated to depths of 4-6 ft
to obtain cleanup levels of 150-200 ug/kg (Geissinger, L., 9 December 1987,
personal communication) and contaminated sediment deposits were removed from
the flume in 1985. City Light has plugged all side connections to the flume
to prevent future contamination, and sediment traps were installed in the
flume to collect sediments prior to discharge to Slip 4. City Light plans
to fill the flume to prevent it from being used in the future (Geissinger,
L., 9 December 1987, personal communication).
The source of PCBs in Slip 4 CSO/SD #117 has not been identified to
date. During cleanup activities in Georgetown Flume, City Light collected
sediment samples from Slip 4 CSO/SD and found PCB concentrations as high as
10,000 ug/kg (Smukowski, D., 14 December 1987, personal communication).
Boeing Company worked with Metro to trace contamination in this storm drain
line that crosses their property. However, they were not able to locate a
B-8
-------
PCB source in the area. In 1985, Boeing removed contaminated sediments from
the Slip 4 CSO/SD. This drain has since been rerouted to the pump station on
the Slip 4 SD system and discharges to Slip 4 via the 60-in line (Smukowsld,
D., 14 December 1987, personal communication).
PCB contamination has not been fully investigated in the Slip 4 SD
system to date. Consequently, it is not known whether there is an ongoing
source of PCBs in this drainage basin.
FOX STREET CSO/SD
The Fox Street drain serves an area of about 30 ac located on the west
side of East Marginal Way just south of Slip 3 (Figure B-4). Metro collected
sediment samples from the storm drain and from the Duwamish River upstream
and just offshore of the storm drain, and soil samples in the drainage
basin during 1984-1986 (Sample, T., 23 October 1987, personal communication).
Sampling station locations are shown in Figure B-4.
The results of the sampling and analyses for metals, summarized in
Table B-2, indicate that the drain in the lower part of the drainage basin
contained elevated concentrations of metals. Metals concentrations in storm
drain sediments from Manhole 1 (Figure B-4) located at the junction of the
north and south branch lines, are as much as 150 times greater than the
average concentrations reported in urban street dust (Galvin and Moore
1982). However, metals concentrations in the sediments from Manhole 2 (on
the south branch line; Figure B-4) are only 1.2-6.2 times greater than the
average street dust levels (Galvin and Moore 1982). This suggests that
metals contamination in the Fox Street CSO/SD probably originates in the
north branch line service area (i.e., east of South Fox Avenue). However,
because the distribution and concentration of metals found in the sediments
collected from catch basins on the Marine Power and Equipment property
(Table B-2) are similar to the metals found in the manhole (MH1) at the
entrance to the Marine Power and Equipment facility, it is likely that
fugitive dust emissions from Marine Power and Equipment have contributed to
the contamination observed at Manhole 1.
B-9
-------
TABLE B-2. SUMMARY OF METALS CONCENTRATIONS IN SEDIMENT
SAMPLES FROM FOX STREET CSO/SD 1116 AND SURROUNDING AREA (mg/kg)a
Fox Street CSO/SD 1116 MHbf 1
MH#2
Duwamish River Sedijnents.
Upstream of Drain
Offshore of Drain
Sediment Samples from
Catch Basins0
Mean Street Dust ^evels^
Date
Sampled
4/5/84
2/25/85
3/27/86
3/27/86
4/18/84
4/18/84
2/25/85
__
As
3,800
1,200
1,200
110
21
210
1,000-
3,900d
(2,200)c
25
Cd
4.4
6.7
5.4
6.2
<0.3
0.5
9.5-
19
(14)
1.0
Cu
1,200
900
710
380
60
290
2,300-
7,600
(5,000)
93
Pb
1,400
900
730
620
51
150
950-
1,900
(1,400)
520
Zn
5,600
2,300
2,300
850
160
1,000
6,200-
15,000
(10,000
310
a Stations shown on Figure B-4.
b MH = Manhole.
c Catch basins on the Marine Power and Equipment property that are connected to the Fox
Street drain downstream of Manhole #1.
d Range in concentration for nine stations.
e Mean value from n = 9.
f Galvin and Moore (1982).
Reference: Sample, T. (23 October 1987, personal communication).
B-10
-------
FROM
COMBINED
SYSTEM
SLIP NO. 3
FROM
COMBINED
SYSTEM
OVERFLOW
MANHOLE
boutn Brighton Street
South Willow Street
STORM DRAIN
MANHOLE
RIVER OR CATCH
BASIN SEDIMENT
SAMPLING STATIONS
South Myrtle Street
Figure B-4. Metro sampling stations on Fox Street CSO/SD (#116).
-------
Marine Power and Equipment is a shipbuilding and repair facility, and
occupies the lower portion of the drainage basin immediately downstream of
the junction of the north and south branch lines. Metals concentrations in
the catch basin sediment samples from the property are as high or higher
than concentrations found in the most contaminated storm drain sediments
(i.e., sediments from Manhole 1). During a 1984 site visit, Metro inspectors
reported that sandblast grit was present throughout most of the Marine Power
and Equipment property (Hubbard, T., 15 March 1988, personal communication).
Marine Power and Equipment is currently under a Consent Decree because of
unpermitted discharges of sandblasting materials from their dry dock
facility directly into the Duwamish River. Under the Consent Decree, Marine
Power and Equipment is required to remove contaminated sediments from the
Duwamish River adjacent to their property. In addition, a new NPDES permit
has been issued which requires that Marine Power and Equipment implement best
management practices to control the release of spent sandblast grit from
their facility (Ecology 1987).
DENNY WAY CSO
The Denny Way CSO is the largest and most frequent overflow point in
Metro's combined sewer system. The Denny Way CSO discharges into Elliott
Bay north of the Seattle downtown area at Denny Way. It produces a total
average volume of 500 Mgal/yr from approximately 30 to 60 overflow events.
The service area consists of almost 1,900 ac of mixed residential and
commercial land. Studies from the late 1970s to present have shown
contaminated sediments and adverse effects on benthic communities offshore
from the Denny Way CSO. As a result, the Denny Way CSO was identified in
the Elliott Bay Action Program as a significant problem area (Tetra Tech
1985c; PTI and Tetra Tech 1988).
In 1986, Metro conducted a trial study in the Denny Way CSO drainage
basin to determine if toxicant sources could be identified and reduced
pending a structural solution to eliminate CSO discharges (Romberg et al.
1987). As part of the investigation, Metro developed an inventory of 530
potential sources in the drainage basin based on Standard Industrial Codes
(SIC) and addresses from tax records. A questionnaire on wastewater
B-12
-------
discharges and chemical use was sent to each potential source. Fifty-four
percent of the businesses contacted responded to the questionnaire. Ninety-
six potential sources were visited by Metro inspectors to confirm the
questionnaire survey information and collect information to help develop
practical source control strategies. In addition, sediment and wastewater
samples were collected at key points within the CSO system (Figure B-5) and
analyzed for metals and organic contaminants. Wastewater samples were
collected for two different overflow events at most stations and sediment
samples were collected once at each station (Romberg et al. 1987).
The highest metals concentrations in both wastewater and sediment
samples were measured at stations downstream of two industrial laundries
that discharge wastewater to the Denny Way CSO. In addition, a large volume
of accumulated sediments in one part of the CSO system (Lake Union Tunnel),
located downstream of both laundries, was found to have high metals
concentrations. Both laundries installed new pretreatment equipment in 1986
to reduce the toxicant loadings in their discharges. Based on preliminary
data, metals loadings in sediments and wastewater were estimated to be
reduced by 50 percent for copper, 77 percent for lead, and 24 percent for
zinc after the pretreatment systems were installed (Romberg et al. 1987).
High concentrations of chromium and mercury in in-line discharge
samples were traced to a movie film developing operation. The facility has
been directed to use proper disposal practices, and as a result, the
toxicant input from this source is expected to be eliminated or greatly
reduced (Romberg et al. 1987),
Analyses of organic compounds were generally not as effective in
tracing contaminant sources as analyses of metals because of large variations
in organic compound concentrations between different sampling events at one
station. However, concentrations of toluene, tetrachloroethane, and ethyl
benzene were typically highest (50-200 ug/L) in the wastewater samples
collected downstream of the two industrial laundries (Romberg et al. 1987).
These three volatile organic compounds were also present at relatively high
concentrations (300-800 ug/kg wet weight) in sediment samples collected
immediately downstream of the laundries.. In addition, naphthalene appeared
B-13
-------
INDUSTRIAL
LAUNDRY
w
W , S
Republican
and Boren
Valley
and Boren
INDUSTRIAL
LAUNDRY
Westlake
and Ninth
LEGEND
SAMPLING SITE
COMBINED SEWER
W WASTEWATER SAMPLE
SEDIMENT SAMPLE
Tunnel MS
and Sixth
i Republican
and Pontius
INDUSTRIAL
LAUNDRY
Tunnel
and Ninth
Denny/
Lake union
Elliott and
Harrison
ELLIOTT BAY r~
INTEHCEPTOR1_
Reference: Romborg at al., 1987,
w
Minor and
John
'Melrose
and Olive
w
Westlake
and Denny
LAKE UNION
'TUNNEL
DENNY WAY
REGULATOR STATION
cso
OUTFALL
Figure B-5. Sampling stations in Denny Way CSO source toxicant
investigation.
6-14
-------
to be associated with the industrial laundries because it was only present
(8.5-170 ug/L) in wastewater samples collected downstream of these two
industrial laundries.
Metro is currently evaluating removal of the contaminated sediments in
the Lake Union Tunnel to prevent them from being flushed into Elliott Bay.
In addition, improvements in the stormwater routing program to enhance
inline storage and a notification and control system to reduce source
contaminant discharges when overflows occur are under consideration (Romberg
et al, 1987). Projected stormwater separation measures are anticipated to
reduce the number of CSO events from 50 events/yr to approximately 10
events/yr by the mid-1990s (Romberg and Sumeri 1988).
LAKE UNION AND SHIP CANAL STORM DRAIN
The City of Seattle, as part of a multi-year water quality management
program, conducted an investigation of 20 storm drains discharging to Lake
Union and the Ship Canal (Kennedy/Jenks/Chilton 1987). The study was
designed to 1) characterize the chemical composition of sediments that
accumulate in storm drains, 2) monitor the quality of stormwater discharges
3) model quality and quantity of stormwater runoff, and 4) estimate annual
pollutant loading to Lake Union. Sampling conducted during the investigation
included collecting in-line sediment samples from 11 storm drains, monitoring
flow and water quality during two rainfall events in 4 storm drains, and
hydraulically modeling the storm drain system to estimate average annual
stormwater discharges for each drainage basin. A first flush storm event
was also monitored in one drainage basin that had experienced 45 days of dry
weather prior to the sampling event.
The results of the investigation indicated that stormwater quality in
the Lake Union drains was generally better than that reported for other
urban areas (Kennedy/Jenks/Chilton 1987). However, total Kjeldahl nitrogen
concentrations were higher than other comparison cities, and metals concen-
trations were generally higher than those reported for the City of Bellevue.
Data from stormwater sampling also showed that weather conditions prior to
the sampling event affected the quality of discharge. The basin sampled
B-15
-------
immediately following a 45-day dry period exhibited considerably higher
concentrations for many pollutants when compared with results- from a typical
winter storm event in the same basin. Conventional pollutants (i.e., total
suspended solids, settleable solids, and turbidity) concentrations were up
to six times greater for the first flush event and metals concentrations were
1-3 orders of magnitude greater for the first flush event. Metals concen-
trations in storm drain sediments (Table B-3) exceeded the proposed fresh-
water and saltwater criteria for sediments used in comparisons at most of
the sampling stations. Based on these results, the city recommended that
efforts to control stormwater volume and solids loading would be most
effective in the two largest basins and four medium size basins that
exhibited the highest pollutant concentrations. In addition, three small
basins were recommended for source control investigations because of
elevated metals concentrations observed in the storm drain sediments
(Kennedy/Jenks/Chllton 1987).
B-16
-------
TABLE B-3. SUMMARY OF METAL CONCENTRATIONS IN SEDIMENTS
COLLECTED FROM STORM DRAINS DISCHARGING INTO LAKE UNION
Range Mean3 Detection
Chemical (mg/kg dry wt) (mg/kg dry wt) Frequency
Arsenic
Beryllium
Cadmi um
Chromium
Copper
Lead
Mercury
Nickel
Selenium
Silver
Zinc
0.74-1,700
<0.25-7.3
0.42-39
19-350
22-1,300
210-2,700
0.036-2.29
21-660
0.23-3.0
0.54-9.6
280-7,600
210
1.1
8.2
96
360
1,000
0.71
190
1.4
2.7
180
11/11
4/10
11/11
11/11
11/11
11/11
10/10
10/10
3/7
7/7
10/10
a Mean calculated using the reported detection limit for undetected values.
Reference: Kennedy/Jenks/Chilton 1987.
B-17
-------
APPENDIX C
POLLUTANTS OF CONCERN
-------
TABLE C-l. INORGANIC CONTAMINANTS OF POTENTIAL
CONCERN IN PUGET SOUND3
Antimony
Arsenic^
Cadmium"
Chromium0
Copper*5
Leadb
Mercuryb
Nickel
Silverb
Zinc
Cyanide
Organotinsd
a The elements listed are 11 of the 14 U.S. EPA priority pollutant metals.
The remaining three priority pollutant metals not recommended are beryllium,
thallium, and selenium.
Beryllium and thallium are toxic but have not been found at concentrations
that exceed reference levels in Puget Sound (see Tetra Tech 1986a, Ap-
pendix A).
High selenium concentrations have been reported in sediments in a single
Puget Sound study; these values are considered to be elevated because of
spectral interferences during the particular instrumental analysis used
(see Tetra Tech 1986a, Appendix A). Other studies using alternative
techniques have not found sediment levels of selenium in excess of reference
conditions.
" These elements have previously been suggested as contaminants of concern
in Puget Sound based on elevated sediment concentrations, bioaccumulation
potential, or toxicity (see Konasewich et al. 1982; Jones and Stokes 1983).
c Although not found at elevated concentrations in Puget Sound sediments,
chromium may be of concern in localized areas where chromium-rich wastes are
discharged (e.g., chrome-plating industries).
d Organotins, especially tributyltin, are highly toxic components of some
antifouling paints used on ships. Analytical techniques are not readily
available and very little data are available for these compounds in Puget
Sound waters. Because of the large number of shipyard industries in the
Puget Sound area, organotins may be of concern.
Reference: Tetra Tech (1986c),
C-l
-------
TABLE C-2. ORGANIC CONTAMINANTS OF POTENTIAL
CONCERN IN PUGET SOUND
Phenols
65a phenol0
HSLb 2-methylphenol0
HSL 4-methylphenolc
34 2,4-dimethylphenol
24
31
22
21
55
77
1
HSL
39
84
72
76
74
Substituted Phenols
2-chlorophenol
2,4-dichlorophenol
4-chloro-3-methylphenol
2,4,6-trichlorophenol
HSL 2,4,5-trichlorophenol
64 pentachlorophenold
57 2-nitrophenol
59 2,4-dinitrophenole
60 4,6-dinitro-o-cresole
Miscellaneous Organic Acids (guaiacols/resin acids)^
2-methoxyphenol (guaiacol)
3,4,5-trichloroguaiacol
4,5,6-trichloroguai acol
tetrachloroguaiacol
mono- and di-chlorodehydroabietic acids
Low Molecular Weight Aromatic Hydrocarbons^
naphthalene
acenaphthylene
acenaphthene
80 fluorene
81 phenanthrene
78 anthracene
Alkylated Low Molecular Weight Aromatic Hydrocarbonsd>9
2-methylnaphthalene
1-methylnaphthalene
1-, 2-, and 3-methyl phenanthrenes
High Molecular Weight PAH
fluoranthene
pyrene
benzo(a)anthracene
chrysene
benzo(b)fluoranthene
75 benzo(k)fluoranthene
73 benzo(a)pyrene
83 indeno(l,2,3-c,d)pyrene
82 dibenzo(a,h)anthracene
79 benzo(g,h,i)perylene
C-2
-------
TABLE C-2. (Continued)
Chlorinated Aromatic Hydrocarbons
26 1,3-dichlorobenzene 8 1,2,4-triehlorobenzene
27 1,4-dichlorobenzene 20 2-chloronaphthalene
25 1,2-dichlorobenzene 9 hexachlorobenzene (HCB)
Chlorinated Aliphatic Hydrocarbons
12 hexachloroethane
52 hexachlorobutadiene"
Phthalatesd
71 dimethyl phthalate 67 butyl benzyl phthalate
70 diethyl phthalate 69 di-n-octyl phthalate
68 di-n-butyl phthalate
Miscellaneous oxygenated compounds
54 isophorone . polychlorinated dibenzofuransd'.J
HSL benzyl alcohol1 polychlorinated dibenzodioxinsJ
HSL benzole acidl
HSL dibenzofuran1
Organonitrogen CompoundsK
62 N-nitrosodiphenylamine
9(H)-carbazolel
Pesticides
93 p,p*-DDEdm 98 endrind
94 p,p'-DDDdm 100 heptachlor
92 p,p'-ODTdm 102 alpha-HCH
89 aldrindm. 103 beta-HCH
90 dieldrind 104 delta-HCH
91 alpha-chlordane 105 gamma-HCH (lindane)
PCBs"
Total PCBs (this class includes monochloro-
through decachlorobiphenyls)
C-3
-------
TABLE C-2. (Continued)
45
46
16
44
13
23
10
11
29
30
Volatile Halogenated Alkanes0
chloromethane
bromomethane
chloroethane6
dichloromethane
l»l'-dichloroethane
chloroform
1,2-dichloroethane6
1,1,l-triehloroethane6
6 carbon tetrachloride6
48 bromodichloromethane6
32 1,2-dichloropropane
51 chlorodibromomethane6
14 1,1,2-trlchloroethane
47 bromoform6
15 l,l»2,2-tetrachloroethanee
Volatile Halogenated Alkenes0
vinyl chloride
l,l'-dichloroethene
trans-1,2-dichloroethene
33 cis-l,3-dichloropropene
trans-1,3-dichloropropene
87 trichloroethene
85 tetrachloroethene
Volatile Aromatic and Chlorinated Aromatic Hydrocarbons0
4 benzene
86 toluene
38 ethyl benzene
HSL styrene (ethenylbenzene)
HSL total xylenes
7 chlorobenzene
NOTE: Compounds not recommended from the U.S. EPA priority pollutant list
include:
Halogenated ethers {two volatile and five semi volatile
compounds) are rarely reported in Puget Sound and are not
expected to persist in sediments.
Hexachlorocyclopentadiene has not been confirmed to be present
in Puget Sound sediments, is easily degraded during laboratory
analysis, and has no suspected sources in Puget Sound.
Acrolein and acrylonitrile have not been detected in Puget
Sound sediments and are difficult to analyze for in routine
volatiles analysis.
Other priority pollutants not recommended are indicated in the
following footnotes.
a Indicates U.S. EPA priority pollutant number.
U.S. EPA Hazardous Substance List (HSL) compound.
C-4
-------
TABLE C-2. (Continued)
Phenol, a U.S. EPA-priority pollutant, has been reported at elevated
concentrations in several areas of Puget Sound. Phenol is toxic and may be
associated with effects observed at selected sites in Commencement Bay, but
because of its slightly polar character, does not have a high bioaccumu-
lation potential. Industrial chemical synthesis is one of many sources of
phenol. 2-Methylphenol is an HSL compound and is a known component of Kraft
pulp effluents. 4-Methylphenol is an HSL compound that was reported at high
concentration in numerous areas of Commencement Bay. There are little or no
historical data available for this compound and it is unknown whether
4-methylphenol derives principally from degradation of other compounds or is
present directly in industrial discharges. The occurrence of 4-methyl-
phenol was highly correlated with sediment toxicity and effects on benthic
biota in a problem area near a pulp and paper operation in Commencement
Bay. The compound may also be derived as a ground-water contaminant in
other areas.
Compound or group of compounds has been designated previously as a
contaminant of concern in Puget Sound based on elevated sediment concentra-
tions, bioaccumulation potential, or toxicity (Jones and Stokes 1983,
Konasewich et al. 1982, Quinlan et al. 1985).
6 Compound is seldom or not reported, possibly due to analytical problems
presented by the compounds or limited number of analyses.
Guaiacol was reported in Commencement Bay and is useful as an indicator
of pulp mill effluent. The chlorinated guaiacols have toxicity comparable
to phenolic priority pollutants, are persistent, and are good indicators of
chlorinated pulp mill effluents. Chlorinated dehydroababietic acids are
also good indicators of chlorinated pulp effluent and are expected to be
toxic and persistent (based on studies of unchlorinated dehydroabietic
acid).
9 These nonpriority pollutant (U.S. EPA) compounds are often detected in
Puget Sound sediments. Although this is not an exhaustive list of alkylated
aromatic compounds, the compounds shown are accessible as analytical
standards and are useful for determining alkylated/non-alkylated ratios used
to indicated PAH sources.
HCBD is a toxic and carcinogenic U.S. EPA-priority pollutant that has
been reported in various regions of Puget Sound. It is largely a byproduct
of chlorinated hydrocarbons (e.g., tri- and tetrachloroethylene) manufacture.
1 Dibenzofuran, benzyl alcohol, and benzoic acid are HSL compounds and have
been detected frequently in Commencement Bay.
C-5
-------
TABLE C-2. (Continued)
J Both classes of compound are of concern because of their severe toxic
affects on higher organisms. Dedicated chemical analyses are required for
these compounds, and few such analyses have been performed on Puget Sound
samples. Thus, the occurrences of these compounds are unknown but are
nonetheless of great potential concern.
k
The remaining seven priority pollution organic bases are seldom detected in
Puget Sound and often present analytical problems (e.g., benzidine and
3,3-dichloro-benzidine).
9(H)-carbazol is a component of creosote and coal tar and has been reported
in Puget Sound regions with these sources.
m DDT and its chlorinated hydrocarbon metabolites, DDE and ODD, are U.S. EPA-
priority pollutants that are persistent, readily bioaccumulated, and very
toxic; DDT itself is a carcinogen. Of the U.S. EPA-priority pollutant
pesticides, these compounds are most frequently reported in Puget Sound
sediments and biota although not nearly as often as the other compounds
recommended. Aldrin, another pesticide-priority pollutant, has not been
widely reported in Puget Sound but is of concern because of its extremely
high acute toxicity.
n PCBs are a class of U.S. EPA-priority pollutants that are widely dis-
tributed among sediments and biota of Puget Sound. PCBs are persistent and
have a high potential to bioaccumulate. PCBs are the only substances present
in Commencement Bay tissue samples that were judged to present a significant
health risk, and were also highly correlated with sediment toxicity.
Commercial PCB mixtures are suspected of containing carcinogens or co-car-
cinogens and were used historically in enclosed systems (e.g., capacitors
and transformers) that have often been discarded into the environment.
0 Some of the volatile organic compounds are of concern because of their
use in industry and their potential for contamination of groundwater.
Reference: Tetra Tech (1986c).
C-6
-------
TABLE C-3. POLLUTANTS OF CONCERN LIST FOR PUGET SOUND
Pollutants of Concern Municipal8
Antimony
Arsenic
Cadmium
Chroml urn
Copper
Lead
Htroury
Nickel
Silver
Zlno
Cyanides
LPAH
Naphthalene
Acenaphthlyena
Acenaphthene
Fluorene
Phenanthrene
Anthracene
HPAH
F1 uoranthene
Pyrena
Ben zo ( a ) anthracene
Chyrsene
Total benzof 1 uoranthenas
Benzo(a)pyrene
Indena(l,2,3,c,d)pyrene
D1benzo(a,h)anthracene
8enzo(g,h,1 )perylene
Total PCBs
Hexachl orobenzene
Hexachl orbut ad 1 one
1 , 3-d1 chl orobenzene
1 , 4-d1 chl orobenzene
4, 4 '-DDT
4,4'-DDE
4, 4 '-ODD
Aldrln
Dlaldrin
gamna-HCH
Phenol
4-Mathyl phenol
A
A
A
A
A
A
A
A
A
A
A
A
A
C
B
8
A
A
A
A
A
A
A
A
A
A
B
A
B
C
C
B
B
C
C
C
C
A
Point Sources
Industrial11
C.CA.LS.OR
C,OR,LS,(S)
CP,C,(M)
F.CP.(S)
P,C,CP,OR,CA,LS,(M),(L),(S)
C.OC.CA.QR
CA.B.OC.CA.OR
C,CA,OC,(M)
(CP)
C,OC,CA,OR,LS,(H)
CP,C,(F),(H)
L,(M)
L,P
L
L
L
L
L
L,(M)
L
L
L
L
L
L
L
L
L
OC
OC,IC,OR,P,L,LS
(P)
CSOsc
A
A
A
A
A
A
A
A
A
A
A
A
A
B
B
A
A
B
A
A
A
A
A
B
A
B
B
A
C
C
B
B
C
C
C
A
Nonpol nt
Sources
UR.IR
UR
UR
UR
UR.IR.GW
UR.IR.SW
UR.iR.ey
UR
UR.IR.6W
UR
UR
UR
UR
UR
UR
UR
UR
UR
IR
IR
AR
AR
AR
AR
AR
UR.AR
UR.IR
Spills8
OS
OS
C
OS
OS
C,OS
OS
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
C
C
C-7
-------
TABLE C-3. (Continued)
Pentaohlorophenol B
01 benzofuran
2-Methoxyphenol
2-Mathylnaphthal ene A
N-n1trosod1phenylamina C
Trlehloroethene A
Tetrachl oroethene A
Ethyl benzene A
Chloroform A
2,3,7,8-Tetrachl orodlox1 n
Organotin
P,OC,IC,L
L
(P)
P,OC,CA,(OC5
P,OC,IC,CA,(OC)
UR,!R
A QW
A QW
A 0
A
a Municipal
A » Chemical occurs 1n >25 percent of samples from Puget Sound municipal discharges.
B Chemical occurs 1n <25 percent of samples from Puget Sound municipal discharges,
C » Chemical not detected based on available Information,
Blanks Indicate that there are Insufficient data to categorize,
° Industrial: Industries In which chemical may be found.
S Ship building/repair.
P - Pulp mills.
C Copper smelters.
CP » Chrome plating, silver plating.
F = Ferro, silicon, chrome Industries.
CA » Chi oral kali plants,
B Bleach plant,
L Log/wood treatment facility,
OC Organic chemical manufacturing.
1C * Inorganic chemical manufacturing.
LS « Log sort yards.
M » Primary production of farrous and nonferrous metals.
OR = 011 refining.
DC - Dry-cleaning.
Codes In parentheses Indicate Industries that are potential sources but have not been documented In Puget
Sound.
Blanks Indicate that there are Insufficient data to categorize.
0 CSOs
A » Chemical occurs In »25 percent of samples from Puget Sound CSOs.
B => Chemical occurs In <25 percent of samples from Puget Sound CSOs.
C - Chemical not detected based on available Information.
Blanks Indicate that there ara Insufficient data to categorize.
Nonpolnt Sources: Types of nonpolnt sources where chemical may be found.
UR » Urban runoff
AR " Agricultural runoff
IR - Industrial runoff
GU *> Groundwater
Blanks Indicate that there are Insufficient data to categorlza.
8 Spills: Kinds of spills where chemical may be found,
0 - 011 spills.
C » Miscellaneous product spills.
OS « Ore spills.
Blanks Indicate that there are Insufficient data to categorize.
C-8
------- |