SPE 125903 Horizontal Wells with Multistage Fracs Prove to be Best Economic Completion for Many Low-Permeability Reservoirs
B.W. McDaniel and Keith Rispler, Halliburton
ght 2009, Society of Petroleum Engineers This paper was prepared for presentation at the 2009 SPE Eastern Regional Meeting held in Charleston, West Virginia, USA, 23–25 September 2009. This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.
Abstract Like any other business, return on investment (ROI) drives the decision-making processes in the oilfield. The “cut-cost” approach can only be successful when resulting production is adequate. Especially during the past few years, it has been observed that most North America operators using horizontal completions in low-perm oil or gas reservoirs have abandoned the “lowest-cost” approach in favor of a “maximize-production” mindset. Maximizing ROI requires operators to continually evaluate both cost and effectiveness of various completion and stimulation options. With low- to ultralow-perm reservoirs, it has been proven in many fields in North America that using a long lateral section combined with effective, controlled placement of large, multistage propped fracturing treatments can offer the best economic return. Often being combined with pad drilling, the trend continues to be using fewer wells while maximizing the volume of reservoir rock that is fracture stimulated within each completion. There are two interrelated choices that can be combined in several ways to decide how best to complete a lateral section to achieve maximum benefit: ? Assuming a solid liner is installed, should the operator cement, seal off the annulus into sections by some method, or leave it unsealed? ? What method should be used to provide frac-stage isolation within the liner? This paper will provide an overview of several different horizontal completion methods and stimulation techniques most commonly used in North America for low-perm reservoirs during the past few years. Included are two different operators’ experiences with multiple application methods, but the common denominator is that they included multiwell overviews. Managing risk properly will usually be more than a “current-well” mentality and requires a more field-wide approach, including the cost of completion interruptions from unscheduled/unexpected events. One case takes the comparisons through completion cost and all the way to ROI results, where all wells had more than six months of production. Another case illustrates a situation where the operator concluded that well production is more dependent on reservoir quality than on his choice for completion methods. This case includes more than 75 wells with six to nine fractured intervals per well, comparing not only costs for four methods, but also showing the representative cost variations or overruns. Generally, once a drillsite has been chosen, the most important variable that can be affected is effectively placing individual hydraulic fractures at (and only at) preselected locations along the completed lateral section. Choosing the method to be used is best made before drilling the well, but might have to be revisited if formation properties are different than anticipated or drilling problems result in a wellbore the is a poor fit for the completion plan originally selected. Obtaining effective isolation of stimulation stages is often the primary goal required to achieving adequate production response and effective reservoir exploitation while managing the costs to achieve best ROI on a field-wide basis. Introduction In North America, possibly the most significant change in operator mindset of the current decade has been the economics of completing wells in low- to ultralow-perm formations. The most dominant factor driving the drilling craze was higher gas prices. With that element no longer present, it seems the best opportunity for best economics is long laterals with multistage fracturing treatments. Not many areas outside North America have yet observed this change in mindset, but it is sure to come. More than ever, the industry attitude toward a new tight gas play is to immediately consider if this is another opportunity to use long lateral completions with multistage hydraulic fracturing as the basic completion approach. Unless the reservoir is made up of numerous thin pay sands spread over hundreds of vertical feet, then it is no longer simply assumed vertical
completions will be used. Additionally, once lateral completions are considered, it seems that operators then immediately try to determine if there is a practical limit to the length of lateral to be completed. It seems unusual (but refreshing!) to see our “staid, old oilfield” so quickly change general perceptions. As few as six to seven years ago, it was short work to make a list of low-perm gas and oil plays that were predominantly horizontal completions. Presently, most in the oilfield could probably list almost ten on instant recall. Additionally, most in their list would be among the most active drilling plays. With the recent crash of oil and gas prices since the third quarter of 2008, we see further proof of the importance of these low-/ultralow-perm fields where completions are predominately long lateral sections. The US drilling-rig trends during the past two years are shown in Fig. 1, and this clearly illustrates that when profits are squeezed by low hydrocarbon prices, it is the horizontal completions that are showing the greatest tenacity. On the right in Fig. 1 is a pie chart with a July 3, 2009 snapshot that best illustrates current trends. As the vertical rigs typically drill wells in fewer days, these rig percentages don’t reflect total well counts. However, as recently as April 2008, the number of active vertical rigs was twice the number of horizontal rigs. Nearly all these horizontal rigs are operating in fields that are low- to ultralow perm (many are shales) where long laterals and multistage hydraulic fracs are critical components in the completions. This implies many operators are seeing (or expecting) better economics with their horizontal drilling programs.
Fig. 1—USA rig-count data (from BakerHughes.com) shows the economic importance of horizontal drilling as oil and gas prices started their free-fall in the marketplace.
Not only is there a proliferation of operators choosing to complete low-perm reservoirs with long laterals and multistage fracs, but in many such fields in North America, there is also a push for longer laterals and using more frac stages. This trend might have first grown to near-epidemic status in either the Barnett shale of north Texas or the Bakken reservoir in Elm Coulee field of eastern Montana (Wiley et al. 2004). It has also been witnessed in many non-shale reservoirs, such as in the very tight Cotton Valley sands of east Texas and several of the Canadian tight sands, such as the Montney. It has come full circle back to the Bakken in several areas of North Dakota (Miller et al. 2008). In early 2009, some operators reported an increasingly dominate relationship observed with Haynesville shale completions (Fig. 2) that higher IPs at completion strongly correlates to larger volumes of proppant being placed, and in many cases this was achieved by longer laterals and more frac stages. Also, Fig. 2 shows the lower typical performance of vertical wells as compared to the potential offered by long lateral horizontal completions. Currently, some operators appear to only be drilling vertical completions in Haynesville reservoirs when the primary purpose is either data gathering or for acreage holding.
Fig. 2—Comparison of vertical and horizontal wells to show correlation between production-rate potential and quantity of total proppant placed during completion (EnCana Corp. 2009).
Cost, Completion Risk, and ROI Two of these three terms are common to almost every business endeavor, but how they will specifically be used in this paper needs to be addressed. Historically, the stimulation costs attributed to the pumping-service provider have been relatively simple to isolate from other well-completion costs, but for cases where a significant number of separate stimulation operations are necessary, this becomes more difficult. Some processes require significant prestimulation operations, or even special completion installations unique to the stimulation method but possibly provided by another vendor. For many years, operators “designed a frac job,” whereas today, for most low reservoir-perm horizontals, operators more typically “design a completion approach” with numerous imbedded “design-a-frac-job” steps that often are fairly repetitive designs, as long as formation response was judged to be desirable on the previous stages. Factors critical to isolation of completed stages and wellbore preparation for the next frac stage are considered more often now. Of course, a premature screenout (SO) during any frac stage can disrupt this sequence, and the duration of the following nonproductive time (NPT) needed to recover from such an event are important factors. An event that disrupts the planned event sequence or timing can increase completion costs, or it can cause a reduction in the number of stimulation stages that are achieved by the completion operations in order to maintain a cost limit. The term “risk” is far too broad to be used with respect to only a single portion of oilfield operations. Although it will be highly dependent on the geology and geomechanics of every zone penetrated, it is generally considered that the drilling operation will incur highest risk, followed by the completion operation as the next level of risk, as these are the most critical times for experiencing not only the loss of the well but also possible loss of additional surface equipment and even lives. Next, there is the risk that the well will be completed but then abandoned as having no productive potential or judged to be a dry hole and abandoned before expending completion costs (typically no meaningful safety risks for this outcome which could multiply costs). The next risk level is that a completed well does not produce at economic levels. One more risk level is of importance: Failure to maximize ROI. With most low to ultralow perm reservoirs, the reservoir quality can vary significantly, and we often have limited understanding of the quality before completion. This can mean that production rates and rate of decline may not reflect whether or not we maximized ROI for a specific well. Well operators must review many more variables with multi-well databases before fully understanding what completion plan is best for them. Fortunately, a high percentage of horizontal completions in North America during this decade have provided enough production to eventually recover most of the well investment, although some have been sub- to marginally economic. Such operator company management statements as “this is a play without dry holes” is one heard often in a few of the present shale plays in the US. To address the challenge to maximize ROI, operators generally will evaluate more than one method for lateral completion and possibly several options related to hydraulic fracturing methodology for multistage stimulation, as well as stage volumes, fluid systems, proppant volumes, and possibly other proppant variables (sieve size, sand vs. manmade, and maybe resin coatings). Quite a challenge. This paper will limit discussion and review to only annular-sealing methods and methods for controlling stage placement along the lateral. Looking beyond any “learning-curve” problems with the different approaches, there are three primary criteria that help operators maintain control of their completion economics: Quantity of nonproductive time (NPT), cost of recovery from unplanned events, and loss of completion zones cause by stimulation-related problems. The term “completion risk,” when applied to each method used to achieve multistage fracture-treatment placements, can be used to reflect the significance unscheduled post-drilling events might have on final well economics. With respect to
varied reservoir conditions, each completion/isolation method can be more fragile or more robust with respect to a specific challenge. For general discussion here, factors related to the unknown geological variations are not included. Lateral Length and Frac Spacing The length of the lateral, the spacing of fracs, and the size of frac stages are not topics that can be responsibly addressed without first isolating the discussion to a specific reservoir, and in some cases even to the economic metrics of a specific well operator. Although it seems simplistic, it is usually effective to model each specific frac stage as though it were being created from a “virtual” vertical wellbore at that specific location within the reservoir. Few (if any) of commonly used frac-design software packages can incorporate the effect of any other hydraulic fracture previously created at another point along the lateral, especially because the elapsed time since the generation of a previous hydraulic fracture would also be important. Depending on the method used, it could have been only minutes to multiple days beforehand. In some instances, some parts of a lateral section can present more concern for avoiding excessive fracture-height growth. Additional to concerns of communicating with wet zones or layers, which are porous but depleted, with long laterals it is possible that a geologic anomaly that should have been avoided might have been drilled through. Again, every wellbore-completion plan should be reconsidered after it is drilled. Some authors have addressed a few specific field cases to make recommendations for lateral lengths and frac spacing distances. However, most have used generalized reservoir descriptions that, even if the reservoir model can define significant layering, it still will assume lateral homogeneity of the reservoir which is seldom representative of most low-perm reservoirs. Reservoir simulation of long, multifracture stimulated lateral wells is an art that is rapidly evolving, and many of the low- to ultralow-perm reservoirs are known to be extremely variable. Many times, the frac spacing along a specific lateral will be adjusted to high-grade frac placement using wellbore data while typically maintaining some (operator chosen) lower and upper limit to the distance between frac locations. The validity of reservoir-production modeling also is subject to just how well known is the fracture angle to the wellbore axis. This does not appear to be commonly known, except for a few fields, unless extensive microseismic monitoring has been performed. Typically, the need for a monitor wellbore available to deploy the geophone array often limits the possibility of using microseismic monitoring, especially early in the development of a field. Optimum length for laterals usually is another item determined by well performance. Occasionally, the maximum lateral can be dictated by some type of governing limits. However, in many recent fields the limits are based on ability to effectively fracture stimulate the entire lateral drilled. This appears to have been a major factor for many Bakken formation completions in Montana and North Dakota for several years, where 10,000 ft lateral lengths have become common, and at least one of 23,000 ft lateral is on the books. Several operators in the Barnett shale have been rewarded when choosing longer laterals than were common at that time. The same results seem to be occuring very recently in some areas drilling the Haynesville shale, and possibly also the Bossier shale. As seen earlier in Fig. 1, at least one Haynesville well operator is reporting a significant correlation between a well’s production potential and the quantity of proppant placed. Looking a little deeper into this data reveals that both more proppant per frac stage and increased lateral length allowing for more stages in newer completions are combined to achieve the greater volumes of proppant placed per well. Globally, multistage fracturing of moderate to long laterals in moderate- to low-perm reservoirs is slowly becoming more common. Often, many of these operators are challenged by their lack of understanding of available technology for fracture stimulation in horizontal wells, even though they are more often encountering the need for it. This need can more often be seen in nonprolific offshore completions, where partially depleted existing wells now need fracture stimulation but are not completed in a way that allows for many of the better options, such as the methods discussed below. Lateral Completion Options This paper assumes that bullhead frac treatments pumped with full access to the lateral section have been adequately addressed by other papers. Whether the lateral is some form of openhole completion or has a perforated cemented liner, if the fracturing operation will be bullheaded into the lateral, then it will not be reviewed here. For low-perm reservoirs, such completions rarely have resulted in optimized production response. A second limitation for this paper is assuming the lateral will have some type of liner, so there is not the added question of wellbore integrity related to formation properties. There have been several past papers related to multistage fracs in openhole completions (i.e., bare or slotted/preperforated liner), and an extensive listing is presented in Table 4 of McDaniel et al. 2008. An additional recent reference is Jakobsen et al. 2009. Liner and Frac-Stage Isolation: To perform effective multistage fracturing operations on a lateral, a well operator must decide what method will be used to achieve frac-stage isolation. In essence, this could be thought of as two decisions, although they are intimately linked together: 1. Will the lateral be completed using a nonisolated openhole (bare, slotted, or perfed liner), a solid liner without any annular isolation, or a liner with some type of annular isolation (i.e., cemented or some form of external casing packers [ECP])? Some discussion here will include mechanical ECPs, inflatable ECPs, and swell packers. What staging method should be used to ensure fracture stages are isolated from each other inside the wellbore?
Although this paper will not attempt to describe every potential lateral completion option, it does include six categories (Table 1) with several specific methods of application that are familiar to the authors. Several of these will be further discussed later in the paper. These options are reviewed assuming they were selected by the well operator primarily to allow effective multistage fracture stimulation as the dominant goal of the completion program.
TABLE 1—LISTING OF SIX HYDRAULIC FRAC MULTISTAGE ISOLATION METHODS WITH ABBREVIATIONS AND A METHOD NUMBER FOR SIMPLIFIED REFERENCE IN TABLE 2. Frac Staging Type of Frac Staging Isolation Method Abbreviations for Frac Staging Isolation Method Number Method HydraJet Frac HJF 1 Wireline/Pump Down Ceramic Bridge Plug WL CBP+PerfGuns* 2 and Perf Guns Same as above, but with Flow-Through WL Flow-Thru CBP+PerfGun* 3 CBP CT operated Sliding Sleeve Tools CT + SS Tools 4 Ball Activated Sliding Sleeve Tools BASS Tools 5 Hydrajet Perf-Annulus Frac + Sand Plug HPAP + SPD 6 Frac Isolation
*Both of these methods would be referred to as “Perf and Plug” operations. Table 2 provides a listing of the lateral-isolation design (first three have no isolation, last four have at least partial isolation) and then lists in the second column the “available” methods for achieving fracture-staging separation when using the specific lateral-isolation design. The third column lists (to the authors understanding) the most commonly applied fracstaging method number.
TABLE 2–LATERAL-COMPLETION DESIGN AND CORRESPONDING USE OF VARIOUS MULTISTAGE FRACTURING Lateral-Isolation Design Available Frac Staging-Method No. Most Commonly Applied Staging-Method No. Bare openhole 1 1a Preperfed or slotted OH liner 1 1a,b Solid liner only 1b to 6 2, 5, or 6 Cemented liner 1b to 6 2, 5, or 6 Mechanical ECP 1b to 6 2 or 5 Inflatable ECP 1b to 6 2, 5, or 6 Swell packers 1b to 6 2, 5, or 6 a When Method 1 is not used, most likely no method is used. b Method 1 can be rate restricted if liner OD is less than 5 in.
Annular Isolation of Horizontal Liners Cemented Liner: There is still extensive use of cement to provide annular-space seal for liners in lateral sections. The fracturing implementation might be through conventional shape-charge perforating or through perforations made by hydrajetting operation. Acid-soluble cement is sometimes used (McDaniel et al. 1999) to improve communication with the reservoir after using conventional perforations in a staged-frac completion of a horizontal. More recently, this process has been extended to use when BASS tools will provide the wellbore openings for a frac stage but the cement sheath is still present in the annulus after the frac ports have been slid open. Some type of acid (depending on formation conditions) is used to dissolve away the special cement blend and give a very short openhole window to the formation. Liner Not Sealed: In some cases, solid liners are installed with no annular seal. A few operators have reported consistent success using limited-life viscous fluids as a temporary annular sealant, but the authors could only find published references of this type for fracture-acidizing applications instead of multistage proppant fracturing of horizontal wells. Some of these have published cases where no annular supplemental-seal mechanism was used other than what can be created by the fluid dynamics of using hydrajet-assisted fracturing (HJF) to affect adequate control, placement, or distribution of hydraulic fractures along a lateral (McDaniel et al. 2008). External Packers on Liner String: If there are relatively “gauge hole” sections of the lateral, and if the formation itself remains competent for a reasonable time after being drilled, it is usually believed that isolation of annular zones can be achieved using some type of external casing packer. Very recently, Roundtree et al. (2009) closely reviewed the possible fracture initiation-related effects that could be present using external casing packers. Achieving a seal using a very short packer-element length can require that the sealing material be forced against the drilled lateral wall with considerable force. The ranges of pressure listed below that could be expected are discussed in Roundtree et al. (2009). These are probably the three types of external packers most predominantly being used (currently and the recent past) in North America: ? Mechanical-set packers—This requires tool movement to extrude the packer elements to provide an annular seal, exerting a significant force against the formation (5,000 to 10,000 psi) over a relatively short length (~4 to 8 inches).
? Inflatable packers—The liner is pressurized to extend the packer elements to effect an annular seal, with 1,000 to 3,000 psi of stress against the wellbore wall for a length of 4 to ~10 ft (shorter length usually uses higher pressures). ? Swellable packers—Have rubber elements bonded to a short section of pipe and will slowly swell in the presence of hydrocarbon (or water, in some cases), engineered to have adequate swelling to fully contact the wellbore wall, possibly requiring several days, and typically exerting much smaller stress (<200 psi) over a far greater length (~3 to 10 feet). This may be the most effective packer for overcoming poor wellbore quality Concerns for Installation of Liner with External Packers: When the OD of a packer (or other hardware) is significantly larger than the liner OD, it is understood there is a significantly higher risk of failure to the completion string to be completely installed (i.e., a completion risk). The first obvious criteria is the larger the OD, the greater the risk. Many vendors offer products within the three external-packer categories above, but it is generally acceptable to assume that the swellable packers (to date, at least) offer the least-challenging OD for placement of a liner string. Of course, the swell packers offer the lowest mechanical risk, they must be exposed to fluids that activate the swelling after placement, and normally several days will be needed. Additionally, they offer no opportunity to test them or validate their sealing performance before observing pressure data during the multistage fracturing process. Liner Hanger: The applications discussed in this paper all include hydraulic fracturing operations of some type inside the liner, meaning a seal is generally needed at the liner hanger. An operator’s options for effective multistage fracturing will be greatly limited without a liner seal. Additionally, there is added value to a liner-hanger design that can still be sealed if the completion string cannot be fully deployed. Without the liner annulus sealed, inflatable packers cannot be set, and if the string is stuck there might not be options for setting mechanical packers. A liner hanger designed to circumvent such problems was discussed in a recent paper (Vargus et al. 2008). The type of liner hanger employed affects the completion risk for the method. Providing Isolation of Fracturing Stages When the lateral is completed openhole (bare, slotted, or preperforated liner), there might be only one established method to effectively control the location of and isolation between hydraulic-fracturing stages, which is the hydrajet-assisted fracturing (HJF) method. With many papers in the literature already describing applications of the process, it is not discussed here (McDaniel et al. 2008; Jakobsen et al. 2009). The HJF process (as any other) has it strong points and limitations, especially in the presence of small lateral size or other wellbore-ID restrictions. Also, low-strength or age-weakened casing could prevent its application. Adequate ID of the lateral section is a prime requirement unless only low injection-rate fracturing treatments are desired. Currently in North America operations, this is seldom the case. As formation permeabilities being drilled began to consistently be below 0.01 md, the use of high-rate, large-volume waterfrac treatments began to proliferate. To achieve adequate frac-injection rates, a well operator must provide adequate casing and liner ID. Many recent completions are further complicating this problem with extremely long laterals, and the frac stage near the toe needs frac-injection rates similar to zones near the heel. First Frac (Toe) With low-perm formations, until a significant hydraulic fracture is present to allow meaningful fluid injection, essentially the only methods for causing something to happen, such as perforating or moving a sleeve to open frac ports, is the use of jointed pipe or a CT string, or to have a tool installed in the liner toe area that responds to fluid pressure build-up. With a CT string, either a conventional- shape charge perforating gun or a hydrajetting tool could be used to provide wellbore communication for the first frac stage. Without CT use, the method of choice is usually to install a pressure opening (operated) valve (POV) in the liner string near the toe. Typically, pressuring to some value, usually based on shear pins, will slide a sleeve to open ports and the first frac stage can be pumped. After the first fracture is completed, then the wellbore-displacement stage of the frac can be used to carry down items, such as ceramic bridge plug (CBP), with perforating gun attached or a ceramic ball to both seal and cause a sleeve to slide open.. Pump-Down Bridge Plugs In vertical-well applications, the more highly dependable, easier-to-drill-out wireline settable ceramic bridge-plug (CBP or frac plug) technology has allowed the retirement of cast-iron bridge plugs from most stimulation operations. With respect to number of well applications in North America, this is probably the most common current (mechanical) method for multistage frac operations. In a smaller number of applications, a “flow-through” design (using a free-floating or a captured ball) is popular. With a small design change by the vender companies, the wireline CBP became a commonly applied pump-down deployed tool that also could pull down a perforating gun in horizontal wellbores. This assumes the toe area of the wellbore can accept some fluid for the pump-down operation As long as the wellbore is free of solids, this is rated as a low-risk installation. However, when pumped closely behind the wellbore flush of a fracture-stimulation treatment, this is not always a “clean wellbore” condition. In some cases, operators will choose to use a CT string to deploy the plugs and perforating charges. In other cases, where wireline guns might have failed to adequately fire, CT-string deployment of a perf gun can be required to avoid having to
skip the completion of one stage. A significant amount of costly nonproductive time can accrue, or in the most extreme cases, large sections of the planned completion can get skipped when there are problems with premature CBP sets because of wellbore trash, or inadequate perforating occured, even when the plug is set at the designed lateral point. In many of the deeply buried low- to ultralow-perm formations, there can be increased difficulty with formation breakdown. Many of the tight sands and shales are very nonhomogeneous, and some areas within the reservoir can be more difficult to achieve fracture initiation. In other cases, there may be an excessive number of fracs that will extend simultaneously, giving shorter, poorly propped fracs. If this is extreme, it can cause a premature screenout that ends the stimulation treatment at that wellbore location. In some fields where such problems occur very often, operators sometimes use the flow-through plug design, which is commonly used in vertical well applications but more sparingly in horizontal completions. This option would allow additional pump-down operations even after the frac plug has been set. Once perforating and formation breakdown operations are satisfactory, a ball is pumped just ahead of the next frac stage to seat on the top of the frac plug. Additional monitoring for controlling flowback between frac stages is required if a flow-through frac plug is used. This can require the use of a flowback manifold with chokes. As the “perf-and-plug” method is probably the most commonly used of the multistage frac methods, it will be the first one discussed. The introduction and rapid development of ceramic bridge plugs was a radical improvement over cast-iron bridge plugs in use for many years. Because they can be easily run, set, and later drilled out, the use of a ceramic frac plug can now be the nearest thing to a standard practice for temporary wellbore isolation during multistage frac operations in both vertical and horizontal wells. In possibly most low-perm horizontal wells, the pump-down plug will carry down the perforating gun so that, after setting the plug, the next frac target above the plug can be perforated. Once the guns are retrieved and judged to have fired properly, it is time to start the next frac stage. If the gun inspection is disappointing, or if achieving formation breakdown is a problem, there might be no way to reperforate the same location, or very close above the plug just set, unless the frac plug is the flowthrough type that is sealed by pumping down a ball. Although these are occasionally used, they are not very common. If a solid frac plug was used, and the inadequate perforations will allow some fluid entry into the reservoir, then a choice must be made: try to “open up” the perfs with acid or with proppant slugs, or immediately try to pump down another perf gun or another frac plug with attached perforating guns. If the completion is a cemented liner, this is less of a concern. However, the space for shooting higher perforations will be somewhat limited if the annulus is sectioned/sealed by external packers. If perforations cannot be shot within the area between the two packers defining the annular space, a frac stage can be lost in addition to added material and service costs and other cost effects of the lost time. The variables discussed above are a part of the completion-risk factor for using a specific type of completion with pump-down CBP for frac-stage isolation. A typical completion procedure using pump-down plugs and perf guns could be: 1. Perforate first stage with CT-deployed tubing guns or open a POV at the toe position 2. Perform a dynamic fluid injection test (DFIT) (typically only before first frac stage unless premature screenouts are a common problem) 3. Pump first fracturing stage 4. Run plug and gun assembly to kick-off point with wireline. 5. Pump down plug and perf guns at an average rate of 1 to 4 bbl/min (slowly through build section) 6. Pump plug in place using line speed of ~ 100 ft/min. (if experience has proven that acid might be needed to ensure effective formation breakdown, then alternating stages of water and acid (15 to 25 bbl) are used to help ensure acid is placed across the zone of interest 7. Set plug ~ 70 to 80 ft below point for next set of perfs 8. Set CBP and pull wireline perf gun to next interval, pressure-test plug, and fire perf guns 9. Pull wireline out of hole 10. Frac second zone and repeat process A typical completion procedure using CT-deployed plugs and perf guns could be: 1. Perforate first stage with CT-deployed tubing guns (or through an installed POV near the toe) 2. Perform DFIT (injection/falloff test) 3. Pump first fracturing stage 4. Run in hole with CBP and perforating assembly and set plug 5. Pull perf guns to next interval. Pressure-test plug and perforate second interval with CT-deployed tubing guns. 6. Spot 15 to 25 bbl of acid and pull CT out of hole (see acid comments above) 7. Frac second zone and repeat process Sliding Sleeves The installation of sliding-sleeve tools during the makeup of the lateral-liner string can allow placement control for each hydraulic-fracturing stage when combined with some type of annular-sealing mechanism, most commonly with mechanicalor hydraulic-actuated external casing packers spaced along the lateral during installation, with the liner sealed against the casing at the top of the liner. This adds a degree of completion risk over cemented-liner completions, with larger-OD tools
along the string and the need to achieve complete installation of the liner to obtain a liner seal at the top to ensure the liner can be manipulated to set mechanical packers or pressured adequately to activate inflatable packers. There have been two primary methods for opening ports on sliding-sleeve tools: CT string manipulation or ball-drop seating against a baffle to slide them open. With the assistance of the CT string, operators can actually both open and close sleeves. Using a CT string to open/close sliding sleeves offers the option of using any number of stages, and with the CT string present near the frac ports being used, it can serve as a real-time bottomhole gauge—a luxury most fracturing operations do not have. Additionally, if an unplanned screenout were to occur, the CT string can save many hours of nonproductive time in achieving cleanout operations and preparation for the next event, such as trying to refrac the problem zone or preparing the wellbore for the next scheduled frac location. There is a more simple and faster sliding-sleeve implementation (if all goes according to plan) using progressively larger balls to both seal above a previous frac stage and also open ports to a new zone in the annulus for the next frac stage. As mentioned earlier, for ball-activated sliding-sleeve (BASS) tools, a POV would be installed for the toe-section frac, and each following ball drop would be the leading part of the wellbore displacement of each frac stage, except the final one. To catch the proper ball but allow some to pass through, each BASS tool must have a properly sized baffle, with the smallest baffle in the tool just above the POV and progressively larger baffles moving toward the heel. The smaller the lateral ID, the more limited the number of ball/baffle steps available. One of the most important concerns for a well operator using BASS tools is that they operate properly—each one opens ONLY when the proper ball is seated on its baffle—and that this always occurs at the proper stage sequence during the stimulation operations. The BASS tool-staging method for fracturing operations can even be used with a cemented liner by incorporating acidsoluble cement (Fig. 3). Figure 3 shows one well of a five-well test to compare using BASS tool completions directly to the “field standard” completion, which is to use four frac stages with a cemented liner and conventional-shape charge perforating with the perf-and-plug approach for frac-stage isolation. The method used for creating an open flowpath from the wellbore to the reservoir after a POV sleeve or a BASS tool is opened is to dissolve the cement and create a small annular void. At higher formation temperatures, the use of an organic acid (such as 10% acetic) might be preferred over mineral acid (HCl acid). This chemical reaction generally requires only minutes. The treating data, as well as the 22% cost reduction reported, is typical for all five wells within this comparison group. The 30- and 60-day production comparisons to offsets were reported by the operator as “similar to better” than perf-and-plug completions. However, this is too small a sample number to make claims for production improvement additional to the cost savings. Hopefully, this will be the subject of future reports.
Fig. 3—Illustration of completion using BASS tools with a cemented-liner completion.
Earlier discussion of liner annulus-sealing methods mentioned that they can give an operator more control of the specific placement of hydraulic fractures and possibly reduce the likelihood of multiple fractures growing simultaneously during a single frac stage. However, in many of the shale formations (and especially in the Barnett shale), multiple simultaneous fracture growth is preferred. As formation permeability increases and greater fracture conductivity is needed, limiting multiple fracs is necessary to achieve greater effective fracture lengths. This seems to be the best answer for many of the recent completions in the Haynesville shale. As is usually the case, the way to learn how to better apply hydraulic fracturing is to better understand the reservoir and how it produces. In some instances, especially for higher injection-rate fracturing treatments and when a large number of frac stages are planned, the baffles could combine to excessively increase wellhead pressure. This is of highest concern for the first few stages, where frac fluid must ball through all the baffles for Stage 1 at the toe, and all but the smallest baffle for Stage 2. Even if wellhead-pressure limits are high enough that the added pressure increases from flowing through the baffles is allowable, another limitation could exist. The pressure drop across each baffle exerts a piston force. In many of the sliding-sleeve tool
designs, the sleeve is held in its original (closed) position with shear pins to resist any movement of the sleeve until desired. The pin number and shear strength must hold the sleeve immobile until a desired force magnitude is reached, hopefully only when the proper ball has seated on the baffle of that sleeve. Each installation should be designed for the specific completion and stimulation plan, where frac-injection rates and the number of baffles needed are two of these critical factors. To review a hypothetical case (similar to an actual well application), a design spreadsheet is used that calculates the pressure drop across each baffle at a specified rate, then determines the force exerted against the pins based on each baffle’s net piston force area (total area minus flow area). Because added wellhead pressure resulting from flow through the baffles is a concern, as well as the piston force effect, a maximum delta pressure across any baffle can be specified, and also a maximum total additive pressure allowable. Fig. 4 gives an example using 700 psi as our maximum delta pressure allowed for any single baffle, with 11 potential baffle sizes shown on the x-axis (this will vary depending on the sliding sleeve manufacturer or other variables). Field experience has proven the difference in progressively larger ball diameters cannot be too small or the results become less dependable. If using a POV for Stage 1 and this specific BASS tool for the remaining stages, this example could have up to 12 frac stages as the tool-design limit.
Fig. 4—Example calculation showing pressure vs. injection rate through varying baffle openings.
Additional to considering a limit to the pressure drop through each specific baffle, operators must also design for some limit as a maximum added pressure to wellhead pressures expected if no baffles were in the frac-fluid flow paths. For this example, we are using 1,500 psi maximum added pressure from all baffles and 70 bbl/min will be used as the targeted maximum fracturing injection rate in a 4.5-in. liner completion. The data from Fig. 4 shows that only seven baffles could be incorporated in the completion design unless significantly lower injection rates can be accepted for the first stage (or more). Next, the cumulative-pressure drop must be totaled through all the baffles. The red squares in Fig. 4 can only be summed to find this answer if the 2.31- through 3.56-in. baffles are used at 70 bbl/min, although 2.06-in. baffle can be added in and still be reasonably close. In practice, it is suggested to re-run the spreadsheet using the actual maximum rate designed for each stage. For example, if one more baffle was used, the 1.81-in. size, and Stage 1 was limited to about 45 bbl/min, recalculate all baffles using this rate. When this is done, a total added pressure effect of 865 psi is calculated. As these calculations are based on an orifice-flow equation, it is also realized that fluid density and coefficient of discharge (Cd) are variables. The fluid density of 8.33 lbm/gal is shown on Fig. 4, and those data points are also based on a Cd = 0.85. Considering that density rises as proppant is added, it is necessary to recalculate allowing for the maximum concentration of proppant added. If a maximum of 4 lb of proppant is added to each gallon, the new pressure effect is 1,090 psi. Another point is that when using this higher density in the calculations, a rate below 47 bbl/min was required to not exceed the 700 psi limit through the 1.81-in. baffle. Before pumping Stage 2, a ball is seated on the first BASS tool to open its ports and isolate the Stage 1 frac. Now, the smallest baffle is the 2.06-in. size, and using the 4 lbm/gal proppant-concentration density, we can pump up to 60 bbl/min without exceeding 700 psi across it. Summing up the pressures with 60 bbl/min through the six baffles still open gives a total added pressure of 1,247 psi. After the second ball drop opens the next BASS tool and seals above the Stage 2 frac, the desired injection rate of 70 bbl/min can finally be achieved, and the added pressure at this rate from the five remaining baffles is 1,104 psi based on the 4-lbm/gal stage slurry density.
Dropping Balls When using only a few ball drops, such as in the example shown in Fig. 3, inserting the balls into the wellbore tubing at the exactly proper time, and keeping the ball-size sequence correct, can be accomplished in a “manual” method employing a dedicated pumping unit and manual valving. However, for more dependable operations, and especially when many treatments will use significantly more than two or three balls, like the seven or eight balls that would be needed in the Fig. 4 example, it is best to use a specifically designed ball-injection machine. A remote-control ball injector rigged for a capacity of up to eight balls is shown in Fig. 5. The ball injector is shown as mounted on the wellhead, and the control box used to remotely operate the ball-injection action is illustrated on the right. The installation of the specific balls can be done on location or in a workshop environment if that might better ensure the correct ball sizes are installed with the proper sequence.
Fig. 5—Left side shows wellhead-mounted ball injector; right side shows remote-operation panel for controlling ball-drop action.
Hydrajet Perforating Methods The use of hydrajetted perforations has experienced a massive rebirth in oilfield operations this century. Starting in the mid1990s, a few in the industry began to champion new, more sophisticated applications of the technology, eventually developing an entirely new conceptual application (Surjaatmadja 1998) that combined hydrajet perforating with hydrajet assisted fracturing (HJF) to provide multistage hydraulic fracturing in horizontal wells. Within a few years, hundreds of well applications had occurred (McDaniel et al. 2008). By 2002, the lessons learned and encouraging outcomes drove the desire to bring practical, cost-effective applications of hydrajetting to the vertical-well market. The resulting method uses a CT string to deploy the hydrajet perforating step as part of a multistage fracturing process using the CT-casing annulus as the flow path for the fracturing stages. The multistage frac approach using hydrajet perforating and annulus fracturing (HPAP) was the key to making the application economic for a large number of operators.Table 5 of McDaniel et al. (2008) presents an extensive listing of papers related to applications of the HPAP method. Perforating by hydrajetting instead of explosive shape charges provides extensive benefits which are probably the single best advantage of using the process. Near-wellbore pressure problems during fracturing essentially are nonexistent. Even though the process application uses very few perforations, they are large, very clustered, and nondamaged. Concerns for inflow convergence caused by the small number of holes have proven to be a nonfactor. This is probably because it is not generally understood how poor of a fracture-to-wellbore communication is normally achieved using common perforating practices. In this regard, because the process involves effectively packing around large, open perforations with wellbore sand
plugs, the chances are further enhanced that the fracture will have an open, high-quality communication path that exceeds what would likely be achieved with conventional perforating. The effects mentioned above are discussed more extensively in another recent paper (McDaniel and Surjaatmadja 2009). Additionally, in that paper there is extensive discussion related to the concern that some authors have expressed for the very small wellbore footprint typically created with the HPAP frac method, the possibility of production being significantly restricted by “inflow convergence” effects. With thousands of intervals now completed using this process, this concern is simply not supported by field evidence known to the authors. However, in McDaniel and Jurjaatmadja (2009) the authors discuss their awareness of a large number of fracturing cases in lower-perm reservoirs through conventional perforations where operators later observed serious inflow restrictions. Knowing BH Fracturing Pressure in Real Time Another valuable benefit of this HPAP method is that the coil is generally left parked less than 100 ft above the perforated zone during the fracturing operations, giving the operator a real-time bottomhole gauge to clearly evaluate what pressures are revealing about the fracturing operation. There is no need to “estimate” the bottomhole pressure when it can be measured using this method. This allows for a higher degree of confidence in knowing what formation-pressure responses tell us as to how the fracture reacts to increasing slurry concentrations. When it is desired to maximize the proppant concentration, this knowledge is vitally important. The earliest deployment approach of the HPAP method included the use of a resetable packer just below the jetting tool to allow isolation of current frac operations from previously stimulated zones. In 2003, an improved method removed the packer component and used wellbore sand plugs as the wellbore isolation method. Like any other stimulation method, it was not economic for all well applications, but its general success was very notable, and taking the process to horizontal wells was the obvious next step. By mid-2009, well over a hundred horizontal wells had been fracture stimulated with this process, even though the setting of sand plugs is far more challenging. In vertical wellbores, gravity is a great friend to this process but almost an arch-enemy in setting plugs in horizontal wells. Additionally, keeping them in place as the operator perforates and fracs more zones toward the heel increases the total challenge. In most North America deployments, the jetting BHA is designed such that recirculation up the coil is an option, and often this can enhance, or at least shorten, operations in getting effective sand plugs set. In certain applications, such as in the presence of H2S, a required check valve in the CT string will prevent this. In some offshore environments, safety rules require a check valve for many CT applications. Completion Risk Can the intended completion plan be accomplished? It is accepted that this might never be known with 100% certainty, but effort is made to understand the degree of risk for each option considered. There are general foreknown risks, and with experience using a specific completion method it is further understood that a formation or reservoir can serve to increase or lessen the importance of some of these. When the drilling of each lateral is completed, more data is available to again evaluate the risk factors, and it might even be necessary to alter the completion plan to the specific wellbore just created if potential failures or even just shortcomings in the plan can be seen. For example, many of the completion techniques discussed previously have a higher or lower tolerance for wellbore irregularities. Fig. 6 shows two examples of wellbores where operators were unsuccessful in installing the preplanned completion because of these irregularities. The left side of Fig. 6 shows a side-view plot of the drilling survey data for the lateral. Unfortunately, before attempting to install the completion, this specific plot was not generated, but instead a similar plot using fewer of the points from the survey data. Using all the survey data points for the plot clearly shows an example where the drill bit gave a “porposing motion” over a short length. The desired installation of using a liner with several external inflatable packers in conjunction with BASS-type tools could not get past this problem area. As this was only the well operator’s second application of this technology, the net result (of inexperience) was getting the installation stuck with about 60% of the completion string in the lateral (some of the string was worked past this problem point). The liner hanger was unable to seal off the annulus, so the inflatable packers could not be activated nor the POV at the toe opened. The operator pumped two frac stages with ball drops, but had no idea where the fracs might have gone. An attempted third stage screened out at a low proppant concentration. Not surprisingly, the well produced at noneconomic rates. The right side of Fig. 6 shows an overhead view of a drilled lateral. Unfortunately, this was not a view that had been plotted by the operator personnel before attempting to install the completion string. The completion string had to be pulled, dissembled, and stored for a future well application.
Fig. 6—Side-view lateral path in a well and different well with view of lateral from overhead.
Can Annular Isolation Method Fail After Successful Installation? Completion risk was briefly discussed above as it is associated to achieving satisfactory installation so the multistage fracturing operations can proceed as planned. However, ultimate success and optimum well productivity are also highly dependent on the fracturing operations performing as intended. Roundtree et al. (2008) showed how problems of the mechanical (tangential) stress from mechanical (especially) and inflatable (somewhat lower) packers that can cause fractures created from an openhole wellbore to initiate with little or no respect to the far-field preferred fracture plane. In the extreme, the authors concluded that, in some cases, such packers “… might not provide isolation because of the fracture they create longitudinally where they are placed.” This should be considered by a well operator. The authors also concluded that, “Swellable packers appear to not fracture the formation and have the chance to provide better isolation than conventional packers.” However, as was mentioned earlier, the swell packers require adequate exposure to a swell-inducing fluid and extended time (days) to reach effective levels of swelling to properly seal from the pressure differentials present during multistage fracturing operations. Other points offered by Roundtree et al. (2009): ? Tangential stress varies around the circumference of a horizontal wellbore with the minimum compressive stress occurring at the top and bottom of the hole in a normal stress environment, such as the Bakken. This will result in the induced hydraulic fracture initiating longitudinally from the wellbore. ? Virtually all conventional packers, mechanical or hydraulic (inflatable), can easily exceed the tangential stress and rock tensile strength computed at the top and bottom of a horizontal Bakken well. ? Conventional packers are used for isolation and diversion, With respect to BASS-tool applications, there is a high dependency on proper order of insertion for the graduated-size balls. In smaller-ID wellbores when larger numbers of sleeves are installed, very little clearance can be required for some balls to pass through the baffle just uphole of the intended landing location. The unexpected presence of wellbore trash can increase the chance that a ball might seat on the wrong baffle. ROI In one field test reported in an earlier paper (McDaniel et al. 2008), an operator chose several new wells in one area of a field to provide a direct comparison of the “field standard” six-stage perf-and-plug treatments (i.e., conventionally perf, frac, set bridge plug, perf, etc.) performed on several wells with offset wells completed with 13 stages using the HPAP method. Both the HPAP-with-packer-diversion method and then the HPAP-with-sand-plug-diversion (SPD) method were each used on several offsets to the “perf-and-plug” completed wells. Average total completion costs for the six-stage perf-and-plug wells were 25 to 30% lower than the 13 stage HPAP-method wells (Fig. 7). Fluid type, volumes, and total proppant were held constant on a per-well basis. However, production response was much better on the HPAP wells, with the HPAP-using-SPD wells showing 2.3-fold more cumulative production after six months. This resulted in the cost to produce a barrel oil equivalent (BOE) for the HPAP-with-SPD wells proved to be more than 50% lower, based on six-month cumulative production. All these comparative treatments used for this specific comparison were performed by the same pumping service provider, and total proppant used per well was held constant. The operator
concluded that “… statistical analysis clearly points that currently the CT technology is superior over the conventional perfand-plug technique.” The HPAP-with-SPD method became the primary completion technique used in this field development project.
Fig. 7—Charts showing comparisons for six-stage perf-and-plug completions with both methods of applying the HPAP method that used 13 frac stages.
More than one year later, the operator observed production costs per barrel to be 30% lower in the 100+ wells completed with the HPAP process than for 30+ wells using the perf-and-plug completion method during the same time frame. Had this operator chosen the completion method based on initial costs, the results would have had a very negative impact on profits. Although this comparison was in vertical wells, it is one of only a few available examples where the well operator had closely controlled data for a large number of wells to accurately judge the ROI impact of choosing a specific completion method for a specific reservoir. Much more complete comparitons were later published from work in this field (Hejl et al. 2006). Thompson et al. (2009) was referenced earlier with respect to Montney reservoir completions in a field in Canada where, during 2007 and 2008, the operator completed 13 wells using pump-down frac plugs and 19 wells using CT- deployed frac plugs and CT-deployed perforating guns. Additionally, 11 wells were completed using ECPs (two mechanical, nine with swell packers) in conjunction with BASS tools and 27 wells using the HPAP method with SPD and swell packers. This last completion method discussed in that paper became the operator’s major focus for best economics and lowest completion risk. It also included the application of swell packers to achieve annular isolation of the liner along with frac-stage isolation using the HPAP method. This process uses CT-deployed hydrajetting for perfs through which fractures will be created and wellbore sand plugs for post-frac isolation of each frac stage. In almost every implementation of the HPAP-with-SPD method, this hydrajetting application has experienced a high degree of success when sand plugs can be efficiently set. Thompson et al. 2009 discusses in detail this facet of their Montney horizontal-completion process. The Montney formation was more challenging than many because the threat of H2S in the reservoir required all CT operations to include a check valve in the string. This means that reverse circulation up the coil was never possible. This factor removed an option that often can shorten the time requirements to set effective post-frac sand plugs in horizontal completions. However, the service operator developed methods specifically in this field which allowed for consistent and dependable setting of sand plugs, even to the point that the HPAP-with-SPD completions were statistically much more trouble free than the other three completion methods used. When field production data from these four different completion types are carefully compared, it is the well operator’s conclusion that production trends appear to be more related to reservoir quality variations. Therefore, the focus of the operator has been to use well construction and completions costs and associated risks as the primary driver in selection of an optimal completion process. The completion costs and associated time for the four methods the operator employed from 2006 through 2008 are presented in Fig. 8. The “clean costs” represent the average completion cost per fractured interval with no unplanned events. The contingency costs are the added cost of unplanned events averaged over total number of fractured intervals. “Frac days” represent the average time per completion method to fracture eight intervals per well. The “best cost” was described as the lowest completion cost per interval on a per-well average in the wells studied. At this time, the HPAPwith-SPD method is demonstrating lower cost, lower cost associated with risks, and faster completion time than other methods.
Fig. 8—Comparison of completion costs per interval for four methods.
As evidenced by the data in Fig. 8, the HPAP method using sand plug isolation was the cost-effective leader. Additional benefits not directly reflected in this data are other positive factors. One benefit is that the use of the swell packer annular isolation allows for a less challenging installation of the completion string than the closest cost competitor. When using the HPAP method for multistage fracturing, the CT string stays in the hole, positioned a few hundred feet above the perforations during the proppant slurry-injection stages. By using a constant (low-rate) injection down the CT string, the surface-pressure reading on the CT is a real-time bottomhole gauge to allow instant knowledge about the pressure response the fracture is giving during each stimulation stage. As foam fluids are the operator’s choice for this formation, the other multistage frac methods provide the frac engineer very little confidence of knowing actual downhole frac pressures. Conclusions ? While cost control is imperative to any business venture, in low- to ultralow-perm reservoirs an operator must seek to maximize the volume of reservoir rock that can be stimulated with each wellbore drilled. ? For low- to ultralow-perm reservoirs in North America, well economics are often best when using long laterals and effective placement of large-volume propped fracturing treatments using methods to control the placement of multiple frac stages. ? The completion design must often include predetermined choices for how multifrac staging will be accomplished before the well is drilled. ? Drilling events or reservoir inconsistencies can create added problems for some fracture-placement control methods, requiring a change in completion design to avoid costly delays or problems that would sacrifice the stimulation of some sections along the lateral. ? The reservoir-quality variations from well to well in many low- to ultralow-perm reservoirs can make it difficult to determine if certain completion methods actually produce more hydrocarbons. ? Liner-installation risk is reduced using swell-packer isolation. ? The expandable-seal type of liner hanger lessens liner installation risk. ? Most currently used methods for fracture-placement control will either increase complexity and/or completion risk during the installation timeframe or later during the stimulation-implementation timeframe, so the operator must always combine these two aspects (and costs) when comparing effects on final well economics. ? In the planning stages, the perf-and-plug method for multiple fracture-placement control will often appear to be the option with most predictable costs and implementation consistency, but in some applications this will prove false. ? In many cases, the most cost-effective method for controlling placement of multiple fracturing stages will be the one offering the fewest operational difficulties.
Acknowledgements The authors wish to thank Halliburton for its support and permission to prepare this paper.
EnCana Corp., Oil & Gas Journal; June 15, 2009. P 34–35. Hejl, K.A., Madding, A.M., Morea, M.F., Glatz, C.W., Luna, J., Minner, W.A., Singh, T., and Stanley, G.R. 2006. Extreme Multistage Fracturing Improves Vertical Coverage and Well Performance in the Lost Hills Field. Paper SPE 101840 presented at SPE Annual Technical Conference and Exhibition held in San Antonio, TX, USA, 24–27 September. DOI: 10.2118/101840-MS. Jakobsen, E., Dreisig, H., and McDaniel, BW. 2009. Pinpoint Hydrajet Fracturing in Multilayered Sandstone Formation Completed with Slotted Liners. Paper SPE 119444 presented at the SPE Hydraulic Fracturing technology Conference held in The Woodlands, TX, 19– 21 Jan. DOI: 10.2118/119444-MS. McDaniel, B.W., Willett, R., and Underwood, P. 1999. Limited-Entry Frac Application on Long Intervals of Highly Deviated or Horizontal Wells. Paper SPE 56780 presented at the SPE Annual Technical Conference and Exhibition, Houston, Texas, 3–6 October. DOI: 10.2118/56780-MS. McDaniel, B.W., Surjaatmadja, J.B., and East, L.E. 2008. Use of Hydrajet Perforating to Improve Fracturing Success Sees global Expansion. Paper SPE 114695 presented at the COPC/SPE Gas Technology Symposium, Calgary, Alberta, Canada, 16–19 June. DOI: 10.2118/114695-MS. McDaniel, B.W. and Surjaatmadja, J.B. 2009. Hydrajetting Applications in Horizontal Completions To Improve Hydraulic Fracturing Stimulations and Improve ROI. Paper SPE 129554 presented at the SPE Eastern Regional Meeting, Charleston, WV, 23–25 September. DOI: 10.2118/129554-MS. Miller, B., Paneitz, J., Mullen, M., Meijs, R., Tunstall, M., and Garcia, L. 2008. The Successful Application of a Compartmental Completion Technique Used to Isolate Multiple Hydraulic-Fracture Treatments in Horizontal Bakken Shale Wells in North Dakota. Paper SPE 116469 presented at the SPE Annual Technical Conference and Exhibition, Denver, CO, 21–24 September. DOI: 10.2118/116469-MS. Roundtree, R, Eberhard, M., and Barree, R. 2009. Horizontal, Near-Wellbore Stress Effects on Fracture Initiation. Paper SPE 123589 presented at the SPE Rocky Mountain Petroleum Technology Conference, Denver, CO, 14–16 April. DOI: 10.2118/123589-MS. Surjaatmadja, J.B. 1998. Subterranean Formation Fracturing Methods. U.S. Patent No. 5765642. Thompson, D., Rispler, K., Stadnyk, S., Hoch, O., and McDaniel, BW. 2009. Operators Evaluate Various Stimulation Methods for Multizone Stimulation of Horizontals in Northeast British Columbia. Paper SPE 119620 presented at the SPE Hydraulic Fracturing Technology Conference held in The Woodlands, TX, 19–21 January. DOI: 10.2118/119620-MS. Vargus, G., Howell, M., Hinkie, R., Williford, J., and Bozeman, T. 2008 Completion System Allows for Intervention-less Stimulation Treatments in Horizontal Wells With Multiple Shale Pay Zones. Paper SPE 115476 presented at the SPE Annual Technical Conference and Exhibition. Denver, CO, 21–24 September. DOI: 10.2118/115476-MS. Wiley, C., Barree, B., Eberhard, M., and Lantz, T. 2004. Improved Horizontal Well Stimulations in the Bakken Formation, Williston Basin, Montana. Paper SPE 90697 presented at the SPE Annual Technical Conference and Exhibition held in Houston, TX, 26–29 September. DOI: 10.2118/90697-MS.