Symposium Program

Wave buoys were installed along Australia's east coast in the 1970s following severe coastal erosion events and now provide invaluable long-term records for understanding coastal impacts. These observations enable investigation of trends in extreme wave events and assessment of whether climate change, and associated shifts in weather patterns, influence the highest waves recorded each year, measured as the annual maximum significant wave height.
In this study we apply non-stationary extreme value analysis to buoy records to detect changes in the probability and magnitude of extreme waves. To complement these in-situ observations, we use the 40- year (1981-2020) Coupled Coastal Hazard Prediction System (CCHaPS) hindcast, which provides simulated wave records with broader spatial and temporal coverage, including areas without long-term buoy data.
The robustness of detected trends is assessed in the context of data limitations, including uncertainties in early satellite observations and the absence of high-quality hourly buoy records prior to the early 1980s.
Results show that extreme wave heights have increased at locations along the east coast. Interestingly, while some of these observed increases are supported by the hindcast data, in some cases the trends differ from recent studies on the changes in extratropical lows and from global climate projections, which generally suggest a weakening of extreme wave conditions in the subtropical east coast of Australia.
This study highlights the value of integrating long-term observational records with hindcast modelling to improve understanding of regional wave climate change through non-stationary extreme value analysis. It also shows that local trends can diverge from global, coarse-resolution projections, reinforcing the need for sustained coastal monitoring, and regional climate and coastal projection modelling to inform hazard risk assessments and adaptation planning.

Australia possesses some of the best offshore wind resource in the world, which is yet untapped. There has been progress towards establishing an offshore wind industry since the passage of the Offshore Electricity Infrastructure Act 2021, with the first offshore wind turbines in Australian waters envisaged to be installed in the early 2030s. The offshore wind industry traces its origins to the North Sea, where the wave climate is rather different from the Australian wave conditions. This talk will cover the implications of Australian wave climate on design of offshore wind support structures due to the different wave-induced loading.
The Australian wave climate along the southern margin is dominated by long-period swell waves. In this regime, non-linear wave loads, which act at frequencies different to the main energy-bearing wave components, are amplified (compared to shorter-period waves, more typical of the North Sea). This has design implications for both bottom-fixed and floating offshore wind structures, since this nonlinear forcing can be a significant fraction of the total load and can, moreover, induce resonant structural and motion responses with consequences for fatigue damage.
The year-round persistence of waves on Australian shores from the Southern Ocean is also strikingly different to the North Sea conditions, where considerable seasonal variation exists with calm summer conditions yet more severe winter storms. Offshore wind structures in Australian waters therefore need to be designed to withstand this continual loading. There are other important differences related to directional characteristics and coupling to local wind conditions, for example. Their relevance for offshore wind structures will also be outlined.

The bimodal wave climate of the semi-protected, shallow Gulf St Vincent in South Australia is analysed using a forty-year (1980-2020) wave hindcast. Sea and swell wave components were modelled both independently and through an integrated model that accounts for both partitions. The hindcast was validated against two wave buoys located off the Adelaide metropolitan coast, and key wave parameter anomalies were calculated across the Gulf. Climatic teleconnections were examined, revealing that the Southern Annular Mode (SAM) Index shows the strongest correlation with wave anomalies. Seasonal correlations were also identified between wave anomalies and fluctuations in the Southern Oscillation Index (SOI) and the Dipole Mode Index (DMI). Future projections of these climate drivers from existing literature were used to assess potential impacts on the gulf's bimodal wave climate. A positive trend in the SAM is expected to reduce both overall wave height and the westerly component of wave energy in the Gulf. In contrast, increased variability in the SOI and DMI is projected to result in more extreme wave conditions during winter and spring. Understanding these trends enables coastal managers to proactively address seasonal and annual wave impacts, while also offering insight into potential long-term changes in the region's wave climate.

Future projected extreme ocean surface wave uncertainties challenge the predictions of future coastal extreme sea level projections. These uncertainties stem from climate natural variability and limitations in General Circulation Models (GCMs), particularly their ability to simulate extreme wind speed events due to physical constraints and coarse spatial resolution. To refine estimates of future wind wave extremes and assess their impact on extreme sea levels across the Australasian region, we employed a state-of-the-art global ocean wave climate model (Meucci et al., 2023, 2024) forced by downscaled surface wind fields from two CMIP6 GCMs. The downscaling was carried out using the BARPA-R land-surface Regional Climate Model (Howard et al., 2023) and the CCAM coupled atmosphere-ocean RCM with spectral nudging techniques (Thatcher et al., 2009). The two GCMs were selected based on their performance in simulating global wind wave climates (Meucci et al., 2024) and the priorities set by the CORDEX Australasia community. Focusing on extreme wind waves, we conducted a stationary and non-stationary Extreme Value Analysis (EVA) to examine sensitivity to the downscaling models, temporal variability, and three climate change scenarios: SSP1-2.6, SSP3-7.0, and SSP5-8.5. The analysis shows that increasing the wind speed spatial resolution adds complexity to the result interpretation, emphasising the impact of the chosen downscaling methodology in interpreting uncertainties in future wind wave extreme projections. Furthermore, the Australasian domain presents a particularly compelling test case, as it spans three distinct climatic regions: Tropical, Sub-Tropical, and Southern Ocean, each contributing to a wide range of extreme sea states. From Tropical Cyclones to Extra-Tropical Cyclones and synoptic low-pressure systems, this diverse exposure poses a significant challenge for wave climate models, testing their ability to capture the full spectrum of extreme wave events.

The Our Changing Coasts (OCC) program is a comprehensive initiative to quantify how climate-driven ocean processes, sea-level rise, and vertical land movements collectively dictate coastal geomorphic change in New Zealand. A core component of this effort is the generation of high-resolution, multi-model projections for wave and storm surge climates, which are critical inputs for understanding and predicting these coastal transformations.
For wave climate, we use a nested modeling suite driven by two CMIP6 global climate models, ACCESS-CM2 and EC-Earth3. A global WAVEWATCH III model provides spectral boundary conditions for a regional 5-km SWAN model, which is forced by NIWA's downscaled atmospheric data to significantly improve coastal wind representation. Storm surge projections employ an unstructured SCHISM grid with a resolution up to 1-km, also utilising the high-resolution NIWA winds. This modeling suite provides continuous projections from 1985 to 2100 under historical, SSP2-4.5, and SSP3-7.0 scenarios.
A key output is the generation of high-resolution, 3-hourly 2D wave spectra at over 4,000 global and 1,300 New Zealand locations. These spectral data, along with gridded wave parameters, provide unprecedented detail for downscaling and advanced spectral analysis. All data will be disseminated through the Oceanum Datamesh platform, facilitating broader research and application, including the establishment of shoreline change and sediment budget models within the OCC program.

Accurately capturing tropical cyclone (TC) wind fields is critical for wave and storm surge modelling in coastal regions. This study assesses the impact of three parametric TC wind models (Kepert, McConochie, and Chavas-Lin-Emanuel (CLE15)) on the simulation skill of wave and surge events in Queensland, Australia. For comparison, we also evaluate the performance of the 4-km BARRA-C2 regional atmospheric reanalysis. Using the coupled hydrodynamic-waves model SCHISM-WWMIII on an unstructured grid, we simulate historical TC cases, validating significant wave heights and surge levels against wave buoy, satellite altimeters and tide gauge observations.
Our results show that parametric models are essential for resolving the intense inner-core wind structure of severe TCs, which remains underrepresented in regional reanalysis products, even high-resolution ones as BARRA-C2. While all three parametric approaches improve wave and surge predictions relative to the reanalysis winds, there are notable differences. The McConochie (double vortex) model overestimates the spatial footprint and duration of hazardous conditions, leading to biased wave and surge predictions. In contrast, both the CLE15 and Kepert models demonstrate a strong performance in wave and storm surge validations, with CLE15 providing slightly more accurate and consistent results across most cases.
These findings underscore the importance of wind field selection in operational and research applications of TC hazard modelling. Further, we propose that validating TC wind fields against wave height observations offers a more faithful assessment of wind accuracy (than point-based wind measurements), as waves act as natural integrators of the wind field, reflecting the cumulative effects of wind over broad spatial and temporal scales.

Ocean engineering models for the kinematics of large ocean waves simply reduce those for a uni-directional wave field by a factor to account for directional spreading, typically ~ 0.95x for extra-tropical winter storms and ~ 0.9x for tropical cyclones. Is this adequate?
NewWave is a linear focussed wave group describing the average shape of extreme wave crests in any random sea-state. It has already been used as an initial condition for numerical simulations of waves to gain knowledge on the local shape and kinematics of extreme waves in winter storm sea-states, with a single mean wave direction and spreading around this mean. In contrast, waves in tropical cyclonic seas have the mean wave direction changing rapidly with frequency as well as directional spreading around this mean direction, introducing a "net twist" in the waves as demonstrated by Young and others. We use NewWave a few periods before focus as the initial condition and simulate fully nonlinear wave evolution using the well-validated solver (OceanWave3D).
We expected to find that the nonlinear interactions of the waves are smaller than the same counterpart with no twisting, as the wave components have less time to interact with each other as focusing occurs. However, current results show that the linear components of the wave group are not much different compared to the non-twisted cases, both showing small shifts of focus away from the predicted point based on linear theory and no significant extra (linearised) peak elevation. The effects of the twist become more obvious in the higher order terms, crest-trough asymmetry and beyond. This research also explores the full kinematics profiles with depth below the largest crests, as this velocity field drives the drag loading on jacket-type structures for both oil & gas and fixed wind applications.

The northwest shelf region of Western Australia is well known for its mineral resources and is an important region for the resource industry. Severe weather events, such as tropical cyclones, pose a significant and recurring risk for essential operations for offshore industries. For example, analysis of tropical cyclone track data indicates an average occurrence of 6 events per season for the northwest shelf region. The Bureau has developed a prototype coupled atmosphere-ocean-wave model spanning parts of the northwest shelf within the regional coupled working group under the unified model partnership. At kilometre scale, the model can explicitly represent processes related to eye-wall formation, the interaction between outer flow and inner core, and asymmetry under the influence from surface gravity waves, ocean currents, ocean temperature and tides. This presentation describes the modelling system, its model components, and coupling configuration. A series of illustrative examples and key insights will be given for a severe weather event in 2019, tropical cyclone Veronica.

Coastal vegetation is increasingly recognised not just for its ecological value, but as a vital natural infrastructure for shoreline protection. As climate-driven hazards intensify, the role of mangroves, saltmarshes, and seagrasses in buffering wave energy and stabilizing sediments is gaining global attention as a resilient solution for coastal adaptation. Critical to the utilisation and ultimately design of mangroves in coastal protection is an understanding of the wave attenuation performance in situ under extreme conditions when their potential value is greatest. Furthermore, due to the vast variations in intra-species allometry, obtaining locally relevant observations is crucial to build confidence with designers, engineers and stakeholders.
An array of pressure sensors, current meters and offshore wave buoy were deployed in January through April 2025 in Deception Bay QLD across a gradient in mangrove life-stage. On 8th March 2025 the study site was directly impacted by a Category 1 Tropical Cyclone (TC Alfred). Significant wave attenuation was observed, compared to waves incident on an exposed, adjacent seawall. This presentation will provide an overview of the wave attenuation experiment, mangrove vegetation measurements and within canopy hand-held lidar observations as we work towards in situ estimates of life-stage dependent frontal area observations and associated wave drag coefficients.

The existence of 'freak' or 'rogue' waves is an important question in ocean engineering and related to local vs. global nonlinearity in wave evolution. Commonly, sea-states are approximated as a linear random Gaussian process with known wave spectrum (usually JONSWAP) and directional spreading (often cos^2s Θ/2 or wrapped normal), leading to waves near Rayleigh distributed.
However, for narrow-banded and uni-directional waves on deep water, 4-wave resonances and Benjamin-Feir type instabilities produce localised extremes. Thus, the Peregrine breather in the NLS equation has a local peak 3x the regular wave background. However, it is well-known that broadbandedness, directional spreading, and finite water depth all act to weaken Benjamin-Feir type dynamics.
NewWave is a simple model for the average shape of extreme wave crests arising by chance in a linearly dispersive sea. We use fully nonlinear NewWave simulations starting 12 periods before linear focus to give the local shape and kinematics of extreme waves for winter storm sea-states. The wave spectrum is JONSWAP (T_p = 15.3s, γ = 2, upper cut-off for the linear components at 2.5ω_p, RMS spreading angle 22.5°), and water depth 92m (k_p d = 1.7).
Even for very tall focus events (linear crest up to 18m), the structure of the fully nonlinear waves around focus in time and space is close to linear with small nonlinear changes in the underlying group shape – slight contraction in the mean wave direction, slight expansion laterally, but no extra elevation. The expected bound harmonics produce crest-trough asymmetry, the surface set-down and return flow beneath the wave group. In contrast, the localised kinematics in the crests are very severe, with the peak horizontal velocity approaching the crest speed. Thus, for intermediate water depth and realistic sea-state properties, even very steep waves are only locally nonlinear, so unlikely to be characterised as 'rogue' or 'freak'.

When wind waves break in the nearshore, free long (infragravity) waves are generated through two mechanisms: breakpoint forcing and bound wave release. Previous studies have highlighted that lower-frequency groups breaking on steep slopes favour long wave generation by breakpoint forcing, whereas higher-frequency groups breaking on mild slopes favour bound wave release. However, no study has established and demonstrated quantitative thresholds for when each mechanism will dominate. In this paper, we use a one-dimensional linear numerical model to quantify each mechanism and assess their relative dominance in free long wave generation, as a function of wave and bathymetry parameters. The results show that the individual contributions of breakpoint forcing and bound wave release do not add up linearly to the total long wave energy generated from both mechanisms due to the phase differences between free long waves generated from each mechanism. The combination of mechanisms always results in a net smaller long wave amplitude. The normalised bed slope parameter is shown to be effective in differentiating between the dominance of the breakpoint-forced and bound wave release mechanisms. The efficiency of breakpoint forcing is found to be independent of bed slope, wave group frequency and short-wave period. Therefore, the relative importance of each mechanism to long wave generation depends mainly on the efficiency of the bound wave release, which varies with these parameters. Overall, the bound wave release mechanism tends to dominate under most conditions, except for cases with low infragravity frequencies that occur on steep slopes (e.g., typical of reef environments).

Coastal areas with heterogeneous geomorphological and ecological features present challenges for understanding wave dynamics and their implications for sediment transport and shoreline change. This study focuses on Victor Harbor (South Australia), a highly variable coastal setting characterised by offshore islands, reefs, a shore platform, seagrass meadows, and human-built structures such as seawalls, a causeway, and a breakwater. These elements interact to modulate wave energy through diffraction, dissipation, and refraction, resulting in complex spatial wave transformation.
To investigate these dynamics, two Spotter wave buoys were deployed in early winter 2025: one offshore at 30 m depth and another inshore at 15 m, capturing the evolution of wave conditions across the bays. Additionally, five pressure sensors and one ADCP were installed along the nearshore to measure variations in wave height near the coast. This instrument setup supports a detailed analysis of wave propagation across this mixed natural-anthropogenic environment.
While a long-term wave buoy in a more exposed offshore location provides over 20 years of wave climate data, its applicability to nearshore conditions within the embayed, reef-sheltered system of Victor Harbor is limited. The present deployment offers a rare opportunity to resolve how waves evolve as they move from deep, exposed waters to complex, sheltered coastal environments. This finer spatial resolution is essential to improving our understanding of wave-driven processes in diverse ecological and geomorphological settings.
Preliminary analysis of wave data across five beaches within two adjacent bays reveals notable differences in wave conditions despite a consistent offshore wave climate. Tidal fluctuations and wind direction play key roles in modulating nearshore wave heights. Ongoing data collection is promising, with at least three storm events captured; full retrieval and processing of this dataset is planned for mid-August 2025. These findings will support future modelling of wave-driven sediment transport and coastal change.

Along the coast of Robe, South Australia, a series of pocket beaches sit within a headland-bay system exposed to a highly variable, bimodal wave climate: long-period swell from the southwest and shorter-period sea waves from the south to the north. As waves refract, diffract, and break around the headland, they drive site-specific sediment transport patterns that differ significantly from open coastlines.
Using in situ wave observations from two wave buoys and a validated numerical model, this study examines how wave forcing affects alongshore transport gradients and beach volume fluctuations. Results show consistent seasonal signals: erosion during high-energy winter conditions with energetic sea waves from the north and accretion during calmer summer periods. However, transport pathways vary across beaches due to differences in orientation, exposure, and proximity to coastal structures.
Looking forward, climate-driven changes in wave energy and direction raise questions about future beach stability. Numerical modelling under projected scenarios suggests that more intense westerly storm events could enhance sediment transport in exposed areas and more sheltered beaches may face increased net sediment loss.
This presentation will discuss how resolving nearshore wave processes in embayed settings is critical for predicting sediment transport and informing adaptive coastal management under future wave climates.

Natural and artificial reef structures provide effective coastal protection by attenuating incident wave and current energy. Anticipated effects of climate change, such as sea level rise, threaten coastal zones and are predicted to reduce the protection provided to coastlines by reef structures. The coastal protection that a reef structure can provide through its impact on wave and current dynamics is dependent on a variety of factors (i.e., morphology, geometry). Research related to the coastal protection offered by reef structures tends to focus on coral and artificial reef structures, neglecting limestone reefs which are common features of the more than 9000 kilometres of the southern Australian coastline, and can present vastly different morphologies, geometries, and roughness factors compared to coral or artificial reefs. Investigating the influence of limestone reefs on wave and current dynamics is vital in understanding their potential present-day protective role and how this may change in changing climate conditions. This research addresses the need for research on wave and current dynamics around limestone reefs by presenting in-situ observations from a temperate limestone reef at Town Beach, Robe, South Australia. Pressure sensors and acoustic doppler current profilers were deployed on a cross-reef transect to investigate rates of wave attenuation, shifts in spectral wave energy, and changes to current speed and direction. Field measurements indicate that the Town Beach reef effectively dissipates approximately 42% of gravity wave energy, resulting in a domination of lower frequency, infragravity wave energy within the landward reef-protected lagoon. Current speeds landward of the reef are approximately one third lower than those observed seaward of the reef. Bidirectional flows (north-south and east-west, respectively) occur at higher water levels, changing to unidirectional flows (south and west, respectively) at lower water levels.

Artificial reef structures can be designed to provide coastal protection comparable to natural reefs and, in some cases, to traditional engineered structures such as submerged rubble mound breakwaters. The transformation and attenuation of surface wave energy across submerged structures occurs through a combination of (1) partial reflection of waves offshore, (2) depth-limited wave breaking, and, for reefs with high internal porosity, (3) dissipation by drag forces associated with flows within the reef. The relative contribution of these mechanisms is controlled by geometric properties of the artificial reef, which differ from typical submerged rubble mound breakwaters, as well as by the incident wave and water level conditions. We present findings from a series of experimental studies conducted in the University of Western Australia Coastal and Offshore Research Lab (CORL) wave flume across a broad range of wave conditions and water depths using different artificial reef modules. Results show how wave reflection is strongly influenced by different resonant responses associated with larger-scale reef geometry. The relative contribution of wave breaking and drag dissipation shift as conditions transition from deep to shallow freeboards, with slightly varying observations when the reefs module porosity and surface roughness change. These findings provide new insights into the mechanisms of wave transformation across artificial reef structures, supporting the development of more effective nature-based strategies for coastal protection.

Sustained, near real-time observations of waves in coastal regions are a fundamental pre-requisite to monitor, predict, and forecast coastal hazards that threaten coastal populations, infrastructure and marine operations. Additionally, near real-time temperature observations in coastal regions are required to understand the impacts of environmental disturbances (both extreme events and long-term changes) to coastal ecosystems. Thus, the establishment of a nationally coordinated network of surface wave and temperature observations is required to support a wide range of end user needs, including research, coastal managers, and marine industries. In 2024, Australia's Integrated Marine Observing System (IMOS) established the Coastal Wave Buoy facility (IMOS CWB), comprising a central facility based at the University of Western Australia, several sub-facilities managed by a variety of institutions across Australia, and a steering committee of representatives from both those institutions, IMOS and the Australian Ocean Data Network (AODN). Together, they are responsible for maintaining 23 IMOS funded and 34 co-invest sites spread around Australia's nearshore region that are collecting near real-time surface wave and temperature observations. The generated information is processed in Near Real-Time mode, as data is transmitted via satellite or cellular communications, and Delayed mode, as the instruments are retrieved and internally stored data is collected. Processing is performed by the central facility, when data undergoes quality control checks following QARTOD guidelines before being made available via the AODN. Near real-time data is also accessible through AusWaves, a website maintained by the IMOS CWB central facility that provides visualizations for wave parameters and 2D spectra. Deployments commenced late 2024, and within the first few months of operation, the network has already captured critical data from multiple tropical cyclones and a significant marine heatwave. Here, we present the facility's structure and main objectives, as well as highlight the value of near real-time observations during extreme events.

This study evaluates the accuracy and quality of wave data products from the Surface Waves Investigation and Monitoring (SWIM) instrument onboard the China France Oceanography Satellite (CFOSat), using extensive multi-platform wave buoy measurements.
SWIM, launched in 2018, is the first satellite-based wave scatterometer capable of measuring directional wave spectra globally. It uses near-nadir radar beams to scan the ocean surface and produce mean slope spectra and wave statistics, along with conventional altimeter-derived significant wave height and wind speed.
The SWIM data are particularly valuable in regions with sparse in-situ measurements, such as the Southern Ocean. This study compares SWIM wave statistics ‐ significant wave height, peak wave period, mean wave period, and peak direction ‐ against three buoy datasets: the National Data Buoy Centre (NDBC), Sofar drifting spotter buoys, and the Southern Ocean Flux Station (SOFS). These datasets span various timeframes and buoy configurations, enabling broad validation across diverse oceanic conditions.
Results show strong agreement for significant wave height and peak direction across all datasets. Mean and peak wave periods show lower accuracy, with mean wave period being the least reliable. However, applying a neural network-based method significantly improves mean wave period estimates. Initial comparisons of wave spectra are promising, with lower accuracy in the low-frequency range, possibly due to known spectral artifacts in SWIM data. Performance varies across SWIM's off-nadir beams, with the 10-degree beam and a weighted average of all beams yielding the best results. Among buoy datasets, the Sofar drifting buoys show the highest agreement.
Validated SWIM data with established accuracy and error metrics offer substantial value for global ocean wave research, especially in data-sparse regions.

This presentation discusses the development and application of a motion compensation technique for turbulence intensity (TI) measurements obtained from an RPS LiDAR Buoy deployed offshore.
The RPS LiDAR Buoy, constructed in Perth by RPS and other local manufacturers, employs a ZX 300M LiDAR system designed for measuring the wind profile up to full turbine blade height (~300 m). The buoy is used for measuring wind profiles at potential offshore wind farm sites, aiding in wind energy yield assessment.
While buoy motion has minimal impact on the mean 10-minute wind speed measurement, it artificially inflates TI estimates by increasing the variability in raw velocity data.
RPS have developed and implemented a motion compensation technique which is applied to the raw ZX LiDAR data with the aim of converting TI measured in the presence of buoy motion to the equivalent static LiDAR measured TI.
This advancement enables more accurate TI characterisation offshore, improving the reliability of wind variability assessments essential for optimising turbine performance and maintenance strategies in offshore wind farms.

Spectral wave forecasts, produced by national forecasting agencies, are critical for offshore industries, including shipping, oil and gas, renewable energy, and emergency response. However, numerical forecasts exhibit errors, arising from uncertainty in the model's initial and boundary conditions, as well as numerical approximations. The evolution of error over the forecast horizon is condition-dependent, acting as a function of wave state, due to the chaotic nature of the ocean system.
To reduce prediction error, conventional statistical post-processing approaches, such as Model Output Statistics (MOS), are effective in correcting systematic bias, yet limited by their inherent linearity and rigidity. Deep learning presents a flexible, nonlinear post-processing approach that is well suited to the large volumes of high-dimensional data produced by operational wave forecasts. This study evaluates the extent to which deep learning used as a post-processing system can improve the skill of site-specific offshore wave forecasts.
The deep learning approach is formulated as a supervised regression problem, where the model learns site-specific prediction residuals over the temporal horizon (residuals are the difference between ground-truth buoy observation and operational forecast). The deep learning model input is composed of archived operational wave forecasts of bulk wave parameters at a synoptic scale (spatial extent of atmospheric phenomenon covering 1000 km or more) and site-specific buoy observational measurements as the wave-state ground-truth. The gridded multivariable forecasts as model input (in which variables act as channels) motivate the selection of a convolutional neural network, capable of capturing complex spatio-temporal patterns. The evaluation of model performance compares corrected forecasts against baseline operational skill at each observation site.
Based on the proposed post-processing methodology, we expect to enhance site-specific forecast skill over the temporal horizon compared to current operational predictions. To assess this improvement, we will present a comparative analysis of operational and post-processed forecasts at three locations in Western Australia, focusing on swell events which are critical to the offshore industry.

The interaction between long-distance swell waves from Southern Ocean, locally generated wind waves (sea waves), and other environmental factors such as sea breezes, ocean currents, reflection, and refraction create a highly complex and variable wave climate along the Western Australian coast. To improve the prediction and analysis of this complex wave climate, a high-resolution coupled ADCIRC+SWAN model was developed to simulate wave heights, water levels, and surge heights along the WA coast. Over the past six months, the model has operated as an experimental forecasting system. Its predictions have been validated against near real-time data from tide gauge and wave buoy observations provided by the Department of Transport, WA. The model successfully captured large wave height events associated with three winter storm fronts between April 15 and May 10, 2005. These events caused significant coastal erosion along the Lancelin Coastal Stretch.
The model results indicate that wave induced water level setup along the coast vary significantly both spatially and temporally. Notably, substantial wave setup was observed along the southwest and southern coastlines in response to incoming Southern Ocean swells. As large swell waves propagate eastward, residual water levels from wave setup along the southern coast behave like forced shelf waves. For detailed information on the coupled model configuration, and experiment forecast model outputs, please refer to the following link (external).

The Integrated Marine Observing System (IMOS) Coastal Wave Buoy (CWB) facility, established in 2024, supports the national effort to monitor coastal ocean conditions and validate emerging technologies for marine observations. One of the key objectives of the facility is to assess innovative and cost-effective monitoring solutions that can complement or potentially replace traditional approaches. This study focuses on evaluating the performance of a Sofar smart and half smart mooring configurations in comparison to conventional deployments.
Specifically, we compare observations from a smart mooring that integrates a Sofar Spotter wave sensor with a Aanderaa current meter into a single mooring line, against those obtained from separate, standalone deployments of Sofar Spotter (waves) and Nortek Signature1000 (waves and current profiles) instruments. The deployments were conducted from May to November 2025 at Floreat Beach, Western Australia (~21m depth), a site exposed to a wide range of wave, water level, and current conditions. In addition, half smart mooring configurations were tested at Cape Bridgewater in Victoria and Hillarys in Western Australia. These setups integrated Sofar Spotter sensors with temperature loggers to assess their suitability for wave and temperature monitoring.
The aim of this study is to assess the suitability of the smart and half smart mooring configuration for long-term coastal monitoring by examining the consistency, quality, and reliability of the wave, current and temperature data. We investigate how mooring configuration influences the measurements and identify any potential trade-offs associated with the integrated setup. Additionally, the study evaluates operational benefits, including cost savings, deployment simplicity, and reduced equipment footprint.
Findings from this work will provide valuable insights into the feasibility of smart moorings for national-scale ocean observing systems and inform future infrastructure planning. The results will also contribute to the broader effort of adopting scalable and efficient technologies in support of sustainable ocean monitoring.

There are many ways to measure 'depth' however this paper will focus on the use of pressure-based loggers to determine water level measurements for sea level monitoring, tidal averaging, and pressure derived wave measurements. The paper will also discuss the use of pressure-based loggers for monitoring oceanographic moorings for knock down and evidence of trawling, and their use on profiling bodies and autonomous vehicles.
Although other equipment manufacturers will also be included in the analysis of different methodologies, the paper will use the RBR range to demonstrate the difference between piezo-resistive based loggers and Digiquartz® based loggers in terms of accuracy, resolution, data quantity, endurance, and cost.
'Pros and cons' of the different methods of water level determination will be considered, along with a discussion on the difference between accuracy and resolution, understanding drift and the importance of calibration. This will also include factors affecting accuracy and methods to get the highest accuracy out of your instrument.
Processing of water level data will be detailed including determining tide and/or wave parameters from the raw data, but most importantly, how to remove the atmospheric pressure component from the pressure signal recorded at the sensor. Examples of different deployment programs using pressure-based loggers will be given, along with deployment methodologies.
Finally, the paper will look at some recent developments including the integration of bottom based pressure sensors into a 'Smart Mooring' to not only allow real-time data transmission to shore, but comparison between that data and the buoy determined data recorded at the surface buoy.

Surface wave measurements by accelerometer-based oceanographic buoys are regarded as the "ground truth" for the validation of sea-state prediction models, providing key inputs for offshore and coastal engineering. Historically, engineering practice has relied on bulk statistics of wave motion, so H_s, T_z, frequency spectrum. However, a modern wave buoy can provide continuous full 3D vector displacement motion for individual waves in time, opening new possibilities for wave analysis.
We investigate the measured motion of a wave buoy in three orthogonal directions, and explore the relationships between these, for both severe and relatively benign sea-states. A NewWave-type analysis is used to investigate the average shape of the large events across the measured time histories. In combination with a conditioning analysis, we give a reciprocity relation between the vertical displacement of the wave buoy and those in the horizontal plane. The relationship is of value, as it connects the observed wave kinematics in the horizontal plane to the vertical motion. Significant nonlinear low and high frequency contributions are observed in the measured motion, larger horizontally and smaller vertically - despite the widely held view that buoys move close to linearly. Much of this nonlinearity is consistent with 2^nd order wave-wave interactions, so can be removed from the motion, but the mooring system may also play a role at low frequency.
This data-driven analysis paves the way for wave-by-wave prediction in steep seas with applications to the active control of wave energy converters, floating wind turbines and perhaps also marine current turbines.

An adequate understanding of swell propagation and decay across oceanic basins remains one of the greatest limitations in spectral wave models. Although theories exist for swell decay, adequate experimental confirmation of theory is lacking. To date, a paucity of insitu measurements and the measurement limitations of remote sensing techniques have limited validation and calibration of such theories. This presentation will consider data from the Sofar drifting buoy network in the Pacific. Due to the number of buoys distributed across the ocean, it is possible to identify high latitude swell generation sources and track the swell along Great Circle paths for up to 10,000km. In many cases, as many as 20 buoys provide point measurements of the directional wave spectrum along the propagation path. As such, this dataset represents a unique step change in available data to understand swell decay. The presentation will investigate swell arrival times, swell decay as a function of frequency, the role of island blocking and changes in directional spreading. The results will also investigate the adequacy of swell decay terms used in spectral wave models by using Wavewatch III to model the observed swell events. A detailed error analysis will investigate the importance of accurate prediction of the location and intensity of generating storm systems in determining the magnitude and arrival time of the swell.

Infragravity (IG) period oscillations (periods between 30 and 300s) in harbours and marinas in Western Australia often result in interruption to port operations due to excessive vessel movements. Many previous studies globally have shown relationships between wind (sea/swell) waves and IG waves. For example, it has been suggested that swell waves propagate as well-defined groups from deep water to water depths less than a few meters and significant amount of wave energy can be transferred from the wind waves to the IG waves. This implies that IG wave energy is generally low in deep water and increases where the depth decreases such as near offshore reefs and at the shoreline. In this paper, data from continuously recording pressure sensors (RBR Solo recording at 2Hz) from offshore Watermans Bay in water depths ~2m are analysed to examine the relationship between IG waves and wind waves in different bands (sea < 8s; swell 8-15s; long period swell > 15). Repeated deployments at the same location enables the definition of seasonal variability in IG wave energy. The results indicated that IG energy was: (1) present throughout the year with maxima during winter; (2) negligible during local storms under low swell conditions; (3) always associated with swell waves > 15 s; and, (4) enhanced during storm events when the local sea waves interacted with swell waves. There were strong linear relationships with swell and IG wave heights.

Major update of the physics of the third-generation models will be presented. The new source terms for wind input, whitecapping dissipation, interaction of waves with adverse winds (negative input), swell attenuation, nonlinear interactions, wave-bottom interactions and wave-ice interactions have been developed and implemented in WAVEWATCH-III (ST6, IC5), SWAN and WAM models, including operational versions and public releases of these model. Physics and parameterisations for the new source functions are based on observations, which allowed us to reveal features and processes previously unknown and not accounted for. For extreme conditions, physics of the wind input and whitecapping dissipation terms exhibit additional features irrelevant or inactive at moderate weather. In order to test the source functions independently, and control the flux balance in the model, additional observation-based constraints are implemented. At each time step, the total momentum input is verified to match an independently known wind stress. The new versions of the models have undergone extensive testing by means of academic tests, regional and global wave hindcast, modelling extreme conditions ranging from tropical cyclones to the marginal ice zone.

Data assimilation in wave models has lagged behind that in atmospheric and ocean circulation models, largely due to sparse observations in remote regions such as the Southern Ocean. This region generates some of the world's most energetic waves, with swells that travel thousands of kilometres into the Indian, Pacific, and Atlantic Oceans. Accurate forecasts of these swells are vital for maritime safety, offshore operations, and coastal planning, yet systematic biases remain, especially in predicting arrival times. Recent increases in wave observations, combined with advances in assimilation techniques, now create new opportunities. We present an extension to the open-source WaveWatch-III model that assimilates wave height and period from satellite altimetry and buoy data using an efficient optimal interpolation scheme. Implemented within the Australian Bureau of Meteorology's AUSWAVE global system, this module delivers measurable improvements in swell forecasts with minimal computational overhead. The framework leverages global datasets, including a unique drifting buoy array across the Southern and Indian Oceans (available via https://wawaves.org), together with established buoy networks worldwide. Ongoing work will expand the system to assimilate newly derived wave period estimates from satellite altimeters, providing further potential to advance global swell prediction skill,

Accurate wave forecasts, especially for swell, are vital for various maritime and coastal applications. Modern spectral wave models generally provide accurate forecasts but model errors in some parameters, such as swell arrival time, and amplitude remain a significant issue. Hence, Data Assimilation (DA) can provide a means to reduce forecast errors without altering the model source terms.
This study presents recent developments in assimilating wave observations into the WaveWatch III (WWIII) model using an enhanced Optimal Interpolation (OI) DA framework. Significant wave height (Hs) data from more than 150 drifting buoys and eight satellite altimeter datasets were assimilated globally. The data from buoys and altimeters were assimilated both separately and in combination to evaluate the relative and combined benefits of each platform. Altimeters offer high spatial coverage but low temporal resolution, whereas buoys provide long-duration measurements at fixed locations but with limited spatial coverage. This study investigates how these characteristics of the two platforms impact the forecast results. We also investigate how forecast improvements vary as a function of the spatial distance and time elapsed since the most recent nearby assimilation. Results show that forecast errors reduce more significantly when verification buoys are closer to recent assimilation points. Additionally, we assess the sensitivity of forecast improvements to different spatial structures of the background error correlation matrix. Findings indicate that the length scale of the correlation matrix has a greater impact on forecast accuracy than its shape or amplitude. Assimilation of altimeter data alone led to approximately a 25% reduction in the root mean square error (RMSE) of forecasted Hs, with an additional 5% improvement on the first forecast day when buoy data were included. Notably, similar improvements can be achieved with fewer buoys if they are strategically located, particularly in the Southern Indian Ocean and Southern Ocean where storm activity is highest.

Assimilation of wave buoy observations into third-generation spectral wave models is becoming increasingly common as more in situ and satellite wave measurements become available. Traditionally, these efforts have focused on bulk parameters assimilation, where integrated quantities such as significant wave height, mean direction, or peak period are used to geometrically correct the model state (e.g. energy scaling, directional rotation, frequency shifting). While effective in some cases, this approach limits the capacity to adjust more detailed spectral features. By contrast, assimilating observations in spectral space offers the potential to improve model performance more substantially, enabling refined adjustments to the energy distribution across frequency and direction that better capture the physical structure of the sea state.
Assimilating spectral measurements into wave models is challenging due to the high dimensionality of the system state, which includes the spatially distributed directional wave spectrum discretized across frequencies and directions. Classical Gaussian-based assimilation methods offer a proper Bayesian framework to deal with the indirect nature of spectral buoy observations-typically limited to the one-dimensional variance density spectra and low-order directional Fourier coefficients. But they become impractical as they require estimating high-dimensional background error covariance matrices which models correlation of the correction across space, frequency, and direction. Existing approaches handle this dimensionality issue either through assimilation directly in the observation space followed by an update of the model background or through partition-based assimilation after partition cross-assignment. Both approaches involve multi-step processes and heuristics aimed at preserving the geometric structure of the wave field.
Inspired by point cloud registration from computer vision literature, we propose a novel single-step assimilation strategy that circumvents the need for frequency-direction covariance estimation. The method defines the updated wave spectrum (i.e. the analysis) as a parametrized geometric transformation of the model background, rather than adjusting individual frequency or directional components. Parameters of the transformation are estimated through the classical variational assimilation scheme minimizing only the observational error, which is the quadratic penalty on the mismatch between observations and the model, weighted by the inverse of the observational error covariance matrix.
We will present numerical experiments with synthetic and real-world wave buoy data that demonstrates our approach accurately assimilates spectral measurements while preserving the geometric integrity of the directional spectrum.

We evaluate the effects of downscaling, source terms, and tidal interactions on numerical wave forecasts in Aotearoa New Zealand. We utilised a set of three nested domains (from global to regional scale) to examine significant wave height (Hs), mean period (Tm01), and peak wave direction at two coastal locations, Banks Peninsula and Baring Head. Downscaling markedly improved forecast accuracy at Baring Head, a tidally constricted region, reducing Hs forecast error by 25%. However, improvements at Banks Peninsula were minimal, likely due to its open coast characteristics which are adequately represented even by lower resolution models. Source term enhancements using default ST6 parameters generally improved Hs predictions on the west coast but worsened them on the east, indicating a geographical dependency in model performance. This variability was also evident in the Tm01 predictions, with notable improvements in bias reduction through model downscaling, particularly at Baring Head. Tidal influences were significant, especially at Baring Head, where they enhanced the forecast accuracy of wave height and direction due to the strong tidal currents and relative vorticity characteristic of this location. In contrast, at Banks Peninsula, tidal effects were less pronounced. The study underscores the importance of tailored modelling approaches that consider local geographical and hydrodynamic conditions to optimise wave forecasting.

Accurate coastal ocean and wave modeling is critical for coastal management, hazard forecasting, and climate impact assessment. Yet configuring, executing, and managing numerical models like SWAN and SCHISM remains technically demanding. ROMPY (Relocatable Ocean Modelling in Python) is a Python-based framework designed to streamline and unify this process through declarative configuration, template rendering, and integration with scientific-data ecosystems. At its core, Rompy introduces a layered architecture comprising:

1. A configuration layer, based on templated cookie-cutter models and Pydantic-validated configuration classes
2. A grid and data abstraction layer for geographic representation and structured classes for managing inputs
3. An execution orchestration layer, which automates generation of input files, model execution (locally or in containers), and post-processing in composable pipeline stages.

ROMPY currently supports SWAN wave modeling with built-in template packages and notebook-based demonstrations; SCHISM and XBeach wrappers are under active development, enhancing ROMPY's flexibility across coastal model systems.
By leveraging cookie-cutter templates, xarray semantics, and Pydantic validation, ROMPY enables users to declaratively define model configurations, manage spatial/temporal filtering of data inputs, and orchestrate workflow components reproducibly.
In this presentation, we will illustrate how ROMPY helps users construct replicable workflows for coastal wave simulations using SWAN, including configuration management, bathymetry and forcing data processing, boundary condition setup, and result visualization within Jupyter notebooks. We will also preview ongoing extension work for SCHISM and XBeach, and invite community contributions via plugin interfaces and execution backends.
ROMPY offers a unified, flexible, and reproducible approach to coastal ocean modeling in Python, accelerating research and operations across the marine modeling community. The code is open source and freely available.

Accurate parameterization of bottom roughness is critical for the performance of spectral wave models such as SWAN, particularly in shallow and reef-lined coastal environments. However, traditional calibration approaches, often based on trial and error or manual tuning, are time-consuming, potentially suboptimal, and rely heavily on repeated field measurements. While methods such as adjoint modelling and inversion techniques have been proposed for parameter estimation, there remains a gap in open-source, scalable tools for automatically calibrating spatially variable bottom friction and vegetation parameters in wave models.
This study presents a novel implementation of Approximate Bayesian Computation (ABC) to automatically calibrate spatially heterogeneous bottom coefficients within SWAN. The method was tested along a temperate reef system in Dawesville, Western Australia using three months of field observations (Dec 2023–Mar 2024), including pressure transducers, Spotter buoys, and Acoustic Wave and Current Profilers (AWACs). Wind sea and swell partitions were treated separately in the ABC framework to better resolve their differing sensitivities to bottom friction and vegetation drag. The proposed approach aims to improve model skill while reducing calibration effort, particularly in environments with strong spatial gradients in benthic substrates.
Initial results show that while constant friction performs reasonably well, accounting for spatial variability leads to improved model performance in key regions. Furthermore, the automated method reduces calibration time significantly compared to manual approaches. This study offers a promising path toward efficient, scalable and physics informed calibration of bottom parameters in coastal wave modelling, with implications for nature-based solutions for coastal systems, and operational forecasting.

In wave-dominated coastal regions such as New South Wales (NSW), variations in wave energy significantly impact shoreline dynamics. To protect coastal infrastructure from shoreline recession and plan effective mitigation strategies, wave projections that consider human-induced climate-change are crucial.
Current wave projections rely on coarse-resolution GCMs that are inadequate given the region's complex coastline and climate. This study projects 21^st century coastal waves by using a high-resolution NARCliM RCMs ensemble and a WaveWatchIII based wave modelling system. Two versions of ensembles, NARCliM1.5 and NARCliM2.0, are utilized for a range of future climate scenarios. The study examines three key aspects of future wave climate: modal conditions (50^th percentile), high wave conditions (95^th percentile of significant wave height, Hs), and severe wave conditions (99^th percentile of Hs). These wave projections are corrected for wave directionality biases using DAGQM method.
Wave projections from the six-member NARCliM1.5 ensemble under RCP8.5 indicate that Hs in modal conditions may increase slightly, up to +0.25 m. In contrast, Hs in high (95^th percentile) and severe (99^th percentile) wave conditions, though subject to model uncertainties, could rise by up to 0.4 m and 0.55 m, respectively, particularly in the central region of the NSW coastline. Peak wave periods may decrease by up to 0.5 seconds for modal conditions but could increase by up to 1 and 1.5 seconds for high and severe conditions, respectively. Peak wave direction changes are minimal for modal conditions but tend to rotate clockwise for high and severe conditions, reaching 14° and 18°, respectively. The reduced magnitudes under the RCP4.5 scenario highlight the importance of climate targets in mitigating adverse impacts.
Ongoing investigations using the latest 10-member NARCliM2.0 ensemble aim to further refine these projections. These high-resolution simulations will enhance our understanding of future waves and associated uncertainties, informing coastal engineering design and long-term planning.

The present work investigates the effects of wave-current interactions in a tidally dominant harbour, utilising Darwin Harbour in Australia as a case study. A series of numerical experiments was performed using a two-way coupled FVCOM modelling system across five distinct configurations: a hydrodynamic model, a fully coupled wave-current interaction, a wave model, a fully coupled model excluding wind forcing, and a fully coupled model without offshore wave boundary input (no swell). The findings indicate that the fully coupled model generates the strongest and spatially diverse velocity fields during both flood and ebb tides in nearshore zones. A comparative analysis of Eulerian, Stokes, and Lagrangian velocities indicates that Lagrangian transport patterns predominantly align with Eulerian circulation, primarily influenced by tidal flows. At the same time, surface Stokes drift plays a significant role in shallow areas. Directional patterns of surface and bottom current fields indicate wave-induced modification of flow vectors near wave-exposed boundaries and shallow zones, resulting in spatial variation in current direction and intensity between different regions of the harbour during flood and ebb phases. In addition, the spatial autocorrelation analysis of wavelength reveals a sharp contrast between outer and inner harbour dynamics, with swells ensuring coherent long-period wave behaviour offshore, while local wind forcing leads to shorter wavelengths and higher variability inland. Fully coupled simulations demonstrate an increased significant wave height during both flood and ebb tides in shallow areas. In contrast, wave-only scenarios show modified wave energy in regions characterised by strong current refraction and shoaling. The results emphasise the importance of two-way coupling for accurately depicting wave transformations, flow patterns, and directional energy redistribution in morphodynamically complex, tide-dominated harbours. The findings provide essential insights for enhancing coastal circulation modelling.

While wave energy attenuation by emergent vegetation has been extensively studied, research on the reduction of wave runup and its components, such as wave setup, sea-swell swash, and infragravity wave swash, is scarce. This study presents a combined laboratory and numerical investigation into the potential of emergent vegetation canopies (such as those formed by intertidal salt marsh vegetation and mangroves under high tide or storm surge conditions) to reduce wave runup and its components on coastlines. Wave runup across a range of regular and irregular wave conditions is found to be considerably reduced due to the presence of vegetation canopies in the surf and swash zone. Due to complexity in observing wave rundown and setup experimentally, a phase-resolving, non-hydrostatic wave model was validated using the runup observations and subsequently used to quantify the influence of vegetation drag on sea-swell and infragravity swash as well as wave setup. The model was further used to expand to a broad range of scenarios covering variations in wave characteristics, beach slope and vegetation density. It was found that the effectiveness of the vegetation canopies in reducing wave runup height and its components (i.e., setup, sea-swell-band swash and infragravity-band swash) decreases with increasing Keulegan-Carpenter number. A simple empirical formulation was derived that allows for runup prediction based on the offshore wave conditions, beach slope and vegetation properties. Overall, this study provides new insights in how coastal intertidal habitats such as salt marshes and mangroves may considerably reduce wave runup under high tide and storm surge conditions, and therefore can play an important role in reducing risk of wave-driven coastal flooding.

Coastal ocean extremes result from the dynamic interaction of storm surge, waves, astronomical tides, and sea level rise (SLR). These events drive erosion and inundation, threaten infrastructure, disrupt communities and impact ecosystems. Australia, with its population and infrastructure concentrated in coastal regions is particularly exposed. Regional variations in tidal range, wave exposure, and climate drivers further complicate the picture. For instance, the north is dominated by high tides and tropical cyclones, the south and southwest by westerly frontal systems and high wave climates, and the southeast by east coast lows.
To better understand and predict these hazards, we developed the Coupled Coastal Hazards Prediction System (CCHaPS) through the Australian Climate Service (ACS). CCHaPS is a high-resolution (~250 m) national-scale, coupled hydrodynamic-wave model that simulates waves, storm surge, tides, currents, and sea levels. It incorporates new boundary forcing datasets, including:

• Global Wave Hindcast for the ACS (WHACS)
• Downscaled atmospheric forcing (12.5 km),
• CMIP6-based global wave climate projections, leveraging enhanced atmospheric resolution in the Australian region
• AR6 regionalised SLR projections
• Updated national digital elevation models (DEMs)

We used CCHaPS to produce a 43-year hindcast (1981-2023) and 20-year historical (1995-2014) and future (2081-2100) climate (projection) simulations. These simulations will help quantify historical and projected changes in coastal extremes and their dynamic interactions, allow exploration of regional differences, and provide inputs to further downscaling process models, such as for erosion or compound flooding and subsequent risks. In this presentation, we provide an overview of the CCHaPS underpinning numerical model configuration and verification as well as simulation data structure and availability.

The Australian Climate Service (ACS) was established in 2021 to improve data intelligence on climate hazards and risks, which led to the development of a new global wave dataset WHACS: the Wave Hindcast for ACS, a multi-decadal global wind-wave hindcast that provides invaluable data for applications in coastal management, climate research, and renewable energy projects, ultimately helping communities and industries make informed decisions to improve safety, efficiency, and resilience regarding wave conditions. This dataset features a near-global spherical multi-cell (SMC) grid that aligns with the Bureau operational wave forecast model with a resolution of 1/8° (excluding poles) down to 1/16° (approximately 6 km) in continental shelf regions and depths shallower than 200 metres. The model has been calibrated to better represent extreme wave conditions by improving the representation of extreme winds. Spanning from 1979 to near present, WHACS available output consists of multiple hourly bulk and spectral partition wave parameters for the native SMC grid, as well as regular global and regional regrids of bulk wave parameters. For the Indo-Pacific, a gridded output of full spectral data is available across exclusive economic zones at a resolution of 1 degree in the open ocean and down to 30 arcminutes at depths shallower than 500 metres, and 15 arcminutes within 25 km of coastlines.

Understanding how extreme coastal wave events are evolving is essential for assessing coastal risks under climate change. While previous studies have highlighted trends in average wave conditions and extreme individual wave heights, no global assessment has yet examined how the most energetic discrete coastal wave storm events have changed over time. Cumulative wave energy, a function of wave height and duration, has been found to be a reliable indicator of coastal erosion potential, particularly along sandy, wave-dominated coastlines. This study investigates whether cumulative wave energy of extreme individual coastal wave events has increased over recent decades at the global scale.
Global coastal wave data from ERA5 (1979‐2024) were used to identify wave storm events using a Peaks Over Threshold method, applying a 95th percentile wave height threshold. Cumulative wave energy was computed as the time-integrated energy over each event. Events shorter than 12 hours were excluded, and those with peak wave heights occurring within 48 hours were merged. At each location, the Top 10 and Top 50 events were selected based on cumulative wave energy and compared across two time slices: 1979‐2001 and 2002‐2024.
Results reveal a general global increase in cumulative energy of wave storm events, although with significant regional variability. Notably, Western Africa exhibits a substantial rise, while Western North America and parts of Europe show a decline. A strong positive correlation between changes in cumulative energy and event duration indicates that longer-lasting storms are the primary driver of intensification. In contrast, weaker correlations with peak or average wave height suggest a secondary role in driving energy increases.
This global analysis provides key insights into regional hotspots related to shoreline erosion associated with individual storms and highlights that storm wave energy of the most extreme events has increased in some regions over recent decades.

The first pass of Australia' National Climate Risk Assessment identified coastal erosion and shoreline change as one of ten priority hazards. Indeed, with a large proportion of the population residing along the coast and increasing uncertainty around future storm patterns, there is a pressing need for reliable tools to support preparedness, disaster risk reduction, and climate adaptation.
This study introduces the Coastal Hazard Index (CHI) – a single composite parameter that combines the influence of three key drivers of storm-driven coastal erosion together into a single dimension-less value. The CHI consists of three terms to represent each ingredient to the hazard: wave energy flux, wave direction, and water level.
In the development of CHI, we utilised two recently developed hindcasts: the Wave Hindcast for the Australian Climate Service (WHACS) and the Coupled Coastal Hazard Prediction System (CCHaPS). From these hindcast, the CHI was calculated at virtual wave buoys located approximately 10 km offshore along the Australian coastline, and the outputs used as a proxy for potential coastal erosion. The CHI results were then assessed against well-known past events with significant impacts.
This nationally consistent model-based analysis provides critical insights to support coastal hazard risk assessments and inform policy and management strategies at both regional and national levels. It establishes a baseline for improving damaging surf thresholds, contributing to the Bureau of Meteorology's Coastal Hazard Warnings.

Accurate forecasting of swell arrival times in the open ocean remains a challenge, with errors often reaching several hours. One potential source of this uncertainty is the influence of ocean currents, which are frequently present along swell propagation paths. Ocean currents can modify both the group velocity and direction of swell propagation, yet their effects are not explicitly accounted for in most wave models. In this study, we investigate the role of ocean currents in swell propagation through a case study approach, combining satellite observations, buoy measurements, and model outputs with a wave ray tracing method. This allows us to explore how current-induced changes in propagation may contribute to discrepancies in swell arrival time forecasts. Our preliminary results suggest that incorporating ocean current effects into wave models can improve the accuracy of swell arrival time forecasts, particularly over long-distance propagation scenarios. Furthermore, improving the spatial resolution of ocean current fields is also beneficial for better capturing swell-current interactions.

It has been shown that the morphodynamic behaviour of subsea pipelines and cables can be quite sensitive to the local ambient and storm ramp-up conditions. On a number of recent subsea cable projects, Aurora Offshore Engineering has had the opportunity to develop bespoke sets of ambient and storm metocean design criteria through the reanalysis of the project metocean hindcast datasets. It has been found from these studies that the use of bespoke site-specific criteria produces more accurate and more reliable on-bottom stability outcomes compared to the use of generic metocean criteria. This presentation summarises the findings of a case study where bespoke metocean criteria were developed for a cable route in the eastern region of the English Channel (La Manche). The English Channel is renowned for its high tidal range and complex tidal regime. Furthermore, tidally induced currents dominate the hydrodynamic and sedimentary processes in this area, with current strength strongly related to tidal phase. To create an accurate representation of the site morphodynamic conditions to which the subsea cable would be exposed, the metocean criteria developed incorporated a range of processes such as wave-tide interactions, correlation of tide with surge during storm events and variation in wave refraction with water level. Furthermore, storm events were extracted at a range of tidal phases to incorporate high and low water levels. Although the bespoke metocean criteria were developed for a specific location, many of the insights and behaviours are applicable to other projects in the English Channel where infrastructure is being installed in areas with large and complex tidal regimes.