Objectives In order to provide theoretical basis for improving rice yield and N use efficiency, we studied the effect of different nitrogen (N) management models on rice growth, N accumulation and distribution, and yield formation.

Methods ‘Zhendao 11’, a widely planted rice in Jiangsu Province, was used in this study to investigate the biomass, yield, N accumulation, N transport and their relationships under different N managements at four growth stages of rice, including tillering, shooting, flowering and maturity. The experiment was conducted in the long-term fertilizer experiment field from 2008 in the Agriculture Science Research Institute of Rugao, Jiangsu Province. The three N fertilizer management models were designed as follows: free N fertilizer control (CK), farmer’s N fertilizer practices (N 350 kg/hm², N application of basal: tillering: shooting was 4:4:2, FFP), and optimum N fertilizer managements (N application of basal: tillering: shooting: flowering was 4:2:2:2, OPTs) that including optimal N treatment (N 240 kg/hm², OPT), the N rate was substituted by 20% N with organic fertilizer base on OPT treatment (OPT1) and the N rate was substituted by 20% N with organic fertilizer after reducing 20% N base on OPT treatment (OPT2).

Results The average yield of optimum N managements (OPTs) increased by 8.4% compared with FFP. In OPTs, the kernels per spike, seed setting rate and 1000-grain weight were higher than FFP treatment through improving the accumulation of biomass and N after anthesis. Grain N accumulation in FFP treatment was mainly resulted due to the increased amount of remobilized N from plant N absorption before anthesis, while in OPTs, it was mainly resulted from the increased amount of N uptake after anthesis. Furthermore, the N use efficiency of rice was decreased with increasing application levels of N fertilizer. Compared with FFP treatment, partial factor productivity of N (PFP_N), N agronomy efficiency (AE_N) and N recovery efficiency (RE_N) in OPTs were increased by 99.4%, 137.6% and 70%, respectively, and the N use efficiency was further increased in OPT1 treatment. Additionally, the results also showed that the rice yield contribution stages of different N managements, high yield and high efficiency stage from FFP to OPTs was more effective for improving yield than low yield and low efficiency stage from CK to FFP.

Conclusions The optimum N managements can be coordinated to achieve high yield and high N use efficiency of rice plants by regulated the amount of N fertilizer with side dressing and substituted chemical fertilizer-N through organic fertilizer-N.