Talk:Heat index

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Temperature extremes (Rated Start-class, Mid-importance)
WikiProject icon This article related to extreme heat or cold is part of the temperature extremes sub-project of WikiProject Meteorology and weather events, an attempt to standardize and improve all articles related to weather or meteorology. You can help! Visit the project page or discuss an article at its talk page.
Start-Class article Start  This article has been rated as Start-Class on the project's quality scale.
 Mid  This article has been rated as Mid-importance on the project's importance scale.

Discussion before 2006[edit]

I'm in the process of writing an application that is going to read all available data from the US National Weather Service and produce a report. As part of checking my code to make sure it was functioning correctly I implemented the first two algorithms on the page and compared the output to that of the KWeather program and also the data on The first algorithm matched within 2 degrees fahrenheit, while the second algorithm was consistently 10 degrees fahrenheit above the first (with a .2 degree margin). The code I wrote to test it is available. What I've been wondering since spotting this difference is: Is this difference known and (aside from the equation involving the matrix math) which is actually preferred? -- Dshadowwolf 02:10, 4 August 2007 (UTC)

The part I'm guessing at, in case anyone's wondering, is how they figure the numbers for the heat index. It might be that it's based on theoretical modeling of the thermodynamics of the human body, but I'm thinking they would have taken a "does it feel hotter in this room, or in that one?" approach, or something like it. The links provided don't seem to give any information how they arrived at the curves, so if anyone else can find anything more authoritative than my guesswork, help yourself. -- John Owens 05:22 May 9, 2003 (UTC)

Is this link answer your question? The mathematical forge:Humidex
MiguelTremblay 18:01, 22 May 2005 (UTC)
First of all, look for copies of "The Assessment of Sultriness. Part I: A Temperature-Humidity Index Based on Human Physiology and Clothing Science" by R. G. Steadman (published July 1979 in the Journal of Applied Meteorology, Volume 18 pp. 861-873 as well as "The Assessment of Sultriness. Part II: Effects of Wind, Extra Radiation and Barometric Pressure on Apparent Temperature" same author same volume, pp. 874-885.
Second, the formula that the NWS uses for its Heat Index calculations is a simplification of Steadman's work assuming things like a constant wind speed, a person of a certain height, etc. Basically they hold everything constant except the temperature and the relative humidity. Clegett 01:03, 16 July 2006 (UTC)

What's the difference between heat index and dew point?

I'm not an expert, but they are related but quite different. Roughly, the dew point is the temperature at which condensation forms, given the air's relative humidity and barometric pressure. The heat index is how hot it feels, depending on the temperature and relative humidy, because if it is humid, water can't evaporate from your skin as easily to cool you down. Generally, as the dew point goes up the heat index also goes up, but they are not linearly related. Bubba73 18:03, July 19, 2005 (UTC)
The heat index is defined by the National Weather Service as "the temperature the body feels when heat and humidity are combined." This is due to the fact that the higher the moisture content of the air, the slower sweat will evaporate off the body and therefore the slower evaporative cooling will take place.
On the other hand, the dew point is defined by the American Meteorological Society to be "the temperature to which a given air parcel must be cooled at constant pressure and constant water vapor content in order for saturation to occur." In other words, "all things being equal, how cold does the air have to be before it can't hold any more water vapor than it already does."
One is a measurement of how hot it feels, the other a measurement of atmospheric moisture content. Clegett 01:03, 16 July 2006 (UTC)

Where is there a similar table showing the comfort effects of Dew Point like the table in the Heat Index article?

All I know about dew points and comfort what is mentioned in the Dew point article. I've written a Windows program to calculate heat index, dew point, and wind chill. I will make it available soon as an external link, within the next few days. Bubba73 14:44, July 27, 2005 (UTC)
I've added an external link to my webpage where the program can be downloaded. Bubba73 03:05, August 1, 2005 (UTC)


The article formerly stated "The humidex was developed by Canadian meteorologists in 1965"; however, the highest humidex was recorded in 1953. I've removed the former statement. ᓛᖁ♀ 19:34, 15 August 2005 (UTC)

They could have calculated it retroactively. Bubba73 19:45, August 15, 2005 (UTC)
It was calculated retroactively; relative humidity was recorded even then, although it was measured using hair... Jeffrey 23:53, 15 August 2005 (UTC)
It was calculated by holding certain variables constant from the calculations contained in "The Assessment of Sultriness. Part I: A Temperature-Humidity Index Based on Human Physiology and Clothing Science" by R. G. Steadman (published July 1979 in the Journal of Applied Meteorology, Volume 18 pp. 861-873 as well as "The Assessment of Sultriness. Part II: Effects of Wind, Extra Radiation and Barometric Pressure on Apparent Temperature" same author same volume, pp. 874-885. Clegett 01:03, 16 July 2006 (UTC)

Danger zone[edit]

Many sources give a heat index of 130F as the start of "extreme danger", which probably comes from the US National Weather Service. I'm surprized that it is so high. Based on my experience with 105-110 heat index, I'd expect "extreme danger" to start at 120F or maybe 115F. Bubba73 23:48, August 25, 2005 (UTC)

The National Weather Service merely states that above a heat index of 130F heat stroke becomes highly likely with continued exposure. Clegett 01:03, 16 July 2006 (UTC)

I seriously question the Dhahran stats. I've visited Weather Underground several times just to check on Dhahran figures, and the humidity is never anywhere near 67 percent. I suggest either broken equipment or a typo and suggest the whole section of Dhahran info be removed.

More reason to question the July 8, 2003 Dhahran stats. According to Weather Underground, max/min/average relative humidity that day were 100/58/81 percent.

Let's compare that to a day before, a day after, a year before, a year after.

July 7, 2003: 84/15/47. July 9, 2003: 84/24/43. July 8, 2002: 70/10/30. July 8, 2004: 46/12/30. On all four of these dates, the AVERAGE relative humidity is less than the MINIMUM was on the allegely record-shattering July 8, 2003.

Remove the Dhahran comments, unless they can be substantiated by another weather organization measuring from ANOTHER SITE in Dhahran. —The preceding unsigned comment was added by (talkcontribs) 19:06, 16 July 2006.

I've seen dewpoints in the upper 80s and even low 90s F in other coastal cities in Saudi Arabia, on both coasts actually (Red Sea and Persian Gulf). Two cities that seemed to always be very steamy were Jeddah(June 2006) and Gizan (or Jizan)(June 2006). I'm sure I'm forgetting other locations though. The humidity here is highly dependent on wind direction, as some of the most humid conditions preceed a change in wind direction from the dry desert, whereas a wind from off the sea brings very humid conditions. Keep in mind that some of the highest open water sea surface temperatures in the world have been recorded in the Persian Gulf (which Dhahran is next to), in the mid to upper 90s F, so air coming off of this body of water could possibly have a dewpoint in the mid 90s when it reaches the coast.
It appears that the internet now has many references to this episode in Dhahran, some of which are clearly mirror sites, whereas others are probably not (e.g. USA Today). The source here is a valid one and to use your above analysis to try to invalidate it would be original research. However, if you feel that this issue is worth pursuing, I would suggest that you contact a meteorological agency to get them to change the data or something equivalent. I have done this before in the United States and it can be painstakingly slow, but it is possible (at least in the U.S.). I'm not sure if you could contact anyone in Saudi Arabia; perhaps you could try the World Meteorological Organization. Good luck. Ufwuct 21:51, 25 October 2006 (UTC)

Inverse relationship[edit]

This clause is confusing: thus there is an inverse relationship between maximum potential temperature and maximum potential relative humidity, therefore making, say, a simultaneous temperature of 120 °F (50 °C) and 90% relative humidity physically impossible. There cannot be an inverse relationship between two maximums. And it is not clear why the second part is proven by the first. I am not sure what this sentence is trying to accomplish, but if you do I urge you to rewrite it without the "relationship" language. -- cmh 03:43, 27 May 2006 (UTC)

I think what was meant by this is that high humidity and extremely high temperatures (50C and up) tend not to happen at the same time. From what I understand, places that get this hot are, most often, in arid desert-type climates, and all of the solar radiation heats the air, rather than heating surrounding bodies of water as well. Such heating of surrounding water would increase water vapor in the air, making it humid, but would have less tendency to overheat the air. In a sense, there is a relationship between max humidity and max temperature. Areas that are more humid tend to have more rain, which itself cools the air as it passes through it. Areas that receive little rain therefore have nothing, other than decrease in the intensity of sun rays (night time, winter time) to cool the air. In other words, it's almost as if the solar radiation is "split" between two things in humid areas (water vapour and air temperature) but focused solely on air heating in dry areas. For example, a city like Riyadh will get temperatures easily above 45C many times throughout the summer. The dewpoints however, tend to stay very low. As I look now Riyadh has a temperature of 42C and a dewpoint of -7C. Very dry! As I look at Miami, the temperature is 29C, but the dewpoint is 24C. It's obvious that much more solar radition goes into heating the water surrounding Miami than it does heating the air itself. (for the record, miami has never recorded above 40C) —Preceding unsigned comment added by (talk) 15:00, 22 June 2008 (UTC)


this article doesn't sound like encyclopedic entry. i am not a native speaker of english and can't fix it properly. the informative aspects are good, although occasionally confusing. thus i put up the cleanup tag.

Heat index = humidex?[edit]

I've added a disputed tag because I don't think heat index and humidex are the same thing. The NWS heat index is calculated with temperature and relative humidity, while the Environment Canada humidex is derived from temperature, dewpoint and vapor pressure ( -- Mwalcoff 01:37, 1 June 2006 (UTC)

Side note on that: Relative Humidity can be calculated by using vapor pressure and a constant. They could be basically the same thing going through slightly different math.

IMO this isn't a dispute, this is you not liking a sentence in the article. (A dispute is where two parties can't come to an agreement.) It makes no sense to tag this whole article as disputed just because you think "heat index and humidex" are not synonymous... the rest of the article is fine. I have reworded the first sentence to address your concern and removed the tag. In future, if you think something is wrong please simply edit the article to improve it, rather than putting dispute tags on things. -- cmh 03:00, 16 July 2006 (UTC)

I'm not a meteorologist and was hoping that experts could fix the article. I've removed the references to the humidex until this can be sorted out. -- Mwalcoff 03:05, 16 July 2006 (UTC)

The Humidex is not the same as the heat index - it is based on vapor pressure and dew point only whereas the Heat Index takes multiple factors into account. The Humidex is not scientific and exaggerates the effect of humidity at low temperatures and underestimates at high temperatures. It is used in Canada - unfortunately the committee to develop a common index for both US and Canada has dissolved.Deajohn1 (talk) 17:36, 19 June 2012 (UTC)

I agree. Having travelled only in the US and Canada, but extensively in each, I have only heard "heat index" used in the US, and "humidex" used in Canada. The "humidex" values tended to produce higher values for the same situations on the border, when you do the conversions between the "Fahrenheit" heat index and "Celsius" humidex. I doubt that "humidex" is ever used to mean "heat index" today in the US. But supposing it is, I simply highlighted that "humidex" has multiple definitions, and linked to the Canadian one, which I'm certain is what most people will be looking for if they look up "humidex". (talk) 14:56, 18 July 2013 (UTC)


The top of the page mentions how the "factual accuracy of the article is disputed". I don't see any ongoing dispute here on the discussion page though, what am I missing?

See the comment above yours. -- Mwalcoff 23:06, 6 July 2006 (UTC)
I agree that humidex does not seem to equal heat index. The formula given for heat index does not match the one posted on the Canadian Weather Office page here
The Humidex formula is based on the work of J.M. Masterton and F.A. Richardson at the Atmospheric Environment Service (now MSC) of Environment Canada in 1979. It is a standard for Canada, but variations are used around the world. The dew point temperature should be given in kelvins (temperature in K = temperature in °C + 273.1) for the formula to work. The magic number 5417.7530 is a rounded constant; it's based on the molecular weight of water, latent heat of evaporation, and the universal gas constant.
e = vapour pressure in hPa (mbar), given by:
e = 6.11 * exp [5417.7530 * ( (1/273.16) - (1/dewpoint) ) ]
h = (0.5555)*(e - 10.0);
humidex = (air temperature) + h
--DRead 00:37, 12 July 2006 (UTC)

Confusing language[edit]

I can't tell what this means:

"...essentially the same temperature colder than which wind chill is thought to commence"

I'd fix it myself if I knew what the author was trying to say. 04:28, 24 July 2006 (UTC)

Heat index at high temperature and humidity[edit]

I was just wondering if anyone had any information as to why heat index tables, such as the one on NOAA's webpage: [1] don't consider the result of high humidity at high temperatures. For example, I'm currently in Qatar and the temperature is about 113F(45C) with a humidity of around 90%. As I've read in some of the other comments, I'm sure some would say this isn't possible, but believe me it's true. Using a seemingly unofficial online calculator at: [2] these values yield a heat index of: 272F(133C). Realizing that even though it's really hot here, this isn't possible, why isn't the equation designed to consider extremes such as this? Any information would be appreciated.

All I can say is, no, it's not possible. Something is wrong with the humidity reading. It may have been 90% humidity in the morning when the temperature was lower, but it won't be 90% humidity at mid-day because the amount of moisture in the air would have to increase drastically for that to occur, and that just doesn't happen in the real world. Haplolology 19:40, 25 May 2007 (UTC)

Yes - relative humidity is always higher in the morning but decreases when the temperature goes up during the day (usually). The thing to understand is that the dewpoint of the air (the temperature at which the air would have 100% humidity) tends to stay fairly consistent during short periods of time, given no other meteorological events (rain, intense storms). If the dewpoint is 15 at dawn, it will probably hover around 15 (maybe going up or down a little bit) throughout the day. This is why humidity is usually higher in the morning and less in the afternoon. Ex 18C with 15C dewpoint= 87% humidity. 45C with 15C dewpoint= 29% humidity —Preceding unsigned comment added by (talk) 14:00, 23 June 2008 (UTC)

Vapor Pressure Units[edit]

The article defines V, "vapour pressure", but doesn't specify the units to be used. ?mmHg ?mbar

'Tis done ʍαμ$ʏ5043 17:33, 30 May 2007 (UTC)


I removed the hypothetical about Kent because the humidity was nowhere near 50% that day at the time the 101.3F was recorded. If it had been, the dewpoint would have been about 80F. In London, on the same day (8/10/2003), the dewpoint was in the low 40s Fahrenheit and when the temperature was 98.6F, yielding a relative humidity below 20% and a heat index of about 95F.[3] So, there's no reason to get into hypotheticals. Ufwuct 21:05, 25 October 2006 (UTC)

Weather Underground[edit]

All I want to say is that I would take any reading coming from Weather Underground very very cautiously, even if it looks correct. I could give you a list of dozens of readings and long-term averages that anyone should be able to see are plainly wrong but which they list as officially correct. In their defense, however, they have tagged many of the more unreliable stations as being "not official NWS values". Haplolology 19:43, 25 May 2007 (UTC)

I would definately agree with this, it appears that their local reading station gets fully exposed to sunlight in the mid afternoon resulting in it often reading 10 F higher than NWS / NOAA. Jon (talk) 14:35, 6 March 2009 (UTC)


Are the terms "heat index" or the colloquial "feels-like temperature" used otuside of the U.S.? -FZ 15:03, 12 July 2007 (UTC)

Here in the UK, "feels like" values are occasionally given in winter, but that's about it. Our generally mild climate means that we hardly ever experience conditions where a precise wind chill/heat index value is really important. (talk) 00:57, 28 July 2008 (UTC)

WikiProject class rating[edit]

This article was automatically assessed because at least one WikiProject had rated the article as start, and the rating on other projects was brought up to start class. BetacommandBot 09:53, 10 November 2007 (UTC)

At least one formula is horribly wrong[edit]

At least one of the formulas must be horribly wrong, as a simple check of the value for T=0, H=0 shows. The first says it's in Fahrenheit, but the second formula wouldn't make sense if it were in Celsius either, because -42.379°F just isn't anywhere near 16.923°C. — Sebastian 21:57, 13 December 2008 (UTC)

You might check to see if the formula(s) would make since if the temperature were in Kelven units as well since that is also a common unit by mathematicans. Jon (talk) 14:32, 6 March 2009 (UTC)

The formulae are only considered accurate above 80 degree and 40 % humidity. Values in that range are relatively accurate (talk) 23:34, 31 March 2009 (UTC)Fiona

Heat Stroke versus Sun Stroke[edit]

Heat Stroke and Sun Stroke are the same thing, both being Hyperthermia. Where essentially, the bodies ability to regulate it's own temperature becomes overwhelmed and fails. In the article it reads the following under Effects of heat index (shade value)

Danger — sunstroke, heat cramps, and heat exhaustion are likely; heat stroke is possible

I will correct this quote to correct the statement that implies that sunstroke occurs at one stage and heat stroke may be possible after. Medic48 (talk) 03:43, 12 May 2009 (UTC)

Miami/Phoenix comparison is inaccurate, heat index criticism[edit]

First of all, what range of dates or months does "summer" cover? Also, according to, the average humidity for Phoenix in July is 24%, nowhere close to 10%. Also, Miami is at 67%, not 75%. The temperatures are more or less correct. If you run the calculations, Phoenix's heat index on a typical July afternoon would be several degrees higher than Miami.

A criticism of the heat index in general is that it assumes you are in the shade. This happens a lot less frequently in a place like Phoenix than Miami. Hence, the heat index makes Phoenix look a lot better than it should.

Kuvopolis (talk) 21:31, 17 July 2009 (UTC)


I converted some degrees farenheit values to degrees celsius. I hope that relative heat can be converted. No formulae were provided for the original figures. See my IP's edits. -- (talk) 18:47, 12 February 2010 (UTC)

Formulae: 16 term polynomial?[edit]

This whole heat index concept is only a crude approximation, so to calculate more than even a few terms is specious. (talk) 20:42, 10 March 2010 (UTC)

When HI should be calculated.[edit]

"The heat index is calculated only if the actual temperature is above 27 °C (80 °F), dew point temperatures greater than 12 °C (54 °F), and relative humidities higher than 40%[4]."

This statement is contradicted by NOAA:

NOAA says DP should be 60F —Preceding unsigned comment added by PeterK2003 (talkcontribs) 22:51, 6 April 2010 (UTC)

The NOAA lists a much more comprehensive formula (that does not give absurd results at low temperatures) in the client side JS code. This should be reflected in the article. As of now, the formula shown will produce 98 as a heat index for a temperature of 50 and %50 humidity. — Preceding unsigned comment added by (talk) 05:12, 13 June 2012 (UTC)

Finnish Sauna Example is Incorrect[edit]

The heat in Finnish sauna is from 60-100 C. Take if from me, I've lived in Finland for almost four decades and we sure don't go into sauna before it's at least 60 C, preferably 80 C. 100 C is uncomfortable to most people.

So please fix that part of the article. I'd do it myself, but can't comment on the actual heat index, because I know so little about it.

Please not that I only talk about temperatures in the above, not heat index. —Preceding unsigned comment added by SuurMyy (talkcontribs) 17:09, 11 July 2010 (UTC)

Definitely. Sauna temperatures are usually around 70...140 °C, humidities 5..30%. If someone finds an extension of Heat index to cover this range it could be reported here. Relative humidity of 100% and a temperature of 38 °C would make it a Roman caldarium. Turkish hammam is slightly less humid, but changing the heading "Turkish bath example" would be close enough. (talk) 07:54, 1 August 2010 (UTC) timo

Not only that, but heat index values for the Finnish Sauna example don't match with those provided by the NOAA Heat Index Calculator: so I am changing them to the NOAA values. —Preceding unsigned comment added by (talk) 20:14, 23 December 2010 (UTC)

Wait, wouldn't you die having your whole body at over 70°C, even only for a few minutes, or maybe you are talking about the heat index of those saunas (talk) 06:53, 11 April 2014 (UTC)

Incorrect example of formula and a broken link[edit]

The last formula given, the simple one that gives a very rough approximation, is either incorrectly stated, or the example usage is wrong. The formula given is HI = 1.9 * T * R + 10. The example uses T=90 and R=73% and gives a result of 124 F. The only way to wrench that result from the formula is to treat R as a fraction (it was defined as expressed in percent in the prior two formulas) AND to ignore the constant 10. Should the formula say R/100 rather than R? Should the example result be 134? Since I'm a reader of this article, not an expert, I can't correct it, especially since the link given twice in the article to the site that details the various formulae is broken ( So I can't even check if the experts there give the correct version of the formula (in which case I'd do the edit myself). Could someone knowledgeable please fix the link and the formula? —Preceding unsigned comment added by Paulalancaster (talkcontribs) 22:42, 12 August 2010 (UTC)

I found an online heat index calculator elsewhere on NOAA's site: If someone is interested, they could parse the Javascript source to determine the equation. These two links ( and found at the bottom of seem useful, but they are also dead, unfortunately. Vidalian Tears (talk) 19:14, 26 August 2010 (UTC)

The wikimedia tool provides a template for heat index. That template based upon a polynomial that has more terms. I've added the new formula to the article and credited Richard Stull. This seems to be the more reliable formula. gregh3285 (talk) 23:40, 08 July 2012 (UTC) — Preceding unsigned comment added by (talk)

Wind vs Heat index[edit]

I found the following discussion of the heating and cooling effect of wind on HI. -- which I leave here for others more-better than I.

 HalFonts (talk) 21:18, 17 June 2011 (UTC)

Examples needed[edit]

There is a loooong formula for calculating HI, but no examples. For the common user, this formula (using lots of extended small values) in effect means that the article is a dead-end in terms of getting actual HI values. How about a table with some values? Kdammers (talk) 00:44, 31 July 2011 (UTC)

Done! Thanks for the suggestion. Duoduoduo (talk) 16:17, 31 July 2011 (UTC)
Great! That really helps. Kdammers (talk) 05:21, 1 August 2011 (UTC)

Function of (dew-point or relative-humidity) -- HI vs Humidex Wikipedia articles[edit]

In this Wikipedia Heat Index article it states: "The humidex differs from the heat index in using the relative humidity rather than the dew point." In the Wikipedia Humidex article the same sentence is the opposite: "Humidex differs from the heat index used in the United States in being derived from the dew point rather than the relative humidity." Both can not be correct. I'm no expert, so I made no changes. One needs fixin. (22Dec2011) HalFonts (talk) 05:22, 23 December 2011 (UTC)

Scaling error of "R" in first formula[edit]

Just below the first HI formula is the phrase "R = relative humidity (decimal fraction of 1.00, i.e. percentage/100)". I can't get results from this formula to come out anywhere close to the NOAA table shown in the article, unless I use "R" values scaled as percentages (i.e. 0-100). Then the results match within +/- 1 degree F, as far as I have checked. I also confirmed the table's results match this online calculator: .

Reference #8 (, at the bottom of pg. 2, states "R = relative humidity (integer percentage)."

I didn't look into the scaling of "R" in the second formula (with the 16 terms.) BBmex (talk) 20:07, 16 February 2013 (UTC)