Green manuring is a potential method for reducing chemical N fertilizer input and reactive N losses to the environment that can achieve dual goals of food security and green development of agriculture. In this study, three long-term field experiments in south China and process-based model (WHCNS_LIXIM model) were combined to examine effects of long-term green manure substitution on soil N supply, crop N uptake, and N losses in rice production system. Results showed that green manure substitution significantly improves soil mineral N by 27.4–94.9% and 1.0–45.0% at jointing and maturity stages, respectively, compared with the use of chemical N alone. Moreover, 18.4–62.2% of crop N uptake increased at the maturity stage at Xinyang and Yujiang sites. The model demonstrated satisfactory performance in modeling dynamics of soil mineral N, crop N uptake, crop dry matter, and grain yield of the rice–green manure system with a determination coefficient (R2) > 0.967 and slope of linear regression equation (β) close to 1.0. Simulated N budgets presented that green manure reduces 20–40% of the chemical N fertilizer input and decreased 5.5%, 25.1%, and 30.6% of N losses by ammonia volatilization, leaching, and runoff, respectively, when green manure application rates (fresh matter) are less than 30 t ha−1. However, the total N loss increased with the increase of green manure application amounts when application amounts are larger than 30 t ha−1. These findings indicated that the further reduction of chemical N fertilizer was needed when the application amount of milk vetch was higher than 30 t ha−1. A milk vetch application of 15–30 t ha−1 was recommended to substitute 20–40% of chemical N fertilizer to improve crop N uptake and reduce N losses, and develop green rice production system in the study areas.
Read full abstract