Objectives The main pathways of controlled-release urea in decreasing ammonia volatilization in paddy fields, and the application dosage for the effectiveness were studied, aiming for the best environmental returns of controlled-release urea.

Methods Pot experiments were conducted in the greenhouse of Hunan Agricultural University. The tested rice cultivars were Zhongzao39 and Taiyou390, the controlled-release fertilzier (CRU) was resin coated urea, and the tested soil was tide sandy mud paddy soil. There were six treatments, including no nitrogen control (CK), commercial urea (U), CRU treatments with N equal to urea (CRU1), 10% less (CRU2), 20% less (CRU3) and 30% less (CRU4). 60% of nitrogen were basal applied before transplanting and 40% were top dressed at 10 days after transplanting. All the pots stood in isolated glass containers, and ammonia volatilization was monitored by continuous measuring intermittent airflow. The NH₄⁺-N, NO₃^–-N concentrations and pH in the field surface water and the soil temperature at 10 cm depth were measured at the same time.

Results Application of CRU in rice fields significantly reduced ammonia volatilization loss in paddy fields. Among the nitrogen treatments, the cumulative ammonia volatilization loss of CRU3 in rice was the lowest, followed by CRU4, CRU2, CRU1 and U, respectively. Compared with U treatment, CRU significantly reduced the peak rate of ammonia volatilization, and reduced the cumulative ammonia volatilization loss by 50.3%–70.1%. The loss rate of ammonia volatilization in CRU treatment was 5.6%–8.13%, and the ammonia volatilization loss rates in CRU3 and CRU4 in early and late rice were lower than those in the others. Compared with U treatment, the peak NH₄⁺-N concentration in CRU were decreased by 74.5%–80.4% and 53.4%–76.0% after basal fertilization in early and late rice, and decreased by 69.5%–89.1% and 67.3%–80.3% after topdressing. The average pH in surface water of paddy field in U, CRU1, CRU2, CRU3 and CRU4 were 7.26, 7.22, 7.25, 7.32 and 7.14 in early rice, and there was no significant difference among treatments. The pH in late rice were averaged 7.85, 7.71, 7.72, 7.72 and 7.66, and pH in CRU was significantly lower than that in urea treatment. The ammonia volatilization rate of U treatment showed a very significant and positive correlation with NH₄⁺-N concentration (r = 0.8813), a significant negative correlation with NO₃^–-N (r = –0.5319). The ammonia volatilization rates in CRU3 and CRU4 were significantly and positively correlated with the NH₄⁺-N concentration of surface water (r = 0.5388, 0.4245), and each treatment was not significantly correlated with water and soil temperature.

Conclusions Application of controlled-release urea in rice fields can significantly reduce the NH₄⁺-N concentration and lessen the pH increase caused by fertilization in the surface water, therefore, could significantly reduce the loss of ammonia volatilization. Decreasing 20%–30% of controlled-release urea input could bring the best environmental effect in both early and late rice.