Strategies for Multiscale Seismic Monitoring in Alaska

Abstract Earthquake monitoring balances competing demands. Organizations must track small-scale seismicity visible on perhaps just a handful of sensors while assessing large earthquakes recorded globally. They must deliver authoritative information within seconds, while curating a research-grade catalog over decades. They need to remain operational precisely when communications, power, or cloud infrastructure may be degraded. These challenges are in tension and are particularly acute in Alaska. The region comprises 5%–10% of global seismicity. Station spacing varies by two orders of magnitude. Earthquake depths range from above sea level to below 200 km. A substantial fraction of seismicity occurs offshore and is essentially out of network. In late 2025, the Alaska Earthquake Center completed a ground-up rebuild of our acquisition, analysis, and product dissemination systems. We address the challenges above with a series of major design choices. Parallel on-premises and cloud-based deployments provide mostly independent paths for producing information. Detection algorithms are tuned to capture the full range of phase arrivals, whether they originate in our mission region or elsewhere in the world. Multiple association pipelines accommodate the wide range of station densities and tolerances for latent data. Together, they handle teleseismic earthquakes that spread arrivals across the region, routine in-state events, and in-state earthquakes large enough to benefit from using data from far outside our area of responsibility. Finally, we balance response time and event quality with a catalog workflow that allows the highest-quality real-time solutions to be published automatically, while lesser-quality events are held for review by the on-call seismologist, or for postprocessing during the final steps of catalog curation. Though our implementation is tailored to Alaska and remains a work in progress, our hope is that these strategies can help inform networks facing similar competing demands.