Accurate and timely monitoring of biochemical and biophysical traits associated with crop growth is essential for indicating crop growth status and yield prediction for precise field management. This study evaluated the application of three combinations of feature selection and machine learning regression techniques based on unmanned aerial vehicle (UAV) multispectral images for estimating the bio-parameters, including leaf area index (LAI), leaf chlorophyll content (LCC), and canopy chlorophyll content (CCC), at key growth stages of winter wheat. The performance of Support Vector Regression (SVR) in combination with Sequential Forward Selection (SFS) for the bio-parameters estimation was compared with that of Least Absolute Shrinkage and Selection Operator (LASSO) regression and Random Forest (RF) regression with internal feature selectors. A consumer-grade multispectral UAV was used to conduct four flight campaigns over a split-plot experimental field with various nitrogen fertilizer treatments during a growing season of winter wheat. Eighteen spectral variables were used as the input candidates for analyses against the three bio-parameters at four growth stages. Compared to LASSO and RF internal feature selectors, the SFS algorithm selects the least input variables for each crop bio-parameter model, which can reduce data redundancy while improving model efficiency. The results of the SFS-SVR method show better accuracy and robustness in predicting winter wheat bio-parameter traits during the four growth stages. The regression model developed based on SFS-SVR for LAI, LCC, and CCC, had the best predictive accuracy in terms of coefficients of determination (R²), root mean square error (RMSE) and relative predictive deviation (RPD) of 0.967, 0.225 and 4.905 at the early filling stage, 0.912, 2.711 μg/cm² and 2.872 at the heading stage, and 0.968, 0.147 g/m² and 5.279 at the booting stage, respectively. Furthermore, the spatial distributions in the retrieved winter wheat bio-parameter maps accurately depicted the application of the fertilization treatments across the experimental field, and further statistical analysis revealed the variations in the bio-parameters and yield under different nitrogen fertilization treatments. This study provides a reference for monitoring and estimating winter wheat bio-parameters based on UAV multispectral imagery during specific crop phenology periods.

Nitrogen (N) is an essential nutrient element for crop productivity. Unfortunately, the nitrogen use efficiency (NUE) of crop plants gradually decreases with the increase of the N application rate. Nevertheless, little has been known about the molecular mechanisms of differences in NUE among genotypes of wheat. In this study, we used RNA-Sequencing (RNA-Seq) to compare the transcriptome profiling of flag leaves at the stage of anthesis in wheat NILs (1Y, high-NUE, and 1W, low-NUE) under normal nitrogen conditions (300 kg N ha⁻¹, corresponding to 1.6 g N pot⁻¹). We identified 7023 DEGs (4738 upregulated and 2285 downregulated) in the comparison between lines 1Y and 1W. The responses of 1Y and 1W to normal N differed in the transcriptional regulatory mechanisms. Several genes belonging to the GS and GOGAT gene families were upregulated in 1Y compared with 1W, and the enhanced carbon metabolism might lead 1Y to produce more C skeletons, metabolic energy, and reductants for nitrogen metabolism. A subset of transcription factors (TFs) family members, such as ERF, WRKY, NAC, and MYB, were also identified. Collectively, these identified candidate genes provided new information for a further understanding of the genotypic difference in NUE.

The development of crop cultivars with high nitrogen use efficiency (NUE) under low-N fertilizer inputs is imperative for sustainable agriculture. However, there has been little research on the molecular mechanisms underlying enhanced resilience to low N in high-NUE plants. The comparison of the transcriptional responses of genotypes contrasting for NUE will facilitate an understanding of the key molecular mechanism of wheat resilience to low-N stress. In the current study, the RNA sequencing (RNA-seq) technique was employed to investigate the genotypic difference in response to N deficiency between two wheat NILs (1Y, high-NUE, and 1W, low-NUE). In our research, high- and low-NUE wheat NILs showed different patterns of gene expression under N-deficient conditions, and these N-responsive genes were classified into two major classes, including “frontloaded genes” and “relatively upregulated genes”. In total, 103 and 45 genes were identified as frontloaded genes in high-NUE and low-NUE wheat, respectively. In summary, our study might provide potential directions for further understanding the molecular mechanism of high-NUE genotypes adapting to low-N stress.

Quantifying the vertical distribution of leaf chlorophyll content (LCC) is integral for a comprehensive understanding of the physiological status and function of winter wheat crops, having significant implications for crop management and yield optimization. In this study, we investigated the vertical LCC trait of winter wheat during two consecutive field growth seasons using proximal multispectral imaging measurements to evaluate vertical variations of LCC within winter wheat canopies. The results revealed the non-uniform vertical LCC distribution varied across the entire growth season. The effects of nitrogen fertilization rate on LCC among vertical layers increased gradually from upper to lower layers of canopy. To enhance LCC prediction accuracy, this study proposes a deep transfer learning network model for leaf trait estimation (LeafTNet). It integrates the advantages of physical radiative transfer simulations with deep neural network through transfer learning. The results demonstrate that the LeafTNet achieved remarkable predictive performance and strong robustness. Furthermore, the proposed method exhibits superior estimation accuracy compared to empirical statistical method and traditional machine learning method. This study highlights the performance of LeafTNet in accurately and efficiently quantifying LCC from proximal multispectral data, which provide technical support for the estimation of the vertical distribution of leaf traits and improve crop management.

Remote sensing has been used as an important means of monitoring crop growth, especially for the monitoring of the formation of crop yield in the middle and late growth period. The information acquisition on the yield formation period of winter wheat is of great significance for winter wheat growth monitoring, yield estimation and scientific management. Hence, the main goal of this study was to verify the possibility of monitoring the grain-filling process of winter wheat and its in-field variability using an alternative non-destructive method based on orbital remote sensing. High-resolution satellite imageries (3 m) were obtained from the PlanetScope platform for three commercial winter wheat fields in Jiangsu Province, China during the reproductive stage of the winter wheat (185–215/193–223/194–224 days after sowing (DAS)). Based on the quantitative analysis of vegetation indices (VIs) obtained from high-resolution satellite imageries and three indicators of the winter wheat grain-filling process, linear, polynomial and logistic growth models were used to establish the relationship between VIs and the three indicators. The research showed a high Pearson correlation (p < 0.001) between winter wheat maturity and most VIs. In the overall model, the remote sensing inversion of the dry thousand-grain weight has the highest accuracy and its R² reaches more than 0.8, which is followed by fresh thousand-grain weight and water content, the accuracies of which are also considerable. The results indicated a great potential to use high-resolution satellite imageries to monitor winter wheat maturity variability in fields and subfields. In addition, the proposed method contributes to monitoring the dynamic spatio-temporality of the grain-filling progression, allowing for more accurate management strategies in regard to winter wheat.

Excessive use of synthetic nitrogen (N) fertilizer and lower nitrogen use efficiency (NUE) are threatening the wheat production in the middle and lower reaches of Yangtze River. Excess input of N fertilizers also results in severe environmental pollution, climate change and biodiversity loss. However, the study on reasonable nitrogen application and NUE improvement with the prerequisite of stable and high yield remains unexplored. In our study, the four different levels of nitrogen were applied to find out the nitrogen threshold which could be both friendly to environment and promise the stable and high yield. The experiment was carried out in Yangzhou University (Yangzhou, China). The wheat cultivar Yangmai 23 was selected as the research material. The four nitrogen levels were as follows: 0, 189, 229.5, and 270 kg ha^-1. The results showed that the grain yield under the application of 229.5 kg ha^-1 N was as high as that under 270 kg ha^-1 N level, with the observation of 20.3% increase in agronomic efficiency. The N₂O emission of 229.5 kg ha^-1 N application was as low as that of 189 kg ha^-1 N, but the grain yield and agronomic efficiency were significantly higher (11.9%) under 229.5 kg ha^-1 treatment than the lower one. Taken together, this indicated the nitrogen level at 229.5 kg ha^-1 could be identified as the fertilizer threshold, which will be beneficial for the future fieldwork.