Volcanic eruptions emerge as lead cause for Little Ice Age

Sequences of explosive volcanic eruptions in the tropics were the likely trigger for the Little Ice Age, according to a new study.

The research attempts to answer two longstanding questions swirling around the roughly 400-year span of slightly cooler-than normal temperatures: Exactly when did it begin? And what was its initial trigger?

Previous estimates for the onset of the Little Ice Age range from as early as the late 1200s to as late as the 1500s, the research team notes. Globally, temperatures averaged a modest 0.6 degrees Celsius, or about 1 degree Fahrenheit cooler than usual.

But regionally, cooling could be profound. Glaciers in the Alps grew, bulldozing mountain villages. In Europe, the growing season became shorter, with spring and summers often cold and wet, triggering famines. In China, provinces that for centuries had produced bountiful citrus harvests no longer could provide them. With an additional climate-cooling blast from Mt. Tambora in Indonesia in 1815, North America and Europe experienced the year without summer in 1816.

Researchers have proposed a range of possible causes for the Little Ice Age – from decline in the sun's output, volcanic activity, some combination of the two, or some form of natural variability within the climate system.

The problem either with a decline in the sun's output, which happened during this period, or volcanism is that neither was powerful enough on its own to account for the cooling, says Gifford Miller, a climate scientist at the University of Colorado's Institute of Arctic and Alpine research at the University of Colorado in Boulder.

Recent research on solar activity has indicated the sun didn't dim as deeply as earlier research suggested. And volcanic activity typically affects climate only for a few years after an explosive eruption. It does this by hurling vast amounts of sulfur dioxide into the stratosphere, where the SO2 particles reflect sunlight back into space. Those particles eventually drift back into the lower atmosphere, where they get washed out of the sky during storms.

Instead, Dr. Miller's team proposed that the initial trigger involved several major volcano eruptions occurring within about a decade of one another, followed by another set roughly 150 years later, which intensified the cooling.

The cooling effect of those eruptions, the team posits, probably triggered the growth of Arctic summer sea-ice cover and changes in North Atlantic ocean circulation. These changes reinforced the cooling trend, turning what might have been relatively short cool periods into a centuries-long chill.

The Little Ice Age is the coldest century-scale climate swing in the Northern Hemisphere in the past 8,000 years, Miller says. “Our work says: Here is a potential way that you can explain it.”

The Little Ice Age began suddenly between 1275 and 1300 AD, the team estimates. It bases its estimate on vegetation samples gathered from Canada's Baffin Island during field trips between 2005 and 2010. The location once was covered with ice but lost it to global warming. Because the ice was on relatively flat ground, it preserved the vegetation it buried, without grinding it to powder. Using carbon-14 dating, the team analyzed some 147 samples and found two sudden periods of intense die-offs that took place when the ice cap experienced growth spurts. One occurred between 1275 and 1300, and the other between 1430 and 1450.

The team gathered additional evidence from sediment cores taken from a glacial lake in central Iceland. The team found two bands of sediment that quickly thickened: one dated to the 1300s, and the other to the 1400s. The sediment entered the lake when an ice cap that feeds the lake expanded, increasing the amount of erosion created.

The sediment confirmed the onset of cooling for the periods in question. And it established the cooling as hemisphere-wide.

The team then turned to climate models to figure out how eruptions could effect such a change. With successive eruptions over the course of a decade or so, enough SO2 would get kicked into the stratosphere to cool the temperatures over a longer period than a one-off blast. Modeling indicated that this would chill the Arctic Ocean, allowing sea-ice to thicken in winter. Thicker ice is more likely to survive the summer melt season, reflecting sunlight back into space and preventing the ocean from warming as much as it does when sea-ice cover shrinks. Because the ocean heats and cools very slowly, once a cooling trend sets in, it's hard to reverse.

Even so, more ice also would be available to move into the North Atlantic during the spring and summer. The addition of cold fresh water would have chilled and slowed ocean currents that otherwise would warm eastern North America and Europe in winter.

The new research is consistent with other recent studies of the Little Ice Age, notes Michael Mann, a climate scientist at Pennsylvania State University, in State College. The global cooling was attributable to volcanic activity. But the regional distribution and intensity of cold temperatures owed much to the sun's weakened output, he adds in an e-mail.

The sun produces most of its radiation as ultraviolet light, which sees the largest swings in intensity with the rise and fall of the sunspot cycle. Most of the ultraviolet light is intercepted by ozone in the stratosphere; the UV light builds ozone molecules there. Ozone is a greenhouse gas. When solar radiation declines, so does stratospheric ozone, cooling the stratosphere. This changes wind patterns there. These changes eventually work their way into wind patterns in the lower atmosphere, as well.

In a paper in the journal Science in 2009, Dr. Mann and colleagues showed how these shifting patterns during the Little Ice Age – when the sun entered a pronged period of weakened radiation – would have thrown natural climate swings into one phase more often than the other.

For instance, regional temperature reconstructions from the period indicated that La Niña was appearing more frequently and persisting longer than might otherwise be the case – one sign a weaker sun was shuffling warm and cold regions, rather than contributing significantly to a worldwide cooling.

Indeed, as if to underscore a regional, rather than global, effect from reduced solar radiation, Miller's team didn't vary the sun's output in the modeling phase of its study; they held it constant – implying that the Little Ice Age would have occurred in some form even with the sun's weaker output.