Discrete Fracture Network reconstruction using a tomographic approach
Abstract
Methods of hydraulic and tracer tomography have evolved as promising options for characterization of those geological structures that are relevant for flow and transport. In this study, we focus on hard rock aquifers, where fractures represent preferential pathways for groundwater. The fracture system is described as a discrete fracture network (DFN), which is reconstructed by an attuned inversion procedure using tomographic measurements. Such measurements are, for example, multi-level tracer tests with different depth-dependent tracer breakthrough curves, or multi-level pressure recordings obtained from hydraulic interference tests between adjacent wells. Such set-ups can be utilized for spatially resolving the structures in the cross section between source and receiver wells. In this work, a transdimensional inversion procedure is developed to iteratively adjust the number and the position of fractures in the examined cross section until the flow or transport model with the calibrated DFN yields results consistent with the measurements. It is demonstrated based on synthetic test cases, that especially hydraulic tomography is suitable for reproducing major DFN features, whereas the information from tracer tests is less useful. The findings show the potential and the shortcomings of the current version of the DFN inversion procedure, which is currently expanded as a three-dimensional tomographic approach.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2019
- Bibcode:
- 2019AGUFM.H32A..01E
- Keywords:
-
- 0555 Neural networks;
- fuzzy logic;
- machine learning;
- COMPUTATIONAL GEOPHYSICS;
- 1829 Groundwater hydrology;
- HYDROLOGY;
- 1835 Hydrogeophysics;
- HYDROLOGY;
- 1873 Uncertainty assessment;
- HYDROLOGY