The core of the challenge lies in a trade-off that has long frustrated coastal forecasters. Process-based models such as XBeach, an open-source model developed to simulate the hydrodynamic and morphodynamic response of beaches to storms, can capture the intricate physics of wave runup, overtopping and flooding with impressive fidelity. But a single high-resolution simulation of a storm striking a beach can take hours of computation on a cluster, far too slow for an operational early warning system that must issue alerts while the storm is still offshore. Simpler empirical formulas run instantly but gloss over the site-specific geometry of a beach, the very factor that often determines whether a seawall is overtopped or left untouched.

The researchers’ solution is to let the two approaches do what each does best. They ran XBeach thousands of times, generating a vast library of simulated storm scenarios spanning the range of offshore oceanic conditions that could plausibly strike the study area. Each simulation produced a detailed picture of the resulting hazards and impacts, from wave overtopping discharges to damage levels. These thousands of physically consistent scenario-impact pairs then became the training data for a Bayesian Network, a probabilistic graphical model that learns the statistical relationships between offshore forcing conditions and onshore impacts. Once trained, the Bayesian Network acts as a computationally cheap surrogate of the numerically intensive model, producing probabilistic estimates of storm-induced impacts in a fraction of the time a direct XBeach run would demand.

Bayesian Networks are particularly well suited to this role because they do not merely spit out a single deterministic answer. Instead, they quantify uncertainty explicitly, returning probability distributions over possible outcomes given the observed offshore conditions. For an early warning system, this matters enormously: emergency managers need to know not just the most likely impact but the chance that impacts will cross critical thresholds for pedestrians, vehicles and property. The approach also remains interpretable, a quality emphasized in the machine learning literature on Bayesian networks, which allows coastal scientists to trace which combinations of wave height, water level and beach state drive the most severe predicted responses.

The framework was tested at two contrasting beach profiles along an embayed beach in Santa Catarina, southern Brazil: one strongly urbanized and one semi-natural. This deliberate pairing allowed the team to assess explicitly how storm response varies over short distances along a single embayment, a spatial variability that coarser regional forecasts routinely miss. Embayed beaches, hemmed in by headlands, are notorious for focusing wave energy unevenly, so the same offshore storm can produce dramatically different overtopping at opposite ends of the bay. The extensive training dataset generated for the study provides new insight into this variability and helps identify the critical storm conditions that trigger extreme responses at each profile.

Perhaps the most innovative aspect of the work is how the team confronted the data scarcity that plagues coastal regions worldwide. Accurate XBeach simulations of wave overtopping require quantitative information on the underwater bathymetry, the shape of the seafloor, and on overtopping discharges for calibration, neither of which was readily available at the Brazilian study site. Rather than abandon the effort, the researchers reconstructed beach profiles using an empirical equilibrium profile formulation, drawing on the classic Dean equilibrium profile concept that relates the shape of a sandy beach to its sediment characteristics and wave climate. This allowed physically plausible cross-shore geometry to be generated where survey data were absent.

The second data gap, the lack of measured overtopping discharges, was bridged in an equally pragmatic way: the team used non-professional imagery, photographs and videos typically taken by members of the public during storm events, to estimate observed damage levels and infer the overtopping discharges associated with them. This strategy taps into a growing body of research showing that social media and crowdsourced imagery can meaningfully improve coastal flood forecasts and hazard alerts, turning an everyday byproduct of the smartphone era into a calibration resource for communities that cannot afford dedicated field campaigns.

The significance of the work extends well beyond one beach in Brazil. The United Nations has championed global early warning initiatives as a cornerstone of climate adaptation, recognizing that every dollar spent on early warning saves many more in avoided losses. Yet the regions that stand to benefit most, low-income and rapidly urbanizing coastlines, are precisely those with the least monitoring infrastructure. By demonstrating that a transferable, computationally efficient methodology can explicitly address data limitations, the study offers a template that coastal managers in data-limited settings can adapt without waiting for expensive bathymetric surveys or instrumented seawalls. The authors argue that coupling physically based modeling with data-driven probabilistic inference is what makes the framework both robust and practical for operational deployment.

Technically, the hybrid design also sheds light on the broader debate over surrogate modeling in coastal hazard prediction. Previous studies have benchmarked numerical models for wave overtopping, weighing accuracy against speed, and have explored neural networks and other machine learning tools for overtopping prediction at coastal structures. The Bayesian Network approach distinguishes itself by combining probabilistic output with modest training requirements and by being trained not on sparse field observations but on thousands of physically consistent numerical simulations. This means the surrogate inherits the physics of XBeach while shedding its computational burden, and it can be retrained as new bathymetry, imagery or storm observations become available, progressively sharpening its predictions over time.

For the residents and businesses lining vulnerable shores from Santa Catarina to the Basque coast, the practical promise is straightforward: earlier, more localized and more honest warnings that state not only what a storm is likely to do but how confident the forecast is. As sea levels rise and extreme meteo-oceanographic events intensify, the exposure of low coastal areas to storm surges and energetic waves will only grow. Frameworks like the one developed by Garzon, Husemann and Bonetti suggest that the barrier to protecting these communities is no longer the physics of flooding, which models already capture well, but the clever engineering of prediction systems that can run fast, learn from whatever data exist, and quantify what remains unknown. In that sense, the study marks a meaningful step toward making sophisticated coastal flood early warning systems a realistic option for the places that need them most.

Subject of Research: Hybrid numerical and Bayesian network modeling for rapid coastal storm flooding prediction and early warning in data-scarce coastal environments

Article Title: A hybrid approach for rapid coastal flooding impact prediction in data-scarce environments: an early warning system perspective

Article References: Garzon, J. L., Husemann, P., & Bonetti, J. (2026). A hybrid approach for rapid coastal flooding impact prediction in data-scarce environments: an early warning system perspective. Natural Hazards, 122(20), Article 650. https://doi.org/10.1007/s11069-026-08366-5

Image Credits: AI Generated

DOI: 10.1007/s11069-026-08366-5

Keywords: coastal flooding, early warning systems, XBeach, Bayesian networks, wave overtopping, storm surge, Santa Catarina, data-scarce environments, bathymetry reconstruction, probabilistic modeling, coastal hazard prediction, machine learning

The research, led by L. Carpi of the University of Genova together with N. Oneto and M. Ferrari, tackles a deceptively simple question: which parts of a coastal town will waves actually reach during the worst storms? Answering it required reconstructing four decades of wave history. The team drew on a 43-year hindcast database maintained by the Department of Civil, Chemical and Environmental Engineering at Genoa, which contains numerical simulations of offshore wave conditions produced with the WaveWatchIII spectral wave model and the Weather Research and Forecasting atmospheric model. From the period between January 1979 and December 2022, the researchers selected the twenty largest storms arriving from the southeast and the twenty largest from the southwest, the two dominant and most destructive wave directions in the Ligurian Sea.

Simulating what those forty storms did when they reached the coast demanded an equally detailed picture of the seafloor and the land. Bathymetric data out to depths of twenty metres were collected with a multibeam echosounder offering centimetre-level resolution, while the beach and the seafront urban fabric were mapped with a LiDAR system mounted on an unmanned aerial vehicle. Offshore bathymetry came from regional government datasets. All of these sources were merged and interpolated into a high-resolution digital elevation model that captured not only the natural morphology but also the buildings, seawalls and narrow alleys that shape how floodwater moves through the town.

The heart of the modelling chain is XBeach, an open-source model that solves the time-dependent short wave action balance, roller energy equations and nonlinear shallow water equations, allowing it to reproduce wave groups, run-up and morphological change during storms. The researchers ran XBeach in surfbeat mode on a computational grid covering six square kilometres, with cell sizes shrinking to 2.5 metres in the coastal strip, fine enough to resolve the small streets and gaps between buildings through which water can surge. For storms from the southwest, which strike Camogli directly, XBeach alone proved sufficient. For storms from the southeast, however, the Portofino Promontory partially shelters the coast, and waves had to be translated to the offshore boundary of the model domain using D-Waves, the SWAN-based wave module of the Delft3D suite, applied over a grid of 115 square kilometres encompassing the promontory.

Validating the model posed a familiar problem: no field measurements exist for the historical storms of interest. The team therefore compared simulated significant wave heights and directions with measurements from an oceanographic buoy located inside the model domain during two events in May and August 2025. The agreement was strong. For the August event, the model achieved a coefficient of determination of 0.96, a correlation coefficient of 0.98 and a scatter index of just 0.16 for wave height, while wave direction errors, expressed as circular root-mean-square error, ranged from 12.0 to 19.6 degrees across the two events, figures comparable to those reported in other coastal wave-modelling applications. A second, more dramatic validation came from simulating the storm of 3 to 5 November 2023, which was not part of the exposure dataset. That southwest storm destroyed a beach restaurant built on stilts, and the model’s simulated maximum run-up of 2.6 metres reached precisely the location of the collapsed structure.

With the model chain validated, the researchers computed two run-up indicators for every storm along six coastal transects: the run-up exceeded by two percent of the waves, and the maximum run-up. The contrast between wave directions was stark. For the twenty most intense southeast storms, run-up values almost never exceeded 1.8 metres and typically stayed below 1.3 metres, never crossing the two-metre threshold that marks the elevation above which tourism infrastructure begins to be threatened. Camogli is, in effect, only marginally exposed to storms from that quadrant, a consequence of the sheltering geometry of the Portofino Promontory.

Southwest storms told a very different story. Maximum run-up exceeded two metres in nearly every simulated event, and in five storms the two-percent run-up reached between 2.4 and 2.6 metres. The outlier was the storm of 29 October 2018, widely known as the Vaia Storm, which produced run-up values between 3.89 and 5.28 metres for the two-percent parameter and between 4.94 and 5.72 metres for maximum run-up. It was the only event in the entire 43-year record that pushed waves above the three-metre threshold, corresponding roughly to the maximum beach elevation before the urban settlement begins, where flooding of the town itself becomes a concrete possibility. That finding is consistent with the documented damage from the event and underscores a central message of the study: only long-term wave datasets can capture the rare but catastrophic storms that dominate coastal risk.

The spatial pattern of exposure revealed an uncomfortable structural problem. In earlier work on the nearby bay of Bonassola, a medium exposure class had acted as a buffer zone between low and high exposure, allowing authorities to issue graduated warnings and phase in mitigation measures. Along more than half of the Camogli coastline, that buffer is absent. Exposure jumps directly from low to high, because the beach is only ten to fifteen metres wide, has the steep profile typical of gravel and pebble shores, and is backed immediately by seawalls and building walls that leave no room for a wider dissipative beach. Small variations in wave run-up therefore produce abrupt transitions in exposure level, a dynamic that complicates emergency management and increases the likelihood of frequent high-level alerts.

That abruptness carries a human cost that the authors address head-on. Warning systems that cry wolf too often breed alert fatigue, eroding public trust and responsiveness, while overly precautionary closures of commercial activities impose economic costs and damage institutional credibility. The study found that choosing the more conservative maximum run-up indicator over the two-percent parameter would have generated only one additional flood warning across the entire historical record, limiting the fatigue risk in this particular setting, though the authors stress that the choice of indicator must be evaluated case by case. Because risk perception is strongly shaped by direct personal experience, and because the missing buffer zone is rooted in geomorphology and urban form rather than in any fixable technical shortcoming, the researchers argue that targeted risk communication and community preparedness must complement the modelling work.

The implications stretch well beyond one Ligurian town. As sea levels rise, coastlines like Camogli, with narrow beaches, rocky margins, intense urbanisation up to the waterline and no capacity to migrate landward, are likely to face growing storm exposure, making accurate, site-specific hazard assessment increasingly urgent. The study also flags its own limits: reliable results depend on accurate, up-to-date topographic and bathymetric data, which human development can rapidly invalidate, and on long-term offshore wave datasets, which remain scarce where buoy networks are sparse. Hindcast databases fill that gap, but uncertainties in them can propagate through the modelling chain. Even so, the message is clear and broadly applicable: in crowded coastal towns where a single metre of elevation separates safety from flooding, high-resolution modelling of decades of historical storms is not a luxury but a prerequisite for evidence-based warning thresholds, sensible mitigation planning and, ultimately, the protection of lives and livelihoods.

Subject of Research: Storm-induced coastal exposure modelling using hindcast wave data in a densely urbanised coastal area

Article Title: Modelling storm-induced coastal exposure using hindcast wave data in a densely urbanised coastal area

Article References: Carpi, L., Oneto, N., & Ferrari, M. (2026). Modelling storm-induced coastal exposure using hindcast wave data in a densely urbanised coastal area. Natural Hazards, 122(20), Article 649. https://doi.org/10.1007/s11069-026-08420-2

Image Credits: AI Generated

DOI: 10.1007/s11069-026-08420-2

Keywords: coastal exposure, wave run-up, storm hindcast, XBeach, coastal flooding, Ligurian Riviera, wave modelling, risk assessment, alert fatigue, sea-level rise, digital elevation model, Natural Hazards