As early as 2020, Google launched a project toCrowdsourcing signals from Android phones to predict coming earthquakes. The system uses accelerometers within these devices to collect seismic data, combines this data to detect patterns consistent with earthquakes, and issues timely alerts to people who may be affected.Three years later, researchers involved in the projectA paper was published this week in the journal Science, showed that their earthquake detection technology works well - demonstrating its usefulness in places where there are no seismic stations to study and issue local warnings.
The system works by detecting sudden increases in the phone's acceleration that are consistent with ground movements during earthquakes. Once triggered, the phone sends a signal to a central server, which listens for similar signals from other phones in the same area and at the same time. When the observational data reaches a high enough level of confidence, an earthquake is officially declared.
The system is deployed in 98 countries/regions and has detected an average of 312 earthquakes per month in three years of operation, with magnitudes ranging from a low of 1.9 to a high of 7.8. According to user feedback, 85% of users who received the alert felt shaken, and as many as 36% of the respondents said they received the alert before the earthquake occurred.
Not bad for a system that doesn't require a lot of local infrastructure to keep people safe. The researchers have deployed the system to 2.5 million devices through software that comes preinstalled on all Android phones, meaning it is turned on by default and can be used without user input.
This accelerometer-based crowdsourced system is a more modern version of 2015's ShakeAlert system, in which "GPS receivers in smartphones can detect permanent ground motions (displacements) caused by fault movement in large earthquakes."
The technology is also connected to the Android Earthquake Alert system, which can send alerts directly to mobile phones in areas where seismologists expect earthquakes to occur, with instructions on how to take shelter.
This type of public infrastructure can not only help us understand more large earthquakes and prevent large-scale disasters, but also provide comprehensive observation data of multiple earthquakes to provide reference for regional models. Built-in feedback systems that people can use to determine if they feel an earthquake could also enhance this data and potentially give us a better chance of spotting earthquakes before they happen.
Source: Science