Abstract: Laboratory hydrofracturing stress measurements were carried out in vertically fractured Niagara dolomite under a true triaxial state of stress and different preexisting fracture orientations. The ability to induce new hydraulic fractures, perpendicular to the minimum horizontal principal stress, depended on the applied-stress conditions and the preexisting fracture orientations. Hydraulic-fracture reopening pressures differed significantly from theoretically predicted values. The shut-in pressures interpreted from pressure-time curves using an exponential decay model were better than 90% accurate regardless of the preexisting fracture orientation. We conclude provisionally that when hydrofrac tests result in new hydraulic fractures, the minimum horizontal principal stress can be estimated with little or no effect from preexisting fractures. When preexisting fractures are reopened instead, the shut-in pressure yields a good approximation of the stress normal to the preexisting fracture, a result which can also be used effectively to estimate in-situ stress.
Conventional hydrofracturing stress measurements in borehole segments of intact rock zones have generally been very successful. However, the use of hydrofracturing is now being extended to ever increasing depths, active fault zones, hot rock and other hostile conditions. In fractured or faulted zones, it is sometimes difficult to find intact rock segments to facilitate testing. A major objective of an ongoing research project at University of Wisconsin - Madison is to develop the capability to measure in-situ stresses in fractured zones. Cornet and Valette (1984) studied in the laboratory the influence of flow rate on hydrofracturing of fractured granite under a uniaxial stress field. They found that at very slow flow rates (0.0016 cm3/sec) preexisting fractures tend to reopen irrespective of their orientation; on the other hand, high flow rates (0.0535 cm3/sec) resulted in new hydrofractures perpendicular to the least principal horizontal stress S h, unaffected by the preexisting fracture. Cornet and Valette also suggested a technique, which we call the Fracture Pressurization Method (FPM), which could be used to estimate the in-situ state of stress in preexisting fractures if fracture orientation and the fracture-normal stress are known. We are investigating in the laboratory the conditions that control whether new hydrofractures are induced or whether preexisting fractures are reopened in fractured rock subjected to a true triaxial state of stress and injection fluid flow rates that are comparable to those used in the field (we have used flow rates of 0.02 - 0.04 cm3/sec, equivalent to approximately 2 - 4 1/min in the field). We also seek to verify whether the results of tests in which new hydrofractures are induced can be interpreted in terms of the in-situ stresses in a manner similar to intact rock. To accomplish this we examine the correlation between the experimental shut-in and hydrofrac reopening pressures and the theoretically predicted values in terms of the in-situ stresses. When only old fractures are reopened, we examine whether the shut-in pressures are reliable measures of the fracture-normal stresses.
Niagara Dolomite was selected for this experimental study because it is linear elastic, homogeneous, isotropic, and has very low permeability.