Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.747478
Title: Automated acoustic biodiversity assessment in cities
Author: Fairbrass, Alison J.
ISNI:       0000 0004 7230 9680
Awarding Body: UCL (University College London)
Current Institution: University College London (University of London)
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Abstract:
In the last 40 years more than half of the world’s wildlife populations have disappeared while anthropogenic disturbance continues to push many species to extinction. Cities, which now support over half of the world’s human population, also support biodiversity. Yet the green infrastructure (GI) components of cities are not currently supporting high biodiversity, partly due to the resource-intensity of biodiversity assessment in urban environments. Ecoacoustics, which uses biotic sound as a proxy for biodiversity, could provide an improved way of assessing urban biodiversity, although the use of ecoacoustics in cities dominated by anthropogenic noise remains untested. Here, I demonstrate how ecoacoustics can be used to assess biodiversity in complex and highly disturbed urban environments. I set the scene by using a global terrestrial urban studies database to show that GI does not currently mitigate against biodiversity losses in cities. Then, using an annotated urban ecoacoustics dataset, CitySounds2017, generated from audio data I collected within and surrounding Greater London, UK, I show that several commonly used Acoustic Indices are unsuitable for use in cities without the prior removal of non-biotic sounds from audio data. Next, using CitySounds2017 I develop CityNet, a pair of machine learning algorithms for quantifying biotic and anthropogenic sound in urban audio data. Finally, I show that a strong correlation exists between acoustic and environmental measures in urban GI habitats in London. I anticipate the methods developed here to be a starting point for improved assessment of biodiversity that informs management to maximise the wildlife supported by cities. For example, CityNet could be integrated into urban sensing networks to facilitate large-scale biodiversity assessment. As anthropogenic disturbance increases globally, the need for methods of biodiversity assessment that are reliable in disturbed environments will only increase, and I see these methods as having the potential to support biodiversity assessment globally.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.747478  DOI: Not available
Share: