The Tazhong paleo-uplift is one of the most important hydrocarbon enrichment areas in Tarim Basin. After years of exploration, many reservoirs have been discovered in the deep Paleozoic carbonate karsted rocks. Current research suggests that tectonic evolution and faulting movements have an important impact on reservoir development and hydrocarbon accumulation. But the reservoir prediction reliability is greatly compromised due to the lack of appropriate technical means to ascertain the geometric and kinematic characteristics of fault system more accurately. We solve this difficulty to a great extent by integrating a series of new applicable technologies.

Firstly, the existing seismic amplitude cube was further processed with the technology of "vector high fidelity signal enhance" which enhanced the natural bandwidth of seismic imaged volumes by simultaneously optimizing the signal and frequency content. Then more details of large-scale faults in the target interval were displayed. Secondly, a variety of fault detection methods were used to trace all possible large-scale faults including conventional edge-detection methods such as Variance cube and new method of "end-to-end convolutional neural network (CNN)". Thirdly, "auto fault patch extraction" was performed and the extracted fault patches were combined with some manual work to make sure they followed certain structural patterns according to the regional geological knowledge. Fourthly, the kinematics of the mapped fault system including its grouping and staging were carefully explored based on "discontinuity stability analysis" and traditional geological analysis. The former was based on Mohr-Coulomb criterion considering friction angle and cohesion of the faults, their attitudes and the 3D paleo-stress fields corresponding to different tectonic events. And the latter mainly focused on vertical layering of seismic structures considering vertical variations of fault patterns and strata attitudes, offset formations, stratigraphic unconformities, growth formations and fault patterns in plan view, etc.

It was suggested that the karsted carbonate reservoirs in some regions were mainly related to 3 types of fault damage zones, i.e. strike-slip fault damage zones, reverse fault damage zones and the hybrid ones. Among them, the parts with larger deformation, pull-apart basin, En echelon structure, faulted anticline, and superimposed elements were the most favorable belts to target high performance producers.

It is the first time to integrate so many applicable new technologies to ascertain more clearly the geometry and kinematics and the fault system in the deep karsted carbonate reservoirs which are otherwise blurred on seismic images. This can be applied directly for optimized well placement and can also be referred to for similar industrial projects.

You can access this article if you purchase or spend a download.