Velocity-Porosity Relationships of Basalts from the East Pacific Rise

Abstract

Seismic experiments over young crust consistently measure velocities in the uppermost extrusives substantially lower than that of massive basalt. The differences arise from the high porosity of the rock, but the scale of that porosity (is the porosity from large fractures or from microcracks?) is unknown. To relate porosity at the hand sample scale and smaller to that at the seismic scale we performed physical-properties measurements on 17 basalts from the East Pacific Rise (EPR) near 9°3θ'N, two from Site 864 in the axial summit caldera and the remainder from dredge hauls, for comparison with detailed seismic refraction work that had been performed at Site 864 during site surveys. The elastic properties of a rock are profoundly affected by the shape of any pore space, so knowledge of the distribution of porosity over pore aspect ratio is essential to understanding the seismic velocities of the rock. We measured velocity as a function of confining pressure up to 600 MPa and from the dependence of velocity on pressure inferred a distribution of porosity over different crack shapes. We qualitatively verified our estimates of crack shape from scanning electron microscope images of the samples. The Site 864 samples have the highest porosity (3%), the lowest seismic velocities (Vp - 5.4 km/s at 50 MPa, water saturated), and the highest population of thin cracks. They contrast markedly with the dredge samples from the EPR axis, which have low porosity (<2%) and high velocities (6.0 km/s). Off-axis dredge samples have intermediate properties. The difference in properties seems unrelated to flow type: the dredge samples from the axis are from massive, sheet, and lobate flows; the two Site 864 samples are probably from a massive flow; and the off-axis dredge samples are from pillows. Instead, the key property seems to be crystal size. The Site 864 samples and the slower of the off-axis samples are all fine grained, the remainder of the off-axis samples and all the dredge rocks from the axis are microcrystalline to cryptocrystalline. In every sample, damage from microcracks is most pronounced in the largest crystals. The pattern of microcracks appears consistent with release of confining pressure from low-permeability rocks with high pore pressure; the rocks have been damaged as a consequence of being brought up from the bottom. This problem is probably common to all zero-age basalts. Our measurements in the laboratory must include the effects of any microcracks introduced by the collection process; at the seafloor the concentration of microcracks may be substantially less. Although there will be some variation with flow type, all very young basalts probably have a compressional velocity similar to our uncracked samples, about 6 km/s, in situ. Because seismic measurements of velocity of the uppermost zero-age crust are typically about 2.2 km/s, it is clear that fractures, pillow margins, and other features much larger than the scale represented by laboratory samples are of primary importance in controlling the seismic properties of the ocean bottom.