Michael Vojtesak
National Weather Service Office
Goodland, Kansas





The NEXRAD Weather Service Office (NWSO) in Goodland, Kansas, provides forecasts and warnings for a three-state area. This area includes 13 counties in northwest Kansas and three counties in both far southwest Nebraska and extreme east-central Colorado.

Currently, the wind chill advisory and warning criteria differ across the county warning area. Under the pre- modernization system, in accordance with WSOM C-42, the Weather Service Forecast Offices (WSFOs) at Denver and Topeka adopted separate criteria for wind chill advisory and warning criteria than WSFO Omaha. In Kansas and Colorado, both WSFOs adopted the -20 degree F (-28.9 degree C) threshold for the wind chill advisory and -35 degree F (-37.2 degree C) for the wind chill warning. Meanwhile, WSFO Omaha adopted a policy of -35 degree F (-37.2 degree C) for advisories; -50 degree F (-45.6 degree C) for warnings.

In the past six years, the research literature available has shed considerable light on the public perception of wind chill and the debate surrounding the "quantification" of the wind chill index. This paper proposes a "consistent" wind chill advisory/warning threshold for the Goodland forecast area.



NWSO Goodland has participated in the transition process, in which full county forecast and warning responsibility was transferred from the "parent" WSFOs to NWSOs. Much of this transition process was associated with the Kansas Modernization and Restructuring Demonstration (KMARD). The inherent dilemma for NWSO Goodland management and forecasters will be how to express the state-to-state differences between wind chill advisory and warning thresholds to the public. This also may be quite confusing to the media. During polar outbreaks and/or widespread synoptic-scale wind events, conditions are usually uniform across the three state area.

With two different sets of wind chill criteria existing at the moment, there is a need to unify the criteria across our forecast area. This research proposes adopting a uniform "threshold" for wind chill advisories and warnings across state boundaries. The proposal takes into account climatological similarities based on hourly wind chill data for Goodland, Kansas, and temperature data for Benkelman and McCook, Nebraska.




The Wind Chill Concept

The wind chill index (WCI) was first developed by Siple and Passel in 1945. The "raw" derivation developed to calculate this index is given in Equation 1.

WCF is the wind chill factor. The term was originally designed to measure the freezing rate of one liter of water in a sealed plastic tube suspended at roof level. This is given in Equation 2.

This very basic conceptual model of wind chill unfortunately considered a small plastic tube having roughly the same geometry and surface conductivity of a human body! But it was a start.

In 1971, Steadman is credited with the first adaptation of the wind chill index based on the assumption that heat transfer through clothing and heat loss through breathing are important factors in defining what he called, "the wind chill equivalent temperature." At the time, Steadman worked in the Clothing and Textile Department at the University of Manitoba.

The Steadman article scientifically reasoned for the first time that the comfortable skin temperature (the original Ts of 33 degree C used in Siple and Passel's original WCI) should be 30 degree C based on the average percentage of exposed skin. Steadman showed that the face represents 3 percent of the body surface, with the hands and feet representing an additional 12 percent. He postulated that 85 percent of the human body is covered with the required resistance of clothing to maintain thermal equilibrium.

Figure 1. Wind chill as a function of air temperature for various wind speeds, or equivalent wind chill temperatures (Twc) (Steadman 1971).

Steadman postulated that sunshine effectively raises the ambient temperature by 7 degree C in a strong wind. Figure 1 and Figure 2 summarize the main points of Steadman's work. In both figures, stationary objects at 30 degree C in winds > 10 mph are used. It is an interesting note that the 10-mph guideline is used to this day as the baseline for advisory and warning criteria per WSOM C-42.

Figure 2. Thickness of clothing (mm) required too effectively insulate 85 percent of the body surface (Steadman 1971).

Schaefer (1988) and later Schwerdt (1993, 1995) discussed the Steadman equations in terms of the National Weather Service perspective. These papers discussed heat loss in terms of skin temperatures of the cheek, nose and ear, and developed a face temperature which is an average of the three surface areas. Schaefer postulated that the skin temperature of the face would be representative of a slow walk of between 30 and 60 minutes in length, and he admits that humidity and solar insolation were not considered. However, it is an attempt to develop an alternative to the skin temperatures (30 degree C or 33 degree C) commonly used in earlier models. Equations 3 through 6 and Figure 3 are based on the work of Adamenko and Khairullin (1972), and referenced in the Schaefer and Schwerdt articles. These papers all note that a wind speed of 10 knots and an ambient air temperature of -6.6 degree F (-21.4 degree C)---WCI of -32.8 degree F (-36.0 degree C) can cause frostbite during a short stroll. This currently provides a basis for the issuance of a wind chill warning per WSOM C-42 in Kansas and Colorado.


Tcheek = 0.4T - 4.3v + 53.4



Tnose = 0.4T - 4.3v + 49.6



Tear = 0.4T - 4.3v + 40.8



Tface = 0.4T - 4.3v + 47.9


Figure 3. Nomogram for determining skin temperature of the nose from indicated air temperature (C) and wind speeds (m s-1) after a half-hour to an hour of slow walking outdoors (Adamenko and Khairullin 1972).


 The "Comfort Zone"

From the first theory of "wind chill", it became apparent that the "comfort" of an individual was not a quantity that could be directly measured. Terjung (1966), Steadman (1979, 1984), Schaefer (1988), Rees (1993) and then Coronato (1995) all developed derivations of the first conceptual models of Siple and Passel as well as the adaptations of the early heat loss parameters developed by Steadman in 1971.

Somewhat related to the wind chill index developed in 1945 by Siple and Passel, a geographer from UCLA named Terjung, adopted a subjective "comfort" scale for the United States based on numerous climatic factors including temperature distributions, mean insolation values and humidity. He used the "wind chill" concept as a baseline approach to quantifying his "comfort" index. Adapted from Terjung, Rees (1993) also developed a "comfort" index based on heat loss in terms of W/m2 . Figure 4 shows this "comfort scale."

Figure 4. The "Comfort" Index. eT is the environmental temperature or ambient air temperature (Terjung 1966).

Much of the "debate" on wind chill centers around the index as a useful "measure" of comfort and whether or not wind chill addresses heat transfer theory. With this in mind, quantifying cooling rates and heat loss in terms of human "comfort" may be a better gauge for development of meaningful thresholds concerning wind chill advisories and warnings. Over-exposure to extremes of cold and wind has been studied at great length.

The values of 1400 kcal hr m2 and 1630 W/m2 express the critical "measurable" values commonly used to describe the onset of freezing flesh.

In its present form, the wind chill index accomplishes several things. It does describe the dangers of exposure to wind and cold. And it also allows the public to gauge the severity of the weather based on their own experiences within the climatic zone they inhabit. Coronato (1995) who studied wind chills in Patagonia concluded that "an accurate assessment of the bioclimatic environment requires the balancing of heat losses caused by wind chill and heat gains due to radiation."

In addition, he stated, "there is a true 'need' to assess each climatic zone for the effects of wind chill."

A study developed by Brauner and Shacham (1995) discusses an "upper" and "lower" limit of comfort and two new variables: exposed skin temperature (EST) and maximum exposure time (MET).

This study postulates that from the first moment of exposure skin temperature changes with time until it eventually reaches a steady-state value, or exposed skin temperature. It is this critical skin temperature, not the air temperature and wind velocity that will cause "frostbite." In addition, the maximum exposure time (true onset of frostbite) is a variable based on radiative exchanges and cooling rates of the human body.

The exposed skin temperature (EST) is therefore related directly to the transfer processes between skin and the elements and the maximum time period the skin is exposed. This concept of a maximum exposure "time" is expressed by Equation 7. In extreme conditions (cloudy skies, an air temperature of -50 degree C (-58 degree F) and wind velocity of 20 m/s) the exposure time can be as little as one minute. While on a calm, sunny day, the same air temperature may allow a human to endure for two hours.

Table 1 illustrates several "wind chill" indicators including wind chill (WCI), the equivalent wind chill temperature (Twc) and the exposed skin temperature (EST). Some values used to express the "comfort" index. An * denotes a corrected typo in the original graphic from Brauner and Shacham 1995.

In summary, the wind chill index does provide a measure of instantaneous heat loss from bare skin, but only at the moment the skin is first exposed. What Brauner and Shacham propose is an entirely different concept. For it attempts to "quantify" exposure in terms of human comfort with time. The wind chill index and the "wind chill" equivalent temperature do not "quantify" exposure over time. They simply assess a combination of current wind and temperature conditions.

Also, from Brauner and Shacham, Figures 5 and 6 respectively show the relationship between exposed skin temperature and wind velocity along with the relationship between maximum exposure time and wind velocity. What these researches have concluded is that . . .

"these 'new' variables introduce what can be best described as a measurable quantity that can be improved by objective experiments under well controlled conditions."

Figure 5. Exposed skin temperature in relation to wind velocity.

Figure 6. Maximum exposure time with and without solar radiation.


Current Debate

The current wind chill "debate" also revolves around the criticism that the underlying model does not comply with modern heat transfer theory. This part of the debate frequently includes topics such as effective wind velocities, radiative exchange and skin surface temperature. Likewise, other considerations such as type of clothing worn, breathing, calculated body surface areas (i.e., fingers, toes, etc.) and solar radiation all can provide a substantial impact to human comfort levels. Kessler (1993) provides the most poignant opinion as too why the wind chill index should not be used as a "measure" of human comfort.

"The National Weather Service and the larger meteorological community accept responsibility for presenting meteorological science to the public in useful form. Wind chill equivalent temperature as presently formulated and disseminated is not an example of best effort, and it is bad practice to present potentially misleading information."

For now, the wind chill index will continue to be used to "calibrate" human comfort until another acceptable paradigm is introduced. In spite of its shortcomings, it has proven to be successful in predicting the "comfort" zone of humans and alerting individuals to the onset of dangerous extremes of cold air and wind.



This research of the wind chill index as it applies to the current thresholds used for the Goodland, Kansas public service area considered the period November (1991-1996) through March (1992-1997). Between November and March, the forecast area is periodically affected by cold Canadian air masses and the strong synoptic scale weather systems that usher in this air.

In determining what would be the standard "threshold" across the entire Goodland forecast area, cold air temperatures and the constancy of the wind are the two most important factors. Steadman (1971) plus Brauner and Shacham (1995) have shown that heat loss from the body's surface plays a major role in determining critical exposure to these elements.

Heightening public awareness to the dangers of exposure (the combination of wind and cold) is what the current WSOM C-42 thresholds attempt to accomplish. Table 2 (and in particular the values to the right of the dark line) illustrates the amount of instantaneous heat lost from a number of possible wind speed and temperature combinations. The wind chill index can be related 'subjectively' to the wind and cold present, but rate of instantaneous heat loss is what is most important.

Table 2. Cooling Rates (NOAA, DOC, NWS, 72-15). To convert to W/m2 units, simply use 1.1645 W/m2 for every 1 kcal hr m2.

In addition to physiological factors, it is important to investigate the regional climatology of the Goodland forecast area in determining significant wind chill criteria. Table 3 shows the mean monthly high/low temperatures and wind speeds at NWSO Goodland for the past 30 years (NCDC 1990). What is important to note in Table 3, is that the MEAN monthly wind speeds for each month (November to March) are greater than the 10 mph (minimum) criteria necessary to issue a wind chill advisory.

Table 3. Mean Monthly Wind Speed (kts) and Temperatures (F) for NWSO Goodland, Kansas.

To further demonstrate the constancy of the wind, Table 4 shows the percent frequency distributions for selected wind speed categories. The data represents ALL hours combined for the period November through March. Table 4 shows a relatively high percentage of measured wind speed in excess of 10 knots regardless of the time of day. Approximately 40 percent of the time, the wind is strong enough to meet or exceed the 10-mph threshold for issuance of an advisory.

Table 4. Wind Velocity (kts)---Percent Frequency of Occurrence by Category (all hours) for Goodland, Kansas.

Although the perception of what "feels cold" in personal experience working outdoors is subjective, unification of the wind chill criteria for the entire forecast area should be considered. The "subjective" scale developed by Terjung and the heat loss needed (to cause frostbite injury) as measured in terms of exposure time, and critical exposed skin temperatures do correlate well with instantaneous heat loss rates in Table 2. Furthermore, WSOM C-42 guidelines require a minimum of 10-mph winds. In this study, wind chill calculations for Goodland, Kansas appears to be more dependent on air temperatures than wind speed. In effect, the wind speed serves as the constant with the air temperature serving as the dependent variable.

To illustrate this point, hourly wind chill frequencies at Goodland, Kansas are provided in Table 5. This data set utilizes approximately five years of hourly observations for the period November 1992 to March 1997.

Table 5. Wind Chill Values (F)---Percent Frequency of Occurrence by Category and Hour (LST), Goodland, Kansas.

Using five years of hourly data, Table 5 indicates that the wind chill advisory threshold was exceeded at Goodland, Kansas (wind chills of -20 degree F or less) only 2.95 percent of the time.

Tables 6 and 7 offer a mean daily high/low temperature comparison for Benkelman and McCook, Nebraska. Although a wind speed database is not currently available for these Nebraska sites, temperature climatologies show strikingly similar profiles to those of Goodland, Kansas. McCook and Benkelman should have similar hourly wind chill frequencies as Goodland because of their close proximity and similar temperature climatologies. It appears that constancy of the wind speed is well-established in this part of the country.

Table 6. Mean Temperature Statistics for Benkelman, Nebraska.

Table 7. Mean Temperature Statistics for McCook, Nebraska.

It follows from Tables 6 and 7 that there probably is no appreciable change in the "comfort" index along the Nebraska-Kansas-Colorado borders in the long-term since similar synoptic-scale conditions of wind velocity and air temperature exist. This lends to the notion that the warning and advisory thresholds for southwest Nebraska could be changed to the levels currently in place across Kansas and Colorado without significantly impacting public awareness of hazardous conditions.



Based on the climatological data for Goodland and the literature concerning the wind chill "debate," it appears that WSOM C-42 has a "reasonable handle" on wind chill as a tool to inform the public on issues concerning exposure.

Advisory and warning criteria are designed in WSOM C-42 to address conditions of exposure where the danger or potential danger exists. The thresholds for Kansas and Colorado seem valid based on physiological studies. Evidence of this has been provided in Figure 3 of this research.

Given that the current area of responsibility for NWSO Goodland incorporates two distinctly different advisory and warning criteria, we propose to have the wind chill guidelines for our southwest Nebraska counties changed to those currently in place across Colorado and Kansas.



I want to thank the library staff at NCAR in Boulder, Colorado, who took great care in assembling my literature search. Likewise, the management staff at NWSO Goodland, Kansas, whose constructive criticisms, guidance, and support made this research possible.



Adamenko V. N., and K. S. Khairullin, 1972: Evaluations of Conditions Under Which Unprotected Parts of the Human Body May Freeze in Urban Air During Winter. Bound.-Layer Meteor., 2, 510-518.

Brauner, N., and M. Shacham, 1995: Meaningful Wind Chill Indicators Derived from Heat Transfer Principles. Int. J. Biometeor., 39, 46-52.

Coronato, F.R., 1995: Wind Chill Influences on Thermal Conditions in North Patagonia. Int. J. Biometeor, 39, 87-93.

Kessler, E., 1993: Wind Chill Errors. Bull. Amer. Meteor. Soc., 74, 1743-1744.

____________, 1995: Reply: Wind Chill Errors. Bull. Amer. Meteor. Soc., 76, 1637-1638.

National Climatic Data Center, 1990: Daily Normals of Temperature, Precipitation and Heating and Cooling Degree Days, 1961-1990, Nebraska. Climatography of the United States No. 81, 18pp.

____________, 1990: Daily Normals of Temperature, Precipitation and Heating and Cooling Degree Days, 1961-1990, Goodland WSO AP. Climatography of the United States No. 84, 136pp

NOAA, DOC, NWS, 1972: Wind Chill. Central Region Technical Attachment 72-15, Scientific Services Division, Kansas City, MO, 2pp.

Rees, W., 1993: A New Wind-Chill Monogram. Polar Record, 29, 229-234.

Schaefer, J.T., 1988: The Effect of Wind and Temperature on Humans. Central Region Technical Attachment 88-05, NOAA, DOC, National Weather Service Central Region, Kansas City, MO, 2 pp.

Schwerdt, R.W., 1993: The Effect of Wind and Temperature on Humans-Revisited. Central Region Technical Attachment 93-04, NOAA, DOC, National Weather Service Central Region, Kansas City, MO, 10 pp.

____________, 1995: Comments on Wind Chill Errors, Part III. Bull. Amer. Meteor. Soc., 76, No. 9, 1631-1637.

Siple, P.A. and C.F. Passel, 1945: Measurements of Dry Atmospheric Cooling in Sub-freezing Temperatures. Proc. Amer. Philos. Soc., 89, 177-199.

Steadman, R.G., 1971: Indices of Wind Chill of Clothed Persons. J. Appl Meteor, 10, 674-683.

____________, 1979a: The Assessment of Sultriness, Part I: A Temperature -Humidity Index Based on Human Physiology and Clothing Science. J. Appl. Meteor., 18, 861-873.

____________, 1979b: The Assessment of Sultriness, Part II: Effects of Wind, Extra Radiation and Barometric Pressure on Apparent Temperature. J. Appl. Meteor., 18, 874-884.

____________, 1984: A Universal Scale of Apparent Temperature. J. Appl. Meteor., 23, 1674-1687.

____________, 1995: Comments on "Wind Chill Errors." Part I. Bull. Amer. Meteor. Soc., 76, No. 9, 1628-1630.

Terjung, W., 1966: Physiological Climates of the Conterminous United States: A Bioclimatic Classification Based On Man. Mon. Ann. Assoc. Geographers, 56, 141-179.


USA.gov is the U.S. government's official web portal to all federal, state and local government web resources and services.