The application of remote sensing, which is non-destructive and cost-efficient, has been widely used in crop monitoring and management. This study used a built-in multispectral imager on a small unmanned aerial vehicle (UAV) to capture multispectral images in five different spectral bands (blue, green, red, red edge, and near-infrared), instead of satellite-captured data, to monitor soybean growth in a field. The field experiment was conducted in a soybean field at the Mississippi State University Experiment Station near Pontotoc, MS, USA. The experiment consisted of five cover crops (Cereal Rye, Vetch, Wheat, Mustard plus Cereal Rye, and native vegetation) planted in the winter and three fertilizer treatments (Fertilizer, Poultry Liter, and None) applied before planting the soybean. During the soybean growing season in 2022, eight UAV imaging flyovers were conducted, spread across the growth season. UAV image-derived vegetation indices (VIs) coupled with machine learning (ML) models were computed for characterizing soybean growth at different stages across the season. The aim of this study focuses on monitoring soybean growth to predict yield, using 14 VIs including CC (Canopy Cover), NDVI (Normalized Difference Vegetation Index), GNDVI (Green Normalized Difference Vegetation Index), EVI2 (Enhanced Vegetation Index 2), and others. Different machine learning algorithms including Linear Regression (LR), Support Vector Machine (SVM), and Random Forest (RF) are used for this purpose. The stage of the initial pod development was shown as having the best predictability for earliest soybean yield prediction. CC, NDVI, and NAVI (Normalized area vegetation index) were shown as the best VIs for yield prediction. The RMSE was found to be about 134.5 to 511.11 kg ha⁻¹ in the different yield models, whereas it was 605.26 to 685.96 kg ha⁻¹ in the cross-validated models. Due to the limited number of training and testing samples in the K-fold cross-validation, the models’ results changed to some extent. Nevertheless, the results of this study will be useful for the application of UAV remote sensing to provide information for soybean production and management. This study demonstrates that VIs coupled with ML models can be used in multistage soybean yield prediction at a farm scale, even with a limited number of training samples.

Drought, ultraviolet-B (UV-B), and nitrogen stress are significant constraints for sweetpotato productivity. Their impact on plant growth and development can be acute, resulting in low productivity. Identifying phenotypes that govern stress tolerance in sweetpotatoes is highly desirable to develop elite cultivars with better yield. Ten sweetpotato cultivars were grown under nonstress (100% replacement of evapotranspiration (ET)), drought-stress (50% replacement of ET), UV-B (10 kJ), and low-nitrogen (20% LN) conditions. Various shoot and root morphological, physiological, and gas-exchange traits were measured at the early stage of the crop growth to assess its performance and association with the storage root number. All three stress factors caused significant changes in the physiological and root- and shoot-related traits. Drought stress reduced most shoot developmental traits (29%) to maintain root growth. UV-B stress increased the accumulation of plant pigments and decreased the photosynthetic rate. Low-nitrogen treatment decreased shoot growth (11%) and increased the root traits (18%). The highly stable and productive cultivars under all four treatments were identified using multitrait stability index analysis and weighted average of absolute scores (WAASB) analyses. Further, based on the total stress response indices, ‘Evangeline’, ‘O’Henry’, and ‘Beauregard B-14’ were identified as vigorous under drought; ‘Evangeline’, ‘Orleans’, and ‘Covington’ under UV-B; and ‘Bonita’, ‘Orleans’, and ‘Beauregard B-14’ cultivars showed greater tolerance to low nitrogen. The cultivars ‘Vardaman’ and ‘NC05-198’ recorded a low tolerance index across stress treatments. This information could help determine which plant phenotypes are desirable under stress treatment for better productivity. The cultivars identified as tolerant, sensitive, and well-adapted within and across stress treatments can be used as source materials for abiotic stress tolerance breeding programs.

Soil biological properties are important for the stabilization and preservation of a good soil structure. Management practices can affect the diversity and population of microorganisms, which could beneficially change soil properties and promote a more sustainable dryland crop production. This study was established near Pontotoc, MS, USA (34°07′ N, 88°59′ W) on an Atwood silt loam (fine-silty, mixed, semiactive, thermic Typic Paleudalf) to evaluate the impacts of cover crops, planting dates and fertilizer sources (poultry litter, inorganic fertilizer and no fertilizer) on selected biologically related soil properties in a no-tillage, dryland soybean production. Soil analyses included total carbon and nitrogen, permanganate oxidizable carbon (POXC), easily extractable glomalin-related soil protein (EE-GRSP), water stable aggregate (WSA) and soil pH. Cover crop production and soybean yield were also determined. The results indicated that the fertilizer source had an impact on total nitrogen, EE-GRSP and soybean yield. Total N was 6% higher with poultry litter at the early planting date compared to no fertilizer (control) (p < 0.0018) and at the late planting date, when total N and EE-GRSP were increased by 11% and 13%, respectively, with poultry litter compared to no fertilizer. Additionally, soil pH was reduced by 0.25 units in the poultry litter-amended treatment. Soybean yield was increased by 68% and 51% in early-planted soybean and 42% and 40% in late-planted soybean with poultry litter and inorganic fertilizer, respectively, compared to no fertilizer. This study revealed that biological soil properties and soybean yield were influenced by poultry litter application. The results showed no significant effects of cover crops over the short time period of the study.

Whole cottonseed and its derived products can be used as human food, animal feed, and industrial raw material. Chemical composition of cottonseed is one of the critical parameters for evaluating its quality and potential end use. Especially, protein and dietary fibers are two desirable nutritional properties of cottonseed for animal feed. In this study, we determined their contents in cottonseed harvested from different fertilization managements with or without wheat cover crop in two consecutive years. Seed content of crude protein increased in the order of no fertilizer ? poultry litter ? chemical fertilizer. Both neutral and acid detergent fiber profiles were affected by fertilization and planting year/cover crop in an order reversing that of crude protein. Few statistically significant (P>0.05) impacts were observed on the content of acid detergent lignin. Observations derived from this work will provide information on nutritional value and digestibility of cottonseed as affected by cropping management practices.

The metabolic diversity of soil microbiota embodies diverse functional capabilities that support ecosystem resilience, driving essential biogeochemical processes and facilitating the optimization of sustainable agricultural systems. Integrating cover crops into agricultural systems cultivates a diverse array of metabolic activities among soil microbes, synergistically enhancing ecosystem services and bolstering soil health for sustainable and productive farming practices. In an effort to gain deeper insights and expand our knowledge, we conducted a study examining the effects of cover crops and fertilizer sources, thereby shedding light on their combined impacts on the metabolic activity dynamics of soil microbial communities. In this investigation, we employed a split-plot design with two factors: (a) cover crop with three solo cover crop species—Cereal rye (Secale cereale), wheat (Triticum aestivum), hairy vetch (Vicia villosa), and one mixture of mustard (Brassica rapa) and cereal rye (Secale cereale) (CC-mix), (b) Fertilizer source includes poultry litter, chemical fertilizer, and no-fertilizer treatments. We assessed the metabolic potential of soil microbiota by using carbon substrates utilizing Biolog EcoPlates. The findings revealed that the plots with CC-mix treatment exhibited greater metabolic diversity compared to the other treatments, while among the fertilizer sources, poultry litter demonstrated higher metabolic activity. Furthermore, both treatment factors predominantly metabolized carbohydrates and polymers compared to other carbon substrate categories. The principal component analysis accounted for 46.4% of the variance, collectively represented by PC1 and PC2, emphasizing the substantial contributions of carbohydrates, amino acids, and carboxylic acids to the observed metabolic diversity. Canonical correspondence analysis revealed that pH had positively correlated with microbial functional diversity, whereas total carbon (TC), total nitrogen (TN), and water-stable aggregates (WSA) showed a negative correlation. In conclusion, cover cropping and type of fertilizer source had a notable impact on soil microbial functional diversity, with the cover crop mixture exhibiting a more pronounced influence than the individual cover crop treatments.

The response of soil microbial communities to management practices is composite, as it depends on the various environmental factors which contribute to a shift in the microbial communities. In this study we explored the impact of combinations of soil management practices on microbial diversity and community composition in a dryland soybean production system. Soil samples were collected from the experimental field maintained under no till, cover crops, and fertility treatments, at Pontotoc Ridge-Flatwoods Branch Experiment Station, MS, USA. Targeted amplicon sequencing of 16S rRNA and ITS2 genes was used to study the bacterial and fungal community composition. Poultry litter amendment and cover crops significantly influenced soil bacterial diversity. Fertilizer sources had significantly different bacterial communities, as specific microbial taxa were strongly influenced by the changes in the nutrient availability, while cover crops influenced the soil fungal community differences. Differential enrichment of advantageous bacterial (Proteobacteria, Actinobacteria and Acidobacteria) and fungal (Mortierellomycota) phyla was observed across the treatments. Soil pH and easily extractable glomalin-related soil proteins (EE-GRSP) were correlated with bacterial communities and aggregate stability (WSA) was influenced by the poultry litter amendment, thus driving the differences in bacterial and fungal communities. These findings suggest that a long-term study would provide more inferences on soil microbial community response to management changes in these dryland soybean production systems.