Nonlinear response of soil ammonia emissions to fertilizer nitrogen

Ammonia (NH3) is an important atmospheric pollutant that threatens ecosystem and human health. Synthetic nitrogen (N) fertilizer applications are a major source of atmospheric NH3. Most of current bottom-up estimates assume that the NH3 emission response to increasing N application rates is linear, and thus constant emission factors (EFs) are used. However, increasing evidence suggests that NH3 emissions increase exponentially with increasing N inputs. In the present study, we conducted a meta-analysis to generalize the relationship between N inputs and NH3 emissions. Overall, the change in EF per unit of additional N fertilizer input (Delta EF) was positive from 70 experiments with at least three N application rates, suggesting that NH3 emissions in response to increasing N additions grow at a rate higher than linear. Compared to our Delta EF model, the 10% EF model used by Intergovernmental Panel on Climate Change overestimated NH3 emissions when fertilizer N is applied at low levels, but underestimated NH3 emissions when N is applied in excess. Therefore, our results suggest that replacing the constant EF with the N-rate-dependent EF could improve the accuracy of NH3 emission estimates.