Abstract
The white chalk cliffs of southern England are iconic but highly vulnerable to climate change-triggered accelerations in erosion. This study quantifies long-term trends in cliff retreat rates using cosmogenic radionuclide analysis ($^{10}$Be) and numerical modeling across the late-Holocene. By optimizing a coastal evolution model to topographic data and $^{10}$Be concentrations at four sites, we reveal a strong link between retreat rates and the rate of sea-level rise. Our results indicate that current retreat rates are exceptionally high, matching levels last seen 5,300–6,800 years ago when sea-level rise was significantly more rapid. Furthermore, the study distinguishes the role of intertidal weathering in chalk versus sandstone, highlighting its critical influence on shore platform development and the "wave versus weathering" debate.
Introduction
Predicting how coastlines will respond to climate change is a significant challenge in geomorphology, primarily because most available data comes from historical records (maps and aerial photography) spanning only about 150 years. This relatively short-term view makes it difficult to distinguish between natural variability and genuine acceleration caused by modern climate shifts. Additionally, a long-standing debate persists regarding the primary drivers of cliff retreat: is it the mechanical force of waves or the chemical and physical weathering of the rock? This research addresses these gaps by looking back 7,000 years to establish a millennial-scale baseline for cliff retreat in weak lithologies like chalk, comparing it to previous findings in more resistant sandstone.
Methods and Data
The study employed a multi-objective optimization approach to reconcile numerical models with physical evidence across four chalk sites on the south coast of England: Seven Sisters, St Margaret’s at Cliffe, Hope Gap, and Beachy Head. These sites were compared against previously studied sandstone sites at Bideford and Scalby, UK.
Cosmogenic Radionuclide ($^{10}$Be) Analysis
Researchers measured the concentration of Beryllium-10 in flint nodules found on shore platforms. Because $^{10}$Be accumulates when rock is exposed to cosmic rays, its concentration serves as a geochronological clock. By analyzing these concentrations across the shore platform, the researchers could determine the speed at which the cliff has moved landward over thousands of years.
Numerical Modeling and Optimization
A process-based Rock Platform Model (RPM) was utilized to simulate coastal evolution. Unlike traditional steady-state models, this transient model allows retreat rates to change over time in response to varying sea levels. The researchers used Markov Chain Monte Carlo (MCMC) simulations to find the best fit between the model's predictions and two real-world datasets: the measured $^{10}$Be concentrations and high-resolution LiDAR topographic data.
Key Insights and Results
The Weathering Divide
A significant finding of this research is the stark difference in erosion drivers between rock types. In sandstone sites, subaerial weathering has been negligible during the Holocene, with wave action performing the bulk of the erosion. In contrast, for chalk sites, intertidal weathering (processes like wetting, drying, and salt crystallization) is 2 to 3 orders of magnitude more active. This weathering plays a vital role in weakening the platform, facilitating its subsequent removal by waves and driving the long-term development of the shore platform system.
Sea-Level Rise (SLR) Coupling
Long-term cliff retreat rates are closely coupled with the rate of sea-level rise. The fastest retreat rates in the record occurred approximately 7,000 years ago when SLR was at its Holocene peak. The model results suggest that the relatively rapid historical cliff retreat rates observed at these sites over the last 150 years last occurred between 5,300 and 6,800 years ago.
Modern Acceleration
Modern retreat rates (22–39 cm/yr) are an order of magnitude faster than the Holocene average. During the period 5,300–6,800 years ago when similar rates were observed, the rate of relative sea-level rise was a factor of 5–9 times more rapid than during the recent historical record. This suggests that modern factors are driving erosion at rates disproportionate to current sea-level rise speeds.
Limitations and Future Directions
While the model successfully captured millennial trends, it struggled to replicate the "stepped" topography often observed on chalk shore platforms. These steps are likely caused by alternating layers of hard flint and soft chalk, indicating that internal rock stratigraphy (heterogeneous lithology) is a key factor currently overlooked in process-based models.
Furthermore, the model does not yet account for beach dynamics. Beaches act as a "double-edged sword," protecting cliffs from waves while also providing abrasive material that can accelerate erosion. The mismatch in $^{10}$Be data near cliff bases suggests that beach thinning in recent centuries may be a primary driver of the observed acceleration in retreat rates.
Conclusion
By combining cosmogenic dating with transient numerical modeling, this study demonstrates that England's iconic chalk cliffs are retreating at rates not seen for over 5,000 years. The results highlight that while sea-level rise drives long-term trends, rock-specific factors like intertidal weathering are dominant controls that must be integrated into future coastal evolution models to accurately forecast the impacts of climate change on fragile rock coasts.
Summary updated