ShakeAlert, the name given to the statewide system, is somewhat shared by the city of Los Angeles, which in early January introduced ShakeAlertLA — an earthquake early warning app that in less than a week of availability, made Apple’s Top 10 downloads list and became the most-downloaded app for Android devices. The statewide system, by contrast, has been in development since 2016, funded by $41 million over three state budget years and with ongoing annual operational costs of more than $16 million, Ryan Arba, branch chief for the California Office of Emergency Services (CalOES) seismic hazards program, told Techwire. For the 2019-2020 fiscal year that began July 1, state officials added another $60 million, much of which will likely go toward education and outreach. Among the takeaways:
• The statewide ShakeAlert system should make its debut sometime in October, Sarah Minson, a research geophysicist at the U.S. Geological Survey (USGS), told Techwire, noting that it will be expanded later to Oregon and Washington.
“I think the most important thing is the timeline, and we anticipate being able to do a public rollout in October,” said Minson, a seismologist who works on ShakeAlert. She noted it’s still being considered whether to instead begin the rollout in urban areas like San Francisco and Los Angeles “where the seismic networks are particularly dense and we feel like the performance would be best, and then rolling out further later.” Other system partners include the Berkeley Seismology Lab at the University of California, Berkeley; the California Institute of Technology; and the California Geological Survey.
Arba described the October timeframe as “a date that we’re setting as a target to evaluate our progress.” Asked whether the system was 70 percent complete, a figure Gov. Gavin Newsom has used in the last week, he confirmed the number but said it refers specifically to the insulation of the seismic instruments, a necessary step to ensure their proper function. Instrument installation, Arba said, should continue through June 2021 with sensor placement ultimately being 10 kilometers apart in urban areas and 20 kilometers apart in rural areas. Around 780 of 1,115 seismic instruments have been installed statewide, he said.
• The project is somewhat bleeding edge, Minson said, because unlike alerts for other natural disasters such as tornadoes, hurricanes and floods, earthquake alerts must be low-latency due to the speed with which they develop.
“We are really at the edge of what is technically possible, and we keep coming across things that we just don’t know the answer to. How long it takes to actually get emergency alerts from [the Federal Emergency Management Agency] to your television, your radio, your phone is not something that has been quantified before. Obviously, you are used to getting push notifications on your phone but how does that scale up? Can you do a push notification to the entire city of Los Angeles without overwhelming the network?” Minson said.
• It’s possible that vendor opportunities could come through infrastructure needs, Arba said, noting the official desire to connect the early warning system to utilities and large-scale facilities like the Aliso Canyon underground natural gas storage facility in Southern California.
State officials are working with USGS to “make it easier for vendors to tie in,” Arba said, adding: “The goal — anything where there’s a way to receive a signal and make use of it.” Examples, he said, could include slowing trains and temporarily shutting down water district pumps during a temblor.
"Certainly, there’s a desire for business in California to be able to tie in and make use for the automated actions,” Arba said. He pointed out the actual earthquake detection equipment market is small and somewhat cloistered in terms of vendors worldwide capable of producing seismic instruments, and buyers like state- and federal-level agencies that need to purchase them.
• The question of whether earthquake early warnings will continue to arrive via wireless emergency alert; or, as in the case of ShakeAlert LA, they will be delivered via third-party apps that make use of the same federal- and state-level data, has not yet been resolved.
“There has been some discussion of whether or not there should be official ShakeAlert software that gets distributed to users, or an official ShakeAlert app. At the moment, the idea has been to have just an API that’s available to everyone. And to let the whole marketplace get involved and let that competition sort (that) out,” Minson said, adding: “That doesn’t preclude us from deciding later that we really want to make sure there is at least one official ShakeAlert product out there."
Arba said the opposite viewpoint may prevail.
“There’s a desire at the moment, at least in these early stages of development, to have a level of oversight," Arba said. "The governor was pretty clear this weekend from his perspective, the idea of each county or each city having its own app may defeat the purpose, so that’s why we’re pursuing the statewide solution.”
