Km. Fischer et al., SEDIMENT-INDUCED AMPLIFICATION IN THE NORTHEASTERN UNITED-STATES - A CASE-STUDY IN PROVIDENCE, RHODE-ISLAND, Bulletin of the Seismological Society of America, 85(5), 1995, pp. 1388-1397
We employed ambient-noise measurements to assess the potential for sei
smic site response in sediment-filled valleys that intersect beneath d
owntown Providence, Rhode Island, At eight valley stations and at two
sites on an adjacent bedrock highland, we recorded ground motion from
two types of sources: pile drivers at a local construction site and am
bient microtremors. At all valley sites where sediment thicknesses exc
eed 10 m, spectral ratios contain amplitude peaks at frequencies of 1.
5 to 3.0 Hz. In contrast, spectral ratios from the two sites on the be
drock highland where sediment cover is less than 4-m thick are relativ
ely flat within this frequency range. A variety of borehole logs ident
ified two fundamental sediment types (soft sediment and a consolidated
glacial till) and were used to map layer thicknesses over the entire
study region. Refraction data constrained P-wave velocity in the bedro
ck to be 3680 +/- 160 m/sec and indicated two soft-sediment layers wit
h P-wave velocities of 300 +/- 50 and 1580 +/- 120 m/sec. Using a one-
dimensional reflection matrix technique, we matched the spectral-ratio
peak observed at each valley site with the frequency of fundamental r
esonance predicted for local layer thicknesses and velocities. A posit
ive correlation between the best-fitting soft-sediment velocities and
bedrock depth may reflect greater compaction in the deepest sediments
or a locally two-dimensional sediment resonance at the deepest sedimen
t sites. We conclude that unconsolidated sediment layers under downtow
n Providence have the potential to amplify earthquake ground motion at
frequencies damaging to engineered structures.