Abstract: The conventional advance detection of tunnels using magnetic resonance technology can only obtain onedimensional depth imaging, making it difficult to accurately reflect the spatial distribution characteristics of nonlaminar waterbearing structures and causing safety hazards in tunnel construction. Therefore, using magnetic resonance technology, a twodimensional imaging method based on complex envelope inversion for tunnel rotation detection of water hazards is proposed in this study. First, the complex envelope inversion formula and twodimensional complex envelope inversion expressions are derived for tunnel space based on the detected envelope data of the tunnel using magnetic resonance technology. Then, the water content and relaxation time are obtained by calculating the optimal solution by Gikhonov regularization, Gaussian Newton iteration method, and smoothing constraints. Finally, a case study is conducted on the Maizhishan tunnel of the BaojiLanzhou highspeed railway and the process of magnetic resonance rotation detection is simulated. The obtained imaging results of waterbearing structures in the tunnel roof, floor, left and right sidewalls are numerically calculated. Results of the analysis are the following: (1) The twodimensional imaging results of accurate water content and relaxation time can be obtained in advance detection using magnetic resonance technology. (2) The twodimensional imaging detection results using magnetic resonance are in good agreement with the actual engineering conditions, which can precisely locate the positions of water hazards and predict the spatial distribution characteristics of nonlaminar waterbearing structures, and are reasonable and operational.

Key words: tunnel, magnetic resonance technology, rotation detection, water hazard, high precision imaging