Spatio-temporal variability of soil moisture in alpine areas : high-resolution detection based on the combination of remote sensing data, a digital terrain model and ground-based measurements

Soil moisture is a critical environmental factor that affects evapotranspiration, surface runoff, and plant water availability. At the Global Observation Research Initiative in Alpine environment (GLORIA) Project master site on the Schrankogel, research on plant dispersal and development in high mountains has been ongoing for over 30 years. A limiting factor for plants is their water availability during the snow-free period. Therefore, the objective is to investigate the study area’s spatio-temporal variability of soil moisture. Terrain data, like slope, aspect, elevation, topographic wetness index (TWI), curvature, remote sensing data (Sentinel-2 and Landsat 8) and in-situ soil moisture measurements could be combined by the machine learning algorithm Random Forest. The resulting final products are different soil moisture maps of the study area with different distances to the last precipitation event. The calculated model accuracy reaches the maximum of R² = 0.67 one day after the precipitation. The accuracy decreases with increasing distance to the precipitation event. Evaluation of the maps revealed rapid drying on the south-facing slopes and longer-lasting moisture in excavations, canyons, and ditches (high TWI). After a precipitation event, soil moisture is heterogeneously distributed. One reason for this is assumed to be the high variability of grain size distributions and soil depths, for which little research knowledge is available for the study area. In addition, it was found that the most influential variables on soil moisture are exposure to the sun and the TWI. Satellite-based remote sensing is unsuitable for estimating water content due to the high small-scale soil moisture variability. Finally, recommendations for further research on soil moisture in the study area are collected.
by Marco Wegler, B.Sc., B.A.
Zusammenfassung in deutscher Sprache
Masterarbeit University Innsbruck 2023