Scientific target selection optimizes horizontal shale completions

With no control over commodity prices, the only path to profitability is through operational efficiency, optimized well productivity, and cost savings. Most operators are focused on achieving maximum operating efficiency and have had success, particularly in well construction. Technology has enabled many to minimize nonproductive time. Techniques like batch drilling and completion have shaved precious hours off field operations. And new combinations of bits, drilling fluids, and drilling parameters that have enabled drilling from shoe-to-shoe on a single bit have taken a major bite out of costs.

Time to look at targets

Suspicious that completions might not be performing to their full potential, a few operators decided it was time to take a closer look at the targeted formations. They ran a through-tubing horizontal and deviated well production logging system that measures oil, water, and gas volume fraction stage-by-stage along the completion. The results were astounding.

FIGURE 1. Production results from four horizontal multistage completions are summarized. Perforation clusters’ contribution to production are shown in the abscissa. Even the fourth well, which looks to be the best, shows 31% nonproductive stages. (Images courtesy of Schlumberger)

In well after well, the log identified anemic production from a majority of perforation clusters (Figure 1). In fact, most wells turned out to be producing more than half of total hydrocarbon flow from just a few of the stages of a multistage completion. Were the perforations missing their assigned targets? Closer examination showed the perforations penetrated the casing exactly where they were supposed to; the problem was most perforations were not flowing. The perforation clusters were simply missing the heart of the completion zone.

FIGURE 2. High anisotropic stress impairs producibility of nine of 11 zones in this well; the top two zones, shot in the lower-stress rock, are responsible for 69% of total production.

Three of four wells sampled revealed symptoms of ineffective completions. The first well showed only 19 of 36 stages to be productive, with one stage responsible for 46% of the well’s production. Another well had 15 productive stages out of a total of 27, with only two stages producing more than 4% of total flow. The third well studied had 14 of 29 stages contributing, with another five stages contributing less than 1% each. The fourth well, the best of those tested, had 20 of 29 stages contributing, but still 31% of the stages perforated were nonproductive. Clearly, more diagnosis was needed.

Improving aim

When targets are hit but no production is forthcoming, logic suggests the target itself may have been under-evaluated before the completion was designed. Recently, notable successes have been achieved when reservoir stress anisotropy was determined in advance of final completion design. Since production performance is not solely attributable to perforator performance but more likely to perforation placement relative to potentially productive formations, a study was launched to see how overall aim could be improved. The study had dual objectives: pick the most potentially productive targets and scratch potentially nonproductive ones. If achieved, completion plans would result in production improvements due to more zones contributing to flow and would save money by eliminating targets shown to be likely marginal producers.

FIGURE 3. Sonic Scanner and pulsed-neutron logs tractored into the well after it was cased provide a low-risk technique to evaluate reservoir quality and pinpoint low stress zones for completion design.

Stress imaging, sufficient to estimate with reasonable accuracy a zone’s potential, can be performed by the Sonic Scanner acoustic scanning platform. The wireline-conveyed device measures the formation axially, azimuthally, and radially to deliver a 3-D acoustic characterization that can address both intrinsic and stress-induced anisotropy. Images from the log reveal the locations, orientation, and concentration of natural fractures penetrated by the well bore. Studies have shown that the presence of natural factures greatly enhances the formation’s productive performance. Hydraulic fractures that intersect natural fracture systems typically exhibit higher productivities.

Even so, reservoir quality and completion quality are not always linked. In many cases a fairly homogeneous section will have variable rock stress. Sections with lower anisotropic stress produce much better than those with higher stress.

An example proves the point (Figure 2). Even though the bulk volume analysis (reservoir quality track) shows homogeneity, the completion quality, as predicted by the comparison of isotropic and anisotropic stress maps generated from the logs, reveals a comparatively lower stress zone in the region of 3,280 m (10,750 ft). After all stages were treated, flowmeter logs showed the upper two stages of 11 total stages to be contributing 69% of total flow. Stages treated in the higher stress zones barely contributed any flow at all. Had completion design been finalized using the stress log data, low potential zones would be identified beforehand and attention could focus on the zones with highest potential.

Confidence in stress prediction from logs is augmented by whole core geomechanical analyses. In many cases, stress analyses can be extrapolated to nearby offset wells.

In the example shown in Figure 2, an offset well whose completion was designed using stress mapping delivered the same production rate as the original well but with 50% cost reduction in completion costs due to selective perforating.

Practical considerations

FIGURE 4. Under difficult logging conditions, the ThruBit solution using slim quad combo logs provided sufficient information to complete a stress map and refocus perforation clusters to target more productive zones.

Operators naturally ask, “Science is nice, but what is the price?” It is a fair question, notwithstanding the impressive production numbers. Fortunately, there are options that can help get the desired results using a practical combination of logs and cores acquired in the vertical section of a well or extrapolated from a nearby offset well and minimal data from the long lateral section. Recent experience has shown that specific logs acquired in cased laterals at minimal risk can provide the requisite stress analyses when interpreted using formation parameters measured in the vertical borehole section.

In a single trip for one well, a pulsed neutron spectroscopy (PNS) service was run using a wireline tractor after the well had been cased. The completion model based on the PNS service matched very well with the design that would have been chosen based on the mechanical rock strength derived from the Sonic Scanner, which had been deployed in open hole using coiled tubing (Figure 3).

When parameters obtained from the vertical hole are projected onto the lateral log, a thorough stress map is obtained with zones of high, intermediate, and low stress. Selective perforation clusters targeting the red zones should produce optimal results compared with the shot-in-the dark techniques of the past.

In deviated, horizontal, or extended-reach wells, where getting to bottom with wireline logging strings may be problematic, there is a solution. A special slim-profile logging string consisting of triple-combo and sonic logs can be pumped down the drillpipe and into the openhole as a “stinger” through a ported drill bit. The logging tools acquire data and store these in memory as the drillpipe is tripped out of the hole. Logs deployed this way can be used to derive stress maps (Figure 4).

In this example from a horizontal Eagle Ford well in South Texas, the rock properties derived from the quad-combo log are used to more effectively select targets for perforation. The orange bar along the top of the log shows a typical geometrically selected stage pattern. But extreme stress anisotropy threatens stages 5, 10 to 13, and 16. The green and red bar show one possible optimization using the available data to group the stages based on similar properties. In addition, there is no need to design stages of equal lengths. Stages 1, 2, 15, and 16 have been extended to match the observed stress conditions. By avoiding zones of mixed stress and focusing on the lower stressed zones, optimized total production should be attained.

Attempting to fracture high-stress zones can contribute to screenouts when the job is pumped. A recent multiwell test was conducted to validate the conclusions drawn from this new design technique. Not only was production increased, but when the hydraulic fracture treatment was pumped, zero screenouts were experienced. The operator confirmed that costs to mitigate a premature screenout exceeded US $300,000 per incident.