Integration of UAV-sensed features using machine learning methods to assess species richness in wet grassland ecosystems
Wet grasslands are crucial components of terrestrial ecosystems, known for their biodiversity and provision of ecosystem services such as flood attenuation and carbon sequestration. Given their ecological significance, monitoring biodiversity within these landscapes is of utmost importance for effec...
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-11-01
|
| Series: | Ecological Informatics |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1574954124003558 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850266154435608576 |
|---|---|
| author | Clara Oliva Gonçalves Bazzo Bahareh Kamali Murilo dos Santos Vianna Dominik Behrend Hubert Hueging Inga Schleip Paul Mosebach Almut Haub Axel Behrendt Thomas Gaiser |
| author_facet | Clara Oliva Gonçalves Bazzo Bahareh Kamali Murilo dos Santos Vianna Dominik Behrend Hubert Hueging Inga Schleip Paul Mosebach Almut Haub Axel Behrendt Thomas Gaiser |
| author_sort | Clara Oliva Gonçalves Bazzo |
| collection | DOAJ |
| description | Wet grasslands are crucial components of terrestrial ecosystems, known for their biodiversity and provision of ecosystem services such as flood attenuation and carbon sequestration. Given their ecological significance, monitoring biodiversity within these landscapes is of utmost importance for effective conservation and management strategies. This study, conducted in a wet grassland of Brandenburg, Germany, utilized unmanned aerial vehicles (UAVs) to facilitate the estimation of species richness by the integration of remotely sensed canopy features such as canopy height (CH), spectral data (Vegetation Indices, VI), and texture features (Gray-Level Co-occurrence Matrix, GLCM) using two machine learning methods (Partial Least Square regression (PLS) and Random Forest (RF)). Data was collected over two growing seasons under three different grass cutting regimes, employing multispectral sensors to capture detailed vegetation characteristics. The analysis revealed that the performance of the machine learning methods varied with the feature combinations. Models combining VI and GLCM features demonstrated the highest predictive accuracy, particularly in frequently cut grasslands, as indicated by higher R2 values (up to 0.52) and lower root mean square errors (rRMSE as low as 34.9 %). RF models generally outperformed PLS models across different feature sets, with the CH + VI + GLCM combination yielding the best results. These findings underscore the potential of spectral and textural data to effectively capture the ecological dynamics of wet grasslands, providing valuable insights into biodiversity patterns. |
| format | Article |
| id | doaj-art-d5a6d11eefc04500a3cf41a7b27bdcec |
| institution | OA Journals |
| issn | 1574-9541 |
| language | English |
| publishDate | 2024-11-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Ecological Informatics |
| spelling | doaj-art-d5a6d11eefc04500a3cf41a7b27bdcec2025-08-20T01:54:15ZengElsevierEcological Informatics1574-95412024-11-018310281310.1016/j.ecoinf.2024.102813Integration of UAV-sensed features using machine learning methods to assess species richness in wet grassland ecosystemsClara Oliva Gonçalves Bazzo0Bahareh Kamali1Murilo dos Santos Vianna2Dominik Behrend3Hubert Hueging4Inga Schleip5Paul Mosebach6Almut Haub7Axel Behrendt8Thomas Gaiser9Institute of Crop Science and Resource Conservation (INRES), University of Bonn, Katzenburgweg 5, 53115 Bonn, Germany; Corresponding author.Institute of Crop Science and Resource Conservation (INRES), University of Bonn, Katzenburgweg 5, 53115 Bonn, GermanyInstitute of Crop Science and Resource Conservation (INRES), University of Bonn, Katzenburgweg 5, 53115 Bonn, Germany; Agrosphere Institute, IBG-3, Forschungszentrum Jülich GmbH (FZJ), Juelich, GermanyInstitute of Crop Science and Resource Conservation (INRES), University of Bonn, Katzenburgweg 5, 53115 Bonn, GermanyInstitute of Crop Science and Resource Conservation (INRES), University of Bonn, Katzenburgweg 5, 53115 Bonn, GermanyFaculty of Landscape Management and Nature Conservation, Eberswalde University for Sustainable Development, 16225 Eberswalde, GermanyFaculty of Landscape Management and Nature Conservation, Eberswalde University for Sustainable Development, 16225 Eberswalde, GermanyFaculty of Landscape Management and Nature Conservation, Eberswalde University for Sustainable Development, 16225 Eberswalde, GermanyLeibniz Centre for Agricultural Landscape Research (ZALF), Gutshof 7, 14641 Paulinenaue, GermanyInstitute of Crop Science and Resource Conservation (INRES), University of Bonn, Katzenburgweg 5, 53115 Bonn, GermanyWet grasslands are crucial components of terrestrial ecosystems, known for their biodiversity and provision of ecosystem services such as flood attenuation and carbon sequestration. Given their ecological significance, monitoring biodiversity within these landscapes is of utmost importance for effective conservation and management strategies. This study, conducted in a wet grassland of Brandenburg, Germany, utilized unmanned aerial vehicles (UAVs) to facilitate the estimation of species richness by the integration of remotely sensed canopy features such as canopy height (CH), spectral data (Vegetation Indices, VI), and texture features (Gray-Level Co-occurrence Matrix, GLCM) using two machine learning methods (Partial Least Square regression (PLS) and Random Forest (RF)). Data was collected over two growing seasons under three different grass cutting regimes, employing multispectral sensors to capture detailed vegetation characteristics. The analysis revealed that the performance of the machine learning methods varied with the feature combinations. Models combining VI and GLCM features demonstrated the highest predictive accuracy, particularly in frequently cut grasslands, as indicated by higher R2 values (up to 0.52) and lower root mean square errors (rRMSE as low as 34.9 %). RF models generally outperformed PLS models across different feature sets, with the CH + VI + GLCM combination yielding the best results. These findings underscore the potential of spectral and textural data to effectively capture the ecological dynamics of wet grasslands, providing valuable insights into biodiversity patterns.http://www.sciencedirect.com/science/article/pii/S1574954124003558Remote sensingVegetationEcological monitoringMultispectralCanopy heightTexture |
| spellingShingle | Clara Oliva Gonçalves Bazzo Bahareh Kamali Murilo dos Santos Vianna Dominik Behrend Hubert Hueging Inga Schleip Paul Mosebach Almut Haub Axel Behrendt Thomas Gaiser Integration of UAV-sensed features using machine learning methods to assess species richness in wet grassland ecosystems Ecological Informatics Remote sensing Vegetation Ecological monitoring Multispectral Canopy height Texture |
| title | Integration of UAV-sensed features using machine learning methods to assess species richness in wet grassland ecosystems |
| title_full | Integration of UAV-sensed features using machine learning methods to assess species richness in wet grassland ecosystems |
| title_fullStr | Integration of UAV-sensed features using machine learning methods to assess species richness in wet grassland ecosystems |
| title_full_unstemmed | Integration of UAV-sensed features using machine learning methods to assess species richness in wet grassland ecosystems |
| title_short | Integration of UAV-sensed features using machine learning methods to assess species richness in wet grassland ecosystems |
| title_sort | integration of uav sensed features using machine learning methods to assess species richness in wet grassland ecosystems |
| topic | Remote sensing Vegetation Ecological monitoring Multispectral Canopy height Texture |
| url | http://www.sciencedirect.com/science/article/pii/S1574954124003558 |
| work_keys_str_mv | AT claraolivagoncalvesbazzo integrationofuavsensedfeaturesusingmachinelearningmethodstoassessspeciesrichnessinwetgrasslandecosystems AT baharehkamali integrationofuavsensedfeaturesusingmachinelearningmethodstoassessspeciesrichnessinwetgrasslandecosystems AT murilodossantosvianna integrationofuavsensedfeaturesusingmachinelearningmethodstoassessspeciesrichnessinwetgrasslandecosystems AT dominikbehrend integrationofuavsensedfeaturesusingmachinelearningmethodstoassessspeciesrichnessinwetgrasslandecosystems AT huberthueging integrationofuavsensedfeaturesusingmachinelearningmethodstoassessspeciesrichnessinwetgrasslandecosystems AT ingaschleip integrationofuavsensedfeaturesusingmachinelearningmethodstoassessspeciesrichnessinwetgrasslandecosystems AT paulmosebach integrationofuavsensedfeaturesusingmachinelearningmethodstoassessspeciesrichnessinwetgrasslandecosystems AT almuthaub integrationofuavsensedfeaturesusingmachinelearningmethodstoassessspeciesrichnessinwetgrasslandecosystems AT axelbehrendt integrationofuavsensedfeaturesusingmachinelearningmethodstoassessspeciesrichnessinwetgrasslandecosystems AT thomasgaiser integrationofuavsensedfeaturesusingmachinelearningmethodstoassessspeciesrichnessinwetgrasslandecosystems |