Multivariate analysis of socioeconomic profiles in the Ruhr area, Germany

Figures & data

Figure 1. (a) Green and recreational land use. (b) Urban, commercial and industrial land use. (c) Schematic view of the eight historical zones of the Ruhr area (d) The location of the Ruhr area in Germany. Data source: Urban Atlas of the Copernicus Land Monitoring Service (Urban Atlas, 2019)

Industrial, commercial and urban land uses are concentrated mainly in the Emscher and Hellweg zones, located in the central parts of the Ruhr region. Green and recreational areas dominate the outskirts, where industrial production was substantially less extensive throughout the different phases of industrialization.

Figure 3. Distribution of the twenty-five socioeconomic SOM clusters over the Ruhr region

We performed the SOM analysis for the 732 neighbourhoods or Stadtteile. There is an evident pattern of more red and yellow neighbourhoods in the central areas (Emscher and Hellweg zones) with grey and blue colours dominating the outer neighbourhoods.

Figure 2. The ten SOM component maps. Values are defined as percentages (%) or indices ranging between 1 (very low) an 6 (very high). Data sources: (Microdialog, 2017; Zensus, 2011).

In the ten component maps brighter orange/red colours correspond to higher values and darker blue/cyan colours to lower ones, e.g. neighbourhoods located in the upper left corner are characterized by high intensities of purchasing power, net rent price, share of German population and percentage of only-elderly households.

Figure 4. SOM Hit Map: Distribution of SOM clusters over the 2D model space.

Moving on the surface of the Hit Map from right to left (i.e. from less towards more transparency) we observe increasing median age and decreasing household size, while from the bottom towards the top we encounter rising socioeconomic advantage and decreasing foreign population (from red towards blue). There can be multiple geographical neighbourhoods assigned to one node on the Hit Map.

Urban Atlas. (2019). https://land.copernicus.eu/local/urban-atlas.

 Google Scholar

Microdialog. (2017, September 4). https://www.deutschepost.de/en/d/deutsche-post-direkt/microdialog.html.

 Google Scholar

Zensus. (2011). July 23, 2021. https://www.zensus2011.de.

 Google Scholar

Data availability statement

To support our results and conclusions, we provide data containing the following parameters for each of the 732 neighbourhoods analysed in this article (in $.csv$ format): The x and y coordinates of the centroids (ETRS89 / UTM zone 32N), the SOM cluster-ID (ranging between 0 and 24, see Figures 4 and 3) and we also indicate the city to which they belong (out of the total of 53 cities of the Ruhr Area). Moreover, we make the mean values of the ten variables (Figure 2) available for the twenty-five SOM clusters individually (as $.csv$). Since the German census of 100 × 100 metre resolution follows data protection guidelines (e.g. the precision of the information is reduced by not providing decimal places (Zensus, 2011)), a small error margin is to be expected when averaging for the whole metropolitan level.