Return
Construction of a Nitrogen Fertilizer Decision-Making of Winter Wheat Based on Hyperspectral Estimation of Critical Nitrogen Concentration
M
L
B
X
W
X
DOI:10.3390/agronomy16161545.png)
Abstract
En 中文
Accurately assessing crop nitrogen deficiency could improve nitrogen use efficiency and minimize the environmental issues caused by excessive nitrogen fertilizer application. The critical nitrogen dilution curve and nitrogen nutrition index (NNI) are widely used to de-scribe the nitrogen status of crops during growth. Hyperspectral estimation research has mostly focused on the vegetative growth stage, with the reproductive growth stage, in which nitrogen accumulation and transfer directly affect grain yield, receiving less attention. This study combined hyperspectral technology to build a critical nitrogen concentration model for winter wheat during its vegetative and reproductive growth stages and diagnose its nitrogen nutritional status. The results show that the optimal estimation models for leaf nitrogen concentration (LNC) and leaf dry matter (LDM) were both competitive adaptive reweighted sampling–random forest (CARS-RF), with R2 values of 0.8475 and 0.7665 on the validation set. By combining the quadratic curve relationship between the NNI and relative yield, a precise nitrogen application decision model was obtained: nitrogen rate NR = (a LDM−b−LNC) × LDM/NUE, where NUE is the nitrogen use efficiency. Thus, a decision-making model for determining the nitrogen application rate for winter wheat based on UAV hyperspectral data was realized. The critical nitrogen concentration model of winter wheat constructed in this study covers the vegetative and reproductive growth stages, providing a reference for accurate decision-making regarding the use of nitrogen fertilizer in winter wheat production.
Keywords:
winter wheat
hyperspectral
critical nitrogen concentration model
nitrogen fertilizer decision-making
Journal
A
IF:
3.4
Papers:
1.7W
Citations:
5.0W
