Elastic full-waveform inversion application using multicomponent measurements of seismic data collection

Abstract

Recent computational improvements allowed us to simulate elastic wavefields in a 3D manner and undertake the challenge of executing elastic full-waveform inversion (EFWI). The 3D SEG/EAGE overthrust synthetic data were used to run the initial tests, which included using all three components for the simulation. The inversion targeted two regions: the channel system and the overthrusted zone, which proved the effectiveness of EFWI to delineate geology in terms of [Formula: see text] and [Formula: see text] velocity fields. For the field data experiment to demonstrate the technologies, we elected to use a Gulf of Mexico ocean bottom cable data set, which allowed us to take advantage of relatively large offsets along with the 4C acquisition. The input data were minimally processed mostly through noise removal, and the initial model was a Gaussian smoothed version of grid tomography output, which is done by a prestack migrated gather flattening process. During EWFI, a multiscale approach was followed to ensure convergence, and the early stages of the [Formula: see text]/[Formula: see text] ratio were constrained by the mud rock-line ratio. When the last sets of inversions were executed, this constraint was eliminated to ensure the simultaneous update of the [Formula: see text] and [Formula: see text] velocity fields. The density was kept constant to keep the inversion at a simple level, which allowed us to draw essential conclusions. The velocity fields were validated through an imaging algorithm of the elastic reverse time migration, and the imaging shows clear structural improvements when inputting the inverted velocities in conjunction with the measurements. If full-waveform inversion can provide multiple earth parameters, the user can use the process to detect gas zones along with sand and shale content of the subsurface, which will further assist the drilling decisions. We achieved this by simulating the earth more accurately with the elastic wave propagation in the algorithms.