2011-11-06 05:44:23Going Down the Up Escalator, Part 2
dana1981
Dana Nuccitelli
dana1981@yahoo...
69.230.101.55

Part 2.

Going Down the Up Escalator, Part 2

2011-11-06 05:59:07
Sphaerica

Bob@Lacatena...
76.28.5.93

I'd really love to see you take a shot at Spencer for his "polynomial fit ... for entertainment purposes only."

Or, as he says:

The 3rd order polynomial fit to the data (courtesy of Excel) is for entertainment purposes only, and should not be construed as having any predictive value whatsoever.

2011-11-06 10:03:30
dana1981
Dana Nuccitelli
dana1981@yahoo...
69.230.101.55

Hmm not sure how to fit that in there, Sphaerica.  Got any suggestions?

2011-11-06 13:12:28
dana1981
Dana Nuccitelli
dana1981@yahoo...
69.230.101.55
Note to self: I need to point out that in every case here, the "skeptics" are relying either on HadCRUT or RSS/UAH, which we just showed are all biased or potentially biased low.
2011-11-06 14:07:45
Sphaerica

Bob@Lacatena...
76.28.5.93

The only way to do it I think is too cheap a dig... after your discussion of step functions, suggest that even more elaborate functions can certainly be attempted by the very creative who are unconstrained by the limits of truth and proper methods, such as a third polynomial fit.  As far as you know only one scientist has taken things that far -- and even then, only for "entertainment purposes," whatever that means.  One can only presume that that web site is intended for entertainment purposes rather than the presentation, advancement and teaching of science.

2011-11-06 16:44:04
dana1981
Dana Nuccitelli
dana1981@yahoo...
69.230.101.55

Yeah I think that's too much of a dig, because he specifically says it's for entertainment purposes.  Yeah it's dumb, and it's clearly an underhanded tactic (putting the idea in peoples' minds anyway), but it's hard to criticize him for something that he says isn't intended seriously.  He's got a built-in excuse.  And I did take a little shot at him in the post for blaming global warming on ENSO, so it's not like he's getting off scott free.

Second note to self - at the beginning I need to add a link to the weekly digest, which will discuss all the links to Part 1 and the escalator graphic.  We've got a tweet from Michael Mann, a tweet and re-post by Romm, and a link from Andrew Revkin's Facebook page so far.  Monday we'll have a re-post on TreeHugger, but we'll have to save that for next week's digest.

2011-11-06 17:01:14link to an early AMO paper
jyyh
Otto Lehikoinen
otanle@hotmail...
85.76.7.139

"The concept of a short-term "climate shift" may have originated in Tsonis et al. 2007 and Swanson and Tsonis 2009," ,

The Regime Shift idea is much older, dating to the time of the first (danish?) AMO-studies (early 1990s?, but I've forgotten by whom) but as you say 'may' it may be ok.

F.e. http://journals.ametsoc.org/doi/pdf/10.1175/1520-0442(1998)011%3C1906%3ANHIVAC%3E2.0.CO%3B2 considers regimes and decadal cycles, but does this only for Northern Hemisphere, which would (as I understand it) be acceptable and good science even nowadays. Global Warming is global and hemispherical phenomena are just that.

2011-11-06 20:53:17
alan_marshall

alan.from.tas@gmail...
114.73.32.59

Looks good to me.

2011-11-06 21:17:30
MarkR
Mark Richardson
m.t.richardson2@gmail...
134.225.187.197

Perhaps add another self generated graph?

 

 

If you have a linear trend + a sine (or a few sines) + noise, you can easily fit 'step functions'. If you show that you made this up, and that post-hoc you can claim it looks like a set of steps then it's more obvious that it's not a bright thing to do. At least, to me...

2011-11-07 04:06:14
KR

k-ryan@comcast...
69.138.165.234

I would suggest a 'Tamino' style approach:

Generate a series with a linear trend and random noise, then show that you can fit some step functions to it.

2011-11-07 04:08:25
Riccardo

riccardoreitano@tiscali...
188.152.84.240

Fig. 4 does not come from Church et al 2011 as referenced. The image itself is quite heavy (1 MB).

2011-11-07 07:34:33
dana1981
Dana Nuccitelli
dana1981@yahoo...
69.230.101.55

Okay, I created the simulated temperature dataset suggested by Mark and KR - a linear warming trend plus two cycles plus noise, and fit it with a step function, then with a linear trend (another GIF).  I think it makes the point quite nicely.

I also added the note that the "skeptics" are relying on data that is either known to be or very well may be biased low, and added a 'those who forget history are doomed to repeat it' reference at the end.  I also changed the text to just say that the 'climate shift' concept might originate from Swanson and Tsonis, since the concept is older, as jyyh notes.  I'll add a link to the Weekly Digest when John H posts it.

Riccardo - I clarified that the data from that figure comes from Church and White (which John used to create it).

Any other comments?

2011-11-07 10:40:20
Rob Painting
Rob
paintingskeri@vodafone.co...
118.93.143.213

Dude, you keep eviscerating the deniers like this, and you'll deflect a lot of the denier-hate away from JC. Which is nice.

2011-11-07 10:42:21Having fun with animated graphs
John Cook

john@skepticalscience...
130.102.158.12
I can see you're having fun with animating graphs, Dana. We need to get you on Macromedia Flash - who knows where that might lead :-)

Seeing as the step function argument is actually appearing around the place, I'd suggest adding it as a new argument and adapting this blog post into the rebuttal.

2011-11-07 10:54:29
dana1981
Dana Nuccitelli
dana1981@yahoo...
69.230.101.55

Rob - tell me about it.  I think John's been enjoying all the hate directed my way instead of at him for a change.

Haha yeah John, I loves me some GIFs :-)  You can thank KR for giving me the idea and pointing me to some free GIF software.  Easy to make and effective - can't go wrong there. 

Good idea about the rebuttal to the step function.  Kind of a strange myth though - "global warming is just a step function caused by natural cycles"?  Or how about "It's a climate shift step function caused by natural cycles"?  I'll try that - let me know if you can come up with a better myth title.

2011-11-07 11:21:25
Rob Painting
Rob
paintingskeri@vodafone.co...
118.93.143.213

What free GIF software are you using Dana? 

2011-11-07 11:48:30
dana1981
Dana Nuccitelli
dana1981@yahoo...
69.230.101.55

It's called GIFBuilder carbon.  It's a little glitchy to be honest - the preview rarely works, for example.  But it gets the job done.

2011-11-07 11:48:55
Daniel Bailey
Daniel Bailey
yooper49855@hotmail...
97.83.150.37

Nice job.  Particularly liked Fig. 6

2011-11-07 13:31:31
KR

k-ryan@comcast...
69.138.165.234

What is the vertical dotted line in the animation, on the 'step' fade-in?

2011-11-07 13:39:02
KR

k-ryan@comcast...
69.138.165.234

Re: "...thank KR for giving me the idea" - I appear to have created a monster!

Seriously, though, a bit of animation can combine multiple images into one, connecting them in a very visceral fashion. Hence, I suspect, the Le Monde use of the Sphaerica graph/escalator illustration. They certainly can be overdone - if the animation doesn't add to the communication, if it's cutsie rather than communicative, and (a horrible diagnostic) if it includes any dancing teddy bears or happily smiling bees, it's a bad idea. Just something to watch for...

Animated GIFs are very easy to generate, and don't take a lot of backing to be effective - Flash does a good job too, but (IMO) tends to require too much software to drive them. iPhones, for example, are still not happy with Flash.

But do not, do not let your 12yr old daughters advise you on graphic decorations, color choices, or accompanying sound tracks.

2011-11-07 15:32:30
dana1981
Dana Nuccitelli
dana1981@yahoo...
69.230.101.55

KR - it's just a vertical line separating the steps.  Not really necessary, but just something to guide the eye.

2011-11-07 20:42:12
Riccardo

riccardoreitano@tiscali...
192.84.150.209

This theory of the step function amounts to no more than an average over a variable time range plotted as a histogram; if you join the dots (also metaphorically) you get a trend. But while averaging of a fixed range means smoothing the data, leaving it variable has no meaning at all, neither physically nor statistically.

2011-11-07 20:51:51Me enjoying Dana hate? Never!!
John Cook

john@skepticalscience...
58.170.57.120
Say it ain't so! Actually, I do take a perverse pleasure in it but not in the "thank goodness it's not me" sense. More because hatred from science deniers is a measure of your impact so More intense hatred = more effective Dana.

Re the myth title, take the most succinct version from the denialosphere examples.

2011-11-07 20:53:32
nealjking

nealjking@gmail...
91.33.120.187

dana,

It's my impression that Flash will never work on Apple products: for marketing reasons, Apple doesn't want it to.

2011-11-08 01:27:06
KR

k-ryan@comcast...
69.138.165.234

Oh, and thumbs up

2011-11-08 07:40:10
Sphaerica

Bob@Lacatena...
76.28.5.93

Dana, if you're a Macintosh user (which is my guess if you use GifBuilder carbon) then GraphicConverter is a good tool to add.  It's more manual than GifBuilder, I think, but adds a ton of capabilities.  It can be a tiny bit buggy, but not horribly so.

It's a general graphics utility with basic drawing functions, elaborate color/depth manipulation functions, and a good frame-by-frame editor for animated GIFs.

If you buy the registered version its batch conversion function is awesome (letting you, for instance, with one click change the image type or compression of a whole folder full of images, even applying filters for scaling, contrast and such along the way, or converting a series of still images into a Quicktime movie).

2011-11-08 15:35:58
dana1981
Dana Nuccitelli
dana1981@yahoo...
69.230.101.55

Thanks Sphaerica, I am indeed a Mac user, and I'll check that out.

2011-11-08 16:08:56
Sphaerica

Bob@Lacatena...
76.28.5.93

nealjking,

Flash also sucks.  Resource hog.  Buggy.  Requires an idiot-savant to program.  But, more than anything, HTML5 functionality combined with Javascript will replace 99% of what it does, but using standardized technology so one doesn't have to count on one company (Adobe, Microsoft, Google, Apple, whomever) to maintain it.  Gone will be the days of having to constantly install a new version of Flash Player.  Hopefully gone some day will be the days of 10 animated advertisements consuming 90% of your CPU resources and making a page take as long to load now in the age of broadband as it did when we had to dial in over phone lines.

Can you tell I'm a Mac user, and I don't particularly like Adobe and Flash?

Now if we could only also kill Internet Explorer, the world would be one step closer to having matured enough to tackle climate change.