Background/Question/Methods
Climate and land-use change can have a profound influence on vegetation community composition, but it is less clear how this will, in turn affect soil processes and microbial communities.  Previous research on interactions between tree species composition and soil properties has focused on mono or mixed-culture plantations, an important area of study because it represents a growing type of land-use change.   However, it is unknown whether lessons from plantations can be generalized to natural forests.  In order to investigate this question, we conducted a study of how tree species composition affects soil processes in both plantations and natural primary forests in the eastern Amazon basin near Santarem, Brazil.

We selected eight tree species represented on both forest and plantation.  All sites (three forest and one plantation) were on clay-rich (>80%) soils and all measurements were conducted in the rainy season, when the greatest soil flux variability has been observed. We measured all trees and soil, pH, BD, moisture content, temperature, and gas fluxes close to large (> 35 cm DBH) trees in the forest and centrally in monoculture plots on the plantation.

Results/Conclusions
The plantation soil was more compacted (higher bulk density), drier, and warmer than forest soil, which in part caused low soil N₂O fluxes (24±2 vs. 77±3 μg-N m^-2 h^-1 on plantation and in forest), CO₂ fluxes (212±6 vs. 238±5 mg-C m^-2 h^-1), and CH₄ uptake rates (-11±4 and -52±6 mg-C m^-2 h^-1).  We found tree species significantly influenced soil physio-chemical parameters as well as N₂O fluxes in the forest and CH₄, CO₂, and N₂O fluxes on the plantation.  However, tree species influences were not consistent in plantation versus forest, as species associated with high or low fluxes were not the same in the forest and plantation.    None of the variables measured could explain more than 15% (species on the plantation) of the CO₂ flux variability anywhere.  Tree species explained most of the N₂O flux variability both on the plantation and the forest (R²=0.35 and 0.17, respectively) and we found a strong negative relationship between tree growth rates and mean soil N₂O fluxes at both sites.

Our results demonstrate that tree species and potentially tree growth rates can influence soil processes in tropical forests and plantations, but that species differences on plantations do not carry any predictive capability for within forests.