All smartphones have accelerometers, which are tiny devices that help with identifying the direction a smartphone is oriented, which allows the phone to know how one may hold it or where it is pointed. Additionally, these devices are also useful for detecting vibrations in nearby areas.
Using aggregated accelerometer data across many smartphones, many of us can now better access a large-scale earthquake warning system that informs us within seconds if an earthquake in our area has occurred and how powerful that earthquake might be.
Typically, seismometers are installed, such as in California, across a wide area and these seismometers detect movement in the earth that can then be sent as a warning signal.
Different applications have existed for some time that enabled people to get alerts on possible earthquakes. However, installing such seismometers may not be feasible everywhere.
Using accelerometers within phones, on the other hand, may be a sensible, low-cost option given that a crowdsourcing effort using large groups of smartphones can uses movement patterns and signals from accelerometers to determine earthquakes.
Google has partnered with leading seismologists so that Android phones, using the ShakeAlert system, in different countries and states could now automatically alert if an earthquake has happened. If one now searches on Google to search ‘earthquake’, at least for participating areas, they will be able to see if there is an earthquake detect in their area using aggregated (i.e., crowdsourced) accelerometers and get information about the earthquake and any warnings.
Other similar efforts have also been done, such as the MyShake application that can be installed in not only Android phones but also iOS phones. Over 300,000 people have downloaded this application and it has been demonstrated to detect an earthquake, including its magnitude and origin location, within 5-7 seconds, with alerts being set to phones within 1-5 seconds.
This application has been mainly limited to California, although there are plans to expand the effort. The application also uses accelerometers, with the Earthquake Network also being another application that uses similar technology and alert system sent to phones.
The key algorithms used to determine the presence of earthquakes applying crowdsourced data is a density-based spatial clustering of applications, with a noise spatial clustering algorithm used to also to determine the likelihood that a potentially detect event is an earthquake.
Magnitudes under 5, such as the 2018 Berkeley earthquake, can be detected relatively accurately, indicating that moderate to large earthquakes could lead to the system providing accurate information to app users.
What is unique for the MyShake application, relative to earlier efforts that used smartphones, is that the application combines different sources of data. This includes global earthquake catalogs, earthquake early warning (EEW) alerts using traditional seismic networks, and combing data using artificial neural networks artificial intelligence to determine likelihood of an earthquake to then send alerts to users.
Combining built-in sensors with smartphone data gives the application relatively high accuracy and avoid false-positive results for potential earthquakes.
Recent applications and even developments by Google are making it possible for many of us to have our smartphones act as important information nodes for early detection of earthquakes. Having a few extra seconds can make a large difference. Fortunately, smartphone signals travel faster than earthquakes, which means alerts can now be send within seconds of an earthquake, giving many people valuable time to prepare before a potentially disastrous earthquake strikes.
The next steps are to expand this coverage globally so that remote regions can now also have access to an earthquake alert system.
 For more on this new service using accelerometers on Android, see: https://blog.google/products/android/earthquake-detection-and-alerts/.
 For more on the MyShake system, see: Allen, R.M., Kong, Q., Martin-Short, R., 2020. The MyShake Platform: A Global Vision for Earthquake Early Warning. Pure and Applied Geophysics 177, 1699–1712. https://doi.org/10.1007/s00024-019-02337-7.
 For more on the algorithms used for MyShake, see: Kong, Q., Martin-Short, R., Allen, R.M., 2020. Toward Global Earthquake Early Warning with the MyShake Smartphone Seismic Network, Part 1: Simulation Platform and Detection Algorithm. Seismological Research Letters 91, 2206–2217. https://doi.org/10.1785/0220190177.
 For more on MyShake data sources, see: Strauss, J.A., Kong, Q., Pothan, S., Thompson, S., Mejia, R.F., Allen, S., Patel, S., Allen, R.M., 2020. MyShake Citizen Seismologists Help Launch Dual-Use Seismic Network in California. Frontiers in Communication 5. https://doi.org/10.3389/fcomm.2020.00032