River impoundments are commonly cited as key disturbances to freshwater aquatic ecosystems. Dams alter natural hydrological regimes, homogenize river system dynamics at a global scale, can act as barriers for migratory species and may facilitate species invasions. In this synthesis, we examined the short- and long-term effects of impoundment on fish population dynamics and community structure. At the population level, we tested the "trophic surge hypothesis", which predicts a hump-shaped response of fish abundance through time after impoundment. We tested the hypothesis on 40 recruitment time series and 125 adult abundance time series from 19 species and nine reservoirs distributed in temperate and boreal regions. At the community level, we compared diversity metrics (richness, evenness, diversity) on two datasets: 1) between reservoirs and reference ecosystems (lakes, rivers, and streams) and 2) over time (before and after impoundment and over time). At the population level, the trophic surge hypothesis was supported in more than 55% of the time series but we observed significant variation across species, reservoirs and regions. Fish recruitment increased substantially during reservoir filling and shortly after impoundment, and was usually followed by an increase in adult fish. The surge was transient and vanished after 3-4 years for recruits and after 10 years for adults. However, we are lacking long time series to conclude about population patterns in the trophic equilibrium phase. At the community level, we did not find any strong directional patterns in species diversity metrics when comparing reservoirs to reference lakes but found higher diversity and evenness in reservoirs and impounded streams/rivers relative to unimpounded streams/rivers. We did not find directional patterns when looking at a change over time. Variability in the reported diversity results across studies may be related to the ability to tease apart the unique effects of impoundment and water regulation from other stressors such as propagule pressure and eutrophication, as well as the comparability of the reference system. In conclusion, fish populations benefited quickly but transiently from impoundment, and longer time series are needed to conclude about population dynamics and equilibrium in aging reservoirs in order to develop management recommendations.