No abstract

The Upper Rhine Graben (URG) is an active rift with a high geothermal potential. Despite being a well-studied area, the three-dimensional interaction of the main controlling factors of the thermal and hydraulic regime is still not fully understood. Therefore, we have used a data-based 3D structural model of the lithological configuration of the central URG for some conceptual numerical experiments of 3D coupled simulations of fluid and heat transport. To assess the influence of the main faults bordering the graben on the hydraulic and the deep thermal field, we carried out a sensitivity analysis on fault width and permeability. Depending on the assigned width and permeability of the main border faults, fluid velocity and temperatures are affected only in the direct proximity of the respective border faults. Hence, the hydraulic characteristics of these major faults do not significantly influence the graben-wide groundwater flow patterns. Instead, the different scenarios tested provide a consistent image of the main characteristics of fluid and heat transport as they have in common: (1) a topography-driven basin-wide fluid flow perpendicular to the rift axis from the graben shoulders to the rift center, (2) a N/NE-directed flow parallel to the rift axis in the center of the rift and, (3) a pronounced upflow of hot fluids along the rift central axis, where the streams from both sides of the rift merge. This upflow axis is predicted to occur predominantly in the center of the URG (northern and southern model area) and shifted towards the eastern boundary fault (central model area).

Geothermal simulations are widely used in both scientific and applied industrial contexts. Typically, the temperature state is evaluated on the basis of the heat equation, with suitable parameterizations of the model domain and defined boundary conditions, which are calibrated to obtain a minimal misfit between measured and simulated temperature values. We demonstrate the essential need for global sensitivity studies for robust geothermal model calibrations since local studies overestimate the influence of the parameters. We ensure the feasibility of the study by using a physics-based machine learning approach, that reduces the computation time by several orders of magnitude.