XAI-driven flood risk assessment: Integrating machine learning and hydrological model

Abstract

Increasingly frequent extreme climate events have intensified urban flood risks, underscoring the urgent need for accurate, interpretable assessment methodologies. This study establishes an explainable artificial intelligence (XAI) framework for flood risk assessment in the Guangdong-Hong Kong-Macao Greater Bay Area (GBA), integrating the LISFLOOD-FP hydrodynamic model with Gradient Boosting Decision Tree (GBDT). To resolve model opacity, Local Interpretable Model-agnostic Explanations (LIME) quantifies the contributions of critical disaster-inducing indicators. The framework achieves over 91% predictive accuracy, revealing a 1.33% expansion of very high-risk zones and a 3.80% increase in high-risk areas under the 100-year flood scenario, with the most affected cities including Guangzhou, Shenzhen, Zhuhai, and Foshan. LIME-based interpretability analysis under this scenario underscores the dominant influence of hydrological and topographic variables, with FD (flood depth), SD (submerge duration), and DEM (Digital Elevation Model) collectively contributing over 60% of the total explanatory contribution. This XAI approach significantly enhances flood risk prediction precision, delivering actionable insights for evidence-based resilience planning across the GBA.

Keywords

Flood / Risk assessment / LIME / XAI / GBA introduction

CRediT authorship contribution statement

Meihong Ma: Conceptualization, Methodology, Writing - original draft. Ting Wang: Methodology, Funding acquisition. Jianhua Yang: Conceptualization, Writing - original draft. Zhuoran Chen: Data curation, Visualization. Jinqi Wang: Investigation, Visualization. Ronghua Liu: Formal analysis, Supervision. Xiaoyi Miao: Data curation, Visualization, Writing - review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This research was supported by National Natural Science Foundation of China (42371086, 42271095, 42501092) and National Key Research and Development Program of China (2021YFC3001000).

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.gsf.2025.102244.

References

[1] Akshitha, K., Roopashree, R., Kodipalli, A., Rao, T., 2024. Utilizing explainable AI methodologies:LIME and SHAP, for the classification of natural disasters through machine learning algorithms. In: Paper Presented at the 2024 Parul International Conference on Engineering and Technology (PICET), Waghodia, Gujarat, pp. 1-5.

[2] Bisht, D.S., Chatterjee, C., Kalakoti, S., Upadhyay, P., Sahoo, M., Panda, A., 2016. Modeling urban floods and drainage using SWMM and MIKE URBAN: a case study. Nat. Hazards 84 (2), 749-776. https://doi.org/10.1007/s11069-016-2455-1.

[3] Bui, Q.D., Luu, C., Mai, S.H., Ha, H.T., Ta, H.T., Pham, B.T., 2023. Flood risk mapping and analysis using an integrated framework of machine learning models and analytic hierarchy process. Risk Anal. 43 (7), 1478-1495. https://doi.org/10.1111/risa.14018.

[4] Cantoni, E., Tramblay, Y., Grimaldi, S., Salamon, P., Dakhlaoui, H., Dezetter, A., Thiemig, V., 2022. Hydrological performance of the ERA 5 reanalysis for flood modeling in Tunisia with the LISFLOOD and GR4J models. J. Hydrol. Reg. Stud. 42, 101169. https://doi.org/10.1016/j.ejrh.2022.101169.

[5] Chen, J., Huang, G., Chen, W., 2021. Towards better flood risk management: Assessing flood risk and investigating the potential mechanism based on machine learning models. J. Environ. Manage. 293 (1), 112810. https://doi.org/10.1016/j.jenvman.2021.112810.

[6] Costa, R., Bolte, E., Sharp, C., Bowers, C., 2024. Repeated and localized flooding is an underestimated challenge for urban disaster risk management. Nat. Cities 1 (9), 587-596. https://doi.org/10.1038/s44284-024-00107-8.

[7] Ding, C., Cao, X., Liu, C., 2019. How does the station-area built environment influence Metrorail ridership? Using gradient boosting decision trees to identify non-linear thresholds. J. Transp. Geogr. 77, 70-78. https://doi.org/10.1016/j.jtrangeo.2019.04.011.

[8] Ding, Y., Wang, H., Liu, Y., Lei, X., 2024. Urban waterlogging structure risk assessment and enhancement. J. Environ. Manage. 352, 120074. https://doi.org/10.1016/j.jenvman.2024.120074.

[9] Dong, B., Xia, J., Wang, X., 2025. Comprehensive flood risk assessment in highly developed urban areas. J. Hydrol. 648, 132391. https://doi.org/10.1016/j.jhydrol.2024.132391.

[10] Han, F., Yu, J., Zhou, G., Li, S., Sun, T., 2024. A comparative study on urban waterlogging susceptibility assessment based on multiple data-driven models. J. Environ. Manage. 360, 121166. https://doi.org/10.1016/j.jenvman.2024.121166.

[11] Koks, E.E., Jongman, B., Husby, T.G., Botzen, W.J.W., 2015. Combining hazard, exposure and social vulnerability to provide lessons for flood risk management. Environ. Sci. Policy 47, 42-52. https://doi.org/10.1016/j.envsci.2014.10.013.

[12] Li, J., Yuan, L., Hu, Y., Xu, A., Cheng, Z., Song, Z., Zhang, X., Zhu, W., Shang, W., Liu, J., Liu, M., 2024. Flood simulation using LISFLOOD and inundation effects: a case study of Typhoon In-Fa in Shanghai. Sci. Total Environ. 954, 176372. https://doi.org/10.1016/j.scitotenv.2024.176372.

[13] Liang, W., Luo, S., Zhao, G., Wu, H., 2020. Predicting hard rock pillar stability using GBDT, XGBoost, and LightGBM algorithms. Mathematics 8 (5), 765. https://doi.org/10.3390/math8050765.

[14] Liu, J., Zhao, X., Chen, Y., Sun, H., Gu, Y., Xu, S., 2025. A novel flood conditioning factor based on topography for flood susceptibility modeling. Geosci. Front. 16 (1), 101960. https://doi.org/10.1016/j.gsf.2024.101960.

[15] Ma, M., Zhao, G., He, B., Li, Q., Dong, H., Wang, S., Wang, Z., 2021. XGBoost-based method for flash flood risk assessment. J. Hydrol. 598, 126382. https://doi.org/10.1016/j.jhydrol.2021.126382.

[16] Molnar, C., 2021. Interpretable Machine Learning: A Guide for Making Black Box Models Explainable ( Z. Mingchao, Trans.). Electronic Industry Press, Beijing, p. 312 (in Chinese).

[17] Ribeiro, M. T., Singh, S., Guestrin, C., 2016. ‘‘Why should I trust you?” explaining the predictions of any classifier. In: Paper presented at the Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, pp. 1135-1144.

[18] Shaw, J., Kesserwani, G., Neal, J., Bates, P., Sharifian, M.K., 2021. LISFLOOD-FP 8.0: the new discontinuous Galerkin shallow-water solver for multi-core CPUs and GPUs. Geosci. Model Dev. 14 (6), 3577-3602. https://doi.org/10.5194/gmd-14-3577-2021.

[19] Singha, C., Chakraborty, N., Sahoo, S., Pham, Q.B., Xuan, Y., 2025a. A novel framework for flood susceptibility assessment using hybrid analytic hierarchy process-based machine learning methods. Nat. Hazards 121 (11), 13765-13810. https://doi.org/10.1007/s11069-025-07335-8.

[20] Singha, C., Sahoo, S., Mahtaj, A.B., Moghimi, A., Welzel, M., Govind, A., 2025b. Advancing flood risk assessment: Multitemporal SAR-based flood inventory generation using transfer learning and hybrid fuzzy-AHP-machine learning for flood susceptibility mapping in the Mahananda River Basin. J. Environ. Manage. 380, 124972. https://doi.org/10.1016/j.jenvman.2025.124972.

[21] Sosa, J., Sampson, C., Smith, A., Neal, J., Bates, P., 2020. A toolbox to quickly prepare flood inundation models for LISFLOOD-FP simulations. Environ. Model. Softw. 123, 104561. https://doi.org/10.1016/j.envsoft.2019.104561.

[22] Tehrany, M.S., Pradhan, B., Mansor, S., Ahmad, N., 2015. Flood susceptibility assessment using GIS-based support vector machine model with different kernel types. Catena 125, 91-101. https://doi.org/10.1016/j.catena.2014.10.017.

[23] Wang, Y., Liu, C., Wang, Y., Liu, Y., Liu, T., 2024. Climate risk assessment and adaption ability in China’s coastal urban agglomerations - a case study of Guangdong-Hong Kong-Macao greater bay area. J. Clean. Prod. 452 (1), 142036. https://doi.org/10.1016/j.jclepro.2024.142036.

[24] Wang, Z., Lai, C., Chen, X., Yang, B., Zhao, S., Bai, X., 2015. Flood hazard risk assessment model based on random forest. J. Hydrol. 527, 1130-1141. https://doi.org/10.1016/j.jhydrol.2015.06.008.

[25] Wen, L., Miao, X., Wang, T., Wang, J., Yang, J., Liu, R., Ma, M., 2025. A novel multi-scenario mitigation model for rainstorm flood disasters. Int. J. Disaster Risk Reduct. 119, 105321. https://doi.org/10.1016/j.ijdrr.2025.105321.

[26] Wu, Z., Zhou, Y., Wang, H., Jiang, Z., 2020. Depth prediction of urban flood under different rainfall return periods based on deep learning and data warehouse. Sci. Total Environ. 716, 137077. https://doi.org/10.1016/j.scitotenv.2020.137077.

[27] Xu, H., Gao, J., Yu, X., Qin, Q., Du, S., Wen, J., 2024. Assessment of rainstorm waterlogging disaster risk in rapidly urbanizing areas based on land use scenario simulation: a case study of Jiangqiao Town in Shanghai, China. Land 13 (7), 1088. https://doi.org/10.3390/land13071088.

[28] Xu, K., Han, Z., Xu, H., Bin, L., 2023. Rapid prediction model for urban floods based on a light gradient boosting machine approach and hydrological-hydraulic model. Int. J. Disaster Risk Sci. 14, 79-97. https://doi.org/10.1007/s13753-023-00465-2.

[29] Yang, L., Ji, X., Li, M., Yang, P., Jiang, W., Chen, L., Yang, C., Sun, C., Li, Y., 2024. A comprehensive framework for assessing the spatial drivers of flood disasters using an Optimal Parameter-based Geographical Detector-machine learning coupled model. Geosci. Front. 15 (6), 101889. https://doi.org/10.1016/j.gsf.2024.101889.

[30] Ye, C., Xu, Z., Lei, X., Liao, W., Ding, X., Liang, Y., 2022. Assessment of urban flood risk based on data-driven models: a case study in Fuzhou City, China. Int. J. Disaster Risk Reduct. 82, 103318. https://doi.org/10.1016/j.ijdrr.2022.103318.

[31] Zhang, T., He, W., Zheng, H., Cui, Y., Song, H., Fu, S., 2021. Satellite-based ground PM (2.5) estimation using a gradient boosting decision tree. Chemosphere 268, 128801. https://doi.org/10.1016/j.chemosphere.2020.128801.

[32] Zhao, W., Zhang, Z., Khodadadi, N., Wang, L., 2025. A deep learning model coupled with metaheuristic optimization for urban rainfall prediction. J. Hydrol. 651, 132596. https://doi.org/10.1016/j.jhydrol.2024.132596.