South African Journal of Geomatics

Temporal Characterization of Land Use Change and Land-scape Processes in Informal Settlements in the City of Cape Town, South Africa

Perpetua I Okoye, Jörg Lalk — Wed, 10 Jul 2024 00:00:00 +0000

This study conducted a Land Use Change (LUC) analysis on informal settlements in Cape Town, South Africa, using bi-temporal steps, S1 (2010) and S2 (2016), to characterize land use (LU) conversions and landscape processes for informed policymaking. Utilizing the 2011 national land cover dataset and post-classification methods, two LU datasets and maps, D1 for S1 and D2 for S2, were derived. These classifications achieved an overall accuracy exceeding 95%, with Kappa coefficients above 0.9. The analysis employed change trajectories and conversion labels to evaluate LU changes and landscape dynamics, providing a thematic representation of LUC within informal settlements. Landscape reclamation processes, including abandonment, urban development, and RDP (Reconstruction and Development Programme) development, constituted approximately five percent of the total LU conversions, while degradation processes like persistence and intensification dominated, affecting approximately 93% of the area. Partial reclamation, notably through interspersed RDP (RDPi), accounted for about two percent of conversions. These findings highlight the importance of accurate and timely LUC data reporting in informal settlements to address socioeconomic challenges effectively and support policy decisions to enhance these communities' physical and socioeconomic infrastructure.

Error Analysis in Multibeam Hydrographic Survey System

Basil Daniel Devote — Wed, 10 Jul 2024 00:00:00 +0000

Hydrographic surveying involves the integration of a depth-measuring sonar (Sound navigation and ranging) with a positioning system or Global Navigation Satellite System (GNSS); a motion sensor or Inertia Measuring Unit (IMU); and an azimuth sensor (gyroscope). The various sensors acquire data in terms of their respective reference frame and time. The challenge lies in integrating the various sensor frames and time, and in transforming the vessel frame coordinate system into a terrestrial reference frame. The integration of the various sensor frames and time is necessary to minimize systematic errors in the bathymetric data that result from latency, and calibration uncertainty. The focus of this research is to model the systematic bias associated with the integration of the various sensor reference frames. In so doing, the quality of the acquired data is enhanced, and error budgeting and uncertainty prediction can be effectively carried out during the preparation, acquisition, and processing stages of the bathymetric exercise. As such, the required project specification and hydrographic standards, as defined by the International Hydrographic Organization (IHO), are met.

Characterising the evolution of the urban form of zones that accommodate warehousing clusters in the City of Cape Town municipality

Masilonyane Mokhele, Brian Fisher-Holloway — Wed, 10 Jul 2024 00:00:00 +0000

Modern economies are characterised by increasing globalisation, e-commerce, and a growing number of logistics facilities. Despite insightful research on the changing locational patterns of logistics facilities epitomised by logistics sprawl, there is a lack of literature analysing changes in the urban form of areas that accommodate warehousing clusters. The paper, therefore, aims to analyse changes in the urban form of zones that accommodate warehousing clusters in the City of Cape Town municipality. The study was based on three main types of secondary data: georeferenced 1:50 000 topographical maps from 1942 to 2010, current and historical spatial planning policy applicable to the City of Cape Town, and historical literature on the spatial economic characteristics of the zones that accommodate warehousing clusters. The topographical maps were loaded onto ArcGIS 3.10, after which large buildings were traced to ascertain changes in the urban form of the warehousing cluster areas over the respective decades. The study found that changes in the urban form of the contemporary warehousing cluster areas were linked to the growth of industrial zones and the transport infrastructure. However, the spatial policy for the time under consideration, although cementing the growth of industrial zones in the municipality, did not explicitly consider the placement of warehousing facilities. In light of the findings, the City of Cape Town municipality is urged to anticipate and plan for the growth of warehousing relative to the industrial zones and transport infrastructure. To ensure efficient and sustainable land utilisation, derelict industrial buildings in accessible areas could be redeveloped to accommodate warehousing facilities.

Classification of 3D Sonar Point Clouds derived Underwater using Machine and Deep Learning (CANUPO and RandLA-Net) Approaches

Simiso Ntuli, Mulemwa Akombelwa, Angus Forbes, Mayshree Singh — Fri, 12 Jul 2024 00:00:00 +0000

The techniques of point cloud classification in aquatic environments have various applications such as landslide hazard mapping, recovery of lost objects, underwater infrastructure inspection, exploration of mineral resources on the seabed, underwater cultural heritage documentation, environmental preservation and conservation purposes. This study combines acoustic (Sonar) and laser-based (Lidar) remote sensing technologies in an aquatic environment with two machine and deep learning approaches to illustrate the techniques to identify submerged objects. Firstly, the relative accuracy of the underwater imaging system, the BlueView BV5000 Mechanical Scanning Sonar, is evaluated at close range. Secondly, the supervised CANUPO and RandLA-Net classification approaches are used to classify submerged sonar point clouds. Common objects of interest, namely tyres and chairs, were selected for classification. Relative accuracy measurement results showed a centimetre-level root mean square error (RMSE) value, with good accuracies recorded when the scanner is positioned close to objects. The best results were achieved when the target objects were placed at a minimum distance of 2 m from the acoustic scanner. Subsequently, the results of point cloud classification were satisfactory for both approaches. An overall accuracy of 79.81% and an F₁ score of 79.80% were achieved using the CANUPO classification approach. On the other hand, an 80.72% overall accuracy and an 80.63% F₁ score were obtained using a RandLA-Net approach. These analyses provide a reasonable framework for the parameters that can be used when applying these techniques in natural aquatic environments.