Browsing by Author "Musakwa, Walter"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Evaluating the potential of Earth observation for supporting sustainable urban land use planning
    (Stellenbosch : Stellenbosch University, 2013-12) Musakwa, Walter; Van Niekerk, Adriaan; Stellenbosch University. Faculty of Arts and Social Sciences. Dept. of Geography and Environmental Studies.
    ENGLISH ABSTRACT: In many developing countries, rapid urbanisation continues to substantially transform land from agricultural and rural land uses, as well as natural landscapes into urban areas. This leads to significant changes to the socio-economic fabric and nature of the natural environment. Data to monitor such transformation is often out of date, unreliable, in unstandardised format, cumbersome and expensive to collect or simply unavailable in urban centres of many developing countries. These characteristics inhibit local authorities‘ and other stakeholders‘ capacity to monitor and leverage resources toward sustainable urban development. Sustainable urban land use planning is a major objective of urban planning, but it is difficult to put into practice. This study investigates the efficacy of earth observation (EO) for collecting information required for sustainable urban land use planning and proposes the use of decision consequence analysis (DCA) as a simple and structured way to put sustainable urban development into practice. The study focuses on three central determinants of sustainable urban land use, namely (1) land use change and land use mix, (2) urban sprawl and (3) the urban built-up area. Consequently, urban sustainability indicators of these three components were identified. EO data for Stellenbosch, a town in the Western Cape province of South Africa, was gathered and used to perform spatio-temporal analyses of the indicators in a geographic information system (GIS). This enabled the establishing of the positive or negative trajectory made toward achieving sustainable urban land use planning. The study demonstrates how the use of EO data, DCA, urban sustainability indicators and GIS can enhance local authorities‘ capacities for monitoring urban sustainability. EO data and urban sustainability indicators were used to develop an urban sustainability toolbox which facilitates evidence-based decision making. The results also show that urban sustainability indicators derived from EO are valuable in providing synoptic, up-to-date, standardised and normalised information on urban areas. Such information would be expensive and cumbersome to collect without the use of EO and GIS. As a result, earth observation will continue to play a key role in monitoring urban sustainability, particularly in developing countries.
  • Thumbnail Image
    Item
    A synthesizing land-cover classification method based on Google Earth Engine : a case study in Nzhelele and Levhuvu catchments, South Africa
    (Springer, 2020-07-07) Zeng, Hongwei; Wu, Bingfang; Wang, Shuai; Musakwa, Walter; Tian, Fuyou; Mashimbye, Zama Eric; Poona, Nitesh; Syndey, Mavengahama
    This study designed an approach to derive land-cover in the South Africa with insufficient ground samples, and made a case demonstration in Nzhelele and Levhuvu catchments, South Africa. The method was developed based on an integration of Landsat 8, Sentinel-1, and Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM), and the Google Earth Engine (GEE) platform. Random forest classifier with 300 trees is employed as land-cover classification model. In order to overcome the defect of insufficient ground data, the stratified sampling method was used to generate the training and validation samples from the existing land-cover product. Likewise, in order to recognize different land-cover categories, the percentile and monthly median composites were employed to expand input metrics of random forest classifier. Results showed that the overall accuracy of the land-cover of Nzhelele and Levhuvu catchments, South Africa in 2017–2018 reached to 76.43%. Three important results can be drawn from our research. 1) The participation of Sentinel-1 data can slightly improve overall accuracy of land-cover while its contribution on land-cover classification varied with land types. 2) Under-fitting problem was observed in the training of non-dominant land-cover categories using the random sampling, the stratified sampling method is recommended to make sure the classification accuracy of non-dominant classes. 3) When related reflectance bands participated in the training process, individual Normalized Difference Vegetation index (NDVI), Enhanced Vegetation Index (EVI), Soil Adjusted Vegetation Index (SAVI), Normalized Difference Built-up Index (NDBI) have little effect on final land-cover classification result.