A new study has recently been published that looks at the ecology of bristlecone pine growth at Sheep Mountain, and the tree ring signal those trees produce, at high altitudes in the Southwestern US. This is important because tree rings are an often used proxyindicator for reconstructing past climates. Those who keep track of the paleoclimate research will recall, for example, that tree rings were one of the proxyindicators used by Michael Mann and his team in constructing the famous “Hockey Stick” graph showing a dramatic increase in the Earth’s temperature since the onset of industrial times, when the instrumental record (of the last century plus) is put in context of previous centuries. Since then, tree rings have played an important role in past climate reconstruction.

I asked Malcolm Hughes, who was an author on that earlier hockey stick work and an author on the recent study discussed here, how his recent work informs us of the validity of Mann et al, especially in relation to bristlecone pine data used in that early seminal study. He said, “Back in 1999 we (Mann et al) made the best available choices with the information and data we had. Now, more than 15 years later, with a Bristlecone Pine record that extends back 5000 years, the original results hold up remarkably well.”

So, now, let’s discuss tree rings a bit and then see what the new paper offers.

Tree Rings And Past Climate

Past climates can be reconstructed by using proxyindicators of past conditions, much like more recent climate can be characterized by using instruments (thermometers, etc.) and data form satellites. One of these proxyindicators is tree ring width. Trees with seasonal growth may produce woody tissue at higher or lower rates depending on a limiting factor, such as available water or temperature. An individual tree may be limited by water availability if it is growing in a certain microhabitat, but a tree of the same species may be limited by temperature if it is growing in a different microhabitat. It is even possible for one limiting factor to control growth for a period of time, then, as conditions change, a different limiting factor takes over. Saplings and small trees growing in a forest may be limited by light, and later, as they grow tall enough (or gaps in the forest canopy open), they may be limited by moisture or temperature.

The tree rings from certain sites seem to properly reflect temperature variability up until around 1960, and after that, the usability of the signal from that proxy can be reduced. This pattern has been noticed in a number of different tree ring records; the phenomenon is widespread enough that it has a name. It is called the “divergence problem.”

Several explanations have been offered to explain the divergence problem, and there is a fair amount of literature on it. It is possible that the post industrial increase in atmospheric CO2 has affected tree growth (acting, essentially as airborne fertilizer) in such a way that the tree ring widths no longer reliably indicate temperature. Changes in the pattern of snow melt at high altitudes, which is where the temperature-sensitive trees are generally found, may affect growth patterns. Changes in minimum or maximum temperature distributions could be a cause. It is also likely that the amount of atmospheric dust (aerosols) has an impact on tree growth, so recent pollution could be a factor. The anomaly could also be an artifact of sampling large trees, in order to sample the longest time periods, if older trees have different growth patterns. In the case of one of the key proxy tree species, bristlecone pines, divergence may have to do with the fact that there are two different growth patterns at the tree’s surface, referred to as strip-bark and whole-bark.

In 2009, Matthew Salzar, Malcolm Hughes, Andrew Bunn, and Kurt Kipfmueller published a paper that looked at a possible divergence problem in bristlecone pines at three sites the American Great Basin. They looked at trees at the upper limit of elevation and found that “…ring growth in the second half of the 20th century … was greater than during any other 50-year period in the last 3,700 years.” This confirms that whatever the cause of divergence is, it is likely something going on during the most recent decades, which strongly suggests a cause related to Industrial Era pollution or warming. They were able to rule out changes in tree growth patterns and fertilization by added atmospheric CO2. They note that “[t]he growth surge has occurred only in a limited elevational band within ?150 m of upper treeline, regardless of treeline elevation,” and concluded that “[i]ncreasing temperature at high elevations is likely a prominent factor in the modern unprecedented level of growth for Pinus longaeva at these sites.”

A New Study On How Tree Rings Work

More recently an overlapping set of authors have published an important study (“Changing climate response in near-treeline bristlecone pine with elevation and aspect”) that looks at this problem in more detail. From the abstract:

In the White Mountains of California, eight bristlecone pine (Pinus longaeva) tree-ring width chronologies were developed from trees at upper treeline and just below upper treeline along North- and South-facing elevational transects from treeline to ~90 m below. There is evidence for a climate-response threshold between approximately 60–80 vertical m below treeline, above which trees have shown a positive growth-response to temperature and below which they do not. Chronologies from 80 m or more below treeline show a change in climate response and do not correlate strongly with temperature-sensitive chronologies developed from trees growing at upper treeline. Rather, they more closely resemble lower elevation precipitation-sensitive chronologies. At the highest sites, trees on South-facing slopes grow faster than trees on North-facing slopes. High growth rates in the treeline South-facing trees have declined since the mid-1990s. This suggests the possibility that the climate-response of the highest South-facing trees may have changed and that temperature may no longer be the main limiting factor for growth on the South aspect. These results indicate that increasing warmth may lead to a divergence between tree growth and temperature at previously temperature-limited sites.

Generally, trees from higher latitudes (farther north) and higher altitudes both make better temperature proxies, because the two factors (altitude and latitude) both have temperature as a common thread. You learned this in your middle school Earth Science class. Altitude and latitude mimic each other to a large degree, which is why mountain glaciers can be found on the equator (or, at least, were found there before global warming mostly melted them away). But this new research also shows that topographical position in relation to the sun (south facing vs. north facing) further modify the microenvironment the trees are growing in.

Three earlier studies by Salzer and others (in 2009, 2013, and 2014) show is that the 20th century growth increase at the very highest elevations is temperature related, so in that sense, there has not been a divergence problem in the bristlecone pine, although one may now be emerging on the south- facing slopes, but not on the north-facing slopes near treeline.

The importance of microenvironment is well illustrated in this figure:

This shows the tree ring values for two sampled sites from 1980-2009. Until the mid 1990s, the tree ring values correlate tightly. After this point in time, however, they demonstrate the divergence problem among the samples, in this case, caused primarily by the specific direction the locations are facing (north vs. south).

The most important conclusion of this paper (which is not a reconstruction of past climate, but rather, a study of the ecology of tree ring growth) has to do with how tree ring data are assembled and used. Multiple tree ring samples may be taken from a given site, with the assumption that that site has similar conditions for all the sampled trees, so the assembled and combined tree ring data will have a similar climate related signal. (Note that there is a fair amount of internal variability within a given tree that is hopefully erased when more than one sample from a given site is used). Salzer et al show that at high elevations bristlecone pines can vary from each other considerably if they are sampled form elevations that differ in several tens of meters, or in the aspect (direction) of the slope they grew on, and that this sensitivity is tied to the treeline at on a given slope. And, of course, the position of the samples (as noted above) affects the degree to which tree ring data reflect temperature as opposed to other factors.

We have shown that approximately 60–80 m of vertical elevation can be sufficient to create a change in the climate response of bristlecone pine. Trees below this elevation are not as effective temperature recorders as trees at treeline. Such fine-scale sensitivity, if present at other treeline sites around the world, would have important implications for chronology development and inferences of past climate variability. Treeline site chronologies should be constructed with this vertical heterogeneity in mind. Samples from upper treeline and from trees below treeline should not be mixed to avoid a ‘diluted’ or ‘mixed-signal’ site chronology, particularly at treeline sites that occur in relatively dry environments such as the White Mountains of California. Similarly, treeline samples from differing aspects should not be mixed to avoid problems and uncertainties related to potential ‘divergences’ and to ‘dilution’. Interpretations of existing bristlecone chronologies need to take this into account, particularly when these ring width chronologies are used in climate reconstructions.

I asked Malcolm Hughes and Matt Salzer, two of the study authors, how to best characterize this study. They told me that this paper is, for the most part, “…an ecological study. There are no climate reconstructions, rather mostly comparisons of growth from trees growing in different spots on the landscape. There are paleoclimate implications. We still find temperature sensitive bristlecone pine trees at upper treeline; they simply don’t extend down the mountain as far as we used to think. In addition, some of the treeline south-facing trees seem to be less influenced by temperature in recent years than they used to be. The location of the trees, and understanding what environmental variable is limiting growth at that location, is the key to developing accurate paleoclimatic reconstructions from tree rings. Science is a continuous process of improvement.”

Earlier research by an overlapping team also looked at topography. In “Topographically modified tree-ring chronologies as a potential means to improve paleoclimate inference” by Andrew Bunn, Malcolm Huges, and Matthew Salzer (2011) it is noted that

…a mean ring-width chronology from a particular site may be composed of trees from highly varied topographic positions. Such a “topographically-mixed” chronology can be confounded in terms of its climate signal. For example, ring widths of trees that are primarily recording summer temperature might be averaged with ring widths of trees that are primarily precipitation recorders.

That paper details how researchers can use topographic setting to separate different growth series to produce a cleaner sample for developing a temperature proxy. Like the more recent paper discussed here, Bunn et al is an effort to improve the methodology of using tree rings as a proxy.

Science Denialists Can’t See The Forest Through The Tree Rings

Even as the tree-ring proxyindicator expands in size (more samples) and is better understood (from the above mentioned studies) climate science denialists remain entrenched with their assertion that tree rings are bad proxies, or are being used incorrectly. These criticisms are not legitimate critiques of the science, but rather, combine obfuscation and misinformation to muddle and confound thinking about tree rings. An early example of this comes from the kerfuffle known at “climategate” in which electronic communications among climate scientist were stolen and mined for decontextualized quotes that could be used to lie about the science itself and the motivations and activities of the scientists who developed the Hockey Stick curve. Michael Mann chronicles these events in his book “The Hockey Stick and the Climate Wars: Dispatches from the front lines.”

One e-mail Phil Jones of CRU sent to my coauthors and me in early 1999 has received more attention than any other. In it, Jones both made reference to “Mike’s Nature trick” and used the phrase “to hide the decline” in describing a figure … comparing different proxy temperature reconstructions. Here was the smoking gun, climate change deniers clamored. Climate scientists had finally been caught cooking the books: They were using “a trick to hide the decline in global temperatures,” a nefarious plot to hide the fact the globe was in fact cooling, not warming! …

The full quotation from Jones’s e-mail was …, “I’ve just completed Mike’s Nature trick of adding in the real temps to each series for the last 20 years (i.e. from 1981 onwards) and from 1961 for Keith’s to hide the decline.” Only by omitting the twenty-three words in between “trick” and “hide the decline” were change deniers able to fabricate the claim of a supposed “trick to hide the decline.” No such phrase was used in the e-mail nor in any of the stolen e-mails for that matter. Indeed, “Mike’s Nature trick” and “hide the decline” had nothing to do with each other. In reality, neither “trick” nor “hide the decline” was referring to recent warming, but rather the far more mundane issue of how to compare proxy and instrumental temperature records. Jones was using the word trick [to refer to] to an entirely legitimate plotting device for comparing two datasets on a single graph…

The reconstruction by Briffa, (see K. R. Briffa, F. H. Schweingruber, P. D. Jones, T. J. Osborn, S. G. Shiyatov, and E. A. Vaganov, “Reduced Sensitivity of Recent Tree-Growth to Temperature at High Northern Latitudes,” Nature, 391 (1998): 678–682) in particular …

…was susceptible to the so-called divergence problem, a problem that primarily afflicts tree ring density data from higher latitudes. These data show an enigmatic decline in their response to warming temperatures after roughly 1960, … [Jones] was simply referring to something Briffa and coauthors had themselves cautioned in their original 1998 publication: that their tree ring density data should not be used to infer temperatures after 1960 because they were compromised by the divergence problem. Jones thus chose not to display the Briffa et al. series after 1960 in his plot, “hiding” data known to be faulty and misleading—again, entirely appropriate. … Individuals such as S. Fred Singer have … tried to tar my coauthors and me with “hide the decline” by conflating the divergence problem that plagued the Briffa et al. tree ring density reconstruction with entirely unrelated aspects of the hockey stick.

Note that there wasn’t a “divergence problem” in Mann et al in the sense of Briffa et al. Mann et al match the observational record very well through 1980, which is the end of the calibration interval (owing to the fact that many proxies drop out after 1980). This is something else the deniers tend to get wrong; they try to conflate the Briffa et al post-1960 divergence problem Mann et al’s hockey stick work. There is no such issue with that work, in that there was no detectable divergence through the end of the calibration interval.

Related to this, there was a correction of the Bristlecone Pine data for inflated 20th century increase (which was attributed to CO2 fertilization at the time) in MBH99. So we actually applied a downward correction of the trend in those data. McIntyre doesn’t want people to know that. So need to make sure that is crystal clear.
More recently, climate science denialist JoNova took the new paper by Salzer et al to task using equally mind numbing arguments. JoNova notes that “after decades of studying 800 year old tree rings, someone has finally found some trees living as long ago as 2005. These rarest-of-rare tree rings have been difficult to find … The US government may have spent $30 billion on climate research, but that apparently wasn’t enough to find trees on SheepMountain living between the vast treeless years of 1980 to now.”

I’m sure the scientists involved in tree ring research would like to know where their $30 billion dollars went, but that’s another story. I asked Malcolm Hughes about JoNova’s implication that there has been next to zero research on or with bristlecone pines over these many years. He said, “This post makes a big deal about the lack of updating of bristlecone pine chronologies since 1980. This is simply wrong. She fails to acknowledge that in 2009 we published on bristlecone pine growth rates in PNAS (Proceedings of the National Academy of Sciences) and put tree-ring data from Sheep Mountain out to the year 2005 in a publicly accessible archive.”

JoNova also implies that the lack of tree ring proxy use for periods after 1980 is somehow suspicious, but as detailed at length above, the divergence problem is, well, a problem. Also, further work such as that reported here is likely to revive some of that data and allow it to be used, eventually. At the very least, future work with high altitude/latitude tree ring data will be improved by these methodological and ecological studies.

Climate science denialist Steve McIntyre has also weighed in on Salzer et all’s research. His post is truly mind numbing, as he treats Salzer et al as a climate reconstruction paper, and critiques it as such, but the paper examines the methodology of tree ring proxy use and the ecology of tree rings. McIntyre shows the same figure I show above (Figure 5 from that paper) and critiques the researchers for failing to integrate that figure or its data with Mann et al’s climate reconstructions. But they shouldn’t have. That is not what the paper is about. Another very recent paper by the same team is in fact a climate reconstruction study (published in Climate Dynamics) but McIntyre manages to ignore that.

Science writer and failed banker Matt Ridley has also applied his abysmal understanding of paleoclimate science with a critique of some of this research.

I’m sure you find these esoteric details of tree ring chronologies fascinating, but there is a point that needs to be made that is even more interesting.

Ever since Mann et al published the famous “hockey stick graph” those in the business of denying or (inadequately) refuting the growing consensus of climate science have made much hay out of both the divergence problem and a sense of suspicion of the tree ring record. For examples of this, read through the comments on this post. If you read comments by those who seem bent on the idea of refuting the reality of global warming, you’ll get the impression that there are only a few tree ring chronologies, we have no idea how they work, they don’t work, that climate scientists pretend they stop working at some point when really they are working but show cooling (which we know didn’t happen because we have thermometers) and, generally, that tree ring science is some sort of exercise in voodoo.

What has really happened, however, is that tree ring data have behaved pretty much as any other paleoclimate proxy behaves. There are conditions under which any given proxy works, and there are conditions under which the proxy can’t be trusted and should not be used. Decades ago, when the Hockey Stick was first formulated, the loss of signal in the tree ring record was somewhat mysterious, though numerous good ideas explaining it were out there. Subsequently, there has been a considerable amount of research adding new tree ring data, and some of that research has focused on teasing out the methodological and ecological details of this particular proxy. Interestingly, the last 10 years or so of tree ring research has failed to force the conclusion that tree rings are not good sources of past temperature data; the divergence problem is replicated in other records showing it again to be a recent phenomenon; change in regional (and global) temperature is increasingly implicated as the cause of the divergence problem; and much finer details of how this all works, at the scale of tens of meters elevation and at the level of details of topography have been worked out.

References and related items

• Briffa, K.R., F. H. Schweingruber, P. D. Jones, T. J. Osborn, S. G. Shiyatov, and E. A. Vaganov, “Reduced Sensitivity of Recent Tree-Growth to Temperature at High Northern Latitudes,” Nature, 391 (1998): 678–682.

• Laden, Greg. 2014. Matt Ridley, Anti-Science Writer, Climate Science Denialist blog post

• Mann, Michael E. (2012-01-24). The Hockey Stick and the Climate Wars: Dispatches from the Front Lines (p. 263). Columbia University Press. Kindle Edition.

• Mann, Michael E. The Hockey Stick and the Climate Wars: Dispatches from the front lines.

• Mann, Michael E., Raymond Bradley, Malcolm Hughes. 1998. Global-scale temperature patterns and climate forcing over the last six centuries. Nature 392.

• Matthew W. Salzer, Malcolm K. Hughes, Andrew G. Bunn, and Kurt F. Kipfmueller. 2009. Recent unprecedented tree-ring growth in bristlecone pine at the highest elevations and possible causes. PNAS 2009 106 (48) 20348-20353; published ahead of print November 16, 2009, doi:10.1073/pnas.0903029106

• McIntyre, Steve, 2014, “Sheep Mountain Update” blog post

• Ridley, Matt, 2014, “Science is being corrupted by political bias” blog post.

• Salzer, Matthew, Evan Larson, Andrew Bunn, and Malcolm Hughes. 2014. Changing climate response in near-treeline bristlecone pine with elevation and aspect. Environmental Research Letters 9(11)

• Skeptical Science. “Clearing up misconceptions regarding ‘hide the decline’” web page
  Wikipedia. The Divergence Problem.