Coastal observatories integrate operational numerical modelling and data to produce targeted tools for coastal early warning and risk management. They explore the usage of computational services and web-based tools to produce timely and interactive products in support of coastal management, building the foundation for Digital Twins from the ocean to the river basin (Rodrigues et al., 2021). Examples of computational services in this context include OPENCoastS (Oliveira et al., 2020) and SURF (Trotta et al., 2021), for on-demand predictions, and WebGIS platform applications, customized to meet user requirements.
In Copernicus Marine Service National Collaboration Programme’s project CONNECT, a multi-purpose coastal service was established for two sites, based on coastal observatories technology (Rodrigues et al., 2021), merging high-resolution model predictions from OPENCoastS deployments and data from the national infrastructure CoastNet (Figure 1). By providing both pre-configured products (such as maps, automatic quality assessments and indicator results) and the capacity for users to build products on-the-fly (such as probing results as time series and building tailored dashboards with user selected information), the CONNECT service is one of the first core coastal Digital Twins applicable for flooding and coastal pollution management (Figure 2).
Figure 1 – CONNECT’s coastal service architecture
Figure 2 – Sample user services in the Tagus estuary: Virtual sensors and configurable dashboard.
Computational services are also applied to the construction of the river predictions (Jesus et al., in review). Particularly suited for dam-controlled river basins, a deep learning service was used to predict river flows at the upstream boundary of CONNECT sites, based on Multilayer Perceptron algorithms, potentially reducing severe phase errors during extreme events in the estuary than conventional persistence approaches. An on-demand platform is being built to allow users to setup their own AI model, with multiple choices for deep learning algorithms, input and output data , and training and validation periods. These tools will be validated in two sites in Africa, addressing co-creation challenges with local communities, and in two sites in Portugal, to address urban-driven contamination and shellfish farm needs.
References
Rodrigues M., Martins R., Rogeiro J., Fortunato A.B., Oliveira A., Cravo A., Jacob J., Rosa A., Azevedo A., Freire P. , 2021. A Web-Based Observatory for Biogeochemical Assessment in Coastal Regions. Journal of Environmental Informatics, 38(1), 1-15, https://doi.org/10.3808/jei.202100450
Oliveira A., Fortunato A.B. Rogeiro J., Teixeira J., Azevedo A., Lavaud L., Bertin X., Gomes J., David M., Pina J., Rodrigues M., Lopes P., 2019. OPENCoastS: An open-access service for the automatic generation of coastal forecast systems, Environmental Modelling & Software, 124, 104585, https://doi.org/10.1016/j.envsoft.2019.104585.
Trotta F., Federico I., Pinardi N., Coppini G., Causio S., Jansen E., Iovino D., Masina S., 2021. A Relocatable Ocean Modeling Platform for Downscaling to Shelf-Coastal Areas to Support Disaster Risk Reduction , Frontiers in Marine Science, 8, pp. 317.
Jesus, G., Mardani, Z., Alves, E., Oliveira, A. Under review, Deep Learning-Based River Flow Forecasting with MLPs: Comparative exploratory analysis applied to the Tejo and the Mondego rivers, Sensors.
How to cite: Oliveira, A.: Computational services for coastal observatories: the building blocks in support of coastal resilience, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-8962, https://doi.org/10.5194/egusphere-egu25-8962, 2025.
A coastal digital twin of the ocean (C-DTO) is a powerful framework that integrates Earth-Observation (EO) and ground data with predictive models, providing a dynamic, actionable view of coastal environments. It serves as a critical tool for monitoring, forecasting, and planning responses to natural and human-induced changes, fostering more resilient and sustainable coastal management.
We present recent advancements in developing a C-DTO that incorporates five interlinked cores—waves, circulation, sediment transport, vegetation, and flooding. These cores enhance realism by including interactions among Earth system processes. Critical feedback mechanisms, such as wave-current, wave-sediment, and current-vegetation interactions, are essential for accurately representing coastal dynamics.
Built on deterministic foundations, the system integrates diverse observational data, from bathymetry and vegetation characteristics to ocean parameters such as waves, tracers, and sea level, supporting calibration, validation, and data assimilation. The integration of machine learning further enhances system capabilities, enabling the simulation of more complex processes and scenarios.
Accessibility and flexibility are central to the framework, allowing deployment across diverse geographic areas and temporal scales. It supports process studies, forecasting, event-based and long-term simulations, and what-if scenario testing, accommodating both gray (engineered) and green (nature-based) adaptation strategies.
Case studies illustrate the framework’s versatility and effectiveness. Examples include characterizing extreme events such as storm surges during the Ianos Medicane, evaluating the potential of seagrass as a nature-based coastal protection strategy, assessing the impacts of breakwater construction, identifying optimal sites for Posidonia oceanica meadow restoration, and analyzing the effects of the MOSE barrier closure in the Venice Lagoon.
This system empowers policymakers and researchers to assess the impacts of climate change and human interventions on coastal systems. By simulating complex scenarios, identifying risks, and investigating processes, it supports informed decision-making for enhanced coastal resilience and effective ecosystem conservation and restoration.
How to cite: Causio, S., Federico, I., Shirinov, S., Alessandri, J., Piermattei, V., Bonamano, S., Piazzolla, D., Mentaschi, L., Boetti, M., Takeo Carvalho, J., Marcelli, M., Coppini, G., and Pinardi, N.: Integrated Coastal Digital Twin framework for enhancing sustainable, science-based coastal resilience and adaptation strategies, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-8906, https://doi.org/10.5194/egusphere-egu25-8906, 2025.
Inspired by Walter Munk’s proposition of an ‘orbital gap’ in climate variability, this study seeks to determine the linkage between cycles within such orbital gap and ‘sunny-day’ coastal flooding. The orbital gap represents astronomical variability between the nodal lunar cycle (18.6 y) and the shortest Milankovitch cycle (Earth’s axial precession, ~20,000 y). Such astronomical variability arises from Earth-Moon-Sun orbital non-linear interactions, some of them represented by multiples of the lunar nodal cycle and the radiational periods of the sun (solar activity, ~10-11 y). Periodicities <100 y associated with these astronomical effects are fitted in this study to the signal of daily maximum in water levels of tide stations in the eastern United States with records greater than 100 y. In Boston, for example, the lunar nodal cycle by itself explains 73% of the variance of the daily maxima in water level. Adding twice, thrice and six times the nodal cycle, plus twice the solar period, yields a fit that explains 81% of the variance. This fit is repeated for other locations on the eastern United States and allows projections for periods of most vulnerability to sunny-day flooding in the rest of the 21st century. This approach is likely to be applicable in other parts of the world to provide early warnings of susceptibility for ocean-induced coastal zone flooding.
How to cite: Valle-Levinson, A. and Yang, J.: Early warning for sunny-day flooding in the East Coast of the United States: global implications, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-7350, https://doi.org/10.5194/egusphere-egu25-7350, 2025.
How to cite: Palmer, M., Mansfield, T., Kay, S., Wihsgott, J., Tilstone, G., Menon, P., and Ford, D.: SyncED-Ocean: Towards a Digital Twin Coastal Ocean Ecosystem, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-11836, https://doi.org/10.5194/egusphere-egu25-11836, 2025.
This study focuses on a test case during the summer of 2024 to assess the impact of marine heatwaves on the Gulf of Hammamet ecosystem.
The Gulf of Hammamet, located along the eastern coast of Tunisia, is a vital ecological and socio-economic region in the Mediterranean basin. Renowned for its rich marine biodiversity, productive fisheries and Mariculture, and thriving tourism industry, the Gulf supports the livelihoods of coastal communities and plays a key role in regional economic stability. However, this ecosystem is increasingly threatened by the intensification of marine heatwaves, a phenomenon driven by climate change.
Marine heatwaves, characterized by prolonged periods of abnormally high sea surface temperatures, have become more frequent and severe in recent decades. These events disrupt the delicate balance of marine ecosystems, leading to biodiversity loss, habitat degradation, and altered fisheries productivity. For the Gulf of Hammamet, these impacts are particularly concerning, as they exacerbate existing vulnerabilities and pose significant challenges to sustainable coastal management. Understanding the mechanisms and consequences of marine heatwaves in this region is essential for enhancing coastal resilience. This study aims to bridge critical knowledge gaps by assessing the physical and ecological impacts of heatwaves in the Gulf of Hammamet, providing valuable insights to inform risk mitigation strategies and adaptive management practices.
A high-resolution hydrodynamic model is employed, fed at the open boundaries by Mediterranean Sea Physics Reanalysis provided by Copernicus Marine Service, to capture localized physical processes while maintaining consistency with broader-scale ocean dynamics. The model setup incorporates detailed boundary conditions and region-specific parameters to enhance its predictive capabilities. Key variables, such as sea surface temperature, currents, and heat fluxes, are simulated to analyze the onset, intensity, and progression of marine heatwaves during the specified period. Validation of the model is achieved through comparisons with global datasets and climatological records relevant to summer 2024, ensuring reliable insights without relying on in-situ measurements. This approach enables a robust analysis of how heatwaves influence physical conditions and their cascading effects on the Gulf's ecosystem, including shifts in water column stratification and potential impacts on marine biodiversity.
However, we acknowledge that in-situ data are indispensable for a more robust and comprehensive validation of the model, and future efforts will aim to incorporate such datasets to improve accuracy and reliability.
How to cite: Gana, S., Cucco, A., and Mkhinini, N.: Marine Heatwaves in the Gulf of Hammamet: A Case Study from Summer 2024, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-12382, https://doi.org/10.5194/egusphere-egu25-12382, 2025.
Ocean services to society gain in their importance as we strive to intensify the multiple uses of the sea, to cope with the ever-increasing coastal population flows, and to meet the higher and more widespread expectations for better life standards. Social concerns related to the sea demand further integration and merging of marine data and research priorities to enhance their societal and economic potential. There have been significant changes in the way policies, marine resource management, coastal planning and efficient marine operations are perceived and implemented, such as integrated marine policies and spatial planning approaches. Today the quest for environmental security, based on the concepts of sound ocean governance, knowledge sharing and the controlled use of resources, is the enabler of prosperity, sustainability and peace. There is a greater understanding that actions must be based on informed decisions, which requires an integrated approach based on networked management and decision support systems, and the sustained delivery of reliable and routine marine data.
Indeed, the digital era has opened new realms for ocean data delivery. More users are dependent on reliable information deriving from multiple data sources, while non-professional users are increasing in numbers with different demands from those of professional users. As technology further feeds the value addition chain of data to information, knowledge and intelligence, innovative downstream services are evolving to popularise the uptake of data, catering the demand of a knowledge-based society seeking faster and selective access to information. The STREAM project builds on these aspects to provide at fingertips, on-demand, pixel-based data to general users, including the common citizen, via popular smart mass media without requiring data science skills. The main concept in STREAM aligns with the target to popularise data for use and re-use to the benefit of society at large.
STREAM’s intuitive design allows users to seamlessly access data with a simple, personalised “view, select, click, and go” facility, eliminating the need to download large volumes of data. The platform delivers precise data for the chosen location and time window, quickly, free of charge, and in a user-friendly data format. By combining satellite observations with numerical simulations, STREAM offers three key data modes: Before, Now, and Next. Users can access historical and climatological data (spanning 30 years), updated real-time observations, and predictions, all from a single platform. Additional features include the ability to save and revisit selections, receive mobile notifications for user-defined thresholds and locations, and monitor change over selected sites and time slots. These functionalities empower users to assess variations, obtain precise data for assessments, and enable informed decision-making.
The service covers the sea area around the Maltese Islands and the southern Sicilian coast. It can be accessed via the STREAM web platform (www.stream-srf.com) or through the mobile app available for Android (https://play.google.com/store/apps/details?id=com.stream_srf.app.twa&hl=en) and iOS (https://apps.apple.com/us/app/stream-srf/id6648788013?platform=iphone) devices.
STREAM further provides additional web services on its portal (https://www.stream-srf.com/products/) in the form of routine updated data products and forecasts. Two such services are the Marine Heat Wave Service and the ROSARIO Marine Forecast.
How to cite: Drago, A.: Data-sharing services to benefit society, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-13535, https://doi.org/10.5194/egusphere-egu25-13535, 2025.
In the United States, the National Oceanic and Atmospheric Administration’s National Ocean Service (NOAA/NOS) has developed a statistical model to predict the daily risks of high tide flooding (HTF), for forecast leads of up to one year, at 98 tide gauge locations along the US coastline. NOAA/NOS predicts the daily probability of exceedance of hourly water levels above a specified flood threshold by combining the tide prediction with the (extrapolated) linear trend of mean sea level and a probabilistic prediction of the anomalous hourly non-tidal residual (NTR). In turn, the NTR anomaly prediction is made up of two components: (1) a prediction of monthly mean NTR, currently based upon the observed autocorrelation function of linearly detrended NTR, with uncertainty based upon the observed standard deviation of monthly NTR; and (2) a prediction of the probability distribution function (PDF) of hourly NTR anomalies, which uses observed historical dependence upon the total water level and is assumed to be Gaussian. These forecasts are available at https://tidesandcurrents.noaa.gov/high-tide-flooding/monthly-outlook.html.
In this presentation, we introduce an updated version of the NOAA/NOS HTF framework, with three key improvements: (1) the trend estimate is determined empirically and is allowed to be nonlinear; (2) monthly mean SLA is predicted by either an empirical or dynamical climate forecast model, and includes an ensemble spread; and (3) the PDF of hourly NTR anomalies is non-Gaussian and determined separately for each month from past observations using a “stochastically-generated skewed” (SGS) distribution. Skill of the updated version is compared to the original (currently operational) version at all tide gauge locations, and the impact of each of the improvements on skill is diagnosed. Further prospects for improvement of the HTF framework are also discussed.
How to cite: Newman, M., Albers, J., Callahan, J., Colin, M., Dusek, G., Hovenga, P., Kavanaugh, K., Sweet, W., and Wang, Y.: Development of NOAA’s Next-generation Prediction System for High Tide Flooding Risk on Subseasonal to Annual Timescales, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-13714, https://doi.org/10.5194/egusphere-egu25-13714, 2025.
The Horizon Europe FOCCUS project ( Forecasting and Observing the Open-to-Coastal Ocean for Copernicus Users, foccus-project.eu), brings together 19 partners from 11 countries. In collaboration with Member State Coastal Systems (MSCS) and users this interdisciplinary and international effort aims to improve the existing capabilities of CMEMS and develop innovative coastal products that will contribute to resilience against coastal hazards and climate change. Recently endorsed by the UN Ocean Decade and part of the CoastPredict program (coastpredict.org), FOCCUS develops innovative coastal products through 3 key pillars: i) developing new coastal observations, ii) developing advanced hydrology and coastal models, and iii) establishing coastal applications for enhanced coastal risk management. FOCCUS develops novel high-resolution data products by integrating multi-platform coastal observations (in-situ coastal data and satellite and land-based remote sensing) with simulated data to enhance weather and marine services, and monitor ocean health, ecosystems and coastal changes. New coastal products are created by data fusion, merging diverse data sources, implementing new approaches with numerical models and Artificial Intelligence (AI) methods for image processing and data analysis. FOCCUS builds on existing pan-European hydrological models and develops a pan-European ensemble to provide improved river discharge (sediment, nutrients, water volume and active tracers) for ocean and coastal ocean models and their two-way feedbacks at the land-sea interface. New methodologies are being tested, taking advantage of stochastic simulation, ensemble approaches, and AI. Working to improve coastal management, FOCCUS will facilitate advanced, seamless ocean monitoring and forecasting, from CMEMS global/regional systems to coastal systems, through demonstrations of new products and improved co-produced services. This includes focused coastal applications, co-designed with stakeholders, to address three areas of coastal protection: i) coastal management and protection (including the prediction of coastal erosion risk, marine pollution, and sediment tracking), ii) enhancement of the blue economy (including the co-use of wind and aquaculture resources), and iii) building resilience to coastal climate change (including tracking marine heatwaves, monitoring ecosystem degradation and harmful algae blooms, and predicting storm surge/waves). FOCCUS aims to provide the marine knowledge needed to support Marine Protected Areas, and address natural hazards and extreme events.
FOCCUS is funded by the European Union (Grant Agreement No. 101133911). Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Health and Digital Executive Agency (HaDEA). Neither the European Union nor the granting authority can be held responsible for them.
How to cite: Johnson, K., Staneva, J., Reyes, E., Bonaduce, A., Verri, G., Federico, I., Bartosova, A., Terskii, P., Christensen, K. H., Jamet, Q., Melet, A., Garcia Hermosa, I., Mészáros, L., and El Serafy, G.: Forecasting and Observing the Open-to-Coastal Ocean for Copernicus Users (FOCCUS): Advances in Coastal Monitoring and Forecasting to Enhance Europe’s Coastal Hazard Resilience, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-16397, https://doi.org/10.5194/egusphere-egu25-16397, 2025.
Coastal inundation presents a substantial risk to human lives and economic assets, driving the need for effective mitigation strategies. Recently, Nature-Based Solutions (NBS) have emerged as sustainable and adaptive alternatives to traditional "gray" infrastructure for coastal hazard management. Accurate modeling of coastal flooding and its interactions with NBS is essential for effective risk assessment, but challenges remain due to data limitations and modeling uncertainties.
Flood modeling techniques range from simplistic bathtub models to advanced hydromorphodynamic approaches. Simplified dynamic models, such as LISFLOOD-FP, which solve shallow water equations for floodplain processes, offer a practical balance between computational efficiency and accuracy.
This study enhanced LISFLOOD-FP model's ability to simulate coastal flooding by incorporating wave contributions (setup and swash), and their interactions with protective features like temporary dunes, along potential erosion and failures. These advancements were tested in Cesenatico, a coastal town in Emilia-Romagna, Italy, where seasonal dunes are constructed each winter as temporary defenses against flooding.
The enhanced model was validated using two storm events: the 2015 Saint Agatha Storm, which breached dunes and caused extensive flooding, and the 2022 Denise Storm, during which intact dunes mitigated flood impacts. The enhanced LISFLOOD-FP model significantly improved flood simulations, particularly for the 2022 event, accurately reproducing flooded areas in the presence of temporary dunes. These findings underscore the model's ability to capture the protective effects of NBS and highlight the importance of appropriately sizing such defenses.
The study also underscores the critical impact of data uncertainty on coastal flood modeling. Specifically, the lack of detailed topographic data on the location and dimensions of temporary dunes introduces significant uncertainty, with small variations in dune height—on the scale of centimeters—potentially determining whether dunes collapse or resist storm impacts. This uncertainty is compounded by the scarcity of observational flood maps, which limits rigorous model validation and reliability assessments.
This work represents a significant step toward developing a digital twin of coastal NBS, providing a robust framework for coastal management. Digital twins enable the exploration of "what-if" scenarios, optimization of defenses, evaluation of strategies, and generation of probabilistic flood forecasts, marking an important advancement in sustainable, science-driven coastal resilience planning.
How to cite: Lopes, I., Mentaschi, L., Federico, I., Vousdoukas, M., Perini, L., Pinardi, N., and Coppini, G.: Advancements in coastal flood modeling with LISFLOOD-FP: incorporating the dynamic of waves and dune failure, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-16448, https://doi.org/10.5194/egusphere-egu25-16448, 2025.
Coastal regions are increasingly vulnerable to extreme weather events and rising sea levels, posing significant risks to human lives, infrastructure, and ecosystems. Enhancing resilience in these areas demands innovative and cost-effective solutions for early warning systems. This study explores the integration of low-cost and do-it-yourself (LC+DIY) sensor devices and networks with high-resolution coastal prediction models to improve early warning capabilities for at-risk coastal communities. By leveraging advancements in IoT technology, the proposed sensor networks can monitor key environmental parameters, such as water levels, waves, and water temperature and salinity, in real time. Several examples of sensors, along with their applications across various continents, highlight the suitability of low-cost sensors, particularly in scenarios requiring extensive data collection and in geographically diverse contexts such as developing countries. These LC+DIY examples range from laboratory experiment comparisons to the development of monitoring system networks in Mozambique (eastern Africa). The system's affordability and scalability make it accessible to resource-constrained regions, addressing gaps in traditional (i.e. commercial) monitoring systems. This approach underscores the potential of integrating low-cost technologies with advanced modelling to safeguard coastal communities and ecosystems against climate-related hazards. Moreover, these initiatives also present significant opportunities, including fostering citizen science through collaborative approaches, such as integrating open-source platforms. The next steps include conducting further inter-comparisons with commercial devices, empowering local communities through an open science approach, and the ongoing development and refinement of LC+DIY prototypes to enhance their functionality and accessibility.
Acknowledgments
This work is funded by ECOBAYS (PID2020-115924RB-I00) from the Agencia Estatal de Investigación - Spain
How to cite: Grifoll, M.: Enhancing Resilience Through Low-Cost Sensors and High-Resolution Coastal Predictions for Early Warning Systems, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-3586, https://doi.org/10.5194/egusphere-egu25-3586, 2025.
Coastal hazards such as erosion and flooding are intensifying and becoming more frequent due to climate change, posing significant threats to many low-lying coastal areas. Historically, interventions have focused on grey infrastructure, including seawalls, breakwaters, and revetments, which present challenges such as high construction costs and negative environmental impacts. Consequently, there is a growing drive towards adopting Nature-based Solutions (NBS), such as the use of mangroves. Utilizing the Digital Twin of the Ocean (DTO), the effectiveness of such NBS can be simulated through advanced models. This study explores What-if Scenarios (WiS) using mangroves as NBS to mitigate coastal erosion in the Volta Delta region, an area particularly lacking comprehensive observational data. The integration of the DTO framework bridges this data gap by providing high-resolution simulations and predictive capabilities. The approach adopted is based on a robust model chain integrated within the DTO to simulate different configurations and densities of mangroves. 1D and 2D -Xbeach model is used to explore three categories of WiS: the beach without mangroves, mangroves positioned at the back of the shoreline, and mangroves placed within the intertidal zone. Model validation against measured coastal profiles shows good agreement with observed erosion trends, providing accurate predictions of sediment volume changes. From the results, a significant reduction in erosion is observed, with mangroves at varying densities offering varied protection levels between 18% and 100%. High densities of mangroves introduced in the intertidal zone resulted in the complete stabilization of the shoreline. These simulations highlight the potential of mangroves as a dynamic coastal defense strategy, with DTO applications providing a valuable tool for testing and optimizing NBS interventions. This study contributes to the ongoing development of mangroves as a NBS for coastal defense, demonstrating how DTO applications can effectively test and optimize interventions. By addressing the scarcity of observational data, the DTO framework enhances our understanding and predictive capacity for coastal dynamics.
How to cite: Jayson-Quashigah, P.-N., Staneva, J., Chen, W., and Djath, B.: Mangroves and Coastal Resilience: A Model-Based Evaluation, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-17068, https://doi.org/10.5194/egusphere-egu25-17068, 2025.
Effective storm surge prediction is vital for safeguarding coastal communities and enhancing disaster preparedness, particularly as climate change amplifies the frequency and intensity of extreme events. Despite the growing application of Machine Learning (ML) in storm surge downscaling, systematic comparisons with high-resolution dynamical models and focused assessments of extreme events remain underexplored. This study bridges these gaps by comparing advanced dynamical modeling with ML techniques to improve storm surge forecasting in the Northern Adriatic Sea.
High-resolution simulations were conducted using the SHYFEM-MPI model, leveraging optimized physical configurations and high-quality forcing datasets. This benchmark model demonstrated strong accuracy in representing storm surge dynamics and extremes, serving as a reference for evaluating ML-based approaches. To explore ML potential, models ranging from Multivariate Linear Regression (MLR) to the more advanced Long Short-Term Memory (LSTM) networks were developed and tested. A novel validation metric, the corrected mean absolute deviation (MADc) [1], and a tailored loss function (MADc2) were employed to improve model performance, particularly for extreme event prediction.
Results highlighted that while MLR offered computational efficiency, it struggled to capture non-linear dynamics and extremes. In contrast, LSTM networks excelled at modeling temporal dependencies and non-linearities, particularly when trained using the MADc2 loss function. Training ML models on outputs from the dynamical model revealed that MADc2-based architectures aligned closely with observations, offering a cost-effective alternative to traditional downscaling when high-quality forcing data is unavailable. Moreover, direct training on observed data at key sites such as Punta della Salute and Trieste showed that ML models, including LSTM, could outperform the dynamical model on critical metrics, underscoring the value of observational data.
This study underscores the promise of ML approaches in storm surge prediction, especially when integrated with high-quality data sources. By offering accurate predictions with significantly lower computational demands, ML techniques present a compelling case as efficient alternatives to traditional numerical models. As data accessibility and computational methods continue to advance, ML approaches may redefine the future of storm surge forecasting, enabling more sustainable and cost-effective solutions for coastal resilience.
[1] Campos-Caba, R., Alessandri, J., Camus, P., Mazzino, A., Ferrari, F., Federico, I., Vousdoukas, M., Tondello, M., and Mentaschi, L. (2024). Assessing storm surge model performance: what error indicators can measure the model’s skill? Ocean Science 20, 1513-1526. https://doi.org/10.5194/os-20-1513-2024.
How to cite: Mentaschi, L., Campos-Caba, R., Alessandri, J., Camus, P., Mazzino, A., Ferrari, F., Federico, I., Vousdoukas, M., Tondello, M., and Coppini, G.: Storm surge prediction in the Northern Adriatic Sea: a comparison between Machine Learning and numerical modelling, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-17094, https://doi.org/10.5194/egusphere-egu25-17094, 2025.
On 15 December 2024, a severe storm in the Kerch Strait led to catastrophic incidents involving two Russian oil tankers, Volgoneft-212 and Volgoneft-239. The Volgoneft-212 broke apart, spilling approximately 4,900 tonnes of mazut into the Black Sea, while Volgoneft-239, damaged and aground, leaked an additional 2,400 tonnes.
To assess and mitigate the environmental impact, simulations were initiated immediately using the Medslik-II oil spill model. These simulations utilized analysis and forecast data from the Copernicus Marine Service (Black Sea currents and sea surface temperature) and ECMWF-IFS winds provided by the Italian Air Force Meteorological Service. Four operational bulletins were generated during the first week, informed by evolving observations and reliable event reports. These reports were sent to authorities and environmental entities, explaining the constraints of the simulation and the expected forecast.
On 18 December at 03:00 UTC, COSMO-SkyMed satellite imagery, distributed and processed by e-GEOS (a Telespazio and Italian Space Agency Company) (based on COSMO-SkyMed satellites by Agenzia Spaziale Italiana and Ministero della Difesa) detected an oil slick near the Kerch Strait. Comparative analysis revealed strong agreement between Medslik-II simulations and satellite observations in both shape and trajectory, validating the model's accuracy during the initial response phase. Forecasts continued until 21 December, predicting a reversal of current patterns that would transport oil westward. This forecast aligned with subsequent reports of oil pollution as far as Sevastopol, approximately 250 kilometers from the spill's origin.
Post-event analyses incorporated satellite imagery and media reports to refine simulations and assess long-term impacts. These efforts highlight the importance of integrating operational modeling, remote sensing, and reliable field data for real-time decision-making and post-incident analysis. Lessons learned from the Kerch Strait accident can serve as pathways to enhance oil spill response strategies and mitigate environmental risks in future maritime emergencies.
How to cite: Atake, I., Coppini, G., Daffinà, F., Ramos, J., Bravo, S., Dissanayake, A., Scuro, M., Hoxhaj, M., Mannarini, G., and Liubartseva, S.: Operational oil spill monitoring and forecasting in the Kerch Strait accident in December 2024, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-17594, https://doi.org/10.5194/egusphere-egu25-17594, 2025.
Dealing with storm surges always involves working with forecasting systems, as they can provide crucial information on the evolution of sea level phenomena, fundamental for coastal flooding risk prevention.
This is one of the main reasons why ISPRA developed and continuously updates an integrated operational monitoring and forecasting system, specifically focused on the North Adriatic Sea and the Venice Lagoon. The system consists of three integrated components: the monitoring system, which is based on the National Sea Level Measurement Network (36 stations) and the North Adriatic and Venice Lagoon Sea Level Measurement (33 tide gauges); the forecasting component, based on the deterministic hydrodynamic finite element numerical model SHYFEM; the probabilistic module for evaluating uncertainty in sea level predictions. The numerical model provides 8 different forecasts each day (up to 144 hours), depending on the spatial resolution, the input meteorological data (ECMWF or BOLAM) and the assimilation of real time observed data. Furthermore, the statistical bayesian processor (Model Conditional Processor MCP), recently integrated in the operational chain, estimates the forecast uncertainty in terms of the probability of an event exceeding a fixed threshold; in the first version (v1) it has been directly applied to the total tidal height, while in a new recent development (v2) it has been provided with information regarding only the meteorological component that predominantly determines the uncertainty, and thus improving the performance. These two information (i.e. the deterministic value and the probability), when combined, make a significant difference, as they provide decision-makers with a deeper understanding of whether or not to take action during a storm surge. In other words, the decision triggering threshold will not be based only on different sea level thresholds (warning level, alert level, flooding level), but rather on different probabilities of a threshold to be overtopped. A very detailed analysis has been carried out to better understand the performance of different model configurations, particularly during the storm surges that occurred between January 2022 and April 2024 (approximately 50 events), and the preliminary results are presented in this study.
Finally, the ex-post analysis of one of the most impactful and recent events (November 22nd, 2022, with sea levels reaching up to 200 cm in the North Adriatic Sea) is presented here to highlight the need for a widespread distribution of measurement stations and very high-resolution forecasts. These are essential to allow a detailed analysis of the effects at both large and small scales (e.g., the lagoons of Venice, Marano-Grado, and Sacca degli Scardovari – Po river delta), even in closely located areas. Once again, the integration and improvement of in situ observations and the modeling system provide a virtuous example of efficiency and functionality in the prevention and mitigation of the impacts of floods and extreme weather-marine events on the coastal environment.
How to cite: Morucci, S., Coraci, E., Bonometto, A., Mel, R. A., Crosato, F., and Gyssels, P.: Facing storm surges in Venice: operational system and uncertainty , EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-19028, https://doi.org/10.5194/egusphere-egu25-19028, 2025.
Marine and coastal ecosystems (MCEs) are vital to human well-being, playing a significant role in climate regulation, carbon sequestration, while protecting coastal areas from sea level rise and erosion. However, these ecosystems are increasingly threatened by the combined effects of anthropogenic stressors (e.g., pollution) and climate change-related pressures (e.g., rising sea temperatures and ocean acidification). Cumulative impacts arising from this complex interplay threaten MCEs' ability to deliver critical ecosystem services, compromising their health and resilience.
Machine Learning (ML) has emerged as a valuable tool for assessing ecological conditions under multiple pressures. Algorithms like Random Forest (RF) and Support Vector Machine (SVM) have demonstrated their effectiveness in identifying patterns and predicting changes in ecosystem health. However, these models often fail to account for spatial dependencies between data points, which are crucial for understanding the interconnected nature of marine environments. Graph Neural Networks (GNNs), a more recent advancement in ML, overcome this limitation by explicitly modelling spatial relationships, making them highly suitable for analysing complex MCE dynamics.
This study explores the application of GNN-based models to assess the impact of multiple pressures on seagrass ecosystems in the Italian coastal areas. To this aim, a comprehensive dataset was constructed, including key variables influencing seagrass health, such as nutrient concentrations, temperature, and salinity, derived from open-source platforms (e.g., Copernicus CMEMS, EMODnet). Data were synthesized into a 4km raster grid, with each pixel representing seagrass presence or absence. GNNs were constructed by considering each pixel as a node and connecting it to neighbouring pixels to capture spatial relationships. Experiments evaluated different GNN architectures, such as Graph Convolutional Networks (GCNs) and Graph Attention Networks (GATs), alongside traditional ML models like RF, SVM, and Multi-Layer Perceptron.
The results showed that GNNs outperformed traditional models in terms of F1-score and accuracy, particularly in spatially complex scenarios. Traditional models often misclassified regions with intricate spatial dependencies, such as boundaries between seagrass patches, whereas GNNs demonstrated superior capability in leveraging spatial context. Despite these advantages, the study faced challenges due to the limited availability of high-resolution, temporal datasets, constraining the full exploration of dynamic ecosystem processes. However, by addressing the challenge of spatial resolution in ecological data, GNNs represents a transformative approach to understanding ocean dynamics. Their integration into a Digital Twin of the Ocean has the potential to transform ecosystem management and significantly advance coastal resilience efforts. This framework would enable detailed simulations and predictions of processes like ocean currents, extreme weather events, and the cumulative impacts of climate change and human activities. Moreover, the combination of GNNs and Digital Twins would provide deeper insights into the complex interplay of factors shaping marine and coastal ecosystems ecological state and processes and their resilience overall. This synergy empowers scientists and policymakers with actionable intelligence, fostering effective decision-making and the development of strategies to mitigate ocean hazards, while safeguarding biodiversity and enhancing the resilience of coastal communities. As future efforts move towards incorporating high-resolution data, this integrated approach holds promise for advancing the sustainable management of MCEs globally.
How to cite: Bianconi, A., Vascon, S., Furlan, E., and Critto, A.: Toward the integration of Graph Neural Networks and Digital Twins: Transforming marine ecosystem management and coastal resilience, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-19556, https://doi.org/10.5194/egusphere-egu25-19556, 2025.
In the context of climate change, future coastal flooding will become increasingly impactful, driven primarily by sea level rise and the intensification of extreme precipitation events along coastlines, rather than changes in wind regimes, wave dynamics, or hydrodynamic circulation (IPCC, 2021; Vousdoukas et al., 2018). Due to the vast geomorphological diversity of coastlines and the varying exposure of assets at risk, advanced tools are necessary for precise risk assessments, particularly for urban areas and human settlements.
Coastal flooding dynamics cannot be adequately described solely by barotropic phenomena such as wave setup but require explicit modeling of wave impacts through phase-resolving models. In this work, we propose a compound flooding assessment integrating, at an urban scale and very high resolution (1-5 meters), the effects of storm surge, coastal wave-current interactions, and explicit wave runup simulations using phase-resolving coastal models.
Urban-scale flood risk assessment not only allows the modeling of wave interactions with natural and man-made structures but also incorporates urban elements such as groynes, breakwaters, roads, and buildings. This modeling approach supports the design of ecosystem-based solutions to enhance coastal resilience (Cheong et al., 2013). Additionally, we discuss the integration of hydraulic flooding simulations caused by rainfall and river overflow into compound flooding models, demonstrating how coastal hydraulic risk can be better described—beyond traditional extreme value statistics—using impact-based metrics for compound events (Bevacqua et al., 2019).
We conduct a combined assessment of sea level and wave effects on flooding using a selection of extreme events with varying wave and sea-level conditions. This is achieved through a multimodel approach comparing simulation outputs from FUNWAVE and XBEACH models (Shi et al., 2012; Roelvink et al., 2009). These outputs are used to validate results obtained from a simplified modeling approach developed within the SCORE project, which employs traditional hydraulic models for unsteady flow simulations coupled with phase-averaged wave models for the marine component.
These tools have been applied to the study site of Marina di Massa (Italy) to assess coastal flood risk trends for future scenarios and to design early-warning systems. In Marina di Massa, this approach is supported and validated by an observational system that includes coastal webcams, wave radars, and meteomarine observation systems (buoys and ADCPs).
How to cite: Sacco, M., Mocali, R., Bendoni, M., Taddei, S., Cucco, A., Caparrini, F., Perna, M., Vitale, G., Ortolani, A., and Brandini, C.: Compound Coastal Flood and impact-based Risk Assessment in the Context of Climate Change Using a Multimodel Approach to Enhance Coastal Resilience, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-19940, https://doi.org/10.5194/egusphere-egu25-19940, 2025.
In today’s world, the accessibility of operational large-scale regional ocean models from platforms like the Copernicus Marine Environment Monitoring Service (CMEMS), combined with advanced computing infrastructures such as cloud computing and high-performance computing (HPC), is enabling the creation of high-resolution, on-demand digital representations of the ocean. These advancements are driving international interest in implementing high-resolution shelf-coastal numerical models to deepen our understanding of marine systems and their sensitivities to climate change. Such models are essential for capturing fine-scale processes that coarse-resolution global and regional models cannot resolve.
The Structured and Unstructured grid Relocatable Ocean platform for Forecasting (SURF) is an innovative, open-source ocean modeling platform designed to set up, execute, and analyze high-resolution nested ocean models in any region within a large-scale Ocean Forecasting, Analysis, and Reanalysis System. SURF integrates two state-of-the-art ocean models: NEMO: A structured-grid model optimized for open ocean and shelf applications. SHYFEM-MPI: An unstructured-grid model tailored for accurately modeling complex coastal dynamics.
SURF has been successfully implemented and validated in various regions of the world’s oceans, downscaling from large-scale ocean prediction systems, such as global and regional CMEMS products. The nested high-resolution models have shown better performance compared to their parent coarse-resolution models.
SURF provides a high-level, user-friendly interface to conduct an ocean downscaling experiment from start to finish, including input data acquisition and pre-processing, model execution, and post-processing for visualization and analysis of results. The platform is distributed as a Virtual Machine and Container Images, using portable virtualization technology for easy deployment across various computational environments, ensuring accessibility for a wide range of users, including educational institutions and commercial enterprises.
SURF is a valuable tool to supports Decision Support System (DSS) by providing high-resolution ocean forecasts crucial for applications like oil spill monitoring, search and rescue operations, navigation routing, fisheries and tourism. A recent application was its deployment during the Manila Oil Spill accident on July 24, 2024, where high-resolution ocean circulation fields generated by SURF were integrated with the WITOIL oil spill simulation platform. This integration improved trajectory predictions, accurately depicting the northward drift of the oil slick and closely aligning with satellite observations.
On-demand regional and coastal high-resolution models can be beneficial to diverse end-users, including coastalmanagers, harbour authorities, civil protection agencies and maritime communities. By providing high-resolution ocean forecasts, SURF can play a crucial role in mitigating risks, protecting communities, and reducing potential losses.
How to cite: Trotta, F., Giunti, L., Federico, I., Causio, S., Scuro, M., Vicente Cruz, R., Pinardi, N., and Coppini, G.: SURF: A Relocatable Platform for On-Demand High-Resolution Ocean Modelling for the Digital Twins , EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-19949, https://doi.org/10.5194/egusphere-egu25-19949, 2025.
Coastal variables monitoring and study are crucial activities for researchers to develop and enhance the knowledge about how climate changes effects are modifying the maritime hydrodynamics.
This study focuses on local effects on the hydrodynamic circulation and significant wave height in the Gulf of Trieste (GoT) through coastal high frequency radars (HFRs) measurements and wave buoys. In a semi-enclosed basin, characterized by an average depth equal to 18m and maximum depth of 25m, the variables investigated are surface currents direction (θcurr) and eastward and northward velocity component (ucurr,vcurr), mean wave direction (θw) and spectral significant wave height (Hm0).
The HFRs network installed in GoT are composed by four WERA (WEllen RAdar) systems installed in the east and south part of the gulf and operating at a frequency of 24.5 MHz. Two wave buoys are installed in the central and north parts of the GoT at different depths, recording the wave energy spectral distribution characterizing the area. Finally numerical method to estimate power spectral density are implemented in this study with high resolution allowing to increase the knowledge of spatial and temporal hydrodynamic evolution inside the GoT.
The HFR measurements can return a spatial information about waves and currents variable, improved by the calibration process. This information is crucial to reproduce the coastal hydrodynamic. Coupling all data is possible to underline the wave spectral evolution inside the gulf, the hydrodynamic peculiarity and have a focus on the extreme events effect on semi-enclosed basins.
Joining the HFR and wave buoys data and implementing the numerical model, this research shows innovative methods to deepening knowledge and monitoring the extreme events that characterise the study area.
How to cite: Ingrassia, E., Ursella, L., Lo Re, C., Capodici, F., and Ciraolo, G.: Coastal monitoring in Gulf of Trieste through the integration of oceanographic instruments data and numerical models, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-19978, https://doi.org/10.5194/egusphere-egu25-19978, 2025.
Research interest on tsunami propagation and coastal impact in the Mediterranean Sea has recently grown due to recent catastrophic events, like the ones in Indian and Pacific Oceans.
A novel operational approach for evaluating tsunami-induced inundation, based on a generalization of Green's law and a chain of intermediate and small-scale simulations, has been recently proposed (Lalli et al. 2019; Melito et al., 2022).
The original Green’s law provides the amplitude of a long wave at a given water depth, accounting only for the wave shoaling, while neglecting both diffraction and refraction effects. The new formulation of Green’s law allows one to overcome such limitations, by introducing a semi-analytically computed coefficient α, encompassing the effects due to refraction and diffraction phenomena, and other interactions with natural obstacles and artificial structures, which play a crucial role in the case of natural, complex bathymetry. Therefore, α represents a proxy for coastal susceptibility to tsunami impact. Melito et al. (2022) performed intermediate-scale simulations to identify the distribution of α along the south-eastern Italian coasts. To validate the employed procedure, small-scale modeling of coastal flooding has been also performed by Melito et al. (2022), for two case studies: the Esaro river estuary (Calabria) and the nearshore of Bari (Apulia). A good agreement was found between the inundation levels obtained by small and intermediate-scale modeling, the latter allowing for a significant reduction of the computational costs. The present work aims at extending the research by applying the described procedure to all the Italian Tyrrhenian coasts.
References
Lalli, F., Postacchini, M. & Brocchini, M. (2019) Long waves approaching the coast: Green’s law generalization. Journal of Ocean Engineering and Marine Energy, 5, 385–402. https://doi.org/10.1007/s40722-019-00152-9.
Melito, L., Lalli, F., Postacchini, M., & Brocchini, M. (2022). A semi-empirical approach for tsunami inundation: An application to the coasts of South Italy. Geophysical Research Letters, 49, e2022GL098422. https://doi.org/10.1029/2022GL098422.
How to cite: Lombardi, V., Lalli, F., Melito, L., Cassese, M. L., Bruschi, A., and Brocchini, M.: A Semi-Empirical Approach for Tsunami Inundation:Application to the Italian Tyrrhenian coasts, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-20006, https://doi.org/10.5194/egusphere-egu25-20006, 2025.
The interconnected nature of catchment hydrology and marine circulation processes poses significant challenges to the numerical modelling of river and coastal flood/drought conditions in the catchment-sea continuum.
Finite Element Modelling (FEM) provides an advanced solution, offering the ability to handle cross-scale and multi-scale processes with adaptive unstructured meshes, which are crucial for accurately representing complex coastlines and varying bathymetry. This makes FEM well-suited for seamless modelling of inland and marine water systems.
Compound flooding and drought events are becoming increasingly frequent and intense across many catchment areas draining into the Mediterranean basin. To address these challenges, we used a seamless numerical modelling of the river-sea continuum based on a Finite Element code (SHYFEM-MPI-ZSTAR, Micaletto et al 2022, Verri et al 2023). By progressively refining the SHYFEM-MPI-ZSTAR experimental settings, we aim to deepen our understanding of compound flooding and drought events occurring in the Po River delta system, which is Italy's longest river and the second-largest freshwater source for the Mediterranean basin.
A four-year experiment (2019 to 2023) was conducted to simulate significant events, including the November 2019 flood and the July 2022 drought. Model findings were validated against available in-situ and satellite observations.
We explored the role of non-linear combination of multi-scale and cross-scale forcing mechanisms to enhance the modeling accuracy and the understanding of the complex physical processes underlying such extreme events.
How to cite: Worou, O. L., Verri, G., Viola, F., and Pinardi, N.: Compound river and coastal flooding/drought events in the Po delta area, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-6224, https://doi.org/10.5194/egusphere-egu25-6224, 2025.
CoastPredit general aim is : co-design and implement an integrated coastal ocean observing and predicting system adhering to best practices and standards, designed as a global framework and implemented locally.
Current solutions and services for global coastal areas often do not include the establishment of monitoring and prediction systems to evaluate impacts ranging from specific events to long-term climate trends. Components of such systems do exist but do not necessarily provide the accuracy needed. For instance, storm surge prediction solutions have predominantly relied on depth-integrated modeling, which limits the direct consideration of climate change-induced sea level trends resulting from ocean warming and freshening, processes that are different on the surface and at depth.
The GlobalCoast CoastPredict initiative aims at providing integrated coastal ocean monitoring and prediction systems as part of a comprehensive, science-based approach for global and local solutions under international standards. Additionally, solutions and services have not been systematically compared across coastal regions. For example, the prediction and monitoring systems for the loss of coastal coral reefs vary across the world’s oceans without clear justification.
The CoastPredict GlobalCoast initiative aims to address this gap by identifying global coastal areas where similar solutions can be implemented, tested, and refined, or alternatively, where distinct solutions are necessary.
To achieve its general aim and its implementation principles, CoastPredict has established the GlobalCoast Network, defining pilot sites to implement CoastPredict solutions and enhance coastal resilience services . The GlobalCoast survey will be reviewed where the resilience challenges specific to each Pilot site were identified.
How to cite: Coppini, G., Kourafalou, V., Tintore, J., Helsop, E., O'Donovan, M., and Pinardi, N.: The GlobalCoast Initiative of CoastPredict: from operational oceanography to management solutions, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-8124, https://doi.org/10.5194/egusphere-egu25-8124, 2025.
Coastal flooding triggered by tropical cyclones is a frequent and devastating threat in low-lying coastal areas. The risk of inundation escalates when storm tides interact with river systems and are compounded by intense rainfall during the cyclone. This vulnerability is further heightened when cyclones land near estuaries, river deltas, or adjacent rivers along the coast. Consequently, understanding these interactions and accurately quantifying their contributions to coastal inundation is crucial for effective inland flood mapping and disaster management. ADCIRC model is one of the practical tools in computing coastal inundation, but it needs to consider precipitation, which plays a major role during flooding. The HEC-RAS model is coupled with ADCIRC to solve this issue to provide realistic coastal flooding. Validation of inundation is a very tough task during storm surge events due to the unavailability of an inundation map at the time of landfall. In our experiment, we used coupled ADCIRC and HEC RAS over the significant river estuaries (Hooghly, Mahanadi, Krishna, and Godavari) along the east coast of India since these regions are very vulnerable to storm surges. Significant cyclone landfalling over or near these river systems is selected for computing inundation. To calculate the inundation, storm tides from the ADCIRC model are used as input to the HEC-RAS model. Other parameters like river discharge and gridded precipitation are also incorporated.
Further model capability is enhanced by adding land cover, soil, and infiltration data over these river systems. Fani cyclone is one of the devastating cyclones that significantly impacted the Mahanadi basin. Inundation from the model is validated with the satellite map, which was available two days after the landfall. Model inundation is adjusted by altering the precipitation factor depending on the observed value. The same factor is used for other river basins. Results show that the model is validated reasonably well with the observation and is best suitable for assessing compound flooding along the river basin over east coast of India.
How to cite: tiwari, P., Rao, A. D., Pandey, S., and Pant, V.: Compound Flood Modeling: Coupling ADCIRC and HEC-RAS for Enhanced Risk Assessment along East Coast of India, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-425, https://doi.org/10.5194/egusphere-egu25-425, 2025.
The Korean peninsula's intricate coastline, distinguished by its numerous islands and extreme tidal variations, is particularly susceptible to storm surges-especially when typhoons coincide with high tides, posing significant threats to both life and property. Accurate prediction of storm surges is crucial for mitigating these potential disasters. Current operational forecasting systems, implemented by various agencies, typically rely on single atmospheric model inputs as their forcing mechanism. However, such deterministic approaches often exhibit considerable variability in their predictions due to inherent uncertainties in atmospheric modeling processes.
This study focuses on developing an Ensemble Regional Tidal and Storm Surge Model (ETSM), integrating forecast outputs from 26 atmospheric ensemble members to enhance storm surge predictions along the coasts of Korea. The model was evaluated through case studies of major typhoons, including HINNAMNOR (2022) and KHANUN (2023), by comparing observed water levels against both deterministic and ensemble model predictions. The deterministic model tended to both overestimate and underestimate, whereas the ensemble spread encompassed the observed water levels, demonstrating that the ensemble model provided better predictions in representing actual storm surge events.
Additionally, the study analyzed the characteristics of storm surge heights observed during the summer of 2024, establishing threshold values. The analysis revealed significant regional variations in storm surge height distributions across the West, South, and East Coasts of Korea. To assess the model's predictive performance, probabilistic validation using Brier Scores and ROC (Receiver Operating Characteristic) metrics was performed. The results indicate reliable predictive performance while also revealing for further improvement.
How to cite: Kim, J., Park, J.-H., Oh, S. M., and Cho, I. H.: Development of an Ensemble Regional Tide and Storm Surge Model (ETSM) for the Coasts of Korea, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-7765, https://doi.org/10.5194/egusphere-egu25-7765, 2025.
Due to the high densities of human populations, diverse human activities, and a variety of anthropogenic pressures, coastal zones rank among the most heavily impacted areas of the global ocean. With climate change impacts escalating and the European Union's prioritization of a sustainable blue economy, comprehensive and high-resolution coastal observations have become critical to enhance community resilience. The FOCCUS EU-funded project (Forecasting and Observing the Open-to-Coastal Ocean for Copernicus Users) directly address this challenge by enhancing existing ocean monitoring and forecasting capabilities in coastal regions as part of the Copernicus Marine Environment Monitoring Service (CMEMS), and developing innovative coastal data products.
FOCCUS will develop novel observation-based data products through advanced multi-platform in-situ and remote sensing observations, merging multi-platform observing data, leveraging satellite multi-sensors, exploiting satellite mission synergies (e.g., SWOT), and utilizing data fusion techniques with artificial intelligence for image processing and data analysis. These efforts will yield improved estimates of essential climate variables, coastal ocean circulation, biogeochemical anomalies, harmful algal blooms (HABs), seagrass and macroalgae coverage, high-resolution topo-bathymetries, beach profiles, waves, and shoreline positions.
The new high-resolution data products generated by FOCCUS will significantly enhance weather and marine services, as well as monitor the health of the ocean, ecosystems, and coastal changes. Insights from these high-resolution observations will support hydrological and coastal model assessments, satellite calibration and validation, and the development of targeted coastal applications such as HABs and marine heat waves monitoring. These initiatives are driven by EU policies like the European Green Deal and Horizon Europe Mission, which mandate Member States to actively monitor, assess, manage, and protect their coastal waters.
By aligning with the UN Ocean Decade CoastPredict Program, FOCCUS addresses the urgent need for improved coastal management strategies. Integrating these data into CMEMS contributes to a sustainable blue economy, empowering informed decision-making for enhanced coastal resilience in the face of intensifying environmental challenges.
FOCCUS is funded by the European Union (Grant Agreement No. 101133911). Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Health and Digital Executive Agency (HaDEA). Neither the European Union nor the granting authority can be held responsible for them.
How to cite: Bonaduce, A., Reyes, E., Staneva, J., and Johnson, K. and the FOCCUS Project: New insight of high-resolution coastal observations: FOCCUS project: New Insight into High-resolution Coastal Observations for Enhancing Models and Applications, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-15636, https://doi.org/10.5194/egusphere-egu25-15636, 2025.
The implementation of Early Warning Systems (EWS) is a highly effective approach to mitigating the impacts of dredging activities. These systems combine predictive numerical models and real-time data to forecast sediment dispersion and assess potential impacts on marine species and habitats protected under the EU Habitat Directive. As part of the Renovate research project, an EWS was developed for the coastal area of Civitavecchia (northeastern Tyrrhenian Sea, Italy) to address the effects of dredging activities related to new port infrastructure on Posidonia oceanicameadows and coralligenous habitats located within Sites of Community Importance (SCIs) near the port. During the initial phase of development, the EWS incorporated a hydrodynamic model and a wave model to analyze the coastal hydrodynamic environment. Both models employ a finite-difference curvilinear grid, enabling high spatial resolution near the shoreline while maintaining lower resolution offshore. This configuration facilitates effective downscaling from Copernicus Marine System models, which operate at approximately 4 km spatial resolution. A distinctive feature of the EWS is its integration with two monitoring fixed stations located north and south of the port. These stations are designed to measure turbidity levels in areas affected by dredging operations. Each station is equipped with custom-assembled sensors capable of continuously monitoring physical, chemical, and bio-optical water parameters, supported by a real-time data acquisition and transmission system. The EWS is activated when turbidity levels detected by the monitoring stations exceed thresholds established by national or international regulations. Its predictive outputs enable the identification and planning of mitigation measures to address dredging impacts. To assess the direct effects of increased turbidity on protected habitats and species, the system incorporates risk indicators based on species-specific stressor tolerance curves and thresholds derived from targeted laboratory experiments. This study demonstrates how the Civitavecchia EWS can significantly enhance risk assessment and response strategies, providing a valuable tool for local stakeholders (such as the Port Authority of the Central-Northern Tyrrhenian Sea) and contributing to the strengthening of coastal resilience. This EWS integrates multidisciplinary data, models, and knowledge into a standardized governance framework, employing an ecosystem impact forecasting system inspired by the concept of a coastal digital twin.
How to cite: Piermattei, V., Bonamano, S., Madonia, N., Madonia, A., Fersini, G., Federico, I., and Marcelli, M.: Development of an Early Warning System to Mitigate Dredging Impacts on Coastal Ecosystems, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-16679, https://doi.org/10.5194/egusphere-egu25-16679, 2025.
Coastal regions face severe impacts from their proximity to the sea, such as sea level rise and climate variability. Currently, over 2 billion people worldwide live in near-coastal areas [L. Reimann et al., 2023], where industrial activities coexist with urban development, port traffic, and tourism.
The Venice Lagoon exemplifies such an environment, with its high population density and extensive industrial and tourism development, along with unique marine, atmospheric, and geomorphological features. This area is particularly vulnerable to high tides, which become even more hazardous when combined with extreme climate events. To mitigate the damage caused by storm surges in Venice, the MoSE (Modulo Sperimentale Elettromeccanico) regulated barriers were developed. This system consists of mobile barriers located at the three main inlets connecting the lagoon to the open sea.
In November 2022, a combination of meteo-marine phenomena—including astronomical tides, seiches, and strong southeasterly winds—resulted in one of the strongest marine surge events on record [R.A. Mel et al., 2023], and the barriers were activated for this event.
Here, we present storm surge forecasts through a synergistic modeling chain. This process begins with the regional-scale accuracy of MedFs (capable of predicting the surge peak outside the lagoon up to three days in advance) and extends to the urban scale with a downscaling model based on SHYFEM-MPI. The downscaled model was enhanced by developing immersed boundary (IB) conditions—a common Computational Fluid Dynamics (CFD) technique for simulating solid bodies—enabling the inclusion of the MoSE barriers in the modeling chain, allowing real-time simulations that can activate or deactivate these barriers as needed.
The modeling chain is put to the test with the November 2022 event in the Venice Lagoon, where the barriers were activated and deactivated four times. The outputs of the local urban model showed strong agreement with tide gauge networks, both those located outside the lagoon and, more importantly, those inside it, validating the effectiveness of the IB method used. Sea level maps and time series for simulations with and without the barriers demonstrate a reduction in total water levels by up to one meter inside the lagoon. Furthermore, the methodology proposed represents an important forecasting tool capable to perform what-if scenarios regarding the number and timing of MoSE barrier openings/closings, as well as partial closures that reproduce potential malfunctions or failures in the mechanical apparatus of MoSE.
This modeling chain could serve as a critical early warning system for decision-makers, providing essential information on local dynamics inside the lagoon during extreme events.
How to cite: Boetti, M., Federico, I., Causio, S., Goglio, A. C., Clementi, E., and Coppini, G.: Storm surge forecasting in Venice: what-if scenario with regulated barriers, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-18863, https://doi.org/10.5194/egusphere-egu25-18863, 2025.
This work seeks to enhance the physical representation of coastal-ocean dynamics through integrated numerical models, advancing the understanding of intricate Earth system processes. It specifically focuses on two critical aspects within coastal zones: the influence of vegetation on wave dynamics and the morphodynamic processes driven by sediment transport.
Modeling the intricate interplay between waves, seagrass, currents, and sediment processes is crucial for developing a comprehensive and realistic digital twin of the ocean. The absence of robust in-situ observational systems can result in insufficient representation of these highly dynamic environments. We aim to integrate numerical simulations with an observational system design, emphasizing the critical importance of continuous data collection and the cohesive application of empirical measurements within numerical models.
The augmented wave model, featuring a refined seagrass representation that incorporates flexibility, seasonal growth patterns, and phenotypic traits informed by site-specific measurements, is applied to the case study in the coastal zone of Civitavecchia in the north-eastern Tyrrhenian Sea, Italy. This study examines the restoration of Posidonia oceanica meadows, and their impact on wave attenuation, utilizing insights derived from the numerical model results. The sediment transport is tested in both an idealized tidal inlet scenario and along the coast of Fiumicino, south of Civitavecchia, with the aim of integrating a three-dimensional model capable of accurately capturing bedload transport influenced by local bathymetry and the advection of suspended sediments from the Tiber River mouth. The respective contributions of these factors to seabed evolution are quantified, and a feedback mechanism is further considered within the circulation and wave models.
Ultimately, this synergy aims to improve predictive capabilities in dynamic marine environments, advancing the numerical modeling of coastal-ocean processes to better forecast environmental extremes and enhance our understanding of the underlying physics.
How to cite: Shirinov, S., Federico, I., Bonamano, S., Causio, S., Biocca, N., Piermattei, V., Piazzolla, D., Alessandri, J., Mentaschi, L., Coppini, G., Marcelli, M., and Pinardi, N.: Modelling coastal-ocean morphodynamics and wave-vegetation interactions, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-18990, https://doi.org/10.5194/egusphere-egu25-18990, 2025.
