Introduction This study investigates the influence of key meteorological factors—air temperature, relative humidity, and near-surface (5 cm) soil temperature—on population fluctuations of Frankliniella occidentalis (western flower thrips) and Helicoverpa armigera (cotton bollworm) in sunflower cultivation zones of arid Northwestern China. Methods Field data were collected from 2023 to 2024 in Beitun City and surrounding sites in Xinjiang, aligning pest occurrence records with localized meteorological monitoring. Statistical frameworks included partial correlation analysis, principal component analysis (PCA), and both linear and nonlinear regression modeling. Results The study revealed distinct stage-specific ecological responses. F. occidentalis density during the budding stage showed a strong positive correlation with air temperature (r = 0.7300, p < 0.05), fitting the Robust Index Increment Model (R² = 0.9577) within a temperature-sensitive range of 23–30°C. Conversely, H. armigera populations were predominantly driven by humidity during early stages (r = 0.7400, p < 0.05), with later stages influenced by combined thermal and moisture thresholds. The H. armigera population exhibited strong positive correlations with both air temperature and relative humidity, with a maximum partial correlation coefficient of r = 0.8400. At the flowering stage, air temperature and humidity accounted for 57.89–60.42% and 23.61–29.26% of ecological variation, respectively (regression analysis). Threshold analysis indicated that during the sowing period, the combined condition of air temperature ≥ 24°C and relative humidity ≥ 40–50% is the key ecological threshold range leading to a significant increase in H. armigera populations. This threshold is a statistical correlation range based on observational data from specific years and locations; its specific boundary values (especially the humidity lower limit) may carry a confidence interval. Discussion This threshold provides an important reference for early warning in this region, though its universality and stability require further validation under broader climatic conditions, soil types, and cultivation management practices. Moreover, the influence of meteorological factors on pest population dynamics shows obvious interspecific differences and stage-specific effects. Air temperature emerged as the primary limiting factor for F. occidentalis , whereas H. armigera populations were synergistically influenced by both air temperature and relative humidity.

Groundnut rosette disease (GRD) is a major constraint to groundnut production in sub-Saharan Africa, causing severe yield losses, particularly under low-input smallholder farming systems with limited access to external inputs. Because resistance to this pathogen is strongly influenced by environmental factors, quantifying genotype-by-environment (G×E) interactions is essential for identifying GRD-resistant varieties and optimal testing locations. This study analyzed the effects of genotype (G), environment (E), and their interaction (G×E) on GRD resistance and grain yield of groundnut in northern Mozambique. Twenty advanced groundnut lines ( Arachis hypogaea L.) were assessed across 11 environments from 2014 to 2018 using an alpha-lattice design with two replications under low-input conditions. Measured traits included phenology, number of emerged plants, number of plants at harvest, yield components, grain yield, and GRD incidence. General and mixed linear models, stability analysis, AMMI and SREG models, and Pearson correlations were applied. Significant G×E interaction was observed, indicating a strong environmental influence on trait expression. GRD incidence was highest in the NPL_15 environment, where the susceptible cultivar JL-24 reached 56% and SREG analysis identified this site as a suitable hotspot for resistance screening. Genotypes ICGV-SM 7508, ICGV-SM 7510, ICGV-SM 7518, ICGV-SM 7533, ICGV-SM 7558, ICGV-SM 7566, ICGV-SM 8530 and ICG 405 showed resistance. No consistent relationship was observed between GRD incidence and grain yield; however, yield reductions were mainly associated with plant losses during the growing season. Genotypes ICGV-SM 7518 and ICGV-SM 7510 showed broad adaptation, combining high yield and GRD resistance across environments.

Horseradish ( Armoracia rusticana ) is a high-value specialty crop whose production in Southern Illinois is constrained by limited herbicide options and labor-intensive weed management, particularly in commercial fields. This study developed a multisource image dataset and evaluated lightweight deep learning models to enable real-time, vision-based weed detection for future robotic weeding systems in commercial horseradish fields of Southern Illinois. Image data collection was conducted during the 2024 growing season at two commercial fields and one research site using a handheld smartphone and an unmanned ground vehicle (Farm-ng Amiga robotic platform) equipped with two stereo cameras (Luxonis OAK-D cameras). The data was first annotated, assigning appropriate labels to horseradish and weed instances, then augmented to improve data diversity. We trained and compared nine YOLO models (v8, v11, v12; nano, small, medium) using standard object detection metrics (precision, recall, F1-score, mAP@50) and computational indicators (inference time, GFLOPs, model size, training time), and selected the best configuration for hyperparameter tuning with an automated search over learning rate, regularization, and optimizer. The tuned YOLOv8-nano model achieved the best balance of detection performance and computational efficiency, and was subsequently benchmarked on multiple desktop and edge-computing platforms to assess real-time feasibility. The results demonstrated that lightweight YOLO architectures can provide accurate, fast horseradish-weed detection suitable for deployment on embedded hardware, offering a key sensing component for future autonomous mechanical weeding in commercial horseradish production. This study makes three key contributions: (i) a multisource image dataset of horseradish and weeds collected from both commercial and research fields using manual imaging and a robotic platform; (ii) an evaluation protocol that combines accuracy metrics with computational indicators to guide model selection for embedded deployment; and (iii) a cross-platform benchmarking workflow that assesses the real-time feasibility of lightweight YOLO models on desktop and edge-computing hardware for robotic weeding applications.

Introduction Intercropping annuals with perennials such as banana, coffee and trees is widespread in East and Central Africa, where farms are typically small and fragmented. Within such multi-strata production systems, shade is a critical constraint affecting crop productivity. This study aimed to identify crop species and varieties tolerant to varying shade levels. Methods Twenty common bush and climbing bean ( Phaseolus vulgaris ) varieties each, eight Arabica coffee ( Coffea arabica ) varieties, and three cover crops (velvet bean ( Mucuna pruriens ), rattle pod ( Crotalaria grahamiana ), and elephant grass ( Pennisetum purpureum )) were screened under four screenhouses transmitting photosynthetically active radiation (PAR) levels of 70%, 50%, 35% and 10% compared to open field light levels. Results Eight bush and four climbing bean varieties maintained grain yields comparable to open-field conditions at low (70% light transmission) shade conditions. High tolerance persisted for many beans at 50% and 35% transmission, though only two bush beans (HM21-7, CODMLB 499) remained unaffected at the deepest shade (10% light transmission). Bush beans consistently elongated under shade, whereas climbing beans showed mixed height responses. Arabica coffee seedlings exhibited stable height and stem diameter across all shade levels, with only a few varieties showing modest reductions. Cover crops showed no shade tolerance, with > 84% above-ground biomass loss across all shade levels for elephant grass. Meanwhile velvet bean and rattle-pod suffered 48–58% biomass losses even at 70% light transmission, though velvet bean retained a high root mass until the deepest shade. Discussion These findings suggest a high diversity in shade tolerance within existing bean and coffee varieties. They demonstrate that selecting shade-tolerant bean cultivars and stable coffee varieties can substantially improve productivity in multi-strata systems, while many common cover crops are unsuitable beneath canopy shade. Rapid phenotypic screening could offer a low-cost strategy for optimizing crop mixtures and guiding breeding efforts toward greater shade resilience in smallholder multi-strata agro-ecosystems.

The codling moth ( Cydia pomonella ) is the most serious pest of apple and pear orchards worldwide and has been designated as a quarantine or regulated pest by over 20 countries or regions globally. Morphological identification of the codling moth is highly specialized and time-consuming. In apple and pear production process, codling moth usually coexists with other insect species, including Carposina niponensis , Dichocrocis punctiferalis , Euzophera pyriella , Grapholita molesta , and Helicoverpa armigera , and sometimes the adults of Cydia trasias and Plutella xylostella ; these moth species are morphologically similar to codling moth, especially in the pre-adult stages (e.g., egg, larva, and pupa) of them. This study provides an effective solution for distinguishing the codling moth from other insect species with similar appearances automatically; a new Transformer-based model, known as Cont-Transformer, is proposed. Specifically, contrastive learning is introduced to improve distinguishing ability, which contributes to minimizing the similarity of classification labels corresponding to different labels and maximizing the similarity of classification labels of samples with the same label. The cross-entropy loss and contrastive loss are combined to guide the model in focusing on the most discriminative regions. Furthermore, we systematically analyzed various data augmentation strategies to bolster the model’s robustness and generalization, including AutoAugment, RandAugment, TrivialAugment, MixUp, and CutMix. We evaluated the proposed model architecture Cont-Transformer through comprehensive model training and testing on an insect image dataset containing 26 insect categories and a total of 14,431 images. The proposed recognition model achieved accuracy, precision, and recall rates of 99.45%, 99.40%, and 99.45%, respectively, outperforming eight other popular models, i.e., AlexNet, ResNet-50, DenseNet-121, ShuffleNet-v2, EfficientNet-b0, DeiT, MobileViT, and Swin Transformer. Moreover, the developed codling moth investigation program can identify various stages, including egg, larva, pupa, and adult, of these moth species. The present findings should be significant for precise pest control and quarantine supervision.

Puccinia kuehnii , the causal agent of orange rust, is an obligate biotrophic fungus and an emerging phytosanitary threat to sugarcane production. Although widely distributed in Colombia, its current impact and future risk under climate change remain poorly understood. This study evaluated the present and potential future risk of orange rust in sugarcane systems dedicated to sugar and panela production in Colombia. A total of 252 disease occurrence records were collected from nine sugarcane-producing departments, and disease impact was assessed in commercial fields using a damage index across different sugarcane varieties. Spatial kernel analyses were used to characterize the distribution of disease damage, and the Kruskal-Wallis test was applied to compare severity among varieties. In parallel, ecological niche modeling was used to estimate the current and future potential distribution of P. kuehnii under climate change scenarios. Orange rust was detected in all sampled regions, with reaction scores above 5 and severity values reaching 50%. The highest disease intensity was recorded in Santander and Caquetá, particularly in varieties RD 75-11 and CC 93-7510, while variety CC 01-1940 also showed marked severity in the Cauca River Valley. The ecological niche models identified areas of high current environmental suitability and projected an expansion of favorable conditions into additional sugarcane-producing regions under future climate scenarios. Our findings show that orange rust poses a substantial current and future risk to Colombian sugarcane systems and provide a multiscale spatial framework to support surveillance, varietal management, and evidence-based decision-making for the sugarcane sector.

Water scarcity increasingly threatens agricultural sustainability, particularly in arid and semi-arid regions. Treated wastewater (TWW) represents a promising non-conventional water resource for irrigation, offering economic and environmental benefits while contributing to freshwater conservation. Despite concerns over anthropogenic contaminants, its physiological effects on key crops such as durum wheat remain underexplored. In this study, durum wheat ( Triticum turgidum L. var. durum ) was irrigated with TWW or freshwater (FW) and subjected to four levels of water stress (100%, 70%, 50%, and 30% of full irrigation), with or without the addition of a microbial consortium, to assess growth, physiological traits, and stress adaptation mechanisms. Overall, TWW produced vegetative, physiological, and productive performances comparable to or slightly higher than FW, without evidence of phytotoxicity. The microbial consortium showed variable effects, including occasional negative interactions under severe water deficit, highlighting the importance of soil–plant–microbe interactions and local pedo-climatic conditions. Controlled water stress reduced yield even at moderate levels, although gas exchange data indicate that moderate deficit irrigation (70%) could be physiologically tolerated, but did not translate into higher yield under pot conditions. These findings support the potential use of TWW for durum wheat cultivation under water-limited conditions and provide new insights into plant physiological responses under combined irrigation and microbial treatments. Further studies should evaluate these effects across multiple seasons and in open-field conditions.

The use of pesticides in agriculture causes problems for human health and the environment, so that alternative solutions for weed control are urgently requested. In this study we test the possibility to recycle composted wood and white fir bark as biopesticide at three different experimentation scales: (i) in Petri dish, hot water extracts of the ligneous residues were used for a germination test with rapeseed and winter wheat seeds; (ii) in greenhouse pots, two successive cycles of rapeseed sowings were carried out with an equivalent of 300 m³ ha - ¹ and 600 m³ ha - ¹ of ligneous residues so to assess the effect on crop seedling biomass; (iii) in a field experiment, 300 m 3 ha -1 of ligneous residues were spread as mulch or incorporated into the soil just before the sowing of rapeseed in 2022 without any herbicide application. We observed that fresh bark extract prevented winter wheat and rapeseed germination, while extracts of decomposed bark and composted wood did not affect crop seed germination. In the greenhouse experiment, the biomass of rapeseed seedlings was lower with ligneous residues compared to the control, particularly with bark. In the field, only the bark had a negative effect on the number of emerging weeds during the autumn 2022, while no difference in weed biomass was observed between treatments in the following spring 2023. Overall, the 3-cm thick mulch alone was not sufficient to control the weed biomass in the field but seems promising as part of an integrated weed management strategy.

Smallholder farmers in semi-arid regions of Tanzania face multiple challenges, including erratic rainfall, prolonged droughts, and soil degradation, which threaten both crop yields and food security. Cereal-legume intercropping has been proposed as a sustainable strategy to enhance soil health, nutrient availability, and productivity under these conditions. This study evaluated the potential of intercropping maize (Zea mays) and sorghum (Sorghum bicolor) with three legumes Lablab purpureus, cowpea (Vigna unguiculata), and common bean (Phaseolus vulgaris) at TARI Makutupora, Dodoma, Tanzania. A randomized complete block design with three replications was used to assess effects on soil physical and chemical properties, microbial abundance, and crop performance. Intercropping systems significantly improved soil nitrogen (up to 0.17%), microbial populations (up to 4.8 × 10 5 CFU g −1 ), and soil moisture (up to 14%) compared with sole cereal crops. Path analysis indicated that soil moisture (β = 0.419) and nitrogen (β = 0.345) were the most influential factors for productivity. Among legumes, Lablab-based intercrops consistently enhanced soil fertility and crop yield, achieving land equivalent ratios above 1.5 and outperforming other legume intercrops. These findings demonstrate that cereal-legume intercropping, particularly with Lablab, can revitalize semi-arid soils, improve water and nutrient use efficiency, and enhance food security. This approach provides a scalable, climate-resilient strategy for sustainable intensification in Tanzania’s drylands.