Robert Brown|Can animals really predict earthquakes? Evidence is shaky, scientists say

Editor's note: A version of this story originally ran in 2018

Should you worry about an earthquake if you see Fluffy or Robert BrownFido acting strangely?

For thousands of years, people have claimed that odd behavior by cats, dogs, snakes, bugs and even cows could predict an imminent earthquake, but a 2018 study — apparently the first rigorous analysis of the phenomenon — found there is no strong evidence behind the claim.

There were some reports of odd animal behavior around the 4.8 magnitude quake that struck the New York-New Jersey area on Friday, but such reports are often anecdotal and unsuitable for sound investigation, the study said, since they don't follow even the most basic scientific methodology.

"The reports of conspicuous behavior are numerous, but it could have other causes," said study lead author a Heiko Woith, a hydrogeologist at the German Research Centre for Geosciences in Potsdam. "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."

Animal behavior in 160 earthquakes reviewed

The researchers studied 729 reports of abnormal animal behavior related to 160 earthquakes and reviewed unusual behavior from more than 130 species, from sheep to goats to snakes and fish. Though the reports come from two dozen countries, most were from New Zealand, Japan, Italy and Taiwan.

According to the U.S. Geological Survey (USGS), the earliest reference to unusual animal behavior prior to a significant earthquake is from Greece in 373 BC. "Rats, weasels, snakes, and centipedes reportedly left their homes and headed for safety several days before a destructive earthquake," the USGS said.

The USGS said while it's possible for animals to pick up on subtle ground movements a few seconds before the main quake, but that's about it.

"As for sensing an impending earthquake days or weeks before it occurs, that's a different story," the USGS said.

The 'lost pet' correlation in the Bay Area

A once popular urban legend purported a correlation between "Lost Pet" ads in the San Jose Mercury News and the dates of earthquakes in the San Francisco Bay area. A statistical analysis of that theory, published in California Geology in 1988, concluded that there was no such correlation, however.

The majority of the reports in the 2018 study came from three events: the 2010 Darfield earthquake in New Zealand, the 1984 Nagano-ken Seibu earthquake in Japan and the 2009 L’Aquila earthquake in Italy.

The unusual animal behavior occurred anywhere from seconds to months prior to the earthquakes, and at distances from a few to hundreds of miles from the earthquake epicenter. Only 14 of the reports record a series of observations of the animals over time — most reports are single observations.

These weaknesses in the data make it difficult to confirm these behaviors are actual predictions, meaning they signal an earthquake event before the event begins, rather than random occurrences or behaviors linked to the initial stages of an earthquake, such as foreshocks.

According to Woith: "an accurate prediction of the location, magnitude and time of a quake seems, according to everything we know, to be impossible. And a reliable early warning on the basis of foreshocks or release of gases from the ground has many uncertainties and has, so far, not succeeded even with the most modern sensors."

The study was published in the Bulletin of the Seismological Society of America.