Ecological and evolutionary consequences of localised predator-prey interactions
In my PhD, I have focused on
(1) how predator home ranges/territories influence predator-prey spatial structure and population dynamics
(2) the resulting selective pressures acting on prey or predator dispersal rate/distance, and predator home range size.
From a technical perspective, I have delved into
My venture into spatial moment equations was joint work with David Murrell (UCL). Some computer code is available here. A review on this topic at Methods in Ecology and Evolution.
An example: evolutionarily stable home range sizes
Using a dynamic spatial point process model, we studied the drivers of home range size (HRS). The evolutionarily stable home range size of a predator depends on its resource characteristics, such as productivity or spatial patterns generated by various dispersal/competition distances, as well as interactions with conspecifics (see a simplified picture of the model on the left, paper here). Among other things, we found that prey spatial autocorrelation was relatively unimportant to HRS while predator-prey spatial cross-correlation was crucial to HRS; and that more 'stay-at-home' predators should have large, overlapping home ranges to dilute competition for resources among neighbours. The methods involved first deriving moment equations from the stochastic and spatial process, and then performing invasion analyses, in the framework of Adaptive Dynamics. In other words, this is optimal foraging theory with population dynamics and evolutionary feedbacks. Another topic we investigated with a similar model is prey dispersal evolution under different predation regimes.More foraging theory: movement, cognition, and functional responses (predator intake= f(prey density))
As predator-prey interactions result from behavioural processes, biased random walk models can be used to study foraging on a finer spatiotemporal scale. We developed a model to understand which cognitive abilities of predators (memory, perception) were most useful in various spatial prey patterns (pdf here). A central place forager alternates between random search, capturing prey, bringing it back to the nest, and homing to memorized profitable places (right-hand figure). We found that long memory was more useful in aggregated spatial patterns (as found before) but also that predators relying on perception were better in more random prey patterns. In turn, this implied the intake rate of 'memorizers' reacted more to prey spatial aggregation and that of 'perceivers' to prey density - while superefficient central place foragers with good memory and perception have a constant functional response (up to the point where prey is completely depleted).
In short, Holling's functional response model may not work so well for central place foragers knowing well their environment; use of memory tends to modify the relationship between intake and prey density. My conclusion from this is that most shapes of functional response are actually possible (so there's really no need to be dogmatic about any, and let as much as possible the available data inform choices in population/community dynamics models). Here's one other example of the many ways movement changes functional responses.
