Here’s the raw(-ish) data you’re analysing:

http://www.woodfortrees.org/plot/uah/mean:6/from:2002/plot/rss/mean:6/from:2002

It looks to me like there’s actually quite a strong 18-month-ish signal here, in both series, which I don’t have any explanation for – but it’s a short sample, and it could just be weather noise! But note it also exists in HADCRUT:

http://www.woodfortrees.org/plot/uah/mean:6/from:2002/plot/rss/mean:6/from:2002/plot/hadcrut3vgl/mean:6/from:2002

so whatever it is, it isn’t satellite-specific!

If one starts the dataset from 2002 when the UAH was changed to the Aqua satellite, (and doesn’t screw up the “window” function &;>), the difference in energy at the yearly harmonic(~7.5) between RSS & UAH is even more pronounced. I wonder if the peaks at 5 and ~16 that both RSS and UAH see are shared processing artifacts or represent real periodic temperature variations?
to save y’all from some typing, the URL is
http://www.woodfortrees.org/plot/rss/from:2002/window/fourier/magnitude/from:2/to:40/plot/uah/from:2002/window/fourier/magnitude/from:2/to:40/plot/nsidc-seaice-n/from:2002/scale:0.01/window/fourier/magnitude/from:2/to:40

Like those claiming warming in a short period, this is much ado about nothing.”

It’s significant because it casts doubt on the CAWG hypothesis, which strongly links increases in CO2 to increases in temperature, and on the robustness of the IPCC’s predictions. The disconnect that is being observed becomes more and more awkward for the CAWGers with each passing year, so our side is right to harp on it.

Also, I’m not quite sure why you’re doing a “mean:2″ on the temperature signal first.

Here’s your graph with this tidied up a bit:

http://www.woodfortrees.org/plot/rss/from:1990/window/fourier/magnitude/from:2/to:40/plot/uah/from:1990/window/fourier/magnitude/from:2/to:40/plot/nsidc-seaice-n/from:1990/scale:0.01/offset:-0.1/window/fourier/magnitude/from:2/to:40

I’m afraid in this case the two harmonic 20 (approx 1 year) peaks look rather more similar… Sorry!

Best wishes

Paul

Nice work by the way. The ice does affect the sensors, the satellites don’t measure the poles but UAH does some infilling. You can see the actual data area though in this video of RSS and UAH

http://noconsensus.wordpress.com/2009/11/03/rss-and-uah-videos/

The reason for the annual signal is the errors in diurnal correction which were also present in UAH before the UAH 2002 change to AQUA. This is why UAH is likely to also require a correction – it’s a more subtle point though and I haven’t heard that from Dr. Christy — yet.

This post should answer some of your questions.
http://noconsensus.wordpress.com/2009/10/26/bias-in-satellite-temperature-metrics/

What you have is NOT analogous to the derivative or anything even remotely close to it. Sorry. Back to Calculus I.

Actually this isn’t about el nino as much as this

You’re right – my original comment was a bit off topic. Sorry.

Yes, the next few years will tell us what the calculated “average temperature of the earth” was for those few years. :-)

They’ll also give us the trend(s) for the preceding years.

As for the future…eeeh, not so much.

I wasn’t sure how to answer the longer comment. The line is an RMS fit which I’m sure you’re familiar with of the equation y = a x + b to the data x.

The straight line is the trend of course as without it there is no definition of trend. Regarding the Fourier spectrum of UAH and RSS I’ve done some work with that from last year. It’s taken a year to figure out why the cyclic variations were occurring in UAH.

http://noconsensus.wordpress.com/2008/10/26/half-year-cyclic-variaition-in-rssuah-and-giss-anomaly/

This linked post above also has implications that UAH may require additional corrections as well.

For those interested, Lucia confirmed the results in the last graph at this link.
http://rankexploits.com/musings/2009/satellite-trends-rss-and-uah/

How could you EVER possibly see evidence of the Earth being in a “long term cooling phase”? You would have to firstly define “long term”. Secondly even if by your definition there was a linear cooling trend over this defined period, you would only be able to “see the evidence” after the period was over and then it would tell you nothing about the next period.

As it happens there is a hypothesis put forward by the alarmists and adopted by the western nations that unstoppably dangerous warming is being caused by anthropogenic CO2. The evidence of the temperature records of the last 7 years, at least, among other evidence, does not seem to support this. But ignoring this and assuming AGW to be gospel we are nonetheless committing to a suicidal battle against this imaginary foe.

I’m not sure I have a clear image in my head of what it is you chaps do with this data.

But looking at your very first graph of the UAH from about 1978 to present; which I presume from the start date is about the sum total of the satellite data set (that they have).

The black line appears to be a straight line. Am I to assume that the rms value of the point by point deviations from that one straight line reaches its absolute minimum value for just that slope and vertical location; the only other constraint being that it is a straight line ?

If that is not the case; then just how is that line defined ?

Any particular reason why a straight line is presumed; since many of these plots seem to show clear indications of some cyclic behavior.

What would the Fourier transform frequency spectrum of that data look like ? I’m wondering if it might reveal the presence of periodic components that are of comparable significance to the zero frequency component.

Well I don’t. Sometimes I know the answer to this. Less frequently to this and that. And hardly ever to that alone. This is one of the frequent times that I don’t have an answer to that.