Search Results (4,097)

Similarity search on image embeddings is a common practice for image retrieval in machine learning and pattern recognition systems. Approximate nearest neighbor (ANN) methods enable scalable similarity search on large datasets, often approaching sub-linear complexity. Yet, little empirical work has examined how ANN neighborhood geometry differs from that of exact k-nearest neighbors (k-NN) search as the neighborhood size increases under constrained search effort. This study quantifies how approximate neighborhood structure changes relative to exact k-NN search as k increases across three experimental conditions. Using multiple random subsets of 10,000 images drawn from the STL-10 dataset, we compute ResNet-50 image embeddings, perform an exact k-NN search, and compare it to a Hierarchical Navigable Small World (HNSW)-based ANN search under controlled hyperparameter regimes. We evaluated the fidelity of neighborhood structure using neighborhood overlap, average neighbor distance, normalized barycenter shift, and local intrinsic dimensionality (LID). Results show that exact k-NN and ANN search behave nearly identically when $efSearch>k$. However, as the neighborhood size grows and $efSearch$ remains fixed, ANN search fails abruptly, exhibiting extreme divergence in neighbor distances at approximately $k\approx 2$–$3.5\times efSearch$. Increasing index construction quality delays this failure, and scaling search effort proportionally with neighborhood size ($efSearch=\alpha \times k$ with $\alpha \ge 1$) preserves neighborhood geometry across all evaluated metrics, including LID. The findings indicate that ANN search preserves neighborhood geometry within its operational capacity but abruptly fails when this capacity is exceeded. Documenting this behavior is relevant for scientific applications that approximate embedding spaces and provides practical guidance on when ANN search is interchangeable with exact k-NN and when geometric differences become nontrivial. Full article

Green space equity is increasingly recognized as a critical environmental condition for healthy aging, yet existing research often overlooks how different green space attributes—accessibility and diversity—are associated with distinct dimensions of older adults’ health. Limited attention has been paid to their nonlinear threshold effects or to the social pathways through which green spaces influence health outcomes. Using the United States county-level panel data from 2020 to 2023, this study integrates fixed-effects models, Extreme Gradient Boosting (XGBoost), and mediation analysis to examine the associations between green accessibility measured by the Two-Step Floating Catchment Area (2SFCA) method, and green diversity measured by the Shannon Index, on the general, physical, and mental health of older adults. Findings indicate that (1) higher green accessibility is associated with better general health, whereas green diversity shows a stronger association with physical health, reflecting its link to more heterogeneous ecosystem service environments. (2) Green accessibility demonstrates the threshold effect, in which the strength of association with health becomes steeper once accessibility approaches higher levels. (3) Green space equity is linked to health partly through social structures. Education clustering and marital stability mediate the associations with general health, while mental health appears to depend more on the social interaction opportunities embedded within green environments than on their physical attributes alone. The study proposes an integrated “physical environment–social structure–health outcome” framework and a threshold-oriented spatial intervention strategy, highlighting the need to prioritize improvements in green accessibility in underserved areas and prioritizing green diversity and age-friendly social functions where accessibility is already high. These findings offer evidence for designing inclusive, health-oriented urban environments for aging populations. Full article

This study investigates the influence of the additive illite on the thermal, tribological, and energy efficiency characteristics of calcium grease (CG) at different concentrations (0.05 wt.%, 0.1 wt.%, 0.2 wt.%, 0.4 wt.%, 0.6 wt.%, and 0.8 wt.%). Thermo-gravimetric analysis under inert and oxidative atmospheres revealed that illite enhances thermal stability by increasing inorganic residue under N₂, but promotes oxidative degradation under O₂, limiting practical thermal use to around 400 °C. Grease with 0.1 wt.% illite (CGI2) performed well in tribological tests by reducing the coefficient of friction and wear scar diameter by 53% and 57%, respectively, compared to the base grease. Fleischer’s energy-based wear model showed that all grease samples operated within the mixed friction regime, and CGI2 exhibited a 93% higher apparent frictional energy density and a substantially lower wear intensity that was 47% lower than the base grease, indicating improved energy dissipation and wear resistance. All samples had the same weld load (1568 N), but CGI2 had a 21% higher load–wear index than the base grease in the extreme-pressure test, indicating better load-carrying capacity. In the energy consumption test, a 6% reduction in current consumption was observed in CGI2 in comparison with the base grease. Overall, illite at an optimal concentration significantly enhances lubrication performance, wear protection, and energy efficiency. Full article

To clarify the spatial–temporal evolution patterns and climate-driven mechanisms of carbon sinks of forest ecosystems under climate change, we calculated the net ecosystem productivity (NEP) of forests in the Guangxi region using remote sensing and meteorological data from 2000 to 2023. By employing trend analysis, spatial clustering, the Hurst index, and climate contribution evaluation, we analyzed the spatial and temporal changes, sustainability, and the relative contribution of climate impacts on forest carbon sinks. The results are as follows: The carbon sink capacity of forests in Guangxi increased continuously from 2000 to 2023, at a rate of 3.57 g C·m⁻²·a⁻¹, reaching 39.19% higher in 2023 than in 2000. The carbon sink capacity was higher in the southwest and lower in the northeast, with hotspots mainly located in evergreen/deciduous broad-leaved forest areas. The Hurst index indicates that 84.44% of regions are likely to maintain this increasing trend, suggesting stability in forest carbon sink function. The climate contribution rate to forest carbon sinks was moderate, with significant temporal fluctuations. Temperature governed annual variation in forest carbon sinks, influencing up to 36.37% of the area. The annual average contribution rate of climate change to forest carbon sinks was 30.28%, but there were temporal fluctuations and spatial heterogeneity. Over time, climate contributions had a positive driving impact; however, extreme climate events tended to produce a negative effect. The pattern of forest carbon sinks in Guangxi showed a “heat sink-coupling” phenomenon, with 16.23% of the hotspots of forest carbon sinks coinciding with temperature control zones, highlighting the enhancing effect of temperature rise on carbon sinks against a background of water and heat synergy. This study provides a scientific basis for the assessment of forest carbon sink potential and climate suitability management in Guangxi. Full article

Droughts are recognized as one of the most devastating extreme climate events, leading to severe socioeconomic losses and ecological degradation globally under climate change. With global warming, the frequency and intensity of extreme droughts are increasing, posing critical challenges to water resource management. The Standardized Precipitation Conversion Index (SPCI) has demonstrated potential in drought monitoring; however, its applicability across diverse climatic zones and multiple temporal scales remains inadequately validated. This study addresses this gap by establishing a novel multi-scale inversion analysis using ERA5-based precipitable water vapor (PWV) and precipitation data. SPCI is selected for its advantage in eliminating climatic background biases through probability normalization, overcoming limitations of traditional indices such as the Standardized Precipitation Index (SPI) and Standardized Precipitation-Evapotranspiration Index (SPEI). We systematically evaluated the spatiotemporal evolution of Precipitation Efficiency (PE) and SPCI across four climatic zones in China. Results show that the first two principal components explain over 85% of the spatiotemporal variability of PE, with PC1 independently contributing from 82.05% to 83.80%. This high variance contribution underscores that the spatiotemporal patterns of PE are dominated by a few key climatic drivers, validating the robustness of the principal component analysis. SPCI exhibits strong correlation with SPI, exceeding 0.95 in the Tropical Monsoon Zone (TMZ) at scales of 1–6 months, indicating its utility for short-to-medium-term drought monitoring. Distinct zonal differentiation in PE patterns is revealed, such as the bimodal annual cycle in the Tropical-Subtropical Monsoon Composite Zone (TSMCZ). This study evaluates the performance of the SPCI against the widely used SPI and SPEI across four major climatic zones in China. It validates the SPCI’s applicability across China’s complex climates, providing a scientific basis for region-specific drought early warning and water resource optimization. Full article

Reliable models for predicting the uniaxial compressive strength (UCS) of rocks are crucial for mining operations and rock engineering design. Empirical methods, including statistical methods, are often faced with many limitations when generalizing in a wide range of lithological types. To address this limitation, this study investigates the capability of grey wolf optimization (GWO)-optimized ensemble machine learning models, including decision tree (DT), extreme gradient boosting (XGBoost), and adaptive boosting (AdaBoost) for predicting UCS using a small dataset of easily measurable and non-destructive rock index properties. The study’s objective is to evaluate whether metaheuristic-based hyperparameter optimization can enhance model robustness and generalization performance under small-sample conditions. A unified experimental framework incorporating GWO-based optimization, three-fold cross-validation, sensitivity analysis, and multiple statistical performance indicators was implemented. The findings of this study confirm that although the GWO-XGBoost model achieves the highest training accuracy, it exhibits signs of mild overfitting. In contrast, the GWO-AdaBoost model outpaced with significant improvement in terms of coefficient of determination (R²) = 0.993, root mean square error (RMSE) = 2.2830, mean absolute error (MAE) = 1.6853, and mean absolute percentage error (MAPE) = 4.6974. Therefore, the GWO-AdaBoost has proven to be the most effective in terms of its prediction potential of UCS, with significant potential for adaptation due to its effectively learned parameters. From a theoretical perspective, this study highlights the non-equivalence between training accuracy and predictive reliability in UCS modeling. Practically, the findings support the use of GWO-AdaBoost as a reliable decision-support tool for preliminary rock strength assessment in mining and geotechnical engineering, particularly when comprehensive laboratory testing is not feasible. Full article

This study presents a comprehensive geospatial framework for assessing coastal vulnerability and ecosystem service distribution along the Greek coastline, one of the longest and most diverse in Europe. The framework integrates two complementary components: a Coastal Erosion Vulnerability Index applied to all identified beach units, and Coastal Flood Risk Indexes focused on low-lying and urbanized coastal segments. Both indices draw on harmonized, open-access European datasets to represent environmental, geomorphological, and socio-economic dimensions of risk. The Coastal Erosion Vulnerability Index is developed through a multi-criteria approach that combines indicators of physical erodibility, such as historical shoreline retreat, projected erosion under climate change, offshore wave power, and the cover of seagrass meadows, with socio-economic exposure metrics, including land use composition, population density, and beach-based recreational values. Inclusive accessibility for wheelchair users is also integrated to highlight equity-relevant aspects of coastal services. The Coastal Flood Risk Indexes identify flood-prone areas by simulating inundation through a novel point-based, computationally efficient geospatial method, which propagates water inland from coastal entry points using Extreme Sea Level (ESL) projections for future scenarios, overcoming the limitations of static ‘bathtub’ approaches. Together, the indices offer a spatially explicit, scalable framework to inform coastal zone management, climate adaptation planning, and the prioritization of nature-based solutions. By integrating vulnerability mapping with ecosystem service valuation, the framework supports evidence-based decision-making while aligning with key European policy goals for resilience and sustainable coastal development. Full article

Although extensive evidence notes a nonlinear relationship between urban greenspace and wellbeing, the conditional role of spatial patterns in this relationship has rarely been examined. To address this gap, this study investigates whether and how landscape metrics moderate the nonlinear association between greenspace coverage and life satisfaction (LS) in urban China. Using nationally representative data from the 2015 wave of the Chinese Social Survey (N = 4319 across 321 subdistricts), this study combines individual-level LS scores with high-resolution GlobeLand30 land use data. Moderated quadratic regression models and formal endpoint slope and turning point tests are applied to identify both the shape and dynamics of the greenspace–wellbeing relationship. The analysis reveals a robust U-shaped curve: LS is lowest at moderate greenspace levels and higher at both low and high extremes. Critically, spatial pattern features, including aggregation index, Euclidean nearest neighbor distance, patch density, and patch richness, significantly moderate this relationship. The turning point of the U-shape moves rightward with greater aggregation and leftward with higher fragmentation or richness. While visual presentation indicates that the curve flips at low patch isolation, further statistical analyses indicate insufficient curve steepness. These findings support that the “more is better” argument should be extended to consider both greenspace quantity and spatial configuration in urban planning for optimal wellbeing outcomes. Full article

In regions with limited meteorological monitoring systems, spatial drought modeling is of importance for efficient water resource management. This study recommends an alternative drought modeling strategy for Standardized Precipitation Evapotranspiration Index (SPEI) prediction at multiple target stations using data from neighboring stations. The Multi-Output Random Forest (MORF) model is implemented in this study to consider the spatial correlations among stations for the simultaneous prediction of SPEI for multiple stations instead of training independent models for each station. The efficiency of MORF is further compared to Multi-Output Support Vector Regression (MOSVR) and three baselines; a single-output RF, a monthly climatology model, and a persistence model. In addition to statistical performance criteria, drought characteristics are evaluated using intensity–duration–frequency analysis for three temporal scales (SPEI-3, SPEI-6, and SPEI-12). Results demonstrate that MORF outperformed MOSVR and RF in approximating observed drought intensity, duration, and frequency under moderate, severe, and extreme drought scenarios. Furthermore, spatial analysis reveals that MORF accurately captured the seasonal evolution of drought conditions including onset and recovery phases. The remarkable success of MORF in contrast to MOSVR and three traditional baselines can be explained by its ability to detect nonlinear and complex interactions of drought condition among various neighboring stations. This study emphasizes the promise of multi-output machine learning algorithms for drought monitoring in water resource management and climate adaptation planning in data-scarce regions. Full article

Compound dry–hot events increasingly threaten ecosystem productivity under global warming. Using ERA5-Land and MODIS NPP (2002–2024) for the Yangtze River Basin, we built climate indices and developed a Copula-based standardized compound dry–hot index (SCDHI) to detect events and examine spatiotemporal patterns. Trend and correlation analyses quantified NPP sensitivity and lag, and an NPP–SCDHI coupling framework assessed resistance and resilience across major vegetation types. Basin-wide monthly NPP increased slightly, while SCDHI decreased, indicating a warmer and drier tendency. Under dry–hot conditions, NPP was mainly negatively related to event intensity in the upper basin but positively related across much of the middle–lower plains. The mean NPP response time was approximately 2 months, with forests and croplands typically lagging 2–3 months. Under extreme stress, forests showed high resistance but limited recovery, whereas shrublands showed moderate resistance and low resilience. Cultivated vegetation exhibited the lowest resistance and weak resilience, grasslands had low resistance but relatively rapid recovery, and alpine vegetation showed moderate resistance and the highest resilience. Cultivated vegetation and grasslands may therefore represent high-risk types for ecological management. Full article

Background/Objectives: Insulin resistance and ambulatory blood pressure monitoring (ABPM) abnormalities represent distinct but interrelated pathways contributing to cardiovascular risk. The triglyceride–glucose (TyG) index reflects metabolic burden, whereas arterial load—captured through arterial stiffness, blood pressure variability, and morning surge—reflects hemodynamic instability. Whether the coexistence of these domains identifies a particularly high-risk ambulatory phenotype remains unclear. To evaluate the independent and combined effects of metabolic burden (TyG) and arterial load on circadian blood pressure pattern and short-term systolic blood pressure variability. Methods: This retrospective cross-sectional study included 294 adults who underwent 24 h ABPM. Arterial load was defined using three ABPM-derived indices (high AASI, high SBP-ARV, high morning surge). High metabolic burden was defined as TyG in the upper quartile. The “double-high” phenotype was classified as high TyG plus high arterial load. Primary and secondary outcomes were non-dipping pattern and high SBP variability. Multivariable logistic regression and Firth penalized models were used to assess independent associations. Predictive performance was evaluated using ROC analysis. Results: The double-high phenotype (n = 15) demonstrated significantly higher nighttime SBP, reduced nocturnal dipping, and markedly elevated BP variability. It was the strongest independent predictor of non-dipping (adjusted OR = 42.0; Firth OR = 11.73; both p < 0.001) and high SBP variability (adjusted OR = 41.7; Firth OR = 26.29; both p < 0.001). Arterial load substantially improved model discrimination (AUC = 0.819 for non-dipping; 0.979 for SBP variability), whereas adding TyG to arterial load produced minimal incremental benefit. Conclusions: The coexistence of elevated TyG and increased arterial load defines a distinct hemodynamic endotype characterized by severe circadian blood pressure disruption and exaggerated short-term variability. While arterial load emerged as the principal determinant of adverse ambulatory blood pressure phenotypes, TyG alone demonstrated limited discriminative capacity. These findings suggest that TyG primarily acts as a metabolic modifier, amplifying adverse ambulatory blood pressure phenotypes predominantly in the presence of underlying arterial instability rather than serving as an independent discriminator. Integrating metabolic and hemodynamic domains may therefore improve risk stratification and help identify a small but clinically meaningful subgroup of patients with extreme ambulatory blood pressure dysregulation. Full article

Background: The original Closed Kinetic Chain Upper Extremity Stability Test (CKCUEST) is a simple assessment tool but does not account for individual differences in hand starting position and fails to provide information on limb asymmetries. The purpose of the study is to evaluate the test–retest reliability of a new single-arm CKCUEST as well as the reliability of the limb symmetry index (LSI). This version normalizes the test based on the participant’s arm length and allows for the assessment of limb symmetry since it is performed one arm at a time. Methods: Twelve healthy young adults provided both verbal and written consent to participate. Participants were excluded if they had sustained an injury in the past three months requiring medical attention and/or resulting in decreased activity for more than three days. Testing was conducted in the push-up position with participants’ thumbs placed parallel and at a distance equal to the length of their dominant arm (measured from the acromion to the tip of the middle finger), and feet positioned shoulder-width apart. Participants were instructed to keep the testing hand stable on the floor while the opposite hand reached across the body to touch the stationary hand and then return to the starting position marked with athletic tape. The goal was to complete as many touches as possible in 15 s, with each touch counted only if the participant touched the stationary hand, returned to the starting position, and maintained the shoulder-width stance. The average number of touches from the three trials was used for analysis. Intraclass Correlation Coefficients (ICC_(3,1)) were computed to determine test–retest reliability. Results: Test–retest reliability of the single-arm CKCUEST individual tests was good to excellent. The ICC_(3,1) was 0.88 (95% CI: 0.74–0.95) for all tests, 0.89 (95% CI: 0.66–0.96) for the dominant arm, and 0.93 (95% CI: 0.78–0.98) for the non-dominant arm. In contrast, the reliability of the Limb Symmetry Index (LSI) was questionable, showing substantial variability with an ICC_(3,1) of 0.53 (95% CI: −0.03–0.83) between Day 1 and Day 2, despite similar mean values (Day 1: 93.6 ± 8.46; Day 2: 94.8 ± 5.77). The Kappa coefficient suggested a substantial level of agreement for the direction of the asymmetry (preferred limb) (Kappa coefficient = 0.62). Conclusions: The new single-arm CKCUEST, which personalizes the hand starting position and measures limb symmetry, demonstrates high reliability among healthy young adults. Full article

Space-based gravitational wave detection relies on laser interferometry to measure picometer-level displacements over ${10}^5$–${10}^6$ km baselines. To integrate ranging and communication within the same optical link without degrading the primary scientific measurement, a low modulation index of 0.1 rad is required, resulting in extremely weak signals and challenging acquisition conditions. This study developed mathematical models for signal acquisition, identifying and analyzing key performance-limiting factors for both Binary Phase Shift Keying (BPSK) and Binary Offset Carrier (BOC) schemes. These factors include spreading factor, acquisition step, modulation index, and carrier-to-noise ratio (CNR). Particularly, the acquisition threshold can be directly calculated from these parameters and applied to the acquisition process of communication and ranging signals. Numerical simulations and evaluations, conducted with TianQin mission parameters, demonstrate that, for a data rate of 62.5 kbps and modulation indices of 0.081 rad (BPSK) or 0.036 rad (BOC), respectively, acquisition (probability ≈ 1) is achieved when the CNR is ≥104 dB·Hz under a false alarm rate of ${10}^{-6}$. These results provide critical theoretical support and practical guidance for optimizing the inter-satellite communication and ranging system design for the space-based gravitational wave detection missions. Full article

Repeated sowing of winter cereals represents one of the adaptive dryland approaches to make more sustainable the rainfed agriculture activities in southern Kazakhstan. This study conducted a multi-year reconstruction of crop transitions using Sentinel-2 imagery for 2018–2025, based on the combined analysis of Normalized Difference Vegetation Index (NDVI) temporal profiles and the Plowed Land Index (PLI), enabling the creation of a field-level harmonized classification set. The transition “spring crop → winter crop” was used as a formal indicator of repeated winter sowing, from which annual repeat layers and an integrated metric, the R-index, were derived. The results revealed a pronounced spatial concentration of repeated sowing in foothill landscapes, where terrain heterogeneity and locally elevated moisture availability promote the recurrent return of winter cereals. Comparison of NDVI composites for the peak spring biomass period (1–20 May) showed a systematic decline in NDVI with increasing R-index, indicating the cumulative effect of repeated soil exploitation and the sensitivity of winter crops to climatic constraints. Precipitation analysis for 2017–2024 confirmed the strong influence of autumn moisture conditions on repetition phases, particularly in years with extreme rainfall anomalies. These findings demonstrate the importance of integrating multi-year satellite observations with climatic indicators for monitoring the resilience of agricultural systems. The identified patterns highlight the necessity of implementing nature-based solutions, including contour–strip land management and the development of protective shelterbelts, to enhance soil moisture retention and improve the stability of regional agricultural landscapes. Full article

Accurate prediction of the Langelier Saturation Index (LSI), an indicator of water’s scaling and corrosive potential, is vital for water treatment and infrastructure maintenance. In this study, five machine learning models (Ridge Regression, Support Vector Machine, Random Forest, Deep Neural Network, and XGBoost) were applied to predict the LSI from physicochemical characteristics of groundwater in the Morava River basin (Serbia). Rigorous data preprocessing (outlier removal, missing data handling, z-score normalization) and feature selection were performed to ensure robust model training. Models were optimized via 10-fold cross-validation on a 70/30 train–test split. All models achieved high predictive accuracy, with ensemble methods outperforming others. XGBoost yielded the best performance (R² = 0.98; RMSE = 0.06), followed closely by Random Forest (R² = 0.95). The linear Ridge model showed the lowest (yet still strong) performance (R² = 0.90) and larger errors at extreme LSI values. Feature importance analysis consistently identified pH as the most influential predictor of the LSI, followed by alkalinity and calcium. Partial dependence plots confirmed that the models captured established nonlinear LSI behavior. The LSI rises steeply with increasing pH and moderately with mineral content. Overall, this comparative study demonstrates that modern machine learning models can predict the LSI accurately, providing interpretable insights through feature importance and dependence plots. These results underscore the potential of data-driven approaches to complement traditional water stability indices for proactive water quality management. Full article