Bishop Hill Der Spiegel
Apr 1, 2010 Climate: CRU Climate: Surface Blimey. Peter Webster's comments in the Der Spiegel article are quite something, aren't they?
While amateur climatologist McIntyre spent years begging in vain for the raw data, Webster eventually managed to convince Jones to send them to him. He is the only scientist to date who has been given access to the data. "To be honest, I'm shocked by the sloppy documentation," Webster told SPIEGEL.
Unnoticed by the public, Webster has spent several months searching for inconsistencies in the Jones curve. For example, it has been known for some time that there are noticeable jumps in ocean temperature readings. The reason for the inconsistencies is that, beginning in the 1940s, water temperature was no longer measured in buckets filled with seawater, but at the intake valves for the water used to cool ship engines.
But when he analyzed Jones's data, Webster discovered suspiciously similar jumps in temperature -- but on land. "Water buckets can't explain this," says Webster.
Curious Inconsistencies
The Jones team attributes another sudden jump in temperature readings to the decline in air pollution since the 1970s as a result of stricter emissions laws. Particles suspended in the air block solar radiation, so that temperatures rise when the air becomes cleaner. Air pollution in the south has always been much lower than in the north, because, as Webster explains, "there is less land and therefore less industry in the Southern Hemisphere."
Oddly enough, however, the temperature increase in the south is just as strong as it is in the north. "That isn't really possible," says Webster.
Webster doesn't believe that inconsistencies like these will invalidate the Jones curve altogether. "But we would like to know, of course, what's behind all of these phenomena." If a natural mechanism were at least partly to blame for the rise in temperatures, it would decrease the share of human influence in current global warming.
Reader Comments (33)
You posted what I was just going to post. Great minds think alike, and all of that.
This is devestating to the assertion that AGW is based on sound, or any, science.
I think you are going a bit too far there. This suggests that CRUTEM is not based on sound science.
"Oddly enough, however, the temperature increase in the south is just as strong as it is in the north. "
Sorry. But what is the evidence for this assertion?
Ships take in cooling water near their bottom, 25 or 35 feet underwater, where the water is cooler, not warmer, then the water taken from buckets. If there is a icreaswe in temerature, it is because the oceans are warmer, as is land.
Well done again for your watchfulness and good science - I'm very interested in how this will pan out.
Jones is more of a geographer than a scientist. Data is not so sacred outside of science.
I suspect there may well be surprisingly few real scientists at the heart of this shenanigans. Lots of geographers, astronomers, computer programmers, economists, and of course activists such as Maurice Strong (way back when?) and 'Napier of the Met' (still at it, even there), and finally, at least one railway engineer.
the bucket / inlet transition handling errors may result in an underestimation of 1942 ff temperatures by 0.3 deg against 1990 ff temperatures in Jones' dataset.
Including the UHI handling errors, the warmest year since the little ice age could then have been even in the 1940s and AGW would then be dead.
Excellent article! But I take exception to Der Spiegel's characterization of Rajendra Pachauri as "a railroad engineer by trade." Kinda sounds like a steam locomotive driver, doesn't it? But Dr. Pachauri graduated from the prestigious Indian Railways School of Mechanical and Electrical Engineering, has a Masters in Industrial Engineering and PhDs in Industrial Engineering and Economics, the latter from NC State.
I am not a fan of Dr. Pachauri, and only reluctantly rise to his defense. But the climate debate is harsh enough without deliberately denigrating someone with legitimate academic credentials by implying that he drives choo-choos.
The infamous Unit Root is a statistical -- actually linear algebraic measure -- of the consistency of data. In particular, the existence of a Unit Root is indicative of a property in the data call non-stationary. This was a puzzlement to me for a while until I found that it meant a shift in the mean and/or variance of the data over a period of time. That is you are dealing with are different populations. This can happen many ways in what VS calls an observational time serious experiment where you collect data, often historic, which you have no experimental control over the variables involved. I.E the weather.
One way is simply due to the fact there are more than one causal factor involved and they vary independently.
Another is you change how you do the measurements. For example, let's say you are measuring the temperature in degrees Fahrenheit and then for some reason someone changes the thermometer to Celsius but nobody notices. So the 50 degree being reported suddenly become 10 degrees and the day to day variance is now 9/5 of the previous data.
Or you could be measuring different locations, like at the intake and outtake of a marine engine's cooling system.
I think VS's discovery of at Unit Root in the AGW data is indicative of such a problem and throws severe doubt on the validity of all the pretty curves and predictions made from it. In short, the DF and ADF and other measures of the Unit Root and therefore stationary data is a SMELL TEST of the data presented and int this case, it fails.
This is not to say that the AGW crowd aren't right, it merely questions their data severely.
Webster is shining light on this issue with his comments.
Personally, I think we need to junk everything that Phil Jones, Hansen and Mann have done and start again. The existence of global warming or not is a serious issue and should not be left to a group of voodoo scientists with vested interests and little understanding of the scientific method, peer review, or statistics.
@ Josh
New cartoon, The Unit Root doing a Smell Test of the data.
I would note that there are fields outside 'pure' science where data is perhaps even more sacred, fields such as aeronautical and structural engineering where missing or erroneous numbers mean someone dies. I think professionals in 'hard' engineering tend to lean toward skepticism because what's known of the climate data collection and filtering process does not even come close to what they'd consider appropriate.
Further, anyone who's been involved in software development in any engineering, medical, or other field with significant regulatory or quality-assurance requirements will recognize what we've seen out of Jones, Hansen, etc. as prototype-quality work unsuited for anything more stringent.
A little help for the hard of thinking please.
Regarding buckets and intakes Webster says "But when he analyzed Jones's data, Webster discovered suspiciously similar jumps in temperature -- but on land. "Water buckets can't explain this," says Webster."
If you strip out what Webster says are suspicious increases on land, the global temp reconstruction from the 40s onwards should be... lower?
Then Webster says for particulates "Oddly enough, however, the temperature increase in the south is just as strong as it is in the north. "That isn't really possible," says Webster."
If you strip out the SH warming Webster says is suspicious the global temp reconstruction from the 70s onwards should be... lower?
Looking at the graphic with the article titled A Heated Dispute is it at all suspicious to anyone else that the two periods Webster is questioning happen to bookend a relative flat line in temps? At a rough guess correcting the reconstruction along Webster's lines would reshape it to something more like a 50-60 year peak to peak pattern with the 40s and late 90s being about equal in temperature.
If Webster's logic is sound, for the Jones reconstruction to be credible would Jones need to show that the change from buckets to intakes *is not* a factor in a change in measured temperature and that a change in particulate levels is also not a factor in measured temperature?
Don Pablo & Gareth
I think that I may have found a similar Unit Root problem in the Australian data.
I also think I may have found what caused it, which is more interesting,
because it would also be true of the global data.
I am working on this slowly, and will broadcast my findings,
if and when I am sure enough that I am right.
It passes the "aha" smell test.
Now for the slow, laborous and boring part of testing and documentation.
My first post on B.H.
As I have worked in the marine field as an electrical engineer for many years now, the measurement of a boat/ship cooling water intake caught my eye.
I would only ask a few questions that other may be able to answer for me.
A temperature measurement would obviously use the ships log for location but
1. Is the ship in port or at high sea?
2. Is the ship loaded or unloaded? (This would obviously change the depth of the sea-chest intake).
3. What type of instrumentation is used and how often is calibration checked/tested
4. Who are the people taking and notating the readings?
5. Are they using Scientific Vessels/Armed Forces vessels or Trading Vessels for the information?
My experience would show that if they use maritime trading vessels data then it would be almost worthless. Most engineering crews at 3rd engineer down (who would collect this data) would be to inept/lazy to step out of the nice air conditioned engine control room! We have already seen that in some Arctic stations, people responsible would not go out to collect the data due to the cold/polar bears/getting lost. Just makes me wonder!
The article is spoiled by this piece of speculation presented as fact:
Unfortunately for Jones, however, McIntyre's supporters eventually included people who know how to secretly hack into computers and steal data.
I was a tired junior engineer taking those temperatures at 3:30 in the morning. The thermometeters were alcohol ones with 5 divisions per 10 degrees Fahrenheit set at an awkward angle with a large parallax error. The thermometer pocket might or might not be oil filled.
As mentioned, the main circulating pump inlet was several metres underwater and on a tanker there were large differences between loaded and light conditions. Some of the ships were registered as 'weather' ships.
We didn't have any air conditioned control rooms and worked in temperatures of 40 to 45 deg C. It was concidered a matter of pride that the 50/60 log readings were memorized and not noted down.
It is a source of amusment to me when I read of anyone taking these reading seriously.
Geckko, your point is well made. Though I don't think, broadly, it's an incorrect assertion (I'm sure it's true that there are climate-sceptical individuals who know far more about such things than they should) I do agree that the inference is perhaps not entirely appropriate. Hackers as heroes is not a universal perception at all.
@SimonH
Journalism shouldn't be a discipline of innuendo. It is a nice smear to suggest that Steve was in any way connected with events that led to the release of the emails etc..
Just as bad in journalistic terms is the repetition of the unsubstantiated and unjustifiable claim that anything was stolen or that CRU computers were "secretly hacked".
It just seems to be an incongruous accusation in an otherwise apparently well researched and well judged piece of journalism.
Another way of putting it with regard to the AGW data so far is "Garbage in, Garage out." If there is a unit root, then something smells bad.
Time to redo it carefully and honestly, as several leading scientists have said.
I commented before about Dr. Judith Curry. I was not aware until the last couple of days that Peter Webster was also at Georgia Tech. I am sure that it is Dr. Peter Webster or soon will be. I am so proud that my school has these two fine scientists. The world would do well to listen to them!
I doubt we would have heard of the bucket/inlet adjustment if there had not been a need to explain away the drop in temperatures in the mid 20th Century.
http://www.independent.co.uk/environment/climate-change/case-against-climate-change-discredited-by-study-835856.html
To summarise:
Phil Jone says inlet water is warmer and adjusts accordingly.
bobmagginis holds that inlet water is cooler.
Chris M (who should know) says we can't trust the measurements.
I think a simple comparative experiment is called for. Where do we get the grant for a cruise?
Pete H and Chris M - the value of first hand experience! Thank you for contributing - those were the type of questions that sprang to mind when I saw bobmaginnis' comment. Global temp. measures are based on many sources which were never established with the current end use in mind.
chris m
source of amusement..
but did you wear a yellow jacket while doing these measurements? If not, I am afraid we'll have to "average" here.
Gekko is quite right about the veiled accusation. As one who has been a programmer for forty years, I do not believe the "hacking" theory of Climategate at all. The most probable explanation is that some clerk spent a lot of time collecting emails (from the administrative backup) and documents to answer a FOIA request; then word came down that the request was denied. The stuff was sitting there, and someone at CRU who was fed up with the lies (perhaps even the same clerk) finally sent it out.
Any other explanation requires too many improbable assumptions.
Anyone reading this who is not convince of the utter scumbaggery undertaken by Jones, Mann, et. al. needs to read John P. Costella's excellent dissertation on the ClimateGate emails - http://assassinationscience.com/climategate/ - and please remember this one salient point...the emails were LEAKED, not HACKED.
"Further, anyone who's been involved in software development in any engineering, medical, or other field with significant regulatory or quality-assurance requirements will recognize what we've seen out of Jones, Hansen, etc. as prototype-quality work unsuited for anything more stringent.
April 2, 2010 | Unregistered CommenterJEM "
I've said this before, but no-one would go up in a plane developed like this, or give drugs developed like this. Stock merket analysts would be in prison with data developed loke this.
Re buckets and thermometres. This is one of the many reasons that paleo-climatology is not a science, it's speculation.
Jack Maloney writes:
"But Dr. Pachauri graduated from the prestigious Indian Railways School of Mechanical and Electrical Engineering, has a Masters in Industrial Engineering and PhDs in Industrial Engineering and Economics, the latter from NC State."
Your defense of Pachauri does him no good service. Industrial Engineering is the major taken by all not quite yet professional athletes at Georgia Tech. As they say, those who can't do engineering do industrial engineering.
For Stan Stendera,
Dear Stan,
Thank you for your kind words. Alas, I graduated sometime ago and hold a joint appointment in Earth and Atmospheric Sciences and Environmental Engineering at Tech. I guess that if we do anything well at Tech it is supporting good ideas and good science and being vocal about issues we feel strongly about. Both Judith and I try to live up to that ethic.
There are simple things we try and do with our research and it was a surprise to find out this simple rule is not followed by all scientists. Both of us have active graduate students engaged in intricate work. Each of them has a lab book and every calculation, model run and etc.. is detailed so that the student, supervisor or interested third parties can replicate results. When something is published, code, data, meta-data and etc are available. What is the point otherwise? Clearly, at some institutions these transparent procedures have not been followed.
Regards
PW
Pete H/Chris M
I was a product quality control engineer in a continuous process manufacturing plant. One of the greatest aggravations was the number of what our operators called 'bulls**t tests' that showed up in the data logs. Operators too busy/lazy to actually leave the comfort of their control room to collect an actual sample would simply make up a number slightly different than the one before and enter it as a bona fide test result. Of course they had no way of knowing whether the previous result was of the BS variety and everybody trusted that everyone else was honest so that they themselves could fudge. Processes would often wander off course and several tonnes of off-grade product would be made because of these made up results. Our plant finally invested in millions of dollars of process control equipment and downsized the non-performing workforce. The quality of the finished product went up, customer complaints went down along with a decrease in cost to produce. When I see data these days I always question whether it was generated by people or instruments. The more often people have touched it, the more likely it will be inaccurate - either through filtering biases, incompetence or laziness. Cynical, I know, but I've watched how people behave when there are no checks and balances.
@jack maloney
The term 'railroad engineer' doesn't have the same resonance on this side of the pond. For us, a railroad (or more likely railway) engineer is one who engineers the railway. So the great IK Brunel or the magnificent Stephenson father and son would be proud to be called railway engineers. We call the guys who makes the train move an 'engine driver'
So der Spiegel was just following common English usage without particular disrespect meant to Pacahuri by this term
But nonetheless, he has no qualifications in anything remotely connected to 'Climate Science'
top wrecking yards and auto recyclers offer premium used engines for sale with 1-year warranty and 90-day lowest price guarantee. Inventory changes daily and includes American, import and Japanese engines, gasoline and diesel engines.
Thanks
Marine Engines
Marine Engines says:
“Interesting post, we shall be following your blog more closely in future! Best Wishes ”
re: "deliberately denigrating someone with legitimate academic credentials by implying that he drives choo-choos"
"deliberately"?
As a native speaker of the American brand of English, I agree that some Americans could unthinkingly read it as choo-choo driver. And if indeed that was deliberate, I'd deplore it --- tacky deniable dishonest weaselry, ugh. But besides the "this side of the pond" point made by another commenter, note that the article was translated to English from a (presumably German) original. To me this looks like the kind of slipped connotation that happens routinely in translation and in transatlantic communication despite people's efforts to prevent it.
Conversely, "deliberately" seems farfetched to me if I try to think about the details of who intended it, and what the motive was. A lone rogue translator choosing to damage his reputation by making an infelicitous translation which could be read as a childish slur? A clever bilingual/bidialectal authorial/editorial/translator conspiracy crafting a neutral-in-German passage designed to be translated transatlantically into a slur which'd deniably mislead? And in either case, motivated by a desire to mislead that enormous audience of Americans who both (a) read Der Spiegel in translation and (b) don't already know enough (enough about Pachauri from other AGW news articles, and/or enough about international/historical ambiguities in the meaning of "engineer" from other eclectic reading) to read right past the ambiguity without noticing, as I did when I first read the article?
i like to read the posts of this blog is very good in providing helpful information to everyone.this is great to make us happy by the information.
thanks