Scott St. George Jan Esper
Department of Geography, Environment and Society, University of Minnesota, Minneapolis, USA
Department of Geography, Johannes Gutenberg University, Mainz, Germany
Received 3 September 2018, Revised 8 November 2018, Accepted 9 November 2018, Available online 15 November 2018.
https://doi.org/10.1016/j.quascirev.2018.11.013 Get rights and content
Highlights
- Tree rings are the backbone of most last millennium temperature reconstructions.
- Maximum density is a superior temperature proxy than ring-width but is less available.
- The newest tree-ring reconstructions agree better with instrumental temperatures.
- They also fit the memory structure of instrumental temperatures more closely.
- It is imperative to develop new, long and up-to-date maximum density chronologies.
Abstract
We review the current generation of large-scale, millennial-length temperature reconstructions derived from tree rings and highlight areas of agreement and disagreement among these state-of-the-art paleotemperature estimates. Although thousands of tree ring-width chronologies are now available from temperate and boreal forest sites across the Northern Hemisphere, only a small fraction of those records are suited as proxies for surface temperature. Maximum latewood density is clearly a superior temperature proxy but is less available, with few densitometric records that are both long and up-to-date. Compared to previous efforts, the newest generation of tree-ring reconstructions correlate more strongly against hemispheric summer temperatures and show better performance in tracking decadal/multi-decadal variability and year-to-year fluctuations. They also fit the observed memory structure of instrumental temperatures more closely than their predecessors. These new estimates still show signs of the so-called ‘divergence problem’ (the apparent loss of temperature sensitivity under recent warming), but do not extend after 2004 and cannot be used to evaluate the impact of the past decade’s warming on northern temperature-limited forests. We caution against averaging together the latest hemispheric-scale reconstructions because they have each been constructed to suit different purposes and share much of the same underlying tree-ring data, especially prior to CE 1500. Past temperatures are recorded more clearly in maximum latewood density than total ring-width, so we recommend the Northern Hemisphere densitometry network be modernized through a new round of field collections and observations.
HT/Caligula Jones
I am constantly amazed that anyone can tease climate trends from tree rings. Do warm dry years look different enough from cool wet from average temp and precipitation that you can tell?
Someone should grow lab control samples controlling for all the variables.
whenever I walk through a clear cut forest I stop and compare the rings of adjacent tree stumps. They never agree with each other.
The rings record completely different growth histories, like they grew 1000’s of kms apart, not merely a few meters.
There are too many unknown variables. Wringing climate data from tree rings is voodoo science.
Tree rings record growth, not climate.
Klem – December 20, 2018 at 1:18 am
But, but, but ….. Klem, that is the “beauty” in using tree growth rings as a proxy for determining past near-surface temperatures.
Ya just “pick & choose” which trees (stumps) in the chosen woodlot that you want to include in your research. 😊
Readers at WUWT and Climate Audit, etc. are already aware, and have been for years:
https://wattsupwiththat.com/2011/04/14/i-can-see-yamal-from-my-house/
Anyone who goes on about how climate change skeptics are funded by “big oil” need to acknowledge stuff like this on the other side:
“we recommend the Northern Hemisphere densitometry network be modernized through a new round of field collections and observations.”
Wonder who will be getting the grants to do this?
whenever I walk through a clear cut forest I stop and compare the rings of adjacent tree stumps. They never agree with each other.
Heck, Klem. Mostly tree ring samples of live trees seem to be from non-lethal cores bored into/through the tree. So try comparing the rings of random chords drawn across a single stump. I try that every now and then and the tree rings do speak to me. They say — “This is a waste of time”
I find the notion that tree rings can (usually) identify the age of a tree branch or trunk to be entirely credible. I’ll even give them that they can probably distinguish between good and bad years for a specific tree. And maybe they can (sometimes) tell why a year was good or bad … maybe.
But the idea that they can tell me the annual average temperature to even a whole degree strikes me as being more preposterous than the oft debunked notion that a “technical” stock market analyst can predict future prices of a security based on various arcane patterns in past prices.
Heck, if you take a core of a single tree at different points around the circumference, the cores won’t agree with each other all of the time.
“HA”, in 30% to 50% of the trees in question there is no need to take a core sample to determine that the “tree rings” are not equal in growth from one side of the tree trunk to the other side.
Just have a look-see at the GROWTH of the canopy (branches).
If the canopy growth is “lopsided” then the larger side is the side that receives the most Sunshine.
Doh! In my part of the country – it can tell you that the prevalent wind is very strong, OR the neighbouring plants ( trees) aren’t all the same height or Density ( grows TO the light), OR the Giraffe / horse has eaten the branches on the “windblown side” etc. Bloated tree ring growth ( since most growth is at night) occurs with adequate water availability ( drawing in nutrients, also. ).
Agreed. And I would go further – treerings within different quadrants from the same tree also show radically different treering histories.
But if trees are not recording climate, then dendrochronology goes down the same scientific sewer as dendroclimatology. If adjacent trees have different treering histories (and I agree with you) then how on earth can you compare trees from radically different regions, to derive a date?
These so-called settled sciences are all snake-oiI science.
R
Plus 1000+
Please note I have been saying this for ten years – – if adjacent trees have different treering histories, then dendrochronology is also pseudo science.
This is a comment of mine from 2013, but there are earlier comments if GoogIe could find fhem.
https://wattsupwiththat.com/2013/08/16/oh-mann-paper-demonstrates-that-tree-ring-proxy-temperature-data-is-seriously-compromised/#comment-1086863
Ralph
I’m thinking that dendrochronology was originally “invented” as a means of reliably “dating” historical artifacts, ….. not for creating FUBAR near-surface temperature proxies.
To wit:
“Dendrochronology – the science or technique of dating events, environmental change, and archaeological artifacts by using the characteristic patterns of annual growth rings in timber and tree trunks.”
Actual, counting tree rings was invented by my dad to keep his two boys from bothering him while he took a break cutting our annual 20 cord of wood…
What is “the memory structure of instrumental temperatures”, please?
Great question.
It sounds like obfuscatory writing to say “unadjusted temperature record”.
Climate Consensus seems to be deeply rooted to the idea of abusing models and maths until the pre-selected conclusion is reached.
I have grown a few trees in my long life, rain when they need a drink is more needed for growth than a few degrees of temperature. The tree rings may make a good rainfall record.
Now THAT may be the most sensible thing written, yet! We are so used to the echo-chamber debates on Temperature. I too have queried the nonsense of tree-ring data – ANYONE working long enough in the Agri / Forestry industry will see that plants grow differently according to their location and locus. Shade, Aspect, Drainage, Exposure, etc. Exposure to weather effects + Parasitic Flora & fauna.
If water is constant, and temperature is the only variable, we should be looking at a chemical reaction WRT tree growth. If I recall correctly a chemical reaction time is reduced 50% for every 10°C increase. With that in mind, tree rings aren’t going to vary much with temperature variations of 1-2°C.
…for places where trees grow. So, deduct the Arctic, Antarctic, mountains, deserts, and the 70%+ that is ocean…
Oh, and they only grow during the day.
Oh, and they only grow (in most regions), from spring to fall (or fall to spring).
Etc.
As the saying goes: if you have the law, pound the law, if you don’t, pound the table.
Or, if you have data, show the data, if you don’t, use pretend data.
Another missing variable is the dates of budding and length of time held in the bud. I have seen late spring frosts and hail storms any time during the summer denude trees. The trees will grow new leaves, but they are small and fewer in number. The first fall frost stops growth. So the growing season varies in length every year. Whether the “growth season” is warm or cold may not be as important as these other critical weather factors. Tree rings (growth records) simply are determined by too many factors to be used for temperature estimates. Using them for dating chronologies is actually fun.
Yes, first frost and last frost are far more important growth factors than averages from a growing season.
I come from Northern Ontario. We have trees. Lots and lots of trees. We know trees, from an early age. (I probably knew more about tree rings in Grade 8 than Mann did in his first year of uni).
One year we had a devastating attack by tent caterpillars. It looked like April in June. Somehow, the forests survived (see: lots of trees. Still.)
Most people with a modicum of sense would know tree rings are good for, as you say, providing a local chronology, but it takes real expertise to extrapolate this into “science”, and a measure of ego to turn it not only into a career, but “distinguished” career.
Trees mostly grow at night… just saying.
Indeed. Strike that one.
\is that true?
trees/my beautiful bougainvillea only grows at night>?
don’t they need UV for photosynthesis to grow>?
I believe it is, but the OP can probably educate us better. My extensive research of clicking on the first few sciency google hits shows that while trees absorb energy during the day, they use it during the night.
I am sure, though, there may be different sciency retorts. Unlike climate science, botany appears to not have settled all its sciency bits yet.
Thx.
Any other comments?
I’m reading all these comments as they come through…. don’t you guys ever ( eva ) watch plants grow in a flower pot ? Plants RESPIRE at night when they burn up the carbohydrates produced by P/s and release CO2 Yes CO2! In daytime they form more Carbos and as they draw in water + Light energy, “Stretch” by elongation and opening of leaves / blossoms, etc …. Wouldn’t say it is all very complicated but one has to understand it – how they inter-relate. was taught this all many moons ago, like as with so much else – yet I wonder WHY I / we ( an older generation) has to suffer all this nonsense from the ECO-loons ? As an Engineer I also have to look at the interactions of Mechanical, Electrical and Physical systems on machinery – an interdisciplinary art, no room for the individual “specialist” – ‘Oh I am an electrical engineer and have no Idea about your hydraulics or mechanics, etc etc….’ Seems that modern mankind aka Society seems to think EVERYTHING can be boxed up individually where , what was t he phrase …. everything is unique and independent of everything else……
How much of a change in growth rates will a less than 1C increase in temperature cause any way?
Beyond that, as others have pointed out tree rings record how good a growing season it was, and most trees don’t grow during the winter.
Early paleotemperature reconstructions from tree rings were used to spread panic and alarm about Global Warming and remove the Mediaeval Warm Period as a comparison. No one was interested in whether they recorded rain, temperature or termites. People are now probably looking at them again to see if anything solid and accurate can be derived from them.
Don’t let Michael Mann near any tree rings. You never know what result you will get.
Or, perhaps the trouble is that you do…
Yes I do—Global Warming
Mann only needed one….
Mann only needed one tree…
And One (tree) Ring to rule them all…
simple physics tells me that as the temperature differential between the poles and equator grows larger due to Since we are the cooling from the top, very likely something will also change on earth. Predictably, there would be a small (?) shift of cloud formation and precipitation, more towards the equator, on average. At the equator insolation is 684 W/m2 whereas on average it is 342 W/m2. So, if there are more clouds in and around the equator, this will amplify the cooling effect due to less direct natural insolation of earth (clouds deflect a lot of radiation). Furthermore, in a cooling world there is more likely less moisture in the air, but even assuming equal amounts of water vapor available in the air, a lesser amount of clouds and precipitation will be available for spreading to higher latitudes. So, a natural consequence of global cooling is that at the higher latitudes it will become both cooler (winters) and warmer (drier summers). It is happening already, is it not?
… that was the intent, but it didn’t work out for Saruman or SaruMann.
“Before his fall, he was the chief of both the wizards and of the White Council (a league of all those opposed to Sauron). His knowledge and skill, especially of Sauron’s devices, was said to be great. However, his deep study of the One Ring and Sauron’s other magic corrupted him, and his overweening lust for power led to his downfall. He is one of the few characters in Middle-earth who is morally “grey” – serving neither good nor evil. He betrays both sides and ultimately works for his own ends.”
There are not two variables that affect tree ring widths, there are many dozens. Cloud cover related to heat stress. Altitude. Salt spray desiccation near sea shores. Insect pests. Soil fertility. Soil moisture. Soil pH.
Soil root pests (nematodes, fungi, etc). Canopy light competition. Tree health and age. Winter freeze duration. Soil salinity. Proximity to streams. Slope angle of incidence. Slope direction. Length of day in shadow From terrain and other trees. Duration of fog and stratus. Relative humidity. Date of first freeze-free day. Length of summer growth availability. Snow pack depth. Root structure. Soil erosion producing root exposure, including slumping. Parasitic lichen infestation. Fire damage and fire heat intensity. Overall climate regime related to months of peak rainfall. You would spend the rest of your life attempting to unravel them, which is impossible. All tree temperature reconstruction is stupid shit. Up to 1985 all tree ring width was related to rainfall only, and only in the gross sense of “dry” versus “drought” versus “wet”. People deconstructing temperature from trees hallucinate too much and probably pick up artifacts from statistical processing more that any real “signal”. If they don’t have an exact, quantifiable answer why “old” rings work and “new” ones don’t, then basically they don’t know what they are doing.
I have had a lot of growth of my trees considerably reduced by bark removal by squirrels and deer.
Grey Squirrels non-native invasive destructive pest. Had some in my loft a few years back, theyvdid a lot of damage incliding gnawing joists. Apparently Pine Martens are partial to Grey Squirrel lunches if you can get hold of a couple.
Townie amti-country activity incomers have aided the increasing population of various species of deer, including immigrant varieties.
I’ve found a pellet rifle to be an effective control of the grey squirrel population.
When they get into your house, and I don’t know of an effective way to keep them out, then they are no longer cute or worthy of protection.
Rat poison works well in the attic, as they are “Tree Rats”.
What flavor, Pop? I can’t get them to eat it.
Rat poison is great so long as you can remove the dead squirrel. It it dies in a place you can’t get to, you’re screwed, for 1 – 2 weeks.
A friend of mine has a cat that has a penchant for hunting. It cleared out three grey squirrels from the attic over the space of a few days!
Thx for the list, DK. I have saved a copy for future reference.
Even changes in local wildlife can affect tree growth.
If a wolf pack sets up a den near your tree, both damage from deer, etc will be reduced and nutrients will increase as the pack does it’s business in the area around the trees roots.
As before in other posts:
https://www.nps.gov/seki/learn/nature/sequoia-fire-history.htm
The data and conclusions are not cited in this study.
Thank you for linking to this, R.S. Brown. There is a lengthy link within this article ABOUT the study to the study itself, namely that the Medieval was the period most heavily plagued by wildfire.
Tree rings can give absolute dates of archaeological objects, as well as living and relict trees going back thousands of years. In this analysis of the sequoia forests, charcoal sediments and trees are cross-referenced and correlated with C-14 dating to show a history of wildfires (fire scarring). Importantly, the modern part of their record of fires and droughts correspond pretty well. It is apparently the oldest record of wild fire frequency from this area, and in fact shows the frequency of fires in the California sequoia forests going back to the Roman period. From this record, they claim that the Medieval period had the most frequency of wildfires. Unfortunately, they try to go further with this…
My only familiarity with Palmer drought index dealt with mapping of 20th century, and it did this with Wayne Palmer’s accumulated historical soil moisture, temps, and humidity records. It (now) shows paleo records of drought which incorporate Ed Cook’s tree ring data to reflect droughts of a thousand years ago. The article shows graphs which allege a close correlation between aridity and fires.
Nevertheless, from what scientists are saying these days, we may be on safe grounds to assume that droughts bear witness to more wildfires than wet years, so… insofar as the tree ring records accurately reflect wildfire years, they surely have their uses for drawing conclusions about climate, and I would be hesitant to throw out that baby with the bathwater of tree rings as proxies for paleo-temps.
\note that we are up for major drought time to come to the great plains of north America.
It happens every 87 years or so…
it was a disaster 87 years ago
http://ocp.ldeo.columbia.edu/res/div/ocp/drought/dust_storms.shtml
and it was a disaster to the bison population 87 years before…
How convenient that trees have developed Climate Change Insensitivity in recent years and can’t be used to compare to previous hot spells. I guess the real answer is that the trees can’t be fooled by ‘adjusted’ temperature readings and flattened hockey sticks, i.e. the trees haven’t responded to the increasing temperatures because they haven’t noticed the 1°C increase over the past 100 yrs.
“trees haven’t responded to the increasing temperatures because they haven’t noticed the 1°C increase over the past 100 yrs.”
Bingo.
If I’m ever put up on charges of climate denialism (hey, the signs are there, at least here in Canada), I’m going to go with the Johnny Cochrane defense: “if the rings don’t fit, you must acquit”.
“the trees haven’t responded to the increasing temperatures because they haven’t noticed the 1°C increase over the past 100 yrs.”
And the trees haven’t noticed the 0.6C cooling over the last couple of years, either.
yes. By my results we are down ca. -0.4 K since the past 2 decades or so,
pity very few people have noticed
reason
terra: unacceptable adjustments
sats: degradation at current solar exposure: minimum solar magnetic field strengths means maximum amount of the most energetic particles released from the sun.
[hence, don’t go to Mars, before they made an earth like atmosphere there….]
My question seems to have disappeared into the aether, so here it is again:
What is ” the memory structure of instrumental temperatures” please?
It’s a new word salad at the Post-Modernist Café!
My understanding is that is is a fancy term for autocorrelation. A well known property of some types of data, and often appears in time series data.
Seems to me that their procedure works like this:
1) They first look at all of the tree ring records.
2) Then they select only those tree ring records that agree with the instrumental temperature record.
3) They use those tree-ring records to make claims about the past.
4) Then they stand around and congratulate themselves on how well the records agree with the instrumental record, and how that shows that their historical reconstruction must be correct.
Color me unimpressed … this is just another attempt to rescue the Hockeystick. If they applied that same procedure to random red-noise, they’d get the same results …
w.
@willis
True. No decorations here. Rings should correlate with rainfall patterns. Global T: not so much. Remember global cooling causes Nile flooding and dry periods at the higher lats. It is happening now. Click on my name to read the report.
Speaking of decorations. I put the Xmas tree up here but so far none of the family have shown an interest to put the decorations up. Is this a global trend? Are we not doing Xmas trees anymore?
My family has. Maybe we’re in the correct Christmas Tree Decorating Temperature Zone while you’re too warm because of AGW?
Ha.ha. true. We r in SH and it is high summer. Too hot. And we have drought again.
Remember drought coming to nh as well. Most probably starting ca. 2019.
Willis
And yet people get paid to do this kind of “work”.
“And yet people get paid to do this kind of “work”.
Indeed:
“we recommend the Northern Hemisphere densitometry network be modernized through a new round of field collections and observations.”
Nice work if you can get it: look for a crappy proxy, miraculously find it, hype it to the gills as saying something it doesn’t, then ask for money to “improve” it.
What is a Climate Scientist idea of a nightmare … one man and a chainsaw.
>>
1) They first look at all of the tree ring records.
2) Then they select only those tree ring records that agree with the instrumental temperature record.
3) They use those tree-ring records to make claims about the past.
<<
Ah, sorry but that is incorrect at step 2. [that is how it should be, but that is not how it was done].
In step 2 they select those ring records that agree with the conclusions they want to make in step 3, claims about the past. Get it?
Explains why those records selected in step 2 (to align with the predetermined conclusions past) may NOT exactly agree with the " instrumental temperature record".
Hence the ‘divergence problem’, which only exists because they did NOT select those records which fully align with know temperature records, but only those that agreed with the conclusions that they wanted to draw…
BTW: My comment above does not necessarily apply to this study, but to most studies done by the alarmists before.
2) Then they select only those tree ring records that agree with the instrumental temperature record
≠==========
Willis, which instrumental record did they use? Local or global averages.
From what I read they are using hemispheric averages, while at best trees are only aware of local averages.
As such the correlation is spurious because there is no possible mechanism and as you note would also work just as well with random noise.
‘Maximum latewood density is clearly a superior temperature proxy but is less available, …’ Error bar: +=5c
Should have had my glasses on. Error bar: +/- 5c
Unless you are coding in ‘C’, then += works just fine.
Yeah it works, but you have now just added 5c to the error bar.
The problem is that summer temperatures don’t tell the important story. In terms of agriculture, what matters is spring and fall temperatures because those determine the growing season.
That’s also the trouble with average annual temperatures in general. A year with extreme summer and winter temperatures will have the same average annual temperature as one with moderate summer and winter temperatures. For agriculture, the effect of the two years could be dramatically different but you wouldn’t know it from the average annual temperature.
Trees make lousy thermometers.
“Compared to previous efforts, the newest generation of tree-ring reconstructions correlate more strongly against hemispheric summer temperatures and show better performance in tracking decadal/multi-decadal variability and year-to-year fluctuations.”
summer temperatures…
Semi-serious question. Does that mean that 75% of each years temperatures are written off as unknowable? At what point in the year do they lose any sense of that year’s record? Seems like more reasons for even more uncertainty about any record of a trend. As if they needed more uncertainty
Proxies have all kinds of limitations.
My favorite proxy is agriculture. During the MWP, agriculture was reliable in Greenland. During the LIA, agriculture ceased to be viable.
Wheat prices are available since the 1200s. That gives an indication of supply and indirectly of the climate.
We also have written data from China.
There are many many proxies that aren’t trees. This WUWT story gives an indication of what’s available and what’s involved.
Trees are only one proxy and they should be taken in context with all the other available data.
Did Mann use proxies other than tree-rings to construct his graph?
“Did Mann use proxies other than tree-rings to construct his graph?”
Which graph? He did a lot of studies.
Talking about hockey stick.
‘Compared to previous efforts, the newest generation of tree-ring reconstructions correlate more strongly against hemispheric summer temperatures’
Climate scientists (sic) do this all the time: in their joy for telling us how they are doing better, they INVALIDATE all their previous work. Justifying all skepticism.
This great ‘newest generation of tree-ring reconstructions’ will be invalidated in a few years. So it is simply nonsense. Surely the educated people producing it know that. Or does part of you brain have to be wired up wrong to enable being a climate scientist?
That applies to academics generally. They spend all their time doing logic and analysis of facts and ignore the part of the brain that considers context and lived experience. link
“These new estimates still show signs of the so-called ‘divergence problem’ (the apparent loss of temperature sensitivity under recent warming), but do not extend after 2004 and cannot be used to evaluate the impact of the past decade’s warming on northern temperature-limited forests.”
How can the tree rings be so smart and become so dumb. Just proves the education system is screwed up, I guess.
Huh hum, just listening to the humdrum waffle over the Drones at Gatwick, UK airport for several hours – yet the science was settled in legislation as to what you could, couldn’t fly near an airport and how the authorities could bring down a drone…
Global Climate Change has STUNNED BRAINS – if trees rings represent Growth due to temperature and NOT Water UPTAKE as opposed to rainfall, THEN a.) What about Ice layers, b.) Soil sedimentary layers, c.) Volcanic Ash Layers, and any other layering of deposition / accretion in whatever objects. …. Hull mann, great topic for anyone wanting to get a PhD at one of the post 1992 Universities ?
Here is a key sentence from the paper:
“If the cause of the divergence phenomenon was not unique to the past few decades, that sort of censored temperature response could cause an underestimation of the magnitude of earlier warm periods and produce biased estimates of climate sensitivity (Hegerl et al., 2006; Jungclaus et al., 2017).”
i.e. trees did not respond much to the recent jump in warmth, hence they cannot properly reveal previous warm periods, such as the MWP, hence this subject is a dead parrot.
LOL, beat me to it. Tree ring destiny is a proxy for temperature. Except when it’s not!
The parrot “is no more”, “has ceased to be”, “bereft of life, it rests in peace”, and “this is an ex-parrot”.
Click your heels three times and say “there’s no place like global warming” while gazing at tree rings. Then you’ll be sure to see the answer.
Time for my “black dog” story. Have I told this before? It shows that we often see what we’re conditioned to see or expect to see or maybe want to see. Here it is:
I came home from work early one day just after dusk and as I pulled into my driveway I saw a black dog run in the driveway, and at the time I thought, “interesting, a black dog.” I went into my apartment and shortly after realized that I’d been broken into, and that the screen on the window adjacent to the driveway had been cut to get in.
Since by chance I’d washed that window the day before (yes, really) there were fingerprints, and I called the police.
The next day I was siting out in the sun, reading, and I suddenly realized, I didn’t see a black dog at all. What I saw was the robber (dressed in black) jumping out my window. I was stunned that I didn’t recognize that fact until the day after. Recalling what I saw, I realized that I had actually seen him jump out the window and run down the driveway but somehow my mind didn’t register the “jump out the window” part. To be fair, I saw this out of the corner of my eye. But, it was right under the window!!
That is why I often say that to a hammer everything looks like a nail: you can see what you think you should see even if you don’t intend to. I know this from experience.
Except, of course, for consensus climate scientists. They always see the objective truth, and right away. In tree rings, too.
Don132
Averages, what averages?
24 inches of rain per year in the UK can keep everyone happy if it falls half an inch every Sunday night.
Farmers are happy as it keeps the crops growing, he has six dry days to do fieldwork and dry his hay.
The barbecuers are happy as they have dry Friday and Saturday evenings.
Compared with rain that comes ‘a foot at a time’ as one farmer said to me.
Both average out at 24 inches, one is useful the other not!
The UK only has 48 weeks in it’s years?
Yeah.
You see, we went ‘Metric’, ish!
Auto
“The divergence problem” is still there though. This tells us one of two things or maybe both things. a) this phenomenon invalidates the proxy – how do we know the fit is otherwise good for the other 500yrs. b) the tree rings are telling us that they detect that the record has been serially adjusted upwards and they refuse to accept this.
Long ago,when I learned that Mann used tree rings to measure temperature I asked my friendly local arborist if that was possible. He laughed at the idea. Nope, they mostly record the availabilty of water. Worse yet, if the tree roots on the North side of the tree get more than the South side roots, the tree rings will be wider on the North side too. Ain’t biology wonderful?
Imagine a factory made thermometers. After manufacture they placed these our in the parking lot. After 1 year those thenimeters that tracked the GISS global average temperature were certified as “calibrated”. These were shipped out to customers.
The thermometers that did not track the GISS global average temperature were marked “failed” and were discarded.
Most people can see why this process is wrong. Yet this is the exact process by which tree rings are “calibrated” to global average temperatures.
Statistically tree ring calibration is called ‘selecting on the dependent variable”. It is a mathematical error because statistics assumes a random sample. Calibration violates the randomness assumption, leading to false conclusions (spurious correlations).
because statistics assumes a random sample.
Henry says
very true. Most people donot realize this. Any trend observed must be from a random sample. Still, you may have to design a sampling procedure depending on what odds you want to assess.
Click on my name to read my report.
Still, you may have to design a sampling procedure depending on what odds you want to assess.
≠=======
Agreed. Random sampling is very helpful when combined with the central limit theorem as you can convert an unknown distribution into the normal distribution.
However the key term is not “sampling”, it is “random sample”. This is why bias is such a problem in experimental design because experimenters always want to throw out the outliers, which removes randomness.
The outliers (for example the divergence problem) is THE MOST FINDING of tree ring data because it reveals something unexpected.
It is the unexpected result that leads to scientific progress. Yet in all the years since it was discovered climate science has not been able to resolve the divergence problem.
instead climate science ignores the divergence and stagnation is the result.
THE MOST IMPORTANT FINDING
======
cell phone typo. a black dog!
When one looks at the history of science, it is the unexpected findings that drive science forward. For example:
1. heavy and light objects fall at the same speed.
2. the speed of light is independent of motion.
These were two unexpected results. Very simple in concept but profound in implication. They form the basis for much of modern physics.
Similarly the divergence problem in climate science is telling us something profound. It is telling us that there is something we do not understand. And as such, any conclusions regarding tree rings and temperature likely cannot be trusted.
The single most important discovery on the horizon for climate science may well be the resolution to the divergence problem. For example, one possibility is that using temperature as the most significant metric for climate may well not be good science.
In most of the world, human settlement is not dependent on temperature. It is dependent upon the supply of water. Thus it may well be that climate science cannot resolve many of the current questions because they are looking in the wrong place.
It could well be that temperature is a dead end for climate studies, and that climate is better explained by moisture. It could well be that the divergence problem is the outlier that will resolve this. But because of experimenter bias, the outliers are being discarded, leading to faulty statistics.
design a sampling procedure depending on what odds you want to assess.
==========
consider drug testing. you have a test that is 99.5% reliable (0.5% false positives). Much more reliable than the 95% used by climate science.
You test 1000 professional athletes for drugs, and you find that 10 athletes test positive for drugs. You would think that given the 99.5% reliability that the results are likely true. These 10 athletes are almost certainly drug users.
But you would be wrong, because your 10 athletes were selected based on the very thing you were trying to measure – drug use. In fact there is only a 50-50 chance that any of these 10 athletes are drug users.
Because your test is 99.5% accurate, you would expect to find 5 false positives when testing 1000 athletes. But you found 10 positives, so likely 5 of the 10 are true drug users and 5 of the 10 are false positives. And since there is no way to seperate the 10 into 2 groups of 5, the best you can say is that there is a 50-50 chance that one of the ten is a drug use. A far cry from the 99.5% reliability you thought you had.
This is what happens when you calibrate tree rings. You are separating the trees into positive and negative based on temperature. which is exactly the same as the above example separating athletes based on drug use. You cannot then use the tree rings to measure temperature because it will make the results look much more reliable than they actually are.
“These new estimates still show signs of the so-called ‘divergence problem’ (the apparent loss of temperature sensitivity under recent warming)…”
— Do you believe your tree ring data? If so, it is much more rational to suspect that there may be a problem with the accuracy of recent temperature data than to suppose that some unknown change has taken place in the biology of all the trees.
If you really do believe your tree ring data, and the data says that there has not been recent warming… then… oh look, a squirrel.
So proxy thats less sensitive to warm temperatures over the last decade likely would miss any warmer periods before instruments. No? The essence of the hockey stick
Bold astounding claims.
Bold astounding claims require astounding proof.
Abstract
And, properly selecting which tree ring-width chronologies to use requires omniscience or witchery.
“The so-called ‘divergence problem'”
Diminish the term and casually dismiss the divergence of tree ring-widths away from temperature significance.
That one phrase demonstrates exactly what part science is actually playing in this article and pseudoscience.
Followed by the ever so typical statement that nothing recent is available and that these records were carefully selected and massaged to fit their purpose.
Circular file, with prejudice; and demand the funds returned!
Commie Bob makes a very good point viz the limitations of proxies taken in isolation or out of context. Archaeologists do not rely on the evidence from a single discipline any more than a TSB investigator would upon finding a single feather in a plane wreck and concluding bird strike was the cause of an incident.
I cannot ever recall arriving at an evaluation site and saying ” Look, bugger all this digging, sampling and research malarkey. Just call in the Dendrochronologist and we’ll make up something based on their findings”.
Dendrochronology is a tool among many in a multidisciplinary team utilising data from pollen, molluscs, geology, material culture, zoology, biology, entomology, soil analysis, stratigraphy, oxygen isotope analysis and much much more.
I doubt there would be many in the field who believe that relying on a single data source is evidence of much of anything.
Indeed for much of prehistory the lived experience is completely lost to the record and at best can only be inferred. A single data source helps not at all when making that inference plausible.
However, the fact that the often heated debate continues is evidence of a healthy scientific community at work. When dialectic ends or is overridden and consensus is declared it should be cause for suspicion that science has ended and something else is taking its place.
@Ferd Berple
The arguments you made tell me that you have understood the [good] science of statistics.
Many people think that you can use statistics to prove anything, which tells me that they have not studied the subject….
I donot agree that global T is not really important. Like I said, it is the differential between the poles and equator that determines the “weather”…
Unfortunately, global T is not only made up by incoming energy,
it also depends on other factors, which I have tried to explain in my final report
[click on my name to read it]
Essentially
to estimate incoming energy: look at Tmax
to estimate energy lost to space: look at Tmin.
this will give you an indication of what is happening
another pointer: don’t trust any data that you have not verified yourself.
In my analysis of global Tmin, I took:
1) equal amount of stations nh and sh
2) all stations sampled are balanced to zero latitude
3) to summarize a station’s result I looked at the derivatives of the least square equations that you get by doing linear regressions over certain pre=determined periods of times, i.e. the speed of the change in temperature versus time in K/annum or C/year. This eliminates the need to rectify for longitude and altitude.
Go figure.