Why California Still Lacks an Earthquake Detector
Posted By Adam Brinklow on Tue, Dec 22, 2015
Earlier this month, Congress approved over $8 million in funding for the West Coast Earthquake Warning System. Once completed, the still-experimental system of sensors will provide advance warning about large earthquakes in progress and trigger a series of automated safeguards to minimize damage (slowing down trains, rerouting utilities, stopping elevators at the nearest floor, etc).
So: yay, Congress.
That's the good news.
The bad news is that this still leaves the system only halfway funded. Other countries, including Mexico, Japan, and Turkey, have had functioning earthquake warning systems for decades. And the proposal is almost 150 years old in California—in 1869, a San Francisco doctor proposed placing telegraph sensors connected to church bells along the coast. (Nobody took him up on it, but he was the first to correctly identify all of the necessary elements of an early warning apparatus.)
So what's the hold-up? Lives are at stake here.
There are two reasons we're still shuffling on this. The first is political — as in not enough Americans have died to spur politicians into action.
"The Mexico system was put in place after the 1985 Mexico City quake that killed 5,000 people," says Doug Given, the U.S. Geological Survey's early warning coordinator. "The Japanese system was put in place after the 1995 Kobe earthquake killed 6,400 people."
America has not had a "killer earthquake" large enough to spur action (not since 1906, anyway). And not even this summer's New Yorker article about the Northwest's ready-to-rupture Cascadia fault has worried Americans on the Pacific Rim enough to lead to action.
"The entire country of Japan sits on faults," says John Vidale, the University of Washington's point man for earthquake warning. "They've poured billions of dollars into their systems because they're constantly reminded that they're at enormous risk."
Our system could be fully funded on what Japan spends annually for their research program's cafeteria, but in the U.S., it's often a matter of "out of sight, out of mind."
The other problem is that America's half-funded proposal is actually much more ambitious than any existing earthquake warning system in the world. The USGS wants to stretch a network of sensors nearly 1,300 miles, from Mexico all the way up to Canada. In the future, the network may even extend beyond the United States' borders, and out into the Pacific Ocean.
The data that the system returns is meant to be much more nuanced and sophisticated than what's provided elsewhere. Japan's system — which was built in a relatively short period of time and has been something of a work in progress every since — treats an earthquake as a single point at its epicenter. But for really big quakes, the ones in the 6.0 and higher range, that's a potentially dangerous distortion, because an earthquake that size is more of a line than a dot, affecting a huge stretch of the fault far from its epicenter. You don't want your earthquake warning system mistaking lines for dots.
Software developed by Berkeley and Caltech can determine the moment when a quake's magnitude grows worrisome, and do it in something like real time. But these programs aren't perfect. They're buggy. Earlier this year, the system thought it was detecting a 5.0 earthquake in California when what it was really feeling was the furthest vibrations of a much larger earthquake overseas.
If earthquakes took hours to develop rather than seconds, experts could analyze how big they'll become and act accordingly, like we do with hurricanes. A computer can analyze earthquake data in seconds, but machines lack the intuition to know how to respond appropriately. So they have to programmed to know what to do in every possible scenario. Otherwise you end up with an automated, seismic hypochondriac prone to fake outs.
"The head of the Japanese system had to go on TV and ceremonially bow his head to the entire country to apologize for false alarms," says Thomas Heaton, director of Caltech's earthquake lab. "I don't think Jerry Brown wants to do that."
So, compared to other countries we're trying to do something much harder, over a much larger area, and with much, much less money. Naturally, this takes a while.
Once we finally hit the magical full funding mark ($16 million a year to run the system full-time), it will be about two years until it's fully implemented and sending warnings out to TVs, radios, and smart phones. In the meantime, small steps are already being taken, like the mechanism that warns BART trains about shaking in progress.
"We're under the gun to make this happen," says Heaton. Nobody wants to be halfway finished when the shaking starts, but the money only goes so far every year.
http://www.sfweekly.com/thesnitch/2015/12/22/why-california-still-lacks-an-earthquake-detector
So: yay, Congress.
That's the good news.
The bad news is that this still leaves the system only halfway funded. Other countries, including Mexico, Japan, and Turkey, have had functioning earthquake warning systems for decades. And the proposal is almost 150 years old in California—in 1869, a San Francisco doctor proposed placing telegraph sensors connected to church bells along the coast. (Nobody took him up on it, but he was the first to correctly identify all of the necessary elements of an early warning apparatus.)
So what's the hold-up? Lives are at stake here.
There are two reasons we're still shuffling on this. The first is political — as in not enough Americans have died to spur politicians into action.
"The Mexico system was put in place after the 1985 Mexico City quake that killed 5,000 people," says Doug Given, the U.S. Geological Survey's early warning coordinator. "The Japanese system was put in place after the 1995 Kobe earthquake killed 6,400 people."
America has not had a "killer earthquake" large enough to spur action (not since 1906, anyway). And not even this summer's New Yorker article about the Northwest's ready-to-rupture Cascadia fault has worried Americans on the Pacific Rim enough to lead to action.
"The entire country of Japan sits on faults," says John Vidale, the University of Washington's point man for earthquake warning. "They've poured billions of dollars into their systems because they're constantly reminded that they're at enormous risk."
Our system could be fully funded on what Japan spends annually for their research program's cafeteria, but in the U.S., it's often a matter of "out of sight, out of mind."
The other problem is that America's half-funded proposal is actually much more ambitious than any existing earthquake warning system in the world. The USGS wants to stretch a network of sensors nearly 1,300 miles, from Mexico all the way up to Canada. In the future, the network may even extend beyond the United States' borders, and out into the Pacific Ocean.
The data that the system returns is meant to be much more nuanced and sophisticated than what's provided elsewhere. Japan's system — which was built in a relatively short period of time and has been something of a work in progress every since — treats an earthquake as a single point at its epicenter. But for really big quakes, the ones in the 6.0 and higher range, that's a potentially dangerous distortion, because an earthquake that size is more of a line than a dot, affecting a huge stretch of the fault far from its epicenter. You don't want your earthquake warning system mistaking lines for dots.
Software developed by Berkeley and Caltech can determine the moment when a quake's magnitude grows worrisome, and do it in something like real time. But these programs aren't perfect. They're buggy. Earlier this year, the system thought it was detecting a 5.0 earthquake in California when what it was really feeling was the furthest vibrations of a much larger earthquake overseas.
If earthquakes took hours to develop rather than seconds, experts could analyze how big they'll become and act accordingly, like we do with hurricanes. A computer can analyze earthquake data in seconds, but machines lack the intuition to know how to respond appropriately. So they have to programmed to know what to do in every possible scenario. Otherwise you end up with an automated, seismic hypochondriac prone to fake outs.
"The head of the Japanese system had to go on TV and ceremonially bow his head to the entire country to apologize for false alarms," says Thomas Heaton, director of Caltech's earthquake lab. "I don't think Jerry Brown wants to do that."
So, compared to other countries we're trying to do something much harder, over a much larger area, and with much, much less money. Naturally, this takes a while.
Once we finally hit the magical full funding mark ($16 million a year to run the system full-time), it will be about two years until it's fully implemented and sending warnings out to TVs, radios, and smart phones. In the meantime, small steps are already being taken, like the mechanism that warns BART trains about shaking in progress.
"We're under the gun to make this happen," says Heaton. Nobody wants to be halfway finished when the shaking starts, but the money only goes so far every year.
http://www.sfweekly.com/thesnitch/2015/12/22/why-california-still-lacks-an-earthquake-detector
--
__._,_.___
No comments:
Post a Comment