Urban flood risk analysis using presence-only machine learning approach: an integrated MaxEnt-cloud model framework in Harbin, China

Jinhui Hu (), Aoxuan Pang () and Changtao Deng ()
Additional contact information
Jinhui Hu: Huazhong University of Science and Technology
Aoxuan Pang: Qingdao Harbin Engineering University Innovation and Development Center
Changtao Deng: Pearl River Water Resources Research Institute

Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards, 2025, vol. 121, issue 14, No 26, 16827-16856

Abstract: Abstract Urban flooding is a devastating global natural disaster requiring robust risk assessment. Traditional methods often struggle with presence-only data (locations where floods occurred, without data on non-flood locations) and clear risk level classification. This study introduces an innovative integrated framework for urban flood risk analysis in Harbin, China, combining Maximum Entropy (MaxEnt) modeling, which excels with presence-only data, with Cloud Model theory for improved risk classification. Methodologically, our framework incorporates: (1) MaxEnt modeling using various environmental and socioeconomic factors; (2) spatial autocorrelation analysis to understand flood distribution; (3) Cloud Model theory implementation, which transforms probability values into linguistic risk levels while representing the uncertainty and fuzziness in these transitions; and (4) a comparative evaluation against other machine learning models. Key findings reveal that the MaxEnt model demonstrated strong predictive performance (AUC = 0.871), comparable to other approaches, despite only needing flood presence data. Human-related factors contributed over half of the total influence on flood risk. Clay loam soil showed the highest flood probability. Spatial analysis identified significant clustering of flood events in low-lying urban areas with inadequate drainage. The Cloud Model classification effectively translated flood probabilities into five risk levels, highlighting that low-risk areas predominate in peripheral high-elevation districts, while high-risk areas are concentrated in Harbin's urban core. This pioneering integration of MaxEnt with Cloud Model theory offers a robust methodology for urban flood risk assessment. The implications include providing a more nuanced, data-driven yet expert-knowledge-informed basis for urban planning, infrastructure development, and emergency management in flood-prone regions.

Keywords: Urban flooding; Maximum entropy; Cloud model theory; Risk assessment; Machine learning (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s11069-025-07452-4 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:nathaz:v:121:y:2025:i:14:d:10.1007_s11069-025-07452-4

Ordering information: This journal article can be ordered from
http://www.springer.com/economics/journal/11069

DOI: 10.1007/s11069-025-07452-4

Access Statistics for this article

Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards is currently edited by Thomas Glade, Tad S. Murty and Vladimír Schenk

More articles in Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards from Springer, International Society for the Prevention and Mitigation of Natural Hazards
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().