Deep learning enables satellite-based monitoring of large populations of terrestrial mammals across heterogeneous landscape
Abstract New satellite remote sensing and machine learning techniques offer untapped possibilities to monitor global biodiversity with unprecedented speed and precision. These efficiencies promise to reveal novel ecological insights at spatial scales which are germane to the management of population...
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2023-05-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-023-38901-y |
| _version_ | 1827939872834322432 |
|---|---|
| author | Zijing Wu Ce Zhang Xiaowei Gu Isla Duporge Lacey F. Hughey Jared A. Stabach Andrew K. Skidmore J. Grant C. Hopcraft Stephen J. Lee Peter M. Atkinson Douglas J. McCauley Richard Lamprey Shadrack Ngene Tiejun Wang |
| author_facet | Zijing Wu Ce Zhang Xiaowei Gu Isla Duporge Lacey F. Hughey Jared A. Stabach Andrew K. Skidmore J. Grant C. Hopcraft Stephen J. Lee Peter M. Atkinson Douglas J. McCauley Richard Lamprey Shadrack Ngene Tiejun Wang |
| author_sort | Zijing Wu |
| collection | DOAJ |
| description | Abstract New satellite remote sensing and machine learning techniques offer untapped possibilities to monitor global biodiversity with unprecedented speed and precision. These efficiencies promise to reveal novel ecological insights at spatial scales which are germane to the management of populations and entire ecosystems. Here, we present a robust transferable deep learning pipeline to automatically locate and count large herds of migratory ungulates (wildebeest and zebra) in the Serengeti-Mara ecosystem using fine-resolution (38-50 cm) satellite imagery. The results achieve accurate detection of nearly 500,000 individuals across thousands of square kilometers and multiple habitat types, with an overall F1-score of 84.75% (Precision: 87.85%, Recall: 81.86%). This research demonstrates the capability of satellite remote sensing and machine learning techniques to automatically and accurately count very large populations of terrestrial mammals across a highly heterogeneous landscape. We also discuss the potential for satellite-derived species detections to advance basic understanding of animal behavior and ecology. |
| first_indexed | 2024-03-13T09:00:57Z |
| format | Article |
| id | doaj.art-ef9eb43ad8ae4eff943a2e2efeca4f8a |
| institution | Directory Open Access Journal |
| issn | 2041-1723 |
| language | English |
| last_indexed | 2024-03-13T09:00:57Z |
| publishDate | 2023-05-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj.art-ef9eb43ad8ae4eff943a2e2efeca4f8a2023-05-28T11:21:15ZengNature PortfolioNature Communications2041-17232023-05-0114111510.1038/s41467-023-38901-yDeep learning enables satellite-based monitoring of large populations of terrestrial mammals across heterogeneous landscapeZijing Wu0Ce Zhang1Xiaowei Gu2Isla Duporge3Lacey F. Hughey4Jared A. Stabach5Andrew K. Skidmore6J. Grant C. Hopcraft7Stephen J. Lee8Peter M. Atkinson9Douglas J. McCauley10Richard Lamprey11Shadrack Ngene12Tiejun Wang13Department of Natural Resources, Faculty of Geo-Information Science and Earth Observation, University of TwenteLancaster Environment Center, Lancaster UniversitySchool of Computing, University of KentDepartment of Ecology and Evolutionary Biology, Princeton UniversityConservation Ecology Center, Smithsonian National Zoo and Conservation Biology InstituteConservation Ecology Center, Smithsonian National Zoo and Conservation Biology InstituteDepartment of Natural Resources, Faculty of Geo-Information Science and Earth Observation, University of TwenteInstitute of Biodiversity, Animal Health, and Comparative Medicine, University of GlasgowU.S. Army Research Laboratory, Army Research OfficeLancaster Environment Center, Lancaster UniversityDepartment of Ecology, Evolution and Marine Biology, University of CaliforniaDepartment of Natural Resources, Faculty of Geo-Information Science and Earth Observation, University of TwenteWildlife Research and Training InstituteDepartment of Natural Resources, Faculty of Geo-Information Science and Earth Observation, University of TwenteAbstract New satellite remote sensing and machine learning techniques offer untapped possibilities to monitor global biodiversity with unprecedented speed and precision. These efficiencies promise to reveal novel ecological insights at spatial scales which are germane to the management of populations and entire ecosystems. Here, we present a robust transferable deep learning pipeline to automatically locate and count large herds of migratory ungulates (wildebeest and zebra) in the Serengeti-Mara ecosystem using fine-resolution (38-50 cm) satellite imagery. The results achieve accurate detection of nearly 500,000 individuals across thousands of square kilometers and multiple habitat types, with an overall F1-score of 84.75% (Precision: 87.85%, Recall: 81.86%). This research demonstrates the capability of satellite remote sensing and machine learning techniques to automatically and accurately count very large populations of terrestrial mammals across a highly heterogeneous landscape. We also discuss the potential for satellite-derived species detections to advance basic understanding of animal behavior and ecology.https://doi.org/10.1038/s41467-023-38901-y |
| spellingShingle | Zijing Wu Ce Zhang Xiaowei Gu Isla Duporge Lacey F. Hughey Jared A. Stabach Andrew K. Skidmore J. Grant C. Hopcraft Stephen J. Lee Peter M. Atkinson Douglas J. McCauley Richard Lamprey Shadrack Ngene Tiejun Wang Deep learning enables satellite-based monitoring of large populations of terrestrial mammals across heterogeneous landscape Nature Communications |
| title | Deep learning enables satellite-based monitoring of large populations of terrestrial mammals across heterogeneous landscape |
| title_full | Deep learning enables satellite-based monitoring of large populations of terrestrial mammals across heterogeneous landscape |
| title_fullStr | Deep learning enables satellite-based monitoring of large populations of terrestrial mammals across heterogeneous landscape |
| title_full_unstemmed | Deep learning enables satellite-based monitoring of large populations of terrestrial mammals across heterogeneous landscape |
| title_short | Deep learning enables satellite-based monitoring of large populations of terrestrial mammals across heterogeneous landscape |
| title_sort | deep learning enables satellite based monitoring of large populations of terrestrial mammals across heterogeneous landscape |
| url | https://doi.org/10.1038/s41467-023-38901-y |
| work_keys_str_mv | AT zijingwu deeplearningenablessatellitebasedmonitoringoflargepopulationsofterrestrialmammalsacrossheterogeneouslandscape AT cezhang deeplearningenablessatellitebasedmonitoringoflargepopulationsofterrestrialmammalsacrossheterogeneouslandscape AT xiaoweigu deeplearningenablessatellitebasedmonitoringoflargepopulationsofterrestrialmammalsacrossheterogeneouslandscape AT isladuporge deeplearningenablessatellitebasedmonitoringoflargepopulationsofterrestrialmammalsacrossheterogeneouslandscape AT laceyfhughey deeplearningenablessatellitebasedmonitoringoflargepopulationsofterrestrialmammalsacrossheterogeneouslandscape AT jaredastabach deeplearningenablessatellitebasedmonitoringoflargepopulationsofterrestrialmammalsacrossheterogeneouslandscape AT andrewkskidmore deeplearningenablessatellitebasedmonitoringoflargepopulationsofterrestrialmammalsacrossheterogeneouslandscape AT jgrantchopcraft deeplearningenablessatellitebasedmonitoringoflargepopulationsofterrestrialmammalsacrossheterogeneouslandscape AT stephenjlee deeplearningenablessatellitebasedmonitoringoflargepopulationsofterrestrialmammalsacrossheterogeneouslandscape AT petermatkinson deeplearningenablessatellitebasedmonitoringoflargepopulationsofterrestrialmammalsacrossheterogeneouslandscape AT douglasjmccauley deeplearningenablessatellitebasedmonitoringoflargepopulationsofterrestrialmammalsacrossheterogeneouslandscape AT richardlamprey deeplearningenablessatellitebasedmonitoringoflargepopulationsofterrestrialmammalsacrossheterogeneouslandscape AT shadrackngene deeplearningenablessatellitebasedmonitoringoflargepopulationsofterrestrialmammalsacrossheterogeneouslandscape AT tiejunwang deeplearningenablessatellitebasedmonitoringoflargepopulationsofterrestrialmammalsacrossheterogeneouslandscape |