Predicting slum dwellers’ deprivations from space: a pilot study on Dhaka

Co-authors:
Christian Borja-Vega, Economist, the World Bank’s Water Global Practice
Amit Patel, Assistant Professor. University of Massachusetts Boston’s McCormack Graduate School for Policy and Global Studies

 
One of the most pressing development challenges we face in today’s world of rapid urbanization is how to respond to the unmet demand for basic infrastructure services, like adequate housing, clean water, and sanitation. Half of the world’s population lives in cities and close to 1 billion live in slums. Megacities in developing countries are growing faster than ever, mostly in an unplanned way. Dhaka, the capital of Bangladesh, is particularly challenged due to congestion, poor infrastructure and regular flooding during heavy rainfall. In fact, according to the Economist Intelligence Unit, it was ranked almost at the bottom of the list of the world’s livable cities, number 137 out of 140.
 
When looking at approaches to and impacts of ways to upgrade slums, a challenge is often a lack of adequate data, particularly a lack of spatial data. Tools like Google Maps, now so ubiquitous in our daily lives, rarely cover growing, unplanned settlements. Conducting household surveys are useful, but they are expensive and not frequently updated. This is a problem when trying to map a dynamic and elusive slum, whose very definition is controversial and context-specific.
 
So what is the solution? It turns out that using earth observation (EO) data – that is, information obtained through remote sensing instruments, such as satellites – may help to address this critical data gap. It is already being used in many other applications, from recognizing cross-border territories in disputes, tracking wildlife, locating hazards and disasters, to identifying migration patterns.
 
Over the past two decades, the temporal and spatial resolution of earth observation imagery has increased dramatically. Along with such imagery, advanced algorithms can be used to detect and describe slums, answering critical questions:

  1. Where are slums/informal settlements located?
  2. How do their appearances change over time?
  3. Do they present spatial characteristics incompatible with basic infrastructure supply (living space, water and sanitation, roads and safety from natural hazards)?

With an eye towards using innovative data analytics for development impact in slums,  we launched a pilot in Dhaka under the ‘Water Supply and Sanitation in Rapid Urbanization’ umbrella about a year ago. We worked in collaboration with the WASH Poverty Diagnostic team in Bangladesh, the European firm GiSat, a member of a EO4SD Urban consortium working with the European Space Agency (ESA), and researchers from the University of Massachusetts Boston. Preliminary results of the study have been presented (see video and slides), explaining how the information of in-depth household surveys can be complemented with geoinformation captured remotely for Dhaka slums.
 
The Details

In our approach, several EO-derived indicators feed into a statistical model to predict an index of household deprivation called Slum Severity Index (SSI). The SSI captures the needs of slum residents for: a) access to safe water, d) access to adequate sanitation, b) durable housing construction, c) sufficient living space, e) security of tenure, and f) metered electricity.
 
Here are some examples of insights obtained from the preliminary analysis of combined EO and survey response data in Dhaka:

  • As the distance from central business district and from major roads increases, the lack of water or electricity increases (known as “peripheralization of slums” phenomenon)
  • Where the percentage of high density residential urban fabric increases, housing deprivation also increases
  • The indicators for building density (e.g. average dwelling size and distance between buildings) are associated with worsening housing and basic services

Mapping these deprivation indicators, and identifying the degree to which they intersect, could help make the prioritization of ways to upgrade slums easier. Citywide programs are not always feasible, so gaining spatial awareness of unmet needs can benefit local governments and utilities to identify the most under-served neighborhoods or to tackle critical bottlenecks, such as tenure insecurity. In addition, this approach can bring valuable information at a fraction of the cost, even with small sample surveys, that could be used to inform the design of complex operations.
 
This initiative is still in progress and we are disseminating the results and tweaking the methodology so it can be used in more cities. The sky is the limit on how far we can develop this to help identify needs and combat poverty.
 
For comments or suggestions on potential applications of the approach, please contact Luisa M. Mimmi and Christian Borja-Vega.

Fig 1 Informal settlements in Dhaka, classified by density, shape & context

Source: EO4SD Urban consortium. 2017. “Service Operations Report – Dhaka”. (Internal document)

Fig 2 Example of slum communities’ locations as derived from WASH-POV

Source: Gisat, 2017.
Note: Slum areas (pink polygons) and buffers around slum communities’ location
(purple circles)

Fig 3 Proportion of Informal Settlements/Slum Areas in flood prone areas with high risk as of 2017 (left) and flooded areas for the historical remarkable flooding event in 2004 (right).

Source: EO4SD Urban consortium. 2017. “Service Operations Report – Dhaka”. (Internal document)

[Read More …]