Currently, pig production faces several challenges in ensuring the individual welfare of animals, especially with the rapid expansion of the sector and the shortage of available labor. Locomotor problems, such as lameness, are among the leading causes of sow culling, economically affecting productivity and overall animal welfare. In this context, Precision Livestock Farming (PLF) has gained prominence as a strategy for introducing technologies into the field that enable the detection of various issues, including locomotor disorders, bringing benefits to both animals and producers and resulting in greater system sustainability. The objective of this study was to conduct a preliminary evaluation of the behavior of piezoelectric sensors under controlled loads as an initial step toward the development of a platform for detecting locomotor variables in intensive production systems. Laboratory tests were conducted without the presence of animals. A sensor system was developed and connected to a microcontroller for acquiring electrical signals generated by the manual application of five standardized masses (100g, 150g, 170g, 190g, and 200g) at regular intervals. The collected signals were smoothed and analyzed based on the average peak amplitude using Python software. The results showed a positive correlation between the increase in applied mass and the average amplitude of the signals, indicating the system's sensitivity to pressure variations. These responses reinforce the potential of the technology to detect loads under various conditions, such as animal body weight, for use in automated monitoring applications.

This study presents the first navigable and georeferenced catalog of olive trees in Apulia (Italy), developed as part of the WADIT (Water Digital Twin) project. Olive canopy detection was performed using the YOLO11n-seg semantic segmentation algorithm, trained on 23,000 manually annotated olive trees across 250 parcels in the Barletta-Andria-Trani province. The model achieved strong performance, with sensitivity and precision exceeding 92%, and a mAP(50) of approximately 95%. Inference was scaled to the entire Apulian region using over 3TB of AGEA2019 orthophotos accessed via WMS services and processed in parallel across 254 threads, covering 460,000 tiles (200m × 200m) in 36 hours.

To enhance model generalization and address challenges such as duplicate detections, omissions, and false positives, an active learning strategy was employed. This iterative approach guided the manual review and targeted re-annotation of ambiguous or error-prone regions, significantly improving the model’s robustness across diverse agricultural landscapes. Post-processing steps included non-maximum suppression, spatial filtering via Dask–Geopandas, and validation using the Copernicus Crop Type 2019 layer to exclude non-olive tree species.

The final estimate of 59 million olive trees in 2019 closely aligns with official pre-Xylella outbreak figures, demonstrating the effectiveness of the proposed pipeline. This high-resolution catalog supports integration of vegetation data into regional water modeling frameworks, contributing to sustainable water resource management. Future work will focus on expanding temporal coverage, improving detection in degraded or high-density canopies, and advancing full automation of the monitoring pipeline.

Mutation breeding offers a rapid and effective approach to create novel genetic variation in rice (Oryza sativa L.) and identify mutants with desirable traits. This study evaluated the radiosensitivity of the long-grain, soft, and aromatic inbred rice variety NSIC Rc 218 (Mabango 3). Seeds were exposed to gamma radiation using a Cobalt-60 source, with doses ranging from 100 to 1000 Gy. For all treatments, 1,500 seeds were irradiated and grown under controlled conditions. Seedling vigor and shoot–root development were assessed at 9, 14, and 21 days after sowing. Our results showed a clear dose-dependent response in early growth traits. Low-to-moderate doses (100–300 Gy) produced variable physiological effects, including slight stimulation of shoot growth at 100 and 200 Gy. In contrast, higher doses (≥500 Gy) caused sharp declines in germination and survival. No seedlings survived beyond 600 Gy, indicating the lethal threshold. Overall survival rates ranged from 59% to 77%, with LD50 estimated at 424.11 Gy. Optimal stimulation of early growth traits occurred between 200 and 300 Gy, suggesting this range as a practical mutagenic window for generating genetic variation while maintaining seedling viability. These findings provide essential baseline radiosensitivity data for NSIC Rc 218. More broadly, such studies strengthen the foundation of mutation breeding in rice and other cereals, where optimized irradiation protocols are critical for developing mutants with improved yield, nutritional quality, and stress resilience. By refining dose–response knowledge across crop species, mutation breeding continues to play a key role in addressing global food security challenges.

Maize (Zea mays), initially domesticated in Central America, is the world's leading crop and is widely cultivated as a cereal grain. It is one of the most versatile emerging crops with wide adaptability. Ascorbic acid was applied to the plants growing under control, 60 and 120 mM NaCl, in two ways: as a foliar spray on leaves or through roots in the soil at two levels, 100 and 150 ppm. At UAF Community College PARS, a pot experiment with three replicates was conducted using a CRD with a factorial layout to investigate the influence of ascorbic acid on a variety of physiological and morphological parameters. After the application of NaCl treatment, various parameters were recorded, like shoot attributes, root attributes, and photosynthetic pigments; nutrient analysis was conducted using a standard procedure. NaCl significantly reduced the growth of the plants. All the considered parameters showed negative growth with high NaCl, but the application of ascorbic acid improved and alleviated the effects of NaCl toxicity on all parameters, except in the roots, where salt accumulated, leading to non-significant results. The leaf area index, root length, shoot length, per plant leaves, maximum plant height, (a and b) carotenoids, SOD, POD, MDA, BOD, CAT, chlorophyll content, nitrogen content Na⁺, K, Ca^2+, protein content, and soluble sugar are some of the parameters that were measured. The collected data was scrutinized using COSTAT software, and treatment means were compared using the LSD test with a probability level of at least 5%. Moreover, Tukey's test was applied to compare the means of the samples.

In arid and semi-arid regions, where water scarcity and soil salinization are major challenges, carob (Ceratonia siliqua L.) has gained attention as a potential drought- and salt-tolerant crop. However, its ability to withstand salt stress is determined by a complex interplay of genetic, physiological, and environmental factors, which requires a comprehensive approach to fully understand. This study aims to explore the resilience of carob to salt stress by integrating advanced multivariate methods, Partial Least Squares Structural Equation Modeling (PLS-SEM), and Artificial Intelligence-assisted Bayesian Inference to identify key determinants of stress tolerance. Through a Multivariate Analysis of Variance (MANOVA), we found significant effects of salt stress (F(5,30) = 7.3637, p < 2.2e-16) and ecotype (F(5,30) = 16.4968, p < 2.2e-16) on physiological and biochemical traits. The Principal Component Analysis (PCA) revealed a distinct separation between stressed and control plants, with the first principal component (PC1) explaining 76.66% of the variance, which was closely related to biomass, water content, and root length. The PLS-SEM model identified root length as the primary factor influencing biomass (coefficient = 0.629, p < 0.05), while water content and chlorophyll had no direct significant effect. Hierarchical Bayesian Inference allowed for the assessment of intra-population variability, showing that the Ouazzane provenance exhibited the highest salt resistance, followed by Safi and Aït Attab, while Khemissat proved the most vulnerable. Additionally, specific environmental effects (E[i]) demonstrated that certain provenances, such as Berkane and Aït Attab, benefitted significantly from co-cultivation with Spergularia salina under 10 g/L NaCl. These findings underline the importance of varietal selection and biotic interactions as effective strategies to mitigate salt stress, highlighting the potential for agroecological adaptation of carob to increasingly saline environments, offering valuable insights for sustainable agricultural practices in areas affected by salinity.

The nutritional profile of ripe and abscised unripe mango fruits of Alphonso, Mabrouka, and Zill varieties have been reported. This has created a paucity of information on the nutritional quality of the fruits, especially the unripe mango fruits, which are potential dietary ingredients for livestock. Hence, this study evaluated the in vitro gas production, methanogenesis, in vitro dry matter degradability, and rumen metabolites of ripe and abscised unripe mango fruits of the Alphonso, Mabrouka, and Zill varieties. The treatments comprised forage (control), unripe Alphonso, ripe Alphonso, unripe Mabrouka, ripe Mabrouka, unripe Zill, and ripe Zill mango fruits in a CRD. The data were analysed using a general linear model of ANOVA, while significant means were compared using DMRT. The results showed that the total gas production (TGP) and in vitro dry matter degradability (IVDMD) of ripe and abscised unripe mango fruits of Alphonso, Mabrouka, and Zill varieties were similar and higher (P<0.05) than the forage. The percentage of CH4 to gas production, TGP/IVDMD, and CH4/IVDMD were higher (P<0.05) in the forage than the ripe and abscised unripe mango fruits of the three varieties. The ruminal pH levels were lower (P<0.05) in mango fruit treatments compared to the forage, while the ripe mango fruit of the Zill variety had significantly higher (P<0.05) volatile fatty acids (mmol/L) and propionic acids, but lower acetic acids production than the forage. There was no difference (P>0.05) observed in the ruminal NH3-N (mg/dL), butyric acid, and iso-butyric acid production among the forage and the mango fruits of the three varieties. The gas production at 3 hours for both the ripe and the unripe mango fruits of the three varieties was higher (P<0.05) than the forage. However, at 24 hours, the gas production of the forage was only lower (P<0.05) than the ripe Alphonso and Mabrouka, while at 48 hours, no difference (P>0.05) was observed in the gas production of the forage compared to the mango fruit treatments. It can be concluded that abscised unripe mango fruits of Alphonso, Mabrouka, and Zill varieties possess competitive nutritional quality to the ripe mango fruits, outperforming the forage as both livestock and climate-smart dietary ingredients. Also, variety had no effect on the in vitro gas production, dry matter degradability, and methanogenesis of mangos for both the ripe and abscised unripe fruits. It is therefore recommended that the nutritional quality and methane-producing potential of abscised unripe mango fruits of the Alphonso, Mabrouka, and Zill varieties be evaluated in vivo.

The ecological context of the Ain Khiar alder forest, a rare and fragile environment, as well as the decline in its food resources, especially in winter, require appropriate herd management to preserve its agroecological balance. To this end, this study compares the daily activities and feeding behaviour of two local ruminant breeds, Berber sheep and Arbia crossbred goats, in this alder forest during the winter period.

The experiment involved visual observation of five two-year-old ewes and five two-year-old goats over ten consecutive days. The main daily activities were recorded using the regular interval observation method, and the quantities ingested were recorded using the ‘bite count’ method. Thus, ewes spend more time feeding and resting (287 and 70 minutes per day, respectively), while goats spend more time moving around (159 minutes per day) in search of food, reflecting their more selective feeding behaviour.

In terms of diet, the total grazing time is identical for both species (480 min/day), but goats consume more shrubs (100% of bites), while sheep adopt a mixed diet (53.5% shrubs, 46.5% grasses). The average weight of each bite and the total amount of dry matter ingested are lower in goats (0.17 g DM and 0.91 kg DM vs. 0.21 g DM and 1.2 kg DM).

In conclusion, there are notable differences in the feeding strategy and use of grazing space between the two species. A better understanding of these specific characteristics offers avenues for optimising the exploitation and management of natural resources, especially shrubs, while preserving the fragile biodiversity of the Ain Khiar alder forest.

In the shade, maize leaves exhibit high senescence because of the suppression of blue light. They also demonstrate oxidative stress traits. The shade effect also depends on light intensity and hydrogen peroxide. H₂O₂ is involved in signaling pathways related to various responses of antioxidants to abiotic stress. A pot experiment was conducted to investigate the effects of shade on the physiology and morphology of plants and presoaked seeds with H₂O₂. The maize variety Malka 2012 was soaked with different levels of H₂O₂ (100µM, 200µM, 300µM, 400µM, water-soaked and unsoaked). These seeds were sown in pots containing soil. After germination, half of the plants were moved to grow under tree shade, and the other half were kept under bright light. The experiment was conducted under a completely randomized design with three replicates per treatment in a factorial arrangement. After two months of sowing, the data were recorded for different growth and physiological attributes. Growth parameters including shoot length, root length, shoot diameter, number of leaves, area of leaves per plant, and fresh and dry weight of shoot and root were recorded. The physiological parameters determined include photosynthetic pigments, total chlorophyll, hydrogen peroxide, soluble phenolic, flavonoids, soluble sugar, anthocyanins, ascorbic acid, K+ , Na+ , Ca2+, Phosphate-P, Nitrate-N and Sulphate-S. Morphological and physiological parameters showed negative growth under shade stress but positive growth under light conditions with the application of 100 and 200µM of H₂O₂. Potassium and sulfur ions of the root and shoot show non-significant differences. Results revealed that shade stress reduces the physiology and morphology of maize. Tukey’s test was used to compare the least significant difference (LSD) at 5% probability levels. The data were analyzed using COSTAT software.

Maize production faces dual challenges of food safety and productivity due to Aspergillus flavus contamination and aflatoxin production, and these risks intensify under abiotic stress conditions. Despite the growing understanding of these threats, the interactive effects of basic agronomic practices on fungal contamination remain poorly understood. Therefore, this study investigated how irrigation and nitrogen fertilization influence maize physiology, A. flavus proliferation, and mycotoxin accumulation. Field experiments employed a complete randomized design with factorial arrangements of irrigation (non-irrigated/irrigated) and nitrogen levels (60, 120, and 180 kg/ha) under Aspergillus flavus-inoculated and non-inoculated conditions. Measurements included kernel number, mold count, mycotoxins (AFB1, FB1, DON, ZEA), and nutritional parameters. Irrigation improved the kernel performance of inoculated plants. The inoculated-non-irrigated group had significantly poorer kernel production (19.09 ± 0.58) compared to the control-non-irrigated, control-irrigated and inoculated-irrigated groups (21.17 ± 0.48, 21.33 ± 0.46 and 21.11 ± 0.60, respectively). Notably, increasing nitrogen from 60 to 180 kg/ha reduced AFB1 levels in inoculated maize from 164.25 ± 74.25 µg/kg to 114.94 ± 80.80 µg/kg while maintaining stable nutritional parameters across treatments. Nitrogen fertilization demonstrates protective effects against fungal proliferation and AFB1 accumulation under biotic stress, highlighting how optimized agronomic practices can enhance maize's resilience to mycotoxin contamination in changing climate conditions.

Acknowledgement

Project No. TKP2021-NKTA-32 has been implemented with support provided by the Ministry of Culture and Innovation of Hungary from the National Research, Development and Innovation Fund, financed under the TKP2021-NKTA funding scheme. This research is funded by the National Research, Development and Innovation Fund of Hungary project No. 2018-1.2.1-NKP-2018-00002.

Crop diseases represent a critical threat to global food security, demanding early and accurate detection to mitigate devastating losses. While AI offers immense potential, its adoption in agriculture is hampered by farmers' data privacy concerns and a lack of trust in "black box" models. Existing solutions combining federated learning for privacy and explainable AI (XAI) for transparency are a step forward, but they remain passive, offering one-way explanations to the user. This research introduces a paradigm-shifting framework, Symbiotic AI, that moves beyond passive explanation to active, collaborative intelligence. We propose a novel Federated Co-learning system where farmers and AI models learn from each other in a continuous, privacy-preserving loop. The core of this framework is an Interactive Counterfactual Explanation module. Instead of merely highlighting what the model saw (e.g., via heatmaps), our system uses a generative model (GAN) to show the farmer a counterfactual: a synthetic image of their own crop, subtly altered to show the minimal change required to flip the model's diagnosis (e.g., "This is what your healthy leaf would need to look like to be classified as having blight").

Crucially, the farmer can then interact with this explanation, confirming its accuracy or providing corrective feedback, such as, "No, the key indicator you missed is this stem discoloration." This expert human feedback is quantified and integrated back into the federated learning process, directly refining not only the central predictive model but also the local generative models that create the explanations. This creates a powerful co-learning dynamic where the AI becomes personalized to the unique environmental conditions and tacit knowledge of each participating farm, without ever compromising data sovereignty. The methodology involves a dual-model federated architecture—a Convolutional Neural Network (CNN) for disease detection and a conditional Generative Adversarial Network (cGAN) for generating counterfactuals—trained across decentralized farm data. Rigorous privacy is ensured through differential privacy. Expected outcomes include a system that not only achieves state-of-the-art detection accuracy (>97%) but also demonstrably improves over time by codifying farmer expertise. This research pioneers a new class of human-in-the-loop AI systems that fosters deep trust, accelerates adoption, and creates a dynamically evolving, resilient, and truly farmer-centric digital agriculture ecosystem.