For all the hubub surrounding Dr. Michael Mann’s hockey stick, the MWP, and throwing out data past 1960 because it didn’t seem to calibrate against the instrumental record, here is a way to put an end to the issue. Have Mann’s, Briffa’s and others tree ring samples submitted to isotope analysis. Given how much UEA and Penn State want to protect their research reputations, it seems to me that this would be an excellent way to settle the issue independently. Unless of course, they threw away the original samples. – Anthony
From a press release by the Arctic Institute of North America
Carbon and oxygen in tree rings can reveal past climate information
Isotope analysis provides accurate information
OTTAWA, DECEMBER 2009 – The analysis of carbon and oxygen isotopes embedded in tree rings may shed new light on past climate events in the Mackenzie Delta region of northern Canada.
Scientists have long looked at the width of tree rings to estimate temperature levels of past years. Larger rings indicate more tree growth in a season, which translates into warmer summer temperatures. But the analysis of carbon and oxygen isotopes in tree rings can also provide accurate data on past climate events, say researchers working in northern Canada.
In a paper published in the most recent issue of the journal of Arctic, Antarctic and Alpine Research, Trevor Porter, a PhD student in Geography and Environmental Science at Carleton University, and three other authors compared temperature data collected in Inuvik, Northwest Territories (NT) since 1957 with their own analysis of isotopes found in white spruce trees in the Mackenzie Delta region of the NT. They found a strong correlation between the two data sets and temperatures.
“Isotope analysis is a good way to measure past climate change,” says Porter about the results.
Isotope analysis is not a new way to measure past air temperatures. However, the method has not been widely used because lab costs have been prohibitive, especially when compared with the examination of tree ring width. Now, however, the cost of equipment has dropped substantially making it more affordable for researchers to use this method.
Porter’s work was carried out on the northern edge of the boreal forest in the NT where trees are small but surprisingly old. “A 15 to 20 cm. tree could be a 300 to 400 year old tree,” says Porter.
This slow rate of growth actually helps researchers because smaller trees stay standing longer. Trees that fall begin to decay making data analysis difficult or impossible.
“Once they get too large, it’s difficult for trees to persist. They are susceptible to wind and ice storms. One of the reasons trees (in the North) persist so long is because they don’t grow as much,” says Porter.
Isotope analysis allows researchers to conduct their work using a smaller sample size than needed when trying to re-construct temperature records using tree ring width. Porter explains that the width of rings can vary considerably between trees even when they are growing in the same stand. This variation can complicate reconstructions of past climate.
A number of factors influence ring size, including the age of the tree and the location of the tree within the forest. Older trees tend to have smaller rings than younger trees. And trees within the same area might not all receive the same amount of light, nutrients or even water.
“Growth is controlled by many things . . . they (trees) can all end up just a little bit different,” says Porter.
Isotope signals, on the other hand, are often very similar between trees. This means researchers can gather accurate data from three or four trees instead of the 20 they might need for tree ring width analysis.
“In ring widths there will be more variability between trees. There will be similar trends, but you have larger differences that you would find between the isotopes of different trees,” says Porter.
Porter is hoping his work will lay the foundation for a model that can be used to investigate the long-term climate history of the Mackenzie Delta region. Although the temperature record for Inuvik only dates back to 1957, the dead and live tree ring record stretches to nearly 1000 years before present. That prospect excites the young researcher.
“The tree ring record goes back almost a thousand years in this area, but it’s never been used for a temperature reconstruction. This is a really exciting time to work in climate research, especially for a young student,” he says adding, “This is a hot topic.”
More information can be found at www.arctic.ucalgary.ca
See this press release in PDF form here
Bill Illis, I didn’t disappear. I went to bed. You mustn’t conflate the Zachos study with the tree ring isotopes. Yes they are both measuring the isotope composition of things but the controlling factors on the isotope composition of these things are different.
Let’s take the Zachos study first. This study reports the isotope composition of the calcium carbonate excreted by marine foraminifera. The isotope composition of the calcium carbonate has two signals: i) temperature and ii) the isotopic composition of the ocean water. The dominant signal is the isotopic com[osition of the source water which is the oceans. At any one time the ocean is relatively well mixed in terms of isotope composition with a range of less than 1 part per thousand variation. However, over time the ocean varies in isotope composition. The dominant effect is ice volume. i.e. when forming large high latitude ice sheets water that is depleted in 18O is locked up as ice. This ice, which is sourced from the oceans, mean that the oceans become enriched in 18O. It is bevcause of this we see the pulse of the ice ages.
It is difficult to disentangle the temperature signal.
For tree rings you measure oxygen, hydrogen and carbon isotope compositions. The oxygen and hydrogen isotopes are sourced from groundwater and the carbon from atmospheric CO2. Thus my statement that the oxygen and hydrogen isotopes are controlled by source water. However, during photosynthesis there is diffusion through the stomata. Diffusion leads to an isotopic separation and a change in the isotope composition of the water. This process is affected by physiological factors such as stomatal density which in turn responds to water availability. Hence my concern that the signal in tree rings will not be readily decoded in terms of temperature.
Measuring the isotope composition of tree rings using automated mass spectrometry systems has been possible for over a decade now. There are a large number of studies of isotopes in tree riings, including many that have a strong plant physiology basis. Few have shown a strong and unambiguous temperature signal. Dave Wendt’s citing of the Nature article shows the problems.
I agree that we need a way to determine proxy temperatures. That method needs to depend on a process that has a strong thermodynamic connection to temperature and our bewst option is using stable isotopes. However, I hope my discussion above shows that the problems are many when dealing with multi-phase systems (water-tree rings, or water-carbonates). Work is progressing on a new isotope thermometer that is referred to as ‘clumped’ isotope thermometry. This relies on measuring the ordering of 13C and 18O isotopes in the carbonate lattice and is very promising. But precision is as yet about +/- 1 to 2 degrees C and the measurement is technically difficult and expensive.
Geo makes a very good point about ‘magic’ trees.
I’m writing as a palaeoclimate scientist who has published over 60 papers in this area. There is a tendency in palaeoclimate science to ‘pick’ those proxies which are perceived to respond to temperature. We’ve seen this to be endemic in the tree ring literature. This is not satisfactory. The bottom line is if we need to posteriori pick proxies then the method has no scientific validity and we must reject the whole sample and method.. This is because we cannot be sure that they have responded to temperature in the past as it does today. Indeed we have a beautiful example in the divergence problem. We have a calibration period and as soon as one moves beyond the calibration period we see that tree rings do not respond to temperature.
One should note, however, that this is also the case for many other proxies. A good example is cave deposits, so called speleothems and their isotope composition. These are often picked because one responds better to temperature than others. Indeed in the same cave it is possible to find 2 or more speleothems responding differently to the presumed same set of environmental factors. One often reads of site specific studies needing to be done to understand the local control on isotope composition. This is really another from of the divergence problem, except here we have speleothems diverging from each other and the poulation is even smaller. Again if a spleothem responds to temperature nowadays there is no guarantee it will in the past. I can show you records from long lived speleothems that grew for the past 200 thousand years that show one oxygen isotope response to the last interglacial and a different one to the present interglacial. This is hardly a recipe for a well behaved thermometer.
It is for this reason that much of palaoclimate science is what I call, arm waving and wiggle matching.
Interestingly the one proxy that we fully understand is that of oxygen isotopes in marine foraminifera and the ice volume record. However, this is not a direct record of local temperature.
The goal has to be to find a proxy that we fully understand in terms of physics and chemistry and spend the money to develop the right analytical methods to make robust measurements that can be used to accurately and precisely assess temperature. It must be possible to choose any sample, at random from a population, to arrive at this robust estimate of temperature. Until we do we cannot assess past temperatures with any degree of confidence.
Quaking aspens give thumbs up to CO2. It could end in disaster says scientist.
“We can’t forecast ecological change. It’s a complicated business,” explains Waller, a UW-Madison professor of botany. “For all we know, this could have very serious effects on slower growing plants and their ability to persist.”
http://www.eurekalert.org/pub_releases/2009-12/uow-ggc120109.php
When we’re talking about tenths of a degree and varying shade/forest cover, then trees can never be used as accurate thermometers for temperature reconstructions. It’s impossible. Scientists should have some integrity and accept that instead of being blinded by fame and money.
For those doubting the O18 isotope method, Andrew Lorrey used that method on New Zealand cave speleo ring analysis which firmly put the MWP and LIA back in the climate record and proved both were global climate phenomena, not just european, and tied the climate swings to long term variations in ENSO, NOT to CO2 levels. I support this method and join in demands that it be applied to Yamal’s cores.
Well done young Trevor Porter. Good luck and good Science in your research.
An outstanding example to your seniors in the field.
Mike Lorrey, the Andrew Lorrey (any relation?) paper looks interesting. Yes, one often finds a good relationship between precipitation amount and the carbon isotope composition of speleothems. Similarly, one also often finds a good relationship between oxygen isotope composition and temperature and this may give good qualitative, and semi-quantitative information on temperature in the past. However I pose the following questions:
1) Has the empirical relationship between cave temperature and oxygen isotope composition of speleothems for the New Zealand caves been determined. I haven’t had a chance to read the paper but my reading of the abstract and contents suggests that the oxygen isotope record has been compared to other proxies and not directly to temperature.
2) How does this relationship compare with what we know of equilibrium oxygen isotope partitioning between water and carbonate? If it is different it indicates that there is disequilibrium in the system that may vary over time.
3) Has the record been subject to testing by use of a calibration period and then a period in which the temperature is predicted and compared to actual temperatures?
4) Are their truly independent and accurate age models for each of the proxies. Perhaps growth layer counts, tree ring chronology, U-series and 14C dating that allow cross correlation of the records? If not then we are ‘wiggle matching’.
If not then as I’ve said in earlier posts the speleothem proxy, whilst very useful, doesn’t allow us accurately and robustly to estimate temperature.
We must ask these questions and be very careful of not falling into the trap of ‘confirmation bias’. Just because the speleo record matches (wiggle matches?) other proxies doesn’t mean it is a true record.
I’m not doubting the value of the Lorrey record and will read it when I can get free access tomorrow. It is another piece of the jig saw but in all probability is not able to answer positively to some, if not all, the questions I pose above.
I hope readers don’t get me wrong. I have been, and am still an speleothem researcher but have a realistic outlook on what these archives can tell us.
So Splice, since there are too many uncertainties in using the isotope data, I think the climate science community should close down all Stable Isotope Labs (like the one at the University of East Anglia) and save some money for better research that has more certainty. Maybe someone wants to write a letter to the UEA about this.
The last times the isotopes got discussed at Climate Audit and Jeff Id’s site, a poster made the exact same points as you did in exactly the same way, using the same terminology and effectively just short-circuited the discussion. Then he disappeared never to post on any other topics. He might have thought he was being helpful or he might have thought it would be good to turn the sceptical community off using the isotope data.
In my last PaleoClimate post, Paul Dennis of the University of East Anglia promised to write an Isotope Primer for us within a week and it is now 7 weeks later. I said I would wait on posting Part II of the series until Paul Dennis posted up his article and Paul Dennis has not posted on any other topics that I am aware of. Obviously, other events might have taken over since then but this just effectively short-circuited the discussion again.
The sceptical community should be using this data since it is far better than tree-rings and we should also move forward on using isotope analysis from trees as proposed in this article.
It might actually be able to prove the WMP and the LIA and the Roman Warm Period were real climate events and then the hockey sticks can be put away for good.
Bill Illis, I think you are misrepresenting my views and taking an unecessarily combative position. The sceptical community should be using the best available science and some of this involves isotopes. But we have to talk from a position of authority and knowledge and that means we have to have a full understanding of these isotope systems.
Now to your point about closing down labs. First stable isotope geochemistry is about much more than palaeotemperatures. It is about tracing sources, sinks and processes in the natural environment. Palaeotemperatures and geothermometry are but a small part of this. Ergo your statement about closing labs down is meaningless and lacks rigor.
Now to back to stable isotopes and palaeothermometry. They are our BEST option of getting anywhere near to being able to determine past temperatures. However, my point is that interpretation is not easy and we would be foolish to think otherwise. You readily cite Zachos and tree rings together and my point was they are measuring different things and not just temperature.
We readily criticise others for confirmation bias and it would be easy to go down that route with respect to isotopes.
Michael (17:36:59) :
Because WUWT is becoming so popular these days, please don’t mind if I contribute some OT points for the lurkers for teaching purposes.
CO2 Contributed by Human Activity: 12 to 15ppmv / version 1
Great Video Michael!
That’s what’s needed. Now we need another that shows people how much temperature change these bozos are talking about/ A Tenth of A Degree, here and there, whether F or C. A diagram of the depth of the atmosphere and the corresponding temperatures as altitudes increase…
******
Gregg E. (21:59:38) :
We show a remarkably constant leaf temperature of 21.4 2.2 °C across 50° of latitude, from subtropical to boreal biomes.
A main assumption underlying the use of 18O to reconstruct climate history is that the temperature and relative humidity of an actively photosynthesizing leaf are the same as those of the surrounding air
Now wait just a minute! How can the leaf temperature be constant across this wide range of latitudes and the same as those (temperature and humidity) of the surrounding air.
That implies that the air temperature and humidity of the air would be constant across this latitude range, which just ain’t so, as any nightly TV weather report shows.
******
I’m no expert, and just going from memory, but my take on this is that leaves are evaporation “machines”, regulated by the stomata openings, and that these openings act to keep leaf temps near a fairly constant ~70F (~21C), no matter where it’s at, tropic or sub-arctic. The purpose, I’d guess, is that photosynthesis is most efficient around 70F.
Remember, sun striking the leaves will, without this evaporation, cause them to be fairly hot, like any dry surface under the sun. So even leaves in the sub-arctic would otherwise be hotter than 70F in the sun even when air temp is lower than that.
So this would seem to throw a monkeywrench into the oxygen or carbon-isotope method. Except that the leaves are building sugars, and cellulose for stems, branch, trunk, etc, is produced in the cambium of the stems themselves (using the sugar as fuel), and the stems supposedly are at or near ambient temps, so perhaps the method could work.
Splice (04:28:07) :
“Mike Lorrey, the Andrew Lorrey (any relation?) paper looks interesting. ”
Yes, he is my cousin, got his BS in Geology from Boston U, MS in hydrogeology from U Maine Orino, and PhD in climatology from U of Auckland. As for your questions about the isotope science specifically, I would refer you to talk to someone like him, whose geology and climatology training combines to be most qualified to answer your questions. He’s very sharp. I recall one time after we had a ski trip to Sugarloaf, he returned to the mountain to do snow cores at the summit and did mm by mm isotope analysis of the snow, defining where the storm for every layer of snowfall originated by the isotope distribution of the dust in the snow.
Gregg E
Re your posting at 20:51 today . I saw the video of Ben Santer at Caltech ( except for the discussion) . He sums up the present science the IPPC has used to conclude AGW.
His lecture does not really address the methods other than atmospheric data collection in detail.
I realize there are big problems with some of the surface area temperatures re the urban heat island effect, but can you or others critique some of the other data he presents, in particular the atmospheric and oceanic data?
Thanks
I never bought the ring-width as a proxy for temperature hypothesis. Too many variables. On my small spread I find shade and precipitation to be two monster variables. The isotope idea sounds interesting, but totally out of my league.
It is also well known among tree survey personnel that vastly varied growth rates can be observed in trees located only a few hundred meters distant from eachother.
A simple visit to one of the huge tracts of pine planted for lumber and paper (all the same age chronologically) will show significant variation among trees.
Furthermore, issues such as soil composition, depth of the water table, precipitation drainage rates, location with respect to floodplain, etc; can and do cause signficant variation within a forest (separations of kilometers).
Dave Wendt (21:44:08) :
I managed to locate a full version of the leaf temp paper I referenced above
http://www.sas.upenn.edu/earth/pdf/nature07031.pdf
Thanks for bringing this up again. I remember discussing it here when it first appeared.
What it is saying actually is that trees have a way of keeping their ambient temperature constant, and actually, if there is any thermometer function in the tree ring width data, that is what it is also saying: trees give constant temperatures.
In view of this, even measuring tree temperatures with isotopes should show this constancy ( and the so called decline, which is a decline, considering the errors, only if one expects a rise).
Let us make a gedanken experiment: suppose one took all those nicely documented bones all over the museums, and go to graveyards that exist from the middle ages and take bone samples and measure the temperature given by the isotope ratio versus time.
You laugh?
Of course, because we know that humans have an internal temperature of 38.something and of course the plot over the years would be constant.
It seems that trees have a constant temperature of 20.something in a similar manner, ie biologically controlling temperature of their growth.
I think it is important that the measurements of the link are repeated and verified by independent experiments. That would end the trees as thermometers attempts.
I can only think of margarine and eggs. Margarine was the healthy substitute for butter until trans fat was discovered. Eggs were the ‘unholy grail’ until it was discovered there was a ‘good’ cholesterol. As we learn more about climate, I feel the current hysteria may be viewed as quite silly.
anna v (12:43:18) :
It is a bit more complicated, because trees are found in many species, but that is a durned good point if a Wal-mart working, denier, beer-swilling redneck has not never seen one.
“The sceptical community should be using this data since it is far better than tree-rings …”
It’s still tree rings. Its just a different way of measuring them. All of the surface temp to biology issues still obtain, they just manifest a bit differently. The leaf temp vs air temp issue presented above is an example. Lack of temp info outside the growing season is another. Length of growing season is a related but different effect. This isnt like reading a geiger counter …
I believe that isotope analysis should be a more reliable tree thermometer than gross ring width. If it works in ice-cores, why not in trees especially when bound fast in carbohydrates. I think it would be best to have all tree-samples prepared with a thorough vacuum drying process to remove all volatile contamination.
I’ve now had a chance to read the Lorrey et al (2007) paper referred to above:
Speleothem stable isotope records interpreted within a multi-proxy framework and implications for New Zealand palaeoclimate reconstruction. Quarternary International, V187, 52-75.
I posed 4 questions above and can answer those now:
“1) Has the empirical relationship between cave temperature and oxygen isotope composition of speleothems for the New Zealand caves been determined.”
No. An empirical relationship between cave temperature and speleothem oxygen isotope composition has not been determined. Thus it is not possible to estimate temperatures other than by using qualitative terms such as warmer, colder etc.
“2) How does this relationship compare with what we know of equilibrium oxygen isotope partitioning between water and carbonate? If it is different it indicates that there is disequilibrium in the system that may vary over time.”
It’s not possible to determine if these speleothems are in isotopic equilibrium or not.
“3) Has the record been subject to testing by use of a calibration period and then a period in which the temperature is predicted and compared to actual temperatures?”
No
4) Are their truly independent and accurate age models for each of the proxies. Perhaps growth layer counts, tree ring chronology, U-series and 14C dating that allow cross correlation of the records? If not then we are ‘wiggle matching’.
Yes
So in summary this paper does provide evidence of a mediaeval warm period, followed by a cooler, wetter period and finally a degree of modern warming.
It is not possible to use the isotope data to estimate temperatures. Indeed the comparison with temperatures in this multi-proxy study is with temperatures estimated from tree rings.
Splice (04:16:10) :
“I’ve now had a chance to read the Lorrey et al (2007) paper referred to above:
…
It is not possible to use the isotope data to estimate temperatures. Indeed the comparison with temperatures in this multi-proxy study is with temperatures estimated from tree rings.”
I’ll note that he seems to have intentionally excluded data later than the early 19th century. I have not discussed this with him so I don’t know why he did that, or whether the speleo source had such data. I suspect, however, that if it was intentional it was done so as to get the paper published by the peerruhveewdliturchur by not directly challenging any modern claims of warming by the Team relative to the actual temperature record, which would obviously raise flags and cause the Hockey Team as evinced in the CRUtape Letters, to go into supression mode. Now the paper is part of the scientific record and helped the MWP and LIA become reestablished as being climate phenomena that existed, and did so beyond northern europe, despite Team claims to the contrary.
Again, this is my personal speculation only.
I suspect that there needs to be some modern standards set as to what isotope levels correlate with what temperatures, if that is possible. O18 should behave the same in response to temperature and altitude around the world, as well as to carbon levels, which is also in the record. The very fact that O18 and C14 levels vary out of synch with each other should be evidence enough that carbon levels, which are a proxy for precipitation, not temperature, do not determine temperature.
Mike, I think Andrew’s paper is very good and an excellent addition to the palaeoclimate science. I hope that he didn’t feel he had to exclude modern data to get the paper published. If so then this is a terrible indictment on science in the present day.
With regard to your last point about 18O levels and temperature. This is not an easy question to resolve. It is not the 18O concentrations in a sample per se that allows us to estimate growth temperatures of speleothems. It is the partitioning of 18O between the parent drip water and the carbonate of the speleothem. Thus to determine temperatures one needs to know the isotope composition of the drip water from which the speleothem grew and the isotope composition of the carbonate. One then needs to prove isotope equilibrium between the speleothem and the water.
To recover the isotope composition of carbonate and its parent drip water is not an easy measurement and has only been successfully done in a handful of studies.
I’m sure if you talk with Andrew about this he can give you the full low down on isotope thermometry and the problems inherent with speleothems. It’s a fascinating problem and one that keeps many scientists occupied full time!
I do agree with you, and all others here that isotopes are probably our best route to temperature estimates but one needs to be aware of the limitations as well as the promise.
There are some very exciting developments in what is called ‘clumped’ isotope thermometry and I predict that over the next 2 to 3 years a number of papers will be published showing robust temperature estimates for speleothem growth. These will have precisions of about +/- 2 degrees C. Such precisions are only really useful for studying major temperature transitions such as from glacial maximum to interglacial optimum. For studies of climate over the last millenium the precision will need to improve to be better than sub degree, and ideally on the order of 0.2 degrees. This is some way off and depends on improvements in measurement technology, either mass spectrometrically or spectroscopically. However, I do believe such improvements are possible.
Obama can sign up for whatever he wants but treaties require a two thirds majority in the Senate and he will never be able to muster the votes. He’s going to backdoor compliance with the treaty by getting the EPA to issue CO2 regs that will produce the same effect.
With CO2 defined as a pollutant, your every breathe will be regulated as well as your lawn mower, barbeque, fireplace, stove, furnace, car, boat etc. Expect the prices on all of those to go up to as the EPA imposes a carbon tax and then the money will get shipped over to Somalia to help fund ship hijacking or China to fund their replacment of our remaining industries there.
“I believe that isotope analysis should be a more reliable tree thermometer than gross ring width. If it works in ice-cores, why not in trees especially when bound fast in carbohydrates.”
Gross tree ring width is carbohydrates. The biological processes of wood production, and their inherent confounding of temp signal, are still depended upon. Ring by ring or atom by atom, it is still a tree. Rushing from one shiny new metric to another doesnt necessarily gain anything. It might, but only time and testing will tell.