Sunday, March 23, 2014

Miss Lime Play seismic detail

Detailed seismic interpretation requires custom tools and a deep geological knowledge of the section under study.  Our work in the Mississippian of NW Arkansas and NE Oklahoma has progressed to the point that we can venture tentative interpretations on a reservoir level to assist oil and gas exploration in the Miss Lime Play. 

In Figure 1, part of a 2D line extracted from a 3D seismic survey in NE Oklahoma.  The Miss interval is about 44 ms thick and well-imaged on this survey with a dominant frequency of 57 Hz at the Miss level.  Top of the Miss on the left side of the image is at 465 ms, represented by the strong positive reflection event.  This is a regional unconformity surface between clastic basal Penn rocks and carbonate rocks of the Mississippian.  Note the gradual decay of amplitude toward higher trace numbers, until the Penn/Miss interface is nearly transparent.  The wiggle trace displays show detail near traces 166 and 238. 

Top of the Miss in this area is know to have two scenarios; Penn clastics can abut either Miss Dense limestone or Miss Chat. Chat is a local term for a zone of deeply weathered rock associated with subarial erosion at the Penn/Miss unconformity.  The name derives from early driller observations that enchanting the zone caused drill pipe to chatter against hard chert fragments.  The Chat is generally considered to be a deep paleosol, probably like a re-litified version of the typical Miss Boone Regolith seen in NW Arkansas and other outcrop areas (Figure 2).

Another important form of chert in the Miss carbonate section as seen in outcrop is Tripolite (Figure 3).  Trip is low density, high porosity (up to 35%), and mechanically weak (low velocity) rock. 

A custom interactive Mathematica interface has been developed at UArk for manual waveform inversion of reservoir-level seismic data (Figure 4).  Unlike automatic inversion schemes (such as full waveform inversion or well-driven impedance inversion), manual inversion uses human expertise to fine-tune results in the presence of uncertainty in rock properties, geological structure, and data noise levels.  All such tools are needed to address the spectrum of scales encountered in modern petroleum exploration.  

Figure 4 shows one possible fit for trace 236 of Figure 1, indicating  100 ft of Chat at the unconformity and a 20 ft interior low impedance layer that may be Trip.  Figure 5 gives a less satisfactory fit.  

Figure 1. Seismic data from 3D survey in NE Oklahoma.  Image shows data window around the Miss interval (465-505 ms adjacent to time label axis). Lower plots show 100 ms wiggle plot detail of the Miss for a few traces. 
Figure 2. Miss Boone reggolith as seen in Hindsville Quarry of NW Arkansas.  Limestone is being progressively leached out of the section leaving characteristic red clay and angular chert fragments.  The Miss Chat encountered in the subsurface of NE OK may be this same kind of material lithified in the early Pennsylvanian period. 
Figure 3. Mississippian Boone formation road cut near Bella Vista AR showing soft Tripolite at the base and dense lime with interbedded hard chert above. This exposure is about 60 ft out of the total 300 ft Boone section, but demonstrates Trip occurring beneath dense lime. 

Figure 4. Interactive fit to trace 236 of Figure 1 assuming the top Miss is at 10 ms event.  Upper graph shows field data (red) and synthetic trace (black).  Layer numbers correspond to basal Penn (1), Miss Chat (2,4), Miss Dense Lime (3,5), and older rocks (6).  Layer properties are averages from full wave sonic and density logs.  Penn and older rock properties are averages over 100 ft interval adjacent to the Miss.

Figure 5. Alternate interpretation of trace 236 that is less satisfactory in associating the 50 ms peak with the top of Miss, in disagreement with Figure 1. This illustrates the inherent non-uniqueness of waveform matching, a issue that can be minimized by geological expertise.  

No comments: