过去25年全球农田N2O排放的时空格局及其驱动机制

Nitrous oxide (N2O), one of the most important long-lived greenhouse gases, has been persistently increasing in the global atmosphere due largely to its emissions from agricultural soils. Recent regional studies suggest that the nonlinear responses of N2O flux to fertilizer nitrogen vary with environmental variations, yet the patterns, trends and causes of global agricultural N2O emissions remain elusive on either regional or global scale. Here we will analyze the global agricultural N2O emissions and their driving mechanisms during the period of 1990-2014 using three different methods: empirical upscaling of ground-based observations, ecosystem models and atmospheric inversions of N2O concentration data. For training upscaling and ecosystem models, we will collect N2O observations and the corresponding environmental factors in global agricultural soils from large-scale projects (e.g., TRAGNET, GRACEnet, NitroEurope, GHG-Europe, NANORP), scientists, and peer-reviewed studies. To reduce the uncertainty of the emission estimates, we will also develop a high-resolution database including fertilizer nitrogen applied in agricultural soils (i.e., synthetic fertilizers, manure, crop residues, and soil mineral nitrogen), cropland patterns (i.e., upland grain crops and paddy rice), nitrogen management options, etc. First, nonlinearity and heterogeneity of global N2O flux as well as their dependences on environmental factors will be analyzed; In addition, utilizing the existing process-based crop model and the support of flux observations and field experiments, the project will improve the diagnostic and prognostic modeling ability through optimizing the key model parameters based on the model-data fusion technique. We will also estimate N2O emissions over global arable lands from 1990 to 2014 with the test of significance of the nonlinearity and heterogeneity. Those models will be further applied to the attributions of anthropogenic causes of the spatial patterns, interannual variability, temporal trends, and growth rates of global agricultural N2O emissions. Overall this project will broaden our knowledge of the interaction between climate change and nitrogen cycle, which is important for refining IPCC default values of emission factors and designing climate-smart agriculture.

氧化亚氮（N2O）是最重要的温室气体之一，其全球大气浓度持续增高。农田排放是最大的贡献源，但其全球尺度的估计、分布及归因存在相当大的不确定性。最新的区域研究表明，N2O排放对施氮量存在非线性和空间差异性响应。这种现象的决定因素是什么？全球排放时空格局及其驱动机制又是什么？本项目拟利用全球通量观测和控制实验、高分辨率的施氮量、氮肥管理措施等数据，阐明全球农田N2O排放对施氮量的非线性和空间差异性响应及决定因素；优化过程模型氮循环模块的关键参数，实现观测和模型的融合，提高模型的模拟精度；采用基于地面观测的升尺度、大气反演和改良的过程模型相结合的方法，定量刻画1990-2014年期间全球农田N2O排放时空格局及其驱动机制，并解析洲际和国家尺度社会经济演变对农田N2O排放量增长速率的贡献。本项目不仅有助于深化理解气候-氮循环的相互作用，且为修正IPCC指南和建立气候智能型农业提供科学依据。

本项目围绕“N2O排放时空格局及其驱动机制”这一重要研究热点开展了系列研究，在数据、模型和过去N2O排放时空格局及其驱动机制等方面取得一些成果，包括：（1）建立了全球农田N2O通量观测和控制实验数据库，解析了全球农田N2O排放因子空间差异的驱动机制，发现土壤和气候是N2O排放因子的决定因素，而不是农艺管理措施；（2）开发了农田N2O排放过程模型-观测数据融合系统，该模型被全球氧化亚氮模型比较计划(NMIP)和全球农作物模型比较计划(GGCMI)所采纳；（3）解析了我国农田N2O排放“增速变缓”的自然与人为驱动机制；（4）采用基于地面观测的升尺度、自上而下的大气反演和自下而上的过程模型3种方法，实现了1961-2014年全球农田N2O排放的统一化评估，被IPCC AR6工作报告的第五章收录，还被未来地球Global Carbon Budget所采纳；（5）开发了全球农田施氮量高分辨率产品，识别了20%的播种面积可以实现60%以上的全球农田N2O减排潜力，为制定减排政策提供科学依据。在项目的支持下，发表中文核心期刊论文1篇和SCI论文10篇（其中第一标注6篇），包括1篇Nature，1篇National Science Review，2篇Global Change Biology杂志论文。