Sea surface warming and cloud feedbacks

I wrote a post a while agoe about a paper that was suggesting that the reason for the difference between the observationally-based estimates for climate sensitivity, and other estimates, was that the pattern of sea surface temperatures can produce different system heat uptake rates for the same change in temperature. A recent paper by Zhou, Zelinka & Klein called Impact of decadal cloud variations on the Earth’s energy budget tries to explain why this happens. They argue that

the global mean cloud feedback in response to decadal temperature fluctuations varies dramatically due to time variations in the spatial pattern of sea surface temperature.

The figure below shows the basic result from a set of model run. The left-hand panel shows the net feedback response, while the right-hand panel shows the cloud feedback only. Essentially if you run a model with prescribed sea surface temperatures you get much more negative cloud feedbacks than the mean from long-term warming runs with either uniform, or patterned, sea surface temperatures. The suggestion is that over time we would expect the feedback response to tend back towards the mean, but we just happened to have experienced a period during which it was more negative than the mean.

Credit: Zhou, Zelinka & Klein (2016)

Credit: Zhou, Zelinka & Klein (2016)

Credit: Mauritsen (2016)

Credit: Mauritsen (2016)

Thorsten Mauritsen has a nice News and Views about this paper. It includes the figure on the right that illustrates what is thought to be happening. When the sea surface temperatures are relatively cool, there will be an inversion at 1-2km and clouds form below this inversion level. If the warmer regions warm more than the cooler regions, the inversion gets stronger and more low-level clouds form. Low level clouds tend to reflect sunlight that would otherwise have heated the Earth and, therefore, more low level clouds means a more negative cloud feedback.

So, it seems that one reason for the mismatch between observationally-based climate sensitivity estimates and other estimates is that the pattern of sea surface warming that we’ve actually experienced has produced less warming than would be expected from some kind of typical warming pattern. On the other hand, Thorsten Mauritsen points out that models rarely produce the pattern that was observed, which could suggest that they don’t properly represent variability, or don’t fully represent the response to increasing greenhouse gas concentrations.

This may also be related to the issue of whether or not we should regard the models as truth-centred, or not. It would certainly seem that over periods during which the variability can be comparable to the forced response, we shouldn’t assume that the observations should be comparable to the multi-model mean. Anyway, that’s all I was going to say. This seems like an interesting result and it does seem as though we’re starting to get a better understanding of how variability can influence the feedback response and how we then warm on decadal timescales.

This entry was posted in Climate change, Climate sensitivity, Research, Science and tagged , , , , , , . Bookmark the permalink.

37 Responses to Sea surface warming and cloud feedbacks

  1. MarkR says:

    I think this is a really nice paper, they link physics and satellite measurements together and explain things clearly.

    Something weird has been going on in the Pacific. It could be aerosols helping (there’s a reference in the paper to that), or it could be just an extreme natural cycle (record-breaking trade wind strength, at least since the early 1900s), or it could be a response to global warming. Or some combination. But our best implementation of the physics (aka climate models) says CO2-driven global warming should show the opposite response with weakening trades (e.g. ).

    It looks like a lot of it is all linked into the Pacific Decadal Oscillation. When that swings into positive territory we should see some of the clouds disperse and more heating. This is also potentially the sort of long-term radiative response to internal variability that people like Spencer (iirc) have been talking about. Usually I hear this suggested as a way to explain the observed global warming so that human activity looks smaller, but it looks like in reality it’s the other way round and natural variability has been hiding some of the warming we’ve caused.

  2. russellseitz says:

    Radiative forcing driven SST variation does not end at sea level; it is driven by sea surface and mixed layer albedo changes arising fom both marine biology ( bright algal and coccolith blooms) and mechanical wave action , from whitecap generation to secondary microbubble generation from jetting as surface bubbles burst.

    These effects range in scale from ocean-wide undershine from an ambient population of ~3 x 10^4 coarse ( 20 to 60 micron) microbubbles , to gale driven circumpolar brightening in the southern hemisphere and annual E. Hux blooms in Arctic seas.

  3. JCH says:

    Mauritsen’s diagram?

  4. JCH,
    Your figure doesn’t seem to want to work.

  5. Mark,

    but it looks like in reality it’s the other way round and natural variability has been hiding some of the warming we’ve caused.

    Indeed, it does seem that this might be the case. The next decade might be interesting.

  6. JCH says:

    Try again:

    My question is, is the Mauritsen diagram indicative of where these clouds are doing their thing, the Eastern Pacific, or is this taking place globally?

  7. JCH says:

    Because if it’s the Eastern Pacific, the next decade is already interesting:

    It’s a regime shift… the ramp up of the positive phase of the PDO… see 1930-1943, 1976-1985.

    Look at those PDO ramp-up warming rates… they’re sky high.

    MarkR, didn’t see your comment last night – I’ve been saying approximately that, what you are saying, on Climate Etc since around 2011-2012, which is when I guessed the PDO was on the verge of going positive… because they is no 65-cycle in the 20th-century record… there is one cycle… the PDO, and it diverged ~1985. This means TCR is likely over 2℃, and ECS could be 4℃.

  8. BBD says:

    Mauritsen (2016) seems to go rather beyond a discussion of decadal variability, with an opening line that reads:

    The slow instrumental-record warming is consistent with lower-end climate sensitivity.

    Usual bloody irritating paywall – can you say over what period Mauritsen thinks the instrumental record shows slow warming which is consistent with lower-end sensitivity?

  9. BBD – Clouds Cooled the Earth

    Climate scientists broadly agree that Earth’s equilibrium climate sensitivity — the global warming that occurs a long time after the atmospheric carbon dioxide has been doubled — is likely to be between 1.5 and 4.5 K. Estimates based on climate models often favour the upper end of the range [1,2], whereas estimates based on instrumental-record warming tend to arrive at the lower end [3,4].

    The footnotes refer to:
    1. Fasullo, J. T. & Trenberth, K. E. Science 338, 792–794 (2012).
    2. Sherwood, S. C. et al. Nature 505, 37–42 (2014).
    3. Otto, A. et al. Nat. Geosci. 6, 415–416 (2013).
    4. Lewis, N. & Curry, J. A. Clim. Dyn. 45, 1009–1023 (2015).

  10. BBD,
    The full abstract is:

    The slow instrumental-record warming is consistent with lower-end climate sensitivity. Simulations and observations now show that changing sea surface temperature patterns could have affected cloudiness and thereby dampened the warming.

  11. BBD says:

    Can anyone who has read the study indicate over what period Mauritsen thinks the instrumental record shows slow warming which is consistent with lower-end sensitivity?

  12. JCH says:

    The slow instrumental-record warming is consistent with lower-end climate sensitivity. …

    Well, it is consistent with that… if you suspend common sense, and do things like hang out with members of the libertarian party, hate greens, hang out with Lamar Smith, keep lists of bad boy scientists, and celebrate the theft of emails.

  13. It goes on to say:

    It is perhaps tempting to think that the patterns of sea surface warming that have such a profound effect on cloudiness are an expression of natural variability associated with the recent slow warming (also termed hiatus) period — the years 1998–2012 when global mean surface temperatures warmed less than expected.

  14. BBD says:

    So it’s another ‘hiatus’ paper claiming low S?

    Thought we were over that now.

  15. BBD says:

    the years 1998–2012


  16. BBD,
    No, I don’t think so. I think it’s suggesting that variability can produce surface warming at a rate that is slower than might be expected.

  17. BBD says:

    I think it’s suggesting that variability can produce surface warming at a rate that is slower than might be expected.

    Ah, okay… so thereby perhaps fooling the incautious researcher into thinking that S is low, etc.

    Thanks for clarifying. The abstract is so brief as to be equivocal. I have the stick the right way round now.

  18. BBD,
    To be fair, his article does have a section that I don’t fully understand. He points out that the observational estimates are not sensitive to whether or not the “hiatus” period is included and that when models are implemented with the observed warming, they indicate that the “slowdown” should have started earlier than it did (i.e., the period during which there were more low-level clouds started earlier than the mid-1990s). I don’t quite get the significance of this, but it seems to be suggesting that the models don’t properly represent the response to increasing GHGs or that they under-represent internal variability.

  19. Thorsten Mauritsen says:

    Thanks ATTP for a nice framing; I fear though that the above discussion is slightly confused.

    I would like to point out that the main new finding in Chen Zhou’s great paper is that cloudiness in the East Pacific have increased over the period 1980-2005 (overlapping period of available observations and CMIP5 model runs) in a way that is consistent between observations and models, and that this is something that can be understood simply from the pattern of warming.

    This is a fantastic achievement, nothing less, but it says nothing about the origin of the pattern.

    So is it an expression of variability? That is certainly possible, but from what we know it is unlikely to be “just the hiatus”; the argument is simply that the deviation from the expected feedback in models exposed to real-world SSTs start much earlier. In Chen Zhou’s simulations in the 1970’s, in Gregory and Andrews (GRL 2016) a bit earlier. There are also other options, for instance even slower timescale variability, time-dependent feedbacks, or negative cloud-circulation feedbacks, and this is simply what I am saying in my News and Views we need to keep in mind. It is too early to jump to conclusion based on the available evidence, but as I also state, I think we are close.

  20. Thorsten,

    I fear though that the above discussion is slightly confused.

    Yes, I did suggest that I was slightly confused 🙂

    I had slightly missed that the main finding was that the observed change in cloudiness matches the models and that this is due to the pattern of the warming, so thanks for clarifying that.

    As regards this

    the argument is simply that the deviation from the expected feedback in models exposed to real-world SSTs start much earlier.

    Do you have a sense if this could indicate that the expected feedbacks are too high (not negative enough) or do you think that the expected feedbacks are what we should expect on sufficiently long timescales?

  21. Thorsten Mauritsen says:

    No, I don’t have a sense, yet, but we are working on it ;o)

  22. No, I don’t have a sense, yet, but we are working on it ;o)

    Thanks, look forward to seeing the results.

  23. JCH says:

    So am I, but if they conclude it’s the AMO my head is going to explode.

  24. Chubbs says:

    I haven’t read the material behind a paywall. Hadcrut warmed at a relatively fast pace of 0.19C/decade between 1980 and 2005. So the paper is saying that warming would have been even faster without this feedback?

  25. So the paper is saying that warming would have been even faster without this feedback?

    I think it is saying that the warming would have been faster if this feedback had been less negative. However, we don’t really know (as Thorsten indicates) if it is simply a period with a more negative feedback response and that the mean over a longer time interval will be less negative, or if this more negative feedback better represents what we should expect.

  26. JCH says:

    Well, I’m not so cautious. It should have warmed faster. One clue, the entire system is warming faster now. Radically faster. The 5-year SLR rate for Jason 2 is 5.19mm per year.

    The cold phase of natural variability… the stadium wave in which Professor Curry holds a religious belief… has come and gone. 2005 to 2012 was exactly the same as 1903 to 1910.. the negative “spike” valley of the PDO.

  27. Pingback: Non-linear climate sensitivity | …and Then There's Physics

  28. JCH says:

    A paleo-perspective on ocean heat content: Lessons from the Holocene and Common Era

    …These records suggests that intermediate waters were 1.5–2 °C warmer during the Holocene Thermal Maximum than in the last century. Intermediate water masses cooled by 0.9 °C from the Medieval Climate Anomaly to the Little Ice Age. These changes are significantly larger than the temperature anomalies documented in the instrumental record. The implied large perturbations in OHC and Earth’s energy budget are at odds with very small radiative forcing anomalies throughout the Holocene and Common Era. We suggest that even very small radiative perturbations can change the latitudinal temperature gradient and strongly affect prevailing atmospheric wind systems and hence air-sea heat exchange. These dynamic processes provide an efficient mechanism to amplify small changes in insolation into relatively large changes in OHC.

  29. JCH says:

    Quark Soup has updated OHC through November 1st.

    I believe there has been a substantially smaller drop in OHC as the result of the 15-16 El Niño than was expected… even by AGW scientists.

    I was expecting a very small drop… and perhaps none at all in a weak El Niño.

    So how much of the energy required to result in a GISS anomaly for 2016 of ~1.02 ℃ came from stored energy in the oceans, and how much came sunlight during each day of the EL Niño interacting with the surface of the Eastern Pacific… perhaps without the interference of the clouds discussed the headline paper?

  30. BBD says:


    I remember Rob Painting raising potential issues with Rosenthal et al. (2013) and it may be that these are relevant to R17. This is just FYI – I’m not taking a position here 🙂

    BBD – The main problem with the Rosenthal paper is that they fail to consider the effects of obliquity on the ocean circulation. Their simplistic assumption is that the circulation has remained pretty much as it is now for the duration of the Holocene.
    A few older climate-modelling papers indicate that this in unlikely, and it has to do with the way that obliquity (Earth’s axial tilt relative to it’s plane of orbit) changes the wind-driven mixing of heat into the deeper layers, and with the speed of the wind-driven ocean circulation itself. Difficult to explain without graphics or animations, but models imply that the tropical thermocline warmed as obliquity increased up to the Holocene Climatic Optimum (HCO), and then cooled as obliquity decreased.
    I don’t doubt for a minute that the climate cooled from the HCO onwards, or that the forams may record this general warming up to the HCO, and then decline afterwards, but much of this is related to obliquity-induced changes of the circulation, not the actual heat content of the Pacific Ocean itself. The tropical ocean surface was actually almost 1°C cooler than present-day during the HCO , despite the tropical thermocline being over 1°C warmer. Same Pacific ocean, but in a very different state compared to today.

    Rob said a bit more later in the thread.

  31. JCH says:

    This is an example of what I am talking about…

    In studying ocean temperature data provided by NOAA covering the past several decades, the researchers found that surface temperatures for the Pacific actually decreased over the past ten years. Recognizing that the heat had to go somewhere, the researchers input the data into models that have been built to show global temperatures, wind movement and other meteorological data. The model showed heat building up in the western Pacific and then being carried by easterly trade winds through the Indonesian archipelago. Upon closer inspection, the team found that surface temperatures (down to 700 meters) in the Indian Ocean had in fact increased to the point that it could account for approximately 70 percent of the heat taken up by the atmosphere over the past decade.

    The finding by the team appears to be both good and bad news. The good news is that it adds credence to global warming theories—the bad news is that it means that it is possible that at some point in the future all that heat in the ocean could be released back into the atmosphere, creating a sudden temperature spike which would almost assuredly cause massive worldwide problems for those of us that caused the problem in the first place.

    It is very hard for the oceans to shed the extra energy accumulating in them due to the increase anthropogenic CO2 in the atmosphere.

  32. Pingback: The warmest year…again | …and Then There's Physics

  33. Pingback: Reconciling climate sensitivity estimates – part III, or IV? | …and Then There's Physics

  34. Pingback: Confounding ECS estimates | …and Then There's Physics

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.