The last century has seen unprecedented rates of human-mediated disturbances in natural ecosystems. Nevertheless, in the last decades, large-scale forest regeneration and active restoration increased native forest cover in many regions of the globe, supporting important ecosystem services such as carbon sequestration and soil protection. In this context, our study identified the human legacies in the provision of ecosystem services related to carbon stocking and soil protection of actively restored forests compared to naturally regenerating forests. Our study was carried out in 7-15 years-old highly fragmented seasonal semideciduous tropical forests in the Atlantic Forest hotspot in southeast Brazil. We installed 900 m² plots in each of three forests types: i) forests under active restoration through seedling planting (AR), and second-growth forests on ii) abandoned pastures (Pa) and iii) abandoned Eucalyptus spp. plantings (Eu). In order to quantify above-ground biomass, we sampled and identified all trees and shrubs DBH≥5 cm. We quantified native understory abundance by counting 1≤DBH<5 individuals in a 120 m² subplot. Soil protection by forests was quantified indirectly by estimating soil cover in seven 1 x 1 m subplots in each plot.
Results/Conclusions
Biomass was higher in AR and Eu than in Pa (AR: 125.4 ± 57.3; Pa: 65.5 ± 40.0; Eu: 107.7 ± 23.4 Mg ha-1). When considering only native species, biomass was higher in AR than in Eu and Pa (AR: 112.2 ± 44.6; Pa: 58.1 ± 27.0; Eu: 34.7 ± 18.8 Mg ha-1). Understory abundance in AR forests was one quarter of other forests (AR: 10.9 ± 8.6; P: 35.7 ± 19.8; E: 40.8 ± 18.0 individuals plot-1). All forests had similar soil cover. There was high variability among plots. The silvicultural methods in AR may have favored greater biomass. However, the sustainability of AR could be jeopardized by the low abundance of individuals in the understory. Most natural regeneration occurs near remnants, while active restoration is carried out in more isolated areas. Therefore it is difficult to disentangle local and landscape factors affecting understory abundance in AR. Depending on the ecosystem service, AR forests can perform similar or better than naturally regenerating forests, however, monitoring is needed to assess the sustainability of AR communities.