Search Results (1,282)

Accurate and timely information on forest disturbance drivers is important for sustainable forest management, global carbon cycle accounting, and climate change response. However, forest disturbance classification is difficult due to two major challenges: limited labeled samples and highly imbalanced disturbance class distribution. In this article, a new framework for multi-type forest disturbance classification based on collaborative semi-supervised learning and sample generation was proposed. First, forest disturbance is detected using long-term remote sensing time series data and disturbance detection algorithms. Spatiotemporal, spectral and terrain features of different disturbance types are extracted. On this basis, to address the problem of imbalanced and small-sample conditions, a collaborative classification strategy is developed. Based on a small number of labeled samples, Support Vector Machine (SVM) and Random Forest (RF) are used to build dual base classifiers. A confident learning (CL) framework is applied to select high-confidence pseudo-labeled samples from unlabeled data. Then, a latent diffusion model (LDM) is introduced to generate high-fidelity pseudo-samples. This increases the sample size and balances the class distribution. Based on the augmented dataset, the dual classifiers are iteratively optimized using a co-training strategy, which improves model generalization under complex conditions. The results show that the proposed framework could generate high-quality pseudo-samples and effectively reduce class imbalance. The overall accuracy (OA) of the proposed framework reaches 93.2%, which is 5.7% and 4.4% higher than single classifier baselines, respectively. After introducing the LDM-based balancing mechanism, performance is further improved by 1.8% compared with the pure semi-supervised framework. This study provides an efficient and reliable solution for large-scale forest ecosystem monitoring. Full article

Fish-derived products are extensively acknowledged for their substantial role in fostering balanced diets and supporting a healthy way of life. This research is aimed at formulating, analyzing and evaluating a fish-based food product. The methodology adopted in this study adheres to contemporary food safety standards, prioritizing the utilization of minimal technological processes and natural ingredients, a focus that is gaining prominence within contemporary industrial practices. Thus, the proposal for a formulation obtained by integrating powders and extracts from plant byproducts (Citrus) represents a concrete application direction with real potential for commercialization. The product has been enriched with biocomponents derived from orange peel, namely orange extract (OE) and orange peel powder (PPO). The research focused on product development and the in situ evaluation of the effects of OE and PPO. The physicochemical composition, bioactive compound content, and antioxidant activity were evaluated, along with the microbiological status under post-opening refrigeration conditions, in order to simulate actual consumer use. In addition, the product’s color parameters and sensory attributes were analyzed. The results highlight significant potential for the development of a clean-label fish-based product, characterized by a simplified and easily implementable formulation, aligned with current production and consumption requirements. Compared to the control sample, both OE and PPO significantly influenced the analyzed parameters. Differences in physicochemical composition were observed in the experimental samples. In addition, PPO increased the antioxidant activity of the samples and the profile of bioactive compounds. Microbiological analysis, performed on day 0 and after 3 and 7 days of storage at 4 °C showed opening, confirmed the absence of Escherichia coli and Staphylococcus aureus in all samples and had an influence on the growth of fungi. The acceptability of fish-based products is often limited by odor perception, which is one of the main factors leading to consumer rejection. Sensory evaluation demonstrated that citrus-enriched samples were distinguished by the perception of particular sensory attributes. This formulation presents a practical solution to address this constraint, thereby enhancing the product’s sensory acceptability. The integration of OE and PPO yielded a more harmonized sensory profile, as evidenced by elevated hedonic scores and an intermediate placement in both principal component analysis (PCA) and external preference mapping. This research furnishes a thorough characterization of a fish-based food product, underscoring its potential as a viable option for balanced dietary regimens. Simultaneously, the findings support the product’s adherence to sustainability principles through the utilization of bioactive compounds sourced from plant byproducts, thus satisfying contemporary requirements for foods that possess an optimal nutritional profile and a diminished environmental footprint. Full article

Alfalfa (Medicago sativa), a globally important perennial forage legume, is widely cultivated in China, where effective weed management is essential for sustainable production. Chemical weed control, primarily relying on the herbicide imazethapyr, represents the most common strategy. Reliance on a single-herbicide program, however, may lead to inconsistent weed control under field conditions and may raise environmental concerns when higher application rates are used. To address this challenge, a two-year field study (2022–2023) was conducted to reduce herbicide inputs and identify new weed management options through tank mixtures. Initial screening identified imazethapyr, prometryn, imazapic, and 2,4-DB as safe and effective against broadleaf weeds. To broaden the control spectrum and reduce total herbicide use, haloxyfop-R-methyl was tank-mixed with each of the four broadleaf-active herbicides. The combinations haloxyfop-R-methyl + imazethapyr (36.5 + 56.3 g a.i. ha⁻¹) provided broad-spectrum weed control without compromising alfalfa performance and, importantly, reduced herbicide input at least by 25% of the recommended label dose. Additionally, the mixture of haloxyfop-R-methyl with 2,4-DB (36.5 + 506.3 g a.i. ha⁻¹) achieved effective, broad-spectrum weed control, increased alfalfa yield, and reduced total herbicide input at least by 25% of the recommended label dose. This mixture offers a useful option for diversifying weed management programs and reducing reliance on repeated imazethapyr applications. These tank mixtures represent sustainable and practical components of an integrated weed management system in alfalfa production. Full article

Low-carbon service outsourcing creates a governance problem in which manufacturers must address hidden emission-reduction capability before contracting and hidden effort after contracting. Consumer low-carbon preference does not automatically translate into market returns, because consumers rely on information disclosure, certification, carbon labeling, and traceability to perceive actual emission-reduction performance. This study develops a principal–agent model for a low-carbon service supply chain composed of a manufacturer and a low-carbon service provider. The baseline model examines screening and effort incentives under dual information asymmetry, the extended static model introduces heterogeneous consumer preferences and information perception, and the dynamic model incorporates consumer trust evolution. The results show that menu contracts enable manufacturers to distinguish service-provider types and induce emission-reduction effort, but truthful self-selection requires information rent. Consumer low-carbon preference strengthens incentive intensity only when disclosure converts actual emission-reduction performance into perceived low-carbon value. Disclosure investment improves the market return of emission-reduction effort, but its effectiveness is constrained by disclosure cost, provider risk aversion, and output uncertainty. Consumer low-carbon trust converges to a steady state supported by sustained emission-reduction effort and credible disclosure. The conclusions apply primarily to low-carbon service outsourcing settings in which provider capability and effort are difficult to observe and market response depends on consumers’ perception of low-carbon information. This study extends principal–agent analysis to low-carbon service supply chains and shows that effective low-carbon governance depends on the coordination of contract incentives, information disclosure, and trust accumulation. Full article

The increasing number of sustainability claims may reduce the perceived reliability of formal eco-labels, creating an environment in which greenwashing can erode institutional trust. This study explores how consumers navigate significant information asymmetry when standardized environmental signals are absent. Using a qualitative research design, we conducted focus group discussions with Hungarian- and Romanian-speaking consumers in Transylvania, Romania, a multiethnic transitioning market. Computational text analysis, including topic modeling, was used to support this interpretive approach and effectively decode the complex typologies of green claim evaluation. The findings suggest that signal degradation among the participants was associated with culturally embedded heuristic substitution rather than a uniform rejection of green claims. Romanian-speaking participants described more analytical, information-seeking heuristics that are tightly integrated into routine purchasing decisions. Conversely, Hungarian-speaking participants articulated a looser connection between generalized skepticism and their purchasing routines. This study contributes to signaling theory and administrative science by suggesting that standardized governance tools may be less effective when they are not aligned with localized trust structures. Reconceiving greenwashing as a failure of signal fit rather than as deceptive marketing communication, the study contributes to a process-oriented understanding of how consumers evaluate sustainability claims under uncertainty. Future research should quantitatively test these heuristic pathways across diverse regulatory and cultural environments. Full article

Background/Objective: Ultra-processed foods (UPFs) have become deeply embedded in global dietary patterns. However, their widespread consumption conceals the dual hidden risks of delayed physiological health effects and long-overlooked environmental externalities. Prior research has largely centered on health-driven dietary behaviors, with insufficient understanding of how perceptions of the environmental burden shape consumer choices, particularly in highly convenient, eating-out-dominated food environments. To address this gap, this study extends the theory of planned behavior (TPB) to examine how dual-risk perceptions influence intentions to reduce UPF consumption. Methods: Drawing on survey data from 362 Taiwanese consumers, this study analyzed the proposed theoretical model using structural equation modeling. Results: The findings show that (1) both health and environmental risk perceptions significantly and positively shape attitudes toward reducing UPF intake; (2) attitude, subjective norms, and perceived behavioral control (PBC) significantly increase reduction intentions, with subjective norms and attitude emerging as the strongest predictors; and (3) environmental awareness produces a counterintuitive diminishing marginal effect, negatively moderating the relationship between environmental burden perception and behavioral intention. Conclusions: These results extend the empirical foundation of the “green TPB” by demonstrating that the internalization of environmental costs complements traditional health motivations. The findings offer actionable implications for public health policy, including the implementation of front-of-pack warning labels and the use of the NOVA food classification system to advance sustainable diets. Full article

Quartz crystal microbalance (QCM) technology provides a powerful, label-free platform for monitoring molecular interactions in real time with nanogram sensitivity. Recent advances in compact instrumentation have enhanced analytical performance while reducing energy consumption, aligning with the principles of Green Analytical Chemistry. In parallel, the European Union has recommended the replacement of animal-derived antibodies with non-animal alternatives, creating an urgent need for sustainable affinity receptors. In this study, we present an innovative application of polynorepinephrine (PNE)-based molecularly imprinted polymers (MIPs) with a compact QCM sensing. PNE, a bioinspired polymer formed under mild aqueous conditions, offers strong adhesive properties and biocompatibility, enabling robust immobilization of imprinted receptors on gold-coated quartz disks. The resulting PNE-MIP/QCM platform combines the ultrasensitivity of quartz microbalances with the selectivity of molecular imprinting, delivering a reproducible and environmentally responsible affinity sensor. The sensor showed a limit of detection of 11.2 nM and enabled accurate IgG quantification in diluted human serum samples. As a proof of concept, the system was applied to Human Immunoglobulin G (IgG1) detection, demonstrating its potential for sustainable clinical diagnostics. Full article

Climate change constitutes a significant environmental challenge, and carbon footprint labeling has emerged as a key policy instrument to promote low-carbon consumption. This study examines long-term trends in consumer purchase intention for carbon-footprint-labeled products in Korea, drawing on nationwide survey data collected over an 11-year period (2009–2019). The results demonstrate that environmental concern, perceived severity of climate change, understanding of the carbon footprint system, and exposure to low-carbon and green media are positively associated with purchase intention. Among these factors, perceived contribution to greenhouse gas reduction is identified as the most influential and consistent determinant over time. Importantly, the analysis focuses on purchase intention rather than actual purchasing behavior. Given the well-documented attitude–behavior gap in sustainable consumption, the findings should be interpreted as indicative of behavioral intention. By documenting long-term changes in consumer perceptions and intentions, this research provides insights into the impact of environmental information on consumer decision-making and offers policy recommendations to enhance the effectiveness of carbon labeling. Full article

Continuous monitoring of the Stand-to-Sit (STS) transition serves as a critical indicator of lower-limb frailty in the elderly, for which millimeter-wave radar provides an ideal privacy-preserving, device-free sensing solution. However, robustly distinguishing between safe Controlled Sits (CSs) and dangerous Uncontrolled Descents (UDs) is severely hindered by the prohibitive cost of subjective expert scoring for fine-grained labels, alongside the pervasive “Clever Hans” effect where existing deep models overfit static environmental clutter rather than learning intrinsic human kinematics. To circumvent these bottlenecks, we formulate STS evaluation as a dynamic boundary detection problem and propose SCA-BiMamba, a linear-complexity bidirectional State-Space Model that utilizes actual fall events as extreme kinematic surrogates for UDs. This forces the network to learn a strict physical boundary between CS and physiological failure without subjective grading. Furthermore, we establish a stringent Real-to-Sim diagnostic audit as a core methodological contribution. By projecting models trained on noisy real-world data onto pure-kinematics simulations—incorporating stochastic temporal phase shifts, kinematic overlaps, and unified physiological tremors—we explicitly quantify feature disentanglement. This protocol serves as a formal ‘probing test’ to expose the ‘Clever Hans’ effect, ensuring the model relies on invariant human physics rather than transient environmental artifacts. Extensive experiments demonstrate that SCA-BiMamba achieves highly robust classification on real-world data (averaging 94.2% Macro F1 with 100.0% Uncontrolled Descent Recall), and achieves a highly robust 99.4% ± 1.1% Macro F1 in the simulated zero-shot transfer. We emphasize that this optimal performance reflects the successful abstraction of extreme kinematic boundaries, rather than a flawless resolution of all clinical complexities. Concurrently, it exhibits strict resistance to shortcut learning and sustains robust real-world scalability using merely 20% of the training data, thereby establishing a promising privacy-preserving boundary-based radar motion classification framework for distinguishing controlled sitting from extreme instability surrogates. Full article

The persistence of greenwashing as a strategic corporate behavior reflects a financial tradeoff between risk and return. Current literature lacks an integrative framework explaining how these risks and institutional arrangements vary across distinct contexts. This study maps the intellectual structure and contextual heterogeneity of corporate greenwashing research through a bibliometric analysis of 818 publications indexed in the Web of Science Core Collection from 2000 to 2025. The results indicate an evolutionary shift in research focus from early ethical and reputational debates toward empirical investigations of capital market consequences, ESG controversies, and the dark side of corporate sustainability. This transition is accompanied by thematic movement from voluntary disclosure and legitimacy concerns toward mandatory compliance, sustainable finance, green bond pricing, and digital detection using artificial intelligence and natural language processing. The analysis reveals substantial structural heterogeneity. Heavy-asset industries are closely associated with technological decoupling under physical and compliance constraints, whereas financial and service sectors rely heavily on information asymmetry, green label arbitrage, and greenhushing. These sectoral patterns intersect with regional governance trajectories shaped by market-driven, regulation-oriented, and state-led contexts, generating distinct incentive structures and risk conditions, while firm-level governance further moderates these behaviors. The findings position greenwashing as a context-dependent corporate strategy and provide a structured synthesis for future research and differentiated regulatory responses. Full article

This study examines whether tourism students’ self-reported time–money use patterns are related to their readiness to collaborate on idea development, and whether sustainability emerges spontaneously in their tourism innovation ideas. Using an anonymised dataset of open-ended questionnaire responses from Slovenian higher education tourism students (N = 597; 2019–2025), we applied deterministic rule-based coding to classify the presence of actionable ideas and sustainability framing, as well as collaboration readiness and conditions. Actionable ideas were common (53.4%), but sustainability framing was uncommon (7.5%). Most respondents were unconditionally willing to collaborate (69.3%), while 30.7% expressed conditional willingness or unwillingness. Time–money behavioural segments were significantly associated with collaboration reservations, whereas segment differences in ideation and sustainability framing were not significant. Among students expressing reservations, topic match and perceived team quality were the most frequently stated conditions. These findings indicate that sustainability-oriented tourism education should support both sustainability integration and low-risk collaboration through clear project briefs, topic-based matching, and team-process supports. The conclusions should be interpreted with reasonable caution as they are context-specific evidence based on self-reported, rule-coded responses, particularly for sustainability framing, where positive cases were rare. In this context, segmentation should be regarded as a diagnostic tool for course design rather than as a basis for labelling students. Full article

The growing interest in clean-label and naturally preserved foods has pushed the scientific community to research essential oils (EOs) as sustainable, multifunctional alternatives to chemical preservatives. These plant volatile compounds exhibit strong antimicrobial and antioxidant activities, making them promising ingredients for natural preservation. Fermented meat products, though highly nutritional, are particularly at risk of microbial spoilage and contamination by foodborne pathogens due to their complex microbiota and processing conditions. This review examines the role of EOs as natural antimicrobials in fermented meat systems, summarizing their mechanisms of action, efficiency against key pathogens, and impact on safety, shelf life, and sensory attributes. Additionally, it discusses technological challenges related to volatility, stability, and sensory alterations, and outlines mitigation strategies such as encapsulation, nanoemulsions, and controlled-release delivery systems. By critically presenting current progress and identifying research gaps such as standardization and matrix interactions, this review contributes to the development of effective, natural, and clean-label preservation strategies. These insights support innovation and sustainability in the meat processing industry by bridging the gap between antimicrobial efficacy and sensory acceptability. Full article

Artificial Intelligence (AI) has traditionally been developed according to a model-centric paradigm, in which progress is driven by increasingly sophisticated learning architectures applied to largely fixed datasets. However, this paradigm exhibits well-known limitations, including sensitivity to label noise, distribution shifts, adversarial perturbations, and limited transparency and reproducibility. These issues indicate that many of the current bottlenecks of AI systems arise from deficiencies in data rather than from model design. In this paper, we adopt and formalize the Data-Centric Artificial Intelligence (DCAI) paradigm, which places data quality, semantic consistency, and representativeness at the core of the AI lifecycle. From this perspective, performance, robustness, interpretability, and regulatory compliance are primarily achieved through systematic data engineering, including data curation, enrichment, validation, and continuous monitoring, rather than through repeated model re-engineering. The contributions of this work are threefold. First, a conceptual framework is provided to clarify the epistemic and methodological foundations of DCAI and distinguish it from traditional model-centric approaches. Second, a data-centric lifecycle is presented, covering training data development, inference data design, and data maintenance and integrating techniques such as semantic data representation, active learning, synthetic data generation, and drift-aware quality control. Third, the role of DCAI in the context of Generative AI is analyzed, showing how data-centric practices are essential to ensure robustness, accountability, and responsible deployment of large-scale generative models. Overall, this work positions DCAI as a coherent methodological and technological framework for the development of trustworthy, resilient, and sustainable AI systems, making a research contribution and providing a reference model for industrial and regulatory contexts. Full article

Spotted-wing Drosophila (Drosophila suzukii), an economically important invasive but widely distributed pest, has developed resistance to multiple insecticide classes, threatening its management in commercial soft fruit production. This study evaluated the synergism of piperonyl butoxide (PBO) with zeta-cypermethrin in two field-collected resistant California populations and a susceptible population with bioassays. Female flies from the two resistant populations exhibited 55-fold and 25-fold resistance, respectively, compared to the susceptible population. PBO co-application significantly enhanced insecticide efficacy in both resistant populations, with synergism ratios of 6.51 and 4.06. However, complete susceptibility at label rates of the insecticide was not restored, indicating that other resistance mechanisms may also be present in these populations. PBO–pyrethroid combinations show promise for improving field efficacy against resistant populations, though they should be integrated with insecticide rotation and other integrated pest management tactics for sustainable resistance management. Full article

Artificial Intelligence systems increasingly shape environmental decision making, infrastructure planning, and resource use across public and urban domains. However, prevailing AI trust and governance mechanisms, including labels, certifications, and assurance schemes, remain primarily focused on ethical and legal accountability, with limited operational attention to environmental sustainability. This paper reconceptualises AI trust mechanisms as socio-technical governance infrastructures that can support both ethical assurance and environmental accountability. Drawing on a comparative qualitative analysis of nine AI trust initiatives, the study develops a three-dimensional analytical framework embedding Environmental Performance Indicators across three governance dimensions: trust-building effectiveness, governance readiness, and sustainable adoption. Applying a systems governance lens, the framework examines how governance instruments structure information flows, institutional practices, and lifecycle feedback relevant to environmental performance. It is analytically illustrated through two urban mobility cases, Helsinki’s Whim application and Barcelona’s smart mobility system, to examine how governance conditions enable or constrain the integration of Environmental Performance Indicators in practice. Findings show that current trust mechanisms lack measurable and publicly visible environmental criteria, indicating a gap between AI assurance and environmental governance. The study contributes a systems-oriented framework for evaluating AI trust mechanisms as governance instruments capable of supporting environmental accountability. While exploratory and based on secondary data, the results indicate that future AI trust mechanisms must incorporate measurable sustainability indicators to support eco-efficient and accountable digital transformation. Full article