Multicomponent Seismic Interpretation and its Relation to Production Response in the Leonardian Drinkard Formation of Vacuum Field, Lea County, New Mexico

RAINES, MICHAEL A., Kinder Morgan CO2 Co., L.P., Midland, Texas and TOM L. DAVIS, Colorado School of Mines, Golden Colorado

    Vacuum Field sits along the margin of the Northwest Shelf of the Permian Basin in central Lea County, New Mexico. Between 1995 and 1999, a portion of this area was the study site for 3D/3C (multi-component) and 4D (time lapse) seismic investigations concentrated on the San Andres Formation, and led by the Colorado School of Mines’ Reservoir Characterization Project industry consortium (RCP). While RCP was investigating the broader geological setting of the study area, certain seismic anomalies became apparent in the Drinkard and Abo formations, especially in the shear-wave dataset. Static reservoir characterization, using both shear and compressional data, was then applied to Drinkard and Abo Formations. These investigations established an empirical relationship between seismic response and production data. Further, the seismic and production information can be explained by considering reservoir porosity development and fracture distribution.

    Abo deposition set the stage for future geometric relations with a sharply defined reef-enhanced shelf edge. During Drinkard deposition, relative sea level rise was slow enough to allow for shelf edge stabilization along the underlying reef trend. Laterally shifting, highly localized, and interfingering facies tracts (including patch reefs, which are now productive at Vacuum Field) were developed at this time. Today, enhanced matrix porosity associated with the patch reefs and fracture distribution (both open and healed) affect production characteristics in the field.

    Production response was somewhat unclear until the data was broken into two categories: Instantaneous and Long Term. Initial Potential (IP) represented the immediate deliverability of each well, and Average Production over the first six months represented the long term storage space available to each well. Both datasets mapped distinct (but different) patterns. Based on a loosely controlled geologic model constructed from the limited core data available, long term production apparently corresponds to patch reef areas, and higher IP areas correspond to regions of open or partially open fracture sets.

    Two seismic datasets exhibited patterns very similar to the production maps. Anisotropy analysis (a comparison of fast and slow shear wave birefringence velocities) matched the IP distribution. A map of the ratio of compressional velocity (VP) to fast shear wave velocity (VS1) gave a response similar to the average production data map. The implication, then, was that VP/VS1 ratio mapping also revealed higher porosity areas, and anisotropy analysis also highlighted regions of open fractures.

    Multi-component static characterization can be a powerful tool, once calibrated to the geology of a specific field. If this type of data were available early in the life of a field, an optimized recovery scheme could be developed, and a specific development drilling plan could be implemented, saving hundreds of thousands of dollars while optimizing field production.