2010-09-03 08:25:20Basic Rebuttal #22: Arctic icemelt is a natural cycle revision 3
doug_bostrom

dbostrom@clearwire...
184.77.83.151

Summary: 

Arctic sea ice has aptly been termed a "canary in the global warming coal mine," a sensitive indicator of climate change; because of its importance as a diagnostic of global warming, climate change skeptics struggle to explain the decline of Arctic sea ice as a natural phenomenon. 

Full rebuttal: 

Satellite measurements of Arctic sea ice extent reveal a rapid decline over the past 30 years, particularly at the end of each year's annual melt season.  The downward trend and the increasing difference between seasons are in keeping with predictions of the effects of global warming. As the Arctic warms, the volume of ice in the region gradually declines, making it less likely ice will survive more than one year and thus exposing more open water at the end of each melt season.

 

(from National Snow and Ice Data Center )

As an explanation for the decline of Arctic sea ice, skeptics hypothesize we're seeing the effects of natural cycles causing deep, decades-long swings in Arctic ice coverage and volume. Lending observational support for such cycles is much more difficult than relying on direct observations of ice extent with contemporary instruments. Still, thanks to ocean sediment cores and some other physical clues left by past climate regimes we have reasonable insight into past Arctic sea ice extent. Combining various information about past climate behavior, we can better understand why changes in ice coverage have occurred in past times, whether those natural variations are happening today, and how those changes compare to today's sea ice trend. 

While it's true that natural variations of the climate have caused significant changes in Arctic ice extent in the past, it's important to note that such such changes are not airtight arguments against anthropogenic global warming causing today's loss of ice. After all, events of the past do not describe newly identified influences by human culture on today's climate. Indeed, comparisons between past and present Arctic climate reveal different reasons for yesterday's and today's Arctic sea ice changes and strongly suggest that today's changes are largely anthropogenic ( Overpeck et al ).  Meanwhile, analysis of several hundred indicators of past Arctic sea ice extent tells us that recent losses appear to have no parallel in records going back many thousands of years ( Polyak et al ).

The past 200 years offers an example of how natural and anthropogenic influences on Arctic sea ice can be distinguished.  The Arctic appears to have undergone an unusually cool period in the early 19th century, certainly natural, with recovery to more normal conditions extending into the 20th century leading to the warming we see today. Referring to the graph above, we can see that after the earlier cool period sea ice extent in the Arctic appears to have largely stabilized, later to begin a steady decline in chorus with other emerging observations of global warming such as increasing air and ocean temperatures. This decline in ice extent is happening even though the causes for natural recovery from the unusual cold of the 19th century are no longer in play, while research strongly suggests these recent reductions in Arctic sea ice are are caused by a new, anthropogenic mechanism ( Johannessen et al ).  

In sum, although natural factors have always influenced the state of Arctic sea ice, research strongly suggests that today's decline is driven by the novel influence of anthropogenic C02 we've added to the atmosphere and thus is unique in Earth's history.

 

 

 


 


2010-09-03 12:30:33Comment
Robert Way

robert_way19@hotmail...
142.162.188.138
Doug,
great post. I'm thinking though that you should consider making the chart yourself so to simplify it. I can provide the data if you like?
2010-09-03 12:59:37suggestions
dana1981
Dana Nuccitelli
dana1981@yahoo...
71.140.3.99

Good content, just a few suggestions.

In the 3rd paragraph of the rebuttal there's a confusingly-worded sentence.

"When Arctic sea ice data are examined for trends, for the entire period of satellite observations Arctic sea ice coverage at the end of the melt season has been shrinking with maximum extent changing less."

I'd rephrase something like this:

"Examining the trends in the Arctic sea ice extent satellite record reveals a rapid decline, particularly at the end of the annual melt season."

Another sentence later in that paragraph says

"As air and ocean temperature conditions in the Arctic warm the volume of ice in the region gradually declines..."

You should add a comma after the word "warm".

I'd also suggest adding an Arctic sea ice volume graph and discussion, because the volume tells us more about the decline than the area or extent of sea ice.

2010-09-03 15:19:41
doug_bostrom

dbostrom@clearwire...
184.77.83.151

Dana, all good suggestions, thank you and I'll weave 'em in.

Robert, you're very kind to offer your help. I'm particularly fond of this graph not only because of its duration which helps it to serve two purposes but also because it includes a running mean superimposed on the "raw" data. The running mean I think helps to illustrate the folly of looking at individual data points.  I'm not sure how it can be simplified while preserving necessary content but if you've got a notion of how that might be done I'd be delighted to see it (or hear of it!). 

2010-09-04 00:59:35Comment
Robert Way

robert_way19@hotmail...
142.162.188.138
The only thing I could do is take the data and just plot the minimum sea ice extents for each year over that time period. If that is sufficient let me know and ill do it in a couple of minutes.
2010-09-04 01:01:54Comment
Robert Way

robert_way19@hotmail...
142.162.188.138
Hate to harp on the subject but there are those who say that arctic ice extent does include the AMO cycle in it as the AMO warms the arctic. We know sea ice extents have gone outside that natural variability now but it explains why there was somewhat low sea ice in the arctic during the early 20th...
2010-09-04 01:35:03
doug_bostrom

dbostrom@clearwire...
184.77.83.151

Trying to keep my defensive hat off, I think explicit treatment of the AMO as well as other factors will need treatment at a higher level. The second section of this rebuttal skates around delving into an explosion of jargon. Even so I can see now on rereading I'm going to have to elaborate on sediment cores a bit. For this level of discussion, following the links to Overpeck and Johannessen will produce a rich efflorescence of details.

Remember, this is for "normal" people, not case-hardened skeptics armed with all the arcana. 


2010-09-04 02:44:14agree with doug
dana1981
Dana Nuccitelli
dana1981@yahoo...
38.223.231.249
I agree, delving into the AMO would be difficult without going beyond a 'basic' level here.  It's something that could be done for the advanced version, for sure.
2010-09-04 02:47:17
doug_bostrom

dbostrom@clearwire...
184.77.83.151

I'm thinking about Dana's suggestion to highlight volume. 

I touch on volume briefly here, noting that reduction of volume is what ultimately makes end of melt anomalies so conspicuous.

Meanwhile, this rebuttal/explanation is principally about past natural cycles of Arctic sea ice and how they're irrelevant to today. We've got proxies beyond visual observation for past extent and so can talk about past variations in extent based on research findings. On the other hand, I'm having a hard time w/extending this argument regarding cycles or past variations to talk about volume because I don't see anything in the literature speaking to past volume of Arctic ice. 

I could sneak in an amplification of my reference to diminishing volume sensitizing Arctic sea ice to reduction in extent, slip in the PIOMAS graph there.  

Any suggestions? 


2010-09-04 03:04:53volume
dana1981
Dana Nuccitelli
dana1981@yahoo...
38.223.231.249

You could add that based on the rate of decrease in volume, we're headed toward an ice-free Arctic (where "ice-free" is defined as more than an 80% drop from the 1979-2000 summer volume baseline) within the next few years.  Here's a useful plot from a presentation by Maslowski et al. in 2010.

Malowski

And that as discussed in a 2005 EOS article by Overpeck et al., this is outside the range of natural variation.

"The Arctic system is moving toward a new state that falls outside the envelope of glacial-interglacial fluctuations that prevailed during recent Earth history.  This future Arctic is likely to have dramatically less permanent ice than exists at present. At the present rate of change, a summer ice-free Arctic Ocean within a century is a real possibility, a state not witnessed for at least a million years."
2010-09-04 03:07:18Comment
Robert Way

robert_way19@hotmail...
142.162.188.138
I wasnt suggesting diving into the AMO or diving into the different contributors. Just making a note of it I guess. Im thinking that since the extent graph goes back further it might be preferable to use that... just a thought though... without volumetric measurements even spanning 50 years how can we make any assumptions supposedly...
2010-09-04 04:10:08extent vs. volume
dana1981
Dana Nuccitelli
dana1981@yahoo...
38.223.231.249

Well we have less than 60 years' worth of extent data vs. 30 years of volume data.  Neither is sufficient to rule out 'natural cycles' by itself, which is why you analyze it further, as in the Overpeck article I linked.  I think it's worth showing the volume data, because extent is more variable since the surface water freezes relatively easily, somewhat masking the rapid decline in volume.  For example, if you try to predict when the Arctic will be 'ice free', you would guess a much later date if you based it on extent than if you did on volume (as Maslowski shows).

It's not a huge difference and I could understand leaving the volume out.  But then again, 'skeptics' love focusing on the sea ice extent data and ignoring the volume data, precisely because the extent is more variable and can be used to argue the sea ice is 'recovering'.  So I like referencing the volume data whenever possible.

2010-09-04 04:41:10
doug_bostrom

dbostrom@clearwire...
184.77.83.151

Fortunately the extent issue is adequately treated by proxy, enough so that Overpeck felt comfortable making comparisons way back in '97.

Meanwhile, Dana has pointed out Overpeck 2005, which I'd missed. That speaks specifically of volume anomaly being outside natural variation, Overpeck is an authority* on this subject so now the course becomes clear. Thank you, Dana!

*I -do- subscribe to the notion that authority exists, JFTR. Anyone referring to "argument by authority" is perfectly free to show how the current authority is wrong, of course.  Also for the record I'm incredibly grouchy right now because I've got some kind of intricate back problem beyond the reach of reasonable narcotic treatment and thus am getting by on about 4 hours' sleep each night. Apologies in advance and meanwhile I'm staying away from comment threads on the main site... :-)

 

 

2010-09-04 17:57:37Is it me?
gpwayne
Graham Wayne
graham@gpwayne...
81.152.234.172

Sorry Doug, don't mean to add to your troubles (and hope you feel better very soon) - but this is the third time today I've felt obliged to suggest a rebuttal is too complex. Can't avoid it, and no point in holding back. I really don't mean to be unconstructive, but my measure of these things - believe it or not - is whether my 84 year old dad would understand the rebuttal, and I don't think he'd get far with this one.

2010-09-04 22:59:43
doug_bostrom

dbostrom@clearwire...
184.77.83.151

Taken on board, Graham. The detailed explanation of various ice metrics is definitely a "power windows and heated mirrors" bit and I was also wondering about the need to specifically name proxies, which of course is useless without even more elaboration. 

 

 

2010-09-05 18:56:54A question
gpwayne
Graham Wayne
graham@gpwayne...
81.152.234.172

I've only just figured out that if you post these rebuttals through the same list where you claim them, you get the 'summary' box on the page - I've been posting them in the author's list, so I didn't do summaries. May need to return to that, or not.

 Anyway, here's my thoughts:

"Warming of the planet is expected to affect upper northern latitudes of the Earth more and sooner;"

You could make this more positive e.g. "As the computer models predicted, the upper northern latitudes are already being clearly affected;" - which is both true, and reinforces the power of the models.

"just as rapidly warming beer in a glass tells us it's a hot day".

Surely better to use ice in a drink, which is also a visual clue to the hot day?

As for elaboration, it's where our good intentions often lead, isn't it?

2010-09-05 23:57:21Some typos
BaerbelW

baerbel-for-350@email...
93.231.152.86

Hi Doug,

here are some typos I spotted:

3rd paragraph of full rebuttal
"Looking at the ice extent graph below, it's possible" --> at is missing

2nd paragraph below chart:
"...reveal different reasons for yesterday's and today's Arctic sea ice changes "

Cheers
Baerbel

2010-09-06 05:51:52
doug_bostrom

dbostrom@clearwire...
184.77.83.151

Thanks guys, suggestions/fixes incorporated. 

Graham I decided to deep-six the analogy w/drinks but thanks for the suggestion anyway. 

2010-09-06 16:56:30Thumbs up
gpwayne
Graham Wayne
graham@gpwayne...
81.152.234.172
OK Doug - nice work in my opinion, and a thumb from me...
2010-09-09 04:34:46approve
nealjking

nealjking@gmail...
84.151.34.204
gt
2010-09-10 06:26:18Rebut 22 Arctic Ice
jimalakirti

jimalakirti@gmail...
71.34.142.115
Looks good to me.
2010-09-10 13:59:22Comment
Robert Way

robert_way19@hotmail...
142.162.205.242
Hello all,
I was thinking the graph seems a little complicated to me for a basic rebuttal (big words and running means). Beyond that though I think it is good to go.


2010-09-27 00:49:54
doug_bostrom

dbostrom@clearwire...
184.77.83.151
It's not really cricket but this has gradually subsided into the sediment so I'm bumping it myself to see if there are any further suggestions or the like.
2010-09-27 02:30:06
Anne-Marie Blackburn
Anne-Marie Blackburn
bioluminescence@hotmail.co...
80.42.215.97
Clear, comprehensive rebuttal. And for me personally, some new information on past changes in the Arctic.
2010-09-27 07:57:51
Riccardo

riccardoreitano@tiscali...
93.147.82.165
Good move Doug :)
2010-09-28 10:27:27Looks good to go, just one question...
John Cook

john@skepticalscience...
124.185.151.34
Would it at all be possible to replot doug's graph to include the 2010 September minimum? Visually, one might conclude from that graph (superficially) that Arctic ice is rebounding. Of course, extent continued it's downward passage in 2010 so I think that graph would be better if it could include the latest data. Robert, is that feasible? (I assume we'd have to wait a few days for September to finish but I also wonder if it's a problem of having different datasets - trying to put a new wine into an old wineskin, so to speak)
2010-09-28 11:16:28
doug_bostrom

dbostrom@clearwire...
184.77.83.151

"...include the 2010 September minimum? "

 

Yes, good idea! How about I do it using Gnuplot as I've been touting that package's virtues? Give me a couple of days, I've got some confounding factors going on here.  

2010-09-28 18:59:14
Anne-Marie Blackburn
Anne-Marie Blackburn
bioluminescence@hotmail.co...
80.42.215.97
Doug, as you seem to know what you're talking about, how difficult is Gnuplot to use? I've been teaching myself how to use R, and with the right book/tutorials, I'm getting somewhere. Would you say Gnuplot is similar in terms of difficulty?
2010-09-29 01:33:53
doug_bostrom

dbostrom@clearwire...
184.77.83.151

Gnuplot is pretty straightforward, is dedicated to general plotting objectives using both scripted expressions as well as externally supplied data, as opposed to the deep intrinsic statistical centering (hah!) of R. The statistics thrust of R is revealed in its graphical output; R has a richer set of visualization tools for stats. 

If you've got momentum w/R, I'd stick with that although I'm speaking through my hat a bit as I've not cracked R. Others convey that R is difficult but if your objective is statistical power GnuPlot is not the tool you want. Cranking out the expressions you'd need in gnuplot to reproduce R's kit would be too cumbersome, not a productive way to spend time.

For mile wide, inch deepers such as myself gnuplot fits my labor allocation budget. :-) 

2010-09-30 21:38:12
Shirley_Rocks
Shirley Pulawski
missfabulous@verizon...
96.243.52.109

FYI the Polyak link doesn't work. I think Science Direct likes to move things about - I've had problems with bookmarks in the past. 

 

I think the rebuttal is well written and clear, but I think it's long for a basic rebuttal. I even think the Summary could be reduced to the last two sentences, or a revision thereof. I think the "canary in the coal mine" might prejudice some readers, the same way the word consensus might. Maybe the same for "... fall of a cliff" - to those looking for "alarmist" buzz words, they might find it there. After watching my sister's eyes dart around avoiding what I was saying but looking for anything in what I was saying she could use against the argument for reality, I've grown to understand just how some people cherry pick. I think some of the sentences are duplicative, too. I've got to head off to campus now, but if I have a chance later, I'll re-visit and make some specific suggestions. Again, I think it's otherwise clear, I would just work at chopping it down to about half its length, and get the graph up higher. 

2010-10-01 02:00:33
doug_bostrom

dbostrom@clearwire...
184.77.83.151

Thanks for pointing out the URL glitch, Shirley. Happily Polyak's paper is also now available in full text form. I added also another Polyak-lead paper w/more background on proxies. 

I hear your concerns about length. In general I'm not writing for die-hard skeptics as that's pointless, rather my objective is inoculation.  That being the case, you'll definitely see a little additional freight here. It's true that I've a tendency to verbosity; I'll give this another close look with an eye to that. 

2010-10-01 14:56:35
doug_bostrom

dbostrom@clearwire...
184.77.83.151

Bit of a dilemma w/ the promised chart replacement.

Sniffing around for data and methods, turns out it's not really going to be possible to capture w/a replot including 2010 the same presentation as the Meier and Stroeve chart from NSIDC now shown in the article. That's a spliced presentation and the method of construction is not fully described. I'm not qualified to create such a splice myself.

I kind of hate to lose it. We're used to the iconic satellite-derived anomaly chart which begins in 1979, but M&S show an obvious beginning of a trend downward a full ten years earlier. Again, this is a spliced presentation, but apparently it's sufficiently defensible as to be presented by NSIDC, on the bull's eye sea ice page for skeptics.

So the choice: short chart so as to get an extra year on the end while dumping the first 26 years, or forget about 2010 and use the existing long chart.  

I suppose fresher data is better. 

2010-10-01 15:12:43
doug_bostrom

dbostrom@clearwire...
184.77.83.151
I should add, the value of that extra 9 or 10 years lies in how the decline appears to have started prior to satellite observations, suggesting the stability of the first 18 or so years of the graph was real. That apparent stability in itself suggests we're not in the middle of a natural cycle that is visible on a decadal scale. 
2010-10-03 06:18:46
doug_bostrom

dbostrom@clearwire...
184.77.83.151

Significantly shortened, focusing more tightly on the core "cycle" matter as opposed to providing a general explanation of the situation in the Arctic. 

Because the Meier graph better addresses the cyclic issue than the typical satellite-derived anomaly graph, I'm leaning to keeping it. It's more communicative of the matter at hand.  

2010-10-03 12:34:35Meier graph
John Cook

john@skepticalscience...
124.186.160.198
This rebuttal is good to go, I'm just waiting on whether we can get an updated version of Meier's graph with the 2010 minimum. shall I keep waiting? Eg - is this a possibility?
2010-10-03 14:23:15
doug_bostrom

dbostrom@clearwire...
184.77.83.151

As the Meier graph only goes through July of 2009 I don't think it's routinely updated, John.

What about making the words "...reveal a rapid decline over the past 30 years..." into a link to pop up a child window containing the updated satellite anomaly graph? A compromise...

2010-10-03 15:59:28Suggestions to simplify language
James Wight

jameswight@southernphone.com...
58.105.164.221

I'll give this the thumbs up though I do think some of the language could be made a bit simpler. Here are a few suggestions.

Change “The increasing seasonal ice coverage swing with a declining trend is…” to “The downward trend, and the increasing difference between seasons, are…”

Change “As air and ocean temperature conditions in the Arctic warm…” to “As the Arctic warms…”

Change “unique in character and promoted by a new mechanism” to “caused by a new mechanism”.

Change “introduced into climate behavior” to “introduced into the climate”.

Also, I think the summaries for basic rebuttals are supposed to be one-liners. (John, correct me if I’m wrong.)

2010-10-03 16:00:37Sea ice volume trends
James Wight

jameswight@southernphone.com...
58.105.164.221
Dana1981, that graph of Arctic sea ice volume seems to contradict the one from the Polar Science Center. The Maslowski graph has the ice remaining stable until circa 1998 then going into a “death spiral”, but the PSC one shows it declining steadily until circa 2001 then accelerating. Any idea what is the reason for the discrepancy?
2010-10-03 23:47:29
doug_bostrom

dbostrom@clearwire...
184.77.83.151

Thanks James, suggestions taken on board.