Extreme Events

Today’s storm gives a good context for the loads that are code-mandated for use in building design and analysis. I’m going to make a point now, at the beginning, and again at the end because it’s so important: I’m not arguing against the loads defined in building codes, I’m simply comparing extreme events with ordinary life.

The worst forecast I’ve seen for snow in NYC today is about 9 inches. That’s what’s known as base snow: undrifted and simply accumulating in one spot. The weight of snow accumulation varies with the snow’s density: we’ve all experienced lighter, drier snow and heavier, wetter snow. The density can vary a great deal, usually taken in the range of 4 to 20 pounds per cubic foot, although the extremes are rare. (In other words, snow density, like so many other things, follows a bell curve distinction.) The design ground snow load for New York is around 20 pounds per square foot; so 9 inches of average-density snow is something like 8 psf or less than half of the code requirement.

Snow’s a tough load to look at because of the varying density. What about wind? We’re expecting gusts in the 40 to 50 mph range today. The New York Building Code gives 98 mph as the base wind speed, and wind pressure is proportional to the square of the wind speed, so we’re looking at wind pressures in the ballpark of one-quarter of the code requirements. (The maximum wind speed recording in the NY metropolitan are during the direct hit from Hurricane Sandy in 2012 was 90 mph in Islip on Long Island.)

What about interior loading? The design live load for a place of assembly – the heaviest load that can come from people congregating in one place – is 100 psf. The average weight for adult men in the US is 196 pounds, and 156 pounds for women. If you look at people from above, we occupy roughly oval spaces that vary in size but are usually taken as an average of about 2 square feet per person at the tightest packing. (It’s more like 5 square feet if you want people to be able to move, but that’s not always the case.) So if you have a crowd of all men, the 100 psf looks about right, but when you mix in women, or you stop assuming that people will pack in like sardines, the 100 psf looks conservative.

Why are our design loads high? The short answer is because of all the unknowns that I’ve been fudging around in my estimates. We don’t know exactly what loads will come, and history shows that we’re not very good at guessing. So we come up with realistic scenarios based on research, and then apply a safety factor to account for what we don’t know.

It always seems possible to make a design or analysis look better by saying “those loads are not realistic” but unless you know for certain that they’re not, it’s a terrible idea. (Also, not allowed by code, which is a fancy way of saying “illegal.”) What does “certain” look like? Since wind can only apply pressure perpendicular to a surface, the engineering community is generally okay with the fact that rowhouses can’t be shown to have the capacity to withstand code wind pressure in their short direction. The houses are not exposed to that wind. If you modify a house, the situation changes and this may not be true anymore, but I’d never argue that we need to take full wind across a house in the middle of a row under the existing conditions.

So in short: design loads seem high, but there’s a good reason for that and we generally have to stick with them.