Abstract:
Permafrost soils functioned as significant carbon (C) sinks for thousands of years, storing an estimated one-third of global terrestrial soil organic carbon (SOC). However, climate change-driven warming causes rapid thawing of permafrost, mobilizing substantial amounts of previously frozen C. This C mobilization allows microbial C decomposition, mineralization, and release in the form of greenhouse gases (GHGs), primarily carbon dioxide (CO₂) and methane (CH₄). Thaw-related increased GHG release can cause a positive permafrost climate feedback with consequences for the global climate. At the same time, climate warming and changing soil conditions upon thaw trigger shifts in the vegetation type, density, and productivity potentially amplifying, mitigating, or even reversing thaw-induced increases in GHG emissions. Understanding the drivers of these ecosystem responses, such as vegetation shifts, is critical for accurately assessing the strength and direction of the permafrost-climate feedback.
Permafrost thawing results in heterogeneous environmental conditions, leading to both drier upland soils and waterlogged wetlands. Wetland conditions in thawed permafrost areas promote anoxic soil conditions, influencing microbial processes. Wetlands are hotspots for microbial CH4 emissions, with CH4 having a stronger warming potential than CO₂. Wetland plants influence microbial activity and GHG emissions through various plant and root properties, referred to as plant and root traits. Examples of these traits include the type and amount of C exuded into the soil, root architecture, and the adaptive root trait to anoxia: the ability to transport gases between soil and atmosphere. These diverse plant and root traits add complexity to plant-soil-microbe interactions, creating uncertainty in predicting how vegetation shifts impact net GHG emissions, especially given seasonal and habitat-specific variations.
This thesis addresses these uncertainties by investigating how vegetation shifts caused by permafrost thaw influence GHG emissions, particularly through changes in plant and root traits and their interactions with microbial communities. Therefore, field observations from thawing permafrost peatlands are combined with controlled laboratory experiments examining individual and combined root traits.
The overarching outcome of the thesis demonstrates that vegetation transitions from shrub-dominated intact permafrost to graminoid (grass-like plants) dominated fully thawed wetlands, i.e. fens, substantially alter GHG dynamics. Graminoids in fens increased CO₂ and CH4 emissions, up to 2.3-fold and 6.8-fold, respectively, compared to non-graminoid areas. In contrast, shrubs in intact permafrost palsas had a modest, insignificant stimulatory effect on CO₂ emissions, increasing them by only 24%. This difference is related to the higher root productivity and C input of graminoids, alongside adaptive root traits such as plant-mediated CH₄ transport. Seasonal analyses revealed that while graminoids initially functioned as CO2 due to photosynthesis in the early and peak growing season, this C sink potential was offset by increased root-driven CH4 emissions and overall GHG emissions toward the end of summer (Chapter 2).
Specifically, graminoid root traits, including high root C exudation, increased microbial CH4 production, and stimulated microbial decomposition of SOC, a phenomenon known as positive priming (Chapters 2-5). These effects depended on soil properties, such as SOC content and mineral-associated organic carbon (OC) fractions. Interestingly, under soil conditions of low SOC content and high mineral-OC associations, simultaneous radial oxygen loss (ROL) from graminoid roots mitigated CH4 and CO₂ emissions compared to scenarios driven by root C exudation only (Chapter 5).
Moreover, this thesis demonstrates that root C exudate chemistry distinctly influences microbial interactions with soil minerals, particularly iron (Fe), affecting the stabilization or release of OC (Chapters 3-5). Simple C substrates, such as glucose, are often used in laboratory experiments, but tend to overestimate CH4 emissions compared to root-derived C exudates. This happens in particular, as they do not capture the temporal dynamics of the interactions of root-derived, natural C exudates with mineral phases, which affect the redox processes. Glucose also poorly mimics complex plant-derived root C exudates by overestimating microbial activity and emissions through artificially stimulating microbial fermentation processes, as well as by omitting the essential role of N-containing exudates (Chapter 4).
Further investigation into root architecture revealed that differences in root branching between shrubs and graminoids influenced GHG fluxes. Branched shrub roots reduced CH₄ emissions likely by trapping CH4 and increasing the activity of CH₄-consuming microorganisms in their rhizospheres. Elongated, deeper graminoid roots, conversely, facilitated CH4 transport to the atmosphere, increasing CH4 emissions. Controlled chamber experiments using three-dimensionally printed root structures provided further insight into how root architecture could affect gas movement, complementing the field-based observations (Chapter 6).
Collectively, these findings challenge the widespread rationale that increased vegetation cover and plant production in Arctic regions, termed "Arctic greening", uniformly increases C sequestration. Instead, this thesis highlights that the net climatic impact of vegetation shifts is more complex and controlled by belowground root traits, which vary on a species-level and spatio-temporally. The transition towards productive graminoid-dominated vegetation in waterlogged soils, while appearing beneficial due to high C uptake, can amplify CH4 emissions, potentially offsetting or even reversing the expected GHG sink effect in seasonal phases of reduced photosynthesis. With this, the outcomes of this thesis are in line with emerging studies that call for careful consideration of belowground plant traits and their interactive effect with the surrounding soils, including microbial communities.