EUVIMEDEUVIMED entdecken
Crossref

A morphometry driven FAHP machine learning framework for flood, erosion and landslide susceptibility in Chitral sub‐basins, Pakistan

Anam Munawwar, Amir Ali, Saira Batool, Sumaira Kousar, Hamid Gulzar, Syed Amer Mahmood, Hania Arif, Olga D. Kucher, Khadeijah Yahya Faqeih, Eman Rafi Alamery, Asma Abdulaziz Abdullah AbalKhail, Aqil Tariq

Earth Surface Processes and Landforms · 2026-07 · 51 (7)

DOI: 10.1002/esp.70365 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Earth Surface Processes and Landforms, online erschienen am 23.07.2026. Verfasst von Anam Munawwar und 11 weiteren Autor:innen. Die Arbeit nennt 51 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study assesses the sensitivity of hydrological and geomorphic hazards in the Chitral sub‐basins of the Hindukush using GLODEM‐30m elevation data. Linear, areal and relief morphometric parameters were derived for nine sub‐basins, including drainage density (D d ), stream frequency (F s ), bifurcation ratio (R b ), basin relief (B h ), relief ratio (R h ) and basin shape indices. The fuzzy analytic hierarchy process (FAHP) was applied to quantify the relative importance of morphometric parameters. Resultant weights were used to compute the Composite Priority Index (CPI) for basin ranking. Machine learning (ML) models of multiple linear regression (MLR), artificial neural network (ANN) and support vector regression (SVR) were used to validate the FAHP‐CPI using leave‐one‐out cross‐validation (LOOCV). Results showed that D d (26.4%), F s (21.6) and Bh (17.8%) were the most influential factors on hazard sensitivity. Drainage‐related indices accounted for about 58% of the total weight. CPI values indicate that high‐relief upstream Basins 7, 8, 9 and 6 are more susceptible to flooding, erosion and landslide hazards, with CPI values of 0.61, 0.58, 0.56 and 0.56, respectively. Whereas elongated basins exhibit water retention potential. Among ANN, MLR and SVR, the MLR showed the best CPI prediction, with absolute errors mostly between 0.00 and 0.02. Unlike conventional morphometric studies, the approach directly links CPI results to flood, erosion, landslide and water retention. The novelty of this study lies in integrating morphometric FAHP weighting with ML‐based CPI validation to prioritize flood, erosion, landslide and water retention in a data‐scarce mountainous basin. The proposed FAHP‐ML framework is transferable and provides practical decision support for disaster mitigation and watershed management.

Zitieren

APA

Munawwar, A., Ali, A., Batool, S., Kousar, S., Gulzar, H., Mahmood, S. A., Arif, H., Kucher, O. D., Faqeih, K. Y., Alamery, E. R., AbalKhail, A. A. A., & Tariq, A. (2026). A morphometry driven FAHP machine learning framework for flood, erosion and landslide susceptibility in Chitral sub‐basins, Pakistan. Earth Surface Processes and Landforms, 51(7). https://doi.org/10.1002/esp.70365

Vancouver

Munawwar A, Ali A, Batool S, Kousar S, Gulzar H, Mahmood SA, et al. A morphometry driven FAHP machine learning framework for flood, erosion and landslide susceptibility in Chitral sub‐basins, Pakistan. Earth Surface Processes and Landforms. 2026;51(7). doi:10.1002/esp.70365

BibTeX
@article{Munawwar2026,
  author = {Munawwar, Anam and Ali, Amir and Batool, Saira and Kousar, Sumaira and Gulzar, Hamid and Mahmood, Syed Amer and Arif, Hania and Kucher, Olga D. and Faqeih, Khadeijah Yahya and Alamery, Eman Rafi and AbalKhail, Asma Abdulaziz Abdullah and Tariq, Aqil},
  title = {A morphometry driven FAHP machine learning framework for flood, erosion and landslide susceptibility in Chitral sub‐basins, Pakistan},
  journal = {Earth Surface Processes and Landforms},
  year = {2026},
  volume = {51},
  number = {7},
  doi = {10.1002/esp.70365}
}
Kostenlos starten →