Following is a brief explanation of the importance of tree-ring data from Ireland.

Why does Earth have a climate?

Climate exists because of heat from the sun. Heat from the sun, i.e. sunlight, is most intense in the tropics. The climate system redistributes that heat: it continually transports heat from the tropics towards the polar regions (some heat is also converted to kinetic energy).

The climate system has two main mechanisms for transporting heat: air (the atmosphere) and water (the oceans). The atmosphere can move quickly, i.e. winds. The oceans move more slowly. The oceans, though, hold much more heat than the atmosphere; e.g. the top 15 cm (6 inches) of ocean waters contain more heat than the entire atmosphere. Over time, say years or decades, ocean heat transport can significantly change. Those changes can have an effect on global heat redistribution, that is, on climate.

Deep water
The circulation of the oceans is driven by two main factors: winds and the formation of deep water (“deep water” is ocean water that is more than about 1500 meters below the surface). A figure illustrating where deep water is formed, and how deep water circulates, is shown below.

Thermohaline circulation

Schematic figure showing the three main areas where deep water is formed: in the Labrador Sea, in the Greenland/Iceland/Norwegian Seas, and in the Antarctic Weddell Sea. Deep water is formed when water cools and sinks, thereby releasing heat to the air. After sinking, the water circulates around the world ocean (eventually rising up in the Indian and North Pacific oceans).

Deep water is formed in the North Atlantic as follows. Warm surface water flows from the tropical South Atlantic, through the Caribbean, and is then transported, via the Gulf Stream and North Atlantic Drift, to the northernmost North Atlantic. En route, there is substantial evaporation; so the water becomes more saline, hence more dense. The water also cools, further increasing its density. Water that reaches the Nordic (Greenland, Iceland, Norwegian) and Labrador seas is additionally densened in winter by salt from sea-ice growth (ice has no salt), and by more cooling. Some water is densened enough to cause it to sink. Water that sinks sufficiently forms deep water.

Deep water formed in the North Atlantic is the dominant source of deep water for the world. Variability in the formation of North Atlantic deep water will lead to climatic change downwind from the northern North Atlantic; it will also influence other areas of the world ocean.

Irish tree rings
Ireland is immediately downwind from the North Atlantic Drift. Changes in the strength and temperature of the Drift—which correlate with changes in the formation of deep water—will strongly affect the climate in Ireland: temperature, precipitation, etc. Such climatic effects will naturally influence trees growing in Ireland. As discussed elsewhere, each ring of a tree indicates what the climate was like during the year in which the ring grew. Thus Irish tree rings would seem to contain annual data on the global climate system.

There are currently tree-ring measurements from Ireland covering the past 7000 years or so. Those measurements are thus valuable for the study of Earth's climate.

I first found out about the foregoing in 1997, when I attended a conference at Cambridge University (this was the conference that led to the founding of the Cambridge Conference Network). One of the speakers at the conference was Michael G.L. Baillie, who is a professor of tree-ring studies at Queen's University Belfast, in Northern Ireland; Baillie is also the researcher who has gathered most of the tree-ring samples in Ireland. During his talk at the conference, he discussed the above.

Empirical tests
Baillie has thus far refused to release most of the data (for discussion of this, see here). A small portion of the data has been released, though; I used some of that to do a partial test of the foregoing. The tree-ring data for the test was from a site in western Ireland, Garryland Wood. I compared that tree-ring data with the average annual temperatures of the Northern Hemisphere. (I ignored the Southern Hemisphere; although the Southern Hemisphere is affected by the climate off Ireland, the effects do not occur in the same year—because oceanic transport and interhemispheric mixing of the atmosphere are too slow.)

The records of annual temperatures were obtained from NASA (GISTEMP). The comparison spanned years 1880 (the first year of the temperature records) through 1997 (the last year of the tree-ring data). The correlation between the temperatures and the tree rings is substantial: 0.41.

The growing season of the Irish trees is about March–October. The trees, though, can be influenced by the climate outside the growing season, e.g. more snow during the prior winter might enhance growth. Furthermore, the growth of a tree during any one year inherently affects how well the tree grows during the subsequent year. Hence, when using tree rings to estimate the temperature in a given year, we should consider not only the ring that grew in that year, but also the ring that grew one year earlier and the ring that grew one year later.

There are many ways to consider the earlier and later rings. As a simple example, we could just add together the data from three consecutive years. Doing the additions gives a new series, whose correlation with the hemispheric temperatures is 0.52. Obviously more sophisticated methods could be used to determine the new series, which would increase the correlation further.

Additional issues
The above correlations are based on the average of data from trees at a single site in Ireland. A problem that can occur when using trees at a single site is that the trees might have been affected by some local event, such as an insect attack, or young trees breaking through the forest canopy and growing with little constraint. Those problems could be at least partially addressed by considering the individual trees at the site, rather than the average for the site, and also by considering trees at other sites in the British Isles. Doing so would presumably lead to additional increases in the correlation.

Another issue is to determine what climatic variables the trees are responding to. The trees obviously cannot be responding directly to the temperature of the whole hemisphere. Could they be responding largely to local temperatures? Local temperature data is available beginning in 1901. This data shows that the tree rings have a correlation with local temperatures that is much less than the rings have with hemispheric temperatures (explaining only about half as much of the variation). So what climatic variables explain the high correlation between the tree rings and hemispheric temperatures?

The regional climatology suggests that the trees are responding to the energy flux from the adjacent ocean to the atmosphere. This energy flux will have a substantial influence on hemispheric temperatures. The energy can be thermal (heat) or kinetic (wind). And it is associated with a variety of climatic effects that influence tree growth (in addition to temperature).

As an example, for air at a fixed temperature, an increase in heat content generally implies an increase in humidity, which can affect tree growth. Increased heat and wind would also tend to increase cloudiness, which again can affect tree growth. Increased heat and wind would additionally increase thunderstorms, which bring much precipitation during a short time and typically in large drops (which are more likely than small drops to break through the forest canopy and undergrowth)—although there is ample precipitation in western Ireland; so it seems unlikely that tree growth is substantively limited by moisture. It has also been suggested (by L. Pearson) that increased lightning, from the thunderstorms, could affect trees, e.g. through the release of nitrogen. Modeling the climatic factors affecting tree growth is extremely difficult, but the Garryland Wood trees do seem to be responding, somehow, to the energy flux.

Finally, although the Irish data seems quite good, it should obviously be combined with other data from different regions of the Northern Hemisphere. Using a small number of sites with high-quality data (e.g. a few sites downwind from each of the three oceans) might well yield a quality temperature reconstruction for the hemisphere, extending back millennia.

Relevance for global-warming studies
The biggest concern with global warming is, arguably, that warming will itself cause further warming. For example, the polar ice caps reflect sunlight back into space (thereby cooling Earth); if the caps shrink, due to warming, then they will reflect less sunlight, and so Earth will warm further. It is claimed to be possible that Earth warms so much that it reaches what is called a “tipping point”, where the global climate system is seriously and permanently disrupted—like when a glass of water has been tipped over, and the water cannot realistically be put back into the glass.

There is much scientific debate over how much Earth has to warm before reaching a tipping point. No one knows for sure. About a thousand years ago, though, there was a time known as the “Medieval Warm Period”, when Earth (or at least the Northern Hemisphere) appears to have been unusually warm. It is not currently known just how warm the Medieval Warm Period was. Clearly, though, the warmth then was below the tipping point, because Earth's climate continued without problem.

Suppose that during the Medieval Warm Period, Earth was 1°C warmer than today. That would imply that the tipping point is more than 1°C higher than today's temperature. For Earth's temperature to increase by 1°C might take a century (at the rate of increase believed to be currently underway). So we would not have to be concerned about an imminent disruption of the climate system.

Finding out how warm the Medieval Warm Period was is thus of enormous importance for the study of global warming. It is one of the most important questions for science today. And the Irish tree-ring data can provide a substantial portion of the solution.

See also

Statistical errors in the “hockey stick” graph
Tree-ring data at Queen's University Belfast

Barry R.G., Chorley R.J. (2010), Atmosphere, Weather and Climate (London: Routledge).

Briffa K.R., Jones P.D., Vogel R.B., Schweingruber F.H., Baillie M.G.L., Shiyatov S.G., Vaganov E.A. (1999), “European tree rings and climate in the 16th century”, Climatic Change, 43: 151–168. doi: 10.1023/A:1005529830082.

Fowler A., Bridge M. (2009), “Climate inferences from oak archaeological samples”, Eurodendro 2009 [conference presentation].

García-Suárez A.M., Butler C.J., Baillie M.G.L. (2009), “Climate signal in tree-ring chronologies in a temperate climate”, Dendrochronologia, 27: 183–198. doi: 10.1016/j.dendro.2009.05.003.

Kuhlbrodt T., Griesel A. , Montoya M. , Levermann A. , Hofmann M. , Rahmstorf S. (2007), “On the driving processes of the Atlantic meridional overturning circulation”, Reviews of Geophysics, 45, RG2001. doi: 10.1029/2004RG000166.

Douglas J. Keenan