Temperature change scenarios over the Chilika Lagoon of India for 200 years (1901-2100) were quantified by the observational data sets of the Climate Research Units (CRU) of UK as well as 39 numbers of simulations from the Couple Model Inter-comparison Project Phase 5 (CMIP5) through Mann Kendall trends analysis. Long-term trend during 1901-2005 over Chilka Lake indicates the highest warming in the pre-monsoon season (1.79°C) and lowest warming (1.09°C) was shown in both the winter and post-monsoon seasons while opposite warming trends i.e., the lowest warming was observed in pre-monsoon season and highest warming was shown in the winter season for recent four decades (1969-2009) data. The performance of the CMIP5 was evaluated over a target point of Chilika Lake. Twelve numbers of models were considered as a group of performing GCMs on the basis of their ability to simulate the long-term trends as well as the mean seasonal correlation with observation. Quantile mapping technique is used for adjusting the bias for the selected GCMs. Improvement in the multi-model ensemble (MME) of bias corrected better performing models compared to MME of 39 was judged with the help of Taylor plot as well as using four different conventional statistical indices viz. correlation (r), index of agreement (d index), Nash-Sutcliffe efficiency (NSE) and root mean square error (RMSE). All the ensemble members commonly available in the four Representative Concentrations Pathways (RCPs) from the better performing selected models show a temperature change of 0.27-3.61°C, 0.38-3.98°C, 0.28-3.72°C, 0.33-3.41°C and 0.22-2.50°C in annual, winter, pre-monsoon, monsoon and post-monsoon seasons respectively at the end of 21st century over Chilika Lagoon.
Read full abstract