500-year flood duly manages risk

-A A +A
By John Bartlit

The term “500-year flood” makes the news as often as sandbag brigades work on levees. Offhand, the term seems to tell the last time and the next time for such a bad flood to hit.

In reality, the term is a beautiful tool for describing and managing risk.

The workaday term does its work mostly out of sight, busily dividing up flood risk among government, individuals, and the free market. The term measures and coordinates the portions of risk handled by each sector. Dividing up risk is a central reason societies organize.

With such a complex task, how the simple term functions is worth knowing. It works at the junction of science, engineering, economics, public policy and law. Let’s see.

Technically, a 500-year flood is the height of flood water expected to be equaled or exceeded every 500 years on average. The 500-year flood is more accurately referred to as the 0.2 percent flood, since it is a flood that has a 0.2 percent chance of occurring in any given year.

Bad floods also come in the smaller, “100-year flood” variety. They have a 1 percent chance of coming in any single year. That is a five-times higher chance.

For any water level, in a given area, a predicted extent of flooding can be mapped out. This flood plain map figures very importantly in building permits, environmental regulations, and flood insurance.

The statistical risk of flooding is used to debate and decide how high, thus how safe, levees should be built. The higher the levee, the less chance it will be overtopped by flood waters.

But the higher the levee, the higher the cost.

The U.S. Army Corps of Engineers uses taxpayer money to design and build lots of levees. How high they build them is set by public policy, say to handle the big 500-year flood or the smaller, but more frequent, 100-year kind.

How much money, from whose pocket, is wisely spent to guard against how small a chance of how much harm? No amount spent can assure no chance of being flooded. A desire for perfect safety is normal, but is not useful for deciding the heights of levees. Good policies meld technical and public perspectives.

Going step by step, the system works as follows. Meteorologists study weather patterns and records – snowpack, spring runoff, rainfall, and hurricanes. Estimates are made of historical and projected water flows in a given drainage basin.

Geographers study land features. From the knowledge of land features and projected water flows in an area, engineers estimate the chance the water level will exceed a certain height. “Chance” means chance per year. Thus, engineers compute the heights of a 500- and 100-year flood in a locale.

From there, public policy takes over. How much tax money should be spent on levees? Of course, we want safe levees. We also want safe bridges, highways, air terminals, neighborhoods, schools, food, water supplies, prescription drugs, treatment options, and children’s toys.

For any total dollars spent, how are they best allocated to these and other risks to minimize total risk for the most people?

To answer, broadly rude and random exchanges occur in the forums of democracy. From the forums, decisions come. A city council or state legislature may specify levees in their jurisdiction must be built to handle a 100-year flood. The council may reject the higher public cost for levees able to withstand a 500-year flood.

The decision puts the populace at risk from anything worse than a 100-year flood, which has near a 1 percent chance of hitting next year. Maybe a sandbag brigade can stop it. Maybe not.

The remaining risk falls to personal business. Insurance companies sell insurance for flood damage. The rates they charge depend on the established maps of flood risk and the portion of it handled by levees in the area. Science, engineering, economics, public policy, and business each does its part to define and divide risk.

Other risks are similar. Roily floods offer clarity in the elements of managing risk.

John Bartlit is with New Mexico Citizens for Clean Air & Water