Dynamical downscaling with COSMO and COSMO-CLM in the Sino-Mongolian Altai region

作者：Kurzrock Frederik; Buerkert Andreas; Byambaa Oyunmunkh; Goenster Sven; Jin Luxi; Ohlwein Christian; Simmer Clemens^*; Simon Thorsten

来源：Meteorology and Atmospheric Physics, 2017, 129(2): 211-228.

DOI：10.1007/s00703-016-0487-4

For the first time, regional atmospheric simulations with spatial resolution down to 6 km have been performed in the Sino-Mongolian Altai region using the COSMO weather forecast and regional climate model. Two 5-year periods (1979-1982 and 2008-2012) have been simulated for a first evaluation of the model in this special region. The added value of a dynamical downscaling with the COSMO regional climate model CCLM towards the driving ERA-Interim reanalysis is investigated by comparison with weather station observations. In the mountainous region, the CCLM simulation much better relates to the observed monthly mean 2 m temperature and maximum monthly precipitation sums in summer than ERA-Interim. In addition, the intensity distribution of sub-daily precipitation amounts becomes more realistic with increasing altitude. CCLM does, however, overestimate convection in the mountains and accordingly simulates too much precipitation. Moreover, wintertime near-surface temperature inversions are underrated in the southern near-Gobi area, which leads to too high 2 m temperatures in that region. To examine the ability of the COSMO model to reproduce the vertical thermodynamic structure of the troposphere, additional simulations with the weather forecast version of COSMO were performed for July 2013 and compared to radiosonde measurements of the WATERCOPE field experiment in this region. The results indicate that the COSMO model is quite capable of qualitatively simulating a range of features of the local tropospheric stratification. Mean differences between observed and simulated dew point and temperature profiles were in the range of only to 1-2 A degrees C in the lower troposphere.

• 出版日期2017-4