Using machine learning and SAR data for the upscaling of large scale modelled soil moisture in the Alps

Konferenz: EUSAR 2016 - 11th European Conference on Synthetic Aperture Radar
06.06.2016 - 09.06.2016 in Hamburg, Germany

Tagungsband: EUSAR 2016

Seiten: 4Sprache: EnglischTyp: PDF

Persönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt

Greifeneder, Felix; Notarnicola, Claudia (EURAC Research Bolzano, Italy)
Wagner, Wolfgang (Vienna University of Technology, Austria)

Knowledge of the spatial and temporal distribution of soil moisture is important for many geoscience disciplines. Currently available remote sensing soil moisture products are not able to fully represent the heterogeneous patterns in mountain areas. In this paper we present a machine learning based approach for the upscaling of coarse scale modelled soil moisture based on Synthetic Aperture Radar data. For this purpose a statistical model was trained using the Gradient Tree Boosting approach. Results show that with the trained model it is possible to reproduce the reference data and increase spatial detailed of mapped soil moisture content.