Datensatz

Zusammenfassung

A successful utilization of deep geothermal resources requires to make accurate predictions about the reservoirtemperature distribution as well as an in depth knowledge of the hydraulic processes exerting a direct influence onthe subsurface temperature distribution and therefore on the productivity of geothermal reservoirs.The aim of this study is to investigate and quantify the influence that regional thermo-hydraulic processes exert onthe geothermal configuration of potential reservoirs in the German federal state Hesse. Specifically, we addressthe question of how the regional thermal and hydraulic configuration influences the local reservoir conditionsand whether it is possible to improve subsurface predictions iteratively by relying on 3D numerical modelingtechniques. Therefore, a 3D structural model of Hesse is used as a basis for coupled 3D thermo hydraulicsimulations of the deep fluid and heat transport. To uncover the effects of process coupling, a stepwise workflow isfollowed. We first simulate the thermal and hydraulic field under steady-state conditions by means of two differentuncoupled simulations and then analyze the results of the coupled thermo-hydraulic steady-state simulations. Ina last effort, we investigate the influence of fluid viscosity and density varying with temperature and pressure intransient coupled simulations.As a result of our numerical simulations, Hesse can be differentiated into sub-areas differing in terms of the domi-nating heat transport processes. In a final attempt to quantify the robustness and reliability of the modeling results,we carry out an analysis of the modelling outcomes by comparing them to available subsurface temperature data.Modelled temperatures show different levels of fit with locally measured well temperatures. These differences inmodel fit indicate the need for either structurally refined models and/or iterative adaptions within realistic rangesof the hydraulic and thermal properties. Structural refinements can often only be handled with smaller-scalemodels, which will, in turn, benefit from the boundary conditions and improved process understanding as derivedfrom the regional modelling approach presented here.