Health & Environmental Research Online (HERO)


Print Feedback Export to File
1160149 
Meetings & Symposia 
Abstract 
The Toms River childhood cancer cluster: Coupled groundwater and water distribution system modeling 
Sykes, J; Normani, S 
2003 
Toms River, New Jersey is the location of a statistically significant childhood cancer cluster. A 1995 cancer investigation indicated that relative to the state, the Toms River section of Dover Township had excess childhood cancer incidence for all malignant cancers combined, brain and central nervous system (CNS) cancers, and leukemia. Children under the age of five were found to have a seven-fold increase in brain and CNS cancer. The community's concern focused on the possibility that exposure to environmental contaminants may be related to the incidence of these childhood cancers. Two Superfund sites in Dover Township were implicated as having a possible impact on the local water supply. One of these, the Reich Farm site, is a source of contaminants to the aquifer that serves a major well field for Toms River. Contaminants in the aquifer include TCE, PCE and styrene-acrylonitrile (SAN) trimer. In 1997, the New Jersey Department of Health and Senior Services and the Agency for Toxic Substances and Disease Registry began an epidemiology study to evaluate the relationship between the environmental exposure pathways and the elevated childhood cancer incidence. Toxicity studies for the SAN trimer were also initiated. Groundwater modeling was undertaken to establish the historical relationship between the Reich Farm site and the municipal well field and to aid in the management and protection of the aquifer and well field to ensure both water quality and quantity. The modeling of the water distribution system for Toms River was also part of the study. Groundwater flow from the Reich Farm Superfund site to the municipal well field for Toms River was modeled for a thirty-year time period using MODFLOW. To account for the growth and development of the well field within the modeling domain, a transient model was constructed. The use of Geographic Information Systems (GIS) and databases to manage, maintain, and compile field observations for model input and calibration was an important part of the work. GIS and databases were important tools in assessing the quality of the data, discovering and correcting errors in the field data (including surveying inconsistencies), as well as providing an efficient and automated means to visualize the data. Model calibration exercises indicated that a more physically based spatial and temporally variable recharge was necessary to account for dramatic fluctuations in water levels due to seasonal variations. The accurate simulation of the transient groundwater flow system was essential for the subsequent prediction of contaminant migration from the superfund site to the municipal wells and then subsequently into the modeled water distribution system. The independent estimation of the adsorption parameters of the SAN trimer on the porous media of the aquifer was an important aspect of the determination of both the average travel time and the breakthrough of the chemical at the municipal well field. The modeling methodology included an uncertainty analysis of the estimated exposure concentration in the water distribution system given uncertain groundwater parameters. Distributed computing with a Monte Carlo analysis was used for this work. The results of the modeling study were used to assist in the definition of the temporal integration periods in the epidemiology study. The predicted historical breakthrough curve of the SAN trimer in the municipal wells correlates with the period with the excess childhood cancer incidence. 
American Geophysical Union Fall Meeting 2003 
San Francisco, CA 
8-12 December 2003