Phase 1 (2007-­2012)

Annual censuses of trees were conducted in field sites in Central Amazonia, Mexico, and Costa Rica. Based on these data, we have tested the validity of chronosequence predictions for a range of response variables including vegetation structure (Chazdon et al. 2007, Letcher and Chazdon 2009 a,b), species richness and diversity (Chazdon et al. 2007, Letcher and Chazdon 2009b), species composition (Norden et al. 2009, 2011), functional group composition (Chazdon et al. 2010, Letcher and Chazdon 2012), recruitment and mortality rates (van Breugel et al. 2007), above- and below-ground competition (van Breugel et al. 2012), phylogenetic community structure (Letcher 2009, Letcher et al. 2012, Norden et al. 2011), and community aggregate functional trait values (Lohbeck et al. 2012).

Our studies are generating new insights into factors affecting successional pathways. Young and intermediate secondary forests with low intensities of prior land use that are close to mature forest remnants in NE Costa Rica show convergence of species composition of woody seedlings and saplings with mature forests over time (Norden et al. 2009). In contrast, frequent burning of pastures prior to abandonment in Central Amazonia produces a Vismia–dominated community capable of arresting natural forest succession (Norden et al. 2011, Wieland et al.2011, Jakovac et al.2012, Williamson et al. in press).

For more details of  Phase 1 results and broader impacts, click here

Phase 2 (2012-2017)

In this second phase, we emphasize analyses of population level changes and functional trait distributions during succession, as these are the basis for changes in community structure and composition. During the next five years we will continue to collect core data from research sites in Brazil, Costa Rica, and Mexico, providing a unique opportunity to examine the functional basis for successional changes, comparing chronosequence trends with long-term plot data. Long-term annual data on species population dynamics during succession collected over 15 to 28 years will be used to characterize vital rates and changes in community weighted mean functional traits within and across study areas.

Additionally, we will compare successional pathways and functional trait distributions from a dry forest region, Oaxaca, Mexico and a semi-deciduous forest region in eastern Bolivia to our tropical wet forests.

We will test three major hypotheses related to our decadal plan to investigate successional pathways:

1) Species-specific functional traits can predict variation in demographic rates across species and within and among successional stages.

2) Community aggregate functional traits and trait frequency distributions of fruits, seeds, leaves, and stems shift during succession in consistent patterns among plots in a chronosequence and within plots over time.

3) Differences in the functional trait distributions and generalist/specialist classifications among study areas are driven by differences in the species pool of the surrounding landscape.