Better late than never …
Any topic or issue reasonably related to climate change is fine, but be forewarned this is not a forum for debating the basics.
Thanks!
Recent comments
Exploring climate science, and climate science misinformation, in Canada and beyond
Deep Climate 
Looks like Spencer/Christy have created a new version of the satellite data, based on a correction to the seasonal cycle you alerted them to.
http://www.drroyspencer.com/2010/03/february-2010-uah-global-temperature-update-version-5-3-unveiled/
I noticed Christy failed to acknowledge your site for this correction, but instead acknowledged Anthony Watts for bringing it to their attention, as if he’s’ the one who discovered the error.
It would be interesting, DC, if you could do an analysis on the seasonal cycle (if it still exists) for v5.3.
[DC: I haven’t seen the data yet, except for 2009-10 as shown by Spencer. The adjustment there appears to reduce the annual cycle considerably (but I don’t think it eliminates it altogether).
It’s not typical to acknowledge anonymous bloggers, and anyway I wasn’t the first to write about the annual cycle, although I may have been the first to demonstrate the resulting wide divergence in trends when calculated month by month. Watts certainly did advise Christy of the problem, but I don’t know if he was the first to do so. Certainly he had nothing to do with discovering or documenting the issue, and I hope no one is crediting him with that.
So it may be interesting to review the history of this adjustment, as well as analyze the actual changes. There’s some evidence to believe that Spencer, at least, was aware of the annual cycle issue before I ever wrote about it. ]
Has anyone else seen this before? I didn’t see an open thread so thought here’s as good as any.
[DC: I’ve reposted it at the new Open Thread as you can see. Thanks for the reminder …]
Guidelines for Responsible Data Management in Scientific Research Funded by: Office of Research Integrity US Department of Health and Human Services
How Long Should Data Be Kept?
There is no set amount of time for which data should be stored. In some cases, the time period is at the discretion of the PIs; however, many sponsor institutions require that data be retained for a minimum number of years after the last expenditure report. For instance, the USDHHS requires that project data be retained for at least 3 years after the funding
Click to access data.pdf
All federally funded universities must have a data policy that addresses this issue. In some cases, as you note, the time data must be kept is surprisingly short.
If March turns out to be hot as well (so far it is), will their headline be that they have switched to version 5.3b?
@ Rob.
Thanks. I did find other policies that say ten years, but they’re current policies and, just like a lot of Nat Met Services have changed their license agreements on use of data (especially in the last few months), it’s hard to tell what was the case, say, twenty years ago. There seems to be this automatic assumption in the world that all data is supposed to be kept forever, which seems daft.
It’s interesting to see how well the different global temperature series match each other – the ‘noise’ of interannual variability is very similar in all of them. If satellite and ground measurements show the same fluctuations then presumably they are accurately reflecting real changes in the climate system rather than just imperfections in our measurements, but to me it seems hard to believe that the temperature of the entire planet can fluctuate by several tenths of a degree in just a year or two. I would have thought that the enormous mass of the oceans, land and atmosphere would have more ‘thermal inertia’ (hope that’s the right phrase) than that.
So, I wonder whether the reality is a bit more subtle, and that perhaps our measurements are too sensitive to heat being shuttled around between the various parts of the climate system? Could it be that the difference between total incoming radiation from the sun and total outgoing radiation to space would show a lot less year-to-year variation (‘noise’) than our figures for global mean temperature, if only we could measure it accurately?
I’m thinking perhaps of heat being sequestered in the deep oceans where we can’t keep track of it very well, or of energy contributing to the melting of ice where it would otherwise be causing warming, neither of which would (presumably) affect the overall temperature or energy content of the climate system.
If we could have a measure of total energy of the entire climate system rather than mean global temperature, would that be a less ‘noisy’ curve? Does anyone attempt to do this? Can it be done from satellite measurements of incoming vs. outgoing radiation, for example?
Cheers…
[DC: There is, as you imply, much more heat in the oceans than in the atmosphere. The way I understand it, changes in ocean surface temperature related to multi-year ocean current oscillations (e.g. ENSO) can show up as large interannual surface air temperature variations globally (even more pronounced regionally). I believe there have been attempts to measure ocean heat content, but I will defer to more knowledgable readers to explicate further and/or point us in the right direction. ]
WOW – “this is not a forum for debating the basics” – now that’s classic….
Levitus 2009, Global ocean heat content 1955-2008 …
I think this is the latest paper …
Looking at Conservative Graphics’s site, I’m sure that by this:
He means we should be debated whether or not “3” is “pi”, etc.
The next question is, of course, “basics” to you but advanced stuff to me so I allow myself the opportunity to ask it 🙂
What exactly is causing the annual absolute temperature cycle that can be seen in the UAH daily (absolute) temperatures (ch05 20-year average about -21.3 Jan => -19,5 July)?
Some people were previously mistaking anomalies for absolute temperature and suggested that earth is the warmest around January due to its orbit (=closest to the sun), but clearly the UAH absolute temperatures show another story?
So do UAH dailies show a real yearly cycle or is it some cyclic error that is adjusted for when calculating the real numbers? And what is actually the annual absolute temperature cycle of earth and why?
Best regards to you all.
[DC: Perhaps I should amend the above. For now, I’ll clarify that there’s a difference between “debating the basics”, i.e. putting forth previously debunked contrarian debating points, and asking reasonable questions. I have no problem with the latter of course. Although I certainly don’t have all the answers (I’m just a lowly blogger), someone will or point to a good reference.
I’d like to avoid false and pointless debates, but reasonable discussion is welcome.
You’re right that it’s important to distinguish between absolute temperatures and anomalies (i.e. departure from a baseline average).
There is an annual cycle in absolute temperature, simply because the northern hemisphere has more land and tends to be warmer in absolute trems than average than the sourthern hemisphere. So overall, July is warmer than January. This is reflected in the raw, “brightness” temperature you see in channel 05.
On the other hand, temperature data sets are expressed in anomalies, where the anomaly for a given month is calculated in comparison to an average for that same month during the baseline period (1979-1998 for UAH). In general temperature data sets show more a little higher *warming* trend in northern hemisphere winter, since the fastest rate of warming is happening in winter over land in extra-tropical regions.
So we have, say, January warming faster overall, but still have July warmer in absolute terms.
To add to the confusion, there can be spurious annual cycles in the anomalies, such as occurred in UAH (partially fixed now). For more, see the post just after this one. ]
@J Bowers
Our policy states five (5) years on data, twenty (20) years for data that has resulted in a patent, and seven (7) years for data that has led to charges of research misconduct. For students, the data must be kept until the student graduates.
This policy follows federal guidelines/laws.
I can’t recall Phil Jones applying for any patents. Many thanks, Rob. I think politicos and lobbyists need to warn scientists up to ten years in advance of any likelihood that they may cast an “Eye of Sauron” over their particular establishment, just in case they need to be implicated for any future manufactroversies. They could even make a model and offer the code to scientists so they can judge how long they should keep data for; a ‘Moranometer’.
Apparently it’s something to do with differences in how the newer AMSU sensors work vs. the older MSU ones.
They remove the average seasonal cycle, according to John Christy. He doesn’t say how in his brief explanation but I’m sure it’s in the published paper trail of their work.
Now they were using the average seasonal cycle data they’d constructed for the old MSU sensors to adjust the AMSU data. It appears that differences in the way the sensors work, then, caused that seasonal adjustment to work poorly for the AMSU data.
Now, they say, they’ve “limited the influence”, presumably by calculating a new average seasonal cycle, possibly using AMSU data, but I’m just guessing at this point.
They’ll probably put out something more detailed and technical at some point.
[DC: There is a longer explanation which is not in the public domain as far as I know, but can be obtained by emailing John Christy, if I recall correctly.
Anyway, from reading all the publicly available commentary, it seems that the issue is intersatellite calibration. Each satellite sensor has its own annual cycle in its response which is presumably evident as a difference in response over the time of overlap between two units. Over time the adjustment calibration from previous to succeeding satellite has introduced an ever-widening and spurious annual cycle.
So Christy and Spencer “start over” with the AMSUs. To me the need for this ad hoc adjustment points out the difficulties inherent in the satellite record in general. I’m also not sure why this annual cycle issue seems more problematic in UAH than RSS, even after the latest adjustment. ]
Ocean heat content … is indeed where the action is – I always teach defensive football players in high school:
ignore the head-fakes (air temperature jiggles), watch the runner’s belt buckle (OHC).
Because: Conservation of Energy is The Law… most of the variable energy is in the ocean, and its oscillations drag the atmosphere around a lot, but the general trend in the ocean displays Earth’s energy imbalance fairly directly, and with less lag than one sees at the surface.
2009:
ftp://ftp.nodc.noaa.gov/pub/data.nodc/woa/PUBLICATIONS/grlheat08.pdf
2008:
http://www.realclimate.org/index.php/archives/2008/06/ocean-heat-content-revisions
2007:
IPCC AR4, p47-48, Fig TS.15 shows the changes in total energy content for 1961-2003 (42 years), and 1991-2003 (12 years).
Bad news, about half the change from 1961-2003 happened from 1991-2003.
The main issue is that we don’t have centuries of modern data, and just like satellites, there have been measurement kinks to work out. Science, as usual.
At Tamino, commenter Brian D said the following, concerning a supposed statement by Andrew Weaver to CanWest reporter Richard Foot, that IPCC chairman Pachauri should resign:
“On a supplemental note, the Edmonton Journal’s print version DID run Dr. Weaver’s correction, but buried it deep in the Technology section, instead of in the front News section where Foot’s story ran.”
I haven’t seen that correction – did it run online (I can’t find it)? If not, does anyone out there have a date and page number for the print version?
(I’ve asked Brian D the same questions at his blog – perhaps he’ll respond here).
Sorry about the delay. My blog’s hosted on a friend’s home server, literally MacGyvered together as an experiment, and one of the courtesies I promised him as a condition of the free hosting was to keep comment moderation on to prevent spamming. (Let’s just say the traffic surge in the wake of my Superfreakonomics criticism led to legitimate concerns for the server’s longevity.) Not that I think I’d get much spam anyway – I’m not even a bit player out here, nor am I terribly prolific – but I only check the queue every few days.
Yes, Holly Stick is linking to the right article, which I found courtesy of BigCityLib before he blogged it, through a mailing list (which I think you’ve been contacted about, though I understand if you’d rather maintain total anonymity. You do incredible work as it is – when I met Weaver himself in December, he mentioned you as soon as I brought up climate blogging – and I wouldn’t want to compromise it). As such, I’m not aware of it being run in the print Journal.
However, my father’s got access to the online scans of the print run; I should ask him to dig through the Jan31 issue to see how little coverage it got compared to the original allegation. (I’m surprised he didn’t inform me about this earlier; he reads the print Journal regularly and is in correspondence with some of their columnists, both in-house and through CanWest – Richard Foot included. This suggests how deeply buried it was.)
Here is the correction I think:
http://www.edmontonjournal.com/technology/Prof+clarifies+position+climate+panel/2504952/story.html
[DC: Thanks, that’s the one. ]
I got into an argument over here in the comments:
http://ken-chapman.blogspot.com/2010/02/how-dare-jean-charest-suck-and-blow-on.html
and found a couple more links:
this interview:
http://www.cbc.ca/thecurrent/2010/201002/20100210.html
and this blogpost:
http://bigcitylib.blogspot.com/2010/02/national-post-one-editorial-two.html
[DC: Also see this analysis by Tyler Hamilton of the Toronto Star, where Andrew Weaver sets the record straight about a number of things.
http://www.thestar.com/business/cleanbreak/article/765519–hamilton-spin-is-in-but-climate-change-still-there
]
From Science News:
Magnetic flows cause sunspot lows, study shows
Newly reported observations of gas flows on the solar surface may explain why the sun recently had such an extended case of the doldrums.
http://www.sciencenews.org/view/generic/id/57180/title/Magnetic_flows_cause_sunspot_lows%2C_study_shows
Regarding data retention:
Jones made himself responsible for the temperature of the world (1 of 4 anyway). Personally, I find global average temperature a useless parameter, but some tout it as reason to restructure the worlds economies, which some other people think is a really, really big deal.
He’s morally obligated to have the data. All of it. And public access to his code/corrections/etc.
I’m a skeptic, but I agree some of the climate-gate revelations are tainted by spin. OTOH, I don’t see any wiggle room for data, data treatment methods, and public access to same – not when the stakes are this high.
[DC: Even the data that is controlled by others and may not be forwarded without permission? Even the data that was forwarded to CRU many years ago on paper?
As far as methods go, those are published in scientific papers. The self-appointed auditors seem to have problems understanding those papers. ]
@jbowers
Re Jones’ data:
Please can you guys get your story straight? I’m beginning to run out of rows on my spreadsheet to keep track of the number of different reasons given why Phil can’t come up with the goodies…
Badly organised office…
You’ll only try to find something wrong with it (you bet your sweet ass matey)….
I have agreements that I can’t find with countries I can’t remember that prevent me from doing so…please don’t ask me which countries they are
…and now that it would be 60 m tall and can’t justified. Perhaps Phil and his crew never even bothered to put these raw numbers into a computer before ‘adjusting’ them? Or they threw away this dataset as being unnecessary?
Just to remind any readers who haven’t followed this in detail. The UK Information Commissioner’s Office (statutory body administering the Freedom of Information Act stated publicly (re a FOI request to Jones)
‘The emails which are now public reveal that Mr Holland’s requests under the Freedom of Information Act were not dealt with as they should have been under the legislation. Section 77 of the Freedom of Information Act makes it an offence for public authorities to act so as to prevent intentionally the disclosure of requested information. Mr Holland’s FOI requests were submitted in 2007/8, but it has only recently come to light that they were not dealt with in accordance with the Act. The legislation requires action within six months of the offence taking place, so by the time the action taken came to light the opportunity to consider a prosecution was long gone’
which is about as damning as it can be.
We can all guess what really happened. Phil threw away the data (either accidentally or on purpose), so has nothing at all to back up his assertions, thought he could get away with it by obfuscation, and has now discovered that he can’t. Hence the three months off work and the sadly aged appearance. Even Mann is beginning to disown him…Curry already has.. his future prospects look grim indeed.
But no excuse to break the law and fail to provide the openness and transparency his publicly funded research and salary required. Greek tragedy..hubris followed by nemesis……
“We can all guess what really happened. Phil threw away the data (either accidentally or on purpose), so has nothing at all to back up his assertions, thought he could get away with it by obfuscation, and has now discovered that he can’t.”
He didn’t need to keep the data. The work had been done, the raw data was kept for a few years after which time it was fine for him to shred it.
Read page 27 of Guidelines for Responsible Data Management in Scientific Research (Office of Research Integrity, US Department of Health and Human Services)
Click to access data.pdf
Acton stated clearly when giving evidence at the Parliamentary hearing that the data need only be kept for between 3 to 5 years.
If nobdy was bothered enough to debunk it in the peer reviewed literature before then, then tough. They should get out of bed earlier in the morning.
“The UK Information Commissioner’s Office (statutory body administering the Freedom of Information Act stated publicly (re a FOI request to Jones)”
So what?
Show me the transcript, or even a formal piece of paper, which are the results of a formal investigation. Don’t throw at us a quote from the equivalent of a copper who thinks so-and-so is a villain without the CPS looking at the evidence, taking it to court, and a judge passing sentence after a trial.
I have agreements that I can’t find with countries I can’t remember that prevent me from doing so…please don’t ask me which countries they are”
Here’s a reminder of a few…
Click to access agreements.pdf
http://data.smhi.se/met/climate/time_series/html/essential20.html
Wrong, wrong wrong.
David Holland didn’t ask for data, he asked for email correspondence related to Ch. 6 of WG1 of the AR4 report. Confidential and legitimately off-limits, according to everybody involved (Jones, Briffa, Ammann, Wahl, Santer, Wigley, several of them feeling pretty strongly about it — check it out) and according to FOI legal practice on both sides of the pond.
Holland would never have gotten his sticky fingers on these mails. Not legally.
Section 77:
Note:
… the information to the communication of which the applicant would have been entitled.
Ex-information commissioner Thomas also pointed this out in the oral session. He should know, don’t you think?
Even if everybody had acted on Jones’ email and deleted all these mails, they would have been in their full legal right to do so.
Making a fool of yourself by letting prejudice override evidence is comedy of sorts — not amusing though.
SE, have you even looked at what is available from GISS and Mann’s SI? What’s missing (be specific)?
To help with your answer, here’s the SI from the 2008 PNAS paper: http://www.meteo.psu.edu/~mann/supplements/MultiproxyMeans07/
So who is the “competent auditor” you write about?
[DC: As I’ve commented before, Mann et al 2008 PNAS provided a wealth information, including very clear methodological description *and* Matlab code. The self-appointed auditor didn’t bother with properly absorbing that information, and made several basic errors in his analysis and replication of Mann’s work.
But that’s not a new problem. The same thing happened from the very beginning: M&M 2003 hard coded the number of PCs to retain in the North American tree-ring network, in direct contradiction of Mann’s explicit methodology of retaining enough PCs to explain about half the variance.
That’s probably enough on SE’s talking points, though. He doesn’t seem to want to discuss the actual issue of this post, so I’m moving his latest comment – and this one – to Unthreaded. This discussion should be continued there, if anyone would like to. ]
If it were my blog, I’d start up a /dev/null thread for SE’s posts …
[DC: You may have a point. We’ll see how it goes. ]
Deep Climate wrote:
I will resist troll feeding, but I do have to say that SE’s response illustrates the point that transparency does not lead to credibility among deniers. Another goalpost on wheels.
“Today’s great gurus (Jones, Mann, Hansen) and their acolytes day….”
You’re including Hansen in this list of discredited ones now? Hansen made the GISS temp data and code available over two years ago and you see fit to declare he’s got no credibility? Whatever mistakes Mann made with MDH98 in releasing code he rectified the situation in his PNAS 2009 paper. By your own standards for being open and encouraging openness these two individuals are doing what’s being asked of them, yet you declare they’ve lost their credibility. I think you’re being highly disingenuous. I also think you’re highly prone to making generalized and imprecise statements.
——
Posted from other thread.
SE,
Is the statement of intent to delete emails the same thing as evidence that deletion actually took place? Is there anything more than prima facie evidence here? There isn’t anything in the stolen emails to prove that the deletion took place. Should the information commissioner be making such bold statements in the absence of an official enquiry? Last time I checked it was innocent until proven guilty.
Just out of interest, do you think that any errors of a significant nature are going to emerge from the released code for the compilation of HADCRUT?
http://www.metoffice.gov.uk/climatechange/science/monitoring/subsets.html
And, just out of interest, any idea where the papers are indicating what the catastrophic errors are in GISStemp? We’ve been waiting over 2 years.
And, just out of interest, what are thoughts on this from a philosophy of science standpoint? How likely is it, without invoking conspiracy theories, that all the available analyses of global average temperature (UK, US, Russian, Japan, and satellite (microwave and IR)) are all wrong in precisely the same way? You sir, are the obfuscater, the one making incredible claims without evidence, and the one wasting everyones time.
Interestingly, the NCDC has 120,000 boxes of paper obs forms (truly), satellite photos, radar data, fiche and so on, all in the process of being digitized.
I wonder why there isn’t a line of denialists waiting to get at that really really raw data. I’d like to see the folks clamoring for it show up and do some volunteer work.
Probably too busy complaining about high taxes, big government, and wasteful government spending …