USGS - science for a changing world

The USGS Water Science School

A Beta version of the new USGS website has been released for public comment.
Use the "Feedback" button at bottom of every Beta page to tell us what you think.

Water Questions & Answers

Measuring Water

Floods, droughts, glaciers:

Q&A home:

Water Questions & Answers
We had a "100-year flood" two years in a row! How can that be?

The term "100-year flood," is used to describe the recurrence interval of floods. As the table below shows, the "100-year recurrence interval" means that a flood of that magnitude has a one percent chance of occurring in any given year. In other words, the chances that a river will flow as high as the 100-year flood stage this year is 1 in 100. Statistically, each year begins with the same 1-percent chance that a 100-year event will occur.

Recurrence intervals and probabilities of flood occurrence
Recurrence interval, in yearsProbability of occurrence in any given yearPercent chance of occurrence in any given year
1001 in 1001
501 in 502
251 in 254
101 in 1010
51 in 520
21 in 250

But, just because a 100-year flood happened last year doesn't mean that it won't happen this year, too. In other words, future rainfall and floods don't depend on the rainfall and floods that happened in the past. The past records are mainly used to show what kind of river flows can be expected. So, when you hear about a 100-year flood, at least you have a general idea that it does mean a BIG flood, and if you hear of a 200-year flood you know that it means one even BIGGER! As an example, in July of 1994, some places in south Georgia received more than 20 inches of rainfall in a few days -- the floods they produced were tremendous... way over the 100-year flood. At Senoia, Ga., the maximum amount of water flowing by the Line Creek gage was 2.4 times greater than the 100-year flood level.

Robert Holmes, the National Flood Coordinator for the USGS, explains it this way when talking about how a "500-year flood" occurred in the Midwest in 2008 and that it was the second 500-year flood in 15 years:

Essentially, I think as hydrologists, we've done ourselves a disservice by calling it a "500-year flood". Essentially it's a probabilistic measure. A lot of people think "OK, if I've had a 500 year flood now, this year, we've got 499 years. We don't have to worry about it again." And that's simply not the case. Essentially, a 500 year flood is just that quantity of water that has the 1 in 500 chance in happening in any one year. Another way to say it would be, there's a .2% chance of a flood of this magnitude occurring in any one year. So, it's essentially a probabilistic measure. We take our annual peak flow values from the USGS gages and we fit a probability model to those and come up with a number for a 100 year, a 500 year a 25 year whatever. You know the 100 year flood for example would be a chance of one in 100 of occurring in any one year.

Accessibility FOIA Privacy Policies and Notices

Take Pride in America logo logo U.S. Department of the Interior | U.S. Geological Survey
Page Contact Information: Howard Perlman
Page Last Modified: Monday, 17-Mar-2014 11:03:31 EDT