An advanced discrete fracture model for variably saturated flow in fractured porous media

• Published in: Advances in water resources 2020-06, Vol.140, p.103602, Article 103602
• Main Authors: Koohbor, Behshad, Fahs, Marwan, Hoteit, Hussein, Doummar, Joanna, Younes, Anis, Belfort, Benjamin
• Format: Article
• Language: English
• Subjects: climate change; Discrete fracture-matrix model; Earth Sciences; Fractured Porous media; Life Sciences; Mass lumping; Method of Lines; Mixed hybrid finite element method; Sciences of the Universe; Variably saturated flow
• ISSN: 0309-1708; 1872-9657
• DOI: 10.1016/j.advwatres.2020.103602

•Variably saturated flow in fractured domains is a computationally challenging problem.•An advanced numerical model is developed to enable large field applications.•Mass lumping technique for fractures leads to a significant reduction in CPU time.•The model is used to predict the effect of climate change on real karst spring system.•Neglecting fractures leads to overestimation of the groundwater resources. Accurate modeling of variably saturated flow (VSF) in fractured porous media with the discrete fracture-matrix (DFM) model is a computationally challenging problem. The applicability of DFM model to VSF in real field studies at large space and time scales is often limited, not only because it requires detailed fracture characterization, but also as it involves excessive computational efforts. We develop an efficient numerical scheme to solve the Richards equation in discretely fractured porous media. This scheme combines the mixed hybrid finite element method for space discretization with the method of lines for time integration. The fractures are modeled as lower-dimensional interfaces (1D), within the 2D porous matrix. We develop a new mass-lumping (ML) technique for the fractures to eliminate unphysical oscillations and convergence issues in the solution, which significantly improves efficiency, enabling larger field applications. The proposed new scheme is validated against a commercial simulator for problems involving water table recharge at the laboratory scale. The computational efficiency of the developed scheme is examined on a challenging problem for water infiltration in fractured dry soil, and compared with standard numerical techniques. We show that the ML technique is crucial to improve robustness and efficiency, which outperforms the commonly used methods that we tested. The applicability of our method is then demonstrated in a study concerning the effect of climate change on groundwater resources in a karst aquifer/spring system in El Assal, Lebanon. Simulations, including recharge predictions under climate change scenarios, are carried out for about 80 years, up to 2099. This study demonstrates the applicability of our proposed scheme to deal with real field cases involving large time and space scales with high variable recharge. Our results indicate that the water-table level is sensitive to the presence of fractures, where neglecting fractures leads to an overestimation of the available groundwater amount. The proposed numerical approa