Hydrogeologic constraints on heat flow along the San Andreas fault: a testing of hypotheses

Abstract

A mathematical model has been developed to study the possible effects of hydrothermal circulation on heat flow along the San Andreas fault. Extensions of laboratory rock strength studies to the state of stress on the fault predict average ambient shear stresses of approximately 50–150 MPa for the upper 15 km of the crust. These high stresses should generate a fault-centered heat flow anomaly of ∼ 42 mW/m 2 (1.0 HFU), but field studies have failed to resolve any anomaly greater than 0.2 HFU. If hydrothermal circulation can diffuse and suppress the frictional heat flow anomaly predicted by high ambient shear stress models, then the apparent contradiction between high shear stress and low heat flow might be resolved. Model results for the 6 m.y. evolution of a 15 km thick seismogenic layer, with a vertical resisting shear stress gradient of 9 MPa/km, demonstrate that the resulting convective heat transfer patterns either are insignificant relative to the conductive heat flow anomaly or tend to accentuate it. These results hold for a wide range of permeabilities (10 −19 m 2 to 10 −15 m 2 or 0.1 μD (microdarcy) to 1.0 mD), and vertical and horizontal permeability anisotropies (2 X to 20 X). If the shear zone is characterized by nearly impermeable rock (10 −24 m 2), 6 m.y. thermal pressurization within the fault zone lowers resisting shear stress and reduces the calculated anomaly by ∼ 30%. However, the effectiveness of this mechanism decreases with decreasing stress, so that the maximum heat flow anomaly allowed by field measurements develops for an average resisting stress of 15 MPa, still a factor of 5 less than laboratory studies indicate. The consistent along-strike regional elevation of the San Andreas fault zone suggests that gravity-induced flow may be a plausible mechanism for attenuating the expected heat flow anomaly. Gravity-induced flow simulations based upon average cross-fault topographic profiles for the San Francisco peninsula, Cholame Hills, and San Gabriel Mts./Mojave Desert regions show an almost complete elimination of the fault-centered anomaly. For gravity-induced flow to be effective, however, average upper crustal permeabilities must be at least 10 −17 m 2. If these permeabilities are characteristic of the rocks adjacent to the San Andreas fault, then lateral variations in topography and permeability may account for the scatter of measured heat flow values. If the average permeabilities are lower than 10 −17 m 2, gravity-induced flow cannot significantly affect heat flow along the San Andreas fault.