Earthquake Early Warning Systems: Methodologies, Strategies, and Future Challenges

Abstract

Earthquake early warning systems are able to issue an alert just after the start of the seismic rupture on the fault plane, but before the ground shaking reaches the sites to be protected. The available lead-time for security actions is very short (seconds to several dozens of seconds), requiring fast and robust methods to make these systems operational. Here, general principles about the two paradigms of early warning systems are presented: the regional and the on-site systems. When these methods are applied to data recorded by very dense networks, such as in near-fault observatories, the blind zone can be shortened to 15–25 km. Next generation of early warning systems should reduce the epistemic variability by implementing modeling of finite-source and site effects. Finally, the combination of on-site and regional systems may lead to a real-time estimation of the ground-shaking maps, overcoming the problem of computing the event location, and magnitude.