Search Results (123)

We introduce TriHex-Cluster, a hierarchical overlapping clustering framework built on the self-similar geometry of the triangular lattice (6-regular planar graph). The primary algorithm is regime C (greedy 2-packing followed by Voronoi completion), a practical hierarchical clustering method producing disjoint clusters with the Voronoi-contact graph $G_{\mathrm{Vor}}^{(k+1)}$ as the next-level graph and aggregation complexity $O(nlogn)$ (embedding cost excluded). On regular triangular domains with near-perfect packings, regime C achieves $n^{(k+1)}\le n^{\left(k\right)}/7+O\left(\sqrt{n^{\left(k\right)}}\right)$ per level; the measured depth on finite data is $K^{\ast }=\lceil {log}_{7}n\rceil \pm 1$. Two variants complete the framework: regime A (full-overlap edge-induced, ${\mathcal{C}}^{\left(k\right)}=V^{\left(k\right)}$) adds native overlap semantics by preserving the EI meta-graph 6-regularity without reducing the vertex count; regime B (deterministic index-7 sublattice, ${\mathcal{C}}^{\left(k\right)}={\mathsf{\Lambda }}_k$ with $\mathbf{a}=2-\omega$) is a theoretical construction establishing an exact sublattice density ratio of 7 per level on the infinite lattice $\mathbb{T}$, and exact termination in $K={log}_{7}n$ levels on finite periodic domains with $n=7^K$. Unconditional results: EI 6-regularity in regime A; perfect star-tiling and exact index-7 structure in regime B; strict hierarchy via Voronoi-completed clusters in regime C; tile-shape alternation proven at levels 1–2 (hexagonal, then triangular-like) and conjectured, with numerical verification, beyond; $h_{\mathrm{Ward}}$ (as an unnormalised SSE) strictly admissible and $h_{max}$ weakly admissible. Aggregation complexity, embedding excluded: $O(nlogn)$ in regime C, $O\left(n\right)$ in regime B, $O(n·K_{max})$ in regime A. We provide a fully reproducible reference implementation (trihex2, MIT-licensed) with extensive parameter sweeps on UCI benchmarks, synthetic Gaussians, non-convex shapes, and overlapping distributions. The genuine contributions of the framework are the multi-scale hierarchical structure with provable geometric guarantees and, in regime A, native overlap semantics that no hard-clustering baseline can provide. A central empirical finding concerns the embedding: an ablation isolating the 2D-lattice projection shows it to be the main bottleneck, and a lattice-free variant that runs the same combinatorial core directly on a k-nearest-neighbour graph in the original feature space—with no embedding and no quantisation—removes the projection entirely and improves accuracy on six of seven pilot datasets. With a frozen, fully unsupervised meta-selection rule (graph-geodesic arbitration between a convex-consensus and a graph-min-cut candidate, no per-dataset tuning), this variant reaches ARI $0.871$ on moons and $1.000$ on circles, where k-means, HAC, and GMM all collapse to $\le 0.43$ and $\le 0.00$, respectively. On a 73-dataset benchmark (23 real UCI, 50 synthetic, all loaded with validated class labels), TriHex is the most frequently best method on the synthetic panel ($46\%$ win rate) and close behind GMM overall ($34\%$ versus $36\%$), while having the lowest mean ARI—the signature of a specialist: it dominates on non-convex structure (rings, spirals, manifolds) and is outperformed on convex tabular data, where we make no claim of superiority. We also report a genuine robustness limitation: with the default configuration, TriHex fails on Cancer (ARI $0.042$, essentially uncorrelated with the ground truth) because the default lattice over-fragments a two-class problem; competitive performance requires a dataset-appropriate configuration, and we report this explicitly rather than only the best configurations. On overlapping Gaussians, regime A detects the boundary points that the data-generating process itself classifies as ambiguous with precision $1.00$ at heavy overlap ($\delta =0.5$); the detector over-flags as the clusters separate (precision falls to $0.43$ at $\delta =3.0$), so its usefulness is confined to the strong-overlap regime. Within that regime, it provides a measurable capability unavailable to hard-clustering baselines. Full article

Climate change is intensifying hydrological extremes, yet most frameworks assess drought and flood hazards independently, limiting integrated risk management. This study proposes a two-dimensional analytical framework to characterize the drought-flood continuum, moving beyond single-index approaches. We introduce the Hydro-Hazard Index (HHI) as a directionality metric (HHI = Flood Severity − Drought Severity) to classify the dominant hazard type, and the Total Severity Index (TSI = Flood Severity + Drought Severity) as a complementary metric to quantify overall hazard magnitude. Analyzing multi-temporal data from 115 hexagonal units (2018–2024), we employed dynamic features (trends, changes, volatility) and four machine learning models to classify areas as “flood-prone” based on validated flood records. Our results show HHI values ranging from −2.44 to 8.81, with 20.9% of areas classified as Flood-Dominated (mean HHI = 4.58) and 79.1% as Normal (mean HHI = 0.76). Crucially, the two-dimensional analysis revealed that areas with identical HHI values can have vastly different TSI values, under scoring the importance of our dual-index approach. Random Forest achieved the highest performance in predicting flood-prone status (Accuracy = 0.913, AUC = 0.967, Recall = 1.00), with flood_volatility as the most important predictor (24.2%). Spatial autocorrelation confirmed strong clustering of high-risk areas (Moran’s I = 0.716, p < 0.001). By analyzing flood and drought as distinct but interacting dimensions, this framework provides a more robust and nuanced tool for integrated risk assessment. While acknowledging limitations related to data availability and the need for further independent validation, the proposed framework supports sustainable water resource management and climate adaptation planning under increasing hydrological uncertainty. Full article

Fine-scale evidence remains scarce regarding where ecological quality has improved or deteriorated in the Yangtze River Delta (YRD) and which landscape conditions are associated with these trajectories. We developed a 1 km² hexagon-based diagnostic framework integrating the Remote Sensing Ecological Index (RSEI), Sen–Mann–Kendall trend analysis, Local Moran’s I clustering, recurrence-based ecological stress typology, and XGBoost–SHAP interpretation for 2000–2025. Annual RSEI was standardized by year to capture relative trajectories of ecological quality rather than absolute change under a fixed loading system. The regional mean RSEI fluctuated markedly and declined only slightly, from 0.639 in 2000 to 0.632 in 2025, suggesting that long-term ecological change was nonlinear. At the hexagon scale, 64.77% of valid units showed positive RSEI trends, with significant improvement covering 15.08% of units and significant degradation covering 5.47%. Local Moran’s I identified distinct High–High and Low–Low clusters; persistent low-quality clusters and stable high-quality areas accounted for 10.0% and 7.8% of valid hexagons, respectively. XGBoost–SHAP results indicated statistical associations between RSEI trends and soil moisture, elevation, impervious surface change, and nighttime light change, rather than direct causal effects. This framework provides a spatially explicit basis for identifying priority monitoring areas, ecological stress zones, and differentiated land management units across rapidly urbanizing megaregions. Full article

Land-use/cover change threatens the ecological integrity of the Miombo region of south-central Africa. In Angola, Miombo ecosystems are of high ecological and socio-economic importance, providing rural populations with woody and non-timber forest products. Fire plays an important role in regional agricultural and silvicultural land-use systems. This study contextualised Copernicus land-cover classes at the regional level to analyse LULC transition pathways and their association with fire occurrence in Central Angola. LULC change was assessed using a post-classification comparison approach combined with pixel-based trajectory analysis. Fire activity was analysed using MODIS-derived ignition points, burned-area data, and a hexagonal-grid aggregation approach. At the same time, spatial clustering was assessed using hot spot analysis based on the Getis-Ord Gi* statistic. Differences in mean fire size among LULC transition classes were tested using the Kruskal–Wallis test followed by Dunn’s post hoc test. The results indicate a gradual reduction in forest cover and conversion to Cultivated Land, associated with the expansion of agricultural frontiers and urban areas. Fire activity was highest in areas affected by LULC conversion, with seasonal patterns varying notably among classes. Mean fire size differed by more than two orders of magnitude among transition types. Overall, fire activity was strongly associated with areas undergoing land-cover transition, highlighting the need to integrate fire management into sustainable land-use policies for long-term Miombo conservation. Full article

Hydrogen is a clean and renewable energy carrier, but its reversible storage near ambient conditions remains a major challenge. Here, density functional theory (DFT) combined with ab initio molecular dynamics (AIMD) is employed to assess the newly predicted 2D orthorhombic diboron dinitride (o-B₂N₂) monolayer, in pristine and Li-functionalized forms, as a hydrogen storage medium. On the pristine surface, H₂ physisorbs with binding energies of −0.158 to −0.174 eV. Li atoms anchor strongly at the hexagonal hollow sites ($E_{\mathrm{bind}}$ from −0.979 to −1.321 eV, strongest at the B-rich H1 site), donate 0.65–0.84 $\left|e\right|$ to the substrate, and render the semiconducting monolayer metallic. A positive cluster formation energy ($+0.171$ eV per Li pair) and a 5 ps AIMD simulation at 400 K confirm that the Li adatoms remain dispersed, without clustering. Each ${\mathrm{Li}}^{+}$ center polarizes and binds up to five H₂ molecules, with average adsorption energies of −0.207 to −0.336 eV/H₂, within the optimal window for room-temperature reversible storage. The 4Li@o-B₂N₂(20H₂) system attains a theoretical gravimetric capacity of 15.12 wt% and a practical capacity of 10.99 wt% under realistic operating conditions (charging at 30 atm/25 °C; release at 3 atm/100 °C). These results establish Li-functionalized o-B₂N₂ as a promising hydrogen storage material that merits experimental exploration. Full article

Rapid urbanization and stringent ecological protection policies in China have reshaped spatial competition among urban, agricultural, and ecological spaces. However, existing studies often overlook how this competition evolves across different slope structures. To address this, this study establishes a fine-scale analytical framework using H3 hexagonal grids and slope spectrum analysis to investigate slope structure evolution and spatial competition patterns from 1990 to 2023. The results reveal a distinct topographic stratification: urban space dominates low-slope regions (<6°) but exhibits a pervasive “upslope expansion” trend, with its average slope increasing from $1.81°$ to $2.07°$, equivalent to an annualized increase of approximately $0.008°{\mathrm{yr}}^{-1}$; agricultural space characterizes the transition zones (6–$15°$), showing an “upslope migration” in the Southeastern Hills associated with urban expansion pressure in low-slope areas; and ecological space functions as a stable barrier in steep terrains (>15°) but faces encroachment in transition zones. Furthermore, cluster analysis identifies significant regional heterogeneity aligned with China’s macro-topography, including “low-slope agglomeration” in the Eastern Plains, “interwoven upslope” patterns in the Southern Hilly Regions, and ecological dominance in the Western Highlands. Association analysis using GeoDetector and Multiscale Geographically Weighted Regression (MGWR) indicates that competition intensity is most strongly associated with human activity factors, especially human footprint and nighttime lights ($q>0.29$), which show the highest explanatory power among the examined factor groups. The interaction between human activity and elevation further shows relatively high explanatory power ($q=0.41$), suggesting that spatial competition is more pronounced where intensive human activities overlap with topographic constraints. Crucially, this study challenges the traditional flat-projection planning model. We propose a transition to “three-dimensional topographic regulation,” advocating differentiated management strategies—such as strict “slope redlines” for urban-agricultural transition zones—to mitigate intensifying spatial conflicts in complex terrains and safeguard agricultural sustainability. Full article

Timely identification of canopy decline in commercial strawberry production is challenging because visual scouting often misses subtle or spatially heterogeneous symptoms. We developed a plant-level UAV-based monitoring framework that integrates repeated multispectral imagery, canopy-derived metrics, unsupervised clustering, and Random Survival Forest (RSF) time-to-event modeling. The framework was applied across three commercial strawberry fields in Oxnard, California using nine UAV surveys collected from December 2022 to June 2023, yielding 159,220 plant-level monitoring units. NDRE- and Redness Index-based classifications quantified proportional and absolute canopy dieback within standardized hexagonal units and supported survival-based modeling of canopy decline progression. Across withheld test plants from all survey dates, overall concordance indices ranged from 0.88 to 0.95 across fields, indicating strong ability to rank plants by time-to-decline risk under heterogeneous field conditions. Spatial risk maps revealed localized high-risk clusters that expanded over time in fields with greater canopy deterioration, while fields with minimal visible decline exhibited diffuse but stable risk distributions. Post-hoc comparison with operational fumigation rates (280, 336, and 392 kg Pic-Clor 60/ha) showed no consistent association with predicted canopy decline risk. These results demonstrate that framing repeated UAV observations as a time-to-event process enables fine-scale spatiotemporal modeling of canopy decline dynamics and supports risk stratification for targeted field monitoring in commercial strawberry systems. Full article

Investigating aluminum nitride (AlN) clusters is essential for understanding the properties of bulk AlN materials. The incorporation of hydrogen into AlN clusters represents an effective strategy for structural modification and for tuning their physicochemical properties. In this work, we conducted density functional theory (DFT) calculations on the dynamically stable global-minimum (GM) structure of Al₆N₆H₈. Compared to the precursor Al₆N₆ cluster, the incorporation of eight hydrogen atoms achieves coordination saturation of all aluminum and nitrogen atoms, inducing a structural transformation from a hexagonal prism with D_3d symmetry to a cuboid structure with D_2h symmetry. The HOMO–LUMO gap of the Al₆N₆H₈ cluster is increased by 1.85 eV compared to that of Al₆N₆, indicating a remarkable enhancement in stability. Chemical bonding and natural bond orbital (NBO) charge analyses reveal that the Al–N, Al–H, and N–H bonds are predominantly covalent single bonds, with a degree of ionicity arising from electronegativity differences. The hydrogen atoms bonded to Al and N can be substituted with a series of other atoms or functional groups, thereby further tuning the structures and properties of the clusters. To facilitate future experimental characterization, the infrared spectrum of Al₆N₆H₈ was calculated, which shows an overall blue shift in the Al–N bond’s bending and stretching vibrations compared to those in the Al₆N₆ cluster. Full article

Pollination syndromes reflect the convergence of floral traits among plants sharing the same pollinator guild. However, bee-pollinated orchids exhibit striking variation in color and size. This diversity reflects the multiple reward strategies that evolved within the family, each interacting differently with bee sensory biases. Here, we tested whether the complex floral visual displays of orchids differ in signal identity and intensity among reward systems. We also considered intrafloral modularity, measured as the color differentiation among flower parts, and color–size integration. For this, we measured and modeled floral morphometric and reflectance data from sepals, petals, lip tips, and lip bases under bee vision from 95 tropical Epidendroid species to compare chromatic and achromatic contrasts, spectral purity, and mean reflectance across wavebands, plus flower and display size, among reward systems. Reward types included 19 food-deceptive, 8 nectar-offering, 10 oil-offering, 11 fragrance-offering, and 47 orchid species of unknown reward strategy. Principal component analyses on 34 color and 9 size variables summarized major gradients of visual trait variation: first component (19.1%) represented overall green-red reflectance and achromatic contrasts, whereas the second (16.5%) captured chromatic contrast–size covariation. Reward systems differed mostly in signal identity rather than signal intensity. Flower chromatic contrasts presented strong integration with flower size, while achromatic contrasts were negatively associated with display size. While deceptive and nectar-offering orchids tend toward larger solitary flowers with bluer and spectrally purer displays, oil- and fragrance-offering orchids tend toward smaller, brownish, or yellow to green flowers, with larger inflorescences. Rewardless orchids presented more achromatically conspicuous signals than rewarding orchids, but smaller displays. Orchid species clustered by reward both in PCA spaces and in bee hexagon color space. Deceptive orchids were typically associated with UV + White colors, oil orchids with UV + Yellow lip tips, and fragrance orchids with UV-Black lip bases and UV-Green lip tips. Together, these results indicate that orchid reward systems promote qualitative rather than quantitative differentiation in visual signals, integrating display color and size. These long-evolved distinct signals potentially enable foraging bees to discriminate among resource types within the community floral market. Our results demonstrate that color and flower display size are important predictors of reward strategy, likely used by foraging bees for phenotype-reward associations, thus mediating the evolution of floral signals. Full article

Cadmium sulfide (CdS) nanoparticles are classified as II-VI semiconductor materials, used in optoelectronic devices due to a band gap (E_g). In this study, CdS nanoparticles were synthesized by the chemical precipitation method, and a systematic evaluation of pH (4.8–10.1) and temperature (50, 75, and 90 °C) was conducted. The effects of these variables were evaluated by UV-VIS spectroscopy, X-ray diffraction (XRD), and scanning electron microscopy (SEM). Results demonstrate that variables determine crystallite sizes (C_s), cluster sizes, and optical properties. CdS crystallization is more affected by pH conditions than by temperature during synthesis; the change in peak intensity in 2θ = 24–29° suggests the formation of a cubic phase (alkaline conditions) and a transition to a hexagonal phase (acidic conditions). Higher temperature improves the quality of the nanoparticles, as evidenced by the reduction in intensity of the peaks associated with secondary materials. The synthesis conditions of CdS nanoparticles significantly affect E_g, widening the range from 2.21 to 2.40 eV. Both temperature and pH conditions change the size of nanoparticles and clusters. Acid conditions promote the formation of rounded and uniform nanoparticles, while alkaline conditions form the largest crystals of CdS. These findings are useful for developing electronic devices that require different semiconductor profiles. Full article

La-doped ZnO nanoclusters confined within mesoporous SBA-15 were synthesized using an impregnation–calcination method and evaluated for their visible-light-driven photocatalytic degradation of Rhodamine B (RhB). Small-angle X-ray diffraction (XRD) and transmission electron microscopy (TEM) confirmed the preservation of the 2D hexagonal mesostructure of SBA-15 post-loading. In contrast, wide-angle XRD and Fourier-transform infrared spectroscopy (FT-IR) analyses revealed that the incorporated ZnO existed predominantly as highly dispersed amorphous or ultrafine clusters within the mesopores. N₂ adsorption–desorption measurements exhibited Type IV isotherms with H1 hysteresis loops. Compared to pristine SBA-15, the specific surface area and pore volume of the composites decreased from 729.35 m² g⁻¹ to 521.32 m² g⁻¹ and from 1.09 cm³ g⁻¹ to 0.85 cm³ g⁻¹, respectively, accompanied by an apparent increase in the average pore diameter from 5.99 nm to 6.55 nm, attributed to non-uniform pore occupation. Under visible-light irradiation, the photocatalytic performance was highly dependent on the La doping level. Notably, the 5% La-ZnO/SBA-15 sample exhibited superior activity, achieving over 99% RhB removal within 40 min and demonstrating the highest apparent rate constant (k = 0.1152 min⁻¹), surpassing both undoped ZnO/SBA-15 (k = 0.0467 min⁻¹) and other doping levels. Reusability tests over four consecutive cycles showed a consistent degradation efficiency exceeding 93%, with only a ~7 percentage-point decline, indicating excellent structural stability and recyclability. Radical scavenging experiments identified h⁺, ·OH, and ·O₂⁻ as the primary reactive species. Furthermore, photoluminescence (PL) quenching observed at the optimal 5% La doping level suggested suppressed radiative recombination and enhanced charge carrier separation. Collectively, these results underscore the synergistic effect of La doping and mesoporous confinement in achieving fast, efficient, and recyclable photocatalytic degradation of organic pollutants. Full article

Mapping of rice-cropping regimes is crucial for effective irrigation planning and yield monitoring, particularly in data-scarce regions. We analyzed 48 months of 3 m PlanetScope NDVI data, aggregated to a 25 m hexagonal grid, and used Dynamic Time Warping Clustering to segment phenological patterns. Internal validation consistently identified two main clusters, indicating two dominant seasonality modes. Cluster 1 exhibited a higher mean NDVI, fewer low-canopy months, more vigorous growth periods, more peaks, and greater annual cycling, which suggests irrigated double cropping. Cluster 2 exhibited prolonged low NDVI values and a greater amplitude, consistent with single-rainfed systems. The rain–NDVI analysis supported these findings: Cluster 1 responded modestly to rainfall, whereas Cluster 2 exhibited a stronger and delayed response. Independent spatial checks confirmed these classifications. Off-season greenness, measured as NDVI above 0.50 from July to November, was concentrated near main and secondary canals and decreased with distance from intake points. This workflow combines DTW clustering with rainfall lag and off-season greenness analysis, effectively distinguishing between irrigated and rain-fed regimes using satellite time series. These findings are considered indicative rather than definitive, providing an assessment of cropping systems in Timor-Leste and demonstrating that DTW-based NDVI clustering offers a scalable approach in data-scarce regions. Full article

GPS tracking has made ski injury data abundant, yet few studies have mapped where incidents actually occur or how those patterns differ between skiers and snowboarders. To address this gap, we analyzed 8719 GPS-located incidents (4196 skier; 4523 snowboarder) spanning four seasons (2017–2022, excluding 2019–2020 due to COVID-19) at a large West Coast resort in California. Incidents were aggregated into 45 m hexagons and analyzed using Getis–Ord Gi* hot spot analysis, Local Outlier Analysis (LOA), and a space–time cube with time-series clustering. Hot spot analysis identified both activity-specific and overlapping high-injury concentrations at the 99% confidence level (p < 0.01). The LOA revealed no spatial overlap between skier and snowboarder High-High classifications (areas with high incident counts surrounded by other high-count areas) at the 95% confidence level. Temporal analysis exposed distinct patterns by activity: Time Series Clustering revealed skier incidents concentrated at holiday-sensitive locations versus stable zones, while snowboarder incidents separated into sustained high-activity versus baseline areas. These findings indicate universal safety strategies may be insufficient; targeted, activity-specific interventions may warrant investigation. The methodology provides a reproducible framework for spatial injury surveillance applicable across the ski industry. Full article

The development of highly selective and sensitive gas sensors is essential for detecting toxic pollutants, such as carbon monoxide (CO), which pose severe health and environmental risks. In this work, the adsorption of CO molecules on Ni_6−xCu_x (x = 0–6) clusters supported on hexagonal boron nitride quantum dots with nitrogen vacancies (h-BN_VQDs) is explored through density functional theory (DFT) calculations. For this purpose, the stability of the metallic clusters supported on the boron nitride sheet was calculated, and the adsorption properties of the CO molecule on the Ni_6−xCu_x (x = 0–6)/h-BN_VQDs composite were determined. The results demonstrated a high binding energy between Ni_6−xCu_x (x = 0–6) clusters and the h-BN_VQDs sheets, suggesting that Ni-Cu clusters are highly stable on h-BN_VQDs sheets. For CO adsorption, adsorption energy and charge transfer calculations indicated that the Ni₆ and Ni_6−xCu_x (x = 2 and 3) clusters exhibit the strongest CO binding and highest charge transfer, suggesting them as good candidates for CO gas sensing. These findings provide theoretical insights into the rational design of bimetallic catalysts for gas-sensing applications. Full article

Urban areas increasingly face heat-related climate risks, necessitating targeted, nature-based interventions such as tree planting to improve resilience, livability, and public health. This study presents a data-driven workflow to identify urban tree planting potential (TPP) in the city of Dortmund, Germany. The approach integrates high-resolution spatial datasets capturing land cover, shading, thermal comfort, population density, and critical infrastructure. All variables were harmonized within a 50 m hexagonal grid, normalized, and combined into a composite TPP score using weighting schemes informed by expert judgment and sensitivity testing. Spatial and non-spatial clustering were applied to group urban areas by shared characteristics, and a connectivity analysis evaluated the spatial coherence of high-potential cells and their relationship to existing green infrastructure. The findings demonstrate the potential to strengthen urban green infrastructure and guide coordinated planting strategies while addressing both ecological and social priorities. The presented workflow offers a flexible, transferable tool to support municipalities in prioritizing effective greening interventions and integrating climate adaptation objectives into urban development planning. Full article