Monitoring land surfaces with combined DMSP-SSM/I and ERS-1 scatterometer data

Microwave data from the DMSP-SSM/I sensor and the ERS-1 Scatterometer are used to monitor the temporal behaviour of deserts, forests and snow-covered areas. The SSM/I brightness temperatures were spatially interpolated using the algorithm proposed by Poe (1990). In order to detect the different land surface types according to their spectral signature, the surface emissivity has to be derived from the satellite brightness temperatures. This requires the knowledge of the physical temperature of the Earth surface. The authors´ approach for determining the physical surface temperature consists of a linear combination of SSM/I 19 GHz brightness temperatures. Ground measurements of surface temperature are in good agreement with the estimated physical temperature (rms 2.63 K). The agreement is worse for temperatures below 273.15 K. The estimated physical temperatures are used to derive the emissivity of selected land surface types (deserts, forests) from SSM/I measurements. While dry deserts and tropical rain forests showed, as expected, virtually no seasonal variations, large variations in emissivity at 37 GHz for the Siberian forest regions were noticed. Together with the scatterometer data this observation can be interpreted as melting snow within the observed area