Identifying Flow Networks in a Karstified Aquifer by Application of the Cellular Automata‐Based Deterministic Inversion Method (Lez Aquifer, France). Issue 12 (15th December 2017)

• Record Type: Journal Article
• Authors: Fischer, P.; Jardani, A.; Wang, X.; Jourde, H.; Lecoq, N.
• Is Part Of: Water resources research. Volume 53:Issue 12(2017)
• Page Start: 10508
• Page End: 10522
• Publication Date: 2017-12-15
• Subjects: hydraulic tomography -- karst -- aquifer characterization -- distributed modeling -- inverse problem -- fractured aquifer; Hydrology -- Periodicals; 333.91
• DOI: 10.1002/2017WR020921
• Languages: English
• ISSNs: 0043-1397

Abstract

Abstract: The distributed modeling of flow paths within karstic and fractured fields remains a complex task because of the high dependence of the hydraulic responses to the relative locations between observational boreholes and interconnected fractures and karstic conduits that control the main flow of the hydrosystem. The inverse problem in a distributed model is one alternative approach to interpret the hydraulic test data by mapping the karstic networks and fractured areas. In this work, we developed a Bayesian inversion approach, the Cellular Automata‐based Deterministic Inversion (CADI) algorithm to infer the spatial distribution of hydraulic properties in a structurally constrained model. This method distributes hydraulic properties along linear structures (i.e., flow conduits) and iteratively modifies the structural geometry of this conduit network to progressively match the observed hydraulic data to the modeled ones. As a result, this method produces a conductivity model that is composed of a discrete conduit network embedded in the background matrix, capable of producing the same flow behavior as the investigated hydrologic system. The method is applied to invert a set of multiborehole hydraulic tests collected from a hydraulic tomography experiment conducted at the Terrieu field site in the Lez aquifer, Southern France. The emergent model shows a high consistency to field observation of hydraulic connections between boreholes. Furthermore, it provides a geologically realistic pattern of flow conduits. This method is therefore of considerable value toward an enhanced distributed modeling of the fractured and karstified aquifers. Key Points: We apply a newly developed inversion method to characterize a karstified aquifer The approach permits to map the spatial distribution of the hydraulic conductivities in the fractured and matricial areas Our result model can simulate realistic flow behaviors in highly fractured aquifers