Abstract
The tundra-taiga ecotone (TTE), the transitional zone between the boreal forest and the Arctic tundra, spans over 3 million km2 and is dynamically responding to climate change. The morass of interacting and conflicting environmental drivers can make the TTE dynamics difficult to predict. The TTE is crucial for global carbon budgets, as it contains important forest resources growing above large stores of labile carbon held in the unfrozen soil and permafrost. The goal of my dissertation is to examine the growth drivers within the TTE of various plant functional types (PFTs), which are aggregations of species with similar structure and function. Research on vegetation response to climate change in the TTE typically focuses on single drivers in narrow spatial and temporal ranges, leaving a gap in integrating ecosystem drivers and responses across large, heterogenous landscapes. By taking an interdisciplinary, multivariate approach, and leveraging both field data and model outputs, I can assess shifting vegetation dynamics in the TTE which can aid in predicting their impact on global carbon budgets.
In my first chapter, I assess variability in growth drivers of black and white spruce trees in the TTE, using the NOAA International Tree Ring Database. Challenging the assumption that growth increases with warmer temperatures, I examined the strength of climate variables influence on annual growth, and how site-specific environmental variables interact with climate. I found that the climate in the “shoulder” seasons, particularly May temperature, has increased in importance for predicting tree growth, as summer temperature importance has decreased.
My second chapter analyzes the community composition of vegetation in the Canadian Northwest Territories using Canada’s National Forest Inventory dataset applying multivariate analyses. One of the challenges to predicting vegetation dynamics in the TTE is the high heterogeneity (climate, soil, vegetation) of the landscape. By combining the NFI species-level composition dataset and local site measurements, I determined what drives specific plant functional type (PFT) changes and how they manifest in remote sensing records. Each of the 11 represented PFTs responded to different abiotic and biotic drivers, which could result in novel plant communities in the future.
The third chapter reviews the dynamic vegetation models (DVMs) that are currently being used to predict the future of the TTE. As climate shifts in the Arctic, vegetation is responding across the landscape in novel ways. DVMs are able to not only predict vegetation change across remote areas, but also project vegetation composition into the future. However, many of these models are challenging to understand for non-modelers, and the ecosystem elements they are modeling are difficult to determine. I also surveyed experts on what is needed in dynamic vegetation models to improve model predictions and found a demand for models with permafrost and active layer dynamics. Reviewing the state of the science on modeling and outlining these ecosystem processes increase accessibility and understanding of this class of models for future applications.
My fourth chapter conducts a sensitivity analysis of SIBBORK-TTE, to determine the influence of adding three tall shrub genera to the model. I assessed how different combinations of shrub genera influence black and white spruce growth both with historical climate and a warming climate. I found that the shrubs were not only successfully able to compete with the spruce, but they also alter the importance of certain variables under different growing conditions.
To summarize, my first two chapters assess interacting ecosystem drivers of community composition and growth variation utilizing field data sets and multivariate analysis. Applying these findings, the second half of my dissertation focuses on improving dynamic vegetation models for the TTE. My dissertation will advance the state of the science on growth drivers in the Arctic-Boreal Region and will support the further development of model simulations by highlighting which drivers should be incorporated to create better ecosystem facsimiles. The TTE is a crucial biome transition zone and a highly dynamic carbon sink; by understanding how this area is responding to climate change, we will be able to more accurately predict its fate.
