Climate Science Glossary

Term Lookup

Enter a term in the search box to find its definition.

Settings

Use the controls in the far right panel to increase or decrease the number of terms automatically displayed (or to completely turn that feature off).

Term Lookup

Settings


All IPCC definitions taken from Climate Change 2007: The Physical Science Basis. Working Group I Contribution to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Annex I, Glossary, pp. 941-954. Cambridge University Press.

Home Arguments Software Resources Comments The Consensus Project Translations About Support

Twitter Facebook YouTube Mastodon MeWe

RSS Posts RSS Comments Email Subscribe


Climate's changed before
It's the sun
It's not bad
There is no consensus
It's cooling
Models are unreliable
Temp record is unreliable
Animals and plants can adapt
It hasn't warmed since 1998
Antarctica is gaining ice
View All Arguments...



Username
Password
New? Register here
Forgot your password?

Latest Posts

Archives

A basic overview of Antarctic ice

Posted on 29 November 2010 by robert way

Skeptic arguments that Antarctica is gaining ice frequently hinge on an error of omission, namely ignoring the difference between land ice and sea ice.

In glaciology and particularly with respect to Antarctic ice, not all things are created equal. Let us consider the following differences. Antarctic land ice is the ice which has accumulated over thousands of years on the Antarctica landmass itself through snowfall. This land ice therefore is actually stored ocean water that once fell as precipitation. Sea ice in Antarctica is quite different as it is generally considered to be ice which forms in salt water primarily during the winter months.

In Antarctica, sea ice grows quite extensively during winter but nearly completely melts away during the summer (Figure 1). That is where the important difference between antarctic and arctic sea ice exists. Arctic sea ice lasts all the year round, there are increases during the winter months and decreases during the summer months but an ice cover does in fact remain in the North which includes quite a bit of ice from previous years (Figure 1). Essentially Arctic sea ice is more important for the earth's energy balance because when it melts, more sunlight is absorbed by the oceans whereas Antarctic sea ice normally melts each summer leaving the earth's energy balance largely unchanged.

Figure 1: Coverage of sea ice in both the Arctic (Top) and Antarctica (Bottom) for both summer minimums and winter maximums
Source: National Snow and Ice Data Center

One must also be careful how you interpret trends in Antarctic sea ice. Currently this ice is increasing and has been for years but is this the smoking gun against climate change? Not quite. Antarctic sea ice is gaining because of many different reasons but the most accepted recent explanations are listed below:

i) Ozone levels over Antarctica have dropped causing stratospheric cooling and increasing winds which lead to more areas of open water that can be frozen (Gillet 2003, Thompson 2002, Turner 2009).

and

ii) The Southern Ocean is freshening because of increased rain, glacial run-off and snowfall. This changes the composition of the different layers in the ocean there causing less mixing between warm and cold layers and thus less melted sea ice (Zhang 2007).

All the sea ice talk aside, it is quite clear that really when it comes to Antarctic ice, sea ice is not the most important thing to measure. In Antarctica, the most important ice mass is the land ice sitting on the West Antarctic Ice Sheet and the East Antarctic Ice Sheet.

Therefore, how is Antarctic Land-ice doing?

Figure 2: Estimates of Total Antarctic Land Ice Changes and approximate sea level contributions using many different measurement techniques. Adapted from The Copenhagen Diagnosis. (CH= Chen et al. 2006, WH= Wingham et al. 2006, R= Rignot et al. 2008b, CZ= Cazenave et al. 2009 and V=Velicogna 2009)

Estimates of recent changes in Antarctic land ice (Figure 2) range from losing 100 Gt/year to over 300 Gt/year. Because 360 Gt/year represents an annual sea level rise of 1 mm/year, recent estimates indicate a contribution of between 0.27 mm/year and 0.83 mm/year coming from Antarctica. There is of course uncertainty in the estimations methods but multiple different types of measurement techniques (explained here) all show the same thing, Antarctica is losing land ice as a whole, and these losses are accelerating quickly.

NOTE: This post is the Basic Version (written by Robert Way) of the skeptic argument "Antarctica is gaining ice". This post means we now have basic rebuttals for all of the top ten skeptic arguments. More impressively, the prolific Skeptical Science authors have written basic rebuttals to 65 skeptic arguments in total. Go team!

Lastly, just a reminder that Robert Way has also written a comprehensive post on the measurement of land ice loss from Antarctica which puts the recent study by Wu et al into perspective (anyone wishing to cite the Wu study is recommended to read Robert's post first).

0 0

Printable Version  |  Link to this page

Comments

Prev  1  2  3  Next

Comments 51 to 100 out of 108:

  1. For those wishing to understand the uncertainty involved in this sort of work should have a look at this link. It's from an EU funded collaborative effert to constrain GIA estimates. The example they use is quite eye-openning. "For example, a published total mass trend for Antarctica from GRACE is 39±14 km3/yr but with an estimated GIA contribution of 192±79 km3/yr." To me this says GIA uncertainty can be larger than estimates of ice loss.
    0 0
  2. Re: Albatross (49) Actually I thought the whole comment was rhetorical. The part I responded to was something that continually bothers me about "skeptics": the continual nay-saying, unfettered by the need to be constrained by the physical world. The Yooper
    0 0
  3. Paul, I'm not aware of any graphs showing the albedo shifts due to changing sea ice area. However, graphs of sea ice area itself gives part of the picture; Arctic Antarctic Global If you look at the 'zero line' on the anomalies you will see that further back in time the Arctic and global anomalies were predominately above the line... while closer to the present they are predominately below the line (seldom going above it). In short, the ice area in the Arctic and globally has decreased. The Antarctic shows anomalies frequently on both sides of the line throughout, but has a noticeable uptick in recent years... but not sufficient to offset the decrease in Arctic ice area and not as significant for energy absorption due to the winter season it is happening in.
    0 0
  4. fydijkstra wrote : "So, the increase of Antarctic sea ice is also antropogenic? This looks like immunization of the AGW-theory. It does not matter which new evidence is found, it always supports the theory. When a theory reaches this stage, there is no need for further research. The billions of dollars can better be used to build dikes in Pakistan." Sounds very similar to creationist views of evolution : So they accept quill knobs as being evidence of feathers when it fits with their evolutionary paradigm, but they reject such reasoning when it overturns their theories. Inconsistent Reasoning Governs Evolutionary Interpretations of Feathered Dinosaurs
    0 0
  5. I wonder if anybody would like to comment on Velicogna 2009? It seems particularly important given that it is one of the most recent and the largest Antarctic estimates in Fig2? I think Robert is picking some of the numbers in his final paragrqaph from it as well. I wonder if anybody could explain the reasoning behind the way she divides up the data? Here is the Antarctic data graph from the original paper. The data runs from 2002-2009. This is how she describes the early and later rates of ice mass loss in the abstract. “In Antarctica the mass loss increased from 104 Gt/yr in 2002–2006 to 246 Gt/yr in 2006–2009, i.e., an acceleration of 26 ± 14 Gt/yr2 in 2002–2009.” Can anybody give me a logic reason why you would separate the time period the way she did i.e. 2002-2006 and 2006-2009? Especially when the Greenland data for the same period is separated to derive rates for 2002-2003 and 2007-2009. I think I have an explanation. Look at Velicogna fig2 above. Notice the slow down around 2006, you can see it best in the smoothed data (red x's)? You can actually play around with the GRACE data here. I downloaded the data and had a look at the trends and annual changes. In fact if you look at the estimates for ice mass loss on an annual basis then the early years (2003-2005) don’t look so different to the later years (2007-2009). The only way you can generate such large differences in the rate of ice mass loss in Antarctica is by estimating the trends around the year 2006. Below is the annualised data I generated from GRACE for Antarctica from the POET website (link above). The annualised data is simpler than Velicogna, but you can still see the slow down (actually reverse) in 2006. The units are in sea level equivalents (cm) anomalies and GIA and corrections aren’t removed but this doesn’t matter as it’s the trends that are important and as Velicogna says most corrections are the same throughout the time series. The only varying correction is for atmospherics and this is very small. Have a play with the data yourself and let me know whether it's the 2006 data only that is giving such large differences in Velicogna's early and late ice mass loss rates.
    0 0
  6. Humanity @55, You seem to be trying very hard to make something out of nothing, while at the same time making insinuations about poor science or something nefarious going on. The increase in loss (i.e., slope), over this short time window is indeed there-- but this is part of a longer record showing acceleration in Fig. 2 of Robert's post. I recommend that you redo your analysis, this time using all the data and then finding a model that provides the best fit (linear vs. quadratic). In other words reproduce their results. The answer to your question is the green line in Velicogna's Fig. 2 which you pasted in above. Velicogna's results (accelerated loss of ice)have been confirmed by Chen et al. (2009), Bamber and Riva (2010). Also see work by Rignot et al. (2008)". Again, Fig. 2 in Robert's article provides a nice summary of the bigger picture. Really, the lengths some people will go to to convince themselves that there is not a problem.
    0 0
  7. @CBDunkerson Thanks for the graph links - very consistent with article. I was merely thinking about showing rather than just telling about the albedo impacts. Come to think of it, it would be nice to have superimposed plots of ice-mass, ice-extent (area) and reflected energy (albedo) altogether - with separations of supported and unsupported ice even if the numbers had to be interpolated or estimated, just to illustrate all of these distinctions in a single picture.
    0 0
  8. Hi Rob @50, The PDF linked in here may be just what we have been looking for. Here it is.
    0 0
  9. Albatross: Really, the lengths some people will go to to convince themselves that there is not a problem. Or to be more precise, that the problem is with most of the world's scientists. I think the conspiratorial worldview and epistemological nihilism that informs the thinking of most "skeptics" is at least as bleak and alarming as AGW. It's kind of amazing that they find it within themselves to accuse anyone else of pessimism.
    0 0
  10. 56 Albatross Albatross thanks for recognizing I tried hard with that analysis. The one thing I didn't do is insinuate anything or suggest anything nefarious. I did think about doing that but tried to stick with John's comment policy and stick to a critique of the authors results. I think we all know that a difficulty with describing rates of change from very short data sets is that variability can lead to very different trends depending on where you start and stop your analysis. I don't really know why I'm describing this to you because I know you have the skill set to appreciate this very basic point already. Please ignore my data presentation, it was only intended to illustrate my points but know I see it's just a distraction. Look at the Velicogna graph (red x's) I show in #55. Tell me around 2006 there isn't a short term slowdown and reverse in the trend. Tell me that if you start and end analysis then that this won't affect the value of your rates. Please tell me the slope of the data is signficantly different before (2003-2005) and after (2007-2009) this 2006 blip. This can all be done by just eyeballing Velicogna's graph, remember she is suggesting rates of ice mass loss have more than doubled over this time period. Do you see that? The conclusions in the Velicogna paper should stand on their on feet irrespective of what other authors have published, I'm really happy to look at the quality of those other results but I relly think if you are going to defend what Velicogna writes then do it based on the merits of her methods not on what others have done. Look I would turn your own comment back onto you in relation to looking at the quality of Velicogna's interpretation of this data. "Really, the lengths some people will go to to convince themselves that there is not a problem."
    0 0
  11. 59 Phila I think if I made those sort of accusations about climate scientists that you make about "skeptics" my post would be deleted. I see no comspiracy and don't suffer from nihilism. What I see is data and interpretations that are put forward by Velicogna for critical analysis. That's the way normal science works. Rather than trying to get to the heart of "skeptics" mental state why not try looking at the data critically. You can make a much better defence of the data by doing that.
    0 0
  12. Your argument is strange. When one looks at the graphs you provide a good amount of sea ice remains in the southern summer, yet your argument turns on sea ice not melting in the summer thus providing no feedback. Sea ice not melting in the southern summer is patently false and can be seen clearly in your own graphics. Now perhaps when and if the antarctic actually does melt to the edge of the continent a month or two earlier than normal your argument will carry some weight. But for the moment it doesn't seem to explain what it purports to explain is some alleged oversight in skeptic arguments that Antarctica is gaining ice. Even if we limit the argument to sea ice we can see that indeed it is gaining sea ice.
    0 0
  13. Re Billhunter (62) What part of: "In Antarctica, sea ice grows quite extensively during winter but nearly completely melts away during the summer" Do you not understand? The point of the post is that focusing on Antarctic Sea Ice (ASI) is a strawman argument. ASI offers little contribution to the global energy budget, unlike changes in Arctic Sea Ice cover.
    "Even if we limit the argument to sea ice we can see that indeed it is gaining sea ice. "
    Which then melts away come Antarctic summer. The Yooper
    0 0
  14. HR says: "The conclusions in the Velicogna paper should stand on their on feet irrespective of what other authors have published." That only is to be understood in terms of its internal logic and how results are supported by data used. It does not go beyond that. When considering the field of study, the conclusions of a paper should always be weighed with what else has been published before and after on the subject. Why should it not?
    0 0
  15. Re #62 & 63, I posted comments similar to what Bill wrote here (see #26 and 38). My comments were ignored. :( DB(Yooper) -- the fact that Antarctic sea ice melts back to the continent by the end of the austral summer does not contradict the fact that there is sea ice there during the austral summer solstice. Cryosphere today says the current sea ice area there is 10.9 million sq km. There will still be millions on December 21 and even January 21. There could be as much as a half million sq km more ice than the 1979-2008 average over that time period (currently 0.3 million sq km). This anomaly in reflective area may not be as great as is observed in the Arctic during the boreal summer solstice (can someone check?), but we shouldn't pretend that it doesn't exist.
    0 0
  16. HumanityRules, Perhaps the reason Velicogna uses 2006 as a start date is because the rate of ice loss after 2006 was greater and because Rignot et al. 2008 a,b measured accelerations in glaciers for sure between 2006 and 2007. I also noticed that you have yet to address the issue that multiple methods are supporting the same answer. It is because of the aforementioned uncertainties that we should look for independent measurements in agreement. That is the crux of science. We acknowledge there is the potential for error in analyis but when we see it repeated with similar conclusions and different methods then we conclude that there is a foundation upon which claims can be built.
    0 0
  17. 64 Philippe Chantreau My point hasn't got as far as the wider perspective on the subject. It's specifically about the way the data is used in Velicogna 2009. If data is used badly in making an argument there is no defense for that work in the fact that it is in agreement with other studies. I don't disagree with your general argument that there is strength in independant well crafted analysis supporting a particular interpretation. My question is whether we have that in the first place. I'm still looking for a bold defence of V09's method of data analysis given that I recognise important variablity in the data. Of the few people who've responded to my post nobody has even acknowledged whether they see that short-lived upturn around 2006. Philippe please have a go at critiquing my specific criticisms of the data, I'm happy to discuss the merits of mutaully supporting evidence after I've got over the hurdle of whether the V09 estimates are meaningful.
    0 0
  18. Re: Steve L (26, 28, 65) Sorry I missed your earlier questions! You are correct in that austral summer solstice is important (due to the sun being the highest above the horizon at that point), but remember that summer minimum ice extent is also important, as even low-angle sun hitting open water late in the melt season still gets absorbed, contributing to warming. As you noted, this then contributes in the Arctic to an extension of the fall melt season (the fact that melt there is now actually lasting into fall is remarkable). In fact, Arctic area and extent have declined in every month: and ice-edge latitude has moved poleward as well: The money shot for the Antarctic is this: poleward retreat of sea ice goes only so far. The continent, with its lofty elevation and citadel of cold (buttressed by the ozone-hole-strengthened circumpolar vortex) is relatively immune for now to the summer suns' direct effects. The Arctic has no such reinforcements at its back. Latitudinal retreat will continue until the pole is reached, and summer ice known to millennia of mankind will be no more. Cryosat-2 data will provide definitive data on Arctic ice mass & will be the final (until we get something better) arbiter for ice data for both poles. Per Tamino's post, Antarctic summer ice minima increase is not statistically significant. When the Southern Ocean warms enough, then we will see an early and longer melt season in the Antarctic (with trends growing similar to those I showed you for the Arctic). Which is bad news for the WAIS in general and the PIG in specific... Thanks for reminding me about Tamino's post! The Yooper
    0 0
  19. #67: "nobody has even acknowledged whether they see that short-lived upturn around 2006." HR, I'll play. I see it! And I acknowledge it appears just as you describe: 'a short-lived upturn'. Unfortunately, your graph in #55 is in 'sea level equivalents (cm) anomalies', so there's some appling and oranging going on. However, this blip does not eclipse the downward trend, which is present no matter how you slice and dice these data. Trends, its all about trends. Agnostic at #43 made it quite clear that this is one complicated subject and we're trying to sum it up with one number. It's a bit like a conversation that someone might have over whether the stuff in their freezer is thawing evenly -- while their house burns down.
    0 0
  20. #69 muon Thanks for playing. I acknowledged later my data presentation was a distraction from V09 data but I don't think it really matters because as you say it's all about trends. So please stay with V09's original data. It's not just about whether there is a downward trend it's about whether there is an acceleration in that trend. I think V09 (along with Rignot 2008a) appear on the surface to be the strongest supporting data for that outlook based on the Copenhagen Diagnosis figure. So the blip matters here. Look I have to disagree with your final paragraph. Bad data analysis is bad data analysis. There's no running away from that, no matter how inconsequential you think it is.
    0 0
  21. HumanityRules: Rather than trying to get to the heart of "skeptics" mental state why not try looking at the data critically. The comment you objected to wasn't directed at you specifically. Sorry that wasn't clear. As for your other comment, I've been looking at the data critically since the late 1980s. Please don't make the mistake of assuming that people have not been critical until they've embraced whatever species of "skepticism" you favor.
    0 0
  22. HR, yes, at least we can agree that your analysis of the V09 data is a distraction.
    0 0
  23. #70: "Bad data analysis is ..." Here are some key points about the data analysis in Velicogna: For both the Greenland and Antarctica ice sheets, we found that Radj^2 is larger when we use a quadratic fit, i.e., the data are better modeled by a linear increase in mass loss than with a constant mass loss. To verify that the improvement obtained with the quadratic model is significant we used an F-test. The F-test show that the improvement obtained with the quadratic fit is statistical significant at a very high confidence level (99%). Note that if we use the unfiltered GRACE time series instead of the smoothed one, the Radj^2 values drop by 2% and 16% for Greenland and Antarctica, respectively. This illustrates the importance of removing the seasonal variability in the trend estimates. The improvement is much larger for Antarctica than for Greenland. So it doesn't sound 'bad'. But I guess I would accept HR's nitpick: Why wasn't the quadratic function used shown in the paper? That way the 'acceleration' could be determined directly. But here is some 'hedging': The Antarctic filtered data also suggest a slight change in trend around the end of year 2006. It appears that the long term variability could be described by two linear trends, one for the period 2002–2006 and the second during 2006–2009. ... We fitted two straight continuous lines through the data, i.e., connected in the middle. We find that the Radj^2 for the two lines regression model is 0.97, the same than for the quadratic model So to answer HR in #55: It appears that breaking the data into two segments is not arbitrary, nor is it necessarily 'bad'. My key take-away: GRACE supports a flat rejection of the skeptic claim that Antarctica is gaining ice.
    0 0
  24. Hi all, OK, so I have been to the NSIDC FTP site and downloaded some data for sea ice area over the Arctic and Antarctic for January and February (peak of Austral summer) and July through September (Boreal summer) for all years between 1979 and 2010. Daniel and Tamino have also looked at this, but since HR wants us all to be critical ;) Great post Daniel by the way. Some things to note off the bat-- nobody is dismissing the presence of Antarctic sea ice during the austral summer. What I think people have been unsuccessfully trying to communicate to the "skeptics"/contrarians here is the significance of the difference in the rate of change in the sea ice extent. The distribution of land mass is also important, but more about that later. Some numbers: In the Arctic in August and September the rate of loss has accelerated in recent years-- in fact, a quadratic fit provides a much superior fit to the data for July and August and September. But for ease of comparison I will refer only to trends derived using an OLS model. Here statistically significant is set at 95% (p-value < 0.05). Arctic sea ice area trends 1979-2010: January: about -37 000 km/decade (not statistically significant) February: about -34 000 km/decade (not statistically significant) July: about -330 000 km/decade (statistically significant, p-value = 0.000) August: about -390 000 km/decade (statistically significant, p-val =0.000) September: -429 000 km km/decade (statistically significant, p-val =0.000) Antarctic sea ice are trends 1979-2010: January: about +74 000 km/decade (not statistically significant, p-value = 0.219) February: about +75 000 km/decade (not statistically significant, p-value = 0.132) July: about +158 000 km/decade statistically significant, p-value = 0.02) August: about +100 000 km/decade (not statistically significant, p-val =0.072) September: about +100 000 km km/decade (not statistically significant, p-val =0.117) So hopefully this clarifies the point that Robert and others have been trying in vain to make-- the statistically significant and accelerating rate of loss of sea ice during the boreal summer far exceeds the small (statistically insignificant) increase of sea ice during the austral summer. During the boreal summer, the minimum Arctic sea ice area is currently around 3.2 million km^2, compared to 4.5 million km^2 in the early eighties (down 1.3 million km^2). By comparison, the Antarctic sea area minimum during the austral summer is currently around 2 million km^2, compared to 1.8 million km^2 in the early eighties (up 0.2 million km^2 and that trend is not statistically significant). Another issue that I alluded to earlier (here, but I was clearly being too subtle) is that the Arctic ocean is essentially surrounded by a continental land mass which heats up rapidly during the summer. In contrast, the Antarctic ice sheet is thousands of metres high and surrounded by a band of very strong westerlies which essentially isolate it. That is why, for now at least, there has been little polar amplification over most of the Antarctic (not to mention the complicating factor of ozone loss). In contrast, in part because of the albedo feedback, there has been very marked polar amplification of the Arctic. The loss of sea ice from the Arctic during the boreal summer is easily offsetting any gains in the Antarctic sea ice during the austral summer. It will be interesting to see how long it takes before the Antarctic sea ice starts responding to the warming en mass, right now decreases have mostly been regional.
    0 0
  25. Thanks, Albatross! BTW, it appears the level of replies to a 'basic overview' have gotten pretty in-depth. It seems da bottom of da barrel is now being scraped, eh? (yes, that's UP humor) The Yooper
    0 0
  26. Daniel @75, To take your metaphor further, I heard someone the other day say that "skeptics" are now scraping the rust off the bottom of the barrel.
    0 0
  27. 72 Albatross You still prefer to ignore the V09 data and snipe instead?
    0 0
  28. 73 muoncounter So you seem to approve of fitting trend lines to 7-8 years of data with obvious examples of inter-annual variability? Just because a trend line fits this short data set doesn't mean it contains any meaning. I'm not quite sure what we are meant to have confidence in? Chen has another analysis from 2009 which sheds more light on how what occured in 2006 is affecting the interpretation of this data set. ftp://ftp.csr.utexas.edu/pub/ggfc/papers/ngeo694.pdf "GRACE supports a flat rejection of the skeptic claim that Antarctica is gaining ice." I think this is a completely separate question and it depends on how GIA (or PGR) is handled. I'd direct you back to the link in #51 where EU scientist lay out why this is still a contentious issue for the mainstream.
    0 0
  29. #78: "Just because a trend line fits this short data set doesn't mean it contains any meaning." It the trend fits, it fits. The meaning is in the eye of the beholder. If you choose to disagree, fine. But then how do you take such meaning out of the 'blip', which is a far shorter time sample than the data presented by Velicogna. This is akin to the 'it hasn't warmed since 1998' nonsense. "this is a completely separate question" Not really, especially in a thread that starts off "Skeptic arguments that Antarctica is gaining ice". One thing these data do not show is a gain in ice. So your criticism must therefore be directed at those who make conclusions using no data whatsoever.
    0 0
  30. At the well ..., it is worth - this very interesting discussion, supplemented about the significant facts. At the beginning: I recall that one of the objections against Rignot et al. 2008 was: “Corr and Vaughn report that volcanic activity beneath the Antarctic ice may have increased the flow rate of some of the region's largest glaciers.” In addition, I will add that Monaghan et al., 2008., write: “Annual Antarctic snowfall accumulation trends in the GCMs agree with observations during 1960–1999, and the sensitivity of snowfall accumulation to near-surface air temperature fluctuations is approximately the same as observed, about 5% K −1 . Thus if Antarctic temperatures rise as projected, snowfall increases may partially offset ice sheet mass loss by mitigating an additional 1 mm y −1 of global sea level rise by 2100.” Why have not increased snow in the Antarctic? „However, 20th century (1880–1999) annual Antarctic near-surface air temperature trends in the GCMs are about 2.5-to-5 times larger-than-observed [!], possibly due to the radiative impact of UNREALISTIC increases in water vapor.” Also missing in this discussion, reference is made to this latest work: van Ommen et al., 2010. Commenting on his work van Ommen said: “Over the past 30 years, the cores indicated there had been a significant increase in snowfall in that area [coastal East Antarctica] ...” "The snowfall increase we see in the last 30 years lies well outside the natural range recorded over the past 750 years," In paper writes: “A range of factors [drought - Australia, the increase in snowfall - Antarctica], such as natural variability and changes in land use, ocean temperatures and atmospheric circulation, have been implicated in this drought, but the ultimate cause and the relative importance of the various factors remain unclear.” For me, "unclear" is at the same time: whether the (eventual) decrease in Antarctic ice mass is a sign of warming, or maybe, rather, its absence?
    0 0
  31. #80 Arkadiusz Semczyszak at 01:21 AM on 2 December, 2010 [drought - Australia, the increase in snowfall - Antarctica] If Snowfall increase in coastal East Antarctica is linked with Australian drought indeed and Australian drought has nothing to do with AGW, then East Antarctic snowfall is also due to multi-decadal variability alone, isn't it?
    0 0
  32. And the goal post shift, again (re #80 and #81). From the main post (first sentence): "Skeptic arguments that Antarctica is gaining ice frequently hinge on an error of omission, namely ignoring the difference between land ice and sea ice." And later, "All the sea ice talk aside, it is quite clear that really when it comes to Antarctic ice, sea ice is not the most important thing to measure." And then, "There is of course uncertainty in the estimations methods but multiple different types of measurement techniques (explained here) all show the same thing, Antarctica is losing land ice as a whole, and these losses are accelerating quickly." Now we have people talking about volcanoes explaining the loss of land ice, a swipe at the AOGCMS, a link to the questionable SPPI (the much debunked Monckton is their chief policy adviser), talk of how droughts in Australia is explained by changes in snowfall over portions of the EAIS (which is misleading and cherry picking as I will demonstrate below), and the old-time favourite "X can be explained by internal climate modes/variability" (which is also refuted by the study below). It is relevant that at least one prominent "skeptic"/contrarian (McIntyre) thinks that teleconnections are "voodoo science"...I digress. If someone, especially a “skeptics” makes claims about a paper, it is always best to go to the source. Here , again, we have an example of “skeptics” distorting the results from a paper to suite their means. From the abstract from a recent paper in Nature by van Ommen and Morgan (not et al.). “Here we report a significant inverse correlation between the records of precipitation at Law Dome, East Antarctica and southwest Western Australia [edit—not all of Australia] over the instrumental period, including the most recent decades. This relationship accounts for up to 40% of the variability on interannual to decadal timescales [edit-- not all of the variability], and seems to be driven by the meridional circulation south of Australia that simultaneously produces a northward flow of relatively cool, dry air to southwest Western Australia and a southward flow of warm, moist air to East Antarctica. This pattern of meridional flow is consistent with some projections of circulation changes arising from anthropogenic climate change. The precipitation anomaly of the past few decades in Law Dome is the largest in 750 years and lies outside the range of variability for the record as a whole, suggesting that the drought in Western Australia may be similarly unusual.” So on the one hand we have Arkadiusz @80 quoting the researcher to state that the southwestern Australian drought was well outside the range of natural variability, and then another "skeptic" (BP) in the next post is claiming that multi-decadal natural variability is at play. Please do make up you minds guys. This is just yet another example of the incoherence and inconsistency of arguments made by "skeptics". In fact, the usual suspects on this thread seem to be trying to fling as much mud as they can and hope that something sticks, and to try and distract us all from the science and issue at hand. Well that might work at WUWT, but not here.
    0 0
  33. #80: "one of the objections against Rignot et al. 2008 was: “Corr and Vaughn report that volcanic activity beneath the Antarctic ice may have increased the flow rate of some of the region's largest glaciers.”" Ouch. You've quoted directly from the SPPI denial site. Unfortunately for that objection and any credibility that SPPI may once have had, the layer depth dates the eruption at 207 BC±240 years. Hardly a factor in ice loss in this decade, century, millenium or any other relevant time period.
    0 0
  34. 79 muoncounter "This is akin to the 'it hasn't warmed since 1998' nonsense." Are you suggesting it's my argument that's akin to this? Because I thought I was trying to argue the opposite. I was trying to argue that generating trends on the back of short term data sets was dangerous. I specifically asked you if you were happy with fitting short term trends because I am not. If anything, because we have so little data, this is even more susceptible to false interpretation than the global temperature record. We actuall have no idea how interannual variability in the antarctic ice mass balance. So my worry still continues that V09 can make such strong assertions about the rate of ice mass loss doubling over the past decade is suspect. It really all does hinge on the anomalous 2006 data and the fact the author choose to start and end her trends on that year. That is not a matter of opinion, that's a fact. This is not just a problem with V09, Chen 09 comes to the same conclusion by ending and starting her/his trend around the same year. 2006 seems to be the important period for generating these trends. I wonder how you think the 2006 data should be interpreted and handled? Or whether these papers should acknowledge the problem of inter-annual variability?
    0 0
  35. 79 muoncounter "One thing these data do not show is a gain in ice. So your criticism must therefore be directed at those who make conclusions using no data whatsoever." Look I didn't want to avoid this question. I'll agree with you that those predicting gains are on shaky ground but what I'm concerned with is the sureness of those predicting losses. I think it's wrong to suggest that the data is all in one direction. I don't know if you can get round AGU's paywall but here's a new estimate of PGR using GPS from 2009 GEOCHEMISTRY GEOPHYSICS GEOSYSTEMS, VOL. 10, Q10005, 11 PP., 2009 doi:10.1029/2009GC002642 Geodetic measurements of vertical crustal velocity in West Antarctica and the implications for ice mass balance Bevis et al (hopefully not Butthead) Abstract "We present preliminary geodetic estimates for vertical bedrock velocity at twelve survey GPS stations in the West Antarctic GPS Network, an additional survey station in the northern Antarctic Peninsula, and eleven continuous GPS stations distributed across the continent. The spatial pattern of these velocities is not consistent with any postglacial rebound (PGR) model known to us. Four leading PGR models appear to be overpredicting uplift rates in the Transantarctic Mountains and West Antarctica and underpredicting them in the peninsula north of 65°. This discrepancy cannot be explained in terms of an elastic response to modern ice loss (except, perhaps, in part of the peninsula). Therefore, our initial geodetic results suggest that most GRACE ice mass rate estimates, which are critically dependent on a PGR correction, are systematically biased and are overpredicting ice loss for the continent as a whole. " Unfortunately as the paper points out it is impossible for them to know the full extent of this bias but it is likely a "significant fraction of all published ice mass rates derived from GRACE". Let's wait and see but as I said PGR is still contraversial. I'll agree with you that suggestions of ice mass gain are unwarranted but I think it's also reasonable to think that ice mass losses have been over-estimated. It worth pointing out that GRACE actually measures the equivalent of an ice mass gain for Antarctica, it is only the large estimates of PGR that turns this into an ice mass loss so getting PGR/GIA right is very important. The above is only one paper but a review of the subject in 2010 "Improved Constraints on Models of Glacial Isostatic Adjustment: A Review of the Contributionof Ground-Based Geodetic Observations" With a whooping 17 authors commented on the Bevis paper saying "their major conclusions appear robust". More generally on the subject of PGR they say "It is clear that GIA-related surface displacement observation and analysis around Antarctica, as with Greenland, is yet to reach maturity and further developments are required." I'm happy with 'wait and see' but I don't think Robert Way's article is coming to that conclusion so I'll stay in firm opposition to his position.
    0 0
  36. To be fair it's not just Robert, it's Copenhagan Diagnosis and the authors of these ice mass balances who are all underplaying the uncertainties in these processes.
    0 0
  37. #84, #85: HR, you objected to V09's use of a few years of data to generate a trend. Yet you use a self-described 'blip', one year long, to conclude that no 'strong assertions' can be made. "It really all does hinge on the anomalous 2006 data". How is what you are saying any better than what you say Velicogna has done? I merely drew an analogy to the 1998 'blip' of high temperatures, alluding to those who claim that one year will determine that 'it hasn't warmed' -- until that one year's temperature is exceeded. (And then the goal posts will be moved again, but that's a separate story). "and the fact the author choose to start and end her trends on that year." Read the portions I quoted in #73. It doesn't appear arbitrary: We fitted two straight continuous lines through the data, i.e., connected in the middle. We find that the Radj^2 for the two lines regression model is 0.97, the same than for the quadratic model I grant that your points about rebound velocities in #85 are legit and maybe the GRACE estimates are high. However, any rebound in Antarctica is not truly post-glacial in the sense of PGR from the Wisconsin ice sheet, as the ice in Antarctica is still there. If there is rebound in Antarctica, there must in fact be ice loss, because ice loss is what causes rebound. Searching 'antarctic ice loss' in Google Scholar, I find studies documenting measurements made by a wide range of technologies other than gravitational satellite: radar interferometry, shelf glacier acceleration, calving, rapid proliferation of crevasse systems and meltwater ponds, etc. None of these features are consistent with ice gain.
    0 0
  38. Daniel Bailey at 10:44 AM on 1 December, 2010 Do you not understand? "The point of the post is that focusing on Antarctic Sea Ice (ASI) is a strawman argument. ASI offers little contribution to the global energy budget, unlike changes in Arctic Sea Ice cover. Which then melts away come Antarctic summer." The problem as I see it is the anomaly for sea ice in the southern hemisphere is positive (more sea ice) and that is the case for all seasons. The antarctic sea ice expands and contracts just like the arctic sea ice does. This argument seems to speculate that when and if antarctic sea ice does melt totally in the summer continued loss of arctic sea ice will become more important as it continues to melt closer to the pole. But that doesn't seem too likely since antarctic sea ice is expanding rather than contracting and in the mean time the ice at both poles seasonally expand and contract, providing whatever effects that shrinking and expanding sea ice provides. So once again we are not talking science or real ice loss. The above argument argues that arctic sea ice will become more important than antarctic sea ice as the crystal ball is being read. One would expect that if the crystal ball were correct it would be more consistent with the theory that both poles were losing sea ice. But that is the nature of complex systems; simple explanations seldom suffice.
    0 0
  39. Bill, Please read my post @74. Note that, with the exception of July (the austral winter), the trends in the Antarctic sea ice are not statistically significant for the months I looked at. The trends in the Arctic ice are, and more importantly the loss is clearly accelerating, especially during the warm season. Consequently the loss of ice from the Arctic is fast outpacing any gains in Antarctic sea ice. This is probably in part because of polar amplification over the Arctic. So, the greatest loss in Arctic ice is during the warm season when insolation (incoming solar radiation) is high, but the impacts of the ice loss linger into winter, as is explained in the links on polar amplification that I provided. There is a positive feedback at work over the Arctic and it is linked to a lowering of albedo stemming from ice loss (including sea ice). The above situation explains why the trend in sea ice coverage is negative. Finally, multiple, independent lines of evidence show that the Antarctic ice sheet is losing ice, especially the WAIS. Additionally, trends in warm season sea ice coverage in the vicinity of WAIS are also down, and the loss of ice in the Bellinghause and Amundsen seas is critical, because the impacts of the loss of the sea ice which used to buttress ice sheets and glaciers has already been observed.
    0 0
  40. Albatross at 05:11 AM on 4 December, 2010 Bill, "Please read my post @74." Tests of statistical significance may or may not return strong results. In the case of your analysis you returned statistically significant change in two different directions. The fact that you got a distribution of 3 statistically significant ice shrinkage, 1 statistically significant ice gain, and 6 cases of no statistical significant change, it would seem the winner is no statistically significant change. I suspect that if fall and spring was included the number of cases of no statistically significant change would greatly increase. Thus what is happening with ice on the planet would only be remarkable under the assumption that statistically significant change is the rule.
    0 0
  41. "it would seem the winner is no statistically significant change" The above graph of global sea ice anomaly shows a significant decline. One has to go back to 1988 to find a positive anomaly comparable, albeit smaller, to the negative anomalies experienced on a regular basis in the past 10 years. And the baseline includes all years to 2008. Global sea ice area coverage is decreasing.
    0 0
  42. Hi Bill, "..... it would seem the winner is no statistically significant change." Umm, no. I'm afraid that you seem to have completely missed the point of that statistical analysis exercise. Anyhow, the graph @89 is the net result of all the changes/processes and it is showing a distinct downward trend in global sea ice since about 2001. Global sea ice coverage and volume is down. Continental glacier and ice sheet volume is down. Greenland ice sheet volume is down. Antarctic ice sheetvolume (mainly WAIS)is down.
    0 0
  43. Albatross at 09:42 AM on 4 December, 2010 Hi Bill, "Umm, no. I'm afraid that you seem to have completely missed the point of that statistical analysis exercise. Anyhow, the graph @89 is the net result of all the changes/processes and it is showing a distinct downward trend in global sea ice since about 2001." I looked at your post #74 and did not see the test of statistical significance applied to global sea ice Albatross. Since it appears to be statistically insignificant it probably either is or very close to it.
    0 0
  44. "Since it appears to be statistically insignificant." I don't know how one can look at the data and come to that conclusion. Whatever analysis this assertion is based on is in error. The decline of global sea ice is statistically significant.
    0 0
  45. Philippe Chantreau at 03:04 AM on 5 December, 2010 "I don't know how one can look at the data and come to that conclusion. Whatever analysis this assertion is based on is in error. The decline of global sea ice is statistically significant." The claim was made based upon a monthly analysis that ice gain in the Antarctic which also is visibly easy to see on a graph was NOT statistically significant. Thus the appropriate response would be that since there is a positive trend provided by the southern ice that reduces the loss of Arctic ice by nearly half on a month by month basis that the trend globally may also not be statistically significant. When one decreases the percentage of ice loss by both increasing the denominator and decreasing the numerator a mathematical answer is called for. Saying it is easily visible is not an appropriate response when somebody had already responded that something that is easily visible mathematically is not significant. (i.e. antarctic ice gain). The correct response is a third category, which Albatross can easily do as he obviously already has all the data and we can get away from the highly unscientific approach of arguing via cherry picking mathematical tests vs visual tests. All that is required to test this hypothesis is to add the two categories together in post #74 and do the math. An interesting variation if our objective is to test the statistic significance of the "pole" albedo issue would be to add in the land mass of antarctic as virtual sea ice serving the same purpose the core area of the Arctic serves. In that way we could test the statistical significance of "pole" ice. Then we can really begin to treat these issues consistently and better weigh the significance of each argument in a consistent manner without a lot of arm waving.
    0 0
  46. Bill, "Since it appears to be statistically insignificant it probably either is or very close to it." Now the goal posts shift, again. OK, I'll look into that for you. But I have a bunch of things to take care of this weekend...so please be patient. Actually, I do not have the global data for sea ice. Those data I showed earlier were area from NSIDC while the graph I posted was extent data from "Cryosphere today". PS: I have not "cherry-picked" data, nor is applying an OLS model and stating the statistical significance of the model fit "arm waving". The reason for me selecting the months I did is because (if you have read the original post) we have been discussing the impact of changes in albedo around the time of the solstices.
    0 0
  47. Albatross at 05:19 AM on 5 December, 2010 "PS: I have not "cherry-picked" data, nor is applying an OLS model and stating the statistical significance of the model fit "arm waving"." I didn't mean to imply you were arm waving. My reply was to Phillipe. I think our conversation was being productive until Phillipe stepped in and flat made a claim that global ice loss was statistically significant because on a graph it appeared to be. There was nothing unique about such an observation because we know ice melts and we can see it. My similar statement was not based on the visual graph but was a rough rule of thumb mathematical estimate of the effect of reducing the ice loss percentage by 30 to 40% from subtracting the antarctic ice gain from the arctic ice loss and by increasing the extent by 50% by adding the Antarctic maximum extent to the Arctic maximum extent. Should come out to be roughly the same significance as the Antarctic ice gain though it may fall on either side of the arbitrary significance line. Perhaps Phillipe misinterpreted what I was estimating. Consistently pursuing these arguments will remove any arm waving whether intended or not. Then of course to move the conservation forward in the spirit of consistency, it is always important to include all the elements of a theory in your theory. If the theory is feedback from loss of polar albedo (polar amplification) then the antarctic land mass should also be included in the calculations to estimate the significance of polar albedo loss (and polar amplification) to determine if the loss of albedo is in fact currently statistically significant. I think this is entirely consistent with the mathematical approach you offered in post #74 and allows each argument to be fully weighed in and avoid inconsistencies. My comment on "other months" was simply a layman's observation that the arctic ice extent differences seem relatively compressed during the equinoxes as opposed to the solstices. Since albedo is an ongoing year round phenomena I simply thought they should also be included though I realize that significantly ups the workload. I think it is important to get to solid answers as populations and the use of plant resources continue to accelerate. And the only way to get to solid answers is through a combination of critical, complete, accurate, and consistent collection and evaluation of data. Where we can go astray as a specie is through improper weighting of risks. The weighting of risks is inherently (philosophically and functionally) democratic in nature as none of us possess crystal balls and crystal balls is all we have until we get solid answers. Because of that fact, the only thing to be achieved by inconsistency is the exact opposite of that which we wish to achieve. That issue has already been well explored in philosophical and psychological literature and represents an ongoing concern supporting the viewpoint of Judith Curry.
    0 0
  48. Bill, you make it easy to misinterpret. What exactly are you trying to say, in a few words? This: "My similar statement was not based on the visual graph but was a rough rule of thumb mathematical estimate of the effect of reducing the ice loss percentage by 30 to 40% from subtracting the antarctic ice gain from the arctic ice loss and by increasing the extent by 50% by adding the Antarctic maximum extent to the Arctic maximum extent." What is the point of such manipulations? The only way to assess the significance of global sea ice loss is to examine real global sea ice data. Do it daily, monthly, whatever but I doubt that building fictitious quantities by adding numbers at various times of the year can show much about reality. "Since albedo is an ongoing year round phenomena." In the same sentence, you mention Arctic sea ice, so I assume that the albedo statement pertains to Arctic sea ice. Are you serious? I have very little time to devote to climate blogging these days. From a cursory read of your posts above, it does not appear that your assertion that the global sea ice decline is not statistically significant was supported by a real data analysis. Was it the case or not? What I read from you does not appear any better, a priori, than eyeballing a graph. I contend that data analysis from either NSIDC or CT will show that the decline in global sea ice is statistically significant. I have not the leisure of going at length about it. Tamino looked at the numbers again not long ago: "For the Southern Hemisphere, summer minimum has increased at about 9,000 km^2/yr while the winter maximum has increased at about 14,000 km^2/yr. For the Northern Hemisphere, on the other hand, the decrease in winter maximum has been about 42,000 km^2/yr and the decrease in summer minimum has proceeded at about 81,000 km^2/yr." The differences between these rates leaves little doubt IMO but feel free to attempt proving otherwise.
    0 0
  49. Philippe said: "What is the point of such manipulations?" You misunderstand. Those are not manipulations those are roughly the actual numbers. As Albatross did in post #74 he found significant negative ice in the Arctic in 3 of 5 solstice months he tested. And he found significant positive ice in the Antarctic in 1 of 5. So all the paragraph is an estimate of how the combined polar ice, since they are of opposite signs, ends up with ratios similar to the antarctic instead of the arctic (still negative but a negative ratio similar to the antarctic ratio). The percentages when you multiply them show that to be roughly the case. Thus it is likely that mathematically the change in global sea ice is not statistically significant even though it is in some cases regionally.
    0 0
  50. Bill, I have almost completed my analysis of the globals sea ice data....you are not going to like the results, it seems, with just a few months' data left to analyze that Philippe is correct.
    0 0

Prev  1  2  3  Next

You need to be logged in to post a comment. Login via the left margin or if you're new, register here.



The Consensus Project Website

THE ESCALATOR

(free to republish)


© Copyright 2024 John Cook
Home | Translations | About Us | Privacy | Contact Us