A bit more about clouds

A few years ago I posted a video by Andrew Dessler that was discussing whether or not Equilibrium Climate Sensitivity could be less than 3oC. The bottom line was that the best estimate for ECS is about 3oC. Given that we’re quite confident about water vapour feedback, lapse rate feedback, and ice albedo feedback, the main way in which ECS could be much lower than this (say < 2oC) is if there were a strongly negative cloud feedback. Cloud feedbacks are probably the feedbacks about which there is the greatest uncertainty, and so this is not necessarily impossible.

More recently, however, I highlighted a TED talk by Kate Marvel that discussed her work on clouds. The basic conclusion was that the observations are pointing towards clouds acting to intensify the warming – they’re a positive feedback. In fact, Kate Marvel indicates that there is no observational evidence that clouds will substantially slow down global warming.

Credit: Zelinka et al., Nature, 2017.


The reason I’m writing this is because there is a new Nature Commentary called Clearing Clouds of Uncertainty by Mark Zelinka, David Randall, Mark Webb and Steven Klein. Their commentary is really a summary of our recent understanding and – as illustrated by the figure on the right – they conclude that the evidence is converging on the cloud feedback likely being positive. The circles indicate the multi-model average feedback, and the coloured lines show the across model standard deviation. The thin grey lines extend to the model extrema. Essentially, the total cloud feedback is probably positive and has a likely range from about 0.2 Wm-2K-1 to about 0.7 Wm-2K-1.

The implication of this – as Andrew Dessler highlighted – is that it is unlikely that the ECS can be less than 2oC. We have a pretty good understanding of the other feedback processes (water vapour, lapse rate, and ice albedo) and the cloud feedback being positive strongly implies an ECS > 2oC. Some energy balance estimates of climate sensitivity suggest that the ECS is more likely less than 2oC, than above 2oC and – as I think I may have suggested before – I do think that those who promotes these results should put some effort into explaining how this is possible.

If water vaour, lapse rate and ice albedo, by themselves, suggest an ECS > 2oC and if cloud feedbacks probably amplify this, then how can the ECS be less than 2oC? My view is that this is simply because these energy balance estimates are a bit too simple and don’t necessarily capture all the relevant processes. I would, however, be happy to hear some kind of physically motivated argument for ECS < 2oC.

Advertisements
This entry was posted in Climate sensitivity, ClimateBall, Global warming, Research, Science, The scientific method and tagged , , , , , , , , . Bookmark the permalink.

32 Responses to A bit more about clouds

  1. John Hartz says:

    ATTP: Out of curiosity, is fog considered to be cloud?

  2. JH,
    I’m not sure, but I don’t think so. It’s probably too close to the surface to be all that radiatively relevant.

  3. John Harz, in reality or in models?

    There is no clear distinction between clouds and fog, but fog tends to have larger water droplets and they naturally form on the ground when it cools due to radiative heat loss. However, sometimes the base of the fog lifts during the morning and at least the Germans speak of “high fog” (Hochnebel) in that case. I would tend to say that once the base is no longer on the ground, it becomes a cloud, but likely a cloud with rather large droplets.

    In models there is mostly no distinction between clouds and fog. Some newer models do do this and have a higher sedimentation rate for fog droplets. This is mostly used in regional weather prediction models for traffic safety, especially around airports. It is more common, however, to have separate simplified fog models that use the weather of the weather prediction as input and can be tuned to past fog observations. This is simpler and I have yet to see a convincing example that the complex models (which are not tuned) embedded in the weather prediction models do a better job.

  4. Thanks, that’s a far more detailed – and informed – response than mine.

  5. A couple of tweets from the author of the Commentary. Basically, most lines of evidence point towards cloud feedbacks being positive.

  6. JCH says:

    Great that Lindzen takes things like this so well.

  7. Welcome.

    Finally found time to read the overview article itself. I may not be the best judge of it, but I would say it is quire readable and worthwhile to read.

    As young researcher I was working on clouds. In the middle of it, one often had the feeling that we did not make much progress. Now working on something else and hearing less about it, I almost had the impression we had given up, that the clouds are mightier than science. So it is nice to read that we are making progress and are at the stage where we think we know the sign of the feedback. The uncertainties are still large, but that is important progress.

    I am sure that when I retire there will still be enough clouds problems to return to the topic of the good old days as a hobby. I am sure the young people will appreciate the wisdom of a retired electrical engineer.

  8. angech says:

    Thanks Victor,
    Salient points in that that need clarification,
    Clouds on average produce a 46 W/m 2 reduction in incoming energy hence this is evidence that clouds produce an overall cooling effect or negative feedback in general.
    There is more I did not quote yet which ameliorates this but this is an important point at this first step. As this is one of the figures we use to estimate our global heat balance and it can I presume, be more variable but less so in a warming world which produces more water vapour and hence more cloud.
    ATTP ” they conclude that the evidence is converging on the cloud feedback likely being positive.”
    This comment must therefore be directed at the increase in clouds, not their current status as the article demonstrates an original strong negative feedback (18W/m2).
    Part of the argument is dependent on where the clouds are formed.
    High level clouds are said to increase GHG effects, positive feedback whereas ( presumably) low level clouds increase albedo more than they trap heat hence negative feedback.
    Hence the argument drills down into how much more cloud at each level and the difference in albedo v GHG effect at each level going up.
    SInce the first effect is primary ie albedo and since this effect is negative any increase in cloud cover should have a stronger negative than a positive effect.
    Re the definition of a cloud/fog/watervapor. Would it be true that all water molecules in the air for most practical purposes are basically cloud, that is they are all capabilities of reflecting light or is it only when they aggregate into groups. If so how many water molecules does it take to makereflective group and does all water in tHe air exist in aggregates only anyway.
    Hope that your contact might answer some questions if they prove interesting or controversial enough.

  9. izen says:

    @-angtech
    “SInce the first effect is primary ie albedo and since this effect is negative any increase in cloud cover should have a stronger negative than a positive effect.”

    Clouds have both effects, increasing albedo and increasing the GH warming.
    Why would the albedo negative effect be larger than the cloud GH positive effect if they increase ?
    The warming (positive) effect of increasing clouds is continuous, the albedo effect only works half the time.
    The time we call ‘daytime’.

  10. Angech, that cloud have a cooling effect in the current climate is different from them having an amplifying effect when the climate *changes*.

    Let’s take a simpler system: Arctic sea ice has a cooling effect by reflecting sun light. When the climate changes, the Arctic sea ice becomes less, this cooling effect becomes less and the Earth warms more.

  11. paulski0 says:

    angech,

    ATTP ” they conclude that the evidence is converging on the cloud feedback likely being positive.”
    This comment must therefore be directed at the increase in clouds, not their current status as the article demonstrates an original strong negative feedback (18W/m2).

    Positive feedback in this case refers to enhancement of warming rate due to changes in cloud behaviour in response to temperature increase.

    The -18W/m2 figure refers to the net present day radiative effect of clouds, relative to a hypothetical Earth without clouds (all else remaining the same). I guess you could argue that clouds wouldn’t exist at 0K temperature, so the “warming” to present day 288K temperature has resulted in a -18W/m2 negative feedback, but the temperature at which clouds can form is so far below 288K that it makes no sense to extrapolate a continuing increase in cloud amounts with temperature.

    It’s actually fairly easy to dispel any “intuitive” idea that cloud amounts must increase with temperature at present day levels by looking at seasonal cycles. Certainly in the Northern Hemisphere mid-latitudes, Summers are considerably less cloudy than Winters.

    Putting into numbers AR5 concluded with a best estimate of 0.6W/m2/K for net cloud feedback. That means a 1K warming from now would result in a change from -18W/m2 to -17.4W/m2 cloud radiative effect. Don’t know if it’s been done, but might be interesting to produce a diagram giving some idea of how Earth’s net cloud radiative effect, total cloud fraction, low cloud fraction varies with temperature from 0K up to 300K.

  12. Magma says:

    That’s a nice commentary/short review in Nature Climate Change, open-access and well worth reading.

    I have the strong impression that much of the contrarian opinion on cloud feedback originates from the fact that when one is outdoors on a sunny day and a cloud passes in front of the sun, ground level insolation and perceived temperature quickly drop. Coupling this to the undeniable facts that it still gets cold in mid-latitude winters and major coastal cities are still above sea level, it is but series of short logical steps to the “all scientists are liars and climate change is a hoax to bring about one world government” held by many of these rather dim individuals.

  13. Pingback: Collapse Daily 081017 | Loki's Revenge

  14. Eric Michel says:

    Just a quick note of thanks. Not nearly well versed enough to comment on the material, but enjoy reading and learning bits and pieces.

    Eric

    On Sat, Oct 7, 2017 at 3:02 PM, …and Then There’s Physics wrote:

    > …and Then There’s Physics posted: “A few years ago I posted a video by > Andrew Dessler that was discussing whether or not Equilibrium Climate > Sensitivity could be less than 3oC. The bottom line was that the best > estimate for ECS is about 3oC. Given that we’re quite confident about water > v” >

  15. Pingback: "Sales temps pour les glaciers" - Ocasapiens - Blog - Repubblica.it

  16. angech,
    I don’t like having to justify moderation decisions, but I will illustrate this just this once. You said

    SInce the first effect is primary ie albedo and since this effect is negative any increase in cloud cover should have a stronger negative than a positive effect.

    Well, this is wrong, as others have indicated. Yet, you said it as if you understood this well enough to make a strong claim. I don’t really see why I should be expected to post comments that are certain, but wrong.

  17. Andrew Dodds says:

    Magma –

    I must be more cynical than you. My impression is that the skeptic line was ‘Since cloud feedbacks are highly uncertain, we can loudly claim then to be a negative feedback and it’s very hard to easily refute us’..

  18. Eric,
    Glad it’s been useful to some 🙂

  19. angech says:

    ATTP,
    This is a discussion of clouds precisely because they are a very important part of the climate puzzle.
    “Clearing Clouds of Uncertainty by Mark Zelinka, David Randall, Mark Webb and Steven Klein. Their commentary is really a summary of our recent understanding and – as illustrated by the figure on the right – they conclude that the evidence is converging on the cloud feedback likely being positive.”
    From FAR 1990 positive but cloud feedback represented the largest source of uncertainty in climate sensitivity among atmospheric models.
    To SAR Second Assessment Report; 1995), more climate models were predicting the mass of cloud liquid and ice, and generally finding negative cloud opacity feedbacks, albeit of widely differing strengths. The report concluded that it was not possible at that time to judge the sign of the net cloud feedback.
    TAR 2001 …the sign of [the cloud] feedback remains unknown.
    AR4 2007 it is not yet possible to assess which of the model estimates of cloud feedback is the most reliable
    AR5 “The sign of the net radiative feedback due to all cloud types is…likely positive”
    BUT Cloud opacity feedback “is highly uncertain”

    You say,
    “I don’t really see why I should be expected to post comments that are certain, but wrong.”
    I repeated a fact, “Averaged globally and annually, clouds cause cooling ” from the article overview. I said in view of this,
    “any increase in cloud cover should have a stronger negative than a positive effect.”
    The current point of view says this is likely wrong but admit to high uncertainty still in areas like cloud opacity. There is a threefold variation in the global sensitivity parameter FAR 1990 due to
    differences in cloud feedback.
    This is a most important discussion which is not yet settled and needs open discussion.

  20. angech,

    BUT Cloud opacity feedback “is highly uncertain”

    It’s not uncertain enough for there to be a large chance of it making the cloud feedback negative (see the figure in the post). Also, if you look at the tweets from the lead author (which I posted earlier) then observations suggest it is less negative than models suggest.

    I repeated a fact, “Averaged globally and annually, clouds cause cooling ” from the article overview. I said in view of this,
    “any increase in cloud cover should have a stronger negative than a positive effect.”

    Except this is simply wrong. Just because the overall effect is negative does not mean that the response to some temperature change will be negative (i.e., that it will damp, rather than amplify warming). Therefore your claim that “any increase in cloud cover should have a stronger negative than a positive effect” simply does not follow from the fact that averaged globally and annually, clouds cause cooling.

  21. raypierre says:

    Yes, while we need longer and better satellite data to pin down cloud feedback, the record is already getting good enough to rule out a significantly stablilizing cloud feedback, which is the only thing, as you note, that could drive climate sensitivity below 2C. One among many examples is the paper from this year by my and my former student Ian N. Williams:

    http://onlinelibrary.wiley.com/doi/10.1002/2016GL072202/abstract

  22. Ray,
    Thanks, I hadn’t seen that paper.

  23. Michael Hauber says:

    Can we discern anything about cloud feedbacks from Venus vs Earth? I believe that Venus has a higher albedo than Earth largely due to clouds. Does this suggest that in the extreme case cloud feedbacks might average out to be negative? Of course cloud feedbacks could be +ve for temp changes from current earth to +20 C, and then negative from +20C to whatever Venus is….

    I’ve sometimes thought about this test for Lindzen’s Iris. I’ve always suspected that it could only work over a relatively small band of temperatures, but be impossible over a much wider temp change.

  24. Andrew Dodds says:

    Michael –

    When we reach temperatures and atmospheric compositions that allow for sulfuric acid clouds, that may start to be a negative feedback… I’m not sure it;’s comparable, though, because Venus does not have a ocean/cloud system, regardless of composition, so you can’t have higher temperatures causing more evaporation.

    For me, the paleo test shows that stabilizing feedbacks are unlikely – no matter what the mechanism, if we had stabilizers for the climate then we wouldn’t see so much variation in the paleo record.

  25. Andrew,

    For me, the paleo test shows that stabilizing feedbacks are unlikely

    I know what you mean, but I would clarify that there are feedbacks that mean that we will ultimately settle to new quasi-stable state. However, they do not act to damp out perturbations so that they are very small. All the evidence indicates that a perturbation equivalent to a doubling of atmospheric CO2 will ultimately lead to a temperature change of a few oC (probably about 3oC).

  26. verytallguy says:

    I would clarify that there are feedbacks that mean that we will ultimately settle to new quasi-stable state.

    Much confusion on this point comes from the convention in climate science of not including the Planck feedback.

    “Feedbacks” are net positive sounds as though our climate is unstable and will run away, however, by convention that does not include the Planck feedback.

    Including the Planck feedback means that feedbacks are always net negative.

  27. vtg,
    I agree. Often excluding the Planck feedback when discussing feedbacks does generate a great deal of confusion (especially amongst retired engineers 🙂 )

  28. russellseitz says:

    ATTP:re fog.
    “JH,I’m not sure, but I don’t think so. It’s probably too close to the surface to be all that radiatively relevant.”

    Please clarify why, as albedo is where you find it, and sea ice is the surface

  29. Russell,
    I think I was simply wrong.

  30. angech says:

    Michael Hauber says:
    “Can we discern anything about cloud feedbacks from Venus vs Earth? I believe that Venus has a higher albedo than Earth largely due to clouds. Does this suggest that in the extreme case cloud feedbacks might average out to be negative? Of course cloud feedbacks could be +ve for temp changes from current earth to +20 C, and then negative from +20C to whatever Venus is…”

    Earth cloud feedbacks change to negative from +14C.
    Albedo can be due to the right type of clouds, ice, vegetation [both not on Venus], and color of the composition of a surface or atmosphere.
    Venus has a higher atmospheric temperature than Earth due to GHG and proximity to the sun.
    “The atmosphere of Venus is mostly carbon dioxide, 96.5% by volume.”
    Its albedo is due to a different composition of cloud.
    Does it have a higher albedo? Yes.
    Mercury 6%
    Moon is 0.12 12%
    Earth is 0.37 or 37% , all those water molecules ?
    No water has a low albedo 10% but ice is 90%
    “air and seaclouds are made from water ice and have a high albedo.”
    Venus Albedo is due to clouds but not water clouds. “The atmospheric clouds primarily consist of sulfuric acid, which reflect the vast majority of sunlight that is incident upon them. This makes Venus the planet with the highest albedo in the solar system, with a value of 75%”
    Despite reflecting so much energy the stuff that gets through warms the GHG atmosphere up a lot hotter than earth.
    Atmosphere is much denser than earth , is this because CO2 is much heavier than H20?
    The upper layer of troposphere exhibits a phenomenon of super-rotation, in which the atmosphere circles the planet in just four Earth days, much faster than the planet’s sidereal day of 243 days. The winds supporting super-rotation blow at a speed of 100 m/s (~360 km/h or 220 mph)[4] or more. Winds move at up to 60 times the speed of the planet’s rotation, while Earth’s fastest winds are only 10% to 20% rotation speed. Doubt that clouds behave in the same way as on earth given these facts.

  31. angech says:

    On a sad note check out Stoat blog closing down or moving sideways. A bit of support at his site would be good.

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s