Least-squares reverse time migration with velocity errors

Abstract

An accurate migration velocity model is required for reverse time migration (RTM) to correctly predict the kinematics of wave propagation in the subsurface. Leastsquares reverse time migration (LSRTM), which aims to match the amplitudes of the modeled data with the observed data in an iterative inverse procedure, is more sensitive to the accuracy of the migration velocity model. If the migration velocity model contains errors, the final migration images will be defocused and incoherent. As a partial solution, we utilize an LSRTM scheme based on the extended imaging condition, which is called as leastsquares extended RTM (LSERTM). It is well accepted that LSERTM can fit the observed data regardless of the accuracy of the migration velocity model. We further explore this property and find that after stacking the extended migration images along the subsurface offset axis within properly selected ranges, we can obtain an image with better coherency and focusing than the conventional LSRTM. We demonstrate the efficacy of our method with numerical examples on a Salt-like model and the Marmousi model.