Apparent Temperature

Wind Chill Temperature

The National Weather Service (NWS) Wind Chill Temperature (WCT) is based on the rate of heat loss from skin that is exposed to wind and cold temperatures. As the wind increases, the body is cooled faster, causing skin temperature to drop below that of the ambient air. In short, WCT is what it "feels like". WCT only applies to exposed skin. There is no impact on inanimate objects.

WCT is based on a "Human face model", with wind speeds from observations taken at the standard anemometer height of 33 feet reduced to typical face level, five feet above ground, and assumes no solar radiation impact.

WCT = 35.74 + 0.6215*T - 35.75*(V**0.16) + 0.4275*T*(V**0.16)

where WCT and T (air temperature) are °F and
V is anemometer wind speed in mph
http://www.nws.noaa.gov/os/windchill/index.shtml

WCT values in tabular format can be viewed at the above world wide web site. It is clear from this table that the possibility of frostbite or hypothermia are greatly enhanced if a person is exposed to cold and windy conditions, and the NWS is charged with issuing wind chill advisories and warnings at critical thresholds.

It should be noted that the regression is non-linear and is only valid for air temperatures at or below 50deg-F and wind speeds above 3mph. However, to build a comprehensive database of WCT values, albeit for extended winter months only, these limits have been stretched, but only to the point where calculated WCT is less than observed air temperature. If WCT should exceed air temperature then the unadjusted observed value is used in the database.

Hourly station data are obtained from the National Centers for Environmental Information's Integrated Surface Data (ISD) dataset (ftp://ftp.ncdc.noaa.gov/pub/data/noaa/isd-lite/). In this dataset observation times are UTC, but maximum, minimum and average wind chill temperatures are calculated for local standard time days.

The mean base period is 1981-2010, and the "feel cold" threshold is defined as the 15th percentile of daily January and February values in this same period.