A Benchmark Dataset for Sentinel-2 Based Forest Type Classification in the Siberian Summergreen-Evergreen Forest Transition Zone

Circumboreal forests covering about 30% of global forested areas are undergoing significant changes. In Siberia, global warming may reduce the dominance of summergreen larch forest inducing shifts towards evergreen forest types, specifically in the Eastern Siberian summergreen–evergreen f...

Full description

Saved in:
Bibliographic Details
Main Authors: Femke van Geffen, Ronny Hansch, Begum Demir, Stefan Kruse, Ulrike Herzschuh, Birgit Heim
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10971922/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849340821369782272
author Femke van Geffen
Ronny Hansch
Begum Demir
Stefan Kruse
Ulrike Herzschuh
Birgit Heim
author_facet Femke van Geffen
Ronny Hansch
Begum Demir
Stefan Kruse
Ulrike Herzschuh
Birgit Heim
author_sort Femke van Geffen
collection DOAJ
description Circumboreal forests covering about 30% of global forested areas are undergoing significant changes. In Siberia, global warming may reduce the dominance of summergreen larch forest inducing shifts towards evergreen forest types, specifically in the Eastern Siberian summergreen–evergreen forest transition zone. We create a remote sensing training dataset for summergreen and evergreen forest types from the SiDroForest Sentinel-2 image dataset. This new training dataset informed by expert field knowledge includes nearly two million Sentinel-2 pixels across the early summer, peak summer, and late summer phenophases. We create the equivalent seasonal SiDroTest dataset linked to in situ forest plots for benchmarking the seasonal training dataset. To optimize satellite-based monitoring, we train a random forest classifier on the train dataset to map summergreen and evergreen forest resulting in accuracies of 63% for early summer, 89% for peak summer, and 99% for late summer, with an average accuracy of 82% across all seasons. Feature importance analysis highlights the Sentinel-2 shortwave infrared as crucial for distinguishing forest types in all seasons. Additional key features include the normalized difference vegetation index (NDVI) and the red wavelength region for early summer, shortwave infrared and the visible wavelength region for peak summer, and shortwave infrared, near-infrared and NDVI for late summer. This study provides a benchmarked training dataset for mapping boreal forest types in the Siberian summergreen–evergreen transition zone. The random forest classifier performs best in late summer, leveraging distinct spectral differences between evergreen forests’ greenness and the seasonal coloring of summergreen larch forests.
format Article
id doaj-art-e5857a0bb5be486eaa266c41db9d0991
institution Kabale University
issn 1939-1404
2151-1535
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
spelling doaj-art-e5857a0bb5be486eaa266c41db9d09912025-08-20T03:43:47ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118185091852710.1109/JSTARS.2025.356291210971922A Benchmark Dataset for Sentinel-2 Based Forest Type Classification in the Siberian Summergreen-Evergreen Forest Transition ZoneFemke van Geffen0https://orcid.org/0000-0003-3635-6271Ronny Hansch1https://orcid.org/0000-0002-2936-6765Begum Demir2https://orcid.org/0000-0003-2175-7072Stefan Kruse3https://orcid.org/0000-0003-1107-1958Ulrike Herzschuh4https://orcid.org/0000-0003-0999-1261Birgit Heim5https://orcid.org/0000-0003-2614-9391Polar Terrestrial Environmental Systems Research Section, Alfred-Wegener-Institute Helmholtz Centre for Polar and Marine Research (AWI), Potsdam, GermanyMicrowaves and Radar Institute, German Aerospace Center (DLR), Weßling, GermanyBerlin Institute for the Foundations of Learning and Data (BIFOLD), Technical University Berlin, Berlin, GermanyPolar Terrestrial Environmental Systems Research Section, Alfred-Wegener-Institute Helmholtz Centre for Polar and Marine Research (AWI), Potsdam, GermanyInstitute of Environmental Science and Geography, University of Potsdam, Potsdam, GermanyPolar Terrestrial Environmental Systems Research Section, Alfred-Wegener-Institute Helmholtz Centre for Polar and Marine Research (AWI), Potsdam, GermanyCircumboreal forests covering about 30% of global forested areas are undergoing significant changes. In Siberia, global warming may reduce the dominance of summergreen larch forest inducing shifts towards evergreen forest types, specifically in the Eastern Siberian summergreen–evergreen forest transition zone. We create a remote sensing training dataset for summergreen and evergreen forest types from the SiDroForest Sentinel-2 image dataset. This new training dataset informed by expert field knowledge includes nearly two million Sentinel-2 pixels across the early summer, peak summer, and late summer phenophases. We create the equivalent seasonal SiDroTest dataset linked to in situ forest plots for benchmarking the seasonal training dataset. To optimize satellite-based monitoring, we train a random forest classifier on the train dataset to map summergreen and evergreen forest resulting in accuracies of 63% for early summer, 89% for peak summer, and 99% for late summer, with an average accuracy of 82% across all seasons. Feature importance analysis highlights the Sentinel-2 shortwave infrared as crucial for distinguishing forest types in all seasons. Additional key features include the normalized difference vegetation index (NDVI) and the red wavelength region for early summer, shortwave infrared and the visible wavelength region for peak summer, and shortwave infrared, near-infrared and NDVI for late summer. This study provides a benchmarked training dataset for mapping boreal forest types in the Siberian summergreen–evergreen transition zone. The random forest classifier performs best in late summer, leveraging distinct spectral differences between evergreen forests’ greenness and the seasonal coloring of summergreen larch forests.https://ieeexplore.ieee.org/document/10971922/Boreal forestfeature importancemultitemporalrandom forest (RF) classificationSentinel-2summergreen–evergreen forest transition zone
spellingShingle Femke van Geffen
Ronny Hansch
Begum Demir
Stefan Kruse
Ulrike Herzschuh
Birgit Heim
A Benchmark Dataset for Sentinel-2 Based Forest Type Classification in the Siberian Summergreen-Evergreen Forest Transition Zone
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Boreal forest
feature importance
multitemporal
random forest (RF) classification
Sentinel-2
summergreen–evergreen forest transition zone
title A Benchmark Dataset for Sentinel-2 Based Forest Type Classification in the Siberian Summergreen-Evergreen Forest Transition Zone
title_full A Benchmark Dataset for Sentinel-2 Based Forest Type Classification in the Siberian Summergreen-Evergreen Forest Transition Zone
title_fullStr A Benchmark Dataset for Sentinel-2 Based Forest Type Classification in the Siberian Summergreen-Evergreen Forest Transition Zone
title_full_unstemmed A Benchmark Dataset for Sentinel-2 Based Forest Type Classification in the Siberian Summergreen-Evergreen Forest Transition Zone
title_short A Benchmark Dataset for Sentinel-2 Based Forest Type Classification in the Siberian Summergreen-Evergreen Forest Transition Zone
title_sort benchmark dataset for sentinel 2 based forest type classification in the siberian summergreen evergreen forest transition zone
topic Boreal forest
feature importance
multitemporal
random forest (RF) classification
Sentinel-2
summergreen–evergreen forest transition zone
url https://ieeexplore.ieee.org/document/10971922/
work_keys_str_mv AT femkevangeffen abenchmarkdatasetforsentinel2basedforesttypeclassificationinthesiberiansummergreenevergreenforesttransitionzone
AT ronnyhansch abenchmarkdatasetforsentinel2basedforesttypeclassificationinthesiberiansummergreenevergreenforesttransitionzone
AT begumdemir abenchmarkdatasetforsentinel2basedforesttypeclassificationinthesiberiansummergreenevergreenforesttransitionzone
AT stefankruse abenchmarkdatasetforsentinel2basedforesttypeclassificationinthesiberiansummergreenevergreenforesttransitionzone
AT ulrikeherzschuh abenchmarkdatasetforsentinel2basedforesttypeclassificationinthesiberiansummergreenevergreenforesttransitionzone
AT birgitheim abenchmarkdatasetforsentinel2basedforesttypeclassificationinthesiberiansummergreenevergreenforesttransitionzone
AT femkevangeffen benchmarkdatasetforsentinel2basedforesttypeclassificationinthesiberiansummergreenevergreenforesttransitionzone
AT ronnyhansch benchmarkdatasetforsentinel2basedforesttypeclassificationinthesiberiansummergreenevergreenforesttransitionzone
AT begumdemir benchmarkdatasetforsentinel2basedforesttypeclassificationinthesiberiansummergreenevergreenforesttransitionzone
AT stefankruse benchmarkdatasetforsentinel2basedforesttypeclassificationinthesiberiansummergreenevergreenforesttransitionzone
AT ulrikeherzschuh benchmarkdatasetforsentinel2basedforesttypeclassificationinthesiberiansummergreenevergreenforesttransitionzone
AT birgitheim benchmarkdatasetforsentinel2basedforesttypeclassificationinthesiberiansummergreenevergreenforesttransitionzone