Why It's Hard To Plan How To Dress For The Weather In Rapid City

The front entrance of the National Weather Service office
in Rapid City, South Dakota during an extreme
blizzard in October, 2013. An analysis shows the city
has the most extreme weather of any in the U.S.  n

Where I live here in Vermont, there's the old saying, "If you don't like the weather in Vermont, wait a minute and it will change."

I believe they use that phrase in many other parts of the country, too.

But where that cliche fits best is in Rapid City, South Dakota, it turns out.

At  fivethirtyeight.com, they've done an analysis ranking the relative unpredictability of the weather.

The word they should have used is volatility of the weather, but we'll get to that in a bit.

The analysis by fivethirtyeight compares weather patterns in every part of the country against long term averages. The more the weather typically deviates from the "normal" the more "unpredictable" the weather is, in the eyes of Nate Silver and Reuben Fischer-Baum, who wrote the article.

The pair looked at 120 American cities, each representative of the 120 National Weather Service forecast offices in the United States.

Silver and Fischer-Baum looked at temperature, precipitation, snowfall, wind, severe weather and other phenomenon to determine their scale of unpredictable, or predictable weather.

Rapid City turned out to have the most unpredictable, or volatile weather. Honolulu, Hawaii had the most predictable weather.

The cities with the most volatile weather were mostly in the upper Great Plains. Cities with the most settled weather tended to be near coastlines.

That makes sense. Oceans tend to have a modifying influence on local climates, tamping down on extreme high and low temperatures and often discouraging severe thunderstorms. Yes, coastlines tend to have storms, but they're not as frequent as day to day variations in temperatures and conditions as inland.

The Pacific Coast has more predictable weather than the East Coast. Weather systems tend to come from the west, so the Pacific Coast gets a lot of influence from the ocean.

Weather systems from inland, the west, tend to affect the East Coast often, so the weather there is more volatile than the other side of the country.

The cities with the most volatile weather tend not to be very big. The fivethirtyeight analysis says the top 23 most volatile weather cities are not within the metropolitan areas of the nation's 50 biggest cities.

Of the top 50 biggest metro areas of the United States, Kansas City is the most "unpredictable," according to the fivethirtyeight analysis.

I'm a little disappointed my region is not the most volatile. (Weather geeks like me like extremes, I guess.)

Burlington, Vermont, the city closest to my home, ranks as the 38th most unpredictable weather city out of the 120 listed.

I've been batting around the words "unpredictable" and "volatile" in this post to describe what's going on with this analysis.

I agree with the Capital Weather Gang that maybe Silver and Fischer-Baum are mistaken in calling cities with more extreme weather "unpredictable."

To rank the predictability, they should have analyzed National Weather Service forecasts, or those from other forecasting firms, to compare how many of the forecasts turned out to be correct in each city.

So the term we should use in the fivethirtyeight analysis is a ranking of how variable a city's weather is, not its unpredictability.

Here's how the Capital Weather Gang put it:

"...volatility isn't always the best indicator of unpredictability. Big fronts which cause big swings in the weather are actually pretty easy to forecast, whereas more subtle variations can be much more difficult to detect."

For instance, a cold front a couple weeks ago caused temperatures to drop by some 60 degrees overnight in the Great Plains. That's highly variable weather, but the temperature change was very accurately forecast ahead of time.

A couple days later a storm on the East Coast produced rain and snow. Some cities had much more snow than forecast, others had less. Only a subtle change in temperature by one or two degrees from say, 32 to 34, made all the difference.

Not a big change in temperature, but the subtle change made it super hard to forecast the effect of teh storm in any given area.

Speaking of extremes, another interesting statistic came out this week. In November, the United States had, technically, the most extreme weather on earth, based on temperature.

The analysis looked at how far above and below normal temperatures for the month of November were everywhere.

A spot near Cable, Wisconsin had temperatures the most below normal for November, about six degrees lower than average.

Meanwhile, a spot just off the northern Alaska coast in the Beaufort Sea had the biggest warm departure from normal in November, more than nine degrees.

Once again, I missed out in the extremes. Temperatures in Burlington, Vermont in November, 2014 turned out to be exactly normal.

Oh well, better than freezing or roasting, I guess.