Establishment and topdressing validation of a universal critical phosphorus concentration dilution curve for drip-irrigated spring maize in northern Xinjiang of China

Excessive phosphorus (P) fertilizer application has often occurred in recent years, leading to low utilization efficiency in modern agriculture. It is still lacking in the P nutrition diagnosis for the drip-irrigated spring maize in northern Xinjiang, China. In this study, a universal critical phosphorus concentration dilution curve (CPDC) was constructed, suitable for this region. A series of field experiments was also carried out to verify the performance after precise P topdressing. Multi-site field experiments were conducted in the typical drip-irrigated maize areas of northern Xinjiang (Baiyang, Huyanghe, and Shihezi) in 2023. Some treatments were designed with the different maize varieties and P application rates. Intensive sampling was performed to collect the data on the aboveground biomass and P concentration at the various growth stages. A universal CPDC was successfully established using the Bayesian hierarchical model. The results demonstrated that the universal CPDC accurately quantified the dynamic dilution relationship between the maize P concentration and aboveground biomass, with strong parameter stability and a concentrated 95% confidence interval. The verification experiment indicated that the excellent accuracy of the prediction was achieved in 2024, with a root mean square error (RMSE) of 0.052 and a normalized root mean square error (n-RMSE) of 11.0% between the measured and predicted P concentrations. The slope of the scatter trend line was 1.02 with low dispersion, indicating a reliable estimation. At the same time, the optimal topdressing was found in the PR1.2 treatment. The high-yield level was achieved, while the P fertilizer application rate was reduced by 13.1%, compared with the CF. Meanwhile, the maize yield was not significantly different from that of the conventional fertilization. The partial factor productivity and utilization efficiency of the P fertilizer in the PR1.2 treatment increased by 13.8% and 22.1%, respectively, compared with the CF. Additionally, the treatment effectively promoted the accumulation of the aboveground dry matter in maize. The biomass at maturity also shared the minimal difference from that of conventional fertilization. The universal CPDC also quantified the parameter uncertainty using the Bayesian hierarchical model. The specific variety or regional data was independent of the cultivation patterns of the drip irrigation. The real-time diagnosis of the P nutrition and precise topdressing was realized in the entire growth period of maize. The risk of the P loss was reduced for the national food security. The finding can provide the technical support and theoretical reference for the green and efficient production of the drip-irrigated spring maize and the optimal management of the P fertilizer in northern Xinjiang. The promising potential can also lead to the widespread application in modern agriculture.