Kit Calcraft: A Generalised Data-Driven Shoreline Model at the Regional Scale
Kit Calcraft, UNSW Sydney (k.calcraft@unsw.edu.au)
Josh Simmons, University of Sydney
Lucy Marshall, Macquarie University
Kristen Splinter, UNSW Sydney
Introduction and scientific context:
Coastal shoreline models underpin long‐term coastal planning and hazard mitigation but are currently limited by site‐specific calibration and simplified process representations. While machine learning (ML) approaches have recently matched or exceeded the skill of traditional models in shoreline modelling, they have yet to demonstrate generalisation across diverse coastal environments, or any meaningful gain in application.
Methodology:
In this work, we introduce a generalised data-driven regional-scale shoreline model trained across 2,000 km of the southeast Australian coastline. The model adapts the Temporal Fusion Transformer (TFT) architecture, combining a sequence-to-sequence Long Short-Term Memory (LSTM) module with masked multi-head self-attention to learn both short- and long-range temporal dependencies. Static covariates unique to each transect are integrated via gating mechanisms that condition the learned relationship between wave action and shoreline response, enabling outputs to adapt to local site characteristics.
Results:
The model produced stable, state-wide forecasts without site-specific calibration, achieving a median RMSE of 12.92 m on unseen sites across a five-year holdout period. The model has difficulty in regions with complex coastal morphodynamics, while feature analysis further reveals that site-specific information that contextualises the primary forcing relationship between waves and shoreline response is essential for generalisation. Our findings demonstrate that end-to-end deep learning provides a viable pathway for developing regionally trained, generalised shoreline models, and motivates further exploration into the possibility of multi-regional models.
Audience:
Coastal scientists, machine learning researchers working with environmental time series, and practitioners in coastal/climate management.
Keywords:
shoreline modelling, regional-scale forecasting, generalisation, coastal science