Assimilation of Snow Properties Derived from Asar Wide Swath Data in a Hydrological Model of the Neckar Catchment for Improved Flood Forecast

Remote sensing methods to derive snow properties using both optical and SAR data are presented. The assimilation of the remote sensing products in the water balance and rainfall-runoff model LARSIM is demonstrated. The flood forecast centre in Karlsruhe applies LARSIM for the operational runoff forecast of the Neckar in the South West of Germany. A fully automatic algorithm to derive snow covered area from optical data was developed and is applied routinely since 2002. Information on snow cover is processed and provided within a few minutes after online reception of NOAA-AVHRR data. Processing consists of calibration, geometric correction and classification including an automatic cloud detection. The snow classification uses thresholds of multispectral indices within sensor and land use dependent decision trees. In order to derive snow properties from SAR, radiometric and geometric correction tools developed for ERSSAR were adapted and enhanced for ENVISAT’s ASAR sensor in wide swath mode. These tools make it possible to derive radiometric and geometric highprecision products from ENVISAT ASAR scenes. Melting snow (wet snow) is detected using an algorithm based on the image ratio technique. Products are provided near real time since summer 2004. Datasets from winter season 2002/2003 of the Neckar catchment (14.000 km2) were combined with modelled snow-cover and GIS information to demonstrate the innovation of multisensoral analyses of snow properties for operational hydrological modelling. Products from ASAR together with daily snow cover maps were integrated in the hydrological model LARSIM, which enables continuous spatially distributed simulations of all water balance terms in mesoscale catchments. The assimilation of the remotely sensed snowline in the hydrological simulation is already working automatically. First results further demonstrate the high potential of the wet snow classification from ENVISAT ASAR data. The WSM datasets allow the acquisition of surface parameters covering the entire area of the investigated Neckar catchment with sufficient spatial and fair temporal resolution. _____________________________________________________ Proc. of the 2004 Envisat & ERS Symposium, Salzburg, Austria 6-10 September 2004 (ESA SP-572, April 2005)