Enhanced In-Situ Stress Profiling With Microfracture, Core, and Sonic-Logging Data

Abstract

Summary. Knowledge of in-situ stress distribution, especially in the vertical direction, is vital to hydraulic-fracture geometry calculations. The microfracturing technique is recognized as the best method to measure in-situ stress directly. The technique, however, is typically limited to very few measurements and, at times, it is impractical to break down and measure in-situ stress in bounding nonproducing layers. Also, in layers with significant stress variation, microfracture measurements can be reflective of an average value of the variations and thus can be misleading. Alternative techniques like core measurements can suffer from depth discrepancies and lack of measurements at in-situ conditions. Sonic-logging methods can provide continuous in-situ measurements of rock mechanical properties; however, modeling stresses from mechanical properties have certain limitations. In this paper, we present a method properties have certain limitations. In this paper, we present a method that can compensate for the disadvantages of the individual techniques with the advantages of the others and thereby create relationships that can facilitate obtaining representative and continuous in-situ stress data. Data from two wells are presented to illustrate the method's application. Introduction With the decrease in oil price, it has become important to optimize stimulation treatments to maintain profitability. One critical aspect of a hydraulic-fracture treatment design is the prediction of the created fracture geometry. Various fracture-geometry-prediction models are available to the industry. Proper use of these models requires the knowledge of continuous in-situ stress and mechanical properties of the producing formation and bounding nonproducing layers. producing formation and bounding nonproducing layers. The microfracturing technique is considered the best method to measure in-situ stress directly. The technique, however. has certain limitations. At times, it may be impractical to break down and measure in-situ stress in bounding nonproducing layers. In cased holes, this technique depends on good bond between the cement and casing and the formation. Microfracture measured in-situ in a well without adequate cement bond may represent an adjacent layer that was in hydraulic communication with the perforated interval. Use of the technique over intervals of varying st can also lead to inconclusive results. When injected volumes are increased to the point where the fracture migrates into adjacent layers of varying stress, the computed in-situ stress value may no longer represent the originally perforated layer. Alternative techniques include core testing and well logging. Core testing provides stress measurements through stress/strain relaxation methods. Under ideal core-recovery and gauge-resolution conditions, the calculated values can be within an acceptable range. Also, core-measured Poisson's ratios (PR's) (both static and dynamic) can be used with Poisson's ratios (PR's) (both static and dynamic) can be used with elastic-modeling efforts to estimate stress. The major limitations of the core measurement include a lack of measurements at in-situ conditions, the microscopic nature of the core sampling, and a lack of core samples from neighboring nonproducing boundary layers. Sonic-logging methods, measuring compressional- and shear-wave travel time, can be very attractive because of the continuity of the measurement. The technique suffers, however, from the limitation associated with modeling stresses with PR. This limitation also applies to core-derived stress values. To address the above limitations, we present a method that compensates the disadvantages of the individual techniques with the advantages of the others to create relationships that facilitate obtaining representative continuous downhole in-situ stress data. Data from two wells are presented to illustrate the method's application. Methodology Researchers and, more recently, works funded through the Gas Research Inst. (GRI) have shown a strong need to establish criteria that allow one to integrate sonic-logging data with core and microfracturing data to estimate stresses. Refs. 12 and 13 have attempted to present empirical relationships between core- and log-derived data. In our study, we attempt to integrate microfracture, logging, and core data into a method that compensates for the disadvantages of individual techniques with the advantages of others. Such a procedure has allowed us to obtain representative and continuous downhole in-situ-stress data. Briefly, the main aspects of the method are to (1) use the sonic-log-derived and core-measured PR's to select appropriate microfracture-test locations and to help analyze microfracture insitu-stress data; (2) to use available core PR and rock mineral information to perform a regression analysis to calibrate the log-derived PR and in-situ stress; and (3) to compare the continuous stress derived from these core-calibrated data to microfracture information to determine whether further calibration of the microfracture data is necessary. The following sections describe the method in more detail. Use of Sonic Logs and Core Date for Enhanced Microfracture Tests One of the biggest difficulties in establishing a correlation between microfracture-, log-, and core-derived stress data results from irregularities in the depth reference. Microfracture stress may not reflect the same horizon to which the log or core measurement refers. This problem also arises when deciding where to perform a microfracture test. Thus, efforts should be directed to cross-correlate carefully between the various measurements. In a microfracture test, injected volume of fluid typically is limited to about 10 to 20 gal per stage. In such a case, the created fracture is usually confined within several feet around the perforated interval. A significant majority of commercial microfractures, however, are performed with tens to hundreds of gallons of injected fluid. Under such circumstances, it is likely that the created fracture may migrate into adjoining layers or zones. If these layers or zones are at a stress level different from that of the intended zone, then the stress estimation can be significantly influenced. Therefore, extra care should be taken in the design and evaluation of microfracture tests from layers and zones with varying stress levels. During the design phase, the microfracture-test-zone selection should carefully consider how PR varies from the sonic logs and/or available core data. Experience has shown that microfracture-test-zone selection should be such that the PR variation within the interval results in less than 100-psi variation in in-situ stress. Otherwise, the resulting pressure-vs.-time record following test shut-in may not clearly demonstrate the instantaneous shut-in pressure (ISIP). SPEFE P. 243