Coupled hydrogeophysical inversion of electrical resistances and inflow measurements for topsoil hydraulic properties under constant head infiltration

Abstract

Accurate estimation of topsoil hydraulic properties is important for understanding water flow and solute transport in the vadose zone. Coupled hydrogeophysical inversion schemes that enable the use of multiple geophysical and hydrological data for the estimation of soil hydraulic properties have recently been proposed. In these coupled inversion schemes, a hydrological model describing the process under investigation is coupled to a forward geophysical model and hydraulic parameters are directly estimated from geophysical measurements. While these schemes provide a suitable platform for the integration of multiple geophysical and hydrological data, efficient methods to combine these data types for improved parameter estimation still warrant investigation. In this study, we investigated the feasibility of estimating three topsoil Mualem-van Genuchten parameters from the fusion of inflow and electrical resistance measurements obtained under constant head infiltration. In addition to using only inflow or electrical resistances, we investigated three methods of combining these data for improved estimation of topsoil hydraulic parameters. Our results show that using inflow alone does not provide a unique solution to the inverse problem. Better results are obtained with the additional use of electrical resistance data. We show that successful data fusion within the coupled hydrogeophysical inversion framework depends on the choice of an appropriate objective function. We obtained the best data fusion results with an objective function defined as the sum of the root mean square error of both data types normalized by the standard deviation of the respective measurements. In this case, the inverted hydraulic parameters were very comparable to reference values obtained from a multi-step outflow experiment carried out with undisturbed soil cores from the experimental site. It is concluded that the coupled hydrogeophysical inversion framework is a promising tool for non-invasive near-surface hydrological investigations. Parkin et al. 1995;Huisman et al. 2002;Schwartz and Evett 2002) during infiltration events. Although the water content and matric potential of the vadose zone can exhibit large spatial variation in the horizontal and vertical direction (e.g., Flury et al. 1994), tensiometry and TDR only provide limited spatial coverage (0.01-1 dm 3 ), requiring time-consuming measurements at many locations for field scale sampling. Geophysical methods like ground-penetrating radar (GPR; Binley and Beven 2003; Huisman et al. 1994; Faybishenko 2000) and time-domain reflectometry (TDR;

Estimation of Soil Hydraulic Parameters in the Field by Integrated Hydrogeophysical Inversion of Time‐Lapse Ground‐Penetrating Radar Data

“…Mualem (1976) reported that λ equal to 0.5 was an optimal value for a data set of 45 disturbed and undisturbed samples. In many studies aiming to estimate the soil hydraulic properties from coupled hydrogeophysical inversion the parameter λ is fixed to 0.5 to minimize the number of unknowns, e.g., Looms et al (2008); Hinnell et al (2010); Huisman et al (2010); Mboh et al (2011, 2012a,b). Given these considerations, we assumed θ r , θ s , and λ to be known in the hydrogeophysical inversion process.…”

Joint Inversion in Hydrogeophysics and Near‐Surface Geophysics

“…Looms et al [2008] showed that five layers were needed to explain their field data, but that only permeability and one additional fitting parameter could be retrieved for each layer. Hinnell et al [2010] and Mboh et al [2012a] investigated how surface-based ERT monitoring of infiltration tests could improve soil hydraulic property estimation. Hinnell et al [2010] showed in a synthetic example that petrophysical coupling can reduce parameter errors, but only if the conceptual hydraulic model is correct.…”

“…Considering an undisturbed soil core, Mboh et al [2012a] also improved their parameter estimates when combining inflow measurements with ERT data-. They also highlighted the challenge of assigning appropriate weights to each data set in the objective function.…”

Numerical study on CO2 leakage detection using electrical streaming potential data

“…[]; Mboh et al . []. A number of different approaches to model the saturation dependency of the coupling are proposed in the literature [ Guichet et al ., ; Revil and Cerepi , ; Linde et al ., ; Allègre et al ., ; Jougnot et al ., ].…”

“…The approach of Linde et al . [] was confirmed by independent experiments [ Mboh et al ., ]. Jougnot et al .…”