Aarhus Universitets segl

610: Modeling of the evaporation of pesticides from soil and leafs after spraying

Sørensen, P.B., Andersen, H.V., Jensen, N.O., Olesen H.R. & Bossi, R. 2007. NERI Technical report No. 610. 43 pp.

 

Summary

 

This report describes a simple model, AirScreen, for determination of the evaporation of pesticides from soil and leafs after spraying and relates this to potential harmfull effects for water living organisms close to the field. The model is developed for calculation of the evaporation from the field after spraying including the resulting air concentration and possible impact in the surface water environment close to the field. A stocastic modelling approach is applied in order to secure a screening, where a pesticide that truly is unacceptably harmful never will be predicted by the model to be surely harmless. The transportation mechanisms in the model are made as realistic as possible for screening purposes. The model is developed based on pesticide evaporation data found in the closely related report (Andersen et al., 2006) and based on other literature values.

 

The uncertainty related to the calculation of evaporation from soil and plant surfaces are quantified together with the uncertainty intervals of the basic physical-chemical parameters. These two sources of uncertainty are included in the uncertainty calculation by the model. Other sources of uncertainty also exist as e.g. the local condition of turbulence and meteorological conditions. The roughness factor of the landscape around the field is also considered. These other sources of uncertainty are handled in such a way that the results are secured against under estimation of the concentration levels and thus focussing on worst case conditions. The uncertainty of the model in relation to the roughness factor is analyzed and identified as not critical for the total uncertainty when compared with the other sources of quantified uncertainty. The uncertainty calculated by the model yields an upper 99 % fractile value in air concentration that is 10-100 time higher than the 50 % fractile value. So, other uncertainty sources need to be in this order of magnitude before they can be considered as serious. Any model for pesticide evaporation will use the same basic physical-chemical parameters as AirScreen and thus be subject to the same levels of parameter value uncertainty. It is therefore not be possible, by any model, which is more complex than AirScreen, to reduce the level of uncertainty predicted by AirScreen considerably. Contrary, more complex models will tend to include additional uncertainty due to lack in critical supporting empirical knowledge about parameter values and functional relationships.

 

The terrestrial environment can be assessed using the calculated air concentration levels when eco-toxicological data are available. However, the situation is that very few such data are available at present. More data exist in relation to water organisms so AirScreen uses this type of data to interpret the results in relation to the surface water environment. This interpretation is done based on an assumed equilibrium condition for the transport of pesticide between the water and the air. This approach makes a fast and easy screening possible. Thus, the missing consideration of terrestrial eco-toxicological data is not due to any prioritization claiming that the water environment is most important to assess, but just a result of a data gap. It can easily be the case that some terrestrial organisms are directly exposure to the air born pesticides and thus highly relevant to be included in a risk assessment of the pesticide evaporation.

 

The developed software is freely available with reference to this report and can be retrieved by contacting Peter Borgen Sørensen, NERI, Denmark, (email: pbs@dmu.dk).

 

Full report in pdf-format (684 kB).