Pasture soils in the tropical moist life zone of Costa Rica: Implications for secondary forest succession

A study was conducted in the Tropical Moist Forest Premontane Life Zone in the Puriscal area of Costa Rica to examine differences in soil characteristics in the upper 15 cm of twenty-four paired plots of undisturbed primary forest and pasture. Bulk density and pH were different between the forest and pasture plots. Clay content, total organic carbon and total nitrogen were not different. The twenty-four pasture plots were divided into three erosion groups. There were differences in clay content, organic carbon and total nitrogen between the groups. Clay content of the pasture plots increased with increased erosion, while organic carbon and total nitrogen decreased. This was attributed to thinning of the surface horizon through erosion. The erosion groups were also compared to their associated forest plots. The least eroded plots had higher average bulk densities, and increased organic carbon and total nitrogen than did the forest plots. The increased organic carbon and total nitrogen in the pasture were either artifacts from using bulk density to calculate organic carbon and total nitrogen amounts or from changes in nutrient cycling pathways when livestock graze. Moderately eroded plots had total organic carbon and total nitrogen contents similar to those in the forest plots. This was attributed to gains in nitrogen and carbon from increased nutrient cycling or from bulk density calculations being offset by losses from increased erosion. The most eroded pasture plots had smaller amounts of total organic carbon and total nitrogen than the forest plots. This was mainly attributed to erosion.;Soil and site data were used in a forest site-quality assessment model, developed in Costa Rica, to identify pasture plots in which the operational factors of moisture, nutrients, light and heat had changed to the point that tree regeneration might be impeded. There were differences between all four factors for forest plots and pasture plots. There were no differences between the factors of moisture, nutrients and heat for the three erosion groups. However, the light factor was different between erosion group 1 and the other two erosion groups.;The operational factors for the forest, pasture and erosion group plots were compared to ecographs of three tree species, one of each representing species suited to forest gaps, poor sites and fertile sites. The forest and pasture plots fell within the boundaries of the ecographs for the forest gap and poor site species and fell outside the boundary for the fertile site species. Erosion group 3 fell outside of the ecograph boundary for the poor site species, while the other two groups fell within. This indicated that tree regeneration may be impeded in erosion group 3 for the poor site species.