The nitrogen (N) fertilizers overused or misused are the main contributors for water eutrophication of the Nansi Lake in eastern China. A field experiment with different application levels of controlled-release N fertilizer (CRF) was carried out at a rice field in the Nansi lake basin to provide information on the nitrogen uptake rate and use efficiency by rice with CRF application at different growth stages of rice. The fertilization levels for the controlled fertilizer in this study were 0, 300, 337.5 and 375 kg/hm2, respectively, and 375 kg/hm2 for conventional urea (CU). We estimated the N uptake rate in different growing season and apparent recovery of fertilizer in root, stover and grain of rice. The result showed that grain yield increased by higher N rate. However, the N uptake rate did not increase linearly with the nitrogen application amount. The highest N uptake rate was 22.48 mg/(plant x d) under the fertilization amount of 337.5 kg/hm2 during the young panicle differentiation stage. Apparent N recovery efficiency for CU with traditional application amount of 375 kg/hm2 was 21.86%. Apparent N recovery efficiency for CRF with application amount of 337.5 kg/hm2 was highest with 37.17%. Based on the statistical data, with CU of traditional urea application amount of 375 kg/hm2, nitrogen loss and soil N residue at one growing season are 130.07 x 10(3) t for Jining city and 11.4 x 10(3) t for Yutai county of the Nansi Lake basin. If using CRF with the optimized application amount of 337.5 kg/hm2, nitrogen loss and soil residual at one growing season is 10.46 x 10(4) t for Jining city. It could reduce losses of 2.55 x 10(4) t N for Jining city and 2 235.26 t N for Yutai county per year. Because of releasing patterns more closely matched to crop N uptake patterns, controlled release N fertilizer could be a good way to prevent water eutrophication due to nitrogen fertilizer overused or misused in the Nansi Lake.