By Roy W. Spencer, Ph. D., October 11th, 2009
It is claimed by the IPCC that there are ‘fingerprints’ associated with global warming which can be tied to humanity’s greenhouse gas emissions, as if the signatures were somehow unique like real fingerprints.
But I have never been convinced that there is ANY fingerprint of anthropogenic warming. And the reason is that any sufficiently strong radiative warming influence – for instance, a small (even unmeasurable) decrease in cloud cover letting in slightly more sunlight starting back in the late 1970’s or 1980’s– would have had the same effect.
The intent of the following figure from Chapter 9 in the latest (AR4) version IPCC report is to convince the reader that greenhouse gas emissions have been tested against all other sources of warming, and that GHGs are the only agent that can cause substantial warming. (The snarky reference to “proof” is my addition.)
But all the figure demonstrates is that the warming influence of GHGs is stronger than that from a couple of other known external forcing mechanisms, specifically a very small increase in the sun’s output, and a change in ozone. It says absolutely nothing about the possibility that warming might have been simply part of a natural, internal fluctuation (cycle, if you wish) in the climate system.
For instance, the famous “hot spot” seen in the figure has become a hot topic in recent years since at least two satellite temperature datasets (including our UAH dataset), and most radiosonde data analyses suggest the tropical hotspot does not exist. Some have claimed that this somehow invalidates the hypothesis that anthropogenic greenhouse gas emissions are responsible for global warming.
But the hotspot is not a unique signature of manmade greenhouse gases. It simply reflects anomalous heating of the troposphere — no matter what its source. Anomalous heating gets spread throughout the depth of the troposphere by convection, and greater temperature rise in the upper troposphere than in the lower troposphere is because of latent heat release (rainfall formation) there.
For instance, a natural decrease in cloud cover would have had the same effect. It would lead to increased solar warming of the ocean, followed by warming and humidifying of the global atmosphere and an acceleration of the hydrologic cycle.
Thus, while possibly significant from the standpoint of indicating problems with feedbacks in climate models, the lack of a hotspot no more disproves manmade global warming than the existence of the hotspot would have proved manmade global warming. At most, it would be evidence that the warming influence of increasing GHGs in the models has been exaggerated, probably due to exaggerated positive feedback from water vapor.
The same is true of the supposed fingerprint of greater warming over land than over the ocean, of which there is some observational evidence. But this would also be caused by a slight decrease in cloud cover…even if that decrease only occurred over the ocean (Compo, G.P., and P. D. Sardeshmukh, 2009).
What you find in the AR4 report is artfully constructed prose about how patterns of warming are “consistent with” that expected from manmade greenhouse gases. But “consistent with” is not “proof of”.
The AR4 authors are careful to refer to “natural external factors” that have been ruled out as potential causes, like those seen in the above figure. I can only assume this is was deliberate attempt to cover themselves just in case most warming eventually gets traced to natural internal changes in the climate system, rather than to that exceedingly scarce atmospheric constituent that is necessary for life of Earth – carbon dioxide.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Richard: Here’s a little exercise for you – From looking at that figure from the IPCC (and it is probably best to go to the IPCC website and download Chapter 9 so you can blow up the figure), what possible values of tropical tropospheric amplification are each of these three mechanisms compatible with:
(1) solar
(2) aerosols
(3) GHGs
@Dale Rainwater. Shore,
Joel, here is a little exercise for you:
It’s fairly well accepted that there was a rather sudden Arctic warming from 1930-1940. Bengtsson et al (2004) surmised that the warming was due to the NAO and other atmospheric and oceanic circulation changes…
A decadal Arctic anomaly of +1.7°C and a global anomaly of about 0.4°C during a 30-year warm PDO phase totally not cause by CO2.
McConnel et al (2007) presented a fairly solid case for black carbon soot as the main driver of the early 20th century Arctic (and by extension global) warming…
If the NAO and black carbon can both be modeled to have accounted for the early 20th century Arctic warming… Which was it? The Arctic only wamed as much as it warmed… Yet there is strong evidence that two different things caused all of the observed warming,
And… If black carbon could have driven the early 20th century Arctic warming, how can this IPCC chart be accurate? It shows black carbon to have accounted for about 1/9 of the radiative forcing as CO2 from 1750-2005. If CO2 didn’t drive the early 20th century warming or prevent the mid 20th century cooling, how could it possibly have been the dominant radiative forcing mechanism since 1750?
How could it even be the dominant radiative forcing mechanism since 1978, when there is ample evidence that cloud cover and albedo changes can account for most, if not all, of the observed warming of the late 20th century?
Palle et al (2004)
Courtillot found a strong correlation between geomagnetism and climate up until the mid-1980s (Courtillot et al. (2006)) and a strong correlation between solar activity and climate in (Le Mouël et al. (2008)).
If there are multiple other radiative forcing mechanisms that can account for most, if all, of the observed climate change… CO2 cannot be as powerful of a climate driver as the IPCC says it is.
Of course, the fact that the Medieval Warming and Roman Warming Periods were as warm or warmer than today means one of two things: Modern CO2 levels aren’t warming the planet. Or, CO2 levels in the MWP and RWP must have been a lot higher than the ice cores indicate. The same thing applies to the Sangamon Interglacial (~125 KYA)… Either the Earth wasn’t really warmer back then than it is now… Or CO2 levels were higher back then than the ice cores indicate. The Mid-Pliocene Warm Period (~3.2 MYA) was also significantly warmer than today; yet there’s no evidence of CO2 levels being much higher than today’s levels.
The Earth doesn’t suddenly start doing things that it has never done before just because James Hansen makes a computer model and Al Gore gets a Nobel Peace Prize and an Oscar.
P Wilson says:
It is true that the most dramatic cooling is in the early part of the record. The overall rate over the whole satellite period since 12/1979 is -0.407 C per decade. However, even if you start in 1/1993, the trend is -0.113 C per decade. (Starting in 1/1993 lowers it to -.042 C per decade.) I don’t know what the errorbars are on these trends; to compute those correctly would involve taking into account the time correlations in the temperature data…As the time intervals get shorter, of course, the errorbars get larger.
One complication here is that ***lower*** stratospheric temperature is probably not the best thing to look at since this part of the stratosphere is more sensitive to changes in ozone than to CO2. (And, ozone was getting depleted more rapidly before we started to phase out CFCs and other ozone-destroying chemicals.) The middle and upper stratosphere is a better indicator of changes in CO2; it is also probably wise to restrict oneself to lattitudes of less than 60deg, particularly for the southern hemisphere, so that you avoid the most ozone-depleted regions, and I am not sure where the data for that is. (See Scott Mandia’s page for discussion: http://www2.sunysuffolk.edu/mandias/global_warming/greenhouse_gases.html )
John S. says:
There is a confusion between how the net feedback is often discussed in climate and how it is apparently discussed in control theory / system analysis. If you want to define the net feedback in a way where positive feedbacks lead to an outright instability, then you have to include the strong negative feedback that occurs on the radiative imbalance as described by the Stefan-Boltzmann Equation. I.e., an increase in temperature causes an increase in outgoing radiation, which reduces the radiative imbalance.
However, in climate science, people have often described the rise in temperature that the Stefan-Boltzmann Equation implies without any other changes in the climate system as the first-order effect and then talked about the net feedbacks beyond this and how they change the first-order effect. By this definition, a negative feedback reduces the temperature response relative to what the S-B Equation predicts (without feedbacks due to other changes in the atmosphere) and a positive feedback that is less positive than the feedback due to Stefan-Boltzmann is negative will result in a magnification of the first-order temperature response. However, only if the positive feedback due to these other factors is more positive than the feedback due to Stefan-Boltzmann is negative would you get a net positive feedback in the sense that you are talking about and hence outright instability.
To see a discussion in which the response given by the S-B Eq is considered as a negative feedback, see Dennis Hartmann, “Global Physical Climatology”. You can read part of the relevant section on the web here: http://books.google.com/books?id=Zi1coMyhlHoC&pg=PA231&lpg=PA231#v=onepage&q=&f=false This way of talking about the climate system and its feedbacks is apparently more closely aligned with the control theory / system analysis way of talking about things (at least as far as I can judge from the comments of you control theory / system analysis people).
Well, in the common sense of the word, something like the water vapor feedback is called a feedback because the rise in temperature causes an increase in water vapor which feed back on the temperature, causing it to rise more. Not being familiar enough with system analysis lingo, I don’t know exactly how this compares, but it seems like a reasonable use of the term “feedback” from just a basic English language standpoint.
The problems are ones of translating terminology between different disciplines. They are not problems with the actual mathematical calculations themselves.
Those familiar with analog computations using operational amplifiers will tend to compare these types of circuits with climate visualized as a great analog computer. This is misleading, as there is nothing in climate equivalent to an operational amplifier. A better electrical analog to climate would be a network of resistors, capacitors and inductors with the number of elements tending toward infinity and the values of each element varying in time. The only assurance we have of stability is the observation that we are here today in spite of all the disruptions in the climate record.
There is one aspect of climate feedback caused by water vapor that I have not seen mentioned anywhere. This is the consideration of the relative contribution of the two most important greenhouse gases, namely water vapor and CO2. It seems intuitively obvious that the greatest impact of CO2 on the temperature is in the regions where it is the dominant GHG, namely in the polar regions and really dry deserts. Conversely, in the tropical regions, where vater vapor is dominant GHG, the effect of CO2 increase on the temperature would so small that would not be detectable. This presents a problem in postulating a water vapor caused positive feedback. In the regions where the CO2 caused temperature increase is the greatest – polar and desert regions – there is no water to evaporate. Conversely, in the tropical regions, where there is plenty of water vapor generated, the CO2 contribution is so small in relative terms that it would not be noticeable. There are the intermediate temperate regions where both water vapor generation and CO2 caused temperature increase co-exist, so such a simplified argument cannot be used, but the amount of water vapor created positive feedback is likely to be much smaller if the two most important regions – the regions responsible for the most water vapor generation, the tropical regions and regions experiencing the greatest temperature increase due to CO2 increase – are not contributing to it.
Tom: Whether you are correct in where you believe each feedback will be the greatest (which I am not so sure you are), your whole analysis is predicated on the idea that the response to a feedback tends to be localized near where the forcing is. In fact, this is not generally the case. Rather, the predicted temperature response tends to depend only rather weakly on the distribution of the forcing. This is because of the processes that mix the air in the troposphere.
Also, to the extent that such regional variations in forcing do matter, they should in principle be captured and simulated in the full climate models.
By the way, you may be able to find some plots showing the distribution of the radiative forcing due to a change in CO2 over the planet. The book “Global Warming: The Hard Science” by L.D. Danny Harvey shows a plot of the zonally-averaged forcing (and actually shows it generally higher in the lower latitude than the polar regions in some contradiction to what you hypothesize, although one would presumably have to see the full global data to distinguish deserts from tropical regions). The caption implies that the full global data is in principle available (i.e., it says this is from climate model data that has then been zonally average).
Joel Shore I will address the pertinent points in Joel Shore (07:59:13)
Joel Shore (07:59:13) : ..The basic point is that…
The basic point is that there exist negative feedback mechanisms, quite apart from aerosols, that are fundamental to our climate system which keep our climate within certain temperature boundaries. Our temperatures are still well within these boundaries, despite efforts of some alarmists to fudge the data. This negative feedback mechanism the warmist alarmists strenuously try and deny as it goes against their apocalyptic hypothesis.
There was a point I had raised initially if the model simulations point to a hotspot in the troposphere and to a warming of the surface and observations show both to be cooler then this points to negative feedback mechanisms (negative because it works opposite to the “forcing” – though convoluted Joel Shore may say negative is positive and quibble over terminology)
Joel Shore “They don’t directly say that the tropospheric hotspot for the two will be indistinguishable but they certainly don’t say or imply that it won’t.”
Oh really?
What the IPCC says about GHGs is “Greenhouse gas forcing is expected to produce warming in the troposphere,..“. Combine this statement with the GHG graph Figure 9.1 (c), which shows the temperature at 300 hPa as 0.4C warmer than the surface, which could be taken as a “hotspot” and the statement makes sense.
What the IPCC says about solar is “Solar forcing .. (Figure 9.1a) ..in contrast to the response to greenhouse warming, the simulated solar-forced warming extends throughout the atmosphere” Figure 9.1 (a) shows that temperature at 300 hPa as, at most, 0.2C warmer than the surface and far more diffused that the distinct GHG hotspot, which really looks like a “hotspot”, not unlike the red spot of Jupiter and unlike the solar “forcing”.
The contrast between the GHG and solar “forcings” above is consistent with the IPCC statement “The simulated responses to natural forcing are distinct from those due to the anthropogenic forcings”
Joel Shore – You basically come up with the most convoluted interpretation of what the IPCC says as possible, ignoring all the evidence to the contrary. Maybe the IPCC could have been a little clearer but most people try to interpret statements in the way that seems most sensible; you simply do so in the way that suits you. I have noticed this trend in a lot of “skeptics”…
No sir the above is the most straightforward interpretation of what the IPCC has said and it is you and other warmist alarmists like you who ignore evidence. Your interpretation is the most convoluted, in essence saying that what the IPCC clearly have said, they do not in fact say. And its funny but I have noticed this trend in a lot of warmist alarmists.
The IPCC graphs and the RealClimate/ Gavin Schmidt graphs are not about the same things. The IPCC graphs show what is to be expected now, whereas the Gavin Schmidt graphs are when certain conditions are fulfilled (such as a doubling of CO2 – well into the future, or an increase of 2% in the Sun’s irradiance – which will happen god knows when, if ever).
Convolution is in fact what RealClimate and Gavin Schmidt resort to in order to make the GHG and CO2 figures look similar.
1. Maybe someone could check out his graphs from the source he has quoted to verify that they are in fact correct.
2. He has assumed a doubling of CO2
3. He has assumed a 2% increase in Solar “Forcing” .
4. In order to attack the observation that the data does not match model simulations what does he attack? The possibility that the models maybe wrong? Heavens no! How could that possibly be? -Its the data of course. Thats wrong. Do you know what the uncertainty is in the data? How on Earth, with those uncertainties, can you question the certainty of the models?
Joel,
I think you are missing my point. I am not talking about localized feedback effects. I am merely pointing out that the source of postulated positive feedback – and you can take that either globally, or regionally – i.e. water evaporation due to increased CO2 driven temperature increase is not likely to take place at either of the two extremes, i.e the region of the greatest amount of water vapor generation, the tropics, or the greatest relative impact caused by CO2 which are the deserts and polar regions. I am not claiming that the temperate regions may not show some effect, but I am merely subjecting the theory of positive feedback caused by increased water evaporation caused in turn by CO2 driven temperature increase to a very simple sanity check. While I cannot say that the theory fails the sanity check totally, but it does not pass it either.
What I really really need to recognize is that it is impossible to teach people who don’t want to learn. It is simply impossible. When a person’s whole world-view is dependent on thinking about things in a certain way, you can’t break them out of it.
(My last vent applied to Richard’s post, not tom’s.)
Tom, I think that you are missing the point. My point is (1) I don’t even think your estimate of where the CO2 forcing is the greatest is correct and (2) the warming that occurs in response to the CO2 forcing is not localized to where the CO2 forcing is. At the end of the day, what is most important is the radiative imbalance between the Earth’s emission and what it is absorbing from the sun. How that radiative balance is distributed in terms of the location of temperature rises involves more complex atmospheric dynamics. And, then it is those temperature rises that will lead to the water vapor feedback.
Also, to the extent that it is true that, say, a lot of the warming occurs in the polar regions where it is dry, this should be reflected in climate models, so it is not like you are proposing some way in which the models have things wrong. (Or, at least you are not providing evidence that they have it wrong.) You may be right that if the models had more of the warming occurring in the tropical areas, the water vapor feedback would be even stronger. I’m not sure.
Joel Shore (10:49:09):
It’s not just a matter of terminology, but of something much more fundamental. The concepts and mathematical methods of system analysis, which indeed have their historical roots in electrical circuit analysis, have been successfully adapted and assimilated quite intact in wide variety of studies of both inaminate and animate real-world systems. The singularly peculiar exception is “climate science,” whose misguided use of system science terminology (and oversimplified equations) indicates a failure to grasp the basic concepts. This often leads to erudite-sounding, unscientific gobbledygook in explaining system behavior that cannot withstand analytic scutiny. Hartman’s equations and your hand-waving notions of endless chains in explaining “water-vapor feedback” are but two such examples of analytic cluelessness. Increases or decreases of global thermalization relative to the simplistic S-B formulation have nothing to do with feedback, in any scientific sense of the word.
There is a universality to good science that “climate science” does not partake in. It will not advance far beyond its primitive stage in understanding the complexities of climate by standing far apart from the rigorously analytic mainstream.
Re: John S. — (Q about Barkin)
Barkin argues that conventionally-accepted (modeling) assumptions regarding the sphericity, concentricity, & uniformity (of physical properties) of Earth’s shells (core, mantle, hydrosphere, etc.) are overly-simplistic (i.e. they are [for some purposes] insufficient first-order approximations). He appears to be pioneering enhanced models of mutual gravitational interactions (based on relative non-spherical, eccentric mass-centre displacements) between shells and between shells & celestial bodies (sun, moon(s), & planets), with an aim of also accounting for heterogeneous & dynamic elasticity in interactions between shells. Barkin is thus providing a more realistic framework for conceptualizing the spatiotemporal patterns over which factors such as insolation, pressure, & wind integrate. For example, if one shell or some combination of shells shifts north (very slowly) over some decadal-timescale era, this coincides with a north-south asymmetry of pressure (& temperature) in other shells (as dynamic equilibrium adjusts), explaining (to some extent) roughly-see-sawing north-south temporal patterns. Barkin gives specific examples for Earth & other celestial bodies.
Sidorenkov appears to have picked up on the significance of Barkin’s work.
I still have a lot more work to do sifting-through, sorting, & summarizing dozens of Barkin papers, but my impression (at this stage) is that the following Barkin paper is considered seminal:
Barkin Yu.V. (2002). An explanation of endogenous activity of planets and satellites and its cyclisity. Isvestia sekcii nauk o Zemle Rossiiskoi akademii ectestvennykh nauk. Vyp. 9, VINITI, pp. 45-97.
If anyone knows of a free, publicly-available translation of this work, please let us know.
Many of Barkin’s papers deal with the relative motions of solid & interior layers of Earth & other celestial bodies, but I’ve found some evidence that in recent years some of his attention has been devoted to the (terrestrial) hydrosphere & atmosphere (which are the shells Sidorenkov investigates). For example, he refers to the dynamics of 1998 as a “gallop”.
When I have my notes & links more organized (in the weeks & months ahead), I hope to share more with WUWT readers.
Joel Shore (16:19:48) :
You may be right that if the models had more of the warming occurring in the tropical areas, the water vapor feedback would be even stronger. I’m not sure.
———-
If I’m interpreting Tom’s comment correctly, your interpretation is 180 degrees off.
It appears to me that Tom is saying that there is very little or zero anthropogenic CO2-mediated warming occurring in the tropical areas because there is so much water vapor there. The additional CO2 is not capable of additional forcing. Ergo, if there is no warming, there can be no local feedback, either positive or negative, so there is nothing new to incorporate into the “complex atmospheric dynamics” that existed pre-anthropogenic CO2.
I think that passes the sanity check as evidenced by data from one site, at least, in the tropics, as shown:
http://img162.imageshack.us/img162/6073/timcurtintempandco2char.jpg
You said “How that radiative balance is distributed in terms of the location of temperature rises involves more complex atmospheric dynamics. And, then it is those temperature rises that will lead to the water vapor feedback.”
It would follow from Tom’s argument then that any temperature rise and feedback in the tropics would be driven by “complex atmospheric dynamics” from the temperate zones to the tropics. I find that concept somewhat ridiculous on its face.
I’m not suggesting that you said that, or think that, but if Tom is correct, then I for one think that you need a new explanation.
Lastly, surely it is known mathematically what ppms 285 – 385 of CO2 can do in the face of typical water vapor concentrations in the tropics at local temperatures of 80 – 90 F ??
White is black according to the convoluted warmist alarmists. If the IPCC says ““The simulated responses to natural forcing are distinct from those due to the anthropogenic forcings” do they mean it? Certainly not!. How on Earth can you be so naive? That apparently is the most convoluted interpretation of what the IPCC says.
This has to be interpreted by the High Priests of the AGW religion. Because, whereas the IPCC could have been “a little clearer”, there is no difference between the GHG Hotspot and the Solar “Hotspot”, despite the fact that the graphs show that the GHG hotspot is well defined over tropics and the solar is not and the GHG hotspot is at least 0.2C warmer than the solar “hotspot”. But this of course amounts to “no difference”.
The IPCC further says ..the spatial response of surface temperature to solar
forcing resembles that due to anthropogenic greenhouse gas forcing ..Distinct features of the vertical structure of the responses in the atmosphere to different types of forcing further help to distinguish between the different sources of forcing.
Then the IPCC goes on to say “The temporal evolution, and to some extent the spatial and vertical pattern, of the climate response to natural forcings is also quite different from that of anthropogenic forcing. This makes it possible to separate the climate response to solar and volcanic forcing from the response to anthropogenic forcing”
To the uninitiated the above might seem as if they are saying the the spatial and vertical pattern of the GHG hotspot is different to the Solar one but then again you would be horribly naive and foolish to assume so.
Then the IPCC goes onto say in the same document “..a detection and attribution analysis using 13 models from the MMD at PCMDI (Stone et al., 2007a) and an analysis of the National Center for Atmospheric Research (NCAR) Community Climate System Model (CCSM1.4; Stone et al., 2007b)… the response to solar forcing in the model was inferred by fitting a series of EBMs to the mean coupled model response to the combined effects of anthropogenic and natural forcings. In addition, a combined analysis of the response at the surface and through the depth of the atmosphere using HadCM3 and the solar reconstruction of Lean et al. (1995) concluded that.. [in my opinion the conclusion was wrong but is in any case irrelevant to what follows]. This conclusion is also supported by the vertical pattern of climate change, which is more consistent with the response to greenhouse gas than to solar forcing (Figure 9.1)
This again you might assume to be saying that the vertical pattern of change as given in Figure 9.1 between Solar and GHG forcings is different. If they were the same how could the responses be said to be more consistent with one rather than the other. But of course it is actually saying just the opposite.
You have to learn this. And also recognize that it is impossible to teach people who don’t want to learn. It is simply impossible. So you gotta try.
Richard says:
Yes, they mean it. And, from the context in which that statement is imbedded it is clear that they are talking about the fact that the solar forcing warms both troposphere and stratosphere while the anthropogenic forcings warm only the troposphere and cool the stratosphere. Hence the statement, “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 simulated solar-forced warming extends throughout the atmosphere.”
To believe your interpretation, one has to believe that a hotspot in the troposphere is incompatible with the description “warming extends throughout the atmosphere”, which makes no sense at all. The correct interpretation is that cooling of the stratosphere is what is incompatible with the description “warming extends throughout the atmosphere”.
The IPCC estimates that most of the warming that has occurred is due to GHGs, so the GHG response is larger than the solar response EVERYWHERE, i.e., at the surface, in the mid and upper troposphere, etc. What is important for looking at whether the pattern of warming is different is, however, the RELATIVE warming, e.g., by what factor does the mid-upper troposphere warm relative to the surface in the tropics. For GHGs, the graph shows 0.4-0.6 C warming at the tropical surface and 0.8-1.0 C further up in the troposphere…Hence, the amplification factor is somewhere around 2. For solar, the graph shows 0.0-0.2 C warming at the tropical surface and 0.2-0.4 C further up in the troposphere…Alas, this doesn’t tell us much about the amplification factor. It says it is at least 1 but is compatible with any number greater than that. If the surface actually warmed only 0.06 C and the mid- and upper-troposphere warmed 0.30 C, that would be an amplification factor of 5, which is much larger than is seen for GHGs, and such a scenario is compatible with the figure. (In reality, because Gavin has been kind enough to provide a contour plot that is better designed to distinguish the patterns of solar and GHGs, we know that the amplification factor is really about the same for GHGs and solar.)
By the way, if this still confuses you, go to Fig. 9.1 and imagine that you replot the GHG panel but use a contour spacing of 0.6 C instead of 0.2 C. I think you can then see that the GHG panel will look very much like the solar one does now (except for the behavior in the stratosphere, of course). In other words, the reason for the apparent difference between the patterns in the troposphere is that the solar has a lower total magnitude of response and hence the contour intervals are such that the pattern is not well-resolved. This is what can make contour plots a little subtle. You can’t just look at the pretty picture. You have to look closely at the contour interval and think about what the picture means.
It says that they are different but that sentence alone does not say how they are different. When you read the previous discussion in the chapter and when you look at Fig. 9.1 and you are able to understand contour plots, you understand that the difference is in the stratospheric response.
They are not the same. The solar forcing shows warming throughout the atmosphere. The anthropogenic forcings (GHGs alone but even moreso in the case when combined with stratospheric ozone depletion) show a warming in the troposphere and a cooling in the stratosphere. The latter is also what the data show.
John S. says:
So, can you give us a concrete example of how the terminology used in climate science…or the terminology or concepts from system analysis that they have not used…has led the climate scientists astray or has not allowed them to see something important?
philincalifornia says:
And, I am saying that I don’t think this statement about the forcing is correct. In fact, I pointed to a figure (unfortunately not available online as far as I know) that shows that in fact the calculated radiative forcing due to increasing CO2 is larger in the tropics than near the poles. (Since the forcing was latitudinally-averaged in the figure, I don’t know how the forcing compares in dry vs moist areas of the tropics though.)
And in addition to this calculation, I think the logic for arguing that the CO2 is not capable of additional forcing is weak. For one thing, CO2 absorbs at some wavelengths where water vapor is not as major a player. For another, the issue for the greenhouse effect ends up being not whether the absorption is saturated near the surface where the atmosphere is dense but rather what happens high up in the atmosphere where it is much less dense.
Even if tom is correct on the distribution of the forcing, it simply does not follow that the pattern of warming in the atmosphere is proportional to the local forcing. The troposphere is a strongly coupled system. So, even if there is little forcing in the moist part of the tropics, it does NOT follow that there is little warming there and hence little water vapor feedback.
It’s a huge mistake to use a graph from someone like Tim Curtin. First of all, one station does not represent the whole tropics. For another, Tim has not plotted the temperature but rather the change in temperature from one year to the next, so a graph of temperature that showed a steady linear rise (nonzero first derivative but zero second derivative) would be flat on Tim’s graph. For a graph of the temperature anomaly itself at Mauna Loa, see here: http://rabett.blogspot.com/2009/09/first-krammed-then-curtained-one-of.html You will see that there is indeed an upward temperature trend. In fact, the trend is ~0.22 C/decade over 30 years, which is a bit larger than the global trend (although I am sure that there are large errorbars for such an individual station record).
I don’t see why changes in the temperate zone would not tend to affect the flow of energy from the tropics into those zones. However, I also noted that the whole chain of reasoning is incorrect, starting with the assumption that the forcing is lower in the tropical zones.
And, of course, this whole line of argument is basically mute unless you can show that the climate models are getting an aspect of this incorrect. I.e., even if it were true that the pattern of the warming means that there is less feedback from water vapor than if the pattern of the warming were different (which probably is true, in fact), what does it matter as long as the climate models are correctly estimating the pattern of the warming? I.e., if they show less warming in the tropics, then they will naturally have less water vapor feedback occurring there.
“So, can you give us a concrete example of how the terminology used in climate science…or the terminology or concepts from system analysis that they have not used…has led the climate scientists astray or has not allowed them to see something important?”
They dare not estimate error bars like a real scientist would because any error going into a re-iterative GCModel gives an output of pure error bar.
Frankly to talk about a ‘Global Temperature Anomaly’ is simply farcical given the paucity of measuring stations and their uneven spread, and that is before considering the accuracy of their readings in the light of local effects. While on the subject the average of Tmin and Tmax for a day would be an extremely inaccurate approximation to an integrated average with 5 min sampling periods.
These inaccuracies are all of the order of degrees and yet papers get peer-reviewed claiming ‘significant trends’ of tenths of a degree.
It is polite to call climatology a science but I see none of the rigour, discipline and rational thinking that I’d require in a Science. Sadly those climatologists attempting to behave scientifically (ie. providing their raw data) are ostracized by the grant-hunters.
Sandy: I am not sure what your post has to do with system analysis.
At any rate, there are errorbar estimates given for the global temperature anomaly trend in the IPCC report. The errorbars can be smaller than the precision in each measurement because non-systematic errors will tend to average out. Furthermore, the reason that you don’t need that high a density of stations is that temperature anomalies tend to be correlated over a fairly large reason, which is why it is much better to use the anomalies than the absolute temperatures themselves. (See discussion here: http://data.giss.nasa.gov/gistemp/ )
Joel Shore (06:51:50) : The GHG graph shows a distinct hotspot and the solar graph doesnt.
The fact is all the models come up with what should be a hotspot over the tropics between 30 N and S and at a height of 10 km (figure 9.1 f ), which is not there.
The IPCC admits that observations show more warming at the surface than in the troposphere in stark contrast to the models and says that a possible reason is an error or errors common to all models.
I would tend to agree with this explanation and think that the simplest explanation would be feedback mechanisms that work against warming.
Terms get used differently in different fields. I’ve meandered through 7 different fields and people have challenged me to a lot of unnecessary, pedantic discussions. Clear & careful communication is the work-around; people with different mother-tongues find a way. If someone wants to argue that one language should dominate all others on the grounds of universality, that is also a discussion I might evade in favor of more practical pursuits (particularly if all involved parties already appear to understand each other perfectly-well and disagree only on semantics).
Thank you John S. for heightening my awareness of the ‘anomalous’ application of the term “feedback” in Climatology.
Richard says:
You have no way of saying that the solar graph doesn’t. As I have explained, the contour lines are too widely spaced to resolve the structure of the warming for solar. Gavin’s plots show that in fact both mechanisms lead to the same structure of warming in the troposphere.
This first fact is true but independent of the mechanism causing the warming. Also, for both the models and the data, there are errorbars…and for the data in particular, there are some important systematic errors due to artifacts that can affect the measurement of the long-term trend.
Furthermore, the amplification of fluctuations that occur on monthly to yearly timescales as one goes up in the tropical troposphere is in fact seen in the data. Hence, the basic mechanism that produces this does appear to be operating on this timescale. It is difficult to understand how it could be operating on such timescales but not on the longer timescales (since the convective mechanisms that are important in determining the vertical structure in the troposphere operate on timescales of hours to days). This, combined with the known problems with the data for trends on multidecadal timescales is why both the IPCC and the U.S. Climate Change Science Program Report concluded that data errors are the most likely explanation for any remaining discrepancy.
Actually, what exactly the data predicts depends on which analysis of the satellite data or which radiosonde analysis or re-analysis you believe. And, yes, it is possible that it can be due to some error in the models, although the U.S. CCSP concluded that this was less likely than residual errors in the data and for good reasons, as I discussed above.
It is also worth reviewing the history of this field. When the satellite data was first analyzed by Spencer and Christy, it showed cooling globally and this was used as evidence that the surface record was wrong and we were not warming at all. Then, the combination of a longer record and the elimination of errors changed it to warming…but still less warming than at the surface. Then, a still longer record and the elimination of more errors plus the independent analysis by RSS turned it into as much warming as at the surface, at least to within errorbars. The only possible significant left at this point is the trend in the tropics only.
Great. So, explain to me this:
(1) What mechanism do you propose that would have the models and data agree with tropical tropospheric amplification for fluctuations on monthly to yearly-timescales but not for the multidecadal trends?
(2) Given that the most direct effect of there being no “hotspot” is that the models have a negative feedback, the lapse rate feedback, that is not justified if no hotspot exists, why do you think the lack of a hotspot means the feedbacks are more strongly-negative?
(3) In an indirect way, the lack of “hotspot” could be indicative that processes involving the transport of water into the upper troposphere, and hence the water vapor feedback, might not be operating as expected. However, more direct data from the satellites regarding the water vapor feedback show the upper troposphere moistening as predicted. Why would you choose to not believe this data?
Joel Shore (6:56:48):
A clear example of misguided sysytem reasoning is provided by the imputed “positive water-vapor feedback.” For water vapor to enter the atmosphere, evaporation has to take place to begin with, which is invariably a cooling process that correspondingly reduces the surface LW radiation. Once in the atmosphere the vapor does increase the thermal capcitance (though not necessarily the content) of the atmosphere. But it cannot lead to an endless chain of increases, simply because vapor content is physically limited by saturation. And cloud formation invariably reduces the effective insolation, while blocking the return path to the surface for any inceases in atmospheric LW radiation due to the release of latent heat during condensation aloft. Ultimately, precipitation returns the condensation to the surface as cool rain or snow. This, the common hydrological cycle viewed in its entirety, is far from the myopic cry of “tipping points” and “runaway greenhouse.”
For those who might be more familiar with differential equations than with the formal methods of system analysis, an elementary example suffices to illustrate an important point about genuine feedback systems. Let the system be an ideal integrator responding to a forcing signal f(t). It’s output Q(t) then is governed by
dQ(t)/dt = af(t) +bQ(t)
where a is some characteristic response constant and b is the feedback gain. It is readily shown that, for any bounded forcing, the output is bounded only if b is negative. That’s the reason DE texts treat only that case. Positive-feedback control systems are man-made and invariably incorporate artificial limiters. Such systems are nowhere to be found in nature, which operates by its own design.
I hope this puts to rest the present debate, to which I cannot devote more time.
John, excellent post. The water cycle must be too simple for highly educated scientists who have a biased point of some kind to prove.