The UK Foresight Migration and Global Environmental Change Project recently concluded that migration is likely to be increasingly influenced by environmental change in the future through the impact of climate change on economic, social and political drivers. However, the range and complexity of the interactions between these drivers means that it will rarely be possible to distinguish individuals for whom environmental factors are the sole driver. In parallel, within debates about climate change adaptation there has been an emerging trend to position migration as an adaptation strategy to environmental shocks and stresses. However, migration is expensive, requiring a number of assets/capitals. This can result in some populations who experience the impacts of environmental change seeing a reduction in the very capital required to enable a move. Environmental change is therefore equally likely to make migration less possible as more probable. In the decades ahead, millions of people are thus likely to be unable to move away from locations in which they are extremely vulnerable to environmental change. To the international community, this ‘trapped’ population is likely to represent just as important a policy concern as those who do migrate. In determining the extent and location of these populations the non-ubiquity of migration within a community needs to be accounted for in order to separate those people who want to migrate but are unable to do so from those who are unwilling to migrate and instead choose to cope with livelihood stresses in alternative ways. This distinction requires a contextually driven analysis of the decision to migrate or stay. Such a focus on the migration decision is suited to an agent-based modelling approach which takes into account an individual’s intention to migrate, the influence of migration and non-migration behaviour of others and their perceived ability to migrate within a decision mediated by household and community-level factors.