IntroductionClimate migration research often relies on a binary distinction between rapid-onset and slow-onset disasters to predict migration timing. This framework may oversimplify the temporal dynamics linking disaster exposure to migration decisions.MethodsI analyze original survey data from 1,534 households across Guatemala, examining the effects of flood, hurricane, and drought exposure on internal and international migration plans two to seven years after exposure.ResultsFloods increase both internal migration plans (4.3 percentage points) and international migration plans (4.8 percentage points) two to seven years after exposure. Neither hurricanes nor droughts significantly affect migration plans during this timeframe, despite extensive research documenting their effects on other timelines. Migration networks prove critical: flood-exposed households with prior migration experience are fourteen times more likely to plan internal migration than those without.DiscussionBuilding from these findings, I develop the deferred climate-related migration theory, which posits that floods create a unique temporal window for migration by being severe enough to signal persistent risk while allowing sufficient recovery to accumulate resources. This produces a deferral period wherein individuals decide to migrate but take time to increase savings, activate social networks, and coordinate plans. The theory challenges assumptions that onset speed determines migration timing and demonstrates the need for disaster-specific approaches in climate adaptation policy.