Application of the Lagrangian dispersion model to the regional scale using kernel density functions

It will be shown that the the Lagrangian dispersion model has a good performance at variable meteorological conditions in complex terrain in the near field of the source. In order to apply the Lagrangian dispersion model to the regional scale at real time conditions algorithmens have to be introduced to reduce the otherwise lengthy computer time. One reasonable solution is decreasing the number of Lagrangian particles by at least one order of magnitude while giving them a spatial weight to assure stable and smoothed distribution of the concentration in a given spatial grid. Two different kernel density functions will be discussed for the real time application in complex terrain at the regional scale