15:30 〜 15:50
[HCG24-01] Statistical Downscaling of CMIP6 GCMs for the Southeast Greater Mekong Subregion Towards Strengthening Local Climate Impact Assessments
キーワード：statistical downscaling, Southeast Greater Mekong Subregion, extreme indices, local adaptation
High-resolution climate projections are essential for directly assessing localized climate change impacts. However, raw Global Climate Model (GCM) outputs often lack the precision needed for regional adaptation due to their coarse resolution (~hundreds of kilometers). This study applies the Bias Correction Constructed Analogues with Quantile Mapping Reordering (BCCAQ) hybrid downscaling technique to five CMIP6 GCMs (EC-Earth3-Veg, EC-Earth3, NorESM2-MM, MRI-ESM2-0, IPSL-CM6A-LR), selected based on the DN22 performance ranking, over the Southeast Greater Mekong Subregion (SGMS), with daily observational data from the Multi-Source Weather (MSWX)-GloH2O dataset (0.1° resolution, 1981–2014) serving as the reference. Results indicate that BCCAQ significantly improves the accuracy of temperature, precipitation, and relative humidity simulations compared to raw GCM outputs. Performance metrics demonstrate consistent improvements across all validation stations, including bias, index of agreement (IOA), normalized root mean square error (NRMSE), and correlation coefficients. Notably, minimum temperature exhibits a strong spatial correlation (>0.95), while maximum temperature is better represented along Vietnam's central coastal region. BCCAQ effectively reduces relative humidity biases to approximately zero during the JJASO season. The method performs exceptionally well in topographically complex regions, such as Central Vietnam’s mountainous regions and Cambodia’s Phnom Kravanh, and enhances the representation of extreme climate indices, including the Heat Wave Magnitude Index daily (HWMId) and heavy precipitation (R95p), which are vital for disaster risk planning. This study represents a comprehensive application of advanced hybrid statistical downscaling in the SGMS, addressing a critical regional-to-local climate data gap through this approach's contribution through added values. These findings highlight the necessity of high-resolution climate modeling to support sectoral adaptation policies, particularly in mitigating extreme events, managing heat stress, and planning for highly vulnerable regions.