Rezaie et al. (2026) Evaluating the Impact of Climate and Land-Use Change on Flood Susceptibility on a Global Scale

Identification

• Journal: Water Resources Management
• Year: 2026
• Date: 2026-02-01
• Authors: Fatemeh Rezaie, Saeid Janizadeh, Sayed Mohammadreza Bateni, Changhyun Jun, Dongkyun Kim, Zahra Kalantari, Essam Heggy
• DOI: 10.1007/s11269-025-04404-2

Research Groups

• Department of Geophysical Exploration, Korea University of Science and Technology, Daejeon, Republic of Korea
• Department of Geospatial Information Systems, Faculty of Geodesy and Geomatics Engineering, K. N. Toosi University of Technology, Tehran, Iran
• Department of Civil, Environmental, and Construction Engineering and Water Resources Research Center, University of Hawai‘I at Manoa, Honolulu, HI, USA
• UNESCO-UNISA Africa Chair in Nanoscience and Nanotechnology College of Graduate Studies, University of South Africa, Pretoria, South Africa
• School of Civil, Environmental and Architectural Engineering, College of Engineering, Korea University, Seoul, Republic of Korea
• Department of Civil Engineering, Hongik University, Mapo-Gu, Seoul, Republic of Korea
• Department of Sustainable Development, Environmental Science and Engineering (SEED), KTH Royal Institute of Technology, Stockholm, Sweden
• Department of Physical Geography and Bolin Centre for Climate Research, Stockholm University, Stockholm, Sweden
• Institut de Physique du Globe de Paris, Université Paris Cité, CNRS, Paris, France
• Viterbi School of Engineering, University of Southern California, Los Angeles, CA, USA

Short Summary

This global-scale study investigates the combined impacts of climate and land-use change on flood susceptibility in the 21st century using SSP-RCP scenarios and machine learning, finding that Random Forest outperforms other models and projects an increase in high/very high flood susceptibility areas, particularly in Oceania, Europe, and parts of Asia and Africa.

Objective

• To investigate, for the first time on a global scale, the effects of spatial and temporal variability in climate and land use on flood occurrence in the 21st century using Shared Socioeconomic Pathway-Representative Concentration Pathway (SSP-RCP) scenarios (SSP1-RCP2.6 and SSP5-RCP8.5).
• To determine the combined effects of urbanization and climate change on flood-prone areas under various socioeconomic and emission pathways.

Study Configuration

• Spatial Scale: Global scale.
• Temporal Scale: Current period, 2041–2060 (referred to as 2050), and 2061–2080 (referred to as 2070).

Methodology and Data

• Models used: Regularized Logistic Regression (RLR), Boosted Classification Tree (BCT), and Random Forest (RF) machine learning algorithms. An ensemble of 13 Global Climate Models (GCMs) from CMIP6 was used for climate projections.
• Data sources:
  • Flood inventory map: ERTHDATA website (17,347 flood locations worldwide, 1960–2018).
  • Climate variables: WorldClim website (8 dynamic precipitation variables, 2.5 arc-minute spatial resolution).
  • Land-use fractions: Land-Use Harmonization 2 (LUH2) dataset (7 classes, 0.25° × 0.25° spatial resolution).
  • Topographic variables: EarthEnv website (7 factors, 1 km × 1 km spatial resolution), derived from GMTED2010 (250 m resolution) and SRTM4.1dev (90 m resolution).
  • Scenarios: SSP1-RCP2.6 (sustainability) and SSP5-RCP8.5 (fossil-fuel-based development).

Main Results

• The Random Forest (RF) model demonstrated superior predictive performance (AUC = 0.96, sensitivity = 0.91, specificity = 0.86, PPV = 0.88, NPV = 0.90) compared to Boosted Classification Tree (BCT) and Regularized Logistic Regression (RLR).
• Flood susceptibility maps for current and future scenarios (SSP1-RCP2.6 and SSP5-RCP8.5 for 2041–2060 and 2061–2080) project an expansion of areas with high/very high flood susceptibility.
• Regions particularly affected include Oceania (New Zealand, Fiji, Guam, Solomon Islands), Europe, and several Asian and African countries.
• The total global land area with high/very high flood susceptibility is projected to increase from 10.88% (current) to 11.48% by 2070 under the SSP5-RCP8.5 scenario.
• Conversely, the global area with very low flood susceptibility is projected to decrease from 73.63% to 71.79% depending on the scenario.
• Relative importance analysis (using the RF model) identified managed pasture, urban land, and precipitation-related variables (BIO13, BIO16, BIO12) as the most significant drivers of flood susceptibility, while topographic variables played a comparatively minor role.

Contributions

• Generated spatial predictions of flood susceptibility based on statistical and dynamic factors at a global scale.
• Integrated the combined impacts of climate and land-use changes in a unified modeling approach to predict future flood-prone areas under SSP1-RCP2.6 and SSP5-RCP8.5 scenarios for 2050 and 2070.
• Enhanced knowledge of flood modeling by identifying the most influential topographic, meteorological, and land-use factors in global flood susceptibility mapping.
• Provided essential data for informed urban planning, infrastructure development, and risk reduction measures for local and state authorities and policymakers.

Funding

• National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (RS-2025-00518650 & RS-2021-NR060085).

Citation

@article{Rezaie2026Evaluating,
  author = {Rezaie, Fatemeh and Janizadeh, Saeid and Bateni, Sayed Mohammadreza and Jun, Changhyun and Kim, Dongkyun and Kalantari, Zahra and Heggy, Essam},
  title = {Evaluating the Impact of Climate and Land-Use Change on Flood Susceptibility on a Global Scale},
  journal = {Water Resources Management},
  year = {2026},
  doi = {10.1007/s11269-025-04404-2},
  url = {https://doi.org/10.1007/s11269-025-04404-2}
}