On July 17, 2007, Rob Wilson archived British Columbia measurement data obtained in 1997-1998 and used in Wilson and Luckman 2002, 2003 (20 ring width; 7 mxd series). The data is here: ftp://ftp.ncdc.noaa.gov/pub/data/paleo/treering/updates. Wilson et al (JGR 2007), a new article, reports on a new tree ring reconstruction from 1750 on, which Rob claims to mitigate the divergence problem. This includes a very slight SI ftp://ftp.agu.org/apend/jd/2006jd008318 , which contains a new version of the BC reconstruction from Wilson and Luckman 2002, 2003, discussed on an earlier occasion here.
On an earlier occasion, Rob made it clear that any decision to archive would owe nothing to climateaudit (a position expressed with a tinge of spite rather uncharacteristic of Rob.) Nonetheless, this is the first measurement data archived by Rob and I appreciate that this has been done. It is too bad that Rob’s friend, Jan Esper, continues to wilfully withhold measurement data used in various publications.
From the recently archived measurement data, I’ve calculated Jacoby-type and RCS-type site chronologies for the 20 RW datasets and the 7 MXD datasets and compared them to the version used in the Wilson et al 2007 divergence article. As you will see, if you take simple averages of chronologies calculated from the archived data, there is a marked ring width divergence problem with values reaching a maximum around 1940 (as with the Schweingruber data considered by Briffa.) The MXD data shows little trend whatever.
Wilson et al 2007 purport to analyze the divergence problem as follows:
We attempt to address the divergence problem’ noted in large-scale reconstructions by developing a new independent TR based reconstruction of ENH temperatures using published local/regional reconstructions and newly updated/sampled TR chronologies that show no divergence against local temperatures …
I’ve done two separate chronology calculations for the B.C. network of 20 RW and 7 MXD datasets. Figure 1 shows a “Jacoby-type” chronology, calculated by attempting to fit a generalized negative exponential (RW), negative line (MXD) or mean to individual cores. There is a marked divergence problem in the RW average, while the MXD average shows little trend. The series carried forward in Wlson et al 2007, in contrast, has a marked 20th century trend.
Figure 1. Averages calculated from Jacoby-type chronologies
Next, here are corresponding averages from an RCS-type chronology calculation. When I tried to fit negative exponentials to the data, this failed in nearly all cases. I presume that Wilson got similar results. I accordingly fitted the RCS calculation with a spline and used this for the calculation below. These chronologies show more pronounced centennial variability, but the RW average still has a major divergence problem (with a maximum occurring in the 1930s) and there is negligible trend in the MXD data. Neither shows the 20th century trend used in Wilson et al 2007.
Figure 2. From RCS-type calculation using spline fits.
For comparison, here is the illustration from Wilson and Luckman 2003:
I presume that the difference between the Wilson et al 2007 IBC series and the RW and MXD averages comes from the principal components and inverse regression of temperature against several orthogonal series – a highly questionable procedure (common in dendro circles) criticized in connection with an earlier discussion of Wilson and Luckman 2002, 2003. Wilson et al 2007 perpetuates this methodology for the B.C. series as follows:
Multiple linear regression of three orthogonal principal components (derived from 12 RW and 7 MXD chronologies) were used to reconstruct each climate parameter separately. Calibration explained 64% (Tmax) and 39% (Tmin) of the variance in the instrumental climate record (1895—1991). In this study, we utilize the same data set as Wilson and Luckman [2002], except we removed the Harts Pass (Washington State) RW chronology (1585—1991) from the data set.
These statistical operations are not ones that one sees in econometrics or non-dendro statistics; in my opinion, they have negligible intrinsic validity and to be extremely prone to spurious fits and overfits.
The NRC panel recommended the use of averages in carefully selected tree ring networks as an alternative to principal components, I think that Wilson et al 2007 would have done well to have followed this recommendation. By not following the NAS recommendation, they have simply produced an IBC reconstruction that has little to recommend itself to a third party. Worse, had they followed the NAS recommendation, the resulting RW series would have provided further evidence of the divergence problem, evidence that has been swept under the carpet by their principal components-regression methodology. JGR reviewers should have observed this faulty methodology and required Wilson et al to correct it.
References: Rob Wilson’s publications are listed at http://www.geos.ed.ac.uk/homes/rwilson6/Publications/
Climate Audit 
49 Comments
I wonder if it might be worth the time to put up a page with a table that lists all the ‘climate’ publications and mark them irreproducible if they don’t have the data available. There could be a second column that indicates questionable statistical calculations (0-5scale). Other columns to indicate other types of defects: relying on corrections, faulty surface stations ..etc. Studies with out major problems could be marked with a gold star.
Yeah, grade them like they’re back in school… 😉
The script to generate the figures and results is here. This is a script that works on my computer and relies on prior calculations and is provided to be indicative rather than fully transportable.
http://data.climateaudit.org/scripts/wilson/collation.wilson.bc.txt
Publish a formal, short, temperate, objective, comparative, dispassionate, comment. Drop the bit about complying with NRC, that’s lawyer style arguing, not science. This will allow Rob to respond, formally, with justifications for the regression procedure. Will also get the alternately computed chronology (averaged) in the record. (Climate Audit is not the record, this blog could disappear tomorrow, the way Chefen’s blog did.)
“that’s lawyers style” In other words it’s persuasive. Can’t have that! Climate science should not have to respond to persuasive arguments that lead to different conclusions.
The lawyer-style is the appeal to an authority, vice statement of a mathematical, logical argument. What’s more, in this case, it’s one that is just a tribunal, one that has admitted to “winging it”, one that has not published mathematical expositions of its arguments, and one that Steve often finds problems with.
Trust me. The NRC blathering is fine for argumentative blog-wars. It does nothing for moving science forward.
Would anyone have been able to guess that Rob’s B.C. ring width data had such a pronounced divergence problem from any of the publications so far? It would be absolutely impossible. However once the data’s archived, the divergence can be identified with a half day’s work from scratch. It’s good that Rob archived his data so that others can evaluate its meaning more accurately.
However, given the inability to deduce the divergence problem from the information presented in Rob’s articles on these chronologies, it certainly makes one wonder what we would see if Esper ever archived his measurement data.
Could you please give the actual citations? “Wilson et al 2007 (JGR)” or “Wilson and Luckman 2002, 2003” are not suitable references. I wasted a bunch of time trying to find these papers, when a normal research citation would have got me there immediately.
In the past, you’ve done this and it was truly ambiguous (for instance with multiple Von Storch papers).
BTW, if you can keep a bibliography in MS Word (or other programs) so that it is very easy to add endnotes by pushing a button. Or even just cut and pasting. I recommend the Science formats, but any normal endnote format is reasonable.
Wilson and Luckman 2002, 2003 were previously discussed at http://www.climateaudit.org/?p=1324 linking to Rob Wilson’s publications listed at http://www.geos.ed.ac.uk/homes/rwilson6/Publications/
Steve, I think I figured it out by looking through the papers on Wilson’s site (for 2002, 2003, and 2007). However, you should really give proper cites. I shouldn’t have to guess that I have the right paper, that his web bibliography is correct. There should be no question since the page and volume number and coauthors and all that are listed.
Frankly I cannot decide if dendroclimatology is science or hocus-pocus with regards to temperature reconstructions. I’d like to read several good examples of dendroclimatology where the methodology is clear and the results are inarguable, to convince myself that it can work with temperature, even if the circumstances have to be ideal.
Can anyone cite a few such studies available on the internet? Thanks in advance.
RE 8. Some creatures stick their heads in the sand. Like the TCO. Totally Clueless Ostrich.
Google you silly goof. I found the stuff in less than time than it took me to think up
a silly name for you.
I did several googles. That’s what I mean when I said search is required. BTW, try googling “wilson 2007 JGR” on google scholar. You don’t get there from, here…
Asking for a proper cite, is not too much. It’s expected of undergrads. It is butt-basic. I won’t back down on something as simple as that. Oh…and axes on graphs…
David (#11);
Anthony Watts has caused me to wonder about the “instrumental” record, too, though for different reasons.
In any case, kudos to Rob Wilson for archiving his data, and kudos to SteveM for continuing to pursue this issue despite the tortuous path.
14: That’s one of the downsides of posting all those pictures and stuff. An impression is being made (promotionally), but the mathematical argument that the record is off has not been enunciated. Steve even states, cautiously, that he does not expect the need for a major revision of the surface record to be proven by the Watts work. But then he publishes all that in progress work. And the echo chamber here gets all pumped up about it. Heck Watts had to be prompted to even show what the standards are for recording stations. And he has a tendancy to get all excited about air conditioners or jets or whatever. But he has not proven an effect. People need to listen more to the Steve caveat on that work, to settle down and to be very clear about what the hypothesis is, what the design of experiment is, etc. on that work.
RE 13.
Dont use scholar. Regular Google found it in 3 clicks and 10 seconds.
10 SECONDS. that’s twice as long as you last.
Now, you’ve pumped twice and its all over.
TCO, your right. All that is being done with the pictures and sites is the collection of data. We don’t really know much yet. Although it may not be too early to draw conclusions as to the need for better oversight and management. However, it is kinda fun to draw inferences from time to time. Hey, I haven’t seen you here for quite awhile. Missed you. Thought maybe the alligators had finally got you:)
Welcome back TCO, I think. I actually agree with you in your #4. It goes downhill from there. #15 is junk: Anthony has nothing to prove other than whether the NOAA’s guidelines are being met, or not. Presumably they were written for a reason?
Barclay: thanks man. I think you and Steve have been rather careful. And I think Watts has a very kind, sweet personality. I don’t mind some fun or some looking at the half done data. What I have an issue with is the effect on the hoi polloi. I would also feel a bit better with a more thought out, or enunciated overall scope of design (what’s the hypothesis, how will it be tested.)
Joel: Glad we agree on 4. On the latter, not so quick. What is the specification for air conditioner distance. Things like height are clear of course. Even here though, there is one thing to say how many instruments are non-complying and quite another to decide if it has an impact on problems we are trying to figure out. Mathematically what the impact is (in average, trend or standard deviation). Or you are one of these people who just wants to find something wrong, so that you don’t have to interpret imperfect data? Usually these people don’t make a mathematical argument (confidence intervals or the like). Instead they tend to decide that the instruments are not perfect, therefore no discussion can go forward. Like that fellow who thought that accuracy divided by time period and did not understand that increasing the number of instruments/recorded measurements compensates for round off error. Willie or whatever his name was.
re 19
This is a list of questions that has been prompted by Watts’ research. That is the mark of a true scientific effort. Note especially that without Watts, climate science would have carried on without realizing that its underpinnings were full of possible error.
That’s fine if we consider it at that level. But some of the hoi polloi or even Watts’s harumphing about air conditioners 15 feet higher and 40 feet away from an instrument is unjustified harumphing. Read the comments on this site. Hoi polloi are making too much of the pictures.
TCO:
I thought a couple gin and pomegranates was required prior to welcoming you back to the echo chamber. You are sounding much too sober for my liking.
The recent warming seems to be related to the short length of the temperature data wire employed after 1980 or so. The ease in which non-field fools collect data near their office is not very surprising to an old field geologist. As is said, gold is where you find it. Fortunately, for the divergence avoiders, heat is everywhere, therefore morons will always manage to find it.
The owner of this blog is still struggling to get peer reviewed mainstream professors to cough up some raw data. He fails to hew to the modern as noise is the fetish of the age.
Your role as jester to the masses is the lowest of all occupations, no wonder you are a lush.
BTW I sent an email to Rob Wilson mentioning this post to him and saying that I would be happy to make any required corrections if I’ve incorrectly analyzed the data and offering to post up any comments or responses that he wishes to make in the head post (if he wants.)
#15 Please, TCO, spare us. If we want to get a current picture of the quality of the US surface temperature network, we need to get volunteers interested. If we want to prove an effect we first need to gather data. It’s both a community effort and a work in progress, and yet you seem to hold the idea that publishing ‘in progress work’ is somehow the wrong thing to do. As if a internet based community effort could be assembled in secret and then unveiled fully formed. If you don’t want to see work in progress, and you don’t want to see us trying to get interest in the survey, then I would suggest that you should go somewhere else. Perhaps you could go and denounce ‘getting pumped up’ somewhere where people think we should be wiping with a single sheet of toilet paper in order to save the planet?
re 21
TCO acknowledges
So the project is well run and ideveloping useful information that will be helpful to climate science. Rathr a good deal for everybody since it is being done by volunteers
This is as good an auditors-audit that TCO can manage:
C’mon, TCO. That’s ridiculous and you oughtta know it. Just as it’s wrong to harangue Steve about publishing or the tone of his request when the required replication data is not published…
…so to, Andrew doesn’t need to make a mathematical argument about the appropriateness of station siting, when there are quite reasonable siting recommendations out there, and the photographic evidence makes it obvious which sites are easily holding to the spirit of such recommendations.
Hoi polloi or not, a ridiculous site is a ridiculous site. When all is said and done, the Hoi Polloi will be disgusted if the “good” sites selected for the first analysis of Obviously OK Origins have any untoward objects anywhere nearby.
Once we have an Obviously OK dataset from a visual perspective (including evidence that the site has not been moved in a Long Time)… others can have fun analyzing the microsite impacts.
Me? I suspect we’ll learn something valuable from data with Obviously OK Origins. I think it’ll be data producing worth OOO-ing over 😉
Leave the mathematics of air conditioners, brick walls and asphalt to the GCM makers. They’re the ones who love to jigger fancy formulas rather than go into a field and read a thermometer, or go into a library and read about the MWP.
I think it’ll be data producing results worth OOO-ing over. (fingers slipped, sorry!)
Hmm,
The AGW alarmists claim that the sensor network is sufficiently accurate to pick up changes of 0.01C. On the other hand, they then tell us that having a sensor sited on top of an air conditioner blower is of no consequence.
#21 TCO
Good to see you back again. Personally I don’t have a problem with being a member of the hoi polloi (do you mean this as a derisory term? I presume you do?) who visit this excellent blog.
Hopefully you’ve been following a lot of the threads while you’ve not been posting as otherwise you’ve got a lot of catching up to do. Steve M has continued to show just how sloppy the HT members are and with Antony Watts’ help just how equally sloppy the USHCN crowd are also. The bucket corrections thread was a classic and in some of the recent threads Steve has shown just how ludicrous some of the homogeneity corrections are that underpin the claims for an unprecedented increase in the mean global surface temperature in the latter part of the 20th century that AGW hysteria rests upon.
While you’ve been away fellow hoi polloi member, Steve Sadlov, has continued to entertain us with his links to (non Cryosphere Today doctored) satellite images of ice extent in the Artic and Krysten Byrnes has sucessfully predicted the end of the drought in Australia. All in all a busy couple of months on CA.
It seems to me that this is what needs to be done with the surface station data situation:
Step 1: Go out and document each site, noting any deviances from siting standards.
Step 2: Determine the magnitude of the problems at non-conforming sites.
Step 3: Perform an analysis on the impact on the global temperature record from step 2.
Anthony Watts is currently engaged in Step 1 using a network of volunteers and with no funding. Despite what the blogging masses might be saying in thier comments, this is a commendable effort. The fact that it is being done on the internet and the data are available real-time as they are collected makes it even more open than most science done today. This is a good thing, regardless of what TCO thinks.
What should be decried is the fact that so many professional scientists just used this temperature data directly, assuming that it was good to a resolution of .01 degrees without taking the above steps. These stations were not originally set up to collect high-resolution long-term climate data, they were intended to improve the accuracy of local weather forcasting. Using this data for a different purpose without first determining the suitability for the new purpose (characterizing the errors, precision, etc.) is just bad science.
The point is that somebody finally got around to inspecting these sites so that these sort of questions can be asked. Without the site survey, one would just assume there is no need for any sort of adjustment or filtering of non-compliant sites.
Like Tamino’s (sp?) objections, the point of the exercise is completely lost and intentionally obfuscated. You know, take an effort that is mostly a good thing with lots to contribute then find a flaw thereby pronouncing the whole thing flawed. Pathetic really.
…and definitely a lawyerly way to argue.
mosher…stop it you’re killing me. heh. heh.
hey, TCO, this thread is about Wilson, not Anthony Watts work and your interpretation…
#21
Let us not forget when the breeze blows and the building that the air conditioner is attached to creates an eddy that takes the heat from the building and the air conditioner right into the thermometer. Then when the breeze blows from the other direction, it takes the heat from the paved surface into the thermometer.
TCO won’t be back for a while. Just trolling.
>> So the project is well run and ideveloping useful information that will be helpful to climate science
Will the effort lead to anything resembling the accuracy and complete coverage of the satellite data?
Answer: NO! Therefore, the value of Watts effort is to identify a data collection system as corrupt and useless from a climate point of view. It was designed to tell us what we want to know: how hot is it where I live. It was not designed to tell us anything about the climate, and no scientist interested in such information would create a surface based system for that purpose. For all those people living in concrete jungles, moving the sensors to rural areas doesn’t serve them. For real scientists, they already have the satellite system that tells them all they need to know. This point renders all the above discussion about AC unit specs as irrelevant.
RE: #19 – I think you may be getting hung up on the air conditioner element of this thing. It is much, much more than that. Microsite issues are generally a composite of various ills. While air conditioners are a highly attractive symbol of the problem, they are only part of it. Also, I have to call you out on your own exageration – most of the serious issues involved air conditioners that were at about chest or head high and a lot closer than 40 feet.
Yeah, TCO is trolling.
I’d say the question is “Is the network a high quality network?” How do you do that? Find out “Do the stations meet the standards?” We have that answer. Not all of them, no. “How many?” requires more data. “What does it mean?” requires more data. “Are the adjustments valid?” requires more data.
But let’s say that rather than doing a survey to determine the state of the network (which for some odd reason hasn’t been done already…)
Let’s say what is being done is to try and invalidate the network in some denialist plot. Or Let’s say what is being done is to try and validate the network in some alarmist plot. Or even validate/denialist or invalidate/alarmist. Or totally neutral. Whatever.
The results are the same; the first 5 or more pictures in a photographic record of the site, notes on possible biases to the readings, specific long/lat, type of shelter at the location, and some possible answers about site history.
See, the nice thing about this all is the information will be there for anyone to see. It doesn’t matter what you think, you can go look at it.
Any other question (are they suitable for measuring climate) is a different issue, I’d think.
Geez, SteveM, sorry about getting this thread so far off point. I blame it all on TCO:)
But getting back to your point: “Would anyone have been able to guess that Rob’s B.C. ring width data had such a pronounced divergence problem from any of the publications so far? It would be absolutely impossible. However once the data’s archived, the divergence can be identified with a half day’s work from scratch. It’s good that Rob archived his data so that others can evaluate its meaning more accurately.
However, given the inability to deduce the divergence problem from the information presented in Rob’s articles on these chronologies, it certainly makes one wonder what we would see if Esper ever archived his measurement data.”
I am consistently amazed by the failure of some professional scientists to delineate carefully the discrepancies and counter-points in there data, and address them. Ok, Jaye, you’re probably right! It is very lawyerly of them:)
You know, I can’t help wondering about the phenomenon of false tree rings, whereby if a spring/summer has a cold snap, the tree stops growing and produces a false ring that looks like an annual ring, but isn’t (if you follow my drift).
The reason I think of this is the early part of Rob’s reconstruction (RCS method) looks a lot like the beginning of the Central England Temperature Record, except the “double-dip” appears almost 100 years later.
Try it Steve by plotting the two together and you’ll see what I mean.
I can only surmise that in the Little Ice Age, there were frequent unseasonal cold snaps that interrupted the growing season of many things.
#41. My guess is that dendros are pretty good on false rings and that’s not where I’d expect to see problems. My guess is that any similarity resulting from compression is spurious. I’ll try a plot sometime though.
Let’s take the thermometer picture taking discussion to a relevant thread. (Think Steve would want that.)
P.s. My 15 was in response to 14. I was not the first to raise the issue. However, I recognize, however, that when I make a contrary posting, that it has a more derailing impact than the off topic “skeptic view” post. I will try not to respond to such off topic comments.
We can do better than guessing. Dating methodologies specifically address this concern.
I remember reading that one of the reasons why Stradivarius could produce such great tonality in his violins was that the short growing seasons common in the Little Ice Age produced very narrowly spaced rings in the wood he was using. It strikes me that it would be very easy to mistake false rings for annual rings for that timeperiod.
I’ve no idea how to model compression and the phenomenon of false rings, but it just seemed interesting to me that the double dip of the LIA in the CET record appears stretched and shifted further back, as if extra rings had interposed themselves into the record of the 17th and 18th Centuries.
I’m not saying for definite, but its peculiar coincidence.
Go read some of the basics on false rings. It’s a known issue and handled in wiggle matching. You don’t just need to start from ground zero and theorize like Aristotle. You can also read a textbook. Do both, if you want. But the brainless speculation like this, that the hoi polloi picks up and sees as brainstorms is just disgusting.
Um, what’s going on here?
You mean leave the science to the scientists?
#46 Why is it always up to us to answer our own questions? Why don’t you write a paragraph on the subject. I do that, and when I make errors, I try to clarify what I meant or fix what I got wrong. Are you interested in us learning something or not?
#47 No, leave the math of A/C, brick and road cover to the GCM makers, because they’re not interested in accurate measurements, obviously.
The Wilson article (“A matter of divergence: Tracking recent warming at hemispheric scales using tree ring data”) has now been published and is available here.
The abstract:
Wilson tries to address the divergence issue head on, he uses new proxies which do not diverge against local temperatures. He tries to follow the NAS guidelines, and comes up with a few of his own. He has archived the data. All of these seem commendable.
However, to me at least, it’s not surprising that you can find tree rings that match local recent temperature records, but how to be sure they match over the last 4-100 years?
McIntyre and McKitrick are cited neutrally.
One Trackback
[…] been re-visiting some proxy data; I noted last summer that Rob Wilson had archived a considerable amount of B.C. data in Aug 2007 and noticed that he […]