Land cover and polygonal pond classification of the Holocene Terrace on Samoylov Island ...

This dataset contains two sub-datasets: 1) A land cover classification for the Holocene terrace on Samoylov island. The land cover classification is based on a high-resolution near-infrared (NIR) orthomosaic of Samoylov Island, Siberia from summer 2008 (Boike et al., 2012) as well as an existing lan...

Full description

Bibliographic Details
Main Authors: Mirbach, Charlotta, Rehder, ZoƩ, Kutzbach, Lars, Boike, Julia
Format: Dataset
Language:English
Published: PANGAEA 2022
Subjects:
Online Access:https://dx.doi.org/10.1594/pangaea.944071
https://doi.pangaea.de/10.1594/PANGAEA.944071
Description
Summary:This dataset contains two sub-datasets: 1) A land cover classification for the Holocene terrace on Samoylov island. The land cover classification is based on a high-resolution near-infrared (NIR) orthomosaic of Samoylov Island, Siberia from summer 2008 (Boike et al., 2012) as well as an existing land cover classification of part of the Holocene terrace by Muster et al. (2012). 2) A shapefile containing all waterbodies on the Holocene terrace on Samoylov island. Here, we applied minimum bounding geometry to each individual pond in order to calculate its compactness, i.e. the extent to which the ponds approximate a circular shape. The extended land cover classification was produced using ArcMap 10.8 (ESRI Inc., USA). As training data, we used a subset of the existing land cover classification by Muster et al. (2012), and applied a maximum likelihood algorithm to the near-infrared (NIR) orthomosaic of Samoylov island. The land cover classes are: wet tundra, dry tundra, open water, and overgrown water. For the ...