Most of us think of floods as acts of nature. Wrong. The biblical story of the Great Flood notwithstanding, after the skies have opened up, scientists say that what humans have done to the environment has a lot to do with whether all that rainwater turns into flood water.
One of the major ways we make flooding worse comes from one of the ways we try to protect ourselves from flooding in the first place. Levees are big earthen embankments built up along river edges, essentially walls that raise the riverbanks higher, so the water has to rise higher before it can spill over. That seems to make sense. But levees don’t just raise the height of the riverbanks. They also raise the height of the river.
Think of water running down a big wide ditch that’s, say, 50 feet across. Now make that ditch into a narrow little open-topped pipe only 10 feet across. What’s going to happen when the same amount of water that used to be able to spread out in that nice wide ditch tries to run down that narrower tube? With the walls squeezing it in, the water is going to rise higher as it flows. That’s what levees do.
Riverbeds also include their floodplains, those wide flat areas next to the riverbanks that handle the extra water when rivers flood out of their central channels. The overall path of a river includes both the main channel and these floodplains. Together, they are like that wide ditch in our example. But levees cut the river off from its floodplains. They force the same amount of water to run down a narrower pipe. The water level rises. That means that when unusually heavy rains come along, the levees that are supposed to protect us have actually already raised water levels closer to flood levels.
Here’s another problem with levees: Where they’re high enough to contain the river, great. But when the river finally gets downstream to someplace without a levee, the narrow upstream channel means the river arrives at a much higher level. That means the risk of flooding where no levee has been built is much higher. That’s what almost happened to St. Louis in the great Mississippi flood in 1993. Many experts say that downtown St. Louis was spared only because levees upstream failed, allowing lots of water to spill out of the river.
It all started in the 1800s
Levee building really got going in America after Congress, back in the mid-1800s, passed the Swamp Act, encouraging the sale of 65 million acres of “swamp” for as little as $1.25 per acre in 15 Southern and Midwestern states. These swamps were river floodplains. The levees were built to capture this land for development and agriculture.
The Swamp Act might have been good news for developers. It was lousy news for flooding. Floodplains, also known as wetlands, not only give extra water in a river a place to spread out. They’re like sponges. They absorb a lot of water down into the ground, as much as 1.5 million gallons per acre, depending on the nature of the soil. They hold the water and release it slowly.
So if the water can’t get to the “sponge” because of a levee — or if the “sponge” has been filled in and built on — instead of the water being absorbed and then released slowly over time, it all gets concentrated into the main river channel over a much shorter period of time, dramatically increasing flooding.
Between the Swamp Act and other sales, we have eliminated 120 million acres of floodplains in America, 50 percent of what existed 200 years ago.
Building on wetlands has other impacts that increase flooding. Water that falls on the soil is absorbed and released slowly over days and weeks. But rainwater that falls on a pavement, or asphalt parking lot, or roadway, runs quickly into the nearest stream or river. Even office buildings and industrial facilities make things worse. Their wide roofs catch water and send it into pipes that go directly to storm sewers. Human development has accidentally created a gigantic system that captures enormous amounts of rainwater that used to soak into the ground, and sends it directly and quickly into streams and rivers. The faster all the rain that falls in a watershed runs into the river, the faster the river rises, and the more likely it is to flood.
Then there is the problem of bridges. Sure, they allow water to flow under them. Some water. But when the levels rise, sometimes bridges don’t let all the water flow under them that wants to. So the water backs up.
Think of bridges like speed bumps: When traffic is light, a few cars slowing down isn’t a problem. When traffic gets heavy, things can really back up when the first few cars have to slow down. Bridges can back rivers up for miles, increasing the likelihood of flooding.
And then there’s the problem of what we’ve built near rivers — everything from houses to factories to office parks to sewage treatment plants, even whole cities. We don’t much care about a flood if all the river does is spread out into its natural floodplain, as long as there’s nothing but floodplain there. But 10 million people now live in floodplains in America. So not only have we increased the likelihood of flooding, but we’ve put ourselves in harm’s way so that when the floods come along, they do more damage.
What's the solution?
A federal report after the 1993 floods in the Midwest suggested that the best solution to reduce flood damage is to remove most levees, make sure we don’t build on any more floodplains, and simply relocate people and key facilities — like wastewater treatment plants — out of flood-prone areas to higher and safer ground. A federal program to relocate victims of the 1993 flood helped 12,000 families relocate. But most people who were eligible refused. And despite that report, Congress still supports building levees.
Noah may have had the right idea after all. The best thing we can do is to be prepared for floods, not try to stop them. Scientific research, and decades of painful experience, have shown us that nature can deal with floods better than we can. Human efforts to do a better job often make things worse instead.
David Ropeik is a longtime science journalist and currently serves as Director of Risk Communication at the Harvard Center for Risk Analysis.