The 3D inversion of transient electromagnetic (TEM) data remains a highly challenging optimization problem that needs to be resolved in the field of geophysical electromagnetic (EM) exploration. In this study, we present an effective 3D TEM interpretation framework based on time–frequency conversion (TFC) using UNet++ and demonstrate its capability in recovering complex 3D structures as well as its applicability in practical explorations. First, a generalized ellipsoid and Euler rotation are employed to construct 3D geoelectrical models, ensuring that the generated geological bodies possess sufficient geometric complexity while remaining geologically plausible. Second, frequency-domain (FD) data are recovered from time-domain (t-d) data using UNet++, which then serve as pseudo-observation data to be input into a 3D FD inversion code. Finally, the staggered-grid finite difference method is adopted for modeling the TEM responses, and the nonlinear conjugate gradient (NLCG) method is utilized to perform 3D inversion. Taking TEM data from a ground-wire source as an example, a synthetic test on models with complex 3D anomalies confirms that the proposed framework can accurately recover the geometry and position of subsurface bodies. Furthermore, the inversion code is tested on the field data acquired in Qinghai Province, China, and the inversion results are in good agreement with known geological and other geophysical information. Results demonstrate that performing TFC using UNet++ provides a reliable data foundation for 3D FD inversion of TEM data. This framework offers an effective approach for the practical 3D interpretation of TEM data and expands the research path for addressing complex 3D problems in geophysical EM explorations.

Gradient-based inversion methods are widely used for time-domain airborne electromagnetic (AEM) data, but their results are commonly affected by the initial model and regularization-induced smoothing. These limitations are particularly evident when thin layers or alternating high- and low-resistivity structures need to be resolved. To address this problem, we propose a gradient-guided iterative enhancement (GGIE) inversion that combines a limited-iteration Gauss–Newton (GN) inversion with a lightweight U-Net. In each GGIE inversion iteration, the GN module first produces a coarse inverted model (IM) that preserves the main data-driven geoelectric trend but is still affected by regularization-induced smoothing. The trained U-Net then predicts a structurally enhanced model (PM) from the observed data and the IM. A data-misfit-guided adaptive approach is proposed to calculate the weight coefficients of the IM and PM and to construct an update model (UM). These coefficients are further smoothed by a momentum term so that the relative contributions of the IM and the PM are adjusted adaptively during the iterations. This design reduces error propagation from either component alone and dynamically balances learned structural enhancement with physics-based data consistency. The UM then serves as the initial model for the subsequent GN inversion. GGIE inversion is tested on synthetic data, and the results show that it is most beneficial for complex multilayer structures, for which it reduces the mean relative error and root mean squared error (RMSE) by 49.0% and 28.6%, respectively. Compared with the physics-informed neural network (PINN) baseline, GGIE inversion reduces the model relative error, log-domain RMSE, and data misfit by 19.4%, 7.0%, and 71.7%, respectively. Moreover, compared with U-Net alone, GGIE inversion reduces the data misfit by 85.4%. The proposed method is further applied to field data acquired from the Fox River area in Wisconsin, USA. The main advantage of GGIE inversion is its ability to resolve complex multilayered structures, thin layers, and sharp resistivity contrasts with improved accuracy and stability.