Unique Approaches to Analysis of a Cyclic Shelf Dolomite Reservoir

Paul M. (Mitch) Harris
ChevronTexaco Energy Technology Company
6001 Bollinger Canyon Road D-1212, San Ramon, CA, 94583-2324 U.S.A.
Email: [email protected]

The McElroy Field, Central Basin Platform of the US Permian Basin, produces approximately 17,000 BOPD under a mature waterflood from the Grayburg Formation. Core studies document the stacking of numerous small-scale cycles within a larger-scale progradational motif, i.e., upward shallowing, for the main producing zone in the field. Dolograinstones are dominated by intercrystalline/intergranular porosity with a narrow size range of pore throats that results in most of the nearly 20% porosity being effective to oil flow. In contrast, dolopackstones are less porous and contain both moldic and intercrystalline/intergranular porosity. Their bimodal pore system results in a wider range of pore throat size and more ineffective porosity. See Figure 1.

Layering in this type of dolomite reservoir is stratigraphically controlled; therefore a thorough understanding of the stratigraphy is needed for determining reservoir architecture. Lateral and vertical shifts of facies must be understood to assess reservoir variation within layers, as facies boundaries generally equate with subtle variations in dolomite characteristics and associated reservoir quality. The typically fine crystalline dolomite results in low permeability reservoirs, but a long production history for the field attests to good connectivity. Meteoric overprint produced moldic and enhanced intercrystalline porosity leading to patchily distributed zones of higher porosity and permeability, whereas evaporite cementation and replacement further complicates the reservoir quality distribution. Because of its complexity and long production history McElroy field has been investigated in a great amount of detail, including the utilization of some unique approaches to reservoir analysis.

Crosswell Seismic

Geologic “ground-truthing” suggests that cross-well seismic data, when integrated with facies-based porosity models, adds value to reservoir characterization. The coincidence of reflectors with decreases in porosity or gypsum cement from whole-core analysis suggests that total porosity and mineralogy dominantly influence velocity. Reflectors correlate fairly well with major log variations; S-wave reflectors correspond almost exactly with increases in sonic velocity, resistivity, and bulk density, and decreases on the neutron log from high to low porosity (or gypsum). Although major stratigraphic boundaries (sequence boundaries and flooding surfaces) generally coincide with reflectors, lithofacies and small-scale depositional cycles do not relate directly to the seismic data. Comparing geostatistical porosity models directly to the seismic suggests that S-wave reflection images appear to be resolving lateral changes in porosity of less than 56 m but more than 15 m.

Log Facies

A significant result of the diagenetic complexity of the McElroy reservoir is that reservoir quality does not match original depositional facies. Both the seismic and log data respond to the same diagenetic overprint and its resulting petrophysical characteristics; therefore log facies derived from cluster analysis, rather than core lithofacies, better relate to the cross-well seismic. Many of the seismic reflectors correspond to vertical transitions between more and less porous log facies, which indicates the strong relationship between velocity and porosity. In addition, lateral variations in many of the positive-amplitude events can be tied to changes in porosity and differences in log facies between wells. See Figure 2.

Dual Porosity-Permeability Modeling

Heterogeneity is increased significantly in the central portion of McElroy field by thin high porosity-permeability vuggy zones. A method was developed to identify the vuggy zones on logs, create geostatistical models of porosity and permeability incorporating the vuggy zones, and characterize them in simulation models.

The method involved the following: (1) developing a log trace to identify zones of high vuggy porosity, (2) creating a detailed geostatistical model of total porosity using well log data, (3) creating a geostatistical permeability model based on total porosity, (4) creating a separate detailed geostatistical model of secondary porosity, and (5) superimposing exceptionally high permeability in areas of the permeability model defined by high secondary porosity. See Figure 3.

Figure 1