Earthquake early warning with gravitational sensors
Full-fledged, country-wide earthquake alert systems currently exist in Japan and Taiwan. A system is also under development for California, and warnings for specific cities and infrastructure are issued in other countries as well. These systems rely on the early detection of an earthquake by observing the fast, nondestructive wavefront of compressional waves with a network of seismometers, estimating the location, depth and magnitude of the earthquake with real-time updates as more data are analyzed. The earlier the warning, the more people can be warned, the more industry and processes be put on hold for safety, before the arrival of destructive seismic phases. The achievable warning time is limited by the density of the network, or generally, how close to the earthquake epicenter the seismometers are placed. With generally unpredictable epicenters, one must balance the costs and benefits of such a system.
Work by GSSI group members has led to the first predictions of prompt gravity signals from fault ruptures. These were first investigated for a Japanese GW detector prototype, TAMA. There it was found that the km-scale, laser-interferometric detectors, which are sensitive to gravity fluctuations from earthquakes above 10Hz, will not see the gravity perturbation associated with an earthquake until a fraction of a second before the waves arrive at the detector. This makes detectors like LIGO, Virgo or KAGRA useless for earthquake early warning. However, innovative technology for sub-Hz gravity-gradient observations might make it possible in the next years to detect prompt gravity perturbations from earthquakes hundreds of kilometers away from the detector [Juhel et al (2018)]. This capability would turn these instruments into potent early-warning tools since they can increase the warning time by several tens of seconds, especially in countries like Japan where the epicenters are typically out in the ocean, i.e., far from large cities.
One of the activities today, besides instrument development, is to improve the modeling of the prompt gravity perturbations. There is a complicated, nonlinear interaction between the gravity field and deformations of the ground medium. There are numerical (normal mode) simulations that can describe these effects [Juhel et al (2019)], but our theoretical understanding is very limited. Therefore, our current efforts focus on the theoretical modeling of these nonlinear interactions.
