Assimilation of MODIS snow cover area data in a distributed hydrological model

Abstract. Snow is an important component of the water cycle and its estimation in hydrological models is of great importance concerning snow melting flood events simulations and forecasting. The LISFLOOD model is a spatially distributed hydrological model designed at the Joint Research Centre for large European river basins. It is used for a variety of applications including flood forecasting and assessing the effects of land use change and climate change. In order to improve the streamflow simulations of this model, especially with respect to snowmelt induced floods, the assimilation of Snow Cover Area (SCA) has been evaluated in this study. For this purpose daily 420 m-resolution MODIS satellital SCA data have been used, which were then converted in Snow Water Equivalent (SWE) using a Snow Depletion Curve. Tests were performed over the Morava basin, a tributary of the Danube, for a period of almost three years. Two data assimilation techniques, the Ensemble Kalman Filter (EnKF) and the particle filter, were compared, for assimilating the MODIS composites of SCA every seven days. Two approaches were tested, in which the SWE of the model was adjusted either using three altitudinal-based zones or seven sub-basins-based zones. These experiments showed the improvement of the SWE of the model when compared with MODIS-derived snow for both the EnKF and the particle filter. However, on average only the particle filter improved the discharge simulation, because the EnKF imposed too important water balance modifications, which deteriorated the simulation of the discharges during the snow melt periods.