Assessment of the potential of CMIP6 models in simulating the sea surface temperature variability over the tropical Indian Ocean

The ability of Coupled Model Inter-comparison Project phase 6 (CMIP6) in simulating the seasonal and interannual sea surface temperature (SST) variability is evaluated over the tropical Indian Ocean (IO). The historical simulations of eleven CMIP6 models are compared with the observations and standard reference data (Extended Reconstructed SST from 1950 to 2014 and the RAMA buoy) for evaluation purpose. Present analysis shows that most of the models exhibit warm SST bias over the equatorial IO in all seasons whereas MPI-ESM1-2-HAM and MPI-ESM1-2 LR exhibit cold SST bias over the entire IO in all seasons. Out of the 11 models, MPI-ESM1-2-HR, CanESM5, and IPSL-CM6A-LR show less bias over the western Arabian Sea compared to other models suggesting their better ability to capture the upwelling phenomena during the summer monsoon season which mainly drives the phytoplankton bloom over this region. Statistical analysis reveals that all the models perform reasonably well over the Arabian Sea and southern IO. The CanESM5 and IPSL-CM6A-LR perform better over the Bay of Bengal compared to all other models. The analysis of the leading modes of climate variability over the tropical IO suggests that the models are able to represent the Indian Ocean Basin mode more realistically than the Indian Ocean Dipole mode. The entire tropical IO exhibits a warming trend of about 0.018 °C/year estimated from the observation and a comparison shows that although all the models simulate this trend, the rate of warming simulated by CESM2-WACCM, CanESM5, and IPSL-CM6A-LR is close to the observed value.