Assessing Cloud Fraction in the Canadian Regional Climate Model Over North America Using Satellite Data and a Satellite Simulator Package

Abstract Clouds are crucial to Earth's climate system, influencing radiation and contributing to climate projection uncertainties. Here, the simulated cloud fraction by the sixth version of the Canadian Regional Climate Model (CRCM6‐GEM5) was evaluated using CALIPSO lidar retrievals and the second version of the Cloud Feedback Intercomparison Project (CFMIP) Observation Simulator Package (COSP2) for the years 2014 and 2015. Horizontal and vertical distributions of clouds in the CRCM6‐GEM5 model were evaluated using cloud profiles and four cloud categories (total, high‐, mid‐, and low‐level clouds) derived directly from the CRCM6‐GEM5 model and treated using the COSP2 satellite simulator. A seasonal analysis was conducted across specific regions in North America. Results showed that the use of COSP2 is essential for comparing CRCM6‐GEM5 outputs against satellite data to account for variable definitions and signal attenuation of active instruments (e.g., Cloud‐Aerosol Lidar with Orthogonal Polarization: CALIOP). Spatial and vertical cloud distributions and seasonal patterns were generally well represented by the CRCM6‐GEM5 for both winter (December–February) and summer (June–August). High‐ and low‐level clouds were particularly well‐represented, especially in winter. The CRCM6‐GEM5 model demonstrated some difficulty producing enough clouds to accurately represent those at mid‐level. Cloud fraction representation was systematically better during winter than summer. The CRCM6‐GEM5 generally performed well over the whole North American domain for the four cloud categories and COSP2 was confirmed to help mitigate discrepancies in variable definitions. These results contribute to a better understanding of the CRCM6‐GEM5 cloud representations and the use of COSP2 with high‐resolution models.