&EPA
United States
Environmental Protection
Agency
Demonstration of the
AquaBlok® Sediment
Capping Technology
Innovative Technology
Evaluation Report
SUPERFUND INNOVATIVE
liCHXtOGY iVAUMf/Q/V
-------
EPA/540/R-07/008
September 2007
Demonstration of the AquaBlok0
Sediment Capping Technology
Innovative Technology Evaluation Report
Final
National Risk Management Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
-------
Notice
The information in this document has been funded by the U.S. Environmental Protection Agency (EPA)
under Contract No. 68-C-00-185 to Battelle Memorial Institute (Battelle). It has been subjected to the
Agency's peer and administrative reviews and has been approved for publication as an EPA document.
Mention of trade names or commercial products does not constitute an endorsement of recommendation
for use.
11
-------
Foreword
The U.S. Environmental Protection Agency (EPA) is charged by Congress with protecting the Nation's
land, air, and water resources. Under a mandate of national environmental laws, the Agency strives to
formulate and implement actions leading to a compatible balance between human activities and the ability
of natural systems to support and nurture life. To meet this mandate, EPA's research program is providing
data and technical support for solving environmental problems today and building a science knowledge
base necessary to manage our ecological resources wisely, understand how pollutants affect our health,
and prevent or reduce environmental risks in the future.
The National Risk Management Research Laboratory (NRMRL) is the Agency's center for investigation of
technological and management approaches for reducing risks from threats to human health and the
environment. The focus of the Laboratory's research program is on methods for the prevention and control
of pollution to air, land, water and subsurface resources; protection of water quality in public water
systems; remediation of contaminated sites and ground water; and prevention and control indoor air
pollution. The goal of this research effort is to catalyze development and implementation of innovative,
cost-effective environmental technologies; develop scientific and engineering information needed by EPA
to support regulatory and policy decisions; and provide technical support and information transfer to ensure
effective implementation of environmental regulations and strategies.
This publication has been produced as part of the Laboratory's strategic long-term research plan. It is
published and made available by EPA's Office of Research and Development (ORD) to assist the user
community and to link researchers with their clients.
Sally Gutierrez, Director
National Risk Management Research Laboratory
in
-------
Abstract
AquaBlok® is an innovative, proprietary clay polymer composite developed by AquaBlok, Ltd. of Toledo,
OH, and represents an alternative to traditional sediment capping materials such as sand. It is designed to
swell and form a continuous and highly impermeable isolation barrier between contaminated sediments
and the overlying water column, and claims superior impermeability, stability, and erosion resistance and
general cost-competitiveness relative to more traditional capping materials. AquaBlok® is generally
marketed as a non-specific capping material that could encapsulate any class or type of contaminant as
well as theoretically any range of contaminant concentration. Although there is claimed to be no
practicable limit to the depth at which the material would function, AquaBlok® is typically formulated to
function in relatively shallow, freshwater to brackish, generally nearshore environments and is commonly
comprised of bentonite clay with polymer additives covering a small aggregate core. In addition, other
specific formulations of AquaBlok® are available, including varieties that can function in saline
environments and advanced formulations that incorporate treatment reagents to actively treat or sequester
sediment contaminants or plant seeds to promote the establishment or regrowth of vegetated habitat.
Under the U.S. Environmental Protection Agency (EPA) Superfund Innovative Technology Evaluation
(SITE) Program, the effectiveness of AquaBlok® as an innovative contaminated sediment capping
technology was evaluated in the Anacostia River in Washington, DC. Sediments in the Anacostia River
are contaminated with polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs), heavy
metals, and other chemicals to levels that have hindered commercial, industrial, and recreational uses.
The performance of AquaBlok® was assessed through the SITE demonstration by monitoring an
AquaBlok® cap over an approximately three year period using a multitude of invasive and/or non-invasive
sampling and monitoring tools. The performance of AquaBlok® was compared to the performance of a
traditional sand cap relative to three fundamental study objectives, and control sediments were also
monitored to provide critical context to the data evaluations. Specifically, the study objectives were to
determine the physical stability of AquaBlok® relative to the traditional sand cap material, the ability of
AquaBlok® to prevent hydraulic seepage relative to traditional sand cap material, and the impact of
AquaBlok® on bent
native river system.
AquaBlok® on benthic habitat and ecology relative to traditional sand cap material and conditions in the
There were field data collection issues and inherent data uncertainties within the SITE demonstration that
limit the usefulness of certain data and minimize the power of certain evaluations and interpretations, and
the conclusions of the demonstration must be reviewed in this context. However, the overall results of the
AquaBlok® SITE demonstration indicate that the AquaBlok® material is highly stable, and likely more stable
than traditional sand capping material even under very high bottom shear stresses. The AquaBlok®
material is also characteristically more impermeable, and the weight of evidence gathered suggests it is
potentially more effective at controlling contaminant flux, than traditional sand capping material. AquaBlok®
also appears to be characterized by impacts to benthos and benthic habitat generally similar to traditional
sand capping material.
IV
-------
Table of Contents
Foreword iii
Abstract iv
Appendices ix
Figures x
Tables xi
Acronyms, Abbreviations, and Symbols xii
Acknowledgements xvi
Section 1: Introduction 1
1.1 Description of the SITE Program and SITE Reports 1
1.1.1 Purpose, History, and Goals of the SITE Program 1
1.1.2 Documentation of SITE Program Results 2
1.1.2.1 Purpose and Organization of the ITER 2
1.2 AquaBlok® General Technology Description 3
1.3 Key Contacts 4
Section 2: Technology Applications Analysis 6
2.1 Key Technology Features 6
2.2 Applicable Wastes 7
2.3 Technology Operability, Availability, and Transportability 7
2.4 Range of Suitable Site Characteristics 8
2.5 Site Support Requirements 9
2.6 Material Handling and Quality Control Requirements 10
2.7 Technology Limitations 10
2.8 Factors Affecting Performance 12
2.9 Site Reuse 13
2.10 Feasibility Study Evaluation Criteria 13
2.10.1 Overall Protection of Human Health and the Environment 13
2.10.2 Compliance with Applicable or Relevant and Appropriate
Requirements 14
2.10.3 Long-Term Effectiveness and Permanence 14
2.10.4 Reduction of Toxicity, Mobility, and Volume through Treatment 15
2.10.5 Short-Term Effectiveness 15
2.10.6lmplementability 16
2.10.7 Cost 16
2.10.8 State Acceptance 16
2.10.9 Community Acceptance 16
2.11 Permitting 17
-------
Section 3: Technology Effectiveness 19
3.1 AquaBlok® SITE Demonstration Program Description 19
3.1.1 AquaBlok® SITE Demonstration Study Area Description and History 21
3.1.1.1 Physical and Chemical Setting of AquaBlok® SITE
Demonstration 21
3.1.1.2 AquaBlok® SITE Demonstration Cap Design and Construction 21
3.2 AquaBlok® SITE Demonstration Approach and Methods 27
3.2.1 Critical and Non-Critical Measurements 28
3.2.2 Field Activities 30
3.2.3 Field Measurement Tools 31
3.2.3.1 Sedflume Coring and Analysis 31
3.2.3.2 Sediment Coring and Analysis of Contaminants of Concern 34
3.2.3.3 Bathymetry and Sub-Bottom Profiling 35
3.2.3.4 Side-Scan Sonar 35
3.2.3.5 Sediment Profile Imaging 35
3.2.3.6 Gas Flux Analysis 37
3.2.3.7 Sediment Coring and Analysis of Physical Parameters 39
3.2.3.8 Sediment Coring and Analysis of Hydraulic Conductivity 39
3.2.3.9 Seepage Meter Testing 39
3.2.3.10Benthic Grab Sampling and Descriptive and Statistical Benthic
Assays 41
3.2.3.11 Benthic Assessment through Sediment Profile Imaging 42
3.2.4 AquaBlok® SITE Demonstration Specific Approach and Methods 43
3.2.4.1 One-Month Post-Capping Field Event 43
3.2.4.1.1 One-Month Post-Capping Field Event Bathymetry
and Sub-Bottom Profiling 43
3.2.4.1.2 One-month Post-Capping Field Event Side-Scan
Sonar Surveying 44
3.2.4.1.3 One-Month Post-Capping Field Event Sediment
Profile Imaging 44
3.2.4.1.4 One-Month Post-Capping Field Event Seepage
Meter Testing 46
3.2.4.2 Six-Month Post-Capping Field Event 46
3.2.4.2.1 Six-Month Post-Capping Field Event Bathymetry
and Sub-bottom Profiling 46
3.2.4.2.2 Six-Month Post-Capping Field Event Sediment
Profile Imaging 47
3.2.4.2.3 Six-Month Post-Capping Field Event Seepage Meter
Testing 47
3.2.4.2.4 Six-Month Post-Capping Field Event Sedflume
Coring and Analysis 47
3.2.4.2.5 Six-Month Post-Capping Field Event Sediment
Coring and Analysis of Contaminants of Concern
and Physical Parameters 48
3.2.4.3 18-Month Post-Capping Field Event 50
3.2.4.3.1 18-Month Post-Capping Field Event Bathymetry and
Sub-bottom Profiling 50
VI
-------
3.2.4.3.2 18-Month Post-Capping Field Event Side-Scan
Sonar Surveying 50
3.2.4.3.3 18-Month Post-Capping Field Event Sediment
Profile Imaging 50
3.2.4.3.4 18-Month Post-Capping Field Event Seepage Meter
Testing 51
3.2.4.3.5 18-Month Post-Capping Field Event Sediment
Coring and Analysis of Contaminants of Concern,
Physical Parameters, and Hydraulic Conductivity 51
3.2.4.3.6 18-Month Post-Capping Field Event Gas Flux
Analysis 52
3.2.4.4 30-Month Post-Capping Field Event 53
3.2.4.4.1 30-Month Post-Capping Field Event Bathymetry and
Sub-Bottom Profiling 53
3.2.4.4.2 30-Month Post-Capping Field Event Side-Scan
Sonar Surveying 53
3.2.4.4.3 30-Month Post-Capping Field Event Sediment
Profile Imaging 54
3.2.4.4.4 30-Month Post-Capping Field Event Seepage Meter
Testing 54
3.2.4.4.5 30-Month Post-Capping Field Event Sedflume
Analysis 54
3.2.4.4.6 30-Month Post-Capping Field Event Sediment
Coring and Analysis of Contaminants of Concern,
Physical Parameters, and Hydraulic Conductivity 55
3.2.4.4.7 30-Month Post-Capping Field Event Gas Flux
Analysis 56
3.2.4.4.8 30-Month Post-Capping Field Event Benthic Grab
Sampling and Descriptive and Statistical Benthic
Assays 57
3.2.4.5 General AquaBlok® SITE Demonstration Quality Assurance and
Quality Control 58
3.3 AquaBlok® SITE Demonstration Results 59
3.3.1 Objective #1 -Physical Stability of An AquaBlok® Cap 59
3.3.1.1 Objective#1 Results- Critical Measurements 64
3.3.1.1.1 Sedflume Coring and Analysis 64
3.3.1.1.2 Sediment Coring and Analysis of Contaminants of
Concern 66
3.3.1.1.3 Bathymetry and Sub-Bottom Profiling 78
3.3.1.1.4 Side-Scan Sonar Surveying 82
3.3.1.2 Objective#1 Results- Non-Critical Measurements 85
3.3.1.2.1 Sediment Profile Imaging 85
3.3.1.2.2 Gas Flux Analysis 86
3.3.1.2.3 Sediment Coring and Analysis of Physical
Parameters 92
3.3.2 Objective #2 - Ability of An AquaBlok® Cap to Control Groundwater
Seepage 94
Vll
-------
3.3.2.1 Objective#2 Results- Critical Measurements 94
3.3.2.1.1 Sediment Coring and Analysis of Hydraulic
Conductivity 94
3.3.2.1.2 Seepage Meter Testing 96
3.3.2.2 Objective#2 Results- Non-critical Measurements 105
3.3.2 Objective #3 - The Influence of An AquaBlok® Cap on Benthic Flora
and Fauna 105
3.3.3.1 Objective #3 Results - Critical Measurements 106
3.3.3.2 Objective #3 Results - Non-Critical Measurements 106
3.3.3.2.1 Benthic Grab Sampling and Descriptive and
Statistical Benthic Assays 106
3.3.3.2.2 Benthic Assessment Through Sediment Profile
Imaging 107
Section 4: Econom ic Analysis 113
4.1 SITE Demonstration Pilot-Scale AquaBlok® Capping Costs 113
4.1.1 SITE Demonstration As-Built AquaBlok® Cap 113
4.1.2 SITE Demonstration AquaBlok® Pilot Costs 115
4.2 Full-Scale AquaBlok® Application 115
4.2.1 Site-Specific Factors Affecting Cost 115
4.2.2 Issues and Assumptions 116
4.2.3 Full-Scale AquaBlok® Application Cost Categories 117
4.2.3.1 General Cost Categories 117
4.2.3.1.1 Local AquaBlok® Manufacture Costs 117
4.2.3.1.2 AquaBlok® Cap Installation Costs 117
4.2.3.1.3 Construction Quality Control and Documentation
Costs 120
4.2.3.1.4 Engineering Design, Permitting, Contract and Bid
Document Preparation, and Contract Administration
Costs 120
4.2.3.1.5 Operations and Maintenance Costs 122
4.2.4 Full-Scale AquaBlok® Cap Installation Cost Analysis Summary 122
Section 5: Technology Status 124
Section 6: References 127
Vlll
-------
Appendices*
Appendix A Vendor-supplied AquaBlok® Commercial Application Claims
Appendix B AquaBlok® SITE Demonstration Oceanographic Survey Results
Appendix B-1 - AquaBlok® SITE Demonstration One-month Post-capping Oceanographic
Survey Results
Appendix B-2 - AquaBlok® SITE Demonstration Six-month Post-capping Oceanographic
Survey Results
Appendix B-3 - AquaBlok® SITE Demonstration 18-month Post-capping Oceanographic
Survey Results
Appendix B-4 - AquaBlok® SITE Demonstration 30-month Post-capping Oceanographic
Survey Results
Appendix C AquaBlok® SITE Demonstration Sediment Profile Imagery Results
Appendix D AquaBlok® SITE Demonstration Seepage Meter Testing Results
Appendix D-1 - AquaBlok® SITE Demonstration One-month Post-capping Seepage Meter
Testing Results
Appendix D-2 - AquaBlok® SITE Demonstration Six-month Post-capping Seepage Meter
Testing Results
Appendix D-3 - AquaBlok® SITE Demonstration 18-month Post-capping Seepage Meter
Testing Results
Appendix D-4 - AquaBlok® SITE Demonstration 30-month Post-capping Seepage Meter
Testing Results
Appendix E AquaBlok® SITE Demonstration Sedflume Results
Appendix E-1 - AquaBlok® SITE Demonstration Six-month Post-capping Sedflume Results
Appendix E-2 - AquaBlok® SITE Demonstration 30-month Post-capping Sedflume Results
Appendix F AquaBlok® SITE Demonstration Benthic Assay and Statistical Evaluation Results
Appendix G AquaBlok® SITE Demonstration Sediment Coring Logs
AquaBlok® SITE Demonstration Six-month Post-capping Sediment Coring Logs
AquaBlok® SITE Demonstration 18-month Post-capping Sediment Coring Logs
AquaBlok® SITE Demonstration 30-month Post-capping Sediment Coring Logs
Appendix H AquaBlok® SITE Demonstration Sediment Coring Data and Graphs
AquaBlok SITE Demonstration Sediment Coring Data
Polycyclic Aromatic Hydrocarbon Data
Polychlorinated Biphenyl Data
Metals Data
Physical Data
AquaBlok SITE Demonstration Sediment Coring Graphs
Polycyclic Aromatic Hydrocarbon Graphs
Polychlorinated Biphenyl Graphs
Metals Graphs
Physical Graphs
*Copies of Appendices available from the
EPA Task Order Manager at (513) 569-7669
IX
-------
Figures
Figure 1-1. Integrated Conceptual and Actual View of AquaBlok® Capping Material 4
Figure 3-1. Anacostia River Watershed (Anacostia Watershed Society Website) 20
Figure 3-2. Locations of Preliminary AquaBlok® SITE Demonstration Study Areas 22
Figure 3-3. Aerial Image of Preliminary AquaBlok® SITE Demonstration Study Areas 22
Figure 3-4. AquaBlok® SITE Demonstration Area 1 Capping Cell Layout 23
Figure 3-5. AquaBlok® in 2-Ton SuperSack at Staging Area 24
Figure 3-6. Sand Cap Material Stored in Bulk at Staging Area 25
Figure 3-7. Transferring Cap Material to Barge Using Conveyor 25
Figure 3-8. Crane Barge Used to Place Caps in Demonstration Area 26
Figure 3-9. Silt Curtains Deployed Around Demonstration Area Capping Cells 28
Figure 3-10. Sedflume Schematic 33
Figure 3-11. Principles of Acoustic Sub-Bottom Profiling 36
Figure 3-12. Schematic of Sediment Profiling Camera 36
Figure 3-13. Schematic of Typical Submerged Gas Flux Chamber 38
Figure 3-14. Schematic of Ultrasonic Seepage Meter 40
Figure 3-15. Conceptual Cross-Section of Ultrasonic Seepage Meter Flow Tube 40
Figure 3-16. One-Month Post-Capping Field Event Sampling/Monitoring Locations 60
Figure 3-17. Six-Month Post-Capping Field Event Sampling/Monitoring Locations 61
Figure 3-18. 18-Month Post-Capping Field Event Sampling/Monitoring Locations 62
Figure 3-19. 30-Month Post-Capping Field Event Sampling/Monitoring Locations 63
Figure 3-20. Sedflume Coring Locations 65
Figure 3-21. Potomac River (top) and Anacostia River (bottom) River Flows During
Demonstration 67
Figure 3-22. Sediment Coring Locations 68
Figure 3-23. Total PAHs in Control Cell Cores 70
Figure 3-24. Total PAHs in AquaBlok® Cell Cores 70
Figure 3-25. Total PAHs in Sand Cell Cores 71
Figure 3-26. Total PCBs in Control Cell Cores 71
Figure 3-27. Total PCBs in AquaBlok® Cell Cores 72
Figure 3-28. Total PCBs in Sand Cell Cores 72
Figure 3-29. Total Metals in Control Cell Cores 73
Figure 3-30. Total Metals in AquaBlok® Cell Cores 73
Figure 3-31. Total Metals in Sand Cell Cores 74
Figure 3-32. Survey Transects in Demonstration Area for Oceanographic Surveying 79
Figure 3-33. One-Month Post-Capping Bathymetric Cap Thickness Map 81
Figure 3-34. 30-Month Post-Capping Bathymetric Cap Thickness Map 81
-------
Figure 3-35. One-Month Post-Capping Side-Scan Sonar Map 83
Figure 3-36. 30-Month Post-Capping Side-Scan Sonar Map 83
Figure 3-37. Sediment Profile Imaging Monitoring Locations 84
Figure 3-38. Video SPI Camera Penetration Trend 86
Figure 3-39. Gas Flux Monitoring Locations 89
Figure 3-40. Average TOC Concentration in Demonstration Area During SITE
Demonstration 95
Figure 3-41. Seepage Meter Monitoring Locations 98
Figure 3-42. Specific Discharge Rates in AquaBlok® Cell During 30-Month Post-Capping
Survey 101
Figure 3-43. Specific Discharge Rates in Sand Cell During 30-Month Post-Capping Survey 102
Figure 3-44. Specific Discharge Rates in Control Cell During 30-Month Post-Capping
Survey 103
Figure 3-45. Specific Discharge Rates in Demonstration Area During One-Month Post-
Capping Survey 100
Figure 3-46. Bethic Grab Sampling Locations (Including Baseline) 109
Figure 3-47. Abundance of Dero nivea in AquaBlok (AB), Sand (SO), and Control (UC)
Cells 110
Figure 3-48. Abundance of Chironomid Larvae in AquaBlok (AB), Sand (SO), and Control
(UC) Cells 110
Figure 3-49. Total Benthos Abundance in AquaBlok (AB), Sand (SO), and Control (UC)
Cells 112
Figure 3-50. Gas Void Occurrence Trend in Video SPI 112
Figure 4-1. Typical Barge-Mounted Material Conveyor 118
Figure 4-2. Typical Material Barge 118
Figure 4-3. Conceptual Daily Work Cycle for "Typical" AquaBlok® Capping Project
(10-acre AquaBlok® and Sand Cap) 121
Tables
Table 2-1. Summary of AquaBlok® Performance Expectations Relative to CERCLA
Feasibility Criteria 18
Table 3-1. Capping Cell Construction Design and Tolerances 24
Table 3-2. Critical and Non-critical SITE Demonstration Measurements 31
Table 3-3. SITE Demonstration Field Program Details 32
Table 3-4. SITE Demonstration Gas Flux Sampling Observations 90
Table 3-5. SITE Demonstration Gas Flux Sampling Results 91
Table 3-6. Calculated Volumetric and Mass Gas Flux for Individual Compounds 93
Table 3-7. SITE Demonstration Hydraulic Conductivity Results 97
Table 3-8. SITE Demonstration Seepage Meter Results 104
Table 3-9. Results of the Statistical Comparison of Specific Discharge between Cells 106
Table 4-1. Cost Detail for "Typical" AquaBlok® Capping Project (10-acre AquaBlok®
Cap with Sand Cover) 119
XI
-------
Acronyms, Abbreviations, and Symbols
A
ADCP
Ag
AMS
ANOVA
ARAR
ASTM
Athena
AWTA
BMP
bps
°C
CAD
Cd
CD
CDF
CERCLA
CH4
cm
cm2
cm3
cm/s
cm3/s
CNESS
CO2
COC
Cr
CSO
Cu
CV
CWA
cy
d
d
dGPS
DO
£
ECC
area
acoustic Doppler current profiler
silver
Applied Marine Sciences, Inc.
analysis of variance
Applicable or Relevant and Appropriate Requirement
American Society for Testing and Materials
Athena Technologies, Inc.
Anacostia Watershed Toxics Alliance
best management practice
bits per second
degrees Celsius
confined aquatic disposal (facility)
cadmium
compact disc
confined disposal facility
Comprehensive Environmental Response, Compensation, and
Liability Act
methane
centimeter(s)
square centimeter(s)
cubic centimeter(s)
centimeter(s) per second
cubic centimeter(s) per second
chord-normalized expected species shared
carbon dioxide
contaminant of concern
chromium
combined sewer outfall
copper
coefficient of variation
Clean Water Act
cubic yard
day(s)
Margalef s species richness
differential global positioning system
dissolved oxygen
erosion rate
Earth Conservation Corps
xn
-------
EPA United States Environmental Protection Agency
E(Sn) Sander's Rarefaction
FM frequency modulation
F.O.B. free on board (shipping)
FS feasibility study
ft foot/feet
ft2 square foot/feet
ft/s foot/feet per second
g gram(s)
g/cm3 gram(s) per cubic centimeter
gal gallon(s)
gal/s gallon(s) per second
GPS global positioning system
GSA General Services Administration
H' Shannon Diversity Index
HASP health and safety plan
HEC habitat enhancement cap
Hg mercury
HSD Honestly Significant Difference
HSRC Hazardous Substances Research Center
Hz hertz
1C institutional control
IDW investigation-derived waste
in inch(es)
ITER Innovative Technology Evaluation Report
J' Pielou's Evenness Index
K hydraulic conductivity
kg kilogram
kHz kilohertz
L liter(s)
lb(s) pound(s)
Ibs/ft2 pounds per square foot
LSU Louisiana State University
ug microgram(s)
ug/g microgram(s) per gram
ug/kg microgram(s) per kilogram
ug/L microgram(s) per liter
m meter(s)
mm millimeter(s)
m2 square meter(s)
Matrix Matrix Environmental and Geotechnical Services
mg milligram(s)
mg/kg milligram(s) per kilogram
Xlll
-------
MGP
mi
mi2
MLLW
MMT
mol
N
N2
N/m2
NAD
NAVD
Navy
ng
nMDS
NPL
NRMRL
02
ORD
osi
OSWER
O&M
p/P
P
PAH
Pb
PCB
ppb
ppbv
ppm
ppmv
ppt
PSD
q
Q
QA
QAPP
QA/QC
QC
R
RAO
RCRA
RD
Rl
ROD
RPD
manufactured gas plant
mile(s)
square mile(s)
mean lower low water
monitoring and measurement technology
mole(s)
Newton(s)
nitrogen
Newton(s) per square meter
North American Datum
North American Vertical Datum
United States Navy
nanogram
non-metric multi-dimensional scaling
National Priorities List
National Risk Management Research Laboratory
oxygen
Office of Research and Development
organism-sediment index
Office of Solid Waste and Emergency Response
operation and maintenance
pressure
bulk density
polycyclic aromatic hydrocarbon
lead
polychlorinated biphenyl
parts per billion
parts per billion by volume
parts per million
parts per million by volume
parts per trillion
particle size distribution
specific discharge
discharge
quality assurance
quality assurance project plan
quality assurance and quality control
quality control
universal gas constant
remedial action objective
Resource Conservation and Recovery Act
remedial design
remedial investigation
Record of Decision
Redox Potential Discontinuity
xiv
-------
S'
SARA
sec
SITE
SPI
SO
STP
t
T
i
TER
TNMOC
TOC
UC
USCG
USDA
USGS
UT
UXO
v
V
WASA
WINOPS
Zn
ZVI
Bray-Curtis similarity coefficient
Superfund Amendments and Reauthorization Act
second(s)
Superfund Innovative Technology Evaluation
sediment profile imagery/imaging
sand only
standard temperature and pressure
time
time or temperature
shear stress
Technology Evaluation Report
total non-methane organic carbon
total organic carbon
uncapped control
United States Coast Guard
United States Department of Agriculture
United States Geologic Survey
University of Texas
unexploded ordnance
velocity
velocity or volume
Washington Area Water and Sewer Authority
Wndows-based Offshore Positioning Software
zinc
zero-valent iron
xv
-------
Acknowledgements
This report was prepared by Battelle Memorial Institute (Battelle) of Columbus, Ohio, under the direction of
Dr. Edwin Earth, the U.S. Environmental Protection Agency's (EPA) Superfund Innovative Technology
Evaluation (SITE) task order manager at the National Risk Management Research Laboratory (NRMRL) in
Cincinnati, Ohio. Under the direction of Dr. Earth and other EPA technical staff, Battelle was tasked with
designing, conducting, and evaluating the demonstration of the AquaBlok® sediment capping technology.
Contributors and/or reviewers for this report were Mr. John Hull of AquaBlok, Ltd. in Toledo, Ohio, Dr. Joe
Jersak of Biologge AS in Sandefjord, Norway, and Dr. Danny Reible of the Hazardous Substances
Research Center (HSRC) at Louisiana State University (LSU) in Baton Rouge, Louisiana and the
University of Texas (UT) in College Station, Texas. In addition, technical review comments were provided
by Bob Lien, Barbara Bergen, and Terry Lyons of EPA's Office of Research and Development (ORD), Eric
Stern of EPA Region II, and Dr. Carl Herbranson of the Minnesota Department of Health. Staff at the Earth
Conservation Corps (ECC), Washington Area Water and Sewer Authority (WASA), and the General
Services Administration (GSA) in Washington, DC was extremely generous in facilitating field operations.
In particular, Ms. Brenda Richardson and Mr. Glen Ogilvie of ECC, Mr. Charles Wynn and Mr. Carl Banks
of WASA, and Mr. Robert Oliphant of GSA were invaluable in coordinating and successfully implementing
field operations. Facilities operations personnel at the Washington Navy Yard were generous in allowing
the collection of tidal data along their secure bulkhead, and the Harbor Police in Washington, DC and
security personnel at WASA were highly professional throughout the field demonstration project.
xvi
-------
Section 1
Introduction
Under the U.S. Environmental Protection Agency
(EPA) Superfund Innovative Technology
Evaluation (SITE) Program, the effectiveness of
AquaBlok®, a proprietary clay polymer composite
developed by AquaBlok, Ltd. of Toledo, OH that
represents an alternative to traditional sediment
capping materials such as sand, was evaluated
in the Anacostia River in Washington, DC as an
innovative contaminated sediment capping
technology.
This introduction briefly describes the EPA's SITE
program and the reports produced to document a
SITE demonstration project. This introduction
also provides the purpose and general
organization of this Innovative Technology
Evaluation Report (ITER). Background
information on the development of the AquaBlok®
sediment capping technology is also provided,
including a general description of the technology
and its claimed or documented innovative
characteristics, as well as a list of key contacts
who can supply additional information and details
about the technology and the demonstration site.
1.1 Description of the SITE Program
and SITE Reports
This section briefly describes the purpose and
goals of the SITE program and the reports
produced to document the results of SITE
demonstration projects.
1.1.1 Purpose and Goals of the SITE
Program
The primary purpose of the SITE program is to
advance the development and demonstration of
innovative environmental remediation
technologies that are likely applicable to
Comprehensive Environmental Response,
Compensation, and Liability Act (CERCLA; i.e.,
Superfund) and other hazardous waste sites, and
to thereby facilitate the commercial availability
and applicability of such technologies. The SITE
program is administered by the EPA's Office of
Research and Development (ORD) National Risk
Management Research Laboratory (NRMRL) in
the Land Remediation and Pollution Control
Division.
The overall goal of the SITE program is to carry
out the research, evaluation, testing,
development, and demonstration of alternative or
innovative environmental remediation and
treatment technologies that may be used in
response actions at cleanup sites to achieve
long-term protection of human health and the
environment.
Data collected during demonstration projects are
used to assess the performance of technologies
against pre-determined measurement endpoints
to determine applicability and likelihood for
successful implementation at cleanup sites. The
data are used to determine key technology
parameters such as the potential need for pre- or
post-treatment, the types of contaminants,
wastes, and media that could be successfully
addressed, operational design considerations and
limitations, and typically associated capital and
operating costs. Demonstration data can also
provide information on long-term operation and
maintenance (O&M) or monitoring needs as well
as long-term application risks.
Under each SITE demonstration project, a
particular technology's performance is assessed
by how it addresses a particular waste type or
contaminant suite at a particular site. While
successful demonstration of a technology's
performance at the demonstration site is
important in interpreting the applicability and
functionality of the technology, it does not
-------
necessarily ensure the technology's success at
other sites. Data obtained during a SITE
demonstration project can and often do require
extrapolation to estimate an appropriate range of
operating conditions over which the technology
would function effectively and successfully. In
addition, other available case study information
on a particular technology should be used to
extrapolate technology performance conclusions.
SITE demonstration projects typically rely on
cooperative arrangements between EPA, the
technology developer, and the site
owner/operator. EPA is generally responsible for
project planning, monitoring, sampling and
analysis, quality assurance and quality control
(QA/QC), report preparation, and project
information publication and dissemination. The
site owner/operator is generally responsible for
routine site logistics and transport and disposal of
investigation-derived waste (IDW). The
technology developer is typically responsible for
providing the technology to be demonstrated and
for mobilizing and demobilizing equipment
required to deploy the technology.
1.1.2 Documentation of SITE Program
Results
The results of SITE demonstration projects are
documented in four individual reports: a
Technology Demonstration Bulletin; a Technology
Capsule; a Technology Evaluation Report (TER);
and an ITER. The Technology Demonstration
Bulletin provides a brief description of the
technology and SITE project history, notification
that the SITE demonstration was completed, and
key highlights of the demonstration project
findings. The Technology Capsule provides an
even more brief description of the SITE project
and an overview of the project findings and
conclusions.
The purpose of the TER is to consolidate the
information and data generated during the SITE
project, and summarizes the data generated
during the SITE project in comparison with
QA/QC protocols and data quality objectives
(DQOs) relative to measures of data usability,
including accuracy, precision, and completeness.
The TER is not formally published by EPA, but is
retained by EPA as a reference for responding to
public inquiries and for record-keeping purposes.
The Technology Demonstration Bulletin,
Technology Capsule, and TER are produced as
separate, stand-alone documents generally in
parallel with the ITER. The Bulletin, Capsule, and
TER will be produced in this fashion for the
AquaBlok® SITE demonstration documented
herein. The ITER is discussed in more detail in
Section 1.1.2.1.
1.1.2.1 Purpose and Organization of the
ITER. The purpose of the ITER is to assist
decision-makers in evaluating specific
environmental remediation and treatment
technologies for applicability to various cleanup
scenarios and specific cleanup sites. The ITER
discusses the effectiveness and applicability of a
technology and provides an assessment of the
costs associated with the deployment of the
technology. The technology is evaluated on the
basis of data collected during the SITE
demonstration project and, if and where available,
from other case studies. The applicability of the
technology is discussed in terms of contamination
and site characteristics that could affect
technology performance, handling requirements,
limitations, and other important factors. The ITER
represents an important step in the full-scale
development and commercialization of an
environmental remediation technology
demonstrated through the SITE program.
This ITER has been prepared specifically to
summarize the SITE demonstration of the
AquaBlok® sediment capping technology in the
Anacostia River in Washington, DC. Consistent
with the general layout of most ITERs, this ITER
consists of the following sections:
• Section 1 - Introduction: briefly describes the
SITE program in general terms and the
reports produced to document a SITE
demonstration project. Specifically
summarizes the purpose and layout of this
ITER and briefly summarizes the AquaBlok®
technology.
• Section 2 - Technology Applications Analysis:
discusses information relevant to the
application of AquaBlok®, including an
-------
assessment of the technology in the context
of the nine CERCLA feasibility criteria and the
operational and technical limitations of the
technology.
• Section 3 - Technology Effectiveness:
presents information related to the design and
implementation of the AquaBlok® SITE
demonstration at the demonstration site. This
section also summarizes the objectives of the
project, the procedures used in carrying out
the demonstration, and the findings of the
demonstration.
• Section 4 - Economic Analysis: summarizes
the actual costs (within several principal cost
categories) associated with deploying
AquaBlok®, and discusses variables and
scaling factors that may affect the
technology's cost at other sites.
• Section 5 - Demonstration Conclusions:
summarizes the conclusions of the AquaBlok®
SITE demonstration and the status of the
development and commercial availability of
the technology evaluated.
• Section 6 - References: lists the references
used in compiling the ITER.
1.2
AquaBlok® General Technology
Description
AquaBlok® is a proprietary clay polymer
composite developed by AquaBlok, Ltd. of
Toledo, Ohio. AquaBlok® material is designed to
contain and isolate contamination in subaqueous
sediments in predominantly non-terrestrial
settings. In addition, the material can be used for
other applications, such as in retention pond or
wastewater basin lining, well sealing, and erosion
control.
AquaBlok® is a particulate material, with each
particle comprised of an aggregate core covered
by a clay and polymer coating. The clay in most
applications is primarily bentonite, and the
polymer is added to promote adhesion between
the clay and the aggregate core. Specific
formulations that incorporate other clay types
(e.g., attapulgite) or additives (e.g., plant seeds)
are available or can be designed to address site-
specific (e.g., salinity) or action-specific (e.g.,
treatment requirements) needs. The material is
generally applied as a dry product through the
water column to the surface of contaminated
subaqueous sediments and hydrates to form a
continuous and impermeable isolation cap. An
integrated conceptual and actual depiction of
AquaBlok® as a contaminant barrier is provided in
Figure 1-1.
AquaBlok® claims to offer distinct advantages
over materials traditionally used to cap
contaminated sediments (i.e., sand or clean
native sediment). These advantages, as
generally claimed, include:
• Low aqueous permeability and transmissivity
due to low hydraulic conductivity (on the order
of 10"9 centimeters per second [cm/s] for
typical bentonite freshwater formulations);
• High degree of cohesiveness and cap
uniformity due to coalescing of individual
particles on hydration;
• High contaminant attenuation capacity due to
binding capacity of the clays used;
• Contaminant non-specificity due to very low
permeability and uniform isolation coverage;
• High resistance to physical erosion due to
cohesiveness;
• Lower thickness requirements for contaminant
isolation due to physical properties of
material; and
• Compatibility with other remediation elements
and amendments (e.g., reactive components
or seed).
AquaBlok® can be manufactured in specific
blends to accommodate specific cleanup
objectives. It is generally packaged in large bags
-------
AquaBlok
Before Hydration
time
AquaBlok After
Hydration and
Expansion
_ water _
column
AquaBlok cap
contaminated
~~ sediment ~~
— substrate —
not to scale
i.®
Figure 1-1. Integrated Conceptual and Actual View of AquaBlok® Capping Material
(EPA Tech Trends, February 2000)
but can be packaged loose in bulk containers. It
can be transported by truck, rail, or barge, and
can be directly deployed at cleanup sites from
land using typical excavation equipment, from
water using direct-application barges or barge
cranes, or by air using helicopters. It can also be
placed by hand if necessary. The material can be
manufactured on site to meet a specific need or
to achieve cost advantage (i.e., by using local
sources of component materials). AquaBlok® can
be placed, as with more traditional sediment
capping materials, in one or more lifts to achieve
a design cap thickness, and can be armored with
other materials (e.g., sand, gravel, or stone) if
necessary.
The first application of AquaBlok® as an
environmental remediation technology occurred
in an impacted wetlands at a Superfund (i.e.,
CERCLA) site in Alaska known as Eagle River
Flats. The AquaBlok® material was developed for
the Eagle River Flats site in a collaborative effort
between commercial interests and the United
States Department of Agriculture (USDA).
AquaBlok® was subsequently included in the
Record of Decision (ROD) for the site for the in
situ management of impacted sediments. Since
that time, AquaBlok® has, based on information
provided by AquaBlok, Ltd., been successfully
deployed as a sediment remediation technology
at 10 sediment remediation project sites and
evaluated at bench-scale at several others.
1.3 Key Contacts
Additional information on the AquaBlok® sediment
capping technology or the AquaBlok® SITE
demonstration project is available from the
following contacts:
Edwin Earth, Ph.D., P.E., C.I.H.
U.S. Environmental Protection Agency
Office of Research and Development
26 West Martin Luther King Drive
Cincinnati, OH 45268
Telephone: (513) 569-7669
Fax: (513) 569-7158
barth.ed@epa.gov
Andrew Bullard, M.E.M.
Principal Research Scientist
Environmental Restoration
Battelle Memorial Institute
125 Pheasant Run, Suite 115
Newtown, PA 18940
Telephone: (215) 504-5312
Fax: (614) 458-6622
bullarda@battelle.org
-------
John Hull, P.E.
AquaBlok, Ltd.
3401 Glendale Avenue
Suite 300
Toledo, OH 43614
Telephone: (800) 688-2649
Fax: (419)385-2990
ihull(S)aquabloki nfo.com
Danny Reible, Ph.D.
Department of Civil, Architectural and
Environmental Engineering
The University of Texas at Austin
1 University Station C1786
Austin, Texas 78712-0283
Telephone: (512)471-4642
Fax: (512)471-5870
reible@mail.utexas.edu
-------
Section 2
Technology Applications Analysis
This section describes the general applicability
and anticipated effectiveness of the AquaBlok®
sediment capping technology at hazardous waste
cleanup sites. It also describes factors at any
given site that might affect the performance of the
AquaBlok® technology, and summarizes the
expected performance of this technology in the
context of the nine CERCLA criteria used during
feasibility studies to assess the reasonableness
of a potential remediation strategy to accomplish
environmental cleanup at a site.
Additional vendor-supplied information regarding
specific applications, formulations, and
commercial status of the technology is provided
in Appendix A. The information provided in
Appendix A is based exclusively on vendor-
supplied information, and has not been
independently verified.
2.1 Key Technology Features
For contaminated subaqueous sediments, the
most common remediation strategies are
dredging, which involves the removal of
contaminated material (and potentially the
placement of fill material to restore the sediment
surface to its original elevation or to cover
residual contamination exposed by dredging but
economically infeasible to remove), and capping,
which involves the placement of a barrier
between the contaminated sediment and the
overlying water. Capping, subsequently, can be
accomplished using isolation caps, which function
by completely isolating sediment contaminants
from the overlying water, or thin-layer or habitat
enhancement caps (HECs), which function by
creating a clean layer of adequate but minimal
thickness to provide an appropriate level of
isolation while allowing natural physical and
ecological mechanisms to function as a
component of the remedy (e.g., natural recovery).
Generally, capping approaches are less costly
than dredging, but do typically require longer-
term O&M activities to ensure remedy integrity
and the achievement of remedial action
objectives (RAOs).
Capping contaminated subaqueous sediments
can be accomplished using common earth
materials such as sand and gravel, or using clean
sediment similar to that being capped (generally
proportionally finer grained material such as silts
and clays for most contaminated sediment sites).
If necessary, sediment caps can be armored
against physical stresses using an armoring layer
such as gravel or stone.
AquaBlok® is a proprietary clay polymer
composite designed to hydrate and form a
continuous and highly impermeable isolation
layer over contaminated sediments. While it is
claimed there is no practicable limit to the depth
at which the material would function, AquaBlok®
is typically produced for application in relatively
shallow, freshwater to brackish, generally
nearshore environments and is comprised of
bentonite clay with polymer additives covering a
small aggregate core. The bentonite clay is
comprised principally of montmorillonite, and the
proprietary polymer is added to further promote
the adhesion and coalescing of clay particles to
the aggregate core. The aggregate core is used
essentially for weighting to promote the sinking of
the AquaBlok® material to the sediment surface.
AquaBlok® functions by hydrating, swelling, and
forming a continuous and highly impermeable
isolation layer above contaminated sediments.
Based on information provided by the vendor,
AquaBlok® formulations experience a significant
swelling upon placement and hydration, and
freshwater formulations are characterized by
intrinsic permeabilities on the order of 10"9 cm/s.
-------
Sediment caps, which have been employed in the
field at hazardous waste cleanup sites, are an in
situ remediation technology for contaminated
sediment management. Sediment caps have the
general advantage of being low-cost and they do
not generate secondary waste streams requiring
disposition at a landfill or a constructed waste
containment facility (e.g., a confined aquatic
disposal [CAD] facility or confined disposal facility
[CDF]), as with sediment dredging.
2.2 Applicable Wastes
AquaBlok® capping material is designed to
function by swelling and forming a continuous
and highly impermeable isolation barrier between
contaminated sediments and the overlying water
column. As such, it is considered a non-specific
capping material that would function by
encapsulating any class or type of contaminant
as well as theoretically any range of contaminant
concentration.
AquaBlok® formulations can be modified to
include clays that are specifically more
appropriate for a particular environmental setting.
For instance, in a more saline environment,
attapulgite clay could be used instead of
bentonite (i.e., montmorillonite), and in other
situations, organoclays could be used in the
formulation. Formulations of AquaBlok® can also
be made to incorporate specific amendments
designed to react with certain contaminants. For
instance, activated carbon or zero-valent iron
(ZVI) amendments could be integrated into the
material to provide a reactive contribution to
address chlorinated organic (and potentially
other) contaminants. In addition, vendor supplied
information suggests AquaBlok® could be
designed using a "funnel and gate" approach,
where reactive pathways would be deliberately
integrated into an AquaBlok® cap to control
contaminant movement and treat contamination
moving through aqueous and/or vapor flux
mechanisms. However, the type of AquaBlok®
discussed in this ITER is a characteristically
"basic" formulation of bentonite, polymer, and
aggregate, the purpose of which is to provide an
effective isolation barrier for contaminated
sediment.
2.3 Technology Operability,
Availability, and Transportability
As discussed above, AquaBlok® is generally
considered a non-specific capping material
designed to provide a continuous and
impermeable barrier between contaminated
sediment and overlying surface water regardless
of contaminant nature or magnitude. However, in
some cases, it may be desirable to formulate an
AquaBlok® capping material to incorporate a
reactive component to specifically address some
contaminant, and other formulations may be
needed or desired on the basis of local
geochemical characteristics (e.g., salinity).
The overall operability of the technology is not as
strongly influenced by site-specific factors as a
terrestrial remediation approach would be given
its broadcast applicability to various waste types.
However, several factors could affect the
operability of AquaBlok® at a contaminated
sediment site and influence decision-making
related to specific material formulation. These
factors include, but are not necessarily limited to:
• Hydrology (including depth of surface water,
groundwater discharge and recharge
characteristics, and/or local flow velocities
and shear stresses);
• Physical and geochemical properties of the
surface water (including salinity, sediment
depositional characteristics, and/or tidal
characteristics);
• Physical, geotechnical, and ecological
properties of the contaminated sediment site
(including presence and distribution of
intertidal sediments and subtidal sediments,
sediment compressive strength, and/or gas
ebullition potential);
• Ecological properties of the contaminated
sediment site (including presence and
distribution of emergent or submergent plants,
fish, and/or benthos);
• Nature, distribution and magnitude of
contamination (as it relates to decisions
regarding the applicability or desirability of
-------
reactive amendments and required lateral
extent and/or thickness of capping material);
• Climatic conditions (as they relate to
variability in surface water or sediment
characteristics, such as tidal variability and/or
temperature effects on gas ebullition);
• Site characteristics and land use features
(including recreational uses, access concerns
or limitations, ongoing contaminant sources,
site reuse/redevelopment, and/or the need for
institutional or engineering controls);
• Remediation goals (including contamination-
related risk reduction and habitat
enhancement); and
• Short- and long-term monitoring requirements
(including sampling and analysis).
AquaBlok® is a commercially available technology
that has been successfully deployed during
environmental remediation projects. In its
component form, the aggregate and clay
materials used in AquaBlok® formulations are
readily available from common sources of earth
materials. For instance, bentonite is a readily
available material used in the well drilling
industry. The polymers used are proprietary and
developed directly by AquaBlok, Ltd.
The equipment needed to support the application
of AquaBlok® as a sediment capping remedy is
generally standard and not site specific. Such
equipment is, by and large, limited to equipment
required to convey the AquaBlok® material to a
site, move it around the site for staging purposes,
and place on the surface of the contaminated
sediment. Terrestrial earth moving equipment
(e.g., excavators or cranes) or water-based
moving equipment (e.g., barges with or without
cranes or excavator extensions) are both
commonly used to place AquaBlok®. In some
cases, AquaBlok® may be placed manually. In
other unique cases, due to the inability of using
other standard earth moving equipment or
remoteness of a site, tools such as a helicopter
may be required to place AquaBlok®. Equipment
required to monitor the performance and function
of AquaBlok® after placement (e.g., aquatic
geophysical surveying tools) is generally
specialized for the contaminated sediment
management arena, but is also generally
standard and readily available from a number of
vendors who specialize in this area.
AquaBlok® material and the equipment used to
deploy and monitor it can reasonably be
considered easily transportable. AquaBlok® is
generally packaged in large bags (e.g., 1 to 20
ton capacity) and transported to a site via truck or
rail, where it is managed and placed using
terrestrial and/or water-based earth moving
equipment. Alternatively, barges may be loaded
with bulk AquaBlok® material and the material
transported in this manner to a capping site. In
other cases, AquaBlok® could be formulated at a
cleanup site based on the proximity to sources of
earth materials or ease of access to modes of
transportation to move the earth materials
required to make the necessary formulation.
Terrestrial earth moving equipment is easily
transportable to a site by standard over-the-road
hauling, and aquatic earth moving equipment can
generally be navigated to a site along existing
waterways or mobilized using over-the-road
hauling much like terrestrial equipment.
Monitoring tools needed for an AquaBlok®
remedy are similarly easily transported via land or
waterways.
Capping contaminated sediment with AquaBlok®
is considered a single-use technology application.
AquaBlok® deployed at one cleanup site would
not be removed and redeployed at another site,
as might be done with a treatment system for
contaminated groundwater. In this sense,
AquaBlok® is not a "transportable" technology.
However, in the context of a single-use
technology, all materials and equipment used to
implement an AquaBlok® sediment capping
remedy are readily and easily transportable.
2.4 Range of Suitable Site
Characteristics
In general, any site with contaminated
subaqueous sediment would be compatible with
the deployment of an AquaBlok® sediment cap.
However, the practicability and consequent cost-
effectiveness of incorporating AquaBlok® into a
sediment remedy will vary based on the project
location, size of the project, accessibility for
-------
application, and remediation and restoration
objectives. Similarly, the implementability of an
AquaBlok® capping remedy could be constrained
by legal and/or regulatory restrictions or
allowances applicable on a site-specific basis.
AquaBlok® is a technology that reportedly can
provide a wide variety of management functions,
including permeability control, chemical
sequestration, physical stabilization, and
facilitation of in situ treatment. Thus, the specific
product formulation, application rate,
presence/absence of special additives or
incorporation of other materials as part of an
AquaBlok®-based composite cap design (e.g.,
geotextiles, sand, or armoring stone) would be
highly dependent upon site-specific conditions
and specific remediation goals.
An AquaBlok®-based cap can be designed for
many applications to meet multiple remediation
goals and can be used alone or in combination
with other materials based on restoration goals,
accessibility for application, long-term monitoring
goals, availability, regulatory requirements, and
relative cost of other materials such as sand or
armoring stone. In addition, the flexibility of cap
design must be considered in situations where
excessive cap thickness could negatively impact
available floodway cross-sections.
AquaBlok® claims to be effective and
advantageous in capping sediments in deeper-
water and higher-energy regimes, and an
AquaBlok®-based capping solution can address a
range of contaminated sediments including
metals and organic compounds. In addition,
AquaBlok® used in conjunction with "hotspot"
removal activities (e.g., to cap post-dredging
residual contamination) could potentially help
improve project efficiency by supporting the
acceptability of prescriptive removal of a specific
volume with the subsequent addition of an
isolation cap. This remedial strategy could
potentially not only significantly reduce
uncertainties but minimize project costs by
reducing the need for significant sampling and
subsequent re-dredging (both of which are often
required for environmental dredging programs
that typically target specific and conservative
post-dredging levels of residual contamination).
Formulations of AquaBlok® are available or can
be developed to cover a wide range of salinities,
meaning that riverine, lacustrine, deltaic,
estuarine, wetland, offshore, and tidally mixed
nearshore environments are all candidate sites.
In addition, AquaBlok® claims to be highly
resistant to erosion and could therefore be
deployed in environments with variable
hydrologic energy regimes. Furthermore,
AquaBlok® is a non-specific capping material, and
the use of this material is largely not constrained
by the nature or magnitude of the sediment
contaminant load.
Overall, the range of suitable site characteristics
allowing the consideration of an AquaBlok®
sediment cap is based on physicochemical site
setting and is very broad. However, limitations to
the use of AquaBlok® do exist and are discussed
in Section 2.7.
2.5 Site Support Requirements
In general, there are no site support requirements
to effectively deploy an AquaBlok® cap. All
materials and equipment, both to place and
monitor the cap, typically originate from off-site
sources and do not require specific site support.
In some cases, if AquaBlok® is deployed to the
subaqueous environment from land, a controlled
area may be needed to place and operate
equipment and to stage materials during
placement, and controls may be required to
prevent access to such work areas. However,
following construction, there would likely be no
permanent features other than the in-place cap,
and no long-term site support requirements would
therefore exist.
For capping from water using water-based
equipment, there are typically no site support
requirements of any kind, other than the potential
need to transport and deploy water-based
equipment from the landside portion of the site.
In this case, the same staging and site control
considerations may be valid.
It may be necessary to control access to or use of
the water body in the area where an AquaBlok®
cap has been placed. The use of a water body in
-------
the area of a sediment cap is often restricted
through the implementation and enforcement of
administrative control mechanisms and
engineering controls. This could be
accomplished, for instance, by using markings
(such as buoys) or posting signs indicating the
presence of a subaqueous remedy and restricting
site use in any way that could impact the function
of the cap (e.g., recreational uses/ anchoring
restrictions).
Monitoring an AquaBlok® cap following placement
would generally be conducted on a periodic basis
using equipment and personnel imported for each
monitoring event. As such, there would generally
not be permanent monitoring devices left in place
at the site and no site support requirements to
ensure the permanence and function of such
equipment.
2.6 Material Handling and Quality
Control Requirements
The placement of an AquaBlok® cap would
preclude the requirement to handle contaminated
wastes, as the technology is deployed as an
alternative to removing contaminated sediment.
As such, the contaminated material handling
requirements that might otherwise govern the
dredging, transportation, and disposal of
impacted sediment from a site would not be
pertinent.
AquaBlok® itself is a generally inert material
consisting of aggregate, clay, and polymer
additives. However, to the extent that specific
handling requirements would be relevant to the
components in AquaBlok®, these handling
requirements should be followed when placing
the material at a contaminated sediment site. All
generally acceptable practices for working with
and around heavy earth moving equipment
required to place an AquaBlok® cap should also
be adhered to. A detailed Health and Safety Plan
(HASP) should be in place to define the
necessary material handling and hazard
mitigation techniques to be followed when
deploying an AquaBlok® sediment capping
remedy.
When placing any cap, the strength of the
sediments being capped must be considered.
Often, contaminated sediments are fine-grained,
soft, and highly organic, and are not able to
sustain significant vertical loads without
significant mixing of cap material and native
sediment, resuspension of contaminated
sediment to the water column, or mass lateral
movement of native sediment (i.e., in a manner
commonly known as mud-waving). Therefore, it
would be important to approach an AquaBlok®
capping remedy with a clear focus on preventing
unwanted residuals mobilization or contaminated
sediment movement, and material handling
requirements may dictate slow, low energy
placement of the AquaBlok® instead of rapid,
broadcast application of the material through the
water column. Slow, low energy placement could
be accomplished using a crane or excavator to
place individual buckets of AquaBlok® under the
water surface and near the sediment surface.
However, in certain situations, the broadcast
placement of AquaBlok® with a split-bottom barge
or other means might be acceptable. In other
cases, it may be appropriate to slowly and
carefully place a single lift of AquaBlok® and
establish a solid and stable foundation on which
subsequent lifts could be placed less deliberately
by broadcast application. Such considerations
would be very important during the capping
design stage.
2.7 Technology Limitations
The most significant limitation to the application
of AquaBlok® as a viable sediment remediation
technology is its ability to remain hydrated.
Because the technology's function is predicated
on proper and sustained hydration of the clay
polymer material, there are certain environments
that would characteristically not support this
technology (e.g., sediments above the inundated
zone). According to the material vendor, the
capillarity of the material can promote hydration
upslope of the inundated environment (i.e.,
creating a continuously hydrated cap with the toe
of the cap submerged), and it is conceivable that
the clay material could remain hydrated enough
to remain functional during an unanticipated
dewatering event (e.g., an unanticipated drought
that temporarily lowered water levels). In
10
-------
addition, given its plasticity and based on vendor
claims, the material has shown an ability to heal
after dessication and/or freeze/thaw cycles.
Nevertheless, a proper design level consideration
would be that the environment of interest
continuously supports hydration of the AquaBlok®
material.
As with any capping technology, reducing the
effective cross-section of a water body could
have a significant bearing on its applicability and
desirability, particularly if the water body is used
for navigation, ship berthing, or recreation.
Similarly, sediment caps would only be effective if
the specific geomorphology of the site was
amenable to placing a cap. For instance, in a
riverine or coastal environment with very steep
slopes, a sediment cap would potentially be
subject to mass movement and could potentially
not be adequately relied on to remain in place.
Another limitation to the effectiveness of any cap
is the presence of and/or incursion of debris. For
instance, to properly deploy a sediment cap, it
would likely be required that substantial debris
(e.g., tree limbs, large boulders or concrete
rubble) that could represent a potential for cap
failure be removed from the contaminated
sediment surface. This would add mobilization,
operational, and disposal costs and require an
additional incremental amount of time to fully
implement any capping remedy. Similarly, in
environments where significant debris incursion is
anticipated, it could be necessary to provide for
some engineering mechanism to prevent the
potential for debris to influence the effectiveness
of the cap.
Of related concern is the presence of ice in a
water body and the potential for ice-driven scour
on the cap through ice shoves or ice flow, or even
ice-related damage through simple freeze-thaw
cycles leading to ice lenses, blistering, or frost
penetration (see, for instance, published
information on sediment remediation projects in
the Fox River [EPA, 2006], the Grasse River
[Alcoa, 2007], and Ottawa River [Hull &
Associates, Inc., 2002]). As such, while an
AquaBlok® cap, or any sediment cap for that
matter, may be a suitable remedial strategy for a
site in a temperate climate or hydrodynamic
regime without significant threat of ice or
prolonged freezing, it may not be appropriate
where significant ice impacts could occur (unless
specific engineering protections could be
constructed to mitigate this performance risk).
The demonstration summarized in this ITER did
not attempt to evaluate ice impacts in any way.
Gas ebullition from contaminated sediment could
represent a limiting characteristic of an
environment as it relates to the selection of
AquaBlok® as an appropriate remedy. Gas
buildup in contaminated sediment capped with a
highly impermeable material could lead to failure
of the capping material as the pressure of the
accumulating gasses seeks a route of escape. A
similar concern would likely not be associated
with a traditional sand capping material, the
permeability of which would typically allow gases
evolved from underlying sediments to dissipate
without likely compromising cap integrity.
The required monitoring approach for a sediment
remediation site could be an impediment to using
a capping technology. If there are inflexible post-
capping monitoring requirements that call for
significant and repetitive coring, for instance, the
very act of monitoring the integrity of the cap
could substantially decrease its effectiveness
(i.e., by removing a sufficient amount of cap
material to create essentially uncapped areas or
preferential contaminant migration pathways).
However, given its cohesiveness and tendency to
form a uniform and continuous low-permeability
layer, AquaBlok® is claimed to be capable of self-
repairing after being altered through physical
sampling (or gas release), and capping remedies
typically include a requirement for localized cap
repair if and as needed to provide continued
remedy effectiveness.
Common earth materials such as clays used in
environmental remediation applications can
contain traces of the same contaminants creating
the hazardous condition at a cleanup site. For
instance, common clays often contain heavy
metals in some concentration given the
ubiquitous geologic presence of metals and the
high affinity clays have for metals through cation
exchange. However, it is likely that the
hazardous condition at a cleanup site would be
11
-------
significantly greater in terms of concentration
compared to naturally-occurring metal loads in a
clay material applied at a site. In a related sense,
given the high affinity that clays have for metals
and potentially other ionic contaminants, it is
conceivable that an AquaBlok® cap could act as a
sink of such contamination from an underlying
contaminated sediment. In other words, it is
conceivable that contaminants could be "wicked"
into an AquaBlok® cap. However, once the
exchange capacity of clay is saturated, it is
unlikely such a phenomenon would exist, and it is
also unlikely that such sorbed contaminants
would represent a bioavailable source of risk
given the high binding capacity between clays
and metals (and other contaminants).
Given that AquaBlok® is a containment
technology, it could be inappropriate for sites
where there is a regulatory prerequisite for
treatment to reduce volume, toxicity, or mobility.
However, some formulations of AquaBlok® either
already developed or under development could
integrate the common mixture of clay, polymer,
and aggregate with a treatment component such
as activated carbon or ZVI.
Finally, given that AquaBlok® is a highly
impermeable clay material, it could be
inappropriate for sites where there is an abundant
ecological community that relies on a coarser
grained sediment habitat or where there could be
an anticipated detrimental impact on habitat and
ecology associated with placing a substantially
thick layer of ecologically "inert" material.
Similarly, in an environment where vegetation is
common, the growth or regrowth of rooted plants
could detrimentally affect any cap's performance
by creating root paths or by promoting root
uptake of contaminants.
2.8 Factors Affecting Performance
There are factors that could influence the
performance of an AquaBlok® cap at a
contaminated sediment site. By and large, the
factors that could influence the performance of an
AquaBlok® cap are the same issues identified in
Section 2.7 as being potential technology
limitations.
In an environment where water levels fluctuate or
contamination extends beyond the inundated
zone, the performance of AquaBlok® could be
affected by permanent or periodic lack of
complete hydration. In an environment with
significant debris incursion or the buildup of ice,
large debris items or ice flows/dams could scour
AquaBlok® material from the cap or become
lodged in the cap, thereby completely removing
the cap or creating channels through the
impermeable material (as noted in Section 2.7,
the demonstration summarized in this ITER did
not attempt to evaluate ice impacts in any way).
Where gas ebullition is a significant and/or
frequent occurrence, gas pressure buildups could
lead to cap failure, lessening the effectiveness of
the cap for some duration (i.e., until the cap is
able to self-repair or repairs can be made through
O&M design).
During monitoring activities, invasive sampling
(e.g., coring) could create isolated cap failure
regions by removing AquaBlok® material and
creating a channel for contaminant short-
circuiting. Similarly, if a water body in which an
AquaBlok® cap is deployed is used for
recreational purposes or is navigated by
watercraft, it is possible that anthropogenic
activity could undermine the effectiveness of the
cap. Specifically, anchor scour or propeller wash
could be responsible for removing AquaBlok®
material, limiting its effectiveness relative to
remediation design criteria.
Another critical performance-limiting factor
common to all caps is the potential for
contaminated sediment to be deposited on top of
the cap either as a result of resuspension during
cap placement and/or the deposition of new
sediment contaminated by ongoing sources.
Lastly, an AquaBlok® cap could be limited in its
overall effectiveness if remediation performance
requirements include accomplishing contaminant
treatment (unless used in conjunction with other
treatment options) and/or not altering an
ecological equilibrium.
The means of mitigating these potential
performance-affecting factors are generally
fivefold. The first is to properly and reasonably
12
-------
select a site in the context of RAOs before
deploying an AquaBlok® cap. For instance, it
could be appropriate to consider AquaBlok® for a
contaminated sediment site where the
introduction of a clay-based cap would not
significantly alter ecological health as it relates to
substrate if this were a key RAO. The second is
to properly design the capping remedy, including
an appropriate means of deploying the
AquaBlok® cap to limit the resuspension of
contaminated sediment that could subsequently
recontaminate the clean cap surface and
consideration of adequate armoring against
anticipated cap damage. The third is to develop
and implement an adequate monitoring plan,
potentially incorporating a restoration contingency
to repair any damage to the AquaBlok® cap, to
ensure its continued effectiveness. The fourth is
to ensure that potential ongoing sources have
been controlled and/or eliminated. The fifth is to
execute and maintain any and all institutional
controls (ICs) that would serve to limit or prevent
activities that could directly impact the integrity
and effectiveness of the cap. As suggested by
this information, these are common and generally
simple means to mitigate against the potential for
reduced effectiveness, and are typically
considerations of any remedial action.
2.9 Site Reuse
Overall, it is likely that an AquaBlok® cap could be
designed to properly and successfully integrate
into a full spectrum of site reuse scenarios for a
contaminated sediment site. For instance, an
AquaBlok® cap could be designed to provide
appropriate levels of risk management even
within the context of a subsequent construction
plan calling for a marina, pier, or some other
structure.
However, as suggested in the previous section, it
is unlikely that a completely unrestricted site
reuse would be acceptable at a contaminated
sediment site where AquaBlok® is deployed, as
contaminated sediment would be left in place and
a long-term monitoring and maintenance program
would be required to ensure remedy integrity and
function. Accordingly, it is likely that some form
of site use restrictions would be in place in the
form of ICs. For instance, recreational use
restrictions could be executed to restrict the size
of vessels that could pass over the cap or the
speed of these vessels to eliminate the potential
for propeller scour. Moreover, it is possible that
physical access restrictions (e.g., buoys
indicating an exclusion area) could be deployed.
It is also quite likely that site uses that could lead
to the capture and/or consumption of potentially
contaminated food items would be limited or
prohibited. For instance, it is common at
contaminated sediment sites for fishing
advisories to be in place for the duration of a
capping remedy to restrict or prohibit the catch
and consumption of fish, or at least until
appropriate monitoring verifies that no risk
remains through this pathway (i.e., monitoring to
verify fish tissue concentrations at an acceptable
level).
2.10 Feasibility Study Evaluation Criteria
The overall suitability of a remediation technology
for the conditions at any particular CERCLA
cleanup site is assessed in the context of nine
feasibility study (FS) criteria prior to preparing a
detailed remedial design (RD) and actually
constructing the remedy. The following sections
describe the generally anticipated performance of
AquaBlok® as a contaminated sediment
remediation technology relative to each of these
criteria. In general, capping, along with dredging,
is a commonly accepted standard approach for
addressing contaminated subaqueous sediments.
Given this, it stands to reason that capping is
generally characteristically feasible when
evaluated in the context of the CERCLA
feasibility criteria.
2.10.1 Overall Protection of Human
Health and the Environment
Contaminated subaqueous sediments generally
create unacceptable risk in three ways. The first
is a general environmental health risk associated
with the potential for degradation of aquatic or
nearshore habitat resulting from the presence of
contamination in sediment. The second is human
health risk associated with potential direct human
contact with and/or incidental ingestion of
contaminated sediment and/or surface water
13
-------
containing contamination emanating from the
sediment. The third, and generally most critical
from a risk management standpoint for typical
contaminated sediment sites, is associated with
the potential for ecological receptors to be
exposed to contamination in the sediment either
through direct contact with and/or incidental
ingestion of contaminated sediment and/or
contamination in surface water emanating from
the sediment, and/or feeding on lower trophic-
level organisms that themselves are exposed to
contamination in the sediment and/or surface
water. Subsequently, a risk to human health can
be posed by consuming organisms potentially
impacted in this manner.
The AquaBlok® sediment capping technology is
designed to isolate sediment contamination from
the overlying water column, effectively eliminating
the source of contaminant exposure. A concern
at most if not all contaminated sediment sites is
the likelihood that bioturbation or some other
physical or ecological mechanism could mix
contamination into the clean cap interval.
Compared to other common capping materials
used at contaminated sediment sites (e.g., sand
or clean sediment), it is possible that AquaBlok®
would yield a lower probability of mixing given its
cohesiveness. Alternatively, it is possible that
deploying an AquaBlok® cap could upset a site-
specific ecological balance by placing a
substantially thick and ecologically "inert" layer
over the native sediment, or by replacing a
coarser grained substrate to which native flora
and fauna have adapted with a clay material
substrate. This latter potential impact could be
overcome by using other materials as part of a
composite cap (e.g., by covering AquaBlok® with
sand).
Overall, in the context specifically of isolating
contamination, AquaBlok® would be anticipated to
provide for the overall protection of human health
and the environment, and perhaps to a greater
degree than other more commonly used capping
materials (e.g., sand or clean sediment) given its
specific design. With respect to physical effects
on the environment, any capping remedy would
need to be evaluated in the specific context of
compatibility with existing conditions.
2.10.2 Compliance with Applicable or
Relevant and Appropriate
Requirements
Applicable or relevant and appropriate
requirements (ARARs) for a contaminated
sediment cleanup action are generally more
numerous for dredging and the disposition of
removed sediment than capping. However, there
are a number of ARARs that are typically
pertinent to sediment capping approaches,
including water quality standards and biological
resource protection standards that may be
applicable both during and after cap placement.
While certain areas do have promulgated
sediment cleanup standards, remediation goals
are generally developed for a particular site to
protect human and/or ecological receptors on the
basis of risk assessments, and are not generally
ARARs in and of themselves.
Overall, it is anticipated that an AquaBlok® cap
could be designed for any particular sediment
cleanup site where this technology would be well
suited to be compliant with all pertinent ARARs.
2.10.3 Long-Term Effectiveness and
Permanence
Wth an AquaBlok® sediment capping approach,
contaminated sediment would remain in place,
but would be covered by an impermeable and
continuous isolation barrier that would mitigate
against human health and ecological risks. In
addition, it is highly likely that any sediment
capping remedy would be accompanied by the
execution and maintenance of ICs. To ensure
the integrity of such a remedy, a long-term
monitoring plan would typically be required in
addition to a maintenance plan specifying repair
requirements to ensure continued remedy
effectiveness.
As described in Section 2.10.1, a concern at most
if not all contaminated sediment sites is the
likelihood that bioturbation or some other
mechanism could mix contamination into the
clean cap interval. Compared to other common
capping materials used at contaminated sediment
sites (e.g., sand or clean sediment), it is possible
14
-------
that AquaBlok® would yield a lower probability of
this phenomenon given its cohesiveness and
subsequent resistance to mixing. In addition,
AquaBlok® is generally highly resistant to erosion
given its composition, and would likely therefore
be more stable than traditional sand capping
material in environments with high flow and shear
energy. Similarly, given its composition,
AquaBlok® would potentially be more effective
over a wider range of geomorphologic conditions.
For instance, relative to sand, AquaBlok® would
likely be more stable on steeper slopes.
However, it is possible that deploying an
AquaBlok® cap could upset a site-specific
ecological balance by covering existing benthos
or by replacing a coarser grained substrate to
which native flora and fauna have adapted, and
therefore may be ineffective in maintaining or
sustaining a viable ecological setting.
Alternatively, given its grain size composition is
predominantly more similar to the sediment
encountered at most contaminated sediment
sites (i.e., fine-grained), AquaBlok® may be more
effective at promoting the restoration of
ecological equilibrium following capping
compared to a more traditional capping material
such as sand.
Overall, in the context specifically of its ability to
isolate contamination, AquaBlok® is an alternative
that would be anticipated to be highly effective in
the long-term, and perhaps to a greater degree
than other more commonly used capping
materials (e.g., sand or clean sediment) given its
erosion resistance, physical stability, and
impermeability.
2.10.4 Reduction of Toxicity, Mobility,
and Volume through Treatment
An AquaBlok® sediment cap would cover
contaminated sediments left in place. While
there are formulations of AquaBlok® currently
under development that could accomplish some
level of in situ treatment by integrating reactive
components, the common formulation of
AquaBlok® discussed in this ITER would not lead
to any type of treatment of contamination in the
sediment other than potentially simple adsorption
of certain contaminants that have an affinity for a
clay matrix. Rather, AquaBlok® would be used
for simple isolation of sediment contamination
and elimination of the source of exposure to
human or ecological receptors through transport
to the water column. The volume of
contamination would therefore not likely be
materially affected directly by the presence of
AquaBlok®. However, while treatment would not
be responsible, per se, the toxicity and mobility of
sediment contaminants would be reduced by
directly eliminating the pathway between
contamination and receptors. Also, in many
respects (i.e., as relates to the low permeability,
cohesiveness, and erosion resistance of the
AquaBlok® material), AquaBlok® could potentially
reduce toxicity and mobility of contaminants to a
greater degree compared to more common
capping materials (e.g., sand or clean sediment)
by more effectively isolating contaminants and
therefore providing more "contact" between
contamination and active degradation
mechanisms in the contaminated sediment
interval.
2.10.5 Short-Term Effectiveness
The short-term effectiveness of a remediation
technology is generally measured relative to its
short-term impacts on the environment and risk to
the community during construction.
For an AquaBlok® sediment cap, deployment
would likely occur over a relatively short duration,
although construction duration for any sediment
capping technology would be predicated on the
area over which a cap would be placed, the
design specifications of the cap (e.g., thickness),
and/or the geotechnical properties of the
contaminated sediment (e.g., compression
strength). Construction activities would likely be
limited to the water or the nearshore terrestrial
environment, meaning that there would be little
risk of exposing the community to short-term
implementation hazards.
As is generally true for any capping project, the
most significant short-term risks associated with
an AquaBlok® sediment cap would be associated
with transporting material and equipment to a site
and physically placing the cap. A capping
remedy would lead to increased traffic to a site
15
-------
for some duration, which could lead to short-term
risk to the community and/or environment (i.e., by
increased barge traffic through navigable aquatic
environments or increased truck traffic between
suppliers and a site). Placing an AquaBlok® cap
could lead to disturbance of aqueous habitat or
short-term impacts to ecological receptors from
equipment operation, suspension of sediment,
and alterations to general geochemical surface
water quality (e.g., dissolved oxygen [DO]).
Placing an AquaBlok® cap could also suspend
contaminated sediment for some period of time.
In addition, workers involved in constructing an
AquaBlok® cap (or any sediment cap) would be
exposed to work hazards associated with work on
water that are unique relative to more common
terrestrial cleanup work.
The construction duration for an AquaBlok®
sediment capping approach, or any other
sediment capping approach, would likely be short
(assuming cap placement over a limited sediment
area) and would likely be characterized by a
localized construction area. In addition, there are
numerous best management practices (BMPs)
and mitigation strategies that would limit the
short-term risks of an AquaBlok® or any other
sediment capping approach. For instance, silt
curtains could be deployed and water quality
monitoring conducted during cap placement to
minimize the potential to adversely impact
ecological receptors. Workers would be
protected against hazards by a HASP as they
would at any hazardous waste site. Given that
contaminated sediments would remain in place
and be covered there would be a very low overall
risk of being exposed to site contamination.
Overall, therefore, the anticipated short-term
effectiveness of an AquaBlok® sediment capping
approach would be high.
2.10.6 Implementability
In general, all materials and equipment needed to
deploy an AquaBlok® sediment cap are readily
obtainable. In addition, the methods used to
construct, maintain, and monitor an AquaBlok®
cap are all generally standard, as are the
mechanisms typically used to execute and
maintain ICs. Overall, the implementability of an
AquaBlok® cap at any particular contaminated
sediment site would generally be anticipated to
be high.
2.10.7 Cost
Capping, along with dredging, is typically
considered the standard remedial approach for
contaminated sediment sites. These methods
are considered standard sediment cleanup
strategies in part because they are the most cost-
effective methods for addressing the variable
mixture of contaminants typically found in
contaminated sediments and because there is a
general lack of available and proven in situ
treatment alternatives in such cases.
Relative to other typical sediment capping
materials (e.g., sand), AquaBlok® would tend to
be more costly. However, as indicated in Section
1.2, an AquaBlok® cap could potentially require
less thickness to achieve RAOs given its
impermeability and other physical characteristics,
which could offset some of the additional cost of
the material itself. A detailed economic analysis
for the AquaBlok® sediment capping technology is
provided in Section 4.0.
2.10.8 State Acceptance
AquaBlok® was included in the ROD for the Eagle
River Flats Superfund site in Alaska for the in situ
management of impacted sediments. Since that
time, AquaBlok® has, according to vendor-
supplied information, been successfully deployed
as a sediment remediation technology at 10
remediation sites, and has been evaluated at
bench-scale at several others. Because State
acceptance for this technology would likely be
related to the effectiveness of the AquaBlok®
material at providing contaminant isolation, it is
anticipated that the material would be regarded
favorably as a suitable capping alternative.
2.10.9 Community Acceptance
AquaBlok® would potentially be an attractive and
desirable capping option in the eyes of the
community for any given sediment cleanup site
given its impermeability and long-term stability.
In comparison to other traditional capping
materials (e.g., sand and clean sediment), it may
even be considered more desirable in the context
16
-------
of these characteristics. In addition, the
perception that AquaBlok® is a more specifically
engineered capping material compared to sand
or other sediment could play a part in community
acceptance.
Overall protection of human health and the
environment and compliance with ARARs are
considered threshold criteria, in that any remedy
must meet these to be considered appropriate.
The remaining criteria other than state and
community acceptance are considered balancing
criteria that allow remedial alternatives to be
differentiated from one another. State and
community acceptance are considered modifying
criteria generally summarized in remediation
decision documents.
Table 2-1 summarizes the evaluation of
AquaBlok® against the nine CERCLA FS
evaluation criteria, in the context of its anticipated
performance compared to a sand-only sediment
cap.
2.11 Permitting
The applicability of specific permit programs for
installing an AquaBlok® cap at a contaminated
sediment site would be dependent on the type of
waste, the habitat, receptors, and environmental
setting at the site, and the federal, state, and/or
local environmental laws, regulations, and
ordinances in place. For a CERCLA capping
remedial action, an ARARs determination would
be made to define the universe of federal, state,
and/or local environmental laws, regulations, and
ordinances that would guide the remedy
execution. For a non-CERCLA capping action at
a contaminated sediment site, a process similar
to an ARARs determination would be executed to
determine applicable laws, regulations,
ordinances, and permits. It is likely that the
specific ARARs identified for any capping remedy
at any particular site would be largely identical.
17
-------
Table 2-1. Summary of AquaBlok® Performance Expectations Relative to CERCLA Feasibility Criteria
CERCLA
Criterion
Key Factors
Influencing
Determination,
Ranking, or
Probability
Anticipated
Performance
Relative to
Sand-only Cap
of Similar
Thickness
Protective of
Human
Health and
the
Environment
-Not specific
to any type of
contaminant
or degree of
contamination
-Low
permeability
-Highly
cohesive and
stable
-Lower
potential for
bioturbation
mixing
-Potential for
ecological
impacts
SAME TO
HIGHER
Compliant
with
ARARs
-Potential
applicability
of water
quality
standards
and/or
resource
protection
standards
-Permit
programs
SAME
Long-term
Effectiveness
and
Permanence
-Highly stable
and resistant
to erosion
-Effective
long-term
maintenance
and
monitoring
-Institutional
controls to
protect
receptors
-Potential for
ecological
impacts
HIGHER
Reduction of
Toxicity,
Mobility, and
Volume
through
Treatment
-No treatment
per se (without
incorporation
of reactive
amendments
or combination
with other
technologies)
-Toxicity and
mobility
reduction
through
isolation and
sorption
SAME TO
HIGHER
Short-term
Effectiveness
-Increased
traffic for
transportation
of materials
and
equipment
-Likely a
limited
construction
area
-Potential for
habitat
impacts
-Potential for
sediment
resuspension
-Best
management
practices
SAME
Implementability
-Readily available
equipment and
materials
-Standard
methods for
construction and
monitoring
-Institutional
controls generally
easily
implemented and
maintained
SAME
Cost
-Size of
capping area
-Nature and
extent of
monitoring
and
maintenance
requirements
SAME TO
HIGHER
State
Acceptance
-Sensitivity of
habitat
-Recreational
or other value
of site
-Contaminant
isolation
capacity
relative to
contaminant
type,
distribution,
and
concentration
SAME TO
HIGHER
Community
Acceptance
-Sensitivity of
habitat
-Recreational
or other value
of site
-Contaminant
isolation
capacity
relative to
contaminant
type,
distribution,
and
concentration
SAME TO
HIGHER
oo
-------
Section 3
Technology Effectiveness
This section discusses the SITE demonstration
that was conducted to evaluate the effectiveness
of the AquaBlok® sediment capping technology at
pilot-scale at a contaminated sediment site. It
describes the site where the AquaBlok® capping
technology was demonstrated, the physical
construction of the AquaBlok® cap (and other cap
types) evaluated, the measurements and data
acquisition that were completed to evaluate the
effectiveness of the AquaBlok® cap, and the
overall results of the demonstration.
This section is structured as follows: Section 3.1
describes the SITE demonstration program and
its physical location and environmental setting;
Section 3.2 describes the SITE demonstration
approach and methodologies in general and
specific terms; and Section 3.3 describes and
summarizes the SITE demonstration results. The
reader can advance directly to Section 3.3 to
read about the SITE program results only.
3.1
AquaBlok® SITE Demonstration
Program Description
The Anacostia River is a freshwater tidal river
system flowing approximately 8.5 miles (mi) from
Prince George's County in Maryland, through
Washington, DC, to its confluence with the
Potomac River at Mains Point, draining nearly
180 square miles (mi2) in Maryland and
Washington, DC. Flow in the Anacostia River is
generally considered "sluggish", with mean
annual discharge of approximately 1,000 gallons
(gal) per second (gal/sec). Hydrologic records
available since 1986 indicate a minimum
discharge of approximately 13 gal/sec and a
maximum of over 230,000 gal/sec in the
Anacostia River. The Anacostia River watershed
is within the larger Potomac River Drainage
Basin, which in turn empties to Chesapeake Bay.
Figure 3-1 shows the Anacostia River watershed
and the larger Potomac River/ Chesapeake Bay
system.
Sediments in the Anacostia River are
contaminated with polycyclic aromatic
hydrocarbons (PAHs), polychlorinated biphenyls
(PCBs), heavy metals, and other chemicals to
levels that have hindered commercial, industrial,
and recreational uses. Stretches of the
Anacostia River are listed on the National
Priorities List (NPL) of Superfund sites (i.e.,
CERCLA) due to the levels of contamination,
habitat degradation, and risks posed to human
health and the environment. The most likely
sources of contamination in the Anacostia River
are historical and/or present widespread
industrial activity, diffuse urban runoff, direct
discharge of untreated sewage, and military
activity.
Given the economic, logistical, technological, and
ecological limitations of sediment removal and
treatment technologies for the conditions typically
encountered in the Anacostia River, sediment
capping has the potential to afford significant
advantages for contaminated sediment
management. Accordingly, EPA, in cooperation
with the Louisiana State University (LSU)
Hazardous Substance Research Center (HSRC)
and the Anacostia Watershed Toxics Alliance
(AWTA), implemented an investigation of
innovative capping technologies for their use in
the management of contaminated sediments in
the Anacostia River.
AWTA, formed in March 1999 as a voluntary
partnership to focus on addressing toxic sediment
contamination of the tidal Anacostia River, is led
by EPA Region 3, and includes potentially
responsible parties, regulatory agencies including
EPA and the National Park Service, the United
States Navy (Navy), and several industrial
19
-------
The Anatostia
jp ; MARYLAND
. ' .J *K Ifj ^ »
,^,r,e^. J^ ,
' = ,'« ^-iPf •«•»'•'!• \-_
;w^,m'| fj£V| x-,
1 * i\ X^P* I ''I • «•«"•»!' )
. % •'
-------
its affiliates, and AWT A to ensure that similar and
comparable sampling and measurement
techniques, sampling locations, and analytical
methods were used, while not exceeding the
available EPA SITE budget for the
demonstration.
The overall goal of the AquaBlok® SITE
demonstration was to evaluate the efficacy of
AquaBlok® as a potential tool for the
management of contaminated sediments. The
evaluation was completed by comparing the
performance of AquaBlok® against a traditional
sand cap and established control sediments
relative to several measurement endpoints.
This ITER discusses only the AquaBlok® SITE
demonstration and does not discuss the HSRC
study.
3.1.1 AquaBlok® SITE Demonstration
Study Area Description and History
Preliminarily, two study areas in the Anacostia
River adjacent to the Washington Navy Yard in
southeastern Washington, DC were selected for
the AquaBlok® SITE demonstration (see Figures
3-2 and 3-3). Study Area 1 is located near the
south end of the Washington Navy Yard and
northeast of the South Capitol Street Bridge. It is
also located immediately offshore of a combined
sewer outfall (CSO) at the Washington Area
Water and Sewer Authority (WASA) O-Street
pumping station facility and immediately
upstream of the Earth Conservation Corps (ECC)
office that occupies a historical (inactive) surface
water pumping station built on piers in the river.
Study Area 2 is located in the vicinity of a former
manufactured gas plant (MGP) site on the north
end of the Washington Navy Yard. Both study
areas are outside the navigable channel of the
river.
For logistical and budgetary reasons, only Study
Area 1 was selected to implement HSRC's
federally-funded study of active cap technologies
and the EPA SITE demonstration of the
AquaBlok® capping technology. Accordingly,
throughout this ITER, the demonstration area
refers specifically to Study Area 1.
3.1.1.1 Physical and Chemical Setting of
AquaBlok® SITE Demonstration. The
demonstration area is characterized by a
generally shallow water depth (varying between
approximately 4 and 18 feet [ft] below mean
lower low water [MLLW] on average), and is
tidally influenced. Net surface water flow
direction is from the northeast to the southwest,
towards the Potomac River, but flow reversals
are common in conjunction with high tides. From
the shoreline to the navigable Anacostia River
channel, riverbed sediments in the demonstration
area generally slope at an approximately 4%
grade. Baseline flow velocities in the
demonstration area are generally in the range of
0.1 to 0.7 ft per second (s) (ft/s). Sediments in
the demonstration area generally consist of soft,
compressible, highly organic, plastic silty clay to a
depth of at least 10 ft below the sediment
surface.
Given documented contamination conditions in
the Anacostia River, contaminants of concern
(COCs) selected for the AquaBlok® SITE
demonstration were PAHs, PCBs, and metals.
River bottom sediments in the demonstration
area are contaminated with total PAH
concentrations up to 30 milligrams per kilogram
(mg/kg) and total PCB concentrations generally
between 6 and 12 mg/kg. Heavy metal
contaminants identified in the demonstration area
include cadmium (Cd) at concentrations of
generally 3 to 6 mg/kg, chromium (Cr) at
concentrations of generally 120 to 155 mg/kg,
copper (Cu) at concentrations of generally 127 to
207 mg/kg, lead (Pb) at concentrations of
generally 351 to 409 mg/kg, mercury (Hg) at
concentrations of generally 1.2 to 1.4 mg/kg, and
zinc (Zn) at concentrations of generally 512 to
587 mg/kg.
3.1.1.2 AquaBlok® SITE Demonstration
Cap Design and Construction. Figure 3-4
shows the cap study design layout for the
demonstration area, including all of the capping
areas constructed and assessed during the
AquaBlok® SITE demonstration and the HSRC
innovative capping technology evaluation. As
described in Section 3.1, the SITE demonstration
focused on the performance of AquaBlok® while
HSRC evaluated the effectiveness of apatite,
21
-------
I /ftarraad LJ1 Bfl.*ja
Figure 3-2. Locations of Preliminary AquaBlok® SITE Demonstration Study Areas
Figure 3-3. Aerial Image of Preliminary AquaBlok® SITE Demonstration Study Areas
22
-------
Area of material shortage in
the southwest corner of the
AquaBlok® cell responsible
for non-symmetrical as-built
layout.
Figure 3-4. AquaBlok® SITE Demonstration Area Capping Cell Layout
coke breeze, and AquaBlok® caps. However, for
both studies, a traditional sand cap area was
established to serve as a point of comparison
between traditional and innovative technologies,
and an uncapped control cell was established to
provide a baseline for reference comparisons.
These areas are also depicted on Figure 3-4.
The sand-only cap was designed to consist of
approximately 1 ft (30 centimeters [cm]) of clean
sand placed on the contaminated sediment
surface. The AquaBlok® cap was designed to
consist of approximately 4 inches (in) (10 cm) of
AquaBlok® material (after hydration) placed on
the contaminated sediment surface and
approximately 8 in (20 cm) of clean sand placed
atop the AquaBlok® layer. The control sediment
area was established immediately offshore of the
AquaBlok® and sand cap areas (i.e., in the
direction of the navigable river channel). During
cap construction, there were tolerances for
acceptable thickness in each capping cell, as
summarized in Table 3-1 (Home Engineering
Services, Inc. [Home], 2004).
Each cap area was designed to cover an area of
approximately 100 ft by 100 ft (30.5 meters [m] by
30.5 m), for a total area of 10,000 ft2 (930 square
meters [m2]). The uncapped control area did not
receive any capping treatment, but was selected
to cover an area roughly the same size as the
capped cells. During cap construction, a
collective decision was made by key project
personnel (not including Battelle or EPA NRMRL)
to limit the sand and AquaBlok® caps to 8,000 ft2.
The caps, including AquaBlok®, were constructed
by HSRC pursuant to its federally-funded
technology evaluation program and independent
of the AquaBlok® SITE demonstration. As such,
EPA NRMRL and Battelle were not directly
involved in the cap construction activities. All
23
-------
Table 3-1. Capping Cell Construction Design and Tolerances
Cap Cell
AquaBlok181
Cell
Sand
Cell
Material
AquaBlok181
Sand
Sand
Target
Thickness (in)
4 (hyd rated)
8
12
Total Target
Thickness (in)
12
12
Acceptable Thickness Range (in)
Minimum
2
4
4
Maximum
8
12
14
Figure 3-5. AquaBlok® in 2-Ton SuperSack at Staging Area
necessary authorizations were obtained and work
plans developed by HSRC for the cap
construction work.
On March 15, 2004, mobilization was completed
by HSRC for the construction of the
demonstration area capping cells. Materials used
to construct the sand and AquaBlok® caps were
staged at the General Services Administration
(GSA) property adjacent to the Washington Navy
Yard under an agreement between HSRC and
GSA.
AquaBlok® was delivered by flatbed trailer in a
total of 55 palletized SuperSacks (i.e., large
plastic bags) with approximately 2-ton capacity
(see Figure 3-5). Each bag was unloaded with a
forklift at the GSA property. AquaBlok® bags
were placed on and covered with polyethylene
sheeting to prevent contact with precipitation
because of the highly water-sensitive nature of
the product.
Sand was not packaged in any form and was
delivered to the GSA site by standard 20-ton
dump truck (see Figure 3-6). Approximately
1,355 tons of sand was delivered to the site in 64
truckloads. The sand was delivered both before
and during the cap construction period, with
deliveries coordinated to coincide with cap
placement material requirements.
A loader or forklift was used to transfer the
materials stored at the GSA staging area to a
24
-------
- -t _,
Figure 3-6. Sand Cap Material Stored in Bulk at Staging Area
Figure 3-7. Transferring Cap Material to Barge Using Conveyor
25
-------
Figure 3-8. Crane Barge Used to Place Caps in Demonstration Area
nearshore loading area. A belt conveyor system
was then used to transfer the materials from the
nearshore material stockpile area to a material
barge secured along the shoreline (see Figure 3-
7). After loading, a tugboat was used to move
the material barge beside a crane barge that was
used to actually construct the caps (see Figure 3-
8).
The crane barge was secured using long anchor
cable lines attached to anchors deployed outside
of the demonstration area to avoid potential
impacts to cap integrity. The tugboat was always
positioned outside of the capping area to avoid
potential propeller scour in the capped areas. The
material barge was secured to the crane barge
during cap placement.
A two cubic yard (cy) clamshell bucket was used
on the crane boom to slowly release materials
above the water surface. This clamshell bucket
was selected because it could most efficiently
and cost-effectively meet the cap thickness
design requirements. Cap material was applied
to the target cap areas in a broadcast fashion
using a Windows®-based version of the Offshore
Positioning Software (WINOPS) system for
location control. Real-time positioning data from
the WINOPS system assisted the crane operator
in achieving consistent coverage and proper
thickness of cap material across the capping
areas.
The actual placement of cap material began on
March 17, 2004. The sand cap was placed first,
followed by the AquaBlok® cap. The cap
placement procedures consisted of first retrieving
the capping materials from the material barge
with the clamshell bucket, moving the clamshell
bucket to the desired application location using
the WINOPS system, opening the bucket slowly
above the water surface and swinging it across
the targeted area in an arc to allow even
dispersal of cap material, and marking the
26
-------
application location on the WINOPS monitor.
The capping generally started near the shoreline
and worked away from the shoreline, parallel to
the navigation channel, such that previously
capped areas were not disturbed by the capping
operations or movements of the capping
equipment.
Sand cap placement was completed on March
19, 2004. AquaBlok® cap placement (including
the sand cover layer) started on March 22, 2004,
and was completed on March 27, 2004. As
constructed, the sand cap cell covered
approximately 8,000 ft2 (743 m2) and the
AquaBlok® cap covered approximately 7,200 ft2
(680 m2). The AquaBlok® cap footprint was not
completed entirely in the southwest cell corner
(see Figure 3-4). This was related to a material
shortage to achieve the design thickness
throughout the capping cell (i.e., more AquaBlok®
was placed during initial releases than was
needed to achieve the design thickness, leaving
less material for subsequent areas).
Consequently, the size of the area capped was
reduced relative to the original target area. This
material shortfall did not compromise the
demonstration program.
Silt curtains were deployed during cap placement
to create a temporary boundary and reduce the
migration of broadcast-applied cap materials and
potentially resuspended sediments downstream
and/or to other capping cells. The silt curtains
also maximized (by limiting dilution) the ability to
visually determine any increases in turbidity or
potential contaminant levels as a result of
resuspension during cap placement activities.
The silt curtains were placed from shore to shore
(i.e., in an arc) along the outside of the cap
perimeter (see Figure 3-9) which eliminated the
need for a silt curtain along the shoreline. Silt
curtains were also used to separate the various
capping cells to minimize cross-contamination
between work areas.
There were no project specific compliance
requirements for water quality during cap
construction. However, water quality monitoring
was conducted during the capping operations to
verify that turbidity and DO concentrations were
maintained within acceptable ambient water
quality criteria. In addition, target contaminants
(i.e., metals, PAHs, and PCBs) were analyzed for
in surface water samples collected inside the
capping material release areas and outside of the
silt curtains to evaluate the potential for
resuspension of contaminated sediment into the
water column. Water quality monitoring was
conducted during the cap placement using
submersible water quality monitoring equipment
(i.e., a multifunction water quality analyzer). All
water quality monitoring was performed at a
depth of approximately 6 in below the water
surface at established monitoring stations.
Overall, the monitoring information indicated
there was no impact to water quality related to
the capping project (Home, 2004).
3.2 AquaBlok® SITE Demonstration
Approach and Methods
The overall goal of the AquaBlok® SITE
demonstration was to evaluate the efficacy of
AquaBlok® as an innovative remedial approach
for management of contaminated sediments.
The following specific objectives were identified
by EPA NRMRL for study within the context of
the AquaBlok® SITE demonstration:
• Objective #1 - Demonstrate the physical
stability of an AquaBlok® cap in the
Anacostia River under stresses associated
with normal river flows and high-flow
events, and determine theoretical hydraulic
stresses under which the cap could fail.
Compare these stabilities with traditional
sand capping technology and uncapped
(control) sediments.
• Objective #2 - Demonstrate the ability of an
AquaBlok® cap to control groundwater
seepage influenced by regional gradients or
tidal pumping (or both) relative to seepage
through sand-capped and uncapped
(control) sediments.
• Objective #3 - Demonstrate the influence of
an AquaBlok® cap on the benthic flora and
fauna expected to populate Anacostia River
sediments relative to the influence of sand-
capped and uncapped sediments.
27
-------
Figure 3-9. Silt Curtains Deployed Around Demonstration Area Capping Cells
In parallel with the AquaBlok® SITE
demonstration, HSRC conducted an evaluation of
AquaBlok® with a unique but complementary set
of objectives. As described above, this ITER is
intended only to discuss the implementation and
results of the AquaBlok® SITE demonstration and
not the HSRC study.
A series of field monitoring events were
conducted over an approximately three-year
period after the caps were placed. A number of
investigation tools were used during these events
to gather important technology performance data.
The results of these various data gathering
events form the basis of the conclusions
conveyed in this ITER related to the performance
of the AquaBlok® sediment capping technology,
including the relative performance in comparison
to traditional sand-capping technology.
Specifically, field activities were implemented one
month following completion of the cap
construction, and then at six months, 18 months,
and 30 months following cap construction.
Accordingly, the first post-capping field event
occurred in the spring of 2004 and the second in
the fall of 2004. The remaining events occurred
annually thereafter, in the fall of 2005 and 2006.
The critical and non-critical measurements
collected for each of the primary AquaBlok®
demonstration objectives, the various
measurement tools utilized during the four post-
capping monitoring events, and a summary of the
principles of and methods employed for each
measurement tool are summarized below.
3.2.1 Critical and Non-Critical
Measurements
Critical measurements are those that were
deemed of fundamental importance or of
absolute necessity to fully evaluate a
demonstration objective. Non-critical
measurements are those that were deemed to
provide ancillary or incremental value to
understanding a condition or evaluating a
demonstration objective. A series of critical and
28
-------
non-critical measurement endpoints were
developed for each of the primary AquaBlok®
SITE demonstration objectives, as follows:
• Objective #1 - Demonstrate the physical
stability of an AquaBlok® cap.
The physical stability of AquaBlok® in
flowing water depends primarily on the
material's physical strength (e.g., shear
strength) and its ability to withstand shear
stresses imposed by surface water flow field
currents at the cap/water interface. One of
the most critical design characteristics of
AquaBlok® is that, given its high degree of
cohesiveness related to its material
composition, it claims to have a higher
resistance to shear energy compared to
traditional capping materials (e.g., sand).
Critical Measurements
o Sedflume coring and analysis;
o Sediment coring and analysis of COCs;
o Bathymetry and sub-bottom profiling; and
o Side-scan sonar surveying
Non-Critical Measurements
o Sediment profile imaging (SPI);
o Gas flux analysis; and
o Sediment coring and analysis of physical
parameters
• Objective #2 - Demonstrate the ability of an
AquaBlok® cap to control groundwater
seepage.
Tidal forces, regional pumping, or other
hydrogeologic phenomena in surface water
bodies have the potential to impose
significant vertical groundwater gradients
into or out of bottom sediments.
Measurements collected historically within
several areas of the Anacostia River near or
in the demonstration area indicate low but
quantifiable flow velocities both into and out
of the bottom sediments. One of the
primary advantages of AquaBlok® is that it
claims to significantly reduce permeability,
which should be reflected as a reduction in
groundwater seepage flows relative to
seepage in sand-capped sediments and
uncapped control areas.
Critical Measurements
o Sediment coring and analysis of hydraulic
conductivity; and
o Seepage meter testing
Non-critical Measurements
o None
• Objective #3 - Demonstrate the influence of
an AquaBlok® cap on benthic flora and
fauna.
A key concern in applying AquaBlok® as an
innovative sediment capping alternative is
the long-term effect of this material on
habitat for faunal (benthic) communities,
and also on potential habitat for floral
communities (which would depend on site-
specific water levels and suspended
sediment loads as they relate to a favorable
setting for emergent and/or submergent
vegetation).
According to the material vendor, standard
(i.e., non-amended) AquaBlok® material is
inherently low in organic content, and is not
generally designed specifically to support
significant biological growth. However, the
grain size of AquaBlok® is similar to
sediments generally found at most
contaminated sediment sites. In addition,
the AquaBlok® cap constructed during the
SITE demonstration was covered by a sand
layer that would likely support some level of
biological growth and allow for a
comparison between benthic impacts of the
sand-covered AquaBlok® cap and the sand-
only cap (e.g., floral and benthic infaunal
species impacts such as diversity and
richness).
Possible mechanisms by which the basal
AquaBlok® layer could potentially affect the
overlying sand material as benthic habitat
include:
1. The AquaBlok® material could
conceivably become entrained into or
29
-------
mixed with the sand covering layer,
thereby altering the surface
characteristics of the sand covering the
AquaBlok®. Thus, the sand material
covering the AquaBlok® cap could
behave differently than the sand-only
cap due to the entrainment of some of
the AquaBlok® material.
2. The AquaBlok® could create a physical
and/or relatively organically
impoverished barrier to deep burrowing
organisms (e.g., organisms that burrow
deeper than typical bioturbation depths
of approximately 20 cm), thereby
possibly affecting the species
composition and abundance of such
organisms and the diversity of
organisms supported in the cap
material.
3. Given its more similar grain size relative
to native Anacostia River sediments in
comparison to sand, AquaBlok® could
actually be a more preferential habitat
for benthos.
Critical Measurements
o None
Non-Critical Measurements
o Benthic grab sampling and descriptive and
statistical benthic assays; and
o Benthic assessment through SPI
These measurement endpoints associated with
the primary SITE demonstration objectives are
summarized in Table 3-2.
3.2.2 Field Activities
As indicated above, field activities were
implemented one month following completion of
the cap construction, and then at six months, 18
months, and 30 months following cap
construction. Accordingly, the first post-capping
field event occurred in the spring of 2004 and the
second in the fall of 2004. The remaining events
occurred annually thereafter, in the fall of 2005
and 2006.
During each of the post-cap construction field
activities, a robust set of field measurement tools
were utilized to gather information related to the
primary objectives of the AquaBlok®
demonstration project, as follows:
One-Month Post-Capping Field Event (Spring
2004)
• Bathymetry and sub-bottom profiling;
• Side-scan sonar surveying;
• SPI; and
• Seepage meter testing
Six-Month Post-Capping Field Event (Fall
2004)
• Bathymetry and sub-bottom profiling;
• SPI;
• Seepage meter testing;
• Sedflume coring and analysis; and
• Sediment coring and analysis of COCs and
physical parameters
18-Month Post-Capping Field Event (Fall 2005)
• Bathymetry and sub-bottom profiling;
• Side-scan sonar surveying;
• SPI;
• Seepage meter testing;
• Sediment coring and analysis of COCs,
physical parameters, and hydraulic
conductivity; and
• Gas flux analysis
30-Month Post-Capping Field Event (Fall 2006)
• Bathymetry and sub-bottom profiling;
• Side-scan sonar surveying;
• SPI;
• Seepage meter testing;
• Sedflume coring and analysis;
• Sediment coring and analysis of COCs,
physical parameters, and hydraulic
conductivity;
• Gas flux analysis; and
• Benthic grab sampling for descriptive and
statistical benthic assays
30
-------
Table 3-2. Critical and Non-Critical SITE Demonstration Measurements
Demonstration Objective
Objective #1
Demonstrate the physical stability of
an AquaBlok®cap
Objective #2
Demonstrate the ability of an
AquaBlok® cap to control
groundwater seepage
Objective #3
Demonstrate the influence of an
AquaBlok® cap on benthic flora and
fauna
Measurement
Sedflume analysis
Sediment coring and analysis of
COCs
Bathymetry and sub-bottom profiling
Side-scan sonar surveying
SPI
Gas flux analysis
Sediment coring and analysis of
physical parameters
Sediment coring and analysis of
hydraulic conductivity
Seepage meter testing
Benthic grab sampling and
descriptive and statistical assays
SPI
Critical or Non-Critical
Critical
Non-critical
Critical
Non-critical
This matrix of field sampling and monitoring
components for the various post-capping field
events is summarized in Table 3-3. This table
also provides a specific summary of the dates
during which the various tools were implemented
in the field.
3.2.3 Field Measurement Tools
The following subsections describe the general
methods and procedures typically followed to
employ the various field investigation and
monitoring tools that were used during the
AquaBlok® SITE demonstration project. For
simplicity, the field investigation and monitoring
tools are listed in the same order they appear in
Section 3.2.1.
3.2.3.1 Sedflume Coring and Analysis.
Sediment erosion rates typically depend on
sediment bulk density, mean grain size, grain
size distribution, organic content, and relative
cohesiveness. Sediment erosion, however,
cannot be accurately predicted through
knowledge of such sediment parameters alone,
and the relative influences of these parameters
tend to vary depending on the nature of any
substrata involved. Sedflume technology can be
used to determine how the sediment erosion
potential based on these sediment parameters
varies spatially across a study area.
To employ Sedflume technology, sediment cores
must be collected to obtain intact sediment for
testing. Cores in shallow water are typically
collected by manual direct-push techniques, while
cores in deeper water are typically collected
using a vibratory coring unit or comparable
mechanical coring method. Coring is most
commonly completed from a stable boat platform.
Cores collected for Sedflume analysis are
typically rectangular box cores. A rectangular,
transparent, box-shaped coring sleeve with a
nose cone is manually lowered to the sediment
bed by a pole (or by the mechanical coring unit).
Manual or mechanical pressure is applied to the
top of the sleeve and the nose cone. Based on
the combined weight of the coring sleeve and the
applied pressure, the sleeve penetrates the
sediment bed. Upon penetration of the core liner
into the sediment bed, flaps on the nose cone
open upward and allow sediment to enter the
core tube without disturbing the sediment strata.
The coring sleeve then is pushed as far as
possible into the sediment bed or until a suitable
design depth is achieved. The distance of
penetration will vary due to the characteristics of
the sediment (i.e., greater penetration depth will
-------
Table 3-3. SITE Demonstration Field Program Details
Demonstration
Objective
Objective #1
Demonstrate the
physical stability of an
AquaBlok®cap
Objective #2
Demonstrate the ability
of an AquaBlok® cap to
control groundwater
seepage
Objective #3
Demonstrate the
influence of an
AquaBlok® cap on
benthic flora and fauna
Measurement
Sedflume analysis
Sediment coring and analysis of
COCs
Bathymetry and sub-bottom
profiling
Side-scan sonar surveying
SPI
Gas flux analysis
Sediment coring and analysis of
physical parameters
Sediment coring and analysis of
hydraulic conductivity
Seepage meter testing
Benthic grab sampling and
descriptive and statistical assays
SPI
Field Event
Month 6
Month 30
Month 6
Month 18
Month 30
Month 1
Month 6
Month 18
Month 30
Month 1
Month 18
Month 30
Month 1
Month 6
Month 18
Month 30
Month 18
Month 30
Month 6
Month 18
Month 30
Month 18
Month 30
Month 1
Month 6
Month 18
Month 30
Month 30
Month 1
Month 6
Month 18
Month 30
Date(s)
9/17/04-9/23/04
10/17/06-10/22/06
9/20/04-9/25/04
9/27/05-9/28/05
10/17/06-10/19/06
5/12/04
9/14/04-9/15/04
9/15/05
9/19/06
5/11/04
9/14/05
9/20/06
5/13/04-5/14/04
9/16/04
9/16/05
9/20/06-9/21/06
8/25/05-9/26/05
8/14/06-9/13/06
9/20/04-9/25/04
9/27/05-9/28/05
10/17/06-10/19/06
9/27/05-9/28/05
10/17/06-10/19/06
5/17/04-5/22/04
9/27/04-10/7/204
9/19/05-9/23/05
9/25/06-9/30/06
10/17/06-10/19/06
5/13/04-5/14/04
9/16/04
9/16/05
9/20/06-9/21/06
tend to occur in a softer sediment than in a more
consolidated sediment). When the core sleeve is
lifted from the sediment bed, the nose cone flaps
close to retain the sediment core. The coring
sleeve is retrieved, the nose cone is removed, a
plug is inserted into the bottom of the sleeve to
seal the core and later to act as a piston head,
and the core is capped.
A detailed description of Sedflume and its
application is provided in McNeil and Lick (1996).
A Sedflume is essentially a straight flume (see
Figure 3-10) with a test section containing an
open bottom through which the rectangular cross-
section coring sleeve containing sediment is
inserted. The main components of the flume are
the coring sleeve, a test section where the coring
sleeve is advanced, a water storage tank, a pump
to force water through the system, an inlet for the
introduction of uniform, fully-developed, turbulent
flow, and a flow exit section. The Sedflume
system, as with the coring sleeve, is generally
constructed of a transparent material so that
sediment-water interactions can be directly
observed. Water is pumped from the water
storage tank, through a pipe, and then through a
flow converter into the rectangular duct shown on
Figure 3-10.
32
-------
TOP VIEW
PUMP
SIDE VIEW
PUMP
Figure 3-10. Sedflume Schematic (from McNeil and Lick, 1996)
This duct is generally 2 cm in height, 10 cm in
width, and 120 cm in length, and contains the test
section, which receives the box shaped core
sleeve with test sediment. The flow converter
changes the flow field shape from the water
holding tank from a circular cross-section to the
rectangular duct shape while maintaining a
constant cross-sectional area. A three-way valve
regulates the flow so that an appropriate
component of the flow enters the duct while the
remainder returns to the tank. Also, a small valve
in the duct immediately downstream of the test
section can be opened at higher flow rates to
maintain the pressure in the duct and over the
test section at atmospheric conditions.
At the start of each test, the coring sleeve and the
sediment it contains are inserted into the bottom
of the test section. An operator moves the
sediment upward using the piston plate placed
inside the coring sleeve during core retrieval that
is subsequently connected to a hydraulic jack.
The jack is driven by hydraulic pressure that is
regulated with a switch and valve system. By this
means, the test sediments can be raised and
made to enter the test section. The speed of the
jack movement can be controlled at a variable
rate in measurable increments generally as small
as 0.5 millimeters (mm).
Water is forced through the duct and the test
section and over the surface of the test
sediments. The shear produced by this flow
causes the sediments to erode. As the
sediments extruded from the core sleeve erode,
the core is continually moved upward by the
operator so that the sediment-water interface
remains level with the bottom of the test and inlet
sections (a few millimeters of the core does
protrude above the section floor). The erosion
rate is recorded as the degree of upward
movement of the sediments in the coring tube
over time.
Measuring the erosion rate of the sediments as a
function of shear stress and depth is generally
made by running the flume at a specific flow rate
corresponding to a particular shear stress.
Erosion rates are obtained by measuring the
remaining core length at different time intervals
recorded with a stopwatch and dividing the
difference in successive measurements by the
33
-------
time interval between the measurements. Shear
stress is then calculated using the measured
erosion rates and corresponding flume hydraulic
flow rates, as reported in McNeil and Lick (1996).
To measure erosion rates at several different
shear stresses using only one core, an iterative
procedure is used. Starting at a relatively low
shear stress, the flume is run sequentially at
higher shear stresses with each succeeding
shear stress generally being twice the previous.
Generally, about three shear stress intervals are
run sequentially. Each shear stress interval is
run generally until at least 2 to 3 mm but no more
than 2 cm of test sediment are eroded from the
test core. The flow then is increased to the next
shear stress interval, and so on until the highest
shear stress is run. Thus, flows are varied
incrementally, but held statically at each
graduated interval to observe and measure
sediment shear. After a particular interval is
complete, the piston elevates the core so that the
surface is in direct contact with the flume once
again. The intervals are repeated until all of the
sediment has eroded from the test core. If after
three cycles at a particular shear stress interval
an erosion rate of less than 10"4 cm/s is
calculated, that particular stress value is not
considered relevant to the test. Alternatively, if
after many cycles the calculated erosion rates
decrease significantly, higher shear stress
intervals will be introduced.
A critical shear stress can be quantitatively
defined as the shear stress at which a very small
but accurately measurable rate of erosion occurs.
As indicated above, this rate of erosion is
generally chosen to be 10"4 cm/s, which typically
represents 1 mm of erosion in approximately 15
minutes. It would be difficult to measure all
critical shear stresses at exactly 10"4 cm/s, so
erosion rates are generally measured above and
below 10"4 cm/s at shear stresses which differ by
a factor of two. The critical shear stress then is
linearly interpolated to an erosion rate of 10"4
cm/s. Critical shear stress is then a function of
the erosion rate measured as a function of depth.
The following equation (Gailani et al., 2001)
describes the erosion rate £ (cm/s) as a function
of the shear stress i (Newtons per square meter
[N/m2]) and the bulk density p (gram per cubic
centimeter [g/cm3]) and where A, n, and m are
constants related to bulk sediment properties:
E = AT"pm (3-1)
3.2.3.2 Sediment Coring and Analysis of
Contaminants of Concern. To determine
physical characteristics of sediments, as well as
levels of contaminants present in those
sediments, sediment coring is a frequently
utilized sediment investigation tool. Typically, a
pontoon boat or comparable vessel is used as
the sampling platform during sediment coring.
The techniques used for sediment coring include
vibratory coring, piston coring, or manual direct-
push coring.
An accurate global positioning system (GPS) on
the coring vessel is used to define spatial
coordinates for each core sampling location, or
spatial coordinates are determined prior to coring
and uploaded to the GPS system to accurately
locate these positions. Once on station, the
coring vessel is held in position using some
positioning device (e.g., anchors, spuds, or tie-
lines) and the manual or mechanical coring
device is lowered to the sediment surface and
pushed into the sediment to capture a vertical
sediment core. In sediment coring work, some
type of circular, clear core liner (e.g., butyrate)
and a cutter head are typically used inside the
coring device. The cutter head facilitates core
penetration and minimizes sediment loss during
core retrieval, and the core liner is used to
contain the intact core and can later be cut to
produce vertical interval samples for laboratory
analysis.
When the sediment cores are brought to the
surface, the sediment is contained within the
clear sediment core liner, which is typically
capped and stored vertically until the core is
processed in an appropriate fashion either
onboard the coring vessel or at a landside
processing location. Sediment can be visually
assessed in the core liner, or extruded from the
core liner, inspected, and various sampling
intervals transferred to appropriate sample
container(s) for laboratory analysis using
appropriate analytical methods.
34
-------
3.2.3.3 Bathymetry and Sub-Bottom
Profiling. Bathymetry is a method of collecting
accurate water depth information from across a
study area to understand sediment surface
topography and sediment slope. Given that the
surface of a water body can be considered a
static horizontal feature, water depth can be
considered a surrogate for sediment bed
elevation. In addition, bathymetry collected over
successive monitoring episodes can be
compared and provide an understanding of the
net change in sediment topography over time.
Bathymetric data are commonly collected by
following a series of parallel survey lines in a
survey vessel equipped with a survey-grade (i.e.,
high-precision) depth sounding instrument.
Depth sounding instruments function on the
principle of sound wave propagation by sending a
sound pulse (typically at an inaudible frequency)
to the sediment surface and receiving the return
signal from this pulse after reflection. The time
between signal generation and signal return is
geometrically proportional to water depth (after
correction for attenuation) and is recorded either
digitally or on a paper scroll by the depth sounder
device.
Sub-bottom profiling is a method of determining
the specific thickness of multiple layers of
subaqueous material. Acoustic sub-bottom
profiling of sediments, like bathymetry, makes
use of reflected sound waves from different
subsurface sediment layers (see Figure 3-11).
Sediment layers that exhibit different properties of
elasticity and density can sometimes be
distinguished as distinct layers within an acoustic
signal profile.
Sub-bottom profiling is frequently conducted
using a high-resolution subsurface profiler
capable of full-spectrum frequency modulation
(FM), also known as a "Chirp" profiler. As with
bathymetry, sub-bottom profiling data are
generally collected from a survey vessel along a
series of parallel survey lines. The principle of
sub-bottom profiling is similar to bathymetry, in
that an acoustic signal is generated and returned
to the instrument, allowing for a calculation of
depth based on travel time. However, the "Chirp"
profiler emits a signal in a frequency band rather
than a single frequency. The variable frequency
emission allows subsurface penetration and
resolution of different reflective layers based on
sediment lithology and the varying return times of
the varying frequencies and varying degrees of
penetration. Subsurface reflectors (indicating
different sediment depositional layers) can then
be digitized to produce maps of sediment
thickness with distinct lithologic layers.
3.2.3.4 Side-Scan Sonar. Side-scan sonar
works by transmitting sound waves to
subaqueous sediments at an angle. The sound
waves are emitted by a submerged device towed
by a survey vessel called a "towfish", which can
be positioned nearer the sediment surface to
minimize signal attenuation in the water.
Acoustic signals bounce off the sediment surface
and are then detected as a return by the side-
scan sonar instrument. The strength of the return
echo is continuously recorded and varies on the
basis of sediment surface texture, regularity, and
other parameters, which creates a virtual picture
or map of the sediment surface. Objects or
features that protrude from the bottom will tend to
yield a stronger signal, creating a relatively dark
image. The output of a side-scan sonar survey is
typically a plan view map with an appearance
analogous to an aerial photograph. Side-scan
sonar data are generally collected along a
parallel series of survey lines. However, given
the generally wider range of the acoustic signal
emitted from a side-scan sonar "towfish" relative
to bathymetry and sub-bottom profiling, the
survey line spacing for side-scan sonar can
generally be less dense.
Like bathymetry and sub-bottom profiling, side-
scan sonar data collected over successive
monitoring episodes can be compared and
provide an understanding of the net change in a
sediment surface over time.
3.2.3.5 Sediment Profile Imaging. A
sediment profiling camera can be used to
visually inspect sediment for stability, uniformity,
layering, and other important characteristics. SPI
involves the deployment of a highly specialized
camera from a vessel and the penetration of the
camera into the subaqueous sediment. The
sediment profile camera essentially works like an
35
-------
Diagram of acoustic subbottom profiling
Source Reciever
Water surface
-.-
/ /w
j \j Sea bed
Reflector 1
LEGEND
Reflector 2
Multiple
First return
Sound produced at the source reflects off of the seabed surface, reflector 1,
and reflector 2, which are areas of rapid density change.The receiver catches
the reflected sound waves. A multipl e sound wave is received at the same time
as the reflector two sound wave, obscuring reflector 2 on the same seismic record.
Figure 3-11. Principles of Acoustic Sub-Bottom Profiling
DURING
DEPLOYMENT
1-
[
ON THE
SEDIMENT
SURFACE
CAMERA
I
*l{
fflMDOW w ^\
-, MIRROR |-
-I
'DOWN' POSITION
TRANSECTING THE
SEDIMENT-WATER
INTERFACE
/
L
I
Figure 3-12. Schematic of Sediment Profiling Camera
36
-------
inverted periscope. A camera is mounted
horizontally on top of a wedge-shaped, knife-
sharp prism. The prism has a clear, optically
transparent faceplate at the front with a mirror
placed at a 45° angle at the back. The camera
lens looks down at the mirror that reflects the
image from the faceplate. The prism penetrates
the subaqueous sediment to capture a real-time
image of a relatively undisturbed profile of the
sediment.
The prism has an internal strobe mounted inside
at the back of the wedge to provide illumination
for the image. The prism chamber is filled with
distilled water so the camera maintains an
optically clear exposure path. The entire wedge
assembly is mounted on a moveable carriage
within a stainless steel or aluminum frame. The
frame is lowered by a guide line to the seafloor
mechanically from a surface vessel, and the
tension on the wire keeps the prism in its "up"
position. When the camera frame comes to rest
on the sediment surface, the guide line goes
slack (see Figure 3-12), and the camera prism
descends into the sediment at a slow, controlled
rate by the dampening action of a hydraulic
piston to minimize disturbance at the sediment-
water interface. On its descent, the prism trips a
trigger that activates a time-delay circuit to allow
the camera to reach maximum penetration into
the seafloor before the picture is taken.
Alternatively, the camera can be operated
manually to take a series of images as the prism
penetrates the sediment. The resulting images
give the viewer the same perspective as looking
through the side of an aquarium half-filled with
sediment. The strobe can generally recharge in a
matter of seconds.
Using SPI, the thickness of different sediment
deposits can be determined by measuring the
linear distance between the material types (e.g.,
natural deposits and a cap layer) based on the
point of contact between the two layers and a
textural change in sediment composition or a
change in color that should be clearly visible.
Also, sediment grain size can be visually
estimated from the SPI photographs by
comparing to a grain size reference chart at the
same scale. Such a reference chart is generally
prepared by photographing a series of sediments
of known and varying size classes through the
SPI camera prior to a true sediment survey.
Similarly, sediment color can be determined
through a comparison to a detailed reference
color chart, generally available from the
manufacturer of the camera used.
The SPI prism penetration depth is determined by
measuring the longest and shortest linear
distance between the sediment-water interface
and the bottom of the film frame. Software can
be used to automatically average these maximum
and minimum values to determine the average
penetration depth. If needed, weights can be
added to the SPI camera frame to enhance
penetration.
3.2.3.6 Gas Flux Analysis. Submerged gas
flux chambers are designed so that biogenic gas
samples from within the chambers can be
sampled while the chambers are underwater.
Gas flux chambers generally consist of a
modified steel 55-gallon drum with approximately
one-third of the bottom cut away. The lid of the
drum is modified to consist of a detachable steel
"dome-like" top. Each chamber lid is equipped
with a stainless steel, valved female, quick-
connect fitting that is used to obtain gas samples.
The side of the chamber can be outfitted with a
stainless steel "T" that can be left open to relieve
overpressure during gas production and capture
within the chamber.
The cut bottom edge of the 55-gallon drum is
driven into the sediment by a dive team. The
chambers can be pushed through a capping layer
and into the native sediment formation or keyed
into a capping layer. Anchoring points are
usually welded to each side of the chamber and
are used to secure cinder blocks or other objects
for anchoring the chambers on the sediment
surface. Figure 3-13 shows a schematic diagram
of a submerged gas flux chamber that would
commonly be deployed for sediment investigation.
Gas samples are obtained from a gas flux
chamber using a dive team. Samples are usually
collected using a gastight syringe equipped with a
valved stainless steel male quick-connect fitting
on an umbilical cord. The syringe is attached to
the valved female, quick-connect fitting on the top
of the "dome" lid of the chamber via the umbilical.
37
-------
Cinder Blocks
Used toAnchor
3/4 Female Quick-Comect®
With Male Cap
Submerged Dffusion
Chamberwith
Removable Top
Partial 55-gal Drum
"T" With 1-in. to .75-in. Reducer
Plugged on One Side and
Female Quick Connect® on Other
With Male Cap
Side
Viewl
Figure 3-13. Schematic of Typical Submerged Gas Flux Chamber
38
-------
Gas that has collected in the chamber through
ebullition is pulled into the syringe and brought to
the surface through the syringe where it is then
expelled into an appropriate sample container
(generally a Tedlar bag). This procedure is
repeated until all gas has been removed from the
gas flux chamber and water is seen to fill the
syringe. Samples can be sent to a laboratory for
analysis of vapor-phase characteristics or
constituent concentrations. Flux is determined
empirically using equations that incorporate the
gas volume extracted from the chamber, the
volume of the chamber, and the deployment
duration. Such equations are provided and
described in more detail in Section 3.3.1.2.2.
Deployment duration for a submerged gas flux
chamber is generally one month.
3.2.3.7 Sediment Coring and Analysis of
Physical Parameters. The general methods
and approach used during sediment coring are
described in Section 3.2.3.2. In addition to
evaluating COCs, it can be important to
understand some common physical properties of
sediment that can influence contaminant
concentrations or other sediment characteristics.
Some of the most frequently evaluated sediment
physical parameters are total organic carbon
(TOC), particle size distribution (PSD) (also
known as grain size distribution), and moisture
content. TOC is important because sediment
contaminants are most commonly bound in the
sediment organic carbon fraction rather than
directly on inorganic matrix material (e.g., sand
particles). PSD is a measure that can directly
correlate varying sediment layers (i.e., a fine-
grained native sediment versus a coarser-grained
cap layer) to other critical measures (e.g.,
contaminant concentrations). Moisture content is
important in understanding other bulk sediment
properties (e.g., shear strength).
3.2.3.8 Sediment Coring and Analysis of
Hydraulic Conductivity. The general
methods and approach used during sediment
coring are described in Section 3.2.3.2. In
addition to evaluating COCs and physical
parameters, it can be important to understand
common geotechnical properties, namely
hydraulic conductivity (K). This parameter, which
is sometimes referred to as the coefficient of
permeability, is a proportionality that describes
the rate at which water is able to move through a
permeable medium. Obviously, when evaluating
a material that is designed to impart a high
degree of impermeability and resistance to flow, it
is very important to measure K. Samples of
sediment (and other materials) for the evaluation
of K are generally collected in clear core liners as
previously described and submitted to an
appropriate testing facility as intact core sections
containing the sediment interval of interest.
3.2.3.9 Seepage Meter Testing. Ground-
water seepage meters provide continuous
measurement of aqueous flux at high resolution
over an extended period of time. These devices
are frequently modeled after the ultrasonic
seepage meters and funnel collection systems
developed at the Cornell Cooperative Extension
(CCE) marine laboratory in Cedar Beach, New
York.
Using seepage meters, vertical advective flux is
captured by a steel collection chamber with a
square cross-section that is inserted into the
sediment surface, generally by a dive team (see
Figure 3-14). The captured aqueous discharge is
directed via a length of tubing through an
ultrasonic flow tube adapted for use in submarine
environments. The flow tube is angled to allow
trapped gases to escape. The ultrasonic device
houses two piezoelectric transducers at either
end of the flow tube that continually emit
ultrasonic bursts (-400 bursts per second at an
appropriate frequency) from one end of the meter
to the other. The piezoelectric transducers
continuously measure the travel times of the
ultrasonic waves as water enters the flow tube
and passes through the ultrasonic beam path
(see Figure 3-15). The ultrasonic signal that
travels with the direction of flow has a shorter
travel time than the signal traveling against the
direction of flow. The directional perturbation in
travel time is directly proportional to the velocity
of flow in the tube, and both forward and reverse
fluid flows can be measured in real time. The
ultrasonic measurement device is generally
connected to a battery-powered data logger that
records both incremental and cumulative
discharge simultaneously. For field deployment,
the data logger and a back-up battery are usually
39
-------
t t
stainless steel funnel
Figure 3-14. Schematic of Ultrasonic Seepage Meter (not to scale)
flow
inlet
flow
outlet
piezoelectric transducers
Figure 3-15. Conceptual Cross-Section of Ultrasonic Seepage Meter Flow Tube
(from Paulsen et al., 2001)
housed in a floating buoy that is anchored to the
sediment. The battery and back-up battery are
usually selected to provide at least five days of
power to minimize the possibility of equipment
failure during data collection.
The ultrasonic seepage meter records specific
discharge (q) in cm/s. The directional specific
discharge through the capture area at the
sediment-water interface is calculated as follows:
Where
Q
A,
Af
q =
Q_Y^
A,lAf
Af
(3-2)
= discharge (cm3/s);
= area of flow tube (cm2); and
= area of the funnel (cm2)
40
-------
The discharge (Q) is calculated from the flow
velocity through the discharge tube, multiplied by
the inside cross-sectional area of the discharge
tube. Flow velocity (v), in turn, is determined
from the ultrasonic pulse velocity. Travel time for
the upstream propagation of sound waves
against the flow direction is prolonged relative to
that for downstream propagation. The upstream
and downstream travel times are given,
respectively, by:
Tup = L/(V-v) (3-3)
and
Tdown = L/(V + v) (3-4)
Where V = sound velocity (cm/s);
T = time (s); and
L = length of flow tube (cm)
Combining these two equations to solve for v
(cm/s) yields:
v=-
T
up down
_/"T _T
V up down >
(3-5)
Travel times are generally resolved to
nanoseconds to impart suitable sensitivity to the
velocity measurements.
Generally, a seepage meter deployment at one
location lasts from two days (48 hours) to four
days (96 hours). The four-day time period
provides sufficient time for the meters to
equilibrate and capture tidal influences over
multiple diurnal cycles. However, two days may
often suffice. Data are retrieved daily, reduced,
and analyzed on site to assess meter
performance and data adequacy over the
deployment periods.
3.2.3.10 Benthic Grab Sampling and
Descriptive and Statistical Benthic
Assays. A key concern during sediment
capping is the long-term effect on the sediment
bottom as habitat for faunal (benthic)
communities, and also as potential habitat for
floral communities (which will depend on
prevailing water levels and other site-specific
characteristics and their ability to support
emergent and/or submergent vegetation). Faunal
and floral developments can be evaluated via
benthic assays designed to define the type and
density of benthic fauna and flora over time.
To obtain samples for the evaluation of benthic
faunal (and potentially floral) communities,
sediment must be collected from the uppermost
interval where biogenic activity is typically
concentrated (i.e., generally the upper 10 cm of
sediment). Samples for benthic sediment
infaunal analyses are generally collected with a
bottom grab sampler, such as a Van Veen or
Ponar sampler, which are both spring-activated
jaws designed to scoop the upper interval of
sediment on contact. Collection of appropriately
undisturbed sediment samples is critical and is
achieved by careful attention to established
deployment and recovery procedures, including
controlled sampler fall rate and sediment
penetration and slow sampler recovery.
After surface sediment sample collection, the
grab is placed over a bucket, the jaws opened,
and the sample emptied into the bucket. Filtered
river water is used to gently wash the sediment
through a series of fine mesh sieves to remove
inorganic sediment and debris and leave behind
organisms and organic material. Generally, two
sieves are used, with the first having a fine mesh
size and the second having a very fine mesh
size. The material remaining on the sieve(s) is
then placed into a sample container and a fixing
agent is added to preserve the sample. Samples
are submitted to a benthic laboratory and infaunal
organisms are sorted and taxonomic
identifications performed by qualified biologists.
After this laboratory sorting and identification
step, descriptive and statistical ecological metrics
can be applied to further describe the benthic
assemblages in the sediment. Calculating
descriptive ecological measures provides insight
into the overall structure of the community.
These measures typically include simple
parameters such as total abundance and
numbers of species per sample, and may also
include measures of diversity (e.g., the Shannon
Diversity Index [H] or Pielou's Evenness Index
[J]). Statistical tools can then be applied to test
for differences in these measures
41
-------
among/between samples or sites. Statistical
tests may include simple f-tests or analysis of
variance (ANOVA) followed by an appropriate a
posteriori test (e.g., Tukey's Honestly Significant
Different [HSD] test) to identify specific
differences among/between samples or sites.
Correlation analyses may also be used to
evaluate the association between faunal
abundance and measured environmental variable
(e.g., grain size, TOC). Multivariate pattern
analysis may be a final step to examine patterns
among complex combinations of multivariate
data. Multivariate analyses consider the
identities of the species in each sample and
determine degree of similarity in species
composition among samples. Because of this,
these analyses are typically more powerful in
detecting differences among samples compared
to simple or correlation statistical tests which
themselves involve the use of species identities.
Examples of multivariate statistical tests include
the Bray-Curtis or chord-normalized expected
species shared (CNESS) similarity algorithms.
Ordination techniques (e.g., principal component
analysis, non-metric multidimensional scaling, or
multiple discriminate analysis) may also be used
to estimate the set of discriminate functions that
best explain the separation of sites or samples
resulting from the cluster analysis in terms of
selected environmental variables.
3.2.3.11 Benthic Assessment through
Sediment Profile Imaging. The general
methods and approach used during SPI are
described in Section 3.2.3.5. SPI can also be
used for the evaluation of benthic community
parameters and benthic recovery. Specifically,
the following types of information related to
benthic fauna (and flora) can be gathered using
SPI:
• The biogenic disturbance of fine-grained,
cohesive sediments may cause intact
clumps of sediment (mud clasts) to be
scattered on the sediment surface. These
may appear at the sediment-water interface
in SPI images. The abundance,
distribution, oxidation state, and angularity
of mud clasts may be used to infer recent
patterns of disturbance to the sediment
surface in the area.
• The depth of the apparent reduction-
oxidation (redox) potential discontinuity
(RPD) in the sediment column is an
important estimator of benthic habitat
quality. This depth is related to the supply
rate of molecular oxygen by diffusion into
sediment and the subsequent consumption
of that oxygen within the sediment. The
term apparent is used in describing this
parameter because no actual measurement
is made of the redox potential. An
assumption is made that, given the
complexities of iron and sulfate redox
chemistry, reddish-brown sediment color
tones are indications that the sediments are
oxic, or at least are not intensely reducing
(Diaz and Schaffner, 1988). The exact
location of the RPD depth can only be
determined accurately with microelectrodes.
The apparent mean RPD depth can be
used as an estimate of the depth of pore
water exchange, usually resulting from
bioturbation.
• High organic-loading levels in sediment
ultimately may cause methanogenesis to
occur. Methanogenesis results in the
development of methane bubbles in the
sediment column which are readily visible
as voids in SPI images because of their
irregular, circular shape and glassy texture.
If present, the number and total area
covered by all such voids can be measured.
• The successional stage of an infaunal
community is based on the idea that
organism-sediment interactions after a
disturbance follow a predictable sequence
of recovery (Rhoads and Germano, 1986).
This continuum of change in biological
communities after a disturbance has been
divided arbitrarily into three stages. Stage I
is the initial colonizing community (typically
small, densely-populated annelid
assemblages), Stage II is an intermediate
step usually comprised of more types of
organisms (commonly crustaceans and
perhaps some insect larvae) and is the
transitional stage to a mature community,
and Stage III is the mature, equilibrium
community (comprised of deep-dwelling,
head-down deposit feeders).
42
-------
• The organism-sediment index (osi), as
developed by Rhoads and Germane (1986),
is an integrative estimate of the general
ability of the benthic habitat to support
fauna. The osi is based on three
parameters that can be measured using
SPI, namely the mean apparent RPD depth,
presence of methane gas, and the infaunal
community successional stage, and an
indirect estimate of near-sediment DO
levels. Higher osi values are indicative of a
mature benthic community in relatively
undisturbed conditions. Lower osi values
reflect sediments that are azoic and have
high levels of methane and little or no DO.
3.2.4 AquaBlok® SITE Demonstration
Specific Approach and Methods
The specific approach and methods used during
the AquaBlok® demonstration are summarized
below. For simplicity, the summary follows the
same general structure as Section 3.2.2, which
briefly summarized the four field investigation
events implemented during the demonstration.
As appropriate, specific QA/QC procedures
followed during the execution of the field events
are also summarized.
3.2.4.1 One-Month Post-Capping Field
Event. The one-month post-capping field event
was conducted in May 2004. The specific
monitoring tools used during this event consisted
of bathymetric and sub-bottom profiling, side-
scan sonar surveying, SPI, and seepage meter
testing. Specific one-month post-capping field
event information for these individual monitoring
tools is described below.
3.2.4.1.1 One-Month Post-Capping
Field Event Bathymetry and Sub-Bottom
Profiling. On May 12, 2004, an initial integrated
bathymetric and sub-bottom profiling survey was
completed in the demonstration area. For
simplicity and to maximize efficiency relative to
other investigators' activities, all of the capping
cells (i.e., AquaBlok®, sand, apatite, and coke
breeze) and the uncapped reference cell were
surveyed simultaneously using these tools rather
than surveying only the cells of interest for the
SITE demonstration program (i.e., AquaBlok®,
sand, and control). The surveys were conducted
from a survey vessel owned by Ocean Surveys,
Inc. (OSI) of Old Saybrook, Connecticut and
operated by qualified oceanographic
geophysicists from OSI, under the direct
oversight of Battelle.
Prior to conducting the survey, a series of survey
lines were established in the demonstration area,
running parallel to the river bank in a generally
east-west orientation. To provide adequate
coverage of the study area, 29 total survey lines
were established at a nominal spacing of 10 ft. In
addition, prior to the survey, a tide board and
gauge were installed along the bulkhead of the
Washington Navy Yard property immediately
north of the demonstration area to provide data
on tidal level fluctuations in the river throughout
the course of the survey.
The bathymetric survey was conducted by towing
a dual-frequency, high-resolution depth sounder
beside the survey vessel. The sub-bottom
profiling was accomplished simultaneously using
a high-resolution, full-spectrum FM "chirp" profiler
similarly towed beside the survey vehicle. Data
generated using the depth sounder and "chirp"
profiler were continuously uploaded by an
onboard computer. Accurate positional control
was maintained by securing a differential global
positioning system (dGPS) antenna to the cabin
of the survey vessel immediately above the
outboard equipment boom holding the depth
sounder and the "chirp" profiler.
QC procedures for the bathymetric survey
included calibrating the depth sounder against a
metal object placed at a controlled depth beneath
the transducer. Sub-bottom profiling QC included
varying several equipment settings and the
overall signal transmission frequency to resolve
ideal survey parameters. To provide additional
data QC, several survey tie-lines were completed
running perpendicular to the shoreline and the
pre-determined 29 survey transects, ensuring the
overlap of several data points to verify positional
and data accuracy.
Additional details relating to the general principles
of bathymetric and sub-bottom surveying can be
found in Section 3.2.3.3, and details specifically
43
-------
related to the one-month post-capping
bathymetric and sub-bottom profiling survey are
included in Appendix B-1.
3.2.4.1.2 One-month Post-Capping
Field Event Side-Scan Sonar Surveying.
On May 11, 2004, an initial side-scan sonar
survey was completed in the demonstration area.
For simplicity and as with the one-month post-
capping bathymetry and sub-bottom profiling, all
of the capping cells (i.e., AquaBlok®, sand,
apatite, and coke breeze) and the uncapped
reference cell were surveyed simultaneously
using side-scan sonar rather than surveying only
the cells of interest for the SITE demonstration
program (i.e., AquaBlok®, sand, and control).
The survey was conducted from the same survey
vessel owned and operated by qualified
oceanographic geophysicists from OSI as
described in Section 3.2.4.1.1, again under the
direct supervision of Battelle.
The side-scan sonar survey was completed along
10 of the 29 survey lines established in the
demonstration area using a high-resolution side-
scan sonar surveying system. Data generated
using the side-scan sonar were continuously
uploaded by an onboard computer. Accurate
positional control was maintained by securing a
dGPS antenna to the cabin of the survey vessel
immediately above the outboard equipment boom
holding the side-scan sonar towfish.
QC during the side-scan sonar activity included
ensuring that all equipment was in working order
and attempting to maintain a controlled and
constant speed in the survey boat. In addition,
several survey tie-lines were completed running
perpendicular to the shoreline and the pre-
determined 29 survey transects, ensuring the
overlap of several data points to verify positional
and data accuracy.
Additional details relating to the general principles
of side-scan sonar surveying can be found in
Section 3.2.3.4, and details relating to the one-
month post-capping side-scan sonar survey are
included in Appendix B-1.
3.2.4.1.3 One-Month Post-Capping
Field Event Sediment Profile Imaging.
Between May 13 and 14, 2004, a thorough initial
SPI investigation was completed in the
demonstration area. For simplicity and to
maximize efficiencies relative to other
investigators' activities, SPI work was conducted
throughout the demonstration area (i.e., in all
capping cells and the control cell) rather than only
in the AquaBlok®, sand, and control cells.
During the SPI activities, both a high-resolution
still SPI camera and a video SPI camera were
used. Both were deployed from the deck of a
survey vessel owned and operated by OSI with
OSI personnel assisting in camera deployment
and retrieval. RJ Diaz & Daughters (RJ Diaz) of
Ware Neck, Virginia, provided the cameras and
was responsible for operation of the cameras and
related mechanical/electronic equipment.
Battelle provided direct oversight of the SPI
activities.
Prior to implementing the SPI work, a series of
monitoring stations were determined to provide
comprehensive information for each capping cell
and the control cell. A total of nine individual
locations were selected for each cell to be
targeted with the video SPI camera, and three
locations were selected to be targeted by the
high-resolution still SPI camera. At each location,
the survey vessel was piloted until stationed over
the monitoring location of interest using a dGPS
system and OSI's vessel navigation software,
and then the camera was deployed while the
vessel drifted at idle. The camera was deployed
by dropping it slowly while attached to a rope
connected to a winch. Once the frame of the
camera contacted the sediment surface, the rope
was manually moved up and down to further
facilitate penetration until refusal was
encountered.
The cameras (both high-resolution still and video)
were mounted to a steel frame ballasted with lead
weight to improve penetration and containing a
piston arm to dampen the camera's travel (see
Section 3.2.3.5 for a general description of SPI
camera construction and operation). The video
SPI survey was conducted first, and then the
camera housing was changed out for the high-
44
-------
resolution still camera survey. In addition to the
monitoring stations in the capping and control
cells, a single station was selected in the
Anacostia River main channel outside the study
area to provide additional reference information.
While using the video SPI camera, RJ Diaz
remotely initiated a digital recording device upon
contact of the camera with the sediment surface.
The digital recording device was then turned off
by RJ Diaz at the point of refusal. For the high-
resolution still SPI camera, RJ Diaz triggered
individual digital exposures during the camera's
penetration into the sediment, also remotely.
After completing the survey, all SPI images were
evaluated directly by RJ Diaz to determine
specific characteristics related to the physical and
ecological nature of the sediments. For standard
SPI images, the least disturbed image, usually
the last in the series for each cell, was analyzed
digitally using Adobe PhotoShop®. Videotape
recorded from the video SPI images was digitized
and a sequence of still frames was then extracted
with Final Cut Pro®. The still frame sequences so
extracted were then stitched together using
Adobe PhotoShop®. The steps in the computer
analysis of each image were standardized, and
all images were histogram equalized to increase
contrast. All processed sediment profile images,
both standard SPI and video SPI, were analyzed
visually for specific physical features of interest,
as follows:
Prism Penetration - This parameter provided a
geotechnical estimate of sediment compaction,
with the SPI camera prism acting essentially as a
dead weight penetrometer. Greater prism
penetration was generally associated with softer,
finer-grained sediment presumably with higher
water content. Penetration was measured as the
distance the sediment was observed to advance
over the camera faceplate.
Sediment Grain Size - The sediment type
observed in various intervals in each SPI image
was defined on the basis of its major modal grain
size category following the Wentworth
classification system. Relative grain size was
determined by comparing SPI images with a set
of reference images for which mean grain size
had been determined in the laboratory.
Sediment Layering - sediment layering was
assessed by evaluating both color and grain size.
Sediment color in various layers was
characterized by intensity relative to adjoining
layers by using qualitative descriptors (i.e., lighter
or darker). The identification of sediment layering
on the basis of grain size relied on modal grain
size following the Wentworth classification
system, as with the determination of sediment
grain size described above. Where sediment
layering was observed, the average thickness of
layers was described on the basis of both
measures (i.e., color and grain size).
Surface Features - Each SPI image was
evaluated for the presence of distinct surface
features, including a variety of purely physical
(such as bedforms or floe layers) and biogenic
physical features (such as biogenic mounds or
tubes). Surface features were compiled by type
and frequency of occurrence.
Subsurface Features - Each SPI image was
evaluated for the presence of distinct subsurface
features, including a variety of features that
provide evidence about physical processes (such
as burrows, water filled voids, gas voids, or
sediment layering). Subsurface features were
compiled by type and frequency of occurrence.
RJ Diaz also evaluated the processed sediment
images for characteristics representative of
ecological parameters, including the presence of
gas voids, organism burrows, specific organisms,
and osi (see Section 3.2.3.11).
QC procedures implemented during the SPI
program included recording detailed information
on location and frame exposure related to
individual camera drops, periodically checking the
color contrast of the cameras against color
standards, using as close to the same drop rate
and manual line-pulling approach on each
camera drop as possible, and approaching each
deployment location at as close to the same boat
speed as possible.
45
-------
Additional detail related to the one-month post-
capping SPI survey is included in Appendix C.
3.2.4.1.4 One-Month Post-Capping
Field Event Seepage Meter Testing.
Between May 17 and 22, 2004, an initial seepage
meter investigation was implemented in the
demonstration area. Because no other
investigators utilized seepage meter data, this
activity was implemented only for the AquaBlok®,
sand, and control cells.
In each cell, two ultrasonic seepage meters were
deployed and retrieved by divers from Matrix
Environmental and Geotechnical Services
(Matrix) of East Hanover, New Jersey. The
meters were constructed as described in Section
3.2.3.9, and consisted of an angled funnel with a
square cross-sectional area of 0.209 m2 attached
by a 44 cm length of Tygon tubing to the flow
tube. Meters were deployed by removing some
amount of surface sediment material
(approximately 2 in) and then gently pushing
them into the bottom sediments until forming an
appropriate seal, and were weighted down with
ballast. Each meter was then left in place for
approximately three to four days. Data were
continuously downloaded by floating data loggers
attached to each meter, and periodically
uploaded by Matrix (see Section 3.2.3.9 for a
general description of the principles of ultrasonic
seepage meter testing). The loggers were
capable of five hours of continuous operation, but
were connected to backup batteries with a 120
hour lifespan. Battelle provided field oversight
during the seepage meter testing event. One
meter in the control cell was relocated during its
deployment based on readings apparently heavily
impacted by gas ebullition (i.e., those data
demonstrated a highly erratic pattern consistent
with the capture of significant amounts of gas in
the flow tube).
During the seepage meter testing, surface water
temperature and pressure were measured
continuously at one meter location, and
groundwater elevation, temperature, and
conductivity were measured continuously in two
monitoring wells located upland on the north side
of the river.
QC during the seepage meter testing activity
included ensuring that all meters were properly
calibrated in the laboratory prior to deployment.
In addition, two seepage meters were deployed in
each cell during the demonstration to ensure data
usability. Surface water and groundwater
temperature data were collected to facilitate data
corrections for temperature effects, and
groundwater and surface water elevation data
were collected to facilitate resolution of diurnal
tidal cycle impacts on calculated seepage rates.
Additional detail relating to the one-month post-
capping seepage meter survey is included in
Appendix D-1.
3.2.4.2 Six-Month Post-Capping Field
Event. The six-month post-capping field event
was conducted between September and October
2004. The specific monitoring tools used during
this event consisted of bathymetric and sub-
bottom profiling, SPI, seepage meter testing,
Sedflume analysis, and sediment coring for the
analysis of COCs and physical parameters.
Specific six-month post-capping field event
information for these individual monitoring tools is
described below.
3.2.4.2.1 Six-Month Post-Capping
Field Event Bathymetry and Sub-bottom
Profiling. Between September 14 and 15,
2004, a second integrated bathymetric and sub-
bottom profiling survey was completed in the
demonstration area. The six-month post-capping
bathymetric and sub-bottom profiling survey was
completed by OSI under the direct supervision of
Battelle. All methods and procedures followed,
including QA/QC, were generally identical to the
one-month post-capping survey described in
Section 3.2.4.1.1. To ensure the comparability of
data between the six-month event and the one-
month event, the identical survey transects were
used during both events.
Additional detail relating to the general principles
of bathymetric and sub-bottom surveying can be
found in Section 3.2.3.3, and details specifically
related to the six-month post-capping bathymetric
and sub-bottom profiling survey are included in
Appendix B-2.
46
-------
3.2.4.2.2 Six-Month Post-Capping
Field Event Sediment Profile Imaging. On
September 16, 2004, a second thorough SPI
investigation was completed in the demonstration
area. The six-month post-capping SPI survey
was completed by RJ Diaz (with vessel support
provided by OSI) under direct supervision of
Battelle. All methods and procedures followed,
including QA/QC, were generally identical to the
one-month post-capping survey described in
Section 3.2.4.1.3. To facilitate direct comparison
between SPI data from the six-month post-
capping event and SPI data from the one-month
post-capping event, an effort was made to drop
the camera during the six-month event relatively
near the equivalent locations from the one-month
event while not being closer than approximately 5
to 8 ft of the previous locations. During the six-
month survey, two Anacostia River channel
locations were targeted for both high-resolution
and video SPI camera evaluation as opposed to
the one channel reference location evaluated
during the one-month survey.
Additional detail relating to the general principles
of SPI surveying can be found in Section 3.2.3.5,
and details specifically related to the six-month
post-capping SPI survey are included in
Appendix C.
3.2.4.2.3 Six-Month Post-Capping
Field Event Seepage Meter Testing.
Between September 27 and October 7, 2004, a
second seepage meter investigation was
implemented in the demonstration area.
Because other investigators utilized seepage
meters during the six-month post-capping event
and to maximize investigation efficiency, this
activity was implemented in all capping cells and
the control cell rather than only in the AquaBlok®,
sand, and control cells.
In the AquaBlok®, sand, and control cells, two
ultrasonic seepage meters each were deployed
and retrieved by divers from Matrix. All methods
and procedures followed, including QA/QC, were
generally identical to the one-month post-capping
seepage meter testing described in Section
3.2.4.1.4. Meters were deployed for
approximately four to nine days, but deployment
locations were not specifically near the
deployment locations from the one-month post-
capping evaluation. One meter in the control cell
and one meter in the AquaBlok® cell were
relocated during deployment. The control cell
meter was relocated based on readings
apparently heavily impacted by gas ebullition
(i.e., those data demonstrated a highly erratic
pattern consistent with the capture of significant
amounts of gas in the flow tube), while the
AquaBlok® meter was relocated because it was
determined that the meter had inadvertently been
deployed in the portion of the cell that was not
sufficiently capped during initial construction (see
Section 3.1.1 and Figure 3-4).
During the six-month post-capping seepage
meter testing, surface water temperature and
pressure were measured continuously at one
reference location established at the ECC dock,
and, as with the one-month post-capping event,
groundwater elevation, temperature, and
conductivity were measured continuously in two
monitoring wells located upland on the north side
of the river.
Additional detail relating to the six-month post-
capping seepage meter survey is included in
Appendix D-2.
3.2.4.2.4 Six-Month Post-Capping
Field Event Sedflume Coring and
Analysis. Between September 17 and 23,
2004, an initial Sedflume evaluation was
conducted in the demonstration area. The
evaluation was completed for the AquaBlok®,
sand, and control cells by Sea Engineering, Inc.
(SEI) of Santa Cruz, California, under the direct
oversight of Battelle.
During the Sedflume evaluation, 12 individual box
cores were collected from the demonstration
area, four each from the AquaBlok®, sand, and
control cells. One core was collected from each
of four quadrants (i.e., NW, NE, SE, and SW) in
each individual cell by drawing imaginary
bisecting lines both north-south and east-west
and so defining four unique cell areas. The cores
were collected as acrylic box cores measuring 10
cm by 15 cm by approximately 1 m in length.
The cores were collected manually and it was
ensured that the appropriate interval of interest
47
-------
(i.e., the full thickness of the capping material of
interest) was captured in each respective core.
At least 30 cm and no more than 100 cm of
vertical core material was collected at each core
location. A special valve at the top of the core
section ensured that the force of suction from the
native (or capped) material would not prevent the
extraction of a usable core section.
Cores were evaluated immediately in a mobile
Sedflume laboratory provided by SEI and
stationed at the GSA property north of the
investigation area. The mobile laboratory was an
approximately 24-ft enclosed box truck equipped
with a 325-gal water tank and a Sedflume
apparatus. The apparatus and operation of the
apparatus were consistent with the description
provided in Section 3.2.3.1. The Sedflume
apparatus used during the six-month post-
capping event specifically consisted of a 5-cm,
round water inlet pipe and a flow converter to
transform flow from this pipe into a rectangular
flow consistent with the 2 cm by 10 cm and 120
cm long testing duct.
To derive critical shear resistance information
from the Sedflume, it was operated in a manner
consistent with the description in Section 3.2.3.1.
For each core, a series of shear stresses were
applied to the core in the test section of the
Sedflume, and time and erosion were recorded
until at least 1 to 2 mm but not more than 2 cm
were eroded. Initial shear stresses were low and
were gradually increased until a maximum stress
was run for each particular test. Each
subsequent shear stress was generally twice the
previous.
To allow for correlation between shear stress and
important physical characteristics, interval
samples from the cores were periodically
analyzed for PSD and water content in the mobile
laboratory. These samples were collected from
the top of the core after each erosion cycle.
Water content was determined through bulk
density analysis according to American Society of
Testing and Materials (ASTM) method D-2216.
PSD was determined through laser dispersion.
Independent flow field measurements to
characterize the hydrologic profile of the
Anacostia River were not included as part of the
Sedflume approach. Such measurements would
likely have included the use of an acoustic
Doppler current profiler (ADCP) to calculate
specific in-river velocities and sediment transport
characteristics. The reason for not including the
ADCP is that AquaBlok® cap failure due to high
flow rates is unexpected due to the relatively
sluggish flow of the Anacostia River (see Section
3.1) and the Sedflume procedures themselves
included manipulating laboratory flows to levels
higher than and uncharacteristic of the Anacostia
River. Thus, ADCP measurements would have
provided little information regarding the potential
for cap failure that was not gleaned from the
Sedflume laboratory protocol.
QC procedures implemented during the Sedflume
program included recording detailed information
on location related to individual cores, ensuring
the working order and maintenance of all
Sedflume components and equipment, adhering
to appropriate laboratory methods and
procedures, and collecting multiple cores from
each cell for adequate data coverage. All
reusable materials and supplies were
decontaminated prior to reuse.
Additional detail relating to the six-month post-
capping Sedflume study is included in Appendix
E-1.
3.2.4.2.5 Six-Month Post-Capping
Field Event Sediment Coring and Analysis
of Contaminants of Concern and Physical
Parameters. Between September 20 and 25,
2004, an initial sediment coring investigation was
completed in the demonstration area. Multiple
sediment cores were collected using vibracoring
techniques from a coring vessel owned and
operated by Athena Technologies, Inc. (Athena)
of Columbia, South Carolina. Coring was
overseen by Battelle and sediment cores were
evaluated and processed by Battelle personnel.
During the six-month post-capping field event,
two individual cores were collected from each of
the four unique quadrants in the AquaBlok®,
sand, and control cells, for a total of eight cores
per cell and 24 total cores overall. In addition, to
facilitate evaluations by other investigators,
48
-------
additional cores were collected simultaneously in
these cells as well as the other capping cells (i.e.,
apatite and coke breeze). Cores were collected
from the deck of an aluminum coring vessel
equipped with a coring derrick and a sampling
moonpool. A vibratory head was hoisted by the
coring derrick and connected to a core barrel
containing a butyrate liner. The butyrate liner
used was approximately 3-in in diameter. The
barrel was lowered to the sediment surface and
pushed into the sediment under direct pressure
or by adding vibration until an adequate depth
had been achieved to visually assess sediment
lithology and collect necessary laboratory
samples. The core barrel was then extracted
using a winch on the coring derrick. Upon core
retrieval, the butyrate liner was extracted, cut to
length and as not to disturb the core, wiped
clean, capped on both ends using plastic caps
and electrical tape, labeled, stored upright, and
transported to Battelle for processing and sample
extraction. Accurate positional control was
maintained using a GPS, and the coring vessel
was held in place over each coring station using
spuds lowered gently through the water and into
the bottom sediments.
Battelle established a core processing facility at
the GSA property located north of the
investigation area. The core processing facility
consisted of an approximately 16-ft enclosed
truck equipped with a collapsible processing table
and all expendable materials and supplies
needed.
For each quadrant of each cell, the two replicate
cores were evaluated visually through the
sidewall of the butyrate liner to determine
sediment lithology and determine the thickness of
various sediment layers and the location of
lithologic interfaces. After determining this
lithologic information, Battelle personnel
dissected the cores to generate samples for
laboratory analysis of COCs and physical
parameters according to a previously established
sampling plan. According to this previously
established sampling plan, an attempt was made
to collect the following sample intervals
depending on the actual thickness of various
lithologic intervals:
• AquaBlok cell: collect three individual
samples from the overlying sand layer, one
sample from the interface of the overlying
sand layer and the AquaBlok® capping
layer, three individual samples within the
AquaBlok® capping layer, one sample from
the interface of AquaBlok® and native
sediments, and one sample from the upper
horizon of the native sediment unit, for a
total of nine unique samples per core.
• Sand cell: collect four individual samples
from the sand capping layer, one sample
from the interface of the sand capping layer
and the underlying native sediment, and
one sample from the upper horizon of the
native sediment unit, for a total of six
individual samples per core.
• Control cell: collect three individual
samples from the upper horizon of the
native sediment unit, for a total of three
individual samples per core.
Each sample interval was intended to be a 3 cm
vertical segment of material. A stainless steel
hacksaw was used to cut the cores at intervals
determined to correspond to the desired sample
intervals described above. For each pair of
replicate cores collected from each cell quadrant,
material from each corresponding sampling
interval was composited in a stainless steel
mixing bowl and mixed until of uniform color and
consistency. Subsequently, composited material
was placed in laboratory-provided sample
glassware for analysis of COCs and physical
parameters. Accordingly, actual data were only
generated for one core per quadrant per cell. All
samples were properly labeled and stored on ice
in coolers under appropriate chain of custody
until delivered to the analytical laboratories.
Each sampling interval from each core was
analyzed for six individual metals (Cd, Cr, Cu, Pb,
Hg, and silver [Ag]), PCB congeners, PAHs,
PSD, and TOC. Metals analyses were conducted
by Battelle's Sequim Marine Laboratory (Sequim)
in Sequim, Washington; PCB and PAH analyses
were conducted by Battelle's Duxbury Operations
(EDO) of Duxbury, Massachusetts, and PSD and
TOC analyses were conducted by Applied Marine
Sciences (AMS) of League City, Texas. Metals,
49
-------
PCB, and PAH analyses were conducted using
standard EPA methods, specifically Method
601 OB for all metals other than Hg, Method
7471A for Hg, modified Method 8270 for PCBs,
and Method 8270/8015 for PAHs. TOC and PSD
analyses were conducted using appropriate EPA
or ASTM methods, specifically EPA Method
9060M or ASTM Method D2974-00 for TOC and
ASTM Method D422 for PSD.
QC procedures implemented during the sediment
coring program included recording detailed
information on location related to individual cores,
ensuring the working order and maintenance of
all vibracoring components and equipment,
adhering to appropriate laboratory methods and
procedures, and collecting multiple cores from
each cell for adequate data coverage. All
reusable materials and supplies were properly
decontaminated prior to reuse.
3.2.4.3 18-Month Post-Capping Field
Event. The 18-month post-capping field event
was conducted between August and September
2005. The specific monitoring tools used during
this event consisted of bathymetric and sub-
bottom profiling, side-scan sonar surveying, SPI,
seepage meter testing, sediment coring for the
analysis of COCs, physical parameters, and
hydraulic conductivity, and gas flux analysis.
Specific 18-month post-capping field event
information for these individual monitoring tools is
described below.
3.2.4.3.1 18-Month Post-Capping Field
Event Bathymetry and Sub-bottom
Profiling. On September 15, 2005, a third
integrated bathymetric and sub-bottom profiling
survey was completed in the demonstration area.
The 18-month post-capping bathymetric and sub-
bottom profiling survey was completed by OSI
under the direct supervision of Battelle. All
methods and procedures followed, including
QA/QC, were generally identical to the one-
month and six-month post-capping surveys
described in Sections 3.2.4.1.1 and 3.2.4.2.1,
respectively. To ensure the comparability of data
between the 18-month event and the one-month
and six-month events, the identical survey
transects were used.
Additional detail relating to the general principles
of bathymetric and sub-bottom surveying can be
found in Section 3.2.3.3, and details specifically
related to the 18-month post-capping bathymetric
and sub-bottom profiling survey are included in
Appendix B-3.
3.2.4.3.2 18-Month Post-Capping Field
Event Side-Scan Sonar Surveying. On
September 14, 2005, a second side-scan sonar
survey was completed in the demonstration area.
The 18-month post-capping side-scan sonar
survey was completed by OSI under the direct
supervision of Battelle. All methods and
procedures followed, including QA/QC, were
generally identical to the one-month post-capping
survey described in Section 3.2.4.1.2. To ensure
the comparability of data between the 18-month
event and the one-month events, the identical
side-scan survey transects were used.
Additional detail relating to the general principles
of side-scan sonar surveying can be found in
Section 3.2.3.4, and details specifically related to
the 18-month post-capping side-scan sonar
survey are included in Appendix B-3.
3.2.4.3.3 18-Month Post-Capping Field
Event Sediment Profile Imaging. On
September 16, 2005, a third thorough SPI
investigation was completed in the demonstration
area. The 18-month post-capping SPI survey
was completed by RJ Diaz (with vessel support
provided by OSI) under the direct supervision of
Battelle. All methods and procedures followed,
including QA/QC, were generally identical to the
one-month and six-month post-capping surveys
described in Section 3.2.4.1.3 and Section
3.2.4.2.2, respectively. To facilitate direct
comparison between SPI data from the 18-month
post-capping event and SPI data from the one-
month and six-month post-capping events, an
effort was made to drop the camera during the
18-month event relatively near the equivalent
locations from the previous events while not
being closer than approximately five to eight feet
of the previous locations. During the 18-month
survey, three Anacostia River channel locations
were targeted with the high-resolution SPI
camera and four channel locations were targeted
with the video SPI camera.
50
-------
Additional detail relating to the general principles
of SPI surveying can be found in Section 3.2.3.5,
and details specifically related to the 18-month
post-capping SPI survey are included in
Appendix C.
3.2.4.3.4 18-Month Post-Capping Field
Event Seepage Meter Testing. Between
September 19 and 23, 2005, a third seepage
meter investigation was implemented in the
demonstration area. Because other investigators
did not utilize seepage meters during the 18-
month post-capping event, this activity was
implemented only in the AquaBlok®, sand, and
control cells.
In the AquaBlok®, sand, and control cells, two
ultrasonic seepage meters each were deployed
and retrieved by divers from Matrix. All methods
and procedures followed, including QA/QC, were
generally identical to the one-month and six-
month post-capping seepage meter testing
described in Sections 3.2.4.1.4 and 3.2.4.2.3,
respectively. Meters were deployed for
approximately four to six days, but deployment
locations were not specifically near the
deployment locations from the one-month or six-
month post-capping evaluations. One meter in
the AquaBlok® cell became air locked during
deployment but was cleared and resumed
measurements.
During the 18-month post-capping seepage
meter testing, surface water temperature and
pressure were measured continuously at one
reference location established at the ECC dock,
and, as with the one-month and six-month post-
capping events, groundwater elevation,
temperature, and conductivity were measured
continuously in two monitoring wells located
upland on the north side of the river.
Additional detail relating to the 18-month post-
capping seepage meter survey is included in
Appendix D-3.
3.2.4.3.5 18-Month Post-Capping Field
Event Sediment Coring and Analysis of
Contaminants of Concern, Physical
Parameters, and Hydraulic Conductivity.
Between September 27 and 28, 2005, a second
sediment coring investigation was completed in
the demonstration area. Multiple sediment cores
were collected using vibracoring techniques from
a coring vessel owned and operated by Athena.
Coring was overseen by Battelle and sediment
cores were evaluated and processed by Battelle
personnel.
During the 18-month post-capping field event,
two individual cores were collected from two of
the four unique quadrants in the AquaBlok®,
sand, and control cells, for a total of four cores
per cell and 12 total cores overall. In addition, to
facilitate evaluations by other investigators,
additional cores were collected simultaneously in
these cells as well as the other capping cells (i.e.,
apatite and coke breeze). Cores were collected
according to the same procedures and standards
described for the six-month post-capping coring
event in Section 3.2.4.2.5. To facilitate direct
comparison between coring data from the 18-
month post-capping event and coring data from
the six-month post-capping event, an effort was
made to collect cores during the 18-month event
from relatively near the equivalent locations from
the six-month event while not being closer than
approximately 5 to 8 ft of the previous locations
or closer than approximately 8 to 10 ft of a cell's
edge.
Battelle established a core processing facility at
the WASA property located north of the
investigation area. The core processing facility
was nearly identical to that described in Section
3.2.4.2.5. For each quadrant sampled from each
cell, the two replicate cores were evaluated by
cutting away a lengthwise section of the butyrate
liner to directly view the intact vertical sediment
profile. From this, sediment lithology, thickness
of various sediment layers and the location of
lithologic interfaces were determined. After
determining this information, Battelle personnel
dissected the cores to generate samples for
laboratory analysis of COCs and physical
parameters according to the same previously
established sampling plan described in Section
3.2.4.2.5.
Stainless steel spoons or scoops were used to
remove sediment from the cores from intervals
determined to correspond to the selected sample
51
-------
intervals of interest. For each pair of replicate
cores collected from each cell quadrant, material
from each corresponding sampling interval was
composited in a stainless steel mixing bowl and
mixed until of uniform color and consistency.
Subsequently, composited material was placed in
laboratory-provided sample glassware for
analysis of COCs and physical parameters.
Accordingly, actual data were only generated for
one core per quadrant per cell. All samples were
properly labeled and stored on ice in coolers
under appropriate chain of custody until delivered
to the analytical laboratories.
Each sampling interval from each core was
analyzed for the same constituents as described
in Section 3.2.4.2.5 (i.e., Cd, Cr, Cu, Pb, Hg, Ag,
PCB congeners, PAHs, PSD, and TOC). As with
the six-month post-capping coring activity, metals
analyses were conducted by Sequim, PCB and
PAH analyses were conducted by EDO, and PSD
and TOC analyses were conducted by AMS. All
analyses were conducted using standard EPA or
ASTM methods as described in Section 3.2.4.2.5.
In addition, during the 18-month post-capping
coring activity, two cores each were collected
from locations within the AquaBlok®, sand, and
control cells for evaluation of hydraulic
conductivity. These cores were collected from
two of the four cell quadrants for each cell in the
same manner as those collected for lithologic
evaluation and chemical analysis. During coring,
it was ensured that these cores penetrated a
sufficient vertical distance to obtain the full
vertical thickness of the material of interest for
each cell area (i.e., AquaBlok®, sand, or native
sediment). These cores were left intact (i.e.,
capped within the butyrate liner and not opened
or disturbed), labeled, stored on ice, and
delivered under appropriate chain of custody to
AMS for the laboratory determination of hydraulic
conductivity according to ASTM Method D5084-
D.
QC procedures implemented during the sediment
coring program included recording detailed
information on location related to individual cores,
ensuring the working order and maintenance of
all vibracoring components and equipment,
adhering to appropriate laboratory methods and
procedures, and collecting multiple cores from
each cell for adequate data coverage. All
reusable materials and supplies were properly
decontaminated prior to reuse.
3.2.4.3.6 18-Month Post-Capping Field
Event Gas Flux Analysis. Between August
24 and September 26, 2005, an initial gas flux
investigation was implemented in the
demonstration area. Because no other
investigators utilized this monitoring tool, gas flux
chambers were deployed only in the AquaBlok®,
sand, and control cells. The general principles
and methods of gas flux monitoring are described
in Section 3.2.3.6.
On August 25, 2005, stainless steel gas flux
chambers were deployed at two locations each in
the AquaBlok®, sand, and control cells. The
chambers were generally consistent with the
conceptual image provided in Figure 3-13, but
had only one top valve and a more conical upper
chamber component.
The chambers were installed on the sediment
surface by a dive team from K&M Marine, Inc.
(K&M) of Lusby, Maryland, under the oversight of
Battelle personnel. At each deployment location,
a chamber was transported to the sediment
surface by a diver, pushed into the bottom
sediment until firmly seated, and anchored in
place by two cinder blocks connected with chain
to the chamber at welded anchor points. In
general, the chambers were pushed into the
bottom sediments by approximately 6 to 8 in.
The locations of the chambers were determined
prior to deployment. The surface vessel carrying
the dive team was positioned over the
predetermined locations using GPS and the diver
was released to rapidly descend. At the request
of the EPA, the side valves on each chamber
were left in a closed position.
On September 26, 2005, following an
approximately one-month deployment period,
Battelle personnel collected gas samples from
the submerged gas flux chambers. At each
monitoring location, a diver from K&M descended
to the chamber and attached a male quick
connect fitting to the female quick connect fitting
on the chamber's top. The male quick connect
52
-------
fitting was in turn connected to a length of
polyethylene tubing extending to a surface
support vessel. At the surface, the tubing was
connected to a 1-liter (L) airtight, graduated,
acrylic syringe. The plunger of the syringe was
retracted, and all gas that had collected in the
submerged chamber was extracted and its
volume determined.
Upon extraction, gas samples were injected into
Tedlar bags which were labeled, stored, and
shipped under appropriate chain of custody
protocol to AirToxics, Ltd. (AirToxics) of Folsom,
California. Gas samples were then analyzed for
total non-methane organic carbon (TNMOC),
common gases (i.e., carbon dioxide [CO2],
methane [CH4], nitrogen [N2], and oxygen [O2]),
and reduced sulfur compounds using standard
EPA or ASTM methods. Specifically, TNMOC
was evaluated using EPA Method 25C, common
gases were evaluated using EPA Method 3C,
and reduced sulfur compounds were evaluated
using ASTM Method D5504.
At the conclusion of the 18-month post-capping
flux chamber sampling event, the chambers were
left in place to be used for the subsequent 30-
month post-capping flux chamber sampling event
(see Section 3.2.4.4.7). However, it was
subsequently determined that there was a
reasonable risk that the chambers could be lost
or significantly fouled. Accordingly, on April 14,
2006, the flux chambers were removed by a dive
team from K&M under the supervision of Battelle
personnel and the chambers were stored pending
their use during the 30-month post-capping field
events.
QC procedures implemented during the gas flux
sampling program included recording detailed
information on locations of individual chambers,
ensuring the working order and maintenance of
all diving and chamber-related components and
equipment, adhering to appropriate laboratory
methods and procedures, and collecting
adequate samples from each cell for suitable
data coverage. Other QC procedures related to
the evaluation of gas flux chamber data are
described in Section 3.2.4.5 below.
3.2.4.4 30-Month Post-Capping Field
Event. The 30-month post-capping field event
was conducted between August and October
2006. The specific monitoring tools used during
this event consisted of bathymetric and sub-
bottom profiling, side-scan sonar surveying, SPI,
seepage meter testing, Sedflume analysis,
sediment coring for the analysis of COCs,
physical parameters, and hydraulic conductivity,
gas flux analysis, and benthic grab sampling and
descriptive and statistical benthic assays.
Specific 30-month post-capping field event
information for these individual monitoring tools is
described below.
3.2.4.4.1 30-Month Post-Capping Field
Event Bathymetry and Sub-Bottom
Profiling. On September 19, 2006, a fourth
integrated bathymetric and sub-bottom profiling
survey was completed in the demonstration area.
The 30-month post-capping bathymetric and sub-
bottom profiling survey was completed by OSI
under the direct supervision of Battelle. All
methods and procedures followed, including
QA/QC, were generally identical to the one-
month, six-month, and 18-month post-capping
surveys described in Sections 3.2.4.1.1,
3.2.4.2.1, and 3.2.4.3.1, respectively. To ensure
the comparability of data between the 30-month
event and the previous events, the identical
survey transects were used during the 30-month
post-capping survey as were used during the
previous surveys.
Additional detail relating to the general principles
of bathymetric and sub-bottom surveying can be
found in Section 3.2.3.3, and details specifically
related to the 30-month post-capping bathymetric
and sub-bottom profiling survey are included in
Appendix B-4.
3.2.4.4.2 30-Month Post-Capping Field
Event Side-Scan Sonar Surveying. On
September 20, 2006, a third side-scan sonar
survey was completed in the demonstration area.
The 30-month post-capping side-scan sonar
survey was completed by OSI under the direct
supervision of Battelle. All methods and
procedures followed, including QA/QC, were
generally identical to the one-month and 18-
month post-capping surveys described in Section
53
-------
3.2.4.1.2 and 3.2.4.3.2, respectively. To ensure
the comparability of data between the 30-month
event and the one-month and 18-month events,
identical side-scan survey transects were used
during the 30-month post-capping survey as were
used during the previous surveys.
Additional details relating to the general principles
of side-scan sonar surveying can be found in
Section 3.2.3.4, and details specifically related to
the 30-month post-capping side-scan sonar
survey are included in Appendix B-4.
3.2.4.4.3 30-Month Post-Capping Field
Event Sediment Profile Imaging. Between
September 20 and 21, 2006, a fourth thorough
SPI investigation was completed in the
demonstration area. The 30-month post-capping
SPI survey was completed by RJ Diaz (with
vessel support provided by OSI) under the direct
supervision of Battelle. All methods and
procedures followed, including QA/QC, were
generally identical to the one-month, six-month,
and 18-month post-capping surveys described in
Sections 3.2.4.1.3, 3.2.4.2.2, and 3.2.4.3.3,
respectively. To facilitate direct comparison
between SPI data from the 30-month post-
capping event and SPI data from the previous
post-capping events, an effort was made to drop
the camera during the 30-month event relatively
near the equivalent locations from the previous
events while not being within approximately 5 to 8
ft of the previous locations. During the 30-month
survey, seven Anacostia River channel locations
were targeted with the high-resolution SPI
camera and one channel location was targeted
with the video SPI camera for additional
reference data.
Additional details relating to the general principles
of SPI surveying can be found in Section 3.2.3.5,
and details specifically related to the 30-month
post-capping SPI survey are included in
Appendix C.
3.2.4.4.4 30-Month Post-Capping Field
Event Seepage Meter Testing. Between
September 25 and 30, 2006, a fourth seepage
meter investigation was implemented in the
demonstration area. Because other investigators
did not utilize seepage meters during the 30-
month post-capping event, this activity was
implemented only in the AquaBlok®, sand, and
control cells.
In the AquaBlok®, sand, and control cells, two
ultrasonic seepage meters each were deployed
and retrieved by divers from Matrix. All methods
and procedures followed, including QA/QC, were
generally identical to the one-month, six-month,
and 18-month post-capping seepage meter
testing described in Sections 3.2.4.1.4, 3.2.4.2.3,
and 3.2.4.3.4, respectively. Meters were
deployed for approximately three to four days, but
deployment locations were not specifically near
the deployment locations from the previous post-
capping evaluations. One meter in the
AquaBlok® cell was relocated during deployment
after initial data collected suggested it might have
been located in an area where the cap had been
compromised by other sampling events (i.e., the
data suggested there was potentially a
preferential flow path at the meter location,
essentially short-circuiting any potential hydraulic
control).
During the 30-month post-capping seepage
meter testing, surface water temperature and
pressure were measured continuously at one
reference location established at the ECC dock,
and, as with the other post-capping events,
groundwater elevation, temperature, and
conductivity were measured continuously in two
monitoring wells located upland on the north side
of the river.
Additional details relating to the 30-month post-
capping seepage meter survey are included in
Appendix D-4.
3.2.4.4.5 30-Month Post-Capping Field
Event Sedflume Analysis. Between October
17 and 22, 2006, a second Sedflume evaluation
was conducted in the demonstration area. The
evaluation was completed for the AquaBlok®,
sand, and control cells by SEI under the direct
oversight of Battelle.
During the Sedflume evaluation, 12 individual box
cores were collected from the demonstration
area, four each from the AquaBlok®, sand, and
control cells. One core was collected from each
54
-------
of four quadrants (i.e., NW, NE, SE, and SW)
similar to the approach used during the six-month
post-capping Sedflume study described in
Section 3.2.4.2.4. The cores were collected in
the same manner as the six-month survey.
Cores were evaluated immediately in a mobile
Sedflume laboratory provided by SEI and
stationed at the WASA property north of the
investigation area. The mobile laboratory was
generally identical to that described for the six-
month post-capping evaluation (see Section
3.2.4.2.4). To derive critical shear resistance
information from the Sedflume, it was operated in
a manner consistent with the description in
Section 3.2.31. For each core, a series of shear
stresses were applied to the core in the test
section of the Sedflume, and time and erosion
were recorded until at least 1 to 2 mm but not
more than 2 cm were eroded. Initial shear
stresses were low and were gradually increased
until a maximum stress was run for each
particular test. Each subsequent shear stress
was generally twice the previous.
To allow for correlation between shear stress and
important physical characteristics, interval
samples from the cores were periodically
analyzed for PSD and water content in the mobile
laboratory. These samples were collected from
the top of the core after each erosion cycle.
Water content was determined through bulk
density analysis according to ASTM method D-
2216. PSD was determined through laser
dispersion.
Independent flow field measurements to
characterize the hydrologic profile of the
Anacostia River were not included as part of the
Sedflume approach during the 30-month survey,
consistent with the six-month survey. Such
measurements would have included the use of an
ADCP to calculate specific in-river velocities and
sediment transport characteristics. The reason
for not including the ADCP is that AquaBlok® cap
failure due to high flow rates is unexpected due to
the relatively sluggish flow of the Anacostia River
(see Section 3.1) and the Sedflume procedures
themselves included manipulating laboratory
flows to levels higher than and uncharacteristic of
the Anacostia River. Thus, ADCP measurements
would have provided little information regarding
the potential for cap failure that was not gleaned
from the Sedflume laboratory protocol.
QC procedures implemented during the Sedflume
program included recording detailed information
on location related to individual cores, ensuring
the working order and maintenance of all
Sedflume components and equipment, adhering
to appropriate laboratory methods and
procedures, and collecting multiple cores from
each cell for adequate data coverage. All
reusable materials and supplies were
decontaminated prior to reuse.
Additional detail relating to the 30-month post-
capping Sedflume study is included in Appendix
E-2.
3.2.4.4.6 30-Month Post-Capping Field
Event Sediment Coring and Analysis of
Contaminants of Concern, Physical
Parameters, and Hydraulic Conductivity.
Between October 17 and 19, 2006, a third
sediment coring investigation was completed in
the demonstration area. Multiple sediment cores
were collected using vibracoring techniques from
a coring vessel owned and operated by Athena.
Coring was overseen by Battelle and sediment
cores were evaluated and processed by Battelle
personnel.
During the 30-month post-capping field event,
two individual cores were collected from each of
the four unique quadrants in the AquaBlok®,
sand, and control cells, for a total of eight cores
per cell and 24 total cores overall. In addition, to
facilitate evaluations by other investigators,
additional cores were collected simultaneously in
these cells as well as the other capping cells (i.e.,
apatite and coke breeze). Cores were collected
according to the same procedures and standards
described for the six-month post-capping coring
event in Section 3.2.4.2.5. To facilitate direct
comparison between coring data from the 30-
month post-capping event and coring data from
the previous coring events, an effort was made to
collect cores during the 30-month event from
relatively near the equivalent locations from the
previous events while not being closer than
approximately 5 to 8 ft of the previous locations
55
-------
or closer than approximately 8 to 10 ft of a cell's
edge.
Battelle established a core processing facility at
the WASA property located north of the
investigation area. The core processing facility
was nearly identical to that described in Section
3.2.4.2.5. For each quadrant sampled from each
cell, the two replicate cores were evaluated in
identical fashion to the 18-month post-capping
coring work described in Section 3.2.4.3.5.
Stainless steel spoons or scoops were used to
remove sediment from the cores from intervals
determined to correspond to the selected sample
intervals of interest. For each pair of replicate
cores collected from each cell quadrant, material
from each corresponding sampling interval was
composited in a stainless steel mixing bowl and
mixed until of uniform color and consistency.
Subsequently, composited material was placed in
laboratory-provided sample glassware for
analysis of COCs and physical parameters.
Accordingly, actual data were only generated for
one core per quadrant per cell. All samples were
properly labeled and stored on ice in coolers
under appropriate chain of custody until delivered
to the analytical laboratories.
Each sampling interval from each core was
analyzed for the same constituents as described
in Section 3.2.4.2.5 (i.e., Cd, Cr, Cu, Pb, Hg, Ag,
PCB congeners, PAHs, PSD, and TOC). As with
the six-month and 18-month post-capping coring
activities, metals analyses were conducted by
Sequim, PCB and PAH analyses were conducted
by EDO, and PSD and TOC analyses were
conducted by AMS. All analyses were
conducted using standard EPA or ASTM
methods as described in Section 3.2.4.2.5.
In addition, during the 30-month post-capping
coring activity, two cores each were collected
from locations within the AquaBlok®, sand, and
control cells for evaluation of hydraulic
conductivity. These cores were collected from
two of the four cell quadrants for each cell, and
were collected in identical fashion to those
collected for lithologic evaluation and chemical
analysis. During coring, it was ensured that
these cores penetrated a sufficient vertical
distance to obtain the full vertical thickness of the
material of interest for each cell area (i.e.,
AquaBlok®, sand, or native sediment). These
cores were left intact (i.e., capped within the
butyrate liner and not opened or disturbed),
labeled, stored on ice, and delivered under
appropriate chain of custody to AMS for the
laboratory determination of hydraulic conductivity
according to ASTM Method D5084-D. To
facilitate direct comparison between hydraulic
conductivity data from the 30-month post-capping
event and from the 18-month coring event, an
effort was made to collect hydraulic conductivity
cores during the 30-month event from relatively
near the equivalent locations from the previous
event while not being closer than approximately 5
to 8 ft of the previous locations.
QC procedures implemented during the sediment
coring program included recording detailed
information on location related to individual cores,
ensuring the working order and maintenance of
all vibracoring components and equipment,
adhering to appropriate laboratory methods and
procedures, and collecting multiple cores from
each cell for adequate data coverage. All
reusable materials and supplies were properly
decontaminated prior to reuse.
3.2.4.4.7 30-Month Post-Capping Field
Event Gas Flux Analysis. Between August
14 and September 13, 2006, a second gas flux
investigation was implemented in the
demonstration area. Because no other
investigators utilized this monitoring tool, gas flux
chambers were deployed only in the AquaBlok®,
sand, and control cells. The general principles
and methods of gas flux monitoring are described
in Section 3.2.3.6.
On August 14, 2006, stainless steel gas flux
chambers were deployed at two locations each in
the AquaBlok®, sand, and control cells. The
chambers were the same chambers used during
the 18-month post-capping gas flux evaluation
(see Section 3.2.4.3.6).
The chambers were installed on the sediment
surface by a dive team from K&M under the
oversight of Battelle personnel. At each
deployment location, a chamber was transported
56
-------
to the sediment surface by a diver and installed in
identical fashion to that described in Section
3.2.4.3.6. To facilitate direct comparison
between gas flux data from the 30-month post-
capping event and from the 18-month coring
event, an effort was made to position the
chambers during the 30-month event immediately
near the equivalent locations from the previous
event while not being within approximately 5 to 8
ft of the previous locations. Consistent with the
18-month post-capping deployment, the side
valves on each chamber were left in a closed
position.
On September 13, 2006, following an
approximately one-month deployment period,
Battelle personnel collected gas samples from
the submerged gas flux chambers. Samples
were collected in identical fashion to that
described in Section 3.2.4.3.6.
Upon extraction, and as with the 18-month post-
capping event, gas samples were injected into
Tedlar bags which were labeled, stored, and
shipped under appropriate chain of custody
protocol to AirToxics. Gas samples were then
analyzed for TNMOC, common gases (i.e., CO2,
CH4, N2, and O2), and reduced sulfur compounds
using the same standard EPA or ASTM methods
used for the 18-month post-capping event.
At the conclusion of the 30-month post-capping
flux chamber sampling event, the chambers were
removed by the dive team from K&M.
QC procedures implemented during the gas flux
sampling program included recording detailed
information on location related to individual
chambers, ensuring the working order and
maintenance of all diving and chamber-related
components and equipment, adhering to
appropriate laboratory methods and procedures,
and collecting adequate samples from each cell
for suitable data coverage. Other QC procedures
related to the evaluation of gas flux chamber data
are described in Section 3.2.4.5 below.
3.2.4.4.8 30-Month Post-Capping Field
Event Benthic Grab Sampling and
Descriptive and Statistical Benthic
Assays. Between October 17 and 19, 2007, 36
sediment grab samples (three samples from each
of the four quadrants within the AquaBlok®, sand,
and control cells) were collected to evaluate
benthic infaunal communities in the
demonstration area.
Benthic samples for infaunal analyses were
collected using a stainless steel 0.04-m2 modified
Van Veen grab sampler deployed from the
sampling vessel owned and operated by Athena
simultaneously with the 30-month post-capping
sediment coring activities. The sampling platform
was equipped with dGPS connected to a laptop
computer running navigational software for
positional control. Sample collection coordinates
were stored electronically on the laptop in real-
time during field operations.
The open grab was lowered to the river bottom
from the sampling vessel. When the line went
slack, a mechanical, counterweighted latch
released the arms, allowing them to close the
grab as the line was retrieved. Once the closed
grab was returned to the sampling vessel, the top
covers of the grab were opened and the contents
inspected. If the sample was adequate in volume
and quality, it was deemed acceptable. The
sediment sample depth within the grab was
measured to the nearest 0.1 cm and recorded for
sample volume calculations. The sample was
transferred to a pre-marked sample tray for
storage and transport to shore. Three grab
samples were collected from the bow of the
sample vessel at each location (one from the
port-bow corner, one from the center-bow, and
one from the starboard-bow corner). After the
three grab samples were collected, they were
transferred to shore for processing. Once on
shore, each benthic sample was rinsed with river
water over nested 1.0- and 0.5-mm sieves to
remove fine sediment particles. Material retained
on the sieves was transferred carefully into
labeled polyethylene bottles. Samples were fixed
in the field by adding buffered 10 percent (%)
formalin solution (3.7% formaldehyde) to each
sample bottle.
Infauna was removed from the sediment grab
samples and taxonomic identifications were
performed by Cove Corporation (Cove) of Lusby,
Maryland. The 36 individual samples were sorted
57
-------
to remove at least 95% of the infaunal organisms.
QC was accomplished by re-sorting a complete
sample once for at least every nine samples
sorted. Data from the three grab samples
collected from each cell quadrant were pooled
(summed) for statistical analysis. The equivalent
sample size of the pooled grab samples was 0.12
m2. A total of 12 pooled observations were
completed for the demonstration area (i.e., three
pooled grabs in four quadrants each for three
cells).
The primary ecological metrics used to evaluate
infaunal communities during the 30-month post-
capping benthic assessment were total
abundance, total species, Sander's Rarefaction
(E[Sn]; modified by Hurlbert, 1971), species
presence-absence (occurrence), major taxon
abundance, and the most abundant species.
Several other traditional ecological metrics,
including H' (calculated using Iog2), J', log-series
alpha diversity (May, 1975), and Margalefs
species richness (d), were calculated for each
sample and reported because of their general
ecological interest (Ludwig and Reynolds, 1988).
The software package Primer 5 for Windows
(Version 5.2.9, ©2002, Primer-E, Ltd.) was used
to calculate all of these metrics except Sander's
rarefaction. BioDiversity Professional, Version 2
(© 1997 The Natural History Museum/ Scottish
Association for Marine Science) was used to
calculate the rarefaction values.
Standard descriptive statistics including the
mean, median, and coefficient of variation (CV)
were calculated, and boxplots generated for
selected ecological parameters for each pooled
sample by using Microsoft® Excel or Minitab™
software. Several ecological parameters and the
abundances of selected key taxa were evaluated
by using the Kruskal-Wallis test of equal median
response.
Similarity analyses, using the Bray-Curtis
similarity coefficient (S1) as a measure of distance
between stations (described in Clarke and
Warwick, 2001), were also performed using
Primer™. Abundance data were square-root
transformed, but not standardized prior to
analyses. The similarity matrix was converted to
a dendrogram using the hierarchical, unweighted
pair-group mean-averaging method of clustering.
Each similarity matrix was also transformed to a
two-dimensional, non-metric multidimensional
scaling (nMDS) plot, which expresses the Bray-
Curtis similarity in two dimensions such that more
similar samples are spatially close together (Clark
and Warwick, 2001). Primer™ generated each
plot by restarting the nMDS algorithm 30 times
and selecting the lowest stress value (Clarke and
Warwick, 2001). Several physical and biological
parameters were then mapped onto the nMDS
plots to help identify factors that might explain the
similarity patterns identified by the analysis.
Finally, the potential relationships between the
faunal communities in the Anacostia River and
selected physical factors were evaluated by an
approach similar to that conducted for the
biological analyses. Primer™ was used to run a
similarity analysis of a reduced set of physical
parameters with normalized Euclidean distance
as the similarity measure. Primer™ then
generated an nMDS plot based on the resulting
similarity matrix. Several physical and biological
parameters were then mapped onto the nMDS
plots to help identify faunal distribution that might
be explained by the similarity patterns based on
physical habitat characteristics.
The complete details of the sorting and
identification process, the results of QC checks,
and a more detailed discussion of the statistical
methods used to reduce and evaluate the benthic
data are presented in Appendix F.
3.2.4.5 General AquaBlok® SITE
Demonstration Quality Assurance and
Quality Control. Specific QA/QC procedures
for each of the various field monitoring and
sampling tools are described above in Section
3.2.4.
In addition, general QA/QC procedures were
adhered to during the SITE demonstration
program to ensure the representativeness and
usability of all data generated. Specifically,
throughout the demonstration, all efforts were
made to not collect any two samples of any type
or deploy any two monitoring devices of any type
during single events or between different
sampling events any closer than approximately 6
58
-------
ft apart. In addition, all efforts were made to not
locate any sample or monitoring device any
closer than approximately 10 ft from a cell's edge
(with the exception of the control cell) to minimize
potential edge effects. Accurate GPS/dGPS data
were collected throughout all sampling events to
achieve this end. Figures 3-16 through 3-19
show the locations of the various samples
collected and monitoring tools deployed during
the one-month, six-month, 18-month, and 30-
month post-capping events, respectively.
Given the extensive number of monitoring tools
used, the number of sampling events executed,
and the number of individual samples collected, a
limited number of samples were collected from
less than approximately 6 ft apart between
sampling events. In addition, certain monitoring
tools actually required targeting similar or
identical locations between sampling events.
Specifically, oceanographic surveying was
conducted along identical survey lines for each
event by design, and SPI camera drops were
conducted near one another, but not immediately
atop one another, between the various SPI
surveys. In addition, gas flux chambers were
intentionally located in nearly identical positions
between the two gas flux sampling events.
All measurements, observations, and data
generated in the field during the demonstration
were recorded in dedicated field journals or
directly into a laptop computer for later
processing. Data were processed, compiled, and
analyzed both manually and by specific computer
software. All data and derived products were
stored in Battelle and/or laboratory/subcontractor
computers, and copied to compact disc (CD) as
needed. In addition, hard copy deliverables were
produced by most if not all laboratories/
subcontractors.
General and specific QA/QC procedures for the
AquaBlok® demonstration were summarized in
the project Quality Assurance and Project Plan
(QAPP) (Battelle, 2004), and were adhered to
other than as noted specifically in this ITER.
3.3
AquaBlok® SITE Demonstration
Results
The following sections present the results of and
conclusions drawn from the AquaBlok® SITE
demonstration program in the Anacostia River
demonstration area. For simplicity, results and
conclusions are summarized by study objective
rather than by individual sampling/monitoring tool.
3.3. 1 Objective #1 - Physical Stability of
An AquaBlok® Cap
As indicated in Section 3.2.1, the physical
stability of AquaBlok® in flowing water depends
primarily on the material's physical strength (e.g.,
shear strength) and its ability to withstand shear
stresses imposed by surface water flow field
currents at the cap/water interface. One of the
most critical design characteristics of AquaBlok®
is that, given its high degree of cohesiveness
related to its material composition, it claims to
have a higher resistance to shear energy
compared to traditional capping materials (e.g.,
sand).
To evaluate the physical stability of AquaBlok®
relative to sand and native sediments, the
following critical and non-critical measurements
were identified and assessed through data
collection during the various SITE demonstration
sampling events.
Critical Measurements
o Sedflume coring and analysis;
o Sediment coring and analysis of COCs;
o Bathymetry and sub-bottom profiling; and
o Side-scan sonar surveying
Non-critical Measurements
o SPI;
o Gas flux analysis; and
o Sediment coring and analysis of physical
parameters
59
-------
Explanation
Sediment Profile Imaging (SPI) Location
Seepage Meter(SM) Location
AQUABLOK CELL
APATI ECEI
SAND CEI
Control 10 Controls- ANA5-2
COKE BREEZE
CELL
CONTROL
CELL
Month 1 Samp ling Locations,
Anacostia Eiver, Washington, D.C.
ANACOSTIARIVER
* I '4 I 2006AU_SAMPLOCS.CDR I 02,07
Figure 3-16. One-Month Post-Capping Field Event Sampling/Monitoring Locations
-------
Explanation
Sediment Profile Imaging (SPI) Location
Seepage Meter (SM) Location
SerJflume Core (SF) Location
Chemistry Core (CC) Location
SAND CEI
AQUABLOK CELL
AquaBWK 4
WM-Jtff- (01-06)
F-
AquaBlo\7 AquaBloklO
AquaBlok 5
Btok8 AQB3- AquaBlok 3
APATITE CELL
COKE BREEZE
CELL
CS1
CONTROL
Month 6 Sampling Locations,
Anacostia River, Washington, B.C
ANACOSTIA RIVER
Figure 3-17. Six-Month Post-Capping Field Event Sampling/Monitoring Locations
-------
Sediment Profile Imaging (SPI) Location
Seepage Meter (SM) Location
Flux Chamber (FC) Location
0 Chemistry Core (CC) Location
-I Hydraulic Conductivity Core (HCC) Location
AQUABLOK CELL
APATITE CELL
SAND CELL
COKE BREEZE
CELL
CONTROL
CELL
Month 18 Sampling Locations,
Anacostia River, Washington, D.C
ANACOSTIA RIVER
Figure 3-18. 18-Month Post-Capping Field Event Sampling/Monitoring Locations
-------
Explanation
* Sediment Profile Imaging (SPI) Location
• Seepage Meter (SM) Location
Flux Chamber (FC) Location
13 Hydraulic Conductivity Core (HCC) Location
t> Chemistry Core (CC) Location
Baseline Benthic Grab (BB) Location
s Benthic Grab (BG) Location
Sedflume Core (SF) Location
<&*
AQUABLOKCELL
AQB1
AB()2 AquaBlok 11 AQSE
loka* »AQSOT AQSE
3" "AQB2
AquaBlok
COKE BREEZE
CELL
CONTROL
CELL
Month 30 Sampling Locations,
Anacostia River, Washington, D.C
Figure 3-19. 30-Month Post-Capping Field Event Sampling/Monitoring Locations
-------
3.3.1.1 Objective #1 Results - Critical
Measurements
3.3.1.1.1 Sedflume Coring and Analysis.
Sedflume coring and analysis were conducted
during the six-month and 30-month post-capping
surveys, as described in Section 3.2.4. During
each sampling round, 12 total Sedflume cores
were collected (i.e., one per quadrant from the
AquaBlok®, sand, and control cells), as shown on
Figure 3-20, and analyzed on-site in a mobile
Sedflume laboratory.
Several Sedflume cores intended to target the
control cell were collected either outside the
determined boundary of the control cell or
immediately near the boundary. This occurred
presumably because of a misinterpretation of
navigational position by the coring contractor.
However, given that these cores were collected in
identical native sediment material as that within
the control cell, this does not in any way
compromise data usability or representativeness.
In addition, during the 30-month post-capping
Sedflume evaluation, two cores were collected
from the southwest quadrant of the sand cell and
none from the northeast quadrant. Given that
each core still captured the appropriate cap
interval, this also does not compromise the
Sedflume results.
The results of the six-month post-capping
Sedflume analyses indicated that control
sediments (i.e., native river bottom sediments) at
the surface were relatively easily eroded due to
the less consolidated nature of these sediments
and the presence of organic detritus and gas
voids. Erosion rates for the native sediments
decreased at depths approaching and below 10
cm where the native sediments were still
silty/clayey but were more compact and
competent. For the sand cell, the capping
material demonstrated greater erosion resistance
compared to the native sediments, but did exhibit
highly variable erosion rates due to the variable
presence of organic detritus and finer-grained
particles mixed with the sand. At depths
approaching and below 10 cm in the sand cell
Sedflume cores, material generally transitioned to
native sediment and erosion rates were
consistent with the native sediment cores at
equivalent depth. In the AquaBlok® cell Sedflume
cores, the sand covering layer demonstrated
erosion rates quite consistent with those
observed in the sand cell cores. However, in the
actual AquaBlok® material, erosion rates were
exceedingly low and required very high shear
energy to produce erosion. The shear stresses
required to erode the AquaBlok® material were
between 3.2 and 10 N/m2, a range that is
indicative of very high surface water energy at the
sediment/water interface. In addition, this range
is at least an order of magnitude higher than was
required to erode the native sediment interval and
significantly higher than energy required to erode
the sand capping material. Additional specific
data generated from the six-month post-capping
Sedflume survey, including general physical data
(e.g., bulk density and water content) generated
to verify sediment lithology and correlate with
erosion rates, is provided in Appendix E-1.
The results of the 30-month post-capping
Sedflume analyses were generally consistent
with the six-month Sedflume survey. Analyses
conducted on control sediments (i.e., native river
bottom sediments) indicated that the surface was
relatively easily eroded due to less consolidated
nature of the surface sediment and the presence
of organic detritus and gas voids. Erosion rates
for the native sediments decreased at depths
approaching and below 10 cm where the native
sediments were still silty/clayey but were more
compact and competent. For the sand cell, the
capping material demonstrated relatively low
resistance to erosion consistent with or even
lower than the surface of the native sediments.
This was presumably due to the accumulation of
fine-grained detrital sediment between the two
surveys atop the sand capping material. In
addition, sand in the sand cell Sedflume cores
appeared to have sorted to some degree
between surveys. In intervals characterized by
finer grained sands, erosion rates were relatively
higher than in coarser sand intervals. At depths
approaching and below 10 cm in the sand cell
Sedflume cores, material generally transitioned to
native sediment and erosion rates were
consistent with the native sediment cores at
equivalent depth. In the AquaBlok® cell Sedflume
cores, the sand layer demonstrated erosion rates
quite consistent with those observed in the sand
64
-------
O Sedflume Core (SF) Location (Month 6)
fi Sedflume Core (SF) Location (Month 30)
AQUABLOK CELL
SAND CELL
SF-1 O
APATI ECEI
COKE BREEZE
CELL
CONTROL
CELL
Sedflume Core SamplingLocations
Anacostia River, Washington, D.C.
ANACOSTIA RIVER
Figure 3-20. Sedflume Coring Locations
-------
cell cores. However, in the actual AquaBlok®
material, as with the previous six-month post-
capping event, erosion rates were exceedingly
low and required very high shear energy to
produce erosion. The shear stresses required to
erode the AquaBlok® material were at least 3
N/m2 and up to 9 N/m2, which is indicative of very
high surface water energy at the sediment/water
interface. In addition, this shear stress is
approximately an order of magnitude higher than
the energy required to erode the native sediment
interval and significantly higher than the energy
required to erode the sand capping material.
Additional specific data generated from the 30-
month post-capping Sedflume survey, including
general physical data (e.g., bulk density and
water content) generated to verify sediment
lithology and correlate with erosion rates, is
provided in Appendix E-2.
Overall, the results of the Sedflume analyses
conducted during the demonstration indicate that
AquaBlok® is a highly competent and cohesive
material and is unlikely to be eroded even at very
high shear stresses consistent with very high
flow. The results also suggest that traditional
sand cap material (or a sand covering layer over
an AquaBlok® cap) can be less resistant to
erosion when compared to the fine-grained,
organic-rich sediments common in the Anacostia
River (and commonly found at most
contaminated sediment sites), and may be
variably resistant to erosion after layering through
grain sorting. Even where characterized by
erosion resistance greater than typically organic
silty/clayey river bottom sediments, the data
generated suggest sand would not be as
resistant to erosion when compared to
AquaBlok®.
With specific respect to the potential for cap
failure in the Anacostia River system, it appears
unlikely that either an AquaBlok® or a sand cap
would be characterized by such a risk in the
typically sluggish and depositional local
environment of the demonstration area.
Specifically, given that very high river flow events
associated with significant precipitation in the
Washington, DC area were documented in the
Anacostia River during the demonstration and
both the AquaBlok® (see Figure 3-21) and sand
caps remained stable (see Section 3.3.1.1.3), it
would appear that either would be effectively
stable in the range of surface water flow common
to this environment. Beyond the local
demonstration area environment in the Anacostia
River, other research suggests that bottom shear
stresses are not significant and that the likelihood
of sediment movement during even storm events
is not great (Roberts, 2004). Nevertheless,
based on the laboratory Sedflume data generated
during the SITE demonstration, AquaBlok® would
be anticipated to be more stable in higher ranges
of flow and accompanying bottom shear stress at
any given contaminated sediment site.
3.3.1.1.2 Sediment Coring and
Analysis of Contaminants of Concern.
Sediment coring and analysis of COCs was
conducted during the six-month, 18-month, and
30-month post-capping surveys, as described in
Section 3.2.4. During the six-month post-capping
field event, two individual cores were collected
from each of the four unique quadrants in the
AquaBlok®, sand, and control cells, for a total of
eight cores per cell and 24 total cores overall.
During the 18-month post-capping field event,
two individual cores were collected from two of
the four unique quadrants in the AquaBlok®,
sand, and control cells, for a total of four cores
per cell and 12 total cores overall. During the 30-
month post-capping field event, the same coring
approach as the six-month post-capping event
was followed (i.e., two individual cores from each
of the four unique quadrants in the AquaBlok®,
sand, and control cells, for a total of eight cores
per cell and 24 total cores overall). Figure 3-22
displays the sediment coring locations from each
of the sampling events. As indicated on Figure 3-
22, a few sediment cores intended to target the
control cell were collected either outside the
determined boundary of the control cell or
immediately near the boundary. This occurred
presumably because of a misinterpretation of
navigational position by the coring contractor.
However, given that these cores were collected in
identical native sediment material as that within
the control cell, this does not in any way
compromise data usability or representativeness.
66
-------
USGS 816-47609 POTOHRC RIVER RT MISCQNSIN HVE, HRSHINGTON, DC
6.Q ~~
5.0
u 4.0
0)
a
a)
3.9
H 1.8
g
0.0
-1.0
Referenced to Washington Mean Low Water (1.41 ft above Mean Sea Le
2004 2005
el).
Hov 01 Jan 01 Har 01 Hay Ol Jul 01 Sep 01
Referenced to NAVD 88, in feet.
6/11/04 7/11/04 8/11/04 9/11/04 10/11/04 11/11/04 12/11/04 1/11/05 2/11/05 3/11/05 4/11/05 5/11/05 6/11/05 7/11/05 8/11/05 9/11/05
Figure 3-21. Potomac River (top) and Anacostia River (bottom) River Flows During
Demonstration (flood event in top panel is highlighted in green in lower panel; from
United States Geologic Survey [USGS], 2006)
67
-------
Exilian alion
Q Chemistry Core (CC
^Chemistry Core (CC
H Chemistry Core (CC
S3 Hydraulic Conductiv
Hydraulic il unducin.'
Location (Month 6)
Location (Month 18)
Location (Month 30)
ty Core (HCC) Location (Month 18)
tyCore (HCC'i Location (Month 30)
oo
COKE BREEZE
CELL
sediment riyai auiic ^unaucuviiy ana
Chemistrjr Core Sampling Locations,
Figure 3-22. Sediment Coring Locations
-------
During each coring event, the sampling approach
was to collect the following sample intervals
(each sample interval was intended to be a 3 cm
vertical segment of material) depending on the
actual thickness of various lithologic intervals:
• AquaBlok® cell: three individual samples
from the overlying sand layer, one sample
from the interface of the overlying sand
layer and the AquaBlok® capping layer,
three individual samples within the
AquaBlok® capping layer, one sample from
the interface of AquaBlok® and native
sediments, and one sample from the upper
horizon of the native sediment unit, for a
total of nine unique samples per core.
• Sand cell: four individual samples from the
sand capping layer, one sample from the
interface of the sand capping layer and the
underlying native sediment, and one sample
from the upper horizon of the native
sediment unit, for a total of six individual
samples per core.
• Control cell: three individual samples from
the upper horizon of the native sediment
unit, for a total of three individual samples
per core.
During field processing of the cores, the lithology
of the cores was recorded as well as the depths
of analytical samples collected and the specific
analyses to be performed on each sample. This
information was used to generate detailed coring
logs for each sediment core or for at least one
representative core from the set of replicate cores
from each coring location. Coring logs are
provided in Appendix G.
As indicated in Section 3.2.4, all sediment core
samples were analyzed for six individual metals
and a full suite of PCB congeners and PAHs. In
addition, duplicate samples were collected as
appropriate and analyzed for either this same set
of parameters or a subset thereof. The data
generated from the analysis of the sediment
samples collected during the various sediment
coring events are summarized in tabular form for
all individual target analytes in Appendix H. For
duplicate samples, concentrations were
averaged. In addition, graphs in Appendix H
show the concentrations of metals, PCBs, and
PAHs throughout the vertical profile of each core
sampled during each coring event. These graphs
are grouped by contaminant class and within
each contaminant class by cell and quadrant.
For simplicity, PCB graphs show only the six
most commonly detected PCB congeners in the
demonstration area and the PAH graphs show
only the seven most commonly detected PAHs at
the Anacostia study site.
Figures 3-23 to 3-25 show the total
concentrations of PAHs detected throughout the
vertical profile in the composited core from each
quadrant in each cell. Figures 3-26 through 3-28
show the same for PCBs, and Figures 3-29
through 3-31 show the same for metals. For
PAHs and PCBs, the total displayed on these
graphs is the sum of all individual analytes, as
opposed to the limited set of those most
commonly detected in the demonstration area.
For metals, the total is the sum of the six
individual metals analyzed. The evaluation of
data trends in Figures 3-23 to 3-31 is provided
below by compound class. Note that, in general,
these figures demonstrate overall lower
concentrations of all COCs during the 18-month
post-capping event than the six or 30-month
events. This observation is not readily explained,
and may be related to actual differences in COC
concentrations at the variable locations sampled
relative to the other monitoring events or
attributable to simple laboratory variability.
PAHs
As demonstrated on Figure 3-23, total PAH
concentrations in the surficial native sediments in
the control cell during the six-month and 30-
month post-capping coring events were generally
between 20,000 and 40,000 ug/kg and declined
to some extent with depth. Total surficial PAH
concentrations during the 18-month post-capping
coring effort were comparatively lower, which
may have been an artifact of increased
deposition of relatively disproportionately
inorganic new sediment during this timeframe
(see Sections 3.3.1.1.3 and 3.3.1.2.3).
Alternatively, the variability in surficial total PAH
concentrations could be related to simple
laboratory analytical variability and/or the varying
69
-------
Native Sediment Coring Time Series for PAHs
•a
O)
=£
O)
<
a.
50,000
45,000
40,000
35,0004^
30,000 4-
25,000 4-
20,000
15,000
10,000 -
5,000 4-
«s '^-^U.
O $ 03 ^
* ///
D Top Native Sediment
(0-3 cm)
• Native Sediment
(3-6 cm)
D Native Sediment
(6-9 cm)
Figure 3-23. Total PAHs in Control Cell Cores
AquaBlok Coring Time Series for PAHs
t
"3)
DTop Sand
• Middle Sand
D Bottom Sand
pSand/AquaBlok
Interlace
• Top AquaBlok
D Middle AquaBlok
• Bottom AquaBlok
DAquaBlok/Native
Sediment Interface
• Top Native Sediment
Figure 3-24. Total PAHs in AquaBlok® Cell Cores
70
-------
Sand Coring Time Series for PAHs
•o
O)
"3)
(/>
50,000
45,000
40,000
35,000
30,000
25,000
20,000
15,0004-
10,000
5,000
0
*/-**
«0
0-<0°
DTop Sand
• Middle Sand
D Middle Sand
D Bottom Sand
• Sand/Native Sediment
Interlace
DTop Native Sediment
Figure 3-25. Total PAHs in Sand Cell Cores
Native Sediment Coring Time Series for PCBs
D Top Native Sediment
(0-3 cm)
• Native Sediment
(3-6 cm)
D Native Sediment
(6-9 cm)
Figure 3-26. Total PCBs in Control Cell Cores
71
-------
AquaBlok Coring Time Series for PCBs
~ 3,000-k
a 2,500 V
o
D Top Sand
• Middle Sand
D Bottom Sand
D Sand/AquaBlok
Interface
• Top AquaBlok
D Middle AquaBlok
• Bottom AquaBlok
DAquaBlok/Native
Sediment Interface
• Top Native Sediment
Figure 3-27. Total PCBs in AquaBlok® Cell Cores
Sand Coring Time Series for PCBs
n Top Sand
• Middle Sand
D Middle Sand
D Bottom Sand
• Sand/Native
Sediment Interface
D Top Native Sediment
Figure 3-28. Total PCBs in Sand Cell Cores
72
-------
Native Sediment Coring Time Series for Metals
1,600
f///
**//
D Top Native Sediment
(0-4 cm)
• Native Sediment
(3-8 cm)
D Native Sediment
(6-12 cm)
Figure 3-29. Total Metals in Control Cell Cores
AquaBlok Coring Time Series for Metals
_
•o
O)
1
(/)
8
0)
5
o
t—
1,600-r
1,400J
1,200-1
1,000-
800
600
400
200
0
*///
OTop Sand
• Middle Sand
O Bottom Sand
DSand/AquaBlok
Interlace
• Top AquaBlok
O Middle AquaBlok
• Bottom AquaBlok
pAquaBlok/Native
Sediment Interface
• Top Native Sediment
Figure 3-30. Total Metals in AquaBlok® Cell Cores
73
-------
Sand Coring Time Series for Metals
1,600
DTop Sand
• Middle Sand
D Middle Sand
D Bottom Sand
• Sand/Native Sediment
Interlace
DTop Native Sediment
^^ _, ^^~T~~-^;
*,
***v
Figure 3-31. Total Metals in Sand Cell Cores
core locations. Higher total PAH concentrations
at the surface in the control cell could be related
to the fact that PAHs are typically a component of
urban runoff and general urban pollution due to
their presence in urban fill and other products
(e.g., asphalt). Given that the demonstration
area is located in a densely urbanized area and is
immediately near a CSO, ongoing deposition of
PAHs would not be unexpected. Conversely,
declining total PAH concentrations with depth
could be the result of continuous but limited
vertical biogenic mixing combined potentially with
active PAH biodegradation in the shallow
subsurface sediment horizon.
Figure 3-24 shows that in the AquaBlok® cell,
total PAH concentrations in the upper intervals of
the native sediment were consistent with the
control cell. Given the AquaBlok® cap was
covering this cell throughout the demonstration
and the variability in surficial native sediment total
PAH concentrations is generally uniform between
the various coring events in the AquaBlok® and
control cells, it would appear this variability (for
both the AquaBlok® and control cell) is related to
core location and/or laboratory analytical
variability rather than varying degrees of ongoing
contamination from urban sources. In the
interface zone between native sediment and
AquaBlok®, the sample was a physical mixture of
AquaBlok® and native sediments. The
concentration of total PAHs in this interval
generally declined compared to the uppermost
native sediment interval, and was likely driven by
the presence of PAHs in native sediment and the
diluting effect of the AquaBlok®. PAHs were
generally absent throughout the AquaBlok® cap
material interval and over all sampling events,
with the exception of a low total PAH
concentration observed in the bottom AquaBlok®
interval (i.e., the bottom-most sample of purely
AquaBlok® material above the AquaBlok®/native
sediment interface) in one quadrant (i.e.,
Quadrant 1) during the 30-month post-capping
event. This could be indicative of some limited
movement of PAHs upward into the AquaBlok®
capping material, but is very limited in magnitude
both empirically (i.e., a low total PAH
concentration) and physically (i.e., only observed
in one quadrant out of four). Interestingly, over
74
-------
the course of the demonstration, PAHs appear to
have accumulated to some extent on the surface
of the sand covering layer in the AquaBlok® cell,
which is consistent with the nature of PAHs as a
likely component of urban pollution and the
observed accumulation of new sediment over the
course of the demonstration (see Sections
3.3.1.2.1 and 3.3.1.2.3). In addition, there
appears to have been some limited degree of
downward vertical mixing of this PAH
contamination into intermediate levels of the sand
covering layer, potentially via bioturbation.
Figure 3-25 demonstrates that in the sand cell,
total PAH concentration trends were generally
similar to the trends observed in the AquaBlok®
cell. Specifically, total PAH concentrations in the
native sediment interval beneath the sand
capping layer were consistent with the control cell
and AquaBlok® cell results. Also, in the interface
interval between sand cap and native sediment,
the sample was a physical mixture of sand and
native sediments. The concentration of total
PAHs in this interval generally declined compared
to the uppermost native sediment interval, and
was likely driven by the presence of PAHs in
native sediment and the diluting effect of the
sand. In addition, over the course of the
demonstration, PAHs appear to have
accumulated to some extent on the surface of the
sand cap and vertically mixed downward to some
limited extent into more intermediate levels of the
sand cap, potentially through bioturbation. The
most substantial difference between the sand cap
and the AquaBlok® cap appears to be at the base
of the capping intervals. Detectable total PAH
concentrations were observed at the base of the
sand cap (i.e., in the bottom-most sample of
purely sand above the sand/native sediment
interface zone) during both the 18-month and 30-
month post-capping events. In addition, these
total PAH concentrations were consistently
detected across the sand cell quadrants, were
generally higher than the single total PAH
concentration detected in the single quadrant
(i.e., Quadrant 1) during the 30-month post-
capping event in the AquaBlok® cell, and
demonstrated a potentially increasing trend in
extent and concentration between the 18-month
and 30-month post-capping events. This could
be indicative of a greater degree of PAH mobility
upward into sand as compared to AquaBlok®.
However, given the generally low total PAH
concentrations in the basal cap intervals and the
uncertainty surrounding whether some native
sediment material could have been entrained in
any particular sample of sand cap material, it is
not possible to conclusively determine if this
represents a truly varying pattern in contaminant
flux between sand and AquaBlok®. Moreover,
specific statistical testing was not conducted to
evaluate this potential difference.
PCBs
As demonstrated on Figure 3-26, total PCB
concentrations in the surficial native sediments in
the control cell during the six-month and 30-
month post-capping coring events were generally
between 500 and 1,500 ug/kg and increased to
some extent with depth to levels approaching
3,000 to 4,000 ug/kg. Total surficial and
subsurface PCB concentrations during the 18-
month post-capping coring effort were
comparatively lower than those observed in the
six and 30-month events, which may have been
an artifact of increased deposition of relatively
disproportionately inorganic new sediment during
this timeframe (see Sections 3.3.1.1.3 and
3.3.1.2.3) that could have diluted the total PCB
level. Alternatively, the variability in total PCB
concentrations could be related to simple
laboratory variability and/or the varying core
locations. Higher total PCB concentrations at
depth in the control cell are likely related to the
fact that PCBs are unlikely to have any
appreciable ongoing source, and therefore newly
deposited sediment is likely to result in lower
PCB levels at shallower depths. Given the total
PAH results described above for the control cell
cores, it appears likely the PCB variability during
the 18-month period is also associated with
coring locations and/or general laboratory
analytical variability.
Figure 3-27 depicts total PCB trends in the
AquaBlok® cell that are highly similar to the total
PAH trends discussed above. PCBs were
generally absent throughout the AquaBlok® cap
material interval and over all sampling events. A
very low total PCB concentration was observed in
the bottom AquaBlok® interval (i.e., the bottom-
75
-------
most sample of purely AquaBlok® material above
the AquaBlok®/native sediment interface) in one
quadrant (i.e., Quadrant 1) during the 30-month
post-capping event. This could be indicative of
some limited movement of PCBs upward into the
AquaBlok® capping material, but is very limited in
magnitude both empirically (i.e., a very low total
PCB concentration) and physically (i.e., only
observed in one quadrant out of four). Over the
course of the demonstration, PCBs appear to
have accumulated to some limited extent on the
surface of the sand covering layer. Given the fact
that PCBs do not likely have an appreciable
ongoing source, this observation is likely related
to PCBs in suspended native sediments (either
from ongoing sediment transport dynamics or
initial sediment suspension during capping, or
both) being deposited on the capping cell.
Alternatively, given the highly urbanized and
industrialized nature of this portion of
Washington, DC, it is conceivable that there
could be some ongoing contribution of PCBs to
the Anacostia River in diffuse urban runoff.
Figure 3-28 depicts total PCB trends in the sand
cell that are also highly similar to the total PAH
trends discussed above. As with the AquaBlok®
cell, over the course of the demonstration, PCBs
appear to have accumulated to some extent on
the surface of the sand cap, likely for the same
reason(s) as described above for the AquaBlok®
cell. As with PAHs, the most substantial
difference between the sand cap and the
AquaBlok® cap relative to PCBs was at the base
of the capping intervals. Detectable total PCB
concentrations were observed at the base of the
sand cap (i.e., in the bottom-most sample of
purely sand above the sand/native sediment
interface zone) during both the 18-month and 30-
month post-capping events. In addition, these
total PCB concentrations were detected in the
sand cell quadrants more consistently, were
generally higher than the single total PCB
concentration detected in the single quadrant
(i.e., Quadrant 1) during the 30-month post-
capping event in the AquaBlok® cell, and
appeared to demonstrate an increasing
concentration trend between events. This could
be indicative of a greater degree of PCB mobility
upward into sand as compared to AquaBlok®.
However, given the low total PCB concentrations
in the basal cap intervals and the uncertainty
surrounding whether some native sediment
material could have been entrained in any
particular sand cap material sample, it is not
possible to conclusively determine if this
represents a truly varying pattern in contaminant
flux between sand and AquaBlok®. Moreover, as
with PAHs, this potential difference in PCB
concentrations was not evaluated specifically
through statistical testing.
Metals
While the evaluation of metals using the
summation of component elements is not as
meaningful as the summation of total PAHs or
PCBs, it is informative and useful for illustrative
purposes. As indicated on Figure 3-29, total
metals concentrations in the control cell were
generally between 600 and 1,000 mg/kg, and
were generally uniform in lateral distribution and
vertically throughout the upper 9 cm of native
sediments.
Figure 3-30 shows that in the AquaBlok® cell,
total metal concentrations in the upper intervals
of the native sediment were consistent with the
control cell. In the interface zone between native
sediment and AquaBlok®, the sample was a
physical mixture of AquaBlok® and native
sediments. The concentration of total metals in
this interval generally declined compared to the
uppermost native sediment interval, and was
likely driven by the presence of metals in native
sediment and the diluting affect of the AquaBlok®.
Unlike PAHs and PCBs, metals were generally
present at low concentrations throughout the
AquaBlok® cap material interval and over all
sampling events at similar concentrations. This
could be related to the nature of the AquaBlok®
material itself. Being comprised of a natural earth
material (i.e., bentonite clay) that itself typically
contains detectable levels of some metallic
constituents, it is not surprising that metals could
be detected in AquaBlok®. Alternatively,
bentonite clay is characterized by a high metal
exchange capacity and it is therefore possible
that the AquaBlok® cap strongly removes/sorbs
metals from the underlying contaminated
sediment. Over the course of the demonstration,
metals appear to have accumulated to some
76
-------
extent on the surface of the sand covering layer
in the AquaBlok®cell, which is consistent with the
nature of metals as a likely component of urban
pollution and the observed accumulation of new
sediment over the course of the demonstration
(see Sections 3.3.1.2.1 and 3.3.1.2.3). In
addition, there appears to have been some
limited degree of downward vertical mixing of this
metals contamination into intermediate levels of
the sand covering layer, potentially via
bioturbation.
Figure 3-31 demonstrates that in the sand cell,
total metals concentration trends were similar to
the trends observed in the AquaBlok® cell in
some respects. Total metals concentrations in
the native sediment interval beneath the sand
capping layer were consistent with the control cell
and AquaBlok® cell results. Also, in the interface
interval between sand cap and native sediment,
the sample was a physical mixture of sand and
native sediments. The concentration of total
metals in this interval generally declined
compared to the uppermost native sediment
interval, and was likely driven by the presence of
metals in native sediment and the diluting affect
of the sand. In addition, over the course of the
demonstration, metals appear to have
accumulated to some extent on the surface of the
sand cap and vertically mixed downward to some
limited extent into intermediate levels of the sand
cap, potentially via bioturbation. Detectable total
metal concentrations were observed at the base
of the sand cap (i.e., in the bottom-most sample
of purely sand above the sand/native sediment
interface zone) during all sampling events,
demonstrating a potentially increasing trend in
extent and concentration throughout the course
of the demonstration. This could be indicative of
metals mobility upward into the sand. However,
given the low total metals concentrations in the
basal sand cap interval and the uncertainty
surrounding whether some native sediment
material could have been entrained in any
particular sand sample, it is not possible to
conclusively determine if this is indicative of metal
flux into the sand cap. Moreover, as with PAHs
and PCBs, this potential trend was not evaluated
through specific statistical testing.
Overall, the sediment coring and COC analyses
conducted during the demonstration suggest that
the sand cap and AquaBlok® cap have remained
both physically stable (i.e., observations of the
cores indicated no appreciable changes in
lithology from event to event) and have been
effective at preventing the upward movement of
contamination. There appears to have been
some ongoing contribution of PAHs, PCBs, and
metals at the sediment surface evidenced by the
detection of these compounds in the uppermost
core intervals. PAHs and metals are likely
present in diffuse urban pollution emanating from
Washington, DC, and PCBs could also be a
component of ongoing urban pollution given the
highly urbanized and industrialized nature of the
region. In addition, these contaminants could be
present in suspended sediment released from
areas not capped through natural sediment
transport in the Anacostia River system and/or
from sediment resuspended during capping and
subsequently redeposited in the demonstration
area. No specific sampling or monitoring was
conducted during the demonstration to determine
the origin of contaminants at the surface. While
there did appear to have been at least some
downward mixing of contaminants from the
surface, potentially though bioturbation, there did
not appear to be a significant degree of vertical
mixing of contamination from the surface
downward.
The available data suggest that there may be
some increased movement of contaminants from
native sediments upward into the sand cap as
compared to the AquaBlok® cap, but neither cap
demonstrated significant accumulation of
contamination or contaminant breakthrough.
Specifically, PCBs and PAHs were detectable in
the lowest intervals of the sand cap in the sand
cell more frequently and at higher concentrations
than in the AquaBlok® material in the AquaBlok®
cell, and also appeared to show an increasing
concentration trend throughout the demonstration
that was not observed in the AquaBlok® data.
Metals data were more difficult to evaluate given
that common earth materials, and specifically the
clay material used to create AquaBlok®, could
contain metals at varying concentrations and are
characterized by a high exchange capacity that
could lead to ready binding of metals from native
77
-------
sediments. While the sand cap appeared to
demonstrate a similar mobility trend for metals as
compared to PAHs and PCBs, the AquaBlok® cell
demonstrated either an even greater mobility
trend for metals (i.e., related to strong metals
uptake/sorption characteristics) or simply the
signature of metals that are a part of typical
bentonite clays. The evaluation of the chemical
data is complicated by the fact that initial
contaminant displacement/movement and
potential mixing during cap construction was not
specifically studied, and could be a strong
influence on the observed contaminant
concentrations in the capping intervals in the
sand and AquaBlok® cells. Moreover, in general,
the temporal dataset generated through the
demonstration is not sufficient to allow for a
complete determination of the potential mobility of
contamination in either capping cell or the
specific differences between contaminant
migration in the capping cells or between discrete
contaminant classes within and across cells.
3.3.1.1.3 Bathymetry and Sub-Bottom
Profiling. Bathymetric and sub-bottom profiling
surveys were conducted during the one-month,
six-month, 18-month, and 30-month post-capping
surveys, as described above in Section 3.2.4.
Each survey was conducted by traversing the
identical series of 29 survey transects oriented
parallel to the shore (see Figure 3-32). Accurate
positional control was achieved by operating the
survey vessel in a very controlled fashion and by
using a dGPS linked to accurate navigational
software.
The primary objectives of the bathymetric and
sub-bottom profiling were to determine the overall
thickness of capping material in the AquaBlok®
and sand cells as well as the thickness of various
layers in these cells where relevant (i.e.,
AquaBlok® versus overlying sand in the
AquaBlok® cell) and changes in these
thicknesses over time. These measures in turn
were intended to describe the in-place stability of
AquaBlok® relative to sand capping material and
native sediments in a real world flow regime.
During each of the sub-bottom surveys
conducted, the "chirp" profiler was unable to
resolve sub-bottom stratigraphy. This was likely
related to the presence of biogenic gases in the
organic and decompositional environment
characteristic of the site. Such gases typically
represent a barrier to acoustic signal propagation
and prevent the sub-bottom profiling equipment
from penetrating below the interval of gas
production. As the interval of gas production in
the Anacostia River bottom sediments is surficial,
the "chirp" profiler was unable to penetrate even
beyond the very shallow sediment horizon.
Frequently, this condition manifests itself as a
sub-bottom profiler signal return that contains the
sediment surface as a distinct feature followed by
"echoes" of the surface rather than true vertical
lithologic contacts. This data pattern was
observed for all SITE demonstration sub-bottom
surveys. As such, the sub-bottom surveying
component of the SITE demonstration program
was unsuccessful in providing meaningful
information to assess the stability of AquaBlok®
other than by providing a measure of sediment
surface topography redundant with other
measurement tools.
During the one-month post-capping bathymetric
survey, water depths in the demonstration area
ranged from approximately 4.5 ft nearer shore to
approximately 19.5 ft nearer the river channel.
The riverbottom sediment surface exhibited a
northwest to southeast trending slope from the
shoreline towards the river channel at an average
4% grade. To derive the overall thickness of the
cap in each cell from these bathymetric data, the
survey data from a pre-capping survey (note a
pre-capping bathymetric survey was conducted
that was not part of the SITE demonstration
summarized in this ITER) was subtracted from
the one-month post-capping survey data, yielding
essentially a total cap thickness (i.e., all material
placed on the native sediment surface during cap
construction). Given the spacing of the
bathymetric survey transects, this cap thickness
information could be plotted in three dimensions
over the entire plan area of the demonstration
area. The one-month post-capping bathymetric
data indicated that the cap thickness in the sand
cell was 0.25 ft or less around the perimeter of
the cell to a maximum of 1.25 ft in the southwest
corner of the cell. In the AquaBlok® cell, total cap
thickness was 0.25 ft or less in the southernmost
portion of the cell to a maximum of 1.75 ft in the
78
-------
northeastern corner of the cell. These results
generally confirm the information recorded during
cap placement, including the generally thicker
placement of AquaBlok® where capping in this
cell was initiated (i.e., the northeast cell corner)
and the shortage of material to cap the
southwestern cell corner (see Section 3.1.1 and
Figure 3-4). These results also generally
corroborate that design cap thicknesses were
achieved. Figure 3-33 shows the three-
Washington DC
N
100ft
Figure 3-32. Survey Transects in Demonstration Area for Oceanographic Surveying
(Side-Scan Sonar Transects Bolded)
dimensional cap thickness map of the
demonstration area derived from the one-month
post-capping bathymetry data. Other specific
data output from the one-month post-capping
survey is provided in Appendix B-1.
The six-month post-capping bathymetric data
indicated highly consistent water depths (i.e.,
generally between 4 and 20 ft) and the same river
bottom slope (i.e., 4%) as the one-month post-
capping survey. Total cap thickness was derived
by subtracting the baseline bathymetric data from
the six-month post-capping data, and indicated
generally identical total thickness across the
demonstration area as compared to the one-
month post-capping survey. In addition, the one-
month post-capping survey data were subtracted
from the six-month post-capping survey data to
determine the net change in total cap thickness
between these events. This difference operation
indicated that between the one-month and six-
month post-capping surveys there was very little
net change in the total cap thickness in both the
AquaBlok® and sand cells. This net change was
at most +/- 0.25 ft, which is roughly equivalent to
the accuracy of the bathymetric equipment. A
plot specifically demonstrating the total cap
thickness difference between the one-month and
six-month post-capping surveys, as well as other
specific data output relevant to the six-month
post-capping survey, is provided in Appendix B-2.
Similar assessments were completed using the
18-month and 30-month post-capping bathymetry
datasets. Water depths and sediment surface
slopes during these surveys were both highly
consistent with the one-month and six-month
post-capping surveys (i.e., water depths between
79
-------
4 and 18 ft during the 18-month survey and
between 6 and 16 ft during the 30-month survey
and an average grade of approximately 4% for
each).
For the 18-month post-capping bathymetric
dataset, a total cap thickness plot was developed
by subtracting the baseline dataset. The total
cap thickness results from this operation
indicated thicknesses highly similar to the
previous surveys. In addition, a difference plot
was created by subtracting the one-month post-
capping dataset from the 18-month post-capping
dataset, effectively producing a representation of
total cap thickness change since installation.
This assessment indicated a net increase of
material generally throughout the demonstration
area. This net increase in cap material was
generally of limited magnitude (i.e., 0.25 ft or
less) and extent, and generally was not observed
in the AquaBlok® and sand cells where cap
thickness appeared highly consistent with
previous surveys. The net increase in cap
thickness is presumably related to high flow
events that occurred in the Anacostia River
between the six-month and 18-month surveys,
most notably in the spring of 2005 when several
major storm events occurred in the site area.
These high flow events could presumably have
transported material into the demonstration area,
which is characteristically a depositional
environment. However, given the inherent
accuracy of bathymetric survey equipment and
the fact that even storm events in the Anacostia
River are not likely to mobilize sediment to any
significant degree (see Section 3.3.1.1.1;
Roberts, 2004), this observed overall net
increase may have not been real but a simple
artifact of the data reduction. Appendix B-3
provides all of the specific data output from the
18-month post-capping survey, including the
difference evaluations described above and other
comparisons between survey rounds.
For the 30-month post-capping bathymetric
dataset, a total cap thickness plot was developed
by subtracting the baseline dataset as with all
other surveys. The total cap thickness results
from this operation indicated thicknesses highly
similar to the previous surveys. Figure 3-34
shows the three-dimensional cap thickness map
of the demonstration area derived from the 30-
month post-capping bathymetry data. In addition,
a difference plot was created by subtracting the
one-month post-capping dataset from the 30-
month post-capping dataset, effectively producing
a representation of total cap thickness change
since installation. This assessment indicated that
between the one-month and 30-month post-
capping surveys there was very little net change
in the total cap thickness in both the AquaBlok®
and sand cells. This net change was generally
+/- 0.25 ft, which is roughly equivalent to the
accuracy of the bathymetric equipment.
Appendix B-4 provides all of the specific data
output from the 30-month post-capping survey,
including the difference evaluations described
above and other comparisons between survey
rounds.
Overall, the bathymetric data generated during
the SITE demonstration indicate that the
AquaBlok® cap and the sand cap are highly
stable in the demonstration area. However,
these data do not directly describe the AquaBlok®
material itself as the sand surface layer installed
over the AquaBlok® was itself highly stable.
Therefore, the AquaBlok® was not directly
exposed to flow at the sediment/water interface.
In addition, the traditional sand cap was similarly
stable as compared to the sand covering the
AquaBlok® material, meaning that a comparative
evaluation of stability between sand and
AquaBlok® is not possible on the basis of
bathymetric data alone. Moreover, even with
high flow conditions linked to significant storm
events documented in the flow record for the
Anacostia River, the sand covering the
AquaBlok® cap and the traditional sand cap
remained relatively unchanged throughout the
SITE demonstration. This interpretation is,
however, complicated by the inherently limited
resolution of the data collection tools and the fact
that the demonstration area is in a
characteristically depositional environment and
may not itself have been exposed to any
significant degree to the increased energy of high
flow events. Nevertheless, given that the
AquaBlok® cap design for the SITE
demonstration is relatively standard, the
bathymetric data collected support that this
approach yields a stable cap.
80
-------
N
Figure 3-33. One-Month Post-Capping Bathymetric Cap Thickness Map
30-MONTH SURVEY
Figure 3-34. 30-Month Post-Capping Bathymetric Cap Thickness Map
81
-------
3.3.1.1.4 Side-Scan Sonar Surveying.
Side-scan sonar surveys were conducted during
the one-month, 18-month, and 30-month post-
capping surveys, as described above in Section
3.2.4. Each side-scan sonar survey was
completed by traversing the identical series of 10
survey transects oriented parallel to the shore
(see Figure 3-32). These 10 transects were a
subset of the 29 transects used for the
bathymetric surveying. Accurate positional
control was achieved by operating the survey
vessel in a highly controlled fashion and by using
a dGPS linked to accurate navigational software.
The primary objectives of the side-scan sonar
surveys were to determine the general surface
characteristics of the AquaBlok® and sand caps
and the native sediment control cell, and to
support the conclusions derived from the
bathymetric surveying.
Side-scan sonar surveys provide a plan view
image analogous to a high-angle aerial
photograph. The one-month post-capping side-
scan sonar survey demonstrated relatively dark
signal returns characteristic of generally coarse
grained material (i.e., sand) in both the
AquaBlok® and sand cell, and lighter signal
returns characteristic of fine grained material (i.e.,
silt and clay) in the control cell. Overall, the side-
scan sonar image was highly consistent with the
three-dimensional representation of cap
thickness provided by the bathymetric data,
showing the same irregular surface topography.
Figure 3-35 is the side-scan sonar mosaic from
the one-month post-capping survey. Appendix B-
1 provides this same mosaic and additional detail
related to the one-month post-capping side-scan
sonar surveying.
The 18-month post-capping side-scan sonar
survey was generally consistent with the one-
month post-capping survey. Overall, the
AquaBlok® and sand cells were characterized by
darker signal returns indicative of the sand
surface layer in both cells, while the control cell
was characterized by lighter signal returns
indicative of the silty/clayey native sediment
material. The surface topography evident in the
18-month post-capping survey was irregular,
consistent with bathymetric data and the one-
month post-capping sonar survey. Several
objects were identified in the 18-month post-
capping side-scan sonar survey lying on the
sediment surface in the demonstration area.
These objects were presumably debris items
such as tree branches or logs that might have
been deposited as a result of the storm events
documented in the area prior to the 18-month
post-capping field activities. In addition, the side-
scan signal returns over much of the
demonstration area during the 18-month post-
capping survey were slightly darker in nature
compared to the one-month post-capping survey.
This may be indicative of the deposition of a
surface layer of differing texture leading up to the
18-month post-capping survey. Both findings are
generally consistent with the bathymetric survey
data gathered during the 18-month post-capping
evaluation. Appendix B-3 provides a side-scan
sonar mosaic and additional detail related to the
18-month post-capping side-scan sonar
surveying.
The 30-month post-capping side-scan sonar
survey showed generally light returns across the
demonstration area, inconsistent with the return
pattern from the one-month and 18-month post-
capping surveys. In addition, the surface
topography throughout the demonstration area
did not show the irregular characteristic observed
in the previous two surveys but was rather
generally flat. The light sonar returns and flat
surface appearance may be related to the
accumulation of a thin layer of silty/clayey detrital
sediment following the 18-month post-capping
survey, or could be related to the sonar
apparatus being run at an inappropriate setting.
Surface objects consistent with the presumed
debris items observed in the 18-month post-
capping survey were also observed in the 30-
month post-capping side-scan sonar data. Figure
3-36 is the side-scan sonar mosaic from the 30-
month post-capping survey. Appendix B-4
provides this same side-scan sonar mosaic and
additional detail related to the 30-month post-
capping side-scan sonar surveying.
Overall, the results of the side-scan sonar
surveying conducted during the SITE
demonstration program corroborate the results of
the bathymetric surveying summarized in Section
82
-------
Figure 3-35. One-Month Post-Capping Side-Scan Sonar Map
Washington
D.C.
Edge oif-mat
Figure 3-36. 30-Month Post-Capping Side-Scan Sonar Map
83
-------
Explanation
® Sediment Profile Imaging (5PI) Location (Month 1)
O Sedjment Profile Imaging (SPl) Location (Month 6)
Sediment Profile Imaging (SPl) Location (Month 18)
£» Sediment Profile Imaging (SPl) Location (Month 3D)
oo
COKE BREEZE
CELL
Battede
SCALEI^EET
0 26 50 75 100
S ediment Prof ile Ima ging S amp ling Locations,
Anacostia River, Washington, D.C.
ANACOSTTA RIVER
G482024 I 200aALL3AMPL.OC3.CDR I <12M7~
Figure 3-37. Sediment Profile Imaging Monitoring Locations
-------
but do not provide any specific and unique
information related to cap stability in the
AquaBlok® or sand cells that was not gleaned
from the bathymetry.
3.3.1.2 Objective #1 Results - Non-
Critical Measurements
3.3.1.2.1 Sediment Profile Imaging. SPI
surveys were conducted during the one-month,
six-month, 18-month, and 30-month post-capping
surveys, as described above in Section 3.2.4.
During each survey, a total of 12 locations were
evaluated using SPI in the AquaBlok®, sand, and
control cells. Specifically, nine locations were
assessed in each cell using the video SPI
camera, and three locations in each cell were
evaluated using the standard SPI camera. In
addition, reference stations outside the control
cell and either nearer or in the Anacostia River
navigation channel were also assessed to
provide further reference information for
comparisons to the control and capped cells.
Between the various surveys, individual SPI
locations were generally replicated with a
reasonable lateral offset to provide the most
meaningful data comparisons between the
sampling events. Figure 3-37 shows the SPI
locations assessed during each sampling event.
Accurate positional control of the SPI drops
during each sampling event was achieved by
operating the survey vessel in a very controlled
fashion and by using a dGPS linked to accurate
navigational software.
The SPI surveying provided several important
results in the context of evaluating objective #1.
The thickness of the sand layer over the
AquaBlok® capping material remained generally
consistent throughout the multiple surveys based
on SPI attempts that achieved significant
penetration, as did generally the grain size
observed in the SPI images for the AquaBlok®
cell. Similarly, the thickness of the sand cap in
the sand cell and the grain size of this material
remained generally consistent throughout the
multiple surveys. In addition, the thicknesses
observed of the various layers in the various cells
were generally consistent with the design
thicknesses and thicknesses derived from other
measurement tools (e.g., bathymetric surveying).
The nature of the control sediments remained
generally consistent throughout the
demonstration, and the reference sediments
outside the control cell towards the navigation
channel were highly consistent throughout.
During the final SPI survey (i.e., 30-month post-
capping), there appeared to be evidence that the
surface sediments in both the AquaBlok® and
sand cell had accumulated a greater proportion of
fine-grained material, indicative of deposition of
detritus and fine sediment. Also during the 30-
month post-capping SPI survey, the control
sediments appeared to demonstrate a change in
surface texture, actually appearing to be more
coarse-grained. This could have been related to
new sediment deposition, but could also have
been related to the movement of some sand
capping/covering material from the demonstration
area towards the navigation channel.
As indicated in Figure 3-38, depths of camera
penetration were generally greatest in the control
cell, and generally greater in the AquaBlok® cell
compared to the sand cell. While it is intuitive
that camera penetration depth would be greatest
in the uncapped control cell, the reason for
greater penetration rates in the AquaBlok® cell
(which was covered with sand) relative to the
sand cell is not readily explained. Overall
penetration depths in the AquaBlok®, sand, and
control cells generally declined throughout the
course of the SPI surveys (see Figure 3-38),
potentially indicative of grain sorting and
"cementation" that would tend to inhibit physical
penetration. Variations in penetration could also
potentially be an artifact of even minor variations
in the SPI equipment and/or equipment operation
(e.g., specific manual efforts or camera
weighting). The presence of and frequency of
observation of biogenic and purely physical
features in the surface and subsurface sediments
throughout the demonstration area were
generally highly consistent between the multiple
SPI surveys. Such features were dominated by
gas voids, but there were a limited number of
biogenic structures (e.g., infauna tubes) also
observed in the various surveys. There were no
readily apparent differences in the presence of or
frequency of observation of these features
85
-------
70
60
50
40
30
20
10
n AquaBlok
nSand
n Control
in
-10
Month 01
Month 06
Month 18
Month 30
Figure 3-38. Video SPI Camera Penetration Trend (columns represent the mean, n
is the population size, and error bars represent 95% upper and lower confidence
intervals around the mean)
between the AquaBlok® and sand cells from a
purely observational perspective.
Overall, the results of the SPI surveying
conducted during the SITE demonstration
indicate that both the sand and AquaBlok® caps
remained intact and were therefore stable. In
addition, the generally highly consistent grain size
of the sediments in these cells throughout the
multiple rounds of evaluation indicates that there
was not a significant amount of bioturbation or
other surface mixing that could have impacted
cap integrity. While it appears that grain sorting
could potentially have been responsible for
declining penetration rates overtime (i.e., through
a cementing effect), this sorting does not appear
to have been related to cap material loss or
significantly obvious bioturbation. In fact, the SPI
monitoring appears to demonstrate that fine
detrital sediment accumulation occurred over the
duration of the demonstration.
Appendix C provides additional detail related to
the SPI surveys conducted during the AquaBlok®
SITE demonstration as they relate to objective
#1.
3.3.1.2.2 Gas Flux Analysis. Gas flux
sampling was conducted during the 18-month
and 30-month post-capping surveys, as
described above in Section 3.2.4. During each
survey, two chambers each were deployed in the
AquaBlok®, sand, and control cells. Between the
two surveys, individual flux chamber locations
were generally replicated with a reasonable
lateral offset to provide the most meaningful data
comparisons between the sampling events.
Figure 3-39 shows the flux chamber locations
assessed during each sampling event. Accurate
positional control was maintained during flux
chamber deployment during each sampling event
by operating the diving vessel in a very controlled
fashion and by using a dGPS.
86
-------
Table 3-4 presents information about all of the
flux chambers that were deployed during the 18-
month and 30-month post-capping field events.
As indicated on this table, one flux chamber from
each deployment period was not recovered (i.e.,
one chamber from the AquaBlok® cell during the
18-month post-capping event and one chamber
from the sand cell during the 30-month post-
capping event). These chambers may have been
lost due to storm events. The fact that the
missing AquaBlok® chamber during the 18-month
post-capping event was found ashore supports
this hypothesis. In addition, one flux chamber
deployed in the AquaBlok® cell during the 30-
month post-capping event was observed to be
seated at an angle at the end of the month-long
deployment period and the gas that was drawn
from the chamber was only of very limited volume
insufficient for laboratory analysis. Leaks were
observed coming from two flux chambers at
retrieval during the 18-month post-capping field
event. One control cell chamber appeared to
have a bad weld on the flux chamber dome that
allowed gas to escape from the chamber.
Although gas was recovered from this chamber
and the scoped laboratory analyses were
performed, the flux of gas into this chamber could
not be accurately determined given this condition.
In addition, some gas may have been lost during
retrieval from one sand cell chamber while pulling
the sample volume by syringe. Bubbles were
observed coming from the syringe gasket during
the draw-off of gas. The gasket was tightened
after the leak was observed and the total losses
were likely minimal.
The gases drawn from each of the flux chambers
were analyzed for oxygen, nitrogen, methane,
carbon dioxide, TNMOC, and 20 reduced sulfur
compounds (see Table 3-5). In all cases, at least
98% of the gas by volume consisted of nitrogen,
oxygen, and methane. The proportion of oxygen
observed during both events ranged from 1% to
16%. In the 18-month post-capping event, the
gas collected from the control cell exhibited the
highest proportion of oxygen, and the gas from
the sand cap the least. However, during the 30-
month post-capping event, gas collected from
both the sand cell and the control cell contained
less than 3% oxygen. The proportion of methane
observed during both events ranged from 24 to
80%. The lowest values were found in gas
samples collected from the control cell during the
18-month post-capping event and the highest
values were found in gas samples collected from
the sand cell during the month-18 post-capping
event and the control cell during the 30-month
post-capping event. The proportion of carbon
dioxide observed during both events ranged from
0.8% to 3%, with no apparent trends between cell
or deployment event. Concentrations of TNMOC
were generally higher during the 18-month post-
capping event (220-780 parts per million by
volume [ppmv]) compared to the 30-month post-
capping event (24-91 ppmv). During the 18-
month post-capping event, the control cell
exhibited the lowest concentration of TNMOC in
gas and the sand cap the highest with the
AquaBlok® concentration in between.
Of the 20 reduced sulfur compounds that were
analyzed, five were detected in gas collected
from at least one flux chamber. Of these, only
hydrogen sulfide was detected in the gas from
every flux chamber sampled. In addition, only
methyl mercaptan was detected at a level more
then twice the detection limit (84 parts per billion
by volume [ppbv]) in gas from the control cell
during the 30-month post-capping event. Three
reduced sulfur compounds were detected in gas
collected from the AquaBlok® cell. Carbonyl
sulfide and carbon disulfide were detected at low
levels, comparable to gas samples collected from
the sand and control cells. Hydrogen sulfide was
present in much lower concentrations in gas
collected from the AquaBlok® cell (10-11 ppbv)
compared to levels measured in the flux
chambers deployed in the sand and control cells
(ranging from 290 to 18,000 ppbv).
Overall, all of the gases detected in samples from
the AquaBlok® cell during the 18-month post-
capping event were within the ranges observed
for the sand and/or control cell with the exception
of hydrogen sulfide, which was significantly lower
for AquaBlok® compared to both the sand and
control cell. It is possible that AquaBlok® retards
hydrogen sulfide through a mechanism other than
simple physical impermeability, such as sorption,
but this potential phenomenon was not
specifically assessed during the SITE
demonstration. An analysis of the relative
87
-------
presence of the various gases between
AquaBlok® and the sand and control cells was not
possible for the 30-month post-capping dataset
because no gas could be collected from
AquaBlok® during this event.
Table 3-6 shows the volumetric flux calculated for
each gas detected in the sample from at least
one chamber during any sampling event and a
total flux based on the sum of these individual
constituents. For constituents that were not
detected, a volumetric flux rate was determined
from detection limit data as a maximum possible
flux (i.e., using the detection limit as an upper
bound on the potential flux).
Volumetric flux was determined by the following
equation:
Flux = V
AT
(3-6)
V = volume of gas accumulated during
the flux chamber deployment (milliliters
Where:
A = cross-sectional area of the flux
chamber (m2)
T = duration the flux chamber was
deployed (days)
The volume of gas collected from the flux
chambers ranged from 160-10,800 ml_ during the
sampling events, and the deployment time for
each flux chamber ranged from 29 to 32 days
(see Table 3-4). The diameter of each flux
chamber was estimated to be 22.5 in (i.e.,
equivalent to the standard diameter of a 55-gal
drum). Accordingly, the cross-sectional area of
the flux chambers was estimated as 397.6 in2 or
0.2565 m2 (using the equation area = irr2).
Volumetric flux during the SITE demonstration
ranged from 21 to 1,453 ml_/m2-d (see Table 3-6).
No flux of gases was observed at AquaBlok®
chamber 1 during the 30-month post-capping
event, and an insufficient volume of gas was
recovered from the second AquaBlok® chamber
during this event for analysis. This could be
taken to suggest that the AquaBlok® was acting
as an impermeable barrier and preventing the
direct flux of gases from sediment to overlying
water. However, gas was recovered from the
one chamber sampled in the AquaBlok® cell
during the 18-month post-capping event, which
could suggest that there may have been active
ebullition through the AquaBlok® barrier at that
time releasing gases that collected beneath the
cap.
Table 3-6 also presents the mass based flux for
the detected compounds. To achieve this, the
volumetric flux was converted from ml_/m2-d to
m3/m2-d by applying a factor of 1/1,000,000 (i.e.,
1,000,000 ml_ = 1 m3). The Ideal Gas Law was
then used to convert the volume based flux
(m3/m2-d) to flux based on moles (mol) of gas
(mol/m2-d). This was done by taking the Ideal
Gas Law equation PV=nRT and rearranging as
follows:
n
P
(3-7)
where:
n = mol
V = volume
P = standard pressure
(1 atmosphere [atm])
R = universal gas constant
T = standard temperature
(25 degrees Celsius [°C])
This equation could then be solved for the moles
of a particular compound, and then the mass flux
of the collected gas converted from volumetric
flux using the following equation:
m2 -d
m
Xm3
m2 -d
(3-8)
Where:
X = the magnitude of the
volumetric flux after conversion
from mL to m3
Finally, the rate of gas production (mol/m2-d) was
multiplied by the molar mass of the compound
(mg/mol) and the concentration of the compound
(unitless; expressed as a fraction of the whole),
which were initially presented in a variety of units,
such as %, ppmv, and ppbv, for various
compounds. For non-detected compounds, a
mass flux was not calculated. In addition, a mass
88
-------
oo
VO
Flu:: Chamber FC i Location (Month 18)
Flux Chamber (FC) Location (Month 30)
AQUABLOK CELL
APATI ECEI
SAND CEI
COKE BREEZE
CELL
CONTROL
CELL
Flux Chamber SamplingLocations
Anacostia River, Washington, D.C.
ANACOSTIA RIVER
Figure 3-39. Gas Flux Monitoring Locations
-------
Table 3-4. SITE Demonstration Gas Flux Sampling Observations
Field Event
18-Month
Post-Capping
30-Month
Post-Capping
Cell
AquaBlok®
Sand
Control
AquaBlok®
Sand
Control
Flux Chamber
AB01
AB02
SA01
SA02
CN01
CN02
AB01
AB02
SA01
SA02
CN01
CN02
Deployment
Duration (days)
32
N/Aia)
32
32
32
32
29
29
29
N/Aia)
29
29
Recovered Gas
Volume (ml_)
2,500
~
1,400ID)
2,000
3,500
1,800
0
160™
600
~
1,100
10,800
Total Flux
(ml_/m2-d)
303
~
170
242
425
N/AIC)
0
21
81
~
147ie)
1,453
(a) Chamber not recovered; no sampling possible.
(b) Potential minimal loss of gas through leaking syringe.
(c) Flux could not be determined accurately due to significantly leaking chamber.
(d) Insufficient volume of gas to perform laboratory analyses.
(e) Potentially bad weld on chamber, but flux still calculable.
flux for TNMOC and an overall total mass flux
were not calculated, as these would be of very
limited usefulness in understanding the data.
The gas fluxes from native sediment observed
during the SITE demonstration were generally
comparable to those found in the available
literature. Volumetric methane fluxes from
sediments have been observed at other sites
ranging from 0.3 to 2,640 ml_/m2-day (Yuan, 2007
and references therein). In addition, methane
flux has previously been reported from Anacostia
River sediment in the laboratory as a function of
temperature, yielding 0, 341, and 917 ml_/m2-day
at 4, 22, and 35 °C, respectively (Yuan, 2007 and
references therein). By comparison, the fluxes
observed from the uncapped control cell ranged
from approximately 150 to 1,450 ml_/m2-day.
Water temperatures during the 18-month and 30-
month post-capping events were generally
approximately 26 °C.
Overall, it would generally appear that gas
ebullition was least pronounced in the AquaBlok®
cell, in particular on the basis of the lack of
accumulated gas to sample during the 30-month
post-capping event. However, the gas sample
that was collected from the AquaBlok® cell during
the 18-month post-capping event exhibited
generally similar concentrations and volumetric
and mass fluxes compared to the sand and
control cells. In this sample, hydrogen sulfide
was present at a significantly lower concentration
and exhibited significantly lower constituent-
specific volumetric and mass flux than in the sand
and control cells, indicating that AquaBlok® could
potentially have some specific retardation effect
on this compound. In general, the gas flux data
do not indicate that a sand cap alone has a
significant impact on gas ebullition. On the basis
of the data generated, it could be concluded that
AquaBlok® is more stable than sand in terms of
preventing gas migration. Alternatively, given its
high degree of impermeability, AquaBlok® could
be susceptible to a buildup of gases under the
cap and episodic releases of this built up gas if
enough pressure were generated. While this
phenomenon was not directly observed, it could
explain the ability to collect a gas sample from
the AquaBlok® cell during the 18-month post-
capping event (and potentially the loss of one
chamber during this same event).
Clear interpretation of the gas flux data from the
SITE demonstration is complicated by the loss of
certain chambers and the potential for other
sampling issues. These issues also prevented a
robust statistical analysis of the data to ascertain
90
-------
Table 3-5. SITE Demonstration Gas Flux Sampling Results
Analysis
General
Gases
Reduced Sulfur Compounds
TNMOC
Oxygen
Nitrogen
Methane
Carbon Dioxide
Hydrogen Sulfide
Carbonyl Sulfide
Methyl Mercaptan
Ethyl Mercaptan
Dimethyl Sulfide
Carbon Disulfide
Isopropyl Mecaptan
tert-Butyl Mercaptan
n-Propyl Mercaptan
Ethyl Methyl Sulfide
Thiopene
Isobutyl Mercaptan
Diethyl Sulfide
n-Butyl Mercaptan
Dimethyl Disulfide
3-Methylthiophene
Tetrahydrothiophene
2-Ethylthiophene
2,5-Dimethylthiophene
Diethyl Disulfide
Detection Limit
and Units
50 ppmv
0.10%
0.10%
0.10%
0.10%
4 ppbv
4ppbv
4 ppbv
4 ppbv
4 ppbv
4 ppbv
4 ppbv
4 ppbv
4 ppbv
4 ppbv
4 ppbv
4 ppbv
4 ppbv
4 ppbv
4 ppbv
4 ppbv
4 ppbv
4 ppbv
4 ppbv
4 ppbv
AquaBlok® Cell
AB01
Month 18
630
8.3
38
55
0.81
10
15
<10
<10
<10
14
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
AB01
Month 18
Duplicate
NA
NA
NA
NA
NA
11
17
<10
<10
<10
17
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
Sand Cell
SA01
Month 18
780
1.2
20
78
3.0
18,000
<300
<300
<300
<300
<300
<300
<300
<300
<300
<300
<300
<300
<300
<300
<300
<300
<300
<300
<300
SA01
Month 18
Duplicate
NA
NA
NA
NA
NA
18,000
<400
<400
<400
<400
<400
<400
<400
<400
<400
<400
<400
<400
<400
<400
<400
<400
<400
<400
<400
SA01
Month 30
24.0
2.59
44.9
50.9
1.65
7,000
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
<40
SA02
Month 18
410
5.0
38
57
1.8
12,000
<200
<200
<200
<200
<200
<200
<200
<200
<200
<200
<200
<200
<200
<200
<200
<200
<200
<200
<200
Control Cell
CN01
Month 18
420
12
46
40
1.2
840
18
<10
<10
<10
18
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
CN01
Month 30
80.9
1.22
18.7
78.3
1.74
2200
16
84
<12
19
<12
<12
<12
<12
<12
<12
<12
<12
<12
<12
<12
<12
<12
<12
<12
CN02
Month 18
220
16
58
24
0.91
290
19
<10
<10
<10
12
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
CN02
Month 18
Duplicate
220
16
58
25
0.90
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
CN02
Month 30
91.4
1.58
21.9
75.7
0.81
11000
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
<100
-------
temporal, spatial, or inter-cell trends. Specifically,
potential disruption of the seal between flux
chambers and the sediment they were keyed into
was observed in certain cases. It is, therefore,
not certain that the integrity of the chamber seals
was maintained for the entire duration of the
chamber deployments. If the chamber seal for
the AquaBlok® sample collected during the 18-
month post-capping event was compromised, it is
possible that gas may have flowed into the flux
chamber via lateral transport (i.e., short-circuiting)
rather than through the AquaBlok®. Furthermore,
it is unlikely that the gas flux assessment
captured all potential gas movement over the cap
areas, given that the flux chambers were located
in small isolated regions that could not have
captured ebullition cap-wide. As such, while a
quantitative interpretation of the gas flux data is
provided herein, these data should be evaluated
in the context of the uncertainties associated.
Specifically, while the data suggest that ebullition
did occur, the quantitative analysis should not be
taken to suggest that the gas flux evaluation was
able to quantify the duration, volume, or
concentration of all vapor flux.
Also, conceptually, a cap installed over
contaminated sediment would tend to eliminate
the accumulation of new organic-rich sediment on
the contaminated sediment surface. Accordingly,
it is likely that over time, the rate of biogenic gas
production in the contaminated sediment interval
would decrease. In addition, a cap could
conceivably be designed to specifically integrate
active or passive venting of biogenic gas.
3.3.1.2.3 Sediment Coring and
Analysis of Physical Parameters. Sediment
coring and analysis of physical parameters was
conducted during the six-month, 18-month, and
30-month post-capping surveys, as described
above in Section 3.2.4 and in direct conjunction
with the sediment coring described in Section
3.3.1.1.2.
As indicated in Section 3.2.4, all sediment core
samples were analyzed for TOC, PSD, and
moisture content. In addition, duplicate samples
were collected as appropriate and analyzed for
either this same set of parameters or a subset
thereof. The data generated from the analysis of
the sediment samples collected during the
various sediment coring events are summarized
in tabular form for all individual analyses in
Appendix H. In addition, graphs are provided in
Appendix H that show the average PSD
throughout the vertical profile of the cores
sampled from each cell. The PSD results were
averaged given the high degree of consistency
between sampling events. The PSD graphs are
grouped by sampling event and then by cell
within each sampling event.
As indicated in the PSD graphs in Appendix H,
sediment in the control cell was dominated by
silts and clays, with a decreasing amount of sand
with depth. A trace to small amount of gravel
was observed at the surface of the native
sediments in some cases. In the sand cell,
sample intervals in the sand capping layer were
generally nearly 100% sand with trace to small
contributions of silt/clay and gravel. In the
interface between the sand capping layer and
native sediment, the sample was dominated by
sand with an increasing amount of silt and clay,
and in the upper native sediment layer, the
sample was generally predominantly silt and clay
with only some sand. For the AquaBlok® cell, the
sand covering layer was generally nearly 100%
sand with trace to small contributions of silt/clay
and gravel, consistent with the sand capped cell.
In the interface between the sand layer and
AquaBlok®, the amount of gravel increased, and
in the AquaBlok® layer itself, the sample intervals
were highly dominated by the gravel and silt/clay
fractions. In the interface between AquaBlok®
and native sediments, the proportion of gravel
generally declined along with an increase in
silt/clay content. In the upper native sediment
layer, the sample was generally predominantly
silt and clay with only some sand and trace
gravel, consistent with the other cells. These
observations are consistent with all other
observations of the sediment type in the various
cells (i.e., SPI and visual assessment of core
logs) as well as information related to the actual
composition of the various materials used during
capping (i.e., AquaBlok® is a clay material with a
gravel core).
Figure 3-40 provides a comprehensive graphical
summary of the average TOC concentration
92
-------
Table 3-6. Calculated Volumetric and Mass Gas Flux for Individual Compounds
Compound
TNMOC
Oxygen
Nitrogen
Methane
Carbon Dioxide
Hydrogen Sulfide
Carbonyl Sulfide
Methyl Mercaptan
Dimethyl Sulfide
Carbon Disulfide
TOTAL
AquaBlok" Cell
AB01
Month
18
0.2
(N/A)
25
(33)
115
(132)
167
(109)
2.0
(4.4)
3.0E-6
(4.2E-6)
4.5E-6
(1.1E-5)
<3.0E-6
(N/A)
<3.0E-6
(N/A)
4.2E-6
(1 .3E-5)
303
(N/A)
Month
30(a)
N/A
(N/A)
N/A
(N/A)
N/A
(N/A)
N/A
(N/A)
N/A
(N/A)
N/A
(N/A)
N/A
(N/A)
N/A
(N/A)
N/A
(N/A)
N/A
(N/A)
N/A
(N/A)
Sand Cell
SA01
Month
18
0.1
(N/A)
2.0
(2.7)
34
(39)
132
(87)
5.1
(9.2)
3.1 E-3
(4.3E-3)
<5.1E-5
(N/A)
<5.1E-5
(N/A)
<5.1E-5
(N/A)
<5.1E-5
(N/A)
170
(N/A)
Month
30
1.9E-3
(N/A)
2.1
(2.7)
36
(42)
41
(27)
1.3
(2.4)
5.7E-4
(7.9E-4)
<3.2E-6
(N/A)
<3.2E-6
(N/A)
<3.2E-6
(N/A)
<3.2E-6
(N/A)
81
(N/A)
SA02
Month
18
0.1
(N/A)
12
(16)
92
(105)
138
(91)
4.4
(7.8)
2.9E-3
(4.0E-3)
<4.8E-5
(N/A)
<4.8E-5
(N/A)
<4.8E-5
(N/A)
<4.8E-5
(N/A)
242
(N/A)
Control Cell
CN01
Month
18
0.2
(N/A)
51
(67)
195
(224)
170
(111)
5.1
(9.2)
3.6E-4
(5.0E-4)
7.6E-6
(1 .9E-5)
<4.3E-6
(N/A)
<4.3E-6
(N/A)
7.6E-6
(2.4E-5)
425
(N/A)
Month
30
1 .2E-2
(N/A)
1.8
(2.4)
28
(32)
115
(76)
2.6
(4.6)
3.2E-4
(4.5E-4)
2.4E-6
(5.8E-6)
1.2E-5
(2.4E-5)
2.8E-6
(7.1E-6)
<1.8E-6
(N/A)
147
(N/A)
CN02
Month
30
0.1
(N/A)
23
(30)
318
(364)
1,100
(721)
12
(21)
1 .6E-2
(2.2E-2)
<1.5E-4
(N/A)
<1.5E-4
(N/A)
<1.5E-4
(N/A)
<1.5E-4
(N/A)
1,453
(N/A)
Volumetric flux precedes mass flux in parentheses
Units for volumetric flux = ml_/m2-day
Unites for mass flux = mg/m2-day
(a) No gas recovered from chamber
N/A = not applicable (because of sampling issue or calculation is not appropriate)
detected throughout the vertical profile of the
cores collected from each cell during each
sampling event. For the AquaBlok® cell, intervals
AB1 through AB3 represent samples of the sand
covering layer, intervals AB5 through AB7
represent samples from the AquaBlok® material,
and interval AB9 represents the upper horizon of
native sediment, while interval AB4 represents
the interface between the sand covering layer
and AquaBlok® material and AB8 represents the
interface between AquaBlok® material and native
sediment. In the sand cell, intervals SO1 through
SO4 represent sand capping material, interval
SOS represents the interface between sand and
native sediment, and interval SO6 represents the
upper horizon of the native sediment. For the
control cell, all intervals are obviously native
sediment.
As indicated on Figure 3-40, in the control cell,
TOC content was generally quite high for all
events and throughout the upper 9 cm of the
native sediments. The range of TOC was
generally between 6 and 12%, and declined with
depth in the upper 9 cm. This is generally
consistent with the likely deposition of new
organic detrital material at the surface and a
limited degree of subsurface mixing through
biogenic activity in addition to biogenic
consumption of organic material in the deeper
surface layers. The differences in TOC content
between events could be related to differences in
the deposition of new organic-rich sediment. For
instance, TOC at the surface of the control cell
appears to have declined in the 18-month post-
capping monitoring period, which could be
attributed to high river flow events that may have
93
-------
deposited relatively disproportionately inorganic
material rather than fine detrital material.
In the sand cell, TOC concentrations were
generally very low in the sand layer, and
increased with depth in the interface between
sand and native sediments. In the basal
sampling interval in the sand cell (i.e., the upper
native sediment horizon), TOC levels were
consistent with the native sediment material
sampled in the control cell. TOC levels were
slightly higher at the surface of the sand capping
layer than in the rest of the sand layer, but there
was no indication of significant vertical mixing at
the surface. These results are consistent with the
profile of the sand cell and the likely deposition of
new, more organic-rich sediment at the surface
than the relatively organically-inert sand used
during cap construction.
In the AquaBlok® cell, the TOC trend in the sand
covering layer was highly consistent with the
sand cap in the sand cell, and the TOC trend
between AquaBlok® and the native sediment was
generally consistent with the trend between sand
and native sediment in the sand cell. In the
AquaBlok® material itself, TOC concentrations
were generally higher than in the sand cover
layer or the sand capping cell in the month 6 and
month 18 data, ranging generally between 2 and
6%. In the month 30 data, levels of TOC in the
AquaBlok® cell were generally very low and
consistent with the inert sand covering layer.
Given than typical, unamended AquaBlok® is low
in organic content, it would appear that the month
6 and month 18 data were influenced potentially
by the entrainment of organic-rich native
sediment in certain samples. Alternatively, the
presence of higher levels of TOC in the
AquaBlok® material during the month 6 and
month 18 events could have been real and then
depleted by month 30.
Overall, the physical data generated through
sediment coring during the demonstration confirm
the physical stability of the sand and AquaBlok®
caps and corroborate other lines of evidence (i.e.,
SPI and oceanographic surveying) that indicate
the same. Moreover, the results from the
physical dataset appear to suggest that
AquaBlok® may have a greater sorption capacity
relative to sand given its greater proportion of
clay/silt and higher TOC content than generally
organically-inert sand (i.e., clay material and
organic carbon are capable of retarding organic
and inorganic contaminants through sorption
mechanisms).
3.3.2 Objective #2 - Ability of An
AquaBlok® Cap to Control
Groundwater Seepage
Tidal forces, regional pumping, or other
hydrogeologic phenomena in surface water
bodies have the potential to impose significant
vertical groundwater gradients into or out of
bottom sediments. One of the primary
advantages of AquaBlok® is that it is claimed to
significantly reduce permeability, which would be
reflected as a reduction in groundwater seepage
flows relative to seepage in sand-capped
sediments and uncapped control areas.
To evaluate the ability of AquaBlok® to control
groundwater seepage relative to sand and native
sediments, the following critical and non-critical
measurements were identified and assessed
through data collection during the various SITE
demonstration sampling events.
Critical Measurements
o Sediment coring and analysis of hydraulic
conductivity; and
o Seepage meter testing
Non-critical Measurements
o None
3.3.2.1 Objective #2 Results - Critical
Measurements
3.3.2.1.1 Sediment Coring and
Analysis of Hydraulic Conductivity.
Sediment coring and analysis of hydraulic
conductivity was conducted during the 18-month
and 30-month post-capping surveys, as
described above in Section 3.2.4 and in direct
conjunction with the sediment coring described in
Section 3.3.1.1.2. During each coring event, two
individual cores each were collected from two of
94
-------
Months 6,18, & 30 Coring Results
Average TOC Composition of Quadrants 1-4
Figure 3-40. Average TOC Concentration in Demonstration Area During SITE Demonstration (x-axis
represents vertical core profile from shallowest at left to deepest at right)
the four quadrants in the AquaBlok®, sand, and
control cells. Figure 3-22 displays the hydraulic
conductivity sediment coring locations from each
of the sampling events. As indicated on Figure 3-
22, some of the hydraulic conductivity sediment
cores intended to target the control cell were
collected outside the determined boundary of the
control cell. This occurred presumably because
of a misinterpretation of navigational position by
the coring contractor. However, given that these
cores were collected in identical native sediment
material as that within control cell, this does not in
any way compromise data usability or
representativeness. In addition, during the 30-
month post-capping hydraulic conductivity
evaluation, both conductivity cores from the sand
cell were collected from the same quadrant.
However, given that these cores both yielded the
appropriate interval of interest for analysis (i.e.,
the sand cap), this also does not affect the
demonstration results.
As indicated in Section 3.2.4, hydraulic
conductivity cores were preserved intact and
analyzed at the laboratory for hydraulic
conductivity. Table 3-7 summarizes the hydraulic
conductivity data generated in the AquaBlok®,
sand, and control cells. As indicated in Table 3-
7, the hydraulic conductivity of the native
sediment material during both events was very
low, on the order of 10"8 cm/s. The low
conductivity in the native material is likely
attributable to the cohesive, fine-grained nature of
the sediment. The hydraulic conductivity of the
AquaBlok® material during both events (i.e., 10"7
to 10"8 cm/s) was very similar to the range of
values determined for the native sediment and
consistent with the documented range for this
capping material (see Section 2.1). Hydraulic
conductivities in the range determined for
AquaBlok® (and native sediment in the
demonstration area) are indicative of a highly
impermeable material. Alternatively, the
calculated hydraulic conductivity for the sand
capping material from the sand cell (i.e., 10"3 to
10"4 cm/s), while it did demonstrate some
decrease between the 18-month and 30-month
post-capping events, was several orders of
magnitude greater than AquaBlok®.
95
-------
The hydraulic conductivity data generated during
the demonstration clearly indicate that AquaBlok®
is significantly less permeable than sand and
therefore likely to be characterized by far less
fluid flow and the potential for contaminant
movement in this fluid flow compared to the more
traditional sand capping material. Moreover,
while the AquaBlok® material demonstrated
similar conductivity when compared to native
sediments, it is likely that AquaBlok® would have
a lower intrinsic permeability given the greater
potential for preferential flow paths to develop in
native sediments from biogenic activity.
The results of the hydraulic conductivity testing
conducted during the SITE demonstration raise
an important question. Specifically, the data
suggest that certain native sediments might be
equally effective in terms of impermeability
compared to AquaBlok®. During any sediment
capping remedial design, a designer would
certainly want to evaluate all potential sources of
capping material to identify one with the greatest
probability of meeting performance objectives
and minimal cost.
3.3.2.1.2 Seepage Meter Testing.
Seepage meter testing was conducted during the
one-month, six-month, 18-month, and 30-month
post-capping surveys, as described above in
Section 3.2.4. Each seepage meter event was
conducted by deploying at least two ultrasonic
flux meters in the AquaBlok®, sand, and control
cells and collecting flux data from the meters for a
few to several days. The meters were deployed
and retrieved by divers, and accurate positional
control during seepage meter deployments was
achieved by using a GPS. Locations of the
submerged meters are provided on Figure 3-41.
Relevant meter location information, including
monitoring problems associated with each
seepage meter testing event, is described below.
Month 1
One meter location in the control cell during the
one-month post-capping seepage meter testing
event was moved during the deployment period
(i.e., from location ANA5-1 to location ANA5-2;
see Figure 3-41) due to significant data instability
presumably from gas ebullition.
Month 6
One AquaBlok® cell meter during the six-month
post-capping event was located improperly (i.e.,
in the portion of this capping cell that was
inadequately covered during construction). This
meter was moved as appropriate (i.e., from
location AQB1 to location AQB3; see Figure 3-
41) so that representative data could be
gathered. In addition, one meter location in the
control cell during the six-month post-capping
seepage meter testing event was moved during
the deployment period (i.e., from location CS1 to
location CSS; see Figure 3-41) due to significant
data instability presumably from gas ebullition.
Month 18
No meters required repositioning during the 18-
month post-capping seepage meter testing event.
Month 30
One AquaBlok® cell meter during the 30-month
post-capping event was moved during the
deployment period (i.e., from location AQB2 to
location AQB3; see Figure 3-41) based on data
variability that suggested the AquaBlok® cap may
have been compromised at the original location,
potentially by other sampling methods (e.g.,
coring).
Once the data from each meter were uploaded, a
representative 24-hour tidal cycle from at least
one meter location per cell was selected to
calculate a range of and average specific
discharge rate. The specific meter locations
relied on to perform these calculations were as
follows (see Figure 3-41):
Month 1
A representative 24-hour tidal cycle was selected
for location ANA1 in the AquaBlok® cell, location
ANA4 in the sand cell, and location ANA6 in the
control cell. These locations were selected as
they tended to exhibit less impact from gas
96
-------
Table 3-7. SITE Demonstration Hydraulic Conductivity Results
Cap Cell
AquaBlok®
Control
Sand
Material
AquaBlok®
Native
Sediment
Sand
Sampling Event
18-month Post-
capping
30-month Post-
capping
18-month Post-
capping
30-month Post-
capping
18-month Post-
capping
30-month Post-
capping
Quadrant
NE
SE
NE
NW
SW
outside cell
SW
outside cell
NW
NE
NW
NW
Hydraulic
Conductivity (cm/s)
1 7E-8
1.7E-7
4.1E-8
7.7E-8
6.6E-8
5.8E-8
1 .4E-8
8.7E-8
5.7E-3
8.3E-3
2.9E-4
1 7E-4
ebullition compared to the other meter in each
cell.
Month 6
A representative 24-hour tidal cycle was selected
for both properly located meters in the AquaBlok®
cell (i.e., AQB2 and AQB3) and both locations in
the control cell (i.e., CS1 and CS2). A
representative 24-hour tidal cycle was selected
for location SC2 in the sand cell, while the other
sand cell meter experienced data instability
related to gas ebullition. A 24-hour tidal cycle
was also used to complete calculations for the
improperly located AquaBlok® cell meter (i.e.,
AQB1) to provide reference.
Month 18
A representative 24-hour tidal cycle was selected
for both meters in the AquaBlok® cell (i.e., AQB1
and AQB2) and both locations in the sand cell
(i.e., SC1 and SC2). A representative 24-hour
tidal cycle was selected for location CS2 in the
control cell, while the other control cell meter
experienced a cable failure.
Month 30
A representative 24-hour tidal cycle was selected
for both locations in the sand cell (i.e., SC1 and
SC2) and both locations in the control cell (i.e.,
CS1 and CS2). A representative 24-hour tidal
cycle was selected for location AQB1 in the
AquaBlok® cell, but location AQB3 did not yield a
full 24-hour dataset. A 24-hour tidal cycle was
also used to complete calculations for the
potentially improperly located AquaBlok® cell
meter (i.e., AQB2) to provide reference.
Table 3-8 summarizes the calculated discharge
rates for the various meters deployed in the
AquaBlok®, sand, and control cells during the
SITE demonstration. As indicated in this table,
for each sampling event, the mean, minimum,
and maximum calculated discharge rates over
the representative 24-hour tidal period were
generally lowest for the seepage meters
deployed in the AquaBlok® cell. In addition, the
mean discharge rate measured in the meters
deployed in the AquaBlok® cell tended to be
negative, indicating an average flux from surface
water into the sediment as opposed to from the
sediment to the overlying water column. For the
most part, discharge measured in the control cell
was low but on average positive, indicating a
typically net flux from the sediment to surface
water in the native sediments. However,
variability from event to event tended to be
greatest in the control cell, which is not
unexpected given tidal variability and the realistic
expectation that native sediments would be least
effective at dampening tidal impacts on seepage.
Calculated discharge rates in the sand cell were
generally higher than in the AquaBlok® cell and
the control cell, suggesting the most significant
vertical movement of fluid from sediment to
surface water in this cell. It is not clear if there is
97
-------
oo
Seepage Meter (SM) Location (Month 1
O Seepage Meter (SM) Location (Month 6
ft Seepage Meter (SM) Location (Month 18)
Seepage Meter (SM) Location (Month 30)
AQUABLOK CELL
SC1 SC1 «
o
SC2J. QANA3
SC2S
ANA4E
SC2
AQB20
AQB1S AQB1
AQB3
O
AQB2
0» o OANA2
AQB2 ANA1
APATI ECEI
SAND CEI
COKE BREEZE
CELL
CONTROL
CELL
Seepage Meter Samp ling Locations,
Anacostia River, Washington, D.C.
ANACOSTIA RIVER
Figure 3-41. Seepage Meter Monitoring Locations
-------
a mechanism, such as hydrostatic pressure
buildup coupled with the permeable nature of
sand, that could be responsible for exacerbating
upward fluid flow in the sand cell. Alternatively, it
is possible that increased upward fluid flow in the
sand cap cell could have resulted from the
diversion of flow under the AquaBlok® cap
towards the adjacent sand cell. However, neither
of these specific potential phenomena were
assessed through the SITE demonstration.
Table 3-8 also suggests that the magnitude of the
difference in seepage between the AquaBlok® cell
and the sand and control cells was generally less
pronounced beyond the one-month post-capping
event. The reason for this apparent trend is not
known, but could be associated with the
increasing effects of gas ebullition beneath the
cap cells, or alternatively, with the incremental
increase in insults to the caps through invasive
sampling, which could have had the most
pronounced effect in the AquaBlok® cell by
mitigating the ability of the clay cap to control fluid
flow. Specifically, successively more cores that
penetrated the AquaBlok® cell could have
generated sand-filled channels as these voids
were then filled with the sand covering material.
Subsequently, seepage meters could have been
located at or near such locations where the
potential seepage control of AquaBlok® would
have been compromised.
During each sampling event, an empirical
harmonic analysis was also completed on the
data to determine the relative variability in
discharge compared to tidal phase. These
analyses generally indicated that the AquaBlok®
capping material was more effective at
dampening tidal influences on flux relative to the
sand capping material (i.e., the sand cell), and
also that the lag in flux induced by tidal phase
was shortest for the sand cell even compared to
the control cell sediments. It is not clear if there
is a mechanism, such as hydrostatic pressure
induced gradients, that could be responsible for
lessening tidal lag effects in the sand cell, and no
specific assessment was completed to resolve
this question.
Figure 3-42 shows the measured specific
discharge rates in the various cells, along with
the tidal phase and harmonic analyses for the
one-month post-capping seepage meter dataset,
and demonstrates the lower discharge through
AquaBlok® as well as the dampening effect of
AquaBlok® on tidal phase. Figures 3-43 through
3-45 show the same for the 30-month post-
capping data (graphed separately for each
individual cell). Appendix D provides additional
detail related to the individual seepage meter
testing events and application of the harmonic
analyses, including the rationale and methods for
the harmonic assessement.
In addition to the general evaluation of discharge
rates calculated during the SITE demonstration,
detailed statistical analysis was conducted for the
specific discharge measured through the
AquaBlok®, sand, and control cells to determine
whether there were any statistically significant
differences in seepage through the caps. The
statistical analysis was performed by fitting a
series of statistical models to the specific
discharge data. The data used for the statistical
analysis were the same 24-hour tidal cycle data
selected for each appropriate meter in each cell
to derive the general summary calculations
described above. The statistical models were of
increasing complexity to adjust for several
ancillary variables that could have affected the
measured specific discharge. The fitted models
are best expressed as:
Di]t = ju + Ct + L]{1} + sl]t (3-9)
j~~\ . f~i . T . DT1 i /O H (\\
Dm =V + C,+L ,m + pTt_g + eat (3-10)
D1Jt =
Where:
Lm
l]t
(3-11)
Dijt = specific discharge from location; in cap / at
time t;
p = average specific discharge (over all caps,
locations, and times);
C, = difference between average specific
discharge for cap / and the overall average;
LKi) = difference in specific discharge due to
location; within cap/;
% = random error in specific discharge
measurement at location; in cap / at time t,
99
-------
14
g 13
"CD
CD
CO
CD
O
CO
12
11
10
T3
o
Oi
CD
E> 10
CO
.c
o
to
Q 5
o
o
CD
Q.
CO
O
-/--
139.8 140 140.2 140.4 140.6 140.8 141 141.2
20
Control
Sand Cap
AquaBlok Cap
139.8 140 140.2 140.4 140.6
Julian Day 2004
140.8
141
141.2
Figure 3-42. Specific Discharge Rates in Demonstration Area During One-Month
Post-Capping Survey (top panel is tidal phase, lower panel shows discharge as
points, harmonic fit as solid curve, and trend as dashed line)
100
-------
8
7
-6
-------
(D 5
CL ,
269.2
269.4
269.6
269.8
270
270.2
•a
E
10
o
CD
Q_
269.2
269.4
269.6
269.8
270
270.2
269.2
269.4
269.6 269.8
Julian Day 2006
270
270.2
Figure 3-44. Specific Discharge Rates in Sand Cell During 30-Month Post-Capping
Survey (top panel is tidal phase, middle panel is station SC1, lower panel is station
SC2; discharge shown as points, harmonic fit as solid curves, and trend as dashed
lines)
102
-------
269.9
270.1
270.3
270.5
270.7
270.9
271.1
—. 10
T3
E
£, 5
Q.
269.9
270.1
270.3
270.5
270.7
270.9
271.1
10
E
o
d
269.9
270.1 270.3 270.5 270.7
Julian Day 2006
270.9
271.1
Figure 3-45. Specific Discharge Rates in Control Cell During 30-Month Post-Capping
Survey (top panel is tidal phase, middle panel is station CS1, lower panel is station
CS2; discharge shown as points, harmonic fit as solid curves, and trend as dashed
lines)
103
-------
Table 3-8. SITE Demonstration Seepage Meter Results
Sampling
Event
One-month
Post-
capping
Six-month
Post-
capping
18-month
Post-
capping
30-month
Post-
capping
Cell
AquaBlok"1
Sand
Control
AquaBlok®
Sand
Control
AquaBlok®
Sand
Control
AquaBlok®
Sand
Control
Meter
Location
ANA1
ANAS
ANA6
AQB1 (a>
AQB2
AQB3
Average™
SC2
CS1
CS2
Average
AQB1
AQB2
Average
SC1
SC2
Average
CS2
AQB1
AQB2IC)
SC1
SC2
Average
CS1
CS2
Average
Specific Discharge (cm/day)
Mean
-0.10
1.67
7.24
2.81
-0.45
-0.21
-0.33
0.30
-0.30
0.20
-0.05
-0.56
-0.68
-0.62
2.30
-0.81
0.75
0.63
-0.32
0.41
0.17
2.25
1.21
0.90
2.81
1.86
Minimum
-2.32
-0.30
5.34
0.99
-2.16
-0.65
-1.41
-1.24
-1.49
-0.76
-1.13
-1.98
-3.14
-2.56
0.00
-2.67
-1.34
-1.43
-1.81
-4.09
-1.29
0.15
-0.57
0.07
0.00
0.04
Maximum
2.05
5.54
10.83
4.81
1.08
0.42
0.75
3.07
0.60
0.91
0.76
0.54
2.21
1.38
5.50
2.03
3.90
3.16
0.97
4.13
3.01
6.70
4.86
1.55
5.32
3.44
Standard
Deviation
0.50
1.26
1.42
0.90
0.82
0.28
~
0.78
0.51
0.34
~
0.43
1.10
~
1.24
1.19
~
1.34
0.41
1.96
0.82
1.50
~
0.36
1.34
~
(a) Location AQB1 during the six-month post-capping event was outside the AquaBlok cell; data are
provided for reference and are not suitable for averaging with other data.
(b) Location AQB1 is not included in the average.
(c) Location AQB2 during the 30-month post-capping was in area where the AquaBlok® cap was
potentially compromised; data are provided for reference and are not suitable for averaging with
other data.
- Standard deviation not calculated for average.
Tt-e= measured tide at time t-e (with shift e
averaged over all caps);
Tt_ei = measured tide at time f-e,, with a cap-
specific shift;
/3 = effect of tide on discharge (average over all
caps and locations); and
/3, = effect of tide on discharge in cap /
Model 1 (Equation 3-9) is the simplest model,
expressing the specific discharge as a function of
the capping cell, with random differences due
only to meter location within each cell. Model 1
was fitted separately to the entire dataset for
each of the relevant 24-hour data periods, using
two different assumptions concerning the error:
(a) that the errors are independent, and (b) that
the errors are related. Resulting Model 1a is the
simpler model, with independent errors for each
observation, while resulting Model 1b allows for
correlated errors. Model 2 (Equation 3-10)
extends the basic model to account for the effects
of tidal stage on the specific discharge,
specifically by an amount proportional to the tide
height. In addition, preliminary examination of
the data showed that the effects of the tide were
offset in each cell, usually by approximately four
104
-------
to five hours (i.e., "tidal lag"). Thus, Model 2 also
incorporated this offset. In Model 2, the tide
effect and offset were kept constant across all
caps and monitoring locations. In Model 3
(Equation 3-11), additional complexity was
introduced by allowing the tide effect and tide
offset to vary for each individual monitoring
station.
Statistical analysis consisted of fitting the model
to the data and performing multiple comparisons
between the three treatments (i.e., AquaBlok®,
sand, and control cells) within the individual
models. This was done using Tukey multiple
comparisons after accounting for all model terms.
Table 3-9 shows the estimated mean specific
discharge for each cell for the various sampling
periods and models. These means were
compared using Tukey multiple comparisons, and
there were no statistically significant differences
at any p-value found between the means for the
various cells within any model.
The use of sequentially more complex models
provided increasingly more accurate attempts to
allocate the variability in the model to the different
factors that likely affected the specific discharge.
It was hoped that as the models became more
sophisticated, the capping methods would,
indeed, show statistically significant differences.
This was not the case. However, Table 3-9 does
show an interesting result. The estimated mean
specific discharge did not vary between cells
within the various models with the exception of
Model 3. In this case, because separate tide
offsets were used for each capping cell, there
was often less usable data for the analysis,
whereas the amount of usable data for the other
three models was consistent. As a rule,
therefore, the estimated means for Models 1a,
1b, and 2 may be more representative of true
conditions simply given the more robust datasets
applicable to each. In addition, on general visual
inspection of the sample data, it does appear
likely that the AquaBlok® cap allowed smaller
discharge on average than the sand cap and
native sediments (i.e., Table 3-8 shows that
mean and maximum specific discharge rates
were empirically lower in the AquaBlok® cell
compared to the sand and control cells for all
events), although this can not be statistically
shown at a reasonable level of significance.
3.3.2.2 Objective #2 Results - Non-critical
Measurements. There were no identified non-
critical measurements for this objective.
3.3.3 Objective #3 - The Influence of An
AquaBlok® Cap on Benthic Flora
and Fauna
As indicated in Section 3.2.1, a key concern in
applying AquaBlok® as an innovative sediment
capping alternative is the long-term effect of this
material on habitat for faunal (benthic)
communities, and also on potential habitat for
floral communities (which would necessarily
depend on site-specific water levels and
suspended sediment loads as they relate to a
favorable setting for emergent and/or submergent
vegetation). Standard (i.e., non-amended)
AquaBlok® material is inherently low in organic
content, and is not generally designed specifically
to support significant biological growth. However,
the AquaBlok® cap constructed during the SITE
demonstration was covered by a sand layer that
would likely support some level of biological
growth and allow for a comparison between
benthic impacts of the sand-covered AquaBlok®
cap and the sand-only cap (e.g., benthic infaunal
species impacts such as diversity and richness).
In addition, AquaBlok® is quite similar in terms of
grain size to the native sediment found in the
demonstration area (and at most contaminated
sediment sites), and could therefore itself prove a
viable habitat for benthos that rely on a fine-
grained substrate. Given the water depth and
turbidity in the study area, floral assemblages
were not found, and were therefore not assessed.
To evaluate the influence of AquaBlok® on
benthic communities relative to sand and native
sediments, the following critical and non-critical
measurements were identified and assessed
through data collection during the various SITE
demonstration sampling events.
Critical Measurements
o None
105
-------
Table 3-9. Results of the Statistical Comparison of Specific
Discharge between Cells (statistically significant differences were
not observed at any confidence level)
Sampling
Event
One-month
Post-capping
Six-month
Post-capping
18-month
Post-capping
30-month
Post-capping
Model
1a
1b
2
3
1a
1b
2
3
1a
1b
2
3
1a
1b
2
3
Mean Specific Discharge (cm/day)
Control
7.2776
7.3461
7.2717
14.0044
-0.0506
-0.0421
-0.0974
-1.5597
-0.8126
-0.8003
-0.7706
-3.2289
1.8550
1.8503
1.7743
0.9908
Sand
0.3513
0.3353
0.3573
-7.1242
0.3196
0.3511
0.3480
2.9017
0.7403
0.7284
0.7757
-2.0849
1.2104
1 .2250
1 .2497
0.2301
AquaBlok"
-0.0998
-0.1668
-0.1057
7.9703
-0.3206
-0.2891
-0.3510
-0.1541
-0.6259
-0.7290
-0.6831
3.4187
0.1160
-0.0348
0.1552
1 .9486
Non-critical Measurements
o Benthic grab sampling and descriptive and
statistical benthic assays; and
o Benthic assessment through SPI
3.3.3.1 Objective #3 Results - Critical
Measurements. There were no identified
critical measurements for this objective.
3.3.3.2 Objective #3 Results - Non-
Critical Measurements
3.3.3.2.1 Benthic Grab Sampling and
Descriptive and Statistical Benthic
Assays. Benthic grab sampling and descriptive
and statistical assays were conducted only during
the 30-month post-capping field event, as
described in Section 3.2.4. Thirty six total
sediment grab samples (three samples from each
of the four quadrants within the AquaBlok®, sand,
and native control cells) were collected to
evaluate benthic infaunal communities. Figure 3-
46 shows the locations where benthic grab
samples were collected. Figure 3-46 also shows
the sampling locations from a baseline pre-
capping benthic survey that was conducted but
was not part of the SITE demonstration
summarized in this ITER. Nevertheless, this
baseline ecological survey was critical in deriving
conclusions related to benthic impacts of the
capping activities.
The main objective of the infaunal study was to
examine the potential influence of the AquaBlok®
cap on the benthic community structure expected
to populate Anacostia River sediments (as
demonstrated by native control sediments)
relative to the influences of sand capping
material. The benthic assays and statistical
analyses conducted demonstrate that there were
substantial physical habitat differences between
the control cell and the two capped cells (i.e.,
AquaBlok® and sand) during the 30-month post-
capping benthic survey. This was not surprising
given that the native, primarily silty/clayey river
bottom was covered by sand or a combination of
sand and AquaBlok® at the capped cells. The
AquaBlok® and sand capped cells were relatively
similar to each other, but still were noticeably
distinct. The difference probably was primarily
related to differences in the fine sand particle size
fraction (higher in the sand cell) and the gravel
particle size fraction (higher in the AquaBlok®
cell).
106
-------
Despite the differences in habitats between the
control cell and the AquaBlok®and sand cells, the
faunal communities identified during the 30-
month post-capping event showed somewhat
surprising overall similarity among the cells.
Specifically, all of the sites sampled shared 15 of
22 ecological taxa. Some taxa encountered at
the site (e.g., Limnodrilus) are capable of
burrowing deeply into sediments, but most
observed typically occur at depths up to or less
than 10 cm in the sediment and have relatively
little effect on sediment properties below that
depth (Mermillod-Blondin et al. 2003). The
oligochaete worm Branchiura sowerbyi, which is
a deep-deposit-feeding species that burrows to
depths of 20 cm (Wang and Matisoff 1997), was
not found at either the AquaBlok® or sand capped
cell, but was found in the control cell. The
relatively high similarity among all three cells
probably resulted from the widespread
occurrence of six or seven species that varied in
abundances among sites. In addition, despite the
general similarity among cells, the control cell
community was clearly distinct from the
communities at the AquaBlok® and sand capped
cells. Most of the stations within the AquaBlok®
and sand capped cells were quite similar to each
other, but again showed distinct differences
between cells. The most likely factors explaining
the differences between the AquaBlok® and sand
capped cells were the relative abundances of
Dero nivea and chironomid larvae, both of which
were more abundant in the AquaBlok® cell (see
Figures 3-47 and 3-48, which are both box plots
demonstrating general summary level statistical
qualities of the data common of this type of visual
data display).
Overall, benthic habitats and faunal communities
in the AquaBlok® and sand cells were more
similar to each other than to those in the control
cell, but retained differences that clearly
separated them from each other. In particular,
the AquaBlok® stations had relatively equal or
greater abundances of individuals in the major
taxonomic groups found to occur in the sand cell
(see Figure 3-49, which is a box plot
demonstrating general summary level statistical
qualities of the data common of this type of visual
data display). This could be taken to indicate that
the AquaBlok® cell was a more suitable habitat
for benthic recolonization. However, this
conclusion is made tenuous by the presence of
the sand cover over the AquaBlok® material
which essentially presented the same habitat as
the sand capped cell (with the exception of total
thickness). In addition, the 2006 survey data
represent single snapshots of infaunal
communities that can vary seasonally and
annually and therefore are of limited use in
accurately predicting a longer term response of
the benthos to the AquaBlok® and sand caps.
The concentrations of metals in the native
sediment were at least 10 times greater and up to
30 times greater than the concentrations of
metals in the AquaBlok® and sand capping
materials. PAHs were 30 to 40 times greater in
the uncapped sediment, and total PCBs were up
to 95 times greater than they were in the
AquaBlok® and sand cap materials. Because the
capping materials were much less contaminated
than the native sediment, ecological
assemblages more commonly associated with
less contaminated sediments may increase over
time in cap materials, potentially differentiated by
physical cap material attributes (i.e., the sandy
nature of the sand cap material versus the clayey
nature of AquaBlok®). However, if the depth of
the fine sediment fraction observed to be
accumulating over the AquaBlok® and sand caps,
and potentially the associated concentrations of
organic contamination in this surficial layer, were
to increase through time, ecological assemblages
in both caps could potentially converge and
become more similar to the control cell.
The SITE demonstration did not include the
specific benthic assessment of an
uncontaminated reference area in the Anacostia
River and, thus, it is not possible to compare
benthic recolonization in the AquaBlok®, sand,
and native sediment cells from the demonstration
area to unimpacted areas.
Appendix F provides additional detail related to
the benthic assays and statistical evaluations
conducted during the AquaBlok® SITE
demonstration as they relate to objective #2.
3.3.3.2.2 Benthic Assessment Through
Sediment Profile Imaging. SPI surveys were
107
-------
conducted during the one-month, six-month, 18-
month, and 30-month post-capping surveys, as
described above in Section 3.2.4. During each
survey, a total of 12 locations were evaluated
using SPI in the AquaBlok®, sand, and control
cells. Specifically, nine locations were assessed
in each cell using the video SPI camera, and
three locations in each cell were evaluated using
the standard SPI camera. In addition, reference
stations outside the control cell and either nearer
to or in the Anacostia River navigation channel
were also assessed to provide further information
for comparisons to the control and AquaBlok® and
sand capped cells. Between the various surveys,
individual SPI locations were generally replicated
with a reasonable lateral offset to provide the
most meaningful data comparisons between the
sampling events. Figure 3-37 shows the SPI
locations assessed during each sampling event.
Accurate positional control of the SPI drops
during each sampling event was achieved by
operating the survey vessel in a very controlled
fashion and by using a dGPS linked to accurate
navigational software. All sediment profile
images were processed as described in Section
3.3.1.2.1, and were subsequently analyzed
visually for specific ecological features of interest,
including the presence of organisms and physical
indicators of the presence of organisms (e.g.,
burrows or tubes).
During the SPI assessment, certain ecological
measures could not be evaluated, as follows:
• Mud clasts - Mud clasts were not present in
adequate quantities to measure during the
demonstration. This was not surprising due
to the sand and gravel sized capping
material in the AquaBlok® and sand cells.
• RPD depth - There was no observable RPD
layering in sediments due to an apparent
gradual oxygen reduction gradient. This is
not atypical for a shallow water, tidally
influenced, riverine environment.
• Infaunal successional stage - The criteria
defining the developmental stages present
in the Anacostia River were not visible
during the demonstration. Invertebrate
assemblages were not observed in
adequate detail through SPI to measure
meaningfully, which is common in a
physically dominated environment.
• osi - Due to the inability to define RPD
depths and successional stage, it was not
possible to calculate osi during the
demonstration.
From an ecological standpoint, the relevant
general findings of the SPI surveys conducted
during the demonstration include:
• Gas filled voids were a prominent
subsurface feature during all sampling
periods. It appears that sediments at most
monitoring stations contained adequate
concentrations of organic detritus to support
a high rate of methanogenesis as
evidenced by the occurrence of gas filled
voids at most stations.
• There were significant declines in gas void
occurrence over time, which may have been
related to the weight of cap material
squeezing gas out of the sediments at a
rate faster than microbes generated gas
(see Figure 3-50). The decline was likely
not related to reduced microbial activity in
the river, as the presence of gas within the
control cell and channel stations remained
high through time. Gas voids were
generally observed at the lowest rate in the
sand cell, and no voids were observed in
the sand cell during the final two monitoring
events.
• Biogenic activity of infauna was not a
predominant factor in structuring subsurface
sediments at any monitoring station. There
were few biogenic structures, other than the
aforementioned gas voids and small tubes.
• Some small infaunal organisms were
observed in the sand capping material in
the sand cell and in the sand cover and
AquaBlok® material in the AquaBlok® cell.
The presence of infaunal worms at
monitoring stations in the demonstration
area indicates that benthic habitats were
supporting benthos populations.
108
-------
Baseline Benihir Oral: (BB) Loc alien (Baseline)
Benthic Grab (BG) Loction (Month 30)
AQUABLOK CELL
APATI ECEI
SAND CEI
COKE BREEZE
CELL
CONTROL
CELL
Benthic Grab Sampling Locations,
Anacostia River, Washington, D.C.
ANACOSTIA RIVER
Figure 3-46. Benthic Grab Sampling Locations (Including Baseline)
-------
300-1
Q.
E
CO
C/D
i_
CD
Q.
CD
O
C
CO
T3
CO
CD
'E
o
CD
Q
200-
100-
Site
o
to
o
Figure 3-47. Abundance of Dero nivea in AquaBlok (AB), Sand (SO), and
Control (DC) Cells (red dot represents the mean and horizontal line
represents the median)
d Abundance per Sample
Chironomi
100
90-
80-
70-
60-
50-
40-
30
20 —
10-
Site
I=T3
I I I
moo
Figure 3-48. Abundance of Chironomid Larvae in AquaBlok (AB),
Sand (SO), and Control (DC) Cells (red dot represents the mean and
horizontal line represents the median)
110
-------
• The prominent functional group observed
during the demonstration was the
subsurface deposit feeding oligochaetes.
Oligochaetes typically dominate tidal
freshwater systems but generally do not
occur in densities high enough to overcome
physical processes in the structuring of
sediments (Diaz and Schaffner, 1990).
The presence of even the limited number of
infauna observed in the AquaBlok® material
indicates a theoretical potential for bioturbation to
reach below the cap and mobilize contaminants
from native sediments. However, homogeneous
benthic habitats typical of tidal freshwater
typically do not support a diverse benthos
capable of significantly bioturbating COCs.
Overall, the surface sediment environment within
the tidally influenced freshwater environment at
the site appears to be controlled by physical
processes and does not appear to support a
highly structured benthic community, limiting the
power of image data in directly understanding
ecological recovery rates. Hence, biogenic
activity does not appear to be a factor, negative
or positive, in cap longevity. Benthic organisms
in the system appear from the SPI imagery to be
dominated by very small oligochaetes that were
generally unquantifiable in the images, and larger
benthic organisms were not observed. However,
snails were observed during coring activities and
surface grab sampling for benthic infaunal
analysis. Indirect indications of the benthos
community, such as RPD and color, show no
observable restrictions to benthic community
colonization within the AquaBlok® cap system,
other than grain size, when compared to the
contaminated native sediments or the sand only
cap. The SITE demonstration did not include the
specific benthic assessment of an
uncontaminated reference area in the Anacostia
River and, thus, it is not possible to compare
benthic recolonization between the demonstration
area image data and unimpacted areas.
Were significant benthic communities to be
present in the cap areas, it is possible that
bioturbation could potentially affect the ability of
the caps to provide effective contaminant
isolation (i.e., through vertical mixing or the
creation of preferential contaminant migration
pathways). However, cap thickness in both the
AquaBlok® and sand cap areas was consistent
with or greater than the typical bioturbation depth
of most common benthos, suggesting that even a
robust population of benthos would not likely
have significantly impacted contaminant
distribution during the demonstration.
The AquaBlok® cap was covered with a sand
layer, and the presence of benthos in the
AquaBlok® cell was most closely related to the
sand covering layer as opposed to the AquaBlok®
cap material itself. However, given that the grain
size composition of the AquaBlok® material is
generally consistent with the native sediment in
the Anacostia River (i.e., with the exception of the
gravel size component, AquaBlok® is almost
entirely fine grained as is the native sediment), it
does not appear that there is a physical grain size
limitation to the benthic colonization of
AquaBlok®. This is supported by the observation
of some benthos in the AquaBlok® material during
the SPI surveying. AquaBlok® caps are
frequently designed with a sand covering layer to
accommodate a required minimum cap thickness
(i.e., while only a thin layer of AquaBlok® may be
required to form a suitable isolation barrier, there
could be a design requirement to achieve a
greater minimum thickness, and, assuming it
would be resistant to erosion, sand would
potentially be used to make up the additional
required thickness) or to provide a more suitable
habitat compatible with existing ecology (i.e.,
where AquaBlok® is used to cover a
characteristically more coarse-grained
environment) but it is not a design requirement
that this sand layer be included. Importantly,
absent a sand covering layer, benthic recovery in
an AquaBlok® cap may not, in environments
where native sediments are characteristically
fine-grained, be subject to physical constraints of
grain size. This SITE demonstration did not
provide data suitable to determine the relative
benthic recovery rates in AquaBlok® itself.
Appendix C provides additional detail related to
the SPI surveys conducted during the AquaBlok®
SITE demonstration as they relate to objective
#2.
Ill
-------
1300-
™ 1200
"o.
£ 1100-
03
^ 1000-
0>
S" 90°-
I 800-
•Q
ci ynn
< 600-
03
o 500-
i
400-
Site
•
B
,
1
1 1
moo
< O> Z)
Figure 3-49. Total Benthos Abundance in AquaBlok (AB), Sand (SO), and
Control (DC) Cells (red dot represents the mean and horizontal line
represents the median)
40
35
30
«f 25
c
-=• 20
V)
D AquaBlok
nSand
D Control
I
15
10
1 5
-10
Month 01
Month 06
Month 18
Month 30
Figure 3-50. Gas Void Occurrence Trend in Video SPI (columns represent the
mean, n is the population size, and error bars represent 95% upper and lower
confidence intervals around the mean)
112
-------
Section 4
Economic Analysis
The primary purpose of this economic analysis is
to summarize costs incurred in deploying
AquaBlok® at pilot-scale through the Anacostia
River SITE demonstration, and estimated costs
for deploying a standard formulation of
AquaBlok® at full-scale at a contaminated
sediment site for the purpose of isolating
sediment contamination and mitigating human
health and ecological risks. The majority of the
information in this section was provided directly
by AquaBlok, Ltd., and was not independently
verified. It is therefore true and accurate to the
extent that AquaBlok, Ltd. provided true and
accurate information.
4.1 SITE Demonstration Pilot-Scale
AquaBlok® Capping Costs
The primary objectives of the evaluation of
AquaBlok® as an innovative sediment capping
technology under the SITE program were (1) to
assess its relative stability, (2) to assess its ability
to provide a low-permeability barrier to flux of
contamination, and (3) to assess its effect on flora
and fauna. These measurement endpoints were
evaluated using a number of field sampling tools,
including the deployment of seepage meters and
the collection of multiple sediment cores.
Consequently, in initiating the SITE
demonstration there was an overall emphasis on
ensuring adequate coverage of the cap area and
a sufficient cap thickness for sampling purposes.
Section 3 provides additional detail on the overall
objectives of the SITE demonstration and the
measurement tools used to assess AquaBlok®
performance.
A detailed analysis of installation costs was not
completed for the AquaBlok® pilot-scale cap due
to the nature of the SITE program, the use of
common equipment for construction of multiple
pilot test caps using different materials, the small
scale of the application, and particular project
emphasis on the iterative evaluation process
necessary to better identify and describe the
relative performance attributes of the AquaBlok®
material. That is, as the nature of the project
placed more emphasis on the performance of the
AquaBlok® cap relative to that of more traditional
capping material, and because common
construction equipment and techniques were
selected for the overall project, the aspect of
overall efficiency in cap construction, which
translates directly into overall capping costs, was
secondary. Nevertheless, some observations
can be provided regarding the SITE program
costs relative to the importance of the level of
care taken during the process of construction
monitoring to ensure construction of the cap
design (and minimum thickness) as intended.
4.1.1 SITE Demonstration As-Built
AquaBlok® Cap
The specific AquaBlok® formulation used for the
SITE demonstration was "3070 FW, a
nomenclature that indicates the material was a
freshwater formulation consisting of 30% clay and
70% aggregate on a dry weight basis. The
AquaBlok® incorporated a No. 8 (i.e., 0.094 to
0.375 in diameter) aggregate as a core. Using
this particular formulation, and based on
preliminary laboratory column testing conducted
by the vendor in support of the SITE
demonstration in the Anacostia River, a "dry"
product thickness of approximately three in was
required to construct a basal AquaBlok® layer at
the targeted hydrated (expanded) thickness of
approximately 4 in (see Section 3.1.1.2). At a dry
product bulk density of approximately 85 pounds
per square foot (Ibs/ft2), this target 3-in dry
thickness translated to a target dry application
rate of approximately 21 Ibs/ft2.
113
-------
Given the final target cap placement area of
approximately 8,000 ft2, a total of approximately
85 tons of AquaBlok® product was needed for cap
construction. The contractor responsible for cap
construction (as indicated in Section 3.1.1.2, the
AquaBlok® cap was not constructed by Battelle
on behalf of EPA NRMRL, but rather by a
separate contractor acting pursuant to an
investigation separate from the SITE
demonstration summarized and discussed in this
ITER) requested delivery of a total of 110 tons of
product to the site. The extra quantity of product
ordered (i.e., approximately 25 tons) was
intended to address potential on-site wastage and
other contingencies, and was also based on an
initial target cap placement area of 10,000 ft2
(rather than the 8,000 ft2 cap area ultimately
placed - see Section 3.1.1.2).
Confirmation core samples were collected by the
contractor during actual construction of the
AquaBlok® cap. Over the approximately three-
day period during which AquaBlok, Ltd. personnel
were on site during cap construction, the
contractor was observed to collect a total of 19
confirmation core samples, generally within 10
minutes to several hours from product placement.
Eighteen of the cores were observed to contain
measurable quantities of "dry" AquaBlok®, with
measured initial material thicknesses (while still in
coring tubes) ranging from approximately 2.5 in to
approximately 7 in. For the majority of cores,
AquaBlok®/sediment interfaces appeared
relatively distinct (indicating minimal mixing),
although some degree of penetration of
AquaBlok® particles into the underlying sediment
was observed by AquaBlok, Ltd. personnel to
occur in most cores (up to depths of
approximately 0.5 in). Furthermore, a number of
the 18 cores clearly displayed the presence of
previously placed (i.e., hydrated) AquaBlok®
overlain by newly placed (non- to only slightly
hydrated) product, indicating overlapping or
"double" coverage of the product in some areas.
Overall, using the contractor's on-site
measurement data from cap placement
confirmation sampling, the mean value for initial
("dry") AquaBlok® thickness within the 18 core
samples was approximately 3.1 in. This
calculation considered total cap material
thickness (including any overlapping) but did not
include the quantity of AquaBlok® observed to
penetrate the native sediment surface. The mean
thickness for dry AquaBlok® appeared consistent
with the dry thickness of product required for
construction of the hydrated cap.
By the end of cap construction, and based largely
on survey data generated during cap placement,
AquaBlok® product had been placed across
approximately 90% of the total 8,000 ft2 target
area. Less-than-complete coverage of the entire
8,000 ft2 area can be attributed to a combination
of factors, including a dwindling supply of bulk
product available towards the end of the
construction phase coupled with a desire on the
part of the cap construction contractor to not
extend construction beyond a limited number of
working shifts. Specifically, equipment and
material barges were left essentially stationary
over one corner of the 8,000 ft2 AquaBlok® cap
footprint. Coverage of the final 10% of the total
cap area would have required repositioning
equipment and material-holding barges, and
there was an apparent reluctance to do this to
avoid extending cap construction into another
consecutive work day.
Based on simple mass-per-area calculations, the
dwindling supply of product available towards the
end of the construction phase was largely due to
over-application of product despite the
approximately 3-in average value measured in
the 18 initial confirmatory core samples. That is,
a total of 110 tons of product (220,000 Ibs) was
placed across approximately 7,200 ft2, translating
to a site-wide application rate of approximately 31
Ibs/ft2, which is well above the targeted
application rate of 21 Ibs/ft2. At this site-wide
application rate, dry thicknesses of AquaBlok®
actually measured in most core samples should
have been closer to about 4.5 to 5 in (rather than
the actual measured average of approximately 3
in). The fact that measured thickness data do not
reconcile directly with thicknesses predicted from
mass-per-area calculation is likely because areas
where AquaBlok® was placed at greater than
design thickness were apparently not adequately
represented during the construction monitoring
and confirmation core collection process and
quantities of AquaBlok® material observed to
114
-------
penetrate the sediment surface were not included
in the contractor's thickness measurements.
Other, earlier demonstrations of AquaBlok®
placement, particularly the Ottawa River Capping
Demonstration funded by the Ohio Lake Erie
Commission, have been specifically designed to
demonstrate effective placement over larger
areas (2.7 acres in the case of the Ottawa River
project) at lower application rates (as low as 12.5
to 16 Ibs/ft2) and to document application costs
and production rates for a variety of application
techniques (Hull & Associates, Inc., 2002). In
addition, AquaBlok® pilot applications and full-
scale projects completed by others have
achieved application rates on the order of the
original Anacostia SITE demonstration target rate
within relatively good tolerance levels and at
equivalent placement rates of approximately one
acre per day.
4.1.2 SITE Demonstration AquaBlok®
Pilot Costs
Product cost for the Anacostia SITE
demonstration pilot-scale AquaBlok® cap (i.e.,
AquaBlok® and covering sand) was initially
estimated at approximately $2 to $3 per ft2. This
initial cost estimate was based on placement of a
3-in thick dry AquaBlok® layer, which, as
discussed above, corresponds to a dry product
application rate of approximately 21 Ibs/ft2. A
product cost of $170 per ton (i.e., the actual
AquaBlok® material cost for the Anacostia cap
construction phase), assuming placement at the
target thickness over the entire capping area,
yields a cost of approximately $1.80 per ft2 for the
AquaBlok® capping material evaluated during the
SITE demonstration.
4.2 Full-Scale AquaBlok® Application
4.2.1 Site-Specific Factors Affecting Cost
Site-specific factors that will affect overall full-
scale costs for AquaBlok® capping projects
include (in approximate order of relative
sensitivity):
• Salinity;
• Project location;
• Project size;
• Performance criteria;
• Composite cap design elements; and
• Regulatory constraints
AquaBlok® formulations designed to function in
full-strength seawater require clay mineral blends
(such as attapulgite) that can materially increase
formulation costs.
Geographical project location and accessibility
can impact project costs primarily as a result of
greater shipping and packaging costs. Local
manufacture can eliminate many of these costs,
but typically it is only practical for sites over two to
three acres in size. Similarly, installation costs
can be greater if access is not readily available,
requiring mobilization and use of specialty
equipment, or lengthy barge transport cycles to
feed the installation equipment.
Project size will influence per-acre costs in
several ways. Smaller projects often do not
economically justify mobilization for local
manufacture, resulting in the need to package
and ship capping materials to the site.
Transportation costs can add substantially to an
overall project cost. For example, the relatively
small amount of material for the Anacostia River
SITE demonstration pilot project cost $31,300 via
free on board (F.O.B.) shipping to the site
($18,700 for AquaBlok® material and $12,600 for
packaging and shipping).
Similarly, costs for mobilization and
demobilization of construction application/
installation equipment must be recognized in
addition to actual application costs, and must be
applied over the entire application area. Thus, a
relatively small project area that cannot
adequately be reached by shore-based
equipment might require the mobilization of
barges, or if water depths are insufficient to float
barges, might require the use of aerial application
methods (e.g., helicopter). Such factors could
result in significantly higher per-acre installation
costs for smaller sites as compared to larger
application areas.
Although AquaBlok® can and has been
successfully used in construction of "monolayer"
115
-------
cap designs, most applications of AquaBlok®
incorporate its use as part of a composite cap,
potentially including the use of a sacrificial
absorbent or stabilization layer beneath the
AquaBlok® cap, followed by a sand layer for
benthic isolation purposes or stone armoring for
increased protection from scour in higher-energy
environments. The cost of the AquaBlok®
component will be impacted directly by the
relative thickness of the AquaBlok® portion of a
composite cap. While a total freshwater cap
thickness of approximately 6 in is often desired to
address ecological risks, a thinner (e.g. 3 to 5 in
thick) hydrated AquaBlok® cap at the base of a
sand cap can often provide adequate chemical
isolation of sediment-borne contaminants. As in
the Anacostia SITE demonstration capping
project, AquaBlok® material has been
successfully applied at other project sites with a 2
to 3 in application (pre-hydrated) within
acceptable tolerances.
Performance criteria will also impact the relative
cost of AquaBlok®. If amendments are required
to provide enhanced reduction of contaminant
flux (particularly for organic compounds or
methylated mercury), or if the cap design requires
treatment gates to address gas ebullition or
groundwater upwelling, the material costs will be
greater in proportion to the cost of the
amendments required.
Other regulatory constraints, such as access
restrictions due to fish spawning (i.e., "fish
windows") or migratory waterfowl use, could
impact the timing of cap installation, possibly
requiring multiple site mobilizations or
constrained working hours. In addition, the
applicability of permit programs could constrain a
capping remedy. Other constraints could include
physical access issues, tidal periods, the
presence of high hazard conditions (e.g.,
unexploded ordinance [UXO] or free-phase
product) that could require special materials and
methods.
4.2.2 Issues and Assumptions
Similar to the Anacostia River study area where
the SITE program was conducted, many
sediment cap applications can be accomplished
to address environmental risk issues resulting
from common contaminants such as heavy
metals, PAHs, and PCBs. These contaminants
are often found in combination to some degree as
a legacy of commercial and industrial activities,
are often co-located as a result of
contaminant/sediment transport and deposition
phenomena, and typically have a common affinity
for fine-grained matrix particles such as typical
subaqueous sediments. For purposes of
examining relative AquaBlok®-based cap
implementation costs, various application
parameters can be assumed for which more
detailed cost assumptions and estimates can be
provided. However, it is important to note that,
because the Anacostia River site is fairly typical
of many sites, the assumptions that follow could
reasonably apply to a full-scale implementation of
a cap at such a typical location as well.
For a "typical" application scenario, it is
reasonable to assume a 3 to 4 in thick, hydrated
layer of the 3070 FW AquaBlok® formulation
(achieving an in situ permeability of 10"8 cm/sec
or less) would be appropriate to provide chemical
isolation, absorption, and/or suitable cation
exchange capacity for metals. Such an
application would also have sufficient bearing
capacity to support an overlying 3-in sand layer
intended to provide additional bioturbation
isolation and benthic restoration capacity. For
purposes of this cost application, it is assumed
that ebullition of gases is not a significant
performance issue.
The cost analysis for this scenario assumes that
the full-scale site is ten acres in size, located on
the East or West Coast, Upper Midwest or Gulf
Region (logisitics of shipping raw materials for
local manufacture should not vary over 15% for
this scenario). Approximately 4,900 tons of
AquaBlok® would be required. The size of this
project would justify local manufacture to
eliminate packaging cost and minimize
transportation costs. Although a mobilization cost
for local manufacture would still apply, as would
transport to the project site (assumed to be within
five miles), the offsetting savings compared to
packaging and transport from a remote
manufacture site would more than offset costs to
establish and support local manufacture.
116
-------
Based on the dry bulk density of AquaBlok® 3070
FW, this "typical" scenario would require an
application rate of approximately 22.5 Ibs/ft2. For
purposes of this full-scale scenario evaluation, it
is assumed that the application would be through
15 ft of water by barged-based conveyor (see
Figure 4-1), with a supply barge(s) (see Figure 4-
2) reloaded by a shore-based conveyor. A
conveyor application has been demonstrated to
be highly efficient for larger-scale applications.
An application rate of 500 tons per day is
assumed for both the AquaBlok® and overlying
sand material, which would require 20 days total
application time. The Anacostia SITE
demonstration pilot-scale application used 110
tons of AquaBlok® applied over three to four
hours of actual construction time over a two-plus
day period, with much of the intervening time
spent in training, addressing field logistics, and
holding strategic discussions. For reference,
another AquaBlok® pilot program was completed
using a clamshell with a larger bucket and a
supply barge with a larger capacity than were
used in the Anacostia project applying
AquaBlok®, and, at a similar application rate,
accomplished a 0.75 acre cap placement over
approximately six hours total. QC for pilot
purposes generally does necessitate a slower
application rate than could be experienced on a
full-scale application. Also, given the relatively
short application time, it is assumed no additional
costs for restricting or controlling access by boat
traffic would be necessary. Based on monitoring
completed during the Anacostia SITE project and
other AquaBlok® applications, it is assumed that
(unlike many dredging projects) additional
controls to minimize and control turbidity and
resuspension of sediment (e.g. silt curtains)
would not be required.
It is also assumed that sufficient material would
be manufactured ahead and stockpiled on-site so
that after application is initiated, the process
would continue through single (extended) day
shifts until project completion.
4.2.3 Full-Scale AquaBlok® Application
Cost Categories
4.2.3.1 General Cost Categories. Project
costs for an AquaBlok®-based composite cap can
be estimated for the following general categories:
• Material costs associated with local
manufacture of AquaBlok® and purchase of
sand;
• Installation costs, including equipment,
labor, general overhead and profit;
• Construction QC and documentation; and
• Engineering design, permitting, contract and
bid document preparation, and contract
administration
4.2.3.1.1 Local AquaBlok® Manufacture
Costs. A 10-acre capping project would justify
local or near-site manufacture. Facility costs
associated with such an activity would include a
four-month lease period to cover set-up, raw
material accumulation, and manufacture prior to
and during actual application. In addition,
mobilization of key manufacturing equipment and
installation of the equipment and utilities would
result in a cost, as would miscellaneous items
such as insurance and security. While most
manufacturing could reasonably be accomplished
with local labor, a production supervisor and lead
manufacturing QC personnel would incur travel
and per-diem costs during the four-month period.
Finally, transport costs for the local manufacture
site are estimated based on a one-way transport
distance of five miles and assuming a 20-ton
payload. Table 4-1 includes estimated costs to
cover these project components. For purposes of
this analysis, it is assumed material costs are
purchased directly by the project owner. If they
are purchased by the installation contractor, they
would typically be subject to a mark-up.
4.2.3.1.2 AquaBlok® Cap Installation
Costs. Installation activities would include
mobilizing appropriate construction equipment to
117
-------
- *- fc
V •+*' •* - -A'-
t* o
•
<"
%-*4 "
1
J?
-
Figure 4-1. Typical Barge-Mounted Material Conveyor
Figure 4-2. Typical Material Barge
118
-------
Table 4-1. Cost Detail for "Typical" AquaBlok® Capping Project
(10-acre AquaBlok® Cap with Sand Cover)
Item
Rate
Amount
Units
Cost
Materials
facility lease/insurance/security
manufacturing setup/mobilization
AquaBlok® manufacture
Sand (delivered)
AquaBlok® transport (5 mi)
SUBTOTAL
$2,500
$125,000
$180
$30
$70
10
1
5,000
6,000
250
weeks
lump sum
tons
tons
loads
$25,000
$125,000
$900,000
$180,000
$17,500
$1,247,500
Installation
equipment mob/demob
support trailer w/ utilities/security
equipment rental
GPS
conveyor
backhoe
terrain loader
front-end loader
barges (2)
work boat
equipment fuel/maintenance
conveyor
backhoe
terrain loader
front-end loader
work boat
labor
conveyor operator
backhoe operator
terrain loader operator
front-end loader operator
work boat operator
general laborers
supervisor/foreman
SUBTOTAL
contractor
bond/insurance
overhead/profit
SUBTOTAL
$150,000
$1,000
$190
$788
$2,038
$736
$2,520
$5,250
$1,050
$1,200
$1,600
$800
$1,800
$2,500
$3,098
$2,113
$2,033
$2,112
$4,988
$1,583
$1,699
1
5
25
10
5
5
5
5
5
10
5
5
5
5
10
5
5
5
5
10
5
2%
15%
lump sum
weeks
days
weeks
weeks
weeks
weeks
weeks
weeks
weeks
weeks
weeks
weeks
weeks
weeks
weeks
weeks
weeks
weeks
weeks
weeks
equip/labor
equip/labor
$150,000
$5,000
$4,750
$7,880
$10,190
$3,680
$12,600
$26,250
$5,250
$12,000
$8,000
$4,000
$9,000
$12,500
$30,980
$10,565
$10,165
$10,560
$24,940
$15,830
$8,495
$382,700
$7,655
$57,405
$65,060
Construction Quality Control and O&M
QC scientists (2)
final observation report
miscellaneous expenses
O&M (1 -yr, 5-yr, and 1 0-yr)
SUBTOTAL
$10,000
$25,000
$2,000
$10,000
10
1
1
3
weeks
lump sum
lump sum
events
$100,000
$25,000
$2,000
$30,000
$157,000
Engineering Design
engineering design
permits
bid prep/contract administration
SUBTOTAL
TOTAL
$120,000
$30,000
1
1
7%
lump sum
lump sum
installation
$120,000
$30,000
$31,346
$181,300
$2,034,000
Costs from RS Means, Putzmeister, vendor sources, and/or engineering estimates.
119
-------
the project area, preparing a material
laydown/unloading area, providing ramp access
for work boats, providing site security and safety
amenities as appropriate, and establishing a
project trailer and utilities.
Equipment rental and labor costs to operate the
equipment, including contractor overhead and
profit, are also included in this category, as are all
construction permits, access fees, bonds and
contractors' insurance requirements.
As noted previously, it is assumed that
application of the AquaBlok® and sand capping
material could be accomplished at a base rate of
one acre per day for each material, resulting in a
five week period for a 10-acre site.
Following cap completion, an additional cost for
demobilization and site restoration must be
budgeted. For purposes of this analysis,
demobilization costs are included in the
mobilization cost estimate.
Insurance and bond fees are assumed to be an
average of 2% of the total material and
construction costs, and general overheard and
profit are assumed to be 15% of the construction
costs.
Equipment rental needs are assumed as follows:
• One front-end loader;
• Two mobile articulated/telescoping
conveyors with hoppers (one barge-
mounted for material application, one shore-
based to load supply barges);
• One backhoe to load material from supply
barge to application barge;
• One supply barge and one application
barge; and
• One workboat to position barges
Specific construction labor personnel for the
project are assumed to be as follows:
• Two conveyor operators;
• Three operators (backhoe, front-end
loader, and terrain loader);
• Three work boat operators;
• Two miscellaneous laborers; and
• One supervisor/foreman
Specific production rates and a conceptual daily
resource leveling schedule are provided on
Figure 4-3, and resulting material and
construction-related costs are detailed on Table
4-1.
4.2.3.1.3 Construction Quality Control
and Documentation Costs. QC during
installation is a fairly straightforward process
which would consist of a team of
scientists/technicians obtaining core samples on
a periodic yet systematic basis, ascertaining dry
and hydrated cap thicknesses, reviewing daily
positioning data, compiling records, and at the
end of the project preparing a final report. These
professionals would incur travel, labor, and per-
diem costs. In addition, the sampling crew would
require the use of a small boat for the five-week
installation period. Table 4-1 includes estimated
costs to cover these project components.
4.2.3.1.4 Engineering Design,
Permitting, Contract and Bid Document
Preparation, and Contract Administration
Costs. It is assumed that although a typical
project would have already undergone a site
investigation and experienced costs consistent
with a remedial investigation (RI)/FS, additional
engineering design would be required to establish
construction parameters specific to the selected
remedy; in this case, an AquaBlok®-based
composite cap.
Typically, engineering design costs for an
AquaBlok® cap would be incurred in establishing
more detailed site bathymetry, particularly for QC
and material pay quantity determinations and
permit drawings. Relevant permits may include
Clean Water Act (CWA) Section 404 and 401
permits/ certifications (see Section 2.11). Costs
to prepare necessary bid and construction control
documents and provide construction contract
administration are estimated at 7% of the
construction costs, which is a reasonable
engineering assumption. Table 4-1
120
-------
Activity
Setup and transportation
07:00
Work boat operation
Conveyor loaded by front-end
loader
08:00
09:00
10:00
11:00
Barge loaded by conveyor
Conveyor loaded by backhoe
Material moved by terrain loader
Cap installed with conveyor
Figure 4-3. Conceptual Daily Work Cycle for "Typical" AquaBlok® Capping Project (10-acre AquaBlok® and Sand Cap)
-------
includes estimated costs to cover these project
components.
4.2.3.1.5 Operations and Maintenance
Costs. A properly designed and installed
AquaBlok® cap would require virtually no ongoing
O&M activities or costs other than might be
required as part of an extended QA/QC program.
If the cap is appropriately designed to consider
the sediment substrate bearing capacity, potential
bioturbation, erosional forces (including
recreational or commercial boating activities,
scour, etc.) and chemical resistance, the inert and
"geologic" nature of the prime components would
provide extensive service.
An extended O&M activity for the cap would be to
complete minimal site inspections at one-, five-,
and perhaps ten-year intervals to ascertain the
condition of the cap, potentially in conjunction
with other monitoring programs (such as body
burden analyses) that would likely be common to
all major sediment remediation projects
regardless of the remedy selected. Table 4-1
includes estimated costs to conduct a post-
capping review at one-, five- and ten-year
intervals. Due to the relatively low per-event
costs and short overall duration of the O&M
program, no net present value analyses were
considered.
As with any remedy, if the cap were not designed
to incorporate in situ degradation or permanent
chemical treatment to render COCs harmless to
the environment within a specified remedy life,
more costs would be incurred for longer term site
inspection.
Maintenance, where required, could be
accomplished without significant effort if the cap
were damaged in a small area by adding
additional material to the area. As positive
attributes of AquaBlok® include its ability to self-
heal and self-compact, extensive maintenance
would not be expected.
4.2.4 Full-Scale AquaBlok® Cap
Installation Cost Analysis Summary
The material and installation cost for the assumed
"typical" AquaBlok® cap placed over a 10-acre
site would be approximately $1,695,260
($1,247,500 for AquaBlok® and sand, and
$447,760 for installation) for an average cost of
$3.90/ft2. These costs are summarized in Table
4-1. Note that Table 4-1 provides all unit cost
information to allow a basic comparison between
the cost of a "typical" AquaBlok® cap and a sand-
only cap, bearing in mind that the thickness, and
accordingly volume, of a sand-only cap intended
to accomplish the same RAOs as this "typical"
AquaBlok® cap would likely be significantly
greater (i.e., a sand-only cap may need to be 1 ft
or more thick).
A summary of costs in component square foot
and percentage of total project cost, including
construction QC, engineering design, permitting
and contract administration, is provided as
follows:
Component Cost/ft2
AquaBlok® $2.45
sand $0.41
installation $1.03
construction QC $0.36
design/permitting $0.43
TOTAL $4.68
% Project Total
52
9
22
8
9
100
While the costs for this "typical" capping scenario
and the real costs incurred during the SITE
demonstration can be used as a reasonable basis
to demonstrate a site-specific range of costs for
any given project, the reader is reminded that, as
with any project, costs will vary on the basis of
project-specific factors.
The assumed "typical" scenario is neither
conservative nor liberal as a representative
project. For example, if a significant portion of
the area could be reached from shore using a
conveyor approach, application rates could be
lower and produce significantly lower costs.
Similarly, a larger project and favorable terrain
might support the construction of a simple chute
to load supply barges directly from the shore
without the need for a conveyor and operator for
this purpose. If a project is located in a major
port area where larger supply barges can be
mobilized effectively, the application time for the
assumed 10-acre site could be reduced
significantly using the same level of equipment
122
-------
and labor, with corresponding reductions in
equipment rental and labor cost.
Material costs could increase significantly if
remote manufacture is not appropriate and
packaging and long-haul shipping are required.
In addition, other freshwater blends and saline
blends of AquaBlok® are more expensive than the
"typical" 3070 FW material considered. Similarly,
the use of AquaBlok Gate™ or AquaBlok+™ (see
Appendix A), which can provide in situ chemical
treatment of contaminated sediments, could
increase material costs by 20-40%. The cost of
an organoclay modified AquaBlok® can be even
greater.
Conversely, the use of a lesser thickness of
AquaBlok® (and more sand), or the use of an
AquaBlok Blended Barrier™ approach (see
Appendix A) could significantly reduce material
costs (by as much as 40%).
Finally, as with any technology, the maturation of
the AquaBlok® technology and its potential
selection and performance at contaminated
sediment sites over time would most certainly
lead to free market impacts of supply and
demand that could ultimately influence its cost.
The precise impact of free market forces on unit
cost for deployment of AquaBlok® cannot be
predicted, but it is likely that if this approach were
to become used more commonly, costs would be
driven down.
It is anticipated that lower equivalent material
costs, mobilization costs and shipping costs
would be experienced, as with most new
technologies, at a point after multiple applications
were completed and economics of scale of
manufacturing and distribution realized, including
on-site production. Similarly, if different
AquaBlok®-based designs are implemented and
used in conjunction with other materials for a
wider range of application goals, it is likely that
overall AquaBlok® costs would come down. The
tendency to overdesign (i.e. thicker than
necessary cap layers), due to lack of experience,
could continue to contribute to higher costs in the
near term, but as hypothetically more projects are
deemed successful, designers and engineers will
undoubtedly push the design envelope to provide
solutions relying on minimal product or leaner
mixtures that result in ultimately acceptable
remedies. The Blended Barrier™ approach (see
Appendix A) of blending AquaBlok® with
specifically sized untreated aggregate (obtained
in the locale of the project) to provide a lower
permeability barrier is a clear example of
improvements to the technology that could result
in overall lower project costs by using the main
AquaBlok® product in a more efficient manner,
where warranted.
Finally, as the demand for material becomes
more consistent and continuous, with multiple
larger projects or repeated smaller projects,
AquaBlok, Ltd. would likely develop customized
equipment for more efficient production of
products. The development of new application
techniques by the construction industry in general
can accomplish installation more simply, with
more uniform and efficient applications, perhaps
further reducing material needs while also
resulting in lower installation and QA/QC costs.
123
-------
Section 5
Demonstration Conclusions
The overall performance of AquaBlok® as an
innovative contaminated sediment capping
technology was evaluated in the context of three
primary SITE demonstration program study
objectives through an extensive field assessment
program implemented in the Anacostia River in
Washington, DC over the course of approximately
three years. Four individual field sampling events
were conducted (i.e., one, six, 18, and 30 months
following cap construction), and a multitude of
individual sampling tools and monitoring devices
were utilized.
The data generated during the SITE
demonstration suggest that AquaBlok® material is
highly stable. Oceanographic surveying (i.e.,
bathymetry and side-scan sonar surveying)
indicated that the AquaBlok® cap was not
substantially physically altered in any way during
the three year evaluation period. In fact, over the
course of the demonstration, it appears that fine,
organic-rich new sediment was deposited in the
demonstration area, effectively increasing the
overall thickness of the sediment caps, albeit
slightly and generally at a magnitude consistent
with the inherent resolution of the oceanographic
measurement tools. In the specific demonstration
area environment, the sand-only cap and even
native sediments were generally physically
unaffected during the course of the
demonstration, which would suggest that
AquaBlok® may not have a distinct advantage in
this particular environmental setting and relative
only to the measure of physical stability provided
by oceanographic surveying. SPI data as well as
sediment coring and laboratory physical
parameter data further confirmed the integrity of
the AquaBlok® and sand capping materials in the
specific demonstration area environment by
demonstrating a consistent grain size distribution
versus depth in the capped areas. However,
Sedflume analyses conducted during the
demonstration indicate that AquaBlok® is more
resistant to shear stress compared to traditional
sand capping material. Moreover, the Sedflume
data indicate that AquaBlok® is a highly
competent material and is unlikely to be eroded
even at very high shear stresses consistent with
very high flow conditions that are uncharacteristic
of the generally sluggish Anacostia River
demonstration area environment.
COC data generated during the sediment coring
suggest there was ongoing deposition of new
sediment on the capping cells that contained
contamination. The sediment coring data
suggest that newly deposited sediment in the
Anacostia River contained detectable levels of all
of the primary COC classes, which is not
surprising given the location of the site in a highly
urbanized/industrialized portion of Washington,
DC. The specific source of this contamination
was not studied (e.g., suspended sediment from
areas outside the demonstration area being
deposited in the study area through natural
hydrodynamics in the river, inputs from ongoing
diffuse urban pollution, and/or redeposition of
sediment suspended during actual capping
activities).
In addition, there may have been some relatively
minor increased rate of contaminant flux from the
underlying native sediment into the basal portion
of the sand cap as compared to the AquaBlok®
material, as evidenced by generally higher and
more consistently detectable concentrations of
PAHs and PCBs at the base of the sand cap
compared to the base of the AquaBlok® cap.
However, this observation was not specifically
verified using statistical testing, and given the
generally very low levels of contamination present
in the basal portion of the sand and AquaBlok®
capping materials, this conclusion is not
necessarily indicative of variability in the potential
124
-------
for contaminant movement upward through the
two cap types. In addition, this observation is
somewhat complicated with respect to metals and
the possibility that AquaBlok® material
preferentially sorbs metals and/or itself contains
metal constituents dependant on its source
material (i.e., the particular clay used to
manufacture the product). Given the strong
sorption capacity of AquaBlok® material, it may
also bind certain organic and inorganic
contaminants. Alternatively, given its strong
sorption capacity, and assuming contaminant flux
into both AquaBlok® and sand were occurring,
AquaBlok® may be more effective at preventing
the subsequent breakthrough of contamination
and exposure of sensitive ecological or human
receptors.
The gas flux data generated through the SITE
demonstration, while limited in utility due to
several field deployment and retrieval issues and
the likely inability to capture all potential vapor
phase flux across the study area, appear to
indicate that AquaBlok® is characterized by little
to no net flux through gas ebullition while the
traditional sand capping material is characterized
by at least some flux. In addition, it appears at
least plausible that AquaBlok® is capable of
retarding the movement of certain vapor phase
(i.e., sulfur-based) compounds while this same
effect was not observed for sand. While a
quantitative analysis of gas flux data was
accomplished, it is noteworthy that the gas flux
study design could not have captured all gas
ebullition potentially occurring in the study area
and may have specifically not targeted areas
where increased gas ebullition was occurring,
and should therefore be evaluated in the context
of this significant uncertainty.
In terms of the ability of AquaBlok® to prevent
seepage, hydraulic conductivity measurements
indicate AquaBlok® is highly impermeable and far
more impermeable relative to traditional sand
capping material, which is not surprising given the
very different grain-size composition of these two
capping materials. In addition, while the data
suggest AquaBlok® is characterized by hydraulic
conductivities generally similar to those in native
Anacostia River sediments, it appears likely that
the intrinsic permeability of AquaBlok® is lower
than the native sediments given the greater
potential for preferential flow paths in native
sediments related to biogenic activity and the
claimed ability of AquaBlok® to heal. Seepage
meter testing generally confirms this conclusion,
with visual evaluation of the data indicating that
aqueous flux through AquaBlok® was lower than
through traditional sand capping material.
Moreover, aqueous seepage through AquaBlok®
was determined to be, on average, vertically
downward from surface water to sediment as
opposed to from sediment to surface water (i.e.,
as with sand). The seepage meter data actually
appear to potentially indicate that traditional sand
capping material may have acted in some way to
exacerbate fluid flow through sediment during the
demonstration, although no mechanism for this
effect was directly observed or studied, and this
observation may actually have been an artifact of
fluid flow diversion from beneath the AquaBlok®
cap to the adjacent sand cell. Overall, the
seepage data did not exhibit a statistical trend
that clearly indicated a difference between the
performance of AquaBlok® and sand, but the
weight of evidence gathered through the
demonstration (including an evaluation of the
seepage data from a purely empirical
perspective) does appear to suggest AquaBlok®
would be a more effective barrier to fluid flow.
With respect to benthic ecology, the surface
sediment environment within the tidally influenced
freshwater environment at the site appeared to be
controlled by physical processes and did not
appear to support a highly structured benthic
community. Biogenic activity, therefore, did not
appear to be a factor, negative or positive, in cap
longevity. Accordingly, SPI data were of only
limited power in directly understanding ecological
recovery rates or impacts of capping material on
benthic communities. Benthic organisms in the
system appeared from the SPI imagery to have
been dominated by very small oligochaetes.
Specific assays and statistical evaluation of
benthic habitats and faunal communities in the
AquaBlok® and sand cells indicated that small
deposit-feeding organisms were dominant in the
demonstration area and that the AquaBlok® and
sand cells were more similar to each other than
the control cell in terms of ecology. However,
ecological assemblages in the AquaBlok® and
125
-------
sand cells retained differences that clearly
separated them from each other. In particular,
AquaBlok® demonstrated relatively equal or
greater abundances of individuals within the
major taxonomic groups found to also occur in
the sand capping material in the sand cell. This
could be taken to indicate that the AquaBlok® cell
was a more suitable habitat for benthic
recolonization, perhaps by providing a more
effective barrier against porewater contaminant
flux into surficial sediments where most benthos
occur, or by being a more similar grain size
relative to existing native sediments. However,
the AquaBlok® cell was covered by a sand layer,
and the benthic sampling was conducted in
surface sediments. It is therefore difficult to infer
from the available data what impact AquaBlok®
itself had on benthic recovery as the benthic
comparisons were essentially between the same
cap material (i.e., sand in the sand cell and sand
covering material in the AquaBlok® cell). It can,
however, be inferred from the data generated that
AquaBlok® does not appear to have a detrimental
effect on benthic recovery.
The SITE demonstration of the AquaBlok®
technology was designed to answer fundamental
questions about its performance relative to more
traditional sediment capping material. In
answering these questions, a significant amount
of data collection and data analysis were
conducted. While the data collection and data
analysis were robust and appropriate, the data
were not necessarily evaluated in every fashion
possible. In addition, there are obvious field data
collection issues and inherent data uncertainties
that limit the usefulness of certain data and the
power of certain evaluations and interpretations,
and the conclusions of the demonstration must be
reviewed in that context.
The results of the SITE demonstration do open
for consideration several complimentary lines of
questioning and potentially beneficial avenues of
further study. For instance, additional study may
be warranted to determine if AquaBlok® is
susceptible to significant failure from the buildup
of gas pressure and subsequent short-circuiting
through preferential pathways or catastrophic gas
releases. In addition, it is possible that
AquaBlok® material could act to divert
contaminant flux (fluid or vapor phase) to the
periphery of a capped area, potentially biasing
and concentrating the flux of contamination in
discrete locations even beyond the original
contaminant footprint (note there are sites where
a net neutral flux could be the equilibrium
condition, meaning that an impermeable cap
would likely not lead to lateral contaminant
diversion; also, as described in Appendix A, there
are commercially available forms of AquaBlok® or
formulations in development that could potentially
counteract the lateral diversion of contaminant
flux by integrating reactive components or "funnel
and gate" concepts). Ice scour and freezing
conditions are generally acknowledged to be a
potential limitation of sediment capping
alternatives, and understanding the full effect of
ice-related conditions on an AquaBlok® cap could
be a critical developmental need. Moreover, a
more complete understanding of benthic recovery
would likely be gained by assessing community
structure over more time than three years, and
potentially by assessing an AquaBlok®-only cap
instead of a sand covered AquaBlok® cap. It was
not an objective of the SITE demonstration to
evaluate these potential phenomena.
126
-------
Section 6
References
Alcoa. 2007. Grasse River information. Available
at: http://www.thegrasseriver.com.
Anacostia Watershed Society. 2007. Anacostia
River information. Available at:
http://www.anacostiaws.org.
Battelle Memorial Institute (Battelle). 2004.
Quality Assurance Project Plan for the SITE
Demonstration of the AquaBlok® Sediment
Capping Technology at the Anacostia River,
Washington, DC. Prepared for EPA NRMRL.
March.
Clarke, K.R. and R.M. Warwick. 2001. Change
in Marine Communities: An Approach to
Statistical Analysis and Interpretation.
Plymouth Marine Laboratory, Plymouth, UK
859 p.
Diaz, R.J. and L.C. Shaffner. 1988. "Comparison
of Sediment Landscapes in the Chesapeake
Bay as Seen by Surface and Profile Imaging."
M.P. Lynch and E.G. Krome, eds.,
Understanding the Estuary: Advances in
Chesapeake Bay Research. Chesapeake
Research Consortium Publication 129,
CBP/TRS 24/88. pp. 222-240.
Gailani, J.Z., L Jin, J. McNeil, and W. Lick. 2001.
"Effects of Bentonite on Sediment Erosion
Rates." DOER Technical Notes Collection
(ERDC TN-DOER-N12), U.S. Army Engineer
Research and Development Center, Vickburg,
MS. Available at:
www.wes.army.mil/el/dots/doer.
Home Engineering Services, Inc (Home). 2004.
Revised Draft Cap Completion Report for
Comparative Validation of Innovative "Active
Capping" Technologies, Anacostia River,
Washington, DC.
December.
Prepared for HSRC.
Hull & Associates, Inc. 2002. Summary Report
of the Ottawa River AquaBlok® Application
Demonstration.
Hurlbert, S.H. 1971. "The Nonconcept of
Species Diversity: A Critique and Alternative
Parameters." Ecology 52:577-586.
Ludwig, J.A. and J.F. Reynolds. 1988. Statistical
Ecology: A Primer on Methods and
Computing. John Wley & Sons, New York,
New York.
May, R.M. 1975. "Patterns of Species
Abundance and Diversity." M.L. Cody and
J.M. Diamond, eds., Ecology and Evolution of
Communities. Belknap Press, Cambridge,
MA, pp. 81-120.
McNeil, J., C. Taylor, and W. Lick. 1996.
"Measurements of Erosion of Undisturbed
Bottom Sediments with Depth." Journal of
Hydraulic Engineering 722(6): 316-324.
Mermillod-Blondin, F., M. Gerino, V. Degrange,
R. Lensi, J.L. Chasse, M. Rard, and M.C. des
Chatelliers. 2003. "Testing the Functional
Redundancy of Limnodrilus and Tubifex
(Oligochaeta, Tubificidae) in Hyporheic
Sediments: An Experimental Study in
Microcosms." Canadian Journal of Fisheries
and Aquatic Sciences 58(9): 1747-1759.
Paulsen, R.J., C.F. Smith, D.E. O'Rourke, and T-
f. Wong. 2001. "Development and Evaluation
of an Ultrasonic Groundwater Seepage
Meter." Groundwater Nov-Dec Vol. 39 No. 6:
904-911.
127
-------
Pielou, E.G. 1966. "Species Diversity and
Pattern Diversity in the Study of Ecological
Succession." Journal of Theoretical Biology
10:370-383.
Rhoads, D.C., and J.D. Germane. 1986.
"Interpreting Long-Term Changes in Benthic
Community Structure: A New Protocol."
Hydrobiologia 142: 291-308.
Roberts, K.L. 2004. "Modeling of River
Hydrodynamics and Active Cap Effectiveness
in the Anacostia River." Master's Thesis.
Louisiana State University.
Sanders, H.L 1968. "Marine Benthic Diversity: A
Comparative Study." American Naturalist
102:243-282.
U.S. Geologic Survey (USGS). 2006. River
stage information. Available at:
http://waterdata.usgs.gov/nwis.
U.S. Environmental Protection Agency (EPA).
2000. "Field-Scale Testing of a Composite
Particle Sediment Capping Technology."
Tech Trends February.
U.S. Environmental Protection Agency (EPA).
2006. Lower Fox River/Green Bay Site
Technical Memorandum, Current Plan and
Proposed Plan. November.
Wang, X-S. and G. Matisoff. 1997. "Solute
Transport in Sediments by a Large
Freshwater oligochaete, Branchiura
sowerbyi." Environmental Science &
Technology 31 (7): 1926-1933.
Yuan, 2007. "Experimental and Modeling Studies
of Contaminant Transport in Capped
Sediments during Gas Bubble Ebullition."
PhD Dissertation. Louisiana State University.
128
------- |