So is Houston just on a historically unlucky run of flooding, to be followed by a return to normal soon? Or was there some miscalculation of how frequently these massive flooding events occur? Or, most alarmingly, is something else happening that suggests these catastrophic weather events are becoming much more common?
Let's start with what it means to be a "500-year” flood.
A 500-year flood isn't necessarily something that happens once every five hundred years. Rather, a 500-year flood is an event that has a 1 in 500 chance of occurring in any given year. “For a 500-year flood, there is a 0.2 percent chance of having a flood of that magnitude occurring” in any given year, according to the National Weather Service.
Think of it as more a statistical term than a meteorological one. The probability that any extreme storm will happen is completely independent of the probability of the next one. “If one of those events occurs, it has no effect on future events occurring,” according to the National Weather Service. “In other words, if a 100-year flood event occurs, that does NOT mean that people are 'safe' for 99 years. The risk of having the flood in any given year is the same," regardless of whether it occurred recently. Ditto for 500-year floods.
Practically speaking, that means you can have multiple 500-year flood events happen essentially back-to-back. Indeed, that appears to be happening in Houston right now, with the flooding in 2015, 2016 and today.
Here's the other big caveat: The term is applied to a local area, not to the United States as a whole. So when meteorologists say the Houston is experiencing a 500-year flood, they mean there is a 1 in 500 chance of it happening in any given year in Houston.
The United States has 3.5 million miles of rivers and streams and 95,000 miles of shoreline, which means lots of potential for flooding — even flooding that's historic, by local or regional standards — in any given year. As it turns out, the country experiences multiple 500-year flood or storm events (that is to say, an event that in had a 1 in 500 chance of occurring in that given place) every single year. Including Harvey, the country has experienced at least 25 such 500-year rain events since 2010, according to the National Weather Service.
Before Harvey, the most recent one happened in Missouri in May, when some areas were soaked with a foot of rain in a few hours. Other recent 500-year rain events include Hurricane Matthew, which roared up the Carolinas last fall, and the torrential rain that destroyed much of Ellicott City, Md., last summer. Harvey will be at least the fifth major 500-year flood to hit Texas since 2010.
Here's a map of where some of the major ones have occurred since 2010, as compiled by the National Weather Service.
This brings us to another variable in the equation: Climate change.
It's unwise to try to link climate change to any specific storm or even string of storms. Even if we weren't heating the planet with a steady stream of human greenhouse gases, all of these storms would have likely happened anyway. But climate scientists do believe that global warming is creating conditions that allow these storms to become more powerful, and perhaps even more frequent.
Climatologists say the mechanism by which this is happening is fairly straightforward. “Warmer air can contain more water vapor than cooler air,” according to the 2014 Climate Assessment produced by the U.S. government. “Global analyses show that the amount of water vapor in the atmosphere has in fact increased due to human-caused warming. This extra moisture is available to storm systems, resulting in heavier rainfalls.”
That assessment includes this eye-opening chart on the rising prevalence of “heavy precipitation events” — defined in this case as five-year rain events, or events that have a 20 percent change of occurring in any given year.
This charts the number of these events in a given decade, relative to the average number for the period of 1900 to 1960. The 1990s saw 30 percent more of these heavy rainmakers than the typical decade between 1900 and 1960 did. In the 2000s there were 40 percent more of these events.
There's one final issue: Not all estimates of 100-year or 500-year events are equal.
Terms like 100-year and 500-year began to be applied to floods with the creation of the National Flood Insurance Program in 1968. The program needed to identify which areas of the United States were at risk of flooding and which weren't. Floodplains came to be defined as land likely to flood during a 100-year rain event.
Climatologists base their estimates of 100-year and 1,000-year rainfall estimates based on a statistical analysis of prior years' rainfall data. The more data you have, the better your estimate is. Some places have more data available than others. As new data rolls in every year, estimates everywhere can be updated.
All of that is alarming news for residents of Houston, New Orleans and the millions of other people who live in areas vulnerable to extreme flooding. It cumulatively suggests that what we once projected would be extreme weather is more common than we thought and, as the climate cooks, is only getting more so.
The latest data we have on extreme precipitation events, through 2015, shows that the trends outlined above are likely to continue.
This story has been updated.