Search for a command to run...
There is current debate about the potential for secondary regrowth to rescue tropical forests from an otherwise inevitable cascade of biodiversity loss due to land clearing and scant evidence to test how well active restoration may accelerate recovery. We used site chronosequences to compare developmental trajectories of vegetation between self-organized (i.e., spontaneous) forest regrowth and biodiversity plantings (established for ecological restoration, with many locally native tree species at high density) in the Australian wet tropics uplands. Across 28 regrowth sites aged 1-59 years, some structural attributes reached reference rainforest levels within 40 years, whereas wood volume and most tested components of native plant species richness (classified by species' origins, family, and ecological functions) reached less than 50% of reference rainforest values. Development of native tree and shrub richness was particularly slow among species that were wind dispersed or animal dispersed with large (>10 mm) seeds. Many species with animal-dispersed seeds were from near-basal evolutionary lineages that contribute to recognized World Heritage values of the study region. Faster recovery was recorded in 25 biodiversity plantings of 1-25 years in which wood volume developed more rapidly; native woody plant species richness reached values similar to reference rainforest and was better represented across all dispersal modes; and species from near-basal plant families were better (although incompletely) represented. Plantings and regrowth showed slow recovery in species richness of vines and epiphytes and in overall resemblance to forest in species composition. Our results can inform decision making about when and where to invest in active restoration and provide strong evidence that protecting old-growth forest is crucially important for sustaining tropical biodiversity.