The autonomous underwater vehicle (AUV) plays an essential role in scientific research and offshore exploration. Numerous risk issues exist during operation, however, which makes the AUV mission risky. The operating environment is one of the factors that seriously affects its safe operation. In long-range AUV missions, unavoidable changes in the operating environment impact the vehicle risk. Therefore, it is critical to determine the effect of the dynamic environment on AUV operation. The current work aims to provide a method for assessing the risk of AUV operation, considering the effect of changes in the operating environment. In this study, the Bayesian approach is applied, capturing various potential risk factors and their relationships. In order to take the influence of environmental changes on the AUV missions into account, the study applies a dynamic Bayesian network to incorporate online location information. Given the online information of the AUV´s operating environment, a dynamic risk value can then be analysed and determined. The proposed method provides operators with an overview of the environmental impact on AUV operation. It can work as a guide on how to choose the mission path before the operation, and it also provides the vehicle and operators with a dynamic risk indicator, which can support the AUV’s decision making.