Jayawardane et al. (2026) Comparative urban flood hazard mapping using GIS-integrated multi-criteria decision analysis: A remote sensing approach for Colombo, Auckland, and Valencia

Identification

• Journal: Environmental Earth Sciences
• Year: 2026
• Date: 2026-03-30
• Authors: Paboda Jayawardane, Lalith Rajapakse, Chandana Siriwardana
• DOI: 10.1007/s12665-026-12888-3

Research Groups

• Department of Civil Engineering, University of Moratuwa, Moratuwa, Sri Lanka
• School of Built Environment, Massey University, Auckland, New Zealand

Short Summary

This study applies a GIS-integrated hybrid Analytical Hierarchy Process (AHP) framework to map urban flood hazard zones in Colombo (Sri Lanka), Auckland (New Zealand), and Valencia (Spain). It reveals that 13.64% of Colombo, 25.64% of Auckland, and 17.63% of Valencia fall under extremely high flood hazard levels, offering insights for area-specific mitigation strategies.

Objective

• To assess the relative influence of flood susceptibility factors (FSFs) and identify urban flood hazard (UFH) zones in Colombo, Auckland, and Valencia using a GIS-integrated multi-criteria decision analysis.
• To apply and validate a novel cross-regional hybrid Analytical Hierarchy Process (AHP) framework across three climatically and geographically distinct urban environments.

Study Configuration

• Spatial Scale: Colombo Metropolitan Area, Sri Lanka (27.165 km²); Auckland, New Zealand (4,938 km²); Valencia, Spain (10,512 km²).
• Temporal Scale: Land Use and Land Cover (LULC) maps for 2023; FOA-UNESCO Soil Map updated 2023; real-time precipitation data and historical annual average rainfall values; flood events up to 2024.

Methodology and Data

• Models used: GIS-based Multi-Criteria Decision Analysis (MCDA), Hybrid Analytical Hierarchy Process (AHP) integrating Traditional AHP, Fuzzy AHP, and AHP-TOPSIS, Weighted Overlay Analysis (ArcGIS Pro V3.3).
• Data sources:
  • Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM) (30 m resolution) for topography (elevation, slope, drainage density).
  • Landsat 8 Operational Land Imager (OLI) / Sentinel 2 remote sensing data (10 m resolution) for Land Use and Land Cover (LULC).
  • Real-time precipitation data / Meteorological data from Meteorological Departments for rainfall maps.
  • FOA-UNESCO’s Soil Map for the World (updated 2023) for soil types.
  • Expert judgments from professionals in hydrology, urban flood management, and geospatial analysis (from Sri Lanka and New Zealand).

Main Results

• The study successfully identified urban flood hazard (UFH) zones in Colombo, Auckland, and Valencia, classifying them into four levels: Extremely High, High, Moderate, and Low.
• Colombo: 13.64% (3.70 km²) of the area is classified as Extremely High hazard, and 18.19% (4.94 km²) as High hazard. Key drivers include flat terrain (82% between 0°–5° slope, 36% in 0–5 m MSL elevation), high built-up density (85.6%), and proximity to water bodies.
• Auckland: 25.64% (1,266 km²) of the area is classified as Extremely High hazard, and 14.32% (707 km²) as High hazard. High hazard levels are concentrated in the city area due to dense built-up environments, proximity to urban river networks, and relatively low elevation and slope.
• Valencia: 17.63% (1,833 km²) of the area falls under extremely high hazard levels, while 32.11% (3,375 km²) is classified as high hazard. High susceptibility is found in Valencia City and downstream reaches of the Turia and Jucar Rivers, driven by extreme precipitation (DANA events), low elevation, low slope, high built-up density, high drainage density, and alluvial soil types.
• The hybrid AHP approach (integrating traditional AHP, Fuzzy AHP, and AHP-TOPSIS) demonstrated robustness and consistency in assigning weights to flood susceptibility factors, with minimal deviations between methods.
• Validation of land use maps using the Kappa coefficient showed "almost perfect agreement" for Colombo (0.9683), Auckland (0.8625), and Valencia (0.9125).

Contributions

• Unique application of a cross-regional hybrid AHP framework for urban flood hazard mapping across three climatically and geographically distinct cities (Colombo, Auckland, Valencia), which was previously unexplored.
• Methodological improvement by integrating traditional AHP, Fuzzy AHP, and AHP-TOPSIS for refining and validating flood susceptibility factor (FSF) weights, reducing subjectivity and uncertainty in expert judgments.
• Demonstrated the transferability and applicability of a unified hybrid AHP–GIS framework across contrasting urban flood regimes (coastal, riverine, island-continental hydrological settings).
• Provided valuable, area-specific flood hazard maps and insights for urban planners and disaster management authorities to implement sustainable flood mitigation strategies, particularly in data-scarce and uncertainty-prone conditions.

Funding

No funding was allocated for this research.

Citation

@article{Jayawardane2026Comparative,
  author = {Jayawardane, Paboda and Rajapakse, Lalith and Siriwardana, Chandana},
  title = {Comparative urban flood hazard mapping using GIS-integrated multi-criteria decision analysis: A remote sensing approach for Colombo, Auckland, and Valencia},
  journal = {Environmental Earth Sciences},
  year = {2026},
  doi = {10.1007/s12665-026-12888-3},
  url = {https://doi.org/10.1007/s12665-026-12888-3}
}