[From the U.S. Government Printing Office, www.gpo.gov]

COASTAL ZONE GRANT

RISK ASSESSCEN  AND EVAIUATION
OF SELECTED VIRGINIA SITES
WInITHN THE COASTAL ZONE              I4#~

FINAL-REPORT

Risk Assessment and Evaluation

of Selected Virginia Sites

Within the Coastal Zone

(SE-VDWM-4222-89)

U. S. DEPARTMENT OF COMMERCE
NOAA
COASTAL SERVICES CENTER
2234 SOUTH HOBSON AVENUE
CHARLESTON , SC 29405-2413

Submitted to

Dr. K. C. Das, Director
Administration and Special Programs
Commonwealth of Virginia
Department of Waste Management
Richmond, Virginia
Property of csc Library
by
1 I
.  Q





i        .   m>




4-

0 I1
The Center for Risk management of Engineering Systems
University of Virginia
Charlottesville, Virginia


January 15, 1990
This report was produced, in part, through
financial support from the Council on the
Environment pursuant to Coastal Resources
Program Grant No. NA88AA-D-CZO91 from the
National Oceanic and Atmospheric
Administration.

















PROJECT STAFF






Ralph Allen, Director, Environmental Health and Safety, Project Director

Yacov Y. Haimes, Lawrence R. Quarles Professor and Center Director

Duan Li, Research Assiotant Professor

W. S. Lung, Professor

W. Kelly, Research Associate

Mahesh Shah, Graduate Student

I	~Michael Lockhart, Research Associate

3	~Christopher Hamlett, Research Assistant

Risk Assessment and Evaluation
of Selected Virginia Sites Within a Coastal Zone


FINAL REPORT

I.	Summary of Proj ect GoalsI

IT.	Evaluation of Suffolk Chemical Site	2

A.	Summary of perceived problem	2
B.	Summary of field survey and sampling plan	2
C.	Analytical results	5
D.	Interpretation and modeling	6
E.	Conclusions	7
F.	Recommendations	3

III.    Evaluation of Alliance Fertilizer Site

A.	Summary of-perceived problem	18
B.	Summary of field survey and sampling plan	18
C.	Analytical results	20
D.	Interpretation and modeling	22
E.	Conclusions	2~4
F.	Recommendat iono	25

IV.     Evaluation of Republic Creosoting Site
(McLean Construction)

A.	Summary of perceived problem	37
B.	Summary of fi-eld survey and sampling plan	37
C.	Analytical results	39
D.	Interpretation and modeling	39
E.	Conclusions 4
F.	Recommendations                       ~41

V.	Framework for Decision Support System                              146

VI.	Computer Assisted Decision Support System (ERIES
Environmental Risk Information and Evaluation System)             50

References

Appendices:

1.	Actual analytical data for all three sites
2.	Field visit reports
3.	Monte Carlo estimation procedure
4.	Model used for Alliance and Suffolk sites
5.	Role of risk assessment in site evaluation

I.SUMMARY 0OF PROJECT' GOALS




.1            ~~~~From the inception of this program, the participants from
the Center for Risk Management of Engineering Systems have
-5,       ~~assumed two major goals for the project.

First, there were three specific sites that we were to
investigate and, based upon our risk assessmaent, we were to
give guidance to the Department of Waste Management on how to
proceed with these sites. Specifically, the sites were the
Suffolk Chemical Company near Suffolk, Alliance Chemical near
Hay-nesville, and the former Republic Creosoting Company (now
McLean Construction) located near Chesapeake, Virginia.

Second, we were to develop a systematic approach that
-         ~~co'Uld be used by the limited staf f of the Department of Waste
Management to organize information, make decisions, and plan
strategies in dealing, with the iaany other non-NPL sites in
Virginia.
I

I
I.EVALUATION OP SUFPOLX CHEMICAL SITE



A.   SUMMARY OF PERCEIVED PROBLEMS

The Department of Waste Management summarized the potential
problems at this site as follows:

The Suffolk Chemical Company has been operating an
industrial chemiical distribution center on this site in
Suffolk, Virginia, since 1970. Liquids known to be com-monly
handled include solvents such as methyl ethyl ketone, 1,1,1-
trichloroethane, and acetone, as well as nu-merous acids and
bases. The site contains a clay-lined lagoon which receives
rinse wastes and storm run-off and an area where solvents are
dispensed.    These  areas would be the main  location  of
contaminants. Sampling of on-site monitoring wells in 1986
detected significant levels of inorganics, such as lead,
arsenic, and cadmium, and organics, most of which are
sol-vents. No soil samples have been taken.

The site is located neaLr sensitive habitats.   It is
within one mile of the Great Dismal Swamp National Wildlife
Refuge and extensive wetlands associated with the Nansemond
River and Shingle Creek, both of which flow into Lake Meade,
which is used as a public water supply. Shingle creek. flows
within 200 yards of the facility and receives shallow
groundwater moving from the site.   There are some shallow
wells within a quarter mile of the site.

Monitoring wells an site which tap into the shallow
groundwater system were found to contain levels of lead and
cadmium that are above Maximnum Contaminant Levels established
under the Safe Drinking Water Act. Arsenic, a probable human
carcinogen,  was  also present  in the  samples.   Numerous
organics were detected in the monitoring well samples,
including various benzene derivatives and a numiber- of
carcinogens, including bis (2-chloroethyl) ether and n-
nitrosodimethylamine, compounds used as solvents in various
industrial processes.


B.   SUMMARY OF FIELD SURVEY AND SAMPLING DATA

As a result of the evaluation of previous data collected
and our site visit at Suffolk Chemicals (December, 1988), we
identified four major areas of concern and developed a
sampling plan to investigate each area.   This plan was
2

implemented on our second visit to this site (February, 1989).
The four areas are:

1.   The lagoon sludge:  No data established the status of the
lagoon as either haazardous waste or as a potential source of
groundwater contamination.  The sampling involved carefully
digging down to the lagoon liner and removing samples of the
liner sludge.  Analysis was for both EP TOX metals and for
total metals as well as several other inorganic species.

2.	The groundwater monitoring wells:	Previous analyses of
the	'wells   were   irregular,	with	several   analytical
discrepancies.    Re-examination	of  these  wells, was  also
necessary to provide stati sti cal support f or imodeling ef forts.

only three wells could be saimpled.  Well #3 (furthest
upstream) was below surface water level, making sampling
impossible. The other three wells, numbers 1,2, and 4, were
all analyzed for Primary Drinking Water metals, several
targeted  inorganics,  and  acid-base/neutral   extractable
organics (GC/IMS).

Preliminary assessment of the results shows well #2 to
hav-e the greatest inorganic contamination, low pH, high
chloride, nitrates, and sulfates, as well as trace amounts of
4-chloro-3-methylphenol, 12, 4-trichlorobenzene, and butyl
benzyl phthalate. (Phthalates are used as plasticizers and
under different solvent conditions, or acid environments,
etc., they could have leached from the PVC well casings.)
Well #2 is approximately 20 feet away from the sulfuric acid
tank (probably the source of the low pH and sulfates
observed).

Well #14 is unique, showing relatively high levels of
copper, zinc, and chromium. Well #4 also showed 20 to 60 ppb
levels of bis(2-ethylhexyl) phthalate and Poly Aromatic
Hydrocarbons. (Since these are commonly found in fuel oils,
they imay be a result of run-off from the nearby parking lot.)
(Note: Well #4 was partially filled in with sludge and bailing
the well out only made the sludge contamination worse.)_-

3.   The neighboring junkyard:   Previous data at well T#2
showed moderate levels of specific metals. Since this well
is directly downstream from what had been an automobile
junkyard, we wanted to test this as a source for metals.
There was much debris and metal an the ground, making it
necessary for us to insure that the junkyard was not
contributing to the site's groundwater contamination.	Two
sampies  were  taken  of the  soil..   EP TOX  analysis	was
performed.

4.   Surface water run-of f from beh ind the drum washing area:
3

I
I
I
I
 
I




I
I
I
 
1
I



.I
Ii
I
FIGURE # 3
SITE MAP - SUFFOL'K CHEMICALS








ko~
/   I VA 04

The drums are recycled, which requires washing them. An
evaluation of the run-off in this area of high contamination
potential was necessary to insure the washing station was not
causinq any contamination. There is a drainage ditch behind
the washing station. Sludge from this area was collected and
analyzed.



C.   ANALYTICAL RESULTS

The analytical data is summarized in Tables 1-12.
Locations of samples are shown in Fig. 3. New data are given
in Appendix 1. Each type of sample will be discussed below.
1. Soils
Two suspected -contaminated areas were the lagoon sludge and
the junkyard. The results were not remarkable~ for most
chemicals tested, except for slightly elevated levels of
copper and zinc. Note that the levels of total metals (Table
1) and sulfate are higher in the lagoon than in the soils near
the junkyard (Table 3). However, the EPTOX results more
realistically represent the leachate into the groundwater;
the6se results show a very low level of contamination (Table
2). Samples from the junkyard show no leakaqe: this result is
consistent with the fact that it was cleaned up.
2. Surface Water
The surface run-of f from the drum washing is likely ,to
contaminate the surrounding area. The drainage ditch behind
the drum washing area was analyzed; the results were not
remarkable (Tables 4-5).

3. Groundwater

Analysis of well samples has been performed twice: once in
1986 (LES) and again in 1989 (Havens Laboratory). These
results are displayed in Tables 6-12. High levels of chloride,
ammonia, sulfuric acid, and total orqanic carbon were detpLcted
in the 1986 study. An assessment of the 1989 study shows that
well #2 has the highest inorganic contamination, low pH, high
chloride, nitrate, and sulfate, as well as trace amounts of
4-chloro 3-methylphenol, 1,2,4 -trichlorobenzene and butyl
benzyl phthalate. Well #4 is unique, showinq relatively hiqh
levels of copper, zinc, and chromium, Poly Aromatic
Hydrocarbons and bis(2-ethylhexyl) phthalate. Well #1 was not
remarkable. Well #2 showed high levels of sulfates and low pH.
5

D). INTERPRETATION AND MODELING
Available data for the site was evaluated. Evaluation
of the soil and surface water data have yielded no conclusive
results. High levels of sulfate, chloride, nitrate, copper,
zinc, and some organics were found in the groundwater. These
contaminants are prevalent in the uppermost aquif er, which is
not used for drinking water purposes.
The contamination of groundwater at Suffolk is serious
but the contaminants themselves are not highly toxic. The top
aquifer is not utilized for drinking water purposes (LES,
1986). However, the middle aquifer supplies drinking water
(LES, 1986) and the chances of these chemicals reaching
aquifer from the top cannot be excluded. Data collected at
present is inadequate to predict the migration of contaminants
to the middle aquifer from the top acquifer. In order to
estimate migration to the middle aquifer and eventually to
drinking water wells, data will need to be collected
concerning the thickness af the confining layer between the
two aquifers, the hydraulic conductivities of the confining
layer and the middleF aquifer, and the potentiometric surface
of the middle aquifer.

The potential for on-site gjroundwater contaminants to
migrate to the nearby Shingle Creek was predicted utilizing two
computer ground water models. Several modeling systems were
investigated and these two were chosen based on their problem
solving ability, their in-group familiarity, and availability.

The first is a widely used but somewhat complex model, method
of characteristics (MOC) model developed at the U.S. Geological
Survey (Konikow and Bredehoeft, 1978). MOC can handle anisotropic
and heterogeneous media and non-point sources, and can account for
unsteady flow. This model was used (Scherer, 1989) for simulating
the migration off contaminants to Shingle Creek. Many simplifying
assumptions (e.g., steady-state flow, constant hydraulic gradient,
location, and leakage rate of sources assumed) had to be made to
run the program. Utilizing worst case defaults, it was found that
less than 2%-, of the contaminants reached Shingles Creek in 10
years.
The second model is an analytical solution for the
migration of contaminants from a point source (CONMIG; WaLlton,
1989). This model assumes steady-state flow and allows for
multiple point sources. This model, although simpler to use,
does not allow for probabilistic elements which are necessary
for risk analysis. Thus the basic equation used in this model
was utilized for deriving a distribution function for the
concentration at any point from the source, given the
distribution function of the source concentration. Appendix
6

4 describes these derivations and the application to the
Suffolk problem.

Parameters required for the model are given in Appendix 4
were estimated as shown in Table 13.  Assume that at well #2 a
sulfuric acid source is located which injects 100,000 gallons of
sulfuric acid at the present time. The concentration of the acid
has a distribution function which must be assessed. In the absence
of appropriate data, the distribution can be assumed to be of a
triangular form (Kelton and Law, 1982):
--------------------------------------------------------
= 0	when Co < a
Fco(Co)    = (Co-a)2/[(B-a)(r-a)]	when a < C0 < T
 1 - (B-C) /[(B-a)(B-T)]	when r < Co < B
- 1	when C0 > B


[,-B]  = interval in which c is believed to lie
r = mode; the most likely value
---------------------------------------------------------

From Equation (4) the distribution function for the
concentration at Shingle Creek, which is 600 ft away (distance
estimated from the U.S.G.S. 7.5 minute topographic map), is
 given by:
= 0	when C < af
F_(C)    = (C/f-a)2/[(B-a)(r-a)]	when af < C < rf
= 1 - (B-C/f) /[(B-a) (B-r)]	when Tf < C < Bf
~~~= 1	~when C > Bf

It is assumed that a=500 mg/L,	B=3000 mg/L and T=2000 mg/L.
Graphs of the input and output concentrations are shown in
Figures 1 and 2. From Figure 2 it becomes clear that the
chance of the concentration at the Creek exceeding 60 mg/L is
negligible.


E.   CONCLUSIONS
Based on the available data, the following statements about the
Suffolk site can be made:
(1)  Soil contamination of the junk yard and the drainage
ditch behind the tank washing building was not
significant, but because of the number of samples taken
and sampling conditions, the posibility of contamination
cannot be completely ruled out.
(2)  High levels of sulfate, chloride, nitrate, copper, and
zinc are present in some wells at groundwater level. In
particular, well #2 contains about 2000 mg/L of sulfate.
7

These chemicals could reach the drinking water system by
migrating to the Yorktown Aquifer.

(3) The models indicate that the possibility of groundwater
contaminants reaching Shingle Creek during the next 20
years i6 low, but cannot be ruled out. Specifically, if
a 100,000-gallon spill of sulfuric acid occurs near well
#2, whose concentration is described by a distribution
functions shown in Figure 1, the distribution of
concentration of the acid in Shingle Creek after 20 years
will  be  as  shown	in  Figure	2.	For	example,	if
concentration  near	well  #2	is	2000	mg/I,	the
concentration near Shingle Creek will be 35 mg/L after
20 years.

(4) High metal concentration found in well 12 is likely due
to an automobile salvage yard located adjacent to the
property. This area has been cleaned up and there was no
evidence of metals contamination of the surface soils
taken from the former junkyard off the Suffolk property.

(5) Metals concentrations in well #2 were much lower in the
samples analyzed. It therefore appears that the metals
contamination problem was suitably remedied by the clean
up of the junkyard.


F. RECOMMENDATIONS

1.   Although sulphate has been removed from the list of
hazardous  chemicals,     there  might  be  significant
quantities of ammnonium sulphate in the wells, since
ammonia and sulphate levels are high. Leakage from  the
filler hoses from the sulfuric acid tank (as well as
other   tanks) needs to be stopped by installing a drip
pad.

2.   Two of the four wells need repair. Well #3 needs to be
evaluated to remove the possibility of surface water
contamination. Well #4 has been partially filled in and
needs to be repaired.

3.   Regular monitoring of pH and several common inorganics
(sulphate, nitrate, ammonia, and chloride) needs to be
performed.

4.   Although the chemicals can imigrate to the iuiddle aquifer,
we do not expect this to happen because of the
intervening layer. However, investigation needs to be
done to confirm that the inter-vening layer is not
leaking. Sampling of wells withdrawing water from the
middle aquifer can also be used to check the quality.
8

Figure 1: CDF for Concentration of Sulfuric Acid at Well #2
1.2




.8

-    .6



.4

.2                                     /


0         500   I100            1500      2000        2500       3000       3500
Co -- Concentration of Sulfuric Acid
Figure 2:   CDF Function of Sulfuric Acid at Shingles Creek (after 20 years)
1.2 .




.8.




.4


.2


0    5    10   1       2    20   2  30   35   40   45   50   5'5   60   65
C (mg/L) -- Concentration of Sulfuric Acid

Table 1: Total Metals in the Soil (mg/L)
Chemical
Lagoon
Standard
Arsenic	<0.5	0.05
Barium	<10	1.0
Cadmium	1.4	0.010
Chromium	840	0.05
Copper	200	N/A
Lead	240	0.02
Mercury	<0.05	0.002
Selenium	<0.5	0.01
Silver	<1	0.05
Zinc	0.74	N/A
Table 2: EPTOX Metals in the Soil (mg/L)
Chemical
Lagoon
Junkyard
#1     #2
Standard
Arsenic
Barium
Cadmium
Chromium
Copper
Lead
Mercury
Selenium
Silver
Zinc
<0.002
<0.2
<0. 01
<0.1
<0.04
<0.1
<0. 0001
<0.002
<0.03
0.12
<0.002
<0.2
<0.01
<0.1
<0.04
<0.1
<0.0001
<0.002
<0.03
0.05
<0.002
<0.2
<0.01
<0.1
<0.09
<0.1
<0.0001
<0.002
<0.03
0.78
0.05
1.0
0.010
0.05
N/A
0.02
0.002
0.01
0.05
N/A



Table 3: Inorganios in the Soil (mg/kg)

Chemical          Lagoon         Junkyard
#1	#2

Ammonia (as N)	<0.8	<0.8	<0.8
Chloride	0.014	0.014	0.005
Nitrate (as N)	0.8	<0.1	<0.1
Sulfate	640.0	60.0	40.0
=========Q= =========== =========== =========== ==-=======
10

Table 4: Drainage Ditch EPTOX Metals Analysis (mg/L)
Chemical            Concentration                  Standard

Arsenic	<0.002	0.05
Barium	<0.2	1.0
Cadmium	<0.01	0.010
Chromium	<0.1	0.05
Copper	<0.04	N/A
Lead	<0.1	0.02
Mercury	<0.0001	0.002
Selenium	<0.002	0.01
Silver	<0.03	0.05
Zinc	0.13	N/A



Table 5: Drainage Ditch Inorganic Analysis (mg/Kg)

Chemical            Concentration

Ammonia (as N)	<0.8
Chloride	0.014
Nitrate (as N)	5.2
Sulfate	60.0
===================================-- ========ï¿½---
11

Table 6: Inorqanics (LES, Jan., 13, 1986)
(Unless otherwise stated, all parameters are in mg/L. NT
= not tested, N/A = not applicable.)
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Chemical                      Well#l                 Well#2              Well#3    Well#4
------------------------------------------------------------
Alkalinity         4
Ammonia
Chloride 2
Total Dissolved
Solids
Nitrate
pH                7.
Chromium,
dissolved
Mercury,
dissolved
Copper,
dissolved
Lead, dissolved
Zinc, dissolved
Arsenic,
dissolved
Barium, dissolved
Cadmium, dissolved
Volatile Organics
Total Organic
carbon
Cyanide
Sulfate
458.0
10.97
293.7

1103
<0.01
.1 @ 10 C

<0.04

<0.001

<0O.01
0.09
0.06

0. 015
<0.01
0.023
<0.010
832
8.04
1043

4072
0.40
6.3 @ 10

<0.04

<0.001

<0.01
0.21
0.11

0.039
0.15
0.023
<0.010
NT
NT
NT

NT
NT
C    NT

NT

NT

NT
NT
NT

NT
NT
NT
NT

NT
NT
NT
NT
NT
NT

NT
NT
NT

NT

NT

NT
NT
NT

NT
NT
NT
NT

NT
NT
NT
NT
NT
NT
NT
NT
NT
=== == == == == == ===-= == == == == === == == == == == ==
12

Table 7: Inorganics (LES, May 13, 1986)
(Unless otherwise stated, all parameters are in mg/L.
NT = not tested, N/A = not applicable.)
Chemical
Well#1
Well#2
Well#3
Well 4
Alkalinity        389.9
Ammonia	9.18
Chloride	372.0
Total Dissolved
Solids	737
Nitrate	<0.003
pH	6.4 @ 26
Chromium,
dissolved          <0.04
Mercury,
dissolved         <0.001
Copper,
dissolved	0.02
Lead, dissolved	0.09
Zinc, dissolved	0.08
Arsenic,
dissolved	0.007
Barium, dissolved	0.67
Cadmium, dissolved <0.007
Volatile Organics   NT
Total Organic
carbon	41.8
Cyanide	NT
Sulfate	89.96
894.6
11.5
1203
2310.5
2.52
11.3
502.2
3.56
398.3

1537
0.023
@ 26 C

<0.04

<0.001
2250
<0.003
11.0 @ 27  5.3
4750
0.022
C    4.7 @ 27
C
<0.04

<0.001

0.03
0.21
0.11

0.023
<0.01
0.029
NT

757.1
NT
987.74
<0.08

<0.001

0.13
0.37
0.09

0.006
<0.01
0.011
NT

54.5
NT
23.59
0.02
0.32
0.10
0.005
3.88
<0.007
NT

281.9
NT
13.28
===--===================================
13

Table 8: Organics (LES, Jan. 13, 1986)
(Unless otherwise noted, all the parameters are in mg/L.
ND = not detected, NT = not tested).

Chemical                          Well Number
1         2           3           4
1,2ï¿½Di----------                 ---------------
1,2 Dichlorobenzene	ND	ND	NT	NT
1,3 Dichlorobenzene	ND	ND	NT	NT
1,4 Dichlorobenzene	ND	ND	NT	NT
bis (2-chloroethyl)
ether	0.021	0.027	NT	NT
hexachlorethane	ND	0.123	NT	NT
hexachlorobenzene	ND        ND
n-nitrosodi-n-
propylamine              ND      0.041           NT          NT
bis (2-ethylhexyl)
phthalate	ND      0.027	NT	NT
n-nitroeodimethylamine	ND        ND	NT	NT
bis (2-chloroisopropyl)
ether	ND	ND	NT	NT
nitrobenzene	ND	ND	NT	NT
1,2,4-Trichlorobenzene	ND	ND	NT	NT
hexachlorocyclpentadiene ND	ND	NT	NT
2-chloronaphthalene	ND	ND	NT	NT
acenaphthene	ND	ND	NT	NT
dibutyl phthalate	ND	ND	NT	NT
14

Table 9: Organics (LES, May 6, 1986)
(Unless otherwise noted, all the parameters are in mg/L.
ND = not detected, NT = not tested).
Chemical
Well Number
2
1
3
4
0.026
ND
ND

ND
ND
0.011

ND

ND
0.012

0.015
ND
ND
ND
ND
ND
ND
0.018
0.023
ND

ND
ND
ND

0. 011

ND
ND

0.464
ND
0.011
ND
ND
ND
ND
1,2 Dichlrobenzene
1,3 Dichlorobenzene
1,4 Dichlrobenzene
bis (2-chloroethyl)
ether
hexachlroethane
hexachlorobenzene
n-nitrosodi-n-
propylamine
bis (2-ethylhexyl)
phthalate
n-nitroeodimethylamine
bis (2-chlroisopropyl)
ether
ND
ND
0.015

ND
ND
ND

ND

0.032
0.053
0. 044
ND
0. 014

0.126
0.978
ND

0.190

0.018
ND

0.109
0.199
0.015
0.018
0.016
0.035
0.011
ND
nitrobenzene	ND
1,2,4-Trichlrobenzene	ND
hexachlorocyclpentadiene ND
2-chloronaphtalene	ND
acenaphthene	ND
dibutyl phthalate	ND
-----------------------------------_--_=
===============r_=_--,-,--,-------------
Table 10: Total Metals (Havens Laboratory, 2/22/89) (mg/L)

Chemical       Well#1   Well#2a   Well#2b    Well#4b    Standard


------------------------------------------------------------____
Arsenic
Barium
Cadmium
Chromium
Copper
Lead
Mercury
Selenium
Silver
Zink
<0.002
<0.2
<0.01
0.1
<0.04
<0.1
<0.0001
<0.002
<0.03
<0.08
<0.002
<0.2
<0. 01
0.5
0.14
0.1
<0.0001
<0.002
<0.03
0.75
0.05
1.0
0.010
_0.05
N/A
0.02
0.002
0.01
0.05
N/A
<0.002
<0.2
<0.01
<0.1
<0.04
<0.1
<0.0001
<0.002
<0.03
0.13
<0.002
<0.2
<0.01
0.2
<0.04
<0.1
<0.0001
<0.002
<0.03
0.06
---------------------------_==_----_--__
=a=========_L_----_---------------------
15

Table 11: Inorganic Analysis (Havens Laboratory, 2/22/89) (mg/L)
Chemical       Well#l    Well#2a    Well#2b   Well#4b

Alkalinity
(as CaCo3)	350	370	350	330
Ammonia (as N)	9.9	83	96	4.2
Chloride	210.0	585.0	615.0	448.0
Solids, total
dissolved	1120	4670	5130	1640
Nitrate (as N)	0.525	4.550	3.300	0.700
pH	6.4	5.0	5.0	5.9
Sulfate	120.0	1900.0	2100.0	105.0
Cyanide	<0.01	<0.01	<0.01	<0.01
Conductivity
(umho/cm)        1400       3300        3300       1700



Table 12: Organic Analysis (Havens Laboratory, 2/22/89) (mg/L)
(samples were extracted by EPA method 3510, analyzed
by EPA method 8270, GC/MS using a DB-1 Col. ND = none
detected).

Chemical                     Well#l	Well#2a    Well44b

4-chloro 3-methylphenol	ND	0.023	ND
1,2,4-trichlorobenzene	ND	0.009	0.034
acenaphthelene	ND	ND	0.027
fluorene	ND	ND	0.008
fluoranthene	ND	ND	0.060
pyrene	ND	ND	0.039
butyl benzyl phthalate	ND	0.044	ND
bis(2-ethylhexyl)phthalate	ND	ND	0.116
16

I
I


Table 13: Risk Analysis Parameter Values

Parameter                           Value              Remarks
--1
Hydraulic conductivity, K        15 ft/day	conservative

Effective porosity, n              0.3	conservative

Hydraulic gradient , dh/dx       0.002 ft/ft	conservative
(
L
E
S
,

1
9
8
6
)
Aquifer thickness, m	30 ft	average

Long. dispersivity, A	10 ft	subjective
e
s
t
i
m
a
t
e

Trans. dispersivity, A              2 ft              subjective
e
s
t
i
m
a
t
e

seepage velocity, v                0.1 ft/day	computed

volume injected, v               100,000 gal.	conservative

Retardation factor, R                1	conservative
(no
ret
ard
ati
on)

Half life, h                     infinity             conservative
(
n
o

d
e
c
a
y
)






I

III.- EVM1UATIOAN OF ALZIA-2IC F21tTIL1."Z2R COMPANY SIT2
A.   SUMMARY OF PERCEIVED PROBLEM

The Department of Waste Management summarized the
potential problems at this site:
The Alliance Fertilizer site, locatad in Richmond County
near Haynesville, has been in operation since 1977.   The
parent company is Alliance Agronomics Inc. of Mechanicsville,
Virginia.  Fertilizer is blended on-site and herbicides and
pesticides are added during this process.   The site is the
source of nutrients and several herbicides that are leaching
into the shallow groundwater and moving into nearby surface
waters.  Pesticides, including toxaphene and dieldrin, were
detected in soil samples.

The shallow aquifer in the nearby area is utilized as a
drinking water source for private residences. A nearby town
obtains its water from a deeper aquifer in the area.   The
herbicides atrazine, alachlor, metolachlor, and dicamba have
been detected in significant concentrations in the shallow
aquifer.   Several of them are possible human carcinogens.
Nitrate is also present at levels that could present a human
health hazard.	One well has already been abandoned due to
conta-mination.	A nearby spring is also contaminated wiith
nutrients and herbicides at concentrations that are toxic to
some freshwater organisms. This spring eventually drains into
the Rappahannock River, approximately three miles downstream.


B.   SUMM4ARY OF FIELD SURVEY AND SAM~PLING PLAN

Our sampling and analysis had several basic goals:

1.  To test the possibility of a "hot spot" at the left front
of the Alliance property as a potential source of the
contamination of the Davis well on the neighboring
property.

2.   To remeasure all monitoring wells and the Davis well.
To check concentration of previously identified problems
and look for other possible problems.

3.   To examine the sp-ring to determine current level of
contamination and potential sources of loading.
is

4.   To examine drainage ditches for organic pollutants in an
attempt to evaluate source of contamination in front of
property (possibly leadingj to Davis well contamination) .



Initial sampling at the Haynes-ville site occurred on May 24,
1989, and samples were taken in three categories:

1.   Surface water at the head of Purcell Springs, behind the
Alliance site.

2.   Well water from all four on-site monitoring wells and
the Davis property well.

3.   Three soil samples, two from drainage ditches at the
front and the right side and one from the driveway at a
"hot spot" (approx. 2' deep) where an earlier spill was
suspected.

4.   To locate the "hot spot", we used a H-Ntl PI 101 gas
analyzer hooked up to a KV soil gas probe. The soil gas
was drawn by a 0.5 I/rn pump from approximately 18-2011 of
depth. The area of the left front part of the property
was laid out in a 4-part grid and the analysis conducted
at each corner, with two samples taken from the middle
of the grids. No hot spots were identified. one sample
had a reading between 1. 5 and 2 ppm, so a core sample
was taken and analyzed.


Water samples were analyzed for IpH, conductivity, TDS,
P04,J  NO3,   and  priority   pollutants   (acid/base-neutral
extractables only). Soil samples were analyzed for only
acid/base-neutral extractable priority pollutants.

Preliminary stream modeling of Purcell Spring indicated
that the results of the previous sampling showed a great drop-
off in concentration from sampling location #1l to #2. This,
along with our earlier inspection of the spring, suggested
that the stream was not running during sampling; therefore,
our efforts to identify the loading location were invalid.

Therefore, a second sampling of Purcell Spring was deemed
necessary and accomplished on a rainy day when the stream was
running.  We were also able to get several loading samples
from behind Alliance and the Davis property.
19

C.  ANALYTICAL RESULTS
The locations of samples are identified in Figures 4-6.
Figure 7 is a site map showing the monitoring well locations,
relative grade of the property, locations of other wells to the
site, and the area of the suspected "Hot Spot". Analytical data are
summarized in Tables 14-36 (the data collected for this study are
included in Appendix 1). Each type of sample is reviewed below:

1.   soils

Soil samples were tested twice: once in February, 1988, by the
Virginia State Water Control Board and then by Havens
Laboratory in 1989. Both times, the samples were tested for
pesticides and herbicides (atrazine, metolachlor, alachlor and
dicamba in 1988; alachlor and -metolachlor in 1989) but
generally these compounds were not detected (see Table 14).
The three samples taken during this study (form drainage
ditches and the front of the site) did not contain detectable
levels. The owner indicated the possibility of a hot spot in
the left front of the property. This was analyzed using soil
gas analysis and no containination was found.

2.   Surface water

The following surface water bodies have be-en tested for
contamination: the holding pond, Purcell Spring, and the
stream which runs from the site into Totuskey Creek.

Holdincr Pond: The 1986 samples (Tables 15-16) showed high
levels of chloride, TKN', ammonia, nitrate, sulphate, atrazine,
lasso and dual.  No further analysis was conducted because
Alliance does not have a discharge permit nor do they
discharge  from this pond to  the stream.   This was not
perceived as a problem.

Purcell SiDrina~: High le-vels off ammonia, nitrate, and sulphate
were detected in '77 and -'81 samples (Table 17). The 1986
samples showed a drop in the nitrate and ammonia level. The
1981 samples showed high levels of alachlor, metolachlor, and
atrazine (Table 18) and metals (Table 19). The high levels of
these chemicals are possibly due to the run-off from Alliance
and nearby fields.   Purcell Spring continues as a stream
behind Alliance.
Stream: The stream near Alliance was sampled thrice: in 1981,
1988, and 1989 (see Tables 20-24). There is a discrepancy in
the location of stations in 1981 and 1988 (as noted in the
correspondence with Keith Fowler of the. State Water Control
Board in Appendix 2 with the Field Visit Reports) and it is
difficult to decipher the relative locations of stations in
20

1981 . High levels of nitrate, ammonia, alachlor, metolachlor,
and atrazine were recorded at Station no. 3 (which probably
is the same as station no. I of 1988) in the 1981 studY
(Tables 21-22). Station no. 4 also showed high levels of those
chemicals. In the 1988 study, there was a high level ot
Nitrate (29.9 mg/L) at station no.l. Levels of alachlor,
metolachlor, and atrazine were signif icantly less compared to
those of 1981. It was not clear from the 1981 and 1988 data
where these contaminants were coming from, hence additional
nitrate samples were taken by Havens Laboratory on June 2 1,
1989, on a rainy day (Table 19).

Figure 9 shows a plot of nitrate concentration along the
stream. From the 1988 data it becomes clear that there is a
sudden drop-off after the initial sample location and then
the concentration remains constant. However, the 1989 data
shows an initial drop-of f and then a sudden rise.  This is
probably due to a washload which comes from the drainage ditch
near the hog lot (see Figures 4 and 5 for the locations of the
stations for '88 and '89 respectively) ; a concentration level
of 14.0 mg/L measured in that ditch confirms this hypothesis.
Another source is at the imouth of Purcell Spring (Station no.
1) . From the run-of f samples at the back of Alliance property
(maximum concentration of 2.26 mg/L), it appears that the high
concentration of nitrate at station number I in 1988 is not
due to surface run-off from Alliance. It may be because of
groundwater seepage or because of evaporation, which had
concentrated the nitrate in the standing water where station
no. I was located.

3.   Groundwater

Of the four residential 'wells (Haynie, King, Lawrence Davis,
and Davis), only the Davis well showed significant levels of
contamination (Tables 25-28). High levels of nitrate,
chloride, sulphate, metolachlor, alachlor, dicamba, atrazine
and some heavy metals have been detected since 1977. Figure
10 shows a plot of nitrate and chloride concentration in the
well as a function of time. As can be seen, there is a rise
in the contaminants from 1977 to 1986 from 20 mg/L to 60 mg/L
and then a drop in 1989 to 12 mg/L.

Figure 11 shows a plot of nitrate concentration in the four
monitoring wells. As can be seen, the nitrate concentration
is quite high in the wells from August 1986 to September 1987,
with well no. 4 showing the highest concentration. Samples
taken by Havens Laboratory in M~ay 1989 showed very low levels
of nitrate in the four wells.   The results for inorganic
contaminants in the four monitoring wells are summarized in
Tables 29-32.

The four monitoring wells also show high levels of
21

mnetolachlor, alchlor, dicamnba, and atrazine (summarized in
Tables 33-36).
D.   INTERPRETATION AND MODELTNG

Examination of our second sampling results of the Purcell
Spring indicates that much of the nitrate contamination is
from the hog lot at the back of the Davis property.   High
levels of nitrate at the head of the spring do not infulence
other results of the spring, because the spring only runs when
it rains. There is a pond at the head location (station no.
1) that evaporate-s and concentrates what probably were low
levels into many times higher levels.

surface examination, which included soil gas analysis in
search of the hypothesized "hot spot," did not indicate any
signif icant contamination.   Interpretati on of the plotted
nitrate concentrations in the monitoring wells may indicate
a peak in the contamiination in midyear 1987 and a drop-off
since. For this to be a viable hypothesis, further sampling
of the monitoring wells will be needed to establish the trend.
There is considerable uncertainty concerning ~the source
of the nitrate (N03-) and other contamination in the Davis
well. The "hot spot" near the buildings (in the front of the
Alliance property) was ruled out because no contamination was
detected. However, in case the source of contamination was
on the Alliance property, we have modeled the possible
migration of contaminants to this well from any source an the
Alliance property.  Two computer models for solving solute
transport poroblems were used to assess the potential for
contamination of the Davis well due to activities at Alliance
Fertilizer: CONMIG and MOC. CONMIG (Walton, 1989) is a simple
analytical solution for the advection-dispersion equation for
point-source pollution. The Method of Characteristics (MOC)
code (Konikow and Bredehoeft, 1978) is a numerical model which
couples the groundwater flow equation with the solute-
transport equation. The models are independent of the actual
chemical  contaminant.   We have used nitrate as a model
compound.

Data requirements for the models include estimates of
aquifer characteristics such as porosity, thickness, flow
velocity, transmissivity, head values, longitudinal and
transverse dispersivities, location of the contamination
source, and injection rate and concentration of the
contaminant of interest.   Both models are also capable of
modeling simplified chemical reactions by incorporating
sorption and/or decay constants.
The Groundwater Management Plan prepared for Alliance
Fertilizer provides estimates for aquifer thickness (100
22

feet), flow velocity (3.5 - 10ft/yr)1, and transmissivities
(0. 000521 - 0. 0490 fta/s) . Therem is some question as to the
validity of the flow velocities quoted in the report.   A
conservatively large value (40 ft/yr) was used in the
simulations. Transmissivities were determined by slug tests
and  showed   considerable  variability.        A  reasonably
conservative value of 0.02 ft2-/s was used. A typical porosity
f or unconsol idated sand deposits was employed (0. 3) . In order
to be conservative, large values for dispersivities,
contaminant   injection  rate,   and  an  initial   nitrate
concentration of 1000 mg/L were used at the source.  Using
these values, a steady-state nitrate concentration of
approximately 11 mg/L i s attained in the Davis well after
about 14 years (Figure 12) .  Large nitrtate concentrations
can be attained if smaller velocities are used. This allows
greater spreading of the plume in a lateral direction because
the containment is not allowed to migrate away from the source
as rapidly.  On the other hand, the amount of time for the
plume to reach the Davis well greatly increases. For example,
if a nitrate concentration of approximately 20 mg/L reached
in the Davis well using a velocity of Ift/yr is considered,
a steady-state nitrate concentration in the Davis well greater
than 100 mg/L is attained after 100 years. (At 50 years, the
concentration is less than 10mg/L.)

The MOC code was used for simulations in which the Davis
well was pumped at a rate of 1000 gallons/day with a constant
concentration of nitrate at the source of 500 mg/L. Steady
state was reached in less than 15 years with a nitrate
concentration of approximately 46 mg/L in the Davis well
(Figure 13).

Assumptions inherent in these simulations include: (1)
the aquifer is homogeneous and isotropic (i.e., constant
hydraulic  conductivity);  (2)  the  aquifer  thickness  is
constant; (3) the contaminant is well-mixed throughout the
aquifer; (4) the velocity is constant; and (5) contamination
loading and concentration is constant.

With respect to nitrate concentrations in the Davis well,
the most important variables in the models are transverse
dispersivity, injection rate, and concentration of nitrate at
the source, so extremely conservative values were selected.
Dispersivities are poorly understood in a physical sense and
are typically used as fitting parameters.   A longitudinal
dispersivity of 100 feet is near the uppermost limit typically
used by hydrogeologists and the ratio of longitudinal to
transverse dispersivity of five used in the simulations is
smaller than values suggested by experts (Anderson, 1979).
The result is that the dispersivity values used in these
simulations represent a worst-case scenario with respect to
spreading  of  the plume.    For  example,  if a transverse
23

dispersivity of 10 feet is used in the CONM~IG simulation, the
steady-state nitrate concentration in the Davis well is
approximately halved. Other parameters may also be important.
The role of pare velocity was discussed above.   If aquifer
thickness and/or porosity are significantly less than the
values   used,   then   nitrate   concentrations   could   be
significantly higher because the contaminant is diluted in a
smaller volume of water.

It is apparent from the use of very conservative data and
assumptions in the models that Alliance Fertilizer is probably
not the source of nitrate contamination to the Davis well (and
other contaminants as well) . Additional evidence support this
conclusion.   The considerable variability in the measured
nitrate values suggests that contamination is not primarily
due to migration of groundwater from a source more than 400
feet away. This behavior is more likely due to surface water
entering the well.   If this hypothesis  is correct,  the
contamination may be from the Davis form itself (note that a
ditch separates the Alliance site from the Davis well) and
would limit the possibility of surface water contamination
from Alliance. In addition, nitrate is typically a reactive
species in shallow aquifers, and a considerable amount would
be expected to decay during groundwater migration.



E.   CONCLUSIONS

1. Levels of ammonia, nitrate, and sulphates were high in the
surface water samples taken at Purcell Spring, probably due
to the run-of f from nearby fields. Samples examined by Havens
Laboratory for nitrate concentrations, do not indicate that
run-off from this site is contributing to the spring
contamination problem.

2. The soil sample analysis (both soil gas and solid) do not
indicate any contamination.

3. A massive spill with concentrations of 1000 parts per
million in the soil would be unlikely to he able to
contaminate the Davis Well at the identified levels (It may
be a contributing factor). M~odels of these types do not take
into consideration specific molecular characteristics, like
hydrophobicity,   ionic  strength,   etc.   They  treat  all
contaminants as similar agents. There are differences between
organics and inorganics migration/transportation mechanisms
(the oil industry is spending fortunes evaluating these
conditions) in terms of adsorption/asbsorption partition
coefficients and many other liquid mediu-m interactions.

4. it is unlikely that Davis well is contaminated from the
24

Figure 4: Location of Stream Sampling Stations
(1988; Keith Fowler, SWCB)
(see Table 18 for concentrations)









a .,..  *.*.** ....           .   .
24a
p                                                 * /~~~~~~~~~~~~~~~~~~~~y






Location off Stream Sampling Stations
(1989, Havens Laboratory)
(see Table 19 fcr concentrations)




214b









.. ..... ...








. .. .......J...

I








































Fi~~~~~~~~~~gur 6LoatPison ofSra amplin Sat                 ions

*~~~~~(91 S W      B       ) (see  Tal     6




* A                            .ï¿½e                               b
*  ~ ~ ~ ~ SS              0      .~~~~24

I II'"~i"i IiIIII/I.A

+ Mathie shallow
supply well
(abandoned)
Agricultural Field






Plant	-]Pool
Office       Well  I Plant	d         67         /66    P     Apprt v im ,at
L_J~~~~~~~~~~~~~~~~~~u~i  67am'
o
0o



N	O
QP	0
P-
/--
rupuu ty  LIIiu
1                                                  ..                       65


A                   S
A -_ MW #_2 _-/                                      -            mw,!-

Legend
66                                                     o	monitoring well location
/	supply well location
,ultural Field	contour line
5    Davis shallow
6~~~~pI wir%N1tuial
 MW #




"Hot Spc
Agric
supply weii
(abandoned)
/
Figure 7:   Alliance Fertilizer - Haynesville, Virginia





4
I.
a


a






*
0














*                                                                    -,







I
-7

2


.// //jiIj& s_ -
_
I                                            ï¿½'
yKZ




(  Figure 8 j_ _
-.N      ..   _

I
I:.
30




25-	0  1988
:.       ...	ï¿½ ,*1989

I.l           E
.:  20-

co
0

|    15-
-I

Z    10 -

I



5-    \ _




0
0          0.2         0.4          0.6          0.8
Distance (miles)



Figure 9:  Nitrate in the stream near Alliance
24r

I









I l

ii l
II


"! 100-


s                                               ~~~~~~~~~~~~--- Chloride
-- l  80-                                !Nitrate

; |  ~~80                              //                            OhIr\d
C,                        /

*z~ï¿½   40 -                     f//\

0                        /
I~~~~~~~~
I ~~c    60 /
O                    //'A~~

/







I                   ' "


,1.,	,    ~	~     ~	~     ~	~~~~~~~~~~ I
	I	&
>1                 77	80	83	86	89
0                                    ~~~~~~~~~~Year
I
Figure 10: Concentration of chemicals in the Davis Well
I

24g
a

160




140




120




80




60




40




20
I
I

.I
, .







71

!
I


i
0I
E
.I







'is
a





1
I


I
I
i

I
SWCB
* 1
10 2
 3
o 4
A&L
\

I
I
t
%

\%N%    1


NI-


I
I
I
II
4
8/86    12/86   3/87  6/87  9/87
12/87   3/88
Ti me
6/88  9/88  12/88
5/89
Figure 11:  Concentration of nitrate in monotoring
wells.
24h


Figure 12:  CONMIG results for NO3 - migration from the Alliance
site. Contours are in mg/L NO3.





24i


Figure 13:  MOC results for NO3 - migration from the Alliance
site. Contours are in mg/L NO3.


~~~~~~~~~24j~~~O

24 j

front part  of the  site.   it  is possible  that  it was
contaminated in the past because of contamination in the Davis
property.

F.   RECOMMENDATIONS

1. The holding pond should not be discharged into the stream.
2. An ongoing monitoring program of the wells should be
implemented.
25

Table 14: Soil Samples at Alliance (1-7 VSWCB, 2/2/88),
(8-10 HLI 5/89)
(ND** = not detected)
Station Number*
Chemical(ppm) 1   2   3   4    5    6   7    8   9   10
-------------------------------------------------------------
Atrazine	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
Metolachlor	ND	ND	0.59	ND	ND	0.04	ND	ND	ND	ND
Alachlor	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
Dicamba	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
Pesticides	ND	ND	ND	ND	ND	ND	ND	ND	ND	ND
* STATIONS:

1: Agricultural field east of Alliance
2: Agricultural field west of Alliance
3: East of drainage ditch
4: Front lot of route 360
5: Near pond (?)
6: Dirt lot in back of Alliance
7: Former discharge location in back of Alliance
8: Drainage ditch in east of property
9: Drainage ditch in front of property
10: Hot spot source, left front of property
.1




*X
ij

,:i

I
I
:

; I
** Detection limits:
Atrazine, Metolachlor, Alachlor,
Pesticides - 0.02 ppm
Dicamba - 0.01 ppm
Table 15: Holding Pond (4/7/86, VSWCB)
(Inorganics, mg/L)
Chemical
Concentration
pH
Conductance (umho/cm)
Diss. Solids
Chloride
TKN
Ammonia (as N)
Nitrate
Nitrite
Sulphate
8.8
255
15,828
4,000
4,000
2,900
786.5
1.0
144
===========================--===-======
26

Table 16: Holding Pond (4/7/86, VSWCB)
(Pesticide, ug/L)

Chemical         Surface          Mid
---------------------------------------
Atrazine	14,600	9,708
Lasso	3,900	2,934
Dual	14,700	11,400





Table 17: Purcell Spring (12/6/77 - 4/7/86, VSWCB)
(Inorganics, mg/L, NT = not tested)

Chemical    12/6/77    4/14/81       7/22/81    7/30/81    4/7/86
--------------------------------------------------------------__-
pH	5.7	4.9	4.9	5.6	4.8
Conductance	NT	NT	NT	NT	504
Total Solids	119	NT	NT	NT	NT
Volatile	62	NT	NT	NT	NT
Fixed	57	NT	NT	NT	NT
Susp. sol.	9	NT	NT	NT	NT
Diss. Sol.	NT	NT	NT	NT	219
Chloride	15	NT	NT	NT	4.0
TKN	<0.1	175.0       96-102.5	160.0	21.0
T. Phos.	<0.1	0.2	NT	0.1	NT
0. Phos.	<0.01	0.1	NT	0.04	NT
Ammonia (as N) <0.1	160.0	96-102.5	137.5	19.0
Nitrate	2.3	200.0	105.0	175.0	27.45
Nitrite	0.01	0.37	0.33	0.44	0.05
Sulphate	28	NT           NT	NT	32.8






Table 18: Purcell Spring (4/14/81 - 7/30/81, VSWCB)
(Pesticides, ug/L)

Chemical            4/14/81            7/22/81           7/30/81
-------------------------------------------------------------__
Alachlor	16	44	23
Metolachlor	49	166	87
Atrazine	54	121	15
=========  _ ===========----------------========_=
27

Table 19: Purcell Spring (4/14/81 - 7/22/81, VSWCB)
(Metals, ug/L)

Metal             4/14/81             7/22/81

Arsenic	4	3
Cadmium	<10	<10
Chromium	<10	<10
Copper	<10	<10
Iron	NT	210
Lead	<2	2
Magnesium	NT	15,700
Manganese	NT	2,900
Mercury	<0.3	<0.3
Nickel	<100	10
Potassium	NT	14,600
Zinc	130	40




Table 20: Totuskey Creek Tributary Sampling (Inorganics,
VSWCB, 7/30/81)
(Concentrations are in mg/L)

Station Number*
Chemical	1	2	3	4	5	6

pH	6.4	5.8	5.6	6.6	6.2	6.3
Alkalinity	7	4	19	19	7	10
TKN	0.9	0.5	160.0	31.0	1.0	0.9
T. Phos	0.1	0.1	0.1	0.3	0.1	0.2
O. Phos	0.08	0.05	0.04	0.29	0.04	0.18
Ammonia	0.6	<0.1	137.5	30.0	0.9	0.6
Nitrate	3.9	0.7	175.0	5.1	4.4	2.5
Nitrite	0.05	<0.1	0.44	0.39	0.05	0.02

* See Figure   for location of the stations.




Table 21: Totuskey Creek Tributary Sampling (Organics,
VSWCB, 7/30/81)
(Concentrations are in ug/L. ND = not detected,
detection level is 0.1 ug/L. NT = not tested.)

Station Number*
Chemical	1        2	3	4	5	6

Alachlor	0.14	0.19	23.0	4.0	0.2	0.04
Metolachlor	1.0	0.08	87.0	40.0	1.6	ND
28

Trifluralin	NT	0.26	NT	NT	NT	NT
Atrazine	ND	ND	15.0	24.0	1.3	ND

* See Figure   for location of the stations.



Table 22: Totuskey Creek Sampling (Inorganics, 2/2/88, VSWCB)
(Concentrations are in mg/L. NT = not tested, ND = not
detected.)

Station Number*
Chemical	1	2	3	4       5	6

pH	4.4	5.9	5.8	6.0	6.2	6.1
Alkalinity	NT	NT	NT	3.2	5.0	3.2
Acidity	46	5	7	5	3	2
Dissolved solids,	217	67	70	73	50	40
total
TKN	14.0	0.4	0.4	0.3	0.5	4.5
Total Phosphorous	0.2	0.1-	0.1-	0.1-	0.1-	0.1-
Ortho Phosphorous	0.01-	0.01-	0.01-	0.01	0.03	0.01
Ammonia	4.0	0.04-	0.07	0.05	0.04	0.05
Nitrate	29.9	4.5	5.5	5.5	4.0	3.5
Nitrite	0.08	0.01-	0.01-	0.01-	0.01-	0.01-
Conductivity	469	154	116	145	88.6	107
(umho/cm)
=================== --==================
* See Figure    for the location of stations.



Table 23: Totuskey Creek Sampling (Organics, 2/2/88, VSWCB)
(Concentrations are in ug/L. NT = not tested, ND = not
detected.)

Station Numbers*
Chemical               1          2       3       4      5        6

2,4-D	<0.3	<0.3	<0.3	<0.3	<0.3	<0.3
Linucon	<1.0	<1.0	<1.0	<1.0	<0.1	<1.0
Alachlor	<0.1	<0. I	<0.1	<0.1	<0.1	<0.1
Metolachlor	5.2	<0.1	<0.1	<0.1	<0.1	<0.1
Atrazine	1.0	<0.2	<0.2	<0.2	<0.2	<0.2
Picamba            <6.0(?)	<0.1	<0.1	<0.1	<0.1	<0.1
===================================* See Figure for location of stations.
* See Figure    for location of stations.
29

Table 24: Totuskey Creek Tributary Sampling for Nitrate (Havens
Lab., 6/21/89)

Station*                      Concentraion (mg/L)

1. Top of Purcell Spring	9.20
2. Branch # 3	3.30
3. Branch # 2	2.28
4. Foamy spot 100 ft past 5	11.2
5. Connect for 6	10.8
6. Run-off ditch behind hog lot	14.0
7. Old Stream	9.92
8. run-off ditch into Purcell spr.	0.51
9. Back Center, Alliance Property	0.51
10.Behind dike, Alliance Property	2.26

* See Figure   for the location of stations




Table 25: Davis Well Inorganic Sampling (12/6/77 - 3/13/86)
(Units are mg/L unless otherwise stated. NT=not tested.)

Chemical          12/6/77	4/14/81           3/13/86
VSWCB	VSWCB	RCHD

Nitrate	20.0	20.0	51.0
Chloride	14	36	75.1
Sulphate	15	18	33.2
Total Solids	231	272	550
Volatile	147	165	288
Fixed	84	107	262
TKN	<0.1	0.2	0.9
T.Phos	<0.1	<0.1	<0.01
O.Phos	<0.01	0.01	<0.05
Ammonia (as N)	<0.1	<0.1	0.7
Nitrite	<0.01	<0.01	0.05
BOD	1	NT	NT
TOC	2	4	NT
Fluoride	NT	<0.1	<0.1
Halo. Hydrocarbon	NT	<0.1	<1.0
(ug/L)
Arom. Hydrocarbon  NT             NT                <1.0
(ug/L)
PH	5.8	5.7	6.7
Alkalinity	NT	NT	NT
Conductance	NT	NT	728.2
(umho/cm)
=================a==============_~=============
30

Table 26: Davis Well Inorganic Sampling (3/31/86 - 5/24/89)
(Units are mg/L unless otherwise stated. NT=not tested.)
Chemical
3/31/86
RCHD
4/23/86
RCHD
2/2/88
VSWCB
5/24/89
Hav. Lab
Nitrate
Chloride
Sulphate
Total Solids
Volatile
Fixed
TKN
T.Phos
O.Phos
Ammonia (as N)
Nitrite
BOD
TOC
Fluoride
PH
Conductance
(umho/cm)
52.5
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
60.0
81.4
NT
541
274
267
INT
0.1
0.01
1.2
0.04
NT
NT
<0.1
6.35
NT
32.5
NT
NT
351
NT
NT
<0. 1
<0.1
0.01
<0.04
0.01
NT
NT
NT
6.5
31
12.1
NT
NT
678
NT
NT
NT
0.074
NT
NT
NT
NT
NT
NT
6.5
640
==_=====---==                                                             ==                  ==                  ==                 ==                  _=                  ==                 ==                  ==                  ==                 ==                  ==                  ==                 ==                  =
Table 27: Davis Well Pesticide Sampling (4/14/81 -
(Concentration in ug/L. NT=not tested.)
6/18/89)
Pesticide
4/14/81
VSWCB
3/13/86
RCHD
4/23/86	6/18/89
RCHD	Hav. L
Metolachlor
Alachlor
Dicamba
Atrazine
Endrin
Lindane
Methoxychlor
Toxaphene
2,4-D
2,4,5-TP
Paraquat
Oryzalin
Carbofuran
Disulfoton
<0.1
<0.1
<0.1
<0.1
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
58
3.8
6.1
28
<0.04
<0.1
<0.2
<0.5
<0.1
<0. 1
NT
NT
NT
NT
98
7.4
13
38
<0.04
<0. 1
<0.2
<0.5
<1.0
<1.0
<5.0
<1.0
<2.0
<0.5
<0. 1
<0.1
<0. 1
<0. 1
<0.08
<0. 1
<0.2
<0.5
<1.0
<1.0
NT
NT
NT
NT
31

Table 28: Davis Well Metals Sampling (RCHD, 3/13/86)
Metal                                    Concentration (mg/L)
--------------------------------------------------------
Arsenic	0.001
Barium	0.31
Cadmium	0.003
Chromium	0.0017
Lead	0.003
Mercury	<0.0003
Selenium	<0.001
Aluminium	0.07
Calcium	48.0
Iron	0.32
Magnesium	34.4
Manganese	0.37
Strontium	0.42
Zinc	0.05
Copper	0.11
Potassium	14.5
Sodium	<35.2
Nickel	<0.01
Boron	0.01
Calcium Hardness	119.9
Magnesium hardness	141.5
Ca/Mg hardness	261.0
Total hardness	264.0
Antimony	<0.0005
===      _== === z=== z=== -== ==== === == _==== ===========
32

Table 29: Monitoring Well No. 1 Sampling Results (Inorganics)
(Concentrations are in mg/L. NT = not tested.)
Chemical
8/86	8/86
SWCB	A&L
12/86
A&L
3/87
A&L
6/87	9/87	5/89
A&L	A&L	Hay. L
Nitrate
Ammonia
TKN
T. Phos
0. Phos
Diss. Sol.
Nitrite
pH
Conduct.
(umho/cm)
Alkalinity
Acidity
50
3
NT
NT
0.46
420
NT
NT
63
NT
NT
NT
NT
470
NT
4.3
81
NT
NT
NT
NT
300
NT
4.1
48
NT
NT
NT
NT
380
NT
4.5
57.48
0.2
3.9
4.0
0.03
603
0.02
4.9

574
0.3
20
22
NT
NT
NT
NT
225
NT
4.9

NT
NT
NT
9.0
NT
NT
0.004
NT
290
NT
6.7
NT
NT
NT
NT
NT
NT
NT	NT	420
NT	NT	NT
NT	NT	NT
33

Table 30: Monitoring Well No. 2  Sampling Results (Inorganics)
(Concentrations are in mg/L, NT = not tested.)
Chemical
8/86	8/86
SWCB	A&L
12/86
A&L
3/87
A&L
6/87	9/87	5/89
A&L	A&L	Hav. L
Nitrate
Ammonia
TKN
T. Phos
0. Phos
Diss. Sol.
Nitrite
pH
Conduct.
(umho/cm)
Acidity
49
7
NT
NT
0.30
450
NT
NT
57.42
5.0
5.0
8.0
0.02
590
0.08
4.4

652
110
63
NT
NT
NT
NT
470
NT
4.3
60
NT
NT
NT
NT
700
NT
4.9

NT
NT
91
NT
NT
NT
NT
416
NT
4.1
99
NT
NT
NT
NT
760
NT
4.5
1.78
NT
NT
0.218
NT
999
NT
6.7

420
NT
NT
NT
NT
NT
NT	NT
NT	NT
-------_----- ------c----_-_--------------------------
-------_------~-------------------------
Table 31: Monitoring Well No. 3 Sampling Results (Inorganics)
(Concentrations are in mg/L. NT = not tested. NC = not
clear.)
Chemical   8/86	8/86	12/86	3/87	6/87	9/87	5/89
SWCB	A&L	A&L	A&L	A&L	A&L	Hav. L
----------------------------------------------------------------
Nitrate
Ammonia
TKN
T. Phos
O. Phos
Diss. Sol.
Nitrite
pH
Conduct.
(umho/cm)
Acidity
53.73
35.0
35.0
10.0
0.05
651
0.02
4.5

NC
31
48
88
NT
NT
0.20
760
NT
NT
61
NT
NT
NT
NT
770
NT
5.8
61
NT
NT
NT
NT
1540
NT
4.0

NT
NT
86
NT
NT
NT
NT
770
NT
4.1

NT
NT
45
NT
NT
NT
NT
670
NT
4.3
0.26
NT
NT
0.120
NT
1880
NT
6.0
NT
NT
NT
NT
NT	1900
NT	NT
34

Table 32: Monitoring Well No. 4 Sampling Results (Inorganics)
(Concentrations are in mg/L. NT = not tested. NC = not
clear.)
Chemical
8/86	8/86
SWCB	A&L
12/86
A&L
3/87
A&L
6/87	9/87	5/89
A&L	A&L	Hav. L
Nitrate
Ammonia
TKN
T. Phos
O. Phos
Diss. Sol.
Nitrite
pH
Conduct.
(umho/cm)
Acidity
149.9
77.5
85.0
28.0
0.01
1202
0.10
4.6
110
112
NT
NT
1.12
1050
NT
NT
138
NT
NT
NT
NT
1400
NT
4.9

NT
NT
30 (NC)
NT
NT
NT
NT
680 (NC)
NT
4.0

NT
NT
109
NT
NT
NT
NT
830
NT
4.5
93
NT
NT
NT
NT
1150
NT
4.5
4.6
NT
NT
0.115
NT
383
NT
6.3
1852	NT
119	NT
NT
NT
NT	420
NT	NT
=====               _====               =====               ============ --------
Table 33: Monitoring Well No. 1 Sampling Results (Organics)
(Concentrations are in ug/L. NT = not tested.)
Chemical
8/86
VSWCB
8/86	12/86	3/87	6/87	9/87
A&L	A&L	A&L	A&L	A&L
2/88  6/89
VSWCB Hav. L
Metolachlor
Alachlor
Dicamba
Atrazine
2,4-D
Linucon
125   104
1.7	1
8.6	4
45	58
NT	NT
NT	NT
61
NT
NT
NT
NT
NT
22.4
NT
NT
NT
NT
NT
30
NT
NT
NT
NT
NT
86.3
NT
NT
NT
NT
NT
90
1.7
62
50
0.5
21
16
2
<0. 1
<0. 1
<1.0
NT
--I-_----------_---c--------'-----------
----_----------_------------------------
Table 34: Monitoring Well No. 2 Sampling Results (Organics)
(Concentrations are in ug/L. NT = not tested.)
Chemical
8/86   8/86
VSWCB A&L
12/86 3/87 6/87
A&L    A&L   A&L
9/87  2/88  6/89
A&L VSWCB Hav. L
Metolachlor 8.9
Alachlor	<0.1
Dicamba	<0.05
Atrazine	<1
2,4-D	NT
Linucon	NT
3
<1
<1
1
NT
NT
5	3.3
IT	NT
IT	NT
5
NT
NT
3
NT
NT
9.3
NT
NT
3.9
NT
NT
7.1
<0.1
NT
1.8
<0.3
<1.0
<0. 1
<0.1
<0. 1
<0. 1
<1.0
NT
N
N
1
NT
NT
1.4
NT
NT
-------------------------------------_--
-------------------------------------_--
35

Table 35: Monitoring Well No. 3 Sampling Results (Organics)
(Concentrations are in ug/L. NT = not tested.)
Chemical  8/86   8/86
VSWCB A&L
12/86	3/87	6/87	9/87	2/88  6/89
A&L	A&L	A&L	A&L	VSWCB Hav. L
Metolachlor 88
Alachlor	2.1
Dicamba	0.4
Atrazine	42
2,4-D	NT
Linucon	NT
141
2
1
44
NT
NT
40
NT
NT
41
NT
NT
86.5
NT
NT
49.1
NT
NT
131
NT
NT
33
NT
NT
104.5
NT
NT
41.9
NT
NT
38
2.6
<0.1
27
<0.3
18
16
1
<0.1
<0.1
<1.0
NT
-----=--------=========== _==========
Table 36: Monitoring Well No. 4 Sampling Results (Organics)
(Concentrations are in ug/L. NT = not tested.)
Chemical
8/86	8/86	12/86	3/87	6/87	9/87	2/88  6/89
VSWCB	A&L	A&L	A&L	A&L	A&L	VSWCB Hav. L
Metolachlor 150
Alachlor	11.6
Dicamba	NC
Atrazine	60
2,4-D	NT
Linucon	NT
128
6
2
60
NT
NT
36
NT
NT
37
NT
NT
54.5
NT
NT
21.3
NT
NT
33
NT
NT
18
NT
NT
50.8
NT
NT
28.2
NT
NT
21
5.3
<0.1
18
<0.3
5.7
26
4
<0. 1
<0. 1
<1.0
NT
=      =        =        =        =        =        =         =        =        =        =        =        =        =        ==--------=   =
36

IV. EVALUATION OF REPUBLIC CREOSOTING (MCLEAN CONSTRUCTION)
A.   SUMMARY OF PERCEIVED PROBLEMS

The Department of Waste Management summarized the
problems as follows:

The site, operated by Republic Creosoting Company from
1917-1972, is located in Chesapeake, Virginia, on the south
branch of the Elizabeth River. During that time the property
was owned by Reilly Tar and Chemical Corp. It is presently
owned by McLean Contracting Company which uses it as a supply
yard. for theair imarine construction operations.   The main
activity at the site by Republic was creosote and tar
treatment of wood. This also involved refining coal, tar, and
creosote.   Two open deteriorating tanks, which contain a
sludge of nearly 100% polynuclear aroimatic hydrocarbons
(PNAs), remain on site.  There is also a four-acre area of
mulch which came from treated lumber shavings. The soils of
this area are contaminated with PNAs up to 34%0.  Aqueous
samples from a drainage ditch running through the mulch area
conitained significant amounts of lead, cadmium, cyanide, and
mercury. Lead was also found in high concentrations in the
soil near one of the sludge tanks.
The high levels of PNAs are a hazard if contacted. They
are severe dermal irritants and can cause skin tumors. They
are readily absorbed through the skin, where they exert toxic
and/or carcinogenic effects. PNAs bind tightly to soil but
they may be carried with it into the surrounding waters. In
most organisms they are metabolized quickly, preventing
bicaccumulation, but shellfish are an exception.   Lead,
cadmium, cyanide, and mercury are toxic to aquatic organisms
at low concentrations.   Samples were also taken from the
drainage ditch, which empties into a 'marshy area adjacent to
the Elizabeth River.



B.   SUMMARY OF FIELD SURVEY AND SAMPLING PLAN

This site is expansive, the previous history is vague,
and the new property owners may be daily contributing to the
contamination problemi.  The original PAH problem is buried
with 0 to 2 feet of sedimnent  (potentially contaminated)
--dredged  from  the  Elizabeth  River,	which  makes  source
quantification much imore difficult.	We have defined five
problem areas, based on our evaluation of existing data and
our site visit. They are:
37

1. Quantitative assessment of the PAH source areas., Both
the area where the creosote holding tanks were and the
woodchip-mulched areas needed to be investigated as
potential sources.   Extensive sail gas analysis was
planned to be' used to outline contaminated areas, but
high groundwater levels precluded our doing so. Coring
was used to define depth and contamination levels.
Targetted PAH analysis was done using FID/GC.
Attempts were made to core out samples for PAH analysis.
The mulch is covered with dredgings from the Elizabeth
river, neither of these materials provides any support
for coring. The samples were taken with the assistance
of a backhoe. The water level was so high that sub-
mulch samples were not practical to take, in place of
them we took water samples from the holes dug by t h e
backhoe. These samples were filtered and extracted and
analyzed-for PAH levels.
2.   Groundwater:  Little was known about the groundwater at
this site, so modeling will not be reliable. PAH's are
relatively insoluble and would not be readily transported
but could migrate if high enough concentrations were
-present.   There  is also the question of other yet
unidentified contaminants.   Priority pollutant screens
were therefore used to test several areas. The Elizabeth
River borders one side of the property.   Four wells,
spaced evenly across the site, would give us insight into
the  groundwater  contamination  status.    Because  the
aquifer is undefined here two additional wells off-site
(opposite from the Elizabeth River) could help us
establish bass quantities and head values.
Note: We requested drilling quotes from several
firms for a class C well, 2"1 ID, PVC-cased.   Verbal
quotes were given to us: 60' to 80' wells would cost
approximately $2,200 each, assuminq no greater depth was
necessary and no hard rock encountered. The costs could
be less if the aquifer is enc-ounte-red at shallower
depths.   We did not drill themse wells in the present
study as we did not have sufficient funds to complete the
wells and the analysis program. As noted later, we t~hink
that this mnust be done to characterize the site.
3.   Surface waters,  drainage ditches,  and the Elizabeth
River:   Sampling of the Elizabeth River upstream and
downstream from the Republic site was conducted.   In
addition, run-off in the drainage ditches (both sides of
the property) was sampled and analyzed. These samples
were target-analyzed for PAH's, to determine the extent
of  PAH  contamination,  and  screened  for  priority
pollutants to see if there was anything else present
(keeping in mind that PAH's are not readily soluble in
38

water) .
4.   Biological  contamnination  from PAH's:   PAH's  readily
accumulate in adipose tissues of animals and in plants.
A measure of the extent of the contamination is to
determine how much the PAR's have accuamulated in local
aquatic species, land animals, and plants. Collection
of these biological specimens, to test whole body
digestion and subsequent PAR analysis, should be very
useful in this evaluation. Both grass and crab/fish
samples were taken. Neither sets showed any evidence of
PAH contamination.

5.   McLean Construction as a source of contaminat ion to the
site: During our site visit, we observed that there were
numerous empty 55-gallon drums on site, containing what
appeared to be asbestos-containing materials and tons of
dredged up sediment. We believe that these all need to
be evaluated.   The question we addressed is whether
McLean Construction is adding to the problem?  if so,
what and how 'much?


C.   ANALYTICAL RESULTS

Sampling locations are shown in Figure 14.  soil gas
analysis (all of which were negative) were carried out at
locations indicate-d by dot(-)  The water level being so high
< 24"1 in some places did not allow us to successfully use soil
gas analysis as a means of localizing the PAH contamination.
No meaningful results were obtained.

Total metals analysis and EPTPOX metals analysis
performed by Havens Laboratory shows very little contamination
for many samples (see Tables 37-38 for results of the metals
determinations and Table 39 reports on the PAR analysis). As
a matter of fact, -most metals are below the detection limits.
However, in the areas where the mulch is buried, the
contamination is very large.
Bulk asbestos analysis (Table 40 for resultz; Figur-e 14
for locations) was negative. other fibers such as cellulose,
fibrous glass, synthetics, and hair were detected.


D.   INTERPRETATION AND MODELING

The mulch area is very contaminated with PAH's. At this
time, no analytical data collected showed evidence that the
PAH contamination has spread from the mulch area or is
spreading. The toxicity of PAH's is well documented. The
Elizabeth River has several sources of PAR contamination (one
39

REPUBLIC CREOSOTING (McLEAN CONSTRUCTION) SITE MAP
0 ~~  ~A   P AALYStS  .oATeJ         IO


~~~ VA ~~~~~~~i
no ~~~~rt is ",GU th~ ~ ~ ~ ~~~-


16~~~~~~~~





D%t~~~~~~~~~











13 1 xx~~~~~~~~~~~~~~~~~~~U





.10~~~~~~~~~~~~~~~~~~~~~~~~~~~~(

- I
I
.21,11
r'.2 a -% -0
SLIMA Sell#~
OZ -1 - -'11-30

Table 37: Total Metal Analysis (Havens Laboratory, 8/11/89).
Concentrations are in parts per million (mg/L).
Locations of stations are shown in Figure 1.


-1 |Sample           Ag     As      Ba     Cd      Cr     Hg       Pb      Se

3115	<0.001 <0.001 <0.02 <0.001 <0.002 <0.0002	<0.005	<0.001

3117	<0.001	<0.001 <0.02 <0.001 <0.002 <0.0002	<0.005	<0.001

3119	<0.001	<0.001 <0.02 <0.001 <0.002 <0.0002	<0.005	<0.001

3121	<0.001	<0.001 <0.02 <0.001 <0.002 <0.0002	<0.005	<0.001

*       3123	<0.001 <0.001 <0.02 <0.001 <0.002 <0.0002	<0.005	<0.001

3125	<0.001  <0.001 <0.02 <0.001 <0.002 <0.0002	<0.005	<0.001

3127	<0.001 <0.001 <0.02 <0.001 <0.002 <0.0002	<0.005	<0.001

3129	<0.001	<0.001 <0.02 <0.001 <0.002 <0.0002	<0.005	<0.001
if,1      3131	<0.001	<0.001 <0.02 <0.001 <0.002 <0.0002	<0.005	<0.001
3131	<0.001	<0.001 <0.02 <0.001 <0.002 <0.0002	<0.005	<0.001


 l3135	<0.001	<0.001 <0.02 <0.001 <0.002 <0.0170	0.010	<0.001
3135	<0.001	<0.001 <0.02 <0.001 <0.002 <0.0002	0.010	<0.001

4135	<0.001	<0.001 <0.02 <0.001 <0.002 <0.0002	0.010	<0.001

41435	<0.001	<0.001 <0.02 <0.001 <0.002 <0.0002	0.008
<0.001

~--    4150	<0.001	<0.001 <0.02 <0.001 <0.002 <0.0002	0.005
<0.001























~~~~~~~* *~42

Table 38: EP TOX Metals Analysis (Havens Laboratory, 8/11/89).
Samples analyzed by EPA method 3010, EP Toxicity.
Concentrations are in parts per million. Locations of
stations are,as shown in Figure 1.
.:I
Sample     Ag      As     Ba      Cd     Cr      Hg      Pb      Se

4131	<0.001	<0.001 <0.02 <0.001 <0.002 <0.0002	0.032	<0.001

4133	<0.001	<0.001 <0.02 <0.001 <0.002 <0.0002	<0.005	<0.001

*I       4137	<0.001	<0.001 <0.02 <0.001 <0.002 <0.0002	0.010	<0.001

4139	<0.001	<0.001 <0.02 <0.001 <0.002 <0.0002	0.026	<0.001
4141	<0.001	<0.001 <0.02 <0.001 <0.002 <0.0002	0.007	<0.001

41415	<0.001	<0.001 <0.02 <0.001 <0.002 <0.0002	<0.005	<0.001

4147	<0.001	<0.001 <0.02 <0.001 <0.002 <0.0002	0.016	<0.001

ii1      4149	<0.001	<0.001 <0.02 <0.001 <0.002 <0.0002	0.007	<0.001
I
i

I














*                                          43

Table 39: PAH Analytical Results (Havens Laboratory, 8/11/89).
HLI I	Acena-   Fluoran- Naphtha- Pyrene Benzo(a) Chrysene Anthra-
(fidl)	phthene thene	lene                Pyrene                cene


74466                          -	-
3116

74468                          -         -              -
3118

74470                -         -        -          -        -            -          -
3120

74472                -         -         -                               -          -
3122
74474                -         -         -         -        -            -
3124

74476                -         -         -         -        -            -
3126

74478                -                   -          -       -            -
3128
74480
3130
74482
3132
74484
3134
74486
3136

74487-489
3137-39
74490
3140
74491
3141
74492
4130
74494             9800
4132
74496             120
4134
74498
4136
74500             11200
4138
74502             7400
4140
74504             235
4142
74506             8900
4144
74508             8700
4146
74510             9600
4148
all results are in ppm
analysis by FID/GC
-	600


4200	160000

-	300

-	50


17500

160




32000

31000




25000

10000

10000


500
11000  -
4000	200500

5800	185000

-	950


-	110000

-	I	95000

-	85000
1200
16000


15000
1500



-          11000   -


11000 -


-	11000	-
-	1 1 0 0 0
	-
- Z.,

Bulk Asbestos Analysis (Havens Laboratory, 8/11/89).
Detection limits = 1%. ND = None detected; ACM = Asbestos
Containing Material. Sample Locations are as in Figure
Table 40:

1.
Sample

2929




2921
% ACM & Type

ND




ND
% Other Fibers
5 cellulose
2 fibrous glass
15 synthetics
2 hair

2 cellulose
20 synthetics
% Other
76




78
I











.. -
1







4 -  1


I
















I




































. I
..

m
45

V.   FRAMEWRORK FOR DECISION SUPPORT SYSTEX
This framework was designed to assist the Department of
Waste Management (DWM) in performing evaluations of sites
suspected of potential risks.  The framework does not make
decisions but is only a decision support approach. It will
format data and provide simple or elaborate mechanisms of
exposure evaluation, in an effort to allow the knowledgeable
evaluator access to pertinent data (or to let him/her know
that pertinent data is missing). This effort is not meant to
replace CERCLA site evaluation protocols, but to parallel
them. The Framework is as follows:
Phase I (Supported by the computer software ERIES):
1.   COMPILATION OF DATA - Data can be acquired in many forms
and types, at different times and locations, and from
many sources. Data collected under different conditions
using different procedures must be compared. Subjective
and objective information is available. which makes
examining the data difficult at best. This first part
of the framework is basically an interactive program
designed to format the data in a simple, uniform manner
so that it can be more easily examined. Tt also will
assist in insuring that all the available data is
collected by helping define the pertinent data needed.
The format should allow clerical staff to enter data.

The data is compiled into two general groups:
analytical data and site characteristics data. Either
analytical data is target-selected or it can be collected
generally, utilizing large unknown parameters (like
priority  pollutant  analysis  rather  than  selected
chlorinated hydrocarbons). Site characterization data is
necessary for exposure evaluation.

2.   REVIEWING  AND  COMPARING  DATA  - This  multicomponent
section starts by defining a specific format for looking
at data. Q-uantitative data can be displayed as a two-
dimensional plot of multi-dimensional data and viewed as
follows, for example:
46

.ANALYTE: Arochlor 1254   SAM4PLE TYPE: Surface Water
Plot of concentrations in ppm


2  0.026   nd     0.005
Times
1  0.045	nd	0.018

1	2	3
Locations


Qualitative   data,   both   analytical   and   site-
characteristic, will be in lists. At all times within
this format, subjective statements on the quality of data
can be entered. This can include the age, reliability
of the data, etc., and should correspond to a 1-to-5
quality scale where I = good and 5 = bad. The review can
be as simnple as a list of compiled data (both analytical
and characteristic) or comparisons of analytical data.
Characteristic data is needed for modeling and exposure
assessment. The ability to compare related data allows
one to track trends (e.g., to determine if the
concentration is going up or down at a specific
location).

3. USE OF THE INITIAL DECISION TOOLS TO EVALUATE EXPOSURE -
This is broken down into two basic groups of procedures:

a)   Best case/worst case analvsis of c'ontaminants at the
sites. This is a powerful tool allowing one to
predict the worst possible case and subsequently
compare it to threshold values. In imany cases the
evaluation process may be terminated at this point,
when the worst case does not exceed the minimum
threshold. When minimal data has been collected and
or when there are numerous gaps in characteristics
data, creating larger uncertainties in the modeling
and evaluation processes, exposure evaluation can
still be estimated using this "Monte Carlo"
analysis.

This analysis program begins by making an
estimate of the variables and assuming either a
uniform or log-normal distribution.   By making
numerous calculation runs, randomly choosing the
distributed variables using the low parameters and
then the high parameters, one can generate a worst
case/best case probability plot of distribution
versus concentration. Subsequent insertion of TLV
information on this plot allows one to judge the
47

probability of whether the concentration has the
potential to exceed the threshold values anywhere
on the site.


Plot of Probability of Contamination versus
Concentration of Contamination













110






1 ~ ~ ~ ~ ~ ~ ~ ~l00



PrbabSilfiemolsto  assist in peictn movmen
and evnua l exo ue Simlfe m oel dono






spcii simplified models buasilt in pecton  movaluate


-Sracqur.Tee warenuerou models toavailuable. bthernnn


and standing water bodies.
-   A groundwater model to evaluate point source
contamination (2-dimensional).
-   A point source surface conta:~inants model to
show effects on undefined aquifer systems (3-
dimensional).

Phase 2   (Not included in the data base program, although we
have used EPA's program called PCGENS for this
analysis.)

4.   COMPLEX MODELING - The goals are to ensure the effective
use of models for appropriate problems. it is necessary
to defiJne all quantitative inputs required by the program
and to define the variability in default values and
ranges as a function of output. Most coimplex models
48

VI. COMPUTER ASSISTED DECISION SUPPORT SYZSTEM (ERIES)
ERIES, Environmental Risk Information and Evaluation
Syste-m is a phase I (edition 1. 0) program designed to provide
maximum utility in support of the first half of the
"Framework".   ERIES is a knowledge based (KB) system.   it
rearranges multiple sets of data bases and compiles them into
readily evaluatable formats of information.  The comparison
mode allows the user to look at trends, determine
localization, and track concentrations with respect to both
location and time.
ERIES is a menu driven, user friendly KB decision support
system. A brief description of the program follows:

ERIES, Environmental Risk Information and Evaluation System

I.   Title axid- Abstracts

These pages allow an evaluator the ability to screen
sites. You can menu-select a site (by title or nuimber)
and then examine the following basic information.

Title Pacre - basic information about the site: name,
address, contacts, etc.

sources Pacre - an interactive listing of sources of
information used to develop teh~ KB and those which are
available.

Abstract - brief description of problem and notes on any
litigation or pending regulatory action.

II.  Knowledge Base Data Entry

This section is where entry of analytical and site
chara-cteristic data are made.

Analvtical and Characteristic Data - includes five menu
options:
1. Source surface data base
2. Surface water data base
3. Unsaturated zone data base
4. Groundwater data base
5. Analytical Reference data base

Source Surface Data Base
- How imany sources have been identified?
- source #1?
- Status of source? stopped, still. contaminating?
- Analytes and data
50

1114 Population Data Base
This section includes a brief description of the area
including land usage and population density information
as well as plant and animal species in the area.

IV.  General Output

This section allows for a complete view of all collected
data or any i.ndividual groups (e.g., groundwater only).
This data is listed by groups and then subdivided by
types. It also includes all locations and times.

V.   Compare Data

With analytical results from several different sources
at several different locations and times,  it can be
strategically important to view selected data in several
formats.  This portion of the program will do just that.
It looks at analytical data and can view them two-
dimensionally.

This sort program can display any analytical paraLmeters
from the source/ surface, surface water, unsaturated zone,
and ground water.   It will ask user to def ine which
paramenters to display.  It will request you choose an
analyte group, then a specif ic analyte. The program will
then automatically generate a table all owing you to
compare location versus date.

Great ef fort went into developing a readily expandable
prograim. ERIES is capable of taking on many more tasks. The
program will require initial installation and a user
supporting document will be provided with the softwgare.
52

APPENDIX I


Actual Analytical Data for All Three Sites





Risk Assessment and Evaluation of Selected
Virginia Sites Within a Coastal Region







EP TOX METALS

Samples were analyzed by EPA Method 3010, EP Toxicity. Results are reported
as concentration, parts per million (mg/L), in the sample extract.
ANALYTE   SPL# LAGOOII	JUNKYDIA	JMUNKI	I sc
OLTI 70511	70513	70515	70517
I
Arsaenia
Barium
I	Cadmium
I	Cbromium
Copper

I Lead
Mercury
 I
Selenium
Silver
zinc
< 0.002
< 0.2
< 0.01
< 0.1
< 0.04
< 0.1
< a.0001
< 0.002
( 0.03
0.12
< 0.002
< 0.2
< 0.01
< 0.1
0.09
< 0.1
< 0,0001
< 0.002
< 0.03
0.78
< 0.002
< 0.2
< 0.01
< 0.1
< 0.04
< 0.1
< 0.0001
< 0.002
< 0.03
0.13
< 0.002
< 0.2
< 0.01
< 0.1

< 0.04

< 0.1

< 0,0001
< 0.002
C 0.03
0.05
I
2
APPENDIX 1

INORGANIC ANALYSIS
Results expressed in parts per million (mg/L).

ANALYTE                   SPLi Welltl	Well2A	Wellt2B	Well14B
HLII 70497	70500	70503	70509
Alkalinity (as CaC03)
Ammonia (as N)
Chlor ide
Solids, Total Dissolved
Nitrate (as N)
pe
Sulfate
Cyanide
Conductivity (umho/cm)
350
9.9
210.0
1120
0.525
6.4
120.0
< 0.01
1400
370
83
585.0
4670
4.550
5.0
1900.0
< 0.01
3300
350
96
615.0
5130
3.300
5.0
2100.0
< 0.01
3300
330
4.2
448.0
1640
0.700
5.9
105.0
< 0.01
1700
Results expressed in parts per million (mg/kg).

ANALYTE                     SPL# 101
SLI# 70512         70514         70516         70518
Ammonia (as N)
 Chl oride
Nitrate (as N)
Sulfate
< 0.8
0.014

0.8

640.0
< 0.8
0.014
< 0.1

60.0
< 0.8
0.005
< 0.1
40.0
< 0.8
0.014

5.2

60.0
3
APPENDIX 1

I.>
ORGANIC ANALYSIS

SaMples  were  acid/ base-neutral extracted by EPA method 3510 and  analyzed
by RPA method 8270, GC/XS using a DB-1 Col.
note (-)a none detected

*-I   COMPOUND               SPLt Welltl      Well2A        Well#4B
HLZf 70498         70501           70510
4-chloro 3-methylphenol
1,2,4-trichlorobenzene
acenapbthelene
fluorene
fluoranthene
pyrene
butyl benzyl phthalate
bis(2-ethy1hexyl)phthalate
0.023
0.009








0.044
0.034
0.027
0a008
o * 60
0.039


0.116
I
I
I
I
4
APPENDIX I


-   -  ---    -  -   --    - p
1130 East Market Street, Charlottesville, VA 22901
(804) 293-6000

ANALYSI S REPORT
----------------------------------------------------------------------------
HLI REPORT	: RE-144-4
I|   ~     REPORT DATE	: June 18, 1989
ACCOUNT	: 156
NAME	: Center for Risk Assessment
m             COMPANY	: UVA, School of Engineering &
Applied Science
ADDRESS	: Thornton Hall
Charlottesville, VA 22903
I0 |       PHONE #	: (804) 924-3954
PROJECT	: Alliance Fertilizer
SAMPLING DATE	: 5/24/89
SAMPLE TYPE	: Well Waters and Soils
I
ANALYTE            Well1#l	Welll	Well12	Well#3	Well4	Creek	Davis
73049	73055	73058	73061	73064	73067	73070
-
pH	6.7	6.5	6.5	6.0	6.3	6.2	6.5
 Cond. (umho/cm)	420	430	1100	1900	420	410	640
TDS (mg/L)	290	288	999	1880	383	387	678
1 P04-P, Total	0.004	0.002	0.218	0.120	0.115	0.112	0.074
(mg/L)
>mNO3-N (mg/L)          9.0       8.8     1.78     0.26       4.6      4.7
12.1
ORGANIC RESULTS
I HLI     -	DESC.	ANALYTE	CONC.
73051	Well #1	Alachlor	0.002
Metolachlor	0.016
DibutylPhthalate	0.028
373057	Well *2	Dibutylphthalate	0.010
73060	Well U3	Alachlor	0.001
Metolachlor	0.016
|3~~~~~ ~~Dibutylphthalate	0.016

73063          Well 14                 Alachlor	0.004
Metolachlor	0.026
!~~~~~ ~~~Dibutylphthalate	0.016
I73066	Creek                      -
 73069	Davis Well
I173074	Side Ditch	-
73075	Front Ditch	-
73076	Side of Drive	-
11 results in ppm unless otherwise stated
 rganic analysis-acid/base-neutral extract. by GC-MS


I APPENDIX 1                               5

HAVENS LABORATORIES,  INC.
1130 East Market Street, Charllottesville, VA 2Z901  (804) 293-6000


--   ARNALYSIS REPORT

HLI REPORT #	: RE-172-2
REPORT DATE	: July 5, 1989
ACCOUNT #	: 156
 NAME               Center for Risk Assessmernt
COMPANY	: UVA, School of Engineering & Applied Science
ADDRESS	: Thornton Hall
Charlottesville, VP 22903
PHONE #	: (804) 924-3954
PROJECT	: Alliance Fertilizer
 SAMPLING DATE	: 6/21/89
SAMPLE TYPE	: Runoff water


HLI #	DESCRIPTION                                     NO3-N (mg/L)

73667	3101, Behind Dike,  Alliarce property	2.26

73668	3102, Back center, A liance property	0. 51

 7:3669	33103, Top of Purcell Sprirng	9. 0

73670	3104, Run-off ditch into Purcell spring	0. 51
fromr l Alliar,ce-Davis pro:perty line

73671	3105, Run-off ditch behind hog lot	14. C0

i;i  73C72	3106, Connect for 3105	1C0. 8

73673	3107, Foarliy spot in creek 10:' past 3106	11.2

73674	3108, Branch #2. 28

 | 73675	3109, Old streaml	3. 9

73676	31101, Branch #3	3. 30

.1
I



APPENDIX 1                         6

1130 East Market Street, Charlottesville, VA 22901  (804) 293-6000

----------------------------------------
ANALYSIS REPORT


HLI REPORT #	: RE-194-1
REPORT DATE	: August 11, 1989
ACCOUNT i	: 156
NAME	: Center for Risk Assessment
COMPANY	: UVA, School of Engineering & Applied Science
ADDRESS	: Thornton Hall
Charlottesville, VA 22903
PHONE t	: (804) 924-3954
PROJECT	: Republic Creosote
SAMPLING DATE	: 6/12/89
iI              SAMPLE TYPE	:
page 1 of 3
1 TOTAL METALS
Concentrations are expressed in parts per million (mg/L).

4HLI    Ag         As        Ba       Cd        Cr        Hg       Pb        Se
(fldi)

.:! 74465  <0.001 <0.001   <0.02    <0.001   <0.002  <0.0002   <0.005   <0.001
3115
74467  <o.ool001   <o.ool001   <0.02    <o.ool01   <0.002  <0.0002   <0.005   <o.oo00
3117

 74469	<0.001	(<0.001	<0.02	<0.001	<0.002	<0.0002	<0.005	<0.001


74471	<0.001	<0.001	<0.02	<0.001	<0.002	<0.0002	<0.005	<0.001
3121
74473  <0.001   <0.001   <0.02    <0.001   <0.002  <0.0002   <0.005   <0.001
3123
74475  <0.001   <0.001   <0.02    <0.001   <0.002  <0.0002   <0.005   <0.001
I 3125
74477  <0.001   <0.001   <0.02    <0.001   <0.002  <0.0002   <0.005   <0.001
I 3127
74479  <o.oo001   <o0.001   <0.02 <   o0.001 < 0.002 <  0.0002   <o0.005oo    <o0.001oo
, |3129

74481  <0.001   <0.001   <0.02    <0.001   <0.002  <0.0002   <0.005   <0.001
3131
 74483  <0.001   <0.001   <0.02    <0.001   <0.002  <0.0002   <0.005   <0.001
3133
7
APPENDIX 1

IRE-194-1, cont.


IHLI t    Ag         As        Ba        Cd        Cr       Hg        Pb        Se
(fldi)
ï¿½---------------------_-----------------

1 74485  <0.001   <0.001   <0.02    <0.001    0.017  <0.0002    0.010   <0.001
3135

174497  <0.001   <0.001   <0.02    <0.001   <0.002  <0.0002    0.010  <0.001
4135

174505  <0.001   <0.001   <0.02    <0.001   <0.002 <0.0002 <0.0002    0.008    0.001
4143

i74512  <0.001   <0.001   <0.02    <0.001   <0.002  <0.0002   <0.005   <0.001
4150

EP TOX METALS
Samples were analyzed by EPA Method 3010, EP Toxicity. Results are expressed
s concentration, in parts per million (mg/L), in the sample extract.

(ILI t    Ag         As        Ba        Cd       Cr        Hg	Pb       Se

174493  <0.001   <0.001   <0.02    0.002    <0.002  <0.0002	0.032   <0.001
4131

I 4495  <0.001   <0.001   <0.02    0.001    <0.002  <0.0002   <0.005   <0.001
4133

;  4499  <0.001   <0.001   <0.02    0.002    <0.002  <0.0002    0.010   <0.001
137

,1R4501  <0.001   <0.001   <0.02    0.003    <0.002  <0.0002    0.026   <0.001
139

I 4503  <0.001   <0.001   <0.02    0.001         0.004  <0.0002    0.007   <0.001
141

4507  <0.001   <0.001   <0.02    0.001    <0.002  <0.0002   <0.005   <0.001
145


1147

74511  <0.001   <0.001 <0.02    0.001    <0.002  <0.0002    0.007   <0.001
1149
8
APPENDIX 1

I RE-194-1, cont.



I   --BULK ASBESTOS ANALYSIS REPORT,  EPA TEST METHOD 600/M4-82-020

REMARKS: Detection Limits = 1%. "-" = None Detected
I|~~ ~ACM = Asbestos Containing Material
IHLI #    FIELD SAMPLE #
.U    ISAMPLE DESCRIPTION
% ACM & TYPE
% OTHER FIBERS
% OTHER
74513  2929,
.I
5 cellulose
2 fibrous glass
15 synthetics
2 hair
76

I 74514  2921,
2 cellulose
20 synthetics
78
9
APPENDIX 1

__ - . ,aa  o. Lt:f   -narlotteSVllle, VA iZ4U1  (804) 293-6000

I                 -A--------------------ANALYSIS REPORT
HLI REPORT t
REPORT DATE
ACCOUNT t
NAME
COMPANY
ADDRESS

PHONE t
PROJECT
SAMPLING DATE
SAMPLE TYPE
: RE-194-1.B
: August 11, 1989
: 156
: Center for Risk Assessment
: UVA, School of Engineering & Applied Science
: Thornton Hall
Charlottesville, VA 22903
: (804) 924-3954
: Republic Creosote
: 6/12/89
: PAH results
page 1 of 2

HLI t	Acena-  Fluoran- Naphtha- Pyrene Benzo(a) Chrysene Anthra-
(fldt)	phthene thene    lene             Pyrene             cene
-I-
74466
 3116

' 74468

3118
.U! 74470
3120

1 74472
3122
r  -1
4474474
3124
i-. 74476
 3126
7 4478
128

74480
- 3130
74482
13132

74484
13134

74486
1 3136
I
0o
APPENDIX 1

RE-194-1.B cont.                            :-

LI 9	Acena-  Fluoran- Naphtha- Pyrene Benzo(a) Chrysene Anthra-
(fldi)	phthene thene    lene                Pyrene              cene

El4487-489
3137-39

14490              -        -        -         -
3140
'wi4491              -        _        _        _       _           _        _
141

4492                   -        -        -        -              -          600
130
4494
132
17500	500


160	-
11000
4200
160000
9800
300
I 4496
1134
.4498
I . 136
74500
.. 1138
74502
: 140
14504






4146

:1510
A18
120
50
200500


185000
11200   32000   1200
16000  1500
4000
5800
15000
7400


.235
31000




25000

10000

10000
950

110000
8900
11000
95000
11000
8700


9600
85000
11000
I results are in ppm
alysis by FID/GC
I
1I
APPENDIX 1
I

U
_
I
I
1
I
I
ii
I
I
(I
I
I
I
I
I
I
APPENDIX 2





Field Visit Reports












Risk Assessment and Evaluation of Selected

Virginia Sites Within a Coastal Region





I~~~~~~~FIELD	VISIT REPORTS

u    ~Suffolk Chemical field trip:	February 1989

Team Leader - Mike Lockhart
Field Personnel -'Sandra Neuse
Michele Sherer
Joey Romanoli

Sampling plan was developed to evaluate the following targeted
potential problem areas:
- possibility that lagoon area could contaminate ground water
- possibility that lagoon sludge itself is a hazardous waste
I        ~~- possibility that junk yard is contributing to groundwater
contamination
- potential that surface run-off from drum washing area is a
problem
- evaluate cur-rent state of groundwater
- evaluate any other problems that may appear

.1        ~~sampling was  conducted  as represented  in the  analytical
report. Analyses were performed using SW-846 and Standards Methods
14th Editioni proceedural guides.

Several problems and observations were noted:
- the weather was overcast and raining throughout sampling
I            ~~~~procedures
- well T#3 was under surface water and not able to be sampled
- well #4 had extensive silt and sediment at the 8' imark
I        ~~~- groundwater level was approximately loll
- the fill nozzle of the sulfuric acid tank extends out beyond
the drip/leak protection
- no visible debris on junk yard lot
- wells #2 and #4 have extensive odor
- all wells bailed a minimum of five volumes before taking
I             ~~~~samples
APPENDIX 2
2
I

Alliance Fertilizer field trip:
Ma,98
may, i989
Team Laader - Mike Lockhart
Field Personnel -Stan Havens
Therrell Hall
I(arl Klein

Sampling was performed in two teams. Team I (Mike Lockhart
and Karl Klein) sampled the following:
- front left corner of property for the presence of
potential "hot spot" of pesticide/herbicide containination
- collected 2x soil samples in drainage ditches to examine
surface water contamination

Team 2 (Stan Havens and Therrell Hall) collected monitoring
well, stream, and Davis well water samples.

Sampling was conducted as represented in the analytical
report. SW-846 and Standard Methods prceedures were followed for
the analysis.

Several problems and observations were noted:
- the weather was overcast.
- the entire region of the front left corner of the
property was soil gas analyzed using a KVA soil gas probe
and a HNU PID analyzer.   Only one sample showed any
deflection above background. A soil sample was taken for
priority pollutant analysis at this location. samples
were taken at 18 to 24 inches depth.
- Purcell Spring was not running, the level of water was
low.
APPEN~DIX 22
2

Alliance Fertilizer field trip 7#2:
Jue518
June 5,1989
Team Leader - Mike Lockhart
Field Personnel -John Martin

Preliminary modeling of the source loading of Purcell Spring
showed a potential source of sampling error. That being, it was
hypothesized that the stream was originally sampled in pooled areas
- not running. Data from previous reports and our most recent data
supported this conclusion.	This result makes determining the
loading source imupossible.	We needed to sample Purcell Spring
after a period of rain to evaluate loading sources.
The secretary at Alliance called us and let us know when it
was raining in Haynesville.   We conducted sampling at several
locations along the stream and tributaries and at several potential
loading (run-off) l-ocations.

Observations:
- raining and hot
- high suspended solids content
3
APPENDIX 23

Republic Creosoting (McLean Construction) field trip: June 12,1989
Team Leader - Michael Lockhart
Field Personnel -Stan Havens
John Martin
Jim Smith
Clint Butts
Carla Gauss
NOTE:   Mr Glen Metzl er, with the Virginia Department of Wqaste
management, assisted in the field trip and sampling.

Sa-mpling was performed in three teams:
Team #1 (Mike Lockhart, J'im Smith) evaluated and sampled the
following:
- soil gas analysis of mulch area to localize area of
highest PAH contamination
-    surface debris evaluation to see if any new problems may
exist

Team  -#2  (John Martin,  Clint Butts,  Carla Gauss)  sampled the
following:
- all surface water and river samples
- fish, crab, and grass samples
Team #3 (Stan Havens, Glen Metzler) sampled the following:
-    core sampling of mulch area
- surface soil sample drainage area
Sampling was conducted as represented in the analytical
report. SW-846 and Standards Methods proceedures were followed for
all analyses.

Several problems and observations were noted:
-	weather was cloudy
-	it rained hard for the second half of the field trip
-	coring' through  the mullch  area was  not  a realistic
approach. The mulch was too soft and irregular.
-    the on-site fdremnan offered us the use of a bac-khoe and
we dug holes to varying levels (and sampled) usin' it.
We will probably need to dig down the base soil in the
same fashion when installing monitoring wells.
-    the-re are dozens of drums and tanks with varying contents
scattered throughout the site.
-    there are  several  areas where  there  appears to be
asbestos-like insulation on pieces of scrap
-    samples #3135 and #3136 were taken from the adjoining
property with permission
-    sample #4150 was taken across the railroad tracks on the
other side of the road in front oE the property in a
drainage ditch
APPENDIX 24
4

REPUBLIC CREOSOTING (McLEAN CONST.)
Field Trip to Analysis Report Correlation Table

HLI#       Field#       Code         Description          Analysis
I
74465
74466
74467
74468
74469
7  74470
74471
74472
74473
74474
74475
74476
74477
74478
74479
74480
74481
74482
74483
74484
74485
74486
74487
74488
74489
74490
74491
74492
74493
74494
74495
74496
74497
74498
74499
74500
74501
74502
74503
74504
7450 5
74506

74507
745098
74510
74511
74512
74513
74514
3115
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
3140
3141
4130
4131
4132
4133
4134
4135
4136
4137
4138
4139
4140
4141
4142
4143
4144
4145
4146
4147
4147
4148
4149
2929
2921
water
water
water
water
water
water
water
water
water
water
water
water
water
water
water
water
water
water
water
water
water
water
grass
grass
grass
fish
crab
a	61 surface
a	6" surface
a	3' wood shavings
a	3' wood shavings
a	5' ground water
a	5' ground water
b	2" surface
b	2" surface
c	wood shavings
c	wood shavings
d	3' wood shavings
d	3' wood shavings
d	5' ground water
d	5' ground water
e	wood shavings
e	wood shavings
f	wood shavings
f	wood shavings
g	surface soil
g	surface soil
drainage ditch
pipe wrap
pipe wrap
metals
PAH
metals
PAH
metals
PAH
metals
PAH
metals
PAH
metals
PAH
metals
PAH
metals
PAR
metals
PAH
metals
PAH
metals
PAR
PAH
PAH
PAH
PAH
PAH
organic
met
organic
met
organic
met
organic
met
organic
met
organic
met
organic
met
organic
met
organic
met
organic
met
organic
asbestos
asbestos







































































hole
hole
hole
hole
hole
hole
hole
hole
hole
hole
hole
hole
hole
hole
hole
hole
hole
hole
hole
hole

CORRES PONDENCES
Enclosed at the end of APPENDIX 2 are copies of several
correspondences we had with others pertaining to these sites.
6
APPENDIX 26

July 9, 1989
TO:	Mr. David Siedle, McLean Construction

FROM:	Michael Lockhart, Field Operations, Systems
Engineering, Uniiversity of Virginia (804)924-0960

SUBJECT: Field Evaluations and Sampling at McLean Construction
Site - General Overview



This site is expansive, the previous history is vague, and
the property may contain other problems contributing to the
contamination problems. The original PAH problem is buried in 0
to 2 feet of sediment (potentially contaminated) dredged from
the   Elizabeth   River.	This   is  going	to	make   source
quantification  much  more	difficult.   We	have	defined	five
problem  areas based on our evaluation of existing data and	our
site visit. They are:


1. Quantitative Assessment of the PAH- Source Area(s) : Both the
area where the creosote holding tanks and the wood chip land
applied -area(s) were need to be investigated as sources.
Extensive soil gas analysis will be used to outline areas and
coring will be used to define depth and contamination levels.

2. Ground Water: Little is known about the ground water at this
site,  so  modeling  is  not going to be  reliable.   PAll's  are
relativqely insoluble and will not be readily transported, but
could migrate if high concentrations are present. There is also
the question of other yet unidentified  contaminents.   Priority
Pollutant screens are needed.   Due to cost restrictions, ground
water monitering will not be performed at this time.

3. Surface Waters,  Drainage Ditches,   and the Elizabeth River:
Sampling of the Elizabeth River upstream and downstream from the
Republic  site will be conducted.   In addition,  run off in the
drainage ditches (both sides of the property) will be sampled
and  analyzed.   These samples will be target analyzed for PAHl's
to determine the extent of PAHl contamination and screened for
Priority Pollutants to see if there are other contaminents
present (keeping in mind that PAHl's are not readily soluble in
water).

4. Biological Contamination For PAll's: PAEl's readily accumulate
in --plants and in adipose tissues of animals.   A measure of the
extent of the contamination to date is the determination of the'
concentration of the PAR's which have accumulated in local
aquatic species, land animals, and plants. Collection of these
biological specimens, whole body digestions, and subsequent PAS
analysis will be very useful in this evaluation.

5. Other Sources of Contamination to the Site: During our site
I    *visit  we  observed that there were numerous empty(?) 55
gallon
drums on the site and what appeared to be Asbestos-Containing-
materials on several pieces of equipment as well as tons of
dredged sediment.   These all need to be evaluated.  The overall
I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


I        ~~The following generalized sampling plan has been
devised to
address the above problems:

1. a)   Use  soil  gas  analyzer  to  outline  areas  of   PAR
con tam ination.

b)  Take  soil samples (up to six feet in depth) to  quantify
the PAH contamination. This will include up to 50 PAH and 5
Priority Pollutant sampDles.

I    ~2. No groundwater testing at this time.

3. Analyze runoff for Priority Pollutants (6-10) and PAHI's (15-
I     ~~20).
4. a) Collect ten soil samples from targeted locations. Analyze
for PAH's and metals.

b) Collect foliage samples from each of these ten locations
and analyze for PAll's.
c) Collect two to four animal/fish samples from both zones A
and B and analyze for PAHl's (sampling zones A and B are along
I     ~~the Elizabeth River on th upstream/downstream portions of
the
property).

5. a) Inspect surface debris for asbestos and take samples as
needed.

b) Take samples of the containers and other possible sources
I     ~~of contamination (unknown at this time).

c) Take several random samples of dredgings.


We will be prepared to take up to:

*1      ~100 - soil/water PAll samples

25 - Priority Pollutant samples (acid/base-neutral
I	-~-    extractables)

*	~~35 - metal samples (for total and EP toxicity metals)
25 - other related inorganic and general samples

U
I^~~~~~~~~~ ~~~Alliance Agronomics, Inc.
6526 Mechanicsville Turnpike
Iw V~~~~~~~~~~~~~~~~~ ~Mechanicsville, Virginia 23111
([804) 730-2900
I1A
April 6, 1989



Dr. Ralph O. Allen
University of Virginia
Chemistry Building
McCormick Road
Charlottesville, VA 22901

Dear Dr. Allen:
Enclosed is the
Technologies in
the changes and
groundwater management plan prepared by Environmental
1988 for our Haynesville site. I have also inctuded
an addendum to the plan.
The crop history of the fields adjacent to our site has been difficult
to determine.  There are three fields that your group asked about--two
to the west and one to the east of our site. We are unable to
reconstruct the crop history or the crop protection chemicals used
on the two fields to the west of us. These fields are owned by Mr.
Wilson I. Davis and are located in front of and behind his home.
I believe he rents out these fields to saneone in the cocnunity.
The 20-acre field to the east of
before 1987. In 1987, 1988, and
farmed it.
us was farmned by Mr. Douglas Lewis
1989, I believe Mr. George Self
Normal
Rate

1 1/4 pint/A
3 pints/A
1 quart/A

1/4 pint/A
3/4 pint/A
Year
Crop
Chemical
1982



1983
Dual (Metolachlor)
Aatrex (Atrazine)
Toxaphene

Banvel (Dicamba)
2,4-D
Corn



Wheat
Beans
Dual
Lorox
1 1/4 pint/A
1 lb/A
1984

1985
Corn
Same as 1982
Same as 1982

Same as 1983

Same as 1983
Wheat
Same as 1983
Sarme as 1983
Beans
I   Creating More Profit Per Acre for the Farmer.

I
I
I



f
l
]
I
;"1
I
I-
Dr. Ralph 0. Allen
Page 2
April 6, 1989
Normal
Rate

1 1/4 pint/A
3 pints/A

?

1 1/4 pints/A
1 lb/A

Same as 1986

?
Year

1986
Crop

Corn


Wheat

Beans
Chemical

Dual
Aatrex

Unknown

Dual
Lorox

Sarme as 1986

Unknown
1987
1988

1989
Corn

Wheat
I hope this information is beneficial. I am sorry for the delay in getting
this to you.

Yours truly, I
G. W
Pres:
bjw

cc: Mr. Glen Metzler

i/
:I

IRN


COiMMONWE1ALTH of VIRQINIA -
STATE WATER CONTROL BOARD
2111 Hamilton Street

Ple4se imply 1:  TdewterU Rtgion- Kimunwoor Otie
P. O. Bofx 563
Chuwch Sure*t
KiUmamock. VigilWa 22482-0669

Jamiary 12,    s88
Richaid N. uwton
Excutiuv Director

Post Offics aBx 11143
Ricrmond. Virgin.a 23230-1143
t041 367-0058
Mr. mahash P. Shah.
Center for Pisk .anagIA  t
of   i         Systes
Appliid Math a2ildirj, 1&oon 103
University of Vixginia
Charlott,esvi%Je, VA 22901

e$/1~'VA
)eaa Mr. Shah:

In rsp=nsa to y=r lettar, attached is an enlarged  p shwa &q the esuct
stream sam?ti5 lcati=,s for the 2/2/88 atudy.  Also atr.-adue is a cop of the
\v    file maP for the 7/30/81 st-eam sa=lin  stations.  Sirna I was mt present
uirn  that sau2plig, the attachad =p is the best I can p=ride.  The 7/81
saMles veze co1llaetad by T. L. Switzer, who is n rc 1igar with this agency.
Please nota that Mr. Switzer's labeLirt g of the 1'kauthorized Discharge
Lzcation" arn "PuReall Spr-irjr, (Station 3) is iorz=crt. The ura.tthbrized
discharge 4   into -race=ll Spring, wc  is identifisd as Station 1 on my 2/as
map.  I asa=e tthat his rmainirg stations are correcty labeled, and that his
Station 4 co rezrcrAs with my Station 4, and his Station 5 is sligatly
dcwsream of my Station 6.
(804) 367-6763 if yCU. have fuzther
questions
Plaase contact David GLLscan at
raqaztd. this facility.
Sitmeraiy,


B.  Fs lei
Miviroruental Specialist
cc: David Gus=an

APPENDIX 3


Monte Carlo Estimation Procedlure





Risk Assessment and Evaluation of Selected
Virginia Sites Within a Coastal Region

As mentioned earlier in this report, it is often difficult to
accurately predict the interaction of the various factors that
govern the transport of che-mical contaminants in the environment.
Because imany of these factors have a probabilistic distribution
(such as wind direction and speed, rainfall over a given period,
and rate of groundwater flow), it is impossible to provide precise
deterministic basis for analyzing the fate of chemicals released
into ecological systems. For this reason, a model used to analyze
contamination needs to have a probabilistic component if the
decision-maker wishes to analyze anything other than the expected
exposure to contamination.

Such a model needs to have, at its heart, random numbers
generated according to known probability distributions that can
then be used in analytic equations governing chemical transport
and decay.-  The use of such a model is known as Monte Carlo
simulation because of the association between probability
distributions and games of chance.

As an aid to understanding the usefulness of such a
simulation, we should consider the problem facing current
deterministic -models.  For example, in the surface water model for
riverine systems in PCGEMS asks, in the environment input file,
for the averag~e levels of:

- Oxidant Radical Concentration
- Rainfall (per month)
- Cloudiness (on a scale of 0.0 [clear] to 10.0 [full cover])
- Ozone in the Atmosphere
- Relative Humidity
- Atmospheric Turbidity

It then uses these average levels of environmental factors to
calculate chemical breakdown from such things as photolysis and
organic adsorption. Over a long period of time with the system in
steady state, you would certainly expect to see results similar to
those predicted by this model. However, these expected values do
not provide a single clue as to the variability of chemical
concentration.    it  can  easily  be  seen  that  each  of  the
environmental characteristics listed (and there are many others
not included in this list) can vary over a relatively wide range
on any given day, and each has a 50!% chance of being above or below
the mean given in the input file, although none of this variability
is taken into account by the program itself.   The ef fect of
combining the differences from the mean of each of the input


I This refers to EXAMS-11 (Exposure Analysis Modeling System) ,
developed by the Environmental Protection Agency for use in PCGEMS
(Personal Computer Version of the Graphical Exposure Modeling
System) as a tool to analyze the expected concentrations of
chemicals introduced into surface water environments.
APPENDIX ~31
I

characteristics will alter the concentration of a chemical ever
time, although the mean concentration will remain about the same.
The greater the -variance that each of the input parameters has, the
I  greater  the  variance  will  be  in  chemical  concentrations;
conversely, in temperature regions with only smali aeviations from
average weather conditilons, there will be only a small variation
I    ~from the mean chemical concentration.
The point of a Monte Carlo simulation, then, is to find
probability distributions for the input chararcteristics. The model
input is then taken as a set of random variables f rom these
distributions. The variables interact according to the equations
governing transport and decay already established in the model's
U    ~code and the result is a system for determining the
distribution
of the chemical concentration. To begin with, one run is maae with
input numbers taken from the random variables (for rainfall and
relative humidity, etc.) . The output from the model is a level of
concentration. Taking another run of the mnodel with another set
of numbers from the same distributions will give a different level
of concentration. A number of successive runs will give data that
I   ~represent the distribution of chemical exposure, which can be usea
for an analysis of variation, or extreme event analysis, or a
number of other useful techniques that cannot be performed with an
I   ~expected value alone.
The crucial aspect of this method is the importance of the
input probability distributions. The more accurately they can be
found, the better the results from the simulation.  Fortunately,
much of the data can be found in geological and weather reports
made by outside agencies.   Unfortunately,  the distribution of
I   ~chemical release from the source itself is unknown and must be
determined for tChe input. often the source itself is indeterminate
and must be estimated; making measurements can be extremaely costly
I   ~and time-consuming (particularly for groundwater contamination).
With only a few measurements of source concentrations, it is
difficult to find the actual distribution of chemical release--
although it can (and -should) be estimated, for example with a
triangular distribution.
This is where the advice of experts and research in other
I   ~fields can be valuable.  If studies of similar chemicals reveal a
tendency towards one type of release distribution and the data
taken from the actual site agrees well with that, then that type
of distribution should be used.   Similar reasoning follows for
using the ad-vice of technical experts; if an expert cani determine
a particular type of distribution for some chemical release and
give a valid supporting argument, that distribution is a strong
possibility for use in the input parameters.
on the down side of the Monte Carlo technique is the argument
I   ~that it requires a great deal of computer time.  This is often
true
when modeling an annual cycle of chemical contamination--after all,
APPENDIX 32
2

the more runs that are made, the bette-r the results will be.
However, initial groundwork on the PCGEMS models running on a 20
MHz 80386 computer (with the required 80387 math coprocessor) shows
running time for an annual surface water model expected value
analysis to be under a minute. Such speed certainly opens up the
possibility of a great many runs if some method can be derived for
automating the update of the input file based on the required
probability distributions.

If other, less complex, models can be found to approximate
chemical transport with fewer inputs, then the same reasoning can
be used to adapt them to this type of simulation in order to
provide the decisionmaker with a useful distribution of chemical
contamination.
APPENDIX 33
3

APPENDIX 4
Model'Used for Alliance and Suffolk Sites





Risk Assessment and Evaluation of Selected
Virginia Sites Within a Coastal Region
h

Model for Risk Analvsis
The common two-dimensional equation governing contaminant migration
in uniform one-directional flow from a slug point source without
I     adsorption and radioactive decay is (Hunt, 1983; Walton, 1989):

C = f(x,y,v ,Ys,m,n,A ,AT,t)C2                     (1)
where f( ) = 1.064xl- v           -((X-Vst) /4ALvst] + y /(4ALvSt))/
rmnonys(ALA   t]
C0 = difference Tetween solute concentration injected into
 aquifer and native solute concentration in mg/L
If~ C  = change in aquifer solute concentration due to solute
injection in mg/L
x,y = cartesian coordinates of monitoring wells in feet
I|~ ~       m = aquifer thickness in feet
vc = volume of injected mass in gallons
vs = seepage velocity without adsorption in feet/day
 = (K/n)dh/dx
K  = hydraulic conductivity in feet/day
AL = longitudinal dispersivity without adsorption in feet
AT = transverse dispersivity without adsorption in feet


To account for adsorption AL, AT, v5 , and C are divided by a
retardation factor defined as (Marsily, 1986; Walton, 1989):

where      Rd = 1 + [(Dbs/np)Kd]                     (2)
I
Rd = retardation factor
Dbs = bulk density of dry aquifer skeleton in g/cm
n	= aquifer actual porosity
K	= distribution coefficient
d
Radioactive decay is simulated as (Marsily, 1986):
C  = Ce-Zt                                  (3)
 = 0.693/h1
Cr = concentration of solute with radioactive decay
Us~~~ ~in mg/L
C = concentration of solute without radioactive
decay in mg/L
I|~ ~t = time after radioactive decay started in days
h1 = half-life of substance in days
APPENDIX 4

Derivation of Distribution Function
The initial concentration, C0 can be treated as a random variable
(because of uncertainty in our knowledge about its value) with a
distribution function FC (Co). Let Fc(c) represent the distribution
function of C. Then
Fc(c) = P(C < c)
= P(fc0 < C)
= P(C < c/f)
= Fc (c/f)                             (4)


ADolication

The above model was applied to the Suffolk problem. Parameters
required for the model were estimated as shown in Table 1 of this
Appendix. Assume that a sulfuric acid source is located at Well
no. 2, which injects 100,000 gallons of sulfuric acid at present,
with a concentration whose distribution function is to be assessed.
In absence of appropriate data, the distribution can be assumed to
be of a triangular form (Kelton and Law, 1982):

= 0	when C  < a
Fco(Co)   = (Co-) /[(O-a) (T-2)]	when a < CO < T
- 1 - (B-Co) /[(B-a)(B-T)]	when T < C  < B
= 1	when Co >

[a,3]  = interval in which c is believed to lie
T = mode; the most likely value

From Equation (4) the distribution function for the concentration
at Shingles Creek, which is 600 ft away (distance estimated from
the U.S.G.S. 7.5 minute topographic map) is given by:

= 0	when C < af
Fc(C)     = (C/f-a)2/[(I-a)(r-a)]	when af S C < Tf
= 1 - (B-C/f) /[(B-a)(B-T)]	when if < C < Bf
=                i	when C > Bf
It is assumed that  a=500 mg/L,  B=3000 mg/L and  r=2000 mg/L.
Graphs of the input and output concentrations are shown in Figures
1 and 2. From Figure 2 it becomes clear that the chance of the
concentration at the Creek exceeding 60 mg/L is negligible.
APPENDIX 4
2

Table 1: Risk Analysis Parameter Values
Parameter
Value
Remarks
Hydraulic conductivity, K
Effective porosity, n
Hydraulic gradient ,
15 ft/day
0.3
0.002 ft/ft
conservative
conservative
conservative
(LES, 1986)
Aquifer thickness, m
Long. dispersivity, AL

Trans. dispersivity, AT
30 ft
10 ft

2 ft
average
subjective
estimate
subjective
estimate
seepage velocity, vs
volume injected, vc
0.3I ft/day
100,000 gal.
computed
conservative
Retardation factor, Rd
1
conservative
(no retardation)
Half life, h1
infinity
conservative
(no decay)
References

Hunt, B., Mathematical Analysis of Groundwater Resources,
Bulterworth & Co., Ltd., 1983.

Konikow, L. F., and Bredehoeft, J. D., "Computer Model of Two-
dimensional Solute Transport and Dispersion in Groundwater,"
Book 7, Chapter 2, United States Government Printing Office,
1978.
Simulation Modelincr and Analysis,
Law, A.M., and Kelton, W.D.,
1982.
Law Environmental Services, Division of Law Engineering Testing
Company, Report of Hydrogeologic Investigation - Phase II,
LES project number SS6630, Suffolk, Virginia, 1986.

Marsily, G. de, Ouantitative Hvdroaeolocv, Academic Press, 1986.

Scherer, M., "Modeling of Groundwater Contamination at the Suffolk
Chemical site," Undergraduate Thesis, Systems Engineering
Department, University of Virginia, 1989.

Walton, W.C., Analytical Goundwater Modelinc, Lewis Publishers,
1989.
3
APPENDIX 4

APPENDIX 5


Role of Risk Assessment in Site Evaluation





Risk Assessment and Evaluation of Selected
Virginia Sites Within a Coastal Region
*
I
I
 I
I
 I

I
I
I
I
I

As in any project involving risk analysis, there must first
be a process of site characterization for each area being
investigated. This characterization of the problem is necessarily
an important aspect of any successful risk analysis. in the case
of chemical release into the environment, this phase of the study
involves answering four questions in detail:

1.	What are the conditions of exposure?
2.	What are the adverse effects resulting from exposure?
3.	What is the relationship between exposure and effect?
4.	What is the overall risk?

The first problem, that of gefining the conditions of
exposure, is often the most difficult one for ecological systems.
The overall goal is to determine who (or what, in the case of
wildlife) will be exposed to chemical releases from the site in
question--and in what amounts for how long. The problem faced in
addressing these -questions is the inherent difficulty in
determining the ultimate fate of cheimicals released into the
environment.  Given that it is impossible to accurately predict
and analyze all of the factors in an ecological system, we must
use estimates and averages in determining chemical transport: an
average windspeed or aquifer flow, an average pH level in the soil,'
or an estimated annual rainfall. it is because of the uncertainty
inherent here that we must be aware of the variability in possible
exposure levels as well as the average amounts of exposure.

This dif ficulty in establishing analytic relationships between
the various factors in a chemical transport mechanism often makes
it necessary to turn to computer simulation in order to derive a
distribution for exposure. A Monte Carlo simulation (of which more
will be said later) can be particularly useful in this regard--
especially since it is often difficult even to accurately determine
the source of chemical releases.

Another topic to be addressed is the method of chemical
transportation.   Is it an atmospheric pollutant,  or does it
contaminate groundwater? Will it continuously evaporate or does
it only break down in direct sunlight? Much of this information
can be obtained in laboratory experiments on the chemicals being
analyzed, but often measuremients on-site reveal the predominant
mode of chemical movement, as the varying concentrations between
surface water, groundwater, the atmosphere, and the soil can all
be measured to some degree. The extent of biological absorption is
more difficult to discover--it is of-ten impossible to -measure
chemical concentrations in wildlife without killing the animals in
question.   Biological absorption, then, must often be estimated
froma properties of the chemical itself.      Indirect exposure is
a necessary consideration in addition to the hazard of direct
exposure.   As an example,  consider chemical waLste dumped into
rivers leading to oyster beds.   People may never swim in that
water, but they can potentially be poisoned by eating enough of

APPENDIX 5                       1
0

the oysters.   Indeed, this is a primary source of concern for
certain types of chemical releases and may thus indicate where
attention should be foc-used in studying contamination.
once we have identified what is being released into the
environment, how it is being transported, and who (or what) is
being exposed to it, we must define the adverse effects that
result. Will the chemical cause cai~er? Can it alter human or
animal  genetic patterns?   Will  it physiologically damage the
exposed population?  These are impDortant questions in analyzing
the risk of exposure; presumably., this type of information is
available for the chemicals being studied. That is, the sites in
question are being investigated because they are potential sources
of chemicals that are known to have adverse effects on the
environment. If the effects of the chemicals being released are
not known, then determining these effects should be a primary focus
of this phase of study.
The nature of the chemical also comes into play here. Does
the chemical gradually decay once in the body, or do successive
doses build up, possibly to lethal levels? If the substance is
chemically stable in the environment, populations could,be subject
to a low level of exposure over a long period; in other instances
there could be a brief exposure of a large magnitude. Different
types of exposure can lead to vastly different effects. There is
also the problem of delayed effects.   There may be a long time
between initial exposure and the manifestation of adverse effects
as, over time, the chemical builds up or works its insidious change
in the human population. Just because there is no indication of
a health hazard at a site presently does not necessarily mean that
no hazard exists.
The third step is, logically, to relate exposure with adverse
effects.   Obviously a small dose of a chemical will not be as
dangerous as a large dose; however it is necessary to ascertain
whether chemical releases are within reasonable amounts. At this
point some kind of relationship between an amount of exposure and
an amount of adverse effect needs to derived in order to determine
the overall consequences of environmental release. The possibility
that some of the exposed population may be more susceptible than
the average must be considered. People with special allergies imay
have a far greater reaction to an introduced substance than others;
what may be a "safe" level in one person could conceivably be
lethal in another. In any event, if such data is not available for
chemicals being released at the sites under scrutiny, studies
should be made to determine these effects.
-. The interaction between different populations must be
considered as well. in any ecosystem there is a delicate balance
in the food chain; a chemnical that only affects speckled trout can
still ixidirectly affect all the animals that feed on or depend on
speckled  trout for their existence.    Thus,  while  it may be
APPENDIX 52
2

impossible to test all of the animal or plant species that may be
exposed to a certain contaminant, an indication of adverse effects
in even one species must be carefully considered.

it is entirely possible that a che-mical can be found to have
little or no effect on the environment even in the maximum
concentrations found at the source. Although this may be the case,
the prospect of multiple sources should be evaluated. For example,
a factory may be discharging waste into a stream at what are
thought to be reasonable levels.  If, however, there are fields
downstream with pesticide runoff, the chemical toxicity can be
pushed to a dangerous level. While there 'may be nothing that can
be done about the pesticide, cleaning up the factory's- discharge
may be warranted, to protect the environment.

The final problem in site characterization is estimating the
risk, both to individuals and to society as a whole. It is here
that the various effects studied earlier are compared, to ascertain
which are the most undesirable, which affect the most people or
greatest area, and which pose the greatest threat to the natural
order of things. This process of evaluating risk can often become
the political one of judging the acceptability of risk--but that
is not its purpose.   It should only provide a framework within
which decisionmakers can determine what presents the greatest
hazard and is thus in greatest need of correction.   Tt is in
combining the risks found in studying various sites that
decisionmakers can determine which ones are in the greatest need
of attention.



















The-primary reference for this appendix was:

Lowrance, William W., Of Accentable Risk. William Kaufman, Inc.:
Los Angeles, CA, 1976.
3
APPENDIX 53