The Blackboard

You left out the UAH upper confidence limit in Figure 1.

UAH uses data from Aqua, RSS doesn’t. Aqua has station keeping capability so no orbital drift which eliminates the diurnal correction from orbital drift. However, the anomalies are calculated from averages of data before Aqua was launched so it’s likely that the residual errors in the correction for the older satellite data show up as a small annual cycle in the current UAH data and a big cycle, starting about 2002, in the difference between RSS and UAH.

Dewitt

You left out the UAH upper confidence limit in Figure 1.

No I didn’t.

The order of operations with EXCEL affects which line is on top of which. The brown line is sooooo close to the blue dashed line you can’t see the blue dashed line. I know the values, and the brown line is just inside the blue dashed line.

However, the anomalies are calculated from averages of data before Aqua was launched so it’s likely that the residual errors in the correction for the older satellite data show up as a small annual cycle in the current UAH data and a big cycle, starting about 2002, in the difference between RSS and UAH.

This is a possibility. In this case, the uncertainty in computing the monthly averages from the 20 year baseline are frozen into the current anomalies.

Of course, we can’t be certain this is the reason. The other possibility is that problems are introduced patching from instruments during the baseline period. Both RSS and UAH had to patch. They each did it differently and obtain slightly different long and short term trends as a result of that patch.

I have to say that even though it might seem odd for me to favor a data set which appears to agree better with models over the alternative, I think there is pretty good evidence in the literature to favor UAH, in particular:
http://www.agu.org/pubs/crossr.....8864.shtml
I have seen no rebuttal of this argument that UAH is superior to RSS.

Another interesting point: Craig Loehle uses an interesting method to remove the annual cycle from the OHC data to get at the underlying trend:
http://www.ncasi.org/publicati.....px?id=3152
WHich may be applicable to analyses such as these as well.

I would expect differences in cross-calibration between satellites to produce offset and slope changes. You can see that when you plot the difference between UAH and RSS anomalies. But which geographical area you choose makes a big difference as well. For global anomalies, there seems to be a step change between 1992 and 1994. But that change is not at all obvious if you just look at NH or Tropical anomaly differences (land + sea). It does show up in the SH anomaly difference, though.

Aqua came on line in 2002 so the appearance of cyclic behavior after that date seems to be more than coincidental.

Aqua came on line in 2002 so the appearance of cyclic behavior after that date seems to be more than coincidental.

Yes. But it wouldn’t tell us which is “right” and which is “wrong”! If the issue is due to AQUA, it tells us AQUA is seeing a different cycle than the previous instruments. But does that mean AQUA is now getting the cycle right? Or does it mean AQUA is now getting it wrong? Both would result in the same symptom.

[...] Click here for the rest of her post. [...]

Lucia,

Given your background and diligence in monitoring the global average temperature, I thought you’d find the following to be very interesting:

http://news.concordia.ca/main_story/014941.shtml

Concordia researcher finds definitive link between carbon dioxide emissions and global warming

Damon Matthews, a professor in Concordia University’s Department of Geography, Planning and the Environment has found a direct relationship between carbon dioxide emissions and global warming.

Matthews, together with colleagues from Victoria and the U.K., used a combination of global climate models and historical climate data to show that there is a simple linear relationship between total cumulative emissions and global temperature change.

I’d bet my hat that AQUA is getting things right and the rest are wrong.

Frank K, no doubt you have read their paper in Nature? And you therefore know that the key results come from climate models, not from observations? Or do you just read press releases?

Just in case anyone still follows GISS temps, the may figure is out:

+55 vs April’s +46

http://data.giss.nasa.gov/gist.....s+dSST.txt

GISSTEMP increase…mostly due to their Antartica data which is crap??

Hi all,
Lucia, can you help me with the phrasing for general readers of examiner.com? If I said something like,

“Warming does seem to be occurring, but for the past decade the general rate of warming is at 2 degrees Celsius per century” is that a) correct, b) understandable and c) complete, in the sense that earlier trends don’t swamp it?

I’m sure you’ve probably written this a thousand times, but what is the per century rate since 1979, for example?

Many thanks

Tom:
Warming does seem to be occurring, but for the past decade the general rate of warming *appears to be less than* 2 degrees Celsius per century.

For UAH the trend since inception in December of 1978 is 1.26 C/century. and provided in the footer of their data page (http://vortex.nsstc.uah.edu/pu.....glhmam_5.2)

RSS starts in January 1979, and doesn’t post the trend themselves. I get a trend of 1.54 C/century. Their raw data is here: http://www.remss.com/data/msu/....._v03_2.txt

Jonathan (Comment#14427) June 11th, 2009 at 1:03 pm

“Frank K, no doubt you have read their paper in Nature? And you therefore know that the key results come from climate models, not from observations? Or do you just read press releases?”

Hey Jonathan – yes I know the results come from (mostly) climate models and this is just more AGW BS. I thought Lucia would get a kick out of the “simple linear relationship” between CO2 and global temperature…

By the way, the CO2 you just exhaled just raised the earth’s temperature .0000000001 K ! LOL

Okay–you guys get a preview:

Global Warming–at a glacial pace

The University of Alabama at Huntsville and the Remote Sensing Systems measure the Earth’s temperature indirectly, being satellite measurements of radiance at particular wavelengths.

Both sets of satellite measurements do show global warming–but not as much and not as quickly as the most hysterical of warming alarmists would have you believe.

According to Lucia Liljegren, an atmospheric researcher with the Ames Laboratory at Iowa State University, trend analyses of global temperatures since 1979 (when satellite measurements started) are between 1.26 and 1.54 degrees Celsius per century.

As I wrote in a previous article, Svante Arrhenius predicted a 2.1 degree Celsius rise in the earth’s temperature if we doubled CO2. Of course, this was in 1905, and maybe science has improved since then… but the work of one or two other scientists of that era has proved fairly robust, so maybe we should start paying attention.

The sad thing is, as I also wrote recently, if we double CO2 and the temperatures rise 2 degrees Celsius, it doesn’t stop there. We don’t turn off our engines and say we’ve finished the race. We, umm, keep on emitting CO2.

The problem is real. It just isn’t as bad as some people want you to think.

Tom–
The resume is off! My husband was an atmospheric researcher at Ames Lab when I was an assistant professor in the AEEM department. I work part time at Argonne National Lab, but mostly I blog. I avoid using my professional affiliations at the blog because we are supposed to avoid that when speaking purely for ourselves.

Frank K, apologies, must recalibrate my sarcasm detector; it’s been a long day.

The idea of measuring CO2 output in femtoKelvin has been around for a while. I think I first saw it in the original version of the essay republished at http://kino-ap.eng.hokudai.ac.jp/fK.html

That’s what happens when you scrape info off these intertube thingies… noted! Any other impurities unclarities imprecisions and dark matter in general?

Global Warming–at a glacial pace
The University of Alabama at Huntsville and the Remote Sensing Systems both analyse measure satellite measurements of the Earth’s temperature, albeit indirectly, as the satellites measure radiance at particular wavelengths.
Both analyses of satellite measurements do show global warming–but not as much and not as quickly as the most hysterical of warming alarmists would have you believe.
According to Lucia Liljegren, principal contributor at The Blackboard (a weblog that is truly invaluable to anyone following this debate), trend analyses of global temperatures since 1979 (when satellite measurements started) are between 1.26 and 1.54 degrees Celsius per century.
As I wrote in a previous article, Svante Arrhenius predicted a 2.1 degree Celsius rise in the earth’s temperature if we doubled CO2. Of course, this was in 1905, and maybe science has improved since then… but the work of one or two other scientists of that era has proved fairly robust, so maybe we should start paying attention.
The sad thing is, as I also wrote recently, if we double CO2 and the temperatures rise 2 degrees Celsius, it doesn’t stop there. We don’t turn off our engines and say we’ve finished the race. We, umm, keep on emitting CO2.
The problem is real. It just probably isn’t as bad as some people want you to think.

(I will be using quite a few links in the story.) Good to go?

umm, first sentence corrected to read ‘both analyse measurements.’

Well, any other fact checking will be on the live story–check it out (and feel free to comment–I can still edit for accuracy…) at http://www.examiner.com/examin.....acial-pace

Thanks for your help!

Lucia: “RSS starts in January 1979, and doesn’t post the trend themselves. I get a trend of 1.54 C/century.”

They post the trend on this page. It agrees with your trend. (In my opinion, the linear trends are virtually meaningless.)

Tom–
If it’s your opinion, you can say “isn’t” at the end of “the declines indicated in the paper may or may not be occurring, and the regional locations may not dovetail with locations where wind energy is favorable. So, the declines that might be occuring may not have that strong effect.”

But “may not be” is more accurate.

Hi Lucia,

I didn’t get that deep into it, so that point is moot–until the next article…

That’s ok. I was just answering “but what is the per century rate since 1979, for example?” The values since Jan 1 1979 or Dec 1978 are available from RSS and UAH, but I can also always compute them based on the data.

Whether or not everyone likes least squares trends, it is a metric.

Nitpick: RSS and UAH start in Dec 1978. For some reason, RSS doesn’t report the first data point.

Chad,
Interesting. I assumed that for some reason, they just don’t do anything in Dec. 1978. I wonder why they don’t report it? Strange. . .

Ordinarily I’d suggest emailing Christy with any questions regarding satellite temps, but in this case Mears or Wentz seem more likely to know why they don’t report an anomaly for a month that UAH does.

Try support@remss.com maybe?

I’m happy to see as many measurements as possible. I didn’t mean to discard UAH altogether but just that the cycle will definitely show up on a monthly basis. And indeed I think we can safely say that the two satellite datasets are largely in agreement, no need to support one against the other.
Accepting or not the last eight years of trend as a demonstration that models are off is a completely different matter.

[...] Lucia’s The Blackboard tracks global temperatures since 2001, as measured against predictions of a warming of 2 degrees this century (brown line): [...]

Tom Fuller,
“Svante Arrhenius predicted a 2.1 degree Celsius rise in the earth’s temperature if we doubled CO2. ”
Are you entirely satisfied that this is what Arrhenius predicted? If you checked again, maybe you would find that there is even a better fit between Arrhenius’ prediction and UAH/RSS observations.

Hi Jan,

My understanding of Arrhenius’ calculations is that in 1896 he predicted an increase of5 degrees Celsius, and that upon recalculating in 1905 he reduced his predictions to 1.6 degrees Celsius supplemented by half a degree from various postulated feedbacks.
Does this not agree with your information? My source is Wikipedia.