Summary

Fluid composition is a valuable addition to the battery of "static" data available during reservoir appraisal that can be used to predict the dynamic behavior of the reservoir later in field life. This is because fluid data are not truly static; natural fluid mixing is a dynamic process that occurs over a long (geologic) time scale. Oil compositional differences, especially those that parallel changes in density, should be mixed rapidly by convection; their preservation indicates barriers to fluid flow. Water variations, now measurable on conventional core samples by use of residual salt analysis (RSA), help identify barriers to vertical fluid flow in oil and water legs.

Introduction

A compartmentalized reservoir is one that is subdivided into segments that behave as separate flow units during production. Reservoir compartmentalization is caused by barriers to fluid flow. A reservoir may be compartmentalized laterally by sealing faults or lateral variations in reservoir quality. Vertical compartmentalization occurs where reservoir zones are separated by laterally extensive zones of low-permeability rock, such as shales, carbonate cemented zones, or tar mats. Flow barriers can be of different strengths, ranging from relatively minor features that might inhibit fluid flow to major features that will not allow any fluid communication. Compartmentalization is part of the natural anatomy of a reservoir that controls the spatial distribution of reserves and the optimal way to produce them.

Compartmentalization frequently becomes evident during production as pressure data accumulate and production can be history-matched with the reservoir model. Many oil companies have their own horror stories of fields where unexpected compartmentalization was encountered during production, wrecking field economics because of the need for more wells and facilities that were inappropriate for the fluids produced. For example, the unexpected presence of a compartment with a high GOR might cause oil production rates to be constrained by the topside gas-handling capacity.

Ideally, therefore, reservoir compartmentalization should be mapped during reservoir appraisal so that this knowledge can be factored into field commerciality decisions, development planning, and facility design. The problem with this is that the dynamic data so useful for identifying compartmentalization during production are usually lacking at the appraisal stage. Therefore, making the best use of the data that are available during reservoir appraisal is important. Some of these data are truly static. For example, a 3D seismic survey may reveal the presence of faults but does not in itself provide information about whether the faults will behave as fluid-flow barriers. However, other types of data can be said to be "quasidynamic," and these include reservoir fluid compositions, the focus of this paper. Reservoir fluid compositional variations provide information about how fluids have been able to move through the reservoir on a geologic time scale. If this quasidynamic information can be extrapolated to a reservoir production time scale, it can be used to predict the dynamic behavior of the fluids during production. These types of data not only add confidence to the interpretation of more-conventional static information but also become available as soon as wells are drilled, providing information on compartmentalization early in field life.

This paper deals with how compositional heterogeneities in reservoir fluids (oil, gas, water) arise, how such heterogeneities mix, and how the mixing process can be modeled. Examples are given to show how such information can be used to assess reservoir compartmentalization.

How Fluid Variations Arise

Petroleum.

Perhaps the most common process that leads to field-scale variations in petroleum compositions is the effect of progressive source-rock maturation. During petroleum filling of a reservoir, a source rock that is progressively being heated will expel oil with a gradually changing composition. With time, the oil becomes gassier [and gas has a lower condensate/gas ratio (CGR)] and API gravity and geochemical-maturity parameters increase. The crestal part of the field often receives the earliest-generated, least-mature petroleum, with maturity increasing down-flank toward the part of the field closest to the source-rock kitchen, which will contain the most recently arrived petroleum. In some cases, more than one source-rock kitchen contributes oil to a field. These may have different compositions or temperatures, leading to differences in oil composition between the parts of the field filled from each kitchen.

Another process that can lead to petroleum compositional variations is biodegradation, which leads to preferential removal of the lighter petroleum fractions. If this occurs nonuniformly across a field, compositional differences will result.1 Leakage of gas from a field that carries dissolved light petroleum fractions with it may also be a cause of spatial compositional variations. As a result of these processes, many oil fields have inherited variations in oil composition laterally between different areas in a field1,2 and vertically between different reservoir layers.3–5

To use oil compositional variations to investigate reservoir compartmentalization, the ability to measure them is necessary. This requires samples and oil components that can be used to fingerprint oil composition. A variety of data types can be used to reveal the existence of reservoir fluid variations: PVT data, such as GOR, CGR, bubblepoint pressures, and density at reservoir conditions. Such data are frequently obtained from drillstem test (DST) samples, often from several wells in a field and sometimes from more than one depth in a well. Other geochemical information obtainable from gas chromatographic (GC) and GC-mass-spectrometric analysis of DST oils includes maturity parameters and statistical analyses of GC peaks, so-called GC fingerprinting. 1,3 These kinds of data can, in some cases, also be derived from core extracts, enabling investigation of finer-scale variations.

Formation Water.

The composition of formation water in sedimentary rocks is constantly evolving. This is evident from the study of mineral cements that have formed in the rock at different times in its history. These frequently display an evolution in isotopic composition (e.g., oxygen or strontium isotopes)6 or in the composition of fluid inclusions trapped in them.7 Such variations occur principally by water/rock interaction. Dissolution of minerals in the reservoir or enclosing mud rocks can modify the isotopic composition of the water, while dissolution of nearby salt will modify salinity. The Piper and Forties fields provide examples of such variations.8,9

p. 175–180

Petroleum.

Perhaps the most common process that leads to field-scale variations in petroleum compositions is the effect of progressive source-rock maturation. During petroleum filling of a reservoir, a source rock that is progressively being heated will expel oil with a gradually changing composition. With time, the oil becomes gassier [and gas has a lower condensate/gas ratio (CGR)] and API gravity and geochemical-maturity parameters increase. The crestal part of the field often receives the earliest-generated, least-mature petroleum, with maturity increasing down-flank toward the part of the field closest to the source-rock kitchen, which will contain the most recently arrived petroleum. In some cases, more than one source-rock kitchen contributes oil to a field. These may have different compositions or temperatures, leading to differences in oil composition between the parts of the field filled from each kitchen.

Another process that can lead to petroleum compositional variations is biodegradation, which leads to preferential removal of the lighter petroleum fractions. If this occurs nonuniformly across a field, compositional differences will result.1 Leakage of gas from a field that carries dissolved light petroleum fractions with it may also be a cause of spatial compositional variations. As a result of these processes, many oil fields have inherited variations in oil composition laterally between different areas in a field1,2 and vertically between different reservoir layers.3–5

To use oil compositional variations to investigate reservoir compartmentalization, the ability to measure them is necessary. This requires samples and oil components that can be used to fingerprint oil composition. A variety of data types can be used to reveal the existence of reservoir fluid variations: PVT data, such as GOR, CGR, bubblepoint pressures, and density at reservoir conditions. Such data are frequently obtained from drillstem test (DST) samples, often from several wells in a field and sometimes from more than one depth in a well. Other geochemical information obtainable from gas chromatographic (GC) and GC-mass-spectrometric analysis of DST oils includes maturity parameters and statistical analyses of GC peaks, so-called GC fingerprinting. 1,3 These kinds of data can, in some cases, also be derived from core extracts, enabling investigation of finer-scale variations.

Formation Water.

The composition of formation water in sedimentary rocks is constantly evolving. This is evident from the study of mineral cements that have formed in the rock at different times in its history. These frequently display an evolution in isotopic composition (e.g., oxygen or strontium isotopes)6 or in the composition of fluid inclusions trapped in them.7 Such variations occur principally by water/rock interaction. Dissolution of minerals in the reservoir or enclosing mud rocks can modify the isotopic composition of the water, while dissolution of nearby salt will modify salinity. The Piper and Forties fields provide examples of such variations.8,9

p. 175–180

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