Use of toxicity identification evaluations to determine the pesticide mitigation effectiveness of on-farm vegetated treatment systems.

Evidence of ecological impacts from pesticide runoff has prompted installation of vegetated treatment systems (VTS) along the central coast of California, USA. During five surveys of two on-farm VTS ponds, 88% of inlet and outlet water samples were toxic to Ceriodaphnia dubia. Toxicity identification evaluations (TIEs) indicated water toxicity was caused by diazinon at VTS-1, and chlorpyrifos at VTS-2. Diazinon levels in VTS-1 were variable, but high pulse inflow concentrations were reduced through dilution. At VTS-2, chlorpyrifos concentrations averaged 52% lower at the VTS outlet than at the inlet. Water concentrations of most other pesticides averaged 20-90% lower at VTS outlets. All VTS sediment samples were toxic to amphipods (Hyalella azteca). Sediment TIEs indicated toxicity was caused by cypermethrin and lambda-cyhalothrin at VTS-1, and chlorpyrifos and permethrin at VTS-2. As with water, sediment concentrations were lower at VTS outlets, indicating substantial reductions in farm runoff pesticide concentrations.

Spatial relationships between water quality and pesticide application rates in agricultural watersheds.

Pesticide applications to agricultural lands in California, USA, are reported to a central data base, while data on water and sediment quality are collected by a number of monitoring programs. Data from both sources are geo-referenced, allowing spatial analysis of relationships between pesticide application rates and the chemical and biological condition of water bodies. This study collected data from 12 watersheds, selected to represent a range of pesticide usage. Water quality parameters were measured during six surveys of stream sites receiving runoff from the selected watershed areas. This study had three objectives: to evaluate the usefulness of pesticide application data in selecting regional monitoring sites, to provide information for generating and testing hypotheses about pesticide fate and effects, and to determine whether in-stream nitrate concentration was a useful surrogate indicator for regional monitoring of toxic substances. Significant correlations were observed between pesticide application rates and in-stream pesticide concentrations (p < 0.05) and toxicity (p < 0.10). In-stream nitrate concentrations were not significantly correlated with either the amount of pesticides applied, in-stream pesticide concentrations, or in-stream toxicity (all p > 0.30). Neither total watershed area nor the area in which pesticide usage was reported correlated significantly with the amount of pesticides applied, in-stream pesticide concentrations, or in-stream toxicity (all p > 0.14). In-stream pesticide concentrations and effects were more closely related to the intensity of pesticide use than to the area under cultivation.