Abstract: Debris flow hazard assessment serves as an important foundation for effective disaster prevention and land-use planning. Current research on debris flow hazard assessment has primarily focused on statistical analysis at large regional scales, while studies addressing potential hazard zones and intensity in complex watersheds remain relatively limited. A key challenge lies in effectively analyzing the uncertainty associated with debris flow initiation and disaster formation under multi-source initiation conditions. To address this issue, this study integrates physical-mechanical mechanisms of source initiation, Bayesian theory, and dynamic process numerical modeling to propose a comprehensive method for assessing debris flow hazard and intensity under multi-source initiation conditions. The approach includes: conducting stability analysis of upstream source materials using the limit equilibrium method; applying Monte Carlo simulation and frequency analysis to determine the probability of source instability triggering debris flows; and employing Bayesian analysis to evaluate the probability of debris flow occurrence under multiple initiation scenarios. Subsequently, a numerical model based on shallow water equations was used to simulate the dynamic processes of high-probability debris flow scenarios. Finally, hazard assessment and zoning were performed through deposition and impact intensity evaluation. To validate and illustrate the proposed methodology, the 2010 Yingxiu Hongchun Gully debris flow event was analyzed as a case study, including source area instability initiation analysis and hazard assessment. The results provide debris flow hazard classification and zoning under multi-source initiation conditions, offering a reference framework for debris flow disaster prevention and mitigation planning.