Climate Science Glossary

Term Lookup

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

Settings

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

Term Lookup

Settings


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

Home Arguments Software Resources Comments The Consensus Project Translations About Donate

Twitter Facebook YouTube Pinterest

RSS Posts RSS Comments Email Subscribe


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



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

Latest Posts

Archives

How reliable are climate models?

What the science says...

Select a level... Basic Intermediate

Models successfully reproduce temperatures since 1900 globally, by land, in the air and the ocean.

Climate Myth...

Models are unreliable
"[Models] are full of fudge factors that are fitted to the existing climate, so the models more or less agree with the observed data. But there is no reason to believe that the same fudge factors would give the right behaviour in a world with different chemistry, for example in a world with increased CO2 in the atmosphere."  (Freeman Dyson)

Climate models are mathematical representations of the interactions between the atmosphere, oceans, land surface, ice – and the sun. This is clearly a very complex task, so models are built to estimate trends rather than events. For example, a climate model can tell you it will be cold in winter, but it can’t tell you what the temperature will be on a specific day – that’s weather forecasting. Climate trends are weather, averaged out over time - usually 30 years. Trends are important because they eliminate - or "smooth out" - single events that may be extreme, but quite rare.

Climate models have to be tested to find out if they work. We can’t wait for 30 years to see if a model is any good or not; models are tested against the past, against what we know happened. If a model can correctly predict trends from a starting point somewhere in the past, we could expect it to predict with reasonable certainty what might happen in the future.

So all models are first tested in a process called Hindcasting. The models used to predict future global warming can accurately map past climate changes. If they get the past right, there is no reason to think their predictions would be wrong. Testing models against the existing instrumental record suggested CO2 must cause global warming, because the models could not simulate what had already happened unless the extra CO2 was added to the model. All other known forcings are adequate in explaining temperature variations prior to the rise in temperature over the last thirty years, while none of them are capable of explaining the rise in the past thirty years.  CO2 does explain that rise, and explains it completely without any need for additional, as yet unknown forcings.

Where models have been running for sufficient time, they have also been proved to make accurate predictions. For example, the eruption of Mt. Pinatubo allowed modellers to test the accuracy of models by feeding in the data about the eruption. The models successfully predicted the climatic response after the eruption. Models also correctly predicted other effects subsequently confirmed by observation, including greater warming in the Arctic and over land, greater warming at night, and stratospheric cooling.

The climate models, far from being melodramatic, may be conservative in the predictions they produce. For example, here’s a graph of sea level rise:

Observed sea level rise since 1970 from tide gauge data (red) and satellite measurements (blue) compared to model projections for 1990-2010 from the IPCC Third Assessment Report (grey band).  (Source: The Copenhagen Diagnosis, 2009)

Here, the models have understated the problem. In reality, observed sea level is tracking at the upper range of the model projections. There are other examples of models being too conservative, rather than alarmist as some portray them. All models have limits - uncertainties - for they are modelling complex systems. However, all models improve over time, and with increasing sources of real-world information such as satellites, the output of climate models can be constantly refined to increase their power and usefulness.

Climate models have already predicted many of the phenomena for which we now have empirical evidence. Climate models form a reliable guide to potential climate change.

Basic rebuttal written by GPWayne

Last updated on 26 February 2014 by LarryM. View Archives

Printable Version  |  Offline PDF Version  |  Link to this page

Further reading

Update

On 21 January 2012, 'the skeptic argument' was revised to correct for some small formatting errors.

Comments

Prev  1  2  3  4  5  6  7  8  

Comments 351 to 398 out of 398:

  1. #354 CBDunkerson at 19:22 PM on 1 May, 2011
    we are instead currently seeing accumulation of heat in both hemispheres

    No, we are currently not seeing anything like that. Rate of heat accumulation in the upper 700 m of oceans in 2003-2010 (when it is measured properly by ARGO) is 5.5±6.5×1020 J/year. That is, it's flat.

    Even if we go with the upper bound (12×1020 J/year), it takes 900 years to warm up this layer by 1°C, which is a warming rate of 0.11°C/century. However, it is entirely possible that the upper ocean is actually losing heat.

    Note that rate of change in surface temperatures in general can't be far removed from rate of upper ocean heating, as heat capacity of oceans is three orders of magnitude higher than that of the atmosphere.

    We also have (rather low quality) data for 1955-2010. If we take it on face value and believe the error bars provided by Levitus et al. are correct, average rate of heat accumulation in the upper ocean during this 56 year period is 25.3±0.5×1020 J/year. At this rate it takes at least 450 years to heat it up by 1°C, which is 0.22°C/century.

    It is much smaller than the alleged late 20th century warming rate of the surface, so that could only be a transient phenomenon mostly due to redistribution of heat in the climate system (and also failure in properly taking into account local warming close to land based measurement sites due to land use changes - a.k.a. UHI).
  2. 353, trunkmonkey,

    See here and here for info on models.

    In particular, last section of the first link is titled "Can I use a climate model myself?".
  3. "I care deeply about this. It may be the defining issue of our time."

    How about reading Ch6 (Paleoclimate) of the WG1 then? I think you will find what want in the referenced papers.

    Nature of DO's. Well DO events appear to only have strong climatic effect when exiting a glacial. Its not the adding of CO2 that changes things but loss of the ice sheet.

    Candidates for Bond/Heinrich/DO events - solar and changes to thermohaline circulation. Any of those happening to explain the current warming? Nope.
  4. 354.

    I was rounding up. The models are initialized at 280ppm in 1750 or whatever, but humans have been burning since fire was domesticated. How much CO2 would you figure was released by the burning of the European oak forest (much of it to slake lime)? It is believed neolithic (and probably paleolithic) hunters routinely set uncontrolled fires to promote savannah at the expense of forest and chapparal.

    DO/Bond

    Don't you expect some surface cooling with the fresh meltwater that won't sink? Isn't this heat redistrubution?

    Why is the Atlantic deepwater way more oxygenated than the Pacific?
    Response: [DB] As for your first part, if I see where you're going with that, you should look into Ruddiman's Hypothesis. Ruddiman himself has a recent (and still active) thread over at Real Climate where Ruddiman has been answering questions.
  5. "Don't you expect some surface cooling with the fresh meltwater that won't sink? Isn't this heat redistrubution?"

    That would be why DO have climate effects coming out of interglacial but not now?
  6. trunkmonkey wrote: "How much CO2 would you figure was released by the burning of the European oak forest (much of it to slake lime)?"

    First, I don't see how this is relevant to CO2 levels in the models... given that those are based on measured atmospheric CO2 values, CO2 levels in air bubbles in ice cores, and various proxies.

    Second, you are ignoring the other half of the equation... when you burn down forestland you get grassland or cropland. All that new vegetation needs alot of carbon... which it gets from the atmosphere. The net effect is very little change in atmospheric CO2... though Ruddiman and others have suggested that the ~20 ppm rise over the ~8000 years prior to the industrial revolution was due to such 'land use' changes.
  7. scaddenp 359: "That would be why DO have climate effects coming out of interglacial but not now?"

    Does the system really care about glacial/interglacial stage? Isn't meltwater, meltwater, whenever it ocurrs? If our efforts have accelerated the melting, does it matter that the triple net Milankovitch "forcings" languish in the future?

    Berenyi Peter, 355., is saying that he sees no SST increase.

    (Alley 2005) has an interesting discussion of this. He states that several models predict meltwater, but that the results are equivocal.
  8. 360.

    The signature of the burning must be included in the ice cores, so I concede this, except for the possibility that the models are initialized a bit too high.

    Mizmi is the real expert but grasses are C4 plants adapted to the declining CO2 levels of the Pleiocene. They use less CO2 that the forest they replace.
  9. "Isn't meltwater, meltwater,"

    Volume matters. At end of glacial, you have large amount of ice to melt below the arctic circle. The issue of interest for now wrt to sealevel rates, is that rate of warming is far higher than exiting a glacial but amount of ice available to melt is far less.

    As to BP, look at the data yourself.

    As to role of meltwater - this is unsettled science. There is good evidence of disturbance to thermahaline cycle and ditto for solar forcing. What causes the disturbance, relationships and timing is not settled. If you are really interested, read Wally Broecker on the subject.

    Relevance to now? Well no solar forcing and no disturbance to thermahaline cycle detected as possible causes of warming.
  10. 363.

    We established a while back that we know nothing at a decadal scale because the models can't resolve the irregular and powerful influenses of the ocean sloshings; ENSO, PDO, AMO, IOD (It's almost like Lake Tahoe must have a dipole as well). These influences are stronger than the expected warming so we could concievably have a decade of cooling or a decade of warming much greater than the models predict and it would mean nothing (except politically).

    The implication of DO/meltwater is profound, because if it is internally driven, we might not know anytinig at a centennial or millenial scale either.
  11. trunkmonkey - So, a little uncertainty, and we know nothing? That's not a reasonable statement, trunkmonkey. I don't think that's been established anywhere.

    Tamino has demonstrated how to look at ENSO and other local variations, and remove their influence. The results? The warming we expect from the CO2 we've added, at the trends we expect from the physics.

    A wee uncertainty is reason for resolving how much we know despite the uncertainty, not reason for throwing up our hands and giving up.

    That said, an overturning of the thermohaline circulation due to increased fresh water would be a large change in climate state. But that hasn't happened for a long time, and won't happen unless due to our influence, barring major changes in natural forcings which are well out of what we've seen in millenia.
  12. Slight difference. We can't predict ENSO, PDO,AMO in models. These internal variabilities exist in models. You can certainly run a model and get PDO index out of it. However, you cant initialise a model to predict them.

    I don't of any paper which puts a case for DO being internally driven. Do you? Everything has physical causes.
  13. trunkmonkey#364: "we know nothing at a decadal scale"

    We know this at a decadal scale: Global temperatures are increasing at ~0.15C/decade; northern hemisphere temperatures at 0.3-0.5C/decade. We know of no reasons why that will decrease anytime soon.
  14. trunkmonkey wrote: "The implication of DO/meltwater is profound, because if it is internally driven, we might not know anytinig at a centennial or millenial scale either."

    Again, this would be observed as a transfer of 'heat' from one hemisphere to the other. Yet all records (i.e. surface thermometers, ocean water temps, satellite readings, et cetera) show that both hemispheres have been warming. Ergo, the observed warming is unquestionably NOT D-O related. And ditto ENSO, PDO, et cetera. These are all examples of energy moving around within the climate system... whereas what we have observed over the past 100 years or so is an increase in energy throughout every part of the climate system.

    Essentially, your argument is the equivalent of saying that if you pour a gallon of water into multiple different containers it can turn into two gallons of water. In reality it doesn't matter how much you move the water (or heat) around... the amount doesn't change.
  15. "I have seen many projections of the models into the future. You claim that the models are hindcast, but I have never seen a graphic to demonstrate."

    Running a full climate model over 800,000 years or so? Um. That would be tricky so you need simplication. More common to do full runs for specific period of interest like LGM, PETM, YD etc. Again, IPCC WG1 is place for the references.

    You could see the Hansen and Sato for a much simpler calculation covering last 800,000 years.
  16. Here some new stuff about climate models.

    According a paper from J.Hansen, M.Sato and P.Kharecha
    most IPCC models and GISS modelE-R mix heat too efficiëntly downward through the oceans. The result is that models respond slower on forcings as the real world does.

    On page 18 in Hansen et al 2011 I read "Below we argue that the real world response function is faster than that of modelE-R. We also suggest that most global climate models are similarly too sluggish in their response to a climate forcing and that this has important implications for anticipated climate change."

    Adjustments towards less ocean heat uptake are in better agreement with observed OCH trends. Despite the slow response IPCC models do a good job in mimicing global warming. The IPCC underestimates aërosol cooling effects and some models have too large GHG forcings according Hansen. Since the oceans are NOT the favourable place for large antropogenic aërosol influences the answer may be quite more simple: Climate models are too sensitive to GHG (and other forcings) and thus OVERestimate AGW.

    You can read more about this in chapter 5 t/m 7 of the paper.(Excuse me if this subject is discussed before).
  17. KR 365.

    The models are unable to make specific predictions at a decadal scale to be contradicted or confirmed, except that it will be warmer at the end.Various alternative celestial and cyclical hypotheses are likewise untestable at a decadal scale.

    Those who wait breathlessly for yearly GAT data are making essentially the same mistake as one who guages global warming by his backyard thermometer.
  18. CBDunkerson 368.

    I don't see THC as hemispheric balancing. The Antarctic has plenty of cold salty water. Some say it even undercuts the Atlantic bottom water. I would describe the THC as an Antarctic beltway, both bottom and surface water, with three feed loops into the Pacific, Atlantic, and Inian Oceans. It balances overall ocean temperature by cooling the Pacific and Indian Oceans and warming the Atlantic, the only ocean directly connected to Arctic bttom water.

    I am not trying to reconstruct Singer and Avery.They have done that ably, and it is interesting, but I am tired of the notion that adding CO2 does NOTHING.

    It would be nice if we could get a model to reproduce the Younger Dryas or the 8.2ky event by adding meltwater. My suggestion: try it again with the CO2 knob backed off.
  19. Trunkmonkey - try Wally Was Right, Alley 2007.

    I have extremely limited internet access at moment, so hard to further. I dont really understand what you mean by "CO2 knob" backed off. Firstly, YD is too fast for CO2 feedback to be much of a factor. Secondly, all CO2 does in a model is change radiative forcing. I dont think there is an Ar4 model that could realistically predict CO2 change as a feedback - can only feed in what actually happened.
  20. trunkmonkey #372: There is a well documented hemispheric see-saw to DO events... whether you see it or not.

    In any case, your description of "cooling the Pacific and Indian Oceans and warming the Atlantic" is just fine for making my point too... because the Pacific and Indian Oceans did not cool over the course of the 20th century. They warmed. Just like every other ocean on the planet. Total ocean heat content increased. Total atmospheric heat content increased. Ergo, none of these changes can be put down to 'internal variability'... because that would require decreasing temperatures somewhere else and there just isn't any data showing that.
  21. CBDunkerson 374.

    Perturbing the THC by jumping on one end and finding the other end rises does not mean that DO events funtion to balance hemispheric temperatures.

    I see your point that universally higher SST's would contraindicate DO as an explanation for current warming.

    Bear in mind that the deep water now upwelling is very old. Much of it emerging now began sinking during the Medieval Warm Period as the Vikings sacked England, and some may be 1600 years old.

    There is a lot of cold salty water in that pipeline that is insulated from warming at the surface.
  22. scaddenp 373.

    Thanks for the Wally link. That is a very impressive paper. The models get the jist of the YD but they don't relicate it's full amplitude.

    From what little progress I have made in the decade of study you guys have prescribed I've gleaned that models have their own logic when they find a stable sweet spot. You guys know better than anyone that when you do certain things the model gets crazy and runs out of bounds like a kid who's eaten too much candy.

    So we have this happy model that refuses to reproduce the amplitude of YD by adding a reasonable amount of meltwater. Models have a good track record in these situations. So either the proxy data are wrong, there is anther stable point the model should be initialized at, or another unknown factor is contributing.

    I still do not understand exactly how CO2 works in the models. I played with the edgcm model but all it will let me do is imput a ppm for CO2.

    In the idealized model in my mind I would be able to right click on CO2 an pull up its properties. A screen would pop up showing all the relationships for CO2 and the values applied to these relationships. For example the absorbtion of outgoing IR would show a relationship to air temperature at 6 w/m^2 or so and there would be diminishing relationships to soil, ocean and plant sequestration, an exaggerating relation to temperature as it feeds back to increased microbial activity, etc.

    My suggestion on the control knob is to back off these values to bare bones. Can YD be too fast for CO2 if sucking 380 ppm of it out of the atmosphere today would drop GAT 6 degrees in a year?
  23. Truckmonkey, you are missing the CO2 handle. Firstly, in paleoclimate CO2 is ONLY a feedback. It responds to temperature and amplifies whatever else is happening to temperature by changing the radiative forcing. There are two things to note about this.
    1/ The CO2 feedbacks are SLOW. They are thought to have minimal if any effect over climate in next century. This is NOT to say that the effect of CO2 forcing is slow - only that the change in concentration of CO2 in the atmosphere in response to temperature is slow. When CO2 concentration does change, then the effect in instantaneous more or less.
    2/ Most AR4 models dont even consider the CO2 feedback. AR5 will, but with what skill? For paleoclimate, studies do have to consider that. However, at every time step in the model, you have to know boundary conditions. One of these is CO2 concentration in atmosphere. Now to determine climate in say 2-300 years, you would need to what the CO2 concentration is. You have do this with a combination of both scenario - how CO2 are humans likely to emit - and a carbon cycle model - how much will CO2 change due to temperature rise. For paleoclimate though, you dont have to have a carbon-cycle model at all. You can just put in the CO2 concentration at that time. Of course if you are trying to understand what happens to carbon cycle, then you need model, but my understanding is these are so far somewhat unconstrained - there are many ways to reproduce the CO2 response to temperature change without so far easy ways to favour one versus the other.

    You comment on models parameter implies you really need to study how these models work. The only thing in climate model that changes with CO2 concentration is the radiative forcing. There is no link in the code between that and the other factors you mention (which are mostly carbon cycle model parameters and not in models for reasons above). Climate models get complicated by feedback and this is related to temperature, not directly to CO2. There is no control knob in the model like you imply to "turn down". Alley comments are an observation about what model imply, not a description of the model.
  24. I think, many problems concerning the value of numerical prediction arise from the term "precise". It is widely used here in a binary manner as an all-or-nothing gauge. It would be much better to talk about probabilities and ranges.
    Concerning the comparison of complex simulation models with polynomials: it is right, that both, because of input data uncertainty ranges, have to be fitted to the past. But in the physical models is so much more information integrated about how the world works, that common sense tells us, that their result is (again probability!) most probably more reliable than a simple polynomial.
    I am a physicist - I don't pretend to understand every intricacy of climate science though. But that all model output is invalid because of software bugs seems (again!) improbable to me. The models deliver results with a limited precision. But their imprecision is unknown and may lie in either direction. So for me their results represent the center of probability according to the information available. All the more as they are in line with the common sense argument, that if you put in more energy than you let get out, the thing becomes warmer. (- Yes, I know, it's the degree of warming that is disputed. If you want to estimate the degree, you start calculating, and end up at - surprise - numerical models.)
    And a global AGW conspiracy - seems (again probability!) not very probable to me.
  25. Does anyone have any comments on the following paper?

    R. Fildes and N. Kourentzes, Validation and forecasting accuracy in models of climate change, Working Paper
  26. Forecasters at it again. However, I see you asked over at Realclimate as well (good idea), so it's worth noting Gavin's response:

    "Actually it isn't that terrible. They clearly spent more time trying to understand the science than previous forecasting researchers and they do a reasonably constructive job of trying to see whether you can improve on climate model projections. They slip up a little in mixing up decadal intialized predictions with the wider climate model enterprise but it is a reasonable first effort."
  27. Scott I had a look at it a while back and was not impressed by it. As Gavin's comment at RealClimate suggests the main problem is that they didn't spend enough time learning about the climatology or the way in which models operate and are used. This meant they ended up doing things like evaluating model predictions on station level data. It is well known that models can't be expected to do so (as the climate at station level depends on local geography on a scale much smaller than the typical grid box of a GCM) and in practice modellers use downscaling methods. So they are criticisng the use of models for something models are never actually used. I rather doubt their statistical methodology is robust either, but I can't remember the details off-hand, I'll have a look for my earlier post.

    The thing that I found irritating though was the constant mention of the IPCC, when none of the issues raised in the paper had any bearing on what the IPCC have already done, just on what they are planning to do for the next WG1 report. However reading the paper you might think it somehow casts doubt on the accuracy of existing reports.

    There has been at least one comment paper submitted to the journal. I was thinking of submitting one myself on the statistical aspects, but I just don't have the time to run the simulations etc.

    In essence, it was a rather poor paper, that shouldn't have been published in its final form, IMHO.
  28. This page could be improved a lot by considering at least one global circulation model. Pick one that's referenced in IPCC AR 4, describe the initialization state, describe the data series being fed as inputs, show the output over the time period between when the model was written and the present, then compare its predictions to predictions of less complex models like a time series.

    I actually think a lot of folks who come to this page will be surprised by the lack of such a comparison.
    Response: [Dikran Marsupial] You can probably find most of this at climateprediction.net, although you would have to add the time series models yourself. I am not sure there is much point in comparison with the time series models, for a start would they give coherent spatial predictions? This sort of comparison has been done, and the model runs etc archived (search for CMIP3). However writing a good blog article on this would be a huge amount of work and beyond my capabilities (and I have worked with the models!).
  29. 382, rcglinski,

    I'm not sure that anyone could do what you ask except for a primary researcher working with the model in question, and then I think that compiling and presenting that information is a fair amount to ask of that person, even with their familiarity.

    I also think the models are so complex that the result may be many, many pages long, and require a lot of supporting information.

    All in all... I think (I could be wrong) that your request is way, way out of line with what is reasonable. Anyone who has that level of interest needs to go to the models and look at them themselves.

    Beyond that, there is a huge, huge wealth of information below these comments, under Notes, including links to papers, pages about modeling, and blog articles.

    Maybe a list of links to the pages for the GCMs, as well as those that can be downloaded and run by the really ambitious, should be added to that, but again... if you're that involved that you're going to do that much work, then you can use Google to find them.
    Response:

    [DB] Bob, I don't believe the Notes section is viewable.  If there's a relevant section in it, you may want to post it in a comment.

  30. For Anthony Mills - you might like to consider that you can make useful prediction with even very simple models. eg Broecker's model

    However, it is also fair to say that climate models have no skill at short period prediction, not even decadal, for reasons that include your concerns.
  31. And yet another thought - if we were getting an extra 2W/m2 of solar radiation (twice the range of the 11 year solar cycle), would you say that models were incapable of predicting that this would warm the planet?
  32. "The Conversation" has a sceptic asking the following. Anyone got an answer?

    Doug Cotton starts with this...
    < quote >
    Speaking of error bars, Michael, in this paper http://www.agu.org/pubs/crossref/2004.../2003JD004457.shtml Zhang et al claim to "reduce the overall uncertainties from 10–15 to 5–10 W/m2 at TOA and from 20–25 to 10–15 W/m2 at SRF."

    Now, empirical data appears to suggest that the difference between upwelling and downwelling flux at TOA tends to range between about plus and minus 0.5% of total incoming insolation.

    It seems to me that those error bars are, in most cases, too large to even confirm with certainty whether the net flux is positive or negative, that is, whether we should expect warming or cooling.
    < end quote >

    Then claims this:

    < quote >
    Doug Cotton
    Doug Cotton

    Maths and Physics university tutor

    Score: +1

    insightful +
    report abuse • unconstructive -

    I don't think that is really the case Michael. The post below was just deleted within an hour from today's thread about media reporting.

    Yes, I agree there is a "consensus among scientists" but history is full of examples when "the science" of the day, or "the medicine" of the day has ultimately been proven wrong. Majorities are not always right. So why should the media block out those who put forward legitimate questions about the science? Without such questioning (which should not be equated with scepticism) nothing will ever change in this world and errors will be perpetuated.

    There are four main areas, as I see it, which need careful consideration in the climate debate and which I, for one, question:

    (1) The degree of accuracy in the models used by the IPCC and others is simply not sufficient to prove that warming should be happening. In general, they determine the difference between downwelling and upwelling radiation and it is only this difference which (depending on its sign) indicates warming or cooling. But the difference in fact is rarely more than 0.5% (plus or minus) of the total incoming solar insolation. The error in the figures used to determine such a difference is greater than that. Hence there is no valid proof from the models than we have +0.5% rather than, say, -0.5%.

    (2) An as yet unpublished study of temperature data from hundreds of boreholes (being prepared by myself and colleagues) reveals that there is a very strong correlation between surface temperatures and the temperatures determined by an extrapolation of the underground temperature plot determined only from measurements more than 200 metres underground which are well beyond the influence of solar insolation. The probability of this happening at random is absolutely infinitesimal. Hence we can deduce that underground temperatures supported by heat flowing out from the core are the forcing factors, rather than any processes relating to solar insolation or gases in the atmosphere.

    (3) Trenberth's trend (shown at the top here http://climate-change-theory.com ) shows a curved line now past its maximum and starting to decline. Adding data to 31 Aug 2011 shows that downward trend continuing. (Sea surface temperatures (using highly accurate NASA data) are probably the best indicator because about 90% of heat above the crust is stored in the oceans and sea ice.) The gradient of this curved trend is now statistically significantly different from the lowest gradient of IPCC projections. This proves such projections incorrect with 99% probability.

    (4) The role of greenhouse gases in radiating away heat obtained by collision with non-greenhouse molecules would not appear to be considered in the models. Oxygen in particular is a very stable molecule which radiates very little itself at atmospheric temperatures. Nitrogen is a close second. Quantum mechanics shows why molecules can only radiate the frequencies (wavelengths) which they can also absorb.
    < end quote >


    And this:
    < quote>
    "Yes, we know IR radiation is captured by GH gas molecules, and further photons are then re-emitted. The emission of even more photons takes place as the GH gas molecules (including CO2 of course) cool off. Some of the radiation goes back to Earth, then heats the Earth and more conduction and radiation occurs as a result. Heat is carried upwards partly by radiation and partly by physical movement of molecules - ie convection. Eventually, between 99.5% and 100.5% of all incoming solar insolation is radiated to space. So yes, GH gas can delay the process by a few minutes, maybe an hour or two, but it can also speed up the process of cooling 98% of air molecules. Who knows which dominates? By night nearly all heat will escape, except in local summer when the oceans will warm, but lose their extra heat again by winter. The models appear to "overlook" the above potential cooling effect. But, even if I'm wrong on that (and someone else show me where) the models are still not accurate enough to be able to determine whether it is 99.5% or 100.5% and that makes all the difference between warming and cooling. The world has been misled by bad statistical accounting for margins of error. You simply can't take a difference of two numbers each over 300 (with errors about plus or minus 5) and prove that the difference is +1 rather than -1 for example. "
    < end quote >

    Anyone got any papers on this?
    Response:

    [DB] Eclipse, Mr. Cotton feels his own special pet hypothesis, using as-yet-undiscovered physics, proves that the Earth is warm due to heat escaping from it's core. 

    Thus, he (Mr. Cotton) is right despite a lack of any published studies to support his position.  And therefore hundreds of years of research by hundreds of thousands of scientists is wrong, despite an overwhelming amount of physical evidence to the contrary.

    Dialogue with Mr. Cotton is thus impossible, as science says that 2+2=4.  Mr. Cotton says it equals 16 and also that on Tuesdays water flows uphill, the sky is green and the Moon is actually made of cheese (Edam, I believe...as we could smell it from here if it was Limburger).

  33. This was my (rather embarrassing and ill informed) tentative reply.

    ***
    I'm no scientist Doug, but isn't it the case that thermodynamics states heat energy can't move from a hot source to a hotter source? Doesn't heat always travel 'down' temperature grades?

    If greenhouse gases are emitting IR heat that they've prevented from radiating out into space, then surely the GHG's are the ones warming the other gases, not the other way around?
    Response:

    [DB] "isn't it the case that thermodynamics states heat energy can't move from a hot source to a hotter source?"

    The 2nd Law refers to Net heat exchange...

    "Doesn't heat always travel 'down' temperature grades?"

    For convection, yes.  Back radiation is 360 degrees...

  34. DB inline @386, if only Doug Cotton where so succinct, and so sane in his theories.
  35. Eclipse @386:

    1) Doug Cotton is certainly not in a position to judge this (see point 4), and indeed is completely wrong. Specifically, climate models very accurately predict the change in radiative forcing due to changes in greenhouse gas levels.



    To give you an idea how accurate line by line models are in those predictions, here is a comparison of model data (dotted line) and observed data (solid line) over the Gulf of Mexico:



    Global circulation models are not quite that exact, but more than exact enough to narrow the expected temperature rise per doubling of CO2 to the range of 1.5 to 4.5 degrees C. Copious physical evidence from diverse sources narrows that still further to 2 to 4.5 degrees C, with a most probable value of 3 degrees C.

    2) An as yet unpublished paper that refutes a well established theory is worth no more than the paper it is currently printed on. In fact,from clues about the contents of that paper Cotton has left on this website, the paper is not only unpublished but unpublishable as it shows no knowledge of basic physical laws, including those relating to thermal conduction (the theory on which the paper is supposedly grounded), but also directly contradicts well determined measures of heat flux from the core to the surface. (I have linked you to my posts, but other posters have equally effectively rebutted Cotton's nonsense.)

    3) Cotton just made that 99% figure up. In fact, the reduced rate of global warming is not unexpected given high aerosol emissions by China, a recent solar minimum lower than anything since 1910, and three strong La Nina's in just four years. But China is curtailing its emissions, La Nina's come and go, and the Sun is now well into its next solar cycle so expectations of anything but renewed warming are just wishful thinking.

    4) Curiously I am responsible for this belief by Cotton. He came on this site saying the majority of thermal emissions from the Earth's atmosphere were from oxygen and nitrogen. This is an absurd falsity. When I demonstrated to him that he was wrong, he without pause or consideration switched to this new theory. He had just made a massive change in his theory but it made no difference at all to his conclusion, ie, that global warming is false. It is safe to conclude that Cotton want's to retain that conclusion, and no near detail of fact or logic will be allowed to prevent him from doing so. He is one of those unfortunate people of whom it can be said that he is always in error, but never in doubt.
  36. Thank you both. I've copied your replies back to the "The Conversation" and have bookmarked this thread for further reading. Great job!
  37. Say I wanted to play around with the data, to look for relationships in a pretty amateur way.
    There's:
    Sea Level
    Surface/Air temp
    C02 ppm
    Sunspot number
    SOI
    What am I missing and where will such a simple model fail?
  38. #391: watch out for cause-effect relationships between different factors, but you'll also want volcanic forcing in there, and perhaps aerosol forcing too. With a reasonably simple model you can reconstruct surface temperature changes using CO2, SOI, volcanic - see Tamino's Open mind for some examples, one here and a better (superb) one here. I'm not sure about sea level, though a recent post on this site speaks of the impact of ENSO on short-term variations.

    Simple models don't capture the complexities of the interactions in each system, but they have their value in identifying some of the key elements to a system.
  39. Thanks a bunch Sky :)
  40. Tristan if you want to play with simple phenomenological model with say multiple regression, then have a look at Benestad and Schmidt 2009 for an example.

    Sunspot number is very crude - you should use one of latest TSI proxy constructions (eg look at here.

    Very importantly though, you need an aerosol term, both industrial and volcano. For internal variability, you should include an ENSO index.
  41. This is an interesting abstract. Does anyone here have a journal handy to read it and know if it is worth the subscription rental?

    Climate models don't show the warming in the early 20th century
  42. Camburn @395, I cannot comment on the paper, but I can comment on your misrepresentation of the abstract. The relevant sentence, just one item out of many discussed is:

    "Few models reproduce the strong observed warming trend from 1918 to 1940. The simulated trend is too low, particularly in the tropics, even allowing for internal variability, suggesting there is too little positive forcing or too much negative forcing in the models at this time."


    There is a very large difference between the claim that "Climate models don't show the warming in the early 20th Century" and the actual claim in the abstract that the warming shown by most models is not as great as that observed. There is also a difference between your blanket "Climate models" (indicating all Climate models) and the abstracts concession that a few models do in fact show the correct trend.

    It is difficult to not believe that your misrepresentation of the contents of the abstract is deliberate.

    Further, your choice of just one sentence to highlight out of the abstract also shows bias. Why not, for example, discuss this sentence:

    "Over the whole of the 20th century, the feedback strength is likely to be underestimated by the multimodel mean."


    The answer, I am sure, is that you do not want people thinking about the possibility that climate sensitivity is more than that which the models indicate.
  43. Tom:
    The hindsight of the models do a poor job replicating the temp pattern in the early 20th century.

    That is obvious from the abstract.

    As to my question, is the paper worth paying the rental fee for? Someone may be able to access this and give an opinion. Abstracts are a hint, but the meat of an issue is in the paper itself.
  44. Tom:
    As far as feedback, that is in question. The statement of the abstract that the models do not do a good job of hindcast is a fact.
  45. 397, Camburn,

    Lame response. You've been caught red-handed misrepresenting a paper abstract to try to imply doubt about climate science. Then you compound the error by acting as if your misrepresentation is still a reasonable interpretation.

    This is typical denialism, laid bare for anyone with half a brain to look at and recognize. Thank you for the demonstration.
  46. Camburn#398: "The statement of the abstract that the models do not do a good job of hindcast is a fact. "

    Of what use is a selectively chosen, isolated fact, without context or mechanism?

    This paper calls for higher positive feedback; Camburn is on record siding with Spencer on the side of low feedback and therefore low sensitivity. I would think you'd be running away from this paper as fast as possible - if you agree with it, you are contradicting your support of Spencer and tacitly siding with Dessler. Surely that's not your intent?
  47. Camburn @397, the performance of the models against the early 20th century has been known for a long time:



    As can be seen, the trend of the observed temperature changes in the early twentieth century is very close to the modeled temperature changes. Exceptions can be seen in 1909-10 and 1915-17 when the observed temperatures are significantly below the modeled temperatures, but both of those periods coincide with < ahref="http://www.bom.gov.au/climate/current/soihtm1.shtml">strong La Nina years (exceptionally so in 1917). A further exception can be found in the period 1938-1945 in which the observed temperatures lie well above the modeled line. This is partially explained by a strong El Nino in 1940-41.

    The unexplained increase represents approximately 10% of the increase in temperature between the 1910's and the 1940's. It may well be explained by a dip in anthropogenic sulfates at the time, or indeed by a sudden influx of black carbon aerosols. Regardless, trying to interpret an approx 10% at one point as "Climate models don't show the warming in the early 20th Century" (my emphasis) is bizzare. Your statement was both unequivocal and wrong. Your follow up that "models do a poor job replicating the temp pattern" seems to come down to this - Climate models do a poor job at retrodicting the exact year of ENSO fluctuations (as opposed to their frequency), and the onset of wars and depressions.

    Well, your probably right on that, but I don't think a failure to predict WWII (or the exact amount of black carbon released by the blitz) constitutes a serious problem for climate modelers.
  48. muoncounter:
    So you have read the paper? Does it provide anything new that is worthy of paying the rental fee?
  49. I have been corresponding with an ex-engineer with regards to skepticism regarding anthropogenic climate change. Here is one of his remarks.

    "We certainly do not have the computing power to perform the ANOVA for the effect of orbit/tilt, solar output, geological processes, etc. on GTA. That makes attempts to correlate recent measurements of GTA with its inputs otiose."

    Basically, he believes that we are applying way too much certainty with regards to CO2, and understating our uncertainty with regards to other variables. I can debunk this at a qualitative level, but not quantitatively because I truly have no expertise in computer science. Can anybody help me out here, or is he right?
  50. chuidburg The answer is yes and no. We do have enough computer power to run the simulations that demonstrate quite unequivocally that anthropogenic climate change is real and that natural forcings are unable to explain much of the observed warming. However, our characterisation of the uncertainties involved will continue to improve the more experiments we perform. So whether you think there has been enough depends on where you put the goalposts (which your friends has been extremly vague about).

    I suggest you challenge your friend to read chapters 8 and 9 of the most recent IPCC WG1 report and make a specific suggestion of an experiment that they haven't pursued, that relates to a scenario or theory that is plausible and has some support from observational evidence.

    Ask him to specify exactly what simulations would need to be run to perform the ANOVA to his satisfaction (an ANOVA is not the right tool anyway, as correlation is not causation, but that is another matter).

Prev  1  2  3  4  5  6  7  8  

Post a Comment

Political, off-topic or ad hominem comments will be deleted. Comments Policy...

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

Link to this page



The Consensus Project Website

TEXTBOOK

THE ESCALATOR

(free to republish)

THE DEBUNKING HANDBOOK

BOOK NOW AVAILABLE

The Scientific Guide to
Global Warming Skepticism

Smartphone Apps

iPhone
Android
Nokia

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