Search for: All records

SUMMARY We present a computational technique to model hydroacoustic waveforms from teleseismic earthquakes recorded by mid-column Mermaid floats deployed in the Pacific, taking into consideration bathymetric effects that modify seismo-acoustic conversions at the ocean bottom and acoustic wave propagation in the ocean layer, including reverberations. Our approach couples axisymmetric spectral-element simulations performed for moment-tensor earthquakes in a 1-D solid Earth to a 2-D Cartesian fluid–solid coupled spectral-element simulation that captures the conversion from displacement to acoustic pressure at an ocean-bottom interface with accurate bathymetry. We applied our workflow to 1129 seismograms for 682 earthquakes from 16 Mermaids (short for Mobile Earthquake Recording in Marine Areas by Independent Divers) owned by Princeton University that were deployed in the Southern Pacific as part of the South Pacific Plume Imaging and Modeling (SPPIM) project. We compare the modelled synthetic waveforms to the observed records in individually selected frequency bands aimed at reducing local noise levels while maximizing earthquake-generated signal content. The modelled waveforms match the observations very well, with a median correlation coefficient of 0.72, and some as high as 0.95. We compare our correlation-based traveltime measurements to measurements made on the same data set determined by automated arrival-time picking and ray- traced traveltime predictions, with the aim of opening up the use of Mermaid records for global seismic tomography via full-waveform inversion. 

Abstract Floating seismographs (Mobile Earthquake Recorder in Marine Areas by Independent Divers project “MERMAIDs”) record the data at depth at a location that is determined by linearly interpolating between the Global Positioning System positions when surfacing, assuming a constant drift velocity at depth. We study the influence of a changing drift velocity between surfacings and of a curvature of the drift trajectory. We separate localizations that directly follow a triggered ascent from those that are interpolated later. The first ones have on average a mislocation of 99 m due to curvature of the drift, against 685 m for interpolated localizations. Mislocations due to nonconstant velocity are somewhat smaller. Equivalent time errors have a distribution with heavier tails than Gaussian. The halfwidth of the 95% interval for equivalent arrival-time errors is smaller than 27 ms if the seismogram recording triggers an immediate ascent. If the recording is transmitted at a later surfacing, the interpolation is less precise with a 95% confidence interval halfwidth of 222 ms, but 67% of the errors are below 44 ms. We conclude that the localization errors have no significant impact on the accuracy of picked arrival times.