Objectives The availability of four foliar nutrient diagnosis approaches in litchi (Litchi chinensis Sonn.) was compared. The proper diagnosis method was chosen and the ranges of nutrient in litchi leaves were proposed.

Methods Feizixiao is the most widely cultivated litchi cultivar in China. Twenty-two typical Feizixiao orchards, located in the main production regions of South China, were chosen in this study. The collected leaf sample number was 193 at the fruit swelling stage (FSS) of 2016, 186 at the last autumn shoot maturing stage (LASMS) of 2016 and 159 at FSS of 2017, respectively. Foliar nutrient (N, P, K, Ca, Mg, S, Zn and B) concentrations of all samples were detected. The fruit yield was recorded for each tree at harvest stage in 2016 and 2017, and the fruit yield in each orchard was calculated by multiplying fruit yield per tree and plantation density in the orchard. The high yield population for nutrient diagnosis in both years was determined by squared Euclidean distance. The relationship between fruit yield and leaf nutrient concentration was fitted. Four methods including the critical value approach (CVA), sufficiency range approach (SRA), modified diagnosis and recommendation integrated system (M-DRIS), and compositional nutrient diagnosis (CND) were used to evaluate the foliar nutrient status of litchi at the three stages.

Results Alternative fruit bearing was commonly observed in most sampled orchards. The maximum concentrations of foliar nutrients were 2-fold to more than 10-fold higher than the minimum levels in the sampled trees at the same growth stage. Significant difference was observed in foliar nutrient contents at the FSS between 2016 and 2017, however, the high yield population in both years was characterized by relatively constant foliar nutrient concentrations. Based on the relationship between fruit yield and foliar nutrient concentration, diagnosis indices of a few foliar nutrients were obtained by CVA. The necessity of foliar nutrients was qualitatively instead of quantitatively evaluated by M-DRIS and CND, moreover, the diagnosis accuracy for nutrients was related with the diagnosis stage and the severity of nutrient deficiency or abundance, with significant annual variation. In contrast, diagnosis indices for all nutrients could be calculated by SRA, and maintained relatively stable at FSS of both 2016 and 2017, regardless of the annual variation of fruit yields.

Conclusions SRA is accepted to compute the foliar nutrient diagnosis norms for litchi due to its small annual variation, universal diagnosis index and practicability. According to the SRA, the optimal foliar nutrient contents for Feizixiao are recommended as N 16.7–19.2 g/kg, P 1.06–1.25 g/kg, K 5.1–6.7 g/kg, Ca 7.7–11.0 g/kg, Mg 2.5–3.7 g/kg, S 1.51–1.81 g/kg, Zn 19.6–32.6 mg/kg, B 11.5–19.2 mg/kg at FSS, and N 19.7–22.0 g/kg, P 1.69–1.95 g/kg, K 10.8–12.7 g/kg, Ca 3.0–4.1 g/kg, Mg 2.5–2.9 g/kg, S 1.38–1.57 g/kg, Zn 15.0–18.9 mg/kg, B 10.8–16.8 mg/kg at LASMS.