Name
Distributed modelling of the Canadian Prairie snowpack: model evaluation and demonstration
Description
Melt of the seasonal Canadian Prairie snowpack drives the majority of prairie runoff and is a critical soil water input necessary for dryland agricultural production. Over the winter the prairie snowpack is highly redistributed by blowing snow transport and ablated by sublimation and mid-winter melts so that it is spatially very heterogeneous. Spatial distributions are further influenced by stubble management, topography and the presence of wooded sloughs and shelterbelts. Snowcover depletion and areal melt rates during the spring freshet are controlled by this spatial distribution and have great influence on runoff and infiltration. Despite its importance and variability, there is a lack of detailed information on the current, or forecasted, state of prairie snowpacks. Here, the flexible, multi-physics, and modular, Canadian Hydrological Model (CHM) was forced by ECCC GEM forecasts and run over 32,000 km2 at resolutions down to 50 m to quantify prairie snowpack dynamics. CHM downscales GEM using radiation and wind fields to account for effects of topography and vegetation on snow model forcing. It includes physics-based blowing snow and energy balance processes in a high performance computing package that runs on an adaptive mesh. Detailed meteorological and snowpack observations collected at Global Water Futures Observatories sites near Saskatoon were used to diagnose model performance and identify improvements at field scales. The improvements to CHM will inform the expansion of the SnowCast product (snowcast.ca) to provide high resolution snow forecasts for the Canadian Prairies.