Modeling the influence of intermittent rain events on long-term fate and transport of organic air pollutants

The deposition of particles and substances in air is under strong influence of the precipitation patterns of the atmosphere. Most multimedia models, like type III Mackay models, treat rain as a continuous phenomenon. This may cause severe overestimation of the substance removal from the atmosphere through wet deposition and an underestimation of travel distances, leading to the following questions: How strong is the influence of the intermittent character of rain on concentrations, residence times, deposited fractions, and characteristic transport distances of different substances in air? Is there an expression that can provide an accurate approximation to be used in steady-state multimedia models? Assuming a periodically intermittent rain, the mass of an emitted substance that is present in the air compartment is calculated as a function of the deposition rate constants during dry and wet periods and the durations of these periods. In this paper, results for 300 different organic chemicals are presented and illustrated in more detail for four typical substances, showing the following: (i) Deposition velocities can be up to 4 orders of magnitude higher during rain events than during dry periods, especially for persistent substances with low Henry constant. (ii) For substances with a short reaction time (residence time as determined by atmospheric degradation alone) (e.g., propoxur), the assumption of continuous rain may lead to an underestimation of the atmospheric residence time and travel distance by up to 3 orders of magnitude. For this group of substances, the residence time during dry period provides a good estimate of the overall atmospheric residence time. (iii) For substances with reaction times close to the duration of the dry period, the behavior is driven by the length of the time interval between two rain events, as for example, for methomyl. (iv) For very persistent substances such as pentachloronitrobenzene or carbon tetrachloride, the continuous rain approximation provides a good estimate. On the basis of these findings, an accurate but simple approximation is provided by eq 17 for the incorporation of intermittent rain behavior in steady-state multimedia models.