The implication seems to be that 100mm error can be vastly reduced by taking 60,000 readings. My personal feeling is that if the 100mm is all a bias in accuracy, no number of readings will correct for it. So do we quote change of 3mm a year globally for a device accurate to 100 mm? Try building a home with a ruler graduated only in half yards, or as some of us express in a more enlightened decimal way, about 50 cm.

You can go beyond your personal feeling, Geoff. What you state is a fact. Increased sample size only reduces sampling (i.e. random) error. Increased sample size does not reduce systematic error (i.e. bias).

Thank you, oh thank you, for picking up the example of my lousy typing. I now know that at least one person read the post.

On the more substantial issue, can anyone add extra information about the absolute accuracy of satellites used for the determination of sea level? Is it not delusionary to model figures which spend their lives in and out of error bounds? Besides, how are satellite altitudes measured and corrected? By bouncing signals off land and sea. The land/sea ensemble is almost circular, at least for one orbit.

I’d dispute their assessment of the relative ratings of the two journals.

A few stunning pieces of wisdom cited from Roger Pielke Jr. with regard to Vecchi and Soden:

Roger A. Pielke Jr., a professor of environmental studies at the University of Colorado, recently noted the very different reception received last year by two conflicting papers on the link between hurricanes and global warming. He counted 79 news articles about a paper in the Philosophical Transactions of the Royal Society, and only 3 news articles about one in a far more prestigious journal, Nature.

Guess which paper jibed with the theory — and image of Katrina — presented by Al Gore’s “Inconvenient Truth”?

It was, of course, the paper in the more obscure journal, which suggested that global warming is creating more hurricanes. The paper in Nature concluded that global warming has a minimal effect on hurricanes. It was published in December — by coincidence, the same week that Mr. Gore received his Nobel Peace Prize.

And who can forgot the historically quiet hurricane season in the Northern Hemisphere: (and 2007 is over, by the way)

When Hurricane Katrina flooded New Orleans in 2005, it was supposed to be a harbinger of the stormier world predicted by some climate modelers. When the next two hurricane seasons were fairly calm — by some measures, last season in the Northern Hemisphere was the calmest in three decades — the availability entrepreneurs changed the subject. Droughts in California and Australia became the new harbingers of climate change (never mind that a warmer planet is projected to have more, not less, precipitation over all).

The author of this piece may need to go into the Witness Protection program for straying so far away from his comrades at the Times. I am stunned and at the same time, impressed with the breadth of the article. Is this a new year or what?

When I return to my home computer I’ll check ACE per storm versus AMM.

My sense is that the AMM-related phenomena affect storm duration and intensity while storm count is driven by something else.

David, did you restrict your storms to those with an ACE greater than 2 for your graph and correlations?

Back to thread. A CSIRO climatologist emailed me about satellite measurements for seal level changes, email early in 2007.

Would seal level changes be used as a proxy for temperatures below sea level? I thought seals had a wide range of swimming depths.

The purist scientist in me objects to this quote -

Steve Sadlov: Come on, you know there’s no such thing as proof when it comes to science!

Would it not be more charitable to say that “scientific truth is not absolute for a subject, but it is the aim of purer scientists to explain and account for the maximum amount of known uncertainty”.

BTW, in your # 53 the spelling is “gases” not “gasses”, with fair certainty, source – many reputable English dictionaries. One letter can make a difference. Inulin is not insulin.

Mea culpa too, I am a lousy typist.

Back to thread. A CSIRO climatologist emailed me about satellite measurements for seal level changes, email early in 2007. I’ll just quote him verbatim, it’s easy to get what I mean.

keeping satellites on track – there I meant in the horizontal
plane, ie steering the satellite to overfly the same ground tracks. This
only needs to be within a few km, since the footprint of the sounding is
many km anyway.

Noise floor: 100mm is the noise level for 1Hz observations
(which are averages of the original 20Hz observations).Each ‘snapshot’
of global sea level takes 10 days to sample. That’s 864000*(sea fraction
of planet)= about 600,000. This does not reduce the noise as much as you
might hope because some of the sources of error have large covariance
scales.

The implication seems to be that 100mm error can be vastly reduced by taking 60,000 readings. My personal feeling is that if the 100mm is all a bias in accuracy, no number of readings will correct for it. So do we quote change of 3mm a year globally for a device accurate to 100 mm? Try building a home with a ruler graduated only in half yards, or as some of us express in a more enlightened decimal way, about 50 cm.

Nothing profround. It confirms that AMM correlates a little better with hurricane activity than does SST.

This one shows Atlantic SST variation (Gulf of Mexico, Western Tropical and Eastern Tropical Atlantic) plus an AMM index (related to the north-south SST gradient near the equator). What impresses me is the oscillatory nature of all of these parameters, with roughly a decade from peak to peak. I have not found an explanation for this.

The other point is that the pattern becomes more distint in the 1970s – I wonder if this is tied to improved SST measurement as satellite data began to supplement ship data in the late 1970s.

Something to ponder.