The deep learning-based gravity anomaly inversion method addresses the complex challenge of deriving subsurface density variation models from surface gravity anomaly data. In order to generate various geological environments and their corresponding surface gravity anomaly datasets, three-dimensional density models considering different spatial locations and density variations are created in this paper. At the same time, the residual module and spatial attention mechanism are introduced into the U-Net architecture to improve the learning ability and inversion accuracy of complex geological structures. Experimental results demonstrate that the proposed method achieves the high-precision reconstruction of density variation models in complex anomaly environments, with a model residual error lower than 3%. Additionally, the inversion results of the density change and the gravity change in the Longshoushan fault zone show that the 2022 Menyuan MS6.9 earthquake is in the middle of the positive and negative density changes, which verifies the applicability of the U-Net network in the field of gravity change data, highlighting the method’s value in the real-world environment.

Geophysical forward modeling serves as a fundamental theoretical approach for characterizing subsurface structures and material properties, essentially involving the computation of gravity responses at surface or spatial observation points based on a predefined density distribution. With the rapid development of data-driven techniques such as deep learning in geophysical inversion, forward algorithms are facing increasing demands in terms of computational scale, observable types, and efficiency. To address these challenges, this study develops an efficient forward modeling method based on voxel discretization, the enabling rapid calculation of gravity anomalies and radial gravity gradients on multiple observational surfaces. Leveraging the parallel computing capabilities of graphics processing units (GPU), together with tensor acceleration, Compute Unified Device Architecture (CUDA) execution, and Just-in-time (JIT) compilation strategies, the method achieves high efficiency and automation in the forward computation process. Numerical experiments conducted on several typical theoretical models demonstrate the convergence and stability of the calculated results, indicating that the proposed method significantly reduces computation time while maintaining accuracy, thus being well-suited for large-scale 3D modeling and fast batch simulation tasks. This research can efficiently generate forward datasets with multi-view and multi-metric characteristics, providing solid data support and a scalable computational platform for deep-learning-based geophysical inversion studies.

The Eastern Himalayan Syntaxis (EHS), which is located at the southeastern edge of the Qinghai–Xizang Plateau, is a key region for understanding mountain‐building and subduction processes. Bouguer gravity anomalies derived from the Earth Gravitational Model 2008 free‐air anomaly data following topographic corrections, were analyzed. Wavelet multiscale decomposition and normalized full gradient methods were employed to analyze the data, revealing complex crustal density structures and fault depth characteristics in both the lateral and vertical dimensions. The results indicate significant lateral density variations within the crust, attributed to the wedging of the Indian Plate and partial melting of deep materials. The article also revealed several major faults that extend to 40 km in depth that potentially play crucial roles in regional stress release and material migration between the crust and mantle. Additionally, smaller faults at middle to shallow depths may influence the near‐surface stress distribution and the evolution of shallow crustal structures. The analysis of Bouguer gravity anomalies in the EHS provides new insights into the complex crustal density architecture of the EHS and improves our understanding of mountain‐building processes and subduction dynamics in this geologically significant region.