The Earthquake Network Project: Toward a Crowdsourced Smartphone-Based Earthquake Early Warning System

The Earthquake Network research project aims at implementing a global scale earthquake early warning system using smartphones as accelerometers. Software at the central server declares an earthquake when the number of smartphones reporting shaking during the past 30 s exceeds a threshold based on the expected number of reports in the absence of shaking due to earthquake waves. In this article, data collected by the smartphone network of the project are analyzed to understand whether they are informative enough to derive preliminary estimates of an earthquake epicenter. Data consist of the spatial position of the smartphones and the server receiving times. From the preliminary analysis of the smartphone data, it is clear that the signal times do not show systematic moveout as expected for seismic phases. For this reason, a new methodology that exploits the high number of smartphones in the network is developed. In particular, the epicenter is estimated by considering only the spatial locations of the smartphones that reported the earthquake and the spatial locations of the smartphones that did not. A nonparametric approach and a model-based approach are introduced and their epicenter estimation capabilities are studied by means of data from simulated earthquakes and data from 49 earthquakes detected by the system in Chile.