14:45 〜 15:00
[HCG26-05] Investigation of Uncertainties in Added values and Multi-variable Bias Adjustment of Drivers of Heat Stress
キーワード：Regional Climate Modelling, Multi-model Ensemble, Added Value, Multi-variable Bias Adjustment
Regional climate models are widely used to dynamically downscale general circulation models. Downscaled products, to some extent, provide a clearer understanding of surface-induced processes compared to the parent model. However, there are several uncertainties associated with downscaling. Some uncertainties are related to the structural differences in climate models, while some are related to the biases in global- and regional- climate models. Post-processing methods such as univariate bias correction are widely used to reduce the bias in the individual variable. However, in the context of compound events such as heat stress, multiple drivers of, surface air temperature and relative humidity play crucial roles. Thus, a multi-variable bias adjustment is necessary to retain the interdependence between the drivers and obtain reliable information on heat stress. The present study focuses on a Multi-variable Bias Adjustment method adapted from topography adjustment of surface-dependent surface air temperature and relative humidity obtained from a multi-model ensemble. Variations in the magnitude of bias before and after adjusting the variables are compared with high-resolution observed data. The efficiency of multi-variable bias adjustment is assessed for its sensitivity to the reference data used for the adjustment, and it shows some gains and some losses throughout the process. In addition, the added values of the regional climate model are also examined. Added values show pseudo nature over some regions after the bias adjustment. Overall, the bias adjustment shows improvement in the drivers over low-altitude urban regions, encouraging its application to assess heat stress.