Three-dimensional hydrogeological parametrization using sparse piezometric data

Abstract

When modelling contamination transport in the subsurface and aquifers, it is crucial to assess the heterogeneities of the porous medium, including the vertical distribution of the aquifer parameter. This issue is generally addressed thanks to geophysical investigations. As an alternative, a method is proposed using estimated hydraulic parameters from a 2D calibrated flow model (solely reliant on piezometric series) as parametrization constraints for a 3D hydrogeological model. The methodology is tested via a synthetic model, ensuring full knowledge and control of its structure. The synthetic aquifer is composed of five lithofacies, distributed according to a sedimentary pattern, and functions in an unconfined regime. The level of heterogeneity for hydraulic conductivity spans 3 orders of magnitude. It provides the piezometric chronicles used to inverse 2D flow parameter fields and the lithological logs used to interpolate the 3D lithological model. Finally, the parameters of each facies (hydraulic conductivity and porosity) are obtained through an optimization loop, which minimizes the difference between the 2D calibrated transmissivity and the transmissivity computed with the estimated 3D facies parameters. The method estimates values close to the known parameters, even with sparse piezometric and lithological data sampling. The maximal discrepancy is 45 % of the known value for the hydraulic conductivity and 6 % for the porosity (mean error 26 % and 3 %, respectively). Although the methodology does not prevent interpolation errors, it succeeds in reconstructing flow and transport dynamics close to the control data. Due to the inherent limitations of the 2D inversion approach, the method only applies to the saturated zone at this point.