Bias-corrected UKCP18 Convection-permitting model projections for England

Abstract. The UKCP18 Convection-Permitting Model (CPM) provides the latest high-resolution climate projections for the UK. Compared with regional climate model projections, the CPM projections are more capable of simulating small-scale atmospheric convection particularly during extreme weather events such as intense rainfall and localized storms. However, systematic biases still exist in these projections. To improve the reliability of these projections, bias correction is crucial. In this study, we applied a quantile mapping (QM) method to correct hourly precipitation and daily temperature for four selected ensemble members (EM01, EM04, EM07, EM08) of the UKCP18-CPM for England. The raw UKCP18-CPM simulations exhibit wet precipitation biases, particularly in northern England, with annual mean biases ranging from 4.6 % to 18.3 %, and cool temperature biases, with annual mean biases from −0.87 °C to 0.02 °C. Bias correction substantially improved agreement with observational datasets, increasing R² values for the 95th percentile of hourly precipitation from 0.80–0.88 to 0.98 and achieving near-perfect alignment (R² = 1) for temperature extremes. Future projections for the 2070s indicate notable increases in annual maximum precipitation by 25.1–39.1 % and mean daily temperature by 3.1 °C to 4.5 °C, highlighting the potential for more intense climate-related events. These results emphasize the effectiveness of bias correction in reducing model biases and improving the reliability of the CPM climate projections, thereby supporting more reliable future high-resolution climate and hydrological impact assessments in England.