The biophysical and carbon-climate feedbacks of shrub expansion in the Arctic

Open Access
Cahoon, Sean Michael
Graduate Program:
Doctor of Philosophy
Document Type:
Date of Defense:
October 21, 2015
Committee Members:
  • Eric S Post, Dissertation Advisor
  • Eric S Post, Committee Chair
  • David Eissenstat, Committee Member
  • Armen Ricardo Kemanian, Committee Member
  • Matthew David Hurteau, Committee Member
  • Arctic
  • Betula nana
  • climate change
  • CO2 exchange
  • Poa pratensis
  • shrub expansion
Recent warming in the Arctic has been twice the global average, leading to substantive changes in terrestrial ecosystem structure and function. Of primary concern is the vast amount of soil carbon (C) in the region, which, if mineralized, is expected to provide a significant positive feedback to global climate in the form of increasing atmospheric CO2 concentrations, and accelerate climatic warming. Alternatively, C uptake may be increasing in the region due to extended growing seasons and changes in plant community composition. The expansion of woody shrubs, for example, has been widely noted throughout the region, and is expected to impact ecosystem C budgets and alter soil C pools. However, the timing, magnitude and direction of changes in the C cycle in response to shrub expansion remain uncertain. Addressing the C cycle impacts of shrub expansion requires an examination of CO2 exchange processes at multiple spatiotemporal scales. At the broadest scale, Chapter 2 analyzes mid-summer CO2 exchange among herbaceous and woody tundra across 21 sites spanning 16° of latitude across the Arctic and Boreal zone. Variation in canopy temperature explained approximately half the variation in ecosystem respiration (ER), which, in turn, drives patterns of variation in net ecosystem CO2 exchange (NEE) across ecosystems, at least during the peak of the growing season. In addition, woody sites were slightly stronger sinks for atmospheric C; however, when soils were above 10° C, these sites became a net source for CO2. Meanwhile, herbaceous tundra remained a moderate C sink, regardless of soil temperature. These results indicate that an increase in shrub abundance may result in a net loss of C in the middle of the growing season if soils temperatures also rise. Identifying the primary drivers of CO2 uptake is a vital component of ecosystem C cycling, and may elucidate key physiological differences between shrubs and graminoids that confer a competitive advantage to a particular species. In Chapter 3, I combined leaf gas exchange with stable isotope analysis of two common arctic species, Betula nana and Poa pratensis, to reveal important difference in C uptake processes. I found greater drought sensitivity in the graminoid species, P. pratensis, which displayed reductions in stomatal conductance and photosynthesis during periods of high atmospheric vapor pressure deficits and low soil water content. Leaf Δ13C and Δ18O showed a negative relationship, indicating a strong stomatal influence on CO2 uptake. In contrast, Betula displayed less sensitivity to drought, and leaf Δ13C and Δ18O were not correlated with one another for this species. These key differences in C-H2O relations between co-existing species may confer an advantage to Betula if the climate becomes warmer and drier. Partitioning soil respiration (RS) into heterotrophic (RH) and autotrophic (RA) sources may provide valuable insight into the breakdown of soil organic matter (SOM) and short-term C flux from the rhizosphere. Chapter 4 presents a multi-year analysis of soil respiration (RS) partitioning among four common vegetation types in West Greenland. Although there were major differences in RS between vegetation types, the autotrophic proportion (RA / RS ≈ 47%) was remarkably similar among vegetation types and between seasons. My comparative analysis suggests that RA / RS may be more conservative than what might be estimated based on global models, and that partitioning RS may be a useful diagnostic to identify ecosystems that are gaining or losing soil C. In Chapter 5, I examine the effect on the seasonal C cycle at the ecosystem scale of increasing abundance of Betula nana in West Greenland. Sites dominated by Betula were a stronger C sink than those dominated by graminoids due to higher gross ecosystem photosynthesis (GEP) and lower ER; however, water stress may have played a role in reducing mid-season GEP for both vegetation types. Vegetation along the transition between Betula and graminoid patches was commonly the strongest C sink, suggesting complementary in resource use. I detected a strong early season phenological influence on net C uptake, which coincided with recent warming in the region. A retrospective analysis revealed a net gain of 1.3 and 2.1 g C m-2 season-1 in Betula and graminoid tundra respectively, since 2002. These results suggest GEP has responded more strongly to recent warming than ER, resulting in an extended growing season that has increased net C uptake in West Greenland and provided a negative feedback. My research represents a comprehensive examination of C cycle processes across multiple scales in one of the most common types of Arctic tundra. However, there are clear gaps in our knowledge that remain unfilled. For instance, there may be important interactive effects between greater leaf area, litter input and changes in soil temperature associated with shrub expansion that will feedback to alter nutrient turnover and ultimately, plant productivity. My research has identified the importance of species-specific responses to climate perturbations and the need to focus future work on expanding, contracting and stable patch edges to elucidate physiological controls on vegetation change. The chapters presented herein represent a significant contribution to our understanding of how structural and physiological differences between woody and herbaceous species affect the magnitude of carbon-climate feedbacks in the Arctic.