Explore the vast range of titles available.

This study presents OceanCastNet (OCN), a machine learning approach for wave forecasting that incorporates wind and wave fields to predict significant wave height, mean wave period, and mean wave direction. We evaluate OCN's performance against the operational ECWAM model using two independent data sets: NDBC buoy and Jason‐3 satellite observations. NDBC station validation indicates OCN performs better at 24 stations compared to ECWAM's 10 stations, and Jason‐3 satellite validation confirms similar accuracy across 228‐hr forecasts. OCN successfully captures wave patterns during extreme weather conditions, demonstrated through Typhoon Goni with prediction errors typically within ±$\\pm $ 0.5 m. The approach also offers computational efficiency advantages. The results suggest that machine learning approaches can achieve performance comparable to conventional wave forecasting systems for operational wave prediction applications. Plain Language Summary Predicting ocean waves is important for maritime safety and coastal planning, but current computer models require powerful computers and take many hours to run. This study developed OceanCastNet, a machine learning model that learns to predict wave height, period, and direction by studying patterns in historical wave and wind data. We tested our approach against ECWAM, a widely used operational wave forecasting system, using two different types of validation data: ocean buoy measurements, and satellite observations. The results show that our machine learning approach can predict waves about as accurately as conventional models, performing better at 24 buoy locations compared to the conventional model's better performance at 10 locations, and successfully capturing wave patterns during Typhoon Goni in 2020. The machine learning approach also runs much faster, completing forecasts in seconds rather than hours, suggesting that artificial intelligence offers a promising alternative for wave forecasting applications. Key Points A deep learning wave model shows forecast skill comparable to operational systems with high computational efficiency An auto‐regressive process with wind forcing enables forecasting of waves during extreme weather events Satellite validation shows regional model strengths with the new model excelling in the West Pacific

IntroductionOcean wave conditions forecasting is crucial for reducing wave-related disasters and enhancing prevention and mitigation capabilities in China's coastal regions.MethodsThis study develops a convolutional neural network (CNN) model optimized by a random search algorithm to predict significant wave height (Hs) and mean wave period (Tm). Using SWAN nearshore wave simulation data as input, the model analyzes the impact of different historical input step lengths and various forecast lead times on prediction performance. It is applied to the Yantai Fishing Zone, China.ResultsThe optimal input step length for the model is 3 hours, achieving correlation coefficients (CC) of 0.9997 for Hs and 0.9969 for Tm. The mean absolute errors (MAE) are 0.0075 m and 0.0562 s, and the root mean square errors (RMSE) are 0.0149 m and 0.2014 s, respectively. Based on the 3-hour input step length, forecasts were conducted for lead times of 3, 6, 9, and 12 hours. As the forecast lead time increased, prediction accuracy declined, but the model still effectively captured the main trends of Hs and Tm.DiscussionThe model errors remain within an acceptable range, and its computational efficiency is significantly superior to traditional numerical methods. This demonstrates the model's good applicability in the study area, indicating its potential to effectively enhance fishery production efficiency and optimize fishing operation scheduling.

A wave forecasting system using FUNWAVE-TVD which is based on the fully nonlinear Boussinesq equations by Chen (2006) was developed to provide an accurate wave prediction in the Port of Busan, South Korea. This system is linked to the Korea Operational Oceanographic System (KOOS) developed by Park et al. (2015). The computational domain covers a region of 9.6 km×7.0 km with a grid size of 2 m in both directions, which is sufficient to resolve short waves and dominant sea states. The total number of grid points exceeds 16 millions, making the model computational expensive. To provide real-time forecasting, an interpolation method, which is based on pre-calculated results of FUNWAVE-TVD and SWAN forecasting results at the FUNWAVE-TVD offshore boundary, was used. A total of 45 cases were pre-calculated, which took 71 days on 924 computational cores of a Linux cluster system. Wind wave generation and propagation from the deep water were computed using the SWAN in KOOS. SWAN results provided a boundary condition for the FUNWAVE-TVD forecasting system. To verify the model, wave observations were conducted at three locations inside the port in a time period of more than 7 months. A model/model comparison between FUNWAVE-TVD and SWAN was also carried out. It is found that, FUNWAVE-TVD improves the forecasting results significantly compared to SWAN which underestimates wave heights in sheltered areas due to incorrect physical mechanism of wave diffraction, as well as large wave heights caused by wave reflections inside the port.

Equatorial waves are a major driver of widespread convection in Southeast Asia and the tropics more widely, a region in which accurate heavy rainfall forecasts are still a challenge. Conditioning rainfall over land on local equatorial wave phases finds that heavy rainfall can be between 2 and 4 times more likely to occur in Indonesia, Malaysia, Vietnam, and the Philippines. Equatorial waves are identified in a global numerical weather prediction ensemble forecast [Met Office Global and Regional Ensemble Prediction System (MOGREPS-G)]. Skill in the ensemble forecast of wave activity is highly dependent on region and time of year, although generally forecasts of equatorial Rossby waves and westward-moving mixed Rossby–gravity waves are substantially more skillful than for the eastward-moving Kelvin wave. The observed statistical relationship between wave phases and rainfall is combined with ensemble forecasts of dynamical wave fields to construct hybrid dynamical–statistical forecasts of rainfall probability using a Bayesian approach. The Brier skill score is used to assess the skill of forecasts of rainfall probability. Skill in the hybrid forecasts can exceed that of probabilistic rainfall forecasts taken directly from MOGREPS-G and can be linked to both the skill in forecasts of wave activity and the relationship between equatorial waves and heavy rainfall in the relevant region. The results show that there is potential for improvements of forecasts of high-impact weather using this method as forecasts of large-scale waves improve.

A 72-h fine-resolution atmosphere-wave-ocean coupled forecasting system was developed for the South China Sea and its adjacent seas. The forecasting model domain covers from from 15°S to 45°N in latitude and 99°E to 135°E in longitude including the Bohai Sea, the Yellow Sea, the East China Sea, the South China Sea and the Indonesian seas. To get precise initial conditions for the coupled forecasting model, the forecasting system conducts a 24-h hindcast simulation with data assimilation before forecasting. The Ensemble Adjustment Kalman Filter (EAKF) data assimilation method was adopted for the wave model MASNUM with assimilating Jason-2 significant wave height (SWH) data. The EAKF data assimilation method was also introduced to the ROMS model with assimilating sea surface temperature (SST), mean absolute dynamic topography (MADT) and Argo profiles data. To improve simulation of the structure of temperature and salinity, the vertical mixing scheme of the ocean model was improved by considering the surface wave induced vertical mixing and internal wave induced vertical mixing. The wave and current models were integrated from January 2014 to October 2015 driven by the ECMWF reanalysis 6 hourly mean dataset with data assimilation. Then the coupled atmosphere-wave-ocean forecasting system was carried out 14 months operational running since November 2015. The forecasting outputs include atmospheric forecast products, wave forecast products and ocean forecast products. A series of observation data are used to evaluate the coupled forecasting results, including the wind, SHW, ocean temperature and velocity. The forecasting results are in good agreement with observation data. The prediction practice for more than one year indicates that the coupled forecasting system performs stably and predict relatively accurate, which can support the shipping safety, the fisheries and the oil exploitation.

Severe storms produce ocean waves with periods of 18–26 s, corresponding to wavelengths 500–1,055 m. These waves radiate globally as swell, generating microseisms and affecting coastal areas. Despite their significance, long waves often elude detection by existing remote sensing systems when their height is below 0.2 m. The new Surface Water Ocean Topography (SWOT) satellite offers a breakthrough by resolving these waves in global sea level measurements. Here we show that SWOT can detect 25‐s waves with heights as low as 3 cm, and resolves period and direction better than in situ buoys. SWOT provides detailed maps of wave height, wavelength, and direction across ocean basins. These measurements unveil intricate spatial patterns, shedding light on wave generation in storms, currents that influence propagation, and refraction, diffraction and reflection in shallow regions. Notably, the magnitude of reflections exceeds previous expectations, illustrating SWOT's transformative impact. Plain Language Summary Wind storms at sea make waves that increase in size with wind speed, and with the distance over which the high winds have been able to amplify the waves. Once generated these waves propagate as swell around the world ocean: in that stage the wave period remains constant while the wave height decay away from the source. Waves with periods longer than 18 s are relatively infrequent, but they are an important source of seismic waves and coastal impacts. However, current remote sensing techniques miss long waves under 0.2 m high. The Surface Water Ocean Topography (SWOT) satellite mission changes this, spotting 25‐s waves with heights as low as 3 cm. SWOT maps wave height, wavelength, and direction worldwide, revealing the influence of winds, currents and water depth. For example, We found stronger than expected coastal reflection, which will help revise wave forecasting models and their application in seismology. Key Points Surface Water Ocean Topography (SWOT) data provide the first open ocean spatial measurements of phase‐resolved swells with wavelength 500–1,050 m Swells with heights as low as 3 cm are well detected by SWOT, allowing tracking across oceans Swell reflection off the coast can be separated from incident waves

The establishment of the Great Lakes wave forecast system is an early success story inspiring the introduction of open-innovation practices at the U.S. National Oceanic and Atmospheric Administration (NOAA). It shows the power of community modeling to accelerate the transition of scientific innovations to operational environmental forecasting. This paper presents an overview of wave modeling in the Great Lakes from the perspective of its societal benefits. NOAA’s operational wave modeling systems and development practices are examined, emphasizing the importance of community- and stakeholder-driven collaborative efforts to introduce innovations such as using advanced spatial grid types and physics parameterizations, leading to improved predictive skill. The success of the open-innovation approach, set in motion at NOAA by initiatives such as the Great Lakes wave forecasting system, accelerated the transition of innovations to operations. The culture change to operational modeling efforts became part of the foundation for establishing the Unified Forecast System and, more recently, the Earth Prediction Innovation Center. Open-innovation initiatives will improve operational weather and climate forecast systems through scientific and technical innovation, reducing the devastating impacts of hazardous weather and supporting NOAA’s mission of protecting life and property and enhancing the national economy.

A predictability study on wave forecast of the Arctic Ocean is necessary to help identify hazardous areas and ensure sustainable shipping along the trans-Arctic routes. To assist with validation of the Arctic Ocean wave model, two drifting wave buoys were deployed off Point Barrow, Alaska for two months in September 2016. Both buoys measured significant wave heights exceeding 4 m during two different storm events on 19 September and 22 October. The NOAA-WAVEWATCH IIIⓇ model with 16-km resolution was forced using wind and sea ice reanalysis data and obtained general agreement with the observation. The September storm was reproduced well; however, model accuracy deteriorated in October with a negative wave height bias of around 1 m during the October storm. Utilising reanalysis data, including the most up-to-date ERA5, this study investigated the cause: grid resolution, wind and ice forcing, and in situ sea level pressure observations assimilated for reanalysis. The analysis has found that there is a 20% reduction of in situ SLP observations in the area of interest, presumably due to fewer ships and deployment options during the sea ice advance period. The 63-member atmospheric ensemble reanalysis, ALERA2, has shown that this led to a larger ensemble spread in the October monthly mean wind field compared to September. Since atmospheric physics is complex during sea ice advance, it is speculated that the elevated uncertainty of synoptic-scale wind caused the negative wave model bias. This has implications for wave hindcasts and forecasts in the Arctic Ocean.

Forecasting of waves under extreme conditions such as tropical cyclones is vitally important for many offshore industries, but there remain many challenges. For Northwest Western Australia (NW WA), wave forecasts issued by the Australian Bureau of Meteorology have previously been limited to products from deterministic operational wave models forced by deterministic atmospheric models. The wave models are run over global (resolution 1/4∘) and regional (resolution 1/10∘) domains with forecast ranges of + 7 and + 3 day respectively. Because of this relatively coarse resolution (both in the wave models and in the forcing fields), the accuracy of these products is limited under tropical cyclone conditions. Given this limited accuracy, a new ensemble-based wave forecasting system for the NW WA region has been developed. To achieve this, a new dedicated 8-km resolution grid was nested in the global wave model. Over this grid, the wave model is forced with winds from a bias-corrected European Centre for Medium Range Weather Forecast atmospheric ensemble that comprises 51 ensemble members to take into account the uncertainties in location, intensity and structure of a tropical cyclone system. A unique technique is used to select restart files for each wave ensemble member. The system is designed to operate in real time during the cyclone season providing + 10-day forecasts. This paper will describe the wave forecast components of this system and present the verification metrics and skill for specific events.

We analyze the sea state conditions during which the accident of the cruise ship Louis Majesty took place. The ship was hit by a large wave that destroyed some windows at deck number five and caused two fatalities. Using the wave model (WAM), driven by the Consortium for Small‐Scale Modelling (COSMO‐ME) winds, we perform a detailed hindcast of the local wave conditions. The results reveal the presence of two comparable wave systems characterized almost by the same frequency. We discuss such sea state conditions in the framework of a system of two coupled Nonlinear Schrödinger (CNLS) equations, each of which describe the dynamics of a single spectral peak. For some specific parameters, we discuss the breather solutions of the CNLS equations and estimate the maximum wave amplitude. Even though, due to the lack of measurements, it is impossible to establish the nature of the wave that caused the accident, we show that the angle between the two wave systems during the accident was close to the condition for which the maximum amplitude of the breather solution is observed. Key Points A hindcast of the Louis Majesty accident sea state condition have been performed The accident took place during crossing sea state conditions Rogue wave solutions describing the crossing sea condition are derived