Disaster such as hurricane, by nature, involves uncertainties in many facets, from the time of its occurrence to magnitude of its impacts. Due to its highly erratic movement, uncertainty during a storm event can quickly cascade. Failure to incorporate these uncertainties when forming any emergency response operations can significantly affect the efficiency and effectiveness of the operations. Understanding that the storm hazards such as strong winds, torrential rain, and storm surge can inflict significant damage on the transport network affecting population’s ability to move during/after the storm event, we proposed a cascading network failure model to accentuate this mobility issue. The model takes the scenario-level storm impacts generated by the data-driven probabilistic scenarios model from our previous work as inputs to predict uncertainties in the land transport network states during the storm event. We tested the model on Hurricane Irma case study to determine the mobility states of the Tampa Bay network over the 72-hour time horizon. The proposed model serves as a mean to predict uncertainty in the mobility states over the course of a storm event – a critical factor in forming effective and efficient response operation models.