Quantifying Bat Detection in Terms of Distance from a Roost

Institution: University of Exeter & Cardiff

Lead Supervisors: Fiona Mathews & Thomas Woolley

Project Description

To improve knowledge of bat ecology and achieve more effective conservation, it is frequently necessary to identify roosts. This is because bats are patchily distributed in the landscape and roost communally. Unfortunately, finding roosts is extremely difficult, both for practitioners and scientists. To understand and maximise the success of this process, we propose to combine the ecological expertise of Prof. Fiona Mathews (FM) and the modelling skills of Dr. Thomas Woolley (TW), in order to:

1) gain data on the dependence of bat detection rate on the distance of detectors from known roosts;
2) create a stochastic random walk model;
3) parameterise the model using data generated by the student;
4) simulate and predict bat spatial spread; and
5) (further work) reverse engineer the model to predict locations of unknown roosts.

This fits with FM’s research interests in the impacts of anthropological pressures on bat populations. This work requires detailed study of bat behaviour and population dynamics. TW is a mathematical biologist specialising in stochastic population motion and spatiotemporal complexity. He has worked on numerous generalisations of models of population movement. FM and TW have already worked together to produce a preliminary analysis (Figure 1), which is of interest to stakeholder as it shows that the probability of recording bats using acoustic surveys is strongly dependent on roost proximity.

The student will be jointly supervised, with FM providing the biological background and TW consulting on the modelling components. The student will be based at the Exeter. FM will provide the necessary training, logistical support and equipment.

The student will learn skills including ecological surveys for bats and acoustic analyses. These will be complemented by numerical modelling and stochastic simulation skills that will naturally incorporate the field data in order to parameterise the model.

There are no intellectual property concerns.









Indicative timescale for project (subject to change)


Powered by WordPress. Designed by WooThemes