Least-squares reverse time migration (LSRTM) has emerged as a powerful tool for generating superior subsurface images. There is a growing trend for LSRTM moving from merely producing stacked images toward generating angle gathers. The primary challenge in creating LSRTM angle gathers lies in the precise implementation of demigration. Conventional methods, which conduct demigration based on stacked images or angle gathers, often suffer from inaccuracy or computational inefficiency. Here, we develop an alternative method for generating LSRTM angle gathers. This method uses the prestack approach and the amplitude-preserving migration operator to construct a proxy that replaces the angle-dependent scattering potential in the demigration operator with the shot-dependent scattering potential, thus leading to a precise and convenient implementation of the demigration process. We then use angles calculated via Poynting vectors to map extracted imaging values or updates at each imaging time to the angle domain, thereby producing angle gathers. This strategy effectively eliminates the reliance on stacked images or angle gathers in the implementation of the demigration process, and it also provides flexibility for generating angle gathers on a relatively sparse grid or focusing only on specific target areas. The effectiveness of our algorithm is verified through a synthetic data set and a field data set. The results show that, even after a few iterations, the presented method produces high-quality images and angle gathers with enhanced amplitudes, improved illumination, more continuous events, and higher resolution.

You do not have access to this content, please speak to your institutional administrator if you feel you should have access.