Buy

Books
Click images for more details

Support

 

Twitter
Recent comments
Recent posts
Currently discussing
Links

A few sites I've stumbled across recently....

Powered by Squarespace
« Money without accountability | Main | McKibben's mindpoop »
Thursday
Jan142016

Wilson trending

Rob Wilson emails a copy of his new paper (£) in QSR, co-authored with, well, just about everybody in the dendro community. It's a tree-ring based temperature reconstruction of summer temperatures in the northern hemisphere, it's called N-TREND and excitingly it's a hockey stick!

I gather that Steve McIntyre is looking at it already, so I shall leave it to the expert to pronounce. But on a quick skim through the paper, there are some parts that are likely to prompt discussion. For example, I wonder how the data series were chosen. There is some explanation:

For N-TREND, rather than statistically screening all extant TR [tree ring] chronologies for a significant local temperature signal, we utilise mostly published TR temperature reconstructions (or chronologies used in published reconstructions) that start prior to 1750. This strategy explicitly incorporates the expert judgement the original authors used to derive the most robust reconstruction possible from the available data at that particular location. 

But I'd like to know how the selection of chronologies used was made from the full range of those available. Or is this all of them?

If so, there are not many - only 54 at peak. Moreover, the blade of the stick is only supported by a handful. On a whim, I picked one of these series at random - a reconstruction of summer temperatures for Mount Olympus by Klesse et al, which runs right up to 2011. Here's their graph:

You can see the uptick in recent years, but you can also see an equally warm a medieval warm period. This is only one of the constituent series of course, but I think it will be worth considering this in more detail as it does raise questions over the robustness of the NTREND hockey stick blade. 

PrintView Printer Friendly Version

Reader Comments (171)

"We pick the stats that support the conclusion. That is how evidence-based science works" (A.N. Climatologist)

Jan 14, 2016 at 10:34 AM | Unregistered Commenterbill

From the summary of N-TREND:

[...]we utilise mostly published TR temperature reconstructions (or chronologies used in published reconstructions) that start prior to 1750. This strategy explicitly incorporates the expert judgement the original authors[...]
So, is there a list of these 'authors' and TR series, Bish? Does it by any chance include anything beginning MBH?

Jan 14, 2016 at 10:38 AM | Unregistered CommenterHarry Passfield

As hockey sticks go, that looks more like some of the pool cues I've played with.

Jan 14, 2016 at 10:43 AM | Unregistered CommenterBloke down the pub

Harry Passfield

Yes, there's a list, but Mann doesn't prepare TR chronologies, so no he's not on it.

Jan 14, 2016 at 10:51 AM | Registered CommenterBishop Hill

Explanation copied in twice.

Also, where's the link to the paper?

Jan 14, 2016 at 10:52 AM | Registered CommenterM Courtney

Hi BH followers
please read the paper carefully before making comments - it will help you formulate objective criticism.
All available here: https://ntrenddendro.wordpress.com/
regards
Rob

Jan 14, 2016 at 10:53 AM | Unregistered CommenterRob Wilson

This strategy explicitly incorporates the expert judgement mistakes of the original authors used to derive the most robust reconstruction possible from the available data at that particular location.

Fixed it for you

Jan 14, 2016 at 10:54 AM | Unregistered CommenterTerryS

...This strategy explicitly incorporates the expert judgement mistakes bias of the original authors used to derive the most robust reconstruction possible from the available data at that particular location....

There. Fixed that fix for you...

Jan 14, 2016 at 11:38 AM | Unregistered CommenterDodgy Geezer

As a complete ignoramus of tree ring "technology" I must say it is extremely refreshing (unique) for the author to comment on here.
Quodus to Mr Wilson and give him credit for his appearance on B H.
It would be a step in the right direction for some of his peers to do likewise.

Jan 14, 2016 at 11:41 AM | Unregistered Commenterpatrick healy

How is the anticipated increased CO2 fertilisation effect on tree ring growth distinguished from the higher temperature growth effect?

Jan 14, 2016 at 11:42 AM | Registered CommenterPharos

To me, the Medieval Warm Period & Little Ice Age are clearly visible!

Jan 14, 2016 at 11:45 AM | Unregistered CommenterAlan the Brit

How do they eliminate all the other factors that affect tree ring growth, like water, grazing, fertilisation, fires, etc? I'm not an expert in tree growth, so I would like to know.

PS I have lots of trees, including a couple of hundred I planted about 9 years ago and the growth seems to be mostly affected by grazing animals, squirrels and by how dry the summer is.

Jan 14, 2016 at 11:48 AM | Registered CommenterPhillip Bratby

Hi Rob Wilson,

Your link just takes me to an 'About' page, one which seems to exist in isolation from anything else, not least your paper.

'N-TREND' is a pretty stylish name though for your group.

Jan 14, 2016 at 11:56 AM | Registered CommenterJohn Shade

While I remain sceptical about the magnitude of the late 20th Century uptick, the graph does clearly show the Little Ice Age, and also the 'seven ill years' of the 1690s - which were so cold in Scotland that many harvests failed, resulting in famine and in some areas a significant decline in population (15% overall and in some areas up to a third). So why are we worried about a little warmth and recovery from the LIA?

Jan 14, 2016 at 11:56 AM | Registered Commenterlapogus

Mr Shade,

The site is a wordpress blog, you should get a menu on the left including a link to a publications page, where you can get the paper

https://ntrenddendro.wordpress.com/publications/

Jan 14, 2016 at 12:01 PM | Unregistered CommenterPhil Clarke

Thanks Rob Wilson for the link to your paper.

John Shade - click on "publications" link in the side bar to the left - this page has a link to Wilson et al 2016 both the paper and the supplementary information.

Jan 14, 2016 at 12:06 PM | Unregistered CommenterMartin

Yes, I'm looking it. But in the meantime, I agree with the above commenters about the error in your remark about the MWP. The graphic that you've shown above only starts ~1500, long after the MWP, which peaks in the 11th century and ends ~1250 in conventional dating e.g. AR5.

Jan 14, 2016 at 12:19 PM | Unregistered CommenterSteve McIntyre

Alan, the MWP is clearly visible (in the first pic) to the authors also. From the abstract:

"N-TREND2015 indicates a longer and warmer medieval period (∼900–1170) than portrayed by previous TR NH reconstructions and by the CMIP5 model ensemble"

For those who aren't familiar with Rob Wilson, he has commented here several times before.
He's also known to have written somewhat heretical things to "the team" such as this in 2005:

"There has been criticism by Macintyre of Mann's sole reliance on RE, and I am now starting to believe the accusations"

See also a longer remark about how screening creates hockey sticks from noise here.

Jan 14, 2016 at 12:20 PM | Registered CommenterPaul Matthews

If so, there are not many - only 54 at peak. Moreover, the blade of the stick is only supported by a handful.

I counted 21 series still in use in 2000, and 8 in 2010.

Frankly, Fig.4 looks like an extraordinarily strong vindication of MBH.

Jan 14, 2016 at 12:25 PM | Unregistered CommenterPhil Clarke

It appears at no point do you show how a tree ring is a proxy for temperature independently. It's kind of fundamental since you are saying a tree ring can provide information like a thermometer yet provide no data as to the temp to ring relationship.

Fitting ring growth to local temperature wouldn't even do it. You need to characterise the species of tree. Otherwise you need to apply a large uncertainty which makes this exercise moot in terms of useful substance. It's a good theory paper though.

Jan 14, 2016 at 12:34 PM | Unregistered CommenterMicky H Corbett

"strong vindication of MBH"

Does that mean that Mike's Nature trick was unnecessary..?

Jan 14, 2016 at 12:50 PM | Registered Commenterjamesp

Tree rings are not temperature proxies, full stop.

It's pointless chewing over the methodology.

Jan 14, 2016 at 12:53 PM | Unregistered CommenterTim Spence

Bad climate science never dies, it just fades away only to be recycled a few months later

Jan 14, 2016 at 12:53 PM | Registered Commenteromnologos

The thing about 'Summer only' is that todays Summer temps in the northern hemisphere are not remarkably different even from 1930 according to Berkeley Earth here (admittedly using my mark 1 eyeball):
http://berkeleyearth.org/graphics/physical-effects-of-warming/

Jan 14, 2016 at 12:54 PM | Unregistered CommenterJamesG

In carrying out my own research and reconstructions of CET I do not recognise large parts of the graph. Firstly, whilst technically the first half of the 1500s were not the MWP THE period did share many of the warm settled characteristics of the mwp with some very warm summers and notable heat waves.

Similarly the 1730s were the warmest in the record until the 1990s according to Phil Jones.

Here I have this recurrent problems with using tree rings as thermometers. Why? They are perhaps useful as an indicator of moisture or generally good and bad growing conditions but they are highly prone to microclimates issues.

I have read papers, books, gone on short courses but try as I may I do not get this reverence some pay to the merits of tree rings as a thermometer of some accuracy
tonyb

Jan 14, 2016 at 12:54 PM | Unregistered Commentertonyb

It must be right, it has the word 'robust' in it.

Jan 14, 2016 at 1:05 PM | Unregistered CommenterRoger Tolson

The one thing that always stands out for me in these studies is the total lack of people like arborist who are the experts on tree growth . It’s almost as if they simply not want to be told that the idea of tree rings has a useful means to know past temperatures is ‘problematic ‘ to say the least . Now I wonder why that is ?

But as ever we find that settled science , is in fact built on data whose quality good be best summed as ‘better than nothing ‘

Jan 14, 2016 at 1:08 PM | Unregistered Commenterknr

MWP point accepted and fixed.

Jan 14, 2016 at 1:13 PM | Registered CommenterBishop Hill

I don't agree with this in the conclusions:
"It is quite understandable that when few data existed, there was no real option but to amalgamate all possible records together, ignoring the local/regional seasonal signal they represented, or their quality and individual biases."

Bad science is worse than none at all! The proper response to a lack of data is to obtain the data first. Even today the sks trolls don't realise that proxies without local calibration are not actually proxies at all. This issue invalidates all of these alarmist IPCC recons. Of course why would a scientist do such a silly thing as to use non-proxies as proxies? To influence policy? To get funding and a nice promotion by pandering to the fashionable catastrophe meme? Or just plain stupidity maybe. Lordelpus the most hockey-stick-like proxies didn't even match the trees surrounding them, never mind the local temperature. The first thing to do here is check which of the individual proxies have a hockey-stick shape. If there are only a small percentage that do, then the recon is probably just data-mining.

Jan 14, 2016 at 1:16 PM | Unregistered CommenterJamesG

You can take climate science out of a hockey stick-shaped graph but you can't take hockey stick-shaped graphs out of climate science.

Lame, I know. So, is climate science.

Andrew

Jan 14, 2016 at 1:18 PM | Unregistered CommenterBad Andrew

Tree-ring archives are annually resolved (facilitating robust validation), precisely dated (Stokes and Smiley, 1968) and the interpretation of their measurements is supported by a wealth of ecological and biological process-based knowledge of how climate variability in fluences ring formation (Fritts, 1976; Korner, 2003; Vaganov et al., 2006; Deslauriers et al., 2007; Vaganov et al., 2011; Rossi et al., 2012, 2013, 2014; Palacio et al., 2014).

Several commenters have said there is no justification for using trees as thermometers. While I'm certainly not going to try and track down all the above references, it seems premature to dismiss them without doing exactly that.

Jan 14, 2016 at 1:21 PM | Registered Commentersteve ta

"the interpretation of their measurements is supported by a wealth of ecological and biological process-based knowledge of how climate variability in fluences ring formation"

steve ta,

Note the first two words of this excerpt. It means the information from tree rings is subject to the introduction of human error. Science is supposed to remove human error, not introduce it.

Andrew

Jan 14, 2016 at 1:29 PM | Unregistered CommenterBad Andrew

In carrying out my own research and reconstructions of CET I do not recognise large parts of the graph. Firstly, whilst technically the first half of the 1500s were not the MWP THE period did share many of the warm settled characteristics of the mwp with some very warm summers and notable heat waves.
Similarly the 1730s were the warmest in the record until the 1990s according to Phil Jones.

You will be aware that while the CET starts in 1659, it is not considered reliable enough for comparisons with present day until 1772, because there were gaps in the record which were filled in with readings from Utrecht, and readings were taken indoors, amongst other reasons.

Jan 14, 2016 at 1:31 PM | Unregistered CommenterPhil Clarke

steve ta - yes tread carefully. Note the difference between "climate variability" and the narrower metric of temperature. And
always include searches for confounding material.

Jan 14, 2016 at 1:34 PM | Unregistered Commenternot banned yet

More drivel. A thermometer reading in an unheated 'indoor' environment (compare, for example, with the draughty, unheated indoor environment of a Stephenson screen) in 1659 would have a much better chance of reflecting the broader environment than one placed in the jet wash of a Boeing passenger airliner at Heathrow in 2015.

As for the 'better than nothing' argument, this is the logical filth that gives us 'ensemble means', where the average of dross is presented as gospel truth. Like the black line that runs up the middle of all those those failed, worthless CMIP models.

Jan 14, 2016 at 1:55 PM | Registered Commenterflaxdoctor

Rob, sorry but you lost it when you explicitly use subjective criteria.
The rest is just more rent seeking long winded climate arm waving.

Jan 14, 2016 at 2:03 PM | Unregistered Commenterhunter

Tree rings as temperature proxies just fails the smell test. Yes I can accept tree rings as corroborating evidence that one period was warmer/wetter/colder/dryer than another, in that broad-brush way, but when the Mannikins tell me that 1755 was 2 degrees F warmer at a certain location than in 1760 because the tree rings say so, then I reserve my right to be sceptical. I am sure at present they are in a position to call me an ignoramus, however .....lets wait and see. I am painting a picture in my head of Trenberth at my tomb asking for forgiveness, Gavin quietly weeping, Jones beating his breast, Mann with his head hung in shame.....

Jan 14, 2016 at 2:15 PM | Unregistered Commenterbill

Phil Clarke: You don't seem to have much confidence in CET. Do you think a tree ring is a better proxie? (serious question).

Jan 14, 2016 at 2:16 PM | Unregistered CommenterHarry Passfield

'Re: 'Hi BH followers
please read the paper carefully before making comments - it will help you formulate objective criticism.
All available here: https://ntrenddendro.wordpress.com/
regards Rob Jan 14, 2016 at 10:53 AM | Unregistered Commenter Rob Wilson'

Thanks, Rob, for making the paper available to those outside the paywall. Much appreciated.

Jan 14, 2016 at 2:17 PM | Unregistered CommenterColdish

I appreciate the work that goes into things like this, but with all due respect, it's just *another* squiggly line.

Andrew

Jan 14, 2016 at 2:28 PM | Unregistered CommenterBad Andrew

Steve ta

If they believed that each tree ring set was adequately representative of temperature then they should provide the equation plus error in the supplementary material. Sadly if you use a tree as a tool you need to define how it works as a tool.

Otherwise it's jazz hands time.

Jan 14, 2016 at 2:31 PM | Unregistered CommenterMicky H Corbett

Whatever this study reveals about the changes in temperature, one thing is beautifully clear : the present climate is as well adapted for the growth of plant life as it has been for over a 1000 years. The ever growing population of the world will be grateful for that.

Jan 14, 2016 at 2:31 PM | Unregistered Commentermikewaite

Hard to take any reconstruction seriously that shows the last decade to be warmer than the 1930s. Also not encouraging to see certain names in the author list. Climate "scientists" need to bear in mind that, like most everybody else, they are judged by the company they keep.

Jan 14, 2016 at 2:33 PM | Unregistered Commenterchris moffatt

Perhaps this is the money paragraph:

"1161-1170 is the 3rd warmest decade in the reconstruction followed by 1946-1955 (2nd) and 1994-2003 (1st - see Table 2). It should be noted that these three decades cannot be statistically distinguished when uncertainty estimates are taken into account. Following 2003, 1168 is the 2nd warmest year, although caution is advised regarding the inter-annual fidelity of the reconstruction..."

Jan 14, 2016 at 2:36 PM | Unregistered CommenterEditor of the Fabius Maximus website

Invited review - Last millennium northern hemisphere summer temperatures from
tree rings: Part I: The long term contextpaper at http://www.st-andrews.ac.uk/~rjsw/N-TREND/Wilsonetal2016.pdf

Jan 14, 2016 at 2:52 PM | Unregistered CommenterEugene S. Conlin

Phil Clarke
Hmmm. It is normally Eli rabbett I have to put right on CET. I have discussed this with the met office and David parker who as you know created the 1772 record which is daily as opposed to the monthly 1659 record created by Manley. The record is good and robust. The indoor records are few, other proxies are used, The infilling is limited but comes from a climatically similar area. Which is more than can be said for much of the interpolation used by climate science.

I am in Austria and using a seven inch tablet so do not have access to my extensive research but find this smallscale work far too fiddly to carry on a conversation.it is not helped by the mad way in which the tablet fills in my words. Eli came out as eliminate Garnett...
However you can explain why you think tree rings have more scientific merit that CET

tonyb

Jan 14, 2016 at 2:52 PM | Unregistered Commentertonyb

While there are definitely things which deserve praiseabout this paper, especially the archiving of the data used, it is rather unfortunate the authors offered no explanation for how they actually choose which series to use. This post highlights the quote:

For N-TREND, rather than statistically screening all extant TR [tree ring] chronologies for a significant local temperature signal, we utilise mostly published TR temperature reconstructions (or chronologies used in published reconstructions) that start prior to 1750. This strategy explicitly incorporates the expert judgement the original authors used to derive the most robust reconstruction possible from the available data at that particular location.

But what it fails to notice is those "original authors" referred to obliquely in that quote are largely the co-authors of this paper. I haven't checked the entire list of data series used yet, but I counted something like 30 of them that were provided by the authors of the paper themselves. That means this quote is, to a not insignificant effect, saying "We picked series using a stratagey which explicitly incorporates our own, personal judgment, which is quite expert."

In fact, it appears a full third of the series are taken from a single paper, the PAGES 2K reconstruction for Asia. It'd be natural for people to wonder why those were all used while other data was not. My understanding is those 17 series were gridded results created by a reconstruction which used 229 tree ring series (that are not publicly archived), meaning it may be okay to use them all due to their gridded nature meaning they're spread out, but... some explanation of the decision seems warranted. If a concerted effort from a large group of people is going to be carried out, they would presumably have a plan and process they could share.

Incidentally, I feel kind of weird praising the authors of this paper for archiving their data. Because they're frequently the source of the series they used in this paper, they're also the ones responsible for data used to create some of these series not being archived :P

Jan 14, 2016 at 2:56 PM | Registered CommenterBrandon Shollenberger

Yes, Same questions as @Pharos

How is the anticipated increased CO2 fertilisation effect on tree ring growth distinguished from the higher temperature growth effect?
..........................................................Jan 14, 2016 at 11:42 AM | Registered Pharos

Jan 14, 2016 at 3:10 PM | Unregistered Commenterstewgreen

@ Rob Wilson

Kudos for showing up.

Probably these questions are answered elsewhere on the 'net already, but: how do we know tree rings record historic temperature, rather than - say - rainfall; or hours of sunshine; or something like that? Have tree rings been checked against known values of these in recent times, and found to reflect temperature accurately, rather than something else?

What I am wondering is how you control for gross local effects on the accuracy of the record. If you chop down half the area of a forest, presumably those now on an edge that were previously in the shadier middle would get more light and hence grow more for the next few years. If the nearest tree to your thermometer treee fell over (struck by lightning, whatever), presumably the same effect would occur. How do you avoid mistaking such growth spikes for increased [whatever it is] when in fact some physical local change occurred?

Jan 14, 2016 at 3:11 PM | Unregistered CommenterJustice4Rinka

By the way, for anyone who looks at the data file, don't get excited at seeing "Yamal." It's not the same Yamal series which was used in tons of previous reconstructions due to its strong hockey stick shape. It's an updated series that was created after the problems with the old Yamal series were too much to bear anymore. It looks very different from the infamous Yamal proxy series.

Or at least, I'm pretty sure that's right. It's difficult to keep all the Polar Urals (which also makes an appearance) and Yamal stuff straight. You have the original Yamal and Polar Urals proxies and the new Yamal chronology, but then there's the Polar Urals-Yamal regional chronology. Plus there were other alternatives considered instead of Urals-Yamal chronology that they came up with, and those inter-relate with neighboring areas like Taimyr, which also makes an apperance in this data set. It's a jumbled mess with no clear answers.

I can't imagine how a single, objective standard could have been applied here and have come up with this list. It looks like the authors just stuck in the ones they wanted.

Jan 14, 2016 at 3:11 PM | Registered CommenterBrandon Shollenberger

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Post:
 
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>