EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Weighted Hybrid Clustering of Dynamic Time Warping and Energy Distance: Identifying Biocapacity Heterogeneity Among EU Member States

Monika Hadaś-Dyduch

Sustainability · 06.11.2026 · 18 (21) · 11350

DOI: 10.3390/su182111350 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sustainability, online erschienen am 06.11.2026. Verfasst von Monika Hadaś-Dyduch. Die Arbeit nennt 66 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Biocapacity is a key indicator of sustainable development. This study identifies homogeneous groups of 21 European Union member states characterised by similar temporal and distributional profiles of biocapacity over 1961–2025. The main methodological contribution is a hybrid clustering framework combining Dynamic Time Warping (DTW) and Energy Distance through an optimised linear weighted distance, followed by hierarchical clustering and bootstrap-based stability assessment. The framework is general; however, for the preferred fine-grained solution the optimisation selected the boundary value α*=1.0, meaning that the Energy Distance component receives zero weight at k=12. This is reported as a substantive empirical finding: for a fine-grained partition, temporal similarity alone suffices. For coarser solutions (k=3–6), the optimal α ranges from 0.45 to 0.65, indicating that both components contribute when the partition is less granular. A hierarchical typology is therefore proposed: a parsimonious macro-level classification (k=3–5, with genuinely hybrid weights) is used for the main conclusions, while the 12-cluster solution is retained as a secondary, fine-grained classification. The analysis identified 12 clusters among the 21 countries. Of these 12 clusters, 7 are singletons (Austria, Lithuania, Denmark, Croatia, Estonia, Latvia, Sweden) and 3 are two-element clusters (Belgium–Italy, Cyprus–Malta, Poland–Romania), meaning that 10 of 12 clusters (83.3%) and 13 of 21 countries (61.9%) fall into single- or two-element groups. The difference between the highest and lowest cluster mean is 9.793 gha per capita (17-fold). Scandinavian and Baltic countries achieved very high values (6.57–10.37 gha per capita), while Mediterranean and some Western European countries recorded values below 1.0 gha per capita. The solution showed high bootstrap stability (mean ARI =0.9589). External validation using independent indicators (forest share, agricultural share, population density, GDP per capita) confirms that the macro-level clusters differ significantly on these variables (Kruskal–Wallis, p

Zitieren

APA

Hadaś-Dyduch, M. (2026). Weighted Hybrid Clustering of Dynamic Time Warping and Energy Distance: Identifying Biocapacity Heterogeneity Among EU Member States. Sustainability, 18(21), 11350. https://doi.org/10.3390/su182111350

Vancouver

Hadaś-Dyduch M. Weighted Hybrid Clustering of Dynamic Time Warping and Energy Distance: Identifying Biocapacity Heterogeneity Among EU Member States. Sustainability. 2026;18(21):11350. doi:10.3390/su182111350

BibTeX
@article{HadaDyduch2026,
  author = {Hadaś-Dyduch, Monika},
  title = {Weighted Hybrid Clustering of Dynamic Time Warping and Energy Distance: Identifying Biocapacity Heterogeneity Among EU Member States},
  journal = {Sustainability},
  year = {2026},
  volume = {18},
  number = {21},
  pages = {11350},
  doi = {10.3390/su182111350}
}
Kostenlos starten →