Pressure prediction from seismic data: implications for seal distribution and hydrocarbon exploration and exploitation in the deepwater Gulf Of Mexico

Abstract

Pore-pressure prediction before drilling is critical for several reasons. It is required to assess “seal” effectiveness, map hydrocarbon migration pathways and analyze “trap” configuration and geometry of a prospective basin. Furthermore, it aids in the well planning process by providing proper casing and mud program design which can help prevent dangerous “blow-outs”, lost circulation of drilling fluids and stuck pipes. The conventional techniques for pressure prediction are limited by two factors: establishing a “normal” trend of an attribute (usually a porosity indicator) and a set of calibration curves relating “overpressure” to deviation from the normal trend of that attribute. Thus, these techniques cannot be used in rank wildcat areas and areas such as the deep water environment (water depth greater than 330 m) of the Gulf of Mexico where normal compaction trends are usually non-existent. At BP, a new technique for pressure prediction has been developed. The essentials of this technique are as follows. It uses a proprietary transform that relates velocity to effective stress (defined as the difference between overburden and pore-pressure), temperature and gross lithology directly. It takes into account the major causes of overpressure mechanisms: undercompaction, clay dehydration and transformation, buoyancy and charging of fluids in dipping, permeable beds. It does not require local calibration and predicts effective stress directly, which is the most fundamental quantity for pressure prediction. In this paper a brief description of this technology is presented together with several examples from the deepwater environment of the Gulf of Mexico. Applications are made in 1-D, 2-D and 3-D and have enabled explorationists to define “seal” failure risks in deepwater prospects. Drilling experiences have shown that this technology can predict pressures to within 0.5–0.75 pounds per gallon (ppg) at target depths, provided the “low-frequency” trends of seismic interval velocities are of good quality and “close” to well velocities to within 5–10%. The quantitative reliability of the method depends on two factors: availability of high quality seismic velocity data and an understanding of the rock properties. The vertical (temporal) resolution is limited by the available bandwidth of the seismic velocity data whereas the spatial resolution is dictated by the acquisition parameters and the frequency of velocity analysis (CDP spacing and panels for analysis).