Summary

The MICAS project "Modelling and Intensive Computation for Aquifer Simulations" deals with the use of high performance computing for understanding hydraulic properties of heterogeneous media and for modelling aquifers. It contributes to the debate about macro-dispersion, interpretation of drawdown signals, interactions between rock and fractures.

This project is supported by the French Research Agency (Agence Nationale de la Recherche ANR), in the framework of the ANR-CIS-07 program (Calcul Intensif et Simulation).

Objectives

The MICAS project, "Modelling and Intensive Computation for Aquifer Simulations" is designed to solve great challenge in hydrogeology and to develop a generic open source software.

Numerical modelling is an important key for the management and remediation of groundwater resources. Natural geological formations are highly heterogeneous, leading to preferential flow paths and stagnant regions. The contaminant migration is strongly affected by these irregular water velocity distributions. In order to account for the limited knowledge of the geological characteristics and for the natural heterogeneity, the MICAS project relies on Uncertainty Quantification methodes. In previous and current work, we use a classical Monte-Carlo method, with random permeability fields and random Discrete Fracture networks. In this stochastic approach, numerical simulations consist in computing the velocity field over large spatial domains and solving solute transport over large temporal scales. This approach must overcome two main difficulties, memory size and runtime, in order to solve very large linear systems and to simulate over a large number of time steps. High performance computing is thus necessary to carry out these large scale simulations.

The objectives of MICAS are to get outstanding results in seven well-identified topics, and to continue the development of the software platform H2OLab for integrating all the modules developed in the project.

The platform H2OLab is used by the team GEOSCIENCES for several closely related scientific projects.

Computing resources

• The teams CDCSP, SAGE and LMPG have access to the GENCI computers since october and june 2009 (accepted projects)
• The teams SAGE and LMPG have access to the Grid'5000 facilities for experiments on parallel and grid computing
• The team LMPG has a cluster called MecaClus, partly funded by ANR with the Micas project
• The team CDCSP has a cluster, partly funded by ANR with the Micas project
• Each team has a powerful multiprocessor machine for development and validation

National academic relations

• The teams SAGE, LMPG and CDCSP participated in the GNR MOMAS in 2008-2009 (three projects).
  The team SAGE coordinates a two-years project (2010-2011) on numerical methods for reactive transport.
• The team SAGE participates in the HEMERA group at INRIA. Hydrogeology is one scientific challenge for grid computing.
• The teams GEOSCIENCES and SAGE participate in the RISC-E regional network, connected to the RNSC national network (Reseau National des Systemes Complexes).

International academic relations

• The teams GEOSCIENCES, LMPG and SAGE have collaborations with the university of Barcelona.
• The teams GEOSCIENCES and SAGE have collaborations with and the university of Leipzig.
• The team GEOSCIENCES has collaborations with the university of San Diego.
• The teams GEOSCIENCES, SAGE and CDCSP have collaborations with the university of Tunis.
• The teams SAGE and GEOSCIENCES participate in the european project CO-ADVISE (Marie Curie People program).
  This project is closely related to the HYDROMED network (program INRIA-Med), a follow-up to the HYDRO 3+3 INRIA-Med network.
  Other partners are the universities of Tunis, Rabat, Kenitra, Annaba.