Examining Daily Snow Depths at the Catchment Scale in Canada Using CMIP6

The lack of reliable data on daily snow depth (SD) is a significant challenge for studying water systems, ecology, and resources. Climate models present a potential solution for generating daily SD data, but the literature has not thoroughly explored how accurately they simulate this data. This study investigates the capabilities of CMIP6 climate models to replicate daily SD characteristics in eleven major Canadian catchments. The results depict that CMIP6 simulations overestimate the average SD values by a median of 57.7% (6.9 cm). In the Arctic and Pacific regions, this overestimation becomes particularly pronounced. However, the simulations align more closely with observations in smaller catchments with homogenous land characteristics. This finding suggests a shortcoming in how these models simulate different land types within the grid. Additionally, the models appear to overestimate the snow cover duration, with a median underestimation of 30.5 days. This overestimation could be due to the models failing to accurately account for the rates at which snow accumulates and melts away. However, the models perform relatively well when predicting extreme SD conditions. This study carries valuable implications for refining the outputs of climate models and effectively utilizing them in impact studies.