Skip to navigation Skip to content

Refined Pesticide Exposure Modeling for Endangered Species in Flowing Water Habitats

Refined pesticide exposure modeling for endangered species in flowing water habitats.

Authors: Michael Winchell, Naresh Pai, and Lauren Padilla, Stone Environmental; Raghavan Srinivasan, Texas A&M University; Patrick Havens, Dow AgroSciences LLC; Jeff Giddings, Compliance Services International


In their recent Biological Evaluations for three organophosphate insecticides, the USEPA concluded that improved pesticide exposure modeling approaches are needed for simulating pesticide exposure in flowing water habitat. To address this need, a species-specific, spatially distributed watershed scale exposure modeling approach at the 12-digit hydrologic unit code (HUC12) resolution was applied to evaluate chlorpyrifos exposure to flowing water habitats. Over 80,000 HUC12 watersheds cover the continental United States, including streams and rivers representing the low, medium, and high flow generic habitats proposed by the USEPA and the Services. The modeling approach combined the PRZM5 model, the traditional field scale landscape model used by the EPA, and the Soil and Water Assessment Tool (SWAT) watershed scale water quality model. The PRZM5 model was applied in a distributed manner to represent the variability in landscape and agronomic conditions within each HUC12 watershed. The parameterization of the PRZM5 simulations incorporated best available spatial datasets describing the land use and cropping patterns, soils, and weather. The timing of chlorpyrifos applications was based on regional agronomic practices and varied across watersheds, reflecting differences driven by field conditions, resource availability, and pest pressure. Conservative estimates of the percent of crops treated with chlorpyrifos were also included. The PRZM5 simulations provided surface runoff, erosion, and pesticide fluxes entering each HUC12 stream, while the SWAT model simulated realistic time-varying baseflow contributions to streamflow for each HUC12 watershed, providing a significant improvement to the assumption of constant baseflow. The channel system modules of SWAT simulated the downstream routing of the HUC12-level stream inputs. In-channel pesticide environmental fate processes were accounted for, including water and benthic degradation, equilibrium partitioning, settling and resuspension, and volatilization. The uncertainty in several parameters affecting chlorpyrifos environmental fate, baseflow, and channel routing were addressed by generating an ensemble of model simulations representing the expected range of these parameters. Probability distributions of chlorpyrifos exposures were derived for each species range and habitat type and can be used in combination with probabilistic effects determinations to inform risk characterization for the chemical.