Deep Knox gas play grows in Black Warrior basin

June 3, 2002
Interest has been intermittent in the Knox and other objectives in the deeper portions of the Black Warrior basin for several years.

Interest has been intermittent in the Knox and other objectives in the deeper portions of the Black Warrior basin for several years.

It was not until completion of the Sanders-1 well in 1998 by Fina Oil & Chemical Co. in Maben field for an average 6.5 MMcfd rate that this play was truly initiated. Three primary objectives should be considered in this play. They are the Lower Knox, the Upper Knox, and the Devonian, in ascending order. Upper Knox is being developed at Maben field and is the only zone of the three with proven production. Significant gas shows have been encountered in the other two zones, so similar potential exists for them.

Maben gas production

Click here to enlarge image

As seen on a regional map of the Black Warrior basin (Fig. 1), the deep Knox play extends about 130 miles long and 20-50 miles wide from Panola County, Miss., on the northwest to Sumter and Greene counties, Ala.

Click here to enlarge image

Based on the early performance of the wells in Maben field (Fig. 2), ultimate recovery estimates for the entire play are 10 tcf of gas. This is roughly double the proved reserves of the basin shallow shelf to the northeast.

Production from the Sanders well commenced at 200 MMcf/month, or about 6.5 MMcfd, in 1999 and has declined to about 120 MMcf/month, or 4 MMcfd. Projecting the decline curve indicates ultimate production of 16.5 bcf over 33 years.

The Richardson-2 well, completed in early 2000, started at about 165 MMcf/month, or 5.5 MMcfd, and is presently making about 60 MMcf/ month, 2 MMcfd.

The Love Heirs-1, completed in late 2000, is the best well to date with an initial production of about 300 MMcf/ month, or 10 MMcfd. It is now making 200 MMcf/ month, 6.7 MMcfd.

The Brown-1, completed in late 2001, commenced at about 240 MMcf/month, 8 MMcfd.

Initial production rates for these wells average 225 MMcf/month, 7.5 MMcfd, and ultimate reserves average an estimated 15-20 bcf; hence, at a cost of $4.5 million/well payout would occur in 1 year with 14 bcf of reserves remaining. This is pretty good economics.

Structure, basin history

Through most of Early Paleozoic time, Cambrian through Silurian, the area was part of a broad carbonate platform, which resulted in thick sequences of limestone and dolomite.

The basin first became evident in Devonian time when it began to sag along an axis on the southwest side, as siliceous lime and chert accumulated in a thick wedge of as much as 1,200 ft in the southwest and pinching out in the northeast corner of the basin.

In Mississippian time gentle downwarping continued. Lower Mississippian carbonates were deposited on the shallow shelf areas in the north and dark shales in the deeper waters of the basin in the south. During the Upper Mississippian, this depositional pattern was modified by a significant introduction of clastics. A major stream carrying muds and quartzose sands from a northern source area, probably originating in southeastern Canada and flowing southward through the Illinois basin, deposited these sediments in a series of large deltas on the northern shelf of the basin.

Basin subsidence accelerated during Pennsylvanian time and was accompanied by a greater influx of clastics. This thick sand-shale series was deposited in a complex sequence of deltas in the northern and northeastern parts of the basin and grade to deeper shales and silts to the southwest.

The source of these sediments appears to be primarily from the north and northeast; however, it is obvious that a significant contribution came from the southeast from uplifts associated with the initial stages of the Appalachian orogeny.

It was during the latter stages of the downwarping in Pennsylvanian time that the prominent hingeline fault systems were developed. These systems are characterized by large down-to-the-basin faults and smaller up-to-the-basin complementary faults. It is this structural pattern that is responsible for the hydrocarbon traps for most of the more important fields in the basin, including Maben field. Hence, there should be many similar traps along these fault trends in the deeper play.

Following the filling of the basin during Pennsylvanian time, folding and thrust faulting developed from compressional forces on both the southeast and southwest margins of the basin. These folded belts are parts of the Appalachian and Ouachita orogenies, respectively. The entire basin was uplifted, eroded, and tilted to the southwest during late Paleozoic and early Mesozoic times.

Exploration objectives

Objectives in the deep Knox play range in depth from 2,000 ft to 20,000 ft and include strata of Pennsylvanian, Mississippian, Devonian, Ordovician, and Cambrian ages.

The Knox dolomite of Cambro-Ordovician age is a thick sequence of dolomite and dolomitic limestone 6,000 ft thick and offers the greatest potential of all objectives in the play. It is an equivalent of the Arbuckle limestone of Oklahoma and the Ellenburger dolomite of Texas, both prolific producers of oil and gas.

In the Black Warrior basin, however, only minor production had been established in the Knox prior to Fina's ongoing development of Maben field, but wells drilled to the Knox are sparse, especially in the deeper portions of the basin. For ease the Knox can be divided into two units, Upper Knox and Lower Knox, the Lower Knox being equivalent or largely equivalent to the Copper Ridge dolomite of the Appalachians.

The Upper Knox consists of about 400 ft of dolomite at the top and about 2,500 ft of dolomitic limestone. The Lower Knox is about 3,000 ft thick and is predominantly dolomite.

Click here to enlarge image

The limestone intervals are too tight for commercial production, and the dolomites generally have less than 5% of primary intercrystalline porosity. However, karst zones with well-developed solution cavities are fairly common in the dolomites (Fig. 3), and permeabilities are generally enhanced by natural fractures.

These zones are believed to result from the lowering and rising of sea levels, alternately exposing limestone to subaerial weathering, hence the dolomitization of the limestone and the development of solution cavities on a regional scale. It was from such a cavernous zone in the Lower Knox that the Davis Petroleum Corp. Williams 9-6 No. 1 well in northeastern Oktibbeha County tested 2.4 MMcfd of gas in 1997 before loading with salt water.

The Devonian is a thick sequence, 1,000 to 1,200 ft, of chert and siliceous limestone. It is an important oil and gas producer in almost every US Paleozoic basin, such as Hunton limestone in the Anadarko basin, Oklahoma.

Click here to enlarge image

Although porosities are generally low, 6% or less, these brittle rocks are very susceptible to fracturing, especially in fault-associated structures (Fig. 4). Gas shows have been reported from the Devonian in a number of wells in this area, and noncommercial amounts of gas have been tested from four of these wells.

However, perhaps the most significant shows remain untested, as in the Love Heirs-1 at Maben field. Good gas shows were recorded throughout the Devonian in this well with the mud being cut to 8.6 lb/gal from 10.7 lb/gal at one point and losing 10-15 bbl/hr of mud at another point.

Although only marginal productive possibilities have been tested, it is believed that with better recognition of the fracture zones, major gas reserves and production can be established. The Mississippian, averaging 1,500 ft thick, is comprised of marine and deltaic sediments, primarily dark shales, several shoreline sandstones, and a few thin limestones.

Mississippian sandstones have been the primary objectives in the Black Warrior basin, accounting for over 90% of the total production to date. The principal sands in ascending order are the Lewis, Rea, Abernathy, Sanders, and Carter. While most of this production has been farther updip on the shallower shelf, limited production occurred along the northern margin of this deeper play area.

Although these sands are considered to be secondary objectives in this play, the Carter sand has proven productive at four wells in Siloam field, the Sanders sand in one-well Abbott field, and the Rea sand in one-well at Siloam field. Porosities are generally low in these sands, 12% or less, but good gas rates generally follow frac treatments.

The Pennsylvanian consists of a deltaic sequence of sandstone and shale 5,000 ft to over 10,000 ft thick. Turbidites may be included in deeper areas.

Although shows have been encountered in a number of wells in this area, the only commercial production has been in three wells in Siloam field and in one-well Pine Bluff field. The most promising zones occur in the upper part of the Pennsylvanian where porosities of 12-15% are seen in the better developed sands. In general, there is less sand in the lower part of the Pennsylvanian, and the sands that are present tend to be tighter.

Developments

Several companies are active in the area. In the Maben area Fina has completed five productive wells and one dry hole and is completing one well, drilling one well, and has two locations permitted. Also, seismic work is under way in several areas to supplement older conventional seismic data, and lease blocks are being assembled.

With the improvements in drilling and completion technologies as demonstrated by Fina, especially the hydraulic fracturing of these relatively tight rocks, good production rates can be established and large reserves proved. Therefore, it is anticipated that this play should continue for years.

The author

Click here to enlarge image

Stewart Welch is co-owner of Browning & Welch Inc., a privately owned company operating out of Jackson, Miss. He was employed 7 years with the US Geological Survey and 7 years with Shell Oil Co. as an exploration geologist before creation of a partnership in 1965 that was incorporated as Browning & Welch in 1981. He is a graduate of Oklahoma State University.