Improving the assessment of the State of the Carbon Cycle in North America by integrating inventory- and process- based approaches: A case study for Canada

Abstract:

Regional and continental carbon stock and flux estimates differ among assessments depending on the scaling approach used and the budget components considered. This is particularly manifest across the vast circum-boreal region, which has experienced substantial modification of the major driving forces of the carbon cycle in recent decades, including pronounced climate warming and associated increases in the frequency and severity of disturbances. In Canada, inventory-based estimates suggest a small carbon sink for its managed forest, but do not include unmanaged lands nor capture major driving forces such as climate change and atmospheric chemistry. On the other hand, estimates from process-based models vary widely and often do not consider critical disturbance and management impacts. Here, we demonstrate results from an updated approach that integrates inventory-based information on management and disturbances with process-level representation of ecological dynamics using a terrestrial biogeochemistry model.

The integrated approach facilitates more comprehensive diagnosis of Canada’s land-based carbon budget within a framework that also allows for attribution of the major driving forces and prediction under future scenarios. Using this framework, we diagnose an approximately 30 Tg C yr⁻¹ sink in Canada over the first decade of the 21st Century, which represents a significant reduction in the strength of the CO₂ sink estimated for previous decades. This decline in sink strength is attributed primarily to CO₂ emissions from the substantial area disturbed by wildfire and insect outbreaks in recent years. Such changes are predicted to create positive feedbacks to the climate system that accelerate global warming. Compared to other assessments, our results suggest that CO₂ uptake by the region’s ecosystems may not be as strong as estimated by atmospheric inverse approaches, which are highly uncertain over the high latitudes, or by process-based models that do not consider permafrost dynamics, disturbance and management impacts, or other important drivers in these regions. Although there are benefits in retaining independence among approaches for estimating and comparing carbon fluxes, our approach demonstrates the significant progress to be made by more formally integrating them.