Guest Post by Willis Eschenbach
There’s a paper out by, inter alios, our good friend Judith Curry. The paper is “Impact of Declining Arctic Sea Ice on Winter Snowfall”, by Jiping Liu, Judith A. Curry, Huijun Wang, Mirong Song, and Radley M. Horton (PDF, hereinafter L2012). Judith has a thread discussing the paper at her excellent blog. Their claim is that reducing Arctic sea ice leads to heavier winter snowfall. Inherently, this seems to make sense. Less ice means more evaporation, and because what goes up must come down, more evaporation means more snow. Case closed … or not …
Unfortunately, the paper doesn’t live up to its promise. Oh, it has lots of pretty pictures. Here’s one of them:
Figure 1. According to L2012, this shows the difference between the outputs of two climate model runs. I would call this is pretty conclusive evidence, perhaps even the long-sought “smoking gun”, that clearly establishes that the two climate model runs were indeed different.
Here’s what their abstract has to say (emphasis mine):
Abstract
While the Arctic region has been warming strongly in recent decades, anomalously large snowfall in recent winters has affected large parts of North America, Europe, and East Asia. Here we demonstrate that the decrease in autumn Arctic sea ice area is linked to changes in the winter Northern Hemisphere atmospheric circulation that have some resemblance to the negative phase of the winter Arctic Oscillation. However, the atmospheric circulation change linked to the reduction of sea ice shows much broader meridional meanders in mid-latitudes and clearly different interannual variability than the classical Arctic Oscillation. This circulation change results in more frequent episodes of blocking patterns that lead to increased cold surges over large parts of northern continents. Moreover, the increase in atmospheric water vapor content in the Arctic region during late autumn and winter driven locally by the reduction of sea ice provides enhanced moisture sources, supporting increased heavy snowfall in Europe during early winter, and the northeastern and mid-west United States during winter. We conclude that the recent decline of Arctic sea ice has played a critical role in recent cold and snowy winters.
So … what’s not to like? Reduced ice causes cold surges, leading to more snowfall. Case closed … or not …
For me, the first clue that something is wrong in a study is often that they don’t start out with a historical look as far back as the records may go. In this case, the satellite ice area as records go back to 1978. But in this study, they only show snow records going back as far as the antediluvian year of 2007/2008 … at that point, the bells started ringing for me. I always start with the longest overview of the question that I can find.
So let me remedy that, and we can see if declining sea ice really does lead to cold, snowy winters. The upper panel of Figure 2 shows the actual ice and snow data (normalized to an average of zero and a standard deviation of one). Below that, the lower panel shows the anomalies in those same normalized datasets once the monthly averages have been removed.
Figure 2. Arctic sea ice area (blue) and Northern Hemisphere snow area (red). Upper panel shows actual data. Lower panel shows the anomalies of the same data, with the same units (note different scales). The R^2 of the snow and ice anomalies is 0.01, meaninglessly small. The R^2 of the first differences of the anomalies is 0.004, equally insignificant. Neither of these are significantly improved by lags of up to ± 6 months. SNOW DATA ICE DATA
I’m not going to say a whole lot about this graph. It is clear that in general the arctic ice area has been decreasing for twenty years or so. It is equally clear that the northern hemisphere snowfall has not been increasing for the last twenty years. Finally, it is clear that there is no statistical relationship between decreased ice and increased snow.
I will leave it to the reader to decide if, as the authors of L2012 say in the Abstract, ” the recent decline of Arctic sea ice has played a critical role in recent cold and snowy winters.” I certainly don’t see it in the historical record. And this is why graphing the full record of both variables is so important. There may be some effect there … but if so, it is a very small effect, it’s invisible at this level.
In a more general sense, I see this as studying “how many snow-storms can dance on the head of an iceberg”. There have been no breakthroughs in climate science in thirty years, and I can see that the people searching for the “smoking gun” establishing a “human fingerprint” are getting mighty frustrated. But that is no reason to give up on the important questions and work on this kind of trivia. If there were a significant effect of decreasing ice causing increasing snow area, it would be visible in Figure 2. So at best, they are studying some tiny, third-order phenomenon. There’s nothing wrong with doing that once a field has no big questions left unanswered.
The thing is, climate science is nothing but unanswered questions, big questions. And until those questions have answers, for them to be wasting their valuable time and their trained scientific curiosity on this kind of small potatoes?
I suppose it’s meaningful in some universe … not mine.
w.
PS—The authors do deserve kudos, however. The paper nowhere contains the words “human influence”, “AGW”, “anthropogenic”, or “CO2”. That alone is shocking enough that it should get a medal of some kind.
PPS—Joe D’Aleo discussed the L2012 paper on WUWT here. Unfortunately, he didn’t show a direct comparison between ice and snow either.
PPPS—The title is from Mae West, who said “I used to be Snow White … but I drifted.”
Judging by the lower graph, it is absolutely clear to any sceptic and the AGW obscurantist alike, that snow fall in the winter is highly correlated to the mid summer temperatures. No doubt.
http://www.vukcevic.talktalk.net/CET-Jun.htm
🙂
The thing is melting ice in the Arctic is not new. Happens all the time, its sea ice, don’t they know?
It melts in summer and closes over in the other months when there is little sun to warm anything.
I thought Willis had left something out of his CV…
It reminded of that old comics, a bald mustachioed guy sitting in a bar saying to a lady “You cannot imagine how strange my life has been”. “Oh, I’ve seen many things”, she replies, “nothing can ever surprise me”.
To that, the guy counters: “It all started when I was a little girl this tall…”
NSW is currently underwater. Maybe that is where the artic sea ice has gone. Or maybe to the ski resorts of Colorado-perhaps they should be subsidising research into elevating global warming. Or maybe the sea level has stopped rising because it is now transferred from the artic ice to fill NSW dams. Or maybe, just maybe, a couple of slightly more snowy winters has nothing at all to do with any of them.
Can one get past peer review with a boring conclusion, such as: no correlation, no meaning and no relevance?
T
Willis: The paper nowhere contains the words “human influence”, “AGW”, “anthropogenic”, or “CO2″.
No need to. The faithful NYT, Guardian, et al, will run with it and fill in the blanks as they see fit. Good for circulation, and L2012 authors get their name in print while being able to avoid any direct flack for alarmism. It’s a win-win.
Doesn’t make sense to me.
Go back to the 70s and see that sea ice was not diminishing, but big amounts of snow were falling.
Gosh, I could figure out about 10 papers in the following hour. One would be how the sea ice extent correlates with stock market. I would also start in 2007/2008…
Ecotretas
Nonetheless, Willis, despite the award-winning absence of meme terminology, it still says “warming means cooling”…but not in so many words. It’s exhausting.
Maybe I’m seeing things that aren’t there, but doesn’t it look, in the bottom panel, as if the snow area anomaly is leading the ice area anomaly (by a few months)? If so, that would not really support causation in the direction that the authors suggest.
Wilis,
May I state the ‘journalists code’ here which applies here?
“Why let facts get in the way of a good story.”
And after all they are academics so have got to ‘publish or be damned’. Well done for showing that this is yet another example of poor/failed peer (pal) review and ‘computer model science’ (The Emperor) trying to pretend to be real science (‘have clothes’).
Regards
KevinUK
Inter alia is Latin for “among other things”; when referring to human authors, the correct (and more respectful) Latin is inter alios (if they’re masculine or a mix of masculine and feminine) or inter alias (if they’re all feminine).
[Thanks, fixed. Always more to learn. -w]
Well, your analysis ignores that the largest drop in sea ice area occurred in 2007 and the has never really recovered, thus your analysis seems the simplistic one here.
Having read most or all of your and her posts for years, I pretty clearly pick her analysis as being more detailed, more thoughtful, and based on a much deeper understanding of science as well as her having more knowledge in general. This post included.
If I were Mr. Watts I would pull this simplistic post, almost childish in its analysis.
Brad – I’m sorry to hear you have been order to read WUWT by your doctor. Best wishes for a speed recovery!
So we must have created HUGE amounts of sea ice in the last 12 months because Minnesota is having one of its warmest winters with next to no snowfall. Or… something like that anyway.
Brad says:
March 6, 2012 at 1:24 am
If the theory advanced by the paper were correct, that 2007 drop in sea ice (which contrary to your claim is clearly visible in my graph) would be accompanied by a large increase in snowfall. I don’t see that. If you see that big snowfall increase, point it out.
So your conclusion is that, despite there being no visible correlation of any kind between snow and ice variations, that she has a “deeper understanding of science” … she may well have that, Brad, she’s a very sharp woman.
But the relative level of our individual science-fu is not the question here. The question is, is there actually an effect where a drop in ice area leads to an increase in snow? That doesn’t depend on either her or my understanding, deep or shallow. If you see evidence of such an effect, you’ll have to point it out to me. I don’t see it. Surely, if decreasing ice meant increasing snow, it would be visible in the record … so point it out.
Y’know, Brad, with that unpleasant accusation you sound a lot like the lady who said to Winston Churchill, “If you were my husband, I’d put arsenic in your tea.”
To which he replied, “Madame, if I were your husband, I’d drink it.”
w.
PS—You make a lot of accusations, Brad. You call my analysis “almost childish” … yet you have singularly neglected to find one single tiny thing wrong with it.
I notice that with folks a lot. The more the bombast, the more the ugly accusations, the less actual science in the person’s post.
So congratulations, you’ve taken that trend to its logical extreme.
Brad says:
March 6, 2012 at 1:24 am
You missed Willis’s point: come back in 10 or 20 years time and see what has happened to ice extent. None of us know the future but looking at a longer historical time frame now makes sense to this analytical geezer.
I’ll leave those better able than me to determine whether Curry or Eschenbach is the “better” and if you are more perceptive than most here. I will continue to read both since Ms. C became a little less alarmist.
Finally, whether one thinks a particular person in the debate is better than another particular person also misses the point: we need to look at the science as a whole and when I do I find myself totally on the real science side and highly sceptical about The Team and The Cause. Did I just come up with a better definition of a sceptic?
Frank DJ says:
March 6, 2012 at 1:13 am
The best correlation ( R^2 = 0.06, still meaningless) is with the snow leading the ice by three months, as you suggest … but the correlation is both positive and in the wrong direction.
The strongest negative correlation ( R^2 = 0.001 ) is with ice leading the snow by five months. But that’s so small it’s a joke.
None of that, not a bit of it, is statistically significant. I find no significant correlation anywhere, no matter which one leads or lags.
w.
Nice try Willis, but Brad has nailed you. He picked up right away on the fact that you didn’t have four co-authors, garner a $120,000 NSF grant, and spend eight months gluing your post together.
Poor Liu, J. et al. Gotta be discouraging to go to all that effort to publish a paper, then have somebody poke a hole in it with a couple easily assembled charts.
Brad,
Well you are not Anthony and this is not your blog so tough!
I’ve been on WUWT since the very beginning and sorry mate but I’ve never read any of your comments here on WUWT. Unlike you I have read pretty much all of Willis’s threads here on WUWT (I’ve even re-constructed a number of his analyses) and they are always well written and very much to the point. This one is no exception. Willis has clearly pointed out what happens when you cherry pick data in order to suit your pre-written conclusions.
Do you understand the principle of falsifiablity Brad? Willis has clearly shown that when you use all of the available data there is clearly NO correlation between Artic sea ice area and NH snowfall area. The L2012 conclusions are thus falsified.
KevinUK
LOL. Indeed.
And at least one of them must be wrong.
Willis, why do you think that they did not study the whole data range from 1979 to 2010?
What happens to your “analysis” if you use winter snow data only?
Even 34 years data may not be enough given the probable 80 year Arctic cycle but Fig 2 does demonstrate little to zero connection. The UK has had a mild winter, contrary to that of the rest of Europe, and the UK Met. Office claim that this probably proves a warming planet.
The Arctic ice is now increasing, a natural cyclic event.
Deadman says: March 6, 2012 at 1:19 am
Inter alia is Latin for “among other things”; when referring to human authors, the correct (and more respectful) Latin is inter alios (if they’re masculine or a mix of masculine and feminine) or inter alias (if they’re all feminine).
That must be lawyer Latin.
Best I can come up with is inter aliae for a dative feminine plural.
Yes Willis, you are right, the paper is flimsy. Judith is on the defensive over this paper. In her defence she does appear to relate it to weather phenomena, rather than an assault on causes of decreasing cover. She has previous papers in this regard that could also be accounted for in the mix.
http://judithcurry.com/2012/03/05/impact-of-declining-arctic-sea-ice-on-winter-snowfall/#comment-181922
I have problems with this argument. It may just mean less freezing.
It seems to me that there is not much evaporation at the poles. Overall, evaporation happens elsewhere, but not at the poles. Just an impression, though.
Brad says:
March 6, 2012 at 1:24 am
Well, your analysis ignores that the largest drop in sea ice area occurred in 2007 and the has never really recovered,
Beware the ice of March!
Well done Willis. I thought the article was pretty weak when I saw it. In the UK our cold weather (and Europe’s at the same time) is brought about by negative AO and NAO but in my experience this is random and to link the cold spells of the recent winters to the drop in Arctic ice was very simplistic and lacked the longer view. Thank you for adding perspective and showing it to be spurious.
But…but…the MODEL was significant. We don’t need the NWS telling us what the weather is out there. Or how snowy it will be next winter. Heck, I don’t even need my finger to tell me which way the wind is blowing. We have models now. Sounds familiar doesn’t it. Burn the books and don’t call your folks anymore. Just watch your walls.
‘Y’know, Brad, with that unpleasant accusation you sound a lot like the lady who said to Winston Churchill, “If you were my husband, I’d put arsenic in your tea.”’
To which he replied, “Madame, if I were your husband, I’d drink it.”
Willis, it is a good thing for an uneducated man to read books of quotations.”
Excellent ride by shooting, Brad, like Audy Murphy on speed. Shame you missed, but the camera angles will no doubt change that to a hit in your mind. Now get back on stage in the Rocky Horror show where you so obviously belong.
I must say I have every sympathy with Dr. Curry. When the science is settled and the debate is over and the consensus is in there is not much left for the good doctor to do.
I am a great admirer of her blog and her honest approach to science. I get two impressions from her paper, firstly that it started out as a conclusion looking for data to fit and it finished up as a “weather” paper rather than a “climate” paper because the long term data do not correlate but the short term “weather” data do have a measure of agreement with the conclusion.
I still advise her and everyone to ask the old folks about the actual weather and climate over the last 70 years. Look out the window occasionally. The world has not ended.
Add extra commas and parenthesis, double-dashes to your taste.
I would like to see more articles that contain less of “what the other side said”.
That meme is getting rather boring. Could you please write an article without interjections like “warmists” and “then he/she said” and the like? Just let the data and the science speak.
This blog is getting worse by the minute, stooping to a new low in every article.
I used to like to come here everyday, not so anymore. This once fine blog is drifting towards school yard bullying.
I have to say, I think Willis makes a good point here and I would love to have Judith’s take on it. However;
– it may be that the correlation is not direct. For the circulation pattern to change may require – dare I say it – a tipping point. That is, the needs to be sufficient warming to cause the pattern to switch. If that is the case you wouldn’t see a direct correlation.
– the authors ought really have acknowledged the uncertainty. Considering this is Judith’s main drum she quite rightly beats, I do feel the abstract implies greater certainty than is justified. But the mechanism itself seems quite plausible.
The whole conjecture of declining sea ice = more snow fall, fails the sniff test.
Firstly, snow is caused when warm, moist air from the lower lattitudes collides with cold air masses from the higher lattitudes, not the other way round.
Secondly, the amount of water vapour that could be carried in frigidly cold arctic winter air, in perpetual winter darkness, must be very small indeed. Have they even measured it? Have they done basic calculations to check if there is sufficient water vapour to create several feet of snow?
Thirdly, the conjecture implies that if there is a large expanse of arctic sea ice, the snowfall would be very little, which clearly does not correlate with historical records.
Willis has done a nice job of showing the lack of correlations, and the cherry picked data, but he is really delivering the coup de grace to an already dead bull. This is just so much junk science I am flabbergasted that Judith would put her name to it.
Willis, two other thoughts:
1. Your graph is showing sea ice area and snow extent.
It may be that it is not the extent that is important but the total snowfall. That is, the volume of snow that falls in total as a function of ice extent rather than how far what snow that does fall extends to.
2. One point made in the paper is that it is autumn sea ice extent that is important for this effect. Maybe it is the autumn extent that may correlate better to winter NH snowfall?
It seems to me that before you can say what causes a certain climatic phenomenon, you need to prove that natural causes are not responsible.
I cannot see how this can be done, unless you first are able to explain the action of natural causes in the past.
In this case, have the authors explained why some winters in the past have been cold and snowy, while others were not?
If” arctic decline ” makes for cold winters, then 2007 should have been the most brutal winter in history. Indeed it wasn’t.
At that stage their argument was the opposite, that winters would be a thing of the past.
In either scenario, they are not making a fair assessment and both cases are fairly erroneous
Just one question, Willis – how does snow area relate to snow volume, is it similar to ‘ice area/ice volume’, and are either confidently measured? Three, as it happens.
Congratulations Willis, your science is excellent. I would not say the same about the authors of the paper, as it looks like cherry picking to me.
Something can be simple and profound at the same time. Just because something is easy to understand, doesn’t mean it isn’t right. Just because something is hard to understand, doesn’t mean it is right. Always, dear Brad, remember the story of the Emperor’s New Clothes. Anyway, one should not trust people just because they are experts. As Philip Tetlock has pointed out, experts are extremely fallible.
Willis has presented a question that Judith et al. will have to deal with if they want their paper to retain its credibility. They can probably get another paper (maybe even a research grant) out of it. 😉
When I view the incredibly complex maze of sophisticated modeling and analysis that has led to our current sorry state of “climate science”, I find “simplistic” and “almost childish” to be terms of endearment.
Willis.
Here in Iceland we have had mild winters with little snow for the past several years. At least in South-Iceland. This winter was a little different however in Reykjavik as we had snow on the ground for 60 days in December and January. November and February have been mild and more or less without snow.
Just a small point.
If you look at the NORSEX Sea Ice Area and Extent graphs (link on the right at the Sea Ice Page) both graphs show that, far from reducing, receding, or doing all the other things which the alarmists would have us believe, Arctic sea ice seems to be heading very smartly towards the 1979-2006 level.
Where this leaves the ‘Less sea ice, more snow’ argument I’m not sure – but what I AM sure of, is that sea ice in the Arctic is doing what it always does – increasing until the middle of March.
Willis,
I have two questions for you.
1. Do you believe that, if less Arctic ice leads to greater snowfall, there should be a direct year to year correlation between ice extent and snow (2007 ice extent low, 2007 snow high);
2. If winter has more snow, shouldn’t summer have more rain?
3. Isn’t summer ice extent meaningless if moisture is being taken up during the winter months over open ice and then deposited (almost immediately) on land?
4. Is it possible to store moisture for more than one year?
5. Isn’t winter ice extent fairly constant in the Arctic (not as extreme as summer ice)?
Steve
….just to add:
http://judithcurry.com/2012/03/05/impact-of-declining-arctic-sea-ice-on-winter-snowfall/#comment-182157
Dr Curry points out much the same thing, that the paper states nothing quantitatively about cover. Also, the effect “contributes” to changes in circulation patterns, which is not suggesting that it is responsible for them.
I ought to point out too, that I think the abstract too certain, but the discussion contains sufficient equivocations. I think Dr Curry agrees with me on that too.
I always believe in the conservation of energy:- if ice melts in the arctic causing precipitation of snow in the northern hemisphere then unless the relative humidity in the air in the north hemisphere increases there is no net energy gain: latent heat of melting + evaporation must equal the latent heat of condensation plus the latent heat of freezing – so no net energy increase just a redistribution QED
Agnostic says:
March 6, 2012 at 3:28 am
Point 2 is important (now that I’ve read the paper).
But this paper is easy to confirm. From their Fig. 3 they show four years of autumn ice extent with 2007, 2009 and 2010 having almost exactly the same (lower) extent and 2008 showing a distinct higher extent. So was 2008 snowfall normal and 2007, 2009 and 2011 above normal? If so then case closed on a good paper.
…..and, “The snowiest two year period on record (1978-1979) also had the peak Arctic sea ice of the last century.”
real life contradicts this paper………….
http://www.real-science.com/georgia-tech-study-takes-goal
The discourse here is rather funny. If anyone disagrees or says anything negative they are completely open to attack by everyone from the mods to the board owner, but vice versa is absolutely not allowed. It is simply an interesting insight into the mind of true believers, they really do not want the truth, they long to have their deeply held belief confirmed at all costs.
It is tantamount to insanity, and you guys are no better then Mann or Gleick.
Willis, concerning Brad’s accusation of this being a childish post: “The wolf will live with the lamb, the leopard will lie down with the goat, the calf and the lion and the yearling together; and a little child will lead them.” Isaiah 11:6. A picture peace and understanding in the millennial kingdom. Sometimes it is very simple.
As Willis logically concluded, it is not good to see such intellectual talent diverted. Much grander truths need to be known about earth’s climate system.
Understanding first the historic record and natural variability should be foremost in climate analysis. Rather than current AGW bias conclusions.
Clearly, the science is not settled, the debate is not over, and there is no concensus.
Thank you Willis for your enlightening post.
I don’t trust Judith Curry anymore. The best way to describe how I feel about her approach to the whole climate discussion at this point: totally creeped out. I would advise journalists to steer clear of her opinions.
Sincerely.
Time series of Ice to snow. By ignoring the daily humidity exchange and looking at the seasonal numbers it seems one might miss what really happened.
While you are at it, why don’t you try to refute the actual conclusion of the paper, instead of the strawman you have thrown up. BTW – I assume she has the basic correlations run here, any descent scientist would have, and Dr. Curry is.
The real conclusion:
“supporting increased heavy snowfall in Europe during early winter, and the northeastern and mid-west United States during winter. We conclude that the recent decline of Arctic sea ice has played a critical role in recent cold and snowy winters.”
Do you have evidence refuting the snowfall statement in early winter? If you have not addressed this question (your post is completely irrelevant to it) then you have not addressed the paper.
Thanks, you apology is accepted.
A few quotes from the paper -” potential contributor – it is more likely – the likelihood – suggests –
might be potential contributor – suggests that the moisture source for these regions” Not convincing language. Following on this approach “data” is then confidently run through various computer models.
It is claimed that Arctic humidity is greatly increased but I could not find any supporting data. (It may be there: it wasn’t too long before I started jump reading.) Does anyone here have real data for that?
Sorry, also the total snowfall in those particular regions, not in general.
I find those that can do science do not throw up basic correlations irrelevant to the point at hand – at least after their first or second year in grad school.
[snip . . repeated post . . kbmod]
In the list of authors, Radley Horton from Columbia University is mentioned. Isn’t modeler and RC host Gavin Schmidt there also? Did the L2012 authors use Schmidt’s model, hence, the outcome a foregone conclusion? It makes me wonder.
The authors do deserve kudos, however. The paper nowhere contains the words “human influence”, “AGW”, “anthropogenic”, or “CO2″. That alone is shocking enough that it should get a medal of some kind.
This because, were the paper correct in its conclusions, it would be clear evidence of climate cooling (increased ocean heat loss), with a positive (cooling) feedback (increased snow covered land albedo).
Otherwise excellent debunking of a very weak paper.
Dr. Curry just uploaded the final version of the article. This shows now the actual figures and not the supplementary figures. The figures were not included in the earlier version, only the supplementary ones.
Maybe Willis should now read again the article and then try to make a proper comment about it and stop misleading the people here with false science.
Mike McMillan says:
March 6, 2012 at 2:28 am
“[Inter alios] must be lawyer Latin.
Best I can come up with is inter aliae for a dative feminine plural.”
Low though my tolerance may be for using Latin–and for the pendantry that so often accompanies it–when a perfectly good English phrase will serve, I’ll throw in this relic of my misspent youth: “alios” and “alias” are indeed the accusative plurals, and “inter” takes the accusative, so Deadman is correct. (And I believe that the dative plural is “aliis.”)
Agnostic seems to have got this mostly right: you’ve done the wrong analysis. Curry doesn’t claim a relation between sea ice and total snow extent, or total snow volume. You can see this from the pic you’ve taken from here paper: there is as much red as blue. What Curry is claiming is *changes* in the snowfall pattern.
One excellent article above and one below the one that promises the “Matt Ridley Prize for Environmental Heresy.
Hope you are planning to frame it. – You’re worth every penny of the prize – so go for it- its as good as in the bag!
My reading of Brad’s first post is that he just didn’t know that a correlation between snow and ice coverage was even under discussion. He’s waltzed in half-cocked wanting to see skeptics denying warming so that’s what he saw.
Now he’s reverting to damage control.
You confuse the conclusion with the mechanism:
Conclusion:
Mechanism:
As always, you should “Keep your eye upon the donut and not upon the hole” Willis has shown that the conclusion is wrong. The mechanism doesn’t matter.
Mike McMillan says:
March 6, 2012 at 2:28 am
:::
That must be lawyer Latin.
Best I can come up with is inter aliae for a dative feminine plural.
############
nope: inter comes with the accusative.
And by the way: correlateion coefficient r² for cyclic signals: sounds AFAIR this sounds odd for me.
Maybe my math and stats are a little rusty.
Steve from Rockwood says:
March 6, 2012 at 5:56 am
“If so then case closed on a good paper.”
Ooo, I wouldn’t go that far. If I understand correctly the mechanism proposed here is not new, but provides some evidence. But just because one season or a small handful of autumns of low sea ice and greater snowfall correlates, it does not mean that it is significant.
What I think the paper tries to show is that it is a contributing factor, in particular to disturbing the usual circulation pattern, but if other contributing factors aren’t apparent it isn’t sufficient to bring about the high snowfall.
You would need to identify other contributing factors, find where they are manifest alongside low sea ice extent and sea whether you then get high snowfall. And you would need to do that quite a few times over before you could have confidence that this idea is correct.
In the absence of the evidence, which is hard to get, it is a plausible theory that is worthy of falsifying and giving some indication of what might happen on seasonal basis.
MFKBoulder says:
Your comment is awaiting moderation.
March 6, 2012 at 7:00 am
Mike McMillan says:
March 6, 2012 at 2:28 am
:::
That must be lawyer Latin.
Best I can come up with is inter aliae for a dative feminine plural.
############
nope: inter comes with the accusative.
But who cares: Fred Singer can’t even quote the frist seven words of Cecar’s De Bello Galloco (in his last “paper” on WUWT).
And For Wiilis:
And by the way regardin Willis’ paper: correlateion coefficient r² for cyclic signals: sounds AFAIR this sounds odd for me.
Maybe my math and stats are a little rusty.
It is important to see what happened in the past. Analyzing just a few years, as done in this paper, can be quite misleading.
What we know is that the year 1985-1986 snowed quite a lot in Europe and the snow reached Rome, see here
as it happened last month. A lot of snow in Rome is extremely rare.
According the theory proposed in the paper, the years 1984 and 1985 had to have a very small arctic sea ice cove in the fall. But according the figure in Willies article, nothing serious happened to the arctic sea ice cove during those falls.
The things appear to be more complex than what the paper suggests.
I tend to agree with the proposed mechanism, but like Willis, question the significance. The mechanisim has to do with the processes of evaporation/condensation and freezing/thawing and the associated vast amounts of energy exchange. Our study of the Arctic Ocean is probably the key to our understanding of the rate of energy lose to space. http://www.kidswincom.net/CO2OLR.pdf.
A side note: the more cold exposed ocean water, the greater the global sink rate for CO2. Is that a “negative feedback”?
Brad,
Lets pin this down, you want Willis to recast his Fig 2. as Fig 2a Ice/”Early” European Snow and Fig 2b Ice/NorthEastern and mid-wester North America Snow. Thus factoring the analysis over this historical record into a geographic and inter-seasonal clustering.
Does the underlying data support such an analysis over the 1978 to present record?
Brad says on March 6, 2012 at 1:24 am:
“Well, your analysis ignores that the largest drop in sea ice area occurred in 2007 and the has never really recovered, thus your analysis seems the simplistic one here.
Having read most or all of your and her posts for years, I pretty clearly pick her analysis as being more detailed, more thoughtful, and based on a much deeper understanding of science as well as her having more knowledge in general. This post included.
If I were Mr. Watts I would pull this simplistic post, almost childish in its analysis.”
=============
Willis says: “Finally, it is clear that there is no statistical relationship between decreased ice and increased snow.”
And I say: Yes exactly – and why should it? — What can melting sea-ice in the Artic summer possibly have to do with snow-fall in the NH’s winter?
Does sea ice truly reduce evaporation?
By how much? Where is the data to support this? Or is it simply an assumption? In the freezer, ice cubes shrink. What is the evaporation rate for the polar ice cap? What is the evaporation rate for cold sea water?
Brilliant! Very simple and very clear. You can’t argue with that.
Afte reading the press release on the paper I had a quick look at the snow/sea-ice balance and its impact on albedo. I concluded that effect of declining sea-ice is not compensated by relatively constant snow cover.
http://www.climatedata.info/Discussions/Discussions/opinions.php
…and scoll down to ‘Sea ice and Snow.’
The data set time series in the paper is definitely too short. But if you look at their models results (which are meant to mimic the last snowy years) one of the strongest positive anomalies is in the conterminous US states. If you check those data ( http://climate.rutgers.edu/snowcover/files/monthly.uslower48 ) the correlation with ice cover is even worst. Peak snow cover was in 1978 and then 1979 and 1985, while the minimum of the whole series was in 1981.
Winnipeg set an all-time record for winter snowfall in 1955-56. Was that caused by ice disappearing?
Nicola Scafetta wrote:
“What we know is that the year 1985-1986 snowed quite a lot in Europe and the snow reached Rome, see here…
…as it happened last month. A lot of snow in Rome is extremely rare.
According the theory proposed in the paper, the years 1984 and 1985 had to have a very small arctic sea ice cove in the fall. But according the figure in Willies article, nothing serious happened to the arctic sea ice cove during those falls.”
August/september extent & area:
http://klimaforskning.com/forum/index.php?action=dlattach;topic=337.0;attach=1705;image
http://klimaforskning.com/forum/index.php?action=dlattach;topic=337.0;attach=1707;image
http://youtu.be/IIAdHEwiAy8
Willis, you think that you have debunked the paper with your very simple comparison of ice extend charts and snow extend charts.
The paper says:
1. “Here we demonstrate that the decrease in autumn Arctic sea ice area is linked to changes in the winter Northern Hemisphere atmospheric circulation”
Did your comparison debunk this statement? Of course not.
2. “This circulation change results in more frequent episodes of blocking patterns that
lead to increased cold surges over large parts of northern continents.”
Did your comparison debunk this statement? Of course not.
3. “Moreover, the increase in atmospheric water vapor content in the Arctic region during late autumn and winter driven locally by the reduction of sea ice provides enhanced moisture sources, supporting increased heavy snowfall in Europe during early winter and the northeastern and midwestern United States during winter.”
Did your comparison debunk this statement? Of course not. Pay attention to: “heavy snowfall in Europe during early winter and the northeastern and midwestern United States during winter”.
In addition you indicated that only 2007-2010 data was used and not the full available data set. Wrong again.
I would be interested to learn how do you understand for example the Figure 1B in the full article? Do you understand what it shows?
Personally I do not believe that all this has anything to do with CO2 levels. Dr. Curry did not say that either.
Nice post. The snow graphic flat lined. The study is perhaps flat lined as well. Snow fall always varies, the jet stream varies, the high and low pressure systems vary in location, the barometric pressures vary, time of year varies – there are so many variable inputs it will be a very long time before a “model” comes close to reflecting reality. And the reality is I need to go start the tractor and plough that displaced white stuff off my driveway and go hit the mountains as there are a few feet of that fresh displaced white stuff on the ski hills in the region (Canadian Rockies and Selkirks/Monashees).
Love your graphs. They put things nicely into perspective. Maybe they don’t show why, or maybe some people would have preferred them to represent something else. But what they do, is cause people to think and from that we get some interesting responses to cause us to think further. What could be better?
Thank you.
Terry Marsh and Catherine L. Harvey report on the empirical data in “The Thames flood series: a lack of trend in flood magnitude and a decline in maximum levels”
http://www.iwaponline.com/nh/043/nh0430203.htm
that “Since routine flow measurement began in 1883, the Thames basin has seen a substantial rise in air temperature and a tendency for both winter rainfall and annual runoff to increase. There is no trend in fluvial flood magnitude however, partly reflecting a decline in snowmelt contributions to major floods”.
In other words winter snowfalls, at least in the Thames basin, are not what they used to be, contradicting the assertion of “increased heavy snowfall in Europe during early winter.”
If Willis were not scientific, and instead promoting the skeptical agenda, he would have focused on this logical sequelum of the study:
If less sea ice cover in the Arctic leads to greater snow cover at lower NH latitudes, a negative feedback is established. Namely, a lower albedo in a region with weaker sunshine (watery Arctic in winter) is causing a higher albedo in stronger sunshine (non-Arctic, snow-covered NH). Thus this process would have a net cooling effect for the the Earth’s heat budget, and counter man-caused global warming, if there is such a thing.
Instead, Willis is scientific, and letting the facts fall where they may lay, and good for him, and those of us who enjoy his posts.
Brad says:
March 6, 2012 at 6:15 am
It is simply an interesting insight into the mind of true believers, they really do not want the truth, they long to have their deeply held belief confirmed at all costs.
Talking about yourself again, Brad?
Back in 1963-64 when I studied Geomorphology as part pf my Geology/Geophysics degree at UBC there was a theory that with the continued melt back of the arctic ice pack there would be more evaporation. The higher humidity would result in more snow and the growth of glaciers. This in turn could trigger a new ice age. I’m still waiting for the ice age.
Wow willis you really screwed this up.
best read the paper and attend to the details and make a correction.
three words: regional; equivalent water
AdderV says:
March 6, 2012 at 3:12 am
AdderV, I didn’t say “what the other side said”. I didn’t refer to “warmists”. I didn’t say “then he said”. I didn’t say “then she said”.
Doing what you have done is called “misquoting” someone, AdderV. When you put words in quotations marks, that means you are claiming someone said exactly that. Not something like that. Exactly that.
Now, obviously, you have some objection to this post. However, since all you have done is misquote the post, I don’t have a clue what it is that you are objecting to. So far, all you’ve done is object to THINGS I DIDN’T SAY.
If you think that makes you look all smart and on-topic … think again.
w.
Brad says:
March 6, 2012 at 6:24 am
“While you are at it, why don’t you try to refute the actual conclusion of the paper, instead of the strawman you have thrown up. BTW – I assume she has the basic correlations run here, any descent scientist would have, and Dr. Curry is.
The real conclusion:
“supporting increased heavy snowfall in Europe during early winter, and the northeastern and mid-west United States during winter. We conclude that the recent decline of Arctic sea ice has played a critical role in recent cold and snowy winters.””
1. The years of data were cherry picked to support her predetermined conclusion.
2. Looking at all of the data it is clear that there is no relationship.
Simplistic, maybe, but enough said.
Replicant says:
March 6, 2012 at 2:21 am
I don’t have a clue what they studied, just what they graphed and demonstrated.
Nothing. Look at the graphs. There are no unusual spikes and little secular variation in the winter snowfalls.
w.
Brad says:
March 6, 2012 at 1:24 am
… If I were Mr. Watts I would pull this simplistic post, almost childish in its analysis.
========
There is a much simpler way to look at “Impact of Declining Arctic Sea Ice on Winter Snowfall”. Let’s take a look at Northern Hemisphere snowfall over the extended period Willis proposes and see if there is any trend; http://climate.rutgers.edu/snowcover/chart_seasonal.php?ui_set=nhland&ui_season=1
As Willis points out, the trend is insignificant.
However if one looks at Eurasia vs North America, Eurasia is an insignificant trend but North America is showing a slightly increasing trend with 4 of the last 5 years above the trend. This supports Willis’s point about only including the last 5 years (need to include all available data) and generally shows how insignificant the issue is.
IMO, it would be interesting to reflect specific snowfall in relation to specific areas where sea ice isn’t forming over time for the winter real sea ice to snowfall conclusions.
Agnostic says:
March 6, 2012 at 3:28 am
Thanks, Agnostic. I doubt greatly that “increased cold surges over large parts of northern continents” would increase the amount but not the area of the snowfall. That makes no sense.
If that were so, we’d have seen it as a delayed (lagged) response. But as I pointed out above, lagging the response doesn’t increase the correlation in any meaningful way.
w.
Steve from Rockwood says:
March 6, 2012 at 5:22 am
Thanks for the questions, Steve. Sure. If it didn’t we wouldn’t say that “less Arctic ice leads to greater snowfall”.
No clue, and I doubt greatly if there is a general rule.
No clue, and I doubt greatly if there is a general rule..
On the ground, yes. In the air, no.
I just graphed the Arctic ice area above. In general it’s highly correlated to Arctic ice extent, so we can use it as a proxy if you like (correlation = 0.991).
That’s in Figure 2 in the upper panel. You see the parts of the graph where there is more ice? That’s the winter. Is the winter ice are “fairly constant”? Your call …
That’s my answer to the two questions …
w.
Brad says:
March 6, 2012 at 6:15 am
You got pwned, and now you want to whine about it. Man up, put on your big boy pants, and come back when you have something other than whining to contribute.
w.
Brad says:
March 6, 2012 at 6:24 am
Actually, the real conclusion was in the Abstract, where they said that “the recent decline of Arctic sea ice has played a critical role in recent cold and snowy winters.”
Seems like what you mean is that I’ve refuted their main conclusion, but not one of their minor conclusions. Part of the problem is, it’s never clear exactly what their “real conclusion” is. Does their “real conclusion” just apply to Europe and the “northeastern and mid-west US” as one part of their paper says? Or does it apply the whole northern regions, as their abstract says? We don’t know. Does it just apply to autumn ice, or does it apply to the “recent decline” of Arctic ice in general? We don’t know.
So … if you have a falsifiable statement of theirs that you’d like me to look into, I’ll consider it. Bear in mind that a vague claim about the “northeastern and mid-west US”, with no geographical or temporal boundaries to what they are referring to, is not falsifiable.
Were you born an unpleasant jerkwagon, Brad, or do you have to work at it? If you cut the attitude, you might actually pass for a real human being instead of a particularly unpleasant Turing Test.
w.
AR5 needs peer-reviewed literature to explain why nature isn’t cooperating. This is a keep-the-show-on-the-road paper. It only needs a veneer of science for polishing, and a well known name for credibility.
It doesn’t have to be right.
Replicant says:
March 6, 2012 at 6:40 am
So you’re going to bust me for using what I had available at the time? Then you’re not going to link to the full study? Well, ain’t you just the cutest thang?
In any case, I find nothing in the full study that is significantly different from the ARXIV version Judith posted before. I do note that indeed, she is talking (as I am) about snow EXTENT and not snow thickness.
I also note that you call what I have written “false science”, but you haven’t pointed out a single flaw in what I wrote. You sure you understand how this “science” thing works?
w.
The hypothesis sounds plausible. It agrees with my skeptical argument that feedbacks are negative. In the Arctic free of ice theory, not only are feedbacks negative, but they’re HUGELY negative!
Unfortunately, I think that the hypothesis is countered by facts:
1613: Inuit (Eskimo) found in kayak near Hull, eastern England
1682-1684: Inuit landings on Orkney Islands, northern Scotland
If lower ice in the arctic led to glaciers growing farther south, those eskimos would not have been traveling as far south as Scotland and England during the little ice age.
Less Arctic ice makes more snow….
…and more Arctic ice makes more snow
Where have we seen that before? warm/cold, wet/dry, drought/flood, rain/snow…………
“The snowiest two year period on record (1978-1979) also had the peak Arctic sea ice of the last century.”
Willis:
“Thanks, Agnostic. I doubt greatly that “increased cold surges over large parts of northern continents” would increase the amount but not the area of the snowfall. That makes no sense.”
Actually it was one of the things that Skeptical Science attacked Monckton for, that just because there had been increased snowfall, snow extent had not increased…but i digress…
I am inclined to agree with you Willis, but if you think about it, it actually does not necessarily follow that increased precipitation means an increase, or a significant increase in coverage. And it’s important with regards to snow because it is regarded as contributing to albedo. If the weather patterns bring lots of snow to a particular area – more than usual – but no more to an area that does not get much, then albedo has not increased since even a thin layer of snow can increase albedo. And such a case existed in Europe this year, with not very much snow in the UK (except parts of the North) and much more than usual in East Europe.
Judith actually pointed out that what they were measuring was volume not extent. It is something they clearly thought about and probably worth your while checking. In any case your graph is misleading although your conclusion may well be valid. They are not claiming extent and sea ice correlation, but volume.
Interactive chart showing the complete satellite Arctic Sea Ice record (30% lower limit) without an imposed average. Cryosphere Today also does a great job of segmenting Arctic Sea Ice into regions and highlights the deviations both positive and negative within each region.
http://arctic.atmos.uiuc.edu/cryosphere/arctic.sea.ice.interactive.html
I really love this interactive chart; h/t to UofI for the willingness to show the actual data in an insightful way.
If you progressively turn Off each year from the legend at the right of the chart from the most recent to the beginning of the satellite record, you’ll see at the minimum all but a couple of years that haven’t followed the Arctic Sea Ice decline at the Sea Ice minimum.
The problem I have with the satellite record is the realization that it may only represent one half of a “normal” cycle and any conclusions associated with warming are nothing but obvious within the cycle.
The next 30 years will be the tell. We live in interesting times and thanks for the great post Willis!
In his comments, Brad says:
“Thanks, you (sic) apology is accepted.”
And
“I find those that can do science do not throw up basic correlations irrelevant to the point at hand – at least after their first or second year in grad school.”
Where does the anger originate?
Examining real data, graphing, correlation, pondering. Are doing these things not doing science?
Willis puts his writiings up for review, isn’t peer review part of science?
There is a famous story among sediment transport and hydraulic engineers that H. A. Einstein (son of Albert A. Einstein) would take his new graduate students to his hydraulics laboratory and ask them to take data and learn from them for a few months before they started their formal studies.
Northern hemisphere winter snow extent has increased over recent years, but only possible link that relates to rest of data was the extended solar minimum. Arctic sea ice has been declining for decades so if this was the cause, then there would be a general increase in snow extent throughout the data. The probem with this judgement is also the true Arctic ice 80N+ hardly changes and remains around zero most of peak summer. The arctic decline has been regions concerning less than 80N, so other than a little local lake affect snow this should hardly affect NH weather.
William M. Connolley says:
March 6, 2012 at 6:44 am
What Curry is claiming is that increases in snowfall, not changes in patterns but increases in snowfall, will result from lowered ice area. Emphasis mine.
You say “Curry doesn’t claim a relation between sea ice and total snow extent”, but that doesn’t hold up when you look at what she said. You see the part about the increase in snowfall that is supposed to come from reduced sea ice? That’s called a “relation between sea ice and total snow extent”.
She waves her hands and talks about “increased heavy snowfall in Europe during early winter, and in the northeastern and mid-west United States during winter”. That’s all wonderful … but which states, and which parts of the winter? If I look at the winter (DJF) North American snowfall vs the autumn (SON) Arctic ice, I get the following:
The p-value of the relationship is 0.10, no significant relationship between the two.
Is there some kind of teleconnection between arctic ice somewhere and snow extent somewhere, at a statistical significance of less than p = 0.5? Most likely there is. If you break anything down into small enough pieces, and you only look for P less than 0.05, you’ll find a relationship about one out of twenty tries just by chance. And …?
This is the problem with things like her graphs of large areas, where the “regions within contours denote the regression above 95% confidence level”. Remember that we expect 5% of the region to show FALSE STATISTICAL SIGNIFICANCE just from random variations … and they make no attempt to separate out such randomness and show that the relationship is more than just chance.
I’m sorry, William, but I just don’t find the meat in the study. YMMV.
w.
MFKBoulder says:
March 6, 2012 at 7:00 am
What cyclical signals are you referring to?
w.
I have issues with the references to Canadian geography (or the lack thereof) in this paper.
The Abstract seems to skip the 2nd largest country (aka “Americas Hat”) on the planet entirely. How the snow extent can differentiate between northeastern US and southeastern Canada is an interesting conclusion — maybe snow can read a map of the irregular border for 1/2 the continent better than I can.
Also: “The only notable exception is northeastern Canada and Greenland, where weak westerly winds favors more frequent incursions of warm air masses from the North Atlantic. This leads to warm anomalies there (Fig. 2C), helping to explain extremely low ice coverage observed in Baffin/Hudson Bay, Davis Strait, the Labrador Sea, and Gulf of Saint Lawrence in recent winters,
particularly in 2009–2010 and 2010–2011” — this is a pretty big exception since I believe the area they refer to is greater than Europe.
This macro-assessment of Canadian geography is totally counter to what any Canadian and Environment Canada knows about the Canadian climates (emphasis on multiple climates) e.g. every major city (>200K’s people) in Canada is pretty well in its own climate zone never mind the Northern hinterlands. Papers that refer to Canada’s with this kind of granularity lose credibility with me personally — and may be why so many skeptics are Canadian.
Replicant says:
March 6, 2012 at 8:27 am
No, I don’t think that. I rarely even use the term “debunked” because it is so emotionally loaded. Your attempt at mind-reading what I think is an abject failure, don’t quit your day job.
w.
steven mosher says:
March 6, 2012 at 9:22 am
Steven, you seem to specialize in drive-by, mildly unpleasant, and most importantly, often unintelligible posts.
If you want to clarify that pile of meaningless nonsense, please do so. You’re a smart guy, and you may have a point, but if so, it is invisible.
Otherwise, I’ll ignore it.
w.
PS—The phrase “equivalent water”, and the word “equivalent”, do not appear anywhere in the paper. You see the problem with your bogus posting style? It just makes you look like you are clueless. I’ll assume you are talking about “SWE”, or snow water equivalent. This is the amount of water contained in a certain amount of snow.
But since Judith never uses the term … then why on earth are you talking about it?
And “regional” is used twice, once in the reference list, and one referring to one of the references.
You must think your posting style makes you look mysteriously oracular. It just makes you look dumb.
Three words; explain; your ideas.
w.
Split flow jet streams and blocking patterns? That’s sooo 2010/2011. Didn’t you know the latest result of global warming is “Superjets?” Yeah. “Superjets.” Formerly known as a “phased jet stream,” that is until yesterday…
From the blog of Jeff Masters:
“Dr. Jonathan Martin of the University of Wisconsin-Madison is doing interesting research on the type of situation we saw with some of the recent severe tornado outbreaks, when two branches of the jet stream, the polar jet and the subtropical jet, merge to form a “superjet.” In a December 2011 interview with sciencedaily.com, he said: “There is reason to believe that in a warmer climate, this kind of overlapping of the jet streams that can lead to high-impact weather may be more frequent.””
So, you see, the “Superjet” is supposed to become more frequent, not blocking. At least it will be until the next heat wave.
Agnostic says:
March 6, 2012 at 10:40 am
Here’s what the paper says that they are using:
See the part where it says “snow extent”? That’s where I got my figures from that I used in the graph. I rest my case that they are using extent.
w.
“Here we demonstrate that the decrease in autumn Arctic sea ice area is linked to changes in the winter Northern Hemisphere atmospheric circulation that have some resemblance to the negative phase of the winter Arctic Oscillation. However, the atmospheric circulation change linked to the reduction of sea ice shows much broader meridional meanders in mid-latitudes and clearly different interannual variability than the classical Arctic Oscillation.”
So we have it is regional claims on this thread not NH, but the AO and NAO affects the entire NH not just these regions of NE USA, SE Canada, parts of Asia and Europe. A broader meridional in mid-latitudes does give different yearly variability than the classic AO, but the AO needs to be going in the same direction that this claim is made. Over the long term since arctic ice was declining the AO has become increasingly positive not negative. The increased snowfalls require a negative AO for this increased different atmospheric circulation to be possible.
Yet we have declining Arctic ice since the 1970’s and wait for it an increasing postive AO during this time, except the last few years. Again like in my previous post the only possible link related to other data time period is the extended solar minimum.
http://upload.wikimedia.org/wikipedia/commons/1/13/Arctic_Oscillation.svg
Ágúst in Iceland and Willis’ PDF discuss the accuracy of the Temperature measurement.
To transmit (or record) the data some form of Analog to Digital (A/D) conversion is required which will limit the precision. For example a typical industrial A/D card converts to a 12 bit word i.e. a resolution of 1/4096 (1/(2**12)). The measured resolution will, of course, vary with the range of temperatures the card is set up for. Add to this the accuracy of the electronic design and components.
Some other sources of error (among many) may be found in allowing for non linearity in temperature to sensor output, thermocouple cold junction compensation and simple transmission errors.
Finally, I will mention processing errors such as: data entry, fixed or floating point arithmetic, truncation or rounding….
Sign me a retired controls engineer with too many scars
Willis
I have supplied Judith with masses of sea ice data for the 1918 to 1939 arctic melting and also snow extent. If her hypothesis holds she should be able to determine if this episode of arctic melting did what she thinks it should.
I don’t have a dog in the fight, just interested to see if the facts match the theories.
tonyb
This paper seems to inter-change snowfall and snow cover (even though all the data, as Willis points out, is snow cover related). These two are not the same and once again as a Canadian I will throw the challenge flag on this. You can draw an East-West line across Canada where above it you are expected to have snow cover (e.g. >95% chance) for January and February. But this doesn’t tell you anything about the amount of snow. For example, here in Ottawa we have had snow cover since before Christmas and still do but we are over 1 meter blow our average snowfall (50% of normal). Note: this is not abnormal as our snowfall varies a lot — what would be abnormal would be to see the grass in your yard (no snow cover).
Willis
I think Mosh might have picked up the phrase ‘equivalents’ from the comments section of the Curry thread. I have just come across it.
“steve fitzpatrick | March 5, 2012 at 9:08 pm | Reply
Judith,
Interesting paper. One thing I wonder about is if there might be a heat balance calculation which would be relevant. Lake effect snow near the great lakes results from substantial warming of the lakes in the summer months and heat carryover causing evaporation into the early part of the following winter. But I wounder, is enough heat being collected in the Arctic ocean as a result of lower ice cover to account for the increased (total mass of) snowfall? The arctic doesn’t really warm very much above 0C over the summer. The final figure in the paper shows modeled deviation of snowfall. The units say ‘mm’, but I am guessing mm equivalent water not actual mm of snowfall. Do I have that right?
curryja | March 6, 2012 at 8:18 am | Reply
Steve, yes mm equivalent water. The major impact on increased snowfall is from the circulation changes; secondary from the open water in the Arctic Ocean. The later needs more investigation. Cold winter air (maybe 30C) blowing over open water even at 0C will give a large evaporative flux.”
So can’t see it was in the original paper but it came up in the clarification in comments
tonyb
w – you say “There have been no breakthroughs in climate science in thirty years”.
Wrong. Perhaps you should have said: There have been no breakthroughs in “climate science” in thirty years.
There have been a number of breakthroughs in real climate science, the latest is
http://arxiv.org/abs/1202.5156v1 (submitted to Physical Review Letters)
comment on it is here
http://calderup.wordpress.com/2012/03/01/yet-another-trick-of-cosmic-rays/
AdderV said (March 6, 2012 at 3:12 am):
This blog is getting worse by the minute, stooping to a new low in every article.
I used to like to come here everyday, not so anymore. This once fine blog is drifting towards school yard bullying.
It pains me to realize that I agree with AdderV’s statement. Before reading any comments, I found myself wishing that Willis had omitted his last three paragraphs; comparing L2012 how many snow-storms can dance on the head of an iceberg” is “inciteful” rather than insightful. Ad hominem slurs reduce the power of argument and reduce respect for the author.
I fear Willis’ posts should be reviewed by an calm editor before appearing in this worthy blog.
REPLY: I reviewed it and pressed the button. Blame me then, but I didn’t see the commentary as “inciteful” at all. If you want “inciteful” read Grist or ClimateProgress. Here, we tend to go with comedy and fun analogy – Anthony
AdderV and the other trolls,
Here we have yet another poster who drives by ‘eats shoots and leaves’. Like Brad anonymous AdderV you are clearly only here to disrupt the thread. Do us all a favour and find somehere else to troll.
In regards to JC, some of us have been around on climate blogs a lot longer than most others here. I personally remember JC’s first posts on CA and challenged her back then about her advocacy along with Kerry Emmanual. To this day she still hasn’t explained why she jumped on the ‘Hurricane Katrina’ band wagon along with Trenberth & Co.
Septic Matt
“I can’t see where you have disputed any of the basic claims of the authors. ”
Willis is clearly disputing L2012’s computer model pre-written conclusions in this thread. I presume you mean ‘refuted’ rather than ‘disputed’?
KevinUK
Can you imagine the storm of frenzied AGW jowl-flapping that would erupt if Willis or Anthony were to propose an hypothesis based on only 5 years of data, from an available dataset of 30 years? The endless parroting of the “cherry-picking” accusation? But since its cutesy Curry with a troupe of camp-followers, instead its all obsequious praise and flattery from the AGW courtesans. I dont think so – this “study” is a load of infantile crap. A 2-year old could tell you that – “when there´s less ice there´s more snow?” Pull the other one!
I re-examined the paper and Willis is at least partly right: the only reference is the Rutger Data set which I happen to be familiar with and it is for extent only. The only mention of snowfall by volume is at figure 7 but it’s not clear what the source is. Clearly they looked at extent as well, but they were doing something more than examining it in conjunction with sea-ice which is what Willis has done with his graph.
Other than a reasonable and plausible mechanism, the paper seems to suffer from some information missing that might have cleared up the confusion.
Inter alia is common usage in English and other European languages. Inter alios/alias while technically correct as a reference to people is pedantic and tends to obfuscate.
Besides, in this instance, one could well be referring to people’s papers rather than the people themselves in which case inter alia is much more appropriate.
phlogiston wrote:
“I dont think so – this “study” is a load of infantile crap. A 2-year old could tell you that – “when there´s less ice there´s more snow?” Pull the other one!”
Accumulation on the Greenland ice cap from the historical GISP2 records. Y-axis is meter pr. year:::
http://klimaforskning.com/forum/index.php?action=dlattach;topic=206.0;attach=656;image
“I used to be Snow White”
But you drifted, Willis? ;-p
Willis — enjoy the read, have you looked at correlations between the *rate* of loss w.r.t. snowfall?
Barbara Skolaut says:
March 6, 2012 at 3:37 pm
“I used to be Snow White”
But you drifted, Willis? ;-p
======
Best Muse of the Day ; )
Let’s keep it simple fun in science.
Sadly, at the end of the day, the obvious fact of all this effort points to ignorance (aka –> Al Gore).
“I would call this is pretty conclusive evidence, perhaps even the long-sought “smoking gun”, that clearly establishes that the two climate model runs were indeed different.”
Priceless, Willis!
WUWT: Come for the science, stay for the fun.
I might be preaching to the converted, but doesn’t everyone know that the Gulf Stream keeps Northern Europe, America, and UK from freezing? However, before every mini ice age or glacial period the Gulf Stream stopped or changed direction because of the increase of fresh water coming from melting ice or rivers and land ice, that caused the places to freeze up. The planet did warm up before it froze up again. Plus other things of course, many variables that make it hard for weather men to give 100% long term predictions. But the sun don’t shine in Antarctica and the Arctic or the Arctic Circle for months during the winter and early spring. Then sometimes 22 hours during summer.
Hardly typical of the rest of the world, eh? Yes we are getting heaps of rain in Oz, and flooding, but that is typical of this ancient continent. But people will build on flood plains? Or on water fronts, volcanoes, and seismic trenches. It will be bush fires next as the added green growth dries out and some idiot comes along and lights a fire deliberately.
I feel very sorry for Professor Judith.
She seems to be such a nice lady, trying to do honest work, while living in, what shall I say?
“A vipers’ nest”? Nay.
“A covern of villians”?
Perhaps, “extended family of witch doctors” sums them up best.
Or the Catholic church at the time of friend Luther, perhaps?
Take your pick.
Life is not easy for a consensus laden, yet honest broker, trying to do good, honest work.
Oh, what a shame!
Give it up, dear professor!
Time to crawl out of the closet into the light and call it as you really see it.
Find a place to stand tall with you feet firmly placed on solid land and face the world bravely.
Here’s the abstract again, With bolding, mine not Willis’: anomalously large snowfall in recent winters has affected large parts of North America, Europe, and East Asia. Here we demonstrate that the decrease in autumn Arctic sea ice area is linked to changes in the winter Northern Hemisphere atmospheric circulation that have some resemblance to the negative phase of the winter Arctic Oscillation. However, the atmospheric circulation change linked to the reduction of sea ice shows much broader meridional meanders in mid-latitudes and clearly different interannual variability than the classical Arctic Oscillation. This circulation change results in more frequent episodes of blocking patterns that lead to increased cold surges over large parts of northern continents. Moreover, the increase in atmospheric water vapor content in the Arctic region during late autumn and winter driven locally by the reduction of sea ice provides enhanced moisture sources, supporting increased heavy snowfall in Europe during early winter, and the northeastern and mid-west United States during winter . We conclude that the recent decline of Arctic sea ice has played a critical role in recent cold and snowy winters.
I can’t see where you have disputed any of the basic claims of the authors. Clearly the emphasis is on changes in the distribution of cold and snow following declines in Arctic ice. You have clearly shown that something they clearly did not claim is clearly unsupported by the data. They could still be wrong, I won’t deny that, but your analysis does not show it.
Septic Matthew/Matthew R Marler says:
March 6, 2012 at 6:47 pm
“They could still be wrong, I won’t deny that, but your analysis does not show it.”
Willis’s graph is slightly misleading but actually his analysis is reasonable; that to show the existence of a mechanism, there needs to be some correlation between snowy NH winters and low sea ice in the arctic in autumn. While the reasoning and some of the supporting discussions are perfectly plausible and make good sense, the supporting evidence is not clear.
We “know” that the last 4 or 5 winters in the NH have been anomalously cold but they only cite the Rutger snow extent data. And I do think that the broad thrust of Willis’s criticism is valid that there should be an effort to find other examples of low ice/snowy winters in the past at least to support the conclusion, even if the evidence is only anecdotal.
The key issue in the paper, is changes to the atmospheric circulation pattern due to open arctic water in autumn, but as one other critic of the paper notes, you can strike a bell with anything and get the same note.
> You say “Curry doesn’t claim a relation between sea ice and total snow extent”, but that doesn’t hold up when you look at what she said.
But notice, in reply, that you only quote some ambiguous words – not any of the actual results from the paper. No figures, no numbers – those are the actual content of the paper, as always. The words are just wrappers around the numbers and the figures.
> This circulation change results in more frequent episodes of blocking patterns that lead to increased cold surges over large parts of northern continents.
Cold surges, not snow amount, and notice the restricted geographical range. Some gets colder, sometimes. Some gets warmer. No claim about total snowfall, or total extent.
> Moreover, the increase in atmospheric water vapor content in the Arctic region during late autumn and winter driven locally by the reduction of sea ice provides enhanced moisture sources, supporting increased heavy snowfall in Europe during early winter, and the northeastern and mid-west United States during winter.
More snow, but only in restricted regions. This is what her figures show – not an increase in overall snowfall.
> If I look at the winter (DJF) North American snow…
You haven’t defined the region “North American” that you use. The paper doesn’t use that region; what it says is “northeastern and mid-west United States” though I think those are words about the figure (the one from the paper that you have reproduced). Looking at that figure, and the gradient between red and blue over “North America”, it is easy to see that you’re unlikely to see much using the analysis in your comment.
You’ve read something into the paper that isn’t there. You’ve looked for an effect that wasn’t claimed, and you haven’t found it, leaving you in agreement with the actual paper, though possibly not with your mental image of the paper.
In the end we did Northern Hemisphere snow extent to death in the thread http://wattsupwiththat.com/2010/03/02/2001-2010-was-the-snowiest-decade-on-record/
To summarise the statistical discourse from that thread :
The data set in play http://climate.rutgers.edu/snowcover/files/moncov.nhland.txt
shows no significant change in mean snow cover at the 95% confidence level in any of the following scenarios: whole year(01-12), early winter (11,12,01) winter (12,01,02) early spring/late autumn (02,03,09,10) between 1966 and 2010. It does indicate increased winter(12,01,02) variability that occured between 1977 and 1987 diminished between 1988 and 1998 and was undetectable in 1999-2010.
Using the standard widget makers techniques outlined in ANSI/EIA-557-A-95 NH Winter Snowfall has the charecteristics of normal variability about a stable mean. But it is possible to identify abnormal Winter snowfall years as 1978 (Abnormally High), 1980 (Abnormally Variable) and 1981(Abnormally Low). !979 Arctic Sea Ice minima 7.1 M Km 1980 ASIM 8.1 M Km2 I can’t find a figure for 1977. So in two of the instances where we are looking for an Assignable cause Arctic sea ice extent was substantially higher than it was in the period 2007 – 2011 where no exceptions occurred.
An increase in Minimum sea ice between ’79 and ’80 was followed by light winter snowfall but it would be hard to assign causality given the presence of other climate perturbing events in the same time period (Major eruptions at Etna and Mt St Helens for example).
Do not you find yourself constantly surrounded by people like that?
Today’s countries are full of them!
People walking all day, every minute of the day, worried about everything! Concerned about air, the water, worried about the soil. Concerned on insecticides, pesticides, food additives, carcinogens. Concerned about radon gas, asbestos. Concerned about saving endangered species!
We’re so cocky. So so cocky! Everyone is going to save something now: “Save the trees, save the bees, save the whales, save those snails.” And the greatest arrogance of all, “Save the planet” WHAT? Is that [SNIP: Language. -REP] people teasing me? Saving the planet? We do not even know yet how to take care of ourselves. We have not learned to care for one another and are we going to save the [SNIP: Language. -REP] planet?
Curry was over here. She did not post a reply. I guess she is a bit upset with the fast shotgun approach to her paper. Willis this is the moment where a smart guy starts thinking about sending flowers. Dont forget women are always right. I know it is a lie just like AGW but that one stood the test of time.
I took the liberty to quote her reply on her own blog “There is nothing there to infer anything quantitatively about NH total snow cover, which is what Willis analyzed. Most notably, our analysis clearly says nothing about any change in large regions of Russia and Canada. The moisture source from the open water in the arctic ocean is secondary to the atmospheric circulation change.
In the conclusion, the words used are “The results of this study add to an increasing body of evidence . . .” are the appropriate words in my opinion. The “we conclude” statement in the abstract (where word numbers are limited) comes across a bit strong, but I think the more thorough statements in the the Discussion are appropriate. Again, some of these words are not mine such as “load the dice”.
IMO the significance of this paper is in understanding seasonal snowfall variability, which is a goal of seasonal forecasting.”
Agnostic:Willis’s graph is slightly misleading but actually his analysis is reasonable;
I agree that Willis’ analysis is reasonable: he clearly showed that something they clearly did not claim is clearly not supported by the data.
LucVC says:
March 7, 2012 at 8:55 am
Thanks, Luc. Judith can get flowers from me any time, I appreciate her general approach.
And if all that her paper says is that in years of low ice, some small parts of the US and Europe experience different snow patterns, well, that may be so … so?
Let me repeat what I said:
We have a branch of science where there is no overarching theory other than
∆T = λ ∆F
and that is clearly wrong … and meanwhile people, brilliant people like Judith, are spinning their wheels on trivial questions like this one.
Is there some relationship between decreasing ice, and snow levels? I would be surprised if there were not such a relationship at some point at some time. Just keep splitting the spatial field and the time into smaller and smaller sections. No relationship worldwide? Lets look by continents. No relationship over Eurasia or North America? Lets look by countries and smaller areas. No relationship in the whole US? Lets look by parts of the US …
HEY, WE FOUND A RELATIONSHIP! It’s between sea ice levels and snowfall, not on the planet, not on the continent, not in the country, but in the “northeastern and mid-west United States” … Oh, and we found a relationship for Europe as well … but not for the whole winter, just for some parts of the winter …
Doing that kind of thing, when there is no theory of climate and there are huge unanswered questions in the field, is the waste of a brilliant woman.
Anyhow, that’s my flowers for Judith.
w.
PS—People in climate science always forget that the more you divide up your fields, the more likely it is that you will find a “statistically significant” relationship that is nothing of the sort.
Suppose, for example, that you are looking for a relationship where p is less than 0.05. That is (foolishly but usually) taken as significant in climate science.
The odds of that happening by chance are one in twenty.
So … how many situations do you have to look at until you have 50/50 odds of hitting one of those “happened by chance” areas of FALSE STATISTICAL SIGNIFICANCE?
The answer is the solution to the equation 0.95^x = 0.5. The solution is log(0.5) / log(.95), which is about 13. At that point you have a 50/50 chance of finding a spurious result.
Now look at what Judith et al. have done. They have divided the winter up into early, middle, and late. They have divided the world up into (at a minimum) Europe, Asia, Canada, Alaska western US, and eastern US.
That gives us no less than 18 categories already … so we have better than a 50/50 chance of finding a “statistically significant” result in there that is not significant in the slightest, but is an expected random occurrence.
Any real statistician would have allowed for this in the experimental design. Judith and her co-authors, as far as I can tell, have ignored the issue entirely.
So while she gets flowers from me, it’s not for her statistics or experimental design …
Sorry, but the snow extent in the NH is quite important in this case because these regions that get more snow from this atmospheric pattern are always the areas that don’t get any snow most of the time away from mountains. When the AO is positve the extent of NH snow is greater than when it is negative. The positive NAO always has the least amount of NH snow extent, whereas the negative NAO always the greatest amount of snow extent. With this evidence even though the NH snow extent wasn’t mentioned in the paper this becomes an important matter. My last post describes even concentrating on the change in atmospheric pattern that the claims are still very weak.
My previous post should show this below.
“When the AO is negative the extent of NH snow is greater than when it is positive.”
Willis Eschenbach: HEY, WE FOUND A RELATIONSHIP! It’s between sea ice levels and snowfall, not on the planet, not on the continent, not in the country, but in the “northeastern and mid-west United States” … Oh, and we found a relationship for Europe as well … but not for the whole winter, just for some parts of the winter …
Doing that kind of thing, when there is no theory of climate and there are huge unanswered questions in the field, is the waste of a brilliant woman.
Most of scientific research is like that. These relationships may not be as exciting as what everyone wants, but they contribute to the total amount of knowledge. 100+ years ago was the purely empirical Balmer series of no known importance, and by 1939 Lisa Meitner and Otto Hahn had published dozens of mostly boring (to us now, and to most researchers then) papers documenting the breakdown pathways of radionuclides. I agree that this paper is not likely a part of a momentous theory such as followed those works, but practically nothing is.
Meanwhile, you have successfully refuted a claim that they did not make. Evidently, on your report, you did this because you were not interested in what they paper actually (pending replication) showed.
@Willis, the paper deals with a not so trivial issue.
Firstly, changes to snowfall (but more particularly snow extent) is important because an increase means higher albedo and thus a cooling affect. What they show here is that a reduction in sea ice or more open water in autumn leads to a . So the irony of alarmists saying that snowy winters are ’caused’ by global warming may actually be quite true. Global cooling is caused by global warming….
Secondly, from the point of view of seasonal forecasting, this paper adds to the body of knowledge and points to where further research may be warranted. That may include looking for correlations in the historical record, looking at sea ice distribution and prevailing winds to see if there is a pattern that might affect where higher snowfall precipitates.
@Septic Matthew/Matthew R Marler says:
March 7, 2012 at 9:38 am
Well actually the reason I think willis’s objections are reasonable is because the supporting data is unclear. Contrary to what William Connelly states up thread, they do suggest higher snow fall and colder winters as a result of the effect they are analysing. It may be that the data makes better sense to readers really familiar with it, but I know the Rutger data set well having done my own analysis of it, and it is the only source quoted, and it deals with extent not total fall, although they are undoubtedly correlated, they are not the same thing.
A lot of the confusion could be cleared up if the supporting evidence was set out more clearly. Willis might not have been drawn to make the analysis he did had that been the case. One of his big frustrations that comes through in many of his posts is that many climate papers just don’t clearly establish a connection between what they discuss and the evidence they cite to support it, and I am afraid I think this paper is one of them.
That said, I agree with the paper. It makes a clear, logical and reasonable case and I would be very surprised to find that the mechanism they propose does not exist or has no effect. So while it may annoy Willis because it lacks proper rigour, it is one of those things to ‘keep your eye on’ as it were.