Use and challenges of electrical resistivity tomography to study processes in agro-ecosystems

Garré Sarah Javaux, Mathieu [UCL] Manhaeghe Thibault Hermans Thomas Nguyen Frederic

Sustainable management of agro-ecosystems requires a thorough understanding of the interaction between physical, chemical and biological processes at play. In addition, processes at pore scale are linked to field scale phenomena, but this connection is often poorly understood. Electrical resistivity tomography (ERT) is increasingly used in the context of agriculture since the measured resistivity distribution can be linked to soil moisture, soil structural characteristics or pore water salinity. Due to its minimally invasive character, its spatial coverage and its monitoring abilities, ERT can be used to study field heterogeneity and competition between plants, quantify water fluxes throughout a growing season or distinguish preferential flow pathways in soils. Is resolution is well-below classical soil imaging techniques such as X-ray CT or MRI, but its spatial coverage much larger. This highlights the potential of ERT to link our knowledge obtained from pore scale data to field scale processes. Nevertheless, a lot of challenges still remain. A Tikhonov-type regularization approach is often used to solve the ill-posed, inverse problem linked to ERT, resulting in a smoothed resistivity distribution. However, in reality strong contrasts can exist due to e.g. compacted soil layers due to ploughing, water infiltration fronts, etc. and in that case other operators have been proposed to regularize the inversion. Taking into account spatial heterogeneity of petrophysical characteristics and providing a realistic uncertainty estimation are additional challenges, which can be addressed using stochastic approaches. Monitoring data provides further elements to constrain the inverse problem: data can be replaced by data difference and regularization may incorporate the temporal dimension for instance. However, such constraints require their compatibility with the studied temporal process, which is not always straightforward. Several alternative strategies are being developed, such as coupled hydrogeophysical inversion, or stochastic approaches using a prior falsification/validation method following a PopperBayes philosophy. In this talk, we will address some of these challenges and give some recent applications in the field of agro-geophysics.