Statistical Downscaling of Maximum Temperature in Hoshangabad District of India

Abstract

The Global Climate ModelsCanESM2 and CGCM3 were utilised to downscale the maximum temperature for Hoshangabad district of Madhya Pradesh, India. The area of study comprises to be of 6704 km2. The predictors employed for CanESM2 were ncepmslpgl, ncepp500gl, ncepp850gl and ncepmslpas, ncepp500gl, ncepp850gl were the predictors fixed for CGCM3. The total duration of the study was from the years 1979 – 2001. The two GCMs, CGCM3 and CanESM2 were checked for their capability in downscaling the maximum temperature climatic parameter. The GCM outputs were evaluated on Nash Sutcliffe Efficiency (NSE) and coefficient of determination (r2) criterias. The period of calibration was taken to be 1979-1995 and 1996-2001 was chosen as the period of validation. GCM CanESM2 obtained NSE of 0.77, 0.75 and r2 of 0.79, 0.79 during the periods of calibration and validation respectively. It was concluded that CanESM2 model is found comparatively more suitable for downscaling of maximum temperature for Hoshangabad region. The GCM can be further employed to generate the future scenario of maximum temperature in the region.