Abstract: Uranium possesses both radiological and chemical toxicity, and its mining, especially through in-situ leaching (ISL) processes, has raised widespread environmental concerns, particularly regarding the contamination of groundwater systems. To accurately assess the potential risks of uranium pollutant migration in sandstone aquifers following the decommissioning of CO₂+O₂-based ISL uranium mines, it is essential to investigate the transport behavior of uranium under realistic subsurface geochemical conditions, especially in the presence of bicarbonate ions (\mathrmHCO^-_3 ), which play a key role in enhancing uranium mobility through complexation. However, direct migration experiments using undisturbed sandstone cores are often constrained by significant technical difficulties, including challenges in core preparation, experimental operation, and limited accessibility to dynamic migration data. To address these limitations, the comprehensive batch adsorption experiments were first conducted using sandstone particles under various \mathrmHCO^-_3 concentrations in this study. Both isothermal and kinetic adsorption tests were performed, and the fitting results show that uranium adsorption behavior best follows the Langmuir isotherm model and the pseudo-second-order kinetic model, with correlation coefficients (R²) exceeding 0.97. Subsequently, a three-dimensional uranium transport model incorporating adsorption effects was developed using the lattice Boltzmann method (LBM) with the D3Q19 lattice configuration. A high-resolution digital core model was reconstructed from micro-CT scans of the undisturbed sandstone sample, capturing the actual pore structure as the simulation domain. To verify the accuracy of the proposed model, the simulation results were compared with the uranium migration data that were obtained from physical core migration experiments conducted under different \mathrmHCO^-_3 conditions. The results demonstrate that uranium migration is significantly accelerated in the presence of \mathrmHCO^-_3 and notably delayed in its absence. The simulation outcomes are in strong agreement with experimental observations, confirming the validity of the derived parameters and the model’s predictive capability. Additionally, the results indicate that even at the core scale, uranium migration is still controlled by particle-scale adsorption mechanisms. The lattice points in contact with the pore surface follow both the Langmuir isotherm and the pseudo-second-order kinetic model, suggesting that micro-scale adsorption behavior has a significant impact on macroscopic transport processes. In conclusion, the digital core-based three-dimensional uranium transport model developed in this study serves as an effective and reliable supplement or alternative to traditional core-scale migration experiments. It also provides a digital twin framework for predicting uranium migration behavior in aquifers and assessing groundwater contamination risks during and after the closure of ISL uranium mining operations. This approach offers strong theoretical and methodological support for future environmental assessments and remediation planning.