Assimilation of short wavelength satellite observations into an agrometeorological model

One important issue of agro-economy is the estimate of crop productivity. Process models have been developed that simulate the growths and biomass production of the crops. These models are not perfect and need adjustments; they also need initialisation. The paper presents a method to use satellite data for this purpose. First the growth model is coupled with a radiative transfer model via the leaf area index. The radiative transfer model allows to compute the reflectance of the crop. The interannual variations of the measured radiometric signal were simulated. By comparing the simulated reflectance in the visible and near infrared spectral bands with measurements made in space the process model can be adjusted in its temporal development. One example is presenting how this adjustment works with high resolution SPOT data for a uniform field. Often the high resolution satellite data are not available at times that are most critical for the adjustment procedure. In these cases daily NOAA-AVHRR data could serve. But the signal of one pixel often contains a mixture of vegetation types. An example shows how the annual course of the measured NDVI compares to the NDVI simulated by the combined process and reflectance models