To investigate near-surface site effects in granite rock, we drilled 300-m-deep boreholes at two sites which are collocated with stations from the digital array at Anza, California. The first borehole was sited at station KNW (Keenwild fire station), which is located along a ridge line about 8.7 km east of the San Jacinto Fault zone. Station PFO (Piñon Flat Observatory), chosen for the second site, is another 6 km further to the east of station KNW and is located on a gently sloping hillside. We logged each borehole for P- and S-wave velocities, as well as for crack density and orientation.
P waves were generated by striking a plate with a hammer at the surface. A tool consisting of weighted anvils driven by compressed air against end plates along a 3.5-m beam was used to generate shear waves. Signals were recorded downhole with a three-component sensor package at 2.5-m intervals from the surface to 50 m depth, and at 5-m intervals from 50 m depth to the bottom of the hole. Velocities were determined by differencing the measured arrival times of first arrivals or peaks over each interval in depth. Travel times were computed for the first breaks at shallow depths, however, below about 100 m depth, times were computed for the first peaks rather than for first breaks since the first arrival was no longer clearly distinguishable. The KNW site yielded a shear velocity of 1.9 km/sec by only 30 m in depth and reached close to 2.6 km/sec at the bottom of the hole. P-wave velocities at KNW were also high at 5.4 km/sec starting at 120 m depth. The PFO site had similar but slightly higher shear-wave velocities. The bottom-hole shear-wave velocity reached 3.0 km/sec, and the P-wave velocity was 5.4 km/sec.
Shear-wave attenuation was computed using both the pulse rise time and spectral ratio methods. At station KNW, attenuation was significant only in an interval between 17.5 and approximately 40 m in depth. Over the top 50 m, attenuation corresponding to a Q of about 8 was obtained. A total T* of 0.004 sec was measured for this interval. Pulse rise times also increased rapidly in this zone. The spectral ratio data for station PFO yields two peaks in attenuation above 50 m. Similar to the attenuation found for station KNW, the peak in attenuation corresponds to a Q of about 11, averaged over the top 50 m. Spectra of the seismic pulses produced by the hammer give good signal between 20 to 80 Hz.
Significant motion perpendicular to the polarizations of the first shear-wave arrival was recorded within a few meters of the surface. Apparently, the rock structure is sufficiently complicated that body waves are being converted (SH to SV at oblique incidence) very close to the surface. The presence of these elliptical particle motions within a mere few m of the pure shear-wave source suggests that the detection of polarizations perpendicular to the main shear arrival at a single location at the surface is not, by itself, a good method for detecting shearwave splitting within the upper few tens of kilometers of the earth's crust.
Crack densities and orientations were determined from televiewer records. These records showed cracks with a preferred direction at station KNW and of a greater density than at station PFO. At station PFO, crack densities were smaller and more diffuse in orientation.