Wind Chill Calculator

Enter the air temperature and the wind speed and the calculator returns the wind chill — the feels-like temperature winter forecasts quote — plus a frostbite risk line read from the National Weather Service chart. The number models how quickly exposed skin loses heat, expressed as an equivalent calm-air temperature; it is not the air temperature, and nothing outdoors ever cools down to it. The defaults, 20 °F with a 15 mph wind, return 6.2 °F, which the printed chart rounds to 6. Switching to °C and km/h reinterprets the boxes rather than converting them, so type metric values after switching.

How the wind chill formula works

Since 1 November 2001 the National Weather Service has computed wind chill with the formula from the interagency Joint Action Group for Temperature Indices, JAG/TI:

WC = 35.74 + 0.6215T − 35.75V^0.16 + 0.4275TV^0.16

where T is the air temperature in °F and V the wind speed in mph. The metric form of the same model, used by Environment and Climate Change Canada, is

W = 13.12 + 0.6215T − 11.37V^0.16 + 0.3965TV^0.16

with T in °C and V in km/h. Both are regressions of one physical model, and the exponent is exactly 0.16 in both. Per the NWS, the index is defined only for temperatures at or below 50 °F (10 °C) and winds above 3 mph (5 km/h), so outside that domain the calculator reports an error. The 3 mph floor is the NWS calm-wind breakpoint and the walking speed the model takes as its still-air reference, so at that speed wind chill lands close to the air temperature: 50 °F and 3 mph gives 49.7 °F. The formula expects the standard observation at 10 m (33 ft); the developers took face-level wind to be two-thirds of that, following Steadman's 1971 open-field analysis, so a handheld reading at face height corresponds to a formula input roughly 1.5 times larger.

A worked example against the official chart

Take the NWS example on its wind chill chart: 0 °F air with a 15 mph wind. The wind term is 15 raised to the power 0.16, which is 1.5423. The T terms vanish at zero, so the calculation collapses to 35.74 − 35.75 × 1.5423 = 35.74 − 55.14 = −19.4 °F. The chart prints −19, and the NWS notes that at this wind chill exposed skin can freeze in 30 minutes. A second cell checks the temperature terms: at 20 °F and 30 mph, 30 to the power 0.16 is 1.7232, and 35.74 + 12.43 − 61.60 + 14.73 = 1.3 °F, which the chart prints as 1. In metric, −10 °C with a 30 km/h wind gives 13.12 − 6.215 − 19.59 − 6.83 = −19.5 °C, which rounds to −20.

Bottles of water in Antarctica

The term wind chill was coined by Paul Siple, whose unpublished 1939 Clark University dissertation, Adaptation of the Explorer to the Climate of Antarctica, already contained a rudimentary first index. The published index came from measurements on the United States Antarctic Service Expedition of 1939 to 1941, on which Siple commanded West Base. There, as Randall Osczevski and Maurice Bluestein tell it in their 2005 history in the Bulletin of the American Meteorological Society, he and Charles F. Passel timed how long water took to freeze in a small plastic bottle suspended from a post on the roof of the expedition building. The results appeared in 1945 in the Proceedings of the American Philosophical Society, and the Wind Chill Index they defined was not a temperature but a cooling rate — a three- or four-digit number in kilocalories per square meter per hour that the same history calls the best-known result of a century of Antarctic research.

Water freezes faster than flesh, which generates its own heat, so, as the NWS puts it, the old index underestimated the time to freezing and overestimated the chilling effect of the wind. Osczevski and Bluestein catalogued the other faults: great variability in the data points, too little data at high winds, neglect of the container's own thermal resistance, and an assumed skin temperature that was too warm, compounded by 10 m winds, which blow significantly harder than at face height. By the mid-1970s equivalent temperatures had supplanted the raw index across most of North America, and the early ones ran spectacularly cold: the 2005 history records a 1964 Strategic Air Command table calling −1 °C with a 40 km/h wind equivalent to −40 °C in still air; Charles Eagan's redefinition of the still-air reference that same year raised the figure to −18 °C, and the 2001 model raised it again to −9 °C.

The 2001 revision

The revision crystallized at an Internet Workshop on Windchill, held the week of 3 April 2000 under Environment Canada and World Meteorological Organization sponsorship. In its wake the US OFCM established JAG/TI, chaired by the NWS, and the group designated Osczevski, of Toronto's Defence and Civil Institute of Environmental Medicine, DCIEM, and Bluestein of IUPUI to build a model of the exposed face. Heat loss is computed for the upwind side of a vertical cylinder 18 cm across, wind is corrected from 10 m to an adult face height of 5 ft, the person walks into it at 3 mph, and the sky is a clear night with no solar warming. In early summer 2001, twelve volunteers, six men and six women, walked treadmills in a chilled wind tunnel at DCIEM with thermal transducers on their faces. The NWS brochure says the results improved the formula's accuracy; the developers write that the trials' primary purpose was to fix one parameter, the thermal resistance of the cheek, and call the idea that they validated the model a misconception. Per the OFCM's report on the project, Canada's weather service began using the index in forecasts on 2 October 2001, with a ceremony on 30 October at DCIEM, and the NWS followed on 1 November. Reported wind chills became warmer, and at the weather services' request the charts were marked where frostbite on exposed skin is expected within 30 minutes; the printed NWS chart shades three bands of 30, 10 and 5 minutes. The bands rest on a facial-cooling model by Tikuisis and Osczevski, and the developers caution that frostbite risk does not track a fixed wind chill: skin cools faster in a strong wind at a moderate temperature than in a light wind at a very cold one.

What the number does and does not mean

Wind chill is a statement about skin. Wind strips away the warmed air next to any surface and speeds cooling toward the air temperature, but cannot push anything below it: a parked car on a −5 °C night reaches −5 °C sooner in a gale, never lower. Frostbite obeys the same physics — the NWS cold-weather FAQ states that the air temperature has to be below freezing for frostbite to develop on exposed skin — so this calculator reports a low frostbite risk whenever the air itself is above freezing. The bands deserve a similarly careful reading: per the JAG/TI implementation report they are drawn for the most susceptible 95th percentile of the population on a worst-case clear night, with a threshold skin temperature of −4.8 °C carrying roughly a 5 percent frostbite risk, so they are planning thresholds rather than personal countdowns, and nothing here is medical advice. A weather app's feels-like figure may blend in humidity; NWS wind chill uses temperature and wind alone, and humidity's counterpart is the warm-season heat index. Osczevski and Bluestein conceded that wind chill is not a neat and simple package, and that the equivalent-temperature format the public prefers is a deceptive simplification that only seems easier to understand.

Assumptions and conventions

The headline shows one decimal where the NWS chart prints whole degrees, so a one-degree disagreement with a printed cell is rounding. The chart spans 40 °F to −45 °F and 5 to 60 mph; beyond the printed grid the calculator simply follows the formula. The frostbite line is anchored on the NWS example of −19 °F: at or below it the calculator reports the band where frostbite can occur in about 30 minutes or less, folding the 10 and 5 minute shadings into one hedged statement, and metric results are converted to Fahrenheit for the comparison. Because the chart's shading does not follow a fixed wind chill, a result within a degree or two of −19 °F is a borderline call. Environment Canada publishes its own Celsius bands, reproduced by the Canadian Centre for Occupational Health and Safety: moderate risk from −10 to −27, skin freezing in 10 to 30 minutes from −28 to −39, and in under 2 minutes at −55 and colder; this page follows the NWS chart only. Because the boxes keep their values when the unit segment changes, metric mode shows the domain error until °C and km/h figures are typed in.

Frequently asked questions

At what wind chill does frostbite become a risk in 30 minutes?

The NWS anchors its chart with the example of 0 °F air and a 15 mph wind, which produces a wind chill of −19 °F; at that level, exposed skin can freeze in 30 minutes. The chart shades three frostbite-time bands of 30, 10 and 5 minutes, and per the JAG/TI implementation report they are drawn for a worst case, the most susceptible 95th percentile of the population on a clear night, with a frostbite threshold of about 5 percent risk, so the times are planning thresholds rather than personal predictions.

Why is wind chill only defined below 50 degrees Fahrenheit?

The NWS states that wind chill temperature is only defined for temperatures at or below 50 °F and wind speeds above 3 mph. The index models cold-weather heat loss from exposed skin, so outside that range the quantity it measures stops being meaningful, and this calculator shows an error instead of a number.

Can wind chill freeze pipes or cool a car below the air temperature?

No. Wind speeds up cooling toward the air temperature but cannot cool any object below it, so a pipe on a windy 35 °F night ends up at 35 °F, just sooner. The same physics protects skin above freezing: the NWS cold FAQ notes that the air temperature has to be below freezing for frostbite to develop on exposed skin, however low the wind chill sounds.

Why did wind chill numbers change in 2001?

The old index, built on Siple and Passel's Antarctic freezing-water experiment, overstated the effect of the wind; the JAG/TI implementation report says it made people think it felt colder than it really was, partly because it used wind measured at 33 ft rather than face height. The 2001 replacement models heat loss from an exposed human face and reports warmer values: one combination once called equivalent to −40 °C now comes out at −9 °C. Canada's weather service began using the new index in its forecasts on 2 October 2001 and the NWS adopted it on 1 November 2001.

What wind speed should I enter into a wind chill calculator?

Use the forecast or observed wind, which is measured at the standard 10 m anemometer height; the formula internally corrects that to face level. The JAG/TI model takes the wind at face level to be about two-thirds of the 10 m value, so a reading taken at face height, such as one from a handheld meter, corresponds to a formula input roughly 1.5 times larger.

Does humidity affect wind chill?

No. The NWS wind chill formula uses only air temperature and wind speed. Humidity's role in how weather feels belongs to the warm season, where the NWS publishes the heat index as the counterpart measure, and weather-app feels-like figures that blend in humidity are a different quantity from official wind chill.