Saturday, May 17, 2014

New paper shows cloud radiative effects are negative, not positive as assumed by climate models

A paper published yesterday in Climate Dynamics illustrates just a few of the large unresolved problems in modeling clouds, the largest source of climate modelling uncertainty. The authors attempted but were unable to reconcile cloud radiative effects with global precipitation and the atmospheric energy budget due to doubled CO2 concentration, unless the sign of cloud radiative effects is changed from positive [as falsely assumed by all IPCC models] to strongly negative. 

According to the paper, 
"because clouds are a large source of modelling uncertainty, we consider whether resolving errors in cloud simulation could reconcile modelled global mean precipitation trends of about 1–3 % K−1 with some estimates of observed trends of 7 % K−1 or more. This would require the radiative effect of clouds to change from one that increases atmospheric radiative absorption by about 0.5 Wm−2 K−1 to one that decreases it by −3.5 Wm−2 K−1 . Based on our results, this seems difficult to achieve within our current rationale for the tropics at least."
In other words, the net radiative effect of increased clouds from warming & increased evaporation of water vapor would be to decrease IR radiative absorption & increase IR radiative cooling of the atmosphere, an anti-greenhouse effect. Thus, IPCC climate models don't have either the magnitude or even the sign of radiative forcing from clouds correct, one of several reasons why the models greatly exaggerate warming and have been falsified at confidence levels of 95-98%+.

As Dr. Roy Spencer notes,
"The most obvious way for warming to be caused naturally is for small, natural fluctuations in the circulation patterns of the atmosphere and ocean to result in a 1% or 2% decrease in global cloud cover. Clouds are the Earth’s sunshade, and if cloud cover changes for any reason, you have global warming — or global cooling."
Climate Dynamics May 2014

The cloud radiative effect on the atmospheric energy budget and global mean precipitation

F. Hugo Lambert, Mark J. Webb, Masakazu Yoshimori, Tokuta Yokohata


Abstract
This study seeks to explain the effects of cloud on changes in atmospheric radiative absorption that largely balance changes in global mean precipitation under climate change. The partial radiative perturbations (PRPs) due to changes in cloud and due to the effects of the pre-existing climatological cloud distribution on non-cloud changes, known as “cloud masking”, are calculated when atmospheric CO2 concentration is doubled for the HadSM3 and MIROC models and for a large ensemble of parameter perturbed models based on HadSM3. Because the effect of cloud on changes in atmospheric shortwave absorption is almost negligible, longwave fluxes are analysed alone. We find that the net effects of cloud masking and cloud PRP on atmospheric absorption are both substantial. For the tropics, our results are reviewed in light of hypotheses put forward to explain cloud and radiative flux changes. We find that the major effects of clouds on radiation change are linked to known physical processes that are quite consistently simulated by models. Cloud top height changes are quite well described by the fixed anvil temperature hypothesis of Hartmann and Larson; cloud base heights change little, remaining near the same pressure. Changes in cloud geographical location and cloud amount are significant, but play a smaller role in driving radiative flux changes. Finally, because clouds are a large source of modelling uncertainty, we consider whether resolving errors in cloud simulation could reconcile modelled global mean precipitation trends of about 1–3 % K1 with some estimates of observed trends of 7 % K1 or more. This would require the radiative effect of clouds to change from one that increases atmospheric radiative absorption by about 0.5Wm2K1 to one that decreases it by 3.5Wm2K1 . Based on our results, this seems difficult to achieve within our current rationale for the tropics at least.


4 comments:

  1. i just got answer rfom Dr. Lambert

    "- You're correct of course that the website completely distorts our
    message. If there is evidence for negative cloud feedback, then it
    certainly isn't in this paper."

    it seems you have missunderstood the paper you are blogging about here.

    ReplyDelete
    Replies
    1. Please post your entire message to the author and all of his comment. What specifically is he disputing? I have simply quoted the paper itself:

      " Finally, because clouds are a large source of modelling uncertainty, we consider whether resolving errors in cloud simulation could reconcile modelled global mean precipitation trends of about 1–3 % K−1 with some estimates of observed trends of 7 % K−1 or more. This would require the radiative effect of clouds to change from one that increases atmospheric radiative absorption by about 0.5Wm−2K−1 to one that decreases it by −3.5Wm−2K−1 . Based on our results, this seems difficult to achieve within our current rationale for the tropics at least."

      Delete
    2. "Dear Daniel,

      Thanks for the
      heads-up. I don't really get involved with climate sceptics, but I'm glad that somebody does it. Certainly, you won't convince them of anything, but I
      know there must be others watching from the fence who you will.

      I'm hard pressed at the moment with exam marking etc, but here's a quick
      summary:

      - You're correct of course that the website completely distorts our
      message. If there is evidence for negative cloud feedback, then it
      certainly isn't in this paper.

      - The point of the paper is to describe the effect of clouds on the net
      cooling of the atmosphere, which has top of atmosphere (TOA) and surface
      components. Radiative cooling is balanced by (among other things)
      changes in precipitation. We want to understand the balance between
      different processes in the atmosphere. It is not about TOA cloud
      feedbacks as claimed by the website.

      - The paragraph that they highlight points out that we could potentially
      "fix" modelled precipitation to match two rather shaky satellite-based
      approximate estimates of recent global mean precipitation change by
      messing around with clouds. But,

      (i) Those precipitation estimates are really not good at all. Where we
      have good observations (only over land), models do fairly competently,
      albeit with some notable exceptions. Using the satellite estimates to
      try to correct our physical understanding would be very dubious.

      (ii) If modelled precipitation is substantially in error, then it's by
      no means necessarily the case that it's all down to clouds anyway. Even
      if it is down to clouds, it's probably the cloud effect on surface
      radiation (not well-observed) rather than at the top of atmosphere
      (modelled high clouds respond to the same physical processes as observed
      ones according to recent studies, cirrus cloud aside I would guess). And
      the cloud effect on surface radiation does not affect cloud feedbacks on
      temperature, which is what the article is trying to claim.

      (iii) The clouds we are really uncertain of are the low clouds in the
      subtropics. These clouds are very important to *shortwave* feedbacks,
      which are most important to overall feedback uncertainty, but don't
      affect longwave feedback at TOA. (They also don't affect atmospheric
      radiative cooling much, which was what we were interested in...)


      best wishes,

      Hugo
      "

      Delete
    3. "I don't really get involved with climate sceptics, but I'm glad that somebody does it. Certainly, you won't convince them of anything"

      All scientists should be skeptics. That's how scientific progress is made. Read Feynman, Einstein, Happer, etc.

      "Certainly you won't convince them of anything" is blatantly false and appears to be projection on the part of the author. Maybe if the author would debate directly with skeptics instead of arrogantly berating them and refusing to debate, maybe "both sides" could learn something. Many skeptics are scientists in other fields and just might have something to contribute.

      2. I never stated this this paper claims to show negative cloud feedback. There are, however, many that do.

      3. As is typical in climate science, if the models don't match the observations, the observations are blamed. This is the opposite of the scientific method, rife within climate science [but not in any other field of science], and this arrogant pseudo-science has led to models which greatly exaggerate warming.

      4. "Using the satellite estimates to
      try to correct our physical understanding would be very dubious."

      Uh, no it wouldn't, the models make many dubious physical assumptions, including specific humidity will increase while relative humidity stays the same, which has been proven false by observations. If specific humidity goes up, relative humidity goes down to compensate. This affects cloud formation as well. This is only one of many, many false physical assumptions in the models.

      5. "If modelled precipitation is substantially in error, then it's by no means necessarily the case that it's all down to clouds anyway."

      I never said it was

      6."the cloud effect on surface radiation does not affect cloud feedbacks on temperature, which is what the article is trying to claim."

      Nonsense. If surface radiation affects temperature, it affects evaporation, which affects cloud formation, which in turn affects surface radiation.

      6. "The clouds we are really uncertain of are the low clouds in the subtropics. These clouds are very important to *shortwave* feedbacks, which are most important to overall feedback uncertainty, but don't affect longwave feedback at TOA. (They also don't affect atmospheric
      radiative cooling much, which was what we were interested in."

      This is another assumption of the models that may also be false as questioned by a number of papers.

      Delete