I’m kind of excited! I think someone has decided to jump in and start doing calculations, and should he permit me to mooch some of his results, that will speed me along doing comparisons! Naturally, I’m hoping I can get people to help Chad.
What’s Chad doing?: Chad’s is examining global mean surface temperatures and and comparing them to IPCC projections.
As there is quite bit of effort involved in locating model data, downloading, and post processing, he’d like to be able to:
a) communicate results (preliminary and/or final) and
b) get people to locate scripts, methods and /or check whether the results he has seem correct.
I suggested one good way to do this is to set up a blog, and ask people. The blog gives him a way to describe what he’s done, and ask concrete questions in a way that people can understand. You can visit Chad’s blog here:
Scientific Prospective and read the questions he’s asking. These include:
- Compiling cdf2asc.
- He’s having trouble applying OLS and getting the correct autocorrelation for the lagged residuals on the “model data”. (I think he’s using Matlab.)
Meanwhile, you can tell Chad is goal oriented– he does have these figures:
Other issues
.
Reading Chad’s method of creating the figures for anomaly, it sounds like his approach is fine. However, there are quite a few steps involved in going from monthly lat/longitude gridded data to obtaining the anomalies. So, if any interested third parties are interested, they could try replicating the process of getting the anomalies, that would be great. Afterwards, we can keep track of scripts to do things, and show they have been verified and validated. And.. we can also have reasonable “trusted” monthly anomalies people can use. (By “people”, I mean,… oh.. say me! But there are other people who I believe are interested too. 🙂 )
Otherwise, if anyone could locate resources where monthly surface temperature anomaly predictions from individual models are already available from “standard” groups (NASA, IPCC etc.), that would be great. (The second step was on my “to-do” list, immediately after dealing with the issue of overlying measurement uncertainty on hypothesis tests. Vistiors have dropped links in comments, but I hadn’t followed up on all of them yet, because I didn’t want to interrupt the specific self-assigned task I was doing! )
As we find different resources, I can also create a list of resources, along with text actually describing what’s there, and the hurdles required to get the data!
Update: WordPress won’t let Chad’s file load properly. So, we’re trying it at my blog. If you want his data, click : Chad’s Zip file (http://rankexploits.com/musings/wp-content/uploads/2008/07/post1_code_and_data.zip)
I hope Chad won’t be left hanging (sorry couldn’t help myself). CA would be a good source for scripts but they’re mostly in R.
His GISS Model output seems to suggest a huge solar effect (2006 was the expected minimum back in 2000) – much larger than Gavin or others have suggested. This could be interesting.
Thank you for the link Lucia. The blog is called “Scientific Prospective” not “Scientific Progress”.
You may have to justify using GISS data in any temp analysis from now on. Is this a fair comment in the light of CA’s analysis and others see here: http://www.telegraph.co.uk/opinion/main.jhtml?xml=/opinion/2008/07/27/do2708.xml
Chad,
I wish you every success with your venture.
Be aware though of what you might encounter. I just got banned from Tamino’s ‘Open Mind’ blog because I queried the concept of GMST and the surface coverage of GISS. ‘Open Mind’ – forget it!
Chad’s blogroll seems to indicate that he considers anyone who is skeptical of the AGW orthodoxy an idiot. In fact, Lucia’s link looks a little out of place there.
And why he is using GISS as opposed to HadCrut3, considering all of the weaknesses that have been pointed out about GISS, is another good question.
Tilo-
Everyone gets to look at data. If he finds and explanation for the flat bit that’s consistent with 2C/century, that’s fine with me!
Lucia:
“Everyone gets to look at data.”
Yes, and some people get to torture it until it confesses.
“If he finds and explanation for the flat bit that’s consistent with 2C/century, that’s fine with me!”
That would require a physical explanation. Not something that he is likely to find in the models or Hansen’s dubious temp record.
Tilo:
I didn’t use gisstemp in anything in that post. I noted that they only give a mean gmst for their base period and not for each month thus I couldn’t use it in any analysis. I’d like to get such numbers for all the data sets so I can adjust them to have the same baseline. Yes, I think gisstemp has some problems that would make any analysis utilizing it fairly suspect in many people’s minds. No, I don’t believe anyone who is “skeptical of the AGW orthodoxy an idiot.” I do think there are plenty of people who just don’t understand the science behind it to such an extent that idiotic things tend to come out of their mouths.
Dave:
I saw your exchange with Tamino. I gotta tell ya, I like his blog a lot. I especially like it when he blows up on people. I think he was pretty tame with you.
Chad,
Sorry my dust up with Tamino was only ‘tame’. I suppose he could have been a lot nastier, as he has with others in the past. The problem I have with that is, a) it is not necessary to be nasty – how many people whom Tamino has done that to do you think go away and say yeah, he was right, I’ll change my viewpoint? and b)I thought science proceeded on the basis of scepticism, yet sites like RealClimate and Open Mind don’t really want to debate issues, they only want to reinforce their own, some might say myopic, point of view.
Chad– I found a resource that already provides monthly averages. I was too busy to blog yesterday– but this will save us both a lot of time.
Lucia, I know how to extract NetCDF files into R. It’s not a bad format once you get the hang of it. I have some templates and, if you give me an exact url of one of the files that’s being sought ,I can probably set up an extraction script in a few minutes.