Skip to Main Content
The validation of coarse-scale remote sensing products like SMOS (ESA's Soil Moisture and Ocean Salinity mission) L2 soil moisture or L1c brightness temperature data requires the maintenance of long-term soil moisture monitoring sites like the Upper Danube Catchment SMOS validation site situated in Southern Germany. An automatic framework has been built up to compare SMOS data against in situ measurements, land surface model simulations, and ancillary satellite data. The uncertainties of the different data sets used for SMOS validation are being assessed in this paper by comparing different microwave radiative transfer and land surface model results to measured soil moisture and brightness temperature data from local scale to SMOS scale. The mean observed uncertainties of the modeled soil moisture decrease from 0.094 m3 m-3 on the local scale to 0.040 m3 m-3 root mean squared error (RMSE) on the large scale. The RMSE of anomalies is 0.023 m3 m-3 on the large scale. The mean R2 increases from 0.6 on the local scale to 0.75 on the medium scale. The land surface model tends to underestimate soil moisture under wet conditions and has a smaller dynamical range than the measurements. The brightness temperature comparison leads to a RMSE around 12-16 K between microwave radiative transfer model and airborne measurements under varying soil moisture and vegetation conditions. The assessed data sets are considered reliable and robust enough to be able to provide a valuable contribution to SMOS validation activities.