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-yr) harvest rotations. We investigated the impacts of forest management at a 1,700-ha privately-owned area located at the Araripe Plateau, in the semi-arid interior of northeastern Brazil, focusing on the population parameters of eight 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-yr 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 three 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 NDVI and 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-yr 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 five species unaffected and three species declining. Although our sampling occurred halfway 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.     

These data were collected from acoustic bird census using the distance sampling method. This method was applied to 11 trails that divide a dry forest area in the Brazilian Caatinga under a forest management regime. In the field, bird individuals were monitored across the transects and the following data were collected: species, type of detection, distance and angle of the transect, the total length of the transect, the numbers of the forest stands on each side of the transect, the logging status of each forest stand and the extent of the logged and unlogged area for forest management. All data were obtained in both the wet and dry seasons of 2016.

Readme file is included with the dataset.