Investigating soil water dynamics using repeated quantitative rigid boom mulit-coil EMI measurements and inversions

Electromagnetic induction (EMI) systems measure the apparent electrical conductivity (ECa)of the soil, which can be related to texture, salinity, and soil water content (SWC). Repeatedmulti-coil EMI measurements and inversions can be used to obtain lateral and vertical changesin SWC over time. The present study uses calibrated ECa data measured over nine monthsin a quantitative EMI inversion scheme to obtain the depth-specic electrical conductivitychanges, which are transformed into the vertical and lateral soil water changes along a sixtymeter long transect that crosses a buried river channel structure. The ECa values measuredwith a multi-coil EMI system are weighted averages over specic depth ranges of investigation,which data can principally be inverted to obtain a layered electrical conductive model. However,external inuences such as cables and GPS systems alter the ECa measurements suchthat these data need calibration prior inversion. A novel time-lapse VES-based calibrationscheme was developed to nd time-independent calibration parameters. When using theseparameters, quantitative time-lapse transect EMI data were obtained and inverted using aMaxwell-based full solution forward model in a shued complex evolution inversion scheme.This provided a highly accurate description of the subsurface where dierences in layer electricalconductivity (EC) and especially in the paleo-river channel were observed. Whereasthe paleo-river channel structure was less visible in the rainy autumn to winter 2017, the ECof the channel increased up to May 2018 most likely due to retarded water recharge. Sincetemperature also impacts the electrical conductivity, a temperature correction using temperaturesfrom time-lapse depth-dependent modeling was novelly applied to inverted time-lapseEMI data. To obtain soil water dynamics from this time-lapse EC data, a modied Archie'slaw was used. The modeled SWC was correlated to recharge rates obtained from meteorologicaldata. Summarizing, this work quanties soil water dynamics using new approachesto correct both for external inuences to ECa and temperature inuences to inverted ECdata. The resulting SWC estimates are in good agreement with gravimetrically obtainedSWC and estimated recharge rates, which consequently indicates robust calibration and areliable temperature correcting approach in time-lapse studies such that EMI measurementsand inversions can be used to investigate soil water dynamics