158 Geophysics in Geothermal Exploration Based on new gravity data acquired in Northern Alsace and their comparison to the older Bouguer anomaly, a qualitative data analysis reveals several negative Bouguer anomalies suggesting a decrease of the bulk density at the depth that fits with potential geothermal reservoirs like the Lower Trias and the top basement (Abdelfettah et al., 2020). A more quantitative analysis of gravity data combined with 3D geological models outlined areas with low density values that could be explained either by the variation of petrography within the basement and/or the occurrences of highly fractured zones associated with geothermal fluid affecting the bulk density values. Due to their sensitivity to fluids and particularly brine water in rocks, passive electromagnetic (EM) techniques (e.g. Magnetotellurics or MT) or active (Controlled-Source Electromagnetic or CSEM) have been traditionally used to investigate the subsurface conductivity. Therefore, MT surveys were conducted in Northern Alsace, respectively, close to Soultz in 2007-2008 and Rittershoffen in 2013-2014. MT data collected in the Soultz area were combined with other geophysical data for estimating temperatures at depth below the existing geothermal wells drilled at 5000 m. The main result of this analysis based on MT, was the forecast of a very deep convective cell below GPK-2 at around 6000–8000 m. Results from continuous MT measurements done at Rittershoffen in 2013-2014 suggested transient variations in subsurface conductivity due to the occurrence of fluids at depth. Furthermore, by using MT response versus time, it revealed that fluids could migrate in a NE direction from the injection well GRT1. Therefore, MT is not only a method for geothermal exploration or for assessing temperatures at depth but could be used as a monitoring tool during hydraulic stimulation or geothermal exploitation. EM methods have shown to be effective to characterize geothermal reservoir geometry in volcanic area, hydrocarbon reservoir geometry in offshore sedimentary area or onshore mineral exploration but not really in EGS. Nevertheless, the ability of EM methods to image targets with high geothermal potential in deep fractured reservoir and in a high man-made noise environment still needs to be demonstrated. Indeed, CSEM sources must compete with high noise levels and a conductive sedimentary cover resulting in low signal to noise ratio. At SsF, a full-scale 3D CSEM campaign done in 2020 demonstrated the ability of the technique to image resistivity variations underneath a thick sedimentary cover (>1400 m). An assessment of subsurface rock mineral compositions derived from their physical properties measured through geophysical logging, employing a combination of statistical and machine learning techniques has been applied to the Triassic sedimentary reservoirs from the URG (Pwavodi et al., 2024). Based on various geophysical data from the geothermal SsF and Rittershoffen wells, mineral composition was spatially predicted and compared with existing mineral descriptions. This approach based on machine learning helps in deciphering complex mineralogical compositions and geological structures within subsurface geothermal reservoirs from the URG.
RkJQdWJsaXNoZXIy NjA3NzQ=