EPA Superfund Record of Decision: Pester Refinery Co. EPA ID: KSD000829846 OU 02 El Dorado KS


                                    EPA/ROD/R07-98/173
                                    1998
EPA Superfund
     Record of Decision:
     PESTER REFINERY CO.
     EPA ID: KSD000829846
     OU02
     EL DORADO, KS
     09/29/1998

-------
EPA 541-R98-173

                                              RECORD OF DECISION

                                   PESTER REFINERY CO./PESTER BURN POND SITE
                                          GROUND WATER OPERABLE UNIT

                                               El Dorado,  Kansas

                                               DECISION SUMMARY

-------
                                          RECORD  OF DECISION


                          PESTER  REFINERY CO./PESTER BURN POND SITE
                                    GROUND WATER  OPERABLE UNIT


                                           EL  DORADO,  KANSAS


                                               DECLARATION

Statement of Basis and Purpose

This decision document presents the selected remedial action for the Ground Water Operable Unit
(OU2) of the Pester Refinery Co. Site (commonly referred to as the Pester Burn Pond Site).  El
Dorado, Butler County, Kansas. The selected remedy was chosen in accordance with the
Comprehensive Environmental Response.  Compensation and Liability Act  (CERCLA),  as amended
by the Superfund Amendments and Reauthorization Act (SARA)  and the National Oil  and Hazardous
Substances Pollution Contingency Plan (NCP).  This decision document is based on  the information
contained in the administrative record for this Site.  The State of Kansas concurs with the
selected remedy.

Assessment of the Site

Based on the conclusions of the OU2 Risk Assessment (RA)  and the OU2 Remedial Investigation
and Feasibility Study (RI/FS), the identification of response objectives, and the identification
and evaluation of potential response actions, the U.S.  Environmental Protection  Agency (EPA)  and
the Kansas Department of Health and Environment (KDHE)  have determined that no additional action
other than ground water monitoring is necessary for the Ground Water Operable Unit of the Pester
Refinery Co. Site to protect human health and the environment.

Description of the Selected Remedy

The EPA and KDHE have selected "No Action" as the remedy for the Ground Water Operable Unit
of the Pester Refinery Co./Pester Burn Pond Site.  The selected "No Action" remedy for OU2
involves no additional measures to eliminate, reduce,  or control threats to human health and the
environment other than mitigative measures implemented as components of the Soil and Sludge
Operable Unit (OU1)  Record of Decision.  Those measures include institutional controls in the
form of a deed restriction controlling development of the property and a fence to restrict site
access, the treatment and/or removal of the source material (oily sludge in the  ponds on site),
and the operation of the underground interceptor trench.  The underground interceptor trench was
constructed to prevent separate-phase hydrocarbon and dissolved-phase seeps from the ponds at
the Site from reaching the river. KDHE and EPA anticipate that the process of extracting,
treating, and returning the seep water to the ponds for recirculation will also  reduce the
concentrations of contaminants in ground water through oil-water separation, physical
filtration, biodegradation, and other natural processes that will act to reduce  contaminant
concentrations.

Actions associated with the selected "No Action" remedial alternative will consist of ground
water monitoring and sediment sampling.  Ground water monitoring of the alluvial, upper bedrock,
and lower bedrock aguifers will be conducted to verify that the concentrations of contaminants
in the alluvial aguifer and the underlying, less-contaminated upper bedrock aguifer continue to
decline and that the concentrations of the chemicals of concern in the underlying deep bedrock
aguifer do not increase. Sediment sampling in the West Branch Walnut River will  be conducted to
augment the limited information on potential impacts historical releases from the Site may have
had on the river.  Ground water monitoring and sediment sampling will be conducted on a
guarterly basis to provide a clearer assessment of baseline conditions, account  for seasonal
variations in ground water level and contaminant concentrations, and permit an assessment of the
effectiveness of the OU1 remedy in restoring the ground water conditions at the  Site and
protecting the West Branch Walnut River.  After two years,  the freguency of monitoring and
sediment sampling will be re-evaluated by KDHE and EPA.

-------
Statutory Determinations

As required by CERCLA for sites where any hazardous substances, pollutants, or contaminants
remain on site, reviews of the remedy must be conducted by KDHE and EPA at least every five
years to verify that human health and the environment are being protected by the remedy. For
sites with multiple operable units, one five-year review would be conducted for the combined
operable units. The first five-year review for OU1 and OU2 of the Site will be conducted no
later than five years from the initiation of the remedial action for OU1 and an additional
five-vear review will be conducted at least every five years thereafter. The five-year reviews
will continue as long as hazardous substances remain on site in the alluvial and upper bedrock
aguifers.

The EPA and KDHE have determined that no further action is necessary to protect human health and
the environment from exposure to the ground water at the Site. In addition, the fact that the
alluvial aguifer is not a usable drinking water source, the expectation that contamination
levels at the Site will continue to decline as a result of the OU1 remediation activities, and
the belief that the operation of the seepage interceptor trench as a component of the OU1 remedy
is effectively preventing the off-site migration of contaminated ground water, and that
institutional controls will prevent any future residential use or well drilling at the Site
contribute to KDHE's and EPA's determination that no additional measures are necessary at the
Site at this time.

Based on the conclusions of the OU2 RA and the OU2 RI/FS, the identification of response
objectives, and the identification and evaluation of potential response actions, KDHE and EPA
have determined that no additional action other than ground water monitoring and sediment
sampling is necessary for the Ground Water Operable Unit of the Pester Refinery Co. Site to
protect human health and the environment.

-------
                                              RECORD OF DECISION

                                   PESTER REFINERY CO./PESTER BURN POND SITE
                                          GROUND WATER OPERABLE UNIT
                                               El  Dorado,  Kansas

                                               DECISION SUMMARY
                                               TABIiE OF CONTENTS

TABLE OF CONTENTS 	i

1.0   SITE NAME, LOCATION, AND PHYSICAL SETTING 	1
      1.1  Site Name 	1
      1. 2  Site Location  	1
      1.3  Physical Setting 	1

2.0   SITE HISTORY AND ENFORCEMENT ACTIVITIES  	2

3.0   HIGHLIGHTS OF COMMUNITY INVOLVEMENT 	4

4 . 0   SCOPE AND ROLE OF THE GROUND WATER OPERABLE UNIT  	4

5.0   SITE CHARACTERISTICS AND INVESTIGATIVE FINDINGS  	5
      5.1  Original Site Characterization 	5
      5.2  OU1 Remedial Investigation 	6
      5 . 3  OU1 Feasibility Study	7
      5.4  OU2 Focused/Abbreviated Remedial Investigation  	8

6. 0   SUMMARY OF SITE RISKS 	11
      6.1  The Risk Assessment 	11
      6.2  Contaminant Transport 	13
      6.3  Ecological Risks 	15
      6.4  Remediation Goals 	15

7. 0   DESCRIPTION OF THE SELECTED "NO ACTION" REMEDY 	18

8.0   SIGNIFICANT CHANGES FROM THE PROPOSED PLAN  	19

GLOSSARY

Figure 1 - Site Location Map
Figure 2 - Site and Vicinity Map
Figure 3 - Geologic Cross Section
Figure 4 - Site Features Map
Figure 5 - Alluvial Aguifer Potentiometric Surface Map, 6/17/94
Figure 6 - Conceptual Cross-Section of Ground Water Contamination Distribution

Table 1  - Summary of Compounds Detected in Ground Water
           Maximum Concentrations - June 1994 Sampling Event
Table 2  - Chemicals of Potential Concern

-------
1.0 SITE NAME, LOCATION, AND PHYSICAL SETTING

1.1 SITE NAME

Althouuh the Site is called the "Pester Refinery Co." Site on the EPA Comprehensive
Environmental Response, Compensation and Liability Information System (CERCLIS) data system, the
National Priorities List (NPL) Site as originally described in site ranking documents focused
somewhat more narrowly on a group of waste (burn) ponds owned by the bankrupt Pester Refining
Co., and not the refinery as a whole, which is operated by Coastal Refining and Marketing Co.
Accordingly, this OU2 Record of Decision addresses these ponds and their environs rather than
the entire refinery. The site has CERCLIS ID No. KSD00829846. This is a state of Kansas-lead
site with PRP cleanup ongoing as Operable Unit 1 of the Site.

1.2 SITE LOCATION

The Site is located on a 10-acre tract located in El Dorado, Kansas, approximately 30
miles northeast of Wichita, Kansas (see Figure 1). The Site is located in the southwest guarter
of Section 26, Township 25 South, Range 5 East, Butler County, Kansas.

Industrial and agricultural lands surround the Site (see Figure 2). Wastewater treatment
and aeration ponds owned and operated by Coastal Refining and Marketing, Inc. (Coastal) are
located south of the Site. West of the Site is an active refinery owned and operated by Coastal
and a Santa Fe Railroad spur that services the refinery. The West Branch Walnut River flows
along, the north and east edge of the area to be remediated. Agricultural land lies east of the
Site across the river. As noted above, the refinery and wastewater treatment and aeration ponds
owned by Coastal are outside the focus of this remedy, which addressed the former Pester Burn
pond and connected ponds and their environs.

1.3 PHYSICAL SETTING

The Site lies within the Osage Plains section (Flint Hills Upland subsection) of the Central
Lowland Physiographic province. In general, the topography is characterized by flat-topped,
steep-sided hills capped by chert-bearing limestone. The Site is underlain by terrace and
alluvial sediments of Pleistocene-Recent age deposited by the West Branch Walnut River and
Permian age units of the Barneston Limestone Formation. There are three aguifers beneath the
Site: 1) an alluvial aguifer ranging in thickness from 2 to 17 feet and consisting of clayey
silts and fine sands with local gravel beds: 2) an upper limestone bedrock aguifer (Fort Riley
Limestone Member of the Barneston Limestone) consisting of thin to massively bedded
fossiliferous limestone and clayey shale; and 3) a lower limestone bedrock aguifer (Florence
Limestone Member of the Barneston Limestone) consisting of fossiliferous limestone with Member)
separating the upper and lower bedrock aguifers (see Figure 3).

Ground water in the region is drawn primarily from the shallow bedrock aguifers (Fort Riley and
Florence limestones), with lesser amounts from the shallow alluvial aguifer because that aguifer
is less commonly present. The bedrock aguifers are characterized by jointed and fractured
limestone that may be confined.

The direction of ground water flow in the alluvial aguifer at the Site is generally northeast
and east toward the river with the possibility of some radial flow to the south and southwest of
the burn pond. Data suggest that the alluvial aguifer and the Fort Riley Limestone aguifer are
hydraulically connected and are locally recharged by the ponds on the Site. Ground water within
the Fort Riley Limestone aguifer is interpreted to flow northeast and east from the ponds with
partial discharge into the West Branch Walnut River. Ground water in the Florence Limestone
aguifer is interpreted to flow eastward from the Site.

2.0 SITE HISTORY AND ENFORCEMENT ACTIVITIES

The refinery occupying the area immediately west of the Site was constructed in 1917, soon after
the discovery of oil at El Dorado in 1915. The refinery and surrounding area were purchased by
FINA in 1958. The burn pond was built by FINA around the time of the purchase. FINA disposed
of petroleum waste products generated by normal refinery operations by running a pipe from the
refinery to the burn pond. The pond was used to store various refinery byproducts such as slop

-------
oil emulsion solids. API separator sludge, and heat exchanger bundle cleaning sludge. When the
waste products were of a very gaseous nature. FINA would ignite the waste product as it came out
of the pipe. Whatever did not burn was discharged out of the pipe into the pond.

The Site historically contained a burn pond, a stormwater pond, and a smaller settling pond (see
Figure 4). The dike separating the burn pond and the larger stormwater pond was breached,
resulting in a U-shaped pond. Eventually the dike between the stormwater pond and the settling
pond also was breached, creating common water between all three ponds. Except for the historical
references, all references to the "burn pond" in this document refer to all three interconnected
ponds at the Site.

On January 1, 1977. Pester purchased the refinery from FINA and continued refinery operations.
Pester filed for bankruptcy on February 25, 1985. Subseguent to Fester's bankruptcy, Coastal
Derby Refining Company (now Coastal Refining and Marketing, Inc.) purchased the refinery with
the exception of the tract of land containing the burn pond. The tract occupied by the burn pond
is still owned by Pester.

On February 28, 1986, KDHE Administrative Order =86-E-16 was issued reguiring Pester to
conduct a site investigation of this surface impoundment, perform monitoring, and submit a Burn
Pond Closure Plan.

The Pester Refinery Co. Site was placed on the N P L on January 31, 1989, by the EPA pursuant
to its authority under CERCLA as amended by SARA. Following initial investigations, a Consent
Order was signed between Pester, FINA, and KDHE (April 19, 1990) to conduct the remedial
investigation/feasibility study (RI/FS) activities at the Site.

During late March 1992, a subsurface interceptor trench was constructed on the north and east
sides of the burn pond between the pond and the West Branch Walnut River to prevent the seepage
of contamination from the burn pond into the river in those areas. This trench extended east
and south of the existing open interceptor trench. The subsurface interceptor trench was dug
into weathered bedrock and sloped to a central collection point. Appreciable thicknesses of oil
that accumulate at the central collection point are periodically skimmed off of the water in the
trench and disposed. Water extracted from the subsurface trench system is discharged back to
the burn pond.

A Record of Decision (ROD) was issued for the Site in September 1992. The ROD split the Site
into two operable units, a soil and sludge operable unit (OU1) and a ground water operable unit
(OU2), because ground water contamination at the Site had not been adeguately characterized at
that time. The main remedial elements specified in the ROD for OU1 included: excavation of
sludge from the three interconnected ponds; separation of the sludge into Recovered Refinery
Feedstock (RRF oil), water, and residual solids; treatment, transportation, and off-site
disposal of the residual solids; transportation of the RRF oil for incorporation into the
refining process; and bioremediation of the contaminated soils in the ponds.

The KDHE, FINA, and Pester entered into a Consent Order to complete a Remedial Design/Remedial
Action (RD/RA) for OU1 of the Site in September 1993. The treatability study for the soil was
completed in the fall of 1994. The pond sludge dredging and removal and recycling of oil
contained in the sludge began in December 1995 and continued through March 1996. The
bioremediation design document for the soil portion of OU1 establishes the organization and
technical basis for the bioremediation of the pond soils. Bioremediation of the pond soils is
being conducted in a phased approach: Phase I has been completed and focused on remediating half
of the stormwater pond while simultaneously gathering bioremediation performance data: Phase II
is underway and is focused on addressing the remainder of the stained soil in the ponds.

In December 1993, FINA and KDHE entered into a Consent Order to conduct RI/FS activities for
OU2.

-------
In August 1996, FINA requested that KDHE permit the construction of an extension of the
northwestern end of the subsurface interceptor trench at the Site to replace the existing open
interceptor trench.  The open trench is subject to flooding by the West Branch Walnut River and
has overflowed on several occasions, releasing wastes into the river. KDHE approved the interim
measure and the trench extension has been completed.

                         3.0      HIGHLIGHTS  OF  COMMUNITY INVOLVEMENT

A draft Community Relations Plan was prepared to the Site in June 1990 by the responsible
parties and was implemented by KDHE. A public meeting was held after community interviews in
October 1990, prior to the start of the OU1 RI/FS activities. The OU1 RI and FS reports were
released to the public in 1991 and 1992, respectively.  On August 27, 1992, a public meeting was
held in El Dorado, Kansas, to present the Proposed Plan for OU1.  Notice for the public meeting
was published in the El Dorado Times.

On September 2, 1998, a public meeting was held at the Bradford Memorial Library in El Dorado.
Kansas, to present the Proposed Plan for OU2 and a concurrently offered Explanation of
Significant Differences for OU1.  Notice for a 30-day public comment period and the public
meeting was published in the El Dorado Times and was provided to a number of local and regional
public and private entities that might have interest in the Site. Information repositories
containing the Administrative Record for the Site are available at the Bradford Memorial Library
in El Dorado, at the KDHE office at Forbes Field in Topeka, and at the EPA Region VII office in
Kansas City, Kansas.

                4.0      SCOPE  AND  ROLE  OF  THE  GROUND  WATER  OPERABLE UNIT

The September 1992 ROD for the Site split the Site into two operable units, a soil and sludge
operable unit  (OU1) and a ground water operable unit (OU2), because ground water contamination
at the Site had not been adequately characterized at that time. Certain aspects of the OU1
remedy have contributed to the mitigation of ground water contamination and have diminished the
potential risk posed by that contamination. Exposure controls implemented as elements of the OU1
remedy include institutional controls in the form of a deed restriction controlling development
of the property and a fence to restrict site access, the removal of the material that served as
the original source of ground water contamination  (oily sludge in the ponds),  and the operation
of the underground interceptor trench.

With the dredging of the oily sludge from the ponds, the source for ground water contamination
has likely been removed. The levels of contamination can reasonably be expected to remain
constant or decline following the removal of the source. With the presence of an active
bioremediation system in the ponds and the recirculation of biologically active ground water
through the pond,  alluvial aquifer, interceptor trench, and treatment train, it is anticipated
that treatment and biodegradation and other natural attenuation processes will continue to
reduce the concentration of chemicals of concern in alluvial ground water  (Table 2).

The shallow alluvial aquifer is the most significantly impacted water-bearing zone at the Site,
but is not employed as a drinking water source in the area and does not yield sufficient water
to serve as a domestic water supply. Aquifer testing conducted as part of the OU2 RI  indicated a
hydraulic conductivity for the combined alluvial and upper bedrock aquifers of 3.7 X 10 -4 cm/s.
Given the maximum saturated thickness of approximately 20 feet for the combined saturated zones
of these aquifers that are in communication with each other, the projected maximum sustained
yield of the aquifers would be insufficient to serve as a permitted domestic water supply in
Kansas.  In addition,  the total dissolved solids content of water from the thin veneer of
saturated alluvium at the upgradient edge of the Site exceeds the Secondary Maximum Contaminant
Limit for that parameter, suggesting that the water would be non-potable as a result of taste,
odor, color, or other non-aesthetic effects. The underlying upper bedrock aquifer while
significantly less impacted, contains elevated concentrations of volatile organic compounds
(VOCs), arsenic, and barium.  Both the alluvial aquifer and the upper bedrock aquifer normally
discharge to the West Branch Walnut River but are currently being captured by the underground
interceptor trench. No chemicals of concern for the Site have been detected at significant
concentrations in the lower bedrock aquifer, a potential drinking water aquifer. The presence of
the shale zone within the upper bedrock aquifer appears to act as an aquitard, isolating the
lower bedrock aquifer to some extent from the downward migration of contaminants present in the

-------
overlying aquifers.

An open interceptor trench was installed in the late 1950s or early 1960s to intercept seepage
from the burn pond to the West Branch Walnut River. The trench was excavated to the top of
weathered bedrock and sloped to the east where fluids were collected and pumped back up to the
ponds on site. Although typically effective, the trench occasionally overflowed and carried
contaminants into the river. During late March 1992, the original subsurface interceptor trench
was constructed on the north and east sides of the burn pond between the pond and the West
Branch Walnut River to prevent the seepage of contamination from the burn pond into the river in
those areas. This trench extended east and south of the existing open interceptor trench. The
subsurface interceptor trench was dug into weathered bedrock and sloped to a central collection
point. Appreciable thicknesses of oil that accumulate at the central collection point are
periodically skimmed off of the water in the trench and disposed. Water extracted from the
subsurface trench system at the collection point is treated by oil-water separation and
mechanical filtration and is discharged back to the burn pond.

In August 1996 Fina requested that KDHE permit the construction of an extension of the
northwestern end of the subsurface interceptor trench at the Site to replace the existing open
interceptor trench. KDHE approved the interim measure and the trench extension has been
completed. KDHE and EPA anticipate that the process of extracting, treating, and returning the
seep water to the ponds for recirculation will also reduce the concentrations of contaminants in
ground water through oil-water separation, physical filtration, biodegradation, and other
natural attenuation processes.

5.0 SITE CHARACTERISTICS AND INVESTIGATIVE FINDINGS

5.1 ORIGINAL SITE CHARACTERIZATION

Pursuant to the 1986 KDHE Administrative Order, Pester retained Mid West Environmental
Consultants (MWEC) to investigate the Site and provide site characterization of the pond.

ground water, and soil. The MWEC investigation covered the entire refinery. MWEC identified
constituents in the sludge, soil, ground water, and surface water. These constituents included
lead, chromium, phenols, and oil and grease in the sludge and the soil under the pond.
Separate-phase hydrocarbons and dissolved-phase benzene, toluene, xylenes, ethylbenzene, and oil
and grease were identified in the ground water, and dissolved-phase phenols, benzene, toluene,
ethylbenzene, and oil and grease were identified in the pond water. KDHE has allowed the pond to
be periodically de-watered and the effluent treated at the Coastal wastewater treatment plant
contingent upon meeting National Pollutant Discharge Elimination System (NPDES) permit limits.
Chromium and lead were not detected in any surface water samples. Chloride and selenium were
detected in a spring adjacent to the river. Surface water sampling in the West Branch Walnut
River indicated the presence of oil and grease, phenols, sulfide, chloride, arsenic, barium, and
selenium at concentrations below the water quality standards for the protection of human health
(i.e., drinking water Maximum Contaminant Levels [MCLs] and lifetime Health Advisory
concentrations). Sediment sampling in the river indicated that concentrations of oil and grease
increased along the reach of the river adjacent to the Site and decreased immediately
downstream. Phenols and sulfide increased along the length of the Site in river sediments.
Concentrations of arsenic, lead, and selenium increased along the length of the Site and
decreased downstream. Concentrations of barium increased downstream. The results of the site
investigation were to be used by Sunbelt Environmental Management (Sunbelt) to develop a closure
plan for the burn pond. In 1989, Groundwater Technology, Inc. (GTI) was retained by FINA to
summarize previous data collected by MWEC, Sunbelt, and KDHE.

5.2 OU1 REMEDIAL INVESTIGATION

FINA retained Metcalf and Eddy (M&E) to perform the initial RI/FS of the Site under the April
19, 1990. Consent Order between Fina, Pester, and KDHE. The RI consisted of five primary
components: 1) review of existing data; 2) installation of five additional monitoring wells to
complement the seven existing monitoring wells; 3) verification of the volume and chemical and
physical nature of the sludge; 4) dye-tracer testing of the storm sewers; and 5) completion of
the sludge, soil, ground water, and surface water sampling program.

-------
Sludge samples were collected at three sampling locations within the burn pond to characterize
for VOCs and metals.

Contaminants identified in the sludge samples included xylenes, 2-methylnaphthalene,
benzo(a)anthracene, chrysene, phenanthrene, pyrene, arsenic, barium, chromium, and lead.

Soil samples were collected at four sampling locations beneath the sludge in the ponds. Organic
contaminants detected in the soil samples included ethylbenzene, toluene, xylenes,
2-methylnaphthalene, benzo(a)anthracene, chrysene, naphthalene, phenanthrene, phenol, and
pyrene. Barium was the only metal detected in the Toxicity Characteristic Leachate Procedure
(TCLP) metals analyses of soil samples; it was detected at a non-hazardous concentration.

Analytical results for samples collected from wells screened in the alluvial aguifer around the
Site indicated that contamination is present in the alluvial aguifer. Elevated concentrations of
VOCs and metals were detected in ground water in the alluvial aguifer to the northeast, east,
and south of the ponds. Samples from a monitoring well screened in the upper bedrock aguifer
(the Fort Riley Limestone) also contained elevated concentrations of VOCs and slightly elevated
levels of metals relative to background.  No contaminants were detected in the lower bedrock
aguifer (the Florence Limestone).

Samples collected from a spring emanating from the west bank of the river northeast of the ponds
contained VOCs and SVOCs at concentrations slightly above analytical detection limits. A ground
water seep located northeast of the Site at the southwest edge of the river showed elevated
concentrations of SVOCs and metals. Many of the contaminants detected in the seep are detected
in the seep.

5.3 OU1 FEASIBILITY STUDY

The original FS evaluated five general response actions which could be applied to the
contaminated media and conditions known to exist at the Site. The general response categories
included:  (1) no action;  (2) reuse; (3) capping;  (4) stabilization; and  (5) thermal treatment.
The FS identified and screened remedial action technologies associated with each general
response action previously identified. The screening criteria used for the analysis included
effectiveness, implementability,  and cost of each remedial action technology. Those remedial
action technologies failing to meet the pre-defined criteria were screened out of the process.

Remedial action technologies were screened for applicability to the specific affected media
types  (sludge and soil). The remedial action technologies evaluated for sludge included: (1) no
action; (2)  reuse in asphalt or refinery;  (3) capping; (4) stabilization on or off site; (5)
thermal treatment on or off site;  (6)  bioremediation; and (7) removal/landfill. The remedial
action technologies evaluated for soil included:  (1) no action; (2) capping;  (3) stabilization
on or off site; (4) thermal treatment on or off site; (5) bioremediation;  (6) removal/landfill;
and (7) in situ ("in place") soil flushing.

Several remedial alternatives combining the technologies and process options that passed initial
screening were evaluated in the FS report. The EPA ROD split the Site into two operable units
and selected removal and reuse of sludge at a refinery followed by in situ flushing/
bioremediation of the remaining soils as the remedy for the sludge and soil operable unit.   The
existing interceptor trench was incorporated into the treatment of the soil; seepage from the
ponds that is captured in the interceptor trench system is currently being recirculated into the
ponds.

It was determined subseguent to the execution of the ROD that the remedy described therein was
not implementable because there were no available refineries which held a permit to recycling,
making the transportation of sludge materials off site infeasible. An Explanation of Significant
Differences  (BSD)  was prepared to document changes to the ROD and the reasons those changes were
made.   The remedy was modified by the BSD to incorporate an alternative method of treating the
sludge material. The modified remedy included three-phase separation of the pond sludge on site.
The water phase was to be sent to the Coastal Derby wastewater treatment plant which operates
under a NPDES permit.  The oily phase was to be taken off site to be recycled. The residual
solids, or "filter cake", were to be further treated on site to meet Best Demonstrated Available
Technology standards to meet the land ban reguirement for land disposal. The treated filter cake
was then to be disposed at a RCRA-permitted Treatment, Storage, and Disposal facility in

-------
compliance with EPA' s off-site policy.

5.4 OU2 FOCUSED/ABBREVIATED REMEDIAL INVESTIGATION

The Focused/Abbreviated RI for OU2 of the Site (OU2 RI) was directed toward augmenting
information on the nature and distribution of groundwater and groundwater contamination at the
Site collected during the original Site-wide RI. Fina retained Sharp and Associates, Inc.
(Sharp) to perform the OU2 RI/FS activities. Sharp identified the following goals for the OU2
RI:

• to gain further understanding of groundwater flow at the Site:
• to assess interaction between aquifers; and
• to define the extent of groundwater contamination.

The field work for completion of the OU2 RI was conducted in June 1994.

The following three aquifer units were defined for OU2:

alluvial:
upper bedrock (Fort Riley Limestone); and
lower bedrock (Florence Limestone).

Ground water flow in the alluvial and weathered zone of the upper bedrock aquifers was generally
to the northeast toward the West Branch Walnut River. The lower portion of the upper bedrock
aquifer also flowed from the Site to the east in the direction of bedrock dip. A well on the
east side of the river indicated that the ground water flow direction in the upper bedrock
aquifer in that area is also toward the river.

A hydrologic connection between the alluvial aquifer and the upper portion of the upper bedrock
aquifer was inferred due to similar potentiometric surfaces, i.e., ground water in the two
aquifers is mingled. The alluvial aquifer and the upper portion of the upper bedrock aquifer
flow toward the river which acts as a hydrologic sink or a gaining stream. The lower bedrock
aquifer exhibited artesian (confined) conditions due to the presence of an aquiclude (Oketo
Shale) between the upper and lower bedrock aquifers.

Contamination was generally confined to the alluvial and upper section of the upper bedrock
aquifer. The lower bedrock aquifer was free of contaminants in excess of MCLs. Further
discussion of the occurrence of ground water contaminants in the aquifers is included in Section
6.1 below.

Figure 4 illustrates the burn pond, surrounding active wastewater ponds and the wells in the
area. There are three potential sources for ground water contamination in the alluvial aquifer
on and off site: the burn pond, the existing refinery located upgradient of the Site, and the
active refinery wastewater treatment ponds south of the Site.

To resolve the potential upgradient contaminant source at the existing refinery, well W38 was
drilled during the OU2 RI. Well W38 is hydraulically upgradient of wells W23, W24, and the
piezometers nearest to the ponds (P3, P4, and P5). Well W38 contained substantially higher
concentrations of dissolved-phase BTEX contamination than wells W23 and W24. The OU2 RI
concluded that the contamination in W38 is related to a source upgradient of the Site. KDHE is
in the process of evaluating this area with the current refinery operator.

Another area of potential contamination exists south of the burn pond in the vicinity of wells
W3 and W22. Additional piezometers (P6, P7, and P8) were installed to resolve the ground water
flow direction in the alluvial aquifer. Based on the June 1994 alluvial potentiometric map, the
ground water in the vicinity of the final refinery wastewater treatment pond and the south end
of the burn pond flows toward the river (see Figure 5). The active wastewater treatment ponds at
the Coastal refinery also contain oil being treated through an aeration/biodegradation process.
The conclusion drawn in the OU2 RI from these data is that the wastewater ponds may be potential
contributors to ground water contamination in the vicinity of well W22. The separation of
dissolved phase contamination plumes in the alluvial aquifer at the southern boundary of the
Site is based on ground water flow potential rather than chemical composition because there was
no distinct chemical fingerprint for either potential source to distinguish between the two.

-------
A summary of the major contaminants and maximum concentrations in the June 1994 samples by
aquifer is shown on Table 1. Figure 6 illustrates a conceptual cross-section of contaminant
distribution in OU2.

Contamination in the alluvial aquifer, which is not considered a drinkinq water aquifer based on
potential yield and inorqanic quality, extends from the upqradient edqe of the Site property to
the river to the east and north and to the southern property boundary to the south based on flow
direction. The existinq subsurface interceptor trench which was constructed in connection with
the OU1 remedy serves as a barrier to qround water contaminant miqration to the river. The
alluvial aquifer terminates at the river. The hiqhest concentrations of VOCs were present in the
alluvial aquifer. VOCs identified in the alluvial aquifer durinq the OU2 RI included benzene,
ethylbenzene, toluene, and total xylenes. Benzene is present at concentrations in excess of the
MCL. SVOCs identified in the alluvial aquifer included low concentrations of polycyclic aromatic
hydrocarbon compounds, phenols, phthalates, naphthalene, and methyl naphthalene. Arsenic,
barium, chromium, and lead were present above backqround concentrations in the alluvial aquifer.
Separate-phase hydrocarbons were present in several wells durinq the OU2 RI samplinq.

The upper bedrock aquifer (Fort Riley Limestone) is locally recharqed by the ponds on the Site.
Ground water is interpreted as flowinq north and east from the ponds and discharqinq to the West
Branch Walnut River. Trace concentrations of benzene, toluene, and total xylenes were detected
in a well completed in the upper bedrock aquifer upqradient of the Site in 1990. No VOCs were
detected in the sample collected from the upqradient well durinq the OU2 RI except acetone which
was tentatively attributed to laboratory contamination. A sample collected in 1990 from a well
located downqradient of the burn pond and screened in the upper bedrock aquifer contained hiqher
levels of benzene, toluene, and total xylenes relative to the upqradient well and low levels of
arsenic and barium; subsequent samples have not contained elevated concentrations of metals or
VOCs. Low concentrations of several SVOCs were detected in the upper bedrock aquifer durinq the
1994 OU2 RI.

The lower bedrock aquifer (Florence Limestone) is separated from the upper bedrock aquifer by an
aquiclude, the Oketo Shale. A backqround sample collected in the lower bedrock aquifer in 1990
contained trace concentrations of toluene and total xylenes; no VOCs were detected above
detection limits at this location durinq the OU2 RI. Samples collected from a well completed in
the lower bedrock aquifer downqradient of the ponds contained no VOCs or SVOCs above detection
limits. Arsenic and barium were detected at concentrations below the MCLs for those substances
in 1994 and may reflect the naturally-occurrinq backqround concentrations for those metals in
the aquifer.

Surface water samples were collected from the West Branch Walnut River durinq the OU2 RI and
analyzed for VOCs; no VOCs were present above detection limits.

A subsurface interceptor trench exists at the Site to prevent further miqration of contaminated
qround water. The interceptor trench was desiqned to capture and recirculate seeps from the
ponds to maintain the aqueous bioremediation system; the trench also serves to contain the
qround water plume and to prevent further miqration of contaminated qround water from the
hydraulically connected alluvial and upper bedrock aquifers to the river. The number of oily
seeps discharqinq to the river east of the interceptor trench decreased after its installation.
No oily seeps have been noted alonq the river bank adjacent to the Site durinq recent
inspections.

Both a step test and a constant rate aquifer pumpinq test were conducted on the trench in June
1994. The connection between the alluvial aquifer and upper portion of the upper bedrock aquifer
was confirmed as well as the connection to the river. A hydraulic conductivity of 3.7x10 -4
cm/s was determined. Durinq active pumpinq in the trench, the qround water flow direction
between the trench and the river reverses toward the trench.

The effectiveness of the interceptor trench was evaluated. The east side of the trench acts as
a partial barrier to qround water miqration due to the presence of a liner installed into the
bedrock surface. This barrier was found to be approximately 20% effective at preventinq qround
water miqration with the pumps off and 130% effective with the pumps on, meaninq that with the
pumps operatinq the trench not only prevented water from flowinq downqradient but actually drew
some water from the downqradient side of the liner. A pumpinq rate of 12 qallons per minute was
determined to be sufficient to exert hydraulic control over the lenqth of the trench.

-------
6.0 SUMMARY OF SITE RISKS

6.1 THE RISK ASSESSMENT

As part of the RI/FS, a Risk Assessment (RA) for OU2 was prepared by PRC Environmental
Management, Inc. on behalf of KDHE. The OU2 RA was carried out to characterize, in the absence
of remedial action, the current and potential threats to human health and the environment that
may be posed by contaminants released from the ponds and their impacts to ground water and
surface water. The RA provides information to help determine whether remedial action is
necessary for OU2 of the Site.

The OU2 RA is primarily based on sample data collected from the Site during the OU2 RI. Ground
water and surface water data were collected during the RI. Contaminants detected in these media
were used to characterize the nature and extent of the human and ecological risks posed by OU2.

For this risk assessment, PRC evaluated data from three aguifers: alluvial, upper bedrock, and
lower bedrock. The aguifer with the highest number of contaminants and the highest
concentrations was the alluvial aguifer; this aguifer was the focus of the risk assessment.
Chemicals of potential concern found in the Pester Refinery Co. OU2 are VOCs and metals. Ground
water data showed the major contaminants to be benzene, toluene, ethylbenzene, xylenes,
naphthalene, and arsenic. A total of 18 substances were evaluated in the RA as chemicals of
potential concern in ground water (see Table 2).

Surface water samples were collected from the West Branch Walnut River during the OU2 RI;
surface water sampling showed no detectable concentrations of COCs.

Pathways by which humans could be exposed to the chemicals of potential concern in ground water
at the Site were evaluated based on reasonable assumptions about current and future land use.
The exposure assessment identified receptors that potentially may be exposed to contaminants.
Future on-site workers and future residents were evaluated as receptors that might receive
exposure to site contaminants. It should be noted that, consistent with the OU1 ROD for soil and
sludge, deed restrictions have been emplaced on the property that prevent future use of the Site
for residential purposes; conseguently, an industrial land-use scenario was used as the standard
for remedy evaluation for OU2. The Site is presently being used for industrial purposes,
although industrial activities are limited to maintenance of the ponds and the interceptor
trench. The Site property is currently zoned industrial. The Site is also fenced and posted with
warnings to prevent trespass. Future residential exposure scenarios will not be further
considered in this document.

For future worker exposures, the OU2 RA assumed that ground water from the alluvial aguifer
would be used and that workers would be exposed through ingestion, dermal contact, and
inhalation of volatilized contaminants during showering. The RA characterized this assumption as
conservative since City of El Dorado Public Water Supply water is available at the Site.

Available risk factors for carcinogens and noncarcinogens were obtained from the EPA Integrated
Risk Information System (IRIS) and the Superfund Technical Support Center Environmental Criteria
and Assessment Office Risk Assessment Issue Paper for: Provisional Oral Reference Dose for
Naphthalene for the chemicals of potential concern in OU2. The majority of the chemicals of
potential concern have a noncarcinogenic effect on humans; only benzene and arsenic are
carcinogenic.

A Hazard Index (HI) was calculated for each pathway evaluated in the OU2 RA. A HI of less than 1
indicates that the noncarcinogenic risks associated with that pathway are within the acceptable
range for human health risk as determined by EPA. The HI for on-site workers did exceed 1 for
the ground water ingestion exposure pathway as evaluated for exposure to contaminants in the
alluvial aguifers. Arsenic was the most significant contributor. Barium, chromium, and
2-methylnaphthalene were more significant contributors to non-carcinogenic risk than the
remaining chemicals of potential concern.

The carcinogenic risks were also calculated for the evaluated pathways in OU2. The carcinogenic
risk from exposure to a chemical is described in terms of the probability that an individual
receiving chronic exposure to the chemical will develop cancer as a result of that exposure.

-------
Risk estimates are presented as excess cancer risk per unit of population. For example, a risk
estimate of 1 X 10 -4 is equivalent to a risk of one excess occurrence of cancer per 10,0000
exposed individuals in a given population.

The total excess lifetime cancer risk for future on-site workers was estimated for exposure to
ground water from the alluvial aguifer. The highest cancer risk was seen to the future worker
ingesting ground water, 6.7 X 10 -5. The cancer risk for dermal contact by future on-site
workers was 2.7 X 10 -6. That for inhalation during showering by future on-site workers was 3 X
10 -5. The total cancer risk for all exposure pathways involving future on-site workers was 1 X
10 -4. The contaminants contributing most to these risks were arsenic and benzene.

PRC did not calculate potential risk resulting from human exposure to ground water in the
bedrock aguifers. The chemicals found in the upper and lower bedrock aguifers were limited and
at concentrations two orders of magnitude lower than levels found in the alluvial aguifer. This
indicates that risk associated with exposure to the bedrock aguifers would be significantly less
than that posed by contamination in the alluvial aguifer.

EPA uses the general 1 X 10 -4 (one in 10,000) to 1 X 10 -6 (1 in 1,000,000) lifetime excess
cancer risk range as a "target range" within which the EPA strives to manage risks as part of a
Superfund cleanup. Generally, where the risk assessment indicates that a cumulative site risk to
an individual using reasonable maximum exposure assumptions for either current or future land
use exceeds the 10 -4 lifetime excess cancer risk end of the risk range, action under CERCLA is
warranted at the Site. Although the National Contingency Plan (NCP) suggests the 10 -6 level as
a starting point for acceptable risk, a site-specific risk estimate around 1 X 10 -4 may be
considered acceptable if justified based on site-specific conditions. For sites where the
cumulative site risk to an individual based on reasonable maximum exposure for both current and
future land use is less than 1 X 10 -4, action generally is not warranted, but may be warranted
if a chemical-specific standard that defines acceptable risk is violated, noncarcinogenic risk
falls within an unacceptable range, or an adverse environmental impact exists that warrants
action. It should be noted that the calculation of potential risk posed by ground water
contamination at the Site in the OU2 RA should probably be viewed as overly conservative, since
the projected maximum sustained yield of the combined alluvial and upper bedrock water-bearing
zones would be insufficient to serve as a permitted domestic water supply in Kansas, and the
ingestion of, dermal contact with, and inhalation of contamination from those zones by future
workers is conseguently unlikely to occur.

6.2 CONTAMINANT TRANSPORT

The original source of contaminants at the Site was the burn pond sludge, now removed from the
Site, which contained volatile and semi-volatile organic compounds as well as several metals.
The first 12 to 24 inches of soil beneath the ponds were stained as a result of contamination,
as were up to 5 feet of soil in the bottom of the aguifer between the eastern boundary of the
burn pond and the river. The amount of contaminated soil was originally estimated at 70,000
cubic yards.

Based on the findings of the OU2 RI, transport of contaminants from the burn pond sludge through
the vadose zone to ground water has occurred. Ground water flows radially to the north, east,
and south from the burn pond in the alluvial aguifer and discharges into the West Branch Walnut
River. The ground water flow velocity through the alluvial aguifer is estimated to range from
0.2 to 60 meters/year depending on whether water is moving within the silty clay or through
local clayey gravels. The rate of contaminant migration will be highest in the gravel-rich zones
within the alluvial sediments.

Benzene, toluene, xylene, arsenic, and barium were identified in the Fort Riley aguifer in wells
located downgradient of the burn pond area. This condition indicates that contamination has
migrated from the ponds and the alluvial aguifer to the Fort Riley limestone. The Fort Riley
limestone aguifer also discharges to the West Branch Walnut River.

No chemicals of concern for the Site have been detected at significant concentrations in the
Florence Limestone aguifer. A sample collected in 1994 from the Florence Limestone contained
slightly elevated concentrations of arsenic and barium (below their respective MCLs), but those
detections may reflect the naturally-occurring background concentrations for those metals in the
aguifer. The presence of the shale zone within the Fort Riley Limestone appears to act as an

-------
aquitard, isolating the Florence Limestone to some extent from the downward migration of
contaminants present in the overlying aguifers. Trace concentrations of toluene and total
xylenes were detected in a background sample collected from the Florence Limestone downgradient
of the adjacent refinery and upgradient of the Site in 1990, suggesting that there may be some
limited communication between the overlying saturated zones and the Florence Limestone.

Oil and dissoived-phase contamination is accumulating in the subsurface interceptor trenches on
the east, northeast, and north sides of the ponds. The contaminants appear to migrate to the
top of the bedrock surface and then travel via gravity to subsurface topographic depressions.
The bedrock slopes to the east and has a surface exposure on the north side of the Site. The
subsurface interceptor trenches are completed into the bedrock surface. Migration is believed to
be due at least in part to the hydraulic head supplied by the water in the ponds.

Organic chemical contamination was detected in the alluvial sediments in 1986, indicating that
contamination has migrated via surface runoff, aguifer/stream interconnection, or trench
overflow events into the West Branch Walnut River.

6.3 ECOLOGICAL RISKS

The data from the Site were limited for evaluating the potential ecological risks posed by
contamination in OU2. At the time of the RA preparation there were seeps discharging to the West
Branch Walnut River; however, those that are adjacent to the ponds have seen a significant
decrease in flow since the installation of the interceptor trench. No seeps are visible at this
time. No VOC contamination has been detected in surface water during recent sampling events.
There are no current data available for the sediments, although it is anticipated that SVOCs
could accumulate in the sediments. Based on the available data, no impacts to the ecosystem have
been detected.

The environmental receptors in the vicinity of the Site include common wildlife such as rabbits,
snakes, migrating waterfowl, and aguatic life, including fish. Wildlife might be affected by
exposure to contaminated sediments or surface water in the West Branch Walnut River.

An Ecological Assessment (EA) was conducted as a part of the OU1 RI/FS for the purposes of
determining possible effects from contamination to the Site ecological system. The EA states
that there is a possibility that wildlife accessing the burn pond area could be affected, but
that the severity of the effect could not be quantified without further information. Any effect
would be dependent upon wildlife accessing and using the area and ingesting significant
quantities of contaminated material. Contaminated ground water was not indicated as a threat to
animal life in the vicinity of the Site. Three endangered species, the bald eagle and two
species of snake, are mentioned in the EA as located near the Site. Bald eagles are known to
winter in El Dorado State Park one mile from the Site. The two species of snake, Heterodgn
platyrhinos and H. hasicus, have verified locations from counties on three sides of Butler
County (Harvey, Cowley and Greenwood counties).

The EA determined that the principal on-site ecological threat was posed by the sludge contained
in the burn pond. The sludge has been treated and removed from the Site and stained soils within
the burn pond are being addressed through bioremediation consistent with the OU1 ROD.

6.4 REMEDIATION GOALS

Remediation goals for OU2 have been defined on the basis of the OU2 RI/FS findings in
conjunction with the conclusions of the OU2 RA. The OU2 RA established that, for the Ground
Water Operable Unit, the total excess lifetime cancer risk posed by all exposure pathways
involving future on-site workers was 1 X 10 -4. The cumulative HI for all non-carcinogenic
chemicals of concern was determined during the OU2 RA to exceed the threshold criteria of 1.0.
These risks are based on exposure to contamination identified in the alluvial aquifer. The
concentrations of contaminants in the upper bedrock aquifer (the Fort Riley Limestone) were
lower than those in the alluvial aquifer upon which the findings of the OU2 RA were based. No
significant contamination has been detected in the lower bedrock aquifer (the Florence
Limestone). The risk evaluated for OU2 during the OU2 RA have been effectively elimimited by
existing institutional controls that were implemented as aspects of the OU1 remedy i.e., the
deed restriction preventing development of the property for residential purposes and the
presence of a maintained fence to control access to the Site. Furthermore, the alluvial aquifer

-------
that was evaluated in the OU2 RA is not used as drinking water sources in the vicinity of the
Site, and the OU2 RI/FS indicates that the combined alluvial and upper bedrock aquifers would
not yield sufficient water at the Site to serve as a domestic water supply source, and may be
non-potable due to background inorganic quality issues. While a hydrologic connection has been
determined to exist between the contaminated alluvial aquifer and the less-contaminated upper
bedrock aquifer, the lower bedrock aquifer,  which may serve as a drinking water source in the
region, is separated from the upper bedrock aquifer by a competent shale confining layer and
exhibits artesian properties on site; in other words, ground water in the upper two aquifers can
mingle, but ground water in the lower bedrock aquifer appears to be geologically isolated from
the upper aquifers. Finally, while no supporting analytical evidence is available at this time,
the operation of the underground interceptor as an aspect of the OU1 remedy is probably
reducing the concentrations of ground water contaminants in the alluvial aquifer through the
removal of oil and the filtration of particulates; ground water monitoring included as a
component of the selected remedy for OU2 will confirm this contaminant reduction. KDHE and EPA
have therefore concluded that ground water contamination in the alluvial aquifer does not pose
unacceptable risk to human health.

Unacceptable potential risk could be posed if contamination from the alluvial and upper bedrock
aquifers were to impact the lower bedrock aquifer. As a drinking water aquifer, the relevant and
appropriate water quality standards for the lower bedrock aquifer are MCLs.  Primary remediation
goals for OU2 are therefore to monitor the lower bedrock aquifer to make sure that contaminant
levels do not trend upward, so that it does not become contaminated in excess of MCLs for the
chemicals of concern in OU2; and to prevent future human exposure to contaminated ground water
within the upper bedrock and alluvial aquifers. Ideally this latter goal would be accomplished
by reducing the concentrations of contaminants in the overlying aquifers to concentrations that
no longer pose a significant threat to human health or water quality in the lower bedrock
aquifer. While KDHE and EPA have determined that the potential risk posed by ground water
contamination in the upper saturated zones at the Site does not warrant a ground water remedy,
operation of the interceptor trench and the treatment of the trench effluent that is being
performed as an aspect of the OU1 flushing/bioremediation remedy will probably continue to
reduce the concentrations of chemicals of concern in those saturated zones.

Significant Remedial Action Objectives  (RAOs) identified in the OU2 FS for OU2 included: 1)
reducing contamination in environmental media with completed exposure pathways such that the
cumulative HI is less than 1 and the cumulative excess lifetime cancer risk is less than 1 X
10 -4; 2) protecting the Florence Limestone bedrock drinking water aquifer from becoming a
contaminated in excess of MCLs; 3) preventing separate-phase hydrocarbons from entering the West
Branch Walnut River; and 4) preventing water with levels of contamination above the applicable
water quantity standards from discharging into the West Branch Walnut River. As previously
indicated, no completed exposure pathways currently exist for contaminated ground water at the
Site, and the existing institutional controls effectively eliminate the potential risk posed by
exposure to contaminated ground water. The operation of the underaround interceptor trench as
part of the OU1 remedy prevents the discharge of separate-phase hydrocarbons or significant
concentrations of other contaminants to the river.

The primary remediation goals that remain are therefore to prevent degradation of water quality
in the lower bedrock aquifer by the chemicals of potential concern, prevent the degradation of
surface water quality in excess of applicable surface water quality standards, and prevent the
discharge of visible films resulting from seeps of separate-phase hydrocarbons into the river.
An additional remedial goal, to prevent human ingestion of ground water from the alluvial
aquifer that exceeds acceptable risk levels, has been determined by KDHE and EPA to not apply to
the Site; however, this goal will be met through the device of institutional controls (i.e., the
existing deed restriction preventing the use of the Site property for other than industrial uses
and the existing site access controls) and through the ongoing operation of the subsurface
interceptor trench as an aspect of the OU1 ROD. In addition, though not required for protection
of human health and the environment. KDHE has requested that the property owner amend the
existing deed restriction to include a restriction preventing the installation of water wells
for purposes other than remediation as an additional conservative measure. This restriction will
run with the property and could only be removed with KDHE's consent.

In summary, because the risk from ground water falls within EPA's acceptable range, KDHE and EPA
believe no further action is necessary to protect human health and the environment from exposure
to the ground water at the Site. In addition, the fact that the alluvial aquifer is not a usable

-------
drinking water source, the expectation that contamination levels at the Site will probably
continue to reduce as a result of the OU1 remediation activities, and the belief that the
operation of the seepage interceptor trench as a component of the OU1 remedy is effectively
preventing any off-site migration of contaminated water, and that institutional controls will
prevent any future well drilling at the Site contribute to KDHE's and EPA's belief that no
additional measures are necessary at the Site at this time.

Based on the conclusions of the OU2 RA and the OU2 RI/FS, the identification of response
objectives, and the identification and evaluation of potential response actions, KDHE and EPA
have determined that no additional action other than ground water monitoring and sediment
sampling is necessary for the Ground Water Operable Unit of the Pester Refinery Co. Site to
protect human health and the environment. If future monitoring of the Florence  Limestone
aguifer were to indicate a consistent trend toward increases in the concentrations of chemicals
of concern, an active remedy will be warranted. Similarly, if sediment sampling in the West
Branch Walnut River were to indicate significant impacts related to historical Site activities,
additional assessment or remedial action will be warranted.

                 7.0      DESCRIPTION OF  THE  SELECTED  "NO  ACTION" REMEDY

The OU2 FS evaluated several different remedial alternatives including the "No Action"
alternative for the Ground Water Operable Unit of the Pester Refinery Co. Site based on the
findings of the OU2 RI. Details of the remedy evaluation are included in the OU2 RI and FS in
the Administrative Record file for the Site.

The selected "No Action" remedy for OU2 will involve no additional measures to eliminate,
reduce, or control threats to human health and the environment other than mitigative measures
implemented as components of the OU1 ROD. Those measures include institutional controls in the
form of a deed restriction controlling development of the property and a fence to restrict site
access, the treatment and/or removal of the source material (oily sludge in the ponds), and the
operation of the underground interceptor trench. The underground interceptor trench was
constructed to prevent separate-phase hydrocarbon and dissolved-phase seeps from the ponds from
reaching the river and to recirculate trench effluent to the ponds to maintain the agueous
bioremediation system. KDHE and EPA anticipate that the process of extracting, treating, and
returning the seep water to the ponds for recirculation will also reduce the concentrations of
contaminants in ground water through oil-water separation, physical filtration, biodegradation,
and other natural attenuation processes.

Though not reguired for protection of human health and the environment, KDHE has reguested that
the property owner amend the existing deed restriction to include a restriction preventing the
installation of water wells for purposes other than remediation as an additional conservative
measure. This restriction will run with the property and could only be removed with KDHE's
consent.

Actions to be taken under the selected "No Action" remedial alternative will consist of ground
water monitoring and sediment sampling. Ground water monitoring of the alluvial, upper bedrock,
and lower bedrock aguifers will be conducted to verify that the concentrations of contaminants
in the alluvial aguifer and the upper bedrock aguifer (the Fort Riley Limestone) continue to
decline and that the concentrations of the chemicals of concern  (Table 2) in the Florence
Limestone drinking water aguifer do not increase. Sediment sampling in the West Branch Walnut
River will be conducted to augment the limited information on potential impacts historical
releases from the Site may have had on the river. Ground water monitoring and sediment sampling
will be conducted on a guarterly basis to provide a clearer assessment of baseline conditions,
account for seasonal variations in ground water level and contaminant concentrations, and permit
an assessment of the effectiveness of the OU1 remedy in concomitantly restoring the ground water
conditions at the Site and protecting the West Branch Walnut River.  After two years, the
freguency of monitoring and sediment sampling will be re-evaluated by KDHE and EPA.

As reguired by CERCLA for sites where any hazardous substances, pollutants, or contaminants
remain on site, reviews of the remedy must be conducted by KDHE and EPA at least every five
years after the initiation of the remedy to verify that human health and the environment are
being protected by the remedy. For sites with multiple operable units, one five-year review will
be conducted for the combined operable units; conseguently, the first five-year review for OU1

-------
and OU2 of the Site will be conducted no later than five years  from the  initiation of the
remedial action for OU1, and an additional five-year review will  be conducted at  least every
five years thereafter.   The five-year reviews will continue as  long as hazardous  substances
remain on site.

                    8.0      SIGNIFICANT  CHANGES FROM THE  PROPOSED  PLAN

The Proposed Plan for the Ground Water Operable Unit of the Pester  Refinery Co./Pester Burn  Pond
Site was released for comment in August 1998. That Proposed Plan  identified the No Action
alternative as the preferred alternative for OU2.  Public comments on the Proposed Plan were
evaluated at the end of the public comment period and are addressed in the  attached
Responsiveness Summary. This Record of Decision proposes no significant  changes for OU2 from the
alternative presented in the Proposed Plan.

-------
                                                   Table 1
                                Summary of Compounds Detected in Ground Water
                              Maximum Concentrations - June 1994 Sampling Event
                                       Pester Burn Pond Superfund Site
                                              El Dorado, Kansas
       Compound
                          MCL
Alluvial Aquifer
Fort Riley Aquifer
 (Upper Bedrock)
Florence Limestone
 (Lower Bedrock)
VOCs
Acetone
Beazene
Toluene
Ethylbenzene
Xylenes, total
Carbon Bisulfide
SVOCs
2 , 4-Dimethylphenol
2-Methylphenol
4-Methylphenol
Phenol
Naphthalene
2-Methylnaphthalene
Metals
Arsenic
Barium
Chromium
Lead

NE
5
1000
700
10,000
NE

NE
NE
NE
NE
NE
NE
Total
50 125
2000 1850
100 226
NE 182

46
230
9
24
48
120

30
<10
<10
39
270
420
Filtered
50.9
391
1.1
2.2

30
<5
<5
<5
<5
<5

<10
<10
<10
<10
<10
<10
Total
NA 27.2
NA 383
NA 34.5
NA 2.7

<10
<5 (<1)
<5
<5
<5
6

NA
NA
NA
NA
NA
NA
Filtered
4.1
338
<1.0
<2.0
All results in microqrams per liter
NA = not analyzed
NE = not established
Note that not all estimated quantities are reported in the table.  The diluted sample results are
reported for concentrations that exceeded the instrument calibration.

-------
            Table 2
Chemicals of Potential Concern
Pester Burn Pond Superfund Site
       El Dorado, Kansas

     Chemicals of  Concern

         Acenaphthene
            Arsenic
            Barium
            Benzene
       Carbon Bisulfide
            Chromium
         Dibenzofuran
      Dimethyl Phthalate
      2,4-Dimethylphenol
         Ethylbenzene
            Fluorene
             Lead
      2-Methylnaphthalene
          Naphthalene
         Phenanthrene
            Phenol
            Toluene
            Xylene

-------
                                          RECORD  OF  DECISION


                          PESTER REFINERY CO./PESTER  BURN  POND  SITE
                                    GROUND  WATER OPERABLE UNIT


                                           El  Dorado,  Kansas

                                       RESPONSIVENESS SUMMARY

1.0 Introduction

This Responsiveness Summary for the Ground Water Operable Unit (OU2)  of the Pester Refinery
Co./Pester Burn Pond Site responds to comments received  during the formal public comment  period
on the August 27,  1998. Proposed Plan of Action. The Proposed Plan described the Environmental
Protection Agency (EPA) and Kansas Department  of Health  and Environment (KDHE)  proposal for
remediation strategy for OU2.  This summary also presents EPA and KDHE responses to comments
received during the formal public comment period from August 21  to September 26,  1998.

On September 2, 1998,  a public meeting was held at the Bradford  Memorial Library in El Dorado.
Kansas, to present the Proposed Plan for OU2 and a concurrently  offered Explanation of
Significant Differences for OU1.  Notice for a  30-day public comment  period and the public
meeting was published in the El Dorado Times and was provided to a number of local and regional
public and private entities that that have interest in the Site.  No  oral comments were received
during the public meeting; transcripts of the  meeting have been  added to the Administrative
Record for the Site. Information repositories  containing the Administrative Record for the Site
are available at the Bradford Memorial Library in El Dorado,  at  the  KDHE office at Forbes Field
in Topeka, and at the EPA Region VII office in Kansas City,  Kansas.

Pursuant to Section 117 of the Comprehensive Environmental Response,  Compensation and Liability
Act (CERCLA, or Superfund). 42 U.S.C.  ° 9617,  EPA and KDHE considered all of the comments
received during the public comment period in making the  final decisions for OU2.  This
Responsiveness Summary is part of the Administrative Record for  the  site and an attachment to
the Record of Decision (ROD) for the Site.

2.0    Stakeholder Issues and EPA Responses

No oral comments were presented by any interested parties at the September 2,  1998, public
meeting. Several written comments were received during the formal public comment period:
summaries of these comments and the EPA and KDHE responses are included below.

Comment from the City of El Dorado: The City of El Dorado reports that  periodically their
wastewater treatment plant operators have noted oil sheens on the West  Branch  Walnut  River
while performing river gauging at some distance downstream of the Site.  The City goes on  to
recommend that testing be conducted upstream and downstream of the Site.  "This testing
(biomonitoring) may, be instrumental in detecting damage to the  aguatic life in the west  branch
of the Walnut River, and also in pinpointing a source of contamination  that might possibly be
blamed on the El Dorado Wastewater Treatment Plant some  time in  the  future."

Response: It is unclear from the City's comment whether  the reported incidences of oil sheens
on the West Branch Walnut River are directly attributable to the Pester Burn Pond Site. It
should he noted that there are a number of other current and historical potential sources for
low-volume seepage releases to the river, including, many decades of crude oil exploration and
production along the entire reach of the West  Branch Walnut River. The  Pester  Burn Pond Site
is unigue among industrial facilities along the lower reach of the river in that an active
hydraulic control system is operating at the Site to prevent the discharge of  separate-phase
hydrocarbon seeps to the river. While the Site had historically  resulted in the discharge of
oily-phase material from around water to the river,  those seeps  have decreased dramatically
since the installation of the underground interceptor trench,  and inspections  within  the  last
year have not identified any seeps along the reach of the river  adjacent to the Site.

-------
The EPA and KDHE do not believe that biomonitoring, of the West Branch Walnut River is necessary
at this time as a feature of the Pester Burn Pond Site remedy. No seeps are currently observed
from the Site. Surface water sampling conducted as part of the 1994 OU2 Remedial Investigation
did not detect any VOCs in the river. The OU2 Risk Assessment concluded, based on the limited
data available for the Site, that the primary on-site ecological risk was due to the sludge in
the ponds, which have been removed as part of the Soil and Sludge Operable Unit remedy. Again,
although Site conditions have historically resulted in the discharge of oily wastes to the
river, given the nature of current and historical operations along, the river, it seems unlikely
based on recent observations that any downstream impacts to the river can be clearly attributed
to the Pester Burn Pond Site alone; furthermore, this Site has an active system in place to
prevent any future discharges to the river other than those permitted under the substantive
reguirements of the National Pollution Discharge Elimination System (NPDES).

Sediment sampling will be conducted along the West Branch Walnut River as part of the selected
"No Action" remedy for the Site to augment the limited information available on impacts the Site
may have had on the river. The identification of a significant trend in enrichment of site
contaminants in the river sediments could trigger additional characterization of the impact the
Site has had on the river and its aguatic biota.

As a preliminary suggestion, it would be helpful for the City employees to report the occurrence
and location of any observed oil sheens to the KDHE South Central District Office in Wichita
for investigation. District office staff may be able to identify the source of an active
discharge if the discharge is reported in a timely manner. In the meantime, the City's comment
will be forwarded to the KDHE Bureau of Environmental Field Services for possible follow-up.

Comment from Danny R. Kite, Fina Oil and Chemical Company: The steady state pumping
rate for the underground interceptor trench should be 12 gallons per minute rather than the 17
gallons per minute rate stated in the Proposed Plan on page 11.

Response: The text of the ROD has been modified to reflect the correct steady state pumping
rate for the underground interceptor trench.

Comment from C.T. Dickerson, Coastal Refining and Marketing, Inc.: Coastal Refining and
Marketing reguests that you remove the word "generally" from line 25 on page 2 of the OU2
Proposed Plan.

Response: The ROD has been modified to remove the word "generally" from the corresponding
text.

-------
ATTACHMENT 1
ARARS -
Glossary of Terms

Applicable or Relevant and Appropriate Requirements - Cleanup standards, standards
of control or other environmental protection requirements.
AR File - Record File - includes all pertinent documents and site information which forms the
basis and rationale for selection of a remedial alternative.

Aquifer - A saturated subsurface zone.

Carcinoqen - A compound or substance that increases the incidence of cancer.
Chronic
Exposure -
CERCLA-
COCS -
EPA -
A persistent, recurrinq, or lonq-term exposure. Chronic exposure may result in
health effects that are delayed in onset, occurrinq lonq after exposure has ceased.

Comprehensive Environmental Response, Compensation and Liability Act of 1980. The
federal "Superfund" law.

Chemicals of potential concern, chemicals that are potentially site-related and
whose data are of sufficient quality for use in a risk assessment.

United States Environmental Protection Aqency - The support qovernment aqency for
the Pester Refinery Co. Site.
FS -
Feasibility Study. The study used to evaluate various alternatives to clean up
contamination.
IN SITU - In place, as in treatment of contaminants in their location of occurrence without
excavation or extraction.

KDHE - Kansas Department of Health and Environment. The lead qovernment aqency for the
Pester Refinery Co. Site.

MAHs - Monocyclic Aromatic Hydrocarbons - a qroup of hydrocarbon compounds known for their
similar properties.

MCL - Maximum Contaminant Level - The maximum permissible level of a contaminant in water
which is delivered to any user of a public water supply system.

NPDES - National Pollutant Discharqe Elimination System - a permit that sets standards for
the discharqe of potentially contaminated water.

NCP - National Oil and Hazardous Substances Pollution Continqency Plan. The procedures
used to address the response powers and responsibilities created by the federal
Superfund law.
NPL -
National Priorities List. A list or the most contaminated sites as determined by the
NCP.
OU -
Operable Unit - A discrete area within a larqer site distinquished on the basis of
some defined criteria such as contaminant type, contaminated media, or some other
identifiable demarcation.
PAHS - Polycyclic Aromatic Hydrocarbons - a qroup of hydrocarbon compounds known for their
similar properties.

PRP - Potentially Responsible Party - The party identified by the U.S. EPA which is
potentially responsible for contamination.

-------
RA -
Risk Assessment - Provides an evaluation of the potential threat to human health and
the environment in the absence of remedial action.
RCRA - Resource Conservation and Recovery Act of 1976 - legislation that established
cradle-to-grave accountability for hazardous wastes, from point of generation to
point of ultimate disposal.

RD/RA - Remedial Design/Remedial Action - The remedy phase in the CERCLA process which
involves the description and design of the remedy and the subseguent implementation
of approved plans.
RI -
ROD -

SARA -

Toxic -

VOCS -
Remedial Investigation - The investigative phase in the CERCLA process which
involves field investigation to determine the nature and extent of contamination at
a site, the identification of ARARs for the site, the characterization of risk posed
by contamination at a sit, and the performance of pilot treatability studies for
possible treatment remedies, as necessary.

Record of Decision - The official document by U.S. EPA which selects the remedy to
clean up a Superfund site.

Superfund Amendments and Reauthorization Act of 1986. The federal law which amended
and extended authorization of the original Superfund law (CERCLA).

A descriptor indicating harm to human health or other living organisms.

Volatile Organic Compounds - generally man-made chemicals that are found in many
household, commercial, and industrial products. They are used widely in industrial
processes. VOCs in ground water are a concern due to their potential health effects.

-------