Monday, June 03, 2013

This just in: Earth is round!!!

The April 2013 issue of Physics Today features an article entitled Earth's land surface temperature trends: A new approach confirms previous results, by Barbara Goss Levi, on the Berkeley project for measuring and reconstructing the Earth's temperature, started by UC Berkeley physicist Richard Muller, and his daughter Elizabeth, enlisting the collaboration of scientists at UC Berkeley, Livermore, and Oregon State University. Their project was modestly entitled BEST: Berkeley Earth Surface Temperature. (BEST has not addressed ocean temperatures, but they plan to do so.)
Before BEST, there were three earlier (and continuing) projects: (1) the Climatic Research Unit (CRU) of the University of East Anglia, U. K. working with the Hadley Center of the UK Met Office; (2) NASA's Goddard Institute for Space Studies (GISS), and (3) NOAA's National Climatic Data Center (NCDC). These three projects obtained results that are in agreement with each other. And they all show a global warming trend in the last century or so, and they are in agreement with each other.

Muller's BEST is the fourth effort at finding the Earth's global temperature and its trend. The BEST results are in agreement with the earlier three results for land temperature.

The reaction may be, "ho hum". Is there a point to the BEST project? I think so. The Berkeley team addresses some specific criticisms made of prior analyses: not enough data and allegedly flawed procedures for correcting data discontinuities (e.g., move of a station to a new location). In addition, Muller has pulled together 14 previously compiled databases. Gavin Schmidt of NASA's GISS credits BEST for releasing the data in a consistent way. BEST has put on its website not just the raw data but also the computer code for analyzing the data.

Some 20 months ago, I covered this study, right here on the pages of Rabett Run. And I was hardly snarky at all. Well, maybe a little bit.

This really ought to be the end of the "trend deniers", who think the Earth is not warming. It should be the last of four nails in the coffin.


David B. Benson said...

But there is also the Japanese global temperature product. I think it is called JAXA.

David B. Benson said...

Well, round is in nearly an oblate spheroid.

John said...

David Benson, thanks for the tip. As far as I can tell, JAXA is the Japanese counterpart of NASA. The JAXA Wikipedia page claims that JAXA launched a Greenhouse Gas Observing Satellite (GOSAT) in 2008, but doesn't say anything about monitoring temperatures.

Ron Broberg said...

JMA's global temperature index

Martin Vermeer said...

I seem to recall there are also a number of amateur efforts, also producing the same result.

Martin Vermeer said...

...and then of course there are the satellite results, though they measure a slightly different thing. They also agree as well as can be expected.

Martin Vermeer said...

Sorry for rubbing it in

Martin Vermeer said...

> Their project was modestly entitled BEST

Actually it was titled, not entitled, which betrays a sense of entitlement ;-)

(Regrettably the English language has moved on to accepting this wrong-headed usage :-( )

Nick Stokes said...

There's also ISTI. They have been a bit quiet lately, but I'm sure we'll hear more from them.

Anonymous said...

entitle (also US intitle)

SOED gives:

1 Give (a book, picture, composition, etc.) a title or (formerly) a heading or superscription


M-W's Collegiate gives:

1 To give a title to

And it's of middle English/late middle English, so not exactly modern usage either.

Cymraeg llygoden

Anonymous said...

It turns out that it is not very difficult at all to replicate the NASA/NOAA/JAXA/Whatever global-average temperature results.

The global-warming signal in the surface temperature data is so strong that it practically "jumps out at you" even with the crudest processing techniques.

And for those who are interested (and don't mind waiting around to download a ridiculously large file -- 1GB), I have a "Global Temperature Virtual Machine" available for download at:

It combines a rudimentary gridding/averaging procedure with a clickable Google Map front-end that allows folks to "roll their own" global-average temperature results by clicking on stations on a global map.

I started off with working javascript/html kindly provided by Nick Stokes and used it to create a popup "control panel" that allows you to sift/sort stations by data record length, rural/urban status, etc.

For example, with the control panel you can display, say, only stations with at least 120 years of data. This allows you to "drill down" to the long-record stations needed to get results going back a century or more.

You can then generate results by clicking on individual stations or by using all the filtered stations ("batch mode").

I bundled everything up in a VirtualBox virtual-machine (VM) file so that the app could be installed/run on almost any Windows/Mac/Linux PC/Laptop with minimal hassles. (That's why the download is so big -- all the VM overhead).

Not the most elegant or efficient way to do things, but it was easy for me to put together from the resources I had at hand (zero cost, and I didn't have to make a career out of it), and it works pretty well.

All software and data (GHCN V3) are bundled together and ready to use almost "out of the box". Any newer PC/Laptop (about 6 years old or newer) with at least 2GB memory should work. OSX 10.6 or newer recommended for Mac users. Just install Oracle's free VirtualBox package (, import the "Global Temperature Virtual Machine" file via File->Import Appliance.., hit the VirtualBox "Start" button, and wait for the VM to boot up and initialize.

With the app, you can generate your own results from raw and homogenized data and then compare those results directly with the official NASA "meteorological stations" results.

Experiment with the app a bit, and you will find that it's virtually impossible to get results that contradict that NASA results, from homogenized *or* raw data. (provided that you select stations with decent global coverage).

Pick individual stations on the map -- watch your results converge amazingly quickly to the official NASA results as you add more stations.

The upper plot in the data plot panel shows the global average results; the lower plot shows how many of your selected stations actually reported data for any given year -- that way, you can correlate the quality of your results for any time-period with the number of stations that actually reported data for that time period.

Results are displayed via GnuPlot (launched with popen()).

You will see that once you get to about 30 or so stations scattered around the world, your results will line up with the NASA results quite nicely.

Full "quickstart" instructions are provided at the documentation link on the right side of the download page.

Once again, the download link is:

--caerbannog the anonybunny


Palaeoclimatology ? What palaeoclimatology?

Watts has just taken down the latest in this unending series of game changing scientific blockbusters , which , while it lasted , claimed the Greenland ice record must be all wrong because glacial ice only<a href="> started to accumulate at the onset of the Little Ice Age.</a>


Here's the fixed link

Anonymous said...

Russell, that is one premier collection of climate Galileos. Awesome work.

Rib Smokin' Bunny

Sou said...

You've nailed it, Russell. I was thinking of having a competition for the weirdest 'theory' later this year. They are all serious contenders.

The only way I figure to pick the real deniers from the fake deniers (Poe) is how long they keep it up, and whether every now and again they make a half intelligible comment. I've been convinced a couple were fake but then realised I was wrong, probably :)