Overview
We received for review the Brunswick Wood Preserving Site Final Design
Report dated October 14, 2004. This document discusses in detail the planned
project for remediating the Brunswick Wood Preserving Site.
The Design follows previous models for incorporating toxic site soils, and
the remaining toxic waste piles, into two caps for reducing rainwater
movement. The caps would be placed on top of old pond areas where
contamination remains in contact with groundwater. Barrier-wall impoundments
would be constructed around underground waste to slow release of site
contaminants into groundwater.
In addition to the Design this document includes the results of recent
studies at the site, and previously unavailable discussion of the site
groundwater and underlying geology. The document provides the most complete
view of the site yet released.
The Environmental Protection Agency stated in the Record of Decision and at
a formal Public Meeting that the limestone layer beneath the site would act
as a fully confining floor for the proposed barrier walled impoundments. It
appears from the actual data this presumption was too optimistic and the
limestone layer is not full protection for the lower aquifers.
Site History
The Brunswick Wood Preserving Superfund site (“BWP” or the “site”) is an old
wood treatment facility. The plant produced creosote and CCA
(copper-chromium-arsenate) soaked wood products from local timber. A large
contaminated area includes heavy metals, aromatic hydrocarbons (some are
known to cause cancer) and dioxins. Cleanup operations by both the U.S.
Environmental Protection Agency (EPA) and the State of Georgia left
considerable toxic waste at the site. Two toxic plumes of groundwater flow
underground from the site. Toxic releases into adjacent Burnett Creek are
still common. There have been numerous studies and meetings for this site.
A Record of Decision (ROD) and Public Meeting were held earlier to discuss
the remedy.
Overview of the
Proposed Remedy
There are six basic parts to the remedy proposed in the ROD:
1)
Construction of two underground confining areas in the old ponds by digging
trenches down to a limestone layer and backfilling with a slurry of
materials to form barrier “walled” impoundments;
2)
Consolidating contaminated on-site surface ground materials onto the top of
the confined pond impoundments;
3)
Digging out areas of contaminated creek sediments from Burnett Creek and
placing them within the consolidated areas on top of the old pond
impoundments;
4)
Placing a rainwater-shedding cover on top of the impoundment mounds;
5)
Treating remaining groundwater areas outside of the impoundments; and,
6)
Long-term monitoring and site maintenance.
Overview of the
Design Report
The Brunswick Wood Preserving site is not stable: there are groundwater
contamination plumes spreading west and south from the site; dense
non-aqueous layers (DNALs) are sinking and moving beneath the site; there is
surface runoff into Burnett Creek; there are groundwater releases to Burnett
Creek; there are contaminated materials in Burnett Creek moving downstream;
and, trespassers are exposed to site contamination through faulty security.
The proposed barrier wall impoundments will confine a large percentage-- but
certainly not all-- of the waste within the borders of the former pond
areas. This will reduce, not end, the release of toxins into the surrounding
groundwater. This is not a treatment. The toxins will still be present, but
over time it may reduce the amount of waste release. It is not practical
with the information provided to determine how much waste will continue to
enter the system or how soon residents will see a difference.
Building a rainwater-reducing cap over the impoundments will somewhat reduce
water movement. However, the cap is too small to really change the amount of
rainwater adding to the groundwater, so flow will not be impacted
significantly. Further, the Dense Non-Aqueous Layers of pollutants are
moving as a result of gravity, not driven by groundwater flow. The cap
should have no impact on that material.
Consolidating the surface soils is not a cleanup, merely moving waste from
one part of the site to another provides no real advantage. There may be
some benefit to animals using the site, although EPA has done a poor job
quantifying environmental exposure.
The proposed cleanup of Burnett Creek is by far the best part of the Design.
The actual removal of waste from this stream can have big impact on the
health of area waters.
The in situ treatment of organic wastes in groundwater is beneficial
to reducing the off-site plumes that now extend under the highway and
beneath adjacent properties. Combined with the slurry wall impoundments used
to reduce wastes the in situ treatment has the potential to reduce
off-site contamination. However, this is a future benefit; there will be no
short-term reduction in the footprint of site contamination.
Throughout the history of this site there have been reports the EPA was
unable or unwilling to maintain the integrity of site fences, post effective
signage, and control access. The proposed long-term site security plan
requires more work to build genuine barriers to entry.
Findings
The Design provides additional information on groundwater flow, toxin
movements and the geological makeup of the site that was not previously
available during the Remedial Investigation phase.
Plumes:
The plumes of contamination that originate in the former ponds extend much
further than previous documents indicated. One plume originates in the
IM-1/IM-2 pond area and the second plume begins in the IM-4/IM-5 area. The
larger of the two plumes is the Eastern Plume that starts in the IM-4/IM-5
pond area and moves easterly towards Perry Lane Road and Burnett Creek. This
plume contains significant amounts of polycyclic aromatic hydrocarbons (“PAH”’)
a mixture of many chemicals, some of which are known carcinogens. The plume
from the IM-1 area reaches Burnett Creek, but apparently does not pass it.
The best explanation for the data is that the toxins enter Burnett Creek. At
this time there are two few observations to provide velocity of contaminates
to the creek or to estimate chemical mass calculations.
Barrier Wall
Construction:
The purpose of the barrier walls (often called slurry walls) is to prevent
the horizontal plume movement of toxins in groundwater. The old ponds are
the source of toxins in groundwater plumes. Hydrocarbons (the oils in the
creosote used to waterproof the logs) and toxins (a variety of chemicals
were added to reduce rotting) left in the ground are moving both downward by
the pull of gravity and moving away from the site in the groundwater
currents flowing under the site. The downward vertical migration will not be
stopped by the building of barrier walls or caps. However, the horizontal
movement in groundwater could be stopped by creating large circular
containment walls underground.
To produce a slurry wall, a trench is dug from the surface down to a
limestone layer beneath the ponds. The trench would be 3 feet (about 1
meter) wide and form a complete ring around the waste. There would be two
such trenches, one around the ponds in the IM1/2 areas, the other around the
IM4/5 areas. The dimensions of the walls for the IM1/2 impoundment are
unchanged from the Record of Decision phase. For that impoundment, the
critical eastern boundary was defined by proximity of Perry Lane and could
not be expanded. For the IM4/5 ponds the wall was moved further outward to
better enclose high concentrations of groundwater contaminates.
The
slurry material poured into the trench to form the wall would be fill dirt
mixed with clays, such as bentonite. Fine clays have a high surface area and
are not very wettable—they do not take up water easily. Material that
becomes stuck to the surface of clay tends to stay attached and clay
particles neither shrink nor swell. Properly constructed man-made
clay-containing walls are known to be stable for generations.
The entire design calls for a rainwater-shedding cap over the impoundments
after construction of the underground walls. The thought by the EPA is that
with little rainwater entering the impoundment there should be little water
leaving the impoundment; accordingly, there should be no release of toxins
to the environment. However, rainwater and horizontal groundwater flow are
not the only potential sources of water that could enter the impoundments.
The floor of the impoundment will allow movement of water into and out of
the impoundments—unless there is a perfect seal between the upper and lower
aquifers. The data provided by EPA suggests that the limestone layer is not
such a perfect seal. In a best case scenario voids and areas of high
permeability within the limestone floor would allow water to migrate upwards
into the impoundment container, and then slowly out through the walls of the
impoundment. This would greatly reduce influx of contaminates into the upper
aquifer, probably at the levels where natural dilution would render them
harmless. A worse-case scenario is that water from the lower aquifer would
slosh up and down with fluctuations within the impoundment, exchanging clean
groundwater with contaminated groundwater and resulting in a deep plume
within the lower drinking water aquifer. A scientifically constructed
monitoring system with independent oversight is needed to assure the public
that this remedy works and is safe for the long-term.
Treatment of
Hazardous Plume Material:
After the slurry walls are complete and the areas beneath the ponds are no
longer sources of contamination there will still be toxic materials in
groundwater outside of the slurry wall impoundments. This material will
continue to flow off-site for the foreseeable future. The Design states the
groundwater toxicity will be reduced by injecting treatments into the
groundwater. This discussion remains unchanged from previous statements in
the ROD and at the public meeting. The rate at which pollutants would be
removed would depend on how aggressive the treatment is; the more injection
points and the higher the concentration of treatment reagents the faster the
plumes would be reduced to non-toxicity. Unfortunately, the Design does not
provide specific details on the treatment chemicals and concentrations, or
where injection pumps may be placed, or how many pumps would be used, or
what flow rates will be used. However, although the Design document does not
allow estimates of how fast the groundwater may be cleaned, underground
treatment of this type of toxin is a mature technology and should work.
DNAL Containment:
It would be fair to say that the success of this remedy relies on the
ability of the bottom of the containment impoundments to retain toxins.
According to the ROD, when a barrier walls is entrenched into the limestone
layer that forms the floor it should act like a giant bucket, with the
consolidation and cap forming the lid. There is little doubt the slurry
walls can be built as described, and landfill caps are well known
technology. However, there is no engineering to the bottom of the “bucket”,
it either works or it does not.
The Design document describes nine geological layers, numbered from the
surface down about 75 feet below the surface. Each “layer” is a collection
of sands, gravels, clays, rocks, and/or solid limestone that is
characteristic for the layer. The first six layers are not relevant to this
discussion since the bottom of the slurry wall could not form a seal in any
of those layers. For example, stratum 6 (layer 6) is described as dark sand
(with) little to some fine gravel, shell fragments (ancient mollusk shells),
and trace to little fines. Trenching into this sandy layer and backfilling
with slurry would have little effect on waste mobility, groundwater would
flow into the “impoundment” under the wall, and toxins would just continue
to flow out. Ideally, the stratum used for the containment should be like
cement, and should be like cement over the entire impoundment.
Stratum 7 is the layer described as ideal in the Record of Decision.
However, further studies show this layer is not consistently thick. Further,
stratum 7 in some areas is two layers and in others a single layer. EPA has
designated some areas to have layers 7a and 7b, where 7b is more like hard
limestone and 7a is limestone “weathered” to the consistency of gravel. Some
areas studied lack a stratum 7b. For example, on Design Sheet BW-3 there is
no layer 7b at the boring BRU-GT-3, only the weathered limestone layer 7a,
and it is not especially thick either, only a few feet thick there, and even
thinner at nearby boring BRU-GT-4. Note also that even the limestone of
Stratum 7b is described as having voids (open spaces with water) and areas
of imbedded sand, gravel and clay—it is not always solid rock. It is clear
the weathered limestone layer described in the Record of Decision is not
adequate protection as an impound floor.
To get around the problems with the limestone layer, EPA states the barrier
wall will pass through the limestone into layer 8, a stratum of mixed sands
and clays. Laboratory studies on the clays indicate this layer is a low
permeability layer, similar to material proposed for the slurry walls. If
the layer was thick enough and consistent enough it would form an effective
bottom to the slurry wall impoundment “buckets.” However, data casts doubt
on this for the IM4/5 impoundment, which raises questions regarding the
sufficiency of this layer for the entire project. In the IM4/5 impoundment
footprint some areas of stratum 8 are too thin and the layer contains sands
and materials that can conduct water. In these regions the barrier walls
might not form a good seal with the strata. It is important to note that it
is not just the intersection of the barrier wall and bottom layer that
results in a stable impoundment. Rather, the entire floor must be free of
channels, gravels, sands and other artifacts. Water follows the path of
least resistance underground as on the surface and the entire base must have
water resisting features or there will be movements of water into and out
through the impoundment floor. The variability of the strata and the large
size of the areas to be confined indicate that neither stratums 7 or 8 are
sufficient to guarantee waste confinement.
The Design plan calls for the engineer to make judgment calls during slurry
wall construction, ending it in either layer 7 or layer 8 depending on
results of the digging. In the Record of Decision and at the Public Meeting
there was no mention of such guesswork, the public was told unequivocally
the limestone layer was more than adequate and would work unconditionally.
It is likely the impoundments will be only a first step, and some form of
technology will be necessary to actually treat material within the
impoundments. EPA notes the possibility that additional treatment could be
needed, but the Design documents lack specific information on further
treatments or monitoring criteria that would trigger a decision to begin
treatment within the impoundments.
Burnett Creek
Cleanup:
During and after plant operation toxic material washed into Burnett Creek.
Some of this material has detoxified by natural processes—dilution,
biological degradation, photochemical breakdown, and similar processes. Some
“pockets” of contaminated sediment remain and must be removed. The estimates
provided in the Design documents indicate that 300 cubic yards must be dug
out of the creek and returned to the site for processing. This amount is
probably low. For reference, a small “dumpster” trash bin is about 2 cubic
yards, so the removal of 300 cubic yards is not a large volume of sediment
considering the length of the creek. However, that amount would aid the
creek’s recovery when combined with the site no longer being a source of
continued contamination.
Soil Consolidation:
This section is little changed from the Record of Decision. Essentially,
contaminated surface and subsurface soils from areas on site and some
sediment from Burnett Creek will be “consolidated” (dug up) and placed on
top of the slurry wall impoundments. There is no net decrease in total
toxicity at this site, and the waste is not “clean” in any way after
consolidation. However, areas that were highly contaminated at the surface
will be less so after consolidation. Consolidation could reduce toxic runoff
to Burnett Creek. Consolidation’s main benefit would be to wildlife and
trespassers that are exposed from activities on the site. Both groups are
poorly quantified by the EPA, so the degree of potential benefit is hard to
estimate.
Cap Construction:
After soil consolidation, the pond areas will have a grade that is higher
than the surrounding landscape—they will no longer be ponds they will be low
mounds. The purpose of the landfill caps is to reduce or prevent rainwater
from entering the impoundments through the consolidated soil mounds. If the
impoundments are successfully constructed, they will act like buckets buried
in the ground. If significant amounts of water were allowed to enter from
the top, the impoundment would fill and spill over, causing the consolidated
material on top to become unstable and possibly erode. Caps placed on top of
the impoundments should shed water—this would not decrease or increase plume
movements-- merely keep the impoundments from catching rainfall. Rainfall
would continue to be diverted to Burnett Creek through surface channels. The
scenarios provided all should work to shed water as described.
Monitoring/Institutional Controls:
Long-term monitoring is required to know if the remedy is working.
Monitoring includes checking chemical concentrations and movements within
the groundwater plumes outside of the impoundments. These chemical levels
would be expected to diminish over time. If the remedy is successful, the
plumes extending off site should withdraw from their present tracks, and
then dissipate. Deep monitoring outside of the impoundments is needed to
ensure the impoundments retain toxins and are not leaking to the deeper
aquifers.
Monitoring inside the impoundments would expect to show stability-- no
diminishment over time. It will be very important to monitor fluctuations in
water levels within the impoundments. Each impoundment will need multiple
measuring devices and the readings should be monthly for the first five
years.
The height and side slopes of the consolidation and cap should settle during
the first few years, and then become stable. Burnett Creek monitoring should
show no contamination after a few years. Institutional controls include
fences to restrict access, repair of the cap from weather erosion—and there
will be weather erosion-- and landscaping to control destructive vegetation
on the caps.
Conclusions
The
geological layers underneath the areas proposed for the slurry walled
impoundments do contain some regions that can block toxin movements;
however, it is also clear some areas will not provide a continuous “floor”
to the impoundments. Just as a bucket with a single hole will eventually
leak all of its contents these impoundments will not hold material
indefinitely. Using the bucket analogy, the rate at which the contents leak
depends on the number and size of the holes in the bottom. There is
insufficient information in the Design to predict with accuracy how well the
impoundments will work.
It can be agreed that the walls will be effective if properly constructed;
three-foot wide walls of high content clay material can be effective
chemical barriers. The notion that excluding rainwater will improve DNAL
migration is not supported by scientific studies at other sites. The DNAL
migration is driven by physical interactions and gravity, not by rainwater
intrusion or flow. If anything, the DNAL movement may speed up when the
slurry walls are in place since lateral groundwater movement would cease.
The Design lacks sufficient detail on the monitoring system envisioned for
the interior of the impoundments. Annual monitoring of a single well within
the impoundments would be grossly inadequate. Multiple monitoring wells at
several points within each impoundment are needed, and the wells should be
sampled monthly for at least the first few years to assure stability within
the impoundments.
Since it now seems likely technology will be needed to treat material within
the slurry walled impoundments there should be a thorough discussion of the
types of technologies that would be employed. A decision mechanism needs to
be added to the Design for implementation of treatment if water levels
within the impoundments are seen to be changing, of if monitoring within the
deep groundwater indicates migration of contaminants. Specific legal
language should be added to the Record of Decision and Scope of Work
documents regarding the limestone and deeper layers to be used as the floor
of the impoundments, since these are not as described to the public.
Cleanup of Burnett Creek appears limited in scope. However, removal of even
300 cubic yards is a big improvement over the current condition.
Consolidation of toxic surface materials on-site provides no advantages in
terms of reduction of waste, removal would be better. It will improve the
quality of life for wildlife that accesses the site. Trespassers continue to
enter the site. Site security and signage needs to be better defined than
currently described in the Design.
Studies on the geological layers indicate there really are no options if the
trenching fails to find a secure layer to act as a solid base. Those
sections of the wall inadequately keyed to limestone or clay will simply
leak waste at that spot.
This is a remedy with no margins for error on a geology marginally suited
for the technology.
In
the long-term, the safety of this Design is invested in the natural floor of
the impoundments and the security of the fences. Neither is adequately
defined in this Design.
Written by R. Kevin Pegg, Ph.D.; edited by Dr. Mary S. Saunders. Copies of
the newsletter are available from the GEC, at the Glynn County library, or
at www.enviro-issues.net on the Internet.
"This project has been funded wholly or partly by the U.S. Environmental
Protection Agency under Assistance Agreement Number
1984482-98-0
to The Glynn
Environmental Coalition, Inc. The contents of this document do not
necessarily reflect the views and policies of the U.S. Environmental
Protection agency, nor does mention of trade names or commercial products
constitute endorsement or recommendation for use." |
P. O. Box 2443