Climate Science Glossary

Term Lookup

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


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


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...

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

Latest Posts


Climate Hustle

Christy Crock #6: Climate Sensitivity

Posted on 7 June 2011 by dana1981, James Wight

Christy Crocks (200 x 70 pixels)In recent months, climate scientist "skeptic" John Christy seems to have grown an appetite for the limelight.  First he testified before U.S. Congress, then he appeared on an Australian radio talk show to discourage their country from implementing a carbon tax, and now he has appeared on a Canadian radio show to downplay the link between global warming and extreme weather.  In the process, he made one of the "skeptics" favorite arguments: climate sensitivity is low.

"We are finding that the climate is not very sensitive to CO2 and those kind of gases"

This statement is strikingly similar to one which we examined in Lindzen Illusion #4:

"If we doubled CO2, it's well accepted that you should get about 1 degree warming if nothing else what we've seen so far suggesting that you have more than that, and the answer is no."

We've also addressed the "low sensitivity" claims of Roy Spencer, David Evans, and Christopher Monckton.  It's not surprising that this is a favored "skeptic" argument, because low sensitivity is absolutely critical to arguing that global warming is of no concern, and opposing carbon emissions reductions.  I've previously even gone so far as to describe it as the "skeptic" endgame.

Christy and Lindzen's similar quotes are particularly puzzling because of their use of the term "we".  When I hear a climate scientist use the phrase "what we've seen" or "we are finding" when discussing a climate research issue like sensitivity, I assume they are referring to the body of climate science research.  But as we have discussed ad nauseam, the body of climate science research is in very strong agreement that climate sensitivity is not low, but rather likely between 2 and 4.5°C for doubled atmospheric CO2 (Figure 1).  In fact there are very few exceptions to this strong agreement, which are generally limited to fundamentally flawed papers by these same few "skeptics", like Lindzen and Choi (2009) or Roy Spencer's blunders and 'silver bullets'.

Various estimates of climate sensitivity

Figure 1: Various estimates of climate sensitivity (Knutti and Hegerl 2008).

It becomes a bit frustrating that these few "skeptics" continue to claim with such certainty that sensitivity is low, and in such a misleading fashion (i.e. saying "we" rather than "the work of a few like Lindzen and Spencer"), even though we have debunked this myth many times before.  But Christy's repetition of the  myth does give us the opportunity to discuss an interesting recent draft paper which examined the issue of climate sensitivity: Hansen and Sato's Paleoclimate Implications for Human-Made Climate Change.

Hansen and Sato (2011)

In the paper, Hansen and Sato evaluate fast-feedback climate sensitivity based on the radiative forcings in the last ice age (~20,000 years ago) compared to those during the Holocene prior to the Industrial Revolution (Figure 2), knowing that the planet was near equilibrium during these two periods:

"Climate, averaged over a few millennia, must be in near-equilibrium during the last ice age (~20 ky ago) and in the current interglacial period prior to introduction of substantial human-made climate forcings. Any planetary energy imbalance was at most a small fraction of 1 W/m2, as shown by considering the contrary: an imbalance approaching 1 W/m2 would be sufficient to melt all ice on Earth or change ocean temperature a large amount, contrary to numerous paleoclimate data records."

Fig 2

Figure 2: Climate forcings during the ice age 20 ky ago relative to the pre-industrial Holocene from Hansen and Sato (2011)

As you can see in Figure 1, the radiative forcing in the last ice age as compared to the pre-industrial Holocene is -6.5  +/- 1.5 W/m2.  The average global temperature was 5 +/- 1°C cooler during the past ice age, which means the climate sensitivity is 5°C/6.5 W/m2, or approximately 0.77°C per W/m2.  Since the radiative forcing associated with doubled CO2 is 3.7 W/m2, this implies a fast-feedback climate sensitivity of 2.8°C. Note that this is an empirical result based on real-world observations of climate change in the relatively recent past.

Hansen and Sato break down the approximate contributions of CO2 and feedbacks to this doubled CO2 warming:

"If Earth were a blackbody without climate feedbacks the equilibrium response to 4 W/m2 forcing would be about 1.2°C (Hansen et al., 1981, 1984). The water vapor increase and sea ice decrease that accompany global warming can be simulated reasonably well by climate models; together these two feedbacks approximately double the blackbody sensitivity. The further amplification is the net effect of all other processes, with aerosols, clouds, and their interactions probably being the most important of the remaining feedback processes."

The paper goes on to note that during the Pliocene, which was only 1-2°C hotter than current global temperatures, and also the last time CO2 levels were as high as today, average sea level was 25 meters higher (a subject which John Cook recently discussed). 

In another of Hansen’s draft papers, Earth’s Energy Imbalance and Implications, he concludes that slow feedbacks greatly amplify the 3°C sensitivity from fast feedbacks. The exact value depends on the climate state you start with, which feedbacks you include, what timescale you’re interested in, and what assumptions you make. Adding other greenhouse gases increases climate sensitivity to 4°C for doubled CO2, and adding ice albedo gives 8°C for doubled CO2 (as long as the Earth has ice sheets). These feedbacks have historically occurred over centuries to millennia, but could become significant this century. Including CO2 itself as a feedback would make climate sensitivity even higher, except for the weathering feedback which operates over hundreds of millennia.

To avoid kicking off slow feedbacks, Hansen advises that we return the Earth to energy balance by eventually reducing atmospheric CO2 levels to 350 ppm. Of course if we listen to the 'low sensitivity' crowd and buy into the wishful thinking that CO2 emissions are of little concern, we will blow post the 2°C danger limit (approximately 450 ppm CO2), let alone giving ourselves a reasonable chance to eventually reduce CO2 levels to 350 ppm.


Astute Skeptical Science readers may have noticed that "climate's changed before" recently supplanted "it's the Sun" as the #1 most used "skeptic" argument.  In the same radio interview, Christy made a similar argument, first specifically regarding extreme weather events:

"In looking at the best climate data through the past, these events are not outside the range that of what has already occurred"

But he then expanded the argument further:

"Extremes events are not the key that you look for, you look for a large global number in the heat storage of the atmosphere and ocean and that is rising slowly but it is not rising catastrophically or dramatically and certainly does not point to a high sensitivity of the climate to things like greenhouse gases."

This is simply wrong, because as we saw in Lindzen Illusion #1, the amount of warming we've observed is consistent with a climate sensitivity of approximately 3°C for doubled CO2.  With these two statements, Christy seems to be trying to weave several climate myths together to make an argument something like "climate has changed before, the current warming isn't very significant, it points to low climate sensitivity, and it could just be natural variability."  This sort of storyline is consistent with Christy's recent claims in the media and before US Congress, and similar to the alternative hypothesis to anthropogenic global warming that Lindzen put together.

The problem, aside from being wrong, is that it's self-contradictory.  The climate certainly has changed before, and by large amounts.  How can we explain the ~5°C increase in average global surface temperature between glacial and interglacial periods if climate sensitivity is low?  Natural variability can only account for fractions of a degree change, and the Milankovitch orbital forcing isn't that large, especially if sensitivity is low. 

As Hansen and Sato and many other studies have showed, these past climate changes are consistent with a climate sensitivity of 3°C.  The larger the past climate change, the larger sensitivity is required to explain them.  So arguing both for low sensitivity and large past climate changes is self-contradictory.

Misleading Christy

In short, Christy and the other "skeptics" who are claiming high certainty that climate sensitivity is low are not only contradicted by dozens of studies on the subject, including Hansen's empirically-based paleoclimate assessments, and by their own claims of large past climate changes, but they are advocating that we continue on an extremely risky path. 

It's particularly disturbing when these "skeptics" claim in the media that "we" are seeing evidence that climate sensitivity is low, implying that they are referring to the body of climate science research, when in fact the opposite is true.  In doing so they mislead the public about the actual state of the scientific research on climate sensitivity, which may lead people to the mistaken conclusion that urgent action is unnecessary.  In fact Christy seems to be going on an international public misinformation tour, having recently made similar misleading statements to the American, Australian, and Canadian public: a disturbing trend.

0 0

Bookmark and Share Printable Version  |  Link to this page


1  2  Next

Comments 1 to 50 out of 78:

  1. Dana & James:

    Kudos on an excellent article!

    As Sgt. Joe Friday was wont to say:

    "Just the facts ma'm. Just the facts."
    0 0

    [dana1981] Thanks very much.  The "skeptics" have given us a lot of practice in rebutting "climate sensitivity is low"!

  2. Great informative write up.

    As for Christy using "WE" perhaps he has merely adopted Lord M's habit of talking in the 'regal' first-person plural form.

    It lends an air of superiority don't you know.
    0 0
  3. A post on a previous site stated that likely was defined as 66%. Is this correct? If so, is very likely 90%?
    0 0
  4. Eric, do you mean this:
    virtually certain >99%; extremely likely >95%; very likely >90%; likely >66%; more likely than not > 50%; about as likely as not 33% to 66%; unlikely <33%; very unlikely <10%; extremely unlikely <5%; exceptionally unlikely <1%.

    from introduction to AR4 synthesis
    0 0
  5. It's great that we're living in a global CO2 experiment so will be able to see for ourselves the results of significant increases in CO2
    0 0

    [dana1981] Lucky us?

  6. I don't really understand why we speak of "the" climate sensitivity, as if it were a unique "scalar" , like the electron mass. When applied to intra-annual variability (comparing summer to winter) , the boreal summer/austral winter is warmer than the austral summer, although the solar input is significantly lower (more than 6 % = 20 W.m-2 less ! ). This means that the "sensitivity" would be negative, which is absurd. This is due of course to the fact that lands reacts much more than oceans, and that they represent a larger area in the Northern hemisphere. So the average temperature doesn't depend only on "global" energy input, but also on its spatial repartition (which varies following the astronomical variations). In the same way, the fact that climate has delayed responses implies that the "sensitivity" is actually frequency dependent (much like a complex impedance). This is not T=S.F but rather T(omega) = S(omega).F(omega) (relations between Fourier components) , which is not the same at all and does not imply a proportionality between instantaneous values. And I even don't speak about unforced variations that cannot be described by "sensitivities" at all, since the global temperature can change with little or no change of forcings.

    So I'm rather surprised to see Hansen computing an "average" sensitivity by summing forcings that don't have the same spatial pattern (GHG and snow albedo for instance), which means actually adding apples and oranges, and even more surprised to see him comparing sensitivities on very different timescales. I think these discussions about "the" sensitivity are actually very confusing.
    0 0
  7. okatiniko@6 "Climate sensitivity" refers to the equilibrium sensitivity of the climate to a change in the forcings, i.e. it describes the change in global average temperatures once all regional variations (due to a change in forcings) have had time to balance out. Climate sensitivity describes long term changes in climate due to a change in forcings.

    Unforced variations are indeed not described by climate sensitivity. Unforced variations cause short term quasi-cyclic changes in climate.

    Your surprise about Hansen's averaging appears to be due to not appreciating what climate sensitivity actually describes. The discussion of "the" sensitivity is not confusing once you know that it refers to the equilibroum sensitivity of globally averaged temperatures to a change in forcings. Of course you could define many other sensitivities if you so desired, but the reason climatologists discuss "the" equilibrum climate sensitivity is that it is the definition that is relevant to a discussion of long-term climate change.
    0 0
  8. #6 - what Dikran said, and to add to the equilibrium climate sensitivity's uniqueness, it's value is not the same every time the Earth is forced from 280-390ppm. In a future Earth, where the Greenland Ice Sheet and Arctic sea ice have been melted, yet we return CO2 levels to pre-industrial values, Earth's equilibrium sensitivity will be different because of the lack of ice sheet and sea ice, even if all the other forcings were the same. Each particular state of the Earth (continent configuration, ice sheet configuration etc) has a particular equilibrium sensitivity to CO2 forcing. It's highly likely that ours is high enough to be very concerned about the quantity of carbon we are putting into the atmosphere.
    0 0
  9. Agree sky,
    The sensitivity is not a constant, but a value for here and now. This is part of the reason why calculations based on data during the last glacial maximum different from those millions of years ago. Sea ice is one issue, vegetation is another, and atmospheric CO2 concentrations also comes into play. The sensitivity is lowest during the last millenium, but that could be data limited.
    0 0
  10. I’d think climate sensitivity is not even a constant in the here and now, just an average over typical conditions. It’s like the average speed of a large truck (lorry) traveling through a hilly region. As this truck passes through the Solar Cycle Hills or climbs Enso Mountain, the driver of the truck shifts at times between low and high gear, which affects the actual speed.
    0 0
  11. #9: why would equilibrium sensitivity be lowest during the past 1000 years? Do you have a source for that statement? With easily changeable sea ice areas, the climate of the past 1000 years has a demonstrably substantial sensitivity, as we are proving now (~0.8C transient change for ~35% Co2 increase, so ECS 3C or more per doubling when slow feedbacks have operated). A climate without significant Arctic sea ice would likely have a slightly lower sensitivity that our present climate, as there would be less scope for albedo feedbacks.

    #10: You mis-interpret. Equilibrium sensitivity is not affected by events as transient as individual solar cycles or ENSO. It is the aggregated effect once slow feedbacks have had time to operate.
    0 0
  12. #11: See graph above from Knutti and Hegerl.
    0 0
  13. Eric #12 - the Knutti and Hegerl figure shows a large range of possible values for the past millenium. You'll note it also shows a smaller range for the 'current mean climate state', with a most likely value of 3°C.
    0 0
  14. True, but the instrument record showed a similar sensitivity with higher confidence than other methods.
    0 0
  15. I've discussed implications of the instrumental record on climate sensitivity a few times, most recently in Lindzen Illusion #1. Bottom line, it's not far from 3°C, but the aerosol uncertainty complicates things so it's hard to narrow down.
    0 0
  16. Agreed Dana. That is why I refrain from nailing down the sensitivity at 3°C, and do not rule out lower values (I do nto rule out higher values either, but seldom get an arguement on that side).
    0 0
  17. #11 - Surely when we discuss climate sensitivity unqualified, we are discussing the fast feed back climate sensitivity, that is, the warming to be expected within 20 or 30 years of a doubling of CO2 concentration, estimated to be about 3°C. At least that’s what I mean. This is a sensitivity that could be affected by short term climate bumps. I’ve seen model runs that had up to 20 year flat spots even using the CO2 emissions forcing expected later this century (just one possible outcome, of course).

    Does anyone really care about equilibrium climate sensitivity other than paleoclimatologists? Certainly not policy makers – it’s too far off.

    “About 40 percent of the equilibrium response is obtained within five years. This quick response is due to the small effective inertia of continents, but warming over continents is limited by exchange of continental and marine air masses. Even after a century only 60 percent of the equilibrium response has been achieved. Nearly full response requires a millennium.” – James Hansen here.
    0 0

    [DB] You should re-read your comment and think about what perceptions that other readers are apt to draw from it.  After reflection, you may wish to issue a revision to this comment.

  18. 17, SoundOff,
    Does anyone really care about equilibrium climate sensitivity other than paleoclimatologists? Certainly not policy makers – it’s too far off.
    That's amazingly callous. You would knowingly leave our descendants with a marginally livable planet, only because it won't happen in your own lifetime?

    The lives of your children and grandchildren, and everyone who comes after, are of no importance?
    0 0
  19. SoundOff - I certainly care. Equilibrium sensitivity tells us the minimum amount of warming we've committed ourselves to. It's also the measure used by international climate negotiations (i.e. the goal is to limit equilibrium warming to no more than 2°C). So I disagree that nobody cares - it's actually a very widely-used parameter.
    0 0
  20. DM #7 : I have a question then : assume that the Earth would be in a perpetual "boreal summer" state , with a distance to the sun equal to aphelia, and the North pole directed towards the sun (State A), and let all the equilibria be set up, after a very long time. Now assume the Earth to be in a perpetual "austral summer" state, with the distance equal to perihelia, but the South Pole directed towards the sun (state B). The input energy depending only on distance, and not on the inclination of rotation axis, F(A) < F(B). Would T(A) be higher or lower than T(B) ?
    0 0
  21. okatinko@20 You are missing the point, climate sensitivity is to do with a change in temperature caused by a change in forcings. In both scenarios there is no change in forcings, so it is meaningless to talk of equilibrium climate sensitivity in either context.

    Of course if the configuration of the planet changes radically, that may alter the baseline temperature, but that doesn't mean that the climate sensitivity (the amount climate changes in response to a change in forcing) will be different.

    Mathematically t(A) not being equal to t(B) does not imply that dt(A)/df does not equal dt(B)/df, where t is temperature, f is forcing and A and B represent the two scenarios in your example. The dt/df is climate sensitivity, not the t.
    0 0
  22. Don’t exaggerate the meaning of my rhetorical question. I just mean that you aren’t going to make any progress towards mitigation efforts if you talk about something that will happen a 1000 years from now – which is how equilibrium CS is defined by Hansen (my source for such matters).

    Concern over such a long time frame is beyond personally caring about your children and grandchildren. It’s beyond caring about even your great-great-great-grandchildren, and is into sainthood territory. You may care about the planet’s future in theory, but arguing for benefits that are a millennium away says to me that you don’t really care in practice, or you are quite gullible about how real politics works. Stay focused on what will get action, that’s fast feedback climate sensitivity and its evidence.

    dana1981, your [IPCC’s?] equilibrium sensitivity sounds more like my [Hansen’s] fast feedback climate sensitivity. Do we have terminology confusion? Maybe we need a whole SkS article (hint) just to clarify all the following, each of which I’ve seen defined differently in literature, sometimes in the same paper …

    Equilibrium climate sensitivity
    Fast feedback climate sensitivity
    Transient climate sensitivity
    Effective climate sensitivity
    Long-term climate sensitivity
    Earth system climate sensitivity
    Slow feedback climate sensitivity
    0 0
  23. soundoff@22 Actually, it isn't a rhetorical question, in economics it is called "discounting", and apparently it is an argument that is genuinely considered in discussions about mitigation efforts.

    1000 years may be too long a timescale for you to personally care, but not for others. 1000 miles is enough distance for many not to personally care as well, but does that mean that problems that ocurr in (say) the next decade in the third world due to climate change are not our concern? No.

    IMHO, the "discounting" argument stinks.
    0 0
  24. SoundOff #22 - yes you're right, the IPCC equilibrium sensitivity is the same as Hansen's fast feedback sensitivity. My mistake. We may take your suggestion and do a post clarifying these different concepts.

    Dikran #23 - discounting makes some sense from an economic perspective, but it's a tricky question. There's quite a bit of debate regarding what the discount rate should be. The relatively few 'skeptic' economists (i.e. Richard Tol) think the discount rate should be incredibly high, which thus makes reducing GHG emissions now potentially more expensive than adapting to climate change in the future.
    0 0
  25. @dana1981 don't worry, I have learned over the years that me not liking something doesn't actually make it wrong! ;o)

    In cricket when you are batting, you don't leave scoring runs to your team mates lower in the order, you take responsibility and (try to) get the job done yourself. You can be reassured by the solidity of the lower order batsmen, but you shouldn't rely on them. Leaving future generations to deal with the problem seems rather irresponsible to me. Howver this is heading off-topic...
    0 0
  26. Yes this is going off topic, but I will say that I agree with you that discounting can result in screwing future generations, and thus you can certainly argue that high discount rates are irresponsible (I tend to agree). Most economists don't use terribly high discount rates, which is why the vast majority agree with taking action to reduce emissions now from a purely economic perspective.
    0 0
  27. 22, SoundOff,


    Sorry if I overreacted, but you seem to have very quickly adopted Hansen's thousand year scenario as gospel for us, and hence an argument (whether you intended it as such, or not) for inaction.

    There are a few problems with that.

    First, it's only one paper.

    Second, you stopped reading too early. In the very next paragraph he goes on to say:
    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.
    Third, your premise silently hinges on the idea that we'd stop at a mere doubling, and climate sensitivity is at worst in the middle around 3˚C. Even then, 60% of that is an uncomfortable 1.8˚C. But if actual sensitivity turns out to be 4˚C, and we go well beyond a doubling, then 60% of say 5˚C is 3˚C, and suddenly it is a lot of suffering for our own children and grandchildren.

    Fourth, Hansen's time frame estimate is based on runs of a single model trying to project something that's never happened in the history of the planet. I'm not saying this makes it entirely unreliable, but the error bars are logical as well as quantitative. Lord help us if the model hasn't recognized certain elements (such as a potential fast-methane feedback) that are difficult to model properly.

    I guess, in summary, I'd say that while people might look at that thousand year projection and honestly say "who cares, it's not my problem," I still think any reasonably careful, sane and caring person would look at all of the error bars involved, and then be too responsible to take that sort of risk.
    0 0
  28. DM#21 : but as I understand it, Hansen compares temperatures in two different situations A and B, which may not be as different as in my extreme example, but are nevertheless different : because glaciations are due to astronomical variations that do influence the local repartition of solar irradiance, and because the change of forcing due to ice coverage doesn't have the same spatial distribution of, for instance, GHG gases. So he is really evaluating something like (tA(fA) - tB(fB))/(fA-fB) , and not local variations with "the same situation" dtA/dfA or dtB/dfB.

    Is it not a concern to assimilate global differences with local derivatives ? as I said, "global differences" between boreal winter and summer would yield a negative sensitivity if computed in the same way as Hansen seems to do it.
    0 0
  29. okatinko@28 Hansen is assuming that climate sensitivity (dt/df) is essentially the same during the last ice age as it is in the current interglacial, so dt/df \approx dt(A)/df(A) \approx dt(B)/df(B). This means you can reasonably approximate dt/df as the ratio of the change in t and the change in f between the two scenarios, in other words,

    sensitivity = dt/df \approx [t(A) - t(B)]/[f(A) - f(B)]

    If we take the limit as A goes to B, then by the fundamental definition of the derivative the approximation becomes exact.

    The second paragraph was dealt with in my earlier reply. The difference in your thought experiment is not a difference in forcing, and hence is irrelevant to a discussion of climate sensitivity.
    0 0
  30. Okatinko,
    Yes, T(A) > T(B). The greater land mass in the NH would result in significantly more heat being absorbed by the surface. This is one of the premises in the Milankovitch theories. These changes are thought to be responsible for the ice ages.
    0 0
  31. DM#29 : precisely I doubt the validity of this assumption since the various forcing have different spatial repartitions - or more precisely I haven't seen any strong justification for this assumption. Concerning my thought experiment, it is undoubtedly a change of solar energy input , so it has exactly the same effect than the possible influence of solar activity, or a change of albedo. My point is that the temperature can be higher with a lower energy input, so there is nothing like a simple relation T(F) : it must be T(F, x, y...) where x, y.. are parameters describing for instance the repartition in latitude, the oceanic circulation, and so forth. So there is also nothing like a single derivative dT/dF . It could be a partial derivative ∂T/∂F |x,y... but this coefficient depends on what you choose as the "other coordinates" x, y, to be held constant. Again nothing like "one" single sensitivity valid under any circumstances.
    0 0
  32. #8, 9, and 10 make the point that climate sensitivity to a doubling of CO2 is not a constant and will vary depending on changes of other factors such as albido. Are they right? I think not. Surely the ability of CO2 to absorb and radiate energy is a constant which is not affected by such externalities. Ergo, ceteris paribus, the effect of doubling CO2 in the atmosphere will be constant.

    Average global temperature is affected by the level of CO2 in the atmosphere but obviously externalities such as slow feedbacks, particularly release of Arctic methane and diminution of albido due to melting of snow and ice, also have an effect, as indeed do clouds and aerosols. The net effect of these changes may be greater than that arising from a doubling of CO2.

    The point to be made – and made well by Hansen – is that a CO2 concentration of 450ppm is unsafe because is causes slow feedbacks to accelerate beyond human control. Indeed, that point has already been reached and nothing short of a reduction in atmospheric CO2 can change it, hence his call for a reduction to 350ppm.

    Hansen tells us that we are fast approaching the Pliocene temperatures, when sea level was 5m. higher than to-day, simply warns us of what lies ahead if we aim to limit CO2 concentrations to 450ppm – disaster and dangerous, uncontrollable climate change. It is a shame this message and its implications escape the comprehension of our political leaders and the general public. None of them can claim they were not warned! Few can claim they did their best to respond.
    0 0
  33. Agnostic @32, the no feedback forcing of doubling CO2 is close to a constant throughout history. I believe that at very high and very low concentrations of CO2 it does change, but over the range of CO2 concentrations over the last 500 million years, it is effectively constant.

    However the feedback responses need not be constant through history. In particular, distributions of continents, ocean currents and shallow seas can make a large difference. This can be clearly observed in the different climate sensitivities in the Arctic and the Antarctic. Clearly if the Antarctic continental configuration was duplicated in the Arctic, climate sensitivity would be lower, while if the Arctic configuration was duplicated in the Antarctic it would be higher.

    Another difference arises from mean global surface temperatures. Given glacial conditions, climate sensitivity will be higher. This follows from the difference in insolation at, for example, 50 degrees North compared to 60 degrees North. A retreat of permanent or seasonal snow and ice by one degree at 50 degrees will result in a greater net warming than the same one degree retreat at 60 degrees. The difference in surface areas of a one degree latitude band also makes a large difference between those two latitudes.

    None of this, however, shows that climate sensitivity has ever been less than 2.5 degree per doubling for a no ice albedo feedback. In fact, given the CO2 in the atmosphere accounts for 20% of the total greenhouse effect, averaged over the temperature range between snowball Earth and current temperatures, and ignoring ice albedo effects, the climate sensitivity per doubling of CO2 has been 5 degrees. Further, as the water vapour feedback increasingly dominates the CO2 greenhouse effect with increasing temperature, climate sensitivity ignoring albedo effects will rise with increasing global temperatures, not fall.

    With those two facts in mind, I would say Hansen's estimate of short term climate sensitivity (no ice albedo feedback) from the last glacial maximum is a fair estimate for current conditions, but that his slow feedback is probably an overestimate. The slow feedback will certainly be greater than the fast feedback until there is no permanent ice sheets on Antarctica, and only winter snow inside the Arctic or Antarctic circle. But it will not be as great as that during the height of the glacials (or even the average of glacials and interglacials).
    0 0
  34. Agnostic,
    I echo Tom's response that no one is arguing the no-feedback effect. The direct effect attributaed to CO2 has remained virtually unchanged (although there may be saturation effects at very high levels). I disagree that the sensitivity has never been less than 2.5 or been as high as 5. Also, can you reference that 20% value?
    0 0
  35. okatinko wrote "precisely I doubt the validity of this assumption since the various forcing have different spatial repartitions"

    As I have already pointed out, those different spatial repartitions have no effect on the equilibrium global average temperature, as the averaging averages them out.

    "My point is that the temperature can be higher with a lower energy input, "

    Yes, but that is a statement about the absolute temperature, not the rate of change of temperature with a change in the forcing. CLimate sensitivity is the latter, not the former.

    "T(F, x, y...) where x, y.. are parameters describing for instance the repartition in latitude,"

    In my original notation I used A and B as index variables representing the state of the planet. It was you that incorrectly changed t to be a function of f alone, not me.

    Climate sensitivity appears to be exactly what yopu have in your notation when your write ∂T/∂F |x,y..., where x and y etc. describe the configuration of the planet independent of the forcings, which has not changed between the last ice age and the current interglacial, but has changed between e.g. the Triassic and now. Hence on geological timescales climate sensitivity does change, but only on geological timescales. Note glaciation is not a change in planetary configuration as it is a consequence of a change in forcings (i.e. a feedback within an essentially constant system).
    0 0
  36. agnostic Indeed "the ability of CO2 to absorb and radiate energy is a constant", however this describes the radiative forcing due to increased CO2, climate sensitivity describes the response of global temperatures to that additional forcing.
    0 0
  37. 34, Eric the Red,
    I disagree that the sensitivity has never been less than 2.5 or been as high as 5.
    Evidence and citations, please.
    0 0
  38. On climate sensitivity in general, I would point out that there are only four factors (well, five, if you introduce humans, and even then... but I'm getting ahead of myself)...

    Sorry. I would point out that there are only four factors which affect climate, and are themselves unaffected by climate:

    1) The sun
    2) The locations of the continents
    3) The orbital factors of the earth (inclination, precession, obliquity, etc.)
    4) Volcanic activity

    [The last, volcanic activity, obviously is only a continuous forcing if the earth enters a sort of "new volcanic activity regime" such as a sudden explosion or quieting of the Circle of Fire.]

    For these factors, if the climate changes, it does not alter their inputs, while for all other inputs, although sometimes on massive time scales, a change in climate in turn changes their inputs.

    CO2 in the atmosphere, ice albedo, clouds, water vapor, lapse rate, everything will changes (over large time scales, in some cases) if temperature changes.

    One comment mentioned a different sensitivity, for instance, if the Greenland and Arctic ice sheets were already melted. I don't think, in properly complete climate time scales, that this matters. If climate changes, those ice sheets would reform, albedo would drop, and that would be a positive feedback (unless their absence was a result of land mass configuration, which is the point I'm trying to make).

    I believe (and by believe I mean that my logic and thought process is obviously limited, without doing extensive studies and verifications to confirm that my logic is accurate) that while climate sensitivity may vary based on alterations in long term, non-climate affected inputs (sun, land disposition, orbit, volcanoes), within any set regime it will be pretty constant, and I actually do not see much reason for much variation in sensitivity even with changes to those four parameters.

    Obviously there would be some, but I see no reason to believe that the degree would be too great (except, perhaps, in the case of landmasses, which by adopting a configuration which prevents the formation of ice caps also prevents a major feedback, ice albedo, and by removing one feedback from the equation could alter sensitivity by a notable degree).

    But everything else depends on physics which is in turn affected by temperature; lapse rate, water vapor, cloud formation (maybe also affected by land masses), CO2 balance, etc.

    Now to the new, fifth forcing: humanity. At first, I thought of that as a true forcing because it does not also respond as a feedback. We burn CO2 regardless of whether the temperature goes up or down. We don't decide to burn more CO2 because the planet warms, or stop burning it when the planet cools.

    Then I thought about it on geologic time scales, and realized that maybe that's not true. I'm not sure that modern civilization would ever have formed during a glacial, so maybe our ability to poor CO2 into the atmosphere is a one-time positive feedback response to a warming planet.

    And we can't keep doing it forever. The increase in temperature will either kill us, or force us to wise up and ease off the gas, so the speak. In which case rising temperatures will cause us to stop (not a negative feedback, but instead a stop-forcing response).

    So we're back to the original 4: sun, land configuration, orbit, and volcanic activity, as the only true forcings that affect climate and might also vary sensitivity.... but I still don't see it varying by much, because except for ice albedo and land mass configurations, the physics is not going to change.
    0 0
  39. Is this the first article in the Christy Crock series to be posted? If so, why is it #6?
    0 0
  40. Sphaerica, I would agree that there are four inputs (and I would agree than any thing else is a result of those inputs), with any time of galactic input being an anomaly. In the past, volcanoes have had short-term effects, except for the few super volcanoes. That leaves three primary inputs.
    The location of the continents has been shown to affect input, but change on such a long time scale to be only relevant to paleoclimatology. Surface albedo may play a much greater role.
    Orbital factors cause changes on a much shorter time scale; thousands up to a hundred thousand years.
    That leaves the sun. Total solar irradiance has been shown to vary only slightly, and is unliekly to be a large climate influence. Sunspots are much more variable, as is the solar magnetic field, and the results are inconclusive.
    I would argue that CO2 is only an input with regards to cloud formation and blocking incoming solar radiation. Most of the effects concern restricting output. Also, CO2 is affected by temperature; ocean dissolution and plant respiration.
    The physics is not going to change. While your four inputs may have changed significantly in the past, the sensitivity to CO2 has probably been fairly constant, except during the early Earth, when atmospheric concentrations approached 1%.
    0 0
  41. Badgersouth, no the earlier items in the series can be found here.
    0 0
  42. 40, Eric the Red,

    Your comment on volcanoes in recent times is certainly correct, but I'm talking about geologic time scales, and the full history of the earth. In that history, periods of extreme volcanic activity (far in excess of what we see today) have been responsible for boosting CO2 in the atmosphere (Precambrian). I've also seen theories (i.e not entirely substantiated) that great increases in volcanic activity have caused sustained climate change in recent times (a theory that it caused Neandertals to go extinct) and further back (that it was implicated in the extinction of dinosaurs, through climate change).

    Your points about time scales is certainly true, and I thought obvious, but not relevant. I wasn't making any statement about current climate change causes, only about the likelihood of variability in climate sensitivity due to configurations, and the fact that these are the only parameters that would affect sensitivity, since all other aspects are themselves temperature dependent.

    As such, my point was more that any variability in climate sensitivity would only come into play when using paleoclimate data from the very distant past (i.e. different configurations), and even then, any variation would I expect (my opinion, not fact) be slight.

    Your final statement that physics is not going to change, and that the sensitivity to CO2 has probably been fairly constant, is exactly the point I was trying to make.

    So we agree on something!!!!

    The next round's on me!
    0 0
  43. Eric the Red,

    I'm still waiting for evidence and citations for your statement about sensitivity at 34, however.
    0 0
  44. DM#35 : you write "
    As I have already pointed out, those different spatial repartitions have no effect on the equilibrium global average temperature, as the averaging averages them out."
    but of course they have an effect, since this is precisely the reason why the boreal summer is warmer than the austral one, despite a lower input flux ! spatial repartition of forcings must be taken into account for sure !

    "Yes, but that is a statement about the absolute temperature, not the rate of change of temperature with a change in the forcing."
    rate of change ? you mean a time derivative dT/dt ? as far as I know there is nothing like a time derivative in the definition of sensitivity - it's just the equilibrium value.

    Climate sensitivity appears to be exactly what yopu have in your notation when your write ∂T/∂F |x,y..., where x and y etc. describe the configuration of the planet independent of the forcings, which has not changed between the last ice age and the current interglacial, "

    why do you say it has not changed ? my point was precisely that it HAS changed, at least first because astronomical precessions and second because of the ice albedo that doesn't have the same spatial repartition as, for instance, the GHG forcing. May be I'm wrong, but you didn't really show me why.
    0 0
  45. Sphaerica,
    The issue of lower cliamte sensitivity in past eras has less to do with the physical properties of CO2, than in the biological and chemical reactions which increased removed of CO2 from teh atmosphere. Here is one paper that demonstrates that occurrance (note a specific climate sensitivity is not given, but can be presumed).
    0 0
  46. okatiniko @44, the Northern Hemisphere summer is warmer than the Southern Hemisphere summer despite greater insolation in the SH summer because the ocean has a much greater thermal inertia than does land. Because there is more land than water in the NH, and the reverse in the SH, it takes longer for the SH to approach the equilibrium temperature than does the NH. If insolation where held constant in a SH Summer Solstice conditions, the resulting global temperature would be significantly higher than if the same where done for the NH winter Solstice.

    In contrast with the seasonal variation, the temperature fluctuations between glacial and interglacial are sufficiently slow that treating temperatures as having approached equilibrium for the level of forcing is a reasonable approximation. Consequently Hansen's method of determining climate sensitivity is valid for that data, but would be invalid if applied to seasonal data where such a presumption is straight forwardly false.

    So far as I can tell, your entire argument against Hansen's sensitivity calculation is based on the invalid analogy between centenial and seasonal temperature variations. To the extent that is true, you have no valid criticism of Hansen. If it is not true, surely you can repost your argument without appeal to that invalid comparison as a distraction.
    0 0
  47. Sphaerica #42: a small nit - theories are the parts of science that are generally well-substantiated and tested, so presumably you meant hypotheses?

    I like your post at #38. I would add that there is an importance in considering the fate of the ice sheets. Regardless of the effects of presence or absence of ice on overall sensitivity (probably small but noticeable), we need to be aware of the fact that the ice sheets would not grow back as they are under present climate conditions. That is, if warming is sufficient that the ice sheets are melted, and then in the far future we are able to stabilise and manage global temperature, the ice sheets would not magically return to their present state merely by returning temperatures to those of the 19th or 20th Century. You would need to cool the climate very substantially lower than at present if you wished to lower the snow equilibrium line enough to grow the sheets back, due to the lost thickness of ice. It's part of the reason the ice sheets are at great risk from current warming, but it's not something that has a great bearing on the question of our climate sensitivity.
    0 0
  48. 45, Eric the Red,

    You are reading a lot (when you say "has less to do" rather than "may have less to do") into a letter which admits to and glosses over strong arguments that contradict the findings of their own work, and for which they admit there are large error bars in many areas.

    That said, the letter is also discussing climate 500 million years ago. I think that certainly there are some scenarios which would wildly vary climate sensitivity and the effects of certain mechanisms (in that case, CO2) then versus today's values (for instance, the lack of an extensive, land based life ecosystem 500 million years ago, to either accelerate or balance many of the chemical interchanges).

    The authors say as much in their conclusion when they say "climate models ... are calibrated to the present ... and may thus be unable to reproduce correctly the past climate modes". To me, this is less of a comment on climate sensitivity, and more of a comment on the fact that the actual dominant mechanisms at play could have been very different (again, on a planet 500 million years removed from this one, almost as if it were another planet in the solar system).

    Basically, trying to do climate work today is very hard. Trying to do it for 500 million years in the past is a daunting task, and not what I'd use as a yard stick for evaluating the quality of other science, or values of climate sensitivity. I'm not saying paleoclimate is not valuable, but 500 million years is really stretching it.

    You also still have not yet posted your evidence and citations for the position stated in post 34 that climate sensitivity is low.
    0 0
  49. 47, skywatcher,

    Yes... I actually typed "hypothesis," and thought about it, and forget why I changed it to "theory," but you're right.

    On the ice sheets coming back... I think my main (natural) view is that in normal climate sensitivity time scales are utterly huge, meaning that thousands of years is a very short time frame. Human time frames of even hundreds of years, let alone ten or five, are utterly silly.

    So the point is that the ice sheets would grow back, and increase albedo, in their own good time, which to mother earth is more than fast enough. The same thing goes for the very slow transition in CO2 levels under normal circumstances (abrupt glacial termination being a still unresolved exception).

    At the same time, it puts serious emphasis on exactly how crazy we are to be doing what we're doing. The earth's climate is a machine with a few knobs and levers and dials that the pilot lovingly adjusts, only a nudge here or a smidgeon there, and still he steers the planet between pretty marked extremes.

    We just barged into the cockpit, grabbed the biggest lever, and shoved it all the way forward, from lowest to highest, shouting "Yeehah! Flank speed! Go, baby, go!"
    0 0
  50. Eric the Red @34, for the 20% figure:

    Attribution of the present‐day total greenhouse effect
    Gavin A. Schmidt, Reto A. Ruedy, on L. Miller, and Andy A. Lacis

    Received 30 March 2010; revised 27 July 2010; accepted 3 August 2010; published 16 October 2010.

    [1] The relative contributions of atmospheric long‐wave absorbers to the present‐day global greenhouse effect are among the most misquoted statistics in public discussions
    of climate change. Much of the interest in these values is however due to an implicit assumption that these contributions are directly relevant for the question of climate sensitivity. Motivated by the need for a clear reference for this issue, we review the
    existing literature and use the Goddard Institute for Space Studies ModelE radiation module to provide an overview of the role of each absorber at the present‐day and under
    doubled CO2. With a straightforward scheme for allocating overlaps, we find that water
    vapor is the dominant contributor (∼50% of the effect), followed by clouds (∼25%)
    and then CO2 with ∼20%. All other absorbers play only minor roles. In a doubled CO2 scenario, this allocation is essentially unchanged, even though the magnitude of the total greenhouse effect is significantly larger than the initial radiative forcing, underscoring
    the importance of feedbacks from water vapor and clouds to climate sensitivity.

    (My empahsis PDF)

    Some indication of the climate sensitivity across a range of conditions is given by work by Pierehumbert et al (2011) from which Chris Colose shows this graph:

    The climate sensitivity at each point is given by the slope. In the upper section (non-snowball), climate sensitivity is never much below 3, and becomes significantly greater at higher levels of CO2. In the snowball Earth condition, climate sensitivity falls as low as 1.5 to 2 degrees per doubling of CO2. The increase in sensitivity for higher levels of CO2 is due to an increasingly powerful water vapour feedback.

    Chris Colose advices us to take exact numbers from these graphs with a large grain of salt, but the slopes at least are indicative.

    Pierrehumbert's graph is, of course, the product of a model. An evidence based approach to historical climate sensitivities can be found in Royer et al 2007. They take the fact that CO sequestration rates are controlled by temperature. Consequently, different CO2 sensitivities will result in different rates of accumulation of CO2 through out history. They model the CO2 cycle for different climate sensitivities and compare the results to the CO2 proxy record. As you can see, low climate sensitivities are inconsistent with the CO2 proxy record through out the phanerozoic, as are very high climate sensitivities.

    Admittedly the evidence is tenuous due to limitations in the proxy CO2 and temperature record in the past. But the evidence we have suggests a climate sensitivity around 3 degrees as a robust feature of the Earth's history.
    0 0

1  2  Next

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

The Consensus Project Website


(free to republish)

© Copyright 2019 John Cook
Home | Links | Translations | About Us | Privacy | Contact Us