UNDERSTANDING THE ROLES OF CLIMATE, DISTURBANCE, AND FUNCTIONAL DIVERSITY IN THE TERRESTRIAL CARBON CYCLE: LINKING MECHANISMS FROM REGIONAL TO GLOBAL SCALES

Abstract

Terrestrial ecosystems currently sequester ~25% of human carbon emissions annually, yet changes in climate and disturbance have the potential to impact the integrity of the terrestrial carbon sink. Despite its fundamental role in mitigating anthropogenic CO2 emissions, the terrestrial biosphere remains one of the largest sources of uncertainty in current predictions from global climate models because key physiological processes are often poorly understood and/or poorly represented in terrestrial biosphere models. In this thesis, I use observations and terrestrial biosphere models to gain a better insight into (a) the mechanisms driving the terrestrial biosphere response to changes in temperature, water availability, nutrients, atmospheric CO2, and disturbance and (b) the uncertainty associated with projections for how climate change will impact the terrestrial carbon sink.

Chapter 2 demonstrates that the representation of water-limited photosynthesis in terrestrial biosphere models comprises a large and uncertain component of the terrestrial carbon cycle, comparable to 3-286% of current global productivity. Chapters 3-4 focus on the boreal biome, which contains >30% of terrestrial carbon and has the potential to experience shifts in composition and decreased soil carbon storage over the next century due to rapid warming, intensifying disturbance, and drought. In Chapter 3, I find a significantly negative growth response stemming from atmospheric drought across Alaska. Then in Chapter 4, I parameterize a terrestrial biosphere model for the North American boreal forest and use it to show that a temperature perturbation of 4°C can decrease total ecosystem carbon by ~40% after 300 years due to aboveground-soil carbon feedbacks. In Chapter 5, I use the same vegetation model to demonstrate that the high fire survival rate of thick-barked, large trees is key in the resistance of ecosystem carbon to increasing fire frequency.

Throughout this thesis, I highlight the importance of interactions between climate, disturbance, and functional diversity when understanding the trajectory of the land carbon sink. I further underscore important avenues for improvement and observational benchmarking of model processes. Projections of climate change impacts are not straightforward. However, incorporating process-oriented drought and disturbance mechanisms into models has the potential to significantly reduce uncertainty in carbon sink projections.