Variation in stand structure, light and seedling abundance across a tropical moist forest chronosequence, Panama

• Published in: Journal of vegetation science 2000-04, Vol.11 (2), p.201-212
• Main Authors: Denslow, Julie S, Guzman G., Sandra
• Format: Article
• Language: English
• Subjects: aerial photography; Barro Colorado Island; basal area; canopy; chronosequences; Diversity; Forest canopy; Forest ecology; Forest regeneration; Forest stands; forests; growing season; Logistic regression; Old growth forests; PAR; photosynthetically active radiation; prediction; secondary succession; Seedlings; Semivariance analysis; Spatial pattern; species diversity; Stand age; stand structure; Succession; transmittance; trees; Tropical forests; Tropical rain forests; Understory
• ISSN: 1100-9233; 1654-1103
• DOI: 10.2307/3236800

We asked whether forest structure and understory light environments across a tropical moist forest chronosequence followed predictions of a 4‐phase model of secondary succession (establishment, thinning, transition and steady‐state) and whether seedling density and diversity were functions of light availability as predicted by this model. Using aerial photographs, we identified eight second‐growth stands (two each aged ca. 20, 40, 70, and 100 yr) and two old‐growth stands within Barro Colorado Nature Monument, Panama. Trees and seedlings were sampled in nested, contiguous quadrats in 2 160‐m transects in each stand. Light was measured as percent transmittance of diffuse photosynthetically active radiation (TPAR) at each seedling quadrat and by estimation of percent total incident radiation during the growing season from hemispherical canopy photographs. Basal area, tree density, and canopy height followed predictions of the 4‐phase model. Percent total radiation, but not TPAR, declined with stand age as did seedling density. While seedlings were more likely to occur in quadrats at higher light levels, much variation in seedling density was not related to light availability. Seedling patch sizes were small irrespective of light patches, estimated as semivariance ranges. Seedling species richness was a function of seedling density; estimates of species diversity unbiased by density did not vary systematically as a function of stand age. Proximate seed sources, efficient dispersal mechanisms, and appropriate establishment conditions can promote establishment of species‐rich communities early in successions of heterogeneous tropical moist forest.