What question did this study set out to answer?

This research aims to assess novel hybrid MCDM approaches for analyzing flood hazards in South Iran.

June 11, 2026Journal of Hydrology Regional Studies0 citationsOpen Access

Capability assessment of the novel hybrid MCDM GIS-based approaches and machine learning algorithm for flood hazard analysis in South of Iran

KSKarim SolaimaniSari Agricultural Sciences and Natural Resources University SDShadman DarvishiUniversity of Kurdistan

Key Points

This research aims to assess novel hybrid MCDM approaches for analyzing flood hazards in South Iran.
Evaluated the capability of hybrid MCDM methods (Fuzzy PIPRECIA, FuzzyCoCoSo, WASPAS, OWA, COPRAS) for flood hazard maps.
Used Sentinel-1 and Sentinel-2 imagery along with the Random Forest algorithm to assess vulnerability of land use.
Analyzed flood conditions based on percentage areas affected by different hazard levels.
Fuzzy PIPRECIA model showed a correlation coefficient of 0.8393 for flood hazard mapping.
34.40% to 48.50% of the Southern Baluchistan basin is identified in very high and high flood hazard conditions.
Floods in 2019 and 2022 caused significant damage, affecting 1056.47 km² of rangeland and other land areas.

Abstract

Study region Southern Baluchistan basin, Iran. Study Focus Presenting new methods is one of the essential steps in developing a methodology for flood occurrence, which is a crucial measure in environmental planning for effective flood management. In this study, the capability of several hybrid approaches, including extended fuzzy-pivot pairwise relative criteria importance assessment (Fuzzy PIPRECIA), fuzzy-combined compromise solution (FuzzyCoCoSo), weighted aggregated sum product assessment (WASPAS), ordered weighted average (OWA), and complex proportional assessment (COPRAS), was investigated to prepare flood hazard maps (FHM). Furthermore, the vulnerability of land use categories (LUC) in FH conditions and flooded LUC was evaluated using Sentinel-1 and Sentinel-2 images and the Random Forest algorithm. New hydrological insights for the region: The Fuzzy PIPRECIA model, with a Spearman rank correlation coefficient of 0.8393, and the new hybrid MCDM, with an average ROC-AUC coefficient of more than 0.81, had very good and excellent performances for FHM preparation. Based on the FuzzyCoCoSo, FuzzyPIPRECIA-FuzzyCoCoSo-OWA, FuzzyPIPRECIA-WASPAS and FuzzyPIPRECIA-COPRAS models, about 34.40% (16,712.69 km 2 ), 41.20% (20,014.48 km 2 ), 48.50% (23,562.26 km 2 ), and 31.54% (15,323.25 km 2 ) of the southern Baluchistan basin are in very high and high FH conditions, 29.67%, 31.06%, 36.14% and 34.82% are in medium FH, and 25.99%, 20.24%, 17.18% and 33.59% are in low FH conditions, respectively. Additionally, the 2019 and 2022 floods in the southern parts resulted in damage to rangeland (1056.47 km²), bare ground (477.96 km²), built areas (6.17 km²), and crops (16.15 km²). These situations demonstrate the critical flood condition and the demand for appropriate flood management planning in the southern Baluchistan basin.

问 AI

Bookmark

View Full Paper