This repository contains tree canopy cover loss information between January 2018 and April 2021 for Germany at monthly resolution. The analysis is based on monthly image composites of the disturbance index (DI) derived from Sentinel-2 and Landsat-8 time series. Deviations from a 2017 reference median DI image exceeding a threshold are recorded as losses.

The dataset is available as 8-bit GeoTIFF in the ETRS89 / LAEA Europe projection (EPSG:3035) with a pixel size of 10 m. Aggregated results at district level are available as annual statistics per forest type.

The grey levels in the raster file indicate the month of tree canopy cover loss. Their meaning is as follows:
0 = intact forest
1 = January 2018
2 = February 2018
.
.
.
40 = April 2021
100 = non-forested

The vector data contains following attributes:
Name_2 = district name
TYPE_2 = district type 
Name_1 = federal state
nr = number
lk_area = district area (ha)
forestr = initial forest area per district (ha)
decall = losses in deciduous forest from January 2018 to April 2021 (ha)
dec2018 = losses in deciduous forest in 2018 (ha)
dec2019 = losses in deciduous forest in 2019 (ha)
dec2020 = losses in deciduous forest in 2020 (ha)
dec2021 = losses in deciduous forest in 2021 (January to April) (ha)
conall = losses in coniferous forest from January 2018 to April 2021 (ha)
con2018 = losses in coniferous forest in 2018 (ha)
con2019 = losses in coniferous forest in 2019 (ha)
con2020 = losses in coniferous forest in 2020 (ha)
con2021 = losses in coniferous forest in 2021 (January to April) (ha)
allall = losses in all forest types from January 2018 to April 2021 (ha)
all2018 = losses in all forest types in 2018 (ha)
all2019 = losses in all forest types in 2019 (ha)
all2020 = losses in all forest types in 2020 (ha)
all2021 = losses in all forest types in 2021 (January to April) (ha)
p_frstr = initial forest area per district (%)
p_decll = losses in deciduous forest from January 2018 to April 2021 (%)
p_d2018 = losses in deciduous forest in 2018 (%)
p_d2019 = losses in deciduous forest in 2019 (%)
p_d2020 = losses in deciduous forest in 2020 (%)
p_d2021 = losses in deciduous forest in 2021 (January to April) (%)
p_conll = losses in coniferous forest from January 2018 to April 2021 (%)
p_c2018 = losses in coniferous forest in 2018 (%)
p_c2019 = losses in coniferous forest in 2019 (%)
p_c2020 = losses in coniferous forest in 2020 (%)
p_c2021 = losses in coniferous forest in 2021 (January to April) (%)
p_allll = losses in all forest types from January 2018 to April 2021 (%)
p_l2018 = losses in all forest types in 2018 (%)
p_l2019 = losses in all forest types in 2019 (%)
p_l2020 = losses in all forest types in 2020 (%)
p_l2021 = losses in all forest types in 2021 (January to April) (%)

An open access publication with all technical details can be found here:
Thonfeld, F., Gessner, U., Holzwarth, S., Kriese, J., da Ponte, E., Huth, J., Kuenzer, C., 2022. A First Assessment of Canopy Cover Loss in Germany’s Forests after the 2018–2020 Drought Years. Remote Sensing 14, 562. https://doi.org/10.3390/rs14030562

For further information please contact: frank.thonfeld@dlr.de.