NCD980602163_19831020_Warren County PCB Landfill_SERB C_Background on Landfill Site-OCR

..
Ti; q n,7 l'I! lr'f) ·
·-" . j '
PCB LANDFILL SITE IN WARREN COUNTY ----OCT 201~8.'~
Background . ~~~k~~~wu
In 1978, approximately 210 miles of North Carolina roadsides were
contaminated with PCB' s through illegal disposal. PCB' s are most noted
for long persistence in the environment, and their acute human health
effects, which include a skin rash known as chloracne; and liver toxicity
resulting in alteration of liver enzymes. In addition, PCB's have been
shown to be carcinogenic in animal studies, and there is concern that they
may be carcinogenic in man. Added to the fact that they are already
ubiqui taus in the environment from indiscriminate use and disposal, and
that most people already-contain some body burden of PCB's, it is desirable
to prevent further bioaccumulation of these compounds in man. For this
reason, it was deemed necessary to isolate these compounds, first by
fixing them in place along the roadside and then by disposing of them.
EPA currently approves only two methods of disposal: destruction by
incineration and isolation by burial. Destruction by incineration was not
feasible due to the large volume of soil contaminated; therefore, the only
other al~ernative, burial, was selected. The landfill site in Warren County
was chosen, based on its overall suitability. (Factors involved in that.
decision included geologic and hydrologic characteristics, soil permeability,
isolation from population centers, availability of the land, etc.). The
landfill was designed and constructed to contain the PCB-laden soil indefinitely .
'!'he landfill area encompasses approximately three acres, and approximately
20 acres are contained in the perimeter around the landfill. Approximately
120 additional surrounding acr~s were deeded to Warren County as an additional
buffer zone . The landfill itself is elevated and constructed with the latest
technology to prevent migration of the PCB's. The contaminated soil is
•.
-2-
encapsulated in a 5-foot clay liner that exceeds the EPA permeability guidelines
of 1 x 10-7 by a factor of approximately 10. '!his means that soil permeability
studies indicate that water will migrate through the clay approximately .01
foot per year instead of the EPA guidelines of 0.1 foot per year. PCB's
should migrate at a much slower rate due to their affinity for clay soil
and their extremely low solubility in water of 25 ppb. In addition, two
30 mil plastic liners encapsulate the clay liner. '!he landfill contains a
leachate collection system that will be monitored every 30 days to detect
PCB's. Outside the landfill, the four monitoring groundwater wells in the
perimeter, as well as the surface water and sediment from the streams that
drain the topography, will be monitored every 6 months for PCB's. (Prior
-to landfill construction, background levels were taken from these wells,
streams and sediments and no PCB' s were detected.)
The majority of the PCB-laden soil has been delivered to the landfill.
Six separate composite soi~ borings, four .inches in diameter and five feet
deep, have revealed PCB concentrations ranging from a low of 46 ppm in one
boring to a high of 206 ppm in another .boring. '!he average concentration in
the six borings was 135 ppm. Two additional corrposite grab samples con-posed
of 10-12 random grabs of soil have shown 706 ppm and 4466 ppm. Replication
of the analysis showed 252 ppm and 117 ppm respectively, indicating "hot
spots" and non-uniform concentration of PCB's in the soil. If the previously
mentioned borings and the average concentration of 135 ppm obtained from them
are representative of the average concentration of PCB's in the landfill, one
should keep this in perspective with the fact that prior to 1978, PCB concen-
trations less than 500 ppm were not considered to be PCB-hazardous waste by
EPA and these concentrations were routinely disposed of into the environment
and into landfills. Since 1978, EPA has considered PCB concentrations of more
•.
-3-
than 50 ppm to be hazardous waste, necessitating disposal by one of the two
previously mentioned methods. In addition, these levels should be compared
to the extremely high levels of several hundred thousand ppm present in the
offshore sediments near New Bedford, Massachusetts, a major fishing area.
_ In light of this background information, it is reasonable to conclude
that the chance of population exposure to PCB's from the Warren County landfill
is negligible. This deduction is supported by the remote location of the land-
fill from population; ·. its design with the protective clay and vinyl liners
built into it, the nature of the surrounding topography, geology, and hy-
drology; the leachate collection system; and the groundwater and surface
water;sediment monitoring schedule, as well as the innate chemical charac-
teristics of PCB's that will tend to keep them in place--namely, low solubility
and affinity for soil particles.
Public concern over the landfill has raised the questions of whether or
not rasidentinl water well monitoring for PCB's is feasible and justified;
whether or not population rnoni toring for PCB' s via blood samples is feasible
and justified; and whether microbiological detoxification of PCB's is feasible
and justified. The relative merits of each of these proposals is discussed
below.
A. Residential Water Well Monitoring for PCB's. From a scientific stand-
point, justification for this is difficult, based on the above discussion
and the long distance from the landfill to the nearest home. 'lliere is one
unlikely scenario that might lead to contamination of residential well~ and that
would be a major earthquake, fracturing the landfill and communicating it with
the groundwater table. Such an event would be recognize4 and appropriate
sampling could be done at that time. One justification for sampling private
water wells sometime in the future (say, five years) would be to provide
•.
proof that PCB' s have not contaminated the surrounding groundwater supplies,
which should relieve the anxiety of Warren County residents. However, it is
reasonable to expect contamination of the monitoring wells (which will be
continuously sampled every six months) before residential wells, unless a
major fault line exists in which groundwater would bypass the monitoring
wells. The cost for sampling individual wells includes the cost for collection
and the cost for analysis. Present per sample cost incurred by the DHS lab
is approximately $60 for a qualitative (yes or no) analysis and $120 for a
quantitative analysis. If residential well sampling is proposed, it would
be necessary to test area wells now in order to get "before and after" com-_
parisons. The total cost could markedly escalate due to in-migration of
population around the landfill area.
B. Population Monitoring Via Blood Analysis for PCB's. Again, from
a scientific standpoint, little, if any, justification for this can be made,
based on the following reasor.s:
1. Population monitoring should be based on sound evidence
that a population has been exposed to an agent or is likely to be
exposed to an agent, unless the monitoring is being done to establish
baseline population data or the monitored population is serving as a
control population. As for the former, there is no evidence to suggest
people in Warren County have been exposed to PCB's originating from
the contaminated soil during its transport to the landfill. In
addition, potential routes of exposure to PCB's from the landfill
are not present. The possibility of dermal absorption from contact
with the PCB's, inhalation of airborne PCB's, and ingestion of food-
borne PCB' s originating from the landfill is nonexistent and the proba-
bility of ingestion of waterborne PCB's via surface or groundwater as
previously discussed_ is negligible. As for the latter, monitoring of
•,'
t ' -s-
the population to establish baseline data, etc., requires a sub-
stantial investment in time, manpower, and financial resources,
for this represents a prospective cohort epidemiological survey.
The capability for DHS to carry out such a study is not present;
·-neither is the scientific justification for such a study.
2. Population blood monitoring for PCB's requires a before
and after analysis. It is kna,m that virtually all people have
serum concentrations of the organohalides (includes PCB 's) in the
40 to 60 ppb range with a bell-shaped curve representing the popu-
lation. Breast milk PCB concentrations range in the area o_f 1 to 8
ppm fat basis of milk. There is fairly good correlation between
breast milk concentrations and serum concentrations. Estimated
average total body burden of PCB's (extrapolated from breast milk
sample concentrations) is in the range of 20 mg. The health signifi-
cance of these levels is not known at this time. It is generally
accepted that the only excretion pathway for PCB's is through breast
milk, hence a molecule of PCB absorbed is a molecule retained. '!here
currently is no good data on how serum PCB levels vary with age, race,
sex, stress,. hormone levels, etc. Some inferences may be able to be
drawn from data collected in Michigan on PBB's, in which knCMn
exposure took place through ingestion of food contaminated with PBB's .
3. In order to see changes in serum PCB concentrations of the
magnitude necessary to place them outside 1 or 2 standard deviations from
.\\v-the meanof baseline serum PCB bell-shaped curve, one Tt1ould have to I\
have a very large exposure from the PCB's in the landfill .
4. The Centers for Disease Control in Atlanta presently have 6
contract labs doing PCB analysis. A report scheduled to be released
-6-
in the December issue of ENVIRONMENTAL HEAL'ffi PERSPECTIVES will
apparently reveal significant variability in both interlab and
intralab quality control. ·· There are significant problems with
reliability (reproducibility) of results. In addition, the sensi-
tivity and specificity of results leave much to be desired, based
on the fact that DDE, a metabolite of DDT, ubiquitous in the
environment and in the human population, produces a spike in gas
chromatography . where spikes for PCB's occur, therefore increasing
the chance for false positive and false negative interpretations.
'!here are apparently one or two labs that have fairly good repro-
ducibility and sensitivity/specificity (e.g., Raltech, Inc., of Madison,
Wisconsin).
5. currently no lab is certified or licensed (?FDA) to do
clinical testing for PCB's .. Those currently doing PCB analysis
do so only for research purposes and agreements are signed with
the ·promise not to use the results of the tests for clinical
purposes.
.. 6. Issues raised in numbers 2 through 5 leave
substantial problems for those appointed to interpret lab results
to the public. In addition, they may signficantly increase the
anxiety and psychological stress of the residents in Warren County.
7. The cost for PCB serum analysis for research purposes is
approximately $100 per analysis. MJnitoring a population of 5,000
persons with before and after samples (e.g., now and 5 years from
now) would cost $1 million.
8. Interpretation of the results again would be difficult.
Should increases be seen in later samples (one would expect this
to occur), it must be determined if the increases are artifactual
(due to improved analytic methods, quality control, etc.) or real.
.. ,
'· .
If real, do they represent a shift of PCB's from the fat to
the serum due to weight loss, change in hormone level, or other
physiologic factors due to age, race, sex, stress, etc. Finally,
if serum increases are real and they represent increased body
burdens of PCB's, where did these increases come from--the environ-
ment, where PCB's are ubiquitous and we will all continue to absorb
small amounts; or from the landfill, from which the chance of
exposure is negligible. The burden of proof will probably rest
with the state, and the only way to obtain a defensible argument
that the increases did not come from the landfill is to design
a prospective, controlled clinical epidemiologic study involving
millions of doliars in resources and extensive scientific manpower.
C. Microbiological Detoxification of PCB's. There have been suggestions
that microbiological detoxification via biodegradation be employed at the
PCB landfill when the technology ·to do this becomes feasible. Presently,
information sources indicate biodegradation of PCB's is possible under
controlled laboratory conditions. However, the degree and extent of biode-
gradation is unclear. More work appears to be needed in the area of radio-
isotopic labelling and following the biodegradation process. Consideration
of the following points should take place before attempts to biodegrade the
PCB's in Warren County landfill.
1. What bacterial strains will be used? Are they non-pat~ogenic
to man? What is their capacity to mutate to pathogenic bacteria?
2. Are the bacteria substrate-specific for PCB's, or can they
attack other chemical compounds (e.g., plastics, rubber, etc.)?
can substrate specificity change by mutation?
3. What is their reproductive capacity, given unlimited substrate
and ideal environmental conditions?
4. Is the average concentration of PCB's in the landfill (e.g.,
-8-
135 ppm) sufficient for effective biodegradation?
5. What effect do temperature, humidity, oxygen availability,
etc. have on the rate of biodegradation?
6. Is biodegradation complete (e.g. , to simple molecules such
as co2 and H2o) or are intermediate metabolites formed? What is
the toxicity of these metabolites?
7. Will biodegradation necessitate removing the dirt from the
landfill layer by layer? '!his may be necessary because addition
of bacteria to the upper layers will certainly have no effect
on the more deeply buried PCB's.
8. What is the cost of biodegradation?
The answers to these questions are not available at this time. Until
they are, one should "let sleeping dogs lie. 11
Prepared for Dr. 1-::wine by Dr. Smi th
10/13/82