Table of Contents          Next Section

BACKGROUND

Contamination of groundwater resources in Dover Township, Ocean County, New Jersey, including public and private water-supply wells, was identified in the 1960s (Toms River Chemical Corporation 1966) and subsequently documented in the 1970s (ATSDR 2001a,b,c,d). Water-quality analyses, conducted since the mid-1980s, indicate that this contamination has generally consisted of volatile organic compounds such as trichloroethylene (TCE) and semi-volatile organic compounds such as styrene-acrylonitrile (SAN) trimer (ATSDR 2001d). The reader is referred to the following reports for a description and analysis of contamination of groundwater resources in the Dover Township area: ATSDR (1988, 1989, 2001a,b,c,d), Malcolm Pirnie, Inc. (1992), Pinder, et al. (1992), and Sykes (1992, 1995, 2000). The primary source of potable water for the area is groundwater and it is withdrawn primarily from the shallow Kirkwood-Cohansey aquifer. To a lesser degree, the deeper Piney Point and Potomac/Raritan/Magothy aquifers are also used as sources for groundwater (Maslia et al. 2000a, Table 1). Approximately 85% of current Dover Township area residents are served by a public water-supply system (as opposed to privately owned domestic wells). Based on public health assessments conducted for the Dover Township area, ATSDR and NJDHSS have determined that completed human exposure pathways to groundwater contaminants have occurred through private and community water supplies (ATSDR 2001a,b,c,d). Therefore, an analysis of the potential for distribution of contaminants through the water-distribution system was deemed necessary as part of the exposure assessment component of the epidemiological study.

Because the focus of the epidemiologic investigation is on children, exposure at residential locations is deemed the most important exposure opportunity to investigate, although other exposure opportunities, such as at schools and other public facilities, may have occurred. Exposure to water sources that study subjects received (well or well fields) from the water-distribution system can be estimated using the results of the historical reconstruction of water-distribution system operations and residential histories. Given the multiple number of wells and well fields in the distribution system serving the Dover Township area, the ability to track the percentage of water originating from a well or well field was considered a useful analytical tool to help estimate exposure. For the current study, the EPANET 2 water-distribution system model was applied in a diagnostic mode to reconstruct historical water-distribution system operations. Prior to conducting the historical reconstruction analysis phase of the investigation, model simulation results were compared to spatially and temporally varying field measurements in order to better understand and quantify the reliability of model predictions.

Table 1. Water-distribution system configuration, Dover Township area, New Jersey, 1962–96
[Except for number of pipeline segments, all data from annual reports of the Board of Public Utilities, State of New Jersey (1962–96) and Flegal (1997)]

PREVIOUS INVESTIGATION

During the earlier phase of this investigation (Figure 2), ATSDR and NJDHSS gathered synoptic, system-wide hydraulic and operational data in March and August 1998 in order to characterize, as completely as possible, the water-distribution system under present-day operating conditions. Results of these field-data collection activities and the water-distribution system model calibration and testing are described in the report, "Analysis of the 1998 Water-Distribution System Serving the Dover Township Area, New Jersey: Field-Data Collection Activities and Water-Distribution System Modeling" (Maslia et al. 2000a). Specifically, this report describes the following activities:

• Data gathered during field tests conducted in March and August 1998;
• The development, calibration, and testing of the water-distribution system model (EPANET 2) for present-day (1998) conditions;
• A constituent-transport simulation of a naturally occurring conservative element, barium, to further test the reliability of the model calibration; and
• The simulation of the proportionate contribution of water from wells and well fields to various locations throughout the distribution system under 1998 operating conditions.

Results of these activities support the assertion that: (1) the model presented and described is calibrated and is an acceptable and reliable representation of the waterdistribution system operations during 1998, and (2) that constituent transport within the water-distribution system is reasonably simulated by the calibrated model. A more concise technical summary of the analysis is also presented in Maslia et al. (2000b).

DESCRIPTION OF THE PRESENT-DAY (1998) WATER-DISTRIBUTION SYSTEM

The Dover Township area water-distribution system being analyzed has been operating since 1897 and is currently operated by United Water Toms River, Inc. (UWTR). It serves the residents of Dover Township, New Jersey, and communities outside of Dover Township including the borough of South Toms River and a portion of Berkeley Township (Figure 3, Plate 2). At the end of 1998, the water-distribution system served approximately 45,000 customers from a population of about 94,000 persons. The distribution system consists of (Board of Public Utilities, State of New Jersey 1998):

• 488.2 miles (mi) of mains, ranging in diameter from 2 inches (in.) to 16 in.;
• 3 elevated and 6 ground-level storage tanks with a total rated storage volume of 7.35 million gallons (Mgal);
• 23 municipal groundwater wells in 8 well fields with a total rated capacity of 27 million gallons per day (MGD) (18,750 gallons per minute [gpm]); and
• 12 high-service or booster pumps.

A list and description of the present-day water-distribution system storage tanks, wells, and high-service and booster pumps serving the Dover Township area is provided in Maslia et al. (2000a, Table 1). As presently configured, 9 wells discharge directly into the distribution system (wells 15, 20, 31-35, 38, 43); whereas, the remaining 14 wells (21, 22, 24, 26, 28, 29, 30, 37, 39, 40, 41, 42, 44, 45) are used to fill storage tanks (such as the Parkway well field ground-level or the North Dover elevated). High-service and booster pumps are used to supply the distribution system with water from the storage tanks. Not all extracted groundwater receives the same treatment. Components of the treatment system may include filters; aeration; and the addition of lime, chlorine, alum, or permanganate. The reason for this treatment is for filtration, pH control, or purification (Board of Public Utilities, State of New Jersey 1998). The type of water treatment and the reason for the treatment by well and well field is listed in Table 2.

Table 2. Type of water treatment used by the present-day (1998) water-distribution system, Dover Township area, New Jersey
[Data from Board of Public Utilities, State of New Jersey (1998)]

Diurnal or 24-hour demand for water in the Dover Township area, as measured during the 1998 field-data collection activities, is characterized by two typical demand patterns. A minimum- or winter-demand pattern, typical of data collected in March 1998 (Figure 4A), generally occurs from November through mid- May, and a maximum- or summer-demand pattern, typical of data collected in August 1998 (Figure 4B), generally occurs during the summer season from the end of May (Memorial Day) through September. The diurnaldemand patterns obtained from the measured data in 1998 (Figure 4) were used to characterize the historical diurnal-demand patterns for the historical reconstruction analysis. Total water production during the historical period was based on production information obtained from the Board of Public Utilities, State of New Jersey annual reports (1962�96), NJDHSS data searches (Michael P. McLinden, written communication, August 28, 1997), and water-utility databases (Flegal 1997).

An average demand can be approximated by taking the mean of the minimum- and maximum-demand period data. Based on field data collected in March and August 1998 (Figure 4), the average demand is 11.7 MGD; whereas, the average demand for October 1998 is 11.8 MGD, based on data obtained from the Board of Public Utilities, State of New Jersey (1998). Similar computations using monthly water-production data obtained from the annual reports of the Board of Public Utilities, State of New Jersey (1962�96) for every month of the historical reconstruction period indicate that October production consistently approximates the average yearly production.

EXTERNAL EXPERT REVIEW

Throughout this investigation, ATSDR has sought external expert input and review of this project. On November 14, 2000, ATSDR convened an external expert panel to review the approach used in conducting the historical reconstruction analysis and to provide input and recommendations on the preliminary modeling results (ATSDR 2001e). The panel was composed of experts with professional backgrounds from government and academia, as well as the private sector. Areas of expertise included numerical model development and simulation, hydraulic and water-quality analysis of water distribution systems, model calibration, and water-distribution system optimization. Panel members considered the modeling approaches梐 manual adjustment process which conforms as closely as possible to actual water-distribution system operations, and a Genetic Algorithm (GA) optimization approach. The experts indicated that these two approaches were technically sound given data limitations, and provided the following recommendations for enhancing the modeling approaches and historical reconstruction analysis (ATSDR 2001e):

• To identify past demand, ATSDR investigators reconstructed the water-distribution system assuming that past demand was proportional to demand measured in 1998. Investigators should review data to ensure that distribution lines were not incorrectly assigned a demand during the reconstruction analysis and to identify major water users who may have initiated or terminated demand during the historical period;
• ATSDR investigators assumed that wells operated on a 24-hour pumping pattern. Although other operating patterns were possible, this assumption was consistent with available information that described the water-distribution system operations. Investigators should, however, consider how other on-and-off cycling patterns may affect water-distribution patterns;
• The GA approach derived pumping patterns that allowed wells to operate at a fraction of their pumping capacities. Fractional pumping capacities were permitted to provide flexibility to the GA approach and to achieve balanced-flow operating conditions. However, a well is either on or off. Therefore, investigators should relax the pressure and storage tank water-level requirements to increase simulation flexibility. If relaxing these constraints increases flexibility, investigators should reassess water-distribution system patterns using pumping capacities that more closely reflect the on-and-off cycling of wells; and
• Panel members suggested that investigators conduct sensitivity analyses to determine if other possible operating patterns would result in vastly different water-distribution patterns. Sensitivity analyses could be conducted by: (1) applying the GA approach to water-distribution data collected in 1998 to evaluate whether predicted operating patterns match observed operating patterns, or (2) applying the GA approach to find an operating pattern as different as possible from the operating patterns used in the manual adjustment process and assessing resulting differences in water-distribution patterns.

The recommendations of the external expert panel were implemented as part of the historical reconstruction analysis effort. Results of these efforts are presented in conjunction with specific data needs, descriptions of the historical reconstruction simulations, and sensitivity analyses in the report sections that follow.

Table of Contents          Next Section