Maritime Autonomous Surface Ships (MASSs) constitute highly interconnected and tightly coupled multistate systems. Incorporating the coupling effects of both interactions and dependencies is centrally important to ensure navigation safety of MASSs. This paper proposes a framework for examining the coupling effects in the operational modes (OM) of MASSs. Failure Modes (FMs) of MASSs related to interactions with the environment and subsystems are identified using 24Model developed based on conventional ship accidents involving grounding in the west of Shenzhen port, China. FMs related to grounding are classified into the five risk types of human, organization, ship, environment, and technology, considering three OMs (states), i.e., manual control, remote control, and autonomous control. Based on this classification, the N-K model is applied to calculate the risk coupling values of scenarios involving multiple risk types. These scenarios are then ranked by the risk coupling values of the three OMs, and values are compared among different OMs. The results indicate that for grounding in the west of Shenzhen port most risk coupling values increase with increasing risk type diversity. Furthermore, the risk coupling mechanism in remote control differs significantly from that in manual control and autonomous control, thus risk management of MASS OMs should vary.