Assessment of Electrical Anisotropy Effects on Magnetotelluric Modelling and Inversion at Annecy, French Pre-Alps

The magnetotelluric (MT) method is widely used for reservoir exploration, including hydrocarbons, magma chambers, and other conductive fluids. However, this technique is sensitive to electrical anisotropy, which can complicate the determination of the target's dimensionality, potentially leading to inaccurate modelling and inversion. Identifying whether a region exhibits anisotropy is non-trivial, as the direction of anisotropy may be independent from the geoelectrical strike of the target structures.

In this study, we aim to evaluate the impact of anisotropy on MT modelling and inversion for reservoir identification of a hydrothermal system in a fractured zone. For this, we present a dataset of 45 broadband magnetotelluric sites, collected over a 100 km² area in the urban region of Annecy, France, within the Western Alpine Molasse Basin (WAMB). The recordings span from 1 to 12 days, with sampling rates from 256 Hz to 65 kHz. This area is of particular interest due to the potential presence of a low-enthalpy hydrothermal reservoir located within Jurassic marlstone-limestone units at depths of 1.5–2 km. The region also features brittle, seismogenic structures such as the Vuache Fault.

Fluid circulation in a fractured region might occur at preferred orientations, which signature can lead to electrical anisotropy that could be mistaken for a geoelectrical strike. To address this, we propose a methodology that incorporates magnetotelluric forward modelling and inversion under various dimensional scenarios (1D, 2D, and 3D) and geological settings. We begin with a dimensionality analysis, followed by isotropic and anisotropic forward modelling and inversion based on the area's geology to achieve the best fit to the observed data.

The dimensionality analysis yielded inconsistent results, suggesting that electrical anisotropy might be influencing the data. Consequently, we conducted 1D, 2D, and 3D modelling. Results from 1D forward and inverse models incorporating anisotropy show acceptable fits to the field data compared to isotropic models. These findings support the presence of conductive layers consistent with the hypothesis of a hydrothermal reservoir at those depths. In contrast, 2D models, whether isotropic or anisotropic, could not fit the data at the target frequencies related to the potential water boundary. Ongoing work includes 3D anisotropic forward modelling and inversions, which, in preliminary results, suggest indications of possible electrical anisotropy in the region.