As mentioned in Comment 52, I wrote a complaint to the BBC news website regarding their selective reporting of the meeting. I’m not too convinced by the ‘science’ journalist’s comments, but at least he took the effort to reply:

“Peter
It is indeed a huge meeting with very many presentations – oral and
poster; some of which have been peer reviewed and some which have not.
As a guide on deciding what to cover, I prefer to go with material that
has been seen and approved by the widest number of experts in their
field. That is to say, it is not one AGU poster by one researcher, but
material that is representative of a wider input of expertise.
The material to which you refer was presented to journalists by a group
that represents the largest collection of Arctic researchers in the US -
it is called ARCUS. http://www.arcus.org/ Further, it reflects part of
the new NOAA synthesis document “State of the Arctic 2006″
http://www.arctic.noaa.gov/soa2006/
If there was at the AGU an opposing view – built upon an equally large
body of peer-reviewed work put together by a separate but equally
prestigious set of individuals – I’m afraid I missed it.

With thanks for your mail, time and interest.

Jonathan Amos
Assistant Editor, Science and Nature
BBC News Interactive
http://www.bbcnews.com/science”

For the record, here’s my message to the BBC:

“I complain about your selective bias, which directly undermines your policy of impartiality and balance. You led your main news website this morning with a story about potential Arctic ice melting. This was based on one single presentation at a meeting containing 100s of presentations. To simply pick a story with a scary headline is not ‘science’. The AGU meeting contains scientific debate with presentations from across the spectrum of the “global warming’ debate. While many politicians and policy-makers are claiming there is no longer a “debate’, as a scientist myself, I can assure you there still is, and the AGU meeting is an example of that. It is not for the BBC to “choose’ the truth, but to report on the arguments in a balanced manner.”

Mann said that he had “moved on” to Truncated Total Least Squares.

leaving behind rev.#2 of the BàƒÆ’à‚⻲ger and Cubasch verification paper at CPD:

Consider, for example, their use of truncated “Total Least Squares” (TLS). This is not an optimal approach to regularizing TLS. Under the assumption of homogeneous relative errors in the standardized data as discussed above, an optimal regularization of TLS leads directly to ridge regression, and not truncated TLS, which is indeed the reason Schneider(2001) employed ridge regression in the RegEM algorithm.

Figure 7 demonstrates the quality of the range data from the KBR system. The dual frequency (K and Ka band) ranging data from the two satellites are linearly combined to produce a quantity that is free of the time-of-flight or the range changes between the two sensors. Thus this residual quantity is reflective of the noise in tracking system or phase changes that are not desired range change signals; and these residuals are shown to have a standard deviation of a few microns.

Figure 7 reproduced here:

Willis

Are you saying that it would take roughly 120 seconds to see an average realtive change of 1 micron/sec? This would therefore be 120 microns of average change of 1 micron/second over 2 minutes.

RE: #207 – “What is likely to be less reliable is the stretching interpretations made based on near-the-noise-limit measurements from the satellite.”

What does that remind us of?

Seems theoretically possible, although you’re talking about a measurement that is accurate to about one part in 10^12

Most of this ties in with what I would expect. I found a spec sheet on the USO (ultra stable oscillator) and it is quoted as being stable “on the 1 second to 100 second scale” to approx. 10^-13. This ties in with the estimates. Most of the concerns I highlighted (temperature drift etc.) have clearly been considered in the link provided above by BKC.

This doesn’t surprise me; RF comms, radar and ranging is a “hard” science, although parts of RF (especially around the microwave frequencies and higher) is still a bit of a black art, it is quite possible to do real tests and measurements (unlike in climatology), so the analyses tends to be reliable.

What is likely to be less reliable is the stretching interpretations made based on near-the-noise-limit measurements from the satellite.