Accurate quantification of nitrogen (N) loss is essential for N management in agricultural production and consumption system. Taking the Yangtze River Delta (YRD) region as the study area, N loss characteristics of agricultural production and consumption system was estimated based on substance flow analysis, and sources of uncertainty were quantitatively analyzed using error propagation equation, and multiple regression analysis was used to analyze the transmission of uncertainty. The results showed that from 2011 to 2020, the total amount of N loss in the agricultural production and consumption system in the YRD initially increased from (1841.0±150.4) Gg/a in 2011 to (1874.1±154.2) Gg/a in 2013, and then decreased year by year to (1636.4±144.6) Gg/a in 2020. The cropland and livestock subsystems were the two largest sources of N loss, accounting for 37.5% and 31.0% on average, respectively. Moreover, the proportion of N loss of them both showed a decreasing trend, while that of human consumption and waste management subsystem showed an increasing trend. It was mainly attributed to the decrease of agricultural production scale and urbanization process in the YRD. Atmospheric environment was the primary sink of N loss in the system, with an average proportion of 52.2% in the 10 years. Based on calculation, the uncertainty of total N loss in the system was 8.1%-8.8%, in which the waste management subsystem and surface water were the sources and sinks of N loss with the highest uncertainty, respectively. In terms of uncertainty sources, the uncertainties of N loss introduced by human activity level data and N flow parameters were 1.2%-1.3% and 8.0%-8.8%, respectively, indicating that the latter was the main source of uncertainty. In addition, N loss from livestock subsystem and to surface water environment contributed the most to the total N loss uncertainty of the system, responsible for 27.4% and 50.0%, respectively. Multiple regression analysis showed that the N loss flux of each component and its uncertainty both significantly affected the uncertainty of the total N loss of the system, and the N loss flux of each component was the predominant factor affecting the uncertainty propagation. The influence of the N loss flux of each component on the total N loss uncertainty was about 2.3 times that of the latter. The findings could provide a reference for reducing the uncertainty of N loss in agricultural production and consumption system.