Advanced CRS Application by Model Based Attribute Guidance
In time domain the NMO stacking technique has been dominant for many decades. However the simple subsurface model in combination with a user derived velocity model has been very inaccurate ever since. An alternative and much more accurate approach is the CRS methodology (Common Reflection Surface) which assumes subsurface reflector elements with a certain dip, depth and curvature. Here, the corresponding CRS operator is not limited to a single CMP gather, but it collects the reflection energy of a subsurface element from all contributing traces, including neighboring CMPs of the imaging location. In general, the CRS technology is used to regularize and interpolate reflection seismic data in one step and to improve the overall signal-to-noise ratio. The wavefield attributes that describe the traveltimes of a local seismic event are being computed for each seismic sample automatically and fully data driven. The main tools to steer the CRS impact on the data are the rms velocities, which are used as a guide function, and the CRS aperture sizes.
A new implementation of the CRS technology is a model based/driven computation in data cases where the general dip trends of the recorded wavefield are sufficiently known. Either a simplified layer based model, but also a more complex geological model can be used as a guide for the automatic dip search. By using a structural model as guide function, the coherency based dip search becomes much more stable, as now especially linear noise can be eliminated much more efficiently. Additionally, areas with low coverage or poor signal quality can be further improved by limiting the search for coherent signals to a small-range dip which is directly linked to the known subsurface structure. Depending on the data quality and accuracy of the geological model, a narrower or wider dip search will be applied. This new technology works very good in structural complex areas like parts of Latin America (e.g. mountainous areas of Bolivia, Colombia and Mexico). Examples of standard processing, conventional CRS processing and model guided processing results will be shown.
AAPG Datapages/Search and Discovery Article #90260 © 2016 AAPG/SEG International Conference & Exhibition, Cancun, Mexico, September 6-9, 2016