IPCC and the Law Dome Graphic

Re-reading Climategate and AR4 Review Comments, I noticed an interesting discussion about handling the Law Dome O18 record – a series used in Mann and Jones (2003) and Jones and Mann (2004) with a very elevated MWP.

The Law Dome O18 series was illustrated in Jones and Mann 2004 as follows (although the digital data for most series in this article was commendably archived, the digital version of the Law Dome O18 series wasn’t. After a couple of years of effort, I obtained it from Tas van Ommen.) Law Dome O18 was one of three series used in the Mann and Jones (2003) Sh “reconstruction” ( Cook’s Tasmanian tree ring chronology and Thompson’s Quelccaya, Peru ice core were the others.) The Mann and Jones (2003) SH reconstruction was discussed in the AR4 Second Draft as follows (language unchanged in the final version).

6.6.2 Southern Hemisphere Temperature Variability: There are markedly fewer well-dated proxy records for the SH compared to the NH (Figure 6.11), and consequently little evidence of how large-scale average surface temperatures have changed over the past few thousand years. Mann and Jones (2003) used only three series to represent annual mean SH temperature change over the last 1.5 kyr.

AR4 Second Draft Figure 6.11 purported to show the “locations of temperature-sensitive proxy records with data back to 1000, 1500 and 1750”, but, for some reason, didn’t include Law Dome, Quelccaya and other sites. The caption was as follows;

Figure 6.11. Locations of temperature-sensitive proxy records with data back to 1000, 1500 and 1750 (instrumental records: red thermometers; tree-ring: brown triangles; boreholes: black circles; ice-core/ice-boreholes: blue stars; other records including low-resolution records: purple squares). All proxies used in reconstructions [R1] to [R11] of Northern Hemisphere temperatures (see Table 6.1 and Figure 6.10) or used to indicate Southern Hemisphere regional temperatures (Figure 6.12) are included.

Neither was the Law Dome O18 data shown in Figure 6.12, illustrating SH proxy histories.

IPCC AR4 Second Draft Figure 6.12

The IPCC stated of this data:

Taken together, the very sparse evidence for Southern Hemisphere temperatures prior to the period of instrumental records indicates that warming is occurring in some regions. However, more proxy data are required to verify the apparent warm trend.

SOD Review Comments
One (and only one) IPCC noticed that SH proxies in R1 and R2 (Jones et al 1998, Mann et al 1999) were missing from Figure 6.11:

6-1168 B 30:5 30:5 Figure 6.11a does not show many proxies used in R1, R2 [Jones et al., 1998; calibrated by Jones et al., 2001; Mann et al., 1999 ] : e.g. Rio Alerce, Lenca, Morocco tree rings, Quelccaya, Law Dome [Stephen McIntyre (Reviewer’s comment ID #: 309-63)]

IPCC accepted this criticism (and did add missing proxies), mentioning in passing that they had removed the term “temperature sensitive” from the caption – I hadn’t made this request. In retrospect, this was sort of an odd thing to do, given that these are supposed to temperature proxies. More on this later.

Accepted – Figure (6.11) now shows a more comprehensive picture of proxy series locations used in the references cited. The reference to “temperature sensitive” proxies has been removed in the caption and additional series, as indicated by the reviewer, have been shown.

I also wondered what had happened to the Law Dome proxy and why it wasn’t shown in Figure 6.12:

6-1231 B 34:12 34:12 What happened to the Law Dome proxy? Why isn’t it shown? [Stephen McIntyre (Reviewer’s comment ID #: 309-115)]

This prompted a surprisingly elaborate answer – one that is considerably clarified by the Climategate emails.

Past temperature variations at Law Dome have been inferred from isotopic and borehole records. (1) Jones and Mann (2004) showed an isotope record from Law Dome based on O18. This record has a “cold” present-day and “warm” 1000-1750 period. Dahl-Jensen et al. (1999) showed temperature variations at Law Dome obtained by inverting the borehole temperature profiles. This record has a colder interval (peaking in 1250 and 1850) relative to the recent period, followed by a steady recent warming. Therefore, the opposite trends recorded in these reconstructions do not allow reaching a final consensus on temperature variations at Law Dome during the past millennium.

Although I hadn’t previously noticed this point, Dahl-Jensen (1999) – Dahl-Jensen, D., V.I. Morgan, and A. Elcheikh, 1999: Monte Carlo inverse modelling of the Law Dome (Antarctica) temperature profile. Ann. Glaciol., 29, 145–150. – was not mentioned in the Second Order Draft.

Climategate Letters
The Climategate Letters in July 2006 proved to have a surprisingly lengthy discussion about how to refuse my request that the Law Dome proxy series be shown.

On June 30, 2006 (704. 1153167959.txt), Overpeck wrote:

Figure 6.12 …
2. consider adding Law Dome temperature record – Ricardo is investigating, but perhaps Keith/Tim can help figure out if it’s valid to include. Feel free to check with Valerie on this too, as she seems to know these data at least a little

On July 14 (700. 1152912026.txt) Overpeck wrote Briffa, Osborn, Villalba, Jansen and Masson-Delmotte revisiting Figure 6.12:

Subject: figure issues
Hi all – including Eystein, whom I haven’t been able to talk with on
these issues yet:
1) I’d like to get your status report on Fig. 6.12 – based on
feedback from Henry Pollack, we will keep the borehole curves and
corresponding instrumental data. I believe we are also going to add
the new recon from Law Dome – Valerie was going to send. Do you have
everything needed for this figure revision?

Later on July 17, 2006 16:25 (704. 1153167959.txt) Osborn replied to Overpeck, also commenting on his June 30 questions (cc Briffa, Jansen, Villalba, Joos):

Subject: Re: Special instructions/timing adjustment
Hi all,
I’m halfway through these changes and will get the revised figures out to you probably tomorrow, except maybe the SH one, because: I’m not sure if the van Ommen (pers. comm.) data shown by Jones & Mann and suggested by Riccardo are the data to use or not. Is it published properly? I’ve seen the last 700 years of the Law Dome 18O record published, so perhaps we should show just the period since 1300 AD? That period appears in: Mayewski PA, Maasch KA, White JWC, et al. A 700 year record of Southern Hemisphere extratropical climate variability ANNALS OF GLACIOLOGY 39: 127-132 2004 and Goodwin ID, van Ommen TD, Curran MAJ, et al. Mid latitude winter climate variability in the South Indian and southwest Pacific Regions since 1300 AD CLIMATE DYNAMICS 22 (8): 783-794 JUL 2004 …
Cheers, Tim

Late in the evening of July 17 (Arizona time), Overpeck reverted to Osborn that they thought that they would go with borehole data (which didn’t have an elevated MWP – though it had elevated values in earlier periods):

Hi Tim et al (especially Valerie) – again, sorry for the confusion, but hopefully the emails sent and forwarded from Valerie and me this evening helps figure this out. I think we’re going with borehole for Law Dome, but you guys need to confirm it’s the way to go. I’m cc’ing to Valerie in the hope she can try to provide more guidance in this – with a confirmation that it’s the best way to go and will stand up to criticism. If we have multiple conflicting temp recons from Law Dome, and one can’t be shown from the literature as being the best, then we should state that, and show neither – just an idea. BUT, I think Valerie was pretty sure the borehole was best. She should be more available in a day or so.

The next day, (709. 1153233036.txt), Overpeck wrote that Masson-Delmotte had sent them references a couple of weeks earlier and that the Law Dome borehole data was available and should be the data used in a revised Figure 6.12 (various references from Masson-Delmotte were shown in the email):

Hi Tim, Ricardo and Keith – Valerie just reminded me that she sent this to us all (minus Tim) back in June. There is plenty below for discussion in the text, and the Law Dome borehole data can be obtained at the site below (http://www.nbi.ku.dk/side95613.htm). This is the record that should be added to the SH figure.
Thanks, Peck

Soon afterwards on July 18 (709. 1153233036.txt), Osborn wrote to explain that, if they showed the borehole data with its supposedly cool MWP, they would be criticized for not showing the isotope data with its warm MWP. Osborn suggested a Sir Humphrey solution of showing neither – instead talking around the problem verbally in the text;

Hi all,

(1) Jones/Mann showed (and Mann/Jones used in their reconstruction) an isotope record from Law Dome that is probably O18 (they say “oxygen isotopes”). This has a “cold” present-day and “warm” MWP (indeed relatively “warm” throughout the 1000-1750 period). The review comments from sceptics wanted us to show this for obvious reasons. But its interpretation is ambiguous and I think (though I’m not certain) that it has been used to indicate atmospheric circulation changes rather than temperature changes by some authors (Souney et al., JGR, 2002).

(2) Goosse et al. showed Deuterium excess as an indicator of Southern Ocean SST (rather than local temperature). Goosse et al. also showed a composite of 4 Antarctic ice core records (3 deuterium, 1 O18). Neither of these comes up to the 20th century making plotting on the same scale as observed temperature rather tricky!

(3) Dahl-Jensen showed the temperatures obtained by inverting the borehole temperature profiles. This has a colder MWP relative to the recent period, which shows strong recent warming.

I have data from (1) and now from (3) too, but not from (2) though I could ask Hugues Goosse for (2). Anyway, (1) and (2) aren’t calibrated reconstructions like the others in the Southern Hemisphere figure, so plotting them would alter the nature of the figure.

But if we show only (3) then we will be accused of (cherry-)picking that (and not showing (1) as used by Mann/Jones) because it showed what we wanted/expected.

Can I, therefore, leave the SH figure unchanged and can we just discuss the Law Dome ambiguities in the text?


Overpeck liked Osborn’s tactic of dealing with the inconsistency verbally (rather than letting readers actually see the inconsistency). Overpeck congratulated his team for their “nice resolution”, but didn’t overlook the opportunity to make a snide comment about the “experts” who had the temerity to wonder why the Law Dome data hadn’t been illustrated:

Subject: Re: Law Dome figure
Hi Tim, Ricardo and friends – your suggestion to leave the figure unchanged makes sense to me. Of course, we need to discuss the Law Dome ambiguity clearly and BRIEFLY in the text, and also in the response to “expert” review comments (sometimes, it is hard to use that term “expert”…). Ricardo, Tim and Keith – can you take care of this please. Nice resolution, thanks.
best, Peck

AR4 left its mention of Mann and Jones (2003) unchanged as follows:

There are markedly fewer well-dated proxy records for the SH compared to the NH (Figure 6.11), and consequently little evidence of how large-scale average surface temperatures have changed over the past few thousand years. Mann and Jones (2003)used only three series to represent annual mean SH temperature change over the last 1.5 kyr. A weighted combination of the individual standardised series was scaled to match (at decadal time scales) the mean and the standard deviation of SH annual mean land and marine temperatures over the period 1856 to 1980.

Following the CRU strategy, Figure 6.12 was left unchanged with the following sentence added to the text:

Contrasting evidence of past temperature variations at Law Dome, Antarctica has been derived from ice core isotope measurements and from the inversion of a subsurface temperature profile (Dahl-Jensen et al., 1999; Goosse et al., 2004; Jones and Mann, 2004). The borehole analysis indicates colder intervals at around 1250 and 1850, followed by a gradual warming of 0.7°C to the present. The isotope record indicates a relatively cold 20th century and warmer conditions throughout the period 1000 to 1750.

The closing sentence of the SH section was left unchanged:

Taken together, the very sparse evidence for SH temperatures prior to the period of instrumental records indicates that unusual warming is occurring in some regions. However, more proxy data are required to verify the apparent warm trend.

Just in case you wondered what Osborn and Overpeck didn’t want you to see, here it is:

Note that the inversion of borehole temperatures raises interesting questions equivalent to principal component retention – see prior posts on this. I’ve asked the Danish institution for the pre-inversion borehole measurements. See http://www.climateaudit.org/?tag=borehole especially Truncated SVD and Borehole Reconstructions

Update- Dahl Jensen responded in April 2010:

From: Dorthe Dahl Jensen
Sent: April-13-10 10:39 AM
To: Steve McIntyre; tas.van.ommen@xxx; Mark Curran
Subject: Re: FW: Law Dome Borehole

Dear Steve McIntyre

The temperature measurements of the Law Dome are the property of the
Hobart Glaciology group and I propose you contact them to obtain the data

yours sincerely
Dorthe Dahl-Jensen


  1. EdeF
    Posted Apr 5, 2010 at 1:12 PM | Permalink


    Thanks for your nice “resolution” of this.

  2. Answer up
    Posted Apr 5, 2010 at 1:16 PM | Permalink

    How come you don’t have time to answer Tom’s questions about methods, but have time for this crap, Steve?

    Steve: Tom observed that there was an inconsistency between a caption and the running text in my submission to the Parliamentary Inquiry. I determined that the caption contained the correct information; I posted information on the correction on the post reporting my submission, I placed an inline notification to Tom P that I had done so on an active thread and I sent a corrigendum to the committee.

    • clazy8
      Posted Apr 7, 2010 at 9:37 AM | Permalink

      Very generous of you to answer this obnoxious question, Steve.

  3. Charlie A
    Posted Apr 5, 2010 at 1:20 PM | Permalink

    Climategate e-mails are the gift that just keeps on giving. 🙂

    If only the government funded climate scientists would put as much energy into furthering our knowledge of climate as they appear to be putting into efforts to promote and defend their already established and fixed beliefs.

  4. Posted Apr 5, 2010 at 1:25 PM | Permalink

    Dahl-Jensen et al. (1999) showed temperature variations at Law Dome obtained by inverting the borehole temperature profiles.

    I’m curious about the “inverting” technique as well. Is like Mann’s upside down technique?

    • TAG
      Posted Apr 5, 2010 at 1:39 PM | Permalink

      It is a mathematical inversion – take teh bore hole readings and derive temperature from them

  5. Charlie A
    Posted Apr 5, 2010 at 1:37 PM | Permalink

    Jason — in this context the inversion is NOT flipping the temperature profile upside down. It is a group of mathematical techniques for going from observed borehole temperatures backwards (inverse) to the surface temperature inputs that could have resulted in the observed borehole temperatures.

    Inverse in this context is the same as in the sine and inverse sine functions you will find on your calculator. Sine function goes from angle to a value. Inverse sine goes the other direction, from a value to an angle.

    Steve: this description of the inverse problem is not really on point – the inversion problem is that there are multiple solutions that look very different from one another. The “solution” shown in the recons includes some arbitrary conditioning assumptions that can’t be shown to apply.

    • Craig Loehle
      Posted Apr 5, 2010 at 1:44 PM | Permalink

      It has been suggested that the borehole inversion problem is ill-conditioned (non-unique) (discussed here at CA). I suspect this is likely, and one can clearly see how smooth the borehole temperatures are compared to other proxies, so there is certainly some loss of info.

      • Paul Linsay
        Posted Apr 6, 2010 at 1:16 PM | Permalink

        The heat equation is a diffusion equation which guarantees complete loss of information with time. From the bits and pieces I’ve picked up reading here over the years, it’s surprising that any information can be extracted simply because of the physical state of the boreholes and the uncertainties in the underground environment.

  6. S. Geiger
    Posted Apr 5, 2010 at 1:53 PM | Permalink

    are there (a lot of) cases where borehole temp reconstructions via ‘inversion’ and alternate isotopic methods provide good matches?

  7. Steve McIntyre
    Posted Apr 5, 2010 at 2:01 PM | Permalink

    Here is a good post(IMO) on borehole inversion, connecting the problem to truncated SVD.

    Are there cases where borehole inversion and alternatives “match”? I’ve never seen such a comparison. One of the ongoing frustrations in this field is the failure of specialists to reconcile inconsistencies between different proxies – something that is obviously evidenced in this case.

  8. Posted Apr 5, 2010 at 2:01 PM | Permalink


  9. Posted Apr 5, 2010 at 2:05 PM | Permalink

    Why is there so much difference between the isotope reading and the borehole reading except between about 1250 an 2000?
    Is the borehole as “robust” as tree ring data?

    Steve: read the linked post on borehole inversion and truncated SVD. It might be less robust.

  10. Hmmm
    Posted Apr 5, 2010 at 3:13 PM | Permalink

    The correct answer from science 101 would be:
    Show the results and the uncertainty. Even if the resultant reconstruction doesn’t provide precise temperatures/trends, that in itself is useful knowledge.

    But instead from climate science 101 we get:
    Ignore inconvenient series (or even just the inconvenient portions of series) to reduce the concluding uncertainty.

    It really is as simple as that. Even when directly called out on it. Then they actually call it good work and congratulate themselves. I can’t trust this field. I don’t think most of them even understand what they did was wrong.

    It makes me wonder what happens to the inconvenient data we don’t know about? It took so much work and attention to detail just for an outsider to reveal this crap! What went (and goes) on where he can’t look?

  11. pete m
    Posted Apr 5, 2010 at 7:41 PM | Permalink

    “expert” – so clever.

    Says it all really about their view of the outside world.

    Get over yourselves guys!

  12. rjtomes
    Posted Apr 5, 2010 at 8:15 PM | Permalink

    Where can I download the data for Law Dome that was used above?
    (and in the graphic https://climateaudit.files.wordpress.com/2005/02/lawdom1.gif for example).

    Steve: borehole data at http://www.gfy.ku.dk/~www-glac/data/ddjtemp.txt . I got the isotope data by an email from Tas van Ommen in 2006 after a couple of years. At the time, he said that he anticipated making the data public (with update to 2004) in the near future. I just checked and he replied that the data remains non-public. There is a short data set to 2000 at the Australian

  13. Robert
    Posted Apr 5, 2010 at 8:40 PM | Permalink

    This is OT but I was wondering whether anyone knew how to get the v2.mean.Z (mean monthly temperatures) uncompressed from the Global Historical Climate Network. I’m not so big on computer coding but I think it is some sort of linux compression and i thought maybe i could download some program (for my windows 7) which could help me uncompress it to see the monthly data.

    • pjm
      Posted Apr 5, 2010 at 9:13 PM | Permalink

      Yes, thing.z files have a type of unix/linux compression. You can unpack them with 7-zip, which is freeware. See http://www.7-zip.org/

    • Tim Channon
      Posted Apr 5, 2010 at 9:21 PM | Permalink

      Try http://www.7-zip.org/ and yes it specifically mentions W7. GUI integrates with Windows shell without getting in the way. I’d take the older version.

      Whether the insides of the .Z archive are much use to you is a different matter.

  14. Posted Apr 6, 2010 at 3:27 AM | Permalink

    There are two points to mention in the bore hole proxy and 18O differences:

    – Borehole temperatures flatten out over time, thus reverse engineering of the temperatures is less and less accurate going back in time.
    – Borehole temperatures only reflects local temperature of the (snow) precipitation/surface at the place where is drilled, while 18O (and dD) reflect the temperature of the oceans where the water vapour originated + the temperature of the air masses where the clouds/snow were formed. This may give a quite different temperature profile.

    I should prefer the 18O/dD profile, as that reflects the temperature changes over a larger sea surface area, while the borehole only reflects local point temperatures, except if one has many such points or from an area where local temperature reflects the temperature of a large area (like the Greenland inland ice summit).

    Despite that, the Law Dome O18 temperature profile still is rather restricted to the Southern Antarctic ocean and heavely influenced by regional ENSO-like ocean/wind patterns, which makes that the Peninsula temperatures behave opposite to the rest of Antarctica. The more inland ice cores reflect much larger ocean surface temperatures, but their resolution (Vostok: 600 years) is not sufficient to show what happened over the past millenium.

    See the Holocene dD record for the Holocene:

    Click to access masson.pdf

    More background on 18O profiling can be read here:

  15. David A
    Posted Apr 6, 2010 at 7:22 AM | Permalink

    This appears telling…

    “But if we show only (3) then we will be accused of (cherry-)picking that (and not showing (1) as used by Mann/Jones) because it showed what we wanted/expected.”

    Does this reflect a strong possibilitie that if the three data methods were not well known, then the team would have choosen “3”, and only the fact that they knew they were being watched by outside “experts” prevented this choice?

  16. Posted Apr 6, 2010 at 7:40 AM | Permalink

    Law Dome is not the only Antarctic temperature proxy that CLEARLY shows WARMER MWP than today.

    We also have the “Remote Plateau” located at the very South Pole showing warm MWP:
    See black graph “37 South Pole”:

    Look fig 7 in this article:

    The “Remote Plateau” or “37 South Pole”
    Mosley-Thomson 1996 : “Holocene climate changes recorded in an east Antarctica”

    Article here:

    Please have patience to download (!) and then checkout fig 6.
    These high quality datapoints for every 2-3 years (ten times better than Vostok..) clearly shows 2 big warming peaks in the medieval that is warmer than today. Ont the very South pole.

    These data are O18 – BUT – in the very article, it is written how to translate into temperature, see pages 276-77 i think.

    So some of the best data for Antarctica is hard to download, etc. but the message is clear.

    (Or take a gooood look at figure 5 here, sum of several proxies of Eastern Antarctica:

    Finay, Steve McIntyre, i need your help.. 🙂
    I may be clumsy etc. but I cant find the place to download the full length of RAOBCORE NH TLT 1958-today data. I know tha University of Wienna has somthing on this, but.. I couldnt see the download data option or just a graph for these data. (Why dont they show a graph for the different raobcore temperature data…. ? Is it “hidden”?)

    Anyway, i see that you have had luck downloading Raobcore data, so if you could post a link?
    (And hopefully all other raobcore data)

    K.R Frank Lansner

    Steve: I have an R-script that downloads Raobcore. Remind me.

    • ferdiegb
      Posted Apr 7, 2010 at 1:05 PM | Permalink

      Thanks Frank for the links, I was not aware of the detailed dD records of the inland Antarctic ice cores…

  17. Posted Apr 6, 2010 at 8:27 AM | Permalink

    It all consistent with how the IPCC works. They pick their proxy data pretty much the way they pick the references they cite in the complete AR4. There will be more coming out on this soon.

  18. Ron Cram
    Posted Apr 6, 2010 at 9:31 AM | Permalink

    Nice work on this. It goes to show the IPCC “experts” are more interested in “getting rid of the MWP” than they are in providing an accurate view of the science. If you cannot get away with photoshopping the graphic, then refusing to show the graphic is the next best resolution. The IPCC at work – hiding the science.

    • Shallow Climate
      Posted Apr 6, 2010 at 11:33 AM | Permalink


    • nono
      Posted Apr 6, 2010 at 2:20 PM | Permalink

      Have you noticed that Law Dome MWP (1000-1750) is below present temperature?

      • nono
        Posted Apr 6, 2010 at 2:27 PM | Permalink

        ooops, I had misinterpreted the graph. Sure enough, they could have included both graphs (O18+borehole) instead of only talking verbally about the problem.

  19. mpaul
    Posted Apr 6, 2010 at 2:40 PM | Permalink

    Its another example where evidence that supports the consensus view gets waved through with virtually no review, while evidence that does not support the consensus gets subjected to mob scrutiny with the clear objective of finding a reason, any reason, to exclude said evidence.

    Once the approach becomes acculturated (as it has in climate science) then research outcomes evolve over time to match the predetermined viewpoint. It’s not really a overt conspiracy — it’s a cultural defect in the climate science community that results in bad science.

    You’ve got to conclude that the ‘bunker mentality’ has led to a heightened sense of homophily among these scientists. As Eisenberg and Riley (2001) found: “a person’s identity [within a community] is not found in behavior or in the reactions of others, but in the capacity to keep a particular narrative going”.

  20. rw
    Posted Apr 6, 2010 at 4:34 PM | Permalink

    Wrong post for this I know, but thought you would be interested this. Steve your site is quoted.


  21. Geoff Sherrington
    Posted Apr 6, 2010 at 11:03 PM | Permalink

    Hmmm. From one of Ferdinand’s papers above, a sentiment with which I can align, re oxygen isotopes. I have no independent knowledge of how this paper was received by the top guns, but here is some of it:

    “……stressed, that only the changes in delta18O should be shown in the diagrams, never the change in temperature. Even today, after almost 40 years of ice core research, the problem remains unsolved.
    There are many different reasons for the persistent uncertainty. The delta18O is influenced by many other factors apart from air temperature, such as the original isotope ratio of the ocean water (which was different from today during ice ages), the origin of precipitation and the seasonal distribution of snowfall events.” Which is what I have been writing for some years in less formal places. The part played by sublimation (if any) to differentiate oxygen isotopes in water is not a prominent feature of relevant papers; and Law Dome has a hugely larger accumulation rate than most other stations studied in the Antarctic (Masson et al, ibid, fig 2b). Uniformitarianism takes ue where?

    Next, being in a gloomy mood, I think it verges on the ridiculous to try to recreate temperatures from ice boreholes over several hundred years. Those with better maths than I might calculate the thermal diffusion between layers, to find that the heat dissipates through conduction to the period where no signal remains. I suspect this period can be counted in years on the fingers and toes. Yes, I read in detail many past blogs on the ill-posed nature of the inversion mathematics and I have to agree.

    Some Huon pine dendro work from Tasmania, another main proxy, has been criticised because of the weather differences between where the pines grow and where the calibrated weather stations are. Too far away for comfort. But, I don’t have a closed mind, so I might be convinced that it’s OK.

    So where does this leave us? With the Southern Hemisphere in doubt as to whether (m)any reliable proxies exist beyond 150 years before present? Time to call Houston?

    • John Murphy
      Posted Apr 14, 2010 at 5:18 PM | Permalink

      I agree with you. I think boreholes are probably nonsense and no reliance should be placed on them as evidence for, for example, the MWP.

      When I get some time in the next few weeks, I’ll simulate the heat flow in some boreholes with long term surface temperature changes. What would be interesting – ice boreholes or dirt/rock boreholes, or both?

  22. Ron Cram
    Posted Apr 7, 2010 at 7:45 AM | Permalink

    OT, but I cannot find an active Unthreaded thread.

    I appreciate greatly all of your work on Climategate emails, but I came across something you may want to look into more closely at some point. See http://www.americanthinker.com/2010/04/was_the_arctic_ice_cap_adjuste.html The update at the bottom of his article has a more accurate understanding, but I would be very interested to see your assessment of NSIDC data.

  23. Readertom
    Posted Apr 7, 2010 at 11:04 AM | Permalink

    Is there a good primer for this whole debate?
    I have to confess ignorance of methodology, particularly about bore holes.

    I need something for the liberal arts types.

    Thanks for any help.

    • Brooks Hurd
      Posted Apr 7, 2010 at 2:50 PM | Permalink

      This website if well orgnized for searching. There is a tremendous amount of information here. Look in the upper left for categories. A good starting point would be “MBH98,” since this is the paper which got Steve’s attention.

    • TAG
      Posted Apr 7, 2010 at 3:09 PM | Permalink

      The best primer that I have found is the book by Bishop Hill” whosee proper name is Andrew Montford. It is called ‘The Hockey Stick Ilusion” and is available from the UK version of Amazon.

  24. Frank
    Posted Apr 7, 2010 at 11:36 AM | Permalink

    This post raises the interesting issue of what 4AR should have said about the Southern Hemisphere. It looks as if the conclusion was written first, and then information to support the conclusion was gathered: “Taken together, the very sparse evidence for SH temperatures prior to the period of instrumental records indicates that unusual warming is occurring in some regions. However, more proxy data are required to verify the apparent warm trend.”

    What should 4AR have actually said? Steve’s suggestion (that additional proxy records should have be added) misses a bigger problem. If the peer-reviewed literature contained any useful reconstructions of SH temperature, these reconstructions should have been featured in place of individual records. The only published reconstruction may be Mann and Jones (2003), which CONTRADICTS 4AR’s conclusion that unusual warming is occurring. http://www.spaceweather.ac.cn/publication/jgrs/2003/Geophysical_Research_Letters/2003GL017814.pdf

    Instead of reporting what peer-reviewed reconstructions actually said about the SH, 4AR provides a selection of proxy records purporting to illustrate trends in SH climate. Steve’s post provides conclusive evidence that the authors were reluctant to include proxies that conflicted with their pre-determined conclusion. (It is interesting to note that Mann and Jones (2003) rejected the NZ proxy record shown in 4AR because of lack of correlation with the instrumental record.)

    If the authors of 4AR believed that the peer-reviewed literature had nothing useful to say about naturally-occurring climate variation in the SH over the last few millennia and they thought that it was important to say something, they should have published their own reconstruction using all of the available proxies that met a pre-determined set of criteria.

  25. KuhnKat
    Posted Apr 7, 2010 at 5:43 PM | Permalink


    they should have said there were insufficient studies to say anything useful.

  26. oneuniverse
    Posted Apr 8, 2010 at 3:34 AM | Permalink

    I’d thought that boreholes were the one good non-proxy temperature record we had. Having read some of the CA threads, I feel disabused.

    Are there any boreholes studies which have been done with a careful approach from the start, before drilling began (rather than exploiting pre-existing ‘holes of opportunity’ 🙂 ) ? I vaguely remember reading that T reconstructions created from boreholes made in non-permeable rock showed that groundwater effects were small and ‘manageable’. Is that just wishful thinking ?

    d180 : it seems to be more a proxy for precipitation rather than temperature ? (thank you for the links, Ferdinand). Are the two closely connected?

    Steve: over and above groundwater problems, the inversion procedures use arbitrary regularization assumptions – there are always multiple solutions that are consistent with the observations.

    • oneuniverse
      Posted Apr 9, 2010 at 4:17 PM | Permalink

      Might it be possible to resolve this (at least theoretically, assuming the profile measurements are meaningful) by drilling two or more boreholes in the closest allowable proximity, with a period of some years or decades between each drilling ? Please note that my knowledge of statistics, and much else, is quite basic, so I apologise for the mistakes that doubtless follow :

      The idea is that if the multiple solutions/reconstructions from a borehole’s residual T profile are all extended by the same time series (any series), and then cast back to residual profiles, then it seems plausible that the residual profiles won’t be identical for the extended series (except maybe in very rare cases).

      Ideally, the above should be formally proved, but it can easily be tested to some confidence-building extent, or disproved, by getting a computer to inspect a large number of randomly-generated cases.

      If the measured temperature profiles of the holes appear to be unrelated, then one raises a red flag and stops. Otherwise, one could calculate the set of possible solutions for the boreholes for the last N-D and N years respectively, where D is the period of time between drilling the boreholes, and truncate the longer reconstruction so that the temporal spans match.

      Assuming that
      a) the ratio D/N is small
      b) the number of solutions is countable
      c) the same number of solutions exist for residual temperature profiles T1 and T2 if the difference between T1 and T2 is small
      d) a 1-1 relationship exists between the individual solutions for T1 and T2 in (c)
      e) small changes in the measured profile lead to small changes in the multiple solutions for the reconstructions (do they?)

      ..then if the inversion is true, only one solution should remain relatively unchanged (it may be necessary to do more than two sequential drills).

      Additionally, but more tenuously, if one has good confidence in the inversion (possibly because the exercise described abpve has already been carried out and the putative technique for finding the right solution is found to be valid), then instead of drilling a 2nd borehole, one could extend the reconstruction with the real temperatures (adjusted to fit the variance of reconstruction) measured for the period since the 1st borehole, and then cast back to form the residuals for the 2nd now virtual borehole (& 3rd etc). The analysis between the boreholes can then be carried out as before. (For this lazy method, there’s no need to drill a second borehole, but time still needs to pass to gather the temperature readings).

  27. Dany
    Posted Apr 8, 2010 at 3:47 PM | Permalink

    Interestingly, Valerie Masson-Delmotte is currently organising with J. Jauzel a sort of reponse of french scientists against alledged unfair treatment by sceptics ; they wrote a petition (400+ names) to the science and research minister, and to any research institute in France ; the minister suggested a fair debate through the French academy of science ; the group of 400 is supposed to publish a 50 page paper (expected on April,7, but apparently late) ; this comes after a book by Courtillot (end of 2009) and one by C. Allegre (the global warming scam), both denouncing the GW hysteria.
    The insight given by Steve’s most recent posts on activities by Masson-Delmotte and Jauzel in the course of AR4 reviewing and final drafting is consequently particularly interesting as they claim to constantly apply pure scientific standards in their work whilst Courtillot and Allegre would not.

3 Trackbacks

  1. […] “hiding the decline” 6 04 2010 Steve McIntyre points out some inconvenient data from Law […]

  2. By Top Posts — WordPress.com on Apr 6, 2010 at 7:28 PM

    […] IPCC and the Law Dome Graphic Re-reading Climategate and AR4 Review Comments, I noticed an interesting discussion about handling the Law Dome O18 […] […]

  3. […] In passing, I’ll remind readers of my efforts as an IPCC AR4 reviewer to get the IPCC authors to show the Law Dome data (which I’d received in March 2006 just before review comments) – see CA post here. […]

%d bloggers like this: