What?
The SLUMAP project (Remote Sensing for Slum Mapping and Characterization in sub-Saharan African Cities) funded by the STEREO-III programme of the Belgian Science Policy (BELSPO) and ACCOUNT funded by the Dutch Research Council (NWO),
develops methods for mapping and estimating the population of such often invisible spaces
The overall objective of SLUMAP is to propose an open-source framework that allows for the processing of remote sensing images for (i) providing information on the location and extension of deprived areas (e.g. slums) within a city
and (ii) characterizing their physical environment (i.e., in terms of greenness, built-up density, etc.) at limited cost
More info: https://slumap.ulb.be/ and https://slummap.net/
Why?
According to UN-Habitat, around one billion people live in 'slums'. The urban Sustainable Development Goal (SDG 11) has the "proportion of urban population living in slums, informal settlements or inadequate housing" as its first indicator
to measure progress towards sustainability. Unfortunately, data for this indicator is commonly not readily available for supporting local or global monitoring
Up-to-date maps on the location, characterization of the living environment and population data are urgently required, for local SDG monitoring, planning and service provision, humanitarian response or to address dramatic differences
in health outcomes
How?
First, for modelling at city scale the deprivation probability of a 50 by 50 m grid, we use free-cost Sentinel-1/2. We develop a machine learning workflow using GRASS GIS and R, in a Jupyter Notebook. Ancillary open global datasets include
SRTM, OSM and WSF 2019
Second, in addition to the deprivation probability at the city scale, we included data on the percentage of built-up using Open Buildings (https://sites.research.google/open-buildings/)
and High Resolution Population Density Maps (HRL) (https://dataforgood.facebook.com/dfg/tools/high-resolution-population-density-maps).
These additions allow to compare and combine deprivation probabilities with built-up densities and population estimates at a 50 by 50 m grid cell
Third, at settlement scale, we made use of VHR satellite images (WorldView-3) and used Geographic Object-Based Image Analysis (GEOIBA) to create land use/cover maps for deprived areas. The classes include buildings, low and tall vegetation,
water, ground cover, and waste piles
Fourth, we provide an environmental characterization at settlement scale and population estimates at the settlement scale at 50 by 50 m grid cells. This layer includes the percentage of waste, built-up densities, percentage of vegetation
cover and a bottom-up population estimate. For the bottom-up population estimate, the buildings and local population data are used
The tool is developed and hosted by the Slumap team!