Discussion of “Variation of Site Response at the UCSB Dense Array of Portable Accelerometers” by J. H. Steidl

Abstract

In a recent paper, Steidl (1993) addresses the estimation of earthquake site response, or sediment amplification, when the motion input to the sediments is estimated from nearby bedrock observations. In particular, Steidl compares spectral-ratio estimates of site response to those based on the cross spectrum. By introducing cross-spectrum estimates as the «statistically optimum method» (p. 295) and by stating in the conclusions that «Cross-spectrum estimates should be considered more reliable [than spectral ratios] when the reference site is a single rock station» (p. 299), Steidl gives the impression that spectral ratios are inferior estimates of site response. However, previously published studies by myself and others (Field et al., 1991, 1992a) have reached different conclusions. This discrepancy is addressed in this letter, and an example is included for illustrative purposes