Saturday, April 14, 2007

We got big trouble in ocean heat content measurement. . . .

Ocean heat content measurements are proving to be especially troublesome. While the new ARGO float system holds the promise of improved data and is almost in place

problems have emerged with a large number of the floats, particularly in the North Atlantic that impose a cold bias on the measurements. This now appears to be understood and can be corrected for. That is the good news.

ARGO is a relatively young system with great promise. Deployment started in 2000 and has almost reached the 3000 float target. Each little bugger bobs up and down between the surface and 2000 m every 10 days measuring temperature, salinity and the flow of currents then phones home with the data

Expendable bathythermographs (XBT) are toss it over the side with a couple of wires that bring the data to the surface torpedo like gizmos that have been used since the 1950s. In Geophysical Research Letters, Gouretski and Kolterman argue that they have a warm bias of 0.2 - 0.4 K, and it may not be possible to correct the data which is a large part of the older measurements of oceanic temperature profiles.

This knocks a number of things into cocked hats. G&K estimate that

Using bias-corrected XBT data we argue reduces the ocean heat content change since the 1950s by a factor of 0.62. Our estimate of the ocean heat content increase (0–3000 m) between 1957–66 and 1987–96 is 12.8·1022 J. Because of imperfect sampling this estimate has an uncertainty of at least 8·1022 J
This leaves studies which relied on the XBT and ARGO data up an interesting river without a propulsion system. In particular, the Lyman, Willis and Johnson 2006 study which described a RecentCooling of the Upper Ocean, has a 2007 submitted correction that describes the effect of both problems. In short, the cooling described in the 2006 study is now seen to be an artifact.

Now, among the Friends of Rabett Run, ClimateScience has been the one most heavily invested in Lyman, Willis and Johnson, using it to argue strongly against the IPCC WG1 Summary for Policy Makers. To his credit Roger Sr., owner operator of Climate Science, noted the coming correction week or so ago. Eli commented
I would recommend caution with something like this. It is going to take a while for the calibration and other kinks in the ARGO float data to be worked out, and in the meantime there is great potential for egg on the face as was the case with the MSU fiasco’s. ARGO has the huge advantage that it was designed for the type of measurements that are being done. Moreover extrapolation with such a short data set is particularly risky given variability.
The response was to attack the surface temperature record, via a 95 pager that apparently has been accepted by J. Geophys. Res. One wearies.


Anonymous said...

I have drawn Roger's attention to this paper, also from GRL:
I am left wondering though, whether the results reported here are also compromised by the findings of the Gouretski paper, or if they in fact can be matched with it.
When I post a reference to this paper on CS, nobody answers.

Roger promotes ocean heat content (measured in Joules) as the best measure of global warming, but the database is so short, and the data as yet unreliable, so it is hard to see how this would advance our understanding of the climate, as things stand.

Anonymous said...

Eli, thanks for this post. There is one other thing which should be emphasized.

We have about a century of land based temperatures, but we know very little about ocean temperature, even though that compries around 75% of the earth surface. Hopefully these new sensors will help clear some of that up, but it's going to take time before we have a reliable record.

Woods Hole took a basic body of one of these sensors and made it mobile. It's a pretty cool system which can cross large expanses of ocean.

Mus musculus anonymouse

Anonymous said...

Fergus, from the timing it would appear that those results would have to be invalidated. OTOH a straightforward fix may be possible with the corrected ARGO data. Would you be willing to just email the lead author and ask, and then post the response? Such folks tend to be gratified that anybody outside a tiny circle of their colleagues cares about their work.

As for the use of ocean heat content as a substitute for surface temps, the several obvious flaws have been pointed out to Roger on many occasions. He seemed to see his responses mainly as an opportunity for enhancing his reputation for obtuseness. But for a little fun, try asking him, "So what's the ocean heat content today, Roger?"

Anonymous said...

Steve; as they are neighbours (more or less) in Southampton, I'll give'em a buzz & see what they say.
Now, stop baiting The good Prof.

Anonymous said...

"Roger promotes ocean heat content (measured in Joules) as the best measure of global warming, but the database is so short, and the data as yet unreliable, so it is hard to see how this would advance our understanding of the climate, as things stand."

Yes, we clearly need ten more years of research on oceans before we can say anything about global warming.

By that time, of course, we will be on to a brand new argument in favor of some other type of data collection, based on the premise that "ocean heat content is meaningless, since as everyone (named John, at least) knows that 'not even the smallest drop of water is ever in local thermodynamic equilibrium and thus all water temperatures are meaningless' ".

Each time climate scientists discover something about the climate, it seems that the current data sets suddenly become suspect to some.

If physics were climate science, instead of building physics from the ground up, scientists would have thrown out data at each stage with the idea that it could not tell them anything useful.

Newton would have declined to develop his 3 laws, since he did not yet have data from String Theory (and alas, never would).

In other words, we'd still be back in the dark ages. (But hey, at least we'd have our strings to strum.)

Anonymous said...

Well, Eli and mice may not be fans, but Roger Pielke Sr.'s article is getting picked up on junkscience.

Who would have ever imagined that?

Anonymous said...

As we all know, for a scientist, the only thing more prestigious than the Nobel Prize is the Milloy Prize.

Hank Roberts said...

Well, as Einstein told us, if you spin fast enough, you drag the frame around with you.

Anonymous said...

He also taught us that "If you are dense enough, you also attract everything to yourself -- including all the kooks."

JunkScience is a singularity. Anything that makes it onto that site never escapes.