Presenter
Shuvajit Bhattacharya, Ph.D.
Research Associate
Bureau of Economic Geology
Jackson School of Geosciences
The University of Texas at Austin
Description
Petrophysical evaluation is fundamental to unconventional resource assessment and development. The study discusses some petrophysical challenges, methods, and results from the Wolfcamp and Spraberry formations in the Midland Basin. The major petrophysical challenges are well-log availability/quality/vintage, the need for location/property-specific core data, the variation of formation character related to the scale of heterogeneities, formation water salinity, overpressure, geomechanics, and lack of integration, etc. Based on our well-log evaluation and comparison to core data, we recognize the need for well-log-based multi-mineralogy optimized solver models for such rocks compared to deterministic and cutoff value-based approaches that are more appropriate for conventional reservoirs. We use a rock-type-based and thermal maturity-adjusted petrophysical log modeling approach to enhance core-log calibration quality. As a part of the workflow, we have incorporated machine learning and deep learning at several steps, for example, log QA/QC, rock typing, and rock property prediction. Results show the heterogeneities of the Wolfcmap and Spraberry formations, in terms of lithology, porosity, Sw, TOC, and pore pressure. The variation of thermal maturity at a basin scale greatly impacts kerogen properties and corresponding well-log expressions that need to be accounted for deriving other petrophysical properties, such as porosity. These formations are a mixture of multiple minerals, including quartz, feldspar, illite, chlorite, calcite, dolomite, and pyrite. Although Wolfcamp and Spraberry have low porosity, some Spraberry sections (e.g., Jo Mill) behave mostly like a conventional reservoir. In general, basal Wolfcamp has higher pore pressure, whereas Dean and Spraberry are mostly in the hydrostatic and underpressure conditions.
