Invasive alien plants (IAPs) are a great challenge for biodiversity conservation and management. Multi-temporal landscape analysis has a great potential for describing plant invasion, however conservation solutions accounting of landscape dynamics are still limited. This research aims to explore the spatial-temporal pattern of the IAP Carpobrotus spp. in relation with landscape changes on Mediterranean coastal dunes. Based on detailed multi-temporal Carpobrotus spp. and local land cover maps, we described coastal dune landscape changes of invaded areas using transition matrices. We then calculated a set of non-redundant class and landscape pattern metrics for the compared data maps and explored invasion trends through trajectory analysis. We analyzed the relationship between the spatial pattern of alien invasion over time (e.g., expansion or reduction) and landscape pattern trends by examining the temporal delta values of Carpobrotus spp. pattern indices in relation to the delta values of landscape pattern metrics, using a machine learning technique followed by Partial Dependence analysis. The spatial-temporal characteristics of Carpobrotus spp. patches and the contextual landscape varied across coastal tracts experiencing IAP expansion and reduction. Carpobrotus spp. invasion expanded with more numerous, bigger and more connected patches in tracts with stable seashore and increasing artificial surfaces. In coastal tracts affected by coastal erosion, Carpobrotus spp declined, /remaining in few, small and regularly shaped patches. Multi-temporal landscape analysis provides a sound framework for understanding invasion dynamics across landscapes influenced by a variety of factors such as seashore dynamics and urban sprawl. This approach offers valuable insights for developing tailored management strategies that account for specific contextual nuances and enables informed planning of recovery actions.