Rice yield estimation is vital for enhancing food security, optimizing agricultural management, and promoting sustainable development. However, traditional satellite/aerial and ground-based/tower-based platforms face limitations in rice yield estimation, and few studies have explored the potential of UAV-borne hyperspectral remote sensing for this purpose. In this study, we employed a UAV-borne push-broom hyperspectral camera to acquire remote sensing data of rice fields during the filling stage, and the machine learning regression algorithms were applied to rice yield estimation. The research comprised three parts: hyperspectral data preprocessing, spectral feature extraction, and model construction. To begin, the preprocessing of hyperspectral data involved geometric distortion correction, relative radiometric calibration, and rice canopy mask construction. Challenges in geometric distortion correction were addressed by tracking linear features during flight and applying a single-line correction method. Additionally, the NIR reflectance threshold method was applied for rice canopy mask construction, which was subsequently utilized for average reflectance extraction. Then, spectral feature extraction was carried out to reduce multicollinearity in the hyperspectral data. Recursive feature elimination (RFE) was then employed to identify the optimal feature set for model performance. Finally, six machine learning regression models (SVR, RFR, AdaBoost, XGBoost, Ridge, and PLSR) were used for rice yield estimation, achieving significant results. PLSR showed the best R² of 0.827 with selected features, while XGBoost had the best R² of 0.827 with full features. In addition, the spatial distribution of absolute error in rice yield estimation was assessed. The results suggested that this UAV-borne imaging hyperspectral-based approach held great potential for crop yield estimation, not only for rice but also for other crops.

Echinochloa crus-galli (L.) P. Beauv (common name: barnyard grass) is a major weed in rice-growing areas and has evolved resistance to multiple herbicides. Florpyrauxifen-benzyl (trade name Rinskor) is a novel synthetic auxin herbicide that was approved in China in 2017 and is widely used in rice production to control resistant weeds, including barnyard grass. We identified a florpyrauxifen-benzyl-resistant E. crus-galli biotype with a resistance index (RI) of 11.89 using screen house herbicide experiments. To understand the phytotoxicity mechanisms of florpyrauxifen-benzyl, we used transcriptomics technologies to compare the gene expression profiles of florpyrauxifen-benzyl treatment on phytohormone transduction between florpyrauxifen-benzyl-resistant and -susceptible barnyard grasses (Echinochloa crus-galli (L.) P. Beauv). A total of 1810 DEGs were identified in the S comparison setting (FTS vs. UTS), and 915 DEGs were identified in the R comparison setting (FTR vs. UTR); 464 genes overlapped between the two comparison groups. Approximately sixty-nine hormone-related DEGs were detected after treatment with florpyrauxifen-benzyl in both R and S biotypes. At 24 h after florpyrauxifen-benzyl treatment, compared with the R biotype, the S biotype showed a stronger auxin response and higher expression of related genes involved in ethylene and abscisic acid biosynthesis and signal transduction. In addition, a brassinolide receptor gene was upregulated after florpyrauxifen-benzyl treatment and had higher expression in the S biotype than in the R biotype. This study is the first transcriptome analysis of the differential effects of florpyrauxifen-benzyl treatment between florpyrauxifen-benzyl-resistant and -susceptible E. crus-galli. It reflects the difference in phytohormone biosynthesis and signal transduction between R and S barnyard grasses in response to florpyrauxifen-benzyl treatment and will be helpful for understanding the phytotoxicity mechanisms of florpyrauxifen-benzyl.

Roegneria kamoji Ohwi (Poaceae), a wild relative plant of wheat which is widely distributed across China, has become a dominant and problematic weed in wheat fields in some regions. We have previously confirmed that R. kamoji is highly tolerant to foliar-applied acetyl-CoA carboxylase (ACCase) and acetolactate synthase (ALS) inhibitors in wheat (Triticum aestivum L.). The sensitivity of R. kamoji to pre-emergence (PRE) herbicides and the basis of fenclorim increase selectivity to butachlor between wheat and R. kamoji were evaluated in this study. Screenhouse bioassay showed that R. kamoji exhibited similar sensitivity to wheat to PRE herbicides at their recommended field doses (RFD); it also showed that buatchlor provides the highest relative control for R. kamoji (53.4% emergence and 81.5% fresh weight reduction, respectively), while it had no impact on seedling emergence of wheat among the six PRE herbicides. When butachlor was applied at four-fold RFD, no R. kamoji seedlings emerged; however, it significantly reduced the above-ground biomass of wheat compared with the non-treated control. Pre-treatment with herbicide safener fenclorim by seed soaking increased the ED₁₀ value of butachlor to wheat from 221.8 to 1600.1 g a.i. ha⁻¹, thus increasing the selectivity index from 9.6 to 68.9 between wheat and R. kamoji. The activities of α-amylase activity and protein content during germination, and glutathione-S-transferase (GST) and β-ketoacyl-CoA synthase (KCS) in the seedlings, could be induced by butachlor in both wheat seeds with or without fenclorim pre-soaking. These results suggested that butachlor provides the highest control for R. kamoji and did not affect germination and emergence in wheat. The basis of fenclorim-increased selectivity to butachlor was associated with the induced GST and KCS-mediated enhanced herbicide metabolism in wheat.

Penoxsulam is an acetolactate synthase (ALS)-inhibiting herbicide usually applied by post-emergence foliar spraying (PFS) for the control of Echinochloa crus-galli and numerous annual weeds in paddy fields. Herbicides applied by foliar spraying can have negative impacts on the environment, ecosystems, and human health. In this study, the response of E. crus-galli and rice to the PFS and post-emergence water-dispersal (PWD) applications of penoxsulam, and the differences in the detoxification displayed by them between the two treatment methods were compared. The results showed that the PWD application of penoxsulam provides a similar control efficacy against E. crus-galli as PFS at the 1-, 3-, and 5-leaf stages. Meanwhile, the PWD application had a higher safety for the rice. After being treated with 30 g a.i. ha⁻¹ penoxsulam, residues were not detected in the rice treated by the PWD application method, whereas, with the PFS treatment, there was 59.0 µg/kg penoxsulam remaining. With the PFS application, there were many more residues of penoxsulam in the E. crus-galli than with the PWD method; the amount of residues was 32-fold higher 12 h after treatment. The in vitro enzyme activity assays revealed that the activities of ALS, glutathione-S-transferase (GST), and cytochrome P450 monooxygenases (P450) were increased in the PWD treatments, and were 1.5-, 1.3-, and 2.3-fold higher than with PFS 72 h after treatment. The real-time quantitative PCR (qRT-PCR) revealed that the GST1 and P450 genes, CYP81A14, CYP81A12, CYP81A18, and CYP81A21 were upregulated with the PWD application versus PFS in the E. crus-galli. In summary, these results demonstrate that the herbicidal activity was not affected by the upregulation of target and metabolic enzyme activities with the PWD application of penoxsulam. This research could contribute to application strategies reducing the risk of rice injury and environmental impacts by using water-dispersal formulations of penoxsulam.