Skeptoid PodcastSkeptoid on Facebook   Skeptoid on Twitter   Skeptoid on Stitcher   iTunes   Google Play

Members Portal

Store

 

Get a Free Book

 

SKEPTOID BLOG:

Going Ape over Earthquakes

by Craig Good

August 26, 2011

Share Tweet Reddit

Donate A recent story about the devastating D.C. quake dredged up the old claim that animal behavior can predict earthquakes.
Animals at the Washington National Zoo started behaving oddly minutes before Tuesday's magnitude 5.8 earthquake hit.
Quakes seem to dislodge quack claims as much as they do dust and debris.

There are plenty of resources showing why this claim is dubious. National Geographic:
American seismologists, on the other hand, are skeptical. Even though there have been documented cases of strange animal behavior prior to earthquakes, the United States Geological Survey, a government agency that provides scientific information about the Earth, says a reproducible connection between a specific behavior and the occurrence of a quake has never been made.
And Skepticblog.
The idea that animals can predict earthquakes has been carefully analyzed and has failed the test again and again. Animals may be more sensitive than humans to the P-waves, which are the fastest seismic waves and arrive several seconds before the destructive S-waves in regions far from the epicenter—but this gives warnings of only a few seconds in any place that is likely to experience strong shaking. If animals are sensitive to other disturbances in the earth's crust that happen more than a few seconds before the quake itself, it has never been reliably corroborated. In addition, this method runs into the same problem that most short-term earthquake prediction methods have encountered: no two earthquakes are alike. Some have precursors, and others don't. Thus, if animals did act strangely before a particular quake occurred (just as some geophysical precursors have been observed on some quakes), there is no evidence that they reliably predict most quakes (just as many quakes don't have precursors).
The story I linked to at the top of this post provides good fodder for a skeptical thinking analysis.
First to respond to some hidden signal, says the zoo, were the red-ruffed lemurs, which sounded an alarm call about 15 minutes before the quake and then again just after it occurred.
The writer assumes that lemur alarm calls never happen at other times and for other reasons and, in an interesting twist on a post hoc fallacy, further assumes that 15 minutes is close enough to the quake for the two events to be related. Which is in direct conflict with the next paragraph.
About five to ten seconds before the quake, many of the apes, including Kyle, an orangutan, and Kojo, a western lowland gorilla, abandoned their food and climbed to the top of the tree-like structure in the exhibit.

About three seconds before the quake, a gorilla called Mandara let out a shriek, collected her baby, Kibibi, and did the same.
So 15 minutes to 3 or 5 or 10 seconds. The latter is at least more credible because of that P-wave idea. I wouldn't be surprised if just better low-frequency hearing explained this. But, again, this assumes that orangutans never abandon their food and climb trees at other times and for other reasons.
Another orangutan began 'belch vocalizing', making a noise normally reserved for extreme irritation, before the quake and carried on afterwards. Snakes writhed, flamingos huddled, and creatures from a shrew to a Komodo dragon hid.
Carrying on afterwards is neither odd nor predictive, of course. Somehow the writing, huddling and hiding don't seem different from what I've seen on normal days at the zoo. Now comes the most telling paragraph.
However plenty of animals faled to react at all, including the Przewalski's horses, the scimitar-horned oryx and the giant pandas.
Pandas are just mellow. Everybody knows that.
Many animals appear to be able to sense disasters coming. Last year, a toad researcher - yes, they do exist, reported that her subjects scarpered a full five days before a major earthquake struck 74 miles away in L'Aquila.
Wait. I was complaining about 15 minutes for animals a short distance from the quake?Now anything strange within 5 days and 74 miles is a prediction? When psychics try this we call it cherry picking. I'm pretty sure somewhere in Virginia a cabbie honked his horn at somebody within 5 days of the quake. Should that count as a prediction?
It's been suggested that the animals may be responding to the release of radon gas, triggered by changes in the ionosphere.
Suggested by whom? There are people working on using radon detectors to predict quakes, but I can't find anything suggesting that animals are good radon detectors. "It's been suggested" is an even bigger red flag than "some say".

Remember: Just because a story appears on the "science" page doesn't mean you can let down your critical thinking guard.

by Craig Good

Share Tweet Reddit

@Skeptoid Media, a 501(c)(3) nonprofit

 

 

 

Donate