The Physics of Climate Modelling

This is just a pointer to a ‘Quick Study’ guide on The physics of climate modelling that appears in Physics Today this month, and to welcome anyone following through from that magazine. Feel free to post comments or questions about the article here and I’ll try and answer as many as I can.

132 comments on this post.
  1. L. David Cooke:

    RE: #48

    Hey Hank;

    Actually, I have a number of links that will clearly demonstrate that any accuracy resolution greater then 5 W/m^2 in a clear sky night time measurement is not currently available regardless of your 1st links suggestion of 2 watts.

    The tools available, can be up to 50W/m^2 in error from calculated, depending on the environmental conditions. The recommended improvements suggested to reduce the level of error range from, Lidar, to ensure the various water vapor and aerosol participation, to different detectors altogether. I could share these though you can find them yourself at the site if you will search for “down welling clear sky night”.

    The point is even with the latest devices such as the SPECTRA AERI tools there remains a question. Either the frequency band or the temperature range of the equipment is exceeded by the real world values to be measured.

    For example even the IRT versus the PIR discussion that have been deployed as recently as Oct 2006 are both subject ground IR skew. Though the IRT is more accurate at night, the bottom of the range being in the 220 K range when the actual down welling values can be 180 K or less. When you include the participation of water vapor and CO2 being in the 900nm to 2.3um range is missed by the IRT as it is focused in the 9-11um range.

    This only goes to repeat the earlier concern that we have to obtain a set of reliable, discrete frequency band detectors designed with the necessary operating temperature/conditions range. A major concern is that it is unlikely that we will achieve the sensitivity necessary with a well shielded detector head or a constant temperature detector oven in the ambient temperature range that measures differentials between the calibration source and the down welling night time sky.

    To address your post, the presentation by Dr. Long (found in the above search as well) is much better as an introductory into atmospheric radiative measurement data. The follow ups that have occurred in relation to the equipment and data qualities also found in the above search leave major opportunities in relation to future improvements.

    (Note: Putting ideas into simple language, such as “CO2 vibrations”, are intended to help with communications (one of my personal challenges I am attempting to overcome) and are not intended to assist Google searches.)

    Dave Cooke

  2. Hank Roberts:

    I’m afraid this is just word salad to me, and apparently also to Google:
    Your search – “down welling clear sky night” – did not match any documents.
    * Make sure all words are spelled correctly.
    * Try different keywords.
    * Try more general keywords.

  3. Hank Roberts:

    Okay, here’s something — when I drop the quotes from your example and limit to recent Google Scholar hits, some papers turn up. Perhaps if you focus on one article, the authors might care to reply. Some of these are familiar names.

    I guess you’re saying that this is an area not well understood (agreed).

  4. John Dodds:

    OK people we are NOT all talking about the same thing or even from the same educational starting point. (responses to #38) I am going to explain where I am coming from & try to answer the comments made above as I go thru.

    According to a Comment by Eli Rabett – 13 Jul 2005 @ 11:32 am (read the whole thing to get more info! Eli, I truly thank you for taking the time for that piece of education 18 months ago & as I just read, for the update in #50- JD)

    “Briefly put, the process can be defined as a CO2 molecule absorbing a ~650 cm-1 photon (equivalent to a thermal energy of about 900 K), and losing that energy to the surrounding bath of atmospheric gases….Because collisional energy transfer to and from the excited molecules is rapid, the chunk of energy (650 cm-1) rapidly degrades into the heat bath of the atmosphere.” (Hank, this says that the laser excitation/deexcitation is only a trivial part of the energy transport process- the major part is absorption, collision to other air molecules, full spectrum at the specified temp radiation from the warmer air, reabsorption etc etc – IF the only process is absorption and radiation to another CO2 … then there is no way for the other air molecules in the bath to heat up.- you would end up with 900 K CO2 and 287K air- which is not possible for very long )

    BUT this does not account for warming because for every absorption (or capture) there is an equivalent loss (or release) – conservation of energy. SO where does the energy for the warming come from? Answer- when you add extra CO2 & increase the number of absorptions and if each absorption/release/collision takes a few to ten microseconds (per Eli), then the energy being transported out to space takes LONGER due to added absorptions which translates to more energy in the air at the lower levels, BUT less energy being transported out to space over the same time interval in order to conserve energy. ie we get the energy dis-equilibrium with height or imbalance (TOA is cooler than at pre CO2 equilibrium with the energy-in) that is pictorially shown in Hansen et al 2005 fig 2e – ie ground warming, TOA cooling by an equivalent total amount of energy. ie The GHE does cause warming!! (now this raises another contradiction- Hansen shows the equilibrium point (temp where energy was formerly released to space) has moved DOWN, whereas James (#46) says it goes higher & Dave Cooke (#47) asks for evidence. – well Dave perhaps the evidence is in my proposal that the GHE exists but the higher temperature from the absorption and delay results in a larger driving force (SBL feedback- hotter air rises faster, or hotter objects radiate faster) to bring the earth system back to equilibrium just as fast as it warms up – in which case there would be NO evidence.- other than differences caused by changes in solar energy input (sunspot cycles) and any other source of variable input energy. Given the variability of sunspots/solar insolation I personally doubt/guess that you could measure the differences well enough- too much noise as Eli said.)

    Which brings me back to “English’s ” paraphrasing/understanding (#41). NO your paraphrase of what I said is NOT what I meant. First on your reference (at the pco-bcp site) & challenge philosophy – YES we are adding CO2, yes adding CO2 causes more warming, BUT I do not agree that there is consensus that this results in significant global climate changes. Only the GCMs say that this is the cause. I personally think there is an error or two in the GCM calculations (not counting the uncertainties in parameters) and I am challenging this rather than trying to challenge the validity of the various parameters etc- everyone else is already doing this & getting nowhere. I am trying to point out that the results generated by the GCMs are internally inconsistent or wrong or do not agree with observed reality, & I am trying to point out what I think is the error. As I have learned, painfully, I have discovered that most of the Scientists doing this research are VERY VERY good- if they put it into the GCMs then it is probably valid, maybe with some uncertainties but valid in direction and pretty close to the correct magnitude, so I look for what they missed.

    In one sentence, the proposal I am trying to make is that the GHE causes warming at the ground level and cooling at the TOA level, per Arrhenius and Hansen 2005, BUT by this very act this creates a temperature differential dependant driving force calculated by the SBL that results in a feedback that returns the atmosphere at all heights to the former equilibrium to energy-in conditions. ie When the GHE causes warming, (& cooling at the TOA- – adding GHGs can NOT create energy it has to come from somewhere – ie up near the TOA,) then these new temperature differences from equilibrium (even when it is only the warming caused by a single absorption) create the driving force that return the atmosphere to equilibrium conditions. This (SBL Feedback for lack of another term- I wouldn’t dare call it the John Dodds effect, you get too much grief- besides Stefan identified it first :) ) is a self adjusting NEGATIVE feedback that will automatically exactly negate the GHE & WV positive feedback effects while adjusting to changes in energy-in caused by solar, albedo, volcanoes etc. The GCM says that the warming creates a permanent (or at least multi-year until CO2 disappears) energy imbalance at the TOA that will “permanently” add (solar) energy to the Earth. SO why wouldn’t adding this solar energy just eliminate the GHE caused energy imbalance at the TOA? This is the reality every morning as we warm up. Besides if the energy imbalance is permanent with higher CO2 (ie global warming) then HOW do we get to equilibrium conditions twice a day? (Visualize the annual average equilibrium temp as being 288K./15C At night we cool below this – since the earth is still hotter than the energy-in which is zero at night (CO2 absorption operates 24 hrs/day) . Then during the morning/day the solar energy-in is more than energy out so we warm up to try to keep up with the steadily increasing equilibrium – BUT we HAD to pass thru the equilibrium point- BUT the GCM says that due to GHE warming we are always warmer (or cooler @TOA) & hence have a multi-year energy imbalance (ie the GCM conditions are contradicted by reality) During the daily cycles whenever there is an imbalance (be it caused by GHE or the sun) the driving force to return to equilibrium is the amount of energy transfer calculated by the SBL. The SBL canNOT differentiate between GHE warming & solar warming- just the temperature & Eli’s transfer mechanism/collisions guarantees that all the air adopts the same temperature, subject to weather variations). The return to equilibrium to energy-in is a fact observed twice a day. It is very fast- we can go from the daily high temp past equilibrium to the daily low temp in a few or 12 hours at most, but we are limited to the amount of energy transported out as calculated by the SBL (So, James #46- an object in space will NOT instantaneous cool – it will radiate the SBL calculated energy over time (SBL units are Joules/(sec-m^2) & keep cooling until it is at the equilibrium temp- because IF it keeps radiating past equilibrium with energy-in, then the SBL says it will radiate less than what is coming in & it will warm back up to equilibrium. The SBL strength and direction varies depending on the temperature differential from equilibrium )

    The SBL Feedback is a self regulating equilibrium enforcer, that happens to operate much much faster than the CO2 warming does ( CO2 is a few degrees over a century). It forces equilibrium to energy-in twice a day regardless of if we are warmer or cooler. I think the GCMs ignore/forgot the radiation part of the SBL Feedback – Back in “Tom Fiddaman â�� 3 Nov 2006 @ 1:36 pm “, Tom identified that convection and conduction are small compared to radiation as a transport mechanism (but, James (in 46) – conduction in the air exists- think thunder & lightning! & all three combine to get energy to the TOA where it is radiated to space) AND Gavin agreed that there is a convection feedback due to the higher GHE caused temperature (- ie Hotter air rises faster & it is included in the GCMs as a lapse rate change within the water vapor feedback – see 12:57 pm ) SO FAR I have not seen anyone address the radiation part of the SBL feedback- hotter objects radiate faster??? BUT it has to work since that is what causes the loss of energy at night. AND it returns to the equilibrium to energy-in (which is constantly moving), not to several degrees above (ground) or below (TOA) per the GCM. Also if hotter air @ ground radiates more and faster, and cooler air @TOA radiates less and slower (both up & down) – then won’t they meet somewhere in between which is equilibrium before the slower GHE started? Since the rate of energy transfer is dependant upon the temperature differential (imbalance) then the SBL Feedback will self adjust to even out the balance of both solar and GHE over a day, and as I visualize it- it will adjust for the GHE as soon as the first little delta T is created. ie Arrhenius’s GHE exists but is neutralized by the SBL feedback virtually instantaneously. Gavin said that if we have equilibrium then we (effectively) do not have a GHE. & I agree- BUT I now think we still have a GHE (Arrhenius was right) it is just that the SBL Feedback cancels it within seconds. Which leads to Julian’s (39) no warming but we can add CO2 – at least until the CO2 ruins the ocean pH or the burning of CO2 depletes the O2 in the air so we can’t breathe (ie CH4 (smallest hydrocarbon) + 2O2 (from air)= CO2(to air)+ 2H2O.- which actually yields a trivially lower air density and trivial global cooling due to adding CO2 per the ideal gas law which applies when there is no net energy transport into or out of the system – Sorry Gavin, – but I can’t resist quoting your reasons/responses back to you- I apologize.)

    OK I hope this explains what I’m trying to say. More comment/questions please.

  5. L. David Cooke:

    RE: #50

    Hey Eli;

    Thank you for the general physical process you shared. As you indicated that the photon release is generally non-directional and will be released in 360 degrees, though the input is primarily the reflected or radiated ground radiation/up welling at between 15 and 20 um. I don’t know about your 6% activly emitting energy, though. I suspect any temperature above the solid state temperature of CO2 would suggest most all of the molecules would be active emitters. To some degree even after the solid form of CO2 is achieved some percentage on the boundaries would be significantly active, as well.

    I guess my major concern of radiative measure is that there has been significant work recently on ECHAM5 to account for the short wave physics and it seems we really do not have a good understanding of the radiative physics yet. It does not mean that what you describe is not valid; however, it would appear to occur more often at the surface altitudes then higher in the air/grid column. This is part of where the adiabatic processes and radiant energy physics or convective along with radiant processes interact and confuse the mechanics.

    I have been spending time on the UKweatherworld site and in association with a PeterH there, I have come to see that the convective processes actually have greater influence in heat transport as opposed to radiant transport. If this is accurate I suspect that the radiant influence of CO2 is much smaller then I have seen described in most studies. If this is accurate it may require a review of the model constructs as it may be that the radiative interaction is over rated. Hence, my desire to determine the direct baseline value of real world, night time, clear sky radiant energy that can be associated with GHG.

    (Note: As I related above though professional language is more precise and descriptive, it reminds me of taking newbies out in my sailboat. If I have to Jibe or Tack, I am supposed to use the term “Jib Ho” or “Helms a Lee” to inform everyone that the boom is about to swing across the boat. The first time I took a newbe out he got bumped in the head and said, “Why didn’t you yell Duck!, that I understand.”.)

    Dave Cooke

  6. John Dodds:

    Re 50,55
    Instead of the statement 6% of the CO2 actively emit energy, would it be more accurate to say that 6% actively emit energy in the spectral range that can be absorbed by CO2 during the next absorption?
    This way ALL the air in the air soup packet will actively emit energy just at different wavelengths (or maybe even as more collisions(?), 94% of which will be sent to space directly or absorbed by WV etc & only 6% of it will see the next CO2???
    I think I asked that right. or I may not fully understand Eli’s physics.

  7. James:

    Re #54: This seems so wrong that I’m not sure where to start. Let me begin by correcting a couple of misunderstandings. First, what I said about an object instantaneously cooling was meant as a sarcastic (see the smiley?) response to your statement that “…radiation of energy is as fast as the speed of light.” I really don’t understand what you’re trying to say here. In radiative cooling, the photons of course _leave_ at the speed of light, but that says nothing at all about the _rate_ of cooling, which depends on the number of photons leaving and their energies.

    Second, when you say “…conduction in the air exists- think thunder & lightning…”, I think you’re being confused by language. Neither of those involves heat conduction. While air (and gasses in general) do conduct some heat, the amount is very, very small. That’s why most insulating materials are just ways to hold air in pockets small enough that it doesn’t form convective cells.

    I think your major problem is in thinking that the Stefan-Boltzman law applies to the Earth. It applies to black bodies, and the Earth most definitely is not a black body. If it was, there’d be no greenhouse effect of any sort. Beyond that, about all I can do is repeat the simple first-order explanation: The atmosphere is transparent to visible light, partly opaque to infrared. Energy comes in as visible light, warms the ground, which re-radiates in the infrared. But the infrared can’t get out easily, so that raises the temperature. Eventually it increases enough that a new equilibrum is formed.

    It’s not that different from putting on a sweater when you’re cold. The insulation slows the rate at which heat leaves your body, so you get warmer. Same with the greenhouse effect. Imagine you have two planets at the same temperature, say 280 K. A has CO2, the B doesn’t. Now take away the Sun. Both planets will cool, but B will cool faster. After 12 hours, it might be at 275 K, while A is at 276 K. Bring back the sun, both planets get the same amount of energy input (say enough to raise their temperature 5 K), and what happens? B is back at 280 K, but A is now at 281 K.

    See? Global warming made simple :-) Beyond that, I think I have to let the professionals try to explain, if they like.

  8. L. David Cooke:

    RE: #52 and #56

    Hey All;

    First statement goes to Hank, good sir if you want to see the data that I have been referencing for the past 7 years feel free to go to and perform the search as I had suggested. Google is likely not going to be of any assistance in trying to collect information that is published specifically on a site. Sorry, if you want the data you have to go to the well, there is no way you are going to prime the pump from the faucet.

    Secondly, Mr. Dodds a great treatise; however, a photon is not focused to be received by any specific molecule. If a molecule releases the energy it will be as a point source and as a wave front it will be detectable providing there is not a shield or reflector between the detector and the emitter. Put another way, energy emissions that are generated in 3D from a point source are going to be detectable unless the density of the material between the source and the detector absorbs the total emission and subsequent emissions are directed away from the detector. It appears you are confusing radiant photon transmission (at the speed of light) and convective transmission (around the speed of conduction) with the transportation of the higher thermal content moving away from the detector.

    (Keep in mind that the expansion of fluid or the expansion differences of clad dissimilar metals in most thermometers are based on the interception of part of the localized energy that is radiated. In short, temperature is the food coloring and the tank or bowl of fluid is the medium. Your detection of the temperature is going to relate to detecting a difference of the physical properties in at least two materials. The accuracy of the measurements are going to be related to how those materials react to the same stimuli frequency and energy band of the sampled physical phenomena.)

    Dave Cooke

  9. Hank Roberts:
    Search there uses Google; 7 hits for that search string:

  10. Hank Roberts:

    >48, 51, Dave Cooke wrote: “I have a number of links that will clearly demonstrate that any accuracy resolution greater then 5 W/m^2 in a clear sky night time measurement is not currently available regardless of your 1st links suggestion of 2 watts….”

    But using your search at ARM, I find this:

    “Clear-Sky Longwave Radiation: Clear-sky longwave radiative transfer appears to be largely a solved problem…. the uncertainty in the ccalculated longwave flux at the surface is better than 2 W m-2
    for the range of measured precipitable water vapor values (Turner et al. 2003b). Because there are no major spectral errors in the flux calculation, we expect that model calculations will yield accurate cooling rate profiles in the spectral interval from 4 to 20 µm.”
    —— end quote—–

    I can’t find what you say you found — at the sites and using the search terms you say should support your info. What I find contradicts what you say you found somewhere in the same set of documents.

    Can you find your source again, and point specifically to it? Knowing what you read, its date, and its cites would help.

  11. Ike Solem:

    Look, all the word salad comments on radiative physics “not being well understood” are just nonsense, not to put too fine a point on it. Take this physical example: the sun has set, and it is a cloudless night. What happens next? The earth’s surface was heated during the day, and is now emitting infrared radiation as it cools off.

    The infrared radiation interacts with the matter (gas molecules) in the atmospheric column. The optical cross-section of absorption for different molecules varies; this means, for example, that O2 and N2 do not interact with infrared radiation, but CO2, CH4 and N2O all do. See the absorption spectrums here (infrared to the right). As a result of that interaction, the energy levels of those ‘greenhouse gases’ are excited, as they fall back down to ground level they emit infrared radiation in all directions. This has all been very well understood for quite some time!

    So, the ground, which would have cooled more rapidly at night, is recieving back-radiation from the sky, and when morning comes, it is warmer then it otherwise would have been. Now the sun comes out… and if you have solar panels on your house, the solar radiation excites the energy levels in the N,P-doped silicon structure (or in some other material) and due to the electronic structure of the semiconductor system, excited electrons can only return to ground state by flowing through an electrical circuit – providing you with a useful power source. Algae in the oceans and plants on land do something very similar with their highly complex photosyntheic apparatus – they store the energy in the primary form of ATP and NADPH, and get their electrons by tearing apart water molecules with the aid of sunlight.

    In any case, energy that would have been released to space is instead stored up in the land, oceans, ice sheets, and atmosphere – and the oceans and atmosphere are convective fluid systems, so the heat is transported here and there, and evaporation rates increase, and that puts more H2O in the atmosphere, and the radiation trapping effect grow stronger – and to quantify all that, and make future predictions, various real time observations are made using satellites,etc, and very complex mathematical models are developed for oceans, ice sheets and the atmosphere.

    Now, what climate models don’t do is attempt to make numerical projections of future CO2 responses; rather they work off given scenarios for CO2 emissions. The CO2 emission rate is dependent on things like human behavior as well as the biosphere responses, but it has been steadily increasing over the past few decades – both the rate and the amount. This factor, along with N2O and CH4 emissions, seems to represent the greatest uncertainty in future predictions – with the IPCC projections being on the conservative side. Its also a factor that humans can control by switching from fossil fuels to renewable energy sources.

    Some people complain that “the models are constantly being tweaked, so how can you rely on them?” but that’s how weather models work – the weather stations constantly send data out, which is incorporated into the models for the latest round of predictions. Old ice sheet models are similarly being replaced by more accurate modern versions which incorporate new observations about dynamic ice sheet responses, as I understand it.

    Endless technical jargon often serves no other purpose than to confuse an issue and exclude ‘outsiders’ from the discussion – which seems to be the intent of some of the above posts.

  12. L. David Cooke:

    RE: #59

    Hey Hank;

    The problem is my word processor, link the down and welling together and you get 159 hits:


  13. L. David Cooke:

    RE: #60

    Hey Hank;

    At the risk of not getting the links to pass the filter, here they are with a few editorial notes:
    (Pay particular attention to page 3 and the recent SW and LW graphs.)
    (This was part of the caliabration resolution work that happened after the noted issue)
    (This is the reason that down welling measurements could be questioned as
    unknown aerosols can offset the predicted values. With an unknown
    reason for night time expected value being as much as > 30 W/m^2 lower
    then projected, raises questions regarding the current equipments
    ability to make these measures. These two issues appear to complicate
    the ability to confirm clear sky nigth radiant measurements.)
    (The issue examined here is not unique, depending on the size or design
    of the detection cavity radiative rate of change can lead to
    inaccuracies when the detector is operating at the edge of the
    temperature or bandwidth capabilities of the detector array.)
    (The recommendation here appears to suggest to not rely on ARM data until
    Lidar are installed at the detection sites. The AERI Homepage referenced
    in this paper is found here: )
    (Even with the latest equipment there appears to remain issues, as it
    appears the frequency band and the temperature range necessary, exceed
    the capability of the equipment.) (See Specifications)

    These were in the long list with only a few links requiring you to register, hopefully these will help. Pay particular attention to the date stamps as the more recent indicate that the issues remain and the planned means to deal with them. Hopefully, I have not missrepresneted anything and Good Luck!

    Dave Cooke

  14. L. David Cooke:

    RE: #61

    Hey Ike;

    The point is your night time radiation of GHG does not appear to be detectable by the current radiative detector systems. If you have a reliable source that demonstrates this over time, on a daily basis and can docuement the clear sky nightly radiative down welling curves it would be very welcome. (Please keep in mind that we need a minimum of hourly discrete full spectrum measures, with a detector that is stable within 3db across the desired badwidth and a range of 175K to 350K over the local 12 hour period related to the night side of the planet.)

    Dave Cooke

  15. Hank Roberts:

    > Please keep in mind that we need a minimum of …

    What’s your basis for that? Who says “we need a minimum of” — again, I can’t find your source on the site you point us to. Is that your requirement, or one from the research?

    “Clear-sky longwave radiative transfer appears to be largely a solved problem…. the uncertainty … is better than 2 W m-2 … (Turner et al. 2003b)…. we expect that model calculations will yield accurate cooling rate profiles in the spectral interval from 4 to 20 µm.”

    Why do you say that’s an impossible level of accuracy to achieve? You still aren’t giving specific sources, just pointing to large papers about many instruments, which of course take work to put together into a monitoring system.

  16. L. David Cooke:

    RE: #65

    Hey Hank;

    Let’s start with your first question. A description of the curve of radiant decay is a graphic that is easy to over lay for comparitive purposes. To be have a high resolution and meet standard deviation or statistical confidence levels the desire would be to collect 30 samples nightly.

    The collection of 1 per hour meets the requirement for describing the mean statistically and because you have solar influences on the upper atmosphere when the ground is in darkness the 10 hours of samples in darkness meet the min. requirement best as long as you are measuring winter clear sky night time values near the Vernal or Autmnal Equinox. So yes, sample rates of 1 per hour is my minimum requirement to meet a minimum level of statistical confidence regarding the description of the mean of a known environment / population. (Note: If you read the papers I have provided you will find that ARM makes measurements at 15 min. where Spectra AERI the sample rate is 30 min. Both meet the min. required sample rate, the ARM sample rate meets the minimum desired.)

    As to your second question, you have to read these papers to see the discussion regarding the error and the calculated versus measured deviations. Virtually every reference that I have annotated as such, there is a description of a question of quality of the data support my conclusion. The specific data contained within these papers will point to the limitations of the detectors in regards to the frequency range, absolute temperature range and measurement conditions.

    Do you have sources that deny the issues raised in these papers, where I have noted specifics? Do you have sources that have sufficient accuracy and sample rate required to define the full spectrum downwelling energy for a clear sky night time measure?

    (Your reliance on a 2003 paper when there is clearly more recent papers that question the “monitoring systems”, causes me concern. Your motives appear to be suspect, do you wish to clarify your point?)

    Dave Cooke

  17. Hank Roberts:

    I’m not arguing with you about this, I’m showing where I got that direct quote following your search. You’re quarreling with their cite, not mine.

  18. L. David Cooke:

    RE: #67

    Hey Hank;

    Funny, you appear to purposefully ignore data that is contrary to your position or has been superceded. On that stand point I guess there is no reason for further conversation. Good Luck to you in your endevors.

    Dave Cooke

  19. Hank Roberts:

    I don’t have a ‘position’ and don’t understand what you’re arguing with — I followed your pointers, read the 2004 summary that said “Clear-sky longwave radiative transfer appears to be largely a solved problem…. the uncertainty in the calculated longwave flux at the surface is better than 2 W m-2” — that’s not the new stuff, that’s the foundation work.

    Do you disagree with the report there? That’s the first part of the program, done on a few sites with the first set of instruments. That was the beginning work.

    They got that — they say — down to a level of accuracy you don’t believe. I can’t say why they claim one thing and you claim another — but I am pointing out that you’re talking about their more recent work, with different instruments, in more locations — and that of course they will be reporting less accuracy for a while in new locations with new instruments.

    I don’t see an argument here. Look, they reported getting that high accuracy level and cited a 2003 paper. That’s not the same instrument that you’re looking at for 2005 or 2006 — it’s the first stage work. They got to that level — then said their next step is rolling out a greater variety of instruments in mobile labs and going to many more sites.

    So — with new equipment in new locations — they’re reporting more variability. That’s what you’d expect from the new mobile labs and new sites — what’s to argue about there? New instruments and new locations are going to give more variable data. Publishing the problems is how progress is made.

    Did you read the reports done tracing the Space Shuttle launch plume, a few weeks ago? (at Head in a Cloud, here: where he talks about “PUMA, a field campaign for making measurements of the chemistry/microphysics of the space shuttle plume from aboard the NASA WB-57′

    That’s not final publication, that’s a scientist blogging his daily work — including the uncertainties about the data collection flight. This is _exciting_ for an amateur to watch. It’s not something that proves the science is wrong, eh?

    Don’t you think these folks are doing something wonderful, and right out in public? We can cheer them on because the facts that emerge are what’s real. We’re the audience, we need to be an inviting one if they’re going to talk to us. Same for the folks whose work you’re pointing us to — if they’re going to want to discuss their work, we need to understand it enough to talk about it. So — one accuracy level in 2003 at the first stage; and later in 2005-6, a variety of different accuracy reports from the second stage. It’s not that either is wrong, it’s a question of what were they doing, where, with what tools. So let’s charm them into talking to us. Why not?

    And that’s really far too much from me. It’s not about me. Back to listening and reading.

  20. Ike Solem:

    So, for those who want to enter the highly technical world of atmospheric radiation calculations, the best place I’ve found to start is with Spencer Weart’s excellent website based on his book, “The Discovery of Global Warming” – and having read a lot of introductory books on global warming, that one seems to be the best – if you want to introduce someone to the topic, buy them that book! The supporting website has this essay on Basic Radiation Calculations in the Atmosphere.

    As far as measuring the downwelling radiative flux (RE#64)…you’d have to measure it over the whole planet, wouldn’t you? Recall that the reason that Charles Keeling set up his CO2 measurement station in the middle of the Pacific was to avoid local fluctuations in CO2 concentration. Then you’d have to worry about ‘separating out’ the CO2 signal from the water vapor signal from other signals (CH4 – is there a gas flare nearby, or a swamp? An old leaky fridge putting out CFC’s? N2O generation? Aerosol content?). Of course, you can tell that it’s colder at night in the desert then by the coast – because there’s little water vapor in desert regions, so the surface cools off much faster. What if a cloud passes overhead? It suddenly gets warmer…but what about the atmospheric column above the cloud? Even if you got simultaneous data from all over the planet at high resolution, how would you know what was due to CO2? You’re not in a lab, you’re out in a field – on the ocean – on a glacier – and so on.

    So, what do you do? The first thing to do is to take the temperature of the atmosphere and monitor it over time – that’s what the radiosonde ballon network attempted to do, followed by the microwave sounding units on satellites, which at first appeared to show that the atmosphere was not warming at the rate predicted by the climate models – the radiative-convective models of the atmosphere, terribly complicated beasties that they are. That whole topic has been put to rest, and the details of the issue are explained with the usual clarity by realclimate: Et Tu, LT?

    Beyond this, it is possible to sort out different radiative signatures using satellites; see The Radiative Signaure of Upper Tropospheric Moistening, Science, Soden et. al 2005 So the claim in #64 (radiative detection systems do not exist) is simply untrue.

    Given a clear theoretical reason why the atmospheric temperature would increase, plus an actual measurement of that temperature increase, it seems rather obvious that the conclusion is solid. You can’t rule out invisible aliens from a far-away galaxy heating the atmosphere with giant ray guns as part of an elaborate practical joke… nor can you rule out deities placing fossils in the earth to test the faith of true believers – but come on, now.

    As with many other aspects of climate science, the observations of atmospheric temperature are now matching the modelled predictions of atmospheric warming. The water vapor feedback effect is also matching the modelled predictions of an increase in atmospheric water vapor.

    Those who feel the need for more information on why radiative models are so complicated could start with Richard Feynman’s short text, QED, on the physics of the interaction of light and matter. Happily, someone recorded those lectures on video at , so you can read along and watch at the same time. Then go back and read the above “basic radiation calculations” link. Repeat this process several times… and that’s just the front end of the climate models. The fact that they are reproducing observations is good evidence of their success; these are the most complex computer models ever created, as far as I know.

  21. Jeff Brown:

    I’m a physicist, and new to this, so pls bear w/ me. I understand that due to computational considerations, it isn’t possible to generate error statistics for climate predictions in the “usual” manner (e.g. Monte Carlo simulations or what have you). Can anyone point me to some references that deal w/ these questions: how ARE error stats derived and how are they rigorously shown to be equivalent to the “usual suspects”? Thanks.

  22. Ike Solem:

    Jeff, I’m not sure what the ‘usual manner’ is for computing errors in models, though there are many… not being a modeller, but someone who has used models (QM-classical mixed models and protein structure prediction models), it seems that the issue is indeed addressed in some detail in climate modelling studies; hope these two references help:
    (Interpolating Climate Model Experiments)
    (Intercomparison of Arctic Regional Climate Models)

  23. John Dodds:

    James Re 38 & #57.

    1. Re Conduction: Energy transport is by 3 mechanisms Convection (movement of hotter atoms) Conduction (movement of hotter electrons), and radiation (movement of photons.) Heat is a form of energy In the air both conduction and convection are relatively SLOW heat/energy movement processes, but they do result in energy transport from the ground to space. Radiation since it consists of photon at the speed of light, followed by absorption & then another photon, another absorption etc is very fast in going thru the atmosphere to space. – according to Eli Rabett a few microseconds for each absorption – so we are talking a less than seconds for radiation to transfer energy from ground to space. so most energy transport is by radiation. Conduction exists in the air because electrons get carried from the ground to the tops of clouds & are also created in the air, – which is why we have charged clouds which result in lightning. but we can ignore it since it is small.

    2. Re Stefan-Boltzmann: Yes the SBL applies to black bodies where the emissivity constant e =1, but for non black bodies the constant just gets adjusted. For the earth it is 0.95 per a lecture on global warming from Columbia University. So the principle still applies- The energy flux (watts/m2) IS calculated by the SBL . If you know the peak temperature of the radiating energy spectrum then you can calculate the outgoing energy flux. If you know the flux you can calculate the temperature. Hotter bodies radiate more – look at the sun.
    When the GHE for added CO2 raises the ground level temperature then the ground/air will radiate MORE back up than what was being radiated pre GHG addition. IF the GHE lowers the TOA temperature, then the Earth at the TOA will radiate LESS energy both up to space and down. If we started pre-CO2, at equilibrium conditions, then there will be a negative energy imbalance at the TOA meaning that the Earth system will absorb more solar energy than it radiates in addition to the GHG part that is recycled. We will warm up.

    3. Next your simple GW model – it is fine up to the last sentence.”Eventually it increases enough until a new equilibrium is formed.” Equilibrium for the earth is defined as Energy-in equals energy out. – in the case of our GW computer models energy-in is ONLY what comes from the sun.. If you do not change the energy-in then changing the energy-out by CO2 absorption can NOT create a NEW equilibrium. What has been created in the computer models is a non-equilibrium condition where the CO2 absorptions delay or slowdown the rate of transmission of energy to space. The longer transit time results in higher ground temps (global warming), and lower TOA temps to conserve energy. See Hansen et al 2005 for these results.

    4. In your planet A & B example. The fallacy is in the “B will cool faster statement.” IF the temperature is the same- then the planets will transport energy out at the same rate, because the only thing that impacts the transport rate is the temperature per the SBL. A conceptual analogy, if 100gpm of water is pushed thru a 10 ft long pipe, and you add a recycle loop that takes 1 gpm from near the outlet and runs it back to near the inlet, the water will still flow in and out at100gpm because that is what you are adding at the front end, just the friction resistance has changed & it takes more power to push the water thru. It WILL take a few extra seconds for the water to transition thru the extra recycle loop, but AT EQUILIBRIUM the 100gpm will flow thru because we did not change the input. (This analogy will fail if you try to model the real environment.)

    5. My QUESTION for the experts is can the atmosphere stay in this non-equilibrium situation in 2/3 above for 20-30 years per the GCMs, or 20,000 if the CO2 keeps increasing as it has since the CO2 low at the bottom of the ice age?

    6.The GCMs say YES. Even with the TOA temp being lower than equilibrium, they say that the excess energy that accumulates goes into the ocean and the TOA temp remains unchanged at below equilibrium for years.(I assume until the CO2 is removed to the ocean?) So why don’t the daily changes in solar energy flow just eliminate the imbalance & return to equilibrium?

    7.The SBL Feedback theory says NO. Equilibrium is restored every day. It postulates that because the very real GHE raises the ground temp, then the raised temp results in more pump power or SBL Feedback to push the energy back out, MORE energy being radiated (& convected & conducted) back up to the TOA where it came from. This larger pumping will push up enough energy to return to equilibrium and accommodate the extra energy that is returned to the ground by the added CO2. The question is will this return the air to equilibrium? In parallel, the SBL feedback is the same mechanism that returns the earth back to its solar equilibrium on a daily basis.- ie hotter air rises and radiates faster as the temperature warms every morning. The SBL can NOT differentiate between GHG or solar caused temperature increases.. It is postulated that the SBL effects to return to equilibrium are not properly included in the GCM. The conclusion is that GHG warming is negated by the SBL feedback. CO2 can increase without changing the temperature. (see 112, 118, 126, 148.. in the section on Al Gore’s movie distribution to schools.)

    OK GCM experts – a comment please

  24. Hank Roberts:

    This may help:

    John, consider why the heat at the core of the Sun doesn’t rush straight out into space:

    “… total Solar energy is determined by the temperature of the Sun’s visible surface, or photosphere, which is about 6000 deg K. This in turn is determined by the Sun’s core temperature of about 15,000,000 deg K, which arises from a balance of inward pressure from gravity and outward pressure from the inner nuclear reactions, and the radiative and convective transports of energy from the core to the photosphere. “

  25. Eli Rabett:

    #70, the way you look at the Earth’s radiative signature is put a satellite out at L2.

  26. Jeff Brown:

    Re 72.
    Thank you. Now I’m going to go one step further and continue to display my ignorance. By the “usual” methods I mean something along the following lines: in a particle physics experiment one accumulates a large # of events and ultimately derives error bars for whatever you’re trying to measure. The more events you have, the better (generally speaking) the result (the smaller the error bars). I gather that it’s not feasible to do this w/ climate models, so how does one assign error bars to, say, sea level rise? (10m +- whatever). Anyway, I’ll check these refs, and any other info is greatly appreciated. Thanks again.

  27. Neal J. King:

    73. John Dodds:

    No, the SBL does not apply with an adjusted constant to frequency-dependent emissivity. If you have a gray body, with non-frequency-dependent emissivity, that constant emissivity can be multiplied by the signma-T**4 to give SBL. But it just doesn’t work if the emissivity is not constant as a function of frequency.

    Don’t believe me? Just go straight to the Planck radiation law and try to integrate it over frequency. If there is a non-constant emissivity factor, you can’t do the integral, so you can’t get to SBL. End of story.

    The fact that hotter bodies give off more radiation doesn’t mean they satisfy the SBL, which is a specific dependence.

  28. Ike Solem:

    That gets to the heart of the problem, because in climate science there is only one experimental event – the ongoing evolution of the land-sea-ice-atmosphere system, with no controls or ability to repeat the experiment. Thus, all experiments must be done ‘in silico’ – imagine if in your particle physics experiments, you just had one chance to collect data from one single event – you’d want to collect as much data as possible, of course.

    The chief problem with climate models, as with other models and experiments, is distinguishing between systematic errors and random errors (say you had a tiny piece of metal in your cloud chamber; that’d create the systematic skewing of results, even though the random variance might be very low after hundreds of observed events). That 2nd reference seems to use multiple independent models to test for systematic errors in the climate models; random variance is apparently tested by running the same model over and over.

    This is why we should have far better data collection systems for monitoring the Earth system – more satellites and more ocean sensors (bottom moored ones deployed on a global basis to measure subsurface ocean temps and currents) and more in-situ measurements of ice sheet dynamics in the Antarctic and Greenland, as well as measurements of the permafrost behavior- it’d be a better use of resources than another ‘race to the Moon’. The only real way to test the models, after all, is to compare their predictions with actual data on an ongoing basis.

  29. John Dodds:

    Re 74 Hank,
    Your quote about the sun contains the key point- There is a BALANCE of inward pressure and outward nuclear radiations….
    This is the key for CO2 caused global warming also. Nature/physics requires a balance, energy-in equals energy-out. The equilibrium exists. (UNLESS you are adding or subtracting energy and adding CO2 does NOT add energy to the earth system))
    The SBL forces a balance in the atmosphere and at all points within the atmosphere. IF it gets warmer the SBL radiates more energy out to return to the balance, if it gets cooler the SBL radiates less out to return to the balance. The SBL is mother natures equilibrium enforcer.

    The Earth on an annual average is at this balance point. It passes thru this point twice a day as the solar energy-in increases and decreases.
    The GCM computer programs require that the the ground level air be warmer than the equilibrium AND the TOA air be cooler. (per Hansen) THIS IS NOT a balance. Even Hansen says that the earth is at an energy imbalance that lasts for years. The GCMs are wrong. Apparently the GCMs do NOT account for the return of the CO2 absorbed energy to the TOA.

    In fact the SBL (hotter air radiates more) will force the balance to return the CO2 greenhouse effect warming to the TOA and return the atmosphere to equilibrium balance. ie NO NET CO2 caused warming. You only get warming if you increase the energy-in, because the SBL will return you to a balance with the energy-in.

    [Response:John, your many comments’ dogmatic statements about what is or is not in GCMs continue to amuse. I appreciate your enthusiasm, but rather than make loud declamations, why not just check it out? There are mutliple GCM codes available for download (GISS, NCAR, EdGCM etc.), so try looking at them and searching through to find places where SB is used (and you’ll find it in all of them). Long-wave radiation is indeed the main ‘balancer’ against warming effects as has been known and used in all models of the climate (even energy balance models) for decades already. No more nonsense please. -gavin]

  30. English:

    re #79

    Sorry to drag this out but I THINK Mr Dodds’ has made two main points.

    One is that dawn temperatures should be largely unaffected by CO2 as there should be enough time at night for any extra heat to go away. (There would still be an increase in mean temperature but this is not the whole story, at least not for Mr Dodds.) Can anyone say if this is as measured?

    His other point seems to be that the lower atmosphere IS heating but this extra heat will eventually be passed into the upper atmosphere by conduction and radiation. That is, if we stopped putting any more CO2 into the atmosphere, the mean temperature would eventually fall (perhaps slightly) after some indeterminate time. There would be some overshoot in the lower atmosphere temperature for a step change in CO2 concentration. Does anyone know if the models agree with this statement?

  31. Dan Hughes:

    re: #70. “… these are the most complex computer models ever created, as far as I know.”

    I’m certain that AOLGCMs are not the most complex computer models ever created. How wold you like to start comparing the complexity of computer models of inherently complex physical phenomena and processes that occur in complex physical geometries. Close to 100 thousand lines of code is my estimate of the entry level computer model for such applications. Truly complex codes consist of a few million lines of code. The AOLGCM codes will fall closer to the former estimate than to the latter.

  32. Bruce Hall:

    Those darn Chinese are trying to use crude modeling to say that the next 20 years brings cooling instead of warming and that CO2 increases will not be that relevant.

    Multi-scale analysis of global temperature changes
    and trend of a drop in temperature in the next 20 years

    Obviously a Communist plot.

    [Response: No, just rubbish. -gavin]

  33. John Dodds:

    Gavin, Thanks for acknowledging.
    I keep asking because you know better what is in the GCM (would I ever find that there are 365.0 days in a model year?). I do not want to become an expert.
    AND philosophically IF my view is to prevail then I have to convince the experts (ie you) that the GCM is wrong. It does me no good to argue about parameter values like Temperature hockey sticks, that Monkton & Junkscience and most of the other denyers do because I can’t win that argument. I HAVE to show that the model is wrong or inconsistent or missing something OR accept that it works. (but a ray of hope for you- you & readers have guided me thru the entire process of the model from WV feedback to the actual absorption & energy return to the air (per Eli) process – I am NEAR THE END in that I am now challenging if Arrhenius(& the GCMs) included the SB process to return to equibrium of not. My conclusion is STILL that something is wrong. You have NOT refuted my points – just ignored them.

    Why can’t you just respond to the physics itself?
    I do not understand why the energy imbalance (ground warmer, TOA cooler) would not be equalized by the SB forces? After all the SB forces MUST change and be larger because the ground temp is warmer. It seems to me that the SB forces to equalize, would have to be faster than the CO2 absorbtions, because they act on all the air, not just the GHGs, and they act on daily temperature changes that are much larger than GHGs – AND we still return to equilibrium. The SB forces & especially what I refer to as SB Feedback is automatic to return the system to equilibrium. The only reasons I can see to explain them is that somehow they are missing (misapplied?) in the GCMs, hence I offer that explanation. – I will stop asking if the GCMs missed them, as you requested.

    BUT please explain the apparent GCM inconsistencies that I identified in #38. Where are they wrong? To paraphrase you – you are dogmatically insisting that the GCMs are correct (which I understand since you have spent so much time on them)- I identify inconsistencies and you ignore them. – you do NOT refute them.- at least way back when, when you explained, I could move on. As it is I have come back to my discussion points from 15 months ago where you said I assumed equilibrium & you said that was wrong (but not WHY). I did assume then , BUT now I see that the SB forces should reestablish it. – see 38- so WHY is it wrong?)

    Quite frankly, the idea that the inequilibrium/energy imbalance could last for many or 20,000 years or as long as the CO2 is increasing just seems impossible. After all we go thru equilibrium every day. The amount of energy transferred by GHGs is downright trival compared to what flows thru the earth at all times. The equilibrium concept must be dominant – the documentation for the research (publish papers) even identifies it as a requirement. Hank’s sources identify equilibrium as a requirement for the sun & earth energy flows and balances. So how can the GCM not allow equlibrium for the entire run length?

    OR try this analogy (I know, analogies always seem to fail eventually – but at least this demonstraes why I think we return to equilibrium!)
    Think of the energy transport as a frictionless pipe carrying water. The energy/water in from the sun is 100gpm, at equilibrium the energy/water at the outlet is 100gpm. If I add a GHG recycle/bypass line that takes 2 GPM of water /energy from near the outlet & recycles it to near the inlet, then I have simulated the atmosphere with the potential to double or change CO2 (igmore the actual numbers – use the concept) . If I suddenly open a valve to allow water to go into the recycle line, then it immediately creates the GCM conditions (ie more water at the inlet, less water at the outlet/TOA during the NON-equilibrium response,) BUT as soon as the system returns to equilibrium (as soon as the GHG recycle line fills up) I am left with equilibrium, 100gpm in, 100gpm out BUT a total of 102 GPM going down in the pipe (faster SB energy flow to compensate for the GHGs slowing down the outward flow) with a 2gpm flowing back up the recycle line – ie net 100gpm within the pipes total.- yes the flow rates in the pipes increase as more GHGs are added, but the equilibrium conditions are maintained because the increased water/energy flow returns the system to equilibrium immediately.
    The question – where does the water/energy accumulate over time. My view it doesn’t. the increased downflow just speeds up & returns the system to equilibrium where the CO2 flow is returned to where it came from at the outlet/TOA.

    Philosphically, IF the equilibrium conditons canNOT prevail and be reestablished in the earth, then ANY little change will upset the balance forever. A return to equilibrium (ie no CO2) would never be possible without reversing the the actual change. Such a situation does not seem to model the earth. The earth has compensating mechanisms and we are talking about energy which can assume many different forms. The GCM does NOT even include the energy contained in the Earth’s magnetic field or in the solid earth or the core. How do you know that they do not come into play in the balance? The Earth Mag field has reduced by 10% in the last century. What happened to that energy that was used to maintain the field?

    Sorry I’m a skeptic who has to find out why. The GCM model just doesn;t feel right.

  34. James:

    Re 80: You say “…dawn temperatures should be largely unaffected by CO2 as there should be enough time at night for any extra heat to go away”, but that’s obviously not true.

    What exactly is this “extra” heat? Seems to me anything much over the 3K microwave background must be considered extra, and would radiate away given time – and indeed, the surface of the moon drops to somewhere around 100 K during the night. Or think about why the polar regions get cold during their sunless winters.

    It seems that you’ve missed the important point, which is that CO2 changes the rate at which heat radiates away. It’s not all that different in principle from adding more insulation to your attic.

  35. Barton Paul Levenson:

    Re “The SB forces & especially what I refer to as SB Feedback is automatic to return the system to equilibrium. The only reasons I can see to explain them is that somehow they are missing (misapplied?) in the GCMs, hence I offer that explanation. – I will stop asking if the GCMs missed them, as you requested.”

    They aren’t missing. You have a wrong idea of what the Stefan-Boltzmann feedback means and how it works, and your misconception is what’s missing from the models. If you follow your idea out logically, you seem to be saying that nothing can ever change its temperature, because the SB feedback would immediately put it back in equilibrium. That’s not what equilibrium means and that’s not how it works.

    If you want to understand how the Stefan-Boltzmann law works, work out an example QUANTITATIVELY. Do the math. And as the math teachers say, show your work.

  36. Jeffrey Davis:

    re: 82

    The Chinese are adding some ungodly number of coal fired power generating plants every year. They’ve got maybe the worst environmental record on the planet. They’re authoritarian and dictatorial.

    What are the odds that they’d produce something to justify what they’ve been doing and where they’re going?

  37. Hank Roberts:

    Look again at the sun. You grasped the word ‘balance’ but you’re taking it from the part of the sentence referring to physical pressure from gravity balancing pressure from the nuclear reactions, yes, it says that. You don’t think gravity holds the infrared photons inside the sun, of course. The sentence talks about the issue you were focusing on up til that word grabbed your focus — conduction and radiation transfer heat. That was your big issue

    How long does it take a photon to move from the surface of earth into space? You claim above it happens at the speed of light.

    How long does it take a photon to move from the center of the sun into space?

    What’s different about the two conditions? Gravity, temperature, pressure, density and composition of the medium.

    I’m not a physicist; I am relying on a lot of published work. I read many of the references here as well as the basic article:

    To the physicists — I suggest this does need to be explained and the math set out explicitly, because looking around there are far more websites proclaiming that the physics isn’t right than I’d imagined, most of them very recent. It seems like there’s a great number of people lately who are denying this research has been done or that it can be possible there’s a problem.

    I _think_ the terms needed may be ‘minimum free path’ — how long a photon travels before interacting — and then how long it’s held in a CO2 molecule before it’s re-emitted in a random directdion.

    Random direction means, likely to hit another CO2 molecule. I doubt anyone can arrange for the atmosphere’s CO2 molecules to emit their infrared photons in one consistent direction. If so it’d be one hell of a weapon.

  38. Bruce Hall:


    Before everyone dismisses the Chinese as duplicitous with regard to the publication on potential global cooling, here is a post by Lubos Motl, Harvard physicist, that cites Russian and Ukranian scientists saying some very similar things… except they are even more extreme in their predictions.

    One cannot attribute completely to politics those scientific conclusions that you don’t want to accept. One must confront the positions with other than ad hominem dismissals… dare I say one must use scientific review of that which is published and provide specific grounds for rebuttal? Since one of the physicist is acting as a contact and has provided and email address, it would seem reasonable to make the effort to contact him for more information about the study and the actual data used.

    If one bothered to actually read the Chinese publication cited in #82, one might then be able to address the issues… and realize that the publication did say that CO2 accounted for approximately 40% of the change in the analysis… just less than current “global warming correct” adherents demand.

    A closed mind is a terrible thing to waste.

    [Response: A dismissal of a Russian or a Chinese publication or media report is not a dismissal of their entire scientific output or their government’s policy. There is much to be commended in both countries scientific processes. There has been a tendency for certain commentors to grab at any straw available that seems to cast doubt on the mainstream consensus – that’s fine of course, but those straws need to be examined critically. When it comes to comments in Moscow News that are not backed up by any published work or demonstrated expertise, that dismissal is easy. The Chinese publication is simply a statistical analysis of two temperature time series with no physics at all, and so it’s potential to give good predictions in the future is zero (there being an infinite number of statistical models of equal or better skill that would produce any prediction one would like). Attribution of temperature changes and prediction can only be acheived through physical modelling – absent that, you might as well cast runes. -gavin]

  39. English:

    re #84

    I was asking if anyone knew of any empirical evidence to support Mr Dodd’s statements or otherwise.

    By “Extra Heat” I meant the increase in heat in the lower atmosphere caused by anthropogenic CO2.

    The overshoot may be of interest as it may mean that the models are projecting slightly too high a temperature. The difference is very likely to be well within model errors, anyway, so it’s an academic point but an interesting one for someone who has been involved in modelling.

    “It’s not all that different in principle from adding more insulation to your attic.” That is mostly a convection effect and Mr Dodd’s does not seem to argue with the current views of convection effects in the atmosphere.

  40. Steve Latham:

    To John Dodds (#83),
    First you have to note, John, that the models aren’t “right”. No model is right. The question is whether or not they have all the important ingredients in proportions that will allow an adequate representation such that we can learn about the thing being modeled. Apologies in advance, from here this posting gets murkier….

    I dunno how much interest you’ll get in your analogy. I tried to figure out what you were saying, though, because your inquiry seems honest enough. Unfortunately, I’m only a biologist, but let me see if I understand what you wrote. The water represents heat. Let’s say the GHG loop represents the extra heat in the system. So, the GHG’s don’t change the amount coming from the sun, but the total amount in the Earth’s system is increased before flowing out to space (or out the end of the hose which still equals the same amount coming in). Okay, to me there doesn’t seem to be a problem there. Now take into consideration some feedbacks and your analogy might begin to take shape.

    Some people have done analogies with water pouring, but let’s try something different, an experiment you could try in your home. I have no idea if this will work:

    Take two equal wires (unwound coathangers perhaps?) and shut one end of each into your fridge. Now you need two cans at room temperature with thermometers in them. The other end comes out and into a soda can that has been filled with water (each can has to be the same, side by side, with the same amount of water, etc). Now, if the wires are conducting heat away from the cans, the thermometers should be reading lower than they were when they were at room temperature. Now insulate one of the cans in bubble wrap or something. Because this can is now gaining heat from the room at a lower rate, it will cool faster than the other can. It will reach equilibrium eventually, but its equilibrium temperature will be lower than that of the other can. Maybe in reverse this would be a better analogy, with the wires touching a hotplate and the insulated can getting warmer. But hopefully you get the idea. Now, what has to be done to this system to take into consideration your concerns?

  41. L. David Cooke:

    RE: #89

    Hey English;

    I suspect that insulation is more likely related to conduction then convection. I agree that on the surface there appears to be the possibility that model error and possible mathematical calculations might carry an accumulating error, it is still very likely that there is atmospheric warming occurring, though it is likely not as great as was attributed in the IPCC 2001 report.

    (In relation to my earlier posts regarding the direct measurement of low level radiant IR, if a 2 watt (White/Blue GaAs(P or Ni)) LED or Solid State Laser was suspended at 1000 feet in the air I might be able to see it. (Gallium and Arsenide (GaAs) emits infrared light. Gallium Arsenide Phosphide (GaAsP) emits either red or yellow light, and Gallium Phosphide (GaP) emits red or green light.) But, at 10,000 feet it would be less likely and again at 70,000 feet it is more unlikely, even in a clear night sky as the photon density is less by a factor of 4 for twice the distance. Based on this, I suspect if there is heating due to GHG, it must be centered in the lower several thousand feet of the atmosphere (troposphere).)

    The current theory is that CO2 would be acting as heat trap and reducing the radiation of the terrestrial IR into the upper atmosphere and into space. The heat trap would be different from an insulator in that the insulator would be preventing convection or conduction. The heat trap or absorption and re-emission of the IR by GHG is supposed to be reducing the radiation path to the upper atmosphere. As to the other heat paths, some form of temperature inversion would have to explain the convection issue and conduction would be blocked due to the low density of the medium. So I don’t know if Mr. Dodd is addressing the terrestrial heat transfer path correctly yet.

    Dave Cooke

  42. Bruce Hall:


    I recognize that climate modelling using supercomputers is a different effort than statistical analysis.

    Often statistical models will show strong positive correlations between data that may or may not have a logical causal relationship. Whether or not the Chinese or Russian and Ukrainian studies were more than “quick and dirty” statistical analyses or legitimate analyses of major climate influences based on “real physics” certainly should be determined.

    It is important to not disregard macro analyses in favor of models econtaining multiple thousands of lines of code and assumptions, especially if the macro analyses include the major climate influencing factors. If historical data show that CO2 concentrations are not correlated with absolute temperatures and lag major temperature increase trends by up to 800 years(and geologically that is the case), then it is important to be able to say why that is not the case this time. Just because a short-timeline model fits now, doesn’t excuse the model if it doesn’t backfit geologically.

    Macro models can be useful for a reality check… even if they don’t have 100,000s lines of code.

    [Response: The reason why we know CO2 is not following temperature this time is because we know that we’re putting more into the atmosphere than is staying and that the excesses are going into the oceans (mainly). Statistics from the ice ages (which show a strong correlation, not none!) have nothing to say on the matter (but just as an exercise, calculate what the predicted increase in CO2 would be over a multi-century timescale assuming such an ice-age-based statistical model was valid – then come back and try and make your point again). More to the point, where are these sophisitcated Russian statistical models of which you speak? Find a paper and I’ll analyse it. Until then this is all smoke and mirrors. – gavin]

  43. Bruce Hall:


    I don’t have the source documents for the Russian studies which are only abstracted in (which I’m sure you are familiar). You can probably obtain them with a little effort. This forum is fine for lighter weight discussions and offering up related news, not complete examinations of models. You don’t have to pursue the leads; others will, no doubt.

    Once again, the point of geological studies as a checkpoint is that absolute concentrations of CO2 have not been correlated with absolute temperature levels… something different from a correlation between temperature changes and CO2 changes that follow.

    Dr. Patterson can be contacted at Carleton University ( if you really want to examine the geological studies that make those points. I don’t dispute your knowledge of the dynamics of atmospheric gases. I’m suggesting that other disciplines have something to say about climate change, too… like geology and astrophysics. The earth is not a closed system or a static one geologically.

    Or, you can just keep calling anyone who might offer up differing perspectives a hack and go on with close-system thinking.

    [Response: As an exercise in jumping to conclusions your posts are pretty good. Where did you ever get the idea that I advocate not talking to geologists? (Try looking at my publication record to see the intereactions with paleo-climatologists for instance). The best records that have both temperature and CO2 changes are the ice cores, and I am flabbergasted that you think they show no correlation between CO2 and temperature over the last 650,000 years. On the contrary, the relationships are very obvious and causal in both directions (though you need more information than just the two records to determine that). Still doesn’t have much to do with the change in CO2 occuring now (as you would see from doing the calculation I suggested above).

    Going back to our Russian colleague – the abstract linked discusses only potential changes to the solar forcing. This is fine topic for discussion, but one should state straightaway that this is only one of many forecasts for the upcoming solar cycles (which differ enormously). Most importantly however, you should note that no comparison of these changes with the other forcings in the system have been made. Since the climate will react to the net forcing, that is essential to working out whether a cooling should be expected. Since the measured solar forcing (solar max to min) is so small (0.1% ~ 0.24 W/m2), any change of that magnitude is not going to make any difference in the next few decades. -gavin]

  44. Hank Roberts:

    Dave Cooke writes: “The heat trap or absorption and re-emission of the IR by GHG is supposed to be reducing the radiation path to the upper atmosphere.”

    I believe that’s backwards — increasing CO2 increases the radiation path, each photon has a longer path/distance/time to travel before it escapes the planet, because there are more opportunities for each photon to be captured, particularly at low pressure/high altitude where the absorbtion bands are more effective. See Weart’s AIP article link above.

  45. Bruce Hall:


    Check you email.

    These discussions are too terse and too easy to have “tones” that are not part of normal discussion… which I think we both did here.

    My “tone” was not regarding your collaboration with other disciplines which is obvious from your publications and statements elsewhere in this site. It was more directed to the initial “rubbish” comment.

  46. L. David Cooke:

    RE: #94

    Hey Hank;

    You are correct, I had meant to say reduced, the ability of transport of the radiant energy to the upper atmosphere. Or as you would put it,
    increase the time that the terrestrial IR energy is trapped in the lower atmosphere.

    Dave Cooke

  47. James:

    Re #89: “By “Extra Heat” I meant the increase in heat in the lower atmosphere caused by anthropogenic CO2.” – Which I what I thought, so my argument still applies. The night is nowhere near long enough for the dark side of Earth to radiate down to thermal equilibrium. The bottom line is that, all else being equal, if some area starts out warmer at sunset, it will be warmer at dawn.

    “”It’s not all that different in principle from adding more insulation to your attic.” That is mostly a convection effect…”

    OK, if you want a better analogy, it’s like using low-E glass in your windows :-)

  48. Ike Solem:

    Dan, aren’t weather and climate models themselves “models of inherently complex physical phenomena and processes that occur in complex physical geometries.”?

    In fact, if you assign a temperature value to a point (x,y,z) in space you have already created a 4-dimensional output F(x,y,z). Let’s say you want to calculate the local humidity as a function of pressure(P), temperature(T) and location(x,y,z) – and then allow the whole system to evolve in time(t), and what you have is humidity as a function of (x,y,z,P,T,t) – which is something like what weather models try to do when they come up with rainfall predictions, isn’t it? 6-dimensional geometries calculated over the entire planet – isn’t that what you are talking about – i.e. climate and weather models?

  49. English:

    re #97

    Thanks for the reply but I was asking for actual measurements.

    I can only give my own measurements – the temperature inside my shed-cum-greenhouse (when I had one) with greenhouse glass reached the same as the outside temperature before dawn.

    By my maths, the temperature of the lower atmosphere should decay exponentially towards an aymptote, so dawn temperatures should be affected less than dusk temperatures. I’m not sure if the asymptote is the 3K background or the temperature of the stratosphere.

    But it would be interesting to know.

  50. James:

    “…the temperature inside my shed-cum-greenhouse… reached the same as the outside temperature before dawn.”

    Maybe that has something to do with the fact that your greenhouse is a lot smaller than the atmosphere, and so has a lot less thermal mass to store heat. Look at any source on passive solar design, and you’ll find they’ll stress the importance of having enough thermal mass to store heat overnight.