Search

Reliable estimation of crop yield is vital for refining farming operations and guaranteeing food supply. This investigation tested the capacity of machine learning techniques to estimate tomato yield using meteorological measurements, spectral reflectance bands, and vegetation indices subjected to different nitrogen levels and bio-stimulant applications to generate variation in crop development. UAV-mounted sensors captured spectral information across seven acquisition dates spanning October 27 to December 15, 2023, matching critical developmental windows: vegetative establishment (acquisition 1), anthesis (acquisitions 2 and 3), fruit set and enlargement (acquisitions 4, 5, and 6), and initial maturation (acquisition 7). Discriminating explanatory variables were selected via the Pearson correlation metric (r > 0.65, P < 0.05), incorporating Near Infrared (NIR), Red Edge, and Red wavebands, together with vegetation indices comprising NDVI, GNDVI, NDRE, and SAVI. Remotely sensed canopy signatures gathered throughout the fruit set and enlargement window (acquisitions 5 and 6) displayed the tightest association with harvested yield (r = 0.66–0.74), reinforcing the necessity of spectral intelligence from the middle to latter parts of the cycle. When comparing predictive algorithms, both linear regression (LR) and XGBoost yielded the strongest predictive skill, producing root mean squared error (RMSE) figures of 16.13 kg and 16.15 kg, respectively, matched with R² values reaching 0.63. Support vector machine (SVM) and decision tree (DT) implementations also delivered competent outcomes, achieving RMSE scores of 17.15 kg and 17.18 kg, respectively. By comparison, the deep learning architecture lagged notably (RMSE = 23.49 kg, R² = 0.23), plausibly attributable to data scarcity. This research spotlights the prognostic capability inherent in spectral wavebands and underscores the centrality of growth-stage-aligned spectral acquisition for yield forecasting.

The quantification of actual evapotranspiration (ETₐ) is fundamental to sound water resource governance, yet current estimation methods pose significant constraints. Established methodologies, among them eddy covariance and energy balance computations reliant on remote sensing, are frequently beset by prohibitive costs, patchy spatial and temporal coverage, and deteriorating forecast fidelity, particularly for older empirical constructs. Although machine learning has emerged as a viable solution, it is not without drawbacks, especially a systematic bias toward low ETₐ predictions during high-temperature episodes. We trace this deficiency to inadequate model training on the scarce but critically significant extreme ETₐ values, paired with meteorological datasets that are limited rather than massive. Within this paper, we illustrate how few-shot, meta-learning frameworks (MAML), purpose-built to deliver superior generalization when data are constrained, can eclipse elementary machine learning formulations in the upscaling of ETₐ originating from two major in-situ observation networks, namely Ameriflux and Euroflux. Working with a sparse set of remotely sensed land surface variables sourced from the METRIC-EEFlux system alongside a narrow suite of climatic drivers, we establish that the selected models yield measurable benefits when assessed under the utility-based regression philosophy, guiding meaningful real-world decisions. Our preliminary findings show that EEflux ETₐ strays markedly from in-situ tower readings from the Ameriflux and EEflux installations (R² = 39%). By comparison, MAML attains the strongest alignment with ETₐ, outperforming both elementary machine learning architectures and EEFlux (R² = 71% on the full test corpus, R² = 0.88 under the Csa climate class, R² = 0.79 under the Cfa climate class, and R² = 0.78 for the CSH vegetation category), and steadily improves without succumbing to overfitting even when trained on a comparatively modest data pool. Its elevated F2 score (96%) indicates that MAML maintains high precision and recall for atypical instances, a feature of considerable consequence for irrigation scheduling. As a noteworthy result in its own right, this investigation corroborates that even a sparse set of remotely sensed EEflux outputs contributes materially to understanding ground-referenced ETₐ and can thus serve a valuable role in environments where the procurement of abundant, high-quality data is limited.

Artificial intelligence is increasingly presented as a transformative force in agriculture, promising more precise decisions, better resource use, improved productivity, and more resilient farming systems. Yet much of this promise has been shaped around large-scale, capital-intensive, data-rich farms in high-income settings. This creates a structural mismatch between dominant agricultural AI imaginaries and the everyday realities of smallholder farming. Smallholder farmers operate under conditions that are often treated as barriers to technological adoption rather than as core design conditions. Limited connectivity, intermittent electricity, restricted access to smartphones, low disposable income, linguistic diversity, and uneven access to advisory support all shape how digital tools are encountered and used. When these realities are ignored, agricultural AI risks widening rather than narrowing existing inequalities. This article develops a Smallholder-Centred AI Model for agriculture. The model reorients agricultural AI around four foundational principles: low-resource design, inclusion and affordability, context-awareness, and farmer participation. It argues that these principles must not be added after technical development but must structure the design, governance, and evaluation of AI from the beginning. The article concludes that AI can serve smallholder agriculture only when smallholders are positioned as active participants in the intelligence of the system. A smallholder-centred approach requires technologies that work with limited infrastructure, respond to local agroecological and social contexts, respect farmer knowledge, and remain affordable and accountable. The future of agricultural AI must therefore be judged not only by predictive accuracy but by its capacity to support inclusive, locally grounded, and farmer-led transformation.

Circular food systems require a shift from linear models of production, consumption, disposal, and remediation toward systems that continuously recover value from biological materials. This conceptual article argues that such a transition cannot rely on policy, behavioural change, or technological substitution alone. It also requires an intelligence layer capable of sensing waste generation, interpreting material quality, identifying recovery options, and coordinating decisions across fragmented food-system actors. Artificial intelligence offers this intelligence layer by enabling predictive, diagnostic, and decision-support capacities across the food chain. Rather than treating AI as a narrow automation tool, this article positions AI as cognitive infrastructure for circular food economies. Its role is to connect data on food loss, by-product composition, logistics, environmental impacts, recovery technologies, and stakeholder constraints. The article develops a conceptual framework for AI-enabled circular food systems. The framework integrates five functions: waste detection, valorisation pathway selection, resource recovery, sustainability-oriented decision support, and cross-actor coordination. These functions are presented as mutually reinforcing layers that allow food materials to be tracked, evaluated, redirected, and valorised before they become unmanaged waste. The article contributes to sustainability and circular economy scholarship by linking food-waste valorisation with material flow intelligence and multi-criteria decision support. It highlights how AI can support real-time evaluation of waste streams, match heterogeneous residues with suitable recovery pathways, and improve the coordination of producers, processors, retailers, consumers, waste managers, and policymakers. It also clarifies the risks of fragmented digitalisation when data systems are not aligned with circular outcomes. The proposed framework is intended for researchers, practitioners, and policymakers seeking to design circular food systems that are adaptive, evidence-based, and sustainability-oriented. It does not present new empirical data. Instead, it provides a conceptual basis for future empirical testing, digital platform development, policy design, and interdisciplinary research on AI-enabled food circularity.

Food recalls remain essential instruments for protecting public health when unsafe, contaminated, mislabelled, or otherwise non-compliant products enter food supply chains. Yet many recall systems still operate primarily as downstream corrective mechanisms after hazards have already moved through production, distribution, retail, food service, or household channels. This reactive orientation can delay containment, increase exposure, and weaken public confidence in food safety governance. Conventional recall management is increasingly challenged by globalised sourcing, rapid distribution, e-commerce, multi-ingredient products, outsourced processing, fragmented traceability records, and uneven information sharing among actors. Weak risk signals may appear in laboratory data, consumer complaints, inspection findings, supplier records, online reports, or public health surveillance before they are recognised as recall-relevant evidence. When these signals remain disconnected, recall decisions may occur only after risk propagation has already widened. This article proposes a conceptual model for intelligent food recall systems that integrates early detection, risk signal fusion, network propagation mapping, coordinated response, accountability, and communication feedback. The model reframes recall management as a system intelligence problem rather than a narrow compliance, logistics, or product-removal task. It explains how anticipatory analytics and network-aware decision support can strengthen recall scope definition, actor alignment, and public communication. The contribution of the article is a structured conceptual framework for designing recall systems that are more anticipatory, traceable, coordinated, accountable, and adaptive. The model does not report new empirical data; rather, it synthesises food safety, artificial intelligence, supply-chain traceability, risk communication, and crisis-response perspectives into an integrated recall intelligence architecture. It offers a foundation for future empirical testing, digital system design, regulatory innovation, and food-industry implementation.

Food systems increasingly generate consumer data through digital platforms, shopping records, dietary applications, wearable technologies, loyalty schemes, and personalised nutrition services. These data can reveal patterns of preference, dietary behaviour, product use, affordability constraints, and health-oriented needs. However, consumer data are often used for commercial targeting while remaining disconnected from product reformulation, nutrition improvement, and public health goals. AI-enabled food intelligence is frequently positioned as a tool for predicting consumer preferences or delivering personalised recommendations. This narrow use limits the potential of AI to support healthier food environments, more acceptable reformulated products, and more transparent consumer communication. A broader framework is needed to connect consumer insight with nutrition evidence, product innovation, ethical governance, and public trust. This article proposes a consumer-centred food intelligence framework linking AI, nutrition personalization, product reformulation, trust, transparency, and public health. The framework treats consumers not only as data sources but also as beneficiaries, participants, and evaluators of food-system innovation. It explains how individual and aggregated consumer intelligence can inform dietary advice, product design, reformulation priorities, and responsible communication. The article contributes a structured conceptual model for ethically transforming consumer data into health-oriented and trust-sensitive food-system decisions. It argues that AI should not only recommend diets or predict purchasing behaviour, but should also help translate nutritional evidence into acceptable products and transparent innovation pathways. Consumer-centred food intelligence is presented as a foundation for more responsible, healthier, and publicly trusted digital food systems.

Artificial intelligence is increasingly positioned as a driver of agri-food transformation through sensing, prediction, optimisation, automation, and decision support. Across farms, supply chains, advisory systems, and governance settings, AI-enabled tools are changing how agricultural information is gathered, interpreted, and translated into action. These changes create important opportunities, but they also raise deeper theoretical questions about what kind of intelligence agri-food systems should develop. Much discussion of agri-food AI emphasises productivity, efficiency, technical accuracy, and optimisation. Yet sustainable agri-food transformation cannot be judged only by whether AI increases yields, lowers costs, or accelerates decisions. It must also be evaluated by whether intelligent systems strengthen resilience, support equity, and respect ecological limits. This article develops a theory of sustainable intelligence in agri-food systems. Sustainable intelligence is defined as the AI-enabled capacity of agri-food systems to sense, interpret, learn from, and respond to changing conditions while balancing productivity, resilience, equity, and ecological care. The theory treats these four dimensions not as separate goals, but as interdependent constructs that must be continuously aligned. The article builds theory by integrating scholarship on digital agriculture, food-system sustainability, resilience thinking, equity, ecological responsibility, and responsible AI. It focuses on the mechanisms through which sensing, prediction, optimisation, feedback, participation, and governance can either mediate or intensify tensions among agri-food system goals. In doing so, it shifts attention from AI as a technical instrument to AI as a socio-technical balancing capacity. The proposed theory explains sustainable intelligence as an emergent property of data integration, adaptive feedback, participatory governance, ecological sensitivity, and accountability. It argues that AI-enabled agri-food systems become sustainable not when they maximise a single performance metric, but when they help actors recognise trade-offs, negotiate priorities, and preserve the conditions for just and ecologically responsible production. Sustainable intelligence therefore offers a theoretical foundation for agri-food futures that are productive, resilient, equitable, and ecologically caring.

Food security early warning systems increasingly draw on artificial intelligence, remote sensing, predictive analytics, climate information, crop monitoring, market indicators, and vulnerability data to identify emerging risks before they become severe crises. These systems are expected to provide timely signals about threats to food availability, access, utilisation, and stability across diverse ecological, economic, and social settings. Yet many early warning approaches remain fragmented. Climate risk, market instability, crop stress, and human vulnerability are often analysed in separate technical workflows, which can weaken interpretation, delay action, or produce alerts that are insufficiently linked to response capacity. This article develops a conceptual architecture for AI-enabled food security early warning systems. The architecture treats early warning as a layered process that connects data ingestion, risk-domain analytics, signal integration, uncertainty interpretation, alert generation, decision support, response targeting, and feedback governance. The proposed architecture explains how multi-source data and predictive models can be translated into interpretable and actionable food security warnings. It also emphasises that AI-enabled early warning can fail when data are incomplete, model outputs are opaque, vulnerable populations are underrepresented, or alerts are disconnected from institutions capable of acting.

Intelligent agriculture increasingly uses artificial intelligence, sensors, remote sensing, automation, modelling, and decision-support systems to monitor and manage agricultural production. These tools can strengthen observation, prediction, and coordination across farms and landscapes. Yet their contribution to agricultural transformation depends on the ecological aims toward which they are designed and governed. Regenerative agriculture requires more than improved efficiency or narrower optimisation of inputs. It calls for soil restoration, biodiversity protection, ecological feedback, and long-term system health. Intelligent systems therefore need to be evaluated not only by productivity gains, but also by their capacity to support ecological recovery. This article develops a conceptual framework for regenerative intelligence in agriculture. Regenerative intelligence is defined as the AI-supported capacity of agricultural systems to monitor, interpret, and support soil restoration, biodiversity protection, and long-term system health through adaptive and ecologically accountable decision pathways. The framework explains how sensing, ecological analytics, modelling, decision support, monitoring, feedback, and governance can contribute to regenerative agricultural transformation. It also clarifies that AI-supported agriculture is not automatically regenerative. Its regenerative value depends on ecological indicators, farmer participation, responsible governance, biodiversity sensitivity, soil-health interpretation, and sustained learning from ecological outcomes.

Agri-food systems increasingly operate under compound crises in which climate shocks, supply disruptions, market instability, ecological stress, and social vulnerability interact. These crises do not affect production, distribution, livelihoods, and food access in isolation. They move through connected risk pathways that can destabilise farms, supply chains, public institutions, and households at the same time. Artificial intelligence can support prediction, monitoring, early warning, and decision support in these contexts. However, technical forecasting alone does not constitute resilience when risks cascade across multiple domains. Agri-food resilience requires that predictive information be translated into adaptive decisions, coordinated response, and food security protection. This article develops a conceptual model of AI-supported agri-food resilience under compound crises. The model links climate shocks, supply disruptions, farmer adaptation, and food security governance as interacting domains. It treats AI as a decision-support function embedded within social, institutional, ecological, and market systems rather than as a stand-alone technological solution. The model identifies resilience mechanisms that connect risk detection, signal integration, uncertainty communication, adaptive decision support, institutional coordination, feedback learning, and accountability. It also clarifies boundary conditions that determine whether AI strengthens or weakens resilience. These include data quality, interpretability, local relevance, farmer participation, institutional response capacity, equity, and governance accountability. The article contributes a structured conceptual model for understanding how AI can support agri-food resilience when analytics are aligned with adaptation needs and food security objectives. It argues that resilient agri-food AI systems must connect prediction with human judgement, local knowledge, public action, and accountable coordination. The proposed model offers a basis for future empirical testing, policy design, and responsible innovation in crisis-sensitive food systems.

Explainable artificial intelligence is becoming central to precision agriculture because farm-management decisions increasingly depend on algorithmic recommendations generated from sensor, remote-sensing, weather, soil, crop, and machinery data. However, technical performance alone does not make agricultural AI trustworthy. A recommendation may be statistically accurate yet remain unusable if farmers cannot understand why it was produced, how uncertain it is, whether it fits local conditions, or how it can be questioned. This article develops a farmer-centered approach to explainable AI for precision agriculture. It argues that explainability must connect model logic with farmer interpretation, contextual judgment, decision confidence, contestability, accountability, and feedback learning. The proposed perspective treats farmers not as passive recipients of algorithmic advice but as active interpreters, evaluators, and co-decision-makers within human-in-the-loop agricultural systems. By reframing explainable AI as a socio-technical decision-support requirement, the article contributes a foundation for trustworthy algorithmic decision support that is understandable, usable, context-sensitive, and accountable in real farm-management conditions.

Artificial intelligence is increasingly embedded in precision agriculture through sensing, remote observation, predictive modelling, automated classification, and decision-support applications. Technical performance alone, however, is insufficient for determining whether these systems are responsibly designed or appropriately used. This article proposes an original governance framework that connects agricultural context definition, farm-data governance, transparency, explainability, accountability, sustainability assessment, farmer participation, human oversight, risk safeguards, implementation-readiness review, monitoring, and adaptive feedback. The framework treats transparency as a requirement for documenting data provenance, data quality, model assumptions, system limitations, and decision pathways, while explainability is positioned as a task-specific means of supporting interpretation rather than proof of model validity. Accountability is structured through explicit allocation of responsibilities among developers, platform providers, data stewards, advisors, farmers, and oversight actors. Environmental, social, and economic sustainability are assessed alongside digital inclusion, usability, local relevance, farmer autonomy, and trust-building conditions. The framework also addresses algorithmic opacity, bias, model drift, poor local calibration, data misuse, cybersecurity, automation dependence, accountability gaps, and implementation overclaiming. Its principal contribution is a layered, adaptive governance model for responsible agricultural decision support. The framework supports governance planning and research design but does not establish empirical validity, legal compliance, policy acceptance, farmer approval, sustainability performance, trust, certification, or deployment readiness.