Abstract Description: Incorporating distributed acoustic sensing (DAS) into the Southern California Seismic Network (SCSN) offers significant enhancements for earthquake monitoring, especially using submarine cables. DAS transforms fiber-optic telecommunication cables into dense seismic arrays, spanning tens of kilometers. Modern DAS units can interrogate multiple fiber strands, enabling broader network deployment. However, managing and processing the vast data produced, up to tens of gigabytes daily, poses a challenge. Thus, traditional seismic network operations require new workflows tailored for DAS data.
In 2024, we began integrating DAS into SCSN operations using a DAS instrument in Ridgecrest, California, where a 100-km telecommunication cable forms a 10,000-channel array. We apply the machine-learning algorithm PhaseNet-DAS to extract precise P- and S-wave arrival times from local events, leveraging modern computational architectures for real-time phase picking. Selected waveforms and phase times from 18 DAS channels are injected into existing earthquake monitoring processes, with the data archived and utilized during routine earthquake reviews. This workflow exemplifies the seamless integration of DAS data into established seismic networks, a critical advancement as more fiber strands are incorporated into earthquake monitoring, particularly in urban and submarine environments.
.jpg)
