Concept and method of drought disaster risk measurement based on drought propagation level--a case study of AnHui Province

Abstract

In view of the difficulties in quantification of drought and its potential impact on drought disaster, as well as the challenges of applying scenario simulation technology in regional drought risk assessment, we proposes a fuzzy identification model of drought disaster coupling improved Monte-Carlo model and Grey Wolf Optimization-Back Propagation (GWO-BP) neural network based on the disaster-forming process, aiming to seek a quantitative relationship between drought and drought disasters through numerous drought scenario simulations. Meanwhile, we also constructed a joint probability distribution model of drought and drought disasters, which can calculate the conditional probability of different grades of drought leading to different grades of drought disasters, then, we can calculate the drought disaster risk measurement(DDRM) based on the level of drought propagation, by combining the achievements of the above model and the determination of inhibitory or promoting mechanism in the disaster-forming process. Finally, we conducted an empirical study of the new method using Anhui Province as the research object. The results show that: (1) The improved Monte-Carlo model incorporates the Copula function and rank correlation coefficient, which can generate indicator sequences with correlation and maintain a similar correlation level to the original samples. (2) The fuzzy identification model of drought disaster can not only elucidate the propagation pattern of drought factors to drought disaster through numerous drought scenario simulations, but also effectively address the limitations of traditional drought risk research that heavily relies on long sequences of statistical data. (3) The DDRM can absolutely quantify the magnitude of regional inhibiting or promoting effects during the disaster formation process, which can make the assessment results comparable among the drought risk in different regions. The resultant findings offer crucial insights for pertinent decision-makers to alleviate the adverse effects of drought disaster.

Keywords

• Drought propagation mechanisms
• Correlation between indicators
• Monte-Carlo
• GWO-BP
• Copula function
• Risk measurement