Geothermal Exploration with Ambient Noise Tomography, Gravity Data and a 3-D MT Resistivity Model in a Joint Inversion Approach

Abstract

Geophysical exploration of unconventional geothermal resources is challenging due to the lack of any well known specific geophysical signature such as the clay cap observed in volcanic geothermal systems. Also, unconventional geothermal reservoir usually set in a complex geological environment. There, the purpose of surface exploration is to characterize possible permeable paths such as fault networks. In this context, geophysical models should provide some image at a scale large enough to understand the complex geology but with the adequate resolution to resolve features such as faults. One approach to overcome this difficulty is to use several geophysical methods providing complementary insight into the geology. We developed a 3-D joint inversion technique of gravity and ambient noise tomography constrained by a resistivity model from a prior 3-D MT inversion. This approach was applied to data collected for the geothermal exploration of a prospect in a crustal fault zone located in Massif Central, France. In 2017, a dense temporary seismic network with 299 vertical component seismic nodes was deployed in the area. We present the joint inversion of this new seismic data set with MT and gravity data. The geophysical data set consists of 48 MT soundings, 627 gravity measurements and 579 surface wave dispersion curves. We first inverted the 48 full MT tensors to obtain a 3-D resistivity model. We then jointly inverted the gravity data and the surface wave dispersion curves with our 3-D joint inversion approach. The joint gravity and seismic solutions are constrained by the 3-D resistivity model. Couplings between resistivity, density and shear wave velocity models are based on linear correlation. The correlation is enforced at the model scale, not locally. This strategy allows models to converge toward a solution with correlated parameters within some of geological domains only while locally free to adjust independently to fit the data. We identify a major fault zone extending North/South and separating two major geological units connected to a deep conductor.