Objectives Excessive nitrogen fertilizer application, the unreasonable mixing ratio of organic and inorganic fertilizers, and frequent irrigation are common in facility vegetable production. Here, we proposed three N application modes and studied their effects on tomato yield, soil N residue, and N balance. The results will provide a scientific basis for efficient tomato production and optimized N fertilization.

Methods The experiment was carried out in a vegetable greenhouse in Huimin, Shandong Province. We irrigated the greenhouse at 80% (390 mm) of the average irrigation (482.5 mm) in the study area. Tomato was cultivated with film mulching. Using the basal application of pig manure at N 65 kg/hm², the conventional nitrogen application rate (N 1000 kg/hm2, TF) and three treatments with 50% nitrogen reduction were set: chicken manure treatment (OF), ordinary urea treatment (CF) and coated urea treatment (CRF). Soil samples were collected at 0–180 cm depth before transplanting and after harvest. Before each fertilization, soil samples were collected to a depth of 0–90 cm (at 30 cm interval) to determine nitrate and ammonium nitrogen contents. After tomato harvest, plant and fruit samples were collected to determine biomass and nutrient content.

Results There were no significant differences in tomato yield and N uptake among the treatments. However, the N rate was one fold higher in TF than in CF treatment. CF, OF and CRF treatments maintain appropriate soil N supply levels during the growing period. Correspondingly, the residual NO₃^–-N amount in 0–60 cm soil layer of CF, OF, and CRF treatments were 190.1, 227.2 and 310.5 kg/hm². These values were 56.61%, 43.35%, and 22.59% lower than those recorded in TF treatment. The apparent N loss in CF, OF, and CRF treatment were 416.6, 443.7, and 352.3 kg/hm², 45.72%, 42.20% and 54.10% lower than those in TF treatment, respectively. Compared with TF, soil N balance surplus in CF, OF, and CRF (P < 0.05) decreased by 34.26%, 33.40%, and 61.78%, respectively. Coated urea maintained more N in the upper soil layer than the other treatments, and its NO₃^–-N accumulation in 0–30 cm soil layer was 43.0% higher than that in CF treatment. We found two obvious accumulation processes of NO₃^–-N in the soil surface layer during the growing period of tomato. The first was during the 20–60 days period, and the second occurred during the 110–120 day period after transplanting. Further, there was a depletion process in the 30–60 cm soil layer during the 80–100 day period.

Conclusions Under the experimental conditions, reducing half of the conventional N application rate for tomato can effectively reduce the NO₃^–-N residue and leaching in the soil of greenhouse tomato, and will not cause a decrease in yield. The application of coated urea is more beneficial to reduce the leaching of N than poultry manure or ordinary urea, and more beneficial to achieve a win-win situation of yield and environmental benefits.