Integration of ground-based and remote sensing data with deep learning algorithms for mapping habitats in Natura 2000 protected oak forests
Landscape changes caused by climate change require new methods for forest research, analysis, mapping, and monitoring. This study aims to combine ground-based and remote sensing data utilising deep learning techniques to map protected forest habitats and communities within the Natura 2000 network. T...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-03-01
|
| Series: | Basic and Applied Ecology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1439179125000064 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823856841956786176 |
|---|---|
| author | Lucia Čahojová Ivan Jarolímek Barbora Klímová Michal Kollár Michaela Michalková Karol Mikula Aneta A. Ožvat Denisa Slabejová Mária Šibíková |
| author_facet | Lucia Čahojová Ivan Jarolímek Barbora Klímová Michal Kollár Michaela Michalková Karol Mikula Aneta A. Ožvat Denisa Slabejová Mária Šibíková |
| author_sort | Lucia Čahojová |
| collection | DOAJ |
| description | Landscape changes caused by climate change require new methods for forest research, analysis, mapping, and monitoring. This study aims to combine ground-based and remote sensing data utilising deep learning techniques to map protected forest habitats and communities within the Natura 2000 network. The study also seeks to evaluate the accuracy of this approach, specifically in oak-dominated forests, as well as identify the optimal time period within a year for effective habitat identification.Using the specialised software NaturaSat, automated segmentations were performed based on the coordinates of phytosociological relevés and forest strands defined in database. Oak-dominated forest habitats were differentiated solely through multispectral data obtained from Sentinel-2 satellites. A dataset was selected for the training of a deep learning algorithm called the Natural Numerical Network on the basis of the analysis results. This algorithm aims to create a prediction map of habitats dominated by Quercus cerris, which is also known as the relevancy map.Through the utilisation of the Natural Numerical Network, a training accuracy of 95.24% was achieved. Field validation, which was conducted at randomly generated locations within the relevancy map, yielded an accuracy of 98.33%. The most distinguishing differences in band characteristics between the two oak-dominated habitats were observed during the autumn months.This study presents a framework that integrates terrestrial and remote sensing data. This method can serve as a basis for mapping forest habitats and observing changes related to climate change. Moreover, it contributes to the documentation of nature conservation and the mapping of landscapes. |
| format | Article |
| id | doaj-art-911598224fa343b39963a843d8f284fc |
| institution | Kabale University |
| issn | 1439-1791 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Basic and Applied Ecology |
| spelling | doaj-art-911598224fa343b39963a843d8f284fc2025-02-12T05:30:45ZengElsevierBasic and Applied Ecology1439-17912025-03-0183136146Integration of ground-based and remote sensing data with deep learning algorithms for mapping habitats in Natura 2000 protected oak forestsLucia Čahojová0Ivan Jarolímek1Barbora Klímová2Michal Kollár3Michaela Michalková4Karol Mikula5Aneta A. Ožvat6Denisa Slabejová7Mária Šibíková8Institute of Botany, Plant Science and Biodiversity Center, Slovak Academy of Sciences, Dúbravská cesta 9, 845 23 Bratislava, Slovakia; State Nature Conservancy of the Slovak Republic, Little Carpathians Protected Landscape Area, Štúrova 115, Modra 900 01, Slovakia; Corresponding author.Institute of Botany, Plant Science and Biodiversity Center, Slovak Academy of Sciences, Dúbravská cesta 9, 845 23 Bratislava, SlovakiaDepartment of Botany and Zoology, Faculty of Science, Masaryk University, Brno, Czech RepublicDepartment of Mathematics, Slovak University of Technology, Radlinského 11, 810 05 Bratislava, SlovakiaInstitute of Botany, Plant Science and Biodiversity Center, Slovak Academy of Sciences, Dúbravská cesta 9, 845 23 Bratislava, SlovakiaDepartment of Mathematics, Slovak University of Technology, Radlinského 11, 810 05 Bratislava, SlovakiaInstitute of Botany, Plant Science and Biodiversity Center, Slovak Academy of Sciences, Dúbravská cesta 9, 845 23 Bratislava, Slovakia; Department of Mathematics, Slovak University of Technology, Radlinského 11, 810 05 Bratislava, SlovakiaInstitute of Botany, Plant Science and Biodiversity Center, Slovak Academy of Sciences, Dúbravská cesta 9, 845 23 Bratislava, SlovakiaInstitute of Botany, Plant Science and Biodiversity Center, Slovak Academy of Sciences, Dúbravská cesta 9, 845 23 Bratislava, Slovakia; Geobotany Research Center, Karpatské námestie 10/A, Bratislava, SlovakiaLandscape changes caused by climate change require new methods for forest research, analysis, mapping, and monitoring. This study aims to combine ground-based and remote sensing data utilising deep learning techniques to map protected forest habitats and communities within the Natura 2000 network. The study also seeks to evaluate the accuracy of this approach, specifically in oak-dominated forests, as well as identify the optimal time period within a year for effective habitat identification.Using the specialised software NaturaSat, automated segmentations were performed based on the coordinates of phytosociological relevés and forest strands defined in database. Oak-dominated forest habitats were differentiated solely through multispectral data obtained from Sentinel-2 satellites. A dataset was selected for the training of a deep learning algorithm called the Natural Numerical Network on the basis of the analysis results. This algorithm aims to create a prediction map of habitats dominated by Quercus cerris, which is also known as the relevancy map.Through the utilisation of the Natural Numerical Network, a training accuracy of 95.24% was achieved. Field validation, which was conducted at randomly generated locations within the relevancy map, yielded an accuracy of 98.33%. The most distinguishing differences in band characteristics between the two oak-dominated habitats were observed during the autumn months.This study presents a framework that integrates terrestrial and remote sensing data. This method can serve as a basis for mapping forest habitats and observing changes related to climate change. Moreover, it contributes to the documentation of nature conservation and the mapping of landscapes.http://www.sciencedirect.com/science/article/pii/S1439179125000064Earth observationHabitat mappingNatural numerical networkNatura 2000Quercus cerris |
| spellingShingle | Lucia Čahojová Ivan Jarolímek Barbora Klímová Michal Kollár Michaela Michalková Karol Mikula Aneta A. Ožvat Denisa Slabejová Mária Šibíková Integration of ground-based and remote sensing data with deep learning algorithms for mapping habitats in Natura 2000 protected oak forests Basic and Applied Ecology Earth observation Habitat mapping Natural numerical network Natura 2000 Quercus cerris |
| title | Integration of ground-based and remote sensing data with deep learning algorithms for mapping habitats in Natura 2000 protected oak forests |
| title_full | Integration of ground-based and remote sensing data with deep learning algorithms for mapping habitats in Natura 2000 protected oak forests |
| title_fullStr | Integration of ground-based and remote sensing data with deep learning algorithms for mapping habitats in Natura 2000 protected oak forests |
| title_full_unstemmed | Integration of ground-based and remote sensing data with deep learning algorithms for mapping habitats in Natura 2000 protected oak forests |
| title_short | Integration of ground-based and remote sensing data with deep learning algorithms for mapping habitats in Natura 2000 protected oak forests |
| title_sort | integration of ground based and remote sensing data with deep learning algorithms for mapping habitats in natura 2000 protected oak forests |
| topic | Earth observation Habitat mapping Natural numerical network Natura 2000 Quercus cerris |
| url | http://www.sciencedirect.com/science/article/pii/S1439179125000064 |
| work_keys_str_mv | AT luciacahojova integrationofgroundbasedandremotesensingdatawithdeeplearningalgorithmsformappinghabitatsinnatura2000protectedoakforests AT ivanjarolimek integrationofgroundbasedandremotesensingdatawithdeeplearningalgorithmsformappinghabitatsinnatura2000protectedoakforests AT barboraklimova integrationofgroundbasedandremotesensingdatawithdeeplearningalgorithmsformappinghabitatsinnatura2000protectedoakforests AT michalkollar integrationofgroundbasedandremotesensingdatawithdeeplearningalgorithmsformappinghabitatsinnatura2000protectedoakforests AT michaelamichalkova integrationofgroundbasedandremotesensingdatawithdeeplearningalgorithmsformappinghabitatsinnatura2000protectedoakforests AT karolmikula integrationofgroundbasedandremotesensingdatawithdeeplearningalgorithmsformappinghabitatsinnatura2000protectedoakforests AT anetaaozvat integrationofgroundbasedandremotesensingdatawithdeeplearningalgorithmsformappinghabitatsinnatura2000protectedoakforests AT denisaslabejova integrationofgroundbasedandremotesensingdatawithdeeplearningalgorithmsformappinghabitatsinnatura2000protectedoakforests AT mariasibikova integrationofgroundbasedandremotesensingdatawithdeeplearningalgorithmsformappinghabitatsinnatura2000protectedoakforests |