Towards back-projection earthquake rupture imaging with ocean bottom distributed acoustic sensing data

Distributed Acoustic Sensing (DAS) along seafloor fiber optic cables offers high-density and wide-aperture seismic data close to seismic sources, at a lower cost than conventional cabled ocean bottom seismic networks. It is thus a very promising approach to develop offshore monitoring systems for hazard mitigation and to obtain deeper insights into earthquake mechanics. We introduce a workflow for back-projection earthquake rupture imaging based on ocean bottom DAS data off the Chilean coast, taking full advantage of DAS data features to greatly refine the resolution and accuracy of source parameter estimation of local earthquakes. The workflow includes a number of steps that improve the back-projection performance. To reduce the negative effects of wave scattering on waveform coherence, we apply spatial integration to convert DAS strains into displacements. We refine travel time accuracy through shallow-sediment time corrections. We apply array processing on multiple overlapping cable segments (sub-arrays) to get the apparent slowness. The information from all sub-arrays is used jointly to locate the earthquakes using a 1D local velocity model. Through systematic synthetic tests, utilizing the 120-km-long cable configuration off the coast of Chile, we identified a high-precision, high-resolution source region, which is also less sensitive to uncertainties of the velocity structure. This region extends to about 80 km laterally from the cable and reaches depths of up to 15 km, a range likely attributable to optimal signal focusing from various angles and that can be extended by increasing the cable length. We apply our method to data of roughly 50 local earthquakes with magnitudes from 1.5 to 3. We consistently obtain sharp back-projection images with high spatial accuracy, within 1 to 4 km, for earthquakes occurring within this defined region. Such precision is comparable to the location uncertainties of the seismic catalog. The true strength of our approach is its potential for imaging the rupture process of larger earthquakes. We apply our method to the synthetic waveforms of a magnitude 7 earthquake constructed from multiple empirical Green's functions. We demonstrate that strong coda waves do not compromise the precise detection and location of subsequent sub-sources, if we apply a travel time calibration. The rupture speeds and locations of sub-sources are accurately recovered, even for concurrent multiple sources. We are currently improving the calibration of travel times to increase the location accuracy and resolution. These include waveform alignment with static calibration, 3D velocity model travel time tables, and slowness bias measurements and calibrations for each source-subarray pair. Collectively, these methods will increase the resolution and accuracy of our method, along with more sophisticated back-propagation methods for individual arrays. Our work holds promise for the development of earthquake and tsunami early warning, provided that we can effectively address the issue of amplitude saturation of DAS data.