Analysis of regional seismograms and 3D synthetic seismograms for the 2016-01-24 Mw 7.1 Iniskin earthquake in southern Alaska

Abstract

I perform two analyses to identify cases of seismogram clipping or other problems (e.g., data gaps) for the 2016-01-24 Mw 7.1 Iniskin, Alaska, earthquake. The first analysis is a comparison of synthetic and observed seismograms: three-component, displacement seismograms filtered between periods 4-80 s. The subset of 141 stations is limited to an oblique rectangular region that is 1200 km x 600 km (Figures 1 and 2) and used in a seismic wavefield simulation with a three-dimensional seismic velocity model. I identify 60 out of 141 stations that are suspected of clipping or other problems. Of the 81 good stations, only 8 are within 250 km of the Iniskin epicenter, and all 8 stations are outside of Cook Inlet basin, which strongly amplifies ground motion (both in data and in synthetics). The second, much simpler, analysis is to identify clipping based on the maximum counts on the waveforms. The max-counts approach reveals general agreeement with the classification based on long-period data and synthetics. The analysis suggests that (1) some recorded waveforms that exceed clipping levels may still be usable for some modeling purposes, and (2) some recorded waveforms that appear to be suitable for modeling purposes should probably be discarded due to clipping at high frequencies. The identification of suspected stations, along with the waveform comparisons, may help network operators assess the possibility of unexpected performance during the Mw 7.1 slab earthquake.