2010-10-21 05:05:05New article on GISP2 - Need feedback
Rob Honeycutt


I still need to pull together the figures and a few links but you'll get the gist of it.  No references to TIS this time.  

Please don't hesitate to be tough reviewers. 


Getting to the Crux of a Core.

Ladies and gentlemen I present to you another hockey stick.  This one may have slipped by you previously but there it is in all it's glory.  Well, actually it's slightly different than a real hockey stick.  It's more like a sock.  Let's call it a "Sockey Stick."  Here it is.  All the evidence you need to show that the earth is warmer today than in the past.

Fig 1  -  The "Sockey Stick"

Let's pull back a little further.  Yikes!  Um, I'm a little embarrassed.  It looks like maybe this one doesn't show current warming as being warmer than the past.  That, there, must be the infamous MWP.  Curse you MWP!

Fig 2  -  Sockey stick with MWP

Let's pull back even further.  Oh no!  More periods of even warmer times in the past?  What's going on?  My sockey stick is unraveling before my very eyes!  

Fig 3  -  Unraveling the sockey stick

Dare I pull back any further?  I'll have to cover my eyes.  [peeking]  Oh no!!!  This just can't be!  My world is falling apart.  It's all got to be a big lie.

Fig 4  -  4000 years of the GISP2 ice core record

OMG!  It's real.  This is Dr. Richard Alley's actual data.  It has to be real (we know none of the lead authors of the IPCC report would ever fabricate data).  

So, there you have it.  The sockey stick is fully unraveled.  This is proof positive that there really is a huge cabal of scientists all conspiring with the UN to create a new world order and take away the livelihoods of honest, hardworking fossil fuel industry CEO's.  Ken Cuccinelli is the hound dog who has the scent and is on the trail of the perpetrators.  It's as clear as the sock on my foot.  It's those gal-dang socialists!

Okay, I can hear the symphony of rolling eyeballs, palms slapping foreheads and deep heavy sighs.  This site is "Skeptical" Science so I know you guys well.  You're being appropriately skeptical.  You may be viewing this first bit I've written here as a touch of comedy but this is essentially exactly what I have encountered here and here and here and here and here and here. [need to fill in links for all the 'heres']  If there is any comedy it is a very dark comedy because I, for one, have run into many many people on the internet who believe exactly this with all their conspiratorial hearts.

But, being a skeptical site let's try to get to the crux of this core.  


First let's examine where the data for this chart comes from.  GISP2 stands for the Greenland Ice Sheet Project 2.  It was a 5 year drilling project on the Greenland ice sheet that reached "3053 meters in depth, the deepest ice core recorded at the time."  The work was completed in 1993 including some 30 different study programs.  It is obviously the result of a lot of hard work by a large number of scientists and support staff.  The GISP2 home page link above lists many of the people and projects.


Fig 5  -  Location of GISP2 at the summit of the Greenland ice sheet 

Step 1 in our skeptical exercise here is to do the obvious.  Examine the GISP2 data on the NOAA website.  The graph used in each of the examples cited above is an accurate representation of the GISP2 data.  But when looking at the actual data the first thing we find is that the data ends at "0.09514... thousand years before present."  Being that the paper is presented in 2000 that means at very least the end of the graph is representing the turn of the 20th century.  Maybe 1905.  So, our sockey stick is not really a sockey stick that bears any relationship to warming from man made CO2.  

Step 2, being a non-scientist, was a little more of a challenge for me.  The red flag that went up for me was whether GISP2 could be considered a proxy of global temperature, as all the the presentation of the GISP2 data noted above seem to imply.  NASA's Earth Observatory provides this explanation of how the temperature data is collected from ice core samples.  The measurement of O16 and O18 ratios give us what is considered to be a very accurate measurement of temperatures.  But the example in this article seems to be primarily looking at how these are measured in the Antarctic ice cores and it doesn't give me a strong sense of how widespread or localized the temperature readings are.  They suggest that mountain glacier measurements are more complicated and I would imagine that Greenland has it's own set of complicating factors.  As well, is the temperature proxy reading a function of the distance the isotopes have traveled or are they a function of the point where they are frozen into the ice?

So, I opted to pull a tactic straight from Dr Abraham's playbook.  I wrote an email to Dr. Alley with a couple of brief questions.  What a guy!  He responded.

One question was whether GISP2 could be considered a proxy of global temperature.  Richard said unequivocally, "GISP2 is GISP2, not the world."

He goes on to say, "Parts of GISP2 look like the north Atlantic--the Younger Dryas was cold in many places including GISP2, but it was warm in Antarctica.  Parts of GISP2 are noise--a snowdrift may give the impression of some more important event."  He also did a write up for Andrew Revkin at the NYT explaining more of this in detail.

I also asked Dr. Alley about the end date of the data.  Another SkS poster suggested to me that radiometric dating generally references 1950 as the "present" in the term "before present."  Richard confirms this saying that, "GISP2 used 1950 as the reference."  

So, what we have here is a picture that is vastly more complex that is being presented on the internet using GISP2 data.  We live in a time when people want clear, simple answers.  Good cop/bad cop.  Black and white.  Either it's yes or no.  Real science doesn't operate in absolutes.  Real science doesn't always make things more simple.  When we get truly skeptical in science a more complicated - but also more accurate - picture begins to emerge.  And the picture always leaves the door open to better explanations.

What we have with GISP2 is a 50,000 year temperature record of a single location on the planet.  It is not the entire planet.  GISP2 also stops in the mid-1800's and therefore shows no data related to modern warming trends.  GISP2 is a single, yet very important, piece of the vast paleoclimate puzzle.


I've only taken 2 very small skeptical steps here to get a better understanding of one ice core project.  Below I'm including a list of the many ice core projects around the world.  The amount of digging one could do to learn more about these projects and ice cores in general is astounding... more that that, it's exciting!  I can think of few things so exhilarating as expanding my own knowledge of this very complex topic.


I want to extend special thanks to Dr. Alley for taking time from his busy schedule to respond to my email.  My socks off to ya, Dr. Alley!


Additional reading:


Other ice core drilling sites...


Station Eismitte, Camp VI, Station Central, Site 2, Camp Century, North Site, North Central, Crete, Milcent, Dye 2, Summit Camp, South Dome, Hans Tausen, Camp III,, Dye 3, Renland, GRIP, NGRIP, NEEM


Plateau Station, Byrd Station, Dolleman Station, Berkner Island, Cape Roberts Project, ITASE, Lake Vida, Vostok, EPICA/Dome C, Dome F, WAIS Divide, TALDICE

Non-polar cores:

Upper Fremont Glacier, Nevado Sajama, Huascaran, Quelccaya Ice Cap, Mt Milimanjaro ice fields, East Rongbuk Glacer

2010-10-21 05:08:08
Rob Honeycutt

Shoot.  I do not know what I keep doing wrong to mess up these links...  Sorry.
2010-10-21 05:18:29
Rob Honeycutt

Okay.  I think I fixed it...
2010-10-21 05:28:39

You gotta do something about the abstract at the end. For people not accustomed to reading scientific papers, that is like eating a 3-pound steak: It will sit in the gut, indigestible, for hours.
2010-10-21 05:33:21
Rob Honeycutt


Neal...  I agree.  I found it interesting to read again because it seems as though what is important about GISP2 has little to do with temperature anomalies and more to do with climate events like the Younger-Dryas.

But, you're right, it's probably superfluous. 

2010-10-21 05:42:00

Otherwise, it's not possible to predict the reaction of the reader without seeing the graphics. They will tell the story, in this case.
2010-10-21 07:11:35
Rob Honeycutt

Neal...  I'll rough out some graphics right now but it's going to take a day or so to pull together something better.
2010-10-21 07:22:35Question: Adding an explanation of what GISP2 actually stands for?


Hi Rob,

I really like this write-up, both as far as its style and its content goes. The only suggestion I have is to include an explanation of what GISP2 stands for. There always are quite a lot of abbreviations (if that is what they are) for these various graphics and it might help to better relate to them if there were at least one upfront mention of what they mean or stand for.

The NYT-article you link to contains the explanation: Greenland Ice Sheet Project 2 but as this is a link in a link not many will bother to go there. Could it make sense to copy the short explanation provided on the GISP2-website into your article (or at least reference it directly via the named link?

Thanks and Cheers

2010-10-21 07:30:57
Rob Honeycutt

Thanks Baerbel...  Will definitely make that adjustment.
2010-10-21 08:59:49



The typical reader is going to stop at Figure 4, just as the nefarious schemes of the climate scientists have been exposed; because Figure 5 looks like it's going to be hard work to understand; and the tone of the writing goes from excited to detached. It's a sure signal to the reader, "You've gotten the best part of the article already - why bother with the rest?"

Unfortunately, "the rest" is where you've put in the science.

Instead of giving the reader a cold bath, the first paragraph after "But, being a skeptical site let's try to get to the crux of this core." SHOULD explain quickly and clearly that the graph ends in 1905; and then should come the next figure, showing what the curve looks like when you include the last 105 years. THAT has got to be the climax moment, not Figure 4. Otherwise, I swear on Odin's nose that 85% of the readers will believe you've just proven GW is fake. You cannot just explain the error in words: You have invested 4 figures in building up the skeptic story, so if you don't blow it away with a shocking figure for the GW story, you will have shot yourself in the foot. Own goal, anyone?

That was step 1. For step 2, you take much too long to get to the main point: One or two locations is just one or two locations, they're not global averages. This is a simple idea, that should not take a long time to state (particularly since you haven't provided any explanation to justify it beyond "Dr. Alley says so"). You would be better off using that extra time explaining why & how these data are STILL useful DESPITE being only one or two locations, and what we learn from them.


2010-10-21 09:26:33
Rob Honeycutt


Really great feedback Neal.  

I think I need to let it rest a few hours and approach how better pull the reader along.  It also feels like it's starting to get a little too long.  I need to trim and strengthen the whole thing.

I'm also looking at this graph that comes from wikipedia to see if this helps to tell the story that the GISP2 data is only one line of evidence among a great many data sets.


I was also contemplating writing a companion piece to this where I attempt the daring feat of adding modern temps onto the GISP data. 

2010-10-21 10:06:51Much better treatment
John Cook


This is much better than the original response to TIM as it addresses several uses of the GISP data including that YouTube movie doing the rounds. By going general, you get a lot more value for money for your post - it can be used on multiple occasions.

I think your Figures 1 to 3 need to be plotted properly with X-axes rather than just blowing up the final graph. Do you have the data in Excel?

Good feedback from Neal - a picture does tell the story more vividly than words. However, even if you add the instrumental record to current time, isn't past periods still greater than today? So the take-home message is still that GISP is a single location, not global temperature. Perhaps comparing GISP to a global proxy would be a good way of showing the contrast - that GISP shows much greater variability than global temperature. Just a thought.

2010-10-21 10:53:44
Rob Honeycutt


John...  Right now these are just place holder images.  Often what I do is pull a diagram into Illustrator and redraw the graph.  

I agree, this is way better than butting heads with Kehr.  (I'm doing that over at his site.)  

Definitely the past data for GISP2 is warmer than today still by a couple of degrees but I think the data sets are not the same.  I just did a graph using just NH data, but that really is still not an apples-to-apple comparison.

In this article I think the gist needs to be that GISP2 is local temps and has to be taken in the context of a much MUCH larger array of other data. 

2010-10-21 16:28:26
Ari Jokimäki


So, did you check from Dr. Alley about the dates? Like I showed in the previous thread, the common meaning in scientific literature for "before present" is before 1950. That would make the GISP2 data to end at 1855. At any case it is not exactly correct when you say:

"But when looking at the actual data the first thing we find is that the data ends at "0.09514... thousand years before present."  Being that the paper is presented in 2000 that means at very least the end of the graph is representing the turn of the 20th century.  Maybe 1905."

While that might be true that Alley might have presented the data in "before 2000" form (against the normal practice), you still need to explain it here. Here you make it seem as it would be obvious that the "before present" in your reference means before the publication year of the data. But it is not obvious. If one would have to decide the end year just based on the comment you quoted ("0.09514... thousand years before present."), the obvious interpretation would be that the end year is 1855 because the standard "before present" is 1950.

2010-10-22 02:56:13
Rob Honeycutt


Ari...  Yes, you were exactly right.  The GISP2 data is referenced to 1950 according to Dr. Alley.  Great call!

On a further note, last night John Kehr (TIS) opted to change his header graphic to a different ice core record that, according to him, "Shows exactly the same thing."  But again, I'm taking him to task because he's using a single location as a proxy for global temperature.

I'm almost done with another rewrite of this piece.  I think it will be ready for prime time later today. 

2010-10-22 03:51:16

I really like this approach. I suspect even some skeptics-1 will find it intriguing and thought-provoking.
2010-10-22 04:51:06
Ari Jokimäki


Rob: " The GISP2 data is referenced to 1950 according to Dr. Alley."

Thanks, that's good to know. I browsed around on this issue and kept finding references to GISP2 data using the standard B.P. of 1950, but I didn't find anything that would have proved it for the temperature reconstruction. Hmm... I think I'll tweet this peace of information because it seems to be quite commonly unknown.