Secondary forests recovering on abandoned lands constitute an increasingly important portion of tropical forestlands. Secondary forests provide valuable ecosystem services such as carbon sequestration and contribute to biodiversity conservation. These forests occur on both public and private lands. Ownership could influence forest structure, biomass and composition through differential management regimes or through the non-random nature of land use. However, few studies have examined this aspect of tropical forest ecology. We surveyed in 82 public and 66 private 0.1 ha forest plots arrayed across forest age and soil gradients in tropical dry forests in Guanacaste, Costa Rica. In each plot, we identified and measured diameter of all trees > 10 cm DBH, collected soil samples (0-10 cm depth) and sampled wood density. Here we present results of analyses of soils, forest structure, aboveground biomass, species diversity and species composition.
Results/Conclusions
Overall, we found considerable overlap in soil chemical properties but identified a subset of public forest plots with particularly rich soils compared to the dataset as a whole; and a split at low end of the soil fertility gradient, with a subset of public with relatively high levels of Mn and low levels of N, P, and K compared to private plots. We found no difference between ownerships in basal area, stem density, aboveground biomass, or Shannon-Wiener diversity. Conversely, we found strong evidence for differences in tree species composition. A NMS ordination revealed differences among ownerships related in part to differences in soils. A classification analysis revealed 24 species (of 166 total) more prevalent in public forest (15%), 15 more prevalent in private forest (9%), 47 with no distributional bias (28%), and 80 too rare to classify (48%). Public forests included valuable timber species (e.g. Swietenia macrophylla [mahogany]) not common in private forests. The most abundant species in private forests was Guazuma ulmifolia, a species widely consumed and dispersed by cattle. Our results suggest that differences in tree composition result from non-random land use with respect to soil fertility and species-specific management differences between public and private forests.