Comparative analysis of bias correction techniques for future climate assessment using CMIP6 hydrological variables for the Indian subcontinent

Abstract

The study focuses on the bias correction of Coupled Model Intercomparison Project Phase 6 (CMIP6) hydrologic variables for the Indian region. The performance of two widely accepted bias correction methodologies, namely Quantile Mapping (QM) and Bias Correction Spatial Disaggregation (BCSD), is compared. The study undertakes to evaluate the application of these popular bias correction methodologies on four important hydrologic variables viz. precipitation, temperature, and surface wind. The QM methodology is employed and compared with BCSD based bias corrected variables obtained from NEX-GDDP-CMIP6 dataset. The selected GCM historical bias corrected climate variables using QM are compared with the NCEP reanalysis variables. The objective is to improve the reliability and accuracy of climate projections by minimizing biases present in the GCM outputs. Through a comprehensive comparative analysis, it is determined that QM exhibits superior performance in reducing biases when compared to BCSD. Thus, use of QM demonstrates higher efficacy by effectively capturing the statistical distribution characteristics of observed data and transferring them to the GCM outputs. The future climate change over the Indian region is observed for both QM and BCSD algorithms for SSP5-8.5, SSP2-4.5, and SSP1-2.6. The result emphasizes the importance of selecting an appropriate bias correction methodology to enhance the reliability of climate projections in the Indian region. Ultimately, the findings of this study contribute to the broader field of climate modeling and impact assessment, providing valuable insights into the selection and application of bias correction techniques for CMIP6 datasets in the Indian subcontinent region.

Graphical abstract