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 Support

Bluesky Facebook LinkedIn Mastodon MeWe

Twitter YouTube 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...

New? Register here
Forgot your password?

Latest Posts


New Study Suggests Future Global Warming at the Higher End of Estimates: 4°C Possible by 2100

Posted on 8 January 2014 by Rob Painting

In everyday terms, climate sensitivity refers to the amount of warming of global surface temperatures with a doubling of atmospheric carbon dioxide - a potent planet-warming greenhouse gas emitted by human industrial activity. In practice, establishing Earth's actual climate sensitivity has proven very problematic. A large part of this likely stems from the emerging realisation that climate sensitivity is not a fixed value, but varies with the background state of the planet (Armour [2012], Meraner [2013]).

The recent (2013) AR5 Intergovernmental Panel on Climate Change (IPCC) Report published a summary of peer-reviewed research on climate sensitivity and found that the likely values (greater than 66% probability) ranged from 1.5°C-4.5°C (for a doubling of atmospheric CO2). This range was lower than the previous (AR4) IPCC assessment because it included climate model/observationally-based research which implied a lower climate sensitivity (see Otto [2013] for example).

However, a research paper just published, Sherwood (2014), suggests that climate sensitivity of relevance today is in excess of 3°C - near the upper range of estimates from the latest IPCC report. Climate models exhibit a large range of climate sensitivities and the main reason for this is down to the way that each model handles cloud feedback. In brief: an increase in cloud cover in response to global warming would act as a negative (counteracting) feedback - reflecting more sunlight back out to space and thereby cooling the Earth, whereas a decline in cloud cover would act as a positive (reinforcing) feedback - as more sunlight reaches the Earth's surface and this leads to greater warming.

The authors of Sherwood (2014) looked at the way that the various climate models handled the cloud feedback and found models with a low climate sensitivity were inconsistent with observations. It turns out that these models were incorrectly simulating water vapour being drawn up to higher levels of the atmosphere to form clouds in a warmer world. In reality (based on observations) warming of the lower atmosphere pulls water vapour away from those higher cloud-forming levels of the atmosphere and the amount of cloud formation there actually decreases. The diminished cloud cover leads to greater warming (a positive feedback). 

Of course, this is just one study on one aspect of climate sensitivity and is certainly not the last word on this topic. But Sherwood (2014) is consistent with, and builds upon, another peer-reviewed paper, Fasullo & Trenberth (2012), which found only those climate models with high climate sensitivity correctly simulated drying in key cloud-forming regions of the atmosphere.

Professor Steve Sherwood talks about his research paper in the video below (thanks to Peter Sinclair at Climate Denial Crock of the Week). If correct, this research indicates that business-as-usual fossil emissions may lead to globally-averaged surface temperatures rising in excess of 4°C by 2100 - a formidable threat to global civilization.  


1 0

Printable Version  |  Link to this page


Comments 1 to 33:

  1. I want to know how Sherwood 2014 influenced the actual numbers of ECS components (forcing+feedbacks) and how the new numbers add up.

    From AR4 estimates on model/observationally-based research (I don't quote latest AR5 because I think they underestimated it), here's a summary of central estimates:

    2xCO2 forcing: +1.25K

    H2O fb: +2.5K

    Snow/Ice albedo fb: +0.6K

    Cloud fb: -1.85K


    Total ECS: 2.5K

    The reason they quoted higher ECS (3.0K) is that other methods (paleo observations) gave higher results (4.0K+).

    Now, Sherwood 2014 claims to have adjusted (Cloud fb: -1.85K) component. According to this article, the adjustment resulted in a change in total ECS: 2.5K -> 4.0K. It means that Cloud fb component was increased by 1.5K. So, the new value of Cloud fb component is now -1.85+1.5 = -0.35K.

    Two observations/questions follows:

    1. Sherwood 2014 brought their model/observational ECS in par with paleo ECS

    2.The Cloud fb is appears still negative (-0.35K) after their adjustement, so why is the new cloud feedback described as "positive" in the text above? To be precise, it should be described as "less negative", or that previous research "overestimated the clouds' cooling effect". Am I missing something here or are my calculations wrong or is the article's  text inaccurate (clouds' overall fb is still negative according to Sherwood 2014)?

    0 0
  2. Good points, chriskoz. Here's a link to a review of the Sherwood article from Nature. Unfortunately, they don't seem to address the issue you point out. (I'm assuming it is just an issue of sloppy wording, but maybe there is something else there.)


    0 0
    Moderator Response:

    [RH] Fixed link that was breaking page format.

  3. I wrote the confused comment below in response to misunderstanding the following: "warming of the lower atmosphere pulls water vapour away from those higher cloud-forming levels of the atmosphere."  Perhaps the article can be edited to emphasize that "lower atmosphere" means "non-cloud-forming altitudes" -- I initially thought the distinction was between "high cloud altitudes" and "low cloud altitudes."  See below to understand my confusion...

    This is confusing to me, at least on a superficial level.  I googled "high cloud low cloud feedback" and went to the basic explanation on SKS for "What is the net feedback from clouds?"  There the text indicates that high level clouds produce a positive feedback and low level clouds produce a negative feedback.  The above indicates that we'll have fewer high level clouds because water vapour is actually drawn from the upper atmosphere into the lower atmosphere (which is warming).   ...  Ah, now watching the video I understand.  This is not about high clouds versus low clouds.  This is only about low clouds.

    0 0
  4. I'm afraid I'm still confused. At minute 1:40 Sherwood says that in some models the water vapor always rises up to 10 or 15 kilometers--wouldn't those be high clouds??
        From clouds wiki: "Clouds of the high family form at altitudes of 3,000 to 7,600 m (10,000 to 25,000 ft) in the polar regions, 5,000 to 12,200 m (16,500 to 40,000 ft) in the temperate regions and 6,100 to 18,300 m (20,000 to 60,000 ft) in the tropical region."
       But at the end he says that it is the low clouds that don't form. So I'm still confused.

    1 0
  5. My understanding of the Sherwood Paper is that it says cloud formation at lower altitudes (>850 hPa) reduces as temperature increases because of mixing with cold dry air from mid troposphere (<750 hPa) dissipating water vapor near the surface.  In other words, the warmer it gets, the less low cloud is formed, reducing its albedo and increasing solar radiation reaching the surface.  Is this wrong?

    0 0
  6. @Rob

    If you could find the time to explain in detail (diagrams and all) where high and low clouds form, where mixing takes place, how mixing influences cloud formation and why mixing will increase when the temperatures go up - this would probably answer many of the questions being asked in this comment section.

    I must admit, I'm still confused.

    0 0
  7. I'm no expert, but I think what they're saying is if the atmosphere heats up due to CO2, where does it heat up?  In heats up in the lower troposphere, mostly.  Hence, where is additional H2O going to make its home?  In the lower troposphere, below cloud-formation level.  This has the unfortunate effect of increasing the lower troposphere further, making it a better home for even MORE H2O.  If much of the additional H2O never makes it high enough to form clouds, its a positive feedback (or at least less of a negative feedback than it would be if you assumed most of it WAS making it up that high).

    0 0
  8. Sorry, I think I put this question under the wrong topic, so I'll put it here where I think it belongs.

    Does equilibrium climate sensitivity include complete equilibration with the hydrosphere? Does this imply a time span of a thousand or so years? Would a transient CS be more appropriate coefficient to discuss for a century timescale?

    0 0
  9. Clouds:

    The paper specifically addresses moisture levels in the 2km to 3km range, meaning low clouds, i.e. those that have a primarily cooling effect by reflecting incoming sunlight (so less such clouds would be a positive feedback).  The paper did not re-run model simulations, but rather examined existing models, and various parameters, to identify those models whose temperature and humidity changes in the 2km to 3km range more accurately reflected actual observations, taking these models to be more accurate in that respect.

    What they found was a very, very strong correlation between how well a model matched observations in that respect, and therefore had fewer low level clouds, and climate sensitivity.

    Those that more closely matched observations consistently had a higher climate sensitivity.

    Those that were less of a match for observations consistently had a lower climate sensitivity.

    Restricting estimates to include only those models whose ensembles more closely match observations leads to the ECS given.

    0 0
  10. tcflood,

    On choosing between ECS vs. TCS:

    Only if we can presume to know the temporal difference between the two, but given that humidity and low cloud changes should be fast feedbacks, i.e. ones which occur very rapidly as a direct consequence of changes in land surface, ocean surface and atmospheric temperatures, then my immediate reaction would be that this is a very big problem, and there is no reason to think that we should ignore the results because final ECS is a thousand years in the future.

    But the distinction has little to do with this paper.  It's more of a general question of "well, do I care about TCS, because I could care less about anyone but my own generation and maybe the next, or do I care about ECS, because I care about future generations, even after I'm gone."

    With that said, if you want to know what difference this makes in the model runs, you'd have to look at the selected and excluded ensembles, and compare their TCS period results.

    0 0
  11. But yes, ECS concerns longer time frames than TCS.  We can't actually say "thousands of years," though, because we can't really know at this time how quickly the slowest feedbacks will take effect, of how much of the total climate sensitivity they will represent.

    0 0
  12. Some of the questions about the mechanism described in the paper are answered by Sherwood in the video (1:08):

    "What we see in the observations is that when air picks up water from the oceans surface and rises up, it often only rises a few km before it begins its descent back to the surface.  Other times it might go up 10 or 15 km.  And those shorter trajectories turn out to be crucial to giving us a higher climate sensitivity, because of what they do to pull water vapour away from the surface and cause clouds to dissipate as the climate warms up.  

    In many of the models this doesn't happen.  The air allways rises up to 10 or 15 kms, and so these models have been predicting a lower climate sensitivity, but we believe they're incorrect.

    What these shallower overturning circulations do is they pull the water vapour away from the part of the atmosphere where clouds form that cool the climate.  If these things are strong, and if they do this more in a normal climate, which is what all the models agree is what happens, then you loose your low clouds and the Earth absorbs more sunlight.

    The question for many years has been what's going to happen to the amount of low cloud.  Does it decrease when the Earth warms up, or does it stay the same, and maybe even increase; and what we found is that it should be decreasing because of this mixing process which pulls water vapour away from the layers were these clouds form, and causes there to be fewer of them in the wamer atmosphere." 

    So, the increased warming is definitely due to the reduction of low cloud.  

    However, Sherwoods account is simplified relative to the paper.  Specifically, in the video, where he refers to the circulation that "only rises a few km", the video shows a circulation rising to 5 km.  In fact the paper mentions two forms of low circulation:

    "As discussed above, air there is either transported directly from the boundary layer with minimal precipitation via lower-tropospheric mixing, or indirectly by ascending in deeper, raining clouds and then descending. The air would arrive cool and humid in the former case, but warmer and drier in the latter case owing to the extra condensation, allowing us to evaluate which pathway dominates by observing mean-state air properties."

    That is, in the lower tropospheric circulation, sometimes air (with its attendant water vapour) is carried only as high as the level of formation of low clouds (around 2 km); but sometimes it is carried higher - precipitating out some of the water vapour, before descending back to the level of low clouds. 

    Sherwood et al measure the different rates of these two types of circulation with the index S:

    "To do this we use an index S, proportional to the differences DT700–850 and DR700–850 of temperature and relative humidity between 700 hPa and 850 hPa (S taken as a linear combination; see Methods Summary) averaged within a broad ascending region which roughly coincides with the region of highest Indo-Pacific ocean temperatures (the Indo-Pacific Warm Pool; Fig. 1)." 

      To help make sense of that, 850 hPa corresponds to an altitude of 1.5 km, and 700 hPa corresponds to an altitude of 3 km.

    In addition to these two forms of low level circulation, Sherwood et al describe a higher level circulation:

    "We next turn to the large-scale lower-tropospheric mixing, which we associate with shallow ascent or flows of air upward through the top of the boundary layer that diverge horizontally before reaching the upper troposphere. Although air ascending on large scales over warm tropical oceans typically passes through nearly the whole troposphere, over cooler oceans its ascent often wanes with altitude, showing that this type of mixing indeed occurs in the Earth’s atmosphere (Fig. 3). The associated mid-level outflows are well documented for the central and eastern Pacific and Atlantic Intertropical Convergence Zone and some monsoon circulations13,14. Although these are indeed the regions where shallow ascent is steadiest, and hence clearest in monthly-mean data (Fig. 3), in daily reanalysis data, shallow ascent is equally strong outside the tropics owing largely to contributions from extratropical storms."

    Clearly this form of circulation will also dry the region around 2 km altitude where low level cloud forms.

    Sherwood et al define an indice of the ratio of large circulation to low level circulation:

    "We quantify the large-scale lower-tropospheric mixing more thoroughly by calculating the ratio D of shallow to deep overturning (see Methods Summary) in a broad region encompassing most of the persistent shallow ascent (see Fig. 3). This index D varies by a factor of four across 43 GCMs (see below). Interestingly, however, D and S are uncorrelated (r50.01), confirming that the two scales of mixing are controlled by different aspects of model design."

    Because low circulation provides the numerator, and high circulation the denominator, D will be larger when low circulation dominates over the high circulation.

    It turns out that the differences in low circulation (indexed by S) and the ratio of low to high circulation (indexed by D) equally contribute to reduced low cloud.  However, the model spread relative to S is not biased relative to observations.  In contrast, the model spread relative to D is, with models consistently underestimating D, that is, the ratio of low to high circulation.  Consequently it is differences in D that explain why low climate sensitivity models are probably inaccurate (according to Sherwood et al).  That is probably why Sherwood in the video concentrates on explaining the factors relevant to D, essentially ignoring S.  If you find all this confusing, it is probably better to also ignore S, and simply rely on Sherwood's description as transcribed above.

    1 0
  13. tcflood @8, strictly speaking it does take thousands of years to reach ECS, but in practise nearly all of the Equilibrium Climate Response is achieved in about 200 years.  To confuse things, however, by two hundred years you are starting to feel some of the effects of slow feedbacks, such as melting ice sheets and changes in vegetation.  Arguably we are already feeling some of those effects.  Consequently it is not safe to assume the change in temperature in human time scales is limited by the ECS.

    For practical purposes, if we cease all emissions, then temperatures will not rise much above the TCR to the peak CO2 concentration - but will not fall much below it for thousands of years either.  However, if we retain emissions at just 9% of current levels, CO2 concentrations will not fall and we will face the equilibrium climate response within two centuries, and temperatures will keep on rising slowly for thousands of years to come.  It follows that the relevance of the ECS depends on our future policies, and therefore it is at least relevant in helping us determine those policies.  

    0 0
  14. Bob Lacatena @10 and 11:  

    Thanks for your responses.

    I must say, though, that I found rather perplexing your introduction of the concept whose generation we care about into an answer to what I thought was a fairly straightforward technical question.    

    Tom Curtis @13: 

    Thanks for your helpful response.

    Just to clarify my question, let’s assume some arbitrary, hypothetical numbers. Suppose the globally averaged surface air temperature in 1850 were 286 K, and [CO2] were 280 ppm. Then in, say, 2050 the [CO2] were 560 ppm and that [CO2] were to remain constant at that value until year 3000 (preposterous, I know).  If one postulates an ECS (primary effects and all feedbacks, fast and slow) of, say, 4 degrees, what does this imply for the likely global SAT in, say, 2100?   I suppose that one might react differently if models projected an SAT of 290 K in 2100 versus in 3000. Whenever I read papers or discussions about the ECS, I am always left with this lack of clarity.  Perhaps it’s just me.

    0 0
  15. Thanks, Tom; those quotes really help. The one I am still left puzzled by is from the second quoted paragraph in your #12: "The air allways rises up to 10 or 15 kms, and so these models have been predicting a lower climate sensitivity..."
      If the air in these models always rises to this height bringing moisture up that high to form high clouds, shouldn't that bias those models toward a higher climate sensitivity, since it is the high clouds that do more heating, while the low clouds reflect sun more effectively and keep in less heat?

       Thanks again for your continued efforts in helping us undestand this important paper. What is your source for the second paragraph in 13. I'd like to point this out to people at other sites, but it would be nice to be able to site somthing more convincing than "a poster named Tom on another site."

    0 0
  16. Wili, re high clouds... No, because it's not an all or nothing thing, and it's not discrete.  We're not talking about a "block of moist air" that either rises 10km up, taking all of it's moisture with it, or not.  What is involved instead is the parameterized estimation of how much moisture rises to what altitude at what temperature, how much remains as vapor, how much condenses into cloud, and how much precipitates out.

    The models in which air predominently rises 10km and higher does not take all of the moisture with it (remember, the ability of air to hold moisture drops with temperature, which in turn drops wih altittude).  In those models, instead, the moisture precipitates out or is distributed (based on whatever criteria/profile the model may use) throughout the atmospheric column, leaving some for low clouds, some for middle clouds, and some for high clouds.  More moisture (in a warmer climate) will then increase clouds at all levels, a positive feedback for high and middle clouds, but a negative (reflecting more sunlight) for low level clouds.

    In contrast, a model which better simulates up and down drafts will leave the lower levels of air in the boundary layer (2km) dryer, leading to more cloud formation at middle and high levels (a positive feedback) and less low cloud formation (another positive feedback).

    1 0
  17. Additional note... some air (and moisture) will obviously also be transported horizontally, which is a major mechanism in the climate system for transporting moisture from the equator poleward.  So again, it's not as simple as more clouds or less clouds in a column.

    0 0
  18. willi @15, other than the first quote, which is a transcript from the video, all my quotes were from the actual paper.  I think the most important one is that first one, ie, the transcript, as it gives the clearest statement of the mechanism involved, which appears to have been giving some people (including myself) difficulty understanding.  However, the paper deals with the issue in more detail than the does Sherwood's discussion on the video.

    With regard to your first question, I have nothing to add to Bob Lacatena's answer. 

    0 0
  19. Bob at #16 wrote: "the moisture precipitates out or is distributed (based on whatever criteria/profile the model may use) throughout the atmospheric column, leaving some for low clouds, some for middle clouds, and some for high clouds."
       Thanks tons. That helps a lot. As usual, these things are more complex than what can easily be squeezed into a two minute video.

    0 0
  20. Tom, so your paragraph at #13 that goes:

    "For practical purposes, if we cease all emissions, then temperatures will not rise much above the TCR to the peak CO2 concentration - but will not fall much below it for thousands of years either. However, if we retain emissions at just 9% of current levels, CO2 concentrations will not fall and we will face the equilibrium climate response within two centuries, and temperatures will keep on rising slowly for thousands of years to come. It follows that the relevance of the ECS depends on our future policies, and therefore it is at least relevant in helping us determine those policies."

    is a direct quote from the Sherwood paper?

    0 0
  21. wili,

    Tom was not quoting Sherwood in that comment.  Sherwood does not deal with emissions or the results of different emissions scenarios.  His paper deals solely with models, low cloud cover, and climate sensitivity.

    0 0
  22. wili @20, Bob @21 is correct.  You will find that my quotations of other people are always enclosed in inverted commas, and typically indented unless they are quotations of less than a full sententence.  Consequently my quotations should be easy to distinguish from my own comments.

    With regard to the 9% emissions comments, that comes from reflection on implications of various models of the carbon cycle, as for example, the geocarb model placed online by David Archer.  

    Using that model, if you set the "transition CO2 spike" to 5000 Gt C, and the simulation "CO2 degassing rate" to 70 x 10^12 mol/yr *, leaving all other values at the default values, you will see that the initial peak is 2394 ppmv of CO2, and that it falls to 1326 ppmv at 650 years, before rising to 77,095 ppmv (ie, 7.75 of the atmosphere) at a million years.  70 x 10^12 mol/yr is 9% of current anthropogenic emissions plus ongoing natural emissions.  The curve is not linear indicating the value will stabilize, but it clearly still rising at a million years so has some time to go for stabilization at a million years.

    The Geocarb model dumps its CO2 into the atmosphere as a single pulse, making it hard to model ongoing releases.  You can partially model those releases using the simulation "CO2 degassing rate", but are constrained to a single value rather than an increasing value.  In real life, the gradualy increasing emissions since c1850 has lead to a situation where 55% of emissions if we count industrial emissions only (ie, CO2 from fossil fuels or cement manufacture), or 44% from all anthropogenic emissions including land use changes such as deforestation has been retained in the atmosphere.

    Taking the values above, and the 272.6 ppmv 'natural CO2 concentration' (scare quotes rather than quotation**) we see that the increase in atmospheric concentration  for a 5000 Gt C slug of CO2 is 2120 ppmv.  The increase at minimum concentration is 1055 ppmv, ie, approx 50% of the overall increase.  That is, my estimate over estimates the standard of CO2 not falling.  In actual life it will do worse than that because as we approach 5000 GtC the percentage of CO2 retained in the atmosphere will increase significantly.  On the other hand, it does considerably better than that (and worse for us) at 1000 GtC, the level of emissions we notionally should not exceed to keep the global temperatures increase below 2 C. 



    (* Note that the units are 10^12 mol/yr, so you only need to type 70 in the box.

    ** The 'natural level of emissions' used in the model in fact includes a significant level of emissions from human agricultural activity and pre-industrial fossil fuel and cement use.)

    0 0
  23. Tom Curtis @13

    Layperson's question, where did you get the 9% of CO2 emissions = no ppm increase from?

    I was under the impression that around 45% of our CO2 emissions are absorbed by the natural carbon cycle, so I always figured if we cut back to that level then ppm will stop increasing.

    0 0
  24. Sorry Tom, didn't see your reply to Willi @20. I'll take my question as answered :)

    0 0
  25. macoles, there is one estimate of consequences of cutting back CO2 emissions summarized on RealClimate.

    0 0
  26. Tom Curtis @ 13

    If nearly all of the ECS response is attained at a couple hundred years, then one could use the TCS and ECS pretty much interchangeably.  This was the impression that I had because of the apparent lack of distinction in common use in discussions. 

    But the reason I focused on the hydrosphere @ 8 is because of the “pause” in SAT over the last 15 years being explained by supposing that periodic oscillations in ocean-atmosphere thermal coupling can lead to faster than usual heat transfer from the lower troposphere into the oceans below 700 m or so. 

    Does periodic slowing of SAT increase by periodically enhanced AO coupling imply that those models that don’t do the AO coupling especially well could significantly overestimate the rate of SAT increase over, say, a 50-100 year period?  So could the TCS in fact be significantly smaller than the ECS?  Could this have led to an over-estimation of likely rate of SAT heating in the 21st century?  

    0 0
  27. Tom Curtis@22,

    You modelled your scenario in GEOCARB model a little bit inaccurately. Originally, you said @13:

    if we cease all emissions, then temperatures will not rise much above the TCR to the peak CO2 concentration - but will not fall much below it for thousands of years either. However, if we retain emissions at just 9% of current levels, CO2 concentrations will not fall

    (emphasis yours)
    It mean that your scenario is to cease the CO2 emissions now but keep the 9% residue of CO2 emissions (70E12 mol/yr as "degassing simulation") forever. Such scenario translates to the Transition CO2 Spike of sth like 500GtC (cumulative emissions to 2011) in GEOCARB. Yours 5000GtC input is unrealistic.

    So, with 500GtC spike input and degassing simulation of  70E12 mol/yr, the GEOCARB output is 505ppm CO2 spike (obviouly higher than today's real value because of shorter timeframe - 50y - of release in GEOCARB) but the minimum CO2 reached is 455ppm (in 400y), therefore somr than 50ppm higher than today. So I think, according to your argument, the residual degasing rate does not need to be as high as 9% of the current rate, so that CO2 concentrations "will not fall". In fact, I modified the degassing simulation of 40E12 mol/yr (5% of current levels) and I got the minimum CO2 of 398ppm in 700y, which is incidentally the current level.

    So I argue that CO2 degasing of just 5% of current levels, if kept indefinitely, will ensure that the current 400ppm "will not fall". Further, please note that this 40E12 mol/yr degassing includes the volcano output (7.5E12 mol/yr spinup parameter) estimated from the current geo configuration. Therefore, the "human C residual outpout" in this scenario is only 32E12 mol/yr, or 4% of current levels.

    For those who want to understand how the numbers are spun by GEOCARB on a shorte and long timescale, check out the rock weathering simulation by choosing "Silicate Thermostat" as an output graph. Before the simulation (years -100 to -50), the silicate weathering rate (WeatS) is in balance with volcano degassing (Degas), which is equal to the spinup value of 7.5. That equilibrium results in 272ppmCO2. If you show "Silicate Thermostat" for 1 million years, you sea that WeatS still would not catch up with Degas (33.7 vs. 40 in my simulation), therefore CO2 in the A would still be rising causing T rise and slowly speeding up WeatS. Such scenario (CO2 degassing at 5 times the natural level for 1My) has nothing to do with AGW and I'm not a geologist to judge how likely it is (i.e. if enough C and the mechanisms of its elevated level of release) exist in the system. It is just to show how miniscule CCycle changes are in the long-term natural system, as opposed to the "disaster-like", abrupt anthropo disturbance.

    0 0
  28. chriskoz @27, I quite agree, and am happy to see your expansion on my point that:

    "On the other hand, it does considerably better than that (and worse for us) at 1000 GtC, the level of emissions we notionally should not exceed to keep the global temperatures increase below 2 C."

    My original point was that for any reasonable estimate of cumulative emissions by 2100, ongoing emissions after 2100 of just 9% of current emissions will be sufficient to prevent CO2 levels from falling.  I did not claim they would not rise.  As also noted @22, that level is sufficient for an effectively indefinite rise in CO2 levels in the very long term (by human time scales).

    0 0
  29. As an addendum to my @28, and Chriskoz's @27, increasing background emission rates by 1.51 x 10^12 mol per year (2.4% of the current increase of emissons over background levels) with a 0 Gt C spike results in a CO2 concentration of 400 ppmv after 1 million years.  That is an excess emissions rate of 0.02 Gt C, and represents a reasonable estimate of the maximum safe emissions in the very long term.  For all intents and purposes, that means we need to target 0 net emissions in the long run.  An intermediate target less than 5 - 9% of current emissions will be necessary to ensure we avoid the full ECS response to increased CO2.  I am sure Chriskoz and I will agree that above that is madness, and the closer to 0 net emissions the better.  

    0 0
  30. Macoles, 24,

    Your understanding is incorrect.  The "natural carbon cycle" is just that, a cycle, not a reservoir.  About half of human emissions go into the ocean, causing acidification which may turn out to be as or more dangerous than climate change.  A big chunk goes into the atmosphere.  Most of the rest goes into expanded vegetation.

    It can't and doesn't just disappear.  It took nature hundreds of millions of years to bury it in the ground.  That won't happen easily, or quickly.  There are some mechanisms by which carbon will be deposited in the deep oceans, but that will happen very slowly.

    worse yet, we can't even count on it continuing to go into either the ocean or vegetation.  As the ocean warms, it's capacity to absorb CO2 is reduced.  Eventually, if it warms enough, it may release some of that absorbed CO2 -- a positive feedback.  The same goes for vegetation.  While for a while, it may show more growth due to mildly warmer temperatures, increased precipitation, and higher CO2 levels, that's hardly a permanent trajectory.  Eventually, expansion of the deserts and droughts, especially if it happens too quickly, will reverse some or much of that growth (worst case would be, for example, the transition of major parts of the Amazon to savanna).  The subsequent release of carbon is yet another positive feedback.

    If we reduce emissions, there is no reason to think one particular sink (atmosphere, ocean, vegetation) to absorb more than another, and as the planet will continue to warm until it reaches a new equilibrium temperature, any of those positive feedbacks listed above could still come into play.

    Interestingly, even if we found a magical, technological way to suck CO2 out of the atmosphere, the ocean would still replace some of it, and those positive feedbacks might still kick in.  It's a very dangerous game that we're playing... Carbon Roulette.

    1 0
  31. Thanks Bob, I hadn't concidered that the ocean soaking up the better part of that 45% CO2 would require ongoing acidification. I foolishly just assumed that if the ppm reached a manner of equalibrium then its effects would too.

    Thanks aswell to both Toms for the links and explanations.

    0 0
  32. @12 Tom Curtis

    Thanks Tom for your explanation. I’m certainly less confused now, although I will probably still have to draw what you said as I’m more visually oriented.

    0 0
  33. The effect describes sounds like a climate system that is not in equilibrium i.e. the land surface and atmospheric temperatures warming through greenhouse gases at a faster rate than the sea surface temperatures can equilibrate to. In a steady state climate, either during brief periods of global temperature slow-down or upon curbing carbon dioxide, methane, CFCs and other greenhouse gases the sensitivity from this cloud effect may reverse.
    It seems to me that the system described is a climate state where the specific water vapour concentration in the atmosphere is inefficiently distributed into the atmosphere - indicating also that there is a longer than anticipated lag-time for atmospheric moistening and also delayed upper tropospheric moistening. As water vapour is a greenhouse gas, this may reach a critial point in time limit at which the rate of warming is buckled before rising sporadically in a staged approach. Pushing the Earth's hydrological cycle into an imbalance can only cause more weather extremes and enhance internal climate variability further.

    0 0

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 2024 John Cook
Home | Translations | About Us | Privacy | Contact Us