Wind chill dropped as low as 52 below zero in parts of the Midwest on Thursday, with similar conditions expected for early Friday. Meanwhile, parts of northern Texas may be hit with a wind chill of between minus-1 and minus-9 degrees—the coldest local weather in 12 years. In this column, first published in 2007 and reprinted last winter, Daniel Engber explains that "wind chill" is little more than shameless puffery.
If the weather makes headlines only when it's horrendous out, wind chill is its PR agent. This week, when temperatures in New York City dropped to single digits, newspapers and TV meteorologists breathlessly reported that the wind chill had hit minus 11. In Ohio, they told us, the thermometers read close to zero, but gusts of cold air made it feel like 25 below. Banner stories proclaimed a wind chill of 35 below in Chicago.
The weathermen trot out these arctic, pumped-down numbers to put an exclamation point on the banality of winter. Wind chill readings make excitement out of mere inconvenience; they imbue a miserable day with the air of epic calamity. A temperature of 5 degrees is unpleasant. A wind chill of 20 below—well, that's something to talk about.
The gaudy negative numbers do more than describe the weather; they try to tell us how we experience it. The reporting of wind chill carries with it a paternalistic impulse to explain not just how cold it is, but how cold we'll feel. Well, I've been out in the cold every day this week, and I know exactly what it's like. If wind chill can tell me only what I've already experienced—my cell phone hand too numb to dial a number, my moustache freezing on my face—then we should just get rid of it altogether.
The weatherman's favorite alarmist statistic has been around for more than 60 years. Its ignoble history began with a pair of Antarctic explorers named Paul Siple and Charles Passel. In 1945, the two men left plastic bottles of water outside in the wind and observed the rate at which they froze. The equation they worked out used the wind speed and air temperature to describe the rate at which the bottles gave off heat, expressed in watts per square meter.
In the 1970s, the Canadian weather service started reporting numbers based on Siple and Passel's work. These three- and four-digit values meant little to the average person, however—the "wind chill factor" might have been 1,200 one day and 1,800 the next. American weathermen took a more pragmatic approach, converting the output from the Siple-Passel equation into the familiar language of temperature—statements like "it's 5 degrees outside, but it feels like 40 below." What exactly did these phrases mean? The meteorologists would figure the rate of heat loss in watts per square meter and then try to match it up to an equivalent rate produced in low-wind conditions. For example, the rate of heat loss in 5-degree weather and 30 mph wind matched up with the one for minus-40-degree weather and very little wind. So, 5 degrees "felt like" 40 below.
As the use of equivalent temperatures spread, people started to notice inconsistencies between real temperatures and their wind chill counterparts. For some reason, a day spent in a minus-40 wind chill was a lot easier to handle than a minus-40-degree day with no wind. Around 2000, two researchers—Randall Osczevski in Canada and Maurice Bluestein in the United States—began looking closely at this problem. Before long, they discovered that the adapted Siple-Passel equations grossly overestimated rates of heat loss.
The two nations' weather services formed a committee to address the problem. By 2001, the Joint Action Group on Temperature Indices had created a new system that toned down wind chill readings across the board. After the recalibration, conditions that were once said to feel like minus 40 now "felt like" minus 19. (Click here for a sidebar that explains how Osczevski and Bluestein came up with their new wind chill table.)
The updated model patches over the worst flaws of the old wind chill system, but it's not anything close to perfect. Osczevski and Bluestein made a set of new assumptions to determine wind-chill-equivalent temperatures. Namely, they geared their calculations toward people who are 5 feet tall, somewhat portly, and walk at an even clip directly into the wind. They also left out crucial variables that have an important effect on how we experience the weather, like solar radiation. Direct sunlight can make us feel 10 to 15 degrees warmer, even on a frigid winter day. The wind chill equivalent temperature, though, assumes that we're taking a stroll in the dead of night.
Even the variables that Osczevski and Bluestein did include might be wildly off base. Air temperatures tend to remain fairly stable throughout the day, but wind speeds fluctuate a great deal. (It's much less breezy in the morning and at night, for example.) Wind speed also varies depending on where you are. Obstacles on a city street—like buildings, cars, and kiosks—can block the flow of air and reduce its average speed. But wind-chill-equivalent temperatures use a single number to represent all this variability.
Other meteorologists have tried to work out more-involved schemes to account for these flaws. One group is even trying to combine every possible variable—temperature, wind, humidity, sunlight, and so forth—to create a universal weather index. (A researcher in Australia devised a model that takes into consideration an individual's height, weight, and style of dress.)
But no amount of tweaking will make wind chill more comprehensible. The language of "equivalent temperatures" creates a fundamental misconception about what wind chill really means. It doesn't tell you how cold your skin will get; that's determined by air temperature alone. Wind chill just tells you the rate at which your skin will reach the air temperature. If it were 35 degrees outside with a wind chill of 25, you might think you're in danger of getting frostbite. But your skin can freeze only if the air temperature is below freezing. At a real temperature of 35 degrees, you'll never get frostbite no matter how long you stand outside. And despite a popular misconception, a below-32 wind chill can't freeze our pipes or car radiators by itself, either.
The recent fiddling with wind chill has only made the numbers less useful. The old system might have overstated the numbers when it said that 5 degrees could feel like minus 40. But after three decades of practice, we all got pretty good at translating from the outrageous numbers in the weather reports to our own experience. When the weather service recalibrated the system in 2001, we had to start all over and rebuild our frame of reference from scratch.
Rather than trying to patch up wind chill's inconsistencies, we should just dump it altogether. The best algorithm we'll ever have for determining how cold it feels comes from our own experience. A look out the window gives us most of the variables we need to compute our own, personal weather index. The sight of a few leafy trees will tell us how windy it is on our corner and whether the breeze is swirling or gusting. We'll see if the sun is shining or if the sky is overcast. We'll also know how we're dressed, how tall we are, how much we weigh, and how quickly we walk down the street. We can even stick our hand outside for a moment, to get a sample of the ambient air temperature.
That's more than enough data to know how it might feel to step outside our front door. After all, our brains have been tallying up these variables for our entire lives. Weather reports can give us more specific information than we can get on our own, like predictions on what the wind and temperature will be in the future. But there's something absurd in the notion that the weatherman can tell us how we feel. Even the most rigorous meteorological model just mimics the one we build for ourselves.