Green manure application is a common practice to improve soil fertility in China. However, the impact of different green manure-fertilizer combinations on the soil microbial communities in the low-fertility immature red soil in southern China remains unclear. In this study, we conducted a pot experiment using two common green manure crops, ryegrass (Lolium perenne L.) and Chinese milk vetch (Astragalus sinicus L.), along with a fallow treatment. We also considered three combined fertilizer management strategies, including mineral, humic acid, and organic manure fertilizers. We evaluated the soil microbial biomass, activity, communities, functional prediction and their correlation with soil properties during green manure growth and incorporation periods, to assess the potential alterations caused by different green manure and fertilizer combinations. Our findings indicate that green manure application, particularly in combination with organic fertilizers, increased the alpha diversity of the soil bacterial community, while the opposite trend was observed in the fungal community. The application of green manure altered the soil microbial communities during both growth and incorporation periods, especially the taxa that participate in carbon, nitrogen and sulfur cycles. Notably, ryegrass significantly increased the relative abundance of bacterial phylum Firmicutes and fungal phylum Ascomycota, whereas Chinese milk vetch significantly stimulated the bacterial phylum Acidobacteria and fungal phylum Glomeromycota. Compared with fallow treatments, green manure application significantly increased the soil pH by 4.1%–12.4%, and microbial biomass carbon by 29.8%–72.9%, regardless of the types of combined fertilizer. Additionally, the application of green manure resulted in a 35.6%–142.6% increase in urease activity and a 65.9%–172.9% increase in β-glucosidase activity compared to fallow treatments, while led to a 22.5%–55.6% decrease in catalase activity. Further analysis revealed that the changes in both bacterial and fungal communities positively correlated with soil pH, soil organic matter, total nitrogen and alkali hydrolyzed nitrogen contents. Moreover, the relationship between the soil microbial community and soil enzyme activities was regulated by the specific green manure species. In conclusion, our results provide insight into the effects of different green manure-fertilizer combinations on soil microorganisms and their underlying mechanisms in improving soil fertility in the low-fertility immature red soil.

Milk vetch (Astragalus sinicus L.) is a winter-growing plant that can enhance soil fertility and provide essential nutrients for subsequent season crops. The fertilizing capacity of milk vetch is closely related to its above-ground biomass. Compared to the manual measurement methods of milk vetch biomass, remote sensing-based estimation methods have the advantages of rapid, noninvasive, and large-scale measurement. However, few studies have been conducted on remote sensing-based estimation of milk vetch biomass. To address this shortcoming, this study proposes combining unmanned aerial vehicle (UAV)-based hyperspectral imagery and machine learning algorithms for accurate estimation of milk vetch biomass. Through the analysis of hyperspectral images and feature selection based on the Pearson correlation and principal component analysis, vegetation indices (VIs), including near-infrared reflectance (NIR), red-edge spectral transform index (RE), and difference vegetation index (DVI), are selected as estimation metrics of the model development process. Four machine learning methods, including random forest (RF), multiple linear regression (MLR), deep neural network (DNN), and support vector machine (SVM), are used to construct the biomass models. The results show that the RF estimation model exhibits the highest coefficient of determination (R²) of 0.950 and the lowest relative root-mean-squared error (RRMSE) of 14.86% among all the models. Notably, the DNN model demonstrates promising performance on the test set, with the R² and RRMSE values slightly superior and inferior to those of the RF, respectively. The proposed method based on UAV imagery and machine learning can provide an accurate and reliable large-scale estimation of milk vetch biomass.

Astragalus sinicus is an important winter-growing cover crop. It is widely utilized, not only as a cover crop for its benefits in fertilizing the soil but also as a landscape ground cover plant. Anthocyanins are involved in the pigmentation of plants in leaves and flowers, which is a crucial characteristic trait for A. sinicus. The formation of anthocyanins depends significantly on the enzyme chalcone isomerase (CHI). However, research on the CHI gene of A. sinicus remains unexplored. The rapid amplification of cDNA ends (RACE) approach was used in this research to clone the CHI sequence from A. sinicus (AsiCHI). The expression profiles of the AsiCHI gene in multiple tissues of A. sinicus were subsequently examined by qRT-PCR (Quantitative Real-Time PCR). Furthermore, the function of the AsiCHI was identified by the performance of ectopic expression in Arabidopsis (Arabidopsis thaliana). The outcomes revealed that the full-length cDNA of the AsiCHI gene (GeneBank: OQ870547) measured 972 bp in length and included an open reading frame of 660 bp. The encoded protein contains 219 amino acids with a molecular weight of 24.14 kDa and a theoretical isoelectric point of 5.11. In addition, the remarkable similarity between the AsiCHI protein and the CHI proteins of other Astragalus species was demonstrated by the sequence alignment and phylogenetic analysis. Moreover, the highest expression level of AsiCHI was observed in leaves and showed a positive correlation with anthocyanin content. The functional analysis further revealed that the overexpression of AsiCHI enhanced the anthocyanidin accumulation in the transgenic lines. This study provided a better understanding of AsiCHI and elucidated its role in anthocyanin production.

Low soil fertility and low microbiological activity have been widely observed in certain newly cultivated lands. To develop effective agricultural management practices that can improve soil quality rapidly, a field experiment was conducted on paddy fields reclaimed from unused, low-slope, hilly wasteland in Quzhou City, Zhejiang Province, Southeast China. The six treatments that were applied included (1) control (CK), no crops were planted and no fertilizers were applied; (2) R, rice planted without fertilization; (3) RR, rice–ryegrass rotation without fertilizer application; (4) RRM, rice–ryegrass rotation with the application of mineral fertilizer; (5) RRMO, rice–ryegrass rotation with the application of mineral fertilizer and organic manure; and (6) RRMH, rice–ryegrass rotation with mineral fertilizer and humic acid application. The results showed that the application of fertilizer with ryegrass cropping improved the amounts of soil organic C (SOC) by 115–296% and particulate organic C (POC) by 162–256% (p < 0.05). Additionally, the RRMO treatment had the highest soil quality index value (0.81), which was significantly higher (p < 0.05) than that of the other treatments. The random forest model showed that N-related properties (including urease activity, total N, and particulate organic N), C-related properties (including SOC, POC, and cellulase activity), and available P could significantly (p < 0.05) explain the response of rice yield to different fertilization management strategies. Compared with CK, the cultivated land quality grades of RRMO and RRMH increased from Grade 9 to Grade 7. However, the RRMH obtained the highest net profit from farmland quota trading. Overall, RRMO and RRMH are suggested to rapidly improve soil productivity and maximize the economic benefit of newly cultivated land utilization, respectively.