Maritime transportation is central to the global economy, and analyzing its large-scale behavioral data is critical for operational planning, environmental stewardship, and governance. This work presents a spatio-temporal analytical framework based on discrete-time Markov chains to model vessel movement patterns in the Gulf of St. Lawrence, with particular emphasis on disruptions induced by the COVID-19 pandemic. We discretize the maritime domain into hexagonal cells and construct mobility signatures for distinct vessel types using cell transition frequencies and dwell times. These features are used to build origin-destination matrices and spatial transition probability models that characterize maritime dynamics across multiple temporal resolutions. Focusing on commercial, fishing, and passenger vessels, we analyze the temporal evolution of mobility behaviors during the pandemic, highlighting significant yet transient disruptions to recurring transport patterns. The methodology we contribute to this paper allows for an extensive behavioral analytics key for transportation planning. Accordingly, our findings reveal vessel-specific mobility signatures that persist across spatially disjoint regions, suggesting behaviors invariant to time. In contrast, we observe temporal deviations among passenger and fishing vessels during the pandemic, reflecting the influence of social isolation measures and operational constraints on non-essential maritime transport in this region.

Routing vessels through narrow and dynamic waterways is challenging due to changing environmental conditions and operational constraints. Existing vessel-routing studies typically fail to generalize across multiple origin-destination pairs and do not exploit large-scale, data-driven traffic graphs. In this paper, we propose a reinforcement learning solution for big maritime data that can learn to find a route across multiple origin-destination pairs while adapting to different hexagonal grid resolutions. Agents learn to select direction and speed under continuous observations in a multi-discrete action space. A reward function balances fuel efficiency, travel time, wind resistance, and route diversity, using an Automatic Identification System (AIS)-derived traffic graph with ERA5 wind fields. The approach is demonstrated in the Gulf of St. Lawrence, one of the largest estuaries in the world. We evaluate configurations that combine Proximal Policy Optimization with recurrent networks, invalid-action masking, and exploration strategies. Our experiments demonstrate that action masking yields a clear improvement in policy performance and that supplementing penalty-only feedback with positive shaping rewards produces additional gains.