A comprehensive framework for assessing the spatial drivers of flood disasters using an Optimal Parameter-based Geographical Detector–machine learning coupled model

Abstract

Flood disasters pose serious threats to human life and property worldwide. Exploring the spatial drivers of flood disasters on a macroscopic scale is of great significance for mitigating their impacts. This study proposes a comprehensive framework for integrating driving-factor optimization and interpretability, while considering spatial heterogeneity. In this framework, the Optimal Parameter-based Geographic Detector (OPGD), Recursive Feature Estimation (RFE), and Light Gradient Boosting Machine (LGBM) models were utilized to construct the OPGD–RFE–LGBM coupled model to identify the essential driving factors and simulate the spatial distribution of flood disasters. The SHapley Additive ExPlanation (SHAP) interpreter was employed to quantitatively explain the driving mechanisms behind the spatial distribution of flood disasters. Yunnan Province, a typical mountainous and plateau area in Southwest China, was selected to implement the proposed framework and conduct a case study. For this purpose, a flood disaster inventory of 7332 historical events was prepared, and 22 potential driving factors related to precipitation, surface environment, and human activity were initially selected. Results revealed that flood disasters in Yunnan Province exhibit high spatial heterogeneity, with geomorphic zoning accounting for 66.1% of the spatial variation in historical flood disasters. The OPGD–RFE–LGBM coupled model offers clear advantages over a single LGBM in identifying essential driving factors and quantitatively analyzing their impacts. Moreover, the simulation performance shows a slight improvement (a 6% average decrease in RMSE and an average increase of 1% in R²) even with reduced factor data. Factor explanatory analysis indicated that the combination of the essential driving factor sets varied across different subregions; nevertheless, precipitation-related factors, such as precipitation intensity index (SDII), wet days (R10MM), and 5-day maximum precipitation (RX5day), were the main driving factors controlling flood disasters. This study provides a quantitative analytical framework for the spatial drivers of flood disasters at large scales with significant heterogeneity, offering a reference for disaster management authorities in developing macro-strategies for disaster prevention.

Keywords

Flood disaster / Spatial driving factors / Spatial heterogeneity / Machine learning / Optimal Parameter-based Geographical Detector / Yunnan Province

CRediT authorship contribution statement

Luyi Yang: Formal analysis, Methodology, Writing – original draft. Xuan Ji: Conceptualization, Funding acquisition, Writing – review & editing. Meng Li: Data curation. Pengwu Yang: Data curation. Wei Jiang: Writing – review & editing. Linyan Chen: Writing – review & editing. Chuanjian Yang: Writing – review & editing. Cezong Sun: Writing – review & editing. Yungang Li: Conceptualization, 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.

Acknowledgments

This work was supported by the National Key Research and Development Program of China (Grant No. 2022YFF1302405), the Yunnan Province Key Research and Development Program (Grant No. 202203AC100005), the National Natural Science Foundation of China (Grant No. 42061005, 42067033), and Applied Basic Research Programs of Yunnan Province (Grant No. 202101AT070110, 202001BB050073).

References

A. Ahani, S.S.M. Nadoushani, A. Moridi. Regionalization of watersheds by finite mixture models. J. Hydrol., 583 (2020), Article 124620

H.E. Aydin, M.C. Iban. Predicting and analyzing flood susceptibility using boosting-based ensemble machine learning algorithms with SHapley Additive exPlanations. Nat. Hazards, 116 (2023), pp. 2957-2991

J. Bac-Bronowicz, P. Grzempowski. Regionalization of geographical space according to selected topographic factors in reference to spatial distribution of precipitation: application of artificial neural networks in GIS. Environ. Earth Sci., 77 (18) (2018), p. 631

K. Bhattacharjee, B. Behera. Does forest cover help prevent flood damage? Empirical evidence from India. Glob. Environ. Chang., 53 (2018), pp. 78-89

T. Cai, X.Y. Li, X. Ding, J. Wang, J. Zhan. Flood risk assessment based on hydrodynamic model and fuzzy comprehensive evaluation with GIS technique. Int. J. Disaster Risk Reduct., 35 (2019), Article 101077

Y.F. Cao, H.L. Jia, J.N. Xiong, W.M. Cheng, K. Li, Q. Pang, Z.W. Yong. Flash flood susceptibility assessment based on geodetector, certainty factor, and logistic regression analyses in Fujian province. China. ISPRS Int. J. Geo-Inf., 9 (12) (2020), p. 748

J.L. Chen, G.R. Huang, W.J. Chen. Towards better flood risk management: Assessing flood risk and investigating the potential mechanism based on machine learning models. J. Environ. Manage., 293 (2021), Article 112810

J.F. Chen, Q. Li, H.M. Wang, M.H. Deng. A machine learning ensemble approach based on random forest and radial basis function neural network for risk evaluation of regional flood disaster: a case study of the Yangtze River Delta, China. Int. J. Environ. Res. Public Health, 17 (1) (2020), p. 49

Y. Chen, J.L. Wang. Ecological security early-warning in central Yunnan Province, China, based on the gray model. Ecol. Indic., 111 (2020), Article 106000

Z.J. Cui, X.X. Qing, H.X. Chai, S.X. Yang, Y. Zhu, F.F. Wang. Real-time rainfall-runoff prediction using light gradient boosting machine coupled with singular spectrum analysis. J. Hydrol., 603 (2021), Article 127124

Q.W. Duan, M.H. Tan. Using a geographical detector to identify the key factors that influence urban forest spatial differences within China. Urban for. Urban Green., 49 (2020), Article 126623

Z.X. Fan, A. Bräuning, A. Thomas, J.B. Li, K.F. Cao. Spatial and temporal temperature trends on the Yunnan Plateau (Southwest China) during 1961–2004. Int. J. Climatol., 31 (14) (2011), pp. 2078-2090

H. Fan, J.M. Hu, D.M. He. Trends in precipitation over the low latitude highlands of Yunnan. China. J. Geogr. Sci., 23 (6) (2013), pp. 1107-1122

B.J. Fu, G.H. Liu, L.D. Chen, J.R. Li. Schee of ecological regionalization in China. Acta Ecol. Sin., 21 (1) (2001), pp. 1-6

I. Guyon, J. Weston, S. Barnhill, V. Vapnik. Gene selection for cancer classification using support vector machines. Mach. Learn., 46 (2002), pp. 389-422

M. Haklay, P. Weber. Openstreetmap: User-generated street maps. IEEE Pervasive Comput., 7 (4) (2008), pp. 12-18

B.S. He, X.L. Huang, M.H. Ma, Q.R. Chang, Y. Tu, Q. Li, K. Zhang, Y. Hong. Analysis of flash flood disaster characteristics in China from 2011 to 2015. Nat. Hazards, 90 (1) (2018), pp. 407-420

S. Hochrainer-Stigler, F. Laurien, S. Velev, A. Keating, R. Mechler. Standardized disaster and climate resilience grading: A global scale empirical analysis of community flood resilience. J. Environ. Manage., 276 (2020), Article 111332

H.Y. Hong, P. Tsangaratos, I. Ilia, J.Z. Liu, A.X. Zhu, W. Chen. Application of fuzzy weight of evidence and data mining techniques in construction of flood susceptibility map of Poyang County. China. Sci. Total Environ., 625 (2018), pp. 575-588

F.S. Hosseini, B. Choubin, A. Mosavi, N. Nabipour, S. Shamshirband, H. Darabi, A.T. Haghighi. Flash-flood hazard assessment using ensembles and Bayesian-based machine learning models: Application of the simulated annealing feature selection method. Sci. Total Environ., 711 (2020), Article 135161

Y.H. Huang, D. Jiang, J.Y. Fu. 1 km grid GDP dataset of China (2005, 2010). Global Change Research Data Publishing and Repository, v1 (2014),

M. Kaiser, S. Günnemann, M. Disse. Regional-scale prediction of pluvial and flash flood susceptible areas using tree-based classifiers. J. Hydrol., 612 (2022), Article 128088

Ke, G.L., Meng, Q., Finley, T., Wang, T.F., Chen, W., Ma, W.D., Ye, Q.W., Liu, T.Y., 2017. Lightgbm: A highly efficient gradient boosting decision tree, in: Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, CA, 3149–3157.

K. Khosravi, H. Shahabi, B.T. Pham, J. Adamowski, A. Shirzadi, B. Pradhan, J. Dou, H.B. Ly, G. Grof, H.L. Ho, H. Hong, K. Chapi, I. Prakash. A comparative assessment of flood susceptibility modeling using multi-criteria decision-making analysis and machine learning methods. J. Hydrol., 573 (2019), pp. 311-323

Y.G. Li, D.M. He, J.M. Hu, J. Cao. Variability of extreme precipitation over Yunnan Province, China 1960–2012. Int. J. Climatol., 35 (2) (2015), pp. 245-258

D. Liu, Z.R. Fan, Q. Fu, M. Li, M.A. Faiz, S. Ali, T.X. Li, L.L. Zhang, M.I. Khan. Random forest regression evaluation model of regional flood disaster resilience based on the whale optimization algorithm. J. Clean Prod., 250 (2020), Article 119468

Y.S. Liu, Y.H. Huang. Why Flash Floods Occur Differently across Regions? A Spatial Analysis of China. Water, 12 (12) (2020), p. 3344

L. Liu, Z.X. Xu. Regionalization of precipitation and the spatiotemporal distribution of extreme precipitation in southwestern China. Nat. Hazards, 80 (2) (2016), pp. 1195-1211

Y.S. Liu, X.M. Yuan, L. Guo, Y.H. Huang, X.L. Zhang. Driving force analysis of the temporal and spatial distribution of flash floods in Sichuan Province. Sustainability, 9 (9) (2017), p. 1527

Y.S. Liu, Z.S. Yang, Y.H. Huang, C.J. Liu. Spatiotemporal evolution and driving factors of China’s flash flood disasters since 1949. Sci. China Earth Sci., 61 (12) (2018), pp. 1804-1817

Lundberg, S.M. and Lee, S.I., 2017. A unified approach to interpreting model predictions, in: Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, CA, 4768–4777.

Lundberg, S.M., Erion, G.G., Lee, S.I., 2018. Consistent individualized feature attribution for tree ensembles. Preprint at https://arXiv preprint arXiv:1802.03888.

P. Luo, Y.Z. Song, P. Wu. Spatial disparities in trade-offs: economic and environmental impacts of road infrastructure on continental level. Gisci. Remote Sens., 58 (5) (2021), pp. 756-775

Z.Y. Luo, J. Tian, J. Zeng, F. Pilla. Resilient landscape pattern for reducing coastal flood susceptibility. Sci. Total Environ., 856 (2023), Article 159087

D.J. Marceau. The scale issue in the social and natural sciences. Can. J. Remote Sens., 25 (4) (1999), pp. 347-356

B. Merz, H. Kreibich, U. Lall. Multi-variate flood damage assessment: a tree-based data-mining approach. Nat. Hazards Earth Syst. Sci., 13 (1) (2013), pp. 53-64

H. Mojaddadi, B. Pradhan, H. Nampak, N. Ahmad, A.H. Ghazali. Ensemble machine-learning-based geospatial approach for flood risk assessment using multi-sensor remote-sensing data and GIS. Geomat. Nat. Hazards Risk, 8 (2) (2017), pp. 1080-1102

A. Mosavi, P. Ozturk, K.W. Chau. Flood prediction using machine learning models: Literature review. Water, 10 (11) (2018), p. 1536

F.N. Nkeki, E.I. Bello, I.G. Agbaje. Flood risk mapping and urban infrastructural susceptibility assessment using a GIS and analytic hierarchical raster fusion approach in the Ona River Basin. Nigeria. Int. J. Disaster Risk Reduct., 77 (2022), Article 103097

M. Panahi, E. Dodangeh, F. Rezaie, K. Khosravi, H.V. Le, M.J. Lee, S. Lee, B.T. Pham. Flood spatial prediction modeling using a hybrid of meta-optimization and support vector regression modeling. Catena, 199 (2021), Article 105114

Y. Qiao, X.Q. Wang, Z.X. Han, M. Tian, Q. Wang, H. Wu, F.T. Liu. Geodetector based identification of influencing factors on spatial distribution patterns of heavy metals in soil: A case in the upper reaches of the Yangtze River. China. Appl. Geochem., 146 (2022), Article 105459

O. Rahmati, H.R. Pourghasemi, H. Zeinivand. Flood susceptibility mapping using frequency ratio and weights-of-evidence models in the Golastan Province. Iran. Geocarto Int., 31 (1) (2016), pp. 42-70

M. Saber, T. Boulmaiz, M. Guermoui, K.I. Abdrado, S.A. Kantoush, T. Sumi, H. Boutaghane, D. Nohara, E. Mabrouk. Examining LightGBM and CatBoost models for wadi flash flood susceptibility prediction. Geocarto Int., 37 (25) (2021), pp. 7462-7487

R. Sayre, J. Dangermond, C. Frye, R. Vaughan, P. Aniello, S. Breyer, D. Cribbs, D. Hopkins, R. Naumann, B. Derrenbacher, D. Wright, C. Brown, K. Butler, L. Bennett, J. Smith, L. Benson, D. Sistine, H. Warner, J. Cress, A. Grosse. A new map of global ecological land units—an ecophysiographic stratification approach. Association of American Geographers, Washington, DC (2014), p. 46

Y.Z. Song, J.F. Wang, Y. Ge, C.D. Xu. An optimal parameters-based geographical detector model enhances geographic characteristics of explanatory variables for spatial heterogeneity analysis: Cases with different types of spatial data. Gisci. Remote Sens., 57 (5) (2020), pp. 593-610

S. Stefanidis, D. Stathis. Assessment of flood hazard based on natural and anthropogenic factors using analytic hierarchy process (AHP). Nat. Hazards, 68 (2) (2013), pp. 569-585

H.Z. Sun, J.Y. Wang, J.N. Xiong, J.H. Bian, H.A. Jin, W.M. Cheng, A.N. Li. Vegetation change and its response to climate change in Yunnan Province. China. Adv. Meteorol., 2021 (2021), p. 8857589

T. Tanaka, K. Kiyohara, Y. Tachikawa. Comparison of fluvial and pluvial flood risk curves in urban cities derived from a large ensemble climate simulation dataset: A case study in Nagoya. Japan. J. Hydrol., 584 (2020), Article 124706

A.J. Tatem. WorldPop, open data for spatial demography. Sci. Data, 4 (2017), pp. 1-4

M.S. Tehrany, B. Pradhan, M.N. Jebur. Flood susceptibility mapping using a novel ensemble weights-of-evidence and support vector machine models in GIS. J. Hydrol., 512 (2014), pp. 332-343

M.S. Tehrany, S. Jones, F. Shabani. Identifying the essential flood conditioning factors for flood prone area mapping using machine learning techniques. Catena, 175 (2019), pp. 174-192

B. Tellman, J.A. Sullivan, C. Kuhn, A.J. Kettner, C.S. Doyle, G.R. Brakenridge, T.A. Erickson, D.A. Slayback. Satellite imaging reveals increased proportion of population exposed to floods. Nature, 596 (7870) (2021), pp. 80-86

UNDRR, CRED, 2020. Human Cost of Disasters: An Overview of the last 20 years: 2000–2019. CRED, UNDRR, Geneva, 30 pp.

K. Vogel, L. Weise, K. Schröter, A.H. Thieken. Identifying driving factors in flood damaging processes using graphical models. Water Resour. Res., 54 (11) (2018), pp. 8864-8889

N. Wang, W.M. Cheng, B.X. Wang, Q.Y. Liu, C.H. Zhou. Geomorphological regionalization theory system and division methodology of China. J. Geogr. Sci., 30 (2) (2020), pp. 212-232

J.F. Wang, X.H. Li, G. Christakos, Y.L. Liao, T. Zhang, X. Gu, X.Y. Zheng. Geographical detectors-based health risk assessment and its application in the neural tube defects study of the Heshun Region. China. Int. J. Geogr. Inf. Sci., 24 (1) (2010), pp. 107-127

S. Xiao, L. Zou, J. Xia, Y. Dong, Z.Z. Yang, T.C. Yao. Assessment of the urban waterlogging resilience and identification of its driving factors: a case study of Wuhan City. China. Sci. Total Environ., 866 (2023), Article 161321

J.N. Xiong, C.C. Ye, W.M. Cheng, L. Guo, C.H. Zhou, X.L. Zhang. The spatiotemporal distribution of flash floods and analysis of partition driving forces in Yunnan Province. Sustainability, 11 (10) (2019), p. 2926

J.N. Xiong, Q. Pang, C.K. Fan, W.M. Cheng, C.C. Ye, Y.L. Zhao, Y.R. He, Y.F. Cao. Spatiotemporal characteristics and driving force analysis of flash floods in Fujian province. ISPRS Int. J. Geo-Inf., 9 (2) (2020), p. 133

Xu, Z.X., Yang, X.J., Zuo, D.P, Chu, Q., Liu, W.F., 2015. Spatiotemporal characteristics of extreme precipitation and temperature: a case study in Yunnan Province, China, in: Proceedings of the International Association of Hydrological Sciences, Prague, Czech Republic, 121-127.

K. Xu, Z.T. Han, H.S. Xu, L.L. Bin. Rapid Prediction Model for Urban Floods Based on a Light Gradient Boosting Machine Approach and Hydrological-Hydraulic Model. Int. J. Disaster Risk Sci., 14 (1) (2023), pp. 79-97

Q.L. Xu, Q. Wang, J. Liu, H. Liang. Simulation of land-use changes using the partitioned ANN-CA model and considering the influence of land-use change frequency. ISPRS Int. J. Geo-Inf., 10 (5) (2021), p. 346

J.T. Yang, C. Song, Y. Yang, C.D. Xu, F. Guo, L. Xie. New method for landslide susceptibility mapping supported by spatial logistic regression and GeoDetector: A case study of Duwen Highway Basin, Sichuan Province, China. Geomorphology, 324 (2019), pp. 62-71

Y.H. Yu, J.L. Wang, F. Cheng, H. Deng, S. Chen. Drought monitoring in Yunnan Province based on a TRMM precipitation product. Nat. Hazards, 104 (3) (2020), pp. 2369-2387

X.F. Yuan, J.C. Han, Y.J. Shao, Y.H. Li, Y.S. Wang. Geodetection analysis of the driving forces and mechanisms of erosion in the hilly-gully region of northern Shaanxi Province. J. Geogr. Sci., 29 (5) (2019), pp. 779-790

D.Y. Zhang, Y.C. Gong. The comparison of LightGBM and XGBoost coupling factor analysis and prediagnosis of acute liver failure. IEEE Access, 8 (2020), pp. 220990-221003

X. Zhang, F. Yang. RClimDex (1.0) user manual. Climate Research Branch Environment Canada, Ontario (2004), p. 22

Y.H. Zhao, L. Liu, S.Z. Kang, Y. Ao, L. Han, C.Q. Ma. Quantitative analysis of factors influencing spatial distribution of soil erosion based on geo-detector model under diverse geomorphological types. Land, 10 (6) (2021), p. 604

G. Zhao, B. Pang, Z.X. Xu, J.J. Yue, T.B. Tu. Mapping flood susceptibility in mountainous areas on a national scale in China. Sci. Total Environ., 615 (2018), pp. 1133-1142

J.Y. Zheng, Y.H. Yin, B.Y. Li. A new scheme for climate regionalization in China. Acta Geographica Sinica, 65 (1) (2010), pp. 3-12

L. Zhou, F.N. Hu, B. Wang, C.Z. Wei, D.Q. Sun, S.H. Wang. Relationship between urban landscape structure and land surface temperature: Spatial hierarchy and interaction effects. Sust. Cities Soc., 80 (2022), Article 103795

X.Z. Zhou, H.J. Wen, Y.L. Zhang, J.H. Xu, W.G. Zhang. Landslide susceptibility mapping using hybrid random forest with GeoDetector and RFE for factor optimization. Geosci. Front., 12 (5) (2021), Article 101211