Imaging complex near-surface structures in Yumen oil field by joint seismic traveltime and waveform inversion

Abstract

The first-arrival traveltime tomography is a standard approach for near-surface velocity estimation.However, it cannot resolve complex near-surface structures and will produce a smooth velocity model with low resolution.Early arrival waveform inversion is a robust tool for imaging the near surface structures, but it requires a good initial model to avoid cycle skipping between the predicted and observed data.Furthermore, waveform inversion requires substantial computation efforts.Therefore, we present joint seismic traveltime and waveform inversion method, and we expect the joint inversion method retains the advantages of both traveltime inversion and full waveform inversion and overcomes their respective drawbacks at the same time.The objective function includes both the traveltime and waveform misfit.At each iteration, the traveltimes are calculated by wavefront raytracing, and the waveforms are computed using a finite-difference method.The nonlinear optimization problem is solved by the conjugate gradient method.We apply the joint inversion method to study complex near-surface area where shallow overthrust and rugged topography present a significant challenge for applying traveltime inversion and waveform inversion alone.We test synthetic data to verify the advantages of the joint method, and then apply the method to a 2D dataset acquired in Yumen Oil field, China.The inversion results suggest that the joint traveltime and waveform inversion helps constrain the very shallow velocity structures and also resolve complex overthrust with large velocity contrasts.