Aim: To investigate the influence of multiple canopy openings on the composition and diversity of recruited saplings in a forest frequently disturbed by typhoons.

Methods: We conducted tree-by-tree censuses (diameter at breast height ≥ 1 cm) and mapped gaps (canopy height < 5 m) in 1993, 2000, 2008, and 2013 in a tropical mountain zonal foothill evergreen broad-leaved forest in Nanjenshan Nature Reserve, Taiwan. We analysed the composition and diversity of recruited saplings within a 2.1 ha plot (840 sampling quadrats (5 m × 5 m)) with variable numbers of canopy openings recorded during the study period.

Results: Composition of recruited saplings was dissimilar between quadrats that stayed opened and those that stayed closed throughout the study period (pairwise similarity estimates C₀₂ = 0.52, 95% CI = 0.38–0.66). The quadrats under closed canopy had high diversity when weighting rare species (species richness), whereas quadrats with one or two gap opening records during the past 20 years had high diversity when weighting the abundance of species. Although canopy openings provided establishment conditions for saplings of some shade-intolerant species, due to the nature of small gap size, such habitats don’t favour the dominance of shade-intolerant species.

Conclusions: Even in a frequently-disturbed forest, species composition and richness of recruited saplings were mainly contributed by shade-tolerant species. Although multiple canopy openings facilitated the establishment of shade-intolerant species, species diversity in the study forests is possibly mainly mediated by coexistence mechanisms of those shade-tolerant species rather than light-gap-related species strategies.

This dataset is the basal area (m²) of recruited saplings (diameter at breast height ≥ 1 cm) in the 2000, 2008, and 2013 censuses of a tropical mountain zonal foothill evergreen broad-leaved forest in Nanjenshan Nature Reserve, Taiwan. There are 840 sampling quadrats (5 m × 5 m) within the 2.1 ha study plot. Many individuals had multiple stems, and some stems may die or recruit during the census period. Thus, we sum up the basal area (radius² × π) of the stems for each individual to indicate its survival status. The canopy opening history type of each sampling quadrats were also indicated in the column Type.9313. We also included a worksheet with species abbreviation and species photosynthetic capacity (Amax; μmol CO2 m^-2 s^-1) measured by Kuo and Yeh (2015) and Kuo et al. (2021).