What does permafrost thaw mean for future Arctic ecosystems?
by Alison Cassidy
Flying over the Arctic by helicopter, some areas of the tundra look as if they are falling apart. The landscape is crumbling, which from above looks like a group of dark spots. As you get closer, the sound of rushing water and frequent loud cracks can be heard. These are examples of active layer detachment slides and retrogressive thaw slumps, which are landslides caused by permafrost thaw.
In our recent paper published in Arctic Science, we explored the impacts of active retrogressive thaw slumps on the amount of carbon absorbed and released by the tundra. With climate change, permafrost disturbances are happening more frequently and over a larger area, increasing the movement (flux) of carbon from the tundra to the atmosphere. We found that in some cases these disturbances absorbed less carbon than undisturbed tundra and that they were distinguished by different soils and vegetation. Given that these disturbances are expected to increase over time, they may turn the tundra from a carbon sink to a carbon source over the growing season. Understanding these carbon fluxes can help us predict future ecosystem changes associated with a warming climate.
During the summers of 2012 to 2014, we studied the recovery and revegetation of different permafrost disturbances on the Fosheim Peninsula of western Ellesmere Island. Our research site was close to Hot Weather Creek, which was once a High Arctic Global Climate Change Observatory operated by the Geological Survey of Canada. This is an area of ice-rich permafrost, where a number of permafrost disturbances have been studied since the 1990s.
In the High Arctic, the growing season of the tundra ecosystem, characterized by dwarf shrubs and grasses, is rather short: from mid-June to mid-August. Despite this limitation, the area has been called the “Garden Spot of the Arctic” due to the high diversity of vegetation. Researchers have found over 140 vascular plant species in this polar oasis!
We spent 2012 doing aerial surveys and started on-the-ground work in 2013. Upon our arrival on the Fosheim Peninsula, we set up camp with a view of the Sawtooth Mountains to the west. We flew into Eureka, approximately 50 km away, in a Twin Otter and were then taken by helicopter to our study site. We had to bring enough equipment and food for the duration of our stay, so effective planning was essential. One forgotten piece of equipment could mean failure to capture a key piece of the puzzle.
Given all our gear, we had to sling it into our field site using a large net hanging below the helicopter. Our minimal camp was comprised of seven white and yellow tents. The silence at such a remote location was astounding and provided ample opportunity for nature viewing. We had regular visits from local Arctic fox, Arctic hares, ermine, and a pack of Arctic wolves.
We used the eddy covariance technique to find out whether ecosystem fluxes from a heavily disturbed tundra landscape were different than from undisturbed tundra. Eddy covariance measures gas fluxes, including carbon dioxide (CO2) and wind speed and direction. Using this information, we can figure out whether the landscape is, on average, absorbing or releasing carbon.
From our initial aerial surveys in 2012, we selected two retrogressive thaw slumps that we monitored throughout the growing season. To understand both local carbon fluxes and regional (landscape-level) variations, which are especially important in mixed environments like Arctic tundra, we used portable chamber systems and eddy covariance measurements.
We used a portable chamber system both at the site of the thaw slumps and at a recovered slump to measure the amount of carbon released by microorganisms within the soil (i.e., respiration). We also installed two flux towers at the edges of the active slumps for the duration of the growing season. When winds were coming from the direction of the slumps, fluxes were attributed to the disturbed terrain; winds from other directions represented fluxes from the undisturbed landscape. We also examined the impact of permafrost thaw on regional soils and vegetation by measuring soil characteristics (moisture and nutrients) and vegetation cover.
During the growing season, we found greater CO2 uptake in undisturbed than disturbed tundra, showing that eddy covariance can successfully detect fluxes from disturbed terrain. When compared with undisturbed tundra, we also found wetter soils, increased nutrient availability and respiration, and less vegetation at the slumps. At the recovered slump site, we found wet soils and increased respiration similar to the disturbed site, but we also found more vegetation. The dynamic nature of these disturbances was evident as an area within the slump reactivated partway through the growing season. I was trapped by the viscous material several times and needed help to pry myself out of the mud!
During the 2014 growing season, we continued to explore the impacts of disturbance on carbon fluxes by co-locating flux towers on one active retrogressive thaw slump located in an otherwise undisturbed landscape, just east of our 2013 research site. Here we found that, during the growing season, the undisturbed tundra acted as a weak carbon sink while the disturbed tundra acted as a carbon source.
From the air, it’s obvious that permafrost disturbances are significant features on the landscape, and their frequency is increasing. Given that they can be carbon sources instead of sinks, climate change model predictions need to take them into account to accurately model ecosystem responses to climate change.
Alison Cassidy received her Ph.D. from the Department of Geography at the University of British Columbia.
Read the full paper — Impacts of active retrogressive thaw slumps on vegetation, soil, and net ecosystem exchange of carbon dioxide in the Canadian High Arctic in Arctic Science.
