Home

Urbanization is an important consideration in all aspects of a project's life cycle at the World Bank, from design, through implementation, to monitoring and evaluation. As part of the many projects the GOST team have supported, we have developed a number of tools and processes for quantifying urbanization, as the majority of the urbanization numbers used by the UN and the World Bank are self-reported by governments, and not all governments define urbanization in the same way.

Analyzing urbanization

The first step in understanding urbanization is to quantify urbanization. The World Bank endorsed a methodology to quantify urbanization using a consistent methodology developed by our friends at the European Commission. We have a more detailed description of the methodology here in the wiki, of both the standard three-class definition of urban and the updated [seven-class definition] (https://github.com/worldbank/GOST_Urban/wiki/Developing-L2-urban-classification-in-python).

For most of our work, we rely on the endorsed three-class methodology, which operates by applying thresholds to gridded population data, and summarizing total population in the resulting settlements (contiguous cells matching the minimum thresholds). This results in three classes of urbanization:

Class	Pop Density (people per km2)	Total Pop
Urban Center	1500	50000
Urban	300	5000
Rural (everything else)	0	0

The results of this analysis are a series of vectors representing both urban centers (major cities) and urban areas. These vectors can then be used to summarize a number of other datasets and perform additional analysis.

Additional Analysis

1. Summarize change in nighttime lights
2. Summarize change in the evolution of the built environment
3. Calculate the Landscape Expansion Index (LEI), which quantifies the nature of urban expansion