A simulation study of pesticide drift from an air-assisted orchard sprayer

In this paper a random-walk atmospheric dispersion model is used to examine the problem of pesticide drift from a radial flow air-assisted orchard sprayer. The spray source is represented in the model as an extended source surface close to the sprayer but remote from the influence of the mean forced air from the sprayer. Two different source surfaces (i.e. a plane surface down wind of the sprayer and a curved surface enclosing but displaced from the sprayer) are examined. The model takes into account the effects of droplet inertia on turbulence, droplet evaporation, the sink effect Of the crop and the efficiency of typical collectors used to measure drift. In addition, spray drift experiments with different nozzles are simulated by using droplet size distribution measurements to provide appropriate weighting to different droplet size classes and the results compared with experimental data for two types of crop canopy (cereal stubble and an orchard crop with 2 m tall trees). The model is also used to simulate the effect of a practical drift collector (i.e. a 2 mm diameter static line collector used to measure the airborne spray flux during typical drift trials). It is shown that the normalized vertical drift flux profiles given by these simulations of practical collectors agree with both experimental data and the simulations of a perfectly efficient collector. However, the predictions of bulk airborne flux distributions with distance downwind of the sprayer did not show agreement between the perfectly efficient collectors and 2mm diameter line collectors. The simulation suggests that these drift profiles have different forms dependent upon collector efficiency and the drift levels are underestimated by a factor of at least two for the 2 mm dia. collector at 50 m for a fine spray and wind shear velocities between 0·4 and 0·6 m/s (i.e. the maximum wind speed conditions associated with commercial spraying practice in the UK).