COS 239-1 - Dominance and demography of Lecythidaceae in the Central Amazon

Background/Question/Methods

Why certain species are common and others rare remains an unanswered questions in ecology, and the extent to which a small subset of species dominate is particularly striking in the tropics. Several ecological processes are known to influence species demography and therefore abundances, and may be associated with species dominance. These include demographic tradeoffs, such as that between species maximum growth rate and mortality, as well as between stature and recruitment. In addition to demographic rates, the spatial location of demographic events, notably mortality, is significant in tropical forests where conspecifics are less likely to establish and survive close to each other due to host-specific pathogens. We ask whether particular demographic strategies, being less prone to demographic tradeoffs, or demonstrating less spatially clustered mortality is associated with Lecythidaceae dominance in the central Amazon.

We use demographic data from a large forest plot near Manaus, Brazil where Lecythidaceae individuals were recensused after 30 years. We use linear regression models and principal component analyses to determine whether hyperdominant species demonstrate a particular demographic strategies or are less prone to ecological tradeoffs, and mark-connection functions to determine whether hyperdominant species display less clumped mortality than nondominant species.

Results/Conclusions

Analyses thus far demonstrate that hyperdominant species display demographic strategies that include show growth and high recruitment, suggesting they have high reproductive investment. There are no hyperdominant species with high growth rates within the clade, suggesting that “fast” and “long-lived pioneer” strategies do not lead to dominance among Lecythidaceae. Dominant species tend toward high mortality given slow growth, but mortality does not seem to increase with increasing mean growth, thus they may be less affected by the growth mortality tradeoff than other species. Dominant species demonstrate less clumped mortality than nondominant species, including those with similar plot abundance, suggesting they may be less prone to conspecific negative density dependent effects.

A better understanding of patterns of dominance throughout the Amazon, and the cause of those patterns, is critical to predict how forests will respond to future change. Given the outsize contribution of the Amazon region to global biogeochemical cycles and carbon sequestration, and the disproportionate contribution of Lecythidaceae to these processes, this understanding is vital if we are to define the most effective land use and management strategies for maintaining diversity in an area that is being rapidly altered by humans.