Résumé : There are large gaps in the coverage of critical ecological processes related to the movement of individuals or genes (i.e. dispersal), which is critical for determining the spread and persistence of populations across space. In this dissertation we investigate understudied but important aspects of the dispersal process in mangroves, with as the main objective the reduction of parameter and model uncertainty. Models rarely incorporate realism and complexity at the level of emigration, transfer and immigration phases, hampering reliable predictions of dispersal patterns and long-term population dynamics under different climate change scenarios.

Mangrove ecosystems function at the edge of land and sea, often covering large intertidal areas along (sub)tropical coastal regions worldwide. Mangroves can live in these highly dynamic and demanding environmental conditions via a series of remarkable adaptations. They produce buoyant seeds and fruits (propagules) that disperse at the ocean surface (i.e. hydrochory - see cover image).

Despite their ecological and economical value, about 40 % of original mangroves have been lost worldwide during the last 50 years due to excessive exploitation and development. Deforestation, degradation and conversion to other land uses like intensive shrimp farming and agriculture have reduced and fragmented these ecosystems at an alarming rate. Climate change, probably most pronouncedly via changes in sea level, poses another important threat.