Multivariate Canadian Downscaled Climate Scenarios for CMIP6 (CanDCS- M6)

Multivariate Canadian Downscaled Climate Scenarios for CMIP6 (CanDCS- M6)

Stephen R. Sobie1 | Dhouha Ouali1 | Charles L. Curry1,2 | Francis W. Zwiers1,3

1Pacific Climate Impacts Consortium, University of Victoria, Victoria, British Columbia, Canada 2School of Earth and Ocean Sciences, University of Victoria, Victoria, British Columbia, Canada 3Nanjing University of Information Science and Technology, Nanjing, China

This study presents a new dataset, Multivariate Canadian Downscaled Climate Scenarios for CMIP6 (CanDCS-M6), which provides statistically downscaled simulations of global climate models from the Sixth Coupled Model Intercomparison Project (CMIP6). The authors developed a new multivariate downscaling method called N-dimensional Multivariate Bias Correction (MBCn) to improve the representation of compound climate events, which involve interactions between multiple climate variables. This dataset uses PCIC-Blend, a new calibration dataset that combines existing gridded observational datasets to provide a more accurate representation of precipitation and temperature conditions across Canada. The authors evaluated the performance of MBCn compared to other downscaling methods and found that MBCn outperforms other methods, particularly in capturing dependencies between variables, which is essential for simulating compound climate events. The CanDCS-M6 dataset provides daily simulations of precipitation, maximum temperature, and minimum temperature at a high resolution for 26 global climate models, covering the historical period (1950–2014) and three future Shared Socioeconomic Pathways (SSPs) representing different future emissions scenarios. This dataset is intended to facilitate climate impact assessments, hydrologic modelling, and analysis tools for presenting climate projections for Canada.