Functional Urban Areas – Application of the Concept on Traditional Longitudinal Data for Intra- and Inter-City Analysis

Wenxiu Du , Ecole Polytechnique Fédérale De Lausanne (epfl-eth)
Andrew Ding, University of Waterloo
Dorothee Beckendorff, EPFL
Mathias Lerch, Ecole Polytechnique Fédérale De Lausanne (epfl-eth)

The use of satellite-based data has greatly improved our understanding of the world. In demography, urban planning, and city science, one of the most important satellite-based datasets to consider is the Global Human Settlement Layer (GHSL). However, the image-based data does not show individual-level information, therefore with no qualitative insights on who lives in a given city, how the population grow etc, which are only possible through surveying or data collected by governments. Among individual-level data, census data, which are extensively used around the world for planning, are especially rich in information. Due to privacy issues, census data are often anonymised and grouped into administrative regions, so that individuals cannot be identified. Administrative borders are frequently changed by the governments, making longitudinal research on a particular region challenging to conduct. On top of that, census data are country-specific in terms of defining the criteria for classifying an area as urban or rural and whether an area belongs to a city or metropolitan area. These factors make cross-national longitudinal studies challenging when performing urbanisation or city-related studies. Because the newly available satellite-derived data are rich in nature and enable analyses that were previously impossible to perform, it is empirical to harmonise the geographical divisions of each country through different observation periods and to develop novel approaches of integrating newly available satellite-derived data into research using traditional data. This research is an attempt to bridge this gap between satellite-derived GHSL data and census data for longitudinal urban studies.

See extended abstract

 Presented in Session 13. Flash session Data Infrastructures for Population Research