• Climate Science
• Instrumental Record

1,285 Responses to “CRU Hack: More context”

1. 1
   Timmy says:

   2 Dec 2009 at 9:41 AM

   Here’s something I’ve yet to see addressed: If the famous hidden “decline” is irrelevant, then how accurate can any of the historic tree ring data be?

   It seems to me that such a major “decline” would indicate that there is no link between tree ring data and historic climate.

   Isn’t Tree ring data somewhat important to AGW theory?
2. 2
   Mesa says:

   2 Dec 2009 at 9:52 AM

   Well, steps 1-4 are OK, but steps 5 and 6 is where all the action will be in the comments. The truth is that most people who have studied the issue at all understand that a doubling of CO2 will lead to a temperature increase of 1.2 C or so **all other things being equal**. What’s not at all clear is how the feedbacks should operate, with by far the biggest feedback effect that of water vapor/clouds. If the 3 C sensitivity number comes from Annan’s bayesian paper – it’s very, very skinny science. If it comes from model simulations – well, it’s apparent even from IPCC that the models don’t handle water vapor/cloud feedbacks very well. I think it’s a hard problem, and it needs more study, but to wave hands and say that the effect with feedbacks is triple that of the bare effect is not going to impress. I think it’s possible that natural variability has masked the drift over the past 10 yrs, but with too many more lost warming years, different ideas and models will be needed.

   [Response: Water vapour and clouds are very different things. The first is very well characterised, the second not so much, and you are correct – the climate sensitivity is a more interesting discussion than the earlier issues – but the top-down constraints on that are not ‘hand waving’. – gavin]
3. 3
   Olive Heffernan says:

   2 Dec 2009 at 9:59 AM

   Timmy,

   You don’t need the tree-ring records as evidence of anthropogenic climate change if you don’t trust those records, for whatever reason. Last year, there was a reconstruction (by Mike Mann et al) of the global temperature record over the past 2,000 years, with and without tree-ring records.

   Without inclusion of the tree-ring measurements, the data still showed that recent
   warming is greater than at any point in at least the past 1,300 years. Instead of tree ring data this reconstruction used sources such as corals, cave deposits, sediments etc.

   See: Proc. Natl Acad. Sci USA 105, 13252–13257 (2008)
4. 4
   Mesa says:

   2 Dec 2009 at 10:09 AM

   OK – sorry to use the term hand waving – not productive. Here’s a topical question – how much of the derived sensitivity depends on the CRU/GISS instrument record? In other words, if total sensitivity with feedbacks is now thought to be 3+ C per doubling of CO2, if the instrumental temperature rise over the past 100 yrs was half of what it now appears to be, what would that do to the implied consensus sensitivity? Suppose there was no significant temperature rise except for in the Arctic over the past 100 yrs? Thanks.

   [Response: As we’ve discussed often, the problem with trying to constrain sensitivity over the 20th Century is the uncertainty in the aerosol contribution. If it is at the high end, then you get very large sensitivities indeed, while if it is at the low end, then sensitivities are lower (but not that low). Issues with ocean uptake are also relevant. Thus the constraints from the surface temperature record in the last century are not that useful. The LGM is much more useful. – gavin]
5. 5
   Jim Bouldin says:

   2 Dec 2009 at 10:11 AM

   re#1:
   Tree ring data is critical in documenting former land surface temperatures, but AGW concepts rest on much more than just tree rings, or any collection of proxy data for that matter–in particular on the thermodynamics of atmospheric gases in relation to the magnitude of other radiative forcings.

   There has been much discussion here and elsewhere on the divergence problem. First, it does not present itself in all locations (see, e.g. the recent “Treeline story” article here). Second, when it does occur, there can be non-climatic explanations, particularly related to the mathematics of the method of standardization chosen. Third, even if it is in fact due to reduced sensitivity of the trees, this does not mean that the historic T estimates are necessarily in error, as a substantial, valid calibration period still exists in almost all cases. Although it does indeed raise the possibility that formerly warm periods could be under-estimated, and attention to this possibility needs to be paid, it is not as though this is unrecognized. If you have access to the literature, start with Esper and Frank, 2009, Divergence pitfalls in tree ring research, Climatic Change 94:261-66. Also a paper last year by D’Arrigo et al.
6. 6
   caerbannog says:

   2 Dec 2009 at 10:13 AM

   
   You don’t need the tree-ring records as evidence of anthropogenic climate change if you don’t trust those records, for whatever reason. Last year, there was a reconstruction (by Mike Mann et al) of the global temperature record over the past 2,000 years, with and without tree-ring records.
   

   And the tree-ring data correlate well with these alternative proxies for pre-1960 time periods. It’s just *certain* tree-ring data diverge from the temperature record and the other proxies post-1960. As long as they are cross-checked with other proxies, even the divergent tree-ring data can provide useful information.

   And of course, people shouldn’t confuse “using tree-ring data” with “relying solely on tree-ring data” (as “skeptics” seem to do all too often).
7. 7
   Jim Bouldin says:

   2 Dec 2009 at 10:14 AM

   D’Arrigo et al., 2008, On the ‘Divergence Problem’ in Northern Forests: A review of the tree-ring evidence and possible causes. Global and Planetary Change 60: 289–305.
8. 8
   cc81 says:

   2 Dec 2009 at 10:16 AM

   I really tried to find if this had been taken up before but I might have missed it:

   I got intrigued by one of the hacked CRU emails, from the Phil Jones and Kevin Trenberth to Professor Wibjorn Karlen. In it, Professor Karlen asked some very pointed questions about the CRU and IPCC results. He got incomplete, incorrect and very misleading answers. Here’s the story, complete with pictures. I have labeled the text to make it clear who is speaking, including my comments.

   wattsupwiththat.com/2009/11/29/when-results-go-bad/

   Is this “a smoking gun” or just manipulations from a blogger? Thank you for taking your time Gavin.
9. 9
   Adam says:

   2 Dec 2009 at 10:19 AM

   “Isn’t Tree ring data somewhat important to AGW theory?”

   Not really, no. It can be useful in comparing current to past temperatures, but tells more of the régime we’re entering than how we got there.

   As for the post 1960s issue (divergence), I’m just a layman so this will be simplified but AIUI it basically comes down to two issues.

   1) Tree rings form in the centre of the tree and show a different width relative to the growth factors (such as temperature) than the older rings which are further out. There are techniques to account for this, but it does not cope well with the rings at either end of the scale (very inner and very outer rings). Thus the most recent set of rings are not going to be as useful as the earlier (but perhaps not the earliest) rings.

   2) Other human impacts. Some areas have had considerable human impact in more recent years (last century, say) such as pollution, CO2 percentage, land use changes, etc. These may have an effect on tree ring growth that hides what was once a clear temperature dependency.

   These issues are being investigated and techniques worked on. However, the earlier rings can be compared to earlier temperature measurements, ice cores, and other proxies and generally holds up pretty well.
10. 10
    Scott A. Mandia says:

    2 Dec 2009 at 10:20 AM

    Regarding sensitivity:

    Bony, et al. (2006). How well do we understand and evaluate climate change feedback processes? Journal of Climate, 19, 3445 – 3482.

    www.met.sjsu.edu/~tesfai/RESULTS/Journals/how%20well%20do%20we%20understand%20and%20evaluate%20climate%20change%20feedback%20processes.pdf

    Knutti, R. & Hegerl, G. (2008). The equilibrium sensitivity of the earth’s temperature to radiation changes. Nature Geoscience, (1), 735 – 743.

    climatechange.pbworks.com/f/The+equilibrium+sensitivity+of+the+Earth's+temperature,+2008.pdf

    It does appear that the uncertainty is at the upper bound and not the 2C lower bound with 3C showing as the most likely. Of course, with the release of the Copenhagen report, 3C is starting to look like the lower bound. Not good.
11. 11
    Bart Verheggen says:

    2 Dec 2009 at 10:21 AM

    Mesa,

    If the real sensitivity were just the CO2-only value of 1.2 by any miraculous cancellation of feedbacks, you’d have a very hard time explaining how ice ages could possibly have occurred. Just to name one constraint.

    James Annan has a blog too, so you can take his “skinny science” up with him if you want/dare: julesandjames.blogspot.com/2006/03/climate-sensitivity-is-3c.html

    I have yet see a hole being poked in it, so I’m all ears to your argument.
12. 12
    Sebastian says:

    2 Dec 2009 at 10:24 AM

    I’d be curious your thoughts on Phil Jones stepping down. Personally I see it as disappointing; I know the administration had said they wouldn’t accept his resignation if he tendered it. Seems like adding fuel to the fire.

    Meantime, I was debating a denialist re: the MBH98 numbers their assertion that corrections issued suggest some sort of obstructionist trench warfare tactics on the part of climatologists in general. Can you help me specifically counter that?
13. 13
    David Harrington says:

    2 Dec 2009 at 10:31 AM

    Was it actually a hack? Or was it a leak from an insider. A leak would be a much more serious problem than a hack as it would tend to imply that someone on the inside nwas not happy with what they were seeing.

    Personally I think it was a leak.

    [Response: My information is that it was a hack into their backup mailserver. – gavin]
14. 14
    VM says:

    2 Dec 2009 at 10:34 AM

    I’m a beginner in this debate, trying to understand the basics. Going back to your six easy steps. Step 3 is a non-sequitur. You have not shown the duration of the changes you discuss or that they are anthropogenic. sorry but your argument needs shoring up.

    [Response: That’s actually the easiest thing to demonstrate (we should have a link there though). How do we know recent CO2 increases are due to human activities. – gavin]
15. 15
    Mesa says:

    2 Dec 2009 at 10:39 AM

    The problem with Annan’s paper is that it takes a lot of wildly uncertain estimates and uses bayesian methods to try to narrow the distribution – I just don’t buy it. I guess I would say that I think the error bars on the original estimates are (much) wider than he does. It’s also possible that the climate doesn’t have “one sensitivity” – ie looking at it as a single number is not useful, just as looking at “global temperature” may not be that useful. I think he’s a smart guy, and I understand part of the idea was to chop off the higher tails in the sensitivity estimates.
16. 16
    Jim Bouldin says:

    2 Dec 2009 at 10:40 AM

    Didn’t Charney et al, NAS, have the sensitivity at right about 3C, 30 years ago?
17. 17
    Timmy says:

    2 Dec 2009 at 10:40 AM

    Thank you Olive, Jim and Adam for trying to sort out the importance of Tree Ring data for me, but i still have a couple unanswered questions:

    Olive: Do you have a link to Mann’s reconstruction minus the tree ring data? (A graph would be preferable, since I’m a layman and work better with pictures.) Every reconstruction I’ve been able to locate includes tree ring data.

    [Response: Here (Blue line). See the supplemental data on Mann et al 2008. – gavin]

    Jim & Adam: I understand that there is far more to AGW theory than historic temperatures, however, the reason it is considered a crisis is because of the assumption that current warming is an unique event. If the modern tree data underestimates the current observed warming, it also could have underestimated it in the past. Which would indicate that the current warming is neither unique, nor a crisis.
18. 18
    VB says:

    2 Dec 2009 at 10:42 AM

    I was under the impression that this is supposed to be a neutral website discussing climate. Are you honestly telling me that only global warming skeptics make mistakes in their papers? I find that a little difficult to stomach.

    [Response: Not at all. The distortion comes in how much press/spin accompanies the mistaken papers which seem to cast doubt on the mainstream science. Good peer review can catch mistakes on both sides though (albeit imperfectly). – gavin]
19. 19
    Olive Heffernan says:

    2 Dec 2009 at 10:42 AM

    Sebastian has already asked this, but I’d also be interested to know your thoughts on Phil Jones’s decision to temporarily hand over the role of director to Peter Liss while the investigation is under way.
20. 20
    Jim Bouldin says:

    2 Dec 2009 at 10:48 AM

    It’s also possible that the climate doesn’t have “one sensitivity” – ie looking at it as a single number is not useful, just as looking at “global temperature” may not be that useful.

    You have to start with a global number and try to get that right before you can start looking at regional sensitivities, it seems to me.
21. 21
    walter says:

    2 Dec 2009 at 10:53 AM

    Timmy, you should also note the tree-ring data was not used for _extrapolation_. Extrapolations based on noisy signals are very tricky and can lead to hilarious errors.

    An example; lets say on april 13 temperature outside is 14 degrees. On april 14 it’s 13 degrees, on the 15th its 11 degrees. This can happen, as weather is very noisy. But if we want to forcast the temperature on april 21th, we should not use said data to extrapolate: the conclusion would be some 5 or 6 degrees!
22. 22
    Adam says:

    2 Dec 2009 at 10:56 AM

    Timmy #comment-146368

    There certainly have been warmer periods in the past, the Cretaceous is one example. However whether there have been warmer periods during the Holocene is I guess the point you’re trying to make. Well tree rings aren’t the only proxies for starters, and as shown above, the reconstructions are not dependant on them.

    “the current warming is neither unique, nor a crisis.”

    What has uniqueness got to do with it? It’s not like we’ve got the evacuation of half of Bangladesh down to a fine art.
23. 23
    Tom Wiita says:

    2 Dec 2009 at 10:58 AM

    Is everyone on this “we do science” site braindead? re: comment 9, no, champ, tree rings form on the outside of the tree as layers of growth are added to the wood already grown. Didn’t anybody here take 7th grade science?
24. 24
    Jim Bouldin says:

    2 Dec 2009 at 10:59 AM

    re # 17:
    If the modern tree data underestimates the current observed warming, it also could have underestimated it in the past..

    Yes, it could have at certain locations. See my response to that point above.

    Which would indicate that the current warming is neither unique, nor a crisis.

    No. The fact that the current warming may not be unique in some locations does NOT mean that the cause of the current global warming is not due to GHG forcing. And “crisis” is a matter of subjective description.
25. 25
    caerbannog says:

    2 Dec 2009 at 11:00 AM

    
    I was under the impression that this is supposed to be a neutral website discussing climate. Are you honestly telling me that only global warming skeptics make mistakes in their papers? I find that a little difficult to stomach.
    

    No, but the “skeptics” make mistakes that would earn college students F’s on their exams. For example, taking the time-derivative of temperature and SOI data before correlating them and then drawing the wrong conclusion about how much of the observed warming is due to the SOI might be such a mistake (I see a good exam question for college students here).
26. 26
    Joe says:

    2 Dec 2009 at 11:01 AM

    ReT_P_Hamilton on the previous thread ( www.realclimate.org/index.php/archives/2009/11/the-cru-hack-context/comment-page-22/#comment-146333 ):

    “Why don’t the deniers just make their OWN homogenization process, and publish in the scientific literature? That would be far better. Perhaps they are incompetent.”

    You clearly miss the point about homogenization. Let’s say, in a very simple case, I collect temperature measurements directly from my carden and my friend in California does the same. We then both send our data to someone in, let’s say, Uzbekistan becuase he wants to do some research.

    Now, my measurements have been taken at a certain altitude (in metres), using an old mercury thermometer that shows celsius cos that’s all I can afford. My friend has taken his at a different altitude (in feet cos he’s American), in farenheit (again, cos he’s American) using a state-of-the-art thermocouple system (again cos he’s American and they like their gadgets).

    The person in Uzbekistan has to correct for the different altitudes and temperature scales and make allowance for the difference in accuracy of the two thermometers and the fact that my Californian friend has an aircon unit which might heat his garden enough to make a difference. That’s the homogenization process.

    Immediately there’s a problem that the only way he can account for the aircon unit is by making an estimate, informed by current knowledge, (an assumption) of the effect it has. That’s fine from an integrity POV as long as his estimate is realistic but it might be realistic and wrong, or someone might come up with new information that affects what would be “realistic”.

    He does all that, in good faith, but genuinely forgets to convert the American altitude from feet to metres. So he’s placed the American thermometer roughly 3x as high as it really is.

    That means that his (honest) conversion will make the American reading seem hotter than it should be because he will have allowed too much for altitude. There is no possible way to spot that error from the dataset he produces. To find such errors you MUST have access to the original data that my friend and I supplied. To confirm that any assumptions were reasonable (including honest mistakes and changes to the state of understanding) you must also have details about what assumptions were made.

    So, no, those sceptical of AGW can’t simply “make their own process” because that process is entirely dependent on what the raw data was and why it needed homogenizing in the first place.

    I have the utmost respect for people working in climate science – it’s a vast, multi-disciplinary, area where no one person can possibly “understand it all” so an awful lot has to be taken on trust, especially in terms of available data.

    The very fact that it’s such a complex and (possibly) vitally important field makes it all the more essential that all data and methods are open to scrutiny and that there is not even a hint of suppressing alternative theory. Otherwise there can be no scientific basis to trust anything published anywhere.

    Incidentally, I also believe that anyone honestly suggesting some global Big Brother conspiracy between Science and Government is a first class fruit-cake with a side-order of nuts (probably served in tinfoil).
27. 27