While Joe Romm and Mark Serreze bloviate about the current Arctic sea ice being “lowest in history”, science that doesn’t have an agenda (or paying thinktank) attached says otherwise:
“More importantly, there have been times when sea-ice cover was less extensive than at the end of the 20th century.”
From the Hockey Schtck: Paper: Current Arctic Sea Ice is More Extensive than Most of the past 9000 Years
A peer-reviewed paper published in the Canadian Journal of Earth Sciences finds that Arctic sea ice extent at the end of the 20th century was more extensive than most of the past 9000 years. The paper also finds that Arctic sea ice extent was on a declining trend over the past 9000 years, but recovered beginning sometime over the past 1000 years and has been relatively stable and extensive since.
Although it seems like a day doesn’t go by without an alarmist headline or blog posting obsessing over the daily Arctic sea ice statistics (and never about Antarctic sea ice extent which reached a record high this year), this paleo-climate perspective takes all the wind out of alarmist sails. Satellite assessment of sea ice conditions is only available beginning in 1979 (around the time the global cooling scare ended), with only sparse data available prior to 1979. The alarmists at the NRDC fraudulently claim in a new video that due to “climate destruction,” Arctic sea ice reached the lowest in history in 2010 (actually the low since 1979 was in 2007 and 2010 was the 3rd or 4th lowest depending on the source). Probably wouldn’t bring in many donations if they mentioned the truth: the 21st century has some of the highest annual Arctic sea ice extents over the past 9000 years.
The figure below comes from the paper, but has been modified with the red notations and rotated clockwise. The number of months the sea ice extent is greater than 50% is shown on the y axis. Time is on the x axis starting over 9000 years ago up to the present. Warming periods are shown in gray with the Roman and Medieval warming periods (RWP/MWP) notated, a spike for the Minoan Warming Period about 5000 years ago, and two other older & unnamed warming periods. The last dot on the graph is the end of the 20th century and represents one of the highest annual sea ice extents.
Holocene fluctuations in Arctic sea-ice cover: dinocyst-based reconstructions for the eastern Chukchi Sea Canadian Journal of Earth Sciences, 45: 1377-1397
Authors: J.L. McKay, A. de Vernal, C. Hillaire-Marcel, C. Not, L. Polyak, and D. Darby
Abstract: Cores from site HLY0501-05 on the Alaskan margin in the eastern Chukchi Sea were analyzed for their geochemical (organic carbon, d13Corg, Corg/N, and CaCO3) and palynological (dinocyst, pollen, and spores) content to document oceanographic changes during the Holocene. The chronology of the cores was established from 210Pb dating of near- surface sediments and 14C dating of bivalve shells. The sediments span the last 9000 years, possibly more, but with a gap between the base of the trigger core and top of the piston core. Sedimentation rates are very high (*156 cm/ka), allowing analyses with a decadal to centennial resolution. The data suggest a shift from a dominantly terrigenous to marine input from the early to late Holocene. Dinocyst assemblages are characterized by relatively high concentrations (600–7200 cysts/cm3) and high species diversity, allowing the use of the modern analogue technique for the reconstruction of sea-ice cover, summer temperature, and salinity. Results indicate a decrease in sea-ice cover and a corresponding, albeit much smaller, increase in summer sea-surface temperature over the past 9000 years. Superimposed on these long-term trends are millennial-scale fluctuations characterized by periods of low sea-ice and high sea-surface temperature and salinity that appear quasi-cyclic with a frequency of about one every 2500–3000 years. The results of this study clearly show that sea-ice cover in the western Arctic Ocean has varied throughout the Holocene. More importantly, there have been times when sea-ice cover was less extensive than at the end of the 20th century.
 |
|
| Arctic summer sea surface temperatures are also currently lower than much of the past 9000 years |
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
JK: I read the graph at 7000 years BP to have ~6.5 months with sea ice <50%, and with the other study suggesting persistent ice free conditions in the eastern Arctic for prolonged periods just proposed the possibility of an ice free Arctic for at least part of the year. Just a hypothesis and that is why I used the word "may." Regardless, the point is the western Arctic has had "far" less sea ice than many periods in the past without causing a tipping point.
Barry: well, I guess you better write to the 6 authors asking for a retraction of their statement as the concluding line of the abstract, “More importantly, there have been times when sea-ice cover was less extensive than at the end of the 20th century.”
HockeySchtick
Why? I have no issue with their paper (how could I? I’m no expert on the subject). But I don’t need to be an expert to see that your commentary is selective and misleading. You have provided yet another example in your last post. All one has to do is read the sentence before the one you quote,
As my point is that the study is assessing one region of the Arctic Ocean – and I have no problem with their results for that regio0n – then your treatment that this represents the entire Arctic is shown to be fallacious just by reading the paper.
Then there’s the conclusion.
The Holocene record from site HLY0501-05 illustrates the sensitivity of hydrographical conditions in the western Arctic Ocean. The data show a long-term warming that is opposite to what is reconstructed for the eastern Arctic and point to a bipolar behavior of the Arctic Ocean at the timescale of the Holocene. The millennial-scale variability in the eastern Chukchi Sea is characterized by quasi-cyclic periods of high SSS, high SST, and reduced sea-ice cover, which most probably reflects variations in the stratification of the upper water column. Such changes may be related to tidal forcing and (or) large-scale mechanisms, such as AO/NAOlike
oscillations. It is important to note that the amplitude of these millennial-scale changes in sea-surface conditions far exceed those observed at the end of the 20th century.
The bulk of this study, and the conclusions it rests on data from one site in the Cukchi Sea. The paper discusses opposite effects occurring elsewhere in the Arctic, and the same goes for their more recent paper. There is no justification whatsoever for a bald assertion that Arctic-wide sea ice cover was definitely less than current – not based on these papers.
Sea ice cover was likely less in the early Holocene than current. I’m sure you could find a paper making qualified conclusions on that. Even NSIDC points that out. It’s not a particularly controversial item. Why do you hold the idea that less ice cover on the geologic past would ‘take the winds out of the sails’ of ‘alarmists’ (whoever they may be)? In the long-distant past there was no summer sea ice at all, and, there was probably a ‘snowball Earth.’ a few times. What point are you trying to make?
[clicking links]
Ah – I see the issue seems to be with an eco-video saying Arctic sea ice is at an “all time low”. OK, that’s a fine example of false rhetoric from the other side of the debate from someone with a different agenda. Truth is obscured in propaganda wars, and this discussion is indicative.
Paul Birch says:
September 25, 2010 at 3:43 am
Validation of proxies
Proxies can be validated in different ways. What I write below is most relevant to reconstructions based on microfossil assemblages, but perhaps also some others. The past environmental conditions are inferred from the fossil assemblages using the relationship between the modern assemblages and the modern environment in a collection of modern samples known as the training set.
The most important, and often only, validation, is cross-validation of the training set. In the simplest case, one site is omitted from the training set and the remaining sites used to estimate it’s environmental conditions. This is then repeated for all sites, then the predicted and measured environmental variables can be compared statistically. If there is a low error, and high r2 then this is good evidence that the proxy is reliable. The training set used by McKay and others has been cross-validated, and appears to have good performance. However, I would contend that the performance statistics they report are seriously overoptimistic because spatial autocorrelation in their training set violates the assumption of independent observations. This group doesn’t agree with me, but have failed to provide a rational of why their data should be exempt from a problem known to all other statistical fields.
A second way that proxies can be validated is by comparing the reconstruction with the historical record. This is a powerful test, but the opportunities for doing this are very limited, especially in the ocean where time series of observational data are usually short, and sedimentation rates normally fairly low. Were we to limit use of proxies to sites where they can be validated against the historical record the method would be of little utility. Even if the proxies validated in the period with instrumental data, this would not be an absolute guarantee that it remains reliable throughout the record. What is possible, is to compare the core top reconstruction with the modern value. If it is close it is encouraging, but spatial autocorrelation can bias this test.
A third way to validate a proxy is to compare it’s reconstructions with reconstructions from other proxies in a multiproxy study. If all the proxies concur (and are independent) it is good evidence that they are reliable.
Of these three tests, robust cross-validation is the most essential. If the proxy fails this, it is useless, if it passes, we can be fairly confident that it is useful. The other validation techniques can give extra assurance, but are not required. The key word here is robust. If the cross-validation is not robust, then all sorts of junk can appear to be possible to reconstruct.
But…but…Al Gore in 2008 predicted that “the entire north polar ice cap will be completely gone in five years!” And geez, he’s famous, with a Noble Prize and everything, isn’t he. And he done a movie to prove it, too.
Barry says : “Look again at 9.1, and you will see enhanced heating (brighter yellow) in the tropical troposphere area of the solar panel. The reason TT is only slightly warmer is because the forcing is not nearly as great as for GHGs. … the temp change at the TT is almost entirely a result of the change in adibiatic lapse rate in response to temperature change of the total atmosphere. This does not hinge on the type of forcing (as far as we know).“.
For the benefit of those who don’t have the IPCC Report readily to hand, here is the Fig 9.1 we have been discussing.
http://members.westnet.com.au/jonas1/IPCCFig9p1_LowRes.jpg
Caption : Figure 9.1. Zonal mean atmospheric temperature change from 1890 to 1999 (°C per century) as simulated by the PCM model from (a) solar forcing, (b) volcanoes, (c) wellmixed greenhouse gases, (d) tropospheric and stratospheric ozone changes, (e) direct sulphate aerosol forcing and (f) the sum of all forcings. Plot is from 1,000 hPa to 10 hPa (shown on left scale) and from 0 km to 30 km (shown on right). See Appendix 9.C for additional information. Based on Santer et al. (2003a).
Barry, I find it quite difficult to work out what you are getting at. To my simple mind the essentials are quite clear : The IPCC (or their climate models) have calculated where the warming occurs, and Fig 9.1 in the IPCC Report AR4 presents it visually. As you say, the solar forcing is not nearly as great as for GHGs, and the contribution from aerosols is slightly negative (according to the IPCC), in fact all except GHGs (panel (c)) are negligible. If the IPCC are correct about AGW, then the total forcings from all sources are as per panel (f), and therefore examination of the TT (tropical troposphere) should reveal this very specific warming pattern of ~1 deg C per century. Well, the TT has been examined, and that warming pattern is not there. Assuming that it was looked for correctly (a non-trivial assumption I admit) the only possible conclusion is that the IPCC got it wrong. All that matters is panel (f) and the TT measurements. AGW is falsified. Period.
Discussion of the adiabatic lapse rate is irrelevant, because that is all part of the IPCC case and the IPCC case has just been falsified. I just don’t know how to put it any more clearly.
Your addendum “SPM2 figure expresses values of various forcings (GHG, solar, aerosol, ozone etc) over a time period twice as long as 9.1” :
I have seen these figures, they show that CO2 forcing is 1.66 Wm-2 and all the other forcings roughly balance out for a total net forcing of 1.6 Wm-2. SPM2 doesn’t identify the TT component, so I have to assume that Fig 9.1 is the spatial representation of SPM 2 (it certainly tallies with SPM2 wrt the importance of CO2). We have just seen that Fig 9.1 is wrong – it does not actually happen – so SPM 2 must be wrong too.
Well, if the IPCC have got it wrong, what is the real answer? As Smokey and others have pointed out, it is not up to us to say what does happen and why. It is sufficient to falsify AGW in order to know that whatever it is, it isn’t what the IPCC says it is.
But I would say that there are clues. In the satellite age, the surface has warmed more than the TT (except temporarily during El Ninos). It would seem that whatever has been warming the globe has warmed it at the surface, not in the TT. That rules out CO2 as the main driver. But we know that the sun primarily heats the surface, not the troposphere. We know that solar radiation taken on its own does not vary enough, but even though the IPCC got their positive “feedbacks” wrong, maybe there are positive “feedbacks” for solar radiation instead. We know that quite small changes in cloud cover have a big effect on how much sunlight reaches the surface (the IPCC spell that out quite clearly). Maybe that would be a good place to look next.
barry said: “There is no justification whatsoever for a bald assertion that Arctic-wide sea ice cover was definitely less than current”
There is no justification for misquoting and distorting what I said either. At no point did I say “the entire Arctic” or “Arctic-wide” sea ice cover was “definitely less than current” and it was immediately obvious to anyone who read the abstract provided exactly where the drilling site was located. The authors themselves extrapolate their findings to the “western Arctic” as stated in the abstract I provided.
But thanks for providing the link to the NSIDC which states, “A recent study suggests that 5,500 years ago, the Arctic had substantially less summertime sea ice than today.” Now would their “post” about this unnamed study happen to be referring to the entire Arctic, western Arctic, eastern Arctic, or? Doesn’t say. Do you know what paper they are referring to?
What point am I trying to make?
That Arctic sea ice cover has been “far” less than current in the western, or eastern, or both (see fig 6 in my comment above; low at both east & west in the 1500’s-1600’s) portions without causing a planetary-albedo-positive-feedback-tipping-point that is claimed to be occurring now with better sea ice conditions, such as by taxpayer funded ‘scientists’ like Mark “death spiral” Serreze.
HockeySchtick,
I didn’t see that argument in the post at your site. Your thesis there is about the claim Arctic sea ice is at an all time low being incorrect.
But seeing as we’ve moved on, are you able to cite Mark Serreze saying that low sea ice conditions are currently causing a planetary-albedo-positive-feedback-tipping point? I think that his comments have been limited to the reduction of Arctic sea ice from local feedbacks (like more open water/albedo), and not about observed enhanced global temperature from ice loss at the poles. AFAIK, this is a theoretical component that is difficult to observe currently, and Serreze’s comments have been speculative. EG,
Published January 2010 – http://onlinelibrary.wiley.com/doi/10.1111/j.1523-1739.2009.01408.x/full
Once again, I don’t think your commentary matches what is being said. Serreze notes that temperatures have gone up (year round, not summer, much) stronger in the Arctic, and that this is nominally consistent with Arctic feedback, but that the seasonal timing and amplitude make difficult positive determination that this is due to an ice albedo effect, rather than other effects at this time.
Cliff,
In my September 25, 2010, 4:17 am post I corrected your misconception that “Evidence is anything supporting a theory.” As I explained, that is incorrect, and I provided Karl Popper’s explanation to help understand why.
I just now got back in town, and I’ve read your newest misconception:
“And the models too are empirical evidence.”
That is simply wrong. You really need to get up to speed regarding the scientific method. It is the reason we don’t go to witch doctors any more to cure diseases, so it is worth making the effort to understand it.
Computer climate models are not empirical evidence. Models are simply tools — and they are very inaccurate tools. They can not make accurate predictions, and they certainly have been unable to falsify the hypothesis of natural climate variability, which fully explains the current climate without the need for extraneous entities such as CO2.
There is a critical difference between empirical, testable evidence, and a tool. Evidence under the scientific method is verifiable, raw [unadjusted], testable physical data and observations — evidence does not consist of the tools that are used, whatever they may be. Alarmist scientists love their GCMs, because their models will say whatever they are programmed to say, just as a crescent wrench can be adjusted to fit different sized nuts.
The alarmist crowd throws a tantrum when their feet are held to the fire of the scientific method. Why? Because the scientific method does not support the CO2=CAGW conjecture. If it did, the debate would already be over. But as it happens, the more evidence that is found concerning the effect of CO2, the more insignificant it turns out to be. There is zero empirical evidence showing that this harmless and beneficial trace gas is leading to runaway global warming.
Alarmists constantly attempt to denigrate scientific skeptics by putting quotation marks around the word skeptic, and they never acknowledge the fact that all honest scientists are skeptics, first, last, and always.
Dishonest scientists, OTOH, deliberately avoid the scientific method. They refuse to disclose their data, methodologies and metadata, so their results cannot be either verified or falsified. And they always begin with the presumption that human-emitted CO2 is the primary cause of any observed warming. Rather than see where the data leads them, they beat it into submission through endless adjustments, until it says what they want it to say. That is their way to grant heaven. But it is not any closer to science than Scientology.
Understanding what empirical evidence is, and more importantly, what it is not, is essential knowledge required for understanding the scientific method.
When there is a conflict between the CO2=CAGW model-based hypothesis, and empirical evidence from unbiased physical observations, the alarmists always go with the [repeatedly falsified] model hypothesis, while skeptical scientists accept the raw data — the empirical evidence.
This skeptical scientist explains why alarmist models vs observations are wrong.
phlogiston says:
September 25, 2010 at 2:15 am
EFS_Junior says:
September 24, 2010 at 5:53 pm
So what you are really saying is that you refuse to do the actual science.
Yet you claim to understand the scientific method, where clearly you do not fully understand the scientific method, as is abundently clear from your posts here at WUWT.
The “jar of fleas” argument was not a council of despair. I do not argue that we cannot understand at least some key elements of the climate system. I have no problem with doing real experimental science – on the contrary, CAGW skeptic arguments tend to be based on experimental measurements while CAGW rests on a foundation of computer modeling. [snip] …
yadda, yadda, yadda, I know a bunch of dead philosophers, from long ago, I’ll cherry pick a few that fit my biases, yadda, yadda, yadda.
……
_____________________________________________________________
But briefly, experimental science is the eminent domain of the scientist, not the septic tankers.
You don’t do ANY experimentation yourselves.
Just about all (99.44%) of the climate science data is collected by climate scientists, NOT the septic tanlers.
You are TEAM AUDIT!
If you don’t like inductive reasoning and theory, that’s your problem, not mine.
“Oh, lordy, lordy, the world is too complex to model, heaven forbid, at any scale, at any time, at any place. Because I said so, although I could never prove it.”
The above paraphrased quote about sums up your deeply flawed argument.
REPLY: Junior, your arguments aren’t going anywhere, and amount to taunts. I’d suggest an extended time out. – Anthony
richard telford says:
September 26, 2010 at 5:12 am
“…Validation of proxies”
Unless there is overlap between proxy data and actual observations – a direct comparison between examples of the proxy and historical records – you do not have a validated proxy. You may have a model or theory which you believe to be robust, but experimental confirmation is lacking. This is quite general, a basic requirement of the scientific method. All the cross-correlations and “training sets” in the world cannot change it. Perhaps you did mean to imply that such direct comparisons have been utilised here, but I’m afraid I found your reply to be ambiguous and needlessly convoluted. So I will ask you straight out: is there a direct confirmation of the dinocysts for sea ice proxy, between actual cores and actual observations? If so, what dates and areas are covered by this?
Paul Birch says:
September 27, 2010 at 3:39 pm
Ideally we would be able to validate proxies against historical data, but this is often not possible due to the short instrumental record in many regions. In these cases cross-validation (not cross correlation) is adequate. Cross-validation is a space for time substitution, the idea being that if the proxy can reconstruct the modern environment in different locations, it is valid to use it to reconstruct the environment in the past. There are some problems with this approach, for example when assemblages without good modern analogues are encountered (a particular problem for pollen reconstruction).
If you want to argue that proxies must be validated against historical records – a demand that severely limit our ability to understand past climate – you need to show that cross-validation is inadequate.
The dinocyst-salinity transfer function fails robust cross-validation, I would be surprised if the sea-ice transfer function performs well.
richard telford says:
September 28, 2010 at 4:29 pm
“…”
Kindly just answer my question: is there a direct confirmation of the dinocysts for sea ice proxy, between actual cores and actual observations? If so, what dates and areas are covered by this?
Wayne Delbeke – re Antarctic sea ice – see “Polar see-saw” in
http://www.spacecenter.dk/research/sun-climate/Scientific%20work%20and%20publications/resolveuid/86c49eb9229b3a7478e8d12407643bed
We can expect the warmists to switch their attention from the Arctic to the Antarctic soon …..
(there is no end to this pain)
Paul Birch says:
September 29, 2010 at 1:03 pm
Passing cross validation is a necessary and sufficient condition. Dinocysts fail this test.
richard telford says:
September 30, 2010 at 2:44 am
“Passing cross validation is a necessary and sufficient condition. Dinocysts fail this test.”
Why are you so unwilling to answer a simple question, without trying to turn it into something else? If you don’t know, say you don’t know. And if you don’t understand the question, ask. The question is not about the robustness, accuracy or uncertainties of the proxy, nor about dinocyst proxies in general, but about whether this particular proxy (dinocysts for sea ice) has been shown to be a valid one by means of actual observations; and it is about the proxy itself, not the use of it in this study.
So, for the third time of asking: Is there a direct confirmation of the dinocysts for sea ice proxy, between actual cores and actual observations? (This can be answered “yes”, “no”, or “I don’t know”). If so, what dates and areas are covered by this?
Paul Birch says:
September 30, 2010 at 3:38 am
Your question is of little relevance. Since the dinocysts have not been shown to pass robust cross-validation, they shouldn’t be used for reconstructing sea ice.
There are few multi-century observational ice cover records available for validation. There is one north of Iceland that has been used to validate a geochemical marker of sea ice algae, but I don’t think it has been used to test dinocysts.
Barry says “I don’t think your commentary matches what is being said. ”
http://climatechangepsychology.blogspot.com/2010/09/serreze-i-stand-by-my-previous.html
NSIDC director Serreze: “The volume of ice left in the Arctic likely reached the lowest ever level this month.”
Serreze: “I stand by my previous statements that the Arctic summer sea ice cover is in a death spiral. It’s not going to recover.”
“The reason so much (of the Arctic ice) went suddenly is that it is hitting a tipping point that we have been warning about for the past few years.” James Hansen, 2007
richard telford says:
September 30, 2010 at 5:12 am
“Your question is of little relevance. ”
No, it isn’t. Stop wriggling and answer it. If you can’t, just admit you can’t.
For the fourth time of asking: Is there a direct confirmation of the dinocysts for sea ice proxy, between actual cores and actual observations? (Answer “yes”, “no”, or “I don’t know”). If so, what dates and areas are covered by this?
Paul Birch says:
October 1, 2010 at 3:16 am
I fail to understand why you are so keen to know if the dinocyst-sea ice transfer function has been validated against historical data (in any case, the answer to your quest should be obvious from my previous). The dinocysts have not been shown to pass robust cross-validation. This is a major problem. If they cannot pass cross-validation, it is pointless to ask if they can be validated. It is like demanding to know if a car with no wheels has an MOT.
richard telford says:
October 3, 2010 at 3:18 am
“I fail to understand why you are so keen to know if the dinocyst-sea ice transfer function has been validated against historical data (in any case, the answer to your quest should be obvious from my previous). The dinocysts have not been shown to pass robust cross-validation. This is a major problem. If they cannot pass cross-validation, it is pointless to ask if they can be validated. It is like demanding to know if a car with no wheels has an MOT.”
You don’t actually need to understand why I’m asking. All you need to understand is the question. If you know the answer, all you have to do is to give that answer. If you don’t know the answer, all you have to do is to say you don’t know. Until you have answered that question, without trying to morph it into something slightly different, I can have no confidence in anything you say about what you claim to be “a major problem”, or in the relevance of whether you consider dinocyst proxies to be robust.
If I ask whether your car has an MOT, that’s what I want to know. Not whether or not it has wheels. Many cars with no wheels still have MOTs.
So, for the fifth time of asking: Is there a direct confirmation of the dinocysts for sea ice proxy, between actual cores and actual observations? (Answer “yes”, “no”, or “I don’t know”). If so, what dates and areas are covered by this?
Paul Birch says:
October 4, 2010 at 4:05 am
As I alluded to before, I am not aware of any attempt to correlate dinocyst-ice reconstructions with historical records, there are very few places where this could be done. There may be attempts to validate dinocysts against independent proxies. I should imagine if such data existed, the dinocyst community would have used them to rebut Telford (2006) rather than relying on a series of invalid statistical tests.
richard telford says:
October 5, 2010 at 2:12 pm
“I am not aware of any attempt to correlate dinocyst-ice reconstructions with historical records, there are very few places where this could be done. ”
Thank you.
Surprise … you actually should read this paper.
It says and I quote…
…However, sea ice has continued its rapid decline, since the AO returned to a more neutral state in the late 1990s, suggesting that anthropogenic warming of surface air temperatures is playing a role in the loss (Overland and Wang 2005), as now recognized by the Intergovernmental Panel on Climate Change (IPCC 2007).”
This paper you are calling the death of AGW actually says it is AGW is occuring … IN THE PAPER ITSELF.
Pardon me if I don’t jump for joy.
Texanjeff,
You do understand they are simply laying out the arguments for AGW not endorsing that position? They explicitly reference Overland and Wang 2005 for this argument. Papers do this all the time, they present the opposing argument or the source of debate their paper may be addressing. This has nothing to do with their conclusions.