|Maria Baumer (08/2011-01/2012)|
Vergleich zweier Lagarnge'scher Modelle zur Bestimmung des Footprints über heterogenem Gelände
|Support: Wolfgang Babel, Thomas Foken|
Measurements of vertical fluxes above emitting surface are often used to determine the surface emission rate. Therefore the question awakes which area is responsible for the observed flux. This can be calculated and displayer with the so called footprints, that is a function which connects the source of emission with the observed flux. The following study compares two Lagrangian stochastic models and their generated footprints with each other above a heterogeneous area. This conceptual area is inspired of a real measurement site at NamCo Lake in China. The separation of the two areas is designed crosswind and the sensor’s height is 3m. After some preliminary analysis concerning applicability of both models, and also of applicability of the modified version created for this study, the focus is on comparing the predicted footprints above heterogeneous and homogeneous surfaces. Especially for the decision to use the backward model in the routine for characterization of heterogeneous measurement sites, a comparison test between backward and forward model is performed. The preliminary analysis shows that the models can be compared accurate in spite of different input options of some parameters. Furthermore, the analysis shows that instead of lpdm-b_homo the lpdm-b_hetero could be used. The modified version created for this study hardly shows any effects by the presented cases. Nevertheless, it is used for the further course of the investigation. Comparison of the models lpdm-b_IBL and lpdm-b_homo shows that first effects occur on the footprint at different roughness lengths of surfaces, which are still matching well. Great contrasts in the dimensions of footprints arise from different Obukhov lengths, and thus thermal conditions. In comparison to their homogeneous equivalents these differences can be explained logically. Comparison of the models lpdm-b_IBL and forward model already differ in their footprints within homogeneous surfaces. It is clear that depending on the stability both models predict different spreads of footprints and thus comparison of predicted pattern matches only under certain conditions. This is why comparison is complicated with lpdm-b_IBL and forward model. However the results resemble those of the comparison with lpdm-b_IBL above homogeneous surface that is the reason why the conclusion is the same. You can see wide differences between the footprints created with lpdm-b_IBL above heterogeneous surface, but all these differences can be explained logically. Overall, the study concludes that the LS model lpdm-b_IBL should be preferred above heterogeneous surfaces for a more precise prediction of the footprint area. Concerning the incorporation of the backward model into the routine for characterization of heterogeneous surfaces, has to be said that for some individual cases it is very applicable, but increasing the number of differences to two areas causes a huge number of required pre-calculated lookup tables. For that reason it might be impracticable.