Sunday, December 21, 2008

My holiday earthquake theory

Two blogs back I wrote on earthquake (seismicity) patterns. Well, I have a pet theory that earthquakes tend to occur on holidays. Cases in point:
  1. Lisbon earthquake, 1755 - All Saint's Day. According to Rob Zaretsky, professor of French history at the University of Houston, over 100,000 people died. The earthquake also pounded Europe's political and cultural landscapes. The brightest light of the Enlightenment, Voltaire, used the earthquake in his wonderful satire, Candide. From a ship in Lisbon harbor, Candide watched helplessly as the good drowned and the wicked survived.
  2. Alaskan earthquake, 1964 - Good Friday. This is the second biggest earthquake ever recorded, surpassed only by the 1960 Chile quake. According to the Alaska Earthquake Information Center, the area of significant damage covered about 130,000 square kilometers. It was felt over 1,300,000 square kilometers (all of Alaska, parts of Canada, and south to Washington). The four minute duration of shaking triggered many landslides and avalanches. Major structural damage occurred in many of the cities in Alaska. The damage totaled $300-400 million dollars. You betcha.
  3. Sumatra-Andaman earthquake, 2005 - day after Christmas (that's kind of a holiday, isn't it?). The IRIS special report states that approximately 1200 kilometers of the India-Burma plate boundary slipped, with an average displacement on the fault plane of about 15 meters. The resulting tsunami claimed 250,000 lives.
  4. Lancaster, Pennsylvania, 1984 - Easter Sunday. This is actually the only earthquake I have ever experienced. It was studied by our own group's Charlie Scharnberger (see booklet on Earthquake Hazard in Pennsylvania) and John Armbruster. Actually, this is my favorite kind of earthquake - sensible, exciting, and useful for scientific research, yet it only caused a little bit of cracked plaster.
I like the web site at left, Today in Earthquake History. Check out other holidays. What do you think - does my theory amount to anything?

Hoping your solstice and upcoming holidays are disaster-free.


  1. Ah, the old observer bias. You note the big earthquakes that happened on holidays but not big earthquakes that did not happen on holidays. Since you set the range from 1755, and as USGS notes: "According to long-term records (since about 1900), we expect about 18 major earthquakes (7.0 - 7.9) and one great earthquake (8.0 or above) in any given year," there should be approximately 250 great earthquakes. Four is a very small hand-picked set, and really it is only three great earthquakes, and one "light" (magnitude 4-4.9).
    For every magnitude 6 earthquake there are 10 magnitude 5, 100 magnitude 4. If you assume the rate holds for magnitude 7=8, and the average rate of worldwide magnitude 8+ earthquakes for the past century is 1/yr, the rate of a magnitude 4 is ~10000/yr or approximately 27/day. The NEIC FAQ says they locate 12-14,000 earthquakes per year, but it's not clear if that's only USGS or all contributing networks (like Lamont where we're reporting tens of events per year, almost all below magnitude 4).

    An odd outlier in the numbers. Either USGS is missing a lot of worldwide earthquakes of magnitude 4, or the rate of M 7-7.9 earthquakes is not in scale (~18/yr) which would mean 18000 magnitude 4 earthquakes per year. But if the relationship is really closer to a factor of 8 than 10, the figure for average number of magnitude 4 events becomes 18*8^3 or 9216 events (+1152 M5, 144 M6, 18 M7, 1 M8+), or a total of around 10500 events/year. Pretty good chance of hitting a holiday then. - Common Myths About Earthquakes

  2. I recently came across your blog and have been reading along. I thought I would leave my first comment. I don't know what to say except that I have enjoyed reading. Nice blog. I will keep visiting this blog very often.


  3. There was a magnitude 1.8 quake in northeastern Massachusetts this past Sunday evening ( just as Hanukkah was beginning. I (jokingly) rest my case.

  4. Now that I could believe... perhaps a change in thermal gradient near the surface.

  5. More credence - our automatic detector picked up three events last night from some bigger events.

    5.9 2008/12/25 08:12:00 49.132 -128.526 10.0 VANCOUVER ISLAND, CANADA REGION

    6.2 2008/12/25 03:20:30 5.772 125.535 207.5 MINDANAO, PHILIPPINES

    Those three events are really just two, but the depth phase of the Phillipines located as another event about 9 minutes later and in a different location.

    12/25/2008 3:19:49.859 -4.1216 122.7746 h=30
    12/25/2008 3:28:46.214 32.3948 145.4381 h=30

    The 10 degree mislocation of the 3:19 event isn't too bad considering the poor azimuthal coverage and distance of stations to the event.


Wave to us!