US Geological Survey


Document Type


Date of this Version



Published in JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 97, NO. B4, PAGES 4991-4994, APRIL 10, 1992


Over the past few decades a generally accepted conceptual model for the thermomechanics of the San Andreas fault was based on three factors believed to have been supported by observation: (1) maximum horizontal shear stress increases with depth in the crust at the rate of 7-8 MPa/km (as indicated by measurements in relatively shallow boreholes), roughly the same rate of shear stress increase as in other regions [e.g., McGarr et al., 1982], (2) the maximum horizontal stress direction in the vicinity of the fault was oriented at about 300-45 ° to its strike (as indicated by the predominance of right-lateral strike-slip earthquakes along it), and (3) there is no detectable frictional heat generated by the San Andreas (as implied by heat flow observations in numerous shallow boreholes both near and far from the fault along the length of the San Andreas system). Assuming that laboratory- derived coefficients of friction in the range 0.6-1.0 [e.g., Byeflee, 1978] are applicable to all faults in the crust (sometimes referred to as "Byedee's law" [after Brace and Kohlstedt, 1980]) and that approximately hydrostatic fluid pressure exists at depth, it is straightforward to show that while the second observation is related directly to the first, the stress state they imply at depth would result in a conspicuous heat flow anomaly; a direct contradiction of observation 3. This contradiction has come to be known as the San Andreas stress/heat flow paradox.