Global soil nitrous oxide emissions since the preindustrial era estimated by an ensemble of terrestrial biosphere models: Magnitude, attribution, and uncertainty

Our understanding and quantification of global soil nitrous oxide (N₂ O) emissions and the underlying processes remain largely uncertain. Here, we assessed the effects of multiple anthropogenic and natural factors, including nitrogen fertilizer (N) application, atmospheric N deposition, manure N application, land cover change, climate change, and rising atmospheric CO₂ concentration, on global soil N₂ O emissions for the period 1861-2016 using a standard simulation protocol with seven process-based terrestrial biosphere models. Results suggest global soil N₂ O emissions have increased from 6.3 ± 1.1 Tg N₂ O-N/year in the preindustrial period (the 1860s) to 10.0 ± 2.0 Tg N₂ O-N/year in the recent decade (2007-2016). Cropland soil emissions increased from 0.3 Tg N₂ O-N/year to 3.3 Tg N₂ O-N/year over the same period, accounting for 82% of the total increase. Regionally, China, South Asia, and Southeast Asia underwent rapid increases in cropland N₂ O emissions since the 1970s. However, US cropland N₂ O emissions had been relatively flat in magnitude since the 1980s, and EU cropland N₂ O emissions appear to have decreased by 14%. Soil N₂ O emissions from predominantly natural ecosystems accounted for 67% of the global soil emissions in the recent decade but showed only a relatively small increase of 0.7 ± 0.5 Tg N₂ O-N/year (11%) since the 1860s. In the recent decade, N fertilizer application, N deposition, manure N application, and climate change contributed 54%, 26%, 15%, and 24%, respectively, to the total increase. Rising atmospheric CO₂ concentration reduced soil N₂ O emissions by 10% through the enhanced plant N uptake, while land cover change played a minor role. Our estimation here does not account for indirect emissions from soils and the directed emissions from excreta of grazing livestock. To address uncertainties in estimating regional and global soil N₂ O emissions, this study recommends several critical strategies for improving the process-based simulations.