Abstract: Nitrogen (N) leaching is a major pathway of N losses in paddy fields. Here, an experiment was conducted to assess the effects of the reduction of controlled-release urea application on grain yield and N leaching in a double-cropping paddy field. Fertilization treatments included zero-N (CK, control, 0 kg N·hm^-2), conventional urea (CU, 180 kg N·hm^-2), and four polymer-coated urea fertilization levels, i.e., 1.0CRU, 0.9CRU, 0.8CRU, 0.7CRU, which represented 0, 10%, 20% and 30% reduction of fertilizer inputs relative to CU, respectively. Leachate was collected at the soil depth of 60 cm using field leakage pool method. Nitrogen leaching peaked shortly after fertilization, implying that measures should be taken to prevent N leaching in the early period. Nitrogen losses from leaching were 42.3 kg N·hm^-2 for 0.8CRU, and by 37.7 kg N·hm^-2 for 0.7CRU, significantly lower than the leaching in CU (53.9 kg N·hm^-2). Nitrogen leaching in 0.7CRU was significantly lower than that in 1.0CRU (51.3 kg N·hm^-2). 11.9%-13.5% of the fertilizer N was lost via leaching across the six treatments with comparable N loss rates. Rice yields, N utilization efficiency and N harvest index were significantly higher in 0.8CRU and 0.7CRU relative to CU. Our findings suggested that the use of CRU would permit a reduction in N application by 20%-30%, which could maintain the rice yield and obtain a reduction in N leaching.