An estimated 2.15 billion people live in the near-coastal areas worldwide, 42% of whom live in the low-lying coastal zones [1]. Of these, nearly 190 million are expected to be prone to coastal environmental risks by 2100 CE [2]. While the most catastrophic environmental threat to coastal zones is tsunamis [3], there is also a globally ongoing process of coastal erosion and accretion that is most prominent where sandy shore sediments are mobilized by waves, tidal and littoral currents, and deflation [3, 4]. It is possible that nearly half of the global sandy beaches may go extinct under the current trajectory of erosion and the anticipated acceleration of climate change [6]. Understanding shoreline evolution is becoming increasingly important to the society and economy of coastal regions, as erosion typically destroys infrastructure, negatively affects land values, and can strongly influence ecological processes (e.g., where coastal wetland systems become inundated with salt water), but the mechanisms underlying erosion events are often complex and difficult to predict.
Coastal erosion occurs when the landward portion of the swash zone mobilizes the sandy substrate [7]. The wave setup is the way in which energy interacts with the sandy substrate as waves propagate into shallower water and cause elevated sea levels after the wave breaking point. The physics that conserves the radiation stress (basically the energy mass balance) of the water drives the formation of rip currents, which are an important part of sand redistribution. The mechanism is coupled with the dynamics of the swash zone and the concept of wave runup [8]. In a simple 2-dimensional conceptual framework, beach gradient, wave height and period, tidal effects and the local sea level rise associated with atmospheric low-pressure systems during storms will determine where sand will be eroded and where it will be deposited. This is captured in Bruun’s rule [9] which describes the physics of beach gradient responses to wave regimes. Most beach systems have seasonal (storm season vs non-storm season) and lunar period (spring vs neap tide) variations in the gradient of the beach. These typically operate around a long-term median, but some interactions exacerbate the erosive nature of waves.
As wave runup is the most tangible predictor of coastal erosion, models of coastal erosion vulnerability use this as the key parameter. Several coastal erosion vulnerability indices have been developed [10] and almost all are parameterized using the same input variables [11]. They typically use coastal geomorphology and slope, relative sea level change, shoreline erosion/accretion rates, tidal range, and wave height, which all are essentially wave runup variables, but some also consider the role of the foredune barrier system [11, 12] and sea level rise through climate change [12]. Although input variables from climate and wave models, tidal models, and remotely sensed shoreline characteristics can be obtained over large spatial scales, the models are typically only locally applicable [11]. The models can become computationally challenging with the addition of other variables, but most achieve satisfactory approximations without further parameterization.
When wave runup extends beyond the swash regime to coastal foredune barriers, the potential exists for a collision regime in which sand from the base of the foredune is eroded and deposited elsewhere, leading to coastal erosion [7]. Incorporating foredune barrier parameters into coastal vulnerability models is important because while wave runup models indicate coastal erosion risk, coastal vulnerability is related to the consequences of the erosion and flooding. Boruff et al. [14] developed a coastal social vulnerability index that considers the socio-economic impacts of sea level rise, and the IPCC Common Methodology for Vulnerability Assessment [15] uses sea level increase forecast to extrapolate wetland loss, climate change risk and adaptation, community vulnerability, and coastal vulnerability. Coastal foredune barriers are the final mitigation before erosion leads to impacts.
Shoreline position is a representative feature of beach dynamics and is useful in describing coastal erosion and accretion [16]. Remote sensing-based approaches and automatic computation in GIS-based platforms have become a preferred choice to map and monitor shoreline changes [17] due to their synoptic-scale coverage, consistent acquisition of sensitive spectral bands for mapping surface features, and their cost-effectiveness compared to spatially limited field-based techniques [18]. Recent remote sensing techniques enable the determination of shoreline position from open-access mid-resolution satellite products using advanced subpixel extraction algorithms [19, 20].
Sunny et al. [21] compared the performance of Sentinel-2, Landsat, and MODIS for quantifying shoreline changes at the regional scale and their results showed that Sentinel-2 is more effective due to its finer resolution. The Sentinel-2 coverage and the algorithms that calculate changes in shoreline position are sensitive to changes in beach gradient, but they are not designed to estimate the changes that extreme wave runup has on foredune barriers.
Using unmanned aerial vehicle (UAV) photogrammetry surveys, local erosion hotspots can be mapped in detail [22], and sand redistribution during foredune barrier erosion can be determined empirically. The accuracy of low-cost UAV systems and photogrammetry enables rapid reconstruction of complex environments and is particularly suitable for reconstructing dune profiles, as heterogenous sandy textures are ideal for the photogrammetric process [23, 24].
Sixteen Mile Beach on the west coast of South Africa is a long sandy beach stretching from Yzerfontein town in the south to the Langebaan Peninsula in the north (Fig. 1), and it has been the subject of long-term sand dynamics research [25–27] as well as historical analysis of beach sand dynamics [28], which provided empirical evidence that this coastline is vulnerable to coastal erosion. In June 2017, a severe storm caused extensive damage to a section of the dune cordon and destroyed properties and businesses along the eroded dune [27]. Another storm in September 2023 led to further erosion of the foredunes. Understanding the erosion caused by such events can inform future shoreline changes and vulnerabilities but detailed analysis also reveals erosion dynamics that are of broader relevance.
Here we analyse shoreline and foredune morphology changes in response to the 2017 and 2023 storms at Sixteen Mile Beach. Our objective is to determine whether coastal erosion is spatially uniform, and, if not, to determine the underlying cause of the spatial patterning. Assessing the age of eroded sediments, we consider the novelty of these coastal erosion event relative to the short-term context of monthly and seasonal stochastic variability of coastal morphology changes. The study provides new insights into the sand supply mechanisms underlying coastal erosion.