Estimates of in situ stress in G Tunnel Rainier Mesa, Nevada Test Site, have been obtained with hydraulic fracturing techniques. This work represents a nontraditional use of hydraulic fracturing in that it was performed primarily in horizontal boreholes drilled into the formation from access drifts tunneled into the mesa rather than the usual operation performed at depth in vertical boreholes drilled from the surface. Several operations were performed in essentially orthogonal triads of boreholes located at a point. A significant feature of this work is the mineback operation in which the borehole is mined out to reveal the actual fracture.

During the hydraulic fracturing operation, colored dye was added to the fracturing fluid which left an easily detectable stain on the fracture surface. Direct observation of the fracture orientation away from the borehole establishes the direction of the minimum compressive in situ stress and the plane of the other two principal stresses. In one borehole, which was not aligned along a principal stress direction, a modified mineback operation revealed a fracture plane that twisted as it grew away from the borehole, aligning itself finally in a plane determined by the in situ stresse. This twisting is consistent with theoretical predictions of borehole stresses under these conditions. This same mineback also showed that fracture of the formation initiated at or under one of the packers, indicating that the packers may have an effect on in situ stress estimates. General observations of over 100 tests performed in G Tunnel show that under the sloping portion of the mesa, fracture planes are not vertical but dip in a direction tending to parallel the mesa slope. Deep into the tunnel and well under the flat part of the mesa, fractures are essentially vertical with strikes approximately N45 ¿E, which substantiates the usual vertical fracture assumption in traditional fracture operations performed at depth in vertical boreholes. Advantages, limitations, and problem areas associated with extracting in situ stress fields from hydraulic fracture pressure records are discussed in detail.