The effect of noise on seismograms

Abstract

Summary. We describe a method which provides an estimate of the accuracy to which time‐domain features of seismic signals can be measured in the presence of noise. Observed seismograms are simulated by adding random noise with the same frequency spectrum and signal‐to‐noise ratio to matching synthetic seismograms. The effect of noise on synthetic and observed P‐wave first motions is used as an illustration. It is shown that the apparent reliability of such observations, as determined by visual estimation, is often illusory. Copyright © 1977, Wiley Blackwell. All rights reserved