Technology
Transfer for Bioremediation of Complex Environmental
Systems
Peter Adriaens, The University of Michigan, Ann arbor, MI
Confirmatory
Sampling Approach for New Bedford Harbor
Ronald J. Marnicio, Foster Wheeler Environmental
Corporation, Boston, MA
The
Role of Risk Management Decisions in the Development of
Sediment Quality Guidelines
Julie A. Rothrock and Paul D. Anderson, AMEC Earth and
Environmental, Westford, MA
Modeling
and Mapping of Sediment Contaminant Distribution: A
Univariate Geostatistical Perspective
Kandiah Ramanitharan, Tulane University, New Orleans, LA
Dr. Laura J. Steinberg, Tulane University, New Orleans, LA
Compositional
Changes in Nearfield and Farfield Massachusetts Coastal
Sediments Attributed to the MWRA Wastewater Treatment
Plant: A Comprehensive Comparison of Pre- and
Post-Discharge Periods
Stephen
Emsbo-Mattingly, Deirdre Dahlen, Carlton Hunt, Battelle
Memorial Institute, Duxbury, MA
Ken Keay, Ph.D., Massachusetts Water Resources Authority
Diffusion
Approach to Model Desorption Resistance of Organic
Contaminants in Soils/Sediments
Yunzhou
Chai, Louisiana State University, Baton Rouge, LA
Danny Reible, Louisiana State University, Baton Rouge, LA
Alexander Kochetkov, Louisiana State University, Baton
Rouge, LA
Technology
Transfer for Bioremediation of Complex Environmental
Systems
Peter
Adriaens, Professor, Environmental and Water Resources
Engineering, Department of Civil and Environmental
Engineering, The University of Michigan, Ann Arbor, MI
48109-2125, USA
During
the last decade, advances in our understanding of
biodegradation principles, and lessons learned from
bioremediation applications, have increased our knowledge
base to help transition laboratory-based processes to the
field. Within the group of organic anthropogenic compounds,
chlorinated aromatics, chlorinated volatile organic
compounds, and petroleum hydrocarbons are among the most
frequently encountered pollutants in soils, groundwater
aquifers and sediments.
Whereas the scientific knowledge base regarding
process understanding is maturing, the transitioning of
this information to effectively remediate contaminated
sites at best has accumulated a spotty record.
This
treatise will illustrate the guiding principles of
multi-disciplinary project management using
well-documented success stories from groundwater and
surface water sediment systems remediation.
Successful technology transition hinges on: (i)
appropriate definition of technology transfer objectives
and needs; (ii) formulation of success metrics; (iii)
seeking out the required integration of team expertise;
(iv) a priori agreement on team expectations and
communication levels. Emphasis will be placed on the
challenges posed by, and methodologies used during, the
transfer of biodegradation principles to field
application. Whereas
the fundamental processes of biodegradation have been well
established, harnessing these processes for in situ
application is fraught with substantial uncertainty.
These uncertainties are mainly due to site
specificities associated with expression of the desired
biodegradation activity, the difficulty with establishing
causal relationships between biological activity and
geochemical (organic and inorganic) changes, and the lack
of data for a convincing cost-benefit analysis associated
with bioremediation applications.
Successful technology transfer examples will be
based on the development of bioreactive barriers for
control of chlorinated solvents, and the design of
remedial units for dioxin-contaminated surface water
sediments based on the interpretation of natural
attenuation patterns.
Despite
well-documented success stories in the field, the required
design parameters that would allow engineers to
effectively use the proposed technologies are generally
not available. Moreover,
there is no general agreement on the metrics of success
(chemical and biological analysis, soil quality
indicators) of bioremediation technologies, leaving
informed decision making by regulators a difficult task
which can at best be accomplished on a site-specific
basis. Finally,
there is an urgent need to incorporate uncertainty
analysis to bridge site assessment and laboratory-based
data with the purpose of developing a risk-based decision
framework for technology implementation.
Confirmatory
Sampling Approach for New Bedford Harbor
Ronald
J. Marnicio, Ph.D., PE, Foster Wheeler Environmental
Corporation, 133 Federal St., 6th Floor, Boston, MA 02110, Tel: 617-457-8262, Fax: 617- 457-8499, Email:
rmarnicio@fwenc.com
Foster
Wheeler, under the U. S. Army Corps of Engineers New
England Total Environmental Restoration Contract and in
conjunction with the U. S. EPA, is conducting engineering
design, remediation, and restoration activities at the New
Bedford Harbor (NBH) Superfund Site.
Sediments at NBH were contaminated with PCBs as the
result of releases from industrial operations over many
years. A
sediment characterization program was performed to support
the development of dredging and excavation plans to bring
NBH into compliance with the target PCB levels specified
in the Site’s Record of Decision.
A confirmatory sampling (CS) program is being
implemented to test and demonstrate that post-removal
sediment quality meets the specified clean-up targets.
This program is unique in scale, complexity and
duration relative to CS approaches that have been applied
in smaller, simpler projects.
This
presentation describes the design, development and initial
implementation of the NBH CS approach.
The CS objectives are identified, and CS is
contrasted with sampling for other purposes.
A survey of recent similar environmental dredging
projects in the U.S and abroad was used to establish best
practices relative to CS in this context, and identify
possible precedents relative to many CS program
components. A
statistically based Data Quality Objective approach was
used to define data sufficiency for compliance testing.
Aspects of sample collection, depth and spatial
pattern of sampling, and the analytical methods and
procedures to be applied were evaluated and specified.
The direct relationship between the CS components
and the approaches used to remove and manage the
contaminated sediments is highlighted.
Also discussed are a number of special conditions
that must be accommodated by the CS program to ensure the
success and credibility of the CS effort and the overall
project. A
status on the application of the CS approach in the
removal effort is presented.
The
Role of Risk Management Decisions in the Development of
Sediment Quality Guidelines
Julie
A. Rothrock and Paul D. Anderson, Ph.D AMEC Earth and
Environmental, 239 Littleton Road, Suite 1-B, Westford,
Massachusetts 01886, Tel: 978-692-9090,
Fax: 978-692-6633
Sediment
quality guidelines (SQGs) are chemical concentrations in
sediments that are assumed to be associated with no or
acceptable levels of adverse effects in benthic organisms.
They have been proposed and published for numerous
chemicals using various methodologies over the past few
decades, and their use has inspired considerable debate
within the scientific community. One commonly applied
methodology for developing SQGs is the compilation and
combination of various databases containing sediment
chemistry measurements and co-located effects in aquatic
organisms (i.e., Effects Range-Low [ER-L] and Effects
Range-Medium [ER-M] values used in the National
Oceanographic and Atmospheric Administration Status and
Trends Program). This methodology has been adapted
recently to derive "consensus-based" SQGs (such
as threshold and probable effect concentrations [TECs and
PECs]) by compiling and combining previously derived SQGs.
The derivation of such SQGs incorporates numerous unstated
risk management decisions including the selection of which
SQGs to use in the calculation of
"consensus-based" SQGs, the choice of method
used to determine central tendency, and the interpretation
of toxicity test results (such as evaluating multiple
species, multiple endpoints, and control versus reference
locations). Moreover,
the application of SQGs to contaminated sediment
remediation is also based on unclear risk management
decisions. As a result, rather than providing the means
for an objective evaluation of sediments, these SQGs and
their application incorporate a great deal of
subjectivity. The implications of using such SQGs in
ecological risk assessments of aquatic systems are
discussed.
Modeling
and Mapping of Sediment Contaminant Distribution: A
Univariate Geostatistical Perspective
Kandiah
Ramanitharan, Department of Civil & Environmental
Engineering, Tulane University, New Orleans, LA 70118,
Tel: 504-865-7313, Fax: 504-862-8941
Dr. Laura J. Steinberg, Department of Civil &
Environmental Engineering, Tulane University, New Orleans,
LA 70118, Tel: 504-862-3254, Fax: 504-862-8941
Characterization and remediation of a contaminated sediment
site depends on the sampling strategies and the consequent
contamination modeling. This paper presents an exploratory
study on the applications of geostatistical techniques in
mapping sediment contaminant distributions in water
bodies. How the physical shape of the water bodies,
hotspots and multiple data measurements per data point
influences the geostatistical modeling is analyzed using
sediment contaminant concentration data sets from five
water bodies (Seal Beach, Tampa Bay, Hudson river,
Duwamish river and Lake Geneva). The datasets from the
first two sites are randomly sampled, and the others are
of near-grid sampling design. In the geostatistical
models, whenever a concentration data set is of non-normal
distribution, an appropriate data transformation
(log-normal, or Box-Cox, or Normal Score Transformation)
is implemented. Micro-spatial correlation is modeled with
Spherical, Gaussian, Exponential, and Nugget structures.
Geostatistical models are developed to incorporate the
noise due to the multiple measurements at data points.
Mean Squared Error (MSE), Mean Error (ME) and prediction
kriging standard errors of cross validation are used as
the best model parameter-selecting criteria for the
models. It is found that trend and anisotropy are much
influenced by the shape of the water body. When the
measurement noise is high or the sampling density is low,
all geostatistical models perform more or less the same.
Representing hotspots only with a few extreme metal
concentrations has a negative impact in fitting a
geostatistical model. Selection of data to fit a
geostatistical model for the water bodies with physical
barriers is discussed. The use of studying cokriging and
the use of geostatistical models in the sampling design
are identified as the future research areas.
Compositional
Changes in Nearfield and Farfield Massachusetts Coastal
Sediments Attributed to the MWRA Wastewater Treatment
Plant: A Comprehensive Comparison of Pre- and
Post-Discharge Periods
Stephen
Emsbo-Mattingly, M.S., Deirdre Dahlen, Carlton Hunt,
Ph.D., Battelle Memorial Institute, 397 Washington Street,
Duxbury, MA 02332, Tel: 781-934-0571
Ken Keay, Ph.D., Massachusetts Water Resources Authority,
100 First Avenue, Charleston Navy Yard, Boston, MA 02129,
Tel: 617-242-6000
The
Massachusetts Water Resources Authority (MWRA) planned and
built a large state-of-the-art wastewater treatment plant
on Deer Island during the 1980’s and 1990’s to abate
the discharge of untreated and partially treated sewage to
Boston Harbor. Effluent
from the plant is discharged through a pipe running nine
miles offshore to further improve the water quality in
Boston Harbor and facilitate the dilution of effluent with
minimal impact to the offshore environment.
MWRA commissioned a comprehensive testing program
of Massachusetts coastal sediments to monitor potential
impacts of effluent from the Deer Island Plant.
Monitoring stations were selected near Boston
Harbor and the new outfall (nearfield) and throughout
Massachusetts and Cap Cod Bays (farfield).
This presentation discusses the compositional
changes in selected sediment parameters that occurred
during the pre- and post-discharge periods (1992 to 2000
and 2001 to 2002, respectively).
Baseline
data collected in Massachusetts Bay from 1992 to 2000
showed multiple regions defined by physical and chemical
composition. Within
the MWRA nearfield stations of Massachusetts Bay, there is
a series of heterogeneous sediments in relatively close
proximity to the primary historic source of contaminants
(i.e., Boston Harbor).
Nearfield stations were generally equidistant from
the harbor. The
major factors influencing the concentration of
contaminants and sewage tracers are primarily related to
grain size, which suggests different sediment depositional
environments. The
primary factor explaining the variance in the data is sand
content, which correlated inversely with organic and
inorganic analyte concentrations. The secondary factors were associated with anthropogenic
analytes (e.g., chlorinated pesticides and Cadmium), which
were measured at higher levels in the early 1990s, and
fine particles.
In
contrast, the sediments collected from the farfield
stations are more spatially dispersed and compositionally
distinct. As
in the nearfield data, sand content strongly influenced
the variance in the farfield data.
In addition, the station proximity to Boston Harbor
(the historic source of sewage contaminants) influenced
the concentration of tracers of sewage, Clostridium
perfringens and total linear alkyl benzenes (LAB), at
farfield stations. The
distribution of fine particles and the analytes associated
with them (total organic carbon and selected metals) also
helped compositionally define some of the more remote
sampling locations. In
short, the composition of sediments at sampling locations
distant from Boston Harbor may reflect inputs to the
sediment that are distinct from the historic sewage
discharge from Boston Harbor.
Thus, the sediments in Masschusetts Bay are defined
by two compositionally distinct areas.
Results
from correlation analysis between contaminants and bulk
sediment properties (percent fines and TOC) also support a
system with two disparate regions.
Correlations between contaminants and bulk sediment
properties in the nearfield are relatively high, with r2
of 0.5 or greater for most parameters.
The correlations were generally weaker for farfield
stations (organic contaminants in particular), further
suggesting that the primary controlling variables in the
farfield system are governed by both proximity to Boston
Harbor and by regional depositional patterns.
On
average, contaminant concentrations remained relatively
constant over the past decade and were well below MWRA
(2001) monitoring thresholds.
In contrast, concentrations of the sewage tracers, Clostridium
perfringens and total LAB, decreased in the late 1990s
for stations in Massachusetts Bay located closer to the
Harbor. The
observed decreases can be attributed to a variety of
factors including various MWRA facility upgrades, which in
turn have resulted in reduced discharges of effluent
solids. Results
from 2001 and 2002, the first two years since activation
of the new outfall, showed a general increase in Clostridium perfringens
abundances across the entire nearfield, with largest
increases observed at those stations located within 2-km
of the western end of the new outfall.
These findings suggest that effluent discharge at
the new outfall is having a modest effect on nearby
sediments.
Diffusion
Approach to Model Desorption Resistance of Organic
Contaminants in Soils/Sediments
Yunzhou
Chai, Louisiana State University, Hazardous
Substances Research Center South/Southwest, Department of
Chemical Engineering, Louisiana State University, Baton
Rouge, LA 70803m Tel:
225-578-4072, Fax: 225-578-1476
Danny Reible, Louisiana State University, Hazardous
Substances Research Center South/Southwest, Department of
Chemical Engineering, Louisiana State University, Baton
Rouge, LA 70803, Tel:
225-578-6770, Fax: 225-578-1476
Alexander Kochetkov, Louisiana State University, Hazardous
Substances Research Center South/Southwest, Department of
Chemical Engineering, Louisiana State University, Baton
Rouge, LA 70803, Tel:
225-578-4072, Fax: 225-578-1476
Desorption resistance of
organic contaminants in soils and sediments is often
observed, which results in a relatively rapid release and
degradation of organic contaminants in soils and sediments
initially followed by a period of slow change. The
desorption resistance has been attributed to the
sequestration of the contaminants in the soils or
sediments organic matrices, which has been attributed to
effects of different sorbing nature of various soil or
sediment organic matrices, and conformational or other
changes of the contaminants in the matrices. The objective
of this work is to compare various models of desorption
resistance in an attempt to develop a consistent model for
partitioning of organic contaminants in soils or
sediments. Diffusion of contaminants in various sorbent
matrices is utilized to model the desorption resistance,
considering different physical/chemical properties such as
porosity and capacity of various soil or sediment organic
matrices. This diffusion-based model is able to predict
the observed desorption isotherms, desorption resistance
and aging effects. It is a truly predictive model for
availability of contaminants in assessing potential
exposure and risks from contaminated sediments. Detailed
model development and results as well as experimental data
will be presented.
Vacuum
dredging was performed in the fall of 2002.
The permits required for the project contained
numerous conditions which presented challenges prior to
the initiation of vacuum dredging activities including: no
fill was allowed in the resource areas, diesel motors were
required to have after-engine exhaust controls, and there
were strict filtering requirements for dewatering fluids.
In addition, numerous challenges arose during
project execution: worker mobility was severely hampered
by soft sediments, a combination of rainstorms and
unexpected dissolved copper in the pond presented water
management issues, the consistency of the sediments
significantly slowed progress, cold weather freezing
issues arose as the schedule slipped, a beaver established
itself downstream backing the stream up into the work
areas, and numerous QA/QC issues inherent with sediment
sampling arose. All
of these challenges were successfully resolved during the
course of the project.
Upon
completion, a total of approximately 491 tons of sediment
were vacuum dredged.
Post-excavation sediment sampling results indicate
that remediation goals were met and
post-remediation surface water sampling
demonstrated dissolved metals were below ambient water
quality standards. Final
wetlands restoration will be completed in the spring of
2003. A permanent solution for the site was achieved.
Top
|
