Abstract

Accurate determination of remaining oil saturation is critical for the successful implementation of enhanced recovery programs planned for the Yates field. Determination of oil saturation in this strongly heterogeneous, mixed-wet, and vuggy carbonate reservoir, however, is difficult. Conventional methods generally fail to produce robust results because of the variations in the Archie exponents m, and n, both in space and time. The saturation variations originate from the double displacement processes employed for improved recovery, which have generated a sizable gas cap and a moving oil/water contact.

Because of its direct reservoir fluid characterization capabilities, NMR logging was earlier identified as a potential alternative for the direct determination of remaining oil saturation at the Yates Field. A series of MRIL logs was run in 1998, where the main objective of the logging program was to exploit the diffusion contrast between oil and water. Results from this logging program were unsatisfactory because the apparent T2 contrast between the oil and water phases was not sufficiently large to allow an accurate determination of the various fluid fractions.

An extensive modeling study was then initiated to study the feasibility of NMR logging for oil saturation measurements. Existing MRIL logs and limited laboratory data were used as input into the models. Modeling showed that interecho times in excess of 6 ms. could increase the apparent T2 contrast between the oil and water phases. However, modeling also showed that such an increase in T2 contrast would be accompanied by severely lowered T2 resolution, which would in turn increase the uncertainty in computed saturations. Further modeling showed that the combination of higher magnetic field gradients and long interecho times could yield sufficient T2 contrast without compromising T2 resolution. It was then decided to modify an existing MRIL-C tool to achieve these objectives.

An observation well that was included in the 1998 program was relogged in late 1999 with the specially configured MRIL-C tool. Interpretation of the data acquired with this special tool and the remaining oil saturation determined from an NMR only technique agreed very well with the existing field model and petrophysical data. Expansion of the pilot study to a multiple-well evaluation program is currently being planned.

Introduction

The Yates Field lies near the southern termination of the Central Basin Platform, between the twin Permian Basins of West Texas1. It produces primarily from the middle Permian-age San Andres Formation, which comprises a series of mostly massive dolomite layers. The reservoir has a complex diagenetic history, including extensive karstification and fracturing which have made it a world class deposit.

The average porosity in the field is 15%, and the predominant pore textures vary from sucrosic (intercrystalline) to moldic, from vuggy to cavernous. While the average matrix permeability is in the order of 100 md, reservoir flow is dominated by fracture flow. Fracture porosity averages 2% with permeabilities exceeding 1 darcy.

After many years of production, a secondary gas cap has evolved in the reservoir, and a moderately strong aquifer has invaded from below. These natural forces have led to the current production method of double-displacement: displacing oil with water from below, followed by artificial inflation of the gas cap to force gravity drainage. Other enhanced production methods that have been utilized in the field include polymer-augmented water flood, extraneous gas injection, CO2 flooding, and a current steam injection pilot.

This content is only available via PDF.
You can access this article if you purchase or spend a download.