An evaluation of snowmelt and sublimation over short vegetation in land surface modelling

Sublimation and snowmelt in a short brush canopy is examined with a two-layer canopy model. The model's performance is evaluated with observations taken over a brush site in the North Park basin of Colorado, USA. Two cases are selected for detailed examination. In the first case, a thin snow layer covers the ground surface but the canopy is snow free. In the second case, a thicker layer of snow covers both the ground surface and part of the brush canopy. Sensitivity tests are performed to investigate the effect of the subcanopy mixing parameterization and the formulation of intercepted snow on the sublimation and snowmelt. In short canopies, snowmelt is sensitive to the subcanopy mixing parameterization, and turbulent heat flux is sensitive to the intercepted snow due to modification of the albedo and water availability. Comparisons are made with additional eddy-correlation measurements over a snow-covered surface with no vegetation. Greater roughness and heating at the sparse canopy site lead to faster snow ablation compared with the site with no vegetation.