A 100-year flood does not only happen once every 100 years. So why is it called that?
This term can be misleading. The phrase, “100-year flood” is an example of a return period for a flood event. Return periods are a common, yet widely misunderstood, metric for the occurrence of natural hazards such as floods, earthquakes, and eruptions. They describe how likely a hazard event is to occur at, or above, a specific intensity within a time frame defined by a probability.
A longer return period (for example, 100 vs. 20 years) suggests a lower probability that an extreme hazard will occur in any single year.
This tool is meant to help convert the return period for an event into a probability, showing that a “100-year” event could occur in any year, or reoccur multiple years in a row. While developing the tool, we made certain assumptions, such as defining adulthood as 20 years and a lifetime as 80 years, for simplicity's sake.
The most common misconception is that a 100-year flood will only occur once per century, but that is not true. There is a small probability that such an intense event could occur every year. If a 100-year flood happened last year, it can happen again before the next century, or even this year. It is also possible for such an event to not occur within a 100-year period.
As a recent example, the city of Houston, Texas, USA, has experienced 500-year floods three years in a row, including one caused by Hurricane Harvey, which has prompted a revision of the city’s zoning regulations to account for changes in the flood drainage basins around Houston.
We expect to see many similar revisions as climate change alters the frequency and intensity of extreme events, and the rapid sprawl of cities shrinks the natural areas available to absorb floodwater.