Shoreline modelling on timescales of days to decades

Abstract

In the context of increased probability of coastal erosion and flooding associated with climate change, there is a pressing need to predict future shorelines at both short-(daily) and mediumterm (decadal) timescales. Such predictions are essential for the assessment of the climateresilience of the world's coastlines and the delivery of effective, economic and data-informed coastal management. Coastal managers currently lack these predictions and there are many different modelling approaches to inform where increased coastal protection, adaption measures or future infrastructure developments should be focussed. Promising modelling advances have recently been made, particularly in the context of reduced complexity models. This paper reviews various numerical modelling approaches to predicting shoreline and coastal morphological change, comments on some of the most promising methods used to-date, provides some guidance on model selection, and highlights important future research directions and challenges to progress.

A Model Integrating Satellite‐Derived Shoreline Observations for Predicting Fine‐Scale Shoreline Response to Waves and Sea‐Level Rise Across Large Coastal Regions

“…However, since its initial development in southern California, the Yates et al. (2009) model has been proven to be skillful across diverse coastal settings (Castelle et al., 2014; Hunt et al., 2023; Montaño et al., 2020). Poorer model performance is generally encountered in northern California, particularly across Humboldt County, which we hypothesize is due to large signals of fluvial sediment input and the presence of large‐scale sand waves (∼200–1,000 m wavelength) in the region, whose dynamics are not well resolved in the context of the model governing equation.…”

“…Innovations to improve the fidelity of coastal physics‐based models have had a noticeable impact on the skill of coastal‐change simulations during individual storm events, but so far have arguably not had the same effect on long‐term simulation of beach processes. On the other hand, simplified, parametrized, and increasingly probabilistic coastal change models, which are most often based on the concept of “equilibrium” (e.g., Davisdon et al., 2013; Hunt et al., 2023; Miller & Dean, 2004; Wright & Short, 1984; Yates et al., 2009), have provided the biggest recent innovation in the prediction of long‐term (e.g., multi‐annual to decadal+) coastal change. Although both physics‐based and parameterized (reduced‐complexity) coastal‐change models will benefit from increased availability of observations, we believe the simplified models will receive the greatest returns from data‐integration efforts for a number of different reasons: (a) simplified models can be readily calibrated to real‐world, site‐specific shoreline observations in contrast to more expensive, monolithic models, which also require full bathymetric and topographic surveys for validation, (b) simplified models, mainly due to their significantly shorter runtimes, can be readily applied in a probabilistic sense (e.g., using Monte Carlo methods), and thus will excel in propagating, quantifying, and balancing uncertainty (in both modeling and observational components) in contrast to more expensive and consequently more deterministic models, (c) simplified models can be readily adapted to produce multi‐model ensemble predictions, and (d) simplified models are amenable to data‐assimilated operational modeling (e.g., based on EnKF methods) as well as scenario‐based modeling of future coastal change.…”

“…However, this is perhaps expected since the equilibrium shoreline-change model, that is, Yates et al (2009), which dominates the short-term signal of change, was conceived from local shoreline behavior and observations thereof at Torrey Pines Beach in southern California. However, since its initial development in southern California, the Yates et al ( 2009) model has been proven to be skillful across diverse coastal settings (Castelle et al, 2014;Hunt et al, 2023;Montaño et al, 2020). Poorer model performance is generally encountered in northern California, particularly across Humboldt County, which we hypothesize is due to large signals of fluvial sediment input and the presence of large-scale sand waves (∼200-1,000 m wavelength) in the region, whose dynamics are not well resolved in the context of the model governing equation.…”

“…Accurate predictions of coastal erosion in response to sea‐level rise, changing wave conditions, and reduced natural sediment supplies are increasingly sought by coastal managers to assess the impacts of climate change on beaches (Le Cozannet et al., 2019; Masselink et al., 2016; Vitousek, Barnard, & Limber, 2017). Several well‐tested and emerging models are capable of simulating coastal erosion (Hunt et al., 2023; Montaño et al., 2020; Ranasinghe, 2020; Roelvink, 2011; Toimil, Camus, et al., 2020). However, most models inherently capture a limited number of processes occurring at a narrow geographic scope because of computational or data availability constraints.…”

“…Recent advances in satellite remote‐sensing analysis provide a window into the recent past and current state of the world's beaches (Luijendijk et al., 2018) and their large‐scale vulnerability to climatic forces like El Niño (Vos et al., 2023). By leveraging the large streams of data offered from satellites, reduced‐complexity coastal‐change models seem poised for success in a challenging field of study owing to the newly found “treasure trove” of data (Barnard & Vitousek, 2023; Hunt et al., 2023; Vitousek et al., 2023a).…”