Well, it turns out that a recent poke at this by Eli, brought a pointer to a place where this was mentioned three years ago,
Dr. Christy and Dr. Spencer of UAH asking about the public availability of the source code used to process UAH data. Dr. Christy replied:and in turn mentioned a name of someone who might have some information
We are in a program with NOAA to transfer the code to a certified system that will be mounted on a government site and where almost anyone should be able to run it. We actually tried this several years ago, but our code was so complicated that the transfer was eventually given up after six months.
So UAH source code isn't currently available, but they're in the process of working with a NOAA program to make it available. I followed up asking if there was a general ETA for this availability. He replied:So Eli googled John Bates and got a reply
I talked with John Bates of NOAA two weeks ago and indicated I wanted to be early (I said the "first guinea pig") in the program. He didn't have a firm date on when his IT/programming team would be ready to start the transition, so I don't know.
you can find details on code and download the code itself for Christy and for RSS MSU from this web page:
The last link sent me to a webmail address...
ReplyDeleteThanks fixed
ReplyDeleteNot sure if all the code is actually there. It seems to call in some Matlab stuff.
ReplyDeleteLet's make a little list. . .:)
ReplyDeleteWhat's with the line that says
ReplyDelete"goto Antarctica"
??
~@:>
Eli,
ReplyDeleteIt is going to take a long time to wade through their code-- Braswell does some aggregation and processing, those data are passed to Spencer's code and the results from Spencer's analysis are then finally passed to Christy's code. But then again, no one has suggested that estimating temperatures from brightness data would be trivial.
Very odd that Spencer and Christy were so shy about their code being out there, especially given that they knew people were interested.
From what I just read it is mostly in FORTRAN. Now some, not this bunny for sure, might suggest something nefarious is going on when Christy says in the comments "special treatment for hightest latitudes" [see tlt_3_5.4 2011.07.01.f] ;) Is that like Mike's "trick"?
There is a lot that needs to be vetted and cross checked in their code. So have at it bunnies. We want to get this right, right?
There is actually a real point to this and its not just UAH but also RSS. There are inconsistencies and it would be good to be confident that it ain't demons in the code. Eli will post something on this tomorrow.
ReplyDeleteMe? Audit code? You Josh!
ReplyDeleteSeriously, I don't "audit code". I never have. Don't ever plan to. Churn away at their code? As. If.
I must have missed it (and still be missing it), but where was the suggestion above that Lucia might "audit code"?
ReplyDelete~@:>
Lucia churns away. Steve audits. Both give Spencer and Christy a free pass.
ReplyDeleteAnd Michaels, Lindzen, Soon, Baliunas, Singer....
I eagerly await the guest poster on WUWT who will bemoan the lack of Independent Verification and Validation of the code, and the fact it's written in FORTRAN.
ReplyDelete"written in FORTRAN"?
ReplyDeleteThe idiot known as Poptech will be all over it. That was his main argument against climate models.
The high latitudes/Antarctic stuff may be a cluge to repair the problems that Eric Swanson found in 2003
ReplyDeleteHm... nice find Eli.
ReplyDeleteThere was a blogger, http://magicjava.blogspot.com, who spent a fair amount of effort trying to replicate the satellite record (as you will find, he was fairly "conservative" in his views on climate). He finally gave up due to road blocks. To bad he wasn't aware of John Bates efforts.
I know this is probably heretical, but I actually don't regard sub-annual variations in the UAH or RSS temperature indexes to be "true" temperature. The reasons for why I think this are too complicated to fully communicate in a short blog comment, but I'd suggest that anybody interested in the problem start with working out how they derive an estimate for TLT (it's far from a direct measurement).l
Then think about how different layers of the atmosphere respond differently to high-frequency phenomena like sudden stratospheric warming or Julian-Madden oscillations. My suspicion is the effective weighting of the atmospheric temperature profile that gives TLT is time varying (but probably fairly constant over periods of years, discounting effects of instrumentation drift)... so IMO for shorter duration fluctuations, it's not a "true" temperature though probably a climate-related quantity.
Also, I've no problems with people coding in Fortran. At least you don't have to deal with obnoxious memory limits like you do with the R language.
There is a pointer in the post to magicjava and indeed that was the hint that got this started. When mj was working on the problem the software was not available. It only showed up last fall and evidently was transferred to NOAA sometime in Fall 2011.
ReplyDeleteAUDIT ALL THE CODES!
ReplyDeletehttp://judithcurry.com/2013/03/16/open-thread-weekend-11/#comment-303426!
You might remind folks of who exactly told you about magic java's effort Eli.
ReplyDelete