A probabilistic screening model for evaluating pyrethroid surface water monitoring data.

Synthetic pyrethroids have been detected in recent California surface water monitoring. Filtration is avoided during sample workup because pyrethroids are extremely hydrophobic, tending to sorb to most surfaces. The resultant analytical pyrethroid concentrations reflect both dissolved pyrethroid and pyrethroid associated with suspended sediment in the water column. Such "whole-water" analytical data are not directly comparable to aquatic acute toxicity effect concentrations measured in laboratory sediment-free water. Consequently, any potential aquatic toxicity risk is indeterminate. In this study a simple probabilistic model was developed to allow a screening-level assessment of pyrethroid whole-water monitoring data. The results suggest that water column toxicity of pyrethroids is possible in California's agriculturally-dominated tributaries, and indicate that additional monitoring to better characterize pyrethroid water-column concentrations are warranted. Model refinement will depend on future work that more firmly establishes the relationship between pyrethroid partitioning and bioavailability, and that addresses the potential influence of dissolved organic carbon on pyrethroid sorption and bioavailability.