The sudden chills violent volcano eruptions cast over the world centuries ago effectively erased themselves from the historical climate record produced by examining tree-rings. So suggests a team led by Michael Mann from Pennsylvania State University, who famously used 1,000 years of tree-ring measurements in the “hockey stick” graph showing how unusual today’s temperatures are. Michael warns the skipped years could affect scientists’ estimates of how much the world warms in response to greenhouse gases in the atmosphere, known as its climate sensitivity. But other than the volcano years, the scientist notes that tree-ring data is a remarkably accurate match with the climate models they used for comparison. “Interestingly, the effect has little influence on long-term trends, including conclusions about how previous temperatures compares to modern ones,” he told me. “Instead, it appears only to have implications for how strong past short-term cooling events were.”
A tree’s age can usually be told from the rings that form across its trunk representing each year’s growth. How thick each ring is shows how much the tree grew in the year in question, which is influenced by the temperatures that tree experienced. That means examining the thickness of rings in old trees can provide a way to tell temperatures back through history. Many challenges have already been overcome in turning this simple-sounding idea into a history of the world’s temperature, but Michael was still troubled by one particular detail.
Sleuthing through history
In 1258 AD a large volcano, whose location is unknown, exploded and threw masses of sulphur-rich gases into the atmosphere. Those gases form particles that reflect the energy normally reaching the Earth back out into space, which cooled the world and reportedly led to crop damage and famines. Similar but smaller cooling events have been seen more recently, like the eruption of Mount Pinatubo in 1991. But the 1258 eruption’s effects don’t show up as strongly as they should on the historical tree-ring graph, and appear around five years late. That prompted Michael to ask: “Why is the cooling of what we believe to be the most explosive tropical eruption of the past millennium – four times larger than Pinatubo – not shown in tree-ring reconstructions of past temperatures?”
To answer this, Michael, along with fellow Penn State meteorologist Jose Fuentes and Scott Rutherford from Roger Williams University in Rhode Island, compared tree-ring measurements with two climate models. In a paper published in research journal Nature Geoscience on Sunday they show how they fed estimates for the energy flowing into the atmosphere, known as radiative forcing, into these models. The models produced a temperature graph that is very similar to the tree-ring record back as far as 1850, including sudden cooling in years with notable volcano eruptions. But reaching back to 1200 there are three separate eruptions whose effects are greater and earlier in the models than tree rings show. In particular, the models see a sharp 2°C cooling for the 1258 eruption. By contrast, the maximum rapid cooling the tree rings suggest was around 1°C.
Michael said he, Jose and Scott then used more models to do “a bit of sleuthing” to understand why this should be. They found that because lots of different trees are used to get measurements, “noise” that could be caused by how temperature is deduced from ring measurements wasn’t a good explanation. Looking at how the trees respond to temperature biologically, by contrast, gave “a strikingly close reproduction” of the record actually seen, including the 1°C maximum cooling.
The success of the biological approach comes because there is typically a minimum temperature at which trees will grow. To ensure that temperature is the most important factor in how wide tree-rings used to construct historical records are, measurements are usually taken in areas that are near this minimum. The main place fitting this description is on the tree line beyond which forests stop but mountaintops continue upwards. But when temperatures fall too far there is no growth, and therefore no ring for that year.
“This would explain the delayed cooling seen in the tree-rings relative to the timing of the eruptions,” Michael said. “We didn’t honestly think that we would be able to match the actual tree-ring reconstruction so closely. We were pretty amazed that our hypothesis was as successful in explaining the observations as it appears to be. It paves the way to better incorporating these sorts of biases when reconstructing the past climate.” The close match between models and tree-ring measurements underlines how accurate both tools can be, he added.
While large volcano eruptions only occur occasionally, Michael points out that they are currently relevant because they help work out climate sensitivity. This is a measure of what temperature increase we can expect if we keep on putting greenhouse gases into the atmosphere. Climate sensitivity is normally given as the temperature increase a doubling of the concentration of CO2 in the atmosphere would cause. CO2 concentrations in the atmosphere were 280 parts per million (ppm) before we started burning fossil fuels on an industrial scale and is now 393 ppm. Scientists therefore haven’t had the chance to directly measure the effects of doubling CO2.
The energy that greenhouse gases trap in the atmosphere increases radiative forcing, and so radiative forcing can also be used to calculate climate sensitivity. To do this scientists must compare temperatures at current radiative forcing levels to those during periods when radiative forcing was low. The sulfate particles volcanoes eject into the atmosphere lower radiative forcing, making eruption years attractive for this purpose. At least one such comparison uses temperatures for volcanic eruption periods from tree ring data as its lower end. If those figures are artificially high, as Michael suggests, then it would in turn give a temperature rise for a doubling of CO2 that is too small. “We’re now pursuing the issue of how this effect may have influenced previous estimates of climate sensitivity,” he said.
- Michael has a new book about the attack he has come under because of his “hockey stick” graph available for preorder, entitled “The Hockey Stick and the Climate Wars”.