Citation

Deng M, Li Z, Tao F. Int. J. Environ. Res. Public Health 2022; 19(15): e9497.

Copyright

(Copyright © 2022, MDPI: Multidisciplinary Digital Publishing Institute)

DOI

10.3390/ijerph19159497

PMID

35954847

Abstract

Rainstorm disasters have had a serious impact on the sustainable development of society and the economy. However, due to the complexity of rainstorm disasters, it is difficult to measure the importance of each indicator. In this paper, the rainstorm disaster risk assessment framework was systematically proposed based on the disaster system theory and a system of corresponding indicators was established. Furthermore, the genetic algorithm optimized projection pursuit and XGBoost were coupled to assess the rainstorm disaster risk and to measure the relative importance of each indicator. Finally, the Yangtze River Delta was taken as the case study area. The results show that: the rainstorm disaster risk in the eastern and southeast is higher than those in the central and northwest of the Yangtze River Delta; the total precipitation from June to September and the top ten indicators contribute 9.34% and 74.20% to the rainstorm disaster risk assessment results, respectively. The results can provide references for decision makers and are helpful for the formulation of rainstorm adaptation strategies.

Language: en

Keywords

risk assessment; projection pursuit; rainstorm disaster; XGBoost; Yangtze River Delta