NEW 4/10/09: There is an update to this post, see below the “read the rest of this entry” – Anthony
Guest Post by Richard Lindzen, PhD.
Alfred P. Sloan Professor of Meteorology, Department of Earth, Atmospheric and Planetary Science, MIT
This essay is from an email list that I subscribe to. Dr. Lindzen has sent this along as an addendum to his address made at ICCC 2009 in New York City. I present it here for consideration. – Anthony
Simplified Greenhouse Theory
The wavelength of visible light corresponds to the temperature of the sun’s surface (ca 6000oK). The wavelength of the heat radiation corresponds to the temperature of the earth’s atmosphere at the level from which the radiation is emitted (ca 255oK). When the earth is in equilibrium with the sun, the absorbed visible light is balanced by the emitted heat radiation.
The basic idea is that the atmosphere is roughly transparent to visible light, but, due to the presence of greenhouse substances like water vapor, clouds, and (to a much lesser extent) CO2 (which all absorb heat radiation, and hence inhibit the cooling emission), the earth is warmer than it would be in the absence of such gases.
The Perturbed Greenhouse
If one adds greenhouse gases to the atmosphere, one is adding to the ‘blanket’ that is inhibiting the emission of heat radiation (also commonly referred to as infrared radiation or long wave radiation). This causes the temperature of the earth to increase until equilibrium with the sun is reestablished.
For example, if one simply doubles the amount of CO2 in the atmosphere, the temperature increase is about 1°C.
If, however, water vapor and clouds respond to the increase in temperature in such a manner as to further enhance the ‘blanketing,’ then we have what is called a positive feedback, and the temperature needed to reestablish equilibrium will be increased. In the climate GCMs (General Circulation Models) referred to by the IPCC (the UN’s Intergovernmental Panel on Climate Change), this new temperature ranges from roughly 1.5°C to 5°C.
The equilibrium response to a doubling of CO2 (including the effects of feedbacks) is commonly referred to as the climate sensitivity.
Two Important Points
1. Equilibration takes time.
2. The feedbacks are responses to temperature – not to CO2 increases per se.
The time it takes depends primarily on the climate sensitivity, and the rapidity with which heat is transported down into the ocean. Both higher sensitivity and more rapid mixing lead to longer times. For the models referred to by the IPCC, this time is on the order of decades.
This all leads to a crucial observational test of feedbacks!
The Test: Preliminaries
Note that, in addition to any long term trends that may be present, temperature fluctuates on shorter time scales ranging from years to decades.
Such fluctuations are associated with the internal dynamics of the ocean- atmosphere system. Examples include the El Nino – Southern Oscillation, the Pacific Decadal Oscillation, etc.
These fluctuations must excite the feedback mechanisms that we have just described.
The Test
1. Run the models with the observed sea surface temperatures as boundary conditions.
2. Use the models to calculate the heat radiation emitted to space.
3. Use satellites to measure the heat radiation actually emitted by the earth.
When temperature fluctuations lead to warmer temperatures, emitted heat radiation should increase, but positive feedbacks should inhibit these emissions by virtue of the enhanced ‘blanketing.’ Given the model climate sensitivities, this ‘blanketing’ should typically reduce the emissions by a factor of about 2 or 3 from what one would see in the absence of feedbacks. If the satellite data confirms the calculated emissions, then this would constitute solid evidence that the model feedbacks are correct.
The Results of an Inadvertent Test
Above graph:
Comparison of the observed broadband LW and SW flux anomalies for the tropics with climate model simulations using observed SST records. The models are not given volcanic aerosols, so the should not expected to show the Mt. Pinatubo eruption effects in mid-1991 through mid-1993. The dashed line shows the mean of all five models, and the gray band shows the total rnage of model anomalies (maximum to minimum).
It is the topmost panel for long wave (LW) emission that we want.
Let us examine the top figure a bit more closely.
From 1985 until 1989 the models and observations are more or less the same – they have, in fact, been tuned to be so. However, with the warming after 1989, the observations characteristically exceed 7 times the model values. Recall that if the observations were only 2-3 times what the models produce, it would correspond to no feedback. What we see is much more than this – implying strong negative feedback. Note that the ups and downs of both the observations and the model (forced by observed sea surface temperature) follow the ups and downs of temperature (not shown).
Note that these results were sufficiently surprising that they were confirmed by at least 4 other groups:
Chen, J., B.E. Carlson, and A.D. Del Genio, 2002: Evidence for strengthening of the tropical general circulation in the 1990s. Science, 295, 838-841.
Cess, R.D. and P.M. Udelhofen, 2003: Climate change during 1985–1999: Cloud interactions determined from satellite measurements. Geophys. Res. Ltrs., 30, No. 1, 1019, doi:10.1029/2002GL016128.
Hatzidimitriou, D., I. Vardavas, K. G. Pavlakis, N. Hatzianastassiou, C. Matsoukas, and E. Drakakis (2004) On the decadal increase in the tropical mean outgoing longwave radiation for the period 1984–2000. Atmos. Chem. Phys., 4, 1419–1425.
Clement, A.C. and B. Soden (2005) The sensitivity of the tropical-mean radiation budget. J. Clim., 18, 3189-3203.
The preceding authors did not dwell on the profound implications of these results – they had not intended a test of model feedbacks! Rather, they mostly emphasized that the differences had to arise from cloud behavior (a well acknowledged weakness of current models). However, as noted by Chou and Lindzen (2005, Comments on “Examination of the Decadal Tropical Mean ERBS Nonscanner Radiation Data for the Iris Hypothesis”, J. Climate, 18, 2123-2127), the results imply a strong negative feedback regardless of what one attributes this to.
The Bottom Line
The earth’s climate (in contrast to the climate in current climate GCMs) is dominated by a strong net negative feedback. Climate sensitivity is on the order of 0.3°C, and such warming as may arise from increasing greenhouse gases will be indistinguishable from the fluctuations in climate that occur naturally from processes internal to the climate system itself.
An aside on Feedbacks
Here is an easily appreciated example of positive and negative feedback. In your car, the gas and brake pedals act as negative feedbacks to reduce speed when you are going too fast and increase it when you are going too slow. If someone were to reverse the position of the pedals without informing you, then they would act as positive feedbacks: increasing your speed when you are going too fast, and slowing you down when you are going too slow.
Alarming climate predictions depend critically on the fact that models have large positive feedbacks. The crucial question is whether nature actually behaves this way? The answer, as we have just seen, is unambiguously no.
UPDATE: There are some suggestions (in comments) that the graph has issues of orbital decay affecting the nonscanner instrument’s field of view. I’ve sent a request off to Dr. Lindzen for clarification. – Anthony
UPDATE2: While I have not yet heard from Dr. Lindzen (it has only been 3 hours as of this writing) commenter “wmanny” found this below, apparently written by Lindzen to address the issue:
“Recently, Wong et al (Wong, Wielicki et al, 2006, Reexamination of the Observed Decadal Variability of the Earth Radiation Budget Using Altitude-Corrected ERBE/ERBS Nonscanner WFOV Data, J. Clim., 19, 4028-4040) have reassessed their data to reduce the magnitude of the anomaly, but the remaining anomaly still represents a substantial negative feedback, and there is reason to question the new adjustments.”
I found the text above to match “wmanny’s” comment in a presentation given by Lindzen to Colgate University on 7/11/2008 which you can see here as a PDF:
http://portaldata.colgate.edu/imagegallerywww/3503/ImageGallery/LindzenLectureBeyondModels.pdf
– Anthony
UPDATE3: I received this email today (4/10) from Dr. Lindzen. My sincere thanks for his response.
Dear Anthony,
The paper was sent out for comments, and the comments (even those from “realclimate”) are appreciated. In fact, the reduction of the difference in OLR between the 80’s and 90’s due to orbital decay seems to me to be largely correct. However, the reduction in Wong, Wielicki et al (2006) of the difference in the spikes of OLR between observations and models cannot be attributed to orbital decay, and seem to me to be questionable. Nevertheless, the differences that remain still imply negative feedbacks. We are proceeding to redo the analysis of satellite data in order to better understand what went into these analyses. The matter of net differences between the 80’s and 90’s is an interesting question. Given enough time, the radiative balance is reestablished and the anomalies can be wiped out. The time it takes for this to happen depends on climate sensitivity with adjustments occurring more rapidly when sensitivity is less. However, for the spikes, the time scales are short enough to preclude adjustment except for very low sensitivity.
That said, it has become standard in climate science that data in contradiction to alarmism is inevitably ‘corrected’ to bring it closer to alarming models. None of us would argue that this data is perfect, and the corrections are often plausible. What is implausible is that the ‘corrections’ should always bring the data closer to models.
Best wishes,
Dick
Sponsored IT training links:
Best quality 70-448 prep material is available for download. Pass the real exam using JN0-350 guide and E20-361 lab tutorial.
“”” Kevin (11:47:43) :
I’m having a problem with this:
“The wavelength of the heat radiation corresponds to the temperature of the earth’s atmosphere at the level from which the radiation is emitted (ca 255oK).”
Is he saying that the Earth emits radiation at 255K? Because that’s just not true. I’m sure I’m just misunderstanding something, and would appreciate someone ’splaining it to me.
Thanks! “””
Kevin, I couldn’t directly see a response to your query so let me take a whack at it.
If you read Prof Lindzen’s essay carefully, I think you will see when he talks about the earth’s “heat radiation” (oedantically incorrect terminology) he mentions the level at which that radiation is effectively emitted.
Now 255K (again a slight technical blip), that of course is about -18 deg C, which is certainly not the average surface temperature of the earth. Actually, the earth surface can be radiating at temperatures between about -90 deg C (Vostok Station) all the way up to about +60 deg C and probably higher in the tropical deserts, and it could be doing that all at the saem time, since NH midsummer, is Antarctic winter night. But takign the mean surface temperature as being +15 deg C (this is a nonsense concept and there is no way we can measure that presently), then that would be 288K, and a black body radiation for that temperature would have a spectral peak wavelength of about 10.1 microns (From Wien’s Displacement Law).
So let’s say a large fraction of the surface happens to be at about 288 K; 15 C. A lot of that radiation gets captured by GHGs such as water vapor and CO2, warming (HEATing) the atmosphere, which ultimately re-radiates long wave radiation, some of which comes back to the surface and some goes upwards towards space. It keeps getting absorbed and re-emitted, and of course the atmosphere gets coolre (for a while) as you go up. What Lindzen is saying; and I have no way of refuting him, is that, in the end analysis, the radiation that does escape as a net loss to the planet, appears to have been emitted by a source that is about 255K, so he is talking about some higher stratum than ground level as being the effective source of the lost radiation.
That’s a great simplification of what is a very complex iterative process, and the way Prof Lindzen has put it, is not a bad simplification. The full modelling of the “exact” physics, I don’t think has ever been accomplished but a lot of people have tried. It’s reasonable to consider the earth as radiating at some effective temperature, but that temperature evidently does not happen to be the so-called mean global surface temperature.
So -18C for a high altitude effective final source of earth’s thermal radiation is not too bad a picture.
I hope that answers your query.
George
JamesG (12:40:32) :
John Philip/Phil/Chris
Yes he does seem to have used an uncorrected dataset without mentioning it, which is disappointing. However we’ve all seen this “correction” story before and it always smells of confirmation bias (no, not a conspiracy Chris). That the authors of the paper are highly pro-agw means they’d be more than happy to homogenize the data just like Willis’s heat content data. But if the other 4 papers confirm the original data then that would excuse the use of the uncorrected data by Lindzen, especially given the purely speculative nature of Trenberth’s original comment and the as yet unknown error calibration technique. I guess we’d have to check those other papers, or hope that Prof. Lindzen clears it up.
The error and their correction is described here:
http://asd-www.larc.nasa.gov/~tak/wong/f20m.pdf
The UK has large areas that are 1metre or less above sea level. If the temp rises a few degrees and sea levels rise then much of this land will be underwater or subject to erosion. Much fertile land will be lost. I’m sure this applies in many areas of the world.
A couple of degrees of global rise may not seem much but it is the heatwaves that will kill and disrupt.
A real world example in UK (from the docment I referenced above
The hot summer of 1995
Cereal yields and quality were good. The protein content of grain rose, which was good for bread-making wheat but not for malting barley. The harvest was large and early, allowing farmers to make a cost saving with fewer working days and less grain drying,…
Cattle enterprises suffered because of a shortage of grass and forage, although upland forage for beef was less restricted. Many areas could only take one silage cut, and grass production in the driest areas dropped by 20-30%. Maize yields were down by 30%, and other feeds such as potatoes were in short supply. All this resulted in food supply problems over the winter, with farmers buying in feed or selling off animals.
One beneficial effect of the summer was that, although hay and forage prices were high, cereals had produced a lot of good quality straw, which some farmers used to supplement rations. Dairy cattle had reduced milk yields. Conception rates in dairy and beef were 5% lower during the summer. It is
difficult to know whether cattle suffered from heat stress or whether decreased production was due to fodder, drinking water and disease problems.
Sheep also suffered from forage supply and heat problems, and supplementary grass feeding was often necessary. Increased numbers were slaughtered as the summer wore on, reducing the market price. Pigs and
poultry both responded to the heat by reducing feed intake. This resulted in lower slaughter weights for pigs and reduced egglaying, low broiler growth rates and increased mortality in poultry.
AND
Posted 9/25/2003 10:56 AM Updated 9/25/2003 9:23 PM
PARIS (AP) — The death toll in France from August’s blistering heat wave has reached nearly 15,000, according to a government-commissioned report released Thursday,
If these “heat waves” become more frequent and just 2degC hotter do you think that Europe will cope and this is a temperate climate?
Now go to equatorial countries where agriculture is already heat stressed what will happen there?
Some places will be better off others will go under leading to forced migration polewards. Politics will not allow this.
Bill
George E Smith says:
A^4 + B^4 <= (A + B)^4
bill,
Don’t worry about it, the world is getting safer every year: click
“Barry Kearns (08:50:54) :
Regarding positive feedbacks:
It is entirely possible to have a system with positive feedbacks, and not have that system go into a “runaway” mode. Such systems are “stable”, but responses to inputs are amplified. In some systems, this is a highly desirable characteristic.”
No arguments with this, per se. When I referred to a “rail”, this simply meant the point at which the system lacks the energy to further supply the positive feedback, which could occur even with a feedback between zero and 1. The federal reserve ratio operates as the type of feedback that you present, for example. A bank is required to keep a certain percentage of its deposits and loan the rest – say keep 10% loan 90%. The 90% loaned gets deposited in a bank somewhere, 90% of that gets re-loaned, etc. This means that a given injection of cash by the federal reserve into the banking system theoretically creates a geometric progression of subsequent injections that gradually reduces to zero, so the sum of the series is a finite value, but much greater than the original injection.
The assumptions here, however, is not only that the feedback to an input signal is less than 1, but that there is an unconstrained source for the feedback – by unconstrained , I don’t mean infinite, but just that the feedback is not at a point where it is otherwise limited. In the federal reserve example cited above, you get the desired feedback so long as there is enough demand for loans at the interest rates available. If the fed were to inject so much cash into the system, that with all previous multipliers no one needed to borrow money anymore because the economy is so awash in it that there is no desire to take out a loan and pay interest, then the feedback isn’t as effective as it was in the past, even though the multiplier is less than 1. The interest rate thus constrains the supply for the feedback – the more money you inject, the lower interest rates need to be to get the same effect. Eventually, however, even lowering interest rates won’t generate enough demand for loans to sustain the positive feedback, and there is no more source for the positive feedback to reinforce cash injections.
There is, however, at that point a very large source for positive feedback when taking money out of the economy. When the fed takes cash out, interest rates go up, meaning that people in a cash-flooded economy want to borrow even less under the higher interest rate, and in fact have a further incentive to buy government bonds because they are now more profitable, further sucking money out of the economy, etc. But eventually as interest rates rise rise and the cash supply drops, the higher interest rates get less and less effective at selling bonds and discouraging loans because the economy needs a certain amount of money just to function.
In the same sense that positive feedback loops should get less effective as you swing to an extreme, negative feedbacks should get more effective. Wind resistance increases with the square of velocity, for example, such that when you increase horsepower to your car by amount X when driving at 10mph the resulting increase in drag isn’t nearly as much as that same HP increase X will produce if you were driving 100 mph. This is why gas mileage drops so rapidly after about 45 – 55 mph.
These same rules should apply to the climate system. Even if the positive feedbacks inherent in the system have multipliers less than 1, the physical phenomenon that provide the feedbacks will have their own constraints that limit the net amount of feedback they are capable of providing, and will also limit the rate at which the feedback is even capable of occuring. It seems more than reasonable to propose that, late in an interglacial period, positive feedbacks are insignificant and have more capability to reinforce a sustained cooling input than a sustained warming input, while negative feedbacks are likely to be significant. While in the transition between an ice age and an interglacial, positive feedbacks are likely to be the dominant cause of temperature changes, changes at either of the two extremes is likley to be dominated by the inputs rather than the feedbacks.
bill-they’ll deal like they are already dealing-they’ll adapt. The stupid people hypothesis is bunk:
http://www.worldclimatereport.com/index.php/2008/02/14/few-french-fried-in-2006/
[snip] you don’t seem to get 1. it is the lower temperatures which warm fastest and 2. the tropics hardly change temperature at all. Food can be ~traded~-if agriculture in the tropics really is threatened, the solution is FREE TRADE-something I doubt you want…
Apologies-this “bill” person got to me. I’m taking a cooling off period.
kurt (17:48:34) :
What you state is very reasonable, except that if what you described occurred, there would be a gradual transition (a gradual reduction of positive feedback, and/or increase or negative feedback). This would show up as a decrease in the rate of change of temperature coming out of an ice-age. However, looking at the little and local data we have (and therefore ‘possibly’ not representative) this is not what happens. From the ice core data it appears that the rate of change of temperature does slowly decrease, but remains pretty much constant until the very end. Possibly ice-core data is not representative … or possibly your theory is incorrect and positive feedback ‘stays on’ and until we hit the limit of the system (saturation).
Mark T (11:02:58) :
Phil. (10:02:45) :
Here’s an example that is close to the ghg effect.
It looks like you actually learned something from the last time you brought this example up (steady state references and that it takes time to get there). Kudos. Just as a check, you do realize that the feedback in your examples are both less than unity, correct (0.5 and 0.6 respectively)?
I don’t recall saying anything about the time, that would just depend on the thermal mass of the absorber. I chose the feedback value, 0.5, deliberately since that is the ~value needed for the GH effect for the earth.
That should be “from the ice core data it appears that the rate of change of temperature does NOT slowly decrease,”
I vaguely recall something from high school science books some 30 odd years ago, that there’s no such thing as “cold” only a lack of heat.
What’s that got to do with anything? That “land” has been underwater many many times in the past, and will be in the future, regardless of what humans do. Do you propose we stop Plate Tectonics because it’s slowly killing the planet?
Jeff Alberts-(I know I said I was cooling off, but its hard to resist) You are right about that thing about cold. As for bill’s comment, what on Earth is the concern over a meter of sea level rise about? Best projections, it will be measurable in tens of centimeters at most. It will take centuries to inundate jolly old England. I was concerned about the issue to-the vast majority of the state of Florida is pretty much at sea level. But there is no basis for concern over it. I’ll zip away now.
Oh, wait one moment. Some supporting calcs:
http://www.appinsys.com/GlobalWarming/GW_4CE_SeaLevel_files/image002.jpg
at 16 milimeters per decade, that’s 160 millimeters per century, or 16 centimeters, or 1.6 decameters, or .16 meters. That’s a little over six inches BTW.
Doesn’t it strike you (or any of the other incurious lot here this evening) as odd that this is/is near the middle of the atmosphereic window (8 – 14 um for the readers out there) here on earth given the transmittance properties of CO2 and H2O in that range?
Could I not then liken Planck’s spectral curve leading to ‘spillover from a bucket’ as more LWIR energy transmits through this window (DIRECTLY into space I might add, unblocked by CO2 or H2O) as temperatures approach and exceed about 216K ( 0 deg F)? *
Did I mention that total radiative energy under the curve (for purists) – seems to be proportional to: Temperature_to_the_4th_power )MEANING that energy under the curge and the surge in the curve REALLY FAST)?
Anyone else notice this particular coincidence? Anyone?
Bueller?
.
.
.
.
BEST interactive demo I have seen of Planck’s curve and temperature (vary the temp and watch Planck’s curve surge and grow!): http://profhorn.meteor.wisc.edu/wxwise/AckermanKnox/chap2/planck_curve.html
.
.
.
.
(*I owe credit for this thought to poster “Andrew” who has hinted at this a few times.)
.
“Jeff Alberts (18:52:22) :
I vaguely recall something from high school science books some 30 odd years ago, that there’s no such thing as “cold” only a lack of heat.”
So funny, Jeff… when I tried to explain that to one of my older brothers many years ago, he said, “That is so stupid Mike, heat is also only a lack of cold.” No amount of explanation could make him waver even a little…
Mike
I’m assuming people are aware that there has been no sea level rise in the last five years in the Atlantic, Mediterranean, or Pacific oceans. The only ocean basin with any sea level rise is the Indian Ocean which has started declining now after being loaded up by successive El Ninos in previous years.
This data correlates fully with the stable Ocean Heat Content, the lack of any temperature rise and the end of “warming in the pipeline” over the last five to ten years as well.
Watch the surface temp measurements over the next few months. GISS, NCDC and the Hadley Centre have been struggling in an effort to produce stable let alone increasing temperature data over the last year. The satellite temps have been affected by the record Sudden Stratospheric Warming event and, consequently, won’t show declining temps until April, but the surface temp measurers/adjusters will have a hard time keeping the trend from declining another 0.1C or 0.2C over the next few months.
We’re not that far off of Zero anomaly right now and they can’t afford to have that show up in the surface temp numbers.
Since the principal criticism of Prof. Lindzen’s little essay is that he used data before it was subsequently corrected for technical reasons, which corrections effectly destroy his thesis,
and since the response seems to be the suggestion that this correction was an underhanded way of better fitting the data to the AGW models,
then perhaps it would be helpful to have a post from someone relatively impartial (if such a person with the relevant expertise exists) to assess these two competing claims—ideally separate from this now enormous thread, which is mostly devoted to the nature of feedback.
/Mr Lynn
Oops—that should be “data before they were subsequently corrected. . .
/Mr L
For those that are interested, here are some graphs and examination of ice-age temperature changes and why this strongly suggests positive feedback driving the system to saturation with no significant negative feedback in sight. At least that is what it suggest to me 😉
http://www.ianschumacher.com/iceages.html
Phil. (10:02:45) :
Illuminate an grey surface with visible radiation and the temperature will reach an elevated steady state value.
Place a dichroic mirror (which transmits vis and reflects 50% of the IR) this will feedback IR to the surface and heat it up, the system will reach equilibrium when the IR passing through the mirror equals the input.
Reminds me of J.Peden’s green house oven
http://www.vermonttiger.com/content/2008/07/nasa-free-energ.html
You realize that the mirror will stop the incoming that is setting up the steady state of the grey surface by x%? So the surface will lose x% of incoming and get back 50% of less reflected IR,less than it had without the mirror, the temperature will drop, no steady state on previous temperature.
Therefore Input=100, IR from surface=200, IR from mirror to surface=100, IR though mirror=100.
That is positive feedback and stable.
Increase the feedback by replacing the dichroic with one which reflects 60% of the IR and the feedback increases to 150 so a new higher ss temperature will be reached, you’d only get runaway increase if the mirror reflected 100% of the IR.
As explained at /chriscolose.wordpress.com
“Lindzen’s analysis is based on outdated data that has been revised since 2002, and these revisions are not exactly recent, so he should have been aware of them. Using the more recent data would not allow him to make his argument as presented as WUWT.”
This was documented by Wong et al 2006, Journal of Climate which inexplicably Dr Lindzen did not refer to.
Bill Illis – You are quite quite simply flat out wrong about sea level rise.
Whereas sea level is not rising uniformly, it is none the less rising. The current global average trend is 3.3mm/year. The observed rise is just higher than the IPCC predictions. (The IPCC sea level models apparently did not include ice melt (because it isn’t understood well enough yet), and observed rises in CO2 concentrations are currently tracking higher then their worse case scenario).
See Australian Commonwealth Scientific and Industrial Research Organisation (CSIRO) sea level rise pages for data, plots and maps. These are presented for the last few decades, for the last two centuries and for the last 140,000 years.
Download the data, map, plot and analyse it for yourself. Enjoy!