Gravimetry for geothermal exploration: methodology, computer programs and two case studies in the Swiss Molasse basin

In Switzerland and neighboring countries, most deep geothermal projects target deep aquifers associated with major fault zones. The objective of this thesis is to develop and test a methodology for the estimation and localization of such high porosity rock volumes in the underground using gravity data and 3D geological modeling. To reach that aim, the computer program GInGER – GravImetry for Geothermal ExploRation - was developed. It allows computing rapidly and efficiently the intensity of the gravity field in surface from 3D geological models. Then the user can compare the calculated response with measured gravity data and strip the effect of geological structures that are not relevant for the project. One of the main originality of GInGER is that it permits to directly assess the porosity of fault damage zones when providing their geometry. The software is also endowed with a graphical user interface and density inversion capabilities to be user friendly. In addition, for sensitivity analysis, an independent sister software named GInGERSP, based on the same principles allow to compute the gravity effects of simple geometrical shapes which can be superposed to a given geological model. This tool is particularly designed for rapid sensitivity analysis. The two computer programs and the underlying methodology were tested on two ongoing geothermal projects in Switzerland, the projects of Sankt Gallen and Eclépens. Both projects are targeting deep aquifers associated with major faults zones affecting the Mesozoic sediments under the Swiss Molasse Basin. In both case studies, a preliminary step was to conduct a sensitivity analysis to determine the prerequisites allowing the assessment of faults damage zone porosity. This was done in 2D for the case study of Sankt Gallen and it appeared that, if the geometry of the fault zone is constrained enough, if the damage zone of the fault represent a sufficient volume and the error on gravity measurement is kept as small as possible, typically under 0.1 mGal. It should be possible to assess the porosity induced by the damage zone of the fault using gravity data and that even if the induced porosity is rather small, under 1-2 % if the affected volume is big enough. In Sankt Gallen, the 3D geological model was established using 3D seismic data. Comparing the gravity effect of this model with gravity data specifically acquired, we were able to make a first estimation of fracture porosity that compared well with later measurements obtained from drilling. However, the presence of a Permo-Carboniferous graben in the vicinity of the fault zone prevents us from removing all ambiguities on the result. In Eclépens, the 3D geological model was established using only 2D seismic data and oil exploration wells. As no accurate geometry of the different fault zones exist, it was not possible to assess precisely the porosity of the damage zones. However, using gravity measurements, acquired during this study, and the two computer programs, GInGER and GInGERSP, it was possible to make a sensitivity analysis, compare the possible effects of the various fault zones porosity and highlight the most suitable targets. This demonstrates how a user could use these two tools for any case study. Note also, that this work also revealed that a major Permo-Carboniferous graben of regional extent likely crosses the northern part of the model. This could, to a certain extent, explain the high geothermal gradient found in the Eclépens-1 oil prospection well which is located in the central part of the targeted area of the geothermal project. For these two case studies, the application of the proposed methodology and the combined use of 3D modelling, gravity data, GInGER, and GInGERSP allowed to bring new and useful information for the geothermal projects