Enhancing Electrical Resistivity Tomography Detection of Landslide Structures Through Multisource Geological Constraints
The electrical resistivity tomography (ERT) method is a powerful tool for inferring subsurface structures in landslide investigations. However, its effectiveness is often limited by the inherent nonuniqueness of inversion and the severe smearing artifacts caused by conventional regularization, particularly for delineating thin structures like slip zones. To address these challenges, this article proposes a unified and application-oriented constrained inversion framework for ERT that flexibly integrates multisource prior geological information. Building on established constrained inversion concepts, the framework combines structurally guided directional constraints, borehole resistivity constraints, differentiated regional constraints, and localized borehole-derived interface and dip constraints within a consistent Gauss–Newton inversion workflow. The framework integrates different types of constraints into the inversion process, including structural orientations, known resistivity points from boreholes, and predetermined interface positions. We first demonstrate the efficacy of each constraint type through numerical simulations on a theoretical landslide model. The method was then validated with a comprehensive field study at the Huangtupo landslide in the Three Gorges Reservoir area of China. In situ resistivity tests within a tunnel system provided characteristic resistivity values for key structural elements (slip zone, slip mass, and bedrock), forming the basis for the reference model. Constrained inversions were performed on four ERT survey lines by incorporating interfaces from seismic ambient noise tomography (S-ANT) and orientations from multiple boreholes and tunnels. The results show that, compared with unconstrained inversion, the proposed constrained strategy reduces inversion ambiguity associated with equivalence and suppresses the smearing effects, enabling a clearer identification of multilevel landslide structures that are typically obscured in standard ERT images. Finally, a detailed and reliable 3-D model of the landslide was constructed by integrating the interpreted 2-D profiles. The proposed constrained inversion approach mitigates the common pitfalls of standard ERT and provides a practical methodological basis for landslide characterization and stability analysis.