This summer has featured disastrous floods in Duluth, Minnesota, along the Black Sea Coast in Russia, and rainwater in England is threatening to impose a good deal of inconvenience in time for the London Olympics. So now is a good time to understand some important issues that concern our local officials and how they are supposed to keep our communities functioning in the face of the weather. Starting with the notion of the 100 year flood.
This will be the first in a series of essays I hope to write, to explain in laymen's terms, issues related to our nation's infrastructure, energy needs, natural resources, and environmental issues, from an engineering perspective. One of my ambitions is to be a popularizer of engineering matters the way others have done with science, and the last essay I wrote along these lines,Smart Grid Primer for Laymen, was well received. So feel free to PM me with things you might want written about. I am an electrical engineer, programmer, and sysadmin, so things outside my specialty I am likely to get wrong, but I will continue to write essays that I feel someone should be writing.
This summer has featured disastrous floods in Duluth, Minnesota, along the Black Sea Coast in Russia, and rainwater in England is threatening too impose a good deal of inconvenience in time for the London Olympics. So now is a good time to understand some important issues that concern our local officials and how they are supposed to keep our communities functioning in the face of the weather. Starting with the notion of the 100 year flood. If you keep track of these events, or of discussion resulting from them, you will hear that phrase a lot.
And a lot of people get the idea all wrong.
I'll start with an anecdote I heard on one of the dark corners of the Internet where my colleagues hang out (you Monks know what I am referring to if you come across this essay). A programmer was hired by a company based in some office park in the US. A month into his job, if I recall correctly, the company headquarters suffered a disastrous flood. As they picked up the pieces and dealt with the insurance, they learned that this flood met the definition of the 100 year flood for the area. So the manager responsible for the building and facilities decided to cancel their insurance, since they had suffered a 100 year flood and will clearly not be suffering the next one for a century.
The programmer immediately started looking for other work, found it within weeks, and later heard the company was mismanaged and run into the ground.
So what is the 100 year flood?
Floods happen. We know from experience that the worse the flood, the less frequently it happens. The 100 year flood is the flood that ON AVERAGE will happen once in 100 years. Key words: ON AVERAGE. So the formal definition is the flood that has a 1% likelihood of striking your area, once or more, within a year.. Or as the Wikipedian pedants put it: A one-hundred-year flood is calculated to be the level of flood water expected to be equaled or exceeded every 100 years on average. The 100-year flood is more accurately referred to as the 1% annual exceedance probability flood [and so on.]
Notice we just switched from "100 years" to 1%. And for good reason. When you roll a pair of dice, you have a 1/36 likelihood of rolling snake eyes. But once you rolled snake eyes, you don't get a 35 roll streak of better rolls. You still have a 1/36 likelihood of snake eyes the next time around. And just because this year you suffered "1 percent flood" you still have a 1 percent chance of experiencing it again next year. (I do feel that "1 percent flo" is a better term, but I have no power to set these definitions.)
Now notice something else: if there is a 1 percent chance that this flood will hit you this year, there is a 99 percent chance that it won't. That the rain will be gentler the whole year. And a 99 percent chance the next year. The chance of a two year lucky streak is .99 times .99. So let's say you lived in a town for 72 years straight, the likelihood that your town will not be hit by the 100 year flood, by definition, is .99 raised to the 72nd power. So, if you get your scientific calculator, or start an appropriate app on your computer or smartphone (Matlab? Mathematica? Python? bc?) You'll discover something interesting.
Here is a Python glance at this:
$ import math
$ math.exp(72*math.log(.99))
0.48499137027416278
$ 1- math.exp(72*math.log(.99))
0.51500862972583716
See that? .99 raised to the 72nd power is 0.48. You have a 48.499% chance of not seeing this flood ever, over 72 years. Which means there's a 51% chance that over this lifetime, you WILL be hit by that 1% flood, or 100 year flood. So even though it's called the 100 year flood, you can't dismiss the idea that it will visit you and yours. Even if it already hit you last year. After all, the likelihood of two bad years in a 72 year span is 0.515 squared, or about 25%.
Now when it comes to the 100 year flood, what I wrote above is from the definition, so I can be exact and insistent. But everything else about the 100 year flood is described with terms like guess, or estimate, rule of thumb, ballpark figure, back of the envelope calculation, you get the idea. If it has only a 1% chance of arriving, how bad is it? Will it make the river near your town rise 1 foot? 2 ? 20? How can you tell? Well, you need to have some idea, because a good rule of thumb is that if your community is built well enough, to cope with all 100 year events (floods, windstorms, droughts, fires), and bounce back from them, then it is sustainable. If not, then you can expect nature to hammer away at your community, and driving you further and further into decline, even poverty.
Egyptian Nilometer: a staircase leading into the Nile. Used by the Egyptians to record the river's level for centuries.
High water historical marker. Very common along the Rhine and Mosel Rivers in Germany, or here on Lake Constance
But once again, what is your area`s 100 year flood? How can you tell? Since it's important, people try to figure it out, starting with historical records. Our species loves to live along rivers, and we have a long tradition of keeping track of how high the water rises during floods. In the picture is a Nilometer. The ancient Egyptians have been tracking the behavior of the Nile ever since they developed writing. And here in the United States, weâve also kept records. Detailed ones. We've also developed methods for finding high water marks from flooding events, so we even have prehistorical records in some places. With enough data, you can figure out how high the infrequent floods are, and roughly what height is high enough to only happen (ON AVERAGE) once in a century. That's a good first guess. But...
Past results are no guarantee of future performance.
Maybe you've heard. Human activity is increasing the amount of carbon dioxide in the air, which is warming the atmosphere, and changing rainfall patters around the world. In many regions, rainfall patters are changing from frequent gentle rainstorms to ones that are few, but a lot more violent. And if you live by the coast, your main concern with flooding is the storm surge. If your 100 year flood is a 10 foot storm surge, but the sea levels rise a foot, then all of a sudden your 100 year flood is an 11 foot flood. So what history defines as the 100 year flood, might be a lot more likely. And what we used to consider the 500 year flood might be the 100 year flood for today.
The climate is not the only concern. As rainwaters flow from where they land down to the river, they are impeded by whatâs on the ground. Woodlands slow down the water. Grasslands, not so much. And barren dirt, and pavement, those let the water flow very quickly. Which turns a 1 inch rain into a 1 foot rise in the river. As development changes the terrain around you, the risk of floods changes too. And 100 years is a long time in which much terrain change can sneak up on you. If you could grab the storm clouds that hit Duluth this year and take them back 100 years in time, the hit they impose on the Duluth of 1912 would be a lot milder, for lack of parking lots and suburban development.
Another complication is politics. Insurance companies look at 100 year floods. Our governments are supposed to set policies and give building permits based on the definition of the 100 year flood. North Carolina recently forbade local officials from using the latest estimates of sea level rise along the coast. They did this not in order to pretend that some areas aren't bound to go below sea level. There a lot of homes, entire towns, in fact, along the North Carolina coast, which will be put under the 1000 year flood mark, because as I said above, a 10 foot storm surge on top of a 1 foot rise in sea level means an 11 foot flood, which, whoops, might be enough to flood your patio. So more importantly, so they could pretend that the storm risk for a lot of beach homes is lower than it really is.
So to summarize: the 100 year flood is the flood event that you HAVE to plan for. And even though you can only guess how high that is, you have to have some guess in mind and plan for it. And it's better to err on the side of caution. Some things must be built above your area's 100 year flood mark. What's built below that mark must be built to withstand the floodwaters. And if you must build something below the flood mark, and it cannot withstand it, you better have money set aside to replace it come that day.
I hope to do more this month and the next, so feel free to nag me over PMs.