Welsh Corgi
Do dogs sense earthquakes before humans do? -Representational imageTIMOTHY A. CLARY/AFP/Getty Images

Dogs, cats, cows, and other animals cannot really sense earthquakes before they hit. The myth is a long and storied one, but there is no real proof, apart from anecdotal evidence to actually support it.

A study on this phenomenon has concluded that there is no strong evidence to back this claim. Most evidence points out to singular observations and stories from people that cannot really be tested, notes a report by Phys.org (PO). Authors of the paper, first published in the Bulletin of the Seismological Society of America, suggest that a series of questions that researchers going forward to study abnormal animal behavior.

The right way to determine whether there is indeed a clear link between earthquakes and abnormal animal behavior should be based on clearly defined rules. Some of them include the distance of the animal from the epicenter of the quake, magnitude of the impending event, if this "abnormal behavior" has been observed without an earthquake coming, and a number of other variables, like the health of the animals in question.

Once these questions are answered, there has to be a statistical testing hypothesis in place to either prove or disprove the data collected, says Heiko Woith and his team of researchers at the GFZ German Research Centre for Geosciences.

Questions like the ones mentioned are rarely ever asked, so it becomes difficult to prove or even disprove the concept of animal prediction. For this study, researchers analyzed 729 reports linked with 160 earthquakes. "Many review papers on the potential of animals as earthquake precursors exist, but to the best of our knowledge, this is the first time that a statistical approach was used to evaluate the data," said Woith.

Geologists collected reports of several animals that seemingly had the ability to predict earthquakes ranging from silkworms to elephants. A good number of the quake stories came from three events, notes the report; 1984 Nagano-ken Seibu earthquake in Japan, 2009 L'Aquila earthquake in Italy, and 2010 Darfield earthquake in New Zealand. Even in these instances, most accounts were merely anecdotal.

The stories ranged from seconds before the earthquakes hit, to months before the event and at distances that covered a few hundred kilometers. Of the reports, only 14 of them were series of records over time, the rest of them were single, one-time observations, say researchers.

In the available data, there is a strong cluster of abnormal behavior and foreshocks, suggesting that there might be some behavior related to seismic events. Another option could be secondary effects triggered by the foreshocks, like changes in groundwater or release of gases from the ground which might be sensed by the animals.