Montane to lowland differences in biomass, turnover, and productivity are closely associated with tree species functional traits in tropical forests of northern South America

Abstract:

Previous work has suggested the existence of a slow–fast continuum in lowland tropical forests where resource-rich forests have higher woody productivity (AGWP), faster stem recruitment and mortality (r, m), but often lower biomass (AGB) than resource-poor forests, and with these stand-level metrics being closely associated to the traits of their constituent tree species. Here, using data from long-term sample plots from western Venezuela, we extend this approach to examine whether such considerations also apply to the montane-lowland forest contrast. Specifically, we expect lowland forests to be dominated by acquisitive traits (e.g. high Specific Leaf Area – SLA), influencing faster stem dynamics and AGWP, while conservative traits (e.g. high leaf mass per area - LMA) would govern stand dynamics in high-elevation forests with slower turnover, reduced AGWP but higher AGB. We employed principal component analysis combined with linear mixed models, using data from 10 functional leaf and stem-level traits to describe the functional trait space in two contrasting tropical environmental settings, and analyze the differences in stand-level turnover rates (i.e. recruitment – r and mortality – m), aboveground biomass (AGB) and productivity (AGWP), using traits, soil nutrient content and chemistry, and stem density as potential predictors. The sites contrast strongly with regards to their structure, dynamics and function. Species’ traits in high elevation forests were mostly associated with adaptation to high soil moisture, greater shade tolerance and thus lower growth rates, whereas for lowland species, traits were more associated with higher resource availability. We found that different traits have effects over different responses with the Community Weighted Mean (CWM) of leaf nitrogen simultaneously involved in all our studied response variables. Combined, these results can be used to inform climate models and/or for restoration efforts that where site-specific data is needed.