DAVID MICHAEL COHEN, Managing Editor
 |
| Cattle graze on a meadow overlooking a downgrade where the Trinidad 123 rig drills a Haynesville shale well for Encana in October 2009. Photo by Dale Shank, courtesy of Newpark Drilling Fluids. |
|
There is no question that sustained low prices for natural gas have taken their toll on the Haynesville shale play of northwestern Louisiana and East Texas. The number of rigs working in the Haynesville has fallen steadily during the past 15 months, and there are reasons to be concerned that activity could continue falling off. Producers have largely completed the drilling needed to hold onto their expensive leases in the play, and two attractive, liquids-rich shale targets elsewhere in Louisiana have been attracting recent attention: the Tuscaloosa marine shale in the middle of the state and the Brown Dense or Lower Smackover shale to the north. Judging from the rapid pace of development in South Texas’ liquids-rich Eagle Ford shale, these two new prospects could soon be seriously competing for rigs with the dry-gas Haynesville.
However, Haynesville drilling in East Texas has stayed relatively stable during the last year in the face of the Eagle Ford boom, unlike its North Texas neighbor, the Barnett. Despite its extreme depth and high-pressure, high-temperature (HPHT) operating conditions, resulting in well costs often exceeding $9 million, the Haynesville’s very high initial production (IP) rates and estimated ultimate recoveries (EURs) have kept some of the biggest players very active there. Furthermore, many operators are testing the highly prospective overlying Bossier shale, which could add production to keep Haynesville wells profitable even if gas prices continue to stagnate around $4/Mcf. Another good sign relates to Australian mining giant BHP Billiton’s purchase this summer of heavily Haynesville-invested independent Petrohawk for $15.1 billion in cash. In September, Petrohawk submitted drilling applications for 18 wells in Swan Lake field overlapping Caddo and Bossier Parishes and five wells in Caspiana field of Bossier Parish, in the gas-factory configurations that have become customary in the play.
After less than three years of development, Haynesville production surpassed that of the Barnett in March, making it the highest-producing US shale play at slightly more than 5 Bcfd, according to an analysis by the US Energy Information Administration (EIA) based on data from market analysts Bentek Energy. Compare that performance with more than nine years that it took the Barnett to reach the 5-Bcfd mark. In the long term, indications are for Haynesville production to soar even higher, with a number of companies investigating projects to make use of the gas.
In May, Cheniere Energy won US Department of Energy approval to export liquefied natural gas at its Sabine Pass LNG terminal in Louisiana, which could liquefy up to 2.2 Bcfd of gas. Terminals in Lake Charles, Louisiana, and Freeport, Texas, are seeking similar approval. Additionally, Encana is developing LNG and compressed natural gas infrastructure in Louisiana that could be the vanguard of a new market for Haynesville gas. The company opened its first CNG fueling station in Red River Parish last November, and in April it struck a deal to supply fuel for 200 new LNG trucks belonging to Heckmann Water Resources, a California company that provides water-hauling services to Haynesville producers. Along similar lines, Chesapeake has agreed to invest $150 million toward the construction of 250–300 LNG truck fueling stations as part of a $1 billion plan to increase long-term US gas demand.
Even if persistent low gas prices compared with oil complicate the economics of CNG or LNG, increased Haynesville gas production could find a market through conversion to pricier liquid fuels. In September, South African energy and chemicals group Sasol announced an 18-month feasibility study for an $8–10 billion gas-to-liquids plant to be built near Lake Charles, Louisiana. The facility would be the first of its kind in the US and would produce either 2 million tonnes per annum (mtpa) or 4 mtpa of diesel fuel and related products from up to 840 MMcfd of feed gas. Construction could start as early as 2013, with completion by 2018.
GEOLOGY AND RESOURCES
Underlain by the Smackover limestone formation and overlain by sandstones of the Cotton Valley group, the Upper Jurassic-aged Haynesville shale takes up an area of about 9,000 sq mi primarily in Louisiana and stretching into East Texas and southern Arkansas. With net thickness ranging 200–300 ft, the Haynesville is encountered at depths of 10,500–13,500 ft (top) and is being developed with horizontal wells that typically have laterals of 4,000–5,000 ft, for a measured depth range of about 16,000–19,000 ft. Completions typically have 10–15 fracture stages per lateral, with slickwater treatments that contain 300,000–400,000 lbm of proppant pumped at up to 80 bbl/min.
Being the deepest unconventional play in development, the Haynesville is subject to HPHT conditions not generally seen in other shales. According to David Breeden, lab manager at Newpark Drilling Fluids, bottomhole temperatures are generally 350°F and higher, with the coolest sections in the shallower Texas part of the play being about 325°F. In Red River Parish, Louisiana, BHTs up to 400°F have been encountered. Bottomhole pressures exceeding 12,000 psi have been reported, and treatment pressures approach 15,000 psi. According to a paper presented at last year’s SPE Annual Technical Conference and Exhibition by authors from Hexion Specialty Chemicals (now Momentive), closure stresses in the Haynesville range 9,000–12,000 psi, compared with 6,000 psi or less in the Fayetteville and up to 9,500 psi in the Bakken. Closure stress is the stress required to hold a fracture open.
Bossier potential. E&P companies active in the Haynesville have begun directing some capital toward the Bossier shale, which overlies much of the Haynesville. Originally a name used for the Texas part of the Haynesville, the Bossier has come to denote a separate shale formation directly overlying the Haynesville on both sides of the Texas-Louisiana border. It is also known as the Middle or Lower Bossier. (The Upper Bossier shale, composed mostly of sands, lies farther southwest.)
According to Robert Hutchinson, an oil and gas investor focused on the Haynesville, very few rigs have been directed exclusively at the Bossier. “Because the Bossier is above the Haynesville, there is very little Bossier drilling activity since a successful Haynesville well holds the Bossier by production,” Hutchinson told World Oil. “While some operators have drilled exploratory Bossier wells to estimate reserves, there are few Bossier wells drilled for the sake of production.” Nonetheless, several active Haynesville operators have touted their Bossier results within investor presentations, eager to show the added value this new shale interval brings to their Haynesville assets.
New resource assessment. In August, the EIA released a new appraisal of technically recoverable shale gas and oil resources across the undeveloped portions of 20 US shale plays. The study estimates that there was 750 Tcf of technically recoverable shale gas in the US as of Jan. 1, 2009, of which 10% (74.7 Tcf) was in the Haynesville. The agency noted, however, that these estimates could be way off, since many of the plays are in early stages of development and production from future wells might vary widely from that of existing wells. This is particularly true in the Haynesville, which was only discovered in 2007 and has seen production increase much more rapidly than other shale plays.
The assessment further stated that, in the 3,547 sq mi of Haynesville shale that has been developed, the average EUR per well was 3.57 Bcf, higher than in any other area studied except for the Cana field portion of the Woodford, in western Oklahoma, which saw EURs averaging 5.20 Bcf. EURs in the top 10% of Haynesville wells average 13 Bcf, followed by 9.75 Bcf/well for the next 20%, 6.50 Bcf/well for the next 30%, and 3.25 Bcf/well for the bottom 40%.
Geophysical properties. Recently, CGGVeritas has applied its integrated geophysical workflow, based on amplitude-vs.-offset (AVO) analyses of pre-stack azimuthal seismic data calibrated with well logs and core measurements, to its Tri-Parish multiclient library in the Haynesville shale. The company’s approach seeks to characterize shale gas reservoirs by providing a quantitative understanding of rock properties such as acoustic impedance, Poisson’s ratio and Young’s modulus—which are related to reservoir properties such as porosity, total organic content (TOC) and mineral content. The goal is to predict optimal drilling and completion locations, in order to increase productivity and reduce exploration risk. According to company literature, “Relative production estimates across the field can be derived by combining seismic estimates of lithological, geomechanical and stress properties, correlated to existing well measurements. The selection of geomechanical and lithological parameters that provide the best correlation with production will vary from one shale play to another and must be derived for each survey using multi-attribute correlation.”
CGGVeritas says the overall regional stress regime in the Haynesville has an east-west maximum horizontal stress orientation, and there are observable lateral variations in the local stresses. Zones with relatively high brittleness (derived from estimated properties such as isotropic Young’s modulus and lambda-mu-rho) have been identified, and their associated differential horizontal stress ratio (DHSR, important for predicting hydraulic fracture behavior), fracture initiation pressure and closure stress have been estimated.
Multi-attribute analysis of the Tri-Parish data suggests that better development locations have a combination of certain key rock properties, such as high porosity, siliceous mineral content and TOC. Detailed rock property analyses showed that properties such as Poisson’s ratio and lambda-rho (incompressibility) are useful for identifying such areas. Areas of low Poisson’s ratio indicate the more siliceous, low-carbonate content normally associated with better porosity development. Bulk volume of gas can be estimated by combining these properties via multi-attribute analysis. These seismically derived predictions were calibrated with existing production, well core and well test measurements to determine optimal zones for drilling and completion. In general, no single attribute was found to provide sufficient information to correlate seismic data to production. However multiple elastic- and stress-related attributes (Fig. 1) were used to predict observed production (compensated by horizontal well length) with 95% correlation, showing several undrilled areas with potentially high productivity.
 |
| Fig. 1. CGGVeritas’ multi-attribute analysis within its Tri-Parish multiclient library in the Haynesville: Dynamic Young’s modulus (red represents high values) with plates showing DHSR and maximum horizontal stress orientation and bubbles showing initial production. |
|
OPERATOR ACTIVITY
The number of rigs working in the Haynesville and Bossier shales reached a peak of 184 in July 2010, according to Hutchinson. The current rig count, as compiled on Hutchinson’s blog haynesvilleplay.com, is over 40% down from that high level, at a Sept. 26 count of 106 rigs, Fig. 2. Of that total, 76 are working in the Louisiana part of the play and 30 across the Texas border, maintaining a split that has remained relatively steady since Hutchinson began reporting the count in January 2010.
 |
| Fig. 2. The Haynesville shale rig count has fallen over 40% from its peak in July 2010. |
|
Many large and small operators are active in the Haynesville; this report highlights the activities of just a few of the most active players. The rig counts used for each company are from haynesvilleplay.com, Table 1.
| TABLE 1. RIG COUNTS FOR TOP HAYNESVILLE DRILLERS |
 |
Chesapeake. Oklahoma City-based shale gas pioneer Chesapeake drilled what is widely considered the Haynesville discovery well in 2007 (Dallas, Texas-based independent Cubic Energy also claims to have first drilled into the shale), following geoscientific investigation in 2005 and 2006. In 2008, Chesapeake formed a joint venture agreement with Houston-based Plains Exploration & Production, to which it sold 20% of its Haynesville and Bossier assets for about $3.2 billion in cash and drilling carries.
Since its first Haynesville well, Chesapeake has built up its production in the play to an average of 1.085 Bcfd net (1.7 Bcfd gross) in July. “That’s been built 100% organically in just four years,” said CEO Aubrey McClendon in a second-quarter conference call. “If Chesapeake’s Haynesville asset were a stand-alone company, it would remarkably be the seventh largest gas producer in the US by itself.” Chesapeake’s Haynesville production in July represented more than one-third of the company’s total output, and included 15 MMcfd from the Bossier formation. The company holds 495,000 net Haynesville acres, about one-third of which (190,000 net acres) is overlain by the Bossier. Chesapeake’s net Haynesville acreage contains 4.157 Tcfe of proved reserves. The company’s drilling program is constrained by a relatively wide 650-acre spacing, and involves laterals of about 4,500 ft.
For a long time the play’s top driller by far, Chesapeake cut its drilling activity in the Haynesville substantially during the third quarter, and now is just slightly ahead of runner-up Exco Resources with 20 rigs running in late September. That’s down from an average of about 35 active Haynesville/Bossier rigs in the first half, which were needed to hold by production the acreage. The company plans to continue cutting down to 15 Haynesville/Bossier rigs by end of year, and to hold at that level until gas prices go up. “That will have an effect on our production next year,” said Chief Operating Officer Steve Dixon in the conference call.
Exco Resources. Unlike Chesapeake, Exco Resources has kept a relatively stable number of Haynesville/Bossier rigs running throughout 2011, so that the Dallas-based independent is now running neck-and-neck with Chesapeake at 19 rigs. Though down from the first half, that’s still higher than the 16.8 rigs that Exco average in 2010. The company holds 76,000 net acres with Haynesville/Bossier shale potential, primarily located in DeSoto and Caddo Parishes in Louisiana and in Harrison, Panola, Shelby, San Augustine and Nacogdoches Counties in Texas. A substantial portion of the acreage is held by existing Haynesville, Cotton Valley, Hosston and Travis Peak production.
Exco’s active rigs are split between its Holly core development area in DeSoto Parish, Louisiana, and the Shelby trough area on the Texas side. In the core area, Exco has moved from delineation to full manufacturing mode, using multiwell pad development in an 80-acre spacing—tightened from the company’s previous 160-acre spacing in the area. In the Shelby area, the focus is on delineating the company’s position and holding acreage. Shelby is expected to transition to full manufacturing mode in 2012.
The company completed 47 gross wells (20.4 net) in the Haynesville/Bossier play during the second quarter, and plans to drill a total of about 163 operated horizontal wells this year. That would be a 37% increase from the 119 Haynesville wells spudded by Exco during 2010, primarily in core DeSoto Parish area. The company’s Haynesville wells typically have laterals of 4,000–5,000 ft.
Continuous improvement in drilling time and optimization of frac design have helped Exco keep its costs relatively flat. This effort has focused on the company’s northern Louisiana assets, where well costs have fallen from a high of $10.8 million in second-quarter 2010 to $9.35 million in second-quarter 2011, due to variables including bit selection, construction efficiencies, reducing nonproductive time, proppant type/volumes, horsepower, equipment rentals and perforation spacing. The company’s fastest well to date was drilled in 28 days. On 11 wells, the curve and entire lateral were drilled in a single bit run, and, in one well, Exco drilled from surface through the intermediate section with one bit run.
Exco had 706.8 Bcfe of proved reserves in the Haynesville shale as of Dec. 31, 2010, largely due to strong delineation results in the Shelby area. IP rates in the area have averaged 18 MMcfd, with two Haynesville wells having IP rates over 30 MMcfd and one Bossier well over 25 MMcfd. In the Highlander segment of the Shelby area, average IP rates exceed 28 MMcfd.
In August, Exco reported operated Haynesville/Bossier production of 1.2 Bcfd gross (365 MMcfd net), with 232 operated and 123 non-operated wells producing. The company estimated that an average of about 25–30 MMcfd net was curtailed due to plant shutdowns after an incident at its 50%-owned TGGT Holdings treating facility in Red River Parish killed one worker and injured another. Production was expected to remain curtailed through the end of the third quarter.
In May 2010, Exco and BG Group jointly bought Common Resources, consisting of properties in Shelby, San Augustine and Nacogdoches Counties, Texas, for $442.1 million ($221.0 million net to Exco). Two months later, the two companies jointly purchased additional properties from Southwestern Energy for $357.8 million ($178.9 million net to Exco).
Encana. Canadian gas producer Encana has seen strong growth in the Haynesville, where production was up 70% to 487 MMcfed in second-quarter 2011 from a 2010 average of 287 MMcfed, partially offsetting production decreases due to divestitures elsewhere in US. In 2009, the company averaged just 61 MMcfed from the Haynesville.
Encana is shifting its drilling focus from lease retention to expanding and optimizing its pad drilling (or “hub”) activities, a transition that includes seeking regulatory approvals for longer laterals and building on its low-cost completion program. On the company’s drilling hubs, drilling times have been reduced 20% in the last year to 40 days, and a number of wells this year have been drilled in 35 days. In the first quarter of 2011, the shift to hub activity resulted in about a 25% reduction in drilling costs compared with lease-retention drilling. Further cost reductions are expected through the deployment this year of fit-for-purpose pumping equipment and service supply agreements.
Drilling is down somewhat from last year, with 45 wells drilled in first-half 2011 and a planned total of 85 wells by year-end, vs. 106 net wells drilled in 2010. The company is operating 12 rigs in the Haynesville, down by almost half from a 2010 average of 22.4 rigs running. This demonstrates Encana’s increased drilling efficiency: In 2010 it drilled an average of 4.7 wells per rig running, compared with a projected 6.4 wells per rig in 2011 (if the first-half well counts and rig averages were applied to the whole year).
During second-quarter 2011, Encana obtained regulatory approval for its first cross-unit, alternate-unit wells in Louisiana. The approval allows the company to drill horizontal wells with laterals of 7,500 ft across existing blocks of three sections. Encana plans to spud its first cross-unit well before the end of the year.
ExxonMobil/XTO. In its 2010 annual report, ExxonMobil called the Haynesville and Bossier shales its “fastest growing unconventional play.” Exxon’s 240,000 net acres are operated by XTO, which the major acquired in mid-2010. That leasehold includes 67,000 net acres added last year, 46,000 acres of which came through the $695 million cash purchase of Boulder, Colorado-based Ellora Energy last October (along with production and pipeline capacity). The remaining 21,000 net acres were added through a 108,000-gross-acre joint venture with Encana. Exxon has been busy consolidating these Haynesville acquisitions into XTO, along with recently acquired Eagle Ford and Marcellus acreage.
A relatively active drilling program throughout 2010 resulted in gross operated production of 250 MMcfd by the year’s end, more than four times that of year-end 2009. The company ran an average of 11.7 rigs in 2010, with development focused in its prolific southern core area. Bossier shale testing also began last year. XTO’s Haynesville/Bossier rig count has dipped somewhat this year, averaging 10.8 in the first half and falling to seven in late September.
In a speech on June 14, XTO President Jack Williams said Exxon is applying its Fast-Drill technology in the Haynesville. According to company literature, “This physics-based process combines real-time digital analysis of the drilling system’s energy consumption with a structured approach to well planning and design to ensure that a well is drilled as efficiently and quickly as possible. Applying this method has resulted in a rate-of-penetration improvement of up to 100 percent.” Among other applications, the method helped Exxon to achieve record ROPs on extended-reach wells in the Sakhalin-1 project in Russia.
El Paso. As El Paso Corp. prepares to spin off its E&P business into a publicly traded company called EP Energy by the end of the year, successful Haynesville drilling has been a major driver of the company’s overall production increases. As of June 30, El Paso had 83 operated wells and 260 MMcfed of total production related to its Haynesville program—more than one-third of the company’s total US production. The company currently has five rigs running in the play, and plans to keep about that many running through the rest of the year. The company estimates that it has 415 future drilling locations on its 43,000 net acres in the Haynesville, with resource potential of 800 Bcfe. It spent $197 million there in first-half 2011, 27% of its total capex budget and second only to the Eagle Ford.
Of El Paso’s four major E&P operational areas, the Haynesville is the only one that is gas focused. This reflects the company’s confidence in the economics of the play even at low gas prices, especially in the busy Holly area of DeSoto Parish, Louisiana, where El Paso has about half of its Haynesville acreage and is one of several companies employing a gas-factory concept using multiwell pads. El Paso estimates that wells in the Holly area will cost $8.7–9.3 million to drill and complete, and will yield an EUR of 6–7 Bcfe, resulting in an internal rate of return between 20% and 30% at $4/MMBtu. By way of comparison, the company estimates wells in its other two Haynesville areas, Bethany Longstreet and Logansport, to have rates of return between 5% and 15%.
Petrohawk. BHP Billiton’s $12.1 billion purchase of Petrohawk, closed in August, would seem to reflect at least in part a high level of confidence in the long-term profitability of Haynesville gas. After all, Petrohawk holds the second largest net leasehold in the play, after Chesapeake, at 368,000 acres—294,000 in Louisiana and 74,000 acres in Texas. It is also one of the top producers, averaging 684 MMcfed in the second quarter of this year.
During the same time period, Petrohawk drilled 21 operated Haynesville wells, as well as 67 non-operated Haynesville wells and three Bossier wells. Non-operated activity exceeded expectations, in terms of both activity level and capex, primarily due to the transition to full section development by some industry partners.
The company has achieved cost reductions in Haynesville completions of about $600,000–800,000 per well, largely as a result of changes in well design that require two fewer frac stages per well, lower overall sand requirements per well, and improved pricing for resin-coated sand. During the second quarter Petrohawk averaged slightly less than 45 days spud to spud, a reduction of five days compared with the preceding quarter. Significant additional improvements are expected as the company moves toward pad drilling and full section development toward the end of 2012. Meanwhile, the company substantially drew down its Haynesville rig fleet in the third quarter, to a late September count of four rigs, compared with averages of 12.5 in the first half and 14.2 in 2010.
Improvements in water handling and usage have contributed to more flexibility in water sourcing. About half of all Petrohawk-operated wells in the play have been completed with 20% recycled water. In the first half of 2011, the company pumped over 47,000 bbl of recycled wastewater on Haynesville completions.
Others. Shell, through its 100% owned subsidiary Shell Western Exploration and Production (SWEPI) is the owner of a 50% interest in leases that cover about 400,000 Haynesville acres in DeSoto, Red River, Sabine and Natchitoches parishes, the company told World Oil. Otherwise, the major has been exceptionally tight-lipped about its Haynesville assets, declining to comment on production, reserves or drilling activity. The company has kept a steady average of about nine rigs running in the play throughout 2010 and 2011.
EOG Resources has directed 20% of its 2011 capex toward dry gas drilling to hold acreage, with a substantial portion of that drilling in the Haynesville/Bossier play. The Houston-based independent holds a 190,000-net-acre position in the play, containing 11 Tcf of potential (i.e., not proved) reserves—5.6 Tcf in the Haynesville shale and 5.4 Tcf in the Bossier. The company currently has six rigs operating in the play, down from an average of 7.2 in the first half and 10.4 in 2010.
TECHNOLOGY ADVANCES
Most of the technologies developed specifically for Haynesville have emerged in response to the extreme HPHT conditions of the formation in comparison with other producing shales. An instructive example of this is Newpark Drilling Fluids’ Evolution water-based drilling fluid system, developed in 2009 as an alternative to the oil-based muds (OBMs) that had become the default in the play. A number of operators had attempted different water-based muds (WBMs) in the Haynesville in order to reduce the high costs of drilling in the play (oil-based muds are more expensive, and more difficult to dispose of after use), but these attempts had failed.
After considerable sampling and analysis of the shale, Newpark determined that the shale was not particularly reactive, as had previously been hypothesized. Rather, the major issues were the high temperature and carbon dioxide content. At the high temperatures found in the Haynesville, the standard viscosifying polymer of choice, xanthum gum, deteriorates rapidly.
“We also discovered a significant amount of CO2 in with the gas, such that if you use a conventional bentonite mud, it flocculates severely, giving you a viscosity hump that significantly limits your lubricity,” lab manager David Breeden told World Oil. “The only way to treat out the CO2 is with lime, and at bottomhole temperatures of 350° and above, you don’t want very much lime because you’ll make cement. … So you’re constantly going over the two viscosity humps of CO2 and calcium.” The high mud weights required in the Haynesville, exceeding 15 lb/gal, further complicated the search for alternatives.
Newpark went to work to formulate a water-based polymer system that exhibited thermal stability at temperatures in excess of 400°F, provided contaminant resistance to CO2 and drilled solids, and provided lubricity comparable to that of OBMs. The resulting system, incorporating a polymeric viscosifier and suspension agent, an HPHT lubricant, a rheology modifier and a general-purpose fluid conditioner, performed successfully in the field and became the first water-based system used repeatedly in the Haynesville.
Just as the Haynesville’s high temperatures have caused problems for conventional muds, its high pressures have caused problems for conventional proppants, i.e., uncoated sand. At the formation’s high closure stresses, these proppants tend to crush and form large quantities of fines, which clog pore throats and limit productivity. As a result, stronger, resin-coated sand proppants have been increasingly popular in the Haynesville, with several suppliers struggling to keep up with the high demand.
While most Haynesville-related technology advances have been related to getting gas out of the HPHT formation, the most important advances going forward will likely be those that present new ways of marketing the produced gas. Thus, planned investments in LNG export, LNG- and CNG-fueled vehicles and gas-to-liquids technologies may be the most hopeful sign yet for the play. 
|
