The present and future of US forest carbon allocation and turnover time

The turnover time of forest biomass carbon is highly dynamic across space and time and is projected to decrease due to the acceleration of land use change and disturbance. However, turnover time may also shift due to changes in within-tree carbon allocation and species composition, processes that are highly unresolved. Using trait datasets and forest surveys, we developed US-wide maps of carbon contained in tree structural pools (leaves, stems, coarse roots, and fine roots), from which we derived forest biomass carbon turnover time. We found that hotter and wetter forests across the US experience lower carbon turnover time, primarily due to differences in tissue longevity and carbon allocation across species. We then tested the extent to which two mechanisms – shifts in carbon allocation and species composition – may affect carbon allocation and turnover time into the future using species distribution modeling and an individual-based tree model that can simulate changes in allocation. Turnover time generally decreased in the future across all methods, but the magnitude of this change, along with its underlying mechanisms, differed greatly depending on model type. This work underscores the need for expanded observations of carbon allocation in field settings on mature trees, and hints at the promise of optimality-based allocation models. Importantly, our results can be used to identify hotspots of carbon sequestration and constrain the sources of uncertainty in future forest carbon turnover time.