Ashraf et al. (2025) Assessing hazardous flash flood susceptibility using multivariate zonation mapping techniques in Pishin District, Balochistan province of Pakistan

Identification

• Journal: Natural Hazards
• Year: 2025
• Date: 2025-10-13
• Authors: Muhammad Ashraf, Q. A. Shah, Adnan Arshad, Ghulam Murtaza
• DOI: 10.1007/s11069-025-07723-0

Research Groups

• Department of Disaster Management and Development Studies, University of Balochistan, Quetta, Pakistan
• State-Key Laboratory Herbage Improvement and Grassland Agroecosystem, Key Laboratory of Grassland Livestock Industry Innovation, Ministry of Agriculture and Rural Affairs, Engineering Research Center of Grassland Industry, College of Pastoral Agriculture Science and Technology, Ministry of Education, Lanzhou University, Lanzhou, China

Short Summary

This study utilized an integrated Analytical Hierarchy Process (AHP), Geographic Information Systems (GIS), and Remote Sensing (RS) approach to map flash flood susceptibility in Pishin District, Pakistan, identifying that 29.7% of the area is highly to very highly susceptible with a model accuracy of 0.993 Area Under the Curve (AUC).

Objective

• To build a GIS and AHP technique-based model for mapping flash flood susceptibility zones in the Pishin District, after conducting a multicollinearity assessment of flood conditioning parameters.
• To assess the efficacy of the AHP method using sensitivity analyses.
• To determine the model's accuracy using Receiver Operating Characteristic (ROC) - Area Under the Curve (AUC) assessment.

Study Configuration

• Spatial Scale: Pishin District, Balochistan province, Pakistan (approximately 66° 13′ to 67° 50′ East longitude and 30° 04′ to 31° 17′ North latitude), with elevations ranging from 1440 to 3196 meters above sea level.
• Temporal Scale: Analysis of a flood event that occurred on August 28, 2022, using data acquired up to October 29, 2024.

Methodology and Data

• Models used: Analytical Hierarchy Process (AHP), Geographic Information Systems (GIS), Remote Sensing (RS), Weighted Overlay, Stillwell Ranking Technique (SRT), Single Parameter Sensitivity Analysis (SPSA), Map Removal Sensitivity Analysis (MRSA), Receiver Operating Characteristic (ROC) analysis, Variance Inflation Factor (VIF), Tolerance (TOL), Inverse Distance Weighted (IDW) interpolation.
• Data sources:
  • Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM) (30 meter resolution) from NASA Earth data.
  • United States Geological Survey (USGS) Landsat 8 OLI/TIRS satellite imagery (30 meter resolution).
  • Average annual rainfall map from PARSIAN CCS (0.4° × 0.4° resolution).
  • Sentinel-1 images and field surveys for flood inventory mapping.
  • Nine flash flood conditioning factors: elevation, slope, drainage density, topographic wetness index (TWI), modified normalized difference water index (MNDWI), stream power index (SPI), distance to rivers, rainfall intensity, and Normalized Difference Vegetation Index (NDVI).

Main Results

• Multicollinearity analysis confirmed no significant multicollinearity issues among the nine flash flood susceptibility parameters (Tolerance > 0.1, VIF < 10).
• The AHP model demonstrated consistency with a Consistency Ratio (CR) of 0.07.
• The flood susceptibility zonation revealed the following distribution:
  • Low susceptibility: 27.41%
  • Medium susceptibility: 23.39%
  • Very Low susceptibility: 19.50%
  • High susceptibility: 16.33%
  • Very High susceptibility: 13.37%
• Areas identified as highly to very highly susceptible to flash floods include Malezai, Manzari, Dad Khanzai, Manzaki, Malikyar, Khairabad, and Torah Shah.
• The most influential susceptibility factors, based on AHP weights, were elevation (0.32), slope (0.22), drainage density (0.13), and distance to the river (0.13).
• Sensitivity analyses (Stillwell ranking, single parameter, and map removal) confirmed the significant influence of elevation and slope on flood susceptibility.
• Model validation using ROC-AUC analysis yielded an accuracy of 0.993 (99%), indicating high robustness and effectiveness in generating the flood susceptibility map.

Contributions

• Provides a comprehensive flash flood susceptibility zonation map for the Pishin District, Balochistan, Pakistan, a region often characterized by data scarcity.
• Integrates Analytical Hierarchy Process (AHP) with Geographic Information Systems (GIS) and Remote Sensing (RS) to offer a robust methodology for flood hazard assessment.
• Incorporates rigorous model performance checks through multicollinearity analysis and multiple sensitivity analyses (Stillwell ranking, single parameter, and map removal) to validate the assigned weights and enhance the reliability of the AHP model.
• Offers practical insights and a valuable tool for planners, hydrologists, and water resource managers to identify flood-vulnerable zones and formulate effective flood mitigation strategies.
• The developed methodology is flexible and applicable to other flood-prone areas globally, particularly in data-limited environments, contributing to broader disaster risk reduction efforts.

Funding

No specific funding projects, programs, or reference codes were provided in the paper text.

Citation

@article{Ashraf2025Assessing,
  author = {Ashraf, Muhammad and Shah, Q. A. and Arshad, Adnan and Murtaza, Ghulam},
  title = {Assessing hazardous flash flood susceptibility using multivariate zonation mapping techniques in Pishin District, Balochistan province of Pakistan},
  journal = {Natural Hazards},
  year = {2025},
  doi = {10.1007/s11069-025-07723-0},
  url = {https://doi.org/10.1007/s11069-025-07723-0}
}