Forest inventories
We established one 20 × 45 m plot in each of the aforementioned forest stands, totaling 30 plots. For second-growth forests and restoration plantations, which usually had a small and elongated area, we allocated the inventory plots as centralized as possible in the forest stand in order to avoid edge effects. In old-growth forest stands, which were larger than the aforementioned forest typologies, we took the main trail that crossed these forests an established the inventory plot when distant at least 50 m from the forest edge. For all plots, we avoided forest gaps or patches affected by human disturbances (e.g. fire or logging).
We measured the diameter at breast height (DBH) of all trees and shrubs (DBH ≥ 5 cm) within each plot and identified them to species level, without distinguishing between planted or spontaneously regenerated trees, and used these data to calculate above-ground biomass based on the equations developed by Chave et al. (2014), with wood density values mostly compiled from Chave et al. (2009) and César et al. (2017). Above-ground carbon stocks were calculated by multiplying above-ground biomass by 0.46, the average carbon content of dry woody biomass of Atlantic Forest trees (Ferez, Campoe, Mendes, & Stape, 2015).