A Few Things Ill Considered

A layman's take on the science of Global Warming featuring a guide on How to Talk to a Climate Sceptic.

Monday, April 17, 2006

send this to... Digg it! | Technorati | Del.icio.us | Reddit | Furl | Spurl

No Past, No Present

(Part of the How to Talk to a Climate Sceptic guide)

This article has moved to ScienceBlogs

It has also been updated and this page is still here only to preserve the original comment thread. Please visit A Few Things Ill Considered there. You may also like to view Painting With Water, Coby Beck's original fine art photography.



  • At February 20, 2007 7:29 PM, Anonymous Anonymous said…

    Seems this one rings true for alot of people, but it shouldn't be hard to clear up such a misunderstanding.


    What is so hard behind the concept of, "Observation and data can lead to model creations." Its an integral part of modern science, do you dislike modern science?

  • At March 07, 2007 1:29 AM, Anonymous Anonymous said…

    The IPCC isn't as confident as you about our knowledge of many of those things... including the sun.

  • At March 26, 2007 11:45 PM, Anonymous Anonymous said…

    Hi coby, great site.

    This seems as good a place as any for this post.

    My question comes from a GW critic (who says the denialist label is 'demonizing').

    In table SPM-2 of the latest IPCC report, confidence on trends for warm spells and heat waves for the future (by way of modeling) is greater than recorded (past) trends of the late 20th century. Greater confidence is given for other future trends than past in that box, too. The question is, how does the panel arrive at a higher confidence for future projections based on models than for data already recorded? IOW, why is there more uncertainty regarding recent past trends (where we have been able to record data), than on future trends where we can only rely on modeling?

    I searched the net for an answer but found nothing specific on that. I speculated (I'm no scientist) that the greater uncertainty in the recent past might be a result of the difficulty in determining what constitutes a 'heat wave' (making me wonder why an ill-defined phenomenon made it into the report in the first place), and that the parameter of the subject is global, while measurements have been limited to regions, or that the data from various regions is questionable. I read elsewhere (realclimate) about heat waves in Europe and North America - if I remember correctly, it was stated that world wide heat measurements in that regard have been ongoing since before 1900.

    What do you think?

    (emailed this question to realclimate also)

  • At March 27, 2007 12:25 AM, Blogger coby said…

    how does the panel arrive at a higher confidence for future projections based on models than for data already recorded?

    I think the answer for this lies in the fact that we are only at the beginning of significant warming and just now rising above what falls in the range of natural variability. In other words, even while the data may clearly show trends it is still to small to completely rule out random flucuations and even when that is ruled out there is uncertainty in all the contributing factors. Once GHG levels have doubled and beyond it becomes a very certain matter of physics.

    Hopefully RC will give you a more detailed and authoritative reply!

    Thanks for the comment.

  • At April 03, 2007 7:49 AM, Anonymous Anonymous said…


    I emailed realclimate on the question and this is the reply I got (from Gavin).

    "...there are a number of reasons why future attribution might appear to be more confident than in the past. Mostly, it is because the forcings in the past (prior to the satellite era) are much more poorly constrained (aerosols + solar in particular) - so the attribution of the ~0.75 warming so far is more tricky than calculating the estimated change in temperature in the future (which of course assumes some GHG trajectory)."


  • At May 11, 2007 5:36 AM, Anonymous Anonymous said…

    One of the biggest uncertainties for the science relate to the feedback mechanisms, from water vapour and cloud changes in particular, that arise when the atmosphere warms. Many of the feedbacks are believed to be the same whether the cause of warming is greenhouse gases, solar or albedo changes from melting ice.

    For example, the models that predict up to 4-5C warming this century do so because they predict strong positive feedbacks from clouds and water vapour. But scientists such as Lindzen make the argument that these feedbacks are wrong, and that they are more likely to be negative, leaving us with minor warming of 1C or less.

    But the reconstructions of paleoclimate, though not well understood, shows us that the climate can diverge by 5-10C. If feedbacks from water vapour and clouds are negative or zero, then the changes required to cause such swings in temperature are huge - equivalent to a change in the suns strength of 10% for the recent ice age cycles, and possibly upto 30% for the mid-cretaceous maximum, 100 million years ago. I have not found any other theories that account for such changes.

    So rather than the uncertainty being a thorn in the side of global warming theory, they provide pretty good evidence that the models are about right.

    Steve Milesworthy

  • At May 11, 2007 5:40 AM, Anonymous Anonymous said…

    PS. the figures in the above were based on:

    "Deriving global climate sensitivity from palaeoclimate reconstructions" Hoffert and Covey, Nature Vol 360, 10th December 1992.

    but I'm working through a stack of papers that make other predictions using similar methods.

    Steve Milesworthy

  • At May 11, 2007 10:23 AM, Blogger coby said…

    Hi Steve,

    Those are good and important points, thanks for posting them.


Post a Comment

<< Home