Speaker
Description
Karst aquifers are characterized by highly heterogeneous conduit networks that control groundwater flow. Explicit simulation of flow in large karst networks, often composed of tens of thousands of conduits, remains computationally challenging due to their size and topological complexity.
In this work, we propose a graph-based framework for the upscaling of karst conduit networks, where conduits are represented as weighted edges and junctions as nodes. The objective is to construct reduced (coarse) graph models that preserve the hydraulic behavior of the original network while significantly decreasing its computational cost. We investigate coarsening strategies based on spectral clustering of the graph Laplacian and effective resistance metrics, which are well suited to capture long-range flow interactions in networks exhibiting strong heterogeneity and cycles.
The proposed approach is assessed on synthetic and realistic karst networks by comparing fine and coarse models in terms of hydraulic potentials, flow rates, and effective resistances. The results demonstrate that the coarse graphs can reproduce key global flow properties of the original networks, while offering substantial reductions in model size. This framework provides a scalable and physically meaningful tool for large-scale karst flow simulations.
| Country | France |
|---|---|
| Student Awards | I would like to submit this presentation into both awards |
| Acceptance of the Terms & Conditions | Click here to agree |
