United States
           Environmental Protection
           Agency
              Office of
              Emergency and
              Remedial Response
EPA/ROD/R01-89/039
September 1989
&EPA
Superfund
Record of Decision
            Kellogg - Deering
            Well Field, CT

-------
50272-101
 REPORT DOCUMENTATION
        PAGE
1. REPORT NO.
     EPA/ROD/R01-89/039
                                           3. Recipient's Accession No.
 4. Title and Subtitle
   SUPERFUND RECORD OF  DECISION
   Kellogg-Deering Well Field, CT
   Second  Remedial Action
                                           5. Report Date
                                              09/29/89
                                           6.
 7. Author(s)
                                                                     8. Perfoiming Organization Rept No.
 9. Performing Organization Name and Addreoa
                                           10. ProiectfTask/Worlt Unit No.
                                                                     11. Contract(C) or Grant(G) No.

                                                                     (C)

                                                                     (G)
 12. Sponsoring Organization Name and Address
   U.S. Environmental Protection  Agency
   401 M Street,  S.W.
   Washington, D.C.   20460
                                           13. Type of Report A Period Covered

                                                800/000
                                                                     14.
 15. Supplementary Notes
 16. Abstract (Limit: 200 words)

  The  Kellogg-Deering Well Field site is  a  10-acre contaminated municipal well field  in
 Norwalk,  Fairfield County, Connecticut,  which includes  adjacent  source areas  that
 contribute to the  well field  contamination.   The site  lies within  the Norwalk River's
 100-year  floodplain and consists of four municipal drinking water  wells that  supply
 approximately 25 percent of the water  for  the 45,000  residents in  Norwalk.  The city
 detected  elevated  TCE levels  in the wells  during routine monitoring in 1975 and
 subsequently shut  down unacceptably contaminated wells  until a redwood slat treatment
 system  was installed in 1981.   EPA conducted a remedial investigation between 1984 and
 1986  leading to a  1986 Record of Decision  (ROD)  which  required the city to operate an
 air stripping unit.   EPA conducted a subsequent remedial investigation in 1987 which
 further defined the contaminated ground  water area and  identified  the Elinco/Pitney
 Bowes/Motleis Court Complex,  located east  of the site and upgradient of the well field,
 as a  source of ground water contamination.    This ROD,  the second  of three planned
 operable  units, addresses the major source  area.  The primary contaminants of concern
 affecting the soil and ground water are  VOCs including  benzene,  PCE,  TCE, toluene, and
 xylenes.
 (See  Attached Sheet)

 17 Document Analysis a. Descriptors
    Record of  Decision  - Kellogg-Deering Well Field,  CT
    Second Remedial Action
    Contaminated  Media: soil,  gw                                                     .
    Key Contaminants: VOCs  (benzene, toluene, PCE, TCE,  xylenes)
   b. Identifiers/Open-Ended Terms
   c. COSATI Field/Group
 18. Availability Statement
                            19. Security Class (This Report)
                                   None
                                                      20. Security Class (This Page)
                                                     	None	
21. No. of Pages
    124
                                                                                22. Price
(See ANSI-Z39.18)
                                      See Instructions on Reverse
                                                      OPTIONAL FORM 272 (4-77)
                                                      (Formerly NTIS-35)
                                                      Department of Commerce

-------
                   DO NOT PRINT THESE INSTRUCTIONS AS A PAGE IN A REPORT


                                                     INSTRUCTIONS
Optional Form 272, Report Documentation Pag* la based on Guideline* for Format and Production of Scientific and Technical Reports,
ANSI 230.18-1974 available from American National Standards Institute, 1430 Broadway, New York, New York 10018.  Each aeparately
bound report—for example, each volume in a muftlvolume set—shall have Ita unique Report Documentation Page.

  1.  Report Number. Each Individually bound report shall carry a unique alphanumeric designation assigned by the performing orga-
     nization or provided by the sponsoring organization In accordance with American National Standard ANSI Z39.23-1974, Technical
     Report Number (STRN). For registration of report code, contact NTIS Report Number Clearinghouse, Springfield, VA 22161. Uae
     uppercase letters, Arabic numerals, slashes, and hyphens only, aa In the following examples: FASEB/NS-75/87 and FAA/
     RD-75/09.
  2.  Leave blank.

  3.  Recipient's Accession Number. Reserved for use by each report recipient

  4.  TNto and Subtitle. Title should Indicate clearly and briefly the subject coverage of the report, subordinate subtitle to the main
     title.  When a report la prepared In more than one volume, repeat the primary title, add volume number and Include subtitle for
     the specific volume.

  5.  Report Data. Each report ahall carry a date Indicating at leaat month and year. Indicate the basis on which it waa selected (e.g.,
     date of Issue, date of approval, date of preparation, date published).

  6.  Sponsoring Agency Code. Leave blank.

  7.  Authors).  Give name(s) In conventional order (e.g., John R. Doe, or J. Robert Doe). List author's affiliation If It differs from
     the performing organization.

  8.  Performing organization Report Number.  Insert If performing organizaton wlahea to assign this number.

  9.  Performing Organization Name and Mailing Addreaa. Give name, street, city, state, and ZIP code. Uat no more than two levels ol
     an organizational hlerachy. Display the name of the organization exactly aa It should appear In Government Indexea such as
     Government Reports Announcements & Index (GRA & I).

 10.  Project/Task/Work Unit Number. Use the project, task and work unit numbers under which the report waa prepared.

 11.  Contract/Grant Number. Insert contract or grant number under which report waa prepared.

 12.  Sponsoring Agency Name and Mailing Address. Include ZIP code. Cite main sponsors.

 13.  Type of Report and Period Covered. State Interim, final, etc., and, If applicable, inclusive datea.

 14.  Performing Organization Code. Leave blank.

 15.  Supplementary Notes.  Enter information not Included elsewhere but useful, such aa: Prepared In cooperation with... Translation
     of... Presented at conference of... To be published in... When a report is revised, Include a statement whether the new
     report supersedes or supplements the older report.

 16.  Abstract Include a brief (200 words or less) factual summary of the most significant Information contained in the report.  If the
     report contains a significant bibliography or literature survey, mention it here.

 17.  Document  Analysis, (a).  Descriptors. Select from the Thesaurus of Engineering and Scientific Terms the proper authorized terms
     that Identify the major concept of the research and are sufficiently specific and precise to be uaed aa Index entries for cataloging.

     (b).  Identifiers and Open-Ended Terms. Use Identifiers for project names, code names, equipment designators, ale. Use open-  -
     ended terms written in descriptor form for those subjects for which no descriptor exists.

     (c).  COSATI Field/Group.  Field and Group assignments are to be taken form the 1964 COSATI Subject Category Uat Since the
     majority of documents are multldlsciplinary In nature, the primary Field/Group assignment(s) will be the ape rifle dtedpilne,
     area of human endeavor, or type of physical object  The applications) will be cross-referenced with secondary Reid/Group
     assignments that will follow the primary poatlng(s).

18. Distribution Statement Denote public releaaabillty, for example -Release unlimited", or limitation for reasons other than
    security. Cite any availability to the public, with address, order number and price, if known.

19. & 20.  Security Classification. Enter U.S. Security Classification In accordance with U. S. Security Regulations (i.e., UNCLASSIFIED).

 21.  Number of pages. Insert the total number of pages, Including introductory pages, but excluding distribution list If any.

22.  Price.  Enter price In paper copy (PC) and/or microfiche (MF) It known.

 A GPO   I983 0 -  381-526(8393)                                                                       OPTIONAL FORM 272 BACK
                                                                                                   (*-77)

-------
EPA/ROD/R01-89/039
Kellogg-Deering Well Field, CT
Second Remedial Action

16.  Abstract (Continued)

 The selected remedial action for this site includes onsite in situ vacuum extraction
followed by carbon adsorption to remediate soil underlying the major source area at the
court complex; ground water pumping and onsite treatment using air stripping followed by
offsite discharge; and periodic sampling of ground water,  soil vapor, and indoor air.
The estimated present worth cost for this remedial action is $9,100,000, which includes a
present worth O&M cost of $3,034,000 for 30 years.

-------
                        RECORD OF DECISION
                      DECLARATION STATEMENT
            KELLOGG-DEERING WELL FIELD 8UPERFUND SITE
                 OPERABLE UNIT 2 - SOURCE CONTROL

                          September 1989

SITE NAME AND LOCATION

Kellogg-Deering Well Field Superfund Site
Norwalk, CT

STATEMENT OF PURPOSE

This Decision Document presents the selected remedial action for
the second operable unit for source control of this National
Priorities List site 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 Contingency Plan, 40 C.F.R. Part 300 et
seq. (1987).  The Regional Administrator for Region I of the
United States Environmental Protection Agency (EPA) has been
delegated the authority to approve this Record of Decision.

The State of Connecticut has concurred on the selected remedy and
has determined that it is consistent with Connecticut laws and
regulations.

STATEMENT OF BASIS

This decision is based on the Administrative Record for the site
developed in accordance with Section 113(k) of CERCLA.  The
attached index identifies the items that comprise the
Administrative Record.

The Administrative Record is available for public review at the
Norwalk Public Library and the EPA Region I Waste Management
Division Records Center at 90 Canal Street in Boston,
Massachusetts.

DESCRIPTION OF THE SELECTED REMEDY

The response action for the second operable unit at the Kellogg-
Deering site is a comprehensive remedy for the second operable
unit that combines components of source control and management of
migration within an area of high contaminant concentrations.  The
source control component will include an in-situ soil vacuum
extraction system to remediate contaminated soils underlying
buildings and parking lots at the Elinco/Pitney Bowes/Matheis
Court Complex.  Soils will be cleaned to levels that will protect
the quality of groundwater in the event that the buildings and
parking lots are removed at some time in the future.

-------
The management of migration component consists of groundwater
extraction, treatment and discharge within a highly contaminated
portion of the aquifer known as the Source Area. The aquifer is
composed of unconsolidated overburden and bedrock.  Groundwater
will be pumped with the objective of achieving Safe Drinking
Water Act Maximum Contaminant Levels in the aquifer.
DECLARATION

The selected remedy is protective of human health and the
environment and attains Federal and State requirements that are
applicable or relevant and appropriate.  In addition, the
selected remedy is cost effective; satisfies the statutory
preference for remedies that employ treatment that reduces
toxicity, mobility, or volume as a principal element; and
utilizes permanent solutions and alternative treatment
technologies to the maximum extent practicable.


The selected remedy will result in hazardous substances remaining
on-site above health based levels.  Therefore, a review will be
conducted within five years after commencement of remedial action
to ensure that the remedy continues to provide adequate
protection of human health and the environment.
                                  Paul G. Keough,
                                  Acting Regional Administrator
                                  EPA Region I
                                ii

-------
                    RECORD OF DECISION SUMMARY
            KELLOGG-DEERING WELL FIELD 8UPERFUND SITE
                 OPERABLE UNIT 2 -  SOURCE CONTROL

                          September 1989

                             CONTENTS

Declaration	..	..  i

Location and Description	  1

Site History	..-....*	  2
     Response History	..	  2
     Enforcement History	  3

Community Participation	  3

Scope and Role of Operable Unit	  4

Site Characteristics	  5
     Groundwater  	...	  5
     Soil	  7
     Indoor Air	  8

Summary of Site Risks	  8
     Groundvater	  8
     Soil		 12

Documentation of Significant Changes	 12

Development and Screening of Alternatives	 16
     Statutory Requirements/Response Objectives	 16
     Technology and Alternative Development and Screening .... 17

Description of Alternatives 	 	 19
     Alternative I		 19
     Alternative II	 20
     Alternative III	 21
     Alternative IV	 22
     Alternative V	 22
     Alternative VI	... 23
     Alternative VII	 25
     Alternative VIII	 25

Selected Remedy	 25
     Remedial Action Objectives/Cleanup Goals	 26
          Groundvater	 26
          Soil	 27
     Description of Components of the Selected Remedy	 28
          Contaminated Soil Treatment 	 28
          Contaminated Groundwater Extraction and Treatment .. 29

                               iii

-------
     Rationale for Selection	 31

Summary of Comparative Analysis of Alternatives	 32
     Overall Protection of Human Health and the Environment... 32
     Compliance with ARARs....	 33
     Long-term Effectiveness and Permanence	 34
     Reduction of Mobility, Toxicity or Volume 	 35
     Short-term Effectiveness.	 35
     Implementability	 36
     Cost	 36
     Community Acceptance	 37
     State Acceptance	 37
     Conclusion	 37

Statutory Determinations	 38
     Protection of Human Health and the Environment 	 38
     Compliance with ARARs	 39
     Cost Effectiveness 	 41
     Permanent Solution/Alternative Treatment/
     Resource Recovery 	 41
     Preference for Treatment	»	 42

State Role	 42
                             FIGURES

1.   General Location Map	 43
2.   Kellogg-Deering Well Field Superfund Site Map	 44
3.   General Cross Section Diagram 	 	 44
4.   Area of Groundwater Contamination	 45
                              TABLES

1.   Indicator Chemicals 	 46
2.   Summary of Remedial Alternatives 	 47
3    Groundwater and Soil Cleanup Goals 	 48
4.   Estimated Costs for the Selected Alternative 	 49
5.   Comparison of Estimated Costs for Alternatives 	 50
6.   ARARs for the Selected Alternative 	 51
                            APPENDICES

Responsiveness Summary	Appendix A
Administrative Record Index 	 Appendix B
State Concurrence Letter	Appendix C
                                iv

-------
                    RECORD OF DECISION SUMMARY
            KELLOGG-DEERING WELL FIELD 8UFERFUND SITE
                 OPERABLE UNIT 2 - SOURCE CONTROL

                          September 1989


LOCATION AND DESCRIPTION
     The Kellogg-Deering Well Field Superfund Site is located in
Norwalk, CT.  The "site" consists of an approximately 10-acre
municipal well field  (the Kellogg-Deering Well Field) and
adjacent areas that contribute to the well field contamination.
The well field portion of the site (Figure 1) is located along
the western bank of the Norwalk River in southwestern Fairfield
County, Connecticut and is bordered on the east by the Norwalk
River, on the north by residences along Broad Street, on the west
by residences along Lakeview Avenue, and on the south both by
wooded acreage and residences in the vicinity of East Lakeview
Drive and Nutneg Place.

     The site, considered to be within the study area shown in
Figure 2, also includes light industrial and residential areas on
the east side of the Norwalk River.  The 100 year floodplain of
the Norwalk River encompasses the Kellogg-Deering Well Field and
low-lying bands along the banks of both sides of the river,
including portions of the site.

     The Kellogg-Deering Well Field, which is owned and operated
by the Norwalk First Taxing District (NFTD) Water Department,
consists of four municipal water supply wells that supply
approximately 25 percent of the water for the 45,000 residents in
the city of Norwalk. The estimated safe yield of the Kellogg-
Deering Well Field is estimated to be approximately 3.5 million
gallons per day on a long-term basis as reported in Safe Yield
Evaluation of the Kelloqg-Deerinq Well Field. Norwalk. CT by
Leggette, Brashears and Graham in 1981.   The water supply wells
are located in an aquifer that is classified as II-A under the
EPA Ground Water Protection Strategy and GAA under the State of
Connecticut's Water Quality Standards.  EPA's II-A classification
means that the groundwater is currently and potentially a source
of drinking water and other beneficial uses.  The Connecticut GAA
classification encompasses the immediate vicinity of the Kellogg-
Deering Well Field and indicates that the Well Field is an
existing or potential public drinking water supply.   The aquifer
outside of the immediate vicinity of the Well Field is classified
by Connecticut as GB/GA which indicates that although the
groundwater presently may not be suitable for direct human
consumption without treatment due to chemical contaminants or
land use impacts, the State maintains a goal to restore the

-------
ground water to drinking water quality.  A more complete
description of the  site can be found in Section 1 of the
supplemental Remedial  Investigation Report (RI).

     the site is underlain by two water-bearing geologic
formations:  overburden and bedrock (Figure 3).  The overburden
aquifer, composed of primarily sand and gravel, is very thick
near and under the  Norwalk River (approx. 100 feet) and tapers
out toward the east.   The depth to the water table varies from
between zero feet to approximately 35 feet below ground surface.
All four of the Kellogg-Deering water supply wells are located in
the overburden aquifer.  The bedrock aquifer underlies the
overburden aquifer  and is closer to the surface as the overburden
aquifer thins toward the east.  The bedrock aquifer is composed
of fractured rock.  Groundwater movement in the bedrock aquifer
occurs through interconnected fractures.

     Operation of the  water supply wells at the Kellogg-Deering
Wells Field effects the local hydrology in two important ways.
First, some surface water from the Norwalk River is drawn
downward from the Norwalk River as it passes near the Well Field.
Secondly, water from portion of the aquifer located to the east
side of the river is drawn underneath the river bed before
entering the operating supply wells.  As a result, approximately
86 percent of the water entering the supply wells originates as
surface water from  the Norwalk River while most of the remaining
14 percent comes from  the aquifer.


SITE HISTORY

Response History

     Elevated levels of trichloroethene (TCE) were first detected
in the wells at the Kellogg-Deering Well Field during routine
monitoring by the Norwalk City Health Department in 1975.  The
wells with unacceptable concentrations of TCE were shut down
until a redwood slat treatment system was installed by NFTD in
1981.  The site was placed on the National Priorities List (NPL)
in 1984.  An initial Remedial Investigation and Feasibility Study
(RI/FS) was conducted  at the site between 1984 and 1986.  The
initial RI/FS supported the first operable unit Record of
Decision (ROD) issued  in September 1986, that required completion
of construction and operation of a new air stripping facility
(well head treatment)  at the Kellogg-Deering Well Field.  The
well head treatment removes TCE and other volatile organic
compounds (VOCs) from  the contaminated groundwater prior to
discharge into a conventional water treatment plant and the water
distribution system.   The well head treatment was designed to
supply safe, potable water to the public.

-------
     A supplemental RI/FS began at the site in 1987 to provide
further  information regarding the source(s) and extent of
groundwater contamination at the site.  In addition to better
defining the area of groundwater contamination found during the
initial  RI, the supplemental RI also identified a major source
area of  groundwater and soil contamination at the Elinco/Pitney
Bowes/Hatheis Court Complex (the complex) located at 272 and 282
Main Avenue in Norwalk.  The supplemental RI concluded that the
contamination at the complex is contributing to the contamination
at the Kellogg-Deering Well Field and the aquifer supplying the
Well Field.  The supplemental RI also identified soil gas
contamination sources at the complex and provided an analysis of
indoor air quality.  The complex is considered to be a major soil
and groundwater contamination source area.  A more detailed
description of the site history can be found in the supplemental
RI.
Enforcement History

Since July 29, 1986, EPA has notified approximately nine parties
who are either past or present owners or operators of the
facility of their potential liability with respect to the site.
In 1986, EPA issued an administrative order to the Norwalk First
Taxing District  (NFTD) ordering the NFTD to complete construction
of and begin operation of an existing air stripping at the
Kellogg-Deering Well Field.  By letter dated May 11, 1988, EPA
notified NFTD that it had satisfied the requirements of the
order.

     The potentially responsible parties (PRPs) have been active
in the remedy selection process for the site.  Comments were
submitted by the PRPs during the public comment period.  A
summary of these comments and EPA's responses to these comments
are included in the Responsiveness Summary in Appendix A.  The
PRPs1 comments, the summary of the comments and EPA's responses
to these comments are all included in the administrative record.

     Special notice has not been issued in this case to date.


COMMUNITY PARTICIPATION

     Through the Site's history, the level of community concern
and involvement has fluctuated.  EPA has kept the community and
other interested parties apprised of the Site activities through
informational meetings,"fact sheets, press releases and public
meetings.

-------
     EPA held several public meetings and hearings in connection
with the first operable unit for the site.  Details of community
relations activities in connection with the first operable unit
are described in the first operable unit Record of Decision.
In June 1986, EPA released a community relations plan which
outlined a program to address community concerns and keep
citizens informed about and involved in activities during
remedial activities.  In January 1988, EPA issued a press release
announcing the completion of the Supplemental RI/FS workplan.

     On October 31, 1988, EPA made the administrative record
available for public review at EPA's offices in Boston and at the
Norwalk Public Library.  The administrative record was updated
with information relating to the second operable unit at both
locations on July 26, 1989.  EPA published a notice and brief
analysis of the Proposed Plan in Norwalk Hour on July 21, 1989.
On July 26, 1989 the Proposed Plan was made available to the
public at the Norwalk Public Library.

     On July 26, 1989, EPA held an informational meeting to
discuss the results of the Supplemental Remedial Investigation
and the cleanup alternatives presented in the Supplemental
Feasibility Study and to present the Agency's Proposed Plan for
remedial actions for the second operable unit.  Also during this
meeting, the Agency answered questions from the public.  From
July 27, 1989 to August 25, 1989, the Agency held a 30-day public
comment period to accept public comment on the alternatives
presented in the supplemental Feasibility Study and the Proposed
Plan and on the other documents which were a part of the
administrative record for the site.  On August 14, 1989, the
Agency held a public hearing to accept any oral comments.  A
transcript of this meeting and the comments and the Agency's
response to comments are included in the attached Responsiveness
Summary.

     This decision document presents the selected remedial action
for the second operable unit at the Site, chosen in accordance
with CERGLA, as amended by SARA and, to the extent practicable,
the National Contingency Plan.  The decision for the second
operable unit is based on the Administrative Record.

SCOPE AND ROLE O7 OPERABLE UNIT

     The remedy described in this Record of Decision addresses
the second of three planned operable units at the Site.  The
first operable unit Record of Decision, signed in September of
1986, required completion and operation of an existing air
stripping unit at the Kellogg-Deering Well Field (wellhead
treatment).  The wellhead treatment is currently operational and
is designed to provide safe drinking water to the residents of
Norwalk.  This second operable unit Record of Decision addresses
a major source of soil and groundwater contamination upgradient

-------
of the Kellogg-Deering Well Field.  The source area is known as
the Elinco/Pitney Bowes/Matheis Court Complex and is considered
to be a source of the contamination of the aquifer that supplies
the Kellogg-Deering Well Field.

     The selected remedy for the second operable unit was
developed by combining a management of migration alternative with
a source control alternative.  In summary, the remedy addresses:
1) highly concentrated residual groundwater contamination
emanating from the complex in both the overburden and bedrock
aquifers, 2) residual soil contamination existing below the
buildings and parking lots at the complex and 3) the introduction
of contaminants into the buildings at the complex from underlying
contaminated soils.

     As described below in the Documentation of Significant
Changes section, EPA will address a third operable unit for the
site in the future.  The third operable unit will, at a minimum,
address the area of contamination located downgradient of the
Source Area addressed by this Record of Decision.


SITE CHARACTERISTICS

     The significant findings of the initial and supplemental
remedial investigations are summarized below.  A complete
detailed discussion of the site characteristics can be found in
the initial and supplemental Remedial Investigation reports.

    i TCE, PCE, and other chemicals were detected in soils and
groundwater at the site and in indoor air in the buildings at the
site.  Surface water and sediment samples were collected and
analyzed only during the initial Remedial Investigation.  No
significant levels of contaminants were found in either the
surface water or the sediments to warrant further investigation
during the supplemental Remedial Investigation.

     As stated above, the Elinco/Pitney Bowes/Matheis Court
Complex is a major source of groundwater and soil contamination
which is contributing to the contamination of the Kellogg-Deering
Well Field and the aquifer supplying the Well Field.  The
existence of other source areas contributing to contamination of
the portion of the aquifer that supplies the Kellogg-Deering Well
Field is not precluded.


Groundvater

     Groundwater data collected to date indicates that a
significant source area of contamination exists below the complex
located at 272 and 282 Main Avenue in Norwalk.  A plume of
groundwater contamination is emanating away from the complex in a

-------
••si-radial pattern from the northwest to southwest directions.
Contaminant transport occurs in both the overburden and bedrock
aquifers.  The southwest extent of the plume is migrating below
the Norwalk River and intersects the Kellogg-Deering Well Field.
The vest and northwest edge of the plume appears to extend to
Deering Pond and the Norwalk River.

     Both the initial RI and the supplemental RI support a
finding of migration of groundwater contaminants from the complex
through the unconsolidated overburden aquifer and through deep
and shallow fractures of the bedrock aquifer to the wells of the
Kellogg-Deering Well Field.  Some of the contamination has
entered bedrock fractures immediately below the buildings at the
complex.  Contaminants discharge from the bedrock into the
overburden aquifer in the vicinity of the Norwalk River and the
CL&P Landfill.  While the actual location of every fracture
containing contaminated groundwater is uncertain, groundwater
data collected to date indicates that a "piping system" of
bedrock fractures is responsible for contamination migration from
the complex to the overburden aquifer near the Kellogg-Deering
Wellfield.

     A generalized map of groundwater contamination is shown in
Figure 4.  This map combines groundwater quality data from both
the bedrock and overburden aquifers.  The map divides the area of
contamination into two parts:  A Source area and a Downgradient
area.  The Source area is defined as the approximate area in
which TCE concentrations exceed 6,600 micrograms per liter
(ug/1).  The Downgradient area is the approximate area in which
TCE concentrations have been found to be between 5 ug/1 and 6,600
ug/1.  The concentrations of 5 ug/1 TCE is the highest
concentration permitted in compliance with federal drinking water
standards.

     The most pronounced groundwater contaminants from the
standpoint of frequency of occurrence and concentration include
PCE, TCE, TCA and toluene.  These compounds were detected in 52,
79, 42 and 18 of the 115 groundwater samples obtained during the
Supplemental RI.  Concentrations of these compounds ranged as
high as 68,000, 430,000, 2,000 and 14,000 ug/1, respectively.
TCE was the most concentrated and widely detected contaminant.
Ethyl benzene and 1,2-dichloroethene were detected in 10 and 8
samples, respectively, at concentrations ranging as high as 5,800
ug/1 and 4,800 ug/1, respectively.  By contrast, none of the
other volatile chemicals detected were found at concentrations  in
excess of 65 ug/1 (with the exception of 1 detection of 4-methy1-
2-pentanone at 990 ug/1).

-------
     Based  on the  distribution of the dense chlorinated organics
in groundwater below the complex, it is apparent that density
effects probably play a major role in contaminant movement.  High
concentrations of  PCE (up to 68,000 ug/1) and TCE (up to 430,000
ug/1).were  detected  in groundwater samples collected from
monitoring  wells installed on the complex property.  Invariably,
concentrations in  samples from deeper wells exceeded those in
samples fron  shallow wells at the same well cluster locations.
It appears  possible  that a dense non-aqueous phase liquid (DNAPL)
may be present below the complex buildings.

     PCE  and  TCE were also frequently detected in downgradient
monitoring  wells.  Both overburden and bedrock monitoring well
samples revealed the presence of chlorinated solvents.
Concentrations decreased from the source area wells to
downgradient  wells.   However, TCE concentrations ranged as high
as 37 ug/1  in one  of the production wells sampled (Layne 2).

     The  groundwater contamination above 5 ug/1 TCE encompasses
approximately 54 acres.  It is calculated that 316,000,000
gallons of  groundwater are affected and contaminated with
approximately 19,400 pounds of soluble TCE plus less quantities
of other  contaminants.  In addition, DNAPL may be present at the
complex and possibly downgradient of the complex.  Of the total
quantity  of contaminated groundwater, it is estimated that 15
percent is  in bedrock and 85 percent is in overburden.


Soil

     Source characterization at the complex was conducted through
a soil-gas  investigation and subsurface soil sampling and
analysis.   As a result of the physical nature of the complex
property,  (three large buildings) characterization of the soil
source area was limited.  The soil-gas investigation indicated
that the  majority  of the residual contamination at the complex
exists beneath the complex buildings.  PCE and TCE were the most
frequently  detected  and most highly concentrated chemicals in the
soil-gas  samples.  Eight soil gas samples were collected under
the buildings.  These samples were several orders of magnitude
greater than  those in outside the complex buildings.

     Approximately 30 subsurface soil borings were taken outside
the complex buildings.  The PCE subsurface soil sample results
indicated that releases to the subsurface occurred in the
vicinity  of the former processing areas within the buildings.
TCE was found in 28  of 104 soil samples in levels ranging from 7
to 3,903  ug/1;  PCE was found in 22 of 104 soil samples ranging in
concentrations  from  3  to 8,688 ug/kg.  EPA acknowledged that
certain soil  samples  collected at the complex and analyzed by the
Contract  Laboratory  Program as part of the supplemental Remedial
Investigation were suspect (See Matthew Hoagland July 17,  1989

-------
                                8

memorandum and NUS Corp. July 20, 1989 letter to EPA in
Administrative Record).  However, EPA maintains that the
estimated 44,000 cubic yards of contaminated soils based on an
average depth to bedrock of 12 feet, as concluded in the
supplemental Remedial Investigation, is correct.


Indoor Xir

     An indoor air study performed in the three buildings at the
complex was limited to TCE and PCE due to their frequent
occurrence in groundwater and soil gas samples beneath and in the
vicinity of the complex.  Since both PCE and TCE exhibit a strong
potential for volatilization, and are thus considered to be
mobile in the environment, they were identified as the
contaminants of concern.  These contaminants were detected in the
three study buildings at concentrations ranging from 20 ug/m  to
764.3 ug/m3 (TCE),  and from below the detection limits  to 45.2
ug/m3 (PCE).


SUMMARY OF SITE RISKS

     A Risk Assessment was performed as part of the supplemental
Remedial Investigation to estimate the probability and magnitude
of potential adverse human health and environmental effects from
exposure to contaminants associated with the site.  Sixteen
contaminants of concern or "indicator compounds" listed in Table
1 were selected for evaluation in the Risk Assessment.   These
contaminants constitute a representative subset of the more than
25 contaminants identified at the site during the supplemental
Remedial Investigation.  The 16 contaminants were selected to
represent potential onsite hazards based on toxicity,
concentration, frequency of detection, and mobility and
persistence in the environment.

Qroundvater

     Groundwater contaminants at the site include relatively
water soluble volatile organics.  Monocyclic aromatics (e.g.,
benzene, toluene, etc.) and chlorinated straight-chain
hydrocarbons (e.g., TCA, PCE, TCE, etc.) constitute the
predominant groundwater contaminants.  All of the potential
carcinogenic substances detected in groundwater were included as
indicator compounds, regardless of their frequency of occurrence
or concentration, and several substances displaying only
noncarcinogenic effects (e.g., toluene, ethylbenzene, 1,1,1-
trichloroethane, acetone, and 4-methyl-2-pentanone)  were also
included as indicator compounds.

-------
     Hunan health risks were characterized both quantitatively
and qualitatively in the risk assessment with regard to
groundwater exposure.  Qualitative risks were developed through
the comparison of concentrations and/or doses with the
appropriate standards and guidelines.  The qualitative risks are
summarized in Table 6-2 of the supplemental Remedial
Investigation. Trichloroethane (TCE) one of the primary
contaminants of concern exceeded the Maximum Contaminant Level
(MCL) of 5 ug/1 and the Maximum Contaminant Level Goal (MCLG) of
0 ppb at 79 of 115 samples taken during the supplemental Remedial
Investigation.

     Potential human health effects associated with the
contaminants of concern in groundwater were estimated
quantitatively through the development of several hypothetical
exposure scenarios.  Incremental; lifetime cancer risks and a
measure of the potential for noncarcinogenic adverse health
effects were estimated for the various exposure scenarios.
Exposure scenarios were developed to reflect the potential for
exposure to hazardous substances based on the characteristic uses
and location of the Site.  Factors of special note that are
reflected in the Risk Assessment are that the Site is located in
a residential and light industrial urban area and that the area
is served by a public water supply.  Groundwater from the
Kellogg-Deering Well Field, blended with water from surface water
reservoirs as part of the water supply, is currently being
treated for contaminants that are associated with the contaminant
plume being addressed in this Record of Decision.

     The quantitative risk assessment considered potential
exposures associated with household use of potable water.  There
are numerous routes of exposure associated with household use of
contaminated water, including ingestion, inhalation of volatiles
emitted from showers, dishwashers, washing machines, bathing, car
washing etc.  However, since previous experience has shown that
ingestion and inhalation of volatiles during showering are the
predominant exposure mechanism in the home, these are assessed
quantitatively for this exposure route.

     EPA estimated potential health effects for future use
groundwater using both future conditions (worst case) and future
conditions (plausible case) scenarios.  Doses under the worst
case scenario were developed using the indicator chemical
concentrations detected at upgradient locations.  The maximum,
arithmetic average, and geometric mean groundwater concentrations
were used.  This exposure scenario provides an indication of
groundwater quality at the source and in the aquifer in general.
This scenario does not consider the existence of the air
stripping column and therefore corresponds to a case under which
air stripper failure or decommissioning occurs or if installation
of residential wells in locations near the source takes place in
the future.  For estimating risks from the Well Field,  this

-------
                                10

scenario does not include a dilution factor for the Morvalk
River; it assumes the dilution offered by the river is removed
(due to silting, dry weather, etc.) The plausible case exposure
scenario is similar to the worst-case scenario.  Maximum
observed, arithmetic average, and geometric mean groundwater
concentrations were used.  However, it is expected that the air
stripping column will continue to operate, and the stripper
removal efficiency (99.9%) was included.  This scenario included
a dilution factor, since only a minor portion of all wellfield
water actually originates from the upgradient aquifer.  This
exposure scenario corresponds to a condition in which the
contamination at the complex acts as a continuous source of
wellfield contamination.

     Quantitative risk estimates are generated through the use of
Carcinogenic Potency Factors and Reference Doses. Risks from
exposure to chemicals that are associated with noncarcinogenic
health effects are assessed by calculating the ratio of the
estimated dose from exposure to each noncarcinogenic compound to
a reference dose.  A reference dose is the dose below which no
adverse health effect is expected to occur.  If this value
exceeds unity (1.0), there is a potential health risk associated
with exposure to that particular chemical.  This risk estimate is
not a mathematical prediction of the severity of toxic effects;
it is simply a numerical indicator of the possibility of the
occurrence of noncarcinogenic effects.  The overall potential for
the occurrence of effects is assessed by summing the risks to
individual chemicals.  The resulting summation is the Hazard
Index.  This approach assumes additive dose and effects for all
contaminants.

     Carcinogenic risk estimates are provided in the form of
incremental cancer risks and are determined by multiplying the
potency factor for a specific carcinogen by the intake level.
The resulting number (risk) is a unitless expression of an
individual's incremental likelihood of developing cancer as a
result of exposure to the carcinogenic indicator chemicals.  An
incremental cancer risk of 1 x 10  indicates that an exposed
individual has a 1/1 x 10* (or 1 in 1  million)  chance  of
contracting cancer.

     Total carcinogenic risks for multiple compounds are
presented as the summation of the risks for the individual
contaminants.  Calculating risks in this manner assumes that
individual intakes are small, that there are no antagonistic/
synergistic effects between chemicals, and that all chemicals
produce the same result (i.e., cancer).  Cancer risks from
various exposure routes are also additive, if the exposed
populations are the same.

-------
                                11

     The  cumulative  incremental cancer risk for groundwater
inhalation  and  ingestion  for adult receptors under the two
exposure  scenarios and  existing conditions are as follows:


    Existing Condition       2.8 x 10"9

    Future  Worst Case        3.2 x 10"1   (Maximum Concentrations
                                          of Contaminants)


    Future  Plausible Case  4.7 x 10"5   (Maximum Concentrations
                                        of Contaminants)   ;

     The  major  components of the incremental cancer risk result
from the  presence of PC-E .and TCE.

     Noncarcinogenic effects are unlikely under current
conditions  since the hazard index is well below 1.0.
Noncarcinogenic effects would be quite likely under the future
worst case  conditions.  The major component of the potential
noncarcinogenic effect  is PCE.  Under the future plausible case
scenario  the hazard  indices do not exceed unity.

     In conclusion,  both  the Hazard Indices and incremental
cancer risk estimates reveal that groundwater contamination at
the Kellogg Deering  Well  Field would pose a significant threat to
the public  health if not  for the dilution offered by the Norwalk
River and the removal of.contaminants by the existing treatment
system (i.e., the air stripper).  The contamination in the
aquifer in  the  vicinity of the source area is outside the EPA
risk range  of 10"* to 10  and poses a threat to human health  in
the event it was used for private water supply wells.

     Although the risks based on the maximum observed
concentrations  may be overestimates under present site
conditions, it  is anticipated that continued migration of
contaminants to the  Well  Field could result in a long-term threat
to the public health.   In the event the contaminated groundwater
was used  for private water supply it could pose a risk to human
beings.   The risks generated using arithmetic average and
geometric mean  concentrations may be better estimates of future
risks in -the near term.   However, based on the concentrations
encountered in  the source area and evidence suggesting the
presence  of a Dense  Non-Aqueous Phase Liquid below the complex,
it is considered possible that the well field concentrations
could attain levels  approaching the source concentrations over
the long-term.   In this respect, the contaminated bedrock
groundwater at  the complex could act as a continuous source of
downgradient contamination.  In the event that the dilution
offered by  the  river is removed (as a result of silting,
prolonged dry weather,  or other unforeseen circumstances)  or the

-------
                                12

•tripper removal efficiency  is impeded  (as a result of fouling or
equipment  failure),  it  is possible that Well Field production
would have to be discontinued.


Bell

     Chemical contaminants in the soils below buildings and the
parking lot in the complex area currently present no health risk
from exposure through direct contact or incidental ingestion of
contaminated soil.   The potential does exist for human inhalation
of contaminants that are being released from soils beneath the
complex to the air within the buildings.  The levels of
contaminants measured during one sampling event are well within
the standards set for industrial safety by the Federal
Occupational Safety  and Health Administration.  However, the
risks in the Elinco  building from exposure to these levels (4.0 x
10 ) exceeds the Superfund target risk  range  (10** to  10  )  using
a reasonable worst case scenario.  If, in the future, the
buildings  were demolished and/or soil excavated, the potential
risks to public health  and the environment would increase
substantially.  Finally, contamination in the soils will continue
to leach into groundwater under the complex and further
contribute to groundwater contamination unless the levels of
contamination in the soils is reduced.

     No significant  levels of contaminants were found in either
surface water or sediments during the initial Remedial
Investigation, so no further sampling or risk assessment was
performed  during the supplemental Remedial Investigation.

     A complete discussion of Site risks can be found in Section
6 of the supplemental Remedial Investigation.  Actual or
threatened releases  of  hazardous substances from the Site, if not
addressed  by implementing the response action selected in this
ROD, may present an  imminent and substantial endangerment to
public health, welfare  or the environment.


DOCUMENTATION OF SIGNIFICANT CHANGES

     EPA published a Proposed Plan for remediation of the Site on
July 26, 1989.  The  preferred alternative consisted of a source
control component and a management of migration component.  The
source control component consisted of the treatment of
contaminated soils at the complex using in-situ vapor extraction.
The management of migration portion of the Proposed Plan covered
alternatives for the treatment of contaminated groundwater.  In
developing the management of migration components,  the area of
groundwater contamination at the site was divided into two zones,
the Source Area and  Downgradient Area, according to the
concentration levels found in the supplemental RI.   The Source .

-------
                                13

Area is characterized by TCE concentrations greater than 6,600
ug/1 and consists  of an area that includes most of the complex as
well as an adjacent area extending 400 feet west and about 500
feet southeast  of  the complex. See Figure 4.  The Downgradient
Area consists of the remainder of the TCE groundwater plume in
excess of 5 ug/1,  the Safe Drinking Water Act maximum contaminant
level  (MCL) for TCE.  The Downgradient Area as mapped in the
supplemental FS extends westward either to the Norwalk River or,
where influenced by the pumping of the Kellogg-Deering Wells,
under and beyond the Norwalk River.  The management of migration
component of the preferred alternative included the extraction
and treatment of groundwater from both the Source Area and
Downgradient Area.  The Proposed Plan indicated that there were
four potential  treatment technologies for groundwater, including
air stripping,  carbon adsorption, ultraviolet uv/chemical ' • '.
oxidation and steam stripping.  The Proposed Plan also stated
that air stripping was the likely choice of treatment
technologies in the ROD.  All four treatment technologies would
involve: 1) the installation of wells to extract contaminated
groundwater at  the site; 2) the pretreatment of extracted
groundwater to  remove suspended solids and metals that could
reduce the effectiveness of the principal treatment unit; 3) the
treatment of contaminated groundwater to remove VOCs; and 4) the
discharge of treated groundwater.

     The Proposed  Plan indicated that the disposal.method for the
treated groundwater would be determined during the.design phase
and would consist  of either discharge to the local publicly owned
treatment works (POTW), discharge to the Norwalk River through a
new pipe, discharge to the Norwalk River through existing storm
sewers or reinjection into the aquifer, either solely or in
combination. The Proposed Plan also stated that if after an
adequate period of performance of the preferred alternative,
complete restoration of the bedrock aquifer was determined to be
technically impracticable, then EPA would consider amending the
Record of Decision to waive applicable or relevant and
appropriate requirements (ARARs) in the aquifer.  In addition,
the Proposed Plan  stated that the cleanup goals may be adjusted
if chemical contaminant concentrations reach a constant value and
contaminants are no longer being removed at significant levels.

     The remedy selected in this ROD adopts the same source
control component  that was presented in the Proposed Plan.   For
the management  of migration component, however, this ROD contains
the following changes:

     *    Extraction and treatment of Source Area groundwater
          only  will occur pursuant to this ROD; EPA will make a
          decision regarding remediation of the Downgradient Area
          as a  separate operable unit at a later date;

-------
                                14

     *    Air stripping has been  chosen as the groundwater
          treatment technology for Source Area groundwater; and

     *    EPA will consider, during the design phase, the use of
          either ultraviolet(UV)/chemical oxidation, carbon
          adsorption or steam stripping as alternative
          groundwater treatment technologies for Source Area
          groundwater if it can be shown that the proposed
          alternative technology will meet the four statutory
          criteria for CERCLA cleanups and will provide a similar
          balance in the nine evaluation criteria as air
          stripping.


Each of the changes listed above will be discussed in turn in the
remainder of this section.

     EPA has determined that the construction of a new highway,
Route 7, will impede the performance of remedial design
activities in the near future in the Downgradient Area.
Therefore, in order to avoid any delay in remedial actions for
contaminated soil and Source Area groundwater, EPA has decided to
postpone a decision on remediation of the Downgradient Area until
a later date.  Remediation of the Downgradient Area will
constitute a third operable unit  for the site.  The selection of
a remedy in the Downgradient Area will depend on several factors
which may include 1) the effectiveness of the Source Area
groundwater extraction and treatment system in remediating
contamination in the Downgradient Area groundwater, 2) any future
data indicating that the contamination plume at the site is
having a negative impact on the Norwalk River and 3)
accessibility of the Downgradient Area after the construction of
Route 7 is complete.

     In addition, many individuals raised issues during the
public comment period which challenged the effectiveness and
protectiveness of extracting and treating groundwater from the
Downgradient Area.  While it is EPA's intent to address the
contamination in the Downgradient Area groundwater, EPA does see
merit in further evaluating options for remediating this area
while implementing the Source Area cleanup.  Therefore, EPA has
decided to refrain from making a decision on a remedy for the
Downgradient Area until the concerns raised during the public
comment period can be more fully evaluated.

     As a result of its decision to create a third operable unit
to address the Downgradient Area, EPA will not evaluate
Alternatives III, IV, VII and VIII in this Record of Decision
because these alternatives differ from Alternatives I, II, V and
VI only with regard to the inclusion of the Downgradient Area
groundwater extraction and treatment system.  All significant
comments submitted to EPA during the July 27, 1989 to August 25,

-------
                                15

1989 public  comment period that pertain to the Dovngradient Area
groundwater  extraction  system will be more thoroughly responded
to at the time a  final  decision is made on remediation of the
Dovngradient Area.

     As -mentioned above, EPA has decided to address the
Downgradient Area as a  separate operable unit.  EPA believes that
this approach is  environmentally sound and logical for several
reasons: the highest levels of contamination at the site are
associated with the Source Area; cleanup of the Source Area will
prevent the  further migration of contaminants from the Source
Area into the Downgradient Area; and further evaluation of the
impact of Source  Area remediation on Downgradient Area
remediation  will  ensure that the Downgradient Area cleanup will
be protective and effective.

     The Proposed Plan  indicated that there exist four potential
groundwater  treatment technologies for the site, including, air
stripping, ultraviolet(UV)/chemical oxidation, carbon adsorption
or steam stripping.  EPA has chosen air stripping as the
groundwater  treatment technology for Source Area groundwater in
this ROD because  it vill meet EPA's cleanup goals, will provide
overall protectiveness  of public health and the environment, will
provide long and  short  term effectiveness, is implementable, will
provide reduction of toxicity, mobility and volume and is cost
effective.   EPA will consider, during the design phase, the use
of either ultraviolet  (uv)/chemical oxidation, carbon adsorption
or steam stripping as alternative groundwater treatment
technologies for  Source Area groundwater if it can be
demonstrated that another alternative will meet the four
statutory requirements  for cleanups under CERCLA and will provide
a similar balance of the nine evaluation criteria as air
stripping.   Any decision to utilize one of these alternative
treatment technologies  will be documented and an explanation of
significant  differences issued, if appropriate.

     The Proposed Plan  suggested that EPA would reevaluate the
remedy if after an adequate period of performance of the remedy
complete restoration of the aquifer is determined to be
technically  impracticable and that cleanup goals might be
readjusted if chemical  contaminant concentrations reach a
constant value and are  no longer being removed at significant
levels.  EPA is not including these reevaluations as a part of
this ROD.  This remedy  will attain EPA's cleanup goals, will meet
ARARs and assure  protectiveness of human health and the
environment.  EPA, retains the right to amend this Record of
Decision or  change its  decision on the remedy for the Source
Area, as provided by CERCLA.

-------
                                16

     None of the changes presented in this ROD differ
significantly  from those presented in the Proposed Plan.
Furthermore, to the extent that differences exist, they are
the logical outgrowth of the Proposed Plan and were reasonably
forseeable.  The Agency does not believe that it is necessary to
reissue the Proposed Plan for further comment and provides the
following rationale: 1) the supplemental Feasibility Study and
the Proposed Plan discussed the alternatives of proceeding with
Source Area groundwater extraction and treatment only and source
area soil treatment only; 2) the public already had an
opportunity to comment on this alternative, which is merely a
different combination of a source component and management of
migration component which were already evaluated in the FS; 3)
the supplemental Feasibility Study and the Proposed plan
discussed the  alternatives of air stripping; 4) the suggested
changes, do not alter the overall remedial objectives for the
Site, as presented below; and 5) the eventual decision on
remediation of the Downgradient Area will again be subject to
public comment.


DEVELOPMENT AND SCREENING OF ALTERNATIVES

Statutory Requirements/Response Objectives

     Prior to  the passage of the Superfund Amendments and
Reauthorization Act of 1986 (SARA), actions taken in response to
releases of hazardous substances were conducted in accordance
with CERCLA as enacted in 1980 and the revised National Oil and
Hazardous Substances Pollution Contingency Plan (NCP), 40 C.F.R.
Part 300, promulgated in the Federal Register on November 20,
1985.  Although EPA proposed revisions on December 21, 1988, to
the NCP to reflect SARA, until those proposed revisions are
finalized, the procedures and standards for responding to
releases of hazardous substances, pollutants and contaminants
shall be in accordance with Section 121 of CERCLA and, to the
maximum extent practicable, the current NCP.

     Under its legal authorities, EPA's primary responsibility at
Superfund sites is to undertake remedial actions that are
protective of  human health and the environment.  In addition,
Section 121 of CERCLA establishes several other statutory
requirements and preferences, including: a requirement that EPA's
remedial action, when complete, must comply with applicable or
relevant and appropriate environmental standards established
under federal  and state environmental laws unless a statutory
waiver is invoked; a requirement that EPA select a remedial
action that is cost-effective and that utilizes permanent
solutions and  alternative treatment technologies or resource
recovery technologies to the maximum extent practicable; and a
statutory preference for remedies that permanently and
significantly  reduce the volume, toxicity or mobility of

-------
                                .17

hazardous substances over remedies that do not achieve such
results through treatment.  Response alternatives were developed
to be consistent with these Congressional mandates.

     A number of potential exposure pathways were analyzed for
risk and threats to public health and the environment in the Risk
Assessment and Indoor Air Addendum to the supplemental Remedial
Investigation report.  Guidelines in the Superfund Public Health
Evaluation Manual  (EPA, 1986) regarding development of risk
analyses for remedial alternatives were used to assist EPA in the
development of response actions.  As a result of these
assessments, remedial response objectives were developed to
mitigate existing  and future threats to public health and the
environment.  These response objectives are:

      1.   Treat soils to minimize the risk from exposure
           by direct contact with the contaminants and minimize
           the leaching of contaminants to the groundwater;

      2.   Restore the overburden and bedrock aquifers to
           drinking water standards;

      3.   Minimize the further introduction of contaminated
           groundwater from the bedrock aquifer to the overburden
           aquifer supplying the Kellogg-Deering Well Field;

      4.   Protect the Norwalk River from the adverse effects
           of contaminated groundwater discharges; and

      5.   Prevent exposure to contaminated groundwater.

     The remedial  objectives listed above are overall objectives
for the entire site at the completion of all operable units.  The
specific remedial  objectives for the operable unit associated
with this Record of Decision are listed with the discussion of
the selected remedy.


Technology and Alternative Development and Screening

     CERCLA, the NCP, and EPA guidance documents including, the
"Guidance on Feasibility Studies Under CERCLA" dated June 1985,
the "Interim Guidance on Superfund Selection of Remedy" [EPA
Office of Solid Waste and Emergency Response (OSWER)],  Directive
No. 9355.0-19 (December 24, 1986), and the Interim Final
"Guidance for Conducting RIs and FSs under CERCLA," OSWER
Directive No. 9355.3-01, set forth the process by which remedial
actions are evaluated and selected.  In accordance with these
requirements and guidance documents, a range of treatment
alternatives were  developed for the Site, a containment option
involving little or no treatment, and a no-action alternative.

-------
                                18

     Section 121(b)(l) of CERCLA presents several factors that,
at a minimum, EPA  is required to consider in its assessment of
alternatives.  In  addition to these factors and the other
statutory directives of Section 121 of CERCLA, the evaluation and
selection process  was guided by the EPA document "Additional
Interim Guidance for FY'87 Records of Decision" dated July 24,
1987.  This document provides direction on the consideration of
SARA cleanup standards and sets forth nine factors that EPA
should consider in its evaluation and selection of remedial
actions.  The evaluation criteria are:

1.   Overall Protection of Human Health and the Environment.

2.   Compliance with Applicable or Relevant and Appropriate
     Requirements  (ARARs).

3.   Long-term Effectiveness and Permanence.

4.   Reduction of  Toxicity, Mobility or Volume.

5.   Short-term Effectiveness.

6.   Implementability.

7.   Cost.

8.   Community Acceptance.

9.   State Acceptance.

     Chapters 10 and 11 of the supplemental Feasibility Study
identified, assessed and screened technologies based on
effectiveness, implementability and cost.  These technologies
were combined into source control (SC) and management of
migration (MM) alternatives.  The purpose of the initial
screening was to narrow the number of potential remedial actions
for further detailed analysis while preserving a range of
options.

     In Chapter 12 of the supplemental Feasibility Study, the
"Detailed Analysis of Overall Remedial Action Alternatives,"
remedial alternatives are developed by combining the technologies
identified in the  previous screening process into the categories
required by OSWER  Directive No. 9355.3-01.  This process involves
combining the source control component with the Source Area and
Downgradient Area  management of migratfon components and no-
action to produce  a range of remedial alternatives.   Table 2
displays the eight Remedial Alternatives that undergo the
detailed analysis.

-------
                                19

DESCRIPTION OF ALTERNATIVES

     This section presents a narrative summary and brief
evaluation of the alternatives that do not involve management of
migration components  in the Dovngradient Area.  The alternatives
which incorporate treatment of groundwater in the Downgradient
Area (i.e. Alternatives III, IV, VII and VIII) are not described
in this Record of Decision because EPA is not making a formal
decision regarding remediation of the Downgradient Area at this
time.

     The summaries below include brief evaluations of the first
seven of the nine evaluation criteria described above.  The last
two criteria, State and Community Acceptance, are described later
during the comparative analysis of alternatives.  Table 12-20 of
the supplemental Feasibility Study presents a detailed tabular
assessment of each of the original eight alternatives.  Table 12-
20 is supported by text in Chapter 12 of the supplemental
Feasibility Study.

     In October of 1987, Elinco and the Connecticut Department of
Environmental Protection (DEP) entered into a consent agreement
to implement a groundwater extraction and treatment system within
the boundaries of the complex.  This system, composed of four
extraction wells, is  presently under construction.  Elinco's
extraction system was not considered in EPA's supplemental
Feasibility Study because the Elinco system was under design at
the time of the supplemental Feasibility Study's writing.
Alternatives described in the supplemental Feasibility Study and
below that involve no action for Source Area groundwater (i.e.,
Alternatives I and V) do not include the Elinco extraction
system.  Alternatives that involve Source Area groundwater
treatment (i.e., Alternatives II and VI) would involve an
evaluation by EPA of  the Elinco extraction system during Remedial
Design.


Alternative I;  No Action

     Alternative I, the no action alternative, means leaving the
site in its present condition.  This alternative does not include
the remediation system to be undertaken by Elinco Associates.
Contaminated soils would not be treated. Long term institutional
controls including restrictions on deeds, building permits and
private wells would be enacted.  Aquifer restoration would rely
primarily on natural  aquifer flushing because only a portion of
the contaminants in the groundwater would be intercepted and
treated by operations at the Kellogg-Deering Well Field.  The
remainder of the contaminants in the groundwater outside of the
effective pumping radius of the Kellogg-Deering Well Field would
either persist in dead end bedrock fractures or migrate through
the aquifer and eventually discharge into the Norwalk River.  In

-------
                                20

addition, contaminants in soils of the unsaturated zone at the
complex could percolate downward to the water table if
infiltration was not prevented by the buildings and parking lots.

     Environmental monitoring of groundwater and of indoor air
within the buildings at the complex would occur indefinitely.
This monitoring would be implemented to aid in identifying
potential receptors and minimize potential health risks.  Details
of the monitoring program would be developed in the Remedial
Design phase.

     Alternative I would not provide overall protection of human
health and the environment because no cleanup of contaminants in
soils and groundwater would occur and further release of
contaminants to groundwater from soils would not be minimized.
Alternative I is impleroentable and would provide short term
effectiveness.  However, long-term institutional controls
including restrictions on deeds, building permits and private
wells would be needed.  There would be no reduction of the
mobility, toxicity and volume of contaminants in the unsaturated
soils of the complex.  The mobility of the contaminants in
groundwater would not be reduced.  The toxicity and volume of
contaminated groundwater would not be reduced except by well head
treatment.  Alternative I would not provide long-term
effectiveness or permanence or achieve groundwater quality ARARs.

Estimated Capital Cost: $0
Estimated Operations and Maintenance Cost (net present worth):
$240,000
Estimated Total Cost $240,000


Alternative lit  Groundwater Treatment of Source Area

     Alternative II was developed for the purpose of controlling
the contaminant migration only in the Source Area which is the
most heavily contaminated portion of the aquifer.  This system,
would involve intercepting contaminated groundwater to the
maximum extent practicable in the source Area.  The contaminated
groundwater would be treated by air stripping and discharged
offsite.  An evaluation of the Elinco wells to remediate the
contamination within the complex boundaries would be part of
predesign activities.  Contaminated soils would not be treated.
Long term institutional controls, including restrictions on deeds
building permits and private wells, would be enacted.

     Alternative II would reduce the overall mass of contaminants
in the bedrock aquifer and eventually reduce the input of
contaminants into the overburden aquifer in the vicinity of the
Norwalk River.  Contaminants in soils of the unsaturated zone at
the complex could percolate downward to the water table if
infiltration were prevented by the buildings and parking lots.

-------
                                21

     Monitoring of  indoor air within the buildings at the complex
would occur  indefinitely to minimize potential health risks.
Groundwater  monitoring would occur in both the Source and
Downgradient Areas  to evaluate the progress of the remedy.
Details of the monitoring program would be developed in the
Remedial Design phase and will include at a minimum, overburden
and bedrock  monitoring wells.

     Alternative II would provide protection of human health and
the environment through the cleanup of contaminants in Source
Area groundwater.   Alternative II would comply with groundwater
ARARs, but not soil ARARs.  The groundwater treatment component
of Alternative II would substantially reduce the toxicity and
volume of contamination.  The mobility of contaminants would be
reduced by the extraction wells.  Since Alternative II does not
include soil treatment/ mobility, toxicity and volume of
contaminants in soils would .not be reduced.

     Alternative II is implementable and would provide short term
effectiveness.  Alternative II would provide limited long-term
effectiveness or permanence because soil contamination would
continue to  contribute to the groundwater in the Source Area.

     The time required to attain the groundwater cleanup goals
cannot be estimated with precision since volatile organic
compounds are expected to, be present as dense non-aqueous phase
liquids in bedrock  fractures.

Estimated Time for  Completion Including Design, Bidding,
Construction and Operation: Exceeds 10 years
Estimated Capital Cost: $4,600,000
Estimated Operations and Maintenance Cost (net present worth):
$3,100,000
Estimated Total Cost $7,700,000


Alternative  III; Groundvater Treatment of Source and Dovnaradient
Areas

     Alternative III would involve groundwater treatment of the
Source Area  and the Downgradient Area without treatment of soils
at the complex.  This alternative is not being considered in this
Record of Decision  because EPA has decided not to issue a final
decision on  the Downgradient System at this time.  Alternative
III, minus the Downgradient System, is essentially the same as
Alternative  II which is-considered in this Record of Decision.
The reasons  for not including the Downgradient System in this
Record of Decision  can be found above in the Documentation of
Significant  Changes section.

-------
                                22

Alternative TVs Groundvater Treatment of Dovnaradient Area

     Alternative IV would involve groundwater treatment of the
Downgradient Area without groundwater treatment of the Source
Area or treatment of soils at the complex.  This alternative is
not being considered in this Record of Decision because EPA has
decided not to issue a final decision on the Downgradient System
at this time.  Alternative IV, minus the Downgradient System, is
essentially the same as Alternative I, the no action alternative,
which is considered in this Record of Decision.  The reasons for
not including the Downgradient System in this Record of Decision
can be found above in the Documentation of Significant Changes
section.
Alternative V: Soil Treatment

     Alternative V would involve treatment of contaminated soils
only.  Groundwater would remain much the same as described in
Alternative I, the no action alternative.  The in-situ soil
vacuum extraction system would utilize unsaturated zone
extraction wells, vacuum pumps, associated piping and vapor
treatment.  The vacuum extraction process develops a vacuum in
the unsaturated zone which induces contaminant flow in the form
of a vapor.  The contaminant-bearing vapor is forced through the
soil matrix to extraction wells and withdrawn.  The process
essentially air-strips volatile organic compounds from the soils
due to the mass transfer of contaminants to the soil vapor.
Moisture is removed from the contaminated soil vapor in a
condenser or separator.  Vapors are then drawn through activated
carbon beds where volatile organic contaminants are absorbed
before discharging the clean air into the atmosphere in
compliance with Federal and State standards.  The separated water
will be treated and discharged.

     Alternative V would remove contaminants from soils of the
unsaturated zone below the complex thereby reducing further the
upward migration of contaminants into the indoor air of the
complex buildings.  If in the future the buildings and parking
lots were ever removed, Alternative V would reduce the increased
potential for downward migration of contaminants to groundwater
and would minimize the potential human health risk from exposure
to contaminated soils.

     Periodic sampling and analysis of indoor air within the
buildings at the complex would occur to monitor the progress of
soil remediation.  Wellhead gases, process gases and gas
emissions would be monitored throughout the period of operation
to measure the effectiveness of the vacuum extraction system.
The progress of the remedy would also be measured by a monitoring
well network.  Details of the soil vapor monitoring program would
be developed during the Remedial Design phase.

-------
                                23

     Groundwater monitoring in both the Source and Dbwrigradient
areas would be needed  indefinitely.  Details of the monitoring
program would be developed in the Remedial Design phase.

     Alternative V would provide protection of human health and
the environment through the cleanup of contaminated soils and by
minimizing the release of contaminants to the underlying
groundwater.  The mobility, toxicity and volume of contaminated
groundwater would not  be reduced by Alternative V.  The mobility,
toxicity and volume of soil contaminants would be reduced by the
soil vapor extraction  and treatment.

     Alternative V is  implamentable and would provide short term
effectiveness.  Due to the fact that contaminants would remain in
groundwater, institutional controls would need to be enacted
during and after the time of in-situ soil vacuum extraction
treatment.  These institutional controls would include
restrictions on deeds  and building permits.

     Alternative V would neither attain groundwater ARARs nor
provide long-term effectiveness or permanence because groundwater
contamination would persist.  Additionally, the levels of
contamination may increase in the downgradient areas of the plume
which could potentially impact the ecosystems in Deering Pond and
the Norwalk River.                                       ;

     Soil remediation  at the complex using in-situ vacuum
extraction would take  approximately 2.5 years.  However, the
ability of VOC contaminants to move upward in soils from
groundwater would likely extend the remediation time than would
be the case for soil treatment conducted concurrently with source
area groundwater treatment.

Estimated Time for Completion Including Design, Bidding,
Construction and Operation: 2.5 years
Estimated Capital Cost : $1,400,000
Estimated Operations and Maintenance Cost (net present worth):
$200,000
Estimated Total Cost $1,600,000


Alternative VIt Soil and Source Area Groundvater Treatment

     Alternative VI was developed to control the contaminant
migration in the Source Area, the most heavily contaminated
portion of the aquifer, and remove contaminants from soils at the
complex.  The groundwater system would be essentially the same as
described for Alterative II, that is,  it would involve
intercepting contaminated groundwater in the Source Area.  The
contaminated groundwater would be treated by air stripping and
discharged offsite.  An evaluation of the Elinco wells to control
the contamination within the complex boundaries would be a part

-------
                                24

of predesign  activities.  Institutional controls including
restrictions  on deeds, building permits and private wells would
be enacted.

     The source control component, in-situ vacuum extraction,
would be essentially the same as described in Alternative II.
The in-situ soil vacuum extraction system would utilize
unsaturated zone extraction wells, vacuum pumps, associated
piping and vapor treatment.  The vacuum extraction process
develops a vacuum  in the unsaturated zone which induces
contaminant flow in the form of a vapor.  The contaminant bearing
vapor is forced through the soil matrix to extraction wells and
withdrawn.  The process essentially air-strips volatile organic
compounds from the soils due to the mass transfer of contaminants
to the soil vapor.  Moisture is removed from the contaminated
soil vapor in a condenser or separator.  Vapors are then drawn
through activated  carbon beds where volatile organic contaminants
are absorbed  before discharging the clean air into the atmosphere
in compliance with Federal and State requirements.  The separated
water will be treated and discharged.

     Alternative VI would reduce the overall mass of contaminants
in the bedrock aquifer and, over the long term, reduce the input
of contaminants into the overburden aquifer in the vicinity of
the Norwalk River.

     Contaminants  would be removed from soils of the unsaturated
zone at the complex which would further aid the groundwater
restoration effort, reduce the upward migration of contaminants
into the indoor air of the buildings of the complex, and reduce
human health  risks if the buildings were ever removed and soils
became exposed.  Periodic sampling and analysis of groundwater,
soil vapor and indoor air would be required to monitor the
progress of the cleanup.  Details of the monitoring program would
be developed  in the Remedial Design phase.

     Alternative VI would provide protection of human health and
the environment because cleanup of contaminants in soils and
groundwater would  occur and further release of contaminants to
groundwater from soils would be minimized.  Treatment of the
extracted water and soil gas under Alternative VI will
substantially reduce the overall toxicity and volume of site
contamination.  Alternative VI would reduce the mobility of
contaminants  in the saturated or unsaturated zones through the
use of extraction  wells.

     Alternative VI is implementable and would provide short term
effectiveness.  Alternative VI would provide long-term effective-
ness or permanence because both Source Area groundwater and
contaminated  soils would be addressed.

-------
                                25

     Soil remediation at the complex using in-situ vacuum
extraction would take approximately 2.5 years.  The time required
to attain the groundwater cleanup goals cannot be estimated with
precison since volatile organic compounds are expected to be
present as dense non-aqueous phase liquids in bedrock fractures.

Estimated Time for Completion Including Design, Bidding,
Construction and Operation: Exceeds 10 years
Estimated Capital Cost : $6,000,000
Estimated Operations and Maintenance Cost (net present worth):
$3,000,000
Estimated Total Cost $9,000,000


Alternative VII; Soil and Groundwater Treatment

     Alternative VII would involve groundwater treatment of the
Source and Downgradient Areas and treatment of soils at the
complex.  This alternative was EPA's Preferred Alternative in the
July 1989 Proposed Plan, but is not being considered in this
Record of Decision because EPA has decided not to issue a final
decision on the Downgradient System at this time.  Alternative
VII, minus the Downgradient System, is essentially the same as
Alternative VI, the selected alternative.  The reasons for not
including the Downgradient System in this Record of Decision can
be found above in the Documentation of Significant Changes
section.


Alternative VIIIt Soil and Dovnaradient Area Groundwater
Treatment

     Alternative VIII would involve groundwater treatment of the
Downgradient Area and treatment of soils at the complex.  This
alternative is not being considered in this Record of Decision
because EPA has decided not to issue a final decision on the
Downgradient System at this time.  Alternative VIII, minus the
Downgradient System, is essentially the same as Alternative V
which is considered in this Record of Decision.  The reasons for
not including the Downgradient System in this Record of Decision
can be found above in the Documentation of Significant Changes
section.
SELECTED REMEDY

     The selected remedy is Alternative VI which consists of in-
situ vacuum extraction for soils; pumping, treating and
discharging of contaminated groundwater in the Source Area; and
implementation of institutional controls.

-------
                                26

     The overall goal of the cleanup action at the Kellogg-
Deering Well Field Superfund Site is to restore the groundwater
in the aquifer to its beneficial uses  (or health based levels)
within a reasonable period of time.  The selected remedy for the
second operable unit is a comprehensive approach to Source Area
remediation that combines components of source control and
management of migration.


Remedial Action O^eetives/Cleaaup Qeals

The selected remedy was developed to satisfy the following
remedial objectives which will guide the design of the remedy and
be used to measure the success of the remedy.

Qroundvater - The remedial objectives for contaminated
groundwater at the Source Area are as follows:

     *    Prevent further introduction of contaminated ground
          water from the Source Area to the Downgradient Area and
          ultimately to the production wells at the Kellogg-
          Deering Well Field and the Norwalk River.

     *    Restore the Source Area aquifer to drinking water
          quality.

     *    Reduce the mass of contaminants at the Source Area.

     *    Prevent human consumption of or contact with
          contaminated ground water above the cleanup goals
          presented in this ROD.

     EPA believes that active restoration of the groundwater is
appropriate for the Kellogg-Deering Well Field Superfund Site.
The groundwater cleanup levels at the site are primarily based
upon the classification of the groundwater as a current drinking
water source.  Therefore, the Maximum Contaminant Levels
promulgated under the Safe Drinking Water Act are relevant and
appropriate to the aquifer.

     EPA's decision to attempt restoration of the groundwater at
the site to drinking water standards is based on several factors.
In addition to the requirement to attain Federal and State ARARs,
EPA considered its Groundwater Protection Strategy (Office of
Groundwater Protection, August 1984) which provides guidance
concerning how different groundwaters throughout the country
should be classified and to what extent cleaning up a particular
class of groundwater is appropriate.  The policy under the
Groundwater Protection Strategy establishes groundwater
protection goals based on "the highest beneficial uses to which
groundwater having significant water resources value can
presently or potentially be put."  Guidelines for protection of

-------
                                27

aquifers are differentially based, relative to characteristics of
vulnerability, use and value.  Under the Groundwater Protection
Strategy classification, the groundwater at the Kellogg-Deering
Well Field Superfund Site is considered to be Class II-A, a
current drinking water source.

     EPA also considered the Agency's Guidance on Remedial
Actions for Contaminated Ground Water at Superfund Sites
(December 1988).  This guidance directs the Agency to consider a
10*4 to  10   range of risk levels in selecting the appropriate
risk level for the groundwater at the site.  The Source Area
groundwater will be pumped and treated until MCLS are met.  At
that point in time, groundwater will be sampled to determine if
the cumulative carcinogenic risk falls within a range generally
considered by EPA to be protective at Superfund sites and does
not exceed unity on the Hazard Index.  If the level of risk does
not fall within EPA's risk range or exceeds unity on the Hazard
Index, then additional treatment will be conducted in order to
meet EPA's risk levels.  A supplementary decision document will
be issued, as appropriate.

     The area of attainment for the groundwater goals shall be at
every point within the Source Area.  The groundwater cleanup
goals are presented in Table 3.

Soil -  The remedial objectives for contaminated soil at the
complex are as follows:

     *    Prevent public contact with contaminated soils above
          cleanup goals.

     *    Prevent leaching of contaminants to the groundwater at
          values that affect the quality of groundwater.

     Soil cleanup goals were developed in order to reduce the
potential future risk to public health and the environment and to
protect groundwater.  The soil cleanup levels presented in Table
3 are calculated by using a soil leaching model in order to
achieve the target groundwater levels.  This approach utilizes a
soil/water sorption coefficient which describes the distribu-
tion of an indicator chemical between soil and groundwater at
equilibrium.  The type of soils and amount of organic carbon
present in the soils have a large impact on the amount of
contamination that will be attenuated in those soils and,
therefore, unable to migrate and contribute to groundwater
contamination.  The soil cleanup levels listed in Table 3 are
presented as ranges because the soil types differ within the area
of soil contamination.  Further soil sampling and analysis will
be undertaken during Remedial Design to refine the soil cleanup
levels.  A more complete description the soil leaching model can
be found in Section 6 of the supplemental RI.

-------
                                28

Description of Components of the Selected Remedy

     EPAfs selected remedy consist* of source control through the
implementation of in-situ vacuum extraction at the Elinco/Pitney
Bowes/Matheis Court Complex, management of migration through
implementation of groundwater pumping, treatment and discharge
for the Source Area, and institutional controls to prevent
potential exposure to contamination during implementation of the
remedy. ' Institutional controls will involve restrictions on soil
excavation in areas of soil contamination and restrictions on
veil installations in areas of groundwater contamination.  If EPA
cannot obtain the required institutional controls, EPA will
consider treatment technologies that do not require institutional
controls.

7?ntfnJP>ted Soil Treatment - The source control component will
involve designing, installing, operating and maintaining a soil
vapor vacuum extraction system in the unsaturated zone at the
Elinco/Pitney Bowes/Matheis Court Complex which will intercept
laterally and vertically all areas of subsurface soil
contamination so as to attain the soil cleanup goals presented
above and in Table 3.  The area of attainment for the soil
cleanup goals will be everywhere on the complex property.

     The vacuum extraction system will utilize extraction,
injection and monitoring wells.  It is anticipated that
extraction wells would be located in or near the three processing
areas of the two Zell and Elinco Buildings as well as in the two
exterior corridors.  The number and location of these wells will
be finally determined during design.  Each of the extraction
wells will be surrounded by approximately 15 passive air
injection wells.

     The vacuum extraction system will utilize carbon adsorption
for vapor treatment for protectiveness consistent with OSWER
Directive 9355.0-28 and Region I guidance.  The contaminated
water from the separator will be treated as necessary and
discharged in accordance with the discharge limitations required
by the chosen groundwater discharge method as described below.

     Predesign work will include sampling to refine the volume of
soil to be treated.  This sampling program will include analysis
for total organic carbon so that soil cleanup goals can be better
refined in accordance with the soil leaching model described
above.

     An adequate and effective sampling and analysis program will
be designed and implemented to monitor the performance of the
vacuum extraction system recovery and treatment and process gas
discharges.  It is currently estimated that the soil cleanup
goals can be achieved within 2.5 years for the estimated 44,000
cubic yards of contaminated soils.  Consideration will be given

-------
                                29

to sequencing  soil  and groundwater remedial components to avoid
recontamination  of  soil  from contaminated groundvater.
Completion monitoring will also be conducted.  The costs
associated with  this source control component are listed in
Table 4.
Management of Migration component will include designing,
installing, operating and maintaining a groundwater extraction,
treatment and disposal system for the Source Area shown in Figure
4.  The remedial action would be combined with institutional
controls consisting of restrictions on the installation and use
of private wells in the area of groundwater contamination in the
Source and Downgradient Areas.  The costs associated with this
Management of Migration component are listed in Table 4.

     The Source Area extraction system will utilize two
remediation well networks.  One remediation well network will be
located at the complex property.  EPA will evaluate during design
the groundwater treatment system currently being constructed by
Elinco at the complex to determine if some, or all, of that
system may be incorporated in the groundwater treatment system
required by this Record of Decision.  The second remediation well
network will be installed near the center of the Source Area
contamination near the railroad tracks west of Main Ave in
Norwalk.  Extraction wells at this network will be installed to
capture contaminants along the entire width of the groundwater
contamination plume at the railroad tracks.  The remediation well
networks together may include as many as 25 wells, possibly more,
to remove contaminants from both the overburden and bedrock
aquifers.  Bedrock wells may extend as far as 250 feet or more
below ground surface.  Extraction rates for the system have been
estimated at approximately 370 gallons per minute.  The specific
pumping rates, number of wells and locations of wells will be
developed during design.

     Air stripping will be utilized to treat the groundwater to
remove VOCs.  The system would utilize carbon adsorbtion for
vapor treatment for protectiveness consistent with OSWER
Directive 9355.0-8 and Region I guidance.  Pretreatroent to remove
solids may be required prior to air stripping.  Any pretreatment
sludge will be tested to determine if it is subject to RCRA and
the RCRA Land Disposal Restrictions.  The sludge will be managed
in compliance with RCRA and the Land Disposal Restrictions if
they are applicable.

     The Proposed Plan indicated that four different treatment
technologies could potentially be used to address groundwater
contamination at the site.  These include air stripping,
ultraviolet (uv)/Chemical Oxidation, carbon adsorption and steam
stripping.  EPA's chosen groundwater treatment method in this ROD

-------
                                30

is air stripping.  EPA will consider, during the design phase,
the use of ultraviolet  (uv)/chemical oxidation, carbon adsorption
or steam stripping as alternative groundwater treatment
technologies  for Source Area groundwater if it can be
demonstrated  that these alternatives will meet the four statutory
requirements  for cleanups under CERCLA and will provide a similar
balance of the nine evaluation criteria as air stripping.

     The Proposed Plan indicated that a decision on the discharge
option for treated groundwater would be made during the design
phase from among four options.  These options include discharge
to the local  publicly owned treatment works, discharge to the
Norwalk River through a new pipe, discharge to the Morwalk River
through existing storm sewers or reinjection into the aquifer,
either solely or in combination.  A decision among these
discharge options will be made on the basis of the statutory and
regulatory criteria for CERCLA cleanups.  This decision will be
issued in a supplementary decision document.

     Predesign activities will include at a minimum: 1)
comprehensive groundwater sampling to refine and confirm the
nature and extent of contamination at the site in both the
bedrock and overburden; 2) an evaluation of the adequacy of the
Elinco groundwater extraction system to control the groundwater
contamination within the complex boundaries; 3) an evaluation of
the need for  pretreatroent (as described above) of extracted
groundwater prior to air stripping; 4) pump tests to determine
well yields,  remediation pumping rates, and location and number
of extraction wells; and 5) bench and pilot testing of the
presented and/or proposed treatment technologies.

     Groundwater monitoring of the overburden and bedrock
aquifers within the Source Area, and the Downgradient Area will
occur during  implementation of the remedy in order to determine
compliance with the cleanup goals.  A specific monitoring program
will be developed during design that will include overburden and
bedrock monitoring wells including, at the least, those wells
that have been installed as part of the remedial investigation.
Monitoring wells will be sampled at least quarterly.  Treatment
system influent and effluent concentrations will be monitored at
least once per day.  The objectives of monitoring are to define
the mass of contaminants extracted over the life of the remedy,
to evaluate the efficiency of the remedy, and to ensure
compliance with appropriate Federal and State requirements.
     Pumping rates at each extraction well will be monitored
during the course of the remedial action.  An adequate array of
piezometers will be installed to adequately measure in three
dimensions the hydraulic gradient between adjacent pumping wells
to insure an inward hydraulic gradient is maintained at each

-------
                                31

remediation well  network and that no portion of the plume is free
to migrate from the  Source Area.  Extraction wells may need to be
repositioned occasionally to maximize the extraction efficiency.

     Remediation  of  the Source Area can involve the intermittent
operation, or pulse  pumping, of the remediation well fields to
improve the efficiency of contaminant removal.  The purpose of
the pulse pumping is to allow a sufficient resting phase between
pumping phases when  1) contaminants can diffuse out of low
permeability zones and into adjacent high permeability zones, 2)
maximum concentrations can be achieved in high permeability
zones, or 3) non  aqueous phase liquid residuals can reach
equilibrium concentrations.  As a result, active phases will
remove a minimum  amount of contaminated groundwater at maximum
possible concentrations.  The cycle of pumping and resting will
continue for each extraction well until it is determined that
contaminant levels in the extraction wells are not rising above
the cleanup goals for individual contaminants.  At that point in
time, groundwater will be sampled to determine if the cumulative
carcinogenic risk falls within a risk range generally considered
by EPA to be protective at Superfund sites and does not exceed
unity on the Hazard  Index.  If the level of risk does not fall
within EPA's risk range or exceeds unity on the Hazard Index,
then additional treatment will be conducted in order to meet
EPA's risk levels.   A supplementary decision document will be
issued as appropriate.

     A completion monitoring program will be implemented which
will consist of quarterly monitoring for three years.  The
completion monitoring program will be defined during design and
refined during the remedy implementation.  The completion
monitoring program may require the installation of additional
wells.  This program will be implemented at the point the cleanup
goals are achieved at all wells that are part of the
implementation program.

     To the extent required by law, EPA will review the Site at
least once every  five years after the initiation of remedial
action at the Site if any hazardous substances, pollutants or
contaminants remain  at the Site to assure that the remedial
action continues  to  protect human health and the environment.
EPA will also evaluate risk posed by the Site at the conclusion
of the completion monitoring program to guarantee it is
protective (i.e.,  before the Site is proposed for deletion from
the NPL).


Rationale for selection

     The rationale for choosing the selected alternative is based
on an assessment  of  each criteria listed in the evaluation of
alternatives section of this document.  In accordance with

-------
                                32

Section 121  of CERCLA, to be considered as a candidate for
selection  in the ROD, the alternative must have been found to be
protective of human health and the environment and able to attain
ARARs unless a waiver is invoked.  In assessing the alternatives
that met these statutory requirements, EPA focused on the other
evaluation criteria, including, short term effectiveness, long-
term effectiveness, implementability, use of treatment to
permanently  reduce the mobility, toxicity and volume, and cost.
EPA also considered nontechnical factors that affect the
implementability of a remedy, such as state and community
acceptance.  Based upon this assessment, taking into account the
statutory  preferences of CERCLA, EPA selected the remedial
approach for the Site.


SUMMARY OF COMPARATIVE ANALYSIS OF ALTERNATIVES

     In this section, the four alternatives are compared against
each other in relation to the nine evaluation criteria.  A more
complete comparison of the Alternatives according to these
criteria is  provided in Table 12-20 and Section 12.3 of the
supplemental Feasibility Study.


Overall Protection of Hmpan Health and the Environment

     Section 121(d) of CERCLA provides that remedial actions
shall attain a degree of cleanup of hazardous substances,
pollutants,  and contaminants released into the environment and of
control of further release at a minimum which assures protection
of human health and the environment.  Alternatives I, II, V and
VI provide varying degrees of protection of human health and the
environment.  Although, institutional controls would be
implemented  (to prevent excavations in areas of soil
contamination and well installations and use in areas of
groundwater  contamination) under all these alternatives, such
institutional controls can fail.  Therefore, absent engineering
controls,  institutional controls fail to provide overall
protectiveness.  The use of institutional controls to restrict
use or access should not substitute for active response measures
(e.g. treatment and/or containment of source material,
restoration  of groundwaters to their beneficial uses) as the sole
remedy.
     Alternative I, the no-action alternative, would not be
protective of human health and the environment because
contaminants in the groundwater outside of the effective pumping
radius of the Kellogg-Deering Well Field would persist in dead
end bedrock fractures or migrate through the aquifer and

-------
                                33

discharge to the Norwalk River.  In addition, contaminants would
remain in the soils at the complex and would provide a continuing
source of groundwater contamination and soil gases.  Alternative
I would include institutional controls which, absent engineering
controls, fail to provide overall protectiveness.

     Alternatives II and VI provide for greater protection of the
environment than Alternatives I and V because the groundwater
extraction wells will remove significant amounts of contaminants
from the aquifer.  Alternatives II and VI would thereby reduce
the size of the contaminant plume, expedite aquifer restoration,
and provide greater overall protectiveness should institutional
controls fail.  Alternatives I and V would rely on natural
flushing for groundwater restoration.

     Alternatives V and VI would remove contaminants from soils
of the unsaturated zone below the complex thereby reducing
further the upward migration of contaminants into the indoor air
of the complex buildings.  If in the future the buildings and
parking lots were ever removed, Alternatives V and VI would
reduce the increased potential for downward migration of
contaminants to groundwater and would minimize the potential
human health risk from exposure to contaminated soils.

     Alternative VI would provide the greatest protection of
human health and the environment by removing contaminants from
soils in the unsaturated zone at the complex and by removing
contaminant mass and reducing contaminant concentrations within
the bedrock and overburden aquifers in the Source Area.


Compliance with Applicable or Relevant and Appropriate
Requirements

     Section 121(d) of CERCLA requires that remedial actions
comply with all applicable or relevant and appropriate Federal
and State standards, requirements, criteria or limitations that
apply to the Site.

     Alternatives I and V, the no-action and soil treatment
alternatives, rely on aquifer flushing and the treatment system
at the Kellogg-Deering Well Field to alleviate groundwater
contamination at the site.  These alternatives are less likely to
assure that groundwater in the Source Area will meet Safe
Drinking Water Act MCLs for groundwater within a reasonable
timeframe.  Alternative-VI is more capable of meeting groundwater
quality ARARs than Alternative II because the soil treatment
component of Alternative VI will reduce the potential for
additional leaching of contaminants from the unsaturated zone to
the saturated zone.

-------
                                34

     The air stripping system proposed under Alternatives II and
VI are equally capable of meeting Federal and State requirements
for air emissions.  These systems are also equally capable of
treating extracted groundwater to levels required for discharge
to surface water or groundwater.

     Alternative VI is the most capable of meeting the relevant
and appropriate RCRA closure requirement, especially the Closure
Performance Standard of 40 CFR Part 264.111
Long-term Effectiveness and Permanence

     Long-term effectiveness and permanence addresses the long-
term protection and reliability of an alternative and the
magnitude of residual risks.

     Alternative VI would remove the largest amount of
contaminants from the aquifer, would provide the greatest control
of migration of contaminants, and would restore the Source Area
portion of the aquifer in the shortest period of time.  In
comparison to Alternatives I, II and V, Alternative VI would be
the most capable of all the other alternatives evaluated in this
Record of Decision to meet soil and groundwater cleanup goals and
thereby provide a permanent remedy.

     Long-term environmental monitoring of indoor air and
groundwater would be implemented for each alternative to measure
effectiveness of the remedy and to identify potential receptors.
Institutional controls would be implemented with each
alternative, although such controls are not considered a
permanent remedy.  Alternative VI, which addresses contaminated
soils and groundwater in the Source Area, would require the least
in terms of long-term institutional controls.  Maintaining
institutional controls on private well use for the duration of
the aquifer restoration period would be more difficult under
Alternatives I and V because they would require a greater amount
of time to cleanse the aquifer through natural aquifer flushing
than with the aid of extraction wells.  Maintaining long-term
contaminated soil excavation institutional controls would be more
difficult under Alternatives I and II.

     Alternatives II and VI are capable of removing more
contaminants from the Source Area, and at a faster rate than
Alternatives I and V, which rely mostly on dilution and natural
aquifer flushing.  The natural aquifer flushing that would be a
part of Alternatives I and V are likely to result in increased
contaminant levels in downgradient portions of the groundwater
plume before cleansing by flushing is complete.  Therefore,
Alternatives II and VI would be more effective at restoring the
aquifer and reducing the long term risk to potential receptors
downgradient.  Alternative VI would be more effective than

-------
                                35

Alternative  II because the soil treatment component of
Alternative  VI will  reduce the potential for additional leaching
of contaminants  from the unsaturated zone to soils to the
saturated  zone.
Reduction of Toxicitv, Mobility or

     This evaluation criterion relates to the ability of a
remedial alternative to reduce the toxicity, mobility or volume
of hazardous substances at a site and to thereby control the
risks associated with such hazardous substances.  Reducing the
level of toxicity  in the aquifer and soils will reduce the public
health risk posed  to future users of the aquifer should
institutional controls fail.

     Alternative VI which is the only alternative that addresses
both contaminated  soil and groundwater would provide the greatest
and quickest reduction of mobility, toxicity and volume of
contaminants at the site.  Under Alternatives II and VI, toxicity
and volume of groundwater contaminants would be reduced by the
treatment.  The mobility of contaminants would be reduced by the
extraction wells.  Under Alternatives I and V, the toxicity and
volume would not be reduced.  Alternatives I and V would continue
to allow some of the contaminants to continue to migrate
doWngradient of the Source Area.

     Under Alternatives V and VI the toxicity and volume of soil
contaminants would be reduced by soil vapor treatment.  The
mobility of soil contaminants would be reduced by the extraction
wells.  Under Alternatives I and II, contaminants remaining in
the unsaturated zone could possibly migrate upwards into the
indoor air of the  complex buildings or percolate downward to the
water table (if the buildings and parking lots were removed).


Short-Term Effectiveness

     The short-term effectiveness criterion relates to the time
required to meet remedial objectives, and the impacts of the
implementation of  the remedy.

     The time required to meet the groundwater cleanup goals in
the aquifer cannot be determined with specificity for any of the
alternatives due to the probable presence of DNAPL's and the
complexity of the  fractured bedrock system.  However, the time
required to meet the groundwater cleanup goals would be
substantially longer under Alternatives I and V than under
Alternatives II and VI, which, in addition to natural aquifer
flushing, rely on  groundwater extraction wells to actively remove
contaminants from  the aquifer.

-------
                                36

     Implementation of Alternative I would have very minor short-
term impacts to workers and to the environment.  Implementation
of Alternatives II, V and VI could pose some risk to workers due
to the possible release of contaminant vapors during drilling and
installation of soil vapor and/or groundwater wells.  This
potential  for release would pose a minimal risk to the community.
Alternatives V and VI would require some coordination with
existing industrial activities within the complex buildings
during the construction of some of the soil vapor extraction
equipment.  The short-term adverse impacts associated with
Alternative VI would be greatest in comparison to the other
alternatives because Alternative VI involves the most
construction.

     Alternatives II and VI may, if no practicable alternative
exists, have short tern impacts on wetlands or the floodplain
during installation and operation of a discharge pipe or
injection  wells.  All impacts from these activities will be
mitigated  at the time the remedial action is completed.


Implementability

     Implementability addresses the ability to implement and
operate each alternative from design through construction and
operation  and maintenance.

     Alternative I would be the easiest to implement since long-
term environmental monitoring would be the principle element.
Alternatives II, V and VI would be more difficult to implement
because these alternatives would have substantially greater
design, construction, and operation and maintenance requirements
than Alternative I.  Also, Alternatives I, V and VI would require
trained personnel to perform operation and maintenance
activities.  Alternative VI would be the most difficult to
implement  since it includes both an in-situ vacuum extraction
system and groundwater extraction system.

     The equipment and resources under Alternatives I and II are
readily available.  The availability of vendors to implement the
in-situ vacuum extraction system for Alternatives V and VI is
limited.
cost

     The cost evaluation of each alternative is based on the
capital cost (cost to construct), long-term monitoring, operation
and maintenance (O&M), and present worth costs.  Table 5 presents
estimates of these costs for Alternatives I, II, V and VI.

-------
                                37

Community Acceptance

     This evaluation criterion addresses the degree to which
members of the local community support the remedial alternatives
being evaluated.  Three general groups from the community
commented on the remedial alternatives for the site.  These
groups include: local citizens, the Norwalk First Taxing District
(NFTD) Water Department and Potential Responsible Parties  (PRPs).

     Local citizens favor timely restoration of the aquifer to
the maximum extent practicable.

     The NFTD supports the in-situ vacuum extraction system for
soil treatment (Alternative V).  The NFTD has expressed concern
with regard to the cost-effectiveness of groundwater extraction
and treatment systems within the bedrock aquifer.

     PRPs have raised numerous technical and legal concerns that
tend to support Alternative I, the no-action alternative .  In
.general, many of the PRPs feel that the groundwater extraction
and treatment system currently being constructed by Elinco at the
complex is a sufficient effort toward aquifer restoration.


State Acceptance

     This evaluation criterion addresses the concern and degree
of support that the state government has expressed regarding the
remedial alternatives being evaluated.

     The State of Connecticut believes that Alternative VI is an
appropriate remedial action to address the Source Area, and that
this remedy is consistent with Connecticut's groundwater quality
standards and goals.  Connecticut disagrees with EPA on technical
aspects of the remedial action originally proposed for the
Downgradient Area.

     Technical comments presented by DEP on the Downgradient Area
will be discussed and evaluated during the remedy selection
process for the third operable unit.  A letter of concurrence
with this Record of Decision for the second operable unit
(consisting of the source area)  is attached as Appendix C.

Conclusion

     Alternative VI, the selected remedy, provides the best
balance of tradeoffs among the alternatives with respect to the
evaluation criteria.

-------
                                38

STATUTORY DETERMINATIONS

     The remedial action selected for implementation at the
Kellogg-Deering Well Field Superfund Site is consistent with
CERCIA and, to the extent practicable, the NCP.  The selected
remedy is protective of human health and the environment, attains
ARARs and is cost effective.  The selected remedy also satisfies
the statutory preference for treatment which permanently and
significantly reduces the mobility, toxicity or volume of
hazardous substances as a principal element.  Additionally, the
selected remedy utilizes alternate treatment technologies or
resource recovery technologies to the maximum extent practicable.

Protection ef Human Health and the Environment

     The remedy at this Site will permanently reduce the risks
presently posed to human health and the environment by removing
contaminants from soils in the unsaturated zone beneath the
buildings and parking lots at the Elinco/Pitney Bowes/Matheis
Court complex and by removing contaminant mass and reducing
contaminant concentrations within the bedrock and overburden
aquifers in the Source Area.  This remedial action will:

     *    reduce the level of contamination in the soil at the
          Source Area to cleanup goals, thus preventing  exposure
          to contamination that may present a risk to human
          health and the environment if any excavation occurred
          at the Complex;

     *    eliminate the leaching of soil contamination to the
          groundwater at levels in excess of groundwater cleanup
          goals;

     *    reduce the level of contamination in the soil at the
          Source Area, and thus eliminate the migration of soil
          vapors into the buildings at the complex and prevent
          human exposure to this vapor;

     *    reduce the contamination in the bedrock and overburden
          aquifers in the Source Area to cleanup levels;

     *    prevent the migration of contaminated groundwater from
          the Source Area;

     *    reduce the contamination in the aquifer which will
          decrease long-term reliance on institutional controls
          on future well installations to protect public health;

     *    minimize the future introduction of contaminated
          groundwater from the bedrock aquifer to the overburden
         . aquifer supplying the Kellogg-Deering Well Field; and

-------
                                39

      *     reduce  the  risk to  the  Norwalk River and the life it
           supports.

The Source Area groundwater will  be pumped and treated until MCLS
are net.   At  that point  in time,  groundwater will be sampled to
determine  if  the  cumulative carcinogenic risk falls within a risk
range generally considered by EPA to be protective at Superfund
-sites and  does not exceed unity on the Hazard Index.  If the
level of risk does not fall within EPA's risk range or exceeds
unity on the  Hazard Index, then additional treatment will be
conducted  in  order to meet EPA's  risk levels.  A supplementary
decision document will be issued, as appropriate.


The Selected  Remedy Attains ARARs

      This  remedy  will meet or attain all applicable or relevant
and appropriate federal  and state requirements that apply to the
Site.  Environmental  laws from which ARARs for the selected
remedial action for the  second operable unit at the Kellogg-
Deering Superfund Site include:

      *     Resource Conservation and Recovery Act (RCRA)

      *     Clean Water Act (CWA)

      *     Safe Drinking  Water Act

      *     Executive Order 11988 (Floodplain Management)

      *     Executive Order 11990 (Protection of Wetlands)

      *     Clean Air Act  (CAA)

      *     Occupational Safety and Health Administration
           (OSHA)

      *     Connecticut Water Quality Standards and Classification

      *     Connecticut Standards for the Quality of Public
           Drinking Water

      *     Fish and Wildlife Coordination Act

      *     Connecticut Public  Health Code

      *     Connecticut Inland  Wetlands and Water Courses
           Regulations

      *     Connecticut Hazardous Waste Facility Siting Rules

      *     Protection  of  Archaeological Resources .

-------
                                40

     *    Connecticut  Air Pollution Control Regulations
     *    Connecticut  Water Diversion Regulation
     *    Connecticut  Discharge Permit Regulations
     *   . Connecticut  Hazardous Waste Rules
     *    Federal  Department of Transportation Rules for
          Transportation of Hazardous Materials
     Table  6 provides  a  brief synopsis of the requirements and
outlines the action  to be taken to attain these ARARs.
     The list  below  includes some of policies, criteria, and
guidances that will  also be considered (TBCs) during the
implementation of  the  remedial action:
     *    Connecticut  Water Quality Management Plan
     *    Connecticut  Draft Clean Water Strategy
     *    Connecticut  Water Quality Standards and Classification
     *    EPA  Risk Reference Doses (RfDs)
     *    EPA  Carcinogen Assessment Group Potency Factors
     *    RCRA Proposed  Air Emission Standards for Treatment
          Facilities (52 FR 3748, Feb. 5, 1987)
     *    EPA  OSWER  Directive 9355.0-28; Air Stripper Control
          Guidance
     *    EPA  Groundvater Protection Strategy
     *    Connecticut  Standards for the Quality of Public
          Drinking Water
     *    EPA  Health Advisories and Acceptable Intake Health
          Assessment Documents

     *    Safe Drinking  Water Act Maximum Contaminant Level Goals
          (MCLGs)
     *    OSHA Threshold Limit Values (TLVs)
Additional  information regarding TBCs can be found in Tables 8-1,
8-2, 8-3, and  12-15  of the  supplemental Feasibility Study.

-------
                                41
The  Selected  Remedial Action  ia Cost Effective

     Once  EPA identifies alternatives that are protective of
human health  and the environment and attain ARARs  (unless a
waiver  is  invoked), EPA evaluates each of those alternatives to
determine  their cost effectiveness.  Each of the alternatives
underwent  a detailed cost analysis to develop costs to the
accuracy of -30 to +50 percent.  In that analysis, capital and
operation  and maintenance costs have been estimated and then used
to develop present worth costs.  In the present worth analysis,
annual  costs  were calculated  for thirty years (estimated life of
an alternative) using a five  percent interest rate factor and
were based on 1988 costs.

     Of those remedial alternatives that are protective and
attain  ARARs,  and satisfy the preference for treatment to the
maximum extent practicable, EPA selected a remedy that is cost-
effective  in  mitigating the risks posed by the soil and
groundwater in the Source Area within a reasonable period of
time.   Overall, the total cost (present worth) of the selected
remedy  is  estimated at $9,100,000, as presented in Table 4.

     This  cost is:higher than that of some of the other
alternatives;  however, none of the less expensive technologies
can  ensure that the treated soil and groundwater will reach the
target  cleanup goals and also provide long-term effectiveness and
permanence.   Additionally, EPA has determined that this remedy
will yield results that are in proportion to its cost in terms of
effectiveness.  Thus, while other alternatives evaluated are
cheaper than  the selected alternative, they do not provide the
same degree of effectiveness.

     Table 4  provides an itemized breakdown of the total cost of
the  remedy by elements, capital cost, operation and maintenance
costs and  present worth.  Some changes in cost figures may be
made as a  result of the remedial design and construction
processes.
The Selected Remedy utilizes Permanent Solutions and Alternative
Treatment Technologies or Resource Recovery Technologies to the
Maximum Extent Practicable

     The selected remedy utilizes permanent solutions and
alternative treatment technologies or resource recovery
technologies to the maximum extent practicable.

-------
                                42

     In-situ vacuum extraction and treatment with carbon
adsorption  is an alternative treatment technology which provides
permanent removal of the mass of volatile organic contamination
in soil, thereby permanently and significantly reducing the
toxicity, mobility and volume of contamination.  Contaminant
reduction efficiencies of 99.99% have been achieved at other
sites using in-situ vacuum extraction with treatment by carbon
adsorption.

     The groundvater extraction/treatment portion of the selected
remedy also provides permanent removal and reduction of the mass
of volatile organic contaminants in groundvater through
groundvater recovery and treatment via air stripping and carbon
adsorption.  Carbon columns vill remove contaminants from the
airstream before being released to the atmosphere.  Treated
groundvater vill be discharged in compliance vith discharge
limitations required by ARARs.
The Selected Remedy Satisfies the Preference for Treatment as a
Principal Element

     The principal element of the selected source control remedy
for contaminated soil is in-situ vacuum extraction and carbon
adsorption. The principal elements of the selected management of
migration component for contaminated groundvater are air
stripping and carbon adsorption.  These elements are all
technologies that use treatment to address the human health and
environmental threats at the Site resulting from contamination of
soil and groundvater.


STATE ROLE

     The Connecticut Department of Environmental Protection has
revieved the various alternatives and has indicated its support
for the selected remedy for the Source Area.  The State has also
revieved the supplemental RI, Risk Assessment and FS to determine
if the selected remedy for the Source Area is in compliance vith
the applicable or relevant and appropriate State environmental
lavs and regulations.

     Connecticut concurs vith the selected remedy for the second
operable unit of the Kellogg-Deering Well Field Superfund Site,
vith the understanding that a third operable unit consisting of a
Dovngradient Area betveen the source and the veil field vill be
addressed separately.  Connecticut concurs that this remedy
attains State ARARs.

-------
                               vfptJ-i
                               'MtSFOtf
   GENERAL LOCATION OF
 KELLOGG-PEERING WELL FIELD
•'• ^^-^-^^^^^ ^£'-
••;• ^S^^^s^SSL^m
w TO3iiF ;if^p^^»N
4&?f;% ••CTws§raeHS
                          Wrniwi.!!
                        "K" I T\
BASE MAP IS A PORTION OF THE U.S.G.S. NORWALK NORTH, CONN.- N.Y. AND NOR'wALK SOUTH, CONN. QUADRANGLES
(7.5 MINUTE SERIES, I960. PHOTOREVISED 1971 ) CONTOUR INTERVAL 10*.
SOURCE- NUS CORPORATION 1966
              NOTE- SEE FIGURE 1-2 FOR DETAIL.
       GENERAL LOCATION MAP
  KELLOGG -PEERING WELL FIELD. NORWALK. CT
          SCALE: l's 2000'

-------
                             44
                         FIGURE 2
                   Kollogg-Dttrlng Sltt Location Map
          Amomun truot MU MUNOMT
          Ktueaa-Aunma wtu mi*
•  KlUOaa-OIUMM WIUS
                          FIGURE 3
              General Cross-Section Diagram from
                    Line Looking Northwest
Ktllogg-OMring

-------
                              FIGURE 4

 Generalized Map of Areas of Groundvater Contamination  at the
Kellogg-Deering Well Field  Superfund site
              •r*u»«vttir vitk TOI •o»o»mtrttl»»§
                  tkaa
                     «ttk Kl ••••
                                                       SCALE IN FEET
             NARJIOAO

-------
          TABLE 1
      INDICATOR CHEMICALS
KELLOGG-DEERING SUPERFUND SITE
     NORWALKr CONNECTICUT
Indicator Chemicals
Carcinogens
tetrachloroethene
trichloroethene
1,1-dichloroethene
vinyl chloride
chloroform
methylene chloride
1 , 1-dichloroethane
1,2-dichloroethane
benzene
Noncarcinogens
1 , 1 , 1-tr ichloroethane
1,2-dichloroethene
toluene
ethylbenzene
xylenes
acetone
4rmethyl-2-pentanone



-------
                             TABLE 2

                 SUMMARY OF REMEDIAL ALTERNATIVES
Remedial
Alternative Soil Source Downgradient
I
n
m
IV
V
^ VI
>• vn
vm




•
•
•
•

•
•


•
•



•
•


•
•
Blank Spaces indicate no action
  •  Soil treatment using in-situ vacuum extraction
  •  Groundwater extraction/treatment/discharge
  >  EPA's Preferred Alternative in July 1989 Proposed Plan
  •»  EPA's selected Alternative

-------
               48
Groundvater and Boil Cleanup Goals
   KELLOGG-DEERING SOPERFUND
        NC?.W;.LK, CONKICTI CUT
SITE
Contaminant
benzene
toluene
ethylbenzene
1,1, 1-tr ichlorocthane
1 , 2-dichloroethane
tetrachloroethene
trichloroethcne
1 f 2-dichloroethene
1 , 1-dichloroethene
vinyl chloride
4-methyl-2-pentanone
Groundwater
Cleanup Goal
(ug/D
5
2,000
680
200
5
5
5
70
7
2
350
Soil Cleanup Goal
(ugAg)
1.2 - 36.7
5,523 - 169,552
13,771 - 422,750
560 - 17,332
0.6 - 7.9
33 - 1,036
12 - 358
76 - 2,321
8.3 - 256
0.3 - 9
1,246 - 38,243

-------
                                   49
                                TABLE 4
Estimated capital,  o&M, and net present worth costs for the
Selected Remedy  (Alternative VI)  for the  second operable unit at
the  Kellogg-Deering Well Field Superfund  Site.
 Alternative VX:
      •  Soils  - Jo-situ Treatment
      •  Groundwater - Source area treatment  using air  stripping
 1.  Capital Cost
Item
In-situ vapor extraction
Air stripping system (source area)
River discharge system
Total
Cost
$1,396,000
3,508,000*
1.133.000
$6,037,000*
 2.  0(M Costs
Task
Air Monitoring
Groundwater
Monitoring
Outfall Monitoring
5-Year Site Review
Energy
Maintenance
Operators
(8 man-hr/day)
Total OfcM
Present-Worth
Year
1
2-30
1-30
1
2-30
1-30
1-30
1-30
1-30

Cost per
Event
$4,650
1,000
6,600
350
350
20,000
57,900
62,000
60,000

Frequency
Quarterly
Annually
Annually
Weekly
Monthly
Every
5 years
Annually
Annually
Annually

Present-Worth
5% for 30 years
$18,600
14,300
101,200
18,200
61,600
55,600
890,000
953,000
922.400

$3,034,900
 2.  Total NPW
[capital + OEM NPW
$9,100,000*
   Rounded up from $9,071,900.

-------
                     50
                   TABLE 5




COMPARISON OF ESTIMATED COSTS OF ALTERNATIVES
•tternativa
-
I
n
V
«t
Capital
Costs
f 11. 0001
AabMH**
0
4M1
1396
6,037
Long-Tar* Ground
Watar Monitoring
f ivofloi

101
101
101
101
Long-Tar* Air
Monitoring
{11.0001
AA
• 1
87
49
33
Annual
Ot M
(f 1.0001


200
13
200
OM
30-yaar
f 81.0001
238

3,088
206
3.034
Net Prasant
Worth Cost
ftl.OQO)
2&0

7,700
1,600
9,100

-------
                                           TABLE  6
               APPLICABLE OR RELEVANT AND APPROPRIATE REQUIREMENTS  (ARARS)
                             FOR THE SECOND OPERABLE UNIT AT
                       THE KELLOGG-DEERING WELLFIELD SUPERFUND SITE
ARARS
Federal Safe Drinking
Water Act regulations
establishing maximum
contaminant levels (MCLs)
(40 CFR 141.11-141.16)
Federal Resource Conser-
vation and Recovery Act
(RCRA) Groundwater Pro-
tection Standard (40 CFR
Part 264, Subpart F)
Connecticut Water Quality
Standards and Classifica-
tion (22a-426)
REQUIREMENT SYNOPSIS
Establish contaminant
concentration levels
in public drinking water.
They are relevant and
appropriate to the aquifer.
Established for ground-
water monitoring at RCRA-
permitted treatment,
storage and disposal facil-
ities.  This standard is
relevant and appropriate
to the aquifer.

Establishes designated
uses for surface water
and groundwater.  This is
applicable to aquifer
restoration and to any
discharges to the Norwalk
River and the aquifer.
ACTION TO BE TAKEN
TO ATTAIN ARARS

MCLs for indicator compounds
are the target cleanup levels
for groundwater at the source
area.  Attaining the Soil
Cleanup Goals will ensure that
any future migration of resi-
dual contaminants in the soil
will not exceed MCLs in the
source area.

Treatment will be conducted to
achieve RCRA MCLs in ground-
water.  Groundwater monitoring
will be conducted in com-
pliance With 264.97, 264.98
and 264.99.
MCLs and public health code
levels will be attained to
restore the aquifer to its
designated use as a drinking
water aquifer and surface
water discharges will meet
NPDES limitations.

-------
ARARS

Standards for Quality
of Public Drinking Hater;
Connecticut Public Health
Code
Federal Clean Water Act
(CWA) 404 (33 U.S.C. 1344,
40 CFR 230)
       Table 6 - page
REQUIREMENT SYMQPSIS
Federal Fish and Wildlife
Coordination Act, 16 U.S.C.
661
RCRA Location Standards
(40 CFR 264.18)
Establishes water quality
standards for water sup-
plies.  (CT Public Health
Code Regulation 19-13-B102)
This requirement is relevant
and appropriate to the aquifer.
ACTION TO BE TAKEN
TO ATTAIN ARARS
Cleanup of the aquifer will be
conducted in accordance with
these standards.
This requirement is appli-
cable to any dredging or
filling activities in the
wetlands or the river.
This requirement is appli-
cable to any action which
modifies the Norwalk River.
This requirement is appli-
cable if a RCRA facility
is located in the 100 year
floodplain.
If any dredging or filling
action is required in the
wetlands then Section 404 will
be compiled with.  The govern-
ing regulations favor practi-
cable alternatives that have
less impact on wetlands.  If
no mitigated practicable
alternative exists, impacts
must be mitigated.

If it is determined that the
remedial action potentially
modifies the Norwalk River,
actions will be coordinated
with the U.S. Fish and
Wildlife Service,

It is assumed that remediation
facilities will be located
outside floodplains.
Temporary staging areas or
remediation facilities that
are located by a floodplain
will be designed to allow
quick mobilization out of the
area and to prevent damage
caused by initial floodwaters.

-------
ARARS


Wetlands Executive Order
(E.O. 11990)
Floodplalns Executive Order
(E.O. 11988)
Connecticut Inland Wetlands
and Water Courses Regula-
tions  (Title 22a)
       Table 6 - page 3

REQUIREMENT SYNOPSIS
Under this Executive Order,
federal agencies are re-
quired to minimize the des-
truction, loss or degrada-
tion of wetlands, and pre-
serve and enhance natural
and beneficial values of
wetlands.  If no practi-
cable alternative exists,
impacts must be mitigated..
This is applicable to any
activity in the wetland.

Federal agencies are re-
quired to reduce the risk
of flood loss, to minimize
impact of floods, and to
restore and preserve the
natural and beneficial
value of floodplains.
This is applicable to any
activity in the floodplain.

These requirements are
applicable for alternatives
that affect wetlands.
These requirements limit
activities that deposit
material in, alter, or
pollute inland wetlands
and water courses.
ACTION TO BE TAKEN
TO ATTAIN ARARS

EPA will choose a discharge
method for treated groundwater
that avoids impacts to the
wetlands.  If no practicable
alternative exists, impacts to
the wetlands will be miti-
gated.
EPA will choose a discharge
method for treated groundwater
that avoids impacts to the
floodplain.  If no practicable
alternative exists, impacts to
the floodplain will be miti-
gated.
These requirements will be met
for any activity in the
wetland at the site.
                                                                                                01
                                                                                                U)

-------
                                     Table 6 — page 4
ARARS
Connecticut Hazardous Haste
Facility Siting Rules
(Title 22a)

Federal Clean Water Act
(40 CFR 122-125) National
Pollutant Discharge and
Elimination System (NPDES)

Occupational Safety and
Health Act (OSHA) - General
^Industry Standards (29 CFR
1910)
Occupational Safety and
Health Act  (OSHA) -
Safety and  Health
Standards   (29 CFR 1926)
Occupational Safety and
Health Act  (OSHA)  - Record-
keeping, Reporting and
Related Regulations (29 CFR
1904)
REQUIREMENT SYNOPSIS
This requirement is appli-
cable where more stringent
than Federal Regulations.

These requirements regulate
discharges to surface water
are applicable to treated
groundwater at the site.

This applicable regulation
specifies the 8-hour, time-
weighted average concentra-
tion for various organic
compounds and 2 PCB com-
pounds; site control proce-
dures; training and protec-
tive clothing requirements
for worker protection at
site reinediations.

This applicable regulation
specifies the type of safety
equipment and procedures to
be followed during con-
struction and excavation
activities.

This applicable regulation
outlines the recordkeeping
and reporting requirements
for an employer under OSHA.
ACTION TO BE TAKEN
TO ATTAIN ARARS

These requirements will be met
when siting the treatment
facility.

Water discharges from the
treatment systems will be in
compliance with the Clean
Water Act.

Proper respiratory
equipment will be worn if it
is not possible to maintain
the work atmosphere below
these concentrations.
All appropriate safety equip-
ment will be on site pro-
cedures will be followed dur-
ing groundwater monitoring.
These regulations are appli-
cable to the company con-
tracted to execute site reme-
diation.

-------
ARARS
Clean Air Act (CAA) -
National Air Quality Stan-
dards for Total Suspended
Particulates (40 CFR
129.105, 750)
Protection of Archaeo-
logical Resources
(32 CFR 229)
Connecticut Air Pollution
Control Regulations
(22a - 17a)
Connecticut Water Diversion
Regulation (22a-365)
       Table 6 - page 5

REQUIREMENT SYNOPSIS
This applicable regulation
specifies maximum primary
and secondary 24-hour con-
centrations for particulate
matter.
These regulations develop
procedures for the protec-
tion of archaeological
resources.
These regulations are
applicable during treat-
ment operations.
Diversion of water of the
state will be allowed only
when such diversion is
found to be necessary and
is compatible with the
state's long-range water
resource planning and
management.  These regula-
tions are applicable.
ACTION TO BE TAKEN
TO ATTAIN ARARS

Fugitive dust* emissions from
site activities will be main-
tained below 150 ug/m3 (sec-
ondary standard) by water
sprays and other dust
suppressants.

If archaeological resources
are encountered during exca-
vation, these regulations will
be applicable and the area
must be reviewed by federal
and state archaeologists.

Air emission limits from the
the treatment systems will be
in compliance with the
Connecticut Air Quality
Standards.

The groundwater extraction
system will be operated as to
minimize impacts on the well
field and Norwalk River flow
rate.  Partial shut down of
extraction wells may be re-
quired during extremely dry
precipitation events.
o
u

-------
ARARS

Connecticut Discharge
Permit Regulations
Connecticut Hazardous
Waste Rules (22a-449)
(Title 22a-430)
FEDERAL DOT - Rules for
the Transportation of
Hazardous Materials (49
CFR 107, 171.1 - 171.500)
CWA Pretreatment Regula-
tions  (40 CFR 403)
      Table  6  - page 6

REQUIREMENT SYNOPSIS

These applicable require-
ments supplement the CWA
NPDES permit requirements.
This regulation outlines
requirements for the con-
struction, operation, and
location of a hazardous
waste facility.  These
regulations, where more
stringent than federal re-
quirements, are applicable
to the site.

Specific requirements for
markings, vehicle regis-
tration, manifests, and
transportation of hazar-
dous wastes and chemical
substances are applicable.

This applicable regulation
specifies pretreatment
standards for discharges to
publicly-owned treatment
works (POTWs).
ACTION TO BE TAKEN TO
ATTAIN ARARS
Groundwater treated onsite and
discharged to a surface water
will need to comply with the
water quality standards and
complete routine monitoring
and recordkeeping activities.

These requirements will be
complied in constructing and
operating the treatment
systems.
All on- and off-site transport
of excavated soil treatment
sludgesf and other materials
will follow these standards.
If treatment system discharge
is sent to a POTW, the POTH
must have an approved pre-
treatment program and the
discharge must comply with the
POTH program.
                                                                                                  in
                                                                                                  
-------
ARARS
RCRA Facility Standards and
Land Disposal Restrictions
(40 CFR 264)
RCRA General Facility
Requirements (40 CFR
264.10-264.18)
RCRA - Preparedness and
Prevention  (40 CFR 264.30-
(40 CFR   264. 30-37)
       Table 6 - page 7

REQUIREMENT SYNOPSIS
This regulation is appli-
cable to land disposal of
certain RCRA wastes and of
materials contaminated with
or derived from RCRA wastes.

General facility require-
ments outline general waste
security measures, inspec-
tions, and training require-
ments.  These regulations
are applicable.
This regulation outlines
requirements for safety
equipment and spill con-
trol.  These requirements
are applicable.
ACTION TO BE TAKEN
TO ATTAIN ARARS

These requirements will be
complied with in disposing of
material on-site and off-site,
Substantive requirements will
be applicable to PTD and vapor
extraction facility.  Periodic
analysis of the influent
groundwater and treatment
residues will be conducted in
accordance with 264.13.
Locked facilities, fencing and
the presence of operators
should be in accordance with
264.15.  The facility will
undergo regular inspections by
the operators in accordance   :
with 264.15.  The operators
will, be trained in accordance
with 264.16.  Any of the PTp
systems effluent piping near
the river will be designed in
accordance with 264.18.

Substantive requirements will
be applicable to PTD and vapor
extraction facility.  The
treatment facilities will be
designed and operated in
accordance with 264.31.

-------
ARARS
RCRA Contingency Plan
and Emergency Procedures
(40 CFR 264.50 - 264.56)2
RCRA Manifesting, Record-
keeping, and Reporting
(40 CFR 264.70 - 264.77)2
RCRA Container Require-
ments (40 CFR 264,
Subpart I)
       Table 6 - page 8

REQUIREMENT SYNOPSIS
This regulation outlines
the requirements for emer-
gency procedures to be
used following explosions,
fires, etc.  These require-
ments are applicable.

This regulation specifies
the recordkeeping and re-
porting requirements for
RCRA facilities.  These
requirements are applicable.
This regulations sets forth
RCRA requirements for use
and management of con-
tainers at RCRA facilities.
These requirements are
applicable.
ACTION; TO BE TAKEN
TO ATTAIN ARARS
Support equipment in com-
pliance with 264.32 will be
provided for the facilities.
This equipment will be tested
and maintained in accordance
with 264.33.  State and local
authorities will be educated
as to th« hazards at the
facilities in accordance with
264.37.

Substantive requirements will
be applicable to PTD and vapor
extraction facility.  Contin-
gency plans will be developed
in accordance with 264.50 to
264.55.

Substantive requirements will
be applicable to PTD and vapor
extraction facility.  Record-
keeping utilized at the
facilities will be in
accordance with 264.73 and
264.74.

Packing and accumulation of
treatment sludges and other
materials will adhere to
these standards.
                                                                                                 00

-------
ARARS
       Table 6 - pace 9

REQUIREMENT SYNOPSIS
ACTION TO BE TAKEN
TO ATTAIN ARARS
RCRA - Closure and Post-
Closure (40 CFR 264.111)
RCRA - Tanks (40 CFR
264.190-264.199)
^Connecticut Public Health
Code (19a-36)
These requirements are
relevant and appropriate
to soil cleanup and
groundwater.

This requirement may be
applicable to tanks of
treatment systems.
Provides controls to
restrict groundwater use as
potable water.  This re-
quirement is applicable to
the site.
The facilities will be closed
in accordance with the closure
performance standard 264.111.
This requirement will be com-
plied with to the extent that
it is an ARAR which depends on
discharge option implemented.

This requirement will be com-
plied with to restrict private
well use at the site.

-------
          RECORD OF DECISION
KELLOGG-DEERING SITE - OPERABLE UNIT 2
            SOURCE CONTROL
              APPENDIX A

        RESPONSIVENESS SUMMARY

-------
                         KELLOGG-DEERING
                      RESPONSIVENESS SUMMARY
I.   INTRODUCTION

This responsiveness summary addresses the comments received by
EPA on the Supplemental Feasibility Study and Proposed Plan for
the Kellogg-Deering Well Field Superfund site in Norwalk,
Connecticut.  The comments were received from citizens, the State
of Connecticut  and also from a number of potentially responsible
parties  (PRPs)  during the 30-day comment period held from July
27, 1989, to August 25, 1989.

Written  comments were received from ten parties and two people
presented oral  comments at the public hearing held in Norwalk on
August 14, 1989.  The transcript of the public hearing is
provided as Exhibit B to this Responsiveness Summary.  The
comments are summarized and organized by subject.  EPA responses
to each  comment, or set of like comments, are provided.

As a result of  the decision made in this Record of Decision (ROD)
to create a third operable unit to address the Downgradient Area,
EPA did  not address the comments that were received pertaining to
the Downgradient Area at this time.  At the time a final decision
is made  on the  remediation, the Downgradient Area comments that
have been received will be more thoroughly responded to.


II.  BACKGROUND ON COMMUNITY RELATIONS ACTIVITIES

Community concern and involvement at the Kellogg-Deering Well
Field Superfund site has fluctuated since the Well Field was
first discovered to be contaminated in 1975.  Initial concern in
the community was very high, as residents feared they had been
using unsafe drinking water.

Public interest escalated in 1983 when the site was proposed to
the National Priorities List of Superfund sites that are eligible
for federal funding for investigations and cleanups.  A local
group of citizens, known as the Waterforce, became active in
monitoring EPA  cleanup plans.  In 1984, EPA prepared a community
relations plan  to assess community concerns and develop
activities to respond to informational and public involvement
needs related to those concerns.  Most community concerns related
to possible adverse health effects, the adequacy of the water
supply,  the need for more public information on the site, and
concern  about the length of time necessary for studies before
cleanup  action  could be taken.  A public meeting was held in July
of 1984  to answer questions from about 60 people who attended the
meeting.

-------
The community  relations plan was updated in June of 1986.
Although  citizen  concern vac not as intense as it had been when
contamination  was first discovered at the Well Field, citizens
interviewed  said  that having a safe drinking water supply was
still their  principal concern.  Other issues noted by residents
included  concern  about the  following:

     • future  constraints on growth caused by water supply
          problems;
     - possible dangerous emissions from the air stripper
          installed to treat the water supply wells;
     - potential  spread of  contamination as a result of the
          construction of Route 7 where it passes near the well-
          field;
     - protection of wetlands on the site;
     - monitoring of the quality of the water supply by the
          Norwalk First Taxing District; and
     - the amount of time spent studying the site.

The community  has not been  visibly active in raising concerns
about the site since initial questions were responded to in 1984
and the Norwalk First Taxing District began its own well-head
treatment program in 1986 to control the pollution of the water
supply wells.  The community has, however, expressed an interest
in being  informed of activities at the site and the schedule for
decisions to be made to advance site cleanup.

To keep the  public informed and involved in decision points in
the Superfund  process, EPA  has held numerous public meetings and
issued written material and press releases to the Norwalk
community.   A  complete chronology of community relations
activities conducted by EPA is attached as Exhibit A of this
Responsiveness Summary document.  After EPA mailed a fact sheet
and held  public meetings to explain the first of two remedial
investigations and EPA's options for a well-head treatment
system, EPA  invited public  comment.  During the public comment
period conducted  in 1986, EPA received one citizen comment and
several comments  from potentially responsible parties at a public
hearing and  in writing.  A  Record of Decision to improve the
existing  well-head treatment system was approved in September
1986 as the  first step in the cleanup of the Xellogg-Deering
site.

There has been little or no community action relative to EPA
activities since  the 1986 Record of Decision was signed.  In
January 1988 EPA  released a fact sheet on the results of a study
to determine the  sources of groundwater contamination.  In July
1989 EPA  mailed a proposal, the Proposed Plan, to address the
sources of contamination to interested and affected individuals.
The release  of the Proposed Plan was followed by a public
informational  meeting and a public hearing on the proposal.
Approximately  30  people attended the meeting and the hearing.

-------
The comments received from citizens, local officials, the state
of Connecticut, and potentially responsible parties are
summarized and responded to in the following sections of this
document.  A transcript of comments made at the public hearing
held August 14, 1989, is attached as Exhibit B of this
Responsiveness Summary.


III. SUMMARY OF COMMENTS AND EPA RESPONSES

A.  SUMMARY OF COMMENTS FROM PRPs

Provided below are comments from parties that EPA considers to be
potentially responsible parties (PRPs) for the site, and EPA's
responses to those comments.  The comments were provided by three
PRPs:  Electric Indicator Company, Inc. (Elinco), EDO Corporation
and Pitney Bowes.

Beveridge & Diamond and Versar commented on behalf of Elinco;
Sidley & Austin and Goldberg, Zoino & Associates (GZA) commented
on behalf of EDO Corporation; and Murtha,  Cullina, Richter and
Pinney, and Dames & Moore commented on behalf of Pitney Bowes.

All PRP commentors asked for an extension to the public comment
period, however none was granted by EPA.


Complex as a Source of Contamination

1.   Comment:  The remedial investigation and feasibility study
     (RI/FS) fails to demonstrate that the Elinco/Pitney
     Bowes/Mathies Complex (the Complex) is a major source of
     Well Field and aquifer contamination, basing its conclusion
     on general qualitative, not quantitative, analysis.  The
     bedrock between the Complex and the Well Field has been
     treated as a "black box" accounting for TCE migration,
     without showing actual transport mechanisms.

     The RI/FS found fractured bedrock to be the major migration
     pathway from the Complex to the northwest, west, and
     southwest, while explicitly acknowledging that bedrock
     fractures at Kellogg-Deering are tight and widely spaced,
     and that contaminant flow is therefore difficult to trace.
     Further, the RI/FS has failed to identify major bedrock
     fractures that can be shown to be carrying contaminants
     outward from the Complex, and instead relies on regional
     fracture orientation for this purpose.  It is technically
     just as plausible that the contaminants in the fracture came
     from other sources located in different directions and at
     different distances from that fracture zone.

-------
     EPA Response:  Groundwater data obtained by EPA indicates
     that the Complex is a significant source of contamination.
     Sampling and analysis demonstrated that a plume of
     contamination is emanating away from the Complex in a semi-
     radial pattern from the northwest to southwest.  Moreover,
     this testing indicates that contaminant movement occurs
     initially in the bedrock and then in both the overburden and
     bedrock aquifers.  The southwest extent of the plume is
     migrating below the Norwalk River and intersects the
     Kellogg-Deering Well Field.  The west and northwest edge of
     the plume extends to the Deering Pond and the Norwalk River.

     Both of EPA1s field investigations, documented in the
     initial and supplemental remedial investigation reports,
     show that groundwater contaminants are moving through the
     unconsolidated overburden aquifer at the Complex and through
     deep and shallow fractures of the bedrock aquifer to the
     sand and gravel aquifer that supplies the wells of the
     Kellogg-Deering Well Field.  Some of the contamination has
     entered bedrock fractures immediately below the buildings at
     the Complex.  Contaminants discharge from the bedrock into
     the overburden aquifer in the vicinity of the Norwalk River
     and the CL&P Landfill.  While the actual location of every
     fracture containing contaminated groundwater is uncertain
     and other sources of contamination may be present, EPA
     believes that a "piping system11 of bedrock fractures from
     the Complex allows contamination to migrate toward the
     Kellogg-Deering water supply wells and the data clearly
     indicates that the Complex is a major source of
     contamination.

2.   Comment:  Pitney Bowes indicated that they do not store,
     use, or dispose of significant quantities of TCE in the Zell
     Building Complex, and did not have TCE on-site until 1983-
     84.  Their entire inventory of TCE since occupying the
     building has totaled 55 gallons; one gallon purchased in
     1983, and 55 gallons purchased in 1984.

     EPA Response:  These comments relate to enforcement issues
     that are not the subject of this ROD. In this Responsiveness
     Summary, EPA must address significant comments relateing to
     the selected remedy.  Comments on enforcement issues need
     not be responded to in this Responsiveness Summary and con
     be dealt with during negotioations.

3.   Comments  Pitney Bowes provided a number of comments
     indicating that Elinco is the source of the TCE
     contamination found at the Complex.

     EPA Response:  These comments, like the previous one, relate
     to enforcement issues that are not the subject of this ROD.

-------
     Comments on enforcement issues are not the subject of this
     Responsiveness Summary.


Groundwater Contamination

4.   Conaents  The RI/FS does not actually identify a pathway to
     the wells west of the Norwalk River, but merely hypothesizes
     one by assuming unidentified localized open fractures.
     Until a link is proven, PRPs cannot be liable for Well Field
     contamination, given several precedents.

     EPA Response:  Determining if a given party is liable for
     the contamination of the Kellogg-Deering Well Field is not
     the subject of this Record of Decision, which focuses on the
     selection of a remedy for the site.  However, as indicated
     in EPA's response to Comment #1 above, sampling and analysis
     show that contaminants move through both the overburden and
     bedrock aquifers from the Complex toward the Well Field.

5.   Comment:  The RI/FS incorrectly uses the greatest value for
     hydraulic conductivity from tests.  The number used is both
     incorrectly calculated and is not representative of
     hydraulic conductivity throughout the study area bedrock.
     Similarly, the RI/FS used stratified drift hydraulic
     conductivity measurements from the Well Field and applied
     those throughout the study area, although hydraulic
     conductivity for stratified drift varies considerably
     throughout the study area.  Both approaches are technically
     inappropriate.

     An alternative model of hydraulic conductivity shows that
     any contaminant release from the Complex on the dates
     alleged in the RI/FS would not yet have arrived at the Well
     Field or at other areas of high contaminant concentration in
     the aquifer beyond the immediate vicinity of the Complex.
     This model is supported by EPA's tracer studies, which could
     detect no tracer salts downgradient from the Complex in the
     short term, and in only one instance in the long term.  The
     one positive sample suffered from three technical
     difficulties that make it an improper basis for establishing
     a migration pathway.

     EPA Response:  EPA recognizes that hydraulic conductivity
     may vary with the different lithology in the aquifer;
     however, the values EPA used and how they were applied were
     not inappropriate.   EPA utilized the average transmissivity
     and hydraulic conductivity values for the stratified drift
     deposits and the geometric mean for the conductivity in the
     bedrock aquifer.   For bedrock flow, a high value was used to
     illustrate the potential transport capabilities of the rock
     mass.

-------
The model EPA utilized is appropriate for this situation.
Although an alternative model that uses a lower hydraulic
conductivity value nay show that contamination from the
Complex could not have reached the Well Field, this is not
supported by extensive sampling and analysis that shows that
contaminants (primarily TCE), are move through both the
'overburden and bedrock aquifers toward the Well Field.
Moreover, the alternative model presented ignores highly
permeable bedrock fractures that could serve as a major
route for contaminant flow to the Well Field.

The alternative model, presented by the PRPs, which predicts
maximum travel distances of 125 meters for contaminants
discharged at the site for over 40 years, relies on
assumptions and input values that EPA believes are not
valid.  The model describes the glacial deposits in the area
adjacent to and down-gradient of the source area as till
that have very low hydraulic conductivities, when in fact
they are stratified glacial drift deposits with much higher
hydraulic conductivities than tills.  Also, the hydraulic
conductivity value used (0.35 ft/day) was incorrectly
calculated from an aquifer test (the "K-19A slug test1*).
The correct value, 144 ft/day, results in contaminant flow
rates that are about 250 times greater than predicted by the
alternative model.

Although it may be difficult to draw definite conclusions
from the tracer study conducted by EPA, the alternative
model presented is not necessarily supported by the study.

EPA believes the tracer study illustrated the complexity of
groundwater flow at the site, but did not demonstrate
definitively whether or not the Complex is a source of the
contamination found at the Well Field.  Additionally, EPA
did not rely on a single reading from the tracer study to
conclude that there is a hydrogeologic link between the
contamination found at the Complex and at the Well Field.
The tracer study was only one part of a very extensive
investigation conducted by EPA to define the nature and
extent of contamination found at the site.  In addition to
the tracer work, the lithology of the aquifer was studied,
and extensive sampling and analysis was conducted that
showed that contaminants move through both the overburden
and bedrock aquifers toward the Well Field.  In fact, the
distribution of TCE in the groundwater was the "tracer" that
EPA used to conclude that the Complex is a source of
contamination for the Well Field.

Commentt  The RI/FS concludes that the migration pathway is
demonstrated by a general increase in TCE concentrations in
five bedrock wells as one moves from the Well Field toward
the Complex.  However, the conclusion rests on an inadequate

-------
     data base, and in fact suggests that the contaminant flow is
     both more complex than suggested in the RI/FS, and may point
     to other contamination sources.

     EPA Response:  EPA concurs with the comment that groundwater
     flow and contaminant transport at the site are very complex
     due to the interaction of the overburden and fractured
     bedrock aquifers and does not dispute that there may be
     other sources that contribute to contamination in the
     aquifer and at the Well Field.  However, the data base that
     EPA utilized to determine that the Complex is a significant
     source of contamination and that this is affecting the Well
     Field water quality is adequate.  Analytical data collected
     from monitoring wells in and around the Complex show the
     highest concentrations of contamination.  This data, as well
     as groundwater level measurements throughout the area,
     clearly show that the Complex is a major source of
     contamination and that that contamination is flowing from
     the Complex toward the Well Field.

7.   Comment:  The areal extent of the source area groundwater
     remediation is too large.  The source area is defined as
     that area with TCE concentrations greater than 6,600 ppb
     because that is the maximum level EPA deemed the Well Field
     air stripper could safely treat.  However, the air stripper
     can easily be modified to treat higher concentrations,
     making possible the use of TCE concentrations above 6,600
     ppb as a source area boundary. If EPA's definition of the
     source area is based on health risks to Well Field users,
     the only receptors, given admittedly effective institutional
     controls, those risks are better and more cost-effectively
     addressed by modifications to the air stripper at the Well
     Field than by groundwater treatment at the source area. This
     is true because 1) air stripping at Well Fields is a tried,
     effective, and readily implementable remedial action, 2)  the
     Well Field air stripper can be modified easily and at low
     cost, and 3} the effectiveness of air stripping at the
     source area as proposed in the RI/FS is highly uncertain at
     best given the recognized difficulty of treating groundwater
     in bedrock fractures.

     Even if the 6,600 ppb delineation is judged appropriate,  the
     area designated in RI/FS is too large, as it includes many
     monitoring wells that do not show TCE concentrations above
     6,600 ppb. In particular, although EPA's mobile lab found a
     TCE concentration of 37,840 ppb in a sample from the K-8
     monitoring well 250 feet southwest of the Complex, a second
     sample taken at the same time and analyzed using more
     accurate contract laboratory procedures measured only 4,700
     ppb. Thus, it is inappropriate to extend groundwater
     treatment beyond the immediate vicinity of the complex.
                                8

-------
     XPA Response:  The Source Area groundwater extraction and
     treatment system will be developed to attain ARARs (e.g.,
     MCLs) in the aquifer. The area is defined by the 6,600 ug/1
     TCE contour line as shown in the supplemental RI/FS.  The
     areal extent of this system will be refined as part of the
     pre-design activities to maximize contaminant removal rates
     and to pinpoint the location and number of extraction wells.
     In any event, there will be two sets of extraction well
     networks to be installed:  one within the Complex boundaries
     and one near the railroad tracks to the west of the Complex.


8.   Commentt  EPA states without explanation that 80 percent of
     the soluble contaminants originating at the Complex have
     already migrated beyond the property line.  That statement
     is without apparent basis, in light of the much higher
     levels of TCE measured in monitoring wells at the Complex as
     compared to those in the downgradient portion of the source
     area.

     ZFA Response:  Although the highest concentrations of TCE
     and PCE are found at the Complex, most of the contamination
     is found outside of the its boundaries at lower concentra-
     tions, but still exceeding MCLs, in the huge volume of
     groundwater in the surrounding aquifer.  Because a
     significant portion of the contamination has moved from the
     complex, the pumping and treatment of the aquifer is
     necessary to achieve the cleanup goals.

9.   Commentt  Based on a computer model of contaminant
     transport, the cleanup goal for the source area bedrock
     aquifer should be 160 ppb.  As a result of dispersion and
     assuming no retardation, a TCE concentration of approxi-
     mately 160 ppb at the Complex would be reduced to 25 ppb at
     the Norwalk River.  Then, assuming 80% dilution of the
     groundwater by clean water from the Horwalk River
     (reasonable and similar to the dilution factor used by NUS),
     the contaminant concentration would be lowered to 5 ppb, the
     MCL for TCE.

     EPA Response:  The cleanup is based on achieving MCLs and
     state ARARs throughout the aquifer and not only at the Well
     Field.  With MCLs and state ARARs as the goal, the effect of
     dilution on the contaminant concentrations is not relevant
     to the aquifer cleanup levels.  In addition, RCRA
     regulations on closure require that the soils be
     decontaminated so as not to result in unacceptable
     groundwater contamination.

10~  comment:  Their appears to be some inconsistencies in the
     calculated dilution of water at the Well Field by the
     Norwalk River and in the assumptions that EPA presented in

-------
     the RI/FS about this.  EPA has not used the lowest
     assumptions in deriving its future (plausible case)
     scenario.  Neither future scenario depicts the true risks
     posed by the site and thus basing the need for further
     remedial action on either future scenario is unwarranted.

     EPA Response:  Dilution of the Well Field by the river is
     dependant on seasonal changes and the amount of pumping and
     types of wells being used.  Although dilution of
     contaminated groundvater due to recharge from the river will
     reduce contaminant levels, if the treatment system at the
     Kellogg-Deering Well Field were not in use through
     mechanical failure or otherwise, the contaminants in the
     aquifer could pose a risk to human health.  In addition, the
     cleanup must attain legally applicable or relevant and
     appropriate requirements set by state and federal laws,
     including MCLs set under the federal Safe Drinking Water
     Act.

11.  Comment:  In Section 12 of the RI/FS, EPA makes assumptions
     about how each remedy will reduce the toxicity, mobility, or
     volume of TCE.  These assumptions, however, are not
     supported by earlier statements in the RI/FS about the
     nature and extent of contamination.

     For example, on p. 176, the RI/FS states: "of the total
     quantity of contaminated groundwater, 15% is in the bedrock
     and 85% is in the overburden."  This statement contradicts
     the statement on page 147: "although contamination has been
     detected in both overburden and bedrock monitoring wells, it
     is apparent from the distribution of contamination that
     migration in the bedrock groundwater regime is the primary
     contaminant migration pathway."

     With such internally contradictory remarks, there is no way
     to adequately determine if EPA's preferred alternative is
     justified. The effectiveness of each remedial alternative
     will obviously be affected by the assumptions made
     concerning the nature and extent of contamination. Since
     EPA's preferred alternative relies on assumptions
     contradicted elsewhere in the RI/FS,  there is no rational
     basis for the alternative.

     EPA Response: There is no contradiction in the remarks that
     are quoted, in fact the two statements deal with different
     topics.  One addresses the distribution of contamination, a
     majority of which is found in the overburden, and the other
     deals with contaminant transport, which occurs primarily in
     the bedrock.

     EPA's cleanup plan for the site is justified and is based on
     a careful evaluation as outlined in the ROD.   That

                               10

-------
     •valuation includes a complete analysis of how the cleanup
     satisfies the statutory and regulatory criteria for
     selecting remedial actions under CERCLA.

12.  Consents  The preferred alternative assumes that no dense
     non-aqueous phase liquid (DNAPL or "free product") TCE
     exists in the groundwater beneath the Complex, but the RZ/FS
     concludes that some DNAPL TCE must exist there.  Because it
     may continue to release aqueous TCE, the presence of DNAPL
     TCE .may prevent the preferred alternative from accomplishing
     its 'objective, possibly to the point where TCE levels do not
     decrease.  Thus, a remediation plan such as the preferred
     alternative that does not address specific removal of "free
     product" from the bedrock aquifer will be no better than
     continuing the current treatment of groundwater by NFTD.

     BPA Response!  The remedy includes pre-design work to refine
     the understanding of the nature and extent of contamination.
     If free product, or DNAPL,  is found, the design of the
     groundwater treatment system will take this into account.
     Additionally, even if free product is found, the cleanup
     goals for the remedy will not change.

13.  Comment:  The RI/FS notes that because of complex
     hydrogeology and potential presence of DNAPL TCE, no
     estimates can be made for the natural flushing of the source
     area, which directly contradicts the logic of the preferred
     alternative.

     EPA Response:  Recognizing the real limitations of the
     understanding of the hydrogeology and distribution of
     contaminants does not contradict the logic of the cleanup
     plan.  The remedy is an aggressive plan to cleanup a
     contaminated aquifer.  Its design will take into account the
     complexity and uncertainties of the situation, including
     contaminant distribution and the site's hydrogeology.  As
     stated in the ROD, the estimated time for remediation may
     exceed 10 years.

14.  Comments  Given the area's complex hydrogeology, it is not
     known, and is not reasonable to assume, that the 250 ft.
     deep pumping wells will be in contact with the same bedrock
     aquifer now being monitored.  Further, without knowing where
     the DNAPL TCE exists, the deep wells could introduce TCE to
     greater depths and/or other aquifers. Even if the deep wells
     were to contact the aquifer now being monitored, it is not
     known whether they will be able to extract DNAPL TCE.  In
     short, pumpinj and treating the bedrock aquifer will not be
     effective using existing geologic information.

     EPA Responses   As responded to earlier, the remedy includes
     pre-design work to refine the understanding of the hydro-

                               11

-------
     geology and the nature and extent of contamination.  If free
     product, or DNAPL, is found, the design of the groundvater
     treatment system will take this into account.  Additionally,
     wells will be installed in a manner to ensure that
     contamination is not introduced from the overburden into
     bedrock.
Other Potential Sources of Contamination

15.  Comment:  EPA has failed to investigate adequately other
     potential source areas of Well Field and aquifer
     contamination.  EPA investigated only two other potential
     contaminant sources in the study area, although provided
     with extensive information regarding both other industries
     that have used large amounts of solvents, and solvent spills
     and discharges, and provided with specific suggestions for
     alternate source investigations.  The Agency should
     thoroughly evaluate those potential sources of contamination
     or, provide an explanation of its decision not to.

     EPA data supports the possibility that other sources may be
     responsible for contamination.  For example, the lack of TCE
     in bedrock monitoring well K-22B coupled with the elevated
     TCE concentrations in well K-21, located downgradient of
     well K-22b, suggests a contamination source other than the
     Complex.

     EPA Response:  EPA does not dispute that there may be other
     sources that contribute to contamination of the aquifer and
     the Well Field; however, the Complex is a significant source
     of contamination that requires remediation.  Sampling and
     analysis shows that a plume is emanating away from the
     Complex in a semi-radial pattern from the northwest to
     southwest.  Moreover, this testing indicates that
     contaminants move in both the overburden and bedrock
     aquifers to the Well Field.  Extensive sampling of over 120
     wells and borings at the site , as well as work done by the
     Connecticut Department of Environmental Protection have
     failed to identify other sources of contamination.

     Two other potential sources of contamination (Wesco and a
     Shell gas station)  were evaluated by EPA in the supplemental
     RI because they are identified in the 1986 RI as potential
     sources of contamination.  They were, however,  found not to
     be TCE or PCE contaminant sources after further analysis.

     EPA may consider whether to evaluate other potential sources
     of contamination.

     The lack of TCE at well K-22B does not necessarily indicate
     that a source other than the Complex is responsible for

                                12

-------
     contamination in down-gradient locations.  Because of the
     fracture orientation in the bedrock, some wells in up-
     gradient locations may be free of contamination while other
     nearby wells could be highly contaminated.


Risk Assessment

16.  Commentt  EPA's risk calculation for workers at the Complex
     is overly conservative because 1) it does not take into
     account worker movement, but rather assumes that workers
     stay directly above contaminated areas all day; 2) it
     assumes a 40-year working period, 3) it overestimates
     inhalation rates, and 4) it does not address likely changes
     in contaminant concentration with the seasons.  Further, the
     risk assessment does not consider the possibility that
     measurements were influenced by contaminated surfaces near
     former operational areas, rather than soil gas; nor does it
     take into account the possible use of TCE in certain of the
     buildings during sampling of building air.

     EVA Response:  EPA's risk calculation is meant to be
     conservative to ensure protection of human health.  It
     assumes that workers could be exposed to contamination eight
     hours per day over forty years.

     Although other sources of contamination could contribute to
     the air pollution problem found, given the concentration of
     TCE and PCE found in the Complex area, contaminant transport
     from the soil to the building air is significant.

17.  Comment:  Soil remediation is not necessary under the three
     buildings at the Complex, as they already meet Occupational
     Safety and Health Administration (OSHA) Permissible Exposure
     Limits (PELs) for PCE and TCE.  PELs are the correct
     enforceable standard, not EPA's target excess cancer risk
     range, which should be used only when a standard such as
     OSHA's does not exist.

     EPA Response:  Soil remediation is necessary for three
     reasons:  First, contamination in soil contributes to
     groundwater contamination that exceeds Maximum Contaminant
     Levels (MCLs), EPA's cleanup goal for groundwater.  Second,
     although the levels of contaminants measured were within the
     standards set for industrial safety by the Federal
     Occupational Safety and Health Administration, the risks
     from exposure to these levels (4x10 ) exceeds EPA's
     Superfund target risk range of 10  to 10  , using  a
     reasonable worst-case scenario.  Third, the Resource,
     Conservation and Recovery Act (RCRA) closure regulations,
     which are relevant and appropriate to the site, require that
     soil be cleaned so that contaminated leachate is not

                               13

-------
     generated that could result in unacceptable groundwater
     quality.
18.  Comment:  Should remediation to meet EPA standards be
     necessary, sealing the Complex building's floors and related
     measures would be a more reliable and cost effective
     remediation method.

     EPA Response:  EPA does not dispute that sealing floors and
     other similar actions could address the inhalation risks
     posed in the buildings; however, this would not address the
     contribution that the contaminated soil has to pollution in
     groundwater and this would not attain RCRA closure
     requirements/ an ARAR at the site.

19.  Comment:  Concerns about future human contact with
     contaminated soil through building demolition and soil
     disturbance can be met with institutional controls, such as
     have been successfully used elsewhere, and by means of an
     analysis of the need for soil remediation measures at the
     time a change in use of the Complex property dictates that
     such an analysis is needed.

     EPA Response:  CERCLA requires EPA to select permanent
     remedies that employ treatment to reduce the mobility,
     toxicity or volume of the contamination (if such a remedy is
     practicable), over remedies that rely on non-treatment based
     mechanisms such as institutional controls and containment.
     Additionally, EPA must assure that the remedy provides long-
     term protectiveness.  If soils are not remediated, the
     groundwater cleanup goal of achieving MCLs would not be met
     and, also,: given the potential failure of institutional
     controls, institutional controls alone would not provide for
     adequate protectiveness.

20.  Comment:  The risk assessment used to determine goals for
     soil remediation alternatives is seriously flawed.  The
     baseline risk assessment concentrates on the risk that
     contaminants in the soil serve as a continuing source for
     contamination of groundwater.  The risk assessment therefore
     estimates that amount of TCE and PCE that may remain in the
     soils without resulting in TCE and PCE groundwater
     contamination above 5 ppb.  The assessment assumes that all
     buildings and asphalt are removed from the property.

     First, the use of MCLs for TCE to determine groundwater
     cleanup levels is inappropriate, since the MCL is not itself
     an ARAR for groundwater at Kellogg-Deering, as detailed
     elsewhere. Similarly, the 5 ppb groundwater cleanup level
     set for PCE is inappropriate for setting soil cleanup
     levels,  as the PCE groundwater goal set by EPA is not an
     ARAR, nor does it appear to be based on any meaningful


                               .14

-------
assessment of PCE's threat to human health or the
environment.

Second, the assumption that soil contamination should not be
permitted to contaminate groundwater at levels above 5 ppb
of TCE and PCE is unreasonable in light of two other Agency
findings.  EPA has agreed that institutional controls will
prevent exposure to, and therefore risk from, contaminated
groundwater.  Moreover, even in the absence of such
controls, the Agency has admitted that it is extremely
unlikely that the aquifer can be cleaned up to referenced
levels.

Third, EPA1s assumption that the Complex has been stripped
is unreasonable, given current capping and EPA's ability to
use institutional controls to limit infiltration should the
use of Complex property change.

EPA Response:  The risk assessment that EPA utilized was not
flawed.  The assumptions used ensure that a protective
remedy will result.  Additionally, the risk assessment was
not the only tool that EPA utilized to set the soil cleanup
goals.  In addition to mitigating the risks posed by the
contamination in the aquifer, Section 121(d) of CERCLA
requires that the remedial action attain legally applicable
or relevant and appropriate requirements set by state and
federal laws.  At Kellogg-Deering, EPA has determined that
MCLs are relevant and appropriate throughout the aquifer.
Further, EPA has determined that restoration of the
groundwater is required to achieve them.

The cleanup levels were also set to ensure that leachate
generated by water percolating through the contaminated soil
would not result in unacceptable groundwater contamination
(i.e., MCLs), and to meet the RCRA closure requirements.

If the present Well Field treatment system were not in use,
the chemical contaminants in the aquifer could pose a risk
to human beings. . Human beings could be exposed to site
contamination by drinking the water, skin contact or
inhalation of contaminants while showering if there were no
treatment system at the Well Field or if drinking water
wells were installed in the area of groundwater
contamination.

As responded to earlier, the MCL for TCE is a relevant and
appropriate standard in groundwater and is the goal of the
remedy.  Although PCE does not have a MCL, 5 ppb is an
appropriate cleanup level.  This level was selected to be
protective and because PCE has chemical, physical and
toxicological properties that are similar to TCE.
                           15

-------
     Contrary to the statement in the comment, EPA has not
     admitted that the cleanup goals for TCE and PCE can not be
     met.  The remedy, utilizing soil and groundwater treatment,
     is an aggressive cleanup that can achieve these goals.
     Institutional controls would not attain ARARs and are
     therefore not an acceptable remedy for the site.  Also,
     'given the potential failure of institutional controls,.
     institutional controls alone would not provide for adequate
     protectiveness.


Contamination in Soil

21.  Comment:  The contaminants in the groundwater, soil,  and
     soil gas under the Complex are in equilibrium.  Hence, a
     portion of the contamination detected in the soil gas is a
     result of off-gassing from the contaminated groundwater and
     does not reflect contamination from a soil source. EPA does
     not determine the extent to which soil gas reflects primary
     contamination above the water table, rather than secondary
     contamination caused by off-gassing from contaminated
     groundwater. In other words, soil contamination at the
     Complex is likely, to a large extent, to be a lingering
     residual effect rather than a continuing cause of
     groundwater contamination.              .  .

     Because of the potential for this groundwater off-gassing
     effect, soil cleanup may not be effective until groundwater
     remediation at the Complex has been largely completed.
     Otherwise, the soils could be continually recontaminated.
     Although EPA recognizes this in the Supplemental RI/FS, it
     nevertheless uncritically concludes, erroneously, that the
     short-term effectiveness of in situ vacuum extraction at the
     Complex is high.

     EPA Response:  EPA concurs that the contamination found in
     the soil,  soil gas and groundwater is in equilibrium;
     however, the chemicals in the soil at the Complex are a
     continuing source of contamination to groundwater.  The
     concentrations of TCE and PCE measured in the soil gas study
     were greater than could be caused by off-gassing from the
     groundwater.  The release of very concentrated spent
     solvents from the ground surface into the soil accounts for
     the observed levels.

     The remediation of the aquifer at Kellogg-Deering requires
     both the vacuum extraction of soil and pump-and-treatment
     for groundwater to.be effective.   This two-component source
     control remedy will remove contamination from the highly
     contaminated soils at the Complex and in the groundwater
     flowing from that area,  and will result in the rapid
     remediation of the aquifer.   The operation of the remedy

                               16

-------
     will take into account the potential of re-contamination by
     off-gassing.  The pumping and treatment of groundvater and
     vacuum extraction of the contaminated soil will be sequenced
     to prevent contamination from moving from the groundvater
     back to clean soil.

22.  Comments  Under current site conditions, the amount of TCE
     contamination in the soil is relatively small and immobile
     in comparison to the amount of TCE contamination in the
     groundvater.  Moreover, only a small percentage of the
     contamination in the soil beneath the Complex buildings is
     likely to enter the groundvater.  A large portion of the
     area is covered by buildings and asphalt parking lots vhich
     prevent precipitation from filtering into the ground and
     entraining contaminants into the groundvater.  The only
     contaminated soil that vould release contaminants into the
     groundvater is that small volume of soil flushed by the
     groundvater through seasonal fluctuations in the vater
     table.

     Any soil contaminants released into the groundvater vill be
     intercepted and treated by the source area groundvater
     treatment system.  Thus, any contaminants in the soil vill
     not migrate off-site and vill not adversely affect human
     health or the environment.  Therefore, collecting and
     treating groundvater at the Complex vill intercept any soil
     contamination entrained into the groundvater and vill
     effectively reduce the majority of the mobile contamination
     at the Complex.  Eventually, the flushing of the soil
     contaminants by the fluctuating vater table should
     substantially cleanse the soil and minimize any remaining
     soil source.  Although failure to treat the soil nov may
     result in the need for a longer groundvater pump and treat
     period at the source area, the additional operation and
     maintenance operations are less costly, and equally
     effective, as the soil treatment proposed by EPA.

     EPA Response:  The concentrations and amount of TCE found in
     the soil is significant and acts as a continuing source of
     groundvater contamination.  To achieve MCLs in the aquifer,
     a combination of contaminated soil remediation and
     groundvater treatment vill be needed.  Although the source
     area groundvater treatment system vill intercept some of the
     contamination released from the soil, vithout the soil
     remediation component of the remedy, the soil vill act as a
     continuing source of contamination and vill result in
     unacceptable aquifer restoration times.  Additionally,  the
     RCRA closure requirements for tie soil vill not be met by
     groundvater treatment alone.

     Although the paved parking lots at the Complex may reduce
     some of the rainwater that infiltrates into the soil,

                               17

-------
     relying on this as part of the remedy is unacceptable.  Some
     significant amount of water still washes through the soil
     and there is no long-term way to ensure that the parking
     lots will remain  in-place and that institutional controls
     will be complied  with.

23.  Comment:  Because of questions regarding the integrity of
     analytical data for 17 soil samples, analyzed in a mobile
     laboratory, EPA should reevaluate the conclusions in the
     RI/FS based on those 17 samples.

     EPA Response:  EPA has re-evaluated how the samples were
     used in the RI/FS and determined that they do not affect the
     conclusions that  the Agency reached regarding the nature and
     extent of contamination or have any bearing on the
     evaluation and selection of the remedy.  Besides these
     samples, EPA relied on validated, accurate data from
     hundreds of samples collected at the site since work began
     in 1984.
Applicable or Relevant and Appropriate Requirements fARARs)

24.  Comment:  EPA has misidentified and misapplied ARARs for
     soils at the Complex.  It is unclear from the preferred
     remedy exactly what cleanup criteria EPA seeks to impose at
     the Complex, as the Supplemental RI/FS sets forth different
     numbers in different places.  For example, soil cleanup
     goals for TCE are stated to be between 12 and 358 ppb in the
     text, but are identified variously as 6.5 ppb and between
     6.9 and 151 ppb in the design calculation appendix.  It is
     clear, however, that whatever numbers EPA is using, it
     considers them to be driven by RCRA closure standards, which
     the Agency identifies as an action-specific ARAR for soils,
     although "applicable" or ''relevant and appropriate" is
     unclear.

     RCRA closure standards are clearly not applicable if,  as in
     this case, a RCRA hazardous waste is not newly placed at a
     site after the effective date of the closure standards.
     Moreover, the RCRA closure standards do not require
     treatment or removal of wastes to comply with closure
     requirements.  Instead, waste can be left in place, provided
     that post-closure groundwater monitoring of the type
     proposed by EPA for this site occurs.

     EPA Response:   The soil cleanup levels were developed using
     a soil/water sorption coefficient which describes the
     distribution of chemicals between soil and groundwater.
     They were set in order to reduce the potential future risk
     to public health and the environment and to protect ground-
     water by reducing contamination levels to the point where

                               18

-------
     they no longer contribute to groundvater contamination above
     MCLs.  The ranges are given because the soil types differ
     within the area of soil contamination.  The particular type
     of soil and amount of organic carbon present in the soils
     attenuate contaminant migration and result in the different
     cleanup goals provided.

     The closure standard cited above refers to landfill closure
     and not to the closure standards which EPA utilized.  EPA
     has determined that portions of the RCRA regulations on
     closure that require decontamination of soils are relevant
     and appropriate for the site (40 CFR 264.110 - 264.120).
     EPA determined that the release of hazardous substances at
     the site was similar to releases regulated by RCRA and
     because of this, closure performance standards under RCRA
     are both relevant and appropriate for the cleanup.

25.  Commentt  EPA has misidentified and misapplied applicable or
     relevant and appropriate requirements for source area
     groundwater remediation.  The Agency wrongly relies on
     inappropriate cleanup standards, given that there is
     virtually no risk-based justification for source area
     remediation.                    -

     SDWA's MCLs are inappropriate as ARARs given CERCLA's
     Section 121(d) requirement that cleanup criteria selection
     must consider the "circumstances of the release,11 the
     characteristics of the remedial action, the physical
     circumstances of the site, the type of place regulated, and
     the objective and origin of the requirement. Judged by these
     criteria, MCLs are not relevant and appropriate requirements
     for restoration of the study area aquifer.

     EPA Response:  It is within EPA's authority to determine
     ARARs at the site and to determine cleanup levels necessary
     to attain those ARARs.  With respect to achieving drinking
     water standards in the aquifer, EPA states in 40 CFR Part
     300, National Oil and Hazardous Substances Pollution
     Contingency Plan; Proposed Rule (Fed. Reg. 53:51394-51520,
     December 21, 1988) that "it has been the policy of EPA's
     Superfund program for several years to operate within the
     framework of EPA's Groundwater Protection Strategy in
     determining the appropriate remediation for contaminated
     groundwater at CERCLA sites".  The water supply wells are
     located in an aquifer that is classified as XI-A under the
     EPA Ground Water Protection Strategy.  This classification
     means that the groundwater is currently and potentially a
     source of drinking water and other beneficial uses.  EPA
     states further that, "for groundwater that is or may be used
     for drinking water (Class I or II), the maximum contaminant
     levels (MCLs) set under the Safe Drinking Water Act or more
     stringent promulgated State standards are generally

                                19

-------
applicable or relevant and appropriate standards" (ARARs).
The aquifer at Kellogg-Deering is a valuable resource both
under EPA's Guidelines for Ground-Water Classification and
under the State classification for groundwater, which
classifies the aquifer as Class GAA at the Well Field — the
State's most stringent classification for groundwater
protection — and GB/GA in other portions of the aquifer at
the site, which indicates that although the groundwater
presently may not be suitable for direct human consumption
without treatment due to chemical contaminants or land use
impacts, the State maintains a goal to restore the
groundvater to drinking water quality.  Given these factors,
EPA has determined that MCLs are relevant and appropriate to
aquifer remediation at the site.

There are six limited circumstances where ARARs may be
waived Under CERCLA § 121(d)(4).  These are:

     i)   the remedial action selected is only part of a
     total remedial action that will attain such level or
     standard of control when completed;

     ii)  compliance with such requirement at that facility
     will result in greater risk to human health and the
     environment than alternative options;

     iii)  compliance with such requirements is technically
     impracticable from an engineering perspective;

     iv)  the remedial action selected will attain a
     standard of performance that is equivalent to that
     required under the otherwise applicable standard,
     requirement, criteria,  or limitation, through use of
     another method or approach;

     v)   with respect to a State standard, requirement,
     criteria, or limitation,  the State has not consistently
     applied (or demonstrated the intention to consistently
     apply) the standard,  requirement, criteria,  or
     limitation in similar circumstances at other remedial
     actions within the State; or

     vi)  in the case of a remedial action to be undertaken
     solely under section 104  using the Fund, selection of a
     remedial action that attains such level or standard of
     control will not provide a balance between the need for
     protection of public health and welfare and the
     environment at the facility under consideration,  and
     the availability of amounts from the Fund to respond to
     other sites which present or may present a threat to
     public health or welfare or the environment,  taking
                          20

-------
          into consideration the relative immediacy of such
          threats.

     None of these six conditions apply at the Kellogg-Deering
     Well Field Superfund site.  Thus, EPA is justified in
     setting action levels that will result in the attainment of
     MCLs in the aquifer throughout the site.

26.  Comment*  If EPA does treat MCLs as relevant and appropriate
     requirements, it should waive the requirement that they be
     met pursuant to CERCLA Section 121(d)(4), as complete
     restoration to health-based levels is not feasible here
     where EPA has determined that contaminants have migrated
     into fractured bedrock with low transmissivity.  EPA
     recently waived groundwater ARARs in the Sullivan's Ledge
     Record of Decision (ROD) because there, as here,
     contamination was located in difficult-to-locate bedrock
     fractures.  An EPA decision not to waive groundwater ARARs
     in this instance would be arbitrary and capricious, given
     the dictates of its own guidance and the factual similarity
     with bedrock contamination at Sullivan's Ledge.

     EPA Response:  EPA is aware of the difficulty in predicting   \
     the performance of the groundwater remedies, especially for
     sites that have complicated hydrogeology and contamination
     transport mechanisms.  However, EPA will not waive MCLs in
     this ROD because the statutory criteria for a waiver of
     ARARs, as set forth in EPA's response to the preceding
     question, has not been met.

     Although EPA waived some requirements for groundwater
     cleanup at Sullivan's Ledge, the aquifer at that site is not
     a drinking water supply source like Kellogg-Deering and a
     number of different site conditions exist.

27.  Comment!  EPA misapplies the RCRA groundwater protection
     standards as relevant and appropriate for the Kellogg-
     Deering site.  The RCRA standards require compliance at the
     edge of a discrete waste management area for containment and
     early detection reasons that do not apply to a Superfund
     site where contamination has already spread beyond a source
     boundary.

     EPA Responses  EPA did not misapply the RCRA groundwater
     protection standards.  Although the remedy was developed to
     meet the closure requirements of 40 CFR 264.111 and attain
     MCLs and state ARARs in the aquifer, EPA also determined
     that the RCRA groundwater protection standards are relevant
     and appropriate.

     Under RCRA regulations, the groundwater protection standard
     establishes a safe level of contamination in groundwater in

                                21

-------
     the vicinity of a waste disposal site.  Under these
     regulations, the protection standard can be set at MCLs,
     ACLs or at background.  EPA has determined that the release
     of hazardous substances at the site was similar to releases
     regulated by RCRA.  Therefore, EPA has determined that MCLs
     are relevant and appropriate at the site.

     In addition to the groundwater protection standard, Subpart
     F provides relevant and appropriate groundwater monitoring
     requirements and procedures that EPA will utilize in the
     remedy (40 CFR 264.97 through 264.99).

28.  Comment:  EPA has also aisidentified the Connecticut Water
     Quality Standards for groundwater as an "applicable"
     requirement at Kellogg-Deering.  "Applicable" standards or
     requirements must be enforceable, and a review of the WQS
     statute reveals that the standards are neither requirements
     nor legally enforceable.  Moreover, the WQS require that
     groundwater be restored to drinking water quality without
     treatment only "whenever feasible" and "to the extent
     possible."  Given the admitted technical impracticability of
     remediating Kellogg-Deering groundwater to MCL standards,
     WQS cannot be considered an ARAR from groundwater
     remediation.

     EPA Response:  Section 121(d)(2)(A) of CERCLA, as amended,
     states that remedies must comply with "any promulgated
     standard, requirement, criteria, or limitation under a State
     environmental or facility siting law that is more stringent
     than any Federal standard, requirement, or limitation" if
     applicable or relevant and appropriate to the hazardous  .
     substance or release in question.  EPA proposes in 40 CFR
     Part 300, National Oil and Hazardous Substances Pollution
     Contingency Plan; Proposed Rule (Fed. Reg. 53:51394-51520,
     December 21, 1988) "to define promulgated State requirements
     as those laws or regulations that are of general
     applicability and are legally enforceable."  "The phrase
     'legally enforceable* refers to State regulations that are
     statutes which contain specific enforcement provisions or
     are otherwise enforceable under State law.  EPA expects that
     State laws or standards which are considered potential ARARs
     have been issued in accordance with State procedural
     requirements.  The phrase 'of general applicability* is
     meant to preclude consideration of State requirements
     promulgated specifically for one or more CERCLA sites as
     potential ARARs . . .  For a State requirement to be a
     potential ARAR it must be applicable to all remedial
     situations described in the requirement, not just CERCLA
     sites.   General State goals that are contained in a
     promulgated statute and implemented via specific
     requirements found in the statute or in other promulgated
     regulations are potential ARARs. "

                               22

-------
     EPA has determined that the Connecticut Water Quality
     Standards are applicable to aquifer restoration at the
     Source Area to the extent that they provide a level or
     standard of control that is promulgated and more stringent
     than Federal requirements.  The aquifer at Kellogg-Deering
     is a valuable resource both under EPA's Guidelines for
     Ground-Water Classification and under the State
     classification for groundwater, which classifies the aquifer
     as Class GAA at the Well Field, the State's most stringent
     classification for groundvater protection, and GB/GA in
     other portions of the aquifer at the site.  The GB/BA
     classification indicates that although the groundvater
     presently may not be suitable for direct human consumption
     without treatment due to chemical contaminants or land use
     impacts, the State maintains a goal to restore the
     groundwater to drinking water quality.  The Connecticut
     Water Quality Standards were adopted pursuant to Connecticut
     General Statutes 22a-426 and are enforced by the State
     through other provisions in Title 22.  The Water Quality
     Standards are of general applicability and apply to all
     remedial situations described in the requirement, not just
     CERCLA sites.  For the reasons stated above, the Connecticut
     Water Quality Criteria are applicable to the cleanup of the
     aquifer to the extent that they provide a level or standard
     of control that is promulgated and more stringent than
     Federal requirements.

29.  Comment:  If the Connecticut Water Quality Standards are
     considered ARARs, they should be waived, for the same
     reasons that MCLs should be waived as detailed above.
     Furthermore, CERCLA authorizes the waiver of state standards
     that have not been consistently applied, or demonstrated an
     intention to consistently apply, the standard in similar
     circumstances at other remedial actions.  Connecticut has
     not applied the WQS consistently at similar sites, nor has
     it demonstrated any intention of applying them consistently
     in the future.

     ZPA Response:  It is within EPA's authority to determine the
     ARARs for the site and to determine when a waiver is
     appropriate.  As stated in EPA's response to comment number
     25 above, there are six potential circumstances in which a
     waiver may apply.  EPA has determined that none of these six
     conditions apply at the Kellogg-Deering Well Field Superfund
     Site.
Remedy Components

30.  Comment:  Although EPA has determined that vapor phase
     carbon adsorption will be used to treat emissions from the

                                23

-------
     in situ vacuum extraction system, it does not justify this
     requirement.  If one were to calculate the maximum allowable
     stack concentration for TCE under Connecticut Air Pollution
     Control regulations using EPA's design parameters for the
     extraction system's air discharge rate, the maximum
     allowable stack concentration would be no less than 88 ppm
     by volume (ppmv).

     EPA Response:  Carbon adsorption will be used for the vacuum
     extraction system to ensure that the hazardous substances
     are treated to the maximum practicable and to ensure
     protectiveness consistent with OSWER Directive 9355.028 and
     Region I guidance.

31.  Comment:  Although EPA's preferred remedy ostensibly leaves
     open the possibility that source area groundwater may be
     discharged to the Norwalk River or the local publicly-owned
     treatment works (POTW), or injected into the aquifer, EPA
     analyzes alternative groundwater remedial components using
     only the Norwalk River option, which requires piping the
     treated water to the Norwalk River south of the Well Field.
     EPA's rationale that discharges upstream of the Well Field
     would have to meet MCLs is not supported by any analysis.

     As the Agency notes, discharge from both source and
     downgradient air strippers would meet appropriate water
     quality criteria in the Norwalk River, and projected in-
     stream TCE concentrations comply with the standards of the
     Connecticut Discharge Permit Regulations and risk-based
     calculations for consumption of aquatic organisms from the
     Norwalk River.

     EPA Response:  The method for disposal of treated ground-
     water will be determined during Remedial Design.  The method
     of disposal specified in the FS was used for costing and as
     a basis for the comparison of the other alternatives.

     Options that will be considered in the design individually
     or in combination include 1) reinjection back into the
     aquifer, 2)  discharge into the Norwalk River through a new
     pipe, 3) discharge to the Norwalk River through existing
     storm sewers, or 4) discharge to the local Publicly Owned
     Treatment Works (POTW).  The treatment standards will comply
     with the applicable or relevant and appropriate requirements
     for the disposal or discharge method chosen.

32.  Comment:  EPA has prematurely and erroneously determined
     that air emission control is necessary for groundwater
     source area air stripping.  It is impossible to determine,
     without further design, whether a source area air stripper
     would violate Connecticut Air Pollution Control requirements
     without air emission treatment.  It is not possible to

                               24

-------
     accurately determine whether an air stripper will neat
     maximum allowable stack concentrations before the air
     stripper actually goes on line or is pilot tested, since
     input variables to the maximum allowable stack concentration
     calculation  (e.g., air emission rate) will not be known
     until that time.

     The RI/FS acknowledges that the calculated air stripper
     emissions fall below the level requiring carbon off-gas
     filters, but then proceeds to call for them anyway. There is
     no ARAR or risk-based justification for the use of carbon
     off-gas filters at either the Complex or the Well Field.

     EPA Responses  Air emission control by carbon adsorption
     will be used for the vacuum extraction system to ensure that
     the hazardous substances are treated to the maximum
     practicable and to ensure protectiveness consistent with
     OSWER Directive 9355.028 and Region I guidance.

33.  Comment:  Areas chosen in the RI/FS for extraction wells do
     not exhibit contamination action levels chosen elsewhere in
     the RI/FS.

     EPA Response:  The exact number and location of the
     extraction wells and their respective pumping rates will be
     determined in the remedial design to attain the cleanup
     goals of the remedy.  The proposed configuration in the
     RI/FS was appropriate for the development and relative
     comparison of groundwater treatment alternatives.

34.  Comment:  The RI/FS states "no reliable estimates can be
     made for the overall remediation time required." (p. 358)
     The only guess exceeds thirty years, but EPA does not really
     know.  This willingness to begin pumping and treating
     groundwater is unjustified in light of the questionable
     effectiveness of the system and EPA's inability to determine
     how soon the aquifer would clean itself.  It is possible
     that natural flushing would be faster without pumping and
     treating.

     EPA Response:  Pumping and treating contaminated groundwater
     will be effective in reducing the size of the plume and the
     mass of contaminants in the aquifer.  Moreover, this will
     enhance natural flushing of the aquifer.  EPA believes that
     approximately 2.5 years will be needed to remediate the soil
     at the Complex and, although the time needed to remediate
     groundwater can not be precisely set, pumping and treating
     groundwater may have to be operated for ten or more years.

35.  Comment:  The RI/FS does not discuss the potential impact of
     extensive pumping of the overburden and bedrock aquifers in
     the source area, and subsequent reinjection. Such pumping

                               25

-------
     and reinjection in a concentrated area has been shown to
     cause structural failure of foundations and other rigid
     structures as a result of the subsidence and mass movement
     of sediment, extensive water damage of structures and even
     micro-seismic events.

     EPA Response:  Subsidence is not expected to be a problem.
     Also, the design of the remedy will take these factors into
     account and will be developed to avoid de-watering that
     could result in structural damage and subsidence.


Institutional Controls

36.  Comment:  The Phase II RI notes risk due to the possibility
     that future private wells may be drilled near the source of
     contamination, exposing humans to the contamination.
     However, Connecticut state law requires all properties
     within 200 feet of a public water supply system to hook up
     to that system, and does not allow the installation of
     private wells under such circumstances. These institutional
     restrictions are present here and are unlikely to change
     given the areas's recognized contamination and
     impracticability of restoring the aquifer to MCLs.

     EPA Response:  EPA chose the cleanup because CERCLA requires
     the selection of permanent remedies that employ treatment to
     reduce the mobility, toxicity or volume of the contamination
     (if such a remedy is practicable),  over remedies like the
     one outlined above that rely on non-treatment based
     mechanisms such as institutional controls and containment.

     Additionally, institutional controls would not attain ARARs
     (e.g., MCLs and RCRA closure requirements)  and is,
     therefore, not an acceptable remedy for the site.  Also,
     given the potential failure of institutional controls,
     without adequate engineering controls institutional controls
     alone would not provide for adequate protectiveness.


State Ordered Groundwater Treatment System

37.  Comment:  Even assuming a hydrogeologic connection to the
     Well Field, source area remediation is not necessary because
     the RI/FS has determined that the Well Field air stripper
     can meet Safe Drinking Water Act (SDWA) MCLs even under
     "reasonable worst case scenarios."  The State-ordered
     treatment system at the Complex provides a further safety
     margin.

     EPA Response:  As discussed in a number of earlier
     responses, the selected remedy provides that MCLs must be

                               26

-------
     attained throughout the aquifer, not only at the Well Field.
     The State-ordered groundwater treatment system, even in
     combination with the veil head treatment, will not attain
     ARARs or EPA's remedial objectives at the site.

38.  Comments  The RZ/FS does not explain why the groundwater
     treatment and extraction system installed under the DEP
     order is not sufficient for source area remediation.

     EPA Responses  The RZ/FS prepared by EPA developed and
     evaluated a range of cleanup alternatives to restore the
     aquifer at the Kellogg-Deering site.  The system installed
     pursuant to the State Order addresses only a limited portion
     of the highly contaminated portion of the aquifer and does
     not provide for soil treatment.  Therefore, it is unlikely
     that the system can meet EPA's cleanup goals ~ it would not
     intercept enough of the contaminated groundwater to achieve
     a cleanup to MCLs.  However, as part of the remedy's design,
     EPA will consider use of the system installed under the DEP
     order and, if possible, will incorporate it into EPA's
     cleanup plans.


Alternative Remedies

39.  Comments  A number of alternative remedies were proposed
     that the FRPs contended were more cost-effective than the
     selected remedy.  The first would modify the existing Well
     Field air stripper, or add another stripper, to ensure that
     water extracted from the Well Field continues to meet risk-
     based standards in the event that contaminant levels
     increase at the wellhead.

     EPA Response:  For any alternative to be more cost-
     effective, it would have to achieve the same results as the
     selected remedy but at lower cost.  None of the proposed
     alternatives do this.  Further, EPA determined that the
     selected remedy will yield results that are in proportion to
     its cost in terms of the implementability and
     protectiveness.

     The premise of this first alternative is that MCLs do not
     have to be attained throughout the aquifer but only at the
     Well Field.  Again, as discussed earlier, at Kellogg-Deering
     EPA has determined that MCLs are relevant and appropriate
     throughout the aquifer and that restoration of the
     groundwater is required to achieve them.

40.  Comments  A second alternative remedy would have the
     following components:
                                27

-------
 - The groundwater pump and treat air stripping system
approved by DEP at the Complex, which will clean up the
highest contaminant levels observed in the study area and
prevent the unlikely possibility that soil contaminants
migrate into groundwater and then move offsite;

- Discharge of treated groundwater from the Complex air
stripper into the storm sewer, which flows into the Norwalk
River;
    •   • '  '.
- Continued operation of the existing Well Field air
stripper until contaminant levels entering the Well Field
distribution system drop below safe drinking water levels
without treatment by the stripper;

- Removal of any contaminated flooring and sealing of any
flooring cracks within the southernmost building at the
Complex as. necessary to minimize exposure to any possible
air emissions from contaminated soil gas beneath the
building;

- Institutional controls 1) at the Complex to prevent direct
exposure to soil contaminants, and 2)  throughout the study
area to/prevent use of unsafe contaminated groundwater; and

- Necessary and cost-effective monitoring of concentrations
of VOCS in 1) groundwater to determine the effectiveness of
the cleanup effort and 2) air inside the southernmost
building at the Complex to ensure that acceptable
concentration levels exist.

EPA Response:  This alternative is based on the premise that
either the determination that MCLs are not ARARs throughout
the aquifer, or if they are ARARs, then they should be
waived.

Again, as discussed earlier, EPA has determined that MCLs
are relevant and appropriate throughout the aquifer.
Further,  EPA has determined that restoration of the
groundwater is required to achieve them and to provide a
protective remedy.

In light of the criteria that EPA uses to evaluate remedial
alternatives, this proposed remedy is not acceptable because
it would not satisfy the statutory and regulatory criteria
for CERCLA cleanups,  including, but not limited to,
providing acceptable overall protectiveness and long-term
effectiveness,  or attaining ARARs.

Although EPA is aware of the difficulty in predicting the
performance of the groundwater remedies; EPA will not waive
MCLs because the Kellogg-Deering aquifer is a significant

                           28

-------
     resource utilized as a public drinking water source and
     because the statutory criteria for a waiver have not been
     vet.

41.  Consent*  The third alternative remedy proposes the
     following remedial actions:

     - Air quality

          1. Fill the pit and trenches in the Elinco building;
          2. Remove the TCE contaminated wood and sandblast or
               encapsulate contaminated floor areas, or remove
               wooden flooring as necessary;
          3. Monitor air quality to ensure compliance with
               applicable OSHA standards;
          4. Modify building ventilation if standards are
               violated.

     - Soil

          1. Impose land use controls on the Complex to prevent
               site cover stripping and abandoning as outlined in
               the RI/FS;
          2. Address soil contamination at the time of future
               site development.

     -  Source area groundwater

          1. Place into operation the existing groundwater
               treatment and extraction system, with a goal of
               contamination levels of 5 ppb or less in water
               flowing into the Well Field;
          2. Monitor system performance as an aid to information
               gathering.

     • Downgradient groundwater

          Given the problems of intercepting specific bedrock
          fractures, and the RI/FS's assumption of institutional
          controls to prevent installation of wells on the east
          side of the river, any groundwater treatment program
          should treat water in the overburden aquifer, rather
          than in the bedrock.

     EPA Responset  This alternative is not acceptable.  It
     relies on the State-ordered groundwater treatment system
     which *ill not meet ARARs or EPA's remedial objectives at
     the site.  Land-use institutional controls would not attain
     ARARs or prevent further groundwater contamination.

     Again, as discussed earlier, EPA has determined that MCLs
     are relevant and appropriate throughout the aquifer.

                                29

-------
     Further, EPA has determined that restoration of the
     groundwater is required to achieve them and to provide a
     protective remedy.

     In light of the criteria that EPA uses to evaluate remedial
     alternatives, this proposed remedy is not acceptable because
     it would not satisfy the statutory and regulatory criteria
     for CERCLA cleanups, including, but not limited to,
     providing acceptable overall protectiveness and long-term
     effectiveness, or attaining ARARs.


42.  Comment:  The following case was made for the No-Action
     alternative:

          The RI/FS erroneously focused exclusively on
          remedial activities only where the
          contamination is actually located. The RI/FS
          implicitly assumes that in assessing whether
          a remedy accomplishes a reduction in mobility
          or toxicity of a substance the focus must and
          can only be on accomplishing these goals at
          the contamination location.

          However, nothing in CERCLA or the National
          Contingency Plan dictates this focus. In the
          present case, the Well Field air stripper
          accomplishes all goals of permanently and
          significantly decreasing the toxicity,
          mobility, and volume of TCE. To be sure, the
          Well Field treatment system would not stop
          the spread of contaminants towards the Well
          Field. But by using a wider geographic area,
          an area that extends beyond where the
          contamination is found, the no action
          alternative would meet all remedial action
          requirements without any further remedial
          work.  The no action alternative would
          decrease the mobility, toxicity, and volume
          of TCE moving past the Well Field, and
          satisfy both CERCLA and the NCP.

     EPA Response:  The No Action alternative was not selected by
     EPA because it did not meet the statutory and regulatory
     criteria for selection of a remedy.  First, it would not be
     protective of human health and the environment.  Second,
     there would be no reduction of the mobility, toxicity and
     volume of contaminants in the unsaturated soils of the
     complex.  The mobility of the contaminants in groundwater
     would not be reduced.  The toxicity and volume of
     contaminated groundwater would not be reduced except by well
     head treatment,  and finally, this alternative would not

                                30

-------
     provide long-term effectiveness or permanence or attain
     groundwater or RCRA closure ARARs.


Remedy Selection Process

43.  Comments  EPA has not shown how or why its proposed
     expenditure of 13 million dollars will provide any more
     protection to human health or the environment than that
     currently being provided by the Norwalk First Taxing
     District's (NTFD) treatment system.

     The RI/FS Section 6 confirms that treated water leaving the
     Kellogg-Deering Well Field does not pose any threat to human
     health or the environment.  Further, existing site
     conditions as a whole pose a cancer risk of 1 in 360,000,000
     to an individual with lifetime exposure to the site.  For
     non-carcinogenic risks, the hazard index is well below 1.0
     and non-carcinogenic effects are thus also unlikely.  In
     short, there is no health risk posed by the site as it
     currently exists.

     The only risk-based justification EPA offers for remedial
     work relies on future (worst-case) and future (plausible
     case) scenarios. However, the assumptions underlying both
     scenarios are so untenable that reliance on them would
     constitute a capricious and insupportable basis for further
     action.

     EPA Responset  EPA agrees that the treatment system
     currently operated at the Kellogg-Deering Well Field
     protects public health.  If, however, private wells were
     installed in the area of groundwater contamination or the
     treatment system were not in use, due to equipment failure,
     or otherwise, the chemical contaminants in the aquifer could
     pose a risk to human beings.  Human beings could be exposed
     to site contamination by drinking the water, skin contact or
     inhalation of contaminants could occur while showering if
     there were no treatment system at the Well Field or if
     drinking water wells were installed in the area of
     groundwater contamination.

     In addition to mitigating the risks posed by the
     contamination in the aquifer, Section 121(d) of CERCLA
     requires that the remedial action attain legally applicable
     or relevant and appropriate requirements set by state and
     federal laws.  At Kellogg-Deering, EPA has determined that
     standards,  called Maximum Contaminant Levels (MCLs), set
     under the federal Safe Drinking Water Act are relevant and
     appropriate throughout the aquifer.  Further, EPA has
     determined that restoration of the groundwater is required
     to achieve them.  In addition to the federal standards that

                               31

-------
     the remediation must attain, the cleanup of the aquifer must
     meet State ARARs.

44.  Comment:  EPA has not met the CERCLA Section 121 requirement
     for a cost-effective remedy.  EPA's costs are based on a
     thirty-year remediation estimate that has no rational basis
     in light of its admission that "no reliable estimates can be
     made for the overall remediation time required." (RI/FS p.
     358.) Without a remediation time frame, one cannot
     rationally estimate remedial cost, and thus one cannot
     determine whether the proposed alternative is cost effective
     within the meaning of CERCLA Section 121.

     Further, the uncertain cost must be compared to existing
     site conditions, under which the air stripper at the Well
     Field is removing any threat to human health. This
     comparison demonstrates that EPA has not justified the cost
     effectiveness of the preferred alternative.

     EPA Response:  EPA has determined that the selected remedy
     will yield results that are proportionate to its costs in
     terms of implementabliity and effectiveness.

     Thirty years is not the estimated remediation time for the
     remedy.  Instead, thirty years is a basis for a present-
     worth analysis that EPA used for cost estimating.  Present-
     worth, also called present-value, is a cost accounting
     technique that sums the cost of a remedy that may take a
     number of years to complete as a total in today's dollars.
     This gives a common basis (a present-worth) to compare
     alternatives that have different project lives.

     Thirty years is used in the estimate because costs incurred
     after thirty years are relatively insignificant in terms of
     their present-worth and do not affect the comparison of
     different alternatives.

     As an example of the insignificance of expenses incurred
     after thirty years in the present-worth analysis, if the
     cost to operate a remedy is $100,000 per year,  the thirty-
     year present-worth of this is equal to $9,427,000.   If,
     however, the remedy is operated for another year, only about
     $5,700 (0.06%) is added to the present-worth total.

45.  Comment:  The energy cost of the proposed alternative is
     high, especially given the uncertainty of its effectiveness
     and the limited additional protection it affords. A total
     energy usage for the preferred alternative has not been
     included in the RI/FS. No specific energy cost appears
     associated with the in-situ soil treatment component of the
     proposed remedy although the system surely uses energy. For
     groundwater treatment, the RI/FS assumes a total energy

                               32

-------
     usage of 365,810 kilowatt-hours per year. This translates
     into about 810 barrels of oil per year, assuming 450
     kHh/barrel, or more than 24,000 barrels of oil over 30
     years.

     This energy use raises serious policy questions about
     whether it is worth depleting one natural resource to clean
     another. In the present case, this decision is made somewhat
     easier by the operating treatment system and its removal of
     any threats to human health. Without such a threat, use of
     excessive amounts of natural resources on a highly uncertain
     cleansing system cannot be justified.

     EVA Responses  Although energy costs are not itemized for
     the soil treatment, these costs are included as part of the
     estimated operating costs that are provided.

     No serious policy questions arose over implementing the
     Xellogg-Deering reaedy at the expense of the world's oil
     reserves.
B.   SUMMARY OF COMMENTS FROM THE STATE.  CITIZENS AND OTHER
     INTERESTED PARTIES
     Connecticut Department of Environmental Protection

1.   Comment:  The cleanup goal concentrations cover broad
     ranges, up to two orders of magnitude, and the proposed plan
     does not specify a procedure for selecting remedial goals
     from the cleanup goals.

     EPA Response:  The soil cleanup levels are ranges because
     the soil types differ within the area of soil contamination.
     The particular type of soil and amount of organic carbon
     present in the soils attenuate contaminant migration and
     result in the different cleanup goals provided.  The
     specific goals were set with a soil/water sorption model.
     Details of this are provided in the Remedial Investigation
     and Feasibility Study reports.  The cleanup goals derived
     from the model will insure that MCLs are not exceeded in the
     groundwater as a result of soil leaching.

2.   Comment:  The Connecticut Department of Health Services has
     established an action level of 1,000 micrograms per liter
     for toluene in drinking water supplies.  To properly reflect
     the state's more stringent standards, the site cleanup
     standard should be 1,000 and not 2,000 micrograms per liter.
                                33

-------
     EPA Response: The Connecticut Department of Health Services
     action level is not an ARAR because it has not been
     promulgated.


     Local Citizens and Other Parties

1.   Comments  The Proposed Plan states that wells containing TCE
     were shut down in 1975, and restarted when the NFTD Water
     Department installed treatment systems in 1981.  However,
     numerous newspaper articles from 1984 and 1985 report
     citizen demands that contaminated wells be shut down and
     treatment initiated, and some citizens believe that aeration
     did not begin until 1987.

     EPA Response:  Treatment at the Well Field was started in
     May of 1981 by the Norwalk First Taxing District (NFTD).  In
     1988, further pollution controls were installed at the Well
     Field.
2.   Comment:  The NFTD did not notify its customers when the
     contamination was first found.  What will be done to insure
     that future contamination is not kept secret for years?

     EPA Response:  The Norwalk First Taxing District (NFTD)  is
     required to report contamination that they find to the
     Connecticut Department of Health Services.

3.   Comment:  The November 1980 letter to NFTD Water Company
     customers listed as contaminants 1-4 dioxane,
     methylethylketone or ethylene glycol, yet the RI/FS did not
     mention these chemicals.

     EPA Response:  EPA did not find these chemicals in its
     testing.  The contaminants EPA did find that cause the
     problems at the site are trichloroethene, commonly called
     TCE, and perchloroethene, or PCE.

4.   Comment:  The construction of the new Route 7 so close to
     the Well Field and above the aquifer worries some residents.
     Particular concerns include the alleged use of junked cars
     as fill, Connecticut DOT's oil storage at the construction
     site, current illegal dumping near the Well Field,  chemical
     spills from trucks that will use the highway, and salt use
     in the Winter.

     EPA Response:  Although these are all valid concerns and
     should be addressed, this ROD and Responsiveness Summary
     address the selection of a remedy fro the Source Area.
                                34

-------
5.   Comment:  According to several residents, the RI/FS does not
     reflect the full history of the area.  The following
     information was provided:

          1. The area bounded to the north by Broad Street, to
          the east by the Norwalk River, to the south by East
          Lakeview Drive and Nutmeg Place, and to the west by
          Lakeview Avenue was a large pond until 1965, when it
          was filled in.

          2. The area now known as Kellogg-Deering Pond has long
          been used as a dumping ground by residents and local
          businesses. The configuration of the pond has changed
          many times, and a 1970 map shows the pond covering the
          current location of the Nell Field.

          3. The area between Wesco Ford and the Norwalk River
          was once a wetland lying just above the river, and was
          used until 1981 as a dumping ground for a variety of
          objects including cars, appliances, and drums.
          Numerous other areas along the river have been filled,
          sometimes with a variety of waste.  Possibly
          contaminants are leaching from these areas, or could
          leach in the future.  In addition, manufacturing plants
          north of the Well Field have long regarded the river as
          a waste receptacle, and local residents have long
          spoken about the tainted river.  In short, several
          possible contamination sites exist in the area and
          north of the Well Field. Thus, there is concern that
          the site be constantly monitored even after the
          cleanup.

     EPA Response:  The remedy that EPA has selected will address
     the significant source of contamination found at the
     Complex.  EPA recognizes the uncertainties in treating
     groundwater in bedrock and, although EPA believes that the
     remedy will achieve ARARs and will be protective of human
     health and the environment, the remedy also relies on the
     existing Well Field treatment system (the first operable
     unit ROD for the Site) to ensure protectiveness.

6.   Comment:  Several residents questioned the current safety of
     the water provided by the NFTD.  One believes it unsafe for
     drinking or any other use.  If immediate cleanup does not
     occur, the resident requests bottled water.  Another is
     angry that the NFTD and the State of Connecticut force him
     to consume carcinogenic water as long as contaminant levels
     are kept within recommended ppm limits.  He resents having
     to consume any amount of carcinogenic water and pay for it.
                               35

-------
     EPA Response:  EPA considers the water supplied by the NFTD
     to be within the standards set under the federal Safe
     Drinking Water Act and therefore safe to drink.

7.   Comment: The Noxvalk First Taxing District questions its
     identification as a potentially responsible party, and
     requests either the reason for its listing or removal from
     PRP listing.  Also, the NFTD requested to be kept informed
     about work at the site.

     EPA Response:  This Record of Decision does not address
     liability issues, but deals with the selection of the remedy
     for the Site.  Therefore, EPA will not address in detail
     comments regarding liability.  However, the NFTD is a
     current owner of a portion of the site and therefore, falls
     within one of the classes of potentially responsible parties
     under section 107(a) of CERCLA.

     EPA will keep the NFTD informed about the work at the site.

8.   Comment:  A wetlands restoration expert working on the
     Norwalk River is apparently concerned that either the study
     or remediation area is not large enough, as he has observed
     repeated failures of vegetation in Deering Pond and skin
     irritation from water contact.  He suspects chemical
     contamination is responsible and not addressed in the RI/FS.

     EPA Response:  The studies done by EPA and the cleanup plans
     address the contamination of the Kellogg-Deering aquifer.
     The scope does not include Deering Pond at this time.
     However, remediation of the aquifer will improve, to some
     degree, surface water quality throughout the area.  EPA
     intends to further evaluate surface water quality in the
     future.

9.   Comment:  No mention was made about fines against the
     polluters.

     EPA Response:  EPA is considering its enforcement options
     for the site.  In general under Superfund,  EPA attempts to
     recover the money spent on the studies and attempts to get
     the potentially responsible parties to implement the cleanup
     work.
                               36

-------
                  EXHIBIT A

     EPA COMMUNITY RELATIONS ACTIVITIES
    CONDUCTED AT THE KELLOGG-DEERING SITE
EPA issued a press release announcing funding for study
of the sits

March 1984, EPA issued a community relations plan
identifying community concerns and EPA plans to inform
and involve the public in EPA decisions

July 11, 1989, EPA held a public meeting in Norvalk to
discuss study of the site

October 1985, EPA released a fact sheet describing
field studies to date and plans for further
investigation

April 1986, EPA released a fact sheet presenting the
results of the remedial investigation of the site

May 5, 1986, EPA issued a press release announcing the
availability of the Remedial Investigation report for
public review and announcing a public meeting on the
site

May 22, 1986, EPA held a public meeting to discuss the
results of the Remedial Investigation and answer
community questions

June 1986, EPA issued a fact sheet on the results of
the Feasibility Study evaluating options for well-head
treatment of water pumped from the site

June 1986, EPA released a revised community relations
plan to update information on community concerns and
EPA activities planned to inform and involve the public

July 2, 1986, EPA issued a press release announcing a
public comment period on the cleanup options for the
site to be held from July 18 - August 7; and announcing
EPA's schedule for a public meeting and public hearing
on the cleanup options for the Well Field

July 17, 1986, EPA held a public informational meeting
in Norvalk to.present results of the Feasibility Study
and answer questions

July 31, 1986, EPA held a public hearing to accept oral
comments on the recommendations for well-head treatment

                      37

-------
September 1986, EPA issued a fact sheet and press
release describing the well-head treatment option
selected for the site

May 5, 1987, EPA issued a press release to announce an
Administrative Order under which the Norwalk First
Taxing District would be required to repair, test and
operate the air stripping unit at the site

EPA announced the availability of the Administrative
Record for public review in Norwalk, Connecticut and in
Boston, Massachusetts

January 1988, EPA issued a press release announcing
plans for a Supplemental Remedial Investigation of the
source areas of site contamination, and the
availability of the workplan for public review.  EPA
also released a fact sheet describing the planned
supplemental investigation

July 18, 1989, EPA issued a press release announcing 1)
completion and availability of the Feasibility Study;
2) a public comment period (to be held July 27-August
25) on the Feasibility Study and Proposed Plan for the
second operable unit at the site; and 3) the schedule
for a public meeting and public hearing on the
Feasibility Study and Proposed Plan

July 26, 1989, EPA issued its Proposed Plan
recommending a cleanup option to address sources of
contamination at the site.

July 26, 1986, EPA held a public informational meeting
in Norwalk to present the results of site study and the
Proposed Plan and to answer questions

August 4, 1989, EPA issued a press release providing
information on the public hearing

August 14, 1989, EPA held a public hearing to accept
oral comments on the Proposed Plan to address sources
of site contamination
                      38

-------
            EXHIBIT B

TRANSCRIPT OF THE PUBLIC HEARING
      KELLOGG-DEERING SITE
               39

-------
                        UNITED  STATES OF AMERICA

 1                   ENVIRONMENTAL  PROTECTION AGENCY

 2                          .  . BOSTON REGION

 3

 4   In the Matter of:

 5   PUBLIC HEARING
    EPA PROPOSED CLEANUP  FLAN  FOR
 6   THE KELLOGG-DEERING SUPERFUND  SITE

 7

 8

 9                              Norwalk Pu.Diic Licrary
                               1  Belden Avenue
10                              (Morwalk, Connecticut

11  !                            Monday
                               August 14,  1989
12

13
                  .e aDove er,-;itien matter came on for hearing,
14   pursuant to Notice  at 7:40 p.m.

15

16   BEFORE:     MATTHEW HOAGLAND,
                Remedial  Project Manager
17               MARGARET  LESHEN,
                Connecticut  Superfund Section Chief
18

19

20

21

22

23

24

25

                          APEX  REPORTING
               Registered Professional Reporters
                          (.617') 426-3077

-------
 1                            I_ti_£_I_*

 2    iEfeAKER:                                   PAGE

 3    Ms. LeSken

     Mr. Hoagland

 5    Mr. Barske                                  8

 6    Mr. Wolfe                                   1!

 7

 8

 9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25
                           APEX REPORTING
                 Registered Professional  Reporters
                           (£17:) 426-3077

-------
 2                                            ('.7:40  pm)

 3               MS. LESKEN:  ...for the Super fund  Section,

 4   and I want to welcome everyone here to the Public

 5   Hearing for the Kellogg-Deering Superfund Site.

 6               Can everyone hear me?  -Just  so everyone

     knows, these are mikes for transcribing  the meeting.

 8   They are not amplifying mikes.  So if you heed us to

 9   talk louder, just let us know.

10               I'm going to just tell you about what we're

     going to be doing tonigV.t.                     :

12               It is -- the meeting is going to be  broker:

13   into three carts.  Matt Hoagland, the Froject•Manager,

14   is going to make a very brief presentation about the

15   proposed plan and preferred alternative  for the  site.

16               He did come down here July 2bth.   So you

17   did get an opportunity to hear a more detailed

18   presentation and to ask questions.

19               After Matt makes a presentation, we're

20   going to have the actual formal comment  period.  At

21   that time, I will call whoever has requested the

22   ability to make a comment up, and they will come up and

23   make their comment into the mikes.

24               If at anytime during the meeting' you decide

25   you want to make a comment, I would ask  that you just
                          APEX REPORTING
                Registered Professional Reporters
                          (£17)426-3077

-------
     go to the back, or raise your hand, and Will Schreocer

 2   (phonetic) from our office will have you fill out a

 3   card.  Me do need to know your names so that we  can

 4   read it into the record.

 5               After we do that and everyone has a  chance

 6   to have their -- make their comments into the record,

     we will close the meeting and stay around for any

 8   informal comments and questions -- a short question and

 g   answer period.

10               And as everyone heard at their meeting on

     -- on the meeting on July 2£th, there is a formal

12   public comment period for the Keliogg-Deering Superfund

13   Site for the proposed plan that ran from July 27th to

14   August 25th.  Ycu can re«d your comments into the

15   record tonight on the plan, or you can send written

16   comments to Matt Hoagland, the Project Manager,  and his

17   address is on the second page of the proposed plan that

18   we have in the back.  If anyone needs any copies of the

19   proposed plan, we can just -- Will can get them  to yc-<.;.

2o               With us tonight, we, also, have Chris

21   Atkinson (phonetic) and Mike Powers from the

22   Connecticut Department of Environmental Protection.

23               Does anyone have any questions on how the

24   hearing is going to be run tonight?

25               (Pause.>
                          APEX REPORTING
                Recistered Professional Reporters
                          (.617) 426 -3077

-------
 1                MS. LESKEN:   Yes?

 2                FROM THE FLOOR:   If  you don't understand

 3    something that's in the  report,  when can we ask that

 4    kind of a question?

 5                MB. LESKEN:   In  terms  of a clarifying

 6    question on the report?

 7                FROM THE FLOOR:  Yes.

 8                MS. LESKEN:   At  the  end of the meeting.

 g                FROM THE FLOOR:  At  the end of the meeting"'

10                MS. LESKEN:   Yes.  When the people are.-

11    reading.'the couiir.ents or  questions  into the record,  that

12    is a formal part cf the  hearing, and we will  not be

13    responding.

14             .   We will be c\round  afterwards to answer any

15    clarifying questions.

16                (. Fau.se. )

17                MS. LESKEN:   Okay.   I  don't hear  any other

18    questions.

19                I'm going  to have  Matt make a very brief

20    presentation about the proposed  plan.

21                MR. HOAGLAND:  Okay.   I'd  like to welcome

22    everyone here tonight.

23                As I presented it  earlier,  on July 26thf  we

24    have basically two contaminates  of concern out at the

25    site.  There is trich1 oromethane and tetrachloroethane.
                          APEX  REPORTING
                Registered  Professional  Reporters
                          C6l7:>42e-3077

-------
 1   These are known as volatile organic  compounds.

 2               They have been detected  in soils  beneath

 3   the buildings of an area that we  call the Elinco/Fitney

     Bowes/Matheis Court Complex.  That's an area  that's

 5   just on the east side of -- of Main Street, in Norwalk.

 6               Contaminates have also been detected in the

     indoor air of the buildings, and  in  ground water of the

 8   aquifer.

 9               Tlit EFA's preferred alternative is really

10   ir> three parts.

11               The first part is called an in-situ vacuum

12   extraction,  which is designed to  be  used primarily for

13   contaminants in unsaturated soil, soils that  are above

14   the ground water table.  Those contaminants can be

15   removed using this technology from the soils  themselves:

16   anci also this technology, the EFA believes, will cutoff

17   farther introduction of contaminants to the indoor --

18   into the indoor air of the buildings.

19               Thfe ground water contamination has been

20   divided into -- the area of ground water contamination

21   has been divided into two zones.

22               The -- there is a very highly concentrated

23   area of ground water contamination.  This is  called the

24   source area.  The other area that remains --  that's

25   Beyond the source area or further downgrade in the
                          APEX REPORTING
                Registered Professional Reporters
                          (617)426-3077

-------
                                                            7


 1    source area is called the downgrading area.  The source


 2    area, has higher levels of contamination then the


 3    downgrading area.


 4                The EFA's proposed or preferred


 5    alternative,  as described 'in'.the proposed plan, will  --


 6    discusses how extraction wells would be placed in both


 7    the  source and downgrading areas,..to extract, and

                                    • f' i '
 8    treat,  and discharge contaminated.ground water.  I have


 9    three additional  points to make:..'


10                One is that-.-- that ;there is treatment at


11     the  Kel logg-Deering wells.  Th.'i
-------
 1    that  that  was something that was screened out very

 2    early on  in the feasibility process, and that's real I/

 3    all  I have to say  right now.

 4                MS.  LESKEN:  Okay.  We're going to start

 5    the actual hearing part of the presentation.

 6                Again,  if anyone wants to make a comment,

 7    they  should just give their card or get a card from

 8    Will  in the back.

 9                The first, person that wanted to speak was

10    Val Wolfe.

11                ''.Pause. '.)  •'••"'.

12                --.  U'C_FE;   I have no comment.

13                M5.  LESKEKi•  Oh, none right now.

14                Ckay.   The\-second as P.  Barske.

15                I*.R.  BARSKE:-  It's Barske.

16                MS.  LE3KEM:  Barske.   If you could come

17    forward'"'

18                (Pause. )

19                MR.  BARSKE:  The name you have is

20    Philip Barske,  Fairfield,  Connecticut.

21                 You  may wonder why I'm in Norwalk.   I've

22    been  in Norwalk  since £:30 this morning working near

23    the river.

24                I've gone through the plan briefly.   I'm

25    not a  chemist,  and  I  can  only hope that the  agency has
                           APEX  REPORTING
                Registered Professional  Reporters
                           C617)42£-3077

-------
 1    taken the right one,  but I do have a little odd     ; '  ;

 2    experience along this river that I thought would be

 3    worth presenting,  and I believe I wrote a brief note to

 4    Mr.  Hoagland.

 5                I've been doing some Army .Corps restoration

 6    work at  Deering Pond, and my concern, basically, is the-

 7    clean-up plan  -- where does it start or where should it

 8    start.

 9                Now,  I've been involved in wet land work

10    creation for about  forty years,  and I've never had the

11     odd  experience that I had at Deering Fond over the past

12    three years.

13                Ve-r/ briefly,  I put  in some plantings.   The

14    first year  they failed because there was no b^se

15    material.   Somebody had p-.it seed plant -- upland plants

16    in the water,  and,  of course,  they failed.   There wa-i-

17    no substrata,  so the  first year  it failed.   I wasn't

18     concerned  why,  other  than the  geese,  the swans,  and the

19     muskrats.                               •'-.-••

20                 The second year,  I brought in 100 yards of

21     tested soil  from an excavation site in Trumbull,

22     Connecticut.   I  had that tested  by the Baron

23     Laboratories for heavy metals.   We put in a planting,

24     and  I thought  by fall  we had a fairly successful

25     vegetative  planting which  I  can  illustrate  here.
                          APEX  REPORTING
                Registered  Professional  Reporters
                           C617:>42£-3077

-------
                                                           10

 1                What  I'm driving at is vegetative: reaction,

 2   c.nd  then  some chemical  or medical reaction I've noticed

 3   from the  area.

 4                We  would get -- during the period that I

 5   planted,  when we  had low water,  we had excellent

 6   growth, and  on  occasion -- I've kept very close notes

 7   of this from 7/25/BS.

 8                The first year I wipe off because there wat

 9   no strata -- organic strata.  The second year we- took

10   care of swans,  geese, ad infinitum.   When we had an

11   inundation of this  vegetation,  we have a practically

12   90%  kill.

13                Now,  at  that same time,  on 7/B/E7,  wherr we

14   were planting over  the  organic  matter,  I had «,n odd

15   sense that something was not right there.   The odor,

16   feel,  and I  made  everyone wear  boots and rubber gloves.

17   There is  one young  man  -- if you see him here,  you car,

18   see  why -- he couldn't  wear  boots.   They Just wouldn't'

19   fit  over  his thighs.

20                He  waded in the  water,  and  about three days

21    later  --  I don't  have pictures  of his legs,  but his

22   arms and  his legs erupted in huge welts,  highly

23   irritated, very itchy.   Now,  that happened in '87.

24             •   At  that  same time,  I  got above water therev

25   and  I  got  it  right in the elbow.   That  was excised by c
                          APEX  REPORTING
                Registered  Professional Reporters
                          (£17)426-3077

-------
 1    dermatologist,  and sent to the lab, and found nothing.

 2    Evidently it was a chemical burn of some type.

 3                That -- evidently there was some organism

 4    or  some chemical aspect.  You can see the eruptions.

 5    Tom's leg from knee up -- ankle -- was covered with

 6    this mass of welt.

 7                Well,  very briefly,  the following year we

 8    replanted,  had  a fairly good growth, until we he\d high

 9    water.   With high  water, we practically lost all of the

10    growth,  and on  May 15, the same young man planted for

11     nie,  ancj  that's  wnen he developed the rash on his hands.

12    That  time he had boots.   He was fearless,  with rubber

13    gloves  and  he developed that rash again.

14                Now,  since then the plantings have been i r,

15    poor  shape.   I  worked with plants many years.  I have a

16    feeling  that something is amiss in this area.

17                Now,  I  did talk to hike Noonan,  from the

18    Army  Corps.   I've  forgotten what date.   I  have it in

19    here.

20                (Pause. )

21                 MR.  EARSKE:   Oh,  7/27/E7,  Mike Shean.  He

22     was overseer of  this  project.   I  checked  with him,  told

23     him what  had happened.   He said  that he would contact

24     EPA.

25                 Now, I  never  heard from anyone.   I
                           APEX  REPORTING
                Registered  Professional  Reporters
                       *   ^. £ 17 .> 4 '^! 6 — o) U / /

-------
                                                            i;

 1    contacted  Rich LeClair, Dr. Steve Hitchcock, anc! Brian

 2    Kielbana,  and then they sent a young man,

.3    Michael  Cunningham,  on E/17/87, to look at this area.

 4                Now,  in the river, itself, I don't know.

 5    There is an area  we planted.  There is one under drain,

 6    and  while  I was working on Route 7, on 7/10 -- 10,  11,

 7    I  happened to get an aerial photograph of this proposed

 8    -- of the  highway route,  which went right over Deering

 9    Po.-id,  and  the site that I worked on.   That had been a

10    gravel  pit and then a dump.

11                I had talked  with a Dr.. William Neering,

12    roughly, at I hat  time,  and he said that -- you know,

13    that  was 7/16/37.   ! talked to Dr. Nse'ring, and he  just

14    said  that  w.:-s irritated here.

15                Finally,  he called me at  night, arid he s&iij

16    that  I  had visions of that as a suspected dump sits or

17    toxic  site.   Whether it -Wciis or  not, I don't know.

18    Michael  Cunningham from DEF did come  down,  and he

19    reported no toxics from this area.

20                Now,  all I  can report is  the three years

21     that  I  have watched  this  area.   This  year the growth is,

22    horrible.   At high water,  all  the spring water,  the

23    vegetation just did  not come through.

24                 To me  there is something  in this system.  I

25     know  I'm beyond the  well  field area,  but I  am concerned
                           APEX  REPORTING
                Registered Professional  Reporters
                           <£17:>42<£-3077

-------
                                                           13

 1    with  this,  because thirty years ago I signed the flood

 2    control  small  water shed project for this Norwalk

 3    River.   I  worked in Norwalk a great cieal and will be

 4    the next couple of weeks.

 5                Working in this area,  I just had a sense

 6    that  something was amiss.  It tied in with this program

 7    in the same pond area.   Now,  I won't venture into that

 8    thing without  boots and rubber gloves.   My gloves come

 9    up to her&  now.

10                My feeling is,  "In doing are we neglecting

11     a portion  of the violated system of the river?"

12                Basically,  that is all I have to say.  I

13    can't comment  on the clean-up systems.   I do not know

14    proper techniques,  but  I plan to follow it very

15    c a f e f u 11 y .   T h a n k  y o u.

16                MS.  LESKEN:   Thank you.

17                MR.  BARSKE:   Now,  if you want these,

18    they're  yours.   You may not want them,  but--

19                COURT  REPORTER:   For the re-cord,  I  would

20    like  the sp&llings of the names you used.

21                 MR.  BARSKE:   Of the names?

22                COURT  REPORTER:   Of the people involved.

23                 MR.  BARSKE:   Tom  Rochovanski,

24     R-0-C-H-O-V-A-N-S-K-I.   Dr. William Neering.

25                 Now, those  were medically treated,  and  we
                          APEX  REPORTING
                Registered Professional Reporters
                          (£17:) 426-3077

-------
                                                           14

 1   never  found  out  what they were,  and never found out

 2   what the  bit of  tissue that they excised from my arm.

 3   They did  both bacteria and chemical on it,  but it did

 4   sear into the skin,  and made a definite pit, and I had

 5   something that had  to be removed.

 6                Now,  as I say,  I am concerned with the

 7   whole  river  system  in that area.

 8                Thank you.

 9                MS.  LESKEN:   Thank you.

10                MR.  KCASLAND:   Thank you.

11                MS.  LESKEN:   Is there anyone else that

12   wanted 4.o n.s-.f e &  comment?

13                (Pause. ')

14                MS.  LESKEN:   Don't be shy.   This is your

15   opportunity.

16                '• Pause. )

17                FROM  THE FLODR:    Can I ask another

18   question?

19                MS.  LESKEN:   Sure.

20                FROM  THE FLOOR:   The gentleman  said that

21    there was no more -- no more dumping at the

22   contaminated  area.

23                Is there any attempt  to go  after whoever

24   dumped?   I mean,  what does  it take to  -- to go back and

25   get compensation  for  that?
                          APEX  REPORTING
                Registered  Professional  Reporters
                          <617:>42£-3077

-------
 "'. ;: -                                                         15

  1       •     .  .  MS.. .LESKEN:   Okay.  This  is -- this part  of

  2    the meeting is  for  formal comment. . So we can either

  3    enter that into  the record,  or we  can end this, and

  4    then, we can help;you.

  5                FROM  THE FLOOR:   Okay.  Let me rephrase it.

  6    Why doesn't the  proposal include going after the

  7    dumpers?                                .

  8                MS.  LESKEN:   Okay.  This  proposed plan is. a

  9    proposed plan., for .the -remedial action,  the te'chn.i.c.e 1

10    action at the: site',..-'  ••           .      '• './•      ...

11             •   There :is. a whole other component of ,our . ••- •

12    our work,  whi.ch.. is  enforcement, which  errlail-'s 'those  .

13    types of things.  . If you want to .just  wait until .-t'hj.-5  .

14    part is over,-  'we'  can go  over- that.          '.'  ••'"'   '' .  -

15               ; This  plt^n-'i-f'iat y-'1-1 have is technical in

16    nature,  in terms  of  th.at.                •/'.'..••

17                Ar.& VKere  any other people that  would ii.ke

18    to  make a comment into the record?

19              - ' (Pause.-. ')   .    ''••.'•        .••-'-.'..-

20                MS.  LFSKEN:   Okay.  So we'll  go  back to--

21             ..  MR.. WOLFE:   Val  Wolfe.

22                MS.  LEShCEN:   --Val Wolfe.

23                MR.  WOLFE: .  I'm  not a chemist,  but I have

24    three items that  I  question  on the list  that  lists the

25    contaminants.
                           AFEX  REPORTING
                 Reuistered  Professional Reporters
                            <:£17:)42S-3-077 • . '  •

-------
                                                            ID

 1                One,  is we received a letter in  the  fall  of

 2    1380  from our First Water District Department, sign&d

 3    by  Mr.  Kelly, the previous supervisor.

 4                They indicated that there was  1  to 4

 5    dioxin,  in — in the tests that they had made.   There

 6    was also ethylene glycol, and I can't pronounce  this

 7    other  one.   Perhaps you can?

 8                MR.  HOAGLAND:  Menthyletheyketone?

 g                MR.  WOLFE:  That's right.  I don't see  them

10    listed on this listing of contaminants, and  I wondered

11    why they .are' not  listed there.

12          '.   •   Thank y: u.              .   .  •

13                MS.  LESSEN: . Thank you..  •-' '. •   .

14       _'.••-   .r; u=». )                 '    '

15        -   .     MS.  LESKEN:  Is. there anyone else that

16    we.u.ld  like to aiu-'r..e a cc-iTiment?

17                ».Pause. )

18                MS.  LESKEN:  If not,  we'll close this pert

19    of  the meeting.,  £.nd j-.'.st so you know, all the comments

20    that were received tonight,  as well as any written

21    comments that are received in our office before

22    August 25th  -- and the address as to where to send  them

23    is  in  the proposed plan -- will be responded to  in  a

24    formal nature,  in a document called a Responsiveness

25    Summary,  when we  actually do our  Record of Decision,
                           APEX PEPORTINB
                Registered Professional Reporters
                           <£ 17') 426 -3077

-------
                                                            17

 1    which is a decision document for the  remediation at the

 2    Kel iogg-Dee.ring s'ite. .    ''..-'  .'  •

 3                For anyone that is interested,,  all  of the

 4    information and the reports are downstairs  in the

 5    Library here,  as well as in our office  in Boston,  if

 6    you  want to look at further information.     .      .   .

 7                We are talking comments tonight  and in

 8    writing,  on both the preferred Alternative* .and  any of

 9    the  other alternatives that are described . in-.the :  .

10    feasibility study which is the large  document. .  .  •

11     downstairs.   So if anyone wanted to look at- it.,, they

12    are  iViore than  welcome to look at that at. anytime,'  too.

13                (Pause. ':•

14                r-'S.  LESKEN:  I don't hear any ' other --• is

15    there anyone else with a formal comment-?

16    close this part of the meeting.

17                (Pause.)

18                M3.  LESKEN:  We will be off the  record.

19                '.Whereupon, on August 14th., • 1389, at

20    8:00  pm,  the public hearing was closed.)

21

22

23

24

25
                           APEX REPORTING   .;
                 Registered Professional Reporters*
                           (61.7)42o-2077

-------
 1               CERTIFICATE OF REPORTER AND TRANSCRIBER
 2

 .               This is to certify that the attached  proceedings
 3

 4   before: ENVIRONMENTAL PROTECTION AGENCY

 .   in the Matter of:
    PROPOSED CLEANUP PLAN FOR
 6   THE KELLOGG-BEERING SUPERFUND SITE

 7

 8

 9
          Place:Norwsik, Connecticut
10
          Date: August 14, 1989
n
    were held as Herein appears, and that this  is  the true,

    accurate anc complete transcript prepared from the notes
13
    and/or recorcings taken  of the above entitled proceecing.

15

16

17
    J.  Luff                              E/14/.89
    Reporter                             Date
    D.  Brideau                           8/21/89
    Transcriber                          Date

19

20

21

22

23

24

25
                         APEX REPORTING
               Registered Professional Reporters
                         <:617:> 426-3077

-------
     Kellogg-Deering'Well Field

  NPL Site Administrative Record

                 Index
         Compiled: October 31, 1988
          Updated: July 26, 1989
       ROD Signed: September 29, 1989
               Prepared for

                 Region I
        Waste Management Division
    U.S. Environmental Protection Agency
           With Assistance from

AMERICAN MANAGEMENT SYSTEMS, INC.
 One KendaL Square, Suite 2200 • Cambridge, Massachusetts 02139 • (617) 577-9915

-------
          RECORD OF DECISION
KELLOGG-DEERING SITE - OPERABLE UNIT 2
            SOURCE CONTROL
              APPENDIX C

       STATE CONCURRENCE LETTER

-------
                  STATE OF
        DEPARTMENT OF ENVIRONMENTAL PROTECTION


                                       Sefttsber 29, 1989
Paul Keough
Acting Regional Administrator
OS EPA  Region I
JFK Federal Building
Boston, MA  02203

Dear Mr. Keough;

    The Connecticut Department of Environmental Protection (DEP)  concurs  with
the  remedy  selected   for  the  second  operable  unit  of the  Kellog-Deering
Superfund Site.  Our concurrence  is provided with  the  understanding  that  the
operable unit  for which a remedy  has been selected consists of the "source
area" of the contamination affecting the Kellogg-Deering Wall Field,  as defined
in the  proposed plan,  and  that the downgradient area will be evaluated as  a
third operable unit for which a separate reaedy and Record.of Decision will be
prepared. Furthermore,  DEP1a concurrence  with  the  selected  rtmedy  for  the
second  operable unit   shall  not  be considered  concurrence  with  a no  action
remedial  alternative  for   the  third   operable   unit.   In  providing   this
concurrence,  the  State  finds  that  the  selected  reaedy  will  satisfy  all
applicable or relevant  and appropriate State laws and  regulations.

    Since it is inconsistent  with Connecticut's tfeter  Quality Standards,  goals,
and  policies  to  endorse  any  significant  delay  in   remediation  of  areas
contributing  contamination to the  Kellogg-Deering Vbll Field,  I am  requesting
that within 60 days of  the  signing  of  this  Record of Decision,  technical staff
from EPA and  DEP meet  to discuss areas  within the third  operable unit  where
additional information  is needed, and to possibly  resolve technical  areas  of
disagreement.


                                      Sincerely,
                                      Leslie A.  Carothers
                                      Commissioner
LACtCAsao
                   fboor

                     165 Capitol Await* • Hvtford, ConMcrion 0(106

                          A*

-------
          RECORD OF DECISION
KELLOGG-DEERING SITE - OPERABLE UNIT 2
            SOURCE CONTROL
              APPENDIX B

     ADMINISTRATIVE RECORD INDEX

-------