stepSA News


Tracking spatial change in cities:
New set of fine grained indicators for 9 Cities


The use of a fine-grained uniform spatial resolution to depict economic and population change enabled the CSIR team to create population and economic production change indicators for the nine largest cities in South Africa. The July version of this newsletter reported on the latest State of the Cities Report 2016 (SOCR) release and the assistance provided by the CSIR in developing a number of spatial indicators to measure spatial change, and report on levels of growth and spatial transformation at the sub-city level. The urgency for spatial transformation and need for indicators to track spatial outcomes is clearly highlighted by the SOCR (2016) and the Spatial Planning and Land Use Management Act, 2013.

Even though some spatial information is available in smaller units such as Wards, Sub-places or even Small Area Units, these demarcations often vary from one census to the next which does not allow for direct comparison over time. Using different units to group information presents challenges in comparing spatial change statistically and visually. To enable sufficient pattern detection of spatial features it is necessary that the scale be sufficiently fine (for purposes of detecting spatial transformation for example).

To address this issue, the CSIR explored an approach that uses a single-sized uniform tessellation to create demographic and economic indicators. Hexagons with an edge length of 250m were used. Population and economic information for different census periods was assigned to these zones using a dasymetric assignment process.

Using this fine-grained uniform spatial framework, the CSIR has created population and economic production change indicators for the nine largest cities in South Africa. The CSIR will continue to explore the use of this framework and would also welcome any feedback. .

The finer grained resolution of the spatial unit, used to present the information, hugely enhances what can be portrayed. The new fine grain indicators address problems associated with aggregation (which often results in generalisation that can lead to a loss of detail). This is particularly relevant when information is scale-dependent and where the geographic extent is sensitive to the spatial arrangement. One of the advantages of using such a small uniform tessellation is that information can be represented more effectively in 3D than units that vary substantially in size.

The following two maps are examples illustrating spatial patterns of population concentration across Gauteng's three metros (map 1) and for the same areas reflecting the change in population comparing 1996 and 2011 (map 2).


Map 1: Spatial patterns of population concentration reflected for the three metropolitan areas in Gauteng, 2011.

Map 2: Spatial patterns of population growth reflected for the three metropolitan areas in Gauteng, 1996-2011.

As the spatial units are uniform in size, values along corridors or a given distance from roads can be more accurately calculated. It also enables comparison of change across different time periods. The example below illustrates a linear transect of population density change considering the connection between the Johannesburg and Pretoria CBDs.


Diagram: Linear transect from the JHB CBD to the PTA CBD reflecting the change in population comparing 1996 to 2011 (Source: Napier, Le Roux, & Van Heerden, 2016).

Using this fine-grained uniform spatial framework, the CSIR has created population and economic production change indicators for the nine largest cities in South Africa. The CSIR will continue to explore the use of this framework and would also welcome any feedback. For more information please contact Johan Maritz





Back to top