Modellierung des Ensembleverhaltens von sedimentierenden Hydrometeoren mittels Momentenverfahren

For the treament of clouds and precipitation in weatherforecast or climate models, a direct spectral representation of the microphysical processes is computationally infeasible. Therefore, their effects have to be considered in an approximative, parameterised way. It is then possible that the model results are not sensible in a cloud microphysics context. In order to avoid this, in this dissertation a new parameterisation of sedimentation (two moment method) is developed and tested in a 1D model. The parameterisation is based on a hydrometeor distribution with bounded domain and is implemented in two ways: 1) truncated Gamma-distribution: compared to already existing parameterisations, the model results are considerably improved (for drops as well as for ice crystals). Investigations include the computation time, the sensitivity of the results on the upper hydrometeor size limit and an optimisation concerning this limit. The development of a rain cloud is simulated by adding drop collisions to the test case. 2) Beta-distribution: It contains an additional free parameter. Its diagnostic relationship has a strong impact on the results.