2010 SIE predictions: Goddard and Watts

In anticipation of the first SEARCH Sea Ice Outlook report for 2010 (a synthesis report that describes the September sea ice predictions by several teams of scientists) I'll be posting a few articles with summaries of predictions by scientists and knowledgeable laymen. In this second instalment I'll be discussing the prediction for the 2010 Arctic sea ice minimum extent of none other than Steven Goddard and Anthony Watts from one of the major blogs in the Climate PR war: Watts Up With That.

First off I'd like to stress that I haven't spoken mildly of both Goddard and Watts on other blogs. I'm an alarmist, not just with regards to AGW but to the whole 'crisis cocktail' as I like to call it, but let's stick to the subject at hand: As an alarmist I of course am not a great fan of WUWT and most of its contributors and commenters. Not just because they disagree with my views - I don't think my views are that important or objective - but much more because of the way they choose to impact the public's perception of Global Warming.

So I haven't said and am still not saying nice things about Watts (and to a lesser degree Goddard) on other blogs. But there's a difference between writing comments and writing posts. With writing blog posts come different responsibilities, and so I will try not to make this an inciting post full of snark. In fact, I'll start off with writing something nice about Steven Goddard.

Steven Goddard is a smart guy, infinitely much smarter than I am (I can barely make a graph in Excel). And he has guts. He will go out and write up controversial theories, sometimes he's right, sometimes not, but I commend him for going out on a limb. Steven Goddard's main flaw however is that he does all of that on the basis that Anthropogenic Global Warming isn't real, a hoax, a scam. It's almost as if every time he starts writing he's thinking: something is happening, it cannot possibly be AGW, so it must be this! That mentality gets him into trouble more often than not, even though his line of reasoning can be quite sound.

This time around it's more serious. It's too early to tell, but it could very easily turn out that the Arctic sea ice fooled Goddard and Watts into some premature thumping of chests which they might regret come September. If their prediction turns out to be wishful thinking of course. Perhaps a time line can illustrate the cocky stubbornness that is causing a recent flurry of WUWT blog articles on the Arctic, which appear almost every second day.

The increase, the decline and Goddard's next top model

Everything was just hunky dory on February 9th when Watts and Goddard issued their prediction for the minimum sea ice extent for 2010: another recovery of 500,000 square km, just like 2008 and 2009, no doubt about it.

Their readership in the US and Europe had seen a relatively severe winter and the Arctic ocean had refrozen and showed signs of thicker ice than in previous years. This attitude of 'it's natural variability and natural variability alone' was reinforced on March 31st when a late freeze-up caused the maximum extent to increase considerably at a relatively late date. The NSIDC extent graph at that moment looked like this:

The sea ice extent even grew some more, but just as Steven Goddard had embarked on a series of guest posts called WUWT Sea Ice News, the tables turned. Not right away, we would have to wait till the end of April for the Arctic sea ice to start plummeting, leaving room for one last triumphant post on how normal everything was. It even contained one interesting quote by Watts:

Of course our friends will argue that extent and area don’t matter now, that only volume and ice quality (the rotten ice meme) matters.

Goddard wrote a couple more WUWT Sea Ice News reports, ending with: "The next few weeks will be slow news." Apparently this wasn't the case because all of a sudden the regular updates were being interspersed with extra information on the limited predictive value of sea ice graphs and the influence of shear. As an extra Goddard managed to cook up some graphs and announce us that we wouldn't be seeing an ice-free Arctic before 2065 (a lot more optimistic than the IPCC, it seems).

 

But as the 2010 trend line was cutting through all other years one by one, things were starting to get interesting.

On May 27th a WUWT commenter called Crashex wrote the following in a comment on the Shear Ice Decline post: "The Naval Ice center has a lot of information about the state of the Arctic sea ice. For other 'Ice Watchers', here is an article about the PIPS model used for short term forecasts." And presto, within 24 hours Goddard had a new article up showing that the PIPS 2.0 model showed ice thickness had increased. Yes, you read that right, a model (try and find a sentence on WUWT with the word 'model' in it that doesn't contain the acronym GIGO). In addition to that, Goddard wrote in a comment: "I could make an estimate of the ice volume by counting and numerically integrating pixels." Volume? Who needs ice volume? Only the alarmists talk about volume.

As there is another model that projects sea ice volume, PIOMAS, briefly discussed in my previous SIE prediction post, Goddard now saw it as his duty to attack the other model's credibility. He then proceeded by claiming Arctic sea ice volume had increased by 25% since May 2008, and as always ended his post in the usual spiteful tone:

A few weeks ago, when extent was highest in the JAXA record, our friends were asking for “volume, not extent.” Their wishes have been answered. Ice volume has increased by 25% in the last two years, and those looking for a big melt are likely going to be disappointed.

To anyone with some knowledge of human psychology it looks much more as if Goddard is afraid he might be disappointed. But no matter, as PIPS 2.0 is the way to go now, even though it is being repeatedly pointed out that the model is outdated and inaccurate. Only September will tell ,and so far Watts and Goddard haven't revised their prediction.

To get the gist people could go and read the comment sections of the WUWT articles (watch out for comments by Phil., Tom P., R. Gates, Anu and stevengoddard himself). I haven't linked to all the posts because this was enough work as it is. Let's just say Goddard is a very prolific writer, if prolific is the right word. In just the past week he has spewed forth WUWT Artic Ice News nr 8, an article on how an ice-free Arctic seems implausible and one on concentration vs extent. Perhaps his output will slow down if Arctic SIE goes up again. There might be a correlation there.

His last article contains a stunner that will soon undoubtedly make rounds around the blogosphere. Goddard added a CAVEAT for the first time (or I didn't read carefully enough while wading through the myriad of other posts):

Current conditions continue to indicate a larger minimum ice extent than 2007-2009. This could change if the weather is very warm, windy or sunny during July.

It definitely could change. Some clear skies for an extended period could be a real game changer, let alone winds blowing in the right direction. This caveat looks more like a safeguard.

The prediction

Goddard and Watts have predicted a minimum extent of 500.000 above the 2009 level. I guess they go by the data of IARC-JAXA as their extent graph is featured on WUWT's side bar. It's not entirely sure though if they mean absolute minimum (single date) or average minimum for the whole month of September. These were the September average minimum sea ice extents in previous years according to IARC-JAXA (between brackets is the absolute minimum):

2006: 5.91 (5.78) million square km

2007: 4.38 (4.25) million square km

2008: 4.83 (4.71) million square km

2009: 5.38 (5.24) million square km

So I think it's safe to say that Goddard and Watts predict a minimum sea extent between 5.74 and 5.88 million square km. They might get lucky and turn out to be right, but barring an extremely weird weather pattern unfavourable to melting, it looks as though the gentlemen have dug quite a hole for themselves and give the acronym WUWT a whole new meaning: We Use Wishful Thinking.

The problem will not be so much that their prediction isn't correct. That is only normal. It happens all the time as the Arctic sea ice is impossible to predict a few months in advance, which any polar scientist will tell you. The problem is that they refuse to use caveats in their sarcastic articles and persist in their prediction despite the change in conditions and the fact that if this summer sees weather conditions that are even remotely similar to those of 2007, it's not even sure the 4 million mark will be left untouched. Instead of adding some nuance they go and circle their wagons very aggressively by clasping everything they can get their hands on (like the PIPS-model, the Arctic Roos sea ice extent graph or the Cryosphere comparison maps) and ridicule commenters - like R. Gates, Phil. or Tom P - who post evidence to the contrary.

Time will tell if Watts and Goddard are gambling geniuses of the Galileo-variety or mere mortals like the rest of us. I will write more about my thoughts on that in an upcoming article called the Alarmist's Dilemma.

UPDATE: Goddard seems to have revised his prediction downwards in a comment following his recent WUWT Arctic Sea Ice News #9:

Seems like JAXA 5.5 Mkm2 is a decent WAG at this point.

Comments

[Peter Ellis]

One funny thing is the way Goddard's attempted to discredit the PIOMAS model by using... another model! Specifically, the older and less accurate PIPS2 model. The really funny thing is that (as Tom P shows), PIPS2 actually agrees well with PIOMAS, it's just that Goddard doesn't know how to calculate volume from the PIPS2 data.

Specifically, Goddard calculates volume as (thickness x area), which is the obvious thing to do. Unfortunately for him, it's not the correct thing to do. The PIPS2 output separates out thickness and concentration data.

Thus, looking at the thickness map, you might see a pixel displayed as having 2m ice thickness. If the concentration for that pixel is (say) 25%, then the true average thickness will only be 0.5m. The correct volume calculation for PIPS2 model needs to use both maps, and calculate volume = (thickness x area x concentration). Once you use the correct calculation, PIPS2 and PIOMAS start agreeing with each other.

[Hypnos]

Wayne Davidson over at EH2R is predicting a lower minimum than 2007. He's out there in the Canadian Arctic watching the ice daily, and so far he has a good prediction track record.

He says current conditions in the Arctic are setting up for a cloud-free summer (he calls it "Big Blue") which should greatly enhance the melt.

[Hypnos]

Aaand you mentioned Wayne Davidson right in the previous post. You can discount my previous comment. This one too.

Great blog BTW.

[Neven]

Thanks, Hypnos! What a coincidence you mention Wayne Davidson after I've just published a post on him. If you have any more tips for interesting theories concerning the Arctic, I'd love to hear it.

[Paul van Egmond]

'We Use Wishful Thinking' - Hilarious :-)

Seriously, good job on the blog, Neven! Keep it up and keep an open mind.
Best wishes,
Paul van Egmond

[Kevin McKinney]

Just looked at CT, which--as with NSIDC--is currently updating normally.

Couldn't help but shake my head at Goddard and PIPS. This pattern is not one of "concentrated ice," unless I'm flat nuts.

Nares & Fram both seem pretty "loose" to me, as best as one can tell at such a large scale, and there's a ton of melt in the Western Arctic. Hudson Bay is also declining rapidly.

Thinking about the AO data just posted here, I'm shaking my head once again. Very interesting times, indeed.

And, FWIW, yesterday's UAH lower trop anomaly update clocked in at .9 above last year. Still too early to count chickens, but very interesting indeed. . .

[Lou Grinzo]

Once again, nice site. Keep up the good work. The Reality Enhanced Community needs all the information sources like this we can get.

As for the pseudo-skeptics*, they sometimes strike me as a bunch of exceedingly sleazy used car or real estate salesmen. A car that's been in a flood has a "just scrubbed scent!', a house that was struck by a runaway truck has "a glorious new bay window!", etc.

* edited

[Neven]

Thanks, Paul, although I think it's practically impossible for human beings to truly keep an open mind. It's best to admit your mind isn't completely open and try to find out for yourself why it isn't. Nosce te ipsum. Oh yeah, and then be transparent about it. I'm an alarmist and I might be wrong. In fact, I'm hoping I am. And if not, I'm not really sure what can be done about it. First more people have to be convinced of the need that something needs to change. Then they need to fully understand what the root of the problem is.

Kevin, Nares and Fram look very interesting. I am preparing an animated gif to show the movements of the ice. 2-3 more days should do the trick.

Lou, I've changed your 'deniers' into 'pseudo-skeptics'. Although words like 'denier' and 'denialist' describe these persons very well, it has been spun into an insult. All we need is a good term that makes it possible to discern between the genuine skeptics and the not so genuine skeptics. For now I'm favouring 'pseudo-skeptics', but I might change it to 'people in denial' (which alludes better to the psychological state instead of the historical revisionism). I thought 'delayers' wasn't such a bad term either, but it's very spinnable. I don't have any official blog policy on that, as there haven't been any comments from pseudo-skeptics yet.

As for your used car salesmen analogy: that's precisely the reason I wrote this blog post. I had to get something down 'on digital paper' that describes the behaviour of some prominent pseudo-skeptics. And that's why it so good that Watts and Goddard have committed themselves to a hard number (instead of some meme along the line of 'the globe will start cooling some day'), deceived as they were by that late freeze-up. It has been fascinating to watch their behaviour ever since - especially Goddard's - which to me actually looks exactly like the behaviour the pseudo-skeptics accuse scientists of: arrogant, stubborn, over-confident, and a lot of circling of wagons. Again, fascinating.

It makes one almost wish the Arctic to become ice-free this summer! But more about that in a blog post later today.

[Neven]

Steven Goddard has revised his prediction for the 2010 minimum sea ice extent in his latest WUWT Arctic Sea Ice News (number 9):

"My analysis indicates the highest late summer extent since 2006."

The (absolute) minimum extent in 2006 was 5.78 million square km, so that's a little bit more than 500,000 square km over 2009 (5.74 million square km).

[Neven]

It's official, Steven Goddard has finally committed himself to a hard number and informed us on which data set it will be based:

I’m forecasting a summer minimum of 5.5 million km², based on JAXA. i.e. higher than 2009, lower than 2006.

[Neven]

Another interesting Goddard quote:

The ice extent graph is running parallel to 2006 and is offset downwards due to deficiencies in regions which are normally ice free in September. There is little reason to pay attention to the ice extent graphs until at least July.