The Caatinga, the largest region of seasonally dry tropical forest in the Neotropics, suffers high rates of deforestation and habitat degradation, mostly due to wood extraction. As an alternative to illegal logging, governments have looked at more sustainable management schemes, allowing natural regeneration after logging through relatively long (~25-year) harvest rotations. We investigated the impacts of forest management at a 1,700-ha privately owned area located at the Araripe Plateau, in the semiarid interior of northeastern Brazil, focusing on the population parameters of 8 avian species. The property was subdivided into 22 forest stands, where a different stand has been logged every year since 2004. By 2016, when we sampled the avifauna, 12 forest stands had already been logged and allowed to regrow, creating a landscape of logged and unlogged stands and a 12-year forest recovery chrono-sequence. We conducted distance-based sampling along transects to estimate the density and abundance of these species in logged and unlogged stands. We found that logging impacted 3 of the target species. Two of them (Megaxenops parnaguae and Synallaxis scutata) were less abundant, whereas another (Sclerurus cearensis) disappeared altogether from logged areas. We also found a positive correlation between the normalized difference vegetation index (NDVI) and the avian density of the affected species. However, we failed to observe any significant relationship between forest recovery or NDVI and species densities, suggesting that even after 12 years of forest recovery, species abundance remains lower in logged than in unlogged areas. We found that logging impacted birds in a species-specific manner, with 5 species unaffected and 3 species declining. Although our sampling occurred half-way through the regeneration cycle, we found no evidence of recovery for those species most affected. We suggest keeping unlogged areas intermingled with logged stands, allowing the survival and potential recovery of species in regenerating forests.