High-resolution land cover classification for Freiburg im Breisgau (Germany)

Land cover classifcation for Freiburg im Breisgau (Baden-Württemberg, Germany) with a spatial resolution of 1 m x 1 m in geotiff format based on data from 2019.

Separates between:

ID
Land cover class
Available in 'freiburg_landcover_1m_with_trees.tif'
Available in 'freiburg_landcover_1m_without_trees.tif'

1
paved / sealed
✔︎
✔︎

2
buildings
✔︎
✔︎

3
evergreen trees
✔︎

4
deciduous trees
✔︎

5
low vegetation / soil
✔︎
✔︎

6
bare soil
✔︎
✔︎

7
open water
✔︎
✔︎

The land cover product has been created as part of the Intelligence for Cities (I4C) project for urban climate modelling. Details are described in Briegel et al. 2023. The land cover data was generated by combining multiple sources including:

Open Street Map (OSM)

Copernicus Urban Atlas (Source: https://land.copernicus.eu/en/products/urban-atlas)

Building Outline Data and Lidar data (Source: City of Freiburg - https://www.freiburg.de/pb/-/205348/stabsstelle-geodatenmanagement/oe6008924).

This repository includes two types of data:

Land cover classes with trees ("freiburg_landcover_1m_with_trees.tif") gives the land cover seen from nadir and includes a class for tree canopy. Land cover underneath trees is not considered. The dataset distinguishes between evergreen and deciduous trees.

Land cover classes without trees ("freiburg_landcover_1m_without_trees.tif") represent the surface area below the tree canopy.

The classes in both datasets are given as numbers and the allocation to the corresponding classes can be also found in Readme_Land_cover.txt.

Preferred citation:

Briegel F, Makansi O, Brox T, Matzarakis A, Christen A, 2023: Modelling long-term thermal comfort conditions in urban environments using a deep convolutional encoder-decoder as a computational shortcut. Urban Climate, 47, 101359.✔︎