Dispersion modelling in complex terrain using wind climatologies

Abstract

If annual mean values of concentration in complex terrain have to be determined, a representative wind climatology, which reflects local flow characteristics, is needed. This gives rise to two problems. First, observational data representative of the dispersion area are frequently not to hand. Second, the flow in the dispersion area itself may show a significant spatial structure, so transport is affected by varying wind speeds and directions. Both problems may be overcome by determining local wind climatologies by numerical modelling. The concept is to deduce terrain–dependent wind climatologies near the ground from the geostrophic wind climatology valid for the source region. Surface observations from the model area may also be included if available. If resulting surface wind climatologies are similar to each other within the dispersion area, then the variation of flow caused by the terrain structure need not be considered when modelling dispersion. In this case, Gaussian plume modelling may be used. If, however, the calculated wind climatologies differ significantly within the model area, then dispersion modelling has to take into account the spatial structure of the flow. This requires a Eulerian or a Lagrangian approach, because simpler methods of dispersion modelling are not capable of considering three–dimensional flow characteristics.