W. F. Brace

Stick-slip as a mechanism for earthquakes.

Stick-slip often accompanies frictional sliding in laboratory experi ments with geologic materials. Shallow focus earthquakes may represent stick slip during sliding along old or newly formed faults in the earth In such a situation, observed stress drops repre sent release of a small fraction of the stress supported by the rock surround ing the earthquake focus.

Development of stress-induced microcracks in Westerly Granite

Abstract Ion-thinned samples under the scanning electron microscope reveal new details of the development of dilatant microcracks in stressed Westerly granite at 500 bars confining pressure, room temperature. Most cracks caused by laboratory-induced stress are long, straight and narrow, with sharp ends, in contrast to natural cavities, which are blunt, bridged and discontinuous. New cracks first begin to appear within about 500 bars stress of C′, at grain boundaries and healed transgranular cracks. At around 75% of the peak stress, new transgranular cracks form, starting particularly at high angle interfaces of dissimilar minerals or, rarely, at inclined pre-existing shear cracks. Peak stress coincides with kinking in suitably oriented biotite, which together with magnetite appear to have been the centers of particularly intense brittle cracking. Crack density approximately doubles up to peak stress. The scarcity of shear cracks raises doubts about the usual frictional interpretation of hysteresis in dilatancy; hysteresis could originate at interface-related cracks or at axial cracks themselves.

Permeability of crystalline and argillaceous rocks

Abstract Readily available laboratory, in situ, and inferred values of permeability, k, of crystalline and argillaceous rocks have been compared. For crystalline rocks, in situ k ranged from about 1 μd (10−14 cm2) to 100 md; for argillaceous rocks it was about 0.01 to 1 μd. No systematic decrease of k with depth was evident; over some interval at nearly every well, k was 1 to 100 md; these highly conductive intervals were as deep as 2–3 km. In situ permeability has been inferred from earthquake precursors, anomalous pore pressure, leakage from aquifers or other large-scale phenomena. Where crystalline rocks are involved, k was about 0.1 to 10 md, and thus about the same as the more permeable zones in wells; this is close to the permeability of many sandstones and is about 103 times greater than laboratory,measurements for intact crystalline rocks. For argillaceous rocks, laboratory, in situ, and inferred values all agreed within about a factor of 10. Laboratory study of artificial fractures suggest that in situ values for crystalline rocks are high because of natural fractures; fractures may be sealed or absent in shale. Based on observed variation in wells, k at particular sites in crystalline rock is not predictable within a factor of 105. For crystalline rocks, laboratory values provide little more than the minimum in situ k; for argillaceous rocks they may provide a good estimate of in situ k. Because of the great sensitivity of k to the effective stress, measurement or estimation of k must be tailored to the particular stress state of the application. If, as tentatively suggested by in situ and inferred values of k , average crustal k is about 10 md, pore pressure much greater than hydrostatic seems ruled out in terrains of outcropping crystalline rocks. Apart from hot pluton environments, anomalously high pore pressures seem to require everywhere a thick blanket of clay-rich rocks, as originally suggested for sedimentary basins.

Direct observation of microcavities in crystalline rocks

Abstract Removal of about 30 μm of material from a polished surface by ion thiining reveals microactivities in granite, diabase or gabbro as they exist in the interior of the material. A string of long, thin low aspect ratio cavities often follows grain boundaries in granites; aspect ratio of cavities ranges down to 10−3. In all rock types studied, individual low aspect ratio cavities rarely exceed about a tenth the grain size contrary to theoretical predictions. The strings of cavities could be partially healed early fractures. Nearly equant cavities up to several μm in size are very abundant in sodic plagioclase, less common in quartz and potash feldspar, and irregularly distributed along grain boundaries. They could be the sites of fluids remaining after crystallization, traces of original porosity, or the remains of healed cracks. Mechanical or thermal stress introduces sharpened brittle cracks and other more subtle damage near the ends of longer cavities. Accurate mapping of crack-like cavities in most unstressed rocks will require ion-thinned samples and the SEM.

California earthquakes: Why only shallow focus?

Frictional sliding on sawcuts and faults in laboratory samples of granite and gabbro is markedly temperature-dependent. At pressures from 1 to 5 kilobars, stick-slip gave way to stable sliding as temperature was increased from 200 to 500 degrees Celsius. Increased temperature with depth could thus cause the abrupt disappearance of earthquakes noted at shallow depths in California.

Cracks and Pores: A Closer Look

Most pores and some cracks in several rocks, as directly viewed with a new technique, have a shape that suggests an origin early in the history of these rocks. Thus, behavior in the laboratory may be a reliable indication of behavior in the earths crust, for electrical resistivity, permeability, or other properties that depend on microporosity.

An Experimental Study of Tectonic Overpressure in Franciscan Rocks

Approximately 4 kb tectonic overpressure was required, according to one theory, for the formation of jadeite-aragonite-bearing rocks of the Franciscan. Strength of two common Franciscan rocks—massive graywacke, and thin-bedded shale and graywacke—was determined to see if this tectonic overpressure could have been generated under the conditions of metamorphism; that is, 4 kb confining pressure, 200 to 300° C, in the presence of aqueous pore fluid. The strength was found to depend markedly on both pore and total pressures. The required tectonic overpressure could only have been generated in massive graywacke, and only when pore pressure was less than hydrostatic. With even moderate amounts of interbedded shale, tectonic overpressure could not have exceeded 1 kb.

The effect of speciment size on the mechanical properties of unjointed diorite

Abstract Mechanical properties of unjointed in situ and laboratory specimens of quartz diorite and granodiorite covering a wide range of specimen sizes were determined under static uniaxial loading conditions. A drill and broach technique was developed to prepare the in situ specimens. In situ specimens were right triangular prisms ranging in length from 1 to 9 ft loaded by a flat-jack system. The laboratory samples consisted of triangular prisms from 4·5 to 12 in. in length and cylindrical specimens 3·18 and 4·25 in. in length. Specimens had a length-to-edge or length-to-diameter ratio of 1·5:1·0 or greater. Maximum stress for the quartz diorite decreased with increasing specimen size by a factor of 10, ranging from 10,000 psi for a 2·0 in. specimen to 990 psi for the 9 ft specimen. Maximum stress asymptotically approached a value of 1000 psi for quartz diorite specimens greater than 3·0 ft in length. All specimens failed in shear and exhibited dilation prior to failure. For the quartz diorite, apparent Youngs modulus ranged from 0·322 to 0·645 × 106 psi, apparent Poissons ratio ranged from 0·10 to 0·33. Neither of these properties correlated well with specimen size. The maximum stress for small specimens of granodiorite averaged 27,700 psi; modulus for both in situ and laboratory samples ranged from 2·27 to 8·33 × 106 psi.

Elastic and transport properties of an in situ jointed granite

Abstract In situ elastic and transport properties were measured as a function of compressive stress to 200 bars for a variety of load psths on a 3-m cube of jointed granite near Laramie, Wyoming. The specimen contained 3 vertical joints which are parallel to a set of well-developed microfractures. Measurements were made parallel and normal to the joints across the entire block and within intact areas containing only microfractures. The loads were applied by eight 1.2 × 2.4 m flatjacks, 2 on each of the 4 sides. The measured properties included deformation, compressional velocity, electrical resistivity and fluid permeability. Load paths included uniaxial, biaxial and proportional stress and uniaxial ‘strain’. ‘Direct shear’ tests were also conducted at 2 normal stresses. Field data were compared with measurements made on oriented specimens in the laboratory. Results based on differences in rate of change of fluid flow, elastic moduli and on seismic wave propagation indicate that the joints close at 15–30 bars normal stress, but that microcracks remain open at the highest stresses attained. Changes in fluid permeability along the joint and elastic modulus with stress are greater than those of either seismic velocity or electrical resistivity. Modulus and velocity increased with stress while permeability decreased by a factor of 4 . Even when ‘closed’ permeability along the joint is 3 orders of magnitude greater than for intact granite. Intact granite exhibits a marked anisotropy with respect to both modulus and seismic velocity.

Effect of pressure on electric-resistance strain gages

A simple theory is developed to explain the effect of hydrostatic pressure on electric-resistance strain gages. For relatively incompressible materials,pressure effect, the algebraic difference between true and observed strain, is (α/GF +βG) where α is pressure coefficient of resistance,βG is the linear compressibility, andGF the gage factor of the gage material. For Constantan foil gages, pressure effect should be about +0.70×10−7 bar−1, which is approximately what is observed by Milligan and Gardeen and in experiments described here on materials under pressure of up to 10,000 bars.