Ancient pollen trapped in Greenland ice uncovers changes in Canadian forests over 800 years

Published January in the journal Geophysical Research Letters, the research was led by Sandra Brugger, Ph.D., during her postdoctoral work with Joe McConnell, Ph.D., in DRI's Ice Core Lab. Although the lab has conducted many studies using cores of ice extracted from Arctic landscapes around the world -- using the chemical signals trapped in the ice and air bubbles to track changes in environmental and human history -- this is the first time that pollen was the center of focus. Through painstaking analysis of each pollen grain under the microscope, they found a record of forest changes spanning 850 years, covering both the onset of the Little Ice Age around 1400 and the arrival of European settlers and subsequent intensive logging practices around 1650. The information not only provides glimpses of the past, but can help establish baselines for evaluating changes in forest cover as the global climate warms.

"Our results are exciting because pollen that traveled thousands of miles can illuminate changes that happened on a large scale in those forests," said Brugger, who is now at the University of Basel.

Because pollen is relatively large compared to other atmospheric particles, few pollen grains travel across oceans and get preserved in Arctic ice sheets. In order to obtain enough pollen to piece together the story of historic forest changes, Brugger developed new methods that made the analysis possible, involving carefully evaporating water from the ice core to collect the miniscule grains of pollen. This study marks one of the first times that pollen has been extracted from polar ice cores to examine environmental change over time. Pollen in lake sediment has been used for the same purpose but is reflective of a local ecosystem immediately surrounding the lake, and does not capture changes in vegetation occurring on a region-wide scale.

"We also get a more precise chronology from the ice cores than from most lake sediments, because we can date them almost to the year," Brugger said. "This helps tie the information to, for example, European settlers coming to eastern North America, because we know from historic sources when they arrived and what they did to the landscape."

"Since Greenland is only sparsely vegetated at the margins of the ice sheet and so far from forested regions, we don't usually think about trying to measure pollen in Greenland ice cores," said Nathan Chellman, Ph.D, assistant research professor at DRI and co-author of the study. "It is pretty incredible to think that the history of North American conifer forests can be told by measuring the few pollen grains that manage to make it through the atmosphere to the ice sheet."

Joining DRI's Ice Core Lab to conduct the work allowed Brugger to utilize McConnell's decades of expertise developing and honing techniques for making precise chemical measurements of ice cores to pinpoint specific events in time, including plagues, volcanic eruptions, and the evolution of industrial society.

The ice core was collected from southern Greenland and held pollen primarily from conifer forests on the North Atlantic coast, with smaller amounts from northern tundra vegetation and European forests. Between approximately 1160 and 1400, a period known as the Medieval Warm Period, about 60% of the pollen records came from northern forest species such as pine, spruce, and fir. After the onset of the Little Ice Age circa 1400, the same species compose nearly 80% of the pollen record, demonstrating an expansion of these forests as the climate cooled. Because there is no evidence that the northernmost tree line shifted over this timeframe, the forest spread likely came from changes in forest density, pollen productivity, and a southward expansion into more temperate regions, the authors write.

Pollen from the same conifer species changed again between 1650 and 1760, declining to about 40% of the ice record, which is likely due to extensive logging of the region by European settlers during this time. The study authors ruled out increased forest fires as the cause of the decline by measuring fire indicators like charcoal and black carbon in the ice core. They also cross-referenced their findings with historical records showing that white pine forests were considered valuable timber, leading to many of the forests along the Miramichi, Saint John, and Ottawa rivers to be harvested by 1850.

Around 1760, chemical signatures of fossil fuel pollution began appearing in the ice record, marking the start of industrialization in eastern North America. This is around the time when the pollen record reflects changes in human activities more prominently than climatic influences, suggesting the start of a human-driven ecosystem in the region. Conifer forests then declined further at the beginning of the 20^th century, with continued logging and clearing of forests for farmland throughout eastern Canada. At the same time, pollen records show an increase in ragweed, a shrub known to proliferate in disturbed landscapes. The pollen record indicates that pine forest recovery started around 1950, when a decline in logging coincided with rural farm abandonment and a shift in climate.

"This clarity of the signal showing pine forest expansion at the beginning of the Little Ice Age, and then this retraction once the Little Ice Age is over, and then human impact coming in -- you see this on a large scale," Brugger said. "I did not expect the story to be so clear in the ice."

This study follows previous work by Brugger on another ice core from central Greenland, where the pollen record was markedly different from this southern Greenland ice core. This is because the two locations receive different atmospheric particles, Brugger says. The central Greenland ice core recorded more changes in Arctic vegetation than more distant locales.

The team also worked with colleagues in Vienna to create computer models that could simulate the movement of pollen through the atmosphere. This allowed them to better understand and verify where each Greenland site is positioned within global atmospheric circulation patterns.

"This southern ice core really tracked the boreal forest," Brugger said. "You get a completely different signal than with the previous study I did in central Greenland, where we tracked changes in the Arctic. These two ice core sites track very, very different things and it's all confirmed by atmospheric models, which is beautiful."

Studies like this offer a deeper understanding of environmental history that can also be applied to future changes in climate. Brugger plans to continue using ancient pollen to glimpse into the past, and is working on publishing the results of another study with a much deeper ice core that represents 8,000 years of history.