Search Results (1,318)

Wildfire regimes are closely linked to changes in landscape structure, yet the influence of accelerated land use transitions on fire activity remains poorly understood, particularly in rapidly transforming regions like central Chile. Although land use change has been extensively documented in the country, the specific role of the speed, extent, and spatial configuration of these transitions in shaping fire dynamics requires further investigation. To address this gap, we examined how landscape transitions influence fire frequency in central Chile, a region experiencing rapid land use change and heightened fire activity. Using multi-temporal remote sensing data, we quantified land use transitions, calculated landscape metrics to describe their spatial characteristics, and applied intensity analysis to assess their relationship with fire frequency changes. Our results show that accelerated landscape transitions significantly increased fire frequency, particularly in areas affected by forest plantation rotations, new forest establishment, and urban expansion, with changes exceeding uniform intensity expectations. Regional variations were evident: In the more densely populated northern areas, increased fire frequency was primarily linked to urban development and deforestation, while in the more rural southern regions, forest plantation cycles played a dominant role. Areas with a high number of large forest patches were especially prone to fire frequency increases. These findings demonstrate that both the speed and spatial configuration of landscape transitions are critical drivers of wildfire activity. By identifying the specific land use changes and landscape characteristics that amplify fire risks, this study provides valuable knowledge to inform fire risk reduction, landscape management, and urban planning in Chile and other fire-prone regions undergoing rapid transformation. Full article

As urbanization processes are no longer characterized by simple linear expansion but exhibit leaping, edge-sparse, and discontinuous features, spatiotemporally continuous impervious surface coverage data are needed to better characterize urbanization processes. This study utilized GAIA impervious surface binary data and employed spatiotemporal aggregation methods to convert thirty years of 30 m resolution data into 1 km resolution spatiotemporal impervious surface coverage data, constructing a long-term time series annual impervious surface coverage dataset for the Beijing–Tianjin–Hebei region. Based on this dataset, we analyzed urban expansion processes and landscape pattern indices in the Beijing–Tianjin–Hebei region, exploring the spatiotemporal response relationships of ecological environment changes. Results revealed that the impervious surface area increased dramatically from 7579.3 km² in 1985 to 37,484.0 km² in 2020, representing a year-on-year growth of 88.5%. Urban expansion rates showed two distinct peaks: 800 km²/year around 1990 and approximately 1700 km²/year during 2010–2015. In high-density urbanized areas with impervious surfaces, the average forest area significantly increased from approximately 2500 km² to 7000 km² during 1985–2005 before rapidly declining, grassland patch fragmentation intensified, while in low-density areas, grassland area showed fluctuating decline with poor ecosystem stability. Furthermore, by incorporating natural and social factors such as Fractional Vegetation Coverage (FVC), Habitat Quality Index (HQI), Land Surface Temperature (LST), slope, and population density, we assessed the vulnerability of urbanization development in the Beijing–Tianjin–Hebei region. Results showed that high vulnerability areas (EVI > 0.5) in the Beijing–Tianjin core region continue to expand, while the proportion of low vulnerability areas (EVI < 0.25) in the northern mountainous regions decreased by 4.2% in 2020 compared to 2005. This study provides scientific support for the sustainable development of the Beijing–Tianjin–Hebei urban agglomeration, suggesting location-specific and differentiated regulation of urbanization processes to reduce ecological risks. Full article

The Mapleton Falls National Park transmission line corridor in Queensland, Australia, has received a number of vegetation management treatments over the last decade to maintain and protect the infrastructure and to ensure continuous electricity supply. Recent treatments have included ‘mega-mulching’ (mechanical mastication of vegetation to a mulch layer) in 2020 and targeted herbicide treatment of woody vegetation, with the aim of reducing vegetation height by encouraging a native herbaceous groundcover beneath the transmission lines. We measured vegetation structure (cover and height) and composition (species presence in 15 × 2 m plots), at 12 transects, 90 m in length on the transmission line corridor, to determine if management goals were being achieved and to determine how the vegetation and fire hazard (based on the overall fuel hazard assessment method) varied among the treated corridor, the forest edge environment, and the natural forest. The results showed that vegetation structure and composition in the treated zones had been modified to a state where herbaceous plant species were dominant; there was a significantly (p < 0.05) higher native grass cover and cover of herbs, sedges, and ferns in the treated zones, and a lower cover of trees and tall woody plants (>1 m in height) in these areas. For example, mean native grass cover and the cover of herbs and sedges in the treated areas was 10.2 and 2.8 times higher, respectively, than in the natural forest. The changes in the vegetation structure (particularly removal of tall woody vegetation) resulted in a lower overall fuel hazard in the treated zones, relative to the edge zones and natural forest. The overall fuel hazard was classified as ‘high’ in 83% of the transects in the treated areas, but it was classified as ‘extreme’ in 75% of the transects in the adjacent forest zone. Importantly, there were few introduced species recorded. The results suggest that fuel management has been successful in reducing wildfire risk in the transmission corridor. Temporal monitoring is recommended to determine the frequency of ongoing fuel management. Full article

Landslides are destructive geological hazards, making accurate landslide susceptibility assessment essential for disaster prevention and mitigation. However, existing studies often lack scientific rigor in negative sample construction and have unclear model applicability. This study focuses on Ya’an City, Sichuan Province, China, and proposes an innovative approach to negative sample construction using Geographically Weighted Regression (GWR), which is then integrated with Tabular Network (TabNet), a deep learning architecture tailored to structured tabular data, to assess landslide susceptibility. The performance of TabNet is compared against Random Forest, Light Gradient Boosting Machine, deep neural networks, and Residual Networks. The experimental results indicate that (1) the GWR-based sampling strategy substantially improves model performance across all tested models; (2) TabNet trained using the GWR-based negative samples achieves superior performance over all other evaluated models, with an average AUC of 0.9828, exhibiting both high accuracy and interpretability; and (3) elevation, land cover, and annual Normalized Difference Vegetation Index are identified as dominant predictors through TabNet’s feature importance analysis. The results demonstrate that combining GWR and TabNet substantially enhances landslide susceptibility modeling by improving both accuracy and interpretability, establishing a more scientifically grounded approach to negative sample construction, and providing an interpretable, high-performing modeling framework for geological hazard risk assessment. Full article

The accurate estimation of urban property values is a key challenge for appraisers, market participants, financial institutions, and urban planners. In recent years, machine learning (ML) techniques have emerged as promising tools for price forecasting due to their ability to model complex relationships among variables. However, their application raises two main critical issues: (i) the risk of overfitting, especially with small datasets or with noisy data; (ii) the interpretive issues associated with the “black box” nature of many models. Within this framework, this paper proposes a methodological approach that addresses both these issues, comparing the predictive performance of three ML algorithms—k-Nearest Neighbors (kNN), Random Forest (RF), and the Artificial Neural Network (ANN)—applied to the housing market in the city of Salerno, Italy. For each model, overfitting is preliminarily assessed to ensure predictive robustness. Subsequently, the results are interpreted using explainability techniques, such as SHapley Additive exPlanations (SHAPs) and Permutation Feature Importance (PFI). This analysis reveals that the Random Forest offers the best balance between predictive accuracy and transparency, with features such as area and proximity to the train station identified as the main drivers of property prices. kNN and the ANN are viable alternatives that are particularly robust in terms of generalization. The results demonstrate how the defined methodological framework successfully balances predictive effectiveness and interpretability, supporting the informed and transparent use of ML in real estate valuation. Full article

Early childhood forest and nature education plays a vital role in shaping values and promoting sustainability throughout life. Conceptualized in Denmark, forest and nature childcare groups have been established in Austria for over 20 years, contributing to mental well-being and supporting both Education for Sustainable Development (ESD) and Early Childhood Education and Care (ECEC). With increasing demand for childcare and a growing disconnect from nature—factors linked to physical and mental health challenges—there is a pressing need to expand these groups and integrate them into formal legal frameworks. This study examines the organization, staffing, infrastructure, risk prevention, and hygiene of 79 Austrian forest and nature kindergarten groups, identifying key areas of improvement to ensure safe access for all children, including those in public childcare. A semi-standardized online survey of 72 groups was analyzed using descriptive and statistical methods, including a Spearman correlation, Kruskal–Wallis test, Chi-square test, and ANOVA. Results revealed three main infrastructure types—house, container/trailer, and tipi—with houses offering the most comprehensive facilities. The ANOVA indicated significant effects of sponsorship type (p < 0.01), caregiver numbers (p < 0.001), and their interaction (p < 0.05) on half-day care costs. Currently, legal frameworks exist only in Tyrol and Salzburg. Broader access requires standardized infrastructure and risk assessment guidelines, collaboratively developed with stakeholders, to ensure safety and inclusivity in Austrian forest and nature childcare groups. Full article

Atmospheric corrosion significantly impacts infrastructure worldwide, with traditional assessment methods being time-intensive and costly. This study developed a comprehensive machine learning framework for predicting atmospheric corrosion rates using environmental and material parameters. Three regression models (Linear Regression, Random Forest, and Gradient Boosting) were trained on a scientifically informed synthetic dataset incorporating established corrosion principles from ISO 9223 standards and peer-reviewed literature. The Gradient Boosting model achieved superior performance with cross-validated R² = 0.835 ± 0.024 and RMSE = 98.99 ± 16.62 μm/year, significantly outperforming the Random Forest (p < 0.001) and Linear Regression approaches. Feature importance analysis revealed the copper content (30%), exposure time (20%), and chloride deposition (15%) as primary predictors, consistent with the established principles of corrosion science. Model diagnostics demonstrated excellent predictive accuracy (R² = 0.863) with normally distributed residuals and homoscedastic variance patterns. This methodology provides a systematic framework for ML-based corrosion prediction, with significant implications for protective coating design, material selection, and infrastructure risk assessment, pending comprehensive experimental validation. Full article

The use of the Soil and Water Assessment Tool (SWAT) model driven by China Meteorological Assimilation Driving Datasets (CMADS) for runoff simulation research is of great significance for regional flood prevention and control. Therefore, from the perspective of production-living-ecological space, this article combined multi-source data such as DEM, soil texture and land use type, in order to construct scenarios of territorial spatial change (TSC) across distinct periods. Based on the CMADS-L40 data and the SWAT model, it simulated the runoff dynamics in the Xitiaoxi River Basin, and analyzed the hydrological response characteristics under different TSCs. The results showed that The SWAT model, driven by CMADS-L40 data, demonstrated robust performance in monthly runoff simulation. The coefficient of determination (R²), Nash–Sutcliffe efficiency coefficient (NSE), and the absolute value of percentage bias (|PBIAS|) during the calibration and validation period all met the accuracy requirements of the model, which validated the applicability of CMADS-L40 data and the SWAT model for runoff simulation at the watershed scale. Changes in territorial spatial patterns are closely correlated with runoff variation. Changes in agricultural production space and forest ecological space show statistically significant negative correlation with runoff change, while industrial production space change exhibits a significant positive correlation with runoff change. The expansion of production space, particularly industrial production space, leads to increased runoff, whereas the enlargement of agricultural production space and forest ecological space can reduce runoff. This article contributes to highlighting the role of land use policy in hydrological regulation, providing a scientific basis for optimizing territorial spatial planning to mitigate flood risks and protect water resources. Full article

This paper discusses the possibilities of using stump wood as a raw material for energy generation. The research was based on an analysis of the state of knowledge, forest field studies, and participatory observations. A formula was sought to optimise the procurement cost of stump wood appropriate to Polish conditions. Conceptualisation was carried out in a selected area of the Notecka Forest in the Wielkopolska region, located in western Poland. A pilot study was designed to test a computational formula to assess the profitability of harvesting wood from stump wood resources for energy generation. The potential of stump wood is estimated to be around half a million cubic metres per year from the Notecka Forest area alone. This resource provides an opportunity for business development in both forestry and the renewable energy sources (RESs) sector, despite the barriers and risks shown in this study. Full article

Blasting is an effective technique for fragmenting rock in open-pit mining operations. Blasting operations produce either boulders or fine fragments, both of which increase costs and pose environmental risks. As a result, predicting the mean fragmentation size (MFS) distribution of rock is critical for assessing blasting operations’ quality and mitigating risks. Due to the limitations of empirical and statistical models, several researchers are turning to artificial intelligence (AI)-based techniques to predict the MFS distribution of rock. Thus, this study uses three AI tree-based algorithms—extra trees (ET), gradient boosting (GB), and random forest (RF)—to predict the MFS distribution of rock. The prediction accuracy of the models is optimized utilizing the tree-structured Parzen estimators (TPEs) algorithm, which results in three models: TPE-ET, TPE-GB, and TPE-RF. The dataset used in this study was collected from the published literature and through the data augmentation of a large-scale dataset of 3740 blast samples. Among the evaluated models, the TPE-ET model exhibits the best performance with a coefficient of determination (R²), root mean squared error (RMSE), mean absolute error (MAE), and max error of 0.93, 0.04, 0.03, and 0.25 during the testing phase. Moreover, the block size (XB, m) and modulus of elasticity (E, GPa) parameters are identified as the most influential parameters for predicting the MFS distribution of rock. Lastly, an interactive web application has been developed to assist engineers with the timely prediction of MFS. The predictive model developed in this study is a reliable intelligent model because it combines high accuracy with a strong, explainable AI tool for predicting MFS. Full article

As the demand for more accurate crime prediction and risk assessment grows, researchers have been developing smarter models that blend statistical methods with machine learning. This study compares a hybrid ARIMA-ANN model with traditional classification techniques to see which best forecast monthly crime risk levels in Galați County, Romania. The analysis is based on a newly compiled dataset of 132 monthly observations from January 2014 to December 2024, which combines a broad array of social, economic, and environmental data points. The main variable, ‘Crime risk’, is based on normalized counts of offenses per capita and divided into five balanced levels: very low, low, moderate, high, and very high. The hybrid ARIMA-ANN model merges the strengths of statistical time series analysis with the flexible learning ability of artificial neural networks. Performance is evaluated against multinomial logistic regression, decision trees, random forests, and support vector machines. Overall, the results show that an ARIMA-ANN model consistently outperforms traditional methods, especially in recognizing patterns over time, seasonal trends, and complex nonlinear relationships in crime data. This study not only sets a new benchmark for crime analytics in Romania but also offers a flexible, scalable framework for classifying crime risk levels across different regions. Full article

Background: Brazilian Jiu-Jitsu (BJJ) is a physically demanding sport associated with a notable risk of musculoskeletal injuries. Understanding injury patterns, rehabilitation approaches, and psychological readiness to return to sport (RTS) is essential for prevention and management strategies. This study aimed to investigate injury characteristics among Italian BJJ athletes, assess their rehabilitation processes and psychological recovery, and identify key risk factors such as belt level, body mass index (BMI), and training load. Methods: A cross-sectional survey was conducted among members of the Italian BJJ community, including amateur and competitive athletes. A total of 360 participants completed a 36-item online questionnaire. Data collected included injury history, rehabilitation strategies, RTS timelines, and responses to the Injury-Psychological Readiness to Return to Sport (I-PRRS) scale. A Random Forest machine learning algorithm was used to identify and rank potential injury risk factors. Results: Of the 360 respondents, 331 (92%) reported at least one injury, predominantly occurring during training sessions. The knee was the most frequently injured joint, and the action “attempting to pass guard” was the most reported mechanism. Most athletes (65%) returned to training within one month. BMI and age emerged as the most significant predictors of injury risk. Psychological readiness scores indicated moderate confidence, with the lowest levels associated with playing without pain. Conclusions: Injuries in BJJ are common, particularly affecting the knee. Psychological readiness, especially confidence in training without pain, plays a critical role in RTS outcomes. Machine learning models may aid in identifying individual risk factors and guiding injury prevention strategies. Full article

Historical buildings are exposed to an increased risk of fire. The direct influence comes from the buildings’ structural design and the fire protection level. The fundamental principle for reducing the loss of heritage value in historical buildings due to fire is fire protection, as part of crisis management. This article focuses on selected castle buildings from Slovakia. Three castle buildings were selected based on their location in the country. All of them are currently used for museum purposes. Using an analytical form, we assessed fire hazards and fire safety measures in two parts, calculated the fire risk index, and proposed solutions. Qualitative research, which is more suitable for the issue at hand, was used to evaluate the selected objects. The main methods used in the research focused on visual assessment of the current condition of the objects and analysis of fire documentation and its comparison with currently valid legal regulations. Based on the results, we can conclude that Kežmarok Castle (part of the historical city center) has a small fire risk (fire risk index = 13 points). Trenčín Castle (situated on a rock above the city) and Stará Ľubovňa Castle (situated on a limestone hill outside the city, surrounded by forest) have an increased risk of fire (fire risk index = 50–63). Significant risk sources identified included surrounding forest areas, technical failures related to outdated electrical installations, open flames during cultural events, the concentration of highly flammable materials, and complex evacuation routes for both people and museum collections. Full article

Background/Objectives: Patients with stage IV gastric cancer who develop chronic intestinal failure require home parenteral nutrition (HPN). This study aimed to evaluate the prognostic relevance of nutritional and immune–inflammatory biomarkers and to construct an individualised survival prediction model using machine learning techniques. Methods: A secondary analysis was performed on a cohort of 410 patients with TNM stage IV gastric adenocarcinoma who initiated HPN between 2015 and 2023. Nutritional and inflammatory indices, including the Controlling Nutritional Status (CONUT) score and lymphocyte-to-monocyte ratio (LMR), were assessed. Independent prognostic factors were identified using Cox proportional hazards models. A Random Survival Forest (RSF) model was constructed to estimate survival probabilities and quantify variable importance. Results: Both the CONUT score and LMR were independently associated with overall survival. In multivariate analysis, higher CONUT scores were linked to increased mortality risk (HR = 1.656, 95% CI: 1.306–2.101, p < 0.001), whereas higher LMR values were protective (HR = 0.632, 95% CI: 0.514–0.777, p < 0.001). The RSF model demonstrated strong predictive accuracy (C-index: 0.985–0.986) and effectively stratified patients by survival risk. The CONUT score exerted the greatest prognostic influence, with the LMR providing additional discriminatory value. A gradual decline in survival probability was observed with an increasing CONUT score and a decreasing LMR. Conclusions: The application of machine learning to immune–nutritional data offers a robust tool for predicting survival in patients with advanced gastric cancer requiring HPN. This approach may enhance risk stratification, support individualised clinical decision-making regarding nutritional interventions, and inform treatment intensity adjustment. Full article

Objective: Shoulder dystocia (ShD) is a rare but serious obstetric emergency associated with significant neonatal morbidity. This study aimed to evaluate the predictive performance of machine learning (ML) models based on fetal biometric ratios and clinical characteristics for the identification of ShD in pregnancies without clinical suspicion of macrosomia. Methods: We conducted a retrospective case-control study including 284 women (84 ShD cases and 200 controls) who underwent spontaneous vaginal delivery between 37 and 42 weeks of gestation. All participants had an estimated fetal weight (EFW) below the 90th percentile according to Hadlock reference curves. Univariate and multivariate logistic regression analyses were performed on maternal and neonatal parameters, and statistically significant variables (p < 0.05) were used to construct adjusted odds ratio (aOR) models. Supervised ML models—Logistic Regression (LR), Random Forest (RF), and Extreme Gradient Boosting (XGB)—were trained and tested to assess predictive accuracy. Performance metrics included AUC-ROC, sensitivity, specificity, accuracy, and F1-score. Results: The BPD/AC ratio and AC/FL ratio markedly enhanced the prediction of ShD. When added to other features in RF models, the BPD/AC ratio got an AUC of 0.884 (95% CI: 0.802–0.957), a sensitivity of 68%, and a specificity of 83%. On the other hand, the AC/FL ratio, along with other factors, led to an AUC of 0.896 (95% CI: 0.805–0.972), 68% sensitivity, and 90% specificity. Conclusions: In pregnancies without clinical suspicion of macrosomia, ML models integrating fetal biometric ratios with maternal and labor-related factors significantly improved the prediction of ShD. These models may support clinical decision-making in low-risk deliveries where ShD is often unexpected. Full article