25 Responses to “Question of the Day”

1. Jack Hughes says:

   July 18, 2011 at 4:38 pm

   I’m filing this under “eco-munchausen’s”.
    
   Revkin desperately wants some global problem to care about and keeps flicking through the Big Book of Plausible Problems.
    
   PS: add the error bars on to the graph and that’s all you can see: the error bars.

2. Keith Kloor says:

   July 18, 2011 at 4:55 pm

   Jack,

   I just want to make sure I get this right. Are saying there is no ocean heating?

3. edG says:

   July 18, 2011 at 5:35 pm

   What is the source of that graph Keith?

4. rustneversleeps says:

   July 18, 2011 at 5:49 pm

   I tend to weary of the navel-gazing over whether this communication approach or that approach; or this phraseology or that might be the “silver bullet” that could change everything…
    
   That said, I think there is something to this. Not a lot. But I think there is a subtle distinction in the way the public “hears” those words. I think that “warming” tends to focus them more on the response of the system, while “heating” tends to shift the focus more towards the act of changing the system.
    
   If the room is a little cold, I turn on the “heat”, I don’t turn on the the “warm”. Warm is an adjective, heat is not; heat is a noun, warm is not (except for the terminally obscure); and I’d argue that “heat” is perceived as the more ‘active/forceful’ verb. Maybe there is something like that for the way the public hears the phrases and might in turn frame the issue.
    
   And as a (very) crude analogy, if you have a pot of water on the stove, if you are focussed on the current temperature change as a measure of “warming” you might conclude that nothing much is happening or is going to happen. But if instead you knew that the element had been turned on and was continually being turned up, you would be relatively less concerned about the immediate temperature and yet quite convinced about the inevitably that the pot will get significantly warmer with signficant certainty as time (and the heating) went on. The more you know about the volume of water, the pot, the rate or heating and the rate of turning up of the element, the less you are concerned about the transient temperature. The influence of a nearby kitchen window open to the winter cold would likely be interpreted very differently by the two perspectives.
    
   As to what Revkin is trying to convey with this graphic, I’m not so sure. I know what it means, but I kinda doubt that would have been the breakthrough insight for the public. The ocean holds more heat than the land and atmosphere?
    
   And despite my half-hearted explanation above, I don’t think the phrase “global heating” would have been salvation either. Might have helped a bit. Water under the bridge, more likely…
    
   Nevertheless, maybe there are some lessons about how we convey the science. I’ve mentioned in the past John Sterman’s Understanding public complacency about climate change: Adults’ mental models of climate change violate conservation of matter, and the use of bathtub analogy to overcome these cognitive problems. Maybe emphasizing the “heating” and not the “warming” is another. I think there are bigger issues…
    
   But hey, Keith, on a blog that has devoted so much time recently to investigating how it’s not an information deficit problem – it’s a culture war!!!… this post seems almost charmingly incongruous…

5. Jarmo says:

   July 18, 2011 at 5:54 pm

   IPCC TAR4 graph uses 10 squared 22 J resolution. Looks a bit different to me:
    
   http://www.skepticalscience.com/images/ipcc_ocean_heat.gif
   Figure 3: global annual upper ocean heat content. Black curve from Levitus 2005, red curve from Ishii 2006, green curves from Willis 2004. Red and green error bars denoting 90% confidence interval. Taken from IPCC AR4.

6. Keith Kloor says:

   July 18, 2011 at 5:55 pm

   Rust,

   Sometimes–not often–I am to please. 🙂

   I think you should try and consider this blog in a different context than how you probably view most climate-related blogs.

7. Artifex says:

   July 18, 2011 at 6:04 pm

   I second edG’s question, what is the genesis of this graph and why does it vary so much from what NOAA has released here ? NOAA’s graph has a negative anomaly in the 1955 to 1970s and a distinct flattening in the mid 2000s. Why are these so different ?
    

8. rustneversleeps says:

   July 18, 2011 at 6:06 pm

   @ edG,
    
   I believe the graphic is from Skeptical Science, and in turn was derived from data from Murphy et al, 2009 and Domingues et al, 2008. I think you would find it (and links to the papers) at Skeptical Science if you searched on “Domingues”.
    
   I could be mistaken. The graph itself may have originated with one of those two papers directly.

9. NewYorkJ says:

   July 18, 2011 at 6:20 pm

   There’s also the increase in global ocean heat storage (0-2000 m) found by von Shuckmann.

   http://www.skepticalscience.com/print.php?n=515

10. Jack Hughes says:

    July 18, 2011 at 7:36 pm

    Hi Keith,
    Nobody knows (if there is or is not “ocean heating”). The Argo project is the first serious attempt to measure the heat content of the oceans and it only started this century. It seems to have run aground recently. At one time it showed a cooling trend but this may be a calibration problem or other error.
     
    Or the results of the Argo project may be politically inconvenient. Yes it is that bad. Things are so bad that it’s really hard to trust anything coing out of this field.
    The graph suggests an overall increase in energy of  200 x 10^21 Joules.  It does not say what the starting and finishing numbers were – in other words is this a 10% increase or a 1% increase or a 0.000001% increase ?
     

11. edG says:

    July 18, 2011 at 8:03 pm

    #7,8 – Yes, why is it different than that NOAA graph?

    It is said to be from skepticalscience. I suppose it is the cartoon version.

    Keith, I would suggest that if you are trying to make a point you use something more credible.

12. Gaythia says:

    July 18, 2011 at 8:22 pm

    I think that this might be an interesting case study in scientist/public communication.
    I also was interested in the origins of this graph.  After a bit of scrounging, I came up with the link below. (which is cited in the much more accessible skeptical science piece linked to by NewYorkJ @9 above):
    From Nature:
    http://www.nature.com/nature/journal/v465/n7296/full/nature09043.html
    The explanation as composed by these authors:
    ” Accounting for multiple sources of uncertainty, a composite of several OHCA curves using different XBT bias corrections still yields a statistically significant linear warming trend for 1993″“2008 of 0.64 W m-2 (calculated for the Earth’s entire surface area), with a 90-per-cent confidence interval of 0.53″“0.75 W m-2.”
    So, the key take away message is: “statistically significant linear warming trend for 1993″“2008”
     
    How should it be presented?

13. intrepid_wanders says:

    July 18, 2011 at 9:28 pm

    rustneversleeps is correct on the graph being from the contested Murphy et al 2009:
    http://pielkeclimatesci.wordpress.com/2009/10/14/e-mail-communication-between-josh-willis-and-roger-a-pielke-sr-on-the-murphy-et-al-2009-paper/
     
    It seems that this graph is quite popular with the alarmists, but it appears to NOT be observations as the title of the paper indicates and based on Argo data with calibration issues.

14. Bob Koss says:

    July 19, 2011 at 12:14 am

    The graphic at the top is only up to 2004. They have also calculated joules by 10^21.   Evidently they wanted to make the graphic look more impressive.
    I think 10^22 is more commonly used for joules. Using 10^21 Joules creates a visually sharper rise in the graphic along with a larger discrepancy in the values.
     
    Here is a Levitus 2009 graphic including the flattening in OHC in recent years.
    http://i52.tinypic.com/2b1o40.jpg
    Note they use 10^22 Joules and also note the comparison with their 2005 paper showing substantial differences prior to 2002. I don’t think they really have a good handle on OHC prior to argo.

15. NewYorkJ says:

    July 19, 2011 at 12:33 am

    Bob Koss (#14),

    Using 10^21 Joules creates a visually sharper rise in the graphic along with a larger discrepancy in the values.
    
    The graphs look virtually the same with regards to steepness.   Perhaps you didn’t observe that Levitus, while using 10^22 instead of 10^21, adjusts the vertical axis numbers accordingly (range of roughly 24 instead of 240).  How devious of them!  Those evil scientists need to stop fooling us…

16. intrepid_wanders says:

    July 19, 2011 at 12:58 am

    It appears that Murphy et al 2009 is based on the data from Domingues et al 2008 (and possibly Levitus et al 2005 and 2009).  Science of Doom has a very nice analysis of all papers involved in creating the graph, including error bars 😉  Obviously, it is all crap in, inflated crap out.
     
    On a side note, it is interesting that during the Argo time-frame (2002-2009), Levitus et al 2009 shows only a slight OHC increase… hmmm… Revkin may want to reconsider his stance, no real heating.

17. edG says:

    July 19, 2011 at 1:28 am

    #16. Thanks. Very interesting article at Science of Doom.

    Back on the question:

    “Would things be clearer if the process known as “global warming“ had been described as “global heating“ from the get-go?”

    I don’t think so. Same problem for both when it isn’t warming. And Global Heating sounds like a multinational Lennox Corp.

    ‘Climate disruption’ would have worked because, as is now evident, that can be stretched to cover anything and everything “wierd” or warmcold. All angles covered. But it is a bit too late for that now.

18. Michael Larkin says:

    July 19, 2011 at 1:42 am

    So what’s the total OHC, rather than the anomaly as presented here? IOW, what fraction of the total OHC does the current (for example) anomaly represent? If it’s a very small proportion, then maybe, as suggested above, error bars could swamp the picture. I haven’t been able to Google the total OHC, unfortunately.
     
     

19. NewYorkJ says:

    July 19, 2011 at 2:36 am

    iw (#13): but it appears to NOT be observations as the title of the paper indicates and based on Argo data with calibration issues.

    Strange claim.  Argo data are a result of the Argo observation system.

    http://en.wikipedia.org/wiki/Argo_(oceanography)

    Oddly enough, just a minute ago, courtesy of the latest SkepticalScience post, I read this funny related statement by the always comedic Lord Monckton.

    They [the NOAA] rely only on data from ships dropping canvas buckets down as they randomly pass across the oceans, and pulling up some water and sticking a thermometer in.”
    
    The fact that Monckton says a lot of goofy things is also an observation.  I also observe a tendency for contrarian types to greatly exaggerate uncertainty estimates and misrepresent or distort nearly every piece of data or study they come across that runs counter to their belief system.

20. toto says:

    July 19, 2011 at 4:40 am

    #7,8,11: Actually the graph is very similar to the one at NOAA, except for stopping at 2004 and using a different baseline. It is also quite similar to the Domingues 2008 and Levitus 2009 graphs shown at Science of Doom.
     
    <i>NOAA’s graph has a negative anomaly in the 1955 to 1970s</i>
     
    It’s an anomaly. The overall level is arbitrary and depends on which reference point you use as zero.
     
    #16: <i>2002-2009</i>
    So after starting from 1995, and now from 1998, should we expect that “skeptics” will shift to a new-and-improved rallying cry of “no warming since 2002” ? 😉

21. intrepid_wanders says:

    July 19, 2011 at 10:45 am

    nyj (#19),
     
    When Dr. Josh Willis of JPL replies,

    “Hi Roger,
    Well, at first I was pretty skeptical about those results because the cooling in the Atlantic brings back bad memories of Argo floats that were biased cold there.  In addition, I think we have to be very careful about any results that span the transition from the XBT network to Argo because I know that there are still systematic errors in the XBT dataset. Although we have all made attempts to reduce these errors, I don’t think anyone can claim to have eliminated all of them.  So, we need to be cautious here.
    That said, I also see a sort of transition in the North Atlantic from a period of rapid warming from the mid-1990s through about 2004, followed by a slight cooling during 2005 and 2006 and it has pretty much been level since then. This seems to agree well with the average over the altimeter data for the North Atlantic.  So, perhaps the recent cooling of the North Atlantic is real.
    Cheers,
    Josh”
    
    
    …I am given the impression that the data has been tortured to fit a given model.  That is NOT an observation.  But, feel free to tossed the Monckton red herrings…

22. Sashka says:

    July 19, 2011 at 8:24 pm

    OMG, that’s a mother of all red herring.
    Nobody has a faintest idea of the pre-Argo ocean heat content. (Even Argo doesn’t cover deep ocean, so we still don’t really know.) There was never nearly enough data to say anything meaningful about ocean heat anomaly because even the base was never established with any accuracy.
    With the advent of satellite observations we got an idea of SST but not much about anything below and close to nothing about deep ocean.
    Too bad that Andy fell for this. He used to know better.

23. steven mosher says:

    July 20, 2011 at 2:53 am

    yes.
    The silly focusing on air temps has ended up putting far too much focus on things like the hockey stick, UHI, etc etc. you get much different optiic if you focus on OHC..

24. Sashka says:

    July 20, 2011 at 9:27 am

    I have lot more confidence in hockey stick than in this graph.

25. JohnB says:

    July 20, 2011 at 11:53 am

    @NewYorkJ. What was amusing or non factual in Moncktons comment?

    I must admit to being bemused at the idea of bias or error in the Argo measurements. It’s a thermometer stuck in the water FFS, how hard can it be to get that right? Aren’t these things tested before they drop them over the side?

    Problems with the equipment that takes them down and brings them up, or the data storage and transmission I can understand, but the thermometer?