Share this:
07/01/1995 07:00:00

Caltech Science Question of the Month: Is there Earthquake Weather or an Earthquake Hour?

Answered by Lucy Jones, Seismologist with the United States Geological Survey, and Visiting Associate in Geophysics at Caltech

PASADENA—This is a monthly feature produced by the Caltech Media Relations Office, in collaboration with Caltech's faculty, to answer commonly asked or particularly intriguing questions about science and the natural world.

Question: It seems like earthquakes occur more often during hot, dry weather and during the early morning hours. Is this so?

Answer: Earthquakes occur randomly all year long and at all hours of the day and night. Scientists have looked for patterns in the time of year and time of day that temblors occur, but they haven't found anything significant. People probably associate earthquakes with hot, dry weather because most of the time in California the weather is hot and dry, so a majority of all events, from earthquakes to home sales to car accidents, happen during this type of weather.

Many people also associate hot, dry weather with temblors because the Whittier Narrows earthquake occurred on such a day back in 1987. But sixty years ago, many Angelenos thought earthquake weather was muggy, like the conditions on the day of the Long Beach earthquake in 1933. In fact, earthquakes begin many miles below the region of the soil that is affected by surface temperatures, and so aren't affected by what's happening up top.

As for the time of day, the thousands of smaller temblors, down to magnitude 2.5, that occurred between 1932 and 1994 showed no preference for any particular hour. But because people tend to be sitting still or lying down very late at night, they are more likely to notice small earthquakes that occur then. Earthquakes are also more noticeable in the wee hours because other sources of ground noise and vibrations such as trucks, trains, and cars are much less common at night.

The few large quakes we've had are random but do show a slight clustering in the early morning. For events bigger than magnitude 6.0 from 1932 through 1994, not counting aftershocks, 7 of the 18 earthquakes bigger than magnitude 6.0 (38% of the quakes) occurred between the hours of midnight and 7 a.m. (29% of the day). While that is more than average, the chance that it is random is very high. If the 18 earthquakes were distributed exactly evenly over 24 hours, that wouldn't be a random pattern. Clusters often show up in cases like this where the amount of data is small. As more large events occur the clustering should disappear.

These myths may be common because we want them to be true. Earthquakes would be less frightening if we knew when they would occur. Also, people tend to remember earthquakes that agree with a preconceived idea, and forget those that don't.

The Media Relations Office will accept any burning questions you may have.