Slashdot Mirror
USGS Develops Twitter-Based Earthquake Detection
sprinkletown writes "A team of seismologists at the US Geological Survey has found that Twitter is the fastest way to get information out of an earthquake area, especially in those less densely populated. Seeing the Twitter community as an untapped resource, the USGS has developed a new way to track earthquakes by clustering quake-centric tweets."
I've used Twitter's search before (out of desperation) when my wife thought that we had an earth quake and I didn't.
To my surprise just 3 minutes later (time it took me to exhaust regular search engines), someone tweeted that they're having an earthquake a few miles away from where we live.
Since that day I've been using Twitter's search to find up to the minute updates for topics that interest me (Intel's SSD firmware bug, conferences, etc.).
I think Twitter is shaking up to be a very good source of news/information, if you can manage to find gems in the pile of "I just landed. WOOT!" tweets.
If you can't mod them join them.
Twitter users can be repurposed as sensors for vibration, voltage and even temperature!
http://michaelsmith.id.au
I always knew the whales were going to kill us.
You're a few hours late, but yes, we did have a magnitude 4.2 earthquake here in the Bay Area this morning. Of course, I'm from Cleveland, so I just assumed it was another big truck passing by the building, until I realized that I was nowhere near a major road.
I was impressed, though, at how quickly the USGS did send us all an e-mail detailing the quake, epicenter, magnitude, etc. They are certainly on the ball when it comes to the San Andreas fault, at least.
Now if we could only find a way to get advance warnings... unfortunately a time machine may be the most physically feasible method of doing that.
>> Standing on head makes smile of frown, but rest of face also upside down.
lol
-Clio
Karma: Bad (mostly from not giving a fuck)
Blog: http://clintjcl.wordpress.com
Shock waves in rock travel slower than light. So if the quake can be detected at the source (under ground) a message can be sent ahead of the shock wave to give a few seconds notice.
http://michaelsmith.id.au
Go move to a country where your government DOESN'T warn you that a bunch of people are tweeting "help! i'm dying". Maybe during the extra lag in you getting an emergency notice, your house will fall on you and do us all a favor.
-Clio
Karma: Bad (mostly from not giving a fuck)
Blog: http://clintjcl.wordpress.com
Tweeting about earthquakes is hardly new — at least in twitter years. People turned to twitter during an earthquake in Southern California in July, 2008, after they finding they were unable to make or receive any cell phone calls, they could still use twitter via SMS or another mobile twitter app.
I find this solution to be really silly. After the Northridge Earthquake in Southern California in 1994, no one in the area could even use the phones. There was too many people trying to make calls for anything to work. The earthquake they're referring to was tiny in comparison. People should be looking to a battery based radio or working with their neighbors to figure out what is going on.
Even NPR scooped Slashdot this time:
From December 14, 2009
True, though I was kind of hoping for a warning that would let me at least stay off the roads or something on Earthquake Day. I know people talk about giving enough advance warning to stop driving and pull over, but can you imagine a CA highway packed full of cars traveling at 55 MPH at somewhat-less-than-safe-distance all trying to "stop driving and safely pull over" at once? Eeeeew.
>> Standing on head makes smile of frown, but rest of face also upside down.
it's a web site and you can put small messages on it that other people can read. That's all it is and ever was.
I share your sentiment, but realize, HTTP is a method of transferring files. That's all it is and ever was.
HTML is a method of marking up content. That's all it is and ever was.
Instant Messaging is just email, but faster. That's all it is and ever was. ...until you realize that it's not the technology behind it, but how people use it, that make it what it is. When Twitter is used as a blog site, it's exactly as useless as you suggest. When it's used as a conversation, that's somewhat different.
So no, I don't use it, but I think I'm starting to get what it's about.
Don't thank God, thank a doctor!
A automated, networked, accelerometer-based quake detection process may be more reliable. Sure, a lot of mobile phones would be moving around, but if enough phones in the same areas showed synchronised movements, maybe this could work.
You could get advance warnings if you can get the information out faster than the propagation speed of the wave. I don't imagine it'd be too useful, except to possibly perform some sort of automated pre-earthquake tasks (parking hard drives, maybe?).
Twitter has also been know for false hits as well though, so how can one prevent such a situation?
This is a really bad idea because Twitter can so easily be utilized to spread misinformation. Look at all the "x is dead!!!!!!!!!!!!" shit that happens with celebrities.
When you have lived in California long enough you'll learn that:
1) Tremors move through the Earth's crust just a little bit faster than the speed of sound. To put this in perspective, it's perhaps just about 50-70% faster than a modern jet airliner cruising at high altitude.
2) If you've ever watched a modern jet airliner cruising at high altitude you know that telephone communication, radio and television, and even the internet travel much faster.
3) Following a fairly large earthquake, there will usually be a series of light to moderate aftershocks from which you can calibrate your understanding of the speed of tremors.
4) For instance, if talking across town on the telephone you will often notice that there is a substantial delay between one feeling the tremor and the other feeling the tremor.
5) The more substantial the delay between when one feels the tremor and when the other feels the tremor, the more substantial the difference in level of excitement that will be displayed. The first to feel the tremor will be the more excited party.
6) The epicenters of the aftershocks are not always in the same place, so it is possible that two observers talking on the telephone will reverse roles between (excited) and (calm) on subsequent tremors, sometimes even during a single telephone call.
7) During the period of frequent aftershocks, one of the most enjoyable things you can do is climb to the top of a small hill overlooking a large flat valley and watch the effects of the tremor propagate across the terrain. Unforgetable majesty.
8) These effects may be noted during/following a light to moderate aftershock, however, all bets are off during a large earthquake. You may wonder why a radio or television went off the air or why the power went off until you feel the tremor. But if you feel a huge earthquake first, you will lose power smartly and fail to hear when the radio station finally goes off the air.
I've been within a few thousand feet of the epicenter of four 6.0 or larger earthquakes during my adult life. Never once have I heard anyone scream during a major earthquake. I've heard people scream for 3.0 or smaller aftershocks, however. People are on edge already when aftershocks hit, and behavior changes.
Following the end of the 1994 Northridge quake, the first words out of my mouth were "Good one!", and my wife in effect said immediately "the baby's still asleep", but the extensive house repairs were not completed until four months after workmen began.
Seriously. This is the sort of thing twitter is really good at.
It's not knowing what Britney is eating for breakfast. Or how much a SKANK Malinda next door is. Or how much a bastard Billy is, oh but he's such a hunk. Or what color Aston K's turds are.
Thank goodness twitter popularity is dying.
How long before a group of trolls picks up on this and starts creating false reports?
Fuck the telephone. God damn, it's a damn piece of wire and you can say things on it that other people can hear. That's all it is and ever was. Get the fuck over it already.
Hey, I said hello by telephone! Oh yeah, well I said good morning by telephone! Oh shit, that's new and interesting because it happened on the telephone!
That reminds me of the oil light on a Lincoln Continental. It lights up, then a few seconds later, your engine seizes!
A few seconds warning sometimes just adds to the drama of the inevitable. In the relativity of time, that means if we got a few hours warning of an imminent asteroid impact, it would just be enough time maybe for a little mass-hysteria and to kiss your family good-bye. :)
I agree that a few seconds notice is definitely better than zero notice, but in all reality, a few seconds notice is not likely to measurably increase your chances of survival. It might help people avoid minor injuries such as is often caused by falling debris, etc. But it won't do much to reduce the overall number of fatalities.
If you're indoors, a few seconds is enough time to perhaps crawl under your desk, or get to the nearest doorway, but for the vast majority of earthquake-related deaths, neither of those things would have substantially increased anyone's survival odds. Most earthquake deaths are caused by buildings or other structures like bridges and freeway overpasses that suffer catastrophic collapses.
I'm all for anything that gives people any kind of advance warning of earthquakes. Such warnings will undoubtedly save many people from injury, and maybe even save a few people from death. But it will not substantially reduce the number of deaths, at least not directly. If anything, reducing the number of people with minor injuries inundating local hospitals and overwhelming local emergency services will free up those vital resources to rescue people with more serious injuries.
But to really make a dent in the number of deaths would require a major infrastructure overhaul. Namely, the quake-proofing of the thousands upon thousands of buildings and other structures that are currently vulnerable to a major earthquake. We in the developed world tend to think of high earthquake-related death tolls as being largely a third world problem, but even here in the Los Angeles area (where we really ought to know better), there are tens of thousands of buildings and structures that would suffer severe damage if LA took a direct hit from a large quake.
So, yeah... A few seconds warning is better than nothing, but just don't count on it to save your life.
Also, don't overlook the fact that not everybody is going to receive that few-second warning in time to do anything about it. Unless you (or someone in close proximity to you) happen to be glued to the USGS Twitter page, chances are the shaking will be your first warning that an earthquake is underway.
If you're handy with a soldering iron the you should build this nifty Seismic Reflector. From the website...
This project has two strands, a software and a hardware component. The aim is to build a device which responds to earthquakes being reported in near-real time via the USGS RSS feeds. The device responds by illustrating the magnitude of the reported earthquake via two fairly chunky vibration motors of the kind used in video game controllers. The device is connected to a PC via a virtual com port over USB (thanks to an on board Arduino). On the PC, an application sits there checking the RSS feed periodically and when a new event it posted to the RSS feed, the desktop app parses the data out of it and presents the magnitude of the quake to the Arduino which interpreters this as rate at which to activate the vibration motors.
Also, don't overlook the fact that not everybody is going to receive that few-second warning in time to do anything about it. Unless you (or someone in close proximity to you) happen to be glued to the USGS Twitter page, chances are the shaking will be your first warning that an earthquake is underway.
All good points, but now that mobile phones are pretty ubiquitous I would like to see a universal notification system as a mirror of the universal access to emergency calls they already provide.
http://michaelsmith.id.au
Generally useless.
The shock waves do indeed travel slower than light, but messages would take at least several seconds to be relayed, transmitted, received, and read by a human. An email in my office takes about six seconds to be routed within the building, let alone over a variety of wireless networks. A realtime message might make it out in under a second, but we're talking about shock waves that travel at 15,000-30,000mph and faster, with an damage radius on big quakes of 40 to 60 miles.
A middle case, say, 50 miles at 25k mph is about 7 seconds until the last person at risk feels it. If the message sent out is any longer than the word "earthquake!", the message generation, transmission, receipt, and processing by that guy's brain will end right at the same time the ground starts moving. And everybody closer than him will know before they get the message.
That's even assuming instant identification of an earthquake. In reality, it takes a few seconds to confirm the readings. So even that guy on the outer edge of the damage radius won't realistically get the message in time to do anything before the shaking begins.
People further out, who will feel the quake but aren't at any substantial risk of significant damage might get a few seconds' notice, but they don't really benefit from a warning in the first place, given their relatively minor inconvenience. It's like telling people on the extreme periphery as a hurricane makes landfall that it's going to start raining in a few seconds because a hurricane is making landfall somewhere else. They're not the ones who need the warning.
figures they would come up with such a horribly flawed plan. i remmeber the chief geo at a place i worked trying to explain to me how geologists weren't trained to think in uni, and how that's a good thing...
If you mod me down, I will become more powerful than you can imagine....
Sounds like a good idea, but I wonder if it might, in certain circumstances, be more harmful than helpful. What if you're cruising down the freeway at 70mph and all of a sudden your cellphone and/or radio, along with *EVERYONE* else's, starts screaming "EARTHQUAKE! EARTHQUAKE!" ? Seems like every year or two, you hear about a 100-car pile up caused by something stupid like fog or hail. I shudder to imagine what the possible unintended consequences would be if everyone's cellphone started screaming out an alarm.
And not just on the freeway, either. Having lived in the LA area all my life (37 years), I've been through my share of quakes - most small, but also a couple moderate-to-major ones. It seems like every time we have a quake of any significant size, we hear about at least one or two cases of someone dying of a heart attack (presumably due to the stress caused by the sudden violent shaking). I wonder what a sudden, synchronized blaring of cellphone alarms would do. Not to mention the general panic that would ensue. Imagine being in a crowded mall when the alarm goes off that a major earthquake is imminent. People generally don't seem to respond very well in those circumstances. Usually, there's a small fraction of people who stay calm and try to help others, while everyone else just goes ape shit. I mean, this is a country where people have been trampled to death at Walmart trying to get day-after-Thanksgiving deals.
I'm not saying a general emergency alarm system isn't a good idea. In fact, we already have one here in California, and I don't know why the USGS doesn't just tap into that instead of using Twitter - maybe it's because there are more people glued to their computers these days than to their TVs or radios. But with a device as ubiquitous as a cellphone, I have serious concerns about whether or not it would do more harm than good, ESPECIALLY when you're talking about an event such as an earthquake where you're only getting a few seconds of warning. My guess is that there would be an inverse relationship between the number of seconds of advance notice you give people, and the degree of chaos and panic it causes.
Anyway, just something to think about. I'm glad it's not my job to make such decisions.
Sorry, but that estimate is WAY off. There are several different types of seismic waves, the SLOWEST of which travels at about 3km/s, or approximately 7000mph, which is about 1400% faster than an airliner's cruising speed (assuming a cruising speed of 500mph).
I work in air traffic control, sometimes on the Human Machine Interface and I have done some UI design on that system. Even on our systems this is a finely balanced area. You need to tell the controller that he has an emergency on one of his aircraft, but not create problems for the other aircraft he is controlling at the same time. With full control of the HMI specific colours are used for Alert and Emergency states. They are used nowhere else on the UI.
Now out in the real world the people you are notifying of your emergency could as you point out be asleep, driving, eating, having sex, flying aircraft, controlling nuclear reactors, performing surgery, and so on. The expected response from those people is different in every case. In no case do you want to add to their problems. But in all cases you need to deliver a specific message.
The only answer I can bring from my ATC background is that the alert condition has to be there in some form all the time. Its not something which just appears. It is something which transitions from off to on. Which is not to say I have all the answers either.
http://michaelsmith.id.au
It is possible. Mexico City does it with the Sistema de Alerta Sísmica.
This is a good news!! This is a step in the right direction. It will help in reducing the loss of life. http://ezinearticles.com/?Force-Factor-Reviews---Do-Force-Factor-Supplements-Work?&id=2921490
I just wish the earthquake data provided by the USGS was available through a web API. XML, JSON, whatever. I poked around and there's some quake data available through various obscure programs or protocols, but nothing easy to get at. Nothing I could find, anyway. Maybe someone else knows of something more useful?
"Destroy science and religion. Science would re-emerge exactly the same; but not religion." - Penn Jillette, paraphrased
TWITTER
...
3. intr. To move tremulously, tremble, shake, quiver, shiver...
So, for an alternative article summary: the USGS will use twitterers on Twitter who are twittering about twittering.
[cue chorus of groans]
I've been thinking that if there is a really big one, we'll see a "donut" pattern when you map the data.
This mornings 4.1 (which I felt) was exciting, and tweet-worthy. The BIG ONE will not be tweeted near the epicenter. The power will go out. Even if it doesn't go out, you'll have better things to do.
Eventually the power would come back on and the hole would fill in; but I would think that the existance of the hole in the data would be one indicator of how strong the quake really is.
Has this ever been observed before in "crowdsourced" quake data?
For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares"?
I don't think that they are going to use this to announce an earthquake but to determine location and propagation parameters. Their sensors can tell them for certain IF a quake hit long before twitter will say anything but they need to correlate other measurements to get a good location estimation.
Twitter is fast and useful for that occasional use. Which is why the army of constant micro-bloggers needs to be encouraged. We can then tap into their otherwise pointless garbage about everything that happens every moment of their lives. Because when something does happen, they will tweet about that.
Please do me a favor and don't encourage micro-bloggers. The last thing I need is some dumb-ass walking into the IT field claiming to be a "professional" when the only thing on the resume is "Logged over 200,000 tweets" and "Level 233 Vampire"
Shakin' it over here Boss, Shakin' it over here!
A computer once beat me at chess, but it was no match for me at kick boxing. Emo Philips
This is exactly what I've been thinking about.
The best use case for twitter to me really sounded like machine updates. Mostly because its the only application that seemed sustainable.
A person twittering sounds great and all, until that person suddenly is too busy to update and then its worthless to any followers.
Following a machine, or maybe an organization seemed much more on target.
The best use case for twitter to me really sounded like machine updates. Mostly because its the only application that seemed sustainable.
I'm not sure what you mean by that.
A person twittering sounds great and all, until that person suddenly is too busy to update and then its worthless to any followers.
I think you just missed my point, though -- take the imaginary cocktail party again. Some people will have to go to the bathroom, some will get sick or tired and go home. That doesn't mean the conversation ends, unless you have such a sad social life that you're only talking to that one person.
Again, think of it not as microblogging, but as a slightly slower but MUCH bigger IRC.
Technologically, I actually despise Twitter. There have got to be a dozen different ways it could be done better, relying on existing standards and properly distributed. Socially, I despise the fact that people seem to have picked up on it as the Next Big Thing, much like Facebook, Myspace, blogging, or the Internet itself. Think about it -- "ExecTweets" is a real thing, and it's completely missing the point.
Oddly enough, the best demonstration of Twitter I have seen is fictional. Just remember, it's not a blog, it's a conversation -- you have to actually pick apart the threads to get some of the humor.
Don't thank God, thank a doctor!
This a really very good use of twitter.