山丘区小流域山洪灾害频发，现有研究多集中于区域尺度或常规情景，对村落尺度、桥涵堵塞与下游顶托等复合工况下的山洪风险量化仍显不足。以日照市北陈家沟村为研究对象，依据降雨—洪水冲淹—灾害发生的链式机制，构建了一套基于情景分析的山洪灾害风险综合评估框架。首先设定不同重现期（5 a、10 a、20 a、50 a、100 a）、不同工况（桥涵堵塞50%、下游顶托）等复合情景，并运用Copula函数量化情景发生概率；然后构建水文-水动力耦合及损失计算模型，模拟不同复合情景下的洪水淹没过程及损失情况；在此基础上计算损失期望对风险进行度量。结果表明：在不同重现期条件下，洪水造成的直接经济损失呈阶梯式增长。在一般情景下，5 a、10 a、20 a、50 a、100 a一遇洪水直接经济损失分别为17.81万、89.53万、347.90万、449.84万、998.83万元；复合情景则显著加剧了灾害损失，在100 a一遇情景下，桥涵堵塞、下游顶托及两者叠加复合情景的直接经济损失分别达1 026.29万、1 024.01万与1 129.23万元。山洪灾害期望损失为50.16万元。研究成果可为山丘区沿河村落的洪涝灾害防控与风险管理提供科学依据。

[Objective] Flash floods in hilly small watersheds occur frequently， posing severe threats to the lives and property of residents in riverside villages. Existing research lacks sufficient understanding of disaster mechanisms at the village scale and under compound conditions， making it difficult to accurately reflect real risks. This study develops a comprehensive risk assessment framework based on multiple scenarios to provide scientific support for the governance planning and risk management of riverside villages. [Methods] Based on the disaster chain from rainfall to flood inundation and to disaster formation， we constructed a complete risk assessment framework. First， we collected the basic geographic， hydrometeorological， and socioeconomic data of the study area， and designed different compound scenarios considering field conditions， and then calculated the joint occurrence probabilities of these scenarios using the Copula function. Subsequently， a coupled hydrological hydrodynamic model was built to simulate the flood evolution processes under various scenarios. On this basis， a loss calculation model for affected entities was applied to estimate direct economic losses. Finally， risk probabilities and loss values were integrated to derive a risk index that comprehensively reflects the long-term risk level of the region. [Results] The inundation area expanded significantly with increasing return periods， from 37 980.81 m² under a 5-year return period to 272 788.74 m² under a 100-year return period， with particularly noticeable growth between 20- and 100-year return periods. Compound scenarios markedly aggravated disaster losses. Under a 100-year rainfall condition， compared to the scenario without bridge blockage or downstream backwater （direct economic loss of 9.988 3 million CNY）， the direct economic losses under the scenarios of 50% bridge blockage， downstream backwater， and both combined increased to 10.262 9 million， 10.240 1 million， and 11.292 3 million CNY， respectively. The loss under the combined scenario was notably higher than that under any single-factor scenario， highlighting the amplification effect of compound disaster factors. Additionally， the Copula function indicated that bridge blockage and downstream backwater exhibit significant joint occurrence probabilities under the same return-period rainfall， confirming the objective basis for the compound disaster scenarios involving dual conditions. The expected loss from flash floods in Beichenjiagou Village， Rizhao City， was calculated as 501 600 CNY using the risk degree model. [Conclusion] This study successfully constructed and applied a flash flood risk assessment framework integrating compound scenario analysis incorporating the entire disaster chain process. It comprehensively reflects the entire process of flash flood formation， disaster triggering， and loss generation in small watersheds. By introducing compound scenario analysis， the study enhances the characterization of chain disaster mechanisms， quantifies the amplification effect of bridge blockage and downstream backwater on disaster losses， and underscores the critical role of key hydraulic engineering nodes in risk prevention and control. Furthermore， the study quantifies disaster risk as the expected loss value of the study area， offering a quantitative basis for determining economic inputs in disaster prevention and mitigation， as well as for formulating insurance and risk management strategies.