| Research | ||
Ant-termite predatory
interactions
I am interested in quantifying the degree and specificity of ant predation
on termites. Despite being two of the most abundant and ecologically important
animal groups in tropical rain forests, work to date has tended to focus
on a single pair of species and consequently there is little understanding
of the overall nature of this interaction. At the Natural History Museum,
London, I am using molecular methods to determine the overall frequency
of termite predation and the level of ant predatory specificity in a rain
forest in Gabon.
Ant-plant symbioses
During my PhD I studied the interaction between bird’s nest ferns
(Asplenium spp.) (Fayle et al. 2008 [PDF];
Fayle et al. 2009 [PDF])
and their ant inhabitants. These ferns do not cultivate a relationship
with a single species of ant. Instead, over 70 different ant species nest
in the ferns. This tremendous diversity is in part maintained by a phylogenetic
hierarchy of competitive interactions between ant species. At the other
end of the spectrum of specialisation, I have studied the interaction
between two specialist ant species and the rattan plant they both inhabit
(Edwards et al. 2010 [PDF]).
Effects of habitat change on ant communities and the functions
they perform
Clear-felling relating to agricultural expansion, logging and fragmentation
of forest are major drivers of biodiversity loss (Turner et al. 2008
[PDF]).
I have quantified the way in which habitat modification impacts on ant
communities (Fayle et al. 2010 [PDF])
and the subsequent change in ecosystem function (Fayle et al. 2010 [PDF]).
I am also involved in a new experimental forest fragmentation project
(Stability of Altered Forest Ecosystems – SAFE) that seeks to
understand the relationship between local landscape habitat complexity
and biodiversity and ecosystem functioning.
Development
of techniques for detecting species interactions from presence/absence
data
Null models of species co-occurrence are an increasingly popular way of
assessing interactions between species. Despite this, little is known
about the rates of Type 1 and Type 2 errors for such metrics. I have used
a simulation-based approach to assess the way error rates vary with matrix
size and variance in species abundance for a commonly-used metric, the
c-score (Fayle & Manica In press [PDF on request]).
My PhD supervisors were William Foster and Paul Eggleton. My MPhil supervisors were Rhys Green and William Foster.
For these projects I am funded by the Natural Environment Research Council, the Natural History Museum, the Hanne and Torkel Weis-Fogh Fund and the Eranda Foundation.