Integrating Geomechanical Data Optimizes Completions Design

You have access to this full article to experience the outstanding content available to SPE members and JPT subscribers.

To ensure continued access to JPT's content, please Sign In, JOIN SPE, or Subscribe to JPT

In the process of analyzing treatment responses that occur during hydraulic fracturing, several variances in treating pressure exist that are not readily explained by examining the surface pressures and pipe friction in isolation. These variances are also apparent when looking at bottomhole injectivity. This paper demonstrates how engineers can take advantage of their most-detailed completions and geomechanical data by identifying trends arising from past detailed treatment analyses.


The Eagle Ford Shale was deposited in the Late Cretaceous Period in a marginal to open marine setting. The Lower Cretaceous part can be divided into two second-order transgressive/regressive cycles that have been labeled lower and upper Eagle Ford. The deposition of these units varies across the formation as a result of topography at the time of deposition. Therefore, operators in the Eagle Ford and other shale plays must account for changing stratigraphy and facies to locate horizontal wells properly for optimal perforation intervals.

This study is based on integrated geomechanical data, metered high-­frequency treatment data, and post-job reports. The data set used for this study corresponds to six wells completed in the Lower Bench of the Eagle Ford formation in Brazos and Burleson counties in Texas. Integration of completions, drilling, and geomechanical data has been analyzed before in the literature. Those authors observed an increase in pressure in several stages with higher Young’s modulus and lower gamma ray values, in which they were forced to pump an extra wellbore volume of clean fluid (sweep) to place the designed proppant volume. These sweeps accounted for more than 20,000 bbl of additional fluid pumped.

The wells analyzed in this study presented data showing multiple examples of difficult-to-treat stages, in which unplanned sweeps had to be pumped and screenouts occurred frequently. Because the treating-pressure anomalies were correlated with mechanical rock properties in other unconventional plays, it was suspected that similar investigation may yield savings, efficiency, and productivity improvements in the Eagle Ford.


The six wells of this study similarly consist of an approximately 7,600‑ft lateral completed with 5.5-in. ­casing and approximately 150-ft stage lengths for a total of 283 stages analyzed. The ­fracture-treatment design ­consisted of proppant ramps of 100-mesh and 40/70 white sand placed primarily with slickwater. However, the completion design of four wells included crosslinked gel to place the highest proppant concentrations ranging between 3.0 and 4.0 lbm/gal. All six wells were recorded and processed for geomechanical properties such as Young’s modulus and Poisson’s ratio; a subsequent petromechanical model; and high-­frequency ­fracture-treatment data (treating pressures, rates, and fluid and proppant volumes) and perforation data. The geomechanical data gathered along the wellbore were divided and assigned to each stage, matching the measured depths to compare variables across data sets. Then, these data sets were standardized to a common format, screened for quality control, normalized, and analyzed with a data-management application.

Understanding the distribution of the data available, the clusters within the data set, and which variables can be held constant in the analysis, was critical to start the process. For this purpose, the data set was sorted by various completion-design characteristics (perforation design, fluid type, and well section), fracture-treatment operations (screenouts, pressure builds, and equipment failure), and geomechanically derived properties. While the amount of data available is substantial, by forming a standardized data set, integration becomes a much less intimidating task.

A three-component (3C), high-­frequency accelerometer is deployed in the bottomhole assembly while drilling. The 3C accelerations resulting from the polycrystalline-diamond-compact bit cutting rock are recorded and then converted to 3C stress and strain by applying techniques used in earthquake seismology. These stress and strain relationships provide high-resolution, in-situ rock properties along the wellbore. A typical high-resolution log might contain isotropic values of Young’s modulus and Poisson’s ratio, along with vertical transverse isotropy (bedding compliance) and horizontal transverse isotropy (fracture compliance). The minimum principal stress in an isotropic medium then can be calculated.

Before the data set can be easily integrated, it must be obtained from a reliable and structured source that enables multivariate analysis. For this study, a fracturing-data-management application was used to collect high-frequency treatment data recorded at 1-second intervals while simultaneously structuring and organizing the data to be integrated with other systems.

A review of the high-frequency ­hydraulic-fracturing data revealed distinct and anomalous pressure increases that occurred during a number of stages in the similarly completed wells. For this, each stage was individually evaluated to investigate these pressure responses and how they led to changes in the pumping schedule or if they resulted in screenouts.


Through research conducted with the geomechanics and completions database, a fast analytical procedure was implemented to find substantive correlations and draw conclusions.

An initial analysis of the six wells indicated virtually no correlation between gamma ray values per stage and fracture-treating conditions. However, when evaluating high-resolution mechanical rock properties along the lateral, a much more useful correlation exists between treating pressure and minimum horizontal stress variations within a stage. This suggests that perforation efficiency may be lower (higher treating pressure) in stages with a higher standard deviation (or differential) of minimum horizontal stress across the treatment interval.

The authors hypothesize that a high differential pressure across the treatment interval could cause perforations in lower-stress rock to break down and take the majority of the treatment, with perforations in higher-stress rock potentially being left unstimulated. This scenario would cause an elevation in perforation friction and thus treating pressure. If perforation efficiency is poor, a best practice would be to use an engineered perforation design targeting similarly stressed rock within each interval.

Another metric used in this analysis is bottomhole injectivity. Because surface treating pressure alone does not account for changes in wellbore friction as the treatments progress toward the heel, estimated bottomhole pressure (BHP) was calculated using proprietary empirical data for slickwater pipe friction in 5.5‑in. casing. Average slurry rate and average BHP then were used to calculate bottomhole injectivity.

When comparing treatment responses within a well, a stage with lower injectivity tends to display a higher-than-­expected BHP, a lower-than-expected slurry rate, and a higher occurrence of screen­outs. Injectivity is driven by a complex interplay of geological and mechanical factors that may include poor perforation efficiency, near-wellbore complexity, in-situ stress, and ductility. Elimination of low-injectivity stages and their associated risks may not be feasible, but they may be mitigated by identifying patterns in the dominant controlling factors. With regard to the Eagle Ford target shown in this study, a hypothesis was made that clay-associated ductility played a significant role in injectivity and completions issues; however, measurement-while-drilling (MWD) gamma ray as a proxy for clay volume yields no correlation to injectivity.

To investigate the effect of clays and ductility on injectivity further, a petro­mechanical work flow was designed using vertical-transverse isotropy (VTI), or VTI anisotropy, calculated from the near-bit accelerations. The bedding curve was linearly transformed to a clay volume and then to brittleness on the basis of a relationship determined from cuttings analysis (Fig. 1).

Fig. 1—Brittleness vs. total clay volume obtained from cuttings analysis.


A mineralogy model was then built using the bedding-compliance-derived clay volume and a multilinear regression including Young’s modulus from 3C accelerations and MWD gamma ray for calcite. Although the brittleness estimate from X-ray diffraction was derived from a more-complex mineralogic formulation, it was critical in this instance to simplify to a single input to preserve the independence of each component of the mineralogic solution and of the contrasting mechanical quality component, strength.

Brittleness as estimated by anisotropy had a much higher correlation both to injectivity and to proppant concentration than gamma ray. A strong correlation also exists between very low brittleness and the propensity for screenout for stages with six or fewer clusters. The scatter behavior on stages with seven, eight, nine, or 10 clusters might be related to the fact that these stages have reduced cluster spacing. The closer spacing between these clusters would be expected to result in increased stress shadowing and fracture interference, which may obscure the rock mechanical effect observed in these stages.

The authors hypothesize that the greater correlation seen between anisotropy-derived brittleness and screen­outs compared with gamma ray-derived brittleness or clay volume is because the acceleration data is inherently sensitive to texture as well as mineralogy, while the gamma ray tool is driven by the atomic composition. The implication is that the laminarity of the rock, and not just the mineralogy, creates challenges for the completion.

For all six wells, in stages in which a notable pressure build is observed when a stepped increase in proppant concentration reaches the perforations, up to 7,000 bbl more total fluid and chemical were necessary in sweeps to place the designed proppant volume. Also, these stages were commonly completed with 7–10 clusters and had lower values of brittleness. However, a predictive relationship can be developed in which geomechanical data can be used to identify stages that will be problematic if completed with traditional perforating strategies. With this knowledge, those stages can be redesigned with different staging or perforating strategies to mitigate or eliminate the operational problems. At a minimum, alerting the fracturing crew and operator in advance of problematic stages will facilitate more-timely and -efficient delivery of sufficient water and chemical to place desired proppant volumes.

For a limited time, the complete paper SPE 191768 is free to SPE members.

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 191768, “Integrating Rock Properties and Fracture-Treatment Data To Optimize Completions Design,” by Eric Bruesewitz, SPE, Hawkwood Energy; Jessica Iriarte, SPE, Well Data Labs; Joel Mazza, SPE, Carrie Glaser, and Eric Marshall, Fracture ID; and Scott Brooks, Hawkwood Energy, prepared for the 2018 SPE Liquids-Rich Basins Conference—North America, Midland, Texas, USA, 5–6 September. The paper has not been peer reviewed.

Integrating Geomechanical Data Optimizes Completions Design

01 April 2019

Volume: 71 | Issue: 4



Don't miss out on the latest technology delivered to your email weekly.  Sign up for the JPT newsletter.  If you are not logged in, you will receive a confirmation email that you will need to click on to confirm you want to receive the newsletter.