Dense Temporal Series of C- and L-band SAR Data for Soil Moisture Retrieval Over Agricultural Crops

This paper investigates the potential of multi-temporal C- and L-band SAR data, acquired within a short revisiting time (1-2 weeks), to map temporal changes of surface soil moisture content (m <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">v</sub> ) underneath agricultural crops. The analysed data consist of a new ground and SAR data set acquired on a weekly basis from late April to early August 2006 over the DEMMIN (Durable Environmental Multidisciplinary Monitoring Information Network) agricultural site (Northern Germany) during the European Space Agency 2006 AgriSAR campaign. The paper firstly investigates the main scattering mechanisms characterizing the interaction between the SAR signal and crops, such as winter wheat and rape. Then, the relationship between backscatter and soil moisture content temporal changes as a function of different SAR bands and polarizations is studied. Observations indicate that rationing of the multi-temporal radar backscatter can be a simple and effective way to decouple the effect of vegetation and surface roughness from the effect of soil moisture changes, when volume scattering is not dominant. The study also assesses to which extent changes in the incidence angle between subsequent radar acquisitions may affect the radar sensitivity to soil moisture content. Finally, an algorithm based on the change detection technique retrieving superficial soil moisture content is proposed and assessed both on simulated and experimental data. Results indicate that for crops relatively insensitive to volume scattering in the vegetation canopy (as for instance winter wheat at C-band or winter rape and winter wheat at L-band), m <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">v</sub> can be retrieved during the whole growing season, with accuracies ranging between 5% and 6% [m <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> /m <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> ]. We also show that low incidence angles (e.g., 20-35 ) and HH polarization are generally better suited to m <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">v</sub> retrieval than VV polarization and higher incidence angles.