Deriving peri-urban settlement typologies from Landsat time series analysis: the case study of Mankweng and environs (South Africa)
Sustainable development is a challenge for highly dynamic peri-urban areas like Mankweng and surroundings, located in a former homeland in South Africa. The analysis adopted a two-step approach to test a key assumption. Landsat time series analysis used nine timestamps from 1984 to 2022. Ten land use/cover classes were distinguished including four built-up classes (OA 61.6%). Given the need to integrate historical development into a peri-urban settlement typology, a temporal trend analysis of built-up classes for dedicated areas proved the point. Unsupervised hierarchical clustering was employed to develop four distinct settlement typologies—two related to settlement composition and two related to settlement development. The clustering process used either the most recent timestamp (considering all classes) or the full time series (focusing on built-up classes), and either administrative boundaries or new boundaries (derived from spatially autocorrelated clusters treating the full time series). The settlement development typologies best identify types that correspond to multiple developmental pathways, particularly using the time series-derived boundaries. By incorporating the history of areas into settlement typologies through satellite time series analysis, the study proposes a new approach for deriving typologies for planning in heavily expanding and ever-changing settled former apartheid homeland areas of South Africa.
Authors
- Gertrud Schaab (ORCID: https://orcid.org/0000-0002-1515-5051)
- Deepthi Patric
- Felipe Vásquez Tavera
Institutions
- Karlsruhe University of Applied Sciences (DE)
Publication Details
- Journal
- Big Earth Data
- Published
- 2025-09-29
- DOI
- https://doi.org/10.1080/20964471.2025.2565067
- Primary Topic
- Land Use and Ecosystem Services
- Type
- article
- Field-Weighted Citation Impact
- 0.00