Search

An important constraint on the size and destructive potential of earthquakes is the depth extent of rupture. Laboratory studies of the transition from unstable to stable sliding, along with observed relationships between surface heat flow and the thickness of the seismogenic crust, provide strong evidence for the significance of temperature in determining the maximum nucleation depth of large earthquakes. The June 28, 1992, Mw 7.3 Landers earthquake ruptured fault segments within 20 km of 11 pre-existing heat flow measurements, and shallowing of the base of aftershock seismicity along strike correlates with an increase in heat flow. Crustal geotherms estimated from these measurements place the base of seismicity along the 250°C isotherm. This temperature is consistent with predictions from laboratory studies of the frictional stability of Westerly granite, but estimated temperatures for the seismic-aseismic transition along other faults within the San Andreas fault system are in the range of 350 to 400°C. Variations in country rock and fault gouge composition, together with higher slip rates, may account for this difference, although part of the Landers seismogenic crust might remain unruptured.

The conductivitites of selected gneiss (two) and amphibolite (one) core samples have been measured under conditions of elevated temperature and pressure with a needle-probe. Water-saturated thermal conductivity measurements spanning temperatures from 25 to 300°C and hydrostatic pressures of 0.1 and 34 MPa confirm the general decrease in conductivity with increasing temperature but deviate significantly from results reported from measurements on dry samples over the same temperature range. The thermal conductivity of water-saturated amphibolite decreases with temperature at a rate approximately 40% less than the rate for dry amphibolite, and the conductivity of water-saturated gneiss decreases at a rate approximately 20% less than the rate for dry gneiss. The available evidence points to thermal cracking as the primary cause of the more rapid decrease in dry thermal conductivity with temperature. The effects of thermal cracking were also observed in the water-saturated samples but resulted in a net decrease in room-temperature conductivity of less than 3%. These results highlight the importance of duplicating in-situ conditions when determining thermal conductivity for the deep crust.

Well log measurements of compressional and shear velocity (Vp, Vs), density, and temperature from the 4 km-deep KTB Vorbohrung (pilot hole) were applied in a phonon conduction model for the thermal conductivity of a crystalline solid. The resulting conductivity estimates were compared with conductivities (kLAB) measured on the nearly continuous (91% recovery) core. Previous studies have shown the log-derived conductivity (kLOG) to be within ±15% of kLAB in isotropic or flat-lying anisotropic crystalline rocks. The section penetrated by the KTB pilot hole includes both predominantly isotropic metabasites and highly anisotropic gneisses with foliation dips ranging from horizontal to vertical. The predictions of the phonon model were accurate within ±4% in the metabasites but inaccurate by as much as 23% in the gneisses. The accuracy of the model in the metabasites confirms the utility of the phonon conduction approach in isotropic or weakly anisotropic rocks, but the discrepancies in the anisotropic gneisses remain unexplained. These relatively large discrepancies between kLOG and kLAB correspond to depths at which laboratory measurements of Vs under in situ conditions deviate from the sonic log Vs. This suggests that sonic log determinations of Vs may not be reliable in dipping, anisotropic rocks. Alternatively, the laboratory Vs measurements may not constitute a representative sample, or there may be errors in the phonon conduction model. If the discrepancies can be tied to errors in sonic log Vs measurements, the phonon conduction model may provide a tool for deriving thermal conductivity profiles of the Earth's crust from seismic studies of Vp and Vs.

[1] Detailed thermal measurements have been acquired in the 2.2-km-deep SAFOD pilot hole, located 1.8 km west of the SAF near Parkfield, California. Heat flow from the basement section of the borehole (770 to 2160 m) is 91 mW m−2, higher than the published 74 mW m−2 average for the Parkfield area. Within the resolution of the measurements, heat flow is constant across faults that intersect the borehole, suggesting that fluid flow does not alter the conductive thermal regime. Reanalysis of regional heat flow reveals an increase in heat flow along the SAF northwest of Parkfield. This transition corresponds to a shallowing base of seismicity and a change in fault behavior near the northern terminus of the M6 1966 Parkfield earthquake rupture. The persistence of elevated heat flow in the Coast Ranges to the west appears to rule out frictional heating on the SAF as the source of the SAFOD value.

Geochemical well logs recorded in the Cajon Pass Scientific drillhole provide an estimate of the elemental concentrations of Si, Al, Fe, Ca, K, Th, Ti, S, and the HREE Gd. The continuous geochemical profiles were used to define a lithostratigraphy in the cystalline basement from 501 to 1,829 m. The interval between the faulted top of basement and a thrust fault at 689 m consists of six alternating slivers of granite and granodiorite. Interlayered granitic and granodioritic units were then encountered to another fault at 1290 m, whereas gneisses of granitic to tonalitic composition were logged in the basal portion of the hole. Linear discriminant analysis of major element chemistry, as well as variations in Gd, Th, Ti, and S, were used to demonstrate that each of the major lithostratigraphic units is chemically distinct. Different trace element signatures and variable degrees of metamorphism indicate a complex thermal and hydrological history for the basement rocks encountered in the Cajon Pass well.

Wireline logs from the Cajon Pass drillhole provide observations of crustal properties and processes near the San Andreas fault. Mineralogical profiles developed from geochemical logs are combined with a technique for calculating thermal conductivity from the volume fractions of component minerals. Variations in the equilibrium temperature gradient correlate with conductivity contrasts, and thus with lithology, rather than with fluid circulation. K, U, and Th contents calculated from gamma spectroscopy logs yield heat production values which range from 1.3 to 2.4 µW/m³, averaging 1.8 µW/m³ in the section from 500 to 1800 m. Evidence for depletion of U relative to Th raises the possibility of extensive fluid circulation in the past. However, discontinuity of depletion across faults suggests that the fluid regime responsible for U depletion predated San Andreas faulting.