Climate Science Glossary

Term Lookup

Enter a term in the search box to find its definition.


Use the controls in the far right panel to increase or decrease the number of terms automatically displayed (or to completely turn that feature off).

Term Lookup


All IPCC definitions taken from Climate Change 2007: The Physical Science Basis. Working Group I Contribution to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Annex I, Glossary, pp. 941-954. Cambridge University Press.

Home Arguments Software Resources Comments The Consensus Project Translations About Donate

Twitter Facebook YouTube Pinterest

RSS Posts RSS Comments Email Subscribe

Climate's changed before
It's the sun
It's not bad
There is no consensus
It's cooling
Models are unreliable
Temp record is unreliable
Animals and plants can adapt
It hasn't warmed since 1998
Antarctica is gaining ice
View All Arguments...

Keep me logged in
New? Register here
Forgot your password?

Latest Posts


What causes the tropospheric hot spot?

Posted on 27 June 2010 by John Cook

Previously, we examined how Jo Nova doesn't get the tropospheric hotspot. Recently, Jo posted a response. She points out various semantic errors in my post - how I misquote her and mistakenly describe an effect of a law of physics as a law of physics. Lest we get bogged down in minutiae, I'll concede the semantic mistakes. However, she goes on to repeat the same fundamental confusion - mistakenly thinking the tropospheric hot spot is a signature of the greenhouse effect. She's not alone - this is a common misconception. Here, I'll explain how the hot spot works in more detail (and hopefully with more clarity).

The tropospheric hot spot is due to changes in the lapse rate (Bengtsson 2009, Trenberth 2006, Ramaswamy 2006). As you get higher into the atmosphere, it gets colder. The rate of cooling is called the lapse rate. When the air cools enough for water vapor to condense, latent heat is released. The more moisture in the air, the more heat is released. As it's more moist in the tropics, the air cools at a slower rate compared to the poles. For example, it cools at around 4°C per kilometre at the equator but a much larger 8 to 9°C per kilometre at the subtropics.

When the surface warms, there's more evaporation and more moisture in the air. This decreases the lapse rate - there's less cooling aloft. This means warming aloft is greater than warming at the surface. This amplified trend is the hot spot. It's all to do with changes in the lapse rate, regardless of what's causing the warming. If the warming was caused by a brightening sun or reduced sulphate pollution, you'd still see a hot spot.

There's a figure in the IPCC 4th Assessment report that shows the "temperature signature" expected from the various forcings that drive climate. This figure is frequently misinterpreted. Let's have a close look:   

Figure 1: Atmospheric temperature change from 1890 to 1990 from (a) solar forcing, (b) volcanoes, (c) greenhouse gases, (d) ozone, (e) sulfate aerosols and (f) sum of all forcing (IPCC AR4).

The source of the confusion is box c, showing the modelled temperature change from greenhouse gases. Note the strong hot spot. Does this mean the greenhouse effect causes the hot spot? Not directly. Greenhouse gases cause surface warming which changes the lapse rate leading to the hot spot. The reason the hot spot in box c is so strong is because greenhouse warming is so strong compared to the other forcings.

The hot spot is not a unique greenhouse signature and finding the hot spot doesn't prove that humans are causing global warming. Observing the hot spot would tell us we have a good understanding of how the lapse rate changes. As the hot spot is well observed over short timescales (Trenberth 2006, Santer 2005), this increases our confidence that we're on track. That leaves the question of the long-term trend.

What does the full body of evidence tell us? We have satellite data plus weather balloon measurements of temperature and wind strength. The three satellite records from UAH, RSS and UWA give varied results. UAH show tropospheric trends less than surface warming, RSS are roughly the same and UWA show a hot spot. The difference between the three is how they adjust for effects like decaying satellite orbits. The conclusion from the U.S. Climate Change Science Program (co-authored by UAH's John Christy) is the most likely explanation for the discrepancy between model and satellite observations is measurement uncertainty.

Weather balloon measurements are influenced by effects like the daytime heating of the balloons. When these effects are adjusted for, the weather balloon data is broadly consistent with models  (Titchner 2009, Sherwood 2008, Haimberger 2008). Lastly, there is measurements of wind strength from weather balloons. The direct relationship between temperature and wind shear allows us to empirically obtain a temperature profile of the atmosphere. This method finds a hot spot (Allen 2008).

Looking at all this evidence, the conclusion is, well, a little unsatisfying - there is still much uncertainty in the long-term trend. It's hard when the short-term variability is nearly an order of magnitude greater than the long-term trend. Weather balloons and satellites do a good job of measuring short-term changes and indeed find a hot spot over monthly timescales. There is some evidence of a hot spot over timeframes of decades but there's still much work to be done in this department. Conversely, the data isn't conclusive enough to unequivocally say there is no hot spot.

The take-home message is that you first need to understand what's causing the hot spot. "Changes in the lapse rate" is not as sexy or intuitive as a greenhouse signature but that's the physical reality. Once you properly understand the cause, you can put the whole issue in proper context. As the hot spot is due to changes in the lapse rate, we expect to see a short-term hot spot. We do.

What about a long-term hot spot? With short-term observations confirming our understanding of the lapse rate, that leaves spurious long-term biases as the most likely culprit. However, as observations improve, if it turns out the long-term hot spot is not as strong as expected, the main question will be why do we see a short-term hot spot but not a long-term hot spot?

0 0

Bookmark and Share Printable Version  |  Link to this page | Repost this Article Repost This


1  2  Next

Comments 1 to 50 out of 69:

  1. Just a typo, John. I think 3 paras up from the bottom; last sentence - the word "equivocally" should be "unequivocally"
    0 0
    Response: Fixed, thanks for the tip. Dang semantic glitches.
  2. Re: this sentence:
    The reason the hot spot in box c is so strong is because greenhouse warming is so strong compared to the other forcings.
    I would say " because greenhouse warming has been so strong compared to other forcings over the last 100 years" as that is all the figure is showing. Non-CO2 forcings have been small, so the tropospheric response has been small.

    Perhaps someone could answer this for me: to pose a hypothetical question - if the current warming turns out to have been caused by changes in solar output would we expect to see an identical tropospheric hot spot? Presumably we would, as it's the amount of warming, not the type, that determines the existence of the hot spot, right?
    0 0
  3. Question. Would the "hot spot" be considered a forcing or feedback?
    0 0
  4. According to Santer et al. (2008), International Journal of Climatology 28: 1703-22 [doi:10.1002/joc.1756], "there is no longer a serious discrepancy between modelled and observed trends in tropical lapse rates. This emerging reconciliation of models and observations has two primary explanations. First, because of changes in the treatment of buoy and satellite information, new surface temperature datasets yield slightly reduced tropical warming relative to earlier versions. Second, recently developed satellite and radiosonde datasets show larger warming of the tropical lower troposphere. In the case of a new satellite dataset from Remote Sensing Systems (RSS), enhanced warming is due to an improved procedure of adjusting for inter-satellite biases."
    0 0
  5. RickG,

    Neither, it is simply a signature which the models says should be seen if the earth is warming sufficiently.
    0 0
  6. Thanks David, I was thinking possibly a feedback if anything, but "neither" makes sense. If I understand, correctly its a result rather than a mechanism.
    0 0
  7. RickG,
    if anything, it should be considered as a negative feedback. Indeed it's some forcing that causes the response of a reduced lapse rate which, in turn, increases OLR emission. Note that the opposite is true at mid and high latitudes.
    0 0
  8. J. Neuman, 1955 for an explanation of lapse rate change with latitude and season.

    J.G. Moore, 1956 Distinction between upper and lower troposphere trends.

    I. Thaler et al, 2010 Discussion of current models and recent satellite data. Trend is highly dependent on scenario chosen.
    0 0
  9. To add further information to Johns explanation, there is also another paper on this topic by Santer 2008, and a highly accessible fact sheet that goes with it. Santer also covers some aspects of this debate (and others) in his May 2010 Testimony for House Select Committee on Energy Independence and Global Warming. He concludes that the discrepancy between modeling and observations has largely been resolved.

    The expanding nature of our knowledge of both natural climate change processes in the tropics and of the complexity of effects of anthropogenic forcing and warming on these processes is further illustrated in Sherwood 2010 on the effects of tropospheric humidity changes and polewards shifts of atmospheric zones; Seidel 2007 on observed and modeled widening of the tropical belt, and in Chou 2010 on observed and modeled weakening of the tropical atmospheric circulation. All of these and many more strands of emerging evidence add to the robust global climate trends which are already generally accepted and appear consistent with global warming associated with a major CO2 forcing component.
    0 0
  10. In a previous post I suggested a categorisation of signatures and fingerprints.

    This one appears to fall into the same category as the C13/C12 ratio.

    A secondary signature.

    Your first line of proof/defence should be robust primary signatures, backed up by supporting secondary signatures?
    0 0
  11. Thank you, for this is much, much improved on the previous post.

    I'll note that while it's difficult to nail down this long-term trend in the observations, the short term behavior is possibly consistent with theory.

    After all, the satellite records show amplified variability in the troposphere with El Ninos and La Ninas. I think this observation may be related to the same mechanism, just over the short term.
    0 0
  12. RickG: It ends up being a negative feedback, in a way that is perhaps not obvious.

    For the greenhouse effect to work, there has to be a lapse rate. The emission that makes it back out to space needs to be from a colder place than the surface. But due to this 'hot spot' effect, the lapse rate is reduced. So the greenhouse effect becomes less powerful than it otherwise would have been.

    But there's also a positive feedback in the story, as the increased atmospheric moisture that causes the 'hot spot' itself causes a major positive feedback, as water vapor is a greenhouse gas.

    So in short: the "hot spot" is caused by something that is a positive feedback, but it itself causes a negative feedback.

    Hope that wasn't muddled.
    0 0
  13. I have to say to John, bravo for keeping above the fray. Again, reading the tone of both the article and the discussion on JoNova's site is rather alarming. It really does little to address the scientific issues regarding climate change when people stoop to such personal attacks. I'd like to better understand why Nova and her cohorts reject this stuff and where they're getting their information. I honestly would. But I stop reading when it gets personal.

    I also have to thank those who post on this site with their perspectives that reject AGW for maintaining a sense of decorum. It makes a huge difference. I'm sure you often feel outnumbered here but I hope that you feel you get a fair hearing.

    If you ask me, that is what skepticism is all about. A fair hearing from all sides.
    0 0
  14. Ms. Nova continues be “mistakenly thinking the tropospheric hot spot is a signature of the greenhouse effect”

    There is so much wrong with her aggressive attack on this point. Is she denying that the greenhouse effect exists based on this misunderstanding? The greenhouse effect was around before we humans started emitting GHGs. That there is a “greenhouse effect” is very well established and not disputed, even by reasonable “skeptics”. Or is her misunderstanding of the science leading her to erroneously argue that we are not experiencing an enhanced greenhouse effect on account of higher concentrations of GHGs from human activities? Or is she concerned that the TROPICAL “hot spot” is not as strong as predicted by some models?

    The biosphere has been experiencing an enhanced greenhouse effect for decades now. How that manifests itself has been clearly documented by long term trends in multiple, independent observations. For example, long term warming in the SAT record, long term cooling of the lower stratosphere, long term increase in OHC (see Murphy et al. 2009 and others), and an increase in the height of the tropopause (Santer 2003) as evidenced by radiosonde data.

    IMO, Nova is being disingenuous with here musings on this topic and seems to be doing here best to obfuscate. That is evidenced by the lack of caveats in her musings as to the limitations of both the models and the observations—she seems to be naïve about the real-world complications of working with radiosonde (different sensors and platforms, sensor changes, sparse coverage etc.) data and MSU data. I also find it intriguing that she seems to claim to know more about this complex field than experts such as Santer et al. She should also read the README file for the UAH data sometime and see for herself the multitude of errors, corrections and other issues with that particular product.

    I take issue with the first graph that she presents in her latest post. First, the graph on the right panel is labeled “no hot spot” when in fact, there is warming, it is just not as significant as that predicted.

    The discrepancy between the models and the observations is complex. Partly b/c the maximum warming is predicted to occur near the tropopause, a region not sampled well by MSU products such as RSS b/c of “contamination” from the stratosphere—Ms. Nova does not seem to understand that the satellites do not directly measure temperature but radiances from relatively deep/coarse layers in the atmosphere-- a frustrating problem when one is trying to make measurements near regions of sharp temperature gradients such as between the troposphere and lower stratosphere (region where “hot spot” is expected to be a maximum. Not to mention the fact that the troposphere is warming while the lower stratosphere is cooling, so that further complicates trying to extract meaningful long-tern trends. The UW adjusted RSS product (which address the “contamination” issue) finds that, globally, the mid-troposphere has been warming of +0.15 K/decade since 1979 (from NCDC, annual temperatures up to and including 2009).

    On the model side, the AOGCMs do have issues with handling moist convection (i.e., thunderstorms), primarily because their horizontal grid spacing is too coarse. Consequently, they may be being too aggressive in their tropospheric warming over the tropics where convection predominates.

    IMHO, at most, this is a debate as to much to do with the limitations of the AOGCMs simulate convection and complex issues associated with accurately sampling and resolving temperature changes the upper-troposphere. The observed discrepancies have nothing whatsoever to do with whether or not the greenhouse effect is real, or whether or not we are experiencing an enhanced greenhouse effect. There are more appropriate and reliable measures which support the existence of an enhanced greenhouse effect. As far as I can tell, Ms. Nova’s beloved “missing hot spot” also has nothing to do with global equilibrium climate sensitivity.
    0 0
  15. Robhon @13,

    I would like to echo your thoughts. The vitriol and invective in Ms. Nova's missive is both unacceptable and unprofessional, and only goes to undermine whatever credibility she thinks she might have on this issue.
    0 0
  16. I read Nova's bit carefully and I note that she can produce a reasonable explanation of why a "hot spot" will exist, actually an explanation that worked well for me. She does not dispute that the phenomenon really must exist. Yet she refuses to acknowledge that failure to see the hotspot by instruments is a fundamental problem and that comparisons w/model predictions can't be made until the instrumentation issue is resolved.

    More rhetorical impressionism. I'm close to settling on the term "impressionist" for a fairly major swath of Nova-type persons. Prestidigitation is another possibility but it over-describes.
    0 0
  17. I think it's important to note that in Nova's 'Skeptic's Handbook' she is clearly of the belief that the tropospheric hot spot is a signature specific to AGW; she refers to "the telltale “hot spot” warming pattern that greenhouse gases would leave".

    John, in his initial post, pointed out that this was incorrect, stating "we expect to see an amplified warming trend in the troposphere no matter what’s causing the warming".

    As a result of John's correction, Nova seems to have changed her position, and in her recent article in response to John's post wrote:

    ...strictly any form of warming ought to increase evaporation, increase humidity, and in the world of climate models, raise the level of warming 10 km up over the tropics (ie, create a hot spot).

    I think the fact that Nova has admitted her mistake on this is important. It's not just about semantics.

    If the hot spot was just a signature of AGW, then those arguing that it is absent would have an easier task. They could say, "yes there is warming, but the lack of a hot spot proves it is not humans". The fact that the hot spot acts as signature for warming from any source means that if people want to argue it is absent, they now have to say "no there is no warming, as demonstrated by the absence of the tropospheric hot spot". This would be a difficult argument to make considering all measurements show a warming trend.
    0 0
  18. You're talking about the hot spot that isn't there right?

    It's hard when the short-term variability is nearly an order of magnitude greater than the long-term trend.

    That's a keeper! Thank you for stating the obvious, even while stubbornly refusing to understand it's importance.
    0 0
  19. The depiction of the hot spot I assume is the average of data collected around the full circumference of the globe. Is there any depiction of how the hot spot varies, or should vary across the various regions of vastly different surface conditions?
    0 0
  20. papertiger @ 18: Who's refusing to understand it's importance? That is, after all, the entire point of the post. Short term variability often masks subtle long-term shifts, and this hot spot is no different. Look at this post for an example of the effect, where annual variation in CO2 levels is quite large, although over longer periods the rising trend dominates.

    I get the impression that you regard a long-term shift to be insignificant if it's less than the short-term variability in the system? If so, then I'm sure you wont mind terribly if the average temperature in your neighbourhood increases by 10ºC or so, as that's still less than daily variability...
    0 0
  21. 19, johnd

    The figures above are model results, not observation, but the x-axes are latitude, which is what you seem to want.

    For observation, there are different sources; the last article here linked to one that has charts broken down by NH, tropics, SH.
    0 0
  22. carrot eater at 10:25 AM, it was along the various longitudes I meant. Both the models and the observations should vary somewhat given the different surface conditions and hence perhaps give better understanding to what is expected, and what does or does not actually occur.
    As for the latitude cross section, I would have thought that there would have been some bias in the model given the quite significant difference in the observed northern and southern hemisphere temperatures.
    0 0
  23. It is interesting that the skeptics are so eager to misinterpret Figure 1 that they miss the real GHG signal that it clearly identifies--tropospheric warming combined with stratospheric cooling--which none of the other forcings could possibly produce.* How do they explain the fact that exactly that signature has been observed?

    *Except, in the figure, ozone, which I've been under the impression *is* a GHG.
    0 0
  24. John,

    There are a couple of points in your article that I'd like considered together because they are causing me some concern.

    1) "the hot spot is well observed over short timescales". (I assume you mean the trend here not just the fact the hot spot exists)

    2) "the short-term variability is nearly an order of magnitude greater than the long-term trend"

    I wonder to what extent you can be confident of the first fact given the existence of the second. How can we be sure that the short term trend is not just a happen-chance product of this natural variability? In many other aspects of climate it seems appropriate to be wary of short term trends for this very reason.
    0 0
    Response: Short-term trends are totally due to internal variability. Eg - El Nino. So what we observe is the surface temperature ups and downs due to El Nino are magnified in the tropical troposphere.
  25. The paper below observes a global 0.7% increase in lapse rate for a 1 K warming of surface. It means high above the ground the temperature anomaly should vanish (this surface is supposed to be at a much lower level above the poles than above low latitudes).

    They also say lapse rate is about 6.5 K/km at low latitudes and 4.5 K/km at the polar region, which is roughly consistent with the 0.7% figure (implies a 52.7 K difference in surface temperatures between poles and tropics, which is reasonable).

    How is it consistent with your statement ˇ"When the surface warms, there's more evaporation and more moisture in the air. This decreases the lapse rate - there's less cooling aloft. This means warming aloft is greater than warming at the surface. This amplified trend is the hot spot"?

    A detailed explanation would be welcome.

    Izvestiya Atmospheric and Oceanic Physics
    Volume 42, Number 4 / July, 2006, pp. 430-438
    DOI: 10.1134/S0001433806040037
    Tropospheric lapse rate and its relation to surface temperature from reanalysis data
    I. I. Mokhov and M. G. Akperov
    0 0
  26. HumanityRules,
    I think you might be confusing some matters. Maybe read the links put there, like this one

    The surface record shows a lot of short-term variability, and that variability is even stronger in the troposphere.

    But given that large variability, and the quality of the observations, it is difficult to observe any long term trends.
    0 0
  27. Berényi Péter,
    the hot spot is called tropospheric for a reason. And this is from the article you cite:
    "The correlation of γ with Ts is negative in the equatorial latitudes and midlatitudes over the oceans."
    Hence, following that paper the lapse rate feedback is negative in the tropics but postive overall, more so at the high latitudes.
    0 0
  28. Thanks for the post, John. This was something I've been struggling to understand, too.
    0 0
  29. #27 Riccardo at 22:30 PM on 28 June, 2010
    the lapse rate feedback is negative in the tropics but postive overall

    Let me understand. Are you saying with increasing surface temperatures overall rate of evaporation increases but the area where this happens decreases?
    0 0
  30. 26 carrot eater

    It's not the failure to find a long term trend that I'm concerned about but how, in a highly variable data set, we can see the short term trends. Presumably these short term trends associated with AGW are also an order of magnitude below the variability unless the short term trend is an order of magnitude greater than the expected trend, which seems unlikely.
    0 0
  31. Humanity Rules,
    You are all twisted up here.

    Nobody is talking about "short term trends associated with AGW". That doesn't even have meaning.

    We're talking about the short term variability, related to El Nino, La Nina, etc. Compare some satellite and surface records. You'll see, for example, the El Nino peak of 1998 is more exaggerated in the satellite records than it is in the surface records. That's the sort of thing being discussed here. Or, for a more careful analysis, just look at the linked papers.

    Remember, the 'hot spot' is not unique to greenhouse gases. It should exist, no matter what caused the surface to be warmer - including internal variability like ENSO.
    0 0
  32. First, I wonder if not the enhanced spikes in tropospheric temps relative to surface temps during El Ninos (see e.g. could be illustration of the short term phenomena in question. (Not "trends".)

    Second, the denialist position is logically flawed in this case, too. For what happens to GW if the hot spots don't get very hot? As carrot eater related, there are two feedbacks in operation, a positive WV and a negative LR. Nothing much happens with the positive vapor feedback, BUT the negative lapse rate feeback, which is quite significant in GW, becomes much weaker without hot spots. (It does by no means disappear, though.) SO, the net effect, everything else equal, will be that the same amount of forcing will produce MORE warming without hot spots. Consequently, unless balanced by other effects, data speaking against hot spots are arguments for higher sensitivity.
    0 0
  33. Berényi Péter,
    I wrote "following that paper", so it's not my idea but the results shown there.
    Anyways the answer to your question I think is no, but it's not clear to me where it comes from. Explaining the band (their fig. 3a) of negative correlation in the southern ocean the paper says:
    "negative correlation of γ with Ts can be attributed to a combination of the thermally inertial ocean and quite strong variations in the annual cycle of temperature at the top of the troposphere (with relatively small variations in surface temperature)."
    In other words it appears to be due to the annual cycle and to the associated changes in atmospheric circulation. When talking about the so called tropical hot spot one usually refers to the multi-year trend of the anomaly, which is not the same as what is shown in Mokhov et al. paper.
    0 0
  34. 31 carrot eater

    Here's a quote from Johns previous post "Jo Nova doesn't get the tropospheric hotspot"

    "The result is we expect to see magnified warming trends in the troposphere compared to the surface, both over short intervals (say months to a year) and long intervals (over decades). Indeed over short periods, observations are consistent with expectations - a tropical hot spot"

    John was first to raise the idea of a short term trend. That is what I was questioning. I'm happy to agree with your post.
    0 0
  35. Why would the satellite temperature record be subject to "spurious long-term biases"?

    We seem to be happy with the data when it confirms the surface instrument record or when it confirms stratospheric cooling but this mid section of the data is all wrong. It seems horribly convenient. In fact it appears akin to Antony Watts hunt for those badly placed weather stations.
    0 0
  36. HumanityRules:

    Read 'short term trend' as referring to the observed high frequency behaviour, all the short term wiggles due to ENSO and whatever else. The jump upwards during an El Nino can be described as a short term trend. I did not find John's phrasing at all confusing, but that's what is meant.

    We already have known that satellite records have been subject to long-term biases and calculation errors in correcting for the same. They've been continually corrected in UAH, one by one. The remaining differences between UAH and RSS also beckon.

    Any time you've got satellite drift, or you've got to sew together records from different and non-overlapping satellites (an issue with TSI, isn't it?), you'll have to be careful with long term biases.
    0 0
  37. I was prompted by this post to pay a visit to Jo Nova's site. I don't know much about Ms Nova herself but the comments were a real snake pit - not pleasant.
    One guy got great satisfaction (and approval from others) by countering the argument that the THS is not specifically a fingerprint of AGW by pointing to the IPCC's statements that tropospheric warming/stratospheric cooling is a fingerprint of AGW. But to my layman's eye they are completely different phenomena and irrelevant to this argument.
    0 0
  38. One of the issues at the heart of the matter is the trend differences between UAH and RSS satellite temperature estimates for the tropical lower and mid troposphere, both based on the MSU (microwave soundings) raw data from various satellites.

    Santer 2008
    uses data from 1979 to 2000, Bengtsson 2009 uses a similar methodology with (Ocean only) satellite data updated to 2008, and I have updated the satellite trend values to current, with the latest UAH LT5.3 data. All trend value are degrees C per decade.

    Santers analysis of the satellite trends and surface temperature and Radiosonde trends to 2000 pointed to a much reduced discrepancy between observations and model outputs than found previously (see figure below). He also suggests where models may be lacking.

    Bengtsson 2009 follows on from this but uses the later lower trend values. Based on these and a modeling/statistical approach, the probability of the satellite temperature trends being due to natural causes is given as 27% for the UAH measurements and 2.5% for the RSS measurements. Maybe reasonable odds, but not “robust” yet except perhaps in the case of the RSS trend.

    Bengtsson also argues that the UAH values are closer to observed SST, but conversely Santer 2008 suggests the RSS values are closer to other global temperature series, and other interpretations of the MSU raw data. This relatively small RSS/UAH difference weighs heavily.

    Bengtsson concludes “Observed and re-analyzed lapse rate trends are all positive and for the period 1979-2008 well outside the range of natural variability”, but in terms of temperature trends, “The present 30-years of tropospheric temperature observations are still insufficient to identify robust trends as the internal variability of realistic climate models is larger than the observed trends”

    Has the situation changed since 2008/2009?

    A little. The UAH/RSS divergence has reduced with the latest revisions and data, the RSS trend values are slightly lower but the revised UAH tropical trend values have increased (I should mention UAH global trends did not change with the update) so that they are higher than in Santers original analysis. As the trends have continued (ie troposheric temperatures have continued to rise) we would expect that the updated 2010 data will push further towards (rather than away from) any statistically robust result.

    The following image is from Santer 2008 and summarises the story of the models and measurements as at 2008 quite nicely.

    If you view the JoNova post linked in the article you will see a related but older chart from Santer 2005, which supports the idea of significant divergence despite Jo citing the later 2008 paper in which Santer argues otherwise. Likewise her second figure should be updated in line with more recent work, as science has moved on.

    I also have serious concerns about the one later reference which Jo uses (Paltridge 2009) to support a view that tropospheric relative humidity is falling. This is at odds with the conclusions of the bulk of recent papers I have read, and Paltridge himself states “It is accepted that radiosonde-derived humidity data must be treated with great caution” the data is single source (NCEP) re-analysis.

    As well as Sherwood 2010a listed earlier, and evidence from Sherwood 2010b, he has also completed a recent review of independent work examining tropospheric water vapour using multiple data sets from several types of sensor Sherwood 2010c which adds a more thoroughly referenced and wide perspective on Tropospheric water vapour:

    “Thus, all primary data sets support the conclusion that water vapor mixing ratios in the troposphere are increasing at roughly the rate expected from the Clausius-Clapeyron equation. Although a few analyses have found otherwise, these relied on secondary data sets that are less suitable for quantifying trends”.
    0 0
  39. Peter,

    Really interesting post, thanks. Particularly appreciate the (newer) Santer graph.

    Re: Paltridge, here’s a couple of quotes from the study itself:
    Radiosonde humidity measurements are notoriously unreliable and are usually dismissed out-of-hand as being unsuitable for detecting trends of water vapor in the upper troposphere.
    It is of course possible that the observed humidity trends from the NCEP data are simply the result of problems with the instrumentation and operation of the global radiosonde network from which the data are derived. The potential for such problems needs to be examined in detail.

    Despite these caveats, Paltridge does, quite rightly I feel, argue that “the NCEP data for the middle and upper troposphere should not be “written off”…Since balloon data is the only alternative source of information [as opposed to that taken from satellite measurements] on the past behavior of the middle and upper tropospheric humidity and since that behavior is the dominant control on water vapor feedback, it is important that as much information as possible be retrieved from within the “noise” of the potential errors.”

    However, on the recommendation of the Elliott and Gaffen study (1991), Paltridge’s study only covers the reanalysis data from 1973 to 2007 and limits its examination to particular latitudes between 50° S and 50° N, and atmospheric pressures up to up to 500 hPa everywhere, together with the summer season data from 400 hPa, and the data up to 300 hPa in the tropics. This is because the radiosonde measuring system isn’t accurate enough to measure changes in humidity in locations where humidity is already at comparatively low levels and because any radiosonde humidity measurements prior to 1973 are unusable as a result of instrumental changes and deficiencies.

    Paltridge also bases his findings on a combination of observations and models (you know, the things Nova hates). This report – - notes that “gridded variables, the most widely used product of the reanalysis, have been classified into three classes”; moisture variables, upon which Paltridge would have relied, fall into the category, ‘Type B Variables’, which the report describes as being “influenced both by the observations and by the model, and are therefore less reliable [than Type A Variables which "are generally strongly influenced by the available observations"]“.

    In addition, on both the NCEP reanalysis website and the NCAR reanalysis website a ‘problem report’ is given, discussing the issues associated with the data. One such issue is titled ‘Spurious Moisture Source/Sink’. In brief, it states that “a poor approximation was used for the humidity diffusion which created spurious moisture sources and sinks”; amongst other things this “can be expected to increase/decrease humidity”.

    And of course the radiosonde measurements contradict the satellite measurements -
    0 0
  40. Ah, yes Andrew. I made the same mistake myself. Did they point to where in the IPCC report that it says that? I would be curious.
    0 0
  41. Andrew,
    Your instincts are correct.

    If the surface is warming, then the tropical troposphere should warm faster than the surface, no matter what is causing the warming. So there is no fingerprint there.

    But above that, if the stratosphere is at the same time cooling, that is indeed a fingerprint of enhanced greenhouse effect. Caveat being that ozone loss also causes strat cooling, but that effect is more limited to a certain altitude band.
    0 0
  42. Isn't the GHG fingerprint the _divergence_ of stratospheric and tropospheric temps? I.e. the stratosphere constant or cooling while troposphere is warming, or the stratosphere cooling with the trop. constant or warming? And that fingerprint does not have to be very strong.
    0 0
  43. robhon,

    They did point to the particular passage in the IPCC report - I don't think they misquoted it, it just doesn't support the point they were making.
    0 0
  44. SNRatio at 16:30 PM on 29 June, 2010 (or related anyway)

    Bengtsson 2009 suggests Lower Troposphere Temperature minus Mid Troposphere Temperature (TLT-TMT, or T2LT-T2) is a servicable approximation to lapse rate. Sorry for not pointing this out explicitly above after my table, but this has increased unambiguously over the record. Even between the analysis in the papers referenced. Santer (data up to 2000) gets 0.024 and 0.023 for RSS and UAH data respectively. Bengtsson (data up to 2008) gets 0.035 and 0.036, and the May 2010 values give 0.037 and 0.035 (RSS and UAH respectively). On this the satellite records agree very well.

    It should also be appreciated that the TLT and TMT are not measurements within distinct layers, but represent weighted sums as we ascend so that for example 90% of T2 data is from troposphere surface to 18km, whilst 90% of T2LT is surface to 8km. Radiosondes can of course give data at specific heights, the four most recent primary sonde data sets show clear tropospheric warming higher than at surface, see figure above.
    0 0
  45. The hot spot is not a unique greenhouse signature and finding the hot spot doen't prove that humans are causing global warming.
    That's true, John. The hot spot may be caused by more than one phenomenon. But what does not finding the hotspot mean? That means at least, that the greenhouse signature is weaker than most AGW advocates think. And that is exactly what Jo Nova is telling us. Climate sceptics do not deny the fundemental physics underlying the AGW-theory. Their message is: (1) global warming is less then we are told, (2) the warming is less caused by the greenhouse effect then we are told and (3) the catastrophic effects of the warming are les then we are told.
    The problems that John Cook and others have with identifying the hot spot confirm the second point of the sceptic views.
    0 0
  46. fydijkstra,

    You commend John for stating that "The hot spot is not a unique greenhouse signature", and then assert that not finding it must mean "that the greenhouse signature is weaker than most AGW advocates think".

    Actually, if we accept the large caveat that the hot spot hasn't yet been found (see Santer 2008 amongst others), we can conclude one of three things:

    1) Our measurements have not been accurate enough;
    2) The models that predict the tropospheric hot spot in response to warming are wrong;
    3) There is in fact much less warming than our measurements tell us there is.

    Number one is not popular with the 'sceptics', which leaves numbers two and three. If we go with number two we have to accept that the laws of physics are wrong (at least that's my understanding - I'm not a modeller), and if we go with number three we have to accept that all of our surface and satellite measurements are wrong. Both of these arguments are difficult to make.

    As I pointed out at #17, the fact that the hot spot is not a unique greenhouse signature means that its perceived absence would have far wider ramifications than simply saying humans aren't causing global warming. In light of that, I find it far more conceivable to conclude that the hot spot has in fact been found (see Santer 2008), than that all of our surface and satellite temperature measurements are wrong, or that the laws of physics need revising.

    If the hot spot was unique to the greenhouse signature the 'sceptics' would have a much easier job.
    0 0
  47. Some commenters here are posing the question "would we see a hot spot if the warming was caused by solar output".
    I'm happy to let the IPCC AR4 answer that.

    From the FAQ’s document, page 120-121..

    Models and observations also both show warming in the lower part of the atmosphere (the troposphere) and cooling higher up in the stratosphere. This is another ‘fingerprint’ of change that reveals the effect of human influence on the climate. If, for example, an increase in solar output had been responsible for the recent climate warming, both the troposphere and the stratosphere would have warmed.

    And from page 674 of Chapter 9 WG1

    The simulated responses to natural forcing are distinct from those due to the anthropogenic forcings described above. Solar forcing results in a general warming of the atmosphere (Figure 9.1a) with a pattern of surface warming that is similar to that expected from greenhouse gas warming, but in contrast to the
    response to greenhouse warming"

    The charts regarding this hot spot are on the same page in the report, that is, the charts back-up what the report says. And it says that due to anthropogenic warming, there SHOULD be a hot spot that is about 2 to 2.5 times greater than the surface warming.

    If the surface has warmed by 0.7DegC then the hot spot should have warmed by 1.4 to 1.74DegC Surely large enough to detect.
    If however the hot spot hasn't warmed enough to detect, then the surface couldn't have warmed by 0.7DegC. One or the other, can't be both. Discuss
    0 0
  48. Baa Humbug,

    This is a difficult topic and unfortunately pages 674 and 675 of AR4 Chapter 9
    are causing a great deal of confusion - although that's not to say they're wrong.

    You quote from the FAQs document, page 120 - 121. I repeat:

    Models and observations also both show warming in the lower part of the atmosphere (the troposphere) and cooling higher up in the stratosphere. This is another ‘fingerprint’ of change that reveals the effect of human influence on the climate. If, for example, an increase in solar output had been responsible for the recent climate warming, both the troposphere and the stratosphere would have warmed.

    The quote above says that an increase in solar output, and anthropogenic contributions, should both result in a warming of the troposphere. As your quote testifies, this agrees with observations. It also says that anthropogenic contributions, but not an increase in solar output, should result in a cooling stratosphere. Once again, this prediction agrees with observations.

    The human 'fingerprint of change' to which the IPCC refers is not a warming troposphere alone (which could be induced by changes in solar output), but a warming troposphere combined with a cooling stratosphere.

    Your second quote is necessarily taken out of context, but in the process loses some of its intended meaning. The loss of meaning is exaggerated by the fact that it was poorly and ambiguously worded in the first place.

    The charts accompanying the quote (which are shown in Figure 1 of John's article) are labelled as follows:

    Figure 9.1. Zonal mean atmospheric temperature change from 1890 to 1999 (°C per century) as simulated by the PCM model from (a) solar forcing, (b) volcanoes, (c) wellmixed greenhouse gases, (d) tropospheric and stratospheric ozone changes, (e) direct sulphate aerosol forcing and (f) the sum of all forcings. Plot is from 1,000 hPa to 10 hPa(shown on left scale) and from 0 km to 30 km (shown on right). See Appendix 9.C for additional information. Based on Santer et al. (2003a).

    That is to say that the charts model zonal mean atmospheric temperature change from 1890 to 1999, and assign varying proportions of this change to different forcings. The fact that since 1890 the largest forcing by far has been well-mixed greenhouse gases - or at least that's the 'assumption' the IPCC has based its model on - means (a), and also (f) - the sum of all forcings - show the most prominent hot spot. None of the other forcings have been sufficient in the last 100 years to create a hot spot of similar magnitude. As I pointed out at #2, non-CO2 forcings have been small, so the tropospheric response has been small. Figure 9.1a is not, I repeat not, a demonstration of the possible zonal mean atmospheric temperature changes associated with different forcings. It is merely a modelled image of changes over the last 100 or so years.

    The wording of the IPCC quote you provided is therefore very poor. It perhaps should have read "The simulated responses to natural forcing [over the last 100 years] have been distinct from those due to the anthropogenic forcings described above".

    If we run the GISS model equilibrium with 2xCO2 or a 2% increase in solar forcing (forcings in the same order of magnitude) we get the following pattern of zonal atmospheric temperatures:

    For 2% increase in solar forcing

    For 2xCO2

    As you can see, the main difference occurs in the cooling stratosphere, but the tropospheric hot spot is clearly evident in both.

    You can also see this pattern in picture (a) on the charts you referred to, however, as a result of the lower levels of solar forcing the pattern is not as pronounced.

    So to clarify, any warming of a sufficient magnitide should be expected to create a tropospheric hot spot.

    If you then wish to propose that the hot spot has not been found, you have to argue for one of the three conclusions I highlight in #46.

    By the sounds of it you'd like to argue point 3) - that there is in fact much less warming than our measurements tell us there is. In which case I refer you to 'Are surface temperature records reliable?'
    0 0
    Moderator Response:

    I'd like to correct this post slightly, as I think part of it is misleading.

    As 'andrew adams' and 'e' have alluded to, the wording of the IPCC quote is not in fact 'poor'. That adjective can be reserved for my interpretational ability.

    The following quote is in fact perfectly accurate as it refers to the general pattern of warming troposphere/ cooling stratosphere, as opposed to the existence of a tropospheric hot spot, which is a seperate phenomenon:
    The simulated responses to natural forcing are distinct from those due to the anthropogenic forcings described above.
    The confusion comes because it links to a set of charts which, by necessity, show the tropospheric hotspot in addition to the general pattern of tropospheric warming/ stratospheric cooling.

  49. fydijkstra... I've not read enough of JoNova's post to comment on whether she rejects the basic theories behind AGW but the crowd who posts on her site most certainly does. I spent about two days posting on her site and was barraged with people telling me, in no uncertain terms, that CO2 has absolutely no affect on global temperature or climate. Arrhenius was wrong. When I posted a video of a simple lab experiment using a cylinder filled with CO2 and an infrared camera, it was brushed off as a trick.

    JoNova may not agree with that position but she, and a small handful of people who certainly know better, are not making even the slightest attempt to correct their errors.
    0 0
  50. David and Peter,

    Thank you both for your very informative posts.

    David, in your post at #48, shouldn't the lower figure be for 2xCO2, and the top figure for 2% increase in solar forcing?
    0 0

1  2  Next

You need to be logged in to post a comment. Login via the left margin or if you're new, register here.

The Consensus Project Website



(free to republish)



The Scientific Guide to
Global Warming Skepticism

Smartphone Apps


© Copyright 2014 John Cook
Home | Links | Translations | About Us | Contact Us