Skip to Main content Skip to Navigation
New interface
Theses

Conditionnement des modèles de réseaux de fractures à des données géologiques et géophysiques pour la réduction des incertitudes dans les modèles d’écoulement et de transport

Abstract : The identification of permeable fractures in the subsurface is critical for evaluating potential contaminant transport pathways from deep disposal sites. A common approach is to develop Discrete Fracture Network (DFN) models from field mapping (boreholes and outcrops) and hydraulic experiments. The work accomplished in this thesis attempts at conditioning DFN models to constraints offered by Ground Penetrating Radar (GPR) geophysical data. Two surface-based GPR experiments were first conducted in a tunnel located at 410 m depth in the Äspö Hard Rock Laboratory (Sweden) with the aim to identify very weakly transmissive fractures (≈ 10-9 – 10-10 m²/s) and associated flow paths between two boreholes solicited by a tracer test. We demonstrate that the surface-based GPR method is able to detect about 80% of open and sub-horizontal fractures with areas of 1-10 m² and to detect likely tracer pathways through GPR-inferred fracture-aperture variations caused by the high injection pressure. The 3-D geometrical information of the fractures and their connectivity were then used as conditioning constrains in stochastic local DFN models. We have established validation criteria for our models, based on the connectivity measured on the field, and we show that conditioning allow us to validate up to 40% more realizations than stochastic models. The main outcomes of this research are the demonstrated ability of the GPR method to identify open fractures in formations of very low overall permeability that are targeted to store canisters containing spent nuclear fuel and to provide the first results improving the predictive capacity of stochastic DFN models by conditioning to geophysical data at scales from a few to tens of meters.
Document type :
Theses
Complete list of metadata

https://tel.archives-ouvertes.fr/tel-03238815
Contributor : ABES STAR :  Contact
Submitted on : Thursday, May 27, 2021 - 11:35:08 AM
Last modification on : Thursday, June 2, 2022 - 2:48:15 PM
Long-term archiving on: : Saturday, August 28, 2021 - 6:45:27 PM

File

MOLRON_Justine.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-03238815, version 1

Citation

Justine Molron. Conditionnement des modèles de réseaux de fractures à des données géologiques et géophysiques pour la réduction des incertitudes dans les modèles d’écoulement et de transport. Sciences de la Terre. Université Rennes 1, 2021. Français. ⟨NNT : 2021REN1B004⟩. ⟨tel-03238815⟩

Share

Metrics

Record views

137

Files downloads

60