WS-3893-3893-M_DENR_19890331

•
Colonial Pipeline Company
LENOX TOWERS 3390 PEACHTREE ROAD N.E. ATLANTA, GA. 30326-1108 [phone]
V.A. YARBOROUGH
Senior Manager -Engineering
March 31, 1989
Mr. Stephen Williams
North Carolina DNR & CD
Winston-Salem Regional Office
Division of Environmental Management
Groundwater Section
8025 North Point Boulevard
Suite 100
Winston-Salem, North Carolina 27106-3295
Re: January 23, 1989, Tank Overfill
Greensboro Junction Tank Farm
Guilford County
T.W. CERVINO
Administrator -Environmental
P_ RECEIVED
N.C. Dept, f;�D
APR 3 1989
Winston-Salem
Regional Office
Dear Mr. Williams:
Please find enclosed the proposed Remediation Plan for the above referenced
incident at Colonial's facility. A preliminary investigation has shown that the
extent of contamination is minimal. Emergency response efforts and containment
measures were the reasons for the small environmental impact.
Colonial is confident that the enclosed proposal is satisfactory to accomplish
the remediation goals and objectives. A tentative schedule has been established
to begin work in the remediation area the week of April 10.
Should you have any questions regarding this proposal or inhibitions about the
above scheduled starting date, please contact me at 404-841-2253 or the mailing
address below.
Yours very truly,
Y,
D. V. Pearson
Engineer
pl
Enclosure
cc: T. W. Cervino
W. C. Edwards
J. E. Marder
H. R. Melendy
V. A. Yarborough
MAILING ADDRESS: P.O. BOX 18855 ATLANTA, GA. 30326-0855

t � •
y
Colonial Pipeline Company
REMEDIATION PLAN
GREENSBORO JUNCTION TANK 812 OVERFILL
January 23, 1989
Introduction
On January 23, 1989, Tank 812 at Colonial Pipeline Company's Greensboro Junction
Tank Farm was overfilled, releasing approximately 93,292 gallons bf gasoline into
the environment. Earthen tank dikes contained the spill, limiting the amount
of impacted area (see Figure 1). Emergency response measures, which also aided
in reducing environmental impact, included:
• providing a foam blanket over the product to contain the vapors, thereby
limiting evaporation, and
° introducing a water bottom into the inundated dike to minimize the extent
of product infiltration.
Recovery was expedited by pumping the mix of spilled gasoline and water into a
neighboring tank. The product, after separation, was then reintroduced into the
pipeline system, and the water was treated through Colonial's on -site, permitted
wastewater treatment system.
Soil Remediation Objectives
Even though a water bottom was established in the affected dike as quickly as
possible, some spilled product remains entrained in the soil surrounding the
tank. The extent of soil contamination was initially estimated through soil
sampling on January 31, 1989. Figure 2 delineates the locations and depths of
samples taken to identify the horizontal and vertical Dts:
of the
contamination. The analytical procedure used was EPA Method 4etroleum-
hydrocarbons). The following table summarizes the analytical
Gravel
Depth
Concentration
Thickness
Test Pit
(Inches)
(Inches)
/
1
0 - 5
353
3
5 - 8
112
9 - 11
< 10
2
0 - 5
524
3
5 - 9
35
3
0 - 8
2820
8
9 - 12
22.5
4
0 - 9
6020
9
9 - 12
64.2
15 - 18
14.4
5
0 - 5
600
8
5 - 8
177
9 - 12
246
1

M
•
Colonial Pipeline Company
A background sample was also taken in the same tank dike area near the toe of
the northern dike as shown on Figure 1. This point is upgradient of the spill -
affected area. The petroleum -hydrocarbon concentration for this sample was
<10 ug/g.
These initial soil samples show that the majority of the remaining product is
in the upper eight inches, and most of that eight inches is gravel. Since the
penetration depth was kept to a minimum, Colonial believes that remediation can
be accomplished in place. Additional soil samples were to be obtained on
March 15 to identify the maximum depth of product penetration; however, inclement
weather caused ponding in the dike, which prevented representative sampling.
These deeper samples will be obtained later as described in the Contaminate Plume
Delineation Section. After confirmation of the vertical infiltration, the soil
can then be cultivated to attenuate the contamination and achieve desired
petroleum -hydrocarbon composite concentrations of <100 ug/g.
Hydrogeoloeic Data and Groundwater Quality Concerns
The available geologic information in the area is from two sources:
1. visual inspection at the time of sampling and
2. boring logs, from 1963, obtained prior to construction of the tank farm.
The soil is a reddish -brown, silty clay throughout most of the top ten feet.
The boring logs showed that deeper soils maintained silty composition with
evidence of decomposed rock. From sampling observation, it was noticed that a
distinct grayish clay layer existed at about one to two feet below the surface
on the north and west sides of the tank. The other circumferential area of the
tank also exhibits a relatively impermeable layer within the vertical strata.
This characteristic was evidenced when the additional soil samples were to be
obtained, and all former test pits were holding water from a previous rainstorm.
Most residences and commercial establishments in the vicinity of the tank farm
are connected to city water. The closest drinking water well serves Colonial's
facility control building, which is about 600 feet away from the spill site.
However, this well was installed at a depth of more than 100 feet. The nearest
residential users of groundwater are located on the south side of the tank farm,
about 3,600 feet from the spill site.
Groundwater quality does not appear to be jeopardized from this release. The
nitial soil samples have shown only shallow penetration of the spilled product.
to, grou t fn tit+c area of the- spill iabetW*arV SI*,-and
twenty feet, ccording to visual inspection and previous boring logs. When Test
Pit 4 was dug to a depth of six feet, no water was encountered, „gs
show -an approximate depth of groundwater of about -twenty feet: Therefore, due
to the relatively small volume of residual product in the soil (which will be
reduced by remediation activities), low permeability of the upper soil horizons,
and fairly deep water table, no degradation of groundwater quality should occur.
2

•
0
Colonial Pipeline Company
Contaminate Plume Delineation
The initial effort, before operation of an in -situ treatment process commences,
is the definitive establishment of the vertical and horizontal extents of
contamination. Presently, the downward limit of product penetration has only
been identified at one location. This sample was from TP1 at the 9-11 inch depth
in the presumed confining clay layer. It produced a <10 ug/g result. Additional
soil samples from the other test pit areas will be taken to identify the maximum
depth of vertical penetration. This sampling can be accomplished much easier
with a bucket auger once the gravel is removed. Delineation of the release's
horizontal extent is shown on Figure 1. These boundaries are based on visual
evidence at the time of the spill and encompass most of the lower tier in the
diked area. This area is more than the actual amount of impacted area, but a
conservative approach is best without sampling the entire perimeter. At the time
of the spill, it was observed that the product never reached the interior toe
of the dike walls.
Lateral movement of the contamination is also not probable due to the following
factors:
the small amounts entrained in the soil and
° groundwater not being present at the contamination depths.
In order for significant lateral movement to take place, groundwater must be
present to transport the contaminants. Contaminants generally migrate vertically
downward until the groundwater surface is reached.
Soil Remediation Plan
As stated earlier, gravel removal will begin immediately to facilitate deeper
soil sampling and confirm that all contamination is in a treatable depth zone.
Storage of the gravel will be at the outer boundaries of the tank dike along the
walls, where it will stay until subsequent replacement. This location is logical
because the gravel remains in the affected dike (out of the treatment area),
where any contaminated runoff is contained, and another containment area is not
created requiring additional cleanup.
The treatment area as delineated on Figure 1 encompasses about 31,000 square feet
or 0.7 acres. It will be divided into the three sections shown for composite
sampling reasons. The three sections are an approximate groupin of similar soil
characteristics. Ach-section will have two composite sample• si3t
sawpli:ng points at two depth zones; one 0-6. <inch, sample and one- 6-12!4 :files.
It must be noted that, because of the gravel removal, these depths are now
anywhere from three to twenty-one inches deep when compared with the depths of
the initial soil samples. Previous depths were referenced to the top of the
gravel surface, which varies in thickness. The two samples from each section
should be sufficient to monitor the remed' tion progress. Monitoring of the
soil will begin after the first tillinio�lr
and then on a quarterly frequency until
a <100 ug/g concentration is achieved each depth zone in each section3 Once
the two sample concentrations from a section are <100 ug/g, future sampling in
that section will cease. Cultivation will be scheduled on a biweekly basis as
weather permits and again will be abandoned for a particular section when its
levels of petroleum -hydrocarbons are below 100 ug/g.
3

Colonial Pipeline Company
Sampling will be accomplished using a 3.5-inch bucket auger, mixing the six
sampling points in a clean plastic bucket and transferring the soil into glass
jars for subsequent analysis. The method for the analysis is going to be
maintained as EPA, ,thod 418.1 {petro3,e-hydro"x4Qas)
A report will be prepared after quarterly soil sampling to monitor the progress
of the remediation. The report will include:
° all analytical results from the soil sampling;
° any proposed changes to the remediation plan, if needed; and
° an overview of the degradation progress.
Complete restoration of the impacted area is estimated to take approximately one
year.
Closure
Colonial believes that this proposed remediation plan will sufficiently return
the impacted area back to its pre-existing condition.
Colonial has tentatively scheduled the gravel removal, soil sampling, and initial
tilling to take place the beginning of April..S
DVP:pI
3-31-89
� ,fsJ `fit
Qv �D�
4

Ila
m
�
1
CD
Q)
x
m
r1O W IN
T I
o
Op
00 0 x oo O CD x '�
On 00
m 30 00 00
C�—,ID
X I II Mco
X -
� � I
(� �D U-j . #
Go
Ull
1
00
00 O O
�I O
I�
II -
' 0"
�I m
O L Ln "' o
.--+ - 601
�+ NO H
O (DX
tL
>< CO a x Lo
c
j O O
i
1 I Q O
I ►n ><
c� X o0
00 0
I r� d
x N
O I m O
I tl') X
lL')
m
00 -d c6 p oo -o
C
N Ln x_y__
OD
3 I O

11.
w
N
0�
Y
N