Evaluating Reservoir Architecture in the Northern Gulf of Mexico With Oil and Gas Chemistry

Abstract

Summary A post-development study was carried out at Bullwinkle Field in the Gulf of Mexico to develop and test geochemical methods for evaluating reservoir architecture, establish confidence in reservoir geochemistry technology, and provide a framework for surveillance work at Bullwinkle. In line with previous work, this study shows how fluid composition data can be used to obtain improved reservoir architecture interpretations. Geochemistry is a low-cost direct approach for determining reservoir architecture. When data are integrated with results from other tools and techniques, the final result is to improve decision making and production efficiency for both development and redevelopment projects. Introduction Throughout the life of an oil or gas field—from the time the count and placement of the first development wells are planned, to the last water-flood or sidetract/recompletion—a critical operational issue is the continuity and compartmentalization of reservoir units. Over the last 9 years, a number of papers have demonstrated that a variety of geochemical tools can be used to better define and constrain reservoir architecture interpretations,1–4 especially when resulting data are integrated with other geological, geophysical, and engineering information.1,5-7 The basic tenant applied in petroleum fingerprinting studies is deceptively simple—fluids in single reservoirs are compositionally homogeneous, while fluids in separate compartments show some degree of compositional heterogeneity. Differences are more important than similarities—much like pressure data, compositional homogeneity supports the idea of a continuous well-connected reservoir, although it usually does not prove it.5 Interpretations are complicated by a number of issues, including compositional grading, reservoir filling history, and insufficient time for mixing, but many published field studies support this general model.1-4,6,7 Geochemical fingerprinting technology has a number of advantages: as opposed to many static tools and methods, geochemical fluid data can be a direct indicator of fluid-flow barriers, much like PVT and production data,6,7,12 geochemical tools/methods are inexpensive, and most provide fast turn-around times, and the ability to work with core samples6–8 provides the opportunity for greater coverage, especially in areas with multiple stacked pay such as the Gulf of Mexico. Despite the applicability and promise of this new technology, its true value is not widely recognized in the industry. Consequently, it is not applied as much as it could be in appraisal, development, and re-development projects. An important reason for this lack of recognition/use is that a variety of issues and problems need to be addressed to increase our level of understanding about this technology. We need to know when it works, when it does not and why. Only then will this technology gain credibility and be used to its full potential. To help alleviate this situation, a geochemical study of the Bullwinkle Field was carried out to: identify the optimum analytical techniques and interpretation methods for application studies in the deepwater Gulf of Mexico; understand how to interpret fingerprinting data in the presence of compositional grading; increase Shell's confidence in reservoir geochemistry technology through a series of well-calibrated case studies; and participate in surveillance work aimed at planning subsequent field redevelopment work for the field. Study Area and Methods Bullwinkle field, in Green Canyon blocks 65 and 109 (Fig. 1), is considered a combination stratigraphic/salt-flank trap.9,10 The lower Pleistocene "J" sand package, which contains over 90% of the reserves at Bullwinkle, was mapped originally as four separate pay intervals (J1 through J4).9 Of these, the J2 sand (Fig. 2) contains the bulk of the recoverable reserves and has an estimated column height of about 1600 ft. Partly because of its size and importance, the J2 pay zone was the primary focus of this study. As discussed in this paper, the J2 pay interval is divided into the J2-RA and J2-RB reservoirs (Fig. 2). A comprehensive set of oils (n=18) and associated (dissolved) gas (n=16) samples, taken from nearly all of the initial oil producing wells, were analyzed in this study.

Fluid properties, phase and compartmentalization: Magnolia Field case study, Deepwater Gulf of Mexico, USA

“…Many hydrocarbon reservoirs in the deepwater Gulf of Mexico are known to contain a biogenic methane component (Rice & Claypool 1981;Bissada et al 1990Bissada et al , 2002Kornacki et aL 1996;Westrich et al 1999;Kornacki & Nelson 2000;Beeunas et al 2001;Edman et al 2001) Clayton (1991). whose presence is usually attributed to production in and migration from adjacent immature organically lean sediments.…”

“…In large part the aim of such studies is to aid in well planning (Hwang et al 1994;Smalley & Hale 1996;Beeunas et al 1999;Edman & Burk 1999;Levach6 et al 2000), reconstruction of filling, spilling and alteration histories (Hillebrand & Leythaeuser 1992;Horstad & Larter 1997), and fluid properties characterization and production monitoring and allocation (Kaufman et al 1990(Kaufman et al , 1997McCaffrey et al 1996;Nicolle et al 1997;Bazan, 1998). Geochemical data, irrespective of the application, find their greatest utility when interpreted in conjunction with other data types such as those provided by engineering, geological and geophysical studies (Smalley & England 1992;Larter & Aplin 1995;Westrich et al 1999). Enormous advances in computation speeds increasingly allow data-rich high resolution fluid descriptions to be used in reservoir simulation initializations.…”

New insights into reservoir filling and mixing processes

“…Montel & Gouel 1985;Jacqmin 1990;Westrich et al 1996Westrich et al , 1999Ratulowski et al 2000). Indeed, the inadequacy of diffusive mixing actually p r o m o t e s local convection because it preserves the buoyancy associated with an influx of lightening petroleum.…”

Maximizing volumetric sweep efficiency in waterfloods with hydrocarbon F–Φ curves

“…Interwell connectivity, in waterfloods, has been determined qualitatively using natural tracers and other geochemical fingerprinting. Some of the useful studies that appear in the literature in this context are presented by Smalley et al (1994), Westrich et al (1999), and Kaufman et al (2000Kaufman et al ( , 2002, among others. Recently capacitance-resistance models (CRMs) were introduced to map the interwell connectivity between producers and injectors (Izgec and Kabir, 2009;Sayarpour et al, 2009;Yousef et al, 2006).…”