F. Galiforni-Silva1*, S. Dan2, S. de Vries1, C. Hallin1,3, B. J. A. Huisman4, M. Maarse4, F. van Rees5, A. J. H. M. Reniers1
1 Delft University of Technology, The Netherlands; 2 Flanders Hydraulics, Belgium; 3 Lund University, Sweden; 4 Deltares, The Netherlands; 5 NIOZ Royal Netherlands Institute for Sea Research, The Netherlands
* Corresponding author: f.galifornisilva@tudelft.nl
Introduction
Nature-based solutions are increasingly being adopted for coastal protection worldwide, balancing effective hazard mitigation with reduced environmental impact compared to conventional approaches. While several benefits exist, nature-based solutions are complex and, consequently, hard to forecast given their associated natural dynamics. Digital Twins (DT) emerged as a potential tool integrating process-based and data-driven models to predict system behavior and provide stakeholders with insights into the system’s potential use and reliability. A DT is defined as a digital copy of a physical system based on models and data-driven methods, where continuously updated information from its real counterpart is used as boundary condition for testing and forecasting. Although the concept has been used in different fields (Dal Zilio et al., 2023; Pregnolato et. al, 2022), its use in coastal applications and nature-based solutions is still in its preliminary stages (Jiang et al, 2021).
Objective and Methods
Here, we present and test a first conceptualization of the planned framework to be used for the development of a DT for a hybrid dune-dike system, aiming to be applied for planning and decision-making on coastal management projects. For a hybrid dune-dike system, the virtual twin needs to reproduce not only immediate changes but also the decadal evolution of the system. Thus, we use a coupled shoreline model (ShorelineS, Roelvink, et al., 2020) and beach-dune model (CS model, Hallin et al., 2019) as the base of our DT. We apply the proof-of-concept in several schematic and simplified scenarios and one real case (Hondsbossche Duinen, NL). Schematic scenarios aim to test specific situations, such as increased beach width and dune supply, whereas the real case serves as a general applicability test. For the real case, we simulate the year 2020-2021. Topographic maps made available by Rijkswaterstaat are used to calculate the initial shoreline position and dune characteristics. Wave and wind data made available by Rijkswaterstaat and KNMI, respectively, are used as boundary conditions.
Results
Results show that the proof of concept can simulate most of the overall trends regarding the alongshore spread of sediment and the sediment supply to the dunes. Patterns such as deposition/erosion due to alongshore transport gradients, changes in supply due to beach width variability, and dune erosion due to storm conditions were successfully simulated within the proof-of-concept. For Hondsbossche Duinen, the predictive DT was able to qualitatively simulate the overall beach-dune patterns in terms of volume change. However, even though the overall beach-dune volume trends match the measured data, certain parameters, such as shoreline position, did not perform well for certain coastal stretches, suggesting that better parametrizations for cross-shore sediment redistribution may need to be incorporated into the DT. Further steps include the development of the data assimilation layer and the incorporation of detailed parametrizations from specific process-based models. Nonetheless, the current setup can capture the overall dynamics found in the study area while leaving direct connection points for further development into a DT.
References
Dal Zilio, L., Giardini, D., Carbonell, R., & Wiemer, S. (2023). Harnessing the potential of digital twins in seismology. In Nature Reviews Earth & Environment (Vol. 4, Issue 8, pp. 510–512). Springer Science and Business Media LLC. https://doi.org/10.1038/s43017-023-00469-y
Hallin, C., Larson, M., & Hanson, H. (2019). Simulating beach and dune evolution at decadal to centennial scale under rising sea levels. In J. M. Dias (Ed.), PLOS ONE (Vol. 14, Issue 4, p. e0215651). Public Library of Science (PLoS). https://doi.org/10.1371/journal.pone.0215651
Jiang, P., Meinert, N., Jordão, H., Weisser, C., Holgate, S., Lavin, A., Lütjens, B., Newman, D., Wainwright, H., Walker, C., & Barnard, P. (2021). Digital Twin Earth -- Coasts: Developing a fast and physics-informed surrogate model for coastal floods via neural operators (Version 1). arXiv. https://doi.org/10.48550/ARXIV.2110.07100
Pregnolato, M., Gunner, S., Voyagaki, E., De Risi, R., Carhart, N., Gavriel, G., Tully, P., Tryfonas, T., Macdonald, J., & Taylor, C. (2022). Towards Civil Engineering 4.0: Concept, workflow and application of Digital Twins for existing infrastructure. In Automation in Construction (Vol. 141, p. 104421). Elsevier BV. https://doi.org/10.1016/j.autcon.2022.104421
Roelvink, D., Huisman, B., Elghandour, A., Ghonim, M., & Reyns, J. (2020). Efficient Modeling of Complex Sandy Coastal Evolution at Monthly to Century Time Scales. In Frontiers in Marine Science (Vol. 7). Frontiers Media SA. https://doi.org/10.3389/fmars.2020.00535
