Analysis of the Indian summer monsoon system in the regional climate model COSMO-CLM

The Indian summer monsoon (ISM) influences daily lives and economies in many countries in the South Asian region. This study analyzes the representation of the ISM system in the regional climate model COSMO-CLM. Simulations driven by ERA-40 reanalysis and present-day (1960-2000) data from the global climate model ECHAM5 are investigated. The ability of COSMO-CLM to reproduce the ISM better than the coarser-grid driving models is tested using a set of well-established, complementary monsoon indices: the all-India monsoon rainfall, vertical wind shear indices, and an outgoing longwave radiation (OLR) index. The results show that regarding these large-scale indices the COSMO-CLM simulations are not more accurate than the driving models. Considering the spatial distribution of rainfall, the ERA-40-driven COSMO-CLM simulations show major overestimations (about 100%) for the west coast of India and underestimations (about 50%) for the Himalayan foothills. Large biases occur in the OLR data over the Arabian Sea and the Bay of Bengal where COSMO-CLM shows high convective activity (OLR < 180 W m(-2)) at about 3 times as many days as observed in the monsoon season. In the ECHAM5-driven simulation, underestimations of rainfall also appear at the Himalayan foothills. Nevertheless, the application of COSMO-CLM to ECHAM5 improves the temporal correlations of the modeled ISM indices, and the spatial patterns are better simulated in COSMO-CLM with 0.44 degrees horizontal grid spacing than in the large-scale ECHAM5 data.