GRSG 35th Conference 2024 Abstract

Title: Assessment and classification of coastal erosion responsiveness due to extreme storm surges event using earth observation data in sand beaches of Southeast Sicily

Author: Gianfranco Di Pietro

Organisation: University of Catania

Sand beach shoreline detection from satellite data is a new frontier for remote sensing and geosciences (Vos, 2019). The large quantity of image series provided by space programmes such as EU-Copernicus have offered the possibility of extracting geospatial data over long time series on land changes. The study was conducted by iterating a procedure on a time series of Sentinel-2 data (5 years from 2018 to 2023) and Copernicus Marine Service data for 15 beaches in south-eastern Sicily. As described by Distefano et al (Distefano, 2024), located within the southeastern region of the Hyblaean Plateau, the study area constitutes a portion of the emerged northeastern-southwestern continental bulge of the African foreland. This bulge comprises Mesozoic-Cenozoic carbonate sedimentary successions (Pelagian Block).

The Hyblaean foreland is bordered to the northwest by the Caltanissetta Basin, an asymmetric trough positioned between the foreland and the belt. To the east, the plateau is demarcated by the Malta Escarpment, a feature separating the Pelagian shelf from the Ionian abyssal plain. The Gela-Catania foredeep, a Plio-Pleistocene basin, flanks the Hyblean foreland and extends towards the southern and SW offshore areas of Sicily..

For a first phase, using API of a Google Earth Engine provider it was downloaded all Sentinel-2 data of the beach in ROI and time-slice considered (Gomarasca, 2019). A pre-trained neural network (Vos, 2019) classify pixel in four classes: water, whitewater, sand, non-water. Classification is compared with a specific radiometric index: Modified Normalized Difference Water Index – MNDWI (Xu, 2006). Data is processing by an Otzu’s threshold algorithm for maximize interclass water-whitewater and get the pixel of shorelines. For a final downscaling process, it was used a border segmentation and a marching squares interpolation (Hanisch, 2004).

The second phase was the transect analysis. For each beach, a variable number of transects (as normal to the shoreline as possible) was considered. For each transect and each beach, a shoreline movement signal was created using geometric interpolation between the shoreline position extracted from the specific date and time of each satellite image.

The third phase was the detection of difference before and after storm surges of transects shoreline movement signal. Using the EU-Copernicus Marine Service dataset called Mediterranean Sea Physics Reanalysis (MEDSEA_MULTIYEAR_WAV), it was possible obtain a time-series data of physic parameter for considered beaches with a time-range from 1987 to 2023. These data were filtered using a peak-above-threshold method and finally obtaining a list of extreme events and other event variables.

Finally, for each event we determine a single parameter describing the movement of the coastal shoreline due to an extreme event, called ASM is the average movement of the surface of a beach, given by the difference of the area given considering the two shorelines, normalised in metres of beach length
Analysing the standard deviation of the ASM and Hs data distribution, we observed higher values for non-natural beaches and those near urban areas. Conversely, lower standard deviation values were noted for natural beaches. Intermediate values were found for urban beaches with offshore protection structures.

These findings, currently under refinement, could enable a quantitative assessment of the protective contribution of offshore breakwaters during past extreme events. Furthermore, they could facilitate a classification of sandy beaches based on their resilience to coastal erosion from extreme events. The significance of these results lies in their foundation on a quantitative approach utilizing near-real-time historical data.

Reference:
– Vos, K., Splinter, K.D., Harley, M.D., Simmons, J.A., Turner, I.L., 2019. CoastSat: A Google Earth Engine-enabled Python toolkit to extract shorelines from publicly available satellite imagery. Environmental Modelling & Software 122, 104528. https://doi.org/10.1016/j.envsoft.2019.104528
– Distefano, S., Gamberi, F., Borzì, L., Di Stefano, A., 2021. Quaternary Coastal Landscape Evolution and Sea-Level Rise: An Example from South-East Sicily. Geosciences 11, 506. https://doi.org/10.3390/geosciences11120506
– Gomarasca, M. A., Giardino, C., Bresciani, M., De Carolis, G., Sandu, C., Tornato, A., & Tonolo, F. (2019, June). Copernicus Sentinel missions for water resources. In Proceedings of 6th International Conference on Space Science and Communication. Johor: Malaysia.
– Xu, H., 2006. Modification of normalised difference water index (NDWI) to enhance open water features in remotely sensed imagery. Int. J. Remote Sens. 27, 3025–3033. https://doi.org/10.1080/01431160600589179