Model Wave Heights Research Articles

Improving the accuracy of hurricane wave modeling in Gulf of Mexico with dynamically-coupled SWAN and ADCIRC

Dynamically-coupled SWAN and ADCIRC models have been applied to enhance the predictions of extreme waves and storm surges induced by hurricanes and sea level rise (SLR) in the Gulf of Mexico. The model performance was evaluated using Hurricane Michael, a Category-5 hurricane, as a case study. Modeled wave heights were compared to the observations. Results indicate that the dynamically-coupled SWAN-ADCIRC models substantially enhance the modeling accuracy. By comparing to the maximum observed 2.69 m of wave height near the hurricane landing site, the error is 0.04 m by the SWAN-ADCIRC models in comparison to the 0.39 m by the SWAN stand-alone simulation. Effects of sea level rise on hurricane wave heights were investigated under four SLR scenarios of 0.2m, 0.5m, 1m, and 1.5m. Results indicate that, as sea level rises, wave heights increase non-linearly in shallow waters near the hurricane landing site. At the wave observation station near the hurricane landing site, the ratio of the wave-height change to SLR increases to 117% and the ratio of the combined wave-surge change to SLR increases to 265%. Analysis indicates that this is due to the substantial percentage changes in water depth occurring in shallow water compared to deep water caused by SLR.

An earthquake-triggered submarine mass failure mechanism for the 1994 Mindoro tsunami in the Philippines: Constraints from numerical modeling and submarine geomorphology

Tsunamis have been known to result from a wide range of phenomena, such as earthquakes, volcanic eruptions, submarine mass failures, and meteorite impacts. Of earthquake-generated tsunamis, those arising from strike-slip mechanisms are less common, with the 1994 Mindoro tsunami in the Philippines among the few known examples. The 1994 Mindoro tsunami followed a Mw 7.1 earthquake along the right-lateral Aglubang River Fault. The tsunami affected the coasts surrounding the Verde Island Passage, one of the Philippines’ insular seas located between the islands of Luzon and Mindoro, and east of the West Philippine Sea margin. A total of 78 lives were lost due to the earthquake and tsunami, with 41 being directly attributed to the tsunami alone. Despite the close spatial and temporal association between the 1994 Mindoro earthquake and tsunami, previous numerical modeling suggests the need for other contributing mechanisms for the 1994 tsunami. In this study, we conducted submarine geomorphological mapping of the South Pass within the Verde Island Passage, with particular focus on identifying possible submarine mass failures. Identification of submarine features were based on Red Relief Image Map (RIMM), Topographic Position Index (topographic position index)-based landform classification, and profile and plan curvatures derived from high-resolution bathymetry data. Among the important submarine features mapped include the San Andres submarine mass failure (SASMF). The San Andres submarine mass failure has an estimated volume of 0.0483 km3 and is located within the Malaylay Submarine Canyon System in the Verde Island Passage, ∼1 km offshore of San Andres in Baco, Oriental Mindoro. We also explored two tsunami models (EQ-only and EQ+SMF) for the 1994 Mindoro tsunami using JAGURS. The source mechanisms for both models included an earthquake component based on the Mw 7.1 earthquake, while the EQ+SMF also included an additional submarine mass failure component based on the mapped San Andres submarine mass failure. Modeled wave heights from the EQ-only model drastically underestimates the observed wave heights for the 1994 Mindoro tsunami. In contrast, the EQ+SMF model tsunami wave height estimates were closer to the observed data. As such, we propose an earthquake-triggered, submarine mass failure source mechanism for the 1994 Mindoro tsunami.

Accuracy of numerical wave model results: application to the Atlantic coasts of Europe

Abstract. Numerical wave models are generally less accurate in the coastal ocean than offshore. It is generally suspected that a number of factors specific to coastal environments can be blamed for these larger model errors: complex shoreline and topography, relatively short fetches, combination of remote swells and local wind seas, less accurate wind fields, presence of strong currents, bottom friction, etc. These factors generally have strong local variations, making it all the more difficult to adapt a particular model setup from one area to another. Here we investigate a wide range of modeling choices including forcing fields, spectral resolution, and parameterizations of physical processes in a regional model that covers most of the Atlantic and North Sea coasts. The effects of these choices on the model results are analyzed with buoy spectral data and wave parameter time series. Additionally, satellite altimeter data are employed to provide a more complete performance assessment of the modeled wave heights as a function of the distance to the coast and to identify areas where wave propagation is influenced by bottom friction. We show that the accurate propagation of waves from offshore is probably the most important factor on exposed shorelines, while other specific effects can be important locally, including winds, currents, and bottom friction.

A modified breaker index formula for depth-induced wave breaking in spectral wave models

The breaker index, defined as the ratio of wave height to local water depth for incipient breaking, is a key parameter in contemporary spectral wave models. In the present study, we modified the breaker index of Goda's formula using the SWAN model (Simulating WAves Nearshore). We suggested that the breaker index is a function of the deep-water wave steepness, bottom slope and relative water depth. The formula is evaluated against observational data, and the evaluation indicates that the modeled wave heights using the formula are significantly improved, especially for cases over mild and flat slopes. The formula is further evaluated through comparisons with published formulas and the improved formula overall shows better performance.

Performance assessment of a Persian Gulf wind and wave forecasting system

ABSTRACT Data from three satellites (JASON2, CRYOSAT, and SARAL/AltiKa) and two buoys were used to examine the performance of wind and wave forecasting models in the Persian Gulf. The quality of the satellite data was first examined by comparing them with buoy data. Wind speed data from all three satellites were found to be close to in situ measurements; for wave heights, SARAL/AltiKa data showed the best comparisons. Model forecasts and buoy data indicated correlation coefficients between 0.72 and 0.86 for wind speeds and 0.85–0.94 for wave heights. The forecast was more accurate at the open sea buoy than at the coastal buoy. Relative to SARAL/AltiKa data, which was preferred for assessing modelled wave heights, correlation coefficients were between 0.86 and 0.90. For modelled wind speeds, correlation coefficients with the consolidated satellite dataset were between 0.80 and 0.88. Some anomalous results were partly attributed to the exclusion of air–sea temperature differences. The likelihood of a given forecast representing a given sea state was estimated. Generally, for lead times up to 48 h, a user could have a 78–92% expectation that a SWH forecast > 1 m would lie within ±30% of a sea-state represented by the SARAL/AltiKa measurements.

An analysis of high‐resolution modelled wave heights along the South African south coast suggests recent deterioration of sea state

AbstractUnderstanding climate variability and change together with fishing activities through high‐resolution scientific information and local observations can help implement and sustain improved coastal marine management strategies. Ethnographic data collected between 2010 and 2016 during previous studies with traditional handline fishers on South Africa's south coast suggest that fishers recently observed a deterioration in sea state and a reduction in sea days. Many fishers attributed this deterioration to increased climate variability. The fishers also suggested experiencing increased difficulty when leaving the nearshore through river mouths due to higher waves and stronger winds. The current study investigated these oceanic changes in terms of wave height variability and trends, using high resolution modelled wave data of the south coast for the period 1997–2013. Considering the fishers' perspectives, wave heights near their main launch sites during their main fishing season (austral summer) were focussed on. During the mid‐to‐late 2000s, a positive regime shift of summer nearshore wave heights occurred along the south coast and a statistically significant positive trend in summer nearshore wave heights occurred during the study period. These results support the local fishers’ observations suggesting that the sea state had been deteriorating, in terms of higher summer nearshore waves. While the magnitude of the observed change is not critical to small‐boat fisheries in the area at present, it may contribute to the decrease in available sea days and hence, adds to the stressors in the commercial and small‐scale sectors, further increasing their vulnerability to change.

Meteotsunami-related flooding and drying: numerical modeling of four Adriatic events

Meteotsunamis (localized ocean waves with periods similar to those of tsunamis but caused by meteorological phenomena) have been observed around the globe and are frequently observed in the Adriatic Sea. Numerous studies have focused on numerical modeling of meteotsunamis, but only a few have modeled flooding and drying of the coastal areas that play an important role in risk assessment. In this study, we model four historic meteotsunami events (Vela Luka Bay, June 21, 1978; Siroka Bay, August 22, 2007; Mali Losinj Bay August 15, 2008; Stari Grad Bay February 19, 2010) that occurred in the Adriatic Sea, using ADICRC, a flooding and drying capable ocean numerical model. Comparison of those results with similar simulations that do not use the flooding and drying algorithm was made to determine differences in modeled wave height. Three of the modeled events (Vela Luka, Siroka and Mali Losinj) are more accurately depicted if including the flooding and drying algorithm, suggesting that extreme events can be more realistically modeled than with the more commonly used cut-off depth (i.e., specifying minimum depth larger than expected maximum wave height). Modeling results for the fourth event (Stari Grad) confirm a previous assumption that flooding occurred due to the superposition of a storm surge and a meteotsunami, rather than a meteotsunami alone.

Hydrodynamic Modeling of a Reef-Fringed Pocket Beach Using a Phase-Resolved Non-Hydrostatic Model

The non-hydrostatic wave-flow model SWASH was used to investigate the hydrodynamic processes at a reef fringed pocket beach in southwestern Australia (Gnarabup Beach). Gnarabup Beach is a ~1.5 km long beach with highly variable bathymetry that is bounded by rocky headlands. The site is also exposed to large waves from the Southern Ocean. The model performance was evaluated using observations collected during a field program measuring waves, currents and water levels between June and July 2017. Modeled sea-swell wave heights (periods 5–25 s), infragravity wave heights (periods 25–600 s), and wave-induced setup exhibited moderate to good agreement with the observations throughout the model domain. The mean currents, which were highly-spatially variable across the study site, were less accurately predicted at most sites. Model agreement with the observations tended to be the worst in the areas with the most uncertain bathymetry (i.e., areas where high resolution survey data was not available). The nearshore sea-swell wave heights, infragravity wave heights and setup were strongly modulated by the offshore waves. The headlands and offshore reefs also had a strong impact on the hydrodynamics within the lagoon (bordered by the reefs) by dissipating much of the offshore sea-swell wave energy and modifying the pattern of the nearshore flows (magnitude and direction). Wave breaking on the reef platforms drove strong onshore directed mean currents over the reefs, resulting in off-shore flow through channels between the reefs and headlands where water exchanges from the lagoon to ocean. Our results demonstrate that the SWASH model is able to produce realistic predictions of the hydrodynamic processes within bathymetrically-complex nearshore systems.

Horizontal and Vertical Sediment Sorting in Tidal Sand Waves: Modeling the Finite‐Amplitude Stage

The bed of coastal seas displays a large number of rhythmic bed features, of which sand waves are relevant to study from an engineering perspective. Sediments tend to be well sorted over these bed forms, which is so far poorly understood in terms of modeling of finite‐amplitude sand waves. Using Delft3D, we employ bed stratigraphy and consider four different grain‐size classes, which are normally distributed on the phi scale. The standard deviation (sortedness) is then varied, whereas the geometric mean grain size is kept constant. The results show that, typically, the crests of sand waves are coarser than the troughs. Residual flow causes net sedimentation on the leeside of the crest, and, consequently, the general sorting pattern is distorted. Since larger grains experience a larger settling velocity, they are deposited on the upper lee slope, whereas the smaller grains are found on the lower lee slope. Due to sand wave migration, also, the internal structure of the sand wave is revealed, which follows the same sedimentation pattern as the lee slope surface. These results qualitatively agree with sorting patterns observed offshore. The sorting processes lead to longer wavelengths and lower wave heights, as a function of standard deviation. This relates to the dampening effect of suspended sediment transport for fine grains. Finally, it appears that the modeled wave heights fall in the same range as observations in the North Sea. These results are valuable for, for example, predicting the morphological response after engineering activities and determining suitable aggregates for sand extraction.

Response of an Annular Tuned Liquid Damper Equipped With Damping Screens

Abstract Annular tuned liquid dampers (TLDs) may be installed in slender structures with limited floor space, in which people and utilities must pass through the core, such as a wind turbine or observation tower. This study investigates an annular-shaped TLD equipped with damping screens. A linearized equivalent mechanical model capable of capturing the fundamental sloshing mode response of an annular TLD is presented. An experimental shake table testing program is completed to assess the performance of the model. Thirty-six frequency sweep tests consisting of various TLD configurations, excitation amplitudes, and excitation directions are completed. Good agreement is observed between the linearized equivalent mechanical model and experimental wave heights, sloshing forces, and energy dissipated per cycle that have been filtered to include only the fundamental sloshing mode response. The model is also observed to be in good agreement with experimental data for different excitation directions. The model is coupled to a generalized structure to investigate the response of a structure equipped with an annular TLD. The annular TLD is found to reduce the response of a generalized offshore wind turbine structure undergoing harmonic force excitation. The annular TLD provides performance comparable to an optimal linear tuned mass damper (TMD) with the same properties for a range of force excitation amplitudes.

The influences of the Kuroshio on wave characteristics and wave energy distribution in the East China Sea

In the present study, current effects on waves (CEW) in a strong western boundary current area is investigated. An analytical formula is first derived to address how the kinematic parameters of currents affect wave variations: strain rate, divergence and deformation rate. The formula is applied to numerical wave solutions in the East China Sea (ECS), where the Kuroshio flows through. It is found that modeled wave heights and wave periods with and without the CEW are of significant differences in the Kuroshio area and also carry strong seasonality. In January, when the wave direction is against the Kuroshio, wave heights and periods with the CEW are significantly higher and longer than those without the CEW, while the opposite is noted in July when the wave direction is in the same as the current. The wave energy is also greatly enhanced (lessened) along the Kuroshio axis in January (July), due to the CEW. The analysis of the model results also shows that different terms in the CEW dominate in different seasons. In January, the wave variation caused by the CEW is due to the deformation rate of the current velocity, while in July, it is caused by the strain rate of current velocity. The influence of the CEW on the waves in the Kuroshio region is about 10–20%.

Reconciling Storegga tsunami sedimentation patterns with modelled wave heights: A discussion from the Shetland Isles field laboratory

AbstractThe Shetland Isles represent an ideal field laboratory for tsunami geoscience research. This is due to the widespread preservation of Holocene tsunami sediments in coastal peat deposits. This study uses published accounts of the Holocene Storegga Slide tsunami to illustrate how two different approaches – mapping of tsunami sediments and numerical modelling – produce radically different run‐up heights. The Storegga Slide is one of the world largest submarine slides and took place ca 8150 cal yr bp on the continental slope west of Norway. The tsunami generated by the landslide deposited locally extensive sheets of marine sand and gravel, as well as redeposited clasts of peat across the contemporary land surface. These sediment accumulations have subsequently been buried by peat growth during the Holocene while exposures of the deposits are locally visible in coastal cliff sections. In several areas, the tsunami sediments can be traced upslope and inland within the peat as tapering sediment wedges up to maximum altitudes of between ca 8·1 m and 11·8 m above present sea level. Since reconstructions of palaeo‐sea level for Shetland for ca 8150 cal yr bp suggest an altitude of 20 m below high tide on the day that the tsunami struck, it has been inferred that the minimum tsunami run‐up was locally between 28·1 m (8·1 + 20 m) and 31·8 m (11·8 + 20 m). However, numerical models of the tsunami for Shetland suggest that the wave height may only have reached a highest altitude in the order of +13 m above sea level on the day the tsunami took place. In this paper a description is given of the sedimentary evidence for tsunami run‐up in the Shetland Isles. This is followed by an evaluation of where the palaeo‐sea level was located when the tsunami occurred. Significant differences are highlighted in tsunami inundation estimates between those based on the observed (geological) data and the theoretically‐modelled calculations. This example from the Shetland Isles may have global significance since it exemplifies how two different approaches to the reconstruction of tsunami inundation at the coast can produce radically different results with modelled wave height at the coast being considerably less than the geological estimates of tsunami run‐up.

Simulation of flooding and drying as an essential element of meteotsunami modelling

Meteotsunamis, sea level oscillations of meteorological origin with periods similar to those of tsunamis, can, in extreme cases, reach significant wave heights and cause large damage to property and even a loss of life. In this study we have taken a historic meteotsunami in the Adriatic Sea with a wave height of 6 m and attempted to improve upon previous numerical simulations in order to model it as accurately as possible. The great Adriatic meteotsunami of 21 June 1978 was most pronounced in Vela Luka Bay on Korčula Island where sea level oscillations alternately flooded the town and dried out the harbour. Previous data analysis and numerical modelling have shown that the cause of the flood was a travelling air pressure disturbance that caused Proudman resonance in the open sea, with the waves being further amplified by the basin's funnel shape, reflection at the bay head and constructive superposition with the waves coming in from the open sea. To model the sea level oscillations we used a finite element model with a flooding and drying option included, with the entire Adriatic Sea as the modelling domain. A synthetic air pressure disturbance propagating from the Italian coast and over the island at varying speeds and directions was used as the only forcing field in the model. This new setup resulted in a successful reproduction of harbour flooding and drying and an increase in modelled wave heights.

Examination of Generation Mechanisms for an English Channel Meteotsunami: Combining Observations and Modeling

AbstractOn the morning of 23 June 2016, a 0.70-m meteotsunami was observed in the English Channel between the United Kingdom and France. This wave was measured by several tide gauges and coincided with a heavily precipitating convective system producing 10 m s−1 wind speeds at the 10-m level and 1–2.5-hPa surface pressure anomalies. A combination of precipitation rate cross correlations and NCEP–NCAR Reanalysis 1 data showed that the convective system moved northeastward at 19 ± 2 m s−1. To model the meteotsunami, the finite element model Telemac was forced with an ensemble of prescribed pressure forcings, covering observational uncertainty. Ensembles simulated the observed wave period and arrival times within minutes and wave heights within tens of centimeters. A directly forced wave and a secondary coastal wave were simulated, and these amplified as they propagated. Proudman resonance was responsible for the wave amplification, and the coastal wave resulted from strong refraction of the primary wave. The main generating mechanism was the atmospheric pressure anomaly with wind stress playing a secondary role, increasing the first wave peak by 16% on average. Certain tidal conditions reduced modeled wave heights by up to 56%, by shifting the location where Proudman resonance occurred. This shift was mainly from tidal currents rather than tidal elevation directly affecting shallow-water wave speed. An improved understanding of meteotsunami return periods and generation mechanisms would be aided by tide gauge measurements sampled at less than 15-min intervals.

An assessment of the impact of surface currents on wave modeling in the Southern Ocean

This paper presents an assessment of the impact of the ocean circulation on modeled wave fields in the Southern Ocean, where a systematic positive bias of the modeled wave height against altimetry data has been reported. The inclusion of ocean currents in the wave model considerably reduces the positive bias of the simulated wave height for high southern latitudes. The decrease of wave energy in the presence of currents is almost exclusively related to the reduction of the relative wind, caused by an overall co-flowing current field associated with the Antarctic Circumpolar Current. Improvements of the model results are also found for the peak period and the mean period against a long-term moored buoy. At the mooring location, the effect of currents is greater for larger and longer waves, suggesting remotely generated swells are more influenced by the currents than local waves. However, an additional qualitative analysis using high-resolution currents in a finer grid nested to the global coarser grid shows that typical resolution of global hydrodynamic reanalysis is not sufficient to resolve mesoscale eddies, and as a consequence, the simulation of mesoscale wave patterns can be compromised. The results are also discussed in terms of the accuracy of forcing fields.

Storm surge along the Pacific coast of North America

AbstractStorm surge is an important factor that contributes to coastal flooding and erosion. Storm surge magnitude along eastern North Pacific coasts results primarily from low sea level pressure (SLP). Thus, coastal regions where high surge occurs identify the dominant locations where intense storms make landfall, controlled by storm track across the North Pacific. Here storm surge variability along the Pacific coast of North America is characterized by positive nontide residuals at a network of tide gauge stations from southern California to Alaska. The magnitudes of mean and extreme storm surge generally increase from south to north, with typically high amplitude surge north of Cape Mendocino and lower surge to the south. Correlation of mode 1 nontide principal component (PC1) during winter months (December–February) with anomalous SLP over the northeast Pacific indicates that the dominant storm landfall region is along the Cascadia/British Columbia coast. Although empirical orthogonal function spatial patterns show substantial interannual variability, similar correlation patterns of nontide PC1 over the 1948–1975 and 1983–2014 epochs with anomalous SLP suggest that, when considering decadal‐scale time periods, storm surge and associated tracks have generally not changed appreciably since 1948. Nontide PC1 is well correlated with PC1 of both anomalous SLP and modeled wave height near the tide gauge stations, reflecting the interrelationship between storms, surge, and waves. Weaker surge south of Cape Mendocino during the 2015–2016 El Niño compared with 1982–1983 may result from changes in Hadley circulation. Importantly from a coastal impacts perspective, extreme storm surge events are often accompanied by high waves.

Modelled Impact of Vegetation Heterogeneity and Salt-Marsh Zonation on Wave Damping

ABSTRACT van Loon-Steensma, J.M.; Hu, Z., and Slim, P.A., 2016. Modelled impact of vegetation heterogeneity and salt-marsh zonation on wave damping. This paper analyses the effect of observed vegetation characteristics on modelled wave heights. Detailed information on species composition, as well as on height, number of stems, and diameter of the plant species of a restored salt marsh on the Wadden barrier island of Terschelling was used to parameterize and apply the Simulating Waves Nearshore – Vegetation wave model to a schematized restored salt-marsh zone in front of the dike. The results indicate that wave damping by vegetated forelands is strongly related to vegetation heterogeneity and salt-marsh zonation. The modelling works suggest that at the study site under storm conditions with a frequency of 5–10 times/y, a vegetated foreland of some 90 m in width will dampen the wave height more than 80%, whereas under extreme conditions (1/2000 y) a foreland covered with dense vegetation will dampen the wav...

Numerical modeling of the effects of wave energy converter characteristics on nearshore wave conditions

Modeled nearshore wave propagation was investigated downstream of simulated wave energy converters (WECs) to evaluate overall near- and far-field effects of WEC arrays. Model sensitivity to WEC characteristics and WEC array deployment scenarios was evaluated using a modified version of an industry standard wave model, Simulating WAves Nearshore (SWAN), which allows the incorporation of device-specific WEC characteristics to specify obstacle transmission. The sensitivity study illustrated that WEC device type and subsequently its size directly resulted in wave height variations in the lee of the WEC array. Wave heights decreased up to 30% between modeled scenarios with and without WECs for large arrays (100 devices) of relatively sizable devices (26 m in diameter) with peak power generation near to the modeled incident wave height. Other WEC types resulted in less than 15% differences in modeled wave height with and without WECs, with lesser influence for WECs less than 10 m in diameter. Wave directions and periods were largely insensitive to changes in parameters. However, additional model parameterization and analysis are required to fully explore the model sensitivity of peak wave period and mean wave direction to the varying of the parameters.

Estimation of Shallow-Water Breaking-Wave Height From Synthetic Aperture Radar

The relationship between synthetic aperture radar (SAR) signatures of depth-limited breaking waves and wave height is studied. Wave height is estimated from SAR images using an empirically derived relationship that exploits the azimuthal shift in SAR images associated with moving scatterers. This relationship is derived from in situ measurements rather than from an idealized model of breaking waves as was done in a previous study. We find that the lengths of the SAR signatures are correlated with the observed significant wave height (the correlation coefficient is 0.78) for a range of wave conditions. The relationship between the wave heights and velocity bandwidths from the field data is similar to that between simulated (with a Boussinesq surface wave model) wave heights and velocity ranges (correlation $\mbox{coefficient}=0.82$ ).

Will the Effects of Sea-Level Rise Create Ecological Traps for Pacific Island Seabirds?

More than 18 million seabirds nest on 58 Pacific islands protected within vast U.S. Marine National Monuments (1.9 million km2). However, most of these seabird colonies are on low-elevation islands and sea-level rise (SLR) and accompanying high-water perturbations are predicted to escalate with climate change. To understand how SLR may impact protected islands and insular biodiversity, we modeled inundation and wave-driven flooding of a globally important seabird rookery in the subtropical Pacific. We acquired new high-resolution Digital Elevation Models (DEMs) and used the Delft3D wave model and ArcGIS to model wave heights and inundation for a range of SLR scenarios (+0.5, +1.0, +1.5, and +2.0 m) at Midway Atoll. Next, we classified vegetation to delineate habitat exposure to inundation and identified how breeding phenology, colony synchrony, and life history traits affect species-specific sensitivity. We identified 3 of 13 species as highly vulnerable to SLR in the Hawaiian Islands and quantified their atoll-wide distribution (Laysan albatross, Phoebastria immutabilis; black-footed albatross, P. nigripes; and Bonin petrel, Pterodroma hypoleuca). Our models of wave-driven flooding forecast nest losses up to 10% greater than passive inundation models at +1.0 m SLR. At projections of + 2.0 m SLR, approximately 60% of albatross and 44% of Bonin petrel nests were overwashed displacing more than 616,400 breeding albatrosses and petrels. Habitat loss due to passive SLR may decrease the carrying capacity of some islands to support seabird colonies, while sudden high-water events directly reduce survival and reproduction. This is the first study to simulate wave-driven flooding and the combined impacts of SLR, groundwater rise, and storm waves on seabird colonies. Our results highlight the need for early climate change planning and restoration of higher elevation seabird refugia to prevent low-lying protected islands from becoming ecological traps in the face of rising sea levels.