Assessing the quality of a real-time Snow Cover Area product for hydrological applications

Using reliable observed data is important for performing real-time flood forecasts or hydrological simulation in order to calibrate parameters or to update model variables. Satellite snow products can be one of these data, since snow is a water reservoir with a high impact on the quality of discharge simulation. The satellite Snow Cover Area products are known to be of good quality and are regularly used for studies in the meteorological and hydrological fields. However, these products are reprocessed and thus are not representative of the quality of a real-time product that would be needed for operational applications. The assessment of a real-time Snow Cover Area daily product at 250 m-resolution (the EURAC MODIS SCA product, which is based on the MODIS sensor) by comparing it directly or indirectly to the classical NASA MODIS SCA daily product and to the simulated Snow Cover Area of the distributed hydrological model, LISFLOOD, is realized in this article at the pan-European scale. This real-time product showed an overall good performance compared with the classical product, and a good agreement with the LISFLOOD simulated snow over Europe. The study showed the impact of forest cover on the scores of the compared products, whereas altitude did not have an impact. Using quality flags that are provided with the EURAC product improved its performance by reducing the misclassification of clouds as snow.

► We compared two spatial Snow Cover Area (SCA) datasets and a hydrological model SCA
► We assessed the differences of datasets on several European River basins
► Forest cover is the parameter having the highest impact on differences of datasets
► Real-time SCA data can reach a quality similar to a reprocessed dataset
► Real-time SCA data are of high interest for hydrological use