Three‐dimensional electrical resistivity tomography for earth‐rock dam seepage detection: Integrating real‐world topography, water levels and unstructured meshing

Abstract Seepage in earth‐rock dams poses significant risks to hydrological safety and infrastructure stability, yet conventional two‐dimensional (2D) electrical resistivity tomography (ERT) struggles to resolve complex seepage pathways under irregular topography and fluctuating water levels. This study introduces an enhanced 3D ERT framework, employing a finite element forward model and conjugate gradient inversion, integrated with real‐world topography, water level elevations and unstructured meshing. Compared to traditional methods, unstructured meshing enables precise terrain simulation through localized refinement while overcoming grid‐based electrode constraints, supporting flexible, non‐uniform placement suited to complex dam surfaces. Simulation results show that incorporating water level data corrects direct current electric field distortions, significantly enhancing seepage zone resolution, particularly during low or normal water periods, with field‐validated accuracy improved up to threefold. Applied at Lanxi City, Zhejiang, China, the method completed 3D inversion of real‐world data with topography in 57 min on a workstation, successfully identified two low‐resistivity seepage zones, including a challenging dam abutment anomaly, and was validated by borehole imaging. This approach outperforms traditional ERT methods, offering robust hydrogeophysical insights for dam safety and sustainable water management. Future advancements in node‐based ERT systems and automated inversions will further enhance its efficacy in complex hydrological environments.