Observations of hard-rock site effects

Abstract
Differences were measured in seismograms recorded at eight proximate hardrock sites: five in tunnels and three on outcrops on hills. Tunnel sites affected the amplitude of incident signals by as much as a factor of three, at frequencies inversely proportional to the depth of the site into the tunnel. For example, the maximum effect of a site 12 m deep was at 25 Hz while that of a site 85 m deep was at 11 Hz. Most of the observed tunnel effects could be explained in terms of interference between the incident and surface-reflected waves. Rock outcrop sites on hills affected the amplitude of incident signals by as much as a factor of eight, at frequencies inversely proportional to the dimension of the hill, and more for horizontally than for vertically polarized signals. Hard-rock site effects typically varied by no more than a factor of three for input signals from different earthquakes, having widely different azimuth and incident angles.