reservoir characterization of a authors permian deep … · we thank mr. thomas b. cowden, of the...

AUTHORS

Shirley P. Dutton � Bureau of EconomicGeology, John A. and Katherine G. JacksonSchool of Geosciences, University ofTexas at Austin, Austin, Texas, 78713;[email protected]

Shirley P. Dutton is a senior research scientist atthe Bureau of Economic Geology with researchinterests in sedimentology, reservoir character-ization, sedimentary petrology, and clastic dia-genesis. She received a B.A. degree from theUniversity of Rochester and an M.A. degree andPh.D. from the University of Texas at Austin, allin geology.

William A. Flanders � Transpetco Engi-neering, 110 N. Marienfeld Place, Suite 525,Midland, Texas, 79701; [email protected]

William A. (Bill) Flanders is an engineer withspecial expertise in reservoir engineering,enhanced oil recovery, and reservoir modeling.Flanders is president of Transpetco Engineer-ing of the Southwest, and formerly worked atExxon and at Murphy Oil. He earned his B.S.and M.S. degrees in mechanical engineeringfrom Kansas State University.

Mark D. Barton � Shell Exploration andProduction Inc., 3737 Bellaire Blvd., P.O. Box481, Houston, Texas 77001-0481;[email protected]

Mark D. Barton is a reservoir geologist withShell Exploration and Production. He formerlyworked at the Bureau of Economic Geology,where he conducted the research summarizedin this paper. His expertise is in clastic sedi-mentology and stratigraphy. He holds a Ph.D.in geology from the University of Texas atAustin.

Reservoir characterization of aPermian deep-water sandstone,East Ford field, Delawarebasin, TexasShirley P. Dutton, William A. Flanders, andMark D. Barton

ABSTRACT

Deep-water sandstones of the Delaware Mountain Group in west

Texas and southeast New Mexico contained an estimated 1.8 billion

bbl of original oil in place, but primary recovery from these fields is

commonly less than 20%. East Ford field in Reeves County, Texas,

which produces from the Ramsey sandstone in the upper Bell Can-

yon Formation, went directly from primary production to tertiary

recovery by CO2 flooding. Field production has increased from 30

to more than 185 BOPD. Oil recovery has been improved by the

CO2 flood, but not as much as expected. Geologic heterogeneities

such as interbedded siltstones are apparently influencing reservoir

displacement operations in the East Ford unit.

A depositional model of the East Ford unit was developed using

data from Bell Canyon outcrops and subsurface data. The Ramsey

sandstones were deposited by turbidity currents in a basin-floor set-

ting. The sandstones are interpreted as having been deposited in a

channel-levee system that terminated in broad lobes; overbank splays

filled topographically low interchannel areas. Injection wells located

in splay sandstones apparently have poor communication with wells

in channel sandstones, perhaps because communication is restricted

through levee and channel-margin deposits. The south part of the unit

is responding well to the flood because the injection and production

wells are in the same interconnected lobe depositional environment.

INTRODUCTION

Reservoirs in deep-water sandstones of the Delaware Mountain Group

in the Delaware basin (Figures 1, 2) contained more than 1.8 billion

bbl of original oil in place (Holtz, 1995). Recovery efficiencies of

these reservoirs have averaged less than 20% since production began

Copyright #2003. The American Association of Petroleum Geologists. All rights reserved.

Manuscript received February 19, 2002; provisional acceptance July 8, 2002; revised manuscriptreceived September 23, 2002; final acceptance October 10, 2002.

AAPG Bulletin, v. 87, no. 4 (April 2003), pp. 609–627 609

610 Reservoir Characterization of a Permian Deep-Water Sandstone

mostly in the 1950s and 1960s, and only 340 million bbl of oil had

been produced through 1998. Many of these mature fields are nearing

the end of primary or secondary production and are in danger of

abandonment unless effective, economic methods of enhanced oil re-

covery (EOR) can be implemented.

Because of their historically low recovery, reservoirs in the Bell

Canyon Formation of the Delaware Mountain Group were the tar-

get of a reservoir-characterization and demonstration project fund-

ed by the U.S. Department of Energy as part of the Class III (Slope

and Basin Clastic Reservoirs) Field Demonstration Program (Dutton

and Flanders, 2001a). Our study was focused on East Ford field in

Reeves County, Texas (Figures 1, 3), which produces from the

Ramsey sandstone, the youngest sandstone in the Bell Canyon For-

mation. Primary recovery efficiency at East Ford field was only 16%,

and enhanced oil recovery by CO2 flooding is now being conducted

in the East Ford unit in an effort to increase recovery.

A depositional model of Bell Canyon sandstones was developed

from well-exposed outcrops (Barton and Dutton, 1999; Dutton

et al., 1999b). The outcrop model was used to guide interpreta-

tion of the subsurface data and develop a geologic model of East

Ford field. The geologic model explains much of the production

response to the fluid-displacement operation and provides a frame-

work for modifying the flood to increase recovery.

More than 375 reservoirs have been discovered in the Delaware

basin submarine-fan sandstone play in west Texas and New Mexico

(Galloway et al., 1983; Dutton et al., 2000). Because the reservoirs

in this play have similar depositional, diagenetic, and production

histories, knowledge gained from study of the East Ford field can be

extrapolated to other fields in the play. Lessons learned from the

production history of this mature reservoir in west Texas may also

have application to development of newly discovered turbidite res-

ervoirs throughout the world.

Geologic Setting

Upper Permian (Guadalupian) Delaware Mountain Group strata

compose a 4500-ft-thick (1400-m) succession of slope and basin

deposits in the Delaware basin (Gardner, 1997b). The Delaware

Mountain Group is divided, from oldest to youngest, into the Brushy

Canyon, Cherry Canyon, and Bell Canyon Formations (Figure 2).

Shelf-to-basin-floor correlations of time-equivalent strata indicate

water depths were between 1000 and 2000 ft (300 and 600 m)

during deposition of the Bell Canyon Formation (Kerans et al.,

1992). The cyclic interbedding of sandstones with organic-rich silt-

stones in the Delaware Mountain Group has been interpreted to

record frequent changes in relative sea level (Meissner, 1972; Fis-

cher and Sarnthein, 1988; Gardner, 1992, 1997a,b). During high-

stands in relative sea level, sands were trapped behind a broad,

flooded shelf and were prevented from entering the basin. Thin,

widespread, organic-rich siltstones accumulated on the basin floor

by the slow settling of marine algal material and airborne silt. During

ACKNOWLEDGEMENTS

This research was funded by the U.S. Depart-ment of Energy under contract no. DE-FC22-95BC14936 and by the state of Texas underState Match Pool Project 4201 and as part of theState of Texas Advanced Resource Recoveryproject. DOE project manager was Daniel J.Ferguson. The Bureau of Economic Geologyacknowledges support of this research byLandmark Graphics via the Landmark UniversityGrant Program. We thank Mr. Thomas B.Cowden, of the Cowden Ranch, for permissionto conduct fieldwork on the ranch. Researchwas assisted by Jose I. Guzman, Helena H.Zirczy, Luciano L. Correa, and Daniel L. Mendez.We gratefully acknowledge the efforts of DavidC. Jennette and Lesli J. Wood, who reviewed anearly version of this paper, and of peerreviewers Lee T. Billingsley and Fred F. Meissnerand Bulletin Editor John C. Lorenz. Theirconstructive comments improved this article.Illustrations were prepared by the Graphics staffof the Bureau of Economic Geology under thedirection of Joel L. Lardon, Graphics Manager.Susann Doenges edited the manuscript. Pub-lished by permission of the director, Bureau ofEconomic Geology, University of Texas atAustin.

subsequent lowstands in relative sea level, the carbon-

ate shelf was exposed and sands bypassed to the basin

floor. Many workers have interpreted the Delaware

Mountain Group sandstones as having been deposited

by turbidity currents (Newell et al., 1953; Payne, 1976;

Berg, 1979; Jacka, 1979; Gardner, 1992; Zelt and Rossen,

1995; Bouma, 1996; Barton and Dutton, 1999; Beau-

bouef et al., 1999; Gardner and Borer, 2000). Textural

characteristics of the sands, such as the absence of

detrital clay-sized material and the lack of channels on

the shelf, suggest that wind was an important agent in

transporting the sands to the shelf margin (Fischer and

Sarnthein, 1988). According to this model, dunes pro-

graded to the shelf break during sea level lowstands,

and eolian sands were then carried into the basin by

turbidity currents (Fisher and Sarnthein 1988; Gardner,

1992). Paleocurrent indicators show that the sands

entered the basin from the Northwest shelf and Central

Basin platform (Williamson, 1978) (Figure 1).

Delaware Sandstone Play

Fields in the Delaware sandstone play produce oil and

gas from slope and basin sandstone deposits that form

long linear trends (Figure 1). The reservoir sandstones

in much of the basin dip to the east and northeast, almost

directly opposite original depositional dip, because Late

Cretaceous movement associated with the Laramide

orogeny tilted the Delaware basin eastward (Hills, 1984).

Production from the East Ford unit and other upper Bell

Canyon fields in the central Delaware basin occurs from

the distal (southwest) ends of east-dipping, northeast-

oriented linear trends of thick Ramsey sandstone depos-

its. Most hydrocarbons in these fields are trapped by

up-structure facies changes from higher permeability

reservoir sandstones to low-permeability siltstones (Wil-

liamson, 1979; Ruggiero, 1985, 1993).

Approximately 379 fields produce from sand-

stones of the Delaware Mountain Group in west Texas

and southeast New Mexico (Dutton et al., 2000), in-

cluding 182 large fields that have produced more than

100,000 bbl of oil each. The Bell Canyon Formation has

produced more oil than the Cherry Canyon or Brushy

Canyon Formations. The 79 large fields in the Bell Canyon

Formation (Figure 1) had produced 209 million bbl of

Dutton et al. 611

Figure 1. Location of the largest fields (cumulative production>100,000 bbl) producing from the Bell Canyon Formation,Delaware Mountain Group (modified from Dutton et al., 2000).Field outlines and locations are based on information from Grantand Foster (1989), Kosters et al. (1989), Basham (1996), Lewiset al. (1996), and the Railroad Commission of Texas (un-published data). Field names and other information are given inDutton et al. (2000). Basin axis interpreted from Ewing (1990).

Figure 2. Stratigraphic nomenclature and relative hydrocar-bon production of the Delaware Mountain Group and time-equivalent formations on the surrounding shelves. Modifiedfrom Galloway et al. (1983), Ross and Ross (1987), Kerans andFitchen (1995), and S. C. Ruppel (personal communication, 2000).

oil through 1998 (Dutton et al., 2000). Through 1998,

62 large fields in the Cherry Canyon Formation had

produced 75 million bbl of oil, and 41 large Brushy

Canyon fields had produced 45 million bbl of oil (Dut-

ton et al., 2000). Previous papers have summarized

information about Brushy Canyon and Cherry Canyon

reservoirs in the Delaware sandstone play (Montgomery

et al., 1999, 2000). This paper focuses on the youngest

and most productive interval in the Delaware Moun-

tain Group, the Bell Canyon Formation.

Outcrop Characterization of Bell Canyon Sandstone

Interpretation of the Ramsey sandstone reservoir at the

East Ford unit was based largely on the depositional

model developed by investigation of Bell Canyon sand-

stones exposed in outcrop 25 mi (40 km) to the west

(Barton and Dutton, 1999; Dutton et al., 1999b) (Fig-

ure 1). The outcrop work focused on the first sandstone

below the McCombs limestone (Figure 2), a sandstone

that is environmentally analogous to, but older than,

the Ramsey sandstone. The older sandstone was inves-

tigated because it is interpreted to be a better analog to

the reservoirs at East Ford field, being from a similar

position in the facies tract. The Ramsey sandstone ex-

posed in outcrop represents a more distal depositional

environment than the Ramsey sandstone deposited at

East Ford field. The results of the outcrop study are

summarized below; more detailed descriptions of the

lithofacies, sandstone thickness and geometry, and in-

terpreted depositional processes are published in Bar-

ton and Dutton (1999) and Dutton et al. (1999b).

Six lithofacies were identified in outcrop: litho-

facies 1 is a massive, organic-rich siltstone; lithofacies

2 is an organic-rich, laminated siltstone; lithofacies 3 is a

laminated siltstone; lithofacies 4 is composed of thin-

bedded sandstones and siltstones that are graded or

display partial Bouma sequences (Bouma, 1962); litho-

facies 5 is a massive sandstone having water escape

and load structures; lithofacies 6 is a large-scale, cross-

laminated sandstone (Barton and Dutton, 1999). The

sandstones and massive siltstones are interpreted as

having been deposited by sandy high- and low-density

turbidity currents that carried a narrow range of sedi-

ment size, mostly very fine sand to coarse silt. Lami-

nated siltstones are interpreted to have been deposited

by fine material settling out of suspension.

Depositional elements recognized in the Bell Can-

yon deposits exposed in outcrop include (1) channels,

(2) levees, (3) lobes, (4) overbank splays, and (5) sheets

of laminated siltstones (Barton and Dutton, 1999; Dut-

ton et al., 1999b). Stratigraphic relationships of the dep-

ositional elements indicate that the sandstones were

deposited on the floor of the Delaware basin by a sys-

tem of leveed channels having attached lobes and over-

bank splays (Barton and Dutton, 1999; Dutton et al.,

1999b) (Figure 4). Individual channel-levee and lobe

complexes stack in a compensatory fashion and are

separated by laterally continuous, laminated siltstones.

Paleocurrent measurements indicate that the sandstone

bodies in the outcrop area trend north-south, which is

nearly perpendicular to the shelf margin on the north-

west side of the basin.

Lobe deposits are broadly lenticular sandstone

bodies that generally lack an erosive base (Figure 4c).

They are as much as 25 ft (8 m) thick and 2 mi (3.2 km)


Figure 3. Structure map on Lamar limestone showing loca-tion of the East Ford unit, Reeves County, Texas. Structure datafrom northwest edge of the map are from E. R. Kennedy, Jr.(unpublished data). Current oil-producing and CO2 injectionwells are shown. Other wells produced oil during primaryproduction but are shut in or used as observation wells now.Core was available from the Orla Petco 41R East Ford Unit well.Lamar limestone is about 46 ft above the Ramsey sandstone.Oil has not been produced below an elevation of 88 ft abovesea level in the Ramsey sandstone.

Dutton et al. 613

Figure 4. (a) Depositional model proposed for the Bell Canyon sandstone, showing deposition in submarine channels with levees,overbank splays, and attached lobes. Modified from Galloway and Hobday (1996) and Barton (1997). (b) Photograph of channel-leveedeposits; the two channels are vertically stacked in an offset fashion (from Dutton et al., 1999b). (c) Photograph of lobe sandstones andinterbedded laminated siltstones. (d) Strike-oriented cross section AA0 from Willow Mountain outcrop showing distribution of facies andtraces of key surfaces in a single high-order cycle, Bell Canyon Formation (modified from Dutton et al., 1999b). See Figure 1 for location.

wide (Barton and Dutton, 1999; Dutton et al., 1999b).

Lobes are composed mostly of medium- to thick-bedded

massive sandstones having dewatering features such as

dish and flame structures and are commonly interbedded

with sheets of laminated siltstone. Lobe sandstones are

interpreted as having been deposited at the mouths of

channels by unconfined, highly decelerating sediment-

gravity flows. In a prograding system, lobe facies would

have been deposited first and then overlain and partly

eroded by the channel-levee-overbank system (Figure 4d).

Channels are bound at the base by an erosion surface

and are largely filled with massive and cross-stratified

sandstones (BartonandDutton,1999;Duttonet al., 1999b).

Channels mapped in outcrop range from 10 to 60 ft

(3 to 18 m) in thickness; most are 20–40 ft (6–12 m)

thick. Channel widths are 300–3000 ft (90–900 m).

Aspect ratios (width/thickness) range from 15 to 40.

The channels bifurcate and widen downdip.

Flanking the channels on both sides are wedge-

shaped ‘‘wings’’ composed of thinly bedded sandstone

and siltstone (Figure 4b, d) and interpreted to be levee

deposits (Barton and Dutton, 1999; Dutton et al.,

1999b). In some cases, sandstone beds in the levees can

be traced into the adjacent channel (Figure 4b), and the

bedding relationships indicate that the margins of the

channels were maintained by elevated levee deposits.

In other cases, the levee and channel deposits display

a disconformable contact. Paleocurrents in the levee

deposits deviate by about 15–45j away from the axis

of the channel. Levees are interpreted as having been

deposited by unconfined turbidity currents that spilled

over the margin of the channel. The width of levee

deposits mapped in outcrop ranges from about 500 to

3000 ft (150 m to 1 km) wide. The levees thin abruptly

away from the channel, decreasing in thickness from 20

to 1 ft (6 to 0.3 m).

Overbank splays are composed of massive sand-

stones and display a broad, tabular to irregular geome-

try. They onlap the levee deposits (Barton and Dutton,

1999; Dutton et al., 1999b) (Figure 4d). Convoluted

bedding and dewatering structures such as dish and

pillar structures are common in these beds. The splay

sandstones are 3–25 ft (1–8 m) thick. Splays on the

flanks of the channel system were at least 3000 ft (900

m) wide and possibly much greater, as they extended

beyond the limits of the mappable outcrop. Overbank

splays increase in thickness away from the channel

margins in a compensatory fashion with the levees, in-

dicating that the splays filled topographically low in-

terchannel areas. The locally irregular geometry of the

splay deposits is related to the underlying topography.

Stratigraphic relationships suggest that the splays form-

ed during the final stages of channel filling (Barton and

Dutton, 1999). Volumetrically, they contain much of

the sandstone in the system (Figure 4d). It is not pos-

sible to distinguish overbank-splay deposits from lobe

deposits by lithofacies alone. The location, shape, and

stratigraphic position of the massive sandstones are

needed to distinguish these two depositional elements.

This depositional model is similar in many respects

to the ‘‘build, cut, fill, and spill model’’ proposed for

the older Brushy Canyon Formation by Gardner and

Borer (2000). The lobe deposits, which represent the

‘‘build’’ stage, are cut and overlain by the channel-levee

deposits that represent the ‘‘cut and fill’’ stage. Splay

sandstones represent the ‘‘spill’’ phase of overbank dep-

osition. A difference between our Bell Canyon model

and the Brushy Canyon models of Beaubouef et al.

(1999) and Gardner and Borer (2000) is our interpre-

tation that levees maintained the Bell Canyon channel

margins.

The depositional model developed from study of

well-exposed outcrops in the Bell Canyon Formation

guided our interpretation of the reservoirs at East Ford

field. This depositional model should be widely ap-

plicable to other Bell Canyon reservoirs. The results of

the outcrop study can thus be transferred by operators

to Bell Canyon fields throughout the play and used to

guide reservoir characterization.

RESERVOIR CHARACTERIZATION OF EASTFORD FIELD

The Ramsey sandstone, the main reservoir in East Ford

field, is the youngest sandstone in the Bell Canyon

Formation. It is divided into two sandstones (Ramsey

1 and 2) that are separated by a 1- to 3-ft-thick (0.3-

to 1-m) laminated siltstone (SH1) (Figure 5). The geo-

logic model of East Ford field was developed on the

basis of (1) subsurface data, including one core, from

the East Ford field, (2) an earlier study of the Geraldine

Ford field (Figure 1), for which abundant data, in-

cluding 70 cores, were available (Dutton et al., 1999b),

and (3) the Bell Canyon outcrop study. Eleven other

cores that were taken in East Ford field had been dis-

carded many years ago, and only core-analysis data

were available (Figure 3).

It is difficult to interpret detailed facies relation-

ships, such as those observed in the Bell Canyon out-

crops, from widely spaced well data alone. This dif-

ficulty is compounded in Bell Canyon reservoirs by the


absence of detrital clay in the sandstones; as a result, log

responses are muted, and log patterns are not always

reliable for facies identification. Interpretation of the

sandstone-thickness and log-facies data in the East Ford

field was strongly influenced by the depositional model

developed from the outcrop study.

Sandstone Distribution and Facies

The Ramsey 1 sandstone is thickest on the east side of

East Ford field (Figure 6a). It pinches out along the

west and south margins and reaches a maximum thick-

ness of more than 25 ft (7.6 m) along an elongate,

north-south trend that widens at the south end. The

east side of the Ramsey 1 sandstone is not defined because

few wells were drilled where the Ramsey sandstone

dips toward the water. Typical of Ramsey sandstone

reservoirs, significant production is from within the

oil-water transition zone. Productive oil has not been

Dutton et al. 615

Figure 5. Typical log from East Ford field; well location isshown in Figure 3. Lithology is interpreted on the basis of logresponse because core was not available from this well. Modi-fied from Dutton and Flanders (2001b).

Figure 6. (a) Isopach map of the Ramsey 1 sandstone. (b) Isopach map of percentage of net/gross sandstone in the Ramsey 1interval. Net sandstone was determined from sonic and gamma-ray logs as the number of feet of sandstone having volume of clayless than or equal to 15% and porosity greater than or equal to 17.5%. Sandstone thickness and ratio of net/gross sandstone wereused to interpret facies distribution.

defined below 88 ft (27 m) above sea level. Because log

control to the east is poor, the orientation of the Ramsey

1 sandstone is uncertain, and it may be oriented more

northeast-southwest than is shown in Figure 6a.

The younger sandstone in the Ramsey cycle, the

Ramsey 2 (Figure 7a), is thickest along a north-south

trend that is shifted to the west of the underlying Ram-

sey 1 sandstone. The offset of the Ramsey 2 sandstone

trend suggests that the younger sandstone was depos-

ited in the adjacent topographic depression created by

deposition of the preceding Ramsey 1 sandstone, indi-

cating compensational stacking. The Ramsey 2 sand-

stone is thinner than the Ramsey 1, having a maximum

thickness of 24 ft (7.3 m) at the north end and 10 ft

(3.0 m) at the south end.

Net/gross sandstone maps of the Ramsey 1 and 2

intervals provide an indication of heterogeneity in the

Ramsey sandstones. Because the goal of the maps was

to show the ratio of clean sandstone to total interval

thickness, net sandstone was mapped as sandstone having

porosity greater than or equal to 17.5% and volume of

clay (Vcl) less than or equal to 15%, no matter what the

water saturation (the choice of porosity and Vcl cutoffs

is discussed in Reservoir Properties). Clean sandstones

on the east side of the field have high water saturation

because they are in a structurally low position, not be-

cause of poorer reservoir quality. Well control for the

maps is limited to wells having both sonic and gamma-

ray logs through the entire interval.

The map of net/gross sandstone for the Ramsey 1

interval (Figure 6b) shows that the highest values

(>90%) are along the east side of the field. Net/gross

sandstone values decline toward the west and at the

south end of the field. The Ramsey 2 sandstone has

net/gross sandstone values (Figure 7b) that are some-

what lower, although few wells have sonic logs where

the Ramsey 2 sandstone is thickest and might be ex-

pected to have the highest net/gross values. Net/gross

sandstone decreases both to the east and west, as well

as at the south end of the field (Figure 7b).

A 54-ft (16.5-m) core was taken in the Orla Petco

41R East Ford Unit well; the cored interval extends


Figure 7. (a) Isopach map of the Ramsey 2 sandstone. (b) Isopach map of percentage of net/gross sandstone in the Ramsey 2interval. Net sandstone was determined from sonic and gamma-ray logs as the number of feet of sandstone having volume of clayless than or equal to 15% and porosity greater than or equal to 17.5%. Sandstone thickness and ratio of net/gross sandstone wereused to interpret facies distribution.

from the bottom few feet of the Trap siltstone, through

the Ramsey sandstone, and into the upper few feet of

the Ford siltstone (Figure 8). Several lithofacies iden-

tified in outcrop were observed in the core: (1) organic-

rich siltstone; (2) laminated siltstone; and (3) struc-

tureless or massive sandstones having few laminations

but containing floating siltstone clasts, dewatering fea-

tures, and load structures. Two lithofacies that were

observed in outcrop and in cores from Geraldine Ford

field (Dutton and Barton, 1999), but not in the Orla

Petco 41R East Ford Unit core, are cross-stratified

sandstone and rippled sandstone. The Ramsey 1 and 2

intervals in the core are mostly massive, very fine grained

sandstones. The most common sedimentary structures

are features related to dewatering: dish structures, flame

structures, and convolute bedding.

Depositional Model

Ramsey sandstones in East Ford field are interpreted

as having been deposited in a channel-levee system

with attached lobes and overbank splays (Figures 9,

10). The Ramsey deposits at East Ford are about 2500–

4000 ft (760–1220 m) wide, dimensions similar to those

of the system studied in outcrop. Channel deposits

in East Ford field are interpreted to occur along the

trend of thickest Ramsey 1 and 2 sandstones (Figure

10), where net/gross sandstone is greater than 90%.

Channels in the Ramsey 1 sandstone are about 25 ft

(7.6 m) thick and 950 ft (290 m) to perhaps as much as

2000 ft (610 m) wide (Figure 10a). Ramsey 2 channels

are interpreted to be about 15 ft (4.5 m) thick and

about 1300 ft (400 m) wide (Figure 10b). In outcrop,

many channels were seen to be nested and laterally

offset from each other (Figure 4b). Similar nesting of

multiple channels may occur in East Ford field, but the

well control is not sufficiently close to distinguish

separate channels. Assuming the channels at East Ford

field are single channels, the aspect ratio (width/thick-

ness) of Ramsey 1 channel deposits is 40:1 to as much

as 80:1. Ramsey 2 channel deposits have aspect ratios

of about 85:1.

The sandstones that flank each side of the channel

deposits are interpreted to be levee and overbank-

splay deposits (Figures 9, 10). On the basis of the out-

crop study, we infer that levee deposits are continuous

with the Ramsey 1 and 2 channel sandstones and sep-

arate them from splay deposits. The sandstones that

border each side of the levee deposits are interpreted to

be overbank splays. Volumetrically, the splays contain

much of the sandstone outside of the channels (Figure

10). Good-quality reservoir sandstone occurs in the le-

vee and overbank deposits, but these facies also contain

interbedded siltstones and silty sandstones and thus

have lower values of net/gross sandstone (30 to 90%)

(Figures 6b, 7b).

Dutton et al. 617

Figure 8. Description of core from Orla Petco 41R East FordUnit well (modified from Dutton and Flanders, 2001b). Welllocation is shown in Figure 3.


Figure 9. (a) Strike cross section BB0 of the central part of the East Ford unit. (b) Strike cross section CC0 of the south part of theunit. Location of cross sections is shown in Figure 3.

Deposits that widen out at the south end of the

field are interpreted to be lobe sandstones in both Ram-

sey 1 and 2 intervals (Figure 10). Lobe facies, deposited

by unconfined, high-density turbidity currents at high

suspended-load fallout rates, occur in broad sheets at

the mouths of channels. The presence of massive sand-

stones having abundant fluid-escape structures in the

Orla Petco 41R East Ford Unit core is consistent with

the interpretation that they are lobe deposits.

Because deposition of lobe sandstones was episod-

ic, laminated siltstones are interbedded with the lobe

sandstone sheets. Net/gross sandstone values are high-

est in the center of the lobe deposits (70 to >90%) and

decrease to 30% toward the margins of the lobes (Fig-

ures 6b, 7b).

Channel, levee, splay, and lobe facies in the East

Ford unit could not be distinguished by log patterns in

many cases. The interpretation of depositional facies

shown in Figures 9 and 10 was made on the basis of (1)

sandstone thickness and net/gross sandstone trends in

the unit, (2) the depositional model developed from

outcrop, and (3) production response to the CO2 flood,

which is discussed in CO2 Flood of East Ford Unit.

Reservoir Properties

Ramsey sandstones in the Orla Petco 41R East Ford

Unit well are well-sorted, very fine grained arkoses

having an average composition of Q67F26R7. Primary

porosity estimated from thin-section point counts of

25 sandstone samples averages 19%, and secondary po-

rosity averages 2%. Cements constitute between 1 and

31% of the sandstone volume, with calcite and chlorite

being the most abundant. Calcite cement has an aver-

age volume of 7% and ranges from 0 to 30%. Chlorite

(average = 1%) forms rims around detrital grains, ex-

tending into pores and pore throats.

Porosity and permeability of Ramsey sandstones

and siltstones were determined by core analyses of 368

samples from 12 cored wells (Figure 11). Porosity in

Ramsey sandstones averages 22% and ranges from 4.5

to 30.6%. Sandstone permeability ranges from 0.01 to

Dutton et al. 619

Figure 10. (a) Interpreted facies distribution of the Ramsey 1 sandstone in the East Ford unit, superimposed on Ramsey 1 isopachmap. (b) Interpreted facies distribution of the Ramsey 2 sandstone, superimposed on Ramsey 2 isopach map. See text for discussionof facies interpretation. Heterogeneities at facies boundaries apparently disrupt fluid-displacement operations in the East Ford unit.

249 md; the average permeability is 40 md, and geo-

metric mean permeability is 22 md. Ramsey 1 sand-

stones have higher average permeability than do Ramsey

2 sandstones, 44 versus 34 md, respectively. Volume of

calcite cement is the dominant control on porosity and

permeability (Dutton and Flanders, 2001b). In samples

having less than 10% calcite cement, average porosity is

22.5%, and average permeability is 46 md. Sandstones

having more than 10% calcite cement have an average

porosity of 11.5% and average permeability of 3 md.

On the basis of the Dykstra-Parsons heterogeneity co-

efficient (V), a measure of permeability heterogeneity

(Dykstra and Parsons, 1950), the Ramsey sandstone in

the East Ford unit was found to be moderately homo-

geneous (V = 0.52).

The Ramsey sandstone at the East Ford unit had

high initial water saturation (Sw), and many wells pro-

duced some water at discovery. Log-calculated values

of Sw ranged from 44 to 55% across most of the field

and averaged 48% (Dutton et al., 1999a). The cemen-

tation exponent (m) of 1.83 and saturation exponent

(n) of 1.90 used in these calculations were determined

in the Ford Geraldine unit (Asquith et al., 1997).

Average Sw measured in core analyses of Ramsey sand-

stones was 47%. Water saturation increases to the east

and northeast in the field, which is to be expected be-

cause that direction is down structural dip. The Ford

Geraldine unit averaged 48% Sw at discovery, well above

the irreducible water saturation of 35% (Pittaway and

Rosato, 1991).

Cutoffs to define net pay for the Ramsey sandstone

in the East Ford unit were established for volume of

clay (Vcl), porosity, and Sw (Asquith et al., 1997). The

selection of a Vcl cutoff was based on the work of

Dewan (1984), which suggests a Vcl cutoff of 15% for

reservoirs containing dispersed authigenic clay. Core

porosity and permeability data (Figure 11) were used

to select the porosity cutoff of 17.5%, corresponding to

a permeability of 5 md. Sandstones having permeabil-

ity of greater than or equal to 5 md probably represent

the floodable Ramsey sandstones. Five K ro � K rw

relative-permeability curves from a well in Geraldine

Ford field were used to determine the Sw cutoff of 60%,

because no relative-permeability data were available

from East Ford. At 60% Sw, the relative permeability to

oil (K ro) should be approximately eight times the

relative permeability to water (K rw) (Asquith et al.,

1997).

Average net-pay thickness in East Ford field is 20

ft (6 m). The highest values (>30 ft [9 m]) follow a

north-south trend down the center of the field. Net

pay decreases to the west, where the Ramsey sand-

stone pinches out into siltstone, and to the east, where

the sandstone dips toward the oil-water contact.

EAST FORD PRODUCTION HISTORY

Primary Development

Primary recovery in East Ford field began in October

1960 and continued until June 1995. Several wells

were initially completed in the Olds sandstone (Figure

5), and production from the Olds and Ramsey sand-

stones was commingled. Production peaked at 965

BOPD in May 1966, and cumulative production during

primary recovery was 3,209,655 bbl. The estimated

2.9 million bbl produced from the Ramsey sandstone

represents 16% of the 18.4 million bbl of original oil in

place (OOIP) (Dutton et al., 1999c) (Table 1). East

Ford field, like other Delaware sandstone reservoirs,

was characterized by relatively high amounts of mobile

water at the time of discovery. Among the reasons for

the low primary production are (1) the solution gas/oil

ratio of only 400–600 scf/bbl, which resulted in limit-

ed natural drive energy; (2) the expenditure of con-

siderable solution-gas drive energy in the recovery of

water from the reservoir; (3) the lack of pressure support

from the aquifer because of the limited water influx.

Production during 1994, the last full year of pri-

mary production, was 9734 bbl, and daily production


Figure 11. Cross plot of core porosity versus core perme-ability with porosity-permeability transform for the Ramseysandstone and siltstone in the East Ford unit, Reeves County,Texas. Cored wells are identified in Figure 3.

had declined to 30 bbl. The decline rate, calculated using

an exponential least-squares fit of the production data

from April 1991 through September 1994, was 10.1%.

At that rate of decline, the economic limit of the field

would have been reached soon if the CO2 flood had

not started.

The East Ford unit did not undergo secondary

recovery by water flooding, because water flooding has

not been very successful in other Ramsey sandstone

reservoirs. Water flooding added about 4% of the OOIP

to the total recovery from the Ford Geraldine unit

(Pittaway and Rosato, 1991) and from Twofreds (West

Side) field in Loving County, Texas (Kirkpatrick et al.,

1985; Flanders and DePauw, 1993). Waterflood re-

coveries in Ramsey sandstones have been low because

of poor sweep efficiency caused by (1) abundant mobile

water present when the waterflood was started, (2)

water injection above the formation parting pressure,

(3) lack of proper water filtration, and (4) patterns not

arranged to exploit depositional characteristics.

CO2 Flood of East Ford Unit

Two Bell Canyon reservoirs have already undergone

CO2 floods, Twofreds Delaware unit and Ford Geral-

dine unit (Figure 1). Tertiary recovery began in 1974 in

the Twofreds unit (Kirkpatrick et al., 1985; Flanders

and DePauw, 1993) and in 1981 in the Ford Geraldine

unit (Pittaway and Rosato, 1991). In the Twofreds

unit, 12% of OOIP was recovered by the CO2 flood

(Table 1) (Dutton and Flanders, 2001a). Approxi-

mately 7% of OOIP in the flooded area of the Ford

Geraldine unit was recovered by the CO2 flood (Table

1). Tertiary recovery from the Ford Geraldine unit may

be lower because a higher percentage of OOIP (23%)

was recovered during primary and secondary production

than in the Twofreds unit (16%) (Table 1) because of a

stronger solution-gas drive at Ford Geraldine.

East Ford field was unitized, and CO2 injection in

the East Ford unit began in July 1995 with 8 injectors

and 11 producers; the unit currently has 7 injectors and

15 producers (Figure 3). The number of active wells in

the unit was kept low to reduce costs. In the north part

of the unit, the injectors are positioned on the west side,

but to the south, the injectors are located centrally

(Figure 3).

The first response to the flood was observed in

April 1996, and major production response in the unit

began in December 1997 (Figure 12). Average bottom-

hole pressure in the unit at the start of the project was

723 psi (4.98�106 Pa), whereas minimum miscibility

pressure is 900 psi (6.21�106 Pa). Somewhat higher

pressure occurred around wells 7, 12, 36, and 37 (Fig-

ure 13a), where produced water was reinjected into

the formation. The low pressure in the East Ford unit

at the start of the CO2 flood, combined with the low

reservoir temperature of 83jF (23jC), meant that CO2

would exist as both vapor and liquid phases under these

Dutton et al. 621

Table 1. Summary Production History of Three Fields Producing from Bell Canyon Sandstones*

OOIP

Primary + Secondary

Production to Start

of CO2 Injection

Production After Start

of CO2 Injection to

January 1, 2002

Cumulative Production

to January 1, 2002

(MMbbl) (MMbbl) (% OOIP) (MMbbl) (% OOIP) (MMbbl) (% OOIP)

Ford Geraldine 99 22.3 23 6.6 7 28.9 29

Twofreds 51 8.4 16 6.1 12 14.5 28

East Ford 18.4 2.9** 16 0.22 1 3.1 17

*MMbbl = Million stock-tank barrels; OOIP = Original oil in place.**Estimated primary production from Ramsey sandstone only.

Figure 12. Plot of monthly oil production from the East Fordunit since the field was discovered in 1960. The field was onprimary production until a CO2 flood was begun in July 1995.

conditions. Response to CO2 injection in the field may

have been delayed as a result. At these low tempera-

tures and pressures, liquid CO2 can occur on both the

injection side and the production side.

By May 2001, the production rate had increased to

185 BOPD, along with 345 BWPD and 1.7 mmcf/day

(hydrocarbon gases and CO2) (Figure 14a). Most of the

produced gas and water are reinjected, and to repres-

sure the reservoir, additional water for injection is being

taken from a nearby operator. Cumulative production

through May 2001 was 180,097 stock-tank bbl of oil,

518,000 bbl of water, and 1344 mmcf of gas. Essentially

all the oil production since the start of the CO2 flood

can be attributed to the EOR project. Injection rates in

May 2001 were 3100 mcf/day of purchased CO2, 1425

mcf of recycled CO2, and 375 BWPD (Figure 14b).

Cumulative injection through May 2001 was 9057

mmcf of purchased CO2, 1075 mmcf of recycled CO2,

and 670,000 bbl of water.

The CO2 flood has increased production from the

East Ford unit substantially (Figure 14), but several

production abnormalities have been observed: (1) low

pressure in the center of the field, (2) low production

rates, (3) severe reduction in transmissibility indicated

by a bottom-hole pressure-buildup test, and (4) low

gas production rates in key wells. Some of these abnor-

malities may be caused by mechanical problems, but

others may result from the effect of geologic hetero-

geneity in the field.

Although pressure at the north and south ends of

the field has increased during CO2 injection, low pres-

sure has persisted in the center of the field (Figure

13b). The pressure data were collected in wells used

as observation wells; injection wells were shut in for 48

hr, and the decline in pressure was observed. The pres-

sure distribution suggests that communication is poor

between wells in the center of the field and the nearest

injectors (Figure 13b).

The production rate in some wells, including Orla

Petco 3 East Ford Unit and Orla Petco 4 East Ford Unit

(Figure 13), is lower than would be expected from their

initial potential. For example, well 4 made 106 bbl of

liquids (oil + water) per day (BLPD) during initial-

potential tests, so the current production of about 20

BLPD is surprisingly low. In addition, gas production

from Orla Petco 4 East Ford Unit has leveled off. Well

Orla Petco 3 East Ford Unit has a similarly low pro-

duction rate of 10 BLPD (whereas it flowed 110 BLPD

during the initial-potential test), and pressure in the

well now is greater than 1200 psi (8.27�106 Pa). Gas


Figure 13. Bottom-hole pressure inthe East Ford unit in (a) July 1995 and(b) January 1999. Miscibility pressure is900 psi (6.21�106 Pa). These maps showareas that are not responding to the CO2

flood, where the pressure remains low.

production from Orla Petco 3 East Ford Unit is also

low, although the well is close to the injector Orla Petco

2 East Ford Unit.

Influence of Geologic Heterogeneity on East FordProduction

Oil recovery has been improved by the CO2 flood, but

not as much as had been expected. Production abnor-

malities may indicate that geologic heterogeneities are

affecting reservoir displacement operations. In many

cases, these heterogeneities do not have a major in-

fluence on primary recovery, but they can have a sig-

nificant impact on EOR processes, including water-

floods and CO2 floods.

Interbedded siltstones are probably the most im-

portant cause of heterogeneity in the Ramsey reser-

voirs. Because of the low permeability of siltstones

(average 4 md), limited cross-flow of fluids occurs be-

tween sandstones separated by siltstones. Siltstones

occur as (1) widespread sheets that bound high-order

depositional cycles, (2) a concentration of rounded silt-

stone clasts and, rarely, a drape of massive, organic-rich

siltstone along the base of channels, and (3) beds

interbedded with thin sandstones in the levee depos-

its (Dutton et al., 2000). All of these siltstone beds

have the potential to disrupt displacement operations in

Delaware sandstone reservoirs. For example, cross-flow

of fluids may be limited between a well in an overbank-

splay deposit and a well in a channel deposit, not only

Dutton et al. 623

Figure 14. Plot of oil production, water/oil ratio, and gas/oil ratio in the East Ford unit since 1990. (b) Plot of gas and waterinjection since 1995. STB is stock-tank barrels.

because of interbedded siltstones in the levee, but al-

so because of a siltstone-pebble lag or thin siltstone

drape along the base of the channel (Dutton et al.,

2000). Wells in the same interconnected depositional

facies should have good displacement communication.

In the East Ford unit, a major geologic heterogeneity

is caused by the 1- to 3-ft-thick (0.3- to 1-m) laminated

siltstone (Figure 5) that divides the Ramsey reservoir

into the Ramsey 1 and Ramsey 2 sandstones through-

out the field. The siltstone represents a pause in sand

deposition in the Ramsey interval, when laminated silt

and organic matter were deposited over a widespread

area. Cross-flow of fluids between the Ramsey 1 and 2

sandstones is limited because of the siltstone. Because

CO2 that is injected only into the Ramsey 2 sandstone

interval probably will not penetrate the Ramsey 1 sand-

stone, both injector and producer wells should be per-

forated above and below the siltstone.

Another source of heterogeneity in the East Ford

unit is variation in reservoir quality between wells.

Net/gross maps indicate that channel- and central-lobe

deposits are the most homogeneous, consisting mostly of

clean, porous sandstone (net/gross sandstone >90%),

whereas the levee, overbank, and lobe-margin deposits

are more heterogeneous (net/gross sandstone 30–

90%). In general, better communication would be ex-

pected between wells in the areas of high net/gross

sandstone and poorer communication in areas of low

net/gross sandstone or between areas of high and low

net/gross sandstone.

Finally, 2- to 16-in.-thick (5- to 40-cm) layers of

tightly calcite-cemented sandstone also increase res-

ervoir heterogeneity (Figure 8). Well response and

geophysical log correlations suggest that some calcite-

cemented layers are laterally continuous over a distance

of 1000 ft (300 m) and cause vertical compartmentali-

zation in the reservoir.

The East Ford unit appears to be divided into three

areas of better interwell communication that corre-

spond to the areas of higher pressure at the north and

south ends of the field and the low-pressure area in the

middle (Figure 13b). Communication between wells in

different areas is restricted. The areas may result from

facies changes, subtle structural or bathymetric con-

trols on deposition, or variations in sediment transport

direction.

North Part of East Ford Unit

The area of higher pressure at the north end of the

unit (Figure 13b) contains three injector wells located

along the west side of the area (wells 2, 7, and 14) and

seven producers (1, 3, 4, 9, 10, 13, and 17) (Figure

10). In this part of the field, the Ramsey 2 sandstone is

the main target (Figure 10b). Orla Petco 1 East Ford

Unit has responded well to the flood and is one of the

better wells in the field, producing about 26 BOPD in

March 2000. Production from Orla Petco 4 East Ford

Unit is lower, about 15 BOPD. One possible explana-

tion for the lower production is that a barrier may

restrict communication between this producing well

and the injector, Orla Petco 2 East Ford Unit. Geologic

interpretations suggest the presence of a channel-

levee boundary between wells 2 and 4 in the Ramsey 2

sandstone (Figure 10b).

Orla Petco 10 East Ford Unit, which is interpreted

to be in the same overbank-splay sandstone as injector

Orla Petco 7 East Ford Unit (Figure 10b), is a moder-

ately good well, producing about 19 BOPD in March

2000. Orla Petco 9 East Ford Unit is a poor well,

although it is located in the thickest part of the Ram-

sey 2 sandstone. The presence of a levee between in-

jector well 7 and producer well 9 may explain the poor

response of well 9 (Figure 10b).

Locating a new injector in a north-south orienta-

tion with the existing producers, following the chan-

nel trend, might improve recovery from the thick

Ramsey 2 channel sandstones in the north area. Orla

Petco 6 East Ford Unit could be converted into an

injector to increase the response of Orla Petco 9 East

Ford Unit. Both of these wells are in the thickest part of

the Ramsey 2 sandstone, in an area of high net/gross

sandstone (Figures 7, 10) that is interpreted to be in the

channel facies.

Middle Part of East Ford Unit

Ramsey 1 and 2 sandstones are both targets in the cen-

ter of the unit. The pressure response here has been

slow during the CO2 flood (Figure 13b), suggesting

that this area is in poor communication with injectors

Orla Petco 14 East Ford Unit (and Orla Petco 17 East

Ford Unit before it was converted from an injector to a

producer) to the north and Orla Petco 25 East Ford

Unit to the south. No injectors are located in this area,

and well 19 is the only producer (Figure 13b). Orla

Petco 19 East Ford Unit is not responding to injection

in Orla Petco 14 East Ford Unit or Orla Petco 25 East

Ford Unit; production during March 2000 was only 4

BOPD.

Communication between Orla Petco 25 East Ford

Unit and Orla Petco 19 East Ford Unit may be limited

in the Ramsey 1 sandstone because the channel appar-

ently makes a large bend to the east in this part of the


field (Figure 10a). The Ramsey 1 sandstone is thinner

in Orla Petco 19 East Ford Unit (Figure 6a), net/gross

sandstone is lower than in the wells to the north and

south (Figure 6b), and the percentage of calcite-cemented

sandstone is higher. All these factors may restrict com-

munication between Orla Petco 19 East Ford Unit and

Orla Petco 25 East Ford Unit. By adding an injector

well to this area, such as Orla Petco 20 East Ford Unit

and making Orla Petco 18 East Ford Unit, Orla Petco

21 East Ford Unit, and Orla Petco 22 East Ford Unit

producers may improve production from this appar-

ently isolated area.

South Part of East Ford Unit

The south area of the unit is mostly in the lobe facies of

the Ramsey 1 sandstone (Figure 10a), but lobe deposits

of the Ramsey 2 sandstone (Figure 10b) probably also

contribute to production. This area is responding well

to the existing north-south line of injectors. Wells Orla

Petco 27 East Ford Unit, Orla Petco 28 East Ford

Unit, and Orla Petco 31 East Ford Unit are among the

best wells in the field. Recovery in this area is interpret-

ed to be good because the injection and production

wells are all in a laterally continuous lobe sandstone.

Recovery might be improved by bringing on addition-

al producers.

CONCLUSIONS

The Ramsey sandstone interval of the Bell Canyon

Formation in East Ford field was characterized using

subsurface log and core data, but the interpretation

was guided by the depositional model developed from

study of well-exposed outcrop analogs in the Bell Can-

yon Formation. The Ramsey sandstones are interpreted

as having been deposited in a basin-floor setting by a

system of leveed channels having attached lobes and

overbank splays. Individual channel-levee and lobe com-

plexes stack in a compensatory fashion and are separated

by laterally continuous, laminated siltstones. Reservoir

sandstones consist of (1) broadly lenticular lobe de-

posits, (2) elongate channel deposits, and (3) irregular

splay deposits. The depositional model developed from

outcrop can be widely applied by operators to other

reservoirs that produce from Bell Canyon sandstones.

Enhanced oil recovery by CO2 flood has increased

production from these deep-water sandstone reser-

voirs. As a result of the CO2 flood, production from

the East Ford unit increased from 30 bbl/day at the

end of primary production to more than 185 bbl/day

in 2001. The unit produced 216,6297 bbl of oil from

the start of tertiary recovery through December 2001,

and essentially all this production can be attributed

to the EOR project.

Geologic heterogeneities appear to influence re-

sponse to the CO2 flood in the East Ford unit. The most

important causes of heterogeneity in the Ramsey sand-

stone reservoirs are the presence of siltstone beds, var-

iations in net/gross sandstone, and calcite-cemented

sandstone layers. All these heterogeneities have the

potential to disrupt displacement operations in Dela-

ware sandstone reservoirs by limiting cross-flow of fluids

between injector and producer wells. Locating an ade-

quate number of injectors and producers in the same

depositional facies, whether channels, splays, or lobes,

will maximize the reservoir volume contacted and the

production rate and minimize the displacement across

restrictive depositional barriers and the time to produce

the reservoir. EOR projects undertaken without inte-

grated geologic and engineering reservoir characteriza-

tion will not realize maximum potential production.

REFERENCES CITED

Asquith, G. B., S. P. Dutton, A. G. Cole, M. Razi, and J. I. Guzman,1997, Petrophysics of the Ramsey sandstone, Ford Geraldineunit, Reeves and Culberson counties, Texas: West TexasGeological Society Publication no. 97-102, p. 61–69.

Barton, M. D., 1997, Facies architecture of submarine channel-leveeand lobe sandstones: Permian Bell Canyon Formation, DelawareMountains, west Texas: University of Texas at Austin, Bureau ofEconomic Geology Field Trip Guidebook, 40 p.

Barton, M. D., and S. P. Dutton, 1999, Outcrop analysis of a sand-rich, basin-floor turbidite system, Permian Bell Canyon Forma-tion, west Texas, in T. F. Hentz, ed., Transactions: Gulf CoastSection SEPM 19th Annual Bob F. Perkins Research Con-ference, December 5–8, Houston, p. 53–64.

Basham, W. L., 1996, Delaware Mountain Group sandstone channelorientations: implications for sediment source and deposition,in W. D. DeMis and A. G. Cole, eds., The Brushy Canyon playin outcrop and subsurface: concepts and examples: SEPM(Permian Basin Section) Publication 96-38, p. 91–102.

Beaubouef, R. T., C. Rossen, F. B. Zelt, M. D. Sullivan, D. C. Mohrig,and D. C. Jennette, 1999, Deep-water sandstones, BrushyCanyon Formation, west Texas: AAPG Continuing EducationCourse Notes Series 40, 48 p.

Berg, R. R., 1979, Reservoir sandstones of the Delaware MountainGroup, southeast New Mexico, in N. M. Sullivan, ed., Guada-lupian Delaware Mountain Group of west Texas and southeastNew Mexico, Symposium and Field Trip Conference Guide-book: SEPM (Permian Basin Section) Publication 79-18,p. 75–95.

Bouma, A. H., 1962, Sedimentology of some flysch deposits: Agraphic approach to facies interpretation: Amsterdam, Else-vier, 168 p.

Bouma, A. H., 1996, Initial comparison between fine- and coarse-grained submarine fans and the Brushy Canyon Formation

Dutton et al. 625

sandstones, in W. D. DeMis and A. G. Cole, eds., The BrushyCanyon play in outcrop and subsurface: concepts andexamples: Guidebook, SEPM (Permian Basin Section) Publica-tion 96-38, p. 41–50.

Dewan, J. T., 1984, Essentials of modern open-hole log interpreta-tion: Tulsa, PennWell Publishing, 345 p.

Dutton, S. P., and M. D. Barton, 1999, Application of outcropanalogs to reservoir characterization of Permian deep-watersandstones, Bell Canyon Formation, Ford Geraldine Unit, westTexas (Delaware basin), in T. F. Hentz, ed., Transactions: GulfCoast Section SEPM 19th Annual Bob F. Perkins ResearchConference, December 5–8, Houston, p. 65–76.

Dutton, S. P., and W. A. Flanders, 2001a, Application of advancedreservoir characterization, simulation, and production optimi-zation strategies to maximize recovery in slope and basin clasticreservoirs, west Texas (Delaware basin): University of Texas atAustin, Bureau of Economic Geology, final report prepared forU.S. Department of Energy, DOE/BC/14936-18, 170 p.

Dutton, S. P., and W. A. Flanders, 2001b, Deposition and diagenesisof turbidite sandstones in East Ford Field, Bell CanyonFormation, Delaware basin, Texas: AAPG Southwest Section,2001 Annual Meeting Papers and Abstracts, Dallas, Texas,March 10–13, 12 p.

Dutton, S. P., G. B. Asquith, W. A., Flanders, J. I. Guzman, andH. H. Zirczy, 1999a, New techniques for using old geophysicallogs in reservoir characterization: examples from Bell Canyonsandstones, Ford Geraldine and East Ford units, Delawarebasin, Texas: Proceedings of the 1999 Oil & Gas ConferenceTechnology Options for Producer Survival, Conference spon-sored by the U.S. Department of Energy, Dallas, PublicationDOE/FETC-99/1103, 25 p.

Dutton, S. P., M. D. Barton, G. B. Asquith, M. A. Malik, A. G.Cole, J. Gogas, J. I. Guzman, and S. J. Clift, 1999b, Geologicand engineering characterization of turbidite reservoirs, FordGeraldine unit, Bell Canyon Formation, west Texas: Universityof Texas at Austin, Bureau of Economic Geology Report ofInvestigations no. 255, 88 p.

Dutton, S. P., W. A. Flanders, H. H. Zirczy, and J. I. Guzman, 1999c,Geologic and engineering characterization of East Ford field,Reeves County, Texas: University of Texas at Austin, Bureau ofEconomic Geology, topical report prepared for the U.S.Department of Energy, DOE/BC/14936-11, 113 p.

Dutton, S. P., M. D. Barton, H. H. Zirczy, and W. A. Flanders, 2000,Characterization of reservoir heterogeneity in slope and basinclastic reservoirs, Bell Canyon Formation, Delaware basin,Texas: Transactions, Southwest Section AAPG, publicationno. 2000-107, p. 116–129.

Dykstra, H., and R. L. Parsons, 1950, The prediction of oil recoveryby water flooding, in Secondary recovery of oil in the UnitedStates, 2d ed.: New York, American Petroleum Institute,p. 160–174.

Ewing, T. E., 1990, Tectonic map of Texas: University of Texas atAustin, Bureau of Economic Geology, scale 1:750,000, 4 sheets.

Fischer, A. G., and M. Sarnthein, 1988, Airborne silts and dune-derived sands in the Permian of the Delaware basin: Journal ofSedimentary Petrology, v. 58, p. 637–643.

Flanders, W. A., and R. M. DePauw, 1993, Update case history:Performance of the Twofreds Tertiary CO2 project: Society ofPetroleum Engineers Annual Technical Conference andExhibition, SPE Paper 26614, 10 p.

Galloway, W. E., and D. K. Hobday, 1996, Terrigenous clasticdepositional systems: applications to fossil fuel and ground-water resources, 2d ed.: New York, Springer-Verlag, 489 p.

Galloway, W. E., T. E. Ewing, C. M. Garrett Jr., N. Tyler, and D. G.Bebout, 1983, Atlas of major Texas oil reservoirs: University ofTexas at Austin, Bureau of Economic Geology, 139 p.

Gardner, M. H., 1992, Sequence stratigraphy of eolian-derivedturbidites: patterns of deep-water sedimentation along an aridcarbonate platform, Permian (Guadalupian) Delaware Moun-tain Group, west Texas, in D. H. Mruk and B. C. Curran, eds.,Permian Basin exploration and production strategies: applica-tions of sequence stratigraphic and reservoir characterizationconcepts: West Texas Geological Society Publication 92-91,p. 7–12.

Gardner, M. H., 1997a, Characterization of deep-water siliciclasticreservoirs in the upper Bell Canyon and Cherry Canyon for-mations of the northern Delaware basin, Culberson and ReevesCounties, Texas, in R. P. Major, ed., Oil and gas on Texas statelands: an assessment of the resource and characterization of typereservoirs: University of Texas, Bureau of Economic GeologyReport of Investigations no. 241, p. 137–146.

Gardner, M. H., 1997b, Sequence stratigraphy and hydrocarbonhabitat of the Permian (Guadalupian) Delaware MountainGroup, Delaware basin, west Texas, in R. P. Major, ed., Oil and gason Texas state lands: an assessment of the resource and charac-terization of type reservoirs: University of Texas, Bureau ofEconomic Geology Report of Investigations no. 241, p. 147–157.

Gardner, M. H., and J. M. Borer, 2000, Submarine channel archi-tecture along a slope to basin profile, Brushy Canyon Formation,west Texas, in A. H. Bouma and C. G. Stone, eds., Fine-grainedturbidite systems: AAPG Memoir 72/SEPM Special Publica-tion 68, p. 195–214.

Grant Jr., P. R., and R. W. Foster, 1989, Future petroleum provincesin New Mexico— discovering new reserves: New Mexico Bu-reau of Mines and Mineral Resources, 94 p.

Hills, J. M., 1984, Sedimentation, tectonism, and hydrocarbongeneration in Delaware basin, west Texas and southeasternNew Mexico: AAPG Bulletin, v. 68, p. 250–267.

Holtz, M. H., 1995, Characteristics of the Delaware submarine fansandstone play, southwestern U.S.A.: University of Texas atAustin, Bureau of Economic Geology, Internal Report, 8 p.

Jacka, A. D., 1979, Deposition and entrapment of hydrocarbons inBell Canyon and Cherry Canyon deep-sea fans of the Delawarebasin, in N. M. Sullivan, ed., Guadalupian Delaware MountainGroup of west Texas and southeast New Mexico: SEPM(Permian Basin Section) Publication 79-18, p. 104–120.

Kerans, C., and W. M. Fitchen, 1995, Sequence hierarchy and faciesarchitecture of a carbonate-ramp system: San Andres Forma-tion of Algerita Escarpment and western Guadalupe Moun-tains, west Texas and New Mexico: University of Texas atAustin, Bureau of Economic Geology Report of Investigationsno. 235, 86 p.

Kerans, C., W. M. Fitchen, M. H. Gardner, M. D. Sonnenfeld, S. W.Tinker, and B. R. Wardlaw, 1992, Styles of sequence devel-opment within uppermost Leonardian through Guadalupianstrata of the Guadalupe Mountains, Texas and New Mexico,in D. H. Mruk and B. C. Curran, eds., Permian basin explorationand production strategies: applications of sequence strati-graphic and reservoir characterization concepts: West TexasGeological Society Publication 92-91, p. 1–6.

Kirkpatrick, R. K., W. A. Flanders, and R. M. DePauw, 1985,Performance of the Twofreds CO2 injection project: Society ofPetroleum Engineers Annual Technical Conference andExhibition, SPE Paper 14439, 12 p.

Kosters, E. C., D. G. Bebout, S. J. Seni, C. M. Garrett Jr., L. F.Brown Jr., H. S. Hamlin, S. P. Dutton, S. C. Ruppel, R. J. Finley,and N. Tyler, 1989, Atlas of major Texas gas reservoirs:University of Texas at Austin, Bureau of Economic GeologySpecial Publication, 161 p.

Lewis, P., S. Cross, and G. Hoose, 1996, Delaware MountainGroup production map, in W. D. DeMis and A. G. Cole, eds.,The Brushy Canyon play in outcrop and subsurface: Concepts


and examples: SEPM (Permian Basin Section) Publication96-38, separate map in back pocket.

Meissner, F. F., 1972, Cyclic sedimentation in Middle Permian strataof the Permian basin, west Texas and New Mexico, in J. G. Elamand S. Chuber, eds., Cyclic sedimentation in the Permian Basin,2d ed.: West Texas Geological Society, p. 203–232.

Montgomery, S. L., J. Worrall, and D. Hamilton, 1999, DelawareMountain Group, west Texas and southeastern New Mexico, acase of refound opportunity: Part 1— Brushy Canyon: AAPGBulletin, v. 83, p. 1901–1926.

Montgomery, S. L., D. Hamilton, T. Hunt, and J. Worrall, 2000,Delaware Mountain Group, west Texas and southeastern NewMexico, a case of refound opportunity: Part 2— CherryCanyon: AAPG Bulletin, v. 84, p. 1–11.

Newell, N. D., J. K. Rigby, A. G. Fischer, A. J. Whiteman, J. E.Hickox, and J. S. Bradley, 1953, The Permian reef complex ofthe Guadalupe Mountains region, Texas and New Mexico: SanFrancisco, Freeman and Co., 236 p.

Payne, M. W., 1976, Basinal sandstone facies, Delaware basin, westTexas and southeast New Mexico: AAPG Bulletin, v. 60,p. 517–527.

Pittaway, K. R., and R. J. Rosato, 1991, The Ford Geraldine unitCO2 flood— update 1990: Society of Petroleum EngineersReservoir Engineering, v. 6, no. 4, p. 410–414.

Ross, C. A., and J. R. P. Ross, 1987, Late Paleozoic sea levels anddepositional sequences, in C. A. Ross and D. Haman, eds.,Timing and depositional history of eustatic sequences:

constraints on seismic stratigraphy: Washington, D.C., Cush-man Foundation for Foraminiferal Research, Special Publica-tion 24, p. 137–149.

Ruggiero, R. W., 1985, Depositional history and performance of aBell Canyon sandstone reservoir, Ford-Geraldine field, westTexas: Master’s thesis, University of Texas at Austin, Austin,Texas, 242 p.

Ruggiero, R. W., 1993, Depositional history and performance of aPermian Bell Canyon sandstone reservoir, Ford-Geraldine field,west Texas, in E. G. Rhodes and T. F. Moslow, eds., Marineclastic reservoirs: New York, Springer-Verlag, p. 201–229.

Williamson, C. R., 1978, Depositional processes, diagenesis andreservoir properties of Permian deep-sea sandstones, BellCanyon Formation, Texas— New Mexico: Ph.D. dissertation,University of Texas at Austin, Austin, Texas, 262 p.

Williamson, C. R., 1979, Deep-sea sedimentation and stratigraphictraps, Bell Canyon Formation (Permian), Delaware basin, inN. M. Sullivan, ed., Guadalupian Delaware Mountain Groupof west Texas and southeast New Mexico, Symposium andField Trip Conference Guidebook: SEPM (Permian BasinSection) Publication 79-18, p. 39–74.

Zelt, F. B., and C. Rossen, 1995, Geometry and continuity of deep-water sandstones and siltstones, Brushy Canyon Formation(Permian) Delaware Mountains, Texas, in K. T. Pickering,R. N. Hiscott, N. H. Kenyon, F. R. Lucchi, and R. D. A. Smith,eds., Atlas of deep water environments, architectural style inturbidite systems: London, Chapman and Hall, p. 167–183.

Dutton et al. 627

reservoir characterization of a authors permian deep … · we thank mr. thomas b. cowden, of the...

Documents