Applied nucleation, mostly based upon planting tree islands, has been proposed as a cost‐effective strategy to meet ambitious global forest and landscape restoration targets.

We review results from a 15‐year study, replicated at 15 sites in southern Costa Rica, that compares applied nucleation to natural regeneration and mixed‐species tree plantations as strategies to restore tropical forest. We have collected data on planted tree survival and growth, woody vegetation recruitment and structure, seed rain, litterfall, epiphytes, birds, bats and leaf litter arthropods.

Our results indicate that applied nucleation and plantation restoration strategies are similarly effective in enhancing the recovery of most floral and faunal groups, vegetation structure and ecosystem functions, as compared to natural regeneration.

Seed dispersal and woody recruitment are higher in applied nucleation and plantation than natural regeneration treatments; canopy cover has increased substantially in both natural regeneration and applied nucleation treatments; and mortality of planted N‐fixing tree species has increased in recent years. These trends have led to rapid changes in vegetation composition and structure and nutrient cycling.

The applied nucleation strategy is cheaper than mixed‐species tree plantations, but there may be social obstacles to implementing this technique in agricultural landscapes, such as perceptions that the land is not being used productively.

Applied nucleation is likely to be most effective in cases where: planted vegetation nuclei enhance seed dispersal and seedling establishment of other species; the spread of nuclei is not strongly inhibited by abiotic or biotic factors; and the approach is compatible with restoration goals and landowner preferences.

Synthesis and applications. Results from our 15‐year, multi‐site study show that applied nucleation can be a cost‐effective strategy for facilitating tropical forest regeneration that holds promise for helping to meet large‐scale international forest restoration commitments.

We recorded vegetation point-intercepts at 2-5 m and >5 m heights along eight parallel 48-m transects through our restoration plots at 12 sites. Transects were separated by 6 m and points along transects were separated by 4 m for a total of 96 points. 

Where collected: 13 restoration sites in Coto Brus County, Puntarenas, Costa Rica