Answers are grounded in this paper and cite their sources.
Summary The non-repeatability effects, such as differences in sources and receivers positions and water layer velocity variation, affect the processing of 4D surveys used in reservoir monitoring. To deal with these effects, we propose a new method which combines the double-difference full waveform inversion (DDFWI) with the non-parametric residual statistics FWI (NPR-DDFWI). To show the effectiveness of our technique, we simulate scenarios with perfect repeatability and with non-repeatability effects and perform inversions using the NPR-DDFWI and DDFWI using L2 and L1 norms. We find that our proposal is more robust than L2 and L1 norms, recovering the geometry and magnitude of time-lapse velocity variation at the reservoir in all non-repeatability scenarios.
No takes yet. Share an insight, caveat, or question.
Silva et al. (2023) studied this question.