Predictive Modeling Study on the Critical Nitrogen Concentration and Nitrogen Accumulation in Cut Chrysanthemum Based on the Cumulative Photo-Thermal Effect

Abstract

Background and aims Critical nitrogen (N) concentration (Nc) and critical accumulation (Na) are essential for N status diagnosis and precise N fertilization in crops. However, efficient prediction models for Nc and Na in cut Chrysanthemum remains scarce, limiting precision N management. Methods Five experiments with varying N gradients were conducted from May 2021 to August 2022 using the ‘Nannong Xiaojinxing’ cultivar. We developed and validated dry matter prediction models with various growth and developmental driver variables, established Nc and Na models using dry matter as model driving variable, and created Nc and Na models using optimal driving variable identified from dry matter predictions. Results Among the dry matter prediction models for cut Chrysanthemum, the model incorporating cumulative photo-thermal effect (PTE) demonstrated superior accuracy and stability. We established the Nc and Na models using dry matter as the driving variable. When the above-ground dry matter was 1 g·plant− 1, the Nc and Na were 4.5295% and 45.30 mg·plant− 1, respectively. At the flower picking stage, the Na reached 236.50 mg·plant− 1. The PTE-driven Nc and Na prediction models demonstrated high accuracy, with R2 at 0.9687 and 1.0019, RMSEs at 0.2105% and 17.47 mg·plant− 1, and n-RMSEs at 7.31% and 12.72%, respectively. Conclusions These models can dynamically predict Nc and Na based on light and temperature factors, providing a scientific basis for efficient N diagnostics and precise N fertilizer management for cut chrysanthemum. Moreover, the methodology developed herein could be extrapolated to other crops, contributing to sustainable agriculture and mitigating excessive N fertilizer application.