Invasive species pose a serious risk to the biological, cultural and economic value of marine ecosystems. In an increasingly connected world, maritime traffic has become a key driver of this risk, with vessels transporting species between domestic and international locations. Network analysis is a useful technique for characterizing connectivity in transport systems, and subsequent risk of invasion through these connections. Network analyses are commonly based on static snapshots of a system, however this approach can miss important temporal patterns in systems where connections fluctuate over time, such as marine traffic. Using New Zealand as a case study, we developed a dynamic network model of New Zealand’s commercial maritime transport system. Using traffic volume and vessel characteristics, we quantified seasonal incursion risk at ports and marine protected areas. Using connectivity metrics, we identified which sites, vessel types and times of year would give the greatest overall risk reduction if prioritised for management. Our results can be used to direct the location and timing of biosecurity monitoring and intervention in New Zealand, and this approach is easily adaptable to other locations and scales.