SSS Soil System Sciences
The 2021 Virtual Outstanding Student and PhD candidate Presentation (vOSPP) Award is awarded to Hanna Zeitfogel for the poster/PICO entitled:
Variability across scales – exploring methods for predicting soil properties from multiple sources (Zeitfogel, H.; Feigl, M.; Schulz, K.)
Click here to download the poster/PICO file.
Hanna Zeitfogel is a PhD student at the Institute of Hydrology and Water Management (BOKU, Vienna). The overall goal of her research project is the estimation of spatially distributed groundwater recharge rates and their variability (for the extent of Austria) under changing climate conditions. Therefore, in a first step the regionalization of soil (hydraulic) properties is a major point of interest. Currently, there is no single data product available which covers the whole study area and still displays the variability of local soil observations. The challenge thus is to combine soil data from different sources and resolutions and, at the same time, to preserve the characteristically high spatial variability of soil properties. Preliminary results of this first objective were presented at EGU2021. For the spatial prediction of soil parameters two machine learning models (XGBoost and Feed Forward Neural Networks) were trained with globally available environmental raster datasets and compared to each other. First results show that both models are able to largely reproduce the original data variability.