Wavefield Tomography Based on Local Image Correlation

Abstract

The estimation of a velocity model from the acquired data is a crucial step for obtaining a high-quality image of the subsurface. Velocity estimation is usually formulated as an optimization problem, where an objective function is defined to measure the quality of the image, and its gradient is used to update the model. The objective function can be defined in the data space (as in full-waveform inversion) or in the image space (as in migration velocity analysis). The latter leads to smooth objective functions, which are convex in a wider basin about the global minimum. Nonetheless, migration velocity analysis requires to image the entire survey before assessing the consistency between the velocity model and the data. We present a new measure of velocity errors and a new objective function based on the local correlation of the migrated images that extracts velocity error information directly in the image domain without computing common-image gathers, thus reducing the computational cost. Because of the large size of the problem, gradient-based method (such as conjugate gradient) are used in the optimization procedure. The gradient is computed using the adjoint-state method. We illustrate the method with a simple synthetic example.