Objectives This study aimed to investigate the agronomic and environmental impacts of coated controlled-release nitrogen fertilizer and matrix slow-release nitrogen fertilizer in rice cultivation.

Methods A field experiment was conducted on late-season rice in Jiujiang City, Jiangxi Province, including four treatments: conventional urea with one-time basal and one-time topdressing (U), single application of coated controlled-release nitrogen fertilizer (P), single application of matrix slow-release nitrogen fertilizer (M), and no nitrogen control (CK). following each nitrogen fertilizer application, ammonia volatilization and nitrous oxide emission fluxes were continuously monitored to calculate total gaseous nitrogen losses. Leachate samples were collected every 10 days after transplanting, and surface runoff samples were collected after each rainfall event to determine ammonium and nitrate concentrations, enabling the estimation of nitrogen losses via leaching and runoff. Soil samples from the 0−20 cm layer were collected at five key growth stages to determine ammonium and nitrate nitrogen content. Concurrently, plant samples were taken to assess biomass and nitrogen uptake. At maturity, rice yield and its contributing components were investigated.

Results  Rice yields under P and M treatments increased by 29.3% and 30.2% compared with U, respectively, with nitrogen use efficiency improved by 20.6 and 17.8 percentage points. Compared to U, the P treatment significantly reduced ammonium nitrogen content in paddy fields during the seedling stage, and increased nitrogen accumulation in rice during the tillering−jointing and jointing−heading stages by 12.3% and 39.9%, respectively. It also markedly reduced ammonia volatilization by 33.4%, leaching losses by 12.3%, and runoff losses by 44.5%, resulting in a 33.6% reduction in total reactive nitrogen losses. In contrast, the M treatment increased nitrogen uptake by 109.3% during tillering−jointing stages and enhanced early nitrogen accumulation. It also reduced leaching losses by 16.5%, but increased runoff losses by 74.6% and total reactive nitrogen losses by 16.8%, compared to U.

Conclusion A single application of coated controlled-release nitrogen fertilizer with an approximately 90 days release duration significantly reduced ammonium nitrogen content during the rice seedling stage and enhanced nitrogen uptake during the tillering to heading stage. This shift in nitrogen supply timing from earlier to later stages increased rice yield and nitrogen use efficiency, and reduced in total reactive nitrogen losses, thus offering positive environmental benefits. In contrast, a single application of matrix-based slow-release nitrogen fertilizer increased ammonium nitrogen content in paddy soil during the seedling stage and significantly boosted nitrogen uptake during the tillering to jointing stage, but had no significant effect on nitrogen uptake during the jointing to heading stage. Consequently, matrix slow-release fertilizers exhibit good yield-increasing effects but require improvement in their environmental performance.