COMPARATIVE-ANALYSIS OF MICROHABITAT UTILIZATION BY SAPLINGS OF 9 TREE SPECIES IN NEOTROPICAL RAIN-FOREST

We examined the hypothesis that tropical rain forest tree species differ significantly in the degree to which their regeneration is associated with treefall gaps. Using static and dynamic analyses, we evaluated crown light environments and forest structure around 0.5-5 m tall saplings (N = 424) of two pioneer and seven nonpioneer tree species in old growth tropical wet forest at the La Selva Biological Station, Costa Rica. Canopy photographs were taken with a fisheye lens directly above all individuals, Each photograph was analyzed to calculate a Global Site Factor (GSF), which is analogous to the percentage of photosynthetically active radiation reaching a site relative to a totally open site. The amount and directionality of crown illumination were also estimated visually using a crown position index (Dawkins & Field 1978, Clark & Clark 1992). Forest structure around each individual, a reflection of local disturbance history, was categorized as gap, building, or mature. Saplings of all nine species were found in relatively dark sites (xBAR Global Site Factors 1.9-10.8%). The two pioneer species (Cecropia) were found in higher light microsites than the other species and were significantly different from each other. There were no significant differences in GSF among nonpioneer species. Results were similar for the direct and diffuse components of GSF, and for the crown position index. In contrast, forest structure around saplings varied significantly among nonpioneer species but not between Cecropia species. For six of the nine species short-term height or diameter growth was correlated with GSF or its direct or diffuse components. GSF was weakly but significantly correlated with sapling height and crown position. Analyses over time showed that all nine species had progressively lower crown illumination and fewer individuals in gap/building sites for 2-8 yr after initial encounter. Particularly steep decreases for Hyeronima and Pithecellobium saplings indicate that these species establish under canopy openings that subsequently fill in. Canopy photography, crown position index, and the forest structure index all provided ecologically interpretable information. Only canopy photography distinguished between microsites of the two Cecropia species. Forest structure and crown position indices are rapid and inexpensive methods for obtaining long-term information on microhabitats around individual trees. We found significant interspecific differentiation in microhabitat occupancy along a light gradient of only 0-25 percent of full sun. A crude 3-point index of forest structure revealed even further resolution. This study focused on only three microsite variables and a restricted size range of saplings. Comprehensive understanding of tropical tree regeneration will require long-term concurrent measurements of individual performance and associated environmental variables for all size classes of a broad range of species.