Mapping Urban and Peri-Urban Land Cover in Zimbabwe: Challenges and Opportunities

Abstract

Accurate and current land cover information is required to develop strategies for sustainable development and to improve the quality of life in urban areas. This study presents an approach that combines multi-seasonal Sentinel-1 (S1) and Sentinel-2 (S2) data, and a random forest (RF) classifier in order to map land cover in four major urban centers in Zimbabwe. The specific objective of this study was to assess the potential of multi-seasonal (rainy, post-rainy, and dry season) S1, rainy season S2, post-rainy season, dry season S2, multi-seasonal S2, and multi-seasonal composite S1 and S2 data for mapping land cover in urban areas. The study results show that the combination of multi-seasonal S1 and S2 data improve land cover mapping in urban and peri-urban areas relative to only multi-seasonal S1, mono-seasonal S2, and multi-seasonal S2 data. The overall accuracy scores for the multi-seasonal S1 and S2 land cover maps are above 85% for all urban centers. Our results indicate that rainy and post-rainy S2 spectral bands, as well as dry-season S1 VV and VH bands (ascending orbit) are the most important features for land cover mapping. In particular, S1 data proved useful in separating built-up areas from cropland, which is usually problematic when only optical imagery is used in the study area. While there are notable improvements in land cover mapping, some challenges related to the S1 data analysis still remain. Nonetheless, our land cover mapping approach shows a potential to map land cover in other urban areas in Zimbabwe or in Sub-Sahara Africa. This is important given the urgent need for reliable geospatial information, which is required to implement the United Nations Sustainable Development Goals (UN SDGs) and United Nations New Urban Agenda (NUA) programmes.

Mapping Urban Land Cover Using Multi-Seasonal Sentinel-1 Imagery and Texture Indices

“…We used mean and median seasonal Sentinel-1 imagery because the imagery shows lower speckle than single-date imagery. As a result, we did not perform speckle reduction because it generally reduces spatial resolution (Kamusoko et al 2021). Note that analysis-ready Sentinel-1 data was processed and terrain corrected in GEE.…”

Geospatial Machine Learning in Urban Environments: Challenges and Prospects

“…Consequently, geospatial or mapping agencies in the less developed countries fail to produce timely, reliable, and accurate geospatial information. Furthermore, official development assistance (ODA) funding for major mapping projects has declined over the past decades (Kamusoko et al 2021).…”

Mapping Urban Land Cover Using Multi-seasonal Sentinel-2 Imagery, Spectral and Texture Indices

“…First, built-up areas in sparse urban or peri-urban areas appear identical to fallow cropland and bare areas because these features exhibit high reflectance in the visible-infrared wavelengths (Herold et al 2004;Schneider 2012). For example, spectral similarity between newly developed built-up areas and other land cover surfaces such as fallow cropland fields and bare areas has been reported to cause classification errors, especially in peri-urban areas in developing countries (Kamusoko et al 2013(Kamusoko et al , 2021Schug et al 2018). This is because most built-up structures in peri-urban areas are made of the same materials found in the surrounding areas, which results in low object-to-background contrast (Jensen 2000).…”

“…Therefore, spectral-temporal features derived from multi-seasonal Sentinel-2 data can be used to discriminate built-up areas from cropland and bare areas. Past studies have shown that multi-seasonal or multi-temporal Sentinel-2 or Landsat data improve land cover mapping in urban and peri-urban areas (Kamusoko et al 2021;Yuan et al 2005). This is because built-up spectral responses are largely persistent during the different seasons, while non-built-up (e.g., cropland and bare areas) spectral responses change (Griffiths et al 2010).…”