An analytical study of hindcasts from general circulation models for Indian summer monsoon rainfall

• Published in: Meteorological applications Vol. 21; no. 3; pp. 695 - 707
• Main Authors: Nair, Archana, Mohanty, U. C., Robertson, Andrew W., Panda, T. C., Luo, Jing‐Jia, Yamagata, Toshio
• Format: Journal Article
• Language: English
• Published: Chichester, UK John Wiley & Sons, Ltd 01.07.2014; John Wiley & Sons, Inc
• Subjects: empirical orthogonal function; General circulation models; Indian summer monsoon rainfall; Meteorology; Precipitation; predictability; spectral analysis; variability; Wind; ISW, Indian Ocean; ISW, India; Japan; INW, Japan
• ISSN: 1350-4827; 1469-8080
• DOI: 10.1002/met.1395

ABSTRACT In this paper, precipitation outputs from retrospective seasonal forecasts made by nine General Circulation Models (GCMs) are used to investigate historical Indian summer monsoon seasonal rainfall variability and predictability over India. The observed data is taken from the India Meteorological Department whereas GCMs are obtained from the International Research Institute for Climate and Society, Columbia University, the National Center for Environmental Prediction, and the Japan Agency for Marine Earth Science and Technology. The study focusses on June–September precipitation hindcasts initialized from the 1 May. First, the mean climatology, variance of interannual variability (IAV), and long‐term trends for the nine GCMs were evaluated. Then Empirical Orthogonal Function (EOF) is used to extract major climate modes and spectral analyses method is used to investigate the temporal properties of the leading principal components. It is found that the models are able to reproduce the climatology and IAV to varying degrees. The EOF and spectral analyses of models hindcast reveal that these models are capable of predicting the observed precipitation variability to some extent. In order to study the remote response, the correlation co‐efficient between model predicted rainfall and sea surface temperature (SST) have been calculated. The results suggest that the models show exaggerated remote response to ENSO SST forcing and the Indian Ocean Dipole Mode index has less predictive skill compared to ENSO. The correlation values between the model predicted Monsoon Hadley Index (MHI) and observed MHI reveals that only a few of them could exhibit large scale circulation features well.