We describe an experiment to measure variations in the state of stress within a horizontally bedded Devonian shale/sandstone/limestone sequence in western New York. A total of 75 stress measurements were made in three wells a kilometer or so apart using a wireline‐supported hydraulic‐fracturing system. The stress profiles indicate that a major drop in horizontal stress level occurs in the generally massive shales. This drop occurs principally across the lowermost member of a group of sand beds and corresponds to an offset in S h and S H of 3.5 and 9 MPa, respectively. Above the sands, “thrust” regime conditions prevail, although the amount by which S h exceeds S v is undetermined since instantaneous shut‐in pressures (ISIPs) were clipped at the level of S v due to fracture rotation. Below the sands, the regime is strike slip with both horizontal stresses showing lateral uniformity despite substantial variations in topography. The magnitude of S h in the sand beds themselves and a lower limestone remains at least as great as S v despite the decline in shale stress. Hence stress contrasts between these beds and neighboring shales become pronounced with depth. The contrast in S h and S H between the lowermost sand and the immediately underlying shale is at least 6 and 14.5 MPa, respectively. S H levels in the lower strike‐slip regime are about 1.75 times greater than S h and are less than the value required to initiate slippage on favourably oriented frictional interfaces. For the upper thrust regime and the sand and limestone beds, however, the inferred lower bound on S H is close to the slippage threshold for a Coulomb friction coefficient of 0.6. The orientation of S H is ENE with a standard deviation of 20°. Fracture traces were usually splayed, occasionally spanning 30° of well bore. No systematic correlation between mean orientation and lithology is evident. Significantly different orientations were obtained for adjacent tests in which almost identical ISlPs were observed, suggesting that the fractures quickly reorient themselves to propagate normal to the least principal stress direction. Similarly, vertical traces were observed in those tests where ISIPs apparently reflect S v , suggesting that rotation to horizontal was rapid.
No takes yet. Share an insight, caveat, or question.
Evans et al. (1989) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: