ice cores and the temperture record

As far as I can tell, the article has a high baloney factor.

Opening sentence:

"1934 has long been considered the warmest year of the past century."

Forgot to qualify that with "for the US". It's never been considered the warmest for the globe. But the guys at WattsUp always manage to commingle the two. They have a real thing about 1934 and the US government covering up how hot it was.

"However, December 2010 has been one of the coldest Decembers in a century in many parts of the world, so 2010 probably won’t be warmer than 1998."

Is this for the globe? If so, it's wrong. The second-warmest was 2005, the warmest was 2010, per the GISS timeseries he's just gotten done sneering at.

The discussion of temperature trends (first chart) looks at C02 in isolation, which is clearly not smart.

Whoever this guy is, he certainly has an axe to grind.

Per Wikipedia, he's a retired geology professor who is predicting global cooling.

Then between the first and second graph, he makes the transition from global temperature to the temperature at the spot where a (?) Greenland ice core was extracted. Then compares the rates of change of the two. That's not legitimate.

Then he's got Greenland temperature data, and uses that to say that ... the earth ... has been much warmer than present for much of the last 10,000 years. Again, not legitimate.

It is well know that the high latitudes were warmer roughly 10,000 years ago, in the summer, than they are today. Part of the Milankovitch cycle. Termed the Holocene Thermal Maximum. Maybe that's what's driving his 10,000 year interpretation. I couldn't say. But that's wasn't an indicator of global temperature -- the sunlight that fell on the high latitudes didn't fall elsewhere.

Yeah, here's a graph from ... R.B. Alley, linking Greenland ice sheet melting to July insolation:

http://www.gisp2.sr.unh.edu/DATA/alley1.html

And Greenland shows variable due to ocean currents. So taking Greenland as your sole proxy for world temperature is not sensible. Let me put it this way. All the people who find the various hockey stick recontructions (hemisphere-wide or world-wide reconstruction based on a wide variety of sources) lacking certainly should not accept this one source/one location timeseries as a reliable proxy for global temperature.

At the minimum, I'd want to see what some responsible individual has to say. In particular, he attributes some of the data to R.B Alley. If you've ever seen one of Alley's lectures on this subject, you know that he thinks the earth is warming and that it's a serious problem. So, somehow, he's quoting Alley's data in a way that, as an indication of the Earth's temperature, I'm pretty sure Alley does not agree with.

So, at some point:

The US temperature is not world temperature. The Greenland ice core, in isolation, is not global temperature. Higher summertime temperatures in Greenland 10,000 years ago have been attributed to increased insolation. Temperatures in Greenland have been observed to fluctuate in the past.

Given all that, I don't think there's a lot of useful content there. Near as I can tell, this was just a pre-emptive strike when they realized 2010 was going to register as or near the warmest year in the instrumental record. It was pre-sneering at it, before the results were announced. Hah, only 9999th in the last 10,000 years, or some such.

The IPCC treatment of the paleoclimatology of Greenland is pretty comprehensive. If this is the topic, that's the place to start.

 

There are oxygen-isotope recostructions also from Antarctica, and from continental glaciers. I am not in the ideal situation (time wise) to search for a review article about that, so if somebody else does it I'll be happy to look on.

The Easterbrook guest post at WUWT certainly did have a few screamers, or at least some people would regard them as such. But the centerpiece was the graph mis-atributed to Alley. As I said, I read that Alley paper (among others) and it's not there. I posted a free link so others could see the Alley paper for themselves.

I don't think we can do much more with Easterbrook until we find out where it really came from. As I suggested in the t'other thread, I think we can find out from one of his new grad students.

If we do focus on the latest 10 k years, the Younger Dryas is surely the elephant i nthe room. It was a fascinating rapid climate excursion and has been extensively studies. The impact driver hypothesis has been losing support lately from what I've read (no shocked quartz grains recovered despite lots of searching) A glacial lake outburst flood (GLOF) from N. Amer. into the N. Atlantic, temporarily squelching the Atlantic salt conveyor, is currently the flavor of the month (to somwhat minimize the seriousness of scientific discussins of cause and effect).

A GLOF would be an event of Biblical proportions! Happy to have missed it. But the scablands in Washington State are a good place to go today and see the scale of 'revisions' that a GLOF can cause.

 

Greenland and Antarctica correspond.
So thats 2 areas from different sides of the Earth.
There are also recent temperature proxy studies from sediment which correspond with the latter few thousand years of the ice cores done closer to the Equator.Ill try to find them.
The only thing that doesnt correspond with ANYTHING is the hockey stick.Because its SciFi.

 

Sorry, but I'm just not following this. The original article was clearly incompetent for the reasons I stated. You can't compare US temperature to global temperature to temperature at single location to make your case. No amount of lawyerly wordsmithing changes that.

Do you all think that, somehow, the notion that there were warm temps 10,000 years ago is, like, unknown? Despite the term "holocene thermal maximum"? Or that hockey-stick-type multi-proxy reconstructions somehow fail to show it, but that the ice cores are the true data, or something? Look up some pictures, at least. e.g.,




It's just pretty well established that it's wrong to use the poles, in isolation, as a proxy for global temperature, as outlined here:

6.5.1.3 Was Any Part of the Current Interglacial Period Warmer than the Late 20th Century? - AR4 WGI Chapter 6: Palaeoclimate

Do you think that scientists don't run their models to see how well they match what is known about past climate? Read just the first sentence or two from these:

6.4.1.3 How Realistic Are Results from Climate Model Simulations of the Last Glacial Maximum? - AR4 WGI Chapter 6: Palaeoclimate

IPCC Third Assessment Report - Climate Change 2001 - Complete online versions | UNEP/GRID-Arendal - Publications - Other

As to this phrase " Do they trump the greenland record?", I don't understand that. It seems to imply that somehow the earth must warm or cool in some fairly spatially uniform fashion. So that warming poles are at odds with cooling equator. That's just not so. Consider that, at present, the area of greatest warming is the North Pole, the area of least warming is the South Pole, and that's exactly in line with the predictions of the (ensemble) of GCMs used by the IPCC. So the notion that if two spots on the earth show different temperature trends, one must trump the other, is fundamentally not correct. To return to the example, current rapid warming at the North Pole and current lack of significant warming at the South Pole both validate the skill of the (ensemble of) GCMs at modeling the climate.

 

agree; however are we talking about CO2 in ice cores or atmosphere?

(rest is re-post from phil jones discussion)

Just b/c the CO2 in ice cores is climbing it does not mean that GHG in atmosphere is climbing as well. There could be other explanations to why CO2 levels in ice cores are increasing, such as:
1) atmospheric methane oxygenation
2) increased CO2 solubility and
3) CO2 freezing.

While methane half life in atmosphere is short, the long time CO2 climbing will be consistent with methane belch and permafrost methane release.

Second, look at CO2 solubility in water:


lowering temperatures will produce higher CO2 content in ice cores while the atmospheric CO2 stays constant.

Also CO2 freezes at temperatures which has been observed. The CO2 freeze point at atmospheric pressure is -78C, and the lowest temp recorded are -82.8C in Antarctica and -66.1C in Greenland. Certainly not a big stretch to assume it has been dropping another 12C in Greenland during Ice Age.

There could be other factors at play, for example air trapped in snow will compress, so CO2 could become solid at higher temps, etc

 

"Since you have put your foot in your mouth on "Hockey Stick" let me help to pull it out. No current multiproxy looks like a hockey stick with the mideaval warming period missing. When we look at tree rings they have correlation not only with temperature, but also with percipitation and carbon dioxide concentration. These are known problems with a tree ring only construction. I do not know the problems with a ice core only construction other than spatial sampling. Do the oxygen isotopes also vary with something other than temperture?"

First, here you seem to equate "hockey stick" with "tree ring only". That's wrong. This exposition is as good as any:

Can you make a hockey stick without tree rings?

Note to reference to the papers that combine ice cores with other proxy indicators.

There is a great advantage to tree rings, in that, as I understand it, you can get hold of the higher-frequency components of temperature change that way. But the numerous "hockey stick" reconstructions are not based solely on tree rings, and as noted in that article, you can get the same shape without reference to tree rings at all (e.g., entirely from spelotherms, entirely from boreholes.)

So this whole notion of "hockey stick VERSUS ice core VERSUS other proxies" just has me baffled. If you want to see how they are optimally combined, follow up on the reference above where the reconstruction is based on a combination of ice cores and other proxies.

Second, in the graph I posted, the small inset, with current temps well above the median estimate of the medieval warm period -- that's "the hockey stick". I don't understand your perseveration about whether the handle of the stick was absolutely straight or not. That's not the point. The point is primarily that the blade is steep, and secondarily that the tip of the blade exceeds the best estimate of average medieval warm period temperatures (weak version) or temps for the last 1400 years or so (stronger version).

Is there uncertainty in all of that? Sure. That's still the best existing estimate. If you want to pillory Mann for getting it almost right in his initial reconstruction, fine. Forgive me if I don't grasp why that's not just normal science, but is somehow such an issue.

As far as I know, isotope deposition in ice is not as straightforward as one might think. In particular, I know that the estimated relationship is empirically derived -- it's not based on some pure physics calculation. It's not a gold standard that trumps other sources of information, even given the geographically limited extent of sites from which you can get ice cores.

Just ping Wikipedia, for starters, and see the extent to which the calibration between isotope mix and estimated temperature varies across sites and has been questioned:

Proxy (climate) - Wikipedia, the free encyclopedia

"... empirically calibrated from measurements of temperature and δ as a = 0.67 ‰/oC for [ame="http://en.wikipedia.org/wiki/Greenland"]Greenland[/ame] and 0.76 ‰/oC for East [ame="http://en.wikipedia.org/wiki/Antarctica"]Antarctica[/ame]. ... More recently, borehole thermometry has shown that for glacial-interglacial variations, a = 0.33 ‰/oC (Cuffey et al., 1995) ..."

This is not something I've spent a lot of time on, and Wikipedia is hardly the source of record, but it's clear that the notion that isotope mix is some sort of universal and simple way to estimate temperature accurately (a gold standard) is not, in the main, correct. It is one of many proxies. When it gets incorporated into multi-proxy reconstructions (hockey sticks), I believe it does so in proportion to its correlation with recent temperature change.

EDIT: OK, here's the question that will get to the root of my confusion here. Austingreen, true or false: Mann's "hockey stick" (1998 or 1999, say) was based, in part, on ice core records (both polar and high-altitude temperate). True or false: Mann's later reconstructions used a wider variety of proxies, including ice cores.

My impression, as an outsider, is that when Mann was constructing his earliest series, there wasn't much available other than ice cores and tree rings. So he used what he had available. But, in some part thanks to Mann's analysis, there's a much richer set of proxies now than there was at that time. Unsurprisingly, you now get a more nuanced view of historical temperature changes. I think the most comprehensive inventory is here, but I doubt that any one place has them all:

http://www.ncdc.noaa.gov/paleo/data.html

Anyway, the notion that hockey stick = tree rings never was literally true, and now is completely obsolete. Maybe that's the root of why I'm having such a hard time grasping this. Pretty much ever since temperature reconstructions have been done, they've included ice cores. I assumed that was common knowledge. So the notion of dropping everything but the ice cores -- given how the proxies are constructed, as I understand it, that could only reduce the accuracy of the reconstructions.

 

Mann 2009 cited just above, cut off at the year 1000 to match MBH 1999.

MBH 1999.

 

Oh the Heinrich events! yes..

there was another similar event of [actual] Biblical proportions in the same time frame 5.6K ago, the Black Sea deluge. Don't think anyone studied the climatic impact of 10mi3 a day draining from Mediterranean to Black Sea for a year or so. It had to have an impact on North Atlantic/Mediterranean inflow.

 

First CAGW theory used ice core data to prove its case.
But at finer resolution Ice core data destroys AGW theory ,because temp precedes CO2 level rise.
AGW cheerleaders couldnt find fault with the data.Otherwise every AGW website would say its not reliable.
Instead Richard C Alley invents a Sci Fi Fairytail called "forcing".
Im prompted to post this as Im reading a webpage of propaganda about Tim Ball.http://www.sourcewatch.org/index.php?title=Tim_BallOne of his faults is he claims that
"Ice-core records show that temperature rises before CO2 rises, not because of it."

 

you might have missed these ones (we know that such champion of truth as you will not avoid them for mere "inconvenience"):

 

this is how the statistically based modeling is done: you collect you data points and build a polynomial function based on data points. Works great, as long as it is being used within range, but not so much outside of it. If we knew what the climate will be in 2100 we would have a very accurate presentation for the 2010-2100 range, but since we don't we get an exponential growth aka hockey stick.

We had similar discussion with FHA planners trying to build an interstate through our neighborhood. They had taken a 5% population increase (which was true for last 5-8 years) and tried extrapolate it 40 years in future. Noone even bothered to look that there is not enough land to support Manhattan density population in our area.

I have serious suspicion of any model which does not rely on physical process modeling only on statistic; the only time statistical models work when you know the range, either from-to or low-high and it is more or less closed and stable system.

 

"...we made ourselves an extremely poor experiment.We started to observe meteorology at the coldest spot in the last 10,000 years."
You have to click the title to watch the video.You will learn how they deduce temperature from ice core.Hopefully more.
[ame="http://vimeo.com/14366077"]We live in cold times on Vimeo[/ame]