Designing infrastructure is a tricky endeavor. The day-to-day workings are easy to deal with, but you always have to consider the extreme events. During and after an extreme event is when you’re going to need that infrastructure the most. Early on you have to face up to the fact that you can’t entirely flood proof your bridge, earthquake proof your water treatment plant, or airline proof your skyscraper. You can spend all the money in the national treasury at a structure but there’s always some remote chance that mother nature will throw down an even bigger disaster. Then you have a very expensive pile of rubble and no more money to rebuild it. We won’t even get into all the other stuff you couldn’t build because you spent all your money on the aforementioned pile of rubble…
This is where the concept of acceptable risk comes into play. It’s always delicate using saying ‘acceptable risk’ in a situation where lives might be lost. Most people tend to think in terms of ‘cost doesn’t matter if we can save one life’. Maybe in a perfect world that would be true, but in the real world cost matters. Resources are finite and if you spend an extra million to save one life here then you run the risk of these other four places where tragedy could have been prevented if you had only spent $250,000 at each. Quantifying flood risk is a way of prioritising your efforts to make sure you get the most use out of your scarce resources.
The concept of the 100 year flood was invented to try and quantify acceptable risk. I’m using ‘100 year flood’ as a typical example because it’s fairly commonly used in the media and is threshold used for flood insurance requirements. Engineers regularly work with 2 year floods, 500 year floods, and so on… And it’s a concept that’s applied to other types of natural disasters as well. There are designs for 100 year earthquakes and I’ve even seen it applied to tsunamis.
Before we go further, let’s take a closer look at that phrase, ‘100 year flood’. What does that mean? It means ‘the flood that has a 1 in 100 chance of happening or being exceeded in a given year’. Essentially, you have a 1% chance of getting a 100 year flood (or worse) in any particular year. A 2 year flood means a 1in 2 chance (50%), and a 500 year flood is a 1 in 500 (0.2%) risk. The number of years being used is called the recurrence interval. Larger recurrence intervals mean more severe floods, but they are also less likely to occur.
There is one huge misconception that needs to be addressed. These various flood levels are based on statistics and not science. Surviving a 100 year flood doesn’t mean you’re safe for the next 99 years. The 1% chance is the same every year. The idiosyncrasies of statistics indicate that it’s very unlikely to have 100 year floods in back to back years, but it can happen easily enough. And that’s leaving aside the perils involved in deciding what actually constitutes a 100 year flood. (Which I will be addressing in another post.)
Flood recurrence intervals are often reported in the press, but their primary use is as a design tool. One of the biggest decisions in any infrastructure project is determining what recurrence interval a project should be designed to survive. Now that we’re quantifying the risk of a particular storm, it greatly simplifies the decision. The design storm is based primarily on the size and importance of the project, and the potential damage if something goes wrong.
Let’s take bridges as an example. The average interstate bridge in Metro Nashville has thousands of cars crossing it daily, while a rural county road may have a few dozen cross it. (I’ve worked on a few that had an average daily traffic count in the single digits.) A flood in Nashville may destroy entire subdivisions, strip malls, or industrial plants while a similar size flood in rural Cheatham county might destroy a few acres of corn and drown a few cattle. Since the potential for damage is so much greater the interstate bridge would be designed to stand up to a 100 year storm while a rural county road will be expected to endure a 10 year storm. Most TVA dams are designed to handle events even beyond the 500 year recurrence interval due to the huge potential for damage if something goes wrong.
This is a somewhat simplified discussion, but it gets you started in understanding the design process. In a future post I’ll be discussing how the size of floods of various recurrence intervals are determined.