More sea ice

Nothing much going on with sea ice at the moment, but people are getting excited about it, so why shouldn't I contribute to the smoke?

[A little concept I just created in the comments but am so pleased with I'm going to put it here: is 2007's Arctic sea ice like 1998's global temperature?]

First up in the stupidity stakes is Tim Flannery, for his Words of warming in the Grauniad, in particular for by June 2008 signs of a great melt were emerging and a senior adviser to the Norwegian government was warning that this may be the Arctic's first ice-free year. Even in June, that was blatantly silly. Repeating it now is worse. The Arctic will not be ice free this summer. Or next summer. Or the one after that. Or... The rest of the article is full of the usual over-excitement but isn't very interesting.

The grauniad has another go at sea ice in Meltdown in the Arctic is speeding up, which spends a lot of time saying "we don't know", then continues Professor Wieslaw Maslowski of the Naval Postgraduate School in Monterey, California. Using US navy supercomputers, his team produced a forecast which indicated that by 2013 there will be no ice in the Arctic - other than a few outcrops on islands near Greenland and Canada - between mid-July and mid-September. 'It does not really matter whether 2007 or 2008 is the worst year on record for Arctic ice,' Maslowski said. 'The crucial point is that ice is clearly not building up enough over winter to restore cover and that when you combine current estimates of ice thickness with the extent of the ice cap, you get a very clear indication that the Arctic is going to be ice-free in summer in five years. And when that happens, there will be consequences.' (HT ES in globalchange). But I dont believe it. Nor is it clearly true that ice doesn't have time to rebuild over winter.

Sea ice is still well above last year (e.g. nsidc [update: the IARC pic is better]), even if its taken a little downturn recently. Anyone who actually believes any of these prophets of doom is still welcome to join my betting pool... remember all those people who were advising me to close it a few months ago?

That was all Arctic. Eli has noticed that the SH ice has suddenly gone from a large positive anomaly to a small negative. Thats interesting.

Since I'm G-bashing, I suppose I should continue to On a planet 4C hotter, all we can prepare for is extinction by Oliver Tickell. We need to get prepared for four degrees of global warming, Bob Watson told the Guardian last week. At first sight this looks like wise counsel from the climate science adviser to Defra. But the idea that we could adapt to a 4C rise is absurd... The collapse of the polar ice caps would become inevitable, bringing long-term sea level rises of 70-80 metres. All the world's coastal plains would be lost, complete with ports, cities, transport and industrial infrastructure... Yes: 70m of SLR rise would be a disaster if it occurred within a century. Or even a millenium, at an even rate. But it won't. The entire issue here is how fast it would occur. Current estimates are ~0.5m over the next century. There are arguments that it could be more, but suggesting 70m on a timescale that matters is unrealistic scaremongering. Or just careless; who knows with these people.

[Note: for those following the OT comments: sorry folks, but this is *not* going to degenerate into yet another interminable discussion of the HS. You want to talk about the HS? Fine. Get your own blog - they're free. You're even free to post a link to it here. But this is my site, and one of the facilities I provide is a signal to noise ratio. You want unmoderated discussion? Go and drown in usenet, which I abandoned years ago -W]

More like this

Sea ice again. I was reading Gareth who had been reading Monbiot. And so I did too. After I'd waded through the goo and the dribble about Bush, the first item of substance was A new summary of the science published since last year's Intergovernmental Panel report suggests that - almost a century…
Yay, more nonsense about sea ice: the traditional "US Navy predicts summer ice free Arctic by 201x", where this time x=6. Does anyone actually believe this rubbish? If so, I have money just sitting around, bored, twiddling its little green fingers and waiting to take your bet. If "summer ice free…
Someone refererred be to As Effects of Warming Grow, U.N. Report is Quickly Dated . It's yet another piece of the std.nonsense that one swallow does a summer make, and anyone who feels like quoting the odd warming factoid can throw away the IPCC, whilst of course sneering at the people who say it…
I spent five years of my life in Canada's Arctic and not a day goes by that I'm not reminded just how powerful an impression those years left made on me. When I read the latest news on arctic sea-ice extent, I wish more people would recognize just how important what happens up there is to the rest…

We don't want your smoke, we want mirrors. Seriously what gives in the Antarctic? It wasn;t the ice shelf breakup.

[Dunno. Just looking at the anomaly, I was inclined to suspect a data error. But the long-term pix look sensible. I shall make enquiries -W]

Eli, without even looking I can tell you it must have been an increase in volcanic activity. :)

More seriously, I think the basic answer is that nothing much was going on then and nothing much is going on now.

By Steve Bloom (not verified) on 11 Aug 2008 #permalink

First up in the stupidity stakes is Tim Flannery, for his Words of warming in the Grauniad, in particular for by June 2008 signs of a great melt were emerging and a senior adviser to the Norwegian government was warning that this may be the Arctic's first ice-free year. Even in June, that was blatantly silly. Repeating it now is worse. The Arctic will not be ice free this summer. Or next summer. Or the one after that. Or... The rest of the article is full of the usual over-excitement but isn't very interesting.

But what did Tim actually write?

As I write, the rate of loss, while still well above average, has slowed somewhat. Yet even now it's impossible to predict. We can only project that if this summer's melt trajectory follows recent decades, by September this year the Arctic ice cap will have lost around half of its remaining ice, and be just 2.2m square kilometres.

There's a big difference between 'ice-free' and 'half its remaining ice' . Yes, the latter looks unduly pessimistic at this point, but that's no justification for your pretension that Tim is forecasting an ice-free Arctic this year.

[Tim is, as I said, repeating comments that I argue were stupid at the time they were made, and are even more obviously stupid now. Why is he repeating them? In context, its because he finds them scary and wants us to find them scary. It *is* possible to repeat them in a valid context, but only in the context of "these are silly comments, look how badly this absurd prediction has turned out". Tim isn't doing that. In fact, probably he is repeating them because they are in his quote-file for writing articles about sea ice, which does him no credit etiher -W]


NSIDC is talking about this year, that's one year, and what's happened in the last 10 days has amazed me (it's amazed others including Serreze). It makes me wonder if we have a handle on the real state of the ice up there (As do the Terra/Aqua images I've been looking at this evening.)

Coincidentally I posted on 1 August at RealClimate< /a > arguing for 2007 being a blip of the sort the models produce (Bitz etc). But what has happened since has unsettled me. Already the sustained area gap between this year and last has halved to around 0.3 million sqkm behind last year, when I would have expected it to maintain if not grow.

I'd prefer the sort of figure Stroeve has come up with (2030), but can't find good reason to totally discount Maslowski.

1) Why do you think the drop in perennial extent as shown by fig 3 of Nghiem 2007 will cease? (You can add March 2008 at a further drop of 1 million sqkm from March 2007 i.e 1.6 million sqkm at 03/08.)

2) Why do you find Maslowski's 2013 figure completely unbelievable? (For what it's worth I use 2013 as a lower bound, I don't think it likely.)

PS Orheim's statement (he's the "senior adviser to the Norwegian government") was from way before June, it was Late February, Dr Olaf Orheim head of Norwegian IPY Secretariat: "If Norway's average temperature this year equals that in 2007, the ice cap in the Arctic will all melt away, which is highly possible judging from current conditions."

Nghiem 2007:

[Why do you say Already the sustained area gap between this year and last has halved to around 0.3 million sqkm behind last year, when I would have expected it to maintain if not grow? I really can't think of any way you could have for forecasting that. The only thing I could think of in relation to this years ice is taht its probably thinner than last years and so likely to melt *quicker* not slower.

As to your questions:

1) I expect a long term trend of decrease of ice. But its looking more and more likely (though not yet certain) that 2007 for arctic ice will be like 1998 for global temperature.

2) M-2013 is just obviously wrong, to my mind. It grossly over-exaggerates the positive feedbacks. Mind you - I haven't read his paper. Is it available?


By CobblyWorlds (not verified) on 11 Aug 2008 #permalink

For Eli: What's been going on down south.

On SLR: in my humble opinion you would be nuts to assume that SLR is well constrained to 0.5m over next 100 years. Infrastructure plans should take much more into account - especially if it involves things like nuclear power stations.

[I didn't say it was well constrained. I said that was the best guess (OK, I actually said "Current estimates are ~0.5m " but thats not well constrained either) -W]

It's winter in the Antarctic and the anomaly isn't that ice is melting but that less ice than usual has formed this winter. Looking at absolute area makes this more clear than looking at the anomaly:

[Agreed. When I saw the anomaly pic, I assumed a data error. Looking at the pic you reference, there is no obvious room for error and no obvious discontinuity. I'm inclined to think there must be an interesting explanation for this years ice (or lack of it) - perhaps its warm there. Sadly I no longer have the tools to look myself -W]

Just a question. Isn't sea ice area a better indication as to how much ice there actually is? And if so, why is there so much concentration (excuse the pun) on extent, instead of area?

[Traditionally, its a lot easier to see where the edge is, since thats fairly clear. Attempting to work out whether ice within the pack is 90, 95 or 99% from satellites is tricky (see my papers...) -W]

Makes sense, but does that mean that the CT area graphs have a large(r) error range? It's just that, to me, the extent could be quite a bit larger between two years (generally, I'm not really talking about 2007 & 2008 here), but the area could be very similar and so the actual amount of ice could be the similar?

Does the extent difference, if area is the same, have any effect on future melting (via an effect on refreeze, say)?

As for your papers, do you have a (link to a) list...?

[ -W]

Arctic sea ice loss is slowing. Yesterday, loss was only 56,092 sq km against a loss just five days ago of 145,000 sq km (IJIS).

By HarryTheHat (not verified) on 12 Aug 2008 #permalink

I'm expecting to owe you some money this year, and of course will be happy to pay up. I can pay it out of my winnings from Joeduck. I bet him 50 euros that 2008 NH sea ice area, as reported by CT, would be less than 2005, and at the current rate I should be able to collect on Friday.
The August storms are really shaking things up; my current forecast for the minimum is about half-way between 2005 and 2007.
As for SH, I'm not sure what Eli is seeing. The last-year anomaly chart shows a decline but not really a remarkable one.

By Nick Barnes (not verified) on 12 Aug 2008 #permalink


My reasoning with regards the difference between "now" last year and this year is mainly based around the much more extensive sea surface warming last year. I've not got the the links on me, they're in the RC post I link to (where I screwed up the link in my above post).

Basically last year had already dropped back more during July (higher insolation), and even with the substantial increased heat of the sea, it's rate of loss had started to tail off by this time.

I had been seeing a reduction of rate of retreat polewards of Beaufort/Chucki as the cap got to about the 75th parallel. Now without the "thermal momentum" that was apparent last year there are notable inroads into the pack north of 75degN. Concentrating as I do on area (not extent) this seems significant as the areal minima has been between 27/8 and 15/9.

So yes, it's thinner, but as insolation falls off towards September it's still receding above 75degN where insolation is handicapped by low angle of incidence. I had been expecting the thin ice to have melted more rapidly earlier in the season.

Thanks for your answers:
I am not inclined to write 2007 off as an outlier, in terms of volume and the extent of perenial ice (Nghiem shows what looks to me like a "tipping point"). If I am right in seeing ice thickness as a damper we should see more outliers in the years to come.

Maslowski's paper?
I keep reading references to modelling from Maslowski, but cannot find anything published. That doesn't mean he hasn't been working behind the scenes for the US Navy (who have a tactical interest). The nearest I can get to public reasoning for 2013 from Maslowski is trend extrapolation. I link to a pdf at Hot Topic:…

With regards the Antarctic; I've been too busy to look at it, but Tenney Naumer over at RC's recent Arctic post has mentioned substantial warm anomalies around the Antarctic coast. On the plot she presents there's a lot of red.

By CobblyWorlds (not verified) on 12 Aug 2008 #permalink

Cobbly, try these:…
Has anyone in the modeling field actually asked Dr. M about their modeling work, and why it seems to differ? I'd guess they have access to unreleased ice draft data from the Navy submarine cruises, and can use but not disclose such. Whether it'd make any difference, I dunno.

In fact I think it's safe to conclude they _do_ have unreleased data, from this for example:

"... The skill of the model is evaluated by comparing its ice thickness output to actual sea ice thickness data gathered during the last three decades. This includes the model comparison against the most recently released collection of Arctic ice draft measurements conducted by U.S. Navy submarines between 1979 and 2000. Our model indicates an accelerated thinning trend in Arctic sea ice during the last decade. This trend is robust and independent of timescales for surface temperature and salinity relaxation. The validation of model output with submarine upward-looking sonar data supports this result. This lends credence to the postulation that the Arctic is likely to be ice-free during the summer in the near future."

Key phrase therein:

"... data gathered during the last three decades. ...includes ... the most recently released ... ice draft measurements conducted by U.S. Navy submarines between 1979 and 2000."…
AN: C22A-06
TI: Understanding Recent Variability in the Arctic Sea Ice Thickness and Volume - Synthesis of Model Results and Observations

Simple arithmetic suggests to me this recent publication thus also includes at least eight or nine years of unreleased data, and possibly more if some data from the entire three decades has not been released.

By Hank Roberts (not verified) on 12 Aug 2008 #permalink

Thanks Hank,

That doesn't imply to me it's secret data, both the US and British Navy have publicly released their data, though not specific tracks and dates (AFAIK).

That said, there's another "pessimist" Peter Wadhams who has close ties to the Royal Navy (virtually a free-pass on the Royal Navy's subs). So perhaps you have a point...

I'd love to see the sort of public presentation Zhang's team have done for PIOMAS done for Maslowski. PIOMAS is really rather good from what I've seen, yet Zhang's work seems to support the idea that 2007 is an outlier (I am currently reconsidering this and don't want to say more in a rushed lunchtime post).

By CobblyWorlds (not verified) on 13 Aug 2008 #permalink

OT. A very readable summary of the hockeystick saga.

Needless to say, McIntyre was right all along. The hockeystick is now dead and buried, although unfortunately not soon enough to prevent its exploitation by the ecofascists.

Does anyone seriously wonder why statistically literate skeptics do not trust climate scientists?

[Just looks like the same old ranting to me. Was there anything new? -W]

[Just looks like the same old ranting to me. Was there anything new? -W]

Ranting? The talisman of the IPCC and global warming alarmists everywhere is finally demolished and you call that ranting?

Anyway, yes, there is something new.

[You're not welcome to post vast tracks of McI here. If you're interested in interpreting McI, I suggest getting your own blog space - its free -W]

You're not welcome to post vast tracks of McI here.

I didn't. The first few quotes were from Bishop, which was the original link, and the one you accused of ranting. The quotes were included to A) demonstrate he wasn't ranting, and B) to help point out what was new. They represented a tiny fraction of his post. Then I posted a couple of small quotes from one of McIntyre's two posts on the subject, again a tiny fraction. All the quotes were interspersed with my own interpretation and explanation.

So no "vast tracts" of McIntyre whatsoever. Please put my post back up, or the only conclusion I can draw is that you are as complicit as everyone else in the hockeystick coverup.

[Yes indeed, I'm a signed up member of the Sekret Kabal. But seriously: if you want to post large amounts of other peoples words, you want your own blog, not my comments space -W]

page 6 of 11

Modeled Sea Ice Area, Thickness, and Volume
Decrease from 1997 to 2002:
Ice area by 15-18%,in agreement with observations
Ice thickness by ~35% (or 80cm from 2.3 m to 1.5 m)
Ice volume by ~33% (from 30x103 km 3 to 20x10 3 km 3 ),
which is twice the ice area

If this trend persists for another 10 years (and it has
through 2005) the Arctic Ocean could be icefree
in summer!

page 16 of 18

Between 1997-2004:
- annual mean sea ice concentration has decreased by ~17%
- mean ice thickness has decreased by ~0.9 m or ~36%
- ice volume decreased by 40%, which is >2x the rate of ice area decrease
If this trend persists the Arctic Ocean will become ice-free
by ~2013!

Now am I being stupid or does this not fit together very well.
1. How can the volume decline 33% if mean thickness has decreased 35% and the area has also declined?
(much less pronounced in undated version but I would expect circa 47% decline in volume instead of 40%)

2. Re: 'and it has through 2005' yet 15-18% in 5 years has become 17% in 7 years and 35% in 5 years has become 36% in seven years. Hmm.

3. Looking at graphs, is choosing 1997 as the start point cherrypicking?

4. If you drew parallel lines to the average ice volume decline line but one going through the minimum points and one through the maximum points then AFAICS you end up with a prediction of an ice free arctic in summer by 2007 and an ice free winter by 2017. This is extrapolating beyond ice free in summer so very dubious. Nevertheless a mere 10 years between ice free in summer and ice free in winter seems pretty spectacular and makes me inclined to think the drawn trend is too fast to be believable. (I'm probably mis-interpreting things here.)

So has mean ice thickness really decreased by 0.9m in 7 years?

Even if it has, that is a lot of lost insulation so that more heat can be lost during the winter allowing more ice to form. Is it therefore nonsense to draw a straight line as an extrapolation?

But seriously: if you want to post large amounts of other peoples words,

Hardly "large amounts". Compare to your original post which liberally quotes outside sources. Besides, you asked the question: "what's new"?

Your censorship is highly selective.

Anyway, here's the brief summary of "what's new" in my own words:

Wahl and Amman arbitrarily threw out data in order to get the RE validation statistic for the hockeystick to be "99%" significant. Without throwing out the data, the hockeystick "signal" is indistinguishable from red noise.

Or even more briefly: The hockeystick is bunk.

cce, your history draws false conclusions, particularly in the section titled "Replicating The Hockey Stick".

[Fine. Go argue it on his blog. Or start your own. I'm hoping you're going to get the message, eventually -W]

I'm well aware of the Auditors' conspiracy theories. I'm also aware that despite the thousands of words written by them every day they accomplish virtually no science. If Wahl and Ammann was "disputed" there would be papers inpress as we speak.

I know that this is OT for William's OP, but I just have to thank cce for that history of the Hockey Stick. The best summary I've seen so far. I'll delve into the rest of your site now. :)

By TrueSceptic (not verified) on 13 Aug 2008 #permalink

Heres some questions...
1 is 50m+ sea level rise in a century possible?
2 has it happened before?
3 how often did it happen?
4 what was the worst slr rate to have occurred?
4 why can't it happen now with the unprecedented GW situation?

Fine. Go argue it on his blog.Or start your own. I'm hoping you're going to get the message, eventually -W

You may recall it was you who asked:

Just looks like the same old ranting to me. Was there anything new? -W

So I get the message just fine: you're not actually interested in whether there is anything new.

Re the referenced trend extrapolation by Maslowski, IIRC the results announced last fall were distinct in that they were via an RCM.

mugsy, one reason folks are prone to ignore you on the HS issue is because you don't have much of a grasp on the broad scientific context. Consider e.g. the multiple threads of evidence that over the last ten years led to a fundamental change in our understanding of the MWP and LIA and the forcings that drove them. MBH get rather a lot of credit for having gotten it right at an early date. I should also point out that a more variable MWP and LIA would mean higher climate sensitivity, so it's not at all clear what the point of the whole conspiracy would have been.

While we're on the subject, I should add that you're also a bit thin when it comes to the political context. The IPCC misused the HS graphic (essentially to imply a level of knowledge about recent climate greater than was really the case), but that just makes even more obvious the pointlessness of a technical attack on MBH in the present.

By Steve Bloom (not verified) on 13 Aug 2008 #permalink

Steve Bloom, William asked what was new. I responded. He didn't like it and censored me. That's all. It's his blog. I really don't care.

But FWIW, in my opinion the hockeystick saga is mostly significant for what it says about the scientific integrity (or lack thereof) of the IPCC and Climate Science in general. They got it wrong, then systematically refused to release data and code to allow people to verify their results (or otherwise), closed ranks behind those refusing to release data, smeared anyone who dared question the results, and then fudged followup studies to try and cover up earlier errors.

In terms of the overall scientific AGW picture, I agree: the hockeystick is not a huge deal (although in my opinion it is more scientifically important than some others, eg McIntyre, think). But as far as the political picture goes, the hockeystick has been the talisman for a whole movement.

Personally, I will never yield to those who would use false scientific pretense in order to dictate how I should live.

"a more variable MWP and LIA would mean higher climate sensitivity"
Steve can you please explain this statement? People keep repeating it like a mantra but it makes no sense to me, unless of course one assumes, a priori, that CO2 sensitivity is 3K regardless of natural variation, which is illogical because you can only get a high CO2 sensitivity by downplaying other natural factors in the first place. For example; in the ice core data you get a high CO2 sensitivity only by downplaying the ice albedo amplifier. What am I missing?

[Response = forcing * feedback, to a rough level. If you decide that your historical response is larger, then your estimate of sensitivity from that goes up -W]

[Response = forcing * feedback, to a rough level. If you decide that your historical response is larger, then your estimate of sensitivity from that goes up -W]

I would have thought:

Response = Natural Variability + Forcing * Feedback.

So a more variable MWP and LIA could mean a higher natural variability.

[Natural variability, on the timescale of the MWP/LIA, doesn't just arise from magic. It could be internal noise, but current theory says its probably forced -W]

It could be internal noise, but current theory says its probably forced -W

That's a circular argument if you're trying to use the MWP and LIA to validate current theory.

[No, it isn't. Distinguish theory and obs. Theory is that response = forcing * sensitivity (I oversimplify). Various observations of forcing and response are then used to estimate sensitivity. MWP and LIA have been so used, or rather the whole last 1kyr usually -W]

If the MWP and LIA were real, global events, any one of the three variables (Natural Variability, Forcing, Feedback) could be different from what we currently think. There could be forcings we don't know about, or variability could be higher. The oft repeated glib claim that a more variable MWP and LIA imply higher sensitivity (feedback) presupposes that we already know all there is to know about those three variables, which to my mind rather puts the cart before the horse.

[Of course we don't know all about them. They are estimates, and used to estimate the sensitivity. The point is that changing your estimate of the MWP warmth, all else being equal, changes (increases) your estimate of sensitivity. No-one is claiming that this is the definite answer. If just funny that all the septics are effectively arguing for higher climate sensitivity without realising it -W]

Theory is that response = forcing * sensitivity

It has to be the case that response = natural variability + forcing * sensitivity, otherwise you couldn't have periods (such as at present) where (as far as we know) forcing has been constant yet temperature (response) has plateaued.

Now, it would be remarkable indeed if the natural variability did not have a long-term expectation of zero (short of crazy stuff like cutting a hole in the upper atmosphere to let the heat escape directly). But there may be some big time constants and big amplitudes in that natural variation. It seems to me that such things are at least as well/poorly understood as sensitivity.

So, if the MWP and LIA were large deviations with little change in forcing, that could be evidence that the climate's natural variability has a greater magnitude and larger time constants than is currently understood.

The point is that changing your estimate of the MWP warmth, all else being equal, changes (increases) your estimate of sensitivity.

Or, "changing your estimate of the MWP warmth, all else being equal, changes (increases) your estimate of natural variability.". It just depends on what "all else being equal" means.

If just funny that all the septics are effectively arguing for higher climate sensitivity without realising it -W

Ah yes, the old "septic" insult. Insults aside, I don't see how you can a priori claim that an increased MWP or depressed LIA is evidence of greater sensitivity but not evidence of greater natural variability. It could be either (or it could be there are unknown forcings, although I suspect that is somewhat less likely).

Occam's Razor, mugsy. As I said, you need to consider what else has been learned relative to the time before MBH that affects conclusions about the MWP and LIA. Present understanding is that they were globally modest swings driven primarily by variations in volcanism.

By Steve Bloom (not verified) on 14 Aug 2008 #permalink

Thanks William for the explanation. I see where the idea comes from now but it's wrong to assume that sceptics are unknowingly arguing for a higher sensitivity - they are arguing that the currently accepted theory is overly simplistic. It seems to be a prevailing attitude that what is not properly understood must be either ignored or made into a constant. This likely stems from the oft-repeated philosophical idea that climate is stable without humans but unstable with, when in fact the paleo records would suggest pretty much the opposite - per Ruddiman's ideas.

[Um, you seem to have imbibbed some magic... the oft-repeated philosophical idea that climate is stable without humans but unstable with is your own invention, not a common idea. they are arguing that the currently accepted theory is overly simplistic - no they aren't; at least not the anti-HS brigade, they don't articulate that well -W]

... mugsy ... Present understanding is that they were globally modest swings driven primarily by variations in volcanism.

Evidence, Stevie boy?

they are arguing that the currently accepted theory is overly simplistic - no they aren't; at least not the anti-HS brigade, they don't articulate that well -W

The Anti-HS brigade are just anti-bad-science and anti-corruption.

But as far as interpreting the MWP/LIA goes, oversimplification/lack of understanding is always what the argument has been about: what if they were caused by greater (not currently understood) natural variability, or not currently understood forcings? Ergo, how do you know we're not experiencing something similar today?

OT: The Jehovah's witnesses came by yesterday.

Answering the door, before they could say a word, I asked "Are you proselytizing?" (I could see their bibles clenched in their sweaty hands).

Blank stare in response (I knew "proselytizing" was a risky word to use - Americans are often vocabularily challenged).

Try again: "Are you evangelizing?".

Jehovahs: "We're Jehovah's Witnesses".

Me: "I am a committed atheist, so thankyou very much but I am not interested"

Jehovahs: "Do you care about global warming?"


At least one religion recognizes a kindred faith...

Re the evidence, mugsy, I was assuming some effort on your part to stay up to date with the science. Google awaits you.

By Steve Bloom (not verified) on 15 Aug 2008 #permalink

The following is taken from the latest presentation of Roy Spencer:
Aires & Rossow (QJRMS, 2003): "...the classical technique (from the electrical circuit theory) to analyze feedbacks is, by its hypotheses, very limited in its validity when applied to highly nonlinear, multivariate systems like the climate. Furthermore, the results of this kind of classical analysis are no more than a 'schematic' measure of feedback processes at system equilibrium, which may be very misleading."
Stephens (J Climate, 2005): "..different assumptions about the (climate) system produce very different conclusions about the magnitude and sign of feedbacks."

Hence, Spencer is even quoting peer-reviewed scientific press saying that the method is overly simplistic. If the assumption is wrong then so is the conclusion. I wasn't actually talking about the hockey-stick though because you asked us not to.

On the stable climate idea, yes perhaps in the end it's not scientists who push that stability idea so bluntly, but it does seem implicit in the assumption that any trend in the data must be from CO2. Longer term natural cycles are not considered because we don't have a mechanism for them. In other words a flat line for the 20th century is expected so the anomaly left to be explained is 0.6K and you have argued here that the 100 year trend is due to AGW. I'd say it was a clear assumption of stability in the absence of man is it not?

[but it does seem implicit in the assumption that any trend in the data must be from CO2. Longer term natural cycles are not considered because we don't have a mechanism for them... you still seem to be rather confused about the state of the science. There is no such assumption; and longer term trends are considered. Where exactly do you get this assertion from? -W]

But the crux is - is there such a natural variation in the system for which we have no real explanation - for example the little ice age event and the current recovery from it. Spencer's latest presentation attempts to show that PDO and similar natural circulations can explain most of the warming. If this pdo correlation is valid and it seems to be (D'Aleo and Atmoz also independently showed a temperature-pdo correlation) and since we have no correlation with CO2, then surely we can fairly say that there is a long natural cycle of unknown origin that explain most current warming rather well. The rebuttal would be to argue that pdo etc. results from the warming rather than causes it which is a bit flimsy.

[Sure, Spencer is desperate to find a non-anthro explanation. No-one believes it, for all the obvious reasons (the most obvious of which is that we already have a perfectly good explanation which works, so substituting a badly thought out one that doesn't isn't much of a gain, unless your sole objective is to remove the anthro component) -W]

For another example apart from yourself, Josh Willis is pushing much the same line rather strongly - ie ignore the last 5 or 10 years because the 100 year trend is what shows AGW. Again implicitly assuming that nature should flat-line. Of course the actual IPCC position amounts to contradictory double-speak; a) there is 0.6K/0.8K rise per century to explain, b) man's influence is clear only in the last 50 years.

Do you mean that we have a current explanation that slightly works and when reality deviates from obs we explain it by aerosols or noise or ocean cycles or instrument errors or some other unspecified natural variability, depending on the scientist. One would of course expect most scientists to ignore something that some of them have been pushing for up to 30 years. Human nature. Of course Spencer doesn't believe the anthro component is everything but he argues that natural variation has never actually been ruled out by any study either. All that has been said is that we don't know what else it might be. So to challenge people to come up with an alternative and then rubbish them with little apparent thought is somewhat dogmatic.

Cobblyworlds, thanks for the info (August 13, 2008 6:50 AM).
If you have links I'd appreciate pointers to sources.

William, are there forums where the modelers talk about the kind of thing, or do they pretty much run their own programs and publish? I've been puzzled over why Maslowski and students' work seems so different for quite a while, and over why there's so little mention of that group in public, and guessed it was because it's a US Navy program so probably doesn't hire the more, er, enthusiastic press officer types.

[Not that I know. Scientists talk at conferences -W]

By Hank Roberts (not verified) on 16 Aug 2008 #permalink

>[A little concept I just created in the comments but am so pleased with I'm going to put it here: is 2007's Arctic sea ice like 1998's global temperature?]

[I'll have to have a look at the numbers and draw some lines. There may be some scope for some more interesting bets over the years ahead -W]

1998 lasted over 6 years without being beaten. I'll bet 2007 will be beaten in 6 years or less if you are offering even odds.

What would be necessary for you to give up that view of 2007 like 1998 (Assume the extent is higher this year):
- Area minimum record this year?
- Minimum ice volume record this year?
- Monotonic decline in max ice volume from March 2007 to March 2009 inclusive?
- Minimum extent record in 2009?

Have you looked at those numbers and drawn those lines and got any suggestions for those interesting bets?

[I'd need to look at the stats to be sure. If 2009 beats 2007 (have we all given up on 2008 already?) then its not like 1998. But don't be too literal: I'm not saying the situation is exactly analogous. I'm just trying to put up an obvious point that many people are forgetting in the frothing excitement (tahts not a dig at you): individual anomalous years easily occur -W]


Sorry I missed your request, only just popped back.

The Zhang team's PIOMAS model is discussed by them here:
I can also recomend their study of 2008: See "Ensemble Predictions of Summer 2008 Arctic Sea Ice." on the above page (and pdf of study proper). Retrospectiveon that page has some ice-draft plots for the 1990s.

As for Wadhams: Google "Peter Wadhams arctic" without quotes (if no luck try Like Hansen a lot of what he says publicly is in the press (I suspect disagreement with that from our host). Sorry but what I say of him is based on snippets from loads of sources, he wrote an article in the Independent this June "Every time I visit the Arctic, the ice gets thinner..." that fairly well shows his pessimistic stance.

ARCUS has a forum (which is private). But as William notes a lot of discussion is at conferences (and by email or elsewhere in person i.e. the usual).

By CobblyWorlds (not verified) on 18 Aug 2008 #permalink

Here's another scientist with a view about natural variation. I'm sure you'll recognise him - from Realclimate circa 2005 (easily googled).

"[Response: Thats enough of this nonsense. We've discussed this ad nauseum in past posts. The trend in natural radiative forcing during the 20th century is negative. If you've got something new to add to the discussion, fine. Otherwise, don't bother posting this stuff. - mike]"

So perhaps I'm to conclude that of course it's not assumed to be flat-lining - it's assumed to be decreasing. You guys can't even agree with each other, never mind the IPCC official position, and then you say the rest of us are confused..

[You don't seem to understand the words you're using. "Assumed" means thought-to-be-but-not-checked or somesuch. Mike is saying is-and-has-been-checked (or rather, best-available-obs-indicate. If you're prepared to play fast and loose with the meanings of words, I'm sure you'll be able to get the answer you want, but it won't mean anything -W]

William, are we clear on which metric is being used for the bet?

[Not very clear. My idea is to declare it a draw if different plausible metrics produce different winners -W]

The original post doesn't seem to have specified one, but perhaps it was discussed elsewhere. I raise this issue because it appears to be possible that this year will show different outcomes between area (CT) and extent (NSIDC and others). That very thing has happened before, most recently in 2005.

By Steve Bloom (not verified) on 26 Aug 2008 #permalink

Steve, the bets are for area as at CT according to the post Betting on sea ice?:

I get the no-record side. And we use the record as presented by cryosphere today.

Looking at this:

You might want to check into the futures prices of carrots to reduce your risk.

Wow, this is getting close.

"Watching ice melt" sounds a bit like "watching paint dry" or "watching grass grow".

But if NSIDC and CT are "watching the ice melt", then William and his fellow bettors/bloggers are "watching people watch the ice melt" ... and we lowly commenters might be said to be "watching people watching people watch ice melt"!

It sounds like boredom by proxy, but it's actually sort of thrilling, in a slow-motion kind of way.

Frankly, Eli was going to roll his winnings over into next year, which should be a bit warmer given the solar cycle and with even less multi-year ice.

Aha, per this AP article the wild-eyed Jay Zwally has signed onto a five to ten year ice-free summer. He made some approving noises about Maslowski's results last December, but IIRC he didn't go quite that far. There are some other interesting comments from Zwally and others; tipping points are mentioned.

[Yeah, Zwally is well known as a wild-eyed zealot :-)) "Tipping points" remains a near-meaningless concept. "five to less than 10 years," is presumably quoting Hansen, and remains as meaningless now as it was then. Serrezes later comments are over-interpretation -W]

By Steve Bloom (not verified) on 27 Aug 2008 #permalink

I've generally stayed out of the will it/won't it be a new record minimum ice this year (having publicly just stated a long time ago that if I were to place a bet then I think the odds favoured W's position -- or words close to that effect).

Now, though, barring some weird change to curve shape (and I suppose a severe cold spell could do that), the sea ice extent (NSIDC) looks as if it will go below the 2007 level. What is still not clear is whether sea ice area (CT) will be more or less than 2007. Given that the ice area is close on ~300,000 sq km(??) behind 2007 at the moment, then on the face of it I'd say the 2007 minimum sea ice area record is safe for another year.

But ... what I find interesting recently is comparisons of the CT ice areas 1 year apart with, what, about 2 or 3 weeks (conventional wisdom) of melt to go. The area that represents ~25-70% ice concentration is, by eye (a notoriously bad indicator I'd agree if pushed), about double what it was a year ago at this stage. Of course, this is presumably partly the reason why ice extent appears to be heading south of its 2007 minimum; but, given a fair warm wind and current, perhaps it might also be one reason why 2008 ice area might give W a surprise yet, even if it is currently lagging by few hundred thousand sq km.

Of course, the currently higher ~25-70% ice concentration (this year compared with last year) presumably has another edge, because if it doesn't melt away then it will presumably consolidate to multi-year ice and act as heterogeneous nucleation sites for new first-year ice, perhaps leading to a better recovery in ice area than it did in early 2008.

Not long to go now. And, of course, there's next year to look forward to already, and the next, ....

I understand TPs to be a shift from one climate state into another where the new state exhibits a degree of persistence. It seems somewhat useful. To analogize, we could say that Mark S. seems to have slipped into a new rhetorical state. :)

Re the five to ten year business, they're both referring to (the as yet unpublished) Maslowski, although I understand that others have seen the latter's results.

[How do you know they are referring to M, if its unpublished? More seriously, I'll be interested to see it when it comes out. But I doubt it has much to say: from what I gather, its just another modelling study.

TP: we all know roughly what the vague idea is supposed to be; attempts to formalise it have failed, though, and I know of no use for it other than as a rhetoircal device -W]

By Steve Bloom (not verified) on 28 Aug 2008 #permalink

P., NSIDC says CT's metric isn't reliable due to the difficulty of distinguishing between melt ponds and actual open water. CT says they correct for that, but the recent sharp flattening of their trend line at the same time that temps in much of the high Arctic abruptly slipped below freezing seems like too much of a coincidence.

By Steve Bloom (not verified) on 28 Aug 2008 #permalink

Steve, the reliability of the CT metric is probably not an issue in light of what I was talking about (though it is as far as W's bet is concerned, as bettors all agreed to his terms), since I'm not overly interested in comparing the pros and cons per se of NSIDC over CT data. Unless CT have done something different in the processing between yesterday's sat pic and year-ago-yesterday's sat pic (perhaps they have?), then I'd considered I was comparing like with like with regard to the "~25-70% ice concentration" a year apart.

Looks to me like sea ice area is starting to look likely to match last years. Still might be a late bounce to save William's carrots and beer, so stay tuned. Anomaly has reached 2 million square km.

I'm more interested in extent, and that is somewhat more in doubt.

By Phil Hays (not verified) on 31 Aug 2008 #permalink

A word or two about area vs. extent. But first, my guess before the melt season was for extent not to set a new record this year. Not a huge relaxation, but some.

[Fair enough, but why? I based my argument just on the statistics. You? -W]

The thing about sea ice area observation is that during the melt season, the meltwater on the surface messes up the concentration observations by microwave satellite. It is a persistent low bias in the obs, and not a small one -- up to ca. 30% in concentration. For, say, weather modelling, this is somewhere between minor problem and asset (meltwater looks pretty much like ocean to the atmosphere; unfortunately, the difference isn't purely a meltwater phenomenon). For sea ice study, it's a pain.

Once the pack has broken up enough to have low concentration, chances are good you don't have a lot of meltwater (melt ponds) on top, so extent computations (at least with a 15% concentration cutoff) are significantly more reliable than area. Once the ice begins freezing up, the two go back to having more or less comparable reliability.

Looking at the current Cryosphere Today and NCEP sea ice fields (different details in data flow and processing, though fundamentally quite similar) the thing that stands out to me is that the NH ice cover looks ... mangy. Blobs here and there. The normal (meaning before the last few years) decay pattern in the Arctic was a nice little amoeba -- the edges expanded and contracted something like I imagine an amoeba might do (you might be able to get an idea of this by going to some of the annual mpegs at No blobs left hither and thither around the Arctic. The Swiss cheese collapse was an Antarctic trait. (Abetted by the cover being thin down there.)

As an eyeball test, this seems more important to me than details of exactly what area or extent we see. It's a change of character, not merely amounts. Can't make good objective tests, though, with what strikes my eyes as significant, so keep a look at the extents and, less so, the areas.

[Hmmm, that may be a good point. I have just been watching the area/extent graphs (and not much distinguishing them). Not long to go now... -W]

I hovered between thinking you were more likely than not to win and being unsure. "[But why? - W]" There seemed evidence there was weather involved in 2007 - mentioned in NSIDC sea ice news. But we were starting from a low summer point and there might also be an acceleration in trend involved.

Which was more significant was difficult to tell. So try another tack:

My feeling is that if GHG, SST and other forcings were somehow kept constant it would take some years for the ice to reduce to its equilibrium level. ie I expect there is some inertia there. If there is some inertia then what happens when you give a system a shock that moves it closer to its equilibrium? You don't really expect a big bounce back do you?

This still leaves it as:
The weather probably won't be as unusual, the trend probably isn't much compared to the weather noise and the low starting point is probably also capable of being swamped by extra winter freeze due to the low ice level and weather noise.

Given the swamped by noise considerations above, I don't really see the shock of 2007 inparting extra inertia to the melt. So the weather noise feels like it should be the biggest item and you should be more likely than not to win your bets.

Don't know if that makes much sense, but any comments on the inertia/inertia period or other aspects of the above?

["Inertia" doesn't directly apply, but analogues do. OTOH the sea ice itself appears to have very little "inertia" - at least within Hadcm3, according to work David Schroeder and I did just before I left: if you remove *all* the Arctic (or Antarctic) sea ice from hadcm3 then it recovers very quickly - within a few years; because no-ice is too far out of balance with the overall system -W]

>[if you remove *all* the Arctic (or Antarctic) sea ice from hadcm3 then it recovers very quickly]

Does that include recovery of thickness as well as extent?

[Yes: total volume and area -W]

I do wonder if the models are overtuned to be like that assuming the recent changes in extent are due to recent forcings whereas the land around artic has prevented extent spread and resulted in thicker ice than expected for the extent. The reality being this unusually thick ice has provided more of this inertia analog than the models show. I have no real evidence for this just a feeling it could be the case. It would be good to know if that can be ruled out.

[Overtuned? We struggled to get it right... -W]

My 'method' as such doesn't bear much scrutiny. Just a general tendency for successive years not to have new records, plus an eyeball regression against a couple of simple statistical models (but not quite as simple as putting a straight line through the whole record). I'll have to run the statistical models through something better than eyeball one of these days. Other fires to put out first.

FWIW Eli thinks the only way to recover to the 20th century averages is a fairly large volcano in the Northern Hemisphere. This is a problem with a deep memory effect.