Arctic Sea Ice Loss, part 3

Having looked at extent, area, and volume of Arctic sea ice, then at its thickness, let’s examine some other derived variables.

There are (at least) two derived variables we can construct from extent and area. One, which I’ve called “spread,” is their ratio, i.e., extent divided by area. The other, which I’ve dubbed “split” (although I don’t think that’s such a good term) is their difference, i.e., extent minus area. These are related to how much the ice pack is broken up. If the ice pack were a single, unbroken solid piece then extent would equal area, so that the spread would be 1 and the split would be zero.

First let’s look at their time series. Here’s the spread:


Clearly the spread has increased over time, but only its annual maximum. The annual minimum has stayed surprisingly constant. The large increase in annual-maximum spread is simply due to the fact that sea ice cover has decreased so much, so that the same excess of extent over area (the same “split”) corresponds to a larger ratio of extent over area (a larger “spread”).

The split hasn’t shown the extreme time change which spread has undergone (although, as we’ll see, it does show some changes):


Although split actually does show some trends over time, there are different trends at different times of year:


There has been notable increase during May and June, with notable decrease during September through November (and to a lesser degree in December through March). Essentially, the ice pack has become more fragmented during May and June, less so during September through November.

The trends in spread show increase from April through October, peaking in August, but no significant change for November through March:


The lack of spread trend for November through March reveals that the decline in split has been in the same ratio as the decline in sea ice overall, to keep the extent/area ratio constant. The interesting cases are September and October, for which split has declined, but overall sea ice has declined even faster so that spread is on the increase. Also interesting are July and August, for which split has shown no trend but overall ice has reduced sharply, so the proportion of extent to area (the spread) has increased quite a lot.

Changes in the annual cycles are also revealed by windowed Fourier analysis. For split, the amplitude of the fundamental Fourier component shows changes but the long-term trend is far from linear:


The timing (as indicated by phase of maximum) of the fundamental Fourier component has changed dramatically:


Over the entire time span, the phase has changed by about 0.2 cycles (about 70 days). The most recent years (since 2010) show continuing decrease, the migration of the timing of maximum continuing until the very present. If we compute the average annual cycle for the 1980s, 1990s, and 2000s, we can see some of the migration toward earlier maxima even though this averages out the changes in the 1980s and omits the data since 2010 which continues this trend:


The relative phases (relative to the fundamental) of the higher harmonics of the Fourier series also show decline in the 1980s and continued decline since 2010:


Spread shows the continual amplitude increase which would be expected from its continual increase at maximum with stable minima:


Again we can average the cycle for the decades of the 1980s, 1990s, and 2000s, to illustrate the cycle change:


These two derived variables add to the overall picture of dramatic change in Arctic sea ice. The changes in split (the excess of extent over area) also show continued change after 2010, which further identifies 2010 as a banner year for sea ice change and again argues against the idea that Arctic sea ice has recently been “stabilizing.”

All told, every variable we’ve studied shows significant, in fact remarkable, change in Arctic sea ice. The northern ice pack is truly the “canary in a coal mine” which is warning us of the extremity of changes to come — soon — to the rest of the planet as a result of our throwing a monkey wrench into the machinery of the climate system. The question “Has mankind dramatically altered earth’s climate?” has been answered in the affirmative. The question before us now is, “Will mankind heed the warning signs which are in plain sight?”

16 responses to “Arctic Sea Ice Loss, part 3

  1. “Will mankind heed the warning signs which are in plain sight?”

    Unlikely. Too much psychopathology amongst small percent of people holding most of the wealth/power.

  2. arch stanton

    “…let’s examine some other derived variables…The question before us now is, “Will mankind heed the warning signs which are in plain sight?””

    Apparently not as long as other signs can be found in the bushes of Watts and the like that say: “Full speed ahead. Damn the warning signs posted by those of lessor intellect.”


  3. Tamino, you made me curious about what kind of trends your “spread and split” would show in the Antarctic sea ice, where wind-induced polynyas are known to scatter the ice around. Do you have any plans on posting something about it?

  4. Too pessimistic, I think–we will heed the warning signs, in time. The question is, will it really be “in time?”

    • Hard to know. Before significant harm? Doubt it. These things take time…lot’s of time. It took 50 years essentially to get to where we are today re. tobacco. 30 years of MADD have made a difference…perhaps 50%. It’s really is the way of the world.

  5. > more fragmented during May and June,
    > less so during September through November.

    What’s the physical process there?
    I’d think:
    — ice breaking up into smaller chunks that spread out during May and June, so more ocean has “15 percent or more” ice present
    — smaller chunks of ice then melting so more ocean drops below that 15 percent cover. The more smaller pieces, the more surface area of ice in the water exposed to melting.

    but I’m guessing. I’d love to see someone come up with photographs to correlate with the numbers.

    • As one who lives at the southern edge of the pack (Newfoundland) I have most unscientifically observed this over the past 3 decades in my own region. It is very rare this past decade to get a 100% ice cover here in the spring though 20 to 30 years ago it was an annual occurrence.

  6. Are you familiar with SUYTS blog post. I just had a discussion with him and his followers who are abslutely convinced that the Total Ice coverage is the only relevant factor, and that it is clearly things like AO and AMO that are causing this “down” cycle in the arcitc. The Antarctic is gaining ice, so therefore there can be no real problem, and the arcitc scientists are just making al this stuff up
    I of course have mockingly asked them WHY none of the denier blogs were explaining this before or after 2007. There is apparently a study from 2004 that says ice loss ioss is due to winds and other factors

    Click to access NghiemEtal2007_MYreduction.pdf
    And that total albedo has INCREASED over the last 30 years, and all sorts of proof that the arctic ice loss is meaningless.
    Whole he is a comitted ideologue and convinced that ALL the science pointing to ACC is plain wrong, he is quite a nice person and has always been respectful towards me. He responds to many of my responses (not all of them of course), and i think it would be interesting to have people who understand the science much better than I do interact and see what happened.

    [Response: Being a “nice person” doesn’t make him any more amenable to reason.]

  7. Top notch work. Can you access methane data and have a look at trends in that; emissions are fast rising in the Arctic and this is of enormous importance. Something for starters… Yurganov stuff…
    Cheers, Paul Beckwith

  8. I forgot to include this link from NASA movie posted today on sea ice…

    and this link on how sea ice affects weather extremes, etc…

    Paul Beckwith

  9. Sorry to suggest more work but it would really be interesting to examine the sea ice data, specifically the volume but maybe the others with autocorrelation lag-1 functions. There is work that suggests that ACF can be used to predict tipping points since often in a rapid transition there are early warning signs in slowing of the system. The paper Cimatorius et. al., 2012, called Dansgaard-Oeschger events: tipping points in the climate system appeared in Clim. Past. Discuss. fairly recently. If you cannot access it let me know. Regards, Paul Beckwith

  10. Excellent analysis. I hope we do heed the warnings. Time is probably very short and things are getting just a bit too scarey for my taste.

  11. Thanks Tamino,

    Not had time to properly read through your posts, FWIW I use the August average NSIDC Extent / CT Area as a ‘dispersion index’ as it avoids the September re-freeze while including net seasonal impacts.

    For those who want more Arctic data to dig into. PIOMAS is also available as gridded data available here:
    I’ve broken down each grid box into 25cm thickness categories, and calculated the volume attributed to grid cells of thickness categories. I’ve also broken down into three regions. The resultant data is available via this page:
    Either as CSV style tables on web pages or as an Excel spreadsheet.