Navigating Disturbance Regimes in the New Arctic

NSF Award #1927772

This project focused on understanding the long-term history of disturbance in tundra areas underlain by permafrost to better anticipate how these regions may respond to ongoing change. Specifically, our team investigated how fire and permafrost thaw processes have changed through time by using lake sediment cores obtained from sites on the Alaskan North Slope and the Noatak River Watershed of Alaska. These two tundra ecoregions are particularly sensitive to the high rate and magnitude of warming that has been recorded in the Arctic over the past several decades. For example, both regions contain large ice deposits in the soil, which can lead to the formation of slumps that release mineral soil material to nearby lakes and streams when the ground ice melts. In addition, both areas have experienced wildfires in recent decades, which may be related to increasing shrub abundance that makes the landscape more flammable, as well increasingly warm/dry conditions and abundant lightening strikes that promote burning. Risks associated with increased fire activity and widespread abrupt permafrost thaw make maintaining infrastructure in these regions challenging. In addition, these disturbances impact ecosystem processes, wildlife, and Arctic resources. Thus, understanding how fire and permafrost thaw have varied in the past, and how those processes may interact through time, can help Arctic residents anticipate how these vulnerable tundra systems may respond to continued warming.  

Lake-sediment cores offer insight into how frequently fire and permafrost thaw processes have occurred in the past, as many lakes in the region have been accumulating sediment continuously over many thousand years. Thaw slump features have been observed to form, stabilize, and reactivate in reponse to warm soil temperatures as well as and large storms that remove soil and expose underlying ground ice. However, it is unclear how long these features can continue to reactivate. We obtained lake-sediment cores from twelve sites with thaw slumps on their shorelines in both areas, and used a combination of sediment and geochemical analyses to reconstruct how frequently these slump features released sediments to the lake basins in the past. We found that thaw features can remain active and periodically release sediment to lakes over hundreds to thousands of years, suggesting long-term vulnerability of Arctic landscapes and aquatic systems to ground ice thaw. Moreover, we found that these features can release ancient organic material to lakes, eroding carbon that is tens of thousands of years old. In addition to assessing thaw features, we also reconstructed past fires by analyzing charcoal particles deposited in lakes when surrounding areas burn. We found that fires became more frequent in northern Alaska when climate transitioned to drier conditions ~3000 years ago. We also found that areas with higher shrub abundance on the landscape burn more frequently than areas with less shrub cover, suggesting a link between ongoing Arctic greening and enhanced landscape flammability in the future.

The development of these new empirical datasets offer much-needed baseline data on disturbance processes in remote Arctic regions, where recorded observations are limited to the past ~70 years. We will house all datasets produced from this work in publicly available repositories online so that they are easily accessed by the general public, providing critical long-term historical information for land managers and stake holders in the region who are working on strategies to deal with increasingly disturbed Arctic environments. The datasets and associated publications generated from this work are also of value to the broader scientific community, where there is a major focus on investigating linkages between climate, disturbance, and carbon storage in remote Arctic regions. Beyond providing baseline information on disturbance processes, our research group is continuing to analyze the large datasets produced from this work to investigate interactions between wildfire and the initiation and reactivation of thaw slump features through time. This continuing research is particularly important for assessing feedbacks between disturbance processes in Arctic environments, which are not well understood at present. This project facilitated valuable field-based and laboratory-based scientific training for many undergraduate and graduate students, and formed the foundation of several student research presentations, honors theses, and graduate dissertations. Our research team currently has several manuscripts submitted and in preparation for publication in peer-reviewed journals, many of which are led by student researchers. In addition, results from this research have and will continue to be a part of ongoing outreach efforts to connect and share findings with the public, which include working with school teachers to design educational activities for young students and future scientists.

NSF Award #1928048

Permafrost ecosystems are undergoing rapid transformations due to climate change and shifting disturbance regimes. This research enhances our ability to forecast Arctic landscape transitions by identifying the spatial drivers and controls of past changes. Specifically, we improved predictions of permafrost landscape shifts in response to warming, permafrost collapse (thermokarst), lake drainage, and wildfire interactions.

Thermokarst

Data-driven machine learning models successfully predicted large-scale thermokarst events, such as thaw slumps, with high confidence, as well as the subsequent vegetation recolonization following disturbance. Climate warming and initial feature geometry were key drivers of lake change, while drained lake basins rapidly became dominated by tall shrubs. Additionally, small-scale thermokarst, such as ice-wedge degradation, was also predicted with high confidence. These features proved highly sensitive to the combined effects of warming and fire disturbance, though vegetation recolonization patterns remained less predictable.

Lake Drainage

By tracking lake area changes over 45+ years, we confidently predicted patterns of lake drainage across northern Alaska. Warmer temperatures and lake-specific geometries emerged as primary controls of increased drainage, with exposed basins quickly recolonized by tall shrubs.

Wildfire Interactions

The impact of wildfire on vegetation change varied across landscapes. In lowlands, permafrost collapse often drowned shrubs, leading to a decrease in shrub cover over time. Conversely, in upland areas, wildfires promoted shrub expansion. Notably, the presence of nitrogen-fixing alder accelerated the pace of local shrub expansion, particularly benefiting birch and willows by enriching the soil.

Public Engagement

Project personnel actively engaged with the public through outreach initiatives, including public talks, interactive workshops, and research showcases for middle school, high school, and undergraduate students in Illinois and New York. We participated in the University of Illinois’ Engineering Open House, developed educational materials (blog posts, infographics, social media content), and contributed to K-12 STEM outreach programs. Additionally, we maintained ongoing interactions with underrepresented communities in STEM, broadening access to geospatial and environmental science.