Structural information in subsurface seismic images is critical for reservoir delineation, reserve estimation, and well planning. However, by its very nature, it is also uncertain. One cause of the image uncertainty is the migration velocity model that directly affects the position of migrated events, both laterally and vertically. (The term "velocity" is meant in the broad sense; i.e., it also includes the anisotropy parameters.) We present a method that accounts for uncertainties in a velocity model estimated by tomography and translates them into the migrated domain. Standard-deviation attributes on target horizon positions or layer thicknesses are extracted. The method includes quality controls for validating the estimated uncertainty attributes before integration with other downstream or interpretative information. The method is demonstrated on a North Sea area covered by data from multiple seismic surveys. We observe that uncertainties increase with model complexity or depth and decrease as the illumination diversity increases. The computed uncertainty maps constitute a valuable source of information for hierarchizing (both qualitatively and quantitatively) different areas in the survey. For the purpose of reservoir risk analysis, we combine our technique with other information (e.g., interpretation uncertainties) to map how uncertainties in the depth of the structural spill point impact the gross rock volume (GRV) estimation of a reservoir.
Tomography algorithms using gridded model description and ray tracing have made continuous progress in terms of resolution and efficiency. However one strong limitation is the difficulty to recover strong velocity contrasts encountered in presence of salt bodies, chalk, basalt, carbonates… The conventional solution for velocity model building in such a context is to proceed in a top-down manner from one velocity contrast to the next one. In such a layer stripping approach velocities and horizons are updated layer after layer recursively from top to bottom. Such a workflow is time consuming and prone to velocity errors being propagated into deeper layers as the model building progresses. We present here a solution to remedy these drawbacks. Our solution involves a non-linear tomographic approach combining dense dip and residual move-out picks with horizons describing the main velocity contrasts. While dip and RMO picks are used to update 3D velocity grids inside each layer by non-linear slope tomography, the picked horizons describing layer boundaries are kinematically de-migrated and re-migrated recursively from top to bottom to reposition the major velocity contrasts after each velocity update. We present applications of the method to a marine North Sea dataset and to a land dataset with salt structures and compare the results with the layer stripping approach.
Ray based migration velocity analysis from pre-stack seismic reflection data is based on the characterization of the migrated reflected events by their position, dips and residual move-out. Such approaches update the depth velocity model through an optimization process, where the residual move-out of the picked migrated events is minimized while obeying some regularization constraints related to the depth or to the shape of some horizons or to the smoothness or structural conformity of the velocity field. We propose to introduce an additional term in the cost function involving the dip of kinematically migrated locally coherent events. The velocity is then updated to match the expected dip of the re-migrated offset-dependent events. We develop here the conceptual aspects of this approach within the frame of non-linear slope tomography and present a first application, on a North Sea dataset, to the characterization of very shallow channels creating pull-up and pull-down effects in deeper parts of the migrated image. Due to the very limited offset range of the residual move-out picks in shallow subsurface, these effects could not be solved by residual move-out based tomography. We demonstrate that the introduction of the dip-constrained inversion allows the correction of these pull-up and pull-down effects, resulting in improved depth imaging.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.