Doug Sheridan
“Hardly any weather event term misleads as often as the “100-year” flood. ”
Doug Sheridan
Hardly any weather event term misleads as often as the “100-year” flood. As a statistical matter, in the US the term refers to an event that has a 1-in-100 chance of occurring in a given year. The standard was adopted federally in the 1960s as a basis to calculate risks for the new National Flood Insurance Program.
As an unintended consequence of its use, property owners may believe if they’ve experienced one major flood event recently they should be safe for the next century or so. In reality, a 1-in-100 chance each year equates to a 26% likelihood that a property in a floodplain will be inundated in the next 30 years—the life of a typical mortgage.
Not only does the 100-year standard mislead property owners, it also sets an unrealistically low bar for cities planning for flood mitigation. Take Houston for example. The city soaked under three 500-year floods between 2015 and 2018. With a 0.2% probability for each flood in any given year, the odds of the trifecta of storms that occurred in such a short span was upwards of one in 125 million.
So why do events with such long odds seem so commonplace? One reason is the large number of locations monitored. The USGS tracks water levels at approximately 7,500 locations across the nation. With each site representing an opportunity to register an extreme event, there’s a 90% probability a 100-year flood event occurs somewhere across the network every 11 days.
In short, it's a numbers game—the shear amount of locations and multiple metrics that are monitored greatly increases the odds of a 100-year event being recorded somewhere. This means journalists and climate activists “shopping” for extreme weather events have a rich and ongoing set of opportunities from which to choose.
To find out how rich, we first calculated the frequency of a 100-year event occurring given 50 monitoring locations, one in each US state, for any of three event types—max temperature, 24-hr precipitation and drought severity. The results show a 90% likelihood that at least one 100-year event would occur every 885 days on average.
But that’s assuming just one monitoring site per state. The frequency falls to once every 111 days if the number of locations is increased to ten per state. It falls to once every 7 days if we expand the number of locations to 10,000—a little over three per US county. A single monitoring station in each of the 196 million square miles comprising Earth’s surface would record a 100-year event every 13 seconds on average.
To Sum It Up: Like many artifacts related to the climate, extreme weather metrics have been weaponized by the media and climate activists in an effort to put the public on edge. Don't fall for it—ours is a giant planet with immensely volatile weather patterns. Breathless reporting of naturally occurring outliers, whether intentionally or not, exagerates these events' ties to human-caused climate change. Now you know how.