The Barnett Shale Visitors Guide to the Hottest Gas Play in the US

PICKERING ENERGY PARTNERS, INC.

1

The Barnett Shale Visitors Guide to the Hottest Gas Play in the US

October 2005 r

Jeff Hayden [email protected] (713) 333-2971

Dave Pursell [email protected] 713-333-2962

2

P I CK E RI N G E N E R GY P A R T N E RS, INC.

This report is meant to help investors better understand the Barnett Shale and the future development potential that exists in the core and non-core areas. Later in the report we will also address which E&P companies have the most exposure to the play and what “lessons learned” we can apply to other emerging North American gas shale plays. Some key takeaways are outlined below. ƒ

ƒ

ƒ

ƒ

ƒ

ƒ

ƒ ƒ

Play economics work well at $6/mcf gas – At $6/mcf gas, Barnett drilling is full steam ahead. Economics work for vertical wells (39% cash flow rate of return) and horizontal wells (113% and 70% returns for Tier 1 and 2 non-core areas, respectively). Wells are still economic if gas falls to $5/mcf though vertical well economics are thin (12% return). Tier 1 and 2 horizontals still look good (73% and 38% returns). Uncertainties remain for Tier 2 – Not enough data exists to conclusively determine whether Tier 2 acreage (Erath, Jack, Palo Pinto, etc) will be successful. To date we have seen results from 6 wells (2 EOG, 4 IFNY). While each produced gas, some liquids (oil & condensate) were seen in the EOG wells and the production rates from the majority of the IFNY wells were not impressive. Additional data points and longer production history required to determine how prolific the play will be in Tier 2. Two words…size matters – The Barnett Shale is a highly complex reservoir. Significant variability of well results exists even within concentrated areas. As the industry has yet to figure out how to identify the good wells from the bad (yes…there are bad wells in the play), a large acreage position is a necessity in order to minimize the risks and allow the law of large numbers to take effect. Not every Barnett well is a good one – We know, we know…blasphemy. Just looking at the wells that were mechanical successes (i.e. gas producers), about 23% of the horizontal wells and 32% of the vertical wells drilled in 2004 would have been uneconomic if drilled at today’s cost levels, assuming a $6/mcf long term gas price. That said, the good wells tend to be really good, making the average results economical. Horizontal wells are superior to vertical wells – Probably not a surprise given the increased industry focus on horizontal wells, but the magnitude was surprising. Using $6/mcf as our benchmark gas price, the typical horizontal well generates a 100%+ return while the typical vertical well generates only a 39% return. Johnson County acreage looking good – Though the core area is commonly referred to as Denton, Wise, and Tarrant counties, the true sweet spot has been the Newark East field, which has been extensively drilled. Results outside Newark East have not been as impressive. However another sweet spot appears to be developing in Johnson County, which looks superior to much of the “core” acreage beyond Newark East. Stock Thoughts – Overall the value of Barnett Shale is appropriately discounted in the stocks. EOG looks the riskiest while CRZO looks interesting. Broad resource play implications (beyond the Barnett): ¾ Size matters ¾ Expect variability of results ¾ Learning curve – development progression takes time ¾ Location, location, location – reservoir parameters, gas window, etc. ***IMPORTANT DISCLOSURES ON PAGE 51 OF THIS REPORT***

PICKERING ENERGY PARTNERS, INC.

3

Table of Contents

Barnett at a Glance..................................................................................................................................... 5 Geological Backdrop ............................................................................................................................. 7 Geochemical Backdrop ......................................................................................................................... 9 The Barnett Today ............................................................................................................................... 11 The Players ................................................................................................................................................ 13 The Data.................................................................................................................................................... 14 Horizontal vs. Vertical wells............................................................................................................... 19 Well Performance Analysis by Company ......................................................................................... 22 Barnett Economics – The Bottom Line............................................................................................... 24 Vertical wells ......................................................................................................................................... 24 Horizontal Wells................................................................................................................................... 26 Summing it Up...................................................................................................................................... 32 Reserve Potential...................................................................................................................................... 34 What’s an Mcf worth anyway? ........................................................................................................... 36 Company-specific Barnett upside ...................................................................................................... 39 EOG – a closer look............................................................................................................................ 40 The Unknowns ......................................................................................................................................... 42 Summary/Conclusions............................................................................................................................ 44 Appendix A – Gas Shale Terminology................................................................................................. 45 Appendix B – Comparison of Organic Shales in the US .................................................................. 46 Appendix C – Example of Barnett NPV Model................................................................................. 47 Appendix D – Generalized Company Acreage Maps ........................................................................ 48

4

P I CK E RI N G E N E R GY P A R T N E RS, INC.

Barnett at a Glance The Barnett Shale is one of the largest and most active domestic natural gas plays in the U.S. Production is ~1.2bcf/d and there are ~100 rigs drilling. It’s likely that most investors are already familiar with the play background/basics, but we review them in the following pages. For a comparison of the Barnett to other productive shales, see Appendix B. When most investors (ourselves included) hear about the Barnett Shale, they immediately think of the play in the Fort Worth Basin, but other “Barnett like” resource plays are emerging in the Permian Basin to the west (Culberson/Reeves counties) and the Fayetteville Shale to the northeast. For the purpose of this report, Barnett Shale will refer to the Fort Worth Basin play unless otherwise specified. The charts below illustrates where the Barnett is located and the key counties involved in the play. We also show the growth of the play in terms of active rigs, wells drilled and gas production. Figure 1 . Location of Barnett Shale

Source: Humble Geochemical, Pickering Energy Partners

PICKERING ENERGY PARTNERS, INC.

5

Figure 2. Barnett Shale Production History 3,500

1,500

3,000

1,200 2,500 900

2,000

1,500

600

1,000 300 500

0

0 1990

1995

2000

2005

Source: IHS Energy and Pickering Energy Partners

Figure 3. Barnett Shale Rigcount History 100 90 80

Number of Rigs

70 60 50 40 30 20 10 0 1992

1994

1996

1998

Source: Smith Bits and Pickering Energy Partners

6

P I CK E RI N G E N E R GY P A R T N E RS, INC.

2000

2002

2004

Number Active Wells

Total Gas Production, mmcf/day

Gas Production Active Well Count

Geological Backdrop The Barnett is a Mississippian-aged Shale located at depths of 6,500-8,500 feet deep. The play could be quite large, potentially spanning 10-15 counties in the Fort Worth Basin of north Texas (the shale is bordered to the east by the Ouachita Thrust-fold Belt and the Muenster Arch and to the west by the Bend Arch). Figure 4 shows the stratigraphy of the Forth Worth Basin. As we head northeast in the play, the Barnett is split into the upper and lower Barnett by the Forestburg limestone. Most of the development where this occurs has focused on the Lower Barnett. Figure 4. Fort Worth Basin Stratigraphic Column

Source: AAPG

PICKERING ENERGY PARTNERS, INC.

7

Other key formations to note in the basin include the Marble Falls limestone and the Viola limestone, which provide the upper and lower frac barriers for the Barnett, and the water-bearing Ellenburger formation. The frac barriers are important because the tight Barnett Shale needs to be hydraulically fractured in order to be productive. Fracs that extend out of the Barnett Shale and into a water-bearing formation will result in an uneconomic completion. Figures 5 and 6 show the east-west and north-south cross sections of the Fort Worth Basin. The play thickens and deepens heading north and east. At its thickest (just south of the Muenster Arch), the Barnett is ~1,000 feet thick, and thins to ~30-50 feet thick as it heads south. The figures also show roughly where the lower Viola frac barrier ends. Given these two factors, it’s easy to understand why the industry chose to develop the northeast corner of the play first, especially since the play was being developed exclusively with vertical wells at the time. Figure 5. East-West Cross Section

Source: Humble Geochemical

Figure 6. North-South Cross Section

Source: Humble Geochemical

8

P I CK E RI N G E N E R GY P A R T N E RS, INC.

Geochemical Backdrop The Barnett Shale has been such a productive reservoir due to high proportion of total organic carbon (TOC) despite having likely leaked much of its gas to surrounding formations/reservoirs over time. The TOC of the Barnett averages ~4.5% (immature outcroppings indicate it was as high as 11-13%). This is important because there is a linear relationship between TOC values and gas content. A high TOC value suggests a large potential to generate hydrocarbons. The Barnett Shale is a thermogenic reservoir. In a thermogenic reservoir, hydrocarbons are created by the combination of time, temperature and pressure. The thermal maturity of the reservoir can help determine whether it contains oil, gas, or no hydrocarbons. Thermal maturity is measured in the lab using core samples by vitrinite reflectance (RO), with higher numbers indicating a greater likelihood of gas. A reading >1.0 usually indicates the gas window, while a reading >1.4 indicates dry gas. A reading <0.6 indicates the reservoir is too immature to produce hydrocarbons. Figure 7. Vitrinite Reflectance Table

gas window

immature

0.0

oil window

0.6

wet gas

1.0

dry gas

1.4

2.0

Vitrinite Reflectance (Ro) Source: Pickering Energy Partners

Figure 8 shows the RO for the Barnett Shale. Once again, the best readings tend to be in the northeast. Also, note that northwestern Jack County and eastern Erath County appear to be in the oil window. Ongoing activity is assessing the potential for gas production in areas commonly thought to be oil prone (i.e. RO<1.0). Time will tell.

PICKERING ENERGY PARTNERS, INC.

9

Figure 8. Barnett Shale Isoreflectance Map

Source: AAPG

10

P I CK E RI N G E N E R GY P A R T N E RS, INC.

The Barnett Today The Barnett Shale has come a long way over the last 5-10 years, as light sand fracs and horizontal drilling have driven an explosion of activity in the play, with over 3,800 wells drilled to date. The play has now expanded well outside the original “core” area. Although most of the wells drilled thus far have been in the core, future activity will be focused on the non-core area. Currently, more than 100 companies are active in the Barnett Shale, and the industry has expanded the rig count in the play to ~100 rigs from only ~30 rigs in 2003. We believe the rig count in the play will only move higher, as many of the larger players are adding rigs over the remainder of the year and will likely have to increase their rig count further in the future in order to hold all of their leases. Production from the Barnett is currently ~1.2bcf/d, accounting for >2% of total domestic gas production. Public companies active in the play include BR, CHK, CRZO, DNR, DVN, ECA, EOG, IFNY, KWK, PLLL and XTO. When thinking about the Barnett Shale, we not only split the play into the core and non-core area but also further subdivide the non-core area into Tier 1 and Tier 2. Figure 9. Barnett Shale County Map

Wise

Denton Newark East Field

Parker Hood

Tarrant Johnson

Core Tier 1 Tier 2 Source: Pickering Energy Partners

Core Area. The vast majority of the Barnett Shale production has been from the Newark East Field in the core area. The Newark East Field covers a portion of Denton, Wise and Tarrant Counties. Much of the initial PICKERING ENERGY PARTNERS, INC.

11

development was performed by Mitchell Energy (now part of Devon Energy). Though the term “core area” is typically used to describe all three of these counties, the true sweet spot is Newark East. The core area has been most commonly drilled with vertical wells and completed with large hydraulic fracture treatments. The ability to execute a large fracture treatment is made possible by a limestone barrier (Viola lime) which separates the Barnett Shale from the underlying water-bearing Ellenberger formation. The Barnett Shale is thickest and deepest in the core area…which corresponds to the highest gas-in-place per section in the Barnett (see geological backdrop section). Tier 1. Geographically, south and west of the core area (Johnson, Hood, and Parker Counties). This portion of the Barnett generally lacks the Viola Limestone which separates the Barnett from the underlying waterbearing formations. Vertical wells with large hydraulic fracture treatments risk communicating with the waterbearing Ellenburger formation. Horizontal drilling has been effectively employed (mainly in Johnson County) in conjunction with multiple (typically four or five) hydraulic fracture treatments along the horizontal well section. These smaller fracture treatments are designed to avoid communication with the adjacent waterbearing zones. Tier 2. Geographically, west and south of Tier 1 (counties include Jack, Erath, Palo Pinto, Hill, etc). This is the least developed area of the Barnett. Conventional analysis has suggested that much of the Barnett Shale in Tier 2 has the likelihood to produce oil (uneconomic volumes) instead of gas. Development is slowly ongoing in Tier 2 as companies attempt to identify the western boundary of the oil-gas window. Production results so far are inconclusive. In addition to uncertainties surrounding the western extent of the gas window, the Barnett Shale thins and is shallower to the west and south. This results in lower amounts of gas-in-place and recovery per section than the Core or Tier 1 areas. Moreover, as in Tier 1, a competent fracture barrier does not exist at the base of the Barnett, driving most operators to utilize (more expensive) horizontal wells to develop the resource. This is the riskiest area in the Barnett Shale. There is no long term production performance. Uncertainties surrounding the gas window and the lower resource potential due to thinner and shallower reservoir make widespread commercial development less certain in Tier 2. These uncertainties increase significantly as the industry tries to push the play even further west and south (into counties such as Comanche and Stephens).

12

P I CK E RI N G E N E R GY P A R T N E RS, INC.

0

2,300

77,500

PLLL

XTO

31,750

ECA

KWK

364,000

DVN

0

0

DNR

IFNY

16,000

CRZO

0

0

CHK

EOG

28,000

BR

Core Acreage

38,500

0

67,500

0

90,000

6,350

171,000

20,800

19,750

48,000

44,000

Tier 1 Acreage

39,000

0

162,500

60,700

400,000

88,900

18,000

22,700

29,250

0

17,000

Tier 2 Acreage

103mmcf/d

~2mmcf/d

13mmcf/d

2mmcf/d

70mmcf/d

70mmcf/d

560mmcf/d

15mmcf/d

6mmcf/d

50mmcf/d

80mmcf/d

Current Production

17

1

2

1

9

6

18

3

4

6

6

Active Rigcount 899 Bcfe (7% of proved reserves)

Another large Barnett producer, ECA holds 127,000 net acres in the Barnett. ~25% of the company's acreage lies in the core area, with the remainder spread around but concentrated in Montague, Cooke, and Bosque counties (ECA definitely has the largest acreage position north of the core area).

The company holds 60,700 net acres in the play, split roughly evenly between Erath and Comanche counties (Tier 2). Assuming both counties are productive (still a big risk in Comanche), IFNY’s net upside potential in the Barnett is huge relative to its proved reserve base of 9bcfe.

~1.5 Tcfe Through its acquisition of Antero, XTO became the second largest producer in (25% of proved the Barnett, with net production of ~100mmcf/d. The company holds ~155,000 reserves) acres in the play, located mainly in Tarrant, Johnson, and Parker counties.

36 Bcfe PLLL is one of the smaller players in the Barnett Shale with only 2,300 net acres, (28% of proved all located near Forth Worth in Tarrant County. reserves)

KWK’s acreage lies primarily in Hood and Somervell counties, where the ~1.9 Tcfe company has a large (230,000 net acres), relatively contiguous position. We (200% of proved estimate ~30% of the company’s acreage is in the Tier 1 non-core with the reserves) remainder in Tier 2.

341 Bcfe (huge!)

EOG is one of the largest Barnett acreage holders with ~490,000 net acres. It is also one of the companies at the forefront of pushing the Barnett westward. ~3.5 Tcfe The company holds 90,000 net acres in Johnson County, with the remaining (62% of proved acreage located mainly in Jack, Palo Pinto, and Erath counties. We have also reserves) heard from industry sources that EOG is leasing in Stephens County (due west of Palo Pinto).

785 Bcfe (6% of proved reserves)

Thanks to its acquisition of Mitchell Energy, DVN is far and away the largest ~5.8 Tcfe player in the Barnett Shale, both in terms of acreage and production. Due to its (47% of proved early entrance into the play, DVN holds one of the best acreage positions, with reserves) ~550,000 net acres, the vast majority of which are either in the core area or Tier 1.

387 Bcfe DNR’s acreage is primarily in the non-core area, split roughly evenly between (50% of proved Tier 1 and Tier 2 (mainly Parker and Erath counties). The company holds 43,500 reserves) net acres in the play.

693 Bcfe CRZO holds 65,000 net acres in the play, located mainly in Parker and Tarrant (>630% of counties. About 55% of the company’s acreage position lies in the core area or proved reserves) Tier 1 non-core.

570 Bcfe CHK has grown into a large Barnett player by buying Hallwood (twice). The (12% of proved company currently holds ~48,000 acres in the Barnett, located exclusively in reserves) Johnson County.

Commentary BR has been active in the Barnett for awhile, and is one of the larger producers in the play. The company has 89,000 net acres in the Barnett, with ~30% in the core area, ~50% in Tier 1, and ~20% in Tier 2 (Palo Pinto County).

Upside Potential

The Players

Discussing the merits of a play is interesting, but useless to investors unless it is meaningful to public companies which can be owned by institutional investors. The Barnett Shale has a number of public companies with significant exposure. For company acreage maps, please refer to the appendix.

* Note: acreage splits combination of company information and Pickering Energy estimates

PICKERING ENERGY PARTNERS, INC.

13

The Data Within the Barnett, long-term recovery data are not available due to the immaturity of the play. We use peak monthly production as an indicator of ultimate recovery and “well quality”. This analytical method is well established for technical evaluation of unconventional gas reservoirs. As an example, the chart below highlights data gathered over a decade in the Carthage Field (tight gas) in East Texas. Figure 10. Wells Performance of Carthage Tight Gas Wells

10 Year Cumulative, mmcf

8000

6000

R2 = 0.70

4000

2000

0 0

2000

4000

6000

8000

Peak Monthly Production, mcf/day

Source: IHS Energy and Pickering Energy Partners

To assess the Barnett, we utilized IHS Energy production data to analyze peak monthly production for Barnett Shale wells completed in specific calendar years (i.e. well vintaging). The production data was subdivided for various vintages and different operator classes. The following bubble map shows the areal distribution of peak monthly production. The cluster of data in Wise, Denton, and Tarrant Counties is the Newark East Field. This was the initial Mitchell Energy (now Devon) development. The large dots represent high peak monthly gas production (and high expected recovery) and the small dots correspond to low peak monthly production (and low expected recovery). Even within Newark East Field, the areal variation in peak production/well quality (size of the dots) is significant.

14

P I CK E RI N G E N E R GY P A R T N E RS, INC.

Figure 11a. Areal Distribution of Peak Monthly Production 1996-2004

Source: IHS Energy and Pickering Energy Partners

PICKERING ENERGY PARTNERS, INC.

15

Figure 11. Areal Distribution of Peak Monthly Production 1996-2004 (zoom-in)

Source: IHS Energy and Pickering Energy Partners

16

P I CK E RI N G E N E R GY P A R T N E RS, INC.

Figure 12. Areal Distribution of Peak Monthly Production 2001-2004

Source: IHS Energy and Pickering Energy Partners

PICKERING ENERGY PARTNERS, INC.

17

Figure 13. Areal Distribution of Peak Monthly Production 1996-2000

Source: IHS Energy and Pickering Energy Partners

18

P I CK E RI N G E N E R GY P A R T N E RS, INC.

The figure below shows the trend in median initial well performance since 1999. Well performance peaked in 2000 and has steadily trended down since then. Possible explanations include the Newark East field continuing to mature and companies attempting to expand the field outside of the sweet spot. We expect the production trend to reverse in future years as the focus in the play continues to shift to horizontal wells from vertical wells, which have higher production rates. Figure 14. Median Initial Peak Monthly Performance (Horizontal + Vertical)

978 833

825 729

684 621

mcf/d

517

1999

2000

2001

2002

2003

2004

2005

Source: IHS Energy and Pickering Energy Partners

Horizontal vs. Vertical wells Although long-term (10+ years) production data are not available, there is sufficient production history to determine the typical decline characteristics for Barnett wells during their first few years. Below we look at the decline curves of the typical wells for both horizontal and vertical wells. We care about this because of its impact on time value of money and expected ultimate recovery (EUR), both of which influence well economics (not to mention the obvious production implications). Exhibit 15 shows the decline curves for vertical Barnett wells drilled in 1999 through 2003. As expected, the graph highlights the high initial decline rate followed by flatter decline rate in following years. In 1999 the initial decline rate was only 52% but has since averaged 65%. We think the latter is a better estimate for future forecasts as recent vintage decline rates have consistently been in the mid-60% range. Also as seen in Figure 15, decline rates typically level off around 10% in years 4-5. Sample size isn’t an issue with our vertical well analysis as the lowest number of observations is 78 in 1999 and ranges from 171 to 788 in subsequent years.

PICKERING ENERGY PARTNERS, INC.

19

Figure 15. Vertical well decline curve (core + non-core) 1200 1999 2000 2001 2002 2003

Production (mcf/d)

1000

800

600

400

200

0 Peak

t+1

t+2

t+3

t+4

t+5

Years

Source: IHS Energy and Pickering Energy Partners

Horizontal wells appear to have a shallower decline curve than their vertical counterparts. However we should note that the data is fuzzier for horizontal wells due to the lack of a significant sample size until 2003. We show both 2002 and 2003 data in the Exhibit 16 below but note that 2002 includes only 3 wells. The 2003 sample size is better with 64 wells. Focusing on the 2003 numbers, the initial decline rate of horizontal wells appears to be 50-55%. Longer-term decline rates for horizontal wells have yet to be established. Figure 16. Horizontal well decline curve (core + non-core) 3000 2002 2003 2500

Production (mcf/d)

3 Wells

2000

1500 64 Wells

1000

500

0 Peak

t+1 Years

Source: IHS Energy and Pickering Energy Partners

20

P I CK E RI N G E N E R GY P A R T N E RS, INC.

t+2

While this report focuses on well data from IHS, we realize that horizontal drilling in the Barnett is still in its infancy. Thus the calculated decline rates will likely change over time as additional wells are drilled and more production history is available. A recent example of this is KWK’s update of its “average well” type curve for its acreage in Hood county (Tier 1). The company’s old type curve assumed a shallower initial decline (~45%) than our data suggests. Its updated decline curve now forecasts a steeper initial decline (~65%) followed by a shallower decline in later years. Net-net the company’s model still forecasts the same reserves per well, but the NPV is now lower. It is interesting to note that KWK’s new type curve is similar to the vertical decline curves seen in recent years. See Figure 17 for KWK’s current decline curve assumptions compared to its prior assessment. Figure 17. KWK Decline Curve (Hood county acreage) 1600 OLD

1400

NEW

1200

mcf/d

1000 800 600 400 200

Ye a

r2

5

4

3

r2 Ye a

2

r2 Ye a

1

r2 Ye a

r2 Ye a

Ye a

r2

0

9

8

r1 Ye a

r1 Ye a

Ye a

r1

7

6

5

r1 Ye a

4

r1 Ye a

3

r1 Ye a

r1 Ye a

r1

2

1 Ye a

0 r1

r1 Ye a

r8

r7

r9

Ye a

Ye a

Ye a

r5

r4

r3

r2

r1

r6

Ye a

Ye a

Ye a

Ye a

Ye a

Ye a

Ye a

IP

0

Source: KWK and Pickering Energy Partners

PICKERING ENERGY PARTNERS, INC.

21

Well Performance Analysis by Company The bubble graph earlier highlighted the large variability of well results within the Barnett, so it should come as no surprise that a significant amount of variability exists between the results of different companies. Figure 18 graphs the well results of five of the largest producers in the Barnett, as well as the aggregate of the smaller operators. Looking first at the vertical wells drilled in 2004, Antero (now XTO) appears to have had the best results, with median production of 960mcf/d and average production of 974mcf/d. Private player Chief came in second. It is interesting that the best two performers were private companies. Each of the larger producers delivered better results than the aggregate of the smaller players (others). While we will discuss well economics in more detail in the following section, our math suggests the breakeven monthly peak production level for a vertical well in the core area is ~450mcf/d in a $6/mcf scenario. Using this as the bogey, the average well of each of the larger operators was economic in 2004 while the average of the smaller producers was not. Figure 18. 2004 Vertical Well Performance by Company – Peak Monthly Production

Median Well 960

Average Well

974

789

Mcf/day

666

618 622

670 590

584 551 418 321

Antero

Chief

ECA

Source: IHS Energy and Pickering Energy Partners

22

P I CK E RI N G E N E R GY P A R T N E RS, INC.

DVN

BR

Others

Looking at horizontal wells, Chief delivered the best results in 2004, with median production of 1.9mmcf/d and average production of 2.0mmcf/d. However unlike in the vertical well results, BR and ECA underperformed the smaller-player aggregate. That said each company’s average 2004 well would be economic in our $6/mcf gas price scenario (740mcf/d breakeven). Figure 19 shows the full results: Figure 19. 2004 Horizontal Well Performance by Company – Peak Monthly Production

Median Well

1941

Average Well

2,004 1,786 1,657

1585 1,369

Mcf/day

1300

Chief

DVN

Antero

1285

Others

1166

BR

1,087

1008 1,041

ECA

Source: IHS Energy and Pickering Energy Partners

PICKERING ENERGY PARTNERS, INC.

23

Barnett Economics – The Bottom Line Just how economic is the Barnett Shale play? Public companies often list rates of return from the play in excess of 100%...but is this reasonable? Are the economics better in some parts of the play than in others? Or is the play a slam dunk no matter where a company’s acreage lies? These are some of the questions we will try to answer. The Barnett is a highly complex play that spans a large area. Due to various factors such as depth, pay thickness and optimal type of well drilled, the economics of the play are not uniform across all areas. As a result we will look at the economics for the three types of wells we expect to be most prevalent going forward – Core area vertical, Tier 1 horizontal and Tier 2 horizontal. We caution that the Barnett is a very complex play with large variability of results within each of these three regions as well. The following analysis is our best attempt to look for general trends to help analyze the play. Some operators will be able to deliver better results while others won’t. While the following sections detail our analysis of the well economics for various Barnett wells, the following table summarizes the results, as well as shows the returns in two scenarios not detailed ($5/mcf and $7/mcf gas). In short horizontal wells are superior to vertical wells and Core/Tier 1 horizontal wells are superior to Tier 2 horizontal wells. Figure 20. Comparative Barnett Economics

Peak Monthly Prod. (Mcf/d) Year 1 Decline EUR (MMcf) Well Cost ($M) F&D Cost ($/Mcfe) Rate of Return: @ $5 @ $6 @ $7 NPV per Well: @ $5 @ $6 @ $7

Core Vertical Tier 1 Horiz. 650 1520 61% 53% 733 2356 $1,000 $2,000 $1.71 $1.06

Tier 2 Horiz. 900 53% 1395 $1,500 $1.34

12% 39% 65%

73% 113% 153%

38% 70% 101%

$0.1 $0.4 $0.7

$1.5 $2.3 $3.1

$0.6 $1.0 $1.5

Source: Pickering Energy Partners

Vertical wells The general consensus in the industry (which we agree with) is that vertical wells do not work in the non-core area. As such our analysis of vertical well economics will focus on the core area. Over the last two years, the industry has drilled over 1,100 vertical wells in the core area of the Barnett. The drilling pace did slow noticeably in 2004 relative to 2003 as industry focus shifted to drilling (theoretically) higher-return horizontal wells in the core and non-core. Before we detail the well economics model, we need to discuss the proper inputs. Three of the most important variables to consider are peak monthly production rate, decline rates (these two factors drive 24

P I CK E RI N G E N E R GY P A R T N E RS, INC.

recovery) and well cost. We were able to calculate both the peak monthly production rate and the decline rates from the IHS data set. We turned to operators (both public and private) to get a handle on well costs. Here’s what the typical vertical well looks like in the core area of the Barnett Shale: ƒ ƒ

Well cost – about $1 million to drill and complete. Peak monthly production – 650mcf/d. This is the median peak production of the total vertical wells drilled in 2003 and 2004. Of note, the median rate was ~690mcf/d in 2003 and only ~580mcf/d in 2004. ƒ Decline curve – 60% in year 1, 30% in year 2, 15% in year 3, 10% thereafter. The steepness of the declines surprised us a little bit, but the data doesn’t lie. ƒ Reserves per well – 0.7bcf gross; calculated from average decline curve, using a 30-year life. ƒ F&D cost - $1.71/mcf. (80% net revenue interest) Figure 21 below shows our well economics calculation for the median core area vertical well. We estimate that the median well (650mcf/d) will generate a 39% rate of return in a $6/mcf NYMEX price environment. The average well looks better at ~725mcf/d, which would generate a return of 53%. (Note that the model assumes core area gas receives a ~$0.50/mcf differential to NYMEX for transportation and btu content.) Figure 21. Barnett Shale Core Area Vertical Well Economics Summary Initial Production (Mcf/d) Net Reserves (MMcf) Discounted Cash Flow ($MM) Discounted Cash Flow ROR NPV ($MM) NPV/Mcfe Incremental F&D ($/Mcfe) Capital Cost ($M) Gas Price ($/Mcf) End Period Prod (Mcf/d) Decline (% of Year 1) Net Production (MMcf) Revenue ($MM) LOE ($MM) Production Tax ($MM) Overhead ($MM) DD&A ($MM) Tax ($MM) Cash Flow ($MM) Cash Flow ($/Mcf) P/T Disc. Cash Flow ($MM) Discounted Cash Flow ($MM)

650 586 1.4 39% 0.4 0.66 1.71

Year 1 Year 2 Year 3 Year 4 Year 5 Year 6 Year 7 Year 8 Year 9 Year 10 Years 11-30 1,000 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 251 181 151 134 119 105 93 83 73 65 6 95% 33% 18% 12% 12% 12% 12% 12% 12% 12% 12% 123 63 48 42 37 33 29 26 23 20 164 0.67 0.34 0.27 0.23 0.20 0.18 0.16 0.14 0.13 0.11 0.90 0.06 0.03 0.03 0.02 0.02 0.02 0.02 0.02 0.01 0.01 0.13 0.01 0.01 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.02 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.08 0.21 0.11 0.08 0.07 0.06 0.06 0.05 0.04 0.04 0.03 0.28 0.09 0.05 0.03 0.03 0.03 0.02 0.02 0.02 0.02 0.01 0.10 0.48 0.24 0.19 0.16 0.14 0.12 0.11 0.10 0.08 0.07 0.58 3.95 3.87 3.83 3.81 3.79 3.77 3.75 3.73 3.70 3.68 3.52 0.55 0.25 0.17 0.13 0.11 0.09 0.07 0.06 0.04 0.04 0.18 0.46 0.21 0.15 0.11 0.09 0.07 0.06 0.05 0.04 0.03 0.15

Source: Pickering Energy Partners

We can also see why vertical wells in the non-core area don’t really work. Johnson County should have the best vertical wells of the non-core area because the Barnett Shale is the thick and deep. The median Johnson county vertical well had peak volumes of 500mcf/d, which is only slightly economic at $6/mcf gas (10% return; 45-50% of the wells completed would not have been economic under this scenario). The average well, as expected, looks better at 585mcf/d, generating a 26% return. But here’s the rub…these numbers all assume mechanical success (i.e. a productive well). Any wells which did not flow (due to fracing into the water-bearing Ellenburger, etc.) are not included in the data set. Thus these results should be viewed as a best case scenario since it’s been well documented that penetrating the Ellenberger with a frac is a problem in the non-core area.

PICKERING ENERGY PARTNERS, INC.

25

Mean vs. Median. No it wasn’t a typo when we used median production rate in our analysis. We view the median production rate to be a better tool for analyzing what the average individual future well will deliver. This is because average production rates tend to be skewed upwards by a few very good wells, whereas median rates do not. However when looking at a total fieldwide drilling program in the Barnett, average production rates are the better measure of the aggregate success. Horizontal Wells Horizontal drilling technologies are a major part of the boom in Barnett activity that we see today, which is no surprise when we look at the far superior economics of horizontal vs. vertical wells. In fact the magnitude of the difference is so large that the only vertical wells we’d expect to see going forward are in areas that have already been densely drilled (downspacing) or near lease lines. It’s difficult to nail down what an “average” horizontal well looks like, as factors such as lateral length and completion effectiveness will have a large impact on both reserves/production as well as cost. Though our data focuses on the core area and Johnson County (where we have the best data), we will assume the results also apply to much of Parker and Hood counties, where the shale remains relatively deep and thick. We will examine Tier 2 well economics later in this section. Here’s what the typical horizontal well looks like the core and Tier 1 non-core Barnett: ƒ ƒ

Well cost – about $2 million to drill and complete. Peak monthly production – 1,520mcf/d. This is the median peak production of the total horizontal wells drilled in 2003 and 2004. Similar to the results from the vertical wells, rates in 2004 were lower than in 2003 in each county except Johnson (2.1mmcf/d vs. 0.7mmcf/d). ƒ Decline curve – 55% in year 1, 25% in year 2, 15% in year 3, 10% thereafter. Not quite as steep as in the vertical wells. ƒ Reserves per well – 2.4bcf gross; calculated from average decline curve, using a 30-year life. ƒ F&D cost - $1.06/mcf. (80% net revenue interest) Figure 22 shows our well economics calculation for the median core area or Tier 1 non-core horizontal well. We estimate that the median well (1,520mcf/d) will actually generate a >100% return in the Tier 1 non-core area and a 93% return in the core area (in a $6/mcf NYMEX price environment)…a big jump relative to vertical well economics! Much like we saw with vertical wells, the average horizontal well looks better than the median well, producing ~1,685mcf/d, which would generate returns of 113% and 135% (for core and noncore, respectively). The returns from the core area wells are lower because the gas is dry (~50c lower realized gas price), causing their production to receive a lower price than that of the non-core wells, which produce wet gas (higher btu content gets a higher price).

26

P I CK E RI N G E N E R GY P A R T N E RS, INC.

Figure 22. Barnett Shale Core/Tier 1 Horizontal Well Economics Summary Initial Production (Mcf/d) Net Reserves (MMcf) Discounted Cash Flow ($MM) Discounted Cash Flow ROR NPV ($MM) NPV/Mcfe Incremental F&D ($/Mcfe) Capital Cost ($M) Gas Price ($/Mcf) End Period Prod (Mcf/d) Decline (% of Year 1) Net Production (MMcf) Revenue ($MM) LOE ($MM) Production Tax ($MM) Overhead ($MM) DD&A ($MM) Tax ($MM) Cash Flow ($MM) Cash Flow ($/Mcf) P/T Disc. Cash Flow ($MM) Discounted Cash Flow ($MM)

1520 1885 4.3 113% 2.3 1.20 1.06

Year 1 Year 2 Year 3 Year 4 Year 5 Year 6 Year 7 Year 8 Year 9 Year 10 Years 11-30 2,000 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 718 532 458 414 375 339 307 278 251 227 31 75% 30% 15% 10% 10% 10% 10% 10% 10% 10% 10% 312 181 144 127 115 104 94 85 77 70 644 1.87 1.09 0.87 0.76 0.69 0.62 0.57 0.51 0.46 0.42 3.86 0.25 0.09 0.08 0.07 0.07 0.06 0.06 0.05 0.05 0.05 0.52 0.04 0.02 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.08 0.07 0.05 0.04 0.04 0.04 0.04 0.03 0.03 0.03 0.03 0.31 0.33 0.19 0.15 0.14 0.12 0.11 0.10 0.09 0.08 0.07 0.68 0.30 0.18 0.14 0.13 0.11 0.10 0.09 0.08 0.07 0.07 0.57 1.22 0.74 0.58 0.51 0.46 0.41 0.37 0.34 0.30 0.27 2.38 3.92 4.07 4.04 4.01 3.99 3.97 3.95 3.93 3.91 3.88 3.70 1.45 0.80 0.57 0.46 0.37 0.30 0.25 0.20 0.17 0.14 0.73 1.17 0.64 0.46 0.37 0.30 0.24 0.20 0.16 0.13 0.11 0.59

Source: Pickering Energy Partners

The above well economics calculation assumes no drainage overlap between wells. Given new data released by EOG, it appears that well spacing down to 1000 feet will still result in the economics shown above. 1000foot spacing coupled with a 4,500-foot lateral is equivalent to 100-acre spacing. Data from EOG suggests that a 2,500-foot lateral may be optimal (still resulting in a 2.4bcf well); this would equate to 60-acre spacing. However, the industry is experimenting with even tighter spacing as EOG is drilling pilots on 500-foot spacing (30-acre spacing). Initial results imply that there is overlap of the frac networks on this denser pattern resulting in some rate acceleration. Thus doubling the locations does not double the reserves. While it’s still too early to know precisely how much overlap exists on 30-acre spacing, our initial rough guesstimate is that the per-well reserves would fall ~25% to 1.8bcfe gross (the downspaced well would be 50% new reserves, 50% rate acceleration). Assuming the same well cost and decline rate characteristics, per well returns would only fall to 86% (obviously still attractive). Figure 23 shows our estimate of the well economics for a downspaced Tier 1 well. Figure 23 – Barnett Shale Core/Tier 1 Horizontal Well Economics – Downspaced Well Summary Initial Production (Mcf/d) Net Reserves (MMcf) Discounted Cash Flow ($MM) Discounted Cash Flow ROR NPV ($MM) NPV/Mcfe Incremental F&D ($/Mcfe)

1520 1413 3.7 86% 1.7 1.22 1.42

Source: Pickering Energy Partners

Running well economics for a Tier 2 well is more difficult, because we do not have enough production history from the region to calculate meaningful median/mean production rates or decline curves. Thus the best we can do is work with the limited data we have from public companies and industry sources and assume decline rates will be similar to horizontal wells in the core and Tier 1 non-core. As we mentioned earlier, Tier 2 Barnett PICKERING ENERGY PARTNERS, INC.

27

acreage has less gas-in-place (thinner and shallower) and questions persist regarding the western extent of the “gas window”. The biggest risks to Tier 2 economics are initial production and decline rate. This is what we think the typical Tier 2 horizontal well in the Barnett looks like: ƒ ƒ

Well cost – about $1.5 million to drill and complete; less than core/Tier 1 due to shallower target. Peak monthly production – 900mcf/d. The calculated rate equivalent to an EUR of 1.4bcf/well when assuming the same decline curve as seen in the core/Tier 1 horizontals. ƒ Decline curve – 55% in year 1, 25% in year 2, 15% in year 3, 10% thereafter. Data from core/Tier 1 wells. ƒ Reserves per well – 1.4bcf gross; derived from both EOG’s bottom-up estimate of reserves per well and our top-down estimate of ~75bcf gas-in-place per section. ƒ F&D cost - $1.34/mcf. (80% net revenue interest) Figure 23 shows our well economics calculation for the median Tier 2 non-core horizontal well, which are strong, but not as favorable as in the Tier 1 and core horizontals. We estimate that the average well (900mcf/d) will generate a 70% return in a $6/mcf NYMEX price environment. The median well return will likely be below this. Figure 23. Barnett Shale Tier 2 Horizontal Well Economics Summary Initial Production (Mcf/d) Net Reserves (MMcf) Discounted Cash Flow ($MM) Discounted Cash Flow ROR NPV ($MM) NPV/Mcfe Incremental F&D ($/Mcfe) Capital Cost ($M) Gas Price ($/Mcf) End Period Prod (Mcf/d) Decline (% of Year 1) Net Production (MMcf) Revenue ($MM) LOE ($MM) Production Tax ($MM) Overhead ($MM) DD&A ($MM) Tax ($MM) Cash Flow ($MM) Cash Flow ($/Mcf) P/T Disc. Cash Flow ($MM) Discounted Cash Flow ($MM)

900 1116 2.5 70% 1.0 0.93 1.34

Year 1 Year 2 Year 3 Year 4 Year 5 Year 6 Year 7 Year 8 Year 9 Year 10 Years 11-30 1,500 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 425 315 271 245 222 201 182 164 149 135 18 75% 30% 15% 10% 10% 10% 10% 10% 10% 10% 10% 185 107 85 75 68 62 56 50 46 41 381 1.11 0.64 0.51 0.45 0.41 0.37 0.33 0.30 0.27 0.25 2.29 0.18 0.06 0.05 0.04 0.04 0.04 0.03 0.03 0.03 0.03 0.31 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.00 0.05 0.04 0.03 0.03 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.18 0.25 0.14 0.11 0.10 0.09 0.08 0.08 0.07 0.06 0.06 0.51 0.15 0.10 0.08 0.07 0.06 0.06 0.05 0.04 0.04 0.04 0.31 0.71 0.44 0.35 0.31 0.28 0.25 0.22 0.20 0.18 0.16 1.44 3.84 4.14 4.11 4.08 4.06 4.04 4.02 4.00 3.98 3.95 3.77 0.82 0.47 0.34 0.27 0.22 0.18 0.15 0.12 0.10 0.08 0.43 0.68 0.39 0.28 0.22 0.18 0.15 0.12 0.10 0.08 0.07 0.36

Source: Pickering Energy Partners

We do have a small amount of data about the wells in Tier 2. EOG has announced flow rates from 2 wells and IFNY has announced flow rates from 4 wells. We will ignore the EOG results for now (tested at 0.5 and 0.9mmcf/d) as they were not drilled with full laterals. However the IFNY wells were drilled with full laterals. These wells have averaged 100, 210, 640, and 1,130mcf/d. Of these the last two wells look to be economic in our scenario (we have also heard well costs were >$2 million, higher than our assumptions). Although we expect results to improve as companies improve the completion techniques in the area, we’d be lying if we said the rates didn’t give us some concern. So far most of the wells are noticeably below our assumed average.

28

P I CK E RI N G E N E R GY P A R T N E RS, INC.

How would the steeper decline curve seen in the KWK model impact our economics? For core/Tier 1 horizontals, it would lower the return in our $6/mcf scenario to 85% from 113%. For Tier 2 wells, it would lower the average return to 47% from 70%. Clarity on the actual decline will occur over time with more production history. Peeling Back the Onion One piece of data we found interesting is that the average production rate was consistently above the median production rate. This implies that the well results are not normally distributed…in fact the data appears to be log normally distributed. A log normal distribution simplistically means that there are more below-average wells than above-average wells. However, the good wells tend to be really good, pulling up the average. As the average vertical and horizontal well looks economic in our base case scenario, it is important to analyze the downside cases. Figures 24 and 25 show the distributions of the vertical and horizontal well results for 2004. Figure 24. 2004 Vertical Well Distribution 120%

140 120

100%

# of Wells

100 80

Frequency Cumulative %

60

Average: 652mcf/d Median: 571mcf/d

80%

60%

40% 40 20%

20 0

0% 250

500

750

1,000 1,250 1,500 1,750 2,000 2,250 2,500

More

Peak Monthly Production (mcf/d) Source: IHS Energy and Pickering Energy Partners

PICKERING ENERGY PARTNERS, INC.

29

Figure 25. 2004 Horizontal Well Distribution 50

120%

45 100%

40 35 Frequency Cumulative %

# of Wells

30 25

Average: 1,638mcf/d Median: 1,402mcf/d

80%

60%

20 40%

15 10

20%

5 0

0% 500

1,000

1,500

2,000 2,500 3,000 3,500 4,000 Peak Monthly Production (mcf/d)

4,500

More

Source: IHS Energy and Pickering Energy Partners

In a $6/mcf environment, we estimate the minimum peak monthly production necessary for a horizontal well to break even to be ~650-750mcf/d (depending on richness of gas production). For a vertical well, the minimum peak monthly production would need to be ~400-450mcf/d. Looking back at the wells drilled in 2004, 23% of the horizontal wells and 32% of the vertical wells would be uneconomic in our $6/mcf gas price scenario. Figure 26 shows the breakdown of this data for Denton, Tarrant, Wise and Johnson counties.

30

P I CK E RI N G E N E R GY P A R T N E RS, INC.

Figure 26a. Well Performance Breakdown by County – Vertical Wells Vertical Wells 2003 Production (mcf/d) % Wells Uneconomic County Well Count Avg. Well Med. Well $5/mcf $6/mcf $7/mcf Denton 378 695 637 40% 24% 14% Tarrant 130 937 874 24% 15% 11% Wise 297 772 682 39% 24% 16% Johnson 15 597 620 47% 40% 20% 820 759 690 37% 23% 14% Total 2004 % Wells Uneconomic County Well Count Avg. Well Med. Well $5/mcf $6/mcf $7/mcf Denton 180 614 559 52% 34% 23% Tarrant 126 785 685 34% 21% 14% Wise 145 594 519 59% 35% 26% Johnson 24 578 428 54% 46% 38% 475 652 573 50% 32% 22% Total Vertical Breakeven Economics mcf/d Production (mcf/d) Core Johnson Benchmark Price: $5/mcf 570 500 $6/mcf 445 400 $7/mcf 365 335 Source: IHS Energy and Pickering Energy Partners

Figure 26b. Well Performance Breakdown by County – Horizontal Wells Horizontal Wells 2003 Production (mcf/d) % Wells Uneconomic County Well Count Avg. Well Med. Well $5/mcf $6/mcf $7/mcf Denton 30 1856 1625 13% 13% 10% 17 2134 1967 12% 12% 12% Tarrant Wise 17 1756 1673 29% 18% 6% Johnson 5 936 704 60% 40% 0% 69 1833 1654 20% 16% 9% Total 2004 % Wells Uneconomic Production (mcf/d) County Well Count Avg. Well Med. Well $5/mcf $6/mcf $7/mcf Denton 62 1448 1336 29% 23% 15% 47 1990 1821 15% 11% 6% Tarrant Wise 59 1208 902 53% 37% 34% Johnson 41 2139 2107 22% 17% 10% 209 1638 1474 31% 23% 17% Total Horizontal Breakeven Economics Production (mcf/d) Core Johnson Benchmark Price: $5/mcf 950 830 $6/mcf 740 670 $7/mcf 610 560 Source: IHS Energy and Pickering Energy Partners

PICKERING ENERGY PARTNERS, INC.

31

The prior tables also add color to our earlier discussions of sweet spots in the play. One thing jumps out at us immediately…production from both vertical and horizontal wells got worse in 2004 vs. 2003 in the core area. The poor core area trend likely results from some combination of two factors: the relative maturity and the industry’s continued downspacing of the Newark East field and an increased percentage of the wells drilled outside of the sweet spot. Horizontal wells in Johnson County bucked the poorer year-over-year results as the county is less developed and the industry made significant strides in completion techniques. Looking first at core area verticals, both the average well (656mcf/d) and the median well (581mcf/d) were economic. In both 2003 and 2004, Tarrant County wells were superior to those in Denton and Wise. In 2004 Wise County wells delivered the poorest performance. One possible explanation for this is that a large number of the Wise wells in 2004 appear to be field extension wells well drilled outside of the sweet spot. However this could also indicate that the sweet spot of Wise is more mature than in Tarrant and Denton… If we lower our benchmark gas price assumption to $5/mcf, vertical wells become a lot dicier. In 2004 the median well would have been economic only in Tarrant County. Roughly 50% of the vertical wells drilled in the core area would be uneconomic at $5/mcf gas. If we bump our gas price assumption up to $7/mcf,

it’s full steam ahead in the play, as both the average and median well are quite economic in every county.

As mentioned, horizontal wells look better than vertical wells. In our base case scenario, both the average and median well is economic in each of the four counties. Surprisingly, despite consistently hearing that the core area is the best acreage in the play, the best wells are the Johnson County horizontals (2,139/mcf average rate), with superior economics to both core area horizontals and verticals. The best core area horizontals are in Tarrant County. Lowering our gas price to $5/mcf doesn’t have the same impact to a horizontal drilling program, as average and median wells for the most part remain economic. The exception is Wise County, where ~55% of the 2004 horizontal wells would have been uneconomic. Similar to what we see with the vertical wells, a $7/mcf gas price means everything works. Summing it Up Two words…size matters! The primary take-away from this data is that a large acreage position is necessary. The Barnett Shale is a highly complex reservoir with economics that vary widely from well to well. The data shows that a fair amount of the wells drilled are uneconomic in a $6/mcf environment, but the good wells make the average work. Since the industry has yet to figure out how to distinguish good wells from bad wells prior to drilling, a large, contiguous acreage position is desired in order to minimize the risks in the play. While it’s true that a smaller acreage position could be entirely in a sweet spot, it could also miss it entirely… Horizontals rock. Horizontal wells have consistently generated better returns that verticals wells over the last 2 years. If gas prices fall significantly, expect vertical wells to be the first to get cut back. Overall, the play looks quite good. In a $6/mcf gas price environment, rates of return in the Barnett look quite attractive. Of course, they look far better for horizontal wells than they do for vertical wells, regardless of where they are drilled. Even in a $5/mcf gas price environment, average well returns remain attractive. Now let’s just hope drilling and completion costs don’t go up any more…

32

P I CK E RI N G E N E R GY P A R T N E RS, INC.

Johnson looks sweet. At this point in the play’s development, we’d take random acreage in Johnson County over the core counties (Tarrant, Denton, Wise) any day of the week. The Newark East field looks relatively mature (which is consistent with statements from DVN that its core area production has peaked) and core area results outside of Newark East haven’t been nearly as good. The Johnson horizontals, however, have been quite good, and it appears that another “sweet spot” trend is emerging. Tier 2 – still too early to tell…

PICKERING ENERGY PARTNERS, INC.

33

Reserve Potential There are two ways to calculate reserve potential net to the companies – top down (reserves/section) or bottom up (reserves/well). As it is still unclear what the optimal spacing will need to be or how many stacked laterals will be required to effectively develop the resource base, we choose to use the top down method to estimate reserves. In order to estimate reserve exposure per company, we need five pieces of information: net acres, gas in place per section, recovery factory, drillable acreage and royalty rate. Reserve Potential = Net acres x Gas in place per acre x Recovery factor x % of acres that can be drilled x (1 - royalty rate) Net acres. Self explanatory. How much acreage each company owns. Gas in place. The total resource potential of the Barnett Shale is estimated by the USGS to be ~200tcf. While this is certainly an impressive number, it does little to convey information about the resource potential of an individual company’s acreage position. A more useful number is gas in place per section. Studies on the play’s resource potential have indicated that the Barnett can have gas in place ranging from 50-150bcf per section, depending on the thickness of the shale. Remember that 1 section = 640 acres (1 square mile). Estimates for gas in place in the core area of the Barnett run ~140-145bcf per section. We use the midpoint of this range. As the Barnett shale remains thick in much of the Tier 1 non-core, we will also use this gas in place estimate for Tier 1. As the play moves to the south and west, the Barnett shale gets quite thin and shallows significantly. This affects reserves in place from two angles. First, there is less volume available for the gas to occupy. And second, the shallower depth means there is less pressure, so a smaller quantity of gas will be crammed into an equivalent pore space when compared to the deeper sections. As a result, the gas in place in the Tier 2 noncore area is likely to be much lower than it is in Tier 1. We assume that 75bcf per section will be the average for Tier 2 (the range is likely 50-125bcf per section). As there is little production data from this area, this estimate is quite rough. Our discussion with industry contacts and Barnett operators indicates this is a reasonable starting point given what’s known about the area. Recovery factor. The recovery factor (rf) for the core area of the Barnett Shale has been estimated at 1015%, though recent data indicates this range may be too low. EOG recently announced that it expects to drill 2.4bcf wells on 60-acre spacing in Johnson County. This implies an rf of 18%. That is, ultimate recovery will only be 18% of the original gas in place. As this is the most recent datapoint we have, we use 18% as the average rf in the core and Tier 1 non-core areas for our analysis. Any further increase in the overall recovery factor would be very significant to reserves. Some companies are currently working to further increase the rf by drilling downspaced pilot tests. Our initial take on EOG’s 30acre pilot implies an rf of 26% as the new wells are adding some incremental reserves. Admittedly, this number seems high to us, but it is what it is. As there is still limited data on how extensively downspacing will work and just how many new reserves it will add, we think it’s best to throw it in the unquantified “upside” category for the time being.

34

P I CK E RI N G E N E R GY P A R T N E RS, INC.

As there is still little information available on the Tier 2 non-core, we are not comfortable assuming the higher rf at this time. As a result, we are using an rf of 12% for the acreage until we know whether tighter spacing or shorter laterals will be successful in Tier 2. Drillable acreage. The Barnett Shale is a very complex reservoir. Geologic features such as karsts (sinkholes in layman’s terms) and faults can cause wells to be wet or unproductive. As a result, E&P companies must use 3-D seismic defensively to avoid these structures. This complexity implies that <100% of the acreage in the play will be productive. The real question is just how much less… We have talked extensively with public and private companies that operate throughout the Barnett. Their assessment of productive acreage has been ~50% based on analysis of partial 3-D seismic coverage and “experience”. More recently, a few companies have estimated that up to 85% of their acreage could be productive. While possible over small acreage positions, we think it would be rare for a sizeable acreage position to have so little karsting/faulting. We are using 50% for our analysis. Royalty rate. E&P companies don’t get acreage for free (unfortunately for shareholders). In addition to receiving an upfront lease payment from the operators, leaseholders also typically keep an overriding interest in any production from the lease. On average, we have found this to be ~20%, leaving 80% of the reserves for the E&P company. And the most reserve growth potential belongs to… Figure 27 shows the main public companies with exposure to the Barnett Shale. The table includes all of the companies’ acreage positions, regardless of whether or not we think the acreage will be productive. For example, IFNY has ~31,000 acres in Comanche County. We think this acreage is risky as it may not have been exposed to enough temperature and pressure to be in the gas window. That said, this acreage position is included under IFNY’s Tier 2 acreage. The companies with the most upside reserve exposure to the play are IFNY (huge!), CRZO (>600%), KWK (200%), and EOG (>60%).

PICKERING ENERGY PARTNERS, INC.

35

Figure 27. Barnett Shale Upside Calculation Core/Tier 1 Net acres Sections Bcf/Section Gas in place (Bcf) Recoverable gas (Bcf; @ 18%) Net reserves (Bcf, @ 80%) Tier 2 Net acres Sections Bcf/section Gas in place (Bcf) Recoverable gas (Bcf; @ 12%) Net reserves (Bcf, @ 80%) Total Net acres Net reserves (Bcf) 85% drillable acreage 50% drillable acreage Upside Reserve Potential Total Barnett potential (Bcf) Barnett proved reserves (Bcf)* Cumulative Barnett production (Bcf)* Upside reserves (Bcf) % of total proved reserves Total proved reserves (Bcfe)

BR

CHK

CRZO

DNR

DVN

ECA

EOG

IFNY

KWK

PLLL

XTO

72,000 113

48,000 75

35,750 56

20,800 33

535,000 836

38,100 60

90,000 141

0 0

67,500 105

2,300 4

116,000 181

143 16,031 2,822 2,257

143 10,688 1,881 1,505

143 7,960 1,401 1,121

143 4,631 815 652

143 119,121 20,965 16,772

143 8,483 1,493 1,194

143 20,039 3,527 2,822

143 0 0 0

143 15,029 2,645 2,116

143 512 90 72

143 25,828 4,546 3,637

17,000 27

0 0

29,250 46

22,700 35

18,000 28

88,900 139

400,000 625

60,700 95

162,500 254

0 0

39,000 61

75 1,992 239 191

75 0 0 0

75 3,428 411 329

75 2,660 319 255

75 2,109 253 203

75 10,418 1,250 1,000

75 46,875 5,625 4,500

75 7,113 854 683

75 19,043 2,285 1,828

75 0 0 0

75 4,570 548 439

89,000 2,448 2,081 1,224

48,000 1,505 1,279 752

65,000 1,450 1,232 725

43,500 907 771 454

553,000 127,000 490,000 16,975 2,195 7,322 14,429 1,865 6,223 8,487 1,097 3,661

60,700 683 580 341

230,000 3,944 3,353 1,972

2,300 72 61 36

155,000 4,075 3,464 2,038

1,224 290 35 899 7% 12,007

752 175 7 570 12% 4,902

725 32 0 693 634% 109

454 64 2 387 50% 776

341 0 0 341 3711% 9

1,972 36 0 1,936 200% 968

36 0 0 36 28% 130

2,038 550 13 1,475 25% 5,860

8,487 1,940 752 5,795 47% 12,462

1,097 300 12 785 6% 13,467

3,661 136 3 3,522 62% 5,647

Source: IHS Energy and Pickering Energy Partners *estimated

Two of these companies have made a large bet on the western edge of the play. ~80% of EOG’s acreage is located in Tier 2, primarily in Jack, Erath, and Palo Pinto counties, while all of IFNY’s acreage is located in Erath and Comanche counties. This increases the risk associated with these assets as it is still unclear whether or not the western counties are truly in the gas window. Of these four companies, KWK appears to have the lowest risk acreage position as its acreage is concentrated in the eastern portion of Hood and Somervell counties. CRZO actually has a greater percentage of its acreage in the Core/Tier 1 than KWK, but it is more fragmented while KWK’s is more contiguous. EOG and IFNY both get a qualitative ding for the high percentage of western county acreage. DNR, DVN, PLLL, XTO and CHK also have significant upside reserve potential from the Barnett (~50%, 50%, 30%, 25%, and 10% respectively). PLLL, however, has a relatively small acreage position (which is a concern given prior discussion). BR and ECA each have <10% overall reserve upside from the play. What’s an Mcf worth anyway? Looking back at our well economics calculations indicates that the typical core/Tier 1 horizontal well is worth about $2.3 million or ~$1.20/mcf. A Tier 2 horizontal well is worth $1.0 million or ~$0.95/mcf. A vertical well is only worth $0.4 million or ~$0.65/mcf. Each calculation assumes $6/mcf gas and that the well is drilled today. The tables below show the value of a Barnett well under different pricing and production scenarios.

36

P I CK E RI N G E N E R GY P A R T N E RS, INC.

Gas Price ($/Mcf)

Exhibit 28. Core Vertical Well Value ($MM) Peak Monthly Production (Mcf/d) 0.39 500 650 800 ($0.5) ($0.4) ($0.3) $3.00 ($0.3) ($0.1) $0.0 $4.00 ($0.1) $0.1 $0.3 $5.00 $0.1 $0.4 $0.7 $6.00 $0.3 $0.7 $1.0 $7.00 $0.5 $0.9 $1.3 $8.00 $0.7 $1.2 $1.7 $9.00

950 ($0.2) $0.2 $0.6 $1.0 $1.3 $1.7 $2.1

Source: IHS Energy and Pickering Energy Partners

Average well = 650 mcf/d; $0.4MM NPV @ $6/mcf

Gas Price ($/Mcf)

Exhibit 29. Tier 1 Horizontal Well Value ($MM) Peak Monthly Production (Mcf/d) 2.26 1000 1500 2000 ($0.7) ($0.1) $0.4 $3.00 ($0.2) $0.6 $1.5 $4.00 $0.4 $1.4 $2.5 $5.00 $0.9 $2.2 $3.5 $6.00 $1.4 $3.0 $4.6 $7.00 $1.9 $3.8 $5.6 $8.00 $2.4 $4.6 $6.7 $9.00

2500 $1.0 $2.3 $3.6 $4.9 $6.2 $7.5 $8.8

Source: IHS Energy and Pickering Energy Partners

Average well = 1520 mcf/d; $2.3MM NPV @ $6/mcf

Gas Price ($/Mcf)

Exhibit 30. Tier 2 Horizontal Well Value ($MM) 1.04 $3.00 $4.00 $5.00 $6.00 $7.00 $8.00 $9.00

Peak Monthly Production (Mcf/d) 700 900 1100 ($0.6) ($0.4) ($0.1) ($0.2) $0.1 $0.4 $0.1 $0.6 $1.0 $0.5 $1.0 $1.6 $0.9 $1.5 $2.1 $1.2 $2.0 $2.7 $1.6 $2.5 $3.3

1300 $0.1 $0.7 $1.4 $2.1 $2.8 $3.5 $4.1

Source: IHS Energy and Pickering Energy Partners

Average well = 900 mcf/d; $1.0MM NPV @ $6/mcf However, even with these figures, it is not a simple extrapolation process to determine the value to an individual E&P company’s acreage position. The primary reason for this is time value of money. The NPV of the acreage position is highly sensitive to the number/location of wells drilled each year, so the analysis is clouded by assumptions of future rig counts, location of the rigs, type of well drilled (horizontal/vertical), etc. PICKERING ENERGY PARTNERS, INC.

37

Let’s take KWK for instance. The company has 230,000 acres in the Barnett play. Assuming 50% drillable acreage and just 100-acre spacing, KWK has ~1,285 locations in its inventory. With the two rigs it’s currently using, it would take >40 years to drill its entire acreage position (15 wells per rig-year). Even if the company immediately accelerates to its targeted 6 rigs, it would still take KWK almost 15 years to fully drill its acreage. Clearly, the NPV of the company’s Barnett assets would be different in a 40-year versus a 15-year drilling scenario.

38

P I CK E RI N G E N E R GY P A R T N E RS, INC.

$31 $11 $42 -$33 -43%

$0.57 $0.17 $0.10 $0.02 -$0.04 -$0.12

$9.38 $1.96 $1.23 $1.02 -$0.64 -$2.12

$2.39 $0.64 $0.41 $0.12 -$0.14 -$0.56

$3.44 $0.74 $0.44 $0.19 -$0.29 -$0.75

$0.30 $0.08 $0.05 $0.02 -$0.02 -$0.07

$3.99 $0.67 $0.46 $0.19 -$0.33 -$0.98

$1

$31 $14 $45 -$30 -40%

Barnett Value Sensitivites (from Base Case): $1/Mcf change in gas price $0.81 5% increase in drillable acreage $0.23 1% increase in recovery factor $0.14 1 additional rig $0.05 1% increase in initial decline -$0.05 10% increase in drilling costs -$0.18

$44 $14 $57 -$11 -17%

Upside Case 2 ($6/mcf gas, 50% drillable acreage, 30-acre spacing): Proved Reserve NAV $53 $26 Barnett Potential $4 $3 $39 $11 NAV w/ Barnett Potential $57 $37 Delta vs. Current Price -$25 -$14 % of current stock price -30% -27%

$1

$31 $10 $41 -$33 -45%

EOG $74.90

$37 $28 $65 -$10 -13%

$44 $13 $57 -$12 -17%

Upside Case 1 ($6/mcf gas, 85% drillable acreage, 60-acre spacing): Proved Reserve NAV $53 $26 Barnett Potential $4 $3 $37 $10 NAV w/ Barnett Potential $56 $36 Delta vs. Current Price -$25 -$14 % of current stock price -31% -28%

$1

ECA $58.31

Upside Case 3 ($7/mcf gas, 85% drillable acreage, 30-acre spacing, 50% higher rig count): Proved Reserve NAV $60 $27 $49 Barnett Potential $8 $7 $83 $24 $28 $3 NAV w/ Barnett Potential $68 $51 $77 Delta vs. Current Price -$13 $1 $8 % of current stock price -16% 2% 12%

$44 $10 $53 -$15 -22%

CRZO $29.27

Base Case ($6/mcf gas, 50% drillable acreage, 60-acre spacing): Proved Reserve NAV $53 $26 Barnett Potential $2 $2 $26 $6 NAV w/ Barnett Potential $55 $32 Delta vs. Current Price -$26 -$18 % of current stock price -32% -36%

CHK $38.25

DVN $68.64

BR $81.32

DNR $50.44

Stock Price

$7.83 $1.22 $1.00 $1.18 -$0.67 -$2.14

$41

$17

$24

$17

IFNY $8.15

$5.79 $0.73 $0.48 $1.08 -$0.59 -$1.37

$18 $40 $58 $11 22%

$15 $19 $33 -$15 -31%

$15 $19 $33 -$15 -30%

$15 $15 $30 -$18 -38%

KWK $47.79

$0.38 $0.10 $0.05 $0.03 $0.00 -$0.06

$4

$2

$2

$1

PLLL $13.99

$1.53 $0.51 $0.30 $0.01 -$0.09 -$0.34

$33 $18 $52 $6 14%

$27 $8 $35 -$10 -22%

$27 $7 $35 -$11 -23%

$27 $4 $32 -$14 -30%

XTO $45.32

Company-specific Barnett upside Having highlighted the challenges of the analysis, let’s look at the results. Exhibit 31. Value of Barnett Upside Potential

Source: Pickering Energy Partners

PICKERING ENERGY PARTNERS, INC.

39

The previous table shows the NAV impact of the Barnett in four scenarios. We view our base case as the most likely, but provide three more aggressive scenarios for perspective (and to account for companies like CHK and CRZO arguing that 85% drillable is more likely on their acreage). The delta relative to the current stock price represents the per share value that must come from future growth outside the Barnett Shale to justify the current stock price. For instance, DNR has a large delta in our base case, but it is straightforward to identify additional value in the company’s tertiary recovery potential. We should highlight two of our assumptions. We use a 25% cash tax rate in the analysis. Some companies (such as KWK) have corporate tax shields which will allow them to pay a lower percentage of cash taxes over the next few years. For number of wells drilled per year, we use companies’ expected rig count over the next 12 months and assume each rig can drill 15 Barnett wells per year. The four scenarios we provide obviously don’t cover all the possible or even likely scenarios. For example, it is highly probable that many of the Barnett operators will continue to expand their rig count in subsequent years. As a result, we have included sensitivities to many of the key variables used in the analysis. The sensitivities are calculated from the upside Barnett potential in our base case scenario. Our primary takeaway from this analysis is that Barnett upside is already priced into most of these stocks (with some priced to perfection). An investor has to use aggressive assumptions about the play (not to mention higher commodity prices) to justify buying most of these names on Barnett upside alone. The stock which looks the most interesting as a Barnett play at current stock prices is CRZO (not on our official coverage list). CRZO’s upside Barnett value in our base case is $26/share, not far below its current stock price. Make some more aggressive assumptions about drillable acreage, etc, and you might not need to know much about its other assets. While it appears that IFNY has significant upside potential from the Barnett as well, we caution that it has the riskiest acreage position among the companies listed (100% Tier 2, ~50% in Comanche County). For the purpose of this analysis, we assigned value to all of a company’s acreage position, even if it we viewed it as low probability. We doubt the Barnett will be productive in Comanche County (an opinion we view to be validated by IFNY’s announcement that it will develop this acreage with vertical wells to test the Marble Falls as well), and thus a more prudent way to look at IFNY might be to cut the numbers in the prior table in half. Its remaining acreage position is not without risk either, being concentrated in Erath County. IFNY has a lot of upside potential if its acreage works but at this point that’s a big if. The stock whose valuation concerns us the most is EOG (we exclude BR as it is not really a Barnett story). Even in our most aggressive scenario, the company would still need a significant amount of NAV growth outside of the Barnett to warrant its current stock price. At current levels, EOG’s Barnett position appears fully valued and, perhaps, even overvalued. EOG – a closer look That EOG was still ~$10/share shy of its current price even in our aggressive upside case has made us take a closer look at its valuation. The following table breaks down a fair value for EOG given our Barnett calculations, as well as the assumptions that would be required to justify the current stock price. (Note – the analysis assumes EOG operates 9 rigs in 2005 and 16 rigs thereafter. Also, each rig is assumed to drill 15 wells per year.)

40

P I CK E RI N G E N E R GY P A R T N E RS, INC.

Exhibit 32. EOG Value Calculation EOG ($74.90) Proved Reserve NAV @ $6 and $45/bbl Appropriate NAV Multiple Stock Value of Proved Reserves Barnett Upside Value @ Base Case Base Case EOG Stock Value

$31 1.1x $35 $10 $45

Required Assumptions to justify current stock price: Proved Reserve NAV @ $7/mcf and $50/bbl $38 Appropriate NAV Multiple 1.1x Stock Value of Proved Reserves $42 Barnett Upside Value @ $7/mcf gas $14 30-acre Spacing in Johnson (26% recovery) +$1 85% Barnett acreage drillable +$4 150% higher rigcount than base case +$11 18% Recovery factor for Tier 2 +$3 Stock Value $75 Source: Pickering Energy Partners

As EOG is a well-run company, we believe 1.1x is a fair multiple to place on its proved reserve NAV to account for its going concern value. This multiple generates a per share value for EOG of $35 excluding Barnett upside. Adding this to our base case $10 per share value for EOG’s unbooked Barnett potential yields $45 per share, ~40% below the current stock price. In order to justify the current stock price, we have to make a number of aggressive assumptions. First, we need to increase our benchmark gas price assumption to $7/mcf (in line with what peers are currently discounting). We also have to assume that 30-acre spacing works on all of its Johnson acreage, that 85% of the company’s acreage is productive, that EOG more than doubles its rig count from our forecast 2006 level (to 40 rigs from 16), and that the recovery factor in the Tier 2 non-core can be increased to 18%. We should note that EOG’s well cost assumptions are at the low end of the industry range, and below the average well cost we are using in our model. EOG is assuming a well cost of $1.6 million for its Johnson County acreage (Tier 1) and $1.1 million for its Tier 2 acreage (vs. our assumptions of $2.0 million and $1.5 million). Using the EOG well cost assumptions would increase the NPV of a Tier 1 well to $2.6 million from $2.3 million and the NPV of a Tier 2 well to $1.4 million from $1.0 million. The impact of this cost structure on the above analysis would be to increase the Barnett upside value in our base case to $12 per share from $10, resulting in a base case stock value of $47. In the second case, the lower cost structure would result in a stock value of $81, or $6 higher than with the Pickering Energy cost assumptions. With EOG currently trading at $75, it certainly feels to us that even the most aggressive EOG assumptions are already priced into the stock. Could we be wrong? Of course, as our analysis is based on imperfect data and a number of assumptions. If investors are willing to make more aggressive assumptions than those in the table above (or assume a higher commodity price deck), buying EOG at current levels still makes sense. However, even if one was willing to make those assumptions, we still think it wiser to buy one of the cheaper Barnett players which could offer significantly higher upside potential in such a scenario.

PICKERING ENERGY PARTNERS, INC.

41

The Unknowns Although the industry has been developing the Barnett shale for quite some time now, a number of question still remain, especially in the non-core area. We will not try to discuss all of these, but will highlight five issues that will have an impact on the play economics and reserve potential: drillable acreage, recovery factor, oil/gas window, decline rates, and retrograde reservoir. The first two topics were discussed earlier. As the industry shoots more seismic over the area and gets more well control, it will be able to refine its rough estimate of drillable acreage. For now, that’s just what it is…a rough estimate. As the earlier table shows, if drillable acreage is indeed closer to 85%, we are materially underestimating the upside reserve potential of the play (and Barnett valuations look more palatable). Recovery factory could also continue to improve over time. New data from E&P companies implies the recovery factor may be closer to 18% (well up from the 12% estimate we were using when we started writing this report). Further increases are possible due to downspacing, improvements in technology, etc. That said, it won’t be easy. Recovery improvements could boost both upside reserve potential and drilling returns (if the per well recovery is improved; not the case with downspacing) beyond what our analysis suggests. Oil/gas window. A key question that has yet to be fully answered is where the gas window ends. Conventional wisdom in the industry indicated that the gas window did not extend into the western counties (Jack, Erath, Palo Pinto). Currently, companies such as EOG and IFNY are trying to prove the viability of these counties. Why do we care about oil vs. gas? Simple… Given the larger size of its molecules, oil will not flow very well (if at all) in the ultra-tight Barnett. This means that any gas trapped behind the liquids will not flow either. Currently, adsorbed gas is estimated to be ~25% of the gas in place, but we caution that this is still a very rough estimate. Wells fully in the oil window are easy to identify, as early production will be mostly oil. Where it gets trickier is in the “transition zone”, where oil and gas could both be present in the reservoir. Early production from a well in the transition zone will be mostly gas, as the oil molecules will not flow as easily. This does not necessarily mean that the well is in the gas window. The well may initially look like a gas well, only to have future production decline more rapidly than expected (resulting in lower than anticipated reserves) as the free gas is exhausted. Basically, assuming the above free gas/adsorbed gas ratio is correct, the well would yield 25% lower than expected reserves and lower returns (breakeven production level would jump to 735mcf/d from 670mcf/d in $6/mcf gas scenario). The returns figures also assume that all the lost production comes at the end (ie. production curve looks the same, then drops to zero after free gas is produced), which is probably not a conservative enough assumption. Possibility of retrograde condensate reservoir. Is the rich gas stream in the non-core Barnett a blessing or a curse? Companies currently benefit from the increased heat content created by the liquids in the gas stream, but this does pose a potential problem down the road. Some industry experts postulate that portions of the Barnett “transition zone” are a retrograde condensate reservoir. This means that liquids drop out of the gas stream while still in the reservoir, rather than once they reach the surface. This could pose a problem similar to the oil/gas window as it could reduce eventual recovery. It’s too early to know one way or the other, but it is something that investors should keep an eye on.

42

P I CK E RI N G E N E R GY P A R T N E RS, INC.

Decline rates. We touched on this earlier in the report with the KWK example. The decline rates that will be seen in the non-core area of the play are not precisely known at this time. We have estimates at this point, but these will most likely change as the industry gets more production data. While a changing shape of the decline curve might not impact the reserves in the play (as in the KWK example), it may have a significant impact on economics. Only time will tell if current expectations of the “typical” Barnett decline curve prove to be too conservative or too aggressive.

PICKERING ENERGY PARTNERS, INC.

43

Summary/Conclusions ƒ ƒ ƒ ƒ ƒ ƒ ƒ

ƒ

ƒ

44

Not all acreage is created equally. Significant variability of results exists even within concentrated areas. Early mover advantage. On average, the companies that were early entrants into the play got the best acreage and have the best results. Size matters. For this type of a statistical play (highly variable) to work, a large enough acreage position is required for the law of large numbers to take effect. Not all acreage is created equally (part 2). The Barnett Shale does appear to have sweet spots that will be far more productive than the average acreage position. This “resource play” takes a continuous technical feedback loop to maintain optimal well performance. Operators shouldn’t let the drilling program “get ahead” of well performance. Not all wells are economic. However the statistical nature of the Barnett suggests that, in aggregate, good well more than offset bad wells. Unknowns remain, especially in the non-core area. The industry still does not have definitive proof of where the gas window ends, nor what the decline rates or recovery factor will be. Each of these items will meaningfully impact the play’s ultimate potential. On average, Barnett upside potential appears to be fully reflected in stock prices. More aggressive assumptions (higher gas prices, more drillable acreage, higher recovery factor, etc) are needed to find significant upside from current levels. Barnett stock picks – CRZO (not officially covered) looks interesting while EOG looks riskiest.

P I CK E RI N G E N E R GY P A R T N E RS, INC.

Appendix A – Gas Shale Terminology Gas Shales – Natural gas is generated and stored in a shale in two forms 1) free gas which occupies the pore space (similar to conventional gas reservoirs) and 2) adsorbed gas which is stored on the organic matter. Gas shales are organic-rich, fine-grained sedimentary rocks containing a minimum Total Organic Carbon (TOC) of 0.5% with varying degrees of thermal maturity as measured by vitrinite reflectance RO. TOC in the Barnett Shale is ~4.5%. Generally, RO values <1.0% have a tendency to be oil productive and >1.0% are thought to be gas productive. Shales may have varying degrees of thermal maturity ranging from 0.4-0.6% RO (marginal) to 0.6 to 2.0% RO (mature). Barnett Shale RO values vary widely from >1.5% in the core area to <0.9% in the western area. Because the rock matrix is composed of fine-grained rocks, reservoir permeability (flow capacity) is often low and commercial production of gas from organic shales often requires the presence of natural fractures and/or completions with large hydraulic fracture treatments. Due to the low matrix permeability, recovery factors from organic shales can be very low…sometimes <10% of the original gas in place. Gas shales are often the source rock for oil and gas in adjacent strata in many basins in the US…including the Fort Worth Basin. Definitions: ƒ

ƒ ƒ

ƒ

TOC – Total Organic Carbon is measured in the lab using a technique called pyrolysis, thus core samples of the shales must be available. TOC is the amount of material available to convert into hydrocarbons. A high percentage of TOC implies significant material (generally) available for hydrocarbon generation. Adsorption – The property of some solids to attract a liquid or a gas to their surfaces. In shales, some of the natural gas is adsorbed onto the surface of the shale. Thermal Maturity – Measured in the lab (core samples required) using a technique called vitrinite reflectance. The measurement is given in units of reflectance, % RO, with typical values ranging from 0% RO to 2% RO. It is a measurement of the maturity of organic matter with respect to whether it has generated hydrocarbons or could be an effective source rock. Permeability – Measurement of a rock's ability, to transmit fluids, typically measured in millidarcies. Formations that transmit fluids readily are described as highly permeable and tend to have many large, wellconnected pores. Low permeability formations, such as shales and siltstones, tend to have smaller or less interconnected pores. For comparison, a typical Gulf of Mexico formation may have a permeability of 1,000 md, a tight formation in East Texas has permeability of 0.01 md, and the Barnett Shale <0.0001 md. The flow capacity of a typical GOM reservoir is ten million times higher than an organic shale.

PICKERING ENERGY PARTNERS, INC.

45

Appendix B – Comparison of Organic Shales in the US Comparison of Organic Shale Properties Depth, ft Gross Thickness, ft Net Thickness, ft Bottomhole Temp, ºF TOC, % Total Porosity, % Gas Filled Porosity, % Water Filled Porosity, % Flow Capacity - kh, md-ft Gas Content, scf/ton Adsorbed Gas, % Reservoir Pressure, psi Pressure Gradient, psi/ft Water Production, Bwpd Well Spacing, Acres Recovery Factors, % Gas-In-Place, Bcf/Section Reserves, MMcf

Barnett 6,500-8,500 150-700 100-600 200 4.5 4-5 2.5 1.9 0.01-2 300-350 25 3,000-4,000 0.43 0 60-160 10-20 50-150 500-4,000

Ohio 2,000-5,000 300-1,000 30-100 100 0.0-4.7 4.7 2 2.5-3.0 0.15-50 60-100 50 500-2,000 0.15-0.40 0 40-160 10-20 5-10 150-600

Antrim 600-2,200 160 70-120 75 1-20 9 4 4 1-5,000 40-100 70 400 0.35 5-500 40-160 20-60 6-15 200-1,200

New Albany 500-2,000 180 50-100 80-105 1-25 10-14 5 4-8 40-80 40-60 300-600 0.43 5-500 80 10-20 7-10 150-600

Lewis 3,000-6,000 500-1,900 200-300 130-170 0.45-2.5 3.0-5.5 1-3.5 1-2 6-400 15-45 60-85 1,000-1,500 0.20-0.25 0 80-320 5-15 8-50 600-2,000

Fayetteville 1,500-6,500 50-325 20-200

Source: GTI and Pickering Energy Partners

2005 Estimated Gas Production from US Shales

Appalachia/Ohio: 438 21%

SanJuan: 55 3%

Antrim: 384 18%

Units: mmcf/day

Source: GTI and Pickering Energy Partners

46

P I CK E RI N G E N E R GY P A R T N E RS, INC.

Barnett: 1,233 58%

4.0-9.5 2-8

60-220 50-70 600-2,000

25-60

Appendix C – Example of Barnett NPV Model Below we show an example of our Barnett Shale NPV model. The model assumes that the companies focus their drilling efforts on Core/Tier 1 acreage (higher value) until they run out of locations, then move the rigs to their Tier 2 acreage. The discount rate used in the analysis is 10%. EOG Barnett Shale NPV model Total Core/Tier 1

Tier 2

Total

# Rigs Wells/rig-yr # Wells Value/well ($MM) Value ($MM) NPV ($MM) # Rigs Wells/rig-yr # Wells Value/well ($MM) Value ($MM) NPV ($MM) Value FD Shares $/share

675

1326

2016

1125

Year 1 8 15 120 2.3 272 272

Year 2 14 15 210 2.3 475 432

Year 3 14 15 210 2.3 475 393

1 15 15 1.0 16 16

2 15 30 1.0 31 28

2 15 30 1.0 31 26

Year 4 9 15 135 2.3 306 230

Year 5 0 15 0 2.3 0 0

Year 6 0 15 0 2.3 0 0

Year 7 0 15 0 2.3 0 0

Year 8 0 15 0 2.3 0 0

Year 9 0 15 0 2.3 0 0

Year 10 0 15 0 2.3 0 0

Year 11 0 15 0 2.3 0 0

Year 12 0 15 0 2.3 0 0

7 15 105 1.0 110 82

16 15 240 1.0 250 171

16 15 240 1.0 250 155

16 15 240 1.0 250 141

16 15 240 1.0 250 128

16 15 240 1.0 250 117

16 15 240 1.0 250 106

16 15 240 1.0 250 97

16 15 156 1.0 163 57

2451 243.4 $10.07

Source: Pickering Energy Partners

PICKERING ENERGY PARTNERS, INC.

47

Appendix D – Generalized Company Acreage Maps

48

BR

CRZO

CHK

DNR

P I CK E RI N G E N E R GY P A R T N E RS, INC.

DVN

EOG

ECA

IFNY

*Note – ECA has acreage in Montague and Cooke counties (North of Wise and Denton; indicated by upper shading)

PICKERING ENERGY PARTNERS, INC.

49

KWK

PLLL

50

P I CK E RI N G E N E R GY P A R T N E RS, INC.

XTO

Analyst Certification: I, Jeff Hayden and Dave Pursell, do hereby certify that, to the best of my knowledge, the views and opinions in this research report accurately reflect my personal views about the company and its securities. I have not nor will not receive direct or indirect compensation in return for expressing specific recommendations or viewpoints in this report.

______________________________________________________________________________________ Important Disclosures: The analysts involved in creating or supervising the content of this message (or members of their households) do not own the securities mentioned. This communication is based on information which Pickering Energy Partners, Inc. believes is reliable. However, Pickering Energy Partners, Inc. does not represent or warrant its accuracy. The viewpoints and opinions expressed in this communication represent the views of Pickering Energy Partners, Inc. as of the date of this report. For detailed rating information, please visit our disclosure website at www.pickeringenergy.com/disclosure.asp. These viewpoints and opinions may be subject to change without notice and Pickering Energy Partners, Inc. will not be responsible for any consequences associated with reliance on any statement or opinion contained in this communication. This communication is confidential and may not be reproduced in whole or in part without prior written permission from Pickering Energy Partners, Inc. ______________________________________________________________________________________ Ratings: B = buy, A = accumulate, H = hold, T = trim, S = sell, NR = not rated

PICKERING ENERGY PARTNERS, INC.

51

www.pickeringenergy.com Institutional Research Houston, Texas Research Department 713 333-2960 Institutional Sales Houston C. Paige DiMaggio [email protected] 713-333-2969

Denver Chuck Howell [email protected] 303-300-1902

Lori Yentzen [email protected] 713-333-2974

Jon Mellberg [email protected] 303-300-1995 Jason Foxen [email protected] 303-300-1960 Trading Desk 800-507-2400

Michael du Vigneaud [email protected] 713-333-2976 Matthew Pham [email protected] 713-333-2977 Josh Martin [email protected] 713-333-2982 Email: [email protected] Clearing through Bear, Stearns & Co. Inc. 713.333.2960 Phone ⎟ 713.333.2965 Fax⎟ 1800 West Loop South ⎟ Suite 300 ⎟ Houston, Texas 77027

303.300.1900 Phone ⎟ 303.300.1901 Fax⎟ 2000 S. Colorado ⎟ Suite 3400 ⎟ Denver, Colorado 80222

The Barnett Shale

Uncertainties remain for Tier 2 – Not enough data exists to conclusively ...... Within the Barnett, long-term recovery data are not available due to the ...... 713.333.2965 Fax│ 1800 West Loop South │ Suite 300 │ Houston, Texas 77027.

4MB Sizes 1 Downloads 230 Views

Recommend Documents

Marcellus Shale Gas Drilling - Delaware Riverkeeper Network
Mar 22, 2010 - positive economic impact that such drilling could have for New York State and its counties. However .... top ten gas counties are New York State‟s most “energy focused” counties, it is ... Source: American Community Survey ... in

pdf-12118\the-countess-angelique-book-6-by-sergeanne-barnett ...
Whoops! There was a problem loading more pages. pdf-12118\the-countess-angelique-book-6-by-sergeanne-barnett-marguerite-translator-golon.pdf.

New Information on Barnett Newman's Exhibition History, 1948-52
was a friend of Newman's wife, Annalee; the price was $350, the lowest of any ..... bubbles' attached to it; mobile phones are used freely in the galleries; noise ...

man-167\oil-shale-stocks.pdf
Download. Connect more apps... Try one of the apps below to open or edit this item. man-167\oil-shale-stocks.pdf. man-167\oil-shale-stocks.pdf. Open. Extract.

man-167\oil-shale-reserves.pdf
Sign in. Loading… Whoops! There was a problem loading more pages. Retrying... Whoops! There was a problem previewing this document. Retrying.

Marcellus Shale Gas Drilling - Delaware Riverkeeper Network
Mar 22, 2010 - these employment data are available from County Business Patterns) ..... geologically similar to the Marcellus Shale and the same technology, ...

New Information on Barnett Newman's Exhibition History, 1948-52
expected sales, Newman owed the gallery just over fifteen dollars.4. The paintings were .... smaller works on display in the back office. Connections between the ...

pdf-0716\college-algebra-graphs-and-models-by-raymond-barnett ...
... the apps below to open or edit this item. pdf-0716\college-algebra-graphs-and-models-by-raymond-barnett-michael-ziegler-karl-byleen-david-sobecki.pdf.

Ohio Utica Shale Region Monitor - Maxine Goodman Levin College of ...
Mar 1, 2013 - Analysis of the 2012 employment data shows that employment growth in the ... The eleven Metropolitan Statistical Areas (MSAs) in Ohio were ...

Ohio Utica Shale Region Monitor - Maxine Goodman Levin College of ...
Mar 1, 2013 - Ohio Utica Shale Region Monitor: March 2013 .... Ohio Department of Taxation, http://www.tax.ohio.gov/government/distributions_sales_.aspx.

pdf-12119\albert-barnett-klepper-and-edward-schnur-petitioners-v ...
... the apps below to open or edit this item. pdf-12119\albert-barnett-klepper-and-edward-schnur-p ... ce-company-tanner-us-supreme-court-transcript-of.pdf.

Possibilities of oil shale mining under the Selisoo mire of the Estonia ...
Estonian Academy Publishers, Tallinn, pp 389–390. 13. Koitmets K, Reinsalu E, Valgma I (2003) Precision of oil shale energy rating and oil. shale resources. Oil Shale 20(1):15–24. Page 3 of 14. Main menu. Displaying Possibilities of oil shale min