Home | Curriculum Vitae (pdf) | Research and Publications | Ecology Links | Natural History Photos | Natural History Photos II

RESEARCH INTERESTS

Postdoctoral research: Changes in tropical forest structure and dynamics

Recent studies have revealed that tropical forest are not as stable as once presumed and that they have undergone dramatic shifts in composition and structure over the past several decades.  In particular, tropical forests worldwide are reported to be increasing in standing aboveground biomass which has important implications for the global carbon budget.  Although the cause(s) of increasing biomass remains unresolved, several different global change phenomena have been proposed as potential drivers, including increased atmospheric CO₂, increased nitrogen deposition, increased irradiance, and increased frequency of disturbances including ENSO associated droughts.  Alternatively, forest changes may be due to idiosyncratic or endogenous factors such as altered faunal communities or succession from past large-scale disturbances. 

For my postdoctoral research I am investigating changes in forest dynamics and the relative the contribution of potential drivers using data from a pantropical network of large (50-ha) permanent tree plots administered through the Center for Tropical Forest Science (CTFS).  More specifically, I am conducting detailed analyses of patterns of change in the different components of biomass accumulation (i.e. recruitment, mortality, growth) at various temporal, spatial, and taxonomic levels within and across plots.  These tests allow me to account for variations in forest type, species composition, soil types, disturbance history, etc. that have confounded previous studies.  This project integrates eco-informatics and directed field experiments.

The results of this study will have important implications not only for increasing our understanding of the factors influencing forest dynamics but also for conservation biology.  By determining the impact of various anthropogenic disturbances on these forests, we will be better able to direct efforts at mediating these effects through legislation, policy, or directed conservation initiatives. 

Related publications:

Feeley, K.J., S.J. Wright, S. Davies, M.N.S. Noor, and A.R. Kassim. 2007 (in press). Decelerating growth in tropical forest trees. Ecology Letters. [pdf]

* featured in the Harvard Gazette [pdf]

Feeley, K.J., S. Davies, M.N.S. Noor, A.R. Kassim, and S. Tan. 2007. Do current stem size distributions predict future population changes? An empirical test of intraspecific patterns in tropical trees across two spatial scales. Journal of Tropical Ecology 23: 191-198. [ pdf]

Feeley, K.J., S. Davies, P. Ashton, et al. Succession and biomass dynamics of old-growth tropical forests. In review.

Chave, J., R. Condit…K.J. Feeley, et al. Changes in tropical forest biomass dominance across species and functional group. In review.

Factors determining the lower distribution of tree species: the role of herbivory

The distributional ranges of plant species are the result of a variety of abiotic and biotic factors that decrease fitness of species outside of their range boundaries.  This principle applies not only to latitudinal and longitudinal ranges but also to elevational ranges.  In general, abiotic conditions are more favorable to plant growth in the lowlands (i.e. warmer temperatures, greater nutrient availability, etc.) and thus lower elevational boundaries are hypothesized to be due primarily to biotic interactions, such as increased herbivory, competition, and pathogens, which decrease the fitness of highland plant species at lower elevations. In association with Miles Silman of Wake Forest University and the Andean Biodiversity Consortium, I am testing this hypothesis along an elevational transect in Manu National Park, Peru.  Currently I am measuring rates of herbivory along the transect.  Eventually this data will be combined with experimental transplants of seedlings in conjunction with various environmental manipulations (including herbivore exlosures).

Related publications:

Ibanez, I., J. Clark, M.C. Dietze, K.J. Feeley, M. Hersh, S. LaDeau, A. McBride, N.E. Welch, and M.S. Wolosin. 2006 Predicting biodiversity change: outside the climate envelope, beyond the species-area curve. Ecology (invited) 87: 1896-1906. [pdf]

large island                               small island

Dissertation research: The effects of tropical dry forest fragmentation on floral and faunal communities as mediated through trophic interactions

As a result of ongoing deforestation and land conversion, over half of the worlds tropical forests have already been lost with an additional 2-4% being destroyed each year, inevitably leaving behind complex patchworks of isolated forest fragments embedded in an inhospitable matrix of crops, pasture, or wasteland.  While habitat fragmentation is widely recognized as one of the leading causes of species extinctions, the mechanisms underlying these extinctions remain poorly resolved.

In order to investigate the effects of habitat fragmentation on tropical floral and faunal communities, I conducted a series of experimental and observational studies using recently isolated land-bridge islands in Lago Guri, Venezuela, as a model system.  These studies focused on understanding the diverse effects of distorted trophic communities, specifically looking at the impacts of altered herbivore abundances on ecosystem processes (i.e. nutrient cycling) and how these effects can ramify through the food web to affect the diversity and persistence of other taxonomic groups.  I determined that hyper-abundant populations of herbivorous red howler monkeys (Alouatta seniculus) caused a significant decrease in the availability of soil nutrients  for plant uptake at the whole-island scale, yet also caused a six-fold increase in tree growth rates.  I proposed that this seemingly paradoxical pattern is attributable to the latrining behavior of the monkeys which causes the formation of localized patches of increased nutrient availability.  

I also investigated the potential for changes in plant productivity to affect the communities of forest-interior birds nesting on the islands.  I showed that the diversity of nesting birds is positively correlated with tree growth and that there is a positive indirect effect of howler monkey density on bird diversity.  In addition, utilizing a ten-year record of avian censuses, I investigated the relationship between rates of local extinction and several variables related to island biogeographic and higher-level (i.e. trophic) effects.  The best model describing rates of bird extinction did not include island area or nest predation as is generally predicted, but rather only the density of generalist herbivores including leaf-cutter ants (Atta sp.) and howler monkeys.  Finally, I investigated how the insular communities of birds were structured.  Based on ‘assembly rule’ models, I showed that the communities were only weakly structured by competition and that community composition was determined primarily through selective extinction and differential habitat requirements. 

In 2003, I initiated as series of whole island manipulation experiments designed to directly test the interactions between herbivores, nutrient cycling, and plant communities. On a subset of small islands, I captured and removed all howler monkeys and on a complimentary set of islands I applied treatments of slow-release NPK fertilizer.  This study was in conjunction with the work of my advisor, Dr. Terborgh, who removed leaf-cutter ants and/or rodents from additional sets of small islands.  Unfortunately, this project is currently on hiatus because of the tense political situation in Venezuela.  Hopefully, international relations will improve and we will be allowed to return to Lago Guri and monitor the effects of our experiments in upcoming years.

The implications of this research extend well beyond tropical forest fragments.  Humans are altering herbivore communities throughout the world through both direct and indirect activities (an example of this in temperate forests is the dramatic increase of deer populations in protected areas due to the lack of predators and top-down control).  As my research has shown, changes in these consumer populations will have important implication not only for the vegetation that they feed on, but also for other faunal groups such as insects and birds.

Related publications:

The use of remote sensing to measure forest structure and dynamics

Remote sensing is a powerful tool with the potential to greatly increase our knowledge, especially in tropical forests that are often inaccessible to on-site field studies.  I am working in collaboration with Tom Gillespie of UCLA to test the effectiveness of various satellite born sensors for measuring forest attributes (structure, diversity, composition, successional status, productivity, etc.) at various spatial scales. 

Related Publications:

Feeley, K.J., T.G. Gillespie, and J.W. Terborgh. 2005. The utility of spectral indices from Landsat ETM+ for measuring the structure and composition of tropical dry forests. Biotropica (invited) 37: 508-519. [pdf]

picture from the Harvard Museum of Natural History

Structure of Carabib beetle communities in spruce fir sky islands of the southern Appalachians USA

I am working with Bob Browne and Carmen Chavez of Wake Forest University to test the role of habitat requirements and interspecific interactions in structuring the composition and structure of flightless ground beetles inhabiting the relict spruce fir forest sky islands on the tops of tall mountain in the southern Appalachian mountains.

The effects of treefall gap size and orientation on colonizing plant communities

Treefall gaps are proposed to play a major role in helping to promote and maintain the high diversity of trees found in tropical forests by increasing habitat heterogeneity and thus allowing for the coexistence of different guilds of species.  I am investigating the effects of treefall gap size and orientation on the diversity, composition and growth of the colonizing tree community using a set of 75 simultaneously created gaps in the Amazonian lowlands of Manu National Park, Peru.  Since these gaps were all created by a single disturbance event (a powerful windstorm), the confounding effects of agent of disturbance as well as gap age and season of formation, are minimized.

Reproductive behavior and breeding ecology of the Sula dactylatra

picture from UGA marine lab

Influence of drainage complexity on intertidal salt marsh nekton communities

Long term vegetation dynamics of temperate old growth forests