Using a High- to Low-Frequency Spectral Ratio to Distinguish Variations in Earthquake Source Properties

Earthquakes of similar magnitude can generate different amounts of high-frequency radiation, contributing to variability in peak ground acceleration and uncertainties in ground-motion prediction. These source differences are expressed in their P- and S-wave spectra and affect spectral estimates of earthquake stress drop. Relative variations in spectra and related stress-drop estimates among events are better resolved than their absolute values, and thus provide useful measurements for distinguishing earthquakes and their high-frequency radiation content. Here, we quantify these differences in spectra using the frequency index (FI), defined as the logarithmic ratio of the average spectral amplitude in a high-frequency band to that in a low-frequency band. We then empirically correct for path and station effects to compute a new quantity, the normalized frequency index (nFI). We identify spatial variations in nFI values for aftershocks of the 2019 Ridgecrest earthquakes that correlate with published stress-drop estimates. Our data-driven approach can robustly identify variations in high-frequency energy among different earthquakes and source regions, providing an observation-based alternative to commonly applied stress-drop estimation methods for characterizing source behavior at high frequencies.