Objectives The objective of the study was to clarify the effect of nitrogen (N) application on the tuber yield of sweetpotato and the influencing factors at the national scale, so as to provide references for nitrogen nutrient management in sweetpotato production.

Methods The published litterateurs were collected in China National Knowledge Infrastructure, Wanfang Data, China Science and Technology Journal Database, and Web of Science, using key words “sweet potato”, “N”, “nitrogen fertilizer”, “yield”, “tuber number per plant”, and “single tuber weight”. The literatures were screened using the criteria as: field experiment in China; containing no N control and N fertilization treatment, and all the control and treatments were applied with the same amount of phosphorous (P) and potassium (K) fertilizer; and each treatment had at least three replicates. A total of 45 literatures were obtained, containing 288 groups of yield data,191 groups of tuber number per plant, and 145 groups of single tuber weight. Meta-analysis was used to quantitatively analyze the effect of N application on the tuber yield, yield components, and the influencing factors.

Results N application increased the fresh tuber yield and single tuber weight of sweetpotato by 1.7% and 3.2%, respectively, but reduced the number of tubers per plant (NTP) by 1.2%. N application rate caused significant differences in fresh tuber yield and yield components, generally showing a decreased fresh tuber yield and NTP increment with the increasing of N application rate, and the peak increments were all recorded at N application rate <75 kg/hm². The single tuber weight showed a trend of first increasing and then decreasing, with the highest appeared at N application rate of 75−150 kg/hm². The increasing rates of fresh tuber yield were higher in the Yangtze River Basin region and the southern regions than in the northern regions. The suitable N application rates for the northern, Yangtze River Basin, and southern regions were <75 kg/hm², 75−150 kg/hm², and 75−150 kg/hm², respectively. The levels of tuber yield in the controls negatively impacted the increasing rates of N fertilizer on fresh tuber yield and NTPs. The suitable N application rate was 75−150 kg/hm² under control yield level of ≤25 t/hm², and <75 kg/hm² under control yield level of 25−35 t/hm² and>35 t/hm². The yield response of fresh sweetpotato to N application was lower than that of starch sweetpotato, and the suitable N application rates for fresh and starch sweet potatoes were <75 kg/hm² and 75−150 kg/hm², respectively. The high increment on yield (10.5%) and single tuber weight (14.9%) by N fertilizer were recorded at K₂O application rate 150−225 kg/hm², although the NTP was significantly decreased by 6.3%. And the peak increment of fresh tuber yield, NTP, and single tuber weight by N fertilizer were 7.2%, 4.9%, and 5.8%, respectively, under P₂O₅ application rate ≤60 kg/hm². The appropriate combination application ratio of N, P₂O₅, and K₂O was 1∶0−0.8∶1−3. In the soils with 6.5<pH≤7.5, low available P (≤ 10 mg/kg), low available N (≤60 mg/kg), medium available K (50−100 mg/kg), or high soil organic matter (>20 g/kg), N application produced the highest fresh tuber yield increases, compared with the no N application controls.

Conclusions The effect of N application on the fresh tuber yield of sweetpotato is assured but varied in China, depending on the regions, the levels of tuber yield in the control, P, K application rates, the soil basic fertility, and the cultivar type of sweet potatoes. Therefore, N management should be made accordingly across China.