Classification and Analysis of Multifrequent and Multipolarimetric Airborne E-SAR Data for Hydrological Applications

Multifrequent and multipolarimetric airborne E-SAR data of a small subcatchment of the river Sieg was evaluated to derive land cover and soilmoisture information. The preprocessing included geometric and radiometric corrections. Land cover classifications were carried out in twosteps: multispectral discrimination of farm- and grasland, texture based distinction between settlements and forest areas. For thedetermination of the soil moisture, first a vegetation correction was applied, because the plant coverage induces volume scattering and is notcomparable to bare soil surfaces. A visual inspection of the soil moisture distribution in the test site showed a good correlation with the dryhilltops and the moist floodplain areas.