Constrained 1D joint inversion of seismic surface waves and P-refraction traveltimes

We present a joint inversion scheme that couples P-wave refraction and seismic surface-wave data for a layered subsurface. An algorithm is implemented with a damped least-squares approach. The estimated parameters are S- and P-wave velocities and layer thicknesses, while densities are assumed constant during inversion. The coupling is both geometric and physical: layer thicknesses are the same for Sand P-wave velocity profiles and P-wave velocities enter in both forward algorithms. Sensitivity analysis, performed on synthetic data, reveals that surface-wave dispersion curves can be sensitive also to P-wave velocity of some layers (especially for Poisson’s ratio values smaller than about 0.35), allowing synergic resolution of this parameter. Applications on both synthetic and field data show that the proposed approach mitigates the hidden layer problem of seismic refraction and leads to more accurate results than individual inversions also for surface waves. Additional constraints on the objective function on a priori Poisson’s ratio values allow unrealistic and not admissible VP and VS values to be avoided; such constraints were applied in one field case considering the a priori information available about water-table depth. It is also shown that estimation of porosity can help the selection of the proper constraint on a priori Poisson’s ratio.