Einfluss unterschiedlicher Strahlungsparametrisierungen in Meereissimulationen der Arktis
In this work, the influence of different parameterizations of solar surface radiation on sea ice simulations of the Arctic region has been examined. In this case, the oceanic heat flux and the ocean drift were defaulted as forcing. The ECMWF analyses of the years 1986-1992 were used as atmospheric f...
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | German |
| Published: |
1999
|
| Subjects: | |
| Online Access: | https://oceanrep.geomar.de/id/eprint/54208/ https://oceanrep.geomar.de/id/eprint/54208/1/Dipl.%201999%20Frerichs,W.pdf |
| Summary: | In this work, the influence of different parameterizations of solar surface radiation on sea ice simulations of the Arctic region has been examined. In this case, the oceanic heat flux and the ocean drift were defaulted as forcing. The ECMWF analyses of the years 1986-1992 were used as atmospheric forcing. A special attention was directed towards the clouds as forcing. Both, a climatology of clouds and the ISCCP D2 record have been available for this study. The effect of the two cloud data records on the computation of the solar surface radiation is in the same order of magnitude as the differences between the individual radiation parameterizations. Since the ISCCP D2 record is available for a longer period the study of the influence of the radiation parameterizations on sea ice simulations has been performed with this data set. The parameterizations of the solar surface radiation, originating from investigations of Bennett (1982), Shine (1984a) and Zillman (1972), differ mainly in the required number of input variables. The neural network FluxNet was used as a supplementary radiation parameterization. The results of the radiation computations of all radiation parameterizations were compared with the results of the radiation transport model Streamer. The radiation parameterizations show a very different behavior. The differences between the individual computations are up to 100 wm-2 and up to 60 wm-2 to Streamer. The radiation parameterization of Shine (1984a) shows a very good similarity with the computations of Streamer. FluxN et in turn does not show its close affinity to Streamer. The similarity of FluxN et with Streamer compared to the other radiation parameterizations is not as high as expected. The results from radiation computation can be seen in the melting rates for the individual sea ice simulations. The radiation parameterizations with the highest values of the solar surface radiation also supply the highest melting rates, the radiation parameterizations with the lowest values the smallest melting ... |
|---|