United States
Environmental Protection
Agency
Sediment Issue
Measuring Contaminant Resuspension
Resulting from Sediment Capping
Purpose
Introduction
Site Descriptions
lethods
Results and Discussion
Boston Harbor
Eagle Harbor
Conclusions
References
Purpose
The National Risk Management Research Lab-
oratory (NRMRL) of the U.S. Environmental
Protection Agency (U.S. EPA) is developing
effective, inexpensive remediation strategies for
contaminated sediments. This program theme
includes the evaluation of capping to contain/sta-
bilize contaminated sediments. Studies were con-
ducted by NRMRL to evaluate the resuspension
of surface materials contaminated with polycyclic
aromatic hydrocarbons (PAHs) and polychlori-
nated biphenyls (PCBs). This information, along
with U.S. EPA's sediment guidance document (1),
is intended to: a) be used as a reference for site
managers and U.S. EPA decision makers who are
considering the environmental impacts of capping
contaminated sediments, and b) provide a better
understanding of the techniques and mechanisms
that can be applied to minimize the resuspension
of contaminated material during capping.
Monitoring the Water Column During Capping Activities at Boston Harbor
The results of two NRMRL studies undertaken to evaluate solids
resuspension before, during, and after capping of contaminated
sediments are summarized below. These two studies were both
conducted at marine sites. One study was carried out at the
Boston Harbor/Mystic River Site in cooperation with U.S. EPA
Region 1 and the U.S. Army Corps of Engineers (USAGE). The
other study took place at the Wyckoff/Eagle Harbor Superfund Site
off Bainbridge Island, WA, in cooperation with U.S. EPA Region
10 and USAGE.
Introduction
During sediment capping activities, clean material is commonly re-
leased from a barge at the water surface and falls through the water
column to the sediment surface, providing an uncontaminated sur-
face sediment layer (2). Information on the potential release of in-
situ contaminated sediment during and after capping operations is
sparse; therefore, NRMRL conducted studies as reported in Lyons
et ol. (2) in order to develop a better understanding of the amount
of contaminants released into the surrounding water column
before, during, and after capping. These studies evaluated whether
the placement of conventional sand caps results in the disturbance
of contaminated surface sediments and thus the release of contami-
nants into the surrounding water column through resuspension.
-------
Two sites were examined where different capping methods
(see Table 1 below) were employed for dissimilar sediment
types (2). Data associated with the sites indicated that:
The resuspension of contaminated sediments was measur-
able, remaining in the ng/L range (for contaminants in
the water column), when capping was conducted over
uncapped sediments.
The magnitude of contaminant resuspension decreased
with successive capping layers, suggesting the greatest
potential for resuspension occurred when capping native
uncapped contaminated material.
After capping operations ceased, turbidity plumes dis-
sipated rapidly (generally within hours) due to deposition
and off-site transport.
Site Descriptions
Table 1 summarizes characteristics of the study sites,
including capping techniques, source of capping materials,
and contaminants-of-concern (COCs) at the study sites.
For the Boston Harbor Site, confined aquatic disposal
(CAD) cell M8, measuring 213 m by 61 m, was exca-
vated to a depth of 27 m and had an estimated capacity
of 118,500 m3 of dredged material. CAD cell Ml 9, the
Table 1. Description of Study Sites.
larger of the two cells measuring 244 m by 91 m, was ex-
cavated to a depth of 24 m and had an estimated capacity
of 136,900 m3. The area monitored for the Eagle Harbor
study covered an area of approximately 150 m by 275 m.
Methods
An aquatic monitoring tool was towed behind a boat to
collect and integrate in-situ measurements with continuous
water collection to monitor the effects of sediment suspen-
sion during capping operations. Aquatic monitoring of
the capping events was conducted using the Battelle Ocean
Sampling System (BOSS) deployed from a survey vessel.
The BOSS is an integrated profiling system comprised of
an underwater sensor unit, an electromechanical profiling
cable for delivery of real-time data and continuous water
samples to the shipboard laboratory, and a customized
profiling winch and handling system, as shown in Figure 1.
The BOSS in-situ sensor package (housed inside a towfish)
includes a conductivity, temperature, and depth (CTD)
sensor; a turbidity sensor; an Acoustic Doppler Current
Profiler for vertical profiles of horizontal currents; and a
Teflon/titanium pumping system for sample collection,
which delivered water samples to the onboard laboratory at
12 L/min through a Teflon line. The survey vessel towed
the BOSS at a depth of approximately 1 to 2 m above the
sediment surface to optimize detection of resuspended
sediments.
Study Site
Boston Harbor/
Mystic River
Site
Wyckoff/Eagle
Harbor
Superfund
Site
Location
Boston,
MA
Bainbridge
Island,
WA
Sediment Type
CAD cells (M8 and
Ml 9) filled with
dredged sediments,
typically 85-100%
silt/clay with in-situ
solids ranging from
30-55% (3)
Bedded (specifics
about sediment
unknown)
Capping
Technique
Pushing an open
hopper dredge with
a tugboat over the
area to be capped
High-pressure
washing of
sediments off the
surface of a barge
over the area to be
capped
Capping Material
Sand dredged from the Cape
Cod Canal having modal
grain size of 0.25 mm diam-
eter with an average of less
than 1% fines (3)
Clean quarry sand with the
following properties: 81.1%
passed through a #10 mesh
but retained on a #40 mesh
(0.43- to 2.0-mm-diameter
medium sand); 9.5% passed
through a #40 mesh but was
retained on a #200 mesh
(0.075- to 0.43-mm-diameter
fine sand); and 0.6% passed
through a #200 mesh (less
than 0.075-mm-diameter silt
or clay) (2)
COCs
PCBs
PAHs
PAHs
Contaminant Concentra-
tions in Sediment
Prior to Capping
Average total PCBs and total
PAHs were 220 ug/kg and
64,478 ug/kg,
respectively (4)
Total PAH concentrations
reported as 1,273 ± 2,116
mg/kg in the upper 10 cm
of three sediment cores
collected within 91 m of
the site; farther from the
site, total PAH concentra-
tions decreased to 18.3 ±
6.6 mg/kg in the upper 10
cm of three sediment cores
collected 305 m from the
site (5)
-------
Winch and Handling System
Navigation
Analyzer Data Act1uisition System
Echo-sounder
In situ
Sensors
Figure 1. BOSS and On-Board Components.
A differential Global Positioning System (developed by
Northstar) was interfaced with the BOSS computer to
provide vessel positioning information during sampling
operations.
In order to evaluate the amount of contaminants released
into the surrounding water column before, during, and
after capping, water samples collected by the BOSS were
analyzed for total PCBs (i.e., sum of 18 PCB congeners
[t-PCBs]), total PAHs (i.e., sum of 16 priority PAH
analytes [t-PAHs]), and total suspended solids (TSS).
If suspended sediment was visibly present in the water sam-
ple, a quartz glass fiber filter (1.0 um) was used to remove
larger sediment particles because they were considered to
represent cap material, and because sediments greater than
1.0-um in diameter would settle relatively quickly in the
immediate vicinity of the capping area. Smaller particles
that passed through the filters were more likely to undergo
long-range transport from the site.
Samples were collected before, during, and after capping
activities and were generally defined as:
Pre-Capping Survey: samples taken several days or 1 week
before capping initiated
Pre-Capping Event: samples taken approximately 1 hour
before each capping event
Capping Event: samples taken during capping (i.e., each
time a lift of capping material applied)
Post-Capping Event: samples taken approximately 1 hour
after each capping event
Post-Capping Survey: samples taken days to months after
capping was completed
The sampling events and sample schedule for each of the
studies are summarized in Table 2.
-------
Figures 2 and 3 show the target BOSS transects and target
sampling locations used for background surveys and active
monitoring events (i.e., during capping) at Boston Harbor
and Eagle Harbor, respectively. The top transect repre-
sents the daily Pre-Capping and Post-Capping background
monitoring events. The bottom transect represents a
typical monitoring event during capping. Actual
transects differed significantly based on the barge loca-
tion, capping operations, and turbidity plume migration.
*3r
» "S is
LL CL (0
Typical
Monitoring Ever
Area of Stjdy (~80 m x ~240 m)
Transect line before capping
Current
-4- /I -C ^
^ 1 .1 1 - J
L Transect line during
\^^/ x__^X capping operations
Start
Sampling Station
Figure 2. Transect Line and Sampling Station Locations for BOSS
Surveys and Monitoring at Boston Harbor.
as
'5.
_ Q.
^ « ^
*9 &
P S 1
0- CL O3
Typical
Monitoring Even
Area of Study (~1 50 m x -275 m)
Transect line before capping
Current
S "N-
n f }
L Transect line during
\^S \^ ^y capping operations
Start
Sampling Station
Figure 3. Transect Line and Sampling Station Locations for BOSS
Surveys and Monitoring at Eagle Harbor.
Table 2. Survey Event and Sample Schedule. Reprinted with permission from (2). Copyright 2006, American Society of Civil Engineers.
Type of Event
Number of
Events
Total Samples
Boston Harbor Sample Schedule
Pre-Capping Survey
Pre-Capping Survey1
Day 1, Capping Events 1-3
Day 2, Capping Events 1-3
Day 3, Capping Events 1-3
Days 1-3, Post-Capping Events
Post-Capping Survey
Total
CAD Cell M19
5
CAD Cell M8
5
Capped Area
3
6 to9c
6 to9c
6to9c
15
10
Pre-Capping Events 14
Capping Events 14
Post-Capping Events 14
Pre-Capping Events 5 8
Capping Events 5 8
Post-Capping Events 5 8
Post-Capping Survey
Total
4
4
4
4
4
4
1
3
9
3"
-
-
5
3
9
3
5
12
36
10
12
36
12
10
138
Eagle Harbor Sample Schedule
24
24
24
15
108
a - Sampler struck bottom immediately following sample collection for the first Post-Capping Event 3 sample, and system components fouled with mud. As a result,
the second and third samples for Post-Capping Event 3 could not be collected.
b - Pre-capping samples were collected by divers during a separate site investigation, and the AMT was not used for the pre-capping survey.
c - Nine samples were collected daily during Transects 1 and 2, and six samples were collected daily during Transect 3.
-------
Results and Discussion
Boston Harbor
Two-dimensional turbidity maps using levels detected
by the BOSS were generated to depict turbidity levels in
the area where capping took place, as shown in Figure 4.
Turbidity data generated by the BOSS in-situ sensors were
calibrated using TSS concentrations measured in the water
samples. Cells M8 and Ml9 produced similar turbidity
and TSS data; however, only results for CAD Cell Ml9
are depicted in Figure 4. The highest turbidity and cor-
42.387-
42.386-
42.365
42.3*4-
t = -1 hr
,'CAD M ta
Boston Harbor - PAH
CAD M19 - Pis-Capping Evsnt 1
aj!M2
-------
occurred when cap material was placed on previously
uncapped sediment. Statistical comparisons among the
four capping events were conducted by omitting the data
from the Pre-Capping and Post-Capping Events and using
Tukey multiple comparisons at a fixed significance level of
0.10. For t-PAHs, Capping Event 1 concentrations were
significantly greater than those for Capping Events 2, 3,
and 4, and there were no significant differences between
Capping Events 2, 3, and 4. For t-PCBs, Capping Event
1 concentrations were significantly greater than those for
Capping Events 2 and 3- Capping Event 4 concentrations
could not be distinguished from those for Capping Events
1, 2, and 3, and resided somewhere between these three
capping events.
Turbidity concentration plots for Capping Events 1
through 4, the Pre-Capping Survey, and the Post-Capping
Survey for Cell Ml9 and water sample locations are plotted
in Figure 4. The relationship between contaminant con-
centrations and turbidity was analyzed by plotting t-PAH
and t-PCB against TSS concentrations for water samples
collected during capping operations. The correlation
coefficients (r2 values) for the best-fit linear regression lines
were calculated (2). Despite the visual observation that
higher TSS/turbidity concentrations during Capping Event
1 coincided with higher t-PAH and t-PCB concentrations,
as shown in Figures 5 and 6, a strong correlation between
high TSS concentrations and high organic contaminant
concentrations could not be determined statistically. It is
likely that the contribution of bed sediments to TSS and
turbidity was overshadowed by the TSS from the cap mate-
rial.
Eagle Harbor
TSS concentrations measured in the water samples and tur-
bidity data generated by the BOSS in-situ sensors were used
to develop two-dimensional turbidity maps, shown in Fig-
ure 7, to display turbidity levels in the area where capping
took place. Elevated turbidity levels were observed at vary-
ing distances and along different directions from the barge,
extending beyond the boundaries of the study area based
on analysis of samples collected outside the study area. The
Post-Capping Event map in Figure 7 shows that turbidity
levels quickly decreased to near Pre-Capping Event transect
levels within 1 to 2 hours after capping. As with Boston
Harbor, the contribution ofTSS from the cap material
itself may have partially overshadowed the contribution of
bed sediments to elevated turbidity and TSS levels. None-
theless, in the vicinity of the capping operations, turbidity
and TSS levels were highest during Capping Events 1 and
2, indicating decreased turbidity with successive capping
events. These data suggest that the measured turbidity in-
cludes a significant contribution from in-situ sediment, and
not only capping material during the initial capping events.
Average t-PAH concentrations measured for the successive
sampling events conducted over the 3-day Eagle Har-
bor monitoring period are shown in Figure 8. Elevated
contaminant concentrations were observed during cap-
ping operations, which appeared to decrease with each
successive capping day and dissipated after capping was
completed. Such rapid dissipation likely was the result of
the combined effects of sedimentation and off-site plume
Day 3 - Capping Event 2 it
11,13100
Duration 15:53-17:05
Study Area
-122.51 -122.SOS -122.506 -122.504
Explanation
« Water Sampling Location
TSS Concentration Based on BOSS Turbiditity Sensor Data
Dimensions of study area, 150m x 275m
Figure 7. Turbidity and TSS Maps for Eagle Harbor Day 3. TSS
values were based on turbidity readings and correlations derived
from measured turbidity and TSS samples. Reprinted with permission
from (2). Copyright 2006, American Society of Civil Engineers.
-------
620
ngIL; se ± 467 ngtL
Pre-Capping Event
Capping Event 1
Capping Event 2
Capping Event 3
Post-Capping Event
Pre-
Capping
Survey
Figure 8. Average t-PAH Concentrations at Eagle Harbor. Error
bars represent standard deviations. Reprinted with permission from (2).
Copyright 2006, American Society of Civil Engineers.
migration. A two-way analysis of variance was conducted
to determine whether there were statistically significant dif-
ferences between different days or sampling events within a
single day using the raw data, log-transformed data, and a
significance level of 0.10. Because of high data variability,
no statistically significant differences were found between
the four sampling events (i.e., samples collected during
Capping Events 1, 2, and 3, and the post-capping sample)
within any single day for Days 1, 2, and 3, and no differ-
ences existed between Days 1, 2, and 3-
Scatter graphs plotting t-PAHs against TSS were generated
to determine the relationship between contaminant con-
centration and turbidity. During the first survey day,
r2 values ranged from 0.72 through 0.95, indicating a cor-
relation between turbidity and t-PAHs. However, r2 values
decreased during subsequent capping surveys and, by the
third day, r2 values were less than 0.54, indicating that a
correlation between turbidity and t-PAHs was lacking. As
with Boston Harbor, it is suspected that the suspended sol-
ids generated by the cap material overshadowed the solids
resulting from suspension of contaminated bed sediments
during subsequent capping events (i.e., after capping events
1 and 2).
Conclusions
A comparison of sampling results at Boston Harbor and
Eagle Harbor is provided in Table 3- COC levels were
below detection limits or at very low levels at both sites
before capping. The highest resuspension of contaminated
material was seen during the first capping event at both
sites. In general, contaminant resuspension, although
substantially higher than observed during pre-capping
sampling, was relatively low for all capping events during
both surveys, where contaminant concentrations remained
in the low ng/L range for most samples. Resuspension of
TSS and COCs occurred continuously throughout capping
operations but dissipated to background levels in a matter
of hours following cessation of capping activities.
Data generated during these two studies have helped to
achieve a better understanding of the amounts and patterns
of contaminants released into the surrounding water col-
umn resulting from the capping events. Data from Boston
Harbor and Eagle Harbor indicate that the resuspension of
contaminated sediments was measurable, but relatively low,
when capping was conducted over uncapped sediments.
Based on the results of the two studies summarized here,
resuspension during capping may be reduced by plac-
ing cap material in lifts in which the first lift provides
a uniform layer of clean material using techniques that
minimize potential disturbance. The data presented here
suggest that subsequent lifts could be placed more a
sively once the contaminated sediment is covered.
res-
References
(1) United States Environmental Protection Agency. 2005. Contami-
nated Sediment Remediation Guidance for Hazardous Waste Sites,
OSWER 9355.0-85, EPA540/R05/012. December. (http://www.
epa.gov/superfund/resources/sediment/pdfs/guidance/pdf).
(2) Lyons, T., J A. Ickes, V.S. Magar, M. ASCE, C.S. Albro, L. Gum-
ming, B. Bachman, T. Fredette, T. Myers, M. Keegan, K. Marcy,
and O. Guza. 2006. Evaluation of Contaminant Resuspension
Potential during Cap Placement at Two Dissimilar Sites. /. Environ.
Eng., 132(4): 505-514.
(3) U.S. Army Corps of Engineers. 1988. Navigation Improvement
Study Feasibility Report and Environmental Assessment. Boston Har-
bor, Massachusetts. Mystic River, Chelsea River and Reserved Channel.
New England Division, Waltham, MA.
Table 3. Concentrations of Suspended Contaminated Sediments Before,
During, and After Capping.
Boston Harbor Eagle Harbor
Monitoring Event PCB (ng/L) PAH (ng/L) PAH (ng/L)
Before Capping
During Capping
After Capping
BDL 46 -59 46 - 73
BDL-84 65-5,242 20-3,872
0.4-1.5 41-83 38-159
(4) U.S. Army Corps of Engineers. 1999. Chemistry
Data Report Boston Harbor Navigation and Berth
Dredging, Boston Harbor. August 5.
(5) Brenner, R.C., V.S. Magar, JA. Ickes, J.E. Abbott,
SA. Stout, EA. Crecelius, and L.S. Bingler. 2002.
Characterization and Fate of PAH-Contaminated
Sediments at the Wyckoff/Eagle Harbor Superfund
Site. Environ. Set. TechnoL, 36(12): 2605-2613.
BDL - below detection limits
-------
vvEPA
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA
United States PERMIT NO. G-35
Environmental Protection
Agency
Office of Research and Development
National Risk Mangement
Research Laboratory
Cincinnati, OH 45268
Official Business
Penalty for Private Use
$300
EPA/600/S-08/013
August 2008
www.epa.gov
Recycled/Recyclable
Printed with vegetable-based ink on
pa per that contains a minumum of
50% post-consumer fiber content
processed chlorine free
------- |