CLARKSVILLE FIELD, RED RIVER COUNTY, TEX.: PRODUCTION AND FACIES INTERPRETATION

June 10, 1991
Chris H. Reed Petroleum Geologist Tyler, Tex. Clarksville field was discovered in December 1985 by a well targeting a deeper Paleozoic horizon. Since production began in 1986, this field has produced more than 1 million bbl of oil with the appearance of a considerably longer and more lucrative life. The producing horizon is a Jurassic age lithic conglomerate sitting unconformably on the Paleozoic and Triassic structural front of the buried Ouachita range. Facies correlation out of the basin
Chris H. Reed
Petroleum Geologist
Tyler, Tex.

Clarksville field was discovered in December 1985 by a well targeting a deeper Paleozoic horizon.

Since production began in 1986, this field has produced more than 1 million bbl of oil with the appearance of a considerably longer and more lucrative life.

The producing horizon is a Jurassic age lithic conglomerate sitting unconformably on the Paleozoic and Triassic structural front of the buried Ouachita range. Facies correlation out of the basin indicates this unit to be Louark age.

Mapping and compositional analysis indicate the depositional environment of this unit to be an and climate alluvial fan deposited as a bajada complex. This fan system was laid down at the updip margin of the actively forming Mesozoic embayment where it meets the Ouachita structural front.

The significance of this field is demonstrated by the production yield at a relatively shallow depth of 5,800 ft. At this time, similar production to Clarksville field is yet to be encountered anywhere along the Mesozoic rim of the East Texas basin but does represent a viable exploration trend, in addition to being the gateway for future Paleozoic production in the basin.

UNIQUE FIELD?

Clarksville field is unique to the Mesozoic Gulf embayment in that it is the only Upper Jurassic Louark age producing continental deposit in the entire Gulf basin.

This field was deposited within an and alluvial fan system prograding off the Ouachita structural front into the actively forming East Texas Mesozoic basin.

As vast a trend as this encompasses, it is highly unlikely that this is an isolated condition. Although Clarksville field has shown to have significant production, the search for lookalike fields has been hindered by a poor exploration climate as well as some geological misunderstandings.

Clarksville field lies in the area designated as the overthrust and updip Smackover trend, rimming the north and northwest margin of the East Texas basin (Fig. 1).

Paleozoic Ouachita overthrust and updip Jurassic Smackover production have been sporadic exploration targets the past 30 years within this trend. Comparatively, only a small number of exploratory tests have been drilled with only sparse Smackover production being established and no Paleozoic production.

Recent success in Bowie County has brought some attention to the trend, but it has not enjoyed the same activity as its Alabama counterpart.

TECTONIC SETTING

Clarksville field is in central Red River county due south of the town of Clarksville.

The two major tectonic features affecting the area are the Broken Bow anticline and the Ouachita thrust belt. The field is immediately updip from the pinchout of the Jurassic Smackover carbonates (Fig. 2).

Regionally, there is only limited production established updip of the fault line trend with the majority of the Jurassic and Cretaceous production being associated with the Mexia-Talco fault system, Louann salt structures, or the Sabine platform trend.

Clarksville field is on trend with and just updip from Shelton-Bogata field, which produces from Jurassic Cofton Valley, Haynesville, and Smackover formations. This has contributed to some of the confusion as to just exactly which stratigraphic unit is producing at Clarksville field.

Clarksville and Shelton-Bogata fields produce from a similar depth of 5,300-5,500 ft. The field operators have similar names, with Santa Fe Energy Co. operating Shelton-Bogata field and Santa Fe Minerals Inc. (SFM) operating Clarksville field.

Stratigraphic correlations, seismic analysis, and interpretation of the environment of deposition lend a somewhat better understanding to Clarksville field.

FIELD HISTORY

The Clarksville field discovery well, the SFM 1 Willis, was completed flowing in January 1986, with an initial producing rate of 149 b/d of oil.

SFM immediately started development while intense leasing took off across the trend extending from Little River County, Ark., through Hunt County, Tex.

Since first production was established at Clarksville, 24 wells have been completed in the field with 18 wells being active producers and six being classed as inactive or abandoned without any significant production (Fig. 3).

The 18 producers have a cumulative production in excess of 1.27 million bbl of oil through 1990. Production averaged 625 b/d in 1990.

PRODUCING INTERVALS

The structural cross section from northwest to southeast across the field gives more detail with regard to individual well production and pay character (Fig. 4).

Producing intervals are shown along with the stratigraphic relationships in the structural cross section through the field.

The discovery well, the SFM 1 Willis, had produced 178,000 bbl of oil as of January 1991 and averaged 42 b/d in 1990.

The SFM 3 Elder has a similar production history, although its records have been reported with three other wells on the lease having a cumulative production of 505,000 bbl and a daily average of 220 b/d as of January 1991.

Primary recovery on these wells may be in excess of 225,000 bbl/well.

STRATIGRAPHY

Stratigraphic correlations are most easily made by first finding the base of the Louark interval as it sits unconformably on the Paleozoic and/or Triassic sediments.

Downdip at the Herring 1 Drew test, Smackover carbonates mixed with sandstones and Buckner anhydrites were encountered in cores immediately above the unconformity. Updip, the producing interval is a lithic conglomerate.

The zone of interest is overlain by Jurassic Upper Buckner tight carbonates and shales devoid of anhydrite.

Detailed correlations using full log suites indicate a subtle differentiation of the Haynesville and Cotton Valley sand formations. Both of these units produce at adjacent Shelton-Bogata field.

Updip of Clarksville field, the Louark group thins considerably and gains structure. Structurally, it shows an extreme terrace at the base of the interval with extensive thickening across the terrace through the entire Louark group.

SEISMIC RECORD

Geotrak Line GC 28 parallels the line of cross section through the field (Fig. 5). Geotrak Corp., Houston, has furnished this line from its trade data in the area.

The Paleozoic unconformity shows an intense double wavelet signal. A thin slice of Triassic red beds is shown between the Smackover and the Paleozoic unconformity. Downship, the top of the Smackover signal is an easily recognized intensive double wavelet.

Going updip, this signal becomes a less intense single wavelet, terminating against a wedge shaped zone of almost random wavelet character. This transition is from a Smackover marine carbonate regime, up the shoreface to a detrital clastic mass at the toe of a prominent Paleozoic escarpment.

The Buckner-Haynesville interval shows thinning across the field area to a point of nonrecognition from the overlying Cotton Valley sands.

The lower left part of Line GC 28 shows high angle Paleozoics of the Ouachita group (Fig. 5). Their structure is the result of the Ouachita thrust faulting that occurred through the late Pennsylvanian period.

As the Mesozoic basin formed, Triassic red beds were deposited. Jurassic seas entered the newly forming basin. Jurassic carbonates and evaporates were deposited in an arid sand starved environment.

The penecontemporaneous continental depositional processes occurring updip remained in a very localized stable position with little further transport or alteration under the existing arid climate conditions.

These sediments were eventually drowned by further marine transgression through the remainder of the Louark group deposition.

ROCK UNITS

A stratigraphic nomenclature chart has been devised to show the relationships of the rock units at Clarksville field (Fig. 6).

The interval indicated on the electric log is the producing zone and is shown as the non-marine equivalent to the Smackover and Buckner formations within the upper Jurassic Louark group.

The lithology of this interval is that of a lithic conglomerate with interbedded sandstones. It is felt that the name "Louark conglomerate" would be a more accurate designation for the unit than the current usage of Cotton Valley.

The occurrence of a texturally and compositionally immature lithic conglomerate is more compatible to the arid climate environment of the Louark group than that of the highly altered and transport intensive environment of the Cotton Valley group.

In this area, the Louark conglomerate has been overstepped by the upper Buckner and Haynesville formation within the same group. The Bossier formation of the overlying Cotton Valley group is not present, with only the Schuler sands prograding over the section further into the basin.

LOUARK ON LOGS

A detailed view of the Louark conglomerate is shown on the composite log from the SFM 1 Elder well incorporating the sonic-induction and microlog (Fig. 7).

The producing interval is shown along with a zone that was unsuccessfully tested.

The resistivity curves show little character with Rt ranging from 12-18 fIM. The only character change is a permeability invasion profile of 8-10 fIM between the focus curve and the medium and deep curves. The gamma ray and SP curves show little response.

The sonic curve has no particular quantitative response, reading a homogeneous 73 lisec/ft through the entire unit. This velocity has proved to be an excellent qualitative tool for identifying the Louark conglomerate.

The density and neutron curves were omitted from this paper as they were somewhat less dramatic than the sonic responses.

The caliper log indicates that the overall Louark conglomerate drills as a gauge hole the overlying Buckner-Haynesville section are very prone to washout while drilling. Zones of production build excellent filter cake while drilling.

The microlog is the most diagnostic of all logging tools used to evaluate this interval, demonstrating excellent separation response in zones of production.

The most pervasive responses within the log suite are those effected by permeability. This zone is very difficult to evaluate by open hole logs alone. If not for a marginal mud log show in the discovery well, this zone may well not have been tested.

RESERVOIR PROPERTIES

The 1 Elder was cored as indicated on the composite log.

Due to the unconsolidated nature of conglomerates, coring operations must be undertaken with great care.

Core analysis supports the significance of permeability. Permeability averaged 30 md. Core porosity is 13%, with total and connate water saturations being the same, 40%. Oil saturation is 22.3%.

Interbedded permeable sand lobes associated with conglomerates are most diagnostic of alluvial fan deposits.

FAN SYSTEMS

Dry climate alluvial fans are smaller, more localized, and are deposited at higher angles than their wet climate counterparts. They have three distinct facies tracts: the proximal fan, the mid fan, and the distal fan.

Within the proximal fan facies, the dominant transport mechanism is gravity slide. This facies is characterized by debris flows of poorly sorted sediments and is not conducive to reservoir formation yet provides the necessary seal for hydrocarbon accumulation.

The mid fan facies develops at the intersection point. This is a point at which slope angle of deposition has diminished to such that gravity slide no longer transports the massive debris flows.

At this point in the transport process, the large poorly sorted sediments drop out and in fact act as sieves screening and sorting the remaining sediments that are being transported by streamflow channels and stream floods across the mid fan facies.

These are the coarse to medium grained channel deposits that form the reservoir facies.

The proximal fan is characterized by eolian and playa deposits.

The field cross section shows facies tracts from the arid fan model (Fig. 4). The contributing production from the 3 Elder and the 1 Willis are shown to be in the mid fan facies.

Updip proximal fan production at the Jaco 1 Dawson was not established successfully as was the case in the downdip distal fan facies completion attempt at the SFM 2 Wright.

STRUCTURE, THICKNESS

Structural mapping on the base of the Louark conglomerate shows that the field is situated in a re-entrant with only monoclinal dip.

One colleague described the field as being in a ditch. This re-entrant is bounded on the east and west by positive Paleozoic basement features.

The encroachment of the Smackover shoreline forms a transitional region with both clastic detrital sediments and carbonates being present.

The structure of the top of the Louark conglomerate shows no dramatic change from that of the base. There is a slight change in the rate of dip due to the thickening of the fan.

Isopach mapping of the gross Louark conglomerate interval shows a lobate geometry, thinning updip and downdip. The updip limit is transitional from the proximal fan to the alluvial highlands.

An arbitrary cutoff of 50 ft has been used for the updip limit of the fan. The total gross thickness exceeds 250 ft in the mid fan. The proximal fan thins to zero.

An isoperm map or map of net permeable interval within the Louark conglomerate has been superimposed and highlighted on the isopach map (Fig. 9). Micrologs, invasion profiles, and filter cake were used to determine net permeability.

The thickest interval of net permeability reaches 110 ft. The permeable lobes are distributed within the mid fan facies. Better wells have in excess of 50 ft of net permeable intervals.

FAN INTERPRETATION

Fan facies tracts have been interpreted as shown (Fig. 10).

A point must be made that this is not one single fan but several fan systems fed from several sources. This type of dry fan system is referred to as a bajada system.

Medium to course grained sandstones deposited in the mid fan comprise the reservoirs. These sands have undergone little authigenic alteration and are very poorly cemented.

The appearance of chlorite-sourced from the underlying and updip Stanley shale, which is regionally a phyllite-has been noted in the productive intervals. The chlorite has coated the grains, inhibiting massive cementation.

The distal fan is characterized by finer grain sandstone deposited as streamflood sheet sands and then being reworked by wind and upper shore facies processes.

SUMMARY

The type of production at Clarksville field is a cost effective significant frontier trend model that need not be ignored.

This field may ultimately produce more than 3 million bbl of oil on primary recovery once it is fully developed and closer spacing is determined. This volume could triple after secondary recovery operations are implemented.

Establishing the stratigraphy and morphology of the unit will aid in the search for lookalikes. It is the author's conclusion that it is a Louark age continental deposit being laid down as an arid alluvial fan.

As further exploration for this target proceeds, deeper data from below the Mesozoic will significantly add to the underlying Paleozoic framework.

This could open an entirely new frontier for oil and gas exploration.

Copyright 1991 Oil & Gas Journal. All Rights Reserved.