Integration of World Settlement Footprint

[don-vip]

World Settlement Footprint (WSF) could maybe be integrated to Osmose to detect missing landuse=residential / buildings / highways worldwide.

WSF is an open dataset released by DLR (German Space agency) that identifies settlement footprints based on Sentinel and Landsat imagery. There are two millesimes (2015 and 2019).

WSF 2015:

• download link: https://figshare.com/articles/dataset/World_Settlement_Footprint_WSF_2015/10048412
• licence: CC0
• description:

  The World Settlement Footprint (WSF) 2015 is a 10m (0.32 arc sec) resolution binary mask outlining the 2015 global settlement extent derived by jointly exploiting multitemporal Sentinel-1 radar and Landsat-8 optical satellite imagery. The archive is organized for download in 306 GeoTIFF files (EPSG4326 projection, deflate compression) each one referring to a portion of 10x10 degree size (~1110x1110km) whose upper-left and lower-right corner coordinates are specified in the file name [e.g., the tile WSF2015_v1_EPSG4326_e010_n60_e020_n50.tif covers the area between (10E;60N) and (20E;50N)]. A virtual mosaic file (i.e., WSF2015_v1_EPSG4326.vrt) is also provided which allows visualizing the global product at once on most diffused GIS platforms (e.g., ArcGIS, QGIS, MapInfo). Settlements are associated with value 255; all other pixels are associated with value 0.

WSF 2019:

• download link: https://download.geoservice.dlr.de/WSF2019/files/
• licence: CC-BY 4.0
• description:

  This repository contains the World Settlement Footprint (WSF) 2019, a 10m resolution binary mask outlining the extent of human settlements globally derived by means of 2019 multitemporal Sentinel-1 and Sentinel-2 imagery.

  The dataset is organized in 5138 GeoTIFF files (EPSG4326 projection) each one referring to a portion of 2x2 degree size (~222x222km on the ground) with an extra buffer of 0.1 degree to avoid any discontinuity between neighbour tiles. Each tile is identified by the lower-left corner coordinates specified in the file name [e.g., the tile WSF2019_v1_12_18.tif covers the area between (12E;18N) and (14E;20N)]. Settlements are associated with value 255; all other pixels are associated with value 0.

  The WSF2019 is licensed under CC-BY-4.0.

  A comprehensive publication with all technical details and accuracy figures is currently being finalized. For the time being, please refer to: Marconcini, M., Metz-Marconcini, A., Esch, T., Gorelick, N. (2021). Understanding Current Trends in Global Urbanisation – The World Settlement Footprint suite. GI_Forum, 1, 33-38. https://doi.org/10.1553/giscience2021_01_s33.

Idea:

• each pixel with value 255 should be inside a human-settlement landuse (commercial, construction, education, residential, industrial, etc.)
• we should be able to find a building=* "near" every 255 pixel
• we should be able to find a highway=* "near" every 255 pixel

[frodrigo]

Interesting idea. But probably not to be used at pixel size.

Megre analyser can be derived to get geometry from raster using postgis or ogr.