Mapping and monitoring shoreline changes along South Africa’s southwestern coastline from 1937 to 2020 using multisource remote sensing data and techniques

Abstract

Coastal erosion is a risk presented to sandy beaches around the world. In June 2017, a large storm surge eroded a section of Sixteen-Mile Beach along South Africa’s southwestern shoreline. To understand events like these, and how to manage coastal erosion, one has to analyse how beaches have changed over the long-term. This can be effectively achieved through remote sensing. This study aims to assess the spatio-temporal changes to South Africa’s southwestern shoreline between 1937 and 2020 and projects future shoreline changes. This is achieved by integrating different remote sensing data sources and techniques. Firstly, historical aerial photographs from 1937, 1960 and 1977 were georeferenced, and their shoreline positions were manually digitised. Secondly, shoreline positions were automatically delineated from Landsat 5, 7 and 8 and Sentinel-2 imagery between 1985 and 2020 using the CoastSat toolkit and Google Earth Engine. These two datasets were then combined to quantify the rate and magnitude of shoreline erosion and accretion that has taken place over these eight decades using the Digital Shoreline Analysis System (DSAS). Additionally, the short-term shoreline variability, seasonality and storm surge impact were investigated by creating cross-shore change profiles delineated from Sentinel-2 data. Finally, DSAS was used to model future shoreline positions for 2030 and 2040 based on the long-term changes. The results revealed that the shoreline has undergone dynamic changes between 1937 and 2020, culminating in an average net erosion of 39 m, predominantly due to the June 2017 storm surge. Yet, the shoreline forecasts indicate a slight shoreline retreat for the upcoming two decades. It must be noted that these remote sensing data are limited by their underlying uncertainties. Therefore, recommendations for future studies include integrating additional remote sensing and field data to present a more holistic understanding of the coastal environment