Search Results (14)

This paper presents analytical approximations for the inverse error function, its complementary inverse, and the cumulative distribution function using the Power Series Extender Method (PSEM). The proposed expressions exhibit high accuracy over a wide portion of the domain, particularly in the central region, while maintaining a compact structure based on elementary functions. This formulation ensures practical implementation and computational efficiency without the need for specialized numerical algorithms. The use of strategically selected cancellation points further enhances the accuracy of the approximations, especially in regions of interest. As expected for this class of elementary approximations, a gradual loss of accuracy is observed near the boundaries of the domain due to the asymptotic behavior of the inverse functions. To demonstrate the effectiveness and practical relevance of the proposed expressions, two case studies are presented, involving applications in statistical analysis and engineering contexts. Full article

Soil organic carbon (SOC) storage in forests is governed by complex interactions between site conditions and vegetation. This study quantifies SOC stocks across a gradient of Target Management Sets (TMS) in the Czech Republic (Central Europe) to evaluate the baseline storage capacity of distinct ecological sites and the modifying effects of dominant tree species, specifically Norway spruce and European beech. Utilizing large-scale spatial data, linear mixed-effects models, and piecewise structural equation modeling (pSEM), we analyzed SOC stratification across middle (≈400–600 m a.s.l.) and higher (≈600–800 m a.s.l.) elevational zones. The results indicate that while overall SOC stocks inherently increase with elevation due to climatic constraints, tree species dictate the vertical carbon distribution within the soil profile. Specifically, conifers (i.e., Norway spruce and Scots pine) accumulate SOC primarily in the organic layer, whereas broadleaves (mainly European beech and oak) translocate and stabilize carbon in deeper mineral horizons. The pSEM analysis revealed that beech functions as a ‘calcium pump’, increasing topsoil pH and driving calcium-mediated SOC stabilization in mineral soils. This mechanism is highly effective at middle elevations but partially overridden by abiotic limits at higher elevations. We conclude that inherent site conditions (TMS) determine total SOC capacity, whereas tree species management controls SOC stability. Although no significant differences were observed in total SOC stocks between conifers and broadleaves at the same sites (medians of total SOC ranged from approx. 5 to 16 kg·m⁻², depending on the site), converting purely coniferous stands into broadleaves represents an effective strategy for long-term mineral SOC stabilization, particularly in middle-elevation sites. Full article

Understanding how environmental conditions, functional composition, and species richness jointly relate to phylogenetic community structure is important for sustainable forest management under environmental change. Using 2858 plots from the 7th National Forest Inventory of South Korea, phylogenetic community structure as the standardized effect size of mean pairwise phylogenetic distance (SES.MPD) was quantified for broadleaved, conifer, and mixed stands. Associations between SES.MPD and abiotic factors such as elevation, mean annual precipitation, stand age, as well as biotic factors such as species richness and community-weighted means of specific leaf area and maximum height were assessed using multi-model inference and piecewise structural equation models (pSEM). Across stand types, SES.MPD values in most plots were not significantly different from the null-model baseline under the chosen randomization procedure, indicating weak net departures from null-model-relative phylogenetic dispersion at the national scale; meanwhile, mean SES.MPD tended to be slightly negative in broadleaved stands and positive in conifer and mixed stands. In multi-model inference analysis, the strongest associations with SES.MPD differed among stand types: trait composition—especially community-weighted specific leaf area—showed the strongest association in total stands and broadleaved stands, whereas species richness was the dominant correlate in mixed stands and precipitation showed the strongest association in conifer stands. The pSEM revealed that elevation, precipitation, and stand age were linked to SES.MPD largely through indirect pathways via trait composition and species richness, consistent with trait- and richness-mediated environmental filtering. These results highlight stand-type-specific linkages among environment, traits, richness, and phylogenetic structure and support trait- and phylogeny-informed forest management and restoration to enhance resilience under ongoing environmental change. Full article

Enhancing forest biodiversity and carbon sinks in the face of climate change is a high priority on the global agenda. The aim of our study was to explore the feasibility and potential of enhancing biodiversity and stand biomass productivity, which are strongly linked to forest ecosystem functioning and services in temperate forests. Based on data from the 5th to 7th National Forest Inventory of South Korea, 1760 natural forest plots (0.16 ha) were used, of which 344 plots belonged to conifer stands, 711 plots belonged to broadleaved stands, and 705 plots belonged to mixed stands. Forest succession-related factor (i.e., stand age), and abiotic (i.e., climatic and topographic conditions, and soil properties) and biotic drivers (i.e., species diversity, functional trait diversity, functional trait identity, and stand structural diversity) were jointly included as independent variables in an integrated model to explain variations in stand biomass productivity. In order to reveal the key drivers and relationships that regulate stand biomass productivity across forest stand types, we applied a multi-model averaging approach and piecewise structural equation modelling (pSEM). As a key finding, across all forest stand types, forest stand age-induced tree size inequality (i.e., DBH STD) in all forest stand types commonly increased stand biomass productivity, showing strong positive standardized effects (β > 0.5, p < 0.001). We also found that the functional trait identities controlling stand biomass productivity within each forest stand type differed according to their functional traits of dominant species, and that these mechanisms were controlled directly or indirectly by environmental conditions. Our research suggests that appropriate forest management plans should be developed in accordance with environmental gradients to simultaneously promote biodiversity and stand biomass productivity in different forest stand types. Full article

This study investigates the factors influencing urban residents’ interest in and visits to forests and explores strategies to promote forest space utilization. A survey was conducted among 5000 residents of Tokyo’s 23 wards, one of the world’s most densely populated urban areas, using an online questionnaire. The collected data were analyzed using least absolute shrinkage, selection operator (LASSO) logistic regression, and piecewise structural equation modeling (pSEM). The analysis revealed that nature experiences in current travel destinations, particularly scenic walks, had a significant positive effect on both forest interest (standardized path coefficient = 0.19) and forest visits (0.30). These experiences were also significantly influenced by childhood nature experiences and frequent local walks. Conversely, factors negatively affecting forest visits included the lack of private vehicle ownership (−0.13) and increasing age (−0.21). While previous studies suggest that older individuals tend to visit natural areas more frequently, our findings indicate the opposite trend. One possible explanation is the low car ownership rate among Tokyo residents, which may limit accessibility to forests. These findings provide valuable insights for policy design, particularly regarding strategies to enhance forest accessibility and engagement among urban populations. Full article

This study investigates the psychological factors influencing remote work acceptance during the COVID-19 pandemic using a Bayesian network and probabilistic structural equation modeling (PSEM) approach. The research specifically explores the impact of stress factors induced by the pandemic, including compulsive checking, socio-economic consequences, danger, and contamination, on individuals’ willingness to adopt remote work arrangements. Data were collected from 586 participants with remote work experience, and the Bayesian analysis revealed that compulsive checking had the most significant positive influence on remote work acceptance, followed by socio-economic consequences, while danger and contamination showed no statistically significant effects. The findings suggest that psychological stress factors related to excessive information seeking and economic instability play a stronger role in influencing remote work decisions than direct health-related concerns. These results provide theoretical contributions by extending technology acceptance models to crisis situations and offer practical insights for organizations aiming to implement effective remote work policies. Specifically, strategies such as psychological support programs, financial stability measures, and clear communication regarding workplace safety can enhance employee well-being and productivity in remote settings. Future research should explore long-term effects of stress factors on remote work sustainability and cross-cultural comparisons in pandemic-related work adaptations. Full article

While a flurry of studies and Integrated Assessment Models (IAMs) have independently investigated the impacts of switching mitigation policies in response to different climate scenarios, little is understood about the feedback effect of how human risk perceptions of climate change could contribute to switching climate mitigation policies. This study presents a novel machine learning approach, utilizing a probabilistic structural equation model (PSEM), for understanding complex interactions among climate risk perceptions, beliefs about climate science, political ideology, demographic factors, and their combined effects on support for mitigation policies. We use machine learning-based PSEM to identify the latent variables and quantify their complex interaction effects on support for climate policy. As opposed to a priori clustering of manifest variables into latent variables that is implemented in traditional SEMs, the novel PSEM presented in this study uses unsupervised algorithms to identify data-driven clustering of manifest variables into latent variables. Further, information theoretic metrics are used to estimate both the structural relationships among latent variables and the optimal number of classes within each latent variable. The PSEM yields an R² of 92.2% derived from the “Climate Change in the American Mind” dataset (2008–2018 [N = 22,416]), which is a substantial improvement over a traditional regression analysis-based study applied to the CCAM dataset that identified five manifest variables to account for 51% of the variance in policy support. The PSEM uncovers a previously unidentified class of “lukewarm supporters” (~59% of the US population), different from strong supporters (27%) and opposers (13%). These lukewarm supporters represent a wide swath of the US population, but their support may be capricious and sensitive to the details of the policy and how it is implemented. Individual survey items clustered into latent variables reveal that the public does not respond to “climate risk perceptions” as a single construct in their minds. Instead, PSEM path analysis supports dual processing theory: analytical and affective (emotional) risk perceptions are identified as separate, unique factors, which, along with climate beliefs, political ideology, and race, explain much of the variability in the American public’s support for climate policy. The machine learning approach demonstrates that complex interaction effects of belief states combined with analytical and affective risk perceptions; as well as political ideology, party, and race, will need to be considered for informing the design of feedback loops in IAMs that endogenously feedback the impacts of global climate change on the evolution of climate mitigation policies. Full article

Deep metric learning aims to learn a mapping function that projects input data into a high-dimensional embedding space, facilitating the clustering of similar data points while ensuring dissimilar ones are far apart. The most recent studies focus on designing a batch sampler and mining online triplets to achieve this purpose. Conventionally, hard negative mining schemes serve as the preferred batch sampler. However, most hard negative mining schemes search for hard examples in randomly selected mini-batches at each epoch, which often results in less-optimal hard examples and thus sub-optimal performances. Furthermore, Triplet Loss is commonly adopted to perform online triplet mining by pulling the hard positives close to and pushing the negatives away from the anchor. However, when the anchor in a triplet is an outlier, the positive example will be pulled away from the centroid of the cluster, thus resulting in a loose cluster and inferior performance. To address the above challenges, we propose the Prototype-based Support Example Miner (pSEM) and Triplet Loss (pTriplet Loss). First, we present a support example miner designed to mine the support classes on the prototype-based nearest neighbor graph of classes. Following this, we locate the support examples by searching for instances at the intersection between clusters of these support classes. Second, we develop a variant of Triplet Loss, referred to as a Prototype-based Triplet Loss. In our approach, a dynamically updated prototype is used to rectify outlier anchors, thus reducing their detrimental effects and facilitating a more robust formulation for Triplet Loss. Extensive experiments on typical Computer Vision (CV) and Natural Language Processing (NLP) tasks, namely person re-identification and few-shot relation extraction, demonstrated the effectiveness and generalizability of the proposed scheme, which consistently outperforms the state-of-the-art models. Full article

Pollinator-mediated selection is supposed to influence floral integration. However, the potential pathway through which pollinators drive floral integration needs further investigations. We propose that pollinator proboscis length may play a key role in the evolution of floral integration. We first assessed the divergence of floral traits in 11 Lonicera species. Further, we detected the influence of pollinator proboscis length and eight floral traits on floral integration. We then used phylogenetic structural equation models (PSEMs) to illustrate the pathway through which pollinators drive the divergence of floral integration. Results of PCA indicated that species significantly differed in floral traits. Floral integration increased along with corolla tube length, stigma height, lip length, and the main pollinators’ proboscis length. PSEMs revealed a potential pathway by which pollinator proboscis length directly selected on corolla tube length and stigma height, while lip length co-varied with stigma height. Compared to species with short corolla tubes, long-tube flowers may experience more intense pollinator-mediated selection due to more specialized pollination systems and thus reduce variation in the floral traits. Along elongation of corolla tube and stigma height, the covariation of other relevant traits might help to maintain pollination success. The direct and indirect pollinator-mediation selection collectively enhances floral integration. Full article

Forests regulate air quality and respond to climate change by storing carbon. Assessing the driving factors of forest aboveground carbon (AGC) storage is of great importance for forest management. We assumed that different forest types would affect the relationship between species richness, stand density, individual tree size variation, and AGC. In order to test and verify it, we analyzed the inventory data of 206 fixed plots (20 m × 20 m) of Jingouling Forest Farm, taking advantage of the piecewise structural equation model (pSEM) to explore the effects of species diversity, stand structure attributes, and topography on the AGC storage in the Wangqing Forest in Jilin Province. In addition, in this study, we aimed to investigate whether the fixed factors (species diversity, stand structure attributes, and topography) influenced AGC storage more significantly than the random factor (forest type). According to the results of pSEM, the selected factors jointly explain the impact on 33% of AGC storage. The relationship between stand density and AGC is positive, and the impact of individual tree size variation on AGC storage is negative. Species richness has direct and indirect impacts on AGC storage, and the indirect impact is more significant through individual tree size variation. Both elevation and slope are significantly negatively associated with AGC storage. Forest type explains the impact on 12% of AGC storage, which means the relationship between AGC and predictors varies across forest types. The results provide a scientific basis for the protection and management decision of natural forests in northeastern China. Full article

Japanese cedar (Cryptomeria japonica) is the most important timber species in Japan; however, its pollen is the primary cause of pollinosis in Japan. The total number of pollen grains produced by a single tree is determined by the number of male strobili (male flowers) and the number of pollen grains per male strobilus. While the number of male strobili is a visible and well-investigated trait, little is known about the number of pollen grains per male strobilus. We hypothesized that genetic and environmental factors affect the pollen number per male strobilus and explored the factors that affect pollen production and genetic variation among clones. We counted pollen numbers of 523 male strobili from 26 clones using a cell counter method that we recently developed. Piecewise Structural Equation Modeling (pSEM) revealed that the pollen number is mostly affected by genetic variation, male strobilus weight, and pollen size. Although we collected samples from locations with different environmental conditions, statistical modeling succeeded in predicting pollen numbers for different clones sampled from branches facing different directions. Comparison of predicted pollen numbers revealed that they varied >3-fold among the 26 clones. The determination of the factors affecting pollen number and a precise evaluation of genetic variation will contribute to breeding strategies to counter pollinosis. Furthermore, the combination of our efficient counting method and statistical modeling will provide a powerful tool not only for Japanese cedar but also for other plant species. Full article

In this study, a novel metal matrix composite based on 60 vol% 316L stainless steel and 40 vol% MgO manufactured by powder metallurgy technology was developed. The corrosion resistance of the developed steel–MgO composite material against molten aluminum alloy AlSi7Mg0.3 was investigated by means of wettability tests and long-term crucible corrosion tests. The wettability tests were carried out using the sessile drop method with the capillary purification technique in a hot-stage microscope (HSM). Static corrosion tests were performed in molten aluminum alloy at 850 °C for 168 h to evaluate the impact of pre-oxidation of the composite surface on the corrosion resistance. The pre-oxidation of steel–MgO composites was carried out at 850 and 1000 °C for 24 h, based on preliminary investigations using thermogravimetry (TG) and dilatometry. The influence of the pre-oxidation on the composite structure, the corrosion resistance, and the phase formation at the interface between the steel–MgO composite and aluminum alloy was analyzed using SEM/EDS and XRD. The impact of the steel–MgO composite material on the composition of the aluminum alloy regarding the type, size, and quantity of the formed precipitations was investigated with the aid of ASPEX PSEM/AFA and SEM/EBSD. It was revealed that the pre-oxidation of the steel–MgO composite at 1000 °C induced the formation of stable MgO-FeO solid solutions on its surface, leading to a significant increase of long-term corrosion resistance against the liquid aluminum alloy. Full article

The effect of Mg treatment on the microstructure and toughness of the heat-affected zone (HAZ) of shipbuilding steel after high-heat-input welding was investigated via laboratory and industrial testing. The welding process and Charpy impact tests were also carried out to evaluate the HAZ toughness of steel plates. First, typical inclusion characteristics were characterised with an ASPEX PSEM Explorer. Then, confocal laser scanning microscopy (CLSM) was used to observe the diameters of austenite grains under different holding times. The results showed that when the addition of microalloy elements were in the order of Al–Mg–Ti, this had an effect on dispersing inclusions, the largest proportion of which were micro-inclusions that had a particle size range of 1.0–1.5 μm. This accounted for 25.4% of the total inclusions, which was the highest amount. The micro inclusion particle size that was mainly distributed in the range of 0.5–3.5 μm accounted for 82.8% of all the micro-inclusions. The inclusion structure induced intragranular acicular ferrite (IAF) in austenite as follows: MgO and Al₂O₃ formed the core and Ti₂O₃ adhered to the Al–Mg complex inclusions to produce smaller particle sizes and dispersions of Al, Mg, and Ti complex inclusions. The 40-mm-thick plate obtained in the industrial test after welding had an average impact absorbed energy 2 mm from the weld joint in the heat-affected zone of 198.9 J at −20 °C, while the welding heat input was 150 kJ/cm, compared with the parent material’s low-temperature performance, which exceeded 88%. Full article

Two methods based on the principle of maximum entropy (POME), the ordinary entropy method (ENT) and the parameter space expansion method (PSEM), are developed for estimating the parameters of a four-parameter exponential gamma distribution. Using six data sets for annual precipitation at the Weihe River basin in China, the PSEM was applied for estimating parameters for the four-parameter exponential gamma distribution and was compared to the methods of moments (MOM) and of maximum likelihood estimation (MLE). It is shown that PSEM enables the four-parameter exponential distribution to fit the data well, and can further improve the estimation. Full article