Based on stress-induced wellbore breakouts analyzed with an ultrasonic borehole televiewer, a detailed profile of the direction of the maximum horizontal compressive stress (SH) from 1.75 to 3.5 km depth in the Cajon Pass scientific drillhole is presented. The overall average SH, oriented N57°E, resolves onto planes parallel to the San Andreas fault with a small shear stress, whose left lateral sense is inconsistent with the slip history of the fault. Gradual and discontinuous changes in breakout orientation are abundant in highly fractured sections of the hole and seem to be caused by local perturbations of the state of stress near the borehole due to its interaction with fractures and faults. Relatively stable stress orientations in depth intervals of several hundred meters length are possibly due to the effect of nearby structures on the state of stress near the borehole. Even at great depths, breakouts often do not propagate radially far from the borehole, suggesting that inelastic deformation of the host rock may aid in breakout stabilization.
The Cajon Pass scientific drillhole is located 4.2 km north east of the San Andreas Fault (SAF) in southern California (Figure 1), where the fault strikes about N60°W. The hole is currently 3.5 km deep and has penetrated approximately 3km of granodioritic - gneissic-dioritic assemblages of basement rocks (Silver and James, 1988; L. Vernik, pers. com., 1989). The segment of the San Andreas which ruptured during the large (M~8.2) 1857 earthquake (Sieh, 1978; Sieh et al., 1989), but is now essentially aseismic, extends from Cajon Pass northwestward. The last large San Andreas earthquake to rupture through Cajon Pass was probably that of 1812, with magnitude M>7.0 (Jacoby et al., 1988; Sieh et al., 1989). Heat flow values measured over the years in the vicinity of the SAF have shown no anomaly across the fault (Henyey, 1968; Henyey and Wasserburg, 1971; Lachenbruch and Sass, 1973), which suggests a static frictional strength of no more than about 20MPa (Brune et al.; 1969; Lachenbruch and Sass, 1973, 1980). This concept of a weak fault is based on the assumptions that heat is not lost via convection of groundwater and that seismic stress drops and heat flow are the dominant energy sinks for the SAF system (Lachenbruch and Sass, 1988). In contrast to the heat flow values, frictional fault models based on laboratory friction measurements (e.g. Stesky and Brace, 1973) suggest an average value of fault frictional stress in the upper 15km along the SAF to be on the order of 70MPa (McGarr, 1980; McGarr et al., 1982; Zoback and Healy, 1984). In the context of Mohr- Coulomb failure criterion and laboratory derived coefficients of friction one would expect an angle of 30°-45° between the strike of the fault and the maximum horizontal compressive stress direction. Such a stress direction, along with high stress magnitudes at depth, would contradict the low heat flow measurements, unless the horizontal differential stress (SH - Sh) across the fault is very small.
