Drone Multispectral Imaging Captures the Effects of Soil Nmin on Canopy Structure and Nitrogen Use Efficiency in wheat

Abstract

Drone remote sensing has been increasingly demonstrated its unique advantage of monitoring vegetation and agricultural systems, yet the degree to which the effectiveness of drone imaging based spectral features may differ in assessing the effect of soil nutrient availability on plant canopy is unclear. Soil mineralized nitrogen (Nmin) is a key factor determining crop N fertilization demand, and thus understanding its effect on yield and quality is critical for efficient fertilizer use and precise crop N management and assessing nitrogen use efficiency (NUE). In this study, our objective was to understand how soil Nmin variability is associated canopy characteristics, and whether these associations are also interact with grain yield, quality and NUE. For this purpose, we employed UAV multispectral remote sensing, and plant and grain sampling and sample N content determination in wheat growing at different Nmin levels. We extracted 63 spectral features (25 texture features, canopy height, 5 multispectral bands, and 23 spectral indices) obtained from eight UAV flights of RGB and multispectral images. Effective indicators were selected using random forests and correlation, and 10 machine learning models were used to predict NUE. We identified CH, NGRDI, NDREg, and IPCA as the most predictive factors for NUE (r {\textgreater} 0.7), and the PLSR model (R2 {\textgreater} 0.8) as the best performing model. We found that the impact of Nmin on wheat primarily involves enhancing canopy structure, promoting the expansion of leaf area, thereby increasing grain yield and NUE under moderate nitrogen fertilizer levels. In contrast, there was no significant effect found on grain quality. Results also demonstrated that the red edge spectral indices and color-band indices were effective in predicting NUE as early as the flowering stage. This study demonstrates the potential of UAV remote sensing for capturing soil N effect on canopy traits and for assessing nitrogen use efficiency for the purpose of improving fertilizer utilization and environmental effects.

Jie Wang

PhD student

Using remote sensing to study the impact of climate change and cropping practices on global crop yields and nitrogen use efficiency

Prof. Dr. Kang Yu

Professor of Precision Agriculture

My research interests include precision crop farming, hyperspectral remote sensing, and AI in agriculture.