Gas production from the unconventional Barnett Shale reservoir now exceeds 3 Bcf/d, which is more than 5% of total U.S. dry gas production. Typically Barnett Shale wells exhibit a rapid production decline following the initial hydraulic fracture stimulation treatment, so that, within 5 years, an operator is normally faced with a well producing below its economic threshold. To keep up with current gas demand, operators have moved to an aggressive horizontal drilling and completion program. Additionally, in an effort to increase the productivity of existing wells and book additional reserves at reduced cost, operators have restimulated their older vertical wells, with demonstrable success. This success is providing compelling opportunities to enhance refracture treatment coverage by targeting bypassed and ineffectively stimulated zones in additional vertical wells and even some horizontal wells. Because of the heterogeneous nature of this unconventional gas reservoir, the restimulation of horizontal wells is problematic, and operators have demonstrated limited success using current stimulation techniques. This paper describes a new fracture diversion technique particularly adapted for horizontal well refracture stimulation. During the treatment, a fracture diversion system (FDS) is used to create a temporary bridge within the active fracture networks. That results in differential pressure increase and causes treatment redirection to understimulated intervals along the lateral. This technique enables both fracture diversion without mechanical intervention and, when enhanced with microseismic monitoring, real-time optimization of the fracturing treatment. Refracture stimulation case studies are presented in which this novel diversion technique is successfully applied to horizontal Barnett Shale wells. This paper demonstrates how real-time hydraulic fracture monitoring has enabled operators to make informed decisions that influence fracture geometry, increase lateral coverage, and improve gas recovery. To date, more than 20 fracture diversion designs have been successfully placed. The trial wells have included both cemented and uncemented completions, with drilled azimuths selected to encourage either transverse or longitudinal fracture fairway development. With a continuing optimization of the described refracturing technique, these FDS designs and placement strategies have evolved to the point where they are consistently exhibiting fracture diversion as evidenced by movement of microseismic activity and improved lateral coverage. While this engineered fracture diversion technique is ideally suited for re-fracture stimulations, it is also applicable for stimulation of new wells where the technique enables stimulation of larger wellbore intervals when used in the same fashion as for re-fracture stimulation applications. Introduction The Barnett Shale is a Mississippian-age marine shelf deposit that unconformably lies on the Ordovician-age Viola Limestone/Ellenberger group and is conformably overlain by the Pennsylvanian-age Marble Falls Limestone (Ketter et al. 2006). Formation thickness varies from 200 to 800 ft through the reservoir. The productive rock is typically a black, organic-rich shale with ultralow permeability in the range of 70 to 500 nanodarcy. To attain economically viable production rates, hydraulic fracture stimulation is a necessity.

Summary A 3D numerical model of fracture initiation from a perforated wellbore in linear elastic rock is developed, which allows one to determine the fracture-initiation pressure (FIP) and the location and direction of an initial rupture. The model assumes that the fracture initiates at the point at which the local maximal tensile stress exceeds the rock tensile strength. The 3D boundary-element method (BEM) is used for stress analysis. The model aims to predict the location of initial fractures and the difference in FIP between different perforation intervals in arbitrarily oriented noncemented wellbores. There are many practical applications for this knowledge, but of particular interest for this research is the employment of differently oriented perforations for creating heterogeneity of FIP between wellbore intervals in multistage fracturing treatment. This can enable stimulation of these intervals in a sequential mode and significantly simplify current treatment diversion and completion practices. Comprehensive analysis revealed that the main parameter that can be used for controlling FIP during multistage fracturing treatment is the angle between the direction of the perforation channel and the preferred fracture plane (PFP). The model allows obtaining the range of the angles that is the most suitable for designing and implementation of diversion between the perforated wellbore intervals. The influence of geometrical parameters of perforation (such as length, diameter, and shape) on FIP is substantially less. In addition, we found that against all expectations, increase of perforation diameter can result in higher FIP. It was also discovered that the influence of the intermediate in-situ stress on FIP is comparable with the effect of perforation misalignment, especially in the situation of a horizontal wellbore and properly aligned perforations. On the basis of the model developed, an approximate approach to the evaluation of the effect of wellbore cementation on fracture initiation was suggested. It was discovered that taking into account the state of stress within the cement before well pressurization can result in both an increase and a reduction of FIP, depending on the parameters of perforating and the wellbore orientation. The presented model is a necessary step toward predictable and controllable fracture initiation, which is vital for multistage-fracturing-treatment diversion.

A 3D numerical model of fracture initiation from a perforated wellbore in linear elastic rock is developed, which allows one to determine the fracture initiation pressure (FIP) and the location and direction of an initial rupture. The model assumes that the fracture initiates at the point where the local maximum tensile stress exceeds the rock tensile strength. The 3D boundary element method is used for stress analysis. The model is aiming at predicting the location of initial fractures and the difference in FIP between different perforation intervals in arbitrarily oriented non-cemented wellbores. There are many practical applications where this knowledge is required, but of particular interest for this research is the employment of differently oriented perforations for creating heterogeneity of FIP between wellbore intervals in multistage fracturing treatment. This can enable stimulation of these intervals in a sequential mode and significantly simplify current treatment diversion and completion practices. Comprehensive analysis revealed that the main parameter that can be used for controlling FIP during multistage fracturing treatment is the angle between the direction of the perforation channel and the preferred fracture plane. The model allows obtaining the range of the angles that is the most suitable for designing and implementation of diversion between the perforated wellbore intervals. The influence of geometrical parameters of perforation (e.g. length, diameter and shape) on FIP is substantially less. Addtionally we found that against all expectations increase of perforation diameter can result in higher FIP. It was also discovered that the influence of the intermediate in-situ stress on FIP is comparable with the effect of perforation misalignment especially in the situation of horizontal wellbore and properly aligned perforations. Based on the model developed, an approximate approach to the evaluation of the impact of wellbore cementation on fracture initiation was suggested. It was discovered that taking into account the state of stress within the cement prior to well pressurization can result in both an increase and reduction of FIP depending on the parameters of perforating as well as wellbore orientation. The presented model is the necessary step toward predictable and controllable fracture initiation, which is vital for multistage fracturing treatment diversion.