A two-stage bias correction approach for downscaling and projection of daily average temperature

Abstract

Reliable projection of climate is essential for climate change impact assessment and mitigation planning. General Circulation Models (GCMs) simulations are generally downscaled into much finer spatial resolution for climate change impact studies at local and regional scales. The objective of the present study is to use a two-stage bias correction approach for downscale and project future changes of daily average temperature. The approach was applied for downscaling and projection of daily average temperature of Senai meteorological station located in Johor Bahru, Malaysia using a GCM of Coupled Model Intercomparison Project Phase 5 (CMIP5) under four representative concentration pathways (RCP) scenarios. The two-stage bias correction method was based on correction in mean factor and variability inflation factor. The model performances were assessed using different statistical measures including mean bias error (MBE), mean absolute error (MAE), root mean square error (RMSE), index of agreement (MD), Nash–Sutcliffe model efficiency (NSE) and coefficient of determination (R2). Results showed that the downscaling method could simulate historical daily average temperature at the station very well. The GCM projected an increase in daily average temperature by 1.4ºC, 2.2ºC, 2.5ºC, and 3.4ºC under RCP2.6, RCP4.5, RCP6.0 and RCP8.5 scenarios, respectively in the end of this century. It is expected that the finding of the study would help in climate change impact assessment and adopting necessary adaptation measures.