Missing data

“In humanitarian catastrophes, data about the affected area is needed to increase the effectiveness of responses. Notably, spatial information about the distribution of population and the road network is critical for emergency management, and yet is missing for large portions of the world. The absence of information greatly limits the abilities of organisations to plan and deliver an adequate response in case of emergency. Across many parts of Africa, this essential spatial information is often of limited quality, outdated, or even completely missing.”

Estimating Population Distribution with Landsat Imagery and Volunteered Geographic Information. Available: [accessed May 14 2021].

Being a consultant for an NGO is always challenging: Most of the time an organisation just outlines the problem what they were facing for and the job is to find a useful and cost-effective solution.

And since we are talking about Africa (Disaster relief operations), where most of the time the spatial data is outdated or completely missing, you must find a unique solution for a unique problem.  

What data can we expect in our work?

• OpenStreetMap database (up and down)
  • Add data – more than 16,500 volunteers were involved during the last 5 years and added or modified over 5,250,000 buildings and 295,000 km of the road network, 9800 POIs.
  • Get data
    • Get data by using overpass API
    • Use Geofabrik’s free download server
    • Use export tool from the HOT team
    • Get data by using AWS Athena (check our blog post how to do it)
    • Or you can use some useful plugins for QGIS
• Spatial dataset from the local authorities (health facilities, schools, waterpoints)
• Non-GIS database which can be geocoded or joined to the existing data we have.
• Local survey
  • Maps.me applications uses by local volunteers – POIs directly to OSM
  • KoBoToolbox or Survey123 (local volunteers collect local information like where is the village centre or new water source – for example for “solar irrigation projects”

Other data

Population distribution databases („Gridded Population of the World (GPW)”, and the „Global Rural-Urban Mapping Project (GRUMP)”, Wordpop, LandScan), census, etc.);

Land cover datasets (Global Land Cover (GLC), GLOBCOVER 2005 and 2009, or Global Land Survey, GLS);

Other (Road network, water network (permanent and/or temporary), weather data, etc.)

Satellite pictures (High population density areas can be demarcated by settlement extents provided by processing Sentinel-2.). Low population density areas can be demarcated using VGI – Volunteered geographic information, like MapSwipe or MissingMaps community.

to be continued…