Recovery efficiency and loss of 15N-labelled urea in a rice–soil system in the upper reaches of the Yellow River basin

Chemical N input is essential for high rice yields. However, low recovery efficiency of chemical fertilizer N with flooding irrigation in the anthropogenic-alluvial soil resulted in N lost from the rice field in Ningxia Irrigation Region in the upper reaches of the Yellow River. With the technique of stable isotope 15N-traced urea, we conducted two years experiment to estimate the recovery efficiency and loss of applied chemical fertilizer in a rice field. The three fertilizer N treatments included 300 kg ha−1 (N300, the conventional application rate), 240 kg ha−1 (the optimized N application amount, N240) and no N fertilizer application treatment (CK). We estimated the recovery of 15N-labelled urea in grain, straw and root of rice and residues in soil profile. The 15N not accounted for in the plant and soil was presumably lost. The results showed that more efficient use of N fertilizer could allow current N application rates to reduce by 20%. This would still maintain crop yields while substantially reducing N losses to the environment. The high N fertilizer application increased the N uptake by rice derived from fertilizer, and the amount of N rice taken up from soil reduced correspondingly, which resulted in the higher N surplus in soil. Under the conventional irrigation and fertilizer management level, the recovery rate of 15N-labelled urea in rice–soil system was about 48–49%. The 15N-labelled fertilizer recovery in rice plant (Ndff) was 26–30%. In the paddy soil profiles of 0–90 cm, the residual of 15N-labelled fertilizer in soil (Ndfs) were 54–70 kg ha−1, and N residual rate in soils were 18–23%. The annual N loss from the rice field in the Irrigation Region was 28,865 tons. The distribution of 15N abundance variability in different soil profile indicated that fertilizer N leached into the deep soil layers along with irrigation water as a result of continuous yearly rice planting. Optimization of nitrogen fertilizer can significantly reduce the amount of N residuals and N loss from the paddy field. Compared with N300, optimized nitrogen fertilizer application could decrease the loss of fertilizer N by 22–34 kg ha−1, and reduce the amount of N surplus by 26–33 kg ha−1 while the dry matter of rice increased 8–15% and N uptake by rice increased 2–6%. Considering the high food production and the minimum environmental threat, we should fully take into account the optimization application by reducing fertilizer N inputs. However, the interaction between irrigation management and N application rate on N use efficiency in alkaline anthropogenic-alluvial soil needs to be further studied.