Lake Superior crustal structure-A reinterpretation of the 1963 Seismic Experiment

Abstract

The first-arrival data from the Lake Superior seismic experiment of 1963 have been analyzed by the method of matching structure. This method makes a proper allowance for offset and also enables one to estimate the overburden velocity from the time-distance data alone. For the upper refractor, which lies at a depth of 7–8 km and has a velocity of about 6.7 km/sec, the resulting profile agrees well with profiles previously obtained by the time-term method. But the profile for the Mohorovicic discontinuity disagrees sharply with that previously obtained. This disagreement is shown to be due to improper weighting of the data by the time-term method, which causes the derived structural relief to be nearly twice the true relief and which shifts the main structure 70 km to the west. Good wide-angle reflections from the mantle were recorded, and t2-x2 analyses yield a mean overburden velocity of 6.85 km/sec. This value should be compared with the value of 6.75 km/sec obtained by matching structure. It is emphasized that both these estimates refer to the whole of the crust, so that it is most unlikely that there is any major velocity reversal beneath the upper refractor. The mean discrepancy between depths obtained from the reflections and depths obtained from the first-arrival refraction was 1.5 km. The only other correlatable arrival possibly indicates a reflector that rises from a depth of about 25 km under the east-central part of the lake to about 15 km or less under its western edge and under the Keweenaw peninsula. The amplitudes of the first arrival from the upper refractor were consistent with its being a simple head wave, but the amplitudes from the mantle were at least ten times too large. The simplest explanation is that there is a velocity gradient in the upper mantle of about 5 × 10−3 sec−1. It is recommended that, in the interpretation of refracted-wave time-distance data, the effect of offset should always be allowed for and that mean crustal velocities should be obtained from properly designed reflection spreads. More theoretical studies are needed to determine how the amplitude of Pn is affected by the earth's curvature and by vertical velocity variations on either side of the Mohorovicic discontinuity.