Climate Science Glossary

Term Lookup

Enter a term in the search box to find its definition.


Use the controls in the far right panel to increase or decrease the number of terms automatically displayed (or to completely turn that feature off).

Term Lookup


All IPCC definitions taken from Climate Change 2007: The Physical Science Basis. Working Group I Contribution to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Annex I, Glossary, pp. 941-954. Cambridge University Press.

Home Arguments Software Resources Comments The Consensus Project Translations About Donate

Twitter Facebook YouTube Pinterest

RSS Posts RSS Comments Email Subscribe

Climate's changed before
It's the sun
It's not bad
There is no consensus
It's cooling
Models are unreliable
Temp record is unreliable
Animals and plants can adapt
It hasn't warmed since 1998
Antarctica is gaining ice
View All Arguments...

Keep me logged in
New? Register here
Forgot your password?

Latest Posts


Climate Hustle

Crux of a Core, Part 1 - addressing J Storrs Hall

Posted on 26 February 2011 by Rob Honeycutt

Over the past couple of years perusing the internet on climate issues I have repeatedly come upon various misrepresentations of the GISP2 ice core record here and here and here and here and here and here and here and here and here. (I could literally list at least a 100 more locations where I find this exact same argument on the internet. You can google: "hockey stick gisp2" and see.) Each misrepresentation misses the very basic point that an ice core - any ice core or any other proxy - is a local record of temperature.  Using a single ice core record as a proxy for global temperature is a little like reading the thermometer on your back porch and claiming you know the global average temperature.

Fig. 1: GISP2 as presented on Watts Up With That, conflating a local record with a global record.

One of the major challenges in today's world is that misinformation pops up and propagates like rabid bunnies before the scientific community can effectively address the erroneous information.

The original source of this specific misinformation seems to come from J Storrs Hall, a nano technology engineer from the Foresight Institute. When this blog post hit the internet it quickly made the rounds to all the popular climate skeptic blogs and is now a permanent resident of the "interweb" and continues to misinform people.

Let's get to the crux of this core. First, let's look at where it's located. We find the core comes from the summit of the Greenland Ice Sheet at 72.6 N 38.5 W at an elevation of 3200 meters. This is well above the Arctic Circle and very high in altitude as well. The location is logical because this is where one would expect to drill in order to find the oldest ice possible.

Fig. 2: Google Earth image for the location of the GISP2 project.

Hall presents GISP2 as if it were a global record and makes no attempts to clarify that it is not nor does he even hint that he has any inclination that this is the case. On a whim I decided to send a brief email to Dr Richard Alley, one of the principal investigators for the GISP2 temperature reconstruction. He very kindly abliged me with a response saying, "GISP2 is GISP2, not the world" and also directed me to an article on Andy Revkin's DotEarth blog where he addresses the issue more in depth.

Let's not just take Dr Alley's word from the NYT. What we can do is look directly at the research. In Alley 2010 History of the Greenland Ice Sheet: paleoclimatic insights we find in section 2.2.6, they state "Preferential condensation of the heavier species [of O18 isotopes] causes them to be progressively depleted in air mass, and thus in precipitation, with cooling. Although linked to a site temperature, O18 is affected by the seasonal distribution of precipitation and other factors (Joezel et al., 1997; Alley Cuffey, 2001), requiring additional paleothermometers." This clearly tells us that oxygen isotope ratios (like the ones used for GISP2) are measuring a local record of temperature for the summit of the Greenland ice sheet, and are not a global proxy.

But, what does this mean? Still we are left with data that suggests the Holocene, even if it's just northern latitudes or even just Greenland, has been far warmer than today. When we begin to bring in more lines of evidence we get a clearer picture of the Holocene. More importantly, we actually understand that the gradual cooling since the Holocene Thermal Maximum (HTM) has been driven by changes in the tilt of the planet, so aptly demonstrated via Dr Alley's bald spot (and thereby giving new purpose to my own polar thinning). 

Fig. 3: Composite diagram of Holocene temperature records (data sources here).

In Miller 2010 (section 12.2) they describe the cooling trend over the past 6k years as a "Neoglaciation." They state, "Where quantitative estimates of temperature change are available, they generally indicate that summer temperature has decreased by 1-2C during this initial phase of cooling." This is a far cry from what Mr Hall is attempting to indicate with his presentation of GISP2 data, and arguably we have now warmed the planet again - in spite of an early phase of obliquity forced glaciation - by almost half the natural cooling of the Holocene. This again corroborates Dr Alley's statement during the subcommittee hearing that "Based on about 5 lines of evidence the HTM was about 1C warmer than today." Miller 2010 actually goes a little further and lists the following:

  • Agassiz Ice Cap (Moerner and Fisher, 1990)
  • Devon Island (Fisher 1979)
  • Greenland (Alley and Anandakrishnan, 1995)
  • Greenland (Vinther et al., 2008)
  • Indications from borehole thermometry (Cuffey et al., 1995)
  • Retreat of large marine mammals and warm-water-dependent mollusks from the Canadian Arctic (Dyke and Savelle, 2001)
  • Southward migration of northern treeline across central Canada (MadDonald et al., 1993)
  • and Eurasia (MacDonald et al., 2000b)
  • and Scandinavia (Barnekow and Sandgren, 2001)
  • Expansion of sea-ice cover along the shores of the Arctic Ocean and Ellesmere Island (England et al., 2008)
  • and in Baffin Bay (Lavac et al., 2001)
  • and in the Bering Sea (Cockford and Frederick, 2007)
  • Shift in vegetation communities inferred from plant macrofossils and pollen around the Arctic, including Wrangel Island (Lozhkin et al., 2001; Bigelow et al., 2003)
  • Assemblage of microfossils and the stable isotope ratios of foraminirera (Jennings et al., 2002)
  • and the western Nordic seas (Koc and Jansen, 1994)

(Note: These are not directly related to the diagram in Fig 3.)

If we take a quick stop back at the IPCC to look at radiative forcing from GHG's you can easily see why the planet has warmed since preindustrial times and created a reversal of this neoglaciation.

Fig 4: Chart from IPCC AR4 Working Group I: The Physical Science Basis (Scale: 10k yrs to present)

So, what's the take away?

  • GISP2 is clearly a local record of temperature for the summit of the Greenland ice sheet, not a proxy for global temperature.
  • It requires looking at multiple lines of evidence to piece together a complete picture of the Holocene.
  • The Holocene shows a very slow, gradual cooling trend over the past 6,000 years but the mechanisms behind the cooling are well understood.
  • The cooling during the past 6,000 years globally is on the scale of 1-2C and we have abruptly altered the trend and are now pushing the planet toward warming.

As Churchill is once purported to have said, "A lie gets half way around the world before the truth has a chance to get its trousers on." This was never more true than with J Storrs Hall's misrepresentation of the GISP2 data.

It's our job, folks, to help the truth get its trousers on.


In Part 2 of this article I will address misrepresentations coming from Dr Bob Carter's use of the GISP2 ice core.

- - - - - - - - - - -

Special thanks to Dr. Alley for taking a few moments out of his busy schedule and very important work to answer a few questions. 

Be sure to check out Peter Sinclair's Climate Denier Crock of the Week video on this same topic.


0 0

Bookmark and Share Printable Version  |  Link to this page


1  2  Next

Comments 1 to 50 out of 51:

  1. I struggle to see how anyone remotely intelligent, and presumeably Watts is intelligent, could blend a single local record with a global record and seriously present that as a scientific argument. Its a con job.

    There are two groups of people on this climate change issue. Only two groups. People who only care about making the most of now, and those who also care about the future and future generations. There could be a real reckoning on the whole issue.
    0 0
  2. As i remember average temperature of last 10 years at Summit is about -28 C. 2010 was very warm with temperature -25 - -26 (from SYNOP reports).
    0 0
  3. I've read that Greenland is currently about 1/2 a degree cooler than the MWP (locally) -which agrees with Watts' graph- but that local climate conditions around Greenland have temporarily isolated the continent from about 1 degree of warming which remains in the pipeline as an inevitable consequence of warming of the northern hemisphere due to AGW.
    And of course, Greenland continues to warm. That's the real issue.
    0 0
  4. Thankyou Rob.


    I struggle to see how anyone remotely intelligent, and presumeably Watts is intelligent, could blend a single local record with a global record and seriously present that as a scientific argument.

    You are quite right of course. I would expect nothing better from WUWT but it's difficult to explain why the IPCC did precisely this when they published Lamb's Central England data in their FAR.
    0 0
  5. It's ironic that people refer to scientific analysis (eg oxygen isotope ratios) and willingly accept that from this, temperature can be estimated (and then proceed to fabricate extended conclusions). Yet the same people say 'climate science is a fraud', or words to that effect. It shows that such people are not at all interested in furthering their knowledge. They only want to sift the information provided by scientists and pick out bits and pieces they can misuse to further their political ends.

    BTW, ice cores give researchers a lot of information, not just oxygen isotope ratios.
    0 0
  6. XPLAIN:

    It's should be no surprise that this site addresses that particular argument about a 21-year old report here.
    0 0
  7. Jason Box, PhD from the Byrd polar research center, has as a blog here; - - showing the current 2010 greenland ice sheet temperature anomaly at 2.4 degrees above average.

    It's hardly good science to put in the current curve atop figure 1 for comparison, but no worse than slapping on the global trend for the same period. It would be intersting to see if some one could smooth the contemporary trend into the GISP2 curve to give a better idea where we're at currently.

    Failing that, getting his curve onto this site might be nice, at the very least.
    0 0
  8. Thanks for this article. The ScienceDirect links aren't working for me at this time, but here is a link for Alley_etal_2010_History_of_the_Greenland_Ice_Sheet__Paleoclimatic_insights.pdf (6.25 Mb). It's a beautiful paper scientifically and aesthetically. There is even a picture of ancient coral at Key Largo Florida.
    0 0
  9. nigelj @ #1.... Actually, I think a great deal of Watts' work is relying on the misinformation of others. This one originated from J Storrs Hall. Watts is just the central repository for wrong and poorly researched information.
    0 0

  10. don't mind me - testing the image posting code.
    0 0
    Moderator Response: [Daniel Bailey] Tip: When posting an image, it's considered good form to also post a link to the source for the image. Example for above: Source for image here.
  11. Thank you. I'm remain in denial about how often data is grossly misrepresented. I didn't know the GISP2 data was being abused, and quite elegantly, as shown in Hall's animated GIF.
    FYI: I've recently added GISP2 to my interactive Vostok Viewer. I've juxtaposed GISP2 with EPICA, and you can easily see the drastic mood swings of GISP2 against the gradual changes in EPICA.
    I look forward to Part 2.
    0 0
  12. @mod

    KK - cheers for that - probably not a bad idea since the original is 3 >< larger.
    0 0
  13. Daniel Bailey, would you explain how Comments Policy was violated by this post? I suppose you thought it was, because you have deleted it.
    0 0
  14. jg @11, GISP2 data is being abused far more than is made apparent in the above post. Considering Figure 1, one major error is that the end date (marked as the present day) is shown as 2000. In fact, by convention, the "present" is in fact 1950, and is 1950 in the GISP 2 data. What is more, the GISP 2 data actually finishes 95 years before "the present", or in 1855.

    Further, apart from the obvious fallacy of treating a local climate record with global temperatures, they do so inconsistently in that they use the global record for the last 100 years. If you used the local temperature record throughout, the graph would look like this:

    (Note, the two red crosses mark recent ice care temperature proxies - ending in 2009 - at the nearby GRIP drill site. The grey line marks the temperature of a similar increase in temperature from 1855 to the present at GISP2.)
    0 0
  15. Berenyi Peter @13, I would also like to know the answer to that question. The post seemed fine to me, and I was going to comment on it.
    0 0
  16. BP @ 13... Can I suggest that you also take a look at the Byrd station ice core as well?

    Once again, I think you're trying to pull out one data set that shows what you want to see without doing the full research to see what is going on. It's not a simple answer. Miller 2010 is the paper that addresses the neo-glaciation idea. I suggest that you read that paper.
    0 0
  17. Re: Berényi Péter (13)

    Your comment was deleted due to your contextual usage of the phase "if you actually understand". Those who have a history of pushing the envelope of compliance with the Comments Policy, as you have, will necessarily receive a greater level of comment scrutiny, and less benefit of the doubt, than others do.

    I reproduce your comment below sans offending phrase:

    Berényi Péter at 08:21 AM on 27 February 2011
    "This clearly tells us that oxygen isotope ratios (like the ones used for GISP2) are measuring a local record of temperature for the summit of the Greenland ice sheet, and are not a global proxy"

    I see. Oxygen isotope and Mg/Ca data from foraminifers retrieved from sediment cores in the western tropical Pacific Ocean are probably not global proxies either. It just shows that not only some pretty cold places got even colder, but one of the warmest spots as well.

    2004 Nature, 431, 56-59
    DOI: 10.1038/nature02903
    Decline of surface temperature and salinity in the western tropical Pacific Ocean in the Holocene epoch
    Stott, L., Cannariato, K., Thunell, R., Haug, G. H., Koutavas, A., Lund, S.

    However, if it is cooler today at both places than it used to be several thousand years ago while global average temperature is higher right now, it follows there should be other places on Earth where it was cooler back then. Could you kindly show us several such examples from the literature?

    More importantly, if... the gradual cooling in the tropical western Pacific since the Holocene Thermal Maximum (HTM), as it is shown in Fig. 3. Stott 2004 has been driven by changes in the tilt of the planet, please share this knowledge with us.
    0 0
  18. BP / Daniel.

    I was going to ask whether the "before present" represented the usual 1950 or some other year. So I had a look.

    "....To estimate the magnitude of the salinity change since the early Holocene we stacked the WTP SW18OC and Mg/Ca records, averaging the data at 250-yr intervals. ... "

    in the body of the text, and then this as the final sentence.

    "If so, millennial to centennial scale changes in Holocene ocean thermohaline circulation would be directly affected by ocean–atmosphere processes that have occurred in the tropics."

    These guys are interested in centennial to millenial periods, so a 250 year average is suitable for at least part of their work. Even if their 'before present' year is more recent than 1950, I cannot see where this work has any more than a tangential relevance to atmospheric warming of the last few decades.

    This paper seems to be yet another indication that the planet was on a steady, gradual cooling path from the Holocene optimum until excessive GHG release blocked that path.
    0 0
  19. BP The figure from Stott is an average temperature estimated using data from three sediment cores in the western equatorial Pacific Ocean. If you download the supplementary data from Nature you will see that the binned (averaged) temperature estimates shown in the figure stop at 125 years ago.

    Among the three cores there are a total of three individual data points assigned ages of less than 100 years, not very much to say anything about temperatures "today".
    0 0
  20. Hah. Adelady beat me to the point. Also, to reconcile what may appear to be differences in our statements, each bin (average temperature value) has a duration of 250 years (adelady), so the midpoint of the most recent bin is 125 years ago (boba).
    0 0
  21. I take it the Miller et al. paper (one really should add the et al., that is the least you can do for all the work that goes into a paper) is this long pdf: Miller et al., "Temperature and precipitation history of the Arctic", Quaternary Science Reviews 29 (2010) 1679--1715.
    0 0
  22. Beryényi Péter, are reproduced @17:

    1) The first point to notice is that Stott et al 2004 is one of the plots (dark red) reproduced in Figure 3 of the main article. The full list of data in that figure is here.

    You will notice that the Western Tropical Pacific only rose above the (global) 2004 temperatures anomally 5,000 years ago, and again 10,000 years ago. At those times, the only Southern Hemisphere data in the above graph (Vostok and Epica) were both significantly below modern temperatures. The net effect is that the average is noticably below 2004 temperatures. That probably overstates the Holocene Climactic Optimum temperatures in that there is a NH bias in the data.

    The point is, cherry picking two local temperature indices and claiming they represent the global temperature is no more principled than cherry picking one.

    2) As per the convention, the "present" in paleoclimate reconstructions is 1950. Stott et al do not mention more recent dates for their data, so the most recent datum on the graph is for 1950. As can be easily verified at GISS, current temperatures in the western Pacific are at least 0.5 degrees warmer than they were in 1950 (comparison of 12 year average centered on 1950.5 with 12 year average ending in 2010).

    Again, the Western Pacific has only been warmer than this a couple of times, and not during the HCO.

    The Western Pacific was between 1 and 2 degrees warmer in 2010 than in 1950, making it much warmer than any point on that chart, but issues of resolution mean this may not be an appropriate comparison.

    Finally, in response to your questions:

    a) Several examples of cooler locations at those times can be found in Figure 3 above. Indeed, GISP2 (light blue in figure 3) is anticyclical to the Western Pacific data, so that the large peaks in the Western Pacific temperatures coincide with troughs in the Greenland temperatures, so your question is based on a false premise; and

    b) The anticyclical nature of GISP2 and Western Pacific shows the majority of the variation to be regional in nature, and therefore likely to have primarily regional causes. So, while there may be some connection, you do not show that there is and your question is based on the false assumption of synchrony between GISP2 and Western Pacific temperature peaks.
    0 0
  23. adelady @18, as an addendum to my 22 based on your comments, the difference between the 2001-2010 average and the 1880-2010 average is 0.5 - 1.0 degrees in the tropical western pacific, and would be greater if the average where a 250 year average.

    By Tropical Western Pacific I mean the Western Pacific warm pool from which (by memory) Stott et al drew their data.
    0 0
  24. #17 Daniel Bailey at 12:27 PM on 27 February, 2011
    Your comment was deleted due to your contextual usage of the phase "if you actually understand". Those who have a history of pushing the envelope of compliance with the Comments Policy, as you have, will necessarily receive a greater level of comment scrutiny, and less benefit of the doubt, than others do.

    I see. So I suppose you actually understand why sea surface temperature is decreasing in the tropics during the Holocene (by 1°C or so). Unfortunately I don't.

    You can check at the NASA GISS Monthly Latitude Insolation page that annual average insolation at TOA (Top of Atmosphere) has been increasing steadily (by 1.2 W/m2) along the equator during the last 10 thousand years (from 416.2 W/m2 to 417.4 W/m2). Perhaps you could explain to all how cooling is caused by increasing irradiation.
    0 0
  25. Pete Dunkelberg @ 21... That is the correct Miller el al 2010. Definitely a long el al list on that one.
    0 0
  26. Berényi Péter @24, first, it is by no means clear that global temperatures have declined since the Holocene Climactic Optimum. Indermühle et al (2000) " Atmospheric CO2 concentration from 60 to 20 kyr BP from the Taylor Dome ice core, Antarctica" use the CO2 record as a de facto record of global temperatures, possible because of the changing solubility of CO2 with changing sea surface temperatures. Doing so shows an approx 0.5 degree increase in global temperatures since the Holocene Climactic Optimum.

    This method is not without problems, particularly due to increased CO2 concentrations due to deforestation over that period, however that can be corrected for by comparing rations of C13 to C12.

    Second, taking Figure 3 above at face value (although it is probably an over estimate of HCO temperatures as previously stated), there was a net decline in mean global temperatures of less than 1 degree C from the HCO to the preindustrial era. This May have occurred despite zero change in globally averaged insolation because of a decline in Northern Hemisphere summer insolation resulting in a cooler arctic, and more ice. The additional ice, of course, results in a higher albedo, and hence lower global temperatures.

    An equivalent effect occurs in Antarctica in reverse periods, but it is not as strong because at the relevant latitudes, Antarctica has ice sheets rather than sea ice, so there is much less of a change in albedo.
    0 0
  27. BP,

    I see that you have not decided whether or not to take issue with Dr. Alley's concerns and caveats, but rather try and obfuscate by moving the discussion from Greenland to the tropics. The data in question are those of Dr. Alley and he, more than you or I , or anyone else for that matter, is more qualified to speak to them.

    The point is, is that the GISP 2 data are being misrepresented by "skeptics" to misinform and confuse the public. Surely you do not endorse that? Do you?

    "You can check at the NASA GISS Monthly Latitude Insolation page that annual average insolation at TOA (Top of Atmosphere) has been increasing steadily (by 1.2 W/m2) along the equator during the last 10 thousand years (from 416.2 W/m2 to 417.4 W/m2)."

    That is interesting, I have no reason to doubt the NASA data, but these data do not seem to support the claim that annual tropical insolation has been increasing the last 10 K.

    Anyhow, back to Greenland please....
    0 0

    BP, Using this site linked by Albatross, the 10000yr change in insolation at the tropics has indeed risen at 30S south but fallen at the equator and at 30N. However at 65N which is the important one for determining Greenland summer temperatures fallen from 506w/m2 to 462w/m2. It is the far North temperatures which are the the apparent initial driver of ice age glaciation fluctuations from maximum to minimums since CO2 dropped below ~300ppm, so the earth should be cooling in the far North as all the reconstructions posted seem to indicate and as such due to natural feedbacks so should the whole globe as it did in glacials.

    Also note that average insolation for the earth hasn't changed at all in the last 10000yr but is slowly on the increase over millions of years (the faint sun paradox), people forget this when they peer into the past and it does mean as 55Myr (say) was alot hotter than today that other factors (GHG) must have been heating the earth proportional more then than now.

    Tom 22 and 14, just clarification on graph 14, if current Greenland top of icesheet temperature measurements are graphed they would be at the blue cross (in keeping with the graph shown in 10), 1950's would be the grey line?

    And in West Pacific graph posted from Stott 2004, the 0 or now point is actually 1950 and the real now 2010 or even 1990-2010 average is ~1-2C higher than 1950's and as such to actual present the graph should show a sudden steep rise to above 30C.

    Therefore these two isolated temperature records actual show that on top of the Greenland Ice sheet present temperatures are about 1C warmer than the last 8500yrs and in the Western Pacific for the last 15000yrs, this is keeping with Hansen's most recent estimates that the earth is probably already hotter than the last thermal maximum already.

    WHat was th eworld like in the pliocene cos that was the last CO2 concentration were as now, although as the sun is very very slightly more powerful now than then I suppose the earth might heat a little more eventually than the pliocene, of course put CO2 concentrations up 500 or more and well who knows!?
    0 0
  29. ranyl, the blue crosses mark the 1850-1859 average and the 2000-2009 average at the GRIP site. These temperatures are taken from Box et al 2009. Because it is an average, it does not rise as high as the exceptionally warm 2009 temperature, but the GISP2 ice core, of necessity from how the ice is formed, also represents a multi year average, so we should compare it with multi-year averages rather than single years.

    Gareth Renowden has confirmed with Box by personal communication that the GRIP site is typically warmer than the GISP2 site, by about 0.9 degrees C. On that basis, you could argue that the modern GISP2 temperature would be about -29 degrees C. However, Renowden took the difference between the 1850's temperature and the 2000's temperature at GRIP and added it to the end of the GISP2 record, with that postion being marked by the grey line. I think that is a better procedure. It would probably be better still to say that the modern decadal average is between 1 and 2 degrees warmer than the end point of the GISP2 ice core.

    And to correct an error I made, Box et al reconstruct their temperature from local surface station records rather than an ice core.

    Regarding the Western Pacific data, it shows 250 year averages (see Adeladys'comment @18), so comparison with individual years or even decadal averages is misleading. Modern temperatures are significantly above the last indicated point, but included in a 250 year average, would not significantly move the end point. While there have been no warmer centuries than the current decade, there may have been warmer decades that have been smoothed out by the average, for all we know. That mere possibility is not evidence for BP's position, but we should not neglect it.
    0 0
  30. The author of the post being rebutted, J Storrs Hall, has taken exception to not being addressed as doctor and feels this is a deliberate slight upon him.

    Perhaps the author would be so good as to address this as a courtisy to Dr Halls wishes?
    0 0
  31. The author of the post being rebutted has a "rebuttal to the rebuttal" posted at

    If you have an open mind and would like to be informed, I suggest clicking on over and taking a look.
    0 0
  32. "The author of the post being rebutted, J Storrs Hall, has taken exception to not being addressed as doctor and feels this is a deliberate slight upon him."

    Funny. I have read so many skeptic rants about appeal to authority, use of titles to suggest authority, how one does not need degrees to be qualified, what an expert is, etc. I had construed that skeptics did not care much about academic titles. Silly me.
    0 0
  33. Charlie A, put your serious skeptical hat on and closely read that post yourself. Spot the switch from absolute to anomaly? Perhaps read the papers instead of the quotes.

    As for "the only substantive point was..". Ho, ho.

    The zero point of 1855 isnt a substantive point? The actual temperature measurements? Does the original article still make sense when you look at the correct construction. (eg in #14 )
    0 0
  34. Charlie A @31, first let me second scaddenp's comment which is exactly to the point.

    Let me add, first, that Hall feeling slighted about not being called "Dr" is extraordinarily precious from somebody who dismisses climate scientists as "The Hockey Team" and dismisses (his straw man version of) their theories as "poppycock".

    More substantively, it is very noticeable that when he compares Antarctic and Greenland ice cores, he does so over the period of the Last Glacial Maximum. Unsurprisingly, there is a correlation between temperature differences between the LGM and the Holocene (the last 10 thousand years). But even during the last glacial there are clear anti-correlations between the three ice cores, even between the two Greenland ice cores on a scale of thousands of years. Therefore his failure to show a comparison of GISP2 and Vostok over the period of the Holocene with sufficient resolution is very damming.

    Such a comparison can be made in Figure 3 above. Vostock is shown in Dark Blue, GISP2 in light blue. If you look closely you will see that they are anti-correlated for most of the holocene. That means their average will have far fewer and smaller fluctuations then either seperately. The rapid rises and declines in the Greenland ice core during the Holocene are not global events, but the equivalently precipitous rise in late twentieth century temperatures is.

    Finally, he claims "The 20th-century warming was hardly unprecedented, and doesn’t call for unusual explanations." Let's leave aside that greenhouse warming is not an unusual explanation. The fact of the matter is that any change of climate needs some explanation. For global temperatures to change by up to a degree or more without explanation would violate conservation of energy. As it happens, the only substantive explanation is anthropogenic emissions of Green House Gases.

    (On a side note, Hall's PhD is in "computer science", which may well make him a skilled programmer, but it does not make him a scientist despite his claims.)
    0 0
  35. Thanks for that Tom. I had an unkind thought about how often certain people deign to put "Dr" before many well-known names.

    And "...greenhouse warming is not an unusual explanation... any change of climate needs some explanation." You're right. Far too many people of this ilk talk about global warming "theory" as if it were some discrete academic discipline. "Global warming" is just a sub-set of general climate science.

    Climate science shows that certain events and circumstances have predictable outcomes. The current circumstance, of releasing an unusual amount of GHGs at an unusually fast rate, is displaying fairly predictable responses. Usually called global warming. Those responses being in accordance with the general theory mean that we have a fairly high level of confidence that events in the future will unfold in accordance with the theory.

    More data, better technology, more experience working with the information we've already accumulated just makes it less and less likely that current evaluations of the historical record and any consequential projections for the near or distant future will turn out to be "poppycock".
    0 0
  36. scadenp says "Charlie A, put your serious skeptical hat on and closely read that post yourself. Spot the switch from absolute to anomaly?"

    OK. I followed the link in top of this Skeptical Science post over to the WUWT post Hockey Stick Observed in NOAA Ice Core Data.

    I see the ice core O18 isotope ratio data described as "It gives us about as close as we can come to a direct, experimental measurement of temperature at that one spot for the past 50,000 years." That sounds like a reasonable description.

    Hmmm. Perhaps the article here on SkepticalScience is tilting at windmills are strawmen.

    OK. Then there are a series of graphs for the central Greenland Ice core data. Varying lengths of record are looked at, but they are ALL in plots of estimated temperatures. The maximum length of data for that ice core is plotted.

    He then says "This next graph, for the longest period, we have to go to an Antarctic core (Vostok):" and shows that graph of temperatures. This one in with vertical axis being the temperature anomaly. The graph covers the entire period.

    Do you have some problem with using a plot of temperature anomoly when switching to another location?

    Why does that present a problem? Adding in a fixed offset to all of the graphs for the central greenland ice core would not, IMO, make any difference in the presentation.
    0 0
  37. Rob Honeycutt -- I'm trying to understand your article. In particular, I'm trying to figure out the source of your Figure 1, with your caption of "GISP2 as presented on Watts Up With That, conflating a local record with a global record."

    The only source I have found for that is the link you provided,

    In that link, however, the chart is clearly labeled "Data Source: GISP2 Ice Core, Central Greenland".

    It seems like you removed the "Central Greenland" label, and then accused Watts of conflating local and global records. Did you indeed do this?

    Please identify the source of your Figure 1, so your readers can judge for themselves what sort of conflation is going on.

    The other WUWT link you provided also very clearly talks about this being for 1 point on the earth.
    0 0
  38. Dr Hall shows us temperatures up to 100 years ago and demonstrates that local temperatures had been at least 3C warmer than it was 100 years ago therefor global warming is 'poppycock'.

    This is why they are skeptics. "If it can be shown to have been warmer at some point during the holocene than it was 100 years ago, atmospheric physics is a con job."

    Lets not forget Wattsup is the website that described all those paleoclimatologists who they describe as 'The Team' as belonging to a "Marxist orginisation".
    0 0
  39. Charlie A @37, I'm sorry, I missed that part of Hall's article where he said, "Although I am using a regional temperature only, I am comparing it to a global temperature. Global temperatures show much less variation than regional temperatures because different regions of the world will be warming while others cool. Therefore any comparison between global and regional variation is essentially meaningless. But here goes..."

    I also missed that part where he said, "Although nineteenth century temperatures are much cooler than 20th century temperatures, I am treating the final temperature in this series, from 1855, as though it was representative of late twentieth century temperatures for rhetorical reasons."

    I especially missed that part where he said, "It is known that temperatures in the last decade are at least 1.44 degrees higher than the last point on this graph at the location of the ice core. This makes the late twentieth warming one of the most abrupt on the record, with the distinction that it, unlike the others on this record, was global rather than regional in extent."

    In fact, the only bits I noticed was where he persistently compared local to global temperatures as though that was a reasonable thing to do.

    If he had wanted to do an honest analysis, his graph would have looked like this:

    (With thanks to Gareth Renowden.)

    The reason he didn't actually draw modern local temperatures on the graph is because it would have made it perfectly clear how utterly laughable is his entire argument.
    0 0

  40. The above is from WUWT.
    Note the "Central Greenland: caption in the upper right.

    Compare with the Skeptical Science version, below:

    Fig. 1: GISP2 as presented on Watts Up With That, conflating a local record with a global record.

    Notice the removal of the "Central Greenland" caption.


    Let's move on into the article a couple of paragraphs:
    Rob Honeycutt claims "Hall presents GISP2 as if it were a global record and makes no attempts to clarify that it is not nor does he even hint that he has any inclination that this is the case."

    A simple scan of the article Honeycutt links to,Hockey Stick Observed In NOAA Ice Core Data shows the error of Mr. Honeycutt's statement.

    The data is introduced as "It gives us about as close as we can come to a direct, experimental measurement of temperature at that one spot for the past 50,000 years."
    0 0
  41. I think you are totally missing the point in regards to people concerns regarding the above wuwt curve. It seems apparent that the 'instrumental record' displayed isn't data from the local site, or even the Greenland ice sheet as a whole (which has warmed 2 to 3 degrees over the past century); but appears to be a rather abstracted version of the global record as a whole.

    Whilst you may argue, with a degree of legitimacy, that the global temperature been used for comparison before, this in no way detracts from the fact that it's use on a strictly local curve is highly spurious. Especially when other, much more relevant, data is available, and particularly especially when that data shows a far greater degree of parity with any temperature spikes over the past 10K years since the last glacial period (and before that as well, for obvious reasons).

    Using a global average to compare with a specific local site is never robust - in most cases all it demonstrates is misleading at best, flat out fallacious at worst. That you make plain that those demostrating this are clearly aware of the data's purely localized nature, only serves to cast doubt on the agenda of such a presentation.
    0 0
  42. 70rn, your comments about splicing a global instrumental record onto a single point ice core record are valid. That discussion is the result of a comment.

    That's not what this Skeptical Science article is about.

    Please go back up to the top of this page and read what Rob Honeycutt wrote.
    0 0
  43. The WUWT presentation tries to make the argument that 'global warming' is unprecedented by using a local curve. Whether they identify as such or not still calls into question the entire methodology. You can semantically scrutinize the above as much as you like, and it is possible to argue (fairly enough) that the author is false to presume that the originators tried to hide it's local nature.

    But inspite of all that it is very hard to prove much with one local proxy - and much of the discussion has produced graphs that demonstrate a far more convoluted local/global relationship (e.g fig 3, from the article). This both shows a far more robust array of evidence, and suggests strongly that the current warming trend isn't the usual holocene pattern. Perhaps more importantly it puts current global temperatures against a far broader spatial variety of data curves - which has a greater overall validity. Honeycutt's primary point seems to be that the overall trend is down - by 1 - 2 degrees over the last 10 K years - but that has largely been arrested in a single century or so.

    tbh I find much of the argument about what was said about wuwt to be strictly secondary, but addressing their methodologies less so. Another example is the article here which appears to suffer from a distinct lack of present sight - as it seeks to use the curve to cast doubt upon gw theory with the ice core data, without including any data from the period that global warming is said to have occurred. Small gesures are made along the lines of 'it went up .5 of a degree more', which is not consistent with temeprature records from that area. These kinds of techniques are far more salient an issue than whether or not the above article is wrong to suggest a lack of transparency on the part of these third parties.

    To me the above article was more serving to give a wider context for the GISP2 curve. Sure the sniping is probably misplaced - but I don't think, judging from the concluding dot points - that it was really it's primary intent.
    0 0
  44. Charlie A @ 40... I have to say you are quibbling over an extremely tiny point that has no bearing on the overall statements made in the article. I can go back and create a new screenshot of the chart. When I did the original screenshot I was merely trying to make the chart larger and clearer for the reader. No one is challenging the source of the data. It's the GISP2 data from Alley 2000. Hall says so, and I agree.

    You also have to realize there are hundreds of versions of this same chart floating around the internet.
    0 0
  45. Charlie A @ 42... No. That is exactly what my article is about. Conflating a local record with a global record.

    I state in the very first paragraph, "Using a single ice core record as a proxy for global temperature is a little like reading the thermometer on your back porch and claiming you know the global average temperature."
    0 0
  46. Charlie A, your whole case comes down to not understanding the difference between attribution and analysis. Rob Honeycutt never says that J Storrs Hall did not correctly attribute the GISP2 ice core to the GISP2 site. He clearly indicates, however, that in his analysis, Hall treats the GISP2 records as though it were a global record.

    That Hall does treat the GISP2 record as though it were a global record is obvious in his comparisons between the variability in the GISP2 ice core and the recent rise in global temperatures. That comparison is explicit in Hall's article in which he refers to the "hockey stick", and in which he says temperatures continue "... up in the 20th century at least another half a degree."

    To start with, global temperatures rose by 0.8 degrees from the 1850's to the 2000's (the actual interval between the end point and modern times), or by 0.9 from the 1900's to the 2000's (the interval Hall thought to exist). So even if a comparison between local and global temperatures were valid, Hall has understated the temperatures increase by 40%. That sort of misinformation seems not to vex you, but as it is part of a critical point at issue, it shows a complete disregard for accuracy in Hall's analysis.

    More to the point however, as Hall is using a local record, he should have stated the additional increase in local terms. In other words, the state increase should have been 1.44 degrees (1850's to 2000's decadal average) or 3.7 degrees (1855 to 2010). In either case, the rise in Greenland temperatures in the 20th century would have had only two parallels in the chart, both (I believe) caused by events relating to the melting of the continental Ice Sheets following the preceding glacial.

    Now, either Hall new the difference between local temperatures and global temperatures or he didn't. If he did, then he knowingly understated the 20th century temperature rise by a factor of between 3 and 6 on a crucial point in his analysis; and is consequently a liar. If he did not, then Honeycutt's claim that "Hall presents GISP2 as if it were a global record ...". By arguing that Honeycutt is wrong about that, you implicitly argue that Hall is deliberately dishonest. There is no way around that logic.

    Honeycutt, not one to stoop to ad hominem, has taken the charitable interpretation that Hall is confused on the difference between local and global tempertaures. You apparently are not so charitable. However, whether Hall is dishonest or incompetent is a side issue. Your restricted focus on that issue alone shows that you are trying to distract readers from Honeycutt's devastating critique of Hall's argument.
    0 0
  47. I am not sure if this is mentioned in the comments for this or some or they other Crux of a core posts, but I see another problem in comparing ice cores with instrumental records. Ice cores have another sample rate than instrumental records.
    Instrumental records have a temporal resolution of 1 month or year while ice cores have measurements seperated in time in the magnitude of decades.
    For GISP2 the shortest interval between to measurements is 2.7 years and the largest is 79.67 years.
    This especially a problem when one looks at warming rates, as it is done in this blog post.

    I am interested if you guys here agree on my point?
    0 0
  48. The graph looks like that particular spot in Greenland should have melted away and not been there several times in the past 8000 years. Have the top few feet of ice core been dated by some method other than by counting layers?

    Did the altitude change? I think I missed something. Why does the graph have those peaks above the present? It couldn't have gotten that hot in central Greenland. Too much ice would have melted, throwing off your dates.

    Greenland ice should have melted completely between 8000 and 6000 years ago, leaving no ice sheet. There must be a correction factor somewhere. Otherwise, how could you have such a deep core?

    0 0
  49. Asteroid Miner @48, I am unsure as to which graph you are discussing, but presume it to be one of the many graphs of the GISP2 icecore.  If you look at the first such graph (at the top of the OP), you will see that the temperatures vary between -32.5 and -28.5 centigrade, sufficiently cold to keep the ice frozen.  That is slightly misleading due to the low resolution of the graph.  Ngheim et al (2012) document approximately 65 ice melt events at that site in the last 10,500 years, however some years are missing from their documented periods so that the number may be greater than that, but certainly less than 100.   The melt layers are very shallow, representing single seasons and in the 2012 season, just two days in July.  There is a possibility but no reason to expect that several years data could be melted in particular cases.  This presents no major problem because of the low resolution of the records, and because layers deeper in the core are cross correlated with datable events due to volcanic ash in the ice cores. 

    0 0
  50. Ateroid Miner - are you referring to Fig 3? It is a temperature anomaly graph (with respect to mid-20th C temp), not absolute temperature. Local absolute temperature would be in the negative 20-30s. And yes, multiple dating tools have been used on ice core.

    0 0

1  2  Next

You need to be logged in to post a comment. Login via the left margin or if you're new, register here.

The Consensus Project Website


(free to republish)

© Copyright 2019 John Cook
Home | Links | Translations | About Us | Privacy | Contact Us