PETITIONED PUBLIC HEALTH ASSESSMENT
CONNECTICUT CORRECTIONAL INSTITUTION
(a/k/a SOMERS CORRECTIONAL FACILITY)
SOMERS, NEW HAVEN COUNTY, CONNECTICUT
APPENDIX H - ATSDR's EXPOSURE INVESTIGATION ACTIVITIES RELATED TO PCE CONTAMINATION OF GROUNDWATER SUPPLIES IN THE OSBORN CONNECTICUT CORRECTIONAL INSTITUTION (OCCI) AREA, SOMERS, CONNECTICUT
MORRIS L. MASLIA, M.S.C.E., P.E.
Division of Health Assessment and Consultation
Agency for Toxic Substances and Disease Registry
MUSTAFA M. ARAL, PH.D.,P.E.
School of Civil and Environmental Engineering
Georgia Institute of Technology
In a letter dated February 23, 1995, citizens in the Rye Hill Circle area of Somers, Connecticut,requested that the Agency for Toxic Substances and Disease Registry (ATSDR) assess theirexposure to tetrachloroethylene (PCE). As a result of PCE-contaminated groundwater supplies,many of the residential wells have become contaminated with PCE that exceeds theEnvironmental Protection Agency's (EPA) 5 parts per billion (ppb) maximum contaminant level(MCL) for PCE.
ATSDR's Division of Health Assessment and Consultation (DHAC) conducted an exposure investigation (EI) through its Exposure Investigations Section (EIS) and the Exposure-DoseReconstruction Project (EDRP). The EDRP, a cooperative research effort between ATSDR andthe Georgia Institute of Technology (GATECH), is responsible for developing methods andcomputational tools to quantify levels of contaminants transported through the environment fromthe source of contamination to the receptor populations. Geographic Information Systems (GIS)was also incorporated in this analysis to establish spatial relationships between contaminantsources and distribution and receptor populations.
As part of the (EI), ATSDR requested the U.S. Geological Survey (USGS), through aninteragency agreement, to collect and analyze regional hydrogeologic data not available toATSDR or in the scientific literature at the time of the investigation. This report presents thefindings of the EI and contains the data ATSDR asked the USGS to collect.
ATSDR'S EXPOSURE INVESTIGATION ACTIVITIES RELATED TO PCECONTAMINATION OF GROUNDWATER SUPPLIES IN THE OSBORN CONNECTICUTCORRECTIONAL INSTITUTION (OCCI) AREA, SOMERS, CONNECTICUT
- Site and Off-Site Data
Vertical Migration of PCE Through the Overlying Glacial Till
Characterization of Groundwater Flow in the Bedrock Aquifer
Reconstruction of Historical PCE Concentrations in the Bedrock Aquifer
LIST OF ILLUSTRATIONS
Illustrations are located in Appendix A
- Map showing regional and exposure investigation areas.
- Map showing location of production, domestic, monitoring, and observation wells.
- Hydrogeologic cross section B-B'.
- Graph showing vertical distribution of PCE in the glacial till (one-dimensional analytical solution).
- Histogram showing results of Monte Carlo simulation indicating probability of PCE concentration migrating 55 feet in one year (360 days).
- Map showing comparison of the model calibrated potentiometric surface of the bedrock aquifer and measured water levels, August, 1994.
- Map showing direction of groundwater flow in the bedrock aquifer based on simulation, August, 1994.
- Graph showing comparison of simulated and measured time versus concentration data for wells MW-5D and RHC-082.
LIST OF TABLES
- List of simulations and parameter value ranges used for simulating one-dimensional vertical transport of PCE through glacial till
- OCCI production well data, October 1993
|1.0 foot (ft)||=||0.3048 meter (m)|
|1.0 mile (mi)||=||1.609 kilometers (km)|
|1.0 square mile (mi2)||=||2.590 square kilometers (km2)|
|1.0 million gallons (Mgal)||=||3.785 x 103 cubic meters (m3)|
|=||3.785 x 106 liters (L)|
|1.0 foot per day (ft/d)||=||3.528 x 10-4 centimeter per second (cm/s)|
|0.3048 meter per day (m/d)|
|1.0 gallon per minute (gal/min)||=||6.309 x 10-5 cubic meter per second (m3/s)|
|1.0 gallon per day (gal/d)||=||0.09085 cubic meter per second (m3/s)|
|1.0 million gallons per day (Mgal/d)||=||0.04381 cubic meter per second (m3/s)|
|1.0 inch per year (in/yr)||=||25.40 millimeter per year (mm/yr)|
|1 part per billion (ppb)||=||1.0 microgram per liter (µg/L)|
Listed below are acronyms and their definitions that are used throughout this report.
|ACTS||Analytical Contaminant Transport System|
|ATSDR||Agency for Toxic Substances and Disease Registry|
|CTDEP||Connecticut Department of Environmental Protection|
|DHAC||Division of Health Assessment and Consultation, ATSDR|
|DWEL||Drinking Water Equivalent Level|
|EDRP||Exposure-Dose Reconstruction Project|
|EPA||U. S. Environmental Protection Agency|
|GATECH||Georgia Institute of Technology|
|GIS||Geographic Information System|
|MCL||Maximum Contaminant Level|
|NPL||National Priorities List|
|USGS||U.S. Geological Survey|
|VOC||Volatile Organic Compounds|
ATSDR'S EXPOSURE INVESTIGATION ACTIVITIES RELATED TO PCE CONTAMINATION OF GROUNDWATER SUPPLIES IN THE OSBORN CONNECTICUT CORRECTIONAL INSTITUTION (OCCI) AREA, SOMERS, CONNECTICUT
Tetrachloroethylene (PCE) has been detected in groundwater supplies in the Osborn ConnecticutCorrectional Institution (OCCI) area, including the Rye Hill Circle neighborhood, in Somers,Conn.(3) Contaminant concentrations based on measured groundwater samples were found torange from nearly 5,000 parts per billion on the OCCI property to more than 500 parts per billionin 1 residential well.(4) In the residential wells, PCE concentrations ranged from 545 parts perbillion to below detection limits. Estimates based on analysis of groundwater movement by useof field data and computational methods indicate that residential wells in the Rye Hill Circle areahave probably been contaminated since their installation in 1978 through 1981. Thus, citizenshave probably been exposed to PCE-contaminated water for 16 years from 1978 through 1993,when carbon-activated filters were installed on each well.
Citizens of the Rye Hill Circle area of Somers, Conn., sent a letter to the assistant administrator,Agency for Toxic Substances and Disease Registry (ATSDR), February 23, 1995, asking thatATSDR assess the citizens' exposure to tetrachloroethylene (PCE). As a result of PCE-contaminated groundwater resources, many of the residential wells in the Rye Hill Circle areaand two of the Osborn Connecticut Correctional Institution (OCCI) production wells havebecome contaminated with PCE at levels that exceed the Environmental Protection Agency's(EPA) 5 ppb maximum contaminant level (MCL). In response to the citizens' request, ATSDRinitiated an exposure investigation (EI) as part of the public health assessment process.
ATSDR's Division of Health Assessment and Consultation (DHAC) conducted its investigationthrough the Exposure Investigations Section (EIS) and the Exposure-Dose ReconstructionProject (EDRP). Through an interagency agreement, ATSDR asked the Connecticut District,Water Resources Division, of the U.S. Geological Survey (USGS), to collect and analyzeregional hydrogeologic data not available to ATSDR or in the scientific literature at the time ofthe investigation. This EI report also presents the data the USGS gathered (Appendix B).
The ATSDR EI addresses two issues in response to the citizens' request: (1) estimating thelength of time the residents in the Rye Hill Circle were obtaining PCE-contaminated water fromtheir wells and (2) reconstructing historical levels of PCE contamination to estimate exposure toPCE-contaminated groundwater. ATSDR responded to the requests with the following activities:
(1) analyzed site and off-site hydrogeologic and geochemistry data by
- (a) reviewing data the Connecticut Attorney General's office provided ATSDR and
(b) contracting with the USGS to collect regional hydrogeologic data from the Eastern BorderFault to the Connecticut River (Appendix B);
(2) analyzed the movement of PCE through the glacial material in the OCCI sand filter bed areaby using analytical contaminant transport modeling techniques;
(3) characterized and analyzed groundwater flow in the bedrock aquifer in the OCCI and the RyeHill Circle areas by using numerical groundwater flow modeling techniques; and
(4) reconstructed historical PCE concentration levels for the bedrock aquifer in the OCCI andthe Rye Hill Circle areas and estimated the duration of residential exposure to PCE-contaminatedgroundwater by using numerical contaminant transport modeling techniques.
Previous investigations in the study area have included evaluations of the following:
locating additional groundwater supplies in the surficial and bedrock aquifers (Fuss &O'Neill, 1992a);
soil gas in two potential PCE source areas at the OCCI facility (Fuss & O'Neill, 1992b);
PCE contamination of the overlying glacial aquifer and in the bedrock aquifer on theOCCI property (Fuss & O'Neill, 1993);
the environment in the vicinity of the former sand filter beds at the OCCI facility (Fuss &O'Neill, 1994a);
the hydrogeology and assessment of PCE contamination of the bedrock aquifer in theOCCI area (Fuss & O'Neill, 1994b).
The investigations in 1993 and 1994 inventoried residential wells and determined PCEcontamination in water samples from the wells. ATSDR has completed a health consultation forthe town of Somers (ATSDR, 1994) and most recently a petitioned public health assessment forthe area (ATSDR, 1995).
The OCCI and the Rye Hill Circle area are in Tolland County, town of Somers, north-centralConnecticut, near the Connecticut-Massachusetts border, in the Connecticut Valley lowlandstopographic region (Figs. 1 and 2). The site of the EI has been the subject of ongoinghydrogeologic investigations (Fuss & O'Neill, 1992a; 1992b; 1993; 1994a; 1994b). In additionto these investigations, the USGS collected and analyzed hydrogeologic data in the area from theConnecticut River to the Eastern Border Fault. Appendix B of this report contains the data. Areview of these data indicates that groundwater in the area of the site generally occurs in twoaquifers: (1) the fine-grained drift aquifers of the Upper Connecticut River Valley and (2) anextensive bedrock aquifer, the Portland Arkose. The fine-grained stratified drift aquifers arecapable of providing only limited supplies of groundwater and generally produce yields of 10gal/min or less. The domestic wells in the Rye Hill Circle area and the OCCI production wellsare all cased through the overlying surficial material and obtain water from the underlyingbedrock aquifer. Thus, the surficial aquifer is not a direct source of potable water for site wells. In terms of the EI, the overlying surficial aquifer is considered important only in terms ofproviding recharge to the underlying bedrock aquifer and the degree to which the surficialmaterials retard or enhance the movement of contaminants from land surface to the bedrockaquifer.
Figure 3 is a hydrogeologic cross section (B-B') in the vicinity of the sand filter bed area (seeFig. 2 for location of cross section B-B'). The sand filter bed area has been acknowledged as oneof the most likely sources for the PCE contamination. Groundwater samples from the bedrockaquifer directly below this area have resulted in the highest concentrations of PCE-contaminatedgroundwater for the bedrock aquifer (Fuss & O'Neill, 1993; 1994b). The glacial till in this areais approximately 55 ft thick (including the thickness of the sand filter bed). To estimate the timefor vertical migration of contaminated groundwater, we used ATSDR's analytical contaminanttransport system (ACTS) software package (Aral, 1996) to apply a one-dimensional contaminanttransport equation to a vertical distance represented by section B-B' (Fig. 3). We conductedseveral simulations using both deterministic (single value parameter estimates) and uncertainty(Monte Carlo) analysis to account for variations in site parameter values. We obtained parametervalues used in the simulations from site data (Fuss & O'Neill, 1993; 1994a) and from publishedliterature values (Anderson, 1979; Freeze and Cherry, 1979; Gelhar, et. al., 1992).
Table 1 lists the simulations for the glacial till conducted as part of the EI and the ranges inparameter values used for each simulation. Figures 4 and 5, respectively, present results of thesimulations for Run01 (single value analysis) and Run07_mc (Monte Carlo analysis). Results inFigure 4 indicate that after 1 year (360 days), the concentration of PCE at the top of the bedrockaquifer would be about 70% of the initial PCE concentration in the sand filter bed. Thus, basedon an initial concentration of about 1,800 ppb (Fuss & O'Neill, 1994a), groundwater with a PCEconcentration of greater than 1,200 ppb would reach the top of the bedrock aquifer within oneyear. Results of the Monte Carlo analysis (Fig. 5) indicate that there is a greater than 95%probability that groundwater contaminated with PCE exceeding the MCL of 5 ppb reaches thetop of the bedrock aquifer within one year. The implication of these result is that, for anestimated total exposure period of more than 15 years, the time for movement of PCE-contaminated groundwater through the overlying glacial till is insignificant and can be ignored inquantifying exposure from the PCE-contaminated groundwater of the bedrock aquifer.
|Simulation Number||Simulation Type||Groundwater Velocity (ft/d)||Dispersion Coefficient (ft2/d)|
|Mean||Range||Distribution Type||Mean||Range||Distribution Type|
|Run01_mc||Monte Carlo |
|0.3||0.11 - 0.88||Lognormal||3.7||0.7 - 19.6||Lognormal|
|Run02_mc||Monte Carlo |
|0.3||0.11 - 0.88||Lognormal||3.7||0.7 - 19.6||Lognormal|
|Run03_mc||Monte Carlo |
|0.3||0.11 - 0.88||Lognormal||3.7||0.7 - 19.6||Exponential|
|Run04_mc||Monte Carlo |
|0.3||0.11 - 0.88||Exponential||3.7||0.7 - 19.6||Exponential|
|Run07_mc||Monte Carlo |
|0.1||0.001 - 1.0||Lognormal||5.0||0.1 - 100||Lognormal|
*N/A, not applicable to this simulation type.
The USGS has characterized groundwater flow in the bedrock aquifer in an area of north-centralConnecticut from the Eastern Border Fault to the Connecticut River (Appendix B). The regionalgroundwater flow is generally from the eastern highlands across the lowlands to the Connecticutvalley toward the Connecticut River. In the EI area, wells used for potable water in the OCCIand the Rye Hill Circle area withdraw water solely from the bedrock aquifer. There are fourproduction wells on the OCCI property (PW-1, PW-2, PW-3, and PW-4 on Fig. 2). While allfour wells are available for use, well PW-2 has excessive PCE contamination and operates onlyin emergency situations. Table 2 lists wells depths and water production data from October 1993for the OCCI wells.
|Construction and Production Data||Production Wells|
|Ground Elevation (ft)||305||260||275||250|
|Well Depth (ft)||900||500||500||500|
|Depth to Bedrock (ft)||80||52||59||41|
|Design Pumping Rate (gal/min)||--||150||200||150|
|Actual Pumping Rate (gal/min)||43||0||80||135|
|Actual Daily Yield (gal/d)||49,000||0||48,000||81,000|
*See figure 2 for well locations; data from Fuss & O'Neill (Table 3, 1994b).
**For groundwater flow and contaminant transport modeling analyses (discussed later), well PW-2 is estimated to have a daily yield of 48,000 gal/d.
In the Rye Hill Circle area, there are approximately 92 domestic wells that withdraw water fromthe bedrock aquifer (Fig. 2). We do not know actual water use for individual domestic wellsbefore 1994. During 1994, the Connecticut Department of Environmental Protection (CTDEP)gathered monthly water use data by metering 38 of the wells (R. Fill, written commun., 1995). The data indicate that the average daily household water use for July 1994 was 287 gal/d. As acomparison, the U.S. daily per capita water use (which includes domestic, commercial,industrial, and public use) ranges from 60 to 250 gal/d (Fair, et. al., 1971). Thus, for an average4-person household, the U.S. daily per capita rate would result in water usage of about 240 to1,000 gal/d for the Rye Hill Circle area.
To characterize and better understand the nature of groundwater flow in the area of the EI, wecalibrated a numerical groundwater flow model using hydrogeologic and water-use datadescribed above. We used the steady-state layered aquifer model (SLAM) code (Aral, 1989) tosimulate groundwater flow in the study area. Comparisons of model calibrated water-levelvalues with those measured for August 1994 are in good agreement and are shown in figure 6. The local direction of groundwater flow, based on model simulation appear in figure 7. A reviewof modeling results (Figs. 6 and 7) indicates that groundwater flow in the northern and southernparts of the EI area, is consistent with the regional gradient and is from east to west. However,near the sand filter bed area, groundwater flow is complex, divergent, and nonuniform. In thisarea, groundwater--and presumably contaminants dissolved in groundwater such as PCE--canmove north, south, and to the west of the source of contamination. Thus, contaminationoriginating in the sand filter bed area could be transported and dispersed to the north of this area(OCCI production well area) and to the south (Rye Hill Circle area wells) by the action ofpumping wells and the movement of groundwater through fracture and bedding-plane zonescharacteristic of the bedrock aquifer.
We used the CLAM code (Tang and Aral, 1992), a numerical contaminant transport model, toreconstruct historical levels of PCE in the bedrock aquifer and to quantify exposure to PCE-contaminated groundwater. Among the input used by the CLAM code was the simulated velocityfield (Fig. 7) and aquifer parameter values determined during the groundwater flow modelcalibration process. We made the following assumptions in conducting the transport simulations:
(1) Contamination of the bedrock aquifer began when OCCI operations began and for modelingpurposes this was assumed to be January 1, 1963. All four OCCI production wells wereoperational at this time.
(2) Domestic well usage in the Rye Hill Circle area began on January 1, 1978. All domesticwells and OCCI production wells were operational at this time.
(3) Because of excessive PCE concentration, OCCI production well PW-2 (Fig. 2) ceasedoperating in June 1990. Therefore, for modeling purposes, for the period of July 1, 1990,through December 31, 1993, OCCI production wells PW-1, PW-3, and PW-4 and all domesticwells in the Rye Hill Circle area were operational.
(4) All exposure to PCE-contaminated groundwater ceased after December 31, 1993, because ofthe installation of carbon-activated filters on the domestic wells in the Rye Hill Circle area.
Each time a well was put into or taken out of service, we used the SLAM code to generate a newvelocity (flow) field. We then used the simulated velocity field as one of the inputs for thecontaminant transport simulation. For the study area, we generated three different velocity fieldsbased on the pumping conditions described above in items 1-3.
Because of uncertainty as to when contamination began, the precise location of the source ofcontamination, and the concentration of the source of contamination, we conducted manysimulation scenarios to determine the effect of parameter variation on the estimated levels ofPCE contamination. Based on review of the many simulations, the most likely scenario forcontamination of groundwater supplies in the EI area and thereby exposure to PCE is thefollowing:
(1) OCCI operations contaminated the sand filter bed area with PCE.
(2) The PCE volatilized and went back into solution, contaminating groundwater in the glacialtill.
(3) Contaminated groundwater in the glacial till spread laterally as well as moving vertically,infiltrating the bedrock aquifer.
(4) In addition to PCE-contaminated infiltration from the overlying glacial till, PCE movedvertically downward into the top of the bedrock aquifer because of density effects (Pankow andCherry, 1996) and contaminated the bedrock directly below the sand filter bed area.
(5) Simulation results indicate that when the domestic wells in the northern part of the Rye HillCircle area began to use groundwater in 1978, the bedrock aquifer was already contaminated withPCE that exceeded the MCL of 5 ppb.
Simulation results for Rye Hill Circle well RHC-082, and OCCI monitoring well MW-5D areshown in Figure 8 (see Fig. 2 for well locations). This illustration shows time versusconcentration of PCE for these wells as simulated by the model and the PCE concentration in thewell obtained during water-quality sampling for the third quarter of 1993. These results indicatethat based on available data and modeling scenarios tested, the highest concentrations of PCEoccurred in 1993. In addition, the results show that when the domestic well was installed (1978),the bedrock aquifer was already contaminated. Specific results for other wells will varydepending on several factors including proximity to the OCCI production wells, proximity to thesand filter bed area, depth of well, and pumping rate. However, all wells generally will have thesame characteristics as shown by wells RHC-082 and MW-5D (Fig. 8).
The highest measured level of PCE in the Rye Hill Circle area wells was 545 ppb for well RHC-082 (Fig. 8). This is 9% higher than the PCE Drinking Water Equivalent Level (DWEL) of 500ppb (IRIS, 1994), which is a lifetime exposure level specific for drinking water at which adversenoncarcinogenic health effects would not be expected to occur. However, the DWEL for PCEincludes an uncertainty factor of 1,000 and corresponds to an equivalent dose in humans of 1,000times lower than no effect level observed in animal studies (IRIS, 1994). Thus, it is unlikely thatpast consumption of the most highly contaminated well water in the Rye Hill Circle area wouldhave resulted in any acute toxic effects in the affected residents of this area. Furthermore, basedon available toxicological data and studies (ATSDR, 1994), no long-term adverse human healtheffects, including cancer, are likely to occur in the future as a result of past exposure to andconsumption of PCE-contaminated water by residents in the Rye Hill Circle area. However,because PCE has been shown to cause cancer in animals, continued use of filters or connecting topublic water supplies is a prudent public health practice and will further reduce risks to arearesidents.
Based on review of available data and use of computational models, the exposure investigationactivities resulted in the following conclusions:
- Contaminated groundwater moves rapidly through the glacial till overlying the bedrock aquifer. Simulation results indicate that groundwater contaminated with PCE migrates down to the bedrock aquifer within 1 year at concentrations exceeding the MCL (5 ppb).
- ATSDR's EI study indicates that residents in the Rye Hill Circle area may have been exposed to PCE-contaminated groundwater for at least 16 years. Simulation results indicate that the bedrock aquifer was probably contaminated at the time the domestic wells were installed (assuming a well installation date of 1978).
- Modeling results indicate the following:
- (a) contaminated groundwater in the bedrock aquifer is a consequence of the PCE contaminantsource located in the sand filter bed area, and
(b) the highest measured and simulated concentrations of PCE-contaminated groundwater in theOCCI and Rye Hill Circle area wells occurred during the third quarter of 1993 (for example, wellMW-5D had a measured value of 770 ppb and a simulated value of 846 ppb; well RHC-082 hada measured value of 545 ppb and a simulated value of 521 ppb).
- Evaluation of modeling results suggests that the high levels of PCE found in the domesticwells are probably a consequence of the following:
- (a) hydrogeologic conditions allowing the migration of PCE-contaminated groundwater from theOCCI area to the Rye Hill Circle area,
(b) infiltration of PCE-contaminated groundwater from the overlying glacial material to thebedrock aquifer,
(c) additional undocumented sources of PCE, or
(d) some combination of all of the above factors.
- It is unlikely that past consumption of the most highly contaminated well water in the RyeHill Circle area would have resulted in any acute toxic effects in the affected residents ofthis area (ATSDR, 1994). Furthermore, no long-term adverse human health effects,including cancer, are likely to occur in the future as a result of past exposure to andconsumption of PCE-contaminated water by residents in the Rye Hill Circle area.
Anderson, M.P., 1979, Using Models to Simulate the Movement of Contaminants ThroughGroundwater Flow Systems, Critical Reviews in Environmental Control, v. 9, no. 2, pp. 97-156.
Agency for Toxic Substances and Disease Registry (ATSDR), 1994, Health Consultation forTown of Somers (Somers Correctional Facility), Somers, Connecticut, November 1994.
Agency for Toxic Substances and Disease Registry (ATSDR), 1995, Petitioned Public HealthAssessment, Osborn Connecticut Correctional Institution (a/k/a Somers Correctional Facility),Somers, Tolland County, Connecticut, Cerclis No. CTD980522940, [working draft].
Aral, M.M., 1989, Ground Water Modeling in Multilayer Aquifers, Steady Flow (SLAM), LewisPublishers, Chelsea, Michigan, 114 p.
Aral, M.M., 1996, Analytical Contaminant Transport System (ACTS) Software (version 2.5),submitted to the Agency for Toxic Substances and Disease Registry by the School of Civil andEnvironmental Engineering, Georgia Institute of Technology, Atlanta, Georgia, February 1996[unpublished document].
Fair, G.D., Geyer, J.C., and Okun, D.A., 1971, Elements of Water Supply and Waste Disposal,John Wiley & Sons, Inc., New York, 752 p.
Fill, R., 1995, Facsimile Transmission from State of Connecticut Department of EnvironmentalProtection, Population and Water Usage Information, Somers, Connecticut, December, 1995.
Freeze, R.A., Cherry, J.A., 1979, Groundwater, Prentice-Hall, Inc., Englewood Cliffs, NewJersey, 604 p.
Fuss & O'Neill, 1992a, Hydrogeologic Investigation, Enfield/Somers Correctional Facilities,Enfield/Somers Connecticut, January 1992.
Fuss & O'Neill, 1992b, Results of PCE Source Investigation -- Phase I, Somers MaximumSecurity Correctional Facility, Somers, Connecticut, November 1992.
Fuss & O'Neill, 1993, PCE Source Investigation--Phase II, Somers Correctional Facility, SomersConnecticut, October 1993.
Fuss & O'Neill, 1994a, Environmental Investigation, Former Sand Filter Bed Area, Osborn CCI,Somers, Connecticut, November, 1994.
Fuss & O'Neill, 1994b, Bedrock Aquifer Assessment, Osborn CCI, Somers, Connecticut,December 1994.
Gelhar, L.W., Welty, C., and Rehfeldt, K.R., 1992, A Critical Review of Data on Field-ScaleDispersion in Aquifers, Water Resources Research, v. 28, no.7, pp. 1955-1974.
IRIS, 1994, Integrated Risk Information System, U.S. Environmental Protection Agency, Officeof Health and Environmental Assessment, Environmental Criteria and Assessment Office,Cincinnati, Ohio.
Pankow, J.F., and Cherry, J.A., 1996, Dense Chlorinated Solvents and other DNAPLS inGroundwater, Waterloo Press, Portland, Oregon, 522 p.
Tang, Y., and Aral, M.M., 1992, Contaminant Transport in Layered Aquifer Media (CLAM),Georgia Institute of Technology Report No. CE512, 235 p.
APPENDIX A - Illustrations
Figures 1 - 8
APPENDIX B -
Characterization of Groundwater Flow Between the Eastern Border Fault and the Connecticut River in North-Central Connecticut,Hartford and Tolland Counties, Connecticut
1. Technical Project Officer and Research Hydrologist, Division of Health Assessment and Consultation, ATSDR, 1600 Clifton Road, Mail Stop E-32, Atlanta, Georgia 30333.
2. Principal Investigator and Professor, School of Civil and Environmental Engineering, GeorgiaInstitute of Technology, Atlanta, Georgia 30332.
3. Refer to the Contents section of the report for a list of acronyms, abbreviations, and theirdefinitions.
4. Refer to the Contents section of the report for a list of conversion factors from inch-pound units to International System (SI) of units and the appropriate abbreviations.