An inventory of N 2 O emissions from agriculture in China using precipitation-rectified emission factor and background emission Yanyu Lu a , Yao Huang a,b, * , Jianwen Zou b , Xunhua Zheng b a College of Resources & Environmental Sciences, Nanjing Agricultural University, Nanjing 210095, PR China b LAPC, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029, PR China Received 27 March 2006; received in revised form 10 July 2006; accepted 13 July 2006 Available online 1 September 2006 Abstract Fertilized agricultural soils are a major anthropogenic source of atmospheric N 2 O. A credible national inventory of agricultural N 2 O emission would benefit its global strength estimate. We compiled a worldwide database of N 2 O emissions from fertilized fields that were consecutively measured for more than or close to one year. Both nitrogen input (N) and precipitation (P) were found to be largely responsible for temporal and spatial variabilities in annual N 2 O fluxes (N 2 O–N). Thus, we established an empirical model (N 2 O– N = 1.49 P + 0.0186 P Æ N), in which both emission factor and background emission for N 2 O were rectified by precipitation. In this model, annual N 2 O emission consists of a background emission of 1.49 P and a fertilizer-induced emission of 0.0186 P Æ N. We used this model to develop a spatial inventory at the 10 · 10 km scale of direct N 2 O emissions from agriculture in China. N 2 O emissions from rice paddies were separately quantified using a cropping-specific emission factor. Annual fertilizer-induced N 2 O emissions amounted to 198.89 Gg N 2 O–N in 1997, consisting of 18.50 Gg N 2 O–N from rice paddies and 180.39 Gg N 2 O–N from fertilized uplands. Annual background emissions and total emissions of N 2 O from agriculture were estimated to be 92.78 Gg N 2 O–N and 291.67 Gg N 2 O–N, respectively. The annual direct N 2 O emission accounted for 0.92% of the applied N with an uncertainty of 29%. The highest N 2 O fluxes occurred in East China as compared with the least fluxes in West China. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Emission factor; Background emission; IPCC methodology; Nitrous oxide; Precipitation; Uncertainty 1. Introduction Nitrous oxide (N 2 O), an important atmospheric trace gas, contributes greatly to the global warming and strato- sphere ozone depletion. Although the N 2 O budget remains poorly understood at present, fertilized agricultural soils where N 2 O is naturally produced through the processes of nitrification and denitrification have been believed to be a major source of annual global N 2 O emission (Mosier et al., 1998). Reliable regional or global estimate of N 2 O emissions from agricultural soils depends on an examina- tion of methodologies to reduce the current high uncer- tainty in the estimates (IPCC, 2000). One potential way to do this is to develop predictive flux models, such as a simple empirical model (e.g. the IPCC methodology, IPCC, 2000) and a process-oriented model (e.g. the DNDC model, Li et al., 1992, 2001). The relationship between nitrogen input and N 2 O emis- sion established by Eichner (1990) and Bouwman (1996) motivated the concept of emission factor (EF). The emission factor is defined as a fraction of the nitrogen input released in the form of N 2 O within the current seasonal or annual period. Based on results obtained by Bouwman (1996), the default emission factor of N 2 O for synthetic nitrogen aver- ages 1.25%, ranging from 0.25% to 2.25% and N 2 O back- ground emission is assumed to be 1 kg N 2 O–N ha 1 yr 1 in the IPCC methodology (IPCC, 2000). Although some 0045-6535/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2006.07.035 * Corresponding author. Address: LAPC, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029, PR China. Tel./ fax: +86 25 84396406. E-mail address: huangy@mail.iap.ac.cn (Y. Huang). www.elsevier.com/locate/chemosphere Chemosphere 65 (2006) 1915–1924