Abstract
Individual performance is a function of an individual's traits and its environment. This function, known as an environmental filter, varies in space and affects community composition. However, filters are poorly characterized because dispersal patterns can obscure environmental effects, and few studies utilize longitudinal data linking individual performance to environment. We model the effects of environmental filters on demographic rates of nearly all tree species (99) in a 25-ha subtropical rain forest plot. We develop a hierarchical Bayesian model of environmental filtering, drawing inspiration from classic studies of intraspecific natural selection. We characterize the specific environmental gradients and trait axes most important in filtering of demographic rates across species. We found that stronger filtering along a given trait axis corresponded to less spatial variation in the value of favoured traits. Environmental gradients associated with filtering were different for growth versus survivorship. Species maximum height was under the strongest filtering for growth, with shorter species favoured on convex ridges. Shorter stature species may be favoured on ridges because trees on ridges experience higher wind damage and lower soil moisture. Wood density filtering had the strongest effects on survival. Steep slopes and high available P in the soil favoured species with low-density wood. Such sites may be favourable for fast-growing species that exploit resource-rich environments. Synthesis: We characterized trait-mediated environmental filters that may underlie spatial niche differentiation and life-history trade-offs, which can promote species coexistence. Filtering along trait axes with the strongest effects on local community composition, that is, traits with the strongest filtering, may necessarily have a weaker potential to promote species coexistence across the plot. The weak spatial variation in filters with strong effects on demography may result from long-term processes affecting the species pool that favour habitat generalist strategies.
Original language | English (US) |
---|---|
Pages (from-to) | 722-733 |
Number of pages | 12 |
Journal | Journal of Ecology |
Volume | 101 |
Issue number | 3 |
DOIs | |
State | Published - May 2013 |
All Science Journal Classification (ASJC) codes
- Ecology, Evolution, Behavior and Systematics
- Ecology
- Plant Science