A decade of diversity and forest structure: Post-logging patterns across life stages in an Afrotropical forest

Abstract

Tropical forests are under threat of increasing pressure from income-generating land uses. Selective logging is a compromise that allows use of the land while leaving much of the forest canopy intact across a landscape. However, the ecological impacts of selective logging are unclear, with evidence of positive, negative, and negligible effects on forest structure and diversity. We examined the impact of selective logging on the structure and diversity of evergreen tropical forest in the Monts de Cristal region, a chain of mid-elevation hills in northwestern Gabon. For three size classes (seedling, sapling, and adult) of woody plant species, we tested whether forest structure (canopy openness, stem density, basal area, and relative liana abundances) and diversity were altered in forests that had been logged one year and ten years prior, compared to unlogged forest. In general, we found no large impact of selective logging treatment on the structure and diversity of adult woody plant communities, but the seedling and sapling communities were affected. Compared to unlogged forest, one-year post-logging forest had greater variation in canopy openness and lower sapling stem density. Ten-year post-logging forest had higher seedling and sapling species evenness, higher sapling species diversity, and higher relative abundance of sapling-sized lianas compared to unlogged forest. Our results show that key differences between intact and selectively logged forests persist in the understory at least a decade after logging. Overall, these results contribute an additional data point in the literature on selective logging, specifically representing the impacts of very low impact selective logging in Central African forests. Our study highlights the value of exploring selective logging impacts at multiple time periods of recovery, and makes an important contribution to the knowledge Central African managed forests.