In addition, several new tools will be used to adapt and refine predictions near and at the surface. Very high-resolution two-dimensional surface systems, with 100-m and 20-m mesh size, will cover the Vancouver Olympics area. Using adaptation methods to improve the forcing from the lower-resolution atmospheric models, these 2D surface models better represent surface processes. This leads to improved predictions of snow conditions and near-surface air temperature. Based on a similar strategy, a single-point model will be implemented to better forecast surface conditions at each station of an observing network specially installed for the 2010 events. The main advantage of this single-point system is that surface observations are used as forcing for the land surface models and can even be assimilated to improve initial conditions of surface variables such as snow depth and surface temperatures. Another adaptation tool, based on 2D stationary solutions of a simple dynamical system, will be used to produce near-surface winds on the 100-m grid, corrected for the high-resolution orography. The configuration of the experimental numerical prediction system will be presented along with preliminary results for the 2008 winter.