Abstract: At present, large-scale ecosystem nitrogen flux calculation mainly depends on quasi-empirical models which quantify the nitrogen fluxes by using empirical formulas based on various kinds of collected parameters. However, quasiempirical models have several limitations in application, including (1) completeness of calculation cannot be ensured due to the lack of variables, (2) uncertainty analysis is difficult, and (3) data sharing is lack of standardization. Ontology is a semantic model for describing concepts and their relations, which can unify and enrich the description of data to improve data sharing. In addition, the reasoning mechanism and hierarchy of ontology can provide automatic reasoning for quasi-empirical models to find out unavailable variables and conduct uncertainty analysis. In this paper, the domain ontology for the quasi-empirical models of nitrogen flux calculation (NFCO) was constructed, and a framework of automatic reasoning and uncertainty analysis based on NFCO was proposed, which would improve the quality of large-scale ecosystem nitrogen flux calculation, and build a solid foundation for establishing large-scale ecosystem nitrogen cycle database.  

Key words: dairy effluent irrigation, winter wheat  summer maize rotation, grain yield, phosphorus utilization efficiency.