Recovery of forest structural complexity during secondary succession in the tropics

Martin Ehbrecht , Tim Lehmann, Sebastián Escobar, David Donoso, María-José Endara, Juan Ernesto Guevara-Andino, Nico Blüthgen

Forest structural complexity is an essential determinant of forest ecosystem functions and biodiversity. The natural dynamics of structural complexity of tropical forests remain largely unexplored, especially for naturally regenerating forest during secondary succession. Better understanding the trajectories of forest structural complexity recovery is crucial to inform the development of forest landscape restoration strategies and to predict the reassembly of ecological networks during secondary succession. Here, we investigate the recovery of forest structural complexity during secondary succession following land use abandonment in a human-modified landscape in Ecuador. We employ a terrestrial laser scanning-based index of forest structural complexity to quantify three-dimensional vegetation structure of agricultural lands (cacao plantations and pastures), naturally regenerating sites, and primary old-growth forests along a chrono-sequence of secondary succession. We find that sites recovering after land use abandonment attain levels of forest structural complexity comparable to old-growth forest within 40 years. Changes in forest structural complexity along the successional gradient follow a saturating pattern, with rapid increases in the first years and only minor potential for further increases after 40 years. Increasing tree species diversity during secondary succession is identified as a major driver of the recovery of forest structural complexity. Using a structural equation modelling approach, we find that effects of tree species diversity on forest structural complexity are mediated by its effects on vertical stratification, as the effective number of canopy layers increases with increasing recovery age. Our results suggest that passive restoration is a suitable strategy to restore forest structural complexity in human-modified landscapes.