You are here: EO-DAHR » The project


The hydrological ensemble prediction system of the RMI consists of the forcing of a hydrological model with ensemble predictions of atmospheric variables, e.g. from the European Centre for Medium Range Weather Forecasts (ECMWF), and provides an ensemble of possible discharge series for the next days to estimate the risks of floods and droughts in the main Belgian catchments. Due to errors in initial conditions, forcing variables, hydrological model structure and parameter values, predictions have random errors, systematic biases and lack reliability. These deficiencies can be reduced by using data assimilation and post-processing. Data assimilation combines in the best-possible way the model with intermittent observations and estimates the model initial conditions, parameters and associated uncertainties.

The assimilation of satellite data related to soil moisture and vegetation has long held potential to improve forecasts of weather, floods, fires, nutrient fluxes, and more, but the reported applications have so far been limited. At this time, with the deployed Earth observation systems and their planned next generation, more satellite data meet the requirements for operational use and land assimilation systems have reached enough maturity to adjust and utilize them for operational applications. The EODAHR project specifically aims at fuelling the social benefit of improved early warnings for floods and droughts.