Data

UAV Hyperspectral Moss Health and Lichen Classification Maps at 30 m and 70 m AGL, ASPA 135, Antarctica (Jan–Feb 2023)

Australian Antarctic Division

Sandino, J., Barthelemy, J., Doshi, A., Randall, K., Robinson, S., Bollard, B. and Gonzalez, F. ; SANDINO, JUAN ; BARTHELEMY, JOHAN ; DOSHI, ASHRAY ; RANDALL, KRYSTAL ; ROBINSON, SHARON ; BOLLARD, BARBARA ; GONZALEZ, FELIPE

Viewed: [[ro.stat.viewed]] Cited: [[ro.stat.cited]] Accessed: [[ro.stat.accessed]]

ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rfr_id=info%3Asid%2FANDS&rft_id=info:doi10.26179/r0e3-2j72&rft.title=UAV Hyperspectral Moss Health and Lichen Classification Maps at 30 m and 70 m AGL, ASPA 135, Antarctica (Jan–Feb 2023)&rft.identifier=10.26179/r0e3-2j72&rft.publisher=Australian Antarctic Data Centre&rft.description=This dataset contains georeferenced vegetation classification (segmentation) maps derived from drone‑mounted hyperspectral imaging (HSI) surveys acquired 28 Jan–02 Feb 2023 over ~6.5 ha of transect areas in Antarctic Specially Protected Area (ASPA) No. 135 (Windmill Islands), Antarctica. Flights were conducted at ~30 m and ~70 m above ground level (AGL), yielding ~6 cm and ~10 cm ground sampling distance (GSD) mosaics. Hyperspectral data (core) and concurrent RGB imagery (supplementary) were radiometrically calibrated, orthorectified using Global Navigation Satellite System (GNSS) Real‑Time Kinematic (RTK) ground control, and processed with machine‑learning classification workflows to map moss health condition (Healthy, Stressed, Moribund), Lichen, and Non‑vegetated surfaces.Progress Code: completedStatement: Limitations - Temporal mismatch: Hyperspectral data (used for the classification maps) and ancillary RGB imagery were collected on different days. A snow blizzard occurred between acquisitions. - Surface change: Portions of the mapped transects that were snow‑free during hyperspectral capture were partially snow‑covered in the later RGB imagery (and field photos), leading to visual mismatches. - Interpretation impact: Snow cover can obscure vegetation and alter reflectance; some pixels labelled Non‑vegetated may include snow, and comparisons t&rft.creator=Sandino, J., Barthelemy, J., Doshi, A., Randall, K., Robinson, S., Bollard, B. and Gonzalez, F. &rft.creator=SANDINO, JUAN &rft.creator=BARTHELEMY, JOHAN &rft.creator=DOSHI, ASHRAY &rft.creator=RANDALL, KRYSTAL &rft.creator=ROBINSON, SHARON &rft.creator=BOLLARD, BARBARA &rft.creator=GONZALEZ, FELIPE &rft.date=2025&rft.coverage=westlimit=110.5385254; southlimit=-66.2823365; eastlimit=110.5399896; northlimit=-66.2818034&rft.coverage=westlimit=110.5385254; southlimit=-66.2823365; eastlimit=110.5399896; northlimit=-66.2818034&rft_rights=These data are publicly available for download from the provided URL.&rft_rights=Attribution 4.0 International (CC BY 4.0) https://creativecommons.org/licenses/by/4.0/legalcode&rft_rights=This data set conforms to the CCBY Attribution License (http://creativecommons.org/licenses/by/4.0/). Please follow instructions listed in the citation reference provided at http://data.aad.gov.au/aadc/metadata/citation.cfm?entry_id=AAS_4628_UAS_ASPA135_AERIAL_HSISEG_MOSSLICHEN_2023 when using these data.&rft_rights=This metadata record is publicly available.&rft_subject=imageryBaseMapsEarthCover&rft_subject=biota&rft_subject=EARTH SCIENCE > BIOSPHERE > VEGETATION > VEGETATION COVER&rft_subject=EARTH SCIENCE > BIOLOGICAL CLASSIFICATION > PLANTS > MOSSES/HORNWORTS/LIVERWORTS&rft_subject=EARTH SCIENCE > BIOLOGICAL CLASSIFICATION > FUNGI > LICHENS&rft_subject=EARTH SCIENCE > MODELS > MACHINE LEARNING MODELS&rft_subject=EARTH SCIENCE > CRYOSPHERE > SNOW/ICE > SNOW COVER&rft_subject=EARTH SCIENCE > LAND SURFACE > LAND USE/LAND COVER&rft_subject=EARTH SCIENCE > BIOSPHERE > ECOLOGICAL DYNAMICS > ECOSYSTEM FUNCTIONS&rft_subject=EARTH SCIENCE > SPECTRAL/ENGINEERING > INFRARED WAVELENGTHS > INFRARED IMAGERY&rft_subject=HYPERSPECTRAL IMAGERY&rft_subject=IMAGE CLASSIFICATION&rft_subject=EARTH SCIENCE SERVICES > DATA ANALYSIS AND VISUALIZATION > MACHINE LEARNING > MODEL RESULTS&rft_subject=CAMERAS&rft_subject=ROVS > Remotely Operated Vehicle&rft_subject=UAV > Unmanned Aerial Vehicle&rft_subject=AMD/AU&rft_subject=AMD&rft_subject=CEOS&rft_subject=GEOGRAPHIC REGION > POLAR&rft_subject=CONTINENT > ANTARCTICA > WILKES LAND > BAILEY PENINSULA&rft_subject=CONTINENT > ANTARCTICA > WINDMILL ISLANDS&rft.type=dataset&rft.language=English Access the data

Access data via landing page
http://data.aad.gov.au...

Cite Saved to MyRDA Save to MyRDA

Licence & Rights:

Open Licence view details

Attribution 4.0 International (CC BY 4.0)
https://creativecommons.org/licenses/by/4.0/legalcode

These data are publicly available for download from the provided URL.

This data set conforms to the CCBY Attribution License (http://creativecommons.org/licenses/by/4.0/). Please follow instructions listed in the citation reference provided at http://data.aad.gov.au/aadc/metadata/citation.cfm?entry_id=AAS_4628_UAS_ASPA135_AERIAL_HSISEG_MOSSLICHEN_2023 when using these data.

This metadata record is publicly available.

Access:

Other

Full description

This dataset contains georeferenced vegetation classification (segmentation) maps derived from drone‑mounted hyperspectral imaging (HSI) surveys acquired 28 Jan–02 Feb 2023 over ~6.5 ha of transect areas in Antarctic Specially Protected Area (ASPA) No. 135 (Windmill Islands), Antarctica. Flights were conducted at ~30 m and ~70 m above ground level (AGL), yielding ~6 cm and ~10 cm ground sampling distance (GSD) mosaics. Hyperspectral data (core) and concurrent RGB imagery (supplementary) were radiometrically calibrated, orthorectified using Global Navigation Satellite System (GNSS) Real‑Time Kinematic (RTK) ground control, and processed with machine‑learning classification workflows to map moss health condition (Healthy, Stressed, Moribund), Lichen, and Non‑vegetated surfaces.

Lineage

Progress Code: completed

Statement: Limitations - Temporal mismatch: Hyperspectral data (used for the classification maps) and ancillary RGB imagery were collected on different days. A snow blizzard occurred between acquisitions. - Surface change: Portions of the mapped transects that were snow‑free during hyperspectral capture were partially snow‑covered in the later RGB imagery (and field photos), leading to visual mismatches. - Interpretation impact: Snow cover can obscure vegetation and alter reflectance; some pixels labelled Non‑vegetated may include snow, and comparisons t

Notes

Purpose
The dataset reproduces the products reported in Drone hyperspectral imaging and artificial intelligence for monitoring moss and lichen in Antarctica (doi:10.1038/s41598-025-11535-4) and includes: - GeoTIFF and ENVI classification rasters for each altitude; - Class legend; - Transect boundary files; - Perclass area statistics; - Summary tables used in the publication; - And model performance metrics; Maps are provided in WGS 84 / UTM Zone 49S, EPSG:32749. See the included README for file structure, processing steps, and terminology. These materials are intended to enable long-term reuse for ecological mapping, machine learning training/benchmarking, and change monitoring studies.

Data time period: 2023-01-28 to 2023-02-02