Facilitated wetlands are promoted as a mitigation tool in agricultural landscapes because of their potential to attenuate contaminant losses while increasing biodiversity. Here, we investigate the contaminant removal efficiency of two facilitated wetlands established on a dairy farm and a sheep and beef farm. Each wetland had a sediment trap installed near the inlet to form a sediment trap–wetland complex. Water flows through the wetland complexes were measured continuously over three calendar years. Water samples were collected manually under low flow conditions and by an automated sampling system during runoff events at high flows. Samples were analysed for sediment, Escherichia coli , phosphorus and nitrogen. Effectiveness of the wetlands was assessed based on contaminant concentrations and annual loads. The wetland on the dairy farm did not significantly reduce any contaminants, and this was attributed to the very small size of the wetland (0.06%) relative to the contributing catchment area. The wetland on the sheep and beef farm (0.2% of catchment area) significantly reduced annual loads of sediment (53%), phosphorus (31%) and E. coli (10%). To maximise effectiveness, wetlands should be sized appropriately for the contributing catchment area, and sediment traps should be cleaned before they are completely full.
Large Language Models show promise for agricultural advisory, yet vanilla models exhibit unsupported recommendations, generic advice lacking specific, actionable detail, and communication styles misaligned with smallholder farmer needs. In high stakes agricultural contexts, where recommendation accuracy has direct consequences for farmer outcomes, these limitations pose challenges for responsible deployment. We present a hybrid LLM architecture that decouples factual retrieval from conversational delivery: supervised fine-tuning with LoRA on expert-curated GOLDEN FACTS (atomic, verified units of agricultural knowledge) optimizes fact recall, while a separate stitching layer transforms retrieved facts into culturally appropriate, safety-aware responses. Our evaluation framework, DG-EVAL, performs atomic fact verification (measuring recall, precision, and contradiction detection) against expert-curated ground truth rather than Wikipedia or retrieved documents. Experiments across multiple model configurations on crops and queries from Bihar, India show that fine-tuning on curated data substantially improves fact recall and F1, while maintaining high relevance. Using a fine-tuned smaller model achieves comparable or better factual quality at a fraction of the cost of frontier models. A stitching layer further improves safety subscores while maintaining high conversational quality. We release the farmerchat-prompts library to enable reproducible development of domain-specific agricultural AI.
Judith Vilella-Cantos, Mónica Ballesta, David Valiente
et al.
An optimal solution to the localization problem is essential for developing autonomous robotic systems. Apart from autonomous vehicles, precision agriculture is one of the elds that can bene t most from these systems. Although LiDAR place recognition is a widely used technique in recent years to achieve accurate localization, it is mostly used in urban settings. However, the lack of distinctive features and the unstructured nature of agricultural environments make place recognition challenging. This work presents a comprehensive review of state-of-the-art the latest deep learning applications for agricultural environments and LPR techniques. We focus on the challenges that arise in these environments. We analyze the existing approaches, datasets, and metrics used to evaluate LPR system performance and discuss the limitations and future directions of research in this eld. This is the rst survey that focuses on LiDAR based localization in agricultural settings, with the aim of providing a thorough understanding and fostering further research in this specialized domain.
Ameera Yacoob, Shaeden Gokool, Alistair Clulow
et al.
Water stress significantly threatens sugarcane production, particularly among smallholder farmers in South Africa, where spatially explicit assessments remain limited. This study aimed to improve the quantification of crop water stress by developing a machine learning (ML) model to predict the Normalised Difference Water Index (NDWI), a proxy for vegetation water content. An ML approach was adopted to capture complex, non-linear relationships between structural vegetation indices (SVIs) and NDWI. Sentinel-2 satellite data and UAV-acquired multispectral imagery were integrated, with the model trained using satellite-derived SVIs and NDWI, and then applied to UAV-derived SVIs to predict NDWI. The model achieved high predictive accuracy (R² = 0.95, RMSE = 0.03, MAE = 0.02) and effectively captured temporal variations in sugarcane water status, including post-rainfall stress recovery and increased water retention during early maturation—aligning with changes in leaf area index (LAI), chlorophyll content (CC), and Total Soil Water Profile (TSWP). NDWI also showed a positive correlation with actual evapotranspiration (ETa; R² = 0.60) and a negative correlation with the Water Deficit Index (WDI; R² = 0.62), suggesting its potential to reflect crop water status under certain conditions. When interpreted in conjunction with in situ measurements of precipitation, TSWP, and WDI, the predicted NDWI provides valuable insights into crop water dynamics. This approach demonstrates the potential of ML-driven NDWI estimation to support site-specific irrigation scheduling, enhance resource use efficiency, and promote sustainable sugarcane cultivation. The findings contribute to climate-resilient water management practices tailored to the needs of smallholder systems in water-scarce regions.
Rajhans Singh, Rafael Bidese Puhl, Kshitiz Dhakal
et al.
Deep learning models are transforming agricultural applications by enabling automated phenotyping, monitoring, and yield estimation. However, their effectiveness heavily depends on large amounts of annotated training data, which can be labor and time intensive. Recent advances in open-set object detection, particularly with models like Grounding-DINO, offer a potential solution to detect regions of interests based on text prompt input. Initial zero-shot experiments revealed challenges in crafting effective text prompts, especially for complex objects like individual leaves and visually similar classes. To address these limitations, we propose an efficient few-shot adaptation method that simplifies the Grounding-DINO architecture by removing the text encoder module (BERT) and introducing a randomly initialized trainable text embedding. This method achieves superior performance across multiple agricultural datasets, including plant-weed detection, plant counting, insect identification, fruit counting, and remote sensing tasks. Specifically, it demonstrates up to a $\sim24\%$ higher mAP than fully fine-tuned YOLO models on agricultural datasets and outperforms previous state-of-the-art methods by $\sim10\%$ in remote sensing, under few-shot learning conditions. Our method offers a promising solution for automating annotation and accelerating the development of specialized agricultural AI solutions.
Digital agriculture technologies rely on sensors, drones, robots, and autonomous farm equipment to improve farm yields and incorporate sustainability practices. However, the adoption of such technologies is severely limited by the lack of broadband connectivity in rural areas. We argue that farming applications do not require permanent always-on connectivity. Instead, farming activity and digital agriculture applications follow seasonal rhythms of agriculture. Therefore, the need for connectivity is highly localized in time and space. We introduce BYON, a new connectivity model for high bandwidth agricultural applications that relies on emerging connectivity solutions like citizens broadband radio service (CBRS) and satellite networks. BYON creates an agile connectivity solution that can be moved along a farm to create spatio-temporal connectivity bubbles. BYON incorporates a new gateway design that reacts to the presence of crops and optimizes coverage in agricultural settings. We evaluate BYON in a production farm and demonstrate its benefits.
Amid the challenges posed by global population growth and climate change, traditional agricultural Internet of Things (IoT) systems is currently undergoing a significant digital transformation to facilitate efficient big data processing. While smart agriculture utilizes artificial intelligence (AI) technologies to enable precise control, it still encounters significant challenges, including excessive reliance on agricultural expert knowledge, difficulties in fusing multimodal data, poor adaptability to dynamic environments, and bottlenecks in real-time decision-making at the edge. Large language models (LLMs), with their exceptional capabilities in knowledge acquisition and semantic understanding, provide a promising solution to address these challenges. To this end, we propose Farm-LightSeek, an edge-centric multimodal agricultural IoT data analytics framework that integrates LLMs with edge computing. This framework collects real-time farmland multi-source data (images, weather, geographic information) via sensors, performs cross-modal reasoning and disease detection at edge nodes, conducts low-latency management decisions, and enables cloud collaboration for model updates. The main innovations of Farm-LightSeek include: (1) an agricultural "perception-decision-action" closed-loop architecture; (2) cross-modal adaptive monitoring; and (3)a lightweight LLM deployment strategy balancing performance and efficiency. Experiments conducted on two real-world datasets demonstrate that Farm-LightSeek consistently achieves reliable performance in mission-critical tasks, even under the limitations of edge computing resources. This work advances intelligent real-time agricultural solutions and highlights the potential for deeper integration of agricultural IoT with LLMs.
Anas Abouaomar, Mohammed El hanjri, Abdellatif Kobbane
et al.
In this paper, we presents a novel hierarchical federated learning architecture specifically designed for smart agricultural production systems and crop yield prediction. Our approach introduces a seasonal subscription mechanism where farms join crop-specific clusters at the beginning of each agricultural season. The proposed three-layer architecture consists of individual smart farms at the client level, crop-specific aggregators at the middle layer, and a global model aggregator at the top level. Within each crop cluster, clients collaboratively train specialized models tailored to specific crop types, which are then aggregated to produce a higher-level global model that integrates knowledge across multiple crops. This hierarchical design enables both local specialization for individual crop types and global generalization across diverse agricultural contexts while preserving data privacy and reducing communication overhead. Experiments demonstrate the effectiveness of the proposed system, showing that local and crop-layer models closely follow actual yield patterns with consistent alignment, significantly outperforming standard machine learning models. The results validate the advantages of hierarchical federated learning in the agricultural context, particularly for scenarios involving heterogeneous farming environments and privacy-sensitive agricultural data.
Simar Ghumman, Fabio Di Troia, William Andreopoulos
et al.
Unmanned Ground Vehicles (UGVs) are emerging as a crucial tool in the world of precision agriculture. The combination of UGVs with machine learning allows us to find solutions for a range of complex agricultural problems. This research focuses on developing a UGV capable of autonomously traversing agricultural fields and capturing data. The project, known as AGRO (Autonomous Ground Rover Observer) leverages machine learning, computer vision and other sensor technologies. AGRO uses its capabilities to determine pistachio yields, performing self-localization and real-time environmental mapping while avoiding obstacles. The main objective of this research work is to automate resource-consuming operations so that AGRO can support farmers in making data-driven decisions. Furthermore, AGRO provides a foundation for advanced machine learning techniques as it captures the world around it.
This comprehensive study investigates the intricate relationship between water scarcity and agricultural production, emphasizing its critical global significance. The research, through multidimensional analysis, investigates the various effects of water scarcity on crop productivity, especially the economic water scarcity (AEWS) which is the main factor of influence. The study stresses the possibility of vertical farming as a viable solution to the different kinds of water scarcity problems, hence, it emphasizes its function in the sustainable agricultural development. Although the study recognizes that some problems still remain, it also points out the necessity of more research to solve the issues of scalability and socio-economic implications. Moving forward, interdisciplinary collaboration and technological innovation are essential to achieving water-secure agriculture and societal resilience.
Air pollution from agricultural emissions is a significant yet often overlooked contributor to environmental and public health challenges. Traditional air quality forecasting models rely on physics-based approaches, which struggle to capture complex, nonlinear pollutant interactions. In this work, we explore forecasting N$_2$O agricultural emissions through evaluating popular architectures, and proposing two novel deep learning architectures, EmissionNet (ENV) and EmissionNet-Transformer (ENT). These models leverage convolutional and transformer-based architectures to extract spatial-temporal dependencies from high-resolution emissions data
In recent decades, the German wine market has undergone significant structural changes due to intensifying competition and shifting consumption patterns. Increased imports and declining exports have pressured German wine estates to adapt for survival. The study explores these long-term trends and structural changes in German wine estates, focusing on those marketing bottled wine. It aims to understand how these businesses have adapted to economic pressures in a highly competitive market from 1993 to 2020, using business panel data and regression analysis for 16 key performance indicators (KPIs). At first (until the financial crisis of 2008) estates benefitted from mechanisation and economies of scale, leading to a significant reduction in labour hours per hectare and moderate increases in wine prices, improving labour productivity and profitability. However, yields declined due to a shift towards lower-yield grape varieties in response to market demand. From 2009 onward, rising labour and material costs as well as stagnating yields started eroding profitability gains, leading to an overall stagnation of long-term profitability. When observing differences in developments between size groups, large wine estates experienced a considerably sharper increase in costs per ha than small to medium sized wine estates, from 2009 onward. Nonetheless, this could be counterbalanced by large wine estates also generating significantly higher productivity increases in the same time period, resulting in a significant increase in profitability for large wine estates from 2009 onward, while small to medium sized wine estates stagnated.
Domenico Longo, Serena Guarrera, Delia Ventura
et al.
The need for accurate measurement of CO2 emissions from surfaces arises from various fields, particularly in precision agriculture, irrigation water management, wastewater management, volcanology, geothermal exploration, environmental and climate monitoring. This study introduces a novel, cost-effective closed dynamic accumulation chamber system designed to measure CO2 fluxes from soil and water surfaces. A short review of existing measurements techniques is provided, alongside a detailed explanation of key algorithms used for processing field data. The proposed system collects raw CO2 concentration data via an internal data logger. A custom-developed software suite enables real-time first-approximation flux calculation through a user-friendly Javascript web application compatible with smartphones with any type of operating system and web browser. A freely available Matlab® tool allows for post-processing data analysis for a more accurate flux calculation. After calibration against the commercial PP Systems EGM-5, assumed as a reference, some case studies in agriculture, wastewater treatment and volcanic environments demonstrate the instrument's versatility, showcasing its potential for advance in agricultural field and environmental sustainability.
A field survey was conducted from January to October 2024 to investigate the occurrence of plant diseases caused by fungal, bacterial, and viral pathogens in major horticultural crops cultivated in Cheorwon-gun, Gangwon-do, South Korea. Eight representative crops, including paprika and tomato, were examined using specific primers for pathogenic microorganisms via polymerase chain reaction (PCR) and reverse transcription PCR (RT-PCR). As a result, five fungal pathogens were detected: Botrytis cinerea, Fusarium oxysporum, Phytophthora capsici, P. infestans, and Pythium ultimum. Among them, P. ultimum and P. infestans were predominant, each accounting for 40% of the total fungal detections, jointly representing 80.00% of all fungal cases. Three bacterial pathogens were identified: Clavibacter michiganensis, Pectobacterium carotovorum, and Ralstonia solanacearum, with R. solanacearum showing the highest incidence (75%). Seven viral pathogens were also detected, including broad bean wilt virus 2, cucumber green mottle mosaic virus, cucumber mosaic virus, pepper mottle virus, pepper mild mottle virus, tomato spotted wilt virus (TSWV), and watermelon mosaic virus, with TSWV being the most prevalent (64.86%). Spatial analysis showed that R. solanacearum and TSWV were found across all six surveyed regions of Cheorwon-gun, indicating widespread distribution. Seasonal patterns revealed that fungal diseases were most prevalent in May, bacterial diseases peaked during July and August, and viral infections were primarily detected between May and July. These findings provide baseline data for the development of effective disease monitoring and management strategies in horticultural crop production in Cheorwon-gun.
Abstract Silver is a potent antimicrobial agent which is used in the form of nanomaterial or as metal salts for antimicrobial applications. Antimicrobial agents have a major role in water treatment, chemical industries, food preservation, aquaculture ponds, agricultural productivity and biomedical applications. Presently, due to emergence of nanoscience and technology metallic silver nanoparticles (AgNPs) are used as antimicrobial agent and is synthesized by following various protocols. In this review article, plants and algae mediated AgNPs synthesis is highlighted and their application as an antimicrobial agent is discussed. This review will emphasize the role of biosynthesized AgNPs for its antimicrobial application which will provide further insight towards better health, environment and prevention from infectious diseases.
It is essential to increase the productivity of agricultural and farming processes to improve yields and cost-effectiveness with new technology such as the Internet of Things (IoT). In particular, IoT can make agricultural and farming industry processes more efficient by reducing human intervention through automation. In this study, the aim to analyze recently developed IoT applications in the agriculture and farming industries to provide an overview of sensor data collections, technologies, and sub-verticals such as water management and crop management. In this review, data is extracted from 60 peer-reviewed scientific publications (2016-2018) with a focus on IoT sub-verticals and sensor data collection for measurements to make accurate decisions. Our results from the reported studies show water management is the highest sub-vertical (28.08%) followed by crop management (14.60%) then smart farming (10.11%). From the data collection, livestock management and irrigation management resulted in the same percentage (5.61%). In regard to sensor data collection, the highest result was for the measurement of environmental temperature (24.87%) and environmental humidity (19.79%). There are also some other sensor data regarding soil moisture (15.73%) and soil pH (7.61%). Research indicates that of the technologies used in IoT application development, Wi-Fi is the most frequently used (30.27%) followed by mobile technology (21.10%). As per our review of the research, we can conclude that the agricultural sector (76.1%) is researched considerably more than compared to the farming sector (23.8%). This study should be used as a reference for members of the agricultural industry to improve and develop the use of IoT to enhance agricultural production efficiencies. This study also provides recommendations for future research to include IoT systems' scalability, heterogeneity aspects, IoT system architecture, data analysis methods, size or scale of the observed land or agricultural domain, IoT security and threat solutions/protocols, operational technology, data storage, cloud platform, and power supplies.
AbstractFrom approximately 3.6 percent in 1985–1995 to less than 2 percent in 1995–2005, India's agricultural growth has slowed. This is far below the agriculture sector's goal of 4% annual growth by 2020. Food grain production is the main source of worry. Nanotechnology (NT) has been recognized as a promising technology for revitalizing the agricultural and food industries, as well as improving the life of the poor. Nanotechnology may help a variety of industries, including health care, materials, textiles, information and communication technology (ITC), and energy. Nanotechnology is used in crop production, food processing and packaging, food security and water purification, environmental remediation, crop enhancement, and plant protection in the agricultural industry. Agricultural production may be increased by using nanomaterials to create genetically better animals and plants, site-specific medication and gene delivery at the cellular/molecular level in animals and plants, and Nano array-based genetic alteration in animals and plants under stress. Nanotechnology has the potential to improve disease resistance, plant growth, and nutrient utilization by allowing precise administration of agrochemicals.Nano encapsulated solutions demonstrate the capacity to utilize pesticides, insecticides, and herbicides more effectively and site-specifically in an environmentally benign and greener manner. It has been effectively utilized in postharvest to preserve the freshness, quality, and shelf life of stored products while also avoiding disease outbreaks in a relatively safe manner. Nanomaterials are a relatively new technology in agriculture, and further study is needed. Nanotechnology's application in agriculture has social and ethical implications that must be addressed. The toxicity of nanomaterials must be assessed before they can be commercialized and used in the field.