Improving agronomic practices to reduce nitrate leaching from the rice–wheat rotation system

• Wheat–rice rotation with unique dry–wet cycle water regime favors nitrate leaching.
• N leaching occurred mainly at the early stage of growth in rice season.
• N leaching occurred mainly in the mid-late stage of growth in wheat season.
• We tested the effectiveness of an improved practice for mitigating N leaching.
• We tested the effectiveness of an improved practice for increasing grain yield.

Winter wheat–summer rice rotation with unique dry–wet cycle water regime is one of the most general cropping systems in the Yangtze Delta, which might favor NO3−-N leaching into shallow groundwater due to high fertilizer-N input with low N use efficiency. To alleviate the detrimental impacts caused by N leaching, it is necessary to apply improved management practices. Therefore, the objectives of the present study are to investigate characteristics of N leaching under drained upland crop-flooded rice rotation and illustrate the effectiveness of an improved agronomic practice for mitigating NO3−-N leaching. N leaching measured in situ for the control (CK), conventional (CT), and improved (IT) treatments for one rice–wheat rotation. Improved fertilization and crop cultivation managements have been applied to the IT treatment in an integrated manner, including low N fertilization rate, high fertilization frequency, and dense planting. During the rice-growing season, NO3–-N and dissolved organic N (DON) were the predominant forms of N in percolation water, and contributed over 25% and 59% to TN, respectively. NO3–-N and TN concentrations in percolation water and leaching losses did not differ significantly (p > 0.05) between treatments. N leaching occurred mainly during the early stage of rice seedling transplanting. During the wheat-growing season, N in percolation water was predominantly in the form of NO3–-N, and accounted for over 64.5–82.9% of TN. N fertilization significantly increased NO3–-N and TN concentrations in percolation water. Nitrate and TN leaching losses from the IT treatment (1.15 and 1.38 kg N ha−1, respectively) were smaller than the CT treatment (1.34 and 1.76 kg N ha−1, respectively). The majority of N leaching losses for wheat occurred during the period from late February to final harvest. The IT treatment exhibited no significant effect on wheat yield, but significantly (p < 0.05) increased rice yield (8.60 Mg ha−1 compared to 8.06 Mg ha−1) relative to the CT treatment. Compared with the conventional management practice, the improved management practice appeared to be effective in reducing N leaching and increasing grain yield under the rice–wheat rotation.