Drone-Derived Remote Sensing Datasets for Herbivore-Accessible Biomass Modelling in the Northern Peneda-Gerês National Park (Portugal), 2020–2024
-
1.
German Centre for Integrative Biodiversity Research
-
2.
Martin Luther University Halle-Wittenberg
Description
Abstract / Description
This dataset collection comprises multiple spatial data layers derived from drone-based remote sensing acquisitions over the northern part of Peneda-Gerês National Park (Portugal), spanning 2020 to 2024. The datasets support modeling of herbivore-accessible biomass by providing high-resolution structural and spectral metrics across a ~20 km² study area. The individual datasets include:
-
A LiDAR point cloud acquired in May and September 2024 (DJI Matrice 350 RTK + Zenmuse L2), with ground point classification (processed in DJI Terra).
-
An RGB orthomosaic from summer 2020 (SenseFly eBee Classic), processed via eMotion and Pix4Dmapper.
- An above-ground height (AGH) surface (2024) derived from the LiDAR point cloud (using ground-classified points and subsequent processing via the WhiteboxTools R package).
-
A canopy cover raster (2024), derived from the LiDAR point cloud using the lidR R package (proportion of first returns ≥ 2 m per pixel).
-
Vertical relative point density (2024) rasters computed for height strata between 0 and 5 m (in 0.5 m bins) from the LiDAR cloud via lidR.
- Topographic data (2015) including Elevation, Slope and Aspect (2015) derived from a SenseFly eBee Classic equipped with a Canon PowerShot ELPH 110 HS-NIR camera and processed in Post Flight Terra 3D.
-
An NDVI raster (May 2023) from drone multispectral imagery (SenseFly eBee X, Sequoia sensor), processed with eMotion / Pix4Dmapper.
- A Land-use/Land-cover (LULC) classification raster obtained through EnMAP-Box 3 within QGIS using a Random Forest classifier using RGB, NDVI and AGH rasters as input.
All rasters and point clouds are georeferenced (CRS EPSG:3763 for Portuguese national grid, or EPSG:32629 for imagery) and cover consistent spatial extent. This collection underlies the analyses and modeling in Zuleger et al. (2025) [https://doi.org/10.1016/j.srs.2025.100302]. Users may utilize the data for structural vegetation modelling, biomass estimation, ecological habitat assessments, or similar remote sensing / ecosystem studies.
Temporal coverage
2020, 2023, 2024 (acquisition dates as per each dataset)
Spatial coverage
Northern Peneda-Gerês National Park, Portugal (approximately 20 km²; parishes Castro Laboreiro and Lamas de Mouro)
Data formats
-
LiDAR point cloud: LAZ with ground / first-return classification
-
Rasters (canopy cover, height, vertical point density, NDVI, RGB orthomosaic): GeoTIFF (floating point or integer as appropriate)
- GeoPackages (Slope and Aspect)
Software / Processing workflows
-
DJI Terra (LiDAR point cloud preprocessing and classification)
-
eMotion, Pix4Dmapper (orthomosaic and NDVI processing)
-
R (lidR package, WhiteboxTools, custom scripts)
-
Additional GIS / raster tools as needed
Authors / Contributors
Henrique Pereira, Luise Quoß, Annika Zuleger, Florian Wolf
Related Publications
Related Datasets
Lidar Point Cloud 2024
|
Title |
File |
Description |
Date |
Software |
Authors |
|
Lidar point cloud of Northern Peneda-Gerês National Park (Portugal), 2024 |
Lidar point cloud derived from airborne LiDAR scanning in May and September 2024 for a northern area in the Peneda-Gerês National Park in the parishes of Castro Laboreiro and Lamas de Mouro, Portugal. The point cloud was obtained with the DJI Matrice 350 RTK with a mounted DJI Zenmuse L2 Camera. The points contain a ground point classification, which is carried out as part of the pre-processing of the point cloud with DJI Terra. The data covers an area of around 35 square kilometres and has a point density of XX points/square metre (CRS: EPSG 3763). |
09.-12. May and 30.09. - 04.10.2024 |
DJI Matrice 350 RTK, DJI Zenmuse L2, DJI Terra |
Henrique Pereira, Luise Quoß, Irina Kalmanova, Ann-Christin Jugnickel, Annika Zuleger
|
Drone RGB 2020
|
Title |
File |
Description |
Date |
Software |
Authors |
|
RGB Orthomosaic of Northern Peneda-Gerês National Park (Portugal), 2020 |
RGB Orthomosaic of June/July 2020 for a northern area in the Peneda-Gerês National Park in the parishes of Castro Laboreiro and Lamas de Mouro, Portugal. The imagery was collected with an eBee Classic and processed with eMotion and Pix4Dmapper. The data covers an area of around 35 square kilometres and has a resolution of 11.4 centimetre (CRS: EPSG 32629). |
24.06., 29.06.and 27.07.2020 |
SenseFly eBee Classic, eMotion, Pix4Dmapper, Canon IXUS 127 HS 4.3 Camera |
Henrique Pereira, Florian Wolf, Luise Quoß
|
Canopy Cover 2024
|
Title |
File |
Description |
Date |
Software |
Authors |
|
Canopy Cover of Northern Peneda-Gerês National Park (Portugal), 2024 |
Canopy cover of May 2024 for a northern area in the Peneda-Gerês National Park in the parishes of Castro Laboreiro and Lamas de Mouro, Portugal. This metric is derived from the LiDAR point cloud (s.a) using R and the lidR package. The metric is calculated by dividing the number of first returns above or equal to two meters by the total number of first returns: cc = (z>=2)/sum(z) per pixel (proportion of points z above two meters per pixel). The data covers an area of around 35 square kilometres and has a resolution of one meter (CRS: EPSG 3763). |
09.-12. May and 30.09. - 04.10.2024 |
DJI Matrice 350 RTK, DJI Zenmuse L2, DJI Terra, R (lidR package) |
Henrique Pereira, Luise Quoß, Annika Zuleger
|
Above Ground Height 2024
|
Title |
File |
Description |
Date |
Software |
Authors |
|
Above Ground Height of Northern Peneda-Gerês National Park (Portugal), 2024 |
Above ground height of May 2024 for a northern area in the Peneda-Gerês National Park in the parishes of Castro Laboreiro and Lamas de Mouro, Portugal. This metric is derived from the LiDAR point cloud (s.a.) using R and the whitebox tools. The ground points classification is part of the raw data post processing in DJI Terra. All pixels with values below or equal to 0.001 meters height were set to zero meters of height. The data covers an area of around 35 square kilometres and has a resolution of ten centimetres (CRS: EPSG 3763). |
09.-12. May and 30.09. - 04.10.2024 |
DJI Matrice 350 RTK, DJI Zenmuse L2, DJI Terra, R (whitebox package) |
Henrique Pereira, Luise Quoß, Annika Zuleger
|
Vertical relative point density 2024
|
Title |
File |
Description |
Date |
Software |
Authors |
|
Vertical relative point density between heights x and y of Northern Peneda-Gerês National Park (Portugal), 2024 |
|
Vertical relative point density between heights x and y of May 2024 for a northern area in the Peneda-Gerês National Park in the parishes of Castro Laboreiro and Lamas de Mouro, Portugal. This metric is derived from the LiDAR point cloud (s.a.) using R and the lidR package. The metric is calculated by dividing the number of points between heights x and y by the total number of points: vrd[x,y] = sum(z>=x & z<=y)/sum(z<=5) (proportion of points z between heights x and y per pixel). The vertical relative point density was calculated for the heights between zero and five meters for half-meter intervals. All pixels with no points smaller than five meters are set to zero. The data covers an area of around 35 square kilometres and has a resolution of one meter (CRS: EPSG 3763). |
09.-12. May and 30.09. - 04.10.2024 |
DJI Matrice 350 RTK, DJI Zenmuse L2, DJI Terra, R (lidR package) |
Henrique Pereira, Luise Quoß, Annika Zuleger
|
NDVI 2023
|
Title |
File |
Description |
Date |
Software |
Authors |
|
NDVI of Northern Peneda-Gerês National Park (Portugal), 2023 |
The normalized difference vegetation index (NDVI) of May 2023 for a northern area in the Peneda-Gerês National Park in the parishes of Castro Laboreiro and Lamas de Mouroin in Portugal. The imagery was collected with an eBee X and processed with eMotion and Pix4Dmapper. The data covers an area of around 35 square kilometres and has a resolution of 25.9 centimetre (CRS: EPSG 32629). This data has contributed to the following publication: DOI |
19.05. - 24.05. 2023 |
SenseFly eBee X, eMotion, Pix4DMapper, Sequoia_4.0_1280x960 |
Henrique Pereira, Leana Meder, Annemarie Röske, Theresa Bräuer, Luise Quoß
|
LULC 2020 - 2024
|
Title |
File |
Description |
Date |
Software |
Authors |
|
Land-use/Land-cover types of Northern Peneda-Gerês National Park (Portugal), 2023 |
Land-cover classification for a northern area in the Peneda-Gerês National Park in the parishes of Castro Laboreiro and Lamas de Mouro, Portugal. The classification was conducted in EnMAP-Box 3 (v3.11.1; Jakimow et al., 2023) within QGIS using a Random Forest classifier (100 trees, 3-fold cross-validation). The input raster combined RGB (11.4 cm), NDVI, and above-ground height layers, resampled to 1 × 1 meter resolution. Nine land-cover classes—high shrub, low shrub, oak forest, pine forest, rock, agriculture, urban, water, and roads—were delineated based on field knowledge and aerial imagery. The data covers an area of around 35 square kilometres (CRS: EPSG 3763). |
2020 - 2024 |
EnMAP-Box 3
|
Annika Zuleger, Luise Quoss
|
Topographic data 2015
|
Title |
File |
Description |
Date |
Software |
Authors |
|
Digital surface and terrain models of Northern Peneda-Gerês National Park (Portugal), 2015 |
201508_full_largegrid_2_5_dsm_3763.tif 201508_aspect_largegrid_2_5.gpkg 201508_slope_largegrid_2_5.gpkg
|
High-resolution digital surface and terrain models (DSM and DTM) derived from drone imagery collected in 2015 over a 22.8 km² area in the northern Peneda-Gerês National Park, Portugal. The imagery was acquired with a SenseFly eBee Classic equipped with a Canon PowerShot ELPH 110 HS-NIR camera at 13.4 cm resolution and processed in Post Flight Terra 3D (v3.4.46, Pix4D) using SIFT-based keypoint matching and multi-view stereo reconstruction. A 2.5 m resolution DSM and DTM were produced, from which slope and aspect layers were derived in QGIS (v3.24.2). |
2015 |
SenseFly eBee Classic, Canon PowerShot ELPH 110 HS-NIR, Post Flight Terra 3D (v3.4.46, Pix4D), QGIS (v3.24.2)
|
Henrique Pereira, Florian Wolf
|
Files
Reclassified_LULC_smooth.tif
Files
(27.9 GB)
| Name | Size | Download all |
|---|---|---|
|
md5:c1bb949ac7ac934e3e37e5f0c6bfe8f3
|
15.7 MB | Download |
|
md5:53a827021ab47efbc30f7cc665f92768
|
13.8 MB | Preview Download |
|
md5:ad5c62eccc0089c90b3234b239bfea3b
|
15.9 MB | Download |
|
md5:40284a87f1b20f7bb0a7cf4c57615b20
|
4.2 GB | Preview Download |
|
md5:d5d8c79b661dbf3e160c30ef19181569
|
14.8 GB | Preview Download |
|
md5:0eee460e8672abe45aaed7515fc58966
|
25.1 MB | Preview Download |
|
md5:b10db669828ebc36857ad3d965390293
|
36.8 MB | Preview Download |
|
md5:40aa2b69dcb49bca1d41960d6316573d
|
29.2 MB | Preview Download |
|
md5:50fc424ab33f6980fb7a035419b50139
|
47.7 MB | Preview Download |
|
md5:852c6a5fa7269f48d554fe4a59bbd286
|
23.5 MB | Preview Download |
|
md5:104461475465f686b830a940b9649d12
|
7.2 GB | Preview Download |
|
md5:d3911155c2c8f128a14a0ae4d13973f8
|
1.5 GB | Preview Download |
|
md5:5850e59bd28f4adb892117b07c45a4ac
|
100.6 MB | Preview Download |
Additional details
Related works
- Is source of
- Dataset: 10.5281/zenodo.17347321 (DOI)
- Is supplement to
- Journal article: 10.1016/j.srs.2025.100302 (DOI)