Environmental Geology

, Volume 54, Issue 8, pp 1695–1706

Quick-look assessments to identify optimal CO2 EOR storage sites

Authors

    • Gulf Coast Carbon Center, Bureau of Economic Geology, John A. and Katherine G. Jackson School of GeosciencesThe University of Texas at Austin
    • Chevron Energy Technology Company, Process Technology Unit
  • Mark H. Holtz
    • Gulf Coast Carbon Center, Bureau of Economic Geology, John A. and Katherine G. Jackson School of GeosciencesThe University of Texas at Austin
  • Derek J. Wood
    • Gulf Coast Carbon Center, Bureau of Economic Geology, John A. and Katherine G. Jackson School of GeosciencesThe University of Texas at Austin
  • William A. Ambrose
    • Gulf Coast Carbon Center, Bureau of Economic Geology, John A. and Katherine G. Jackson School of GeosciencesThe University of Texas at Austin
  • Susan D. Hovorka
    • Gulf Coast Carbon Center, Bureau of Economic Geology, John A. and Katherine G. Jackson School of GeosciencesThe University of Texas at Austin
Original Article

DOI: 10.1007/s00254-007-0944-y

Cite this article as:
Núñez-López, V., Holtz, M.H., Wood, D.J. et al. Environ Geol (2008) 54: 1695. doi:10.1007/s00254-007-0944-y

Abstract

A newly developed, multistage quick-look methodology allows for the efficient screening of an unmanageably large number of reservoirs to generate a workable set of sites that closely match the requirements for optimal CO2 enhanced oil recovery (EOR) storage. The objective of the study is to quickly identify miscible CO2 EOR candidates in areas that contain thousands of reservoirs and to estimate additional oil recovery and sequestration capacities of selected top options through dimensionless modeling and reservoir characterization. Quick-look assessments indicate that the CO2 EOR resource potential along the US Gulf Coast is 4.7 billion barrels, and CO2 sequestration capacity is 2.6 billion metric tons. In the first stage, oil reservoirs are screened and ranked in terms of technical and practical feasibility for miscible CO2 EOR. The second stage provides quick estimates of CO2 EOR potential and sequestration capacities. In the third stage, a dimensionless group model is applied to a selected set of sites to improve the estimates of oil recovery and storage potential using appropriate inputs for rock and fluid properties, disregarding reservoir architecture and sweep design. The fourth stage validates and refines the results by simulating flow in a model that describes the internal architecture and fluid distribution in the reservoir. The stated approach both saves time and allows more resources to be applied to the best candidate sites.

Keywords

Gulf CoastCO2 EOR screeningCO2 sequestrationCO2 reserve growthUS oil reservoirs

Introduction

Carbon dioxide geologic storage, or geosequestration, is one possible way to reduce the increase of greenhouse gas concentration in the atmosphere. CO2 emissions from anthropogenic sources, like power plants and refineries, can be injected into the subsurface as a gas or a supercritical fluid under a sealing caprock. This new technology offers a promising solution. However, actions need to be taken promptly to reduce the environmental impact. In order to accelerate the implementation of CO2 geosequestration, it is necessary to identify the opportunities for developing a financial vehicle and infrastructure for large-scale application. The purpose of this study is to develop a methodology that quickly provides the best sites for both sequestering CO2 emissions and increasing the oil production of depleted oil fields.

Sequestration of CO2 in mature oil fields is an attractive alternative to sequestration in brine-bearing formations, not only because of the economic offset through enhanced oil recovery (EOR), but also because oil fields are known to have trapped hydrocarbons for periods of geological time and may therefore present the best environment for public acceptance and regulatory oversight of available storage options. Oil fields also have the advantage of having been well studied so that large volumes of data are available from publications, the industry and state agencies.

The study area for this work was the onshore US Gulf Coast. The data availability allowed the creation of an extensive database, which includes all major oil reservoirs in the area. This database was the starting point for this study. Screening, assessing, modeling and characterizing represent the key elements of the four-stage quick-look approach to identifying optimal CO2 EOR storage sites.

Stage 1: screening and ranking of candidate reservoirs

The focus of this stage is to reduce a very large number of oil reservoirs to a manageable set of candidates that are most suitable for EOR. It is important to note at this point that improved oil recovery translates into larger CO2 sequestration capacities. A number of screening criteria to identify candidate reservoirs for CO2 EOR can be found in literature. Projects on the subject conducted by the Regional Partnerships have been funded by the U.S. Department of Energy.

The Regional Partnerships is a government/industry effort to develop and implement carbon capture and sequestration technologies and regulations in seven different regions of the USA. An overview of the partnerships, achievements and future plans are summarized in a report titled “Carbon Sequestration Technology Roadmap and Program Plan” (2006). An oil field screening study for CO2 sequestration and enhanced oil recovery in the Illinois basin was completed by Korose et al. (1994) through the Carbon Sequestration Regional Partnership Program.

Research groups from universities and national organizations have also contributed with several publications. Kovscek et al. (2002) from Stanford University published a study titled “Screening Criteria for CO2 Storage in Oil Reservoirs”. Other criteria are reported by the Interstate Oil and Gas Compact Commission (1993) in an evaluation conducted in 1993of remaining oil resources in Texas. More screening and ranking methodologies for CO2 EOR/sequestration in oil fields can be found in publications such as Taber et al. (1997) and Ham (1996).

The screening methodology proposed in this paper includes the estimation of the miscibility conditions specific to each reservoir and relevant parameter cut offs that are most applicable to the Gulf Coast. Stage 1 is divided into two steps that provide a set of ranked candidate reservoirs for CO2 miscible EOR and subsequent sequestration. The first step focuses on the technical feasibility for CO2 EOR. Permanence of storage and seal integrity is assumed at this level of the multistage methodology because of existing hydrocarbon trapping. The second step is to rank the resulting set of feasible options in terms of various geologic and economic aspects.

Step 1: CO2 miscible displacement technical feasibility

As previously discussed, several screening criteria have been proposed. Broadly speaking, oil viscosity, oil API gravity (a function of oil specific gravity established by the American Petroleum Institute), reservoir depth, reservoir oil saturation and reservoir heterogeneity are among the most important parameters that define these criteria. Cracoana (1982) suggested oil viscosity values of 1 centipoise (cp) or less and API gravity values greater than 40°. Stalkup (1984) suggested oil API gravities greater than 27° and reservoirs deeper than 2,500 ft. Others have suggested ranges of oil API gravity, most frequently from 11° to 30°. Residual oil saturation has been primarily an economic screen, and values of 20–25% have been suggested by Stalkup (1984).

API gravity and temperature ranges suggested in the mentioned screening methods are constraints used to secure miscibility in the displacement. In this study, miscibility conditions are estimated for each individual reservoir. Thus, minimum miscibility pressure (MMP) becomes the most critical detailed constraint for the applicability of miscible CO2 EOR and is a function of oil properties, reservoir temperature, reservoir pressure and the purity of the injected CO2.

The approach of this study in determining the best possible CO2 EOR miscible flood candidates in the Gulf Coast region is based on a screening criteria published by Holtz et al. (2001). The screening proceeds according to a decision tree, which selects large reservoirs with miscible CO2 flood potential as candidates (Fig. 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs00254-007-0944-y/MediaObjects/254_2007_944_Fig1_HTML.gif
Fig. 1

Decision tree for identifying gas-displacement-recovery candidate reservoirs. (Holtz et al. 2001)

An extensive database that includes major oil reservoirs in Texas, Louisiana, Alabama and Mississippi was developed for screening. An unpublished BEG Texas oil reservoir database with geologic and engineering information was combined with Louisiana and Mississippi data from the Tertiary Oil Recovery Information System (TORIS) database, as well as Alabama reservoir data from the Alabama Geologic Survey. Data for Texas reservoirs were generated by gathering engineering information from numerous sources, including the Atlas of Major Texas Oil Reservoirs (Galloway et al. 1983), Atlas of Major Texas Gas Reservoirs (Kosters et al. 1991) and hearing reports from the Railroad Commission of Texas. The database includes petrophysical and fluid characteristics and geological information, along with production information and location data.

In the first stage of the methodology, MMP from reservoir temperature and C5+ oil molecular weight is determined. A relationship published by Holm and Josendahl (1982) and extended by Mungan (1981), which estimates MMP from molecular weight (MW) of the C5+ components of reservoir oil and reservoir temperature is applied. Equation 1 is a nonlinear multiple regression representative of this relationship that allows the estimation of MMP.
$$ {\text{MMP}} = - 329.558 + (7.727 \times {\text{MW}} \times 1.005^{{\text{T}}} ) - (4.377 \times {\text{MW}}). $$
(1)

The new miscibility information is added to the database at the reservoir level, and only reservoirs that are estimated to be miscible move on to the next step.

General reservoir screening constraints are applied to eliminate reservoirs that are not yet at the stage of their production life where CO2 EOR would be the proper option. Reservoirs that are candidates for CO2 EOR are those that are at an advanced stage of waterflooding or aquifer encroachment. At this production stage, most of the mobile oil has been produced, and the remaining significant volume of oil is residual oil that cannot be produced without EOR.

However, previous waterflooding is not applied as a requirement for large, deep reservoirs where vaporizing gas-drive miscibility can be achieved. The literature, SPE-EOR Field Reports (1982–1992) shows that these reservoirs have had gas displacement EOR applied directly after primary production.

Step 2: ranking methodology for CO2 sequestration candidate reservoirs

Geological character of the candidate reservoirs and economic aspects such as oil production volumes and distance to the stationary CO2 source are evaluated numerically. Each economic and geologic parameter is assigned a number from zero to one, one representing the best value, to quantitatively describe the quality of the reservoirs.

A ranking number generally corresponds to the ratio of the specific parameter value to the largest value in the set. For instance, if a group of reservoirs have cumulative productions ranging from 10 to 100 million barrels, the reservoirs with the smallest and largest production will be assigned ranking numbers of 0.1 and 1, respectively. In cases where parameters vary significantly from reservoir to reservoir, maximum or minimum ranking numbers are assigned to the reservoirs in the extreme ends of the range. These reservoirs, with their already assigned zeros or ones, are taken out of the ranking and are not used as a basis for comparison with the rest of the reservoirs. In this sense, the ranking numbers might not reflect proportionality with respect to their original values and are only used for hierarchical purposes.

The economic ranking is based on parameters that are reflective of the size and activity of the reservoirs, namely cumulative production and previous year annual production, and on the distance to the CO2 source.

The geological ranking is based on the structural regime, structural style, stratigraphic heterogeneity and depositional system. Unpublished work by William Ambrose categorizes complexity in three types (high, intermediate and low). The ranking numbers for these categories are 0.1 for high, 0.5 for intermediate and 0.9 for low. These levels of complexity respond to the relative intensity of folding and faulting and to the facies architecture and inferred reservoir sand body and carbonate lithofacies distribution. Structural regimes are divided into four principal categories, according to dominant stress relationships (collisional/compressional, transpressional/transtensional or strike-slip, and extensional), plus a structural regime characterized by salt tectonics. Structural styles are sub-categories defined within each of the principal four regimes. Table 1 shows structural complexities for different combinations of structural regimes and styles.
Table 1

Structural complexity

Structural regime

Structural style

Complexity

Collisional/compressional

Thrust duplex

High

Collisional/compressional

Fault-bonded anticline

Intermediate

Collisional/compressional

Anticline

Intermediate

Transpressional/transtensional

Shear zones and Wrench faults

High

Transpressional/transtensional

Horsts and Grabens

Intermediate

Transpressional/transtensional

Fault-bonded anticline

Intermediate

Transpressional/transtensional

Anticline

Intermediate

Extensional

Fault-bonded anticline

Intermediate

Extensional

Simple sealing fault

Low

Extensional

Anticline

Low

Extensional

Monocline/homocline

Low

Salt tectonics

Faulted salt dome

High

Salt tectonics

subsalt

Intermediate

Salt tectonics

Turtle structure

Intermediate

Salt tectonics

Onlap

Low

Stratigraphic heterogeneity was inferred from complexity of facies architecture and reservoir geometry within both clastic and carbonate depositional systems. Table 2 shows stratigraphic complexities for different combinations of depositional systems and settings.
Table 2

Stratigraphic heterogeneity

Depositional systems

Setting

Complexity

Slope channel and fan

Slope

High

fluvial (Fine-grained Meanderbelt)

Coastal plain

High

Fluvial (mixed-load)

Coastal plain

High

Tide-dominated delta

Marginal marine

High

Fluvial-dominated delta

Marginal marine

High

Lacustrine

Coastal plain

High

Alluvial fan

Coastal Plain

Intermediate

Fan delta

Marginal marine

Intermediate

Lobate delta (Wave/fluvial)

Coastal plain

Intermediate

Mesotidal shoreface

Marginal marine

Intermediate

Barrier/strandplain

Marginal marine

Intermediate

Fluvial (bedload)

Coastal plain

Intermediate

Basin floor fan: medial

Basin floor

Intermediate

Aeolian

Non- to marginal marine

Low

Wave-dominated delta

Marginal marine

Low

Carbonate shelf Grainstone shoals

Carbonate ramp and inner shelf

High

Carbonale open shelf

Carbonate shelf

High

Tidal channels and Oolite bars

Mixed sandstones and carbonates

High

Carbonate shelf Grainstpones

Carbonate shelf and mounds

High

Finally, the overall ranking is the normalized weighted average of the economic and the geological rankings at a ratio of 2:1 respectively. This methodology can be best understood in the Galveston area practical example given in the section “Results and discussion”.

Stage 2: quick-look storage capacity and EOR potential estimations

Quick-look total CO2 storage capacity

To accurately assess the capacity of a reservoir to store CO2, it is necessary to determine how much of the volume would be filled by CO2 during injection and how much would be bypassed. The quick-look approach is a simple arithmetic calculation implemented in a spreadsheet used when not enough data are available to calculate pore volume from reservoir parameters.

A quick-look CO2 storage potential (capacity) is obtained by analyzing cumulative production of an oil field. Here, it is assumed that the pore volume represented by oil production is available in the reservoir for CO2 storage. Stock-tank oil volumes are converted back to reservoir volumes, and resultant pore volumes are converted to the amount of CO2 that could be put into that volume at initial reservoir conditions (Eq. 2).
$$ {\text{CO}}_{{\text{2}}} {\text{QLSC }}({\text{metric tons}}) = 0.05259 \times N_{{\text{p}}} B_{{{\text{oi}}}} /B_{{{\text{CO}}_{{_{2} }} }} $$
(2)
where CO2QLSC = quick-look total CO2 storage capacity, Np = cumulative oil production (STB), Boi = oil formation volume factor (rbbl/STB), \( {\it B}_{{{\text{CO}}_2}}\) = CO2 formation volume factor (RCF/SCF), rbbl = barrels measured at reservoir pressure and temperature conditions, STB = stock tank barrels (at surface pressure and temperature conditions), RCF = reservoir cubic feet (at reservoir pressure and temperature conditions), SCF = standard cubic feet (at standard pressure and temperature conditions).

An empirical equation was derived to obtain \( {\it B}_{{{\text{CO}}_2}}\). Data for this equation were obtained from Jarrell et al. (2002). The equation is a set of statements and third and fourth order polynomials, and it is a function of reservoir pressure P and temperature T.

The oil formation volume factor Boi is a rarely populated data field in a data set. To overcome this problem, we make assumptions and apply empirical equations. The oil formation volume factor can be estimated from an equation by Standing (1947) (Eq. 3):
$$ B_{{{\text{oi}}}} = 0.972 + 0.000147F^{{1.175}}, $$
(3)
where
$$ F = R_{{{\text{so}}}} {\left( {\frac{{\gamma _{g} }} {{\gamma _{o} }}} \right)} + 1.25T, $$
γο = oil specific gravity, γg = gas specific gravity, Rso = solution gas–oil ratio, T = temperature (°F).
When applying this Standing correlation, values for gas gravity and solution gas–oil ratio are needed, and when these parameters are not known an estimate can be made. For this study, an average gas gravity value of 0.75 was applied and a second Standing correlation was used to estimate Rso (Eq. 4).
$$ R_{{{\text{so}}}} = \gamma _{g} {\left( {\frac{P} {{18(10)^{{Y_{g} }} }}} \right)}^{{1.204}} , $$
(4)
where Yg = 0.00091T–0.0125API

API = function of oil specific gravity used in the oil industry,

P = pressure (psi).

Quick-look total CO2 EOR potential

For this quick-look method, original oil in place (OOIP) is estimated from cumulative production (Np) and primary + secondary recovery (Eq. 5). Each reservoir is assumed to be close to its ultimate primary + secondary recovery. Furthermore, a basin-average primary + secondary recovery factor (Rp+s) is applied. For the US Gulf Coast, with its strong water-drive oil reservoirs, a 50% primary + secondary recovery factor is assumed. The choice of this value is explained in the next section “Waterflooded Reservoirs in Texas”.
$$ {\text{OOIP}} = N_{{_{{\text{p}}} }} /R_{{{\text{p}} + {\text{s}}}} . $$
(5)
Target CO2 EOR reserves \( \rm (N_{{{\text{CO}}_2}})\) are determined by applying a recovery factor, and the ultimate recovery factor from CO2 EOR is taken as a percent of OOIP (Eq. 6). Based on the EOR history in the Gulf Coast, reservoir recovery is assumed to be 15% of the OOIP. This recovery factor is adjusted for each individual reservoir in Stage 3.
$$ N_{{{\text{CO}}_2}} = {\text{OOIP}} \times R_{{{\text{CO}}_2}} .$$
(6)

Waterflooded reservoirs in Texas: recovery factors

To estimate the average waterflood efficiency of Gulf Coast Tertiary sandstone reservoirs, a survey of major Texas oil reservoirs that have undergone secondary recovery waterflood operations was conducted. Only those reservoirs that had undergone waterflood secondary recovery were included. Data were obtained from the Atlas of Major Texas Oil Reservoirs (Galloway et al. 1983). Non-Gulf-Coast reservoirs in Texas were also surveyed to establish a total range of possible values of waterflood recovery efficiency and to place Gulf Coast values of ultimate recovery in perspective. Recovery efficiency values were reported in percent of OOIP. Total range in recovery efficiency values was reported, as well as an average value, which was not weighted by OOIP of each reservoir, but which was considered equally on a reservoir-by-reservoir basis. However, the average value for the East Texas Woodbine play was weighted by OOIP value from the east Texas field because it dominates the play and accounts for the bulk of the play’s oil production.

Reservoirs were summarized primarily by depositional origin and secondarily by individual play. Principal producing Tertiary Gulf Coast plays in southeast Texas that have undergone waterflood secondary-recovery operations are from three plays: Yegua Deep-Seated Salt Domes, Frio Deep-Seated Salt Domes, and Frio Barrier/Strandplain Sandstone (Galloway et al., 1983). Play-average recovery efficiencies in these three plays range from 50.2 to 58.5%, with a total range for individual reservoirs from 28 to 61%. On the basis of these data, an average 50% recovery factor for waterflooded reservoirs in the Gulf Coast area is reasonable. However, one should be aware that recovery efficiency is highly variable, depending on reservoir properties and the optimization of the flood engineering.

Stage 3: quick-look dimensionless model to estimate recovery and storage potentials

Reservoir-specific parameters from the individual prospects generated in Stage 1 are introduced in a newly developed model that improves Stage 2 estimates of oil recovery and CO2 storage potential without the need for comprehensive simulations. This model is described in a recent publication by Wood et al. (2006). In this stage, reservoir geometry is not considered. However, fluid and rock properties are rigorously taken into account. The model is an approximation assuming a homogeneous, Cartesian rock volume with realistic properties where CO2 is injected updip (Fig. 2).
https://static-content.springer.com/image/art%3A10.1007%2Fs00254-007-0944-y/MediaObjects/254_2007_944_Fig2_HTML.gif
Fig. 2

Cartesian geometry of the dimensionless model

The model uses dimensionless groups necessary to describe CO2 flooding for a typical line-drive pattern, which is most applicable to Gulf Coast reservoirs. Past screening models from the literature, Rivas et al. (1992) and Diaz et al. (1996), focused only on oil recovery and simply assigned qualitative rankings to reservoirs, whereas this model focuses on both oil recovery and CO2 storage potential and produces quantitative results for each.

According to Wood et al. (2006), CO2 flooding can be fully described using 10 dimensionless groups:
\( R_{L} = \frac{L} {H}{\sqrt {\frac{{k_{z} }} {{k_{x} }}} } \)

effective aspect ratio

\( N_{\alpha } = \frac{L} {H}\tan \alpha \)

dip angle

\( M^{o}_{w} = \frac{{k^{o}_{{{\text{r}}w}} \times \mu _{o} }} {{k^{o}_{{{\text{ro}}}} \times \mu _{w} }} \)

mobility ratio (water)

\( M^{o}_{g} = \frac{{k^{o}_{{{\text{rw}}}} \times \mu _{o} }} {{k^{o}_{{{\text{rg}}}} \times \mu _{w} }} \)

mobility ratio (CO2)

\( N^{o}_{g} = \frac{{H\Updelta \rho g\cos \alpha }} {{\Updelta P}} \)

buoyancy number

\( P_{{{{inj}}D}} = P_{{{{{{inj}}}} {{}}}} /{\text{MMP}} \)

injection pressure

\( P_{{pD}} = P_{p} /{\text{MMP}} \)

producing pressure.

Soi

initial oil saturation

Sorw

residual oil saturation to water

Sorg

residual oil saturation to gas

where
krgo

 = end-point relative permeability of gas

kroo

 = end-point relative permeability of oil

krwo

 = end-point relative permeability of water

kx

 = horizontal permeability (md)

kz

 = vertical permeability (md).

The effects of capillary forces and dispersion are shown to be negligible in this approach and thus not included in the scaling. Reservoir heterogeneity is neglected in this third stage because of the difficulty inherent in properly scaling it and the relative paucity of heterogeneity data available for most reservoirs. Dimensionless oil recovery is effectively modeled with the dimensionless oil breakthrough time and the dimensionless recovery at three different dimensionless times, whereas CO2 storage potential is calculated only at the final dimensionless time.

The model works by inputting the dimensionless groups into a simulator to produce surface fits, which are generated by fitting an equation to the observed values of a dependent variable using the effects of multiple independent variables. The response surface fits estimate five parameters, four for oil recovery and one for CO2 storage. Each fit contains only six or seven terms, making the model simple to implement. This simplicity is intended to make a tool useful for smaller reservoir operators and improve the estimates in a large database.

The effectiveness of the model is demonstrated by consistent output from different reservoirs when applying the same dimensionless constants (Fig. 3). Each simulation is run until 1.2 pore volumes of CO2 are injected (tD = 1.2). The oil recovery is modeled with four data points. The first point is the dimensionless oil breakthrough time (tDo), at which significant amounts of oil are recovered. Recovery at all points before this time is assumed to be zero. The final three points are the dimensionless oil recovery at t= 0.8 (R1), t= 1.0 (R2), and t= 1.2 (R3). Dimensionless recovery is the fraction of the total pore volume recovered, and not the typical measure of the percentage of the original oil in place (%OOIP) recovered.
https://static-content.springer.com/image/art%3A10.1007%2Fs00254-007-0944-y/MediaObjects/254_2007_944_Fig3_HTML.gif
Fig. 3

Recovery curves from four different reservoirs with constant dimensionless group values

One of the curves in Fig. 3 represents a reservoir of the Gillock field in the Texas Gulf Coast. Results show that the pore volume recovered at a dimensionless time of 1 (or 1 pore volume) equals 0.33. A typical value for oil saturation in this field is 0.75. A new adjusted EOR potential as a percentage of the original oil in place is now easily estimated by simply calculating the ratio of 0.33 and 0.75. This new value of 0.44 (or 44%) is used to improve the storage capacity estimation.

Stage 4: reservoir characterization

Reservoir characterization is the process of building a geologic and engineering model that describes the reservoir’s internal architecture and the distribution of fluids in porous media, which provides needed input for a full reservoir simulation. The model should consider all important geologic scales of heterogeneity, from gigascopic basin-scale characteristics to microscopic pore-level characteristics. The internal architecture delineated is based on the integration of geologic character with measured engineering parameters. A reservoir model can integrate, for example, wire line logs, seismic surveys, basic or advanced core analysis, thin section analysis, production and pressure history, laboratory miscibility determination and fluid geochemistry. This integration results in a model that describes fluid-flow paths and barriers. Identifying the location within the reservoir of both the initial and remaining hydrocarbon resource allows the model to be applied as a tool for accurately assessing CO2 EOR potential and storage capacity.

The purpose of reservoir characterization in the context of the study is to develop a tool that honors factors such as reservoir geometry, heterogeneity, past development or anisotropy that can be applied in determining and optimizing hydrocarbon recovery for reserve growth and the calculation of sequestration volume

Results and discussion

The first two stages of the multistage quick-look methodology not only allow for the efficient screening of large number of reservoirs, but also for the quick-look estimate of sequestration capacities and EOR potential on a regional scale. The analysis shows that the miscible CO2 EOR resource potential along the U.S. Gulf Coast is 4.7 billion stock tank barrels (BSTB), and the CO2 sequestration capacity is as much as 2.6 billion metric tons (BMT).

Figure 4 shows a bar graph including the total number of reservoirs and reservoirs where CO2 miscible displacement is feasible for each of the Gulf Coast States.
https://static-content.springer.com/image/art%3A10.1007%2Fs00254-007-0944-y/MediaObjects/254_2007_944_Fig4_HTML.gif
Fig. 4

Number of total reservoirs and candidate reservoirs per state

Six major groups of oil plays have been identified that contain candidates for CO2 miscible displacement in the Gulf Coast (Fig. 5). Oligocene and Eocene plays extend from central Louisiana, southwestward and parallel to the present-day coastline, all the way to the Mexican border. The Miocene play completely encompasses southern Louisiana and the Mississippi delta in a west-east trend. The Travis Peak-Hosston and the Cotton Valley-Smackover major plays extend from the east side of the Gulf Coast region, in south Alabama and the west Florida Panhandle, to East Texas, ranging from southern Alabama, southern Mississippi and northern Louisiana to central east Texas. Finally, the Pennsylvanian play occurs in north-central Texas, east of the Texas Panhandle and northwest of Dallas-Fort Worth.
https://static-content.springer.com/image/art%3A10.1007%2Fs00254-007-0944-y/MediaObjects/254_2007_944_Fig5_HTML.gif
Fig. 5

Geographic distribution of CO2 EOR miscible potential in the Gulf Coast

A majority of the CO2 EOR candidate reservoirs in southeast Texas are located along the Oligocene play. The large cumulative oil production of the biggest fields in this region comes from Frio (Oligocene) reservoirs associated with deep-seated salt domes and Yegua (Eocene) reservoirs associated with salt-dome flanks. A major group of candidate reservoirs is located in northeast Texas, distributed along the west ends of the Travis Peak-Hosston and Cotton Valley Smackover plays. A third concentration of reservoirs is located in north Texas, bordering Oklahoma and following the Pennsylvanian oil reservoir play trends (Galloway et al. 1983). According to our analysis, Texas Gulf Coast CO2 EOR resources (excluding the Permian basin) add up to 3 BSTB.

In Louisiana, Miocene plays are located mainly in the Mississippi delta region and along the coastline. The rest of the reservoirs are scattered throughout the state and are dispersed in different plays. The Bay Marchand reservoirs have been responsible for the largest cumulative oil production in Louisiana. According to the assessment, the state has 1,500 million stock tank barrels (MMSTB) of CO2 EOR resources.

In Mississippi, candidate reservoirs are located mainly along the Cotton Valley–Smackover plays. Only ten other reservoirs occur south of the major group in the Travis Peak–Hosston play. The Smackover Formation and the Tuscaloosa Group have provided the state with most of the cumulative oil production. Brookhaven is the largest candidate field in Mississippi and produces from the Tuscaloosa Group. The analysis for the state indicates 89 MMSTB of CO2 EOR resource potential.

In Alabama, all the Gulf Coast candidate reservoirs occur in the Cotton Valley–Smackover play. As in Mississippi, gross cumulative volumes have been produced from the Smackover Formation. The largest candidate field in the state is Citronelle, and it produces from the Rodessa Formation. Analysis for Alabama indicates 98 MMSTB of CO2 EOR resources potential.

The largest potential and economic incentive for CO2 EOR occurs in the Texas Gulf Coast, followed by Louisiana (Fig. 6). The magnitude of the resources in the Texas Gulf Coast makes the Alabama and Mississippi results appear small; however, 187 MMSTB still represents a sizable resource to attract development of the use of CO2 for EOR.
https://static-content.springer.com/image/art%3A10.1007%2Fs00254-007-0944-y/MediaObjects/254_2007_944_Fig6_HTML.gif
Fig. 6

Bar graph of miscible CO2 EOR resource potential in the Gulf Coast

CO2 storage capacity associated with miscible CO2 EOR

The estimated volume of storage at abandonment in EOR candidates is over 2,600 million metric tons (MMT) of CO2 (Figure 7). The largest sequestration capacity in these economic EOR reservoirs occurs in Texas, with over 1,300 MMT of sequestration capacity. Louisiana also has a large capacity of over 1,100 MMT. Mississippi and Alabama account for smaller but significant volumes of sequestration capacity. These results indicate that Oligocene and Miocene oil reservoirs represent a large target for sequestering CO2 at the end of CO2 EOR.
https://static-content.springer.com/image/art%3A10.1007%2Fs00254-007-0944-y/MediaObjects/254_2007_944_Fig7_HTML.gif
Fig. 7

CO2 sequestration capacity in miscible oil reservoirs along the Gulf Coast

Example: Galveston Area

Oil reservoirs in the Galveston area were screened and ranked following the methodology described in Stage 1. A power plant in Texas City was chosen to be the anthropogenic source of CO2. In this example, 35 candidate reservoirs resulted to be feasible for CO2 EOR miscible displacement. The reservoirs were ranked and the top 15 options are included in the tables of this example to facilitate the reading. Table 3 lists production volumes and distances from each reservoir to the CO2 source. In this case, reservoirs with very large values of both cumulative production and annual production are given the maximum ranking and the rest of the reservoirs are compared with the largest value in the remaining group. Figure 8 shows the three fields that are out of range and, therefore, assigned a ranking value of 1. This means that the reservoirs within a comparative range are weighted against the fourth largest reservoir, Gillock South, in this example.
Table 3

Economic parameters for top 15 candidate reservoirs

Field name

Reservoir name

Cumulative production in 2003 (1,000 STB)

2003 Production (STB)

Distance from CO2 source (m)

Cedar Point

No name

18,849

59,847

32,942

Chocolate Bayou

Frio Upper

22,611

45,711

21,642

Chocolate Bayou

Alibel

10,183

7,054

21,642

Fig Ridge

Seabreeze

35,155

43,014

71,477

Franks

8900 Sand

8,400

21,059

8,926

Gillock

Big gas

21,416

32,409

12,561

Gillock

East Segment

22,212

5,900

12,561

South Gillock

No name

45,522

8,376

7,910

E. U. Frio Hastings

No name

34,399

85,118

26,111

W. Frio Hastings

No name

365,966

591,496

26,111

Moores Orchard

No name

19,245

3,577

91,000

Oyster Bayou

No name

142,679

53,443

60,373

Red Fish reef

Combined

32,671

21,728

25,228

North Thompson

No name

29,044

100,826

59,770

Webster

No name

595,586

612,760

31,366

https://static-content.springer.com/image/art%3A10.1007%2Fs00254-007-0944-y/MediaObjects/254_2007_944_Fig8_HTML.gif
Fig. 8

Cumulative oil production of top 15 EOR candidate reservoirs in Texas City

Results obtained in Stage 1 of the quick-look assessment are shown in Tables 4, 5, 6, 7. The economic ranking, including the ranking in terms of size and distance, is presented in Table 4. Table 5 shows the geological ranking in terms of the complexity of structural and stratigraphic heterogeneities. The general ranking is presented in Table 6. Figure 9 is a map showing the distribution of the 35 candidate reservoirs in the Galveston area and the top 15 reservoirs enhanced with their field outlines in the map. Stage 2 was also completed for this example. The quick-look CO2 sequestration capacities and CO2 EOR resource potential are presented in Tables 5 and 7
Table 4

Economic ranking of the top 15 candidate reservoirs

Field name

Reservoir name

Cumulative production 2003

2003 Production

Distance from CO2 source

Cedar Point

No name

0.41

0.70

0.80

Chocolate Bayou

Frio Upper

0.50

0.54

0.70

Chocolate Bayou

Alibel

0.22

0.08

0.70

Fig Ridge

Seabreeze

0.77

0.51

0.30

Franks

8900 Sand

0.18

0.25

0.90

Gillock

Big Gas

0.47

0.38

0.95

Gillock

East Segment

0.49

0.07

0.95

South Gillock

No name

1.00

0.10

1.00

E. U. Frio Hastings

No name

0.76

1.00

0.75

W. Frio Hastings

No name

1.00

1.00

0.70

Moores Orchard

No name

0.42

0.04

0.10

Oyster Bayou

No name

1.00

0.63

0.50

Red Fish reef

Combined

0.72

0.26

0.85

North Thompson

No name

0.64

1.00

0.30

Webster

No name

1.00

1.00

0.80

Table 5

Geologic ranking of the top 15 candidate reservoirs

Field name

Reservoir name

Structural heterogeneity

Stratigraphic heterogeneity

Overall verbal heterogeneity

Normalized overall numerical heterogeneity

E. U. Hastings

Frio

High

Intermediate

High–intermediate

0.60

W. Hastings

Frio

Intermediate

Intermediate

Intermediate–intermediate

1.00

Fig Ridge

Seabreeze

High

Intermediate

High–intermediate

0.60

North Thompson

No name

High

Intermediate

High–intermediate

0.60

Gillock

Big Gas

High

Intermediate

High–intermediate

0.60

Oyster Bayou

No name

High

Intermediate

High–intermediate

0.60

Franks

8900 Sand

High

Intermediate

High–intermediate

0.60

Webster

No name

High

Intermediate

High–intermediate

0.60

Red Fish reef

Combined

High

Intermediate

High–intermediate

0.60

South Gillock

No name

Intermediate

Intermediate

Intermediate–intermediate

1.00

Gillock

East Segment

High

Intermediate

High–intermediate

0.60

Moores Orchard

No name

High

High

High–high

0.00

Chocolate Bayou

Frio Upper

Intermediate

Intermediate

Intermediate–intermediate

1.00

Chocolate Bayou

Alibel

Intermediate

Intermediate

Intermediate–intermediate

1.00

Cedar Point

No name

High

Intermediate

High–intermediate

0.60

Table 6

General ranking of the top 15 candidate reservoirs

Field name

Reservoir name

Economic ranking

Geologic ranking

Overall ranking

W. Hastings

Frio

0.96

1.00

1.00

Webster

No name

1.00

0.60

0.89

South Gillock

No name

0.75

1.00

0.85

E. U. Hastings

Frio

0.89

0.60

0.82

Chocolate Bayou

Frio Upper

0.62

1.00

0.76

Oyster Bayou

No name

0.76

0.60

0.72

North Thompson

No name

0.69

0.60

0.68

Cedar Point

No name

0.68

0.60

0.67

Red Fish reef

Combined

0.65

0.60

0.65

Gillock

Big Gas

0.64

0.60

0.64

Fig Ridge

Seabreeze

0.56

0.60

0.59

Chocolate Bayou

Alibel

0.36

1.00

0.59

Gillock

East Segment

0.54

0.60

0.57

Franks

8900 Sand

0.48

0.60

0.53

Moores Orchard

No name

0.20

0.00

0.14

Table 7

CO2 sequestration capacity and EOR potential of top 15 candidate reservoirs

Field name

Reservoir name

Storage potential, CO2QLSP (metric tons)

EOR potential, (1,000 STB)

W. Hastings

Frio

45,813,828

109,790

Webster

No name

68,725,286

178,676

South Gillock

No name

6,062,301

13,657

E. U. Hastings

Frio

8,219,529

10,320

Chocolate Bayou

Frio Upper

3,642,462

6,783

Oyster Bayou

No name

24,943,374

42,804

North Thompson

No name

4,117,161

8,713

Cedar Point

No name

1,703,509

5,655

Red Fish reef

Combined

4,869,760

9,801

Gillock

Big Gas

3,443,931

6,425

Fig Ridge

Seabreeze

4,722,823

10,547

Chocolate Bayou

Alibel

1,352,111

3,055

Gillock

East segment

3,623,467

6,664

Franks

8900 Sand

1,201,794

2,520

Moores Orchard

No name

2,474,472

5,774

https://static-content.springer.com/image/art%3A10.1007%2Fs00254-007-0944-y/MediaObjects/254_2007_944_Fig9_HTML.gif
Fig. 9

Texas City area map with candidate reservoirs

The now reduced set of CO2 EOR opportunities selected using the quick-look methodology has a significant potential and provides attractive options for near term CO2 storage. The available information about these old and depleted reservoirs lowers the risk of the unknown, and existence of the accumulation indicates proven seals. However, the reservoirs need to be further characterized. Stages 3 and 4 were not undertaken for this practical example, but they should be fully completed if this area is selected for a sequestration project. Permanence risks should not be neglected and need to be taken into account. Intermediate risk of CO2 leakage from faulted traps associated with salt domes might be dominant in fields near Texas City. The structure is complex and the fields are compartmentalized. The historic oil and gas seeps near salt domes and stacked reservoirs, some with no gas accumulation, might suggest that leakage could be possible. Old reservoirs have old wells that might need an additional investment. There is also mixed rural and urban land use and it should also be taken into account.

Conclusion

The staged approach quickly reduces the number of possible sequestration sites, thus saving time and allowing more resources to be applied to optimize the candidates. The process sequentially considers (1) the feasibility for miscible CO2 displacement, (2) the location of the miscible candidate reservoirs with respect to CO2 anthropogenic sources, (3) reserve growth and storage capacities, and (4) the reservoir character and heterogeneities at all geologic scales.

A large potential for reserve growth lies along the Gulf Coast through the application of CO2 miscible enhanced oil recovery. Results indicate that there is the potential for approximately 4.7 BSTB of additional oil reserves. This resource lies between the Pennsylvanian and Miocene aged strata, with the main portion of the resources within the Oligocene and Miocene aged reservoirs. Among the Gulf Coast states, Texas contains the greatest oil CO2 EOR potential with a target of over 3 BSTB. This large resource will likely be exploited when a large volume of CO2 is finally captured and made available in this oil province. Results also indicate that the volume of CO2 emissions that could be stored as part of EOR economic activities measures 2,600 MMT.

Acknowledgments

The Gulf Coast Carbon Center, Bureau of Economic Geology acknowledges support of this research by BP America, Chevron, NRG, Entergy, KinderMorgan, Marathon, Schlumberger, Praxair and the John A. and Katherine G. Jackson School of Geosciences. Publication was authorized by the Director, Bureau of Economic Geology.

Copyright information

© Springer-Verlag 2007