The Role of Daily and Monthly Bias Corrected Data in Preserving the Monthly Cross‐Correlation Between Precipitation and Temperature

ABSTRACT Bias correction and statistical downscaling of climatic variables from global climate models (GCMs) is vital in climate application studies. This study focuses on the ability of two bias correction procedures, canonical correlation analysis (CCA) and quantile regression (QR), in preserving the monthly cross‐correlation between precipitation and maximum temperature based on monthly bias corrected data and daily bias‐corrected precipitation and maximum temperature over the CONUS. The analyses show CCA reproduces observed cross‐correlation between precipitation and temperature better compared QR. The removal of the dry bias has also resulted in better performance by all the methods. Further, bias‐corrected data at daily time scale preserves the monthly cross‐correlation better compared to the bias‐corrected data available at monthly time scale. Our analysis shows that bias corrected daily time scale data should be aggregated to monthly time scale, even if the climate‐application studies require monthly forcings for developing sectoral (e.g., water, energy) impact analysis.