A simple probabilistic estimation of spray drift—factors determining spray drift and development of a model

Abstract Spray drift represents a major mode of exposure in off-crop habitats or surface waters after pesticide spray application. Currently, the estimation of exposure by spray drift is based on a deterministic estimation of the amount of drifting residues, either with the use of default drift values or deterministic models, which, however, do not reproduce the entire range of spray drift observed in reality. However, because a series of data from extensive field trials are available, probabilistic methods based on Monte Carlo simulation can reveal realistic estimates of the entire range of exposures. For the development of a probabilistic spray drift model, previously published data from a series of field trials was analyzed to reveal how these data could be used for the parameterization of a probabilistic model. This analysis showed that wind speed, agricultural equipment (nozzle type, spray pressure), and relative humidity showed the strongest effect on spray drift. But remarkably, the effect differed for different distances from sprayed fields. For example, higher wind speed increased spray drift only at larger distances while it even reduced spray drift very close to field borders. Also spray pressure influenced spray drift predominantly close to fields. After identifying the parameters with the strongest effects, a probabilistic model for the estimation of the exposure by spray drift in off-crop habitats was developed. Spray drift can be simulated for any given distance from fields. It is demonstrated how the exposure and the amount of effects can be estimated when applying this model in real landscapes. Results are compared with a deterministic risk assessment.