ABSTRACT
The spectral stress drop is a popular parameter for the simple quantification and characterization of an earthquake source and its expected seismic radiation, enabling investigation of earthquake spatial and temporal variability for larger numbers of events. In addition, spectral measurements are one of the few possible for earthquake characterization and hazard prediction in regions of low seismicity. However, spectral stress‐drop estimates are uncertain, especially as recorded earthquakes may be too complex to characterize ideally with a single parameter. Empirical Green’s function (EGF) approaches to isolate the earthquake source are widely regarded as one of the best for individual analysis of well‐recorded earthquakes. However, analysis decisions related to the selection of stations, EGFs, time windows, frequency bandwidth, and source models can cause discrepancies in resulting estimates of the source spectrum, source time function, and source parameters. We present results following one well‐developed EGF approach, and compare it with those from three other independent methods applied to earthquakes in the 2019 Ridgecrest, California, earthquake, sequence selected for the Southern California Earthquake Center /U.S. Geological Survey Community Stress Drop Validation Study. The common data set consists of two weeks of earthquakes from the 2019 Ridgecrest earthquake sequence, including nearly 13,000 events of M 1 and greater, recorded on stations within 100 km. We obtain estimates of corner frequency and spectral stress drop for 75 earthquakes (M 2.2–4.6) and find varying degrees of similarity among studies. We investigate four events in detail (M 2.7–4.1) and find that we obtain consistent results when the sources are relatively simple. Multiple EGFs produce good ratios and source time functions at stations with good azimuthal distribution. This suggests that there is a role for such approaches to resolve the inherent ambiguity in larger scale inversions between source scaling and attenuation and site effects.