Mountain grasslands of the Central Pyrenees are highly sensitive socio-ecological systems traditionally managed under extensive grazing, where soil properties, vegetation dynamics and microbial processes are tightly coupled. However, the effects of grazing are often assessed separately for either soil or forage, which constrains integrated evaluations of overall ecosystem functioning. In this study, we compared two subalpine grassland plots subjected to contrasting grazing intensities (moderate-high vs. moderate-low), integrating soil physical, chemical and biological indicators with forage nutritional quality. Grazing intensity was characterized using GPS-derived cattle movement data from five adult animals during the grazing season. Soils under lower grazing intensity retained more water, accumulated larger organic carbon pools and exhibited longer carbon residence times together with higher microbial functional diversity. By contrast, higher grazing intensity favored labile carbon dynamics, with more dissolved organic carbon and higher microbial respiration. No significant differences were observed in mesofauna diversity, total glomalin content or β-glucosidase activity. Although forage composition was dominated by mixed grasses and legumes, nutritional quality was low across all plots, due to high fiber content and deficiencies in Zn, Cu and Se. Multivariate analyses showed that soil carbon and hydrological status drive ecosystem variability through coordinated soil–microbial–forage responses. Our results suggest that even subtle variations in grazing intensity generate integrated functional shifts across ecosystem compartments. Under ongoing climate change, such tightly coupled systems may be particularly vulnerable to grazing mismanagement, emphasizing the need for adaptive strategies that integrate soil health, microbial functioning and forage quality to sustain mountain grassland resilience.
Automatising the process of verifying whether the design of a building conforms to a certain regulation is an important task to save time and money. This paper presents SmartNorms4BIM, a building regulatory compliance-checking system that facilitates the automatic evaluation of Building Information Modelling (BIM) models. The system’s architecture follows Eastman’s four stages: translation of norms into a machine-readable format, BIM model preparation, execution of the verification process, and report generation. Our proposal uses Semantic Web technologies (OWL 2 ontologies and semantic reasoners) to represent the knowledge and detect violations of the norms automatically, Large Language Models (LLMs) to interpret and transform building norms into OWL 2 ontologies, fuzzy logic to manage vagueness in the norms, and specific algorithms to reason with geometric knowledge. A Java desktop prototype was developed to determine whether a BIM model complies with a given set of norms. To validate our approach, two real-world use cases were studied.
Poplar wood is an important bioresource used in the manufacture of plywood, fibreboard, paper pulp, and a wide range of other products. The accumulation of poplar sawdust in industrial environments represents a significant hazard when exposed to hot surfaces. Despite the widespread use of this material, fire and explosion safety data for poplar dust remain notably scarce. The Minimum Ignition Temperature of a dust layer (MITL) is a key parameter for assessing its susceptibility to ignition. In this study, a series of hot plate experiments was conducted on different fractions of poplar sawdust to investigate the influence of particle size, layer thickness and bulk density on the ignition temperature. The results indicate that MITL decreased with increasing layer thickness, whereas particle size had little influence on ignition behaviour. The latter finding was not observed for samples of pine pellet dust, which followed the trend commonly reported in the literature. Compaction of the poplar samples reduced MITL by 10 degrees C, thereby increasing their susceptibility to ignition, although one sample did not follow this trend. The results suggest that the process is heat-limited and that bulk porosity -or compaction- may play an important role, potentially explaining the contrasting behaviour of the two materials. In addition, the Frank-Kamenetskii thermal explosion model was applied to the experimental results to estimate the activation energy and extrapolate the MITL values to thicker dust layers. The prediction for 50 mm-thick layers deviated significantly from the general recommendations for dust layers provided in the EN 60079-14 standard.
The construction of higher order uniformly convergent methods for the numerical resolution of any type of singularly perturbed problems, is an interesting task in applied mathematics. The main reason is that those methods permit obtaining good and efficient numerical approximations without increasing the computational cost of the numerical algorithm. Here, we study 2D elliptic weakly coupled systems of convection diffusion type, for which small positive parameters appear in both the diffusion and the convection terms. Moreover, we assume that the diffusion parameters at each equation of the system are different and the convection term has only a non zero component in one of the two spatial directions. Then, the exact solution has parabolic and regular layers in the boundary of the domain, which depend on the value and the ratio between the three small parameters. To solve the continuous problem, an hybrid finite difference scheme is used, which is constructed on an adequate Bakhvalov-Shishkin mesh; then, the scheme is a second order uniformly convergent method; this result is better than the obtained by previous methods existing in the literature to solve the same type of systems. We show the numerical results that the computational algorithm gives for a test problem; from them, we clearly observe the order of uniform convergence of the method and also its efficiency due the low computational cost needed to achieve the numerical results.
Two of the most relevant issues in multifactor decision-making are rationality and representativeness. Rationality refers to the consistency of individuals when incorporating their preferences into the formal model (individual coherence) whilst representativeness refers to the compatibility or the individuals’ acceptance of the collective position (group coherence). This paper presents for the first time in the literature a unified (methodologically) and closed (mathematically) framework for the joint improvement of rationality and representativeness in a local AHP-Group Decision Making context in which the collective position is given by a group pairwise comparison matrix. Based on maximising efficiency, the proposal works with slight modifications in relative terms, uncommon in this field, of the judgements of the collective matrix without modifying the initial individual positions. Optimisation results for the judgement that most rapidly increases the efficiency and the value that achieve the maximum improvement are provided. Significant improvements are attained and valuable information for learning and negotiation is gathered. The flexibility of this unified and closed framework allows its adaptation to different interests and situations. Simulation studies and illustrative examples demonstrate the robustness, effectiveness and potential of the framework.
