Logo Lanfrica
  • Home
  • Atlas
  • Insights
  • Docs
  • Sign in

© 2026 Lanfrica. All rights reserved. All copyrights of the resources shown on the Lanfrica website belong to the original copyright holders, unless explicitly stated otherwise.

The African Wildland-Urban Interface

Domain:

geospatial

Record type:

dataset
Creator:
Pee
Publisher:
Zenodo
Host:avatar

The wildland-urban interface (WUI), where buildings meet wild vegetation, is where socioecological challenges like wildfire risk, zoonotic disease transmission, and invasive species spread are most acute. Understanding the extent, composition and characteristics for of the WUI is important for land use and urban planning to mitigate risks to humans and wildlife. Existing global maps of the WUI (e.g., Schug et al. (2023)) rely on medium-resolution satellite imagery to estimate building density, but in doing so they miss many small, scattered buildings, especially those made of natural building materials (thatched grass, wood, mud etc). These small buildings are especially common in the African WUI. Here, we represent a new dataset of the African WUI that was developed in a recent study by the authors of this dataset (see corresponding publication). We used individual building locations derived from high-resolution imagery to estimate building density. The building locations from two sources: Google's Open Buildings v2, and Microsoft's Global Building Footprints datasets. 

Our African WUI dataset is 8.4 % more accurate than the African portion of Schug et al.'s (2023) global WUI map, with performance gains largely attributable to reduced omission errors in areas with small, scattered buildings. 

Temporal extent

The data contains data representative for ca. 2020.

Data format and units

The data are organized in tiles of 100 km x 100 km and follow the EQUI7 tiling grid and projection system. The images are compressed GeoTiff files (*.tif). There is a mosaic in GDAL Virtual format (*.vrt), which can readily be opened in most Geographic Information Systems. Please consider the generation of image pyramids before using *.vrt files. We also include the tiling grid so that users can select the tiles that correspond to their area of interest. 

The raster dataset contains Wildland-urban interface (WUI) data (one layer), 10 m spatial resolution, 8 discrete classes:

1 - Forest/Shrubland/Wetland-dominated Intermix WU

2 - Forest/Shrubland/Wetland-dominated Interface WUI

3 - Grassland-dominated Intermix WUI

4 - Grassland -dominated Interface WUI

5 - Non-WUI: Forest/Shrub/Wetland-dominated

6 - Non-WUI: Grassland-dominated

7 - Non-WUI: Urban

8 - Non-WUI: Other

Visit

doi.org

Tasks

computer visionimage classification

Licenses

info:eu-repo/semantics/restrictedAccessCreative Commons Attribution 4.0 Internationalhttps://creativecommons.org/licenses/by/4.0/legalcode