This new paper by Kempes et al published in the journal Plos One adds uncertainty to the already uncertain science of dendrochronology dendroclimatology and the attempts at tracking temperature from tree rings. According to this BBC story:
They found that a 2C (3.6F) increase resulted in the average maximum height of trees shrinking by 11%, while a 2C decrease in the nation’s average temperature saw a 13% increase in the predicted maximum height of trees.
Here’s a figure from the paper showing height change with temperature:
The BBC story continues:
“This looks at the basic physics affecting a tree, such as internal fluid flow and the structure of the canopy,” he told BBC News.
“We really wanted something that was based in those mechanisms but at the same time was, conceptually, relatively simple.”
He said tree branches formed a fractal, which meant that if you effectively cut off a branch and then enlarged it, it looked like a whole tree.
“If you nail down that network structure correctly, then you can use it to predict how things change with size.”
From this framework, the team then incorporated local meteorological data, such as rainfall and mean annual temperatures, to allow them to predict the maximum height of trees in the area.
When compared with official data collected by the US Forest Service, the team found that their predictions tied in closely with the actual measurements.
Clearly, there’s more to tree growth than a simple linear relationship with temperature, and this finding shows an inverse relation with temperature to tree height. Maybe this is why Briffa had to truncate uncooperative tree ring data post 1960 and Mike’s Nature trick was used to “hide the decline”.
Here’s the paper abstract, link to the full paper follows.
Predicting Maximum Tree Heights and Other Traits from Allometric Scaling and Resource Limitations
Christopher P. Kempes, Geoffrey B. West, Kelly Crowell, Michelle Girvan
Abstract
Terrestrial vegetation plays a central role in regulating the carbon and water cycles, and adjusting planetary albedo. As such, a clear understanding and accurate characterization of vegetation dynamics is critical to understanding and modeling the broader climate system. Maximum tree height is an important feature of forest vegetation because it is directly related to the overall scale of many ecological and environmental quantities and is an important indicator for understanding several properties of plant communities, including total standing biomass and resource use. We present a model that predicts local maximal tree height across the entire continental United States, in good agreement with data. The model combines scaling laws, which encode the average, base-line behavior of many tree characteristics, with energy budgets constrained by local resource limitations, such as precipitation, temperature and solar radiation. In addition to predicting maximum tree height in an environment, our framework can be extended to predict how other tree traits, such as stomatal density, depend on these resource constraints. Furthermore, it offers predictions for the relationship between height and whole canopy albedo, which is important for understanding the Earth’s radiative budget, a critical component of the climate system. Because our model focuses on dominant features, which are represented by a small set of mechanisms, it can be easily integrated into more complicated ecological or climate models.
Here’s how the model and observations match:
Full paper: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0020551
h/t to WUWT reader “SandyInDerby”
And… if you look closely at the plot, there are no trees in Mississippi.
The study is continental in scope, but the relative grow is not just regional but very, very local.
I live on the eastern side of the Canadian Rockies but drive across to and hike in the the BC side regularly. Tree height is a function of temperature, moisture and wind, and different one side valley from another, as a result of the vagaries of wind and cloud patterns. Yet the pattern upslope from cool temperate to alpine and beyond is consistent. So the maximum height may show up as determined in this paper, but the relative height and growth is probably consistent as Briffa was looking at.
At the same time I can tell you that I did my own dendrochronological analysis of one (1) subalpine fir near at treeline in Alberta (to find out how slowly these trees at that position grow). The cut section showed huge variation one side to another, and some segments were missing growth entirely. Graphing the results made it look like there were years missing, also (step-functions in quadrants existed in 3 of the four quadrants at times, suggesting there were other, 4-quadrant “missing” sites also).
What I learned in the example and in the literature of Douglass, who invented dendrochronology, is that you cannot use a small sample, as individual trees respond individually, and that individual groups of trees respond to their particular, very, very local position. Of course you can have the lucky one tree that is the Rosetta tree of climate study, but that is unlikely. And whatever result you get will be inaccurate as you will blend different local conditions if your grouping gets too big, as well as too small.
Limits to knowledge. They exist everywhere except in the Hansen-Gore mind.
Any attempt to relate this study to “hide the decline” has to be based on misunderstanding. Dendrochronologists are well aware that not all trees are equal as temperature proxies.The use of tree rings as a paleo-temperature proxy is based on selecting trees that, due to their local micro-climate, are limited by temperature and not rainfall, that is, trees near the tree line where rainfall is abundant. Fennoscandia offers such locations. (Don’t ask me how SW US trees qualify, but that’s a separate issue.)
There may well be a nonlinear effect of temperature on growth specific to trees that are temperature limited. But since there is no indication that trees in this study were selected to be temperature-limited, the conclusions do not necessarily apply to trees selected as likely to be temperature proxies.
For those with some knowledge of dendrochronology, this is about as basic as confusing climate with weather. Don’t think I’m defending all of paleo-climatology, just saying that criticism needs to be more robust..
Higher biomass almost always reduces relative tree-ring width, this why detrending is needed to resolve dendrochronologies, for example, and that very detrending also removes low-frequency (long-term) climate change signals from the same record if used as a proxy-this is dealt with in a 2002 Science paper by Jan Esper of the Swiss forestry service, and why Stahle at U. Ark. (who has done interesting work on ENSO using tree-ring materials, mostly from cypress as I recall, in N. Mexico-he would hesitate to use such records also in long-term reconstruction (part of the raeson for the resultant hockeystick, among other intentional factors like cherry-picking Yamal, etc.). Thus the very mathematical function required in dendrochronology reduces its effectiveness as a long-term record, unless sophisticated techniques are used with a high-quality dataset. Many problems observed in this blog result from promoted sub-standard research, so sub-standard results in high-standard journals is systematic in this case. I suspect this is why Kevin T. likes to throw away all proxies controlled by radiocarbon (as the indicative values here are often much better, e.g., lowland calcareous tufa with fractionated 18O in equilibrium with atmospheric temps. from short-term H20 reservoirs), this places greater emphasis on annually resolved proxies of inferior long-term indicative value, an important consideration when you wish to, say, put a spotlight on recent warming at the expense of past trends, as has obviously been the case in so-called climate science.
Am really not complaining, as I realise the page is very slow and tedious to load and navigate (well it is for me) – but I did put this in Tips page on August 4th !
Bruce Stewart says:
August 8, 2011 at 9:24 am
“Any attempt to relate this study to “hide the decline” has to be based on misunderstanding. Dendrochronologists are well aware that not all trees are equal as temperature proxies.The use of tree rings as a paleo-temperature proxy is based on selecting trees that, due to their local micro-climate, are limited by temperature and not rainfall, that is, trees near the tree line where rainfall is abundant.”
Excuse me, but you seem to have forgotten the particulars of “hide the decline.” No critics have claimed that some particular phenomenon explained the decline. What we criticized was the failure of Briffa and others to undertake empirical research that would explain the decline. They did not do it at the time and they have not done it since. All Warmista pretended the problem did not exist.
When I cite this article as evidence against the scientific instincts of Briffa and others, I am simply pointing out that empirical research which would explain the decline is eminently doable and should have been done decades ago. In addition, I am applauding scientists who have undertaken research that is both empirical and experimental in a discipline, paleo-climatology, where outrageous assumptions about the reliability of proxies have been the order of the day for centuries. I hope that paleo-climatology is becoming scientific. The work that went into the Hockey Stick and involved the conscious decision to “hide the decline” was most certainly not scientific.
Friends:
I have read this thread with interest, but I am puzzled that nobody has mentioned the ‘elephabt in the room’.
Anybody with average or above inteligence must conclude from only a few seconds thought that the concept of ‘treemometers’ is plain daft. Many varibles would affect any tree more than temperature. But there were people of at least average inteligence who foisted this idea on the world and a countless number who accepted it.
Those who did the foisting and those not ‘intelectually challenged’ who accepted it must have known it was nonsense. But they did it and they got away with it for years.
Richard
I think that I shall never see a model lovely as a tree. I’ll call your attention to, in particular, Parameter Sensitivity (in Supplement S1), and Figure S2. Several exponents used in the model were estimates based on average values. Exponents have a lot of leverage, mathematically.
Note a deletion: “[Twenty] values less than minus 3 were omitted from the histogram.” Mike’s Nature trick strikes again? Figure 1C is called bimodal in the SI, but it doesn’t look significantly bimodal to me, there’s just a little bump on one shoulder. Was there a second peak that was lopped off? Where is the data?
Were seasonal patterns taken into consideration? I don’t see the words season, Spring, Fall, Summer, or Winter anywhere in the paper. Seasonally, low temperatures may be associated with more rain, boosting tree growth. On the other hand, lightning typically hits taller trees…
Note that use of just the tallest trees is problematic, a very narrow subset of the population from which to draw conclusions about trees. Once a tree grows above the forest canopy, its favored mode of growth may change. Microclimate is another huge factor, as Julian mentions above. Probably the most temperature sensitive factor is sap viscosity. High temperature, lower sap viscosity, more nutrients available for canopy and trunk growth–>shorter, stronger trees for the same effective leaf area. Taller isn’t necessarily better or healthier. Is this paper just more dildoclimatology? It’s hard to say without the model in hand, but I fear the worst.
As one who has been up in Canada’s high Arctic believe me trees are short in the cold and as one travel south trees tend to grow tall, larger, and thicker. Same with height of land – the higher you travel such as the Rockies trees growth eventually ceases entirely and disappears completely. Which means to me for tree growth one needs good soil, temperature, oxygen, moisture, and sunlight. Any one of the above lacking then tree growth is severely impacted. And of course I do not to fail to forget plate tectonics impacting the rise and fall of earth’s land mass nor the sun’s various cycles throughout history. The historical earth shows it had great trees up in the high Arctic when now-a-days any trees up in the High Arctic are a mere dream of what once was. Therefore, to me tree proxies are of no value when choosing an arbitrary date in a mere three or four place on an earth with a multi-billion year old history. What for instance if science were to only use petrified tree ring dating for only the high Arctic – what would be the conclusion? Or even tree dating from North of sixty? Well, that is the exact method today’s scientist use in isolation as proxies for the an entire planet – tree dating from today’s temperate zones. Once again today’s highlighted method are from temperate zones. Wow! Tree grow in temperate zones? Duh? Who knew?
Until science accepts that fact earth has a very dynamic history with multiple temperatures and extremes we all fail to deliver an accurate portrayal of this planet. It is no wonder why we become lost in the clutter –
Theo Goodwin says:
August 8, 2011 at 10:40 am
Thanks for the clarification. More power to all who request more relevant studies. My comment was not meant for them, but rather for those others who mistakenly take this to be a dagger in the heart of treemometers. There may be such a dagger, but this is not it. Suggestions that it is are not robust criticism, for reasons that I indicated. Robust criticism may lead to improved science, weak sauce doesn’t help.
Richard S Courtney says:
August 8, 2011 at 10:41 am
Of course, Richard – the proxy treemometer value is very weak as a direct indicator because of the uncertainties and inherent factors other than temperatures affecting tree growth – but they are another ‘indicator’, and cannot be completely dismissed. Personally, I think they do find a use especially in cross referencing to other proxies – which of course is the merit in all the temp proxies, they are no good as standalone ‘proofs’ but as a group, showing similar trends, they do give us valuable insight.
This is not true except for perhaps a (small) subset of trees. It reminds me of the “golden spiral” nonsense. http://www.lhup.edu/~dsimanek/pseudo/fibonacc.htm
Richard;
I’m not familiar with the mighty elephabt. Could you supply a link to a sketch or photo of one?
😉
Tree height shrinks with increased temperature?
As trees grow they begin to compete for light, moisture and minerals after awhile this competition will begin to slow each others growth, There are far more important and obvious influences on tree size. when It comes to tree height, temperature is insignificant compared to the other variables, when it comes to trees it’s all about the sugar!!
Carbon and oxygen are absorbed from the air, six molecules of water plus six molecules of carbon dioxide produce one molecule of sugar plus six molecules of oxygen.
This Molecule of sugar that the tree produces through photosynthesis is what the tree is after, the sugar gives the tree the energy to grow not temperature.
Less photosynthesis ==>Less Energy from sugar==>Less Growth==>Less Height==>Less photosynthesis. (It’s more complex, but this is the basic idea).
I’m lucky enough to live about 30ft from a Forrest (I can see it as I type this) and 100ft away from a couple of giant sequoia (Sequoiadendron giganteum) trees, these trees are enormous and despite being in a far colder environment they have managed to grow to a height to match giant sequoias from warmer environments.
Some of the biggest influences the environment has on a trees height are (and there are too many to mention) Forrest density and geographical location, the other interesting thing about forests is that they create a kind of artificial climate both in temperature and composition of gases (although it’s a natural process) the understanding that forests regulate their own temperature should be enough for anyone to see error in this explanation for the height of trees.
One more interesting thing about trees is that many types of trees have evolved to take advantage of Forrest fires, the evolution of tree cones is a result of Forrest forest fires over millions of years, The cone its self needs temperatures from a fire to open up and release the seed enclosed inside it.
When I read reports from the “experts” the MSM reporters find, who suggest that a so-called rise of Forrest fires is a result of Man Made Climate Change? I’m Like “Whaaaaat?” 🙂 it’s so funny but not so when you realize they are for real.
Giant Sequoia: http://www.rvforsaleguide.com/images/sequoia-general-sherman-tree-picture.jpg
Has anyone commented on why the temperature changed for the year of higher temperatures? If you garden, you know that your plants are generally lankier/taller in cloudy, cooler weather since the plants reach for the sun. In sunshine, some plants grow faster but bushier; some will grow faster and taller due to more food; many get taller but not as thick. I would imagine that trees respond in the same way. Perhaps different species react in a diverse set of directions.
Sparky;
I dung belief yiz. You cain’t possubly live nexta a forest and not no howta spill it!
Andy: How about a little more candor here? After all, WUWT isn’t RC. Dendroclimatologists interested in reconstructing temperature focus on trees where temperature is likely to frequently be a limiting factor. Thus Briffa focused on sites near the Arctic tree line in Europe and Siberia. Others choose trees at high altitude tree line. Dendroclimatologists interested in reconstructing rainfall choose trees in arid or semi-arid environments. In both cases, everyone recognizes that other factors influence tree ring width and these factors are part of the expected noise in the data analysis. Kempes is interested in the maximum height of trees (possibly as CO2 sinks). A different set of factors is appropriate for this analysis.
Kev-in-Uk says:
August 8, 2011 at 11:40 am
“Personally, I think they do find a use especially in cross referencing to other proxies – which of course is the merit in all the temp proxies, they are no good as standalone ‘proofs’ but as a group, showing similar trends, they do give us valuable insight.”
But doesn’t this take us into the realm of statistical magic and pea hiding? Just asking. Seems to me that what went into the Hockey Stick gives the lie to the use of tree rings as proxies.
Wow
wait a moment
THIS IS A MODEL
where are the empirical measurements to confirm this model has ANY bearing on reality ?
Why is so much of the thread above discussing this MODEL as though it was empirical research ?
OK I haven’t read the paper (why would I – it uses a model) but having been involved in Process Control modeling (not in detail) I know we can NOT model a Fractionation column in an oil refinery that only has some 30 odd WELL KNOWN AND WELL UNDERSTOOD variables. We spend a fortune on creating these models; and understand that they represent a snapshot of the process that is ok for training and SOME help in process prediction but has NO VALIDITY in doing any (repeat any) safety analysis modeling except for giving guidelines as to what should be investigated. Even for Process prediction; the older the model – the more useless it is (but still good for training !). An old model is anything that is more than 6 months old.
Models are not reality. They should only be used to demonstrate the terms of reference of some real world measurements; and as an INDICATION of how the theory being investigated MAY behave
But peter_dtm;
Your safety concerns deal with actual assets and people’s lives. Climate models merely provide projections which will determine the shape of the global energy economy, and whether hundreds of millions die of fuel poverty or not!
Oh, wait …
Brian H says:
August 8, 2011 at 1:20 pm
“Sparky;
I dung belief yiz. You cain’t possubly live nexta a forest and not no howta spill it!”
HaHaHa! too funny!! 1-0 to you 🙂
Momma said life is like a spellchecker you never know what ya gonna get! or was it a box of chocolates I don’t remember, who cares :), BTW look at your own spelling, it’s awful.
Sparky;
Yabbut, my mistooks was dumbliberate.
;p
BTW, the existence of Forrest Gump is prob’ly why yore spellchecker didn’t Czech you, mate!
😀
For the temperature to get warmer by 1⁰C in the atmosphere; first need to get warmer by 0,5⁰C. When the atmosphere gets warmer by 0,5⁰C; especially on sealevel; where the density of the air is greatest – atmosphere expands up by 100m. Up there is minus -90⁰C. Intercepts appropriate extra coldness to equalize / counteract in about 3,5 seconds. That extra coldness falls, somewhere else and becomes colder = overall is same warmth units all the time. So, when discovered warmer some place – declaring that the whole planet is warmer is same as saying: the planet’s atmosphere is warmer at lunch time by 9degrees than before sunrise…?! Not funny. Temperature is controlled by O+N, not by CO2. Oxygen and nitrogen are 998999 ppm. If interested in real truth / proofs / facts http://www.stefanmitich.com.au Otherwise, Skeptics and Warmist are barking up the same wrong tree. Part of the atmosphere can get warmer, and always does. The whole atmosphere cannot get warmer or colder simultaneously! That’s what the laws of physics and my formulas say. The laws of physics are regulating the temperature, not the shonky climatologist!
In same orchard, different health of trees. We are lucky that climatologist become agronomist. Going from east to west coast of Australia or USA; many different heights of trees / same CO2 level. Because H2O controls the climate, not CO2. Unfortunately, prof Plimer’s galahs in Australia cannot notice that. Reason for them the phony GLOBAL warming and the constant climatic changes are one same thing. That’s how they are dignifying the Warmist. That makes them more guilty about the ‘’flat rate carbon tax’’ than the Warmist. The neighbor is irrigating his orchard, I don’t. His trees have bigger rings – they are bigger, more prosperous trees; why? Same sunshine, same CO2 amount.
Verdict: Warmist should shove ‘’their tree ring theories’’ up their own rings!