Nothing seems to attract the inverted love of fake climate sceptics like the word ‘unprecedented’. Perhaps they don’t like to be reminded that our shared climate trajectory is into unexplored and potentially dangerous space. Perhaps because they can often make the trivial claim that the unprecedented event is not really unprecedented because it was warmer/wetter in the Carboniferous
Miller et al (2013) used this sceptic bait-word in the title of their paper published online yesterday: “Unprecedented recent summer warmth in Arctic Canada”. Miller et al radiocarbon date moss and lichen emerging from beneath glaciers on Baffin Island and find that some are over 40,000 years old. An alarming result implying that the glaciers are now smaller than at any time in the last 40,000 years, and by extension, probably since the Eemian, the last interglacial, ~120,000 years ago. Watts and commentators at WUWT are already attacking the paper with the accuracy and precision of a blind sniper using reindeer droppings for ammunition.
The press release for the paper is characteristically useless, not linking to the actual paper. Unfortunately, despite the interest the media interest the paper was likely to generate, it has not been made open access. This might explain why Watts and commentators at WUWT have obviously not read the paper.
Watts argues that the moss now being exposed by the retreating glaciers may have been previously uncovered and reburied. The paper explicitly discounts this possibility.
Our field observations, and the presence of extensive vegetation-free regions surrounding most retreating ice caps [Locke and Locke, 1976] indicate that most long-dead tundra plants exposed by ice recession are rapidly removed from the landscape by wind-blown winter snow or by run-off during the melt season.
Long-dead mosses and lichen are not the most robust of objects, whereas snow shards blown by a gale are very abrasive, so it is reasonable to assume that the plant remains will not survive more than a few years of exposure.
Watts then dives into a evidence-less discussion on whether the glaciers are melting or sublimating, which would have different implications for the climatic mechanisms responsible for their shrinking. A short search would have led him to Zdanowicz et al (2012) who studied summer melt rates on Penny Ice Cap, Baffin Island. In short, there is lots of melting, more than there used to be, and clear evidence of warming. Watt’s had argued there was no recent warming, wrong again.
Watts ends by whining about the word ‘unprecedented’. The authors claim that the current warming is unprecedented when obviously it was at least as warm 120,000 years ago when the moss grew. Yes, the authors ought to qualify their statement, for example ‘unprecedented in the Holocene’ or ‘unprecedented in the last 120,000 years’, but it is pathetic if this is all Watts has to whinge about.
Several of the commentators at WUWT are complaining that whereas the paper suggests the plant remains are possibly ~120,000 years old, radiocarbon cannot be used to date anything over 50,000 years old . This is of course completely true. The half-life of radiocarbon is 5730 years. After one half-life, one half of the original radiocarbon remains, after two, only half of that amount – one quarter. After eight half lives the amount remaining is 1/2^8 or 0.4% of the original, which becomes difficult to measure as even trace contaminants have a major impact. But this is also entirely irrelevant. If a sample has essentially no radiocarbon, then it must be older than ~50,000 years. Both a lump of Carboniferous coal and an Eemian moss will have a radiocarbon date of ~50,000 years. There is little chance that a the ice melted 50,000 years ago, this was in the middle of a long cool period. The last period plausibly warmer than modern is the Eemian, ~120,000 years ago, so this is the most likely date for the moss.
McIntyre showed up to imply that the radiocarbon dates could be wrong, linking to a paper that shows that aquatic mosses can have an old carbon effect, where they use radiogenically dead carbon dissolved from limestones and so appear to be older than they really are. Old carbon effects are a major nuisance for dating lake sediments, but are irrelevant here as the mosses (mainly Polytrichum) and lichens dated are not aquatic. The authors also verify that living mosses have a modern radiocarbon age. A large old carbon effect (anything less would not materially change the results) is also not compatible with the large number of lichens <1000 years old.
Others are arguing that the paper cannot be correct because of the Early Holocene Thermal Optimum. However, the Holocene Thermal Optimum, driven by increased summer insolation due to changes in the Earth’s orbit, was probably rather poorly expressed in this region because of the influence of the decaying Laurentide Ice Sheet. Kaufman et al (2004) has a compilation of palaeo data from the Western Arctic showing that the peak terrestrial temperatures in this region were only about 1°C above ‘modern’ (remarkably similar to the cooling found in the CMIP5 model runs reported by Miller et al.). It is perfectly plausible that the climate has already warmed more than this. Curiously, Miller et al estimate that there has been a much larger cooling, 2.7°C since 5000 years ago, from equilibrium line altitudes. Perhaps the albedo from the ice is acting a local positive feedback, amplifying the slight regional cooling.
Still others are indulging in the evidence-free argument that the ice has been melting since the last glacial maximum and has only just receded back as far as the moss collected by Miller et al. This argument is somewhat at odds with the observation that the glaciers and ice sheets expanded between 5000 years ago and the Little Ice Age, and that the ice caps studied are so thin (and can only have ever been thin) that they will melt rapidly under a warm climate.
All these evidence-free and trivially wrong arguments are mixed with a liberal dose of invective and conspiracy theory. Sadly, this is not unprecedented.
Miller et al (2013) Unprecedented recent summer warmth in Arctic Canada. Geophysical Research Letters
” This might explain why Watts and commentators at WUWT have obviously not read the paper.”
Don’t know, but I have read it and it has holes big enough to drive a train through:
I don’t find your critique persuasive.
Take for example your point about the thickness of ice cap prior to the plants growing there. Who cares? It is not in the least relevant to the paper. What is important is that the ice cap can only be ~70m thick, so can melt in a century or so whenever the climate is warm. So if there had been a warm century in the early Holocene or whenever the ice would have melted, the moss & lichens would have been destroyed.
40kBP is in the middle of marine isotope stage 3. It would be surprising if any time in stage 3 was warmer than the Holocene, so an stage 5 date is more likely. In any case, what is important is that the melt exceeds peak Holocene Thermal Maximum melt.
The other 135 samples. They are just not relevant to the problem. Only the oldest samples are relevant to the question of when it was last warmer as different ice caps will have different melt speeds.
“Take for example your point about the thickness of ice cap prior to the plants growing there. Who cares? ”
Then take it up with the authors, because they’re the ones who raised it that issue–I was just reporting on that.
“The other 135 samples. They are just not relevant to the problem. Only the oldest samples are relevant to the question of when it was last warmer as different ice caps will have different melt speeds.”
You cannot be serious. For the mosses to have been found at the 135 sites and dated to < 5kya means necessarily that it was about as warm there as it is now. That is the whole logical basis of their paper.
I think you are misreading the paper. Where do the authors write about the thickness of the ice “before the dated plants grew there”? There is a relevant discussion of how long the ice would take to melt, which you misinterpret. It would take 140 years to melt a 70m thick ice cap at 0.5m/yr, but it has not all melted – there is a remnant ice cap. So the statement that “a 70-m-thick ice cap would have melted to less-than-present dimensions in 100 years” is plausible.
If the ice caps were in equilibrium with climate then if some were surrounded by >44kBP plants and other were surrounded by Little Ice Age plants, I agree, there would be a problem. However, the ice caps are not in equilibrium with climate. This is trivial to deduce – they are melting at the relevant altitude at 0.5 m/yr. There is no reason to suppose that the different ice caps will approach their equilibrium at the same rate – it will depend on the morphology of the ice cap. Consequently, it is not in the least surprising that the plants being exposed are of different ages.
I’m trying to understand your comment about the ages of the plants, specifically that only the oldest ages are the ones to find as relevant in a conclusions. I don’t have strong experience dealing with the biology aspect of climate science, and I saw other critiques asking why they made most of their conclusions about a small subset of the total samples. At first, I found conclusions of cherry picking to at least seem reasonable.
I’ve since read your critique and you mention that the oldest ages are what care most about, and that we would actually expect several different ages to be exposed.
Do you mind elaborating on why this situation in particular would be one where we would look at just the oldest dated samples instead of doing something more of an average of multiple samples?
I just realized that you elaborated on your reasoning in your subsequent blog post. That seems to answer my question.
OK I’ll try to read it again. I cannot see how the issue of whether or not ice caps are in equilibrium with the climate has anything to do with it. At the very least, some very important conceptual issues are not carefully explained by them.
You say the press release is useless. It seems Anthony Watts didn’t find it useless, but made rather good use of it. The mis-location of Baffin Island as being east of Greenland certainly wouldn’t inspire me to pay to get access to the paper. Might tend to generate some mockery tho’.
The quote below from the PR is sufficient reason for the WUWT discussion without even having the paper. It seems “unprecedented” is the key finding along with certainty warming is all due to GHGs. Hope they tell the IPCC this breaking news so it can be snuck into the final publication of the WG1 in January.
Strangely, Wiley can’t find the Zdanowicz et al (2012) you linked. Either it must have been withdrawn or is too expensive for even Wiley to keep on hand.
Just a suggestion, but you might want to avoid WUWT entirely if such a mild post by AW causes you to have such a conniption. It can’t be good for your health.
If getting everything wrong and revealing himself to be clueless is what you mean by making good use of the press release, then Watts certainly did.
A journalist in a university press office getting confused about Arctic geography. Is that the best you can do?
I’ve corrected the link to Zdanowicz et al (2012).
Bob, mis-stating the relative positions of Greenland and Baffin is just a simple mistake that anybody can make. It doesn’t mean or imply anything. If we all get hung up on small things like that, we are in big trouble, because a useful discussion will be impossible.
Is it really unprecedented. The Miller study tells us about conditions on Baffin during the focus period. That’s about it. What does it tell us about Arctic wide warmth 7 thousand years ago?
As you surely know, in the early Holocene, the Earth’s orbit was different, leading to increased summer insolation and summer warmth. This is well understood and completely irrelevant for the recent warming. This early Holocene warmth is not part of the natural variability that can be expected under the current orbital configuration. Miller et al (2013) shows that the 21st century Arctic climate is already beginning to exceed peak Holocene warmth even thought the orbital configuration is unfavourable. The 21st Century is only just beginning, there are going to be a lot more greenhouse emissions and warming still to come.
The point is that there was less ice. Much less ice. It was therefore much warmer.
Much less ice? Really? What about this little bit that covered a large part of Canada 8400 years ago?
I refer to the links provided by jimjam and your response, regarding sea ice extent as evidence of temperature.
He’s also confusing sea ice with land ice. He’s assuming that having periods with a different orbital configuration that allows for periods of reduced sea ice are somehow analogous to periods of much reduced land ice. As we already know from established history of ice ages, they are not the same.
Pingback: More on Mosses (Miller et al 2013) | Musings on Quantitative Palaeoecology
Good post, Richard. That argument from Watts that moss remains intact while exposed for quite some time, is really, really weak, and coming more from the gut than from the brain. Also, his graph showing SAT on Baffin Island looks a lot different when updated, for instance 2012 had the warmest summer on record (and Clyde isn’t in the Northwest Territories, but in Nunavut).
With regards to the differences with the Holocene Climatic Optimum: you already mentioned that the warmth back than was caused by the Earth’s orbit, and this time it’s most likely caused by an increase in the GHG concentration. This would probably also mean that winters are now warmer because of the GHG blanket. Does the fact that winters were probably colder during the HCO also mean anything with regards to the mosses staying covered?
Thank you for your comment.
Winter temperature is difficult to reconstruct as most biotic proxies are more sensitive to summer temperature. Arctic winter insolation was somewhat lower during the Holocene thermal maximum than pre-industrial (see for example http://pmip.lsce.ipsl.fr/publications/local/wcrp111_009.html) and there is evidence that seasonality was larger (I’ve worked examples from the Norwegian sea), but I’m not aware of evidence that winter temperatures were colder then.
Winter temperature has warmed in recent decades (see for example Zdanowicz et al 2012), and this will precondition the ice for melting the following summer. This will mean that the equilibrium ice cap size will be smaller for the 21st Century than the mid Holocene for equivalent summer temperatures. I am not sure how large this effect is, but I would suspect it is rather small as the ice would tend to be well insulated under a blanket of snow during winter. However, you may have identified more larger uncertainty that all the climate sceptics together.
The Funder paper from 2007 is a good example of what often happens. Dr Funder has since changed his views. Yet while often citing the paper skeptics never point this fact out. “I’m citing a paper the author no longer agrees with.” is a pretty telling argument.
Another of your links, History of Sea Ice in the Arctic, does not say what you imply it to say. It’s my favorite resource on arctic sea ice (Dr Funder was also an author). A complete reading of ‘History of Sea Ice In the Arctic’ does not lead one to the conclusion that the Arctic Ocean was ice-free during the HCO.
There is also evidence that there was *more* ice in some parts of the arctic contemporaneous with *less* ice in others during the HCO. Remember this? Pacific Plankton Crosses to Atlantic … Thanks to Arctic Meltdown. It’s important to look at more than one proxy if we want to make statements for the entire Arctic region.
Pingback: Man with one graph declares war on mosses (more on Miller et al 2013) | Musings on Quantitative Palaeoecology