Micrometeorological and morphological observations of surface hoar dynamics on a mountain snow cover

[1] The formation, growth, and destruction of surface hoar crystals is an important feature of mountain snow covers as buried surface hoar layers are a frequent weak layer leading to unstable snowpacks. The energy and mass exchange associated with surface hoar dynamics is further an important part of land‐atmosphere interaction over snow. A quantitative prediction of surface hoar evolution based on local environmental conditions is, however, difficult. We carried out measurements of crystal hoar size and total surface mass changes in the period between January and March 2007 on the Weissfluhjoch study plot of the WSL Institute for Snow and Avalanche Research SLF, located above Davos, Switzerland, at 2540 m above sea level. For the first time, a direct comparison between eddy correlation measurements of latent heat flux and lysimeter‐like measurements of surface mass change has been made. Results show that the growth of surface hoar crystals is very well correlated with deposition of water vapor during clear‐sky nights as measured by two eddy correlation systems placed close to the ground. By analyzing local meteorological data, we confirm that low to moderate wind speed, humid air, and clear‐sky nights are the necessary ingredients for the occurrence of significant vapor fluxes toward the surface and thus for the growth of surface hoar. We also confirm that surface hoar crystals tend to preserve during daytime, when strong sublimation occurs, although their size significantly reduces. Despite the complexities associated with mountain terrain and snow surfaces, such as nonequilibrium boundary layers and stratification effects, the hoar formation could be predicted by the snow cover model SNOWPACK, which uses a bulk Monin‐Obukhov (MO) parameterization for the turbulent heat fluxes. On the basis of the comparison between direct observations and model predictions, we suggest that neutral stability conditions in the MO formulation provide the most stable and least flawed prediction for surface hoar formation.