Using wind fields from a high-resolution atmospheric model for simulating snow dynamics in mountainous terrain

Wind-induced snow transport has remarkable effects on the snow cover spatial variability and on the temporal dynamics of snowmelt runoff. For accurate snow cover modelling, valid atmospheric forcing fields are essential. Since it is impossible to generate appropriate wind fields by a simple spatial interpolation of station data. a new approach was developed: A modified version of the Penn State University-National Centre for Atmospheric Research (MM5) model was used to generate wind fields with 200-m resolution. Because of the high computational costs of MM5, it was not practicable to include the wind field generation as an operational part of the snow cover modelling. Therefore, an archive consisting of 220 wind fields was generated prior to the simulation of the snow transport processes. These fields represent the most relevant synoptic situations for wind-induced snow transport occurring at our test site. The criteria to generate which wind field to use in a specific snow model time step are mean wind speed and direction at the 700 hPa level derived from German Weather Service/Deutscher Wetterdienst (DWD) Lokalmodell (LM) analysis data. The wind field library provided physically derived wind speed and direction fields that were used to drive a snow transport model (SnowTran-3D) in a high Alpine area in the Berchtesgaden National Park, Germany. In complex Alpine terrain, the procedure provides an alternative to simple interpolation methods, yielding improved physical realism at reasonable computational expense. Copyright (C) 2008 John Wiley & Sons, Ltd.