Google Analytics

Tuesday, August 23, 2011

Predicting earthquake effects with twitter

Twitter follower Brian D'Amico asked me how long it took for today's earthquake to reach us here in New England, and it prompted me to do a little math.

Per the USGS, seismic waves travel at anywhere from 2-8 kilometers per second. From the epicenter of the earthquake to Boston is roughly 853 km, give or take 30 or 40 kilometers. This means it would take the somewhere between 1 minute 46 seconds and 7 minutes for the earthquake to be felt in Boston.

USGS recorded the earthquake happening at 1:51 PM local time. The first tweet I received regarding the earthquake was at 1:57, by @gnimsh. Assuming twitters clocks and the USGS's clocks are pretty close, and that @gnimsh tweeted me right away, this definitely fits into the window. It's also worth noting that @gnimsh was alerted to the earthquake even earlier, since he already knew it occurred in Virginia. Also of note is that the USGS alert email didn't go out until 2:06 or so.

I'm guessing that there were plenty of tweets about the earthquake from people in Virginia in the 6 minutes it took for the shaking to get here. In fact, someone with better skills at searching twitter could probably show a neat trail of tweets as the tremors spread outwards. It seems to me that you (and by "you", I mean a computer program) could roughly judge the severity of a quake by the number of people tweeting about it, and even the approximate speed of it by the spread of tweets, and subsequently issue warnings to people one or two minutes before the quake hits them.

I'm not a computer programmer or seismologist. I'm interested in hearing other people's thoughts on this, though.

3 comments:

  1. I really loved this idea.. it got me thinking and could see it working. It is sort of like the shake map on the USGS in my opinion but more in real time.

    ReplyDelete
  2. I read three tweets about the earthquake before I felt it in Maine.

    ReplyDelete
  3. Thanks, Trin! Rob, do you think being able to read about it first on twitter was helpful?

    ReplyDelete