Long-term simulation of daily rainfall across India: Performance of cumulus convection schemes in regional climate model during southwest and northeast monsoon

The performance of Cumulus Convection Schemes (CCSs) in Regional Climate Model (RCM) is investigated for the long-term (34 years) simulation of daily rainfall over different Homogeneous Monsoon Regions (HMRs) of India. Characteristics of both southwest monsoon (aka summer monsoon) and northeast monsoon (aka winter monsoon or return monsoon) are analysed while assessing the performance. Indian monsoon is largely affected by climate change and studies demand a long-term (climatological scale) simulation and analysis. Furthermore, the vast study area consists of wide variation of topography and climatology that cause variation in best performing CCS across the HMRs. Month-wise and season-wise performances of different CCSs, namely Grell, and its combinations with Emmanuel as mixed schemes, such as Grell over Ocean and Emmanuel over Land (GOEL), and Grell over Land and Emmanuel over Ocean (GLEO) are assessed by using these schemes in the Regional Climate Model Version 4.4 (RegCM4.4) over a span of 34 years (1982 to 2015). The spatio-temporal variation of bias, investigated using the observed and simulated rainfall, shows more realistic performance for the peak monsoon months and positive bias during onset and retreating monsoon. Additionally, the spatio-temporal variation over the HMRs is investigated and the best possible CCS or a combination of CCSs is recommended. The better ability of mixed schemes in simulating the summer monsoon is noticed, particularly late monsoon, as compared to the early monsoon. Furthermore, GOEL performs best as compared to others to capture the spatial variation including peninsular India. The spatial distribution of observed rainfall during return monsoon is also well captured by the mixed schemes. However, GLEO scheme is found to be the best during return monsoon.