Toward catchment vadose zone characterization by linking geophysical electromagnetic induction and remote sensing data

Abstract:

Large-scale information of the crop status can be provided by multispectral remote sensing (RS) products. However, to fully understand the observed RS patterns including plant growth related processes such as water and nutrient availability, knowledge of the vadose zone is necessary, which can be obtained by geophysical methods. We studied a 20 ha test site in Selhausen (Germany), where the upper terrace (UT) sediments consist of sand and gravel, whereas the lower terrace (LT) sediments consist of loamy silt. Leaf area index (LAI) maps that were derived from RapidEye satellite data taken after a drought period showed a high density of undulating structures of higher LAI values within the sand and gravel dominated (and generally lower LAI) UT. These structures were related to better crop performance originating from subsurface loamy silt paleo-river channels. Next, large-scale apparent electrical conductivity (ECa) data were obtained using a multi-configuration electromagnetic induction (EMI) sensor with depths of investigation (DOI) up to 1.8 m. The observed LAI patterns coincided well with the ECa patterns of the 1.8 m DOI measurements, and soil analysis confirmed the presence of silty soil in the deeper subsoil. To gain more knowledge, a novel EMI inversion scheme that inverts for a layered subsurface using multi-configuration EMI data was developed and applied to a one ha large field that contained both UT and LT sediments in the eastern and western part, respectively. The obtained smoothly changing lateral and vertical electrical conductivity model was confirmed by grain size distribution maps and two previously measured 120 m long electrical resistivity tomography (ERT) transects. Conclusively, the combined LAI and EMI analysis can be extended to relatively large areas up to the catchment scale to improve environmental models that aim at improved descriptions of plant growth, water, nutrient and energy processes.