On the simulation of northeast monsoon rainfall over southern peninsular India in CMIP5 and CMIP6 models

P. P. Sreekala,C. A. Babu,S. Vijaya Bhaskara Rao
DOI: https://doi.org/10.1007/s00704-022-04194-2
2022-09-21
Theorectical and Applied Climatology
Abstract:Northeast monsoon rainfall (Oct–Dec) over southern peninsular India (NEMR-SP) has the high coefficient of variation (26%) and low predictability. The poor skill of the ensemble models in simulating the interannual variability of northeast monsoon rainfall over southern peninsular India is well known. Here, we tried to understand the possible reasons behind the poor skill of CMIP5 (31 models) and CMIP6 (15 model) models in capturing the mean and interannual variability of northeast monsoon rainfall over southern peninsular India. Varying skills are observed in simulating the climatological mean (1979–2005) rainfall for different CMIP5 and CMIP6 models. Pattern correlation coefficient (PCC) of NEMR over a broader region (10° S–30° N, 40° E–120° E) is found to be ranging from 0.6 to 0.93 and RMSE is ranging between 1.73 and 3.83 for CMIP5 models, while PCC is ranging from 0.57 to 0.87 and RMSE is ranging from 0.38 to 1.9 for CMIP6 models. The mean rainfall is overestimated over the southwestern Indian Ocean and underestimated over southern peninsular India in both CMIP model ensembles. The warm bias in sea surface temperature (SST) over the western equatorial Indian Ocean and the easterly wind bias over equatorial Indian Ocean are the other biases observed in the mean climatology of CMIP model ensembles. Around 70% (more than 75%) of the selected models were not able to reproduce the observed positive correlation between Niño3.4 SST and NEMR-SP as well as dipole mode index and NEMR-SP in CMIP5 models (CMIP6 models). El Niño and positive phase of Indian Ocean Dipole (PIOD)–related positive rainfall anomalies over the western Indian Ocean and southern peninsular India are underestimated in both CMIP5 and CMIP6 model ensembles. The El Niño and PIOD–related warm SST anomalies over cold tongue region and associated divergent anomalies over warm pool region are found to be shifted westward in the model ensembles. The southern peninsular India is found to be under the influence of these divergent anomalies in the CMIP models, which is contradictory to the observation. The observed southerly component of wind at 850 hPa over the equatorial Indian Ocean, the upper level (200 hPa) anticyclonic circulation and the convergent anomalies over the western Indian Ocean during El Niño and PIOD are not simulated correctly in the CMIP models. All these biases may contribute to the poor skill in the simulation of teleconnection between NEMR-SP and El Niño as well as NEMR-SP and PIOD.
What problem does this paper attempt to address?