A Window on Water Vapor and Planetary Temperature

Here is some interesting news; according to data from NOAA’s Earth System Laboratory, atmospheric water vapor is on the decline globally.

You’ve probably heard many times how water vapor is actually the  most important “greenhouse gas” for keeping our planet warm, with an effectiveness far greater than that of CO2.

It is generally accepted that the rank of important greenhouse gases is:

  • water vapor and clouds which causes up to 70% of the greenhouse effect on Earth.
  • carbon dioxide, which causes 9–26%
  • methane, which causes 4–9%
  • ozone, which causes 3–7%

Note the range of uncertainties, on water vapor some say the percentage goes up to 90% with reduced numbers on the other three.

It is absolutely true that water vapor is the gas most responsible for the “greenhouse effect” of our atmosphere. Greenhouse gases let short-wave solar radiation through the atmosphere, but impede the escape of long-wave radiation from the Earth’s surface. This process keeps the planet at a livable temperature: Without a suitably balanced mixture of water vapor, CO2, methane, and other gases in the atmosphere, Earth’s average surface temperature would be somewhere between -9 and -34 degrees Fahrenheit, rather than the balmy average 59 degrees it is today.

This graph then from NOAA’s Earth System Research Laboratory, showing specific humidity of the atmosphere up to the 300 millibar pressure level (about 8 miles altitude) is interesting for it’s trend:

Click for original source of the graph

[UPDATE2: After reading comments from our always sharp readers, and collaborating with three other meteorologists on the graph, I’m of the opinion now that this graph from ESRL, while labeled as “up to 300mb only” is misleading due to that label. The first impression I had would be from the surface to 300mb i.e. the “up” portion of the label, but on the second thought I believed the label was intended to be numerical meaning “zero to 300mb” or from the top of the atmosphere down as opposed from the surface up as we normally think of it. The values looked like anomaly values, but are inthe range of absolutes for that elevation also.

Thanks to some work by commenter Ken Gregory, looking at other ways this and similar graphs can be generated from the site, it has be come clear that this is a level, not a range from a level. The label ESRL placed “up to 300mB was intended to list the availability of all data levels. Thus there is no 200mb data.

This demonstrates the importance of labeling a graph, as without any supplementary description, it can be viewed differently than the authors intend. A better label would be “at 300mb” which would be unambiguous. ESRL should correct this to prevent others from falling into this trap.]

For some background into atmospheric absorption efficiency of the electromagnetic spectrum, this graph is valuable:

Note the CO2 peak at 15 microns is the only significant one, as the 2.7 and 4.3 micron CO2 peaks have little energy to absorb in that portion of the spectrum.  But the H2O (water vapor) has many peaks from .8 to 8 microns, two that are fairly broad,  and H2O begins absorbing almost continuously from 10 microns on up, making it overwhelmingly the major “greenhouse gas”.

Here is another graph looking at it in a different way:

Click for a larger image

Note that water vapor plays quite a role in keeping the planet cool by absorbing some percentage of incoming radiant energy.

The yellow line is what we’d get without an atmosphere, and the blue  line what we get with it. Sunlit temperatures on the earth’s surface are substantially less than those on the moon (up to 123°C) because our atmosphere intercepts some incoming solar short-wave radiation as well as some outgoing long-wave infrared.

So when we see atmospheric water vapor dropping as shown in the NOAA ESRL graph above, you know it has to have an effect on our overall planetary energy budget, the question that will be argued is; “how much”?

h/t: Thanks to atmospheric physicist Jim Peden,  and also to Barry Hearn, and Alan Siddons for some of the graphs and background to this post.

UPDATE: See part2 of this post here


newest oldest most voted
Notify of
Bill Illis

Thanks Anthony,
I’ve been looking for this data for a long time. If possible can we see other levels of the atmosphere.
Generally, this is another area where the models are wrong as the assumption is that specific humidity remains constant. And this is a key assumption to the sensitivity estimates for GHGs.
It seems the models are getting too many specific assumptions/outputs wrong such as tropical troposphere, Antarctic, and temperature trends, in general, to be reliable.


Doesn’t decreased water vapour fly in the face of positive feedback in global warming theory.
I seem to recall that the worry from increased CO2 is that it causes positive feedback of increased water vapour, and therefore could cause run away global warming.
This seems to contradict that as CO2 levels continue to rise and now water vapour levels are decreasing.


Greenhouse gases let short-wave solar radiation through the atmosphere, but impede the escape of long-wave radiation from the Earth’s surface.
Sunlit temperatures on the earth’s surface are substantially less than those on the moon (up to 123°C) because our atmosphere intercepts incoming solar short-wave radiation as well as outgoing long-wave infrared.
I must be missing something because in the beginning you said greenhouse gases let through short-wave, and then later said they intercept short-wave.
On one hand they warm the planet (pass through solar short-wave and absorb earth-radiated long-wave), and on the other they cool the planet (intercepts incoming solar short-wave).
You’ll have to clear that up for me.
REPLY: The let in some, they intercept some. Its two way partial mirror, otherwise we’d never see stars or stay warm


raypierre @ realclimate talked about anthropogenic GHG’s and strong water vapour feedback in 2005
The authors then subtract off the part of the downward infrared radiation increase attributable to temperature and water vapor increase, and thus estimate the part due directly (as opposed to via feedbacks) to the increase in anthropogenic greenhouse gases such as CO2. They estimate this to be about one third of a Watt per square meter. This is not in bad agreement with estimates from detailed radiation models run by the authors, which say that the change in surface radiation due to the 12ppm CO2 increase between 1995 and 2002 should be about one fourth of a Watt per square meter. It is striking that the changes in the Earth’s surface radiation budget due to anthropogenic greenhouse gases are so profound that they can be directly observed on a regional scale, over such a short time period. So far, so good. Physics seems to be working as it should, and climate scientists seem to be basing their understanding of climate change on rock-solid physical principles….I’m wondering if there needs to be some rethinking on this.


Very interesting. I have a related question. Anthony explained the greenhouse gas very well. If GHGs absorb long wave radiation reflected back toward space, isn’t there a diminshing return as GHG concentrations increase? Think of it this way, I have a window film that blocks 50% of the light. The first sheet I put on the window reduces the visible light by 50%. The next sheet only 25% (50% of 50%), the third 12.5%, the fourth 6.25% and so on.
If that analogy is right, we would expect that the first 100ppm of CO2 to produce the MOST warming. Is there something wrong with my reasoning? At some point adding more GHGs will have almost no effect on the energy absorbed.
REPLY: Your reasoning is spot on, CO2’s effectiveness as long-wave reflector with concentration is logarithmic


That graph is for the global average. It seems that regional patterns would be important to know before drawing conclusions.


Its probably not reasonable to expect the models to match short-term trends. They are not built that way.
We now have roughly 30 years of satellite data though, but I mostly see matches to Hadley or GISS “adjusted” data. The matches there appear to be reasonable for some model runs (the ones that assume CO2 reduction or stasis).
My trust of GISS is somewhat compromised (can’t image why) and MSU data has its own problems, but these pale compared with the surface station problems.
So who is working with MSU data to calibrate their models?


I haven’t looked into this too deeply, but I can’t help noticing that the water vapor graph looks somewhat like the sunspot graphs that we’ve seen. If there was some sort of correlation of sunspots to water vapor that might go a long ways in explaining why a low amount of sunspots or a quiet solar cycle seems to create mini ice-ages.
This is just off the top of my head, so don’t flame me if it actually sounds silly. Just thought I’d put it out there.


“Your reasoning is spot on, CO2’s effectiveness as long-wave reflector with concentration is logarithmic”
As would water vapor, right?
“That graph is for the global average. It seems that regional patterns would be important to know before drawing conclusions.”
Why would regional patterns matter when we’re talking about GLOBAL warming?

This is an amazing assertion by NOAA…. And based on past manipulations by NASA/NOAA, I find it VERY suspect and stinks like a hog farm. For openers, why did NOAA wait so long before announcing the drop in humidity? According to the above chart, the big drop began around 1958.
For the past year, I’ve spent hundreds of hours pouring over research papers on the various forms of water (solid, liquid, and vapor) in our atmosphere for an article on “The Mysterious Climate Project.” Quite frankly, each time I think I have a handle on it along comes another research paper or scientist with still another figure. The above article states water vapor and clouds cause up to 70% of the greenhouse effect on earth. Before I go on, blog members should realize the “volume” and the “effect” are two different animals! You can have all the volume you want (Argon for example), but if it has little or no effect it’s immaterial.
During the course of my research, I’ve received dozens of papers and letters from well known atmospheric physicists and climatologists who believe the “effect” figure should be closer to 95%. So how does this square with the above? I don’t know. Are some of my figures already outdated? Possibly. The truth of the matter seems to be that climatology is such an infant science, we essentially know VERY little of the subject. For the simplistic Pogies to run around like a bunch of chickens with their heads cut off screaming the sky is falling, simply shows how intellectually challenged they really are.
As the Pogies continue witnessing their AGW claims go up as so much smoke, it appears they are embarking on a crusade of minimizing every other variable associated with the climate in order to make CO2 look more important than it is.
If I were young and just entering college, my major would be atmospheric physics. It is virgin, it is challenging, it is intriguing, and it is the wild wild west all over again! To me, atmospheric physics and its associated fields remind me of that knock ’em dead, good looking redhead with a tight butt and a sexy swish in her walk! Okay, okay, maybe not that much, but close to it!
Jack Koenig, Editor
The Mysterious Climate Project
REPLY: There was no “announcement” per se, this was a data plot from the ESRL website that became self evident from the recently released data and upgraded web site that allowed such examinations by members of the public.


Henrik Svensmark has written extensively on the relationship between the sun’s magnetic field and cloud formation; unfortunately, I’m out the door for an appointment and can’t post any links now.

retired engineer

How can this be? Burning all that gas, oil and coal which makes the evil CO2 that warms the Earth also produces vast amounts of water vapor. So it can’t decrease.
Unless human activities aren’t such a big deal.
A bit more OT: It looks like we are close to saturation on inbound absorbtion (for CO2). How close to 100% are we on outbound radiation? Shouldn’t be much at near IR, what about the 15u stuff?

Bill in Vigo

This is a Question.
If we assume (I hate that word always gets me in trouble) that water vapor is at 1000 ppm and that it falls by 50% to 500 ppm the current thought that it causes warming then the warming caused by water vapor would decrease by 12.5% causing lower water vapor content as cooler air tends to be drier. then the air could lose another 50% down to 250ppm and this could cause an additional 25% cooling causing another reduction of temp causing the air to condensate reducing water vapor again if another 50% we lose an additional 50% ability to warm and so forth. I wonder if the modelers put this in their computations. I know that the figures I am using are not correct but just food for thought. It would seem that with the sun possibly having influence on water vapor our problem might be run away cooling not runaway warming. (Ice ages again) I know that my figures are exaggerated but If Gore can as truth why can’t I in a question while admitting that they are exaggerated?
Come on Sun
Bill Derryberry


Ok, GHGs let through some short-wave (visible and very-near infra) and absorb others. They keep in/out the strong majority of the infrared bands, and H2O keeps in/out just about all the microwave range stuff.
Cool, I guess I just got confused on terminology. Blocking a small portion of the short-wave stuff doesn’t sound like it does much “cooling”. Maybe it just doesn’t warm as much as it might, which I guess is pretty much the same thing as having a cooling effect.
Would you happen to have any charts that show the bands of radiation that the Earth puts out? Does Earth radiate infrared pretty uniformly across wavelengths, or does it radiate in particular wavelengths?
I could see a significant difference in effects if Earth happens to primarily radiate at the 6 micron wavelength of infrared where H2O is the only thing that really keeps that in, compared to if Earth radiates more at 4 microns where it’s primarily CO2 that keeps the radiation in.
Or Earth may radiate in a smooth grade of wavelengths, and we can do relatively easy average-calculation of what effect the dropping H2O might have on keeping in infrared radiation.


Anthony, you say “Without a suitably balanced mixture of water vapor, CO2, methane, and other gases in the atmosphere, Earth’s average surface temperature would be somewhere between -9 and -34 degrees Fahrenheit, rather than the balmy average 59 degrees it is today.”
Actually, in his book Roy Spencer has recently reminded people of a paper from 1964 by Manabe and Stickler, where they calculated the surface temperature with major greenhouse gases and natural convection at about 140 F. The lapse rate is also almost double what is actually observed. It is the resulting weather that accelerates the transport of heat from the surface to the top of atmosphere, reduces the lapse rate to the observed values of about 6.5 K/km, and regulates the surface temperature at around 59 degrees F. It is a subtle difference, but shows that weather processes are a huge stabilizing feedback mechanism to the global climate. You don’t need much of a change in cloud cover or precipitation efficiency to have a dramatic impact on surface temperature.
REPLY: Correct, weather is part of the global heat transfer mechanism, hurricanes and thunderstorms do quite a bit of this transport.

Werner Weber

Pardon me, I do not understand, what the globally averaged relative humidity really means. E.g.; values of order 0.2 and somewhat less imply clear sky situations all year long. Or try to estimate the latent heat content in the atmosphere from these numbers, by using globally averaged mean temperatures. You probably get enormous fluctuations which do not make any sense. Humidity saturation increases exponentially with temperature. How meaningful is it to just take the average over exponential functions? Finally, it is the absolute humidity which determines any water vapor greenhouse effect. It is also the absolute humidity, which determines the amount of latent heat, transported into the upper troposphere by convection and which, according to R.S. Lindzen may more or less bypass the blocking of radiation cooling by the greenhouse gases.


Anthony — I believe you are wrong on the primary reason for higher peak temperatures on the moon. While the earth’s atmosphere (clear sky) absorbs about a quarter of incoming solar radiation before it hits the surface (~1365 W/m^2 at the top of the atmosphere, ~1000 W/m^2 at the surface), remember that this energy is still absorbed in the earth “system”.
Remember that the moon’s day/night cycle is a month long in earth terms, so its “daylight period” is two weeks long. This gives plenty of time to get the surface cooking. Think of how hot the earth’s surface would get in the afternoon of a month-long day.
Time-averaged, the moon is cooler than the earth, one of the key pieces of evidence saying there is a natural greenhouse effect on earth.
REPLY: Excellent point, one which I hadn’t considered. That duty cycle does indeed change the issue.

anna v

This is very interesting. As the basic CO2 incriminator is the feedback mechanism between water vapor and temperature rise this is another nail on the coffin of anthropogenic CO2. One does not need a sophisticated analysis to disprove a correlation with the monotonic CO2 rise of the last decades.
It is amazing that they were sitting on such data and never checked their feedback hypothesis.
Thanks for finding it for us.

Do you have a link for the original data for this? I’ve found lots of huge gridded datasets, but nothing as simple as this annual/monthly global mean.
REPLY: This is from web plotting automation similar to yours, I’ll see if I can backtrack the raw data.

Anthony et al, if water vapor is dropping it’s effect should first be felt in the tropics. Do we have cooling there?
Secondly, dropping water vapor is a sign of cooling somewhere, oceans, troposphere, polar regions, temperate regions. So where do we have cooling? We think we have cooling in the seas and the troposphere, South pole any other areas? Perhaps they are enough.


Here is a nice chart on the W/M^2 of sunlight: http://www.spacewx.com/solar_spectrum.html
Notice that most of the energy contained in sunlight is in the 1,000 (10^3) to 10,000 (10^4) nm range (1 to 10 mircon respectively). Where is CO2’s range???? two small isolated peaks below 10 mircons and one larger above 10 microns. And where is the H20 range??? Spanning most of the range of sunlight. That in itself should negate the silly argument about CO2 absorption and emission driving H20. It’s just plain deceptive to even maintain the position.
For those of you math challenged, here is a nm to mircon calculator: http://www.unitconversion.org/length/nanometers-to-microns-conversion.html


GHG alarmists obsess about positive feedbacks. They obsess about positive feedbacks that don’t exist. The ones that do exist generally result in several thousand feet of continental ice covering half of North America.

Locri — yes that was my thought as well. However after looking around for some time, and reviewing papers about what Svensmark has proved, I get this — When you take into account the Svensmark cosmic ray effect then a possible theory begins to come to the fore. Could it be that sunspots are just an indicator, not the prime mover for cooling, the real player might be cosmic rays and low level cloud formation and subsequentially higher precipitation.
Sunspot number would likely be the prime forcing for warming, because of the direct radiative effect of CMEs and increases in the solar wind. So just having sunspot numbers go down, like in a normal solar minimum, may not provide enough forcing for cooling. Consider, there have been periods of low sunspot activity without any cooling correlation before, a normal sunspot minimum for instance.
What we have here is the unique observed event of the lowered sun’s magnetic component.
Try this line of reasoning — The sun goes quite, caused primarily due to a drop in the sun’s magnetic field, increasing the incidence of cosmic rays on Earth’s atmosphere, this in turn increases low level clouds, which produces more rain, which then cools Earth. The missing link was the cosmic rays and cloud formation theory which was provided by Svensmark..


Anthony – the decrease in specific humidity at ~6 miles (300mBars) is very interesting and the site you linked to is pretty neat. However, before jumping to too many conclusion re: feedback etc., you might want to check the trends in specific humidity at lower atmospheric levels. 600 -900 mbars seem pretty flat, but 950 mbars, and particularly 1000 mbars show very clear increasing trends over the time period in question. From eyeballing the g/Kg numbers, it appears that the lower troposphere is gaining more water than the upper troposphere is losing. That’s just eyeballing it though, and maybe it balances out. Keep up the good work.
REPLY: As I understand this data, this is data up to 300 mb, not at 300 mb. I’ll dig a little deeper though to be sure.


What we need is a graph showing heat capacity in watts/ppm at given temperatures and heat absorption rate of change in watts/ppm.
Water vapor absolutely dwarfs CO2.
Water is also a working fluid for heat transfer from the Earth’s land and ocean into space. When it condenses from water vapor into clouds, it releases huge amounts of heat. The reverse is true when fog or clouds dissapate.
Everyone is familiar with a cold can of soda having water condense on it on hot humid days, thus warming the can. Then there is the nighttime convergence of dewpoint and air temps which prevents nights from getting cold.
CO2 does not do this.

Richard deSousa

Oh oh… the drop in water vapor can’t be a good omen. Ice age water vapor was very low… could we be heading towards another Maunder or Dalton Minima?

David Walton

So, how long will it be before the US Supreme Court declares water (in vapor form) a pollutant?
So much for hydrogen fuel alternatives.


I see where you’re trying to take this argument, but I have a few comments.
The major assumption going in here is a syllogism – that specific humidity (which, as a reminder for those of you who don’t know, is simply the ratio of water vapor to dry air in a parcel) is directly proportional to temperature which is logarithmically proportional to CO2 concentration. I won’t fault the logical argument, but I will point out something that I think contradicts this notion:
Extending this argument one step further, higher specific humidity equals a higher amount of precipitation dispensed from the atmosphere. However, this is not a prediction of global warming – climate models have no consensus on what the global precipitation change will be, although there is some consensus on certain regional effects. To illustrate this, look at 1998 on the first graph; why does the warmest year on record not have the highest average specific humidity?
In reality, the water vapor feedback is much more complicated than being acknowledged here. As has been already brought up, water vapor varies dramatically by region and time. It has an extremely short residence time in the atmosphere compared to CO2.
You hypothesize that decreased water vapor means less energy reflected away from the atmospheric system and subsequent warming. Then why is an apparent decrease in average water vapor correlating, as skeptics point out at every chance right now, to a decrease in global temperature rather than an increase?
REPLY: ?? Your last question: It is explained clearly in the article. Note this is up to 300mb so while there may have been an increase in near surface humidity in 1998, it was likely manifested mostly at lower levels of the atmosphere.

David Segesta

Wow! Anthony if graphs can provide food for thought you just gave us a 6 course meal. That graph on electromagnetic absorption is one I’ve been looking for. (I would also like to have the original source for that if possible.) But if I’m reading it right it says in the wavelengths that water or CO2 can absorb, they are already absorbing nearly 100%. So adding more would not make much difference.
And the graph showing water vapor content is also very surprising. The decreasing trend from 1960 to 1975 more or less agrees with temperature drops during that time. But the drop from 1990 to 2000 runs counter to the temperature trend. It sure seems that increasing temperatures would cause more evaporation but the graph from 1990 to 2000 shows the opposite.
Wow- Lots to think about.

Robert Wood

Well, how can the realclimate boys claim a positive feedback from increase in CO2 to increase in H2O vapour? It is obviously non-existance.


So this chart has been available all along, for those who cared to look??
So much for due diligence on the part of the model makers when they “assumed” that CO2 induced warming was causing the atmosphere to hold more water vapor.


Anthony – I am pretty sure the graphs are for the specific humidity at that atmospheric level and the reason it is listed as up to 300 mbars is because they don’t have data for higher atmospheric levels. If you check the plot of specific humidity at 1000 mbar,e.g., you get values of around 7.7 g/kg, at 925 mbars, listed values are around 5.9 g/kg, while at 300 mbars, the values are around 0.2 g/kg. Obviously, this decrease in value with height makes sense since the atmosphere can’t hold as much water vapor at higher altitudes.
I did a quick download of the data for each listed level (1000, 925, 850….300) and summed the annual average values (I do not know if this is the right way to do this analysis! since I am not a meteorologist). What it showed was a modest increase (~2.5%) in the total specific humidity from about 23.4 g/kg to about 24 g/kg for the summed values from 1000 to 300 mbar for the period from 1950 to 2007. I have the data file and a graph but do not have anyway of posting it on the site.
REPLY: Bob check your email, I’m interested in determining if this data plot from ESRL is “at level” or “up to” as they state. The values you cite are right for levels, but I viewed this as an anomaly plot for up to 300mb rather than absolute values for that level.
But the question needs to be settled, I may be in the wrong as to my interpretation.


MattN (07:33:48) ,
Here is my understanding of why regional differences matter to the climate:
Regional patterns matter because physical processes take place at the temperature, pressure, humidity, etc. at the location they are happening. They do not take place based on the average temperature of the earth or any other averages.
Having more or less water vapor on average is useful for tracking trends but not for determining what physical processes are taking place in a certain region.
For instance if cold dry air comes out from high latitude (a cold front) and causes water to be removed from the warmer, more humid air over a warm part of the ocean, more radiation will escape into space from that area of the ocean than from another area of the ocean that still has humid air above it (assuming the ocean temperature is the same in both regions). This is called weather and may not affect the average water vapor in the air if water is evaporating elsewhere and replacing the water vapor that has rained out, but it is important to the radiation budget of the earth.
If the average is done properly (i.e. perhaps a geometric average is necessary for the particular physics of a process — just an example), then the trend can give us some clues about what is happening overall. For instance, if CO2 continues its increasing trend but water vapor continues its decreasing trend we can conclude (assuming all other things are equal, which is the catch) that an increase in CO2 does not induce an increase in water vapor in the atmosphere.
Lyman Horne

Bill Marsh

Well, so much for one of the pillars of positive feedbacks in AGW. As I recall the IPCC treatment of feedbacks in FAR claims a huge amount of ‘warming’ influence from increased water vapor caused by increased CO2 warming. If this is not the case, then the IPCC calculations are in deep trouble.
It does fit with NOAA data that shows an increase in precipitation in the US since 1900 though.
There is one aspect of this that bears scrutiny. IF water vapor is decreasing and water vapor is overlapping a a goodly portion of the CO2 absorption band, then wouldn’t this allow CO2 to absorb more long wave radiation? This would be only true if the remaining water vapor did not still absorb all the radiation available at those wavelengths.
Let’s see here:
IPCC can’t calculate the sensitivity of the climate to CO2 (nobody really can) and may have overstated it by a factor of 3-4.
Water vapor is not providing a positive feedback
Aerosols are not providing a ‘dampening’ of the warming.
Overall not looking good for the AGW crowd.


Yes, temperatures in the tropics are currently at their lowest levels since 1989…with anomalies lower than any other region on earth.


Doesn’t decreased water vapour fly in the face of positive feedback in global warming theory.
And now we know why the AquaSat data is being swept under the carpet:
Not only less vapor (cooling via less GHG), but more low-level cloud cover (increased albedo).
So much for positive feedback.

Tom Bruno

Can someone with more knowledgable put some perspective to the increments in specific humidity on the left side of the graph? The graph shows flucuations from a high of just over 0.21 to a low of about 0.172 over the course of 60 years. How significant is that?


Austin- you say “CO2 does not do this.”
Actually, there may be a place on Earth where CO2 can condense to its solid phase during cooling and sublimate during subsequent warming. But I don’t know how often Amundsen Station in the Antarctic has dipped below -78.5 C!
: )

David Segesta

OOPS. Well maybe I was to hasty in assuming that less water in the air means there was less evaporation. Of course that is one possibility but it is also possible that there has been more precipitation, which drains moisture from the air.
BTW does anyone know if molecules in the atmosphere can gain enough energy to escape the earth’s gravity? That would provide another means by which energy could leave the earth, besides radiation.


“As has been already brought up, water vapor varies dramatically by region and time. It has an extremely short residence time in the atmosphere compared to CO2”
I fail to see what relevance of residence time to what is an on-going process. So one particular molecule of H2O may not stay in the atmosphere for long, but it’s the total concentration at any point in time that has any effect.
Besides, the concentration of CO2 also varies quite widely by region and time. The diurnal range over woodlands can be in excess of 100ppm.

Paddy Lenihan

The new humidity data causes me to wonder if there is a correlation between it and the period of time that atmospheric CO2 is retained. As I understand it, the GCMs assume that CO2 is retained in the atmosphere for about 100 years. This assumption supports the hypothesis that CO2 and water vapor interact so that an exponential increase in temperatures occurs as a result from prolonged long wave retention in increasing amounts.
Conversely, 50 years worth of empirical research indicates that retention is between 4 and 12 years. Does the new data falsify the GCM assumption about the length of the cycle for retention of CO2?
Am I missing something? I look forward to reading new comments on this thread.

Bill Illis

Just noting that Global Warming theory relies on the assumption that “Relative Humidity” would remain constant (the amount of water vapour the atmosphere can hold given the temperature – as relative Humidity approaches 100%, it rains).
In Global Warming theory, as atmospheric temperature increases, the amount of water vapour the atmosphere can and does hold will increase although the Relative Humidity would remain constant.
Anthony’s chart relates to “Specific Humidity” which is the actual mass of water vapour in the air. Specific Humidity should be increasing according to Global Warming as the warmer air is able to hold more water vapour.
So not only is Specific Humidity not increasing with the supposed increase in atmospheric temperature, it is actually falling (which is the nail in the coffin for the +3.5C increase in temperatures for a doubling of CO2 sensitivity estimate.)

[…] 18, 2008 · Filed under Climate crap Anthony at Watts Up With That has some interesting graphs regarding water vapor and planetary temperature. As he points out: It […]


Talking of planetary temperature, the Hadley global temperature for May is in. Are these the wrong figures or have they been Hansonised?
2008/01 0.053
2008/02 0.192
2008/03 0.430
2008/04 0.254
2008/05 0.278 UP???

Ken Feldman

At 8 miles up, isn’t that in the upper troposhere, above the height where most of the water vapor in the atmosphere is? In fact, according to the RSS data, that’s up in the TTS (very slight warming) and TLS (very distinct cooling) temperature bands as measured by the satelites. So if temps are cooling that high up, you’d expect to see water vapor decrease.
Another question. What is the data collection method behind that graph? Given that the trends start in 1940s, I’m guessing radiosondes. There have been bigger discrepencies in the radiosondes with changing instruments, changing locations, time of day, etc… then there have been with the surface temperature stations. You should look into that.
Also, CO2 becomes more important at higher attitudes, because it’s well mixed throughout the atmosphere, unlike water vapor, which is more important near the surface. So increasing the amount of CO2 is going to block more outgoing longwave radiation higher up in the atmosphere, which is going to have an impact on the radiation budget. The planet is going to have to warm up to equalize the amount of radiation received.


“At 8 miles up, isn’t that in the upper troposhere”, but it says “up to” 300mb not “at” 300mb.
Can we assume they are measuring the “specific humidity” from sea level up to 8 miles as an average?


Mr Watt,
From your link: …&level=300…. That’s in the stratosphere, which is cooling.
Go to here: http://www.cdc.noaa.gov/cgi-bin/Timeseries/timeseries1.pl
try it again at say 1000mb height.
Then you get specific humidity going up.
It helps to look in the troposphere (where there’s warming) rather than at 300mb in the stratosphere. Yes the dataset goes up to 300mb, but you selected 300mb as the analysis level.

Ken Feldman

“Can we assume they are measuring the “specific humidity” from sea level up to 8 miles as an average?”
I don’t know what we can presume from that graph, which is why I asked the question. The graph Anthony presents isn’t consistent with the results from peer reviewed science. Take a look at these three articles:
Water Vapor Feedback is Rapidly Warming Europe
Trends and variability in column-integrated atmospheric water vapor
Enhanced positive water vapor feedback associated with tropical deep convection : new evidence from Aura MLS
Water vapor feedback is clearly increasing, and in some cases, much more than predicted by models.
Anthony has used incorrect methods before (his comparison of the anomalies of the different surface and satellite temperature datasets without first putting them on a common baseline for example). I think we should know what his methods are and how he derived this graph before we jump to conclusions.