This data was collected using a DJI P4 Multispectral drone near Þórshöfn in Iceland during the late growing season of 2021. The imagery has been processed using Pix4D software. The images are radiometrically corrected using a Mapir reflectance target during the survey. 6-8 ground control points were collected using a Spectra Precision ProMark 120 GPS system with HMRS/VMRS = ~0.5 m error. 12x orthomosaics were generated and resampled to 5 cm spatial resolution and cut to a 300x300 m area extent per site. This dataset includes 6 orthomosaics per site: red, green, blue, red-edge, near-infrared and ndvi band in geotiff format. 12x land cover maps were classified using Random Forest supervised classification. In the land cover layer, the following numbers correspond to: 1 = barren, 2 = low-stature vegetation, and 3 = medium-stature shrub cover. The training and validation data was collected in the field and derived from high-resolution orthomosaics.
Further information about the survey metadata and the processing details is available in the publication where the data was used.
Kodl et al. 2024 - Arctic tundra shrubification can obscure increasing levels of soil erosion in NDVI assessments of land cover derived from satellite imagery. https://doi.org/10.1016/j.rse.2023.113935
Date made available | 25 Jan 2024 |
---|
Publisher | University of St Andrews |
---|
Temporal coverage | 21 Aug 2021 - 2 Sept 2021 |
---|
Date of data production | 23 Aug 2021 - 2 Sept 2021 |
---|
Geospatial point | 66.159837, -15.635009Show on map |
---|
- UAV
- Multispectral
- Soil erosion
- Shrubs
- Tundra
- Iceland
- Barren
- Arctic
- Orthomosaic
- Land cover map