GIS-Based Spatiotemporal Trend Analysis of Landslide Occurrence in Türkiye: A Comparative Assessment of NonParametric Methods Using a National Inventory (1950–2023)


YURTSEVEN H., ESİN A. İ., ÖZOCAK M., AKAY A. O., AKGÜL M.

Arabian Journal for Science and Engineering, 2026 (SCI-Expanded, Scopus)

  • Yayın Türü: Makale / Tam Makale
  • Basım Tarihi: 2026
  • Doi Numarası: 10.1007/s13369-026-11742-9
  • Dergi Adı: Arabian Journal for Science and Engineering
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, zbMATH, Materials Science & Engineering Collection (ProQuest), Technology Collection (ProQuest)
  • Anahtar Kelimeler: Computational trend screening, Geohazard informatics, GIS-based landslide inventory, Inventory uncertainty, Nonparametric trend detection, Spatiotemporal trend analysis
  • İstanbul Üniversitesi-Cerrahpaşa Adresli: Evet

Özet

This study presents a GIS-supported spatiotemporal analysis of long-term landslide occurrence across 81 provinces of Türkiye for 1950–2023 using normalized annual landslide densities, expressed as events per 1,000 km2. A national landslide inventory was structured within a GIS environment, linked to provincial and regional spatial units, normalized by province area, and analyzed as a standardized spatiotemporal geohazard database. Four nonparametric trend detection methods were comparatively applied: Mann–Kendall, Sen’s slope estimator, Spearman’s rank correlation test, and Innovative Trend Analysis. Statistical significance was evaluated at the 1, 5, and 10% levels. The Mann–Kendall and Spearman tests detected significant trends in 17 provinces and showed complete directional agreement, while Sen’s slope estimator identified significant increasing trends in 8 provinces. Innovative Trend Analysis detected significant trends in 69 provinces, including 38 increasing and 31 decreasing trends, revealing a broader spatial pattern than the classical rank-based tests. Regionally, increasing trends were consistently observed in the Black Sea Region, whereas decreasing tendencies in the Aegean and Central Anatolia regions were mainly identified by Innovative Trend Analysis. Eastern and Southeastern Anatolia displayed mixed patterns, reflecting spatial heterogeneity. The findings show that long-term landslide trend interpretation is strongly method-dependent and sensitive to inventory structure. Rather than establishing the universal superiority of a single method, this study demonstrates how a GIS-structured landslide inventory can support comparative computational trend screening, geohazard informatics, and cautious interpretation of spatially heterogeneous landslide signals under inventory-related uncertainty.