ESA Worldcover Polygons

Overview

This repo contains polygons with different spatial resolutions of the ESA Worldcover dataset, a landuse and landcover raster collection. The polygons can for example be used to generate a landcover vector tileset.

The ESA Worldcover raster images have a spatial resolution of 10 m per pixel. We downsample these images to a resolution of 0.1 / 2 ** k degrees for k=0..8 which corresponds at the equator to 11 km per pixel for k=0 and 44 m per pixel for k=8.

The downsampled raster images are then filtered with a median blur and a sieve, then get polygonized, and subsequently the polygons are smoothed with the midpoint method.

To keep the size of images roughly equal, we subdivide the latitude range -85 to 85 degrees and the longitude range -180 to 180 degrees into 2**k ranges. This means at k=0 we have 1 image, at k=1 we have 4 images, and so on. These images are indexed by i and j.

License

The code in this repository is licensed as MIT.

The ESA Worldcover dataset is licensed as Creative Commons Attribution 4.0 International License, see Terms of use at https://worldcover2020.esa.int/download.

Using the ESA Worldcover polygons from this repo requires attribution to the original ESA Worldcover dataset.

Demo

https://wipfli.github.io/esa-worldcover-polygons

Downloads

GeoPackage polygons (55.35 GB):

https://pub-726b01260c98468a9387cc0dfcb7386b.r2.dev/esa-worldcover-polygons-gpkg.tar

Vector tiles v1, z0 - z11 (6.31 GB):

https://pub-726b01260c98468a9387cc0dfcb7386b.r2.dev/esa-worldcover-polygons.pmtiles

Vector tiles v2, z0 - z9 (401 MB):

https://pub-726b01260c98468a9387cc0dfcb7386b.r2.dev/esa-worldcover-polygons-v2.pmtiles

Steps

Download the ESA Worldcover GeoTiffs with:

cd downloads/
./download.sh
./unzip-all.sh

Create a Virtual Raster (VRT) mosaic.vrt that combines all the ESA GeoTiffs into one file:

gdalbuildvrt mosaic.vrt downloads/ESA_*.tif 

Create tif samples:

python3 sample.py

Filter samples and polygonize them with:

python3 filter.py

To zip:

python3 to_zip.py

Run planetiler:

cd planetiler
./regenerate-swissmap.sh