Présentations Colloques

Session 8.10: Recent developments in groundwater modeling and mathematical tools in Hydrogeology
Ramasomanana Fanilo
On the efficiency of ELLAM for mass transport in fractured porous media- Application to Qatar's aquifer storage project
Water security is one of the main Grand Challenges aligned with Qatar’s 2030 National Vision, which highlights the urgent need to have access to safe, high quality and sustainable water supply. The Aquifer Storage and Recovery (ASR) project aims at artificially storing water in the aquifer for future use. This study contributes to the ASR by developing an efficient and accurate numerical model for flow and transport in fractured porous media. **The developed model is based on the discrete-fracture approach. A variety of numerical methods have been used and developed for the modeling of flow and transport in fractured porous media using the discrete-fracture approach. This approach is known to be accurate but its computation expense is high when applied to field scale problems. Despite the progresses in this field, scientific community is still interested in the development of new numerical methods and techniques to improve the accuracy and the efficiency of the existing numerical codes.**The flow in the fracture system is known to be advection-dominated. Classical numerical schemes, such as Finite Element or Finite Volume, may introduce unphysical oscillations and or numerical dispersion. The Eulerian Lagrangian Localized Adjoint Method (ELLAM) is a successful technique for dealing with such advection-dominated problems as it has widely been used to simulate transport in simple porous media. **In this work, the ELLAM is used for the first time to simulate transport in fractured porous media. Our objective is to adapt and to evaluate the efficiency of ELLAM for modeling transport in fractured porous media. The first results show that ELLAM is well adapted to handle the high contrast of velocity between the fracture and the porous matrix. It provides accurate results with less computational and memory requirements than advanced Eulerian methods.**