Barnett Excerpt

Document Sample
Barnett Excerpt Powered By Docstoc
					Petroleum Frontiers Excerpt

The following is excerpted and adapted from Barnett Shale: A New Gas Play in the Fort Worth Basin, authored by Scott Montgomery, the latest in IHS Energy’s Petroleum Frontiers series, a quarterly investigation into the most promising hydrocarbon horizons and provinces. The 78-page publication includes more than 75 figures and an extensive bibliography. The material presented here is from Chapter 4. For more information on IHS Energy’s Petroleum Frontiers series, call (888) 645-3282.

Barnett Shale: Detailed Discussion
Barnett Stratigraphy Regional stratigraphic relationships for the Barnett Shale are fairly well established. As noted in the previous chapter, the formation is overlain by interbedded shale and limestone of the lower Marble Falls/Comyn interval. It overlies the regional unconformity capping the Lower Paleozoic (Ellenburger, Viola/Simpson), except in the northwestern part of the basin. Here, extending eastward from the Bend Arch, the basal portion of the Mississippian is formed by a sequence of carbonate bank deposits known as the Chappel Shelf (Figure 4.1). The Barnett thins rapidly over these deposits, but remains conformable above them. These relationships are shown by the cross sections in Figure 4.2. Stratigraphic divisions within the Barnett are less well established. Several schemes have been proposed, with varying degrees of complexity and basin-wide applicability (see, for example, Henry, 1982, for the northern basin area). Current usage by operators involved in the Barnett play is shown on Figure 4.3. Divisions include an upper Barnett interval (50-150 ft thick), the Forestburg Limestone (where present, 10-200 ft thick), and a lower Barnett interval (50-400 ft). Definition of upper and lower Barnett units has been made on the basis of two separate intervals of high radioactivity and resistivity, which occur regionally throughout the basin where the Barnett is more than about 100 ft thick in total. The Forestburg Limestone, which divides the upper and lower Barnett, is confined to the northeastern part of the basin. The Forestburg pinches out near the southern limit of Figure 4.1. Regional isopach data, Barnett Shale. Cross Newark East Field (southern Wise County). South of this section lines refer to Figure 4.2. Source: Pollastro

Page 1

PI/Dwights PLUS Drilling Wire™

Petroleum Frontiers Excerpt

(Petroleum Frontiers Excerpt, cont’d)
NE

FORT WORTH BASIN
Parker 10 Hood

SW

A

Denton Wise Denton Tarrant 1 2 3 4 5 6 7 8 9

Somervell Erath Hamilton Lampasas A' 11 12 13 14 15 16

THICKNESS IN FEET

suggest that the Barnett was deposited on a gentle southsouthwest facing slope, which developed a somewhat rapidly subsiding deep adjacent to the incipiently rising Muenster Arch. As such, the Barnett is interpreted to represent the last stage of quiet-water, shelf-type deposition prior to initiation of the Fort Worth Foreland Basin. Lithology Over most of the Fort Worth Basin and eastern Bend Arch, the Barnett Shale is an organic-rich, siliceous shale with variable amounts of limestone, minor dolomite, and a scattering of exotic minerals. Though often described as a “black shale,” this term can be misleading when applied

NW

SE
BEND ARCH
Palo Pinto 5 6 Hood

B
Well 1 No.

CHAPPEL LIMESTONE SHELF
Throckmorton 2 3 4 Stephens

FORT WORTH BASIN
7

B'

Somervell Johnson Hill 8 9 10

THICKNESS IN FEET

W Marble Falls

E

Figure 4.2. Regional cross sections, illustrating thickness trends and stratigraphic relationships for the Barnett Shale, Fort Worth Basin. Modified from Pollastro, 2004.

(Bank Deposits) Pinnacle Reef

Barnett Shelf

Forestburg Ls

field, division between the upper and lower Barnett is not Chappel Ls standardized and is defined somewhat arbitrarily, on the Lower Barnett basis of one or another log marker. Ellenburger Where it outcrops along the northern flanks of the ViolaLs Llano Uplift, the Barnett is relatively thin (<50 ft) and Simpson undivided. It is also undifferentiated over the Bend Arch and Chappel Shelf (see Figure 4.1). In the northeasternNewark most part of the basin, where the Barnett reaches thickEast Field nesses of over 900 ft, the formation contains a number of limestone intervals that complicate the simple upper/ Figure 4.3. Stratigraphic diagram showing common lower divisional scheme. divisions in Newark East Field and areas to the west, Fort Stratigraphic relationships and general lithology Worth Basin-Bend Arch.

PI/Dwights PLUS Drilling Wire™

Page 2

Petroleum Frontiers Excerpt

(Petroleum Frontiers Excerpt, cont’d) to the Barnett. In fact, the formation is rich in silica (35%-50%, by volume) and relatively poor in clay minerals (less than 35%, usually). It is, therefore, lithologically distinct from various well-known black shales such as the Antrim, Bakken, Chattanooga or Woodford. Silica in the Barnett is predominantly derived from radiolarian tests. Important lithologic changes occur in the Barnett from south to north. Where it outcrops along the Llano Uplift, the Barnett is 35-50 ft thick and contains highly petroliferous intervals (Turner, 1957), with up to 13% total organic carbon by weight (Jarvie, 2001). In the subsurface, the formation thickens northeastward from 50 ft, close to the Llano Uplift and along the Bend Arch, to 400 ft in Newark East Field, to over 1,000 ft adjacent to the Muenster Arch. South and west of a line through northern Palo Pinto, Parker and Tarrant counties, the Barnett is generally poor in carbonate material, containing only a few thin limestones. Northeast of this line, however, in Wise, Denton and Montague counties, the amount of carbonate increases rapidly. The Forestburg Limestone changes from a feather edge in southern Wise County to being over 200 ft thick about 15 miles north. Even further northeast, especially in Montague County, the Barnett contains multiple debris-flow limestone intervals, along with calcareous shale, testifying to erosion of exposed Paleozoic carbonates from the Muenster Arch in the late Mississippian. The subsurface Barnett is generally dominated by siliceous shale. The basal portion of the Barnett in the central and eastern part of the basin frequently contains a thin zone (<10 ft) of highly phosphatic material (Bowker, 2002), making for easy distinction on logs and in mud log samples from underlying Viola or Ellenburger deposits. Organic content is highest (3%-10%, by weight) in the most silica-rich intervals. These intervals are also the primary producing facies of the Barnett. They are generally more abundant in the lower part of the interval, but are significant in the upper Barnett as well. According to Bowker (2002), the average composition of this facies (by volume) is as follows: 45% quartz; 27% illite, with minor smectite; 8% calcite + dolomite; 7% feldspar; 5% organic matter; 5% pyrite; and 3% siderite, with trace amounts of native copper and phosphatic minerals. In the northern basin, where the Barnett thickens rapidly, the formation contains an increasing number of carbonate-rich zones, including what appear to be debris flow-type deposits (Bowker, 2002). In the western part of the basin, meanwhile, along the flanks of the Bend Arch, fine-grained calcareous material is fairly abundant in the lower Barnett, due to wave and current distribution of debris from Chappel reefs (Henry, 1982). In addition, the Barnett appears to change character locally over erosional and karst-related highs. Differential compaction apparently has caused thinning and enhanced alignment of clay particles in the formation across such features (Zhao, 2004). This seems to have reduced porosity/permeability in the Barnett still further, and to have increased rock brittleness. Whether other changes— such as relate directly to composition—occur in such settings is not known. Log Character In general, the Barnett is distinguished by very high radioactivity and high resistivity log signatures, which differentiate it easily from overlying Marble Falls/Comyn and underlying Viola/Ellenburger carbonate-bearing intervals. Gamma ray values of up to 300 API units and resistivities of 100-1,000 ohm-meters are common. Gamma ray values, in particular, tend to be highest where organic material is especially abundant. As shown on Figure 4.4, the lowermost portion of the Marble Falls, which consists of interbedded carbonate and shale, contains a rather prominent limestone marker known informally as the “Barnett Limestone.” This marker lies about 25-40 ft above the first high-radioactivity shale bed used to designate the top of the Barnett. The upper Barnett interval commonly includes several zones with gamma ray values above 100 API units and most of the entire interval above 75 API units (Figure 4.4). When present at significant thickness (>20-25 ft), the Forestburg appears on logs as a distinct decrease in gamma ray

Page 3

PI/Dwights PLUS Drilling Wire™

Petroleum Frontiers Excerpt

(Petroleum Frontiers Excerpt, cont’d)

Barnett Ls Marker

Top Barnett

Upper Barnett

Forestburg Ls

Figure 4.4. Type log showing detailed character of upper Barnett. Courtesy: Star of Texas Energy.

PI/Dwights PLUS Drilling Wire™

Page 4

Petroleum Frontiers Excerpt

(Petroleum Frontiers Excerpt, cont’d) values to less than 40 API units. Below the Forestburg, the lower Barnett appears as a second interval of high gamma ray/high resistivity signatures (Figure 4.5). Also notable on Figures 4.4 and 4.5 is the decrease in bulk density through the non-Forestburg Barnett (RHOB values as low as 2.4-2.5 g/cc). Formation evaluation using log data in the Barnett has proved challenging, due to the unconventional nature of the reservoir. One series of approaches has been published by Johnston (2004) and is shown in Table 4.1. This author notes that, in addition to high gamma ray and resistivity measurements, Barnett reservoirs typically show density porosity of 0-16 units and neutron porosity slightly to much higher than this (see Figure 4.4 and 4.5). Johnston (2004) emphasizes the use of resistivity and FMI data for identifying fractures with sufficient aperture to contribute to productivity, and he also outlines the use of calibrated log data for identifying clay-rich and silica-rich intervals. An example of such calibrated data, used to help locate potentially productive intervals, is shown in Figure 4.6. It should be mentioned, however, that the importance of natural fractures to well performance is strongly contested by other workers familiar with the play, who stress that such fractures are normally mineralized and thus act as flow barriers. Thus, the discussion offered by Johnston (2004) may not be representative. Basic Reservoir Characteristics The Barnett Shale is an unconventional reservoir with low porosity and very low permeability. Porosities in productive siliceous shale intervals average 6%, with permeabilities varying from 0.01 md down to the nano-darcy range. The reservoir has no free water; measured water saturations of
Viola

Forestburg Ls

Lower Barnett

Figure 4.5. Type log, lower Barnett. Courtesy: Star of Texas Energy.

Page 5

PI/Dwights PLUS Drilling Wire™

Petroleum Frontiers Excerpt

(Petroleum Frontiers Excerpt, cont’d) 35% correspond to water bound up in clay minerals. Gas is held within the Barnett reservoir in two basic ways. It occurs as free gas, whether in matrix porosity, any remnant fracture porosity and—possibly—in microporosity. Gas also exists in sorbed form on organic matter; such gas is therefore particularly abundant in high TOC intervals. It is assumed that production from the Barnett involves two main stages. Initial flow into induced fractures consists of free gas from matrix porosity, any open natural fracture apertures and—possibly—from microporosity. The formation is known to be somewhat overpressured (0.49-0.54 psi/ft pressure gradient), and since water saturations are generally rather low (e.g. 35%), this overpressuring is interpreted to be a result of free gas. A second stage of production begins once a significant portion of the free gas has been withdrawn and reservoir pressure decreases, stimulating increased desorption of gas from organic carbon in the Barnett. The latter portion of this second stage involves long-lived production (flattened rate of decline) of desorbed gas. Hydrocarbons within the Barnett reservoir change east to west and also north to south across the basin (Figure 4.7). In the far eastern portion, dry gas predominates. This changes westward to wet gas, gas with oil

Log Type Resistivity Gamma ray Density Neutron Sonic Micro-resistivity/ Electric images Spectroscopy

Properties Measured Bound water volume, both clay and pores Clay and organic material volume Minerals and fluids content Clay and gas content Clay and gas content Identify natural and drilling-induced fractures, pyrite, calcite nodules, other geologic features Organic carbon content, clay and carbonaceous minerals

Figure 4.6. Calibrated and interpreted log response, Barnett Shale, correlated against perforated zones (shown on left). Source: Johnston, 2004

Table 4.1. Log type and reservoir characteristics. Source: Johnson, 2004

PI/Dwights PLUS Drilling Wire™

Page 6

Petroleum Frontiers Excerpt

(Petroleum Frontiers Excerpt, cont’d)
Red River Arch
WICHITA

U D

CLAY ARCHER BAYLOR

U D

COOKE

MONTAGUE

Oil
THROCKMORTON

YOUNG JACK WISE

D

U

DENTON

Gas w/Oil

CALLAHAM

EASTLAND ERATH

Dry Gas
SOMERVELL

JOHNSON

COMANCHE BOSQUE COLEMAN BROWN HAMILTON

MILLS

MCLENNAN

LAMPASAS SAN SABA

MASON

Llano Uplift
LLANO

BURNET WILLIAMSON

0 0

25 40

50 mi 80 km

Figure 4.7. Patterns of hydrocarbon occurrence in the Barnett Shale, Fort Worth Basin-Bend Arch area. Modified from Jarvie, 2001.

Page 7

Ou

BELL

ach

McCULLOCH

ita

CORTELL

Thrus t

HILL

Fold

HOOD

Bel

SHACKELFORD

t

STEPHENS

PALO PINTO

PARKER

TARRANT

DALLAS

and—finally, along the flanks of the Bend Arch—to mainly oil. Figure 4.7 also shows that oil is present within the Barnett in the northern part of the basin, changing southward to wet gas/gas w/oil, and finally to dry gas. These patterns of occurrence do not correspond well to present-day depth/burial patterns for the Barnett, suggesting that thermal history in the basin has been complex and perhaps multi-phased (Jarvie, 2001; Pollastro, 2004). For most gas wells, the upper Barnett yields 20-25% of the total production, with the lower Barnett delivering 75-80%. Perforated intervals commonly range from about 35-75 ft in the upper Barnett and 100-200 ft in the lower Barnett, depending on thickness of each interval. Data from stress tests indicate that siliceous intervals respond better to hydraulic fracturing than do more clay-rich intervals (Johnston, 2004). It is known that the upper Barnett has a higher frac gradient than the lower Barnett, particularly where the Forestburg Limestone is present. This gradient is 0.70 psi/ft or above for the upper Barnett and 0.50-0.60 (or slightly higher) for the lower Barnett (Martineau, 2003). In addition, where the Barnett is especially thick (>450 ft) and productive, in northern Newark East Field, different productive zones in the lower Barnett have shown different frac gradients. The specific reasons for such variation are not entirely understood. However, it is assumed by many workers that such differences are related to changes in source potential and gas generation within each zone, and to sealing potential of surrounding shales. Some geologists have also proposed that natural fracturing may be involved. Considerable debate exists over the role of natural fractures with regard to Barnett productivity. Some recent publications (see, for example, Johnston, 2004) have emphasized the crucial role of such fracturing. Early reports from “data wells” in the Barnett also focused on the assumed probability that natural fractures were required for economical production (CER Corp., 1992). These reports, however, indicated that fractures were largely or entirely mineralized with calcite and that

M ue ns r te ch Ar

Bend Arch

PI/Dwights PLUS Drilling Wire™

Petroleum Frontiers Excerpt

(Petroleum Frontiers Excerpt, cont’d) stimulation did not open these structures but instead generated new fracture sets with orientations that differed from natural sets by 90-120 degrees. Workers with considerable experience in the play have stated that cores reveal only sealed fractures, which would effectively reduce deliverability (by acting as local flow barriers) (Bowker, pers. comm., 2004). At present, this important issue remains to be resolved. Most core data from the Barnett are still proprietary, due to the ongoing nature of the play. A future study of fracturing in the Barnett, appropriately supported by both analytical data and relevant images, would contribute significantly to an overall understanding of this complex reservoir.

Select Bibliography
Bowker, K.A., 2002, Recent developments of the Barnett Shale ——— , 2004, Barnett Shale-Conclusion: Reservoir characterization improves stimulation, completion practices; Oil and play, Fort Worth Basin; in Law, B.E. and Wilson, M., eds., Gas Journal, Jan. 26, 2004, vol. 102, no. 4, p. 35-39. Innovative Gas Exploration Concepts Symposium: Rocky Mountain Association of Geologists and Petroleum Technology Martineau, D., 2003, Newark East, Barnett Shale Field, Wise Transfer Council, October, 2002, Denver, CO, 16 p. and Denton Counties, Barnett Shale frac gradient variances; (abs.); 2003 AAPG Southwest Section Meeting, Fort Worth, CER Corporation, 1992, Geological, petrophysical and TX; available online at: http://www.fwgs.org/swsec/ engineering analysis of the Barnett Shale in the Mitchell Energy Corporation T.P. Sims No. 2, Wise County, Texas: Gas techsessions.htm. Accessed: February 15, 2004 Research Institute Contract Report No. 5091-212-2242, 83 p. Pollastro, R.M., 2004, Geologic and production characteristics Henry, J.D., 1982, Stratigraphy of the Barnett Shale (Missisutilized in assessing the Barnett Shale continuous (unconvensippian) and associated reefs in the northern Fort Worth Basin; tional) gas accumulation, Barnett-Paleozoic Total Petroleum in C.A. Martin, ed., Petroleum Geology of the Fort Worth System, Fort worth Basin, Texas. Basin and Bend Arch Area: Dallas Geological Society, p. 157178. Turner, G.I., 1957, Paleozoic stratigraphy of the Fort Worth basin; in Bell, W.C., ed., Abilene and Fort Worth Geological Jarvie, D.M. and L.L. Lundell, 1991, Hydrocarbon generation Societies Joint Field Trip Guidebook, p. 57-77. modeling of naturally and artificially matured Barnett shale, Ft. Worth Basin, Texas, Southwest Regional Geochemistry Zhao, H., 2004, Thermal maturation and physical properties Meeting, Sept. 8-9, 1991, The Woodlands, Texas, 1991, oral of Barnett Shale in Fort Worth Basin, North Texas (abs); presentation. American Association of Petroleum Geologists Annual Meeting, 2004; Session on Unconventional Gas; available Johnston, D., 2004, Barnett Shale-1: Technological advances online at: http://aapg.confex.com/aapg/da2004/techprogram/ expand potential play; Oil and Gas Journal, Jan. 19, 2004, vol. A87090.htm. 102, no. 3, p. 51-59.

PI/Dwights PLUS Drilling Wire™

Page 8