As extreme weather events are becoming more frequent, the chronic poor, being overly exposed to these shocks, risk suffering the highest price. The 2012 flood in Nigeria was the worst in 40 years and hit more than 3 million people. Using nationally representative panel data from LSMS project, I study households’ asset dynamics over about a decade. I find that households hit by the flood converge to multiple equilibria consistent with the poverty trap narrative. In particular, households whose assets fell below the threshold converge to a low-level equilibrium point, whereas better endowed households converge to a high steady state. This is consistent across several empirical methods, ranging from parametric to non-parametric methods, as well as panel threshold estimation. Robustness checks further examine the validity of the finding, testing different asset indexes and flood definitions, as well as controlling for conflict-related events. Identifying a poverty trap is crucially helpful for designing poverty alleviation policies and fostering a country’s development.