PNW Earthquake Early Warning prototype goes live
February 18, 2015
by Steve Malone
At 01:04 am PST this morning (Feb 18) computers that were running the new ShakeAlert-UserDisplay (and were not sleeping at the time) suddenly started squacking with an anoying sound and flashing a count-down message that "no shaking expected in 12 seconds". A computer in the home of PNSN manager, Paul Bodin was one of these and woke Paul up from a deep sleep. By the time he had gotten up and staggered to the computer terminal the "no shaking" waves from the Magnitude 4.3 earthquake near Ellensburg were passing under his house. Thus, the system really worked (unlike the implications in the Seattle Times story and KOMO TV report written by a reporter that missed the whole point about warning before strong shaking, not before the earthquake starts). However, because there are not as many seismic stations in the Ellensburg area as needed for the best warning it came later than is possible, was slightly misslocated at first and the earthquake size was underestimated by a bit. It was able to correct itself and update the warning as the count down occurred such that the real warning time was about 16 seconds. With an appropriate station distribution (such as planned for a full EEW implimentation) there could be more than 25 seconds of warning in the Seattle area for this event or a larger event at the same location.
See for yourself with this the above short video showing the event on the ShakeAlert-UserDisplay. The yellow expanding circle represents the weak P-waves used to generate an alert while the expanding red circle represents the beginning of strong shaking starting with the S-wave. In this case even the "strong-shaking" was not strong enough to be felt in Seattle but was felt in the greater Ellensburg area as reported on the "Did-you-feel-it" maps.
There is nothing like getting a real test of such a system within only hours of its official release at a workshop on Feb 17. This workshop was held at the UW for industry and civil authorities interested in using Earthquake Early Warning (EEW) for the public good. The version of EEW unveiled to the workshop participants is only a prototype version; not complete with additional features and data sources to be added and more testing needed before it can be used by the general public. To start getting feedback on how such a system might be used in a wider context in the Pacific Northwest, selected industry and governmental entities were invited to the workshop and to receive a computer application for displaying earthquake warning information in realtime. This same prototype system has been running in California for the past two years, again with only a small set of users receiving the warnings as a test. This system was adapted for use in the PNW two years ago and after testing and parameter adjusting for our region it is now available for the prototype (Beta-testing) stage.