The DAK Formation Evaluation Model for the Permian Basin Clearfork
E. A. Clerke, K. W. Williams, and L. A. Pearce
Abstract: The Clearfork DAK mineralogy model is a new formation-evaluation procedure developed specifically for the Permian basin Clearfork formation. Using the physical, geochemical, and mineralogic properties of the Clearfork formation, a natural log response analysis frame was developed. The result is a computation of the formation mineralogy and porosity, both of which are accurate and valuable for describing reservoirs with this variable lithology matrix.
The low-permeability and low-porosity Clearfork formation is an economic target for infill drilling and secondary recovery because of its thickness (400 m; 1,300 ft) and discontinuity. The thick Glorieta-Clearfork interval is actually a series of stacked shallowing upward carbonate shelf sequences and intraformational clastic-rich seals.
Clastic material (silt) has periodically been deposited along with the anhydritic-dolomite of the Permian Clearfork formation. This mixed lithology results in a long-standing problem in formation evaluation. The clastic-rich intervals are important because they are commonly reservoir seals and are a time stratigraphic marker. A regional geologic model is used to relate silt mineralogy and log response to the large-scale clastic sediment transport process. Our regional geologic model is tested using core data from the Central Basin Platform TXL and Central Robertson Clearfork Unit and the Northern Platform South Wasson Clearfork Unit.
Applying the DAK formation model results in improved infill assessment and waterflood development. The computed mineralogy and porosity are in excellent agreement with core data. Further, the log-derived mineralogy-facies is a major input to reservoir description, identifying chronostratigraphic supratidal clastics for sequence and flow unit correlations. The application of this model to Clearfork reservoir description results in
NMR in Partially Saturated Rocks: Laboratory Insights on Free Fluid
Index and Comparison with Borehole Logs
C. Straley, C. E. Morriss, W. E. Kenyon: Schlumberger-Doll Research
J. J. Howard: Phillips Petroleum
Abstract: Laboratory nuclear magnetic resonance (NMR) measurements were made on low-permeability clay-rich sandstone samples from formations in western Canada (five wells) and from the Wilcox formation (one well) of the US Gulf Coast. NMR measurements were made with the samples fully water saturated and at residual water saturation obtained by centrifuging. Measurements at various air/water and kerosene/water saturations were made on additional samples.
The measurements were of longitudinal relaxation; i.e., rate of growth of magnetization along the direction of an applied field. The measurements were converted to T 1 distribution curves; i.e., fitted to a sum of many specified single exponentials. In this form, the NMR response at partial saturation is closely related to the physics of capillary drainage and, as a result, the amount of water removed by centrifuging can be estimated in a simple way from the water-saturated T1 distribution curve.
The free-fluid index (FFI) derived from borehole nuclear magnetism logs recorded in five wells is compared directly with laboratory-measured values of centrifugeable water for the corresponding core samples, and the values agree well. The FFI values can be readily understood by comparison with laboratory-measured T1 distribution curves and the underlying physical principles.
Interpretation of Unfocused Resistivity Logs
Walter W. Whitman: Colorado School of Mines
Abstract: Unfocused resistivity tools, in the form of the electrical survey (ES) suite of short and long normals, and lateral log tools were used for hydrocarbon exploration from the late 1920s until the late 1950s in the United States and until the mid 1960s internationally. Software has been developed to simulate the response of the ES suite and to interpret data from these logging suites for bed boundaries and formation resistivity. The software models the borehole effects on the log response and also models surrounding bed effects up to 120 ft (37 m). These unfocused effects can have a significant impact on the log response. The software interprets the log by mathematically modeling the current flow in the borehole and in the geologic formations. After inverting the logs individually for an earth model of bed boundaries and resistivities, an overlay of the associated earth models can be used to indicate such features as invaded zones, impermeable zones, gas/water and oil/water contacts, and bed boundaries with a resolution of 2 to 3 ft (0.75 m). The interpretation has an automatic initial guess for an earth model; no user interaction is necessary, which leads to superior consistency, reliability, and objectivity. The simulation is useful for testing reservoir hypotheses such as depth of invasion, which can have a substantial impact on the analysis of hydrocarbon reserves.
Application of Kruskal Multidimensional Scaling (MSDC) to Rock Type
Indentification from Well Logs
Vaclav Matyas: Consultant, Czech Republic
Abstract: The technique of Kruskal nonmetric multidimensional scaling (MDS) is used to identify rock types from wireline logs. The method is illustrated on a log set from the Slusovice-1 well (located on the southeastern slopes of the Bohemian Massif). Based on other geological information, a relatively short (150-m) interval in this well of sandstone, coal, dolomite, clastics overlies the crystalline basement, which is fractured to varying degrees. MDS plots showed good correspondence of clusters with different rock types. Depth logs of lithology types were drawn based on the the MDS results.