GRSG 34th Conference 2023

Title: Assessing Coastal Granulometry from Satellite Synthetic Aperture Radar (SAR) Data.

Author: Sophie Mann

Abstract:

The UK’s coastline has an increasing risk of erosion and associated hazards. Sediment grain size plays a crucial role in determining beach morphology and shoreline changes, however traditional surveying methods used to obtain grain size are time-consuming and costly. Therefore, the use of remotely sensed data can help us to achieve a ‘real-time’ map of sediment grain size present in coastal areas. The knowledge of grain size in coastal areas can improve understandings of coastal dynamics and better inform beach management and protection policies.

The study explores the use of Synthetic Aperture Radar (SAR) spaceborne imagery as a tool for coastal granulometry assessment. We tested the backscatter variation against field measurements at several beaches across the Southwest and the East coasts of England. Early results show that the backscatter observed from the SAR data increases as grain size increases, displaying a strong visual correlation and moderate visual correlation between SAR backscatter and the median sediment grain size, from Sentinel-1 C-band (5.6cm wavelength) SAR and NovaSAR S-band (9.4cm wavelength) SAR respectively. We attribute the results observed from Sentinel-1 data mainly to the sediment size analysed in this study compared to the wavelength of C-band SAR, along with increasing surface roughness as sediment size increases. A moderate correlation from NovaSAR data is likely due to factors other than grain size because no correlation was found at Chesil Beach, despite clearly graded sediment being present.

Further work focuses on deriving sediment grain size from Sentinel-1 backscatter using Google Earth Engine. Initial results from this analysis show the potential to predict sediment grain size to within 0.15 mm accuracy for some stoney beaches in England. We hope that these findings will allow for quick sediment monitoring and surveying of UK beaches and assessment of coastal vulnerability for coarse sediment beaches.

The UK’s coastline has an increasing risk of erosion and associated hazards. Sediment grain size plays a crucial role in determining beach morphology and shoreline changes, however traditional surveying methods used to obtain grain size are time-consuming and costly. Therefore, the use of remotely sensed data can help us to achieve a ‘real-time’ map of sediment grain size present in coastal areas. The knowledge of grain size in coastal areas can improve understandings of coastal dynamics and better inform beach management and protection policies.

The study explores the use of Synthetic Aperture Radar (SAR) spaceborne imagery as a tool for coastal granulometry assessment. We tested the backscatter variation against field measurements at several beaches across the Southwest and the East coasts of England. Early results show that the backscatter observed from the SAR data increases as grain size increases, displaying a strong visual correlation and moderate visual correlation between SAR backscatter and the median sediment grain size, from Sentinel-1 C-band (5.6cm wavelength) SAR and NovaSAR S-band (9.4cm wavelength) SAR respectively. We attribute the results observed from Sentinel-1 data mainly to the sediment size analysed in this study compared to the wavelength of C-band SAR, along with increasing surface roughness as sediment size increases. A moderate correlation from NovaSAR data is likely due to factors other than grain size because no correlation was found at Chesil Beach, despite clearly graded sediment being present.

Further work focuses on deriving sediment grain size from Sentinel-1 backscatter using Google Earth Engine. Initial results from this analysis show the potential to predict sediment grain size to within 0.15 mm accuracy for some stoney beaches in England. We hope that these findings will allow for quick sediment monitoring and surveying of UK beaches and assessment of coastal vulnerability for coarse sediment beaches.