Puget Sound Estu
ELLIOTT BAY
REVISED ACTION PROGRAM:
Storm Drain Monitoring
Approach
DRAFT REPORT
March 1988
Prepared by
Tetra Tech, Inc.
Prepared for
U.S. Environmental Protection Agency
Region X - Office of Puget Sound
Seattle, Washington
-------
TC 3338-23
Task 5
Draft 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
March 1988
Tetra Tech, Inc.
11820 Northup Way, Suite 100
Bellevue, Washington 98005
-------
CONTENTS
Page
LIST OF FIGURES iii
LIST OF TABLES iv
SECTION 1. INTRODUCTION 1
SECTION 2. BACKGROUND 4
MONITORING STORMWATER RUNOFF 6
SECTION 3. PRELIMINARY INVESTIGATION 9
SECTION 4. PHASE ONE - INITIAL SCREENING 12
SELECTION OF STORM DRAINS 12
SAMPLE COLLECTION 15
IDENTIFYING AND RANKING PROBLEM STORM DRAINS 36
SECTION 5. PHASE TWO - CONTAMINANT TRACING 59
SELECTION OF SAMPLING STATIONS 59
INTERPRETATION OF SEDIMENT CHEMISTRY DATA 63
ADDITIONAL INVESTIGATIONS 66
SAMPLE COLLECTION 68
SECTION 6. PHASE THREE - CONFIRMATION 70
DISCHARGE MONITORING TECHNIQUES 72
SAMPLE COLLECTION 74
REFERENCES 90
APPENDIX 1. STORM DRAIN MONITORING APPROACH COSTS
APPENDIX 2. SUMMARY OF PREVIOUS STORM DRAIN INVESTIGATIONS
APPENDIX 3. POLLUTANTS OF CONCERN
i i
-------
FIGURES
Number Pace
1 Overview of storm drain monitoring approach 3
2 Decision criteria to select storm drains for phase one
initial screening 13
3 Example of station location and sample log form 19
4 Example of summary sampling log 20
5 Example of chain-of-custody record 21
6 Example of sample analysis request form 22
7 Decision criteria for selecting problem chemicals and
ranking problem storm drains 37
8 Schematic of a hypothetical storm drain system 61
2-1 Metals concentration in sediments collected from the
Lander Street drains
2-2 In-line sediment data for stations on SW Florida Street CSO/SD
2-3 Summary of PCB data for Slip 4 drains
2-4 Metro sampling stations on Fox Street CSO/SD (#116)
2-5 Sampling stations in Denny Way CSO source toxicant
investigation
i i i
-------
TABLES
Number Page
1 Traffic-related sources of roadway pollution 5
2 List of equipment needed for storm drain sediment sampling 17
3 Extractable organic compounds recommended for analysis
during phase one screening 24
4 Limits of detection for sediment and water by instrument 27
5 Sample containers, preservation, and recommended holding
times for sediment samples 28
6 Puget Sound AET 39
7 Freshwater sediment criteria 42
8a Summary of metals measured in street dust samples collected
from Seattle and Bellevue 43
8b Summary of organic compounds measured in street dust samples
collected from Seattle and Bellevue 44
9a Summary of metal concentrations in sediments from
Puget Sound reference areas 47
9b Summary of organic compound concentrations in sediments from
Puget Sound reference areas 48
10a Summary of metal concentrations in sediments from Carr Inlet
reference area 53
10b Summary of organic compound concentrations in sediments from
Carr Inlet reference area 54
11 List of equipment needed for storm drain discharging sampling 76
12 Recommended methods, sample containers, preservation, and
holding times for water sample analysis 81
13 Volatile organic compounds recommended for analysis of
discharge samples 82
14 Summary of available water quality criteria 86
iv
-------
1-1 Summary of analytical costs
1-2 Approximate personnel costs for field sampling - sediment
1-3 Approximate costs for sampling equipment - sediment
1-4 Approximate personnel costs for field sampling - discharge
1-5 Approximate costs for sampling equipment - discharge
2-1 Description of drains discharging into Slip 4
2-2 Summary of metals concentrations in sediment samples from
Fox Street CSO/SD #116 and surrounding area (mg/kg)
2-3 Summary of metal concentrations in sediments collected
from storm drains discharging into Lake Union
v
-------
SECTION 1. INTRODUCTION
The 1987 Puqet 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 pollution 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.
In this report, the following phased approach to conducting storm drain
investigations is recommended:
o Preliminary Investigation: Compile available information to
define the storm drain system, drainage basin characteristics,
and conditions in the receiving environment
o Phase One - Initial Screening: Collect in-line sediment
samples from near the mouths of major storm drains to
identify contaminated drainage systems
o Phase Two - Contaminant Tracing: Select problem drains for
further intensive inspection and conduct sampling activities
to trace contaminants and identify the ultimate source(s) of
contami nation
o Phase Three - Confirmation: Confirm contaminant contributions
from individual sources and pinpoint sources by collecting
1
-------
water samples from side connections that discharge into the
storm drain.
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 two procedures.
Study of these small basins could possibly bypass phase two and directly
implement phase three. This situation will likely occur in simple drainage
networks that serve a limit the number of potential sources. For larger
basins, phase one and phase two efforts are designed to limit the size of
the area investigated by eliminating non-contaminated 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. In Section 3, the process
for preliminary investigations is explained. In Section 4, 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. In Section 6, confirmation of contaminant sources
to storm drain systems by further sampling is explained. Potential costs of
the storm drain monitoring approach are outlined in Appendix 1 of this
report.
2
-------
NO
FURTHER
INVESTIGA-
TION
NO
ARE SEDIMENTS IN DRAIN
CONTAMINATED9
YES
IS STORM DRAIN SYSTEM
COMPLEX?
NO
YES
PHASE TWO-
CONTAMINANT TRACING
NO
DO CONCENTRATION GRADIENTS
IN STORM DRAIN SEDIMENTS
INOCATE SOURCE(S)?
YES
YES
NO
DOES SITE INVESTIGATION
CONFIRM SOURCES?
PHASE THREE-
CONFIRMATION
YES
DOES DISCHARGE
MONITORING CONFIRM ONGOING
CONTAMINANT CONTRIBUTIONS?
YES
! NO
NO ONGOING
SOURCE • REMOVE
CONTAMINATED
SEDIMENTS
t FROM DRAIN j
f INITIATE \
SOURCE CONTROU
ACTIVITIES AND
REMOVE
CONTAMINATED
STORM DRAIN J
V sediments/
PRELIMINARY INVESTIGATION
(ALL STORM ORAINS)
COLLECT SEDIMENT SAMPLES
FROM STORM DRAIN SIDE
CONNECTIONS
PHASE ONE-
INITIAL SCREENING
SELECT DRAINS FOR
INITIAL SCREENING
IS CONTAMI NAT ION FOUND
IN SIDE CONNECTIONS?
Figure 1. Overview of storm drain monitoring approach.
3
-------
SECTION 2. 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
(i.e., industrial, commercial, and residential uses in urban areas; agri-
cultural and silvicultural uses in rural areas). Sources of pollutants in
urban stormwater runoff can be categorized as follows:
o Atmospheric deposition (e.g., industrial stack emissions,
uncovered material storage areas, unpaved roads and parking
lots, construction and demolition sites, and landfill
operations)
o Traffic emissions (see Table 1)
o Chemical spills
o Waste and chemical storage and handling practices
o Refuse deposition in streets
o Urban erosion
o Road deicing.
Stormwater 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
4
-------
TABLE 1. TRAFFIC-RELATED SOURCES OF ROADWAY POLLUTION
Pol 1utant
Traffic Related Source
Asbestos
Copper
Chromi um
Lead
Nickel
Phosphorous
Zinc
Clutch plates, brake linings
Thrust bearing, brushings, and brake
1inings
Metal plating, rocker arms, crankshafts,
rings, brake linings, and pavement
materials
Leaded gasoline, motor oil transmission
fluid, babbit metal bearings
Brake linigs and pavement material
Motor oil
Motor oil and tires
Reference: Krenkel and Novotny 1980.
5
-------
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, the U.S. Environmental Protection
Agency (EPA) in 1978 initiated the National Urban Runoff Program (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).
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, and 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, storm water 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 storm drains in
the Puget Sound area because many drains serving metropolitan areas along
6
-------
the sound are tidally influenced. Consequently, sampling must be scheduled
during periods of low tide to avoid tidal interferences. 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.
An alternate method of sampling storm drains has been developed to
avoid the complications of storm water 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 storm water monitoring. First, sediment samples are simply collected
from the storm drain system during dry (i.e., non-rainfall) conditions and no
coordination with rainfall events is required. This makes 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 particular component of stormwater runoff. Sediments deposit in low
energy areas of the storm drain system, accumulating through successive
storms, and are probably flushed out of the system only during intense storm
events. Therefore, they generally provide a composite of several storm
events and can be used to characterize historical contamination in storm
drain lines. As in the case of discharge sampling, sediment sampling is
scheduled during low tide to enable entry to the manhole or drain line for
sample collection.
As with discharge sampling, there are disadvantages to sediment
sampling. First, sediment data cannot be used to calculate pollutant
loadings (measured in lb/day) from the storm drain system. Information on
pollutant loadings is often used as 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
7
-------
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 (Galvin and Moore 1982). Third, like discharge
sampling, 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
envi ronment.
It should be emphasized, however, that storm drain sediment sampling is
intended as a screening tool and has been used by the Municipality of
Metropolitan Seattle (Metro) and the City of Seattle to trace contaminants
in storm drain lines (see Appendix 2). 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.
8
-------
SECTION 3. 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 one for contaminants. The following are major activities to be
conducted during the preliminary investigation:
o Review city utility plans to determine location and layout of
storm drain systems
o Conduct shoreline reconnaissance to verify outfall locations and to
identify unmapped outfalls
o Contact private property owners to obtain storm drain maps
o Trace drainage basin boundaries for each storm drain system,
determine land use characteristics, and determine potential
pollutant sources in each drainage basin
o Compile and review available 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.
9
-------
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 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 one 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
10
-------
program offices maintain lists of permitted facilities and potential
hazardous sites in their region. CERCLIS, a list of Superfund sites in
Region X can be obtained from the Comprehensive Environmental Response,
Compensation, and Liability Act (CERCLA) program office. 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. Washington Department of Ecology (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 state maintains a list of all businesses
by address and Standard Industrial Code (SIC) for tax purposes. A list for
specific areas can be purchased from the state.
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.
11
-------
SECTION 4. PHASE ONE - INITIAL SCREENING
Phase one is designed to initially screen major 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 on only problem (i.e., contaminated) drains.
This procedure is expected to minimize the amount of sampling required by
eliminating non-contaminated storm drains early in the investigation. Phase
one 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.
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 exists, the storm drain system immediately
qualifies for phase two contaminant tracing (Figure 2). If both of these
situations do not exist, the following criteria should be considered:
o Average annual discharge from the storm drain
o Land use characteristics in the drainage basin
o Sensitivity of offshore environment.
12
-------
YES
NO
YES
NO
YES
DOES DRAIN DISCHARGE
LARGE VOLUME OF RUNOFF
TO RECEIVING ENVIRONMENT?
YES
IS DRAINAGE BASIN HIGHLY
DEVELOPED/INDUSTRIALIZED?
NO
NO
YES
NO
HIGH PRIORITY STORM DRAINS
SELECT DRAIN FOR CHEMICAL
CONTAMINANT SCREENING
LOW PRIORITY STORM DRAINS
TARGET DRAIN FOR FUTURE
SCREENING BASED ON
AVAILABILITY OF FUNDING
DO SENSITIVE HABITATS
EXIST IN RECEIVING ENVIRONMENT?
KNOWN CONTAMINATION
IN OFFSHORE SEDIMENTS?
KNOWN CONTAMINANT
DISCHARGES IN DRAINAGE BASIN?
Figure 2. Decision criteria to select storm drains for phase one
initial screening.
13
-------
If available data reveal contamination in offshore sediments that cannot
be attributed to a specific point source (e.g., chemical spill, industrial
discharge), then the storm drain(s) that discharge nearby should automat-
ically be selected for phase one 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 use
exist in the offshore environment, the small drains should be targeted for
14
-------
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
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
criteria in selecting storm drains for phase one (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. Non-industrialized,
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 exist in the offshore environment, a
high priority should be assigned to storm drains serving residential and
undeveloped areas.
SAMPLE COLLECTION
Samp!i no
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-
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
col lection.
15
-------
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) :
o 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.
o 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.
o 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
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.
o 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 "Chemical and Physical Analyses," Section 4) should be
placed in a pre-cleaned stainless steel bucket and brought to
the surface. Document sampling location(s) with a map
16
-------
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, ziploc
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
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
First aid kit
Safety harness and rope
CIipboard
Tide tables
Self-contained breathing apparatus
(SCBA) equipment
17
-------
showing where the sediment sample was collected (e.g., near
discharge pipe in manhole or influent line to manhole).
o If insufficient sediment is found at the proposed sampling
station, select an alternate station farther upstream in the
drain line.
o Thoroughly homogenize the sediment sample in a bucket prior
to filling the sample bottles. 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).
o 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.
o At the end of each day, complete a chain-of-custody record
(Figure 5) and the sample analysis request form (Figure 6)
for al1 samples.
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 non-standardized protocols. The data generated using these
non-standardized protocols were acceptable for individual project objectives,
but the differences in protocols limited comparability of data between
studies. PSEP formulated a compendium of recommended methods (Tetra Tech
18
-------
STORM DRAIN SAMPLING
Station Location and Sample Log
DATE TIME
STATION
LOCATION
METER 02 HNu/OVA
READINGS COMB. GAS 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.
19
-------
SUMMARY SAMPLING LOG
SURVEY:
SAMPLING
DATE
STATION
SAMPLER
HORIZON
SAMPLE
NUMBER
SAMPLES COLLECTED
i ORG
5
u.
a
VOL
AI-HC
VOL
IONS
FIELD
RECORDER: . ORG. CODE: DATE:
Figure 4. Example of summary sampling log.
20
-------
~p* fICLO SAMPLE DATA AND CHAIN OF CUSTODY SHEET
~ Enlorcanwtt/Cuatody Samptare:
PioractCoda Account: ~ Poaaibte To ilc/Hazardous Noiaa:
Nama/Location: ~ Ora ConfidantM
P'oiact Officer ~ Data for Storat Racontar
u
C M
DO
oo
V»tJ
MATRIX
i
51
IT!
i
m
is
¦ - U|
wmo*
5T1TOH
NUMftf*
BATF
—n»R$li1f15NLv
5TJTRS5
otscBimoN
£
t
j
J
3
CNOMtfG oatc
5
Yr
M
•m
Yr
k
lo
Oy
Tlnw
Mo
Oy
Tfcn*
-
--
-
-
-
-
ma
NUMtfH
KFh
1
!
6Sl
MtTO
CO
OA
cooc
rtm
OCG
c
CNOCTVTV
umto/em
MtSCCLLANfOUS
cham of cusToor necono
Vi
m
Sm
-
-
mMKumumw-rn—
mw >i ¦ BmL*
~
-
UUiMblv
¦Mn«_ unW
Hmiyiwuiu unlrwi
-
iBWh^iid ¦*
NHI »««L® AftAt I
-
1
Figure 5. Example of chain-of-custody record.
-------
SAMPLE ANALYSIS REQUEST
PACKING LIST
PROJECT:
SAMPLING OATE(S):
SHIP TO:
FOR LAB USE ONLY
SAMPLING CONTACT:
DATE SHIPPED:
DATE SAMPLES RECEIVED:
(name)
TASK NAME/CODE:
ATTN:
RECEIVED BY:
(phone)
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.
22
-------
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 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:
o Extractable organic compounds (Table 3)
o Metals.
In addition, the following conventional variables are recommended for
analysi s:
o Total solids
o Total organic carbon
o Oil and grease
o Particle size.
The PSEP protocols provide two levels of analysis for extractable
organics: screening and low level. The differences in the level of
analysis are most evident in the detection limits achieved. Detection
limits for the screening level analysis are 500-1000 ppb (dry weight) for
23
-------
TABLE 3. EXTRACTABLE ORGANIC COMPOUNDS RECOMMENDED
FOR ANALYSIS DURING PHASE ONE SCREENING
Acid Extractables
Phenols
phenol
2-methylphenol
4-methylphenol
2,4-dimethylphenol
4,6-dinitro-2-methylphenol
Substituted phenols
2-chlorophenol
2,4-dichlorophenol
4-chloro-3-methylphenol
2,4,6-trichlorophenol
2,4,5-trichlorophenol
pentachlorophenol
2-nitrophenol
2,4,-dinitrophenol
4-nitrophenol
4,6-dinitro-2-methylphenol
Base/Neutral Extractables
Low molecular weight aromatics
naphthalene
acenaphthylene
acenaphthene
fluorene
phenanthrene
anthracene
High molecular weight aromatics
fluoranthene
pyrene
benzo(a)anthracene
chrysene
benzo(b)fluoranthene
benzo(k)fluoranthene
benzo(a)pyrene
indeno(l,2,3-c,d)pyrene
di benzo(a,h)anthracene
benzo(g,h,i)pery1ene
Halogenated ethers
bis(2-chloroethyl)ether
bis(2-chloroisopropyl)ether
bis(2-ch1oroethoxy)methane
4-chlorophenyl phenyl ether
4-bromophenyl phenyl ether
Phthalates
dimethylphthalate
diethylphthalate
di-n-butylphthalate
butyl benzylphthalate
bis(2-ethylhexyl)phthalate
di-n-octylphthalate
Chlorinated aromatic hydrocarbons Miscellaneous oxygenated compounds
1.3-dichlorobenzene isophorone
1.4-dichlorobenzene benzyl alcohol
1,2-dichlorobenzene benzoic acid
1,2,4-trichlorobenzene dibenzofuran
2-chloronaphthalene
hexachlorobenzene
24
-------
TABLE 3. (Continued)
Base/Neutral Extractables (continued)
Organonitrogen compounds
ani1i ne
ni trobenzene
N-nitroso-di-n-propyl amine
4-chloroaniline
2-nitroani1ine
3-nitroaniline
4-nitroani1ine
2,6-dinitrotoluene
2,4-dinitrotoluene
N-nitrosodiphenyl amine
benzidine
3,3'-dichlorobenzidine
N-ni trosodi phenyl ami ne
benzidine
3,3'-dichlorobenzidine
Chlorinated aliphatic hydrocarbons
hexachloroethane
hexachlorobutadiene
hexachlorocyclopentadi ene
Substituted aromatics
2-methylnaphthalene
Pesticides PCBs
p,p1-DDE
Aroclor
1016
p,p'-DDD
Aroclor
1221
p,p1-DDT
Aroclor
1242
aldrin
Aroclor
1248
dieldrin
Aroclor
1254
chlordane
Aroclor
1254
alpha-endosulfan
Aroclor
1260
beta-endosulfan
endosulfan sulfate
endrin
endrin aldehyde
heptachlor
heptachlor epoxide
alpha-hexachlorocyclohexan (HCH)
beta-HCH
delta-HCH
gamma-HCH (Lindane)
toxaphene
25
-------
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 one 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. Total organic carbon is commonly used to normalize contaminant
concentrations to the amount of organic carbon. 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 is analyzed so that contaminant
concentrations can be normalized to the percent fines. Normalization of the
contaminant concentration to the total organic carbon or particle size can
be a useful tool for tracing distribution and fate of contaminants.
The analytical methods for PSEP specifies 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 one. Specified containers are adequate for collection
of the sediment sample plus an amount of sample sufficient for quality
assurance/quality control (QA/QC) samples. Prior to collecting any samples,
the laboratory performing the analyses should be consulted to confirm that
the volumes of sediment collected will be sufficient for the requested
analysis and any QA/QC samples.
26
-------
TABLE 4. LIMITS OF DETECTION FOR SEDIMENT AND WATER BY INSTRUMENT3
Sediment*5
Waterc
ICP
GFAA
ICP
DFAA
GFAA
Antimony
3.2
0.1
0.032
0.2
0.003
Arsenic
--
0.1
0.053
--
0.001
Cadmi urn
4.0
0.1
0.004
0.005
0.0001
Copper
0.6
0.1
0.006
0.02
0.001
Iron
0.7
--
NAd
NA
NA
Lead
4.2
0.1
0.042
0.1
0.001
Mercury
O
o
1—»
(CVAA)
0.0002 (CVAA)
Manganese
2.0
--
NA
NA
NA
Nickel
1.5
0.1
0.015
0.04
0.001
Si 1ver
0.7
0.1
0.002
0.01
0.0002
Zinc
0.2
0.2
0.007
0.005
0.00005
a ICP = Inductively coupled plasma atomic emission spectroscopy.
GFAA = Graphite furnace atomic absorption.
DFAA = Direct flame atomic absorption.
CVAA = Cold vapor atomic absorption.
b ICP data are from Tetra Tech (1984); GFAA 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 digest.
c DFAA and GFAA data are from U.S. EPA (1938b); ICP data are from U.S. EPA
(1984). Values are mg/L.
d NA = Not applicable; iron and manganese are used as natural tracers for
sediments only.
27
-------
TABLE 5. SAMPLE CONTAINERS, PRESERVATION, AND
RECOMMENDED HOLDING TIMES FOR SEDIMENT SAMPLES
Var iables
Mi nimum
Sample Size3
Samp 1e
Container
Preservat ion
and HandI ing
Ho 1d i no
Times"
Semivolat i 1e
organ i cs
50-100 g
16-oz glass jar,
PTFEC-1 ined 1 id
Cool (4° C),
or Freeze
7 days/40 days
1 yr°
Met al s
50 g
8-oz 1 inear poly-
ethylene or boro-
silicate gl ass,
PTFE-1Ined lid
Cool (4° C),
or Freeze
6 mo (Hg 28 days)
6 mo (Hg 28 days)
Total sol Ids,
Total organic
carbon
75 g
8-oz glass or
polyethylene jar
Freeze
6 mo11
Oil and grease
100 g
4-oz glass jar,
PTFE-lined lid
Cool (4° C),
HCI, or
Freeze
28 days'1
6 mo**
Particle size
100-150®
Glass or plastic
jar, or sealable
plastic bag
(approx. 16-oz)
Cool (4° C)
6 mo1*
* 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 adjust ed accordingly.
b 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.
c PTFE = PoIytetraf Iuoroethy I ene.
11 This is a suggested holding time. No U.S. EPA criteria exist for the preservation
of this variable.
' Large grain size samples (I.e., sand) require a larger sample size than silty
samp I es.
28
-------
Prepared sample containers can be obtained through commercial sources
or from the laboratory performing the analyses. Sample containers should be
documented by the supplier as to cleanliness, 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 headspace 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 instrumental analysis) of 40 days have been
established for water samples and have also been recommended for extractable
organic compounds in sediment (Tetra Tech 1986d).
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
adhered to during performance of the project, comparison of the procedures
and results with QA/QC goals can be made to determine data reliability.
29
-------
Sampling programs regulated by U.S. EPA and Washington Department of
Ecology (Ecology) require preparation of a Quality Assurance Project Plan
(QAPP). To ensure compliance with U.S. EPA QA/QC requirements, preparation
of a QAPP is recommended. 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 samples collected in the field include the following:
o Field replicates
o Field rinsate blanks
o Standard reference materials.
Field replicate samples are used to determine sample variability (i.e.,
analytical plus field variability). To collect field replicate samples, 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 replicates should be labeled consistently with other
samples and submitted blind to the laboratory (i.e., the laboratory should
not know the samples are duplicates). One of the replicates can be
designated as an analytical duplicate so that a comparison of field and
laboratory variability can be made. One set of blind field replicate
samples can also be analyzed by a different laboratory to evaluate laboratory
accuracy.
Field rinsate blanks are used to assess potential contamination of
samples during collection and equipment decontamination procedures. A field
30
-------
rinsate blank is collected by pouring analyte-free water (i.e., distilled)
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 blank serves to check
effectiveness of decontamination procedures, and whether contamination
occurred from field sources, or during shipping, storage, and analysis.
The analyte-free water should be stored in a sample container that accom-
panies samples and sample containers through all stages of sampling,
shipping, storage, and analysis.
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 considerably to
project costs. 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.
Standard reference materials (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
analysi s.
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). 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
31
-------
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.
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:
o 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.
o Wash sampling equipment (e.g., spoons, buckets, shovels) with
laboratory grade detergent solution (i.e., Alconox) and rinse
with water. Detergent and rinse water can be disposed in a
nearby sanitary sewer.
o 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.
32
-------
o A final rinse with distilled water is also recommended.
o Outer gloves worn by field personnel should be changed
between each station to prevent cross-contamination of
samples.
Documentati on
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:
o Date and time of starting work
o Names of field task leader and team members
o Purpose of proposed work effort
o Description of sampling station locations, including map
reference
o Details of work effort, particularly any deviation from the
proposed procedures
o Field observations
o 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
33
-------
observations. Each photograph should be documented with the following
information:
o Date and time
o Name of photographer
o Description of station location
o General direction faced and description of the subject
o 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.
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):
o Place sample container in a 2-mi1 thick (or thicker) polyethy-
lene bag, one sample per bag. Position identification tag so
it can be read through the bag. Seal the bag.
o Place sealed bags inside a strong outside container, such as
a lined metal picnic cooler or a Department of Transportation
(DOT)-approved fiber board box. The outside container should
be lined with a polyethylene bag. Surround the sample
containers with noncombustible cushioning material for
stability during transport.
o Seal the large polyethylene liner bag.
34
-------
o Place the laboratory and sampling paperwork in a large
envelope and tape it to the inside lid of the shipping
container.
o 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 fiber board 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 container: "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 (Figure 5) and sample analysis request form
(Figure 6).
35
-------
Schedulina
It is recommended that phase one screening be conducted during a dry
period when rainfall will not greatly affect sediment accumulations 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.
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 the phase two of the sampling program.
More intensive, phase two sampling is recommended for the high priority
storm drains to trace contaminants so that ultimate source(s) can be
identified. In addition, a ranking procedure has been developed to priori-
tize individual problem storm drains to aid in scheduling phase two. A
schematic of the decision criteria recommended for identifying high priority
storm drains is presented in Figure 7.
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-
36
-------
YES
NO
r LOW
PRIORITY FOfl
SAMPLING
r LOW
PRIORITY FOR
SAMPLING i
NO
NO
YES
YES
YES
NO
LOW 1
PRIORITY FOR
SAMPLING
NO
YES
CALCULATE EARs
CALCULATE LOADING INDEX
ARE CHEWICALS ON
POLLUTANT-OF-CONCERN LIST''
DO CONCENTRATIONS IN STORM
DRAIN SEDIMENTS RANK IN
90th PERCENTILES
DO CONCENTRATIONS IN STORM
DRAIN SEDIMENTS EXCEED AET
OR OTHER SEDIMENT CRITERIA?
IDENTIFY PROBLEM CHEMICALS
IN STORM DRAINS
DO CONCENTRATIONS IN STORM
DRAIN SEDIMENTS EXCEED
STREET OUST LEVELS?
RANK PROBELEM DRAINS FOR
CHEMICALS EXCEEDING AET
AND 90th PERCENTILE
SELECT HIGHEST PRIORITY
DRAINS FOR CONTAMINANT
TRACING SAMPLING
AREAETs OR OTHER SEDIMENT
CRITERIA AVAILABLE''
(NEFFet al., 1986;
WISCONSIN DNR, 1985)
Figure 7. Decision criteria for selecting problem chemicals and
ranking problem storm drains.
37
-------
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 indentification 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 leen updated to include
new environmental data sets (Tetra Tech 1987). AET values are based on
sediment chemistry, toxicity (i.e., amphipod, oyster larva, and Microtox
bioassays), and benthic infauna abundance data. For a given chemical and a
specified biological indicator, the AET is the concentration above which
statistically significant biological effects occurred in all samples of
sediments analyzed. 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 contamination 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
concentrations 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).
38
-------
TABLE 6. PUGET SOUND AET
(UG/KG DRY WEIGHT FOR ORGANIC COMPOUNDS;
MG/KG DRY WEIGHT FOR METALS)
Lowest AET Highest AET
LPAHa
5,200
6,100
Naphthalene
2,100
2,400
Acenaphthylene
560
640
Acenaphthene
500
980
Fluorene
540
1,800
Phenanthrene
1,500
5,400
Anthracene
960
1,900
HPAHb
12,000
>51,000
Fluoranthene
1,700
9,800
Pyrene
2,600
11,000
Benzo(a)anthracene
1,300
4,500
Chrysene
1,400
6,700
Benzofluoranthenes
3,200
8,000
Benzo(a)pyrene
1,600
6,800
Indeno(l, 2,3-c,d)pyrene
600
>5,200
Dibenzo(a,h)anthracene
230
1,200
Benzo(g,h,ijperylene
670
5,400
Total PCBs
130
2,500
Total Chlorinated Benzenes
170
680
1,3-Dichlorobenzene
>170
>170
1,4-Dichlorobenzene
110
260
1,2-Dichlorobenzene
35
>350
1,2,4-Trichlorobenzene
31
64
Hexachlorobenzene
70
230
Hexach1orobutadi ene
120
290
Total Phthalates
3,300
>70,000
Dimethyl phthalate
71
700
Diethyl phthalate
>48
1,200
Di-n-butyl phthalate
1,400
>5,100
Butyl benzyl phthalate
63
>470
Bis(2-ethylhexyl) phthalate
1,900
>3,100
39
-------
TABLE 6. (Continued)
Lowest AET Highest AET
Pesticides
p,p1-DDE
9
15
p,p1-DDD
2
43
p,p'-DDT
3.9
11
Phenols
Phenol
420
1,200
2-Methylphenol
63
>72
4-Methylphenol
670
1,200
2,4-Dimethyl phenol
29
>72
Pentachlorophenol
>140
>140
2-Methoxyphenol
930
930
Miscellaneous Extractables
Hexachlorobutadiene
120
290
1-Methylphenanthrene
310
370
2-Methyl naphthalene
670
670
Biphenyl
260
270
Dibenzothiophene
240
250
Dibenzofuran
540
540
Benzyl alcohol
57
73
Benzoic acid
650
>690
n-Nitrosodiphenyl amine
40
220
Volatile Oraanic ComDounds
Tetrachloroethene
140
>210
Ethyl benzene
33
>50
Total xylenes
100
>160
Metals
Antimony
3.2
26
Arsenic
85
700
Cadmium
5.8
9
Copper
310
800
Lead
300
700
Mercury
0.41
2
Nickel
28
>120
Silver
>0.56
5
Zinc
260
1,600
® LPAH = Low molecular weight polycyclic aromatic hydrocarbons.
" HPAH = High molecular weight polycyclic aromatic hydrocarbons.
Reference: Tetra Tech (1987).
40
-------
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 is recommended to
evaluate contamination levels in storm drain sediments. Using this method,
storm drain sediments having a chemical concentration within the highest 10
percent of all 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 percentile ranking. Representative street dust contaminant levels for
urban areas are presented in Tables 8a and b. 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
may be too stringent for these contaminants. Therefore, it is recommended
41
-------
TABLE 7. FRESHWATER SEDIMENT CRITERIA
Metals3
(mg/kg)
Arsenic
10
Cadmium
1.0
Chromium
100
Copper
100
Lead
50
Mercury
0.10
Nickel
100
Zinc
100
Organic Compounds*5
(ug/kg)
Heptachlor epoxide
8
Chlordane
9.8
Dieldrin
21
PC8s
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 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.
42
-------
TABLE 8a.
SUMMARY OF METALS MEASURED IN STREET
COLLECTED FROM SEATTLE AND BELLEVUE3
DUST SAMPLES
Range
Meana
Detection
Chemical
(mg/kg dry wt)
(mg/kg dry wt)
Frequency
Antimony
<1-2.0
1.1
8/12
Arsenic
11-39
25
12/12
Beryl 1i um
0.17-0.34
0.26
12/12
Cadmium
0.6-2.0
1.0
12/12
Chromi um
20-230
97
12/12
Copper
31-260
93
12/12
Lead
90-1300
520
12/12
Mercury
0.02-0.18
0.07
9/12
Ni ckel
20-44
32
12/12
Selenium
<0.6-<3
2
0.12
Si 1ver
0.01-0.5
0.32
6/12
Thai 1i um
<0.2-0.34
0.6
3/12
Zinc
110-970
310
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).
43
-------
TABLE 8b. SUMMARY OF ORGANIC COMPOUNDS
MEASURED IN STREET DUST SAMPLES COLLECTED FROM
SEATTLE AND BELLEVUE3
Mean^
Ranged
Detection
Chemical
(mg/kg)
(mg/kg)
Frequency
Pesticides
A1 pha-Hexachlorocyclohexane
0.014
0.010-0.018
2/14
Gamma-Hexachlorocyclohexane
0.025
0.006-0.043
2/14
DDD
0.005
0.005
1/14
Heptachlor
0.048
0.048
1/14
Haloaenate AliDhatics
Trichloromethane
0.007
0.004-0.015
4/14
Tetrachloroethane
0.024
0.016-0.032
2/14
1,1,1-Trichloroethane
0.013
0.012-0.016
3/14
4-Chlorophenyl phenyl ether
0.24
0.24
1/14
Monocyclic Aromatic Hydrocarbons
Benzene
0.021
0.01-0.032
2/14
Hexachlorobenzene
2.0
2.0
1/14
Ethylbenzene
0.012
0.005-0.025
3/14
Toluene
0.009
0.004-0.019
4/14
Nitrosodimethylamine
0.76
0.76
1/14
Phenolics
Phenol
0.21
0.08-0.47
4/14
Pentachlorophenol
1.76
0.12-3.4
2/14
2,4-Dimethylphenol
0.02
0.01-0.03
2/14
4-Nitrophenol
0.11
0.11
1/14
Phthalate Esters
Dimethyl phthalate
0.78
0.78
1/14
Diethyl phthalate
0.41
0.16-0.89
3/14
Di-n-butyl phthalate
0.70
0.22-2.4
7/14
Di-n-octyl phthalate
0.54
0.23-0.97
4/14
Butyl benzyl phthalate
6.2
0.22-0.35
7/14
Bis-2-ethylhexyl ph 11
38
2.4-90
9/14
44
-------
TABLE 8b. (Continued)
Chemi cal
Meanb
(mg/kg)
Ranged
(mg/kg)
Deletion
Frequency
LPAHC
Acenaphthalene
Anthracene
F1 uorene
Phenanthrene
0.21
0.35
0.23
1.5
0.16-0.25
0.1-0.6
0.2-0.25
0.18-2.4
2/14
5/14
2/14
14/14
HPAHd
Fluoranthene
Pyrene
Chrysene
Benzo(a)pyrene
Benzo(k)fluoranthene
Benzo(a)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).
45
-------
that the average urban street dust concentrations, rather than the AET
values, be used to assess phthalates and PAH.
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 two contaminant
tracing (Figure 7). If chemicals in these drains are on the pollutant of
concern list (Appendix 3), additional sampling under phase two 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.
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 two 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
Puget Sound. These data (Tables 9a and b) are assumed to provide a
46
-------
TABLE 9a. SUMMARY OF METAL CONCENTRATIONS IN SEDIMENTS
FROM PUGET SOUND REFERENCE AREAS
Range
Detecti on
Reference
Chemical
(mg/kg dry wt)
Frequency
Sites3
Antimony
U0.1b-2.79
16/36
1,2,3,4,7,8,9,10
Arsenic
1.9-17
38/38
1,2,3,4,7,8,9,10
Cadmium
0.1-1.9
28/28
1,2,3,4,6,9,10
Chromi um
9.6-255
42/42
1-10
Copper
5-74
32/32
1,2,3,4,5,6,9,10
Lead
U0.1-24
25/32
1,2,3,4,5,6,9,10
Mercury
0.01-0.28
42/42
1-10
Nickel
4-140
30/30
1,2,3,4,5,9,10
Selenium
U0.1-1.0
18/28
1,2,3,4,6,9,10
Si 1ver
U0.02-3.3
28/30
1,2,3,4,5,9,10
Zinc
15-102
30/30
1,2,3,4,5,9,10
a Reference sites:
1. Carr Inlet
2. Samish Bay
3. Dabob Bay
4. Case Inlet 7.
5. Port Madison 8.
6. Port Susan 9.
10.
Nisqually Delta
Hood Canal
Sequim Bay
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).
Mai ins et al. (1980).
Mai ins et al. (1982).
Crecelius et al. (1975).
Crecelius et al. (1975).
Battelle (1985).
(Site 10) Tetra Tech (unpublished)
47
-------
TABLE 9b. SUMMARY OF ORGANIC COMPOUND CONCENTRATIONS
IN SEDIMENTS FROM PUGET SOUND REFERENCE AREAS
Range
(ug/kg Detection Reference
Substance3 dry wt) Frequency Sites"
Phenols
65 Phenol
U3.3-62c»d
8/20
1,2,3,10
HSL 2-Methylphenol
U10
0/11
1,10
HSL 4-Methylphenol
U2-290
7/11
1,10
34 2,4-Dimethylphenol
U1-U14
0/13
1,10
Substituted Phenols
24 2-Chlorophenol
U0.5-U500
0/13
1,10
31 2,4-Dichlorophenol
U0.5-50
0/13
1,10
22 4-Chloro-3-methylpheno1
U0.5-50
0/13
1,10
21 2,4,6-Trichlorophenol
U0.5-U100
0/13
1,10
HSL 2,4,5-Trichlorophenol
U10-U100
0/11
1,10
64 Pentachlorophenol
0.1-U1000
0/13
1,10
57 2-Nitrophenol
0.1-U50
1/9
1,10
59 2,4-Dinitrophenol
U0.5-U50
0/9
1,10
60 4,6-Dinitro-o-cresol
U0.5-U100
0/9
1,10
58 4-Nitrophenol
U0.5-U100
0/9
1,10
Low Molecular Weiaht Aromatic
Hydrocarbons
55 Naphthalene
U0.5-U40
14/27
1-6,10
77 Acenaphthylene
U0.1-U40
2/27
1-6,10
1 Acenaphthene
U0.1-U40
4/27
1-6,10
80 Fluorene
U0.1-U40
7/28
1-7,10
81 Phenanthrene
4-170
18/24
1,2,3,6,7,10
78 Anthracene
U0.5-U40
11/24
1,2,3,6,7,10
HSL 2-Methylnaphthalene
0.3-U22
10/17
1,4,5,6,10
Hi ah Molecular Weiaht Aromatic
Hvdrocarbons
39 Fluroanthene
5-100
24/29
1-7,10
84 Pyrene
5-120
23/29
1-7,10
72 Benzo(a)anthracene
2-U40
15/22
1,2,3,6,7,10
76 Chrysene
4-U40
15/22
1,2,3,6,7,10
74 Benzo(b)f1uoranthene
U5-94
15/25
1-7,10
75 Benzo(k)fluoranthene
4.8-94
15/25
1-7,10
48
-------
TABLE 9b. (Continued)
Range
(ug/kg
Detection
Reference
Substance3
dry wt)
Frequency
Sitesb
Hiah Molecular Weiaht Aromatic Hvdrocarbons (Continued)
73 Benzo(a)pyrene
U0.37-40
16/21
1,3,4,5,6,7,10
83 Indeno(l,2,3-c,d)pyrene
U0.37-30
10/19
1,4,5,6,7,10
82 Dibenzo(a,h)anthracene
0.4-U13
3/12
1,10
79 Benzo(g,h,i)perylene
1.2-20
8/13
1,7,10
Chlorinated Aromatic Hvdrocarbons
26 1,3-Oichlorobenzene
U0.06-U160
1/25
1,2,3,4,5,10
27 1,4-Dichlorobenzene
U0.06-U120
1/25
1,2,3,4,5,10
25 1,2-Dichlorobenzene
U0.06-65
1/25
1,2,3,4,5,10
8 1,2,4-Trichlorobenzene
U0.5-U190
0/13
1,10
20 2-Chloronaphthalene
U0.5-U50
0/13
1,10
9 Hexachlorobenzene (HCB)
0.01-U100
6/19
1,4,5,6,10
Chlorinated AliDhatic Hvdrocarbons
12 Hexachloroethane
U0.5-U.50
0/9
1,10
xx Trichlorobutadiene
U0.03-U25
5/12
1,4,5,6
xx Tetrachlorobutadiene
Isomers
U0.04-U25
5/12
1,4,5,6
xx Pentachlorobutadiene
Isomers
0.03-U400
5/19
1,4,5,6,10
52 Hexachlorobutadiene
U0.03-U25
0.07-8.5
1,4,5,6
53 Hexachlorocyclopentadiene
U200
0/3
10
Haloaenated Ethers
18 Bis(2-chloroethyl)ether
0.3-U20
1/9
1,10
42 Bis(2-chloroisopropy1)ether
U0.5-U10
0/9
1,10
43 Bis(2-chloroethoxy)methane
U10
0/9
1,10
40 4-Chlorophenyl phenyl ether
U0.5-U10
0/9
1,10
41 4-Bromophenyl phenyl ether
U0.5-U10
0/9
1,10
Phthalate Esters
71 Dimethyl phthalate
U0.5-U50
1/12
1,10
70 Diethyl phthalate
9.0-11
4/8
1,10
68 Di-n-butyl phthalate
U20-760
6/8
1,10
67 Butyl benzyl phthalate
U0.5-U25
3/12
1,10
66 Bis(2-ethylhexyl) phthalate
U0.5-58
3/8
1,10
69 Di-n-octyl phthalate
U0.5-U56
5/12
1,10
49
-------
TABLE 9b. (Continued)
Range
(ug/kg
Detection
Reference
Substance3
dry wt)
Frequency
Sites"
Miscellaneous Oxvaenate 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-Tetrach1orodi benzo-p-
dioxin Not Analyzed
HSL Dibenzofuran
U5-14
4/11
1,10
Oraanonitroaen ComDounds
HSL Aniline
U1.0-U20
0/6
1
56 Nitrobenzene
U0.5-U10
0/8
1,10
63 n-Nitroso-di-n-propylamine
U0.5-U10
0/8
1,10
HSL 4-Chloroani1ine
U10-U50
0/7
1,10
HSL 2-Nitroani1ine
U10-U50
0/7
1,10
HSL 3-Nitroaniline
U50
0/7
1,10
HSL 4-Nitroani1ine
U50
0/7
1,10
36 2,6-Dinitrotoluene
U0.5-U10
0/8
1,10
35 2,4-Dinitrotoluene
U.05-U10
0/8
1,10
62 n-Nitrosodiphenylamine
U0.5-U10
0/8
1,10
37 1,2-Diphenylhydrazine
U0.5-U5
0/6
1
5 Benzidine (4,4'-diaminobiphenyl)
U0.5
0/2
1
28 3,3'-Dichlorobenzidine
U0.5-U100
0/9
1,10
Pesticides
93 p,p1-DDE
U1.0-U25
0/12
1,10
94 p, p1 -DDD
U1.0-U25
0/13
1,10
92 p,p'-DDT
U1.0-U25
0/12
1,10
89 Aldrin
U0.5-U25
0/13
1,10
90 Dieldrin
U1.0-U25
0/13
1,10
91 Chlordane
U5.0-U50
0/13
1,10
95 Alpha-endosulfan
U.5-U25
0/8
1,10
96 Beta-endosulfan
U1.0-U25
0/8
1,10
97 Endosulfan sulfate
U1.0-U25
0/8
1,10
98 Endrin
U1.0-U25
0/13
1,10
99 Endrin aldehyde
U2.3-U25
0/9
1,10
100 Heptachlor
U0.5-U50
0/13
1,10
101 Heptachlor epoxide
U0.5-U25
0/9
1,10
102 Alpha-HCH
U0.5-U50
0/13
1,10
103 Beta-HCH
U0.5-U50
0/13
1,10
50
-------
TABLE 9b. (Continued)
Substance3
Range
(ug/kg
dry wt)
Detection
Frequency
Reference
Sites'*
Pesticides (Continued)
104 Delta-HCH
105 Gamma-HCH (lindane)
113 Toxaphene
PCBs
xx Total PCBs (primarily
1254/1260)
Volatile 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. EPA priority pollutant number. TCL indicates
Target Compound List.
b Reference sites: 1. Carr Inlet 4. Case Inlet 7. Nisqually Delta
2. Samish Bay 5. Port Madison 10. Port Susan
3. Dabob Bay 6. Port Susan
c An anomalously high phenol value of 1800 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) Tetra Tech (1985b); Mowrer et al. (1977)
(Site 2) Battelle (1985).
(Site 3) Battelle (1985); Prahl and Carpenter (1979).
(Site 4) Mai ins et al. (1980); Mowrer et al. (1977).
(Site 5) Mai ins et al. (1980).
(Site 6) Mai ins et al. (1982).
(Site 7) Barrick and Prahl (1987); Mowrer et al. (1977).
(Site 10) Tetra Tech (unpublished).
51
-------
reasonable measure of the variability in contaminant concentration for
relatively uncontaminated sediments, but are expected to represent fairly
conservative levels of contaminant concentration for storm drain sediments.
In previous Puget Sound studies (Tetra Tech 1985a,c,d), EARs were calculated
based only on six Carr Inlet reference stations (Tables 10a and b). Only
the Carr Inlet data, rather than the full range of Puget Sound reference
area data, are used in ranking storm drain data for the following reasons:
o 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
o The lowest reference detection limits for most substances of
concern in Puget Sound embayments are available for Carr Inlet
o 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 is possible
o 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 representat-
ive of Puget Sound reference conditions.
EARs 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.
Loading indices are the second method for ranking problem storm drains.
It will not be possible to calculate true discharge loading values for most
52
-------
TABLE 10a. SUMMARY OF METAL CONCENTRATIONS IN
SEDIMENTS FROM CARR INLET REFERENCE AREA
Range Mean3 Detection
Chemical
(mg/kg dry wt)
(mg/kg dry wt)
Freque
Antimony
U0.1-0.14
0.11
4/6
Arsenic
2.4-3.8
3.4
6/6
Cadmium
0.29-1.5
0.95
6/6
Chromium
9.6-24.4
15
6/6
Copper
4.9-8.0
6.4
6/6
Lead
4.4-13
9.2
6/6
Mercury
0.01-0.098
0.04
6/6
Nickel
11-27.6
17
6/6
Selenium
U0.1-U1
0.7
0/6
Si 1ver
0.02-0.12
0.09
2/6
Zinc
15-24.1
19
6/6
a Mean calculated using the reported detection limit for undetected values.
Reference: Tetra Tech (1985b).
53
-------
TABLE 10b. SUMMARY OF ORGANIC COMPOUND CONCENTRATIONS
IN SEDIMENTS FROM CARR INLET REFERENCE AREA
Range
Mean^
(ug/kg
(ug/kg
Detection
Substance3
dry wt)
dry wt)
Frequency
Phenols
65 Phenol
U10-62c»d
33
3/13
HSL 2-Methylphenol
U1-U10
7.0
0/6
HSL 4-Methylphenol
U10-32
13
2/6
34 2,4-Dimethylphenol
U1-U10
6.8
0/6
Substituted Phenols
24 2-Chlorophenol
U0.5-U5
3.5
0/6
31 2,4-Dichlorophenol
U0.5-U10
6.8
0/6
22 4-Chloro-3-methylphenol
U0.5-U10
6.8
0/6
21 2,4,6-Trichlorophenol
U0.5-U10
6.8
0/6
HSL 2,4,5-Trichlorophenol
U10
10
0/4
64 Pentachlorophenol
0.1-U50
33
1/6
57 2-Nitrophenol
0.1-U10
6.8
1/6
59 2,4-Dinitrophenol
U0.5
0.5
0/2
60 4,6-Dinitro-o-cresol
U0.5-U100
67
0/6
58 4-Nitrophenol
U0.5-U100
67
0/6
Low Molecular Weiaht Aromatic Hydrocarbons
55 Naphthalene
1-13
6.8
3/5
77 Acenaphthylene
U0.5-U5
4.1
0/5
1 Acenaphthene
U0.5-U5
4.1
0/5
80 Fluorene
U0.5-U5
4.1
0/5
81 Phenanthrene
5-16
13
5/5
78 Anthracene
3-22
9.1
4/5
HSL 2-Methylnaphthalene
U1-U5
4.2
0/5
Hi ah Molecular Weiaht Aromatic Hvdrocarbons
39 Fluroanthene
11-20
15.4
5/5
84 Pyrene
11-18
14.4
5/5
72 Benzo(a)anthracene
U5-8
8.0
4/5
76 Chrysene
U5-19
10.8
4/5
74 Benzo(b)fluoranthene
3
3
1/1
75 Benzo(k)fluoranthene
5
5
1/1
54
-------
TABLE 10b. (Continued)
Range
Meanb
Substance3
(ug/kg
(ug/kg
Detection
dry wt)
dry wt)
Frequency
Hiah Molecular Weiaht Aromatic Hydrocarbons (Continued)
73 Benzo(a)pyrene
3-7.1
5.7
3/5
83 Indeno(l,2,3-c,d)pyrene
4-U5
4.8
1/5
82 Dibenzo(a,h)anthracene
0.4-U5
4.1
1/5
79 Benzo(g,h,i)perylene
3-U5
4.6
1/5
Chlorinated Aromatic Hydrocarbons
26 1,3-Dichlorobenzene
U0.5-U5
3.5
0/6
27 1,4-Dichlorobenzene
U0.5-U5
3.5
0/6
25 1,2-Dichlorobenzene
U0.5-U5
3.5
0/6
8 1,2,4-Trichlorobenzene
U0.5-U5
3.5
0/6
20 2-Chloronaphthalene
U0.5-U5
3.5
0/6
9 Hexachlorobenzene (HCB)
U0.5-U10
6.8
0/6
Chlorinated Aliphatic Hydrocarbons
12 Hexachloroethane
U0.5-U50
34
0/6
xx Trichlorobutadiene
U0.5-U25
15
0/6
xx Tetrachlorobutadiene isomers
U0.5-U25
15
0/6
xx Pentachlorobutadiene isomers
U0.5-U25
15
0/6
52 Hexachlorobutadiene
U0.5-U25
17
0/6
53 Hexachlorocyclopentadiene
U0.5
0.5
0/1
Haloaenated Ethers
18 Bis(2-chloroethyl)ether
0.3-U10
6.8
1/6
42 Bis(2-chloroisopropyl)ether
U0.5-U10
6.8
0/6
43 Bis(2-chloroethoxy)methane
U10
10
0/6
40 4-Chlorophenyl phenyl ether
U0.5-U5
3.5
0/6
41 4-Bromophenyl phenyl ether
U0.5-U5
3.5
0/6
Phthalate Esters
71 Dimethyl phthalate
U0.5-U50
40
0/5
70 Diethyl phthalate
9.0-11
11
4/5
68 Di-n-butyl phthalate
U20-760
170
3/5
67 Dutyl benzyl phthalate
U0.5-U25
17
0/5
66 Bis(2-ethylhexyl) phthalate
U0.5-U25
17
0/5
69 Di-n-octyl phthalate
U0.5-U25
20
0/5
55
-------
TABLE 10b. (Continued)
Range
Mean^
(ug/kg
(ug/kg
Detection
Substance3
dry wt)
dry wt)
Frequency
Miscellaneous Oxvaenate Compounds
54 Isophorone
U0.5-U25
20
0/5
HSL Benzyl alcohol
U10
10
0/4
HSL Benzoic acid
U25-430
140
3/4
129 2,3,7,8-Tetrachlorodibenzo-p-
di oxi n
U5
5
0/2
HSL Dibenzofuran
U5
3.7
0/4
Oraanonitroaen ComDOunds
HSL Aniline
U1.0-U20
14
0/6
56 Nitrobenzene
U0.5-U5
4.1
0/5
63 n-Nitroso-di-n-propylamine
U0.5-U10
8.1
0/5
HSL 4-Chloroaniline
U50
50
0/4
HSL 2-Nitroani1ine
U50
50
0/4
HSL 3-Nitroaniline
U50
50
0/4
HSL 4-Nitroaniline
U50
50
0/4
36 2,6-Dinitrotoluene
U0.5-U10
8.1
0/5
35 2,4-Dinitrotoluene
U0.5-U5
4.1
0/5
62 n-Nitrosodiphenylamine
U0.5-U5
4.1
0/5
37 1,2-Diphenylhydrazine
U0.5-U5
4.1
0/6
5 Benzidine (4,4'-diaminobiphenyl)
U0.5
0.5
0/2
28 3,3'-Dichlorobenzidine
U0.5-U100
67
0/6
Pesticides
93 p,p'-DDE
U10-U25
10e
0/5
94 p.p'-DDD
U10-U25
10e
0/6
92 p,p1-DDT
U10-U25
10e
0/5
89 Aldrin
U10-U25
10e
0/6
90 Dieldrin
U10-U25
10e
0/6
91 Chlordane
U10-U25
10e
0/6
95 Alpha-endosulfan
U10-U25
10e
0/5
96 Beta-endosulfan
U10-U25
10e
0/5
97 Endosulfan sulfate
U10-U25
10e
0/5
98 Endrin
U10-U25
10e
0/6
99 Endrin aldehyde
U10-U25
10e
0/5
100 Heptachlor
U10-U25
10e
0/6
101 Heptachlor epoxide
U10-U25
10e
0/6
102 Alpha-HCH
U10-U25
10e
0/6
103 Beta-HCH
U10-U25
10e
0/6
56
-------
TABLE 10b. (Continued)
Substance3
Range
(ug/kg
dry wt)
Mean**
(ug/kg
dry wt)
Detection
Frequency
Pesticides (Continued)
104 Delta-HCH
105 Gamma-HCH (lindane)
113 Toxaphene
U10-U25
U10-U25
U10
10e
10e
10e
0/6
0/6
0/2
PCBs
xx Total PCBs (primarily
1254/1260)
<4.3-U7
6
2/6
Volatile ComDounds
85 Tetrachloroethene
38 Ethyl benzene
__f
— —
a Number indicates U.S. EPA priority pollutant number. TCL indicates
Target Compound List.
b Mean calculated using the reported detection limit for undetected values.
c An anomalously high phenol value of 1800 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 -- = Not analyzed.
Reference: Tetra Tech (1985b).
57
-------
storm drains because the necessary flow and water quality data are generally
not available. However, sediment data collected during phase one 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 "Selection
of Storm Drains," Section 4) 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 two. 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.
58
-------
SECTION 5. PHASE TWO - 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 one screening.
When this is done, source identification efforts can be focused on contamin-
ated sections of storm drain lines while uncontaminated sections can be
eliminated from further study. To trace contaminants to the sources, addi-
tional field sampling and continued investigation of land use in the
drainage basin will be required. Phase two will entail collecting additional
sediment samples from manholes throughout the storm drain system to trace
contaminants in the problem storm drains. The phase two sampling effort
will focus on problem chemicals and associated source categories identified
during phase one and the preliminary investigation. The phase two 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 source(s) 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
investigation 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 contamin-
ant tracing program in problem storm drain systems.
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
59
-------
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 non-contaminated
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):
o Isolate subbasins with different land use characteristics
o Determine contaminant gradients along major trunk lines, if
possible
o 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 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:
60
-------
17
15
>
A RESIDENTIAL
COMMUNITY i i
—L f 16
JO.
18
14
6 4 < •"
B COMMERCIAL DISTRICT
~ ^
(f
~*" H
> > 1
20
HLCFIVING WATER
D. HIGHWAY RUNOFF
¦a?
A
19
LEGEND
•
MANHOLE
<
FLOW DIRECTION
> O 10
C. INDUSTRIAL COMPLEX
iJELl,
4~] .
1. ..b
12 13
11 <
POTENTIAL
CONTAMINANT
SOURCE
(e g., abandoned landfill,
chemical storage area,
maintenance shop, etc.)
(DRAWING NOT TO SCALE)
Figure 8. Schematic of a hypothetical storm drain system.
-------
o Subbasin A - Residential Community
o Subbasin B - Commercial District
o Subbasin C - Industrial Complex
o 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 1ine.
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 one. A recent study conducted as part of
Puget Sound Estuary Program (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 developed for the report that linked chemicals
62
-------
with specific source categories and industry types. This information,
provided in Appendix 3, 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 source(s) 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 source(s). 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 line > branch line > side connection > catch basin > source.
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 (AQPP), 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,
assessment the following:
o Method detection limits
o Holding times for analyses
63
-------
o Documentation and chain-of-custody procedures
o Frequency of QA/QC sample checks
o Contamination of field and laboratory blanks by problem
chemicals
o Control limits for laboratory replicate and matrix spike
results
o Control limits for blind field replicate results
o Control limits for Standard Reference Material (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 semi-quantitative 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 because of the greater
absorption capacity of organic matter and fine particulate. To account for
these sample characteristics, data can be normalized to organic carbon
64
-------
content by dividing the contaminant concentration by the total percent fines
obtained from the particle size data.
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 from each drain are not comparable, then data should be
normalized prior to the contaminant concentration comparisons. Further phase
two 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, and 2) the criteria
used to identify problem chemicals (i.e., AET values, proposed freshwater
sediment criteria, 90th percentile, or street dust levels) are at least 5
times greater than the analytical detection limits.
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 foil owing two steps are recommended:
o The concentration of the problem chemical in the sediments
from the upstream and downstream station must be at least 5
times greater than the method detection limit to ensure that
concentrations are in the quantifiable range of the method
65
-------
o The concentration of the problem chemical in the sediments
from the upstream station must be outside the confidence
limits (based on the QA review) for the concentration
reported in the downstream station.
An example of the latter point follows. Assume that the confidence limits
for the downstream station are + 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 control
limits from the QA assessment can be used for the data comparisons.
However, the data are qualified, confidence limits must be established case
by case.
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.
ADDITIONAL INVESTIGATIONS
In some cases, additional investigative activities will be required to
complete phase two and the source identification process. The following
66
-------
additional activities are recommended to support the contaminant tracing
program:
o Distribute questionnaires to businesses in the problem
drainage basin to obtain information on current operations
o 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
o 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). 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:
o Water use and volume (e.g., restroom, rinsing, cooling,
product manufacturing, floor cleaning, washdown)
o Types of connections to the storm drain system (e.g., catch
basins, floor drains, sumps)
o Types of chemicals used or stored onsite
o Type of business (e.g., product manufacture or service).
Inspections of industries will provide detailed information on possible
contaminant sources in the drainage basin. In addition, inspections can be
67
-------
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.
SAMPLE COLLECTION
Recommended storm drain sediment sampling procedures, decontamination
procedures, documentation, sample packaging and shipping requirements, and
scheduling are as described in "Sample Collection," Section 4. It is
recommended that chemicals analyzed for during the contaminant tracing
program be those identified during phase one. Chemical analyses and quality
assurance/quality control (QA/QC) recommendations for phase two are presented
in the following sections.
Chemical and Physical Analyses
Analysis of sediment samples for phase two contaminant tracing should
include, at a minimum, the problem chemicals identified during phase one
screening and the preliminary investigation. If a particular compound or
class of compounds was not detected during phase one screening, or is not
indicated as important during the preliminary investigation, a smaller
number of variables may be analyzed than in phase one 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.
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
68
-------
(e.g., total solids, total organic carbon, and grain size) is recommended
during the contaminant tracing effort to permit comparison to sediment
samples collected during phase one screening.
Quality Assurance/Qua!itv Control
Collection of field QA/QC samples specified for the phase one screening
is also appropriate for phase two contaminant tracing. A detailed discussion
of field QA/QC samples and collection procedures is presented in "Sample
Collection," Section 4, and should be followed during phase two 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 two, 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.
69
-------
SECTION 6. PHASE THREE - CONFIRMATION
The information obtained from phase one screening and phase two
contaminant tracing, combined with the supporting evidence from the site
inspections, is expected to provide sufficient evidence to identify con-
taminant sources for many of the 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 three will require that samples be collected from the actual dischar-
ge^) 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:
o To distinguish between historical and ongoing source contri-
butions
o To confirm sources where volatile organic compounds are
suspected as the major contaminant
o To determine compliance for NPDES-permitted sources
o To document source contaminant loading conditions for
enforcement cases.
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, the adverse effects
on the receiving environment may be reduced by simply cleaning the storm
70
-------
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.
Volatile organic compounds have not been recommended for analysis
during phase one screening and phase two contaminant tracing because
available data indicate that volatile organic compounds are not frequently
detected in storm drain sediments. As part of the Puget Sound Estuary
Program (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 collection. Detection
frequencies for the volatile organic compounds in these samples ranged from
0 to 40 percent. Compounds detected most frequently included trans-1,2-
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 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.
Therefore, to determine compliance with permit conditions, it will be
necessary to monitor a plant's effluent. Although NPDES-permitted facilities
are required to monitor their effluent, toxic contaminants are not usually
included in the variables measured. Therefore, additional discharge
monitoring will be required to confirm the contaminant contributions from
these potential sources.
71
-------
Enforcement cases typically require detailed information on the
source(s), including the type of discharge (i.e., stormwater runoff from
property, process water, illegal discharge), type of contaminants and their
concentration 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 contaminating storm drains should be sufficient to document
contaminant problems associated with stormwater runoff. However, confirma-
tion 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.
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. Storm water 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.
The following sections provide general recommendations for discharge
sampling in the storm drain system. The discharge monitoring program is
designed to complete the source identification process begun during phases
one and two by measuring concentration of contaminants and flow rate in
discharges to the storm drain system. Included in the recommendations are a
general discussion of sample collection procedures, analytical requirements,
and data interpretation techniques.
An important issue in designing a discharge sampling plan is whether to
use bulk water only, or the particulate fraction of the discharge for
chemical analyses. Separate analysis of bulk water and the particulate
72
-------
fraction is normally recommended 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 bulk water only. Separate collection and
analysis of the particulate fraction concentrates contaminants adsorbed to
solids which improves quantification of the contaminants. The following
section provides guidance on selecting between bulk water only and particu-
late fraction analysis.
Bulk Water Vs. Particulate Fraction Analysis
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. By separating and collecting the particulate in the
discharge, contaminants adsorbed to the particulate can be more readily
quantified. If high concentrations of contaminants are expected in the bulk
water samples (i.e., greater than 5 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 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 particulate 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
73
-------
runoff may contain greater than 1,000 mg/L suspended material during an
intense rainfall event. If a discharge consists primarily of cooling water
containing minimal particulate, 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/min. Manual collection 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.
SAMPLE COLLECTION
Sampling conducted during phase three confirmation should follow the
same equipment decontamination, documentation, sample packaging, and
shipping procedures recommended for phase one screening (see "Sample
Collection," Section 4).
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.
74
-------
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
"Chemical Analysis," Section 6) 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 11. General guidance on collecting
discharge samples for bulk water chemical analysis is provided below:
o 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 "Sample Collection,"
Section 4. 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.
o Set equipment to collect samples for the appropriate time
interval (e.g., 12 h for process waste streams). Insure that
sample collection bottles in the automatic sampler contain the
appropriate preservatives (see "Chemical Analyses," Section
75
-------
TABLE 11. LIST OF EQUIPMENT NEEDED FOR STORM DRAIN
DISCHARGING 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, ziploc
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
Kni fe
Sample tray
Kimwipes
Gloves (leather and chemical
resistant)
Coveralls (cotton and chemical
resistant)
Respi ratorsa
(including extra filters)
Waders (two pair)3
Duct tape3
Op/combustible gas meter and tubing3
Pnotoionization detector (PID)a
meter and tubing3
Drager tubes/bellows3
Decontamination sprayer3
Brushes (for decontamination)3
Alconox
First aid kit
Safety harness and rope3
CIipboard
Tide tables
Self-contained breathing apparatus
(SCBA) equipment3
pH meter
Flow meter
Continuous flow centrifuge"
Pump/tubing^
Filtration equipment"
3 Required if personnel must enter manhole to install sampling equipment,
b Required for collection of particulate material.
76
-------
6). Beginning and ending times should be recorded. Sampling
equipment should be checked periodically during the sampling
period to ensure that it is functioning properly.
o 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.
o 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.
o 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.
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
77
-------
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 waste stream 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 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 "Bulk Water Sampling," Section 6. 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.
78
-------
Much larger sample volumes are generally required to obtain low
detection limits for organic compounds. As explained in the example
presented in "Bulk Water Vs. Particulate Fraction Analysis," 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.
Schedulinq
Scheduling requirements for phase three 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.
Chemical Analyses
Chemical analyses of discharge samples for phase three confirmation
should include problem chemicals identified in the storm drain sediment
samples collected during phase one and phase two. In addition, other
chemical compounds (e.g., volatile organic compounds) identified as potential
contaminants in process waste streams or stormwater runoff during the
79
-------
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 three confirmation are listed below:
o Metals
o Extractable organic compounds
o Volatile organic compounds
o Conventionals (i.e., pH, total suspended solids, total
dissolved solids).
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 three confirmation are presented in Table 12.
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 0.1-25 ug/L for acid/neutral compounds
and 0.002-1.0 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
13) 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
80
-------
TABLE 12. RECOMMENDED METHODS, SAMPLE CONTAINERS, PRESERVATION,
AND HOLDING TIMES FOR WATER SAMPLE ANALYSIS
Va r i able
Samp 1e
Con t a i ner
Preservat ion
and Hand 1 ing
Ho 1d i ng T ime'
Met hod'1
Reference
Semi vo 1 a t i 1 e
organi cs
2-L glass bottle;
PTFE •1 i ned cap
Keep on ice
(4° C)
7 days/40 days
Extraction,
GC/MS
U. S. EPA 1984
Pest i c i d es/PC8s
2- L glass bottle;
PTFE-1 ined cap
Keep on ice
(4° C)
7 days/40 days
Extraction,
GC/ECD
U. S. EPA 1984
Volat i 1e organ i cs
Two 40-mL glass
vials; PTFE-lined
si 1 icon septum caps
Fill, 1 eavl ng
no air space,
keep in dark
on ice (4° C)
14 days
Purge and trap,
GC/MS
U. S. EPA 1984
Meta 1 s
1-L glass or 1 inear
HN03 to pH<2
6 mo
ICP, FLAA
Tetra Tech
(total)
Total dissolved
solids, total
suspended solids
Oi I and grease
polyethylene bottle,
PTFE-1 Ined cap
2-L glass or piast i c,
PTFE-1 i ned cap
2- L glass,
PTFE-1Ined cap
Cool (4° C)
Cool (4° C),
H2S04 to pH<2
(Hg 28 days)
7 days
28 days
GFAA, CVAA
Methods 160. 1,
160. 2
1986c
U. S. EPA 1983b
Method 413. 1 U. S. EPA 1983b
or 413. 2
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.
b GC/MS = Gas chromatography/mass spectroscopy.
GC/ECD = Gas chromatography/electron capture detection.
ICP = Inductively coupled plasma atomic emission spectroscopy.
FLAA = Flame atomic absorption.
GFAA = Graphite furnace atomic absorption.
CVAA = Cold vapor atomic absorption.
c PTFE = PoIytetra11uoroethyIene.
81
-------
TABLE 13. VOLATILE ORGANIC COMPOUNDS RECOMMENDED
FOR ANALYSIS OF DISCHARGE SAMPLES
Halogenated Alkanes
ch1oromethane
bromomethane
chloroethane
methylene chloride
1,1' -dichloroetharie
chloroform
1,2-dichloroethane
1.1.1-trichloroethane
carbon tetrachloride
bromodi ch1oromethane
1,2-dichloropropane
chlorodi bromomethane
1.1.2-trichloroethane
bromoform
1,1,2,2-tetrachloroethane
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-dichloroethene
cis- and trans-
1,3-dichloropropene
trichloroethene
tetrachloroethene
Aromatic Hydrocarbons
benzene
toluene
ethyl benzene
styrene
total xylenes
Ethers
2-ch1oroethy1vinylether
Mi seellaneous
carbon disulfide
vinyl acetate
82
-------
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.
Quality Assurance/Quality Control
Field QA/QC samples that should be collected and analyzed during
discharge sampling are summarized below:
o Field replicates
o Field rinsate blanks
o Transfer blanks
o Trip blanks
o Standard reference materials.
Field replicate samples should be collected from a completely mixed
discharge composite sample. When grab samples are collected for volatile
organic compound analysis, the order of collection of the field replicate
volatile organic compound samples should be noted on the summary sampling
log (Figure 4) and in the field logbook. Analysis of a blind field replicate
by an independent laboratory can be used to assess the accuracy of results.
Field rinsate blanks, transfer blanks, and trip blanks should be
collected during discharge sampling to assess potential contamination of
samples during sample collection, shipping, storage, and analysis. Tech-
niques for collecting field rinsate blanks are discussed in "Sample
Collection," Section 4. A transfer blank is a container filled with analyte-
free water that accompanies the sample containers and samples through all
stages of sampling, shipping, storage, and analysis. The transfer blank is
opened in the field concurrently with the collection of a sample, and serves
as a check on possible contamination from field sources, shipping, storage,
83
-------
and analysis. Any preservatives used for samples should also be added to the
transfer blank to assess the potential contamination from this source. A
trip blank is a sample container with analyte-free water that is not opened
in the field. Trip blanks are used when samples for volatile organic
compounds are collected as a check on cross-contamination of samples. The
frequency of collection of field rinsate blanks, transfer blanks, and trip
blanks should be determined by the project manager prior to initiation of
the sampling effort. The overall frequency of field QA/QC sample collection
is normally 5-20 percent of the total number of field samples.
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 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.
Data Interpretation
Contaminant concentrations measured in discharge samples collected
during phase three can be compared with available water quality criteria to
evaluate potential impacts on the receiving environment. Available
freshwater and saltwater criteria (U.S. EPA 1986a) are summarized in Table
84
-------
14. These values are based on acute and chronic toxicity to aquatic life.
Although the ambient water quality criteria are not enforceable standards,
they are commonly used general guidelines for interpreting water quality
data. A discharge sample that exceeds ambient water quality criteria 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, non-exceedance of criteria for a single
sampling event does not confirm the lack of a potential source of contam-
inants. 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, it is recom-
mended that contaminant concentrations measured in NPDES-permitted discharges
be compared with the permit limitations. This 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.
85
-------
TABLE 14. SUMMARY OF AVAILABLE WATER QUALITY CRITERIA (MG/L)
Freshwater Aquatic Lifea
Acute Chronic
Toxicity Toxicity
Saltwater Aquatic Life3
Acute Chronic
Toxicity Toxicity
Metals
Antimony
Arsenic
Beryl 1ium
Cadmi um
Chromium
Copper
Lead
Mercury
Nickel
Seleni um
Silver
Thai 1ium
Zinc
Cyanide
LPAH
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
HPAH
(9,000)
360
(130)
1.8d .
980c/16d
9.2C
34c
2.4
790c
20
1.2C
(1,400)
65c
22
(2.J00)
(1(J00)
b
b
Fluoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Total benzofluoranthenes
Benzo(a)pyrene
Indeno(l,2,3-c,d)pyrene
Dibenzo(a,h)anthracene
Benzo(g,h,ijperylene
(3,
b
b
b
b
b
b
b
(1,600)
190
(5.3)
0.66cJ
120c/lld
6.5C
1.3C
0.012
88c
5.0
(0.12)
(40)
59d
5.2
(6J0)
(SgO)
b
b
b
b
b
b
b
b
b
b
b
¥
43
1,100
2.9
140
2.1
75
300
2.3
(2,130)
95
1
(2,j550)
(9J0)
b
b
b
b
b
b
b
b
b
I6
9.3
50
2.9
5.6
0.025
8.3
V
b
86
1
4.
b
b
a0) (1b6)
b
b
b
b
b
b
b
PAH Total
(300)
86
-------
TABLE 14. (Continued)
Freshwater Aquatic Lifea Saltwater Aouatic Life5
Acute Chronic Acute Chronic
Toxicity Toxicity Toxicity Toxicity
Phenols
Phenol (10,200) (2,560) (5,800) 5
2,4,5-Trichlorophenol
(10,200)
(2,560)
(2,020)
(365)
(30)
b
(2,120)
b
13e
7.9e
b
b
b
b
b
(970)
(230)
(150)
(4,g80)
(2,gOO)
(940)
(3)
b b
2,4-Dichlorophenol
4-Chloro-3-methyl
phenol
2,4-Dimethylphenol
Pentachlorophenol i3e 7.9e 13 (7.9)
2,3,5,6-Tetrachloro-
phenol u u £ (440)
b b
b b
b b
2,4,6-Trichlorophenol D (970) b b
Nitrophenols (230) (150) (4,850) b
2-Chlorophenol (4,380) (2,000) b b
4-Chlorophenol b b (29,700) b
Phthalate esters (940) (3) (2,944) (3.4)
Pesticides
Aldrin 3.0 ^ 1.3 ^
DDT 1.1 0.001 0.13 0.001
DDE (1,050) b (14) b
TDE (0,06) b (3.6) b
Demeton b 0.1 b 0.1
Dieldrin 2.5 0.0019 0.71 0.0019
Endosulfan 0.22 0.056 0.034 0.0087
Endrin 0.18 0.0023 0.Q37 0.0023
Guthion b 0.01 b 0.01
Heptachlor 0.52 0.0038 0.053 0.0036
Hexachlorocyclohexane
(Lindane) 2,0 0.06 0,16 b
Malathion b 0.1 b 0.1
Methoxychlor b 0,03 b 0.03
b n.nni b
Mirex 0.001 D 0.001
Parathion 0.065 0.013
b
Toxaphene 0.73 0.002 0.21 0.0002
PCBs 2.0 0.014 10 0.03
87
-------
TABLE 14. (Continued)
Freshwater Aquatic Lifea Saltwater Aquatic Life9
Acute Chronic Acute Chronic
Toxicity Toxicity Toxicity Toxicity
Volatiles
Acrylonitrile
Acrolei n
Benzene
Trichloromethane
(chloroform)
Tetrachloromethane
(carbon tetra-
chloride)
1,2-dichloroethane
Dichloroethylenes
Dichloropropanes
Dichloropropenes
Ethyl benzene
Halomethanes
Pentachlorinated
ethanes
Tetrachloroethanes
1,1,2,2-Tetrachloro-
ethane
Tetrachloroethy1ene
Toluene
Trichloroethanes
1.1.1-Trichloroethane
1.1.2-Trichloroethane
Trich1oroethy1ene
Misc. Oxygenated Compounds
(7,550)
(68)
(5,300)
(2,600)
(fl)
b
(55)
(5,100)
b
b
(700)
(28,900)
(1,240)
b
b
(35,200)
(118,000)
(11,600)
(23,000)
(6,060)
(32,000)
(11,000)
b
(20,000)
b
(5,700)
(244)
b
b
(50,000)
(113,000)
(224,000)
(10,300)
(790)
(430)
(12,000)
b
b
b
(3,g40)
b
(6,400)
(7,240)
(9,320)
(1.J00)
(3g0)
(2gl)
b
(5,280)
(17,500)
(18,000)
b
b
(45,000)
(2,400)
(8J0)
b
b
(9,400)
(21,900)
(9,020)
(10,200)
(6,jJ00)
(31,200)
b
(2,000)
b
(450)
(5,gOO)
b
b
b
2,3,7,8-Tetrachlorodi- .
benzo-p-dioxin (TCDD) (0.01) (0.00001) b °
Isophorone (117,000) b (12,900)
Oroanonitroaen Compounds
Benzidine (2,500)
b b b
(330) (330) (590) (370)
17,000^ b ^ o
5,850]
1,2-Diphenylhydrazine (270)
Dinitrotoluene
Nitrobenzene (27,000) £ (6,680)
Nitrosamines (5,850) b (3,300,000) ~
1 9_ninhfln\/lhwHy*a7inD (07 C\\ D D D
88
-------
TABLE 14. (Continued)
Freshwater Aquatic Life8 Saltwater Aquatic Life3
Acute Chronic Acute Chronic
Toxicity Toxicity Toxicity Toxicity
Chlorinated Aliphatic Hydrocarbons
Hexachloroethane (980) (540) (940)
Hexachlorobutadiene (90) (9.3) (32)
Hexachlorocyclopenta-
diene (7) (5.2) (7)
Ethers
Chloroalkyl ethers (238,000)
Haloethers (360) (122)
Chlorinated Aromatic Hydrocarbons
b
b
b b b
b b
Chlorinated benzenes (250) (50) (160) (129)
Chlorinated naphtha-
lenes (1,600) b (7.5) °
Dichlorobenzenes 1,120) (763) (1970) "
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.
Reference: U.S. EPA (1986a).
89
-------
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.
Galvin, 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.
Hadley, J. 1987. Wyckoff case shows how hard it is to clean up our nations
pollution. Seattle Post-Intelligencer, 2 November 1987. p. Bl, 2.
Hubbard, T. 15 March 1988. Personal Communication (phone by Ms. Beth
Schmoyer). Municipality of Metropolitan Seattle, Seattle, WA.
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.
Krenkel, P.A., and V. Novotny. 1980. Water quality management. Academic
Press, NY. p. 225.
90
-------
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. 1981. Data from sediments collected from central Puget Sound,
1979-1980. National Marine Fisheries Service, Seattle, WA. 5 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. Contam. Toxicol.
18:588-594.
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.
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.
Puget Sound Water Quality Authority. 1987. 1987 Puget Sound water quality
management plan. Puget Sound Water Quality Authority, 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. 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.
91
-------
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.
Tetra Tech. Unpublished. Puget Sound Estuary Program 1985 data from
Elliott Bay and Everett Harbor action programs. Tetra Tech, Inc., Bellevue,
WA.
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
pollutant 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.
92
-------
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 #HQ-8410-Q1.
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. 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.
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.
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.
93
-------
APPENDIX 1
STORM DRAIN MONITORING
APPROACH COSTS
-------
APPENDIX 1: STORM DRAIN MONITORING APPROACH COSTS
ANALYTICAL COSTS
A summary of the costs for analytical procedures recommended in the
storm drain monitoring approach is presented in Table 1-1. The costs
presented for each procedure can vary depending on the following factors:
o Number of samples submitted for analysis
o Sample characteristics
o Level of services provided
o Sample matrix (soil/sediment vs. water)
o Turnaround time
o 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 chromatography)
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.
-------
TABLE 1-1.
SUMMARY OF ANALYTICAL
COSTS
Vari able
Approximate Cost
Per Sample ($)
Method3
Target Compound List
Volatile organic
compounds
Water: 200-250
Sediment: 250-300
Purge & trap GC/MSb
Extractable ABNC organic
compounds
Water: 375-750
Sediment: 475-800
GC/MS
Pesticides/PCBs
Water: 135-160
Sediment: 160-200
GC/ECD
Priority pollutant metals
Water: 150-210
Sediment: 200-275
AAS, CVAA, ICP, GFAA
Total solids
10-20b
Gravimetric
Total volatile solids
25-40
Gravimetric
Total organic carbon
Water: 30-50
Sediment: 45-65
Elemental analysis
Oil and grease
Water: 40-70
Sediment: 45-65
Gravimetric, spectro-
photometry
Particle size
45-125
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 (1983b, 1984, 1987).
-------
Tabulated analytical results are often the only data a laboratory will
provide without payment of an additional fee. Quality assurance/quality
control (QA/QC) information is necessary to perform data review and
validation. Obtaining QA/QC information 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.
Analysis of sediment samples costs more than the same analysis of water
samples. Increased costs of sediment sample analysis relative to water
sample analysis are 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.
Identification of organic compounds other than priority pollutants and
Target Compound List (TCL) 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.
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.
-------
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.
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 due to sample compositing. Estimated costs for
conducting sediment versus 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 1-2. 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 Puget
Sound Estuary Program (PSEP) sampling efforts, it is estimated that
approximately 1 h will be spent at each station to complete sample collec-
tion, 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.
-------
TABLE 1-2. 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)^
Sampler (l)c
Safety and rescue (1)
Field note taker (1)
Traffic controller (1)
(if needed)
Subtotal 4 h/station 80
Travel Time** (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.
-------
Approximate rental or purchase costs for the major field sampling
equipment are summarized in Table 1-3. 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
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 1-4. 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 1-5. 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:
o Diameter of the storm drain
o Length of the drain lines that need to be cleaned
o Amount of sediment accumulation in the storm drain.
Other factors will also indirectly affect the cost of removal opera-
tions as follows:
-------
TABLE 1-3. APPROXIMATE COSTS FOR SAMPLING EQUIPMENT - SEDIMENT3
Approximate Approximate
Purchase Weekly Rental
Cost ($) Cost ($)
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 Gear
2 Respirators
Filter cartridges
1 Safety harness/rope
2 SCBA
Meters
0?/combustible gas
PID meter
Draeger bellows
H2S tubes
40 NAb
80 NA
5 NA
5 NA
50 NA
60 each NA
2.80/pair NA
3.50/pair NA
3.50/pair NA
50.00/pair NA
240 32
3.33 each c
150 12
2,600 210
1,500 120
6,000 300
200 20
3.50/tube NA
a Costs may vary depending on supplier.
& NA = Not applicable.
c Cost included in respirator rental fee.
-------
TABLE 1-4. APPROXIMATE PERSONNEL COSTS FOR FIELD SAMPLING - DISCHARGE
Estimated Time Requirements/
Sampling Event/Station
Total Person Hour
Requi rementsa
Equipment Mobilization 10 h
10
Sample Collection (13 h/station)^
Samplers (2)c 26 h/station
520
Documentation 1 h/day
20
Equipment Demobilization 10 h
10
TOTAL
560
a Based on the following assuptions: 20 sampling stations, 1 station/day.
b Time requirements per station = 13 h. This includes equipment set up and
decontamination.
c Indicates number of people.
-------
TABLE 1-5. APPROXIMATE COSTS FOR SAMPLING EQUIPMENT - DISCHARGE3
Approximate Approximate
Purchase Weekly Rental
Cost ($) Cost ($)
Sampling Equipment
1 Telescoping extension rod 50 NA^
Coolers 60 each NA
Automatic sampler/flow meter 5,000 800
pH meter 200 25
Flow meter" 2,000 300
Continuous flow centrifuge 27,000 2,500
Pump/tubing 500 75
Generator 300 45
Filtration equipment 400 60
Protective Clothing
Chemical resistant gloves
Inner gloves 2.80/pair NA
Outer gloves 3.50/pair NA
Chemical resistant coveralls 3.50/pair NA
Hip waders, 2 pair 50.00/pair NA
Protective Gear
2 Respirators 240 32
Filter cartridges 3.33 each c
1 Safety harness/rope 150 12
2 SCBA 2,600 210
Meters
Oo/combustible gas 1,500 120
PID meter 6,000 300
Draeger bellows 200 20
H2S tubes 3.50/tube NA
a
b
c
Costs may vary depending on
NA = Not applicable.
Cost included in respirator
supplier.
rental fee.
-------
o Tidal interferences
o Season that cleanup activities are conducted
o 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.
-------
Lander Street CSO/SD
Sediments in the Lander Street drain [combined sewer overflow/storm
drain (CSO/SD) #105] contained lead at concentrations as high as 35 percent.
The lead contamination was traced to atmospheric deposition and surface
runoff from the area surrounding a secondary lead smelter (see Appendix 2).
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 Vactor 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 are 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
Municipality of Metropolitan Seattle (Metro) sampled the SW Florida
Street (CSO/SD #098) drain system in 1984 and reported elevated concentra-
tions of PCBs, pentachlorophenol, arsenic, copper, and PAH. 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 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 Vactor equipment. The remaining 1,449 ft of
36-in to 48-in line was cleaned with a high pressure jet wash and the debris
was removed using Vactor equipment. In addition, all catch basins connected
to the contaminated section of the storm drain were cleaned using Vactor
equipment. All material removed by Vactor equipment was placed in three
lined settling ponds. Decant liquids from the ponds were discharged into
-------
the City of Seattle sanitary sewer system. Solids were removed from the
ponds and temporarily stored on the nearby Wyckoff property (Standifer, J.,
17 May 1985, personal communication; Schwartz, L., 1 August 1985, personal
communication; Clendaniel, B., 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 (see
Appendix 2). In November 1985, a contractor hired by Seattle City Light
removed the contaminated sediments from the flume. All visible 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 place 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 was 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.
-------
APPENDIX 2
SUMMARY OF PREVIOUS
STORM DRAIN INVESTIGATIONS
-------
APPENDIX 2: SUMMARY OF PREVIOUS
STORM DRAIN INVESTIGATIONS
Municipality of Metropolitan Seattle (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. The plan was designed to identify and control
pollution problems in the Duwamish River and was adopted by the Metro
Council in 1984. Metro received a 205(j) grant to implement part of the
plan that focused on 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 4 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.
Samples were analyzed for metals. Results, summarized in Figure 2-1, 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 percent).
-------
Pb =6300
As = 55
Cu = 73
Pb = 370,000
As =2300
Cu =690
to
01
5
sz.
3
o
w
03
3
C
a>
>
<
CO
HARBOR
ISLAND
Southwest Lander Street
Pb s 250,000
As =2500
Cu =460
Pb s 360,000
As =3600
Cu =1200
0 100 200 300 400
feet
meters
50
100
NOTE: Units - mg/kg.
Figure 2-1. Metals concentration in sediments collected from the
Lander Street drains.
-------
The source of lead was traced to stack emissions from a former lead
smelter (Metro 1987) 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 (Metro 1987).
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 (Metro 1987). When U.S. Environmental Protection
Agency (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 a concentration 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
contamination 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. Major contaminants found in the drainage system are
summarized in Figure 2-2. 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 found in wood
preservatives. Profiles of arsenic, pentachlorophenol, and HPAH concentra-
-------
PLAN VIEW
LEGEND
O— CATCH 9ASIN
LOCKHEED
SHIPBUILDING
200 400 600 800
WYCKOFF
feet
IC-01
200
100
16*05 |a,n
-------
tions along the SW Florida Street trunk line show a distinct peak approxi-
mately 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 determined
that Wyckoff was illegally discharging hazardous wastes containing arsenic,
creosote, and pentachlorophenol into a catch basin connected to the SW
Florida Street drain. 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 2-2).
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 1984. 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 (Hadley 1987).
SLIP 4 DRAINS
Elevated concentrations of PCBs have been measured in the surficial
sediments in Slip 4 (Figure 2-3). 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. Five drains discharge into Slip 4 (15 SD, Slip 4 CS0/SD
#117, Slip 4 SD, Georgetown Flume, and East Marginal Pump Station CS0 W043).
Descriptions of each drain are presented in Table 2-1. 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.
-------
South Warsaw Sireet
ISouth Willow Street
sunn**®'
S»l»>kW
. STOP"0
«s?
-------
TABLE 2-1. DESCRIPTION OF DRAINS
DISCHARGING INTO SLIP 4
Name
Outfal1
Diameter (in)
Drainage
Basin Area
(ac)
Description of
Service Area
1-5 SDa
66
30
Drains approximately 1.5 mi
of 1-5 between S. Dawson and
S. Myrtle Streets and part of
Georgetown area.
Georgetown Flume
60
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.
Slip 4 CSOb/SD #117
24
150 (SD)
74.6 (CSO)
Drain for the north end of
King County Airport.
Slip 4 SD
60
170
Serves portions of King
County Airport.
East Marginal Pump
Station CSO W043
36
318
Emergency sewer overflow for
Metro pump station.
a SD = Storm drain,
b CSO = Combined sewer overflow.
-------
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
2-3). 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. These
concentrations exceed the average level reported for urban street dust from
eignt 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 ree -ted 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
2-3). 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 2-3).
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. 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
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 this 60-in line (Smukowski,
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 2-4). 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. Sampling station locations are shown in Figure 2-4.
The results of the sampling and analyses for metals, summarized in
Table 2-2, 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 2-4) 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 2-4) 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).
Elevated metals concentrations measured in sediment samples collected
from catch basins on Marine Power and Equipment property indicate that
stormwater runoff from the property may also be a source of metals to the
Fox Avenue storm drain. 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.
Metal 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,
-------
LEGEND
STORM DRAIN
MANHOLE
RIVER OR CATCH
BASIN SEDIMENT
SAMPLING STATIONS
FROM
COMBINED
SYSTEM
SLIP NO. 3
FROM
COMBINED
SYSTEM
OVERFLOW
MANHOLE
Brighton
MH# 1
South Willow Street
South Myrtle Street
meters
Figure 2-4. Metro sampling stations on Fox Street CSO/SD (#116).
-------
TABLE 2-2. SUMMARY OF METALS CONCENTRATIONS IN SEDIMENT
SAMPLES FROM FOX STREET CSO/SD #116 AND SURROUNDING AREA (nig/kg)a
Date
Sampled
As
Cd
Cu
Pb
Zn
Fox Street CSO/SD #116 MHb#l
4/5/84
2/25/85
3/27/86
3,800
1,200
1,200
4.4
6.7
5.4
1,200
900
710
1,400
900
730
5,600
2,300
2,300
MH#2
3/27/86
110
6.2
380
620
850
Duwamish River Sediments
Upstream of Drain
Offshore of Drain
4/18/84
4/18/84
21
210
<0.3
0.5
60
290
51
150
160
1,000
Sediment Samples from
Catch Basinsc
2/25/85
1,000-
3,900d
9.5-
19
2,300-
7,600
950-
1,900
6,200-
15,000
(2,200)e
(14)
(5,000)
(1,400)
(10,000
Mean Street Dust Levels^
—
25
1.0
93
520
310
a Stations shown on Figure 2-4.
b MH = Manhole.
c Catch basins 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).
15
-------
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 to 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 National Pollutant Discharge Elimination System (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 [Washington Department of Ecology (Ecology) 1987].
DENNY WAY CSO INVESTIGATION
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 products a total
average overflow volume of 500 million gal/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 on have shown
contaminated sediments offshore from the Denny Way CSO and adverse effects
on benthic communities. As a result, the Denny Way CSO was identified in
the interim Elliott Bay Toxics Action Plan as a significant problem area
(Tetra Tech 1985c).
In 1986, Metro (Romberg et al. 1987) 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. 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
discharges and chemical use was sent to each potential source. Ninety-six
potential sources were visited by Metro inspectors to confirm the question-
naire 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 2-5) and analyzed
for metals and organic toxicants. Wastewater samples were collected for two
-------
INDUSTRIAL
LAUNDRY
Y
WO !=
Republican
and Boren
Republican
and Pontius
w
Valley
and Boren
INDUSTRIAL
LAUNDRY
w,s
WestlaKe
and Ninth
LEGEND
•
SAMPLING SITE
—
COMBINED SEWER
w
WASTEWATER SAMPLE
s
SEDIMENT SAMPLE
Tunnel MS
and Sixth
Danny Local
naga
Elliott and
Harrison
ELLIOTT BAY r
INTERCEPTOR~L_
Reference: Romberg et at., 1987.
w,s
Denny
Local
INDUSTRIAL
LAUNDRY
Tunnel
and Ninth
Denny/
Lake Union
w
Minor and
John
Melrose
and Olive
w
Westlake
and Denny
LAKE UNION
TUNNEL
DENNY WAY
REGULATOR STATION
ELLIOTT BAY
cso
OUTFALL
Figure 2-5. Sampling stations in Denny Way CSO source toxicant
investigation.
-------
different events at most stations and sediment samples were collected once
at each station.
The highest metals concentrations in both wastewater and sediment
samples were consistently measured in 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 (Romberg et al. 1987) after the pretreatment systems were installed.
Therefore, it has been determined that the contaminated sediments in the
drain were caused by historic discharges rather than ongoing discharges.
Metro is currently evaluating removal of the contaminated sediments to
prevent them from flushing into Elliott Bay. In addition, improvements to
the storm water routing program to enhance in-line storage, and a notifi-
cation and control system to reduce source toxicant discharges when overflows
occur are under consideration (Romberg et al. 1987).
A discharge of chromium and mercury was traced to a movie film
developing operation based on wastewater analyses in the Denny Way CSO. 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. 1986).
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. 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 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.
LAKE UNION AND SHIP CANAL STORM DRAIN INVESTIGATION
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 storm water 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 storm water sampling also showed that weather conditions prior to
the sampling event affected the quality of discharge. The basin sampled
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 6 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 2-3) 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 storm water volume and solids loading would be most
-------
TABLE 2-3. 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
0.74-1,700
210
11/11
Beryl 1i urn
<0.25-7.3
1.1
4/10
Cadmi urn
0.42-39
8.2
11/11
Chromium
19-350
96
11/11
Copper
22-1,300
360
11/11
Lead
210-2,700
1,000
11/11
Mercury
0.036-2.29
0.71
10/10
Nickel
21-660
190
10/10
Selenium
0.23-3.0
1.4
3/7
Si 1ver
0.54-9.6
2.7
7/7
Zinc
280-7,600
180
10/10
a Mean calculated using the reported detection limit for undetected values.
Reference: Kennedy/Jacobs/ChiIton 1987.
-------
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/Chilton 1987).
-------
APPENDIX 3
POLLUTANTS OF CONCERN
-------
TABLE 3-1. INORGANIC CONTAMINANTS OF POTENTIAL
CONCERN IN PUGET SOUND3
Antimony
Arsen icb
Cadmiumb
Chromiumc
II
Copperb
Leadb
Mercuryb
Nickel
Si Tverb
Zinc
Cyan ide
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,
Appendix A).
High selenium concentrations have been reported
in sediments in a single Puget Sound study;
these values are considered to be elevated
likely 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.
b These elements have been suggested previously
as contaminants of concern in Puget Sound based
on elevated sediment concentrations, bioaccumulation
potential, or toxicity (see Konasewich et al. 1982;
Jones and Stokes 1984).
c Although not found at elevated concentrations
in Puget Sound sediments, chromium may be of
concern in localized areas where chromium-rich
waste are being 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.
-------
TABLE 3-2. ORGANIC CONTAMINANTS OF POTENTIAL
CONCERN IN PUGET SOUND
Phenols
65a phenolc
HSLb 2-methylphenoic
HSL 4-methylphenolc
34 2,4-dimethylphenol
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 pentachlorophenold
57 2-nitrophenol
59 2,4-dinitrophenoie
60 4,6-dinitro-o-cresole
Miscellaneous Organic Acids (guaiacol s/resin acids)
2-methoxyphenol (guaiacol)
3.4.5-trichloroguaiacol
4.5.6-tr ichloroguaiacol
tetrachloroguaiacol
mono- and di- chlorodehydroabietic acids
Low Molecular Weight Aromatic Hydrocarbonsd
55 naphthalene 80 fluorene
77 acenaphthylene 81 phenanthrene
1 acenaphthene 78 anthracene
Alkylated Low Molecular Weight Aromatic Hydrocarbonsd,g
HSL 2-methylnaphthal ene
1-methylnaphthalene
1-, 2-, and 3-methyl phenanthrenes
High Molecular Weight PAH
39 fluoranthene
84 pyrene
72 benzo(a)anthracene
76 chrysene
74 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
-------
TABLE 3-2. (Continued)
Chlorinated Aromatic Hydrocarbons
26 1,3-dichlorobenzene 8 1,2,4-trichlorobenzene
27 1,4-d ich1oroben zene 20 2-chl oronaphthalene
25 1,2-dichlorobenzene 9 hexachlorobenzene (HCB)
Chlorinated Aliphatic Hydrocarbons
12 hexachloroethane
52 hexachlorobutadiened
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 alcoholi polychlorinated dibenzodioxinsj
HSL benzoic acidi
HSL dibenzofurani
Organonitrogen Compoundsk
62 N-nitrosodiphenylamine
9(H)- carbazolel
Pesticides
93 p,p'-DDEd«i 98 endrind
94 p,p'-DDDdm 100 heptachlor
92 p ,p'-DDTdm 102 alpha-HCH
89 aldrindn 103 beta-HCH
90 dieldrind 104 delta-HCH
91 alpha-chlordane 105 gamma-HCH (lindane)
PCBsn
Total PCBs (this class includes monochloro-
through decachlorobiphenyls)
-------
TABLE 3-2. (Continued)
Volatile Halogenated
A1kaneso
45
chloromethane
6
carbon tetrach1 or idee
46
bromometnane
48
bromodichloromethanee
16
chloroethanee
32
1,2-dichloropropane
44
dichloromethane
51
chlorodibromomethanee
13
1,1'-dichloroethane
14
1,1,2-trichloroethane
23
chloroform
47
bromoforme
10
1,2-dichloroethanee
15
1,1,2,2-tetrachloroethanee
11
1,1,1-tr ichloroethanee
Volatile Halogenated AlkenesO
88 vinyl chloride
29 1,1'-dichloroethene
30 trans-1,2-dichloroethene
33 cis-1,3-dichloropropene
trans-1,3-dichloropropene
87 trichloroethene
85 tetrachloroethene
Volatile Aromatic and Chlorinated Aromatic Hydrocarbonso
4 benzene HSL styrene (ethenylbenzene)
86 toluene HSL total xylenes
38 ethyl benzene 7 chlorobenzene
NOTE: Compounds not recommended from the U.S. EPA priority pollutant
list include:
o Halogenated ethers (two volatile and five semivolatile compounds)
are rarely reported in Puget Sound and are not expected
to persist in sediments.
o 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.
o Acrolein and acryl onitril e have not been detected in Puget
Sound sediments and are difficult to analyze for in routine
volatiles analysis.
o 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.
-------
TABLE 3-2. (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-Methyl phenol 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.
d Compound or group of compounds has been designated previously as a contaminant
of concern in Puget Sound based on elevated sediment concentrations, bioaccu-
mulation potential, or toxicity (Jones and Stokes 1984, Konasewich et al. 1982,
Quinlan et al. 1985).
e Compound is seldom or not reported, possibly due to analytical problems
presented by the compounds or limited number of analyses.
f 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 non-priority 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.
"¦ Dibenzofuran, benzyl alcohol, and benzoic acid are HSL compounds and
have been detected frequently in Commencement Bay.
-------
TABLE 3-2. (Continued)
^ Botn 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.
L
* The remaining 7 priority pollution organic bases are seldom detected
in Puget Sound and often present analytical problems (e.g., benzidine and
3,3-dichloro-benzidine),
1 9(H)-carbazol is a component of creosote and coal tar and has been reported
in Puget Sound regions with these sources.
m DOT 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 distributed
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. Coamercial
PCS mixtures are suspected of containing carcinogens or co-carcinogens
and were used historically in enclosed systems (e.g., capacitors and trans-
formers) 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.
-------
TABLE 3-3. POLLUTANT OF CONCERN LIST
roi i ut ant
of
Concern
flntiionv
Arsenic
Caditun
ChrcMiua
Cooper
Lead
Hercury
Nickel
Silver
Zinc
Cyanides
LPAH
Naohthalene
Acenaphthylene
Acenaohthene
Fluorene
Ptienanthrene
Anthracene
HPflH
Fluoranthene
Pyrene
Ben:o(a)anthracene
Chrysene
Total benrofluoranthenes
Berzo(a)pyrene
lnd«no(l,2,3,c.dloyrene
numuoal (a)
A
A
A
A
A
A
A
A
A
A
A
A
A
C
A
A
A
A
A
A
A
A
A
A
?o:nt Sources
Industrial (b>
C.CA.iS.OR
C.OR.LS.(S)
CP.C.(fl)
F.CP.(S)
p.c.cp.or.ca.ls.,(l>.(S)
C.K.CA.OR
CA.B.OC.CA.OR
C.CA.OC.(H>
(CP)
C.E,CA.CR.LS.(M)
CP,C,(F),(H)
L,(«)
L,P
L
L
L
L
L
L. (II)
I
L
L
L
L
L
L
CSOs
-------
TABLE 3-3. (Continued)
Do!iutar:
o+
Concern
Dioen:ci(a.ri!anthracene
Befizooerylene
Total PCBs
Hexachloro6en:e(ie
HexachlorDutadieoe
1.3-dichLorotien:efie
1.4-dichlorotoenzene
4,4'-DDT
4,4'-DDE
4,4 -ODD
flldrin
Oieldrin
ga«u-HCH
PtienoL
4-fletnvlohenol
PentachloroqhenoL
Dibenicrturan
2-flethoxyohefiol
2-t1ethvlnaohthalw«
N-ritrosodiphenyliaine
Trichloroethene
Tetrichloroethert
Ethylbtnzene
Chloro+ori
2,3,7,8-Tetrachlorodioxin
Orgartotin
numciDal (a)
5
A
B
C
c
B
6
C
C
C
C
feint Source?
Inaustrial (6!
L
L
Nonooint
CSOs (c) Sources (d) Soills (e)
OC
X,[C,Ofi,P,L,LS
(P)
P.OC.IC.L
L
(P)
p.oc.ca.idci
P.OC.IC.C/UDC)
IR
IR
en
m
»
m
UR.AR
UR.IR
UR.Ift
6M
SH
-------
TABLE 3-3. (Continued)
a. Nunictoal
ft = Cheaical occurs in >25 percent at sa«;es froa Puaet Sound
aunicioal dtscharaes.
B = Chetnicai occirs :n -.Zi percent of saaoies froa Puaet Sound
¦unianai dismraes.
C - Cheaical not detected baseo on available nforaation.
Blmks indicate that there are insufficient data to categorize.
b. Industrial: Industries in inch cheaical eav be found.
S s Ship buildina/reoair
P = Pulp aills
C = Cooper shelters
CP = Chroae elating, silver platino
F = Ferro, silicon, chroae industries
Cfi - ChloraUah plants
B = Bleach giant
I = loq/nood treatment facility
X = Qrqanic cheaical aanufacturinq
IC - Inorganic cheaical ianufactunnq
LS = Log sort yards
M = Priaary production of ferros and non~ferros aetals
OR = Oil refminq
DC = Dry-ciearung
Codes in parentheses indicate industries wfnch are ootential
sources but have net been documented in Puqet Sound.
Blanks indicate that there are insufficient data to cateoorue.
c. CSOs
A - Cheaical occurs in >25 percent of saaoles froa Puaet Sound CSOs.
B : Cheaical occurs in <25 oercent of saaoles froa Puqet Sound CSOs.
C - Cheaical not detected based on avaiiaole inforaation.
Blanks indicate that tnere are insufficient data to cateoorize.
d. (tonpomt Sources: Types of nonoomt sources »here cheaical nav be found.
UR = Urban runoff
SR = Agricultural ruroff
IR = Industrial runoff
Si - 6roundHat«r
Blanks indicate that there are insufficient data to categorize.
e. Spills: Kinds of soil Is irfiere cheaicai aay be found.
0 = Oil soilis
C * Miscellaneous product soilis
OS 1 Ore soilis
Blanks indicate that there are insufficient data to categorize.
------- |