The Role of Carbon-Nitrogen Interactions for Terrestrial Ecosystem Dynamics under Global Change - a modelling perspective

The nature of future climate change will depend on anthropogenic emissions of CO2, as well as climate- and CO2-mediated feedbacks through carbon (C) cycling in both terrestrial ecosystems and oceans. Terrestrial ecosystems remove presently about 25% of the anthropogenic CO2 fossil-fuel and land-use change emissions, but to attribute which mechanisms cause this uptake, and the key regions where it occurs, is a challenging task. Considerable attention has focused in recent years on whether, and how, interactions of the C and nitrogen (N) cycles affect the future terrestrial C sink. Until relatively recently these interactions were not considered in models of the global C cycle, although in many ecosystems N is believed to be a limiting factor controlling vegetation productivity. The dynamic vegetation model LPJ-GUESS has been extended with a fully coupled dynamic C-N cycle in vegetation and soil, introducing N limitations on plant production and soil decomposition. With N dynamics, LPJ-GUESS simulates the present C and N pools in soil, litter and vegetation in agreement with observation-based and model estimates. Global simulations show a steeper gradient of productivity from high to low latitudes compared with the C-only model version, increasing the ability to correctly reproduce productivity in boreal and tropical ecosystems when evaluated against 75 FLUXNET forest sites. Secondary effects emerge also via ecosystem ecological processes, such as C-N interactions altering the competition between plant functional types, resulting in some differences in the modelled biome distribution, e.g. a more southerly arctic treeline when N cycle dynamics are included. When applying “business-as-usual” scenario of future atmospheric CO2, climate and N deposition, the inclusion of N dynamics results in moderately higher cumulative C sequestration over the period 1850 to 2100 compared to the C-only version of LPJ-GUESS. This result contrasts to some degree with results of earlier studies using other models that are dominated by progressive N limitation in the future at global scale. In LPJ-GUESS, enhanced soil N mineralisation in a warmer climate particularly affects net primary productivity in high-latitudes, enhancing the growth of trees and providing a transient sink of carbon as woody biomass as boreal forests densify and expand. Our results highlight the need to account for C-N interactions not only in studies of global terrestrial C cycling but to understand the underlying interactions on regional scales