Open Mind

Vapor Lock

July 8, 2009 · 48 Comments

Andrew Dessler and Steven Sherwood have a paper in the latest issue of Science reviewing the evidence regaring water vapor feedback. As Dessler writes on Grist,


It is a Perspective, meaning that it is a summary of the existing literature rather than new scientific results. In it, my co-author Steve Sherwood and I discuss the mountain of evidence in support of a strong and positive water vapor feedback.


Water vapor feedback is a straightforward idea. As temperature rises, according to the Clausius-Clapeyron equation (very basic physics) we expect the absolute humidity of the atmosphere to increase. Water vapor is a potent greenhouse gas, so with more water vapor in the air there’ll be further greenhouse warming: this feedback will amplify any warming due to CO2.

Dessler and Sherwood emphasize that the reality of water vapor feedback isn’t just a conclusion of basic physics, or a result of model simulations. We now have sufficient observed data to make it a near-certainty.


Despite these advances, observational evidence is crucial to determine whether models really capture the important aspects of the water vapor feedback. Such evidence is now available from satellite observations of the response of atmospheric humidity (and its impacts on planetary radiation) to a number of climate variations. Observations during the seasonal cycle, the El Niño cycle, the sudden cooling after the 1991 eruption of Mount Pinatubo, and the gradual warming over recent decades all show atmospheric humidity changing in ways consistent with those predicted by global climate models, implying a strong and positive water vapor feedback (9–13).

The actual evidence is contained in the many references. One example is from Held and Soden (2006, J. Climate, 19, 5686) showing the stunning agreement between model-simulated behavior and satellite observations of tropical ocean column-integrated water vapor:

Heldfig1

This example is but one of many; as Dessler says, the evidence is no longer merely a giant heap, it’s now a mountain.

Both theory and computer models lead to the expectation that relative humidity will remain roughly constant. Now we can add observations to that list:


Both observations and models suggest that the magnitude of the water vapor feedback is similar to that obtained if the atmosphere held relative humidity constant everywhere. This should not be taken to mean that relative humidity will remain exactly the same everywhere. Regional variations of relative humidity are seen in all observed climate variations and in model simulations of future climate, but have a negligible net impact on the global feedback.

The magnitude of water vapor feedback is roughly large enough to roughly double the warming effect of increased man-made greenhouse gases alone:


Thus, although there continues to be some uncertainty about its exact magnitude, the water vapor feedback is virtually certain to be strongly positive, with most evidence supporting a magnitude of 1.5 to 2.0 W/m2/K, sufficient to roughly double the warming that would otherwise occur.

But perhaps the most compelling argument that water vapor feedback is real and is strongly positive, the real “lock” on the discussion, comes from the fact that even denialists have a hard time denying it; as Dessler notes,


Interestingly, it seems that just about everybody now agrees water vapor provides a robustly strong and positive feedback. Roy Spencer even sent me email saying that he agrees….”

Doubling CO2, with no feedbacks, leads to warming of about 1.1 to 1.2 deg.C. Double that due to water vapor feedback, and we have 2.2 to 2.4 deg.C. Then of course, there’s ice/snow/albedo feedback, feedbacks in the carbon cycle, …

Categories: Global Warming
Tagged:

48 responses so far ↓

  • Zeke Hausfather // July 8, 2009 at 6:04 pm | Reply

    Quick correction: the paper you linked was in the February 2009 issue of Science rather than the latest one. Got me excited there for a sec until I realized I already read it six months ago :-p

  • TCO // July 8, 2009 at 7:47 pm | Reply

    Tammy:

    The Dressler paper is very nice. So is the graph you posted. I am very intrigued by the level of evidence from observation.

    The “Clausius Clapeyron” touting always does a fingernails on chalkboard to me. I actually took P Chem, so at least touched base with CC. It was some kind of differential equation or something. I would not have the confidence to cite it so flippantlythe way you lot do.

    Also, I think you are kinda using a big fancy word and touting a gas law, when in actuality the Earth is a dynamic system. It’s kind of a different situations from C-C. Sure there might be some good insights (to help create a Bayesian prior, and in fact my “bet” would be constant relative humidity), but it’s not the situation covered by the gas law. You can’t write a derivation, proving the point. So at least there’s some possibility that the response does not happen the way you think. It’s a bit like talking about Le Chatelier’ principle for something that is not a chemical equilibrium.

    [Response: The Clausius-Clapeyron equation is more fundamental than you suspect; it really is basic physics. And Le Chatelier' principle isn't restricted to chemical equilibria. The expectation that humidity increases with temperature is really quite basic.]

  • Mark // July 8, 2009 at 8:40 pm | Reply

    ” The expectation that humidity increases with temperature is really quite basic.]”

    And, if I’m remembering correctly, just another example of chemical dispersion across a density gradient when there is an energy barrier to cross.

    Heck, heat transfer and insulation uses it…

  • Zeke Hausfather // July 8, 2009 at 9:00 pm | Reply

    You would think certain weathermen would have a better understanding of why water vapor is a feedback and not a forcing, but alas. Of all the manifold uncertainties in climate science, the sign and rough magnitude of the water vapor (excluding clouds) feedback is not a particularly large one.

  • Mark // July 8, 2009 at 10:30 pm | Reply

    That’s a lot of words to say very little, TCO.

    Tamino is not using words that are any fancier than necessary for the purpose and he is not “touting a gas law”, whatever that means.

    Do you have a point?

  • Isaac Held // July 8, 2009 at 10:51 pm | Reply

    Tamino,

    Thanks for showing the figure from my paper with Brian Soden (for which Brian deserves credit), but this has relatively little to do with water vapor feedback. The column integrated water vapor is dominated by the vapor in the lowest 1 or 2 kms of the atmosphere, while radiative transfer calculations show that most of the water vapor feedback comes from the upper troposphere, especially in the tropics. This is an important distinction, since there are very solid energetic arguments for why relative humidity cannot change very much near the surface, but these arguments don’t tell us much about the upper troposphere. There is evidence in favor of strong water vapor feedback (especially http://www.sciencemag.org/cgi/content/abstract/310/5749/841) but it is worthwhile keeping this distinction in mind. The increase in water vapor near the surface does have lots of other important implications.

    [Response: I used your graph to illustrate the match between computer model simulations of water vapor changes and direct observations. Is it the case that this isn't a valid illustration of the skill of models of water vapor changes? Or does it only testify to skill in near-surface changes?]

  • suricat // July 8, 2009 at 11:30 pm | Reply

    I have a problem with this presentation. I don’t recognise the precip. water graph, can’t find any mention of diurnal validity and don’t understand how merely precip. water can record the energy transmissions within the latent part of the system without a ref. to either temps, or RH (or preferably precip. water and specific humidity (SH) and temp).

    Temp feedback from water vapour (WV) is given 12/24 from the diurnal cycle, but outgoing long-wave radiation (OLR) is given 24/24 from the diurnal cycle. Thus, the positive feedback factor is evident for only half (or less than half due to the applicable insolation gradient) the duration of the negative feedback factor for a complete diurnal cycle. I can see no representation of a diurnal cycle, perhaps someone could point me towards this please?

    Some of the logic behind the “diurnal cycle” is that from sunrise to about 14.00 hrs – 15.00 hrs local time the RH is playing “catch-up” with temps where water is abundant. Thus, the atmosphere is unlikely to be in a “saturated” WV condition before this hour. I think this point is important because it limits any “maximum” WV feedback mechanism to between this hour and sunset.

    If precip. water was only taken from the hours later than 3.00 PM (local time) this may well obfuscate any outcome.

    Hope this helps and hope someone can help me.

    Best regards, suricat.

  • Isaac Held // July 8, 2009 at 11:50 pm | Reply

    Tamino,

    As Dessler and Sherwood describe, there is plenty of evidence that models’upper tropospheric water vapor behaves about as expected for the seasonal cycle, El Nino, etc. And one has evidence for the expected longer-term trend as well (as in the link I provided above). But we shouldn’t expect the agreement between model and observations in the upper troposphere to be as beautiful as near the surface. The model in the figure you show uses the observed ocean surface temperatures as a boundary condition, so it stands to reason that it will fit atmospheric observations near the surface especially well. (Needless to say, there is no way to get a fit like this in a free-running climate model, generating its own El-Nino’s.) The controls on relative humidity in the upper troposphere are sufficiently different that I think one needs to be careful not to give the wrong impression with this figure, except to lay the encouraging foundation that the model is just fine with regard to water vapor in the lower troposphere. One can turn things around and use this figure to add confidence in estimates of trends in tropical surface temperatures, by the way.

  • Hank Roberts // July 9, 2009 at 12:01 am | Reply

    Is this 2005 paper a better source for documenting upper troposphere changes?

    “… We use satellite measurements to highlight a distinct radiative signature of upper tropospheric moistening over the period 1982 to 2004. …”

    http://www.sciencemag.org/cgi/content/abstract/310/5749/841
    Science Express on 6 October 2005
    Science 4 November 2005:
    Vol. 310. no. 5749, pp. 841 – 844
    DOI: 10.1126/science.1115602

    Reports
    The Radiative Signature of Upper Tropospheric Moistening
    Brian J. Soden,1* Darren L. Jackson,2 V. Ramaswamy,3 M. D. Schwarzkopf,3 Xianglei Huang4

  • Ray Ladbury // July 9, 2009 at 12:52 pm | Reply

    Suricat,
    I’m not sure I understand your problem. First, how do you get that water-vapor feedback is 12/24. Greenhouse forcing works day and night. And while it is true that RH doesn’t equilibrate instantly, we’re looking at higher temps day and night. Indeed, the increased greenhouse effect may even limit the swing, thus reducing the catchup time.

  • Timothy Chase // July 9, 2009 at 3:31 pm | Reply

    suricat wrote:

    Temp feedback from water vapour (WV) is given 12/24 from the diurnal cycle, but outgoing long-wave radiation (OLR) is given 24/24 from the diurnal cycle. Thus, the positive feedback factor is evident for only half (or less than half due to the applicable insolation gradient) the duration of the negative feedback factor for a complete diurnal cycle….

    If the water vapor is in the atmosphere it will be absorbing long-wave radiation, then it will be emitting backradiation which will further heat the ground. And once a given water vapor molecule is in the atmosphere it will take roughly a week for it to get rained out, so the absolute humidity should be roughly constant over the diurnal cycle. And it is actually the absolute humidity which is important in terms of the greenhouse effect. Day or night, it will be there, and higher temperatures due to carbon dioxide forcing imply that more of it will be there, and that will mean feedback.

    However, the greenhouse effect is most important at night. The reason being? Daytime is warmer than nighttime, therefore there will be more moist air convection that will carry heat to higher layers of the atmosphere where it may be more easily radiated into space. This is why climatologists predicted that temperatures would rise more quickly at night than during the daytime given an enhanced greenhouse effect due to higher levels of greenhouse gases, whereas an increase in solar radiance or global brightening due to a reduction in reflective sulfates will increase temperatures more rapidly during the daytime.
    *
    suricat wrote:

    Some of the logic behind the “diurnal cycle” is that from sunrise to about 14.00 hrs – 15.00 hrs local time the RH is playing “catch-up” with temps where water is abundant. Thus, the atmosphere is unlikely to be in a “saturated” WV condition before this hour.

    Relative humidity is roughly constant for the atmosphere as a whole. So as the temperature rises year after year due to carbon dioxide forcing, the absolute humidity is going to continue to climb.

    And the atmosphere doesn’t have to be saturated with things dripping due to condensation or rain for the greenhouse effect to take place. If the relative humidity is at constant average 80% but the temperature is higher so that this relative humidity corresponds to a higher absolute humidity, this will imply water vapor feedback.

    In fact, water vapor feedback doesn’t even require a constant relative humidity, just a higher absolute humidity as the result of higher temperatures. But as the average relative humidity is roughly constant, water vapor is a strong feedback.

  • chriscolose // July 9, 2009 at 6:12 pm | Reply

    Hank’s reference (the same linked by Isaac Held in his first comment) provides a much better signature of water vapor feedback. Lower tropospheric water vapor changes are much more tightly coupled to temperature change through thermodynamic arguments, while the substantial majority of WV feedback is not preciptable water at the boundary layer, but water in the free atmosphere (and in the tropics).

    It’s also important to point out that Clausius-Clapeyron is a necessary consideration but not sufficient, as it essentially just provides the upper bound on the saturation vapor pressure. This is why the constancy of relative humidity is important, and theoretical justification for that to be the case is not that great, but it turns out to be (roughly) what is observed in response to climate change. Again, Soden et al’s Science paper explains this well, and Anthony Del Genio has a good piece in 2002 detailing observational confirmations after the Pinatubo eruption
    http://pubs.giss.nasa.gov/docs/2002/2002_DelGenio.pdf

  • suricat // July 9, 2009 at 11:25 pm | Reply

    Ray Ladbury.

    I don’t know how this site works yet, so I’ll only show quotes within my script.

    “First, how do you get that water-vapor feedback is 12/24″.

    Hope I wasn’t being too obtuse there, but WV is only generated in nature (its vastly most abundant form) during daylight hours (insolation). The absence of daylight (insolation) causes a lowering to temps, a C-C related RH reduction, and thus a reduction in specific humidity (SH) with its corresponding increase of precipitable water (water in the atmosphere).

    “Indeed, the increased greenhouse effect may even limit the swing, thus reducing the catchup time.”

    I doubt it. This is an issue concerning the temporal aspect of the dynamic concerning the energy required to cause a change of phase, which can only be shortened (water to WV) by an increase of insolation intensity, or lenghtened (WV to water) by a decrease of OLR.

    Sorry, but the middle part of your post is either obvious, or doesn’t make full sense to me (I have been known to be a bit thick at times).

    Best regards, suricat.

  • suricat // July 10, 2009 at 12:25 am | Reply

    Timothy Chase.

    I prefer to use the specific humidity (SH) metric, as its less ‘leaky’ than the absolute humidity (AH) ‘volumetric’ measure.

    I don’t think you follow the ‘disconnect’ in this thread. It looks to be a mess to me. I’ll try to explain the reason why I posted here in the first instance.

    The ‘blog’ posted a graph describing ‘Precipitable Water Anomaly %’ that I can’t recognise (sorry Isaac). The reason I can’t recognise this is twofold.

    Firstly, the ‘legend’ to the graph is beyond acceptance! How can ‘Wentz 1997′ offer satellite observations up to 2005? Either this is a misprint, or the observations were made by a ‘Time Lord’!

    Secondly, the relationship of Precipitable Water has a very limited connection to any WV feedback and, as Isaac Held says, “One can turn things around and use this figure to add confidence in estimates of trends in tropical surface temperatures, by the way”.

    Hope this helps your understanding of my post.

    Best regards, suricat.

  • Ray Ladbury // July 10, 2009 at 2:01 am | Reply

    Suricat, You are misunderstanding the feedback. What we are looking at is absorption of IR light by water vapor. If the water vapor increases OVER WHAT IT WOULD HAVE BEEN, you get additional absorption.

    • suricat // July 10, 2009 at 11:35 pm | Reply

      Thanks for your advice Ray, but it isn’t the “feedback principle” that I have a problem with (I accept this in a radiative context). Apparently, it’s the way that the science community present their papers (thanks to help from other inputs here this has been reduced).

      Comming from an engineering discipline, I find the “papers system” a totally nightmarish maze. However, if that’s the way that science links findings I guess I’ll just need to grope my way through it.

      Best regards, suricat.

  • Ian Forrester // July 10, 2009 at 2:07 am | Reply

    Suricat, why don’t you read the two papers you criticize so vehemently?

    You will find that Wentz 1997 describes an algorithm for calculating “geophysical parameters over the ocean from special sensor microwave/imager (SSM/I) observations”. If you had any experience in scientific papers you would realize that there are “methods” papers and “data” papers. If you had actually read the paper you would have realized that Wentz 1997 is a methods paper.

    • suricat // July 11, 2009 at 12:33 am | Reply

      Ian, I’ve read all the papers that were provided with a link. However, the papers that were not linked require a “pay per view” premium for the acquisition of “the knowledge” from what I’ve been able to “Google” (this is worse than the “patents system” for engineers) and the abstract doesn’t reveal everything!

      My response wasn’t intended as “vehement”, merely “disbelief”, or perhaps “not understanding” (because “the knowledge” wasn’t communicated)! Perhaps the science community should consider obviating the distinction between a “data” paper and an “instrument calibration” paper, or “methods” paper. They are all very different, but it’s nice to know that a “Time Lord” wasn’t involved with the satellite observations that the unlinked Wentz 1997 paper suggested.

      Thanks for this.

      Best regards, suricat.

  • Timothy Chase // July 10, 2009 at 2:56 am | Reply

    suricat wrote:

    I prefer to use the specific humidity (SH) metric, as its less ‘leaky’ than the absolute humidity (AH) ‘volumetric’ measure.

    I can see your point…

    Absolute humidity ranges from 0 grams per cubic meter in dry air to 30 grams per cubic meter (0.03 ounce per cubic foot) when the vapour is saturated at 30°C.[1] (See also Absolute Humidity table)
    The absolute humidity changes as air pressure changes. This is very inconvenient for chemical engineering calculations, e.g. for dryers, where temperature can vary considerably.

    http://en.wikipedia.org/wiki/Humidity

    … assuming you are soaking wet. Otherwise it might be time for you to get out a textbook on calculus.
    *
    suricat wrote

    I don’t think you follow the ‘disconnect’ in this thread. It looks to be a mess to me. I’ll try to explain the reason why I posted here in the first instance.

    The ‘blog’ posted a graph describing ‘Precipitable Water Anomaly %’ that I can’t recognise (sorry Isaac). The reason I can’’t recognise this is twofold.

    That is why it might help to read captions.

    Please see:

    FIG. 1. A time series of the tropical-mean (30°N-30°S), ocean-only column-integrated water vapor from satellite observations…

    *
    suricat wrote

    Firstly, the ‘legend’ to the graph is beyond acceptance! How can ‘Wentz 1997′ offer satellite observations up to 2005? Either this is a misprint, or the observations were made by a ‘Time Lord’!

    That is why it helps to look up sources. In this case, if you look at the paper:

    Held and Sode, 2006: Robust responses of the hydrological cycle to global warming. Journal of Climate,

    you find that it is refering to:

    Wentz, F. J. (1997), A well-calibrated ocean algorithm for SMM/I, J. Geophys. Res., 102, 8703– 8718.

    The source is not the source for the satellite data but for the algorithm for processing the data from a special sensor microwave/imager — in order to arrive at column-integrated water vapor.
    *
    suricat wrote:

    Secondly, the relationship of Precipitable Water has a very limited connection to any WV feedback and, as Isaac Held says, “One can turn things around and use this figure to add confidence in estimates of trends in tropical surface temperatures, by the way”.

    Hope this helps your understanding of my post.

    Isaac Held made a couple of points but what you quote is irrelevant to the question of the relationship between water vapor and water vapor feedback. His point was that the distribution of the water vapor in the atmospheric column matters. What you quote was something that he shared about the close relationship between water vapor and tropical surface temperatures. An interesting point but not a flaw in an argument. And the fact that you misunderstood what you quoted as some sort of criticism (there was criticism in his comment, but not there) is rather emblematic of your level of understanding, which – like your prose – is often only slightly above the level of word salad.

    Please see for example:

    This is an issue concerning the temporal aspect of the dynamic concerning the energy required to cause a change of phase, …

    *
    suricat wrote:

    Best regards, suricat.

    Take care.

    • suricat // July 11, 2009 at 2:59 am | Reply

      “I can see your point…”

      Yes. The WV content is altered by both pressure and temp. That’s why pressure and temp are needed to qualify the RH of the fluid.

      “… assuming you are soaking wet. Otherwise it might be time for you to get out a textbook on calculus”.

      Why? The C-C relationship provides all the calculus that we need for this. Saturation is not the issue, but the RH level, thus temp, with the observed SH surely is the issue. However, this requires a temp and pressure (or altitude) input.

      “That is why it might help to read captions”.

      Thanks, but this doesn’t help much if the “captions” are not readily available!

      “Please see:
      FIG. 1. A time series of the tropical-mean (30°N-30°S), ocean-only column-integrated water vapor from satellite observations…”

      This site really messes up my word processor (word pro) when trying to post something semi pro here! Any suggestions for an alternative word processor?

      The only column-intgrated data that I’ve seen in this post is for precipitable water from satellite observations. Would someone please proffer a link to the WV aspect.

      I’ve just discovered that I need to reactivate an “unusual” default for word pro after a particular cut and paste action from this site. It gets better!

      Who am I kidding. I’ve probably tried more than most individuals to maintain a dialogue here. However, my inquiry has been met with nothing less than “derision” and unacceptability.

      This only shows the degree that the science fraternity has become alienated from the masses.

      I’m out of here!

      suricat.

  • Hank Roberts // July 10, 2009 at 3:12 am | Reply

    Suricat, in journal writing, when a piece of hardware is mentioned, a citation is given to some source_describing_the_ instrument_used. I’m sure that’s the case there.

    Look at the full text. You know how to find it?

  • Hank Roberts // July 10, 2009 at 3:22 am | Reply

    In fact to make it easier, just copy the cite you question and paste it into Google Scholar like this:

    http://scholar.google.com/scholar?q=+Special+Sensor+Microwave+Imager+%28SSM%2FI+Wentz+1997%29

    Doing that, you will find out what people have to say about it and see how and why it gets mentioned, even if you don’t read Dr. Held’s actual paper.

    Results … about 672 for Special Sensor Microwave Imager (SSM/I Wentz 1997)

  • Hank Roberts // July 10, 2009 at 3:27 am | Reply

    See also:
    http://ams.confex.com/ams/87ANNUAL/techprogram/paper_117342.htm

    With many satellites, many different sensors, many different orbits, and many different longterm records, all of which require calibration and some of which have ground truth or aircraft measurements or other sources to correlate, NOAA is doing the work to make the pieces come together. Good little description there.

    • suricat // July 12, 2009 at 1:11 am | Reply

      Thanks for your help Hank. I’ll bear this in mind, but an investigative search is time consuming for a reader. I particularly like section 3.4 in the 117342[1].pdf in your link, it emphasises the need for “data, metadata and documentation” to be kept in sync.

      Best regards, suricat.

  • jyyh // July 10, 2009 at 3:53 am | Reply

    This reminds me of a news item here about a year back: A home owner was having trouble with the humidity in his basement during the late 1980s. He got a dehumidifier and began to take measurements of how much water vapor the apparatus extracted from the air. Of course, this is a local measurement, but the results showed a clear upward trend, as would be expected from the climate models for northern locations. Some people are taking note.

  • Chris S. // July 10, 2009 at 4:52 pm | Reply

    Michel, thanks for looking. The reason I posted them was in response to Martinsh’s assertion that we “do not have any data of how did the climate look like 200-300 years ago” this is not the case as these records show. The Marsham data shows that phenology showed no significant change over much of the 200 year period 1736-1947.

    That phenological records are now showing evidence of a shift in the onset of spring (2-7 days per decade being typical across many of the groups so far studied (on a side note the 2.5 days you cite above is per decade not per 30 years)) that was not seen throughout the previous 200 years (as shown at Marsham-thus terribly relevant to what is happening today).

    As for Kyoto, the best reference for it is:
    Lamb, H. H. (1977). Climate Past, Present and Future. London, Methuen. I’d heartily suggest asking your library to get hold of it for you.

    I’d also suggest “Visser, M.E. & Both, C. (2005) Shifts in phenology due to global climate change: the need for a yardstick. Proceedings of the Royal Society B, 272, 2561-2569.” for a good overview of the subject (see here: http://rspb.royalsocietypublishing.org/content/272/1581/2561.abstract )

  • dhogaza // July 11, 2009 at 4:11 am | Reply

    Perhaps the science community should consider obviating the distinction between a “data” paper and an “instrument calibration” paper, or “methods” paper. They are all very different, but it’s nice to know that a “Time Lord” wasn’t involved with the satellite

    Scientists publish for other scientists, who understand these things without being told.

    Changing how scientists work just because a bunch of RW assholes choose to misunderstand science might be useful, in the short term on climate science issues.

    But the long term implication for science is horrible.

  • Curious // July 11, 2009 at 9:38 am | Reply

    I’m not a scientist at all, but for me it is rather intuitive that WV should increase with warming. I understand that a gas phase and liquid phase must be in a dynamic equilibrium. With warmer temperatures there is more kinetic energy in molecules and as a result I can expect more evaporation. Clausius-Clapeyron says that in order to recover the dynamic equilibrium (i.e. as many molecules going into water phase than into gas phase), water vapor should increase (because the atmosphere can hold more without condensation). I cannot speak about the (relevant) distribution on the atmosphere, but being for me something intuitive, I would require some strong evidence if someone means to deny WV feedback.

    I cannot help linking this summary of the literature regarding WV feedback (published in 2000 but quite complete):

    http://www.gfdl.noaa.gov/reference/bibliography/2000/annrev00.pdf

    And highlight that the magnitude of WV feedback has been understood the same for more than 40 years:

    http://www.gfdl.noaa.gov/bibliography/related_files/sm6701.pdf

  • Eli Rabett // July 11, 2009 at 4:26 pm | Reply

    Perhaps this will help (at least it helps Eli), but it is not authoritative

    The source of water vapor in the atmosphere is the marine boundary layer (MBL), the atmosphere, immediately in contact with the oceans. Clausius-Clapyron pretty much holds there, if for no other reason that there is a lot of spray from the sea.

    Above the MBL the atmosphere sinks water vapor, with losses controlled by condensation. If this layer warms, there will be less condensation, again by Clausius Clapyron

    Since the atmosphere is pretty well saturated with nanosized aerosoles that can grow into cloud condensation nucleii, to first order losses will be temperature controlled. We can talk about cosmic rays here, but they appear not to be significantly controlling from observations. The availability of SOx and NOx is orders of magnitude more important for CCN formation and growth

    The two regions, MBL and everything else (ROA), are mixed by convection.

    Than means that the controlling factors are vertical convection and condensation. Increasing the concentration of water vapor in the ROA will increase the temperature by an enhanced greenhouse effect. Unless vertical convection is proportionally reduced this will be a positive feedback. Since the sea surface is also warming, the temperature difference between the sea surface and the rest of the atmosphere above the MBL is the issue.

    • suricat // July 12, 2009 at 1:57 am | Reply

      Hi Eli, I concur. However, the convective thermal transport system is a chaotic dynamic that rules the lower regions of the tropo, and the expected “signature” is in the upper region of the tropical tropo where OLR is open to TOA. It’s hard to understand how this signature can be observed without great increase in altitude for TOA. Thus, an enormous increase in column WV and water, or an increase in CO2 that exceeds the physical diffusion barrier produced by precipitating water.

      Best regards, suricat.

  • Curious // July 11, 2009 at 5:24 pm | Reply

    Broken link in my previous message:
    ———
    I cannot help linking this summary of the literature regarding WV feedback (published in 2000 but quite complete):

    http://www.gfdl.noaa.gov/bibliography/related_files/annrev00.pdf
    Water Vapor Feedback and Global Warming
    Isaac M. Held and Brian J. Soden, 2000
    Annual Review of Energy and the Environment, 25, 441-475

  • Hank Roberts // July 11, 2009 at 7:41 pm | Reply

    Tamino, I hope you get an answer and more from Dr. Held; have you tried email in case he doesn’t see the inline response you made?

  • TCO // July 11, 2009 at 8:40 pm | Reply

    Fingernails on blackboard. There may be all kinds of good reasons for constant relative humidity as a hypothesis. But the “CC touting” just shows ignorance of CC.

    CC does not tell you anthing except states of 100% rel humidity (it’s about the phase boundary). Does not apply to systems that have multiple locations and transport processes between them.

    You can NOT derive or PROVE the anticipated constant rel humidity from a forcing from the C-C differnential equation! Stop just mouthing words and thinking it makes you smart. Be like Feynman in QED and actually understand things intimately.

    p.s. Thanks for backing me up Held and Eli.

  • Dave A // July 11, 2009 at 10:22 pm | Reply

    OK,

    This is anecdotal but real life evidence from my own backyard. When it rains the temperature cools. Certainly, whilst the clouds build and the humidity rises it gets warmer. But eventually it rains and the temperature cools.

    It takes a while, but not too long, for this to happen in my northern latitude location but in the tropics this occurs regularly on a daily basis. So pray where is the empirical evidence for positive feedback in relation to water vapour?

    [Response: You're joking, right?]

  • Timothy Chase // July 11, 2009 at 10:39 pm | Reply

    TCO scolds Tamino:

    Fingernails on blackboard. There may be all kinds of good reasons for constant relative humidity as a hypothesis. But the “CC touting” just shows ignorance of CC.

    …then later:

    p.s. Thanks for backing me up Held and Eli.

    I assume you mean the post three days ago where you wrote:

    The “Clausius Clapeyron” touting always does a fingernails on chalkboard to me. I actually took P Chem, so at least touched base with CC. It was some kind of differential equation or something.

    Actually I doubt that either Held or Eli was giving that much consideration. Both were likely more concerned with explaining the science. Professors do that, you know.
    *
    TCO wrote:

    CC does not tell you anthing except states of 100% rel humidity (it’s about the phase boundary). Does not apply to systems that have multiple locations and transport processes between them.

    A fair restatement of what Eli told us. Shows some thought on you part.

    In any case we all make mistakes. The big question is what do we do when we discover the mistakes that we have made.

  • David B. Benson // July 12, 2009 at 12:02 am | Reply

    Eli Rabett // July 11, 2009 at 4:26 pm — That the atmosphere is “pretty well saturated” with CCN has been called into question. Certainly unlikely to be true in Antrarctica. More importantly, it has been called into question for the Southern Ocean and even tropical mid-Pacific. This last seems highly improbable as one can see the ITCZ there in satellite photos.

    I am under the impression this still remains a research question.

  • paulm // July 12, 2009 at 4:17 am | Reply

    So does anyone have the foggiest why the CO2 starts to decrease once it peaks?

    There must be a feedback system there for the cycles to continue. This would give us some clues in to finding solutions for reducing the effect of the AGW.

    Are there any theories out there?

    • suricat // July 13, 2009 at 2:28 am | Reply

      Hi paulm! I dare say there are many theories on this, but if you’re considering trends the most logical theory for me, I guess, would be the Milankovich Cycles
      http://en.wikipedia.org/wiki/Milankovich_cycles
      coupled with a Maunder Minimum type event.
      http://en.wikipedia.org/wiki/Maunder_Minimum
      Which when concurrent with a Milankovich minimum, BTW, may also be able to trigger an ice age.

      Why these two phenomenon? Water absorbs CO[sub]2[/sub] and the colder the water, the greater the absorption. Also, the purity of the water has a bearing. The greater the water purity, the greater the absorption of CO[sub]2[/sub] and CO[sub]2[/sub] just happens to be the first atmospheric gas to diffuse into pure water.

      Rain precipitate is fairly pure water and absorbs a lot of CO[sub]2[/sub]. Thus, a large part of the CO[sub]2[/sub] in the atmosphere is returned to the Earth’s surface water by rain and stored in the subterranean water-table and ocean by systems within the hydrological cycle (where other processes also take place).

      This enables us to look at this as a catalytic reaction scenario that grasps for the catalyst (CO[sub]2[/sub]) when other attractors do the same (that is, grasp for CO[sub]2[/sub] to use within their own system of catalysis, and an obvious example of another system would be the photosynthesis system employed by flora).

      So, when average global temps are low there is more competition for CO[sub]2[/sub] between attractors because the hydrological cycle’s sink is greater, and I would expect this to cause a reduction in atmospheric PPM of CO[sub]2[/sub].

      We have presently just passed the warm maximum of a Milankovich cycle, but don’t confuse this with our current sunspot minimum. This also doesn’t take fossil fuel combustion into account.

      Best regards, suricat.

      [Response: Bear in mind that Milankovitch forcing is achingly slow compared to most climate forcings operating now. The *fastest* Milankovitch cycle is about 20,000 years.]

  • TomG // July 12, 2009 at 4:40 am | Reply

    Dave A obviously doesn’t live in southwestern Ontario. That’s in southern Canada.
    Unless it’s a strong cold front passing through, a summer rainfall can knock you on your butt with humidity once the sun comes back out.
    I believe it’s called the heat index…

  • Curious // July 12, 2009 at 11:02 am | Reply

    Hank Roberts says:
    “Tamino, I hope you get an answer and more from Dr. Held; have you tried email in case he doesn’t see the inline response you made?”

    I think Dr. Held answered here by saying that we have a good seasonal and long-term picture of upper water vapor, but “we shouldn’t expect the agreement between model and observations in the upper troposphere to be as beautiful as near the surface” (I guess, that short-term variability is more difficult in the upper troposphere).

  • Mark // July 12, 2009 at 9:27 pm | Reply

    David: “That the atmosphere is pretty well saturated with CCN has been called into question.”

    Dimly remembering my cloud physics lectures 34 years ago, cloud condensation nuclei are common enough everywhere to limit the relative humidity that develops in a cloud before condensation starts to no more than a percent (or less, a small number anyway). However the CCN concentration does have an effect on the number of droplets that develops, and therefore on the size to which they ultimately grow, and therefore on the radiative properties of the clouds. But it remains true that, regardless of CCN levels, condensation will occur when the relative humidity of a parcel of air exceeds (slightly more than) 100%.

  • Mark // July 13, 2009 at 4:33 pm | Reply

    David Benson: “Certainly unlikely to be true in Antrarctica.”

    However, an air temp of less than -30C ensures that there will be spontaneous droplet formation.

    If I remember my physical geography course correctly…

  • Hank Roberts // July 13, 2009 at 7:00 pm | Reply

    Paulm wonders if anyone has thought to ask why CO2 starts to decrease eventually.

    Yes.

    Have you read about the many different known biogeochemical cycles? ….. History of Global Warming (Spencer Weart) · IPCC (Intergovernmental Panel on …
    tamino.wordpress.com/2008/08/08/yet-more-co2/

  • Mark // July 13, 2009 at 9:16 pm | Reply

    The Mark who left the comment on 12 July (c’est mois) is not the Mark who left the next comment on 13 July (ce n’est pas mois). Sorry for the confusion. I’ll use a different name in future.

  • Philippe Chantreau // July 13, 2009 at 10:14 pm | Reply

    Mark, my pedant minute, sorry: me is moi. Mois is a month.

  • David B. Benson // July 13, 2009 at 10:16 pm | Reply

    Mark // July 12, 2009 at 9:27 pm — There is a difference between CN (small) and CCN (larger).

    Mark // July 13, 2009 at 4:33 pm — Antarctica is by far the driest continent. For whatever reason.

  • Mark // July 15, 2009 at 4:13 pm | Reply

    Nah, Mark, I don’t post here often and we get different looking icons it looks like.

  • Mark // July 15, 2009 at 4:16 pm | Reply

    “Antarctica is by far the driest continent. For whatever reason.”

    The interior is. And one of the reasons for that is that outer areas may not have any condensation nuclei, it DOES get cold enough quite easily (especially if you go up a few hundred feet) to get spontaneous formation.

    Another reason for the dryness is that the air circulates around the antartic. But as the temperature differential gets bigger, you will get more air breaking the barrier at the pole.

    Both of which show how antartic land ice can expand when it warms…

Leave a Comment