Unravelling the sandy shorelines dynamics derived from satellite images

Abstract

Coastal zones all over the world have become heavily populated and developed due to the aesthetic value and diverse ecosystem services that they provide (Luijendijk et al. 2018). In recent years, however, climate change and anthropogenic pressures have exacerbated unprecedented coastal recession, threatening billions of dollars’ worth of coastal developments and infrastructure (Ranasinghe et al. 2012). In order to counteract the erosion trend and protect shoreline positions, it is necessary to carry out reliable assessments of shoreline dynamics to monitor the erosion process. Therefore, this thesis attempts to explain different spatial and temporal patterns of sandy shoreline evolution.The strict definition of shoreline is the physical land-water boundary (Boak & Turner 2005). However, considering the dynamic nature of water levels and the cross-shore and longshore sediment transport in the littoral zone, many coastal state indicators have been used as proxies to represent the ’true’ shoreline position for practical purposes (Boak & Turner 2005). The satellite derived shorelines (SDS), considered as a new type of coastal state indicator, provide a global shoreline dataset from 1984 to present. Hagenaars et al. (2017) stated that for long-time scales, similarities can be found between coastline dynamics based on the SDS positions and traditional indicators based on topographic measurements. Thus, for any coastal sandy stretch, time series of the SDS can be analyzed to get a first understanding of the coastline evolution in the period of 1984 – present.Three knowledge gaps are identified regarding using the SDS on unravelling shoreline dynamics:Does the time series of the SDS contain signatures of expected shoreline behaviour?To what extent is the limitation of the SDS important for unravelling shoreline dynamics? What is the application range of using the SDS for unravelling shoreline dynamics?The above knowledge gaps were addressed through the following steps. (1) We narrowed down the forcing types that would be focused on, including extreme storms, seasonal forcing, climate variability, land subsidence, sea level rise (SLR) and anthropogenic processes. (2) The shoreline changes at the eleven knowledge-intensive sites were focused on, each of which was selected based on the available documentation to verify if the influence of a specific forcing type could be unravelled through the SDS. The sites of interests include Narrabeen, Moruya and Pedro, Perranporth, Ocean Shores (CRLC), the Nile Delta, Perth, Ocean Beach, Fire Island, Gatseau sandspit and Cap Ferret sandspit (SW France), the Gulf of Valencia and Wrightsville Beach. (3) We decomposed the time series of the SDS with a range of data analysis methods, in order to extract spatiotemporal patterns of shoreline variation and correlate the variation patterns to forcing types at the eleven sites. (4) Three indices were calculated to classify rotational/non-rotational and seasonal/non-seasonal beaches based on the analysis results of the SDS. (5) The knowledge gaps were addressed through comparing the shoreline behaviour patterns derived from the SDS to the analysis within the related literature.We found that the spatiotemporal patterns of shoreline variation on the seasonal scale (due to seasonal forcing), inter-annual scale (due to climate variability) and decadal scale (due to land subsidence) extracted through the SDS are largely in line with the conclusions listed in the related literature, in which the dataset such as field measurements or video monitoring were used. Moreover, the influence of groins and beach nourishment projects on shoreline changes can also be clearly unravelled using the SDS. However, the SDS cannot be considered as the best solution to study shoreline variation governed by storms, considering the low frequency of satellite image acquisition and the influence of the image composite technique. Moreover, only using the SDS to assess the shoreline responses to SLR may not provide promising results, since the limited length of the record (over 30 years) make it hard to capture the influence of SLR on open coasts on long-term scales.To sum up, the SDS could be used as an ideal tool for unravelling coastline dynamics on seasonal, inter-annual and decadal scales governed by the seasonal forcing, climate variability and land subsidence, respectively. The shoreline changes caused by beach nourishment projects can also be clearly unravelled with the SDS. The introduction of new satellite missions can be expected in the near future, which helps to obtain an increasing temporal, spatial and spectral resolution of satellite images (Hagenaars et al. 2018). Therefore, it is highly possible that the SDS will become a more powerful tool for studying shoreline changes governed by extreme storms as time goes by. Furthermore, when the longer-recorded SDS are achieved in the future, it can be expected that the capabilities of using the SDS for studying the long-term shoreline variation governed by SLR will be improved.