RealClimate logo

Musings about models

Filed under: — gavin @ 20 August 2007

With the blogosphere all a-flutter with discussions of hundredths of degrees adjustments to the surface temperature record, you probably missed a couple of actually interesting stories last week.

Tipping points

Oft-discussed and frequently abused, tipping points are very rarely actually defined. Tim Lenton does a good job in this recent article. A tipping ‘element’ for climate purposes is defined as

The parameters controlling the system can be transparently combined into a single control, and there exists a critical value of this control from which a small perturbation leads to a qualitative change in a crucial feature of the system, after some observation time.

and the examples that he thinks have the potential to be large scale tipping elements are: Arctic sea-ice, a reorganisation of the Atlantic thermohaline circulation, melt of the Greenland or West Antarctic Ice Sheets, dieback of the Amazon rainforest, a greening of the Sahara, Indian summer monsoon collapse, boreal forest dieback and ocean methane hydrates.

To that list, we’d probably add any number of ecosystems where small changes can have cascading effects – such as fisheries. It’s interesting to note that most of these elements include physics that modellers are least confident about – hydrology, ice sheets and vegetation dynamics.

Prediction vs. Projections

As we discussed recently in connection with climate ‘forecasting‘, the kinds of simulations used in AR4 are all ‘projections’ i.e. runs that attempt to estimate the forced response of the climate to emission changes, but that don’t attempt to estimate the trajectory of the unforced ‘weather’. As we mentioned briefly, that leads to a ‘sweet spot’ for forecasting of a couple of decades into the future where the initial condition uncertainty dies away, but the uncertainty in the emission scenario is not yet so large as to be dominating. Last week there was a paper by Smith and colleagues in Science that tried to fill in those early years, using a model that initialises the heat content from the upper ocean – with the idea that the structure of those anomalies control the ‘weather’ progression over the next few years.

They find that their initialisation makes a difference for a about a decade, but that at longer timescales the results look like the standard projections (i.e. 0.2 to 0.3ºC per decade warming). One big caveat is that they aren’t able to predict El Niño events, and since they account for a great deal of the interannual global temperature anomaly, that is a limitation. Nonetheless, this is a good step forward and people should be looking out for whether their predictions – for a plateau until 2009 and then a big ramp up – materialise over the next few years.

Model ensembles as probabilities

A rather esoteric point of discussion concerning ‘Bayesian priors’ got a mainstream outing this week in the Economist. The very narrow point in question is to what extent model ensembles are probability distributions. i.e. if only 10% of models show a particular behaviour, does this mean that the likelihood of this happening is 10%?

The answer is no. The other 90% could all be missing some key piece of physics.

However, there has been a bit of confusion generated though through the work of – the multi-thousand member perturbed parameter ensembles that, notoriously, suggested that climate sensitivity could be as high as 11 ºC in a paper a couple of years back. The very specific issue is whether the histograms generated through that process could be considered a probability distribution function or not. (‘Not’ is the correct answer).

The point in the Economist article is that one can demonstrate that very clearly by changing the variables you are perturbing (in the example they use an inverse). If you evenly sample X, or evenly sample 1/X (or any other function of X) you will get a different distribution of results. Then instead of (in one case) getting 10% of models runs to show behaviour X, now maybe 30% of models will. And all this is completely independent of any change to the physics.

My only complaint about the Economist piece is the conclusion that, because of this inherent ambiguity, dealing with it becomes a ‘logistical nightmare’ – that’s is incorrect. What should happen is that people should stop trying to think that counting finite samples of model ensembles can give a probability. Nothing else changes.

167 Responses to “Musings about models”

  1. 1
    Glen Fergus says:

    All good. So how do we get to realistic probability distributions of outcomes? They are needed…

  2. 2
    Osvaldo Lucas says:

    Are there any model predictions (simulations?) for temperatures round 1800 and before?

  3. 3
    Lance Olsen says:

    Do we ask too much of models?

    I think it’s fair to say that we’ll never see a perfect one except the planet system itself. That doesn’t mean or imply that models are not helpful, or needed. They may, however, never meet all the demands that have been placed on them, including by the debunkers who want all action delayed until the perfect model or ensemble is at hand.

    Actually, despite limits demonstrated for accuracy of rate and extent of change, the models deserve some cheers for pointing in the direction we’re headed. And that should be warning enough to set us on a course correction.

  4. 4
    Aaron Lewis says:

    Re 1
    The realistic, probability-distribution models will come after we have made our decisions the old-fashioned way – on the basis on incomplete and mostly wrong information.

    There is so much social inertia and so many lags and delays in climate systems that we may already have made critical decisions; and not yet realize that we have already past those decision points.

    Here, I would cite Beyond the Limits by Meadows, Meadows, and Randers, & earlier works by Jay Forester, Meadows, and Meadows going back into the Club of Rome Report and Limits to Growth.

    (Dynamo was not a good language to program in, but it was better than Formula Translation Language.)

  5. 5
    Forrest Curo says:

    His definition adds unnecessary confusion and misses the main point of a “tipping point”, which is that it’s a value where the system you’re describing abruptly shifts from negative feedback to positive in some range of conditions.

    If you have a normal thermostat, for example, the heater goes on when the temperature falls and turns off as it rises (“negative feedback”, a feature which tends to keep things stable.) If there were a bad design flaw in a thermostat, such that temperatures over 100 F turned the furnace on, that room’s climate would have a tipping point ~100 F and the room would get hard to cool whenever temperatures rose over that (“positive feedback,” at least for awhile.) It wouldn’t necessarily get hotter & hotter until the place burned down; the furnace would have limits & eventually a cold enough night would bring things back to normal, but that 100 F mark would be well worth avoiding…

  6. 6
    Ike Solem says:

    Regarding the Sahara and the Sahel, this news article addresses that in some depth, and even provides a good example of the model ensemble business, and why a ‘model outlier’ might actually be correct.

    In particular, Isaac Held’s group produced model results that indicated that the Sahel wouldn’t be greening but instead would be facing a period of continued drying. They also have a more recent follow-up paper with a statistical approach.

    On the other hand,
    “Last year (2005) US-based researchers Martin Hoerling and James Hurrell looked at all of the most recent climate models, averaged them out, and came to the conclusion that the Sahel’s recent fate would be reversed in the 21st century.” (Note that their analysis predicts increased drought for southern Africa, however).

    The article also finds yet another way to bring tipping into the discussion:
    “To complicate matters, the relative importance of factors affecting the Sahel’s climate is tipped in different directions by the different models.”:)

    Unfortunately, this is a topic that sites like Sherwood Idso’s Western Fuels Association-linked CO2science have jumped on (guess which outcome they’re promoting: greening or drying?). That site seems to do more deliberate distortion of scientific publications than any other.

    Given all this uncertainty, what should policy makers do? There must be something in the ‘policy makers guidebook’ that relates to serious uncertainty – something like : hedge your bets and start some contingency planning for the worst-possible-outcome scenario – just to be on the safe side. There are some more obvious steps – since deforestation in the region has played at least some role, one goal should be to do everything possible to halt and reverse that trend (saving forests appears to be more important than producing biofuels, as well).

  7. 7
    Nigel Williams says:

    Very interesting article by Tim Lenton on tipping points. I note that Tim sees the loss of the Greenland Ice Sheet as the most likely Tipper – giving us 7m sea level rise over the next 300 years or so.

    While he discusses the interaction between the GIS and the WAIS, in his summing up he does not tie the loss of GIS to the WAIS in the way he does in the body of his report. But it seems he is saying that it is most unlikely we will be able to avoid the loss of the GIS (+7m rise), and he means that we will then loose WAIS (4-6m rise) as well – the two will be trundling into the ocean together, as we frantically withdraw our civilisation from the coast.

    With the prospect of 11 to 13m rise over 300 years, it makes well over +1m rise this century seem very very likely, doesnt it? (13m/300 years comes to over 4m/100 years on average. I know it wont be linear, but its going to go with a hiss and a roar, eh!)

    And isnt this well above the politically-correct IPCC predictions?

  8. 8
    Neil B. says:

    The skeptics of GW and AGW often say: processes in nature always try to return to equilibrium instead of runaway amplification, so nature will be take care of it. That is simply false. Look at gravitational collapse, stability of objects in flight unless specially made with canard wings etc., the rust process (more rust makes rusting even easier by opening up surface pits) etc.

  9. 9
    Daniel C. Goodwin says:

    In message 7 above, Nigel Williams expresses something of the cognitive dissonance I’ve felt in trying to sink my teeth into this issue: who is addressing the disconnect between IPCC projections and ice-sheet observations? Luckily for lost, confused souls such as myself, James Hansen (who is a great guru not only among scientists but also among philosophers) has done a fine job, in my humble opinion.

    This review, entitled “Climate Change and Trace Gases” cuts to the chase with remarkable frankness. The “Albedo Flip” Hansen et al warn of is, substantially, what’s going on now (a decade or four ahead of schedule) in the Arctic. So there we are, I suppose.

    But, you say, this article only demonstrates Hansen’s singular genius in his chosen field of climatology. What of my claim that Hansen is also a philosopher of unique insight?

    This paper, called “Scientific Reticence and Sea-Level Rise” is one of the more thoughtful treatments you’re likely to see of the problem: why is a large, consensus organization (such as the IPCC) more likely to underestimate the scope of future disasters?

  10. 10

    There are some important distinctions between Bayesian and Frequentists statistics. What I find refreshing about casting an answer to a question in terms of Bayesian statistics is that one’s assumptions are explicitly part of the answer. Some questions are better answered without pretending there is an absolute probability of being right or wrong. For the problem of climate model parametric uncertainties, one should regard the resulting Bayesian probability as a relative likelihood given the parameters considered and one’s prior assumptions. If one wanted to express no particular preference for a prior for a parameter x, a uniform probability may suffice. However, if the parameter were 1/x then a more appropriate prior (that expresses no particular preference) should be a Gamma distribution. Climate model ensembles can be constructed to reflect exactly what we mean to express in terms of sources of uncertainty.

  11. 11
    Edward Greisch says:

    The problem is genetic. The average human is just not good enough in math to survive the next 200 years. My browser can’t read the Economist web site, but my guess is, you shouldn’t expect accountants to do math. The problem with democracy is, we all get the government the average deserves. The average may “deserve” extinction, or “deserve” is a bum concept.
    I think I know why ET hasn’t arrived yet. His planet died because of global warming before he could get off of it. The end-Permian extinction made it into NASA’s Astrobiology zine. See:
    The Astrobio article refers to the end-Permian as a carbon cycle problem.
    The points Gavin makes are mathematical and way beyond the comprehension of not just the average, but all but the top 0.1%. Even the top 0.1% can’t understand it unless they received the proper training. Gavin is talking to physicists and mathematicians but publishing to the world. Everybody needs to understand what he says. This is an impossible bind. We need to do a million years worth of evolution in the next 5 years.

  12. 12

    Hi Gavin,

    I certainly don’t want to belittle anyones models but I had an interesting conversation last week in California. A computer modeler was explaining to me how models can be wrong. I acknowledged that he is right and computer models can be wrong.

    Then I said, forget all the computer models. They are becoming increasingly irrelevant to the argument of global warming. They are important for understanding things and predictive quality but it’s getting warm outside. Everyone feels it. Everyone says hey this weather seems quite different than when I was young.

    For myself, I remember in Big Bear Lake when they could ice skate on the lake in winter and skiing went until July 4th (that was only around 37 years ago).

    People are arguing about whether the glaciers are melting. Well, the pictures don’t lie.

    When I’m in Switzerland I visit glaciers and its interesting that all these spots have picture of how big the glaciers were before. The amount of shrinkage is striking.

    The fact is that we are getting more Category 5 Tropical Cyclones than ever before. I’ve been watching the SST’s lately and hearing everyone talk about how were not going to get to many hurricanes this year. But it is well known that sea surface temperature drives hurricanes. It takes temps over 28.5 C to get a hurricane but the temps in the gulf today are showing way above that averaging near 30 C.

    As far as the argument goes, I think we are getting beyond models. The sad thing is that if the exponent of acceleration continues as it seems to be doing in the real time data. When the policy makers wake up we will be facing the front of a wave that will seem overwhelming and it will be.

    I continue to hope that we wake up faster as it no longer seems a matter of if it’s too late but rather that we are going to be in damage control for some time to come and all that that implies. I have this message for those that are questioning the veracity of reason or, by some fluke of neuro-chemistry, still doubting that global warming is real. Good luck!

  13. 13
    Bill Tarver says:

    “And isnt this well above the politically-correct IPCC predictions?”

    I’m growing increasingly sceptical about the IPCC’s conclusions wrt sea level rise. Hansen et al think there is good evidence for a 7m rise by the end of the century unless urgent action is taken. Given that the IPCC’s report doesn’t factor in Greenland or the West Antarctic ice sheet owing to their complexity, Hansen’s report seems far more convincing – unfortunately. I’d be very interested to see a response from one of the climatalogists.

    James Hansen et al, 2007. Climate Change and Trace Gases. Philiosophical Transactions of the Royal Society – A. Vol 365, pp 1925-1954. doi: 10.1098/rsta.2007.2052.

  14. 14
    Nick Odoni says:

    Sorry I have not had time to read all the reactions above, and apologies in advance if this repeats anything. However, I think a lot can be achieved by noting the errors in our models (however calibrated), and ‘bootstrapping’ our forecasts. Thus we can include both our parameter (and indeed any other ratio factor) uncertainties and our calibration uncertainty, building ensembles of results for individual models, and hence ensembles of ensembles etc. There are lots of ways of dealing with being cautious with the prediction errors (include jackknifing with the bootstrap, for example – see Davison and Hinkley, 1997) and establishing the extent of equifinality (or convergence if you prefer), and hence to begin considering how much is model artefact, how much is error, and how much is at heart a forecast that we need to take notice of. Nick O.

  15. 15
    cce says:

    Thanks for the article. Very interesting!

  16. 16
    pete best says:

    With regard to Tipping Points, is the 300 year timeline at 382 ppm of CO2 or the project(BAU)scenario of 550 ppm. Is there enough CO2 currently in the atmosphere for Greenland and WAIS to experience this non linear phenomena ?

    Politically this 300 year time line is a joke.

  17. 17

    For feedbacks capable of becoming tipping points, don’t forget greenhouse gases in melting permafrost.

  18. 18
    Eric (skeptic) says:

    I noticed the ‘tipping points’ were all localized. Calling them “large scale” is meaningless, they are either local or global. Often the tipping point proponents here will point to such local phenomenon and imply that the whole earth will get warmer. What they forget is there are negative feedbacks elsewhere, particularly in the weather in a wetter world. Tim pointed one out in the Sahara. That does not mean I believe that Gaia will magically regulate the climate, far from it. There will be local tipping points particularly at the poles where CO2 has the greatest warming effect. For the earth as a whole, doubtful.

  19. 19
    Nigel Williams says:

    Daniel, thank you for those pointers. I had read the first paper, and the one on Reticence is entirely pertinent. Im a transportation planner by trade, and I try and get other engineers to move towards using more sustainable modes of transport. Regrettably not too many are prepared to recommend to their clients that they use horses and carts to build their next skyscraper or dam.

    Everything I have read of Hansen et al lately, and everything that the community is reporting in relation to our declining global climate’s critical parameters points to the same point: A very wet year 2100 for a huge part of humanity.

    I notice that even Hansen is prone to ‘reticence’ as he tells us things will be fine if we sequester virtually all of virtually every greenhouse gas immediately. I think it was in the movie The Princess Bride where – presented with a total load of lies about his future (It was going to be bad) the hero said to his antagonists “We are all gentlemen here..” making it clear that he knew rubbish when he heard it, and thy all knew the truth.

    Hansen has made the same observation among his peers. We ARE all gentlemen here. Frankly the time for being polite is past. So what we loose some funding, or loose out jobs if we speak out. Perhaps the more practice we get making do with less will stand us in good stead before too long!.

  20. 20
    Doug Lowthian says:

    As always a cogent explanation followed by a rousing and intelligent discussion. Exactly why Real Climate is the best climate site around. Slightly off topic, news of this article “Heat Capacity, Time Constant, and Sensitivity of Earth’s Climate System,” by Brookhaven National Lab scientist Stephen Swartz soon to appear in the journal of Geophysical Research is already making the rounds on the skeptic blogs as the final nail in the coffin for AGW. No one has been able to tell me what it actually says though. Any insights from this learned crowd?

  21. 21
    Doug Lowthian says:

    Re: my previous post: The paper in question is available in a preprint at
    Thanks for any comments. My skeptical friends are crowing that this kills the AGW theory. I’m not so sure..

    [Response: You want – William]

  22. 22

    Re #18 where Eric the Skeptic says:

    There will be local tipping points particularly at the poles where CO2 has the greatest warming effect. For the earth as a whole, doubtful.

    The problem is that the global climate is inter-connected. The atmosphere is like a water bed. If you increase the pressure in one place it will pop up somewhere else.

    When the Arctic summer ice disappears, the sea will get warmer and the air will contain more water vapour. This makes it lighter and it will start rising instead of falling in the polar vortex. As a result the Northern hemisphere circulation pattern will be disrupted, and that will overflow into the Southern Hemisphere. Changes in the Arctic air flow could result in changes to the Indian and east African monsoons, El Ninos, and the ITCZ.

    Of course all those changes might not be bad, but if it caused the west African Monsoon to restart and return the Sahara into a bread basket, how are we going to get the farmers and farm machinery from the new US Mid-West desert across to Africa quickly enough to prevent a famine?

  23. 23
    Nick Gotts says:

    RE #18 [I noticed the ‘tipping points’ were all localized.] Some of these “local” (but large-scale) events would have global effects. Among those mentioned in the article the Amazon die-off and the change in the amount of Arctic sea ice are the most obvious over the medium term: the first would lead to a large pulse of greenhouse gas emissions as trees burned and further emissions as soil carbon was released; the second to a change in albedo which would accelerate warming. The greening of the Sahara would presumably have a global effect in the opposite direction (though perhaps not if secondary effects are taken into account: Saharan dust is thought to have an important fertilising effect on the Atlantic and the Amazon). Although it falls outside the article’s definition of a “tipping point”, the release of methane from melting permafrost would have a global effect.

  24. 24
    dean says:

    Re #12

    What is the source for more Cat 5 hurricanes? the Hurricane center has denied any such claim and frankly, they’d know.

    As for computer models… for a decade people have been saying “the models are RIGHT”. now models don’t matter? why is that? because the models aren’t predicting the gloom and doom anymore?

    as far as the 10% comment in the original story goes, if 90% of the models predict no change and 10% predict devastation, then something is wrong somewhere. while science isn’t dependent on consensus, the peer review process does work by developing consensus among knowledgeable people. yes, the 90% may be wrong, but rarely is that the case. when it is the case, i’ll admit that it’s very visible (Galileo comes to mind)

  25. 25
    dean says:

    One thing I’ve noticed is that many of us use the oven analogy. One major flaw with this analogy is that our system isn’t temperature based, it’s time based. the oven’s on for a set amount of time. It makes a lot more sense to think of the world this way… it also is easier to envision the ‘tipping point’.

  26. 26
    Ray Ladbury says:

    A volcano like the Yellowstone eruption can be considered a point source, geologically. Yet it has global and long term effects.

    I haven’t had time to peruse the paper thoroughly, but it would appear to be quite full of questionable assumptions. The values for climate sensitivity and equilibrium response appear to contradict values determined from a variety of other lines of evidence. I would also question his treatment of climate relaxation using “fluctuations” about an equilibrium. Instead, we have a system disturbed from equilibrium by forcer that increases with time. He also conveniently ignores positive feedbacks (e.g. increased water vapor) and most of the actual physics of climate change. The goal of physics should be, we are told, to make a model as simple as it can be and no simpler. It appears he has gone overboard on simplification, so it is not surprising that he is getting absurd results. I’m kind of surprised this got published.

  27. 27
    sidd says:

    Doug Lowthian wrote at 21 Aug 2007,7:54 AM

    i looked at the paper and i see some simplifications:
    1)the outgoing longwave emission is stated to vary as the fourth power of the
    _surface_ temperature. Should this not be the temperature at the radiating surface at the top of the atmosphere ?
    2)there is an assumption of linearity made in eq. 5. i am not sure this will hold
    3)the para before eq 16. admits that there are significant changes in the heat content of the ocean, and posits that the only causes could be variation in planetary albedo or effective emissivity; this is not clear to me since there are other components to the climate system such as icesheets which can sink considerable amounts of heat as well.

    it appears at first reading that some possibly important physics is left out of this model. Perhaps some of our resident climate modellers would care to comment ?


  28. 28
    Bill Tarver says:

    “Re: my previous post: The paper in question is available in a preprint at
    Thanks for any comments. My skeptical friends are crowing that this kills the AGW theory. I’m not so sure..”

    I’m not so sure either. The paper never mentions feedbacks. He concludes that a doubling of CO2 will cause a temperature increase of ~1.1 degC and contrasts this with the IPCC report that gives an equivalent figure of 3 deg C. But didn’t that figure also take into account the several feedbacks?

  29. 29

    Re 18, where Eric says there will be local tipping at the poles,particularly at the poles,but doubtful for the whole Earth.

    Melting of Greenland and West Antarctic ice sheets will raise sea level globally. Melting of Greenland ice would dump large amounts of fresh water into the north Atlantic affecting the Atlantic thermohaline circulation pattern. As was pointed out earlier, the release of CH4 from melting permafrost would be distributed pretty evenly throughout the atmosphere and not remain local. A number of such area wide warming events,will have global consequences.

  30. 30
    tamino says:

    Re: #20, #21

    Yes I’ve seen Schwartz’s paper.

    He estimates the heat capacity of the global climate system, and uses historical temperature data to estimate the “characteristic timescale” of temperature change. Using these data, he claims to derive the climate sensitivity, getting a figure of 1.1 deg.C for a doubling of CO2. This is much less than the current “best estimate” of about 3 deg.C, and would imply that there is *not* any unfelt warming “still in the pipeline” from greenhouse gases we’ve already emitted.

    Even if his result is correct (which it isn’t), it by no means “kills AGW theory.” It simply changes the numbers. In fact, Schwartz’s analysis implicitly assumes that the planet *is* warming, and that the primary cause is man-made greenhouse gases.

    His work depends on (among other things) the assumption that there’s only *one* characteristic timescale for global climate. Physically this would seem to be impossible; each component of the climate system (atmosphere, ocean, cryosphere) will necessarily have a different characteristic timescale. Schwartz’s analysis depends on the fast response of the atmosphere to climate forcing, but quite ignores the slow response of oceans and other components.

    The really odd thing, for me, is that mathematically, the data themselves contradict his hypothesis of a single timescale for climate response. I believe I can show, beyond doubt, that the data he presents *reject* the idea of such a simple behavior, but Schwartz presents an argument (which I regard as invalid) that they confirm it.

    I’m considering writing it up for JGR, and also considering doing a blog post on the subject.

  31. 31
    paul says:

    What are some examples of climate tipping points from the past that demonstrate what actually happens when a tipping point is reached?

  32. 32
    SomeBeans says:

    Ooops, that’s a link to the original paper – James’ comments are actually here:

  33. 33
    Hank Roberts says:

    That’s the same value for climate sensitivity I’ve seen from the string theory physics site and from knowledgeable climate sites as well — it’s the number people get this way: calculated in the absence of any feedback, on the hypothetical twinning of each molecule of CO2 in the atmosphere to make two where there were one, instantly, and having nothing else happen.

    Nice in theory. Of course, if it were possible to cut the total amount of CO2 in the atmosphere by half without any other event occurring, this also would solve a lot of problems. Perhaps pulling on the strings in exactly the right way through the fifth dimension somehow would do it. Til then it’s just a hypothetical.

    The rest of the climate sensitivity above the about 1 degree C is tied to the obvious questions: where does the CO2 come from, and where does it go, and over what time period.

  34. 34
    A.C. says:

    As best I can read the tipping point defn. and comments (and I’m sure I haven’t quite got it), “crucial feature” 1) describes or defines a region of the global system that when subjected to a small change in the short run suffers a large change in the long run because of an abrupt shift in the sign of its feedback and 2) transfers the large change/long run effects to other regions, so as to significanly affect the larger global system.

    Having my brain insisting that every “crucial feature” be something I can point to on a world map is probably not helping me understand. And I have no idea what to make of “parameters controlling the system can be transparently combined into a single control.”

    Anyway, the whole reason I’m even interested in the defn. (which I assume is intended to be read by someone with technical knowledge of climate modeling jargon) is that I’ve started wondering about how climate models handle volcanoes or geology like the Yellowstone caldera–regularly emitting “point sources” that transfer their climate effects globally. And what (if anything) makes geological emissions from Yellowstone different from man-made emissions from Taiwan or L.A.?

  35. 35
    Lynn Vincentnathan says:

    So glad you’re getting around to talking about (I know you hate this :)) the Venus effect, runaway warming…(note, I did not say “permanent runaway warming”).

    The Venus effect is meant to be an analogy (of course, we all know Earth is not Venus), and analogies are only similar in some aspects and to some extent. For the totally unrefined layperson, it is a way to grasp an idea that GW may not be linear (which I sorta think most people are assuming & that we have plenty of time), but may zoom up to big harm, before the climate gets back to today’s temps 1000s of years from now.

    I think scientists prefer the term “hysteresis” for this, because that contains the meaning that the climate eventually levels off (after wiping out much of life) and cools back down again.

    But, I think most people would be more focused on the upswing in GW than its eventual retreat. Except the denialists, who’d probably claim, “Well, that wasn’t so bad, some 30,000 people survived 100,000 years later, and now are back to repopulating the earth with all that great fossil fuel from the anthropocene die out.”

  36. 36
    Ray Ladbury says:

    A dynamical system will respond differently to a point source than it will to a diffuse system. Likewise, a steady perturbation will have a different effect than a delta function in time. Think of it this way: In the vicinity of a Yellowstone type event, the local changes will be drastic, and hence there may be a significant difference in how efficiently these changes are transmitted to the global system–e.g. you may have a significant local temperature drop that results in rainshowers that take at least some of the dust and sulfates out of the air, etc.
    In addition, a volcano emits LOTS of sulfates, dust and other nasties in addition to the CO2. Finally, averaged over time, volcanos produce far fewer ghgs than do humans.

  37. 37
    Daniel C. Goodwin says:

    Regarding “Scientific Reticence” – Nigel Williams, in post 19 above, points out that Hansen et al engage in a bit of the same at the end of “Climate Change and Trace Gases,” where they conclude that sequestering emissions from biofuels is a means of drawing down atmospheric co2. This also strikes me as a bit pollyannish. It would work (if gradually) if the population of the earth were considerably smaller, I suppose. Nobody wants to be accused of being a doomsayer, and so reviews of this nature tend to search (if pathetically) for some ray of hope to grab onto, for some positive message to conclude with, no matter what the state of the evidence.

    Compounding the clumsiness of this obviously inadequate suggestion, the paper concludes on a puzzling footnote, with a whiff of cloying, desperate patriotism:

    The potential of these ‘amber waves of grain’ and coastal facilities for permanent underground storage ‘from sea to shining sea’ to help restore America’s technical prowess, moral authority and prestige, for the sake of our children and grandchildren, in the course of helping to solve the climate problem, has not escaped our attention.

    Just goes to show that even geniuses experience lapses of judgment from time to time…

  38. 38
    Deech56 says:

    RE #12 (John P Reisman): With all due respect, I would disagree with your view that we should forget the models. Yes, the models are imperfect, but that does not mean they are worthless. They are useful, first, to show the causes of the current warming – the comparison of outcomes for natural vs. anthropogenic forcinigs is a powerful argument for AGW. They are also useful to point to what the effects of AGW probably would be. The predictive power of the models does depend on checking them against what is actually happening – and this will be more important when the feedbacks start kicking in (but by then maybe we won’t need the models to tell us we are bad off). I am concerned that the deniers/skeptics use the imperfection of the multiple lines of evidence as a tool to argue against an anthropogenic cause for GW – I see this happening in the current arguments against the surface data. Yes, we can acknowledge the limitations of each line of evidence and of the models, but it does not help us if climatologists abandon these useful tools.

  39. 39
    quadszilla says:

    Re: 35

    Venus and run away global warming. Here’s a serious look at it:

    “Why is the albedo of Venus important? When the albedo is at 0.80, the Global Warming Theory falls apart. . .

    The carbon dioxide levels on Earth have risen from approximately 0.028% to 0.036% in the last few decades. It is a major stretch to compare this with Venus at a 96.500% carbon dioxide level and promote an uncontrollable runaway condition. Earth in its early history, 385 million years ago, had an atmosphere with 10 times the present carbon dioxide levels. Those elevated levels did not produce runaway global warming then, so why should we theorize that it would today?”

  40. 40
    Hank Roberts says:

    > cloying, desperate patriotism

    Er, no. Belated acceptance of responsibility for cleaning up the mess that we profited from, by profiting from being the first to come up with the solutions, so we can sell them overseas.

    He knows Americans want to do well by doing good.

  41. 41

    Re #24

    Hi Dean, while data in NHC, NOAA, NASA, NCAR and other .gov web sites is scattered, the NHC does have a nice paper on the subject including the following:

    Research has shown that the sea surface temperature (SST) alone does not provide a good indication of whether a storm will intensify. (See, for example the SST/Intensity relationships of recent Atlantic tropical cyclones.) However, SST does provide an upper limit to storm intensity. In SHIPS, the Maximum Possible Storm Intensity (MPI) is related to the SST by the equation:

    MPI = 55.6 kt + 108.5 kt exp[0.1813 * {SST – 30.0oC)]

    One of the strongest observed correlative facts is that we are getting more of those ‘rare’ cat. 5 hurricanes, and the water is warmer. Let me know if you have another good explanation for the increase of cat. 5 hurricanes. I’m all ears on that one!

    Also, could you please show me a link where they claim that their own research is incorrect on this matter. Maybe there is a contextual dilemma?

    “As for computer models…” What I said was: “They are becoming increasingly irrelevant to the argument of global warming. They are important for understanding things and predictive quality”

    A lot of the models that are tracking reality pretty well are being attacked because of minute errors. This of course is silly as the models are generally correct and indicative and therefore a good way to indicate or predict probability in relation to human caused global warming. Minute errors are corrected as discovered and do not detract from the overall picture as modeling and measuring improves.

    What I am saying is that it is pretty easy for some to get caught up in the minutia and maybe that is a reflection of individual thought process or bias of some nature, but that generally, it’s getting warmer and that is affecting the weather.

    More heat translates to more moisture in the air so more rainfall, more snow (melting faster though) likely larger climate momentums causing droughts and floods as well as debilitating storms and crop failures.

    This is all just common sense. The world is already experiencing these effects. It’s not like we need a model to see what is happening. Good models help us see what is coming and models are getting better all the time because lots more people are working on it.

  42. 42
    Hank Roberts says:

    Paul, you asked for examples of climate tipping points.

    Your library should have this book:

    If you don’t read the book, at least read the review.

  43. 43
    Aaron Lewis says:

    # 9
    I would like to thank Daniel for pointing out “Scientific Reticence and Sea-Level Rise”. I had not seen it.

  44. 44
    SecularAnimist says:

    Daniel C. Goodwin wrote: “Nobody wants to be accused of being a doomsayer”

    I don’t mind being accused of being a doomsayer. So I’ll say it:

    1. My take on the science is that the warming from the CO2 that humanity has already added to the atmosphere is probably more than sufficient to trigger multiple positive-feedback “tipping points” that will further increase and accelerate the direct anthropogenic warming — and we are already seeing evidence of these feedbacks kicking in; and

    2. Anthropogenic CO2 emissions are presently increasing every year at an accelerating rate, and it is extremely unlikely that humanity will collectively do what is necessary to not only stop that growth in CO2 emissions, but reverse it, and then reduce emissions by 80 percent or more within 5 to 10 years, which is what mainstream climate scientists say is needed to avoid the worst outcomes of anthropogenic global warming. (And per point 1 above, they are probably mistaken to believe that we have that much time anyway.) So,

    3. The rapid, accelerating and extreme anthropogenic warming of the Earth, reinforced by feedbacks, will not only cause the various “disaster scenarios” that will be catastrophic for human civilization (e.g. megadroughts that will wipe out most agriculture, loss of fresh water supplies for billions of people, inundation of heavily populated coastal zones including most of the world’s major cities) but will lead to a global ecological meltdown — a general biospheric collapse — and the mass extinction of most life on Earth.

    So, we are, in a word, doomed. There, I said it.

  45. 45
    Nigel Williams says:

    Daniel, I dont think the Amber Waves comment is an error of judgement, It is a genuine image of how good ot could be if only we did what we know we shuld. If someone (like a president) took that vision to the nation it may work, but I wouldnt hang up my umbrella yet.

  46. 46
    Richard Ordway says:

    [[They (computer models-RO) are important for understanding things and predictive quality but it’s getting warm outside. Everyone feels it. Everyone says hey this weather seems quite different than when I was young.]]

    There are really two other *scientific* ways to predict future global warming trends beside computer models.

    1) Physics.

    2) Paleoclimatology

    I would be really careful on saying that people can observe GW so it must be happening. Remember certain parts of the Earth have been getting cooler…and this is expected….but the average surface temps are going up taken from 10s of thousands of locations world-wide.

    I don’t mean to knock your comments down…but they are simply unscientific. They would not pass for a second in science peer-review journal.

  47. 47

    Re #24

    Hi Dean, here’s some more from NASA on the matter:

    Take Warm Water, Stir:

    Sea surface temperatures must be 82 degrees Fahrenheit (F) or warmer for tropical cyclone formation and sustenance.

  48. 48
    Chuck Booth says:

    Re # 37 Daniel C. Goodwin: “Compounding the clumsiness of this obviously inadequate suggestion, the paper concludes on a puzzling footnote, with a whiff of cloying, desperate patriotism…”

    They also appear to be mimicking the concluding statement of the famous Watson and Crick DNA double helix paper (Nature, 1953): “It has not escaped our notice that the specific mechanism we have postulated…” (

  49. 49

    re #13

    Hi Bill, just a small note on the IPCC. It is a very large process and typically lags the leading edge of new data and modeling in order to process the large scope of data collected. It’s also in Switzerland and its a fairly conservative place. I know because ‘m over there a lot these days.

    The IPCC will lag but lay down arguments that will be harder to argue with. In my opinion, that is helpful because it keeps furthering the leading edge of the research and giving basis for those new arguments, models, predictions and projections.

    When you add the IPCC data with the leading edge, one can see the trends and potentials fairly well. Using both in our considerations I believe will help us formulate policy and reasonable perspective faster.

  50. 50
    Walt Bennett says:

    I’m no expert in these things, but it seems to me that when you rain on a desert you get mud, not a ‘breadbasket’. You still have no soil and no vegetation or animal life to spawn growth.

    Of course those things would develop over time through some processes about which others probably know a great deal. What I do suspect is that we are talking about a very long time; generations, certainly.

    As for those areas which suddenly get no rain, their long-term future will be even less certain, with this exception: those who relied upon the fertility of that habitat will be forced to relocate.

    To the mud flats, perhaps?