Surface Nuclear Magnetic Resonance Inversion Method Based on Quasi-2D Adaptive Laterally Constrained Inversion

Surface Nuclear Magnetic Resonance (SNMR), as a critical high-resolution detection technology for groundwater characterization, offers unique advantages in geothermal resource exploration and hazardous water source identification. Addressing the demand for efficient detection of complex groundwater systems, engineering practices urgently require inversion imaging methods that unify high precision with computational efficiency. However, traditional single-point inversion methods struggle to balance the trade-off between spatial resolution and computational efficiency when resolving laterally continuous variations in groundwater systems. This study innovatively establishes an adaptive constraint mechanism based on groundwater continuity characteristics. By introducing Spearman’s rank correlation coefficient to quantitatively evaluate the geological relevance between adjacent survey points, a dynamically adjusted lateral constraint matrix is autonomously generated along the survey line. This technique implements spatially differentiated constraints according to the groundwater continuity distribution. Numerical simulations demonstrate that, compared to conventional imaging techniques, our proposed quasi-2D adaptive imaging strategy effectively enhances imaging resolution while maintaining computational efficiency.