Wednesday
Jan202016
by 
Bishop Hill
Bishop Hill Xing's bendy hockey stick
Jan 20, 2016 Climate: MWP You wait for ages for a hockey stick to turn up and suddenly two turn up at once. Hot on the heels of the Wilson et al temperature reconstruction comes Xing et al, a new effort from a Chinese team. In their figure, shown below, it's the blue line we're interested in. In truth it's a pretty bendy hockey stick.
;)
A few thoughts and observations based on a skim of the paper:
- The method is a novel one. Caveat emptor.
- They reconstruct annual temperatures from tree rings. Compare this to Wilson et al, who believe that you can only reconstruct summer temperatures from tree rings.
- The usual caveats apply to the blade of the stick - the amount of data falls away precipitously towards the end of the series.
- The overlaying of the instrumental data is mostly there for rhetorical effect. You wouldn't get excited from the proxies alone.
- Data have been screened in a way that requires more careful study for me to grasp. Caveat emptor again.
Reader Comments (50)
Such nice multicentennial variability
I don't care what anyone says, but I remember the MWP period as being warmer and a much nicer climate than the present day. It was a great time to be alive. There were none of those nasty, dirty, fossil fuel contraptions back then. We didn't have tree rings or smoothed reconstructions either.
Alarm bells ring for me due to the use of one of those very complicated signal processing methods. If you throw data at ANY algorithm you will get an answer, but what significance does that answer have? At the very least you should split the data in two, giving two answers, and look to see if those two answers resemble each other.
At the very least you should split the data in two, giving two answers, and look to see if those two answers resemble each other.
Check out Fig 6.
Mann re-re-re-re-re-re-confirmed.
Not a Hockey Stick, but a nice pronounced Sticky Uppy Bit at the end.
As climate models seem to be judged against other climate models, rather than reality, I find it confusing when Climate Scientists can provide alternatives to Mann's Hockey Stick, yet there is no formal response from the 97% Consensus who believe. Perhaps the famous 97% have lost contact with each other.
I do worry about "novel" methods & approaches to things!
Phil Clarke, Mann and his supporters reconfirmed as flogging dead horses.
How much money, time and effort has been put into Denying the MWP?
@golf charlie: Far too much! It's a wicked shame so much taxpayers funding around the world has bn squandered & wasted, when it could have been used to better the world!
smiffy, I remember it well, too, but at the time I read in the Anglo-Saxon Chronicle[*] that the death of King William II in a friendly-fire accident in the New Forest was due to the warm weather, and that we could only expect more of these events in the future.
[* Precursor of the BBC environment and global-warming 'analyst'. They also explained that King Alfred's burning of the cakes in the Somerset Levels was not inconsistent with global warming.]
First eyeballing!
That blue line looks no higher now than around 1940.
Of course the red temperatures appear much higher now! So which is wrong?
michael hart, King John would not have lost his treasure in The Wash if it wasn't for unexpected sea level rise, and his brother, Richard I, wasn't called 'The Polar Bear Heart'
It is nonsensical to believe a summer temperature can be reliably reconstructed from such a proxy as tree rings. How on earth can they produce an annual temperature?
tonyb
Taking it at face value, this chart suggests the following.
There was a marked MWP and there was a marked LIA. The recovery in temperatures was well underway during a period in which CO2 was little changed. Temperatures rose strongly until the Dust Bowl era. They then fell back for three decades or so (as originally reported by NCAR, the US National Science Board and Hansen). Temperatures had another attempt for the summit after the 1970s but peaked with about four to five years of the century remaining, at a level a whisker above the highs of the 1930s.
Ignoring any impact from confidence intervals (which surely must be at least as wide for the past than the present), the message of the MDVM reconstruction is that temperatures at the end of the 20th century were no higher than in the 1930s, and only slightly higher than the peak of the MWP. Moreover, the 1930-2000 “V” shape in the reconstruction implies that the revisions to the 20th century surface series – in particular the post-war period – have in the aggregate ended up incorrectly overstating temperatures for the recent decades.
I’m surprised this got published.
Figure 6 of this paper shows only that hemisperic-AVERAGE "tree health" has "rapid" oscillations that appear similarly from many statistical methods. Does an oscillation in that average reflect some change in regional or hemispheric climate, or does it just reflect changes in a few peculiar trees? The only way to answer that question is to show the degree of consistency amongst different subsets of the data, which the paper appears not to do.
I find this notion of the authors that there was in effect an 800 year long little Ice age from around 1100 to 1900 to be absurd.
There were many periods of great warmth during this period, and whilst there were severe cold interludes there were also blazing hot summers.
I wrote about the intermittent nature of the LIA here. There are several useful graphs which do not begin to match up with the extreme and protracted cold climate the authors have invented
http://judithcurry.com/2015/02/19/the-intermittent-little-ice-age/
tonyb
This paper seems to have three interesting findings from this chart alone.
1) That the Norman Conquest seems to have happened when the world was as warm as today.
2) Stratospheric volcanic sulphate injection has nothing to do with the reconstructed temperature.
3) TSI does seem to correlate a little with the reconstructed temperature, but surprisingly it lags.
So if this is right:
A) Clouds cool the planet.
B) Clouds block solar irradiance effects.an
C) Warmer weather aids amphibious invasions.
Which seems plausible.
The blue line is 1C warmer than the coldest period -- 1630.
Thank god. Or CO2 or whatever.
M Courtney, 4:30 frog invasions have historical precedents, but I am fairly sure that Hannibal used elephants to cross the Alps, having found frogs legs unsustainable for longer journeys.
I have been a sceptic for long enough, I give up. I am now convinced that trees do have rings
-Next!!
It's not a hockey stick of course, and the MBH 98/99 results are once again disconfirmed. But it still suffers from a late 20th century divergence.
Moberg LF is pretty much just Loehle by the way (up to a scale factor and offset). If Xing uses the same, or similar, low-frequency proxies as Moberg, it wouldn't be a surprise to get the same result.
3) TSI does seem to correlate a little with the reconstructed temperature, but surprisingly it lags.
or temperature lags by almost a complete TSI cycle.
Do I see any uncertainties?
"In particular, reconstructions based solely on tree-ring data are prone to underestimate the temperature amplitude during the Medieval Warm Period (MWP) and the Little Ice Age (LIA) (e.g. [3]*)"
".. the combination of tree ring and other proxies in low resolution may bring about significant uncertainties in the final reconstructions [13]*"
The comparisons of considerable reconstructions have suggested that the application of inverse-regression-based methods, like those used in some researches [2, 3]*, is apt to result in large underestimation of low-frequency NH temperature variability
[2] Mann, Bradley, Hughes 98
[3] Mann, Bradley, Hughes 99
[13] Jones & Mann 04
Yet Phil C. takes this as an exoneration of Manns HS rather than harsh criticism of it? None so blind...
On rough reading it seems like they set out to calibrate against Moberg by stats and thusly achieved it.....except that they forgot to 'hide the decline' in figure 7. Oops!
JamesG, if you use enough forcings, anything in Climate Science can be beaten into a Hockey Stick.
Unfortunately for Climate Science, the Hockey Stick has been beaten too often.
Xing's graph above seems consistent with the mid-20th centring cooling. It shows up excellently, in fact, even though the global mean temperature datasets have largely adjusted it away into more of a mid-century plateau.
The graph shows two peaks. The current one shown is 0.1 degC higher than the last. And this is cause for concern?
ssat 6:24 of course this is of concern. How could something in climate science be peer reviewed with something of alleged significance not exceeding a margin of error? Very shoddy work by the peer reviewers for not correcting this breach of consensus.
Another question, what is the actual temperature that is labelled zero? Without that little bit of information the graph is meaningless because there is nothing to compare it with - maybe that is why they are all like this, to cover up their tracks.
From the location map of proxies in the article, it looks like it has lots of Graybill bristlecone pine chronologies included into the algorithm. Graybill bristlecones plus red noise will give hockey sticks. I don't see that there is anything novel here.
ivan, the 00.00 line is about normal, plus or minus a bit, and was all agreed to by the consensus of climate scientists, a little while ago, when somebody thought it would be nice. Since when ever it was, climate science has advanced in hops, skips and jumps, to wherever it is now, to within 00.03%, precisely.
Like all hockey stick graphs it is probably a nonsense built-to-order and based on dendroclimatology best suited to water and CO2 but at least it is good to see the MWP and the LIA are staging a recovery from suppression.
As it happens, I 'studied' dendrochronology in an archaeology class I was thrown into at Glasgow Uni. They didn't take it as seriously as this.
michael hart
'not inconsistent with global warming.'
Good enough for me. Great chronicle !
Steve McIntyre at 9.58pm
The novelty in this reconstruction is in the twentieth century. In Fig 4. (a) there is a very close correspondence between the 11y-smoothed reconstruction and the CRU NH temperature. At the very end the reconstruction diverges, with most recent temperatures no higher than mid-century.
It would be interesting to know how such accuracy was obtained. In particular I thought that with tree-rings temperatures appeared to decline post 1960. We now have two reconstructions in a week where post 1960 reconstructed temperatures rise.
Unlike Mann and IPCC hockey sticks, the color scheme does not hide the modern reconstruction under the temperatures.
Warmer weather has always been propitious to invasions, amphibious or not (mountain passes open, etc). But I appreciate the smile, dear Courtney.
Aila, frogs may lack load carrying capacity when invading, but they are better at jumping over obstacles, than elephants. If the Romans had only known how frightened elephants are of mice, Hannibal could have been defeated, at a cost of peanuts (well, maybe a couple of medium sized bags, bought at a chariot service station)
GC - High Altitude Amphibian Exploration has a proud place in British history. Don't forget Captain Walter Snetterton's 1927 expedition to explore Across the Andes by Frog.
FWIW, there's an oblique acknowledgement of the MWP in the BBC History magazine last month - 'Vikings in America . . . The Greenland colony survived until the mid-15th century when the impact of the Little Ice Age killed it off.' (Article by John Hayward, pp40-44).
Well, Charlie, Romans were aware of caltrops.
Bloody hell.
@ esmiff I don't care what anyone says, but I remember the MWP period as being warmer and a much nicer climate than the present day. It was a great time to be alive. There were none of those nasty, dirty, fossil fuel contraptions back then. We didn't have tree rings or smoothed reconstructions either.
My God, your're even older than I am!
Aila, the standard sized caltrop would have allowed the Romans to cook the legs of two frogs with one prong. Elephant sized Caltrops were less portable, but required larger barbecues, and the Alpine tree line has never regained the dizzying heights of pre-Hannibal times.
Rewriting climate history is even easier with a computer.
I always look for the high temperature excursion in this sort of reconstruction, the 1940-45 blip. Here it appears very smeared.
Why the blip?*
JF
*(C) Tom Wigley
I always look for the high temperature excursion in this sort of reconstruction, the 1940-45 blip. Here it appears very smeared.
Why the blip?*
JF
*(C) Tom Wigley
Julian Flood, was the 1940-1945 "blip" due to higher than average use of airfields? For reasons complicated by human megalamania, airfields with weather measuring equipment became very desirable landscape features during the same period. Due to the primitive technology in use at the time, the offices, aircraft, and meteorological kit, were all crammed into restricted areas, and independent corroboration of data was not encouraged.
Carrick ol friend. MBH 98 went only to 1400 and MBH 99 to 1000. The agreement of this new reconstruction with those two looks not bad over comparable periods. Apples, oranges and limes.
"In particular I thought that with tree-rings temperatures appeared to decline post 1960. We now have two reconstructions in a week where post 1960 reconstructed temperatures rise."
Thinking about what limited growth in the 1960s and then looking for situations where that was not happening.
https://en.wikipedia.org/wiki/Caltrop
"...I thought that with tree-rings temperatures appeared to decline post 1960.."
They do decline. Look at Figure 7.
Eli, would the similarities have any link to similar trees? Or is it just a coincidence?
Will golf charlie and Alla be celebrating the warmist year on record with plate of frog slegs poached in Scottish chablis, or an elephant haggis washed down with Patagonian pinot noir?
I'm a little late to noticing this post, but I want to point out it is wrong when it says:
Rob Wilson and his co-authors did not say this or anything like it. That they believed particular tree ring data was best suited for reconstructing summer temperatures rather than annual ones does not mean they believed tree ring data as a whole is incapable of reconstructing annual temperatures. Wilson et al intentionally chose series they expected to correlate better with summer temperatures. Had they wished to examine annual temperatures, they may well have chosen to use different, more suited, data.