Climate Science Glossary

Term Lookup

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


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


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

Bluesky Facebook LinkedIn Mastodon MeWe

Twitter YouTube 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...

New? Register here
Forgot your password?

Latest Posts


Arctic sea ice low – what does it really mean?

Posted on 14 September 2011 by Verity

This is a cross-post from Carbon Brief

Arctic sea ice has hit the headlines over the last week. Last week the Polar Science Center of the Applied Physics Laboratory at the University of Washington found that Arctic sea ice volume has reached a new record minimum in 2010. This week, following an analysis of satellite data, researchers at the University of Bremen announced that Arctic sea ice extent has reached a \'historic low\' in 2011. Meanwhile, the National Snow and Ice Data Center (NSIDC), who normally release details of Arctic sea ice minima, has yet to determine that this year\'s summer minimum Arctic sea ice extent has been reached.

All of these different sources and claims can get pretty confusing. Just how do these different groups measure sea ice coverage? And why are they not all saying the same thing at the same time?

Arctic populations have been determining sea ice coverage visually for many years, so there are some some written records of ice extent over the last century. However, it is only since the launch of microwave energy-detecting satellites in the 1970s that scientists have been able to reliably quantify sea ice coverage. The satellites detect changes in microwaves emitted from the planet\'s surface. Ice, with its crystalline structure, emits more microwave energy than open seawater, so scientists can determine sea ice versus ocean coverage from the microwave data.

Some confusion arises from how the researchers present their data. Sometimes we hear about Arctic sea ice extent, other times sea ice volume, and sometimes even sea ice thickness. It is important to know which of these parameters is being discussed.

Carbon Brief have explained a bit about the difference between sea ice volume and sea ice extent in this blog. Here, we look at the question in a bit more detail.

The parameter that we tend to hear most about is Arctic sea ice extent - this is a measurement of the area of Arctic Ocean with at least 15% sea ice, and is determined from the satellite measurements discussed above. The National Snow and Ice Data Center (NSIDC) and IJIS (the International Arctic Research Center, working in co-operation with the Japan Aerospace Exploration Agency) post regular updates on Arctic sea ice extent.

The differences between results from separate research groups can arise as the groups may measure the microwaves using different frequency microwave sensors, and can reconstruct the ice coverage using different algorithms. This is why the sea ice reconstructions from researchers at the University of Bremen and the NSIDC differ slightly. The Bremen researchers use data from sensors that can study smaller area, but are more susceptible to interference melting or storm activity. The NSIDC sensors pick up a larger area, so are less detailed, but are not so susceptible to the effects of melting and storm activity. The two techniques do give good agreement though, and tend to show the same overall trends.

Mar Sept Arctic sea ice extent Cryoclim

The change in Arctic sea ice extent, 1979-2010. The graph shows the ice extent in March and September, as compared to the average sea ice extent for the whole period (the grey dotted line). Source: CryoClim

Arctic sea ice volume features less frequently in the news, but is another important parameter. The volume of sea ice is obviously more difficult to assess than the extent of sea ice, but it can be determined in two ways. It is either generated by combining computer modelling and the sea ice coverage determined by microwave satellite data (Pan-Arctic Ice Ocean Modeling and Assimilation System, PIOMAS - at the Polar Science Center), or more directly from altimetry measurements made by the CRYOSAT-2 satellite launched in 2010. The Polar Science Center posts regular updates on sea ice volume here.

PIOMAS Arctic sea ice volume anomaly

Arctic sea ice volume anomaly from PIOMAS updated once a month. Daily Sea Ice volume anomalies for each day are computed relative to the 1979 to 2010 average for that day of the year. The trend for the period 1979- present  is shown in blue. Shaded areas show one and two standard deviations from the trend. Error bars indicate the uncertainty of the  monthly anomaly plotted once per year. Source: Polar Science Center

Arctic sea ice thickness can also be determined using the CRYOSAT-2 satellite, or from the PIOMAS combination of modelling and microwave satellite reconstructions.

Although confusing, the fact that different research groups use different means of measuring the state of Arctic sea ice and may thus make slightly different assessments of this year\'s summer ice coverage minimum at slightly different times is relatively unimportant. The key point is that all of these different techniques show the same scenario - Arctic sea ice is declining, and that natural variation alone cannot explain it. Around half of the decline in Arctic sea ice since 1979 can be attributed to increases in greenhouse gases.

The media tend to focus on the Arctic sea ice minimum in September, and whether it has broken any records. However, this is misleading and draws attention away from more important phenomena. For example, as Real Climate highlighted in a recent blogpost, it is the major loss of Arctic sea ice during the early summer months of this year that will have the biggest climatic impact. Early summer ice-loss prevents the Arctic from remaining cool throughout summer, since the reflective ice is no longer present, allowing the sun to reach the open seawater and heat it. The warm seawater could delay the formation of ice during Autumn.

The early summer sea ice loss did, however, lead to much speculation in the blogosphere about the approaching September minimum, and whether it would break the record set by 2007\'s abnormally low sea ice extent. The unusual low of 2007 has since been ascribed to a very specific set of weather conditions, which exacerbated the increase in melting associated with global warming. Interestingly, as pointed out in an interesting blogpost at Climate Progress, similar weather conditions were present at the start of this summer, but the weather has not been so conducive to melting towards the end of summer, yet the Arctic sea ice extent has continued to hover around similar levels to those seen in 2007.

0 0

Printable Version  |  Link to this page


Comments 1 to 31:

  1. Nice post. Here is another interesting (and scary) graph from PIOMass. Ice Volume rather than Ice Volume Anomally. The bottom of the graph is 0. No Volume! My tip for an ice free Arctic - 2016 Nuff Said...
    0 0
  2. Also, it will be interesting when the Cryosat2 data goes live as a confirmation.
    0 0
  3. I think that the University of Bremen reported a record low based off of a single extent for one day - the NSIDC uses a five-day moving average, so it is possible that the record day may not show up in their data set. In any case it does not matter much, the past record in 2007 is a rather arbitrary line to cross. Whether 2011 breaks it or not does not change the fact that the negative trend continues. What is worth noting, however, is that the year's very-close-to-record minimum is without, if I'm not mistaken, the wind anomalies that contributed to the 2007 record.
    0 0

    [DB] The UB product uses a 6.25km resolution gridding, NSIDC 25km, adding to the difficulties in directly comparing them.

  4. as the Oceanic Heat Content steadily rises with the atmospheric GHG effect the effect on arctic sea ice would be steady thinning through the year, the effect of this thinning to summer ice coverage is steady rise in the acceleration of the yearly minimum if there's no changes in the oceanic currents, thus the yearly maximum would be steadily decreasing and the minimum exponentially (^1.5 faster), this line of reasoning doesn't take the slowing effect of the greenland melt in the extent of the ice into account nor the changing weather systems, f.e. when was the last time there was a stable polar vortex in the summer over arctic? computers' still acting funny so i do not wonder if this rant gets deleted on the way.
    0 0
  5. Decline in the extent of winter ice since 2003 tends to confirm that warming Arctic Ocean water is eroding ice from beneath throughout the year and, at least in winter is warmer than the atmosphere. This infers that sea ice contraction is accompanied by sea-atmosphere heat exchange enhancing land based ice melt. On the other hand, can it be argued that resulting increase in glacier outflow should cool ocean water, thereby reducing the rate of sea ice erosion – which is not what we see. Is this because increased glacier output only has cooling effect in a very limited area?
    0 0
  6. Just some thoughts. I believe, although I don't have this data to hand, that the major warming in the Arctic is a rise in winter minimums. Ice tends to impose a local temperature effect on the air immediately above it that restricts maximum temps until the ice is gone. As the minimums rise, this slows ice growth through the winter since the source of the cooling leading to ice formation is air temps, not water. This can come from both more ice freezing on the top although this is limited since the freezing air is so cold. But also from the thermal gradient through the ice sheet. If the air above is cold enough, the underside of the ice can be cold enough to cause more freezing. Raise the winter minimums and the bottom freezing rate reduces and ice thickness doesn't grow as much through winter. Then the melt starts and progressively a constant melt defeats a reduced freezing rate. Obviously an additional factor is the impact that declining extent has on water temps during summer as there is more ocean to absorb heat. So this could increase summer melt rates. But I can't see how this warming would carry through that much into water temps in winter since autumn freezing would tend to suck up this heat. Then the other impact is mechanical. Ice doesn't just melt. It fractures, splits, compresses, refreezes, drifts, swells, crashes around across the Arctic basin etc. And all of these things can impact, positively or negatively on total ice volume or extent. Ice that can break up more easily, even when it is thick, can more easily float away to warmer climes where it can melt faster. Ice thickness will fluctuate through the season, even at the north pole. If thickness drops below a critical threshold is can either melt completely in Summer, or become mechanically weak because it is too thin. So in the opinion of this arm-chair expert on arctic ice, what we are seeing is rising winter air temps, more summer water heating and mechanical weakening. And the Arctic Death Spiral. Possibly the key measure of what is happening isn't extent or mass. It is the ratio. If mass is low but extent is high we have masses of thin ice - easily melted in Summer, mechanically weak. If extent is low but mass is high, what is left is thick and strong
    0 0
  7. Addendum Have a look at the post at Climate Progress Nice changing graphic of Arctic ice Extent / Area(similar) / Volume. Look at the difference in the slopes of the area vs volume graphs. Volume is sliding faster than area. Ergo. More and more thin weak ice. Not good. Not good at all!
    0 0
  8. Glenn Tamblyn -" I believe, although I don't have this data to hand, that the major warming in the Arctic is a rise in winter minimums." Yup, visible in the graphs at the Danish Meteorological Institute. They don't have an animation set up (sadly) - you have to scroll through the individual years to see the trend.
    0 0
  9. Look for another significant drop by 2016- perhaps 25-30% less then 2007, 2011. See 80-85% less ice compared to the 1979-2000 median in 10 years (2020 or 2021)
    0 0
  10. Any significance to the minimum being earlier this year? The last several years showed the annual minimum occurring in mid to late September, while this year's minimum has apparently occurred in early September. Thoughts? Of course, this would be a premature post if sea ice declined further in the next year or two.
    0 0
  11. If we achieve an ice free Arctic ocean in the middle of the summer one year soon, the ocean becomes a giant solar collector throwing up masses of moist air with it's entrained latent heat. If this couples, like a mini Hadley cell, with the katabatic winds flowing down the slopes of Greenland, we suddenly get a huge import of heat to the ice. Ice melt could jump an order of magnitude. That would have some interesting effects.
    0 0
  12. #10 Not sure what you mean by 'premature post' there Jonathon, as this year goes to show we're on a trajectory, nothing premature about that assessment. As to the minimum being early (assuming that the minimum has happened), it the earliest since... 2008. Someone else may have the refreeze dates going back a few decades, but IIRC on average the dates have crept later and later, but with a great deal of variability. This year happens to be a quite early end, though still 2nd lowest or lowest extent on the record. Funny, talking about a trend with variability. I wonder where else we hear about that?
    0 0
  13. Interesting side note - according to the PIOMAS model results, every single day since December 20, 2009 has set a new record low volume for that day of the year. Of course, we are still waiting on Cryosat to see how close PIOMAS is to more directly measured results. It matched up pretty well with ICESat, but that has been offline for a few years now. In any case, ice volume appears to have dropped quite a bit again this year. If that continued we'd hit zero volume during Summer some time before this decade is over. The longer estimates for a 'nearly ice free' Arctic ocean in Summer all require that volume losses level off some time in the next few years.
    0 0
  14. Is there any data on the sea surface temperature in the Arctic? Would be interesting to see if there is a significant anomaly from the mean. Is there potential for the open ocean in the Arctic increasing the likelihood of another cold winter in Europe? Negative NAO causing Atlantic blocking.
    0 0
  15. Sky, Sorry, meant to say in another week or two, not year. While 2008 did experience a similar minimum date (9/9 for both years according to AMSR), it was almost flat for the rest of September. Dansmark data showed a minimum later in September for 2008. May not mean much anyway.
    0 0
  16. hyperactive - yes there's a lot of SST info. If you go to DMI and check that the map is showing the Arctic, you then choose from the left hand panel. Choose anomaly and the date you're interested in. Then play around with all the other nifty data on the site. (Terrible time eater if you let yourself get carried away.)
    0 0
  17. Jonathan: "Any significance to the minimum being earlier this year? The last several years showed the annual minimum occurring in mid to late September, while this year's minimum has apparently occurred in early September." Basically it's just the weather up there not being amenable to further compaction of the thin ice cover coupled with the freeze season getting underway in parts of the arctic. Nothing significant.
    0 0
  18. A big factor in sea ice is whether it is single or multi-year ice. When sea ice first freezes if forms a mush that thickens and solidifies. Brine drains through this to find its way downwards. This single year ice is not a particularly good conductor. The major heat loss is through the surface during winter darkness and temps ~minus 40degC. In the 1970s and 80s not all the ice melted especially off Northern Canada towards the pole. In this region second and third year growth continued as landfast ice. Multiyear ice has denser crystalline structure and is more difficult to melt. (and a lot more difficult to drill through) From looking at CRREL data and from personal experience, I note spring melt tends to be about 2/3 from the bottom warmer water intrusions. The sun and warmer air temps tend to account for the other third. There was controversy a year to two ago when the sea ice extent was reported as getting larger. A Canadian team who actually went onto the ice reported it was mostly rotten single year ice with very little multi-year ice in existence. (See EOS Transactions of AGU 16 Feb 201 for two reviews:- Ice is rotten in the Beaufort Sea Geophys. Res Letters, doi:10.1029/2009GL041424,2009 Heat flow from the Pacific contributes to Arctic sea ice melt Geophys Res Letters doi:101029/2009GL04121.2010) It is the large volumes of 'rotten' multi-year ice that accounts for the accelerating melting rates and trend towards a totally ice-free Arctic. I have myself measured under ice conditions during breakup of arctic rivers. In the 20min measuring interval water went from -1.8degC at salinity ~40ppt (brine) to 0deg salinity 0 (ice cold freshwater) from the land drainage flowing out over and under the sea ice. Species diversity is low but those that live there survive these conditions without exploding or obvious stress. Bering Sea inflow is almost a an order of magnitude less than the North Atlantic drift up around Norway. But in certain years, related to ENSO, the heat flux through the Bering Strait into Chukchi and Beaufort seas can account for a third of ice lost as it did in 2007. I'm retired now and well remember the ice island T3 that circled in the Beaufort gyre for several years providing a stable ice platform and science base. It eventually left via Denmark Strait east of Greenland. The whole Arctic has changed drastically and the multi-year ice platforms and glacial fragments are not so common. I hope my comments are helpful.
    0 0
  19. Judging from a regression analysis of Cryosat Today’s ice area data, there was no significant trend toward earlier or later minimum dates over 1979-2010. By a thin margin, CT area also reached a new low point in 2011: 2.90m km^2 on Sep 10.
    0 0
  20. Micawber, very interesting. Thanks for the first hand perspectives. I've often wondered about the amount of heat 'imported' into the Arctic ocean from the Atlantic and Pacific and whether this will change as the ice breaks up and new current patterns emerge.
    0 0
  21. More heat for CBD: August was a warm month in the Arctic...and the Antarctic as well: [Source] Looks like da Summertime at bot' da poles, eh?
    0 0
  22. Silly question of the week. Does anyone have a rough idea how much the ice would have been effected if Solar insolation had not dropped since 2002?
    0 0
  23. villabolo, no I can't offer any papers or other objective evidence. Personal opinion? OK. My view would be that the major influence over this period is the steadily warming ocean waters. Regardless of the season, there is always liquid water beneath all but the tightest of landfast ice. And that water is now warmer, year in, year out, than the same inflows and currents were 15 and more years ago. (Warm being a relative term. I'm not starting swimming classes.) That warmth constantly weakens even if it doesn't actually melt ice. It also means that much less insolation and shorter time periods are needed to get the melt season going. Even if we get our act together to cut ghg emissions, we'll have to see several years, if not decades, of less and less ice because of that ocean heat being unable to dissipate before various currents carry it to the ice.
    0 0
  24. It would be good to get a year by year plot of the graph in @1 above. Would this reflect the global temp rise occurring?
    0 0

    [DB] Here are three from contributors on Neven's Arctic Sea Ice blog:




  25. The last graph in #24 shows an interesting bit about how the volume decline has progressed... up through 2006 there was a fairly steady 'slow' decline. Then in 2007 the volume dropped more sharply than it had in any of the previous years. 2008 and 2009 had slightly higher volumes, but still far below all previous years. Then in 2010 the volume dropped sharply again. This year is slightly lower, but overall very close to 2010. To me this suggests that we've entered a phase where, in addition to the slower ongoing decline, the ice is now thin and broken up enough that weather conditions in individual years can cause much more significant drops... which then become 'permanent'. Essentially, while the ice volume of 2007 was shockingly low at the time, the subsequent drop in 2010 has now made 2007 levels a high volume which will not be seen again. The implication of the second graph, that ice volume could be nearly zero from July through November by 2020, is truly shocking. However, if we continue to see sharp drops without recovery like 2007 & 2010 it is entirely plausible. It'd take decades to get to zero volume at the rate of decline seen through 2006, or the change between 2010 and 2011... or just two more years like 2007 & 2010.
    0 0
  26. Thanks for these very interesting graphs and comments The last graph in #24 shows 1979-2001 average max 30 to av min 14 thousand cubic km. On average therefore 16, 000 cu km of ice melts and refreezes during the 1980s and 90s. From the graph we note that 2010 max to min was c23.3-4.2 ie 19,100 cu km melted. Max in 2011 volume was 22,000 cubic km compared to the 23,300 in 2010. So maximum ice volume declined from 2010 to 2011 by about 1,100 cu km. Assuming the same amount of heat input in the two years, we have an excess latent heat to melt 1,100 cu km available to melt more ice or to warm the surface waters in 2011. Since the latent heat (heat to melt a unit quantity) of seawater is about 80 times the specific heat (heat to raise by 1 degC a unit quantity), then we can expect increasing ice melt this year and in successive years. Graphs show this accelerating trend as one would expect on average other things being equal. We would expect this from the excess latent heat from one year to the next. On this basis it is clear that once we reach zero ice volume, then heat goes into raising sea surface temperature. This will be rapid given the ratio of 80 latent to specific heat. In addition we know that global temperatures are increasing and that polar regions average SST is increasing by twice the global average, we can expect less and less ice to form during the winter season. This will result in a warming arctic ocean. Remember also that heat capacity/specific heat of water is over 4,000 times that of air. So arctic air even at forty below zero has little effect on the huge ocean heat capacity and heat loss. There may well be winter ice but seawater does not have the 4degC maximum density that freshwater lakes have at the bottom. Freshwater lakes cannot be compared to seawater lakes and seas for this reason. Maximum density of seawater is below its freezing point. Moreover the great majority of global warming is in the oceans for reasons because of the high heat capacity. This will have huge effects climate. Springs will be much warmer as less and less ice melts taking heat less heat from air during breakup and ice melt seasons. It will be fascinating to watch the UW APL graphs as we journey through these interesting times.
    0 0
  27. In short I agree with the comments above. 1,100 cubic km of excess latent heat will melt the 2010 min of 4,200 or this year's min of c4,000 cu km in less than 4 years. So 2015 is a totally ice free Arctic ocean August-September on current trends and is likely to be sooner. This month a team reached the magnetic north pole entirely by boat. One can expect to reach anywhere in the Arctic by boat in 2015. Reduction of the 2011 maximum ice volume 22,000 cu km by 1,100 cu km per year suggests a maximum of 20 years for a totally ice free Arctic year round on current trends. It will certainly occur much sooner than that.
    0 0

    [DB] One can also visually see the demise of the Multi-Year ice when viewed as in this graphic:

    MY Ice

    "MYI recovery observed in recent years shows a delay relative to
    thermodynamic forcing indicates that MYI is resistant to recovery"


  28. DB Many thanks for this link to online pre-print: Recent changes of arctic multiyear sea-ice coverage and the likely causes by Polyakov et al doi: 10.1175/BAMS-D-11-00070.1 Lots of data to ponder there.
    0 0
  29. For an interesting back and forth between the much-esteemed neven and an Arctic-ice-is-not-melting guy, see this article from the Nunatsiaq Online. The gentleman in question also apparently believes that Apollo mission 'inconsistencies' need to be answered and dinosaurs did not exist: Are we being deceived and brainwashed at an early age into believing a dinosaur myth? Deep probing questions need to be asked of the entire dinosaur business. With that kind of insight, his observations re Arctic ice melt are sure to be spot on!
    0 0

    [DB] Hot-Liquid-Alert-Warning!  Following the links to the materiel therein can be hazardous to computer keyboards and monitors if hot liquids are present!

  30. DB, if that's the equal of a Joe Romm head vise alert, I agree. I read that comment stream the other day and thought I felt something pop! in my frontal lobe. That is one of the great all-time conspiracy lines, though: "Deep probing questions need to be asked of the entire dinosaur business."
    0 0
  31. There's an eerie familiarity to some of his rhetoric. One simply has to replace 'dinosaur' with 'AGW': "Dinosaur" bones sell for a lot of money at auctions. It is a profitable business. There is pressure for academics to publish papers. Museums are in the business of producing displays that are popular and appealing. Movie producers and the media need to produce material to sell to stay in business. The mainstream media loves to hype alleged dinosaurs finds. Much is to be gained by converting a bland non-dinosaur discovery, of a bone of modern origin, into an impressive dinosaur find, and letting artists' interpretations and imaginations take the spotlight, rather than the basic boring real find. There are people who desire and crave prestige, fame and attention. There is the bandwagon effect and crowd behaviour. And then there are people and entities pursuing political and religious agendas.
    0 0

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


(free to republish)

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