Alps and Carpathians Accuracy Weighted Consensus Land Cover Datasets

This repository provides accuracy-weighted consensus landcover maps for the Alps and Carpathians regions, generated using a Random Forest model, Weighted Votes and the Accuracy Confusion method, as described by Tuanmu and Jetz (2014). The dataset includes 10 classes, comprising 9 probability bands for each landcover class and one class label band. The classes represent the following landcover types: built, crop, forest, shrub, grass, bare, water, wetland, and snow. Each probability band reflects the likelihood of a given pixel belonging to a specific class, while the label band provides the final classified landcover type based on the highest weighted probability.

Dataset Structure:

Probability Bands: 9 bands per pixel, each representing the probability (0–1) of a specific landcover class (built, crop, forest, shrub, grass, bare, water, wetland, snow).

Label Band: A single band providing the final landcover classification for each pixel

Spatial Coverage: Alps and Carpathians

Format: GeoTIFF

Resolution: 10 m