Assessing ecosystem resilience through the use of ecoacoustics
Lead Research Organisation:
Imperial College London
Department Name: Design Engineering (Dyson School)
Abstract
Ecoacoustics involves recording and analysing natural soundscapes, which comprise all the sounds in an environment, to monitor biodiversity and ecosystem health. This project aims to advance methods in ecoacoustics in two main areas: (i) the use of spatial audio technology for studying the spatial distribution of sound and thereby gaining an enhanced understanding of ecosystems; and (ii) applying statistical early warning signals (EWS) of resilience loss ahead of a potential ecosystem tipping point to ecoacoustic data. As part of this work, months-long field deployments of the Multichannel Acoustic Autonomous Recording Unit (MAARU), a custom, open-source, fully-autonomous six-microphone recorder, were undertaken in the UK and Malaysian Borneo.
The spatial audio part of the project has three main objectives. First, to replicate real natural soundscapes in a laboratory setting, in order to create a lifelike but more controlled environment for testing multi-microphone field recorders. Second, to refine MAARU's hardware and software for longer-term deployments in harsh environments, such as tropical rainforests. Third, to explore reducing unwanted noise and spatially segmenting soundscapes, such as distinguishing sounds from either side of a forest edge, via beamforming - a technique for steering multi-microphone recordings to hone in on sounds from target directions.
The resilience part has two main objectives. First, to search large datasets of ecoacoustic recordings from tropical rainforests for EWS of tipping points. Tropical rainforests are focussed on given their global importance for biodiversity, carbon storage and livelihoods, coupled with concerns that they may soon tip to savannah-like or treeless states due to human pressures and climate change. The second resilience objective is to verify acoustics-based EWS against those derived from satellite imagery, which indicate rainforests' declining capacity to recover from disturbances in recent decades.
The spatial audio part of the project has three main objectives. First, to replicate real natural soundscapes in a laboratory setting, in order to create a lifelike but more controlled environment for testing multi-microphone field recorders. Second, to refine MAARU's hardware and software for longer-term deployments in harsh environments, such as tropical rainforests. Third, to explore reducing unwanted noise and spatially segmenting soundscapes, such as distinguishing sounds from either side of a forest edge, via beamforming - a technique for steering multi-microphone recordings to hone in on sounds from target directions.
The resilience part has two main objectives. First, to search large datasets of ecoacoustic recordings from tropical rainforests for EWS of tipping points. Tropical rainforests are focussed on given their global importance for biodiversity, carbon storage and livelihoods, coupled with concerns that they may soon tip to savannah-like or treeless states due to human pressures and climate change. The second resilience objective is to verify acoustics-based EWS against those derived from satellite imagery, which indicate rainforests' declining capacity to recover from disturbances in recent decades.
Organisations
People |
ORCID iD |
| Neel Le Penru (Student) |
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| NE/S007415/1 | 30/09/2019 | 29/09/2028 | |||
| 2607363 | Studentship | NE/S007415/1 | 30/09/2021 | 29/09/2025 | Neel Le Penru |