QUANTIFYING RATES OF ROCK CLIFF EROSION USING PS InSAR TECHNIQUE

Bedrock cliffs act as natural barriers against coastal hazards such as storm surges, flooding, extreme waves and runup; but they can fail due to hydrodynamic forcing and weathering. Rockfalls and landslides from eroding cliffs represent a significant, yet understudied hazard to humans and infrastructure. Understanding the complex dynamics of cliff erosion is essential for assessing and mitigating coastal hazards, especially in the context of climate change and sea-level rise, which are expected to change the hydrodynamic load and accelerate cliff retreat.

In Ireland, approximately 56 % of its extensive rocky coastline is rock-dominated. Despite this, rates and mechanisms of rock-cliff recession remain poorly quantified and understood and lag far behind those of soft coasts. This is partially due to difficulties associated with accessing hazardous coastal cliffs. Recent advances in Interferometric Synthetic Aperture Radar (InSAR) provide an opportunity to overcome these limitations by enabling millimetre-scale deformation monitoring over large and inaccessible areas.

In this study, Persistent Scatterer InSAR (PS-InSAR) was applied to assess long-term coastal cliff deformation along the west coast of Ireland. The aim of the study was to test the utility of PS InSAR as an approach for detecting and measuring rock coast erosion.  Two areas of interest (AOIs) were analysed: AOI-1 (20 km, from Doonbeg to Murrooghtoohy North) and AOI-2 (14 km, from Doonbeg to near Kilbaha). A total of 100 Sentinel-1 VV-polarised Single Look Complex (SLC) images were processed for each AOI, spanning from August 2016 to December 2024. Data preparation was carried out using the SNAP2StaMPS workflow, and PS-InSAR processing was performed using the StaMPS algorithm in MATLAB.

The results indicated that both AOIs have remained largely stable over the eight-year period, with maximum cumulative Line-of-Sight (LOS) displacements of –11 mm in AOI-1 and –15 mm in AOI-2. Despite this overall stability, localised clusters of LOS displacement were identified, particularly around Breaffa South, Co. Clare, suggesting ongoing cliff retreat. Active Deformation Area (ADA) analysis, based on a velocity-threshold approach, revealed that active deformation is not randomly distributed but is concentrated mainly in the southern part of the study area. Active PS points predominantly occur on sandstone, siltstone, and mudstone, whereas stable PS points are more commonly associated with mechanically stronger lithologies such as limestone, dolomite, and cherty units. An assessment of terrain geometry further demonstrated a strong dependence of PS detectability on slope, with breakpoint analysis identifying a critical threshold at approximately 21°, beyond which PS generation begins to decrease noticeably.

The PS InSAR results were validated using a high-resolution topographic data derived from drone and Bland–Altman analysis.  The analysis revealed a systematic positive bias of 1.02 mm, indicating that PS-InSAR slightly underestimates displacement. Limitations such as layover, coverage gaps, and reduced sensitivity to rapid deformation were identified to affect the application of PS InSAR. Overall, the findings demonstrated that PS-InSAR is an effective tool for identifying zones of cliff instability, identifying ground displacement over large areas, long-term monitoring and good use as a first step, but must be combined with traditional methodologies.