Back

Combining machine learning and a universal acoustic feature-set yields efficient automated monitoring of ecosystems

Sethi, S. S.; Jones, N. S.; Fulcher, B. D.; Picinali, L.; Clink, D. J.; Klinck, H.; Orme, C. D. L.; Wrege, P. H.; Ewers, R. M.

2019-12-05 ecology
10.1101/865980 bioRxiv
Show abstract

Natural habitats are being impacted by human pressures at an alarming rate. Monitoring these ecosystem-level changes often requires labour-intensive surveys that are unable to detect rapid or unanticipated environmental changes. Here we developed a generalisable, data-driven solution to this challenge using eco-acoustic data. We exploited a convolutional neural network to embed ecosystem soundscapes from a wide variety of biomes into a common acoustic space. In both supervised and unsupervised modes, this allowed us to accurately quantify variation in habitat quality across space and in biodiversity through time. On the scale of seconds, we learned a typical soundscape model that allowed automatic identification of anomalous sounds in playback experiments, paving the way for real-time detection of irregular environmental behaviour including illegal activity. Our highly generalisable approach, and the common set of features, will enable scientists to unlock previously hidden insights from eco-acoustic data and offers promise as a backbone technology for global collaborative autonomous ecosystem monitoring efforts.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

1
Scientific Reports
○ 3612 papers in training set
Top 0.4%
21.5%
Similar papers in this journal
2
Methods in Ecology and Evolution
○ 176 papers in training set
Top 0.2%
14.8%
Similar papers in this journal
3
Nature Communications
○ 5641 papers in training set
Top 21%
7.7%
Similar papers in this journal
4
Royal Society Open Science
● 214 papers in training set
Top 0.8%
4.8%
Similar papers in this journal
5
Scientific Data
○ 209 papers in training set
Top 0.5%
4.8%
Similar papers in this journal
50% of probability mass above
6
PLOS ONE
● 5266 papers in training set
Top 37%
3.2%
7
Communications Biology
○ 993 papers in training set
Top 6%
3.1%
8
Communications Earth & Environment
○ 14 papers in training set
Top 0.1%
3.1%
9
Patterns
○ 78 papers in training set
Top 0.7%
2.7%
10
eLife
● 5828 papers in training set
Top 43%
2.4%
11
Ecological Informatics
○ 33 papers in training set
Top 0.3%
2.4%
12
PLOS Computational Biology
● 1863 papers in training set
Top 12%
2.3%
13
Remote Sensing in Ecology and Conservation
○ 14 papers in training set
Top 0.1%
2.1%
14
Journal of The Royal Society Interface
● 235 papers in training set
Top 2%
1.7%
15
Current Biology
○ 665 papers in training set
Top 7%
1.5%
16
Journal of Applied Ecology
○ 39 papers in training set
Top 0.8%
1.5%
17
Ecology Letters
○ 135 papers in training set
Top 1%
1.5%
18
iScience
○ 1154 papers in training set
Top 24%
1.3%
19
Molecular Ecology Resources
○ 171 papers in training set
Top 1%
1.3%
20
Ecological Indicators
○ 21 papers in training set
Top 0.4%
1.1%
21
Science Advances
● 1243 papers in training set
Top 28%
1.0%
22
PLOS Biology
● 486 papers in training set
Top 10%
0.9%
23
Journal of Animal Ecology
○ 75 papers in training set
Top 2%
0.8%
24
Frontiers in Ecology and Evolution
○ 69 papers in training set
Top 3%
0.8%
25
Nature Methods
○ 385 papers in training set
Top 7%
0.8%
26
Cell
○ 431 papers in training set
Top 11%
0.6%
27
Proceedings of the National Academy of Sciences
● 2444 papers in training set
Top 45%
0.6%

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.