Nonpoint source groundwater contamination: Hotspots in regional-scale plumes.

摘要

This dissertation was conducted to determine if nonpoint source (NPS) agrochemical applications can generate plumes in the saturated subsurface with local hotspot concentrations of more than one order of magnitude above the maximum contaminant level (MCL). The hotspot question was addressed with concept-development simulations for synthetic data that characterize the saturated-subsurface fate and transport of NPS applications of DBCP (1,2-dibromo-3-chloropropane). The synthetic data were generated with information from the Loague et al. (1998a,b) simulation case study for label-recommended DBCP use in Fresno County, California. This dissertation consists of three phases.; The first phase was to critically assess the DBCP-leaching simulations from the Fresno case study (Loague et al., 1998a). The DBCP water-table loading concentrations from the Fresno case study were compared with new estimates simulated with smaller elements (and less upscaling) than was previously used. Because the new water-table DBCP loading concentrations covered approximately the same range of values as the original estimates, the leaching profiles from the Fresno case study (Loague et al., 1998a) were employed in generating the synthetic data sets.; The second phase was to generate synthetic data for NPS water-table loading concentrations, recharge, porous media properties, and pumping conditions. The water-table loading concentrations vary spatially and temporally in response to multiple NPS applications. The porous media realizations result from the same properties arranged for different levels of large-scale continuity (channels, lenses, and random). The pumping data sets involve wells sited preferentially to intercept the highest concentrations within the plumes.; The third phase was to simulate deterministic fluid flow and DBCP transport for the boundary value problems that result from the different combinations of the synthetic data sets. The more than 11,000 simulations produced a total of 15 one-element hotspots. Generating the DBCP hotspots required specific alignments between locally high NPS water-table loading concentrations, extended low permeability networks, and extremely high pumping rates. Given the array of conditions examined and the rarity of the hotspots produced, it was concluded that point sources, such as spills at mixing sites, are likely to be responsible for many high-concentration hotspots observed in low-concentration regional-scale plumes.