Geological structure guided well log interpolation for high-fidelity full waveform inversion

Full waveform inversion (FWI) is a promising technique for inverting a high-resolution subsurface velocity model. The success of FWI highly depends on a fairly well initial velocity model. We propose a method for building a good initial velocity model that can be put into the FWI framework for inverting a nearly perfect velocity structure. We use a well log interpolated velocity model as a high-fidelity initial model for the subsequent FWI. The interpolation problem is solved via a least-squares method with a geological structural regularization. In order to obtain the geological structure of subsurface reflectors, an initial reverse time migration (RTM) with a fairly realistic initial velocity model is conducted, and the local slope of subsurface structure is roughly calculated from the RTM image. The well log interpolated initial velocity model can be very close to the true velocity while containing a small velocity anomaly or oversmoothing caused by the imperfect velocity interpolation. The anomaly and oversmoothing effect can be compensated during the subsequent FWI iterations. We use a relatively simple-layered model and the more complicated Marmousi velocity model to demonstrate the applicability of the proposed approach. We start from a very smooth velocity model and obtain a nearly perfect FWI result which is much better than the traditional FWI result without the velocity interpolation. The migrated images from the RTM method using different velocity models are also compared to further confirm the effectiveness of the proposed framework. Regarding the field deployment, we suggest that future drilling of exploration wells can be seismic-oriented, which can help fully utilize the information of well logs for building initial subsurface velocity model and will facilitate a wide application of the proposed methodology.