Welcome to DU! The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards. Join the community: Create a free account Support DU (and get rid of ads!): Become a Star Member Latest Breaking News General Discussion The DU Lounge All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search
9 replies = new reply since forum marked as read
Highlight: NoneDon't highlight anything 5 newestHighlight 5 most recent replies
End of June Arctic Sea Ice Volume at lowest ever anomaly (Original Post) muriel_volestrangler Jul 2012 OP
Wow! That's an attention-getter! hatrack Jul 2012 #1
It seems to me that the regression line gives a rosier picture than reality HereSince1628 Jul 2012 #2
I think you're probably right; but it's very difficult to chose a type of curve muriel_volestrangler Jul 2012 #3
I suspect that the linear regression analysis is part of the database program HereSince1628 Jul 2012 #4
A guy who goes by "Wipneus" does graphs for this (with various cases): joshcryer Jul 2012 #6
Thanks. HereSince1628 Jul 2012 #7
You may also be interested in Neven's Arctic Sea Ice coverage: joshcryer Jul 2012 #8
Heh, look at 2010, 2011, and 2012 so far. That is crazy. joshcryer Jul 2012 #5
I'm picking out my end of the world pajamas. limpyhobbler Jul 2012 #9

HereSince1628

(36,063 posts)
2. It seems to me that the regression line gives a rosier picture than reality
Sat Jul 7, 2012, 02:29 PM
Jul 2012

If you look at the pattern of error in the residuals it's on the same side at both ends...which is to say the data is more curvilinear than linear.

It sort of looks to me like the data may actually be showing a negative acceleration (possibly a negative exponential trend, but I haven't done the stats).

muriel_volestrangler

(101,265 posts)
3. I think you're probably right; but it's very difficult to chose a type of curve
Sat Jul 7, 2012, 03:50 PM
Jul 2012

when it's a noisy data set - and when it's an anomaly for each time of year, it gets even harder (for instance, the June anomaly, about 10,500 km3, applied to the absolute minimum level, would mean no ice at all - so for the time around September, the anomaly could never go below its present value, but at other times of year it will still be able to go lower.

But plotting a straight line is uncontroversial, so that's what they've done. It's up to us to see that a straight line isn't an adequate description.

HereSince1628

(36,063 posts)
4. I suspect that the linear regression analysis is part of the database program
Sat Jul 7, 2012, 04:20 PM
Jul 2012

used to create the original graph, and that easy access resulted in its use. I doubt anyone thought a first order solution was all that the public could bear.

But I could be wrong.

joshcryer

(62,265 posts)
8. You may also be interested in Neven's Arctic Sea Ice coverage:
Sun Jul 8, 2012, 11:46 PM
Jul 2012
http://neven1.typepad.com/

A lot of the commentators in the comments section saw what you saw.
Latest Discussions»Issue Forums»Environment & Energy»End of June Arctic Sea Ic...