Effect of the entrained air and initial droplet velocity on the release height parameter of a Gaussian spray drift model.

The increased concern about environmental effect of off-target deposits of pesticides use has resulted in the development of numerous spray drift models. Statistical models based on experimental field studies are used to estimate off-target deposits for different sprayers in various environmental conditions. Random-walk and computational fluid dynamics (CFD) models have been used to predict the effect of operational parameters and were extensively validated in wind tunnel. A third group, Gaussian dispersion models have been used for several years for the environmental assessment of the pesticide spray drift, mainly for aerial application. When these models were used for the evaluation of boom sprayer spray drift, their predictions were found unreliable in the short range, were the initial release conditions of the droplets have a significant effect on the spray deposits. For longer ranges, the results were found consistent with the field measurements as the characteristics of the source have a reduced influence on the small droplets drift. Three major parameters must be taken into account in order to define realistic initial conditions of the droplets in a spray drift model: the spray pattern of the nozzle, the boom movements and the effect of entrained air and droplet velocities. To take theses parameters into account in a Gaussian model, the nozzle droplet size distribution measured with a PIV setup to divide the nozzle output into several size classes. The spray deposits of each diameter class was computed for each successive position of the nozzle combining the nozzle spray distribution with drift computed with a Gaussian tilting plume model. The summation of these footprints resulted in the global drift of the nozzle. For increasing droplet size, the release height used in the Gaussian model was decreased from nozzle height to ground level using an experimental law to take into account the effect of entrained air and droplet initial velocity. The experimental law was adjusted on 2 m/s wind tunnel measurements and robustness was evaluated for 1 and 4 m/s.