Article,
Statistical Selection of the Optimum Models in the CMIP5 Dataset for Climate Change Projections of Indian Monsoon Rainfall
Climate, 3 (4): 858--875 (Nov 3, 2015)
DOI: 10.3390/cli3040858
Abstract
Monsoons are the life and soul of India's financial aspects, especially that of agribusiness in deciding cropping patterns. Around 80\% of the yearly precipitation occurs from June to September amid monsoon season across India. Thus, its seasonal mean precipitation is crucial for agriculture and the national water supply. From the start of the 19th century, several studies have been conducted on the possible increments in Indian summer monsoon precipitation in the future. Unfortunately, none of them has endeavoured to discover the models whose yield give the best fit to the observed data. Here some statistical tests are performed to quantify the models of Coupled Model Inter-comparison Project 5 (CMIP5). Then, after, the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method is used to select optimum models. It shows that four models, CCSM4, CESM1-CAM5, GFDL-CM3, and GFDL-ESM2G, best capture the pattern in Indian summer monsoon rainfall over the historical period (1871–2005). Further, Student's t-test is utilized to estimate the significant changes in meteorological subdivisions of selected optimum models. Also, our results reveal the Indian meteorological subdivisions which are liable to encounter significant changes in mean at confidence levels that differ from 80\% to 99\%.
