PB94-964071
EPA/ROD/R04-94/204
February 1995
EPA Superfund
Record of Decision:
Agrico Chemical Co.
(O.U. 2), Pensacola, FL
8/18/1994
-------
RECORD OF DECISION
OPERABLE UNIT 2
SUMMARY OF REMEDIAL ALTERNATIVE SELECTION
AGRICO CHEMICAL NPLSITE
PENSACOLA, ESCAMBIA COUNTY, FLORIDA
PREPARED BY
U. S. ENVIRONMENTAL PROTECTION AGENCY
REGION IV
ATLANTA, GEORGIA
-------
Record of Decision
Operable Unit TWO
Declaration
SITE NAME AND LOCATION
Agrico Chemical Site
Pensacola, Escambia County, Florida . •
STATEMENT OF BASIS AND PURPOSE
This decision document presents the.selected remedial action for
Operable Unit Two- at the Agrico Chemical Site in Pensacola,
Florida, which was chosen in accordance with the Comprehensive
Environmental Response, Compensation and Liability Act of 1980
(CERCLA), as amended by the Superfund Amendments and
.Reauthorization Act of 1986 (SARA), and to the extent
practicable, the National Oil and Hazardous Substances Pollution
Contingency Plan (NCP). .This decision is based on the
Administrative Record for the Agrico Chemical Site.
The State of Florida, as represented by the Florida Department of
Environmental Protection (FDEP), has been the support agency.
during the Remedial Investigation and Feasibility Study process
for the Agrico Chemical Site. In accordance with 40 CFR 300.430,
as the support agency, FDEP has provided input during this
process. Based upon comments received from FDEP, it is expected
that concurrence will be forthcoming; however, a formal letter of
concurrence has not yet been received.
ASSESSMENT OF THE SITE
Actual or threatened releases of hazardous substances from this
site, if not addressed by. implementing the action selected in
this Record of Decision (ROD), may present an imminent and
substantial endangerment to public health, welfare, or the
environment. .
DESCRIPTION OF THE REMEDY
This operable unit will address the.ground water and is the final
action of two operable units for the site. The first operable
unit .addresses .the principal threat at the site by treating the ••
most highly, contaminated soils and waste material and then
consolidates them under a RCRA cap. The implementation of the
first operable unit remedy should eliminate contaminant loading
to the ground water at the Agrico Chemical Site. Therefore, EPA
has selected .a Limited Action, for the second operable unit,
which will monitor the ground water conditions as natural
attenuation, flushing and dispersion, occur. . • .
-------
The major components of the selected remedy include:
o Ground-water monitoring of the Sand-And-Gravel Aquifer
o Surface-water monitoring of the Bayou Texar.
o Door-to-door survey of irrigation wells.
o Request access from private landowners to plug and abandon
impacted irrigation wells.
o Advisory Program.
o Utilization of institutional controls to restrict new wells.
STATUTORY DETERMINATIONS
The selected remedy is protective of human health and the
environment, complies with Federal and State requirements' that
are legally applicable or relevant and appropriate to the
remedial action, and is cost-effective. This remedy utilizes
permanent solutions and alternative treatment (or resource
recovery) technologies to the maximum extent practicable. This
remedy does not employ treatment to reduce the toxicity,
mobility, or volume of the Agrico site ground-water
contamination. However, the reduction of toxicity, mobility, and
volume of the site ground-water contamination will be achieved
through source control (OU1) and natural attenuation (OU2).
EPA views the natural attenuation remedy as more protective of
human health and the environment than the pump and treat
technology alternatives considered. The limited actions
envisioned in the selected remedy and in the contingency remedy
avoid adverse impacts associated with the ground-water extraction
and treatment alternatives. The negative impacts of pump and
treat cleanup methods include: spreading of ground-water
contamination not related to the site,.salt-water intrusion, and
alteration of the ground-water flow patterns in the Sand-And-
Gravel aquifer. Because this remedy will result in hazardous
substances remaining in the ground water above health-based
levels, the five-year review requirement will apply to this
action.
j^
,Jr.
John H. Hankinson, Jr. Date
Regional Administrator
-------
Concurrence Sign-Off Sheet
A6RICO CHEMICAL SITE
RECORD OF DECISION
OPERABLE UNIT 2
4WD-SSRB
GOLDBERG
4WD-SSRB
ABBOTT
4WD-SSRB
MONDR3
ORC
MILLER
WMD
GREEN
4DRA
TOBIN
-------
TABLE OF CONTENTS
1.0 SITE LOCATION AND DESCRIPTION 1
2.0 SITE HISTORY AND ENFORCEMENT ACTIVITIES 1
3.0 HIGHLIGHTS OF COMMUNITY PARTICIPATION 5
4.0 SCOPE AND ROLE OF OPERABLE UNIT . 6
5.0 SUMMARY OF SITE CHARACTERISTICS 7
5.1 GENERAL SITE CHARACTERISTICS 7
5.2 GEOLOGY . 7
5.3 HYDROGEOLOGY ^ .................. 10
5.4 LAND USE . . . 10
5.5 RESULTS OF SITE INVESTIGATION . 10
5.5.1 SITE SOIL/SEDIMENT INVESTIGATION 12
5.5.2 GROUND-WATER INVESTIGATION 12
6.0 SUMMARY OF SITE RISKS 22
6.1 SCOPE 22
6.2 CHEMICALS OF CONCERN 23
6.3 EXPOSURE ASSESSMENT INFORMATION .......... 25
6.4 TOXICITY ASSESSMENT INFORMATION 26
6.5 RISK CHARACTERIZATION INFORMATION . 27
6.6 UNCERTAINTIES IN THE RISK ASSESSMENT PROCESS. . ... 30
6.7 ENVIRONMENTAL RISK ....... 39
7.0 DESCRIPTION OF ALTERNATIVES 39
7.1 REMEDIAL ACTION OBJECTIVES 39
7.2 ARARs 41
7.3 DEVELOPMENT AND SCREENING OF ALTERNATIVES 42
7.3.1 PROCESS 42
7.3.2 ALTERNATIVE 1 - NO ACTION 42
7.3.3 ALTERNATIVE 2 - LIMITED ACTION ....... 43
7.3.4 ALTERNATIVE 5 - GROUND-WATER
EXTRACTION, TREATMENT, REINJECTION ..;.'. 45
7.3.5 ALTERNATIVE 6-1.0 MGD GROUND-WATER
EXTRACTION, TREATMENT, REINJECTION 46
7.3.6 ALTERNATIVE 7-1.0 MGD GROUND-WATER
EXTRACTION, TREATMENT, DISCHARGE TO POTW . . 47
7.3.7 ALTERNATIVE 8-1.0 MGD GROUND-WATER
EXTRACTION, TREATMENT, DISCHARGE TO BAYOU . 49
7.3.8 ALTERNATIVE 10 - FUTURE WELL REPLACEMENT
OR WELL HEAD TREATMENT . 50
8.0 SUMMARY OF THE COMPARATIVE ANALYSIS OF ALTERNATIVES . . 51
8.1 CRITERIA FOR EVALUATING REMEDIAL ALTERNATIVES ... 51
8.2 OVERALL PROTECTION OF HUMAN HEALTH AND THE
ENVIRONMENT 53
8.3 COMPLIANCE WITH APPLICABLE OR RELEVANT
AND APPROPRIATE REQUIREMENTS 54
8.4 LONG-TERM EFFECTIVENESS AND PERMANENCE ...... 54
-------
8.5 REDUCTION OF TOXICITY, MOBILITY OR VOLUME
THROUGH TREATMENT 55
8.6 SHORT-TERM EFFECTIVENESS 55
8.7 IMPLEMENTABILITY 56
8.8 COST 57
8.8 STATE ACCEPTANCE 57
8.9 COMMUNITY ACCEPTANCE 57
9.0 SELECTED REMEDY 57
9.1 CONTINGENCY REMEDY 61
10.0 STATUTORY DETERMINATIONS 65
10.1 PROTECTION OF HUMAN HEALTH AND THE ENVIRONMENT .... 65
10.2 . COMPLIANCE WITH APPLICABLE OR RELEVANT AND
APPROPRIATE REQUIREMENTS 66
10.2.1 CHEMICAL-SPECIFIC ARARs 66
10.2.2 LOCATION-SPECIFIC ARARs . . 67
10.3 COST EFFECTIVENESS 67
10.4 UTILIZATION OF PERMANENT SOLUTIONS 67
10.5 PREFERENCE FOR TREATMENT AS A PRINCIPAL ELEMENT .... 68
10.6 DOCUMENT OF SIGNIFICANT CHANGES 68
Tables
Table 6-1 Reference Doses (RfDS), Cancer Slope Factors
(CSFs), And EPA Cancer Classification
For Constituents Of Concern 28
Table 6-2 Risk Characterization For Current Ingestion
Of Ground Water From The Shallow Zone
Of The Sand-And-Gravel Aquifer 31
Table 6-3 Risk Characterization For Current Ingestion
Of Ground Water From The Deep Zone Of The
Sand-And-Gravel Aquifer 32
Table 6-4 Risk Characterization For Current Ingestion
Of Ground Water Containing Radionuclides 33
Table 6-5 Risk Characterization For Current Ingestion
Of Ground Water From The Shallow Zone Of
The Sand-And-Gravel Aquifer . 34
Table 6-6 Risk Characterization For Future Ingestion
Of Ground Water From The Deep Zone
Of The Sand-And-Gravel Aquifer 35
Table 6-7 Risk Characterization For Future Contact
With Ground Water From The Shallow Zone
Of The Sand-And-Gravel Aquifer 36
Table 6-8 Risk Characterization For Future Contact
With Ground Water From The Deep Zone
Of The Sand-And-Gravel Aquifer 37
Table 6-9 Risk Characterization For Future
Ingestion Of Ground Water Containing
Radionuclides 38
-------
Figures
Figure 1-
Figure 1-
Figure 5-
Figure 5-
Figure 5-
Figure 5-
Figure 5-
Figure 5-
Figure 5-
Figure 5-
Figure 5-
Figure
5-
Appendix A
Appendix B
1 - Regional Area Map and Site Location ......
2 Former Pond Boundaries .
1 Regional Stratigraphic and Hydrogeologic Column
2 Geologic Cross-Section for the Study Area . . .
3 Surrounding Land Use . . .
4 Fluoride Concentrations .... ...
5 Vertical Distribution X-Section for Fluoride .
6 Fluoride Concentrations in the Shallow Zone
7 Sampling Locations Bayou Texar Study .....
8 Fluoride Values Bayou Texar Study . .
9 Fluoride Concentrations in Pore Water
10 Fluoride Concentrations in Pore Water .....
2
, 3
, 8
, 9
11
14
15
16
18
19
20
21
Appendices
Correspondence
Responsiveness Summary
-------
Decision Summary
Record of Decision
Operable Unit Two
Agrico Chemical Site
Pensacola, Florida
1.0 SITE LOCATION AMD DESCRIPTION
The Agrico Chemical Site (hereinafter, "the site") is located in
Pensacola, Escambia County, Florida and covers approximately 35
acres. The site is located at the northwest corner of Fairfield
Drive and Interstate 110 (Figure 1-1). The site is bounded by
Interstate 110 to the east, Fairfield Drive to the south, the CSX
railroad tracks to the west, and includes an abandoned baseball
field to the north.
Site topography is flat and ground-surface elevations range from
about 81 feet to 90 feet National Geodetic Datum (NGVD) . Areal
photographs, field observations, and a topographic map indicate
that surface drainage is currently contained on site. The
concrete foundations of former plant buildings remain on site.
Storage warehouses, in the southern portion of the site, are the
only structures on site. Areas- of the site not occupied by the
storage warehouses, old foundations, and the abandoned baseball
field, are open and characterized by tall grass, brush, and
trees. The site is currently owned by Fred L. Vigodsky, Edwin
Walborsky, and James Lamar Dean, d/b/a MARGOD, a Florida
partnership, and F.A. Baird, Jr.
Four ponds were used for wastewater discharge and designated PFP
I through PFP IV for the purposes of the Remedial Investigation/
Feasibility Study (RI/FS) documents (Figure 1-2). Aerial
photographs taken from 1940 through 1990 indicate that the ponds
were interconnected throughout different time periods, but the
area referenced as PFP II appears to have received the majority
of the sludge. A portion of PFP II is visible at land surface.
In addition, historic aerial photographs show a drainage ditch
through PFP IV across Fairfield Drive. A fence has been
constructed around the former ponds and southern-most baseball
field.
2.0 SITE HISTORY AND ENFfHtrmgNT ACTIVITIES
Industrial processes at the site began in 1889 by a company that
produced sulfuric acid from pyrite. The production of sulfuric
acid continued at the plant until approximately 1920. The
sulfuric acid was manufactured in lead pots, in a building
slightly north of PFP IV. Production of normalsuperphosphate
fertilizer was initiated in 1920. The source rock used in the
process was fluorapatite, which also contained silica and trace
levels of many metals such as aluminum, along with uranium at 20
to 200 parts per million (ppm) as impurities. Superphosphate was
produced through the digestion of the source rock with sulfuric
acid and water. The reaction produced anhydrite and fluoride as
-------
DWC DA1E: OVIO/q?/iq K \ PRJCT NO.: JFO?9I9 [TILE NO.: N/A [DRAWING H/A [
CHECKED: A VIDAl
I APPROVED: C. RORECH IDRAFtER: D. WOOD
PENSACOLA
VILLAGE
APTS.
GULF POWER
RIGHT-OF-WAY
SITE LOCATION -
SEE DETAIL
so.ooo n.
REGIONAL AREA MAP
AND SITE LOCATION
FEASIBILITY STUDY
*VGRICO CHEMICAL SITE
TNSACOLA. TLORIDA
FIGURE
1-1
-------
CALD 01
1300
Former Pond Boundaries
Phase I Remedial Investigation
• Agrico Chemical Site
Pensacola, Florida .
FIGU
1-
-------
byproducts. The anhydrite portion remained with the product and
was sold as- part of it, unlike modern wet process phosphoric acid
plants which filter the anhydrite (phosphogypsum) out and
stockpile it on-site.
The historic site files indicate that the plant purchased and
apparently utilized some volume of spent sulfuric acid during
1967 and 1968. Dinitrotoluenes were constituents of this acid in
concentrations which ranged from 5 ppm to 1600 ppm.
Additional ground-water and soils samples were analyzed for
organics as part of the Phase II RI study (discussed in detail
below) to investigate the possible movement of dinitrotoluenes
from the plant, a potential source of which may have been the
acid.
By early 1957, Pensacola City officials shut down a public supply
well located downgradient of the site. The City's decision to
remove the well from service was based on analyses indicating
declining pH values and elevated levels of fluoride and sulfate
in the ground water. Plant operations ceased in 1975.
EPA conducted a Hazardous Waste Site Investigation in October
1983. The results of the study indicated that the on-site soils
and surface water were contaminated with elevated levels of
fluoride and lead. No attempt was made to install temporary
wells to sample the ground water. However, an effort was made to
locate any private shallow wells in the area; none were located.
The Florida Department of Environmental Protection (FDEP)
conducted a ground-water assessment at the site in January, 1987.
The study concluded that the site contaminants, primarily
fluoride and sulfate, had polluted the area ground water. EPA
listed the site on the National Priorities List (NPL) on October
4, 1989.
Conoco Inc. ("Conoco") and Freeport McMoRan Resource Partners,
Ltd. ("Freeport") entered into an Administrative Order on Consent
(AOC) on September 29, 1989. According to the terms of the AOC,
Conoco and Freeport agreed to conduct the source (soils) and
ground water RI/FS at the site. The RI field study was conducted
in two (2) phases. The first phase was conducted in the summer
of 1990. Subsequent confirmatory sampling was necessary in 1991.
The second phase RI field study was conducted in February 1992 to
more fully define the nature and extent of impacts caused by the
site. The Feasibility Study (FS) for treatment of the site's
contaminated soils was complete in July 1992. A Record of
Decision (ROD) for Operable Unit One (OU1) was issued by EPA,
Region IV on September 29, 1992. The OU1 ROD presents EPA's
selected remedial action for treatment of the soils on site. The
major components of the selected remedy include:
-------
o Excavation and solidification/stabilization of
approximately 32,500 cubic yards of contaminated sludge
and soils from site sludge ponds.
o Consolidation of all stabilized sludge and soils into
one sludge pond (approximately 500,000 cubic yards of
contaminated soils and sludge).
o Construction of a RCRA cap over the sludge pond.
o Construction of a slurry wall around the RCRA cap.
o Implementation of institutional controls to include
security fencing, access and site deed restrictions.
The previous Phase I and Phase II RI field studies included
characterization of the nature and extent of the ground-water
contamination related to the site. However, further
investigation of the ground-water contamination related to the
site was necessary. Specifically, potential impacts to the
nearby Bayou Texar were investigated. The results of the study
were presented in a report submitted to EPA in Hay 1993. In
addition, the Florida Department of Environmental Protection
(FDEP) requested that additional monitoring wells be installed to
further define contaminant levels within the previously defined
ground-water plume. The field work associated with the
investigation of the ground water contamination was completed in
October 1993. The Final RI/FS Report (for ground water) was
approved by EPA in November 1993. This document represents the
EPA's Record of Decision (ROD) for the ground-water contamination
at the site and will be presented as the ROD for Operable Unit
Two (OU2).
3.0 HIJE3T-IGHTS OF COMMUNITY PARTICIPATION
In accordance with public participation requirements of CERCLA
Sections 113(k) (2) (B) (i-v) and 11.7, a comprehensive community
relations program was developed and implemented throughout the
remedial process at the Agrico Chemical site. EPA conducted
community interviews in the winter of 1990.
In June 1990, a Remedial investigation /Feasibility Study (RI/FS)
Kick-Off Fact Sheet was prepared and delivered to interested
citizens and local officials included on the site's mailing list.
The fact sheet explained the overall process of Superfund, the
upcoming RI/FS at the site, and opportunities for community
involvement. A RI/FS Kick-Off Public Meeting was held on June 5,
1992 with approximately 35 interested citizens.of Pensacola,
Florida to discuss the activities that were to take place as part
of the investigation, and to answer any questions the public
might have regarding the upcoming investigation. Television
interviews were conducted with the local ABC affiliate station on
-------
two occasions during the field investigations.
The Public "comment period for the OU1 ROD was from August 6, 1992
to September 5, 1992. A Proposed Plan public meeting was held on
August 13, 1992. No comments were received during the comment
period. However, citizens' questions and comments during the
public meeting were incorporated into the OU1 ROD as the
Responsiveness Summary (Appendix A). EPA issued the ROD for OU1
(soil contamination) on September 29, 1992.
In 1993, a citizens' group (Citizens Against Toxic Exposure or
"GATE") applied for a Technical Assistance Grant (TAG) for the
Agrico Chemical Site. The TAG was awarded to the above group in
May 1993.
The comment period for the Proposed Plan for OU2 (ground water)
was from February 15, 1994 to March 17, 1994. During the comment
period, the Administrative Record was available to the public at
both the information respository maintained at the Pensacola
Public Library and at the EPA Region IV Docket Room in Atlanta,
Georgia.
A Proposed Plan public meeting for OU2 was held on Friday,
February 18, 1994. At the meeting, representatives from EPA
presented the Agency's preferred alternative for ground-water
cleanup of the site and answered any questions the public had
regarding the preferred alternative.
Approximately 50 interested persons attended this public meeting.
Some residents of streets near the Site questioned why EPA had
not investigated their properties for possible soil contamination
from the Agrico site. A response to all comments received
concerning the Proposed Plan for OU2 during the public comment
period, including the comments made at the public meeting, is
Included in the Responsiveness Summary (Appendix B).
Therefore, this decision document presents the selected ground-
water remedial action for the Agrico Chemical site in Pensacola,
Florida, chosen in accordance with CERCLA, as amended by SARA,
and, to the extent practicable, the National Contingency Plan.
The decision for this site is based on the Administrative Record
for the site.
4.0 SCOPE AMD ROT.R QP QPERAttT.B UNIT
As with many Super fund sites, the problems at the Agrico Chemical
site are complex. As a result, EPA organized the remedy into two
operable units (OUs). These are:
• OU1: Contamination of the soils.
• OU2: Contamination of the ground water.
6
-------
As referenced in previous sections of this document, EPA has
selected a.remedy for OU1. The selection can be found in the
Record of Decision for OU1, issued by EPA, Region IV on September
29, 1992. This ROD will address EPA's selected remedy for OU2,
contamination of the ground water. This second operable unit
will be the second and final response action for the site.
5.0 SUMMARY OF SITE CHARACTERISTICS
The purpose of this section is to discuss the general site
physical characteristics and to discuss the results of the ground
water characterization Remedial Investigation. A detailed
discussion of source (soil) contamination is addressed in the
Phase I Remedial Investigation Report, Geraghty & Miller, Inc.,
March 1992, and the Phase II Remedial Investigation Report.
Geraghty & Miller, Inc., August 1992. In addition, a short.
summary of the extent of soil contamination is discussed in this
section. However, a selected remedy for soil contamination has
been selected and is detailed in the Agency's ROD for OU1.
5.1 GENERAL STTB CHARACTERISTICS
Climate in the Pensacola area is characterized by mild winters
and relatively long, humid, warm summers. The. average annual
rainfall is 62 inches. Pensacola is located in the coastal
lowlands, a subdivision of the Coastal Plain province. The
coastal lowlands are relatively level with an elevation of 100
feet NGVD or less. The most significant topographic features are
step-like Pleistocene marine terraces that generally parallel the
coast. The Agrico Chemical site is within a plain of the
Pleistocene terraces and, as a result, is relatively flat. Site
elevation varies from about 81 feet to 90 feet NGVD.
5.2 GEOLOGY
The uppermost sediments in the Pensacola area are marine terraces
and beach scarps, and the Citronelle Formation. The marine
terraces and beach scarps are composed mainly of quartz sand but
contain some clay, silt, gravel,'and iron oxide-cemented
sandstone, locally referred to as hardpan. The sediments were
deposited in marine, deltaic, and fluvial environments and are
Middle Oligocene to Pleistocene in age (Figure 5-1).
.**•
The site geology's predominant lithologies are sands, clayey
sands, and sandy clays. A geologic cross section showing the
variation in lithology over the study area was constructed from
monitoring wells AC-ID, AC-2D, AC-3D, AC-10D, and AC-1ID (Figure
5-2). A shallow layer of sandy clay, present over much of the
area, is encountered at approximately 10 feet below land surface
(bis.) in the vicinity of the site and is about 30 feet thick.
The sandy clay layer tends to thicken towards the south of the
study area where it is found at land surface. Descriptions of
-------
Hill NO AGSIRAI 11 A|i HI) ACSIRAT
ICOMPIItN Ml lOOt MirillSUI I UGH ANDY Mil If R
IDRAI ItR I'AUIIA CAM I
SER6S
PlEISTOCtNl
ttTHOLOOY
FORMATION
EXPLANATTON
UPPER MlOCIHt
UPPER MIOW to
LOWtR Ul'I'lP
M10O.NI
LOWER MIOtlMt
AND
OMCOCINl
MARINt URRACt OtPOSIlS
CIIRONLUE I OHM A DON
- 390
MIOCENE COARSE ClAStlCS
IINCkHCSS TO - 300' WCK PRtUNt
PENSACOl A Cl AY
APPftOKWAIl IIDCKHtSS UPPf* ClAV IS 140 - 680'
S«NO « JO -ttff
IO>HB CIAY 15 IM - 33O'
CHICKASAWHAY LIMESTONE AND
TAMPA FORMATION
»oo -
-------
i.Wii |.M( ;n«/.'//'.y.'i'. VJHMCI no JTOJ-MI.- |im tip II/A _ [finAwinu H/A _ [fig CM n *
MIIUH
lArr-novm o WMnt IHCAIIIH 11
• NOHTHWEST
100 -, A
50 -
o
>
o
z
Ul
UJ
u.
2
0 -
50 -
1*1
d 100
SOUTHEAST
A'
i. •
AC-tO AC-20
I
AC-100
AC-11D
150 -
SOURCE: Wll KINS, t I At.. (198B)
SANDY CLAY
CLAYEY SAMD
1/2
APPR'OX HORIZONTAL
SCALE IN MILES
Geologic Cross-Section for the Study Area
I Remedial Investigation
Agrico Chcmicftl Site
Pensaco/a. Florida
FIGURE
5-2
-------
soil samples taken for the RI indicate that soils on the site
could be classified as Lakeland and Eustis. These soils are
characterized by rapid external and internal drainage, small
inclusions of organic matter, low fertility, and susceptibility
to erosion in sloping areas.
5.3 HYDROGEOLOGY
The upper-most water bearing unit in Pensacola is the Sand-And-
Gravel Aquifer. The upper limit of the aquifer coincides with
the surface of the water table and the lower limit of the aquifer
coincides with the top of the Pensacola Clay. The Pensacola
Clay, which is vertically persistent in the Pensacola area, acts
as a confining unit between the Sand-And-Gravel Aquifer and the
upper portion of the Floridan Aquifer. According to lithologic
logs from wells installed for the Phase I and Phase II RI and
previous studies, the Sand-And-Gravel aquifer is approximately
270 feet in thickness and extends to that depth below land
surface (bis) in the vicinity of the site. The Sand-And-Gravel
Aquifer is characterized by various permeability zones.
Generally,.two permeability zones are identified for purposes of
defining ground water flow direction. The potentiometric surface
contour maps for both shallow and deep zones of the aquifer show
that ground-water flow direction is essentially east-southeast in
the vicinity of the Agrico site, but becomes easterly as it
approaches Bayou Texar.
5.4 LAND USE
The land use surrounding the site within a 1-mile radius consists
of residential, municipal, commercial, and industrial zoned areas
(Figure 5-3). Immediately to the north and west of the site is
an industrial zoned area. The Escambia Wood Treating Company
site is located in this area. EPA has excavated contaminated
soil from the site and stockpiled it on site as an emergency
response action. The eastern border of the Agrico Chemical Site
is an Interstate Highway. East of the Interstate is mainly a
commercial zone. West of the site is industrial and residential.
South of the site is commercial and residential.
5.5 RESULTS OF SITE INVESTIGATION
The Agrico Chemical site RI was conducted in two phases. Field
work for the Phase I RI was initiated in mid-1990. Confirmatory
sampling was required in April 1991, and again in November, 1991.
Field investigations for the Phase II RI were conducted in
February, 1992. Constituents to be addressed during the Remedial
Investigations were identified by locating and sampling the
primary areas of waste deposition. Available records indicate
that the four impoundment ponds had received industrial
wastewater and sludges (Figure 1-2).
10
-------
8
S
.
I
6
6
I
•o
ui
2
5
O
O
i
EXPLANATION
SITE
MUNICIPAL
RESIDENTIAL
COMMERCIAL
INDUSTRIAL
ss
2
SCALE: 0
2000
SURROUNDING LAND USE
AGRICO CHEMICAL SITE
PENSACOLA. FLORIDA
FIGURE
5-3
-------
Soil borings were used to investigate the horizontal and vertical
extent of Haste deposition within the four sludge ponds. Based
on historical areal photographs, isolated areas off-site were
also targeted for sampling and analyses. Samples of soil and
sludge were analyzed for the TAL/TCL parameters to provide a
characterization of the contamination. Elevated levels of
radionuclide activity, if present at a fertilizer production
plant, are generally associated with the gypsum stacks. However,
at the Agrico site the anhydrite (gypsum) portion remained with
the product and was sold as part of the final product.
Nevertheless, EPA required that targeted areas on site were to be
analyzed for radionuclides.
A detailed discussion of the nature and extent of the site's soil
contamination can be found in Section 5.5.1 SITE SOIL/SEDIMENT
INVESTIGATION of the OU1 ROD (September 1992).
An existing monitoring well network, established specifically for
a previous FDEP investigation of the site, was supplemented by
newly installed and other already existing monitoring wells.
During the first phase of the RI, samples of ground water were
analyzed for the TAL/TCL parameters. Ground-water sampling
efforts during the second phase RI focused on site specific
inorganic constituents, organic constituents, and radionuclides.
In addition, the investigation of the ground-water contamination
included the study of the porewater (interstitial water) in the
bottom sediments of the. Bayou Texar, located approximately one
mile hydraulically downgradient of the site. The RI of the
contaminated ground water was completed in November 1993.
5.5.1 SITE SOIL/SEDIMENT INVESTIGATION
The contaminants of concern in the soils and sludges at the
Agrico site are primarily fluoride, lead, and arsenic. The
highest contamination of fluoride is found in the areas of PFP
II, PFP III, and PFP I. Arsenic is found in slightly elevated
levels throughout the above impoundment ponds. The lead
contamination is generally confined to the area of PFP IV. A
detailed discussion of the nature and extent of the site's soil
contamination can be found in Section 5.5.1 SITE SOIL/SEDIMENT
INVESTIGATION of the OU1 ROD.
5.5.2 GROUND-WATER INVESTIGATION
To define the extent of ground-water contamination associated
with the Agrico Chemical site, six (6) two-well clusters,
screened in the shallow and deep zones of the sand-and-gravel
aquifer, were installed during the two phased RI study. In
addition, three (3) new shallow wells and three (3) new deep
wells were added to the existing well network.
The majority of the shallow (s) monitoring wells are screened
12
-------
between 60 and 80 feet below land surface (bis), with three
shallow wells adjacent to Bayou Texar screened between 25 and 45
feet bis. Most of the deep (d) monitoring wells are screened
from 150 to 170 feet bis, while a few are screened from 190 to
220 feet bis.
Finally, in August 1993, the PDEP requested additional monitoring
wells be installed downgradient of the site. The state agency
requested that ground-water samples be collected to further
define and confirm previous off-site ground water quality
characteristics for the lower portion of the main producing zone
for the Sand-and-Gravel Aquifer. A total of thirty (30)
monitoring wells comprise the Agrico Chemical monitoring well
network. ,
The discharge of wastewater has generated a ground-water plume of
contaminants that is approximately 7,000 feet long and 5,000 feet
wide. The monitoring well network indicates that the plume of
inorganic constituents has migrated to the Bayou Texar within the
northern portion of the study area (Figure 5-4). The most highly
concentrated area of the plume is in the lower portion of the
Sand-And-Gravel aquifer, approximately 2,000 feet downgradient of
the site (Figure 5-5). The site ground-water contamination is
less pervasive in the shallow zone of the Sand-And-Gravel aquifer
(Figure 5-6). The above figures represent the distribution of
fluoride in the upper and lower portions of the sand-and-gravel
aquifer.
Concentrations of fluoride, the site's primary ground-water
contaminant, range from below the detection limit of 0.2 mg/1 in
samples from wells located outside of the plume to 98.0 mg/1
detected in the shallow zone at AC-2S. This well is located
approximately 500 feet downgradient of the large on-site PFP II
impoundment pond.
Other inorganic contaminants are present in the aquifer as the
result of plant processes. Concentrations of sulfate range from
below the detection limit of 5 mg/1 to 570 mg/1. The maximum
contaminant level (MCL) for sulfate is 250 mg/1. The highest
sulfate concentration was detected at AC-3D. Concentrations of
nitrate/nitrite range from 0.10 mg/1 to 42 mg/1. The MCL for
nitrate/nitrite combined is 10 mg/1. The highest nitrate/nitrite
13
-------
ent
\
SOUTH
ETC/VIW-4
AC-
Site
AC*
BDL
AC-5
NWO-4
I BDL
AC-23
BDL
AC-7
BDL
• BDL
Gould'mg
.80
|10S
Carpenter Creek
19
AC-27
AC-9
BDL
m
LI
AC-
1.3.
AC
AC-e
0.28
J>
V
HYD
AC-28
Monitor V«U
nuortd*
Conecatntion toocontour (m«/l)
ConecntnUeo* (nc/0
12000
Fluoride Concentrations
in the Deep Zone
Phase H Remedial Investigation
Agrico Chemical Site
Pensacola. Florida
-------
HMIi HA 1C: 0«/2V'J?/H l5 I ''"•*•'
u-'Jlf>
"" N':': V'"*-'1-'
i;- flVlllSI (ltCKtl>. A MIIUR
lAPCROVtD. U
U WOOD
100
-100
-200
-300
< 0.20
TOP OF CONFINING
UNIT
'//,'//, / / / /////////////
////////////,
.
, ///////, /////// , , /,
V^_
I
. \ —
5.5
LEGEND
WATER LEVEL ELEVATION
SCREENED INTERVAL
FLUORIDE CONCENTRATION VERTICAL DISTRIBUTION
-------
Pensacola
Regional
Airport
IMI Runt** C«no»U*Uon
bMMlMr (•(/!)
IMt VhMrtte CMMmtralfaa lmt/0
Fluoride Concentrations
in the Shallow Zone, 1992
Phase II Remedial Investigation
Agrloo Chemical Site
Pensacola. Florida
-------
concentration was detected at AC-3D. The concentrations of
aluminum vary from below laboratory detection limits to 58 mg/1.
There is no MCL for Aluminum. Concentrations of total
recoverable and dissolved aluminum in samples from wells in the
Pensacola area range from below detection limits to 1,000 mg/1.
The concentrations of chloride in samples from the monitoring
wells range from 5.5 mg/1 to 270 mg/1. The maximum contaminant
level (MCL) for chloride is 250 mg/1. Water samples were
analyzed for arsenic, chromium, copper, and lead since these are
trace constituents of plant process raw materials. Each
constituent was below laboratory detection limits or within
federal drinking water standards in the samples, with the
exception of arsenic in Monitoring Wells AC-2S and AC-7S. The
arsenic levels detected in these wells were 0,164 mg/1 in AC-2S
and 0.088 mg/1 in AC-7S. The MCL for Arsenic is 0.05 mg/1. The
results of radionuclide analyses indicate that gross alpha
concentrations are below the primary drinking water standard of
15 picoCuries per liter (pCi/1). The concentrations of radium-
226 and radium-228 ranged from 0.4 pCi/1 to 8.4 pCi/1 and from
0.8 pCi/1 to 12.4 pCi/1 respectively. The current primary
drinking water standard for combined radium-226 and radium-228 is
5pCi/l. However, the proposed regulation radium-226 and radium-
226 is 20 pCi/1 for each.
Organic compounds are also present in the aquifer, but these
compounds appear to originate from the Escambia Wood Treating
Company site, a former wood treatment facility located
approximately 2,500 feet hydraulically upgradient from the Agrico
Chemical Site. Ground-water sampling analyses indicate that
organics are below detection limits in all wells with the
exception of 1.9 ug/1 benzene in AC-ID; 1.5 ug/1 benzene in AC-
7S; 3.7 ug/1 benzene in AC-24D; 2.3 ug/1 benzene, and 8.6 ug/1
2,4 Dinitrotoluene in AC-25D.
A study of the Bayou Texar was conducted to determine if ground
water migrating from the site is discharging to the bayou. This
determination was accomplished by analyzing for Agrico indicator
parameters (i.e., fluoride, chloride* nitrate, and sulfate) in
the pbrewater of the bayou sediments and observing a vertical
increase in concentration with depth. Aluminum was also included
in the analyses, as was ammonia. Figure 5-7 depicts the transect
and sample locations selected for the study.
Out of 27 .cores collected during the study, four cores showed the
presence of fluoride in the sediment porewater which exhibited a
vertical increase in concentration down the core. Aluminum was
also detected in a select number of cores. Figures 5-8, 5-9, and
5-10 depict the fluoride distribution found in the bayou
sediments. Four cores from Bayou Texar (cores 3B, 9B, 12B, and
12C) show a pattern of fluoride concentration increasing in
depth. The fluoride maxima within these four cores are 64, 160,
44, and 240 mg/L, respectively. Fluoride concentrations in the
17
-------
TRANSECT 1
TRANSECT 2
TRANSECT 8-1
TRANSECT 7
TRANSECT 3
r-TRANSECT
TRANSECT 4
TRANSECT
TRANSECT 10
TRANSECT
BAYOU TEXAR
TRANSECT 12
TRANSECT 9
TRANSECT 11
FEET
APPROX. SCALE
LESIMB
AC-21O MONITORING WELL
Figure 5-7'
Sampling Locations
Bayou Texar Study
XXX TRANSECT LOCATION
SAMPLE LOCATIONS
-------
RAYQU TEXAR
FEET
APPROX. SCALE
Figure 5-8
Core Top and Bottom
Fluoride Values
Bayou Texar SJLudy
Pcnsacola, Florida
miQRIDE VALUES. mg/L
-------
Concentrations In mg/l
Contour Interval Is Logarithmic
FILti CHtK\MAPS\UPRFl
TITIE:
Fluoride Concentration in Pore Water
Upper Core Interval
Bayou Texar - Agrico Chemical Site
OWN:
KPL
CHKD:
RGS
DME:
5/4/93
DES.:
APPO:
RGS
REV.:
PROJECT NO.:
402412
FIGURE NO.:
5-9
-------
0 400 800
FEET
niEl CHEM\MArt\lWRn.
TITLE:
Fluoride Concentrations in Pore Water
Lower Core Interval •
Bayou Texar - Agrico Chemical Site
OWN:
KPL
CHKO:
RGS
DATE:
5/4/93
OES.:
APPO:
RGS
REV.:
PROJECT NO.:
402412
FIGURE NO.:
5-10
UK it.: v»J
-------
porewater of Bayou Texar sediments range from below detection
limit (BDL) to 244.3 mg/L. Concentrations below detection limit
(detection "limit = 0.1 mg/L) were recorded for 55 samples in 18
cores.
In addition to fluoride, the Bayou Texar study (Entrix, 1993)
sampled and analyzed several indicator contaminants, including
chloride, sulfate, and nitrate/nitrite. The results of this
study indicated that chloride concentrations ranged from 2.6 mg/L
to 9,600 mg/L. For comparison, standard sea water has a chloride
concentration of 19,000 mg/L. The ranges of chloride measured in
the near-surface porewater of Bayou Texar sediments should have
no adverse effect on the biota since the chloride concentrations
are within normal limits for estuarine habitats such as the Bayou
Texar. Sulfate concentrations ranged from 1.7 mg/L to a maximum
of 1,800 mg/L. Since sulfate is ubiquitous in marine and
estuarine waters, the levels detected in the study are within
limits for a salt water and fresh water interface zone such as
the bayou. Therefore, it is unlikely that sulfate would
adversely impact the bayou biota. Nitrate/nitrite concentrations
ranged from below detection limit to a 8.6 mg/L. The
nitrate/nitrite limits detected in the study are within
acceptable limits for protection of the ecosystem.
In general, the fluoride data indicates that ground water
originating from the Agrico site is discharging to the bayou.
The discharge zone appears to be well-defined and limited in
areal extent.
A review of the Agrico ground-water plume discharge to the bayou
and its impact on biota has been completed by the EPA's Ground
Water Technical Support Unit (GWTSU). The GWTSU dilution
calculations indicate that the fluoride concentrations in the
ground water would have to be 4050 mg/L or greater in order to
exceed the State of Florida's surface water standard of 5 mg/L.
Since the highest level of fluoride found in the ground water is
98 mg/L, it is unlikely that the discharge of the ground water
plume into Bayou Texar will result in impacts to fish or
wildlife.
6.0 SUMMARY OF SITE RISKS
6.1 SCOPE
The baseline risk assessment (BRA) provides the basis for taking
action and indicates the exposure pathways that need to be
addressed by the remedial action. It serves as the baseline
indicating what risks could exist if no action were taken at the
site. This section of the ROD reports the results of the
baseline risk assessment conducted for this site.
A BRA was conducted by EPA as part of the RI to estimate the
22
-------
health or environmental problems that could result if the Agrico
site is not remediated. Results are contained in the Final Risk
Assessment "Report. A BRA represents an evaluation of the "No
Action" alternative, in that it identifies the risk present if no
remedial action is taken. The assessment considers environmental
media and exposure pathways that could result in unacceptable
levels of exposure now or in the foreseeable future. Data
collected and analyzed during the RI provided the basis for the
risk evaluation. The BRA process can be divided into four
components: contaminant identification, exposure assessment,
toxicity assessment, and risk characterization.
The BRA was developed for the source (soils) control and ground-
water RI/FS. The source control risk assessment is discussed in
detail in the Operable Unit One ROD (September 1992). The
ground-water risk assessment is discussed in the remainder of
Section 6 of this Record of Decision.
6.2 CHEMICALS OF CONCERN
Compiled aerial photos (ranging in age from 1940 to 1990)
illustrate that water from the facility operations ponded in four
general areas. These areas are depicted previously on Figure 1-2
(page 3) and will be referred to in this section as PFP I, PFP
II, PFP III, and PFP IV. Other site areas referred to in the
following risk sections are the on-site baseball field and the
former drainage ditch associated with PFP IV. The four ponds are
now devoid of any surface water. Surface drainage is currently
contained on site as a result of natural topography and berms
associated with road construction. For the purposes of this ROD,
ground water is the media of concern at the site.
Chemicals detected on site were evaluated for the selection of
Chemicals of Potential Concern (COPCs). Risks for each COPC were
calculated in the baseline risk assessment. Chemicals in
pathways with risk levels exceeding 1 x 10~4 or Hazard Index
exceeding one (1) with chemical-specific risks greater than 1 x
10~6 or Hazard Quotients exceeding 0.1 or chemicals with
concentrations exceeding a chemical-specific ARAR are considered
chemicals of concern (COCs). A detailed explanation of the risk
level and hazard index numbers are found in the risk discussion
after Table 6-1.
A plume of site-related COCs extends hydraulically downgradient
from the site. A summary of the COCs in the shallow and deep
zones for the Sand-And-Gravel Aquifer is presented below.
23
-------
Shallow Zone of Sand-and-Gravel Aquifer
Chemical of
Concern
Frequency of
Detection
Mi ni mirm—
Value*
Exposure
Point
Concentration*
Arsenic
Fluoride
Manganese
Mercury
Nickel
Nitrate
Benzene
2,4 Dinitro-
toluene
Radium 226
Radium 228
4/10
7/10
1/1
2/7
1/7
10/10
1/5
1/5
1/1
1/1
0.04-0.16
1.7-98
0.33
0.00031-
0.00071
0.055
0.88-15
0.0015
0.025
1.3
1.2
0.16
98
0.33
0.00059
0.035
15
0.0015
0.025
0.8
1.2
1 The numerical values expressed in these columns represent
milligrams per liter (mg/1).
DeeD Zone of Sand-and-Gravel Aouifer
Chemical of
Concern
Frequency of
Detection
Value*
Exposure
Point
Concentration*
Arsenic
Copper
Fluoride
Mercury
Nickel
Nitrate
Benzene
Dibenzofuran
Fluorene
2 Methylnaph-
thalene
1/20
7/20
14/20
2/10
1/10
19/20
3/9
2/9
2/9
6/9
0.01
0.033-0.63
1.1-80
0.00046-0.0015
0.0024
0.10-47
0.0019-0.0037
0.015-0.019
0.025-0.039
0.014-0.11
0.0056
0.3
80
0.0006
0.032
47
0.0029
0.036
0.024
0.11
24
-------
Naphthalene
2,4-Dinitro-
toluene
2 , 6-Dinitro-
toluene
Radium 226
Radium 228
8/9
3/9
1/9
5/5
5/5
0.011-1.5
0.021-0.082
0.024
2.8-8.4
2.4-12.4
1.5
0.073
0.011
8.4
12
1 The numerical values expressed in these columns represent
milligrams per liter (mg/1).
6.3 EXPOSURE ASSESfiMRMT INFORMATION
Two time frames are considered in the baseline risk (or no-
action) assessment: (1) the current risk, or the risk from the
site as it exists today, and (2) the future risk from the site
assuming no remedial actions are taken. The current risk
assessment examines the risks for all pathways which analytical
data indicates that constituents have already reached a point of
exposure. As noted above, this* exposure assessment is for
ground-water contamination only.
The only known wells currently located between the site and the
probable ground-water discharge point (Bayou Texar) are monitor
wells and irrigation wells. For the purpose of assessing
potential current risks associated with ground-water exposure, it
will be assumed that adult and child residents occasionally drink
water (1 time per week at an amount equal to 10 percent of total
daily water consumption) pumped from irrigation wells. This
equates to approximately 200 milliliters (mL) per week for an
adult and 100 mL per week for a child. Thus, adults are assumed
to drink 0.2 L per week, 52 weeks per year for 30 years; children
are assumed to drink 0.1 L per week, 52 weeks per year for 6
years from irrigation wells.
To address the concern that citizens are currently exposed to
contaminated ground water through incidental ingestion or dermal
contact of swimming pool water supplied by an irrigation well, an
irrigation well/swimming pool survey was conducted. The survey
indicated that no one is currently exposed to the contaminated
ground water through a swimming pool water supplied by an
irrigation well. Therefore, the exposure pathway is incomplete
for the current scenario and considered a future exposure.
For the future swimming pathway, it will also be assumed that
adult and child residents are exposed via dermal contact and
incidental ingestion to irrigation water used to fill a swimming
pool. Exposure assumptions for an adult are as follows: (1) 70-
25
-------
kg body weight; (2) incidental ingestion of water at a rate of
0.05 L/hour; (3) body surface area of 18,150 cm2 exposed to
water; (4)~an exposure duration of 2.6 hours/day, and an exposure
frequency of 45 days per year for an exposure period of 30 years.
Exposure assumption for a child are (1) 16-kg body wight; (2)
incidental ingestion of water at a rate of 0.05 L/hr; (3) body
surface area of 7,190 square centimeters (cm2) exposed to water;
and (4) an exposure duration of 4 hours/day, an exposure
frequency of 88 days per year (two days per week in April and
October; three days a week in May and September; and four days
per week in June, July, and August) and an exposure period of 6
years.
No known active potable water supply wells lie in the path of the
contaminated ground-water plume. For the purpose of assessing
future public health risk associated with ground-water exposure,
it is assumed that a potable well is installed within the
constituent plume just downgradient of the site. Hypothetical
future residents (adult and child) are assumed to use the water
as a potable water supply. Therefore, daily ingestion exposure
at the standard ingestion rates of 2 L/day for adults and 1 L/day
for a child are assumed. Exposure periods remain at 30 years for
an adult and 6 years for a child. This is a conservative
assumption, as all drinking-water hookups are through the
Escambia County Utility Authority (ECUA).
The reasonable maximum exposure case is estimated using the
lesser of the 95 percent upper confidence limit on the mean of
the lognormal distribution or the maximum concentrations in
shallow monitor wells (representative of a well in the shallow
zone of the aquifer) and the lesser of the upper 95 percent
confidence limits on the mean of the lognormal distribution or
the maximum concentrations in deep monitor wells (representative
of a well in the deep zone of the aquifer). Constituent
concentrations are assumed to remain constant; thus, current and
future exposure point concentrations are the same.
6.4 TOXICITY ASSESSMENT INFORMATION
Cancer Slope factors (CSFs) have been developed by EPA's
Carcinogenic Assessment Group for estimating excess lifetime
cancer risks associated with exposure to potentially carcinogenic
contaminant(s) of concern. CSFs, which are expressed in units of
(mg/kg-day)'1, are multiplied by the estimated intake of a
potential carcinogen, in mg/kg/day, to provide an upper-bound
estimate of the excess lifetime cancer risk associated with
exposure at that intake level. The term "upper bound" reflects
the conservative estimate of the excess lifetime cancer risk
associated with exposure at that intake level. Use of this
approach makes underestimation of the actual cancer risk highly
unlikely. Slope factors are derived from the results of human
epidemiological studies or chronic animal bioassays to which
26
-------
animal-to-human extrapolation and uncertainty factors have been
applied (e.g., to account for the use of animal data to predict
effects on"humans).
Reference doses (RfDs) have been developed by EPA for indicating
the potential for adverse health effects from exposure to
contaminant(s) of concern exhibiting nonearcinogenic effects.
RfDs, which are expressed in units of mg/kg-day, are estimates of
lifetime daily exposure levels for humans, including sensitive
individuals. Estimated intakes of contaminant(s) of concern
ingested from contaminated drinking water can be compared to the
RfD. RfDs are derived from human epidemiological studies or
animal studies to which uncertainty factors have been applied
(e.g., to account for the use of animal data to predict effects
on humans). Table 6-1 provides Reference Doses (RfDs), Cancer
5lope Factors (CSFs) and EPA's Cancer Classification for
constituents of concern.
6*5 RISK CTTAPAPTORIZATION INFORMATION
For carcinogens, risks are estimated as the incremental
probability of an individual developing cancer over a life-time
as a result of exposure to the carcinogen. Excess life-time
cancer risk is calculated from the following equation:
Risk = GDI X SF
where: risk = a unitless probability (e.g., 2 X 10~s) of an
individual developing cancer;
GDI = chronic daily intake averaged over 70 years (mg/kg-day);
and SF = slope-factor, expressed as (mg/kg-day)*1
These risks are probabilities that are generally expressed in
scientific notation (e.g., IxlO'6 or IE'6). An excess lifetime
cancer risk of IxlO'6 indicates that, as a reasonable maximum
estimate, an individual has 1 in 1,000,000 additional chance of
developing cancer as a result of site-related exposure to a
carcinogen over a 70-year lifetime under the specific exposure
conditions at a site.
The potential for noncarcinogenic effects ife evaluated by
comparing an exposure level over a specified time period (e.g.,
lifetime) with a reference dose derived for a similar exposure
period. The ratio of exposure to toxicity is called a hazard
quotient (HQ). By adding the HQs for all contaminants of concern
that affect the same target organ (e.g., liver) within a medium
or across all media to which a given population may reasonably be
exposed, the Hazard Index (HI) can be generated.
27
-------
Table 6-1
Reference Doses (RfDs), Cancer Slope Factors (CSFs), and EPA Cancer Classification
for Constituents of Concern, Agrico Chemical Site
Constituent
RfD (mg/kg/day)
CSF (mg/kg/day)-1
Oral
Inhalation
Oral
Inhalation
EPA
Cancer
Class
Arsenic
Copper
Fluoride
Manganese
Mercury
Nickel
Nitrate
Benzene
Dibenzofuran
2,4-
Dinitrotoluene
2,6-
Dinitrotoluene
Fluorene
3x10-"
4xlO-2a
6xlO-2a
IxlO'1'
3x10-*'
2x10-"
2xl00b
NO
3xlO-20
NO
ND
4xlO"2a
ND
ND
ND
IxlO'4"
9xlO"5b
ND
ND
ND
NA
ND
ND
NA
2x10°'
NA
NA
NA
NA
NA
NA
0.03"
NA
0.7b
0.7b
NA
5xl0la
NA
NA
NA
NA
0.8a
NA
0.03a
NA
ND
ND
NA
A
D
D
D
D
A
D
A
D
B2
B2
D
28
-------
2 -Me thy 1-
naphthalene
Naphthalene
Radium 226
Radium 228
3xlO-2°
4xlO-2a
NA
NA
NA
NA
NA
NA
NA
NA
lxlO-1
-------
The HQ is calculated as follows:
Non-cancer-HQ = CDI/Rfd
where: GDI = Chronic Daily Intake
Rfd = reference does; and
CDI and RfD are expressed in the same units and represent the
same exposure period (i.e., chronic, subchronic, short-term).
Risk characterization for current and future direct contact with
ground water for resident adult and child is summarized in Tables
6-2 through 6-9. It should be noted that quantitative risk
estimates cannot be made for constituents which have insufficient
data to support toxicity values. Therefore, quantitative risk
estimates for aluminum, chloride, sodium, and sulfate are not
possible.
Excess lifetime cancer risk associated with current exposure to
ground water is 9 x 10"5 (shallow) and 4 x 10"5 (deep) for an
adult and 1 x 10~* and 3 x 10'5 (shallow and deep) for a child,
respectively. Arsenic contributes most significantly to the
excess cancer risk for the shallow and deep ground water.
The combined hazard index (HI) for intermittent ingestion and
swimming exposure for an adult is 2 (shallow) and 2 (deep); the
hazard index for a child was 10 (shallow) and 9 (deep). Fluoride
contributed most to the hazard index. However, a well and pool
survey (Woodward Clyde Memorandum August 1993) indicates this
exposure pathway is incomplete. Therefore, this evaluation
applies only to a future exposure scenario.
Hypothetical future risks from exposure to ground water are
summarized as follows. Excess lifetime cancer risks (ECLR) for
an adult ranges from 9 x 10"* (deep) to 4 x 10"3 (shallow) and 4 x
10"* (deep) to 2 x 10~3 (shallow) for a child. Arsenic and 2,4-
dinitrotoluene contribute most significantly to the ECLR. The
hazard index for an adult ranges from 40 (deep) to 60 (shallow);
the hazard index (HI)for a child is 100 for shallow and deep.
Fluoride contributes most significantly to the His.
6.6 UNCERTAINTIES IN THE RISK ASSESSMENT PROCESS
Uncertainty is inherent in the risk assessment process. Each of
the three basic building blocks for risk assessment (monitoring
data, exposure scenarios, and toxicity values) contribute
uncertainties. Environmental sampling itself introduces
uncertainty, largely because of the potential for uneven
distribution of constituents in environmental media. However,
the use of upper-bound assumptions, no attenuation, and the
conservatism built into the reference doses and cancer slope
factors are believed to result in an over-estimate of human
30
-------
Table 6-2
"RISK CHARACTERIZATION FOR CURRENT INGESTION
OF GROUND WATER FROM THE SHALLOW ZONE OF THE
SAND-AND-GRAVEL AQUIFER
Constituent
.Cancer Effects
Arsenic
Constituent
Non— Cflicer • Effects
Arsenic
Fluoride
Nitrates /Nitrites
EPC
(mg/L)
0.16
EPC .
(mg/L)
0.16
98
15
ARAR*
0.05
Total Risk =
ARAR*
0.05
4
10
Hazard Index
— Calculated Risk —
Adult Child
5x1 Q-5
5xlO'5
— • Hazard
Adult
.2
.7
.004
= 0.9
2xlO'5
2xlO'5
Quotient —
Child
.5
1
.008
2
All levels are set by the Safe Drinking Water Act as Maximum
Concentration Limits (MCLs). . ! ' '
31
-------
Table 6-3
-RISK CHARACTERIZATION FOR CURRENT INGESTION
OF GROUND WATER FROM THE DEEP ZONE OF THE
SAND-AND-GRAVEL AQUIFER
Constituent
EPC
(mg/L)
ARAR*
— Hazard Quotient —
Adult Child
Non-Cancer Effects
Fluoride 80
Nitrates/Nitrites 47
4 .5
10 .01
Hazard Index = 0.6
All levels are set by the Safe Drinking Water Act as Maximum
Concentration Limits (MCLs).
32
-------
Table 6-4
RISK CHARACTERIZATION FOR CURRENT INGESTION
OF GROUND WATER CONTAINING RADIONDCLIDES
Radionuclide
Surf icial Aouj
Radium-226
Radium-228
Deep Aquifer
Radium-226
Radium-228
Activity ARAR*
(pCi/L)
Lfer
0.8
1.2
8.4
12
5"
5*
Total Risk
5*
5*
Total Risk
— Calculated
Adult
3xlO'8
4xlO-8
» 7xlO'8
3xlO'7
. 4xlO'7
= 7xlO'7
Risk —
Child
3xlO'9
4xlO-9
7xlO'9
3xlO'8
4xlO'8
7xlO'8
'Both Radium-226 and Radium-228 have proposed MCLs of 20 pCi/1.
Total Radium has a current MCL of 5 pCi/1.
33
-------
Table 6-5
RISK CHARACTERIZATION FOR FUTURE INGESTION AND INHALATION
OF GROUND HATER FROM THE SHALLOW ZONE
OF THE SAND-AND-GRAVEL AQUIFER
Constituent
Cancer Effects
Arsenic
Benzene
2,4 Dinitro-
toluene
Constituent
Non-Cancer Effects
Arsenic
Fluoride
Manganese
Mercury
Nickel
Nitrates /Nitrites
EPC
(mg/L)
0.16
0.0015
0.025
EPC
(mg/L)
0.16
98
0.33
0.00059
0.035
15
ARAR*
0.05
0.005
n/a
Total Risk
ARAR*
0.05
4
0.05
0.002
0.1
10
— Calculated Risk —
Adult Child
3xlO'3
IxlO'6
2x10'*
= 4xlO-3
— Hazard
Adult
15
47
.009
.06
.05
.3
IxlO'3
5xlO-7
9xlO'5
2xlO-3
Quotient —
Child
33
100
.2
.1
.1
.ii_
Hazard Index =60
100
All levels are set by the Safe Drinking Water Act as Maximum
Concentration Limits (MCLs), which are regulated by the National
Secondary Drinking Water Regulations as Secondary MCLs.
34
-------
Table 6-6
RISK CHARACTERIZATION FOR FUTURE INGESTION AND INHALATION
OF GROUND HATER FROM THE DEEP ZONE
OF THE SAND-AND-GRAVEL AQUIFER
Arsenic
Benzene
2,4 Dinitro-
toluene
2,6 Dinitro-
toluene
Constituent
0.0056
0.0029
0.073
0.011
EPC
(mg/L)
0.05
0.005
n/a
n/a
Total Risk
ARAR*
IxlO"4
2xlO'6
6x10'*
9xlO'5
= 8xlO'4
— Hazard
Adult
5x10"?
9xlO'7
4x1 0-4
Quotient —
Child
Non-Cancer Effects
Arsenic
Copper
Fluoride
2-Methyl-
naphthalene
Mercury
Naphthalene
Nickel
Nitrates/Nitrites
0.0056
0.3 .
80
0.11
0.0006
1.5
0.032
47
0.05
1.3
4
n/a
0.002
n/a
0.1
10
.5
.2
38
.1
.06
1
.5
.8
Hazard Index =40
1
.5
83
.2
.1
2
.1
2_
90
* All levels are set by the Safe Drinking Water Act as Maximum
Concentration Limits (MCLs).
35
-------
Table 6-7
RISK CHARACTERIZATION FOR FUTURE DERMAL CONTACT WITH AND
INGESTION OF GROUND HATER FROM THE SHALLOW ZONE
OF THE SAND-AND-GRAVEL AQUIFER WHILE SWIMMING
Constituent
EPC ARAR*
(mg/L)
— Hazard Quotient —
Adult Child
Non-Cancer Effects
Arsenic 0.16
Fluoride 98
Nitrates/Nitrites 15
0.05
4
10
.2
.6
.003
Hazard Index = 0.8
2
6
.03
8
All levels are set by the Safe Drinking Water Act as Maximum
Concentration Limits (MCLs).
36
-------
Table 6-8
RISK CHARACTERIZATION FOR FUTURE DERMAL CONTACT WITH
AND INGBSTION OF GROUND WATER FROM THE DEEP ZONE
OF THE SAND-AND-GRAVEL AQUIFER WHILE SWIMMING
Constituent
EPC ARAR* — Hazard Quotient —
(mg/L) Adult Child
Non-Cancer Effects
Arsenic 0.0056
Dibenzofuran 0.036
Fluorene 0.024
Fluoride 80
Naphthalene 1.5
Nitrates/Nitrites 47
0.05
n/a
n/a
4
n/a
10
Hazard Index = 1
.007
.03
.02
.5
.4
.07
.2
.1
5
2
.1
8
* All levels are set by the Safe Drinking Water Act as Maximum
Concentration Limits (MCLs).
n/a Not applicable
37
-------
Table 6-9
RISK CHARACTERIZATION FOR FUTURE INGESTION
OF GROUND WATER CONTAINING RADIONDCLIDES
Radionuclide Activity ARAR*
(pCi/L)
— Calculated Risk —
Adult Child
Shallow
Radium-226
Radium-228
Deep
Radium-226
Radium-228
0.8
1.2
8.4
12
5 2xlO'7
5 3xlO'7
Total Risk = 5xlO'7
5 2xlO'6
5 3xlO-6
2xlO-6
2xlO-6
4xlO-6
2xlO'5
3xlO'5
Total Risk = 5xlO'6
4xlO-5
" Both Radium-226 and Radium-228 have proposed MCLs of 20 pCi/1.
Total Radium has a current MCL of 5 pCi/1.
38
-------
health risk. Therefore, actual risk may be lower than the
estimates presented here but are unlikely to be greater.
6.7 ENVIRONMENTAL RISK
Potentially exposed populations in the environment include: (1)
those at the site, (2) threatened or endangered species in the
area if any, and (3) the bayou communities. The focus of this
section is the environmental risk to the Bayou Texar.
Specifically a study (Bavou Texar Study, and the Bayou Texar
Study Addendum Report. Entrix. May -1993 ^ was conducted to assess
the extent of (1) the contaminated ground-water plume associated
with the site and (2) potential impacts to the Bayou Texar. A
detailed discussion of the results of that study is found in
Section 5.5.2, Ground-water Investigation. The results of the
Bayou Texar study indicate that ground water originating from the
Agrico site is discharging to the bayou. The discharge zone
appears to be well-defined and limited in areal extent.
A review of the Agrico ground-water plume discharge to the bayou
and its potential impact on biota has been completed by the EPA's
Ground Water Technical Support Unit (GWTSU). The GWTSU dilution
calculations indicate that the fluoride concentrations in the
ground water would have to be 4050 mg/L or greater in order to
exceed the State of Florida's surface water standard of 5 mg/L.
The highest level of fluoride found in the ground water is 98
mg/L. The above dilution calculations indicate that after
dilution, the level of fluoride entering the bayou is less than 1
mg/L. Therefore, based on the dilution calculations, the
discharge of the ground water plume into Bayou Texar will not
result in impacts to fish or wildlife. In addition, the OU1 ROD
provides for the treatment of the source (on-site contaminated
soils and sludges), which will further mitigate the potential
effects of the Agrico ground-water plume on the Bayou Texar.
Actual or threatened releases of hazardous substances from this
site, if not addressed in implementing the response action
selected in this ROD, may present an Imminent and substantial
endangerment of public health, welfare or the environment.
7.0 DESCRIPTION OF ALTERNATIVES
7.1 pmreniAL ACTION OBJECTIVES
A number of contaminates evaluated in the BRA and the FS are'not
assigned remediation goals; because the contaminant was present,
but in very low concentrations or isolated areas. Also, those
chemicals found at levels that did not pose a health risk are not
assigned remediation goals.
Soil cleanup goals were developed for the site soils and are
39
-------
required for direct contact, ingestion and inhalation of dust
(risk-based). Soil cleanup goals are also necessary which are
protective T)f ground water (leachability-based), for both organic
and inorganic constituents. In addition, risk based remediation
goals were determined for several exposure scenarios. A variety
of methods were used to develop remedial goals for soils and are
discussed in the OU1 ROD (source control).
The ground-water remedial action objectives for protection of
public health and the environment at the Agrico Chemical Site
are:
o Prevent continued degradation of the ground water from
onsite sources;
o Prevent or minimize degradation of the ground-water
resource due to effects associated with the selected
remedy such as the spreading of offsite plumes
including the organics plume emanating from the
Escambia Treating Company (ETC) site and salt water
intrusion;
o- Prevent or minimize future exposure to contaminated
ground water that would result in unacceptable risk.
o Prevent or minimize future impacts to surface water due
to discharge of contaminated ground water to Bayou
Texar.
The following table represents ground-water cleanup goals based
on federal or state primary and secondary drinking water
standards. This list of chemicals includes all chemicals with
unacceptable risks for the current risk scenario. Because of the
accessibility of public water supply in this area, it is unlikely
that residents will be exposed as envisioned in the future risk
scenario. Therefore, the future risk scenario from the baseline
risk assessment is not considered in developing these cleanup
levels.
Remediation
Chemical Levels (mq/L) (ppm)
Fluoride 4 mg/L (ppm)*
Arsenic 0.05 mg/L (ppm)
Chloride" 250 mg/L (ppm)
Sulfate** 250 mg/L (ppm)
Nitrate/nitrite 10 mg/L (ppm)
Radionuclides
Radium-226***
Radium-228*** 5 pCi/L
(Radium-226, 228 combined)
40
-------
* The MCL of 4 ppm for fluoride is the cleanup level for ground
water. The. Florida secondary standard of 2 ppm contained in
Section 17-550.320, F.A.C. will apply at nearby municipal supply
wells as specified in Alternative 10 (contingency remedy).
" Chloride and sulfate were not included in the baseline risk
assessment because no toxicity values exist. The remedial goals
presented for chloride and sulfates are the Florida ARARs.
*"" The proposed MCL for Radium-226 and Radium-228 is 20 pCi/1 for
each.
7.2 ARARs
Under Section 121(d) of CERCLA, a remedial action must attain
legally applicable or relevant and appropriate environmental
standards (ARARs), unless a waiver of such a standard is
applicable. ARARs fall into three categories: contaminant-
specific; location specific; and action-specific. Some rules
which do not specifically apply to a remedial action may provide
guidance because of their subject matter. These rules are called
to-be-considereds (TBCs). There are no action specific ARARs or
TBCs applicable to the Agrico Chemical site.
The Chemical-specific ARARs for ground-water constituents at the
Agrico Chemical site are derived from the Federal and State
primary drinking water standards (MCLs). Specifically, chemical-
specific ARARs for the ground-water treatment for the site
include:
o National Primary Drinking Water Regulations, 40 CFR Part 141
o Florida Primary Drinking Water Standards - Maximum
Contaminant Levels, Section 17-550*310, F.A.C.
o Florida Surface Water Quality Standards, Sections 17-
302.400, 17-302.500, 17^302.510, 17-302.530, and 17-302.560,
F.A.C.
o Florida Secondary Drinking Water Standards for fluoride,
chloride, and sulfate, Section 17-550.320, F.A.C.
Location specific ARARs that are applicable to the site include:
o Statement of Procedures on Wetlands Protection, Appendix A
to 40 C.F.R. Part 6
o Florida Ahtidegradation Policy for Surface Water Quality,
Section 17-302.300, F.A.C.
41
-------
7.3 DEVELOPMENT AND SCREENING OF ALTERNATIVES
7.3.1 PROCESS
As a part of the process, the FS preliminarily evaluates the
number of different technologies for remedial action. The
technologies are generally evaluated on the basis of their
effectiveness, implementability and cost in relation to the
remedial action goals for the site. After the screening, six
major alternatives were determined to be worth developing into
detailed alternatives for evaluation as the final cleanup plan.
For ease of cross reference with the FS, this ROD has maintained
the numbering system used in the FS. The retained alternatives
are as follows:
Alternative 1 - No Action;
Alternative 2 - Limited Action - comprehensive ground-
water monitoring, detailed irrigation
well survey, advisory board for contaminant
tracking, prevention of exposure through
voluntary plug/abandon of irrigation wells;
Alternative 5 - Ground-water extraction and treatment with
evaporation pond and reinjection to aquifer;
Alternative 6 - Modified ground-water extraction/treatment
with evaporation pond and. reinjection;
Alternative 7 - Modified ground-water extraction/treatment
with pond and discharge to water supply;
Alternative 8 - Modified ground-water extraction/treatment
with pond and discharge to surface water;
Alternative 10 - Well replacement or wellhead treatment.
7.3.2 ALTERNATIVE 1 - No Action
The National Contingency Plan (NCP) requires the development of a
no action alternative as a basis for comparison with the other
alternatives. The No Action alternative is presented as a
baseline case. The No Action alternative would include no
ground-water monitoring, no irrigation well survey, no
contaminant tracking, and no attempt to plug and abandon
contaminated irrigation wells. Clearly, this alternative does
not meet the remedial action objectives for preventing dermal
contact, ingestion, or mitigation of ground-water contamination.
Since construction would not be involved in this alternative, the
implementability concerns, engineering, equipment and materials,
42
-------
health and safety, and schedule are not applicable. Because no
ground-water monitoring would be conducted under this
alternative^ there would be no cost associated with quarterly
sampling and analyses. Therefore, there is no estimated present
worth cost; nor is there capital cost included with this No
Action alternative.
7.3.3 ALTERNATIVE 2 - Limited Action
This remedial alternative involves action aimed at limiting
exposure while natural attenuation of the site ground-water
contamination occurs. The remedial alternative primarily
consists of the following: (1) ground-water and Bayou Texar
surface water monitoring, to include installation of 2 additional
monitoring wells adjacent to the bayou; (2) a door-to-dobr
irrigation well survey; (3) institutional controls to include on-
site deed restrictions, groundwater use restrictions, and a
request that private land owners allow the plugging and
abandoning of impacted irrigation wells; and (4) an advisory
program. These activities include restrictions that will prevent
ground-water usage and contact in the vicinity of the Agrico
site. In addition, this alternative utilizes a comprehensive
ground-water monitoring plan to periodically evaluate the
hydrogeologic conditions and quality of the ground.water in the
Sahd-And-Gravel aquifer underlying the site.
The time predicted for the natural attenuation envisioned for
this remedy is 70 years duration. The following hydrogeologic
procedure was conducted to determine the length of time necessary
for the aquifer to undergo natural remediation with respect to
the site ground-water contamination. To evaluate the
effectiveness of various ground-water remediation strategies, a
comprehensive, detailed ground-water flow and solute-transport
modeling was conducted for the site.(Summary of Results -
Analysis of Ground Water Flow and Solute Transport at the Aorico
Plant Site, Pensacola. Florida, McDonald Morrissey Associates,
Inc., April 1992) The modeling yielded information on the
movement of dissolved chemical constituents in ground water and
predicted the fate of contaminants emanating from the site.
The modeling indicates that under existing flow conditions with
no active remediation of ground water, natural attenuation of the
site ground-water contamination will occur within 70 years.
Specifically, the modeling indicates that fluoride levels decline
to 4 mg/L after 70 years. The MCL for fluoride is 4 mg/L. The
modeling assumes no further contaminant loading (treatment of the
source) in its calculations. Utilizing a pump and treat
alternative, the modeling predicts similar levels for fluoride (4
mg/L) after 40 years.
The selection of fluoride over other site chemicals for modeling
purposes follows the rationale that 1) fluoride concentrations
43 c- ' -
-------
are the highest found in the aquifer relative to the federal
standard or-MCL for that chemical; 2) fluoride movement through
the aquifer demonstrates the worst case scenario for fate and
transport behavior of a contaminant emanating from the site.
This remedial alternative envisions a cooperative effort between
the parties conducting the remediation and the local, state, and
federal regulatory agencies to ensure that no one is exposed to
the contaminated ground water associated with the site. Although
potable water is currently supplied by the city, a comprehensive
door-to-door survey will be conducted to verify that previous
well surveys are accurate in the assumption that no one is
drinking water from the ground water within the Agrico
contaminant plume.
With respect to irrigation wells, proposed and "in progress"
irrigation wells will be denied permitting by the Northwest
Florida Management District (NWFWMD) pursuant to FAC 17-524. In
addition, private landowners will be approached for access to
plug and abandon impacted private wells.
In addition, a comprehensive ground-water and bayou surface-water
monitoring program will be implemented and the results will be
incorporated into an advisory program conducted by the NWFWMD for
3-D modeling/contaminant tracking. The monitoring program will
also include installation of an additional two monitoring wells
adjacent to Bayou Texar. The purpose of this advisory program
will be to periodically evaluate the hydrogeologic conditions and
quality of the ground water in the Sand-And^-Gravel aquifer
underlying the site. In addition, the ground-water and surface-
water monitoring specified as part of this remedial action will
verify that contaminant levels are declining through natural
attenuation and flushing as predicted by the above ground-water
modeling.
This remedy does not include a pump and treat technology.
However, this limited action avoids potentially adverse impacts
associated with the ground-water extraction and treatment
alternatives. The negative impacts of a pump and treat cleanup
method would include: 1) spreading of off-site plumes of
contamination such as the Escambia Treating Company (ETC) site
organics plume; 2) salt water intrusion; and 3) alteration of
ground-water flow patterns in the Sand-And-Gravel aquifer. These
adverse affects of active pumping and treating will be discussed
further in remedial alternatives 5 through 10.
Implementation of Alternative 2 in conjunction with source
treatment and containment will protect human health and the
environment. The use of source treatment and containment
currently being implemented, will eliminate the source of ground-
water impacts. In addition, treatment of the contaminated source
will prevent or minimize further impacts to the surface water of
44
-------
the Bayou Texar due to discharge of contaminated ground water to
that water body. The Florida surface water standard for fluoride
is 5 mg/L (ppm). However, the remedial goal specified for this
remedy will be the more stringent federal requirement of 4 ppm.
The capital cost for Alternative 2 is estimated to be $488,000.
The total present worth value of the alternative is $1,661,000.
Compliance with applicable ARARs will be attained over time
through natural attenuation (adsorption, precipitation, dilution)
and flushing of the aquifer. A 70 year period is estimated to
reach applicable federal and state ARARs for the site ground-
water contaminants. This natural attenuation estimate of 70
years is based on a comprehensive, detailed ground-water flow and
solute-transport modeling conducted (Summary of Results -
Analysis of Ground Water Flow and Solute Transport at the Acrrico
Plant Site. Pensacola. Florida. McDonald Morrissey Associates,
Inc., April 1992)i
7.3.4 ALTERNATIVE 5 - Ground-water Extraction and Treatment: with
Evaporation Pond and Reinjection to Aquifer
Alternative 5 consists of the following institutional controls
and treatment technologies: (1) institutional controls to include
on-site deed restrictions and groundwater use restrictions; (2)
Ground-water monitoring; (3) 1.3 MGD Ground-water extraction
wells; (4) Treatment to included Carbon Adsorption, Filtration,
Reverse Osmosis (RO) with RO reject evaporation pond; (5) 3.0 MGD
Reinjection to Sand-And-Gravel Aquifer.
This remedy will actively achieve the remedial goals established
for the site constituents by treatment of the contaminated
ground-water. However, the potential exists for widespread and
uncontrolled aquifer degradation due to the alteration of ground-
water flow resulting in the spreading of organic contaminants
into areas previously not impacted by these constituents, private
irrigation wells, and potentially, ECUA potable water well
locations (both present and future). The potential also exists
for creating a salt water intrusion problem in the area.
Reinjection wells along the bayou and upgradient of the site are
proposed in this alternative to minimize the potential for
migration of other contaminant plumes into the Agrico ground-
water extraction and to minimize salt water intrusion. However,
the hydraulics of ground-water flow in the aquifer around the
bayou are not well understood, and the potential 'impact of
reinjection wells on contaminate plumes adjacent to the Agrico
site are unknown. In addition, a portion of the plume not
intercepted by the reinjection wells hydraulic barrier will
continue discharging to Bayou Texar.
Alternative 5 would use the same controls as Alternative 2 to
45
-------
limit access to contaminated ground water, but would require 5
extraction wells and 14 off-site reinjection wells to pump and
treat 1.3 million gallons per day (MGD) of contaminated ground
water. The extraction wells would be positioned within and
downgradient of the Agrico contaminant plume to prevent migration
of other (not site related) contaminants into nearby irrigation
wells. Therefore, the positioning of the extraction wells would
optimize cleanup of the contaminant plume. In addition, the
placement of 6 reinjection wells along Bayou Texar might minimize
the potential for salt water intrusion. This option would also
use ground-water transmission lines to route ground water to a
treatment plant on site. Ground-water treatment would be done by
granulated activated carbon (GAC) adsorption, a RO unit, and a
spray evaporation pond for disposal of the RO reject stream
(waste from process). A residual sludge collected from the
evaporation pond would go to an off-site landfill.
Alternative 5 will comply with ARARs identified for the site
constituents. The extraction and treatment will capture the
Agrico ground-water plume and reduce the contamination prior to
discharge over an estimated 40 year operational period (McDonald
Morrissey ground-water modeling, April 1992). However, due to
uncertainties related to the hydraulic behavior of the aquifer
and effectiveness of the reinjection wells, actual time to meet
ARARs may be substantially longer. In a letter to the
responsible parties, the ECUA's Executive Director had the
following comment in reference to extracting ground water,
treating and reinjecting into the aquifer."....if reinjection
occurs upstream of the Agrico site, the reinjected ground water
will cause further dilution of the contaminates in the plume and
thus prolong the remedial operation and require even larger water
withdrawals. This likewise is undesirable and
counterproductive".
The estimated capital costs of Alternative 5 is $9,540,000. The
estimated O&M cost is $1,627,000. The present worth value of
this alternative is $41,130,000.
7.3.5 ALTERNATIVE 6-1.0 MGD Ground-water Extraction. Carbon
Adsorption, Filtration, Reverse Osmosis with Evaporation
Pond, Reinjection to Sand-And-Gravel Aquifer
Alternative 6 is similar to Alternatives 2 and 5, in that the
same controls are used, but includes a modified ground-water
extraction and reinjection system for treating contaminated
ground water. This alternative would use only three extraction
wells to pump 1.0 MGD and three reinjection wells, instead of
fourteen wells called for in Alternative 5. The ground-water
extraction wells would be placed downgradient to the maximum
extent possible of the contaminant plume, but not within the
contaminant plume as in Alternative 5. The reinjection wells
would be located on site along the northern boundary.
46
-------
This alternative implements ground water extraction and treatment
technologies which will reduce the threat from the Agrico
contaminants in ground water. The Agrico contaminants will be
reduced over the projected 40 year operational period. However,
due to migration of uncontrolled off-site unrelated plumes
including the ETC site organic plume, which will be induced into
the Agrico extraction system, contamination may still persist in
areas of the aquifer not previously affected after the Agrico
site constituents are removed. .
Residual waste materials will be produced from the treatment
technology proposed for this alternative. The characterization
and ultimate disposal of this material is uncertain. The
treatment of these unknown contaminants may result in the
production of hazardous residual materials. The ultimate
treatment and disposal of these potential hazardous materials
cannot currently be characterized.
This alternative will comply with ARARs identified for the Agrico
site constituents. The projected life of the pump and treat
system is 40 years duration. However, the potential for
effective treatment and removal of other contaminants in plumes
around the Agrico site are unknown. In a November 22, 1993
letter to EPA, the NWFWMD expresses its concern about the
effectiveness of a pump and treat system in the following manner.
Management of ground water resources in Escambia County is a
sensitive, problematic and complicated task; especially
within the additional restrictions imposed by the saltwater
fronts of the Pensacola and Perdido bay systems which bound
the aquifer. . The irony of the aquifer system in the area is
that the high recharge characteristics which allow for a
prolific resource also limit the use of this resource, due
to the relative ease of ground water contamination. Until
an adequate aquifer/wellhead protection program and a civic
education program is developed for the area, further
instances of contamination should be anticipated.
Implementation of plume restoration without some mechanism
to prevent future contamination in the area may be
nonproductive. The potential risks imposed by pump and
treat technology appear to outweigh the benefit of
remediation efforts via this method.
The estimated capital costs of Alternative 6 is $7,402,000. The
estimated total O&M cost is $807, 000. The present worth value of
this alternative is $24,660,000.
7.3.6 Alternative .7 — 1.0 MGD Ground-water Extract ion.
Adsorption, Filtration, Reverse Osmosis with Evaoration
to Potable Water
This alternative is similar to Alternative 5, but considers a
47
-------
modified ground-water extraction system. This Alternative
includes three ground-water extraction wells, instead of the five
utilized in Alternative 5, and no reinjection wells. The ground-
water extraction wells are located downgradient to the maximum
extent possible of the identified contaminant plume. The treated
ground water is discharged to the local potable water supply
system.
This remedy will actively achieve the remedial goals established
for the site constituents by treatment of the contaminated
ground-water. However, the potential exists for widespread and
uncontrolled aquifer degradation due to the alteration of ground-
water flow resulting in the inducement of organic contaminants
into the Agrico plume area private irrigation wells. The ground-
water extraction for the Agrico site plume will induce off-site
unrelated plumes including the ETC site organic plume into the
Agrico site extraction system, impacting areas previously not
contaminated with organics.
This alternative would require the same controls as Alternatives
2 through .6 and would be similar to Alternative 6 except for the
discharge method. Recovery pipelines would be routed to the on-
site treatment plant. This alternative would also include three
ground-water extraction wells pumping 1.0 MGD, but no reinjection
wells. The ground-water extraction wells would be located
similar to those in Alternative 6, but the treated ground water
would be discharged to the local potable water supply system. A
portion of the Agrico plume not intercepted by the extraction
wells would continue discharging to Bayou Texar for a period of
up to ten years. However, dilution calculations results show
that the fluoride concentrations in the Bayou Texar surface
waters from that discharge should not exceed 1 mg/L. Therefore,
the fluoride levels expected in the bayou are below the Florida
surface water quality standard for fluoride of 5 mg/L.
This alternative will comply with ARARs identified for the Agrico
site constituents. The extraction and treatment system will
capture the identified Agrico ground-water plume and reduce the
contamination prior to discharge over an estimated 40 year
operating period. Disposal of treated water and treatment
residuals will be in compliance with applicable regulations. The
potential for effective treatment and removal of other
contaminants in plumes around the Agrico site are unknown.
The estimated capital costs of Alternative 7 is $7,268,000. The
estimated total annual O&M cost is $807,000. The present worth
value of this alternative is $24,573,000.
48
-------
7.3.7 ALTERNATIVE 8 - 1.0 MGD Ground-water Extraction. Carbon
Adsorption. Filtration, Reverse Osmosis with Evaporation
Pond", pisgha-rge to Bayou Texar
Alternative 8 consists of the following controls and treatment
technologies: On-site deed restriction, ground-water use
restrictions, ground-water monitoring, ground-water extraction
wells, carbon adsorption, filtration, Reverse Osmosis with RO
reject evaporation pond, off-site disposal of RO reject sludge
from evaporation pond, and discharge of treated water to Bayou
Texar. Ground-water use restrictions include: proposed and "in
progress" irrigation wells will be denied permitting; plugging
and abandoning existing private irrigation wells will be strongly
encouraged, if shown to be impacted and positioned within the
contaminated area.
This remedy will actively achieve the remedial goals established
for the site constituents by treatment of the contaminated
ground-water. However, potential exists for widespread and
uncontrolled aquifer degradation due to the alteration of ground-
water flow resulting in the inducement of organic contaminants
into the Agrico plume area and private irrigation wells.
This alternative is similar to Alternative 5, but considers a
modified ground-water extraction system and no reinjection wells.
The ground-water extraction wells are located downgradient to the
maximum extent possible of the contaminant plume. The treated
ground water is discharged to the Bayou Texar. An NPDES permit
would be required to implement this alternative and may entail
permitting delays. Bioassay toxicity testing may also be
required as part of the permit conditions which pose problems
after the system is constructed. In addition, recovery pipelines
will be routed to the treatment plant located at the Agrico site.
The recovery piping will be located along:right-of-ways;" to the
maximum extent possible, through the developed
residential/industrial sector of Pensacola.
This alternative will comply with ARARs identified for the Agrico
site constituents. The extraction and treatment system will
capture the identified Agrico ground-water plume and reduce the
contamination prior to discharge over an estimated 40 year
operating period. However, the potential for effective treatment
and removal of other contaminants in plumes around the Agrico
site is unknown.
The estimated capital cost of Alternative 8 is $7,532,000. The
estimated total annual O&M cost is $807,000. The present worth
value of this alternative is $24,713,000.
49
-------
7.3.8 Alternative 10 - Future Well Replacement or Wellhead
Treatment, (if existing potable water supply wells are
Impacted)
Alternative 10 consists of the following institutional controls
and treatment technologies: On-site deed restrictions, ground-
water use restrictions, ground-water monitoring, water supply
wellhead treatment or water supply well replacement. Wellhead
treatment includes filtration, prefiltration, Reverse Osmosis
with RO reject evaporation pond, and off-site disposal of RO
reject sludge from an evaporation pond. Another option under
this alternative is replacement of municipal water supply wells.
The determination of whether wellhead treatment or well
replacement (new water supply well installation) is appropriate
will be made when or in the event the water supply well(s) are
impacted by the Agrico plume. Specifically, if fluoride levels
detected in nearby ECUA water supply wells exceed Florida's
secondary drinking water standard of 2 mg/1, EPA will evaluate
whether implementation of this contingency remedy is appropriate.
This alternative does not create the potential for diversion of
other contaminant plumes around the Agrico site.
This remedy will actively achieve the remedial goals established
for the site constituents by treatment of the contaminated ground
water at existing water supply wells, if impacted by the Agrico
plume in the future. If the affected water supply well is
replaced, remedial goals established for the site will be
attained through natural attenuation, as described in
Alternative 2. However, based on current hydrogeologic
conditions, it is unlikely that nearby municipal water supply
wells will be impacted by the Agrico ground-water plume.
The local water authority, ECUA,-believes this is a viable
contingent solution to the Agrico site, as noted in its
correspondence of March 17, 1993.
If at some time in the future, the plume threatens any ECUA
wells, that those parties responsible for the cleanup of the
Agrico site, at that time, fund the construction and
operation of the necessary treatment facilities to cleanup
the water which is being pumped. The treated water would be
fed directly into ECUA's water distribution system.
Extraction without reinjection avoids a major upset of water
flow/direction in the aquifer. Under this approach
contaminate plume migration and dilution concerns would be
minimal. This would fully use the water pumped resulting in
no loss of a natural resource.
In summary, we believe that the only appropriate way to
handle groundwater clean-up is to only extract the
contaminated groundwater when it becomes an identified
50
-------
threat to the water quality of an existing or proposed well
and then fully utilize the cleaned water rather than to
waste a natural resource.
Alternative 10 is similar to Alternative 5; however, this
alternative considers the installation of a new water supply well
or treatment of ground water at an existing potable water supply
well that has been impacted by the Agrico plume. If fluoride
levels exceed 2 mg/L at the nearby municipal potable water supply
wells, implementation of the contingency remedy will be
evaluated. Essentially, EPA will determine if wellhead treatment
or water supply well replacement will be appropriate. Technical
considerations such as frequency of sampling and/or detection
frequency will be clearly specified in a subsequent Remedial
Design (RD) Report. ..'•..
This alternative will achieve substantial risk reduction by
restricting current and future ground-water use and by treating
water supply wells if they become impacted in the future .
The treated ground water will be discharged into the local
potable water supply system or new water supply wells will be
located outside the area of identified ground-water impact. The
ECUA has done extensive testing- of well sites and has installed
numerous wells to identify several potential water supply well
locations in the area.
This alternative, in conjunction with Alternative 2, will comply
with ARARs through natural attenuation and by treatment at the
wellhead.
The estimated capital cost of Alternative 10 is $4,061,000 for
wellhead treatment and $999,000 for wellhead replacement. The
estimated total annual O&M cost is $1,614,000 for wellhead
treatment . The present worth value of this alternative is
$21,039,000 for wellhead treatment and $999,000 for well
replacement. •
8.0 SUMMARY OF THK COMPARATIVE ANALYSIS OF ALTERNATIVES
This section of the ROD provides the basis for determining which
alternative provides the best balance with respect to the
statutory balancing criteria in Section 121 of CERCLA and in
Section 300.430 of the NCP. The major objective of the FS is to
develop, screen, and evaluate alternatives for the remediation of
the Agrico Chemical site.
8.1 CRITERIA FOR BVAHIATTMC PKMRQIAL ALTERNATIVES
The remedial alternatives selected from the screening process are
evaluated using the following nine evaluation criteria:
51
-------
Overall Protection of Human Health and the Environment - Assesses
degree to which alternative eliminates, reduces, or controls
health and environmental threats through treatment, engineering
methods, or institutional controls.
Compliance with Applicable or Relevant and Appropriate
Requirements - Assesses compliance with Federal/State
requirements.
Cost - Weighs the benefits of a remedy against the cost of
implementation.
Implementability - Refers to the technical feasibility and
administrative ease of a remedy.
Short-Tenm Effectiveness - Length of time for remedy to achieve
protection and potential impact of construction and
implementation of a remedy.
Long-Tenn Effectiveness - Degree to which a remedy can maintain
protection of health and environment once cleanup goals have been
met.
Reduction of Toxicity, Mobility, or Volume Through Treatment -
Refers to expected performance of the treatment technologies to
lessen harmful nature, movement or amount of contaminants.
State Acceptance - Consideration of State's opinion of the
preferred alternative.
Community Acceptance - Consideration of public comments on the
Proposed Plan.
The NCP categorizes the nine criteria into three groups:
• Threshold Criteria - overall protection of human health
and the environment and compliance with ARARs (or
invoking a waiver) are threshold criteria that must be
satisfied in order for an alternative to be eligible
for selection;
• Primary Balancing Criteria - long-term effectiveness
and permanence; reduction of toxicity, mobility, or
volume; short-term effectiveness; implementability, and
cost are primary balancing factors used to weigh major
trade-offs among alternative hazardous waste management
strategies; and
• Modifying Criteria - state and community acceptance are
modifying criteria that are formally taken into account
after public comment is received on the proposed plan
and incorporated in the ROD.
52
-------
The selected alternative must meet the threshold criteria to be
eligible for selection. The Primary Balancing Criteria are the
technical criteria upon which the detailed analysis is primarily
based. The final two criteria, known as Modifying Criteria,
assess the public's and the state agency's acceptance of the
alternative. Based on these final two criteria, EPA may modify
aspects of a specific alternative.
The following analysis is a summary of the evaluation of
alternatives for remediating the ground water related to the
Agrico Chemical Superfund Site under each of the criteria. A
comparison is made between each of the alternatives for
achievement of a specific criterion.
.Threshold Criteria
8.2 OVERAT.T. PROTECTION OF HDMAH HEALTH AND THK ENVIRONMENT
All alternatives except Alternative 1 are protective of human
health and the environment with respect to the Agrico
constituents. However, there are significant differences in
regards to overall protection of human health and the environment
between Alternatives 2 and 10 and the remaining pump and treat
alternatives . There exists the potential for widespread and
uncontrolled aquifer degradation due td the spreading of off -site
plumes from the implementation of Alternatives 5, 6, 7, and 8.
Alternative 2 deals with the plume through natural attenuation
and flushing while Alternatives 5, 6, 7 and 8 actively pump and
treat the plume. However, Alternatives 5, 6, 7 and 8 will also
rely on natural processes to remediate the portion of the plume
not captured in the extraction system, but allowed to discharge
to the bayou. The pumping action included in Alternatives 5, 6,
7 and 8 will have uncontrolled effects that could degrade
portions of the aquifer previously not impacted. Specifically,
the extraction of the Agrico plume will most likely draw into the
deep zone organic contaminants related to other sources that
reside in the shallow zone. Clearly, this would. have the
undesirable effect of introducing more organic contamination into
the main producing zone {deep zone). Alternative 10 will either
provide treatment to the impacted city water supply well or if
more practical, replace the well in a new location.
Institutional controls, through well permitting and recommended
testing procedures by NWFWMD and ECUA, are currently in place.
In addition, the FDEP is in the process of putting further .
constraints on potable well installation in accordance with
Chapter 17-524 F.A.C. Finally, private landowners will be
approached for access to plug and abandon impacted private wells.
Therefore, the routes to exposure will be minimized.
The health risk associated with the ingestion of swimming-pool
53
-------
water supplied by an irrigation well has been examined. A
subsequent ^irrigation well/swimming pool survey has indicated
that no swimming pool within the contaminated plume is supplied
by an irrigation well. Therefore, this potential exposure
pathway is incomplete. A comprehensive door-to-door well survey
will be conducted to verify the accuracy of previous well
surveys.
A study of Bayou Texar sediments and sediment porewater indicates
the Agrico plume discharges to the bayou with minimal potential
impact. Fluoride, the Agrico "signature" contaminant shown to
discharge to the bayou is expected to quickly disperse and not
result in exceedance of the water quality standard. The dilution
calculations conducted by the Agency's Groundwater Technical
Support Unit (GWTSU) indicate that the fluoride concentrations in
the Bayou Texar surface waters from that discharge should not «•
exceed 1 mg/L. To verify the results of the dilution
calculations, the selected remedy would include surface-water
sampling of the Bayou Texar. In addition, treatment of the
contaminated source pursuant to the OU1 ROD will prevent or
minimize further impacts to the surface water of the Bayou Texar
due to discharge of contaminated, ground water to that water body.
For these reasons, all alternatives are expected to be protective
of the bayou environment.
All alternatives will satisfy chemical-specific, location-
specific, and action-specific ARARs with the exception of the "No
Action" alternative. For Alternatives 5, 6, 7 and 8, the ARARs
are estimated to be met within approximately 40 years, while for
Alternatives 2 and 10, the ARARs are estimated to be met within
approximately 70 years. However, with Alternatives 5, 6, 7 and 8
compliance with ARARs for the potential diverted and induced
contaminants from unrelated off-site sources is unknown.
Primary Balancing Criteria
8.4 LONG-TERM EFFECTIVENESS AND PERMANENCE
Coupled with the source containment alternative (OD1 ROD)
currently being implemented, Alternatives 2 and 10 will result in
permanent remediation of the ground water and pose no continued
risk to the surrounding public or the environment upon completion
of the flushing of the Agrico ground-water contaminants.
Alternatives 5, 6, 7 and 8 will potentially have long term
negative impacts on the aquifer by contaminating areas previously
not impacted through the spreading of ground-water contamination
unrelated to the Agrico site. In addition, these Alternatives
will most likely (through the required 1.0 MGD pumping) introduce
54
-------
salt water intrusion from the nearby Bayou Texar. Finally, the
waste by products generated from the treatment processes included
in Alternatives 5, 6,1 and 8 would also pose potential long term
risks/ especially if characterized as hazardous. Removal of the
large volumes of ground-water required for Alternatives 5, 6 and
8, without utilizing it, would not be of beneficial use of this
natural resource.
8.5 REDUCTION OF TOXICITY. MOBILITY OR VOLUME THROUCT TREATMENT
The treatment and containment of on-site contaminated soils and
sludges, as envisioned in the OU1 ROD, will eliminate the source
of contamination to the ground water, thus reducing the toxicity,
mobility and volume of future contamination to the ground water.
The natural, flushing envisioned in Alternative 2 will reduce the
toxicity and volume of the present ground-water contamination.
However, Alternative 2 does not include treatment, nor does this
alternative reduce the mobility of site contaminants currently
residing in the aquifer. The toxicity, mobility and volume of
the Agrico plume will be reduced with Alternatives 5, 6,7, 8 and
10 through ground-water extraction, treatment and discharge.
However, ground-water modeling (Summary of Results - Analysis of
Ground Water Flow and Solute Transport at the Agrico Plant Site,
Pensacola. Florida. McDonald Morrissey Associates, Inc., April
1992) has indicated that ground-water extraction will take
approximately 40 years to attain ARARs. The active pumping
included in these alternatives may "smear" unrelated plumes into
the extraction system for the Agrico site. These contaminants
from multiple other sources will remain in the aquifer after the
Agrico site constituents have been remediated and may pose an
even greater risk. The toxicity and volume of off-site unrelated
plumes that will spread through the implementation of
Alternatives 5, 6, 7 and 8 are unknown.
Alternative 2 and Alternative 10 (if municipal supply wells are
replaced) will rely on natural flushing, attenuation and dilution
to reduce the toxicity and volume of the Agrico plume and attain
ARARs within approximately 70 years (McDonald Morrissey Ground-
water Modeling Report, April 1992). Alternatives 2 and 10, in
conjunction with treatment of the source (OU1), will reduce the
toxicity, mobility, and volume of future contamination to the
ground water. Alternatives 5, 6,7 and 8 will require management
and disposal of potentially large quantities of hazardous
sludge/mineralized brine generated as residuals of treatment.
Alternative 2 will not generate any waste residuals.
8.6 SHORT—TERM EFFECTIVENESS
Alternatives 2 and 10 propose to implement remedial measures at
the site that present the least potential for adverse impact to
the community or on-site workers. Alternatives 5, 6, 7 and 8
involve routing the contaminated ground water via a pressurized
55 ..
-------
involve routing the contaminated ground water via a pressurized
pipeline through a residential community to the treatment
facility. The heavy construction activities of Alternatives 5,
6, 7 and 8 will have the inherent risks of performing these
activities in a residential neighborhood. In addition,
Alternative 8 will also involve construction of a pressurized
discharge line from the treatment facility located on site to the
bayou through the residential community.
Alternative 5 will involve construction of additional extraction
wells and 14 reinjection wells. The use of reinjection wells as
part of Alternatives 5 and 6 may have serious detrimental impacts
due to the spreading of other ground-water contaminant plumes in
the immediate vicinity of the Agrico site. The local water
authority's serious concerns with reinjection of treated water
into the aquifer, as stated in the ECUA correspondence of March
17, 1993 (Appendix A), include potential detrimental impacts to
nearby ECUA water supply wells by altering the ground water flow
patterns, inducing contaminants to migrate into the well capture
zones and prolonging remediation. Based on these concerns, the
permit for reinjection wells may be difficult to attain.
The NWFWMD shares ECUA's serious concerns with the active pump
and treat alternatives, stating in its November 22, 1993 letter
to EPA (Appendix A) .
The potential risks imposed by pump and treat technology
appear to outweigh the benefit of remediation efforts via
this method (pump & treat), and ...with the additional
restriction imposed by the saltwater fronts of the Pensacola
and Perdido bay systems which bound the aquifer. ,
.... Implementation of plume restoration without some
mechanism to prevent future contamination in the area may be
nonproductive.
It is anticipated that Alternative 2 can be implemented within
the shortest period of time (8 months to 1 year). Alternatives
5, 6, 7 and 8 will take an anticipated three years each to
implement after the selection process . Alternative 10 will be
implemented only if required in the future .
8.7
Alternative 2 would be the easiest to implement, since it would
not involve any construction, obtaining of permits, and obtaining
right-of-ways as included in the other alternatives. Alternative
10 could be easily implemented. This alternative would require a
standard design for a treatment system at the wellhead.
Alternatives 5, 6, 7 and 8 would all involve extensive predesign
activities to optimize the location of the extraction wells. For
Alternatives 5 and 6, the location, number, pumping rates and
feasibility of reinjection will also require extensive, detailed
56
-------
predesign and design activities.
80 ~
• O
The comparative present worth costs of the six alternatives are
as follows :
Alternative 2 $1,661,000
Alternative 5 $40,451,000
Alternative 6 $23,974,000
Alternative 7 $23,894,000
Alternative 8 $24,033,000
Alternative 10 $21,039,000 (wellhead treatment)
$999,000 (wellhead replacement)
Modifying Criteria
8.8 STATE ACCEPTANCE
The State of Florida, as represented by the Florida Department of
Environmental Protection ( FDEP ) , has been the support agency
during the Remedial Investigation and Feasibility Study process
for the Agrico Chemical Site. In accordance with 40 CFR 300.430,
FDEP, as the support agency, has provided input during this
process. Based upon comments received from FDEP, it is .expected
that concurrence will be forthcoming; however, a formal letter of
concurrence has not yet been received.
8.9 COMMUNITY ACCEPTANCE
The community had no major concerns about the remedy EPA proposed
in this ROD for OU2. However, some citizens expressed concern
over their proximity to the site and their potential exposure to
contaminated soil . The citizen concerns are referenced in the
previous Section 3.0 Highlights of Community Participation. In
addition, the concerns of the community are discussed in detail
in the Responsiveness Summary, which is Appendix B of this OU2
ROD.
9.0
Based on consideration of the requirements of CERCLA, the NCP,
the detailed analysis of alternatives and public and state
comments, EPA has. selected Alternative 2 as its remedy for
addressing contaminated ground-water related to the Agrico
Chemical Superfund Site. While the other (pump and treat)
alternatives might reduce the toxicity, mobility and volume of
site contaminants, .other contamination from off -site sources
unrelated to the Agrico site, will likely spread, further degrade
the aquifer, and pose an even greater risk. Because the nature .
and extent of other adjacent or nearby ground-water contaminants
is not known, pumping and treating this complex ground-water
57
-------
system at this time is not protective.
This alternative is protective, cost-effective, and will attain
all Federal and State requirements. At the completion of the
above remedy, the risk associated with this site has been
calculated at lxl(T6 which is within EPA's acceptable risk range
of IxlO'4 to IxlO'6. EPA has determined that this risk range is
protective of human health and the environment. Because this
remedy will result in hazardous substances remaining in the
aquifer above health-based levels for a time period exceeding
five years, a review will be conducted within five years after
commencement of the remedial action to ensure that the remedy
continues to provide adequate protection of human health and the
environment. In addition, subsequent five-year reviews will be
conducted as is appropriate due to the extended period of time
required for natural attenuation of the Agrico ground-water
contamination.
The total present worth cost of the selected remedy, Alternative
2, is estimated to be $1,661,000.
This remedial alternative involves action aimed at limiting
exposure while natural attenuation of the site ground-water
contamination occurs. The remedial alternative primarily
consists of the following: (1) ground-water and Bayou Texar
surface-water monitoring, to include installation of 2 additional
monitoring wells adjacent to the bayou and bayou surface-water
sampling; (2) a door-to-door irrigation well survey; (3)
institutional controls to include on-site deed restrictions,
groundwater use restrictions, and a request that private land
owners allow the plugging and abandoning of impacted irrigation
wells; and (4) an advisory program. These activities include
restrictions that will limit ground-water usage and contact in
the vicinity of the Agrico site. In addition, this alternative
utilizes a comprehensive ground-water monitoring plan to
periodically evaluate the hydrogeologic conditions and quality of
the ground water in the Sand-And-Gravel aquifer underlying the
site. However, the installation of the additional monitoring
wells may occur during the Operations and Maintenance (0 & M)
phase of the project.
The time predicted for the natural attenuation envisioned for
this remedy is 70 years duration. The following hydrogeologic
procedure was conducted to determine the length of time necessary
for the aquifer to undergo natural remediation with respect to
the site ground-water contamination and to evaluate the
effectiveness of various ground-water remediation strategies. A
comprehensive, detailed ground-water flow and solute-transport
modeling was conducted for the site (Summary of Results -
Analysis of Ground Water Flow and Solute Transport at the Agrico
Plant Site. Pensacola, Florida. McDonald Morrissey Associates,
Inc., April 1992). The modeling yielded information on the
58
-------
movement of dissolved chemical constituents in ground water and
predicted the fate of contaminants emanating from the site.
The modeling indicates that under existing flow conditions with
no active remediation of ground water, natural attenuation of the
site ground-water contamination will occur within 70 years.
Specifically, the modeling indicates that fluoride in the aquifer
declines to 4 mg/L after 70 years. The MCL for fluoride is 4
mg/L. The modeling assumes no further contaminant loading
(treatment of the source) in its calculations. Utilizing a pump
and treat alternative, the modeling predicts similar levels of
fluoride (4 mg/L) in the aquifer after 40 years.
The selection of fluoride over other site chemicals for modeling
purposes follows the rationale that 1) fluoride concentrations
are the highest found in the aquifer relative to the federal
standard or MCL for that chemical; 2) fluoride movement through
the aquifer demonstrates the worst case scenario for fate and
transport behavior of a contaminant emanating from the site.
This remedial alternative envisions a cooperative effort between
the parties conducting the remediation and the local, state, and
federal regulatory agencies to ensure that no one is exposed to
the contaminated ground water associated with the site. Although
potable water is currently supplied by the city, a comprehensive
door-to-door survey will be conducted to verify that previous
well surveys.are accurate in the assumption that no one is
drinking water from the ground water within the Agrico
contaminant plume.
With respect to irrigation wells, proposed and "in progress"
irrigation wells will be denied permitting by the Northwest
Florida Management District (NWFWMD) pursuant to FAC 17-524. In
addition, private landowners will be-approached for access to
plug and abandon impacted private wells.
A comprehensive ground-water and bayou surface-water monitoring
program will be implemented and the results will be incorporated
into an advisory program conducted by the NWFWMD for 3-D
modeling/contaminant tracking. The monitoring program will also
include installation of an additional two monitoring wells
adjacent to Bayou Texar. The purpose of this advisory program
will be to periodically evaluate the hydrogeologic conditions and
quality of the ground water in the Sand-And-Gravel aquifer .
underlying the site. In addition, the ground-water and surface-
water monitoring specified as part of this remedial action will
verify that contaminant levels are declining through natural
attenuation and flushing as predicted by the above ground-water
modeling.
This remedy does not include a pump and treat technology.
However, this limited action avoids potentially adverse impacts
59
-------
associated with the ground-water extraction and treatment
alternatives. The negative impacts of a pump and treat cleanup
method would include: 1) spreading of off-site plumes of
contamination such as the Escambia Treating Company (ETC) site
organics plume; 2) salt water intrusion; and 3) alteration of
ground-water flow patterns in the Sand-And-Gravel aquifer.
Implementation of Alternative 2 in conjunction with source
treatment and containment will protect human health and the
environment. The use of source treatment and containment will
eliminate the source of ground-water impacts. In addition,
treatment of the contaminated source will prevent or minimize
further impacts to the surface water of the Bayou Texar due to
discharge of contaminated ground water to that water body.
The Performance Standards selected for the chemicals of concern
are as follows:
Contaminant of Concern Cleanup Level (mg/L)
Fluoride 4 mg/L (ppm)*
Arsenic 0.05 mg/L (ppm)
Chloride" 250 mg/L (ppm)
Sulfate" 250 mg/L (ppm)
Nitrate/nitrite 10 mg/L (ppm)
Radionuclides
Radium-226"*
Radium-228*" 5 pCi/L
(Radium-226,228 combined)
* The MCL of 4 ppm for fluoride is the cleanup level for
ground water. The Florida secondary standard of 2 ppm
contained in Section 17-550.320, F.A.C. will apply at nearby
municipal potable supply wells as specified in Alternative
10 (contingency remedy).
Chloride and sulfate were not included in the baseline
risk assessment because no toxicity values exist. The
remedial goals presented for•chloride and sulfates are the
Florida ARARs.
*** The proposed MCL for Radium-226 and Radium-228 is 20
pCi/1 for each.
Because certain performance standards may not be determined until
the Remedial Design phase, the list of performance standards in
this section is and may be subject to addition and/or
modification by the Agency in the RD/RA phase.
Compliance with applicable ARARs will be attained over time
through natural attenuation (adsorption, precipitation, dilution)
and flushing of the aquifer. A 70 year period is estimated to
60
-------
1992).
The Chemical-specific ARARs for this alternative are as follows:
o National Primary Drinking Water Regulations, 40 CFR Part 141
o Florida Primary Drinking Water Standards - Maximum
Contaminant Levels, Section 17-550.310, F.A.C.
o Florida Surface Water Quality Standards, Sections 17-
302.400, 17-302.500, 17-302.510, 17-302.530, and 17-302.560,
F.A.C.
o Florida Secondary Drinking Water Standards for fluoride,
chloride, and sulfate. Section 17-550.320, F.A.C.
Location specific ARARs that are applicable to the site include:
o Statement of Procedures on Wetlands Protection, Appendix A
to 40 C.F.R. Part 6
o Florida Antidegradation Policy for Surface Water Quality,
Section 17-302.300, F.A.C.
This alternative will meet all Federal and State ARARs.
9.1 CONTINGENCY
In considering the requirements of -CERCLA, the NCP, the detailed
analysis of alternatives and public and state comments, EPA has
selected Alternative 10 as its contingency remedy for addressing
contaminated ground water related to the Agrico Chemical
Superfund Site.
This contingency alternative is protective, cost-effective, and
will attain all Federal and State requirements. At the
completion of the above contingency remedy, the risk associated
with this site has been calculated at IxlO'6 which is within
EPA's acceptable risk range of IxlO"4 to IxlO'6. EPA has
determined that this risk range is protective of human health and
the environment. Because this remedy will result in hazardous
substances remaining in the aquifer above health-based levels for
a time period exceeding five years, a review will be conducted
within five years after commencement of the remedial action to
ensure that the remedy continues to provide adequate protection
of human health and the environment.
The total present worth cost of the contingency remedy,
Alternative 10 is $21,039,000 (wellhead treatment) and $999,000
(wellhead replacement).
Based on current hydrogeologic conditions, it is unlikely that
61
-------
Alternative 10 is $21,039,000 (wellhead treatment) and $999,000
(wellhead replacement).
Based on current hydro-geologic conditions, it is unlikely that
nearby municipal water supply wells will be impacted by the
Agrico ground-water plume. However, if at some time in the
future the Agrico plume threatens nearby municipal water supply
wells, Alternative 10 will be implemented. Specifically, if
fluoride levels detected in nearby municipal water supply wells
exceed Florida's secondary drinking water standard of 2 mg/L, EPA
will evaluate whether implementation of this contingency remedy
is appropriate. The determination of whether wellhead treatment
or well replacement (new municipal water supply well
installation) is appropriate will be made when or in the event
the water supply well(s) are impacted by the Agrico plume.
Technical considerations such as frequency of sampling and/or
detection frequency will be clearly specified in a subsequent
Remedial Design (RD) Report.
This alternative will achieve substantial risk reduction by
restricting current and future ground-water use and by treating
municipal water supply wells if they become impacted in the
future.
Alternative 10 consists of the following institutional controls
and treatment technologies: On-site deed restrictions, ground-
water use restrictions j, ground-water monitoring, wellhead
treatment or well replacement. Wellhead treatment includes
filtration, prefiltration, Reverse Osmosis (RO) with RO reject
evaporation pond, and off-site disposal of RO reject sludge from
an evaporation pond.
Ground-water remediation will occur at the wellhead before being
discharged to the public water supply. In addition, ground-water
remediation will occur through natural attenuation, flushing, and
dispersion over a period of 70 years. (Ground-water Modeling,
McDonald Morrissey, April 1992).
This contingency alternative will achieve the remediation goals
established for the Agrico site by treatment of the contaminated
ground water at nearby municipal water supply wells. If municipal
replacement well(s) are installed outside the area of ground-
water impact, remediation goals will be achieved through natural
attenuation, as envisioned in Alternative 2. This alternative
does not create the potential for diversion of other contaminant
plumes around the Agrico site.
The local water authority, ECUA, believes this is a viable
contingent solution to the Agrico site, as noted in its
correspondence of March 17, 1993.
If at some time in the future, the plume threatens any ECUA
62
-------
wells, that those parties responsible for the cleanup of the
Agrico site, at that time, fund the construction and
operation of the necessary treatment facilities to cleanup
the water which is being pumped. The treated water would be
fed directly into ECUA's water distribution system.
Extraction without reinjection avoids a major upset of water
flow/direction in the aquifer. Under this approach
contaminate plume migration and dilution concerns would be
minimal. This would fully use the water pumped resulting in
no loss of a natural resource.
In summary, we believe that the only appropriate way to
handle groundwater clean-up is to only extract the
contaminated groundwater when it becomes an identified
threat to the water quality of an existing or proposed well
and then fully utilize the cleaned water rather than to
waste a natural resource.
Alternative 10 considers the installation of a new municipal
water supply well or treatment of ground water at a nearby
municipal potable water supply well that has been impacted by the
Agrico plume. If fluoride levels exceed 2 mg/L at the nearby
municipal potable water supply wells, implementation of the
contingency remedy will be evaluated. Essentially, EPA will
determine if wellhead treatment or municipal potable water supply
well replacement is appropriate. Technical considerations such
as frequency of sampling and/or detection frequency will be
clearly specified in a subsequent Remedial Design (RD) Report.
Once the contaminated ground water has been treated, it is
discharged into the local potable water supply system or new
water supply wells located outside the area of identified ground-
water impact. The ECUA has done extensive testing of well sites
and has installed numerous wells to identify several potential
municipal water supply well locations in the area.
Performance Standards - The Performance Standards selected for
the chemicals of concern are as follows:
Gontg'm'''nair't of Concern
Fluoride
Arsenic
Chloride**
Sulfate"
Nitrate/nitrite
Radionuclides
Radium-226**
Radium-228**
Cleanup Level (mg/L
4 mg/L (ppm)*
0.05 mg/L (ppm)
250 mg/L (ppm)
250 mg/L (ppm)
10 mg/L (ppm)
(Radium-226/ 228 c<
5 pCi/L
xmbined)
63
-------
The MCL of 4 ppm for fluoride applies to the cleanup
level for ground water. The Florida secondary standard of 2
ppm contained in Section 17-550.320, F.A.C. will apply at
nearby municipal potable supply wells as specified in this
contingency remedy.
** Chloride and sulfate were not included in the baseline
risk assessment because no toxicity values exist. The
remedial goals presented for chloride and sulfates are the
Florida ARARs.
"* The proposed MCL for Radium-226 and Radium-228 is 20
pCi/1 for each.
Because certain performance standards may not be determined until
the Remedial Design phase, the list of performance standards in
this section may be subject to addition and/or modification by
the Agency in the RD/RA phase.
Compliance with applicable ARARs will be attained after treatment
at the wellhead and over time through natural attenuation
(adsorption, precipitation, dilution) and flushing of the
aquifer. A 70 year period is estimated to reach applicable
ARARs. (Summary of Results .- Analysis of Ground Water Flow and
Solute Transport at the Agrico Plant Site. Pensacola. Florida,
McDonald Morrissey Associates, Inc., April 1992)
The Chemical-specific ARARs for this alternative are as follows:
o National Primary Drinking Water Regulations, 40 CFR Part 141
o Florida Primary Drinking Water Standards - Maximum
Contaminant Levels, Section 17-550.310, F.A.C.
o Florida Surface Water Quality Standards, Sections 17-
302.400, 17-302.500, 17-302.510, 17-302.530, and 17-302.560,
F.A.C.
o Florida Secondary Drinking Water Standards for fluoride,
chloride, and sulfate, Section 17-550.320, F.A.C.
Location specific ARARs that are applicable to the site include:
o Statement of Procedures on Wetlands Protection, Appendix A
to 40 C.F.R. Part 6
o Florida Antidegradation Policy for Surface Water Quality,
Section 17-302.300, F.A.C.
This alternative shall meet all Federal and State ARARs.
64
-------
10.0 STATUTORY DETERMINATIONS
Under its legal authorities, EPA's primary responsibility at
Superfund sites is to select remedial actions that are protective
of human health and the environment. In addition, Section 121 of
CERCLA establishes several other statutory requirements and
preferences. These specify that when complete, the selected
remedy for this site must comply with applicable or relevant and
appropriate environmental standards established under Federal and
State environmental laws unless a statutory waiver is justified.
The selected remedy also must be cost-effective and utilize
permanent solutions and alternative treatment technologies or
resource recovery technologies to the maximum extent practicable.
Finally, the statute includes a preference for remedies that
employ treatment that permanently and significantly reduces the
volume, toxicity, or mobility of hazardous wastes as their
principal element.
The pumping and treating of the Agrico ground-water
contamination, as envisioned in Alternatives 5, 6,7 and 8
presents a substantial risk through spreading of off-site plumes
of contamination, salt-water intrusion, and alteration of the
ground-water flow in the aquifer. Therefore, EPA, in
consultation with the State, views natural attenuation of the
Agrico Chemical Site ground-water contamination as being more
protective of human health and the environment.
In addition, the Agency has selected a remedy that will comply
with applicable or relevant and appropriate environmental
standards established under Federal and State environmental laws
through natural attenuation within the guidelines specified in
EPA's Guidance on Remedial Actions for Contaminated Ground Water
at Superfund Sites. December 1988.
Specifically, the above guidance states, "There are special
situations when it may not be practicable or feasible to fully
restore ground water Natural attenuation and wellhead
treatment with monitoring and institutional controls may be the
only feasible remedies for these sites If levels of
contaminants are projected to attenuate, a waiver (of ARARs) may
not be necessary if cleanup levels will be,achieved in a
reasonable time frame (i.e., less than 100 years)." The
following sections discuss how the selected remedy and the .
contingency remedy for the Agrico Chemical site meet these
statutory requirements.
10.1 PROTECTION OF HUMAN HRALTH AND T**F ENVIRONMENT
The treatment of the contaminated soils on site prevent exposure
to soil contamination as well as eliminate contaminant loading,
or leaching, to the ground-water (environment). In addition,
the selected remedy and the contingency remedy prevent the
65
-------
potential for widespread and uncontrolled aquifer degradation
(potentially impacting existing water supply wells) that will
most likely occur if the other (pump and treat) remedies are
implemented. Through this planned ground water management
approach, the selected remedy and the contingency remedy will
serve to ensure protection of human health and the environment.
The current risk associated with exposure to ground water for
non-carcinogenic risk results in a combined Hazard Index (HI) for
potential ingestion and swimming exposure for an adult of 2 for
both shallow and deep ground water. The HI for a child is 5
(shallow) and 4 (deep). Fluoride and PAHs contributed most to
the HI. Excess Lifetime Cancer Risks (ELCRs) for exposure to
shallow ground water is 9 x 10'5 (shallow) and 4 x 10~5 (deep) for
an adult, and 5 x 10"5 to 2 x 10'5 (shallow and deep) for a child,
respectively. Arsenic contributed most significantly to the ELCR
for shallow ground water; benzene and 2,4-dinitrotoluene
contributed most significantly to the ELCR for deep ground water.
However, a subsequent well and pool survey (Woodward Clyde
Memorandum August 1993) indicates this exposure pathway is
incomplete. Therefore, this evaluation applies only to a
hypothetical future risk scenarip.
The hypothetical future risks to adult residents exposed to
ground water via total potable household use and swimming
exposure range from 50 (deep) to 60 (shallow); the HI for a child
was 100 for shallow and deep. Fluoride contributes most
significantly to the His. ELCRs for an adult resident ranged
from 8 x 10'4 (deep) to 4 x 10'3 (shallow) and 4 x 10~* (deep) to 2
x 10"3 (shallow) for a child. Arsenic and 2,4-dinitrotoluene
contribute most significantly to the ECLR.
Through implementation of the selected remedy, risk levels should
be effectively reduced to acceptable levels.
10.2 COMPLIANCE WITH APPLICABLE OR RELEVANT AND APPROPRIATE
REQUIREMENTS
Remedial Action performed under CERCLA must comply with all
ARARs. All alternatives considered for the Agrico Chemical site
were evaluated on the basis of ARAR compliance. The selected
remedy and the contingency remedy will meet or exceed all ARARs.
10.2.1 Chemical-Specific ARARs
The Chemical-specific ARARs for this alternative are as follows:
o National Primary Drinking Water Regulations, 40 CFR Part 141
o Florida Primary Drinking Water Standards - Maximum
Contaminant Levels, Section 17-550.310, F.A.C.
66
-------
o Florida Surface Water Quality Standards, Sections 17-
302.40_0, 17-302.500, 17-302.510, 17-302.530, and 17-302.560,
F.A.C.
o Florida Secondary Drinking Water Standards for fluoride,
chloride, and sulfate, Section 17-550.320, F.A.C.
10.2.2 Location-Specific ARARs
LOGation-specific ARARs that are applicable to the site include:
o Statement of Procedures on Wetlands Protection, Appendix A
to 40 C.F.R. Part 6
6 Florida Antidegradation Policy for Surface Water Quality, .
Section 17-302.300, F.A.C.
This alternative shall meet all Federal and State ARARs.
10.3 COST EFFECTIVENESS
The selected remedy utilizes natural attenuation and ground-water
monitoring which can be easily implemented at the Agrico Chemical
Site. The contingency remedy employs a proven .technology which
can easily be implemented at the Agrico Chemical Site. These
technologies (ground-water monitoring, well head treatment/
replacement) provide the most cost effective remedies when
compared to the other alternatives due to their ability to most
effectively contain and limit further spread of contamination.
10.4 UTILIZATION OF PERMANENT SOLUTIONS AND Ar-TERNATTVE TREATMENT
TECHNOLOGY OR RESOURCE RECOVERY TECHNOLOGIES TO TWR MATCTMDM
EXTENT PRACTICABLE
EPA has determined that the selected remedy represents the
maximum extent practicable to which permanent solutions can be
utilized at the Agrico Chemical site.- Of those alternatives that
are protective of human health and the environment and comply
with ARARs, EPA has determined that the selected remedy and the
contingency remedy provide the best overall balance of tradeoffs
in terms of the five balancing criteria: long-term effectiveness
and permanence; reduction of toxicity, mobility and volume;
short-term effectiveness; implementability; and cost.
Additionally, the selected remedy and the contingency remedy need
not be modified under the two modifying criteria: state
acceptance; and community acceptance.
The selected remedy meets the statutory preference to utilize
permanent solutions to the maximum extent practicable.
67
-------
10.5 PREFERENCE FOR TREATMENT AS A PRINCIPAL ELEMENT
The selected remedy does not envision treatment of the
contaminated ground water related to the Agrico site. The
contingency remedy allows for treatment at the wellhead or
replacement of impacted public water supply wells. However, the
limited action envisioned in the selected remedy and in the
contingency remedy avoids potentially adverse impacts associated
with the ground-water extraction and treatment alternatives. The
negative impacts of a pump and treat cleanup method would include
spreading of off-site plumes of contamination such as the
Escambia Treating Company (ETC) site organics plume, salt water
intrusion, and alteration of ground-water flow patterns in the
Sand-And-Gravel aquifer. Therefore, EPA views natural attenuation
of the Agrico Chemical Site ground-water contamination as being
protective of human health and the environment.
In addition, the Agency has selected a remedy of natural
attenuation pursuant to EPA's Guidance on Remedial Actions for
Contaminated Ground Water at Superfund Sites. December 1988.
Specifically, the above guidance states, "There are special
situations when it may not be practicable or feasible to fully
restore ground water ..Natural attenuation and wellhead
treatment with monitoring and institutional controls may be the
only feasible remedies for these sites." The above guidance
further states that natural attenuation may be preferable under
certain conditions. The guidance further defines a reasonable
time frame for natural attenuation to occur should be within 100
years. As stated previously, ground-water modeling conducted for
this site estimates natural ground-water contamination cleanup to
occur within 70 years.
10.6 DOCUMENTATION OF SIGNIFICANT CHANGES
The Proposed Plan for the Agrico Chemical site was released for
public comment on February 15, 1994. The Proposed Plan
identified Alternative 2 as the preferred alternative for this
action, with Alternative 10 as a contingency remedy. EPA
reviewed all written and verbal comments submitted during the
public comment period. Upon review of these .comments, it was
determined that a significant change to the remedy, as it was
originally identified in the Proposed Plan, is appropriate.
After consultations with the FDEP, EPA has determined that the
MCL for fluoride of 4 ppm is applicable as a remedial goal for
cleanup of the site ground-water contamination. Florida's
Secondary Drinking Water standard of 2 ppm identified in the
Proposed Plan will trigger the evaluation of implementation of
the contingency remedy as it applies to nearby municipal water
68
-------
supply wells. This is because the primary MCL for fluoride is 4
ppm. The State's secondary standard (2 ppm) does not necessarily
require corrective action under Chapter 17-550.320, F.A.C.
69
-------
Appendix A
Correspondence
-------
e Director
Route 1. Boz 3)00. Havana. Florida 32333-9700 • ' •.
(On US. Highway 90. 10 miles west of Tallahassee)
November 22. 1993
-5999
(Suncom) 771-2080
Dave B. Abbott, Chief .
North Florida Remedial Section . -
U. S. Environmental Protection Agency: Region IV
345 Courtland Street. N. E.
Atlanta. GA 30365
Dear Mr. Abbot:
As you know, Uic Northwest Florida Water Management District (NWFWMD) participated in a
meeting with U>c U. S. Environmental Protection Agency; Region IV (EPA), the Florida Department of
Environmental Protection (DEP) and ihc Etcamhia County Utilities Authority (ECUA) on August 25tl> of
this year (DEP; Tallahassee. Florida). The purpose of this meeting was to discuss the appropriate means to
resolve (]>c Agrico Chemicals site contamination problem in Pcnsacola, Florida. As a representative of the
N\VFWMD, 1 outlined our agency's concerns and the underlying reasons for those concerns. At the
conclusion of the event, "a mutual resolution was reached among the participants that additional
information was necessary concerning the fluoride concentrations hydraulically downgradiem from the
Agrico Chemicals site. ;
S' - '
Recently, I received the technical memorandum of the additional field work performed at the
Agrico Chemicals site (Technical Memorandum Off-Site Monitoring Well Installation and Ground Water
Sampling: Agrico Chemicals Site, Pensacola Florida, Woodward-Clyde Consultants, 1993). After
reviewing the results of this activity, I am taking this opportunity to provide you with the Northwest
Florida Water .Management District's (rJWFWMD) analysis of the Agrico contamination problem! In
essence, this correspondence is a reiteration of the NWFWMD's view of the contamination problem as
presented in the previous meeting (DEP; Tallahassee, Florida, August 25, 1993). The results of the
Technical Memorandum complement our previous analysis.. It is anticipated that the NWFWMD's
perspective may aid the U. S. Environmental Protection Agency's and the Florida Department of
Environmental Protection's efforts in rectifying the Agrico situation.
Based on the evidence from the data collection efforts of the Agrico site RI/FS and the recently
completed sampling activity, our conceptualization of the problem is outlined as follows:
,(1). The contamination plume follows the natural hydraulic gradient of the deeper zone (main producing
zone) of the Sand-aod-Gravel Aquifer.
(2). The plume discharges to Bayou Texar and, therefore, a remediative process is active via natural
means. ",*'-'
(3). The highest fluoride concentrations are, as expected, observed along a narrow corridor which
coincides with the most direct flow line from the source to the discharge area. The highest ground water
concentrations are less than the pore water corceatxattons observed at the discharge zone. In addition, the
concentrations along mis flow line do not increase significantly with-distance, downgradient from the
CHARLES W.ROBERTS
Quimun - Bristol
E HENTZ FLETCHER, JR.
Vice Qainntn - Quiney
BENNETT EUBANKS
Sec/Treat. - Bioomaown
JOHNO.DELORGE
Cantonment
-------
source: i.e. concentrations downgradicnt arc not much different than the concentrations observed in well
AC-3. which is closest to the source.
(4). The concentrations attenuate rapidly, transverse 10 the flow direction
(5). Under current conditions. Uic plume docs not pose a threat to existing public water supply wells in the
area.
(6). From Mr. Lee Thomas' analysts (EPA memorandum; August 24, 1993). the fluoride flux, (calculated
from the observed levels of fluoride and ground water discharge) into Bayou Texar may not exceed
surface water regulator)1 requirements for the protection of biota.
(7). From the swimming pool survey study, the contamination appears to pose no health risk to local use
of the ground water for recreational purposes.
(8). The ECU A, winch is responsible for potable water development in the area, is aware of the magnitude
and extent of the contamination.
The Sand-and-Gravel Aquifer underlying Escarnhia County is. the sole source of potable water to
the local communities. The ECUA is the major supplier of potable ground water for Die City of Pensacola
;uid neighboring areas, with permitted average wiihrawal of 45 million gallons per day. Unfortunately.
repeated instances of ground water contamination have demonstrated Uic vulnerability of Uic aquifer and
the complicated nature of ground water resource management in Uie area. This susceptibility arises from
several factors: proximity of the aquifer to land surface; relatively high vertical hydraulic conductivity of
materials found between land surface and the top of ibe aquifer, lack of competent confining units within
the aquifer to retard venical movement of contaminants; and high local recharge rates. These factors work
together to create a situation which makes the aquifer highly vulnerable to the impacts of inappropriate
land use activities.
As a pan of the Water Quality Assurance Act, passed by the Florida Legislature in 1983, the DEP
has developed a cooperative relationship with the five water management districts to establish a water
quality monitoring network in the State. One element of the network is designed to monitor the ground
water quality in selected areas of varying land uses. One of the VISA (Very Intensive Study Area)
networks was established in southern Escambia County. The majority of the monitor wells were installed
within the general area of the Agrico Chemicals site. From a single sampling event of 43 VISA wells, a
variety of organics were detected from 27 VISA wells (Findings of the Pensacola VISA: Preliminary
Report, 1991, DEP). The positive results included chlorinated pesticides, carbamate/urea pesticides.
herbicides, pemacbloropbenols, benzene, xylene, PCE.TCE, constituents from the Agrico plumes and
other organic compounds. Some, but not all of these contaminants may be linked to the Agrico plume.
Given the above scenario, management of ground water resources in Escambia County is a
sensitive, problematic and complicated task; especially within the additional restrictions imposed by the
saltwater fronts of the Pensacola and Perdido bay systems which bound the aquifer. The irony of the
aquifer system in the area is that the high recharge characteristics which allow for a prolific resource also
limit the use of this resource, due to the relative ease of ground water contamination. Until an adequate
aquifer/wellhead protection program and a dvic^ediicational program is developed for the area, further
instances of contamination should be anticipated. Implementation of plume restoration without some
mechanism to prevent future contamination in the area may be nonproductive.
The potential risks imposed by pump and treat technology appear to outweigh the benefit of
remediation efforts via this method. Unfortunately, we have yet to ascertain the magnitude, spatial extent
and geochemical characteristics of all contaminant plumes in the vicinity of the Agrico plume (e.g.
-------
* etc ) Because conventional pumping techniques for ground water remediation alter
uc o'in the aquifer, there is no guarantee that such means of remediation wll no, .nduce
urr degra^L of the resource from these contaminants. To compound the situauon. there « an added
risk to the resource from the salt water intrusion from Bayou Tcxar suuated ,n UK: vc.mty.
The Acrico problem may he best undcrsuxxl when placed within a larger picture of the current
, nJlhv Vifuation in the area. Because of ihc greater concern with the presence of many other adverse
water qualu> suu»uon ^me area ^ contamination may need to be reassessed in
S2^S«r?be option for a Limited Action (Draft Supcrfund Proposed Plan
oSSTsSe. September 1993) has been discussed during our previous meeting
nPP AMMKI •>$ 1993) The NWF\VMD supports this option in addressing the Agrico ground water
SmarnirSion problem. 'l hope *is correspondence clariTufs U,c NXVFWMD's position on the Agrico
cortumination problem. If you have further questions concerning this matter. I am a, your scrv.cc.
Honesto Roaza
Associate HydrogeolojisL P:
cc: Patsy Gol'dbcrc (EPA: Region IV)
Kclscy Hclton'(DEP) •
Zoc Kulakowski (DEP)
Bcrnic Dahl (ECUA)
-------
I'.O. IJOX >53" V-?0 HAMMAN STRUCT IMINSACOLA. FLORIDA 3-'Si-i-o.
-------
Ms. Stephanie Gordlnlcr
March 17. 1993
Page 2
ilil
water received. We believe that this would have a vary serious adverse
economic impact on Pensacola and Escambla County.
B. Treated groundwater from the AgrtCQrSlte would simply dilute the waslcwaler
stream. This clear water has the potential to upset the treatment process
due to the lack of nutrients, and needlessly occupies capacity otherwise
better used to treat residential and commercial wastewatcr inputs.
C. There is also a great concern that if there is a failure of the remedial
treatment process, the product stream could cause major biological and/or
chemical upsets In the wastewater treatments plant. At this time we are not
sure of all the potential contaminants in this flow, therefore we can not
comment on the treatment plants capacity to remove any of the pollutants in
this water. Depending on the nature and severity of any potential process
upset it could lead to EPA/DER citations for ECUA waslewater permit
violations. This is an unacceptable risk, unless EPA is willing to waive all
violations.
D. The local sewer system Into which this flow .-would discharge would have to
be reevaluated. If the sewers and downstream lift stations arc inadequate.
which is most likely the .case, then these facilities would have to be
upgraded. This would be an additional expense and disruption to the
adjacent properties. In the future. If this flow was ever removed from the
system then the expanded sewer lines would be oversized. This would result
in reduced velocities in the system, leading to deposition of solids and the
potential generation of hydrogen sulfide and other objectionable gases.
2. Extract groundwater. treat, and reinject Into the aquifer.
A. We believe that extraction of large volumes of contaminated groundwater has
a high probability of altering the groundwater flow and as a result may
Induce more contaminates Into existing ECUA wen cones of Influence. If the
treated water Is not injected between the Agrtco plume and upstream
contamination plumes, from other sites, the withdrawal of this water will
surely cause an enhanced migration of pollutants from others area. This Is
highly undesirable.
-••
B. On the other hand If the relnjection occurs upstream of the Agrico site the
relnjected groundwater wffl cause further dilution of the contaminates in the
plume and thus prolong the remedial operation and require even larger
water withdrawals. This likewise is undesirable and counterproductive.
C. .Reinjection contributes.no Identifiable real benefit to the environment or
public. Hydraulic studies to date show the Escambla County aquifer
recharge is greatly in excess of today's and projected future withdrawal rates.
-------
Ms. Stephanie Gordlnicr.
March 17. 1993
Page3
:CUA
Salt water intrusion Is not a present or anticipated future problem.
D. ECUA Is the sole public water system drawing from the aquifer immediately
above and below the Agrlco Chemical Site. Along with Escambla County and
the City of Pensacola. ECUA is fomiulating well head protection ordinances.
On a local level, permitabillty of injection of any water source into the aquifer
that could affect contaminate migration...etc.. and affect present potable
water wells is questionable.
It is our understanding that the two scenarios which we commented on above arc the two
alternatives being evaluated at the present time. We would like to add the following
additional comments for your information and consideration.
Presently, there Is no functional three dimensional model of the aquifer in
Pensacola/Escambla County. However. ECUA has entered into an agreement with.
and funded, the Northwest Florida Water Management district to develop such a
model. This work is in process with expectations the model would be available in
six to nine months. We feel that it is Imperative the model be used to evaluate the
impact of any withdrawal and/or reinfection of groundwaler from the Agrlco site.
?v' . .
At the present time it is our understanding that the Agrlco pollutant plume is not
adversely impacting our wells. Therefore, is it advisable to disturb the aquifer or
would it not be better to Just monitor the groundwater for plume movement? All
groundwater monitoring data should then be made available to the Northwest
Florida Water Management district for inclusion in the three dimensional model.
Rather than expending large sums of money to achieve an undefined benefit we
would suggest that monies be spent to assist in the enhancement of the model and
database so this and other pollutant plumes can be tracked.
As Escambla County grows we will require additional wells. However, we do not
pump this well water to a common site for treatment and subsequent redistribution.
Rather, we site wells In the close proximity of the new demands., Therefore.
additional pumpage from the vicinity of the Agrlco site is not anticipated to meet
future needs. We anticipated the construction of any new wells to be in the
northern and western areas of the county.
Given the investment and operating cost to extract groundwater and treat to potable
water quality, the* most cost effective approach for all concerned would be to use any
pumped and treated groundwater as potable water in a public utility. Since ECUA
is the only supplier of water in the vicinity of the Agrico site only ECUA's water
supply Is being impacted. We believe that the wisest move would be to:
A. Increase the monitoring of the,groundwater and input all the data into the
• - ; database for the three dimensional model
-------
Ms. Stephanie Gordlnlcr
March 17. 1993
Page 4
B. If at some lime In the future the plume threatens any ECUA wells, that those
parties responsible for the cleanup of the Agrtco site, at that time, fund the
construction and operation of-the necessary treatment facilities to cleanup
the water which Is being pump. The treated water would be fed directly into
ECUA's water distribution system. Extraction without relnjectlon avoids a
major upset of water flows/direction In the aquifer. Under this approach
contaminate plume migration and dilution concerns would be minimal. Tills
would fully use the water pumped resulting In no loss of a natural resource.
In summary, we believe that the only appropriate way to handle groundwatcr clean-up
projects Is to only extract the contaminated groundwatcr when 11 becomes an Identified
threat to the water quality of an existing or proposed well and then to fully utilize the
cleaned water rather than to waste a natural resource.
We appreciate the opportunity to comment on these proposed cleanup approaches.
Very truly yours.
slh
pc: Bernard Dahl. Director of Wastewater
Otto Prochaska, Director of Water
Tim Haag, Assistant to the Director
-------
Appendix B
Resnonsiveness
-------
RESPONSIVENESS SUMMARY
The U.S. Environmental Protection Agency established a public
comment period from February IS, 1994 through March 17, 1994 for
interested parties to comment on the proposed plan.. EPA
presented its preferred remedial alternative at a public meeting
during the comment period and answered questions from the public
concerning the alternative and the proposed plan.
A responsiveness summary is required by Superfund law and
regulations to provide the public with EPA's response to citizen
comments and concerns about the site raised during the public
comment period. All comments in this document have been factored
into the final Operable Unit Two Record of Decision for the
Agrico Chemical Site.
This responsiveness summary for the Agrico Chemical Site is
divided into the following sections:
I. Overview: This section discusses the recommended
• remedial alternative for operable unit two for the site
and the public reaction to this alternative.
II. Background on Community Involvement and Concerns; This
section provides a brief history of community interest
and concerns regarding the Agrico Chemical Site.
Ill. Su'mn|arv of Ma "lor Questions and Comments Received During
the Public Comment Period and EPA's Responses; This
section presents both oral and written comments
submitted during the public comment period, and
provides the responses to these comments.
IV. Remaining Concerns: This section discusses community
concerns that EPA should be aware of in design and
implementation of the chosen remedial action for
operable unit two.
I. Overview
The proposed plan for this action was presented to the public in
a fact sheet released on February 11, 1994, and at a public
meeting held on February 18, 1994. This action involves limiting
exposure to contaminated ground water, and primarily consists of
the following: (1) ground water and bayou surface water
monitoring (2) a door-to-door irrigation well survey; (3)
voluntary plugging and abandonment of private irrigation wells,
if impacted; (4) restrictions on existing and newly installed
irrigation wells; and (5) an advisory board for 3-D
modeling/contaminant tracking to include a cooperative effort
.between the parties conducting the remediation and the local,
state, and federal regulatory agencies to ensure that no one is
exposed in the future to the contaminated ground water associated
-------
with the site. The estimated capital cost of this action is
$1,661,000.
II. Background on Community Involvement and Concerns
EPA conducted its first community interviews for the site in the
winter of 1990. A RI/FS fact sheet was mailed to interested
citizens in May, 1990. The fact sheet explained the overall
process of Superfund, the upcoming RI/FS at the site, and
opportunities for community involvement. A RI/FS kick-off
meeting was conducted on June 5, 1990 with about 35 interested
citizens of Pensacola, Florida. Television interviews were
conducted on two occasions during the field investigation. EPA
also met with members for whom a new baseball field was built to
discuss potential site risks.
EPA conducted a second public meeting on August 13, 1992. The
purpose of this meeting was to explain the Superfund process, the
proposed remedy for source control at the Agrico Chemical Site,
and to answer questions and concerns of the community regarding
any aspect of the proposed remedy. Two key issues or concerns
were identified in the meeting. They included health effects
from exposure to site contaminants and potential adverse effects
during remedial action construction of contaminated soils. EPA's
response to these concerns can be found in Appendix A of the
September 29, 1992 OU1 ROD.
In 1993, a citizens' group (Citizens Against Toxic Exposure or
"GATE") applied for a Technical Assistance Grant (TAG) for the
Agrico Chemical Site. The TAG was awarded to the above group in
May 1993.
The comment period for the Proposed Plan for OU2 (ground water)
was from February 15, 1994 to March 17, 1994. During the comment
period, the Administrative Record was available to the public at
both the information repository maintained at the Pensacola
Public Library and at the EPA Region IV Docket Room in Atlanta,
Georgia.
A Proposed Plan public meeting was held on Friday, February 18,
1994. At the meeting, representatives from EPA presented the
Agency's preferred alternative for ground-water cleanup of the
site and answered any questions the public had regarding the
preferred alternative.
Approximately 50 interested persons attended this public meeting.
Some residents of streets near the Site questioned why EPA had
not investigated their properties for possible soil contamination
from the Agrico site. Responses to all comments made during the
public comment period, including the comments made at the public
meeting are included in the following section.
B-2
-------
III. Summary of Major Questions and Comments Received Purina the
Public Comment Period and EPA's Responses
The public comment period for the proposed plan for remediation
of the site's contaminated ground water was from February 15,
1994 through March 17, 1994. One letter was received during the
public comment period. In addition, the GATE group of citizens
presented (as submitted by the group's technical advisor) 12
written Questions For The EPA. A summary of public questions and
concerns, beginning with the above 12 questions, along with EPA's
response, follows:
1). Why does the Proposed Plan present no significant
information on site hydrogeology? Certainly, presenting such
information is the normal practice for a Plan that addresses
groundwater cleanup, or does EPA disagree?
EPA Responset In its Superfund guidance, the National
Contingency Plan (NCP) describes how the Agency should write the
Proposed Plan in paragraph [E](2) of Section 300.430 Remedial
Investigation/feasibility study and selection of remedy.
Specifically, this passage states, "The lead agency (EPA), in
conjunction .with the support agency and consistent with Section
300.515(e), shall prepare a proposed plan that briefly describes
the remedial alternatives analyzed by the lead agency, proposes a
preferred remedial action alternative, and summarizes the
information relied upon to select the preferred alternative'."
Consistent with the NCP, the Proposed Plan does summarize the
information EPA relied upon in its selection of the preferred
alternative. Extensive, detailed information concerning site
hydrogeology can be found in the six (6) volumes that comprise
the Remedial Investigation/ Feasibility Study (RI/FS) Report
located at the local site repository at the Pensacola Regional
Library. Finally, site hydrogeology is summarized in Section 5.3
and Figures 5-1 and 5-2 of this. Record of Decision.
2). Why is the only data on groundwater contamination presented
in the Plan for the deep zone and not the shallow zone of the
upper most aquifer?
EPA Responses The scope of the Proposed Plan, as stated in the
previous response, is to summarize the information relied upon to
select the preferred alternative. While there is site ground-
water contamination in the shallow zone of the Sand-And-Gravel
Aquifer in the immediate vicinity of the site and the bayou, the
majority of the Agrico plume resides in the deep zone of the
upper most aquifer. Detailed information on site contamination
in the shallow zone can be found in Section 5.5.2 Ground-Water
Investigation of this document. In addition, extensive
information on site contamination in the shallow zone of the
aquifer can be found in the RI/FS Report in the local
respository.
B-3
-------
3). Have any wells within and outside the site attempted to
identify the presence of contaminants in the deeper aquifers? If
so, have there been any results above detections limits for any
of the contaminants identified at the site?
EPA Response; The Pensacola Clay, which is vertically persistent
in the Pensacola area, acts as a confining unit between the
contaminated Sand-And-Gravel aquifer and the deeper aquifers.
This "aquiclude" ranges from 300 to 600 feet in thickness in this
area. Therefore, the Floridan aquifer has not been the object of
the Agrico Chemical site ground-water investigation.
However, as part of Remedial Action (RA) for this operable unit
two (OU2), a door-to-door well survey will be conducted. If a
water supply well screened in the deeper Floridan Aquifer is
discovered, appropriate well testing and subsequent plugging and
abandoning (if necessary) would be pursued by EPA.
4). Because the only data presented is for the deep zone and
only one set of primary contaminants of concern listed,
presumably the latter are for the deep zone. But other site
documents show a much larger list of contaminants of concern for
the deep zone - 19 versus 7 in the Plan. How does EPA justify
this use of incomplete data?
EPA Response; Initially, EPA determines through investigative
sampling what contaminants are present in the study area.
Subsequent to the investigation, EPA determines what chemicals
will be considered when performing a risk assessment according to
the public health guidance document for use in the Superfund
program, Human Health Evaluation Manual (USEPA, 1989).
As specified in the above manual, several factors are considered
in determining whether a constituent detected in ground water is
included or dropped from consideration as a contaminant of
concern (COC). A constituent not detected at concentrations
above the detection limit is not retained for consideration as a
COC. Contaminants detected only once in a particular media at a
concentration less than twice the detection limit are excluded
from the COC list. Constituents detected in ground water at
concentrations below background concentrations for ground water
in the vicinity of Pensacola are eliminated from consideration as
COCs. Constituents that are essential human nutrients and toxic
only at very high doses(i.e., much higher than those associated
with ingestion of the ground water) are eliminated as
contaminants of concern. This systematic, scientific approach
was used in determining the seven (7) constituents of concern for
the Agrico Chemical site.
5). The (Proposed) Plan presents no data on federal and state
standards (ARARs) for protection of the environment, including
B-4
-------
aquatic species, and refers only to human health effects. Why
has EPA not presented any of the numerical standards for
comparison .with the selected cleanup levels?
EPA Response: Referencing again EPA's. first response, the scope
of the Proposed Plan is not to provide extensive detail on ARARs.
This document (OU2 ROD for ground water) and the RI/FS Report
provide extensive information on appropriate ARARs for the site.
In addition, the dilution calculations performed by EPA indicate
that current levels of site contamination do not adversely impact
aquatic species. In conclusion, the numerical standards (cleanup
goals) listed in the Proposed Plan are the appropriate federal or
state ARAR for the site.
6). In discussing impacts on the ecosystem, EPA refers only to
fluoride. What data and analysis supports the conclusion that no
other contaminant found at the site poses any significant risk to
the ecosystem?
EPA Responset The focus on fluoride as the site's primary
contaminant is appropriate because: 1) fluoride concentrations
in the aquifer are considerably higher than those of other
contaminants in comparison to their respective maximum
contaminant levels (MCLs); 2) fluoride movement through the
aquifer demonstrates the worst case scenario for fate and
transport behavior of a contaminant emanating from the site.
In addition to fluoride, the Bayou Texar study (Entrix, 1993)
sampled and analyzed several indicator contaminants, including
chloride, sulfate, and nitrate/nitrite. The results of this
study indicated that chloride concentrations ranged from 2.6 mg/L
to 9,600 mg/L. For comparison, standard sea water has a chloride
concentration of 19,000 mg/L. The ranges of chloride measured in
the near-surface porewater of Bayou Texar sediments should have
no adverse effect on the biota since the chloride concentrations
are within normal limits for estuarine habitats such as the Bayou
Texar. Sulfate concentrations ranged from 1.7mg/L to a. maximum.
of 1,800 mg/L. Since sulfate is ubiquitous in marine and
estuarine waters, the levels detected in the study are within
limits for a salt water and fresh water interface zone such as
the bayou. Therefore, it is unlikely that sulfate would
adversely impact the bayou biota. Nitrate/nitrite concentrations
ranged from below detection limit to a 8.6 mg/L. The
•^nit rate/nitrite limits detected in the study are within
acceptable limits for protection of the ecosystem.
7). An examination of .more complete information on groundwater
contamination reveals that the plume of contamination is much
larger in the deeper zone. What significance does this have for
understanding the impacts of groundwater contamination?
EPA Response; The site contamination plume follows the natural
B-5
-------
hydraulic gradient of the deeper zone (main producing zone) of
the Sand-and-Gravel Aquifer. The plume discharges to Bayou Texar
and, therefore, (from the stand point of the aquifer itself) a
remediative process is active via natural means.
8). The selection of Alternative 2 - Limited Action rests on the
assumption that "natural attenuation (dispersion) and flushing"
will be effective "within 70 years." Exactly what are the
scientific data (i.e., about the contaminants and the
hydrogeology) and analysis to support this belief?
EPA Response: A comprehensive, detailed ground-water flow and
solute-transport modeling was conducted fSummary of Results -
Analysis of Ground Water Flow and Solute Transport at the Aqrico
Plant Site. Pensacola. Florida. McDonald Morrissey Associates,
Inc. April 1992) to evaluate movement of dissolved chemical
constituents in ground water in the vicinity of the Agrico site,
to determine the fate of contaminants emanating from the site,
and to evaluate the effectiveness of various ground-water
remediation strategies. The above modeling report is included as
an appendix to the Final Phase II Remedial Investigation Report
and the Final Feasibility Study Report. Geraghty & Miller,
November 1993, and resides in the local repository at the
Pensacola Public Library.
The modeling indicates that (under existing flow conditions with
no active remediation of ground water) fluoride concentrations in
the aquifer will decline to 4 mg/L after 70 years. The modeling
assumes no further contaminant loading (treatment of the source)
in its calculation.
The selection of fluoride over other site chemicals for modeling
purposes follows the rationale that 1) fluoride concentrations
are the highest found in the aquifer relative to the federal
standard or the maximum contaminant level (MCL) for that
chemical; 2) fluoride movement through the aquifer demonstrates
the worst case scenario for fate and transport behavior of a
contaminant emanating from the site; and 3) while other
contaminants in the Agrico plume can be linked to other sources,
fluoride levels in the aquifer clearly represent the result of
plant processes at the site.
8a). Over such a long time would the probability of
contamination of the deeper two aquifers increase?
EPA Response; With the treatment of the source (no further
loading to the system), EPA does not believe this probability
will increase. However, as part of the selected limited action
remedy, the plume will be monitored for the extended period of
time it will take the system to attenuate.
9). EPA acknowledges that use of contaminated groundwater could
B-6
-------
lead to unacceptable risk of cancer and non-cancer health
effects. Exactly how can EPA assure no future use of the
contaminated water within the 70 years EPA says the selected
remedy may take?
EPA Response: The selected remedial alternative envisions a
cooperative effort between the parties conducting the remediation
and the local, state, and federal regulatory agencies to ensure
that no one is exposed to the contaminated ground water
associated with the site. Although potable water is currently
supplied by the city, a comprehensive door-to-door survey will be
conducted to verify that previous well surveys are accurate in
the assumption that no one is drinking water from the ground
water within the Agrico contaminant plume.
In addition, a comprehensive ground-water and bayou surface water
monitoring program will be implemented and the results will be
incorporated into an advisory program (conducted by the Northwest
Florida Water Management District) for 3-D modeling/contaminant
tracking. The purpose of this advisory program will be to
periodically evaluate the hydrogeologic conditions and quality of
the ground water in the Sand-And-Gravel Aquifer underlying the
site. Irrigation wells (which are the other primary exposure
source), proposed, and "in progress" irrigation wells will be
denied permitting or, plugging and abandoning of existing private
irrigation wells will be strongly encouraged, if shown to be
impacted and positioned within the contaminated area.
10). EPA makes reference to "other contamination from off-site
sources unrelated to the Agrico site," and takes the position
that a more active groundwater cleanup for Agrico would somehow
cause an even greater risk. Exactly what scientific or
engineering evidence supports this view that an active
groundwater extraction and treatment system for Agrico would have
such a negative impact?
EPA Response; As a part of the Water Quality Assurance Act,
passed by the Florida Legislature in 1983, the Florida Department
of Environmental Protection (FDEP) has developed a cooperative
relationship with the five water management districts to
establish a water quality monitoring network in the State. One
element of the network is designed to monitor the ground water
quality in selected areas of varying land uses. One of the VISA
(Very Intensive Study Area) networks was established in southern
Escambia County.
The majority of the VISA wells were installed within the general
area of the Agrico Chemical site. From a single sampling event
of 43 VISA wells, a variety of organics were detected in 27 VISA
wells. (Findings of the Pensacola VISA: Preliminary Report, 1991,
FDEP). The sampling results indicated the presence of
chlorinated pesticides, carbamate/urea pesticides, herbicides,
B-7
-------
pentachlorophenols, benzene, xylene, PCE, TCE, and other organic
compounds. _ Based on monitoring well sampling upgradient of the
Agrico site, some of the organics can be traced to the Escambia
Wood Treating Plant. The sampling results of the VISA wells
provide clear evidence that other contamination from multiple
sources exists in the Sand-And-Gravel aquifer.
In addition to the constituents listed above, there is
contamination associated with the Tank Inventory Sites (STI)
currently undergoing remediation (or targeted for remediation)
under a program supervised by the Florida Department of
Environmental Protection (FDEP). There are no less than twenty
(20) of these locations within a one mile radius of the Agrico
site. Based on FDEP ground-water sampling of these locations, it
is EPA's understanding that the Volatile Organic Contamination
(VOC) associated with these tanks resides primarily in the upper
most zone of the Sand-And-Gravel Aquifer.
Three or five extraction wells located downgradient of the site
and screened in the lower (main producing) zone of the aquifer
pumping 1 MGD to 1.3 MGD would likely result in the "drawing" of
these other contaminants into the extraction wells. This view is
shared by EPA, FDEP, and those agency (NWFWMD, ECUA) scientists
and engineers who have studied this aquifer on a daily basis for
several years.
In a letter to EPA dated November 22, 1993, the NWFWMD scientist
(co-author of the June 1991, Conceptual Model of the Sand-And-
Gravel Aquifer. Escambia County. Florida) familiar with the
Agrico site states, "...we have yet to ascertain the magnitude,
spatial extent and geochemical characteristics of all contaminant
plumes in the vicinity of the Agrico plume (e.g. Escambia
Treating, etc.). Because conventional pumping techniques for
ground water remediation alter the hydraulic flow in the aquifer,
there is no guarantee that such means of remediation will not
induce further degradation of the resource from these
contaminants. To compound the situation, there is an added risk
to the resource from the salt water intrusion from Bayou Texar
situated in the vicinity."
11). The data shown in Figure 3 of the Plan were obtained 2
years or more ago. Have any more recent groundwater data been
obtained which indicate higher concentrations of fluoride or a
larger plume in the deep zone?
EPA Response; More recent groundwater data has been obtained,
and supports previous fluoride contamination characterization.
At the request of the FDEP, an additional four wells (one
replacing an existing well) were installed downgradient of AC-3
(monitoring well showing the highest level of fluoride
contamination) and upgradient or near the Bayou Texar. The
results of the sampling of the new wells were made available to
B-8
-------
EPA in November 1993. The contaminant levels evidenced in these
new wells were consistent with previous characterization of the
Agrico plume .(with 98 mg/L of fluoride found nearest the AC-3
well and much lower levels found in the new wells adjacent to the
bayou).
12). Have any other plume contour maps been constructed for
contaminants other than fluoride? If so, will they be provided?
EPA Response». Contour maps depicting sulfate concentrations in
the shallow zone, sulfate concentrations in the deep zone, and a
vertical distribution cross section for sulfate can be found in
Figures 5-16, 5-17, and 5-18, respectively in the November 26,
1994 Final Phase II Rl Report. Similarly, maps depicting nitrate
concentrations in the shallow zone, nitrate concentrations in the
deep zone, and a vertical distribution cross section for nitrate
can be found in Figures 5-19, 5-20, and 5-21, respectively in the
above document. As part of the RI Report, these maps were made
available at the local Pensacola Regional Library (Agrico
Repository) on February 15, 1994 (the beginning of the public
comment period).
(Questions 13 through 18 were submitted by an independent
geologist living in the Pensacola, Florida area.)
13). No significant effort has been made in determining the
source of arsenic at the site. The geochemistry and
hydrochemistry of arsenic are exceedingly complex. Arsenic can
exist in four valence states in nature; some are almost totally
insoluble, others are highly soluble. I would submit (it is my
professional philosophy) that it is necessary to more fully
understand the source and history of disposal of arsenic at the
Agrico site prior to finalizing a remedy for ground water
cleanup.
EPA Response: The Agrico Chemical site was an operating facility
from 1889 to 1975. Understanding a detailed history of disposal
at the site would clearly be useful. However, due to the
extended period of operation and lack of record keeping practices
in the early years, that information does not exist. Therefore,
"empirical reasoning" must prevail.
The assertion that the arsenic originated from the pyrite ores is
a reasonable one. In addition, the phosphate rock used in the
production of fertilizer can contain trace amounts of pyrite or
arsenic. Moreover, current literature states that the fate of
arsenic in the aquatic environment is complex and dependent upon
a number of factors including pH, Eh, metal sulfide, and sulfide
ion concentrations, presence of phosphorous minerals, iron
concentration, temperature, salinity, and the distribution and
composition of biota.
B-9
-------
However, a clear understanding of the extent of arsenic in the
soils is more useful than speculation of its origin in finalizing
a. remedy for ground water cleanup. The Operable Unit One Record
of Decision (OU1 ROD) describes the extent of arsenic found in
the site's soils and establishes a conservative cleanup goal of
16 parts per million (ppm) for arsenic. Since the selected
remedy for ground water is natural attenuation (to include long-
term monitoring), the treatment of arsenic in the soils is the
primary concern in achieving cleanup of the Agrico Chemical site.
14). Was the airborne contaminant migration pathway examined at
the site?
EPA Response; The potential for significant transport of
contaminants from the site into the air is minimal. The
contaminants of concern, with the exception of nitrogen, are not
readily volatile. However, the transport of nitrogen from the
soils through the air is unlikely at the site. There is no
evidence of significant wind erosion at the on-site impoundment
ponds. With the exception of portions of the PFP II impoundment
pond, most of the site is covered with heavy brush and other
forms of vegetation. To address the possibility of off-site
migration of particles during soil cleanup activities, stringent
environmental controls will be executed during soil cleanup to
prevent fugitive dust migration.
15). The site map from the fact sheet (Proposed Plan) shows that
pond III underlies the right-of-way and the pavement of
Interstate 110. Was exploration completed in this area? Is
excavation beneath I-110 planned?
EPA Response; The figure in the Proposed Plan fact sheet
indicates where the PFP III impoundment pond boundaries extended
during plant operations. Consequently, surface and subsurface
soils samples were taken in the area of PFP III adjacent to and
on the east of the Interstate 110 right-of-way (Figure 3-4, Final
RI Report, November 1993). The results of these samples indicate
that of seventeen grab soil samples taken, two samples exceeded
the cleanup goal for fluoride (1,400 parts per million). Both
these sample locations are within the site's boundaries. The
soils at the two sampling locations are targeted for cleanup.
All other samples taken adjacent to the Interstate 110 (on-site)
showed fluoride levels well below the cleanup up number of 1,400
parts per million. Soil samples taken on the east side of
Interstate 110 (off-site) showed fluoride levels of <150 parts
per million which is below regional background levels of 200
parts per million.
16). Figure 5-5 from the Operating Unit 1 ROD does not indicate
any monitoring wells around and down gradient from the site, and
it follows that no tests of the "main producing zone" have been
completed east of the Bayou.
B-10
-------
EPA Response; The objective of the OU1 ROD was not to define the
extent of ground-water contamination, but to summarize what was
known about the Agrico ground-water plume at the time the OU1 ROD
was issued (September 1992). Because the investigation of the
soils was complete, EPA was able to proceed with putting a Record
of Decision in place in order to move forward with a Remedial
Design for soils treatment. Since that time, six monitoring
wells have been installed as part of the continuing ground-water
investigation. As part of that ground-water monitoring network,
a well cluster [screened in the shallow zone and deep zone (main
producing)] of the Sand-And-Gravel aquifer was installed east of
Bayou Texar and adjacent to where the plume is discharging to the
Bayou (RI Report, December 1993). The sampling results of the
wells east of the bayou indicate that the Agrico plume does not
extend beyond the bayou.
17). The City of Pensacola is planning to dredge sediment from
the uppermost end of Bayou Texar, from where 12th Avenue crosses
the Bayou, with dredging to be done in both directions. What
advice would EPA give the City with respect to disposal of
dredged material?
EPA Response: The City of Pensacola, as well as other county and
state regulatory agencies, are aware of the ongoing activities
related to the Agrico site. As indicated in the Bayou Texar
Study (in the Administrative Record), the area where the Agrico
ground-water plume has impacted the porewater in the bayou
sediments is clearly defined and limited in areal extent. In
addition, it can be assumed that the City is aware of other
sources impacting the Bayou Texar sediments. EPA will continue
to cooperate with and provide site information to local
regulatory authorities.
18). The ROD for OU1 makes repeated reference to a "slurry wall"
that will be placed around the Capped area on site. It appears
that the slurry wall is being presented as a strictly "cosmetic"
remedy, rather than one that is based oh scientific appraisal.
EPA Response; The Phase II Remedial Investigation (RI) Report,
Geraghty & Miller, Inc., August 1992 provides detailed scientific
evidence of the presence of a "hardpan" and resultant perched
water zones at the Agrico site. In addition, field observation
has confirmed the presence of the hardpan on site. Therefore,
EPA views the slurry wall as a necessary and appropriate part of
the remedial action for the site. However, additional further
geophysical investigations are presently being conducted as part
of the remedial design activities. The purpose of the
geophysical studies is to provide additional information in
determining an appropriate configuration for the proposed slurry
wall.
(The remaining questions/ comments were presented verbally at the
B-ll
-------
Proposed Plan Public Meeting held on February 18, 1994 in
Pensacola,.Florida.)
19). What is the hazardous ranking score for the site?
EPA Response; The Hazardous Ranking Score for the Agrico
Chemical Superfund Site is 44.98.
20). Do you know where the community is located in relation to
the site? Basically, you are stating that you are not going to
clean up the ground water (not do anything) because you don't
know what is there (in the ground water). What are you going to
do to help these people relocate?
EPA Response; EPA is aware of where the community is located in
relation to the Agrico site. Based on the extensive scientific
data that has been generated (Administrative Record - Pensacola
Regional Library) during the field studies, soils and sludges
requiring treatment are confined to within the site property
boundaries. Warning signs have been posted and a security fence
has been constructed to ensure that no one is exposed to the
site's contaminated soil. Therefore, the Agency does not believe
relocation of citizens is appropriate.
However, the Risk Assessment prepared for the OU1 ROD (treatment
of the soils) indicates that treatment of the soils on site is
necessary to protect the public health and the environment. The
remedial action (cleanup) of the soils can begin once the
appropriate technical (Remedial Design and Remedial Action Work
Plan) documents have been finalized.
In response to the statement that the ground-water contamination
related to the site will not be treated because the scientists
don't know what is there, can be addressed in the following
manner. EPA knows the extent and nature of the contaminants in
ground water related to the Agrico site. Through regularly
scheduled well sampling by the county and state, we know that
there are several other contaminant plumes not related to the
Agrico site in the immediate area. What we do not know is the
nature and extent of each of these other plumes.
Because of the presence of these other contaminant plumes, the
high pumping rate required to extract the Agrico plume will cause
further degradation of the aquifer through contaminant
"smearing". In addition, the high pumping rate will most likely
cause salt water intrusion into the extraction wells due to the
close proximity to the brackish (fresh and salt water
combination) Bayou Texar. Therefore, our rationale for not
actively treating Agrico ground-water contamination is based on
the realization that to pump and treat the Agrico plume will
create more ground-water problems that it will resolve. Put
another way, EPA views treatment of the Agrico ground-water
B-12
-------
contamination to be less protective than the limited action
remedy preferred by EPA.
Finally, a well survey has been conducted in the area of
contamination. The well survey indicates that no one is drinking
water from the portion of the aquifer that is contaminated by the
site.
21). Did you take into account that heavy rain, heavy snows
north of here is going to cause a big flow of underground water
moving in the aquifer once all this melts?
EPA Response; The engineers and scientists studying the Agrico
ground water contamination did take into account rain fall
measurement in the Pensacola area and its affect on the aquifer
(i.e., water table levels, recharge rates). In addition, study of
the soils and permeability of the aquifer is evidenced in the
Remedial Investigation (RI) Report (Geraghty & Miller, November
1993). The nearest heavy snow fall location this past winter
might arguably be Louisville, Kentucky. EPA does not believe the
heavy snow fall in that nearest northern area or areas north of
there would have an effect on the aquifer in the Pensacola area.
22). A citizen stated: We understand the extent of time it
takes to clean up a pond, but are you going to let the people
that live right there at point zero, within three hundred feet of
the site, die, because you want to leave them there, or are you
going to allow the personnel back here to relocate? They don't
want to die.
EPA Response; The results of the sampling analyses are
represented in Figures 3-2 through 3-7, and Tables 2-1 through 3-
6 of the Final Phase II Remedial Investigation Report. Geraghty &
Miller, November 1993. These extensive tables reflect off-site
soil analyses that indicate contaminant levels generally below
laboratory detection limits or well below human health based
action levels. Therefore, based on the above information EPA does
not believe relocation of citizens is appropriate.
The Remedial Investigation (RI) of the site soil and ground-water
contamination has been ongoing for the past three years. The
study was conducted in two phases. The purpose of the extensive
field study was to determine the nature and extent of
contamination related to the Agrico site. In the first phase,
all surficial and subsurface samples were analyzed for site
contaminants to include total fluoride, chloride, '
nitrate/nitrite, and sulfate. In addition, samples were analyzed
for metals (i.e., lead, copper, chromium, arsenic).
During field investigations, surficial and subsurface samples
were taken in the neighborhoods south of the site, in the school
yard south of the site, and finally adjacent to and east of
B-13
-------
Interstate 1-110 (see previous comment 15). In addition to the
above locations, a sample was taken as a "control sample"
adjacent to Hermann and Pearl streets and west of the CSX
railroad yard. Based on site information from the above photos,
this location was believed to be a "control sample", or a sample
obtained from an area that had not been impacted by the site.
The rationale for sampling in these locations was based on
information provided through aerial photographs of the site
dating back to 1940.
During the second phase of the investigation, both on-site and
off-site surficial and subsurface soil samples were collected for
analysis of fluoride, lead, organic volatile compounds, semi-
volatile organic compounds, and organochlorine pesticides.
Over 175 surficial and subsurface soil samples were taken on-site
and off-site during the site field investigations. Analyses of
these many locations indicate the site's severe soil
contamination occurs within the former impoundment (sludge) ponds
and decreases as one moves out from (away from) these sludge
locations. In addition, the contaminants targeted in the OU1 ROD
(fluoride, lead, and arsenic) requiring treatment are confined to
on-site soils. Similarly, contaminated soils within and near the
PFP IV area contaminated with fluoride, lead, and the pesticide
dieldrin are confined to on-site locations.
23). A resident asked: Are you aware of all the other chemicals
that the people in these affected communities are coming in
contact with, besides the fluoride, all other chemicals (from
other sources) that are affecting this community? Do you take
that into account? We have lived there for years, for years,
Pearl Street, Herrman Street. There isn't a home on that street
where cancer deaths are not prevalent. There are about forty
homes on those two streets at the present time. There were more
in times past. There isn't a home on that street where cancer
deaths are not prevalent. You should have been able to live
there when that plant was in operation. There were times when
you could not come out of your door. Your eyes burned. Your skin
itched. You had trouble breathing, and people had eye problems.
EPA Response; The Agrico Chemical plant, like many contaminant
sources throughout the country, was not environmentally regulated
for many years prior to the creation of state and federal
environmental protection agencies. In addition, it is not
possible to accurately determine potential human health impacts
from past exposure to chemicals from many sources. Consequently,
EPA is charged with ensuring that one specific facility or site
(in this case, the Agrico site) is thoroughly investigated and
subsequently cleaned up. As part of the Superfund process, the
Agency takes into account what action is protective of the public
health and the environment for the present and the future in
B-14
-------
regards to the one specific site.
As you may be aware, there are other ways to determine if
citizens are being adversely affected by exposure to
contamination through various means. For example, if a person is
concerned that a private ground-water well is contaminated, that
person may request that the county health unit test the well. In
addition, citizens can contact the State of Florida's Department
of Health and Rehabilitative Services (HRS) or the federal Agency
for Toxic Substances and Disease Registry (ATSDR) to request
information on health concerns and, if a situation warrants,
request health studies be conducted.
24). A citizen stated: We have not had fair treatment. Our
technical advisor (under the Technical Assistance Grant) has
stated that the ROD (OU1) clearly establishes cleanup goals for
fluoride, lead, and arsenic only. However, some site soils
contain only dieldrin (a pesticide) at high enough levels to pose
an unacceptable risk. Such soils would not be excavated,
solidified, or placed under the cap.
EPA Response: A total of 20 surficial soil samples and 56
subsurface soil samples (totaling 76) were collected and analyzed
for organochlorine pesticides and polychlorinated biphenols from
on-site and off-site locations. Only four soil samples from
three separate boring locations detected any compound above the
laboratory detection limits. The compound dieldrin, used prior
to 1974 for insect control, was detected in two soil sample from
on-site boring located within the area of PFP I and PFP
II(northern part of site) at concentrations of 0.2
microgram/kilogram (ug/kg) or 0.2 parts per billion (ppb) and 6.9
ppb, respectively. Dieldrin was also detected in the surficial
soil and subsurface soil samples from one off-site boring located
south of the site (south of Texar Drive) with concentrations of
2.4 ppb and 0.1 ppb, respectively.
Based on current EPA risk assessment guidance, levels of dieldrin
ingested in excess of 40 ppb for someone living on the site and
180 ppb for an on-site worker pose an unacceptable risk to human
health. Since the highest level of dieldrin detected at the site
was 6.9 ppb, the soils containing the dieldrin do not pose an
unacceptable risk through ingestion. Although evaluated in the
risk assessment for the site, dieldrin is not assigned a remedial
goal (cleanup number in parts per billion) because of its
isolated occurrence and low detection levels.
25). One retired resident stated: If you'll start at Palafox
Street (near Escambia Wood Treating site) and come from Palafox
to that railroad, you'll see cancer deaths in large numbers. In
my household only, just one household, there was my mother, my
father, my uncle and another one from a heart attack. I had a
baby born with respiratory problems and one stillborn. You
B-15
-------
cannot tell me that soil, that those contaminations did not
affect the.community past that railroad. I will never believe
that.
EPA Response; Thirty-five historic areal photographs dating from
December 1940 through February 1990 provide a detailed record of
the evolution of the impoundment ponds (those areas with the
highest contaminated soils and sludges) due to plant processes at
the site. These chronological photographs indicate that plant
processing never occurred west of the CSX railroad tracks. In
addition, soils samples taken west of the railroad track and in
various locations on the western portion of the site confirm that
elevated, unacceptable levels of contamination are predominantly
confined to the eastern part of the site and within site
boundaries. However, as part of the site investigation, EPA
conducted a risk assessment. The risk assessment results help
the Agency determine what, if any, health effects the site
currently, and in the future, pose to the public. Based on the
risk assessment conducted, EPA views treatment of contaminated
soils on-site as the appropriate response to protect the public
health and the environment.
26). The above resident also stated: We're concerned that you
are not going to keep in contact with the community. We are
concerned that you are going to come in here and do the same
thing to us that they did at the other site (Escambia Wood
Treating site).
EPA Response: EPA is committed to keeping the public informed of
what's happening at the Agrico Chemical Site. The fact sheets
distributed for this site contain a toll-free number to be used
by the citizens to telephone EPA at any time. The cleanup of the
soils at Agrico is tentatively scheduled to begin around June of
1995. The treatment of soils and construction of the cap will
take one to two years to complete. However, several months
before construction is to begin, EPA will issue a fact sheet
announcing that the design of cleanup is complete. In addition,
citizens will be notified that the actual onsite construction
work is beginning. Finally, stringent environmental regulations
will be followed during construction to protect the health of the
surrounding community.
27). A resident stated: The Administrator of EPA has said that
this area is a substantial threat to the residents living here.
Also, we ask you to take a look at the economic degradation that
is associated with the excavation and the cleanup. It's twofold.
Not only are they exposed to the toxins and all types of
hazardous wastes, but economically no one — the value of the
property has depreciated, and they can't get loans or anything
for repair. We're killing them in two ways, and I feel that if
you're really concerned that you'll look at the economic
disproportion, and they should be moved.
B-16
-------
The president of the United States has signed an executive order,
Al Gore has promised that they were going to stop environmental
injustice on neighborhoods such as ours that are close to sites
that's being adversely affected by toxins on those sites. They
promised that environmental racism, environmental inequalities,
genocides would no longer occur. Our children are going to die
year after year. We want removed, we want relocation.
EPA Response; The Administrative Record for the Agrico Chemical
Site which resides in the Pensacola Regional Library in downtown
Pensacola consists of ten volumes of data and analysis supporting
the Agency's Record of Decision for Operable Unit One (soils) and
this Record of Decision for Operable Unit Two (ground water).
EPA has carefully studied the data and subsequently arrived at a
protective environmental solution to the contamination related to
the Agrico Chemical site.
Follow-up discussions with the Environmental Justice Office
personnel at EPA Region IV, EPA staff members assigned to the
Agrico Chemical site and the neighboring Escambia Wood Treating
Site, and an environmental justice activist residing in Atlanta
have continued subsequent to the Agrico Chemical Site public
meeting of February 18, 1994. In addition, the environmental
justice personnel have conducted interviews with citizens living
near the site. Citizens' request to repair fencing and post more
warning signs has been completed. EPA staff has provided verbal
and written notification of current and upcoming activities at
the site to the leader of GATE. In conclusion, EPA
representatives are exploring ways to create an environmental
review board to include EPA staff, citizens, and a mediator to
continue open dialogue with the citizens and to address concerns
regarding the Agrico Chemical and Escambia Wood Treating sites.
28). An attorney associated with CATE stated: Something that
happened in the past is over with. We can't go back on that.
The air pollution and everything that took place. We know the
contaminants are there. We know that something should be done in
the cleanup, both in unit one and in unit two, but the primary
concern is the health of the community is the primary thing, you
know, that the cleanup takes place, that nothing else happens
either in the air or in the soil, or in the grbundwater. If
you'll take that back with you and take this data (first 12
questions/comments of this responsiveness summary) and look at
it, then there is no use in us berating and putting a lot in the
record. Thank you.
EPA Response; The public health is the primary concern of EPA.
We will take this data back and make it part of the record. More
importantly, we will address each of these written concerns as
well as those voiced at this meeting. All questions and comments
will be carefully considered before EPA makes a decision for
cleanup of the ground water at this site.
B-17
-------
29). A citizen stated: I stay in the neighborhood over there.
If there's nothing wrong with the place over there, why are you
dumping so-much money in that hole? Why do you go to so much
trouble to put a fence up around the hole?
EPA Response; We apologize if there is confusion about what
areas are contaminated and which are not. The "hole" or the
impoundment pond (PFP II) is where the majority of gross
contamination occurs on site. The parties responsible for the
cleanup of the site will proceed to cleanup the contaminated
soils on site. In addition, cleaning the soils on site will
eliminate continued contamination from the site to the ground-
water. The fence is there, as you indicated, to prevent anyone
from accidentally coming in contact with soil contamination on
site.
30). A resident asked: If Escambia County Utility Authority
(ECUA) has provided such clean water, why is it that Pensacola's
water has the highest lead level of any city with the exception
of one in South Carolina? Why have some people been told by ECUA
that phenol is in their water? What's happening? The
contamination is getting there somewhere.
EPA Response: The hydraulic properties of the Sand-And-Gravel
Aquifer provide Escambia County with a valuable drinking water
resource. However, these same hydraulic properties render the
aquifer especially vulnerable to varied land use. The land use
in Escambia county is very diversified. The county's history
indicates industry, military, and residential use co-existing
side by side. Unfortunately, unregulated past land use by all of
the above sources has resulted in an aquifer with a variety of
contaminants from multiple sources.
30). How current is the ground-water data that you are using?
1992? The groundwater flow, contaminant flow, how confident are
you that the groundwater flow is in the same direction?
EPA Response: In the summer of 1993, the Florida Department of
Environmental Protection requested four additional wells be
installed down-gradient from the site and adjacent to the Bayou
Texar to ensure that previous characterization of the Agrico
plume was still current. The results of the sampling of these
new wells was made available to the EPA in November of 1993. The
results of this sampling indicated that the direction and
contaminant levels in the Agrico ground-water plume previously
defined were still accurate and current.
31). Have you installed wells north of the site to see if
there's contamination?
EPA Response: An extensive monitoring well network was
established to define the extent of the ground-water
B-18
-------
contamination associated with the site. Sampling both existing
and newly installed wells north and south of the site ensured the
accurate characterization of the Agrico ground-water
contamination. In addition, continued routine ground-water
monitoring by local water authorities has not indicated
contamination from this site outside the delineated Agrico
ground-water plume. In conclusion, extensive ground-water
monitoring will continue as part of the limited action selected
in this ROD for OU2.
32). Is the fluoride content in the Bayou Texar much higher than
the fluoride content in the aquifer?
EPA Response; The highest fluoride level found in the aquifer is
98 milligrams per liter (mg/L) or 98 parts per million (ppm).
The highest level of fluoride found in the porewater (water found
between grains of sand in the bayou's bottom sediment) is 250
mg/L or 250 ppm. , Because ground-water modeling (conducted for
study of fate and transport of fluoride) indicates that fluoride
contamination first reached the bayou as early as 1950, one can
expect to find higher levels "deposited" in the bayou bottom
sediments.
However, the fluoride levels found in the bayou's porewater (250
ppm) are not considered to be significantly higher than the
fluoride in the aquifer, but rather is consistent with what can
be expected. In addition, the dilution calculations conducted to
predict how much fluoride is present in the bayou's surface water
indicates that a level of less than one part per million (1 ppm)
of the contaminant is biologically available for uptake in the
bayou surface waters. Finally, to ensure that the dilution
calculations accurately depict the <1 ppm level of fluoride
expected in the bayou's surface water, analyses of surface water
samples of Bayou Texar will be part of the limited remedial
action for the site's ground-water contamination.
33). A citizen stated: In looking at the chloride level that
the EPA measured in the cluster of wells quite close together,
the chloride level in the water varied by a factor of one
milligram per liter to ninety-five or a hundred. I can't imagine
a hundred to one gradient in this relatively shallow sampling
area. Is the data in error? On repeated sampling of a given
well, at a given location, what is the precision and how much
does that fluoride level change, for instance?
EPA Response; Sampling of AC-3 located directly downgradient of
the site's impoundment ponds has consistently indicated levels of
contamination that are similar (i.e., fluoride found at 98 ppm on
one occasion and 90 ppm for a subsequent sampling event). The
lower levels found are a result of dispersion or diffusion of the
fluoride in the aquifer, in addition, the lower levels of
fluoride found in several wells further downgradient verify the
B-19
-------
presence of lower levels of fluoride as it travels further away
from the source (on-site impoundment ponds).
B-20
------- |