![Sample our Environment & Agriculture journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5934/TOC-Subject-Sample-2021-Environment-Agriculture.jpg"})
Abstract
A method for producing soil moisture maps in mountainous areas by using Environment Satellite Advanced Synthetic Aperture Radar (ENVISAT/ASAR) images at C-band is described in this paper. For this purpose, experimental campaigns were carried out in 2004 in the Cordevole watershed in Italy during ENVISAT passes. Ground truth measurements of soil and vegetation parameters were obtained simultaneously using satellite surveys. A preliminary classification of the area was carried out to mask those zones in which soil moisture measurement was unobtainable. The performance of an inversion algorithm, based on artificial neural networks (ANNs), in retrieving soil moisture content (SMC) from the collected images was then tested and compared with ground measurements. The results obtained on a restricted portion of the watershed show reasonable agreement of backscattering (σ0) with ground truth data and meteorological conditions, thus making it possible to extend the algorithm to the entire test area. The contribution of vegetation cover was then simulated by using a discrete elements model based on radiative transfer theory. Three pixel-by-pixel soil moisture maps of the test site, with four levels of soil moisture, were generated from the available images by using a new ANN that took into account the effects of vegetation.
Acknowledgements
This work was partially supported by the EC project FloodMan (EVK1-2001-00237), the Italian Space Agency (ASI), and the ESA/ENVISAT programme. We thank the personnel of the CVA (Avalanche Defence Centre) in Arabba (Belluno, Italy) for their kind support and cooperation in collecting ground truth data.