【摘要】 开展山洪灾害风险评估及其驱动力分析，对于洪灾风险防范与应对具有现实意义。以洪灾频发的山西省为例，利用层次分析方法有效识别了洪灾风险等级及其空间分异性特征，并利用全局和局部莫兰指数探究了其空间聚集(离散)位置及其分布规律；利用地理探测器技术探明了洪灾风险的驱动机理。结果表明：年最大24 h降水量、多年平均降水量和NDVI是影响洪灾分异性的重要因子，其解释力q值分别高达0.57、0.52和0.24;多风险双因子交互作用对洪灾风险影响显著，且呈增强或非线性增强趋势，尤其是降水和坡度因子交互作用，其q值高达0.71。更多还原

【Abstract】 It is of practical significance for flood risk prevention and response to carry out mountain flood risk assessment and its driving force analysis. Therefore, this paper took Shanxi Province with frequent floods as an example, used the analytic hierarchy process to effectively identify the flood risk level and its spatial differentiation characteristics, and used the global and local Moran’s indexes to explore its spatial aggregation(discrete) location and distribution law. Finally, the driving mechanism behind the flood risk was explored by using the geographic detector technology. The results show that the annual maximum 24-hour rainfall, multi-year average rainfall and NDVI factors are important factors causing flood differentiation, and their explanatory power q values are as high as 0.57, 0.52 and 0.24, respectively. The multi-risk two-factor interaction has a significant impact on flood risk, and it is enhanced or non-linearly enhanced, especially the interaction between rainfall and slope factors, with a q value of 0.71.更多还原