"There appears to be something fundamentally wrong with the way temperature and carbon are linked in climate models."

Oliver and Robert,
Starting with the first point, it’s irrelevant whether the entire warming trend was directly attributable to CO2 or to something else–all that matters is how much warming was triggered by an INITIAL rise in CO2. From the blog post: “Average temperatures worldwide rose by about 7 degrees Celsius — about 13 degrees Fahrenheit — in the relatively short geological span of about 10,000 years.” So a portion of that is still very large. And if you’d actually read the study’s summary, the conclusion is that 3.5 degrees C (half of the warming, and a BIG number) was due to CO2. The rest happened as CO2-caused warming led to additional feedback loops. And remember, those initial 3.5 degrees C were caused by LESS CO2 than we’re currently emitting, and over a time period 100 times longer. So yes, this study proves we’re in dire straits.
As for the second point, it actually does matter which direction the model is wrong in, and it matters what exactly caused the model to be wrong. In this case, most of the models were leaving out an important feedback loop that would have predicted even stronger warming. The logical response isn’t to discount the concept of modeling – it’s to isolate the flaw in the model and correct it. To use the archery analogy, if the wind keeps blowing your shot off target, it doesn’t mean you’re a bad shot – it means you need to change the parameters of your shot (by adjusting for wind); it matters whether you were over or undershooting, because that tells you how to adjust.

Leif Svalgaard (20:51:06) :
Andrew (20:29:55) :
It amazes me how deniers’ religious zealotry is so strong that they can find proof of their case in something that says the exact opposite
That was my sentiment too: 14.07.2009 Leif Svalgaard (19:14:39)
I reached to the same conclusion about the paper after Leif Svalgaard made his observation: (19:25:43) and (21:15:31); however, I don’t see the reaction from bloggers like “religious zealotry” but like a misguiding conclusion from the authors. It’s an AGW trap.
On the other hand, the carbon dioxide does not possess the thermophysical properties enough as to cause any climate change. The climate is always changing. By the time referred by the authors of the paper, the carbon dioxide was diminishing, not increasing. The change of temperature was not 7 °C, but 14 °C and it was included into a long period of warmhouse which had started off during the early Cretaceous, some 150 million years ago. The authors cherry picked a small period and made it to appear like something unusual for the period; however, high temperatures were quite normal during the Cretaceous, the Paleocene, the Eocene and the first third of the Miocene.
Conversely, if you wish we can sustain a scientific dialogue about the thermophysical properties of carbon dioxide obtained through experimentation so you can verify that the carbon dioxide is not a primary source of heat and that the carbon dioxide is incapable of causing any warming. My weapons are all scientific stuff.

Andrew (20:29:55) :
Sure, models could overestimate CO2’s impact on climate, but in this case the models have been UNDERestimating the impact, because they didn’t give enough weight to positive feedback mechanisms. the models are wrong in this case, it lends MORE evidence to global warming’s reality, and MORE cause for concern.
Let’s think the above through logically. You are correct that one possible interpretation is the models underestimate historical positive feedback. So, the logical solution is to increase positive feedback in the models to meet the historical data. Logically, we must also increase the feedback to increasing CO2 occurring now and in the future . Otherwise your corrupt the association, unless you can establish a physical mechanism allowing differential temp response to the feedback effect. Oh, except…… global temps are lower than IPCC predictions based on the positive feedback already incorporated into the current models. Therefore, logically, CO2 does not cause warming, there are no positive feedbacks and the statement “Some feedback loop or other processes that aren’t accounted for in these models” refers to something other than CO2.

Light oxygen – one of the bellweather indicators of ice ages – tracks far more consistently with temperatures than carbon. Light oxygen is not only an indicator of how much water was locked up in vast glacial stades, it also shows how humid the planet is – more humid and hotter during interglacials, drier and cooler during glaciation.
A year ago I plotted prehistoric temperatures against both CO2 levels & light oxygen levels. The temperature & light oxygen plots tracked far more consistently than temperature & CO2 levels.
Light oxygen represents the amount of available water in the Earth’s atmosphere. During ice ages the great stades lock up a great deal water, resulting in a more-arid global climate, a trend reflected by the light oxygen levels from ice core data.

The obvious choice for the missing three or four trillion tons of C13 depleted carbon is the methane hydrates. There are several trillion tons of methane stored in the methane hydrates (also known as methane clathrates), deposited in a zone of stability beneath the ocean floors and in the Arctic.
So, what they’re saying, I think, using scientifically cautious terminology, is that the Paleocene/Eocene Thermal Maximum was likely a mild to moderate methane catastrophe. Essentially what they seem to be saying is that the “clathrate gun” or “methane catastrophe” theory seems to be correct, and that CO2 alone is insufficient to explain the temperature rise.
We are not safe from a methane catastrophe, ourselves, by the way. On February 26,2009, Stanford climate scientist Chris Field talked about the sort of runaway vicious cycle feedback effects, such as melting Arctic permafrost (containing 1.5 trillion tons of carbon), that could very easily trigger dissociation of the oceanic methane hydrate deposits.
Many of the comments on this thread seem to miss that point. This paper implies we could have a methane catastrophe with less than a doubling of CO2 levels.

From Chris Field, DemocracyNow, Feb 26, 2009:

And what we increasingly see is that with temperatures at the upper end of this warming range, we begin to get a large series of very dangerous feedbacks from the earth’s system. In particular, we see tropical forest transitioning from taking up large amounts of carbon to taking up very little or even releasing carbon. And it looks like there’s an increasing risk that high latitude ecosystems that are characterized by these frozen soils called permafrost may release some of the organic matter that’s stored in this permafrost to the atmosphere. So you end up in a situation where, instead of having ecosystems storing large amounts of carbon, their storing very little or releasing large amounts.
The calculations to date are that tropical forests—and this is something that is explored in the IPCC—could, at the higher ranges of temperature forcing, release anywhere from a hundred billion to 500 billion extra tons of carbon to the atmosphere by 2100. And that should be put in the context of understanding that during the entire period from the beginning of the Industrial Revolution until now, all of the world societies have only released a little over 300 billion tons of carbon to the atmosphere….
…AMY GOODMAN: I wanted to ask not only about what’s happening in the Southwest, but a vicious cycle you talked about that could do everything from ignite tropical forests to melt the Arctic tundra.
CHRISTOPHER FIELD: The idea of these vicious cycle feedbacks is that once warming reaches a certain point, the amount of assistance that we’re getting in terms of carbon storage from the land and oceans tends to go down. And this is quite clear from the IPCC models, and it’s clear from a number of other more recent lines of work. In the IPCC, the models characterize a future in which tropical forests at the high range of warming have a potential to release large amounts of carbon to the atmosphere.
One of the new numbers that’s a great concern to me is that we’ve been doing studies of how much organic matter is stored in these frozen soils in northern latitudes, permafrost soils, and the new numbers are that approximately a billion tons [he meant trillion tons, and the latest estimates are actually 1.6 trillion tons of carbon – Leland Palmer] of carbon is stored in the organic matter in these high latitude soils. Climate model projections indicate that at high amounts of warming large fractions of the permafrost could melt, and some of the projections have that at from 60 to 90 percent of the permafrost melting.
And the surprising thing about these permafrost soils is that the organic matter that’s contained within them is not this incredibly stabilized, difficult-to-decompose stuff; it’s basically frozen plants that have been sitting there for, in some cases, tens of thousands of years. And when the permafrost is thawed, these plants decompose quite quickly, releasing their carbon as CO2 to the atmosphere or as methane to the atmosphere, which is a greenhouse gas that, on a molecule per molecule basis, is about twenty-five times as powerful as CO2.
The basic risk is that if we reach a certain point in the warming, what we’ll end up with is a vicious cycle, where the warming causes additional permafrost melt, which causes additional CO2 to be released to the atmosphere, which causes additional warming, which creates this vicious cycle.

What they’re saying in this paper is that Chris Field was basically right, and what he’s afraid of, these runaway positive feedbacks, appear to have occurred during the Paleocene/Eocene thermal maximum.
An even bigger methane catastrophe was likely the Permian/Triassic extinction event, which killed approximately 95 percent of all species on earth.