CAN SMOS SOIL MOISTURE DRY-DOWNS BE USEFUL TO DETECT FLOOD CONDITIONS OVER THE ARGENTINEAN PAMPAS PLAINS?
- 1Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires (FCEN/UBA). Centro de Investigaciones del Mar y la Atmósfera (CIMA/UBA-CONICET). Institut Franco-Argentin d'Études sur le Climat et ses Impacts (UMI 3351-IFAECI/CNRS-IRDCONICET-UBA), Intendente Güiraldes 2160, Ciudad Universitaria, C1428EGA, CABA, Argentina
- 2Quantitative Remote Sensing Group, Instituto de Astronomía y Física del Espacio (IAFE, UBA/CONICET), Intendente Güiraldes 2160, Ciudad Universitaria, C1428EGA, CABA, Argentina
- 3Instituto de Matemática Aplicada de San Luis, Av Ejército De Los Andes 950, San Luis (5700), Argentina
- 4IPGP, CNRS-UMR 7154, Univ. Paris Diderot, USPC, Paris, France
- 5Irstea, UR RiverLy, Centre de Lyon-Villeurbanne, Villeurbanne, France
Keywords: soil moisture, floods, SMOS, dry-downs, Pampas Plains
Abstract. The process of soil drying following a single rainfall input offers an integrated perspective on soil-vegetation water dynamics in responses to atmospheric conditions during periods without rainfall. In this work, the soil moisture dry-down time scale events (τ) was calculated using surface soil moisture data from the SMOS mission, with the objective to explore if the spatio-temporal variability of τ could be used as a proxy for regional flooding and waterlogging characterization. Our working hypothesis is that soil moisture dries up more slowly under flooded conditions as a result of slower surface water elimination by infiltration and capillary rise of water from the saturated zone close to the surface. A clear difference precipitation-moisture coupling was detected between two regions with different flooding dynamics. In a region where flooding is triggered by precipitation excesses on weekly-to-monthly time scales and where the coupling between precipitation and evapotranspiration is strong, a positive correlation between dry-down and 6-month accumulated precipitation anomaly was found for all seasons except winter. By contrast, in the other region where flooding is largely de-coupled from precipitation and evapotranspiration, but rather coupled to ground water table dynamics on time scales from several months to years, no significant correlations were found. These results are based on a short period of data: March 2010 – November 2014.