Arctic sea ice extent

I was somewhat surprised to see the following tweet from Andrew Neil

I have always quite liked Andrew Neil. I think he has an interesting background and comes across as quite sensible. I responded to his tweet and got a rather surprising – and slightly curt – response

So, where does this come from. It appears to come from this Telegraph article which quotes Ken Drinkwater (a researcher at the Institute of Marine Research at Bergen)

“The warming,” he says, “is primarily due to currents. A greater amount of warm Atlantic water is flowing into the North Atlantic and up to the Barents Sea.” He points out that this is just what happened in the 1920s and 1930s, when the ice melted even more dramatically than it has done in recent years, before it recovered again during the decades of what is called “the Little Cooling”.

Firstly, temperatures have risen faster in the arctic than in most other parts of the globe. It is something like 2oC higher than the 1951-1980 mean (see here). This is something like double the rise in mean global surface temperatures. It’s hard to believe that this hasn’t played some role in melting arctic sea ice. However, it’s quite likely that oceans currents can have changed and can have also played an important role in arctic sea ice melting. This, however, doesn’t mean that it’s not due to human-induced climate change. The Telegraph article also mentions the increase in Antarctic sea ice, which – as I discuss here – isn’t really comparable to Arctic sea ice when considering global warming.

What surprised me was the claim that the reduction in Arctic sea ice has happened before, in the 1920s and 1930s (as suggested by the Telegraph article and by Andrew Neil). I did some searching and came across this NOAA Arctic theme page which says During the first half-century, ice extent in all seasons remained essentially constant. It also includes the following figure which seems to show that in 2005, the Arctic sea ice extent was significantly lower than in the 1920s and 1930s. Since 2005, the minimum has dropped by a further factor of 2. So in the 1920s and 1930s, the minimum sea ice extent was about 11 Million km2. Today it is less than 4 Million km2.

20th century Arctic sea ice extent from NOAA.

20th century Arctic sea ice extent from NOAA.

So it seems clear that the sea ice extent in the 1920s and 1930s was not the same as it is today. It hasn’t happened before. The only thing I can find is a suggestion that there was a period during the 1920s and 1930s when the annual change was similar to what it is today (i.e., the difference between the maximum and minimum). However, that’s not really what we’re concerned about. We’re concerned about a continued decrease in sea ice extent and, ultimately, the complete loss of sea ice during the summer. This is not only a strong indicator of global warming but can have a significant impact on our climate (through the release of cold water into the oceans and changes to atmospheric circulation and ocean currents).

I guess Andrew Neil is entitled to tweet whatever he likes but should, ideally, do a bit more research before implying that what we’re currently undergoing is simply some kind of natural variation and is unrelated to human-induced global warming. One suggestion I have is, don’t trust the Telegraph when it comes to climate change or global warming at least.

This entry was posted in Climate change, Global warming and tagged , , , , . Bookmark the permalink.

38 Responses to Arctic sea ice extent

  1. Tom Curtis says:

    You would probably find the chart of the 1938 sea ice extent (ie, the second lowest up to the end of WW2) interesting. It is discussed here. In fact, in the 1930s, the sea ice melt in the European Arctic was quite extensive, and there are some breathless accounts of, for example, the first circumnavigation of Nova Zemlya from the period. These are accounts by sailors for whom the LIA, and sea ice extending as far south as the North Coast of Iceland were the background. However, the annual melts were not comparable to those over recent decades, and the total sea ice extent at minimum was far greater. That is partly because modern melt extends closer to the Arctic in the European arctic, but mostly because of the large melt backs north of Canada and Siberia which were lacking in the early twentieth century.

  2. Thanks, yes that was my impression. There was quite extensive melting in the 1930s but nothing that compared to recent decades and nothing that produced minima anything like we’re seeing today.

  3. ” This is not only a strong indicator of global warming but can have a significant impact on our climate (through the release of cold water into the oceans […])”

    Release of cold water?

  4. Have I misunderstood something? I assumed that the melting of arctic ice will increase sea levels and will be adding water at ooC. Is this wrong or simply not that important?

  5. I’ve never encountered this anywhere as an issues surrounding melting sea ice. The only direct consequence of melting sea ice is that it can change salinity of the water, which then could affect sea currents (but that really depends on the amount of melt).

    The biggest thing that’s mentioned is that less sea ice changes the albedo which then has consequences for temperatures (with all the further fun that can cause).

  6. Absolutely, you’re almost certainly correct. I was thinking along the lines of excess cold water influencing circulation. I completely agree that from a global warming perspective the big issue is the change in albedo which then results in the oceans absorbing more energy that would be the case were the sea ice still present. I should probably be more careful about what I say parenthetically, but the comments are here to correct my mistakes anyway 🙂

  7. Skeptikal says:

    I did some searching and came across this NOAA Arctic theme page

    That really is a spectacular display of ignorance. Sorry Mike, but for someone who claims to be “a professional and active scientist who teaches and carries out research at a university”… I really would have expected you to have at least contacted Professor Drinkwater and asked him for his data. Attempting to discredit someone’s work with sh!t you pull off a theme page, when you haven’t even looked at the work you’re discrediting, isn’t all that scientific… is it?

  8. There was an exchange of views regarding attribution of Arctic sea ice decline earlier this year at
    and a related post by Jos Hagelaars at Bart Verheggen’s blog

  9. My name isn’t Mike by the way, but I suspect that was simply intended as some kind of not so subtle insult. I really wasn’t trying to discredit Ken Drinkwater’s work as such. You do realise that this is a blog, not a peer-reviewed piece of work. It’s a Sunday morning, I don’t do this professionally, and anyone is free to comment and correct my mistakes and mis-interpretations – as Colin has already done. If you wish to correct my mistake, feel free to do so. Referring to data from a NOAA page as “sh!t” doesn’t seem particularly scientific to me.

    If anything, I don’t think I was all that critical of what Ken Drinkwater was quoted as saying. All I was attempting to do was suggest that even if there were periods during the 20s and 30s when there was significant melting it did not lead (according to what I can find) to minimum sea ice extents that were comparable to what we see today. I don’t believe that anyone is suggesting that there couldn’t have been periods in the past when “natural variations” lead to significant melting. However, there does not appear to be a period in recent history when the extent of the melting has lead to sea ice extents (both area and volume) as low as they are today.

  10. Lars Karlsson says:

    I see that the Telegraph article was written by non less that Christopher Booker, who once wrote that “Charles Darwin zealots have made science a substitute religion”.

    I can imagine that the part about “when the ice melted even more dramatically than it has done in recent years”, which is not a direct quote, does not come from Drinkwater, but from Booker. But it might be worth asking Drinkwater directly, as Skeptikal suggests.

  11. Indeed, it would be interesting to do so. What I disagree with Skeptical about is that my intent was to discredit Ken Drinkwater. I simply repeated what the article quoted him as saying (which indeed he may not actually have said and so maybe I should have made that clearer) and commented on whether there was evidence to support what the article is claiming. As far as I can tell, there is not. This post wasn’t intended to be about Ken Drinkwater, it was intended to be about whether or not the observed decrease in arctic sea ice extent over the last 30 years, has happened before (or, at least, has happened before in the last century).

  12. Skeptikal says:

    All I was attempting to do was suggest that even if there were periods during the 20s and 30s when there was significant melting it did not lead (according to what I can find) to minimum sea ice extents that were comparable to what we see today.

    Of course not… there were no satellites taking measurements back then. We can, however, look at past ‘climate’ through other observations. This appears to be what Professor Drinkwater has done. Read this article from 2010. Since we don’t have accurate data on sea ice extents from the 1920’s and 30’s, looking at past ‘climate’ through these other observations, is probably as good a comparison as we can make.

  13. Tom Curtis says:

    The “sh!t you pull off a theme page” is actually the product of a peer reviewed paper, “20th-century sea ice variations from observational data” by Chapman and Walsh (2001). It is based on collated data from ship-borne and air-borne observations and is accurate with regard to the limit of pack ice, but not with regard to ice conditions in the interior (which could not be observed). As a result, it is comparable to modern products with regard to sea ice extent, but should not be used in comparison of sea ice area. It is kept up to date by one of the authors.

    Skeptikal does not care about any of that, of course. He does not like what the data indicates, so he rejects it based on his own ignorance.

  14. Marco says:

    Thanks for that link, Skeptical, as it once again shows what a poor, or maybe even deceitful, interpreter of interpretations Christopher Booker really is. Booker cites Drinkwater talking about *local* water temperatures, but claims it relates to arctic sea ice melt. Booker stops the quote and then states “He points out that this is just what happened in the 1920s and 1930s, when the ice melted even more dramatically than it has done in recent years, before it recovered again during the decades of what is called “the Little Cooling”. “, suggesting Drinkwater has claimed the ice melted even more dramatically than it has done in recent years. Drinkwater did no such thing, as your link makes clear.

    I really, really would like to think that Booker just suffers from confirmation bias, but I am just not certain. It often is too subtle.

  15. Yes, there’s nothing particularly controversial in the article to which Skeptical links and – as you say – doesn’t really say anything about sea ice extent.

  16. Having re-read my post I’ve realised that my narrative isn’t that clear. My tweet to Andrew Neil wasn’t motivated by his original tweet, but by his response to a question from someone else. They asked what was causing the warmer water and Andrew Neil replied by saying A variety of reasons. Has a happened several times in last century.

  17. Pingback: Another Week of GW News, July 7, 2013 – A Few Things Ill Considered

  18. Marco says:

    FYI, Ken Drinkwater has reacted on Realclimate:

    Seems I was right about Booker misrepresenting his views.

  19. Very interesting, thanks. Seems you were indeed correct that Booker has extrapolated wildly from an article that made no claim that higher air temperatures were not responsible for arctic sea ice melt and did not claim that the same had happened in the 20s and 30s. Although I don’t think my post implied that Ken Drinkwater had much such claims, my apologies to him if it did appear that way.

  20. FrankD says:

    The 1938 Sea Ice Extent chart Tom Curtis refers to in the first reply is here:
    Navigate to the parent directory for coverage from 1893 to 1956. Actually observations, surprisingly more accurate than using fish in one small area of the Arctic as a proxy, whatever Skeptikal [sic] claims.

  21. Indeed, I went through quite a few of the different years and the observations and figures are quite impressive. It seems, judging by some of the other comments and by the response given by Ken Drinkwater at RealClimate (linked to above), that the only person (that I’m aware of at least) who has used changes in fish stock in one small area of the Arctic to suggest that current sea ice extents are similar to what was seen in the 20s and 30s, is Christopher Booker. By all accounts, not the most credible of sources 🙂

  22. NevenA says:

    Christopher Booker: “this is just what happened in the 1920s and 1930s, when the ice melted even more dramatically than it has done in recent years”

    Remember who lied to you!

  23. Lars Karlsson says:

    Yes, that was also my suspicion as soon as I saw it was Booker.

  24. Lars Karlsson says:

    Skeptikal, would you like to comment on Drinkwater’s objection to Booker’s misrepresentation?

  25. Yes, the next few years could be an interesting time in terms of who starts back-pedalling and re-interpreting their publicly stated views on climate change and global warming.

  26. Marco says:

    I don’t think Booker will ever do so. His ‘fame’ is directly related to him being controversial. As soon as he isn’t anymore, he will be thrown out of the door, and he knows it. He’s a bit like Peter Duesberg, who knows he will lose the adoration of the few while never regaining the respect of the many.

  27. Pingback: Another Week of GW News, July 7, 2013 [A Few Things Ill Considered] | Blog Submit

  28. BBD says:

    Remember who lied to you!

    The list is getting so long I will have to write it down.

  29. Pingback: Arctic misrepresentations | Blog Post Directory

  30. Lara K. says:

    This is great. Thank you for this blog. I’ve just come from “Watching the Deniers” and I am devastated by the lack of… everything. Thanks again. I’m now reading through your old posts.

  31. Pingback: Andrew Neil and Ed Davey | Wotts Up With That Blog

  32. Pingback: Andrew Neil’s correction to the corrections | Wotts Up With That Blog

  33. Arno Arrak says:

    All that confusion about Arctic ice. It tells me one thing for sure: climate scientists are too lazy to read scientific literature in their own field. This is the kindest explanation I can offer. The alternative would be a mind closed to any new information they don’t like. Two years ago I published an article [1] about Arctic warming that explained all you need to know about it and more. It proved two important facts:
    1. That Arctic warming is not greenhouse warming;
    2. that It is caused by warm Gulf Stream water carried into the Arctic Ocean by North Atlantic currents.

    Arctic warming started at the turn of the twentieth century, after two thousand years of slow, linear cooling. There was no parallel increase of atmospheric carbon dioxide and this rules out greenhouse warming as its cause. Laws of physics just don’t permit it.That is because the infrared absorptivity of carbon dioxide is a property of the gas and cannot be changed. To start a warming you must put more absorbing molecules into the air. Warming paused for thirty years in mid-century, then resumed, and is still going strong. The only logical source for a sudden warming is a re-organization of the North Atlantic current system at the turn of the century that started to carry warm Gulf Stream water into the Arctic. This would also explain why the Arctic today is the only place on earth that is still warming while a pause in global warming exists and has already lasted for 15 years. A direct measurement of water temperature reaching the Arctic in 2010 revealed that it was higher than anytime within the last 2000 years. The warming pause most likely was caused by a temporary return of the older flow pattern of currents. It was not simply a cessation of warming but an actual cooling at the rate of 0.3 degrees Celsius per decade. Nature is fickle and what has happened before can happen again. If, for any reason, a cessation of warming should happen again it would greatly interfere with any plans made for exploitation of the warming Arctic.
    [1] Arno Arrak E&E 22(8):1069-1083 (2011). (Check if Climate etc. still has it)

  34. You sounds remarkably certain about your work. I shall try to find it and give it a read.

  35. Tom Curtis says:

    Arno Arrak makes one clear mistake to begin with, ie, including articles published by Energy and Environment as being in the scientific literature.

    He further, mistakenly, claims there to have been no increase in CO2 in the early twentieth century. In fact, CO2 concentrations increased from 284 to 296 ppmv from 1830 to 1900, with a further increase to 310 ppmv by 1950. That represents an initial forcing of 0.22 W/m^2 from CO2 alone at the start of the twentieth century, and a further forcing of 0.25 W/m^2 over the first half of the twentieth century. That is 11.5% and 13% of total CO2 forcing since the preindustrial period to 2013 respectively. That CO2 was not the dominant forcing in the early twentieth century does not mean it did not exist – a common mistake by deniers.

    Further, Arrak does not discuss (in his comment) any other form of anthropogenic forcing. The big ones here are sulfur dioxide (negative) forcings and Black Carbon (positive). Black Carbon in particular was so prevalent in the late nineteenth century that it had evolutionary effects. It has a particularly large impact in arctic environments. Sulfates, on the other hand, were declining in the early twentieth century relative to CO2 emissions due to the switch from coal to oil for much energy production. Finally, one cannot ignore the dominant natural forcings of that period (a more active sun, and unusually low volcanic activity).

  36. Yes, Tom. I’ve had a quick read through his paper and it does seem that the central argument is that because the ice started melting (and I’m not sure it did TBH) in the early 20th century, that it proves that it can’t be anthropogenic. As you say though, anthropogenic forcings were not negligible in the early 20th century, so the central argument of the paper seems flawed.

Comments are closed.