INTRODUCTION
Hydraulic fracturing was applied in horizontal drill-holes in the Salado salt formation near Carlsbad, New Mexico. Tests were performed approximately 650 m below surface to validate the design of a Waste Isolation Pilot Plant (WlPP) for the disposal of radioactive waste from defense activities of the United States. Hydraulic fracturing was performed primarily to determine whether the virgin in situ stress state at the WlPP site is isotropic and whether the magnitude of the virgin in situ stresses corresponds to the weight of the overburden. Beyond these limited objectives, however, measurements are now being reviewed to evaluate the usefulness of hydraulic fracturing in salt formations in general. Such measurements are desirable to determine stresses induced by mining and to monitor time-dependent stress changes around underground excavations in salt masses. Hydraulic fracturing was chosen because it is the only technique that can be readily applied in virgin ground far from existing mine openings. The method also appeared to yield directly the magnitude of the least compressive principal stress, regardless of the mechanical properties of the rock mass. A knowledge of the least principal compression was deemed sufficient to establish the isotropic character of the virgin in situ stress field under two conditions: (1) the Salado formation was mechanically isotropic, i.e., the fracture toughness did not vary normal and parallel to bedding so as to permit preferred fracture paths, and (2) pressure measurements were combined with determinations of the orientations of hydraulic fractures and, more generally, with observations of the characteristic hydraulic fracture patterns. The latter was possible by locating the drillholes for hydraulic fracturing along the axes of drifts that were mined several months later.
The success of general stress measurements by hydraulic fracturing of salt subjected to anisotropic stress conditions was and remains less certain. While the smallest in situ compressive stresses should still be equal instantaneous shut-in pressures, there may not exist any unique relationships between the greatest and intermediate in situ principal stresses on one hand and the primary breakdown and fracture reopening pressures on other. This problem would be a consequence of stress laxation due to salt creep around any drillhole in which hydraulic fracturing tests are conducted. Moreover, laboratory measurements suggest that creep and stress laxation could be so rapid that it might not be possible interpret the measurements for anisotropic stress conditions even if hydraulic fracturing were performed as soon after drilling as possible. The primary objectives of the stress measurements the WlPP were met by combining hydraulic fracturing tests with finite element analyses of the effects of salt creep, observations of hydraulically induced fracture patterns, and laboratory hydraulic fracturing tests on oriented core. In situ hydraulic fracturing tests in long drillholes were preceded by several trial tests to establish the best test methods and to demonstrate that salt could be fractured. The resolution of anisotropic stresses and special problems, e.g., in the identification of instantaneous shut-in pressures, are currently being addressed in hydraulic fracturing tests in a mine pillar, in tests lowing different delays between drilling and fracturing, and in tests with different frac fluids.
