The potential of combining satellite and airborne remote sensing data for habitat classification and monitoring in forest landscapes
Mapping and monitoring of habitats are requirements for protecting biodiversity. In this study, we investigated the benefit of combining airborne (laser scanning, image-based point clouds) and satellite-based (Sentinel 1 and 2) data for habitat classification. We used a two level random forest 10-fo...
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2023-03-01
|
Series: | International Journal of Applied Earth Observations and Geoinformation |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S1569843222003193 |
_version_ | 1811163479996891136 |
---|---|
author | Anna Iglseder Markus Immitzer Alena Dostálová Andreas Kasper Norbert Pfeifer Christoph Bauerhansl Stefan Schöttl Markus Hollaus |
author_facet | Anna Iglseder Markus Immitzer Alena Dostálová Andreas Kasper Norbert Pfeifer Christoph Bauerhansl Stefan Schöttl Markus Hollaus |
author_sort | Anna Iglseder |
collection | DOAJ |
description | Mapping and monitoring of habitats are requirements for protecting biodiversity. In this study, we investigated the benefit of combining airborne (laser scanning, image-based point clouds) and satellite-based (Sentinel 1 and 2) data for habitat classification. We used a two level random forest 10-fold leave-location-out cross-validation workflow to model Natura 2000 forest and grassland habitat types on a 10 m pixel scale at two study sites in Vienna, Austria. We showed that models using combined airborne and satellite-based remote sensing data perform significantly better for forests than airborne or satellite-based data alone. For frequently occurring classes, we reached class accuracies with F1-scores from 0.60 to 0.87. We identified clear difficulties of correctly assigning rare classes with model-based classification. Finally, we demonstrated the potential of the workflow to identify errors in reference data and point to the opportunities for integration in habitat mapping and monitoring. |
first_indexed | 2024-04-10T15:06:11Z |
format | Article |
id | doaj.art-366e7a792da142f9876fd7268fb9cae3 |
institution | Directory Open Access Journal |
issn | 1569-8432 |
language | English |
last_indexed | 2024-04-10T15:06:11Z |
publishDate | 2023-03-01 |
publisher | Elsevier |
record_format | Article |
series | International Journal of Applied Earth Observations and Geoinformation |
spelling | doaj.art-366e7a792da142f9876fd7268fb9cae32023-02-15T04:27:24ZengElsevierInternational Journal of Applied Earth Observations and Geoinformation1569-84322023-03-01117103131The potential of combining satellite and airborne remote sensing data for habitat classification and monitoring in forest landscapesAnna Iglseder0Markus Immitzer1Alena Dostálová2Andreas Kasper3Norbert Pfeifer4Christoph Bauerhansl5Stefan Schöttl6Markus Hollaus7Department of Geodesy and Geoinformation, TU Wien, 1040 Vienna, Austria; Corresponding author.University of Natural Resources and Life Sciences, Vienna (BOKU), Institute of Geomatics, Peter-Jordan-Straße 82, 1190 Vienna, AustriaDepartment of Geodesy and Geoinformation, TU Wien, 1040 Vienna, AustriaMunicipal Department 22 – Environmental Protection in Vienna (MA22), Dresdner Straße 45, 1200 Vienna, AustriaDepartment of Geodesy and Geoinformation, TU Wien, 1040 Vienna, AustriaAustrian Research Centre for Forests, Unit Remote Sensing, Seckendorff-Gudent-Weg 8, 1130 Vienna, AustriaAustrian Research Centre for Forests, Unit Remote Sensing, Seckendorff-Gudent-Weg 8, 1130 Vienna, AustriaDepartment of Geodesy and Geoinformation, TU Wien, 1040 Vienna, AustriaMapping and monitoring of habitats are requirements for protecting biodiversity. In this study, we investigated the benefit of combining airborne (laser scanning, image-based point clouds) and satellite-based (Sentinel 1 and 2) data for habitat classification. We used a two level random forest 10-fold leave-location-out cross-validation workflow to model Natura 2000 forest and grassland habitat types on a 10 m pixel scale at two study sites in Vienna, Austria. We showed that models using combined airborne and satellite-based remote sensing data perform significantly better for forests than airborne or satellite-based data alone. For frequently occurring classes, we reached class accuracies with F1-scores from 0.60 to 0.87. We identified clear difficulties of correctly assigning rare classes with model-based classification. Finally, we demonstrated the potential of the workflow to identify errors in reference data and point to the opportunities for integration in habitat mapping and monitoring.http://www.sciencedirect.com/science/article/pii/S1569843222003193Habitat MappingNatura 2000Airborne Laser ScanningSentinel-1Sentinel-2Random Forest |
spellingShingle | Anna Iglseder Markus Immitzer Alena Dostálová Andreas Kasper Norbert Pfeifer Christoph Bauerhansl Stefan Schöttl Markus Hollaus The potential of combining satellite and airborne remote sensing data for habitat classification and monitoring in forest landscapes International Journal of Applied Earth Observations and Geoinformation Habitat Mapping Natura 2000 Airborne Laser Scanning Sentinel-1 Sentinel-2 Random Forest |
title | The potential of combining satellite and airborne remote sensing data for habitat classification and monitoring in forest landscapes |
title_full | The potential of combining satellite and airborne remote sensing data for habitat classification and monitoring in forest landscapes |
title_fullStr | The potential of combining satellite and airborne remote sensing data for habitat classification and monitoring in forest landscapes |
title_full_unstemmed | The potential of combining satellite and airborne remote sensing data for habitat classification and monitoring in forest landscapes |
title_short | The potential of combining satellite and airborne remote sensing data for habitat classification and monitoring in forest landscapes |
title_sort | potential of combining satellite and airborne remote sensing data for habitat classification and monitoring in forest landscapes |
topic | Habitat Mapping Natura 2000 Airborne Laser Scanning Sentinel-1 Sentinel-2 Random Forest |
url | http://www.sciencedirect.com/science/article/pii/S1569843222003193 |
work_keys_str_mv | AT annaiglseder thepotentialofcombiningsatelliteandairborneremotesensingdataforhabitatclassificationandmonitoringinforestlandscapes AT markusimmitzer thepotentialofcombiningsatelliteandairborneremotesensingdataforhabitatclassificationandmonitoringinforestlandscapes AT alenadostalova thepotentialofcombiningsatelliteandairborneremotesensingdataforhabitatclassificationandmonitoringinforestlandscapes AT andreaskasper thepotentialofcombiningsatelliteandairborneremotesensingdataforhabitatclassificationandmonitoringinforestlandscapes AT norbertpfeifer thepotentialofcombiningsatelliteandairborneremotesensingdataforhabitatclassificationandmonitoringinforestlandscapes AT christophbauerhansl thepotentialofcombiningsatelliteandairborneremotesensingdataforhabitatclassificationandmonitoringinforestlandscapes AT stefanschottl thepotentialofcombiningsatelliteandairborneremotesensingdataforhabitatclassificationandmonitoringinforestlandscapes AT markushollaus thepotentialofcombiningsatelliteandairborneremotesensingdataforhabitatclassificationandmonitoringinforestlandscapes AT annaiglseder potentialofcombiningsatelliteandairborneremotesensingdataforhabitatclassificationandmonitoringinforestlandscapes AT markusimmitzer potentialofcombiningsatelliteandairborneremotesensingdataforhabitatclassificationandmonitoringinforestlandscapes AT alenadostalova potentialofcombiningsatelliteandairborneremotesensingdataforhabitatclassificationandmonitoringinforestlandscapes AT andreaskasper potentialofcombiningsatelliteandairborneremotesensingdataforhabitatclassificationandmonitoringinforestlandscapes AT norbertpfeifer potentialofcombiningsatelliteandairborneremotesensingdataforhabitatclassificationandmonitoringinforestlandscapes AT christophbauerhansl potentialofcombiningsatelliteandairborneremotesensingdataforhabitatclassificationandmonitoringinforestlandscapes AT stefanschottl potentialofcombiningsatelliteandairborneremotesensingdataforhabitatclassificationandmonitoringinforestlandscapes AT markushollaus potentialofcombiningsatelliteandairborneremotesensingdataforhabitatclassificationandmonitoringinforestlandscapes |