A Clustering Approach for Predicting Dune Morphodynamic Response to Storms Using Typological Coastal Profiles: A Case Study at the Dutch Coast

Abstract

Dune erosion driven by extreme marine storms can damage local infrastructure or ecosystems and affect the long-term flood safety of the hinterland. These storms typically affect long stretches (∼100 km) of sandy coastlines with variable topo-bathymetries. The large spatial scale makes it computationally challenging for process-based morphological models to be used for predicting dune erosion in early warning systems or probabilistic assessments. To alleviate this, we take a first step to enable efficient estimation of dune erosion using the Dutch coast as a case study, due to the availability of a large topo-bathymetric dataset. Using clustering techniques, we reduce 1,430 elevation profiles in this dataset to a set of typological coastal profiles (TCPs), that can be employed to represent dune erosion dynamics along the whole coast. To do so, we use the topo-bathymetric profiles and historic offshore wave and water level conditions, along with simulations of dune erosion for a number of representative storms to characterize each profile. First, we identify the most important drivers of dune erosion variability at the Dutch coast, which are identified as the pre-storm beach geometry, nearshore slope, tidal level and profile orientation. Then using clustering methods, we produce various sets of TCPs, and we test how well they represent dune morphodynamics by cross-validation on the basis of a benchmark set of dune erosion simulations. We find good prediction skill (0.83) with 100 TCPs, representing a 93% input and associated computational costs reduction. These TCPs can be used in a probabilistic model forced with a range of offshore storm conditions, enabling national scale coastal risk assessments. Additionally, the presented techniques could be used in a global context, utilizing elevation data from diverse sandy coastlines to obtain a first order prediction of dune erosion around the world.

Estimating dune erosion at the regional scale using a meta-model based on neural networks

“…Additionally, water level time series at 10 min intervals are available from four tide gauges along the coast (Fig. 2), for which a tidal analysis was performed in Athanasiou et al (2021) and the local tide and storm surge level (SSL, difference between water level and tidal level) were extracted. The wave and water level time series are available from Rijkswaterstaat, an agency of the Ministry of Infrastructure and Water Management of the Netherlands.…”

“…In Athanasiou et al (2021) 1430 of the JARKUS transects which represent sandy beach and dune systems, were chosen and the elevation profiles were extracted for the year 2019. In that study, various morphological and hydrodynamic pa- rameters were calculated per profile and episodic dune erosion volume was simulated per profile with the process-based model XBeach for 10 representative storms.…”

“…A two-phase clustering approach similar to that in Athanasiou et al (2021) was applied to group the available profiles into 100 clusters based on their similarities, each described by its representative profile (centroid) called a typological coastal profile (TCP) (red dots in Fig. 2).…”

“…2). We excluded some of the 1430 profiles from Athanasiou et al (2021), which were found to be at highly dynamic areas (e.g., transect at the tips of the Wadden islands where no clear dune features were identified) or transects at the non-exposed side of the Wadden islands, leaving 1368 transect for our study. The 1368 profiles were first clustered into 300 initial groups, leading to 300 initial TCPs, based on a set of 10 morphological and hydrodynamic parameters that characterize each profile (Table 1).…”

“…Then a second clustering was performed on the 300 initial TCPs, where their simulated dune response to 10 different storms was taken into account, to group them in 100 clusters and obtain the final TCPs. For detailed information on the exact methodology we refer to Athanasiou et al (2021). Most of the resulting TCPs were located at the Delta coast (∼ 55 %), indicating that the similarity of the profiles there is lower in comparison to the Holland (∼ 25 %) and Wadden (∼ 20 %) coasts.…”