Forest structure and litter predict the speed of demographic dynamics in secondary forests of the Atlantic Forest

Abstract

The knowledge of predictive factors related to the demographic dynamics of secondary forests is useful plan strategies for recovery of degraded areas, estimate the production of biomass and to subsidize the planning for rational use of resources. The aim of this study was to investigate possible biotic and abiotic factors that predict rates of demographic dynamics in second-growth vegetation of the Southern Atlantic Forest. We applied a PCA on data from six forests which varied in age from the last disturbance (25, 60, 75, 90 and more than 100 years-old) ordering of demographic rates (mortality, recruitment, abundance turnover, loss in basal area , basal area gain and basal area turnover), and used its ordination as response variable in a regression tree. As possible predictive factors were used: slope, soil chemical variables at different depths, forest structure parameters, amount of litter and spatial patterns. The regression tree revealed that the dynamics rates were better predicted by the forest structure and by the average amount of litter. Sites with a less developed structure in relation to maximum diameter, maximum height and basal area occurred in the youngest forests (25 years-old) and in clearings of the oldest forests (90 and 100 years-old), where the highest rates of demographic dynamics were observed. Areas with more developed structure were frequent in all forests independent of post-disturbance age. In these areas, the changes were slower and the amount of litter on the ground was inversely proportional to trees and basal area turnover.