This Is AuburnElectronic Theses and Dissertations

Reactive Nitrogen and Climate Change in the Terrestrial Biosphere: Data-Model Integration




Pan, Naiqing

Type of Degree

PhD Dissertation


Forestry and Wildlife Science

Restriction Status


Restriction Type

Auburn University Users

Date Available



The addition of excess reactive nitrogen (N) compounds to terrestrial ecosystems has largely stimulated the emissions of nitrous oxide (N2O), the most important depleting substance of stratospheric ozone and a potent greenhouse gas. As a result, atmospheric N2O concentration has increased by more than 20% since the pre-industrial era, and the observed atmospheric N2O concentration has exceeded predicted levels across all scenarios in the Coupled Model Intercomparison Project Phase 6 (CMIP6) for the sixth assessment report of the Intergovernmental Panel on Climate Change (IPCC). From both scientific and climate change policy, therefore, it is essential to quantify how natural and human disturbances have affected historical terrestrial N2O emissions and predict their future changes. Here, we first developed a set of gridded N input datasets and found that the total anthropogenic N inputs to global terrestrial ecosystems increased from 29.1 Tg N yr-1 in the 1860s to 267.2 Tg N yr-1 in the 2010s. Then, we used the newly developed N input datasets to drive the Dynamic Land Ecosystem Model (DLEM). DLEM suggests that global soil N2O emissions significantly increased from 7.72 Tg N yr-1 in the 1850s to 12.46 Tg N yr-1 in the 2010s, and the largest increase of soil N2O emissions was in croplands, accounting for 90% of the total increase, which is larger than the estimate of N2O Model Intercomparison Project (82%). The rising atmospheric CO2 concentration suppressed soil N2O emissions, while all other environmental factors stimulated N2O emissions. In addition, we also compared DLEM estimates with other terrestrial biosphere models (TBM) participated in global Nitrogen Model Intercomparison Project phase 2 (NMIP2) to identify the discrepancies among TBMs. Using both bottom-up (BU) and top-down (TD) approaches, we further investigated regional N2O emissions from two contrasting regions: the human-dominated East Asia and climate-dominated northern high latitudes. The results show that anthropogenic sources dominated East Asia N2O emissions. Although bottom-up approach suggested a larger magnitude of total emissions (1.73 Tg N yr-1) than top-down approach (1.47 Tg N yr-1), both approaches detected significant increases in N2O emissions in recent two decades and top-down approaches suggested a larger increase rate. All anthropogenic N2O emissions showed significantly increasing trends, among which agriculture made the largest contribution. As a comparison with East Asia, soil N2O emissions from the northern high latitudes increased from 333±190 Gg N yr-1 in the 1860s to 570±277 Gg N yr-1 during 2007-2016. Climate change played an increasingly important role with an overall contribution of 37% to the increase, with particularly large influence in the Permafrost Region. Warming dominated the regional-scale climate forcing because precipitation effects offset each other spatially. Top-down models yield spatial patterns of total N2O emissions comparable with the bottom-up estimate but gave a lower estimate of total regional emissions from all sources (668±134 Gg N yr-1 VS 1155±267 Gg N yr-1). Future projections indicate that global soil N2O emissions will increase under bioenergy (SSP126-high), medium N regulation (SSP245-medium), and business-as-usual (SSP585-high) scenarios, while keep fluctuation under the best-case (SSP145-high) scenario, and the average magnitudes of global soil N2O emissions in the 2090s under these four scenarios are 13.07, 13.89, 18.12, and 12.57 Tg N yr-1, respectively. The rapid increase in N2O emissions under the SSP585-low scenario is mainly driven by climate change and N inputs. This study also identifies potential sources of uncertainties in current N2O emissions, and limitations and knowledge gaps in existing information. We proposed several potential approaches to reduce the identified uncertainties and improve our understanding of N2O emissions.