Search Results (873)

Rapid urban expansion in tropical coastal cities has intensified vegetation moisture stress, compromising urban resilience and ecological stability. This study investigates the spatial drivers of the Moisture Stress Index (MSI) in Makassar City, Indonesia, by integrating biophysical indicators and land-use characteristics through multi-scale regression analyses. Utilizing dry-season satellite composites (May–August 2025), the research derived Normalized Difference Vegetation Index (NDVI), Land Surface Temperature (LST), and Normalized Difference Built-up Index (NDBI). MSI was modeled using Ordinary Least Squares (OLS) and Geographically Weighted Regression (GWR) across 240 m, 480 m, and 960 m grids. Results indicate that MSI is highly sensitive to urban morphology and land-use configuration. High moisture stress was concentrated in commercial–industrial and dense residential zones characterized by extreme population densities exceeding 28,000 people/km² and elevated NDBI. In contrast, agricultural zones and open/green spaces provided significant cooling and moisture retention. Comparative performance analysis reveals that the local GWR model significantly outperformed the global OLS model, achieving a substantial reduction in AICc (−10,475.81) and resolving significant spatial autocorrelation to achieve random residuals (z-score = 1.55). The study further confirms that NDBI is the most robust biophysical predictor of MSI. Spatial heterogeneity analysis demonstrated that land-use influences are geographically contingent, with institutional areas showing varied effects based on campus design and canopy presence. These findings emphasize the necessity of scale-aware, climate-adaptive urban planning and demonstrate that GWR provides a high-fidelity tool for identifying neighborhood-level “micro-hotspots” overlooked by global modeling frameworks. Full article

Analysis of game management and trophy game populations in Osijek-Baranja County shows that this region is one of the most valuable hunting areas in Croatia, with rich populations of red deer, roe deer, and wild boar, as well as stable annual population growth. The methodological framework included products and services in hunting based on data analysis from the Croatian Hunting Association, big-game trophy records (ETD forms), the Central Hunting Register, and the official price list of game culling and hunting services. Data on harvests and trophy values indicate long-term population stability and high economic potential of hunting, with red deer generating the highest total revenue (EUR 7.29 million), while roe deer and wild boar contribute to overall stability and harvest volume. The total trophy value over 12 hunting seasons reaches EUR 11.99 million, underscoring the economic importance of hunting tourism for local communities. Differences among hunting ground users suggest that private companies and the state company Croatian Forests Ltd. often achieve higher trophy values, while county hunting associations report more modest results. However, regression analysis shows there is not a strong statistical correlation between management structure and trophy outcomes, highlighting the significant influence of ecological and spatial factors on game quality. International hunters, primarily from Germany and Austria, represent a key segment of demand, confirming the market potential for further development of hunting tourism. Despite the rich natural base, results indicate the need for better marketing approach, digital visibility, and integration of hunting products with other forms of tourism, such as gastronomic, wine, and nature tourism. Effective positioning of Croatia as a competitive hunting destination requires adaptation to contemporary market trends and adherence to international sustainable management guidelines (FAO, ELC, CBD). In conclusion, hunting in Croatia represents an important non-wood forest product and a vital resource for rural and economic development. Sustainable population management, quality promotion, and integration of traditional and innovative practices are essential for reinforcing biodiversity conservation, supporting community livelihoods, and strengthening Croatia’s role in the European and global hunting-tourism market. Full article

Spatial disparities in housing values during the Great Depression reflect not only regional housing market conditions but also deeper inequalities in economic opportunity, social infrastructure, and environmental resilience that are central to place-based sustainability. Despite extensive research on housing inequality during this period, spatial disparities in housing values—particularly in relation to race beyond the neighborhood level—remain underexplored. This study examines county-level spatial disparities in housing values in the United States between 1930 and 1940, framing housing values as an indicator of place-based sustainability. Using spatial visualization, global and local spatial econometric models, and Multi-Scale Geographically Weighted Regression (MGWR), we analyze how economic shocks, environmental stressors, and socioeconomic and demographic factors jointly shaped uneven housing outcomes across space. Our findings reveal distinct regional trends: higher housing values were concentrated in the Northeast, Midwest, and West Coast, while lower values prevailed in the Mountain and Southern regions. Housing values declined from 1930 to 1940, with the Dust Bowl intensifying losses in affected areas. Socioeconomic factors, such as higher illiteracy and unemployment rates, were associated with lower housing values, whereas higher retail sales per capita, a proxy for income, were linked to higher values. Housing values also varied significantly by racial and nativity composition, with persistent disparities disadvantaging Black and other minority populations relative to native White populations within the same regions. By quantifying spatial inequality and identifying uneven regional vulnerability and resilience during a major historical crisis, this study contributes a place-based sustainability perspective on long-term housing inequality and its structural roots. Full article

Google Earth Engine (GEE) has revolutionised remote sensing. The GEE cloud platform lets users quickly analyse large satellite imagery datasets with custom programmes, enhancing global-scale analysis. Crop condition monitoring using GEE would greatly help in decision-making and precision agriculture. Estimating canopy chlorophyll content (CCC) is an effective way to monitor crops using remote sensing because leaf chlorophyll is a key indicator. A hybrid model that combines radiative transfer models (RTMs), such as PROSAIL, with Gaussian Process Regression (GPR) can effectively estimate crop biophysical parameters using remote sensing images. GPR has proven to be one of the best methods for this purpose. This study aimed to develop a hybrid model to estimate CCC from S2 imagery and transfer it to the GEE platform for efficient data processing. In this work, the CCC (g/cm²) data from the S2 biophysical processor toolbox for the S2 imagery of the ICAR-Indian Agricultural Research Institute (IARI) on 23 February 2023 were used as observation data to train the hybrid algorithm. The hybrid model was successfully validated against the 155 input data with an R2 of 0.94, RMSE of 10.02, and NRMSE of 5.04%. The model was integrated into GEE to successfully generate a CCC-estimated map of IARI using S2 imagery from 23 February 2023. An R2 value of 0.96 was observed when GEE-estimated CCC values were compared against CCC values estimated locally. This establishes that the GEE-based CCC estimation with the PROSAIL + GPR hybrid model is an effective and accurate method for monitoring vegetation and crop conditions over large areas and extended periods. Full article

The on-machine measurement (OMM) of aero-engine blades is a critical technology for enabling closed-loop manufacturing. However, when using line laser sensors with a fixed scanning pose to measure free-form surfaces, the variation in surface geometry leads to changing incident angles, which in turn induce non-stationary noise. To address this issue, this paper proposes a multi-sensor fusion method utilizing Adaptive Heteroscedastic Gaussian Process Regression (AHGPR) based on a Spatial-Angle-Balanced Sampling (S-ABS) strategy. The AHGPR explicitly integrates the physical mapping of incident angle errors into its covariance structure, thereby automatically adjusting observation weights according to the local geometric posture. Concurrently, the S-ABS strategy captures the high-error characteristic points with large incident angles while maintaining a globally uniform spatial distribution. The experimental data indicate that this approach addresses the sampling deficiency encountered at the leading and trailing edges and in areas with large incident angles. The proposed approach reduced the impact of optical deviations on measurement accuracy and improved the precision of the process. Full article

Safe and sample-conscious controller synthesis for nonlinear dynamics benefits from reinforcement learning that exploits an explicit plant model. A nonlinear mass–spring–damper with hardening effects and hard stops is studied, and off-plant Q-learning is enabled using two data-driven surrogates: (i) a piecewise linear model assembled from operating region transfer function estimates and blended by triangular memberships and (ii) a global nonlinear autoregressive model with exogenous input constructed from past inputs and outputs. In unit step reference tracking on the true plant, the piecewise linear route yields lower error and reduced steady-state bias (MAE = $0.03$; SSE = $3\%$) compared with the NLARX route (MAE = $0.31$; SSE = $30\%$) in the reported configuration. The improved regulation is obtained at a higher identification cost (60,000 samples versus 12,000 samples), reflecting a fidelity–knowledge–data trade-off between localized linearization and global nonlinear regression. All reported performance metrics correspond to deterministic validation runs using fixed surrogate models and trained policies and are intended to support methodological comparison rather than statistical performance characterization. These results indicate that model-based Q-learning with identified surrogates enables off-plant policy training while containing experimental risk and that performance depends on modeling choices, state discretization, and reward shaping. Full article

Accurate anatomical landmark localization is essential to automate chest X-ray analysis and improve diagnostic reliability. While global context recognition is essential in medical imaging, the inherently high-resolution nature of these images has long made this task particularly difficult. While the U-Net-based heatmap regression methods show strong performance, they still lack explicit modeling of the global spatial relationships among landmarks. To address this limitation, we propose an integrated structural learning framework that captures anatomical correlations across landmarks. The model generates probabilistic heatmaps with U-Net and derives continuous coordinates via soft-argmax. Subsequently, these coordinates, along with their corresponding local feature vectors, are fed into a Graph Neural Network (GNN) to refine the final positions by learning inter-landmark dependencies. Anatomical priors, such as bilateral symmetry and vertical hierarchy, are incorporated into the loss function to enhance spatial consistency. The experimental results show that our method consistently outperforms state-of-the-art models across all metrics, achieving significant improvements in MRE and SDR at 3, 6, and 9 pxl thresholds. This high precision demonstrates the framework’s strong potential to enhance the accuracy and robustness of clinical diagnostic systems. Full article

Thermally modified wood is primarily used in exterior applications due to its enhanced resistance to biotic degradation. However, reduced mechanical performance limits its structural use. This study investigates the structural potential of high-temperature-treated European beech timber (Fagus sylvatica, L.) and evaluates its mechanical properties and grading models for structural design. Timber from 32 beech logs was air-dried and divided into untreated (NoTMW) and thermally modified (TMW) groups. Thermal modification was carried out commercially in an oxidizing atmosphere at 190 °C. All specimens were visually graded according to DIN 4074-5 and assessed using acoustic non-destructive methods before testing in four-point bending following EN 408. Modulus of elasticity (MOE), modulus of rupture (MOR), and density were determined, and characteristic values were calculated according to EN 384. On average, TMW exhibited a 17% reduction in bending strength compared to untreated wood, while both static and dynamic MOE were not significantly affected. The multiple regression model only slightly improved bending strength prediction compared with single linear regression based on global modulus, as the R²-value increased from 17% to 19%. The prediction of stiffness of thermally treated beech timber was greatly improved by combining local and acoustic moduli, explaining 76% of the total variation. Full article

Mediterranean biogeography is characterized as a global “hotspot” for climate change; understanding the impacts of these changes on local agricultural systems through high-resolution analyses has thus become a critical need. This study addresses this gap by evaluating the holistic effects of climate change on site-specific agriculture systems, focusing on the Eğirdir–Karacaören (EKB) and Beyşehir (BB) lake basins in the Lakes Region of Türkiye. This study employed machine learning modeling techniques to forecast changes in the yields of key crops, such as wheat, maize, apple, alfalfa, and sugar beet. Detailed spatial analyses of changes in agro-climatic conditions (heat stress, chilling requirement, frost days, and growing degree days for key crops) between the reference period (1995–2014) and two decadal periods projected for 2040–2049 and 2070–2079 were conducted under the Shared Socioeconomic Pathways (SSP3-7.0). Daily temperature, precipitation, relative humidity, and solar radiation data, derived from high-resolution climate simulations, were aggregated into annual summaries. These datasets were then spatially matched with district-level yield statistics obtained from the official data providers to construct crop-specific data matrices. For each crop, Random Forest (RF) regression models were fitted, and a Leave-One-Site-Out (LOSOCV) cross-validation method was used to evaluate model performance during the reference period. Yield prediction models were evaluated using the mean absolute error (MAE). The models achieved low MAE values for wheat (33.95 kg da⁻¹ in EKB and 75.04 kg da⁻¹ in BB), whereas the MAE values for maize and alfalfa were considerably higher, ranging from 658 to 986 kg da⁻¹. Projections for future periods indicate declines in relative yield across both basins. For 2070–2079, wheat and maize yields are projected to decrease by 10–20%, accompanied by wide uncertainty intervals. Both basins are expected to experience a substantial increase in heat stress days (>35 °C), a reduction in frost days, and an overall acceleration of plant phenology. Results provided insights to inform region-specific, evidence-based adaptation options, such as selecting heat-tolerant varieties, optimizing planting calendars, and integrating precision agriculture practices to improve resource efficiency under changing climatic conditions. Overall, this study establishes a scientific basis for enhancing the resilience of agricultural systems to climate change in two lake basins within the Mediterranean biogeography. Full article

Short-term load forecasting (STLF) underpins the efficient and secure operation of power systems. This study develops and evaluates a hybrid architecture that couples deep neural networks (DNNs) with support vector regression (SVR) for national-scale day-ahead STLF using Croatian load data from 2006 to 2022. The approach employs an hour-specific framework of 24 hybrid models: each DNN learns a compact nonlinear representation for a given hour, while an SVR trained on the penultimate layer activations performs the final regression. Gradient-boosting-based feature selection yields compact, informative inputs shared across all model variants. To overcome limitations of historical local measurements, the framework integrates global numerical weather prediction data from the TIGGE archive with load and local meteorological observations in an operationally realistic setup. In the held-out test year 2022, the proposed hybrid consistently reduced forecasting error relative to standalone DNN-, LSTM- and Transformer-based baselines, while preserving a reproducible pipeline. Beyond using SVR as an alternative output layer, the contributions are as follows: addressing a 17-year STLF task, proposing an hour-specific hybrid DNN–SVR framework, providing a systematic comparison with deep learning baselines under a unified protocol, and integrating global weather forecasts into a practical day-ahead STLF solution for a real power system. Full article

This study investigates the dynamic behaviors of re-entrant unit-cell-shaped steel frames through numerical and experimental modal analyses. Inspired by re-entrant honeycomb structures, individual frame units were modeled to explore how natural frequencies vary with beam cross-sectional dimensions and frame angles. Twenty distinct frame models—incorporating four cross-sectional sizes (4 × 4 mm, 8 × 8 mm, 12 × 12 mm, and 16 × 16 mm) and five main frame angles (120°, 150°, 180°, 210°, and 240°)—were developed using 3D modeling and finite element analysis (FEA) tools, and the first eight natural frequencies and corresponding mode shapes were extracted for each model. The results reveal that lower modes exhibit global bending and torsional behaviors, whereas higher modes demonstrate increasingly localized deformations. It is found that the natural frequencies decrease in the straight frame configuration and increase in the hexagonal configurations, highlighting the critical influence of the frame geometry. Increasing the cross-sectional size consistently enhances the dynamic stiffness, particularly in hexagonal frames. A quadratic polynomial surface regression analysis was performed to model the relationship of the natural frequency with the cross-sectional dimension and frame angle, achieving high predictive accuracy (R² > 0.98). The experimental validation results were in good agreement with the numerical results, with discrepancies generally remaining below 7%. The developed regression model provides an efficient design tool for predicting vibrational behaviors and optimizing frame configurations without extensive simulations; furthermore, experimental modal analyses validated the numerical results, confirming the effectiveness of the model. Overall, this study provides a comprehensive understanding of the dynamic characteristics of re-entrant frame structures and proposes practical design strategies for improving vibrational performance, which is particularly relevant in applications such as machine foundations, vibration isolation systems, and aerospace structures. Full article

Herbage intake by grazers and browsers is of fundamental importance to agricultural ecosystems worldwide but is also notoriously difficult to quantify. The intake process is mediated by herbage comminution in the mouth. The attendant chew actions generate sound bursts that can be detected acoustically and analyzed to help elucidate the entire process. Goats consuming a single plant species were acoustically monitored in order to (i) determine the sensitivity of the chewing effort to the large variation in bite mass and satiety level and (ii) estimate how well the amount of herbage consumed can be predicted by counting chews. Experiments used hand-constructed patches containing bite-sized carob (Ceratonia siliqua L.) leaflets of a pre-determined mass that were presented to six goats, individually, with acoustic sensors attached to their horns. Experiment 1 determined the chewing effort and the sequence of bites and chews for three bite masses across five levels of total intake. Experiment 2 determined the chewing effort and the chew sequence at three levels of satiety, achieved by control of the feeding regime, using a single bite mass across three levels of total intake. In Experiment 1, the global chewing coefficient was ≈4 chews g⁻¹ fresh mass ingested (≈10 chews g⁻¹ dry matter). For an individual animal, the chewing coefficient was fairly stable, being influenced mildly by bite mass, but the variation between animals was large. In Experiment 2, the chewing coefficient was again fairly stable in an individual animal, although the chewing effort was slightly elevated at low satiety. At the population level, and for the most relevant range of intake levels, inverse regression of the pooled data from both experiments estimated the two-sided 95% confidence interval of the predicted intake of carob leaves to be <10% of the predicted value. If chewing coefficients can be estimated locally, usefully precise intake predictions should be attainable for the tested vegetation. These results are promising for the future potential of acoustic monitoring, although significant challenges remain. Full article

This study presents an approach for predicting the hysteresis behavior of shape memory alloys (SMAs) based on automated machine learning (AutoML) integrated with explainable artificial intelligence (XAI). Experimental data from cyclic tests of NiTi wire under loading frequencies of 0.3, 0.5, 1, and 5 Hz were used for model development. The AutoML framework PyCaret enabled automated model selection, hyperparameter optimization, and performance comparison of regression algorithms. The highest prediction accuracy was achieved by the LightGBM model (for 0.3 Hz and 1 Hz) and the CatBoost model (for 0.5 Hz and 5 Hz), both demonstrating a coefficient of determination R² > 0.997 and low MSE, MAE, and MAPE values. Integration of XAI through the SHAP method provided both global and local interpretability of the model’s behavior. The analysis revealed the dominant influence of the Stress parameter, the physically meaningful role of the loading or unloading stage (UpDown), and a gradual increase in the contribution of the Cycle parameter in later cycles, reflecting fatigue accumulation effects. The obtained results confirm the high accuracy, interpretability, and physical consistency of the developed models. Full article

Instance selection is a critical preprocessing technique for enhancing data quality and improving machine learning model efficiency. However, existing algorithms for regression tasks face a fundamental trade-off: non-heuristic methods offer high precision but suffer from sequential dependencies that hinder parallelization, while heuristic methods support parallelization but often yield coarse-grained results susceptible to local optima. To address these challenges, we propose CRDISA, a novel distributed instance selection algorithm driven by a formalized cognitive reasoning logic. Unlike traditional approaches that evaluate subsets, CRDISA transforms each instance into an independent “Instance Expert” capable of reasoning about the global data distribution through a unique difference knowledge base. For regression tasks with continuous outputs, we introduce a soft partitioning strategy to define adaptive error boundaries and a bidirectional voting mechanism to robustly identify high-quality instances. Although the fine-grained reasoning implies high computational complexity, we implement CRDISA on Apache Spark using an optimized broadcast mechanism. This architecture provides linear scalability in wall-clock time, enabling scalable processing without sacrificing theoretical rigor. Experiments on 22 datasets demonstrate that CRDISA achieves an average compression rate of 31.7% while maintaining predictive accuracy ($R^2=0.681$) comparable to or better than state-of-the-art methods, proving its superiority in balancing selection granularity and distributed efficiency. Full article

Reliable probabilistic wind speed forecasts are essential for integrating renewable energy into power grids and managing operational uncertainty. This study compares Quantile Regression Forests (QRF), Bayesian Additive Regression Trees (BART), and Gaussian Process Regression (GPR) under at-site and regional pooled frameworks using 21 years (2000–2020) of daily wind data from eleven stations in New South Wales and Queensland, Australia. Models are evaluated via strict year-based holdout validation across seven metrics: RMSE, MAE, R², bias, correlation, coverage, and Continuous Ranked Probability Score (CRPS). Regional QRF achieves exceptional point forecast stability with minimal RMSE increase but suffers persistent under-coverage, rendering probabilistic bounds unreliable. BART attains near-nominal coverage at individual sites but experiences catastrophic calibration collapse under regional pooling, driven by fixed noise priors inadequate for spatially heterogeneous data. In contrast, GPR maintains robust probabilistic skill regionally despite larger point forecast RMSE penalties, achieving the lowest overall CRPS and near-nominal coverage through kernel-based variance inflation. Variable importance analysis identifies surface pressure and minimum temperature as dominant predictors (60–80%), with spatial covariates critical for regional differentiation. Operationally, regional QRF is prioritised for point accuracy, regional GPR for calibrated probabilistic forecasts in risk-sensitive applications, and at-site BART when local data suffice. These findings show that Bayesian machine learning methods can effectively navigate the trade-off between local specificity and regional pooling, a challenge common to wind forecasting in diverse terrain globally. The methodology and insights are transferable to other heterogeneous regions, providing guidance for probabilistic wind forecasting and renewable energy grid integration. Full article