Climate change projection and drought characterization in Bangladesh

Abstract

One of the biggest threats of the climatic change is aberrant pattern or distribution of rainfall that results to drought. The main objective of this research was to develop a methodological framework to assess the impacts of climate change on seasonal drought characteristics with uncertainty. Bangladesh, one of the most vulnerable countries in the world to climate change was considered as the study area for implementation of the framework. An ensemble of general circulation models (GCMs) of Coupled Model Intercomparison Project phase 5 (CMIP5) were used for downscaling and projection of rainfall and temperature under different Representative Concentration Pathways (RCP) scenarios. Two state of art data mining approaches known as Random Forest (RF) and Support Vector Machine (SVM) were used for the development of downscaling models and Quantile Mapping (QM) approach was used to remove biases in GCMs. The observed and future projected rainfall data were used to characterize the seasonal droughts using Severity-Area-Frequency (SAF) curves developed for different climatic and major crop growing seasons. The results revealed superior performance of SVM in downscaling rainfall and temperature in tropical climate in terms of all standard statistics. Downscaling of CMIP5 GCMs projections revealed a change in annual average rainfall in Bangladesh in the range of -8.6% in the northeast to +11.9% in the northwest, which indicates that spatial distribution of rainfall of Bangladesh will be more homogeneous in future. The maximum and minimum temperatures of Bangladesh were projected to increase in the range of 0.8 to 4.3ºC and 1.0 to 4.8ºC, respectively under different RCPs. Future projection of droughts revealed that affected areas will increase for higher severity and higher return period droughts. Overall, the country will be more affected by higher return period Kharif (May- October) and monsoon droughts, and lower return period pre-monsoon and postmonsoon droughts due to climate change.