My Arctic forecast this year from June 19 was that multi-year ice would make a big jump up again this year. NSIDC has verified that it happened.
I am not going to make a prediction of September ice extent this year….My forecast is that come the end of September, the amount of multi-year ice will again increase relative to last year, as it has done every year since 2008. In 2013 there should be an increase in the amount of five year old ice, because that is when the 2008 ice will have aged five years.
The results are in. there is as much 2+ ice as there was in 1995.
NSIDC keeps making this same silly statement that older ice is declining. It is simple math to predict that four year old ice will increase next September and five year old ice will increase in two years. Every year they push the bar up a year so that they can report bad news. Last year they complained about three year old ice.
Over the past few summers, more first-year ice has survived than in 2007, replenishing the younger multi-year ice categories (2- to 3-year-old ice). This multi-year ice appears to have played a key role in preserving the tongue of ice extending from near the North Pole toward the East Siberian Sea. However, the oldest, thickest ice (five or more years old) has continued to decline, particularly in the Beaufort and Chukchi Seas.
PIOMAS claims that ice thickness has massively declined since 2008. What a joke.
Changing the rules while the game is being played to fit their playing style is SOP for those feeding at the government trough and their defenders.
The defenders are looking for their share or the holier than tho hypocrites that that tell everyone to do what they will not do or there are the walking skeletons that think death warmed over is the latest fad!
..this reminds me of hurricane predictions
We have one year where Florida gets a string of hurricanes, and they predict hurricanes increasing.
2007, the wind blew…..it’s the death of Arctic ice….forevah
Dr. Roy Spencer reports that AMSR-E shut down today
==========================================
And the seas began to heal………………
Does that mean they shut off the Microwave ovens also?
The graph shows PERCENT of one-, two-, three-, etc year ice, not the actual amount of ice. I don’t have the data, but the percent of three year ice could increase at the same time that the AMOUNT of three year ice declined. It might be the old “increasing share of a shrinking market” from “Other People’s Money.”
Up to a point. 3rd year ice increased in 2010, as expected given the large amount of new ice formed in the 2007 re-freeze. However, 4th year ice (i.e. the same exact cohort) declined in 2011. This year must therefore have seen greater than usual losses of 4th year ice, and so there does seem to be an issue with long-term ice survival. That’s not to detract from your successful prediction that net MYI (3rd year and older) is up. It will be interesting to see how well this ice survives another year.
I do however have to take issue with your claim that there is as much MYI in 2011 as in 1995. Sure there is, by percentage of the total extent. Given that the total extent has declined by ~25% in this time, then so has the amount of MYI.
Peter, not sure why Steve keeps making these claims. MYI at the end of the 1995 melt season was 3.63 million sq-km, in 2011 it was 2.0 million sq-km. These are not the same.
Julienne: To be honest, I think it’s partly the way the graph is drawn up, as a percentage of the total. I think it would be more helpful either to display the graph on an absolute scale, or (even better) to include a data table with the numerical values alongside the graph.
Would it be possible to do that in subsequent updates, if not for this one?
I agree completely with your point. I don’t know why Dr. Meier switched to that format since we didn’t use it in our April 2008 posting. But since he has been using it lately, consistency is what they have been going for. But I agree, would be better to show the actual extent #s for each age class. That is what we show in our journal papers, such as the recent Maslanik et al., 2011 paper in GRL.
I didn’t say MYI. I said 2+
Yes. There is the same PERCENTAGE of 2+ year old ice in 2011 as there was in 1996. Given that the total extent is much lower, therefore so is the total amount of 2+ year old ice.
I don’t understand these changes in multi-year ice it seems to disappear from the centre and turn into younger ice,I am not sure how they work all this out.It seems that the amount of MYI can be anything they want it to be.I don’t have any confidence in these estimates and I will only believe what I actually observe happening.
If you look at previous NSIDC end-of-melt-season analysis, they have never distinguish ice age beyond 2+ year old. I doubt that they can.
See:
http://nsidc.org/images/arcticseaicenews/20091005_Figure5.png
http://nsidc.org/images/arcticseaicenews/20101004_Figure6.jpg
Why under moderation?
Did you include more than one URL in your post?
yes?
That’s why Fred. You can wait for Steve to check it, or break up the comment to where there’s only one url/per comment or break the url to where it is obvious the address (such as http :// suyts .wordpress .com / ) and resubmit.
1996 seems to be the year, on a percentage basis, that 2011 has more. What that means, I’ve no idea.
However, comparing to last year and 2008 would be more appropriate, IMHO, because the total ice extent is similar in size.
James, interestingly Steve hasn’t noticed yet that if you were to look only at the Arctic Basin, there was actually less sea ice than in 2011. Overall, 2011 had less ice extent than 2008 but that’s because 2008 had more extensive ice in the E. Greenland Sea.
BTW…on a % basis, summer 1996 had 53% MYI and 2011 had 52%, very close indeed. Of course the total amount of MYI is different, with 2 million sq-km in 2011 and 3.2 million in 1996.
To prevent “mis-interpretations” NSIDC should use absolute values of the amounts of MYI and FYI instead of the misleading percentage of current ice format.
Their Absolute Values are extrapolations with up to 85% error possible due to their >15%=100% calculation.
… but note also that while this may potentially lead to overestimation of ice age, it cannot lead to underestimation of ice age.
Age of sea ice is consistent with playing the Shell Game! Watch the hands and find the pea when the shells stop moving. Mostly guess work! SWAG!
Based on the Gozenta Effect!
If the concentration was 16% last year and 89% this year they would be counted the same, The reverse is also true!
Something that may be a bit more interesting is how much of that MYI was young MYI (2 and 3 yrs) versus ice 4 years or older.
In 1996, 24% of the MYI was 2 and 3 years old.
In 2011, 43% of the MYI was 2 and 3 years old.
Conversely, in 1996, 29% of the MYI was 4+years old, whereas in 2011 it was 9%.
oops, sorry those #s are relative to the total ice (FYI + MYI).
For just relative to MYI the following is true:
45% of MYI was 2 and 3 years old in 1996
83% of MYI was 2 and 3 years old in 2011
55% of MYI was 4+ years old in 1996
17% of MYI was 4+ years old in 2011
Julienne
Why should this be a surprise? Of course there must be a greater proportion of 2-3 year ice now given the 2007 decline.
Next year the 4 year figure will increase and in 2013 the 5 year figure will as well.
Paul, if you think of the end of summer 2007, then in spring 2008 you had a large fraction of first-year ice.
In spring 2009 that would have turned into 2nd year ice, in spring 2010 that would have been 3 yr ice, and in spring 2011 that would have been 4 yr ice.
So the 4 year old ice recovery that Steve has talked about that should have happened by now hasn’t happened. That’s the point I’m trying to make.
Steve also tries to make the argument that he now doesn’t expect the 4 year old ice to recover until next year because he’s saying there was a huge multiyear ice loss in winter 2007/2008. However, this is not true. At the end of the 2007 melt season there was 2.09 million sq-km of MYI. In March 2008 there was 2.07 million sq-km of MYI.
Steve also tries to make the argument that he now doesn’t expect the 4 year old ice to recover until next year because he’s saying there was a huge multiyear ice loss in winter 2007/2008.
This logic doesn’t follow in any event. Ice that was MYI (2nd year and above) in winter ’07-’08 would now be 5th year ice, not 4th year ice. For Steve’s logic to hold, there must have been a proportionally excessive loss of first year ice in 2008, and proportionally excessive retention of first year ice in 2009.
This is where it would be really helpful to visualise the ice broken down by cohort and do Kaplan-Meier survival curves.
By which I mean, take the March value of FYI as the “maximum” value for a given cohort, and plot what percentage of that cohort remains in September (when it’s still FYI), the following March (early 2nd year ice), following September (late 2nd year ice) and so on. You probably have the data to do it month by month, even.
Since the “fringe” FYI in (e.g.) the Bering and Greenland seas will be quite labile and weather-dependent, it would probably be best to consider only the Basin proper.
Alternatively, forget about the FYI and just look at MYI. In that case it would make most sense to set the 100% value for each cohort as the October 2y value, since that’s the maximum size of the cohort, after it’s just survived its first melt season.
Then, once you have the set of “lifetime curves” for the various ice cohorts, you can overlay them (or a selection of them) on the same axis, all starting at 100%. It should at the very least make it immediately visually obvious whether MYI is being predominantly lost in the winter by export or the summer by melt.
Moreover, the point at which the curves diverge from each other may hopefully contain information about the dynamics of the process.
Julienne
Something does not stack up here. Steve’s graph clearly shows the low point of MYI as 2008. Clearly Sep 2008 should not be lower than 2007.
Are we getting confused between first year ice and 1-2 year ice? etc?
So for instance in Spring 2009 the ice built up in winter 2007/8 would not be “2 years old” but 1-2 years. Or in other words by next Spring what now shows as “3yrs” will become 4 years?
I suppose what I am saying is that the 1,2 3,4,5 years definition shown above is the ageing as calculated in Spring 2011?
Or to look at it in another way, the big increase in 3 year ice this year, all things being equal, will translate into a big increase in 4 year ice next Spring?
Paul H: Maybe, maybe not. All other things cannot be assumed equal! 3-year old ice almost doubled in 2010 relative to 2009, and yet 4-year old ice (i.e. the same cohort) declined between 2010 and 2011.
Therefore, you can deduce there was an absolutely massive loss of 4-year ice this year. I’d love to see a month-by-month animation showing when it happened.
Sure, in 1995 there was more Arctic Sea ice, but not very much more…http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=09&fd=29&fy=1995&sm=09&sd=29&sy=2011
In September 1995 the monthly extent was 6.13 million sq-km, in September 2011 it was 4.61 million sq-km, or a 25% reduction.
Interesting…for Oct 3 (the latest CT daily area number), 2011 is at 3854045.4 km^2 vs. 4650936.6 km^2 for 1995, for a reduction in area of 17.1%. Was the ice really that less concentrated in 1995? Thinking about it, I’d expect 1995 to have been more concentrated assuming a solid core with only the “fringe” being low concentration. Thus, the larger total ice, the higher the concentration because the “fringe” becomes a lower and lower percent of the total.
-Scott
Mike, the age estimates are actually very conservative. If 15% of the pixel is MYI and 85% is FYI, it is tagged as MYI.
I guess we’re to just take the “assumed to be” between “is” and “MYI” as elided?
Can I ask a silly question?
How is MYI ice and its age identified ? Is it by thickness or something else?
Paul, it’s done by Lagrangian tracking of ice parcels. So using ice motion vectors, each pixel is tracked through the years and is assigned another age class if it survives the summer melt season. Once the minimum has been reached, each pixel is then aged another year. Make sense?
It seems like an interesting method to explore. How well does it work when compared to actual, physical evidence? Are there any cases where it does a poor job of correctly matching pixes to physical ice age? What sorts of assumptions are made about the ice which may affect (or “impact” if you’re a climate scientist) whether or not the 15% threshold is reached?
Julienne – how well do the results agree with submarine thickness measurements, CryoSat2 results, etc? I’d just worry about small errors propagating over time similar to performing numerical integration…even Runge-Kutta isn’t perfect.
Thanks,
-Scott
Scott: Quite so. Importantly though, this only affects the breakdown between the older age classes. The distinction between first-year ice and ice that has survived at least one melt season gets reset each year, since at the end of September, everything falls into the latter category.
So, the most reliable boundary in that chart is the one between purple and everything else. The other boundaries will have a larger associated error.
Steven , can you tell us what your scientific qualifications are please. P Gosselin claims you are a doctor , but Peter Sinclair is not so sure about that , who is right ?
I have a Bachelor of Science in Geology and a Masters In Electrical Engineering
Paul, yes there was less MYI at the end of the summer melt season in 2008 (what the graph above shows), and it was the least amount of MYI in the data record. What I’m saying is that at the end of the summer melt season in 2007 until the maximum was reached in 2008, there was essentially no reduction in the amount of MYI, so the low MYI observed at the end of the summer in 2008 is not because of export out of the Arctic via Fram Strait in winter. The minimum of MYI in September 2008 points to melting of more of the MYI that summer.
That’s interesting Julienne.
What you are saying is that for whatever reason MYI reached a minimum in Sep 2008 so that there was effectively a year’s lag.
This would mean that under your example earlier we will have to wait for Spring 2012 to see a rebound in 4+ year ice, which is surely what Steve’s graph shows.
Paul, yes the 3 year old ice at the end of this melt season has now been aged so that it’s now 4 year old ice. So depending on the winter circulation patterns, there could be more 4 year old ice before the start of the next melt season than we saw last year.
What’s funny in this particular topic of Arctic ice is that global warming alarmists have tried and tried to split hairs 4 ways over how much ice there is this or that year compared to this or that year over the last 30 years. It’s funny because there is more ice in the Arctic now than there has been for most of the last 9000 years. Arctic ice has been in a growing trend over the last 1000 years even with all this tempest in a thimble about what Arctic ice has been doing “in the satellite era”.
How funny, isn’t it, that alarmists are incessantly trying to scare people over what has happened in Arctic ice over the last 30 years. And some of them call themselves “scientists”. Silliness. The comedy of this world.
I find the September ice age diagram difficult to interpret,but I think it is saying that the first year ice has declined by a large amount as a % of the total ice since 2008/2009 while the second and third year ice have increased by a large amount,ice older then 3 years has decreased slowly.I think then that there need to be an increase in first year ice next year .