Increases in atmospheric nitrogen deposition (Ndep) can strongly affect the greenhouse gas (GHG; CO2, CH4, and N2O) sink capacity of grasslands as well as other terrestrial ecosystems. Robust predictions of the net GHG sink strength of grasslands depend on how experimental N loads compare to projected Ndep rates, and how accurately the relationship between GHG fluxes and Ndep is characterized. A literature review revealed that the vast majority of experimental N loads were higher than levels these ecosystems are predicted to experience in the future. Using a process-based biogeochemical model, we predicted that low levels of Ndep either enhanced or reduced the net GHG sink strength of most grasslands, but as experimental N loads continued to increase, grasslands transitioned to a N saturation-decline stage, where the sensitivity of GHG exchange to further increases in Ndep declined. Most published studies represented treatments well into the N saturation-decline stage. Our model results predict that the responses of GHG fluxes to N are highly nonlinear and that the N saturation thresholds for GHGs varied greatly among grasslands and with fire management. We predict that during the 21st century some grasslands will be in the N limitation stage where others will transition into the N saturation-decline stage. The linear relationship between GHG sink strength and N load assumed by most studies can overestimate or underestimate predictions of the net GHG sink strength of grasslands depending on their N baseline status. The next generation of global change experiments should be designed at multiple N loads consistent with future Ndep rates to improve our empirical understanding and predictive ability.
Keywords: CH 4; N2O; grassland; methane; net ecosystem CO2 exchange; net ecosystem productivity; nitrogen deposition; nitrogen fertilization; nitrous oxide; uncertainty.
© 2015 John Wiley & Sons Ltd.