Guest Commentary by Axel Schweiger, Ron Lindsay, and Cecilia Bitz
We have just passed the annual maximum in Arctic sea ice extent which always occurs sometime in March. Within a month we will reach the annual maximum in Arctic sea ice volume. After that, the sea ice will begin its course towards its annual minimum of both extent and volume in mid-September. This marks the beginning of the ritual of the annual sea ice watch that includes predictions of the extent and rank of this year’s sea ice minimum, as well as discussion about the timing of its eventual demise. One of the inputs into that discussion is the “PIOMAS” ice-ocean model output of ice volume – and in particular, some high-profile extrapolations. This is worth looking at in some detail.
Prediction methods for the sea ice minima range from ad-hoc guesses to model predictions, from statistical analyses to water-cooler speculation in the blogosphere. Many of these predictions are compiled in the SEARCH-sponsored “sea ice outlook“.
This year’s discussions however will be without the input of the father of modern sea ice physics, Norbert Untersteiner, who recently died at the age of 86. Much of the physics in PIOMAS and global climate models can be traced to Norbert’s influence. Norbert was sober-minded and skeptical about the prospects of skillful short-term sea ice predictions, but even he couldn’t help but be drawn into the dubious excitement around the precipitous decline of arctic sea ice and regularly added his own guestimate to the sea ice outlook. Norbert’s legacy challenges those of us who engage in predictions to prove our skill and to understand and explain the limitations of our techniques so they are not used erroneously to misinform the public or to influence policy…more about that later and here.
PIOMAS is the Panarctic Ice Ocean Modeling and Assimilation System. It belongs to the class of ice-ocean models that have components for the sea ice and the ocean, but no interactive atmosphere. There is an active community (AOMIP) engaged in applying and improving these types of models for Arctic problems. Without an atmosphere, inputs that represent the atmospheric forcing (near surface winds, temperature, humidity, and downwelling short and longwave radiation) need to be provided. Typically those inputs are derived from global atmospheric reanalysis projects. The advantage of such partially-coupled models is that they can be driven by past atmospheric conditions and the simulations match well the observed sea ice variability, which is strongly forced by the atmosphere.
This is in contrast to fully-coupled models, such as those used in the IPCC projections, which make their own version of the weather and can only be expected to approximate the mean and general patterns of variability and the long-term trajectory of the sea ice evolution. Another advantage of ice-ocean models is that they don’t have to deal with the complexities of a fully-coupled system. For example, fully-coupled models have biases in the mean wind field over the Arctic which may drive the sea ice into the wrong places, yielding unrealistic patterns of sea ice thickness. This has been a common problem with global climate models but the recent generation of models clearly shows improvement. Because sea ice is strongly driven by the atmosphere, model predictions depend on the quality of the future atmospheric conditions. Therefore an ice-ocean model, like PIOMAS, is much more accurate at hindcasts, when the atmospheric conditions are simply reconstructed from observations, than for forecasts, when atmospheric conditions must be estimated. That is not to say that PIOMAS can’t be used for predictions, it can (Zhang et al. 2008, Lindsay et al. 2008 , Zhang et al. 2010) but it is important to recognize that performance at hindcasts does not necessarily say much about performance at forecasts. This point often gets confused.
Figure 1: PIOMAS mean monthly arctic sea ice volume for April and September. Dashed lines parallel to linear fits represent one and two standard deviations from the trend. Error bars are estimated based on comparison with thickness observations and model sensitivity studies (Schweiger et al. 2011)
PIOMAS was developed and is operated by Jinlun Zhang at the University of Washington. It is the regional version of the global ice-ocean model of Zhang and Rothrock (2003). The sea ice component represents sea ice in multiple categories of thickness and accounts for changes in thickness due to growth and melt as well as mechanical deformation of ice (Thorndike et al. 1975, Hibler 1980).
It has evolved with continual improvements, including the addition of data assimilation capabilities (Zhang et al. 2003, Lindsay et al. 2006) and the development of sister models for new applications (BIOMAS for biology) or specific regions (BESTMAS for the Bering Sea and GIOMAS for the entire globe) (publications). As a modeler working among observationalists from a variety of disciplines, Jinlun has never been short of tire-kickers who probe, push, and challenge his model from all sorts of different angles and identify warts and beauty spots. This is one of the reasons why PIOMAS has evolved into one of the premier ice-ocean models (Johnson et al. 2012), particularly when it comes to the representation of the sea-ice cover.
PIOMAS has been used in a wide range of applications but arguably the most popular product has been the time series of total Arctic sea ice volume which we have been putting out since March 2010 (see also Fig 1). The motivation for this time series is to visualize the fact that the long term Arctic-wide loss of sea ice is not only happening in extent, which is well measured by satellites, but also in thickness, which isn’t. Ice volume, the product of sea ice area and thickness, is a measure for the total loss in sea ice and the total amount of energy involved in melting the ice. Though this is a very small part of the change of global energy content, it is regionally important and investigations into the cause of sea ice need to pin down the sources of this energy.
But why use PIOMAS to show the decline in ice volume when our group of researchers has been involved in measuring, rescuing, and collecting sea ice thickness data from in-situ observations for 30-some years? The answer is that even though wide-spread thickness losses from observations alone have been apparent for some areas or time periods, Arctic-wide thickness losses are more difficult to document because of the sparse sampling in time and space. The problem can be visualized by constructing a “naïve” sea ice thickness time series from in-situ observations:
Figure 2 Naïve sea ice thickness time series. Sea ice thickness observations from the sea ice thickness climate data record (small grey dots), averages for all observations in a given year (large grey dots), and 5-year running mean through those observations. The same calculation for the corresponding PIOMAS simulations at the location and time of observation is shown by the big red dots and line.
Before those claiming that global warming stopped in 1998 have a field day with this figure, they should appreciate that our total volume time series and the naïve thickness time series are entirely consistent. The sampling issues arise from the fact that sea ice is highly dynamic with lots of spatial and seasonal variability so that measurements from individual moorings, submarine sonar tracks, and aircraft flights can only construct an incomplete picture of the evolution of the total Arctic sea ice volume. Progress towards establishing ice thickness records from satellite (ICESat, Envisat, and CryoSat-2) will change this over time, but these sources won’t yield a record before these measurements began and satellite retrievals of ice thickness have their own issues.
PIOMAS is not normally run as a freely-evolving model, but rather it assimilates observations. Ice concentration and sea surface temperature are currently assimilated and we have experimented with the assimilation of ice motion (Zhang et al. 2003, Lindsay et al. 2006). Assimilation helps constrain the ice extent to observations and helps improve the simulation of sea ice thickness. Ice thickness observations are not assimilated into the model. Instead, ice thickness and buoy drift data are used for model calibration and evaluation. So using a model constrained by observations is quite possibly the best we can do to establish a long-term ice volume record.
Model calibration is of course necessary. We need to determine parameters that are not well known, deal with inadequately modeled physics, and address significant biases in the forcing fields. Parameters changed in PIOMAS calibration are typically the surface albedo and roughness, and the ice strength. Once calibrated, the model can be run and evaluated against observations not included in the calibration process. Evaluation does not only mean showing that PIOMAS says something useful but also establishes the error bars on the estimated ice thickness. To establish this uncertainty in the ice-volume record (Schweiger et al. 2011), we spent a significant effort drawing on most types of available observations of ice thickness thanks to a convenient compilation of ice thickness data (Lindsay, 2010). We have also compared PIOMAS estimates with measurements from ICESat and conducted a number of model sensitivity studies. As a result of this evaluation our conservative estimates of the uncertainty of the linear ice volume trend from 1979-present is about 30%. While there is lots to do in improving both measurements and models to reduce the uncertainty in modeled ice volume, we can also say with great confidence that the decline in observed ice thickness is not just an effect of measurement sampling and that the total sea ice volume has been declining over the past 32 years at astonishing rates (for instance a 75% reduction in September volume from 1979 to 2011).
The seasonal prediction issue and the prediction of the long-term trajectory are fundamentally different problems. Seasonal prediction, say predicting September ice extent in March, is what is called an initial value problem and the September ice extent depends both on the weather, which is mostly unpredictable beyond 10 days or so, and the state of the ocean and sea ice in March. Improving observations to better characterize that state, and improving models to carry this information forward in time is our best hope to improve seasonal predictability. The prediction of the long-term trajectory, depends on the climate forcing (greenhouse gases, aerosols, solar variability) and how the model responds to those forcings via feedbacks. A recent model study showed that the crossover between initial-value and climate-forced predictability for sea ice occurs at about 3 years (Blanchard-Wrigglesworth et al. 2011). In other words, a model forgets the initial sea ice state after a few years at which point the main driver of any predictable signal is the climate forcing. In fact, coupled model simulations have shown that even removing all the sea ice in a particular July has little lasting impact on the trajectory of the ice after a few years (Tietsche et al. 2011).
PIOMAS has been run in a forward mode (and hence without data assimilation) to yield seasonal predictions for the sea ice outlook (Zhang et al. 2008) and has also provided input to statistical forecasts (Lindsay et al. 2008) and fully-coupled models. We have also done experiments with PIOMAS in a climate projection mode by scaling atmospheric forcing data from a reanalysis to 2xC02 projections from the CMIP3 models (Zhang et al. 2010). This setup provides more realistic wind fields and spatial thickness distribution but cannot account for important atmosphere-ocean feedbacks.
Global climate model projections (in CMIP3 at least) appear to underestimate sea ice extent losses with respect to observations, though this is not universally true for all models and some of them actually have ensemble spreads that are compatible with PIOMAS ice volume estimates and satellite observations of sea ice extent. With error bars provided, we can use the PIOMAS ice volume time series as a proxy record for reality and compare it against sea-ice simulations in global climate models. This provides another tool in addition to more directly observed properties for the improvement and evaluation of these models and is in our view the best use of PIOMAS in the context of predicting the long-term trajectory of sea ice.
Predictions of a seasonally ice-free Arctic Ocean
The eventual demise of the summer sea ice is a common feature of nearly every climate model projection (the exceptions are models with very inappropriate initial conditions). But the question of when the Arctic will be ‘ice-free’ is imprecise and calls for a clear definition of what ice-free means. Does it mean completely ice-free, or is there a minimum threshold implied? Does it mean the first time the summer sea ice goes beneath this threshold or does it imply a probability of encountering low-ice conditions over a period of time? (e.g. high likelihood of Septembers with less than 106 km2 of ice in a 10-year period). Regardless of whether the concept is actually useful for any practical purpose (say for planning shipping across the Arctic), it is nevertheless a powerful image in communicating the dramatic changes that are under way in the Arctic.
Once defined, predictions of when an ice-free Arctic will occur seem justified. In the published literature there are several papers specifically targeting such predictions (Zhang and Walsh, 2006, Wang and Overland, 2009, Boe et al. 2009, Zhang et. al. 2010) while others include discussion about the timing of ice-free summers (e.g. Holland et al. 2006). Some address the fact that the CMIP3/IPCC AR4 simulations show sea ice declines less rapid than the observations and attempt to correct for it. Published projections, though with varying definitions of what constitutes ice-free, all project an ice-free Arctic ocean somewhere between 2037 (Wang and Overland, 2009) and the end of the century. Predictions of earlier ice-free dates so-far seem to be confined to conference presentations, media-coverage, the blogosphere, and testimony before to the UK parliament.
A different class of predictions are based on simple extrapolation using historical sea ice extent, concentration, or volume. An example is included in the materials presented by the so-called ‘Arctic Methane Emergency Group’ who show extrapolations of PIOMAS data and warn about the potential of a seasonally ice-free Arctic ocean in just a few years. So does it make sense to extrapolate sea ice volume for prediction? In order to do a successful extrapolation several conditions need to be met. First, an appropriate function for the extrapolation should be chosen. This function needs to either be based on the underlying physics of the system or needs to be justified as appropriate for future projections beyond just fitting the historical data.
But what function should one choose? Since we don’t really have data on how the trajectory of the Arctic sea ice evolves under increased greenhouse forcing, model projections may provide a guide about the shape of appropriate function. Clearly, linear, quadratic or exponential functions do not properly reflect the flattening of the trajectory in the next few decades seen for example in the CCSM4 (Fig 3). The characteristic flattening of this trajectory, at first order, arises from the fact that there is an increasingly negative (damping) feedback as the sea ice thins described by Bitz and Roe (2004) and Armour et al. (2011). The thick ice along the northern coast of Greenland is unusually persistent because there are on-shore winds that cause the ice to drift and pile-up there. So extrapolations by fitting a function that resembles a sigmoid-shaped trajectory may make more sense, but even that, as shown in the figure, yields a much earlier prediction of an ice-free Arctic than can be expected from the CCSM4 ensemble.
Figure 3. CCSM4 AR4 ensemble and PIOMAS September mean arctic ice volume. Exponential and sigmoid (Gompertz) fits to PIOMAS data are shown. Note that the 1979-2011 September mean of the CCSM4 ensemble has about 30% higher sea ice volume than PIOMAS. To visualize the difficulty in choosing an appropriate extrapolation function based on PIOMAS data we shifted the CCSM4 time series forward by 20 years to roughly match the mean ice volume over the 1979-2011 fitting period.
But there is a second issue that may foil prediction by extrapolation: The period over which the function is fit must be sufficiently long to include adequate long-term natural variability in the climate system. The goodness of fit over the fitting period unfortunately may be misleading. Whether or not this is the case for sea ice extent or volume is an open question. The sea ice trajectory shows considerable natural variability at various time scales on top of the smoother forced response to changes in greenhouse gases. Periods of rapid decline are followed by slower periods of decline or increases. By fitting a smooth function to a sea ice time series (e.g. PIOMAS) one might easily be tempted to assume that the smooth fit represents the forced (e.g. greenhouse) component and the variation about the curve is due to natural variability. But natural variability can occur at time scales long enough to affect the fit. We have to remember that part of the observed trend is likely due to natural variability (Kay et al. 2011, Winton, 2011) and may therefore have little to do with the future evolution of the sea ice trajectory. This is visualized in figure 4 where ensemble members from the CCSM4 AR4 runs are fit with S-shaped (Gompertz) functions using the 1979-2011 period to estimate the parameters. The differences between the ensemble members, reflecting natural variability, yield vastly different extrapolated trajectories. Natural variability at these time scales (order of 30 years) may very well make prediction by extrapolation hopeless.
Figure 4. CCSM4 AR4 ensemble with sigmoid (Gompertz) fits. Light vertical lines represent fitting period for ensemble members (1979-2011).
In summary, we think that expressing concern about the future of the Arctic by highlighting only the earliest estimates of an ice-free Arctic is misdirected. Instead, serious effort should be devoted to making detailed seasonal-to-interannual (initial-value) predictions with careful evaluations of their skill and better estimates of the climate-forced projections and their uncertainties, both of which are of considerable value to society. Some effort should also target the formulation of applicable and answerable questions that can help focus modeling efforts. We believe that substantially skillful prediction can only be achieved with models, and therefore effort should be given to improving predictive modeling activities. The best role of observations in prediction is to improve, test, and initialize models.
But when will the Arctic be ice free then? The answer will have to come from fully coupled climate models. Only they can account for the non-linear behavior of the trajectory of the sea ice evolution and put longer term changes in the context of expected natural variability. The sea ice simulations in the CMIP5 models are currently being analyzed. This analysis will reveal new insights about model biases, their causes, and about the role of natural variability in long-term change.It is possible that this analysis will change the predicted timing of the “ice free summers” but large uncertainties will likely remain. Until then, we believe, we need to let science run
its course and let previous model-based predictions of somewhere between “2040 and 2100″ stand”
Bitz, C. M., and G. H. Roe (2004), A mechanism for the high rate of sea ice thinning in the Arctic Ocean, J Climate, 17(18), 3623-3632.
Boe, J. L., A. Hall, and X. Qu (2009), September sea-ice cover in the Arctic Ocean projected to vanish by 2100, Nature Geoscience, 2(5), 341-343.
Hibler, W. D. (1980), Modeling a Variable Thickness Sea Ice Cover, Monthly Weather Review, 108(12), 1943-1973.
Holland, M. M., C. M. Bitz, and B. Tremblay (2006), Future abrupt reductions in the summer Arctic sea ice, Geophys. Res. Lett, 33(23), 5.
Johnson, M., et al. (2012), Evaluation of Arctic sea ice thickness simulated by Arctic Ocean Model Intercomparison Project models, J. Geophys. Res., 117, C00D13.
Kay, J. E., M. M. Holland, and A. Jahn (2011), Inter-annual to multi-decadal Arctic sea ice extent trends in a warming world, Geophys. Res. Lett, 38.
Lindsay, R. W. (2010), New Unified Sea Ice Thickness Climate Data Record, Eos Trans. AGU, 91(44), 405-416.
Lindsay, R. W., J. Zhang, A. J. Schweiger, and M. A. Steele (2008), Seasonal predictions of ice extent in the Arctic Ocean, J.Geophys.Res., 113(C2), 11.
Lindsay, R. W., and J. Zhang (2006), Assimilation of ice concentration in an ice-ocean model, Journal of Atmospheric and Oceanic Technology, 23(5), 742-749.
Rothrock, D. A., Y. Yu, and G. A. Maykut (1999), Thinning of the Arctic sea-ice cover, Geophys. Res. Lett, 26(23), 3469-3472.
Schweiger, A. J., R. Lindsay, J. Zhang, M. Steele, H. Stern, and R. Kwok (2011), Uncertainty in modeled Arctic sea ice volume, J. Geophys. Res., 116, C00D06.
Tietsche, S., D. Notz, J. H. Jungclaus, and J. Marotzke (2011), Recovery mechanisms of Arctic summer sea ice, Geophys. Res. Lett, 38.
Thorndike, A. S., D. A. Rothrock, G. A. Maykut, and R. Colony (1975), Thickness Distribution of Sea Ice, J.Geophys.Res., 80(33), 4501-4513.
Wang, M. Y., and J. E. Overland (2009), A sea ice free summer Arctic within 30 years?, Geophys. Res. Lett, 36, 5.
Winton, M. (2000), A reformulated three-layer sea ice model, Journal of Atmospheric and Oceanic Technology, 17(4), 525-531.
Winton, M. (2011), Do Climate Models Underestimate the Sensitivity of Northern Hemisphere Sea Ice Cover?, J Climate, 24(15), 3924-3934.
Zhang, J., D. R. Thomas, D. A. Rothrock, R. W. Lindsay, Y. Yu, and R. Kwok (2003), Assimilation of ice motion observations and comparisons with submarine ice thickness data, J.Geophys.Res., 108(C6), 3170, DOI: 3110.1029/2001JC001041
Zhang, J., and D. A. Rothrock (2003), Modeling global sea ice with a thickness and enthalpy distribution model in generalized curvilinear coordinates, Monthly Weather Review, 131(5), 845-861.
Zhang, X. D., and J. E. Walsh (2006), Toward a seasonally ice-covered Arctic Ocean: Scenarios from the IPCC AR4 model simulations, J Climate, 19(9), 1730-1747.
Zhang, J., M. Steele, and A. Schweiger (2010), Arctic sea ice response to atmospheric
forcings with varying levels of anthropogenic warming and climate variability, Geophys.
Res. Lett, 37 (L20505)
302 Responses to "Arctic Sea Ice Volume: PIOMAS, Prediction, and the Perils of Extrapolation"
Oh, my, he just gish galloped on by and rather than acknowledge that he’d “misinterpreted” his sources, just started hand-waving about models, “lots of [unspecified] papers”, etc.
Yes, good riddance. He’s no seeker of knowledge.
Kevin McKinney says
#249 is a tad confusing–it sounds as though you are talking about Kahl et al., 1993, but the link is to Bengsston et al., 2004. (?)
Dan H. says
I think you are mistaken. I am referring to the paper to which Chris linked (Bengtsson, 2004) and its partner(Johannessen, 2004), not Kahl, 1993. Kevin has it right.
Hank Roberts says
It’s hard to tell what Dan H. is talking about without citations, innit?
This is probably
> Bengtsson, et. al.
The Early Twentieth-Century Warming in the Arctic—A Possible Mechanism
LENNART BENGTSSON, VLADIMIR A. SEMENOV, OLA M. JOHANNESSEN
might be the book that I guessed Jim might mean (21 Apr 2012 at 4:03 PM)
Johannessen, O.M., L. Bengston, et al., 2004. Arctic climate change, observed and modeled temperature …
Jim had misspelled an author’s name but that wasn’t the one he meant.
http://scholar.google.com/scholar?hl=en&q=johannessen+arctic finds many more.
We can’t know what they’re talking about without an actual link or full cite.
Hank Roberts says
One suggestion for those wanting to follow the article instead of the herring
— see the references for the article to further understanding the article. The reference list is at the top of the page, immediately after the main post.
Select an item and paste it into Scholar to see if it’s available; often a PDF or author’s home page copy is among several versions Scholar finds.
Referring to he references to ask about the points made in the article — rather than hinting of, somewhere, a talking point — will help.
I’m as confused as the next reader about this stuff, and batting at the bafflegab wastes life and time better used following the science.
Kevin McKinney says
I scanned both the 2004 papers. The most interesting thing for me was in (IIRC) the *Bengtsson*, where they proposed that the main physical factor in the ‘variability-driven warming/SI decline’ of the ’30s was a persistent circulation pattern resulting in a ‘clearing’ of the Barents (or Barentsz, if you prefer), which in turn fed back into a larger warming. Of course, that result is now 8 years old; I don’t know how well it’s stood up.
But it’s interesting, because currently there’s a rather anomalous clearing of the Barents as well–albeit on a shorter timescale. We’ll see what happens…
Dan H. says
You really need to follow the discussion better. Links to the Bengtsson paper, which we are discussing, have been presented in posts #233, 234, 235, and 250, along with links to the Johannessen paper in #221, 228 and 230. It really is not that hard to follow the discussion once you have read the relevant papers.
Dan H. says
I noticed the same connection. What was missing from the paper is an explanation as to the reversal between the 1940s and 1960s.
Jim Larsen says
Ray L, I meant permafrost and ESAS clathrates would have been extremely cold at the beginning of the Holocene, while now they’re making a move towards slushy. Assuming we do at least as bad as the early Holocene, then our inferior starting position means we could be at risk of a significant methane release.
On ice volume, both from a linear trend and from computer models, sea ice is likely to recover a bit for at least a decade, which could frame the debate for years to come.
But why is ice so low? The weather has been ice-friendly, yet the volume remains record-breakingly low. Why? This seems like an easy way to falsify the exponential hypothesis.
Hank Roberts says
> Hank, You really need to follow the discussion better.
> Links to the Bengtsson paper, which we are discussing,
> have been presented in posts #233, 234 ….
Yes, I’m very grateful to the guy who posted #233 and 234, who guessed that might be what you and Jim were referring to, and provided the links asking you to confirm what you were talking about.
Oh, wait, that was me.
Thanks to Chris, who promptly nailed the cite. That’s how it’s done.
That helps bring us back to the topic at hand here.
Chris Reynolds says
I didn’t ‘promptly’ do it, it took a few days before I woke up and paid attention to the other discussions here. Had I done so sooner…
Now this is the interesting discussion. But for the assimilating models (PIOMAS & NPS), there really wouldn’t be much talk of an imminent sea-ice free Arctic. There was 2011, which very nearly met 2007 despite weather not really being conducive to ice melt, certainly not as much as 2007 itself. Otherwise, post 2007 the sea-ice seems to have settled into a new (pseudo) equilibrium, it hasn’t undergone a rapid succession of crashes.
Then there’s Maslanik’s work with the Drift Age Model.
And Nghiem’s findings of a massive crash in multi-year sea ice.
Maslanik doesn’t find the same precipitous drop as Nghiem, this is because Maslanik counts mixed ice in with perennial, whereas Nghiem discounted it. Considering the two approaches, what they show is that the pack has now transtioned to a seasonally sea ice pack, with a residual amount of multi year ice. This residual is a logical consequence of there still being sea ice growth in the winter, and a significant area of ice at the end of the summer.
These and other issues present a picture of an ice pack that’s now transitioned from largely multi-year ice in a large mass to mainly first year, with multi year becoming more limited in region (off the Canadian Archipelago) and spread out when it leaves that safe zone.
So what is going on? Observations support the idea that we’re in a slow transition with the ice being in a pseudo-equilibrium state of mainly first year ice. While the assimilating models, our only current option for widespread volume indication, imply a precipitous crash of volume.
Short of actual thickness/volume data, is there any way to cut through this issue and clarify what is going on?
Mr. Chris Reynolds wrote on the 24th of April, 2012 at 12:17 PM
“Short of actual thickness/volume data…”
Here we go:
“Cryosat found the volume (area multiplied by thickness) of sea ice in the central Arctic in March 2011 to have been 14,500 cubic kilometres.
This figure is very similar to that suggested by PIOMAS (Panarctic Ice Ocean Modeling and Assimilation System), …”
John Nissen says
At Chris, #261, who asked:
“Short of actual thickness/volume data, is there any way to cut through this issue and clarify what is going on?”
I think I can.
Professor Peter Wadhams, member of AMEG, expert on Arctic sea ice and a reviewer for the IPCC AR5 report, says that the PIOMAS data is based on actual thickness measurements.
[Response: This is not true. Please read the top post. – gavin]
As the sea ice retreats, the water absorbs more sunshine and warms faster. This means that the sea ice formed in winter is thinner and melts faster in summer. As a result, the sea ice volume at its annual minimum has declined 75% over the past three decades. Although the sea ice extent has held up since 2007, the thickness has declined; but the extent cannot continue to hold up indefinitely while the thickness continues to decline. Wadhams expects a collapse in extent within the next few years.
Note that, from the guest commentary at the start of this blog, we read:
“A different class of predictions are based on simple extrapolation using historical sea ice extent, concentration, or volume. An example is included in the materials presented by the so-called ‘Arctic Methane Emergency Group’ [AMEG] who show extrapolations of PIOMAS data and warn about the potential of a seasonally ice-free Arctic ocean in just a few years.”
However, the commentary concludes:
“But when will the Arctic be ice free then? The answer will have to come from fully coupled climate models.”
Wadhams and the AMEG argue that observations are more to be relied on than models, especially when the models have proved unreliable in the past.
[Response: There are no observations of the future yet available, and extrapolation – particularly using exponentials – is fraught with over-confidence and almost certain error. We are well aware of what you are doing to get your result, our complaint is that you haven’t justified why it makes any sense. – gavin]
BTW, Wadhams predictions on sea ice were challenged by Prof Julia Slingo of the Met Office’s Hadley Centre, at the hearing of the Environment Audit Committee on “Protecting the Arctic”. He has written a robust rebuttal, which will shortly be in the public domain.
Hank Roberts says
John Nissen may be misremembering his own page, which needs help from an editor: http://ameg.me/index.php/sea-ice
“Arctic expert, Professor Peter Wadhams (UK), said in November 2011 that the summer sea ice is on track to be virtually gone by 2015.” (No cite)
and below that
“The Arctic Catlin Survey …. results (of direct ice thickness measurements by bore holes coupled with historic data) in October 2009, Professor Wadhams said, …. supports the new consensus view that the Arctic will be ice-free in summer within about 20 years. (CNN)” (No cite; what consensus statement, by what group or organization, where published and when?)
and below that
A duplicate paragraph.
Looking for “Catlin Survey” — nothing in Scholar; the “Catlin Survey” is mentioned as an anecdotal unpublished source thus:
“… anecdotal in-situ observations in spring 2009 [SEARCH Sea Ice Outlook, unpublished, 2009; CATLIN Arctic Survey, unpublished, 2009].”
in “Synoptic airborne thickness surveys reveal state of Arctic sea ice cover” http://epic.awi.de/21995/1/Haa2010b.pdf
That paper (on the use of aircraft instruments) is cited by some familiar names, by the way: http://scholar.google.com/scholar?cites=15201257753537939686
Has anyone found the “Catlin Arctic Survey” data and documentation available somewhere? Info on who maintains it and perhaps whether it will be published at some point?
Hank Roberts says
OK, here it is: http://www.catlin.com/en/Responsibility/CatlinArcticSurvey
I would like to correct my statement of 21st Apr 2012 at 3:06 PM:
melting 300GT/yr as implied by PIOMASS is a heat sink of 1e20J/yr
oddly enuf this is roughly the rate of heat accumulation per year in the Atlantic north of 70 degrees over the last 50 years. (Fig S5 in Levitus 2012 available at
Nigel Williams says
Well, with the global oil exporting countries on track to consuming all the oil they produce by 2025-2030 the rest of the world will be back in the dark ages well before we get an ice-free Arctic. The only knowledge we will have about conditions up there will from stories of distant places from the odd wind-powered whaling ship that ventures through Bering Straight.
As global communications blink out in the post-oil era the havoc an ice-free Acrtic will have on global climate (and vice-versa) will only be evidenced as the seas rise, and where the rain falls or the winds blow, or not.
We will huddle behind our shutters knowing we have unleashed a tiger, but powerless to stop it. Oh darn.
Susan Anderson says
I followed the Catlin Ice Survey – kind of a hike with hardships (British), not meaning to belittle it – seemed fraught with difficulty, as all work in that part of the world would naturally be – a few years back. Fascinating stuff.
I seem to remember some issues with who was funding it, but as far as I could see it was a gallant effort. Of course as with any appearance genuine physical evidence, the deniers were all over it to discredit.
I should, seeing where I’m saying this, follow through, but am very busy so will just contribute this anecdotal note FWIW.
Susan Anderson says
Oh, and I think their diary was published on the BBC at the time. (Catlin)
Brian Dodge says
Here’s why the Arctic sea ice loss is unsustainable. &;>)
Kevin McKinney says
#271–Irony, however, is showing a robust linear growth trend.
Hank Roberts says
There’s presumably a data collection and papers either published or being written; I haven’t found them. There are a lot of different web pages out there for Catlin’s Arctic investigations, different each year. I found hints; I’m sure someone can find the science. I think this is just obscured because it’s being presented in corporate PR format, which means to be informative and probably is, for those who like that kind of thing.
“This endeavour will provide a surface-based dataset, which will then be made available to scientists. Its data will be used to improve the accuracy and reliability of supercomputer models forecasting the timing of the disappearance of the sea ice ….
… the supercomputer model developed by one of the world’s leading research teams at the US Navy’s Department of Oceanography, which focuses on the rate of the sea ice’s declining volume based on ice thickness estimates (as opposed to shrinkage rates), indicates sea ice loss within dramatically less time …”
“The results of the 2010 Catlin Arctic Survey are still being analysed. The results and conclusions will be posted here when they become available.”
Chris Reynolds says
Thanks, I was made aware of that article and the graphic at Neven’s after posting the above. It still doesn’t answer the question of the reality of the last two year’s Spring volume losses. But it does instill confidence in the overall PIOMAS trend.
That has however to be considered against findings such as those by Haas, to which Hank linked in #264. That paper finds:
Which seems to be the case from my amateur analysis of what little observed thickness data I can find for the 2010/2011 period. Although I’ve not been able to find enough to get a broad view of the Arctic, and some data such as Buoys suffers from an observational bias – more substantial floes of ice are chose for the placement of buoys.
However now Cryosat is up and running some of this uncertainty should start to be resolved.
Chris Reynolds says
#263 John Nissen,
I don’t think you’ve cut through the issue at all. Yes, as Stroeve et al 2011 argue, increased open water formation increases the ice albedo effect. But let’s not get things out of proportion.
Whilst there has been a continuing loss of thick multi-year ice (Maslanik – drift age model) after the precipitous drop Nghiem 2008 revealed using QuikScat: This is in line with the arguments of Bitz & Roe – thicker ice thins faster. So once again we’re thrown back on the precipitous drop of volume shown by assimilating models. But this doesn’t necessarily mean that behaviour will continue to zero. Since 2007 we’ve seen continued losses of volume whilst the area has seemed to stabilise somewhat, by this I mean that 2007 didn’t unleash a succession of crashes as might be expected using the simple reasoning of ice albedo feedback. What has happened is a continuation of loss of older thicker ice, with another apparent stabilisation, this time in terms of ice up to four years old.
That there haven’t been further crashes is important because it is telling us something about the processes at play, and ice-albedo is not the only player. In recent years there have been prominent Autumn and Winter near surface warming anomalies (NCEP/NCAR) which Screen and Simmonds interpret as newly open ocean heating the atmosphere. The same process of heat loss is seen in the Tietsche et al modelling study in which ice recovers after its specified removal from the modelled ocean, in that model study the heat loss leads to a recovery of the ice to its equilibrum level. Although as Kevin O’Neill has pointed out on this thread in the real world the reported volume loss of PIOMAS is at odds with Tietsche et al.
Another factor of note is that there is currently vigorous growth of FY ice during the winter. To attain a virtually sea-ice free state substantial amounts of this FY ice would need to melt out completely. I’ve yet to read anything that explains how this might happen. The nearest I’ve seen to evidence of such a process are the Spring melts of the last two years as revealed by a simple examination of the PIOMAS volume data. But I don’t know what the process is, and it’s yet to be seen whether there will be a repeat this year. If you need references for the papers I reference please ask and I’ll give better citations.
So we have a situation in which the more the ice recedes the more of the energy gained during the summer is lost and the more rapidly thin ice grows in the open water. So I’m not persuaded by the “rapid camp’s” arguments, the only thing that’s given me concern you might be right are the massive Spring volume losses of the last two years. If this repeats this year then I’ll review my position, but at present I think a September extent of below 1M km^2 is likely in the latter part of the next decade, and probably won’t become a regular event until into the 2030s.
All this said, I dislike your organisation’s presentation, think you overstate certainty, and that you don’t use sources correctly. On the latter point on this page you quote Stroeve et al (PDF).
However you do not include the closing paragraph of that paper which follows the line you quoted:
I don’t see this as a negligible caveat, and from my reading of that paper I don’t think their idea of ‘rapid change’ is the same as yours.
Chris Reynolds says
I drew a blank with Catlin Arctic Survey too. You’ll find some links to sea ice thickness data on my blog’s most recent ‘Miscellanea’ post. If you want a spreadsheet with Candain Arctic Archipelago thicknesses just ask there and I’ll post my email address for you. Thanks for the admirable searching, as always, the Haas paper was new to me.
T. Marvell says
I find that artic ice exent is dominated by northern hemisphere land temperatures 3 to 12 months previously, with the greatest impact for temperatures 5-10 months earlier. If this is correct, than sea ice will be much greater this summer than in recent years because northern hemisphere temperature anomalies have been low this winter.
Chris, in my amateurish way, I have wondered whether there may be a bit of a slowdown in the rate of ice loss as the oldest ice melts. My reasoning was that, iirc, black carbon has played an important role in the ice loss by changing albedo. If the old ice melts year after year in the way large banks of ice melt in the spring around here–without significant runoff from the surface–you get darker and darker surfaces as the melt proceeds. This process should accelerate over time, with years-old soot coming to the surface and making it ever darker and more heat absorbent, until all the ice is gone.
But new ice would only have this years soot burden to alter its albedo, so would perhaps be ably to resist this accelerating albedo-shift-driven runaway melting.
But since this is just my own armchair theory, I would be happy to have holes poked in it.
On the losses over the last two springs, has that mostly been through physical transport through the Fram Straight?
(reCaptcha: fastruag duobus–sometimes I just love the absurdity of these strings of letters)
T. Marvell says
willi (#277) – apparently soot has little impact on ice melt. See Post #28.
T. Marvell @276
I did a quick plot of Sept Arctic Sea Ice Extent onto NCDC NH land temp anomalies & didn’t see anything that would encourage me to use such temperatures to predict this summer’s ice extent.
Looking at warm episodes, 1990, 1995 & 2007 appear to conform with what you say having low ice. But this was not the case for 1981, 1998, 1999, 2000 or 2004.
Also 2008 & 2010 were low on ice but not warm.
The opposite (cold temps to predict high ice) worked better but only by lenient treatment of time lags, shortened in 1992 & 1996 while lengthened in 1994 & 2001. 1985 was also cold but with the ice nothing unusual. 1980 was high ice but normal temperatures.
So in 16 years, only 7 fit your assertion & 4 of those requiring time lag adjustments. That makes me rather sceptical about your predictor of summer sea ice.
Chris Reynolds says
Despite its popularity in public discussions, I’m not convinced black carbon (BC) is a major factor. For example, see the inline reply by Dr Schweiger on post #28 above. From the paper he links to:
Suspecting inceased Fram Strait export having a role in the volume losses seems reaonable. There is a recent paper that gives Fram Strait export figures for the period in 2010, I don’t have figures for Spring 2011. The paper concerned is “Recent wind driven high sea ice export in the Fram Strait contributes to Arctic sea ice decline” by Smedsrud et al.
Figures 4 and 7 show the recent activity and longer term trend respectively, in the context of recent years Spring 2010 doesn’t stand out. Although note that those are in area, not volume. I’ve not got time to read that paper in its entirety but it doesn’t refer to volume.
It’s worth bearing in mind that despite the increase in area export, volume export through Fram shows no trend, e.g. Spreen et al:
This is because the ice being exported has thinned even as area export has increased.
Could you give more detail? – I presume you’ve used sea-ice area / extent, that’s OK I’ve got them daily back to 1979. But what temperature dataset have you used?
wayne davidson says
Catlin scientists are congenial and very friendly. Try contacting them by E-mail.
Predicting sea ice extent is easy if you can mentally calculate wind variations, momentum, sea currents, multi year ice compression ratios, tidal synergy with weather patterns, the AO, the temperature of ice sea water and air, how cloudy it will be, salinity, pycnocline convection rates, sea surface to air interface, CO2 exchange, ice thickness distributions…..
Piece of cake!, morceau de chocolat..
I predict greater surface salinity, much earlier melt, great adiabatic events at sea surface to air interface, wide early thousand lead expansion events coinciding with at start clear than very cloudy air, a surprising near ice death experience at the North Pole because the North Greenland subduction zone is already very fluid. All time minimum to be exceeded only to be stopped further by clouds. By the way the howitzer in prediction, sun disk expansion comparisons, projects a warmer temperature gain than summer winter 2010.
But sea ice has so many variables, extent is not so easily foreseen. But unlike geologist shy about trying to forecast earthquakes, if we don’t try to do so, we will never understand the subject at hand.
Jim Larsen says
281 Wayne D enters one of the first of this season’s entries for Sept minimum with: “All time minimum to be exceeded only to be stopped further by clouds.”
This sounds like a bold guess in light of the OP. Systems tend to return to the mean and sea ice has a short memory.
If I’m reading the OP correctly, a new minimum is not likely to be set for at least a few years, and it could be decades.
But I wouldn’t bet money against a new record. It’s interesting how the mind works. I’m being “skeptical”.
T. Marvell says
I have predicted that artic sea ice extent this summer will increase greatly because sea ice extent is greatly affected by past land temperatures, which have been unusally low since November (My 279, and responses 279 & 280).
I use NOAA temperature data:
So far I have only used ice extent data. I have not been able to get monthly data for ice volume. Can anybody help me get it? Daily figures don’t help.
The impact is about the same when using the data series for world-wide land temperature & for northern land temperature (most land is in the north). The impact is lagged. That is, temperatures today affect ice some 3 through 12 months into the future, with the largest impact 5 to 10 months. To estimate the impact, one has to factor in the temmperatures for all these 10 or 6 months.
This conclusion comes from regressing ice extent (monthly anomalies) on lags of land temperature (monthly anomolies), with the variables differenced. The results are very strong statistically (prob. lt .00001). Global and hemispheric ocean temperatures have no discernable impact on artic ice extent, other than through its impact on land temperature.
Land temperature account for about 60% of the variance in sea ice extent, which is an extremely high figure in the world of statistic.
The impact is much stronger for summer ice extent than winter ice extent (that is, summer land temperatures don’t have a big impact on winter ice).
By forecasting more sea ice this summer, I am going against the grain; the posts here mainly assume a downard trend (though of course one year does not change in trend). But, by now the sea ice increase is old news. Artic ice extent through yesterday has climbed greatly compared to recent years: See: http://nsidc.org/data/seaice_index/images/daily_images/N_stddev_timeseries.png
And historically the amount of sea ice at this time of year has foreshadowed the summer extent.
So, in saying that sea ice will be much more extensive than in recent years, I’m probably just stating the obvious. But I am giving a reason for the change, which might not have been evident.
I am continuing to research this topic. The next step is pin down the magnitude of the relationship, which would permit a specific prediction of summer sea ice exeent for months April to September.
T. Marvell @283
We appear to be using the same data but then obtaining entirely different results.
My comment @279 was based on examining this graph here. A plot of red squares & blue triangles against each other x/y does give a correlation but it entirely disappears if the ice & temp data is detrended. So I remain more than a little mystified by your assertion of a strong correlation.
Kevin McKinney says
#283–Keep working, Tom–I’m sure you’ll find it a useful and instructive exercise, as perhaps we will here, as well.
That said, I suspect you are a tad over-confident at present. (No offense intended.)
Kevin McKinney says
P.S.–“Tom” was a Freudian slip; “T” is what I should have written. And yes, I read all the Harry Potter books…
wayne davidson says
282, Jim , is OK to be skeptical, but its irrelevant, what will happen will happen. How it happens is what matters. T. Marvell is basing his estimate on the very same ice we look at. Which one looks at the evidence better than the is important, only to be settled come late September. Cloud coverage is the biggest player in this case, without will make a great melt, or if not, totally save sea ice extent from collapsing further. Closely followed by temperatures, salinity, ice thickness and so down a rather large list of variables.
T. Marvel didn’t monitor or forgot to remember that during entire winter Cyclones whisked over Spitzbergen week after week towards the Arctic ocean. Thickness from the start should be less. Arctic first year ice seems not recovering to average, this means that any melt season may exceed 2007 minima. given that 2007 like multiple weather features are repeated, it is certain that the Pole will have some vast open water near by.
Salinity is also a huge factor, there is a significant salt content difference in First year ice melting and old multi year ice. This favors a greater melt or similar to 2011. What was astounding about 2007 was the quick melt of Multi year ice, since 2012 has a whole lot more 1st year ice, any weather conditions approaching 2007 will lead to my estimated result.
As this article suggests, the melts are not linear, ice melt sensitivities are huge, but we have already a very good idea of what will follow, the chances of T. Marvel being right not so good. But I worry about whether I am missing something obvious, if I did my projection was based on errors, and the extent will be more normal. But I never ever mess with sun disk projections, it will be hot, something which melts ice.
Chris Reynolds says
#283 T Marvell,
Re monthly volume, why not just use 12 X 30 day periods and ditch the 5 remaining days? It’ll have a negligible effect and will give as near as damn it monthly figures. In any case PIOMAS volume has every year 365 days, so I’ve adjusted area and extent observations to match that format, I’ve not gone the other way. So the spreadsheet I use (5Mb – Excel – every day from 1979 to 2012 area/extent/volume daily anomalies and subsequent pages such as thickness) would need massive modification to address this, and I’m sorry but I can’t spare the time right now.
I suspect MARodger has hit the nail on the head, if you don’t detrend two series with trends you’ll get large correlations that aren’t really as significant as statisitics based on a staionary series will suggest. Before looking for correlations, lagging, etc you really need to detrend. If the corelation exists in reality it’ll still be there.
T. Marvell says
Concerning my post 283 on the connection between land temperature and ice extent – Probably what is going on is that colder winters create more ice, which takes longer to melt. Seems obvious.
There are bound to many other factors affecting ice extent (see Davidson 287), but that does not rule out a large role for land temperature.
MARodger (#284) Graphs are hard to interpret, but to me there is a connection more often than not. The key here is changes, whether there is a connection between declining temperatures and more sea ice, which is hard to see in a graph. Since there are other factors involved, one would not expect a clear match on the graphs. Also, I find that the major impact is spread over 6 years, beween 5 and 10 years previously. You used a 3 month rolling average displaced 7 years. It would be better to use a 6 month rolling average, displaced 9 or 10 years.
Reynolds (288) – As you said, estimating monthly sea ice volume would be a lot of work. PIOMAS should do it if they want their work to be useful. Most climate data are published in a monthly format.
About detrending the time series – I do that. The variables are first-differenced (current month value minus the prior month variable). The variables are I(1), so regression in levels would make no sense. Also there would be a huge collinearity problem since I enter a lot of lags. But your raising this issue makes me question my statement earlier that temperature explains 60% of the variance in sea ice. That calculation was based on levels data, and is thus incorrect.
T. Marvell says
At the suggestion of Reynolds (288) I constructed a montly ice volume data series from the PIOMAS data set. The results are very similar to those when using the ice extent data set. That is lower land temperatures are followed by more ice volume several months later.
The ice volume data series is a bit odd. I use anomalies (e.g.,ice volume less the montly mean for the volume over 1980-2011) and then take differences (substrct the prior period anomaly from the present anomaly). The regressions contain lagged dependent variables. With nearly all variables I’ve ever used, including ice extent and temperatures, the lagged dependent variables have large negative coefficients, since changes one period tend to lead to “movement to the mean” in the next period. But when using ice volume as the dependent variable, there are large positive coefficients on the lagged dependent variable. It looks like the variable is constructed by starting with prior period values and adjusing them (this could be for only one element of the ice volume estimation), rather than taking independent measurements each time. This can cause data estimates to snowball, and it might account for the apparent accellerating decline of the ice volume time series.
Jim Larsen says
287 Wayne, actually, I used the word “skeptical” in the bad way. I too have no mental qualms with the idea of a new record. However, the OP shows that the best no-weather-known estimate for minimum ice volume for 2012 is around 12k, or THREE TIMES the volume of 2011. Add in a bit of the low-memory-of previous-years and that might drop to 8k, or double 2011’s volume. In any case, the odds of setting a new record are abysmally low, unless the OP is completely wrong.
So, being a non-scientist, I’m left with a conundrum. Either the OP is FOS, or we’re likely in for a HUGE increase in sea ice volume for September. Thus, if you’re right, either the OP is garbage or we’re in for some seriously high sigma weather.
Or, as is often the case, I’m confused. :-)
Chris Reynolds says
I may well look into this when I can make the time. You said: “Probably what is going on is that colder winters create more ice, which takes longer to melt.” Be very cautious with this, in recent Winters mid latitudes were colder than average, but the Arctic was warmer. e.g. GISS Maps.
Converting PIOMAS daily to monthly wouldn’t be that difficult, it’s just it’s not a case of copy down and copy/paste to deal with the mass of data. Once in excel it’s not a major operation to copy down overlapping 30 day averages for all PIOMAS days (apart from the first 29). Use formula of the form ‘=A1&”\”&B1’ to create an index from PIOMAS Year and Day, i.e. Year in An and day in Bn that formula giving an index of the form 1980\239 (forward slash doesn’t work in vlookup). Then use Vlookup() to grab the relevant averages from the index / volume table.
Vlookup gets a figure in a range that’s a number of columns right of the index. All you need to do is sort out the dates you want to grab. Again you’d use the formula type ‘=A1&”\”&B1’, you’d feed that with 12 instances of the year, and the series 1, 31, 61, 91, etc up to 361 and use the formula to make your 1979 values. Then copy the days you want down by referencing 12 rows up, not fill down but e.g. b14 = b2 – that’s how to copy a repeating series down in excel.
Actually now I put it like this, it wouldn’t take me long. If you really want me to do a spreadsheet, just bother me for it at my blog.
ld cooke says
A bit late to the party; however, I wanted to chime in that, IMHO, this is one of the better Posts here.
To actually, characterize the model interdependency or lack there of, certainly helps point out to many readers, the inter-relationships of data as a function of the various model types. To explain that closely coupled models point more towards the median, while loosely coupled models are free to represent great variation is wonderful. It is almost like for the first time we can see the median and up to three sigma of the potentials that can occur at any given time.
IMHO, an excellent job all around…
t marvell says
Thanks for your interest.
Although temperature in the artic has risen over the decades more than elsewhere, this winter has seen a reversal of that trend, due to extreme cold in the Western part and the Baring Straits.
Another reason why land temperature might have a lagged impact on sea ice is river runoff. It will be interesting to see whether sea ice above Siberia stays around this summer, even though the temperatures there are now comparatively warm. There are some large rivers dumping into the artic from Siberia, and Siberia has been very cold this winter, so the river runoff is less likely to melt the ice.
I have converted the PIOMAS daily data into monthly data, using SAS. If you, or anybody, wants a copy let me know. I could not find an email address on your blog. You might want to put the monthly data there.
As I said in my post 290, the PIOMAS series is odd in that lagged dependnet variables, using differenced data, have positive coefficients. That virtually never happens in a real data series, one with separate observations for each period. I replicated the analysis with monthly data (in #290) with the daily data. The lagged dependent variable has a positive coefficient of .69, t ratio = 76.3. That is, the figure for each day is just a tweaking of the figure for the day before. In that case, an initial modeling error can easily get magnified over time. All those attempts to forecast using this series, therefore, are meaningless.
Dan H. says
Going one step further with T. Marvels data, my prediction for this summer’s Arctic sea ice extent will mirror that of 2010. Although slightly off from 2009 in timing, 2011 mirrors 2009 in temperature. The sea ice has mirrored the 2009-2010 winter so far, with a late start to the spring thaw. My prediction is for a similar late minimum (very late Sept.), with sea ice extent reaching a similar minimum to 2010.
Peter Ellis says
That is lower land temperatures are followed by more ice volume several months later.
Congratulations on discovering seasons. You are now approximately as advanced as grass.
The ice volume data series is a bit odd. I use anomalies (e.g.,ice volume less the montly mean for the volume over 1980-2011) and then take differences (substrct the prior period anomaly from the present anomaly).
A remarkably pointless process. (A – avg) – (B – avg) = A – B.
Why are you buggering around calculating anomalies and then throwing them away again? All you’re effectively doing is calculating the difference between time points A and B.
The regressions contain lagged dependent variables. With nearly all variables I’ve ever used, including ice extent and temperatures, the lagged dependent variables have large negative coefficients, since changes one period tend to lead to “movement to the mean” in the next period.
This is so incoherently described as to be meaningless. I think it’s just another way of saying that the volume series has an accelerating downward trend, while the other measures (extent, area) are not easily distinguishable from a linear trend – at least over the (unspecified) period you’re using. This is blindingly obvious from a graph, but there are much better ways of testing and describing it.
Brian Dodge says
Could the Ice models be forced to an “ice free” state at the ides of March, then run backwards to see what the conditions would have to be(IMHO, primarily ocean surface temperature and profile with depth) at the end of the previous September to give this result when run forward?
t marvell says
Dan H. – we’ll see who is right. The daily graph yesterday showed 2012 ice extent dropping below the 1979-2000 average, so maybe you’ll be right. It would be interesting if anybody else cared to make a prediction. How about a contest to see who best guesses the summer ice extent?
I agree that the results of my research are pretty hum-drum (see post 289), but I haven’t seen it stated before, and obvious findings tend to become obvious only after they are pointed out.
About taking differences (current period figures less prior period figures) of anomalies: the anomalies are the value less the monthly mean (i.e., the mean for the particular month over the years, in this case 32 full years), as is the usual practice with climate data (most notably temperature). The two “avg”‘s in your equation are different figures, so A-B on the right hand side is not right.
About lagged dependent variables (which should generally be used in regressions if significant): a simplified regression is x = ax(t-1) + bx(t-2), where x(t-1) and x(t-2) are lagged values of x. When variables are in levels (not differenced) coefficients a and b are positive, since the value in one period is related to that in the prior period. But when the variables are differenced, the a and b are negative, since changes in one direction tend to be counteracted by changes later in the opposite direction (for example, if a short-term factor caused x to increase in time 1 and went away before time 2, then x would drop in time 2, all else equal). As I said almost every variable acts that way. The only exceptions are very regular variables, such as USA population, where a and b tend to be zero.
The fact that ice volume has a large positive coefficient on the lagged dependent variable (with differenced variables) means that something screwy is going on. A change in ice volume one day leads to a change in the same direction in the next day. There is something constantly pushing the variable to move in the same direction, either up or down (although that need not show up in a graph because other factors also affect change). I see no alternative but that the model used for calculating ice volume has a built-in trend mechanism, which may be non-linear. I have no idea what it is. In any event, the ice volume series clearly is not based solely on observations, as is the ice extent series.
The next implication is that those who try to predict ice volume using past ice volume data are largely picking up the artifical trend that is built into the data. As I said, such predictions are meaningless.
Hank Roberts says
> Siberia has been very cold this winter,
> so the river runoff is less likely to melt the ice.
What’s been published correlating summer and winter Siberian temperature with river runoff temperature and the following seasons’ Arctic ice melt? I’ve seen more about air temp. and ocean current temp, but not river temperatures. Looks like ground temperatures don’t change very fast:
“… the coldest Siberian region, and the basin of the Yana River has the lowest temperatures of all, with permafrost reaching 1,493 metres (4,898 ft).”
AMEG’s blog’s quite busy, and the UK Parlaiment is having hearings.
This is new to me; anyone watching this?
“Supplementary written evidence submitted by Professor Peter Wadhams to the Environmental Audit Committee (EAC)
“I am writing in response to information provided recently by Professor Julia Slingo OBE, Chief Scientist, Meteorological Office, firstly in the report ‘Possibility and Impact of Rapid Climate Change in the Arctic’ to the Environmental Audit Committee and subsequently in answering questions from the Committee on Wednesday 14 March 2012. In the responses, the Meteorological Office refers to an earlier presentation to the Committee by myself, made on 21 February 2012.
The following comments are based on the uncorrected transcript of Professor Slingo’s presentation to the EAC, 14 March 2012 session, as at:
1. Speed of ice loss
In response to questions from the Chair, Prof. Slingo ruled out an ice-free summer by as early as 2015. Furthermore, Prof. Slingo rejected data which shows a decline in Arctic sea ice volume of 75% and also rejected the possibility that further decreases may cause an immediate collapse of ice cover.
The data that Prof. Slingo rejected are part of PIOMAS, which is held in high regard, not only by me, but also by many experts in the field. From my position of somebody who has studied the Arctic for many years and has been actively participating in submarine measurements of the Arctic ice thickness since 1976, it seems extraordinary to me that for Prof. Slingo can effectively rule out these PIOMAS data in her consideration of the evidence for decreasing ice volume, when one considers the vast effort and diligence that has been invested ….”
AccuWeather is about weather, reporting individual city temperatures, but talks about a large area of Siberia here:
“Western Siberia Temperature Flip in April
Apr 28, 2010; 10:51 AM ET
A dramatic flip from severe, even record-setting, cold to unusual warmth has happened over that part of western Asia centered upon the west of Siberian Russia….
… the cold broke in March and, in mid-April, unusual warmth blossomed over a vast swath at the heart of Eurasia. In Omsk, this warming culminated on Tuesday in a high of 30.0 C, or 86 F–a hot day even in July. For April as a whole, the mean temperature as of Tuesday rose 5.7 C/10.2 F above normal….”
wayne davidson says
Seldom do we look at Russian products, on my blog on sea ice I show a remarkably good ice map,
which essentially makes the case for a wide open North Pole this melt. The famous continent to continent multi year ice bridge of 2007 melt season, holding up despite the onslaught, is gone….